APPLIED AND ENVIRONMENTAL MICROBIOLOGY, Mar. 2011, p. 1581–1587 Vol. 77, No. 50099-2240/11/$12.00 doi:10.1128/AEM.01960-10Copyright © 2011, American Society for Microbiology. All Rights Reserved.

Detection, Isolation, and Characterization of Helicobacter Species fromthe Gastrointestinal Tract of the Brushtail Possum�

Thosaporn Coldham,1 Kerrie Rose,2 Jani O’rourke,1 Brett A. Neilan,1Helen Dalton,1 Adrian Lee,1 and Hazel Mitchell1*

School of Biotechnology and Biomolecular Sciences, University of New South Wales, Sydney, Australia,1

and Veterinary & Quarantine Centre, Taronga Zoo, Sydney, Australia2

Received 18 August 2010/Accepted 28 December 2010

The presence of Helicobacter species in Australian marsupials was examined systematically using micros-copy, culture, and PCR in different regions of the gastrointestinal tract (GIT) and in the liver of brushtailpossums (BTPs) (Trichosurus vulpecula), a common Australian marsupial that feeds on eucalyptus leaves. Thespatial distribution of Helicobacter species in the GIT sections also was examined microscopically in silver-stained sections and by fluorescent in situ hybridization (FISH) using a Helicobacter genus-specific probe.Helicobacter species were found colonizing the lower bowel of all BTPs studied. Good agreement was observedbetween the detection of Helicobacter species using culture and PCR, which was supported by the microscopicexamination of silver-stained sections and FISH. The lower bowel of BTPs were colonized by one to threemorphologically different (a comma-shaped species with no apparent flagella, a fusiform-shaped speciesentwined with periplasmic fibers and a bipolar sheathed flagella, and an S-shaped species with bipolarsheathed flagella) and potentially novel Helicobacter species, as well as in one case with a potentially novelCampylobacter species, which was a tightly coiled rod with bipolar unsheathed flagella. The isolation andcharacterization of these Helicobacter species in BTPs provides important information regarding the specificnatural niche of these bacteria and their corelationship within their host, and it increases our understandingof the ecology of Helicobacter species.

Helicobacter spp. have been shown to naturally colonize thestomach and intestines of most animals and the liver of someanimals (4, 5, 6, 8). The genus Helicobacter is, in general,classified into two groups, gastric and enterohepatic Helico-bacter spp. Gastric Helicobacter spp. have been found to colo-nize the stomach of humans, dogs, cats, cheetahs, rhesusmonkeys, ferrets, sheep, cattle, whales, and dolphins, whileenterohepatic Helicobacter spp. are more commonly found col-onizing animals such as mice, rats, and hamsters (4, 28). At thecommencement of the current study, no information concern-ing Helicobacter spp. in one of the major classes of Mammalia,the Marsupialia, was known. The common brushtail possum(BTP; Trichosurus vulpecula) is the largest tree-dwelling mar-supial herbivore and feeds on a variety of leaves, particularlyeucalyptus (17). Physiologically the BTP is a cecum fermenter,having a simple stomach and a well-developed cecum andproximal colon (15).

The aim of this study was to investigate the presence ofspiral (Sp)-shaped and fusiform (F)-shaped Helicobacter spp.colonizing the mucus layer of different regions of the gastro-intestinal tract (GIT) as well as the liver of BTPs using micros-copy, culture, and PCR. Spiral- and fusiform-shaped isolateswere characterized and identified using morphological appear-ance, Helicobacter genus-specific PCR, and 16S rRNA genesequence comparisons. The spatial distribution of Helicobacterspecies in GIT sections was examined microscopically in silver-

stained sections of the GIT as well as by fluorescent in situhybridization (FISH) using a Helicobacter genus-specific probe.The number and percentage of Helicobacter species as de-tected by PCR and culture in different regions of the GIT werecompared. In addition, the specific natural niche of these bac-teria and the corelationship between mucus-associated micro-organisms and their brushtail possum hosts was evaluated.

MATERIALS AND METHODS

Animals and collection of specimens. GIT and liver specimens from 11 wildadult brushtail possums (BTP1 to BTP11; four males, four females, and threeunspecified sexes) were collected at the Veterinary & Quarantine Center,Taronga Zoo, Sydney, Australia. All BTPs had been injured or were in ill health,taken to the Veterinary & Quarantine Center, and subsequently died or wereeuthanized for compassionate reasons. For each animal, three samples of tissuefrom each of the following sites were collected: the liver, stomach, mid-ileum,ileum at 3 cm above the cecum (3-ileum), cecum, colon, and rectum. The firstsample from a particular location was frozen at �70°C for DNA extraction.The second sample was frozen in brain heart infusion-glycerol medium(BHIG) and kept at �70°C until cultured. The third sample was fixed informalin for histology.

