Fluorescent oligonucleotide rDNA probes for specific detection of methane oxidising bacteria

Fluorescent oligonucleotide rDNA probes for speci¢c detection ofmethane oxidising bacteria

David G. Bourne a, Andrew J. Holmes a;b, Niels Iversen c, J. Colin Murrell a;*a Department of Biological Sciences, University of Warwick, Coventry CV4 7AL, UK

b Commonwealth Key Centre for Biodiversity and Bioresources, School of Biological Sciences, Macquarie University, Sydney, N.S.W. 2109, Australiac Environmental Engineering Laboratory, Aalborg University, Sohngaardsholmsvej 57, DK-9000 Aalborg, Denmark

Received 21 May 1999; received in revised form 13 September 1999; accepted 24 September 1999

Abstract

Oligonucleotide probes targeting the 16S rRNA of distinct phylogenetic groups of methanotrophs were designed for the in situ detectionof these organisms. A probe, MG-64, detected specifically type I methanotrophs, while probes MA-221 and MA-621, detected type IImethanotrophs in whole cell hybridisations. A probe Mc1029 was also designed which targeted only organisms from the Methylococcusgenus after whole cell hybridisations. All probes were labelled with the fluorochrome Cy3 and optimum conditions for hybridisation weredetermined. Non-specific target sites of the type I (MG-64) and type II (MA-621) probes to non-methanotrophic organisms are highlighted.The probes are however used in studying enrichment cultures and environments where selective pressure favours the growth ofmethanotrophs over other organisms. The application of these probes was demonstrated in the detection of type I methanotrophs with theMG-64 probe in an enrichment culture from an estuarine sample demonstrating methane oxidation. The detection of type I methanotrophswas confirmed by a 16S rDNA molecular analysis of the estuarine enrichment culture which demonstrated that the most abundant bacterialclone type in the 16S rDNA library was most closely related to Methylobacter sp. strain BB5.1, a type I methanotroph also isolated from anestuarine environment. ß 2000 Federation of European Microbiological Societies. Published by Elsevier Science B.V. All rights reserved.

Keywords: Methane oxidising bacteria; 16S rDNA probe; Fluorescent in situ hybridisation

1. Introduction

Methane oxidising bacteria (methanotrophs) are aunique group of methylotrophic bacteria, which utilisemethane as their sole carbon and energy source (see [1,2]for reviews). These organisms have been isolated from awide variety of environments including soils [3], sediments[4], seawater [5^7], peat bogs [8^10], hotsprings [11,12],plant rhizosphere [13] and Antarctic environments [14].

Initial identi¢cation of methanotrophs grouped themaccording to their morphology, type of resting stage, intra-cytoplasmic membrane structure and physiological charac-teristics [3]. Subsequent 16S rDNA sequence analysis hasfurther clari¢ed these phylogenetic relationships and de-¢ned eight genera of methanotrophs, namely Methylococ-cus, Methylomonas, Methylomicrobium, Methylobacter,Methylocaldum, Methylosphaera, Methylocystis and Meth-ylosinus. These genera are divided into two distinct phys-

iological groups. Type I methanotrophs (Methylomonas,Methylomicrobium, Methylobacter, Methylocaldum, Meth-ylosphaera) assimilate formaldehyde produced from theoxidation of methane (via methanol) using the ribulosemonophosphate pathway, contain predominantly 16-car-bon fatty acids and possess bundles of intracytoplasmicmembranes. Type II methanotrophs (Methylocystis andMethylosinus) utilise the serine pathway for formaldehydeassimilation, have intracytoplasmic membranes arrangedaround the periphery of the cell and contain predomi-nantly 18-carbon fatty acids [1]. Members of the genusMethylococcus possess a combination of characteristicsof both type I and type II methanotrophs.

Methanotrophs form coherent phylogenetic clusterswhich share common physiological characteristics. TypeI methanotrophs (including Methylococcus) cluster in theQ-subdivision of the class Proteobacteria, while type IImethanotrophs are found in the K-subdivision of the Pro-teobacteria. The tight phylogenetic clustering of thesegroups has allowed the design of a range of oligonucleo-tides which target a broad range of both methanotrophsand methylotrophs [15^17]. A recent review by Murrell et

0168-6496 / 00 / $20.00 ß 2000 Federation of European Microbiological Societies. Published by Elsevier Science B.V. All rights reserved.PII: S 0 1 6 8 - 6 4 9 6 ( 9 9 ) 0 0 0 7 8 - 1

* Corresponding author. Tel. : +44 (1203) 523553;Fax: +44 (1203) 523568; E-mail : [email protected]

FEMSEC 1084 27-12-99 Cyaan Magenta Geel Zwart

FEMS Microbiology Ecology 31 (2000) 29^38

www.fems-microbiology.org

al. [18], highlighted the 16S rDNA and gene functionalprobes available for detection of methanotrophs andmethylotrophs.

The detection and identi¢cation of methane oxidisingbacteria is increasingly dependent on molecular techniques[9,19]. Fluorescent in situ hybridisation (FISH) has pro-vided a powerful tool to study directly organisms withinthe environment, providing information on cell morphol-ogy, phylogenetic a¤liation and the ability to quantifyorganisms [20,21]. Relatively few studies however, haveused FISH for the speci¢c detection of methanotrophs.Hanson et al. [16] successfully di¡erentiated the methylo-troph subdivisions using £uorochrome labelled probes 9-Kand 10-Q. These £uorescent probes have also been used toclassify a previously unidenti¢ed methanotroph [17] and toenumerate methylotrophs in a natural freshwater environ-ment [16]. Gilbert et al. [13] used the 9-K probe labelledwith rhodamine to detect methanotrophs in the rhizo-sphere of rice plants. FISH techniques have also beenused to detect the abundance of a putative marine meth-anotroph in enrichment cultures [5]. Two rhodamine la-belled oligonucleotide probes (Mm650 and Mm850) weresuccessfully used to determine the response of the marinemethanotroph to changes in culture conditions, therebyoptimising the enrichment culture conditions and isolationstrategies of the organism [5].

