Storage of Ostrich Skin

Storage of Ostrich Skin Effects of preservation methods on skin structure, physical properties & microbial flora

A report for the Rural Industries Research and Development Corporation by Dr Christine A Lunam and Dr Kristy A Weir Flinders University July 2006 RIRDC Publication No 06/054 RIRDC Project No UF-9A

ii

© 2006 Rural Industries Research and Development Corporation. All rights reserved. ISBN 1 74151 317 0 ISSN 1440-6845 Storage of Ostrich Skin: Effects of preservation methods on skin structure, physical properties & microbial flora Publication No. 06/054 Project No. UF-9A The information contained in this publication is intended for general use to assist public knowledge and discussion and to help improve the development of sustainable industries. The information should not be relied upon for the purpose of a particular matter. Specialist and/or appropriate legal advice should be obtained before any action or decision is taken on the basis of any material in this document. The Commonwealth of Australia, Rural Industries Research and Development Corporation, the authors or contributors do not assume liability of any kind whatsoever resulting from any person's use or reliance upon the content of this document. This publication is copyright. However, RIRDC encourages wide dissemination of its research, providing the Corporation is clearly acknowledged. For any other enquiries concerning reproduction, contact the Publications Manager on phone 02 6272 3186. Researcher Contact Details Dr Christine Lunam Department of Anatomy & Histology Flinders University PO Box 2100 ADELAIDE SA 5001 Phone: 08 8204 4704 Fax: 08 8277 0085 Email: [email protected] In submitting this report, the researcher has agreed to RIRDC publishing this material in its edited form. RIRDC Contact Details Rural Industries Research and Development Corporation Level 2 15 National Circuit BARTON ACT 2600 PO Box 4776 KINGSTON ACT 2604 Phone: 02 6272 4819 Fax: 02 6272 5877 Email: [email protected]. Website: http://www.rirdc.gov.au Published in July 2006 Printed on environmentally friendly paper by Canprint

iii

Foreword The aim of this project is to promote the growth and economic sustainability of the Australian Ostrich Industry by optimising storage conditions of ostrich skins prior to tanning. Raw ostrich skins produced in Australia are stored for periods between a few weeks to one year prior to tanning. The development of optimal short and long-term storage conditions prior to tanning is crucial to achieve 1st and 2nd grade finished skins. Factors contributing to downgrading of the skins include dark patches, rough texture, lamination and reduced plumping of the feather follicles. These factors are considered to result from deterioration during storage. This project evaluates the effects of different storage conditions for raw ostrich skins on skin quality after tanning. Treatments addressed are bactericide/fungicide, storage temperature and storage duration. Hide salt is currently used as the main preservative of the skins during storage. However, concern is growing over the detrimental environmental impact of the high salt effluent released from tanneries. Consequently, the effects of storage of raw skins without salting on skin quality is assessed. A pilot study was undertaken to isolate microbial organisms on raw ostrich skins with different storage regimes. The ability of the isolates to downgrade the skins is discussed. This report also provides a detailed description of the structure of ostrich skin at both the macroscopic and microscopic levels, and discusses how the different storage regimes result in perturbation of the tissue. Also reported is the structure and distribution of filoplumes and bristle hairs in ostrich skin. This project was funded from RIRDC Core Funds which are provided by the Australian Government for the program area of New Animal Products. This report, an addition to RIRDC’s diverse range of over 1500 research publications, forms part of our New Animal Products R&D program, which aims to accelerate the development of viable new animal industries. Most of our publications are available for viewing, downloading or purchasing online through our website: • downloads at www.rirdc.gov.au/fullreports/index.html • purchases at www.rirdc.gov.au/eshop Peter O’Brien Managing Director Rural Industries Research and Development Corporation

iv

Acknowledgments

• Dr Peter McInnes for his support and encouragement throughout the entirety of this study. • Mr Geoff Lean, (Industry Partner) for his enthusiasm, the many fruitful discussions of the

work and for the numerous hours devoted to this project. Thank you Geoff for your valued advice on the many and varied technical aspects of the project, organising the slaughtering and purchase of skins as well as hands-on assistance with processing of the skins at Flinders University, at Bute Tannery and at the abattoirs at Pyramid Hill and Myrtleford.

• Mr Trevor Jones, (Industry Partner) manager and owner of Bute Tannery, South Australia, for

undertaking the tanning of the skins in the short-term storage experiment and assisting with grading of the crusts. We are grateful for access to your databank on skin storage and tanning procedures, for your advice on tanning and storage techniques and discussions on problems faced by the tanning industry, particularly the problems in achieving first grade ostrich and emu skins.

• PCA Hodgson Chemicals Pty Ltd, PCA, Clariant Division in Melbourne (Industry Partner) for

supplying the bactericides as well as other chemicals. In particular, we thank Mr Andre Mihelcic, Chief Technician at Clariant for undertaking the physical testing of the tanned skins as well as providing technical data on the chemicals and physical testing procedures.

• Mr Bruce Wetherall, Principal Medical Scientist in the Department of Microbiology and

Infectious Diseases at Flinders Medical Centre. Thank you Bruce for devoting much of your time and expertise in culturing the many swabs and identification of the isolates, photographing the isolates, conducting the gelatinase experiment as well as your advice in interpretation of the data.

• Dr John Plummer, Principal Medical Scientist in the Department of Anaesthesia and Intensive

Care, Flinders Medical Centre for conducting the statistical analyses of the data.

• Dr Jorge Ruiz, Ms Lauren Kingston and Ms Candice Thomson for technical assistance with the storage experiments and preparing the skins for light microscopy as well as quantitation of the histological sections. Thank you also Jorge for preparing many of the samples for scanning electron microscopy.

• Mr Michael Hastings for discussions of the work and for providing information on the

Australian Ostrich Industry and on the current export markets for ostrich skins.

• Mr Jean Marc Bonsujet, Prestige Leather Australia (Vic) Pty. Ltd. Wangaratta for tanning the skins from the long-term storage experiment and for grading the crusts.

• Staff at the abattoirs at Pyramid Hill and Myrtleford for their assistance during processing the

skins at slaughter.

• New Animal Products Research & Development Program, RIRDC for funding the work.

v

Contents

Foreword ............................................................................................................................................... iii Acknowledgments................................................................................................................................. iv Executive Summary ............................................................................................................................ vii 1. Introduction................................................................................................................................. 1

1.1 Production of first grade ostrich skins – industry overview ............................................... 1 1.2 Historical perspective ......................................................................................................... 1 1.3 Parallels with development of A-grade emu skin ............................................................... 2 1.4 Overall aim ......................................................................................................................... 2

2. Short-term storage: effects of temperature & bactericide ..................................................... 4 2.1 Summary............................................................................................................................. 4 2.2 Structure of ostrich skin – relevance to defects after tanning ............................................. 5

2.2.1 Collagen ................................................................................................................. 5 2.2.2 Elastic fibres........................................................................................................... 6 2.2.3 Storage of raw skins............................................................................................... 6 2.2.4 Filoplumes & bristle hairs...................................................................................... 6

2.3 Aims.................................................................................................................................... 6 2.4 Materials & methods........................................................................................................... 7

2.4.1 Sourcing & treatment of skins ............................................................................... 7 2.4.2 Assessment of skin quality..................................................................................... 8

2.5 Results................................................................................................................................. 9 2.5.1 Light microscopy ................................................................................................. 10 2.5.2 Effect of storage on thickness of raw skins.......................................................... 12 2.5.3 Ultrastructure of raw skins................................................................................... 15 2.5.4 Tear strength of ostrich crusts.............................................................................. 17

2.6 Discussion......................................................................................................................... 17 2.6.1 Storage treatment & tissue structure .................................................................... 17 2.6.2 Effects of bactericide pre-treatment & storage temperature on skin thickness ... 18 2.6.3 Tear strength of ostrich crusts – relationship to thickness of raw skins............... 19 2.6.4 Filoplumes & bristle hairs.................................................................................... 19

3. Long-term storage: effects of temperature, bactericide & salt ............................................. 21 3.1 Summary........................................................................................................................... 21 3.2 Background & aims .......................................................................................................... 22 3.3 Materials & methods......................................................................................................... 23

3.3.1 Sourcing & treatment of skins ............................................................................. 23 3.3.2 Assessment of skin quality................................................................................... 24

3.4 Results............................................................................................................................... 26 3.4.1 Light microscopy – tissue architecture ................................................................ 26 3.4.2 Scanning electron microscopy ............................................................................. 28 3.4.3 Tear strength of the tanned skins ......................................................................... 31 3.4.4 Skin grades........................................................................................................... 31 3.4.5 Distribution of filoplumes, bristle hairs & pinholes ............................................ 33

3.5 Discussion......................................................................................................................... 34 3.5.1 Storage treatment, tissue structure & skin thickness............................................ 34 3.5.2 Filoplumes, bristle hairs & pinholes .................................................................... 36 3.5.3 Feather follicles in tanned skin ............................................................................ 37 3.5.4 Tear strength ........................................................................................................ 37

vi

4. Pilot study of microbial flora on skins during storage.......................................................... 39 4.1 Summary........................................................................................................................... 39 4.2 Background & aims .......................................................................................................... 40 4.3 Materials & methods......................................................................................................... 41

4.3.1 Microbial cultures ................................................................................................ 41 4.3.2 Collagenolytic ability of microflora..................................................................... 41 4.3.3 Assessment of skin quality................................................................................... 42

4.4 Results............................................................................................................................... 43 Micrococcus (1) .................................................................................................................................... 43 Bacillus (1)............................................................................................................................................ 43 Micrococcus (1) .................................................................................................................................... 43 Bacillus (1)............................................................................................................................................ 43

4.4.1 Time course of microbial flora with different storage regimes............................ 45 4.4.2 Gelatin liquefication............................................................................................. 48 4.4.3 Potential of microbial flora to degrade the skins ................................................. 51

4.5 Discussion......................................................................................................................... 53 5. References .................................................................................................................................. 56

vii

Executive Summary Production of first grade ostrich skins Raw ostrich skins produced in Australia are stored for periods between a few weeks to one year. At present, 20% of the raw skins are tanned in Australia, whilst the remaining 80% are sold off-shore to Korea and South Africa. These skins are sent chilled and salted and are stored between two months and one year prior to tanning. Raw skins are sold ungraded off-shore and fetch a gross return of $US125-140. The average return for finished skins is $US240 minus the tanning costs of $US60, thus proving a premium of $US40-55 over selling as raw skins. Consequently as a result of the value-added return from sale of the finished skins significant gains are to be made by increasing the number of skins tanned on-shore. At present, only a few farms, representing 10-20% of Australia’s output, produce 90% of their raw skins as 1st or 2nd grade. Based on current trends in the Australian Ostrich Industry, which has moved from the breeder phase into a commercial industry, there will be a substantial increase in the number of raw skins produced per annum. Therefore, to promote the expansion of the ostrich skin industry, it is essential to maximise production of 1st grade skins demanded by local and overseas markets. Rationale for this study The development of optimal short and long-term storage conditions prior to tanning is crucial in achieving 1st and 2nd grade finished skins. Factors contributing to downgrading of the skins include dark patches, rough texture and poor plumping of the feather follicles. These factors are considered to result from deterioration during storage, and in particular to be due to damage inflicted on the collagen fibres by microbial organisms. However, the link between storage conditions of raw skins and skin quality after tanning is largely anecdotal. There is scant scientific data on the effects of different storage conditions and storage duration on skin quality in the ostrich and the emu. Hide salt is currently used as the main preservative of raw skins during storage. Concern is growing at local and state government levels over the potential environmental impact of high effluent salt released from tanneries. Consequently, the effects of storage without salting on skin quality were investigated. Aims & methodology The aim of this project was to promote the growth and economic sustainability of the Australian Ostrich Industry by optimising storage conditions of the raw skins. Anecdotal evidence suggests that cold storage, pre-treatment with a bactericide/fungicide and salting minimises skin deterioration during storage. This work aimed to evaluate the effects of these treatments on the structure of the raw skins and to examine the relationship between changes in structure and skin quality after tanning. A multidisciplinary approach was taken to investigate the effects of different storage regimes on skin quality. Skin quality was assessed by a detailed examination of the structure of the skin using light and scanning electron microscopy, measurement of skin thickness, tear strength and grading after tanning. Two storage durations were included; these were four and nineteen weeks. A pilot study was undertaken to investigate firstly the time course of microbial flora on raw skins during long-term storage and secondly, the potential of these organisms to degrade collagen. The project was designed to address the following questions.

viii

• What structures comprise ostrich skin? How do these structures give the skin its unique properties?

• What is the structure of filoplumes and bristle hairs in ostrich skin? What is the distribution of these structures in individual skins? What do the distributions of filoplumes and bristle hairs suggest about their genetic heritability?

• What changes occur within the skin during different storage regimes? How do these changes contribute to downgrading of the skin?

• How does temperature and bactericide/fungicide treatment affect skin quality during short-term storage?

• How does temperature, bactericide/fungicide treatment and salting during long-term storage affect skin quality?

• What is the relationship between the thickness of raw skin, skin thickness after tanning and skin quality?

• How does salting and temperature during storage affect the growth of microbial flora on the skins? What is the potential of the microbial flora to degrade collagen in ostrich skin?

Relationship between skin structure & skin quality Light and electron microscopy revealed that the surface of ostrich skin consists of a thin epidermis comprised of two to three layers of cells covered by keratin, the stratum corneum. In live birds keratin would provide protection from abrasion, assist in control of water loss and act as a physical barrier to invasion by bacteria and other microbial organisms. As the epidermis is removed during liming it is not a component of tanned skin. However, in raw skins it may act as a barrier to physical damage of the underlying dermis during handling and also to serve to inhibit colonisation of the connective tissue by microbial flora during storage. The main component of ostrich skin was the dermis comprised predominantly of collagen fibres. As in volant (flying) birds, the dermis consisted of two layers, a superficial thin grain and an extensive corium. Both these layers consisted of three-dimensional arrays of collagen fibres orientated perpendicular to one another and predominantly aligned parallel to the surface of the skin. Verhoeff and van Gieson staining revealed very few elastic fibres scattered between the collagen. These fibres were predominantly found either in small bundles deep within the dermis or associated with the attachment of smooth muscles to the feather follicles. The paucity of elastic fibres suggests that they minimally contribute to the elasticity of the skin and that the strength and flexibility of the skin is provided by the collagen fibres and by their cross-weave pattern within the dermis. Therefore it is the integrity of the collagen fibres that is crucial to skin quality. The finding that the grain formed an extremely thin superficial layer of the dermis is consistent with the skin of the emu. Similarly to emu skin, collagen fibres within the grain layer were more compact and thinner compared to those in the adjacent corium. In addition, the grain and corium layers were separated by a narrow band of loose connective tissue. This band of tissue is likely to provide little resistance to shearing forces during processing of the skins and would account for the lamination and loose grain defects prevalent in tanned skin of the ostrich and the emu. Both the band of loose connective tissue and the overlying grain layer contained numerous blood vessels. The high vascularity near the surface of the skin would explain why the skin is readily bruised during transport and handling of the ostriches prior to slaughter.

ix

Filoplumes & bristle hairs Macroscopically filoplumes in the ostrich resembled late immature filoplumes of the domestic chicken. Filoplumes consisted of a single rachis surrounded by two to 12 barbs extending eight to 12 millimetres in length. Barbules were present along the extent of the shaft of the barbs. Distally the barbules of each barb were slightly longer than those along the shaft giving a tuft-like appearance at the tips. Bristle hairs had a single distally tapered rachis extending three to five mm in length. Scanning electron microscopy revealed a cluster of two to four short barbs at the base of each bristle hair with an absence of barbs at the tip of the rachis. Four to six filoplumes were arranged in a semi-circular pattern at the base of the feather follicles in all skins. Bristle hairs were interspersed between the filoplumes. The filoplumes and bristle hairs at the base of the contour feathers were restricted to the caudal side of each feather follicle. In a live bird, the contour feathers are angled in a cranial to caudal direction thus they would obscure the filoplumes and bristle hairs at the base of the feathers. Both filoplumes and bristle hairs were present in the interfollicular space of many skins. Their distribution in individual skins ranged from sparse to covering the entire crown area. Filoplumes and bristle hairs had identical distributions for any given skin. This finding supports the notion that these traits may be genetically linked. As such, filoplumes and bristle hairs may be expressed from two separate gene sets or be different phenotypic expressions of the same pleiotropic gene set. In the current study skins were examined from different flocks separated by a period of two years. Our finding of these miniature feathers at the base of all skins examined, raises the possibility that all ostriches may have filoplumes and bristle hairs at the base of the follicles and that it is only the extent of their distribution between the follicles that is subject to genetic variation. After tanning the pattern of tiny discrete pinholes paralleled the distribution of filoplumes and bristle hairs found in individual skins. This indicates that the pinholes are the openings of the miniature follicles of the filoplumes and bristle hairs. In support of this notion, light microscopy revealed that the tiny follicles of the filoplumes and bristle hairs had a structure similar to the follicles of the contour feathers with an inner lining of epidermal cells surrounded by a band of collagen. This histology accounts for the discrete plumping of the skin immediately surrounding the pinholes at the base of the follicles of the contour feathers as well as the pinholes in the interfollicular space after tanning. Thus, the plumping of the follicles of the contour feathers is augmented by the presence of the pinholes at their base. Short-term storage: effects of temperature & bactericide The effect of storage temperature and bactericide pre-treatment on skin quality was examined following four weeks of storage. Changes in raw skin thickness or deterioration of the general tissue structure after storage using different treatment regimes were compared to the tear strength and thickness of ostrich crusts. Pre-treatment with bactericide Busan 85® prevented a change in thickness of the raw ostrich skins stored at either 4-6°C or 22°C for four weeks. Without bactericide pre-treatment cold stored skins became significantly thinner (p=0.05) whereas skins stored at room temperature showed a significant increase in thickness (p=0.049). This differential response in skin thickness with storage temperature may reflect a time course of degradation, with the rate of degradation inhibited at colder temperatures. Denaturation of the collagen fibres would likely result in a thicker connective tissue layer comprised of a decreased density of the collagen fibres as well as an increased diameter of individual fibres. As this change was prevented with bactericide treatment it is suggested that any loss in the integrity of the collagen fibres is likely to be caused by bacterial infection rather than lysosomal autolysis. Light microscopy revealed all skins had a similar histological appearance with no differences in general tissue architecture before and after storage. Viewed by scanning electron microscopy the

x

collagen fibres showed slight disruption in the alignment of their fibrils as well as increased separation of fibrils after storage. These changes were observed to a similar extent among the treatments. Loss of the ordered array of collagen fibrils after storage suggests some loss of integrity of individual collagen fibres. No differences were detected in tear strength of the crusts between the storage groups in either perpendicular (p=0.08) or parallel (p=0.6) orientations. Similarly, no differences in thickness of the ostrich crusts were found among the different treatment groups (p=0.42). The data indicate that bactericide pre-treatment is essential to prevent changes in the thickness of the raw salted ostrich skins stored for four weeks. Although microscopy revealed some deterioration of the raw salted skins following short-term storage without bactericide pre-treatment, this had no observable affect on skin quality, tear strength or thickness of the ostrich crusts. Therefore, salted skins may be stored either at room temperature or at 4-6°C for a period of four weeks without compromising skin quality after tanning. Long-term storage: effects of temperature, bactericide & salt The effects of temperature, bactericide pre-treatment and salting on the preservation of raw ostrich skins during storage for 19 weeks were determined. Changes in skin thickness and/or deterioration of the general tissue structure after storage were compared to both the tear strength and thickness of the tanned skins. Skin quality was determined by grading after chrome-tanning. The effects of different storage treatments on the structural and physical attributes of the skin were determined. Bactericide/fungicide pre-treatment with either Diamoll® C or sodium hypochlorite had no effect on the thickness of salted skins after two weeks of storage at 4-6ºC (p=0.26). In contrast, unsalted skins pre-treated with Diamoll® C and similarly stored for two weeks, were significantly thicker compared to skins pre-treated with sodium hypochlorite (p=0.047). Unsalted skins stored at 4-6°C became significantly thinner after 14 weeks storage (p=0.037). Salted skins stored at 4-6°C for 14 weeks became thicker by an average of 171 µm while salted skins stored at 22°C became thinner by an average of 934 µm. Light microscopy revealed no histological differences between the groups after two weeks of storage. After 14 weeks however, salted skins stored at 22ºC had greater inter-collagenous spaces within the corium layer compared to skins stored salted at 4-6°C. In contrast, the corium of skins stored unsalted for 14 weeks appeared as an irregularly stained mass with the spaces between the individual fibres often obliterated. Scanning electron microscopy revealed pitting of the keratin in all skins after 19 weeks of storage. Skins from all treatments showed splaying of collagen fibrils within the grain and corium layers. As the extent of splaying was highly variable within a given skin sample, it was not possible to relate storage treatment to the extent of loss of structural integrity of the collagen fibres. No differences were observed in tear strength among the treatment groups in either perpendicular (p=0.12) or parallel (p=0.59) orientations. After 19 weeks storage, no differences were observed in tear strength between skins cold-stored with and without salt or between salted skins stored at 4-6ºC and 22ºC. No differences were found in thickness of the tanned skins among the groups when measurements of skin samples taken parallel and perpendicular to the backbone were analysed jointly (p=0.14). No significant association was found between tear strength and thickness of the raw skins after 19 weeks of storage (p=0.12). Cold storage without salting for 19 weeks markedly decreased the grade of the tanned skins as a result of bacterial damage and flattened follicles. In contrast no difference in grading was observed between salted skins stored for 19 weeks at either 4-6ºC or 22ºC.