Bacterial isolation and biochemical characterization. Samples were culturedon horse blood agar plate (HBA) and on campylobacter selective agar plates(CSA) using both the direct inoculation method and the selective filtrationmethod as described by Robertson et al. (25). The phenotypic characteristic of allisolates was determined using standardized methods as recommended in anextensive identification scheme designed for Campylobacter, Helicobacter, andrelated bacteria (3, 19). Nitrate reduction and gamma-glutamyl transferase ac-tivity were examined using the API-Campy identification system (bioMerieux,Marcy-l’Etoile, France). Indoxyl acetate hydrolysis was examined using the diskmethod of Hodge et al. (14). Alkaline phosphatase and hippurate hydrolysis wasexamined using Rosco diagnostic tablets (UTEC Diagnostics, Denmark). Sus-ceptibility to nalidixic acid (30 �g; designated NA 30; Oxoid), cephalothin (30�g; KF 30; Oxoid), and metronidazole (5 �g; MTZ 5; Oxoid) was determined bydisk diffusion on HBA plates. Strains were determined to be sensitive if there wasa zone of inhibition and resistant if there was no zone (3, 16).

* Corresponding author. Mailing address: The School of Biotech-nology and Biomolecular Sciences, The University of New SouthWales, Sydney, New South Wales 2052, Australia. Phone: (02) 93852040. Fax: (02) 9385 1483. E-mail: H.Mitchell@unsw.edu.au.

� Published ahead of print on 7 January 2011.

1581

on February 3, 2021 by guest

http://aem.asm

.org/D

ownloaded from

DNA extraction for PCR. DNA from the liver and GIT mucus scrapings of all11 BTPs was extracted using the phenol-chloroform method. DNA from bacte-rial cells was extracted using the Puregene DNA isolation kit (Gentra Systems).

Helicobacter genus-specific PCR. A Helicobacter genus-specific PCR was per-formed as described by Riley et al. (23) using primers H276f (5�-CTATGACGGGTATCCGGC-3�; E. coli positions 276 to 293) and H676r (5�-ATTCCACCTACCTCTCCCA-3�; E. coli positions 676 to 658) with minor modifications.Briefly, the MgCl2 concentration was increased from 1.5 to 3 mM, and theannealing temperature increased from 53 to 57°C. The PCRs underwent aninitial denaturation period at 94°C for 5 min, followed by 35 cycles of denatur-ation at 94°C for 5 s, annealing at 57°C for 5 s, and extension at 72°C for 30 s,followed by a final extension at 72°C for 2 min. The PCR was conducted in aSprint thermal cycler (Hybaid, Ashford, Middlesex, United Kingdom).

Nested PCR. Nested PCR was conducted by amplifying the prokaryote 16SRNA gene using the universal bacterial primers F27 (5�-AGAGTTGATCCTGGCTCAG-3�; E. coli positions 8 to 27) and R1494 (5�-TACGGCTACCTTGTTACGAC-3�; E. coli positions 1494 to 1513) (18, 32). Reactions in the first roundof the PCR underwent an initial denaturation period at 94°C for 5 min, followedby 30 cycles of 94°C for 10 s, 50°C for 15 s, and 72°C for 2 min, with a finalextension at 72°C for 7 min. The PCR product (1:10 dilution) of this first-roundPCR then was used as a template to amplify Helicobacter genus-specific DNA inthe second-round PCR, using primers H276f and H676r as described above.

Electron microscopy. Freshly grown bacteria from HBA plates were negativelystained with 2% uranyl acetate and examined with transmission electron micros-copy (TEM) (H7000; Hitachi, Tokyo, Japan).

Histopathology. Four-micron sections of paraffin-embedded samples were cutand stained using a modified Steiner silver stain and a hematoxylin and eosin(H&E) stain. The silver-stained sections were examined for the presences ofspiral- or helical-shaped bacteria and the H&E-stained sections for histopatho-logical changes.

DNA sequencing. The PCR amplification of the 16S rRNA gene was con-ducted using the universal bacterial primers F27 and R1494 (18, 32). The PCRproduct was purified by ethanol precipitation prior to sequencing using the ABIPrism BigDye terminator cycle sequencing ready reaction kit (PE Applied Bio-systems, Foster City, CA) according to the manufacturer’s protocols. The prim-ers used for the sequencing reaction are H276f, H676r (23), F27, R1494, 341R,530F, 1115F, and 1220R (18, 32). Completed sequencing reactions were purifiedby ethanol precipitation. Sequencing products were separated on an ABI Prismmodel 377 DNA sequencer machine and analyzed using the Perkin Elmer ABIPrism sequencing analysis software, version 3.3 (PE Biosystems, FosterCity, CA).

Phylogenetic analysis. 16S rRNA gene sequences were assembled using theABI Prism TM sequencing 2.1.1 Inherit autoassembler program, and the con-sensus was compared to the sequences of other bacteria in the GenBank data-bases using BLASTN (http://www.ncbi.nlm.nih.gov/BLAST/). Sequences werealigned using the GCG program Pile Up, version 8 (Genetics Computer Group)and the multiple-sequence alignment and profile alignment tools in the ClustalXpackage (30). The alignments were only on a pairwise basis. The phylogenetictree was constructed by the neighbor-joining method of Saitou and Nei (26); theoption “positions with gaps in the multiple sequence alignment excluded” waschosen and the tree plotted using NJ plot in the ClustalX package. The level ofsimilarity of sequences was determined using a multiple-sequence alignment asthe input in the HOMOLOGIES program (GCG). All programs in the PHYLIP

package used in the sequence manipulation and phylogenetic analyses wereaccessed via the ECGC extensions to the Wisconsin Package, version 8.1.0, 1996,via the Australian National Genomic Information Service (ANGIS; http://biomanager.info/).