The study of diversity and population structure of meth-anotrophs have mostly depended on both culture-depend-ent methods (enrichment and isolation) and culture-inde-pendent genetic analysis of retrieved DNA sequences. Thedevelopment of techniques for the in situ detection of

methanotrophs within environments to study populationstructure has been limited. Fluorescent oligonucleotideprobes that have previously been developed and reportedfor the detection of methanotrophs and methylotrophshave proved inadequate, hence the requirement for newmethanotroph speci¢c probes. The aim of this study wasto design and test a range of oligonucleotide probes whichtarget mesophilic methanotrophs and apply these probesin £uorescent detection of these organisms. No probeswere designed for the detection of thermophilic [11,12]or psychrophilic [14] methanotrophs. Probes developedhere will be useful for community analysis of methano-trophs in environmental samples in the future.

2. Materials and methods

2.1. Microbial strains and template DNA

The microorganisms used in this study (Table 1) wereobtained from a culture collection of methanotrophsmaintained at the University of Warwick, except forstrains Methylobacter whittenburyi (Y) NCIMB 11128and Methylobacter luteus NCIMB 11914. These strainswere obtained from the National Collection of Industrialand Marine Bacteria. Cultures were grown in nitrate min-eral salts medium with the addition of excess methane(20% v/v in air) as the sole carbon substrate as describedpreviously [3]. The cultures were harvested during mid-exponential growth phase to ensure maximum ribosomecontent [22^24].

Table 1Whole cell hybridisation of a selection of reference organisms against £uorescent-labelled probes MG-64, MA-621, MA-221 and Mc1029a

Strain Probesb

MG-64 MA-621 MA-221 Mc1029

Type I methanotrophsMethylococcus capsulatus (Bath) 6 x x 6

Methylococcus capsulatus (Aberdeen) 6 x x 6

Methylococcus capsulatus (strain M) 6 x x 6

Methylococcus capsulatus (Russian) 6 x x 6

Methylobacter luteus (NCIMB 11914) 6 x x xMethylobacter whittenburyi (NCIMB 11128) x x x xMethylomonas methanica (S1) 6 x x xMethylomonas rubrum 6 x x xMethylomonas sp. 761 H 6 x xs xMethylomonas sp. 13F 6 x x xMethylomicrobium album (BG8) 6 x x xType II methanotrophsMethylocystis parvus (OBBP) x 6 6 xMethylosinus trichosporium (OB3b) x 6 6 xMethylosinus sporium 12 (DF) x 6 6 xMethylosinus trichosporium OB5b x 6 6 xMethylosinus sporium GB4 x 6 6 xE. coli K12 x x x x

aMG-64: 20% formamide in hybridisation bu¡er; MA-621, MA-221 and Mc1029: 10% formamide in hybridisation bu¡er.b6 : positive probe hybridisation and £uorescence; x: negative probe hybridisation and £uorescence.

FEMSEC 1084 27-12-99 Cyaan Magenta Geel Zwart

D.G. Bourne et al. / FEMS Microbiology Ecology 31 (2000) 29^3830

2.2. Oligonucleotide probes

The complete 16S rRNA sequences of 32 strains ofmethanotrophs including representatives of all mesophilicgenera but not thermophiles or psychrophiles, werealigned and scanned for conserved regions within thegene which could provide ideal probe target sites (seeFig. 1 for probe sequences). Potential signature sequencessuitable for in situ probes were compared to the Riboso-mal Database Project database using the CHECK_P-ROBE service (University of Illinois, Urbana) [25], thePROBE_MATCH service of the ARB software package[26] and a BLAST search against the GenBank database,to check for other organisms which may have identicalnucleotide sequences. All probe numbering is accordingto the International Union of Biochemistry nomenclaturefor Escherichia coli 16S rRNA. Probe EUB-338(GCTGCCTCCCGTAGGAGT), complimentary to a re-gion of the 16S rRNA speci¢c for the domain Bacteria,was used as a positive control to test the e¤ciency ofhybridisation [27]. The complimentary NonEUB-338served as the negative control for non-speci¢c binding[28]. Probes were synthesised with a Cy3 £uorochromeat the 5P end (Interactiva Biotechnologie Gmb, Ulm, Ger-many) and stored at 320³C. Rhodamine labelled probesused in initial whole cell hybridisation experiments werepurchased from Genosys Biotechnologies (Europe), (Cam-bridge, UK).

2.3. Whole cell hybridisations

Cells harvested for whole cell hybridisations were ¢xedovernight at 4³C in 4% (w/v) paraformaldehyde in phos-phate-bu¡ered saline (PBS; pH 7.2). Cells were washed inPBS containing 0.01% (w/v) NP40 before storage at320³C in 50% (v/v) ethanol, 20 mM Tris bu¡er (pH 7.2)and 0.1% (w/v) NP40.

Hybridisations were performed in 10-well 6 mm Te£on-coated diagnostic slides (Merck). Fixed cells were spottedonto gelatin coated slides (0.1% (w/v) gelatin; 0.01% (w/v)KCr(SO4)2) and dried at room temperature. Slides weredehydrated in 50, 80 and 96% (v/v) aqueous ethanol sol-utions for 3 min [21,29].

