Microbiology Research - Academic Journals

African Journal of

Microbiology Research

Volume 11 Number 21 7 June, 2017

ISSN 1996-0808

The African Journal of Microbiology Research (AJMR) is published weekly (one volume per year) by Academic Journals.

provides rapid publication (weekly) of articles in all areas of Microbiology such as: Environmental Microbiology, Clinical Microbiology, Immunology, Virology, Bacteriology, Phycology, Mycology and Parasitology, Protozoology, Microbial Ecology, Probiotics and Prebiotics, Molecular Microbiology, Biotechnology, Food Microbiology, Industrial Microbiology, Cell Physiology, Environmental Biotechnology, Genetics, Enzymology, Molecular and Cellular Biology, Plant Pathology, Entomology, Biomedical Sciences, Botany and Plant Sciences, Soil and Environmental Sciences, Zoology, Endocrinology, Toxicology. The Journal welcomes the submission of manuscripts that meet the general criteria of significance and scientific excellence. Papers will be published shortly after acceptance. All articles are peer-reviewed.

Contact Us

Editorial Office: [email protected]

Help Desk: [email protected]

Website: http://www.academicjournals.org/journal/AJMR

Submit manuscript online http://ms.academicjournals.me/

Editors

Prof. Stefan Schmidt Applied and Environmental Microbiology School of Biochemistry, Genetics and Microbiology University of KwaZulu-Natal Pietermaritzburg, South Africa. Prof. Fukai Bao Department of Microbiology and Immunology Kunming Medical University Kunming, China. Dr. Jianfeng Wu Dept. of Environmental Health Sciences School of Public Health University of Michigan USA. Dr. Ahmet Yilmaz Coban OMU Medical School Department of Medical Microbiology Samsun, Turkey. Dr. Seyed Davar Siadat Pasteur Institute of Iran Pasteur Square, Pasteur Avenue Tehran, Iran. Dr. J. Stefan Rokem The Hebrew University of Jerusalem Department of Microbiology and Molecular Genetics Jerusalem, Israel. Prof. Long-Liu Lin National Chiayi University Chiayi, Taiwan.

Dr. Thaddeus Ezeji Fermentation and Biotechnology Unit Department of Animal Sciences The Ohio State University USA. Dr. Mamadou Gueye MIRCEN/Laboratoire commun de microbiologie IRD-ISRA-UCAD Dakar, Senegal. Dr. Caroline Mary Knox Department of Biochemistry, Microbiology and Biotechnology Rhodes University Grahamstown, South Africa. Dr. Hesham Elsayed Mostafa Genetic Engineering and Biotechnology Research Institute (GEBRI) Mubarak City For Scientific Research Alexandria, Egypt. Dr. Wael Abbas El-Naggar Microbiology Department Faculty of Pharmacy Mansoura University Mansoura, Egypt. Dr. Barakat S.M. Mahmoud Food Safety/Microbiology Experimental Seafood Processing Laboratory Costal Research and Extension Center Mississippi State University Pascagoula, USA. Prof. Mohamed Mahrous Amer Faculty of Veterinary Medicine Department of Poultry Diseases Cairo university Giza, Egypt.

Editors Dr. R. Balaji Raja Department of Biotechnology School of Bioengineering SRM University Chennai, India. Dr. Aly E Abo-Amer Division of Microbiology Botany Department Faculty of Science Sohag University Egypt.

Dr. Haoyu Mao Department of Molecular Genetics and Microbiology College of Medicine University of Florida Florida, USA. Dr. Yongxu Sun Department of Medicinal Chemistry and Biomacromolecules Qiqihar Medical University Heilongjiang P.R. China. Dr. Ramesh Chand Kasana Institute of Himalayan Bioresource Technology Palampur, India. Dr. Pagano Marcela Claudia Department of Biology, Federal University of Ceará - UFC Brazil. Dr. Pongsak Rattanachaikunsopon Department of Biological Science Faculty of Science Ubon Ratchathani University Thailand. Dr. Gokul Shankar Sabesan Microbiology Unit, Faculty of Medicine AIMST University Kedah, Malaysia.

Dr. Kamel Belhamel Faculty of Technology University of Bejaia Algeria. Dr. Sladjana Jevremovic Institute for Biological Research Belgrade, Serbia. Dr. Tamer Edirne Dept. of Family Medicine Univ. of Pamukkale Turkey. Dr. Mohd Fuat ABD Razak Institute for Medical Research Malaysia. Dr. Minglei Wang University of Illinois at Urbana-Champaign USA. Dr. Davide Pacifico Istituto di Virologia Vegetale – CNR Italy. Prof. N. S. Alzoreky Food Science & Nutrition Department College of Agricultural Sciences & Food King Faisal University Saudi Arabia. Dr. Chen Ding College of Material Science and Engineering Hunan University China. Dr. Sivakumar Swaminathan Department of Agronomy College of Agriculture and Life Sciences Iowa State University USA. Dr. Alfredo J. Anceno School of Environment, Resources and Development (SERD) Asian Institute of Technology Thailand. Dr. Iqbal Ahmad Aligarh Muslim University Aligrah, India.

Dr. Juliane Elisa Welke UFRGS – Universidade Federal do Rio Grande do Sul Brazil. Dr. Iheanyi Omezuruike Okonko Department of Virology Faculty of Basic Medical Sciences University of Ibadan Ibadan, Nigeria. Dr. Giuliana Noratto Texas A&M University USA. Dr. Babak Mostafazadeh Shaheed Beheshty University of Medical Sciences Iran. Dr. Mehdi Azami Parasitology & Mycology Department Baghaeei Lab. Isfahan, Iran. Dr. Rafel Socias CITA de Aragón Spain. Dr. Anderson de Souza Sant’Ana University of São Paulo Brazil. Dr. Juliane Elisa Welke UFRGS – Universidade Federal do Rio Grande do Sul Brazil. Dr. Paul Shapshak USF Health Depts. Medicine and Psychiatry & Beh Med. Div. Infect. Disease & Internat Med USA. Dr. Jorge Reinheimer Universidad Nacional del Litoral (Santa Fe) Argentina. Dr. Qin Liu East China University of Science and Technology China. Dr. Samuel K Ameyaw Civista Medical Center USA.

Dr. Xiao-Qing Hu State Key Lab of Food Science and Technology Jiangnan University China. Prof. Branislava Kocic University of Nis School of Medicine Institute for Public Health Nis, Serbia. Prof. Kamal I. Mohamed State University of New York Oswego, USA. Dr. Adriano Cruz Faculty of Food Engineering-FEA University of Campinas (UNICAMP) Brazil. Dr. Mike Agenbag Municipal Health Services, Joe Gqabi, South Africa. Dr. D. V. L. Sarada Department of Biotechnology SRM University Chennai India. Prof. Huaizhi Wang Institute of Hepatopancreatobiliary Surgery of PLA Southwest Hospital Third Military Medical University Chongqing China. Prof. A. O. Bakhiet College of Veterinary Medicine Sudan University of Science and Technology Sudan. Dr. Saba F. Hussain Community, Orthodontics and Peadiatric Dentistry Department Faculty of Dentistry Universiti Teknologi MARA Selangor, Malaysia.

Prof. Zohair I. F. Rahemo Department of Microbiology and Parasitology Clinical Center of Serbia Belgrade, Serbia. Dr. Afework Kassu University of Gondar Ethiopia. Dr. How-Yee Lai Taylor’s University College Malaysia. Dr. Nidheesh Dadheech MS. University of Baroda, Vadodara, India. Dr. Franco Mutinelli Istituto Zooprofilattico Sperimentale delle Venezie Italy. Dr. Chanpen Chanchao Department of Biology, Faculty of Science, Chulalongkorn University Thailand. Dr. Tsuyoshi Kasama Division of Rheumatology, Showa University Japan. Dr. Kuender D. Yang Chang Gung Memorial Hospital Taiwan. Dr. Liane Raluca Stan University Politehnica of Bucharest Department of Organic Chemistry Romania. Dr. Mohammad Feizabadi Tehran University of Medical Sciences Iran. Prof. Ahmed H Mitwalli Medical School King Saud University Riyadh, Saudi Arabia.

Dr. Mazyar Yazdani Department of Biology University of Oslo Blindern, Norway. Dr. Babak Khalili Hadad Department of Biological Sciences Islamic Azad University Roudehen, Iran. Dr. Ehsan Sari Department of Plant Pathology Iranian Research Institute of Plant Protection Tehran, Iran. Dr. Snjezana Zidovec Lepej University Hospital for Infectious Diseases Zagreb, Croatia. Dr. Dilshad Ahmad King Saud University Saudi Arabia. Dr. Adriano Gomes da Cruz University of Campinas (UNICAMP) Brazil Dr. Hsin-Mei Ku Agronomy Dept. NCHU Taichung,Taiwan. Dr. Fereshteh Naderi Islamic Azad University Iran. Dr. Adibe Maxwell Ogochukwu Department of Clinical Pharmacy and Pharmacy Management, University of Nigeria Nsukka, Nigeria. Dr. William M. Shafer Emory University School of Medicine USA. Dr. Michelle Bull CSIRO Food and Nutritional Sciences Australia.

Prof. Márcio Garcia Ribeiro School of Veterinary Medicine and Animal Science- UNESP, Dept. Veterinary Hygiene and Public Health, State of Sao Paulo Brazil. Prof. Sheila Nathan National University of Malaysia (UKM) Malaysia. Prof. Ebiamadon Andi Brisibe University of Calabar, Calabar, Nigeria. Dr. Julie Wang Burnet Institute Australia. Dr. Jean-Marc Chobert INRA- BIA, FIPL France. Dr. Zhilong Yang Laboratory of Viral Diseases National Institute of Allergy and Infectious Diseases, National Institutes of Health USA. Dr. Dele Raheem University of Helsinki Finland. Dr. Biljana Miljkovic-Selimovic School of Medicine, University in Nis, Serbia. Dr. Xinan Jiao Yangzhou University China. Dr. Endang Sri Lestari, MD. Department of Clinical Microbiology, Medical Faculty, Diponegoro University/Dr. Kariadi Teaching Hospital, Semarang Indonesia. Dr. Hojin Shin Pusan National University Hospital South Korea.

Dr. Yi Wang Center for Vector Biology Rutgers University New Brunswick USA. Prof. Natasha Potgieter University of Venda South Africa. Dr. Sonia Arriaga Instituto Potosino de Investigación Científicay Tecnológica/ División de Ciencias Ambientales Mexico. Dr. Armando Gonzalez-Sanchez Universidad Autonoma Metropolitana Cuajimalpa Mexico. Dr. Pradeep Parihar Lovely Professional University Punjab, India. Dr. William H Roldán Department of Medical Microbiology Faculty of Medicine Peru. Dr. Kanzaki, L. I. B. Laboratory of Bioprospection University of Brasilia Brazil. Prof. Philippe Dorchies National Veterinary School of Toulouse, France. Dr. C. Ganesh Kumar Indian Institute of Chemical Technology, Hyderabad India. Dr. Zainab Z. Ismail Dept. of Environmental Engineering University of Baghdad Iraq. Dr. Ary Fernandes Junior Universidade Estadual Paulista (UNESP) Brasil.

Dr. Fangyou Yu The first Affiliated Hospital of Wenzhou Medical College China. Dr. Galba Maria de Campos Takaki Catholic University of Pernambuco Brazil. Dr Kwabena Ofori-Kwakye Department of Pharmaceutics Kwame Nkrumah University of Science & Technology Kumasi, Ghana. Prof. Liesel Brenda Gende Arthropods Laboratory, School of Natural and Exact Sciences, National University of Mar del Plata Buenos Aires, Argentina. Dr. Hare Krishna Central Institute for Arid Horticulture Rajasthan, India. Dr. Sabiha Yusuf Essack Department of Pharmaceutical Sciences University of KwaZulu-Natal South Africa. Dr. Anna Mensuali Life Science Scuola Superiore Sant’Anna Italy. Dr. Ghada Sameh Hafez Hassan Pharmaceutical Chemistry Department Faculty of Pharmacy Mansoura University Egypt.

Dr. Kátia Flávia Fernandes Department of Biochemistry and Molecular Biology Universidade Federal de Goiás Brasil. Dr. Abdel-Hady El-Gilany Department of Public Health & Community Medicine Faculty of Medicine Mansoura University Egypt. Dr. Radhika Gopal Cell and Molecular Biology The Scripps Research Institute San Diego, CA USA. Dr. Mutukumira Tony Institute of Food Nutrition and Human Health Massey University New Zealand. Dr. Habip Gedik Department of Infectious Diseases and Clinical Microbiology Ministry of Health Bakırköy Sadi Konuk Training and Research Hospital Istanbul, Turkey. Dr. Annalisa Serio Faculty of Bioscience and Technology for Food Agriculture and Environment University of Teramo Teramo, Italy.

African Journal of Microbiology Research

Table of Contents: Volume 11 Number 21 7 June, 2017

ARTICLES

Technology and microbiology of traditionally fermented food and beverage products of Ethiopia: A review 825 Guesh Mulaw and Anteneh Tesfaye Phenotypic identification of Escherichia coli O157:H7 isolates from cattle at Suleja Abattoir, Nigeria 845 Mailafia, S., Madubuike, S. A., Raji, M. A., Suleiman, M. M., Olabode, H.O. K., Echioda-Egbole, M. and Okoh, G. P. R. Isolation and screening of amylase producing thermophilic spore forming Bacilli from starch rich soil and characterization of their amylase activity 851 Mengistu Fentahun and Pagadala Vijaya Kumari Performance evaluation of oxacillin-resistant Staphylococcus aureus genotypes and taxa on human and animal blood agar culture media 860 Rajesh Prajapati, Shivam Yadav and Neelam Atri Antimicrobial effects of novel fluorous and non-fluorous surfactants 888 Kamonrat Phopin and Barry S. Bean Factors associated with candidemia by non-albicans Candida group in midwest region of Brazil: Eight-year cross-sectional study 908 Hugo Dias HOFFMANN-SANTOS and Rosane C. HAHN

Vol. 11(21), pp. 825-844, 7 June, 2017

DOI: 10.5897/AJMR2017.8524

Article Number: 16547EC64606

ISSN 1996-0808

Copyright © 2017

Author(s) retain the copyright of this article

http://www.academicjournals.org/AJMR

African Journal of Microbiology Research

Review

Technology and microbiology of traditionally fermented food and beverage products of Ethiopia: A review

Guesh Mulaw and Anteneh Tesfaye

School of Graduate Studies, Addis Ababa University, Ethiopia.

Received 14 March, 2017; Accepted 12 May, 2017

Fermented food and beverage products are made globally by using different practices, fresh materials and microbes. Fermented foods have ample sources of essential vitamins, minerals, enzymes, and antioxidants that are all enhanced through the process of fermentation. The advantageous effects related with fermented products have a special prominence during the production of these products in unindustrialized countries like Ethiopia. Therefore, coming to Ethiopia, fermented food and beverage products have practiced in a long history. During the production of traditional fermented food and beverage products, controlled natural fermentation process with the absence of starter cultures are used to initiate it. In Ethiopia, the local fermented food and beverage products are acid-alcohol type of fermentation. Moreover, the preparation of many traditionally fermented food and beverage products is still practiced in a household art thereby a wide variety of fermented foods and beverages are consumed in Ethiopia. Thus, this paper presents on the technology and microbiology of local fermented food and beverage products of Ethiopia. Key words: Fermented food, kocho, keribo.

INTRODUCTION In developed and developing countries, traditional fermented food and beverage products form an important part of the food. Therefore, these food products are prepared from plant and animal materials in which microbes play an important role by altering the material physically and nutritionally. African raw plant and animal materials are predominated by many lactic acid bacteria (LAB) and yeasts. Predominance of specific yeast species seems to be largely product specific (Christoph et al., 2017). Fermentation of row plant and animal material is one of the best and earliest methods of food

preparation and preservation. Thus, f ermented foods have a role in social functions such as marriage, naming and rain making ceremonies, where they are served as weaning foods. In addition, fermentation delivers a natural way to reduce the volume of the material to be transported, to extinguish undesirable components, to improve the nutritive value and appearance of the food, to reduce the energy required for cooking and to make a safer product (Simango, 1997).

Fermented food and beverage products are made

*Corresponding author: E-mail: [email protected].

Author(s) agree that this article remains permanently open access under the terms of the Creative Commons Attribution

License 4.0 International License

826 Afr. J. Microbiol. Res. globally by using different practices, fresh materials and microbes. Therefore, there are only four main fermentation processes which are alcoholic, lactic acid, acetic acid and alkali fermentation (Soni and Sandhu, 1990). Alcoholic fermentation marks in the production of ethanol, and yeasts are the major organisms. Lactic acid fermentation is produced by the presence of LAB. A second group of bacteria of importance in food fermentations are the acetic acid producers from the Acetobacter species. Acetobacter spp. converts alcohol to acetic acid in the existence of surplus oxygen. Alkali fermentation frequently takes place through the fermentation of fish and seeds, popularly known as condiment (McKay and Baldwin, 1990).

The advantageous effects related with fermented products have a special prominence during the production of these products in unindustrialized countries like Ethiopia. These effects resulted in decreased miss of raw materials, minimized cooking time, enhancement of protein quality and carbohydrate digestibility, upgraded bioavailability of micronutrients and removal of toxic and anti-nutritional factors (Sanni, 1993). In addition, the probiotic effects (resistant to low pH and antibacterial activity) and the low rate of pathogenic bacteria seen in fermented food and beverage products are especially important when it comes to undeveloped countries where fermented foods have been stated to reduce the severity of diarrhea (Kimmons et al., 1999). Thus, a better understanding of the intestinal microbial populations will contribute to the development of new strategies for the anticipation and/or treatment of several diseases (De Almad et al., 2015).

Fermented food products have been renowned for their superior dietary value and digestibility compared to their row materials. Hence, fermentation of plant materials such as maize, millet, sorghum and rice, marks in enhanced protein quality, especially the level of available lysine (Hamad and Fields, 1979; Padhye and Salunkhe, 1979). And also fermentation process has the ability to improve the organoleptic properties by making different flavors in different foods (Khetarpaul and Chauhan, 1993). In most of these products the fermentation is natural and involves mixed cultures of microbes. Thus, some microbes may participate in parallel, while others act in a sequential manner with an altering dominant biota during the course of fermentation. The common fermenting bacteria are species of Leuconostoc, Lactobacillus, Streptococcus, Pediococcus, Micrococcus and Bacillus. The fungal genera are Aspergillus, Paecilomyces, Cladosporium, Fusarium, Penicillium and Trichothecium (Steinkraus, 1998). Yeasts have been reported to be involved in various types of local fermented foods and beverage products and the most dominant yeast species accompanying with African indigenous fermented foods and beverage products is Saccharomyces cerevisiae (Jespersen, 2003).

It is known that, as the people of the world, the race of Ethiopia has its own views and approaches relating to foods. Some of these are associated to foods and diseases, while others are to qualities, such as hot and cold or light and heavy foods (Gebrekidan and Gebrettiwat, 1982). However, the absence of a writing culture in most of the country marks their origin difficult to trace. Perhaps, the most accepted of the fermented foods is started at 1970s. The nature of fermentation in Ethiopia is not complex and does not required expensive equipment. During production of traditional fermented food products in Ethiopia, it is common to use and follow controlled natural fermentation process with no defined starter cultures used to initiate it. Ethiopian local fermented foods and beverages are products of acid-alcohol type of fermentation. The preparation of many local fermented foods and beverages is still practiced at household. It is known that a wide variety of fermented foods and beverages are consumed in Ethiopia.

These include injera, ergo, Ititu, ayib, qibe, arrera, kocho, tella, awaze, borde and tejj. Thus, at different time different researchers were conducted studies on the mentioned local traditional fermented food and beverage products. Bearing in mind the rich diversity in fermented food and beverage types in the country; few studies were carried out in widely different parts of Ethiopia, and included the major ethnic groups. Therefore, this paper presents the technology and microbiology of local fermented food and beverage products of Ethiopia. TRADITIONAL FERMENTED PLANT FOODS Injera fermentation Injera is thin, fermented Ethiopian dish made from grains particularly, teff flour by mixing water and starter (ersho), which is a fluid, saved from previously fermented dough. Teff (Eragrostis tef (Zucc) Trotter) is the most widespread grain for making injera, although other grains such as sorghum, maize, barley, wheat and finger millet are sometimes used. Teff has the largest part of area (23.42%, 2.6 million hectares) under cereal cultivation and third (after maize and wheat) in terms of grain production (18.57%, 29.9 million quintals) in Ethiopia (CSA, 2008). Due to its nutrition value, there is an increasing concern in teff grain utilization. For instance, the protein is essentially free of gluten. Gluten is a protein found in wheat, rye, barley and some lesser known grains. Generally, speaking the advantages of using gluten free diet translates to better health. However, people with celiac disease or allergies find the benefits of a gluten free diet to be life sustaining. Therefore, about 66% of Ethiopian nutrition covers of injera and it accounts for about two-third of the daily

Mulaw and Tesfaye 827

Figure 1. Flow diagram for the preparation of injera (Gebrekidan and Gebrettiwat, 1982).

protein consumption of the Ethiopian population (Arogundade, 2006).

The preparation of teff injera comprises of two stages of natural fermentation, which last for about 1 to 3 days depending on ambient temperatures. The method of making injera from its raw materials to the final product involves mixing the ingredients (teff flour and water) to dough, which is fermented and subsequently thinned to a batter. Then, the batter is poured and cooked onto a hot griddle to develop color, flavor and texture. The main quality attribute of a good injera is its somewhat sour taste due to low pH nature of injera. The storage period of injera does not exceed three days at ambient temperature (temperature in the highlands of Ethiopia is between 17 and 25°C). It is a common practice to discard moldy injera. However, in times of food scarcity, moldy injera is sun dried and prepared for consumption (Gashe, 1985).

The teff injera was prepared at household by mixing teff flour with clean water in the ratio 1:2 (w/w) and 16% of starter (ersho) and was kneaded by hand in a bowl in the traditional way. Then the resultant dough was allowed to ferment for 3 days at ambient temperature. And then, the external water formed on the upper of the

dough was discarded. The main dough was thinned by adding water equal to the original weight of the flour and stirred for 15 min. The batter was left covered for 2 h for secondary fermentation. The batter was left for about 30 min to rise (the second fermentation), before baking commenced. Then some more water was added to thin down and form the right batter consistency. Finally, about half a liter of batter was poured onto the hot clay griddle in a circular motion from the outside, working towards the centre. After 2 to 3 min of cooking using traditional baking equipment (metad), the injera was removed and stored in a traditional basket container messob (Figure 1).

According to Askal and Kebede (2013) report, a total of 34 samples from injera batter were collected during 96 h fermentation at 6 h intervals. The teff sample was bought from Hawassa open market. A total of 107 LAB and 68 yeast strains were isolated and identified. The LAB strains were identified as Pediococcus pentosaceus (49.53%), Lactobacillus fermentum (28.04%), Lactococcus piscium (5.61%), Lactococcus plantarum (4.67%), Pediococcus acidilactici (3.74%), Leuconostoc mesenteriodes subsp. mesenteriodes (2.80%), Lactococcus raffinolactis (2.80%), L.

Teff flour

Mix with water and starter

Knead to form dough

Allow to ferment for 3 days

(Primary fermentation)

Discard the surface water

Add water to the remaining dough and allow standing for 1 h

(Secondary fermentation)

Add water

Bake on hot greased clay griddle metal till holes begin to form on top

INJERA

828 Afr. J. Microbiol. Res. mesenteriodes subsp. dextranicum (1.87%), Enterococcus casseliflavus (0.93%), and the yeast strains comprised Saccharomyces cerevisiae (48.53%), Candida humilis (22.06%), Candida tropicalis (17.65%), Saccharomyces exiguus (7.35%) and Pichia norvegensis (4.4%). Kocho fermentation Enset (Ensete ventricosum) is considered as large soft tissue plant, banana-like plant that grows up to 11 m high to the tip of the leaves (Mehtzun and Yewelsew, 1994; Admasu, 2002). Enset placed as cultivated main food crop found in south and southwestern Ethiopia (Berhanu, 1987; Almaz, 2001; Admasu, 2002). Records revealed that enset has been grown in Ethiopia for a long history (Jacob, 2004). Therefore, Enset is used for different purposes such as human food, livestock feed, industrial fiber, as rob material in fences and house-building, for mattresses and seats making, as local packaging material, and as substitute for table plates or umbrellas (Mehtzun and Yewelsew, 1994; Bizuayehu and Peter, 2003).

The plant is drought tolerant and accessible throughout the year thereby, it has the capability to serve for more number of people in the future as staple foods. Currently, the Ethiopian government has started a project on enset plant adaptation to non-enset growing regions (such as Tigray and Amhara) of the country. As a consequence, enset has been increasingly growing in this non-enset region using another food security alleviating food source. In line with this, enset has a potential to use as alternative to alleviate the food security problem throughout the world. Nevertheless, the information on the microorganisms involved and the biochemical changes occurred during the fermentation processes are not well studied.

It is clear that different studies showed that, a number of people hang on this plant for their livelihood. Coming to Ethiopia, kocho is the fermented product of enset, which is the major food product, obtained by fermenting the mixture of the scraped pulp of the pseudo stem, pulverized corm and stalks of inflorescence. And also the other food types obtained from enset are: bulla (extracted juice from edible part of enset) and amicho (non-fermented corm consumed after boiling). As scientific information reported that, the fermentation process of enset is carried out by different microbes. Thus, even though LAB are the dominant microorganisms involved during the fermentation of enset for kocho production, kocho contains a diverse group of microorganisms like aerobic and anaerobic spore-formers, Enterobacteriaceae, and yeast (Berhanu, 1987).

The corms of mature enset plants were used as main

raw material for the preparation of starter culture. The unwanted parts of the corm were removed with knife and then make it ready for fermentation. All the prepared corms were wrapped with fresh enset leaves near the farm site and left at ambient temperature for about 8 days. At the 5th day, it was exposed to the sun for 5 to 12 h and again wrapped with fresh enset leaves and allowed to further ferment for 3 to 5 days. Enset has two traditional processing phases such as phase I (surface fermentation), beginning of fermentation and continued for about 15 days and phase II (pit fermentation), completion of fermentation for about 15 additional days. After comprehensive fermentation of starter culture (gamancho), and ready to use, the selected enset plants were cut and processed in to fresh dough (Tiruha et al., 2014). Hence, the fermented starter was sliced and carefully mixed with fresh enset dough and distributed into treatments. The treatments were: ( A) Traditional kocho fermentation in pit, ( B) Fermenting mass in bucket with starter culture (not buried), ( C) Fermenting mass in bucket without starter culture (not buried), (D) Fermenting mass with starter culture in bucket buried in pit, and (E) Fermenting mass without starter culture in bucket buried in pit (Figure 2).

Different studies indicated that the length of time necessary to complete kocho fermentation varied depending on the locality, environmental temperature and mass of the fermenting dough. Therefore, the majority agreed that 15 to 30 days are required for complete kocho fermentation. Thus, the fermented dough can be kept in a pit up to one year. Nevertheless, during fermentation of kocho, the value of pH was gradually decreased. In line with this, the number of microorganisms during fermentation of kocho was seen gradual increment. The decrease in pH and increase in titrable acidity during the entire kocho fermentation could be attributed to the activities of acid producing microorganisms mainly LAB and yeast. At the initial fermentation Enterobacteriaceae increased and thereafter counts of Enterobacteriaceae reached below detectable level (Tiruha et al., 2014). And also according to Negasi et al. (2017) study on in vitro characteristics of LAB isolated from Ethiopian traditional fermented shamita and kocho for their desirable properities as probiotics, the LAB like Lactobacillus, Leuconostoc, Pediococcus and Lactococcus were found in kocho. Moreover, that Lactobacillus isolates were the most frequently isolated groups from kocho samples followed by Leuconostoc. TRADITIONAL FERMENTED DAIRY PRODUCTS In a long history, fermented dairy products are made from domesticated milk animals. Milk is known as the most nutritious food due to its rich nutrient content. It is an excellent source of proteins, minerals (especially calcium and phosphorus) and vitamins. It is known that,

Mulaw and Tesfaye 829

Figure 2. Flow chart of traditional enset fermentation (Tiruha et al., 2014).

fermented milk products are widely spread throughout the world. Fermentation of milk is carried out by the activities of natural flora present in the food or added from the surroundings. Hence, the microbes mainly encountered on the dairy industry are LAB, AMB, coliforms, Enterobacteriaceae, yeasts, molds and viruses. Some bacteria such as LAB are useful on milk processing, causing milk to sour naturally, and leading to fermented products. These products were an important component of nutritional food in age. Alongside, natural milk can also contain pathogenic bacteria, such as Salmonella species, Mycobacterium tuberculosis, Listeria species, and Brucella species, and can thus transmit disease and produce poor yields of products (O’Connor, 1995).

Traditional dairy fermentation process is conducted natural ly without control l ing the fermentation processes. Due to this, fresh milk is left at room or ambient temperature to assist fermentation process. In rural areas, especially among the pastoralists, raw milk is mostly kept in properly smoked container which is used for killing any organism remaining after washing, improves smell of equipment, improved aroma of butter when used as hair dressing and improved taste. And fermented milk from a previous fermentation uses it as source of inoculums. In addition to this, LAB from the internal surface of the container can also serve as starter culture. Regarding the quality and taste of the

fermented product, the incubation temperature does have significant role in the lowlands. In Ethiopia, the main fermented milk products include ergo, ititu, ayib, kibe, arerra, etc. Ergo (Sour Milk) Ergo is a naturally processed native Ethiopian fermented milk product, which is usually prepared at house level. It is made by natural fermentation of milk under ambient temperature. As a result, the microbial load of fermented milk samples, including Ergo, could vary from sample to sample based on the microbial number and types of microorganisms in the original raw milk (Abdulkadir et al., 2011).

During ergo fermentation, LAB were the dominant all other microorganism, followed by yeasts and then molds. Almaz (2001) showed that ergo fermentation is carried out by the genera, Lactobacillus, Lactococcus, Leuconostoc, Enterococcus and Streptococcus (LAB). And also the same authors also revealed that Micrococcus species, coliforms and spore formers were also present in fairly high numbers during the first 12 to 14 h of fermentation. Their population decreased substantial thereafter, which implies an antimicrobial activity besides low pH in the fermented milk (Savadog et al., 2004). Misganaw and Teketay (2016)

Mature enset plant (Ensete venricosum)

Cutting and cleaning of edible

part

Psedostem corm

Decorticate pulverize

Mix

Add chopped starter and

mix Wrapped tightly with enset leaves and ferment at ambient temperature for 15 days

Preparation of

pit Put in to a pit to ferment for 15 days (Phase II)

Re-mix again fresh enset leaves every

5days

Fermented Kocho ready for baking

830 Afr. J. Microbiol. Res. demonstrated that there is a diversity of LAB (homofermentative and hetrofermentative) in raw cow's milk. The presence of LAB in milk and milk products enhances bioavailability of nutrients, act as a preservative and serve as source for beneficial lactic acid.

According to Anteneh et al. (2011a) report during fermentation and storage of ergo, at the end of fermentation at 72 h, the pure LAB cultures decreased the mean count of the target enteropathogens by 3 log units. In addition to this, the count of all target pathogens was also decreased almost by 2, 3 and 5 log units at 24, 48 and 72 h, respectively during fermentation by mixed LAB cultures. During storage of ready-to-consume ergo at ambient condition, the count of test organisms decreased by 3 to 4 log units at 24 h; and the test strains were totally excluded within 30 to 48 h. In contrast, during storage of ergo at refrigeration condition, the average count of the test pathogens was reduced by 3 to 4 log units at 72 h. The LAB strains survived at counts of log 8.0 cfu/ml or higher up to 72 h during ambient and refrigeration conditions. Moreover, the study results showed that the strains are possible candidates for the formulation of bioprotective starter cultures that can be employed for production of safe and potentially probiotic ergo.

In line with this, Abebe et al. (2013) revealed that two LAB isolate from ergo showed antibacterial activity against clinical and standard human pathogens. The antibacterial activities of these isolates were effectively compared to Ampicillin. Excellent amount of lactic acid and H2O2 was documented at 72 h incubation

time. Thus, fermented ergo was the potential source of LAB that can produce antibacterial agents for the treatment of human pathogens. This can give direction for finding antibacterial agent producing microorganisms from fermented drinks use for preservation of foods as well as prevention of health risk problems associated with food borne diseases like Listeria monocytogenes, Shigella species, Staphylococcus aureus, Staphylococcal enteritis, Streptococcus, Vibrio cholerae, Yersinia enterocolitica, Yersinia pseudotuberculosis, molds and noro virus (Abebe et al., 2013).

Kassahun (2013) revealed that washing of milk and milk products equipment using different kinds of plant materials are a normal practice. Eucalyptus globules and Cymbopogan martini are the major plant materials used by the majority of households in all locations. The major reason is to improve the taste and/or flavor of the milk products and/or to increase the shelf life. Traditional milk preparing techniques involve smoking of processing utensils using residues of Olea africana. This smoking practice is useful to keep better quality of ergo through its inhibitory effect on spoilage and pathogenic organisms. To control the harmful microbes, the effect of lower pH of ergo is more effective after 24 h of incubation. At this time, the ergo is considered

to be too sour for direct consumption since ergo coagulates within 24 h and preferably consumed at this time for its good flavor (Mogossie, 2006). Kibe (traditional butter) Traditionally, Ethiopian butter is the most shelf stable of all traditionally processed of fermented dairy foodstuffs and has always been made from sour milk. It is semi-solid at room temperature with white and sometimes yellowish color, depending on age. It has a typical diacetyl taste and flavor when fresh, but extended storing at room temperatures results in putridity and rancidity. Butter is important component of Ethiopian traditional fermented milk products (Gemechu et al., 2017). This traditional milk product processed and sold by women in every society. It has been used by women for hair dressing and also used as the source of diet both in rural and urban areas, and is also utilized by children of weaning age and the elderly.

The moisture content of the traditional dairy product, batter ranges from 20 to 43% as compared to the international commercial standard for butter of 16%. Putrefactive microorganisms cause spoilage of kibe when stored at room temperature in age. However, it is highly stable against microbial spoilage after 2% salt addition, low moisture and nitrogen ratio. But microbes having lipolytic activity are highly liable for disorders such as rancidity or loss of flavor. In general, the microbiological information on this product is not common in Ethiopia. However, there are some studies published on the microbiological quality of traditional butters from the country (Almaz, 2001).