xi

The data suggests that salted skins may be stored for up to 19 weeks either at room temperature or at 4-6°C without compromising skin quality. In contrast, cold storage for 19 weeks in the absence of salt is insufficient in preventing significant skin damage. Pilot study of microbial flora on skins during storage Microbial flora were isolated from skins stored under different treatment regimes over a storage period of four months. The effects of hide salt, temperature, antimicrobial pre-treatment, and ultraviolet irradiation on the type and persistence of microbial flora were determined. As an indication of their potential to degrade collagen, representative micro-organisms isolated from the skins were subsequently assessed for their ability to liquefy gelatin in vitro. Selected isolates showing varying degrees of gelatinase activity were reapplied to ostrich skin swatches for 45 days to evaluate their ability to degrade collagen in situ. Skin deterioration was determined by qualitative assessment of general tissue structure and the organisation of collagen fibres using scanning electron microscopy. The work revealed clear differences between the types of micro-organisms colonising the salted and non-salted skins. Non-salted skins showed heavy growth of mixed Gram-negative bacilli. Gram-positive bacteria also colonised the non-salted skins but in fewer numbers. In contrast, no Gram-negative organisms were isolated from the salted skins. Staphylococcus, Micrococcus and Bacillus species were cultured from swabs of the salted and unsalted skins. In addition to these Gram-positive bacteria, two fungi, Zygomycete sp. and Alternaria sp. were isolated from the salted skins stored at 4-6°C. Gram-negative bacteria persisted in the non-salted skins throughout the entire storage period. Treatment with 1% sodium hypochlorite and ultraviolet radiation did not prevent the growth of microbial flora on these skins. In contrast, Gram-negative bacteria were never isolated from skins that were salted for the entire storage period. Furthermore, Gram-negative bacteria colonising the initially non-salted skins were not found beyond two weeks of salting. This inhibition of growth of Gram-negative bacteria with salting was independent of storage temperature. Assessment of the ability of selected Gram-positive and Gram-negative bacteria as well as yeast isolates to liquefy 15% porcine gelatin in vitro revealed varying degrees of gelatinase activity. Seventy-two percent of the selected isolates from non-salted skins and 33% of isolates from salted skins showed gelatinase activity. In addition, four out of five of the selected Gram-negative isolates liquefied gelatin. This supports the notion that Gram-negative bacteria are particularly effective at liquefying gelatin and that the microbial flora of non-salted skins are likely to possess greater proteolytic activity than that of salted skins. No correlation was found between the ability of a particular isolate to liquefy gelatin in vitro and their ability to degrade collagen in situ. In summary raw ostrich skins are colonised, in addition to fungi, by several species of Gram-positive and Gram-negative bacteria. Gram-negative bacteria were found exclusively on unsalted skins. Although isolates of either Gram-positive or Gram-negative bacteria were able to liquefy gelatin in vitro, only skins that were stored unsalted were downgraded as a result of bacterial damage. These findings suggest that bacteria responsible for collagen degradation in ostrich skin are Gram-negative and that their growth is inhibited by salting.

xii

Implications

• The skin of the ostrich and the emu are structurally very similar. In both species the organisation of the grain and corium layers and the numerous blood vessels just beneath the surface of the skin accounts for their susceptibility to lamination and bruising.

• The strength and flexibility of skin in the ostrich, as in the emu, is derived almost exclusively from the three-dimensional cross-weave arrangement of collagen fibres.

• Filoplumes and bristle hairs form a semi-circular pattern at the base of the feather follicles in every skin. Their density and distribution between the feather follicles is highly variable among individual birds.

• The data implies all ostriches farmed in Australia have filoplumes and bristle hairs at the base of their follicles and it is their distribution between the follicles that is subject to genetic variation.

• A time course of change in skin thickness occurs with storage duration. Initially the raw skins become thicker followed by eventual thinning. It is predicted that bacteria cause minor disruption of the collagen fibres within the first few days of slaughter. This is followed by dehydration with continued storage.

• Collagen fibres are highly resilient to disruption of their internal organisation. The ultrastructure of the raw skins, in particular collagen fibres, and tear strength after tanning are minimally affected by either storage duration up to five months, salting, storage temperature or bactericide/fungicide treatment.

• Salted skins may be either cold stored or stored at room temperature for four weeks without compromising skin quality after tanning. Bactericide/fungicide pre-treatment of salted skins prior to storage does not improve skin quality.

• For long term storage of several months salting is essential to prevent downgrading of the skins by bacterial damage. Bactericide/fungicide treatment of the raw skins is not effective in preventing bacterial damage with long term storage. For salted skins, cold storage provides no advantage over storage at room temperature.

• In the current study, there was no correlation between the thickness of the raw skins, skin thickness after tanning and skin quality.

• Several species of Gram-positive and Gram-negative bacteria were isolated from the skins. Gram-positive bacteria colonised salted skins stored at either room temperature or at 4-6°C. Salting totally inhibited the growth of Gram-negative bacteria. The data suggest that bacteria responsible for collagen degradation in ostrich skin are Gram-negative.

Recommendations Recommendations for storage are designed to achieve optimal quality of tanned ostrich skin.

• Hide salt is essential to prevent bacterial growth for storage beyond 24 hours. • Optimally skins should be tanned within 24 hours of slaughter. To avoid salting, skins should

be soaked in chilled water containing an antibacterial agent immediately after slaughter prior to transport to the tannery.

• Raw salted skins may be stored for up to four weeks without pre-treatment with bactericide/fungicide. For long-term storage, either pre-treatment with bactericide/fungicide, treatment with ultraviolet radiation or periodic soaking in sodium hypochlorite during storage is not effective in preventing bacterial growth.

• Dilute sodium hypochlorite is an effective anti-bacterial agent for short-term storage. This needs to be removed by soaking prior to tanning as it inhibits even uptake of dye in the finishing process.

• For salted skins there is no advantage in storage at 4-6°C compared to room temperature. This effect is independent of storage duration.

xiii

Future work

• Determine in all flocks within Australia the distribution of bristle hairs and filoplumes in ostrich skin by examination of the raw skins after slaughter as well as scoring of filoplumes in the live bird.

• Further classify the types of Gram-negative bacteria able to colonise raw ostrich skin. • Determine the ability of these Gram-negative bacteria to liquefy gelatin derived from ostrich

skin. • Re-apply the identified isolates to ostrich skin immediately after slaughter to accurately assess

the ability of the isolates to break down collagen in situ. The integrity of collagen fibres need to be examined by both scanning and transmission electron microscopy and the data obtained compared to skin quality and physical properties after tanning.

• Examine the effectiveness of potassium chloride as a preservative of raw ostrich and emu skin. At present potassium chloride is cost ineffective compared to hide salt (98% sodium chloride). In contrast to hide salt, which is highly detrimental to the environment, potassium chloride is considered not harmful to the environment and has been shown in other types of skins to be an effective preservative.

1

1. Introduction 1.1 Production of first grade ostrich skins – industry overview As an emerging industry, the Australian Ostrich (Struthio camelus) Industry faces the task of developing existing domestic and export markets for the skins. Improving the quality of tanned ostrich skins is essential for the sustainability of the industry. Downgrading of the skins results from several factors. These include scarring from cuts during aggressive behaviour, bruising during transport and processing, resulting in “loose grain” or lamination, as well as the presence of tiny pinholes between the feather follicles following removal of the filoplumes. Raw skins are also downgraded by bacterial damage during storage resulting in dark patches, rough texture and poor plumping following tanning. To increase the percentage of 1st grade skins produced in Australia, each of the factors that decrease skin quality need to be addressed. In recent years research has been directed to decreasing skin damage by altered husbandry practices such as de-clawing of ostriches (Glatz 2001) and emus (Lunam & Glatz 2000), and by an epidemiological study of the heritability of filoplumes in the Australian ostrich flock (Kim Bunter, personal communication 2003). However, there is scant scientific data on the effects of different storage conditions and duration of storage on skin quality. Seasonal influences on the breeding stock, the high cost of transport (AU$10 to AU$40 per bird, MacNamara et al. 2003) due to the long distances between the grower farms and the abattoir together with the high cost of slaughter (AU$90 to AU$100, MacNamara et al. 2003) result in the periodic slaughter of batches of ostriches. As there are few tanneries with the expertise to process the ostrich skins, the raw skins may be stored anywhere between a few weeks and several months. Consequently there is an urgent need to evaluate the effects of storage on skin quality. 1.2 Historical perspective Data relating to current trends in the number of ostriches and profits in the ostrich industry were provided by Mr Michael Hastings – President Australian Ostrich Association of Australia - (personal communication 2005) unless otherwise stated. The original Australian ostrich line is termed by the industry as the Australian Grey. This genetic line was imported from South Africa in the late 1860's. The breed types currently farmed in Australia are a composite of the South African Black, Zimbabwe Blue, Kenyan Red and Australian Grey. Ostriches farmed in Australia are therefore genetically different to those in South Africa.

Until recently a paucity of expertise existed in Australia in the tanning of ostrich skins with the result that 80% are sold off-shore as raw skins to Korea and South Africa. This has placed the industry in a vulnerable position in the event of disease being introduced into the Australian ostrich flock as subsequent quarantine of either the meat or raw skins by other countries would seriously jeopardise the viability of the industry. Ostriches are capable of carrying Newcastle disease. An outbreak of Newcastle disease in poultry in 2002 resulted in a ban on the export of ostrich meat. Consequently less ostriches were slaughtered with a resultant decrease in the supply of skins (MacNamara et al. 2003). The raw skins sold off-shore fetch ungraded a gross return of US$125-140. These skins are sent chilled and salted and are stored between two months and one year prior to tanning. The remaining 20% of ostrich skins are tanned and finished within Australia. The main importers of the finished skins are Japan, USA, Italy, India and China. Of the 20% some are made into leather goods on-shore whilst the rest are sent to overseas markets to be made into finished products. Currently finished skins

2

fetch between US$14-19 per square foot with the average skin size approximating 15½ square feet. The average return is thus estimated per tanned skin of US$240 minus tanning costs of US$60 providing a premium of US$40-55 over selling as raw skins. Therefore significant gains are to be made by increasing the number of skins tanned on-shore. This would significantly increase the value-added return from ostrich skin sales. Since the early stages of the industry in Australia, the number of farms has decreased but the farm size has increased. In 1995/6, there were 2000 farms with an average of 10-15 birds each. This was at the breeder phase. The industry has now moved to the commercial phase with 100 farms holding between 500-4000 birds per farm and farms being specialised to different aspects of the production cycle (breeding, chick rearing to 90 days, grower farms 90 days to 300 days). This reduction in the number of farms with increased stock is a reflection of the rationalisation of the ostrich industry during the transition from the breeder phase into a commercial industry. In 2005 it is expected that 15000 ostriches will be slaughtered; which is at the trough of this emerging industry. It is estimated on changing trends in the industry that in 2006 this number will increase to 25000, and will continue to increase again in 2007. Both the domestic and export markets demand 1st grade skins. In recent years there has been a marked increase in skin quality. In 2003, the majority of skins produced in Australia were 3rd and 4th grade (MacNamara et al. 2003). In 2005 a few farms, representing approximately 10 to 20% of Australia’s output, produce 90% of their raw skins as 1st or 2nd grade. This translates to 60% 1st and 2nd grade once tanned. This increase in skin quality is due in part to altered husbandry practices such as de-clawing and decreasing the average age of slaughter thereby reducing the potential for skin damage during the grower phase. In early 2004 the average age at slaughter was 12 months whereas in 2005 some birds are being slaughtered at 10 months of age at 95-100kg live weight. Based on current trends in the Australian Ostrich Industry there will be a substantial increase in the number of raw skins produced per annum and the majority of these raw skins will no longer be sent off-shore but tanned within Australia. To maintain its viability the Australian ostrich industry needs to continue to improve the quality of the tanned skins. 1.3 Parallels with development of A-grade emu skin The Australian Emu Industry faces similar challenges to that of the ostrich industry in that there is a need to improve the quality of tanned skins destined for the local and export markets. The skin of the emu (Dromaius novaehollandiae) faces similar problems to ostrich skin in that both are prone to bruising, lamination and loss of grain. Anectodal evidence suggests that these problems are exacerbated in the emu due to the thin nature of the skin. Recent work in our laboratory aimed to optimise emu skin quality by investigating the effects of different storage conditions on skin structure and physical properties. The current study of ostrich skin provided a unique opportunity to undertake experiments in tandem with that of the emu. The advantage of this parallel approach was that we could compare and contrast the structural and physical differences between ostrich and emu skins as well as determine the effects of different storage conditions on the skins. By this means both skin types serve as a control for the other, thus providing the opportunity to pinpoint causes associated with downgrading of each skin type. 1.4 Overall aim The aim of this project was to promote the growth and economic sustainability of the Australian Ostrich Industry by optimising storage conditions of raw skins prior to tanning. Anecdotal evidence suggests that cold storage, pre-treatment with a bactericide/fungicide and salting results in optimal skin quality. The work aimed to investigate the effect of these treatments on skin quality. Skin quality was assessed using a multidisciplinary approach that included skin morphology at the light and electron microscope levels as well as physical testing and professional grading of the skins after

3

tanning. Two storage durations were investigated; these were four and nineteen weeks. We also undertook a pilot study in collaboration with the Department of Microbiology and Infectious Diseases at Flinders Medical Centre to investigate the microbial flora on the raw skins during long-term storage, and the potential of these organisms to degrade collagen. Hide salt is currently used as the main preservative of the skins during storage. Concern is growing at local and state government levels over the potential environmental impact of the high effluent salt released from tanneries. Consequently the effects of storage without salt on skin quality were included in this study. Data generated from this work will increase our knowledge of the structure of ostrich skin and provide information on the effects of different storage conditions on skin quality. It is anticipated that application of the data will result in greater economic returns from the increased sale of 1st grade ostrich and emu skins.

4

2. Short-term storage: effects of temperature & bactericide 2.1 Summary The effects of storage temperature and bactericide treatment on skin thickness after four weeks of storage on raw salted ostrich skins were determined. Thickness measurements were made on histological sections and confined to the grain and corium layers. General tissue architecture was assessed by light and scanning electron microscopy. Changes in either skin thickness or deterioration of the general tissue structure after storage was compared to the tear strength and thickness of the ostrich crusts. Pre-treatment with bactericide Busan 85® prevented a change in thickness of the raw ostrich skins stored at either 4-6°C or 22°C. Without bactericide treatment cold stored skins became significantly thinner (p=0.05) whereas skins stored at room temperature showed a significant increase in thickness (p=0.049). Investigation of the effect of treatments on skin thickness after covariance adjustment for initial differences in thickness, revealed a significant (p=0.027) interaction between temperature and bactericide. At 4-6°C, the average change in skin thickness was 356 µm greater without bactericide, compared to skins similarly cold stored with bactericide pre-treatment. In contrast, skins stored at 22°C had an average change in thickness of 443 µm greater without bactericide treatment compared to skins similarly stored after treatment. Light microscopy revealed all skins had a similar histological appearance with no differences in general tissue architecture before and after storage. Although some epidermal cells showed morphological changes indicative of degeneration during storage, storage had no apparent effect on the histology of the connective tissue. Scanning electron microscopy revealed an ordered array of compact fibrils within each of the collagen fibres before storage. After storage, some collagen fibres showed slight disruption in the alignment of their fibrils as well as increased separation of fibrils. These changes were observed to a similar extent among the treatments. In addition, pitting of the keratin was found in some skins from each of the groups before as well as after storage. No differences were observed in tear strength of the crusts between the groups in either perpendicular (p=0.08) or parallel (p=0.6) orientations. Similarly, no differences in thickness of the ostrich crusts were found among the different treatment groups. Thickness of the ostrich crusts was similar between the treatments (p=0.42). The data indicate that bactericide treatment is essential to prevent the increase in the thickness of the salted ostrich skins stored for four weeks at room temperature. Denaturation of the collagen fibres would likely result in a thicker connective tissue layer comprised of a decreased density of the collagen fibres as well as an increased diameter of individual fibres. As this change is prevented with bactericide treatment it is suggested that any loss in the integrity of the collagen fibres is likely to be caused by bacterial infection rather than lysosomal autolysis. Loss of the ordered array of collagen fibrils after storage suggests some loss of integrity of individual collagen fibres. Quantitation of the diameter of the collagen fibres is necessary to determine whether the increase in skin thickness observed after storage at room temperature without bactericide resulted from a greater loss in the integrity of the collagen fibrils compared to the other treatments. Filoplumes and bristle hairs were observed in all skins forming a semi-circle caudal to the base of the feather follicles. Some skins had numerous filoplumes and bristle hairs between the follicles. The individual variation in the number and pattern of filoplumes and bristle hairs in the skins supports

5

suggestions within the Ostrich Industry of a genetic heritability of these traits. Further, as filoplumes and bristle hairs had identical distributions for a given ostrich skin, we suggest these traits may be genetically linked. For individual skins, the pattern of distribution of these hair-like structures was similar to the pattern of pinholes after tanning. This indicates that these are “true” pinholes resulting from removal of the filoplumes and bristle hairs. In summary the data indicate that bactericide pre-treatment is essential to prevent changes in the thickness of the raw salted ostrich skins stored for four weeks. Although some minor deterioration of the salted skins had occurred following short-term storage without bactericide pre-treatment, this had no observable affect on either the tear strength or thickness of the ostrich crusts. Therefore, salted skins may be stored either at room temperature or at 4-6°C for a period of four weeks without compromising skin quality after tanning. Filoplumes and bristle hairs were identified in the ostrich skins and their macroscopic and microscopic features described. 2.2 Structure of ostrich skin – relevance to defects after tanning Understanding the structure of ostrich skin and how processing techniques alter its structure is crucial to producing 1st grade tanned skins. However, surprisingly little is known regarding the structure of ratite skin. An extensive literature search revealed only four articles dealing with ostrich skin structure. Lange (1929) and Bezuidenhout (1999) described the organisation of collagen fibres within the dermis of ostrich skin. They reported these were predominantly orientated perpendicular to each other and run parallel to the surface of the skin. The remaining two articles (Menon et al. 1996; Menon & Menon 2000) examined the histology of the epidermis. As the epidermis is removed during liming it is not a component of tanned skin. However the integrity of the epidermis may be important prior to tanning, as it may act as a barrier to physical damage of the underlying dermis during handling and also serve to inhibit colonisation of the connective tissue by microbial flora during storage. Prior to work in our laboratory, there was a paucity of information on the structure of emu skin. Two studies have briefly examined the histology of emu skin (Frapple et al. 1997; Peters 1994). Both studies describe a three-dimensional arrangement of collagen fibres that was similar to the distribution of collagen in ostrich skin. Recent work in our laboratory revealed the epidermal and dermal layers of emu skin have a similar arrangement to other avian species (Weir & Lunam 2004). Furthermore, the thin vascular grain layer was separated from the thicker corium layer by a loose region of connective tissue containing numerous capillaries (Weir & Lunam 2004). We argued that this arrangement of connective tissue within the dermis is consistent with lamination and loose grain defects known to be prevalent in tanned emu skin. The detailed organisation of the dermis of ostrich skin is not known. Anecdotal evidence suggests bruising of the skin during both transport of the ostriches to the abattoir and handling prior to slaughter is associated with lamination after tanning. This pattern of damage suggests a tenuous attachment of a highly vascularised grain layer to the underlying corium exists in the ostrich as in the emu. 2.2.1 Collagen Collagen is a major constituent of skin. The main types of collagen in skin are types I and III. These are known as fibrillar collagens as each collagen fibre is an aggregation of many collagen fibrils aligned parallel to one another. Each collagen fibril is in turn composed of a staggered array of collagen molecules (filaments) consisting of a triple-stranded helix formed by three collagen polypeptide chains rich in proline and glycine. Covalent cross-links bond the filaments. Viewed with the transmission electron microscope each collagen fibril has a cross-banding pattern that is due to the staggered repeating array of filaments. Fibril-associated collagens are thought to link the fibrils together thereby stabilising each collagen fibre. This highly ordered arrangement imparts considerable tensile strength to the collagen fibres (Alberts et al. 2002). Further, the degree of tensile strength is

