Research in the past decade has led to the recognition of Brachyspira (formerly Serpulina) pilosicoli as the primary etiologic agent of colonic spirochetosis (CS), an emerging cause of colitis in humans and animals. Attachment of spirochetes to the epithelial surface of the lower intestine is considered to be the hallmark of CS. However, because B. pilosicoli, B. aalborgi and unclassified flagellated bacteria are found singly or together in humans and non-human primates with CS lesions, attachment of spiral-shaped bacteria may not represent the same etiopathogenetic entity in all hosts. Moreover, North American opossums with CS are infected with B. aalborgi-like spirochetes together with flagellated bacteria, whereas B. pilosicoli is found alone in dogs, pigs, chickens and other species of birds with CS. Conversely, guinea-pigs with CS have unidentified spirochetes that may be B. pilosicoli or B. aalborgi. The pig model of CS suggests that attachment of B. pilosicoli to epithelial cells may be transient. By contrast, persistence of B. pilosicoli in the cecal and colonic crypt lumina, chronic inflammation caused by spirochetal invasion into the subepithelial lamina propria and translocation to extraintestinal sites may be more important than previously thought. This review describes the lesions seen in naturally occurring and experimentally induced CS of animals, and it sets the stage for future research into the pathogenic mechanisms of infection and colitis caused by B. pilosicoli.

Little is known about the outer membrane structure of Brachyspira hyodysenteriae and Brachyspira pilosicoli or the role of outer membrane proteins (OMPs) in host colonization and the development of disease. The isolation of outer membrane vesicles from B. hyodysenteriae has confirmed that cholesterol is a significant outer membrane constituent and that it may impart unique characteristics to the lipid bilayer structure, including a reduced density. Unique proteins that have been identified in the B. hyodysenteriae outer membrane include the variable surface proteins (Vsp) and lipoproteins such as SmpA and BmpB. While the function of these proteins remains to be determined, there is indirect evidence to suggest that they may be involved in immune evasion. These data may explain the ability of the organism to initiate chronic infection. OMPs may be responsible for the unique attachment of B. pilosicoli to colonic epithelial cells; however, the only B. pilosicoli OMPs that have been identified to date are involved in metabolism. In order to identify further B. pilosicoli OMPs we have isolated membrane vesicle fractions from porcine strain 95–1000 by osmotic lysis and isopycnic centrifugation. The fractions were free of contamination by cytoplasm and fla-gella and contained outer membrane. Inner membrane contamination was minimal but could not be completely excluded. An abundant 45-kDa, heat-modifiable protein was shown to have significant homology with B. hyodysenteriae Vsp, and monoclonal antibodies were produced that reacted with five B. pilosicoli-specific membrane protein epitopes. The first of these proteins to be characterized is a unique surface-exposed lipoprotein.

Investigations into the possible causes of colitis and typhlocolitis were carried out on 98 pig units in the United Kingdom between 1997 and 1999. Brachyspira pilosicoliwas identified most commonly, occurring as the suggested primary agent in 18% of the outbreaks but forming part of mixed infections in another 24% of outbreaks. The equivalent figures for other bacterial pathogens were: B. hyodysenteriae, 13% and 16%; Lawsonia intracellularis, 10% and 15%; Salmonella species, 6% and 12%; and Yersinia species, 4% and 10%. Unclassified Brachyspira species of unknown pathogenicity were identified in 12% of outbreaks. The 24 unclassified isolates divided into three groups on the basis of their phenotypic characteristics. In addition, there were 50 atypical Brachyspira species isolates that showed differences between their phenotypic characteristics and genetic identity based on sequence analysis of a section of the 23S rDNA gene. Four representative atypical isolates were found to be pathogenic as a result of an experimental oral challenge study in pigs.