According to Zelalem (2010) report the average total bacterial counts (TBC) ranged from 6.18 log cfu/g in butter samples collected from Selale area to 7.25 l o g cfu/g samples from Sululta. On other studies, the average TBC of 7.49 l o g cfu/g and the presence of high variability a m o n g samples depending on the sources were reported. Samples collected from open markets and rural producers, for instance, had higher counts as compared to that obtained from dairy farms and urban producers of southern Ethiopia. In addition to this, the TBC of fresh butter sampled from rural and public butter markets in Addis Ababa ranged from 8.27 to 4.7 log cfu/g of butter (O’Mahoney, 1998).

Arera (defatted sour milk) Arrera is another byproduct of ergo obtained after removal of kibe after churning. It has a similar color to e rgo, but its appearance slightly smoother, although thicker than fresh milk and basically contains the casein portion of milk. In contrast to other traditional dairy products, arrera has fewer calories. It contains

91.5% moisture, 3.1% protein, 1.4% fat, 3.4% carbohydrate, and 0.6% ash (EHNRI, 1997). And also arrera has a shorter shelf life compared to all other fermented milk products (only 24 to 48 h). It is consumed in all parts of the country where fermented milk is made and it serves as a beverage either plain or spiced. It is chosen by women for consumption as a side dish or as drink. Extra of the products are given to calves, lactating cows and dogs. However, it may indirectly serve as additional income for the women by its use as raw material for cottage cheese (ayib) manufacture, which may be sold in the market. Due to its relatively short shelf life and some traditional beliefs, arrera is not sold in the market for direct consumption (Almaz, 2001).

The average counts of total bacteria, Entero-bacteriaceae and coliforms were greater than 9, 4.7 and 4.2 l o g cfu/ml, respectively of arera sampled from Addis Ababa (Zelalem, 2010). Similar study showed that arera sampled from Wollayta area had total bacterial count of about 9 log cfu/ml (Rahel, 2008). The same author also reported coliform count of 4.86 log cfu/ml. Different species of bacteria were identified in arera samples collected during both dry and wet seasons, which include Klebsiella pneumoniae, Klebsiella oxytoca, Escherichia cloacae, Escherichia sakazakii, Escherichia coli and some species of Salmonella (Zelalem et al., 2007).

Ayib (Ethiopian cottage cheese)

Ayib, a traditional Ethiopian cottage cheese, is a popular milk product consumed by the several local groups of the country. It is prepared from sour milk after the butter is removed by churning. Churning of the sour milk is carried out by slowly shaking the contents of the pot until the butter is separated. The defatted milk is heated to about 50°C until a distinct curd forms. It is then allowed to cool slowly, and the curd is filtered through a muslin cloth. Ayib comprises 79% water, 14.7% protein, 1.8% fat, 0.9% ash and 3.1% soluble milk constituents (O’mahoney, 1988).

In a study on the microbiological quality of ayib (Mogessie, 1990), samples collected from an open market in Awassa had counts of mesophilic aerobic bacteria (AMB), yeasts and enterococci of 10

8, 10

7 and

107

cfu/g, respectively. Above 60% of the samples had psychotropic count of 10 log cfu/g and about 55% of the samples were positive for coliforms and fecal coliforms. The pH values of the samples varied between 3.3 and 4.6 with about 40% having pH lower than 3.7. During preparation of ayib, the high initial count of microbes in milk, which raises the fermentation process, is shown to fall by the combined action of cooking and low pH. The presence of high microbial load of ready-to-consume ayib is assumed to be introduced from plant

Mulaw and Tesfaye 831 parts used for packaging and imparting flavor, and from handlers, too. Its low pH value should also assist in maintaining the low count for a certain period of time (Anteneh et al., 2011b).

Further analysis of ayib micro flora revealed that bacterial and yeast counts did not relate with pH value of ayib samples (Mogessie, 1994). It is clear that ayib samples with pH greater 4.0 contained more bacterial groups than those with pH less than 4.0 (Anteneh et al., 2011b). The Gram-positive rods dominated the aerobic mesophilic bacterial flora, being the most abundant. Enterobacteriaceae and Pseudomonas species constituted the bulk of the Gram-negative rods. The count of LAB was around 10

6 cfu/g and L.

fermentum and L. plantarum dominated the flora. Though the low pH of ayib inhibits the growth of many food-borne pathogens, higher numbers of LAB and yeasts are not desirable in ayib. A considerably lower pH due to the activity of LAB may result in a too sour product with a low sensory quality.

Study done by Anteneh et al. (2011b) showing the incompatible effect of mixed lactic cultures against foodborne pathogens (Escherichia coli, Salmonella Typhimurium DT104 and S. aureus) were evaluated during preparation and storage of ayib. Ayib was prepared by cooking pasteurized milk product with the numerous mixed starter cultures. The test pathogens were separately inoculated in duplicates into 200 g of cooled ayib in sterile stomacher bags to give initial inoculum level of 6 log cfu/g mixed thoroughly and incubated at ambient condition. Separately, ayib samples was similarly processed and stored at refrigeration condition. Enumeration of pathogens was done at 24 h intervals for 9 days. When counts were <log1 cfu/g, enrichment followed by streaking on the nutrient agar plates were done to determine complete inhibition (Mekonnen and Mogosie, 2005).

In milks soured in the presence of mixed starter LAB cultures, the test organisms (TOS) were reduced by 2 and 4 log factors at 24 and 48 h, respectively. The pH of souring milk was about 4.0 at 48 h. In the control milks (milks inoculated with test pathogens in the absence of LAB), the test pathogens grew to 8.4 log cfu/g at 48 h. The mean pH value of control milks dropped from initial 6.44 to 5.43 (Table 1).

Ititu fermentation

Ititu is popular fermented camel milk consumed by pastoral communities of the Kereyu area of the Oromia Region in the eastern part of Ethiopia (Eyassu et al., 2012). The people prepare ititu during the rainy season (Almaz, 2001). Ititu has good nutritional quality and wait for about two months at ambient temperature (25 to 30°C). It is served as side dish with traditional thin-baked cereal chips or consumed as food or drink alone. And also it is considered as one of the special foods and

832 Afr. J. Microbiol. Res.

Table 1. The inhibitory effect of mixed LAB cultures on the test organisms during souring of milk (Anteneh et al., 2011a).

Test organisms

(TOS)-MLCs

Average values

0 h 24 h 48 h

pH Count (log/ml) pH Count (log/ml) pH Count (log/ml)

E. coli - MLCs 6.42 3.57 5.24 5.56 4.21 4.78

S. Typhimurium - MLCs 6.47 3.61 5.23 5.32 4.24 4.74

S. aureus - MLCs 6.47 3.53 5.47 5.65 4.20 5.30

SD1 ±0.01 ±0.17 ±0.31 ±0.19 ±0.14 ±0.36

TOs alone (control) 6.44 3.30 5.97 7.73 5.43 8.48

SD2 ±0.04 ±0.18 ±0.06 ±0.22 ±0.13 ±0.11

SD: Standard deviation.

Table 2. LAB species isolated from the traditional fermented camel milk ititu (Eyassu et al., 2012).

Species Number of isolates % of total isolates

Lactobacillus salivarius 47 32.3

Lactobacillus plantarum 13 8.9

Lactobacillus delbrueckii subsp. Bulgaricus 25 17

Lactococcus lactis subspecies cremoris 10 6.8

Lactococcus lactis subspecies lactis 26 17.8

Enterococcus faecalis 25 17

Total 146 100

served to much respected guests as well as to weaning-age children and elderly.

During the traditional fermentation of ititu, fresh milk is collected in a well smoked fermenting vessel called gorfa. Gorfa is woven from fibers of selected plants into a lidded container with a capacity up to three litters. The lid of the gorfa is treated with leaves of Ocimum basilicum for cleaning and imparting desirable flavor to the product (Kassaye et al., 1991). When the milk coagulates, whey is removed by wooden pipette and an additional volume of fresh milk is added. Eyassu et al. (2012) reported that genus Lactobacillus, genus Lactococcus and genus Entrococcus carried out the souring process of ititu. Thus, Lactobacillus species was the dominant genus and comprised of 58% of the total LAB isolates (Table 2).

Lactobacillus salivarius, L. plantarum, Lactobacillus delbrueckii subsp. bulgaricus, Lactococcus lactis subsp. lactis, and Enterococcus faecalis are the isolated LAB species from ititu. Hence, L. salivarius was a comparatively fast acid producer bringing the initial pH of the skim milk to the final value of 4.6 before 48 h of incubation followed by L. plantarum and L. delbrueckii subsp. bulgaricus.

According to Kassaye et al. (1991) report, ititu had an average pH of 3.65, titratable acidity (as lactic acid) of 1.92%, fat and protein content of 9.05 and 7.17%, respectively. Moreover, these values varied significantly

among samples, though. Ititu had improved contents of free and total amino acids when compared with fresh whole milk and was rich in amino acids such as glutamic acid, alanine, proline, leucine and serine. In a study on farm-made fermented milk in southern Ethiopia, Fekadu and Abrahamsen (1997) reported that ititu had 3.3 to 3.7% fat, 3.3 to 3.6% protein and 3.3 to 3.5% lactose. Kassaye et al. (1991) further reported that the total bacterial count was 10

12 cfu/g, mainly dominated

by LAB.

Dhanaan fermentation Dhanaan is naturally fermented sour milk produced by pastoralists in the areas of Shinile and Jigjiga zones of eastern part of Ethiopia (Seifu, 2007). It has well nutritional quality and stay up to five months. Dhanaan is made by placing fresh camel milk in a smoked container, packaging the container with a piece of cloth and keeping it in ambient temperature place for about 12 to 24 h to allow spontaneous fermentation. Similar products from camel milk were reported from Kenya, Somalia and Sudan. Naturally fermented camel milk products, namely susac and shubat are produced in Kenya, Somalia and Sudan. Similarly, fermented sour milk called gariss is prepared from camel milk in Sudan by placing raw camel milk in a skin bag hitched to the

saddle of a camel that is allowed to go about its business (Abdulkadir et al., 2011).

In Ethiopia, no study was conducted about the bacterial characteristics, manufacturing protocols and potentials of the fermented camel milk product, dhanaan. Pastoralists prepare dhanaan from camel milk because they consider that it has high nutritional value and long shelf life, it enables collection of milk over a few days and thus facilitates delivery of milk to the market, it eliminates seasonal surpluses of milk, its taste is liked by the consumers, it has high demand in the market especially by urban dwellers, and it reduces thirst.

Dhanaan is made by natural fermentation without adding a starter culture. However, some of the producers mentioned that when a small amount of previously fermented milk is added as a starter into fresh camel milk it takes only 6 h to obtain dhanaan (Seifu, 2007). Kenyan researchers reported that the quality of susac improved using selected mesophilic lactic starter cultures rather than spontaneous fermentation; the resulting fermented milk had a uniform taste and a longer shelf life (Farah et al., 1990; Lore et al., 2005). Screening of microorganisms that is responsible for the fermentation and production of the dhanaan would help to develop a commercial starter culture and to standardize the manufacturing method for this product in the future. The producers also mentioned that during making dhanaan, the milk in the container should be kept closed. This suggests that the microorganisms responsible for souring or fermentation of camel milk are probably thermophilic anaerobic types. TRADITIONAL FERMENTED BEVERAGES Tella fermentation Tella is popular Ethiopian traditional beverages, which is made from diverse ingredients. It is, by far, the most commonly consumed alcoholic beverage in Ethiopia. It is assumed that over two million hectoliters of tella to be brewed annually in households and drinking houses in Addis Ababa alone (Shale and Gashe, 1991). Some of them consider as local beer. It is traditionally drunk on major religious festivals, saint’s days and weddings. Depending on the type of cereal ingredients used to make, tella has different names: Amhara tella, Oromo tella, and Gurage tella (Fite et al., 1991). Amhara tella has gesho (Rhamnus prinoides) and concentrated. Gurage tella is delicately aromatized with a variety of spices. Oromo tella has no gesho (R. prinoides), and it is thick and sweet (Vogel and Gobezie, 1983). Generally, tella is brewing from substrates such as barley, wheat, maize, millet, sorghum, teff or other cereals. The quality of tella is variable from local to local, from individual to individual. Even within the same individual, the quality is variable from time to time.

Mulaw and Tesfaye 833

Therefore, the way of making tella varies among the ethnic groups and depends on traditional and the economic situation. The clay container (insera) is washed with grawa (Vernonia amygdalina) and water numerous times and then smoked with wood from weyra (Olea europaea subsp. cuspidate) for about 10 min, in order to get it as clean as possible. Germinated grain of barley, or corn, or wheat (bikil), bought in the local market or prepared at home, are dried and milled. For making bikil, the grains are moistened in water and the moist grains are placed between fresh leaves, left to germinate for 3 days and after that dried. Gesho (R. prinoides), local hops, is available dried in the local market. The leaves of gesho are separated from the stem and dried again in the sun for about ½ h and then pounded. The ground gesho leaves are placed in a clay container with water and left to ferment for 2 to 3 days. Gesho is responsible for the bitter taste of tella. It is also thought to be the source of various chemicals (Sahle and Gashe, 1991; Kebede, 1994). It is assumed that gesho maintains acidic pH during tella fermentation so as to modify the nature of the mash and impedes the growth of unwanted microorganism (Kebede, 1994).

Some of the grains intended for tella preparation are toasted and milled, and then mixed with water and baked on the mitad to prepare what is known as kita (a thin, 5 to 10 mm thick, pancake- like bread). This kita, broken into small pieces, part of the milled bikil and the pounded gesho stems are added to the water and allowed to ferment for 1 to 2 days. The rest of the flour is toasted on mitad, sprinkled with water and toasted until dark brown to form what is known as enkuro. This mixture of enkuro, the rest of the germinated grains (bikil), some gesho, and water are added to the container. The mixture is kept covered overnight, after which more water is added and the container is kept sealed for 5 to 7 days, until when the beverage is ready. Tella can be kept for 10 to 12 days (Figure 3).

According to Shale and Gashe (1991), who made a detailed study of tella fermentation, there are numerous recipes for preparing tella and it appears as if every housewife has her own version of the recipe. The fermenting organisms of tella are composed of S. cerevisiae and Lactobacillus spp. Increase in ethanol content (2.2 to 5% (v/v)) is directly associated with growth in the population of yeasts and decrease in reducing sugar and total carbohydrate. The pH of tella is in the range of 4.5 to 4.8 (Debebe, 2006).

For tella considered to be a good quality, the final ethanol content is in the range of 2 to 8% (v/v) and pH is 4 to 5. The biochemical changes, the microorganisms involved in the fermentation and those which bring about necessary and unwanted changes in the process of tella making are described (Shale and Gashe, 1991). According to the report, the fermentation process of tella is divided into four phases. The first occurs in the original

834 Afr. J. Microbiol. Res.

Figure 3. Flow chart of traditional preparation of tella.

mixtures of ingredients, and the second and third phases occur after successive additions of more carbohydrate materials. The three main carbohydrate materials are mentioned to be bikil, kitta and enkuro. The latter phase is where acidification takes place, which is actually not desirable. Maximum ethanol production occurs during the third phase and at the beginning of the fourth phase.

Shale and Gashe (1991) reported that the extent of heat treatment the asharo (roasted barley) receives and the degree of steaming the enkuro (roasted barley steamed after grinding) is subjected to have the direct bearing on the color of tella, which is determined by the housewife preparing the tella. Tella is actually a beverage of variable viscosity and having a variety of colors (grayish-white to dark brown). Several samples of tella and other traditional alcoholic beverages collected from three regions of Ethiopia (Gojam, North Shoa, and Addis Ababa) were analyzed for their ethanol, methanol, and fuel oil contents by Fite et al. (1991). The mean values for methanol, fuel oil, and ethanol were found to be 35 ppm, 66 ppm, and 3.6%, respectively.

According Abegaz and Kebede (1995) report there were no microbes at the end of tella fermentation phase, especially in tella made with gesho. This is because of the synergic effect of both R. prinoides antibacterial substance, high alcohol concentration,

reduction of pH as fermentation time increases and reduction of nutrient content of tella. Moreover, yeasts were dominated at the middle of tella fermentation phases but at the last phase, they were dramatically reduced. Gesho and bikil (malt) were main sources of yeasts and bikil was the major source of LAB. Acetic acid bacteria were not detected from any ingredient; similarly Enterobacteriaceae and yeasts were not detected in ashero and kita. Therefore, the counts of Lactobacillus, Lactococcus, yeasts and AMB showed increment during the first two phases in both fermenters but gradually reduced at phase IV in both fermenters. The counts of Enterobacteriaceae were high at day zero and not detected at phase II in both fermenters. Acetic acid bacteria were detected at the beginning of phase II in traditional fermenter but at phase III in modified fermenter. In line with this, Belay and Wolde (2014) indicated that based on the nature of the ingredients of tella, the distribution of the microbial community is variable (Table 3).

Tej fermentation

Tej is a honey wine with alcohol content varying from 8 to 14% ABV, which is made from honey, water and leaves of gesho. Previously, upper class were used, but now it is widespread among all social groups, consumed on holidays and at weddings as well as served in hotels and bars across the country. It is a home-based as well as commercially available honey wine. So tej is mainly used for great feasts, such as weddings and the breaking of fasting. Sometimes, widely for commercial purposes, mixture of honey and sugar could be used for its preparation. In cases where sugar is used as part of the substrate, natural food coloring is added so that the beverage attains a yellow color similar to that made from honey (Fite et al., 1991).

Tej fermentation, like other traditional beverages of Ethiopia, is a natural fermentation and no starter culture or other modern techniques are used. So, the fermentation depends upon the microorganisms present in the environment. Thus, to determine the major source of the yeast cells in tej attention was given to honey and gesho. The dominant yeast, S. cerevisiae, counts ranged from 10

2 to 10

3 cfu/g in gesho samples

while 0 to 102

cfu/g in honey samples. And gesho was considered the major source of the dominant yeast in tej because it contained greater number of the yeast than honey.

According to Vogel and Gobezie (1983), during the preparation of tej, the fermentation pot is seasoned by smoking over smoldering R. prinoides stems and olive wood. One part of honey mixed with 2 to 5% (v/v) parts of water is placed in the pot, covered with a cloth for 2 to 3 days to ferment after which wax and top scum is removed. Some portion of the must is boiled with washed and peeled R. prinoides and put back to the

Sleeping Barley

Germination or Malting

Kilning

Milling

Mash mixer

Fermented tella

M u l a w a n d T e s f a y e 8 3 5 Table 3. The distribution of Microflora of the major ingredients of tella (Belay and Wolde, 2014).

Ingredient Enterobacteriaceae Yeasts Lactobacillus Lactococcus Acetic acid

bacteria Aerobic mesophilic

counts

Gesho 5.86 ± 0.01 5.55± 0.21 0.00 ± 0.00 0.00 ± 0.00 0.00± 0.00 5.85 ± 0.91

Bikil 5.88 ± 0.11 5.77 ± 0.10 5.52 ± 0.32 5.79± 0.00 0.00 ± 0.00 5.97 ± 0.63

Kita 0.00 ± 0.00 0.00 ± 0.00 0.00 ± 0.00 0.00 ± 0.00 0.00 ± 0.00 4.8 ± 0.12

Ashero 0.00 ± 0.00 0.00± 0.00 0.00 ± 0.00 0.00 ± 0.00 0.00 ± 0.00 4.83 ± 0.34

Figure 4. Flow chart of traditional preparation of tej (Vogel and Gobezie, 1983).

fermenting must. The pot is covered and fermented continuously for another 5 days, in warmer weathers, or for 15 to 20 days, in colder cases. The mixture is stirred daily and finally filtered through cloth to remove sediment and R. prinoides. Good quality tej is yellow, sweet, effervescent and cloudy due to the content of yeasts. The flavor of tej depends upon the part of the country where the bees have collected the nectar and the climate (Figure 4).

According to Bekele et al. (2006) a total of 200 samples of tej, an indigenous Ethiopian honey wine, was collected from ten production units at different production times. The pH values of samples varied between 3.07 and 4.90 and 77% of the samples had pH values <4.0. Therefore, change in pH value among all samples was significant (p<0.01).The variety for titratable acidity

was 0.1 to 1.03 g/l00 ml and mean values for the different production units were 0.34 to 0.6 g/l00 ml. About 65% of the samples had titratable acidity values of 4 g/l00ml and variations within samples of production units (CV>10%) or among all samples (p<0.01) were significant. Mean total alcohol content for the various production units was 6.98-10.9%. About 58% of the samples had alcohol content of 5 to 10%.

In line with this, the antagonistic activities of LAB studied by Abebe et al. (2013), a total of 18 ergo and tej samples were collected from Gondar town. During fermentation of ergo and tej, lactic acid and H2O2 were

able to hinder the growth of all clinical and standard human pathogens. The antibacterial activity of these was effective compared to ampicillin. Therefore, according to their report, much amount of lactic acid

Rinse fermentation pot

Smoke the pot using olive wood and hop stems

Mix 1 part honey with 3 parts of water and place in pot

Cover the pot and keep in warm place for 2 to 3 days to ferment

Remove wax

Boil washed, peeled hops in a portion of fermenting honey

Return boiled hops to fermenting honey

Cover pot and ferment another 8 days

Stir daily

Filter 3 times through cloth

Tej

8 3 6 A f r . J . M ic ro b i o l . R e s .

Table 4. Determination of lactic acid and H2O2 at 37°C for different incubation hours (Abebe et

al., 2013).

Isolates Incubation hours Measurements in mg

Lactic acid Hydrogen peroxide

I2

24 810.72 1.93

48 873.78 2.14

72 1396.24 1396.24

I3

24 720.64 3.10

48 747.66 3.32

72 810.72 5.35

and H2O2 were recorded at 72 h incubation time. Thus,

isolation and screening of LAB from potential fermented drinks are the sources of antibacterial agents for the treatment of human pathogens (Table 4).

Borde fermentation

Borde is a local beer mostly consumed by people in southwestern parts of the country. It is considered as a drink for people in the lower socio-economic status. Borde is prepared by women from fermented maize, sorghum, barley, or a mixture of the three. Borde can be very thick and serve as a substitute for meals during long trips. According to the villagers attitude borde is also used for medical and ritual purposes. The users consider that it enhances lactation and mothers are encouraged to drink substantial amounts of it after giving birth (Kebede et al., 2002).

Borde is produced by natural fermentation of a diversity of locally available cereal ingredients. It is a gassy whitish-grey to brown colored beverage with thick consistency and sweet-sour taste. Fermentation of borde has four phases marked by the introduction of ingredients into the fermentation pot at different times. In phase I (primary fermentation), maize grits were mixed with water and left to ferment at ambient temperature in a clean insira for 48-72 h. A portion of the fermented grits from phase I (48 h) was roasted on a mitad into enkuro, a well-roasted granular mass. Fresh malt flour and water were carefully mixed by hand in a smoked insira into a pale brown thick mash. This mixture is called tinsis and it was left to ferment for 24 h. A second portion of the fermented grits from phase I (68 h) was slightly roasted into enkuro, carefully kneaded with mixed flour (wheat, finger millet and teff) and water, and then moulded into stiff dough balls. The dough balls were steamed into gafuma and then broken into pieces. Pieces of cooled gafuma were blended with the fermented tinsis and additional water in the same insira to a thick brown mash called difdif. The difif was then allowed to ferment for 18 h.

The last portion of fermented grits from phase I (72 h)

was added into a pan containing a boiling mixture of whole grains of sorghum and water, further boiled into a very thick porridge with continuous stirring and then cooled. The gelled porridge was added to the fermented difdif, along with a small amount of additional malt. After a thorough mixing, the thick brown mash was sieved through a wonnfit (about 1 mm pore size). The residues were then wet milled using traditional grinding stones and sieved 2 times. The filtrates were pooled and poured back into the same rinsed insira and the fermentation was then continued for 6 h (after the addition of porridge). Actively fermenting effervescent borde was then ready for consumption. The production of borde was repeated three times at room temperature (20- 23°C) and the results are average of the triplicates. In addition, preliminary experiments were carried out to compare the following: (1) the 48 to 72 h fermentation with 24 to 48 h at Phase I; (2) earthenware pot with plastic, metal and glass jars; (3) substitution of maize grits and sorghum grains with flour; and (4) roasting of enkuro with baking of kita (flat bread).

According to Kebede et al. (2002) report samples of borde from open markets at five localities in southern Ethiopia showed average aerobic mesophilic count (AMC), LAB and yeast counts of 9.9, 10.1, and 8.1 log cfu/g

respectively. Enterobacteriaceae were <1 to 3.5

log cfu/g. The pH was 3.92±0.14. During the traditional production of borde with its four phases, the proportions of ingredients and cooking temperature were measured. Development of pH, titratable acidity, microbial load and mash temperature were monitored at 6 h intervals. The initial pH of 6.01 fell to 3.84 at end of Phase I. However, the pH increased at the start of Phase II, III and IV fermentations due to addition of malt and/or unmalted cooked ingredients and then decreased to below pH 4.0 at the end of each phase. During Phase I, EB increased from 5.1 to 7.7 log cfu/g

at 24 h, but were not

detected after 48 h. AMC, LAB and yeasts increased from their initial 6.5, 5.3 and 4.5 respectively to 10.5, 10.6 and 7.5 log cfu/g

at end of Phase I. The AMC of

cooked ingredients were 4.6-4.9 log cfu/g, while Enterobacteriaceae, yeasts and LAB were not detected.

Mula w an d T es f a ye 8 37

Figure 5. Flow charts of traditional preparation of borde (Kebede et al., 2002).

After mixing the cooked ingredients and malt, the AMC, LAB and yeasts increased from 7.1, 6.3 and 5.4 atPhase II to 10.5, 10.5 and 8.6 log cfu/g

in borde,

respectively. Enterobacteriaceae decreased from 5.2 to <1 log cfu/g

at Phase II and were not detected in

borde. The major roles of Phase I, II, III and IV are production of an acidic fermented mass, bulk starter production, main and corrective fermentations, respectively (Figure 5). Shamita fermentation Shamita is another traditional beverage of Ethiopia, which is low in alcohol content, made by overnight fermentation of mainly roasted barley flour and, consumed as meal-replacement (Ketema et al., 1999). Shamita is a widely consumed beverage in different regions of Ethiopia. It has a thick consistency and most people who cannot afford a reasonable meal consume it as meal replacement. It is produced by fermenting roasted barely overnight. Malt is not

commonly used in shamita fermentation, although local shamita brewers in Addis Ababa use it frequently, and starch is the only principal fermentable carbohydrate.

The microbes liable for fermentation are mostly from back slopping using small amount of shamita from a previous fermentation as well as from the ingredients and equipment. Ready to consume shamita has a high microbial count made up of mostly LAB and yeast. These microorganisms make the product a good source of microbial protein. However, shamita has poor keeping quality because of these high numbers of live microorganisms and becomes too sour about four hours after being ready for consumption (Mogessie and Tetemke, 1995).

According to Anteneh et al. (2011c) study on antagonism of LAB against foodborne pathogens during fermentation and storage of borde and shamita, pure LAB cultures decreased in average the number of test pathogens by 4 log cycles at 24 h during fermentation shamita. And also the mixed LAB cultures decreased the number of pathogens by 5 log units after 24 h of fermentation shamita. Coming to storage of shamita at

8 3 8 A f r . J . M i c r o b i o l . R e s .

Table 5. Changes in counts (cfu/ml) of major microorganisms during shamita fermentation (Ketema et al., 1999).

Time pH

Lactobacilli

Streptococci Micrococci Staphylococci Bacillus

spp. Homo fermentative

Hetero fermentative

0 5.80 6.44×105 3.0×10

4 <×10

2 1.5×10

5 2.1×10

5 2.4×10

4

4 5.52 6.9×106 3.1×106 <×10

2 7.2×10

5 2.5×10

5 3.6×10

4

8 4.77 7.6×106 1.1×108 <×10

2 1.1×10

6 5.4×10

4 2.6×10

4

12 4.43 3.2×107 1.3×108 <×10

2 1.6×10

6 1.9×10

4 2.2×10

4

16 4.26 4.4×107 1.8×108 <×10

2 2.9×10

6 1.8×10

4 2.1×10

4

24 4.03 1.1×109 2.2×108 <×10

2 5.0×10

4 1.4×10

4 4.6×10

3

ambient temperature, the test pathogens were reduced by 4 log units at 12 h and totally eliminated at 24 h. Therefore, they strongly suggest that the isolates are possible candidates for the formulation of starter cultures that can be used to produce safe and bioprotective products.

According Mogessie and Tetemke ( 1995), report the pH of ready to consume shamita in Awassa town was reported to be 4.2 and the product had high microbial counts (10

6 to 10

7 cfu/ml) consisting mainly LAB and

yeasts. In a microbiological study of shamita fermentation, Ketema et al. (1999) reported that all ingredients and the clay jar rinse water had large numbers of aerobic mesophilic bacteria (>10

4 cfu/ml)

mainly consisting of Bacillus and Micrococcus spp. Barley malt contributed most of the LAB and yeasts, which were important to the fermentation. They dominated the fermentation flora reaching final counts of 10

9 and 10

7 cfu/ml, respectively (Table 5).

In line with this, according to Negasi et al. (2017) study on in vitro characteristics of lactic acid bacteria isolated from traditional fermented shamita and kocho for their desirable characteristics as probiotics, the genera Lactobacillus, Leuconostoc, Pediococcus and Lactococcus were present in shamita. And also Lactobacillus isolates were the most frequently isolated groups from shamita. Keribo fermentation It is known that traditional fermented foods and beverages are those traditionally fermented products based on the skills of the household occupants by indigenous knowledge systems and is produced from a variety of locally available cereal ingredients using traditional techniques by the people of that area themselves. Thus, among the various fermented beverages, keribo is traditional fermented beverage produced mainly from barley and sugar in different parts of the country, including Jimma zone. It constitutes a main part of the beverages being served on holidays, wedding ceremony and also as sources of income of

many households. The popularity of this traditional fermented beverage is more reflected among the religious groups and those do not like alcoholic drinks. It has poor keeping quality with distinct characteristic of the deteriorating beverage at the end of 48 h of fermentation (Kebede et al., 2002).

From the traditional Ethiopian fermented beverages, the fermentation processes and microbial dynamics during fermentation of tella (Samuel and Berhanu, 1991), borde (Ketema et al., 1998) and shamita (Ketema et al., 1999) are described. Moreover, the safety consideration of Ethiopian foods and beverages has shown the possibility of isolating some foodborne pathogens from some fermented products. However, there is no scientifically documented information both on the microbiology and safety of keribo preparation.

During keribo preparation, barley was first washed of broken kernels, chaff and extraneous materials. Then the deeply roasted barley is added to boiling water and continued boiling for 10 to 20 min at 65 to 70°C until the ungrounded grain seems to be dissolved. Finally, it is allowed to cool and sieved. To the filtrate, sugar and yeast were added, thereafter, the container is sealed (to create anaerobic conditions) and left to ferment overnight. The final product can then be poured into bottle for sale or for consumption the next day but before consumption, some amount of sugar was added to the fermented product until it become sweet. The preparation process was easy to follow and it took one day to get it fully fermented under optimum temperature (Figure 6).

According to Rashid (2013) report samples of keribo from open markets and households in Jimma zone showed average LAB, AMB, aerobic spore formers (ASF) and yeasts with mean counts of (log cfu m/l) 2.70 ± 2.07, 2.34 ± 2.37, 4.96 ± 2.80 and 2.01 ± 0.60, respectively. But the mean counts of Enterobacteriaceae, staphylococci and molds were below detectable levels. The early stage was dominated by AMB and ASF. However, the mean counts of LAB increased exponentially for the first 30 h and remain constant thereafter. L. mesenteroides, identified as the most dominant LAB, were found to be susceptible to

Mulaw and Tesfaye 839

Figure 6. Flow chart of traditional keribo fermentation (Abafita, 2013).

Table 6. Microbial count (log cfu m/l) of different microbial groups detected in keribo (Rashid, 2013).

Microbial group Mean±SD CV (%) Minimum Maximum

LAB 2.70±2.07 76.66 0.0 6.89

AMB 2.34±2.37 101.28 0.08 8.31

ASF 4.96±2.80 56.45 0.0 7.97

Yeasts 2.01±0.60 29.85 0.81 3.10

EB ˂2 - - -

Staph ˂2 - - -

Molds ˂2 - - -

LAB: Lactic acid bacteria; AMB: aerobic mesophilic bacteria; ASF: aerobic spore former; Staph: staphylococci; EB: Enterobacteriaceae.

penicillin, gentamicin, ampicilin, chloramphenicol, amikacin, bacitracin and norfloxacin but resistant to vancomycin (Table 6).

Korefe fermentation

Korefe is the name of the traditional indigenous fermented beverage which is prepared in Begemder Province among the Koumant ethnic group. Dehusked barley is left in water overnight, and after that toasted and milled. And then mixed with water and dried gesho leaves, and fermented in a clay container for two to three months. When the beverage is needed, a small quantity of the mixture is taken, more water is added and after a day of fermentation, the beverage is ready for

Figure 7. Flow chart of traditional korefe fermentation.

consumption (Figure 7).

Dehusked barley is left in water overnight

Toasted and milled

Mixed with water and dried gesho leaves

Fermented in a clay container for two to three months

Korefe

840 Afr. J. Microbiol. Res.

Figure 8. Flow chart for the production of wakalim.

TRADITIONAL FERMENTED MEAT PRODUCTS Sausage fermentation Sausages are essential parts of foods in many regions of the world. Coming to our country, Ethiopia, Sausage production has only a recent history. Traditional sausage (wakalim) is the most popular fermented food product in Harari, eastern part of Ethiopia. Its preparation relies on natural fermentation with ingredients as the main source of inocula. The preparation has four-step that includes the preparation of a casing, mincing of meat, stuffing and fermentation. Sausage fermentations are characterized by the succession of microbial groups in the course of fermentation. Although several groups are involved in the initiation of fermentation, only those tolerant to acids and metabolites generated during fermentation survive and dominate the final microflora. In naturally fermenting beef sausage, raw meat yields LAB in low numbers. However, the lactic flora rapidly dominates the fermentation because of the anaerobic environment generated during fermentation (Hammes et al., 1995).

Wakalim is prepared using the following ingredients (g/kg) following traditional techniques: lean meat (700 g), mixed with fat (50 g), salt (20 g), onion (Allium ascalonicum) (160 g), red pepper (Capsicum annuum) (20 g) and 10 g each of Ethiopian cardamom (Aframomum corrorima), black cumin (Nigella sativa), Kemun (Trachyspermum capticum), Ethiopian mustard (Brassica nigra) and garlic (Allium sativum). The ingredients are mixed in a container of 5-kg capacity. About 200 to 250 g of the meat-spice mix is stuffed manually into a prewashed and dried animal casing cut in to 20 cm length, and allowed to ferment at (20 to 25°C) for six days (Figure 8).

