Study of the genetic basis of denitrification in pure culture denitrifiers isolated from activated sludge and soil Kim Heylen Dissertation submitted in fulfillment of the requirements for the degree of Doctor ( Ph.D.) in Sciences, Biotechnology December 2007 Promotor: Prof. Dr. Paul De Vos Co-promotor: Prof. Dr. Ir. Willy Verstraete
197
Embed
Study of the genetic basis of denitrification in pure culture denitrifiers ...
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Study of the genetic basis ofdenitrification in pure culture
denitrifiers isolated fromactivated sludge and soil
Kim Heylen
Dissertation submitted in fulfillment of the requirements for the degree of Doctor ( Ph.D.)
in Sciences, Biotechnology
December 2007
Promotor: Prof. Dr. Paul De Vos
Co-promotor: Prof. Dr. Ir. Willy Verstraete
EXAMINATION COMMITTEE
Prof. Dr. Savvas Savvides (chairman)
Faculty of Sciences, UGent, Ghent
Prof. Dr. Paul De Vos (promotor)
Faculty of Sciences, UGent, Ghent
Prof. Dr. Ir. W. Verstraete (co-promotor)
Faculty of Bioscience Engineering, UGent, Ghent
Prof. Dr. Ir. N. Boon
Faculty of Bioscience Engineering, UGent, Ghent
Prof. Dr. Ir. Mike S.M. Jetten
Faculty of Science, Radboud Universiteit Nijmegen, The Netherlands
Dr. C. Etchebehere
Faculty of Science, Universidad de la República, Uruguay
Prof. Dr. Anne Willems
Faculty of Sciences, UGent, Ghent
"Eindelijk ist gedaan. Oef!"
Dat is zowat het eerste gevoel dat in me opkomt. Niet dat de jaren hier steeds kommer
en kwel waren - ik heb zo genoten van de vele reizen, de leuke (en minder leuke) collega's, de
verhuis van het achtste naart vierde, de (zatte) recepties, de nieuwjaarsetentjes, de
vrijdagavonden bij Christiane, de feestweken... Achterafgezien is alles zeer vlot gegaan, maar
als je er middenin zit, ervaar je dat wel anders. Iemand heeft me ooit gezegd dat naarmate je
meer twijfelt, je beter bezig bent. Voor mij is dat een beetje de samenvatting van vier jaar
doctoreren. Eigenlijk ben ik voor het oog van mijn collega's volwassen geworden... En is het
nu tijd voor het serieuzere werk. Natuurlijk had ik dit doctoraat nooit alleen kunnen volbrengen.
Vele mensen hebben hun steentje bijgedragen, via wetenschappelijk input, of gewoon door er
te zijn in moeilijke en/of zotte momenten. Ik ben niet echt prozaïsch aangelegd, maar ik doe
een poging om iedereen te bedanken.
Eigenlijk is het allemaal begonnen met mijn master thesis bij Geert en Margo. Margo
heeft me al de kneepjes van het microbiologische vak geleerd, en is eigenlijk 'verantwoordelijk'
voor al mijn latere werk. Ik heb me tijdens die thesis super geamuseerd, niet in het minste door
Liesbeth Masco, Robin en Evie. En ik heb er een goei vriendin aan overgehouden, en mijn
ventje...
Na die thesis begon een doctoraat op het labo voor Microbiologie me wel aan te
spreken. Allé, na een beetje rondvragen, kon ik toch bij Paul een doctoraat starten rond
denitrificatie. Dus, een verhuis naar het achtste. Iedereen is dit natuurlijk al vergeten, maar ik
heb me daar toen een half jaar zeer rustig en stil gehouden (ja, dat kan ik ook). Mijn partner
in crime werd Bram. Gelukkig bezat hij alle kwaliteiten die ik nodig had in een buurman:
dezelfde voorliefde voor harde, ietwat cynische humor, bestand tegen mijn gezaag en geklaag
en in staat om me telkens weer vertrouwen te geven in mijn werk. Natuurlijk was er ook
Liesbeth Lebbe, onze ancien. Altijd goed gezind, zo zot als een achterdeur, een echte spring-
in-'t-veld, maar ook steeds bezorgd en een luisterend oor ter beschikking. En dan nog een
verdomd goei laborante. Zo maken ze er geen twee.
Als eerste wapenfeit moest ik van Paul helemaal alleen naar een congres in Marburg.
Tof! Maar eigenlijk was dit een fantastisch begin voor mijn doctoraat. Bij mijn terugkeer ben
ik met veel goesting en volle overgave begonnen aan mijn werk. Dit is het perfecte voorbeeld
van Pauls input voor mijn doctoraat: veel impulsen, veel vertrouwen en veel vrijheid. Waarvoor
veel dank. Natuurlijk waren Prof Verstraete en Nico er ook om mijn wetenschappelijke
wandel te begeleiden. Het was een unieke kans om te mogen samenwerken met een 'instituut'
als Prof. Verstraete. Uw eigenzinnige perceptie en uw immense creativiteit waren een zeer
goede leerschool. Bedankt voor uw vertrouwen in mij. En wat zou ik de laatste vier jaar
gedaan hebben zonder Nico. Steeds maar een telefoontje of emailtje weg, steeds bereid om te
helpen of om een zoveelste draft van een paper na te lezen. Nen hele dikke merci.
Alsof soort soort zoekt, werd ons labootje uitgebreid met nog enkele prettig gestoorde
figuren, zoals Caroline en Joachim, en later ook nog An en Emly. Zij zorgden samen met
Liesbeth voor een enorm plezante werksfeer. Soms was effectief 'werken' een beetje moeilijk
(tijdens het zingen van Strangers in the Night bijvoorbeeld), maar dat werd ruimschoots
gecompenseerd door leuke dynamiek en het sympathieke groepsgevoel. Jeroen Adam bracht
dan weer wat zen opt labo. An heeft gaandeweg een beetje de rol van Bram overgenomen en
fungeerde de laatste maanden als mijn klankbord. Blijkbaar kon ik toch niet zonder. Natuurlijk
zijn er nog andere collega's die een verschil hebben gemaakt. Bij de start van mijn doctoraat
had ik twee voorbeelden, en eigenlijk is dat niet meer veranderd. Door het niveau van hun
werk en hun aanpak heb ik altijd opgekeken naar Peetse en Tom Coenye. En door onze
babbels, was het nu op een vrijdagavond in een bruin café, na een nieuwjaarsetentje of in een
trendy bar in Toronto, hebben ze me een figuurlijke schop onder mijn kont gegeven en
aangezet tot nadenken. De andere doctoraatstudenten waren ook een steun en toeverlaat.
Vooral dan Ilse. Het was enorm leuk samen de toerist uit te hangen. En dan nog Jeanine en
Paul Segers, twee gemotiveerde mensen zonder wie ons labo nooit zo vlot zou draaien.
Uiteraard zijn er naast het werk nog enkele mensen die een groot verschil hebben
gemaakt en mijn leven de laatste vier jaar. Ira en Ariane zijn er altijd geweest. We hebben
elkaar misschien niet veel gezien, vooral het laatste jaar niet, maar gewoon weten dat jullie
steeds voor me klaarstonden als ik het weer eens niet zag zitten of om die eerste publicatie te
vieren, dat was voor mij enorm belangrijk. Sven, wat had ik zonder jou het laatste jaar gedaan.
Een huis verbouwen en doctoreren in hetzelfde jaar, weeral typisch voor mij. Godzijdank dat
je positieve ingesteldheid ons door alles doorsleurt. En last but not least, mijn ouders. Ik weet
eigenlijk niet hoe ik moet omschrijven wat zij voor mij betekenen en betekend hebben tijdens
mijn 'academische loopbaan'. Al de kansen die jullie mij gegeven hebben en jullie voortdurende
onvoorwaardelijke steun heeft mij gemaakt tot wie ik ben. Ik zie jullie graag…
Enorm bedankt,
Kim.
CONTENTS
Background and aim..........................................................................................................9
chemolithotrophs or phototrophs. And although the denitrification process is generally
anoxic, denitrifiers have different oxygen thresholds and some even need oxygen to perform
the process. For example Paracoccus pantotrophus (previously named Thiosphaera
pantotropha) can denitrify under complete air saturation [54], while denitrifying nitrifiers
are obligate aerobes and thus need oxygen to complete denitrification. These denitrifying
nitrifiers are a good example of chemolithotrophic denitrification, which was discovered
when production of N2O from nitrite by Nitrosomonas europaea was observed [71, 82].
Other ammonia-oxidizing bacteria can also perform both processes, suggesting that nitrifier
denitrification can be a universal trait among betaproteobacterial ammonium-oxidizing bacteria
[84]. All other major branches of the nitrogen cycle can also be associated with denitrification,
except ammonification, as mentioned above. Especially denitrification in diazotrophic
bacteria has been reported frequently, for example in Rhizobium [63], Bradyrhizobium
[99], Sinorhizobium [13], Azoarcus [113].
CHAPTER 1
18
1.3. KEY ENZYMES OF THE DENITRIFICATION PROCESS
The conversion of nitrite to nitric oxide by nitrite reductase is the crucial step in denitrification
because it converts fixed nitrogen to gaseous NO. This nitric oxide is further reduced to
N2O through the action of nitric oxide reductase. The control of both reductases in
denitrifying bacteria is regulated coordinately to assure removal of NO by the latter reductase
or, if this is not possible, by down-regulation of nitrite reduction [116]. The steady-state
concentration of free NO during denitrification is in the nanomolar range, because, although
NO metabolism is innate to denitrifiers, NO is also toxic for this group of bacteria.
1.3.1. Nitrite reduction
In denitrifying bacteria, two structurally different types of nitrite reductases occur, which
are distinguishable by their prosthetic groups, either cytochrome cd1 or copper. Both enzymes
are mutually exclusive within one cell. They can be present within the same genus, and
even the same species, as was reported for Alcaligenes faecalis [4]. Their interchangeability
is limited: CuNiR can replace cytochrome cd1 NiR, which was demonstrated using the
CuNiR encoding gene nirK from Pseudomonas aureofaciens in a mutationally cytochrome
cd1-free background of P. stutzeri [32], but requirements for heme d
1 biosynthesis makes
the reverse replacement impossible. It is generally assumed that cytochrome cd1 NiR is
numerically dominant in the environment, while CuNiR can be found in a greater variety of
physiological groups and bacteria from different habitats. These conclusions were based on
pure culture results and therefore could reflect biases of culture methods. Within the
proteobacteria, neither NiR enzyme has been found in exclusive association with a particular
taxon. However, to date, only CuNiR is observed in less conventional denitrifiers, such as
nitrifying bacteria [10], bacilli [20] or archaea [41, 42].
CuNiR are trimetric enzymes located in the periplasm. Each subunit is close to 40 kDa,
contains one type I and one type II Cu and comprises of two domains (Figure 1.2). The
type I Cu site is located on domain I, ligated to His95, Cys136, His145 and Met150
(numbering from “Achromobacter cycloclastes”), where its functional as the electron entry
site. The principal electron donors to CuNiR are azurin [21, 116] and pseudoazurin [48,
56], while cytochromes appear less frequently involved [81]. Type II Cu is the substrate-
binding site of nitrite reductase [55], which is coordinated by three histidines, His100,
His135, and His306, with the latter histidine provided by the adjacent subunit (Figure
1.2.). The electrons donated to the type I Cu center are transferred to the catalytic type II
Cu center through a chemical path involving Asp and His. In the reaction of CuNiR with
INTRODUCTION
19
nitrite to form nitric oxide, a cuprous nitrosyl complex functions as the key intermediate
[40, 92].
In contrast, the respiratory nitrite reductase of Bacillus halodenitrificans [20], B.
azotoformans [89] and Geobacillus stearothermophilus [39] is a dimeric, membrane-bound
Cu-containing enzyme with the catalytic side oriented towards the inside of the cell [98].
Its primary structure is homologeous to that of Gram-negative bacteria, also with type 1
and type 2 Cu. Their electron donors are not known.
A single copy nirK gene encodes CuNiR.
Figure 1.2. Copper-containing nitrite reductase from “Achromobacter cycloclastes”. Schematic representation ofthe trimer and position of Cu atoms. Full circles represent the six Cu atoms coordinated by cysteine (C), hystidine(H), and methionine (M). Domains I and II are related to domains I’, II’ and I”, II” by a crystallographic threefoldaxis. The type II Cu is bound by three histidines between the subunit interfaces. Taken from [114].
The cytochrome cd1 NiR is a homodimeric enzyme with a subunit mass of around 60 kDa,
located in the periplasm. The prosthetic groups are non-covalently bound heme c and
heme d1, which are both present in each subunit to render cytochrome cd
1 a tetraheme
protein. The smallest domain contains the heme c and acts as the electron transfer center.
This c-heme-binding peptide is located nearby the amino-terminus of the protein. A
hydrophobic patch on top of the heme c domain of cytochrome cd1, can form the docking
site for a complementary hydrophobic patch in electron donors (e.g. azurin, cytochrome
c551
, pseudoazurin), bringing the metal centers very closely together [108]. Not being
highly discriminatory, this type of recognition provides a rationale for the interchangeability
of electron donors observed among denitrification components. The heme d1 is localized in
the largest domain and is the catalytic center [107].
CHAPTER 1
20
The ligation of both hemes in oxidized state differs in different denitrifiers [94]. For example,
in Paracoccus panthotrophus, heme c binds to two histidines and heme d1 has two ligands,
a proximal histidine and a distal tyrosine. However, through reduction, the distal ligand of
heme d1 is lost, causing an exchange of one of the axial ligands of the heme c, replacing a
histidine by a methionine. In Pseudomonas aeruginosa, heme c is coordinated by one
histidine and one methionine, and does not modify after reduction, and the distal ligand of
heme d1 is a hydroxyl ion. The nitrogen atom of nitrite is bound to the iron of heme d
1,
while hydrogen bonds bind on of the oxygen atoms to two conserved histidines in the
distal pocket of heme d1. The physiological reaction of cytochrome cd
1 is the protonation
of nitrite and removal of water to yield NO.
The cytochrome cd1 nitrite reductase is encoded by the single copy nirS gene.
1.3.2. Nitric oxide reduction
The N-N bond formation takes place during the reduction of nitric oxide to nitrous oxide.
This step is catalyzed by the membrane-bound nitric oxide reductase (NOR). Three classes
of NOR’s have been identified in bacteria: cNOR, qNOR and qCuNOR (Figure 1.3.).
They mainly differ in electron donors and in the number and type of electron transfer
centers present. The active site is, however, thought to be highly homologous in these
three classes. Compared to the other denitrification enzymes, less is known about NOR
since no tridimensional structure is yet obtained and active site structural data has been
mainly inferred from spectroscopic studies [94]. So, the mechanism of NO reduction to
N2O is not clear, and reports in the literature are contradictory.
The cNOR is a membrane-bound cytochrome bc complex, composed of two subunits of
about 17 and 53 kDa [117]. NORC, the small subunit, is anchored to the cytoplasmic
membrane. It retains one heme c, which is ligated to one conserved histidine and one
methionine, and is responsible for mediating electron transfer from the periplasmic electron
donors, such as cytochrome c and cupredoxin, to the catalytic subunit (Figure 1.3A). NORB,
the largest subunit, contains two b hemes and one non-heme iron, forming one low-spin
heme b and a binuclear active site. The low-spin heme b will act as a provider of electrons
to the active site (Figure 1.3A), and is ligated to two conserved histidines. Four conserved
histidines are possible ligands of this binuclear site. Proximity, provided by the binuclear
site, appears to be a factor in N-N bond formation. The catalytic site is located closer to the
periplasmic than the cytoplasmic face of the membrane, releasing N2O into the periplasm.
This enzyme has only been found in denitrifying bacteria. Other names used are NORB or
INTRODUCTION
21
short-chain NOR (scNOR). The catalytic subunit is encoded by the single copy cnorB
gene.
While possessing a similar primary structure to cNOR, qNOR is a single subunit enzyme
(Figure 1.3B) that accepts electrons from quinols [17]. Primary sequence analysis shows
that qNOR is constituted by an N-terminus extension similar to the NORC subunit, and a
C-terminus region homologous to NORB subunit. This enzyme has been found in both
denitrifying and non-denitrifying bacteria [7], the latter being mostly pathogenic [36].
Other names used are NORZ or long-chain NOR (lcNOR). The qnorB gene encodes this
monomer NOR.
A third type of nitric oxide reductase, qCuANOR, was purified from Bacillus azotoformans
[90] and is formed by two subunits. The largest catalytic subunit is similar to the NORB
subunit, the smaller subunit does not posses a heme c but uses a copper A site to achieve
electron transport to the catalytic subunit (Figure 1.3C). This enzyme can use menaquinol
as well as cytochrome c551
as electron donor, the former is suggested to be active in NO
detoxification, while the latter would be functional in denitrification [90]. This qCuANOR
has only been found in B. azotoformans and the encoding genes are not yet identified.
Bacterial NO reductases belong to the superfamily of the heme-copper oxidases including
cytochrome oxidases, [36], which are the terminal enzymes of the aerobic electron transport
chain that catalyze the reduction of O2 to H
2O. The common phylogeny between heme-
copper terminal oxidases and bacterial NO reductases was proposed because of structural
similarities [35, 80, 100]: (i) the large catalytic subunit displays significant sequence
homology, (ii) crucial residues (including six metal-binding histidines) are conserved, (iii)
topology of the catalytic subunit predicts 12 transmembrane helices, and (iv) both enzyme
types contain a bimetallic center, consisting of a heme-iron and a second metal, which is Cu
in oxidases and qCuANOR, and Fe in the cNOR and qNOR. The common phylogeny is
supported by examples of both reductases using both O2 and NO as alternative electron
acceptors [26, 31].
CHAPTER 1
22
Figure 1.3. Schematic representation of NOR enzymes: (A) cNOR, (B) qNOR, and (C) qCuANOR. Dashedarrows represent the proposed electron transfer pathway from a periplasmic electron donor to the active site.
Taken from [94].
INTRODUCTION
23
1.4. CURRENT DENITRIFICATION RESEARCH
Knowledge on the physiology, biochemistry and molecular regulatory mechanisms of several
pure culture denitrifiers provided information to develop molecular tools for environmental
studies, allowing investigation of the unknown not-yet-cultured denitrifying diversity. As
a result, the last decade, denitrification research focused on environmental (culture-
independent) analysis.
Because of their taxonomic diversity, denitrifying bacteria could not be studied through the
conventional 16S rRNA gene sequence approach, but rather the functional genes were used
for cultivation-independent study. Primers were first developed for nirS and nirK genes [4,
34], so most denitrification research to date focuses on these nir genes. Environmental
studies assessed the nir sequence diversity through clone library sequencing [76, 79, 112],
T-RFLP studies [2, 3, 5, 72, 109, 111], and DGGE analysis [95], mainly in soil and marine
environments. Most studies revealed major environmental gene clusters, showing little
overlap with the clusters harboring genes from isolated strains, suggesting the presence of
yet uncharacterized denitrifiers in the environment [5, 12, 62, 72, 109]. Generally, nirK
sequences were retrieved from more diverse habitats but were more closely related, while
nirS amplicons could not always be obtained from environmental samples but seemed to be
very diverse [5, 72, 111], and represented spatially distinct sequence populations [5, 62].
Yet the reverse was also observed [64, 79].
Next to functional diversity, both quantification and activity of denitrifying bacteria in the
environment are essential to determine the influence of these microbial populations on the
overall denitrification process [69, 70]. Quantification of denitrifiers in the environment
was assessed based on nir genes through southern hybridization [59], PCR-based techniques,
such as competitive PCR [61, 73], MPN-PCR [61], and real-time PCR [37, 38, 49, 103], or
microarrays [93, 110]. This revealed denitrifier abundance between 104 to 109 nir gene
copies per gram soil, which was greatly underestimated by cultivation, but indicated a low
proportion of potential denitrifiers to total bacteria in soils [38]. Also recently, a group-
specific cnorB real-time PCR was developed, for quantification of Pseudomonas and rhizobia
in soil [19]. In spite of all these efforts, the presence of functional genes determined by
DNA probing only indicates the denitrifying potential, regardless of whether they are
retrieved from active denitrifying bacteria in this environment. So, also reverse transcription
PCR amplification was necessary to study only active denitrifiers in the environment on
mRNA level [62, 83].
CHAPTER 1
24
All these molecular approaches tried to correlate diversity, quantity and activity of the
denitrifying community to environmental controls (reviewed by Wallenstein [102]).
Unfortunately, other traits of the organism, not connected to denitrification, but rather to
its phylogeny, can also play a crucial factor in environmental selection. The correlation
between structural and functional biodiversity is one of the major challenges in microbial
ecology. But already in the first studies using functional probes, discrepancies were observed
between nirS gene diversity and phylogenetic diversity of denitrifying pseudomonads
[104, 105]. Functional gene sequences from a dozen or so bacterial genomes [67] and
halobenzoate degrading denitrifying isolates [86] confirmed that functional gene phylogeny
does not always relate to 16S rRNA gene phylogeny. Also, failure to detect nir genes in
denitrifying strains closely related to those in which nir genes were detected [86] suggested
a great nir gene diversity in highly related bacteria. In addition, denitrification genes nirK
and cnorB can be detected in pure culture nitrifiers [10, 11, 28, 45]. Functional genes in
nitrifier denitrifiers can be closely related to those of denitriers, making differentiation
between both metabolic guilds, i.e. denitrification and nitrification, based on functional
sequences very difficult.
Thus, sequence analysis of denitrification genes in cultivation-independent studies is very
useful to assess the functional diversity present in the environment. Yet, the functional
significance of biotic diversity is currently unknown due to the unclear phylogenetic
information content of the denitrification genes. However, to evaluate the importance of
organism diversity versus functional diversity for the denitrification process in the
environment, knowledge of the denitrifiers’ identity is indispensable. For this, researchers
need to return to the basis of the denitrification research, i.e. cultivation, because to date,
the collection of the bacterium is the most straightforward method to obtain information
that can correlate its denitrification capacity, functional genes and phylogenetic position.
INTRODUCTION
25
1.5. CONCEPTUAL FRAMEWORK OF THE THESIS
1.5.1. Cultivation-dependent research and alternatives
As mentioned above, cultivation is the easiest method to obtain information on both the
denitrification capacity, the functional genes and the overall phylogenetic position of an
organism. Unfortunately, it is generally known that cultivation does not retrieve all diversity
present in a given biotope due to ‘the great plate count anomaly’ [87], which is dependent
of the environment of choice [1]. However, DNA-based techniques are also not free of
biases: DNA extraction protocols are not suitable for all bacteria, PCR primers can be
biased towards specific bacterial groups, amplified genes might not be functional, and
extra-cellular DNA can persist in the environment. A renewed interest in cultivation has
yielded new approaches and insights to isolate not-yet-cultivated bacteria, inspired by the
necessity of the microorganism itself to gain in-depth understanding of its physiology or to
access its metabolic pathway [52].
Conventional agar plating on complex media selects for microorganisms that are fast-
growing, grow in high cell densities, are resistant to high nutrient concentrations and are
able to grow on solid media. Dilution culturing [8] or extinction culturing [15, 16] overcome
the interference of these fast-growing ‘bacterial weeds’ and allow isolation of previously
uncultured bacteria. Other techniques, such as diffusion growth chambers, incubate samples
in their natural environment by restricting cell movement but allowing chemical exchange
with the environment [47]. However, also simple adjustments, such as prolonged incubation
time, other solidifying agents, or lower nutrient concentrations can significantly increase
the cultivability of a sample [44, 78, 88].
Most described denitrifier cultivation studies were carried out using complex media. Tiedje
[95] recommended TSA supplemented with 0.1% KNO3 as the most favorable general
medium for isolation of most heterotrophic denitrifiers [23, 65], resulting in a growth rate
two times faster than nutrient broth plus nitrate [18, 57]. Nitrate broth and agar were also
used for soil studies [14, 27, 106]. Several denitrifier cultivation studies were performed on
pyridoxin hydrochloride, 4 mg folic acid, 1 mg lipoic acid in 100 ml 10 mM NaH2PO
4, at pH 7.1; 0, 1 or 2 ml addition
ISOLATION, CHARACTERIZATION AND IDENTIFICATION OF DENITRIFYING BACTERIA FROM THE ENVIRONMENT
40
of riboflavin solution (17) containing 2.5 mg riboflavin in 100 ml 25 mM NaH2PO
4, at pH 3.2; 0, 1 or 2 ml addition
of thiamine solution (17) containing 10 mg thiamine hydrochlorid in 100 ml 25 mM NaH2PO4, at pH 3.4 (0 ml; 1
ml; 2 ml); cobalamin solution (17) containing 50 mg cyanocobalamin per liter distilled water. A pH indicator was
added (10 µM): bromothymolblue for growth media with a pH of 6.5, phenol red for growth media with a pH of 7
or higher. Trypticase soy agar (Oxoid) was supplemented with 10 mM KNO3 and 10 µM phenol red.
IsolationA dilution series (100 to 10-8) of activated sludge was spread plated (100 µl) on 15 different growth media per batch,
as determined by the EA. The inoculated growth media were incubated for two weeks in an anaerobic chamber
(composition gas mixture 8%CO2/8%H
2/84%N
2). From each growth medium and supplemented TSA, 20 isolates
were picked from the highest dilution still showing growth and further purified and sub-cultured on the same medium
(G4M3 was tested in triplicate).
Denitrification testsAll purified isolates were incubated in liquid isolation medium for one week at isolation conditions. Tests for nitrate
and nitrite reduction were performed using the Griess reagents [27]. Selection for denitrifiers was based on the
results of the reduction tests and the pH indicator [19]. This selection approach was validated by the confirmation of
the denitrifying activity of all isolates of the first batch with N2O measurements. All isolates of the first batch
presumed to denitrify were grown in 50-ml culture flasks with 10 ml liquid isolation medium. The headspace of the
vials was replaced with filter-sterilised argon by evacuating five times and refilling. Acetylene (10%) was added to
stop the reduction of N2O to N
2. After one-week incubation, a gas sample (1 ml) was taken with a gas-tight syringe
and N2O was measured with a gas chromatograph (Shimadzu GC-14B) equipped with an electron capture detector,
a precolumn (1m) and a Porapak column (2m, 80-100 mesh).
Fatty acids methyl ester analysis (FAME)A qualitative and quantitative analysis of cellular fatty acid compositions was performed with the gas-liquid
chromatographic procedure as described by Sasser [26]. The resulting profiles were identified with the Microbial
Identification software (MIDI) using the TSBA database version 5.0 (Microbial ID, Newark, DE, USA). In batch 4,
some denitrifiers could not be grown in the standard conditions (medium and incubation time) for FAME analysis.
The genus identification was then obtained by 16S rRNA gene sequence analysis and used in the same way for the
determination of the diversity.
DNA extraction & 16S rRNA gene sequence analysisDNA was extracted from each denitrifying isolate using the guanidium-thiocyanate-EDTA-sarkosyl method described
by Pitcher et al. [22] for fast-growing strains and using alkaline lysis for slow-growing isolates. For alkaline lysis,
one colony was suspended in an eppendorf tube with 20 µl of lysis buffer (2.5 ml 10% SDS; 5 ml 1M NaOH; 92,5
ml MilliQ water). After 15 min at 95°C, 180 ml MilliQ water was added, the tube was centrifuged for 5 min at
13,000 g and the supernatant was transferred to a new tube. DNA extracts were stored at –20°C until use. PCR
amplification was performed as described by Heyrman et al. [9]. The PCR-amplified 16S rRNA gene products were
purified using the Nucleofast® 96 PCR system (Millipore). For each sequence reaction a mixture was made using 3
µl purified and concentrated PCR product, 1 µl of BigDye™ Termination RR mix version 3.1 (Applied Biosystems),
1.5 µl of BigDye™ buffer (5x), 1.5 µl sterile milliQ water and 3 µl (20 ng/µl) of one of the 6 sequencing primers
used [9]. The temperature-time profile was as follows: 30 cycles of denaturation for 15 s at 96°C, primer annealing
for 1 s at 35°C and extension for 4 min at 60°C. The sequencing products were cleaned up, as described by Naser et
al. [20]. Sequence analysis of the partial 16S rRNA gene (first 300-500 bp) was performed using an Applied Biosystems
3100 DNA Sequencer according to protocols provided by the manufacturer. Sequences were assembled using the
BioNumerics 4.0 software (Applied Maths). A reliable identification was obtained in two steps: (i) a BLAST search
[2] of the 16S rRNA gene sequence of an isolate retrieved 50 sequences with the highest sequence similarities to the
query sequence, (ii) all type strains of all species of all genera mentioned in the BLAST report were compared in an
exhaustive pair wise manner with the query sequence of each strain in BioNumerics 4.0. The strains were assigned
to a genus based on the obtained 16S rRNA gene sequence similarity.
Nucleotide sequence accession numbersThe nucleotide sequence data generated in this study have been deposited in Genbank/EMBL/DDBJ with accession
numbers AM083989 to AM084186.
CHAPTER 2
41
RESULTS
EA experiment
An evolutionary algorithm was used to optimize the isolation conditions for denitrifiers.
The influence of eleven medium parameters with different values and their combinations
on the number and diversity of isolated denitrifying bacteria was examined. Sixty different
growth media, i.e. combinations of medium parameter values, were investigated in four
subsequent batches, with 15 growth media per batch. Activated sludge from a municipal
wastewater treatment plant was used as inoculum. Table 2.1 gives an overview of the
composition and the fitness results of each growth medium per batch.
2,48
3,864,09
4,50
0,48 0,540,86 0,87
0
1
2
3
4
5
1 2 3 4
Batch
Fitn
ess
va
lue
Maxim al Fi tness Average Fi tness
Figure 2.1. The average and maximal fitness value for each batch of growth media. The fitness value of a growthmedium represents the success of a combination of medium parameters to render a high (relative) number of denitrifyingisolates, highly diverse in genus assignment.
The success of a growth medium was determined as a fitness value (Figure 2.1). This
fitness selected for (i) a high number of denitrifying bacteria, and (ii) a high diversity of
denitrifying bacteria (see Materials and Methods). For the first batch, the EA randomly
combined medium parameter values into 15 growth media. Batch 1 gave an average fitness
of 0.48. In total 269 isolates were examined and 34 were detected as denitrifiers. The
maximal fitness of batch 1 (i.e. 2.48) was assigned to growth medium G1M1, with a
nitrite concentration of 3mM, a molar C/N ratio of 20, with succinate as carbon source, no
sodium chloride or riboflavin added, addition of 1ml vitamin solution, 2 ml thiamine
solution and 2 ml cobalamin solution, a pH of 6.5 and incubated at 37°C. The EA calculated
a second batch, hereby selecting for those medium parameter values that contributed to a
high fitness in the previous batch. In batch 2, 217 isolates were examined, 33 isolates were
detected as denitrifiers and an average fitness of 0.54 was measured. Results of batch 1 and
ISOLATION, CHARACTERIZATION AND IDENTIFICATION OF DENITRIFYING BACTERIA FROM THE ENVIRONMENT
42
2 appeared very similar, except for the maximal fitness, which increased to 3.86 in batch 2
(Figure 2.1). Growth medium G2M11, giving the maximal fitness, differed from the best
scoring medium of batch 1 only in the pH, 7 instead of 6.5. Some growth media in batch 1
and 2 showed no growth, not even from the undiluted activated sludge sample, while
others showed growth but less than 20 colonies. This greatly limited the total number of
isolates and subsequently the number of denitrifiers in these batches. Batch 3 was calculated
based on the fitness results of batch 1 and 2. For this third batch, the average fitness
increased to 0.86 (Figure 2.1), 315 isolates were examined and 56 denitrifiers were detected,
a clear increase for all three features compared to batch 1 and 2. The maximal fitness (i.e.
