| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1Division of Veterinary and Biomedical Sciences, Murdoch University, Murdoch, Western Australia 6150, Australia 2,3Division of Microbiology and Infectious Diseases, The Western Australian Centre of Pathology and Medical Research2 and Department of Microbiology, The University of Western Australia3, Nedlands, Western Australia 6009, Australia#dReceived 5 November 2002 Accepted 2 March 2003
Correspondence: David J. Hampson (d.hampson{at}murdoch.edu.au)
 |
Abstract |
|---|
 TOP Abstract INTRODUCTIONMETHODSRESULTSDISCUSSIONNOTE ADDED IN PROOFREFERENCES |
|---|
|
INTRODUCTION |
|---|
|
TOPAbstractINTRODUCTIONMETHODSRESULTSDISCUSSIONNOTE ADDED IN PROOFREFERENCES |
|---|
Of the two spirochaete species, B. aalborgi is the more slowly growing and fastidious, requiring at least 2–3 weeks for isolation (Hovind-Hougen et al., 1982). Consequently, B. aalborgi has been relatively poorly studied and detection has relied on the use of PCR amplification of B. aalborgi DNA from colorectal biopsies (Mikosza et al., 1999, 2001a; Kraaz et al., 2000) or faeces (Mikosza et al., 2001b). Isolation of B. aalborgi has been reported only three times, in each case following culture of fresh colonic biopsy specimens taken from Scandinavian patients (Hovind-Hougen et al., 1982; Kraaz et al., 2000; Jensen et al., 2001). The isolation medium in these studies was tryptose soy agar supplemented with 10 % bovine blood (BB), containing 400 µg spectinomycin and 5 µg polymyxin ml-1. To date, B. aalborgi has not been isolated from human faeces, although the organism was recently isolated from the faeces of captive non-human primates (Munshi et al., 2003). The lack of isolates has hampered comprehensive characterization and pathogenicity studies with this organism.
The purposes of the current study were to identify a solid medium that enhanced the growth of B. aalborgi, to modify this as an improved isolation medium and then to use it to isolate B. aalborgi from the faeces of human patients.
|
METHODS |
|---|
|
TOPAbstractINTRODUCTIONMETHODSRESULTSDISCUSSIONNOTE ADDED IN PROOFREFERENCES |
|---|
Reference strains and preparation of inocula.
B. aalborgi strains 513AT (=ATCC 43994T) (Hovind-Hougen et al., 1982) and W1 (Kraaz et al., 2000) were obtained from the culture collection held at the Reference Centre for Intestinal Spirochaetes at Murdoch University. The reference strains were routinely cultured on Columbia base agar containing 6 % equine blood (BA), incubated for 10 days at 37 °C in an anaerobic jar with an atmosphere of 94 % H2 and 6 % CO2 generated with a Gaspak Plus envelope (BBL). For media assessment, spirochaete growth was resuspended in PBS or Brucella broth using a sterile cotton-tipped swab and the suspension was returned to anaerobic conditions until inoculation. The organisms were counted in a Helber counting chamber, the concentration was adjusted to 1 x 108 organisms ml-1 and their viability was assessed by inoculation on BA using the spread plate method. Volumes of 50 µl were added to plates and the suspensions were streaked for single colonies in a standard manner.
Growth media.
The following agar bases, all acquired from Oxoid, were used in the study: brain heart infusion agar (BHIA), Columbia base agar, Brucella base agar plus vitamin K (1 mg ml-1) and haemin (5 mg ml-1), Wilkins–Chalgren agar and anaerobe basal agar. In addition, trypticase soy agar (TSA) from Becton Dickinson was used by itself and as a base for a modification of Kunkle's broth medium (Kunkle et al., 1986). The latter contained 2 % fetal calf serum (FCS), 1.5 % alcoholic cholesterol solution, 1 % (w/v) yeast extract, 0.5 % (w/v) glucose, 0.2 % (w/v) NaHCO3 and 0.05 % (w/v) cysteine hydrochloride monohydrate. Primary supplements that were tested with each of the agar bases were 5 and 10 % ovine blood (OB), 10 % BB and 10 % defibrinated BB. Secondary supplements were 2 % FCS, 1.5 % ethanolic cholesterol solution and 5 % pig faeces extract. The latter was prepared by emulsifying one part faeces obtained from normal adult pigs that had not received antimicrobials in four parts sterile PBS, stirring the emulsion for 24 h at 4 °C and then centrifuging for 1 h at 9950 g at 4 °C. The supernatant was stored at -20 °C until required and filter-sterilized (450 nm) before being added to agar.