FISH. The oligonucleotide probes for fluorescent in situ hybridization (FISH)were EUB338-FITC (eubacterial 16S rRNA probe 5� GCTGCCTCCCGTAGGAGT 3� [1], labeled with fluorescein-isothiocyanate [FITC]) and HRh (Heli-cobacter genus-specific probe 5� TCTCAGGCCGGATACCCGTCATAGCCT3� [9], labeled with tetramethyl-rhodamine-isothiocyanate [TRITC] [Genset Pa-cific Pty. Ltd., Lismore, Australia]). Fixed H. pylori strain SS1 (positive control)and Psychrobacter sp. strain SW5 (negative control) were used as bacterial cellcontrols in every test. The cell control suspension was fixed in 4% paraformal-dehyde and incubated at 4°C for up to 24 h. The cell was pelleted and thenresuspended in a 10% solution of 0.1% nonionic detergent (Igepal CA-630;Sigma Chemical Co., St. Louis, MO.) in phosphate-buffered saline (PBS). Thisstep was repeated using 500 �l of 0.1% nonionic detergent, and finally the pelletwas resuspended in a mixture of an equal volume of storage buffer (40 mM Trisbuffer, pH 7.2, and 0.2% Igepal CA-630) and 96% ethanol and stored at �20°C.

One to 3 �l of the fixed-cell controls were applied to a poly-L-lysine-coatedslide and dehydrated in 50, 80, and 96% ethanol for 3 min each and left to dry.Five-�m tissue sections were deparaffinized in xylene for 10 min and washed with100% ethanol before being air dried. Nine volumes of hybridization solution(40% formamide) and 1 volume of the appropriate probe, EuB338-FITC (0.64�g/�l) or HRh (0.8 �g/�l), were added to wells containing the control cells.Eighteen volumes of hybridization solution and 2 volumes of each of EuB338-FITC and HRh probes were added to the slide containing tissue sections. Theslides then were incubated in a dark, sealed, moist chamber overnight at 37°C.After hybridization, the slides were rinsed with warm milli-Q water, followed bytreatment with warm hybridizing solution, and then incubated for 20 min at 37°C.Prewarmed washing solution II (0.1 M Tris, pH 7.2, and 0.9 M NaCl) then wasadded, and the slides were incubated for a further 15 min at 37°C. Finally, theslides were rinsed in milli-Q water and air dried in the dark. The epifluorescentmicroscopic examination of the slides was conducted using a Zeiss Axioskopmicroscope fitted with an HBO 50-W mercury lamp and equipped with filter sets10 and 15 (Carl Zeiss, Oberkochen, Germany) for detecting the green FITCsignal and the red TRITC signal, respectively. Slides also were visualized usingscanning confocal laser microscopy (SCLM) with an Olympus GB200 micro-scope (Olympus Optical Company Ltd., Tokyo, Japan) fitted with a piezoelectricz stage. The microscope was fitted with a 60�, 1.4-numerical-aperture oil im-mersion lens. An argon laser was used as the excitation source for the fluoro-chromes used. The images obtained were analyzed using Photoshop 6.0 (AdobeSystems Inc., Mountain Vies, CA).

Nucleotide sequence accession numbers. The 16S rRNA accession numbers inGenBank for BTP1C to BTP9C are AY554115 to AY554123, for BTP1F toBTP6F are AY554124 to AY554129, for BTP1S to BTP3S are AY554135 toAY554137, and for the Tcr isolate is AY554142.

RESULTS

Bacterial cultivation. Twenty-eight of 48 spiral- or fusiform-shaped isolates from the BTPs were identified as Helicobacterspecies using the Helicobacter genus-specific PCR. The micro-

TABLE 1. Results of culture and direct and nested PCR obtained from the 11 BTPsa

Animal Liver Stomach Mid-ileum 3-Ileum Cecum Colon Rectum

BTP1 N N N N dP, N dP, N Csp, dP, NBTP2 N N N dP, N Csp, dP, N dP, N dP, NBTP3 N N N Csp, dP, N Csp, dP, N Csp, dP, N Cspf, dP, NBTP4 – – – Csp, N Csp, dP, N Csp, dP, N Csp, dP, NBTP5 – – – – – Cspf, dP, N Cspf, dP, NBTP6 – – – – Cf, N Cf, dP, N dP, NBTP7 – – – – – – Cf, dP, NBTP8 N – – N Csp, N Csp, dP, N Cf, dP, NBTP9 N – – – Csp, N Csp, dP, N Csp, dP, NBTP10 – – N N Csp, N Csp, dP, N Csp, dP, NBTP11 N N N N Tcr, dP, N dP, N Csp, dP, N

a 3-Ileum, ileum at 3 cm above the cecum; Csp, spiral-shaped Helicobacter cultured; Cspf, spiral- and fusiform-shaped Helicobacter cultured; Cf, fusiform-shapedHelicobacter cultured; Tcr, tightly coiled rod cultured; dP, direct PCR positive; N, nested PCR positive; �, Helicobacter culture, direct and nested PCR, all negative.