4.09) was found for growth medium G3M12, differing from the two former best scoring
media in values of most medium parameters: a pH of 7.5, ethanol as carbon source, a low
molar C/N ratio of 2.5, a nitrate concentration of 18 mM, 1ml of thiamin solution, no
cobalamin solution added and incubation temperature of 20°C. The EA calculated batch 4
based on the three preceding batches. Again, an increased number of denitrifying bacteria
was isolated, 69 denitrifiers from a total of 300 examined isolates. The maximal fitness of
4.50 was assigned to G4M3, which differed from G2M11 only in the use of nitrate instead
of nitrite as a nitrogen concentration. Arbitrarily chosen, this growth medium was also
tested in triplicate to investigate the reproducibility of the evolutionary algorithm. The
fitness value differed between the three repeats, due to a difference in diversity of the
isolated denitrifiers (Table 2.1). The average fitness value (i.e. 0.87) reached a plateau in
batch 4, which led to the decision to stop the EA. Supplemented TSA was tested parallel
with each batch. The average fitness for supplemented TSA was 0.625.
CHAPTER 2
43
Table 2.1. Overview of all combinations of medium parameters tested in this study. Values for all medium parameterswere determined by the EA, based on the composition of the most successful combination in previous batches. Theresults of the experimental parameters of the EA are given for each growth medium: RATIO
den, Simpson’s reciprocal
diversity index 1/D and fitness. The average values per batch for the three parameters are also given.
ISOLATION, CHARACTERIZATION AND IDENTIFICATION OF DENITRIFYING BACTERIA FROM THE ENVIRONMENT
44
Experimental course of medium parameters
A detailed look at the experimental course of each medium parameter defined by the EA
revealed convergence to one ‘optimal’ value for five medium parameters (Figure 2.2).
The percentage of growth media with the same medium parameter value is directly correlated
with its contribution to a higher fitness in the preceding batches. Thus, a pH value of 7, a
nitrogen concentration of 3 mM, the addition of 1 ml of vitamin solution, and the exclusion
of sodium chloride and riboflavin solution contributed to the success of an elective growth
medium for denitrifiers (Figure 2.2). The other medium parameters diverged to different
values. Both temperature values were equally selected over four batches, with an increasing
preference for 20°C in batch 3 and 4. Cobalamin converged to either exclusion or the
addition of 2 ml. For the nitrogen source, both nitrite and nitrate were equally selected,
with an increasing preference for the latter in batch 4. For thiamine, all three possible
values were equally selected. Although no ‘optimal’ value could be determined, carbon
source and molar C/N ratio diverged to respectively two (i.e. ethanol and succinate) and
three values (i.e. 2.5, 20, 25), which were therefore more favourable for isolation of
denitrifiers than the other possible values. The best scoring growth medium of batch 1, 2
and 4 incorporated most or all of the ‘optimal’ values determined for the medium parameters;
only the composition of the best scoring medium of batch 3 deviated from this.
Figure 2.2. The percentage of growth media with a certain value for a medium parameter is represented for eachbatch. The experimental course of five medium parameters (A to E) converge to one value: pH (A), nitrogenconcentration (B), sodium chloride concentration (C), vitamin solution (D), and riboflavin solution (E). The percentageof growth media with the same value for a medium parameter is directly correlated with its contribution to a higherfitness in the preceding batches.
A.
0
10
20
30
40
50
60
70
80
1 2 3 4Batch
% o
f gr
owth
med
ia
6,5 7 7,5 8
B.
0
10
20
30
40
50
60
70
1 2 3 4Batch
% o
f gr
owth
med
ia
3mM 6mM 9mM
12mM 15mM 18mM
C.
0
10
20
30
40
50
60
70
80
90
1 2 3 4Batch
% o
f gr
owth
med
ia
0mM 0,34mM
D.
0
10
20
30
40
50
60
70
80
90
100
1 2 3 4Batch
% o
f gr
owth
med
ia
0ml 1ml 2ml
E.
0
10
20
30
40
50
60
70
80
90
1 2 3 4Batch
% o
f gr
owth
med
ia
0ml 1ml 2ml
CHAPTER 2
45
Diversity of the denitrifying populations in activated sludge
192 denitrifying isolates were distinguished in a total of 1101 isolates obtained on the 60
evaluated growth media, while seven out of 80 isolates obtained on supplemented TSA
were able to denitrify. After FAME analysis, 198 denitrifying isolates were reliably identified
onto genus level (Table 2.2) via partial 16S rRNA gene sequence analysis (no 16S rRNA
gene amplicon could be obtained for one isolate). The majority of the denitrifiers belonged
to the Betaproteobacteria (50.5% or 100 isolates). Sixty-eight strains were assigned to
Acidovorax, Alicycliphilus, Comamonas and Diaphorobacter of the Comamonadaceae,
and were isolated predominantly from growth media with ethanol or succinate as carbon
source, coupled with nitrate or nitrite as nitrogen source respectively. Thirty-one isolates
were assigned to Azospira, Azovibrio, Dechloromonas, Thauera and Zoogloea of the
Rhodocyclaceae, the majority of which were isolated on growth medium with succinate as
carbon source and pH value of 7. One isolate belonged to the genus Aquaspirillum of the
Neisseriaceae. The second largest group of denitrifiers belonged to the Alphaproteobacteria
(37.3% or 74 isolates): 22 isolates belonged to Brucella and Ochrobactrum of the
Brucellaceae, 8 isolates to Rhizobium and Sinorhizobium of the Rhizobiaceae, 43 isolates to
Paracoccus and Pannonibacter of the Rhodobacteraceae, and 1 isolate assigned to
Methylobacterium represented the Methylobacteraceae. The Gammaproteobacteria were
represented by 11 isolates belonging to Pseudomonas (5.6%). Four isolates (2%) belonging
to Arcobacter were Epsilonproteobacteria. Eight isolates (4%) belonging to Bacillus,
Trichococcus, Enterococcus, Paenibacillus and Staphylococcus were Firmicutes. One isolate
of the genus Chryseobacterium belonging to the Flavobacteriaceae represented the
Bacteroidetes. No clear trends were observed in the composition of the growth media used
for isolation of members of the Alpha-, Gamma-, Epsilonproteobacteria and Firmicutes.
Table 2.2. (next two pages) An overview of all cultured genera found in activated sludge from a municipal WWTP.All growth media used for isolation of each genus are given, together with the number of isolates.
ISOLATION, CHARACTERIZATION AND IDENTIFICATION OF DENITRIFYING BACTERIA FROM THE ENVIRONMENT
46
CHAPTER 2
Tax
on
om
ical
po
siti
on
T
ype
stra
in w
ith
hig
hes
t 16
S r
RN
A g
ene
seq
uen
ce s
imila
rity
to
qu
ery
seq
uen
ce
Gro
wth
med
ium
(C
lass
, Fam
ily, G
enu
s)
(str
ain
nu
mb
er,
% s
equ
ence
sim
ilari
ty,
acce
ssio
n n
um
ber
) [#
iso
late
s]
Alp
hap
rote
ob
acte
ria
Bru
cella
ceae
Bru
cella
B
ruce
lla o
vis
AT
CC
258
40T
99.5
%
L261
68
G3M
5 [1
]
Och
rob
actr
um
O
chro
bact
rum
ant
hrop
i D
SM
688
2T 97
.8%
-10
0%
D12
794
G1M
3 [1
], G
1M14
[3],
G2M
6 [1
], G
2M7
[1],
G3M
4 [5
], G
3M5
[3],
G3M
G
4M3[
1]
O
chro
bact
rum
inte
rmed
ium
LM
G 3
301T
99.2
%
U70
978
G2M
11 [1
], G
2M12
[1]
O
chro
bact
rum
triti
ci
DS
M 1
3340
T 10
0%
AJ2
4258
4 G
3M5
[2]
M
eth
ylo
bac
tera
ceae
M
eth
ylo
bac
teri
um
M
ethy
loba
cter
ium
suo
mie
nse
DS
M 1
4458
T 95
.1%
A
Y00
9404
G
2M4
[1]
R
hiz
ob
iace
ae
R
hiz
ob
ium
R
hizo
bium
gia
rdin
ii C
IP 1
0550
3T 97
.2%
U
8634
4 G
3M12
[1]
R
hizo
bium
gal
licum
M
SD
J110
9T 97
.6%
U
8634
3 G
1M15
[1]
R
hizo
bium
rad
ioba
cter
A
TC
C 1
9358
T 97
.2%
- 1
00%
A
J389
904
G3M
2 [1
], G
1M15
[1]
R
hizo
bium
sul
lae
DS
M 1
4623
T 97
.6%
Y
1017
0 G
1M15
[1]
Sin
orh
izo
biu
m
Sin
orhi
zobi
um m
orel
ense
LC
04T
97.0
% -
97.
2%
AY
0243
35
G1M
1 [1
], G
2M11
[1],
G2M
12 [1
]
Rh
od
ob
acte
race
ae
P
arac
occ
us
P
arac
occu
s am
inop
hilu
s
AT
CC
496
73T
97.6
%
AY
0141
76
G1M
3 [2
]
Par
acoc
cus
alca
liphi
lus
AT
CC
511
99T
97.6
% -
97.
8%
AY
0141
77
G1M
3 [1
], G
2M7
[1],
G3M
4 [1
], G
3M7
[1],
G3M
12 [1
], G
4M15
[2],
G1
P
arac
occu
s am
inov
oran
s A
TC
C 4
9632
T 97
.4%
- 9
9.7%
D
3224
0 G
1M14
[3],
G2M
3 [3
], G
3M4
[1],
G3M
5 [1
], G
4M3
[1],
G4M
12 [5
], G
4M
Par
acoc
cus
caro
tinifa
cien
s E
-396
T 98
.4%
- 9
8.7%
A
B00
6899
G
3M4
[1],
G3M
7 [2
]
Par
acoc
cus
pant
otro
phus
A
TC
C 3
5512
T 10
0%
Y16
933
G3M
5 [1
], G
3M14
[2]
P
arac
occu
s ye
ei
CC
UG
468
22T
97.8
%
AY
0141
73
G3M
4 [1
], G
3M12
[1],
G3M
13 [1
]
Par
acoc
cus
vers
utus
A
TC
C 2
5364
T 99
.9%
- 1
00%
A
Y01
4174
G
1M7
[1],
G2M
4 [1
], G
2M7
[1],
G3M
5 [1
]
P
ann
on
ibac
ter
Pan
noni
bact
er p
hrag
mite
tus
DS
M 1
4782
T 10
0%
AJ4
0070
4 G
3M2
[1]
Bet
apro
teo
bac
teri
a
C
om
amo
nad
acea
e
Aci
do
vora
x
Aci
dovo
rax
aven
ae s
ubsp
. citr
ulli
AT
CC
296
25T
98.1
% -
98.
3%
AF
0787
61
G2M
6 [1
], G
2M7
[1]
A
cido
vora
x de
fluvi
i D
SM
126
44T
99.4
% -
100
%
Y18
616
G2M
15 [1
], G
3M2
[1],
G3M
10 [2
], G
3M11
[1],
G3M
12 [1
], G
3M13
G4M
13 [1
], G
4M14
[1]
A
cido
vora
x te
mpe
rans
A
TC
C 4
9665
T 97
.8%
- 9
9.5%
A
F07
8766
G
1M1
[4],
G1M
8 [1
], G
2M15
[2]
Alic
yclip
hilu
s
Alic
yclip
hilu
s de
nitr
ifica
ns
DS
M 1
4773
T 97
.7%
- 9
9.8%
A
J418
042
G1M
1 [3
], G
3M1
[1]
Co
mam
on
as
Com
amon
as a
quat
ica
AT
CC
113
30T
99.0
% -
99.
8%
AJ4
3034
4 G
3M1
[2]
C
omam
onas
den
itrifi
cans
A
TC
C 7
0093
6T 98
.3%
- 9
9.9%
A
F23
3877
G
1M15
[1],
G2M
4 [1
], G
2M7
[2],
G2M
9 [1
], G
2M11
[1],
G3M
1 [2
G
3M3
[2],
G3M
9 [1
], G
3M15
[1],
G4M
4 [1
], G
4M5
[3],
G4M
8 [3
], G
4M
D
iap
ho
rob
acte
r
Dia
phor
obac
ter
nitr
ored
ucen
s D
SM
159
85T
99.2
% -
99.
8%
AB
0643
17
G1M
1 [3
], G
2M11
[2],
G3M
1 [1
], G
3M12
[1],
G4M
14 [1
]
47
ISOLATION, CHARACTERIZATION AND IDENTIFICATION OF DENITRIFYING BACTERIA FROM THE ENVIRONMENT
N
eiss
erac
eae
A
qu
asp
irill
um
A
quas
piril
lum
met
amor
phum
D
SM
183
7T 98
.9%
Y
1861
8 G
4M3
[1]
R
ho
do
cycl
acea
e
Azo
spir
a A
zosp
ira o
ryza
e LM
G 9
096T
99.9
% -
100%
A
F01
1347
G
2M11
[1],
G4M
3 [1
]
A
zovi
bri
o
Azo
vibr
io r
estr
ictu
s LM
G 9
099T
100%
A
F01
1346
G
2M9
[1]
Dec
hlo
rom
on
as
Dec
hlor
omon
as a
gita
ta
AT
CC
700
666T
100%
A
F04
7462
G
4M3
[4],
G4M
7 [2
]
Dec
hlor
omon
as d
enitr
ifica
ns
DS
M 1
5892
T 97
.2%
- 9
9.8%
A
J318
917
G4M
3 [1
], G
4M6
[1]
Th
auer
a
Tha
uera
aro
mat
ica
DS
M 6
984T
99.2
% -
99.
3%
X77
118
TS
A [2
]
Tha
uera
am
inoa
rom
atic
a D
SM
147
42T
99.5
% -
100
%
AJ3
1567
7 G
4M3
[5],
G4M
7 [2
], G
4M8
[2]
T
haue
ra c
hlor
oben
zoic
a A
TC
C 7
0072
3T 99
.4%
A
F12
3264
T
SA
[1]
T
haue
ra m
eche
rnic
hens
is
DS
M 1
2266
T 98
.0%
Y
1759
0 G
2M13
[1]
T
haue
ra p
heny
lace
tica
DS
M 1
4743
T 95
.0%
- 9
9,5%
A
J315
678
G2M
13 [1
], G
3M15
[1],
G4M
6 [2
], T
SA
[1]
T
haue
ra s
elen
atis
A
TC
C 5
5363
T 99
.1%
Y
1759
1 T
SA
[1]
Zo
og
loea
Z
oogl
oea
ram
iger
a A
TC
C 1
9544
T 96
.4%
X
7491
3 G
4M6
[1]
Gam
map
rote
ob
acte
ria
Pse
ud
om
on
adac
eae
P
seu
do
mo
nas
P
seud
omon
as a
erug
inos
a A
TC
C 1
0145
T 99
.9%
- 1
00%
A
F09
4713
G
1M1
[1],
G2M
11 [1
]
Pse
udom
onas
alc
alig
enes
A
TC
C 1
4909
T 10
0%
Z76
653
G2M
6 [2
], G
2M7
[1]
P
seud
omon
as m
endo
cina
A
TC
C 2
5411
T 95
.3%
- 9
5.6%
A
J308
310
G3M
9 [1
], T
SA
[1]
P
seud
omon
as n
itror
educ
ens
AT
CC
336
34T
99.1
% -
99.
3%
D84
021
G1M
10 [1
], G
3M9
[1]
P
seud
omon
as p
utid
a A
TC
C 1
2633
T 98
.9%
A
J308
313
G4M
9 [1
]
Pse
udom
onas
stu
tzer
i A
TC
C 1
7588
T 10
0%
AF
0947
48
G2M
11 [1
]
Ep
silo
np
rote
ob
acte
ria
Cam
pyl
ob
acte
race
ae
Arc
ob
acte
r A
rcob
acte
r cr
yaer
ophi
lus
CC
UG
178
01T
99.3
% -
99.
8%
L146
24
G4M
6 [2
]
Arc
obac
ter
skirr
owii
AT
CC
511
32T
94.6
%
L146
25
G4M
6 [1
]
Arc
obac
ter
nitr
ofig
ilis
AT
CC
333
09T
95.4
%
L146
27
G4M
6 [1
] F
irm
icu
tes
Bac
illac
eae
B
acill
us
B
acill
us c
laus
ii
AT
CC
700
160T
99.7
%
X76
440
G4M
3 [1
]
Bac
illus
moj
aven
sis
AT
CC
515
16T
98.6
% -
98.
7%
X68
416
G1M
4 [1
]
G1M
8 [1
]
Car
no
bac
tera
ceae
Tri
cho
cocc
us
Tric
hoco
ccus
floc
culif
orm
is
DS
M 2
094T
100%
A
J306
611
G4M
6 [1
]
En
tero
cocc
acea
e
En
tero
cocc
us
E
nter
ococ
cus
cass
elifl
avus
A
TC
C 2
5788
T 99
.2%
- 1
00%
A
F03
9903
G
1M5
[1]
G
2M11
[1]
P
aen
ibac
illac
eae
P
aen
ibac
illu
s P
aeni
baci
llus
agar
idev
oran
s D
SM
135
5T 98
.6%
A
J345
023
G3M
12 [1
]
Sta
ph
ylo
cocc
acea
e
Sta
ph
ylo
cocc
us
S
taph
yloc
occu
s ho
min
is
AT
CC
278
44T
99.9
%
L37
601
TS
A [1
] B
acte
roid
etes
F
lavo
bac
teri
acea
e
Ch
ryse
ob
acte
riu
m
Chr
yseo
bact
eriu
m g
leum
A
TC
C 3
5910
T 94
.8%
A
Y46
8449
G
2M6
[1]
48
DISCUSSION
Little is known about the denitrifying diversity present in activated sludge, as straightforward
cultivation-independent approaches are not suitable and cultivation-dependent research is
limited. Magnusson et al. [18] performed an isolation campaign on nutrient agar with
activated sludge from five different municipal WWTPs and found only denitrifying
Proteobacteria, belonging to the Rhodobacteraceae, Comamonadaceae and
Pseudomonadaeae. After applying 60 different defined isolation conditions, a much more
important heterotrophic denitrifier diversity was found, although proteobacteria were still
predominant. Denitrifying representatives of Alpha-, Beta-, Gamma- and
Epsilonproteobacteria, Firmicutes and Bacteroidetes were found, and apart from genera
conventionally known to harbour denitrifiers such as Pseudomonas, Ochrobactrum,
Comamonas and Acidovorax, genera less frequently observed in cultivation studies of
denitrifiers were also encountered. The Rhodocyclaceae were well represented, encompassing
besides the genus Thauera also the recently described genera Azospira and Azovibrio [24],
and Dechloromonas [1, 11]. Furthermore, possibly new species belonging to Thauera and
Zoogloea were retrieved. Recent efforts to identify denitrifiers in activated sludge in a
cultivation-independent manner by combining fluorescent in situ hybridisation (FISH)
with microautoradiography (MAR) [35] recognised the Azoarcus-Thauera group of the
Rhodocyclaceae as probably the most abundant denitrifiers in industrial WWTPs. The
genus Arcobacter was previously found in significant numbers in activated sludge [28], but
its function was undetermined. In this study, four denitrifying Arcobacter strains were
isolated, demonstrating that the genus can contribute to the denitrification process in
activated sludge systems. The denitrifying potential of Bacteroidetes and Firmicutes strains
- Bacillus, Paenibacillus, Staphylococcus, Trichococcus and enterococci - known from
cultivation-independent studies to be numerically less important in WWTPs than the
Proteobacteria [13], was also established.
This study shows the applicability of an EA for the optimization of growth media. The
progressive improvement of the average and maximal fitness value in each successive
batch confirms the iterative nature of an EA. The maximal fitness value of each batch of
the newly designed media was significantly higher than the average fitness of supplemented
TSA, which is a widely applied growth medium for denitrifier isolation [33]. Highly suitable
elective growth media were developed, rendering between 40 and 80% heterotrophic
denitrifiers. Comparable data are unavailable for cultivation-dependent studies on activated
sludge; for soil, 10% of all isolates on supplemented nitrate broth were denitrifiers [7].
After evaluation of 60 different combinations of medium parameters, the three best scoring
CHAPTER 2
49
growth media G2M11, G3M12 and G4M3, can be recommended for isolation of denitrifiers
in the future.
The isolation conditions for denitrifiers were optimised heuristically. Convergence of a
medium parameter to one value indicates no interaction with other medium parameters.
The EA determined that five medium parameters converged to one ‘optimal’ value. Because
of their independence of the overall medium composition, these parameters can be fixed at
these values in further optimization studies, while other medium parameters are varied.
Although halotolerant and halophilic denitrifiers are known [16], exclusion of sodium
chloride appeared to increase the isolation of denitrifiers. This observation may be correlated
with the use of activated sludge as inoculum. Riboflavin did not result in an enhanced
retrieval of denitrifiers, which contradicts an earlier report on the reduction of the doubling
time for Paracoccus denitrificans when adding riboflavin under denitrifying conditions
[4]. The same study showed an increase in the nitrite reductase activity, thus decreasing
the accumulation of nitrite, with ethanol as carbon source. The suitability of ethanol as
carbon source for denitrifiers was also confirmed here. In contrast to previous optimization
studies in microbiology with an EA [5, 8, 36], the reproducibility of the fitness was assessed.
The observed non-reproducibility of the genus diversity determination was probably
attributed to (i) the limited number of investigated isolates per growth medium, due to
logistics and time, (ii) the use of FAME analysis for genus identification, and/or (iii) other
possible parameters, not included in the EA.
Weuster-Botz [37] stated that ‘a combination of highly directed random searches to explore
the n-dimensional variable space with a genetic algorithm, and subsequent application of
classical statistical experimental design is recommended for media development’. The here
reported work can be seen as the initial step for elective medium design and development
for denitrifying bacteria and provides the basis for further cultivation-dependent research
on denitrifiers. Furthermore, through this study, new growth media are available that favour
growth of heterotrophic denitrifiers exhibiting a high natural diversity. Also, a large set of
denitrifying isolates has been obtained that can be further subjected to research concerning
denitrification, e.g. functional gene sequence analysis. Similar large-scale cultivation studies
can have future value for physiological interesting bacterial groups that are difficult to
study, for instance filamentous or nitrifying bacteria.
ISOLATION, CHARACTERIZATION AND IDENTIFICATION OF DENITRIFYING BACTERIA FROM THE ENVIRONMENT
50
REFERENCES1. Achenbach LA, Michaelidou U, Bruce RA, Fryman J, Coates JD (2001) Dechloromonas agitata gen. nov.,
sp. nov. and Dechloromonas suillum gen. nov., two novel environmentally dominant (per)chlorate-reducing
bacteria and their phylogenetic postion. Int J Syst Evol Microbiol 51:527-533
2. Altschul SF, Gish W, Miller W, Myers EW, Lipman DJ (1990) Basic local alignment search tool. J Mol Biol
215:403-410
3. Amann, RI, Ludwig W, Schleifer K-H (1995) Phylogenetic identification and in situ detection of individual
microbial cells without cultivation. Microbiol Rev 59:143-169
4. Blaszczyk M (1993) Effect of medium composition on the denitrification of nitrate by Paracoccus denitrificans.
Appl Environ Microbiol 59:3951-3953
5. Boon N, Depuydt S, Verstraete W (2006) Evolutionary algorithms and flow cytometry to examine the
M, Swings J (2005) Phylogeny and identification of Enterococci using atpA gene sequence analysis. J Clin
Microbiol 43:2224-2230
21. Ozturk H K, Ceylan H, Canyurt OE, Hepbaslin A (2005) Electricity estimation using genetic algorithm approach:
a case study of Turkey. Energy 30:1003-1012
22. Pitcher DG, Saunders LA, Owen NA (1989) Rapid extraction of bacterial genomic DNA with guanidium thio-
cyanate. Lett Appl Microbiol 8:151-156
23. Phillipot L (2002) Denitrifying genes in bacterial and Archaeal genomes. Biochim Biophys Acta 1577:355-
376
24. Reinhold-Hurel B, Hurek T (2000) Reassessment of the taxonomic structure of the diazotrophic genus Azoarcus
sensu lato and description of three new genera and new species, Azovibrio restrictus gen. nov., sp. nov.,
CHAPTER 2
51
Azospira oryzae gen. nov., sp. nov. and Azonexus fungiphilus gen. nov., sp. nov. Int J Syst Evol Microbial
50:649-659
25. Rösch C, Mergel A, Bothe H (2002) Biodiversity of denitrifying and dinitrogen-fixing bacteria in an acid
forest soil. Appl Environ Microbiol 68:3818-3829
26. Sasser M (1990) Identification of bacteria by gas chromatography of cellular fatty acids. MIDI Technical
Note #101. www.midi-inc.com
27. Smibert RM, Krieg NR (1994) Phenotypic characterization. In: Methods for General and Molecular
Bacteriology, p. 649. Edited by Gerhardt P, Murray RGE, Wood WA, Krieg NR, Washington: American
Society for Microbiology
28. Snaidr J, Amann R, Huber I, Ludwig W, Schleifer K-H (1997) Phylogenetic analysis and in situ identification
of bacteria in activated sludge. Appl Environ Microbiol 63:2884-2896
29. Stanier RY, Palleroni NJ, Doudoroff M (1966) The aerobic Pseudomonads: a taxonomic study. J Gen Microbiol
43:159-271
30. Stevenson BS, Eichorst SA, Wertz JT, Schmidt TM, Breznak JA (2004) New strategies for cultivation and
detection of previously uncultured microbes. Appl Environ Microbiol 70:4748-4755
31. Sys K, Boon N, Verstraete W (2004) Development and validation of evolutionary algorithm software as an
optimization tool for biological and environmental applications. J Microbial Meth 57:309-322
32. Throbäck IN, Enwall K, Jarvis Å, Hallin S (2004) Reassessing PCR primers targeting nirS, nirK and nosZ
genes for community surveys of denitrifying bacteria with DGGE. FEMS Microbiol Ecol 49:401-417
33. Tiedje JM (1988) Ecology of denitrification and dissimilatory nitrate reduction to ammonium. In:
Environmental microbiology of anaerobes, Zehnder AJB (ed.), John Wiley & Sons, New York, pp.179-244
34. Tyson GW, Banfield JF (2005) Cultivating the uncultivated: a community genomics perspective. Trends
Microbiol 13:411-415
35. Wagner M, Loy A (2002) Bacterial community composition and function in sewage treatment systems. Curr
Opin Biotechnol 13:218-227
36. Weuster-Botz D, Wandrey C (1995) Medium optimization by genetic algorithm for continuous production of
formate dehydrogenase. Process Biochem 30:63-571
37. Weuster-Botz D (2002) Experimental design for fermentation media development: statistical design or global
random search? J Biosc Bioeng 90:73-483
38. Zumft WG (1997) Cell biology and molecular basis of denitrification. Microbiol Mol Biol Rev 61:533-616
ISOLATION, CHARACTERIZATION AND IDENTIFICATION OF DENITRIFYING BACTERIA FROM THE ENVIRONMENT
2.2 DIVERSITY OF HETEROTROPHIC DENITRIFIERSISOLATED FROM SOIL USING A MULTIPLE-MEDIASET
Redrafted from: Heylen K, Boon N, Verstraete W, De Vos P (2007) Functional gene study
on heterotrophic denitrifiers isolated from soil. Microbiol Ecol, Submitted
SUMMARY
Culturing bacteria from environmental samples is still the most straightforward way to
identify the denitrifying diversity present. The aim of this study was to generate more
knowledge on the denitrifying diversity present in soil. Therefore, a multiple-media set,
formed by three defined growth media, developed in a previous study to electivly isolate of
denitrifiers, and the commonly used complex medium TSA, was applied. A total of 112
denitrifiers were isolated and identified through FAME analysis and partial 16S rRNA gene
sequence analysis. Among the retrieved denitrifiers, bacilli, mostly isolated with nitrite as
nitrogen source, were numerically dominant (71%), next to rhizobia (16%). The four media
used in the multiple-media set were compared and evaluated on (i) a high ratio denitrifiers-
to-anaerobes, (ii) a high denitrifier diversity, both retrieved from (iii) the highest dilution
containing denitrifiers. Growth medium G3M12, containing nitrate and ethanol and incubated
at 20°C, scored best. However, the use of a multiple-media set is recommended to cover a
wide denitrifying diversity.
54
INTRODUCTION
Denitrification, i.e. the respiratory reduction of nitrate and nitrite to N2O and nitrogen gas,
is an important bacterial process in soil. It has since long been a concern in agriculture
because nitrogen is often most limiting in crop yields, and fertilizer losses due to
denitrification commonly range between 20 and 30 % (Tiedje 1988). Recently, its production
of N2O has renewed interest in the process, since N
2O contributes to global warming and
the greenhouse effect. Knowledge on the composition of the denitrifying community would
enable in situ monitoring of key players and their response to different environmental
controls, which could lead to management models of soil resulting in optimal fertilizer use
and decreasing N2O emissions.
However, this physiological trait is difficult to study culture-independently. The identity
of a soil bacterium does not yield information on its denitrification capacities, due to the
phylogenetically widespread occurrence of this trait [23]. In addition, the functional gene
sequences contain limited taxonomic information [17]. Other approaches that combine
identifying the taxonomic position and metabolism of microorganisms, such as fluorescent
in situ hybridisation in combination with microautoradiography [12] or stable-isotope
probing [19] are not (yet) suitable for large-scale diversity studies and/or for the scoped
physiological characteristics. As a result, the contributions of different microbial taxa to
the denitrification process in soil are scarcely known [13].
Conventional culturing approach makes a valid alternative for cultivation-independent
analysis, as the denitrifier itself is available for both phylogenetic and functional study. The
aim of this study was to confirm that the newly developed media can recover a higher
diversity of denitrifiers, compared to conventional TSA. Therefore, a multiple-media set,
formed by three defined growth media, developed in a previous study to electivly isolate of
denitrifiers [9], and the commercial complex medium TSA, was applied. The usefullness of
these media to generate new insights in denitrifying diversity in soil was assessed.