Ketema Bacha et al. (2010) reported that wakalim fermentation was dominated by LAB and AMB including staphylococci and members of Enterobacteriaceae. Gram-negative bacteria were under detectable level after day 4 of fermentation. But Staphylococci were detected at low levels (around 4 log cfu/g) until the end of fermentation. Thus, LAB dominated the flora at the end of fermentation. Different species of Lactobacillus and Pediococcus commenced the fermentation and the lactic flora was finally dominated by L. plantarum and P. pentosaceus. The pH of the fermenting wakalim dropped from 5.5 ± 0.22 to 4.1 ± 0.19, while the titratable acidity increased from 0.09 to 0.6% in the course of fermentation. Moreover, moisture content of the fermenting wakalim dropped from 66.5% ± 2.12 to 22.0% ± 0.71 during the 6 days of fermentation.

Assaye and Mogosie (2014) indicated that the mean pH values of retail dry sausages ranged from 6.09 to 6.33 and moisture content values ranged from 35 to 41%. Mean microbial count values (log cfu/g) ranged from 4.87 to 5.18 for AMB, 2.02 to 2.50 for Enterobacteriaceae, 1.73 to 2.24 f o r coliforms, 2.46 to 3.04 f o r enterococci, 3.09 to 3.76 for staphylococci, 5.31 to 5.68 for LAB and 3.28 to 3.87 for yeasts. The aerobic mesophilic bacterial flora of retail sliced dry sausages was dominated by Gram-positive bacteria. Salmonella was isolated from two sausage samples. Spoilage of sliced dry sausages, after the vacuum package was opened, was detected within 3 to 4 days during aerobic storage at ambient temperature (22°C on average) and within 12 to 20 days at refrigeration storage (4°C). The storage conditions were intended to reflect what normally would happen in routine food handling in home kitchen environments and food service establishments. Generally, the majority of retail sliced dry sausages showed the presence of high microbial load, which indicated contamination during or after processing of the products (Table 7).

TRADITIONAL FERMENTED CONDIMENTS Awaze fermentation

It is known that, fermented food, beverage and condiment products are commonly produced throughout the world. Some fermented products produce strong flavor such that the product is not consumed alone, but is added as a condiment to make the food more tasty and enjoyable. In general, different countries of Africa protein-rich food ingredients are often fermented to make condiments. Siljo, awaze and data are among the traditional fermented condiments in Ethiopia and are consumed with other items on the basis of their desired aromas and flavors. Therefore, these condiments result from the microbial fermentations of vegetable-spice mixtures (Hesseltine, 1980).

Lean meat (700g) and fat (50g)

Grinding

Mixing with ingredients 250 to 260g

Stuffing into Merechi

Fermentation Smoking drying (6 days)

Wakalim (ca: 1000g)

Mulaw and Tesfaye 841

Table 7. Microbial counts of retail sliced pork dry sausages from three producers (Assaye and Mogosie, 2014).

Microbial groups Log cfu/g (Mean±SD

Processor 1 Processor 2 Processor 3

Aerobic mesophiles 5.43±1.21a 4.82±0.83

a 4.54±0.91

a

Enterobacteriaceae 3.44±1.46a 1.89±1.01

b 2.17±1.14

ab

Coliforms 3.33±1.45a 1.56±0.92

b 1.82±1.14

b

Enterococci 3.53±1.22a 2.37±1.26

b 1.78±1.09

ab

Staphylococci 3.72±2.18a 2.89±1.69

a 2.66±1.58

a

Lactic acid bacteria 6.19±1.89a 5.98±1.59

a 4.86±1.23

a

Yeast 4.37±0.59a 3.91±0.78

a 3.34±1.28

a

Means in rows followed by the same letters are not significantly different (P > 0.05). SD: Standard deviation.

The main substrates in awaze are red sweet pepper (C. annum), garlic (Allium ursinum) and ginger (Zingiber officinale) with which some proportions of different spices are added. Awaze is commonly known in the north and central Ethiopia and is often consumed with sliced raw or roasted meat and other traditional pancakes. While the microbiology and biochemical properties of several other traditional Ethiopian fermented foods and beverages have been studied (Gashe, 1985, 1987; Sahle and Gashe, 1991; Ashenafi and Mehari, 1995; Bacha et al., 1998), there are no reports on the fermentation of awaze, indigenous Ethiopian condiment.

However, a study on fermentation of awaze indicated that the aerobic mesophilic microflora of the ingredients of awaze was dominated by Bacillus spp. (1.1×10

6

cfu/g) and LAB (4.5×104 cfu/g). The counts of AMB

dropped during the fermentation period. LAB reached the maximum count of 5.9×10

9 cfu/g at day 4 and the count

remained >108 cfu/g throughout the fermentation. The

heterofermentative LAB dominated until day 3; thereafter, the homolactics dominated the fermentation. Yeasts appeared at day 6 and increased to 2.5×10

6 cfu/g.

Hence, fermentation of awaze was accompanied by declining pH and increasing titratable acidity. In addition to this, Salmonella Typhimurium was repressed during the fermentation within 48 h. But awaze had low initial contents of available protein and reducing sugars and did not show marked differences throughout the fermentation (Ahmed et al., 2001).

On the other hand, Asnake and Mogessie (2010) studied that LAB were enumerated and isolated from traditional fermented awaze. According them, a total of 87 LAB strains were isolated from awaze sample. Therefore, the isolates were grouped to different genera with their respective number: Lactobacillus (52), Leuconostoc (1), Pediococcus (27) and Lactococcus (7). In line with this, based on their glucose fermentation profile, the isolates were grouped as homofermentative and heterofermentative. However, the count of LAB for an awaze sample was (9.8 log

cfu/g). Fermentation of siljo Siljo is one of the traditional fermented condiments of Ethiopia made up of safflower (Carthamus tinctorius) extract and faba bean (Vicia faba) flour (Mogosie and Tetemke, 1995). The black mustard powder, which is added after cooking the mixture of the safflower and faba bean, helps as source of starter microorganisms (Mogosie a n d Tetemke, 1995; Zewdie et al., 1995). The fermented product has protein and fat content of 28 and 25%, respectively, with improved protein availability and concentration as a result of fermentation (Mogosie and Tetemke, 1995). The heating step in siljo may be essential in decreasing the level of contamination, but addition of plant materials, for flavoring purposes, to the heated gruel during the process of fermentation, the frequency of serving, and hygienic quality of handlers are factors that contribute to the exposure of siljo to pathogens. Siljo is consumed usually during the long fasting periods when people consume no fatty food of animal origin that may prevent the proliferation of the pathogens (Shin et al., 2002).

During the natural fermentations, the type of fermenting flora is determined by the initial flora of the ingredients. Thus, dif ferent workers have reported diverse microorganisms to be liable for the fermentation of siljo (Zewdie et al., 1995; Mogosie and Tetemke, 1995). In preparation of siljo, a volume of 1600 ml of siljo was made from safflower (Carthamus tinctorius), faba bean (Vicia faba) and black mustard powder. This was divided into 4 screw-capped bottles, each containing 400 ml of the gruel. The gruel was left to ferment naturally at ambient temperature. At around 32 h of fermentation, peeled garlic, ginger, Ethiopian caraway and rue leaves, about 2 g each, were added into each bottle and the fermentation was allowed to continue at ambient temperature.

842 Afr. J. Microbiol. Res. According to Eden and Mogessie (2005) report siljo

was made to ferment naturally and the count of LAB reached 9.9 log cfu/ml on day 5. The pH dropped from an initial value of 5.8–4.65 during this time. The lactic acid flora was dominated by Leuconostoc spp. At ambient temperature storage (18 to 22°C), the product spoiled on day 16. The spoilage was caused by Bacillus spp. At refrigerated storage (4°C), however, the count of Bacillus spp. was below detectable limits (<1 log cfu/ml) until the end of experiment on day 16. When Salmonella Typhimurium DT 104 was inoculated into the fermenting gruel at low initial levels, the count decreased steadily and the test strain was not detected by enrichment on day 5. At higher initial inoculum level (5.5 log cfu/ml), complete elimination was observed on day 7. In a non-fermenting control gruel, count of the test strain increased by about 3 log units on day 7. Datta fermentation There are many traditional condiments in different parts of the world produced by microbial fermentations. Such traditional condiments are used as taste enhancers in many traditional dishes. The majority of these fermentations are accompanied by certain biochemical changes of nutritional importance (Hesseltine and Wang, 1980). Datta is among the traditional fermented condiments mainly in the southern parts of Ethiopia and are consumed with other items on the basis of their desirable aromas and flavors. It is results from the microbial fermentations of vegetable-spice mixtures. But the major substrate in the making of datta is the small chili pepper (C. frutescenc) at its green stage. Datta was also prepared following traditional methods. The small green pepper together with its seeds was carefully washed and cut into pieces. Garlic and ginger, in small proportion, were peeled, washed and cut into small pieces. The pepper, garlic and ginger were mixed with small amounts of fresh sweet basil and seeds of rue. The mixed ingredients were manually wet-milled on a flat smooth traditional stone-mill into a greenish paste. It was then transferred into a 500 ml screw-cap bottle to ferment at ambient temperature (20 to 25°C).

According to Ahmed et al. (2001) study in datta fermentation, the count of AMB remained unchanged during the fermentation. LAB initiated the fermentation at a level of 7.1×10

4 cfu/g and reached 1.2×10

9 cfu/g at

day 7. The homolactic LAB started and dominated the fermentation for the first 2 days and the heterolactics took over thereafter. Datta fermentations were accompanied by declining pH and increasing titratable acidity. Salmonella Typhimurium was repressed during both fermentations within 48 h. Datta had low initial contents of available protein and reducing sugars and did not show marked differences throughout the fermentation.

CONCLUSION Fermentation is one of the best efficient techniques of producing and preserving foods. It is fairly a low-energy requiring conservation technology that improves shelf life of food products. In cases fermentation is important to obtain a certain food, the microorganisms present on the raw ingredients or in the containers spontaneously take care of the process. In most of these products the fermentation is spontaneous and involves different microorganisms.

Generally, clear understanding of the procedures involved in the process of making of the traditional fermented food and beverage products could help to design mechanism for production of an industrially based finished product so as to develop the keeping quality of the products. And it has the advantage of reducing wastage during processing, which is significant at household level. In line with this, eating fermented foods has a beneficial health effect for human beings as well as animals.

CONFLICT OF INTERESTS

The authors have not declared any conflict of interests.

REFERENCES

Abdulkadir B, Tsige K, Ketema B (2011). Antimicrobil susceptibility pattern of lactic acid bacteria isolated from ergo, a traditional Ethiopian milk, Jimma, southwest Ethiopia. Ethiop. J. Educ. Sci. 7(1):9-17.

Abebe B, Feleke M, Berhanu A (2013). Isolation and screening of antibiotic producing actinomycetes from soils in Gondar town, North West Ethiopia. Asian Pac. J. Trop. Dis. 3(5):375-381.

Abegaz B, Kebede T (1995). Geshoidin: A bitter principle of Rhamnus prinoides and other constituents of the leaves. Bull. Chem. Soc. Ethiop. 9:107-114.

Admasu T (2002). On indigenous production, genetic diversity and crop ecology of enset (Ensetventricosum). PhD Thesis, Wageningen University, The Netherlands.

Ahmed I, Tetemke M, Mogessie A (2001). Some microbiological and biochemical studies on the fermentation of awaze and datta, traditional Ethiopian condiments. Int. J. Food Sci. Nutr. 52:5-14.

Almaz N (2001). Diversity and conservation of enset (Enseteventricosum) and its relation to households' food and livelihood security in southwestern Ethiopia. PhD Thesis, Wageningen University, the Netherlands.

Anteneh T, Tetemke M, Mogessie A (2011a). Inhibition of some foodborne pathogens by pure and mixed LAB cultures during fermentation and storage of ergo, a traditional Ethiopian fermented milk. ARPN J. Agri. Biol. Sci. 6(4): 1990-6145.

Anteneh T, Tetemke M, Mogessie A (2011b). The inhibition of some foodborne pathogens by mixed lab cultures during preparation and storage of Ayib, a Traditional Ethiopian Cottage Cheese. World J. Dairy Food Sci. 6(1):61-66.

Anteneh T, Tetemke M, Mogessie A (2011c). Antagonism of lactic acid bacteria against foodborne pathogens during fermentation and storage of borde and shamita, traditional Ethiopian fermented beverages. Int. Food Res. J. 18(3):1189-1194.

Arogundade LA (2006). Functional characterization of teff (Eragostics tef) protein concentrate: Influence of altered chemical environment on its gelation, foaming, and water hydration properties. Food Hydrocol. 20:831-838.

Askal D, Kebede A (2013). Isolation, characterization and identification

of lactic acid bacteria and yeast involved in fermentation of Teff (EragrostisTef) Batter. Adv. Res. Biol. Sci. 1(3):36-44.

Asnake D, Mogessie A (2010). Evaluation of the Probiotic Properties and Antibiotic Resistance of Lactic Acid Bacteria Isolated from Awaze, Qotchqotcha and Tef dough, traditional Ethiopian fermented foods. J. Food Saf.12:187-191

Assaye H Mogessie (2014). Microbiological profile of retail sliced dry sausages in Ethiopia. Int. Food Res. J. 21(6):2473-2479.

Bekele B, Tetemke M, Mogessie A (2006). Yeast and lactic acid flora of 'tej', an indigenous Ethiopian honey wine: variations within and between production units. Food Microbiol. 23:277-282.

Belay B, Wolde A (2014). Fermenter Technology Modification Changes Microbiological and Physico-chemical Parameters, Improves Sensory Characteristics in the Fermentation of Tella: An Ethiopian Traditional Fermented Alcoholic Beverage. J. Food Process. Technol. 5(4):316.

Berhanu AG (1987). Kocho fermentation. J. Appl. Bacteriol. 62:473-477. Bizuayehu T, Peter L (2003). Diversity and distribution patterns of

enset land races in Sidama, Southern Ethiopia. Genet. Resour. Crop Evol. 50:359-371.

Christoph J, Leo M, Dasel WMK, Wambui KM, Peter L, Pierre R, Bernd K, Christophe L, Jan H, Jakob Z, Esther S, Gilbert F, Bassirou B (2017). African fermented dairy products–Overview of predominant technologically important microorganisms focusing on African Streptococcus infantarius variants and potential future applications for enhanced food safety and security. Int. J. Food Microbiol. 250:27-36

CSA-Central Statstical Agency of Ethiopia (2008). Agricultural Sample Survey: Report on Area and Production of Crops. Addis Ababa.

De Almad C, De Almada C, Martinez R, Sant’ana A (2015). Characterization of the intestinal microbiota and its interaction with probiotics and health impacts. Appl. Microbial. Technol. 99:4175-4199.

Debebe G (2006). Determination of ethanol level in beverages. MSc. Thesis, Department of chemistry, Addis Ababa University, Addis Ababa.

Eden E, Mogessie A (2005). Fate of Salmonella typhimurium DT 104 during the fermentation of Siljo, a traditional Ethiopian fermented legume condiment, and during product storage at ambient and refrigeration temperatures. World J. Microbiol. Biotechnol. 21:1259-1265.

EHNRI (1997). Food composition table for use in Ethiopia. Part III. Ethiopian Health and Nutrition Research Institute. Addis Ababa. pp: 34.

Eyassu AA, Mohammed YK, Zelalem Y (2012). Isolation and characterization of lactic acid bacteria from Ititu: Ethiopian traditional fermented camel milk. J. Sci. 5:82-98.

Farah Z, Streiff T, Bachmann MR (1990). Preparation and consumer acceptability tests of fermented camel milk in Kenya. J. Dairy Res. 57:281-283.

Fekadu B, Abrahamsen RK (1997). Farm-made fermented milk and cottage cheese in southern Ethiopia. J. Trop. Sci. 37:75-79.

Fite A, Tadesse A, Urga K, Seyoum E (1991). Methanol, fuel oil and ethanol contents of some Ethiopian traditional alcoholic beverages. SINET. Ethiop. J. Sci. 14:19-27.

Gashe BA (1985). Involvement of lactic acid bacteria in the fermentation of tef (Eragrostis tef), an Ethiopian fermented food. J. Food Sci. 50:800-801.

Gashe BA (1987). Kocho fermentation. J. Appl. Bacterial. 62:473-477. Gebrekidan B, Gebrehiwot B (1982). Sorghum injera: preparations and

quality. In. Proceedings of the International Symposium on Sorghum Grain Quality, 21-31 October 1981 Oct 28. Pp. 55-65.

Gemechu, Alganesh T, Yetenayet BT (2017). Traditional butter and ghee production, processing and handling in Ethiopia: A review. Afr. J. Food Sci. 11(4):95-105.

Hamad AM, Fields ML ( 1979). Evaluation of protein quality and available lysine of germinated and ungerminated cereals. J. Food Sci. 44:456-459.

Hammes WP, Bosch I, Wolf G (1995). Contribution of Staphylococcus carnosus and Staphylococcus piscifermentans to the fermentation of protein foods. J. Appl. Biol. 79:76S-83S.

M u l a w a n d T e s f a ye 8 4 3 Hesseltine CW, Wang WL (1980). The importance of traditional

fermented foods. BioScience 30:402-404. Jacob G (2004). Enset, the 'False Banana' as Food Security. Field

Exchange 22:3. http://fex.ennonline.net/22/enset.aspx. Jespersen L (2003). Occurrence and taxonomic characteristics of

strains of Saccharomyces cerevisiae predominant in African indigenous fermented foods and beverages. FEMS Yeast Res. 3:191-200.

Kassahun M (2013). Sources of milk products, milk and milk products handling preservation and spices added to the milk products in ada‟a woreda, Ethiopia. Int. J. Agric. Res. Rev. 3(1):6-12.

Kassaye T, Simpson BK, Smith JP, O'Connor CB (1991). Chemical and microbiological characteristics of 'Ititu'. Milchwissensch 46:649-653.

Kebede A, Fekadu B, Langsrud T, Narvhus J (2002). Parameters of processing and microbial changes during fermentation of borde, a traditional Ethiopian beverage. J. Food Technol. Afr. 7(3):85-92.

Kebede T (1994). The bitter constituent of Gesho (Rhamnus prinoides) leaves. MSc. Thesis, Department of Chemistry, Addis Ababa University, Addis Ababa.

Ketema B, Tetemke M, Mogosie A (1998). The microbial dynamics of borde fermentation, a traditional Ethiopian fermented beverage. SINET: Ethiop. J. Sci. 21:195-205.

Ketema B, Tetemke M, Mogosie A (1999). Microbiology of the fermentation of Shamita, a traditional Ethiopian fermented beverage. SINET: Ethiop. J. Sci. 22:113-126

Ketema B, Tetemke M, Mogosie A (2010). Microbial dynamics during the fermentation of Wakalim, a traditional Ethiopian fermented sausage. J. Food Qual. 33:370-390.

Khetarpaul N, Chauhan BM (1993). Sensory evaluation of some recipes prepared from fermented pearl millet flour. J. Dairy Foods Home Sci. 12:96-104.

Kimmons JE, Brown KH, Lartey A, Collison E, Mensah PP, Dewey KG (1999).The effects of fermentation and/or vacuum flask storage on the presence of coliforms in complementary foods prepared in Ghana. Int. J. Food Sci. Nutr. 50:195-201.

Lore TA, Mbugua SK, Wangoh J (2005). Enumeration and identification of microflora in Suusac, a Kenyan traditional fermented camel milk product. J. Food Sci. Technol. 38:125-130.

McKay LL, Baldwin KA (1990). Applications for biotechnology: present and future improvements in lactic acid bacteria. FEMS Microbiol. Rev. 87:3-14.

Mehtzun T, Yewelsew A (1994). Study of ensetprocessing and development of ensetprocessing tools in the southern region of Ethiopia (Monograph). ACA/ NORAGRIC research project. Awassa College of Agriculture, Addis Ababa University, Ethiopia. 66p.

Mekonnen T, Mogessie A (2005). Fate of Escherichia coli O157:H7 during the processing of Ergo and Ayib, traditional Ethiopian dairy products and effect of product storage temperatures on its survival. Int. J. Food Microbiol. 103:11-21.

Misganaw W, Teketay W (2016). Isolation and Identification of Lactic Acid Bacteria from Raw Cow Milk. Int. J. Adv. Res. Biol. Sci. 3(8):44-49.

Mogessie A (1990). Microbiological quality of 'Ayib', a traditional Ethiopian cottage cheese. Int. J. Food Microbiol. 10:263-268.

Mogosie A, Tetemke M (1995). Some microbiological and nutritional properties of Borde and Shamita, traditional Ethiopian fermented beverages. Ethiop. J. Health Dev. 9:105-110.

Mogossie A (2006). The microbiology of Ethiopian foods and beverages: A review. Ethiop. J. Biol. Sci. 5:189-245.

Negasi A, Fassil A, Asnake D (2017). In vitro evaluation of lactic acid bacteria isolated from traditional fermented Shamita and Kocho for their desirable characteristics as probiotics Afr. J. Biotechnol. 16(12):594-606.

O’Connor CB (1995). Rural Dairy Technology. International Livestock Research Institute. Addis Ababa, Ethiopia.

O’Mahoney F (1998). Rural Dairy Technology- Experience in Ethiopia. ILCA Manual No. 4. Dairy Technology Unit International Livestock Center For Africa. Addis Ababa, Ethiopia. 64p.

Padhye VM, Salunkhe DK (1979). Biochemical studies on black

8 4 4 A f r . J . M i c r o b i o l . R e s .

gram (Phaseolus mungo L.): III Fermentation of black gram rice blend and its influence the vitro digestibility of proteins. J. Food Biochem. 2: 27-34.

Rahel F (2008). Traditional and improved milk and milk products handling practices and compositional and microbial quality of raw milk and butter in Delbo, water shed of Wolyta zone MSc thesis Hawassa University, Ethiopia.

Rashid AA (2013). Microbiology of Keribo Fermentation: An Ethiopian Traditional Fermented Beverage. Pak. J. Biol. Sci. 16:1113-1121.

Samuel S, Berhanu A (1991). The microbiology of Tella fermentation. Sinet. Ethiop. J. Sci. 14:81-92.

Sanni AI (1993). The need for optimisation of African fermented foods and beverages. Int. J. Food Microbiol. 18:85-95.

Savadog A, Cheik AT, Ouattara H Bassole AS (2004). Antimicrobial activities of lactic acid bacteria streams isolated from Burkinafaso fermented milk. Pak. J. Nutr. 3(3):174-179.

Seifu E (2007). Handling, preservation and utilization of camel milk and camel milk products in Shinile and Jijiga Zones, eastern Ethiopia. Livest. Res. Rural Dev. 19:1-9.

Shale S, Gashe BA (1991). The microbiology of Tella fermentation. Sinet Ethiop. J. Sci. 14:81-92.

Shin R, Suzuki M, Morishita Y (2002). Influence of intestinal anaerobes and organic acids on the growth of enterohemorrhagic Escherichia coli O157:H7. J. Med. Microbiol. 51:201-206.

Simango C (1997). Potential use of traditional fermented foods for weaning in Zimbabwe. J. Soc. Sci. Med. 44:1065-1068.

Soni SK, Sandhu DK (1990). Indian fermented foods: microbiological and biochemical aspects. Indian J. Microbiol. 30:135-157.

Steinkraus KH ( 1998). Bio-enrichment: production of vitamins in fermented foods. In. Wood, B.J.B. (Ed.), Microbiology of fermented foods. Blackie Academic and Professional, London, pp. 603-619.

Tiruha H, Kebede A, Edessa N (2014). The microbiology of Kocho: An

Ethiopian Traditionally Fermented Food from Enset (Ensete ventricosum). Int. J. Sci. 8(1):7-13.

Vogel S, Gobezie A (1983). Ethiopian tej. In. Handbook of Indigenous Fermented Foods.Steinkraus, K.H. (ed). Marcel Dekker, Inc. New York. pp. 363-365.

Zelalem Y (2010). Microbial properties of Ethiopian marketed milk andmilk products and associated critical points of contamination: an epidemiological perspective, East Africa Dairy Development Program, Addis Ababa Ethiopia.

Zelalem Y, Faye B, Gerard L (2007). Occurrence and distributions of species enterobacteracea in selected Ethiopian traditional dairy products: A contribution to epidemiology. Food Control 18:1397-1404.

Zewdie S, Urga K, Nigatu A (1995). Microbiology of Siljo fermentation. SINET: Ethiop. J. Sci. 18:139-142.

Vol. 11(21), pp. 845-850, 7 June, 2017

DOI: 10.5897/AJMR2017.8569

Article Number: 02C4C1664609

ISSN 1996-0808

Copyright © 2017

Author(s) retain the copyright of this article

http://www.academicjournals.org/AJMR

African Journal of Microbiology Research

Full Length Research Paper

Phenotypic identification of Escherichia coli O157:H7 isolates from cattle at Suleja Abattoir, Nigeria

Mailafia, S.1*, Madubuike, S. A.1, Raji, M. A.2, Suleiman, M. M.3, Olabode, H.O. K.1, Echioda-Egbole, M.1 and Okoh, G. P. R.1

1Faculty of Veterinary Medicine, Department of Veterinary Microbiology, University of Abuja, Abuja, Nigeria.

2Department of Veterinary Microbiology, Faculty of Veterinary Medicine, University of Ilorin, Kwara State, Nigeria.

3Department of Veterinary Pharmacology and Toxicology, Faculty of Veterinary Medicine, Ahmadu Bello University,

Zaria, Nigeria.

Received 22 April, 2017; Accepted 19 May, 2017

Escherichia coli O157:H7 is a well-known pathogen of man and animals and a very low infection dose is needed to propagate the infection and clinical disease. In this study, a total of 515 rectal swab samples were collected from cattle and subjected to conventional biochemical tests. Presumptive identification on Eosine Methylene Blue (EMB) yielded an overall prevalence of 83.1%. Cefixime, Tallurite, Sorbitol MacConkey Agar (CT-SMAC) test yielded 18(4.2) isolates while Indole test, Methyl Red, Voges Proskauer and Citrate utilization test (IMVIC) biochemical test showed prevalence rate of 11(61.1%) and Microgen

TM test performed on the 18 Isolates yielded a prevalence rate of 33.3%. A total of 6 of the

isolates were subjected to latex agglutination test, in which 2(0.4%) were confirmed to be E. coli O157. The results of the somatic flagella antigen test performed on the 2 isolates revealed that 1(50%) belonged to E. coli O157:H7. Thus this study is therefore the first research work to confirm the presence of E. coli O157:H7 in cattle presented for slaughter in Suleja abattoir in Nigeria. Humans can contract the infection through exposure, handling and consumption of beef or animal products. Control measures are therefore necessary especially during processing and evisceration of beef, to ensure safety of cattle offal presented to the public for human consumption. Key words: Escherichia coli O157:H7, cattle, Microgen kit, flagella staining, phenotypic identification, cattle, latex agglutination test.

INTRODUCTION E. coli are commensals that inhabit the Gastro Intestinal Tract of (GIT) of healthy animals and man. This bacterium belongs to the family Enterobacteriaceae (Aboh et al., 2015). The organism is gram negative; rod

shaped with 2.0 µm long and 0.25-1.0 µm diameter can survive on a variety of substrates. It can utilize mixed acid fermentation in anaerobic condition, producing lactate, succinate, ethanol, acetate and carbon dioxide (CDC,

*Corresponding author: E-mail: [email protected]. Tel: +234(80)32922883.

Author(s) agree that this article remains permanently open access under the terms of the Creative Commons Attribution

License 4.0 International License

846 Afr. J. Microbiol. Res. 2012). The bacterium is classified into different serotypes based on presence of pathogenic flagella antigens. Chin (2000) identified six pathotypes of E. coli namely: Diarrhoeagenic E. coli (associated with diarrhoea), Shigatoxin producing E. coli (STEC) which shared homology to the cytotoxin produced by Shigella dysenteriae, verocytotoxin producing E. coli (VTEC), Enterohemorrhagic E. coli (EHEC), associated with food poisoning, Enterotoxigenic E. coli (ETEC), Enteropathogenic E. coli (ETEC) and diffusely adherent E. coli (DAEC). The 16S rRNA based phylogenetic analysis has shown close genetic relatedness of E. coli with other members of the Enterobacteriaceae (CDC, 2012).

Human infection with shigatoxin producing E. coli O157:H7 (STECH O157) is relatively rare but the consequences could be serious, especially in the immunocompromised such as the young and the elderly (Beyi, 2017). The outcome associated with STEC O157 infection includes: Diarrhea, intense abdominal pain, hemorrhagic colitis, hemorrhagic uremic syndrome (HUS), kidney failure and eventual death (David, 2015). The infection could be transmitted directly or indirectly through fecal-oral means with organism infecting its victim through braided skin, human or animal feces, contaminated food, water or soil. Out breaks has been associated with poor hygienic measures during slaughter, evisceration and processing of beef. The detection of E. coli O157:H7 is an indicator of fecal contamination and implies presence of other dangerous pathogens which can compromise the wellbeing of consumers (Biruhtesfa et al., 2017).

E. coli O157:H7 was widely distributed in North America, along with other serotypes such as STEC O145, O26, O111 and O103, but studies have inculcated the organism to be found in processed beef in Africa and some other parts of the world (CDC, 2012). Healthy ruminants especially feedlot cattle harbor this organisms in their lower gastrointestinal tract hence, constituting major reservoir of this organisms. Cattle therefore shed these organisms in their faces, thus, disseminating this deadly disease to the environment. Other well-known vectors for the transmission of E. coli O157O:H7 includes houseflies and formites (David, 2015). Some factors are responsible for the pathogenicity of these organisms includes: Season of the year because fecal shedding rates is well known to occur during the summer and early rainfall. Also the age of cattle has shown that there is less shedding of the organisms in cattle of slaughter age than younger cattle (Beyi, 2017). The somatic antigen of E. coli O157:H7 is known to produce potent toxins which are shiga-like in nature and their distribution is based on the seasonal variability especially in healthy cattle which serves as apparent reservoirs of these organisms (Tarr et al., 2005). The organisms could be discharged in meat, offal, milk and dairy products or contaminated water apple drinks, vegetables and bovine manure (Cobbold et

al., 2007).

The risk of transmission of E. coli O157:H7 to man and animals has increased overtime. The fact that low infections dose of the organism as low as ten could trigger serious infection is a signal for more researches to be conducted on this disease. This, coupled with the short incubation period of the bacterium could further exacerbate in the disease especially in the elderly and immune-compromised young individuals below five years of age (Weir and Hay, 2006).

The risk of E. coli O157:H7 from food animals has not been paid much attention in developing countries (Honise et al., 2017). There is also paucity of information regarding the epidemiology of E. coli O157:H7 in developing countries. Animals are commercially slaughtered and dressed in unhygienic conditions which compromise microbiological quality and safety of meat obtained from the animals. This consequently risks the health of the consumers. To the best of our knowledge, there are limited public health surveillance data which characterizes E. coli O157:H7 isolates from cattle presented for slaughter in Suleja abattoir, Nigeria. Our research is the first work to be conducted in the study area in order to phenotypically characterize and identify E. coli O157:H7 serotypes from rectal swabs of cattle presented for slaughter using standard bacteriological methods. This paper therefore serves as a catalyst for the need to the promotion of surveillance programs to identify sources of pathogenic E. coli from non-human origin.

METHODOLOGY

Study area

Suleja is located in Suleja local government of Niger state. The city is situated at latitude 09° 31N and longitude 07° 581 E and about 20 km north of Abuja, FCT, Nigeria. It is about 100 km north east of Minna, administrative headquarters of Niger state. Suleja has a population of 216, 570 (NPC, 2006). It has a sub-humid climate, mean annual rainfall of 1640 mm and a raining season of about 7 months in the year. Agriculture which involves farming of crops and rearing of animals such as sheep and goats for sale as the need arises and also human consumption especially during festive periods.

Experimental design, sample size and sampling

A cross sectional study was conducted based on convenience observation of a selected sample of individuals from a larger population. Thereafter, each individual sample was determined by simultaneous presence or absence of disease or a causative agent of a disease and hypothesized risk factor (Dahiru et al., 2008).

The sample size was determined as described by Thrusfield (1997):

N = zpQ/d2

Where: N = sample size, z = standard normal deviation for 95% confidence interval (1.96); P = prevalence (53%); D = desired precision (0.05).

Nevertheless, 515 samples were collected in this study in order to increase the probability to capture the effect and increase the power of the study. The samples were collected from the rectum using sterilized swab sticks of the cattle presented for slaughter. The samples were placed in separate labeled sample bottles containing 8 to 9 ml Modified Tryptone Soya Broth (MTSB), kept in a cool box containing ice block were then transported to the bacteriology Laboratory of the Department of Veterinary Microbiology of the Ahmadu Bello University, Zaria, where they were processed soon on arrival. Sampling lasted for six months (that is, between July and December, 2016).

E. coli isolation and identification of O157:H7 Serotypes

All samples and media were prepared based on standard bacteriological method as indicated by the manufacturer’s instructions (Oxoid, UK). The bacterial isolation, biochemical identification and confirmation of the isolated colonies were done according to Barrow and Feltham (1993). The rectal swabs were inoculated in pre-enriched on MTSB onto Eosin Methylene Blue (EMB) agar and subsequently incubated at 24 to 48 h at 37°C. The appearance of greenish metallic sheen colonies was suggestive of E. coli. A discrete colony was then picked with sterilized wire loop, and then sub cultured onto Sorbitol MacConkey Agar suplmented with Cefixime and Tellurite (CT-SMAC Oxoid,UK) and incubated at 37°C for 24 h (Cheesebrough, 2000). The colonies on CT-SMAC were considered presumptive E. coli O157 because O157 do not ferment sorbitol. The non-sorbitol fermenters were thereafter picked and inoculated onto nutrient agar slants, incubated at 37°C for 24 h and stored in the refrigerator for further biochemical tests.

Stocked isolates suspected to be E. coli were further verified using conventional biochemical tests as described by Harrigan (1988) on the basis of indole production and motility with SIM medium (Merch, Germany), citrate utilization with simmons citrate agar (Merck, Germany), methyl red and vogues-proskauer using MR-VP medium (Merk, Germany) and urease production using urea agar (Oxoid, UK).