Additional parameters assessed included 30 min pre-reduction of the plates prior to inoculation and incubation either in an anaerobic jar using the Gaspak Plus system or in an anaerobic chamber (80 % N2, 10 % H2, 10 % CO2) (Don Whitley Scientific). Growth in anaerobic jars was compared at 37, 38.5 and 42 °C. The influence of pH was assessed by adding 4 ml 10 mM KCl to 2 ml set TSA 10 % BB (Carson et al., 2000) and the pH was measured after 30 min incubation at room temperature. BA was used as a control medium in all tests.
Isolation media.
The susceptibility of B. aalborgi strains to a range of antimicrobials for potential use as selective agents was determined with Microring MWAC and MWAN discs (Medical Wire & Equipment). The selective compounds and their disc strengths are listed in Table 1. In addition, spectinomycin (S), vancomycin (V), colistin (C), rifampicin (R) and polymyxin B (P) (Sigma) were incorporated directly into the media. Susceptibility to bile (dried ox bile, pure for microbiology; Merck) was assessed by making bile discs containing 5, 10 or 15 mg bile (Weinberg et al., 1983) and placing these on BA swabbed with the suspensions.
|
The selective capacity of the media was examined by adding 1 ml bacterial suspension (108 cells) to 1 g faeces from a healthy donor who had been confirmed free of B. aalborgi and B. pilosicoli by faecal PCR. The suspension was mixed thoroughly and 10 µl was added to a plate with a standard loop and streaked as above. Selection of B. aalborgi from seeded faeces was tested on TSA 10 % BB with all combinations of spectinomycin (400 µg ml-1), vancomycin (25 µg ml-1) and colistin (25 µg ml-1). Additionally, spectinomycin (400 µg ml-1) plus all combinations of polymyxin B (5 µg ml-1), rifampicin (12.5 µg ml-1) and colistin (25 µg ml-1) was tested. The lower limit of selection of B. aalborgi from seeded faeces on spectinomycin plus polymyxin B (SP) was determined by performing 10-fold dilutions on the bacterial suspension. Faeces were seeded, mixed and streaked as above.
Both growth and isolation media were examined for B. aalborgi after 10, 14, 18 and 21 days incubation. Growth was assessed by recording colony size and scoring the extent of growth along streaks on the test plates. A grading method was developed where the best medium was scored ‘1’ at a given time-point, the next best medium ‘2’ and so on, and a mean was determined for each medium after 21 days. All procedures were repeated at least once.
Isolation.
The efficacy of the best medium was tested by attempting isolation of B. aalborgi from fresh faecal samples from 60 apparently healthy individuals taking part in a study examining carriage of intestinal spirochaetes, using faecal PCR as described by Mikosza et al. (2001b). Five of these individuals had positive faecal PCRs for B. aalborgi (data not shown). Isolation of B. aalborgi was attempted on BHIA 10 % BB SP, incubated in an anaerobic jar at 37 °C for 21 days before being exposed to oxygen. Spirochaete growth was subcultured to BA and colonial and phase-contrast morphologies were recorded. For electron microscopy, the cultured spirochaetes were scraped from the plates and resuspended in distilled water. Drops of the suspensions were placed on Formvar-coated reinforced grids and stained with phosphotungstic acid (3 %, pH 7.2). Grids were examined in a Philips CM100 Biotwin transmission electron microscope.
Spirochaete identity was confirmed by adding cells directly to PCRs designed to amplify portions of the 16S rRNA genes of B. aalborgi or B. pilosicoli (Mikosza et al., 1999, 2001a). Cycling for B. aalborgi was carried out at an annealing temperature of 48 °C. The PCR product was sequenced in both directions using a commercially available cycle sequencing kit (ABI PRISM BigDye dye terminator cycle sequencing ready reaction kit, version 1.0; Applied Biosystems), according to the manufacturer's instructions. The sequence was analysed on the ABI PRISM 310 Genetic Analyzer (Applied Biosystems). The sequence data obtained were aligned and compared with 16S rDNA sequences of the B. aalborgi and B. pilosicoli type strains by BLAST search with the National Centre for Biotechnology Information server at the National Library of Medicine. Multiple sequence alignments were performed by the CLUSTAL method using BioManager with the Angis server.
|
RESULTS |
|---|
|
TOPAbstractINTRODUCTIONMETHODSRESULTSDISCUSSIONNOTE ADDED IN PROOFREFERENCES |
|---|
As a supplement, 10 % BB was superior to 10 % OB, which, in turn, allowed larger colonies and faster growth than 5 % OB, where a maximum size of only 0.5 mm was achieved after 21 days. Additional supplements of 2 % FCS, 1.5 % cholesterol and 5 % pig faecal extract offered no growth advantage. Defibrination of BB or pre-reduction of the media did not improve growth of B. aalborgi. Growth of B. aalborgi did not occur at 42 °C. Colonies on BHIA incubated at 38.5 °C were at best only 0.2 mm larger at some time-points than colonies on BHIA incubated at 37 °C. The best pH for growth of B. aalborgi on TSA 10 % BB was pH 7. Growth to pH 8 was satisfactory, but agar below pH 6.5 would not withstand prolonged incubation. The pH of BHIA 10 % BB was 7.6. For all media, incubation in an anaerobic jar resulted in growth of larger colonies, by 0.2–0.5 mm, compared with incubation in an anaerobic chamber.