CHAPTER 2
55
MATERIALS & METHODS
Inoculum and isolation conditionsA soil sample was collected from the upperlayer (1-20 cm) of a luvisol test field in Melle, Belgium. The texture class
of the test field is sand-loam (composition 8.6% clay/11.6% loam/75.8% fine sand/4% rough sand). The soil matrix
was suspended in a sterile solution of 1% NaPO3 (3 g soil in 300ml) by stirring at room temperature [4]. A dilution
series (10-2 to 10-10) of the matrix was spread plated (100 µl) on four growth media: (i) TSA (Trypticase Soy Agar,
Oxoid), supplemented with 10 mM potassium nitrate and 10 µM phenol red, incubated at 37°C; (ii) G2M11,
containing 3mM potassium nitrite, 15 mM sodium succinate and 10 µM phenol red, at pH 7 and supplemented with
vitamins, incubated at 37°C; (iii) G3M12, containing 18 mM potassium nitrate, 22.5 mM ethanol, 10 µM phenol
red, at pH 7.5 and supplemented with vitamins, incubated at 20°C; and (iv) G4M3, containing 3 mM potassium
nitrate, 15 mM sodium succinate, 10 µM phenol red, at pH 7, and supplemented with vitamins, incubated at 37°C.
Media were solidfied with 1.7% agar. Defined growth media G2M11, G3M12 and G4M3 were developed and
described previously [9]. The inoculated growth media were incubated for two weeks in an anaerobic chamber
(composition gas mixture 8%CO2/8%H
2/84%N
2). From each medium, visibly different colony types were randomly
picked and further purified by sub-culturing under isolation conditions.
Nitrogen reduction testsNitrate and nitrite reduction tests were performed in liquid isolation medium as described by Smibert & Krieg [22].
N2O measurements were performed to confirm denitrification. All isolates were grown in 50-ml culture flasks with
10 ml liquid isolation medium. The headspace of the vials was replaced with filter-sterilized argon by evacuating five
times and refilling. Acetylene (10%) was added to stop the reduction of N2O to N2. After one-week incubation, a gas
sample (1 ml) was taken with a gas-tight syringe and N2O was measured with a gas chromatograph (Shimadzu GC-
14B) equipped with an electron capture detector, a precolumn (1 m) and a Porapak column (2 m, 80-100 mesh).
Fatty Acids Methyl Ester (FAME) analysisQualitative and quantitative analyses of cellular fatty acid compositions for all denitrifiers were performed by the
gas-liquid chromatographic procedure described by Sasser [20]. The whole-cell FAME profiles were identified and
quantified with the Microbial Identification System software package (MIS, TSBA database version 5.0).
DNA extraction & 16S rRNA gene sequence analysisDNA was extracted from each isolate using the alkaline lysis method. For alkaline lysis, one colony was suspended
in an eppendorf tube with 20 µl of lysis buffer (2.5 ml 10% SDS; 5 ml 1M NaOH; 92.5 ml MilliQ water). After 15
min at 95°C, 180 ml MilliQ water was added, the tube was centrifuged for 5 min at 13,000 g and the supernatant was
transferred to a new tube. DNA extracts were stored at –20°C until use. The 16S rRNA gene PCR amplification was
performed as described by Heyrman et al. [10]. The PCR-amplified 16S rRNA gene products were purified using the
Nucleofast® 96 PCR system (Millipore). For each sequence reaction a mixture was made using 3 µl purified and
concentrated PCR product, 0.5 µl of BigDye™ Termination RR mix version 3.1 (Perkin Elmer), 1.75 µl of BigDye™
buffer (5x), 1.75 µl sterile milliQ water and 3 µl (20 ng/µl) of one of the 6 sequencing primers. The sequencing
products were cleaned up, as described by Naser et al. [15]. The temperature-time profile was as follows: 30 cycles
of denaturation for 15 s at 96°C, primer annealing for 1 s at 35°C and extension for 4 min at 60°C. Sequence analysis
of the partial 16S rRNA gene (first 300-500 bp) was performed using the Applied Biosystems 3100 DNA Sequencer
according to protocols provided by the manufacturer. Sequences were assembled using the BioNumerics 4.6 software
(Applied Maths). A reliable genus identification was obtained in two steps: (i) a BLAST search [1] of the 16S rRNA
gene sequence of an isolate retrieved 50 sequences with the highest sequence similarities to the query sequence, (ii)
the 16S rRNA gene sequence of all type strains of all species mentioned in the BLAST report were compared in an
exhaustive pair wise manner with the query sequence in BioNumerics 4.6. Isolates were assigned to a genus based
on the FAME analysis and the obtained 16S rRNA gene sequence similarity.
Nucleotide sequence accession numbersThe sequence data generated in this study have been deposited in Genbank/EMBL/DDBJ. The accession numbers
of the 16S rRNA genes can be found in the Table 2.3.
ISOLATION, CHARACTERIZATION AND IDENTIFICATION OF DENITRIFYING BACTERIA FROM THE ENVIRONMENT
56
RESULTS & DISCUSSION
A previous screening of sixty defined growth media showed the high suitability of three
media, G2M11, G3M12 and G4M3, for elective isolation of denitrifiers [9]. Thus far, their
applicability has only been assessed with activated sludge as inoculum. Here, the three
growth media are used for the cultivation of denitrifying isolates from soil. TSA
supplemented with nitrate was also included, as complex commercial growth media are
still frequently used to isolate denitrifiers. This study focused on heterotrophic denitrifiers
– defined growth media contained between 45 and 60 mM carbon – able to start
denitrification from either nitrate, the nitrogen source in G3M12, G4M3, and TSA, or
nitrite, the nitrogen source in G2M11.
Isolation and identification of heterotrophic soil denitrifiers
A total of 249 isolates were obtained from plating the soil suspension on the four media.
112 isolates were capable of performing denitrification in their isolation medium. Half of
the denitrifiers (50,9% or 57 isolates) were retrieved from G2M11, 20,5% (23 isolates)
from G3M12, 22,3% (25 isolates) from G4M3, and 6,3% (7 isolates) from TSA. All
denitrifiers were identified at the genus level through a biphasic identification approach,
using both FAME analysis (data not shown) and partial 16S rRNA gene sequence analysis.
An overview of the isolates, their identification, and other isolation information is included
in Table 2.3.
The majority of the denitrifiers belonged to the Firmicutes (77% or 83 isolates). Three
isolates were assigned to two different Staphylococcus species, while eighty isolates belonged
to Bacillus, showing high sequence similarities with seven different species. The denitrifying
capacity of Bacillus has long been recognized, but is not well investigated [21]. Other
cultivation studies also identified members of Bacillus as as soil denitrifiers [4,24]. Here,
bacilli were mostly retrieved from medium G2M11, the only medium with nitrite as nitrogen
source, indicating a possible preference for starting denitrification from nitrite. Although
this hypothesis is hard to substantiate because literature almost exclusively reports on
(commercial) denitrification tests including nitrate, nitrite tolerance or nitrite dependence
of many Bacillus strains has been described previously [24]. Most of the isolated denitrifying
bacilli were highly related with the dominant active bacteria found in Drentse A grassland
soils through direct ribosome isolation [5], namely B. drentensis, B. bataviensis and B. soli
[11].
CHAPTER 2
57
The other 29 denitrifiers belonged to the Proteobacteria. All 18 isolates of the
Alphaproteobacteria belonged to the family of the Rhizobiaceae, with representatives of
Ensifer, Rhizobium and Sinorhizobium. Rhizobial denitrification is speculated to yield
selective advantage for nodulation in situ through (i) neutralization of the nitrate inhibition
of nodulation, and (ii) generation of ATP for N2 fixation, but also enables survival and
growth of free-living rhizobia when other energy sources are limited [16]. Rhizobia have
not been frequently isolated in cultivation-dependent denitrification studies, despite their
widespread occurrence and a high persistence in agricultural and other soil [16]. Their
underrepresentation in denitrifier isolation campaigns could be attributed to the used growth
media: rhizobia do not grow very well on general complex media [23]. In this study,
rhizobia were mostly isolated from G3M12, containing nitrate and ethanol. Only three
representatives of Cupriavidus and eight of Pseudomonas were isolated. The numerical
dominance of Beta- and Gammaproteobacteria among soil isolates as described previously
[4,6,24] was not observed in our study. Also, this can probably be attributed to the
isolation media used.
ISOLATION, CHARACTERIZATION AND IDENTIFICATION OF DENITRIFYING BACTERIA FROM THE ENVIRONMENT
58
Table 2.3. Identification of all isolated denitrifiers. Isolation medium and dilution is given. Shading indicates thedenitrifying taxa found in the highest dilution on each isolation medium.
S tra in n r G e n us M e d ium D ilu tio n A c c es s io n n r T yp e s tra in with h ig h e s t 16 S rR NA g e n e s equ en c e id e n tific a tio n s im ila r ity to q u e ry s e q ue n ce (% s e q . s im ilar ity, s p ec ie s n am e, s tra in n u m b e r, a cc e ss io n n um be r)
R -3 1 55 3 B a c illu s s p. G 2 M 1 1 -3 AM 6 9 1 5 7 6 9 9 % B ac illus s oli L M G 218 3 8 T (AJ5 4 25 13) R -3 3 77 3 B a c illu s s p. G 2 M 1 1 -3 AM 6 9 1 5 7 9 9 9 ,9% Ba cillus d ren te ns is L M G 21 8 31 T (AJ 5 4 25 06 ) R -3 2 85 1 B a c illu s s p. G 2 M 1 1 -3 AM 6 9 1 5 8 0 1 0 0% B ac illus d ren te n sis L M G 2 18 31 T (AJ 5 4 250 6) R -3 3 77 4 B a c illu s s p. G 2 M 1 1 -3 AM 6 9 1 5 9 2 9 8 ,6% Ba cillus d ren te ns is L M G 21 8 31 T (AJ 5 4 25 06 ) R -3 1 55 4 B a c illu s s p. G 2 M 1 1 -3 AM 6 9 1 6 0 5 9 8 ,6% Ba cillus d ren te ns is L M G 21 8 31 T (AJ 5 4 25 06 ) R -3 1 54 6 B a c illu s s p. G 2 M 1 1 -3 AM 6 9 1 6 0 4 9 9 % B ac illus ba ta v ie ns is L M G 21 83 3 T (AJ 5 42 50 8 ) R -3 2 70 5 B a c illu s s p. G 2 M 1 1 -2 AM 4 0 3 6 3 4 9 8 ,8% Ba cillus d ren te ns is L M G 21 8 31 T (AJ 5 4 25 06 ) R -3 2 84 9 B a c illu s s p. G 2 M 1 1 -2 AM 4 0 3 6 3 5 9 8 ,8% Bacillu s so li L M G 21 8 38 T (AJ 5 42 51 3) R -3 2 78 1 B a c illu s s p. G 2 M 1 1 -2 AM 4 0 3 6 5 1 9 9 % B ac illus ba ta v ie ns is L M G 21 83 3 T (AJ 5 42 50 8 ) R -3 2 77 9 B a c illu s s p. G 2 M 1 1 -2 AM 4 0 3 6 5 0 9 9 % B ac illus ba ta v ie ns is L M G 21 83 3 T (AJ 5 42 50 8 ) R -3 2 70 9 B a c illu s s p. G 2 M 1 1 -2 AM 4 0 3 6 3 1 9 9 ,1% Ba cillus b a tav ien sis L M G 2 18 33 T (AJ 5 4 25 08 ) R -3 2 70 6 B a c illu s s p. G 2 M 1 1 -2 AM 4 0 3 5 8 5 1 0 0% B ac illu s lic he n ifo rm is LM G 12 36 3 T (X 68 416 ) R -3 2 71 5 B a c illu s s p. G 2 M 1 1 -2 AM 4 0 3 6 1 1 9 9 ,1% Bacillu s so li L M G 21 8 38 T (AJ 5 42 51 3) R -3 1 76 9 B a c illu s s p. G 2 M 1 1 -2 AM 4 0 3 6 4 0 1 0 0% B ac illu s lic he n ifo rm is LM G 12 36 3 T (X 68 416 ) R -3 2 70 0 B a c illu s s p. G 2 M 1 1 -2 AM 4 0 3 6 1 0 9 8 ,6% Ba cillus b a tav ien sis L M G 2 18 33 T (AJ 5 4 25 08 ) R -3 1 77 0 B a c illu s s p. G 2 M 1 1 -2 AM 4 0 3 6 1 8 9 9 % B ac illus ba ta v ie ns is L M G 21 83 3 T (AJ 5 42 50 8 ) R -3 2 78 7 B a c illu s s p. G 2 M 1 1 -2 AM 4 0 3 6 2 3 9 9 % B ac illus ba ta v ie ns is L M G 21 83 3 T (AJ 5 42 50 8 ) R -3 2 78 9 B a c illu s s p. G 2 M 1 1 -2 AM 4 0 3 6 3 7 9 8 ,7% Ba cillus d ren te ns is L M G 21 8 31 T (AJ 5 4 25 06 ) R -3 2 51 6 B a c illu s s p. G 2 M 1 1 -2 AM 6 9 1 5 5 7 9 8 ,9% Bacillu s so li L M G 21 8 38 T (AJ 5 42 51 3) R -3 2 52 1 B a c illu s s p. G 2 M 1 1 -2 AM 6 9 1 5 5 8 9 8 ,9% Bacillu s so li L M G 21 8 38 T (AJ 5 4 25 13 ) R -3 2 52 3 B a c illu s s p. G 2 M 1 1 -2 AM 6 9 1 5 5 9 9 8 ,9% Ba cillus b a tav ien sis L M G 2 18 33 T (AJ 5 4 25 08 ) R -3 2 70 3 B a c illu s s p. G 2 M 1 1 -2 AM 6 9 1 5 6 0 9 8 ,9% Ba cillus b a tav ien sis L M G 2 18 33 T (AJ 5 4 25 08 ) R -3 1 85 6 B a c illu s s p. G 2 M 1 1 -2 AM 6 9 1 5 6 1 9 8 ,8% Bacillu s so li L M G 21 8 38 T (AJ 5 4 25 13 ) R -3 2 69 4 B a c illu s s p. G 2 M 1 1 -2 AM 6 9 1 5 6 3 9 9 ,1% Bacillu s so li L M G 21 8 38 T (AJ 5 42 51 3) R -3 3 81 9 B a c illu s s p. G 2 M 1 1 -2 AM 6 9 1 5 6 4 9 9 % B ac illus d re n te ns is L M G 2 1 83 1 T (AJ 5 42 50 6 ) R -3 2 71 7 B a c illu s s p. G 2 M 1 1 -2 AM 6 9 1 5 6 9 9 9 % B ac illus d re n te ns is L M G 2 1 83 1 T (AJ 5 42 50 6 ) R -3 2 70 4 B a c illu s s p. G 2 M 1 1 -2 AM 6 9 1 5 7 0 9 8 ,9% Bacillu s so li L M G 21 8 38 T (AJ 5 4 25 13 ) R -3 3 77 6 B a c illu s s p. G 2 M 1 1 -2 AM 6 9 1 5 7 1 9 8 ,9% Ba cillu s n ov alis L M G 21 83 7 T (AJ 5 42 512 ) R -3 2 70 7 B a c illu s s p. G 2 M 1 1 -2 AM 6 9 1 5 7 4 9 8 ,7% Ba cillus b a tav ien sis L M G 2 18 33 T (AJ 5 4 25 08 ) R -3 2 52 6 B a c illu s s p. G 2 M 1 1 -2 AM 6 9 1 5 7 8 9 9 % B ac illus s o li L M G 2 183 8 T (AJ 54 25 1 3 ) R -3 2 69 5 B a c illu s s p. G 2 M 1 1 -2 AM 6 9 1 5 8 1 9 9 ,1% Ba cillus d ren te ns is L M G 21 8 31 T (AJ 5 4 25 06 ) R -3 2 71 3 B a c illu s s p. G 2 M 1 1 -2 AM 6 9 1 5 8 2 9 8 ,7% Bacillu s so li L M G 21 8 38 T (AJ 5 42 51 3) R -3 2 70 8 B a c illu s s p. G 2 M 1 1 -2 AM 6 9 1 5 8 3 9 9 ,1% Ba cillus b a tav ien sis L M G 2 18 33 T (AJ 5 4 25 08 ) R -3 2 52 8 B a c illu s s p. G 2 M 1 1 -2 AM 6 9 1 5 8 5 9 8 ,6% Ba cillus d ren te ns is L M G 21 8 31 T (AJ 5 4 25 06 ) R -3 2 53 3 B a c illu s s p. G 2 M 1 1 -2 AM 6 9 1 5 8 6 9 8 ,7% Ba cillus b a tav ien sis L M G 2 18 33 T (AJ 5 4 25 08 ) R -3 2 52 2 B a c illu s s p. G 2 M 1 1 -2 AM 6 9 1 5 8 7 9 8 ,9% Ba cillus b a tav ien sis L M G 2 18 33 T (AJ 5 4 25 08 ) R -3 2 69 6 B a c illu s s p. G 2 M 1 1 -2 AM 6 9 1 5 8 8 9 8 ,6% Ba cillus d ren te ns is L M G 21 8 31 T (AJ 5 4 25 06 ) R -3 2 51 7 B a c illu s s p. G 2 M 1 1 -2 AM 6 9 1 5 9 0 9 8 ,7% Ba cillus b a tav ien sis L M G 2 18 33 T (AJ 5 4 25 08 ) R -3 2 53 4 B a c illu s s p. G 2 M 1 1 -2 AM 6 9 1 5 9 1 9 8 ,9% Ba cillus b a tav ien sis L M G 2 18 33 T (AJ 5 4 25 08 ) R -3 2 51 8 B a c illu s s p. G 2 M 1 1 -2 AM 6 9 1 5 9 6 9 8 ,9% Ba cillus b a tav ien sis L M G 2 18 33 T (AJ 5 4 25 08 ) R -3 2 52 9 B a c illu s s p. G 2 M 1 1 -2 AM 6 9 1 5 9 8 9 8 ,9% Ba cillus b a tav ien sis L M G 2 18 33 T (AJ 5 4 25 08 ) R -3 2 52 4 B a c illu s s p. G 2 M 1 1 -2 AM 6 9 1 5 9 9 9 8 ,7% Bacillus so li L M G 21 83 8 T (AJ 54 2 51 3) R -3 1 84 1 B a c illu s s p. G 2 M 1 1 -2 AM 6 9 1 6 0 0 9 9 % B ac illus d re n te ns is L M G 2 1 83 1 T (AJ 5 42 50 6 ) R -3 3 82 0 B a c illu s s p. G 2 M 1 1 -2 AM 6 9 1 6 0 2 9 9 % B ac illus s o li L M G 2 183 8 T (AJ 54 25 1 3 ) R -3 2 70 2 B a c illu s s p. G 2 M 1 1 -2 AM 6 9 1 6 0 3 9 8 ,9% Ba cillus d ren te ns is L M G 21 8 31 T (AJ 5 4 25 06 ) R -3 2 53 1 B a c illu s s p. G 2 M 1 1 -2 AM 6 9 1 6 0 8 9 8 ,7% Bacillu s so li L M G 21 8 38 T (AJ 5 4 25 13 ) R -3 2 52 5 B a c illu s s p. G 2 M 1 1 -2 AM 6 9 1 6 0 9 9 8 ,8% Bacillu s so li L M G 21 8 38 T (AJ 5 4 25 13 ) R -3 2 69 9 B a c illu s s p. G 2 M 1 1 -2 AM 6 9 1 6 1 0 9 8 ,9% Ba cillus b a tav ien sis L M G 2 18 33 T (AJ 5 4 25 08 ) R -3 2 53 5 B a c illu s s p. G 2 M 1 1 -2 AM 6 9 1 6 1 1 9 9 % B ac illus ba ta v ie ns is L M G 21 83 3 T (AJ 5 42 50 8 ) R -3 2 53 0 B a c illu s s p. G 2 M 1 1 -2 AM 6 9 1 6 1 2 9 8 ,9% Bacillu s so li L M G 21 8 38 T (AJ 5 4 25 13 ) R -3 2 78 4 B a c illu s s p. G 2 M 1 1 -2 AM 6 9 1 6 1 4 9 8 ,9% Ba cillus b a tav ien sis L M G 2 18 33 T (AJ 5 4 25 08 ) R -3 2 84 8 B a c illu s s p. G 2 M 1 1 -2 AM 6 9 1 6 2 0 9 8 ,9% Bacillu s so li L M G 21 8 38 T (AJ 5 4 25 13 ) R -3 1 85 5 B a c illu s s p. G 2 M 1 1 -2 AM 6 9 1 6 2 1 9 8 ,9% Bacillu s so li L M G 21 8 38 T (AJ 5 4 25 13 ) R -3 1 85 2 B a c illu s s p. G 2 M 1 1 -2 AM 6 9 1 6 2 5 9 8 ,9% Ba cillus b a tav ien sis L M G 2 18 33 T (AJ 5 4 25 08 ) R -3 2 51 9 B a c illu s s p. G 2 M 1 1 -2 AM 6 9 1 6 2 7 9 8 ,2% Ba cillus m a n na nily tic us AM -0 0 1 T (AB 0 438 6 4 ) R -3 2 77 8 B a c illu s s p. G 2 M 1 1 -2 AM 6 9 1 6 2 9 9 9 % B ac illus ba ta v ie ns is L M G 21 83 3 T (AJ 5 42 50 8 ) R -3 4 18 1 S ta p h yloc occu s s p. G 2 M 1 1 -2 AM 4 0 3 6 3 8 1 0 0% S ta ph y lo co cc us wa rne ri L M G 1 3 35 4 T (L376 0 3) R -3 3 77 1 S ta p h yloc occu s s p. G 2 M 1 1 -2 AM 4 0 3 6 4 3 1 0 0% S ta ph y lo co cc us wa rne ri L M G 1 3 35 4 T (L376 0 3) R -3 2 76 9 S in o rh izob iu m s p. G 3 M 1 2 -5 AM 6 9 1 6 1 5 1 0 0% S in o rh izob iu m m o re len se LM G 2 1 33 1 T (AY 02 43 35 ) R -3 2 54 9 S in o rh izob iu m s p. G 3 M 1 2 -4 AM 6 9 1 6 2 8 1 0 0% S in o rh izob iu m m o re len se LM G 2 1 33 1 T (AY 02 43 35 ) R -3 1 81 7 P s eu d o m onas s p. G 3 M 1 2 -4 AM 6 9 1 6 1 8 9 9 ,8% Ps eu do m o n as k ilo ne ns is D S M 13 54 7 T (AJ2 92 4 26 ) R -3 1 81 5 P s eu d o m onas s p. G 3 M 1 2 -4 AM 6 9 1 6 3 0 9 9 ,8% Ps eu do m o n as k ilo ne ns is D S M 13 54 7 T (AJ2 92 4 26 ) R -3 1 75 9 S in o rh izob iu m s p. G 3 M 1 2 -3 AM 4 0 3 6 4 7 1 0 0% S in o rh izob iu m m o re len se LM G 2 1 33 1 T (AY 02 43 35 ) R -3 2 73 7 S in o rh izob iu m s p. G 3 M 1 2 -3 AM 4 0 3 6 3 2 1 0 0% S in o rh izob iu m m o re len se LM G 2 1 33 1 T (AY 02 43 35 ) R -3 2 54 6 S in o rh izob iu m s p. G 3 M 1 2 -3 AM 6 9 1 6 2 6 1 0 0% S in o rh izob iu m m o re len se LM G 2 1 33 1 T (AY 02 43 35 ) R -3 1 76 3 S in o rh izob iu m s p. G 3 M 1 2 -3 AM 6 9 1 6 3 1 1 0 0% S in o rh izob iu m m o re len se LM G 2 1 33 1 T (AY 02 43 35 ) R -3 1 76 4 S in o rh izob iu m s p. G 3 M 1 2 -3 AM 4 0 3 6 4 8 1 0 0% S in o rh izob iu m m o re len se LM G 2 1 33 1 T (AY 02 43 35 ) R -3 2 54 4 E n s ife r s p. G 3 M 1 2 -3 AM 4 0 3 5 9 3 1 0 0% E ns ife r m ed ic ae A3 2 1 T (Z 78 20 4 ) R -3 1 76 2 R h izo biu m s p. G 3 M 1 2 -3 AM 4 0 3 5 8 4 9 8 ,3% Rh izo biu m rad io b ac te r L M G 14 0 T (AJ 3 8990 4 ) R -3 1 75 8 P s eu d o m onas s p. G 3 M 1 2 -3 AM 4 0 3 6 0 0 1 0 0% Ps e ud o m on as k ilo ne ns is D S M 1 35 47 T (AJ 29 24 26 ) R -3 1 76 5 P s eu d o m onas s p. G 3 M 1 2 -3 AM 4 0 3 6 0 4 1 0 0% Ps e ud o m on as k ilo ne ns is D S M 1 35 47 T (AJ 29 24 26 ) R -3 2 54 1 P s eu d o m onas s p. G 3 M 1 2 -3 AM 6 9 1 6 1 7 1 0 0% Ps e ud o m on as k ilo ne ns is D S M 1 35 47 T (AJ 29 24 26 )
R -3 1 76 1 P s eu d o m onas s p. G 3 M 1 2 -3 AM 6 9 1 6 2 2 9 9 ,8% Ps eu do m o n as b ras sicace a ru m C IP 10 70 5 9 T
(AF 1 0 0 32 1) R -3 3 77 7 S ta p h yloc occu s s p. G 3 M 1 2 -3 AM 6 9 1 5 6 2 1 0 0% S ta ph y lo co cc us e p id e rm id is L M G 10 47 4 T (D83 3 63 ) R -3 2 54 2 S in o rh izob iu m s p. G 3 M 1 2 -3 AM 4 0 3 6 0 6 1 0 0% S in o rh izob iu m m o re len se LM G 2 1 33 1 T (AY 02 43 35 ) R -3 1 75 7 E n s ife r s p. G 3 M 1 2 -3 AM 6 9 1 5 7 2 1 0 0% E ns ife r m ed ic ae A3 2 1 T (Z 78 20 4 ) R -3 2 72 5 R h izo biu m s p. G 3 M 1 2 -3 AM 4 0 3 6 0 5 1 0 0% R hizo b iu m rad iob ac te r L M G 14 0 T (AJ 3 899 04 ) R -3 2 53 9 R h izo biu m s p. G 3 M 1 2 -3 AM 6 9 1 5 8 4 9 8 ,5% Rh izo biu m rad io b ac te r L M G 14 0 T (AJ 3 8990 4 ) R -3 2 55 2 R h izo biu m s p. G 3 M 1 2 -3 AM 6 9 1 5 6 5 9 7 ,7% Rh izo biu m rad io b ac te r L M G 14 0 T (AJ 3 8990 4 ) R -3 2 55 3 P s eu d o m onas s p. G 3 M 1 2 -3 AM 6 9 1 6 1 9 9 9 ,8% Ps eu do m o n as k ilo ne ns is D S M 13 54 7 T (AJ2 92 4 26 ) R -3 2 72 6 P s eu d o m onas s p. G 3 M 1 2 -3 AM 6 9 1 6 2 3 9 9 ,8% Ps eu do m o n as k ilo ne ns is D S M 13 54 7 T (AJ2 92 4 26 ) R -3 1 81 6 S in o rh izob iu m s p. G 4 M 3 -4 AM 4 0 3 6 5 3 1 0 0% S in o rh izob iu m m o re len se LM G 2 1 33 1 T (AY 02 43 35 ) R -3 1 85 7 R h izo biu m s p. G 4 M 3 -4 AM 6 9 1 6 1 6 9 7 ,2% Rh izo biu m rad io b ac te r L M G 14 0 T (AJ 3 8990 4 ) R -3 2 84 5 B a c illu s s p. G 4 M 3 -4 AM 4 0 3 6 4 2 9 9 % B ac illus ba ta v ie ns is L M G 21 83 3 T (AJ 5 42 50 8 ) R -3 2 66 1 B a c illu s s p. G 4 M 3 -4 AM 6 9 1 5 9 3 9 8 ,8% Ba cillus b a tav ien sis L M G 2 18 33 T (AJ 5 4 25 08 ) R -3 1 84 8 B a c illu s s p. G 4 M 3 -4 AM 6 9 1 5 9 4 9 8 ,9% Ba cillus b a tav ien sis L M G 2 18 33 T (AJ 5 4 25 08 ) R -3 2 66 5 B a c illu s s p. G 4 M 3 -4 AM 6 9 1 6 0 7 9 8 ,9% Ba cillus b a tav ien sis L M G 2 18 33 T (AJ 5 4 25 08 ) R -3 1 54 9 R h izo biu m s p. G 4 M 3 -3 AM 4 0 3 6 2 1 9 8 ,5% Rh izo biu m rad io b ac te r L M G 14 0 T (AJ 3 8990 4 ) R -3 1 83 7 R h izo biu m s p. G 4 M 3 -3 AM 4 0 3 6 1 4 9 8 ,5% Rh izo biu m rad io b ac te r L M G 14 0 T (AJ 3 8990 4 ) R -3 2 66 3 B a c illu s s p. G 4 M 3 -3 AM 4 0 3 6 4 6 1 0 0% B ac illus d ren te n sis L M G 2 18 31 T (AJ 5 4 250 6) R -3 1 84 6 B a c illu s s p. G 4 M 3 -3 AM 4 0 3 6 4 5 9 9 % B ac illus d re n te ns is L M G 2 1 83 1 T (AJ 5 42 50 6 ) R -3 1 55 0 B a c illu s s p. G 4 M 3 -3 AM 4 0 3 6 0 2 1 0 0% B ac illus d ren te n sis L M G 2 18 31 T (AJ 5 4 250 6) R -3 1 83 5 B a c illu s s p. G 4 M 3 -3 AM 4 0 3 6 0 1 1 0 0% B ac illus d ren te n sis L M G 2 18 31 T (AJ 5 4 250 6) R -3 1 54 7 B a c illu s s p. G 4 M 3 -3 AM 4 0 3 5 8 3 1 0 0% B ac illus d ren te n sis L M G 2 18 31 T (AJ 5 4 250 6) R -3 1 83 4 B a c illu s s p. G 4 M 3 -3 AM 4 0 3 6 2 4 9 8 ,8% Ba cillus b a tav ien sis L M G 2 18 33 T (AJ 5 4 25 08 ) R -3 1 83 6 B a c illu s s p. G 4 M 3 -3 AM 6 9 1 5 6 6 1 0 0% B ac illus d ren te n sis L M G 2 18 31 T (AJ 5 4 250 6) R -3 1 83 8 B a c illu s s p. G 4 M 3 -3 AM 6 9 1 5 6 8 1 0 0% B ac illus d ren te n sis L M G 2 18 31 T (AJ 5 4 250 6) R -3 2 84 4 B a c illu s s p. G 4 M 3 -3 AM 6 9 1 5 7 3 9 9 ,3% Bacillu s so li L M G 21 8 38 T (AJ 5 42 51 3) R -3 2 65 6 B a c illu s s p. G 4 M 3 -3 AM 6 9 1 5 7 5 1 0 0% B ac illus d ren te n sis L M G 2 18 31 T (AJ 5 4 250 6) R -3 1 55 3 B a c illu s s p. G 4 M 3 -3 AM 6 9 1 5 7 6 9 9 % B ac illus s oli L M G 218 3 8 T (AJ 54 2 51 3) R -3 1 55 2 B a c illu s s p. G 4 M 3 -3 AM 6 9 1 5 7 7 9 8 ,9% Ba cillus b a tav ien sis L M G 2 18 33 T (AJ 5 4 25 08 ) R -3 2 66 2 B a c illu s s p. G 4 M 3 -3 AM 6 9 1 5 8 9 9 8 ,7% Ba cillus b a tav ien sis L M G 2 18 33 T (AJ 5 4 25 08 ) R -3 1 83 2 B a c illu s s p. G 4 M 3 -3 AM 6 9 1 5 9 5 9 8 ,8% Ba cillus d ren te ns is L M G 21 8 31 T (AJ 5 4 25 06 ) R -3 1 85 0 B a c illu s s p. G 4 M 3 -3 AM 6 9 1 5 9 7 1 0 0% B ac illus d ren te n sis L M G 2 18 31 T (AJ 5 4 250 6) R -3 1 84 5 B a c illu s s p. G 4 M 3 -3 AM 6 9 1 6 0 1 9 8 ,9% Bacillu s so li L M G 21 8 38 T (AJ 5 4 25 13 ) R -3 1 84 9 B a c illu s s p. G 4 M 3 -3 AM 6 9 1 6 0 6 1 0 0% B ac illus d ren te n sis L M G 2 18 31 T (AJ 5 4 250 6) R -3 1 54 1 B a c illu s s p. TS A -5 AM 4 0 3 6 3 0 1 0 0% B ac illus d ren te n sis L M G 2 18 31 T (AJ 5 4 250 6) R -3 2 57 5 B a c illu s s p. TS A -5 AM 6 9 1 6 1 3 1 0 0% B ac illus d ren te n sis L M G 2 18 31 T (AJ 5 4 250 6) R -3 1 54 3 C u p ria vid us s p. TS A -5 AM 4 0 3 5 9 9 9 9 ,3% Cu p riavidu s n ec a to r L M G 8 4 53 T (AF 1 9 17 3 7 ) R -3 1 54 2 C u p ria vid us s p. TS A -5 AM 4 0 3 5 9 7 9 9 ,3% Cu p riavidu s n ec a to r L M G 8 4 53 T (AF 1 9 17 3 7 ) R -3 1 54 4 C u p ria vid us s p. TS A -5 AM 6 9 1 5 6 7 9 9 ,5% Cu p riavidu s n ec a to r L M G 8 4 53 T (AF 1 9 17 3 7 ) R -3 1 83 0 B a c illu s s p. TS A -4 AM 4 0 3 6 0 9 9 9 ,5% Ba cillu s ps e ud o m yc oide s L M G 1 89 93 T (AF0 1 3 12 1 ) R -3 2 57 4 B a c illu s s p. TS A -4 AM 6 9 1 6 2 4 9 9 ,2% Ba cillu s ps e ud o m yc oide s L M G 1 89 93 T (AF0 1 3 12 1 )
CHAPTER 2
59
Assessment of applied growth media
Important for universal applicability of elective growth media for denitrifier isolation is (i)
the growth of a high number of denitrifiers, (ii) a high diversity of denitrifiers, both retrieved
from (iii) the highest dilution containing denitrifiers. These three factors can be used for
evaluation of the four growth media, applied simultaneously to the same soil sample.