The sensitivity of culturing Brachyspira hyodysenteriae was determined after sampling with swabs from porcine fecal specimens inoculated with tenfold dilutions of a field strain of these microbes. After storage of swabs, Brachyspira hyodysenteriaewas recovered throughout the first 3 weeks after inoculation from feces with more than 140 cells/g. Viable spirochetes could still be recovered after up to 83 days of storage from feces, with 1.4 × 106 cells or more per gram. Culture for Brachyspira spp. was performed on 285 rectal swabs, which were pooled in batches of five. The number of pooled samples positive for B. hyodysenteriae corresponded with the sum results of individual analysis of the corresponding collections of five samples. A PCR system based on the tlyA gene of B. hyodysenteriae was developed and tested on primary cultures of pooled samples. The results of the PCR assay showed a 97% correlation with the culture results. The prevalence of Brachyspira spp. was determined in five swine herds and found to be highest among breeding gilts and boars aged 13–16 weeks and among 6–12-week-old weaned pigs. In contrast, Brachyspiraspp. were only rarely found in sows, which may reflect the development of immunity by adult pigs to all species of the genus.

Swine dysentery (SD) caused by the intestinal spirochete Brachyspira hyodysenteriae is an economically important disease in pig-producing countries throughout the world. To date, no specific serologic assay is commercially available for the diagnosis of pigs with SD. Several serologic techniques have been identified in the past; however, these tests have all used either whole-cell proteins or lipopolysaccharide (LPS) as the antigen. Whole-cell antigens are plagued with false-positive reactions due to cross-reactivity with common proteins shared with other spirochetes. LPS antigens produce fewer false-positives; however, false-negatives may result due to LPS components being serogroup-specific. Generally, these techniques are useful for detecting infected herds, but are unreliable for the detection of individual infected pigs. In order to develop improved serologic tests it will be necessary to identify suitable diagnostic antigens, in particular immunogenic cell-surface structures which are specific to B. hyodysenteriae but common amongst different strains of the species. Recently, we identified and cloned a 30-kDa outer membrane lipoprotein (BmpB) which is specific to B. hyodysenteriae and is recognized by experimentally and naturally infected pigs. In this review we summarize the available serologic tests for SD, and speculate on the use of recombinant BmpB as an antigen for future development of an improved serologic test for SD diagnosis.

Brachyspira pilosicoli and B. innocens were isolated repeatedly from a herd of 60 sows which mostly produced feeder pigs but also raised some fattening pigs. Postweaning diarrhea had been a severe problem in this herd for years. The B. pilosicoli eradication plan was based on the general guidelines for elimination of B. hyodysenteriae, with some modifica-tions. The eradication measures were run in August 1997. In-feed medication with 200p.p.m. tiamulin lasted for 18–30 days, depending on the age group. The piggery unit was emptied, cleaned, disinfected and dried, and all worn surfaces were repaired. The animals were removed to temporary sheds situated 0–100m from the piggery unit. Only the sows and the boar returned to the piggery unit. All other pigs were sold from the sheds within 3 months after the eradication. Immediately after the eradication, the clinical postweaning diarrhea disappeared. The success of the program was monitored four times bacteriologically, and the last control sampling was in December 1999, 7 months after the total withdrawal of antimicrobial feed additives. The primary cultures from the last three samplings were also analysed with B. pilosicoli-specific PCR. All the samples were negative for B. pilosicoli. However, B. innocens could be isolated from each batch of samples. The analysis of B. inno-cens isolates by pulsed-field gel electrophoresis indicated that at least one genotype persisted in the herd. The clinical and laboratory findings suggest that the eradication of B. pilosicoli had succeeded in this herd