6

dependent on the arrangement of collagen fibres within the dermis. A study of bovine leather revealed that the size and extent of interweaving of collagen fibres are directly related to the tensile and tear strength of the leather (Zapletal et al. 1996). The integrity of collagen, which forms 98% of the fibre weight in tanned skin (Ding 2004), is crucial to skin quality. During the tanning process the collagen fibres are “opened up” by treatment with proteolytic enzymes in a process known as bating that removes many of the interfibrillar proteins. An electron microscope study on the effect of bating conditions on goat skin demonstrated that prolonged bating with high concentrations of the enzyme pancreatin caused excessive disruption of the fibrils resulting in damage to the collagen fibres (Ding 2004). Therefore, during storage of the raw skins, excessive proteolysis of the collagen either by lysosomal autolysis or bacterial proteolytic enzymes could cause disruption to the fine structure of the collagen fibres. 2.2.2 Elastic fibres Elastic fibres form a network within the skin. Mature elastic fibres are composed predominantly of the protein elastin, and similarly to collagen fibres they are rich in proline and glycine that are organised into cross-linked helices. In contrast to collagen fibres however, they have little tensile strength but rather provide elasticity to the skin (Alberts et al. 2002). Elastic fibres are sparse in vertebrate skin and contribute approximately 1.5% of the connective tissue in emu skin (Weir 2004). It is well established that elastic fibres are present after tanning. Contrary to previous reports that tanning renders elastic fibres brittle, a recent study in goat skin reported bating under normal conditions caused minimal perturbation to the ultrastructure of elastic fibres (Ding 2004). Consequently Ding suggested that elastic fibres may be a minor contributor to the suppleness of tanned skin. 2.2.3 Storage of raw skins Ostrich skins tanned within Australia often need to be stored for short periods up to four weeks prior to tanning. Therefore, there is a need for development of effective short-term storage techniques. Salting is a commonly employed preservation method for ostrich skins produced in Australia. Two additional methods commonly used within the industry are pre-treatment with bactericide/fungicide and chilling. However, it is not known whether bactericide/fungicide pre-treatment and/or cold storage provide superior preservation compared to storage in hide salt alone at ambient temperatures. 2.2.4 Filoplumes & bristle hairs Filoplumes and bristle hairs are tiny hair-like feathers that have been identified in several species of bird (Lucas & Stettenheim 1972). However, both their macroscopic and microscopic structure vary between different avian species. These hair-like structures are found at the base of the contour feathers and are thought to convey tactile information to nearby sensory receptors, the Herbst corpuscles. Although Lucas and Stettenheim reported filoplumes are not present in ratite skin, growers, tanners and producers report both filoplumes and bristle hairs occur in the Australian ostrich flock. Removal of the filoplumes and bristle hairs results in tiny discrete pinholes in the tanned skin. The presence of these pinholes at the base of the feather follicles does not affect skin quality, however their occurrence between the feather follicles significantly downgrades the skins. 2.3 Aims

7

The aim of this work was to determine the histological structure of ostrich skin by light and scanning electron microscopy and to assess the effects of different short-term storage treatments on the structure of raw skin and tear strength after tanning. By this means we aimed to determine short-term storage conditions for raw skins that cause minimal perturbation of tissue structure without compromising tear strength. The results from this work will be compared to those from a parallel study of the effect of short-term storage on the structure and physical properties of emu skin. As there is no information available on the histology of filoplumes in ostrich skin, we aimed to determine their microscopic as well as macroscopic structure in the ostrich. The incidence and distribution of filoplumes in the Australian flock is determined by examination of the skin on the live bird. This makes it difficult to determine the exact distribution in any given skin. We aimed to examine each raw skin before storage to determine the distribution of both filoplumes and bristle hairs. The distribution of these hair-like feathers on each skin will be compared to the distribution of pinholes after tanning. Information on the structure and distribution of filoplumes and bristle hairs in individual skins may promote the development of strategies for their eradication from affected flocks, thus optimising the quality of Australian ostrich skins. 2.4 Materials & methods 2.4.1 Sourcing & treatment of skins Twenty skins from adult (14 months of age) ostriches of both sexes were purchased at a commercial abattoir at Pyramid Hill. Ostriches were electrically stunned and killed by bleeding from the carotid artery. Skins were dry-plucked before removal from the carcass. They were then immediately placed into chilled water for ten to fifteen minutes. The skins were flayed to remove as much fat as possible and re-immersed in clean chilled water containing hide salt (98% sodium chloride, 1% boric acid and 1% sodium fluoride; Cheetham Salt Ltd) for twenty minutes. Samples were then excised from the left and right upper belly flaps from each skin and fixed by immersion in either 3% glutaraldehyde in 0.1M phosphate buffer for electron microscopy or in a solution of 2% formaldehyde and 15% saturated picric acid in 0.1M phosphate buffered saline (PBS, pH 7.3) for light microscopy. At the time of sampling, each skin was coded by hole-punching in the region of both the left and right leg. Twelve skins were cut in half along the midline and each skin-half allocated to a different treatment group so each skin-half served as control for the other half. Eight whole skins were allocated to one of the four treatment groups (Table 2.1). Each skin or skin-half was liberally coated with hide salt and each group of skins stacked on separate pallets with the dermal side facing upwards. They were then transported in cold storage to Flinders University. On arrival at Flinders University skins were re-salted and stored for four weeks at either room temperature or 4-6 °C. As a check for possible cross-contamination from other skins on the pallet, two whole skins each from Groups two and four were placed on separate pallets (two skins per pallet) and stored in a different cold room from the half-skins of the same treatment group. Data loggers (T-TEC®, Adelaide, SA) placed in each of the storage rooms recorded the temperature for the four weeks duration. At the end of the storage period, tissue samples were again taken from the upper belly of each whole and half-skin and processed for both light and electron microscopy. All whole and half-skins were then sent to Bute Tannery, SA. After chrome tanning samples were taken from each whole and half-skin from the upper belly flaps and sent to PCA Hodgson Chemicals Pty Ltd, Melbourne for measurement of tear strength.

8

Table 2.1 Treatment groups Group 1 Immersed in cooled hide salt solution (25 g/L) for 30 minutes, stored at 22°C Group 2 As Group1 stored at 4-6°C Group 3 Immersed in cooled hide salt solution (25 g/L) and §Busan 85® (1.5g/L)

bactericide solution for 30 minutes, stored at 22°C Group 4 As Group 3 stored at 4-6°C

§The bactericide Busan 85® (Buckman Laboratories) is a broad spectrum organo-sulfur compound that reputedly minimises damage to collagen fibres. 2.4.2 Assessment of skin quality To assess the effects of temperature and bactericide treatment during four weeks of storage on salted raw ostrich skins, thickness measurements were made on histological sections of the skins before (at slaughter) and after storage. As tanned ostrich skin is comprised exclusively of the grain and corium layers measurements were confined to these layers. Skin quality was also assessed by comparing the general tissue architecture at both the light and ultrastructural levels between the time at slaughter and following four weeks of storage. For each whole or half-skin, any change in thickness or any deterioration of the general tissue structure after storage observed by microscopy was compared to the tear strength of the ostrich crusts. 2.4.2.1 Light microscopy & skin thickness After fixation at 4°C for a minimum of five days, segments of skin (1 cm2) were cut from the original samples. To remove the fat the tissue was cleared in dimethylsulphoxide (DMSO) for 1.5 hours. Samples were then dehydrated through increasing concentrations of ethanol, followed by chloroform overnight before being embedded in paraffin wax. A minimum of twelve, transverse 7 µm-thick sections per sample were cut using a rotary microtome (Leica RM 2135, Leica Microsystems Pty Ltd, Nussloch, Germany) and the medial and lateral edges of the sections were recorded. To avoid over estimation of skin thickness from obliquely cut sections, care was taken to ensure the skin surface in the microtome chuck was oriented perpendicular to the knife-edge. Sections were then mounted on chrome alum gelatin-coated slides. Sections were stained with either haematoxylin and eosin for the investigation of general tissue structure, or Verhoeff and van Gieson for differentiation of collagen and elastic fibres. Skin thickness measurements were made on the haematoxylin and eosin stained sections only. Grain and corium layers are the common use industry terms that refer to the stratum superficiale and stratum compactum respectively, as described by Clark (1993). Thickness of the grain and corium layers was measured with a X4 objective lens using an Olympus EHT light microscope with a calibrated graticule eyepiece. To avoid bias in selection of the region of skin to be measured, the first six sections from each tissue block were measured 4 mm from the left-hand edge of the section. Where the predesignated region was folded or tissue damaged during sectioning, the measurements were taken immediately adjacent to the damaged region. Skin thickness data were analysed by random effects regression. 2.4.2.2 Scanning electron microscopy The tissue was fixed by immersion in a solution of 3% glutaraldehyde in 0.1M PBS. After fixation at 4°C for four hours skin samples were washed three times in 0.1M PBS and stored in 0.1M PBS containing 0.1% sodium azide at 4°C. Samples of skin (0.2 cm2) were cut from the original samples with a sharp sterile surgical blade and post fixed in 1% osmium tetroxide in 0.1M PBS for one hour.

9

The samples were then washed three times in tap water (five minutes each) and dehydrated through 75%, 85%, 95% and 100% ethanol (two changes) with agitation for five minutes each. All samples were then critical point dried (Emscope CPD 750, Kent, UK), mounted on standard pin type SEM mounts (ProSciTech, QLD) and gold-coated using an SEM autocoating unit (Polaron Equipment Ltd. Watford, UK). Samples were examined using a Siemens Autoscan SEM (ETEC Corporation) and images photographed using Ilford FP4 125 (6 x 7 cm) roll film (Ilford Imaging UK Limited, England). Film was processed using Ilfosol 2 developer (Ilford Imaging UK Limited, England) and Ilford rapid fix (Ilford Imaging UK Limited, England). All negatives were scanned at 600 dpi to produce a digital image. 2.4.2.3 Tear strength of ostrich crusts Tear strength measurements were made using an Instron 4302 (Instron Corporation, Canton, USA). Four repeat measurements of tear strength were recorded perpendicular and parallel to the spine for each of the 32 skin samples giving a total of 64 measurements for each direction. Differences in tear strength between the four storage groups were analysed by random intercept regression taking into account the paired half-skins. 2.5 Results The temperature of each of the storage rooms remained relatively constant throughout the storage period (Figure 2.1).

Figure 2.1 This graph is representative of the temperature recorded over the four weeks of storage. The black line indicates the temperature of the storage room for Groups one and three. The red line shows the temperature of the storage room for skins from Group two. The blue line shows the temperature of the storage room for skins from Group four. The green line shows the temperature of storage room containing four whole skins (two each from Groups two and four).

10

2.5.1 Light microscopy 2.5.1.1 Tissue architecture The histology of the skin is shown in figures 2.2 and 2.3. These following features were present in all skins, both at slaughter and after storage. Light microscopy revealed a thin epidermis consisting of two to three layers of epidermal cells covered by a cornified layer, the stratum corneum. The dermis was comprised of the grain and corium layers. These layers consisted of a three-dimensional array of collagen fibres with most fibres predominantly orientated parallel to the surface. The grain layer was highly vascularised and separated from the corium by a thin band of loose connective tissue. Blood vessels, small nerve bundles and small pockets of adipose tissue were observed within the dermis. Smooth muscle bundles (Mm. pennarum), which serve to move the feathers, were observed within the corium. Herbst corpuscles were occasionally found near the epidermal sheath of the feather follicles. Verhoeff and van Gieson staining revealed a few elastic fibres within the corium layer as well as elastic tendons of the Mm. pennarum (Figure 2.3B). Epidermal cells showed morphological changes indicative of degeneration during storage. As a general rule, the epidermal cells stained more densely after storage compared to tissue fixed at the time of slaughter. Furthermore, pyknotic nuclei were observed only in epidermal cells after storage. Sebokeratinocytes, which were clearly visible as having a rounded nucleus surrounded by a clear halo of cytoplasm prior to storage were not distinguishable from other epidermal cells after storage. Storage had no noticeable effect on the architecture of the dermis.

Figure 2.2 Light micrograph showing the superficial region of ostrich skin sampled at the time of slaughter. sc, stratum corneum of the epidermis. Layers of epidermal cells appear as a dark granular band beneath the stratum corneum. Immediately beneath the epidermis lies the dermis consisting of collagen fibres predominantly orientated parallel to the surface. A few collagen fibres can be seen running perpendicular to the surface. In this micrograph only the grain and part of the corium layers of the dermis are visible. Separating the grain and corium layers is a thin layer of loose connective tissue containing numerous blood vessels. Compared to the corium, the grain layer is relatively thin, well vascularised and composed of compact collagen fibres. Red blood cells within capillaries, small veins and arteries are seen as clusters of black dots in the grain and corium layers.

11

Figure 2.3 Light micrographs of ostrich skin. (A) h, Herbst corpuscle; fs, follicle sheath; fp sheath of filoplume. Both the Herbst corpuscle and filoplume lie close to the feather follicle. The epidermis is visible on the right hand-side of the micrograph. Haematoxylin and eosin. (B) Deep layers of the corium. a, adipose tissue; pm, two Mm. pennarum (smooth muscles) running parallel in the corium; e, elastic fibres of a Mm. pennarum. Elastic fibres (arrows) are visible among the dense collagen fibres. Verhoeff and van Gieson. Scale bars are 400 µm. 2.5.1.2 Filoplumes & bristle hairs Light microscopy A small follicle sheath (50 µm in diameter) presumably, that of either a filoplume or bristle hair was occasionally observed lying within 500 µm of the sheath of a feather follicle (Figure 2.3A). Although considerably smaller than the adjacent feather follicle these tiny follicle sheaths had a similar histology to the large feather follicle with an inner lining of epidermal cells encircled by a dense band of collagen fibres. Macroscopic appearance Four to six filoplumes, with one to two bristle hairs interspersed between them, formed a semi-circular pattern on the caudal side of the feather follicles of all skins (Figure 2.4). Individual filoplumes and bristle hairs were frequently found between the feather follicles. The number of bristle hairs and filoplumes in the interfollicular space varied among the skins; ranging from sparse to covering the entire crown area. Bristle hairs were identified by a single distally tapered rachis extending three to five mm in length. Scanning electron microscopy revealed a cluster of two to four short barbs at the base with an absence of barbs at the tip of the rachis (Figure 2.5). Filoplumes consisted of a single rachis surrounded by two to 12 barbs extending eight to 12 millimetres in length. Barbules were present along the extent of the shaft of the barbs. Distally the barbules of each barb were slightly longer than those along the extent of the shaft giving a tuft-like appearance at the tips.

12

Figure 2.4 Raw salted ostrich skin. Filoplumes (arrows) are arranged in a semicircle on the caudal side of two feather follicles. Interspersed between the filoplumes are bristle hairs (arrowheads). A filoplume (arrow indicates shaft) and a bristle hair (arrowhead at shaft) can be seen in the inter-follicular space. Filoplumes and bristle hairs were found at the base of the majority of feather follicles. Crystals of hide salt are visible as white material on the skin surface. Scale bar is 1 cm.

Figure 2.5 Scanning electron micrograph showing a bristle hair. A single barb tapers distally with an absence of barbules along its length. A few short barbs are present at the base of bristle hair (arrow). Scale bar is 500 µm. 2.5.2 Effect of storage on thickness of raw skins Prior to storage there were no differences in skin thickness between the groups (p=0.38, Table 2.2, Figure 2.6). Skins stored for four weeks at 4-6°C without bactericide treatment became thinner (p=0.05) whereas skins similarly cold stored for four weeks with bactericide showed no change in thickness (p=0.95). Skins stored at 22°C without bactericide treatment became significantly thicker (p=0.049). In contrast, skins stored at 22°C with bactericide treatment showed no significant change in thickness (p=0.79). Post thickness values are given in Figure 2.7. Investigation of the effect of treatments on skin thickness after covariance adjustment for initial differences in thickness, revealed a significant (p=0.027) interaction between temperature and

13

bactericide. This is reflected in the means of post - pre differences in thickness shown in Table 2.3. At 4-6°C the average change in skin thickness was 356 µm greater in the absence of bactericide compared to skins similarly cold stored with bactericide treatment. In contrast, skins stored at 22°C had an average change in thickness of 443 µm greater without bactericide treatment compared to skins similarly stored after treatment. The post-storage minus pre-storage thickness among the groups are illustrated in Figure 2.8. Table 2.2 Effects of storage treatment on skin (grain + corium) thickness (μm)

Temperature No bactericide Bactericide Pre-storage 4-6°C 1798 ± 467 (165) 1920 ± 313 (111) 22°C 1715 ± 329 (117) 2017 ± 537 (190) Post-storage 4-6°C 1431 ± 392 (138) 1912 ± 589 (208) 22°C 2211± 699 (247) 2071 ± 437 (155)

Means ± SD (SEM) Table 2.3 Skin (grain + corium) thickness (μm). Post – pre storage

Temperature No bactericide Bactericide 4-6°C -367 ± 429 (152) -11 ± 421 (149) 22°C 497 ± 591 (209) 54 ± 570 (201)

Mean change ± SD change (SEM change)

0

1000

2000

3000

4 °C 22 °C 4 °C 22 °C

Thic

knes

s (µ

m)

No Bactericide Bactericide

Figure 2.6 Box plot showing thickness of the ostrich skin (grain + corium) prior to storage

14

0

1000

2000

3000

4 °C 22 °C 4 °C 22 °C

Thic

knes

s (µ

m)


Figure 2.7 Box plot showing effect of storage treatment on skin thickness

-1000

-500

0

500

1000

4 °C 22 °C 4 °C 22 °C

Cha

nge

in T

hick

ness

(µm

)


Figure 2.8 Box plot showing post minus pre-storage thickness

15

2.5.3 Ultrastructure of raw skins Tissue from all groups prior to storage displayed a similar architecture. Although some differences were observed after storage they were not confined to any individual treatment group. The organisation of the keratin ranged from neat overlapping flakes (Figure 2.9A) to significant sloughing of the flakes. At higher power there was clear pitting of the keratin surface (Figure 2.9B). Sloughing of the stratum corneum and pitting of the keratin was observed in skins from each of the groups before and after storage.

Figure 2.9 Scanning electron micrographs of the stratum corneum. (A) The keratin is arranged as overlapping flakes. (B) Higher magnification shows distinctive pitting on the surface of the keratin. Scale bars are 10 µm.