1582 COLDHAM ET AL. APPL. ENVIRON. MICROBIOL.

on February 3, 2021 by guest

http://aem.asm

.org/D

ownloaded from

scopic examination of the 28 PCR-positive isolates showed 21to be of a spiral (Sp)-shaped morphology and 7 to be of afusiform (F) morphology. In the cecum of one brushtail pos-sum, a non-helicobacter isolate that was a tightly coiled rod(Tcr) also was isolated.

The results of culture, the location from which these isolateswere cultivated, and the direct and nested PCR results ob-tained from the 11 BTPs are shown in Table 1. A comparisonof the number and percentage of the detection of Helicobacterspecies in different regions of the GIT by direct and nestedPCR and culture is shown in Table 2.

Electron microscopy. Eleven spiral to curved isolates and 4fusiform, potentially novel helicobacter and a non-helico-bacter Tcr isolates were examined further using TEM. The

analysis showed that the 15 negatively stained isolates fromBTP fell into three morphological types: comma-shaped iso-lates measuring 0.3 �m by 1.5 to 2.2 �m and without flagella(Fig. 1A) and fusiform-shaped isolates measuring 0.8 �m by2.3 to 3.0 �m and entwined with periplasmic fibers, whichappeared to cover the whole cell. They had multiple (6 to12) bipolar sheathed flagella (Fig. 1B). The S-shaped iso-lates were 0.3 by 2.3 �m and had bipolar sheathed flagella(Fig. 1C). The Tcr isolate measured 0.5 by 4 �m, was atightly coiled rod with no periplasmic fibers, and had bipolarunsheathed flagella (Fig. 1D).

16S rRNA sequencing analysis. The near-complete 16SrRNA gene sequences (�1,450 bp) obtained from nine comma-shaped (BTP1C to BTP9C), six fusiform-shaped (BTP1F toBTP6F), and three S-shaped (BTP1S to BTP3S) isolateswere compared to the sequences in GenBank. The comma-shaped BTP isolates showed 97% similarity to Helicobactersp. MIT 99-5507, isolated from rhesus monkey 2 (accessionnumber AF33334) (7). The fusiform-shaped BTP isolatesshowed 96% similarity to H. marmotae strain MIT 98-6070(accession number AF333341), isolated from the liver of awoodchuck (5). The S-shaped BTP isolates showed 96%similarity to Helicobacter sp. MIT 01-6242, isolated from thegastric mucosa of harp seals (Phoca groenlandica) (13). Thecomparison of the 16S rRNA gene sequences obtained inthis study to other sequences from GenBank confirmed thatthe 18 isolates potentially were novel Helicobacter species.

TABLE 2. Comparison of the number and percentage ofHelicobacter species detected in different regions of the

GIT of the 11 BTPs using culture and direct andnested Helicobacter genus-specific PCR

Region Cultivation (%) Direct PCR (%) Nested PCR (%)

Liver 0/11 (0) 0/11 (0) 6/11 (54.5)Stomach 0/11(0) 0/11(0) 4/11 (36.4)Mid-ileum 0/11 (0) 0/11 (0) 5/11 (45.5)3-Ileum 2/11 (18.2) 2/11 (18.2) 7/11 (63.6)Cecum 7/11 (63.6) 5/11 (45.5) 9/11 (81.8)Colon 7/11 (63.6) 10/11 (90.9) 10/11 (90.9)Rectum 9/11 (81.8) 11/11 (100) 11/11 (100)

FIG. 1. (A) Transmission electron micrograph of a comma-shaped Helicobacter isolate, BTP5C, measuring 0.3 by 1.5 to 2.2 �m. No flagellawere apparent (bar, 500 nm). (B) TEM of a fusiform-shaped Helicobacter isolate, BTP4F, measuring 0.8 by 2.3 to 3.0 �m, entwined withperiplasmic fibers, which appeared to cover the whole cell. It had multiple (6 to 12) bipolar sheathed flagella (bar, 1,000 nm). (C) TEM an S-shapedHelicobacter isolate, BTP1S, measuring 0.3 by 2.3 �m, with bipolar sheathed flagella (bar � 500 nm). (D) TEM of the tightly coiled rodCampylobacter sp. Tcr, measuring 0.5 by 4 �m with bipolar unsheathed flagella (bar, 1,000 nm).

VOL. 77, 2011 HELICOBACTER SPP. FROM GIT OF BRUSHTAIL POSSUM 1583

on February 3, 2021 by guest

http://aem.asm

.org/D

ownloaded from

The Tcr isolate was shown to be a Campylobacter speciesand showed 97% similarity to Campylobacter helveticus (ac-cession numbers AF550647, AF550646, and NR025948)(12).

16S rRNA sequence homology. The sequences from eachgroup of possum isolates showed a high level of homologywithin each group (comma-shaped BTP, S-shaped BTP, andfusiform-shaped BTP). The percent similarity of each of theisolates was 99.5 to 100% within comma-shaped BTP, 99.6 to100% within fusiform-shaped BTP, and 98.8 to 99.6% withinS-shaped BTP. The percent similarity between each of thegroups was �95.3% among the groups.