Fixed cells immobilised on microscope slides were hy-bridised by application of 9 Wl hybridisation solution (0.9M NaCl, 20 mM Tris-HCl, pH 7.2, 0.01% SDS w/v) andformamide concentrations as optimised for each probe(Table 1). Fluorochrome labelled oligonucleotide probes(50 ng) were added to each well of the slides. Hybridisa-tion chambers consisted of petri-dishes containing 3MMpaper soaked with hybridisation solution and sealed withPara¢lm. Hybridisation was carried out for 2^4 h in thedark at 37³C.

Following hybridisation, unbound oligonucleotides wereremoved by gently rinsing with washing bu¡er I (0.9 MNaCl, 20 mM Tris-HCl, pH 7.2; 0.01% SDS w/v). Slides

were subsequently incubated for 20 min at 37³C in wash-ing bu¡er I, transferred to washing bu¡er I containing 50Wg ml31 DAPI staining solution for 5 min at 20³C beforefurther washing in bu¡er II (0.18 M NaCl, 20 mM Tris-HCl, pH 7.2, 0.01% SDS w/v) for 15 min at 37³C. Slideswere rinsed in distilled water and air-dried in the dark.

2.4. Probe check and optimisation of stringency

The hybridisation e¤ciency of labelled oligonucleotideprobes was estimated by monitoring the £uorescence in-tensity of pure cultures of target and closely related non-target organisms. Whole cell hybridisations were per-formed with increasing concentrations of formamide(0%, 5%, 10%, 15%, 20%, 30%, 40% and 50%, v/v) todetermine conditions for discrimination of target andnon-target organisms for each probe.

2.5. Microscopy

After ¢nal air drying, slides were mounted in immersionoil (Citi£uor AF86, Citi£uor Products, Chemical Labora-tory, Canterbury, Kent, UK). Epi£uorescence microscopywas performed with a Zeiss (Oberkochen, Germany) Ax-ioskop microscope ¢tted with a 50-W high pressure bulb.The following Zeiss light ¢lter sets were used: No. 02 forDAPI (excitation, 365 nm; dichroic mirror, 395 nm; sup-pression, 397 nm) and No. 15 for Cy3 (excitation, 546 nm;dichroic mirror, 580 nm; suppression, 590 nm). Colourmicrographs were produced from Kodak Panther 1600¢lm.

2.6. Enrichment cultures and DNA extraction

A 50-ml enrichment culture established from sedimentfrom a Danish estuary was set up in 250-ml Quick¢t £askssealed with rubber Suba seals. The enrichment was grownin nitrate mineral salts (NMS) medium [3], maintained atpH 6.8 and incubated at 20³C. Flasks were supplied withthe equivalent of 200 ppmv methane in the headspace byinjection of a CH4-CO2 (95:5) mixture through the seals.The enrichment culture (10 ml) was centrifuged to concen-trate cells before whole cell hybridisation was performedas explained previously. DNA was extracted from the en-richment by the methods of Marmur [30].

2.7. PCR ampli¢cation

PCR ampli¢cation reactions [19] were performed in 50Wl (total volume) reaction mixture in 0.5 ml Eppendorftubes using a DNA thermal cycler with hotlid (Touch-down model; Hybaid, Teddington, Middlesex, UK). At-tempted optimisation of primers Mc1029, Mmb1121 andMb1125 was performed by varying annealing tempera-tures around the theoretical melting temperature of eachprimer. Each primer was used in conjunction with the

FEMSEC 1084 27-12-99 Cyaan Magenta Geel Zwart

D.G. Bourne et al. / FEMS Microbiology Ecology 31 (2000) 29^38 31

Fig

.1.

rRN

Ase

quen

ceal

ignm

ents

show

ing

targ

etre

gion

sof

prob

esan

dpr

imer

sfo

ra

sele

ctio

nof

refe

renc

em

etha

notr

oph

and

non-

met

hano

trop

hst

rain

s.N

ucle

otid

esar

eon

lyid

enti

¢ed

for

mis

pair

ings

,pa

irin

gsar

ein

dica

ted

bydo

ts.

The

base

ssh

owin

gm

ism

atch

esre

fer

toth

ese

quen

ceof

the

targ

etor

gani

sman

dno

tth

epr

obe.

Pro

benu

mbe

ring

isre

lati

veto

the

E.

coli

posi

tion

.G

enB

ank

acce

ssio

nnu

m-

bers

ofea

chor

gani

smar

epr

esen

ted

inbr

acke

ts.

(*)

Upd

ated

16S

rDN

Ase

quen

ces

for

Met

hylo

sinu

str

icho

spor

ium

OB

3ban

dM

ethy

lom

onas

rubr

umw

ere

obta

ined

from

A.

Cos

tello

and

M.E

.L

idst

rom

and

ther

efor

eac

cess

ion

num

bers

are

not

repr

esen

ted

for

thes

eor

gani

sms.

FEMSEC 1084 27-12-99 Cyaan Magenta Geel Zwart

D.G. Bourne et al. / FEMS Microbiology Ecology 31 (2000) 29^3832

eubacterial universal 16S rRNA primer 27f [31]. OptimisedPCR conditions were as follows: Mc1029 and Mmb1121primers: 94³C for 5 min; Taq polymerase added; 65³C for1 min; 72³C for 1 min; 30 cycles consisting of 94³C for1 min; 65³C for 1 min; 72³C for 1 min; and a ¢nal cycleconsisting of 94³C for 1 min; 65³C for 1 min; 72³C for5 min. Ampli¢cation of the bacterial 16S rDNA gene usedthe conserved bacteria-speci¢c primers 27f and 1492r [31]with ampli¢cation conditions described by Lane [32]. PCRampli¢ed DNA was visualised and checked for size andpurity as described previously [19].