Collagen fibres were predominantly orientated parallel to the surface, with few fibres occasionally running either tangential or perpendicular to the epidermis (Figure 2.10). The relatively thin grain layer, lying immediately deep to the epidermis, was readily distinguishable from the adjacent corium. Often a narrow space separated the grain and corium layers. Although considerably larger than fibres in the adjacent grain layer, the collagen fibres within the corium appeared to vary markedly in both cross–sectional diameter and shape (Figure 2.10). Whereas collagen fibres within the corium were separated by shrinkage spaces, fibres of the grain layer were densely packed with few discernable spaces between them. Fat globules were distributed in all regions of the tissue. They were most often observed trapped in the spaces between the collagen fibres within the corium. High magnification revealed the ultrastructure of the individual collagen fibres (Figure 2.11). These are often referred to as collagen fibre bundles in the general literature. Each collagen fibre consisted of numerous fibrils measuring approximately 300 nm in diameter. The fibrils were tightly packed, running parallel to one-another along the length of the fibre. After storage, some collagen fibres showed slight disruption in the alignment of the fibrils as well as increased separation of fibrils. The changes in the organisation of the fibrils were observed in all treatments.

16

Figure 2.10 Scanning electron micrograph of the superficial layers of ostrich skin. The epidermis, grain and corium layers are clearly distinguishable. sc, stratum corneum. The cellular epidermis is masked by the undulating arrangement of the overlying stratum corneum. Collagen fibres within the grain layer appear smaller and more densely packed compared to collagen fibres within the corium. The corium extends for several hundred µm beyond this field of view. The majority of collagen fibres are orientated perpendicular to the cut surface and are seen in cross-section. Fat globules (arrows) are present in the spaces between the collagen fibres. Scale bar is 50 µm.

Figure 2.11 Scanning electron micrographs of collagen fibres within the corium. Each collagen fibre is comprised of a densely packed array of numerous collagen fibrils aligned parallel to one-another. (A) Three collagen fibres. The fibre extending from the lower left to the upper right of the micrograph is running perpendicular to the fibre exiting the micrograph at the lower right hand side. (B) A single collagen fibre showing fibrils cut in cross-section. Scale bars are 5 µm.

17

2.5.4 Tear strength of ostrich crusts No differences were observed in tear strength between the groups in either perpendicular (p=0.08) or parallel (p=0.6) orientations (Table 2.4). It is notable however that tear strength was significantly higher parallel compared to perpendicular to the backbone (p=0.02) when data for each of the four storage groups were pooled (Table 2.5). No differences in skin thickness were found among the four groups of the crusts (p=0.42, Table 2.4).

Table 2.4 Tear strength of ostrich crust Thickness (mm) Maximum force reached (N) Tear strength (N/mm) Group Parallel Perpendicular Parallel Perpendicular Parallel Perpendicular 1 1.7 ± 0.42 2.1 ± 0.41 79.8 ± 16.0 72.0 ± 11.92 47.0 ± 9.49 34.8 ± 5.22 2 1.8 ± 0.5 1.9 ± 0.19 71.3 ± 13.84 66.3 ± 14.35 41.0 ± 10.8 34.9 ± 6.72 3 1.7 ± 0.68 1.7 ± 0.65 59.1 ± 15.08 70.5 ± 22.35 37.4 ± 7.89 42.9 ± 6.4 4 2.1 ± 0.55 2.1 ± 0.39 76.8 ± 18.82 69.4 ± 14.33 37.1 ± 5.22 33.0 ± 4.74

Values are means ± SD Table 2.5 Tear strength of ostrich crust: parallel & perpendicular to the spine Parallel (N/mm) Perpendicular (N/mm)

Mean 39.03 35.81 Standard deviation 7.89 6.88 Number of skin samples 32 32

2.6 Discussion 2.6.1 Storage treatment & tissue structure The general architecture of ostrich skin is consistent with reports of other workers. Ostrich skin consisted of an outer keratinised epidermis and an extensive connective tissue layer (dermis) comprised predominantly of collagen fibres. The dermis was comprised of two morphologically distinct layers, the grain and corium. Due to flaying of the skins the deepest layers, the stratum laxum and subcutis had been removed. These layers contain extensive adipose tissue comprised of fat cells (Lucas & Stettenheim 1972). Excess fat is removed as it provides a nutritional environment for microbes and treatment with hide salt and bactericides are ineffective (Cooper 2001). The surface of the skin is covered by a thin epidermis consisting of two to three layers of keratin. The epidermal cells produce the overlying keratin that provides protection from abrasion, control of water loss and also acts as physical barrier to invasion by bacteria and other microbial organisms. In addition, sebum produced by the epidermal cells nourishes the epidermis, restricts water loss and has antimicrobial properties (reviewed by Menon et al. 1996; Menon & Menon 2000; Stettenheim 2000). Sloughing of the keratin is likely due to abrasion rather than microbial breakdown as this effect was observed to a similar extent in all groups before and after storage. In addition, pitting of the keratin is likely to reflect the normal physiological degeneration of the keratin at the surface of the stratum corneum. The possibility however, cannot be excluded that regions of abrasion resulting in thinning of the stratum corneum may decrease its effectiveness as a microbial barrier thereby providing a potential route for microbial colonisation of the skins during storage.

18

The dermis consisted predominantly of collagen fibres interspersed with few elastic fibres. The collagen fibres in both the grain and corium were orientated in three directions although they predominantly ran parallel to the epidermal surface. A similar three-dimensional arrangement has been reported in the ostrich (reviewed by Bezuidenhout 1999; Lange 1929) and the emu (Frapple et al. 1997; Weir & Lunam 2004). The paucity of elastic fibres suggests that they minimally contribute to elasticity of the skin and that the strength and flexibility of the skin is provided by the cross-weave pattern of the collagen fibres The finding that the grain formed an extremely thin superficial layer of the dermis is consistent with the skin of the emu (Weir 2004; Weir & Lunam 2004) and feathered skin of chickens (Lucas & Stettenheim 1972). At the light microscope level, similar to emu skin (Weir & Lunam 2004), collagen fibres within the grain layer appeared to be more compact and thinner compared to those in the adjacent corium. These observations were confirmed by scanning electron microscopy. In addition, the grain and corium layers were separated by a narrow band of highly vascular loose connective tissue. This band of loose connective tissue is likely to provide little resistance to shearing forces during processing of the skins. We have reported similar differences in the morphology of the grain and corium in the emu and suggest that the lamination seen in tanned emu and ostrich skin (Cooper 2001) likely occurs at the junction between the corium and grain layers resulting in tearing of the grain from the corium. Scanning electron microscopy revealed an ordered array of compact fibrils within each of the collagen fibres before storage. After storage, some collagen fibres showed slight disruption in the alignment of their fibrils as well as increased separation of fibrils. Loss of the ordered array of collagen fibrils after storage suggests some loss of integrity of individual collagen fibres. However, disruption of the fibrils was found in all skins from each treatment. Quantitation of the diameter of the collagen fibres is necessary to determine whether the increase in skin thickness observed after storage at room temperature without bactericide (discussed below) resulted from a greater loss in the integrity of the collagen fibrils compared to the other treatments. 2.6.2 Effects of bactericide pre-treatment & storage temperature on skin thickness Skin thickness is one of a list of positive criteria for grading skins after tanning. Assuming a positive relationship between the thickness of the raw, salted skins and thickness after tanning, we expected that skins stored at room temperature without bactericide to undergo some degradation and become thinner compared to skins that were stored at 4-6ºC. To reduce bacterial damage during storage it is recommended that skins be pre-treated with bactericide and cold stored (Cooper 2001). Our findings support this recommendation. Pre-treatment with the bactericide Busan 85® prevented a change in thickness of the raw ostrich skins stored for four weeks. This effect was independent of storage temperature. In the current study for non-bactericide treated skins, storage at 4-6ºC resulted in thinning of skins whereas those stored at 22ºC became significantly thicker. This result was unexpected. This differential response in skin thickness with storage temperature may reflect a time course of degradation, with the rate of degradation inhibited at colder temperatures. We suggest that the thinning is likely due to dehydration of the skins as a result of salting, and that this would be followed by an initial period of degradation resulting in thickening of the skin. With continued storage we predict from work on emu skins with similar storage conditions (Weir et al. 2003) that the ostrich skins will become thinner, due to continued dehydration and possible loss of integrity of the collagen fibres. Denaturation of the collagen fibres would likely result in a thicker connective tissue layer comprised of a decreased density of the collagen fibres as well as an increased diameter of individual fibres. As this change is prevented with bactericide treatment it is suggested that any loss in the integrity of the

19

collagen fibres is likely to be caused by bacterial action rather than lysosomal autolysis. Loss of the ordered array of collagen fibrils after storage suggests some loss of integrity of individual collagen fibres. In a similar study in our laboratory we examined the effects of bactericide pre-treatment and temperature on the thickness of raw salted emu skins stored for four weeks (Weir et al. 2003). Similar to our finding in the ostrich, bactericide pre-treatment prevented the change in thickness of emu skins stored at room temperature but had no effect on skins stored at 4-6ºC. However, in contrast to our study on ostrich skin, the emu skins become thinner when stored at room temperature. We suggest that this difference in change of skin thickness after four weeks storage between ostrich and emu skin reflects a differential time course of damage between the two species. Emu skin is notoriously fragile and is highly susceptible to bruising and lamination during processing. These attributes are arguably the result of very thin grain and in particular, corium layers. Although emu skin has a similar histology to the ostrich, the dermis (grain and corium layers) is considerably thinner (approximately 500µm, Weir & Lunam 2004) compared the ostrich (approximately 1800 µm). It is likely therefore, to undergo degenerative changes, resulting in thinning of the skins, at a greater rate than ostrich skin. If this is the case, then the initial thickening of the ostrich skins observed at four weeks would be followed by thinning with longer-term storage. The effects of long-term storage on skin thickness are described in Chapter Three. 2.6.3 Tear strength of ostrich crusts – relationship to thickness of raw skins In the current study, both skin thickness and tear strength after tanning were higher than reported in chrome-tanned crusts from ostriches at fourteen months of age (Cloete et al. 2004). As the ostriches were of a similar age in both studies, these differences in the data may reflect variation of physical properties in different areas of the skin. Cloete and co-workers (2004) sampled from the butt region of the crown whereas we sampled from the upper belly. Alternatively, the differences in the physical characteristics may reflect genetic variation between the flocks. In the present study, no relationship was observed with either the tear strength or thickness of the ostrich crusts compared to the thickness of the raw, salted skins after storage. A similar finding was reported in the emu (Weir 2004). This suggests that any deterioration of the skin was insufficient to affect tear strength after tanning. In addition, no differences were found in the amount of plumping, size or shape of the feather follicles of the crusts from the different treatments. In sum, our data suggest that although some minor deterioration of the salted skins had occurred with storage at room temperature without bactericide pre-treatment, this had no observable affect on skin quality after tanning. 2.6.4 Filoplumes & bristle hairs The macroscopic organisation of filoplumes in the ostrich were similar to late immature filoplumes of the domestic chicken (Lucas & Stettenheim 1972). Four to six filoplumes formed a semi-circular pattern at the base of each feather follicle in all the skins. One or two bristle hairs were interspersed between the filoplumes. A similar distribution has been reported in the ostrich (Phil Glatz, personal communication 2005) and in other species of Aves (Lucas & Stettenheim 1972). It is of interest that the filoplumes and bristle hairs at the base of the contour feathers were restricted to the caudal side of each follicle. In addition, filoplumes and bristle hairs were found scattered in the interfollicular space to varying extents among the skins. The individual variation in the number and pattern of filoplumes and bristle hairs in the skins supports suggestions within the ostrich industry of a genetic heritability of these traits. Further, as filoplumes and bristle hairs had identical distributions for a given ostrich skin, we suggest these traits may be genetically linked. As such, filoplumes and bristle hairs could either be

20

expressed from two separate gene sets or be different phenotypic expressions of the same pleiotropic gene set. After tanning, the pattern of pinholes paralleled the distribution of filoplumes and bristle hairs (Figure 3.5). This indicates that these are “true” pinholes resulting from removal of the filoplumes and bristle hairs and are not defects in the leather caused by bacterial contamination (Cooper 2001).

21

3. Long-term storage: effects of temperature, bactericide & salt 3.1 Summary The effects of temperature, bactericide pre-treatment and salting on the preservation of raw ostrich skins during storage for nineteen weeks were determined. Thickness measurements were made on histological sections and confined to the grain and corium layers. The integrity of the tissue structure was assessed by light and scanning electron microscopy. Changes in skin thickness and/or deterioration of the general tissue structure after storage were compared to both the tear strength and thickness of the ostrich crusts. Skin quality was determined by professional grading after chrome-tanning and compared to the effect of the different storage treatments on its structural and physical attributes. Bactericide pre-treatment with either Diamoll® C or sodium hypochlorite had no effect on the thickness of salted skins after two weeks of storage at 4-6ºC (p=0.26). In contrast, unsalted skins pre-treated with Diamoll® C and similarly stored for two weeks, were significantly thicker compared to skins pre-treated with sodium hypochlorite (p=0.047). When data from each of the pre-treatment groups with either Diamoll® C, sodium hypochlorite or chilled water were pooled, initial storage conditions (salted versus unsalted) did not significantly affect thickness measurements (p=0.35). For skins stored for 14 weeks at 4-6°C, the effect of adding salt was to increase thickness by a mean of 739 µm (95% CI 321-1157, p=0.001). Salted skins stored at 4-6°C for 14 weeks became thicker by an average of 171 µm while salted skins stored at 22°C became thinner by an average of 934 µm. Consequently the difference in change of thickness was 1105 µm in favour of cold storage (95% CI 569-1641, p=0.000). No histological differences were observed between the groups after two weeks of storage. After 14 weeks, salted skins stored at 22ºC had greater spaces between the collagen fibres within the corium layer compared to skins stored salted at 4-6°C. In contrast, the corium of skins stored unsalted for 14 weeks appeared as an irregularly stained mass with the spaces between the individual fibres often obliterated. Scanning electron microscopy after 19 weeks of storage revealed pitting of the keratin in all skins. Skins treated with 1% sodium hypochlorite showed regions entirely devoid of the epidermis. Skins from all treatments showed splaying of collagen fibrils within the grain and corium layers. As the extent of splaying was highly variable within a given skin sample, it was not possible to relate storage treatment to the extent of loss of structural integrity of the collagen fibres. No differences were observed in tear strength among the treatment groups in either perpendicular (p=0.12) or parallel (p=0.59) orientations. Therefore, salting had no effect on tear strength following cold storage for 19 weeks. Similarly storage temperature (4-6ºC or 22ºC) had no effect on tear strength following storage of salted skins for 19 weeks. No differences were found in thickness of the tanned skins among the groups when measurements of skin samples taken parallel and perpendicular to the backbone were analysed jointly (p=0.14). No significant association was found between tear strength and thickness of the raw skins after 19 weeks of storage (p=0.12). The grade of the tanned skins was not decreased by cold storage for two weeks without salt. In contrast, cold storage without salting for 19 weeks markedly decreased the grade of the skins, assessed by the presence of bacterial damage and flattening of the follicles. No difference in grading was observed between salted skins stored for 19 weeks at either 4-6ºC or 22ºC.

22

3.2 Background & aims Long-term storage of raw skins is often necessary within the Australian ostrich industry. As discussed in Chapter One, raw ostrich skins produced in Australia may be stored for periods of up to several months prior to tanning. At present approximately 20% of raw skins are tanned in Australia whilst the remainder of the skins are sent off-shore, these skins may be stored for periods of up to one year prior to tanning. These extended periods of storage necessitate the development of procedures that cause minimal perturbation of tissue structure prior to tanning. Salt is currently used as the main preservative for raw ostrich and emu skins. Salting without bactericide pre-treatment is not adequate to prevent a change in thickness of raw skins in the either ostrich (Lunam et al. 2003) or emu (Weir et al. 2003) following four weeks of storage. As bactericide pre-treatment prevented this change in thickness of the dermis, we suggested this change was due to bacterial action, resulting in disruption of the fine structure of the collagen fibres. In view of the pattern in the change in thickness of raw emu skin stored for four weeks under similar conditions as the ostrich (Weir et al. 2003), we predict that with continued storage degradation of ostrich skins would result in a significant decrease is thickness. Salt poses a threat to the environment (Marcar & Crawford 2003), and concern is growing at local and state government levels over the high effluent salt released from tanneries. Developing alternative storage procedures for either non-salted skins or skins stored in less salt than at present, may prove essential for the sustainability of local industries associated with production of ostrich and emu skins. An array of alternatives to salt preservation of cattle skins have been investigated (see reviews by Bailey 2003; Barrett 1986). The initial studies focused on the effects of freezing, bactericide or bacitracin treatment, acid preservation, and preservation with sodium sulphite. These were considered effective preservatives for short-term storage. More recently the use of potassium chloride (potash) and electron-beam irradiation has been explored (Bailey 1995, 2003; Bailey & Gosselin 1996; Bailey et al. 2001). Potassium chloride has similar physical and chemical properties to sodium chloride but without the detrimental environmental effects of sodium chloride. Potassium chloride, although it did not inhibit growth of Staphylococcus sp., prevented degradation of raw calf and pig skins for up to six months (Bailey 1995; Bailey & Gosselin 1996). At present however, replacement of sodium chloride with potassium chloride is not cost effective. Similarly, electron-beam irradiation, although it is shown to be an effective preservative of raw hides for three weeks (Bailey et al. 2001), is more time-consuming than salting and requires plastic sealing of individual skins prior to irradiation. The effectiveness of these preservation techniques on ratite skin has not been examined. At present the recommended preservation technique for short-term storage of ratite skins is salting combined with chilling. The effect of chilling without salt has not been examined in ratite skin and may present a viable alternative to salt preservation. Chilling is reported to be effective for short-term storage of cattle hides (reviewed by Barrett 1986). However, the efficacy of chilling as a method of long-term storage of ratite skin is yet to be determined. The current work aimed to improve the quality of raw ostrich skins for the local and export markets by investigating long-term storage strategies that are ecologically sustainable. We aimed to determine the effects of temperature, bactericide pre-treatment and salting on preservation of skins during storage for 19 weeks. Changes in raw skin thickness or deterioration of tissue structure after storage were compared to the thickness, tear strength, and grade of the tanned crusts. These results are reported in the current chapter. In Chapter Four we describe the time course of microbial flora isolated from these skins and their potential to cause skin degradation.