The phylogenetic tree obtained for the 18 BTP isolatesand 32 members of the genera Helicobacter, Wolinella, andCampylobacter is shown in Fig. 2. These results further con-firmed that the three groups of BTP isolates could constitutethree novel Helicobacter species. To further investigate thispossibility, the growth, biochemical characterization, andsusceptibility to antimicrobial agents of comma-shaped BTPisolates (BTP1C to BTP9C) and fusiform-shaped BTP iso-

lates (BTP1F to BTP6F) were examined. Both comma-shaped and fusiform-shaped isolates were resistant to nali-dixic acid and cephalothin (inhibition zone, 0 mm) butsensitive to metronidazole (inhibition zone, �50 mm forcomma-shaped isolates and �30 mm for fusiform-shapedisolates). Fusiform-shaped isolates were hippurate hydroly-sis positive. A comparison of a number of these charac-teristics to those of other Helicobacter species is shown inTable 3.

Spatial distribution of mucus-associated bacteria in fixedsections of the liver and different regions of the GIT of BTPs.In the silver-stained liver sections of the 11 BTP, only a fewcurved- to spiral-shaped bacteria were observed in two BTPs(BTP1 and BTP6) (Fig. 3A and B). Very low numbers ofbacteria were observed in the stomach, mid-ileum, and 3-il-eum sections of the majority of BTPs studied. In contrast,large numbers of curved- to spiral-shaped bacteria wereobserved in the mucus layer overlying the epithelium and inthe crypts of the rectal, colonic, and cecal samples. Anexample of the distribution of these mucus-associated bac-

FIG. 2. Phylogenetic tree for 18 BTP isolates and 32 members of the genera Helicobacter, Wolinella, and Campylobacter, performed usingClustal X (30). The scale bar represents 0.01 expected substitutions per site. A phylogenetic tree was constructed using the neighbor-joining method(26).

1584 COLDHAM ET AL. APPL. ENVIRON. MICROBIOL.

on February 3, 2021 by guest

http://aem.asm

.org/D

ownloaded from

teria in the rectal regions of BTP8 in silver-stained sectionsis shown in Fig. 3C.

The spatial distribution of bacteria belonging to the genusHelicobacter was further observed using FISH with a Helico-bacter genus-specific probe, using the formalin-fixed rectal sec-tions of four brushtail possums (BTP6, BTP7, BTP8, andBTP10), in which Helicobacter species had been isolated anddetected by PCR. In these four rectal sections, Helicobacterspecies were shown to localize predominantly in the mucuslayer overlying the surface epithelium and within the crypts ofthe rectum. Images of the hybridized bacteria taken fromBTP10 using confocal scanning microscopy are shown in Fig. 4.Most of the bacteria present in the mucus layer overlying theepithelium and in the rectal crypts were shown to belong to thegenus Helicobacter (Fig. 4B, red, and C, orange).

Histopathology. The histopathological examination of theliver sections from all 11 brushtail possums showed that allwere within the normal range, with only a mild multifocalportal infiltration of small mononuclear cells being evident. Nosignificant lesions were observed in the stomachs of 9 of the 11BTPs. In 2 of the 11 BTPs studied (BTP5 and BTP9), scatteredlymphocytes, plasma cells, and eosinophils were observed inthe lamina propria of the stomach. The lower bowel of allBTPs was normal, with no infiltration being observed in theintestinal mucosa.

DISCUSSION

The presence of Helicobacter species was systematically ex-amined in different regions of the GIT and the liver of Aus-tralian marsupials. This showed, for the first time, that Helico-bacter species are present in the GIT of brushtail possums. Ofthe 11 wild BTPs examined, a total of 28 helicobacter-positivesignals were detected in the lower bowel by culture and PCRamplification. Of these 28 isolates, 18 were further analyzedand shown to belong to the Helicobacter genus by 16S rRNAgene sequence analysis. The Helicobacter spp. isolated from theBTPs differed in morphology and could be separated into threemorphological types by TEM: comma, fusiform, and S shaped.The level of similarity of the 16S rRNA gene sequences of eachmorphological type was �98.8%, with each morphological typerepresenting a possibly novel Helicobacter species. Two un-usual characteristics of the Helicobacter spp. isolated fromBTPs were noted. First, no flagella were apparent in the com-ma-shaped isolates observed under electron microscopy, al-though they were actively motile. Second, the fusiform-shapedisolates were hippurate hydrolysis positive, representing thefirst report of Helicobacter species that can hydrolyze hippu-rate. This finding may facilitate the differentiation of fusiform-shaped isolates from other Helicobacter species. Interestingly,hippurate hydrolysis is important for the differentiation ofCampylobacter jejuni and Campylobacter coli (29).

While bacterial culture detects only viable organisms, thedetection of helicobacter DNA can represent viable and/ornonviable cells. A good agreement was found between thehelicobacter cultivation results and the direct PCR results, asshown in Table 2. This table shows that in most BTPs, thenumber of helicobacters present in the cecal region and theregions above this was very low and below the limit of detec-tion of culture or direct PCR. Nested PCR, which is considered

TA

BL

E3.