MicrogenTM confirmation of E. coli isolates

Commercially available biochemical test strip (MicrogenTM ID, UK) was used to further confirm isolates suspected to be E. coli based on the results of the conventional biochemical tests (OIE, 2008). The isolates to be inoculated were grown on selective media (CT-SMAC Oxoid, UK) using a wire loop, 3 to 4 discrete colonies were emulsified in normal saline solution (0.85%) and adjusted to McFarland turbidity standard of 0.5. The wells of the individual substrates sets were exposed by cutting the end tag of the sealing strip and slowly peeling it back. Using a micropipette with sterile micropipette tips, 100 µl of the bacterial suspension was added to each well in the set. Mineral oil was then used to overlay the substrates in wells 1, 2 and 3 using a Pasteur pipette. The inoculated rows were then re-sealed and properly labeled at the end of the tag with the specimen identification number followed by incubation at 37°C for 24 h. On 24 h incubation period, the sealing tape on the test strips peeled back and evaluated. Results were then recorded on a report form as positive or negative by comparing them with a colour reference to the table of reactions provided. In addition to well 8 and (Indole production), two drops of Kovac’s reagent were added and the results evaluated within 2 min of adding the reagent. To well 10 (Vogues – proskauer reaction), 1

Mailafia et al. 847 drop each VP1 and Vp2 reagents were added and the results evaluated within 15 to 30 min, after addition. To well 12 (Tryptophan Deaminase), 1 drop of TDA reagent was added and the results were evaluated immediately.

In the interpretation of the results, the octal coding system was adopted with each group of the three reactions producing a single digit of the code. Using the results recorded on the report forms, the indices of the positive reactions are circled and the sum of these indices in each group of these three reactions formed a code of four numbers. This code obtained was then entered into the computer-aided identification package and the resulting organisms and its percentage probability was recorded.

Serotyping of E. coli O157:H7 (latex agglutination test)

The isolates that were confirmed as E. coli using the MicrogenTM were further serotyped using bacterial agglutination test using a commercial latex kit for E. coli 0157:H7 (Wellcolex E. coli O157:H7 Kit, Oxoid, UK) which is a rapid Latex Agglutination test for confirming non-sorbitol fermenting colonies and H7 somatic flagella antigen (OIE, 2008).

Detection of E. coli O157

Fresh cultures grown at 37°C for 24 h on sorbitol MacConkey agar (Oxoid, UK), were used for this test. 40 µl of saline was placed in two circles on a reaction card, using a micropipette. With the help of a mixing stick provided, sufficient growth just enough to cover the blunt end of the stick was removed from the plate and emulsified in the saline by rubbing with the flat end of the stick, mixing it thoroughly without damaging the surface of the card. This was done for the second circle on the reaction card already containing saline after which mixing sticks were immediately discarded. For each of the test sample, one drop of O157 test latex was placed in one circle (with the emulsified culture) and one drop of O157 control latex in the other circle by holding the dropper bottles vertically in order to dispense on accurate drop. A mixing stick was then used to mix the contents of the circles, spreading the latex carefully over the entire area of the circle. The card was then rocked slowly for 30 s and observed for agglutination. The used reaction card was then discarded for safe disposal. The agglutination of O157 test latex accompanied by a lack of agglutination of the control latex on observation was used to indicate the presence of O157 antigenic the culture under test. Absence of agglutination both O157 test control reagents was used to indicate the absence of O157 antigen in the test culture (OIE, 2008).

Detection of H7 (Flagella Staining)

Test cultures that were positive for O157 antigen were further tested for H7 antigen by sub culturing in Modified Tryptone Soya Broth (MTSB) Oxiod, UK at 37°C for 24 h. For each sample, 40 ul each of MTSB-grown cultures were placed in two circles on a reaction card, followed by placing a drop of the H7 test latex in one circle (with the broth culture) and a drop of H7control latex in the other circle. The contents of the circles were then thoroughly mixed using a mixing stick, carefully spreading the latex over the entire areas of the circle. The card was then slowly rocked for 30 s and observed for agglutination after which the card was discarded. Agglutination of the H7 test latex accompanied by a lack of agglutination of the control latex on observation was used to indicate the presence of H7 antigen while the absence of agglutination in both reagents was indicative of absence of H7 in the test culture (OIE, 2008).

(1.96) × 0.53 × (1-0.53) N = = 382 (0.05)2

848 Afr. J. Microbiol. Res. Data analysis The results were presented using simple descriptive statistics involving percentages, tables and charts (CDC, 2012).

RESULTS Various reactions to confirm E. coli O157 and E. coli O157:H7 are shown on Figures 1 to 3. Of the 515 rectal swab samples analysed, 428 (83.1%) were presumably positive for E. coli, with characteristic greenish metallic sheen. Sub culturing on cefixime, tellurite sobitol MacConkey (CT-SMAC) agar revealed 18 (4.2%) as non-sorbitol fermenting strains of E. coli O157:H7. However, 87 (20.3%) of the 515 samples did not yield any significant bacterial growth (Table 1). Table 2 showed the results of the biochemical tests (IMVIC), Microgen

TM,

latext agglutination test and flagellar staining of the 18 isolates tested for IMVIC, 11(61.1%) were identified as E. coli. The Microgen

TM GMA + B-ID system, UK showed

that from the 18 isolates tested, 6(33.33%) were positive for Microgen

TMtest. The somatic and flagella staining for 6

isolates yielded 2 (0.4%) of E. coli 0157 while from the 2 of the isolates tested for flagella staining, 1 (50%) yielded positive results.

Figure 1 shows the Microgen TM

test of the groups of triplicate reactions of E. coli isolates. Figure 2 shows that the 4

th circle which contains the positive results to Latex

Agglutination test confirming E. coli O157. Figure 3, on the other hand shows the positive results to flagella staining, with the first two circles dissipating the positive results of presence of E. coli O157:H7. DISCUSSION The presumptive prevalence of 428(83.1%) E. coli isolates signifies a high isolation rate of E. coli as indicated in our studies. The high prevalence rate could be associated with poor hygienic practices right from the abattoir, during handing or transportation of the carcasses to the market. E. coli is a normal commensal of gastrointestinal tract of cattle and man. This observation has been also been made in other animals species including sheep, goats, pigs, birds and non-human primates (Nuno et al., 2011; Rosa et al., 2017). The organism which was thought to be non-pathogenic for several years however has become highly adapted to cause diarrhoea, septicaemia, meningitis, urinary tract infection, wound infections in several animal species and humans. E. coli is well known to cause infection not only in immuno compromised patients but also in situations of impaired gastrointestinal barriers (Keskimaki et al., 1997). The isolation of E. coli from cattle in Suleja abattoir is therefore of public health significance.

The isolation rate of E. coli O157:H7 obtained from this study was found to be 0.4%. This may seem low but

Figure 1. Microgen TM test results showing four groups of triplicate reactions.

Figure 2. Circle 1 and 4 shows positive results to latex agglutination test to E. coli O157.

Figure 3. The first two circles shows positive results to flagella test to confirm E. coli O157:H7.

Mailafia et al. 849

Table 1. Presumptive Identification of E. coli isolates Using EMB and CT-SMAC tests.

Samples collected Isolates +ve on EMB Positive isolates on CT, NSF

100 91(91.0) 1(1.1)

100 89(89.0) 7(7.9)

100 98(98.0) 6(6.1)

100 97(97.0) 3(3.1)

115 53(46.1) 1(1.9)

Total 515 428(83.1) 18(4.2)

CT-SMAC = Cefixime, Tallurite, Sorbitol, MacConkey agar. EMB = Eosine Methylene Blue. NSF = Non-sorbitol

fermenters.

Table 2. Isolation rate of E coli O157 determined with Microgen(TM), latex agglutination and H7 flagella test.

Methods No. of isolates tested No. (%) positive

Conventional biochemical tests 18 11(61.1)

Microgen (TM)

18 6(33.3)

Latex agglutination (0157) 6 2(0.4)

Flagellar (H7) 2 1(50%)

The isolation rate 2 (0.4) was calculated from the total number of samples collected (515).

could present a more challenging effect especially in cases involving human exposure, since a very low dose of the pathogen is required to establish infection (Podolak et al., 2010). The low isolation rate of E. coli O157:H7 at 0.4% observed in our present study concurs to previous reports of Tutenel et al. (2002) with slight variation, who reported low isolation rate of 0.7 and 1.2% from rectal swabs of slaughtered cattle in Poland and Finland respectively. The low prevalence of E. coli O157:H7 in cattle suggests low infection rate in the cattle population studied. This finding can be extrapolated for a larger cattle population in Nigeria with caution. Although it is difficult to trace back the herds’ origin of the cattle presented for slaughter, it is reasonable to assume that the abattoir in Suleja presents a wide catchment area where animals are brought for slaughter across the country. This finding can be confounded by the fact that cattle could be brought from big or small herds; or even infected or uninfected herds. These calls for large scale studies to determine the prevalence and distribution of E. coli O157:H7 under different cattle production systems in Nigeria.

However, Elder et al. (2000) reported a higher isolation rate of 27.8% from rectal swabs of slaughtered cattle in USA; and Dahiru et al. (2008) reported a very high isolation rate of 53% from fresh beef in Kano state, Nigeria. On the other vein, the isolation rate of 2.8% for E. coli 0157: H7 was reported from human feces in Ile-Ife, Nigeria (Odetoyi et al., 2016). These reports have lent credence to the high and increasing endemicity of E. coli. O157:H7 in different geographical locations including abattoirs and slaughter houses globally. However, the

variations in the isolation rates may be attributed coupled with variations in sampling methods and isolation techniques used, as well as the distribution and seasonal variation of the organisms from one geographical location to another or within different countries.

The recto anal junction of cattle serves as the principal predilection site for predilection of E. coli O157:H7 (Rosa et al., 2017). Cattle have been known to be established as natural reservoir for the dissemination of E. coli O157:H7. Cattle play significant role in the epidemiology of human and animal infection, In the present study, the presence of E. coli O157:H7 in cattle presented for slaughter may suggest high possibility of unwholesome practices, leading to the contamination of carcasses by fecal materials during slaughter and processing of beef, which suggests that the current available processing procedures at the abattoir are not reliable to prevent fecal contamination during slaughter. It is therefore necessary that government should intensify regulatory efforts to curb this menace. Epedemiological studies have shown that cattle are healthy carriers of E. coli O157:H7 and that these organisms produce potent toxins which could be shed in feces of apparently healthy cattle (OIE, 2008). Therefore, the findings indicated in our studies are therefore very significant. Prevention and total eradication of this disease is important in improving cattle manage-ment practices and safeguarding the public. Conclusion Though, the study revealed low isolation rate of E. coli

850 Afr. J. Microbiol. Res. 0157:H7 at 0.4%, it also exposes virulent strains of these isolates. This therefore sends a warning signal to relevant regulatory authorities in Nigeria that the disease is very much around. E. coli O157:H7 may spread from cattle to humans. It therefore calls for rampant public health awareness campaign and development of relevant disease control strategies in Nigeria. CONFLICT OF INTERESTS The authors wish to declare that they do not have any conflict of interests. GRANT/SUPPORT This project is sponsored by Dr Stella Madubuike through Dr. Julius Ode and assisted by the Department of Veterinary Microbiology, Ahmadu Bello University, Zaria, Nigeria. Dr. S. Mailafia coordinated, supported and developed the manuscript. Prof M. A. Raji and Dr. M.M. Suleiman played supervisory roles. All the authors supported the research. REFERENCES Aboh EA, Giwa GJ, Giwa A(2015). Microbial Assessment of well water

in Samaru, Kaduna State. Ann. Afr. Med. 14(1):32-38. Barrow GI, Felthan RKA (1993). Cowan and Steel’s manual for the

Identification of Medical Bacteria 3rd ed.), Cambridge University

Press. 128-146:222-238. Beyi AF (2017). Prevalence and antimicrobial Susceptibility of E.coli

O157: H7in beef cattle at butcher shops and restaurants in Central Ethiopia. BMC Microbiol. 217:17-49.

Biruhtesfa A, Degmawi P, Mesele A, Genene T, Dereja H, Sufafel K, Kebede A (2017). Occurrence of E.coli O157: H7 in cattle feces and contaminated carcasses in various contact surfaces in Abbattoir and Butcher shops in Nawasse, Ethiopia. BMC Microbiol. 17:34-35.

Centres for Disease Control and Prevention (CDC) (2012). Human illness caused by E. coli 0157: H7 from food and non-food sources. CDC 27/7 saving lives and protecting people. pp. 3-4.

Cheesebrough M (2000). District Laboratory Practice in Tropical Countries. Part 2, Cambridge University Press. pp. 381-407.

Chin J (2000). Control of Communicable diseases manual, 17th Ed. An

official report of the American Public Health Association. Washington D.C. American Pub. Health Assoc. pp. 90-91.

Cobbold RN, Hancock DD, Rice, DH, Berg J, Stillborn R, Houdec CJ, Besser JE (2007). Recto-anal Junction colonisation of feed lot cattle by Escherichia coli0157: H7 and its association with super shedders and excretion dynamics. Appl. Environ. Microbiol. 73:1563-1568.

Dahiru M, Uriah N, Enabulele SA, Shamsudeen U (2008). Prevalence of

E. coli O157: H7 in fresh and roasted beef in Kano city, Nigeria, Bayero J. Pure Appl. Sci. 1(1):39-42.

David RS (2015). Prevalence of E. coli O157: H7 in beef cattle and slaughter beef carcasses in retail shops in Ethiopia. ASM, Washington DC. Micro. Bioshere. EHEC-0006-2013.

Elder RO, Keen JE, Siragusa GR, Barkocy-Gallagher, GA, Kooh M, Maraier U, lecgried WW (2000). Correlation of Enterohaemorrhagic E. coli 0157: H7 prevalence in faeces, hides and carcasses of beef cattle during processing: Proceedings of the beef cattle during processing. Proc. Natl. Acad. Sci. 97:2999-3003.

Honise L, Punja N, Num S, Nelson D, Hiislop N, Gosselin G, Stashko N, Ditrich C (2017). E.coli O157: H7 infections associated with contaminated pork products. Can. Commun. Dis. Rep. 43(1):21-24.

Keskimaki M, Ikäheimo R, Kärkkäinen P, Scheutz F, Puohiniemi RL, Siitonen A (1997). Shiga toxin producing Escherichia coli serotype OX3:H21 causing hemolytic uremic syndrome. Clin. Infect. Dis. 24:1278-1279.

NPC-National Population Commission (2006). Archived data on 2006 Nigerian Population Census. pp. 55-56.

Nuno S, Igrias G, Goncalves A, Poeta P (2011). Commensally gut bacteria: distribution of Enterococcus Species and prevalence of Escherichia coli phylogenetic groups in animals and humans in Portugal. Ann. Microbiol. 62(2):449-459.

Odetoyi BW, Hofmann J, Aboderin A, Okeke IN, (2016). Diarrheoagenic E.coli in mother and child pairs in Ile-Ife, south western Nigeria. BMC Infect. Dis. Ser. 16:28-30.

OIE-Office of International Epizootics (2008). Vero cytotoxigenic Escherichia coli. Terres. Man. Chapter 2.9.11. pp. 1295-1298.

Podolak I, Galanty, A, Sobolewska D (2010). Saponins as cytoxic agents: A review. Phytochem. Rev. 9(3):425-474.

Rosa A, Woynshet H, Akefete T, Ashenafi F, Getalum EA, Badeso m, Fanos T, Geloye K, Takele B, Teriku J, Bayene LDC, Eric C Bruno MG (2017). Prevalence of E. coli in beef cattle slaughtered in Ethiopia. BMC Infect. Dis. 17:21.

Tarr PT, Gordon CA, Chandler WL (2005). Shiga-toxin producing E. coli and haemolytic uremic syndrome. Lancet 365:1073-1086.

Thrusfield M (1997). Veterinary Epidemiology. 2nd

Edition, Blackwell Scientific Publications, Oxford. pp. 53-55.

Tutenel AV, Perard D, Uradzinski J, Jo-wiki E, Pastuszczak M, Van Hende J, Uyttendaele M. Debevere J. Cheasty T, Van Hoof J, De-zutter (2002). Isolation and characterization of enterohaemorrhagic Escherichia coli 0157:H7 from cattle in Belgium and Poland. Epidemiol. Infect. 129(1):41-47.

Weir E, Hay K (2006). E. coli—sporadic case or an outbreak?. Can. Med. Assoc. J. 174(12):1711.

Vol. 11(21), pp. 851-859, 7 June, 2017

DOI: 10.5897/AJMR2017.8543

Article Number: 1181D2564613

ISSN 1996-0808

Copyright © 2017

Author(s) retain the copyright of this article

http://www.academicjournals.org/AJMR

African Journal of Microbiology Research

Full Length Research Paper

Isolation and screening of amylase producing thermophilic spore forming Bacilli from starch rich soil

and characterization of their amylase activity

Mengistu Fentahun1* and Pagadala Vijaya Kumari2

1Ethiopian Institute of Agricultural Research, National Agricultural Biotechnology Research Center, P.O. Box 31, Holetta,

Ethiopia. 2Department of Biology, College of Natural and Computational Science, Ambo University, P.O. Box 19, Ambo, Ethiopia.

Received 26 March, 2017; Accepted 14 May, 2017

Thermostable amylases are the most important enzymes in present with potential industrial applications. The main objective of this study was to isolate and characterize thermophilic amylases from Bacilli found in starch rich soil. Amylase producing bacilli were isolated and their enzymes were also characterized. Effect of temperature, pH, substrate and salt concentration on amylases activity were determined. All amylases produced by different isolates were hydrolyzed greater than 91% of starch after 60 h of fermentation. There was no significant (P ≥ 0.05) variation in enzyme productivity along with fermentation time. Amylase producing isolates were designated as Isolate-1, Isolate-2, Isolate-3, Isolate-4, Isolate-5 and Isolate-6. Amylases activities of all isolates were reached their optimum at 60°C. Amylases from all bacilli isolates were shown hydrolysis capacity of starch ranging from 91.4 to 95.7%. The optimum enzyme activity of amylase from Isolate-2 was extended from pH 7 to 8 with starch hydrolysis efficiency of 98% but other isolates enzyme activity reaches 99.5 to 100% at pH 8. The crude amylase extract has an activity with inversely proportional with substrate concentration. The bacterial dry weight increases as the course of incubation time increases and NaCl concentration greater than 5 molar significantly decreases the activity of the crude amylase extract. Amylases of this finding with thermophilic, alkalophilic and halophilic characteristics have wide range of huge potential for industrial applications. Besides, further purification of the crude extract could be conducted to meet thermophilic amylase enzyme requirements of pharmaceuticals and clinical sectors. Key words: Alkalophilic, amylase, bacillus, industrial application, thermostable.

INTRODUCTION Amylases are among the most important enzymes widely studied in biotechnology. They constitute a class of

industrial enzymes having approximately 25% of the enzyme in the market (Rajagopalan and Krishnan, 2008).

*Corresponding author: E-mail: [email protected]. Tel: (+251)112370376. Fax: (+251) 112370006.

Author(s) agree that this article remains permanently open access under the terms of the Creative Commons Attribution

License 4.0 International License

852 Afr. J. Microbiol. Res. They have wide area of potential application in baking and bread industry, paper industry, textile desizing, detergent industry, starch liquefaction and saccharification, food and pharmaceutical industries and lastly analyses in medical and clinical chemistry. In addition they play an important role in the biogeochemical cycle of carbon (Gupta et al., 2003). One of the best characteristics of such enzymes is stability at high temperature. Therefore, thermostability is a desired characteristic of most of the industrial enzymes. Thermostable enzymes isolated from thermophilic organisms have a number of commercial and industrial applications because of their stability. Amylases working at high temperature are important for industrial application. This high temperature help to decrease viscosity of the medium, increase substrate solubility and reduce risk of microbial contamination (Kuchner and Arnold, 1997).

Enzymatic liquefaction and saccharification of starch are performed at high temperatures (100 to 110°C) by the help of thermostable amylolytic enzymes. To get such potential thermostable amylolytic enzymes, currently investigation of improved microbial strains from different ecological niches contaminated with starchy substances are being conducted. Amylases produced from such potential microorganism are significant for starch degradation to produce valuable products like crystalline dextrose, glucose, maltose, dextrose syrup and maltodextrins (Asgher et al., 2007).

Almost all microorganisms of the Bacillus genus synthesized thermostable amylase, thus this genus has the potential to dominate the enzyme industry: Bacillus amyloliquifaciens, Bacillus licheniformis, Bacillus stearothremophillus, Bacillus subtilis, Bacillus megaterium and Bacillus circulans (Riaz et al., 2003). The capacity of bacillus strains to produce large quantities of enzymes has placed them among the most important industrial enzyme producers. Indeed, they produce about 60% of commercial enzymes. They are known to be good producers of thermostable-amylase and have been widely used for commercial production of the enzyme for various applications (Prakash and Jaiswal, 2009). They have wide range of useful applications in the food, brewing, textile, detergent and pharmaceutical industries. In detergents production, they are applied to improve cleaning effect and are also used for starch de-sizing in textile industry (Chengyi et al., 1999).

Thermostable-amylases have been reported from several bacterial strains and have been produced through the use of submerged fermentation as well as solid state fermentation (Teodoro and Martins, 2000). Submerged fermentation utilizes free flowing liquid substrates or broths efficiently to produce desired bioactive compounds. Normally, such bioactive compounds are secreted into the fermentation broth. In general, this fermentation technique is appropriate for bacteria that

normally require high moisture content. An additional advantage of this technique is that recovery of products is relatively simple (Subramaniyam and Vimala, 2012). Another best advantage of submerged culture is the technique for sterilization and process control is easier to engineer in these systems (Vidyalakshmi et al., 2009).

In the production of microbial amylase one of the factor is optimization of the culture conditions. A number of investigators have conducted researches to optimize culture condition for amylase production (Saxena et al., 2007). The chemical and physical parameters like pH, temperature, salt concentration and incubation time of microbial fermentation process play great role in enzyme production. Improving the yield of amylase and consequent cost reduction depends on the selection of strains, optimization of the factors affecting biosynthesis, genetic improvement, kinetic studies and biochemical characterization of enzyme (Andualem, 2014).

Isolation and identification of soil microorganisms with best amylase activity could contribute a lot for the discovery of novel potential amylases appropriate for different industrial and biotechnological applications (Mohapatra et al., 2003). The bacilli isolated from soil are considered as an ideal source for the production of bulk extracellular amylase for industrial application (Riaz et al., 2003). Therefore, isolation and screening of thermophilic bacteria from soil samples are significant to discover novel new industrial enzymes. There is a need to isolate and screen amylase producing bacilli bacteria from starch rich soil associated with high temperature environment. Most of the time, the temperature in environment around cereal grinding machine is high and the soil is normally rich in content of starch of cereal flour. The soil in such area may be rich in bacilli species of bacteria with huge potential to produce thermostable amylase (Andualem, 2014).

The purpose of the current investigation is to screen thermophilic Bacillus species from soil sample taken from grain mill houses rich in starchy flours and to study their suitability with regard to thermostable amylase production and characterization of the crude enzyme.

MATERIALS AND METHODS Isolation and screening of amylase producing thermophilic Bacillus strains This study was conducted from January to May, 2011. Samples were collected from Azozo, Arada and Maraki (Gondar town) grain mill house soils rich in content of starchy flours. The sample site was located in Northwest part of Ethiopia. To isolate an aerobic, rod shaped, gram-positive, thermophilic and spore-forming Bacillus sp., from each soil samples 1 g was introduced into 9 ml of distilled sterilized water and heated at 80°C for 10 min and then 1 ml was first enriched on starch broth medium containing 1% soluble starch (w/v), 0.5% peptone (w/v) and 0.5% yeast extract (w/v) at pH 7 and incubated at 45°C for 24 h with constant shaking at 150 rpm. Two percent of the enriched liquid medium was then spread on starch agar medium (1% soluble starch, 0.5% peptone, 1.5% yeast

extract, 1.5% agar (Andualem and Gessesse, 2013). The plates were incubated at 45°C for 48 to 72 h until bacteria typical colonies were obtained. The colonies were further sub-cultured on starch agar plates to get pure colonies. Iodine solution (1% iodine in 2% potassium iodide (w/v) was flooded over the surface of the plate in order to select amylase producing isolates. Those colonies having clear zone (more than 10 mm diameter) were selected for further investigation. Morphological characteristics of isolates were identified using gram staining techniques and colony morphology. The cultures were maintained on nutrient agar slants at 4°C. Enzyme production Amylase activity was assayed using starch as substrate. The selected bacterial isolates (designated as Isolate-1, Isolate-2, Isolate-3, Isolate-4, Isolate-5 and Isolate-6) were separately cultured at 45°C for 120 h in 100 ml of starch broth medium (1% starch, 0.5% peptone and 1.5% yeast extract) at pH 7 in 250 ml flask and constantly mixed using rotary shaker at 150 rpm. The broth from each culture was centrifuged at 6000 rpm for 20 min and the supernatant was collected as crude amylase enzyme extract. The crude extract was used for characterization of the enzyme activity and stability in different conditions (Bajpai and Bajpai, 1998). Enzyme assay and characterization The assay was carried out based on the reduction in blue color intensity due to amylase enzyme hydrolysis of starch (Bajpai and Bajpai, 1988; Oboh, 2005; Andualem and Gessesse, 2013). In this assay, 1 ml crude amylase enzyme (cell free supernatant) and 10 ml of 1% starch solution were mixed and incubated at 45°C for 10 min. The reaction in the test tube was stopped by adding 10 ml of 0.1 N HCl. One more dilution was made by mixing 1 ml of this acidified solution with additional 10 ml of 0.1 N HCl. From this, 1 ml was added to 10 ml iodine solution (0.05% iodine in 0.5% KI). Optical density (OD) of the solution was measured by spectrophotometer at 660 nm. A standard curve was prepared using starch (0 to 2.0 mg/ml) and a linear regression analysis was used to determine the total reducing sugar present as % starch equivalents. The same procedure was done using 1 ml distilled water instead of 1 ml enzyme sample (Yang and Liu, 2004). One unit of activity was defined as the amount of enzyme that reduces the intensity of blue color of starch-iodine solution by 1% at the assay conditions. Stock solution and solution for enzyme assay were prepared from 1% or 1000 mg/100 ml soluble starch. Effect of temperature on amylase activity The effect of temperature on amylase activity was determined at different temperatures (40, 45, 50, 55, 60, 65, 70, 75 and 80°C) at pH 7. One milliliter of crude culture extract enzyme was mixed with 10 ml of 1% soluble starch in sodium phosphate buffer (pH 7) and incubated in a water bath at different temperature for 10 min. The reaction was stopped by adding 10 ml of 0.1 N HCl. One more dilution was made by mixing 1 ml of this acidified solution with additional 10 ml of 0.1 N HCl. Then 1 ml from this solution was added to 10 ml iodine reagent containing 0.05% iodine and 0.5% KI. The OD-value was measured at 660 nm (Andualem and Gessesse, 2013). Effect of pH on amylase activity The effect of pH on amylase activity was determined on starch

Fentahun and Kumari 853 solutions (1%) at different acetate buffer and sodium phosphate buffer; acetate buffer of starch solution of pH 4 and 5 and sodium phosphate buffer starch solution of pH 6, 7, 8 and 9 at 60°C for 10 min. The amylase activity was similarly determined from reduction in blue color intensity (Andualem and Gessesse, 2013). Effect of substrate concentration on amylase activity Amylase activity of various crude amylase preparations was assayed at various substrate concentrations of 0.5, 1.0, 1.5, 2.0, 3.0 and 4.0% starch (w/v) solutions in sodium phosphate buffer at pH 8 and temperature of 60°C for 10 min of incubation. Finally the absorbance was measured at 660 nm using UV spectrophotometer (Oboh, 2005; Andualem and Gessesse, 2013). Effect of NaCl concentration on amylase activity The effect of different NaCl concentration on the amylase activity was done by subjecting the enzymes to different NaCl concentrations (0, 1, 2, 3, 4, 5 and 8 M). The enzymes were incubated in 1% starch solution containing different concentration of NaCl solution for 10 min at 60°C and pH 8 and the absorbance was measured at 660 nm (Andualem and Gessesse, 2013). Growth rate and enzyme production dynamics Growth rate, production and enzyme activity were determined by taking samples aseptically from submerged fermentation at the interval of 20 h (0, 20, 40, 60, 80, 100 and 120 h) (Andualem, 2014). The time course of starch hydrolysis by crude amylase extract The enzyme activity was done by measuring the residual activity of the enzyme after being incubated for specific period at specific pH and temperature (Yang and Liu, 2004). Determination of incubation time was carried out on reaction mixture to estimate the time required for maximum amylase activity. At specific time interval (2, 7, 17, 32 and 47 min) the activity of enzyme was determined (Andualem, 2014). Data analysis The data in this study were analyzed using SPSS version 16.0. Means and standard deviations were calculated using analysis of variance (ANOVA) to analyze the significant differences between the means using Duncan’s multiple range test (p < 0.05) when the F-test demonstrated significance. The triplicates mean values of tests were analyzed. Significant difference was defined as p < 0.05. RESULTS AND DISCUSSION Many species of Bacillus are widely known to produce various kinds of extracellular enzymes having wide range of industrial application. Of those enzymes, amylases are the most significant for industrial application. Characteristics of all bacterial isolates and their amylase activity were determined after 120 h of fermentation (Figure 1). The production of amylase and rate of bacterial isolates growth (Figure 2) were analyzed in the

854 Afr. J. Microbiol. Res.

Figure 1. Enzyme production kinetics or dynamics of amylase producing spore former bacilli against fermentation time (h).

Figure 2. Biomass of amylase producing thermophile bacillus bacteria at different incubation time (h).

medium containing 1% starch as source of carbon and 0.5% peptone as a source of nitrogen. This medium was supplemented with 1.5% yeast extract. The measurement of cell growth rate and enzyme activity of all bacterial isolates were carried out at every 20 h time intervals as shown in Figures 1 and 2.

All isolates which are designated as Isolate-1, Isolate-2, Isolate-3, Isolate-4, Isolate-5 and Isolate-6 were gram positive bacilli and they were screened based on size of clear zone diameter (> 10 mm) formed on starch agar. In all bacterial isolates, amylase production was increased together with cell mass increment after 40 h of fermentation. Moreover, that isolates showing high

concentration of product (fermentable sugars), that is, above 90% after 120 h fermentation were selected for further investigation. With regard to enzyme production kinetic or dynamics of amylase producing spore former bacilli, all isolates were able to hydrolyze > 91% of starch after 60 h of submerged fermentation. High biomass (OD) of amylase producing thermopile Bacillus bacteria was seen after 120 h of incubation period. Isolate-5 was shown the highest (0.36 OD) biomass in comparison with other five isolates (range from 0.28 to 0.34 OD) at stationary phase after 120 h incubation. Among studied isolates, there was no significant (P ≥ 0.05) variation in enzyme productivity but there was significant (P ≤ 0.05)

Fermentation time (h)

Fermentation time (h)

Fentahun and Kumari 855

Figure 3. Effect of temperature on the activity of amylase produced from thermophilic bacillus.

difference in biomass. When the isolated bacteria were cultured in submerged fermenter, gradual increment of biomass up to 120 h of fermentation but the amylases activity were continued to remain up to 120 h of fermentation. In brief enzyme synthesis started after 40 h of fermentation and extended up to 120 h of fermentation process. Determination of period of bacterial growth and amylase productivity are significant to optimize time of product recovery.

Determination of effect of temperature and pH for amylases produced from thermophilic spore former bacilli are significant to optimize fermentation process during enzyme production. The optimum temperature of amylase reaction was analyzed by the incubation of crude amylase extract at temperature range of 40 to 80°C (Figure 3). Crude amylase activity of all isolates was relatively increased from 40°C up to 55°C and sharply reaches their optimum reaction activity at 60°C. After optimum temperature, as temperature increases, amylase activities of all isolates were reduced. This could be due to the breakage of secondary, tertiary and quaternary bonds that maintain the three dimensional structure of enzymes at high temperature and thereby would lead to conformational changes of the enzyme active sites. It is known that enzyme activity increases with increasing temperature up to the optimum temperature as the result of increasing kinetic energy, which can favor rate of collisions between substrate and enzyme during hydrolysis process. In this study, the optimum temperature for amylases produced from all bacteria isolates was 60°C. At this optimum temperature, amylase from Isolate-1, Isolate-2, Isolate-3, Isolate-4, Isolate-5 and Isolate-6 have shown starch hydrolysis capacity of 98.5, 99, 98.8, 98.9, 98.5 and 99%

respectively. High optimum temperature activity of the amylase

offers some advantages for industrial processes such as reducing the cooling cost, lowering viscosity of the substrate, reducing the risk of microbial contamination and provides better solubility at high temperature (Burhan et al., 2003). The amylase activities of all enzymes produced from six bacterial isolates in this study were in line with some scientific reports (Burhan et al., 2003; Afiukwa et al., 2009; Andualem, 2014). However, the optimum temperature of amylases reported in this study was lower in comparison with amylases produced from the spices of Thermus (70°C) (Shaw et al., 1995) and Thermus filiformis (95°C) (Egas et al., 1998).Thus, amylases with optimum activity at 60°C, like in this study, have properties considered to be significant for industrial starch liquefaction. Generally, all amylases produced from all bacterial isolates have not shown significant (P ≥ 0.05) activity along with different temperature treatment.

The effect of pH on the enzyme activity of all isolates in this study is shown on Figure 4. The amylase activity of Isolate-1 was increased from 97.5% at pH 4 to 98.7% at pH 7 and the activity was decreases from 100% at pH 8 to 96% at pH 9. At the same time, amylase activity of isolate-2 was increased from 97.5% at pH 4 to 99.5% at pH 8 and then decreased to 95.5% at pH 9. Thus, its optimum amylase activity was at pH 8. On the other hand, amylase activity of Isolate-3, Isolate-4, Isolate-5 and Isolate-6 was increased from 97 to 100%, 97 to 99.5%, 97 to 100% and 97 to 98% from pH 4 to 8 respectively. The optimum amylase activity (98%) of Isolate-6 was significantly (P ≤ 0.05) reduced in comparison with other isolates at pH 8.

Amylase activity of all bacterial isolates were increased

Fermentation temperature (°C)

856 Afr. J. Microbiol. Res.

Figure 4. Effect of pH on the activity of amylase produced from thermophilic bacillus.

with increment of pH up to optimum and reduced after their optimum pH, that is, pH 8. This is a basic property of all enzymes and is probably due to concomitant alteration in the conformation of the enzyme protein caused by changes in pH of its environment (Afiukwa et al., 2009). The optimum pH of amylases activities found in this study was in line with that of optimum amylase activity reported by Fatoni and Zusfahair (2012). Changes in pH, like that of temperature, could change the three dimensional structure of active sites. And also, the enzymes and substrates binding speed to produce maximum product will be highly reduced due to pH change (Garrett and Grisham, 1999). The optimum pH of the amylase in this study was in line with the optimum pH produced by Thermus sp. (Fatoni and Zusfahair, 2012) and amylase from Bacillus KSM-K38 (Hagihara et al., 2001). However, this optimum pH was higher when compared with that of amylase produced by other Thermus filiformis of 5.5 to 6.0 (Egas et al., 1998) and Thermus sp. of 5.5 to 6.5 (Shaw et al., 1995). In other studies it was shown that optimum pH could range from 5.0 to 10.5. Amylase isolated from such bacteria has high enzymatic activity at alkaline pH which has huge potential for detergent industry. It is known that enzymes in detergents have potential abilities to remove tough stains without any environmental effects. Amylases are the second type of enzymes used for formulation of enzymatic detergents (Gupta et al., 2003; Hmidet et al., 2009). Currently, such type of enzyme formulations are widely used for laundry and automatic dish washing in order to remove starchy food substances derived from gravies, potatoes, chocolate, custard, and other smaller oligosaccharides (Mukherjee et al., 2009).