These results are compared to a cultivation study of activated sludge using the same growth
media [9], and a cultivation study on three agricultural soils using nitrate agar as isolation
medium [4].
(i) An overview of the total number of isolates and their nitrogen reducing capacities per
growth medium is given in Table 2.4. G4M3 and G2M11 yielded a high ratio denitrifiers-
to-total-isolates, appr. 55% and 75% respectively. These ratios were significantly higher
than for TSA and G3M12, due to the high number of nitrate reducers on the latter growth
media, casting doubt on the suitability of G3M12 for denitrifier isolation. Compared to
results from literature achieved with nitrate agar, yielding between 2.2 and 22.8% [4],
media G4M3 and G2M11 numerically scored much better for elective isolation of
heterotrophic denitrifiers from soil. Both media also gave significant better results with
soil than with activated sludge, 20-30% and 43% respectively, while G3M12 and TSA
yielded a similar ratio for both inocula.
Table 2.4. An overview of the number of isolates with the nitrogen reduction capacities specified and grouped perisolation medium. Percentages per isolation medium are given. Exact numbers are mentioned between brackets.
(ii) The retrieved denitrifying diversity per growth medium was scored. The observed
diversity onto genus level - genus identification was based on both FAME analysis and
16S rRNA gene sequence analysis - was represented for each growth medium by the
Simpson’s reciprocal diversity index 1/D,
1/D = N × (N-1)/E[ni × (n
i-1)] (eq. 3)
with N = number of denitrifying isolates per medium, ni = number of denitrifying isolates
per medium belonging to genus i. Growth medium G3M12 retrieved the highest denitrifier
diversity (Figure 2.3). The results for G2M11 and G4M3 were comparable and significantly
Metabolism of nitrogen compounds Growth media for isolation and tests
ISOLATION, CHARACTERIZATION AND IDENTIFICATION OF DENITRIFYING BACTERIA FROM THE ENVIRONMENT
60
lower than TSA, due to the numerical dominance of bacilli retrieved on both media. G2M11,
G3M12 and G4M3 clearly retrieved less diversity from soil than from activated sludge
(Figure 2.3). The difference in retrieved cultured diversity between both habitats may be
correlated with the continuous enrichment of denitrifiers in activated sludge, contrasting
with the controlling factor of denitrification in soil, which is not the process itself but the
competition for carbon with aerobic heterotrophs [23]. Also, the different culturability of
the microbiota of both habitats can be involved: activated sludge contains 1-15% culturable
bacteria, while this is only 0.3% in soil [2]. However, the growth media could also have a
biased towards activated sludge, the sole inoculum during medium development [9]. In
contrast, little difference in retrieved denitrifier diversity was observed between activated
sludge and soil with supplemented TSA, suggesting an intrinsic average diversity index of
2. The cultured denitrifying diversity in the comparable study was nicely spread over five
different taxa, with 15-25% of Pseudomonas, Ralstonia, Frateuria, Bacillus, and
Streptomyces, while we predominantly found Bacillus. Thus, nitrate agar could be more
suitable for denitrifier isolation. Yet, it is difficult to compare results from different soil
samples. First, these cited results are collected from three different agricultural soils, and
the results per soil were not given. Second, because of soil’s three-dimensional matrix,
nutrient conditions, and thus denitrifying activity, are variable in space and time.
4,0
7,8
2,1 2,3
15,0
9,0
1,1
1,4
0
2
4
6
8
10
12
14
16
act ivatedsludge
soil act ivatedsludge
soil act ivatedsludge
soil act ivatedsludge
soil
G2M 11 G3M 12 G4M 3 TSA
1/D
Figure 2.3. Retrieved denitrifying diversity from soil, expressed as Simpson’s Reciprocal Diversity Index 1/D, pergrowth medium. The results of both activated sludge (taken from [9]) and soil (this study) are given.
CHAPTER 2
61
(iii) When taking the isolation dilution as a rough measurement of their in situ abundance
(Table 2.3), the relevance of the retrieved denitrifier diversity can be evaluated. TSA and
G3M12 scored best, with retrieval of at least 106 CFU g-1 denitrifiers, while G4M3 cultured
an order of magnitude less, and G2M11 scored worst with 103 CFU g-1 denitrifiers. Based
on these abundance estimates, Bacillus, Sinorhizobium and Cupriavidus harbor the culturable
denitrifiers dominant in this soil sample, followed by Pseudomonas. Unfortunately,
comparison with other cultivation studies or results from activated sludge is not possible
due to lack of data. However, our results comply with quantitative PCR data of soil using
the nirK gene, which is a single copy gene coding for copper-containing nitrite reductase
[25], reporting nirK densities of 104 to 109 nir gene copies per gram soil, depending on the
technique used and the type of soil [7, 8, 14, 18].
Conclusions
The assessment of a growth medium to cultivate a specific group of bacteria depends on
the criteria used. Here we defined three criteria: the number of denitrifiers retrieved, their
diversity and estimated abundance. Although G2M11, and to a lesser extent, G4M3, retrieved
a high number of denitrifying isolates, G3M12 yieled a much higher denitrifying diversity
from both soil and activated sludge and, based on the used criteria, this is the optimal
medium to study denitrifiers. Also, the results of supplemented TSA seem to be independent
of the studied environment. However, application of a multiple-media set generates more
information than the use of individual growth media. The estimated abundance of
Sinorhizobium, and Cupriavidus and Bacillus would not have been revealed without use
of G3M12 or supplemented TSA respectively. The use of a set of different growth media is
thus recommended for estimating general [3] or specific bacterial diversity. Now, an
investigation of the role and abundance of members of these three genera in situ, for example
with fluorescent in situ hybridization in combination with microautoradiography [12], is
warranted.
ISOLATION, CHARACTERIZATION AND IDENTIFICATION OF DENITRIFYING BACTERIA FROM THE ENVIRONMENT
62
REFERENCES1. Altschul SF, Gish W, Miller W, Myers EW, Lipman DJ (1990) Basic local alignment search tool. J
Mol Biol 215:403-410
2. Amann RI, Ludwig W, Schleifer K-H (1995) Phylogenetic identification and in situ detection of individual
microbial cells without cultivation. Microbiol Rev 59:143-169
3. Balestra GM, Misaghi IJ (1997) Increasing the efficiency of the plate counting method for estimating
bacterial diversity. J Microbiol Meth 30:111-117
4. Chèneby D, Philippot L, Hartmann A, Hénault C, Germon J-C (2000) 16S rDNA analysis for
characterisation of denitrifying bacteria isolated from three agricultural soils. FEMS Microbiol Ecol
34:121-128
5. Felske A, Wolterink A, Van Lis R, Akkermans ADL (1998) Phylogeny of the main bacterial 16S rRNA
sequences in Drentse A grassland soils (The Netherlands). Appl Environ Microbiol 64: 871-879
6. Gamble TN, Betlach MR, Tiedje TM (1977) Numerically dominant denitrifying bacteria from world
soils. Appl Environ Microbiol 33:926-939
7. Henry S, Baudoin E, Lopez-Gutiérrez JC, Martin-Laurent F, Brauman A, Philippot L (2004) Quantification
of denitrifying bacteria in soils by nirK gene targeted real-time PCR. J Microbiol Meth 59:327-335
8. Henry S, Bru D, Stres B, Hallet S, Philippot L (2006) Quantitative detection of the nosZ genes, encoding
nitrous oxide reductase, and comparison of the abundance of 16S rRNA, narG, nirK and nosZ genes in
soils. Appl Environ Microbiol 72:5181-5189
9. Heylen K, Vanparys B, Wittebolle L, Verstraete W, Boon N, De Vos P (2006) Cultivation of denitrifying
bacteria: optimization of isolation conditions and diversity study. Appl Environ Microbiol 72:2637-
2643
10. Heyrman J, Swings J (2001) 16S rDNA sequence analysis of bacterial isolates from biodeteriorated
mural paintings in the Servilia tomb (necropolis of Carmona, Seville, Spain). Syst Appl Microbiol
24:417-422
11. Heyrman J, Vanparys B, Logan NA, Balcaen A, Rodríguez-Díaz M, Felske A, De Vos P (2004) Bacillus
26467), one Acidovorax sp. (R-24336), three Paracoccus sp. (R-24292, R-24342, R-26466), two Comamonas sp.
(R-24447, R-24451), one Thauera sp. (R-24450), one Pseudomonas sp. (R-24261) and one Pseudoxanthomonas
sp. (R-24339). The 21 strains were assigned at the genus level as described previously [11] by performing fatty acid
methyl ester analysis and 16S rRNA gene sequence analysis.
DNA extraction and nir PCRDNA was extracted from each denitrifying isolate using the guanidium-thiocyanate-EDTA-sarkosyl method
described by Pitcher et al. [15] for fast-growing strains and using alkaline lysis for slow-growing isolates. For
alkaline lysis, one colony was suspended in an eppendorf tube with 20 µl of lysis buffer (2.5 ml 10% SDS; 5 ml 1M
NaOH; 92.5 ml MilliQ water). After 15 min at 95°C, 180 ml MilliQ water was added, the tube was centrifuged for
5 min at 13,000 g and the supernatant was transferred to a new tube. DNA extracts were stored at –20°C until use.
For amplification and sequence analysis of nir genes, five primer sets previously described were used: for the nirK
gene nirK1F-nirK5R [2], F1aCu-R3Cu [10] and Cunir3-Cunir4 [4]; for the nirS gene nirS1F-nirS6R [2] and nir3-
nir4 [23]. Temperature-time profiles and protocols were adopted from the original description.
Nir gene sequence analysisThe PCR-amplified nir gene products were purified using the Nucleofast® 96 PCR system (Millipore). For each
sequence reaction a mixture was made using 3 µl purified and concentrated PCR product, 1 µl of BigDye™
Termination RR mix version 3.1 (Applied Biosystems), 1.5 µl of BigDye™ buffer (5x), 1.5 µl sterile milliQ water
and 3 µl (20 ng/µl) of one of the sequencing primers. The amplification primers were used as sequencing primers.
The temperature-time profile was as follows: 30 cycles of denaturation for 15 s at 96°C, primer annealing for 1 s at
35°C and extension for 4 min at 60°C. The sequencing products were cleaned up as described previously [13].
Sequence analysis of the nir gene was performed using an Applied Biosystems 3100 DNA Sequencer according
to protocols provided by the manufacturer.
Phylogenetic analysisForward and reverse strands of nirS and nirK were assembled with the BioNumerics 4.0 software and were submitted
to a BLAST search [1]. Amino acid sequences were deduced with Emboss Transeq (EMBL-EBI) and aligned with
sequences retrieved from the EMBL database using ClustalX [18]. Phylogenetic analyses were performed with
Treecon [26]. Trees were constructed with the neighbor-joining algorithm using the BLOSUM 62 substitution
matrix. Statistical evaluation of the tree topologies was performed by bootstrap analysis with 1000 resamplings.
Nucleotide sequence accession numbersThe 16S rRNA gene sequences have been submitted to the DDBJ/EMBL/GenBank databases under accession
number AM231040 through AM231060. The nirK gene sequences have been submitted to the DDBJ/EMBL/
GenBank databases under accession number AM230812 through AM230887. The nirS gene sequences have been
submitted to the DDBJ/EMBL/GenBank databases under accession number AM230888 through AM230922.
CHAPTER 3
75
RESULTS
Amplification of nirK and nirS from cultured denitrifiers
The nir genes were studied in a set of 227 denitrifying strains (Table 3.1). Eight
representatives of denitrifiers from Alpha-, Beta-, Gammaproteobacteria and Firmicutes
available in culture collections were included as reference strains. Other denitrifiers were
isolated in a previous cultivation-dependent study of denitrifiers from activated sludge of
a municipal wastewater treatment plant [11]: the bulk of these denitrifying isolates belonged
to the Alpha- and Betaproteobacteria , but also included representatives of
Gammaproteobacteria, Epsilonproteobacteria, Firmicutes and Bacteroidetes. Five previously
described primer sets were applied to amplify the two nir genes. The three primer sets used
to amplify nirK - nirK1F-5R, F1aCu-R3Cu, and Cunir3-4 - and the two primer sets used
to amplify nirS - nirS1F-6R and nir3-4.
Table 3.1 provides an overview of all tested denitrifiers, arranged according to their taxonomic
position, and the results for each primer set. Almost half of all included denitrifiers (109 out
of 227) did not render an amplicon with any of the five primer sets. NirK was detected in
76 denitrifiers. Of all nirK amplicons, 75% were found within the Alphaproteobacteria,
14.5% within the Betaproteobacteria, 5.3% within the Gammaproteobacteria, 3.9% within
the Firmicutes, and 1.3% within the Bacteroidetes. The primer set F1aCu-R3Cu clearly
scored best and provided an amplicon for all nirK sequences detected, even for denitrifiers
belonging to less-known denitrifying genera as Enterococcus, Staphylococcus and
Chryseobacterium. Cunir3-Cunir4 did not render any amplicons. NirS was detected in 42
denitrifiers. Of all nirS amplicons, 26.2% were found within the Alphaproteobacteria,
64.3% in the Betaproteobacteria, and 9.5% in the Gammaproteobacteria. For nirS
amplification, the primer set nirS1F-6R scored best. No pure culture denitrifier was found
carrying both nir genes.
Table 3.1. (Next two pages) An overview of the denitrifying strains arranged according to their taxonomicposition. Reliable species identification is given for reference strains from the LMG culture collection.a +, nir amplicon verified through sequence analysis; -, no amplification or unspecific amplification; ?, ampliconscould not be sequenced succesfully; b [2] c [10]; d [23]; e primer set Cunir3-Cunir4 [4] did not render any ampliconand was therefore not included in this overview.
FUNCTIONAL PHYLOGENETIC ANALYSIS OF PURE CULTURE DENITRIFIERS
76
CHAPTER 3
Id
entif
icat
ion
P
rim
ers
nir
K a,
e P
rim
ers
nir
S a
Str
ain
nr.
n
irK
1F-5
R b
F1a
Cu
-R3C
u c n
irS
1F-6
R b
nir
3-4
d
Alp
hap
rote
obac
teri
a
B
ruce
lla s
p.
+ +
-
-
R-2
6895
O
crho
bact
rum
ant
hrop
i +
+ -
-
LMG
213
6
O
chro
bact
rum
sp.
+
+ -
-
R-2
4286
, R-2
4290
, R-2
4291
, R-2
4343
, R-2
4448
, R-2
4618
, R-2
4638
, R-2
4639
,
R-2
4653
, R-2
5018
, R-2
5054
, R-2
5203
, R-2
6465
, R-2
6821
, R-2
6825
, R-2
6890
,
R-2
6892
, R-2
6894
, R-2
6898
, R-2
6900
, R-2
7046
, R-2
7045
, R-2
8410
-
+ -
-
R
-242
89, R
-250
55, R
-268
26, R
-268
91, R
-244
67, R
-268
89
-
-
-
-
R-
2446
8
M
ethy
loba
cter
ium
sp.
+
+ -
-
R
-252
07
R
hizo
bium
sp.
+
+
-
-
R-2
4654
, R-2
6467
, R-2
6835
R
hizo
bium
sp.
-
+
-
-
R
-242
60, R
-243
33, R
-246
58, R
-246
63, R
-268
20
S
inor
hizo
bium
sp
+ +
-
-
R
-246
05, R
-250
67, R
-250
78
P
arac
occu
s de
nitri
fican
s -
-
+
-
LMG
404
9
P
arac
occu
s sp
. +
+ -
-
R
-246
49, R
-246
50
-
+ -
-
R
-243
42, R
-246
23, R
-246
52, R
-250
58, R
-268
22, R
-268
23, R
-268
24, R
-268
41,
R-2
6899
, R-2
6901
, R-2
7049
, R-2
8242
-
-
+ +
R-2
4615
, R-2
4616
, R-2
4617
, R-2
6466
, R-2
6897
, R-2
7041
, R-2
8414
-
-
+ -
R
-246
65, R
-282
39, R
-282
41
-
-
-
? R
-270
43
-
-
-
-
R-2
4292
, R-2
4621
, R-2
5049
, R-2
5059
, R-2
6839
, R-2
6844
, R-2
6888
, R-2
6893
,
R-2
6896
, R-2
6902
, R-2
7047
, R-2
8237
, R-2
6819
, R-2
8238
, R-2
8243
, R-2
8244
,
R-2
8245
, R-2
8294
, R-2
8409
P
anno
niba
cter
sp.
-
-
-
-
R
-270
42
Bet
apro
teo
bact
eria
A
lcal
igen
es fa
ecal
is
-
+ -
-
LMG
122
9
A
chro
mob
acte
r de
nitri
fican
s +
+ -
-
LM
G 1
231
C
upria
vidu
s ne
cato
r -
-
+
-
LMG
120
1
A
cido
vora
x sp
. +
+ -
-
R
-250
76
-
+ -
-
R
-243
36, R
-246
13, R
-246
14, R
-250
52, R
-250
75
-
? -
-
R
-246
67
-
-
+ +
R-2
5212
-
-
-
-
R-2
4607
, R-2
4608
, R-2
5074
, R-2
6831
, R-2
6833
, R-2
6834
, R-2
6837
, R-2
6842
,
R-2
8240
, R-2
6843
, R-2
7044
, R-2
8416
A
licyc
liphi
lus
sp.
-
-
+ -
R
-246
04, R
-246
11
-
? -
-
R
-246
06
- -
-
-
R-2
6814
77
FUNCTIONAL PHYLOGENETIC ANALYSIS OF PURE CULTURE DENITRIFIERS
C
omam
onas
sp.
-
-
+
+ R
-244
47, R
-250
14, R
-250
48, R
-250
66, R
-282
10, R
-282
33, R
-282
93, R
-282
34,
R-2
8235
, R-2
8452
-
-
+ -
R
-282
31
-
-
-
+ R
-244
51, R
-282
24, R
-282
27
-
-
-
-
R-2
4660
, R-2
5057
, R-2
5060
, R-2
6810
, R-2
6812
, R-2
6817
, R-2
6829
, R-2
6886
,
R-2
6887
, R-2
6905
, R-2
7040
, R-2
8209
, R-2
8211
, R-2
8218
, R-2
8220
, R-2
8222
,
R-2
8223
, R-2
8225
, R-2
8226
, R-2
8228
, R-2
8229
, R-2
8230
, R-2
8232
, R-2
8236
,
R-2
8413
, R-2
8449
D
iaph
orob
acte
r sp
. -
+
-
-
R-2
4610
, R-2
4612
, R-2
6815
-
-
-
-
R-2
4661
, R-2
5011
, R-2
5012
, R-2
6840
, R-2
8417
S
impl
icis
pira
sp.
-
-
-
-
R
-284
08
A
zosp
ira s
p.
-
-
+ -
R
-284
47
-
-
-
-
R-2
5019
A
zovi
brio
sp.
-
-
-
-
R
-250
62
D
echl
orom
onas
sp.
-
-
+
+ R
-284
51
-
-
+ -
R
-284
00, R
-284
07
-
-
-
-
R-2
8215
, R-2
8291
, R-2
8317
, R-2
8401
, R-2
8412
T
haue
ra s
p.
-
-
+
-
R-2
4450
, R-2
5069
, R-2
5071
, R-2
6885
, R-2
6906
, R-2
8213
-
-
-
-
R-2
8205
, R-2
8206
, R-2
8207
, R-2
8289
, R-2
8402
, R-2
8403
, R-2
8404
, R-2
8405
,
R-2
8406
, R-2
8312
, R-2
8216
, R-2
8217
, R-2
8292
, R-2
8448
Z
oogl
oea
sp.
-
-
-
-
R-2
8315
Gam
map
rote
ob
acte
ria
P
seud
omon
as a
erug
inos
a -
-
+
-
LM
G 1
242
P
seud
omon
as s
tutz
eri
-
-
+ +
LMG
224
3
P
seud
omon
as s
p.
+ +
? -
R
-268
28
-
+ -
-
R
-246
09, R
-242
61, R
-252
08
-
-
+ +
R-2
5061
-
-
+ -
R
-268
30
-
-
-
-
R-2
4636
, R-2
5016
, R-2
5209
, R-2
5343
, R-2
8208
, R-2
8219
P
seud
oxan
thom
onas
sp.
-
-
-
-
R
-243
39
Ep
silo
nb
acte
ria
A
rcob
acte
r sp
. -
-
-
-
R
-282
14, R
-283
13, R
-283
14, R
-283
16
Fir
mic
ute
s
V
irgib
acill
us h
alod
enitr
ifica
ns
-
-
-
-
LMG
981
8
B
acill
us s
p.
-
-
-
-
R-2
4620
, R-2
4666
, R-2
8399
T
richo
cocc
us s
p.
-
-
-
-
R-2
8212
E
nter
ococ
cus
sp.
+ +
-
-
R
-246
26
-
+ -
-
R
-252
05
P
aeni
baci
llus
sp.
-
-
-
-
R-2
7048
S
taph
yloc
occu
s sp
. -
+
-
-
R-2
5050
Bac
tero
idet
es
C
hrys
eoba
cter
ium
sp.
-
+
-
-
R-2
5053
78
Sequence analysis of nir genes
Sequence similarity analysis of primary DNA sequences confirmed that PCR products of
the expected sizes from denitrifying strains were indeed nirK or nirS fragments. The applied
amplification and sequence analysis protocols were verified by including two reference
strains, Achromobacter denitrificans and Pseudomonas stutzeri, with a nirK sequence
(accession number AJ224905) and a nirS sequence (accession number X56813) already
published. For both strains, the generated nir sequence gave a perfect match with the
available sequence.
The genetic heterogeneity of nirS and nirK was assessed after pairwise comparison of all
retrieved sequences in this study with each other and with publicly available nir sequences
from cultivated and identified denitrifiers. The nirS sequence similarity ranged from 60.01%
to 100.00%, and the nirK sequence similarity ranged from 64.45% to 100.00%. For nirK, a
100% sequence similarity was found between strains from phylogenetically distinct taxa.
For example, the nirK sequences of Paracoccus sp. R-26822, Ochrobactrum sp. R-25203
belonging to the Alphaproteobacteria and Acidovorax sp. R-25075 belonging to the
Betaproteobacteria were identical.
Pairwise comparison showed that nirK sequences from pure culture denitrifiers isolated
from activated sludge of the same wastewater treatment plant were significantly more
similar (t test, P < 0.01) to each other, with an average identity of 88.2% (68.85 to 100%)
than to nirK sequences from denitrifying strains from culture collections or other studies
(nirK: <x> = 78.1%, range, 64.45 to 91.51%). The same observation could not be deduced
from a comparative analysis of nirS sequences.
Phylogenetic analysis
The phylogenetic analysis was performed on amino acid sequences derived from nirK and
nirS gene sequences. Amino acid sequences of cultured and identified strains from EMBL
were also included; environmental clones were omitted.
For the analysis of nirK, 114 sequences were included from Alphaproteobacteria (78),
Betaproteobacteria (23), Gammaproteobacteria (9), Firmicutes (3), and Bacteroidetes (1).
The tree topology of the nirK gene product showed three separate clusters (Figure 3.1).
The large cluster I contained mostly Alphaproteobacteria, but was interspersed with
sequences from Beta-, Gammaproteobacteria and Firmicutes. Three distinct subclusters
could be observed: subcluster Ia with highly similar sequences belonging to different genera
CHAPTER 3
79
(even from different classes), subcluster Ib with representatives of the Rhizobiaceae grouped
together with high bootstrap values, and subcluster Ic with highly related sequences from
representatives of Paracoccus. Cluster II grouped strains belonging to the Alpha-, Beta-,
Gammaproteobacteria and Firmicutes. All Nitrosomonas strains were situated in cluster
III showing distinct sequences, except for Nitrosomonas sp. TA-921i-NH4, which falls in
cluster II. In between cluster II and III, two Rhodobacter strains grouped separate from
other Alphaproteobacteria with a high bootstrap value.
For the analysis of nirS, 89 sequences were included from Alphaproteobacteria (15),
and Bacteroidetes (2). The phylogenetic tree of the nirS gene product was more congruent
with their organism phylogeny. The tree topology showed four distinct clusters supported
with high bootstrap values (Figure 3.2). Cluster I contained the Betaproteobacteria, divided
over two subclusters. Strains in subcluster Ia mostly belonged to the Rhodocyclaceae,
while Comamonadaceae representatives grouped into subcluster Ib. Other strains of the
Rhodocyclaceae formed a separated cluster III, distinct from the other Betaproteobacteria
with high bootstrap values. Cluster II contained all Alphaproteobacteria, all but one of
which belonged to the genus Paracoccus. The Gammaproteobacteria were spread over
cluster IV and V. Cluster IV grouped all Pseudomonas representatives, while cluster V
contained all Marinobacter representatives. Spread over the tree, a number of sequences
did not follow this clear division in bacterial classes. Cluster I of Betaproteobacteria contained
interspersing sequences from representatives of Pseudomonas, Marinobacter,
Flavobacterium and Kocuria, while cluster IV and V of Gammaproteobacteria also contained
sequences from representatives of Burkholderia, Azoarcus, Bacillus and the family of the
Flavobacteriaceae.
FUNCTIONAL PHYLOGENETIC ANALYSIS OF PURE CULTURE DENITRIFIERS
80
Figure 3.1. Phylogenetic analysis of nirK. Unrooted neighbor joining tree was based on partial nirK gene products(114 amino acids; accession numbers between parentheses). Bootstrap values were generated from 1000 replicatesof neighbor joining. Bootstrap values higher than 50% are given. Sequences generated in this study are in bold.
III
Ib
Ia
II
Ic
CHAPTER 3
81
Figure 3.2. Phylogenetic analysis of nirS. Unrooted neighbor joining tree based on partial nirS gene products(253 amino acids; accession numbers between parentheses). Bootstrap values were generated from 1000 replicatesof neighbor joining. Bootstrap values higher than 50% are given. Sequences generated in this study are in bold.
II
III
IV
0.1 subst./site
I
Ib
V
FUNCTIONAL PHYLOGENETIC ANALYSIS OF PURE CULTURE DENITRIFIERS
82
DISCUSSION
Our study reports on the incongruence of nirS and nirK phylogenies with 16S rRNA gene
phylogeny. We recommend that functional nir gene sequence diversity should not be used
to estimate the structural denitrifier diversity but only functional gene diversity in the
environment.
NirK sequences did not form delineated taxon-based groups. The tree topology of the nirK
gene product showed that members of different bacterial classes grouped together, with the
exception of nirK sequences of the Rhizobiaceae and Nitrosomonas that did cluster according
to the taxonomic position of the strains, supported by high bootstrap values. The nirK
sequences of nitrifying genus Nitrosomonas, all except for one, grouped separate from
denitrifiers as was reported earlier [4]. The tree topology of the nirS gene product agreed
more with the 16S rRNA gene phylogeny. The distinct nirS clusters of Alpha-, Beta- and
Gammaproteobacteria were supported by high bootstrap values. Even within the large
clusters based on bacterial classes, further division onto family (Rhodocyclaceae and
Comamonadaceae in the Betaproteobacteria) or genus level (Pseudomonas and Marinobacter
in the Gammaproteobacteria) was quite reliable, not taken into account the few sequences
scattered over the tree and grouping within other bacterial classes. Thus, nir genes, especially
nirS, seemed to contain taxonomic information to some extent, but this appeared to be
taxon dependent and can certainly not be generalised in environmental studies.
NirK sequence analysis revealed two important features absent in nirS: (i) nirK sequences
from denitrifiers isolated from activated sludge of a municipal wastewater treatment plant
were significantly different from other publicly available nirK sequences, despite their
taxonomic position, and (ii) identical nirK sequences were retrieved from taxonomically
distinct strains. Both features suggest that nirK sequence similarity is to some degree
dependent on the habitat of the denitrifiers and not only on their phylogenetic affiliation.