2.8. PCR template DNA

Methanotroph chromosomal DNA used for PCR am-pli¢cation included strains Methylococcus capsulatus(Bath), Methylomicrobium agile (A30), Methylobacterwhittenburyi, Methylocystis sp. strain M, Methylocaldumtepidum LK6, Methylomicrobium album BG8, Methylocys-tis parvus OBBP, Methylosinus trichosporium OB3b. DNAwas extracted from the cells by the methods of Marmur[30]. Nitrifying bacteria DNA also used in PCR ampli¢-cation included Nitrosomonas europaea NCIMB 11850,Nitrosomonas eutropha, Nitrosospira sp. Np22, Nitroso-spira multiformis NCIMB 11849 and Nitrosococcusoceanus NCIMB 11848 and were obtained from J. Prosser(Aberdeen). DNA was extracted from these organisms asdescribed by Giovannoni [31].

2.9. DNA cloning, sequencing and analysis

PCR products of the correct size were cloned using theTA cloning kit (Invitrogen, San Diego, CA, USA), accord-ing to the manufacturer's instructions. Small scale prepa-ration of plasmids were done using the methods ofSaunders and Burke [33]. Double restriction enzyme diges-tion using enzymes EcoRI-RsaI, and Eco RI-Sau3A wasperformed as per manufacturer's instructions (Gibco,BRL). DNA sequencing reactions were carried out bycycle sequencing with the Dye Terminator kit of PE Ap-plied Biosystems (Warrington, Cheshire, UK). The univer-sal primer 357f was used in all sequencing reactions. Se-quences obtained (approximately 460 bp) were comparedagainst the GenBank database using a BLAST GAP 2package.

3. Results and discussion

3.1. Design of oligonucleotide probes

Recently, the database of methanotrophic 16S rDNAsequences has been considerably expanded and this hasallowed us to compare conserved regions and to designprobes aimed at detecting these bacteria at di¡erent phy-logenetic levels. The oligonucleotide probes designed in

this study included MG-64, targeting type I methano-trophs including members of the genera Methylomonas,Methylomicrobium, Methylobacter and Methylococcus.Probes MA-621 and MA-221 targeted type II methano-trophs, which included representatives of the genera Meth-ylocystis and Methylosinus. Probes designed for methano-trophs at the genus level included Mc1029, aimed atrepresentative species of the genus Methylococcus. Dueto the heterogeneity of the remaining genera of methano-trophs, it was di¤cult to design probes representative ofall species. As a result, probes were designed to targetspeci¢c type-strains of methanotrophs in their respectivegenus. Oligonucleotide Mb1125 targeted Methylobactercapsulatus and Methylobacter bovis strains whileMmb1121 targeted the type-strain Methylomicrobium al-bum BG8 and Methylomicrobium agile A30. A di¡erencealignment of the oligonucleotide probe sequences with thecomplementary 16S rDNA regions of target and non-tar-get organisms are shown in Fig. 1.

3.2. Evaluation of type I and type II targeted probes

The applicability of probes MG-64 and MA-621 fordiscrimination between the type I and type II phylogeneticlevels of the methanotrophs was tested using whole cells ofa variety of methanotroph strains. Initial FISH experi-ments used oligonucleotides labelled with the £uoro-chrome rhodamine, resulting in a high non-speci¢c £uo-rescence of non-target cells. In situ analysis with theNonEUB-338 rhodamine labelled probe, which should re-sult in no probe hybridisation, also conferred largeamounts of non-speci¢c £uorescence on cells. Sorensenet al. [34] encountered similar problems during in situhybridisations with rhodamine-linked probes targetedagainst methanogenic bacteria. Poulsen et al. [35] attrib-uted the problem to hydrophobic £uorochromes such asrhodamine which may bind non-speci¢cally to hydropho-bic structures in cells. To overcome these problems, oligo-nucleotides were labelled with the more hydrophilic £uo-rochrome, indocarbocyamine £uorescent dye Cy3.NonEUB-338 Cy3 labelled probes and subsequent hybrid-isations with methanotroph-speci¢c probes demonstratedthat non-speci¢c hybridisation was not a problem withthis £uorochrome label.

The results obtained from whole cell hybridisation ofthe Cy3 labelled probes MG-64, MA-621 and MA-221under optimised hybridisation conditions, against a rangeof type I and type II methanotrophs available in our cul-ture collection is presented in Table 1. Due to the non-availability of some methanotrophs, not all methanotrophspecies for which 16S rDNA sequence information wasavailable were tested in whole cell hybridisations. Like-wise, 16S rDNA data were not available for some meth-anotroph strains used in whole cell hybridisations. How-ever, the selection of organisms presented in Table 1 andFig. 1 represent a broad spectrum of representative species

FEMSEC 1084 27-12-99 Cyaan Magenta Geel Zwart

D.G. Bourne et al. / FEMS Microbiology Ecology 31 (2000) 29^38 33

from which valid conclusions can be drawn on the appli-cation of these probes to the in situ detection of methano-trophs. Importantly, representatives of each type I genus,Methylobacter, Methylomonas, Methylomicrobium andMethylococcus show strong £uorescence when hybridisedwith the MG-64 probe. Whole cell hybridisations usingprobe MG-64 against a mixed sample of type I and typeII organisms demonstrated the speci¢city of this probe fortype I methanotrophs only (Fig. 2a and b).