Characteristics

thatdifferentiate

comm

a-shapedB

TP

andfusiform

-shapedB

TP

isolatesfrom

otherH

elicobacterspecies

a

Taxon

Catalase

productionN

itratereduction

Urease

activityH

ippuratehydrolysis

Alkaline

phosphataseactivity

Glutam

yltransferase

activity

Indoxylacetate

hydrolysis

Susceptibilityto:

Grow

th

Periplasmic

fibersN

o.offlagella

Distribution

offlagella

Cephalothin

(30�

gdisk)

Nalidxicacid

(30�

gdisk)

At

42°CW

ith1%

glycine

BT

P,comm

ashaped

(7/7)

(7/7)

�(0/7)

�(0/7)

�(0/7)

�(0/7)

V(4/7)

R(7/7)

R(7/7)

�(0/7)

(7/7)

NA

NA

NA

BT

P,fusiformshaped

(5/5)

�(0/5)

(5/5)

(5/5)

�(0/5)

(5/5)

�(0/5)

R(5/5)

R(5/5)

�(0/5)

(5/5)

(5/5)

6�12

Bp

H.aurati

�

N

D�

RS

�

7�

10B

pH

.bilis

��

�

RR

3�

14B

pH

.canadensis

V�

ND

��

R

R

ND

�1�

2B

pH

.canis�

��

�

IS

�

�2

Bp

H.ganm

ani(�

)�

��

��

�R

S�

��

2B

pH

.hepaticus

��

�

RR

�

�2

Bp

H.trogontum

�

V

�R

R

ND

5�

7B

pH

.pullorum

�

��

ND

ND

SR

�

�1

Mp

H.m

uridarum

�

�

RR

��

10�

14B

pH

.mustelae

��

�R

SV

��

4�8

PtH

.pylori

�

�

SR

(�)

��

4�8

Bp

a

,100%strains

positive;�

,100%strains

negative;V,variable

results;(�),7

to38%

strainsnegative;S,sensitive;R

,resistant;I,intermediate;N

D,notdeterm

ined;NA

,notapplicable;Bp,bipolar;M

p,monopolar;

Pt,peritrichous.

VOL. 77, 2011 HELICOBACTER SPP. FROM GIT OF BRUSHTAIL POSSUM 1585

on February 3, 2021 by guest

http://aem.asm

.org/D

ownloaded from

to be more sensitive than direct PCR, detected Helicobacterspecies in a number of samples that were negative by cultureand direct PCR (Table 2). In some instances the results ofnested PCR supported the findings of histology. Indeed,curved- to spiral-shaped bacteria were observed in silver-stained sections of almost all nested PCR-positive samples,except for the liver.

There are two possible explanations for the observed dis-crepancies between the detection of Helicobacter spp. in theliver by nested PCR (6/11 BTPs) and the silver stain examina-tion (2/11 BTPs, BTP1 and BTP6). First, in the liver, thebacteria are randomly distributed (usually in small numbers)and thus often are difficult to detect by silver staining (5–6, 8).This contrasts with the GIT, where bacteria are usually inlarger numbers and are located primarily in the mucus layeroverlying the intestines (Fig. 3C). Second, nested PCR detectsonly Helicobacter DNA, which may in these cases have trans-located from the GIT due to the disruption of the gastrointes-tinal ecology as a result of intestinal overgrowth in the GIT,physical damage to the mucosal barrier, or a decrease in theimmune defenses (2). In BTP6, low numbers of curved- tospiral-shaped bacteria were observed in the silver-stain sec-tions of the liver (Fig. 3B), stomach, mid-ileum, and ileum at 3cm above the cecum of BTP6; however, no helicobacters weredetected in the liver by culture or direct or nested PCR. These

results suggest that the observed organisms in the liver ofBTP6 do not belong to the Helicobacter genus.

The detection of Helicobacter species in the liver of animalshas been documented previously, particularly in mice. Indeed,several Helicobacter species have been isolated or detected inextragastric or intestinal sites, including the liver, bile, gallblad-der, and blood of animals and humans (4–6, 8, 10, 11, 31). Ithas been suggested that bacteria access the liver by initialM-cell uptake, with spread to the liver via the portal circulationand finally the discharge of the bacteria into the biliary tract.Another possibility is the direct translocation through entero-cytes or the migration of helicobacters from the lumen of thegut into the bile duct (2, 6).

Few studies to date have described a systematic investigationof the entire GIT of any animal. In the current study, morethan one morphological type of Helicobacter species was iso-lated from a single specimen in a number of the brushtailpossums. For example, both spiral- and fusiform-shaped heli-cobacters were found cocolonizing the colonic and rectal areasof three of the BTPs (BTP3, BTP5, and BTP8). Although notall of the helicobacter isolates were examined by morphologi-cal analysis using TEM, S-shaped isolates were found to mainlycolonize the cecum. The presence of more than one morpho-logical type of spiral-shaped helicobacter in a single location ofsome of these brushtail possums is consistent with previous

FIG. 3. Photomicrographs of silver-stained liver and rectal sections showing small numbers of curved- and spiral-shaped organisms in the liverand high numbers of curved- to spiral-shaped organisms in the mucus layer lining of the rectum of the brushtail possums. (A) The liver of BTP1;(B) the liver of BTP6; and (C) the rectum of BTP8. Magnification for A and B, �1,000; magnification for C, �400.