Distinct separate clustering of nirK clones from the same environment was reported
previously for marine sediment and soil clones[16] and for clones retrieved from the
rhizosphere of legumes [22]. However, no taxonomic information was available to rule out
the grouping of strains belonging to the same taxon due to the culture-independent nature of
these studies.
Earlier nir gene studies on pure culture denitrifiers reported nir gene phylogeny incongruent
with the 16S rRNA gene phylogeny [4, 14, 23]. Recently, a denitrifying strain was found to
contain two nirS genes, from which one was probably acquired through HGT based on the
phylogenetic analysis [6]. The hypothesis that nir genes possibly were horizontally
CHAPTER 3
83
transferred between different taxa could explain several of our observations, besides aberrant
nir phylogenies. First, not all genes itinerate equally [9]; this could account for the different
degree in divergence of nirS and nirK phylogenies. Although the available data are limited
and possibly ambiguous, they suggest that, if HGT has occurred, nirK would have a higher
propensity for transfer than nirS. Second, not all groups of organisms experience HGT to
the same extent [9]. The Rhizobiaceae family and the genus Nitrosomonas would then be
less susceptible to HGT. Third, adaptation to the environment is one of the factors that
determines the occurrence and frequency of DNA exchange between taxa [8]. The significant
influence of the habitat on the nirK phylogeny could be correlated with the possibility to
exchange mobile DNA between strains within the same habitat. Denitrification genes are
probably advantageous for the studied strains in their original habitat, an activated sludge
system with periodic anoxic and aerobic conditions. However, the nir phylogenies were
based on partial gene sequences and phylogenetic examination of genes can give an indication
but is only ambiguous proof for HGT. Gene duplication followed by gene loss can also give
rise to different gene phylogenies, which are indistinguishable from conflicting phylogenetic
signals of HGT [9]. The indications of HGT presented in this study needs to be substantiated
with data on gene organisation and/or phylogenetic analysis of other denitrification genes.
It is generally accepted that nitrite reductase and nitric oxide reductase are genetically
linked [27] in certain denitrifiers, so supplemental sequence analysis on nor genes could
confirm possible HGT. Moreover, the study of denitrification genes should also focus on
the generation of complete gene sequences, to obtain a more complete picture of the
phylogeny of denitrification genes and possible HGT events.
Based on the available data of the nir genes of cultivated denitrifiers, Philippot [14] stated
that, when possible HGT is not taken into account, nirK mainly belonged to the Alpha- and
Betaproteobacteria and nirS to the Gammaproteobacteria. Our study confirmed the
prevalence of nirK in the Alphaproteobacteria but also found that nirS dominated in the
Betaproteobacteria. Both bacterial classes were well represented in the investigated strain
set and these results could be representative of the incidence of nir genes. Only a few
denitrifiers from the Gamma- and Epsilonproteobacteria, Firmicutes and Bacteroidetes
were included in our study, and therefore the detection of nirK in the Firmicutes and
Bacteroidetes and the equal occurrence of nirS and nirK in the Gammaproteobacteria can
give a wrong impression of nir gene incidence in these taxa. However, the basic idea that
nirS is dominantly present while nirK is present in more taxonomically unrelated strains
seems only partially true because nirK was prevalent in our pure culture study. More
cultivation-dependent studies are needed to broaden the information concerning nir gene
incidence and may confirm a correlation between the type of nir gene (and nar or nor gene)
and the bacterial taxon. This will improve our understanding of why different types of
FUNCTIONAL PHYLOGENETIC ANALYSIS OF PURE CULTURE DENITRIFIERS
84
nitrate, nitrite and nitric oxide reductases have evolved and what advantage one may provide
over another.
Three nirK primer sets were used in the here presented work, nirK1F-nirK5R [2], F1aCu-
R3Cu [10], and Cunir3-Cunir4 [4]. All three nirK primer pairs are situated in the more
conserved 3’ region of the target gene. The primer nirK1F targets a region that includes a
three base insertion in some bacteria (located between nucleotides 528 and 529 in the nirK
gene of A. faecalis). This difference corresponds to a single amino acid codon present in
nitrifiers and some rhizobia, and maybe other denitrifiers. More upstream, the Cunir3
primer targets a highly conserved region encoding copper-binding sites. The other forward
primer, F1aCu, targets a more variable region downstream of nirK1F. All three reverse
primers focus on the same region. A re-assessment of these primers indicated F1aCu-R3Cu
as the most suitable primer pair [19], but all three were selected to cover a broad range of
sequence diversity. However, no nirK amplicons were retrieved with Cunir3-Cunir4 when
applied on our strain set, which corresponded with the results of the re-evaluation [19].
This re-assessment did not result in better nirK primers than applied in our study, but did
find more suitable nirS primers than the two sets that were applied here, nirS1F-nirS6R and
nir3-nir4. However, the former primer pair [2] was not re-evaluated [19] with the originally
described touchdown PCR amplification procedure, while the latter pair [23] was not
included in this re-evaluation. Both primer pairs target the 3’ region of the target gene.
Further, this cultivation-dependent study demonstrated the unsuitability of now available
nir primers as broad-range amplification primers. As reported [16], the applied primers
sets for nirS and nirK were developed based on only a few sequences available at the time
and may be biased towards more conserved nir genes. It should be stressed that reliable
primers are a prerequisite for microbial community surveys, since they ultimately determine
what is detected in environmental culture-independent studies. The fallibility of applied
primers in environmental studies should always be kept in mind. More complete nir gene
sequences, which are to date very limited but most needed for primer design, will be
rendered by new and ongoing genome projects. Furthermore, the challenge now is to obtain
the nir gene sequences from those cultivated denitrifiers that do not render an amplicon.
These aberrant sequences are most interesting for future work and for the development of
new primers.
In conclusion, it is becoming apparent that nir gene diversity is not (always) congruent
with 16S rRNA gene phylogeny. Nir gene sequence analysis reveals functional diversity in
pure cultures or environmental studies, but should not be used to deduce the structural
diversity of denitrifying bacteria in culture-independent studies. The hypothesis that
denitrification genes can be horizontally transferred is gathering more indirect evidence.
CHAPTER 3
85
However, substantiation is needed from more in-depth cultivation-dependent studies and
complete gene sequence data of denitrification genes.
FUNCTIONAL PHYLOGENETIC ANALYSIS OF PURE CULTURE DENITRIFIERS
86
REFERENCES1. Altschul SF, Gish W, Miller W, Myers EW, Lipman DJ (1990) Basic local alignment search tool. J Mol
Biol 215:403-410
2. Braker G, Fesefeldt A, Witzel K-P (1998) Development of PCR primer systems for amplification of nitrite
reductase genes (nirK and nirS) to detect denitrifying bacteria in environmental samples. Appl Environ
Microbiol 64:3769-3775
3. Braker G, Zhou K, Wu L, Devol AH, Tiedje JM (2000) Nitrite reductase genes (nirK and nirS) as functional
markers to investigate diversity of denitrifying bacteria in Pacific Northwest marine sediment communities.
2243, Virgibacillus halodenitrificans LMG 9818T. An additional set of 21 denitrifying isolates from activated
sludge was also included [11]: nine Ochrobactrum sp. (R-24286, R-24289, R-24290, R-24291, R-24343, R-
24448, R-24467, R-24468, R-26465), three Rhizobium sp. (R-24260, R-24333, R-26467), one Acidovorax sp. (R-
24336), three Paracoccus sp. (R-24292, R-24342, R-26466), two Comamonas sp. (R-24447, R-24451), one
Thauera sp. (R-24450), one Pseudomonas sp. (R-24261) and one Pseudoxanthomonas sp. (R-24339).
Identification onto genus level was based on fatty acid methyl ester analysis and 16S rRNA gene sequence analysis.
Reliable species identification is given for reference strains from the BCCM/LMG culture collection.
DNA extraction and nor PCRDNA was extracted from each denitrifying isolate using the guanidium-thiocyanate-EDTA-sarkosyl method
described by Pitcher et al. [19] for fast-growing strains and using alkaline lysis for slow-growing isolates. For
alkaline lysis, one colony was suspended in an eppendorf tube with 20 µl of lysis buffer (2.5 ml 10% SDS; 5 ml 1M
NaOH; 92.5 ml MilliQ water). After 15 min at 95°C, 180 ml MilliQ water was added, the tube was centrifuged for
5 min at 13,000 g and the supernatant was transferred to a new tube. DNA extracts were stored at –20°C until use.
For amplification and sequence analysis of nir genes, two primer sets previously described were used: cnorB2F-
cnorB6R, cnorB2F-cnorB7R and qnorB2F-qnorB7R [4]. Temperature-time profiles and protocols were adopted
from the original description.
Nor gene sequence analysisThe nor amplicons were purified using the Nucleofast® 96 PCR system (Millipore). For each sequence reaction a
mixture was made using 3 µl purified and concentrated PCR product, 1 µl of BigDye™ Termination RR mix version
3.1 (Applied Biosystems), 1.5 µl of BigDye™ buffer (5x), 1.5 µl sterile milliQ water and 3 µl (20 ng/µl) of one of
the sequencing primers. The amplification primers were used as sequencing primers. The temperature-time profile
was as follows: 30 cycles of denaturation for 15 s at 96°C, primer annealing for 1 s at 35°C and extension for 4 min
at 60°C. The sequencing products were cleaned up as described previously [15]. Sequence analysis of the nir gene
was performed using the Applied Biosystems 3100 DNA Sequencer according to protocols provided by the
manufacturer.
Phylogenetic analysisForward and reverse strands of cnorB and qnorB were assembled with the BioNumerics 4.0 software and were
submitted to a BLAST search [1]. Amino acid sequences were deduced with Emboss Transeq (EMBL-EBI) and
aligned with sequences retrieved from the EMBL database using Clustalx [24]. Phylogenetic analyses were performed
with Treecon [26]. Trees were constructed with the neighbor -joining algorithm using the BLOSUM 62 substitution
matrix. Statistical evaluation of the tree topologies was performed by bootstrap analysis with 1000 resamplings.
Nucleotide sequence accession numbersThe cnorB and qnorB gene sequences have been submitted to the DDBJ/EMBL/GenBank databases. Accession
numbers are given in Figure 3.3.
FUNCTIONAL PHYLOGENETIC ANALYSIS OF PURE CULTURE DENITRIFIERS
92
RESULTS & DISCUSSION
Taxonomic distribution of denitrification genes in cultivated denitrifiers
The cnorB and qnorB genes were studied in a set of 227 denitrifying strains (Table 3.2).
One primer pair per gene was used, qnorB2F-7R and cnorB2F-7R [4]. The qnorB primers
target the region homologous with cnorB. The qnorB2F primer has an insertion in qnorB
absent in cnorB. The cnorB7R primer targets a conserved site, coding for two putative
histidine ligands. Other primers were described but they target almost identical regions [6],
and thus were not included in this study.
Thirty percent of all included denitrifiers (69 out of 227) rendered an amplicon for the norB
genes (Table 3.2). The primers were designed with the limited number of norB sequences
available in the public databases at that time [4]. Therefore, this low amplification percentage
of norB genes could be expected. The cnorB gene was detected in 48 denitrifiers (Table 3.2).
Of all cnorB amplicons, 30 were retrieved from isolates assigned to the Alphaproteobacteria
and 18 from isolates assigned to the Betaproteobacteria. The number of cnorB gene sequences
available in public databases and the dominance of cnorB in the pure culture denitrifiers in
this study suggest that the majority of the denitrifiers contain the membrane-anchored
dimer nitric oxide reductase NORCB, as was previously suggested [17]. Nevertheless, the
qnorB gene was detected in 20 cultured denitrifiers of which 15 were found within the
Betaproteobacteria, which extended the current knowledge on the occurrence of this norB
type significantly. Table 3.2 also represents the nir gene results of the same set of isolates
[11]. Both a nir and a norB gene were detected in 48 denitrifying isolates. The observed
coincidence of nir and norB genes suggested the dominant coexistence of nirK and cnorB
within Alphaproteobacterial denitrifiers (28 out of 31), as was previously suggested [17].
No distributional pattern could be observed within the Beta- and Gammaproteobacteria.
CHAPTER 3
93
Phylogeny of cnorB and qnorB gene products from cultivated denitrifiers
The sequence analysis was performed on amino acid sequences deduced from cnorB and
qnorB nucleotide sequences. The different steps of the amplification and sequencing
approach were validated by including a reference strain, Alcaligenes faecalis LMG 1229T,
previously shown to contain a qnorB sequence (accession number AJ507329). The here
obtained qnorB sequence of this strain was identical to sequence available in the public
database.
Phylogenetic analysis of norB genes showed the distinct clustering of cnorB and qnorB
gene products (Figure 3.3). It is difficult to assess the level of phylogenetic information
contained by these functional genes. The overall phylogeny of cnorB genes did not agree
with the widely accepted organism phylogeny based on 16S rRNA gene sequence analysis,
as in subcluster I members of different bacterial classes show a 100% amino acid sequence
similarity for the cnorB gene product (Figure 3.3). Indeed, the cnorB sequences of Acidovorax
sp. R-27044, belonging to the Betaproteobacteria, and Paracoccus sp. R-26897 and R-
28245, belonging to the Alphaproteobacteria, were identical, and the cnorB amino acid
sequence of Pseudomonas R-25208, belonging to the Gammaproteobactera, was identical
to the cnorB sequence of several Ochrobactrum strains (R-24291, R-24653, R-26821, R-
25018, R-24289, R-24618, R-25054, R-27045), belonging to the Alphaproteobacteria.
Also, the qnorB sequences separated denitrifying (subcluster V) from non-denitrifying
strains (subcluster IV), but without an internal grouping of taxa. However, the discrepancy
between norB gene phylogeny and organism phylogeny seems to be taxon-dependent. The
cnorB sequences delineated the Rhodocyclaceae, the Brucellaceae and the genus Paracoccus
quite clearly (Figure 3.3). And the functional group of ammonium-oxidizing bacteria could
also be distinguished from denitrifiers, as was described previously[6].
Table 3.2. (next two pages) An overview of denitrifying isolates arranged according to their taxonomic position.a +, nir/norB amplicon verified through sequence analysis; -, no amplificationb Results described previously [11].
FUNCTIONAL PHYLOGENETIC ANALYSIS OF PURE CULTURE DENITRIFIERS
94
CHAPTER 3
Iden
tific
atio
n
nir
K a
,b
nir
S a
,b
cno
rB a
q
no
rB a
S
trai
n n
r.
Alp
hap
rote
obac
teri
a
B
ruce
lla s
p.
+ -
+
- R
-268
95
O
chro
bact
rum
ant
hrop
i +
-
- -
LMG
213
6
O
chro
bact
rum
sp.
+
-
+ -
R-2
4289
, R-2
4291
, R-2
4334
, R-2
4618
, R-2
4638
, R-2
4653
, R-2
5018
, R-2
5054
,
R-2
5203
, R-2
6821
, R-2
6825
, R-2
6826
, R-2
6891
, R-2
6889
, R-2
6894
, R-2
6898
,
R-2
6900
, R-2
7045
, R-2
7046
, R-2
8410
+ -
-
- R
-242
86, R
-242
90, R
-244
48, R
-244
67, R
- 244
68, R
-246
39, R
-250
55, R
-264
65,
R-2
6890
, R-2
6892
M
ethy
loba
cter
ium
sp.
+
-
+ -
R-2
5207
R
hizo
bium
sp.
+
-
+ -
R-2
4333
, R-2
4654
, R-2
4658
, R-2
6467
, R
-268
20
+ -
-
- R
-242
60, R
-246
63, R
-268
35
S
inor
hizo
bium
sp
+ -
+
- R
-250
78
+ -
-
- R
-246
05, R
-250
67
P
arac
occu
s de
nitri
fican
s -
+
- +
LMG
404
9
P
arac
occu
s sp
. +
-
- -
R-2
4342
, R-2
4623
, R-2
4649
, R-2
4650
, R-2
4652
, R-2
5058
, R-2
6822
, R-2
6823
,
R-2
6824
, R-2
6841
, R-2
6899
, R-2
6901
, R-2
7049
, R-2
8242
-
+ +
- R
-268
97, R
-270
41
-
+ -
- R
-246
15, R
-246
16, R
-246
17, R
-246
65, R
-264
66, R
-282
39, R
-282
41, R
-284
14
-
-
+ -
R
-282
44, R
-282
45
-
-
- -
R-2
4292
, R-2
4621
, R-2
5049
, R-2
5059
, R-2
6819
, R-2
6839
, R-2
6844
, R-2
6888
, R-2
6893
,
R-2
6896
, R-2
6902
, R-2
7043
, R-2
7047
, R-2
8237
,R-2
8238
, R-2
8243
, R
-282
94, R
-284
09
P
anno
niba
cter
sp.
-
-
-
-
R
-270
42
Bet
apro
teo
bact
eria
A
lcal
igen
es fa
ecal
is
+ -
- +
LMG
122
9T
A
chro
mob
acte
r de
nitri
fican
s +
-
- +
LMG
123
1T
C
upria
vidu
s ne
cato
r -
+
- +
LMG
120
1
A
cido
vora
x sp
. +
-
- +
R-2
5052
+ -
-
- R
-243
36, R
-246
13, R
-246
14, R
-250
75, R
-250
76
-
+ -
- R
-252
12
-
-
+ -
R-2
4667
, R-2
5074
, R-2
6831
, R-2
6833
, R
-270
44
-
-
- -
R-2
4607
, R-2
4608
, R-2
6834
, R-2
6837
, R-2
8240
, R-2
6842
, R-2
6843
, R-2
8416
A
licyc
liphi
lus
sp.
-
+ -
+ R
-246
04, R
-246
11
-
-
- +
R-2
4606
, R-2
6814
Com
amon
as s
p.
- +
+ -
R-2
8235
-
+ -
- R
-244
47, R
-244
51, R
-250
14, R
-250
48, R
-250
66, R
-282
10, R
-282
24, R
-282
27, R
-282
31,
R-2
8233
, R-2
8234
, R-2
8293
, R-2
8452
-
-
- -
R-2
4660
, R-2
5057
, R-2
5060
, R-2
6810
, R-2
6812
, R-2
6817
, R
-268
29, R
-268
86, R
-268
87,
R-2
6905
, R-2
7040
, R-2
8209
, R-2
8211
, R-2
8218
, R-2
8220
, R-2
8222
, R-2
8223
, R-2
8225
,
R-2
8226
, R-2
8228
, R-2
8229
, R-2
8230
, R-2
8232
, R-2
8236
, R-2
8413
, R-2
8449
95
FUNCTIONAL PHYLOGENETIC ANALYSIS OF PURE CULTURE DENITRIFIERS
D
iaph
orob
acte
r sp
. +
-
- +
R-2
4610
, R-2
4612
, R-2
6815
-
-
- +
R-2
4661
, R-2
5011
, R-2
6840
, R-2
8417
-
-
- -
R-2
5012
S
impl
icis
pira
sp.
-
-
-
- R
-284
08
A
zosp
ira s
p.
-
+ -
- R
-284
47
-
-
- -
R-2
5019
A
zovi
brio
sp.
-
-
+
- R
-250
62
D
echl
orom
onas
sp.
-
+
+ -
R-2
8407
-
+ -
- R
-284
00, R
-284
51
-
-
+ -
R-2
8401
-
-
- -
R-2
8215
, R-2
8291
, R-2
8317
, R
-284
12
T
haue
ra s
p.
-
+
+ -
R-2
5071
, R-2
6906
-
+
- -
R-2
4450
, R-2
5069
, R-2
6885
, R-2
8213
-
-
+ -
R-2
8205
, R-2
8207
, R-2
8289
, R-2
8403
, R-2
8404
-
-
- -
R-2
8206
, R-2
8216
, R-2
8217
, R-2
8292
, R-2
8312
, R-2
8402
, R-2
8405
, R-2
8406
, R-2
8448
Z
oogl
oea
sp.
-
-
- -
R-2
8315
Gam
map
rote
ob
acte
ria
P
seud
omon
as a
erug
inos
a -
+
- -
LMG
124
2
P
seud
omon
as s
tutz
eri
-
+ -
- LM
G 2
243
P
seud
omon
as s
p.
+ -
-
+ R
-242
61, R
-268
28
+ -
+
+ R
-252
08
+ -
-
- R
-246
09
-
+ -
- R
-250
61, R
-268
30
-
-
- +
R-2
5209
-
-
- -
R-2
4636
, R-2
5016
, R-2
5343
, R-2
8208
, R-2
8219
P
seud
oxan
thom
onas
sp.
-
-
-
-
R
-243
39
Ep
silo
nb
acte
ria
A
rcob
acte
r sp
. -
-
-
- R
-282
14, R
-283
13, R
-283
14, R
-283
16
Fir
mic
ute
s
V
irgib
acill
us h
alod
enitr
ifica
ns
-
-
-
-
LMG
981
8
B
acill
us s
p.
-
-
-
-
R-2
4620
, R-2
4666
, R-2
8399
T
richo
cocc
us s
p.
-
-
-
-
R-2
8212
E
nter
ococ
cus
sp.
+ -
-
+ R
-246
26
+ -
-
- R
-252
05
P
aeni
baci
llus
sp.
-
-
-
-
R-2
7048
S
taph
yloc
occu
s sp
. +
-
-
-
R-2
5050
Bac
tero
idet
es
C
hrys
eoba
cter
ium
sp.
+
-
-
-
R-2
5053
96
Figure 3.3. Phylogenetic analysis of norB gene products. Unrooted neighbor joining tree was based on partialnorB gene products (145 amino acids). Bootstrap values were generated from 1000 replicates of neighbor joining.•, bootstrap values higher than 70%; bootstrap values lower than 70% are omitted. Strains analyzed in this studyare in bold. The accession numbers are given between parentheses.
Characterisation and transcriptional analysis of Pseudomonas fluorescens denitrifying clusters
containing the nar, nir, nor and nos genes. Biochim Biophys Acta 1517:436-440
17. Philippot L (2002) Denitrifying genes in bacterial and Archaeal genomes. Biochim Biophys 1577:355-
376
18. Philippot L (2005) Denitrification in pathogenic bacteria: for better or worst? Trends Microbiol
13:191-192
19. Pitcher DG, Saunders LA, Owen NA (1989) Rapid extraction of bacterial genomic DNA with guanidium
thio-cyanate. Lett Appl Microbiol 8:151-156
20. Priemé A, Braker G, Tiedje JM (2002) Diversity of nitrite reductase (nirK and nirS) gene fragments in
forested upland and wetland soils. Appl Environ Microbiol 68:1893-1900
21. Sharma S, Aneja MK, Mayer J, Munch JC, Schloter M (2005) Diversity of transcripts of nitrite reductases
genes (nirS and nirK) in rhizospheres of grain legumes. Appl Environ Microbiol 71:2001-2007
22. Suharti, Heering HA, de Vries S (2004) NO reductase from Bacillus azotoformans is a bifunctional
enzyme accepting electrons from menaquinol and a specific endogenous membrane-bound cytochrome
c551
. Biochemistry 43:13487-13495
23. Suharti S, Strampraad MJ, Schröder T, De Vries S (2001) A novel copper A containing menaquinol NO
reductase from Bacillus azotoformans. Biochemistry 40:2632-2639
24. Thompson JD, Gibson TJ, Plewniak F, Jeanmougin F, Higgins DG (1997) The ClustalX windows
interface: flexible strategies for multiple sequence alignment aided by quality analysis tools. Nucleic
Acids Res 24:4876-4882
FUNCTIONAL PHYLOGENETIC ANALYSIS OF PURE CULTURE DENITRIFIERS
100
25. Tiedje JM (1988) Ecology of denitrification and dissimilatory nitrate reduction to ammonium. In:
Environmental microbiology of anaerobes, Zehnder AJB (ed.), John Wiley & Sons, New York, pp.
179-244
26. Van de Peer Y, De Wachter R (1994) TREECON for Windows: a software package for the construction
and drawing of evolutionary trees for the Microsoft Windows environment. Comput Applic Biosci
10:569-570
27. Yan T, Fields MW, Wu L, Zu Y, Tiedje JM, Zhou J (2003) Molecular diversity and characterization of
nitrite reductase gene fragments (nirK and nirS) from nitrate- and uranium-contaminated groundwater.
Environ Microbiol 5:13-24
28. Yoshie S, Nado N, Tsuneda S, Hirata A, Inamori Y (2004) Salinity decreases nitrite reductase gene
diversity in denitrifying bacteria of wastewater treatment systems. Appl Environ Microbiol 70:3152-
3157
29. Zumft WG (1997) Cell biology and molecular basis of denitrification. Microbiol. Mol Biol Rev 61:533-
616
30. Zumft W (2005) Nitric oxide reductase of prokaryotes with emphasis on the respiratory, heme-copper
oxidase type. J Inorg Biochem 99:194-215
CHAPTER 3
3.3 FUNCTIONAL GENE STUDY ON HETEROTROPHICDENITRIFIERS ISOLATED FROM SOIL
Redrafted from: Heylen K, Boon N, Verstraete W, De Vos P (2007) Functional
gene study on heterotrophic denitrifiers isolated fom soil. Microb Ecol, Submitted
SUMMARY
The key denitrification genes - that is nirS and nirK, encoding nitrite reductase, and cnorB
and qnorB, encoding nitric oxide reductase - were further investigated in a total of 112
heterotrophic soil denitrifiers. A very low number of the denitrifying isolates yielded a nir
or norB amplicon, partly due to the high percentage of bacilli. Isolates of Sinorhizobium,
Cupriavidus, and Pseudomonas, showing no genetic variation, gave nir or norB sequences
without polymorphisms. Detection of denitrification genes in Ensifer and Rhizobium
appeared to be strain-dependent. Unexpected norB sequences phylogeny was observed for
bacilli. These norB sequences clustered together with sequences from non-denitrifiers, and
were only very distantly related with the few norB sequences available from complete
genome projects of other bacilli. Interestingly, a Bacillus isolate was found containing both
a cnorB and qnorB gene, with only the latter showing the expected phylogeny. These
observations demonstrate the usefulness of the cultivation approach in the study of the
denitrification process and its ecology.
104
INTRODUCTION
Denitrification decreases the application efficiency of agricultural fertilizer, contributes to
global warming through N2O emission, but can also remove excess nitrogen from the
environment, counteracting nitrate pollution of ground- and drinking water [22]. It is
primarily a bacterial respiratory process in which four enzymatic reaction steps are catalyzed
by metalloproteins. True denitrifiers contain a nitrite reductase and a nitric oxide reductase,
because they are responsible for converting fixed into gaseous nitrogen. Nitrite can be
reduced to NO through the action of two different enzymes containing either copper or
cytochrome cd1, encoded by nirK and nirS respectively [47]. Three nitric oxide reductases,
which reduce NO to N2O, have been described to date: (i) a dimeric enzyme, only found in
bacterial strains that perform denitrification [3, 16] and encoded by cnorB, (ii) a monomeric
enzyme found in both denitrifying and non-denitrifying strains [8, 11, 25] and encoded by
qnorB, and (iii) a newly described qCuANOR, purified from Bacillus azotoformans [37],
for which the encoding genes are hitherto unknown.
Most denitrification studies qualitatively or quantitatively detect and analyse functional
genes in a cultivation-independent way. They thereby try to relate the occurrence of these
genes to environmental conditions [2, 21, 45, 46], the presence of other metabolic guilds
[23, 40] or other factors [37]. Unfortunately, this approach does not reveal information on
the denitrifier’s identity, because denitrification genes do not contain reliable taxonomic
information [24]. Also, the denitrifying cell is not collected, making further in depth studies
very difficult. In this context, the cultivation approach provides complementary information
for understanding the ecology of the denitrificationn process.
Here, we investigated the key functional genes from pure culture denitrifiers isolated from
soil. An in-depth discussion of the incongruence between nir and norB gene phylogenies
and organism phylogeny was reported previously [17, 18]. This work focuses on the
sequence heterogeneity of denitrifying strains of the same species or genus, and the
unexpected phylogeny of norB genes within the genus Bacillus. The results again confirmed
the unsuitability of available functional primers to detect all present gene sequence diversity.
CHAPTER 3
105
MATERIALS & METHODS
Biological materialThe isolation of heterotrophic denitrifiers from soil is described in Section 3.2 of this thesis. The majority of the
isolates belonged to Bacillus (71%) and the Rhizobiaceae (16%).
Repetitive element PCRBOX-PCR was performed with primer BoxA1R (5’-CTACGGCAAGGCGACGCTGACG-3’) as described
previously [26]. Gels were normalised and analysed with BioNumerics 4.6. Similarity matrices of densiometric
curves of the gel tracks were calculated with the Pearson’s product-moment correlation coefficient. Cluster analyses
of similarity matrices were performed by unweighted pair group method with arithmetic averages (UPGMA). Two
cluster significance tools were applied to express the stability of the error at each branching level and the overall
quality of the cluster analysis: (i) cluster cut-off method, which delimitates relevant clusters at different similarity
levels and corresponds to the lowest similarity value within the entries of a dendrogram; (ii) the cophenetic
correlation, which allows to distinguish reliable from unreliable subclusters.
Functional PCRs and sequence analysis of nir and norB genesFor amplification of nirS, primer pairs nirS1F-nirS6R [6] and nir3-nir4 [35] were used. For amplification of nirK,
primer pairs nirK1F-nirK5R [6], F1aCu-R3Cu [15] and Cunir3-Cunir4 [10] were used. Primer pairs cnorB2F-
cnorB7R and qnorB2F-qnorB7R were used to amplify cnorB and qnorB respectively [7]. Amplification of all
genes was performed as originally described. Because of low amplification percentages, all functional PCRs were
repeated two times on two different PCR machines, GeneAmp® PCR system 9600 (Applied Biosystems) and
MyCyclerTM thermal cycler (Bio-Rad). Further purification and sequencing protocol was described previously
[17]. The different steps of the amplification and sequencing protocol were validated by including reference strains
containing denitrification genes already published, i.e. Achromobacter denitrificans LMG 1231T for nirK
(accession number AJ224905), Pseudomonas stutzeri LMG 2243 for nirS (accession number X56813) and
Alcaligenes faecalis LMG 1229T for norB (accession number AJ507329).
Phylogenetic analysisForward and reverse strands of nirS and nirK were assembled with the BioNumerics 4.6 software and were submitted
to a BLAST search [1]. Amino acid sequences were deduced with Emboss Transeq (EMBL-EBI) and aligned with
sequences retrieved from the EMBL database using ClustalX [41]. Phylogenetic analyses were performed with
Treecon [43]. Trees were constructed with the neighbor-joining algorithm using the BLOSUM 62 substitution
matrix. Statistical evaluation of the tree topologies was performed by bootstrap analysis with 1000 resamplings.
Nucleotide sequence accession numbersThe sequence data generated in this study have been deposited in Genbank/EMBL/DDBJ. The accession numbers
of the 16S rRNA genes can be found in the supplementary table. The accession numbers of the nirS, nirK, cnorB and
qnorB genes are included in Figure 3.4, 3.5 and 3.6 respectively.