The probe sequence alignment in Fig. 1 demonstratedthat there were no mismatches between the 16S rDNAsequence of Methylobacter whittenburyi and the MG-64probe and therefore it was unclear why the probe failedto hybridise to this organism. Alternatively, despite a onebase-pair mismatch between the probe and the Methylo-monas methanica and Methylomomonas rubrum 16S rDNAsequences, strong £uorescent hybridisation was observed.These anomalous results illustrate that despite the abilityto design probes targeted for conserved regions of the 16SrRNA molecule, whole cell hybridisation is the only de¢n-itive way to demonstrate the speci¢city and application ofthese probes.

Database searching revealed the MG-64 probe matchedthe 16S rDNA sequence of species of bacteria from theEctothiorhodospira family and a number of phototrophicbacteria also believed to be from this family (see Fig. 1).The Ecothiorhodospiraceae represent a group of haloalka-liphilic purple sulfur bacteria which are found in marine,hypersaline and haloalkaline environments and require orprefer saline and alkaline growth conditions [36]. Untilrecently, these environments were not considered to bepotential habitats for methanotrophs. However, Khmele-nina et al. [37] isolated two halotolerant alkaliphilic obli-gate methanotrophs, referred to as Methylobacter alcali-philus based on their physiology, from moderately salinesoda lakes in central Asia and therefore caution should beused when examining these environments with the probesin case Ectothiorhodospira species are detected. Furtherdatabase searching also highlighted a Thiobacillus sp.and a L-Proteobacterium environmental clone sequencewhich possessed the same probe binding target (see Fig.1). As more sequences become available within the data-

bases, it can not be ruled out that other organisms maycontain identical probe sites.

The £uorescent-labelled probes MA-621 and MA-221speci¢cally detected type II methanotrophs without thecorresponding detection of type I methanotrophs (Table1). The 16S rDNA sequence of a number of environmentalisolates including Hyphomicrobium species, a Pedomi-crobium sp., a Rhodobium sp. and many unidenti¢ed envi-ronmental clone sequences were shown to match the MA-621 probe (see Fig. 1 for further probe site matches).Although whole cell hybridisations on these species werenot performed, it was concluded that another type II spe-ci¢c probe would be required. The MA-221 probe pro-vided an alternative, which also detected speci¢cally typeII methanotrophs. However, this probe has a one basepair mismatch with the Methylocystis parvus OBBP strain(see Fig. 1). A reduced £uorescent signal was observed inthe whole cell hybridisation of this organism to probeMA-221, though a low formamide stringency of 10% (v/v) made it possible to discriminate between type I and typeII methanotrophs. Typical results from experiments de-signed to di¡erentiate between type I and type II metha-notrophs by the use of probe MA-221 are shown in Fig. 2cand d. Whole cell hybridisations using MA-221 probeagainst Methylocystis species, M. echinoides and M. mini-mus were not performed due to the non-availability of thestrains. However, based on the 16S rDNA comparisonbetween the 16S rDNA sequences from these organismsand the probe MA-221 (see Fig. 1), two and three basepair mismatches respectively are observed and thereforehybridisation is unlikely.

Despite the possible non-speci¢c target sites of theseprobes, their major application will be in the study ofenrichment cultures and environments (i.e. high methaneenvironments) where selective culture pressure favours thegrowth of methanotrophs over other organisms. Environ-ments experiencing high methane concentrations will fa-vour high cell activity, concomitant high ribosome contentand therefore large numbers of probe site targets, whichwill confer distinct £uorescence of methanotrophs relativeto non-methanotrophic organisms. It is unlikely theseprobes or any other probes will have su¤cient discrimina-

CFig. 2. Epi£uorescence photomicrographs of in situ hybridisation with Cy3 labelled 16S rDNA probes (left) and corresponding DAPI stained cells(right). Horizontally paired photomicrographs represent the same ¢eld of view. All hybridisations were performed at 37³C for 2^4 h. Panels a^f repre-sent three species of methanotrophs in a mixed culture, Methylococcus capsulatus (Bath), Methylomicrobium album BG8 and Methylosinus trichosporiumOB3b. Methanotroph species were distinguished by morphology and are labelled in the photomicrographs with numbers. Panels a and b: In situ hybrid-isation with MG-64 type I speci¢c probe showing detection of Methylococcus capsulatus (Bath) (no. 1) and Methylomicrobium album BG8 (no. 2) cellswhile showing no hybridisation to Methylosinus trichosporium OB3b cells (no. 3). Panels c and d: In situ hybridisation with MA-221 type II speci¢cprobe, showing that Methylosinus trichosporium OB3b cells (no. 1) are speci¢cally detected, while Methylococcus capsulatus (Bath) (no. 2) and Methylo-microbium album BG8 (no. 3) cells display only faint background £uorescence. Panels e and f: In situ hybridisation with Mc-1029 probe showing detec-tion of Methylococcus capsulatus (Bath) cells (no. 1) and no hybridisation to Methylomicrobium album BG8 (no. 2) or Methylosinus trichosporium OB3bcells (no. 3). Panels g and h: In situ hybridisation with MG-64 probe on an estuarine enrichment sample demonstrating speci¢c detection of type Imethanotrophs against a background of non-methanotroph organisms.

FEMSEC 1084 27-12-99 Cyaan Magenta Geel Zwart

D.G. Bourne et al. / FEMS Microbiology Ecology 31 (2000) 29^3834

tion value to detect methanotrophs in low methane envi-ronments against background auto-£uorescence.

