SummarySegmented, filamentous bacteria (SFBs) form a group of bacteria with similar morphology and are identified on the basis of their morphology only. The relationships of these organisms are unclear as the application of formal taxonomic criteria is impossible currently due to the lack of an in vitro technique to culture SFBs. The intestine of laboratory animals such as mice, rats, chickens, dogs, cats and pigs is known to harbour SFBs. To see whether this extends to other animal species, intestines from 18 vertebrate species, including man, were examined. SFBs were detected with light microscopy in the cat, dog, rhesus monkey, crab-eating macaque, domestic fowl, South African claw-footed toad, carp, man, laboratory mouse and rat, wood mouse, jackdaw and magpie. These results suggest that non-pathogenic SFBs are ubiquitous in the animal kingdom. Among apparently identical animals, there was considerable variation in the degree of SFB colonization. It is suggested that SFB colonization could serve as a criterion of standardization of laboratory animals.

Segmented filamentous bacteria (SFBs) are apathogenic autochthonous bacteria in the murine small intestine that preferentially attach to Peyer's patch epithelium. SFBs have never been cultured in vitro. We have studied the effects of SFBs on the immune system of the host. Mice monoassociated with SFBs were compared with germ-free mice and with mice without SFBs but with a specific-pathogen-free (SPF) gut flora. SFBs versus no microbial flora raised the number of lymphoid cells in the lamina propria of the ileal and cecal mucosa, raised the number of immunoglobulin A (IgA)-secreting cells in the intestinal mucosa, produced elevated IgA titers in serum and intestinal secretions, and enhanced the concanavalin A-induced proliferative responses of mesenteric lymph node cells. The SPF flora had effects similar to but less pronounced than those mediated by SFBs. The results indicate that SFBs stimulate the mucosal immune system to a greater extent than do other autochthonous gut bacteria.

Segmented, filamentous bacteria (SFBs) are autochthonous, apathogenic bacteria, occuring in the ileum of mice and rats. Although the application of formal taxonomic criteria is imposible due to the lack of an in vitro technique to culture SFBs, microbes with a similar morphology, found in the intestine of a wide range of vertebrate and invertebrate host species, are considered to be related. SFBs are firmly attached to the epithelial cells of the distal ileal mucosa, their preferential ecological niche being the epithelium covering the Peyer's patches. Electron microscopic studies have demonstrated a considerable morphological diversity of SFBs, which may relate to different stages of a life cycle. Determinants of SFB colonization in vivo are host species, genotypical and phenotypical characteristics of the host, diet composition, environmental stress and antimicrobial drugs. SFBs can survive in vitro incubation, but do not multiply. On the basis of their apathogenic character and intimate relationship with the host, it is suggested that SFBs contribute to development and/or maintenance of host resistance to enteropathogens.

In a cross-over trial, five healthy dogs were fed a dry food without or with 1% (w/w) oligofructose to assess any oligofructose-induced effects on the faecal bacterial profile, nitrogen excretion and mineral absorption. The diets were given for a period of 3 weeks. Oligofructose feeding significantly raised the number of Bifidobacteria, Streptococci and Clostridia in faeces. The numbers of faecal anaerobic and aerobic bacteria were raised after ingestion of oligofructose. The faecal pH was unchanged. There was no effect of oligofructose feeding on the route of nitrogen excretion which was associated with a lack of effect on faecal ammonium and urinary urea excretion. It is suggested that the absence or presence of an effect of oligofructose on urinary and faecal nitrogen excretion depends on the background composition of the diet, in particular the content of non-digestible, fermentable carbohydrates. In the diets used, the content of non-digestible, fermentable carbohydrates was not measured. Both apparent magnesium and calcium absorption were significantly raised by oligofructose feeding, but phosphorus absorption was unaffected. The data presented may contribute to the qualification of the use of oligofructose in dog foods.

A technique is described so that mice mono-associated with non-cultivable, segmented filamentous bacteria (SFB's) can be produced for the first time. As SFB donors, mice were used which had an intestinal microflora consisting of both SFB's and bacteria of the genus Clostridium. Recipients were germ-free mice. It was demonstrated that the intraileal inoculation method was more effective than the orogastric route. Therefore, intestinal homogenates of donor mice were treated with filtered ethanol, diluted and administered intraileally to recipient mice. Evidence is presented that cage mates of the recipient mice were mono-associated with SFB's. The availability of these animals, i.e. in vivo monocultures of SFB's, allows taxonomic and functional characterization of SFB's, which was as yet not possible.

SummaryScanning electron micrographs are presented of the ileal epithelium of mice aged 5, 15, 20 and 25 days. During this period the villous pattern develops to full maturity. By the twentieth day of life a segmented filamentous microorganism colonizes the ileal epithelium and is firmly attached via a small segment. During the first days of colonization the segmented filamentous micro-organisms themselves are subcolonized by small rod-shaped bacteria, presumably lactobacilli. At the age of 25 days this subcolonization was no longer observed.

A cultured microflora obtained from the caecum of a 'normal' mouse was given to 4 groups of germfree mice and was supplied 1x, 2x, 3x and 4x respectively at 5-day intervals. Another group received a 10-7 dilution of the caecal flora while a group associated with an 'SPF' flora served as control. The difference (measured by 8 parameters) between mice supplied with the cultured flora or with a 10-7 dilution, both given once only, was small. Supplying the flora 3x resulted in more 'normal' mice compared with mice which received the flora once or twice. The caeca of specified-pathogen-free mice contained more bacteria per gram (microscopic bacterial count), less aerobic and anaerobic bacteria per gram (viable counts), while the yield as percentage of the microscopic bacterial count was lower as compared with the group to which a cultured flora was supplied 4 times.

The effects of the following changes throughout the association of germ-free mice with increasing numbers of anaerobic bacteria were studied: (i) elution patterns obtained by gel-filtration chromatography of caecal diffusates; (ii) concentration of beta-aspartylglycine in caecal and faecal contents; (iii) polypeptide patterns obtained by sodium dodecyl sulphate-polyacrylamide gel electrophoresis of caecal supernatants; (iv) free amino acid content of caecal supernatants; (v) faecal bile acids, analysed by gas-liquid chromatography; (vi) colonization-resistance. The results indicate that monitoring the normalization (association) process can be accomplished in several ways, but the level of colonization-resistance is most easily measured by high-voltage paper electrophoresis of faecal supernatants to determine the concentration of beta-aspartylglycine. During association, the concentration of beta-aspartylglycine decreased and became undetectable after association with 40 to 50 different strains of bacteria. There was a good negative correlation between the level of colonization-resistance and the concentration of beta-aspartylglycine.