The starch hydrolysis activity of amylases from all

bacterial isolates was measured at optimal pH and temperature (pH 8 and 60°C) (Figure 5). In all bacilli isolates, there was a high enzyme activity in the range between 0.5 and 2% of starch concentration. This result was in agreement with that of Alli et al. (1998) and Oboh (2005). In this result, the amylase activity of Isolate-1 was only reduced from 100 to 62% from the range of 0.5 to 4% starch substrate. The activity of Isolate-2, Isolate-3, Isolate-4, Isolate-5 and Isolate-6 were reduced from 100% to 61.1, 63, 61.8, 62.4 and 63% respectively in increasing the substrate 0.5 to 4% of starch. According to this finding optimal hydrolysis of starch substrate may not surpasses over 2%. This data is significant to optimize fermentation process within this range of substrate concentration.

Effect of salt on enzyme activity was presented on Figure 6. Salt tolerance of amylases in this study was evaluated in the presence of different NaCl concentration (0 to 8 M NaCl). The activity of the enzyme or starch hydrolysis capacity of all the isolates was 100% starting from 0 to 5 M NaCl solution. The amylase activity of all the isolates (Isolate-1, Isolate-2, Isolate-3, Isolate-4, Isolate-5 and Isolate-6) were sharply decreased after 5 M Nacl concentration that is 99.2, 99.3, 99.6, 99.7, 99.1 and 99.4% at 8 M NaCl concentration respectively. This result implies that the enzymes of these isolates are halotolerate and are comparable with some halophilic amylases reported (Carvalho et al., 2008). Different slats like that NH4C1, FeC13 and KC1 have an effect on the activation of amylase enzyme. But NaCl which had no effect up to 5 M concentration may probably be due to the presence of chloride ions in the soil sample. Previous studies has shown that metallic chlorides are usually potent activators of amylase (Oboh and Ajele, 1997;

Fentahun and Kumari 857

Figure 5. The effect of concentration of substrate on amylase activity.

Figure 6. The effect of NaCl concentration on the activity of amylase produced from thermophilic bacillus.

Mohapatra et al., 1998).

The incubation time of amylases activity of bacterial isolates were determined and measured at various incubation times (Figure 7). The optimum enzyme activity of all the isolates, Isolate-1, Isolate-2, Isolate-3, Isolate-4, Isolate-5 and Isolate-6, were 99.6, 99.3, 99.5, 99.2, 99.4 and 99.1% respectively, at 32 min of incubation. Determination of time course of starch hydrolysis is significant to hydrolyze the substrate efficiently during industrial fermentation process.

Conclusion In this study, six best amylase producing bacilli were isolated from soil rich in starch at grain milling house. Determination of period of bacterial growth and amylase productivity are significant to optimize time of product recovery. Amylases obtained with optimum activity at 60°C could be significant for industrial starch liquefaction. Amylase isolated from such bacteria in this study with high enzymatic activity at alkaline pH has huge potential

NaCl concentration in mole (M)

858 Afr. J. Microbiol. Res.

Figure 7. The time course of amylase from thermophilic bacteria required to hydrolyze starch.

for detergent industry. Thermostable amylases from six bacilli isolates may be used for starch hydrolysis up to 60°C, which is a basic property of an enzyme for industrial starch hydrolysis. A wide range of pH stability (pH 4 to pH 9) has significant advantage of preserving and handling the enzyme for commercial and industrial process. Generally, high temperature and pH stability of such enzymes have also potential application in industrial food biotechnology. The outcome of this study implies that the enzymes of these isolates are halotolerant. Determination of time course of starch hydrolysis is significant to hydrolyze the substrate efficiently during industrial fermentation process. Finally soil samples could be converted to wealth by developing a standard method of producing amylase and other useful enzymes. If further purification of the crude amylase was conducted it will have the ability to be applicable in pharmaceutical and clinical sectors.

CONFLICT OF INTERESTS

The authors have not declared any conflict of interests.

ACKNOWLEDGEMENTS

This work was supported by University of Gondar, College of Natural and Computational Science, Department of Biotechnology, Molecular Biology Laboratory. REFERENCES Afiukwa CA, Ibiam UA, Edeogu CO, Nweke FN, Chukwu UE (2009).

Determination of amylase activity of crude extract from partially germinated mango seeds (Mangifera oraphila). Afr. J. Biotechnol. 8(14):3294-3296.

Alli AI, Ogbonna CIC, Rahman ATMF (1998). Hydrolysis of certain Nigerian Cereal starch using crude fungal amylase. Niger. J. Biotechnol. 9(1):24-36.

Andualem B, Gessesse A (2013). Isolation and identification of amylase producing yeasts in ‘tella’ (Ethiopian local beer) and their amylase contribution for ‘tella’ production. J. Microbiol. Biotechnol. Food Sci. 3(1):30-34.

Andualem B (2014). Isolation and screening of amylase producing thermophilic spore forming Bacilli from starch rich soil and characterization of their amylase activities using submerged fermentation. Int. Food Res. J. 21(2):831-837.

Asgher M, Asad MJ, Rahman SU, Legge RL (2007). A thermostable α-amylase from a moderately thermophilic Bacillus subtilis strain for starch processing. J. Food Eng. 79:950-955.

ajpai P, ajpai P ( ). igh-temperature alkaline α-amylase from acillus licheniformis T RD - . Biotechnol. Bioeng. 33:72-78.

Burhan A, Unaldi N, Coral G, Colak O, Aygan A, Gulnaz O (2003). Enzymatic properties of a novel thermostable, thermophilic, alkaline and chelator resistant amylase from an alkaliphilic Bacillus sp. isolate ANT-6. Process Biochem. 38:1397-1403.

Carvalho RVR, Correa TLM, Silva JC, Mansur LRCO, Martins MLL (2008). Properties of an amylase from tehrmophilic Bacillus sp. Braz. J. Microbiol. 39:102-107.

Chengyi WH, Ming M, Jiang R (1999). Studies on the properties of alpha-amylase produced by Bacillus pumilus 289 (PBX96). Acta Microb. Sin. 32:400-404.

Egas MCV, da Costa MS, Cowan DA, Pires EMV (1998). Extracellular α-amylase from Thermus filiformis Ork A2: purification and biochemical characterization. Extremophiles 2:23-32.

Fatoni A, Zusfahair (2012). Thermophilic amylase from Thermus sp. Isolation and its potential application for bioethanol production. Songklanakarin J. Sci. Technol. 34(5):525-531.

Garrett R , Grisham M ( ). iochemistry. Saunder’s ollege Publishing, Philadelphia, US.

Gupta R, Gigras P, Mohapatra H, Goswami VK, Chauhan B (2003). Microbial α-amylases: a biotechnological perspective. Process Biochem. 38:1599-1616.

Hagihara H, Igarashi K, Hayashi Y, Endo K, Ikawa-Kitayama K, Ozaki K, Kawai S (2001). Ito S. Novel a-amylase that is highly resistant to chelating reagents and chemical oxidants from the Alkaliphilic

Bacillus isolate KSM-K38. Appl. Environ. Microbiol. 4:1744-1750. Hmidet N, El-Hadj Ali N, Haddar A, Kanoun S, Alya S, Nasri M (2009).

Alkaline proteases and thermostable α-amylase co-produced by Bacillus licheniformis NH1: Characterization and potential application as detergent additive. Biochem. Engl. J. 47:71-79.

Kuchner O, Arnold FH (1997). Directed evolution of enzyme catalysts. Trends Biotechnol. 15(12):523-530.

Mohapatra BR, Baerjee Vic, Bapuji M (1998). Characterization of a fungal amylase from mucor spp. Associated with marine sponge Spirastella sp. J. Biotechnol. 60:13-17.

Mohapatra BR, Bapuji M, Sree A (2003). Production of industrial enzymes (Amylase, Carbo xymethylcellulase and Protease) by bacteria isolated from marine sedentary Organisms. Acta Biotechnol. 23(1):75-84.

Mukherjee AK, Borah M, Raí SK (2008). To study the influence of different components of fermentable substrates on induction of extracellular α-amylase synthesis by Bacillus subtilis DM-03 in solid-state fermentation and exploration of feasibility for inclusion of α-amylase in laundry detergent formulations. Biochem. Engl. J. 43:149-156.

Oboh G (2005). Nutrient enrichment of Cassava peels using a mixed culture of Saccharomyces cerevisae and Lactobacillus spp solid media fermentation techniques. Electrical J. Biotechnol. 9(1):46-49.

Oboh G. Ajele JO (1997). Effect of some Metallic chlorides on the activity of β-amylase. Niger. J. Biochem. Mol. Biol.12:73-75.

Prakash O, Jaiswal N (2009). Alpha-Amylase: An Ideal Representative of Thermostable Enzymes. Appl. Biochem. Biotechnol. 160(8):2401-2414.

Fentahun and Kumari 859 Rajagopalan G, Krishnan C (2008). Alpha-amylase production from

catabolite derepressed Bacillus subtilis KCC103 utilizing sugarcane bagasse hydrolysate. Bioresour. Technol. 99:3044-3050.

Riaz N, Ul- aq I, Qadeer MA (200 ). haracterization of α-amylase by Bacillus subtilis. Int. J. Agric. Biol. 3:249-252.

Saxena KR, Dutt K, Agarwal L, Nayyar P (2007). A highly and thermostable alkaline amylase from a Bacillus sp. PN5. Bioresour. Technol. 98:260-265.

Shaw JF, Lin FP, Chen SC, Chen HC (1995). Purification and properties of an extracellular α-amylase from Thermus sp. Bot. Bull. Acad. Sin. 36:195-200.

Subramaniyam R, Vimala R (2012). Solid state and submerged fermentation for the production of bioactive substances: a comparative study. Int. J. Sci. Nat. 3(3):480-486.

Teodoro CES, Martins MLL (2000). Culture conditions for the production of thermostable amylase by Bacillus sp. Braz. J. Microbiol. 31:298-302.

Vidyalakshmi R, Paranthaman R, Indhumathi J (2009). Amylase Production on Submerged Fermentation by Bacillus spp. World J. Chem. 4(1):89-91.

Yang CH, Liu WH (2004). Purification and properties of a maltotriose-producing α-amylase from Thermobifida fusca. Enzyme Microb. Technol. 35(2-3):254-260.

Vol. 11(21), pp. 860-887, 7 June, 2017

DOI: 10.5897/AJMR2016.8129

Article Number: 8AFE20E64616

ISSN 1996-0808

Copyright © 2017

Author(s) retain the copyright of this article

http://www.academicjournals.org/AJMR

African Journal of Microbiology Research

Full Length Research Paper

Performance evaluation of oxacillin-resistant Staphylococcus aureus genotypes and taxa on human

and animal blood agar culture media

Marcelo Fabiano Gomes Boriollo1,2,3*, Manoel Francisco Rodrigues Netto1,3, Jeferson Júnior da Silva3, Carlos Tadeu dos Santos Dias4 and José Francisco Höfling3

1Laboratório de Farmacogenética e Biologia Molecular, Faculdade de Ciências Médicas, Universidade José do Rosário

Vellano (UNIFENAS), Alfenas, MG, Brazil. 2Centro de Pesquisa e Pós-graduação em Ciência Animal, Área de Patologia e Farmacologia Animal, Universidade

José do Rosário Vellano (UNIFENAS), Alfenas, MG, Brazil. 3Laboratório de Microbiologia e Imunologia, Departamento de Diagnóstico Oral, Faculdade de Odontologia de

Piracicaba, Universidade Estadual de Campinas (FOP/UNICAMP), Piracicaba, SP, Brazil. 4Departamento de Ciências Exatas, Escola Superior de Agricultura “Luiz de Queiroz”, Universidade de São Paulo

(ESALQ/USP), Piracicaba, SP, Brazil.

Received 25 May 2016 Accepted 31 August, 2016

The performance characteristics of growth, morphological aspects and hemolytic activities of oxacillin-resistant S. aureus (ORSA) strains were studied on twelve types of blood agar (BA) culture media (sheep, bovine, horse, rabbit and human). ORSA isolates were also previously characterized by isoenzymes genotyping and genetic and grouping analysis. Variations in the diameter of the colonies were detected among seven sets of BA media. In terms of morphology, 99, 53 and 98% presented shiny, yellow, and glossy colonies, respectively, regardless of the type of BAs. The rabbit BA favored the expression of hemolysins for most isolates (74%), followed by the human BA and other animal BAs. Certain BA media promoted the expression of hemolysins; however, the expression was correlated with a deficit in the colonial growth potential and vice-versa. The data point to the existence of two or more isolates genetically identical or highly related: (i) that either share or do not share the same wild species-specific phenotypes related to appearance without any influence from the external environment, and that are (ii) potentially virulent depending on the external environment. This study also suggestes the use of rabbit BA for the phenotypical characterization of S. aureus. Key words: Colonial morphology, hemolysis, human and animal blood agar, oxacillin-resistant Staphylococcus aureus, clinical microbiology.

INTRODUCTION The technical procedures of isolation and microbiological culture remain as the “gold standard” for clinical diagnosis of numerous bacterial infections, including the species Staphylococcus aureus. However, the characterization of

certain microorganisms requires a blood source as a supplement in culture media. These culture media have been used routinely for the isolation and preliminary identification of S. aureus and other microorganisms of

medical importance (for example streptococci and enterococci) or even in subcultures preceding phenotypic tests for identification and antimicrobial susceptibility (Anand et al., 2000). In addition, culture media containing the defibrinated sheep, horse, pig or goat blood agar (BA) have been recommended for the isolation of Streptococcus pneumoniae and Streptococcus pyogenes (Anand et al., 2000; Centers for Disease Control and Prevention, 1998; Gratten et al., 1994; Johnson et al., 1996; Sharp and Searcy, 2006). Furthermore, the phenotypic characterization of certain virulence factors in S. aureus (Kuroda et al., 2001) can be determined through use of blood agar culture media, for examining

the determination of exotoxins (, , and -hemolysins) (Bohach et al., 1997; Bohach and Foster, 2000; Peacok et al., 2002; Sakoulas et al., 2002), which also have clinical significance in the development of human diseases (Yarwood and Schlievert, 2003).

Given the unfavorable recommendations for the microbial isolation or susceptibility testing, the potential safety risks to laboratory technical experts (e.g., risk of blood infections: Hepatitis B and HIV) and the low rate of bacterial isolation, human blood is not recommended for growing cultures in microbiological laboratories (e.g., human blood may contain anti-microbial agents and antibodies and it may inhibit the microbial growth or cause false haemolysis) (Anand et al., 2000; Centers for Disease Control and Prevention, 1998; CLSI document M07-A9, 2012; CLSI document M02-A11, 2012; Gratten et al., 1994; Johnson et al., 1996; Satzke et al., 2010). Although there is little data on the subject, in many developing countries, the preparation of bacterial culture media from expired human blood, from donors of blood transfusions, has been a common practice and is considered convenient and inexpensive. This practice has also been routinely employed in bacteriology laboratories from seven countries in the Asia-Pacific region, as mentioned in previous studies (Russell et al., 2006).

The blood considered for this purpose should be defibrinated while harvested or collected in bags containing anticoagulant, thus preventing the formation of clots. Citrate phosphate dextrose (CPD) is the commonly employed anticoagulant. In turn, citric acid has also been used in the food industry as an inhibitor of bacterial growth (Young and Foegeding, 1993; Phillips, 1999) and, for this reason, it has been considered inappropriate for use in culture media. In developed countries, commercial animal laboratories use magnetic stirrers to defibrinate the blood during collection procedures. However, this type of specialized equipment tends to be difficult to obtain in developing countries cases, the procedure for

Boriollo et al. 861 blood collection requires a sterile glass container containing glass beads in situ, which is gently agitated manually and rotated during the process of collection. This allows the binding of fibrin around the spheres to prevent clot formation. However, this practice displays limitations for small volumes of blood. According to Russell and associates, in Fiji, human BA is used routinely in bacteriological diagnostic laboratories, given the impracticality of establishing a reliable source of blood for research laboratories, despite the possibility of collecting animal blood in commercially available human blood donor bags containing CPD (Russell et al., 2006). Taking into consideration the data from the literature about the need for isolation and microbiological and molecular characterization of microorganisms of medical interest, the purpose of the present research was to compare agar culture media supplemented with many types of citrated non-commercial human and animal blood sources and commercially available defibrinated sheep blood. We evaluated each BA in terms of their performance characteristics of bacterial growth and production of hemolysis in vitro, in a special manner, for a group of odontological patients and clinical environment (air) isolates of oxacillin-resistant S. aureus (ORSA). These isolates were previously characterized by isoenzymes genotyping (Multilocus Enzyme Electro-phoresis - MLEE) and genetic and grouping analysis (that is, identification and genetic relationship among strains, clusters and taxa, usually established in molecular epidemiological tracking studies) to establish a possible correlation between phenotype and genotype. MATERIAL AND METHODS

Microbiological sampling

A total of ninety-nine bacterial samples of ORSA, from the bacteria collection of the Laboratório de Farmacogenética e Biologia Molecular, Faculdade de Ciências Médicas and Centro de Pesquisa e Pós-graduação (UNIFENAS), Alfenas, MG, Brazil, were kindly provided and used for the present research. These samples were previously isolated from odontological patients and clinical environment (air) (Faculdade de Odontologia, UNIFENAS) and characterized using microbiological methods of identification [that is, stain of Gram, growth in chromogenic medium CHROMagar Staphylococcus aureus®, catalase test, coagulase test (Coagu-Plasma, Laborclin Produtos para Laboratórios Ltda.), clumping factor A test (Staphy Test, Probac do Brasil Produtos Bacteriológicos Ltda., Marnes La Coquette, France), fermentation of mannitol test and DNAse test (Winn et al., 2008)] and antimicrobial susceptibility testing [that is, diffusion disk (CLSI document M02-A11, 2012; CLSI document M100-S22, 2012) and confirmatory triage for resistance to oxacillin (CLSI document M07- A9, 2012)]. Genotyping of oxacillin- and resistant S. aureus was

*Corresponding author. E-mail: [email protected]

Author(s) agree that this article remains permanently open access under the terms of the Creative Commons Attribution

License 4.0 International License

862 Afr. J. Microbiol. Res. previously done by isoenzyme markers and genetic and grouping analysis. Blood and culture media Blood from clinically healthy animals (sheep, bovines, horses and rabbits) from the Faculdade de Medicina Veterinária and/or biotherium (UNIFENAS), in the absence of antibiotic therapy at the time of blood collection, was harvested directly (sheep, bovines and horses) or indirectly (rabbits, cardiac harvest) in sterile blood bags (CPDA1; citrate phosphate dextrose adenine, 3.27 g of citric acid monohydrate, 26.3 g of sodium citrate dihydrate, 2.51 g of monosodium phosphate dihydrate, 31.9 g of dextrose monohydrate, 0.275 g of adenine, injectable water qsp 1000 mL, sodium concentration 275 mM, Na/1000 mL, pH 5.6 ±0.3; MacoPharma, Mouvaux, France) by veterinary experts using aseptic techniques. Immediately after collection, the blood bags were transported (2-

8°C) to the laboratory, centrifuged at 1,820 g (2,500 rpm) at ambient temperature (Centrifuge Sorvall® RC3C Plus), producing a concentrated volume of red blood cells between 80 and 100 mL and then stored at 4±2°C until the time of use (< 30 days). Expired human blood samples (concentrated volume of red blood cells O+, O-, A+, A-, B+, B-, AB+ and AB- between 250 and 330 mL), collected from various donors in sterile manner (sterile CPDA1 blood bags; MacoPharma, Mouvaux, France) and stored at 4±2°C, were kindly provided by the blood bank from the Hospital Universitário Alzira Velano (HUAV), Alfenas, MG, Brazil. According to the clinical laboratory and serological information (Blood Bank HUAV), all the human blood samples tested negative for syphilis, AIDS, Chagas disease, hepatitis B, hepatitis C, HTLV-1, HTLV-2 and hemoglobin S. Prior to the microbiological tests and to ensure the sterility of the blood culture, 10 mL aliquots of each human and animal blood sample were transferred aseptically to BacTalert type bottles, incubated for seven days and analyzed using a BacT-ALERT® 3D system (bioMérieux Inc., Durham, NC).

Petri dishes containing blood agar (BA) culture media were prepared using standard methods (Oxoid Australia Pty. Ltd) using human and animal blood (5% vol/vol) and Columbia Agar Base (Oxoid Ltd.). A total of twelve types of BA [citrated sheep BA (CSBA), citrated bovine BA (CBBA), citrated horse BA (CHBA), citrated rabbit BA (CRBA), citrated human BA O- (CHuBA O-), citrated human BA O+ (CHuBA O+), citrated human BA A- (CHuBA A-), citrated human BA A+ (CHuBA A+), citrated human BA B- (CHuBA B-), citrated human BA B+ (CHuBA B+), citrated human BA AB- (CHuBA AB-) and citrated human BA AB+ (CHuBA AB+)] were produced and then stored at 4±2°C until used.

Characterization of ORSA on BA

For each isolate of oxacillin-resistant S. aureus, an inoculum was prepared from a direct suspension of bacterial colonies

(approximately 1 to 2 108 CFU.mL-1 of 150 mM NaCl according to 0.5 on the McFarland scale) newly grown in BHI Agar (Brain Heart Infusion Agar, DifcoTM) at 35°C for 18 to 24 h. Using a pipette (Eppendorf Reference®, cat. # 4910 000.018. Eppendorf of Brazil

Ltda. São Paulo, SP), aliquots of 5 L of each bacterial inoculum were applied to the Columbia blood agar (5 isolates per culture

medium) previously prepared in 90 15 mm Petri dishes (20 mL of culture media/dishes; medium height in each dish equal to 4±0.5 mm). These dishes were kept at ambient temperature up to 15 min to allow the complete moisture absorption and incubated in reversed mode at 35°C for 24 h. Soon after the incubation, the dishes were observed, and the results were recorded in terms of colonial morphology (by description and photos), colony diameter (mm) and production of hemolysis (by description and photos).

Hemolytic activity (Pz) was quantitatively and qualitatively characterized using a previously described methodology used for the characterization of virulence factors in vitro of C. albicans [that is, exoenzymatic activity (Pz) of secreted aspartyl proteinase and phospholipase on culture media, which seem to play an important role in pathogenicity of C. albicans and others Candida species] (Barros et al., 2008; Boriollo et al., 2009; Price et al., 1982): Pz = dc/(dc + zp), where dc and zp correspond to the diameter (mm) of the colony and external diameter (mm) of the precipitation zone (hemolysis), respectively. These results were interpreted as follows:

(i) Pz = 1, absence of hemolysis (index 0); (ii) 1 Pz 0.64, positive hemolysis (index 1); and (iii) Pz < 0.64, strongly positive hemolysis (index 2). To ensure the typability and reproducibility of tests, S. aureus ATCC® 25923TM reference strain and commercially available defibrinated sheep BA (DSBA commercial. EBE Farma-Biológica Agropecuária Ltd. Niterói, RJ, Brazil) (the “positive control” for the morphological characteristics and virulence) and Columbia Agar Base (CAB) (“negative control” for the morphological characteristics and virulence) were included in the microbiological characterization tests (triplicate inoculum tests). Statistical analysis The results were also subjected to analysis of one-way variance (ANOVA) in a completely randomized factorial scheme design (culture media BA, taxonomic ranking and colonial phenotypes), and the averages were compared with Tukey’s test (α = 0.05) using SAS® version 9.2.

RESULTS

A population of 99 isolates of oxacillin-resistant S. aureus, previously characterized in 79 strains grouped in three taxa and 15 clusters, was evaluated for the size

(mm of ) and the appearance of colonies (shiny or opaque, yellow or white, glossy or dry) and hemolysis activity (Pz obvious or faint) on thirteen different dishes of BA culture media (CSBA, CBBA, CHBA, CRBA, CHuBA O

-, CHuBA O

+, CHuBA A

-, CHuBA A

+, CHuBA B

-, CHuBA

B+, CHuBA AB

- and CHuBA AB

+) and a dish of CAB

medium, in triplicate. In general, phenotypic variability could be observed between the different strains and even between different isolates belonging to the same strain.

For example, there was variability in colony size and -hemolysis activity among the isolates G20.44 and G18.100 that correspond to the same strain ET41, or still, there was variability in colony appearance, colony size

and -hemolysis activity among the isolates G18.104 and G20.45 that correspond to the same strain ET27, depending on the BA media. Such phenotypic variability was also observed among the strains ET41 and ET27 depending on the BA media (Supplemental Table 1).

Variations in the diameter of bacterial colonies (4-11 mm) grown on the different BA media tested could be observed in this population of isolates, including the reference strain of S. aureus ATCC

® 25923

TM (Table 1).

Most bacterial isolates displayed a range of five (DSBA commercial, CSBA, CRBA, CBBA, CHuBA A

+, CHuBA A

-,

CHuBA B+, CHuBA B

-, CHuBA AB

+, CHuBA AB

-, CHuBA

O+ and CHuBA O

-) or six (CAB and CHBA) millimeters in

Boriollo et al. 863

Table 1. Profiles of the diameters of colonies of oxacillin-resistant S. aureus [99 isolates (79 strains/ETs) and reference strain ATCC® 25923TM] on 13 different types of blood agar plates and one Columbia agar base plate.

The letters A, B, C, D, E

and F correspond to the Tukey grouping. The graphic to the right corresponds to the data from Table 1.

diameter, on average. However, significant differences (p < 0.05) were observed between BA media in seven situations: 1. CSBA produced variations in the diameter of bacterial colonies statistically different (p < 0.05) from those observed in CBBA, CHBA, DSBA commercial, CHuBA O

-

, CHuBA O+, CHuBA A

-, CHuBA A

+, CHuBA B

-, CHuBA

B+, CHuBA AB

-, CHuBA AB

+ and CAB;

2. CRBA produced variations in the diameter of bacterial colonies statistically different (p < 0.05) from those observed in CBBA, CHBA, CHuBA A

-, CHuBA A

+,

CHuBA AB-, CHuBA AB

+, CHuBA B

-, CHuBA B

+, CHuBA

O-, CHuBA O

+ and CAB;

3. DSBA commercial produced variations in the diameter of bacterial colonies statistically different (p < 0.05) from those observed in CBBA, CHBA, CSBA, CHuBA A

+,

CHuBA AB-, CHuBA AB

+, CHuBA B

-, CHuBA B

+, CHuBA

O-, CHuBA O

+ and CAB;

4. CHuBA A- produced variations in the diameter of

bacterial colonies statistically different (p < 0.05) from those observed in CBBA, CHBA, CSBA, CRBA, CHuBA B

- and CAB;

5. CHuBA A+, CHuBA AB

-, CHuBA AB

+, CHuBA B

+,

CHuBA O- and CHuBA O

+ produced variations in the

diameter of bacterial colonies statistically different (p < 0.05) from those observed in CBBA, CHBA, CSBA, DSBA commercial, CRBA, CHuBA B

- and CAB;

6. CBBA, CHBA and CHuBA B- produced variations in

the diameter of bacterial colonies statistically different (p < 0.05) from those observed in CSBA, DSBA commercial, CRBA, CHuBA A

-, CHuBA A

+, CHuBA AB

-, CHuBA AB

+,

CHuBA B+, CHuBA O

-, CHuBA O

+ and CAB; and

7. CAB produced variations in the diameter of bacterial

colonies statistically different (p < 0.05) from those observed in others BA media. These variations in the diameter of bacterial colonies were also evaluated among the largest taxonomic ranks of ORSA [that is, taxa A (60 isolates/43 strains), B (33 isolates/30 strains) and C (7 isolates/6 strains) (Table 2) and among the smaller taxonomic ranks of ORSA (that is, clusters from I to XV)] (Table 3). The taxon A comprised isolates/strains with variations in the diameter of bacterial colonies different significantly (p < 0.05) those observed in taxon B. In turn, the taxon B comprised isolates/strains with variations in the diameter of bacterial colonies different significantly (p < 0.05) those observed in taxon C. The taxa A and C were considered statistically identical. As for the lower ranks, significant differences (p < 0.05) were observed between clusters in 10 situations: 1. Cluster XIV comprised isolates/strains with variations in the diameter of bacterial colonies statistically different (p < 0.05) from those observed in clusters I, II, III, V, VI, VII, VIII, IX, X, XI, XII, XIII and XV; 2. Cluster IV comprised isolates/strains with variations in the diameter of bacterial colonies statistically different (p < 0.05) from those observed in clusters I, II, III, V, VI, VII, IX, X, XI, XII, XIII and XV; 3. Cluster VIII comprised isolates/strains with variations in the diameter of bacterial colonies statistically different (p < 0.05) from those observed in clusters I, II, III, V, IX, X, XI, XII, XIII and XIV; 4. Clusters VI and XV comprised isolates/strains with variations in the diameter of bacterial colonies statistically different (p < 0.05) from those observed in clusters I, II, III, IV, V, IX, X, XI, XII, XIII and XIV;

864 Afr. J. Microbiol. Res.

Table 2. Profiles of the diameters of colonies within and among distantly genetically related populations (taxa A, B and C) of oxacillin-resistant S. aureus on 13 different types of blood agar plates.

The letters Aand

Bcorrespond to the Tukey grouping. The graphic to the right corresponds to the data from Table 2.

5. Cluster VII comprised isolates/strains with variations in the diameter of bacterial colonies statistically different (p < 0.05) from those observed in clusters I, III, IV, V, IX, X, XI, XII, XIII and XIV; 6. Cluster II comprised isolates/strains with variations in the diameter of bacterial colonies statistically different (p

< 0.05) from those observed in clusters IV, V, VI, VIII, XI, XIII, XIV and XV; 7. Cluster III comprised isolates/strains with variations in the diameter of bacterial colonies statistically different (p < 0.05) from those observed in clusters IV, V, VI, VII, VIII, XI, XIII, XIV and XV;

Boriollo et al. 865

Table 3. Profiles of the diameters of colonies within and among clusters moderately related and/or distantly genetically related (clusters of I to XV) of oxacillin-resistant S. aureus on thirteen different types of blood agar plates.

The letters A, B, C, D, E, F, G and H correspond to the Tukey grouping. The graphic to the right corresponds to the data from Table 3.

8. Clusters I, IX, X and XII comprised isolates/strains with variations in the diameter of bacterial colonies statistically different (p < 0.05) from those observed in clusters IV, V, VI, VII, VIII, XI, XIV and XV; 9. Cluster XIII comprised isolates/strains with variations in the diameter of bacterial colonies statistically different (p

< 0.05) from those observed in clusters II, III, IV, V, VI, VII, VIII, XIV and XV; and 10. Cluster V comprised isolates/strains with variations in the diameter of bacterial colonies statistically different (p < 0.05) from those observed in clusters I, II, III, IV, VI, VII, VIII, IX, X, XII, XIII, XIV and XV.

866 Afr. J. Microbiol. Res.

Table 3 Contd.

The frequency of bacterial isolates capable of expressing hemolysins in vitro varied quantitatively and qualitatively (Pz: indexes 1 and 2 for obvious or faint and index 0 for absent) depending on the BA culture media tested. The results indicated that the CRBA culture medium allowed the identification of a large number of isolates capable of expressing hemolysin (74% of the bacterial population),

followed by CHuBA (53-63% of bacterial population), CSBA (48% of bacterial population), CHBA (35% of bacterial population) and CBBA (1% of bacterial population).

Surprisingly, the S. aureus isolates were unable to produce any hemolytic activity in vitro when using DSBA commercial culture medium. However, significant

Boriollo et al. 867 Table 3 Contd.

(p < 0.05) were observed among the BA media in five cases (Table 4): 1. CBBA and DSBA commercial provided the expression of the hemolytic activity for S. aureus isolates statistically different (p < 0.05) when compared to CHBA, CSBA,

CRBA, CHuBA A-, CHuBA A

+, CHuBA AB

-, CHuBA AB

+,

CHuBA B-, CHuBA B

+, CHuBA O

- and CHuBA O

+;

2. CHBA provided the expression of the hemolytic activity for S. aureus isolates statistically different (p < 0.05) when compared to CBBA, CSBA, DSBA commercial, CRBA, CHuBA A

-, CHuBA A

+, CHuBA AB

-, CHuBA AB

+,

868 Afr. J. Microbiol. Res.

Table 3. Contd.

CHuBA B

-, CHuBA B

+, CHuBA O

- and CHuBA O

+;

3. CSBA, CHuBA AB-, CHuBA AB

+, CHuBA B

-, CHuBA

B+, CHuBA O

- and CHuBA O

+ provided the expression of

the hemolytic activity for S. aureus isolates statistically different (p < 0.05) when compared to CBBA, CHBA, DSBA commercial, CRBA, CHuBA A

- and CHuBA A

+;

4. CHuBA A-, CHuBA AB

-, CHuBA AB

+, CHuBA B

-,

CHuBA B+, CHuBA O

-, CHuBA O

+, CHuBA A

+ provided

the expression of the hemolytic activity for S. aureus isolates statistically different (p < 0.05) when compared to CBBA, DSBA commercial, CHBA, CSBA and CRBA; and 5. CRBA and CHuBA A

+ provided the expression of the

hemolytic activity for S. aureus isolates statistically different (p < 0.05) when compared to CBBA, CHBA, CSBA, DSBA commercial, CHuBA A

-, CHuBA AB

-,

CHuBA AB+, CHuBA B

-, CHuBA B

+, CHuBA O

- and

Boriollo et al. 869 Table 3. Contd.

CHuBA O

+.

The hemolysis in vitro activities were also evaluated within and between the major taxonomic ranks of ORSA [that is, taxa A (60 isolates/43 strains), B (33 isolates/30 strains) and C (7 isolates/6 strains). The profiles of the hemolytic activities revealed significant differences (p <

0.05) between taxa A and C, as well as B and C. Taxa A and B were considered statistically identical to the hemolysis profiles produced by ORSA based on their ranks on each type of BA medium (Table 5). The in vitro hemolytic profiles evaluated within and among the lowest taxonomic ranks of ORSA (that is, from clusters I to XV) revealed significant differences (p < 0.05) between

870 Afr. J. Microbiol. Res.

Table 4. Percent index of hemolysis activity (Pz) of oxacillin-resistant S. aureus [99 isolates (79 strains/ETs) and reference strain ATCC® 25923TM] on 13 different types of blood agar plates.