FUNCTIONAL PHYLOGENETIC ANALYSIS OF PURE CULTURE DENITRIFIERS
106
RESULTS & DISCUSSION
Denitrification genes in cultivated heterotrophic soil denitrifiers
All heterotrophic soil denitrifiers were subjected to five different nir PCRs, three for the
nirK gene and two for the nirS gene. Only 19.6% of the denitrifiers yielded a nir amplicon
(14 for nirK and 8 for nirS), and 12.5 % for norB (1 for cnorB and 13 for qnorB). These
results are significantly less than previously reported for both genes using the same primer
set (50% for the nir genes and 30% for the norB genes) [17, 18]. However, these very low
amplification percentages could be explained by the bulk of denitrifying Bacillus isolates in
the strain set. Until recently, almost no DNA sequences for denitrification genes from
bacilli were available in the public sequence databases, although a copper-containing nitrite
reductase was described in Virgibacillus halodenitrificans (previously Bacillus
halodenitrificans) [12] and a nirK and qnorB gene was found in the genome of the
Therefore, available primers for the denitrification genes were designed based on functional
sequences from only Gram-negative denitrifiers [6, 7, 10, 15, 35]. Nevertheless, norB genes
were found in ten Bacillus isolates, one of which contained both cnorB and qnorB, and in
three Cupriavidus isolates. Suprisingly, no norB amplicons were detected in Pseudomonas
isolates or members of the Rhizobiaceae, although representatives of taxa closely related to
the latter, such as Bradyrhizobium japonicum and Paracoccus denitrificans, were included
in the primer design [7]. All nirK amplicons were found within the isolated
Alphaproteobacterial rhizobia. Primer set Cunir3-4 was included because of its potential
for amplification of a greater diversity [10], but no nirK amplicons were retrieved, as was
also reported previously [17]. NirS amplicons were retrieved from Cupriavidus
(Betaproteobacteria) and Pseudomonas (Gammaproteobacteria) isolates.
These observations again underline the need for broad range primers, covering the whole
sequence diversity. These primers ultimately determine what is detected in functional
microbial community surveys, and several studies have already reported the inability to
amplify nirS genes [34, 45] from the environment. More complete denitrification gene
sequences, which are to date very limited, will be rendered by new and ongoing genome
projects and become available for primer design.
Table 3.3. (opposite page) An overview of denitrifying isolates arranged according to their taxonomic position,with the PCR results for nir and norB genes. Isolation and identification of denitrifiers is described in Section 2.2.
CHAPTER 3
107
Iden
tifi
cati
on
nir
Kn
irS
cno
rBq
no
rBS
trai
n n
r.
Alp
hap
rote
ob
acte
ria
Ens
ifer
sp.
- -
- -
R-3
1757
+ -
- -
R-3
2544
Rhi
zobi
um s
p. -
- -
-R
-317
62, R
-325
39, R
-325
52 +
- -
-R
-315
49, R
-318
37, R
-318
57, R
-327
25S
inor
hizo
biu m
sp.
+ -
-
-R
-317
59, R
-317
63, R
-317
64, R
-318
16, R
-325
42, R
-325
46, R
-325
49, R
-327
3
Bet
apro
teo
bac
teri
aC
upria
vidu
s s
p. -
+ -
+R
-315
42, R
-315
43, R
-315
44
Gam
map
rote
ob
acte
ria
Pse
udom
onas
sp.
- -
- -
R-3
1765
, R-3
2541
, R-3
2726
- +
- -
R-3
1758
, R-3
1761
, R-3
1815
, R-3
1817
, R-3
2553
Fir
mic
ute
sB
acill
u s s
p. -
- -
-R
-327
05, R
-315
46, R
-318
48, R
-318
50, R
-318
55, R
-325
24, R
-325
25, R
-325
29,
RR
-325
74, R
-326
61, R
-326
99, R
-327
78, R
-327
79, R
-327
81, R
-328
48, R
-328
49, R
R-3
2518
, R-3
1554
, R-3
3774
, R-3
2784
, R-3
1852
, R-3
1541
, R-3
2663
, R-3
1849
, R-3
257
R-3
2706
, R-3
1835
, R-3
1769
, R-3
1830
, R-3
2700
, R-3
1547
, R-3
2707
, R-3
2695
, RR
-328
51, R
-318
34, R
-325
34, R
-327
87, R
-325
28, R
-326
96, R
-326
62, R
-325
17, R
R-3
2703
, R-3
2717
, R-3
3819
, R-3
2523
, R-3
1836
, R-3
1838
, R-3
3776
, R-3
2516
, RR
-328
45, R
-327
04, R
-327
89, R
-315
50, R
-318
45, R
-328
44, R
-315
53, R
-315
5 -
-
- +
R-3
1770
, R-3
1841
, R-3
2526
, R-3
2656
, R-3
2702
, R-3
2709
, R-3
2715
, R-3
377
- -
+
+R
-326
94S
taph
yloc
occu
s s
p. -
- -
-R
-337
71, R
-337
77, R
-341
81
FUNCTIONAL PHYLOGENETIC ANALYSIS OF PURE CULTURE DENITRIFIERS
108
Phylogeny of nir gene products
Sequence analysis was performed on amino acid sequences deduced from nirK and nirS
nucleotide sequences, after confirmation of the presence of their functional domains.
The isolated representatives of Sinorhizobium, all having a 100% 16S rRNA gene sequence
similarity with the type strain of S. morelense, yielded nirK sequences without any
polymorphisms. Four Rhizobium isolates gave a nirK amplicon, while three did not. Also,
only one of the two Ensifer isolates, both with a 100% 16S rRNA gene sequence similarity
to the type strain of E. medicae, gave a nirK amplicon. The phylogenetic tree of nirK gene
products is represented in Figure 3.4. In contrast to a previous report [17], nirK sequences
of members of the Rhizobiaceae did not form a separate cluster but grouped with nirK
sequences obtained from representatives of Ochrobactrum, Pseudomonas,
Chryseobacterium and Alcaligenes, while the nirK sequence of Bradyrhizobium japonicum
USDA 110 clustered separately.
Figure 3.4. Phylogenetic analysis of nirK gene products. Unrooted neighbor joining tree was based on partialnirK amplicons (130 amino acid positions; accession numbers in parentheses). Bootstrap values were generatedfrom 1000 replicates of neighbor joining. Bootstrap values higher than 70% are given. Sequences retrieved fromthis study are in bold.
FUNCTIONAL PHYLOGENETIC ANALYSIS OF PURE CULTURE DENITRIFIERS
Only eight nirS amplicons were found. All three Cupriavidus isolates, highly related to C.
necator LMG 1201T, and all five Pseudomonas isolates, highly related to P. kilonensis
LMG 1242T, gave a nirS amplicon, in which, for both taxa, no sequence polymorphisms
were observed. The phylogenetic tree of nirS gene products is represented in Figure 3.5.
The three Cupriavidus sequences clustered closely to the NiRS sequence of Cupriavidus
necator. The Pseudomonas sequences clustered closely to NiRS of the type strain of P.
aeruginosa. However, other Pseudomonas sequences were grouped in other clusters with
strains belonging to Betaproteobacteria.
Figure 3.5. Phylogenetic analysis of nirS gene products. Unrooted neighbor joining tree was based on partial nirSamplicons (189 amino acid position; accession numbers in parentheses). Bootstrap values were generated from1000 replicates of neighbor joining. Bootstrap values higher than 70% are given. Sequences retrieved from thisstudy are in bold.
Cupriavidus sp. R-31544 ( 403575) AM Cupriavidus sp. R-31543 ( 403574) AM
Thauera linaloolentis 47LolT (AY078265)
Marinobacter sp. NBC31 (AJ626831)Thauera aromatica K172 (AY078256) T
100
85
100
79
100100
99
100
100
100
100
87
100
100
93
110
CHAPTER 3
Cluster analysis of nirK and nirS gene products again confirmed the lack of taxonomic
information contained by these proteins. Repetitive sequence polymorphism analysis was
done for the denitrifying isolates (data not shown) to assess their genetic heterogeneity and
link these results to obtained nir sequence diversity or absence of nir data. This showed
that the Sinorhizobium, Pseudomonas and Cupriavidus isolates all represented one strain
in each taxon, explaining the lack of sequence polymorphisms in nirK and nirS respectively.
For Ensifer and Rhizobium, analysis of repetitive element revealed that the isolates with a
nirK amplicon were genetically different from those without an amplicon; they belonged to
different strains. Thus, not only the process itself [33], but also its detection is strain-
dependent, suggesting significant sequence divergence within these genera, as was previously
reported for the Rhodocyclaceae [35].
111
FUNCTIONAL PHYLOGENETIC ANALYSIS OF PURE CULTURE DENITRIFIERS
Phylogeny of nor gene products
Sequence analysis was performed on amino acid sequences deduced from cnorB and qnorB
nucleotide sequences, after confirmation of the presence of heme and iron binding sites.
Amino acid sequences from other cultured and named denitrifiers available in the EMBL
sequence database were also included. The NORB cluster was divided in the qNORB and
the cNORB cluster. The qnorB gene products fell into two distinct clusters, supported
with a bootstrap value of 100%. In the first qnorB gene product cluster, all Bacillus
sequences grouped together with sequences from non-denitrifying strains or strains for
which the denitrification capacity has not been unequivocally determined (Figure 3.6).
Figure 3.6. Phylogenetic analysis of cnorB and qnorB gene products. Unrooted neighbor joining tree was basedon partial norB gene products (138 amino acid positions; accession numbers in parentheses). Bootstrap valueswere generated from 1000 replicates of neighbor joining. Bootstrap values higher than 70% are given. Sequencesretrieved from this study are in bold.
sp. R-25011 (AM284334) for qnorB. Probes were prepared via amplification with primers cnorB2F-cnorB7R and
qnorB2F-qnorB7R as described previously [1], purification using the Nucleofast® 96 PCR system (Millipore),
dilution to 10 ng/µl and subsequent guidelines of the ‘ECL direct nucleic acid labelling and detection system’
(Amersham Biosciences). The GC content of the selected probes ranged between 58 and 60%. Individual probes and
combinations of either cnorB probes, qnorB probes or both were used.
Dot-blot hybridizationThe ‘ECL direct nucleic acid labelling and detection system’ (Amersham Biosciences) was used for dot-blot
hybridization. Optimization was performed on three positive and three negative controls, and sterile MilliQ as a
blanc. The hybridization temperature was dependent of the kit and set at 42°C. The hybridization stringency was
altered through a range of NaCl concentration, from 0.1 to 1 M. After hybridization, a primary wash for 10 minutes
at 42°C was repeated three times, followed by a secondary wash for 10 minutes at room temperature repeated two
times. Both post-hybridization washes were performed with different stringencies: from 0.1x to 5x SSC.
Autoradiography film was exposed to the hybridized membrane for 40 min and developed. The norB dot-blot of the
selected denitrifiers was repeated three times, with place re-arrangements to check for edge effects.
Phylogenetic analysis of 16S rRNA gene sequencesDNA extraction, 16S rRNA gene sequence analysis and taxon assignment of Paracoccus isolates were described
previously [10, 12]. Sequences were aligned with 16S rRNA gene sequences of closely related type strains of the
genus retrieved from from the EMBL databse using ClustalX [25]. Phylogenetic analyses were performed with
Treecon [26]. Trees were constructed with the neighbor-joining algorithm. Statistical evaluation of the tree
topologies was performed by bootstrap analysis with 1000 resamplings.
SCREENING FOR DENITRIFICATION GENES UNDETECTABLE WITH PCR
128
RESULTS & DISCUSSION
Optimization norB dot-blot hybridization
A dot-blot hybridization procedure was developed to screen for the presence of norB in
pure culture denitrifiers. Several dot-blot hybridization conditions were optimized: i) target
DNA preparation, ii) applied probe mixtures, iii) stringency of the hybridization buffer
through the concentration of NaCl, iv) stringency of primary and secondary washes. During
optimization, the denitrifying strains used for probe production were positive controls
(see further), while three non-denitrifying strains were chosen as negative controls, namely
reactions for all positive controls and simultaneously negative (or very light positive)
reactions for all negative controls.
Both high-molecular genomic DNA and sheared DNA were used as target DNA. Shearing
was thought to ameliorate the target availability in the genomic DNA; however, the results
for both types of target DNA were identical. Six partial gene amplicons were selected for
probe use, three for the cnorB gene and three for the qnorB gene, based on sequence
divergence and taxon assignment. A combined strategy for both genes was chosen based on
their high sequence homology [2, 27]. Both single cnorB and qnorB probes were used, as
also probe mixtures of all three cnorB, all three qnorB and all six norB amplicons. The latter
probe mixture resulted in the best dot-blot reactions for all control strains, with the added
value of only one dot-blot test for both norB genes. The optimal hybridization buffer
contained 1M NaCl, allowing an average mismatch of at least 20% (calculated based on
norB sequences of the probes), necessary to cover the known divergence of norB genes.
Best results were obtained with a primary wash using 5x SSC, followed by secondary wash
using 2x SSC. Although the secondary wash had very low stringency, these results were
most optimal.
CHAPTER 4
129
Detection of norB through dot-blot hybridization
A subset of 80 isolates from activated sludge [12] and soil [9] was selected, for which the
denitrifying capacity was previously demonstrated and no norB genes could be amplified.
The three most represented genera were Bacillus (19 isolates), Comamonas (15 isolates)
and Paracoccus (14 isolates). Also, strain Virgibacillus halodenitrificans LMG 9818 was
included, for which to date no nitric oxide reductase genes have been sequenced. The
optimized dot-blot protocol allowed screening for norB in this subset; a blanc, two positive
and two negative controls were included (Figure 4.1). The results were scored positive after
visual comparison with the negative controls.
Twenty-two denitrifiers (or 27.50%) showed a distinct positive result, indicating the presence
of either cnorB or qnorB, which were previously undetected by PCR. Most included
Paracoccus isolates scored positive, as well as other members of the Alphaproteobacteria,
belonging to the genera Pannonibacter, Ochrobactrum, Rhizobium and Sinorhizobium, and
strains from the Betaproteobacterial Comamonadaeae family (genera Comamonas and
Acidovorax). The results of eleven denitrifiers (or 13.75%) were assessed as doubtful due
to a very light reaction, similar to the faint positive result of K. pneumoniae subsp.
pneumoniae (Figure 4.1). These observations could be attributed to (i) interference of the
norB hybridization with distantly related genes of the heme-copper oxidase superfamily, or
(ii) positive results but suboptimal hybridization conditions. Their positive or negative
character can only be confirmed by genomic sequencing.
All positive-scoring strains belonged to taxa known for their denitrifying capacities. The
detection of these norB genes previously undetected by PCR confirmed the unsuitability of
the available norB primers or PCR protocols [1] for broad-range application. Remarkable is
that, although the primers were designed with the limited norB sequences available at the
time, the positive-scoring strains were relatively closely related to the denitrifiers used for
primer and PCR development, such as Bradyrhizobium japonicum and Paracoccus
denitrificans [1].
SCREENING FOR DENITRIFICATION GENES UNDETECTABLE WITH PCR
130
Figure 4.1. NorB dot-blot hybridization of pure culture denitrifiers as target DNA and partial norB gene ampliconsas probes. Genus assignment and strain number are given in bold for positive results, and underlined for doubtfulresults. A blanc and positive and negative controls were included.
CHAPTER 4
Bac
illu
s sp
. R
-327
06
Bla
nc
Bac
illu
s sp
. R
-327
05A
cido
vora
x sp
. R
-246
13O
chro
bact
rum
sp.
R
-284
10T
haue
ra s
p.
R-2
5069
Tha
uera
sp.
R
-282
16T
haue
ra s
p.
R-2
8213
Com
amon
as s
p.
R-2
8228
Sin
orhi
zobi
um s
p.
R-3
2737
Rhi
zobi
um s
p.
R-3
1549
Com
amon
asR
-284
13
Bac
illu
s sp
. R
-327
79P
seud
omon
as s
p.
R-2
4609
Bac
illu
s sp
. R
-315
47S
inor
hizo
bium
sp.
R
-250
67D
echl
orom
onas
sp.
R
-282
15B
acil
lus
sp.
R-2
6824
Com
amon
as s
p.
R-2
8226
Rhi
zobi
um
sp.
R
-327
25P
arac
occu
s R
-284
09
Dia
phor
obac
ter
sp.
R-2
5011
Bac
illu
s sp
. R
-328
49P
arac
occu
s sp
. R
-246
16B
acil
lus
sp.
R-3
1835
Aci
dovo
rax
sp.
R-2
5075
Dec
hlor
omon
as s
p.
R-2
8291
Com
amon
as s
p.
R-2
8224
Com
amon
as s
p.
R-2
8229
Virg
ibac
illus
ha
lode
nitr
ifica
ns
LM
G 9
818
Pse
udom
onas
sp.
R
-317
58Tr
icho
cocc
uR
-282
12
Esc
heri
chia
col
iL
MG
209
2T
Bac
illu
s sp
. R
-327
81P
arac
occu
s sp
. R
-246
23B
acil
lus
sp.
R-3
1550
Chr
yseo
-ba
cter
ium
sp.
R
-250
53T
haue
ra s
p.
R-2
8292
Com
amon
as s
p.
R-2
8227
Com
amon
as s
p.
R-2
8230
Pse
udo-
xant
hom
onas
sp.
R
-243
39Si
norh
izob
ium
sp.
R
-317
59T
haue
ra s
p.R
-284
48
Kle
bsie
lla p
neum
onia
eL
MG
209
5T
Bac
illu
s sp
. R
-327
89P
arac
occu
s sp
. R
-246
49B
acil
lus
sp.
R-3
1846
Och
roba
ctru
m s
p.
R-2
5055
Com
amon
as s
p.
R-2
8232
Com
amon
as s
p.
R-2
8231
Com
amon
as s
p.
R-2
8236
Aci
dovo
rax
sp.
R-2
4608
Sino
rhiz
obiu
m s
p.
R-3
1764
Com
amon
asR
-284
49
Och
roba
ctru
m s
p.
R-2
4286
Bac
illu
s sp
. R
-328
45P
arac
occu
s sp
. R
-246
50B
acil
lus
sp.
R-3
1541
Ent
eroo
ccus
sp.
R
-252
05B
acil
lus
sp.
R-3
1830
Par
acoc
cus
sp.
R-2
8241
Aci
dovo
rax
sp.
R-2
8416
Pan
noni
bact
er s
p.
R-2
7042
Sin
orhi
zobi
um s
p.
R-3
1816
Com
amon
as s
p.
R-2
8218
Och
roba
ctru
m s
p.
R-2
4290
Stap
hylo
cocc
us s
p.
R-3
3771
Par
acoc
cus
sp.
R-2
4652
Bac
illu
s sp
. R
-326
63P
arac
occu
s sp
. R
-270
49B
acil
lus
sp.
R-3
2700
Par
acoc
cus
sp.
R-2
8242
Par
acoc
cus
sp.
R-2
7043
Com
amon
as s
p.
R-2
6887
Sin
orhi
zobi
um s
p.
R-3
2542
Com
amon
as s
p.
R-2
8222
Par
acoc
cus
sp.
R-2
4342
Par
acoc
cus
sp.
R-2
4665
Par
acoc
cus
sp.
R-2
6822
Bac
illu
s sp
. R
-317
69T
haue
ra s
p.
R-2
8206
Par
acoc
cus
sp.
R-2
6888
Ens
ifer
sp.
R
-325
44C
omam
onas
sp.
R
-282
25
Pos
itiv
e co
ntro
ls
Neg
ativ
e c
ontr
ols
131
To assess whether strains with high 16S rRNA gene sequence similarity resulted in similar
norB results, a 16S rRNA gene sequence analysis was performed for members of the genus
Paracoccus. Paracoccus strains with a norB amplicon, without a norB amplicon, even after
repeated trails, but with a positive results for the norB dot blot, and without a norB
amplicon and with doubtful results for the dot-blot, were included (Figure 4.2). For Paracoccus
isolates affiliated with P. aminovorans, similar affiliation did not correlate with similar norB
detection results. Isolates for which amplicons were retrieved clustered together with isolates
giving positive and doubtful results with dot-blot screening. Six Paracoccus isolates (R-
28409, R-24623, R-24652, R-24650, R-26822, R-24649, R-28245) with positive dot-blot
results clustered together, as well as two isolates with a norB amplicon. Two isolates
affiliated with P. versutus gave different dot-blot results. Thus, high 16S rRNA gene sequence
similarities (>97%, data not shown) were found between Paracoccus isolates with different
results for norB detection. Thus, the discrepancy in norB detection between PCR and dot-
blot can only be explained by high inter- and intraspecies norB sequence divergence. This
was also concluded by Song and Ward [19], after failure to amplify nir genes in denitrifying
strains closely related to those in which nir genes were detected for Rhodocyclaceae.
The next step should be the retrieval of these norB sequences to assess their divergence
causing failure to amplify these genes and the origin of this divergence. Two strategies could
be followed, either norB dot-blot screening of a genomic clone libray and subsequent
sequencing, or genome sequencing of one or several Paracoccus isolates. Both approaches
can provide extra information on gene organization and regulation, which can be used to
substantiate possible horizontal gene transfer (HGT) [17], which was suggested previously
as cause for the high norB gene sequence divergence [1, 11, 14, 17, 27]. Newly gathered
sequence information could stimulate novel development of amplification protocols, possibly
focusing on group-specific strategies, with grouping based on taxon assignment or gene
phylogeny.
SCREENING FOR DENITRIFICATION GENES UNDETECTABLE WITH PCR
132
Figure 4.2. Phylogenetic dendrogram obtained by neighbour-joining of the 16S rRNA gene sequences ofParacoccus. The norB detection results are given for the denitrifying Paracoccus isolates. EMBL accessionnumbers are shown in parenthesis. Rhodobacter sphaeroides ATCC 17023T was used as outgroup. Bootstrapvalues (expressed as percentages of 1000 replicates) are shown at the branch points.
0.05
sub
stitu
tions
/site
Rho
doba
cter
sph
aero
ides
AT
CC
170
23 (
M55
498)
T
Par
acoc
cus
deni
trif
ican
s A
TC
C 1
7741
(Y
1692
7)
T
Par
acoc
cus
amin
ophi
lus
AT
CC
496
73 (
AY
0141
76)
T
Par
acoc
cus
yeei
CD
C G
1212
(A
Y01
4173
)T
Par
acoc
cus
caro
tinifa
cien
s E
-396
(A
B00
6899
)T
Par
acoc
cus
sp. R
-282
42 (A
M08
4137
)
Par
acoc
cus
amin
ovor
ans
JCM
768
5 (
D32
240)
T
Par
acoc
cus
pant
otro
phus
AT
CC
355
12 (
Y16
933)
T
Par
acoc
cus
vers
utus
AT
CC
253
64 (
AY
0141
74)
T
Par
acoc
cus
sp. R
-268
88 (
AM
0840
61)
Par
acoc
cus
sp. R
-243
42 (
AM
2310
51)
Par
acoc
cus
sp. R
-282
44 (
AM
0841
79)
Par
acoc
cus
sp. R
-282
41 (
AM
0841
63)
Par
acoc
cus
sp. R
-270
41 (A
M08
4095
)
Par
acoc
cus
sp. R
-270
43 (A
M08
4092
)
Par
acoc
cus
sp. R
-284
09 (
AM
0841
34)
Par
acoc
cus
sp. R
-246
23 (
AM
0840
03)
Par
acoc
cus
sp. R
-246
52 (
AM
0839
98)
Par
acoc
cus
sp. R
-246
50 (
AM
0840
45)
Par
acoc
cus
sp. R
-268
22 (
AM
0840
57)
Par
acoc
cus
sp. R
-246
49 (
AM
0840
02)
Par
acoc
cus
sp. R
-282
45 (
AM
0841
78)
Par
acoc
cus
sp. R
-268
97 (
AM
0841
00)
Par
acoc
cus
alca
liphi
lus
AT
CC
511
99 (
AY
0141
77)
T
Par
acoc
cus
sp. R
-246
16 (
AM
0840
41)
Par
acoc
cus
sp.
R-2
7049
(A
M08
4080
)
Par
acoc
cus
sp.
R-2
4665
(A
M08
4107
)
100
83
88
82
91
72
95
85
75 74 95
95
100
62
Dou
btfu
l dot
-blo
t res
ult
norB
det
ecti
on
Pos
itive
dot
-blo
t res
ult
norb
am
plic
on
Pos
itive
dot
-blo
t res
ult
norb
am
plic
on
Dou
btfu
l dot
-blo
t res
ult
Pos
itive
dot
-blo
t res
ult
Pos
itive
dot
-blo
t res
ult
norb
am
plic
on
Pos
itive
dot
-blo
t res
ult
Pos
itive
dot
-blo
t res
ult
Dou
btfu
l dot
-blo
t res
ult
CHAPTER 4
133
Absence of norB genes in functional denitrifiers
Unsuitable norB amplification conditions could only explain the undetectable genes for
nitric oxide reductase in less than a third of the studied denitrifiers. Surprisingly, more than
half of the tested denitrifying pure cultures (47 or 58.75%) did not give any reaction with
PCR or dot-blot, indicating the absence of known norB genes. These negative results were
mostly found in Bacillus and Comamonas. However, also Ochrobactrum and Sinorhizobium
strains gave negative results, genera of which other members gave positive results and were
therefore used for probe production. Again, these observations support high inter- and
intraspecies sequence divergence, possibly caused by HGT.
The absence of a norB gene in bacilli could be explained by the presence of qCuANOR. This
nitric oxide reductase was discovered in Bacillus azotoformans, thus far the only species in
which the enzyme is described [22]. Its occurence in other Bacillus strains is plausible, due
to its membrane-bound nature and the limited available periplasmic-like space. However,
confirmation of this hypothesis will only be possible with the discovery of the encoding
genes. Unfortunately, the denitrification process and the relevant enzymes in Gram-positive
bacteria are not well studied [5, 20], notwithstanding their great importance in the environment
[7, 16] and food production [4, 24].
The absence of norB genes in twelve out of fifteen strains (here the doubtful results is
conservatively considered positive) assigned to the well-known Gram-negative Comamonas
genus was unexpected, because of its well-known denitrifying capacity in wastewater
treatment [6, 8]. Surprisingly, the public sequence databases contain only one sequence for
a nitric oxide reductase from a member of this genus, Comamonas sp. R-28235, as published
previously by the authors [11]. The Comamonas isolates with a negative or doubtful dot-
blot result, and with an norB amplicon are all very highly related, showing more than 99%
16S rRNA gene sequence similarity to Comamonas denitrificans LMG 21602T (data not
shown). Thus, the norB detection results are not correlated with their 16S rRNA gene
sequence phylogeny. The qCuANOR could also be present in these denitrifiers, but its
advantages for denitrifying bacilli are unlikely to hold for Comamonas strains. These results
suggest the existence of one or more other undiscovered nitric oxide reductases, making the
denitrification process even more redundant than previously thought.
First of all, the genes encoding qCuANOR are necessary to develop molecular tools for its
detection and the assessment of its prevalence. Second, also more biochemical research and
whole genome projects are needed to confirm the presence of other unknown nitric oxide
reductases in pure culture denitrifiers. This search can be expedited by the availability of
the here presented pure culture denitrifiers potentially harbouring qCuANOR or other new
SCREENING FOR DENITRIFICATION GENES UNDETECTABLE WITH PCR
134
nitric oxide reductases. Tools for their detection are imperative, as the denitrification process
is monitored cultivation-independently through the study of its functional genes [3, 13,
15]. It could well be that an important subset of the denitrifying guild, containing qCuANOR
or other nitric oxide reductases, is now completely ignored.
Conclusions
A dot-blot protocol for the detection of norB genes in pure culture denitrifiers was developed.
Results showed both inadequate broad-range norB primers/PCR protocol and possible
large-scale occurrence of unknown enzymatic redundancy for nitric oxide reduction. The
developed screening method can be used when the available norB PCRs fail and may
provide insight into the prevalence of known norB genes in different bacterial taxa. The
norB detection results could not be correlated to the taxonomic affiliation of the denitrifiers,
and again demonstrate the strain-dependent nature of the denitrification trait. Hopefully,
this study is a step towards further in-depth investigation of the denitrification process in
diverse pure cultures, to complement the current knowledge from reference denitrifiers, for
which relevant biological material is provided here.
CHAPTER 4
135
REFERENCES1. Braker G, Tiedje JM (2003) Nitric oxide reductase (norB) genes from pure culture and environmental
samples. Appl Environ Microbiol 69:3476-3483
2. Cramm P, Pohlmann A, Friedrich B (1999) Purification and characterization of the single component
nitric oxide reductase from Ralstonia eutropha H16. FEBS Lett 460:6-10
3. Dandie CE, Miller MN, Burton DL, Zebarth BJ, Trevors JT, Goyer C (2007) Nitric oxide reductase-
targeted real-time PCR quantification of denitrifier populations in soil. Appl Environ Microbiol
73:4250-4258
4. De Clerck E, Vanhoutte T, Hebb T, Geerinck J, Devos J, De Vos P (2004) Isolation, characterization and
identification of bacterial contaminants in semifinal gelatine extracts. Appl Environ Microbiol 70:
3664-3672
5. Denariaz G, Payne WJ, LeGall J (1991) The denitrifying nitrite reductase of Bacillus halodenitrificans.
were performed within this cluster, using a modification of the microplate method of Ezaki
et al. [6] as described by Willems et al. [23]. A hybridization temperature of 45°C (calculated
with correction for the presence of 50% formamide) was used. R-32768T and R-32729T
showed DNA-DNA relatedness of 44.2% (± 2.8) with each other. R-32768T showed a
DNA-DNA relatedness of 41.3% (± 7) and 35.8% (± 4.7) with S. nitritireducens LMG
22074T and S. acidaminiphila LMG 22073T, respectively. R-32729T showed a DNA-DNA
relatedness of 37.2% (± 6.6) and 38.1% (± 5.1) with S. nitritireducens LMG 22074T and S.
acidaminiphila LMG 22073T, respectively. These results confirmed that R-32729T and R-
32768T belong to two novel genotypic species.
Figure 5.1. Unrooted phylogenetic dendrogram of the 16S rRNA gene sequences (1462 bp) shows the position ofR-32729T and R-32768T among the type strains of Stenotrophomonas species. Neighbour joining analysis wasperformed without corrections. Relevant bootstrap values (expressed as percentages of 1000 replicates) are shownat the branch points. EMBL accession numbers are shown in parenthesis.
Cell morphology, motility and possible sporulation was investigated by phase contrast
microscopy at a magnification of 1000x with cells grown on tryptone soya agar (TSA;
Description of novel bacterial species involved in the nitrogen cycle
146
Oxoid) for 48h at 28°C. Cells were Gram stained and examined for catalase and oxidase
activity. Utilization of carbon sources and enzyme production was tested with API 20NE
(48h, 28°C), API ZYM (4h, 28°C) and API 50CH (inoculated with AUX medium, 48h,
28°C) (BioMérieux) according to the manufacturer’s instruction. Strains R-32729T and R-
32768T were not identified as S. maltophilia with these conventional phenotypic taxonomic
tests. The temperature range (at 4, 15, 28, 37, and 52°C), the pH range (4.5 to 10.5 at 28°C)
and the salt range (0.5 to 5% at 28°C) for growth were recorded after incubation for 72h on
TSA. The ability to reduce nitrate was tested, as described by Smibert & Krieg [17], after
growth for two weeks on tryptone soya broth (TSB; Oxoid) supplemented with 10 mM
potassium nitrate at 37°C and on liquid G3M11 medium containing 18mM potassium
nitrate and 22.5mM ethanol [9] at 20°C - the solid variant of this medium was used as
isolation medium for the Stenotrophomonas strains. For all strains, these results were in
agreement with the nitrate reduction test in API 20NE. Lipolytic activity was tested by the
hydrolysis of Tween 80, as described by Sierra [16]. The phenotypic and biochemical
characteristics of all strains are given in Table 5.1.