23

3.3 Materials & methods 3.3.1 Sourcing & treatment of skins Nineteen skins from adult ostriches at 12 months of age, were purchased after slaughtered at a commercial abattoir in Myrtleford, Victoria. Ostriches were electrically stunned and killed by bleeding from the carotid artery. Skins were dry-plucked before removal from the carcass. They were then immediately placed into iced water for ten to fifteen minutes. Following immersion in the iced water, all skins were fleshed and tattooed in the region of the right and left legs. The skins were then immersed for a further four to five hours in iced solutions containing either the bactericide/fungicide Diamoll® C or sodium hypochlorite, or were placed into clean iced water. Although it is less toxic than chloro- and nitrophenols, Diamoll® C is classified as a class 6.1 toxic substance, is readily absorbed through the skin and may cause allergic reactions (MSDS Clariant, Australia). To minimise any potential health risks by Diamoll® C we used the lowest recommended concentration (Clariant, Australia fact sheet). Ten skins were soaked in 0.1% Diamoll® C and five skins were soaked in 0.25% sodium hypochlorite. After soaking, five skins from each of the two treatments (sodium hypochlorite or Diamoll® C) were halved along the midline. One half of each skin was salted and the remaining half left unsalted. The remaining five whole skins that had been soaked in Diamoll® C were not salted. The four skins re-immersed in the chilled water were coated with hide salt. Skins from each of the six treatment groups were stored on separate pallets with the dermal side facing upwards. To prevent any cross contamination, each group of skins were packed into individual plastic bags and transported by cold storage to Flinders University. Upon arrival at Flinders University one week after slaughter, all salted whole and half-skins were resalted. Each group of skins was placed on a separate wooden pallet and stored at 4-6°C for two weeks. However as all unsalted skins had some degree of bacterial growth, assessed by odour and some spotty discolouration after two weeks, all unsalted skin halves and whole skins were washed in 1% sodium hypochlorite. Groups one to five were stored for a total period of nineteen weeks at 4-6°C whereas Group six was stored for the remaining 17 weeks at 22°C. With the exception of Group four, all unsalted skins were salted during the third week of storage. In an attempt to reduce bacterial growth without salting, skins in Group four, after two weeks of storage, were irradiated with ultraviolet light for two hours on the flesh (dermal) side followed by one hour on the epidermal side. All salted skins were periodically resalted during storage. Data loggers (T-TEC®, Adelaide, SA) placed in each of the storage rooms recorded the temperature for the 19 weeks duration. Treatments of the six groups are summarised in Table 3.1. Due to the variety of the experiments and time consuming nature of preparation of the histology, samples for light and electron microscopy were taken at 14 and 19 weeks storage respectively. Microbial swabs were taken throughout the storage period. Samples were taken for light microscopy from the upper belly of each whole and half-skin at two and again at fourteen weeks of storage. Tissue segments were excised from each whole and half-skin for electron microscopy after nineteen weeks from the upper belly region. Processing of the tissue for either light or electron microscopy was as described previously (see Chapter Two). Swabs were taken for the assessment of microbial flora on the skins at periodic intervals during the storage period. Details of this component of the work are given in Chapter Four. At the end of the storage period all the whole and half-skins with the exception of two whole skins were transported by cold storage to a commercial tannery at Wangaratta, Victoria (Prestige Leather Australia Pty Ltd) for chrome tanning. One of the skins retained at Flinders University was used for further microbiology studies (described in Chapter Four). The remaining skin had extensive filoplumes and was used for further examination of filoplume structure. To minimise cross contamination between the groups during transport, all skins were liberally resalted and each of the six

24

groups of skins was placed into separate plastic bags prior to transport. In addition, skins from Group four (stored unsalted for 19 weeks) were washed in 1% sodium hypochlorite and salted before transport. After chrome tanning and finishing, the skins were graded by a professional tanner, Mr Jean Marc Bonsujet, at Wangaratta and returned to Flinders University. Samples of the skins were then taken from each of the whole and half-skins and sent to PCA Hodgson Chemicals Pty Ltd, Melbourne for measurement of tear strength. Table 3.1 Treatment groups Group 1 Salted and stored at 4-6°C for 19 weeks - half skins C0½R, C1½L, C2½L, C3½L, C4½L

Group 2 Stored at 4-6°C for 2 weeks unsalted, treated with 1% sodium hypochlorite then salted

for 17 weeks - half skins C2½R, C3½R, A0½R, A1½R, A4½R

Group 3 As Group 1 - half skins A0½L, A1½L, A2½L, A3½L, A4½L

Group 4 Stored unsalted at 4-6°C for 2 weeks, treated with 1% sodium hypochlorite, irradiated with ultraviolet light and stored at 4-6°C unsalted for a further 17 weeks – half skins C0½L, C1½R, C4½R, A2½R, A3½R

Group 5 As Group 1 - whole skins D0, D1, D2, D3

Group 6 Stored at 4-6°C for 2 weeks unsalted, treated with 1% sodium hypochlorite, then salted for 17 weeks at 22 °C - whole skins B0, B1, B2, B3, B4

½L & ½R indicate left and right half skins respectively Skin codes A-D are treatments prior to storage: A and B were immersed in Diamoll® C solution; C were immersed in sodium hypochlorite solution and D were immersed in iced water. 3.3.2 Assessment of skin quality Skin quality was assessed by the general appearance at both the gross and microscopic levels and thickness of the raw skins. As tanned ostrich skin is comprised exclusively of the grain and corium layers, measurements were confined to these layers. The effects of pre-treatment with either bactericide or sodium hypochlorite with and without salting on skin quality were assessed by skin thickness measurements after two weeks of storage. To assess the effects of temperature and salting during long term storage of the ostrich skins, thickness measurements were made on histological sections of the skins at two and fourteen weeks after storage. Skin quality was assessed by comparing the general tissue architecture and skin thickness between the different groups at two weeks and after 14 weeks of storage. To assess the effect of long-term storage with different treatments on skin architecture, the ultrastructural appearance (using scanning electron microscopy) of the skins was compared between the groups after nineteen weeks of storage. Six samples from each skin or half-skin were examined at instrument magnifications ranging between x50 and x5000. The six samples were placed on the stubs in different orientations so that all surfaces of the skin could be viewed. Assessment included the integrity of the keratin and presence or absence of the cellular epidermis, the organisation of the collagen fibres within the grain and corium layers as well as the alignment of fibrils within the fibres. In addition the tissue was examined for the presence or absence of fungi and bacteria. For each whole or half-skin, any change in thickness or differences in the general tissue structure after storage observed by microscopy was compared to the tear strength as well the grade of the skins after tanning.

25

3.3.2.1 Quantitation of skin thickness Skin thickness measurements were made on the haematoxylin and eosin stained sections only. Thickness of the grain and corium layers was measured with a x4 objective lens using an Olympus EHT light microscope with a calibrated graticule in the eyepiece. To avoid bias in selection of the region of skin to be measured, the first five sections from each of two tissue segments were measured at two and four millimetres from the left-hand edge of the section. This gave a total number of ten measurements for each skin or skin-half. Where the pre-designated region was folded or the tissue damaged during sectioning, the measurements were taken immediately adjacent to the damaged region. The thickness of each tissue sample per skin was estimated by taking the mean of ten measurements. Skin thickness data were analysed by linear mixed models. Bird (skins) was included in the model as a random factor with pre-treatments with either sodium hypochlorite, Diamoll® C, chilled water and the presence or absence of salt as fixed factors. 3.3.2.2 Tear strength Tear strength measurements were made using an Instron 4302 (Instron Corporation, Canton, USA). A total of six skin samples (three orientated parallel and three orientated perpendicular to the backbone) were excised from the upper belly flap of each of the 24 half-skins. Six skin samples were similarly excised from both the left and right upper belly flaps of each of the seven whole skins. Three repeat measurements of tear strength were then recorded both perpendicular and parallel to the backbone and values averaged for each direction for each half-skin. For each whole skin, the tear strength data for each of the right and left upper belly flaps were treated as separate measurements. Differences in tear strength between the storage groups were analysed by random intercept regression taking into account the paired half-skins. To examine the relationship between the thickness of the raw skins after 14 weeks of storage and tear strength after tanning and finishing, a linear mixed model was fitted including storage group, thickness at 14 weeks and direction of tear strength as fixed factors, and bird (individual whole and half-skins) as a random factor. 3.3.2.3 Grading of tanned skins Skins were graded according to the procedure recommended by Cooper (2001). For whole skins the crown area was divided into quarters to form a diamond-shaped region. The base of the neck, base of the tail region, and laterally the outermost follicles within the crown area formed the extremities of the diamond. Skins were graded according to the number of quadrants with defects, as well as the number and type of defects within each quadrant. Half-skins were graded according the number of defects within each of the two quadrants. Criteria for grading included signs of bacterial damage such as dark patches, plumping and roundness of the follicles, general skin texture and skin thickness, as well as criteria unrelated to the storage conditions such as the presence of filoplumes, bristle hairs and scars.

26

3.4 Results The temperature of each of the storage rooms remained relatively constant throughout the storage period (see Chapter Two, Figure 2.1). 3.4.1 Light microscopy – tissue architecture 3.4.1.1 Effects of anti-microbial pre-treatment after two weeks storage at 4-6°C Skins salted at time of slaughter (Groups one, three and five) Skins pre-treated with Diamoll® C, 0.25 % sodium hypochlorite or iced water showed a similar histology. No differences in the morphology of the epidermis, grain or corium were found between the skins. A thick layer of keratin was found overlying the cellular epidermis. Epidermal nuclei often appeared darkly stained and irregular in shape indicative of pyknosis. Sebokeratinocytes were not distinguishable within the epidermis. The grain and corium layers within the dermis were readily distinguishable from one-another based on the small size and compactness of the collagen fibres within the grain layer compared to the corium. Fibroblast nuclei were observed in both the grain and corium layers. Individual collagen fibres were clearly visible within the dermis with minimal shrinkage spaces between them. Skins unsalted from time of slaughter (Groups two, four and six) The additional treatment with 1% sodium hypochlorite, to reduce the bacterial growth prior to sampling appeared to have a deleterious affect on the epidermis but had no apparent affect on the histology of the dermis. In some skins epidermal nuclei were absent whilst in other skins the epidermis had been completely removed. A few skins contained localised patches of nucleated epidermal cells. Fibroblast nuclei were absent within the dermis of some skins. The organisation of the collagen fibres within the grain and corium layers had a similar appearance among the groups and was indistinguishable from skins that were salted at slaughter (Groups one, three and five). 3.4.1.2 Effects of long-term storage with hide salt at 22°C (Group six) versus 4-6°C (Groups one, two, three & five) After 14 weeks of storage at 4-6°C, the grain and corium layers of skins salted from slaughter (Groups one, three and five) showed a similar histology to those in Group two that were stored unsalted only for the initial two weeks. After 14 weeks, most of the nuclei within the epidermis were pyknotic. The morphology of both the grain and corium layers were similar at 14 weeks compared to two weeks storage. The only observable difference between these groups and Group six stored at 22ºC after 14 weeks storage, was the increased spaces between collagen fibres within the corium of Group six skins compared to skins stored at 4-6ºC. 3.4.1.3 Effects of salting versus non-salting with long-term storage at 4-6°C Group four skins that were stored for 14 weeks unsalted showed differences in their morphology compared to salted skins also stored at 4-6°C. In contrast to the other groups, the corium of skins from Group four appeared as an irregularly stained mass making it difficult to identify individual collagen fibres. Spaces between the individual fibres were often obliterated and the fibres appeared to abut against one-another. The grain however, was readily distinguishable from the underlying corium and had a similar morphology to all other storage groups. A bacterial infiltrate was present in the deep

27

region of the corium in a single skin in Group four. After 14 weeks, Group four skins exhibited numerous localised areas of epidermal damage and an absence of fibroblast nuclei within the dermis. 3.4.1.4 Effects of storage treatment on thickness of the raw skins Skin thickness measurements are given for each of the six treatment groups at two and 14 weeks of storage in Table 3.2. Treatment with either Diamoll® C or sodium hypochlorite alone was not sufficient to prevent obvious bacterial growth on the skins even though they were stored at 4-6°C. Bacterial growth appeared within two weeks from slaughter as evidenced from their odour and the patchy discolouration on the skins. In contrast, all skins that had been treated with either Diamoll® C or sodium hypochlorite and then immediately salted prior to transport to Flinders University had neither bacterial odour nor discoloration. Similarly skins soaked in chilled iced water and then immediately salted showed no signs of bacterial damage. For the salted skins, immersion in iced water, Diamoll® C or sodium hypochlorite had no effect on skin thickness after two weeks of storage as no differences in skin thickness were found among the three treatment groups (p=0.26). In contrast, unsalted skins treated with Diamoll® C were significantly thicker compared to the unsalted skins pre-treated with sodium hypochlorite (p=0.047). The effects of bactericide, sodium hypochlorite and salting on skin thickness are given in Table 3.3. Table 3.2 Skin thickness measurements (µm) for each storage group Group number 2 weeks storage 14 weeks storage 14 - 2 weeks storage 1 1477 ± 378 (169) 1661 ± 282 (126) 183 ± 475 (212) 2 1250 ± 285 (128) 1468 ± 178 (80) 218 ± 240 (107) 3 1262 ± 251 (112) 1373 ± 376 (168) 111 ± 503 (225) 4 1299 ± 322 (144) 732 ± 127 (57) -568 ± 412 (184) 5 1518 ± 196 (98) 1749 ± 506 (253) 231 ± 430 (215) 6 2186 ± 710 (318) 1253 ± 247 (111) -934 ± 826 (369) Means ± SD (SEM) Table 3.3 Effects of bactericide, sodium hypochlorite and salting on skin (grain+ corium) thickness (μm) after two weeks of storage at 4-6°C Treatment Salted Non-salted Diamoll® C 1262 ± 251 (112) 1775 ± 662 (209) NaOCl 1477 ± 378 (169) 1187 ± 331 (148) Iced water 1518 ± 196 (98) Means ± SD (SEM) When data from each of the pre-treatment groups with either Diamoll® C, sodium hypochlorite or chilled water were pooled, initial storage conditions (salted versus unsalted) did not significantly affect thickness measurements (p=0.35, Table 3.4). Unsalted skins stored at 4-6°C (Group four) showed a significant decrease in thickness from two to 14 weeks storage (p=0.037). In contrast, salted skins stored at 4-6°C (Groups one, three and five) showed no significant change in thickness from two to 14 weeks storage (p=0.17). Skins stored salted at room temperature (Group six) tended to become thinner after 14 weeks. This change however, was not statistically significant (p=0.06).

28

For skins stored for 14 weeks at 4-6°C, the effect of adding salt (Groups one, three and five) was to increase thickness by a mean of 739 µm (95% CI 321-1157, p=0.001) compared to Group four that was stored unsalted. Salted skins stored at 4-6°C for 14 weeks became thicker by an average of 171 µm while salted skins stored at 22°C (Group six) became thinner by an average of 934 µm. Consequently the difference in change of thickness is 1105 µm in favour of cold storage (95% CI 569-1641, p=0.000). The change in skin thickness during storage is summarised in Table 3.5. Table 3.4 Skin (grain + corium) thickness (μm) Treatment Salted Non-salted 2 weeks storage at 4-6°C 1412 ± 293 (78)1,3,5 1579 ± 629 (162)2,4,6 14 weeks storage 4-6°C 1553 ± 349 (80)1,3,5 732 ± 127 (57)4 22°C 1253 ± 247 (111)6 - Means ± SD (SEM) Storage groups are shown in superscript Table 3.5 Change in skin thickness (μm): 14 weeks – 2 weeks storage Temperature Salted Non-salted 4-6°C 171 ± 439 (117)1,3,5 -568 ± 412 (184)4 22°C -934 ± 826 (369)6 - Mean change ± SD change (SEM change) Storage groups are shown in superscript 3.4.2 Scanning electron microscopy Qualitative assessment of the general tissue structure revealed differences in the ultrastructure of the epidermis in skins from the different storage treatments. In contrast the dermal layers showed a similar morphology among the treatments. Skin halves that were salted and stored at 4-6°C for 19 weeks showed minimal evidence of deterioration. These skins were found to have numerous overlapping keratin flakes on the epidermal surface with some pitting of the surface (Figure 3.1A). The grain and corium were readily distinguishable by the differences in the density and cross-sectional diameter of the collagen fibres between the two layers. Shrinkage spaces and localised splaying of collagen fibres were evident within the grain and corium layers.

29

Figure 3.1 Scanning electron micrographs showing the epidermal surface of left (A) and right (B) halves from the same skin after 19 weeks of storage at 4-6°C. (A) Group one (salted for entire 19 weeks) shows numerous flakes of keratin on the surface. (B) Group four (treated with 1% sodium hypochlorite and unsalted for 19 weeks) shows almost complete absence of keratin. Scale bars are 40 µm. In contrast to skins salted for the entire storage period (Groups one, three and five) all skins from Group two (unsalted at 4-6°C for the first two weeks of storage, treated with 1% sodium hypochlorite, then salted and kept at 4-6°C) and four (unsalted, 4-6°C) showed localised areas of epidermal deterioration including absence of keratin (Figures 3.1B and 3.2). In some samples the entire epidermis was removed and the fibrous structure of the underlying grain could be viewed from the epidermal surface. Skins from Group six (unsalted at 4-6°C for the first two weeks of storage, treated with 1% sodium hypochlorite, then salted and kept at room temperature) also showed some localised areas of epidermal damage. Bacteria and fungi (Figure 3.3) were occasionally found on the epidermal and dermal surfaces of the unsalted skins.

Figure 3.2 Scanning electron micrograph of the transverse surface of skin after 19 weeks storage at 4-6°C. This skin was stored unsalted for the first 2 weeks of storage, treated with 1% sodium hypochlorite and then salted for the remainder of the storage period. The epidermis (arrow) is absent (see Figure 2.10 for comparison). The grain (g) and corium (c) layers are clearly differentiated. Within the corium, individual collagen fibres can be seen aligned predominantly parallel to the surface. Scale bar is 50 µm.

30

Figure 3.3 Scanning electron micrograph showing the epidermal surface of a skin treated with 1% sodium hypochlorite and stored unsalted at 4-6°C for 19 weeks. Fungal hyphae with conidia (arrows) are shown overlying the keratin. Scale bar is 20 µm. In all skins, independent of storage treatment, the grain and corium layers were readily identified. Splaying of collagen fibrils (Figure 3.4) was found in skins from all storage groups. As the extent of splaying was highly variable within a given skin sample, it was not possible to relate storage treatment to the extent of loss of structural integrity of the collagen fibres. However, the frequency of splaying of the fibrils within any given sample was noticeably greater compared to the splaying observed in skins at the time of slaughter (see Chapter Two).

Figure 3.4 Scanning electron micrograph showing collagen fibres within the corium of a skin stored unsalted at 4-6°C for 19 weeks. Individual collagen fibrils at the edges of the fibres are seen splaying from the main fibre and entangling with those from adjacent fibres. Separation of some of the fibrils is visible within the fibre in the centre of the micrograph. This separation however, may be an artefact caused by damage to the cut edge of the tissue block during preparation of the specimen. Scale bar is 5 µm.

31

3.4.3 Tear strength of the tanned skins No differences were observed in tear strength among the groups in either perpendicular (p=0.12) or parallel (p=0.59) orientations (Table 3.6). Tear strength was not significantly different between parallel and perpendicular orientations (p=0.26) when data for each of the six storage groups were pooled (Table 3.7). No differences were found in thickness of the tanned skins among the groups when measurements of skin samples taken parallel and perpendicular to the backbone were analysed jointly (p=0.14). No significant association was found between tear strength and thickness of the raw skins after 14 weeks of storage (p=0.12). Table 3.6 Tear strength of the ostrich skins

Thickness (mm) Maximum force reached (N) Tear strength (N/mm) Group Parallel Perpendicular Parallel Perpendicular Parallel Perpendicular 1 1.26 ± 0.13 1.24 ± 0.13 78.73 ± 16.54 74.33 ± 18.07 62.37± 10.90 59.97 ± 12.75

2 1.30 ± 0.22 1.26 ± 0.11 77.86 ± 20.04 73.26 ± 17.99 60.40 ± 13.33 56.73 ± 10.07

3 1.16 ± 0.11 1.22 ± 0.19 64.46 ± 25.79 64.87 ± 11.89 54.95 ± 19.74 53.39 ± 7.41

4 1.12 ± 0.05 1.16 ± 0.18 76.93 ± 22.43 86.47 ± 28.21 68.78 ± 20.59 74.11 ± 20.63

5 1.05 ± 0.15 1.17 ± 0.24 67.78 ± 12.48 62.17 ± 17.22 64.50 ± 7.54 53.06 ± 9.41

6 1.34 ± 0.23 1.23 ± 0.18 80.75 ± 12.12 72.75 ± 10.49 60.87 ± 6.31 58.31 ± 14.15

Values are means ± SD. Total number of samples is 34 in each direction. Each sample is an average of three repeat measurements. Table 3.7 Tear strength of ostrich skins: parallel & perpendicular to the spine (N/mm) Parallel Perpendicular Mean 61.95551 58.99725 Standard deviation 12.96325 13.89207 Number of samples 34 34

3.4.4 Skin grades The grading of the raw skins at purchase compared to the skin grade after tanning is given in Table 3.8. In general, the grades of many of the tanned skins were decreased due to the presence of filoplumes, bristle hairs and scarring. Four of the five skins stored for the entire 19 weeks without salt (Group four) were graded as rejects due to bacterial damage after tanning. In contrast to the skins from the other treatment groups, these skins proved difficult to dye during the finishing process. Bacterial damage was not found in the crusts that had been stored salted at 4-6°C for entire 19 weeks (Groups one, three and five). Similarly skins stored unsalted at 4-6°C for the initial two weeks prior to salting (Group two) had no evidence of bacterial damage. Salting of the raw skins stored at 4-6°C for the initial two weeks of storage had no effect on the final grade of the tanned skins. It is notable that some of the skins stored initially without salt (Group two) were classified as first grade after tanning (Table 3.8).