3.3. Genus and species speci¢c oligonucleotide probes

To assess the potential of the genus speci¢c oligonucleo-tide (Mc1029) and the species speci¢c oligonucleotides

(Mb1125 and Mmb1121) for in situ hybridisation, prelimi-nary PCR was performed with these primers and the eu-bacterial universal primer 27f [31], using template DNAfrom a range of methanotrophs and nitrifying bacteria (seeSection 2). The Mb1125 primer failed to amplify any tem-plate DNA including the probe target organism Methylo-bacter whittenburyi, despite a range of PCR conditions

FEMSEC 1084 27-12-99 Cyaan Magenta Geel Zwart

D.G. Bourne et al. / FEMS Microbiology Ecology 31 (2000) 29^38 35

tested. This primer was therefore discounted from any£uorescence in situ application. Alternatively, the primersMc1029 and Mmb1121 speci¢cally ampli¢ed their targettemplate DNA Methylococcus capsulatus (Bath) andMethylomicrobium album (BG8), respectively (results notshown). Non-speci¢c ampli¢cation was not observed foreither primer for the range of methanotroph and nitri¢erDNA templates tested.

Probes Mc1029 and Mmb1121 were labelled with theCy3 £uorochrome and whole cell hybridisation performedfor each on the selection of extant methanotrophs over arange of formamide concentrations. The Mc1029 probespeci¢cally detected cells only from its target genus Meth-ylococcus (formamide concentration 10% (v/v)) (Table 1).The speci¢city of this probe was demonstrated with probehybridisation to and £uorescence of Methylococcus capsu-latus (Bath) cells (Fig. 2e) while no £uorescence was con-ferred to the type I methanotroph Methylomicrobium al-bum (BG8) or the type II methanotroph Methylosinustrichosporium (OB3b) highlighted in the correspondingDAPI stained micrograph (Fig. 2f).

Probe Mmb1121 failed to confer a £uorescence signalagainst any methanotroph strain, including the targetMethylomicrobium album BG8 after whole cell hybridisa-tion. Amann et al. [20] have previously reported that notall target sites are equally suited to whole cell hybridisa-tion. It is therefore likely that the inaccessibility of thetarget site of the rRNA molecule, due to higher orderstructures of the ribosome, prevents binding of the probe.A recent study by Fuchs et al. [38], demonstrated thatprobes constructed for E. coli in this region demonstratedlow relative probe £uorescence.

Signi¢cant errors in deposited sequences of a selectionof methanotrophic strains have recently been identi¢ed (A.Costello and M.E. Lidstrom, personal communication).

These errors may explain some of the anomalous resultsobserved with the £uorescent probes. Correct databaseentries are essential for development of reliable probes.For example, an oligonucleotide probe, which was initiallydesigned for speci¢c detection of Methylosinus trichospo-rium OB3b based on the database entry of the 16S rDNAgene sequence of this organism, failed to hybridise to thetarget strain (Bourne and Murrell, unpublished). Closeanalysis of the region of the probe indicated the sequencedid not match the conserved secondary structures of the16S rRNA molecule. Re-sequencing of the Methylosinustrichosporium OB3b 16S rDNA has demonstrated largedi¡erences to the database entry including the area aroundthe probe site (A. Costello and M.E. Lidstrom, personalcommunication). Analysis of the updated Methylosinus16S rDNA sequence will be required to establish if anygenus speci¢c probes can be designed.

3.4. In situ probing and molecular analysis of an enrichmentculture

A typical small-scale £ask enrichment was establishedfrom a water sample from an estuarine environment andsupplied with methane. The enrichment was maintainedon nitrate mineral salts medium and was shown to oxidisemethane (results not shown). A sample of this enrichmentculture was prepared and analysed with the range of £uo-rescent oligonucleotides (MG-64, MA-621, MA-221,Mc1029) in an attempt to detect the presence of methano-trophs. In situ analysis with the MG-64 probe demon-strated the presence of type I methanotrophs (Fig. 2g).These type I methanotrophs were detected against a back-ground of other non-methanotroph bacteria as observedwith the corresponding DAPI stained micrograph (Fig.2f). No type II methanotrophs could be detected with

Fig. 3. Proportional representation of OTUs or sequence types retrieved from a 16S rDNA gene library of an estuarine enrichment culture. Bacterialidenti¢cation of each OTU group is based on closest matching sequence identity of partial 16S rDNA sequences (approximately 460 bp) after Gene-Bank database comparison with the BLAST Gap 2 software.

FEMSEC 1084 27-12-99 Cyaan Magenta Geel Zwart

D.G. Bourne et al. / FEMS Microbiology Ecology 31 (2000) 29^3836

probes MA-621 or MA-221. Similarly, Mc1029 did notdetect any Methylococcus-like organisms in the enrichmentculture.

From the same estuarine enrichment culture, DNA wasextracted and a small 16S rDNA library containing 40clones constructed. A restriction enzyme pattern of eachclone was compared to allow assignment into operationaltaxonomic units (OTUs) based on unique restriction en-zyme sites caused by sequence di¡erences within theclones. The analysis identi¢ed 11 OTUs and the resultsof this analysis are summarised in Fig. 3. OTU group 1represented the largest number of library clones containing14 out of the 40. The majority of the OTU groups (i.e.OTU groups 5^11) were only represented by a singleclone. Representatives of each OTU group was partiallysequenced (approximately 460 bp from 370^832 E. colinumbering) with the results submitted to a databasesearch (BLAST Gap 2) to provide an indication of thepossible identity of the clone types. The sequence resultsfrom 2 random clones chosen from OTU group 1 wereidentical supporting the restriction enzyme pattern classi-¢cation. The 16S rDNA sequence from these clones mostclosely matched the type I methanotroph Methylobactersp. BB5.1, being 94% similar over 460 bp. Although nota complete 16S rDNA sequence, the divergence fromMethylobacter sp. strain BB5.1 suggests that the clonemay be a novel methanotroph species. Methylobacter sp.BB5.1 had previously been isolated from methane enrich-ments of an estuarine sediment environment [4] similar tothe environment from which this clone sequence was ob-tained. Assignment of the other OTU groups from partialsequence analysis are represented in Fig. 3. No othermethanotroph-like sequences were found within the clonelibrary.