Cells of all strains were identically incubated for exactly 24h at 28°C on TSA. A loopful of
cells was harvested, fatty acid methyl esters were prepared and extracted according to the
standardized protocol of the Microbial Identification System (MIS; Microbial ID Inc.).
The MIDI with the TSBA50 database was used for identification. All Stenotrophomonas
strains contained the characteristic fatty acids iso-C11:0
, isoC11:0
3OH and iso-C13:0
3OH [1,
26], and iso-C15:0
as dominant fatty acid. R-32768T and R-32729T showed highly similar
fatty acid profiles, only differing in the amount of specific fatty acids present, and mostly
containing iso-branched fatty acids. Characteristic fatty acids for R-32768T and R-32729T
were iso-C14:0
(14.2 & 15.7%), iso-C15:1
(4.6 & 2%), iso-C16:0
(8 &12.7%) and iso-C17:1
ω9c
(7.2 & 4.6%). Numerical analysis of the fatty acid profiles (Figure 5.2) showed that R-
32768T and R-32729T formed a distinct cluster, supported with a high cophenetic correlation,
and grouping closely with their phylogenetic nearest neighbours S. nitritireducens LMG
22074T and S. acidaminophila LMG 22073T. The MIDI fatty acid identification system
showed no relevant matches for strains R-32768T and R-32729T. However, the position of
these strains within the group Stenotrophomonas-Xanthomonas was further confirmed by
comparing the fatty acid profiles qualitatively and quantitatively with an in-house database
containing over 60000 fatty acid profiles. The failure of the MIDI identification system to
allocate these strains within the genus Stenotrophomonas was also observed for several
highly similar profiles of Xanthomonas members (unpublished data) and reported previously
for Stenotrophomonas acidaminiphila by Assih et al. [1].
CHAPTER 5
147
Table 5.1. Physiologic characteristics of the novel species and their closest phylogenetic neighbours.
Strains: 1, S. terrae R-32768T; 2, S. humi R-32729T; 3, S. nitritireducens LMG 22074T; 4, S. acidaminiphila LMG
22073T; 5, S. koreensis LMG 23369T; 6, S. maltophilia LMG 958T; 7, S. maltophilia LMG 22072; 8, S. rhizophila
CCUG 47042T. The type strain of S. dokdonensis is not included. Data are from this study unless indicated. +,
Positive; w, weakly positive; -, negative; v, variable.*, LMG 22073T reduces nitrate when grown on TSA supplemented with 10mM nitrate, but does not reduce nitrate
when grown on defined growth medium G3M11; §, LMG 22074T cannot denitrify starting from nitrate, only from
nitrite; †, data from [1]; ‡, data from [5]; #, data from [7]; $, data from [24]; °, data from [27].
All strains characterized in this study were positive for catalase, alkaline phosphatase, esterase, esterase lipase,
trypsin, acid phosphatase, glucose fermentation. All strains characterized in this study were negative for lipase,
DESCRIPTION OF NOVEL BACTERIAL SPECIES INVOLVED IN THE NITROGEN CYCLE
148
Coenye et al. [4] and Hauben et al. [8] suggested that, prior to their description, the
relationship of novel species within the genus Stenotrophomonas with the heterogeneous
species S. maltophilia should be investigated. GyrB-RFLP analysis of S. maltophilia strains
has proven to be in accordance with DNA-DNA hybridization results [4]. Therefore, the
gyrB-RFLP profiles of strains R-32768T and R-32729T were analyzed as described previously
[4], together with an additional set of Stenotrophomonas strains representing gyrB-RFLP
clusters A, B, C, E, F, G and I. Strains R-32768T and R-32729T did not render a gyrB
amplicon, even after several repeats. The same observation was made by Coenye et al. for
two Stenotrophomonas spp. strains (personal communication). It was concluded that this
approach was not suitable for the new strains, which, therefore, must differ from the
strains that did render gyrB amplicons.
Figure 5.2. Numerical comparison of the obtained fatty acid profiles. UPGMA clustering with Pearson’s correlationsimilarity coefficients was performed using BioNumerics version 4.6. The cophenetic correlation tool, whichdistinguishes reliable from unreliable subclusters, was used for cluster significance analysis.
To further substantiate that R-32729T and R-32768T do not belong to S. maltophilia, SDS-
PAGE analysis of whole-cell proteins was performed on all strains included in the gyrB-
RFLP trials. Aerobically grown cells were harvested after incubation at 28°C for 24h on
phosphate-buffered nutrient agar (pH 6.8). A SDS-PAGE banding pattern for all strains
was generated according to previously described standardized protocol [15]. Pearson’s
correlation similarity coefficients were clustered with UPGMA and analyzed with the
cophenetic correlation method in BioNumerics version 4.6 (Figure 5.3). The grouping of
the whole-cell proteins profiles was supported by high cophenetic correlation values but
did not correlate with the gyrB-RFLP grouping, except for gyrB-RFLP cluster G. R-
32729T and R-32768T did not group together. Their phylogenetically closest neighbors, S.
nitritireducens LMG 22074T and S. acidaminophila LMG 22073T, together with strain R-
12772, grouped separate from all other Stenotrophomonas strains.
Pearson correlation coefficient (% similarity)
100
90807060
100
100
88
100
79
85
98
.
.
.
.
.
.
.
.
Stenotrophomonas terrae sp. nov.
Stenotrophomonas acidaminiphila
Stenotrophomonas nitritireducens
Stenotrophomonas rhizophila
Stenotrophomonas maltophilia
Stenotrophomonas maltophilia
Stenotrophomons koreensis
R-32729T
R-32768T
LMG 22073T
LMG 22074T
CCUG 47042T
LMG 22072
LMG 958T
LMG 22369T
CHAPTER 5
149
Figure 5.3. Grouping of normalized digitized SDS-PAGE patterns. UPGMA clustering with Pearson’s correlationsimilarity coefficients was performed using BioNumerics version 4.6. The cophenetic correlation tool, whichdistinguishes reliable from unreliable subclusters, was used for cluster significance analysis. The identification ofnon-type strains and the assignment to gyrB-RFLP clusters were taken from [4].
DESCRIPTION OF NOVEL BACTERIAL SPECIES INVOLVED IN THE NITROGEN CYCLE
150
Based on the polyphasic data presented, the strains R-32768T and R-32729T represent two
novel species in the genus Stenotrophomonas, for which the names Stenotrophomonas
terrae sp. nov. and Stenotrophomonas humi sp. nov. are proposed.
Description of Stenotrophomonas terrae sp. nov.
Stenotrophomonas terrae (ter.ra’e. L. gen. n. terrae, of/from soil)
After 24h incubation at 28°C on TSA, colonies are irregular shaped and light yellow. Cells
are motile, non-spore-forming, rod-shaped, Gram-negative, catalase-positive, oxidase-
positive. Growth is observed at 15-37°C but not at 4°C or 52°C, at a pH of 5-10.5, but not
at 4.5, and at salt concentration of 0.5% to 5%. Anaerobic growth is possible through
nitrate reduction. Enzyme activities and carbon utilization are given in Table 1. Can be
differentiated from the type strains of its closest phylogenetic neighbours, S. humi, S.
nitritireducens and S. acidaminiphila by SDS-PAGE analysis and the presence of leucine
arylamidase, protease, the assimilation of glucose and the absence of N-acetyl-β-
glucosaminidase.
The type strain is R-32768T (= LMG 23958T = DSM 18941T), which has a DNA G+C
content of 65 mol% and was isolated from soil from a university test field in Ghent,
Belgium.
The Genbank/EMBL/DDBJ accession number for the 16S rRNA gene sequence of R-
32768T is AM403589.
Description of Stenotrophomonas humi sp. nov.
Stenotrophomonas humi (hu’mi. L. gen. n. humi, of/from soil)
After 24h incubation at 28°C on TSA, colonies are round, smooth and beige. Cells are
Growth is observed at 15-37°C but not at 4°C or 52°C, at a pH of 5-10.5, but not at 4.5,
and at salt concentration of 0.5% to 4% but not at 5%. Anaerobic growth is possible
through nitrate reduction. Enzyme activities and carbon utilization are given in Table 1. Can
be differentiated from the type strains of its closest phylogenetic neighbour, S. terra, S.
nitritireducens and S. acidaminiphila by SDS-PAGE analysis and the assimilation of malate.
The type strain is R-32729T (= LMG 23959T = DSM 18929T), which has a DNA G+C
content of 64 mol% and was isolated from soil from a university test field in Ghent,
Belgium.
The Genbank/EMBL/DDBJ accession number for the 16S rRNA gene sequence of R-
32729T is AM403587.
CHAPTER 5
151
REFERENCES1. Assih EA, Ouattara AS, Thierry S, Cayol J-L, Labat M, Macarie H (2002) Stenotrophomonas
acidaminiphila sp. nov., a strictly aerobic bacterium isolated from an upflow anaerobic sludge blanket
(UASB) reactor. Int J Syst Evol Microbiol 52:559-568
2. Coenye T, Spilker T, Martin A, LiPuma JJ (2002) Comparative assessment of genotyping methods for
epidemiologic study of Burkholderia cepacia Genomovar III. J Clin Microbiol 40:3300-3307
3. Coenye T, Vanlaere E, Falsen E, Vandamme P (2004) Stenotrophomonas africana Drancourt et al.
1997 is a later synonym of Stenotrophomonas maltophilia (Hugh 1981) Palleroni and Bradbury
1993. Int J Syst Evol Microbiol 54:1235-1237
4. Coenye T, Vanlaere E, LiPuma JJ, Vandamme P (2004) Identification of genomic groups in the genus
Stenotrophomonas using gyrB RFLP analysis. FEMS Immun Med Microbiol 40:181-185
5. Drancourt M, Raoult D (1997) Stenotrophomonas africana sp. nov., an opportunistic human pathogen
in Africa. Int J Syst Bacteriol 47:160-163
6. Ezaki T, Hashimoto Y, Yabuuchi E (1989) Fluorometric deoxyribonucleic acid-deoxyribonucleic acid
hybridization in microdilution wells as an alternative to membrane filter hybridization in which
radioisotopes are used to determine genetic relatedness among bacterial strains. Int J Syst Bacteriol
39:224-229
7. Finkman W, Altendorf K, Stackebrandt E, Lipski A (2000) Characterization of N2O-producing
Xanthomonas-like isolates from biofilters as Stenotrophomonas nitritireducens sp. nov, Luteimonas
mephitis gen. nov., sp. nov. and Pseudoxanthomonas broegbernensis gen. nov., sp. nov. Int J Syst
Evol Microbiol 50:273-282
8. Hauben L, Vauterin L, Moore ERB, Hoste B, Swings J (1999) Genomic diversity of the genus
Stenotrophomonas. Int J Syst Bacteriol 49:1749-1760
9. Heylen K, Vanparys B, Wittebolle L, Verstraete W, Boon N, De Vos P (2006) Cultivation of denitrifying
bacteria: optimisation of isolation conditions and diversity study. Appl Environ Microbiol 72:2637-
2643
10. Heyrman J, Verbeeren J, Schumann P, Swings J, De Vos P (2005) Six novel Arthrobacter species isolated
from deteriorated mural paintings. Int J Syst Evol Microbiol 55:1457-1464
11. Jukes TH, Cantor CR (1969) Evolution of protein molecules, In: Muruo HN (ed.) Mammalian Protein
Metabolism, New York : Academic Press, pp 21-132
12. Kimura M (1980) A simple method for estimating evolutionary rates of base substitutions through
comparative studies of nucleotide sequences. J Mol Evol 16:111-120
13. Mesbah M, Premachandran U, Whitman WB (1989) Precise measurement of the G + C content of
deoxyribonucleic acid by high-performance liquid chromatography. Int J Syst Bacteriol 39:159-167
14. Palleroni NJ, Bradbury JF (1993) Stenotrophomonas, a new bacterial genus for Xanthomonas
maltophilia (Hugh 1980) Swings et al., 1983. Int J Syst Bacteriol 43:606-609
15. Pot B, Vandamme P, Kersters K (1994) Analysis of electrophoretic whole organism protein fingerprints,
In: Goodfellow M, O’Donnel AG (Eds.), Chemical Methods in Prokaryotic Systematics, Chichester:
Wiley, pp. 493-521
16. Sierra G (1957) A simple method for the detection of lipolytic activity of micro-organisms and some
observations on the influence of the contact between cells and fatty substrates. Antonie van
Leeuwenhoek 23:15-22
17. Smibert RM, Krieg NR (1994) Phenotypic characterization. In: Gerhardt P, Murray RGE, Wood WA,
Krieg NR (Eds.), Methods for General and Molecular Bacteriology, Washington: American Society for
Microbiology, pp. 623-624
18. Stanier RY, Palleroni NJ, Doudoroff M (1966) The aerobic pseudomonads : a taxonomic study. J Gen
Microbiol 43:159-271
19. Swings J, De Vos P, Van den Mooter M, De Ley J (1983) Transfer of Pseudomonas maltophilia Hugh
1981 to the genus Xanthomonas as Xanthomonas maltophilia (Hugh 1981) comb. nov. Int J Syst
Bacteriol 40:348-369
20. Thompson JD, Gibson TJ, Plewniak F, Jeanmougin F, Higgins DG (1997) The ClustalX windows
interface: flexible strategies for multiple sequence alignment aided by quality analysis tools. Nucleic
Acids Res 24:4876-4882
21. Van de Peer Y., De Wachter R (1994) TREECON for Windows: a software package for the construction
and drawing of evolutionary trees for the Microsoft Windows environment. Comput Applic Biosci
10:569-570
DESCRIPTION OF NOVEL BACTERIAL SPECIES INVOLVED IN THE NITROGEN CYCLE
152
22. Vanparys B, Heylen K, Lebbe L, De Vos P (2005) Pedobacter caeni sp. nov., a novel species isolated
from a nitrifying inoculum. Int J Syst Evol Microbiol 55:1315-1318
23. Willems A, Doignon-Bourcier F, Goris J, Coopman R, de Lajudie P, De Vos P, Gillis M (2001) DNA-
DNA hybridization study of Bradyrhizobium strains. Int J Syst Evol Microbiol 51:1315-1322
24. Wolf A, Fritze A, Hagemann M, Berg G (2002) Stenotrophomonas rhizophila sp. nov., a novel plant-
associated bacterium with antifungal properties. Int J Syst Evol Microbiol 52:1937-1944
25. Yang H-C, Im W-T, Kang MS, Shin D-Y, Lee S-T (2006) Stenotrophomonas koreensis sp. nov., isolated
from compost in South Korea. Int J Syst Evol Microbiol 56:81-84
26. Yang P, Vauterin L, Vancanneyt M, Swings J, Kersters K (1993) Application of fatty acid methyl esters
for the taxonomic analysis of the genus Xanthomonas. Syst Appl Microbiol 16:47-71
27. Yoon J-H, Kang S-J, Oh HW, Oh T-K (2006) Stenotrophomonas dokdonensis sp. nov., isolated from
soil. Int J Syst Evol Microbiol 56:1363-1367
CHAPTER 5
5.2 ACIDOVORAX CAENI SP. NOV., A NOVELDENITRIFYING SPECIES WITH GENETICALLYDIVERSE ISOLATES FROM ACTIVATED SLUDGE
Redrafted from: Heylen K, Lebbe L, De Vos P (2007) Acidovorax caeni sp. nov., a novel
denitrifying species with genetically diverse isolates from activated sludge. Int J Syst Evol
Microbiol, Accepted
SUMMARY
Four Gram-negative, rod-shaped, non-spore-forming, denitrifying isolates were obtained
from the activated sludge of an aerobic-anaerobic wastewater treatment plant in Belgium.
Analysis of repetitive sequence-based PCR showed that the four isolates were genetically
different. 16S rRNA gene sequence analysis and DNA-DNA hybridization placed them
within the genus Acidovorax and showed their genotypic differentiation from all other
currently known Acidovorax species. Analysis of the whole cell proteins, and physiological
and biochemical tests allowed differentiation from their closest phylogenetic neighbours
and they therefore represent a novel species, for which the name Acidovorax caeni sp. nov.
is proposed, with strains R-24608T (= LMG 24103T = DSM 19327T) as type strain.
156
The genus Acidovorax contains eight species, which can be separated into soil and water
inhabitants, A. facilis, A. delafieldii, A. temperans [22] and A. defluvii [15], and the
phytopathogenic species A. avenae with its three subspecies, A. konjaci [24], A. anthurii
[4] and A. valerianell1a [5]. This separation in occurrence and habitat is reflected in the 16S
rRNA gene sequence phylogeny of these organisms, although separate phylogenetic clustering
of the genus within the Comamonadaceae is confirmed [23].
A previous cultivation-dependent study on activated sludge from an aerobic-anaerobic
wastewater treatment plant used different defined growth media for elective isolation of
denitrifying bacteria [7]. Nineteen denitrifiers were assigned to the genus Acidovorax,
based on partial 16S rRNA gene sequence analysis. The role of members of the
Comamonadaceae in the removal of nitrogen in wastewater treatment plants has been
recognized and described previously [2, 6, 9, 12]. Four Acidovorax isolates, R-24607, R-
24608T, R-24613, R-24614, were retrieved from G1M1, a mineral medium containing 15mM
sodium succinate, 3mM potassium nitrite and different vitamins. These isolates could
possibly belong to a new species within the environmental cluster of Acidovorax, based on
partial 16S rRNA gene sequence similarities and were analyzed further in a polyphasic
study. The type strain and a second representative of each described environmental
Acidovorax species, i.e. A. facilis, A. delafieldii, A. defluvii and A. temperans, were re-
examined for phenotyping, chemotaxonomy and biochemical analysis to guarantee
comparable results.
To avoid studying duplicate isolates of the same strain, genotyping by Random Amplified
Polymorphism DNA PCR analysis [1] and repetitive sequence-based PCR analysis with
REP and BOX primers [8] was carried out. The three fingerprint methods generated different
patterns for isolates R-24607, 24608T, R-24613, and R-24614, indicating genetic differences
between all four isolates (data not shown). The average DNA G+C content of the four
isolates, determined once by HPLC [13] was 64.3 mol% (± 0.8). The nearly complete 16S
rRNA gene sequences of R-24607, 24608T, R-24613, and R-24614 were determined as
described previously [20]. Phylogenetic analysis was performed using TreeCon [19] and
the BioNumerics software version 4.6 after multiple alignment with ClustalX [18]. Cluster
analysis using the neighbour-joining algorithm with or without corrections for evolutionary
distances as described by Jukes & Cantor [10] and Kimura [11] were in agreement with
maximum-parsimony and maximum-likelihood methods. Strains R-24607, 24608T, R-24613,
and R-24614 clustered together with A. temperans LMG 7169T, A. delafieldii LMG 5943T,
CHAPTER 5
157
A. defluvii DSM 12644T and A. facilis LMG 2193T, but clearly formed a separate group,
supported by high bootstrap values (Figure 5.4). Therefore, DNA-DNA hybridization
experiments were performed within this cluster, using a modification of the microplate
method of Ezaki et al. [3] as described by Willems et al. [21]. A hybridization temperature
of 45°C (calculated with correction for 50% formamide) was used. First, R-24607, R-
24608T, R-24613, and R-24614 were hybridized among themselves, to substantiate the
hypothesis of their relatedness on species level. The DNA-DNA similarity ranged between
78.5-88.5% (± 1.3-10.9), thus forming one species, but also confirming the genetic diversity
within the strains. R-24608T was further hybridized with A. temperans LMG 7169T (26.7%
± 5.6), A. delafieldii LMG 5943T (23.9% ± 5.9), A. defluvii DSM 12644T (26.0% ± 0.6) and
A. facilis LMG 2193T (18.2% ± 4.8). These results confirmed that R-24607, R-24608T, R-
24613, R-24614 belong to a novel genospecies.
Figure 5.4. Phylogenetic dendrogram obtained by neighbour-joining clustering of the 16S rRNA gene sequences(without correction), showing the position of the Acidovorax isolates R-24608T, R-24607, R-24613, R-24614among the type strains of Acidovorax species. EMBL accession numbers are shown in parenthesis. Variovoraxparadoxus LMG 1797T was used as outgroup. Relevant bootstrap values (expressed as percentages of 1000replicates) are shown at the branch points. The scale bar represents 0.02 changes per sequence position.
DESCRIPTION OF NOVEL BACTERIAL SPECIES INVOLVED IN THE NITROGEN CYCLE
Acidovorax caen i sp. nov. R-24614 ( 084008)AM
Acidovorax dela fieldii LMG 5943 ( 078764)T AF
Acidovorax avenae citru lli subsp. LMG 5376 ( 078761) T AF
Acidovorax caeni sp. nov. R-24608 ( 084006)T AM
Acidovorax caeni sp. nov. R-24607 ( 084011)AM
Acidovorax caeni sp. nov. R-24613 ( 084007)AM
Acidovorax defluvii DSM 12644 (Y18616) T
Acidovorax avenae cattleyae subsp. LMG 5286 ( 078762)T AF
Acidovorax avenae venae subsp. a LMG 2117 ( 078759)T AF
0.02 subst./site
Variovorax paradoxus LMG 1797 (D30793)T
Acidovorax valerianellae CFBP 4730 ( )T AJ431731
Acidovorax temperans LMG 7169 ( 078766) T AF
Acidovorax facilis LMG 2193 ( 078765) T AF
Acidovorax anthurii CIP 107058 ( 007013)T AJ
Acidovorax konjaci LMG 5691 ( 078760)T AF100
89
53
100
100
86
73
60
57
158
Cell morphology and motility was investigated by electron microscopy (Figure 5.5.) and
phase contrast microscopy (at a magnification of 1000x) respectively, for cells grown on
Tryptone Soy Agar (TSA; Oxoid) for 48h at 28°C. Cells were Gram stained and examined
with light microscopy, and catalase and oxidase activity was determined. Utilization of
carbon sources and enzyme production was tested with API 20NE (48h, 28°C), API ZYM
(4h, 28°C) (BioMérieux) and BIOLOG (24h, 28°C) according to the manufacturer’s
instruction. The temperature range (at 4, 15, 28, 37, 45 and 52°C), the pH range (4.5 to 10.5
at 28°C) and the salt range (0.5 to 5% NaCl w/v at 28°C) for growth were recorded after
incubation for 48h in Tryptone Soy Broth (TSB; Oxoid). The ability to denitrify was
tested, as described by Smibert & Krieg [17], after growth for one week on TSB supplemented
with 10 mM potassium nitrate at 37°C and on liquid isolation medium G1M1 at 37°C, and
confirmed with N2O measurements, as described by Heylen et al. [7]. Lipolytic activity
was determined after 72h by the hydrolysis of Tween 80, as described by Sierra [16].
Biochemical characteristics of all strains are given in Table 5.2.
Figure 5.5. Electron microscopic picture of R-24608T, showing peritrichous rods (about 0.9 x 1.8µm).
CHAPTER 5
159
After a pre-culture, all strains were identically incubated for exactly 48h at 28°C on
TSA. A loopful of well-grown cells was harvested and fatty acid methyl esters were
prepared and extracted according to the standardized protocol of the Microbial
Identification System (MIS; Microbial ID Inc.), and identified using MIDI with the
TSBA database version 5.0. All Acidovorax strains contained the characteristic fatty
acids 3-hydroxyoctanoic acid (C8:0
3OH) and 3-hydroxydecanoic acid (C10:0
3OH) [23].
The dominant fatty acids for strains R-24607, R-24608T, R-24613 and R-24614 were
Summed Feature 3 (38-41%), C18:1
ω7c (22-32%) and C16:0
(25-26.5%). Unfortunately,
the Sherlock MIS software could not clearly resolve Summed Feature 3, referring to the
peaks of C16:1
ω7c and/or isoC15:0
2OH. However, Sherlock lists the closest to the
observed ECL first, which was C16:1
ω7c. And also, comparison fatty acid data in
literature of the already described type strains [15, 22] and our data on the same strains
suggests C16:1
ω7c as the major fatty acid for this peak. No characteristic fatty acids for
the new genospecies were detected (Table 5.3).
SDS-PAGE analysis of whole-cell proteins was performed on aerobically grown cells after
incubation at 28°C for 40h on phosphate-buffered Nutrient Agar (pH 6.8). A SDS-PAGE
banding pattern for all strains was generated according to previously described standardized
protocol [14]. Pearson’s correlation similarity coefficients were clustered with UPGMA
and analyzed with the cophenetic correlation method in BioNumerics version 4.6 (Figure
5.6). The different strains of all species grouped together, supported with high cophenetic
correlation values. Although strains R-24607, R-24608T, R-24613 and R-24614 of the new
genospecies demonstrated significant variation in the whole protein profiles, they formed a
distinct group, separate from the other Acidovorax species.
All four strains were able to denitrify. Their denitrification genes were investigated with PCR
for nir and norB genes and with a dot-blot protocol for norB. R-24613 and R-24614, did
render a nirK sequence (see Section 3.1.). R-24608 and R-24607 did not render any sequence
for all four tested denitrification genes. Only R-24613 and R-24608 were included in the norB
dot-blot screening, including one strain with a nirK sequence and one without any detected
denitrification gene. Both tested positive for the presence of either cnorB or qnorB. Thus, the
nir genes show high sequence divergent within this species, resulting in strain-dependent
detection of nirK. Probably all four strains have a cnorB or qnorB gene that can only be
detected via dot-blot. The results for the functional gene analysis within this novel Acidovorax
species substantiate previous conclusions on strain-dependent detection of denitrification
genes.
DESCRIPTION OF NOVEL BACTERIAL SPECIES INVOLVED IN THE NITROGEN CYCLE
160
Table 5.2. (Opposite page) Physiological characteristics of the novel species and his closest phylogeneticneighbours.Strains: 1, A. caeni sp. nov. R-24607, R-24608T, R-24613, R-24614; 2, A. defluvii DSM 12644T, DSM 12578; 3,A. delafieldii LMG 5943T, LMG 8909; 4, A. facilis LMG 2193T, LMG 6598; 5, A. temperans LMG 7169T, LMG7163. Data are from this study unless indicated. +, Positive; w, weakly positive; -, negative; v, variable with thetest result for the type strain given between parenthesis. #, In contrast to DSM 12578, DSM 12644T does not reducenitrate to nitrite or further (denitrification) in the tested growth conditions (in G1M1 and in supplemented TSA).With the API 20NE gallery, both strains tested positive for nitrate reduction and denitrification; |, DSM 12644Tscored negative for Tween 80 hydrolysis in the Biolog screening (after 24h), but positive when grown on Sierramedium for 72h; °, data from [23]; *, data from [15].All Acidovorax species are positive for assimilation of leucine arylamidase, methyl pyruvate, mono-methyl succinate,β-hydroxybutyric acid, α-keto valeric acid, and D,L lactic acid. All Acidovorax species are negative for argininedihydrolase, urease, lipase (C14), valine arylamidase, cystine arylamidase, α-chymotrypsin, acid phosphatase, α-galactosidase, β-galactosidase, β-glucuronidase, α-glucosidase, β-glucosidase (aesculin hydrolysis), N-acetyl-D-glucosamidase, α-mannosidase, α-fucosidase, indole production, glucose fermentation, and assimilation of L-mannose, phenylacetic acid, N-acetyl-D-galactosamine, adonitol, cellobiose, L-fucose, gentiobiose, m-inositol,α-D-lactose, lactulose, D-melibiose, β-methyl-glucoside, D-raffinose, sucrose, D-trehalose, turanose, xylitol, citricacid, D-galactonic acid lactone, D-galacturonic acid, D-glucosaminic acid, D-glucuronic acid, saccharic acid,glucuronamide, L-histidine, inosine, uridine, thymidine, phenyl ethylamine, putrescine, 2-amino ethanol, 2,3-butanediol, D,L-α-glycerol phosphate, glucose-1-phosphate.
CHAPTER 5
161
DESCRIPTION OF NOVEL BACTERIAL SPECIES INVOLVED IN THE NITROGEN CYCLE
Characteristic 1 2 3 4 5 oxidase + + + + + catalase + v (+) -, v° + + hydrolysis of tween 40 + v (-) + v (-) + hydrolysis of tween 80 + +| + v (-) + nitrate reduction + v#, +* + + + denitrification + v# - - + alkaline phosphatase w - + w + esterase (C4) w + - + w esterase lipase (C8) + w + + + trypsin + - - - - naphthol-AS-BI-phosphohydrolase + - - - - Indole production + - - - - gelatinase - - -, v° v (-), +° - assimilation of glucose - - + -, +° + L-arabinose - - + -, +° - D-mannose - - + -, +° - D-mannitol - - + -, +° + N-acetyl-glucosamine - - - + - maltose - - - + - potassium gluconate - - + -, +° - capric acid - - -, v° - -, v° adipic acid - + - - - malate + - v - + trisodium citrate - - -, v° - - α-cyclodextrin v (-) - v (-) - - dextrin - - v (-) - - glycogen + - - - +, -° D-arabitol - - + - v (-) i-eritritol v (-) - - - - D-fructose - - + v (-) + D-galactose - v (+) + - - α-D-glucose - - v (-) - + D-psicose - - v (-) v (-) + D-sorbitol - - + - + acetic acid v (+) - - - - cis-aconitic acid v (+) - v (+) - - formic acid + - v (+) - - D-gluconic acid - - + - - α -hydroxybutyric acid + v (-) + - + γ-hydroxybutyric acid + v (-) + - + p-hydroxy phenylacetic acid - - + - - itaconic acid - - - - - α -keto butyric acid + v (-) + v (-) + α -keto glutaric acid + v (+) + v (-) + malonic acid v (-) - - v (-) - propionic acid + - v (+) - v (-) quinic acid - - + - - sebacic acid v (+) + + v (-) + succinic acid + + + v (-) + bromo succinic acid + - v (+) v (-) + succinamic acid v (-) - v (+) v (-) + alaninamide v (-) v (+) + v (-) + D-alanine + - + v (-) + L-alanine + + + v (-) + L-alanyl-glycine + v (+) + + + L-asparagine v (-) - + - + L-aspartic acid + - + - + L-glutamic acid + - + - + glycyl-L-aspartic acid - - + v (-) - glycyl-L-glutamic acid - - + - v (+) hydroxy-L-proline - - + - - L-leucine + - + v (-) + L-ornithine - - v (+) - v (-) L-phenylalanine - - v (+) + + L-proline + - + + + L-pyroglutamic acid + - v (+) + + D-serine v (-) - - - - L-serine + - + -, +° - L-threonine + v (-) + - v (-) D,L-carnithine - - v (+) v (+) - γ-amino butyric acid - - + v (+) v (-) urocanic acid - - + - - glycerol + v (+) + - + glucose-6-phosphate v (-) - v (+) - -
162
Table 5.3. Fatty acid content of the novel species and his closest phylogenetic neighbours.Values are percentages of the total fatty acid content. Strains: 1, A. caeni sp. nov. R-24607, R-24608T, R-24613, R-24614; 2, A. defluvii DSM 12644T, DSM 12578; 3, A. delafieldii LMG 5943T, LMG 8909; 4, A. facilis LMG 2193T,LMG 6598; 5, A. temperans LMG 7169T, LMG 7163. Summed Feature 3 contains C16:1ω7c and/or iso-C15:0 2-OH.Summed feature 7 contains C
19:1 ω6c and C
19-cyclo. All data are obtained in this study. /, not present; tr, trace elements
< 1%
Figure 5.6. Grouping of normalized digitized SDS-PAGE patterns. UPGMA clustering with Pearson’s correlationsimilarity coefficients was performed using BioNumerics version 4.6. The cophenetic correlation tool was used forcluster significance analysis.