32

One skin in Group six (stored salted at room temperature) was rejected due to bacterial damage. Although the remaining skins in Group six had low grades after tanning, this was due to the presence of pinholes, scars or filoplumes. Follicles in the skins tended to be raised and rounded. Follicles in skins from Groups one, two, three, five and six, were either well raised or mid-plumped. Storage of salted skins at room temperature, or cold storage without salt for the initial two weeks appeared to have no detrimental effect on the shape or degree of plumping of the follicles in the tanned skins. No differences were observed in either the shape or plumping of the follicles between the corresponding left and right skin halves in Groups one, two and three. In contrast to the other groups, follicles in half-skins stored unsalted at 4-6°C for 19 weeks (Group four) were only slightly raised and often elongated.

33

Table 3.8 Grading of the ostrich skins Raw skins Tanned skins

†Group §Skin number

Grade Comment Grade Comment

C0½R 1 1 C1½L 1 1 C2½L 2 Bristle hairs 2 Knife cut C3½L 1 Filoplumes 1

1

Salted/4-6°C

C4½L 1 1 C2½R 2 Bristle hairs 1 C3½R 1 Filoplumes 1 A0½R 1 5 A1½R 2 reject Scars

2

2 weeks unsalted @ 4-6°C /salted @ 4-6°C

A4½R 2 reject Scars A0½L 1 1 A1½L 2 1 A2½L 2 Bristle hairs 1 A3½L 2 1

3

Salted/4-6°C

A4½L 2 2 Scars C0½L 1 4 Scars C1½R 1 reject Bacteria C4½R 1 reject Bacteria A2½R 2 Bristle hairs reject Bacteria

4

Unsalted/4-6°C

A3½R 2 reject Bacteria D0 2 Filoplumes - Not sent to

tannery D1 3 1 D2 1 Filoplumes 5 Filoplumes

5

Salted/4-6°C

D3 1 Bristle hairs 3 Filoplumes B0 4 4 Scars B1 2 reject Bacteria B2 3 3 Pinholes B3 2 2 Filoplumes

6

2 weeks unsalted @ 4-6°C /salted @ 22°C for 17 wks

B4 4 Filoplumes - Not sent to tannery

§½R & ½L indicates right and left half-skins respectively †Details of treatments are given in Table 3.1 3.4.5 Distribution of filoplumes, bristle hairs & pinholes Filoplumes & bristle hairs Filoplumes and bristle hairs were present at the base of most feather follicles in all raw skins examined. Their density and distribution in the interfollicular spaces varied between individual skins. Pinholes Tiny discrete rounded holes formed a semi-circular pattern at the caudal aspect of the base of the feather follicles in all the tanned skins (Figure 3.5). A noticeable feature was the plumping of the skin in the region of the pinholes. Plumping of the skin was associated with pinholes at the base of the

34

follicles as well as most of the pinholes lying between the follicles. The density of pinholes in the interfollicular spaces appeared to directly relate to the density of the filoplumes and bristle hairs in the raw skins. Skins with very few filoplumes and bristle hairs between the follicles had few interfollicular pinholes. In contrast, skins with numerous interfollicular filoplumes had pinholes scattered across their entire crown region. In these skins, the interfollicular pinholes were more numerous in the cranial quadrants compared to the caudal quadrants of the crown area.

Figure 3.5 Crown area of chrome-tanned ostrich skin. The holes (arrow) from the shafts of the contour feathers are located in the caudal region of the raised follicles. The follicles are orientated to accommodate the contour feathers angled in a cranial to caudal direction. Immediately caudal to the base of each follicle lie several discrete pinholes (arrowheads). The skin is noticeably more plumped in the region of the pinholes at the caudal side of each follicle. An interfollicular pinhole is visible (arrowhead with asterisk). Scale bar is 1cm. 3.5 Discussion 3.5.1 Storage treatment, tissue structure & skin thickness Treatment of the raw skins at slaughter with 0.1% Diamoll® C or 0.25% sodium hypochlorite had no effect on either the morphology or thickness of salted skins after two weeks of storage. This finding is in contrast to our previous study (Chapter Two) where we observed that pre-treatment with bactericide (Busan 85®) prevented a change in thickness of salted skins after storage for four weeks (Lunam et al. 2003). We suggested that the increase in skin thickness at room temperature was caused by bacterial damage to the collagen fibres. We further suggested that the time course of damage to the skins involved an initial thickening followed by thinning of the dermis. A possible explanation is that the differences in data between the two studies reflect the time course of damage. In the current study, the initial sample was taken at two weeks, compared to sampling at the time of slaughter in the previous study. Therefore under conditions of salting and cold storage the data suggests the greatest rate of change occurs within the first two weeks of slaughter. In support of this argument, salted skins stored at 4-6°C showed no significant change in thickness between two and 14 weeks storage.

35

Skins stored salted at room temperature tended to become thinner after 14 weeks. Long-term storage at room temperature would be expected to promote the rate of bacterial growth with a resultant increased potential to degrade the collagen fibres compared to cold storage (reviewed by Bailey 2003). A higher storage temperature may also provide a more suitable environment for the action of collagenolytic enzymes (Lee et al. 2003; Nagano & To 1999). In addition, it is likely that the rate of dehydration of the skins was greater at room temperature compared to storage at 4-6ºC resulting in shrinkage of the skins. The finding of wider shrinkage spaces between the collagen fibres in the dermis of skins stored at room temperature compared to cold stored skins supports this argument. Unsalted skins pre-treated with Diamoll® C were thicker after two weeks storage compared to unsalted skins pre-treated with sodium hypochlorite. Based on our predicted time course of damage to the skins, this increase in skin thickness at two weeks storage suggests Diamoll® C was less effective than sodium hypochlorite in preventing bacterial damage to the connective tissue layers at the concentrations used. However, pre-treatment with either Diamoll® C or sodium hypochlorite was ineffective in preventing microbial growth on unsalted skins stored at 4-6°C for two weeks. In contrast to the salted skins, unsalted skins stored at 4-6ºC had become significantly thinner after 14 weeks. In addition, the skins were rejected after tanning due to the present of bacterial damage. This supports our initial premise that the time course of damage follows an initial thickening of the skins followed by thinning due to bacterial damage. In addition it demonstrates that cold storage for 14 weeks in the absence of salt is insufficient in preventing significant skin damage. It was of interest that the skins that were stored unsalted for two weeks then treated with 1% sodium hypochlorite and salted for the remainder of the storage period showed no signs of bacterial damage after 19 weeks. Furthermore, many of these skins were assessed as first or second grade after chrome tanning. This suggests that short-term storage without salt for two weeks has no visible detrimental effect on the quality of the tanned skins. In addition, our data suggests two weeks was inadequate for penetration of microbial flora beyond either the epidermis or the exposed surface of the corium following its removal from the adjacent stratum laxum during fleshing of the skins. In contrast to our work, Hopkins et al. (1973) reported grain damage in limed cattle skins that had been stored unsalted for 72 hours. Although Hopkins et al. (1973) did not tan the skins it is likely that damage to the grain in the raw skins would be visible after tanning. After cold storage for 14 weeks, the unsalted skins had become significantly thinner than the salted skins. Viewed by light microscopy the corium of unsalted skins revealed an irregularly stained mass of tissue and in contrast to all salted skins, the spaces between the collagen fibres had become obliterated so that individual fibres were not clearly delineated. This change in the morphology of the dermis is consistent with a loss of integrity of the fine structure of the individual collagen fibres. In support of this finding, scanning electron microscopy of these skins revealed fibres scattered within the corium and grain layers in which their fibrils had lost their parallel array. However, similar variations in ultrastructural appearance of the collagen fibres were found in skins from each of the six storage groups. Thus it was not possible to relate any loss of the structural integrity of the collagen fibres at the electron microscope level to any particular storage treatment. To identify any differences in the integrity of the collagen fibres between treatments, quantitation at the electron microscope level is required; this was beyond the scope of this study. We have suggested that the change in skin thickness relates to the loss of integrity of the collagen fibres (Lunam et al. 2003). This loss of integrity is likely due to the denaturation of collagen, which at the molecular level ultimately results in disruption of the cross-banding pattern of individual fibrils within a single collagen fibre. In a healthy fibril collagen molecules are aligned in a staggered fashion to form a distinctive banding pattern viewed by transmission electron microscopy (see reviews by Covington 1997; Montes 1996; Reich 1999). Kobayashi et al. (1977) using transmission electron microscopy, observed a reorganisation in the pattern of the fibrils in skin following application of a variety of collagen-degrading agents. In a similar study, Ding (2004) demonstrated a disruption of the

36

banding pattern of the collagen fibrils in goat skin. Furthermore, this loss of the banding pattern was directly related to the concentration of the bating enzyme. The cross-banding pattern is visualised in a transmission electron microscope by the differential transmission of electrons passing through ultrathin sections. As the scanning electron microscope reveals the surface topography by deflection of electrons from the surface of specimens it does not reveal the cross-banding arrangement of the fibrils. Consequently in the current study using scanning electron microscopy, we were not able to visualise the cross-banding pattern of the individual fibrils. To determine the relationship between skin damage and the integrity of the collagen fibres further work using transmission electron microscopy is required to examine the banding pattern of the collagen fibrils. Scanning electron microscopy revealed pitting of the keratin after 19 weeks of storage. The pitting occurred in skins from all treatments. It was of interest that the pitting appeared to be no more extensive after 19 weeks storage in the current study compared to the amount of pitting after four weeks storage (Chapter Two). This demonstrates that the keratin is highly resilient to degeneration. However, the unsalted skins, which received two treatments with 1% sodium hypochlorite during storage, had focal areas where the keratin and underlying epidermis had been completely removed. Removal of the keratin or cellular epidermis either as a result of treatment or abrasion during handling would provide direct access to the underlying collagen by microbial flora. One of the criteria for determining skin grade is thickness after tanning. Thickness of the tanned ostrich skin is an important consideration for determining its suitability for different end products (Cooper 2001). In contrast to skin thickness at the end of storage there were no significant differences in the thickness of tanned skins between the six storage groups. No relationship was found between the thickness of the raw skins and skin thickness after tanning. Thickness may have been influenced by different amounts of buffing of the individual tanned skins. As there was considerable variation in skin thickness within each storage group (both the raw and tanned skins) a larger number of skins may be required to identify any relationship between their thickness before and after tanning. Raw skins in the current study appeared to be thinner than the skins examined in the short-term storage study (Chapter Two). This difference in thickness may be due to genetic variation between the different flocks, or it may reflect age differences between the birds. In the current study, the ostriches were 12 months of age compared to 14 months of age in the short-term storage study. In support of this, Angel et al. (1997) reported the thickness of tanned ostrich skins increases linearly from 7.3 to 16 months of age. In a similar study Cloete et al. (2004) reported ostrich leather increased in thickness from birds aged five to 14 months. Other than rejection of the unsalted skins due to bacterial damage all other treatment groups had some skins that were graded as first or second grade. With the exception of one skin stored at room temperature, no skins from any other group showed bacteria damage. In these skins downgrading of skin quality resulted from criteria unrelated to storage such as scarring and the presence of filoplumes. This suggests that salted skins may be stored for up to 19 weeks either at room temperature or at 4-6°C without decreasing skin quality. This is supported by a long-term study of cattle skins which demonstrated skins may be cold stored for up to five years in salt without significant detrimental effects on the yield, grade, physical properties or chemical composition of the leather (Frey & Stuart 1941). Furthermore, previous studies have indicated that cattle hides may be stored salted at ambient temperatures for periods of up to eight months without detrimental affects on hide quality (Kritzinger 1950). 3.5.2 Filoplumes, bristle hairs & pinholes Consistent with previous findings in skins from the short-term storage study (Chapter Two), filoplumes and bristle hairs invariability followed the same pattern of distribution within any given

37

skin. This supports our previous suggestion that their phenotypic expression is genetically linked. In addition, the distribution of these feather-like structures paralleled the presence of pinholes in the ostrich crusts. This supports the notion of that these tiny holes are true pinholes formed from removal of the filoplumes and bristle hairs. Although the tanned skins showed individual variation in the number of pinholes within the interfollicular space they were always found at the base of the feather follicles suggesting that all birds in this flock have both filoplumes and bristle hairs. The feathers lie at an angle in a cranial to caudal direction. As the filoplumes and bristle hairs tend to be confined to the caudal region of the follicle base, in a live bird they would be obscured by the feathers and difficult to identify. In the raw skins, both filoplumes and bristle hairs were present. As we could not identify any morphological differences between the discrete tiny pinholes in the crusts it is likely that the pinholes observed in the skins are formed from the sheaths of both filoplumes and bristle hairs. A noticeable feature was the slight discrete plumping of the skin encircling many of the individual pinholes in the interfollicular space. Histology revealed the sheath of the filoplumes is similar to that of the feather follicle, both being lined by stratified epithelium that is in-turn surrounded by a compact layer of collagen. A similar histology of the filoplume sheath has been described in other avian species (reviewed by Lucas & Stettenheim, 1972). Lucas and Stettenheim suggest the histological difference between the sheaths of filoplumes and contour feathers is the absence of the Mm. pennarum attachment to the filoplume sheath. Thus during the tanning process, plumping of the tiny collagen sheaths of filoplumes, and possibly bristle hairs likely occurs, albeit on a miniature scale compared to the plumping of the follicle sheaths of the large contour feathers. 3.5.3 Feather follicles in tanned skin Long-term storage of salted skins had no effect on the roundness or degree of plumping of the follicles in chrome tanned skins. Follicles of all salted skins, stored either at 22ºC or 4-6ºC were highly plumped and rounded. This suggests that any degradation of the collagen fibres within the grain and corium layers was insufficient to affect quality of the tanned skins. Similarly cold storage without salt for the initial two weeks appeared to have no detrimental effect on the shape or degree of plumping of the follicles in the tanned skins. In contrast, skins stored at 4-6ºC without salting for the entire 19 weeks (Group four) had flatted follicles that were often elongate. As their corresponding skin halves that had been stored salted had rounded and plumped follicles, the lack of plumping of the follicles in Group four skins was not due to individual variation among the ostriches but rather resulted from degradation of the collagen fibres within the dermis. 3.5.4 Tear strength Although the morphology, thickness and grade of the skins were each affected by storage treatment, tear strength was similar among all treatment groups. Of particular interest, the unsalted skins with bacterial damage reached a maximum breaking force that was similar to the breaking force of all other storage groups. This suggests that the collagen fibres are highly resilient to the insults of bacterial damage. This does not discount the possibility that many of the fibres may have undergone some degree of denaturation during storage. Resistance to tearing would also be provided by the three-dimensional cross-weave pattern of the collagen fibres. This pattern was clearly apparent in all skins viewed by scanning electron microscopy after 19 weeks storage. As previously discussed (Chapter Two) the three dimensional arrangement of the collagen fibres provides flexibility and strength to the dermis.

38

Similar to our findings, Looney et al. (2002) showed different storage methods have marginal effects on the tear strength of kangaroo skins at both the wet-blue and crust stages. The relative strengths of kangaroo skins stored either chilled, salted or frozen appeared to be dependent on the time of the year. For skins harvested during the winter those that were stored frozen were weaker than skins that were chilled or salted (Looney et al. 2002). In contrast, for skins harvested during the summer those that were stored salted were weaker than skins that were frozen or chilled (Looney et al. 2002). It was not clear from these studies whether the skins were preserved for short- or long-term storage periods or whether the differences in tear strength were statistically analysed. The tear strength of the ostrich skins in the short and long-term studies was unlikely to be affected by seasonal influences as both batches of skins were obtained at similar times of the calendar year. Although the maximum force reached during physical testing of the ostrich skins was similar to that reported in our previous study (Chapter Two), the tear strength per millimetre was higher in the current study compared to skins in Chapter Two. This higher tear strength is a reflection of the thinner skins in the current study. In summary we could not identify any relationship between the maximum tearing force reached and skin thickness in adult ostriches. In summary salted skins may be stored for up to 19 weeks either at room temperature or at 4-6°C without compromising skin quality. In contrast, cold storage for 19 weeks in the absence of salt is insufficient in preventing significant skin damage. Therefore, for long term-storage of several months salting is essential to prevent downgrading of the skins by bacterial damage.

39

4. Pilot study of microbial flora on skins during storage 4.1 Summary Microbial flora were isolated from skins stored under different treatment regimes over a storage period of four months. Skins were swabbed for microbial flora at two, three, five and 16 weeks of storage. The effects of hide salt, storage temperature, antimicrobial pre-treatment (sodium hypochlorite or Diamoll® C), and ultraviolet irradiation on the type and persistence of microbial flora was determined. As an indication of their potential to degrade collagen, representative micro-organisms isolated from the stored skins were subsequently assessed for their ability to liquefy gelatin in vitro. Selected bacteria and yeast cultures showing varying degrees of gelatin liquefication were reapplied to skin swatches for 45 days to evaluate their ability to degrade collagen in situ. Skin deterioration was determined by qualitative assessment of general tissue structure and the organisation of collagen fibres using scanning electron microscopy. Cultures derived from swabs taken following two weeks storage revealed clear differences between the types of micro-organisms colonising the salted and non-salted skins. Non-salted skins showed heavy growth of mixed Gram-negative bacilli including Hafnia alvei, Enterobacter cloacae, Aeromonas species, Enterobacter amnigenus, Acinetobacter calcoaceticus-Acinetobacter baumannii complex, Acinetobacter lwoffii and Proteus species. Gram-positive bacteria also colonised the non-salted skins but in fewer numbers. In contrast, no Gram-negative organisms were isolated from the salted skins. Staphylococcus, Micrococcus and Bacillus species predominated in the salted skins. In addition to these Gram-positive bacteria, two fungal species, Zygomycete sp. and Alternaria sp. were isolated from the salted skins at two and five weeks storage. Gram-negative bacteria persisted in the non-salted skins throughout the entire storage period. Treatment with 1% sodium hypochlorite and ultraviolet irradiation did not affect the ability of the micro-organisms to persist on these skins. In contrast, Gram-negative bacteria were never isolated from skins that were salted for the four-month storage period. Furthermore, Gram-negative bacteria colonising the non-salted skins were absent within two weeks of salting. This effect of salting on Gram-negative micro-organisms was independent of the storage temperature. Assessment of the ability of selected isolates to liquefy 15% porcine gelatin in vitro revealed a varying degree of gelatinase activity. Seventy-two percent of the selected strains from non-salted skins and 33% of strains isolated from salted skins showed gelatinase activity. In addition, four out of five of the selected Gram-negative isolates exhibited gelatinase activity. This supports the notion that Gram-negative bacteria are particularly effective at liquefying gelatin and that the microbial flora of non-salted skins are likely to possess greater proteolytic activity than that of salted skins. No correlation was found between the ability of a particular isolate to liquefy gelatin in vitro and their ability to degrade collagen in situ due to the similar extent of deterioration of all skin swatches. During long-term storage skins were colonised by several species of Gram-positive and Gram-negative bacteria. Gram-positive bacteria colonised the skins in the presence of salt stored at either 22ºC or 4-6ºC. In contrast, salting totally inhibited the growth of Gram-negative bacteria. This effect was independent of the storage temperature. Although isolates of both Gram-positive and Gram-negative bacteria were able to liquefy gelatin in vitro, only skins that were stored unsalted and therefore colonised by Gram-negative bacteria prior to tanning were downgraded as a result of bacterial damage. These findings suggest that bacteria responsible for collagen degradation in ostrich skin are Gram-negative and their growth is inhibited by salting.