Selective media
Both reference strains of B. aalborgi were susceptible to bile, metronidazole, penicillin G, kanamycin and vancomycin and resistant to spectinomycin, polymyxin B, colistin, erythromycin and rifampicin. Equivocal results were obtained for sodium polyethol sulfonate and novobiocin for strain W1, while strain 513AT was susceptible to these compounds. On media with antimicrobials incorporated, clear growth of B. aalborgi occurred on all plates except those containing vancomycin, although colonies were frequently 0.2 mm smaller than on the medium without antimicrobials. Selection of B. aalborgi from seeded faeces was achieved on media containing S, SP, SC, SR, SRC, SPR, SPC and SRPC, and no single medium produced consistently superior growth. The lower limit of detection of B. aalborgi from seeded faeces on SP was around 200 c.f.u. per 10 µl seeded faeces or 2 x 104 c.f.u. g-1.
Spirochaetes were isolated from one of the five faecal samples that were previously shown by PCR to contain B. aalborgi. After 21 days incubation, two types of weakly ß-haemolytic colonies were seen on BHIA 10 % BB: these were grey, spreading colonies of 1.2 mm diameter with a raised, pinpoint centre inside an irregular flat edge (type A) or smooth, pinpoint colonies (< 0.1 mm diameter) (type B). On subculture, a mixture of the two colony types persisted; however, the type A colonies predominated.
PCR and sequencing
PCR of both colony types from the primary isolation plate confirmed their identity as B. aalborgi. A band of 472 bp corresponding to the expected product size in the B. aalborgi 16S rRNA PCR was obtained, while no band was obtained in a 16S rRNA PCR for B. pilosicoli. Sequencing of the PCR product from each colony type showed 99 % similarity with the sequences for B. aalborgi 513T, W1 and nine of the uncultured Brachyspira isolates sequenced by Pettersson et al. (2000). For each colony type, polymorphisms occurred at the base equivalent to position 416 in Escherichia coli (Brosius et al., 1978), where a guanidine residue was replaced by thymidine, and at the base equivalent to position 436, where thymidine was replaced by cytosine.
Phase-contrast and electron-microscope morphology
Under the phase-contrast microscope, the spirochaetes appeared as short, helical cells with a pronounced flexuous motility. Cells from the two colony types had a similar appearance. Electron microscopy revealed two cell types. One was a plump cell type, varying from 2.0 to 6.0 µm long and about 250 nm wide. Cells were typically ‘S'- or comma-shaped with a 2 µm wavelength. The other cell type was similar in length, shape and amplitude, but was only approximately 120–130 nm in width (Fig. 1). A mixture of cell types was observed in the large, grey, spreading colonies, whilst only the thin cell type was observed in the pinpoint colonies. Both cell types had four periplasmic flagella inserted in a row at each end of the cell.
|
|
DISCUSSION |
|---|
|
TOPAbstractINTRODUCTIONMETHODSRESULTSDISCUSSIONNOTE ADDED IN PROOFREFERENCES |
|---|
The best base medium for growth of B. aalborgi was BHIA, supplemented with 10 % BB, whilst selection of B. aalborgi from seeded faeces was achieved with media containing S, SC, SP, SR, SRC, SPR, SPC or SRPC. The medium used in previous isolations of B. aalborgi was TSA 10 % BB SP (Hovind-Hougen et al., 1982; Kraaz et al., 2000; Jensen et al., 2001). As the performance of these combinations appeared equal in this study, SP was chosen because of the history of successful isolation in previous studies. In this investigation, BHIA 10 % BB supported better growth than TSA 10 % BB, and it is possible that the additional nutrients provided by the BHIA allowed the B. aalborgi strain that was isolated in this study to out-compete the other gut microbiota on the selective plate.
TSA with 5 % OB, S (400 µg ml-1), V (25 µg ml-1) and C (25 µg ml-1) has been used to isolate B. pilosicoli from human faeces (Lee & Hampson, 1992). Results from the current study indicate that this medium does not support the growth of B. aalborgi, presumably as a result of the strains being susceptible to vancomycin. Poor growth of B. aalborgi W1 on a very similar medium has been reported previously (Kraaz et al., 2000). These results indicate that studies in which TSA 5 % OB SVC is used to isolate human intestinal spirochaetes are likely to fail to isolate B. aalborgi.