5. Mahne I, Tiedje J.M (1995) Criteria and methodology for identifying respiratory denitrifiers. Appl Environ
Microbiol 6: 1110-1115
6. McKillip JL (2000) Prevalence and expression of enetrotoxins in Bacillus cereus and other Bacillus spp.,
a literature review. Antonie van Leeuwenhoek 77:393-399
7. Meer RR, Baker J, Bodyfelt FW, Griffiths MW (1991) Psychrotrophic Bacillus spp in fluid milk-products –
a review. J Food Protec 54:969-979
8. Mevel G, Prieur D (2000) Heterotrophic nitrification by a thermophilic Bacillus species as influenced by
different culture conditions. Can J Microbiol 46:465-473
9. Park SJ, Yoon JC, Shin K-S, Kim EH, Yim S, Cho Y-J, Sung GM, Lee D-G, Kim SB, Lee D-U, Woo S-H,
Koopman B (2007) Dominance of endospore-forming bacteria on a rotating activated Bacillus contactor
biofilm for advanced wastewater treatment. J Microbiol 45:113-121
10. Suharti, Strampraad MJ, Schröder T, De Vries S (2001) A novel copper A containing menaquinol NO reductase
from Bacillus azotoformans. Biochemistry 40:2632-2639
11. Ternström A, Lindberg A-M, Molin G (1993) Classification of the spoilage flora of raw and pasteurized
bovine milk, with special reference to Pseudomonas and Bacillus. J Appl Bacteriol 75:25-34
12. Yan L, He YL, Kong HN, Tanaka S, Lin Y (2006) Isolation of a new heterotrophic nitrifying Bacillus sp.
strain. J Environ Biol 27:323-326
CHAPTER 6
174
ADDENDUM
Overview of all isolates and strains investigated in this thesis. Strains are sorted according to their taxonomic positionand strain number. Identification is based on cellular fatty acid analysis and 16S rRNA gene sequence analysis.x, gene detected; ?, doubtful detection
177
Identificatie Strain nr. origin
nirK nirS cnorB qnorB included norB
AlphaproteobacteriaBrucella sp. R-26895 activated sludge x xEnsifer sp. R-31757 soilEnsifer sp. R-32544 soil x x ?
Methylobacterium sp. R-25207 activated sludge x x
Ochrobactrum anthropi LMG 2136 reference strain x xOchrobactrum sp R-25018 activated sludge x xOchrobactrum sp R-28410 activated sludge x xOchrobactrum sp. R- 24468 activated sludge xOchrobactrum sp. R-24286 activated sludge x x xOchrobactrum sp. R-24289 activated sludge x xOchrobactrum sp. R-24290 activated sludge x xOchrobactrum sp. R-24291 activated sludge x xOchrobactrum sp. R-24334 activated sludge xOchrobactrum sp. R-24343 activated sludge x xOchrobactrum sp. R-24448 activated sludge x xOchrobactrum sp. R-24467 activated sludge xOchrobactrum sp. R-24618 activated sludge x xOchrobactrum sp. R-24638 activated sludge x xOchrobactrum sp. R-24639 activated sludge xOchrobactrum sp. R-24653 activated sludge x xOchrobactrum sp. R-25054 activated sludge x xOchrobactrum sp. R-25055 activated sludge x x ?Ochrobactrum sp. R-25203 activated sludge x xOchrobactrum sp. R-26465 activated sludge xOchrobactrum sp. R-26821 activated sludge x xOchrobactrum sp. R-26825 activated sludge x xOchrobactrum sp. R-26826 activated sludge x xOchrobactrum sp. R-26889 activated sludge x xOchrobactrum sp. R-26890 activated sludge x xOchrobactrum sp. R-26891 activated sludge x xOchrobactrum sp. R-26892 activated sludge x xOchrobactrum sp. R-26894 activated sludge x xOchrobactrum sp. R-26898 activated sludge x xOchrobactrum sp. R-26900 activated sludge x xOchrobactrum sp. R-27045 activated sludge x xOchrobactrum sp. R-27046 activated sludge x xPannonibacter sp. R-27042 activated sludge x x
Paracoccus denitrificans LMG 4049 reference strain x x
Paracoccus pantotrophus LMG 4218Treference strain
Paracoccus sp. R-24292 activated sludge xParacoccus sp. R-24342 activated sludge x x xParacoccus sp. R-24615 activated sludge xParacoccus sp. R-24616 activated sludge x x xParacoccus sp. R-24617 activated sludge xParacoccus sp. R-24621 activated sludgeParacoccus sp. R-24623 activated sludge x x xParacoccus sp. R-24649 activated sludge x x xParacoccus sp. R-24650 activated sludge x x xParacoccus sp. R-24652 activated sludge x x xParacoccus sp. R-24665 activated sludge x x ?Paracoccus sp. R-25049 activated sludgeParacoccus sp. R-25058 activated sludge xParacoccus sp. R-25059 activated sludgeParacoccus sp. R-26466 activated sludge xParacoccus sp. R-26819 activated sludgeParacoccus sp. R-26822 activated sludge x x xParacoccus sp. R-26823 activated sludge xParacoccus sp. R-26824 activated sludge x x ?Paracoccus sp. R-26839 activated sludgeParacoccus sp. R-26841 activated sludge xParacoccus sp. R-26844 activated sludgeParacoccus sp. R-26888 activated sludge x ?Paracoccus sp. R-26893 activated sludgeParacoccus sp. R-26896 activated sludge
Detection via PCR Detection via dot-blot
ADDENDUM
178
Identificatie Strain nr. origin
nirK nirS cnorB qnorB included norBParacoccus sp. R-26897 activated sludge x xParacoccus sp. R-26899 activated sludge xParacoccus sp. R-26901 activated sludge xParacoccus sp. R-26902 activated sludgeParacoccus sp. R-27041 activated sludge x xParacoccus sp. R-27043 activated sludge x ?Paracoccus sp. R-27047 activated sludge xParacoccus sp. R-27049 activated sludge x x xParacoccus sp. R-28237 activated sludgeParacoccus sp. R-28238 activated sludgeParacoccus sp. R-28239 activated sludgeParacoccus sp. R-28241 activated sludge x x xParacoccus sp. R-28242 activated sludge x x xParacoccus sp. R-28243 activated sludgeParacoccus sp. R-28244 activated sludge xParacoccus sp. R-28245 activated sludge xParacoccus sp. R-28294 activated sludgeParacoccus sp. R-28409 activated sludge x xParacoccus sp. R-28414 activated sludge xRhizobium sp . R-26835 activated sludge xRhizobium sp. R-24261 activated sludge xRhizobium sp. R-24333 activated sludge x xRhizobium sp. R-24654 activated sludge x xRhizobium sp. R-24658 activated sludge x xRhizobium sp. R-24663 activated sludge xRhizobium sp. R-26467 activated sludge x xRhizobium sp. R-26820 activated sludge x xRhizobium sp. R-31549 soil x x ?Rhizobium sp. R-31762 soilRhizobium sp. R-31837 soil xRhizobium sp. R-31857 soil xRhizobium sp. R-32539 soilRhizobium sp. R-32552 soilRhizobium sp. R-32725 soil x x x
Sinorhizobium sp R-24605 activated sludge xSinorhizobium sp R-25067 activated sludge x x xSinorhizobium sp R-25078 activated sludge x xSinorhizobium sp. R-31759 soil x xSinorhizobium sp. R-31763 soil xSinorhizobium sp. R-31764 soil x xSinorhizobium sp. R-31816 soil x x xSinorhizobium sp. R-32542 soil x x xSinorhizobium sp. R-32546 soil xSinorhizobium sp. R-32549 soil xSinorhizobium sp. R-32737 soil x x xSinorhizobium sp. R-32769 soil x
Betaproteobacteria
Achromobacter denitrificans LMG 1231Treference strain x x
Alcaligenes faecalis LMG 1229Treference strain x x
Virgibacillus halodenitrificans LMG 9818Treference strain x
BacteroidetesChryseobacterium sp. R-25053 activated sludge x
Detection via PCR
181
SUMMARY
Denitrification is a step-wise dissimilatory reduction of nitrate and/or nitrite over nitric
oxide and dinitric oxide, also named nitrous oxide, to nitrogen gas, coupled to electron
transport phosphorylation. This modular process is accomplished in four enzymatic steps,
catalyzed by four metalloproteins: nitrate reductase, nitrite reductase, nitric oxide reductase
and nitrous oxide reductase. The denitrification process is of major importance because it
causes fertilizer loss in agriculture, contributes to global warming through release of N2O,
and has an unclear role in bacterial pathogenicity, but it is also responsible for removal of
excess nitrogen from natural and wastewater systems, and can be useful for destruction of
other pollutants such as hydrocarbons.
With the advent of molecular techniques, focus of denitrification research shifted from
physiological and biochemical research on pure cultures to environmental monitoring of
the whole denitrification process in situ. However, the correlation between structural and
functional diversity remains a challenge, because denitrification is a phylogenetically
dispersed trait. Denitrification genes can be used as functional markers, but their taxonomic
information content is not clear, after several reports of differences between functional
gene phylogeny and organism phylogeny. Aim of this thesis was to investigate denitrification
starting from the functional bacterium, with focus on the genetic basis of the process. The
taxonomic value of the genes encoding the key enzymes was assessed, and their detection,
occurrence, prevalence and phylogeny were investigated.
Because most denitrification research was been performed on reference strains, new isolation
procedures from activated sludge and soil were performed. Therefore, sixty different defined
growth media were screened for their elective isolation of denitrifiers from activated sludge.
The influence of eleven medium parameters with different values and their combinations
on the number and the diversity of isolated denitrifiers were examined. All media were
investigated and scored for their selection of a high number and a high diversity of
denitrifying bacteria. The three best scoring media, G2M11, G3M12 and G4M3, together
with trypticase soy agar, were used to isolate denitrifiers from soil. From both inocula, a
higher cultivable denitrifying diversity was retrieved than previously reported in literature.
Also, both studies clearly demonstrated that use of a multiple-media set generated more
information than individual growth media.
The functional genes encoding the key enzymes nitrite and nitric oxide reductase, namely
nirS/nirK and cnorB/qnorB respectively, were investigated in a total number of 227 pure
culture denitrifiers from activated sludge and 112 pure culture denitrifiers from soil. The
183
functional genes were amplified in only a very low fraction of the pure cultures, 50% for
the nir genes and 30% for the norB genes in denitrifiers from activated sludge and even
less for both genes in cultures from soil. Both nir and norB genes showed high genetic
heterogeneity. Based on our data, Alphaproteobacteria mostly contained nirK and cnorB,
while nirS and both cnorB and qnorB were present in the Betaproteobacteria. Too few
representatives of the other bacterial classes were retrieved in the isolation studies to draw
conclusions on the functional gene prevalence.
For both gene types, overall gene phylogeny was not congruent with the widely accepted
organism phylogeny based on 16S rRNA gene sequence analysis. Tree topologies showed
that members of different bacterial classes grouped together. Nevertheless, dependent on
gene type and taxon, functional genes do contain limited taxonomic information. Grouping
based on nirS allowed rough delineation of bacterial taxa, while both norB genes did not
show any congruence with the organism phylogeny. The qnorB genes retrieved from several
denitrifying Bacillus isolates from soil grouped together with sequences from non-
denitrifiers, and clustered distant from qnorB sequences of other bacilli. This data indicates
that functional gene data cannot be generally used for deducing structural diversity
information in environmental studies. Also, comparative analysis of both nir and norB
genes showed that congruence between both gene types couldn’t be extrapolated to all
denitrifiers. Thus, linkage of both genes, either on the chromosome or on a plasmid, is not
a general feature in denitrifiers. Both gene types can be dispersed on the bacterial DNA,
and thus can cover independent evolutionary trajectories.
Next to the discussed incongruence between gene and organism phylogeny, some unexpected
observations were made. First, identical nirK and cnorB sequences were found in
representatives of different bacterial classes. Second, two isolates, Pseudomonas sp. R-
25208 and Bacillus sp. R-32694, contained both a cnorB and a qnorB gene. These are the
first reports of the presence of two different norB genes in one bacterium. Worth noting is
that the qnorB gene phylogeny matched the organism phylogeny in both organisms, while
the cnorB gene phylogeny did not. Third, the environment had a significant influence on
the nirK phylogeny, regardless of the taxonomical position of the denitrifiers. Fourth, the
detection of the functional genes within some taxa was strain-specific, suggesting significant
sequence divergence. All above-mentioned observations can be explained through horizontal
gene transfer (HGT) events of nir and norB between different taxa. However, these data
are only indirect evidence and further substantiation of the HGT theory is necessary.
Therefore, as a first step, we have been able to demonstrate that the cnorB and the nirK
SUMMARY
184
gene of the above-mentioned Pseudomonas sp. R-25208, which both did not match the
organism phylogeny, are located on a plasmid.
The functional genes were amplified in only a very low fraction of the pure cultures. A
possible explanation would be that applied primers and PCR protocols were developed
based on very limited sequences available at the time. Their use in a large strain set showed
their fallibility as broad-range amplification and detection tools. To assess whether the
very low detection percentage was only the result of non-optimal amplification, a dot-blot
method was developed to detect norB. Screening of 80 pure culture denitrifiers showed
that low norB amplification percentages could only be partially attributed to insufficient
primers and PCR protocols. More than half of the functional denitrifiers tested did not
give a positive result, which suggested the presence of unknown enzymatic redundancy
for nitric oxide reduction. Thus, denitrifiers can use enzymes other than those currently
known to reduce nitric oxide to nitrous oxide. More genome analysis and molecular work
is needed to identify those new genes/enzymes and to develop suitable detection tools.
The isolation campaign on activated sludge and soil rendered a large set of isolates,
containing bacteria possibly belonging to novel species or genera. Two new nitrate-reducing
Stenotrophomonas species and one new denitrifying Acidovorax species were described.
Although the denitrification trait is studied extensively, predominantly cultivation-
independently, little information was available on the ecology of the functional genes.
This study provides this much needed, scrutinized ecological information on the
denitrification genes, as well as new insights in the cultivation of denitrifiers, and can be
used as a starting point for further in-depth research on the denitrification process.
SUMMARY
185
SAMENVATTING
Denitrificatie is de stapsgewijze reductie van nitraat en/of nitriet over stikstofoxide en
distikstofoxide tot stikstofgas, gekoppeld aan electron transport fosforylatie. Dit modulair
proces wordt uitgevoerd door vier enzymatische stappen en gekatalyseerd door vier
metaalproteïnen, namelijk nitraatreductase, nitrietreductase, stikstofmonoxidereductase en
distikstofmonoxidereductase. Vanuit een antropogeen standpunt is denitrificatie
tegelijkertijd een negatief en positief proces. Het is namelijk verantwoordelijk voor verlies
van meststoffen in de landbouw, de productie van het broeikasgas N2O en heeft mogelijks
een rol in de bacteriële pathogeniciteit, maar kan ook zowel overtollig stikstof als andere
polluenten verwijderen uit natuurlijke en afvalwatersystemen.
Denitrificatie is sinds de ontdekking van het proces in 1882 reeds uitvoering onderzocht.
En met de ontwikkeling van moleculaire technieken is de focus van denitrificatieonderzoek
verschoven van fysiologisch en biochemisch onderzoek op reinculturen naar opvolging
van het volledige denitrificatieproces in situ. Desondanks blijft het correleren van structurele
en functionele diversiteit een grote uitdaging, meer bepaald door de wijdverspreidheid van
de denitrificatie capaciteit in de bacteriële wereld. Denitrificatiegenen zijn zeker nuttig als
functionele merkers, maar, na verscheidene berichten van incongruentie tussen functionele
gen fylogenie en organisme fylogenie, is het niveau van taxonomische informatie vervat in
deze genen onduidelijk. Daarom heeft deze thesis tot doel de genetische basis van
denitrificatie te onderzoeken, vertrekkend van de functionele bacteriële reincultuur. De
taxonomische waarde van de genen coderend voor de sleutelenzymen en hun detectie,
voorkomen, dominantie en fylogenie werden onderzocht.
Aangezien het meeste onderzoek op denitrificatie in het verleden is uitgevoerd op
referentiestammen, werden nieuwe cultivatiecampagnes opgestart met actief slib en bodem
als inoculum. Hiervoor werden zestig verschillende gedefinieerde groeimedia gescreend
voor de electieve isolatie van denitrificeerders uit actief slib. De invloed van elf
mediumparameters met verschillende waarden en hun combinaties werd nagegaan. Elk
medium kreeg een score op basis van het aantal denitrificeerders en hun diversiteit. In een
volgende studie werden de drie bestscorende media - G2M11, G3M12 en G4M3 – samen
met ‘trypticase soy agar’ gebruikt om denitrificerende bacteriën te isoleren uit bodem. De
denitrificerende isolaten omvatten een grotere diversiteit dan voordien vermeld in de
literatuur. Ook tonen beide studies duidelijk aan dat het gebruik van een multiple-media
set meer informatie genereert dan de applicatie van één medium.
De functionele genen coderend voor de sleutelenzymen nitrietreductase en
stikstofmonoxidereductase, namelijk nirS/nirK en cnorB/qnorB respectievelijk, werden
187
onderzocht in 227 denitrificerende reinculturen uit actief slib en 112 uit bodem. Deze
functionele genen konden slechts geamplificeerd worden in een lage fractie van de
stammenset, meer bepaald in 50% voor de nir genen en in 30% voor de norB genen in
denitrificeerders uit actief slib, en nog minder in reinculturen uit bodem. Beide gentypes
vertoonden hoge genetische heterogeniteit. Alphaproteobacteria bevatten vooral nirK en
cnorB, terwijl in Betaproteobacteria hoofdzakelijk nirS en zowel cnorB als qnorB aanwezig
waren. Onze stammenset bevatte te weinig representatieven van de andere bacteriële klassen
om conclusies te trekken rond het voorkomen van de functionele genen in deze taxa.
Voor beide gentypes was de totale gen fylogenie niet congruent met de algemeen aanvaarde
organisme fylogenie op basis van 16S rRNA gen sequentieanalyse. De clusteringen tonen
aan dat leden van verschillende bacteriële klassen samen groepeerden. Niettegenstaande
blijken de functionele genen wel beperkte taxonomische informatie te bevatten, afhankelijk
van het gen en het taxon. Groepering op basis van nirS laat toe om bacteriële taxa ruw af te
lijnen, terwijl beide norB genen geen enkele overeenstemming vertonen met de organisme
fylogenie. De qnorB genen van verscheidene denitrificerende Bacillus isolaten uit bodem
clusteren samen met sequenties van niet-denitrificeerders, en ver van de qnorB sequenties
van andere bacilli. Deze data tonen aan dat informatie rond functionele diversiteit uit
milieustudies niet algemeen kan gebruikt worden om de structurele diversiteit af te leiden.
Daarnaast blijkt uit vergelijkende analyses van nir en norB genen dat overeenstemming
tussen fylogenie van beide gentypes niet kan geëxtrapoleerd worden naar alle
denitrificeerders. Een koppeling tussen beide genen, of op het chromosoom of op een
plasmide, is dus geen algemeen kenmerk in denitrificeerders. Beide gentypes kunnen
verspreid voorkomen in het bacterieel DNA, waardoor ze onafhankelijke evolutionele
trajecten kunnen doorlopen.
Naast de bediscussieerde incongruentie tussen gen- en organisme fylogenie werden volgende
observaties gemaakt.
(i) Identieke nirK en cnorB sequenties werden teruggevonden in
representatieven van verschillende bacteriële klassen.
(ii) Twee isolaten, Pseudomonas sp. R-25208 en Bacillus sp. R-32694, bevatten
elk een cnorB en een qnorB gen. Dit is de eerste melding van een bacterie
met twee verschillende norB genen. Opvallend was dat de qnorB fylogenie
overeenstemde met de organisme fylogenie, wat niet het geval was voor
cnorB.
(iii) Het habitat had een significante invloed op de nirK gen fylogenie,
onafhankelijk van de taxonomische positie van de denitrificeerders.
SAMENVATTING
188
(iv) De detectie van de functionele genen in sommige taxa was stamafhankelijk,
wat een significante sequentiedivergentie suggereert.
Het frequent voorkomen van horizontale gen transfers (HGT) van functionele genen tussen
verschillende taxa zou bovenstaande observaties verklaren. Echter, deze data zijn slechts
een indicatie en verdere bewijzen voor de HGT theorie zijn noodzakelijk. In een eerste
stap naar een bewijs hebben we kunnen aantonen dat cnorB en nirK van de hierboven
vernoemde Pseudomonas sp. R-25208, waarvan de fylogenie niet overeenstemde met de
organisme fylogenie, gelegen zijn op een plasmide.
Zoals reeds eerder vermeld, werden de functionele genen slechts geamplificeerd in een
kleine fractie van de stammenset. Een mogelijke verklaring hiervoor zijn de gebruikte
primers en PCR protocols, ontwikkeld op basis van de enkele gensequenties toen
beschikbaar. Hun gebruik in een grote stammenset toont nu aan dat ze als amplificatie- en
detectiemechanismen een beperkt bereik hebben. Om te bepalen of de lage
detectiepercentages enkel het gevolg zijn van niet-optimale amplificatie werd een dot-blot
methode ontwikkeld om norB te detecteren. Een screening van 80 denitrificerende
reinculturen toonde aan dat de lage norB amplificatie percentages slechts partieel te wijten
zijn aan ontoereikende primers en PCR protocols. Meer dan de helft van de functionele
denitrificeerders gaven geen positief resultaat, wat wijst op de aanwezigheid van extra,
ongekende enzymen voor stikstofmonoxidereductie. Dus, denitrificeerders kunnen
mogelijks andere dan de nu gekende enzymen gebruiken voor de reductie van
stikstofmonoxide naar distikstofmonoxide. Meer genoomanalyses en moleculair werk is
noodzakelijk om deze nieuwe genen/enzymen te identificeren en geschikte
detectietechnieken te ontwikkelen.
De grote stammenset, gegenereerd door de isolatiecampagnes met actief slib en bodem als
inoculum, bevat bacteriën behorende tot potentieel nieuwe species of genera. Twee nieuwe,
nitraatreducerende Stenotrophomonas species en één nieuwe denitrificerende Acidovorax
species werden reeds beschreven.
Hoewel denitrificatie al uitgebreid bestudeerd werd, vooral dan cultuuronafhankelijk, is er
weinig informatie beschikbaar over de ecologie van de betrokken functionele genen. Deze
PhD studie verschaft de hoogstnoodzakelijke ecologische informatie over
denitrificatiegenen, alsook nieuwe inzichten over de cultivatie van denitrificeerders, en
kan gebruikt worden als startpunt voor verder diepgaand onderzoek van het
denitrificatieproces.
SAMENVATTING
189
CURRICULUM VITAE
PERSONALIAFull name: Kim HeylenAddress: Haanstraat 11, Gent, BelgiumDate of birth: December 29th 1979Place of birth: Borgerhout, Belgium
EDUCATIONAL BACKGROUND2004 – present: Ghent University, Gent, BelgiumPhD studies at the Laboratory of Microbiology, Department of Physiology, Biochemistryand Microbiology (WE10), Faculty of Sciences
2002-2003: Ghent University, Gent, BelgiumMaster of Science in ‘Applied Microbial Systematics’ (graduated magna cum laude)Dissertation: ‘Characterization of chloramphenicol-resistant heterotrophic bacteria fromSoutheast Asian aquaculture environment’
1997-2001: University of Antwerp, Antwerp, BelgiumLicentiate Biology, option physiology (graduated cum laude)Dissertation: ‘Removal of dyes from textile waste water: isolation of dye-degrading bacteriaand their microbial ecology’, in collaboration with VITO (Flemisch Institute ofTechnological Research)
WORK EXPERIENCEFebruary 2002 – June 2002Fulltime Highschool teacher Biology/Maths in second grade at College O-L-V. Ten Doorn,Eeklo, Belgium
SCIENTIFIC OUTPUTA1 PublicationsVanparys B, Heylen K, Lebbe L, De Vos P (2005) Pedobacter caeni sp. nov., a novelspecies isolated from a nitrifying inoculum. Int J Syst Evol Microbiol 55:1315-1318
Vanparys B, Heylen K, Lebbe L, De Vos P (2005) Devosia limi sp. nov., a novel speciesisolated from a nitrifying inoculum. Int J Syst Evol Microbiol 55:1997–2000
Wittebolle L, Boon N, Vanparys B, Heylen K, De Vos P, Verstraete W (2005) Failure ofthe ammonia oxidation process in two pharmaceutical wastewater treatment plantsis linkedto shifts in the bacterial communities. J Appl Microbiol 99:997-1006
Heylen K, Gevers D, Vanparys B, Wittebolle L, Geets J, Boon N, De Vos P (2006) Theincidence of nirS and nirK and their genetic heterogeneity in cultivated denitrifiers. EnvironMicrobiol 8:2012-2021
Heylen K, Vanparys B, Wittebolle L, Verstraete W, Boon N, De Vos P (2006) Cultivationof denitrifying bacteria: optimization of isolation conditions and diversity study. ApplEnviron Microbiol 72:2637-2643
191
Vanparys B, Heylen K, Lebbe L, De Vos P (2006) Pseudomonas peli sp. nov. andPseudomonas borbori sp. nov., isolated from a nitrifying inoculum. Int J Syst EvolMicrobiol 56:1875-1881
Heylen K, Vanparys B, Gevers D, Wittebolle L, Boon N, De Vos P. (2007) Nitric oxidereductase (norB) gene sequence analysis reveals discrepancies with nitrite reductase (nir)gene phylogeny in cultivated denitrifiers. Environ Microbiol 9:1072-1077
Vanparys B, Spieck E, Heylen K, Wittebolle L, Geets J, Boon N, De Vos P (2007) Thephylogeny of Nitrobacter based on comparative rep-PCR, 16S rRNA and nitriteoxidoreductase gene sequence analysis. Syst Appl Microbiol 30:297-308
Verstraete W, Wittebolle L, Heylen K, Vanparys B, De Vos P, Van de Wiele T, Boon N(2007) Microbial Resource Management: The Road to go for Environmental Biotechnology.Engineering in Life Sciences 7:117-126
Geets J, de Cooman M, Wittebolle L, Heylen K, Vanparys B, De Vos P, Verstraete W,Boon N (2007) Real-time PCR assay for the simultaneous quantification of nitrifying anddenitrifying bacteria in activated sludge. Appl Microbiol Biotechnol 75:211-221
Heylen K, Vanparys B, Peirsegaele F, Lebbe L, De Vos P (2007) Stenotrophomonas terrae,sp. nov. and Stenotrophomonas humi , sp. nov., two novel nitrate-reducingStenotrophomonas species isolated from soil. Int J Syst Evol Microbiol 57: 2056-2061
Heylen K, Lebbe L, De Vos P (2007) Acidovorax caeni sp. nov., a novel denitrifyingspecies with genetically diverse isolates from activated sludge. Int J Syst Evol Microbiol,Accepted.
Heylen K, Boon N, Verstraete W, De Vos P (2007) Functional gene study on heterotrophicsoil denitrifiers isolated through a cultivation strategy on soil. Microb Ecol, submitted.
Oral presentationsHeylen K, Vanparys B, Wittebolle L, Boon N, Verstraete W, De Vos P (2005) Diversitystudy of denitrifying bacteria in activated sludge: a culture-dependent approach. COSTAction 856 – Ecological aspects of denitrification, with emphasis on agriculture, April6th-10th, Legnaro (Padova), Italy.
Heylen K, Lebbe L, De Vos P (2007) New insights in the genetic basis of denitrification.BSM symposium – Evolution in the Microbial world, November 23rd, Brussels, Belgium.
Heylen K, Lebbe L, De Vos P (2007) New insights in the genetic basis of denitrification.COST Action 856 - Denitrification ans related aspects, December 5th - 8th, Uppsala,Sweden.
Poster presentationsHeylen K, Vanparys B, Boon N, Verstraete W, De Vos P (2004) Culture-dependent studyon the diversity of the denitrifying microbial community in activated sludge. BSMSymposium – Crossroads of microbiology and informatics, December 3rd, Brussels,Belgium.
Heylen K, Vanparys B, Wittebolle L, Boon N, Verstraete W, De Vos P (2005) Developmentof selective growth media for denitrifying bacteria using an evolutionary algorithm: astrategy outline. ASPD4 – Microbial population dynamics in biological wastewatertreatment, July 17th-20th, Gold Coast, Queensland, Australia.
192
193
Heylen K, Vanparys B, Wittebolle L, Verstraete W, Boon N, De Vos P (2005) Developmentof selective growth media for denitrifying bacteria using an evolutionary algorithm. BSMsymposium – Imaging technology in microbiology cytometric and molecular approaches,November 18th, Brussels, Belgium.
Heylen K, Gevers D, Vanparys B, Wittebolle L, Geets J, Boon N, De Vos P (2006) Theincidence of nirS and nirK and their genetic heterogeneity in cultivated denitrifiers. ISME-11 – The hidden powers – Microbial communities in action, August 20th-25th, Vienna,Austria.
Heylen K, Vanparys B, Boon N, Verstraete W, De Vos P (2006) Bacterial diversity in soilinvolved in the removal of nitrogen demonstrates a high dependence of nitrate and/ornitrite reduction on growth conditions. BSM symposium – Novel compounds and strategiesto combat pathogenic microorganisms, November 24th, Brussels, Belgium. Second nominee.
Heylen K, Boon N, Verstraete W, De Vos P (2007) Validation of previously developedgrowth media for isolation of the denitrifying bacterial diversity from soil.Doctoraatssymposium UGent, April 24th, Gent, Belgium. Second nominee.
Heylen K, De Vos P (2007) Development of a Detection Procedure for DenitrificationGenes Not Amplified by Currently Available PCR Approaches. The 107th General Meetingof the Amercian Society of Microbiology, May 21th - 25th , Toronto, Canada.