40

4.2 Background & aims A number of factors may cause downgrading of ostrich skins. In Australia, downgrading often results from skin damage that occurs prior to or during slaughter of the bird (MacNamara et al. 2003). Improved procedures for the handling, transport and slaughter of ostriches are required in order to reduce bruising and scarring of skins and damage to the feather follicles. Another well-known contributing factor that results in downgrading of the skins is damage caused by the action of microbial flora during skin storage. Microbial action on raw skins is known to result in putrefactive odour, discolouration such as ‘red heat’ and degradation of collagen (Bailey 2003; Cooper 2001; Mitchell 1987; Vivian 1969). These changes are manifested in the tanned product by a number of undesirable characteristics such as deterioration of the grain surface, the presence of holes, loose grain, and blistering (Cooper 2001; Mitchell 1987). To date, no work has been published on the growth and persistence of microbial flora on ratite skins during storage. However, it is well established that the microbial flora able to colonise stored cattle hides includes both Gram-positive and Gram-negative bacteria (Everett & Cordon 1956; Hanlin et al. 1995; Kowalewski 1940). The growth of individual microbial species on stored skins does not necessarily indicate the micro-organism has a deleterious effect on skin quality. Their potential detrimental effects are likely to be determined by their ability to degrade collagen types I and III, which are the major constituents of skin (reviewed by Covington 1997; reviewed by Reich 1999). In general, Gram-negative bacteria are believed to be particularly detrimental to skin quality due to their high collagenolytic activity (Everett & Cordon 1956; Hanlin et al. 1995). However individual isolates of Gram-positive bacteria have also been shown to possess collagenolytic activity (Everett & Cordon 1956). Growth of fungi on raw skins has not been well documented. Therefore, it is not known if fungi cause deleterious effects on skin quality. An early study by Barghoorn (1950) demonstrated inoculation of chrome or vegetable tanned skins with Aspergillus niger resulted in growth of the fungus but no evidence of degradation of collagen was detected by histological examination. In an attempt to reduce salt use within the tanning industry numerous workers have explored techniques for short-term preservation of skins without salt (see Chapter Three). However, few studies have focused on the types of micro-organisms that colonise non-salted skins. Gram-positive bacilli and cocci have been shown to predominate on non-salted cattle hides immediately after slaughter (Hanlin et al. 1995). Following storage of unsalted cattle hides at 25°C for two days the composition of the bacterial population changes so that motile Gram-negative bacilli predominate (Hanlin et al. 1995). Hanlin et al. (1995) suggested that cattle hides should be chilled to prevent this shift in bacterial population towards potentially collagenolytic organisms. Most of the work on microbial colonisation of skins has been performed using short-term storage. It is not known what types of microbial flora persist during long-term storage of skins and how their growth is influenced by different storage conditions. However, a single study revealed long-term storage of salted cattle hides under cold conditions for one, two, three, four or five years does not significantly effect the yield, grade, tensile strength or chemical composition of the tanned product (Frey & Stuart 1941). The current study aimed to identify the types of micro-organisms that colonise ostrich skins during long-term storage under different treatment regimes. The effects of hide salt, storage temperature, antimicrobial pre-treatment, and ultraviolet irradiation on the type and persistence of microbial flora was determined. In addition, we aimed to determine the potential for the isolated microbial flora to degrade collagen within the skins. To achieve this aim, the gelatinase activity of selected isolates was assessed by their ability to liquefy gelatin in vitro. Isolates were subsequently inoculated onto swatches derived from a single raw ostrich skin and their ability to disrupt the ultrastructure of the skins after incubation for 45 days at room temperature was assessed by scanning electron microscopy.

41

4.3 Materials & methods 4.3.1 Microbial cultures 4.3.1.1 Isolation of micro-organisms from skins All whole (nine) and half-skins (20) from each of the six groups described in Chapter Three were swabbed for microbial flora at two, three, five and 16 weeks of storage. Dacron swabs were used to sample the exposed dermis. To ensure a representative sample of the microbial flora present, each swab sampled both fat and non-fat regions, and were stored at 4ºC for a maximum of 24 hours prior to culture. Swabs for microbial flora were cultured on a horse-blood agar plate, a MacConkey agar plate and a Sabouraud's agar slope with added chloramphenicol. The plates were incubated at 35oC in air for 48 hours and the Sabouraud's slope at 28oC in air for 14 days. Bacterial and fungal organisms were assessed by observing colonial characteristics of each morphological type followed by Gram-staining of bacterial colonies and lacto-fuchsin wet mounts of fungi. 4.3.1.2 Identification of bacteria Because of the large number of bacterial species isolated, most were identified in broad descriptive terms such as Gram-negative bacillus, coryneform and coagulase-negative staphylococcus. Some isolates were identified to genus or species levels. Gram-negative bacilli were biochemically characterised by oxidase activity, oxidation/fermentation reactions and by using API 20E and API 20NE systems according to the manufacturer's instructions (API, bioMerieux, Marcy l' Etaile, France). Gram-positive bacilli were identified on the basis of Gram-stain morphology and the presence or absence of spores as Bacillus species or coryneforms respectively. Gram-positive cocci were assessed for coagulase activity using the Staphytect Plus latex agglutination system (Oxoid, Basingstroke, Hampshire, England). 4.3.2 Collagenolytic ability of microflora 4.3.2.1 Gelatin liquefication Thirty-two isolates representing the different colonial types found on the skins from all groups at 16 weeks of storage were purified and tested for their ability to liquefy gelatin. Polycarbonate tubes (90 x 15 mm) containing 3 ml of 15% porcine gelatin in nutrient broth (Oxoid, Basingstroke, Hampshire, England) were inoculated with a straight wire stab containing a pure culture and then incubated at 22oC for 14 days. At the end of this period 16 cultures exhibiting positive or negative gelatin liquefication reactions were selected from the 32 isolates for further studies on the skins. 4.3.2.2 Inoculation of skin swatches Each of the 16 isolates were grown in Tryptone soya broth (Oxoid, Basingstroke, Hampshire, England) containing 0.6% yeast extract for 24 to 72 hours. Incubation was halted when a dense suspension was achieved. As growth of different organisms is known to vary according to temperature, we attempted to maximise growth by incubating the isolates at two different temperatures, at 28ºC and 35ºC. Although not quantitatively measured the density of microbial growth was estimated from experience to be 109 colony forming units/ml. These broth cultures were coded prior to inoculation of the skin swatches.

42

To minimise inter-skin variation a single skin was chosen for reapplication of the microflora. Because of the time constraints of this pilot study it was not possible to obtain a skin from a recently slaughtered ostrich. This would have been preferable as it is expected that a recently flayed skin would show minimal deterioration and/or damage and thus would maximise the potential to detect any adverse effects of the individual microbial isolates. In both the current skin storage experiment and the initial short-term storage study (see Chapter Two) skin quality was highest in salted skins stored at 4-6ºC. Consequently, a whole skin from Group five (skin number DO, Table 4.1) that had been stored since the time of slaughter at 4-6°C in hide salt was chosen. Two skin segments (10 × 8 cm) were cut from the back and rump regions using a sharp sterile surgical blade. This provided a replicate of the application of the broth cultures in two different regions of the skin. A total of 20 skin swatches (2 cm2) were cut from each of the original segments using sharp sterile surgical blades. With the replicate this gave a total of 40 swatches. Samples were placed in individual culture dishes (3.5 cm diameter, 1 cm deep) at room temperature (approximately 23°C). To avoid the swatches drying out, the culture dishes were placed with their lids slightly ajar in one of three semi-humid chambers. The temperature and humidity of each chamber was monitored daily using a digital hygrothermometer (Umcos Trading, South Australia). To ensure all the skin swatches had equilibrated to room temperature they were stored in the semi-humid chambers for seven days prior to placing 200 µl of a broth culture onto the dermal side of the swatch. Each swatch received a single broth culture. Four of the twenty swatches per skin segment (taken from the back and rump) served as controls. These received 200 µl of the broth devoid of the cultured isolates. Following their preparation from the single salted skin, none of the 40 swatches were re-salted during the reinoculation study. 4.3.3 Assessment of skin quality 4.3.3.1 Gelatin liquefication versus tissue ultrastructure For each of the isolates we compared their ability to liquefy gelatin and their potential to damage the skins. Skin damage was determined by qualitative assessment of the general tissue architecture as well as the organisation of collagen fibres using scanning electron microscopy. We attempted to determine whether any of the reapplied microbial isolates were associated with greater ultrastructural changes compared to other isolates. For a given isolate a direct correlation between the ability to liquefy gelatin and degradation of collagen in situ would suggest that they are prime contenders for causing skin damage during storage. Forty-five days after addition of the broth cultures a single 1 × 0.5 cm segment was excised from each of the skin swatches and fixed in 3% glutaraldehyde in 0.1M PBS. After fixation at room temperature for 3 hours, skin samples were washed three times in 0.1M PBS and stored in 0.1M PBS containing 0.1% sodium azide at 4°C. The segments were processed for scanning electron microscopy as described previously (see Chapter Two). Two samples from each skin swatch were examined at instrument magnifications between X50 and X5000.

43

4.4 Results Microflora colonising the skins over the 16 weeks of sampling are summarised in Table 4.1. For each storage group, the microbial flora on the salted skins showed minimal change between the third and fifth week of storage. Consequently, detailed assessments of organisms associated with the different skins were confined to the first few weeks (weeks two and five) and final swabbing at week sixteen (Table 4.1).

Table 4.1 Microbial flora from the skins during storage

Group number

(Treatment)a

Skin numberb

Number and types of micro-organisms isolatedc

Initial swab (week 2) Second swab (week 3) Final swab (week 16)

C0½R Staphylococcus (1) Micrococcus (1)

Staphylococcus (1) Micrococcus (1)

Staphylococcus (1) Bacillus (1)

C1½L Staphylococcus (1) Bacillus (1) Micrococcus (1) Zygomycete (1)

Staphylococcus (1) Bacillus (1) Micrococcus (1)

Micrococcus (2) Bacillus (1) Coryneform (1)

C2½L Staphylococcus (1) Micrococcus (1)


Bacillus (1)

C3½L Staphylococcus (1) Micrococcus (1) Bacillus (1)

Staphylococcus (1) Micrococcus (1) Bacillus (1)

Bacillus (1)

1 (S/4°C)

C4½L Staphylococcus (1) Micrococcus (1) Bacillus (1) Coryneform (1)


Bacillus (1) Staphylococcus (1)

C2½R Gram -ve bacillus (4) Bacillus (1) Bacillus (1) Staphylococcus (1) Coryneform (1) Viridans streptococcus (1)

C3½R Gram -ve bacillus (2) Bacillus (1) Staphylococcus (1)

No growth

A0½R Staphylococcus (2) Bacillus (1) Micrococcus (1) Gram -ve bacillus (1)

Bacillus (1)

Gram -ve bacillus (3)*


A1½R Staphylococcus (1) Bacillus (2) Micrococcus (1) Gram -ve bacillus (1)


Viridans streptococcus (1) Staphylococcus (1) Coryneform (1)

2 (S/Cl/4°C) †

A4½R Bacillus (2) Micrococcus (1)


Bacillus (1) Streptococcus (1) Coryneform (1)

44

A0½L Staphylococcus (1) Micrococcus (1) Bacillus (1)


No growth



No growth



No growth



No growth

3 (S/4°C)

A4½L Staphylococcus (1) Bacillus (1) Micrococcus (1)


Staphylococcus (1) Bacillus (1) Yeast (1) Micrococcus (1) Coryneform (1)

C0½L Gram -ve bacillus (6) Bacillus (1) Micrococcus (1)

Gram -ve bacillus (5)


C1½R Gram -ve bacillus (5) Gram -ve bacillus (4) Proteus (1)


C4½R Gram -ve bacillus (5) Staphylococcus (1)

Gram -ve bacillus (5) Gram -ve bacillus (4) Coryneform (1)

A2½R Gram -ve bacillus (4) Bacillus (1)

Gram -ve bacillus (5) Gram -ve bacillus (3) Yeast (2) Coryneform (1) Micrococcus (1)

4 (NS/CL/UV/4°C)

A3½R Gram -ve bacillus (4) Micrococcus (1) Bacillus (1)

Gram -ve bacillus (4) Gram -ve bacillus (3) Gram -ve cocco-bacillus (1) Coryneform (1) Yeast (1)

D0 Staphylococcus (1) Bacillus (1) Micrococcus (1)

Staphylococcus (1) Bacillus (1) Micrococcus (1)

Bacillus (1) Micrococcus (2) Coryneform (1)

D1 Bacillus (1) Micrococcus (1)



D2 Staphylococcus (1) Bacillus (1) Micrococcus (1)


Staphylococcus (1) Micrococcus (1) Bacillus (1) Streptococcus (1) Yeast (1)

5 (S/4°C)

D3 Bacillus (1) Micrococcus (1)


Staphylococcus (1) Micrococcus (1) Coryneform (1)

45

B0 Staphylococcus (3) Gram -ve bacillus (1)

Staphylococcus (1) No growth

B1 Staphylococcus (2) Proteus (1)


No growth

B2 Staphylococcus (2) Bacillus (1)


No growth

B3 Staphylococcus (1) Bacillus (1) Gram -ve bacillus (1) Proteus (1)


No growth

6 (S/Cl/RT) †

B4 Staphylococcus (1) Staphylococcus (1) Bacillus (1)

No growth

(a) Treatments were: NS, non-salted; S, salted; Cl, sodium hypochlorite; UV, ultraviolet

irradiation. Hypochlorite and UV treatments occurred after the initial swab. Storage temperature was either 4°C (4-6°C) or room temperature (RT).

(b) ½L & ½R indicate left or right half skins; Groups five and six were whole skins. (c) Numbers in parenthesis indicate the number of macroscopically different colonies based on pigment and colonial characteristics of each organism present per agar plate. † Unsalted for initial swab then salted for remainder of storage period. * Not present in swab taken in week five.

All Staphylococcus species were coagulase-negative 4.4.1 Time course of microbial flora with different storage regimes 4.4.1.1 Two weeks storage Skins unsalted from the time of slaughter Initial swab cultures from all non-salted skins in Groups two, four and six showed heavy growth of mixed Gram-negative bacilli, with up to six different colony types being apparent from one half-skin (C0½L). Gram-positive organisms including Bacillus species, Micrococcus species and coagulase-negative staphylococci (Figure 4.1) were also isolated from these skins. These were usually present in fewer numbers than the Gram-negative isolates. To gain an insight into the species of Gram-negative organisms present, the six isolates from skin coded C0½L were further identified using commercial identification kits (API 20E, API 20NE) and conventional biochemical tests. Figure 4.2 illustrates typical growth of a Gram-negative bacillus used for inoculating identification panels, and Figure 4.3 shows examples of two of these isolates inoculated into API 20NE panels. The six isolates were identified as Hafnia alvei, Enterobacter cloacae, Aeromonas species, Enterobacter amnigenus, Acinetobacter calcoaceticus-Acinetobacter baumannii complex and Acinetobacter lwoffii. Two of the Gram-negative isolates from skins in Group six (B1 and B3) were identified as Proteus species.

46

Figure 4.1 Blood agar culture showing typical heavy growth of numerous types of organisms from non-salted skins. Organisms include Gram-negative bacilli, coagulase-negative staphylococci, Micrococcus and Bacillus.

Figure 4.2 Example of a Gram-negative bacillus growing on MacConkey agar, used for identification tests in API kits.

Figure 4.3 API 20NE identification kits for Gram-negative bacilli illustrating different reactions by different species. Top - utilisation of carbohydrates (turbid cupules on right) and inability to liquefy gelatin (7th cupule from left). Bottom - inability to utilise carbohydrates and ability to liquefy gelatin.

47

Skins salted at slaughter Initial swab cultures revealed no growth of Gram-negative bacilli from the salted skins. Gram-positive bacteria predominated in the salted skins and these were of similar type to those found on the non-salted skins (Figure 4.4). A filamentous fungus of the class Zygomycetes was cultured from the initial swab taken from one of the salted skins (C1½L).

Figure 4.4 Typical light growth on blood agar of mixed Gram-positive organisms, predominantly coagulase-negative staphylococci and Micrococcus, found in cultures of salted skins. 4.4.1.2 Three to 16 weeks of storage Non-salted skins As a result of a developing odour and patches of discolouration the non–salted skins in Group four were rinsed in 1% sodium hypochlorite and irradiated with ultraviolet light at two weeks of storage (see Chapter Three). In these skins however, Gram-negative bacteria persisted in large numbers throughout the four month storage period. Although the number of different types of Gram-positive micro-organisms increased in variety, the quantity of overall growth remained about the same. Two additional Gram-positive organisms, coryneforms and yeasts, which were not present in the initial swabs, were isolated after four months of storage. Salted skins Skins that were salted and kept at 4-6°C for the entire storage period (Groups one, three and five) showed minimal changes in the type of microbial flora. Gram-positive bacteria persisted throughout the storage period (Table 4.1). A dermatiaceous fungus Alternaria sp. (Figure 4.5) was isolated from a single skin at week five (data not shown). Yeast species were isolated from the final swab of two of the salted skins (A4½L, Group three and D2, Group five). Gram-negative bacteria were never isolated from any skin that was salted from the commencement of storage. Four of the skins from Group three showed no microbial growth at the time of the final swabbing. Immersion in either Diamoll® C or sodium hypochlorite prior to storage was ineffective in preventing growth of bacteria, yeast or fungi. Initially Groups two, four and six were stored unsalted. As a result of a developing odour and patches of discolouration forming in some of the skins in Groups two and six, these were rinsed in 1 % sodium hypochlorite and coated in hide salt after the initial swabbing at two weeks of storage (see Chapter Three). They were then salted for the remainder of the storage period. Within one week of salting, most of the Gram-negative bacteria had disappeared. These remained in a single half-skin (A0½R) for

48

one week but were absent thereafter. This effect was independent of the storage temperature as Gram-negative bacteria were absent in skins stored either at 4-6°C or 22°C after five weeks. Gram-positive bacteria persisted following salting of these skins. Staphylococcus and Bacillus species were isolated from skins stored at either 4-6°C or room temperature. In contrast, Micrococcus and Streptococcus species were only found on the skins stored at 4-6°C (Group two). The group of salted skins treated with hypochlorite and kept at room temperature (Group six) showed the largest reduction in microbial flora, as no organisms were isolated at the final swabbing. Furthermore, two of these skins showed no growth following five weeks of storage (data not shown). All skins stored at room temperature had become very dry by the end of the 16 weeks of storage. In contrast all skins stored at 4-6°C (Groups one to five) remained moist and pliable.

Figure 4.5 Lacto-fuchsin stained wet-mount preparation of Alternaria sp. cultured on Sabouraud agar showing typical large conidia with both transverse and longitudinal septations (magnification 400x). 4.4.2 Gelatin liquefication Representative colonies from selected non-salted and salted skin cultures at the final swabbing were purified and their ability to liquefy gelatin determined (Table 4.2). This was used as a guide to select isolates for further testing of their ability to damage skins by degradation of collagen. Although gelatinase activity can be mediated by numerous proteases and may not per se indicate the presence of an enzyme capable of degrading collagen, strains that liquefied gelatin were considered as possible contenders for this activity. Of the 32 isolates tested, eight completely liquefied the gelatin in 14 days, five showed partial activity and two showed partial, weak activity. Eight of 11 (72%) strains from the non-salted skins showed gelatinase activity, whilst seven of 21 (33%) strains from salted skins showed this activity. Gram-stains of four of the isolates possessing gelatinase activity are shown in Figure 4.6. Although only a limited number of isolates were tested, these results suggest that microbial flora on non-salted skins are likely to possess greater proteolytic activity than that of salted skins. To determine if gelatinase activity of our isolates correlated with proteolytic activity on skins, eight randomly selected strains displaying varying degrees of gelatinase activity and eight strains without apparent gelatinase activity were coded and cultured in tryptone soya broth for further tests using samples of skin as substrates.

49

Figure 4.6 Gram stains of organisms liquefying gelatin; top left, coryneform bacteria; top right, yeast; bottom left, Gram-negative bacillus; bottom right, Bacillus (magnification 1000x).

50

Table 4.2 Gelatinase activity of representative isolates from selected skins four months post-treatment

Group number

(treatment)a

Skin number Organismb Gelatin hydrolysis (22°C)c

C0½R Micrococcus Gram +ve bacillus (filamentous) Coagulase -ve staphylococcus

- +w -

1 C1½L Micrococcus

Micrococcus Coryneform

- - -

C2½R Coryneform - A0½R Bacillus

Coagulase -ve staphylococcus Coagulase -ve staphylococcus

- +++ +++

2

A4½R Streptococcus +++

3

A4½L Yeast Bacillus Coryneform Micrococcus Coagulase -ve staphylococcus

- + - +w -

C4½R Coryneform

Gram -ve cocco-bacillus - +

A2½R Coryneform Yeast Micrococcus Gram -ve bacillus

+++ +++ + +++

4 A3½R Gram -ve bacillus

Coryneform Gram -ve bacillus Yeast Gram -ve cocco-bacillus

+++ - +++ + -

D0 Micrococcus

Micrococcus - +

5

D2 Streptococcus Yeast Bacillus

- - -

(a) Treatments are as given in Table 4.1. (b) Organisms given in bold were coded and selected for inoculation of skin segments (c) Gelatin, complete liquefication: +++; partial liquefication: +;

weak partial liquefication: +w, no liquefication: -.