*Pz indexes equal to 0 Pz = 1

, 1 0.64 < Pz < 1

and 2 Pz < 0.64

correspond to absent, positive and strongly positive enzyme activity, respectively. The graphic to the right corresponds to the data from Table 4. The letters

A,

B,

C,

D and

E

correspond to the Tukey grouping.

clusters in six cases (Table 6): 1. Cluster IV comprised isolates/strains able to express hemolytic activity statistically different (p < 0.05) from those observed in clusters I, II, III, IV, V, VI, VII, VIII, IX, X, XI, XIII, XIV and XV; 2. Clusters VII, VIII and IX comprised isolates/strains able to express hemolytic activity statistically different (p < 0.05) from those observed in clusters I, III, IV, V, VI, XII and XIV; 3. Clusters XI, XIII and XV comprised isolates/strains able to express hemolytic activity statistically different (p < 0.05) from those observed in clusters III, IV and XIV; 4. Clusters II and X comprised isolates/strains able to express hemolytic activity statistically different (p < 0.05) from those observed in clusters III, IV, XII and XIV; 5. Clusters I, V and VI comprised isolates/strains able to express hemolytic activity statistically different (p < 0.05) from those observed in clusters III, IV, VII, VIII, IX, XII and XIV; and 6. Clusters III and XIV comprised isolates/strains able to express hemolytic activity statistically different (p < 0.05) from those observed in clusters I, II, IV, V, VI, VII, VIII, IX, X, XI, XII, XIII and XV.

The morphological aspects of the bacterial colonies of all ORSA isolates, including the reference strain ATCC

®

25923TM

, were 99% shiny versus 1% opaque, 53% yellow versus 47% white and 98% glossy versus 2% dry, regardless of the type of BA culture media (Table 7). In each taxonomic rank of ORSA isolates (major ranks: A, B and C taxa; minor ranks: clusters I to XV), these morphological aspects were observed regardless of the

BA media type, although with intrinsic characteristics for each rank. Significant differences (p < 0.05) were observed between the taxa A and B or A and C in terms of shiny/opaque and yellow/white aspects, and, in addition, there were differences between the taxa B and C regarding the glossy/dry aspect (Table 8). No significant difference (p < 0.05) was observed between the clusters regarding shiny/opaque colonies. For glossy/dry, differences were observable between clusters in only one situation [that is, cluster X comprised a significant percentage of isolates/strains exhibiting morphological aspects of dry bacterial colonies (11.1%) when compared to other clusters (0.0%)]. For the yellow/white aspect, such differences were observed between clusters in nine situations (Table 9):

1. Cluster VII comprised a significant percentage (p < 0.05) of isolates/strains exhibiting morphological aspects of yellow/white bacterial colonies when compared to clusters I, II, III, IV, V, VI, IX, X, XI, XII, XIII, XIV and XV; 2. Cluster VIII comprised a significant percentage (p < 0.05) of isolates/strains exhibiting morphological aspects of yellow/white bacterial colonies when compared to clusters I, IV, V, VI, IX, X, XI, XII, XIII, XIV and XV; 3. Cluster II comprised a significant percentage (p < 0.05) of isolates/strains exhibiting morphological aspects of yellow/white bacterial colonies when compared to clusters I, IV, V, VI, VII, IX, X, XI, XII, XIII, XIV and XV; 4. Cluster III comprised a significant percentage (p < 0.05) of isolates/strains exhibiting morphological aspects of yellow/white bacterial colonies when compared to clusters IV, V, VII, X, XII and XIII;

Boriollo et al. 871

Table 5. Percent index of hemolysis activity (Pz) within and among distantly genetically related populations (taxa A, B and C) of oxacillin-resistant S. aureus on thirteen different types of blood agar plates.

* Pz indexes equal to 0 Pz = 1

, 1 0.64 < Pz < 1

and 2 Pz < 0.64

correspond to absent, positive and strongly positive enzyme activity, respectively. The three graphics to the right correspond to the data of each taxon from Table 5. The letters A and B correspond to the Tukey grouping.

5. Clusters I, VI and IX comprised a significant percentage (p < 0.05) of isolates/strains exhibiting morphological aspects of yellow/white bacterial colonies

when compared to clusters II, IV, V, VII, VIII, X, XII and XIII; 6. Clusters XI, XIV and XV comprised a significant

872 Afr. J. Microbiol. Res.

Table 6. Percent index of hemolysis activity (Pz) within and among clusters moderately related and/or distantly genetically related (clusters of I to XV) of oxacillin-resistant S. aureus on thirteen different types of blood agar plates.

*Pz indexes equal to 0 Pz = 1

, 1 0.64 < Pz < 1

and 2 Pz < 0.64

correspond to absent, positive and strongly positive enzyme activity, respectively. The three graphics to the right correspond to the data of each cluster from Table 6, respectively. The letters A, B, C, D and E correspond to Tukey grouping.

percentage (p < 0.05) of isolates/strains exhibiting morphological aspects of yellow/white bacterial colonies when compared to clusters II, V, VII, VIII and XIII;

7. Clusters IV, X and XII comprised a significant percen-tage (p < 0.05) of isolates/strains exhibiting morphological aspects of yellow/white bacterial colonies when

Boriollo et al. 873

Table 6, Contd.

compared to clusters I, II, III, V, VI, VII, VIII and IX; 8. Cluster XIII comprised a significant percentage (p < 0.05) of isolates/strains exhibiting morphological aspects

of yellow/white bacterial colonies when compared to clusters I, II, III, V, VI, VII, VIII, IX, XI, XIV and XV; and 9. Cluster V comprised a significant percentage (p < 0.05)

874 Afr. J. Microbiol. Res.

Table 6. Contd.

of isolates/strains exhibiting morphological aspects of yellow/white bacterial colonies when compared to other clusters.

DISCUSSION The use of BA culture media has been demonstrated to

Boriollo et al. 875

Table 6. Contd.

be useful in clinical microbiological diagnosis of several bacterial infections, especially the species S. aureus, for the isolation and preliminary identification of these pathogens of medical importance and the subcultures that precede the phenotypic assays for identification and antimicrobial susceptibility (Anand et al., 2000; Egwuatu et al., 2014; Satzke et al., 2010; Sharp and Searcy, 2006)

and the phenotypic characterization of certain virulence

factors, such as the determination of exotoxins (, ,

and -hemolysins) (Ali-Vehmas et al., 2001; Bohach et al., 1997; Bohach and Foster, 2000; Peacok et al., 2002; Sakoulas et al., 2002) involved in the development of animal or human diseases (Ali-Vehmas et al., 2001; Yarwood and Schlievert, 2003). To the best of our

876 Afr. J. Microbiol. Res.

Table 6. Contd.

knowledge, this is the first study to compare colonial

morphology [size (millimeter of ) and appearance (shiny or opaque, yellow or white, glossy or dry)] and hemolysis activity (Pz obvious or faint) of oxacillin-resistant S. aureus, from odontological patients and clinical environmental (air) samples and characterized genetically

in terms of population, subpopulations (taxa), clusters and strains

ETs using 13 BA culture media with assays

conducted in triplicate (DSBA commercial, CSBA, CBBA, CHBA, CRBA, CHuBA O

-, CHuBA O

+, CHuBA A

-, CHuBA

A+, CHuBA B

-, CHuBA B

+, CHuBA AB

- and CHuBA AB

+),

including a control CAB culture medium. Phenotypic

Boriollo et al. 877

Table 7. Percent index of the morphological features of bacterial colonies of oxacillin-resistant S. aureus [99 isolates (79 strains/ETs) and reference strain ATCC® 25923] on thirteen different types of blood agar plates.

The letters S, O, Y, W, G and D correspond to shiny, opaque, yellow, white, glossy, and dry, respectively. The graphics to the right correspond to the data from Table 7. The letter

A corresponds to Tukey grouping.

variability can be observed between different strains and even between different isolates belonging to the same strains (that is, variability in appearance and size of the

colony and in the -hemolytic activity depending on the BA culture media) (Supplemental Table 1). As for the size of the diameter of the colonies on these culture media,

variations were observed (4-11 mm of ) in all the population of bacterial isolates, with 5-6 mm being the diameter range reached by most of these isolates. However, significant differences were observed between the BA culture media in seven distinct situations (Table 1), among taxa A and B or B and C (taxa A and C were considered statistically identical) (Table 2) and also among clusters in 10 distinct situations (Table 3). The frequency of bacterial isolates capable of expressing hemolysins in vitro varied both quantitatively and qualitatively (indexes Pz) depending on the BA culture media tested (Table 4). The results revealed that the CRBA culture medium facilitated the identification of a large number of isolates able to express hemolysins (74% of the bacterial population), followed by the CHuBA culture media (53-63%), CSBA (48%), CHBA (35%) and CBBA (1%).

Surprisingly, the S. aureus isolates were unable to produce any hemolytic activity in vitro on commercial DSBA culture medium. For this hemolytic activity, significant differences were observed among the BA culture media in five different situations (Table 4), between taxa A and C or B and C (the taxa A and B were considered statistically identical) (Table 5) and among clusters in six distinct situations (Table 6). These results suggest that the expression of hemolysins by oxacillin-resistant S. aureus can be favored by a bacterial intrinsic

mechanism depending on the use of a particular BA culture media designated for bacteriological diagnostics; in addition, there is a deficit in colonial growth potential (for example, the CRBA culture medium). However, the expression of hemolysins by oxacillin-resistant S. aureus can be partially favored or blocked depending on the use of particular BA culture media, but this is associated with a greater potential for colonial growth (for example, the CHBA and CBBA culture media). In turn, the CHuBA and CSBA culture media appeared to behave as intermediates in comparison to the aforementioned examples. In addition to bacterial intrinsic mechanism, the regulation process of the hemolysin activity is usually associated with the synthesis of other virulence factors: for example, a common regulator for virulence factors is mediated by the same gene regulator which responds to environmental stimuli (including hemolysins) (Jonsson and Wadstrom, 1993; Regassa et al., 1992). Bacterial expression of hemolysin genes was also related to respond to changes in oxygen levels, the redox potential, and glutathione concentration of the environment (Bannan et al., 1993; Karunakaran and Holt, 1993; Williams and Austin, 1992). However, intrinsic events of the bacterial regulation and its environmental stimuli could also be elucidated by hematological and biochemical characterization (for example, human and animal blood agar) and bacterial gene expression studies (for example, hemolysins and virulence factors).

The evaluation of in vitro hemolysis activity within each subpopulation of oxacillin-resistant S. aureus (that is, taxa A, B and C) also pointed to CRBA culture medium as being favorable to the expression of hemolysins, as over 70% of bacterial isolates of each subpopulation

878 Afr. J. Microbiol. Res.

Table 8. Percent index of the morphological features of bacterial colonies within and among distantly genetically related populations (taxa A, B and C) of oxacillin-resistant S. aureus on thirteen different types of blood agar plates.

The letters S, O, Y, W, G and D correspond to shiny, opaque, yellow, white, glossy and dry, respectively. The three graphics to the right correspond to the data of each taxon from Table 8. The letters

A and

B correspond to the Tukey grouping.

were able to produce hemolysins. In general, the CSBA, CHBA, CBBA and commercial DSBA culture media

displayed frequencies of hemolysins expression much lower than those observed for CRBA, and most

Boriollo et al. 879

Table 9. Percent index of the morphological features of bacterial colonies within and among clusters moderately related and/or distantly genetically related (clusters of I to XV) of oxacillin-resistant S. aureus on thirteen different types of blood agar plates.

The letters S, O, Y, W, G and D correspond to shiny, opaque, yellow, white, glossy and dry, respectively. The three graphics to the right correspond to the data of each cluster from Table 9, respectively. The letters

A,

B,

C,

D,

E and

F correspond to the Tukey

grouping.

880 Afr. J. Microbiol. Res.

Table 9, Contd.

CHuBA culture media displayed a frequency between those of CRBA and other BA. As a result, such types of

BA are less suited to routine use by the laboratories during the microbiological diagnosis of S. aureus. These

Boriollo et al. 881

Table 9, Contd.

data suggest support the use of CRBA medium in the characterization and microbiological diagnosis of

oxacillin-resistant S. aureus, especially during routine hemolytic activity detection, regardless of the (sub)

882 Afr. J. Microbiol. Res.

Table 9. Contd.

population genetic classifications.

In addition, during the clusters analysis of oxacillin-resistant S. aureus, for almost all clusters, the majority of

the isolates expressed in vitro hemolysins on CRBA culture medium, although there was variation in the Pz and in terms of obvious and/or faint character, and yet,

Boriollo et al. 883

Table 9, Contd.

regardless of the number of isolates or strains present in the clusters (average of 5.33 ± 3.51 isolates by cluster; average of 3.93 ± 2.68 strains by cluster). These data reinforce the hypothesis above about using CRBA medium in the characterization and microbiological

diagnosis of S. aureus, regardless of the clusters are genetically moderately or distantly related and possibly not related epidemiologically. This information also indicates the existence of two or more genetically identical (same strains

ETs) or highly related (common

884 Afr. J. Microbiol. Res. ancestor) isolates that are possibly related from an epidemiological point of view and with the potential for phenotypic expression of virulence, especially in vitro hemolysins, simultaneously due to their intrinsic molecular metabolisms and under the influence of the external environment. The determination of such environmental influence can be based on the observations of variability of hemolytic expression on human and animal BA media by a single isolate or strain. A comparative study between CSBA (citrated sheep blood agar), CHuBA (citrated human blood agar), DHBA (defibrinated horse blood agar) and DSBA (defibrinated sheep blood agar) used for the isolation and antimicrobial susceptibility testing of the strains of S. pneumoniae, S. pyogenes and S. aureus revealed similar colony count values on all culture media, and size of the colonies was generally smaller and accompanied by an absence or a deficit in hemolysin expression on CHuBA for all three species of microorganisms (Russell et al., 2006). At least for S. aureus, our size results support these findings, as the CHuBA produces a smaller colony diameter than CSBA, CHBA and CBBA. However, in contrast to the hemolysis findings, a large number of isolates and/or clinical strains of oxacillin-resistant S. aureus were potentially capable of producing hemolysis in vitro on CRBA plates, followed by CHuBA, CSBA, CHBA, CBBA and commercial DSBA whose hemolysis were quantitatively reduced on these last types of animal BA medium. Given that the size of the colony, the colony morphology, and hemolysis are essentially critical for identification of S. pneumoniae, S. pyogenes and S. aureus, Russell et al. (2006) discussed the large possibility that these microorganisms can be neglected or mistakenly identified when grown on CHuBA, especially when other microorganisms are present in biological samples, such as those from the upper respiratory tract or skin. In addition, the CHuBA demonstrated a performance in antibiotic susceptibility tests that was insufficient when compared with SBA (sheep blood agar). These findings have profound implications in developing countries where expired human blood is commonly used as a culture media supplement. Accordingly, it is likely that clinical laboratory diagnostics of infectious diseases are underestimated by laboratories using CHuBA culture medium (Russell et al., 2006). Therefore, Hu MHA (human Mueller-Hinton blood agar) plates should not be recommended for antimicrobial susceptibility tests or isolation of S. pneumoniae, S. pyogenes and S. aureus (Anand et al., 2000; Centers for Disease Control and Prevention, 1998; Egwuatu et al., 2014; Gratten et al., 1994; Johnson et al., 1996; Satzke et al., 2010), despite their routine use by developing countries.

Another study published on the isolation of Bordetella pertussis on different BA culture media compared Petri dishes containing HBA (horse blood agar), DSBA (defibrinated sheep blood agar) and anticoagulated HuBA (human blood agar) (Hoppe and Schlagenhauf, 1989).

This comparison demonstrated that the HuBA was inferior to the HBA and DSBA. Despite the lack of clarification on the findings related to HuBA, studies in the literature have suggested that human blood can contain antibiotics, antibodies or other anti-infective agents (Johnson et al., 1996), and the lack of hemolysis on HuBA may also be due to age of red blood cells in human blood that has expired or other factors. It is important to note the similar microbiological findings using the DSBA and CSBA dishes (Russell et al., 2006), although it is reported in the literature that citrate displays antibacterial characteristics (Young and Foegeding, 1993; Phillips, 1999). These findings strengthen the hypothesis that CSBA can be safely used for the isolation of S. pneumoniae, S. pyogenes and S. aureus, at a proportion of 1:10 citrate:blood, although it remains unknown whether smaller proportions may affect the patterns of growth and susceptibility of these microorganisms. Accordingly, care should be taken during collection to ensure the correct proportion of blood, and additional studies should examine this issue (Russell et al., 2006). Other studies have demonstrated that defibrinated pig blood and goat blood are viable alternatives as a supplement for S. pneumoniae culture media (Young and Foegeding, 1993; Phillips, 1999). These findings support the increased possibility of the acceptability of citrated blood from animals other than sheep (Russell et al., 2006).

The effect of the different blood (that is, goat, sheep, cow, chicken, rabbit and fresh human blood) on the cultural and morphological characteristics of the bacterial isolates (P. aeruginosa, S. aureus, K. pneumoniae, and β-heamolytic and non-haemolytic Streptococcus) was recently determined (Egwuatu et al., 2014). All these blood agars supported the growth of all these bacterial isolates and without significant difference in the morphology and cultural characteristics (that is, size, colour, pigmentation, elevation, consistency and shape of the colonies). However, some isolates (especially for S. aureus) showed some differences in their abilities to

distinguish - and -haemolytic patterns dependent on blood agar types (Egwuatu et al., 2014). The diagnostic morphological aspects of the colonies

(that are, shiny or opaque, yellow or white, glossy or dry) were invariably displayed in the total population, in subpopulations (taxa) and in the clusters of isolates of oxacillin-resistant S. aureus, regardless of the BA culture media human and animal. Shiny and glossy bacterial colonies predominated in the total population (Table 7) in the (sub) populations (taxa) (Table 8) and in some isolates cluster (Table 9), and the bacterial colony colors of yellow or white were often similar. Although each taxonomic rank of isolates ORSA (taxa and clusters) displayed these morphological aspects regardless of the type of BA media, significant differences were observed between (i) taxa B and A or C and A regarding the shiny/opaque aspects and yellow/white coloration, (ii) the

taxa B and C regarding the glossy/dry aspect, (iii) the clusters in nine distinct situations regarding the yellow/white coloration and (iv) the cluster X compared to the other clusters regarding the glossy/dry aspect. No difference was observed between the clusters regarding the shiny/opaque aspect. These results indicate that human and animal BA culture media does not influence the morphological aspects of the colonies in terms of appearance, particularly where the colonies are shiny, glossy and either yellow or white. These aspects may be observed independently (i) in (sub)populations regardless of whether they are related and genetically and epidemiologically distant, and (ii) they may be observed in clusters that are moderately related or distantly genetically, and even possibly unrelated epidemiologically. However, certain clusters could harbor isolates/strains that are predominantly yellow or white without any exclusivity for this phenotype. These findings indicate the existence of two or more genetically identical (same strain

ET) or highly related (common ancestor)

isolates that are possibly related from an epidemiological point of view that may share the same wild species-specific phenotypes related to appearance (that is, especially shiny, glossy and yellow or white), without any influence from the external environment. Such a statement may be based on the observations of phenotypic invariance in the appearance of colonies on human and animal BA culture media for the same isolate or strain.

These characteristically invariant morphological aspects were also demonstrated by Russell and associates (2006), which examined only two strains of S. aureus [that is, S. aureus ATCC 25923: opaque-white-glossy (HBA), opaque-white-glossy (CSBA), opaque-white-glossy (DSBA) and opaque-white-glossy (HuBA); S. aureus ATCC 29213: Opaque-yellow-glossy (HBA), opaque-yellow-glossy (CSBA), opaque-yellow-glossy (DSBA) and opaque-yellow-glossy (HuBA)] on different types of BA media. However, this invariability cannot be confirmed for S. pneumoniae and S. pyogenes as currently reported [that is, S. pneumoniae ATCC 6305: shiny-grey (HBA), mucoid-grey (CSBA), dull-grey (DSBA) and dull-grey (HuBA); S. pneumoniae ATCC 49619: Shiny-mucoid-grey (HBA), dry-grey (CSBA), dry-grey (DSBA) and shiny-grey (HuBA); S. pyogenes ATCC 19615: Glossy-white (HBA), dry-grey-white (CSBA), dry-grey (DSBA) and glossy-white (HuBA); S. pyogenes strain JC20: glossy-white (HBA), glossy-white (CSBA), glossy-white (DSBA), and glossy-white (HuBA)] (Russell et al., 2006).

The present study evaluates the performance characteristics of bacterial growth (that is, the size of the

and the appearance of colonies) and the production of in vitro hemolysis of a partial population of oxacillin-resistant S. aureus isolates (that is, dental origin from a molecular epidemiological study in progress), grown on non-commercially sourced human and animal citrated BA

Boriollo et al. 885 culture media [that is, citrated sheep BA (CSBA), citrated bovine BA (CBBA), citrated horse BA (CHBA), citrated rabbit BA (CRBA), citrated human BA O

- (CHuBA O

-),

citrated human BA O+ (CHuBA O

+), citrated human BA A

-

(CHuBA A-), citrated human BA A

+ (CHuBA A

+), citrated

human BA B- (CHuBA B

-), citrated human BA B

+ (CHuBA

B+), citrated human BA AB

- (CHuBA AB

-) and citrated

human BA AB+ (CHuBA AB

+)] and commercially available

defibrinated sheep agar (commercial DSBA). The identification of genotypes and genetic relationship between strains, clusters and taxa, were determined using the MLEE method, clustering and genetic analyses to establish a possible correlation between the phenotypic and genotypic characteristics. The MLEE method has proved to be a powerful tool for the typing of S. aureus in epidemiological studies and possess a high discriminatory power and reproducibility. However, given our particular research goals, no epidemiologic inference was performed in this study.

In the total bacterial population, phenotypic variability was observed between different strains and even between different isolates belonging to the same strain depending on the BA media used (that is, variability in

appearance, in the size of the colony and in the -hemolytic activity). The diameter of the colonies was observed to have variations: (i) in the total population of isolates and (ii) within and between taxonomic ranks (that is, taxa or clusters) depending on the BA media used. As for colony appearance (that is, shiny/opaque, yellow/white and glossy/dry), the BA media did not appear to influence colonial morphology among isolates/strains or taxonomic ranks (that is, taxa or clusters). However, certain ranks did harbor strains that were primarily yellow or white without any exclusivity for this phenotype. In regards to hemolytic activity, the rabbit BA favored the expression of hemolysins, followed by the human BA media and the BA media from other animals. The expression of hemolysis revealed intrinsic characteristics in each taxonomic rank and differences between them (that is, taxa or clusters), with the hemolysis occurrence being dependent the BA media used. These data suggest that the hemolysin expression by S. aureus may be favored through the use of a particular type of BA culture to the detriment of colonial growth potential, particularly the CRBA culture media and vice-versa (that is, expression partially favored or blocked depending on the used BA medium but with a greater associated potential for colonial growth, especially for CHBA and CBBA culture media). This study also suggests the use of the CRBA media in the characterization and microbiological diagnosis of oxacillin-resistant S. aureus, especially during routine detection of hemolytic activity and large-scale studies, regardless of the taxonomic classifications of the isolates (that is, taxa and/or cluster). In addition, phenotypic and genotypic correlation studies of bacterial population groups (that is, the groups of microbial genera of medical importance that

886 Afr. J. Microbiol. Res. require a blood source) and the use of the BA culture media could elucidate (i) the microbial behavior in vitro and (ii) facilitate the standardization of methodology, whether in terms of isolation or in terms of species-specific phenotypic characterization. CONFLICT OF INTERESTS The authors have not declared any conflict of interests. ACKNOWLEDGMENTS This study was supported by the Fundação de Amparo à Pesquisa do Estado de Minas Gerais (FAPEMIG process. APQ-3897-4.03/07) and the Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq process no. 157768/2011-2). We thank Elsevier Language Services for help with English language editing. REFERENCES Ali-Vehmas T, Vikerpuur M, Pyorala S, Atroshi F (2001).

Characterization of hemolytic activity of Staphylococcus aureus strains isolated from bovine mastitic milk. Microbiol. Res. 155(4):339-344

Anand CR, Gordon H, Shaw H, Fonseca K, Olsen M (2000). Pig and goat blood as substitutes for sheep blood in blood-supplemented agar media. J. Clin. Microbiol. 38(2):591-594.

Bannan JD, Moran MJ, Macinnes JI, Soltes GA, Friedman RL (1993). Cloning and characterization of btr, a Bordetella pertussis gene encoding an FNR-like transcriptional regulator. J. Bacteriol. 175(22):7228-7235.

Barros LM, Boriollo MFG, Alves ACBA, Klein MI, Gonçalves RB, Höfling JF (2008). Genetic diversity and exoenzyme activities of Candida albicans and Candida dubliniensis isolated from the oral cavity of Brazilian periodontal patients. Arch. Oral Biol. 53(12):1172-1178.

Bohach GA, Foster TJ (2000). Staphylococcus aureus exotoxins. (V.A. Fischetti, R.P. Novick, J.J. Ferretti, D.A. Portnoy and J.I. Rood, Ed.), Washington: American Society for Microbiology, Gram-positive pathogens.

Bohach GA, Dinges MM, Mitchell DT, Ohlendorf DH, Schlievert PM (1997). Exotoxins. (K.B. Crossley and G.L. Archer, Ed.), New York: Churchill Livingstone, The Staphylococci in human disease.

Boriollo MFG, Bassi RC, Nascimento CMGS, Feliciano LM, Francisco SB, Barros LM, Spolidório LC, Spolidório DMP (2009). Distribution and hydrolytic enzyme characteristics of Candida albicans strains isolated from diabetic patients and their non-diabetic consorts. Oral Microbiol. Immunol. 24(6):437-450.

Centers for Disease Control and Prevention CDCP (1998). Laboratory methods for the diagnosis of meningitis caused by Neisseria meningitidis, Streptococcus pneumoniae, and Haemophilus influenza. Atlanta: Centers for Disease Control and Prevention.

Clinical and Laboratory Standards Institute CLSI (2012). Performance standards for antimicrobial disk susceptibility tests; approved standard—eleventh edition. CLSI document M02-A11, Wayne, PA, USA.

Clinical and Laboratory Standards Institute (2012). Methods for dilution antimicrobial susceptibility tests for bacteria that grow aerobically; approved standard—ninth edition. CLSI document M07-A9, Wayne, PA.

Clinical and Laboratory Standards Institute (2012). Performance standards for antimicrobial susceptibility testing; twenty-second

informational supplement. CLSI document M100-S22, Wayne, PA, USA.

Egwuatu TO, Ogunsola FT, Okodugha IM, Jide B, Arewa DG, Osinupebi OA (2014). Effect of blood agar from different animal blood on growth rates and morphology of common pathogenic bacteria. Adv. Microbiol. 4(16):1237-1241.

Gratten M, Battistutta D, Torzillo P, Dixon J, Manning K (1994). Comparison of goat and horse blood as culture medium supplements for isolation and identification of Haemophilus influenzae and Streptococcus pneumoniae for upper respiratory tract secretions. J. Clin. Microbiol. 32(11):2871-2872.

Hoppe JE, Schlagenhauf M (1989). Comparison of three kinds of blood and two incubation atmospheres for cultivation of Bordetella pertussis on charcoal agar. J. Clin. Microbiol. 27(9):2115-2117.

Johnson DR, Kaplan EL, Sramek J, Bicova R, Havlicek J, Havilickova H, Moliova J, Kriz P (1996). Laboratory diagnosis of group A streptococcal infections. Switzerland: World Health Organization, Geneva.

Jonsson P, Wadstrom T (1993). Pathogenesis of bacterial infections in animals. (C.L. Gyles and C.O. Thoen, Ed.), 2nd edn. Iowa, USA. Pp. 21-34.

Karunakaran T, Holt SC (1993). Cloning of two distinct hemolysin genes from Porphyromonas (Bacteroides) gingivalis in Escherichia coli. Microb. Pathog. 15(1):37-49.

Kuroda M, Ohta T, Uchiyama I, Baba T, Yuzawa H, Kobayashi I, Cui L, Oguchi A, Aoki K, Nagai Y, Lian J, Ito T, Kanamori M, Matsumaru H, Maruyama A, Murakami H, Hosoyama A, Mizutani-Ui Y, Takahashi NK, Sawano T, Inoue R, Kaito C, Sekimizu K, Hirakawa H, Kuhara S, Goto S, Yabuzaki J, Kanehisa M, Yamashita A, Oshima K, Furuya K, Yoshino C, Shiba T, Hattori M, Ogasawara N, Hayashi H, Hiramatsu K (2001). Whole genome sequencing of meticillin-resistant Staphylococcus aureus. Lancet. 357(9264):1225-1240.

Peacok SJ, Moore CE, Justice A, Kantzanou M, Story L, Mackie K, O’Neil G, Day NPJ (2002). Virulence combinations of adhesion and toxin genes in natural populations of Staphylococcus aureus. Infect. Immun. 70(9):4987-4996.

Phillips CA (1999). The effects of citric acid, lactic acid, sodium citrate and sodium lactate, alone and in combination with nisin, on the growth of Arcobacter butzleri. Lett. Appl. Microbiol. 29(6):424-428.

Price MF, Wilkinson ID, Gentry LO (1982). Plate method for detection of phospholipase activity in Candida albicans. Sabouraudia. 20(1):7-14.

Regassa LB, Novick RP, Betley MJ (1992). Glucose and nonmaintained pH decrease expression of the accessory gene regulator (agr) in Staphylococcus aureus. Infected. Immun. 60(8):3381-3388.

Russell FM, Biribo SSN, Selvaraj G, Oppedisano F, Warren S, Seduadua A, Mulholland EK, Carapetis JR (2006). As a bacterial culture medium, citrated sheep blood agar is a practical alternative to citrated human blood agar in laboratories of developing countries. J. Clin. Microbiol. 44(9):3346-3351.

Sakoulas G, Eliopoulos GM, Moellering RCJr, Wennersten C, Venkataraman L, Novick RP, Gold HS (2002). Accessory gene regulator (agr) locus in geographically diverse Staphylococcus aureus isolates with reduced susceptibility to vancomycin. Antimicrob. Antimicrob. Agents Chemother. 46(5):1492-1502.

Satzke C, Seduadua A, Chandra R, Carapetis JR, Mulholland EK, Russell FM (2010). Comparison of citrated human blood, citrated sheep blood, and defibrinated sheep blood Mueller-Hinton agar preparations for antimicrobial susceptibility testing of Streptococcus pneumoniae isolates. J. Clin. Microbiol. 48(10):3770-3772.

Sharp SE, Searcy C (2006). Comparison of mannitol salt agar and blood agar plates for identification and susceptibility testing of Staphylococcus aureus in specimens from cystic fibrosis patients. J. Clin. Microbiol. 44(12):4545-4546.

Williams LR, Austin FE (1992). Hemolytic activity of Borrelia burgdorferi. Infect. Immun. 60(8):3224-3230.

Winn WC, Allen SD, Janda WM, Koneman EW, Procop GW, Schreckenberger PC, Woods G (2008). Koneman's Color Atlas and Textbook of Diagnostic Microbiology. 6th edn.: Lippincott Williams & Wilkins, Philadelphia, PA.

Yarwood JM, Schlievert PM (2003). Quorum sensing in Staphylocuccus infections. J. Clin. Invest. 112(11):1620-1625.

Young KM, Foegeding PM (1993). Acetic, lactic, and citric acids and pH

Boriollo et al. 887

inhibition of Listeria monocytogenes Scott A and the effect on intracellular pH. J. Appl. Bacteriol. 74(5):515-520.

Vol. 11(21), pp. 888-907, 7 June, 2017

DOI: 10.5897/AJMR2015.7576

Article Number: 64B986D64628

ISSN 1996-0808

Copyright © 2017

Author(s) retain the copyright of this article

http://www.academicjournals.org/AJMR

African Journal of Microbiology Research

Full Length Research Paper

Antimicrobial effects of novel fluorous and non-fluorous surfactants

Kamonrat Phopin1,2* and Barry S. Bean3

1Center for Research and Innovation, Faculty of Medical Technology, Mahidol University, Bangkok 10700, Thailand. 2Department of Clinical Microbiology and Applied Technology, Faculty of Medical Technology, Mahidol University,

Bangkok 10700, Thailand. 3Department of Biological Sciences, Lehigh University, Bethlehem, PA 18015, USA.

Received 9 May, 2015; Accepted 2 October, 2015

Novel fluorous and non-fluorous surfactants have been synthesized and examined for their potential as microbicides compared with nonoxynol-9 (N-9) by testing their effects on Candida albicans (C. albicans) and Escherichia coli (E. coli). These compounds include nonionic surfactants consisting of F5-triethylene glycol (F5-TEG), F7-triethylene glycol (F7-TEG), C5-triethylene glycol (C5-TEG), C7-triethylene glycol (C7-TEG), and anionic surfactants consisting of F5-propane sultone (F5-PS), and F7-propane sultone (F7-PS). In this study, we investigated the possible effects of fluorous and non-fluorous surfactants on the growth of C. albicans and E. coli cultured in vitro, which were treated individually with different concentrations of these novel surfactants and nonoxynol-9 (N-9). Then, each sample was incubated for 3, 6, 24 and 48 h at 35°C for C. albicans and 37°C for E. coli. After incubation, C. albicans and E. coli colonies were evaluated compared with the control. N-9 and F5-PS had only small effects (25% growth inhibition or lower) on C. albicans but F7-PS, F7-TEG, F5-TEG, and C5-TEG notably inhibited C. albicans growth, and had potential to control their population. C. albicans cells treated with 10% F7-TEG, F5-TEG, or C5-TEG showed no growth; especially, C5-TEG gave the maximum growth inhibition of C. albicans. For E. coli, N-9 had no growth inhibition but F7-PS, F5-TEG, C5-TEG, C7-TEG, and F7-TEG inhibited E. coli growth. Interestingly, F7-TEG showed the maximum inhibition for E. coli starting at a concentration of 1%. Therefore, these surfactants might have potential for prevention or treatment of genital and urinary tract infection from C. albicans and E. coli. Key words: Candida albicans, Escherichia coli, nonoxynol-9, fluorous surfactants, non-fluorous surfactants.

INTRODUCTION Since the 1950’s the nonionic surfactant, nonoxynol-9 (N-9), has been widely used as a contraceptive agent (Savle et al., 1999; Gandour, 2005). It immobilizes sperm in seconds after penetrating into sperm membranes and forming mixed micelles with their lipids and causes sperm

membrane damage (Schill and Wolff, 1981; Wilborn et al., 1983; Doncel, 2006). While effective at killing sperm, many recent studies report that N-9 also damages epithelial cells and normal flora in vagina. N-9 disrupts normal flora such as Lactobacillus and creates imbalance

*Corresponding author. E-mail: [email protected]. Tel: 0830919008.