FIG. 4. Confocal fluorescent photomicrographs of a rectal sample taken from BTP10. (A) All bacteria hybridized with the Eu338-FITC probeshowing a green color. (B) Helicobacter spp. hybridized with the HRh probe showing a red color. (C) Helicobacter spp. hybridized with bothEu338-FITC and HRh probe showing an orange color (mixture of red and green colors); non-helicobacter strains were stained in a yellowish greencolor.

1586 COLDHAM ET AL. APPL. ENVIRON. MICROBIOL.

on February 3, 2021 by guest

http://aem.asm

.org/D

ownloaded from

findings for rodents (20–22, 24, 27). The cocolonization ofHelicobacter spp. and Campylobacter spp. (Tcr isolate) also wasfound in this study. The FISH examination of rectal sectionsfrom four brushtail possums in which Helicobacter spp. hadbeen detected and isolated showed that the majority of mucus-associated bacteria belonged to the genus Helicobacter.

Given that the histopathological examination of sections ob-tained from the liver and gastrointestinal tract of the brushtailpossums showed all animals to be, in general, healthy, theseHelicobacter species are likely to represent indigenous micro-organisms that colonize the lower bowel of wild, normal,healthy brushtail possums.

Brushtail possums are cecum fermenters, with microbial fer-mentation taking place in the cecum/proximal colon site (15,17). From this study, the ecological niche of Helicobacter spe-cies was shown to be the lower bowel, covering the areasbetween the cecum and rectum and, in some animals, expand-ing to the ileum at 3 cm above the cecum. Although thecolonization of Helicobacter species was not confined to thececal-proximal colonic area, an important role of Helicobacterspecies in the fermentation of food cannot be excluded. It alsois possible that Helicobacter spp. gain an optimal nutritionalbenefit from this area, as this region is the preferred ecologicalniche for Helicobacter species colonization in most mammals(28). Currently, however, there is no evidence indicatingwhether the host gains any reciprocal benefit from these bac-teria. The discovery of Helicobacter spp. inhabiting the lowerbowel of the brushtail possum adds to the growing list ofanimals in which helicobacters have been found in the intes-tine. Given that Australian marsupials represent an evolution-ary group distinct from placental mammals, the findings of thisstudy are likely to contribute important information relating tothe evolutionary aspect of these bacteria in their hosts.

ACKNOWLEDGMENTS

This work was supported by a research grant awarded to Adrian Leeby the Faculty of Science, University of New South Wales, Australia.

REFERENCES

1. Amann, R. I., J. Stromley, R. Devereux, R. Key, and D. A. Stahl. 1992.Molecular and microscopic identification of sulfate-reducing bacteria in mul-tispecies biofilms. Appl. Environ. Microbiol. 58:614–623.

2. Berg, R. D. 1995. Bacterial translocation from the gastrointestinal tract.Trends Microbiol. 3:149–154.

3. Dewhirst, F. E., J. G. Fox, and S. L. On. 2000. Recommended minimalstandards for describing new species of the genus Helicobacter. Int. J. Syst.Evol. Microbiol. 50:2231–2237.

4. Fox, J. G. 2002. The non-H. pylori helicobacters: their expanding role ingastrointestinal and systemic diseases. Gut 50:273–283.

5. Fox, J. G., et al. 2002. Helicobacter marmotae sp. nov. isolated from livers ofwoodchucks and intestines of cats. J. Clin. Microbiol. 40:2513–2519.

6. Fox, J. G., et al. 2001. Isolation of Helicobacter cinaedi from the colon, liver,and mesenteric lymph node of a rhesus monkey with chronic colitis andhepatitis. J. Clin. Microbiol. 39:1580–1585.

7. Fox, J. G., et al. 2001. Novel Helicobacter species isolated from rhesusmonkeys with chronic idiopathic colitis. J. Med. Microbiol. 50:420–429.

8. Fox, J. G., et al. 1994. Helicobacter hepaticus sp. nov., a microaerophilicbacterium isolated from livers and intestinal mucosal scrapings from mice.J. Clin. Microbiol. 32:1238–1245.

9. Fox, J. G., et al. 1992. Helicobacter mustelae isolation from feces of ferrets:evidence to support fecal-oral transmission of a gastric Helicobacter. Infect.Immun. 60:606–611.

10. Fox, J. G., et al. 1995. Helicobacter bilis sp. nov., a novel Helicobacter speciesisolated from bile, livers, and intestines of aged, inbred mice. J. Clin. Mi-crobiol. 33:445–454.

11. Franklin, C. L., et al. 1996. Isolation of a novel Helicobacter species, Heli-cobacter cholecystus sp. nov., from the gallbladders of Syrian hamsters withcholangiofibrosis and centrilobular pancreatitis. J. Clin. Microbiol. 34:2952–2958.