The pleuromutilins are the only antimicrobial agents with sufficient minimum inhibitory concentration (MIC) values left to treat swine dysentery in Sweden. Other antimicrobials are either not approved for use against swine dysentery or only partly active against Brachyspira hyodysenteriae. To date, in Sweden two pleuromutilins, tiamulin and valnemulin, are authorized for use in pigs. This study includes a comparison between MICs of tiamulin and valnemulin for Swedish field isolates of B. hyodysenteriae, as determined by broth dilution. For different isolates the MIC of tiamulin was between 0 and 8 times higher than that of valnemulin. No resistance to pleuromutilins was recorded (tiamulin MIC range 0.031–2 μg/ml, valnemulin MIC range ≤0.016–1 μg/ml). In vitro development of tiamulin resistance was also studied. Two B. hyodysenteriae and two B. pilosicoli strains became resistant to tiamulin following reiterated passages on agar containing tiamulin in increasing concentrations. The resistance emerged slowly and three of the strains that went through more than 60 passages increased their tiamulin MICs from 0.031–0.25 to more than 128 μg/ml. The tiamulin MIC for one B. hyodysenteriaestrain that went through 29 passages increased from 0.0125 to 4 μg/ml. One B. pilosicoli strain developed cross-resistance to valnemulin; the MIC increased from 0.25 to more than 64 μg/ml. The valnemulin MIC for one B. hyodysenteriae strain increased from 0.031 μg/ml to 32 μg/ml. Valnemulin MIC was not determined for the B. hyodysenteriae strain that only went through 29 passages. The valnemulin MIC of the other B. pilosicoli strain increased from 0.031 to 4 μg/ml.

Dietary supplementation with 6000 mg of Zn2+/kg of feed has been shown to modify the clinicopathologic expression of Brachyspira hyodysenteriaeinfection in a laboratory mouse model of swine dysentery. However, this concentration impaired the body weight gain of the mice. The purpose of the present study was to determine a minimal prophylactic concentration of feed-grade zinc compounds that would not affect the growth of mice challenge-exposed with B. hyodysenteriae. A total of 440, 6- to 8-week-old, C3H/HeN mice were allocated randomly to groups and fed either a defined diet or a defined diet containing either 1000, 2000 or 4000 mg/kg ZnO, ZnSO4 or zinc-methionine for 7 days before intra- gastric inoculation with B. hyodysenteriae. From days 7 to 35 after inoculation, mice in each group were necropsied at weekly intervals for determination of body weight, presence of B. hyodysenteriae in the cecum, and histological assessment of cecal lesions. Only ZnO fed at 2000 mg/kg had a prophylactic effect against B. hyodysenteriaeinfection without affecting the body weight gain of the mice. The prophylactic effect of Zn2+ against infection with B. hyodysenteriae was also affected by the relative concentration of Fe2+ and Zn2+/Fe2+ratio of the diet.

Brachyspira spp. were isolated from 21 of 32 sampled dogs (66%) in a colony of Swedish beagle dogs with a history of diarrhea and from 3 of 17 Swedish pet dogs (17%) with diarrhea. All Swedish isolates were weakly β-hemolytic and gave a negative indole reaction. Eighty-eight percent showed negative α-galactosidase and hippurate reactions, but a positive β-glucosidase reaction. Two isolates were hippurate positive with a negative β-glucosidase reaction. One additional German isolate diverged by showing a positive indole reaction in combination with a positive hippurate reaction. Sequencing of 16S rDNA indicated that the hippurate-positive isolates belonged to the species Brachyspira pilosicoli. Four representative isolates were examined using pulsed-field gel electrophoresis (PFGE) and compared with six reference strains and five porcine isolates of Brachyspira spp. The canine isolates clustered together in the PFGE analysis. Necropsy examination of a culture-positive B. pilosicoli colony-raised beagle dog revealed macro- and microscopical lesions of colitis with numerous spiral-shaped bacteria in the lumens of the crypts, in goblet cells and within the colonic epithelium.

This paper presents an overview of intestinal spirochete infections of chickens. It focuses particularly on studies in Australia, where recent surveys of 136 layer and broiler breeder flocks have revealed a high rate of infection (>40%) with intestinal spirochetes. Infection was not detected in broiler flocks. Approximately 50% of isolates from infected flocks were Brachyspira (Serpulina) intermedia or B. pilosicoli, with the other isolates being B. innocens, B. murdochii or the proposed species ‘B. pulli’. No isolates of B. alvinipulli were found. Intestinal spirochetes were significantly associated with wet litter problems and/or reduced egg production. Experimental infection of point-of-lay birds with either B. intermedia or B. pilosicoli caused reduced egg production, and, with B. intermedia, a significant increase in fecal moisture content. Infection with B. innocens caused no significant changes. In-water treatment of a flock with a mixed spirochete infection using lincospectin resulted in a slimy diarrhea lasting for 2–3 weeks, followed by absence of spirochetes for 3 months. Birds treated with tiamulin remained healthy, and had a reduced level of infection with intestinal spirochetes (30%) for 3 months. Trials are under way to test the efficacy of antimicrobials in point-of-lay chickens experimentally infected with either B. intermedia or B. pilosicoli.