51

4.4.3 Potential of microbial flora to degrade the skins The temperature and relative humidity of each chamber is given in Figure 4.7. The temperature ranged from 19.3 to 25.7°C and relative humidity ranged from 70 to 88% over the 62 days of storage. The skin swatches remained moist throughout this time with excess condensation being removed twice from the culture dishes to avoid over-saturation of the swatches.

Days

Rel

ativ

e H

umid

ity (%

)

0 10 20 30 40 50 60

70

75

80

85

90

Box 1Box 2Box 3

Days

Tem

pera

ture

°C

0 10 20 30 40 50 60

20

21

22

23

24

25Box 1Box 2Box 3

Figure 4.7 The temperature (°C) and relative humidity (%) of each semi-humid box is shown throughout the 62-day incubation period. Temperature and relative humidity were recorded at the same time each day. The broth cultures were added to the swatches at day 7 and samples were taken for scanning electron microscopy on day 52. Box 1 skin swatches 1 – 13; Box 2 skin swatches 14 – 26; Box 3 skin swatches 27 - 40.

52

4.4.3.1 Scanning electron microscopy Qualitative assessment of the skin swatches using scanning electron microscopy revealed varying levels of skin deterioration. However, there were no consistent differences between skin swatches inoculated with the broth controls and those inoculated with the various broth cultures. Keratin flakes overlying the epidermis exhibited varying degrees of pitting that was independent of the applied broth culture (Figure 4.8). Approximately one-third of the swatches showed evidence of microbial colonisation and/or deterioration of keratin demonstrated by the presence of globular masses scattered across the surface. These appeared in the broth controls as well as some swatches with reapplied microbial cultures. One swatch showed localised areas of keratin covered in an amorphous slime-like mass. This swatch had been inoculated with Gram-negative bacilli isolated from skin A3½R previously shown to completely liquefy gelatin in vitro (Table 4.2). This amorphous covering was only found in one of the two replicate swatches.

Figure 4.8 Scanning electron micrographs of the epidermal surface of three different skin swatches showing keratin flakes on the skin surface. (A) Inoculated with control broth. (B) Inoculated with Gram-negative bacilli isolated from skin A3½R (Group four) previously shown to exhibit complete gelatin liquefication in vitro (Table 4.2). An amorphous mass is shown covering isolated areas of the epidermal surface (arrows). (C) Inoculated with coryneform bacteria isolated from skin A3½R (Group four) which did not liquefy gelatin in vitro (Table 4.2). Scale bars are 50 µm. The dermis cut transversely to the epidermis exhibited the most extensive evidence of deterioration. The dermis was often scattered with clusters of globular material with finger-like projections (Figure 4.9). In extreme cases the dermis was completely covered in an amorphous slime-like mass. It is likely that this amorphous substance is a result of microbial colonisation of the swatches and subsequent degradation of skin proteins. Clusters of globular material were found in both the control swatches and the swatches with reapplied microbial cultures. In contrast, complete covering of the dermal surface occurred only in swatches with reapplied cultures. This effect was independent of the ability of the micro-organisms to liquefy gelatin in vitro. In some swatches the collagen fibres of the grain and corium were completely obscured by this amorphous substance. Consequently, the degree of shrinkage and splaying of collagen fibres could not be assessed in approximately half the skin swatches. When collagen fibres were visible some shrinkage spaces and splaying of fibres was observed in both the control swatches and those with reapplied microbial cultures.

53

Figure 4.9 Scanning electron micrographs of the dermis cut transversely to the epidermis of three skin swatches. e, epidermis; g, grain; c, corium. The actual borders between the epidermis, grain and corium regions are not distinguishable. Scattered over the surface are numerous clusters of fibrous and globular material (arrows). (A) Inoculated with control broth. (B) Inoculated with a Staphylococcus species isolated from skin A0½R (Group two) previously shown to completely liquefy gelatin in vitro (Table 4.2). An amorphous covering obscures the entire dermal surface. (C) Inoculated with a Micrococcus species from skin C0½R (Group one) which did not liquefy gelatin in vitro (Table 4.2). Scale bars are 50 µm. 4.5 Discussion The current study revealed a number of different micro-organisms colonised the ostrich skins during the four month storage period. Staphylococcus, Micrococcus, Bacillus and coryneform species were most commonly found. These Gram-positive bacteria were found to be salt-tolerant as they colonised both the non-salted and salted skins. In agreement with our work, Staphylococcus, Micrococcus, and Corynebacterium species have been reported to colonise stored cattle hides (Hanlin et al. 1995). Fungi (Alternaria sp. and Zygomycete sp.) and yeasts also appeared to be rare colonisers of the salted skins. Although yeasts were found in both salted and non-salted skins, the two types of fungi were only found on skins that were salted and stored at 4-6°C. The significance of fungal growth on these skins is unknown. Our preliminary findings show that Gram-negative bacilli are efficient colonisers of non-salted skins, however this does not appear to be at the expense of Gram-positive organisms. The usual habitat for the identified Gram-negative species (Hafnia alvei, Enterobacter cloacae, Aeromonas species, Enterobacter amnigenus, Acinetobacter calcoaceticus-Acinetobacter baumannii complex and Acinetobacter lwoffii) is the environment (soil, water, sewerage), animals and humans. The presence of Proteus spp. on the non-salted skins may be of particular significance as these Gram-negative motile bacteria have been reported to possess proteinase activity which can potentially cleave proteins in hides during storage (Hanlin et al. 1995). Although previous studies have suggested that storage of non-salted cattle hides at low temperatures may be effective at restricting the growth of Gram-negative bacteria (Hanlin et al. 1995), the current study demonstrates that a storage temperature of 4-6°C without salt does not prevent growth of these potentially proteolytic micro-organisms. Gram-negative bacteria colonising the non-salted skins were absent following salting. The results suggest these salt-sensitive Gram-negative bacilli may be at least in part responsible for the deterioration of the non-salted skins as sodium hypochlorite treatment and subsequent salting significantly reduced their putrefactive odour. Whether salted skins had undergone treatment or not with sodium hypochlorite did not seem to significantly alter the type or persistence of microbial flora over time.

54

No microbial growth was observed towards the end of the storage period in all skins stored salted at room temperature (Group six). It was unlikely that the lack of microbial growth following swabbing of these skins was due to human error as each of the six storage groups were swabbed on the same day for each storage time-point. A possible contributing factor for the lack of microbial growth may be that the drying out of these skins after four months of storage had created an osmotically unfavourable environment for the growth of micro-organisms. However, four out of five of the skins in group three (salted at 4-6ºC) also showed no microbial growth at the end of storage. These skins remained moist throughout the entire storage period. Within each group of Gram-negative bacilli, Gram-positive bacteria and yeasts some but never all isolates exhibited gelatinase activity. These isolates varied in their extent to liquefy gelatin. Each of these groups of microbial organisms have previously been reported to produce collagenolytic enzymes (Everett & Cordon 1956; reviewed by Harrington 1996; Nishimura et al. 2001; reviewed by Watanabe 2004; Woods et al. 1972). A higher percentage of the isolates from the non-salted skins possessed gelatinase activity compared to isolates from the salted skins. Furthermore, four out of five of the Gram-negative isolates from the non-salted skins showed gelatinase activity. This suggests that the Gram-negative isolates are particularly effective at liquefying gelatin. In support of these results, Everett and Cordon (1956) demonstrated that mixed cultures of Gram-negative bacilli and Gram-variable irregular bacilli showed a higher collagenase activity in collagen extracted from cowhide compared to other isolates. Similarly, a study of isolates from tannery soak water, septic tank effluent and surface water showed Gram-negative bacteria had the highest level of collagenase activity (Kowalewski 1940). The current study showed isolates from skins stored in hide salt at 4-6°C for the entire storage period had minimal gelatinase activity. In contrast, micro-organisms with relatively high gelatinase activity were isolated from the final swabbing of skins that were unsalted for the first two weeks and then treated with sodium hypochlorite, salted and stored at 4-6°C for the remainder of the storage period. This indicates that salting should be performed as soon as possible following flaying of the skins in order to reduce the growth of potentially collagenolytic micro-organisms. No correlation was found between the ability of a particular isolate to liquefy gelatin in vitro and degradation of collagen in situ. Scanning electron microscopic evaluation of skin swatches after application of the microbial cultures revealed extensive deterioration in all samples including controls. In extreme cases an amorphous slime-like covering obscured collagen fibres within both the grain and corium layers. To obtain optimal tissue preservation before application of the microbes it would have been preferable to select a skin immediately after slaughter. However, due to time constraints of this pilot study, this was not possible. Consequently, we chose a skin that had optimum preservation based on its histological appearance at 16 weeks storage (D0; salted, 4-6°C). However, it may have been subject to further degradation during continued storage prior to inoculation with the isolates. Therefore, although the isolates may differ in their ability to damage the skin, these differences may have been masked by pre-existing deterioration of the tissue. Alternatively, as the swatches were stored in a humid environment for 45 days after reapplication of the microbial cultures, damage caused by the microbial organisms may have been too extensive to allow detection of any differences in their ability to degrade the tissue. A further problem encountered was the masking of the tissue of some swatches by the extensive microbial colonisation 45 days after inoculation. Further work is required where skin swatches are inoculated with the microbial isolates immediately after slaughter and the time course of both microbial growth and changes in tissue structure determined. As a final comment, caution needs to be applied in interpretation of the ability of micro-organisms to liquefy gelatin as an indicator of their ability to degrade collagen in the skin. Everett and Cordon (1956) has shown a poor correlation between gelatin liquefication and ability to degrade collagen. Therefore, it cannot be assumed that organisms which break down denatured collagen in the form of gelatin will degrade collagen within the skin. Furthermore, the ability to degrade collagen may be species specific, such that microbial organisms able to degrade collagen in one species may have little

55

or no affect on a different species. In support of this argument, work in our laboratory using double-labelling immunohistochemistry, has shown antibodies specific for chicken collagen types I and III (Chemicon International, Inc. USA; Mauger et al. 1982) do not recognise collagen in emu skin (unpublished data). This suggests differences exist in the protein composition between ratite and chicken collagen.

Future more detailed studies on the identification to species level of organisms possessing gelatinase activity and the correlation of this activity with specific ostrich skin collagenolytic activity may assist in establishing control measures to prevent micro-organism-mediated damage during storage.

56

5. References Alberts, B, Johnson, A, Lewis, J, Raff, M, Roberts, K & Walter, P 2002, Molecular biology of the cell, 4th edition, Garland Publishing, Inc, New York & London. Angel, R, Trevino, L, Mantzel, T, Baltmanis, B, Blue-McLendon, A & Pollok, KD 1997, ‘Effect of ostrich age on hide quality’, American Ostrich,April, pp. 25-26. Bailey, DG 1995, ‘Preservation of cattle hides with potassium chloride’, Journal of the American Leather Chemists Association, vol. 90, pp. 13-21. Bailey, DG 2003, ‘The 44th John Arthur Wilson memorial lecture: The preservation of hides and skins’, Journal of the American Leather Chemists Association, vol. 98, pp. 308-319. Bailey, DG & Gosselin, JA 1996, ‘The preservation of animal hides and skins with potassium chloride’, Journal of the American Leather Chemists Association, vol. 91, pp. 317-333. Bailey, DG, DiMaio, GL, Gehring, AG & Ross, GD 2001, ‘Electron beam irradiation preservation of cattle hides in a commercial-scale demonstration’, Journal of the American Leather Chemists Association, vol. 96, pp.382-392. Barghoorn, ES 1950, ‘Histological study of the action of fungi on leather’, Journal of the American Leather Chemists Association, vol. 45, pp. 688-700. Barrett, JC 1986, ‘Short-term preservation of hides in Zimbabwe’, Journal of the Society of Leather Technologists and Chemists, vol. 70, pp. 83-89. Bezuidenhout, A J 1999, ‘Anatomy’, in DC Deeming (ed) The Ostrich Biology, Production and Health, CABI Publishing, London, pp. 13-49. Clark, Jr GA 1993, ‘Integumentum commune’, in JJ Baumel, AS King, JE Breazile, HE Evans & JC Vanden Berge (eds) Handbook of avian anatomy: Nomina Anatomica Avium, second edition, Nuttal Ornithological Club, Cambridge, Massachusetts, pp. 17-44. Cloete, SWP, van Schalkwyk, SJ, Hoffman, LC & Meyer, A 2004, ‘Effect of age on leather and skin traits of slaughter ostriches’, South African Journal of Animal Science, vol. 34, pp. 80-86. Cooper, RG 2001, ‘Ostrich (Struthio camelus var. domesticus) skin and leather: a review focused on southern Africa’, World's Poultry Science Journal, vol. 57, pp. 157-178. Covington, AD 1997, ‘Modern tanning chemistry’, Chemical Society Reviews, vol. 26, pp. 111-126. Ding, H 2004, ‘The relationship between the morphological change of the elastic fibres of goatskin and the effects of bating under different conditions’, Journal of the Society of Leather Technologists and Chemists, vol. 88, pp. 10-14. Everett, AL & Cordon, TC 1956, ‘Resistance of purified collagen to degradation by salt-tolerant bacteria in pure culture’, Journal of the American Leather Chemists Association, vol. 51, pp. 59-67. Frapple, P, O’Malley, P, Snowden, J & Hagan, R 1997, Emu processing and product development, Publication no. 97/66, Rural Industries Research and Development Corporation, Canberra.

57

Frey, RW & Stuart, LS 1941, ‘A five-year cold storage experiment with salted calfskins’, Journal of the American Leather Chemists Association, vol. 36, pp. 650-657. Glatz, PC 2001, ‘Effect of declawing on behaviour of ostrich chicks’, Proceedings of the 6th European Symposium on Poultry Welfare, Zollikofen, pp. 302-304. Hanlin, MB, Field, RA, Ray, B & Bailey, DG 1995, ‘Characterization of predominant bacteria in cattle hides and their control by a bacteriocin based preservative’, Journal of the American Leather Chemists Association, vol. 90, pp. 308-320. Harrington, DJ 1996, ‘Bacterial collagenases and collagen-degrading enzymes and their potential role in human disease’, Infection and Immunity, vol. 64, pp. 1885-1891. Hopkins, WJ, Bailey, DG, Weaver, EA & Korn, AH 1973, ‘Potential short-term techniques for the preservation of cattle hides’, Journal of the American Leather Chemists Association, vol. 68, pp. 426-436. Kobayashi, T, Hentzer, B & Asboe-Hansen, G 1977, ‘Degradation of dermal fibrillar structures: Effects of collagenase, elastase, dithioerythritol and citrate’, Acta Dermatovener (Stockholm), vol. 57, pp. 379-387. Kowalewski, JK 1940, ‘A microbiological study of animal skin. II. The action of proteolytic bacteria on collagen and elastin’, Journal of the American Leather Chemists Association, vol. 35, pp. 656-663. Kritzinger, CC 1950, ‘Preliminary investigations on the effects of prolonged storage upon wet-salted hides’, Journal of the American Leather Chemists Association, vol. 45, pp. 203-211. Lange, B 1929, ‘Uber enige besondere Formen des Faserverlaufes im Bindegewebe der Vogelhaut’, Anatomischer Anzeiger, vol. 67, pp. 452-459.

Lee, J-H, Ahn, S-H, Lee, E-M, Kim, Y-O, Lee, S-J & Kong, I-S 2003, ‘Characterization of the enzyme activity of an extracellular metalloprotease (VMC) from Vibrio mimicus and its C-terminal deletions’, FEMS Microbiology Letters, vol. 223, pp. 293-300. Looney, M, Kyratzis, I, Truong, Y & Wassenberg, J 2002, Enhancing the unique properties of kangaroo leather, Publication no. 02/105, Rural Industries Research and Development Corporation, Canberra. Lucas, AM & Stettenheim, PR 1972, Avian Anatomy – Integument, 2 vols, United States Department of Agriculture, Washington DC.

Lunam, CA & Glatz PC 2000, Declawing of farmed emus. Harmful or helpful? Publication no. 99/177, Rural Industries Research and Development Corporation, Canberra.

Lunam, CA, Weir, KA, Ruiz, J & Plummer, JL 2003, ‘Storage of green ostrich skin’, Proceedings of the Australian Poultry Science Symposium, vol. 15, p. 159.

MacNamara, K, Nicholas, P, Murphy, D, Riedel, E, Goulding, B, Horsburgh, C et al. 2003, Markets for skins and leather from the goat, emu, ostrich, crocodile and camel industries, Publication no. 02/142, Rural Industries Research and Development Corporation, Canberra. Marcar, N & Crawford, D 2003, Trees for saline landscapes, Publication no. 03/108, Rural Industries Research and Development Corporation, Canberra.

58

Mauger, A, Demarchez, M, Herbage, D, Grimaud, J-A, Druguet, M, Hartmann, D et al. 1982, ‘Immunofluorescent localization of collagen types I and III, and of fibronectin during feather morphogenesis in the chick embryo’, Developmental Biology, vol. 94, pp. 93-105. Menon, GK, Maderson, PFA, Drewes, RC, Baptista, LF, Price, LF & Elias, PM 1996, ‘Ultrastructural organization of avian stratum corneum lipids as the basis for facultative cutaneous waterproofing’, Journal of Morphology, vol. 227, pp. 1-13. Menon, GK & Menon, J 2000, ‘Avian epidermal lipids: Functional considerations and relationship to feathering’, American Zoologist, vol. 40, pp. 540-552. Mitchell, JW 1987, ‘Prevention of bacterial damage on brine cured and fresh cattlehides’, Journal of the American Leather Chemists Association, vol. 82, pp. 372-382. Montes, GS 1996, ’Structural biology of the fibres of the collagenous and elastic systems’, Cell Biology International, vol. 20, pp. 15-27. Nagano, K & To, KA 1999, ‘Purification of collagenase and specificity of its related enzyme from Bacillus subtilis FS-2’, Bioscience, Biotechnology and Biochemistry, vol. 63, pp. 181-183. Nishimura, M, Nikawa, H, Yamashiro, H, Nishimura, H, Hamada, T & Embery, G 2001, ‘Cell-associated collagenolytic activity by Candida albicans’, Mycopathologia, vol. 153, pp. 125-128. Peters, DE 1994, ‘The Australian Emu (Dromaius novaehollandiae). In P Smetana (ed) Emu Farming - background information, Department of Agriculture, Western Australia. Reich, G 1999, ‘The structural changes of collagen during the leather making processes - 1998 Atkin memorial lecture’, Journal of the Society of Leather Technologists and Chemists, vol. 83, pp. 63-79. Stettenheim, PR 2000, ‘The integumentary morphology of modern birds – An overview’, American Zoologist, vol. 40, pp. 461-477. Vivian, GW 1969, ‘The preservation of hides and skins against bacterial damage’, Journal of the American Leather Chemists Association, vol. 64, pp. 489-500. Watanabe, K 2004, ‘Collagenolytic proteases from bacteria’, Applied Microbiology and Biotechnology, vol. 63, pp. 520-526. Weir, KA 2004, ‘The microanatomy of emu skin and implications for the Australian emu industry’, PhD thesis, Flinders University. Weir, KA & Lunam, CA 2004, ‘A histological study of emu (Dromaius novaehollandiae) skin’, Journal of Zoology, vol. 264, pp. 259-266. Weir, KA, Plummer, JL & Lunam, CA 2003, ‘Optimising storage conditions of emu skin’, Proceedings of the Australian Poultry Science Symposium, vol. 15, p. 160. Woods, DR, Welton, RL, Thomson, JA & Cooper, DR 1972, ‘The microbiology of curing and tanning processes. Part IV. Collagenolytic activity of bacteria from raw and cured hides’, Journal of the American Leather Chemists Association, vol. 67, pp. 217-225. Zapletal, P, Szarek, J, Weglarz, A & Gardzina, E 1996, ‘Histological structure of hides and crossbreeds by Italian beef breed bulls and the physical properties of the leather’, Journal of the Society of Leather Technologists and Chemists, vol. 80.

Storage of Ostrich Skin

Documents