This 16S rDNA molecular analysis of the enrichmentculture has however demonstrated the presence of a typeI methanotroph and supported the results from our FISHanalysis which also identi¢ed a type I methanotroph. Thisability to monitor and study novel uncharacterised meth-anotroph populations, highlights the advantages of usingthese £uorescent 16S rRNA probes. With the constantexpansion of the 16S rDNA database, continued checkingof the applicability of these probes against a wide range oforganisms is required. Although oligonucleotides speci¢cfor other genera of methanotrophs are still needed, theprobes developed here provide a valuable tool for thestudy of methanotrophic populations in situ.

Acknowledgements

This work was supported under the European Com-munity RTD Programme Biotechnology (Bio 4 CT960419). We thank A. Costello and M.E. Lidstrom, Uni-versity of Washington for providing updated 16S rDNAsequences of two methanotroph strains and J. Prosser,

University of Aberdeen, for the cultures of ammonia-oxi-dising bacteria.

References

[1] Hanson, R.S. and Hanson, T.E. (1996) Methanotrophic bacteria.Microbiol. Rev. 60, 439^471.

[2] Murrell, J.C. (1994) Molecular genetics of methane oxidation. Bio-degradation 5, 145^149.

[3] Whittenbury, R., Phillips, K.C. and Wilkinson, J.G. (1970) Enrich-ment, isolation and some properties of methane utilising bacteria.J. Gen. Microbiol. 61, 205^218.

[4] Smith, K.S., Costello, A.M. and Lidstrom, M.E. (1997) Methane andtrichloroethylene oxidation by an estuarine methanotroph, Methylo-bacter sp. Strain BB5.1. Appl. Environ. Microbiol. 63, 4617^4620.

[5] Holmes, A.J., Owens, N. and Murrell, J.C. (1995) Detection of novelmarine methanotrophs using phylogenetic and functional gene probesafter methane enrichment. Microbiology 141, 1947^1955.

[6] Murrell, J.C. and Holmes. A.J. (1995) Molecular ecology of marinemethanotrophs. In: Molecular Ecology of Aquatic Microbes. NATOASI Series, Vol. G 38. (Joint, I. Ed.), pp. 366^390. Springer-Verlag,Berlin.

[7] Seiburth, J.M., Johnson, P.W., Eberhardt, M.A., Sieracki, M.E., Lid-strom, M.E. and Laux, D. (1987) The ¢rst methane-oxidizing bacte-rium from the upper mixed layer of the deep ocean. Methylomonaspelagica sp. nov.. Curr. Microbiol. 14, 285^293.

[8] Dedysh, S.N., Panikov, N.S. and Tiedje, J.M. (1998) Acidophilicmethanotrophic communities from Sphagnum peat bogs. Appl. Envi-ron. Microbiol. 64, 922^929.

[9] McDonald, I.R., Hall, G.H., Pickup, R.W. and Murrell, J.C. (1996)Methane oxidation potentials and preliminary analysis of methano-trophs in a blanket bog peat using molecular ecology techniques.FEMS Microbiol. Ecol. 21, 197^211.

[10] Ritchie, D.A., Edwards, C., McDonald, I.R. and Murrell, J.C. (1997)Detection of methanogens and methanotrophs in natural environ-ments. Glob. Change Biol. 3, 339^350.

[11] Bodrossy, L., Murrell, J.C., Dalton, H., Kalman, M., Puskas, L.G.and Kovacs, K. (1995) Heat-tolerant methanotrophic bacteria fromthe hot water e¥uent of a natural gas ¢eld. Appl. Environ. Micro-biol. 61, 3549^3555.

[12] Bodrossy, L., Holmes, E.M., Holmes, A.J., Kovacs, K.L. and Mur-rell, J.C. (1997) Analysis of 16S rRNA and methane monooxygenasegene sequences reveals a novel group of thermolerant methanotrophs,Methylocaldum gen. nov.. Arch. Microbiol. 168, 493^503.

[13] Gilbert, B., AMmus, B., Hartmann, A. and Frenzel, P. (1998) In situlocalization of two methanotrophic strains in the rhizosphere of riceplants. FEMS Microbiol. Ecol. 25, 117^128.

[14] Bowman, J.P., McCammon, S.A. and Skerratt, J.H. (1997) Methyl-osphaera hansonii gen. nov., a psychrophilic, group I methanotrophfrom Antarctic marine-salinity, meromictic lakes. Microbiology 143,1451^1459.

[15] Brusseau, G.A., Bulygina, E.S. and Hanson, R.S. (1994) Phylogeneticanalysis and development of probes for di¡erentiating methylotro-phic bacteria. Appl. Environ. Microbiol. 60, 626^636.

[16] Hanson, R.S., Bratina, B.J. and Brusseau, G.A. (1993) Phylogenyand ecology of methylotrophic bacteria. In: Microbial growth onC1 compounds (Murrell, J.C. and Kelly, D.P., Eds.), pp. 285^302.Intercept Ltd., Andover.

[17] Tsien, H.C., Brattina, B.J., Tsuji, K. and Hanson, R.S. (1990) Use ofoligonucleotide signature probes for identi¢cation of physiologicalgroups of methylotrophic bacteria. Appl. Environ. Microbiol. 56,2858^2865.

[18] Murrell, J.C., McDonald, I.R. and Bourne, D.G. (1998) Molecularmethods for the study of methanotroph ecology. FEMS Microbiol.Ecol. 27, 103^114.