Pre-reduction of media did not influence growth of B. aalborgi; however, tests were carried out on laboratory-adapted strains. It is possible that pre-reduction of media might have allowed the isolation of other strains in the samples tested. Likewise, growth of these strains was barely improved at 38.5 °C, despite a previous report of optimal growth at this higher temperature (Hovind-Hougen et al., 1982). This may again be due to the laboratory adaptation of strains, or the difference observed by Hovind-Hougen et al. (1982) was possibly more marked because of the use of TSA as the base medium. The superior growth achieved in the anaerobic jars compared with an anaerobic chamber was unexpected. Humidity levels tended to be higher in jars, and the gas composition differed between the two atmospheres. In particular, anaerobic jars may contain traces of oxygen, and it is known that the growth of Brachyspira hyodysenteriae is enhanced by the presence of 1 % oxygen, associated with the production of NADH oxidase (Stanton & Lebo, 1988; Stanton, 1989). B. aalborgi also has a gene encoding NADH oxidase (Mikosza et al., 1999). It is possible that some or all of these factors may have impacted upon growth of B. aalborgi. It is also likely that these organisms are quite sensitive to changes in the environment, as it was noted that growth of B. aalborgi was inhibited in the presence of large numbers of contaminating organisms.
Antimicrobial susceptibilities of B. aalborgi were not determined, as growth of the organism was poor and inconsistent on the NCCLS-recommended media for testing anaerobes (Summanen et al., 1993). Results of the disc sensitivity tests did not elicit any more compounds that were considered useful for selective media, as B. aalborgi was susceptible to most compounds tested. Resistance to colistin and rifampicin was already suspected and was thus utilized in this study, and erythromycin is not suitable for use under anaerobic conditions. Demonstration of metronidazole susceptibility at 5 µg per disc corresponded to reports of the successful use of this agent in a clinical setting, and this result confirms its role as the drug of choice in intestinal spirochaetosis (Peghini et al., 2000; Heine et al., 2001).
In common with previous descriptions (Hovind-Hougen et al., 1982; Kraaz et al., 2000), two colony types grew from the positive faecal sample. The colony types were similar to those described previously, and the size and morphology of one of the isolates was consistent with previous descriptions of B. aalborgi (Hovind-Hougen et al., 1982; Kraaz et al., 2000; Jensen et al., 2001). Sequencing of PCR products from the two colony types gave identical 16S rDNA sequences, suggesting that the two types were identical in this region. An unusual feature, however, was the identification of some cells that were much thinner by electron microscopy than the ‘typical’ cells. The smaller colony types appeared to be made up of these thinner types, although they were also present in the larger colonies. The differences in cell size might have been an artifact, the result of the presence of two distinct types of organism, or might reflect a genuine difference in the phenotype of a given strain. These possibilities require further investigation.
Isolation of B. aalborgi was achieved from only one of five PCR-positive faecal samples. The lower limit of selection from seeded faeces was quite poor, at 2 x 104 c.f.u. (g faeces)-1, but was comparable to that described for faecal PCR (Mikosza et al., 2001b). While the remainder of these samples may have contained non-viable organisms that therefore were only detected by PCR, these results suggest that further modifications may be required to optimize media and conditions for isolation of all B. aalborgi. The individual from whom the B. aalborgi isolate was obtained was self-diagnosed as having ‘chronic diarrhoea', but was unfortunately lost to follow-up because he moved overseas shortly after the faecal sample was submitted. A faecal sample from his female partner was PCR- and culture-negative for B. aalborgi. The availability of an improved isolation medium should now facilitate further epidemiological studies on the possible role of this spirochaete in human colorectal disease.
|
NOTE ADDED IN PROOF |
|---|
|
TOPAbstractINTRODUCTIONMETHODSRESULTSDISCUSSIONNOTE ADDED IN PROOFREFERENCES |
|---|
|
Acknowledgments |
|---|
|
REFERENCES |
|---|
|
TOPAbstractINTRODUCTIONMETHODSRESULTSDISCUSSIONNOTE ADDED IN PROOFREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  K. R. Margawani, I. D. Robertson, C. J. Brooke, and D. J. Hampson Prevalence, risk factors and molecular epidemiology of Brachyspira pilosicoli in humans on the island of Bali, Indonesia J. Med. Microbiol., April 1, 2004; 53(4): 325 - 332. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. S.J. Mikosza, M. A. Munshi, and D. J. Hampson Analysis of genetic variation in Brachyspira aalborgi and related spirochaetes determined by partial sequencing of the 16S rRNA and NADH oxidase genes J. Med. Microbiol., April 1, 2004; 53(4): 333 - 339. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| INT J SYST EVOL MICROBIOL | J MED MICROBIOL | MICROBIOLOGY | J GEN VIROL | ALL SGM JOURNALS |