Author(s) agree that this article remains permanently open access under the terms of the Creative Commons Attribution

License 4.0 International License

Phopin and Bean 889

Table 1. The chemical structures of surfactants.

Surfactant Chemical structures

Nonoxynol-9 (N-9)

F5-triethylene glycol (F5-TEG)

F7-triethylene glycol (F7-TEG)

C5-triethylene glycol (C5-TEG)

C7-triethylene glycol (C7-TEG)

F5-propane sulfone (F5-PS)

F7-propane sulfone (P7-PS)

in the vagina (Klebanoff, 1992; McGroarty et al., 1992; Stafford et al., 1998; Patton et al., 1999; Gupta, 2005). The imbalance in vagina can enhance the growth of Candida and Escherichia coli, which often become serious pathogens in the vagina and urinary tract. In addition, with prolonged use N-9 can cause tiny abrasions inside the sensitive vaginal and anal walls, leading to increased risk of infections (Raymond et al., 2004; Gandour, 2005).

E. coli, a Gram negative bacterium, is commonly found in the lower intestine of warm-blooded animals. Virulent strains of E. coli can cause vaginal and urinary tract infections (McGroarty et al., 1994). Use of N-9 can increase the risk of E. coli colonization in vagina and increase rates of urinary tract infection by four times (McGroarty et al., 1994; Watts et al., 1999). Candida is yeast and the most common cause of opportunistic infection of the female reproductive tract. It is among the normal flora of skin, mouth, vagina, and intestinal tract. There are many species of Candida that can cause genital candidiasis or vulvovaginitis. The most important causative agent is Candida albicans (Dupont, 1995). Use of spermicides containing N-9 can increase the chance of Candida adhesion to epithelia, with consequent increase in the opportunity of genital tract infection (Gandour, 2005).

Due to adverse effects of N-9, novel fluorous and non-fluorous surfactants have been synthesized and their

potentials as spermicides and microbicides are under investigation. These novel surfactants are able to kill human and mouse sperm, but show only little effect on HeLa cells. Additionally, they might have potential to control overgrowth of C. albicans and E. coli. The structures of the compounds reported here are shown in Table 1 along with their common names.

Here, we report the potential of new fluorous and non-fluorous surfactants as microbicides for C. albicans and E. coli.

MATERIALS AND METHODS

Microbes and culture media

Candida albicans received from St. Luke’s Hospital (Bethlehem, PA) was cultured in Yeast Extract-Peptone-Dextrose (YPD) broth and YPD agar (Becton, Dickinson and Company, MD) and incubated at 35oC. E. coli (ATCC 25922) was cultured in Luria-Bertani (LB) broth and agar (Becton, Dickinson and Company, MD) and incubated at 37°C. All culture media were prepared as the instructions from the manufacture.

Surfactants

Non-ionic nonoxynol-9 (N-9) was procured from Sigma-Aldrich Inc., MO. Other surfactants, F5-triethylene glycol (F5-TEG), F7-triethylene glycol (F7-TEG), C5-triethylene glycol (C5-TEG), C5-triethylene glycol (C5-TEG), F5-propane sultone (F5-PS), and F7-propane sultone (F7-PS), were obtained from the Department of

900 Afr. J. Microbiol. Res.

Figure 1. Growth of C. albicans following 24 h of treatments with six surfactants. 100 µL volumes of YPD broth containing various concentrations of surfactants were inoculated with 2x103-1x104 C. albicans, and incubated with rotation at 35°C for 24 h. Then, 5 µL of each C. albicans suspension (undiluted or 1:3500 diluted samples) were dropped on YPD agar and incubated at 35°C for 48 h. After incubation, C. albicans colonies were evaluated compared with control C1 (untreated positive control) and C2 (negative control). a) C. albicans suspension was treated with different concentrations of N-9, F7-PS, F5-TEG, F7-TEG, and C5-TEG from undiluted samples. b) C. albicans suspension was treated with different concentrations of F5-PS, F7-PS, F5-TEG, C5-TEG, and F7-TEG from 1:3500 diluted samples.

Chemistry, Lehigh University, Bethlehem, PA. The chemical structures of all surfactants are listed in the Table 1. The synthesis and properties of these compounds have been previously reported (Bean et al., 2011).

Microbicidal effects

Several concentrations of each surfactant were made in YPD broth: 0.1, 1, and 5% non-ionic N-9, 0.1, 1, 5, and 10% non-ionic C5-TEG, non-ionic C7-TEG, non-ionic F5-TEG, and non-ionic F7-TEG, 1.5, 3.7, 7.4, and 8.8% anionic F5-PS, and 1.72, 7.2, 8.6 and 13.8% anionic F7-PS. Positive control C1 (98 µL YPD broth and 2 µL C. albicans suspension) and negative control C2 (100 µL YPD broth) were prepared. Yeast cells were suspended in 0.85% NaCl solution and cell densities were estimated by absorbance. Cell density was adjusted with spectrophotometer by adding sufficient sterile saline to increase the transmittance to that produced by a 0.5 McFarland standard at 530 nm wavelength. This procedure yielded a yeast stock suspension of 1-5x106 cells per mL. The stock suspension was diluted by 1:50 in YPD broth medium containing each surfactant, which results in 2x104 to 1x105 cells per mL. Tubes containing these treated cultures were incubated with rotation at 35°C for 3, 6, 24, and 48 h. After incubation, 5 µL of each C. albicans suspension were dropped on YPD agar. C. albicans suspensions were undiluted and diluted to 1:1000 and 1:3500 before plating on YPD agar. Each C. albicans-plated YPD agar was incubated at 35°C for 24 and 48 h. After incubation, C. albicans colonies were evaluated compared with control. Results were

documented and shown in Figure 1. For E. coli investigations, similar concentrations of each of the

above surfactants were made in LB broth positive control C1 (98 µL LB broth and 2 µL E. coli suspension) and C3 (49 µL LB broth and 49 µL sterile distilled water) and negative control C2 (100 µL LB broth) were prepared. E. coli was suspended in 0.85% NaCl solution and adjusted with saline to give a turbidity equivalent to the McFarland 0.5 standard (around 1x108 cells/mL). The stock suspension was diluted by 1:50 in LB broth containing each surfactant to give a final organism density of 2x104-1x105 cells per mL. Tubes were cultured with rotation at 37°C for 3, 6, 24 and 48 h. After incubation, 5 µL of each E. coli suspension was dropped on LB agar. E. coli suspensions were undiluted and diluted to 1:3500 before plating on LB agar. Each E. coli-plated LB agar was incubated at 37°C for 24 and 48 h. After incubation, E. coli colonies were evaluated and compared with control. Results were documented and shown in Figure 1.

RESULTS

Effects of fluorous and non-fluorous surfactants on C. albicans

The effects on growth of C. albicans treated with various concentrations of six surfactants were determined. A sampling of results is shown in Figure 1a for undiluted samples and Figure 1b for dilution at 1:3500 at 24 h and

Undiluted

N9N9 F7PSF7PS F5TEGF5TEG F7TEGF7TEG C5TEGC5TEG

Undiluted

N9N9 F7PSF7PS F5TEGF5TEG F7TEGF7TEG C5TEGC5TEG N-9 F7-PS F5-TEG F7-TEG C5-TEG

a) Undiluted

b) 1:3500

1:3500

F5PSF5PS F7PSF7PS F5TEGF5TEG C5TEGC5TEG F7TEGF7TEG F5-PS F7-PS F5-TEG C5-TEG F7-TEG

Phopin and Bean 901

Table 2. The growth of C. albicans following treatments with different concentrations of N-9, F5-PS, F7-PS, F5-TEG, F7-TEG and C5-TEG for 3, 6, 24, and 48 h.

Treatment duration

3 h 6 h 24 h 48 h

Concentration Concentration Concentration Concentration

N-9 0.01% 0.1% 1% 5% 0.01% 0.1% 1% 5% 0.01% 0.1% 1% 5% 0.01% 0.1% 1% 5%

-Undiluted C C C C C C C C C C C C N/A N/A N/A N/A

- 1:1000 N/A N/A N/A N/A N/A N/A N/A N/A C S S S C S S S

- 1:3500 N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A

F5-PS 1.5% 3.7% 7.4% 8.8% 1.5% 3.7% 7.4% 8.8% 1.5% 3.7% 7.4% 8.8% 1.5% 3.7% 7.4% 8.8%

- Undiluted C C C C C C C C C C C C N/A N/A N/A N/A

- 1:1000 N/A N/A N/A N/A N/A N/A N/A N/A C S S S C S S S

- 1:3500 N/A N/A N/A N/A N/A N/A N/A N/A S S S S N/A N/A N/A N/A

F7-PS 1.72% 7.2% 8.6% 13.8% 1.72% 7.2% 8.6% 13.8% 1.72% 7.2% 8.6% 13.8% 1.72% 7.2% 8.6% 13.8%

- Undiluted C C C C C C C C C C C S C C C L

- 1:1000 N/A N/A N/A N/A N/A N/A N/A N/A C C C L C C C NG

- 1:3500 N/A N/A N/A N/A N/A N/A N/A N/A C C S NG C C S NG

F5-TEG 0.1% 1% 5% 10% 0.1% 1% 5% 10% 0.1% 1% 5% 10% 0.1% 1% 5% 10%

- Undiluted C C C S C C C M C C S NG C C S NG

- 1:3500 N/A N/A N/A N/A N/A N/A N/A N/A C S NG NG C M NG NG

F7-TEG 0.1% 1% 5% 10% 0.1% 1% 5% 10% 0.1% 1% 5% 10% 0.1% 1% 5% 10%

- Undiluted C S M M C S M L C S L NG C S L NG

- 1:1000 N/A N/A N/A N/A C M NG NG C L NG NG C L NG NG

- 1:3500 N/A N/A N/A N/A N/A N/A N/A N/A C L NG NG C L NG NG

C5-TEG 0.1% 1% 5% 10% 0.1% 1% 5% 10% 0.1% 1% 5% 10% 0.1% 1% 5% 10%

- Undiluted C C M NG C C L NG C C NG NG C C NG NG

- 1:3500 N/A N/A N/A N/A N/A N/A N/A N/A C S NG NG C S NG NG

C = result is similar to control; S = small effect (25% growth inhibition or lower); M = medium effect (25-75% growth inhibition); L= large effect (75% growth inhibition or higher); NG = no growth; N/A = no test.

a detailed summary for all treatments is presented in Table 2. There was no effect of all concentrations of N-9 on C. albicans of undiluted samples at 3, 6, and 24 h (Figure 1a, 2a). At dilution 1:1000, N-9 showed a little effect in 0.1, 1, and 5% at 24 h and 48 h (Table 2). F5-PS did not show growth-inhibition effect on C. albicans in undiluted samples at any time periods as shown in Figure 2b. However, Candida growth was slightly inhibited at 3.7, 7.4, and 8.8% of F5-PS in the sample dilution of 1:1000 (at 24 and 48 h) and

1:3500 at 24 h (Figure 1b and Table 2). F7-PS at 13.8% concentration showed small effect (25% growth inhibition or lower) on C. albicans in undiluted samples at 24 h (Figure 1a), but it had a large effect (75% growth inhibition or higher) on C. albicans at 48 h (Figure 2c). Moreover, 13.8% of F7-PS showed large effect (75% growth inhibition or higher) on C. albicans at 24 h, and no growth at 48 h for diluted samples of 1:1000 and we found that it inhibited C. albicans growth effectively at 24 and 48 h for diluted samples of 1:3500 (Figure 1b

at 24 h and Table 2). At 10% concentration of F5-TEG, the results of undiluted samples revealed that there were small (25% growth inhibition or lower) and medium (25-75% growth inhibition) effects at 3 and 6 h, respectively and F5-TEG killed all C. albicans following 24 (Figure 1a) and 48 h of treatment (Figure 2d and Table 2). At sample dilution of 1:3500, 5 and 10% of F5-TEG inhibited Candida growth significantly at 24 (Figure 1b) and 48 h (Table 2). The growth of C. albicans cells treated with 1, 5 and 10% of F7-

902 Afr. J. Microbiol. Res.

Figure 2. Growth of C. albicans treatments with various concentrations of six surfactants in undiluted samples at different time points. a) N-9; b) F5-PS; c) F7-PS; d) F5-TEG; e) F7-TEG; f) C5-TEG.

TEG was decreased in undiluted samples at 3 and 6 h (Figure 2e and Table 2). At 24 and 48 h, C. albicans growth was dramatically decreased at 5% and completely inhibited at 10% (Figure 1a and 2e). For sample dilution of 1:1000, 1% concentration of F7-TEG inhibited growth of C. albicans cells, which were significantly killed in 5%

and 10% at 6, 24 and 48 h (Table 2). The similar results were shown for sample dilution of 1:3500 at 24 (Figure 1b) and 48 h (Table 2). When C. albicans was treated with C5-TEG, 5% of C5-TEG decreased cell growth in undiluted samples at 3 and 6 h and showed no growth in the concentration of 10% (Figure 2f and Table 2).

0

1

2

3

4

5

0 3 6 24

Re

lati

ve g

row

th

time (h)

N-9

0.01%

0.1%

1%

5%

C

S

M

L

NG

0

1

2

3

4

5

0 3 6 24

Re

lati

ve g

row

th

time (h)

F5-PS

1.5%

3.7%

7.4%

8.8%

C

S

M

L

NG

0

1

2

3

4

5

0 3 6 24 48

Re

lati

ve g

row

th

time (h)

C5-TEG

0.10%

1%

5%

10%

C

S

M

L

NG

0

1

2

3

4

5

0 3 6 24 48

Re

lati

ve g

row

th

time (h)

F7-PS

1.72%

7.2%

8.6%

13.8%

C

S

M

L

NG

0

1

2

3

4

5

0 3 6 24 48

Re

lati

ve g

row

th

time (h)

F5-TEG

0.10%

1%

5%

10%

C

S

M

L

NG

0

1

2

3

4

5

0 3 6 24 48

Re

lati

ve g

row

th

time (h)

F7-TEG

0.10%

1%

5%

10%

C

S

M

L

NG

a) b)

c) d)

e) f)

Phopin and Bean 903

Table 3. E.coli treated with different concentrations of N-9, F5-PS , F7- PS, F5-TEG, F7-TEG, and C5-TEG.

Treatment Duration

3 h 6 h 24 h 48 h

Concentration Concentration Concentration Concentration

N-9 0.1% 1% 5% 10% 0.1% 1% 5% 10% 0.1% 1% 5% 10% 0.1% 1% 5% 10%

-Undiluted C C C C C C C C C C C C C C C C

- 1:3500 N/A N/A N/A N/A N/A N/A N/A N/A C C C C C C C C

F7-PS 1.72% 7.2% 8.6% 1.72% 7.2% 8.6% 1.72% 7.2% 8.6% 1.72% 7.2% 8.6%

-Undiluted C S L C S L C S NG C S NG

- 1:3500 N/A N/A N/A N/A N/A N/A C L NG C L NG

C5-TEG 0.1% 1% 5% 10% 0.1% 1% 5% 10% 0.1% 1% 5% 10% 0.1% 1% 5% 10%

-Undiluted C C NG NG C C NG NG C C NG NG C C NG NG

- 1:3500 N/A N/A N/A N/A N/A N/A N/A N/A C C NG NG C C NG NG

C7-TEG 0.1% 1% 5% 10% 0.1% 1% 5% 10% 0.1% 1% 5% 10% 0.1% 1% 5% 10%

-Undiluted C C NG NG C C NG NG C C NG NG C C NG NG

- 1:3500 N/A N/A N/A N/A N/A N/A N/A N/A C C NG NG C C NG NG

F5-TEG 0.1% 1% 5% 10% 0.1% 1% 5% 10% 0.1% 1% 5% 10% 0.1% 1% 5% 10%

-Undiluted C C C NG C C L NG C C NG NG C C NG NG

- 1:3500 N/A N/A N/A N/A N/A N/A N/A N/A C C NG NG C C NG NG

F7-TEG 0.1% 1% 5% 10% 0.1% 1% 5% 10% 0.1% 1% 5% 10% 0.1% 1% 5% 10%

-Undiluted C M NG NG C NG NG NG C NG NG NG C NG NG NG

- 1:3500 N/A N/A N/A N/A N/A N/A N/A N/A C NG NG NG C NG NG NG

C = result is similar to control; S = small effect (25% growth inhibition or lower); M = medium effect (25-75% growth inhibition); L= large effect (75% growth inhibition or higher); NG = no growth; N/A = no test.

Additionally, at 24 and 48 h, 5% and 10% of C5-TEG completely inhibited C. albicans growth in undiluted (Figure 1a, 2f and Table 2) and 1:3500 samples (Figure 1b and Table 2). Based on the experimental results, N-9, F5-PS, and F7-PS showed little growth inhibition (25% inhibition or lower) ability at the same concentration. On the other hand, F5-TEG, F7-TEG, and C5-TEG revealed striking growth inhibition of C. albicans.

Effects of fluorous and non-fluorous surfactants on E. coli

E. coli cells were treated with various concentrations

of different surfactants. The results for all surfactants are listed in Table 3, and examples of plates are shown in Figure 3. There was no growth inhibition effect of N-9 on E. coli at all concentrations (Figure 3, 4a and Table 3). At undiluted samples, E. coli treated with F7-PS showed large effects (75% inhibition or higher) at 3 and 6 h, and no growth in concentration of 8.6% at 24 and 48 h (Figure 3a, 4b and Table 3). Similar results of 1:3000 diluted samples treated with 8.6% of F7-PS were found (Figure 3b and Table 3). In addition, we found that 5, 10%, or higher concentration of C5-TEG and C7-TEG inhibited E. coli growth effectively at 3, 6, 24 and

48 h (Figure 3, 4c, 4d and Table 3). 10% of F5-TEG showed E. coli growth inhibition at 3, 6, 24, and 48 h in undiluted treatments (Figure 3a, 4e and Table 3). Moreover, at 24 and 48 h, 5% and higher concentration of F5-TEG completely inhibited E. coli in all samples (Figure 3, 4e and Table3). In concentration of 1, 5 and 10% of F7-TEG dramatically decreased E. coli growth at 6, 24 and 48 h (Figure 3, 4f and Table 3). All F7-TEG treatments produced similar effects on E. coli in undiluted (Figure 4f) and diluted (1:3500) samples (Figure 3 and Table 3). Therefore, F7- TEG was observed to be the most potent surfactant for E. coli growth inhibition. Based on

904 Afr. J. Microbiol. Res.

Figure 3. Growth of E. coli was treated with various concentrations of six surfactants: N-9, F7-PS, C5-TEG, C7-TEG, F5-TEG, and F7-TEG from undiluted (a) and 1:3500 diluted samples (b). 100 µL volumes of LB broth containing various concentrations of surfactants were inoculated with 2x105 E. coli, and incubated with rotation at 37°C. Samples were taken for evaluation of growth potential following 24 h of treatment. 5 µL of samples of treated suspensions were plated directly (3a), or after 1:3500 dilution (3b) in LB broth. E. coli colonies were compared with untreated positive controls (C1 and C3) and negative controls (uninoculated medium) on each plate following 24 h of incubation.

the results, N-9 showed no effect on E. coli, but F7-PS, F5-TEG, C5-TEG, C7-TEG, and F7-TEG showed effective inhibition on E. coli growth. Interestingly, F7-TEG showed the excellent inhibition growth of E. coli starting at low concentration from 1%. DISCUSSION Based on the results, there was no effect of all concentrations of N-9 on C. albicans and E. coli similar to several reports. Uropathogenic bacteria including E. coli, Proteus mirabilis, Enterococcus faecalis and Staphylococcus species have been found growing at concentration of 25% or higher of N-9 (McGroarty et al., 1994). Watts et al. (1999) studied the effects of N-9 on E. coli and found that the number of women increased the colonization of E. coli in vagina after using N-9 (Watts, et al., 1999). Their results are consistent with many previous researches both in vivo and in vitro studies (Fihn et al., 1985; Foxman and Frerichs, 1985; Watts et al., 1999). After N-9 insertion into vagina without using diaphragm, E. coli colonization increase has been reported in several studies (Rosenstein et al., 1998; Watts, et al., 1999). These lead to increase rates of bacteriuria or urinary tract infection (Percival-Smith et al., 1983; Fihn et al., 1996). C. albicans, another organism that causes vaginal infection, can survive at high concentration of N-9 (McGroarty et al., 1994). Candida colonization in vagina has been found after using N-9 and also causes vaginal burning, vaginal itching, and vulvar

burning (Schreiber et al., 2006). Moreover, N-9 has been found to increase the adhesion of Candida species to human epithelial cell leading to increase in the risk of a serious fungal infection (Gandour, 2005). Additionally, many studies have determined that N-9 increases the risk of infection and also causes vaginal inflammation and ulceration which increase the risk of HIV-1 infection in females (Kreiss et al., 1992; Fichorova et al., 2001; Van Damme et al., 2002; Howett and Kuhl, 2005). Increased rates of vaginal ulceration have been found in the use of the vaginal contraceptive sponge, which has high concentration of N-9 (Kreiss et al., 1992; Watts et al., 1999). Vulvar itching, pain, burning and abnormal discharge have been found after using N-9 (d'Oro et al., 1994; McGroarty et al., 1994). It has been shown that N-9 kills the natural vaginal flora including Lactobacillus causing disturbance of the normal acidic vaginal pH and leading to vaginal infection and urinary tract infection (Hooton et al., 1991; Klebanoff, 1992; McGroarty et al., 1992; Stafford et al., 1998; Patton et al., 1999; Watts et al., 1999; Handley et al., 2002; Brzezinski et al., 2004; Gupta, 2005; Zhou et al., 2010; Ravel et al., 2011).

In this study, C. albicans and E. coli were treated with different concentrations of N-9 and various new surfactants at different periods of times. As mentioned above, N-9 was ineffective on C. albicans and E. coli growth inhibition. All surfactants except F5-PS showed the effective effects on C. albicans growth inhibition. N-9 and F5-PS revealed the low level of inhibition on C. albicans. In E. coli, all surfactants showed effects on growth inhibition except N-9. Based upon the results,

a) Undiluted 1 2

3

b) 1:3500 4

5

F7-PS C5-TEG C7-TEG F5-TEG F7-TEG N-9

Phopin and Bean 905

Figure 4. Growth of E. coli treatments with various concentrations of six surfactants in undiluted samples at different time points. a) N-9, b) F7-PS; c) C5-TEG; d) C7-TEG; e) F5-TEG; f) F7-TEG.

most of our novel surfactants have potential to inhibit growth of E. coli and C. albicans. 0.1% N-9 kills human sperm effectively (data not shown), but cannot inhibit growth of both microorganisms. There is a research showing that N-9 is toxic to HeLa cells and Lactobacillus (Gupta, 2005). The five novel surfactants can kill sperm,

but show only little effect on HeLa cells (Bean et al., 2011). From the experimental results, some novel surfactants, F5-TEG (12%) and F7-TEG (10%), killed sperm and controlled E. coli and C. albicans population, but revealed very small effect on HeLa cells (Bean et al., 2011). Therefore, they might be developed to be used as

0

1

2

3

4

5

0 3 6 24 48

Re

lati

ve g

row

th

time (h)

N-9

0.10%

1%

5%

10%

C

S

M

L

NG

0

1

2

3

4

5

0 3 6 24 48

Re

lati

ve g

row

th

time (h)

C5-TEG

0.10%

1%

5%

10%

C

S

M

L

NG

0

1

2

3

4

5

0 3 6 24 48

Re

lati

ve g

row

th

time (h)

C7-TEG

0.10%

1%

5%

10%

C

S

M

L

NG

0

1

2

3

4

5

0 3 6 24 48

Re

lati

ve g

row

th

time (h)

F7-PS

1.72%

7.2%

8.6%

C

S

M

L

NG

0

1

2

3

4

5

0 3 6 24 48

Re

lati

ve g

row

th

time (h)

F5-TEG

0.10%

1%

5%

10%

C

S

M

L

NG

0

1

2

3

4

5

0 3 6 24 48

Re

lati

ve g

row

th

time (h)

F7-TEG

0.10%

1%

5%

10%

C

S

M

L

NG

a) b)

c) d)

e) f)

906 Afr. J. Microbiol. Res. contraceptive agents that have an ability to protect E. coli and C. albicans infections in vagina and urinary tract and also gentle to epithelium cells. Another surfactant, C5-TEG (12%) killed sperm and inhibited growth of E. coli and C. albicans effectively. From its ability, it might be used as spermicide and microbicide; however, it needs to be tested on HeLa cells. Based on effects on both microorganisms, F5-TEG, F7-TEG, C5-TEG, and C7-TEG were able to inhibit microbial growth better than N-9, F5-PS, and F7-PS.

The structure differences between anionic surfactants (F5-PS and F7-PS) and non-ionic surfactants (F5-TEG, F7-TEG, C5-TEG, and C7-TEG) may be the major point resulting in different effects on C. albicans and E. coli growth inhibition. Non-ionic surfactants gave the better results on growth inhibition for both microorganisms. The dissociation in water may relate to the inhibiting activity of the surfactants. It might involve the surface electrostatic potential of membranes in which the hydrophobic portion of these surfactants is inserted as a consequence of the hydrophobic effect (Vieira and Carmona-Ribeiro, 2006; Vieira et al., 2008). Non-ionic surfactants may insert into cell membranes without charge relationship on the membranes, but anionic surfactant insertion relates to charges on the membranes. However, the mechanism of both anionic and non-ionic surfactants on microorganisms is still not clearly understood.

Non-fluorous TEG provided the excellent inhibiting effects on C. albicans. On the other hand, fluorous TEG showed the large effect (75% inhibition or higher) on E. coli. Based upon the results, surfactant toxicity is not depended on only the chemical structure of the surfactants but also on the nature of cell membranes, which are different in diverse species. The C. albicans membrane has differences in chemical composition and physical properties from the membrane of E. coli (Vieira et al., 2008). Moreover, F7-TEG gave better inhibition effects than F5-TEG for both microorganisms. These show that the amount of fluorine in surfactants is important for inhibiting growth activity. The effect of fluorous surfactant on microbial growth is positively correlated with the number of fluorine in surfactant molecule. Nevertheless, in E. coli, the amount of hydrocarbon in non-fluorous TEG surfactants shows no difference in growth inhibition.

Conclusion Our novel surfactants inhibit growth of C. albicans and E. coli including sperm killing. Based on experimental results, these surfactants might be used as antifungal and antibacterial agents. Moreover, they showed the potential to be developed as contraceptive agents, which might substitute the use of N-9. Therefore, non-ionic and anionic fluorous surfactants may have practical values as fungistatic/bacteriostatic or fungicidal/bactericidal agents and might be useful as vaginal microbicides and

spermicides. However, the effects of these surfactants on normal flora need to be determined and their mechanisms on C. albicans and E. coli will be studied to better understand. In the future, these compounds might be useful for treating genital candidiasis and urinary tract infection in patients. CONFLICT OF INTERESTS The author(s) did not declare any conflict of interest. ACKNOWLEDGMENTS Our friend, colleague and mentor, Barry S. Bean, passed away on June 25, 2012 after contributing to this study. Barry Bean earned a doctoral degree in Life Sciences from Rockefeller University in 1970. He joined Lehigh University in 1973 and began a career studying male reproductive biology. We wish to recognize his dedication to science and to all of those he trained. We gratefully acknowledge the gift of Candida albicans from Ms. Cindy McKellin of St. Luke’s Hospital (Bethlehem PA). We would like to thank Dr. Robert A. Flowers for providing surfactants (Department of Chemistry, Lehigh University, PA), Dr. Robert V. Skibbens and Dr. Marie Maradeo for providing advice and culture media (Department of Biological Sciences, Lehigh University, PA). This study was supported by an award from the Biosystems Dynamics Summer Institute within Lehigh’s Howard Hughes Grant. REFERENCES Bean B, Flowers RA, Venditti JJ, Singh R (2011). Spermicidal and

microbicidal compositions. US 2011/0237676 A1. USA Brzezinski A, Stern T, Arbel R, Rahav G, Benita S (2004). Efficacy of a

novel pH-buffering tampon in preserving the acidic vaginal pH during menstruation. Int. J. Gynaecol. Obstet. 85(3): 298-300.

Doncel GF (2006). Exploiting common targets in human fertilization and HIV infection: development of novel contraceptive microbicides. Hum. Reprod. Update 12(2): 103-117.

d'Oro LC, Parazzini F, Naldi L, La Vecchia C (1994). Barrier methods of contraception, spermicides, and sexually transmitted diseases: a review. Genitourin Med. 70(6): 410-417.

Dupont PF (1995). Candida albicans, the opportunist. A cellular and molecular perspective. J. Am. Podiatr. Med. Assoc. 85(2): 104-115.

Fichorova RN, Tucker LD, Anderson DJ (2001). The molecular basis of nonoxynol-9-induced vaginal inflammation and its possible relevance to human immunodeficiency virus type 1 transmission. J. Infect. Dis. 184(4): 418-428.

Fihn SD, Boyko EJ, Normand EH, Chen CL, Grafton JR, Hunt M, Yarbro P, Scholes D, Stergachis A (1996). Association between use of spermicide-coated condoms and Escherichia coli urinary tract infection in young women. Am. J. Epidemiol. 144(5): 512-520.

Fihn SD, Latham RH, Roberts P, Running K, Stamm WE (1985). Association between diaphragm use and urinary tract infection. JAMA 254(2): 240-245.

Foxman B, Frerichs RR (1985). Epidemiology of urinary tract infection: I. Diaphragm use and sexual intercourse. Am. J. Public Health 75(11): 1308-1313.

Gandour RD (2005). Toward a design of affordable, topical

microbicides: acylcarnitine analogues. Curr. Pharm. Des. 11(29): 3757-3767.

Gupta G (2005). Microbicidal spermicide or spermicidal microbicide? Eur. J. Contracept. Reprod. Health Care 10(4): 212-218.

Handley MA, Reingold AL, Shiboski S, Padian NS (2002). Incidence of acute urinary tract infection in young women and use of male condoms with and without nonoxynol-9 spermicides. Epidemiology 13(4): 431-436.

Hooton TM, Fennell CL, Clark AM, Stamm WE (1991). Nonoxynol-9: differential antibacterial activity and enhancement of bacterial adherence to vaginal epithelial cells. J. Infect. Dis. 164(6): 1216-1219.

Howett MK, Kuhl JP (2005). Microbicides for prevention of transmission of sexually transmitted diseases. Curr. Pharm. Des. 11(29): 3731-3746.

Klebanoff SJ (1992). Effects of the spermicidal agent nonoxynol-9 on vaginal microbial flora. J. Infect. Dis. 165(1): 19-25.

Kreiss J, Ngugi E, Holmes K, Ndinya-Achola J, Waiyaki P, Roberts PL, Ruminjo I, Sajabi R, Kimata J, Fleming TR (1992). Efficacy of nonoxynol 9 contraceptive sponge use in preventing heterosexual acquisition of HIV in Nairobi prostitutes. JAMA 268(4): 477-482.

McGroarty JA, Reid G, Bruce AW (1994). The influence of nonoxynol-9-containing spermicides on urogenital infection. J. Urol. 152(3): 831-833.

McGroarty JA, Tomeczek L, Pond DG, Reid G, Bruce AW (1992). Hydrogen peroxide production by Lactobacillus species: correlation with susceptibility to the spermicidal compound nonoxynol-9. J. Infect. Dis. 165(6): 1142-1144.

Patton DL, Kidder GG, Sweeney YC, Rabe LK, Hillier SL (1999). Effects of multiple applications of benzalkonium chloride and nonoxynol 9 on the vaginal epithelium in the pigtailed macaque (Macaca nemestrina). Am. J. Obstet. Gynecol. 180(5):1080-1087.

Percival-Smith R, Bartlett KH, Chow AW (1983). Vaginal colonization of Escherichia coli and its relation to contraceptive methods. Contraception 27(5): 497-504.

Ravel J, Gajer P, Abdo Z, Schneider GM, Koenig SS, McCulle SL, Karlebach S, Gorle R, Russell J, Tacket CO, Brotman RM, Davis CC, Ault K, Peralta L, Forney LJ (2011). Vaginal microbiome of reproductive-age women. Proc. Natl. Acad. Sci. USA. 108 (Suppl. 1):4680-4687.

Raymond EG, Chen PL, Luoto J (2004). Contraceptive effectiveness and safety of five nonoxynol-9 spermicides: a randomized trial. Obstet. Gynecol. 103(3):430-439.

Phopin and Bean 907 Rosenstein IJ, Stafford MK, Kitchen VS, Ward H, Weber JN, Taylor-

Robinson D (1998). Effect on normal vaginal flora of three intravaginal microbicidal agents potentially active against human immunodeficiency virus type 1. J. Infect. Dis. 177(5): 1386-1390.

Savle PS, Doncel GF, Bryant SD, Hubieki MP, Robinette RG, Gandour RD (1999). Acylcarnitine analogues as topical, microbicidal spermicides. Bioorg. Med. Chem. Lett. 9(17): 2545-2548.

Schill WB, Wolff HH (1981). Ultrastructure of human spermatozoa in the presence of the spermicide nonoxinol-9 and a vaginal contraceptive containing nonoxinol-9. Andrologia 13(1): 42-49.

Schreiber CA, Meyn LA, Creinin MD, Barnhart KT, Hillier SL (2006). Effects of long-term use of nonoxynol-9 on vaginal flora. Obstet. Gynecol. 107(1): 136-143.

Stafford MK, Ward H, Flanagan A, Rosenstein IJ, Taylor-Robinson D, Smith JR, Weber J, Kitchen VS (1998). Safety study of nonoxynol-9 as a vaginal microbicide: evidence of adverse effects. J. Acquir. Immune Defic. Syndr. Hum. Retrovirol. 17(4): 327-331.

Van Damme L, Ramjee G, Alary M, Vuylsteke B, Chandeying V, Rees H, Sirivongrangson P, Mukenge-Tshibaka L, Ettiegne-Traore V, Uaheowitchai C, Karim SS, Masse B, Perriens J, Laga M (2002). Effectiveness of COL-1492, a nonoxynol-9 vaginal gel, on HIV-1 transmission in female sex workers: a randomised controlled trial. Lancet 360(9338): 971-977.

Vieira DB, Carmona-Ribeiro AM (2006). Cationic lipids and surfactants as antifungal agents: mode of action. J. Antimicrob. Chemother. 58(4): 760-767.

Vieira OV, Hartmann DO, Cardoso CM, Oberdoerfer D, Baptista M, Santos MA, Almeida L, Ramalho-Santos J, Vaz WL (2008). Surfactants as microbicides and contraceptive agents: a systematic in vitro study. PLoS ONE 3(8): e2913.