12. Gorkiewicz, G., et al. 2003. Species-specific identification of campylobactersby partial 16S rRNA gene sequencing. J. Clin. Microbiol. 41:2537–2546.

13. Harper, C. G., et al. 2003. Isolation and characterization of novel Helico-bacter spp. from the gastric mucosa of harp seals Phoca groenlandica. Dis.Aquat. Organ 57:1–9.

14. Hodge, D. S., A. Borczyk, and L. L. Wat. 1990. Evaluation of the indoxylacetate hydrolysis test for the differentiation of campylobacters. J. Clin.Microbiol. 28:1482–1483.

15. Hume, I. D. 1999. Hindgut fermenters, the arboreal folivores, p. 149–204. InMarsupial nutrition. Cambridge University Press, Cambridge, MA.

16. Karmali, M. A., S. De Grandis, and P. C. Fleming. 1980. Antimicrobialsusceptibility of Campylobacter jejuni and Campylobacter fetus subsp. fetus toeight cephalosporins with special reference to species differentiation. Anti-microb. Agents Chemother. 18:948–951.

17. Kerle, A. 2001. Possums: the brushtails, ringtail and greater glider, p. 62–74.In T. J. Dawson (ed.), Australian natural history series. UNSW Press, Syd-ney, Australia.

18. Neilan, B. A., et al. 1997. rRNA sequences and evolutionary relationshipsamong toxic and nontoxic Cyanobacteria of the genus Microcystis. Int. J. Syst.Bacteriol. 47:693–697.

19. On, S. L., W. B. Holmes, and M. J. Sackin. 1996. A possibility matrix for theidentification of Campylobacter, Helicobacter and allied taxa. J. Appl. Bac-teriol. 81:425–432.

20. Patterson, M. M., M. D. Schrenzel, Y. Feng, and J. G. Fox. 2000. Gastritisand intestinal metaplasia in Syrian hamsters infected with Helicobacter auratiand two other microaerobes. Vet. Pathol. 37:589–596.

21. Patterson, M. M., et al. 2000. Helicobacter aurati sp. nov., a urease-positiveHelicobacter species cultured from gastrointestinal tissues of Syrian ham-sters. J. Clin. Microbiol. 38:3722–3728.

22. Phillips, M., A. Lee, and W. D. Leach. 1978. The mucosa-associated micro-flora of the rat intestine: a study of normal distribution and magnesiumsulphate induced diarrhoea. Aust J. Exp. Biol. Med. Sci. 56:649–662.

23. Riley, L. K., C. L. Franklin, R. R. Hook, Jr., and C. Besch-Williford. 1996.Identification of murine helicobacters by PCR and restriction enzyme anal-yses. J. Clin. Microbiol. 34:942–946.

24. Robertson, B. R. 1998. The molecular phylogeny and ecology of spiral bac-teria from the mouse gastrointestinal tract. University of New South Wales,Sydney, Australia.

25. Robertson, B. R., J. L. O’rourke, P. Vandamme, S. L. On, and A. Lee. 2001.Helicobacter ganmani sp. nov., a urease-negative anaerobe isolated from theintestines of laboratory mice. Int. J. Syst. Evol. Microbiol. 51:1881–1889.

26. Saitou, N., and M. Nei. 1987. The neighbor-joining method: a new methodfor reconstructing phylogenetic trees. Mol. Biol. Evol. 4:406–425.

27. Shen, Z., Y. Feng, F. E. Dewhirst, and J. G. Fox. 2001. Coinfection of entericHelicobacter spp. and Campylobacter spp. in cats. J. Clin. Microbiol. 39:2166–2172.

28. Schauer, D. B. 2001. Enterohepatic Helicobacter species, chapter 43. InH. L. T. Mobley, G. L. Mendez, and S. L. Hazell (ed.), Helicobacter pylori:physiology and genetics. ASM Press, Washington, DC.

29. Steinhauserova, I., J. CIesikova, K. Fojtikova, and I. Obrovska. 2001. Iden-tification of thermophilic Campylobacter spp. by phenotypic and molecularmethods. J. Appl. Microbiol. 90:470–475.

30. Thompson, J. D., T. J. Gibson, F. Plewniak, F. Jeanmougin, and D. G.Higgins. 1997. The CLUSTAL_X Windows interface: flexible strategies formultiple sequence alignment aided by quality analysis tools. Nucleic AcidsRes. 25:4876–4882.

31. Trivett-Moore, N. L., W. D. Rawlinson, M. Yuen, and G. L. Gilbert. 1997.Helicobacter westmeadii sp. nov., a new species isolated from blood culturesof two AIDS patients. J. Clin. Microbiol. 35:1144–1150.

32. Weisburg, W. G., S. M. Barns, D. A. Pelletier, and D. J. Lane. 1991. 16Sribosomal DNA amplification for phylogenetic study. J. Bacteriol. 173:697–703.

VOL. 77, 2011 HELICOBACTER SPP. FROM GIT OF BRUSHTAIL POSSUM 1587

on February 3, 2021 by guest

http://aem.asm

.org/D

ownloaded from