FEMSEC 1084 27-12-99 Cyaan Magenta Geel Zwart

D.G. Bourne et al. / FEMS Microbiology Ecology 31 (2000) 29^38 37

[19] McDonald, I.R., Kenna, E.M. and Murrell, J.C. (1995) Detection ofmethanotrophic bacteria in environmental samples with the PCR.Appl. Environ. Microbiol. 61, 116^121.

[20] Amann, R.I. (1995) In situ identi¢cation of microorganisms by wholecell hybridisation with rRNA-targeted nucleic acid probes. In: Mo-lecular Microbial Ecology (Akkerman, A.D.L., van Elsas, J.D. andde Bruijn, F.J., Eds.), pp. 1^15. Kluwer Academic Publishers, Dor-drecht.

[21] Amann, R., Ludwig, W. and Schleifer, K.H. (1995) Phylogeneticidenti¢cation and in situ detection of individual microbial cells with-out cultivation. Microbiol. Rev. 59, 143^169.

[22] Amann, R.I., Binder, B.J., Olson, R.J., Chisholm, S.W., Devereux,R. and Stahl, D.A. (1990) Combination of 16S rRNA-targeted oli-gonucleotide probes with £ow-cytometry for analysing mixed micro-bial populations. Appl. Environ. Microbiol. 56, 1919^1925.

[23] Delong, E.F., Wickam, G.S. and Pace, N.R. (1989) Phylogeneticstains: ribosomal RNA-based probes for the identi¢cation of singlecells. Science 243, 1360^1363.

[24] Kemp, P.F., Lee, S. and LaRoche, J. (1993) Estimating the growthrate of slowly growing marine bacteria from RNA content. Appl.Environ. Microbiol. 59, 2594^2601.

[25] Maidak, B.L., Larsen, N., McCaughey, M.J., Overbeek, R., Olsen,G.J., Fogel, K., Blandy, J. and Woese, C.R. (1994) The ribosomaldatabase project. Nucleic Acids Res. 22, 3485^3487.

[26] Strunk, O., Gross, O., Reichel, B., May, M., Hermann, S., Stuck-mann, N., Nonho¡, B., Lenke, M., Vilbig, A., Ludwig, T., Bode, A.,Schleifer, K.H. and Ludwig, W. ARB: a software environment forsequence data. Unpublished data.

[27] Amann, R.I., Krumholz, L. and Stahl, D.A. (1990) Fluorescent-oli-gonucleotide probing of whole cells for determinative, phylogeneticand environmental studies in microbiology. J. Bacteriol. 172, 762^770.

[28] Stahl, D.A. and Amann, R. (1991) Development and application ofnucleic acid probes. In: Nucleic Acid Techniques in Bacterial System-atics (Stackebrandt, E. and Goodfellow, M., Eds.), pp. 205^244. JohnWiley and Sons, Chichester.

[29] Amann, R.I., Lugwig, W. and Schleifer, K.-H. (1992) Identi¢cationand in situ detection of individual bacterial cells. FEMS Microbiol.Lett. 100, 45^50.

[30] Marmur, J. (1961) A procedure for the isolation of deoxyribonucleicacid from microorganisms. J. Mol. Biol. 3, 208^218.

[31] Giovannoni, S.J. (1991) The polymerase chain reaction. In: NucleicAcid Techniques in Bacterial Systematics (Stackebrandt, E. andGoodfellow, M., Eds.), pp. 177^203. John Wiley and Sons, Chiches-ter.

[32] Lane, D.J. (1991) 16S/23S rRNA sequencing. In: Nucleic Acid Tech-niques in Bacterial Systematics (Stackebrandt, E. and Goodfellow,M., Eds.), pp. 115^175. John Wiley and Sons, Chichester.

[33] Saunders, S.E. and Burke, J.F. (1990) Rapid isolation of miniprepDNA for double strand sequencing. Nucleic Acids Res. 18, 4948.

[34] Sorensen, A.H., Torsvik, V.L., Torsvik, T., Poulsen, L.K. and Ahr-ing, B.K. (1997) Whole-cell hybridization of Methanosarcina cellswith two new oligonucleotide probes. Appl. Environ. Microbiol. 63,3043^3050.

[35] Poulsen, L.K., Lan, F., Kristensen, C.S., Hobolth, P., Molin, S. andKrogfelt, K. (1994) Spatial distribution of Escherichia coli in themouse large intestine inferred from rRNA in situ hybridisation. In-fect. Immun. 62, 5191^5194.

[36] Imho¡, J.F. and Suling, J. (1996) The phylogenetic relationshipamong Ectothiorhodospiraceae: a revaluation of their taxonomy onthe basis of 16S rDNA analyses. Arch. Microbiol. 165, 106^113.

[37] Khmelenina, V.N., Kalyuzhnaya, M.G., Starostina, N.G., Suzina,N.E. and Trotsenko, Y.A. (1997) Isolation and characterisation ofhaloltolerant alkaliphilic methanotrophic bacteria from Tuva sodalakes. Curr. Microbiol. 35, 257^261.

[38] Fuchs, B.M., Wallner, G., Beisker, W., Schwippl, I., Ludwig, W. andAmann, R. (1998) Flow cytometric analysis of the in situ accessibilityof Escherichia coli 16S rRNA for £uorescently labeled oligonucleotideprobes. Appl. Environ. Microbiol. 64, 4973^4982.

FEMSEC 1084 27-12-99 Cyaan Magenta Geel Zwart

D.G. Bourne et al. / FEMS Microbiology Ecology 31 (2000) 29^3838