Watts DH, Rabe L, Krohn MA, Aura J, Hillier SL (1999). The effects of three nonoxynol-9 preparations on vaginal flora and epithelium. J. Infect. Dis. 180(2):426-437.

Wilborn WH, Hahn DW, McGuire JJ (1983). Scanning electron microscopy of human spermatozoa after incubation with the spermicide nonoxynol-9. Fertil. Steril. 39(5):717-719.

Zhou X, Hansmann MA, Davis CC, Suzuki H, Brown CJ, Schutte U, Pierson JD, Forney LJ (2010). The vaginal bacterial communities of Japanese women resemble those of women in other racial groups. FEMS Immunol. Med. Microbiol. 58(2):169-181.

Vol. 11(21), pp. 908-914, 7 June, 2017

DOI: 10.5897/AJMR2016.8426

Article Number: DA7981164630

ISSN 1996-0808

Copyright © 2017

Author(s) retain the copyright of this article

http://www.academicjournals.org/AJMR

African Journal of Microbiology Research

Full Length Research Paper

Factors associated with candidemia by non-albicans Candida group in midwest region of Brazil:

Eight-year cross-sectional study

Hugo Dias HOFFMANN-SANTOS* and Rosane C. HAHN

Mycology Laboratory, Faculty of Medicine, Federal University of Mato Grosso, Mato Grosso, Brazil.

Received 29 December, 2016; Accepted 1 March, 2017

The alarming increase in the non-albicans Candida group (NAC) as the etiologic agent of bloodstream infections has made it necessary for the factors associated with candidemia caused by NAC to be elucidated. A cross-sectional retrospective study was conducted which included analysis of microbiological reports, medical records and hospital infection notifications in two tertiary hospitals (Mato Grosso, Brazil) over 8 years (2006 to 2014). Of 144 observed episodes of candidemia, the NAC group represented 64.6%. The prevalence of candidemia caused by NAC was equal to 1.10 × 1,000 admissions, which was statistically different (p<0.001) from and greater than the prevalence of Candida albicans (CA). Hospitalization in the intensive care unit (PR = 1.83; p = 0.05), length of stay ≥42 days (PR = 0.62; p = 0.01) and the use of H2 blockers (PR = 1.75; p = 0.03) were significantly associated with death in patients with candidemia caused by NAC, regardless of gender, use of central venous catheter, treatment with amphotericin B and mechanical ventilation. The incidence of candidemia caused by NAC was 34% higher in men and 40% higher in patients who remained hospitalized for ≥42 days, regardless of prematurity, neutropenia, catheter, mechanical ventilation and age. After 42 days of hospitalization, the chances of survival were 67.3% among patients with candidemia caused by NAC and 56.4% among patients with candidemia caused by CA. This study suggests a different behavior between CA and NAC groups, which should be especially considered for choice of treatment regimen with antifungals. Key words: Candidemia, nosocomial bloodstream infection, epidemiology, tertiary care centers.

INTRODUCTION The frequency of candidemias in tertiary care hospitals is increasing constantly in the US (Pfaller et al., 2012), Europe (Berdal et al., 2014) and Brazil (Nucci et al., 2013). Hospitalization in the intensive care unit (ICU) (Heimann et al., 2015), and the increased use of invasive

medical technologies (Hirano et al., 2015) enables the occurrence of bloodstream infection (BSI) caused by Candida yeasts, contributing to the increase in mortality and morbidity rates (Glöckner and Karthaus, 2011; Cornely et al., 2015).

*Corresponding author. E-mail: [email protected].

Author(s) agree that this article remains permanently open access under the terms of the Creative Commons Attribution

License 4.0 International License

According to Yapar (2014), the incidence of candidemia (per 1,000 admissions) was equal to 0.20 in France, 0.38 in Italy, 0.32 in Sweden, 1.95 in Argentina, 0.33 in Chile, 1.96 in Colombia, 0.90 in Ecuador, Venezuela, 0.45 in Canada, 0.53 in Spain, 1.87 in the United Kingdom and 0.21 in Australia. In Brazil, it reduced from 2.49 (2003-2004) to 1.38 (2008-2010); however, in a study conducted in the Midwest Region of Brazil, it reached 1.80 between 2006 and 2011 (Hoffmann-Santos et al., 2013).

The proportion of the non-albicans Candida group (NAC) as an etiological agent of candidemia has been higher in developing countries than among those developed: 56.1% in the USA (Pfaller et al., 2012), 46.5% in Europe (Cornely et al., 2015), 55.8% in Italy (Caggiano et al., 2015), 57.6% in Japan (Morii et al., 2014), 24.1% in Norway (Berdal et al., 2014), 80.0% in India (Juyal et al., 2013); 72.9% in China (Wu et al., 2011; Nucci et al., 2013) and 61.5% in Brazil (Ng et al., 2015), 54.0% in Mexico (Corzo-Leon et al., 2014), 62.4% in Latin America and 59.5% in Brazil Midwest Region of Brazil (Hoffmann-Santos et al., 2013).

The risk factors for candidemia have been extensively studied worldwide, however, owing to the alarming increase of NAC as the etiologic agent of these infections (Colombo et al., 2007; Ortega et al, 2011; Mondelli et al., 2012; Guimarães et al., 2012; Diekema et al., 2012; Muñoz et al., 2016; Imran and Alshammry, 2016), it is also important that further exploration is performed regarding the factors associated with candidemia caused by NAC, because in Brazil, the C. parapsilosis complex has shown resistance to fluconazole and caused an outbreak in an intensive care unit (Pinhati et al., 2016).

Adult patients who developed candidemia by NAC remained hospitalized for longer than Candidaemia by C. albicans. In this age group, the average cost per hospitalization was about $14,000 dollars higher for patients with candidemia by NAC than among those who developed C. albicans candidemia (Moran et al., 2009). MATERIALS AND METHODS Study design and data collection An epidemiologic, observational, analytical, cross-sectional retrospective study using secondary data, such as microbiological reports (automated detection and species identification system), medical records and hospital infection notifications, was conducted. Because of this, the patients selected were not exposed to any risk. The information collected was stored in the Epi Info 7™ software (CDC, Atlanta, USA).

The sample consisted of cases that had been confirmed in the clinical and laboratory settings as BSI caused by Candida in patients in two university tertiary hospitals in the city of Cuiabá (Mato Grosso, Brazil), during the period from January 2006 to December 2014. Patients who had a second positive blood culture result in a Candida spp. in a time interval lasting ≤30 days or those who did not show clinical signs and symptoms of infection and lacking information for the selected variables or generic laboratory

Hoffmann-Santos and Hahn 909 identification, were not included. Yeast identification were performed by the hospital microbiology laboratory using the VITEK 2 system (bioMérieux, Marcy-I’Étoile, France). Patient characteristics and statistical analysis The following variables were considered: age, gender, length of stay, death, use of antifungals, hospital sector, the presence of infection by Candida spp, prematurity, low birth weight, neutropenia (neutrophil ≤500 cells/mm3), fever axillary ≥ 38°C, previous surgery, the use of catheters, corticosteroids, H2 blockers, parenteral nutrition and mechanical ventilation.

For the age groups, the authors considered those aged less than or equal to 30 days as neonatal patients; those aged between 31 days and 17 years were considered paediatric patients; those aged between 18 and 59 years were considered adults, and those aged 60 years or above were considered elderly.

Through the Stata Statistical Software® version 12.0 (College Station, TX), the Pearson’s chi-square test was performed to check for association between categorical variables; two proportions test to check equality of prevalences; and the unpaired Student t-test to compare two independent means, or its nonparametric analogue, the Mann-Whitney test, when necessary, considering a p-value of <0.05 in the two-tailed test as significant. In order to estimate the strength of association, the prevalence ratio (PR) and the analysis of their respective confidence intervals (95% CI) were applied. This measure was preferred over the odds ratio, since it can overestimate the PR interfering with the inference of analysis. To determine the independent effect of explanatory variables on the dependent variable, Poisson regression with robust variance (Barros and Hirakata, 2003) was used to adjust the covariates, as it is the appropriate multiple model. The variables selected for this model had a p-value ≤ 0.20 in the univariate analysis, or biological plausibility.

Finally, considering the length of stay of each patient, death from all causes such as failure and discharge as censorship, the Kaplan-Meier curve was produced to visualize the distribution of the probability of survival. RESULTS There were 144 observed episodes of candidemia, of which Candida parapsilosis complex was the main etiologic agent (n = 56; 38.8%), followed by Candida albicans (n = 51; 35.4), Candida tropicalis (n = 26; 18.1%), Candida glabrata complex (n = 7; 4.9%), Candida guilliermondii (n = 2; 1.4%), Candida haemulonii (n = 1; 0.7%) and Candida krusei (n = 1; 0.7%). The NAC group represented 64.6% of the total candidemias, a ratio 1.8 times the number of candidemias caused by C. albicans (CA).

The mean age was 29.9 years (95% CI = 25.3–34.5), and the median was 29 years. Men had a mean age of 27.8 years and women, 31.4 years, with no statistically significant difference. The average length of stay was equal to 52.2 days (95% CI = 45.0–59.5) with a median of 42 days. The average length of stay was 56.4 and 49.3 days for men and women, respectively, with no statistically significant difference.

There was no statistical difference between the average length of stay and the mean age of patients who

910 Afr. J. Microbiol. Res.

Table 1. Prevalence of candidemia per 1,000 admissions per yeast group among patients admitted to two tertiary hospitals in Midwest Region, Brazil: 2006 to 2014.

Yeast group No. Prevalence 95% CI p

Non-albicans Candida 93 1.10 0.88–1.34 <0.001

Candida albicans 51 0.61 0.45–0.79

Total 144 1.71 1.45 – 2.02

CI: confidence interval.

Table 2. Percentage distribution, univariate and multivariate of the explanatory variables per yeast group among patients admitted to two tertiary hospitals in Midwest Region, Brazil: 2006 to 2014.

Explanatory variables Total

Univariate Analysis Multivariate Analysis

Non-albicans

Candida Candida albicans

p Prevalence

ratio p

Age group

Neonatal 29.8 27.9 33.3 0.84

Pediatrics 12.5 12.9 11.8

Adult 39.6 41.9 35.3

Elderly 18.1 17.3 19.6

Gender

Male 41.7 48.4 29.4 0.02 1.34 0.02

Female 58.3 51.6 70.6

Antifungals 85.4 84.9 86.2 0.82

Intensive care unit 72.2 70.9 74.5 0.65

Death 50.7 48.4 54.9 0.45

Prematurity 20.1 16.1 27.4 0.10

Low birth weight 21.5 19.3 25.5 0.39

Neutropenia 18.7 15.0 25.5 0.12

Fever 77.1 79.6 72.5 0.33

Previous surgery 47.2 45.2 50.9 0.50

Catheter 79.8 76.3 86.3 0.15

Corticotherapy 52.1 52.7 50.1 0.84

H2 blockers 68.7 68.8 68.6 0.98

Parenteral nutrition 73.6 70.9 78.4 0.33

Mechanical ventilation 67.4 63.4 74.5 0.17

Hospitalization ≥42 days 50.7 56.9 39.2 0.04 1.40 0.01

Multivariate analysis model adjusted for prematurity, neutropenia, catheter, mechanical ventilation and age.

developed candidemia caused by NAC or CA. The average age of patients who died and had candidemia caused by both NAC and CA was statistically similar (p = 0.79). In the group of patients with candidemia caused by NAC, those who died showed a statistically different mean age (p<0.01), which was higher than that of those who were discharged. In the group of patients with CA, this difference was not observed. Among those who had candidemia caused by NAC, the average length of stay was statistically different (p = 0.01) and lower for the patients who died, as compared to the average length of stay of discharged patients. In the group of patients with CA, this difference was also not observed.

The prevalence of BSI caused by NAC was statistically different from and higher than the prevalence of candidemia caused by CA (Table 1). Univariate analysis showed a statistical association of candidemia caused by NAC with gender (PR = 1.64; 95% CI = 1.02–2.64; p = 0.03) and length of stay ≥42 days (PR = 1.45; 95% CI = 1.00–2.13; p = 0.04). The other univariate analyses and the respective percentages of the explanatory variables in multivariate analysis are shown in Table 2.

Of all patients with candidemia, 73 died (50.7%). Among patients who had candidemia caused by NAC (n = 93), 45 died (48.4%), while among those who had candidemia caused by CA (n = 51), 28 died (54.9%),

Hoffmann-Santos and Hahn 911

Figure 1. Survival curve for patients with candidemia caused by C. albicans and non-albicans Candida, showing a median survival rate of 45 and 60 days, respectively.

without a statistically significant difference between groups.

Among patients who had candidemia caused by NAC, the univariate analysis showed the existence of a statistical association between death and the use of parenteral nutrition (p<0.01), admission to the ICU (p<0.01), mechanical ventilation (p<0.01), length of stay ≥42 days (p = 0.02), use of H2 blockers (p = 0.02), catheter use (p = 0.02), gender (p = 0.03) and fever (p = 0.03). In the multivariate analysis, admission to the ICU (PR = 1.83; 95% CI = 1.00–3.33; p = 0.05), length of stay ≥42 days (PR = 0.62; 95% CI = 0.42–0.91; p = 0.01), and the use of H2 blockers (PR = 1.75; 95% CI = 1.05–2.92; p = 0.03) were significantly associated with death in patients with candidemia caused by NAC, regardless of gender, and the use of catheter treatment, amphotericin B and mechanical ventilation.

Among patients who had candidemia caused by CA, the univariate analysis showed the existence of statistical association of death with parenteral nutrition (p<0.01) and gender (p = 0.05). In the multivariate analysis, only the use of antifungals (PR = 0.55; 95% CI = 0.34-0.88; p = 0.01) was statistically associated with death, with the protective effect in patients with candidemia caused by CA, regardless of gender, the use of parenteral nutrition, steroids, catheter, admission to ICU, low birth weight and mechanical ventilation.

Figure 1 shows the comparison of survival curves between patients with candidemia caused by CA and

NAC. The median survival probability (p.50) in patients with candidemia caused by NAC occurred with 59 days of hospitalization, while in the group with candidemia caused by CA, it occurred in 46 days. After 42 days of hospitalization, the chance of survival was 67.3% among patients with candidemia caused by NAC and 56.4% among patients with candidemia caused by CA. DISCUSSION Among patients diagnosed with candidemia caused by NAC, those who died showed statistically different and higher mean age (M = 39.2; 95% CI = 31.2–47.1) than patients who were discharged (M = 21.3; 95% CI = 14.1–28.5). This difference can be explained by the fact that the elderly who died had a higher mean age (M = 70; 95% CI = 67.0–72.9) than the elderly who were discharged (M = 67.2; 95% CI = 61.2–73.3). The depletion of clinical reserves of the patient with advanced age confronting fungemia is believed to be one of the reasons for this difference, since 75.0% of elderly patients with candidemia caused by NAC died.

The predictive nature of a risk factor does not mean that it is necessarily the cause of the disease, as a risk factor can indirectly predict an outcome through association with some other variable that is indeed a determinant of disease. Not being the cause of candidemia does not decrease the value of a risk factor

C. albicans Non-C. albicans

912 Afr. J. Microbiol. Res. as a way of predicting the likelihood of this outcome, but identifies it as a marker for increased probability of this condition, the absence of which may not decrease the risk of developing candidemia.

The epidemiological design of this study did not allow us to infer risk, but allowed us to determine the existence of factors associated with candidemia caused by NAC that, as a BSI agent, was 34% higher in men than in women and 40% higher in patients who remained hospitalized for ≥42 days. Although, not statistically significant on the multivariate analysis (p = 0.056), death was 33% lower among patients who had candidemia caused by NAC and were treated with amphotericin B.

One of the limitations of the epidemiological design of this study is that it did not allow inference of cause and effect for determining both the exposure and the outcome simultaneously, since a temporal relationship between them has not been established. In order to reduce the possibility of non-inclusion of cases that have presented their outcome before the study conducted, long time for data collection (8 years) was considered. The existence of different associated factors in candidemias caused by NAC and CA suggests a clinical behaviour that should be better evaluated by studies with other epidemiological designs.

The literatures has shown NAC as the main cause of candidemias (Wu et al., 2014; Lotfi et al., 2015; Ng et al., 2015; Caggiano et al., 2015), with special attention paid to C. parapsilosis complex among the three major etiologic agents, as was also observed in this study. The finding agree with the result of Imran and Alshammry (2016) showing that NAC isolated from blood samples of ICU patients in Iraq (Middle East) were: C. parapsilosis complex 20.34%, C. membrenifaciens 2.97% and 0.25% for Candida sake, while C. albicans was 12.7%. These Candida spp. were identified based on sequencing analysis of the whole ITS region and the ITS2 of the rDNA.

The incidence of death in patients with candidemia caused by NAC of this study was statistically different from (p<0.01) and higher than that published in the study by Wu et al. (2014), who presented similar epidemiological design and hospital setting (tertiary university hospital in China), in which 238 cases of candidemia were identified in 3 years of study (2009 to 2011). Although, the number of beds in the Chinese hospital is almost three times higher and the number of episodes of candidemia is almost double as compared to this study, with only diabetes mellitus (p<0.001) being observed as an independent variable in the multivariate analysis with the outcome candidemia caused by NAC.

Comparing the annual incidences of candidemia in hospitalized patients aged 65 to 84 years in a study conducted in Texas (Oud, 2016), with the prevalence of candidemia detected by this study for the same age group and year, there was a statistically significant similarity between 2006 and 2009. In 2010, the

prevalence of candidemia in hospitalized patients aged 65 to 84 years of this study (0.91 per 10,000) were statistically different (p = 0.04) and lower than the incidence of candidemia for the same age group and year in the Texas study.

A multicenter study with 227 patients with candidemia (2008 to 2010) in China (Li et al., 2015) stratified the distribution of Candida yeasts isolated into two groups: acquired and not acquired in the ICU. There was no statistical association between the species and the hospital sector, which was also observed in hospitals evaluated in Cuiabá (Mato Grosso, Brazil). The absence of preference of Candida yeasts as the etiologic agent of candidemias in ICU demonstrated by this cross-sectional study was supported by the aforementioned cohort.

In the study conducted in China (Li et al., 2015), patients who acquired candidemia in the ICU had a higher mean age (p<0.05) than those who developed candidemia in other inpatient units. In the present study, however, the opposite behavior was observed, as the patients who remained hospitalized in the ICU had statistically different (p<0.001) and lower (24.3 years; 95% CI 18.9–29.7) mean age than patients who were not admitted to the ICU (44.6 years; 95% CI 37.6 to 51.6). This can be explained by the fact that the Chinese study included only patients older than 16 years, while in the present study, patients admitted to the neonatal ICU were also included.

The prevalence of candidemia (per 1,000 admissions) over 8 years (2006 to 2014) was statistically different from (p<0.05) and lower than that reported in a study conducted in Spain (Aguilar et al., 2015): 19.1 between 2012 and 2013; statistically similar to that observed in Italy (Barchiesi et al., 2016): 1.5 between 2010 and 2014; but statistically different (p<0.01) and higher than that found in France (Richet et al., 2002): 0.29 in 1995.

The Italian study (Barchiesi et al., 2016) showed that older age behaved as an independent variable statistically associated with higher risk of mortality. Of the elderly evaluated in this study (n = 26), 73.1% died, reinforcing advanced age as a predictive factor for high mortality in patients with candidemia, even when including aspects previously discussed.

Epidemiological evidence on the intravascular catheter indicates that its retention is statistically associated with increased risk of death (Fisher et al., 2016) and its removal is associated with an increased survival rate (Tortorano et al., 2002). In this study, death was 30% higher in patients using a central venous catheter (p = 0.02) and had candidemia caused by NAC. There was no association between the use of catheter and death among patients who had candidemia caused by CA.

A review article (Krcmery and Barnes, 2002) synthesized the specific risk factors for candidemia caused by NAC: colonization at two sites, venous catheter, APACHE score II, acute leukaemia, bone

marrow transplantation, antifungal prophylaxis, neutron-penia, malignant disease of hematologic origin, surgery and kidney failure. In the present study, male gender and the length of stay ≥42 days behaved as independent variables associated with candidemia caused by NAC. These variables were not mentioned in the aforesaid article.

In the present study, the factors identified in the multivariate analysis as associated with candidemia caused by NAC distinguished from other studies that showed exposure to azole agents (OR = 3.36; p = 0.03) and surgery to implant artificial materials (OR = 37.5; p = 0.009) by Ding et al. (2015); hematological disorders (p<0.001), mean age (p = 0.02), presence of urinary catheter (p = 0.007) and neutropenia (p = 0.003) by Alp et al. (2015).

Factors associated with death in patients with candidemia identified in this study were different from the group of yeasts. In patients with candidemia caused by NAC, death was 83% higher among those who remained hospitalized in the ICU and 75% higher among those who used H2 blockers, but 38% lower among patients who remained hospitalized for ≥42 days. Among patients with candidemia caused by CA, death was 45% lower among those who used antifungals. The evidence seems to suggest the existence of epidemiological differences between candidemia caused by CA and NAC. This distinct behaviour must be taken into account when choosing antifungal therapy for the treatment of hospitalized patients, if possible, before the identification of species and their respective antifungal susceptibility profile (Ding et al., 2015).

In this scenario, it should be mentioned, the important study by Pfaller et al. (2014), in which the authors observed 2,147 cases of candidemia caused by NAC. The C. parapsilosis complex was the most common etiologic agent among the NAC group, as was also observed in this study. The fact that C. parapsilosis complex represents more than 60% of NAC species in this study and that this species has the characteristic of having one of the greatest chances of survival among the Candida yeasts that act as BSI agents (Morii et al., 2014; Cornely et al., 2015) may explain the less aggressive behaviour by the NAC group when compared with CA, as shown in the Kaplan-Meier curve.

According to Gonçalves et al. (2010), C. parapsilosis complex is the third most commom agent of candidaemia in hospitalized patients in Europe, EUA and Latin American. However, the epidemiology of BSI caused by complex is still not completely understood. Data from the Nationwide Sentinel Surveillance of Candidemia in Brazil (Colombo et al., 2006) identified C. parapsilosis complex with a third cause in eleven medical centers. In this study, the percentage of resistance of C. parapsilosis complex to the antifungal 5-Flucytosine was four times greater than the resistance of C. albicans to the same drug. In adult cases, C. parapsilosis complex was the third most commom agent of candidemia and in pediatric cases,

Hoffmann-Santos and Hahn 913 was the second. In the study conducted in the Midwest Region of Brazil, C. parapsilosis complex was the first cause of candidaemia in adult and neonatal intensive care unit.

Isolates from C. parapsilosis complex in hospitals located in the Southeast of Brazil (Ziccardi et al., 2015) showed two molecular types: C. orthopsilosis and C. parapsilosis. The first species was two times most frequent in blood (C. orthopsilosis = 80.0%; C. parapsilosis sensu stricto = 41.9%), but presented smaller MIC mean for antifungal caspofungin (C. orthopsilosis = 0.139; C. parapsilosis sensu stricto = 0.465; p=0.03). About hydrolytic enzyme production, C. orthopsilosis showed larger mean of caseinolytic activity that the C. parapsilosis sensu stricto, with statistical difference (p<0.01), evidentiating its important virulent character. The study conducted in the Midwest Region of Brazil did not aim to perform molecular identification, but noted that the death in patients with C. parapsilosis complex was 46.4% (26/56), but less than C. albicans with 54.9% (28/51).

The important epidemiological differences between the behaviour of candidemias caused by NAC and CA, such as prevalence, associated factors, and the probability of survival, should be considered by the multidisciplinary team in the clinical management of hospitalized patients. Although, similar associated factors have been observed in studies conducted in different countries, these hospital settings have technological, socioeconomic, structural and microbiological conditions that require continuous monitoring of the hospital epidemiology profile of fungal infections in each health unit, and their particularities should be both valued and compared.

CONFLICT OF INTERESTS

The authors declare that there is no conflict of interest. REFERENCES Aguilar G, Delgado C, Corrales I, Izquierdo A, Gracia E, Moreno T,

Romero E, Ferrando C, Carbonell JA, Borrás R, Navarro D (2015). Epidemiology of invasive candidiasis in a surgical intensive care unit: an observational study. BMC Res. Notes 8:491.

Alp S, Arikan-Akdagli S, Gulmez D, Ascioglu S, Uzun O, Akova M (2015). Epidemiology of candidaemia in a tertiary care university hospital: 10-year experience with 381 candidaemia episodes between 2001 and 2010. Mycoses 58:498-505.

Barchiesi F, Orsetti E, Gesuita R, Skrami E, Manso E (2016). Epidemiology, clinical characteristics, and outcome of candidemia in a tertiary referral center in Italy from 2010 to 2014. Infection 44:205-2013.

Barros AJD, Hirakata V (2003). Alternatives for logistic regression in cross-sectional studies: an empirical comparison of models that directly estimate the prevalence ratio. BMC Med. Res. Methodol. 3:21.

Berdal JE, Haagensen R, Ranheim T, Bjørnholt JV (2014). Nosocomial candidemia; risk factors and prognosis revisited; 11 years experience from a Norwegian secondary hospital. PLoS One 9:e103916.

Caggiano G, Coretti C, Bartolomeo N, Lovero G, De Giglio O, Montagna MT (2015). Candida bloodstream infections in Italy: changing

914 Afr. J. Microbiol. Res.

epidemiology during 16 years of surveillance. BioMed Res. Int. 2015:1-9.

Colombo AL, Guimaraes T, Silva LR, de Almeida Monfardini LP, Cunha AK, Rady P, Alves T, Rosas RC (2007). Prospective observational study of candidemia in São Paulo, Brazil: Incidence rate, epidemiology, and predictors of mortality. Infect. Control Hosp. Epidemiol. 28:570-576.

Colombo AL, Nucci M, et al. and for the Brazilian Network Candidemia Study (2006). Epidemiology of candidemia in Brazil: a Nationwide Sentinel Surveillance of Candidemia in Eleven Medical Centers. J. Clin. Microbiol. 44:2816-2823.

Cornely OA, Gachot B, Akan H, Bassetti M, Uzun O, Kibbler C, Marchetti O, de Burghgraeve P, Ramadan S, Pylkkanen L, Ameye L (2015). Epidemiology and outcome of fungemia in a cancer cohort of the Infectious Diseases Group (IDG) of the European Organization for Research and Treatment of Cancer (EORTC 65031). Clin. Infect. Dis. 61:324-331.

Diekema D, Arbefeville S, Boyken L, Kroeger J, Pfaller M (2012). The changing epidemiology of healthcare-associated candidemia over three decades. Diagn. Microbiol. Infect. Dis. 73:45-48.

Ding X, Yan D, Sun W, Zeng Z, Su R, Su J (2015). Epidemiology and risk factors for nosocomial Non-Candida albicans candidemia in adult patients at a tertiary care hospital in North China. Med. Mycol. 53: 684-690.

Fisher BT, Vendetti N, Bryan M, Prasad PA, Localio AR, Damianos A, Coffin SE, Bell LM, Walsh TJ, Gross R, Zaoutis TE (2016). Central venous catheter retention and mortality in children with candidemia: a retrospective cohort analysis. J. Pediatr. Infect. Dis. Soc. 5(4):403-408

Glöckner A, Karthaus M (2011). Current aspects of invasive candidiasis and aspergillosis in adult intensive care patients. Mycoses 54:420-433.

Gonçalves SS, Amorim CS, Nucci M, Padovan ACB, Briones MRS, Melo ASA, Colombo AL (2010). Prevalence rates and antifungal susceptibility profiles of the Candida parapsilosis species complex: results from a nationwide surveillance of candidaemia in Brazil. Clin. Microbiol. Infect. 16:885-887.

Guimarães T, Nucci M, Mendonça JS, Martinez R, Brito LR, Silva N, Moretti ML, Salomao R, Colombo AL (2012). Epidemiology and predictors of a poor outcome in elderly patients with candidemia. Int. J. Infect. Dis. 16:e442-e447.

Heimann SM, Cornely OA, Wisplinghoff H, Kochanek M, Stippel D, Padosch SA, Langebartels G, Reuter H, Reiner M, Vierzig A, Seifert H (2015). Candidemia in the intensive care unit: analysis of direct treatment costs and clinical outcome in patients treated with echinocandins or fluconazole. Eur. J. Clin. Microbiol. Infect. Dis. 34:331-338.

Hirano R, Sakamoto Y, Kudo K, Ohnishi M (2015). Retrospective analysis of mortality and Candida isolates of 75 patients with candidemia: a single hospital experience. Infect. Drug Resist. 8:199-205.

Hoffmann-Santos HD, Paula CR, Yamamoto AC, Tadano T, Hahn RC (2013). Six-year trend analysis of nosocomial candidemia and risk factors in two intensive care hospitals in Mato Grosso, Midwest Region of Brazil. Mycopathologia 176:409-415.

Imran ZK, Alshammry ZW (2016). Molecular diagnosis of Candidemia of intensive care unite patients based on sequencing analysis of ITS regions. Int. J. PharmTech Res. 9(12):658-668.

Juyal D, Sharma M, Pal S, Rathaur VK, Sharma N (2013). Emergence of non-albicans Candida species in neonatal candidemia. N. Am. J. Med. Sci. 5:541-545.

Krcmery V, Barnes AJ (2002). Non-albicans Candida spp. causing fungaemia: pathogenicity and antifungal resistance. J. Hosp. Infect. 50:243-260.

Li C, Wang H, Yin M, Han H, Yue JF, Zhang F, Shan TC, Guo HP, Wu DW (2015). The differences in the epidemiology and predictors of death between candidemia acquired in intensive care units and other hospital settings. Intern. Med. 54:3009-3016.

Lotfi N, Shokohi T, Nouranibaladezaei SZ, Nasrolahi Omran A, Kondori N (2015). High recovery rate of non-albicans Candida species isolated from burn patients with candidemia in Iran. Jundishapur J.

Microbiol. 8:e22929. Mondelli AL, Niéro-Melo L, Bagagli E, Camargo CH, Bruder-Nascimento

A, Sugizaki MF, Carneiro MV, Villas Boas PJ (2012). Candidemia in a Brazilian tertiary hospital: microbiological and clinical features over a six-year period. J. Venom. Anim. Toxins Incl. Trop. Dis. 18:244-252.

Moran C, Grussemeyer CA, Spalding JR, Benjamin DK, Reed SD (2009). Candida albicans and non-albicans bloodstream infections in adult and pediatric patients: comparison of mortality and costs. Pediatr. Infect. Dis. J. 28:433-435.

Morii D, Seki M, Binongo JN, Ban R, Kobayashi A, Sata M, Hashimoto S, Shimizu J, Morita S, Tomono K (2014). Distribuition of Candida species isolated from blood cultures in hospitals in Osaka, Japan. J. Infect. Chemother. 20:558-562.

Muñoz P, Vena A, Valerio M, Álvarez-Uría A, Guinea J, Escribano P, Bouza E (2016). Risk factors for late recurrent candidemia. A retrospective matched case-control study. Clin. Microbiol. Infect. 22:e11-20.

Ng KP, Kuan CS, Kaur H, Na SL, Atiya N, Velayuthan RD (2015). Candida species epidemiology 2000-2013: a laboratory-based report. Trop. Med. Int. Heath 20:1447-1453.

Nucci M, Queiroz-Telles F, Alvarado-Matute T, Tiraboschi IN, Cortes J, Zurita J, Guzman-Blanco M, Santolaya ME, Thompson L, Sifuentes-Osornio J, Echevarria JI (2013). Epidemiology of Candidemia in Latin America: a laboratory-based survey. PLoS One 8:e59373.

Ortega M, Marco F, Soriano A, Almela M, Martínez JA, López J, Pitart C, Mensa J (2011). Candida species bloodstream infection: epidemiology and outcome in a single institution from 1991 to 2008. J. Hosp. Infect. 77:157-161.

Oud L (2016). Secular trends in utilization of critical care services among candidemia-associated hospitalizations: a population-based cohort study. J. Clin. Med. Res. 8:40-43.

Pfaller MA, Andes DR, Diekema DJ, Horn DL, Reboli AC, Rotstein C, Franks B, Azie NE (2014). Epidemiology and outcomes of invasive candidiasis due to non-albicans species of candida in 2,496 patients: data from the Prospective Antifungal Therapy (PATH) registry 2004–2008. PLoS One 9:e101510.

Pfaller M, Neofytos D, Diekema D, Azie N, Meier-Kriesche HU, Quan SP, Horn D (2012). Epidemiology and outcomes of candidemia in 3648 patients: data from the Prospective Antifungal Therapy (PATH Alliance

®) registry, 2004-2008. Diagn. Microbiol. Infect. Dis. 74:323-

331. Pinhati HMS, Casulari LA, Souza ACR, Siqueira RA, Damasceno CMG,

Colombo AL (2016). Outbreak of candidemia caused by fluconazole resistant Candida parapsilosis strains in an intensive care unit. BMC Infect. Dis. 16:433.

Richet H, Roux P, Des Champs C, Esnault Y, Andremont A (2002). Candidemia in French hospitals: incidence rates and characteristics. Clin. Microbiol. Infect. 8: 405-412.

Tortorano AM, Biraghi E, Astolfi AT, Ossi C, Tejada M, Farina C, Perin S, Bonaccorso C, Cavanna C, Raballo A, Grossi A (2002). European Confederation of Medical Mycology (ECMM) prospective survey of candidaemia: report from one Italian region. J. Hosp. Infect. 51:297-304.

Wu JQ, Zhu LP, Ou XT, Xu B, Hu XP, Wang X, Weng XH (2011). Epidemiology and risk factors for non-Candida albicans candidemia in non-neutropenic patients at a Chinese teaching hospital. Med. Mycol. 49:552-555.

Wu Z, Liu Y, Feng X, Liu Y, Wang S, Zhu X, Chen Q, Pan S (2014). Candidemia: incidence rates, type of species, and risk factors at a tertiary care academic hospital in China. Int. J. Infect. Dis. 22:4-8.

Yapar N (2014). Epidemiology and risk factors for invasive candidiasis. Ther. Clin. Risk Manage. 10:95-105.

Ziccardi M, Souza LO, Gandra RM, Galdino AC, Baptista AR, Nunes AP, Ribeiro MA, Branquinha MH, Santos AL (2015). Candida parapsilosis (sensu lato) isolated from hospitals located in the Southeast of Brazil: Species distribution, antifungal susceptibility and virulence attributes. Int. J. Med. Microbiol. 305:848-859.

African Journal of

Microbiology Research

Related Journals Published by Academic Journals African Journal of Biotechnology African Journal of Biochemistry Research Journal of Bacteriology Research Journal of Evolutionary Biology Research Journal of Yeast and Fungal Research Journal of Brewing and Distilling