Only limited data concerning the prevalence of intestinal spirochetes are available in game birds. This paper describes the prevalence and biochemical reactions of spirochetes isolated from 25 common partridges, 7 pheasants and 16 mallards originating from nine Swedish game-bird farms. The birds, which had been submitted for post-mortem examination due to various problems, showed a variety of underlying diseases. Additionally, fecal droppings from 22 common partridges, 20 pheasants and 20 mallards obtained at one of the farms were included in the study. Intestinal spirochetes were isolated from 85.4% of the game birds and from 71% of the fecal droppings. Seven biochemical types were identified. Seventeen per cent of all isolates were classified as Brachyspira pilosicoli and 3% as B. intermedia. One isolate showed strong -hemolysis and a positive indole reaction that is consistent with B. hyodysenteriae. In addition, three previously unknown biochemical types were found. Unclassified spirochetes in presumed mixed cultures were observed in 21% of all samples of fecal droppings. Histologic examination of spirochete-positive birds revealed numerous spirochetes in the lumen and crypts of the cecum, in some cases densely adhered by one end to the luminal surface. The significance of the findings is discussed.

Intestinal spirochetosis in humans (HIS) is a condition defined by the presence of a layer of spirochetes attached by one cell end to the colorectal epithelium. The pathologic significance of HIS is uncertain, but it has been linked to chronic diarrhea and other abdominal complaints. Two anaerobic intestinal spirochete species have been associated with HIS, namely Brachyspira pilosicoli and Brachyspira aalborgi. Brachyspira pilosicoli, which colonizes many animal species, is common (~30%) in the feces of people from developing countries, including Australian Aborigines, and in HIV+ patients and male homosexuals in Western societies. It is also commonly seen attached to the rectal mucosa of homosexual males. In other groups in Western societies both the presence of B. pilosicoli in feces and histologic HIS are uncommon (~1.5%). Brachyspira aalborgi is an extremely slow growing and fastidious spirochete, which previously had been isolated from an HIS patient in Denmark. Recent studies using polymerase chain reaction amplification of DNA from intestinal biopsies from a series of cases of HIS in the general Western population demonstrated that B. aalborgi, rather than B. pilosicoli, was the main spirochete species involved in these patients. This review outlines recent developments in the study of HIS and the two spirochete species, and identifies priorities for future research.

DNA was extracted from colonic biopsies of 33 patients with and three without evidence of intestinal spirochetosis (IS) in the large bowel. The biopsies were subjected to PCR. A pair of primers, generating a 207 bp fragment, were designed to detect specifically the 16S rDNA gene of Brachyspira. PCR products of the expected size were obtained from 33 samples with histologic evidence of IS. The PCR amplicons were used for sequencing. The sequences obtained were aligned to the corresponding 16S rRNA sequences of five type strains of Brachyspira. The sequences of 23 PCR products were 99–100% identical with the correspond-ing B. aalborgi type strain sequence. Two cases showed 99–100% sequence similarity with the type strain of B. pilosicoli P43/6/78. Six cases could not be referred to any of the known species of Brachyspira. Two PCR products gave incomplete sequences.

The prevalence of human intestinal spirochetosis (HIS) was determined histologically in 402 subjects from South Norway (Kristiansand) who had undergone colorectal resections during the period June 1991–June 1996. The figures were compared with those from a previous mid-Norwegian study (Trondheim)1 comprising 1205 patients with large bowel resections as well as colonoscopic mucosal biopsies examined in 1990. The studies showed a prevalence of HIS of 2.5% in mid-Norway and 3.0% in South Norway, and in both regions there was a considerable male predominance of HIS (70% and 75%, respectively). Furthermore, HIS was more often diagnosed by the pathologists in South Norway (25%) than in mid-Norway (6.5%).