Germfree Guinea Pigs Research Articles

Gnotobiotics: Past, present and future.

Gnotobiotics or gnotobiology is a research field exploring organisms with a known microbiological state. In animal research, the development of gnotobiotics started in the late 19th century with the rederivation of germ-free guinea pigs. Cutting-edge achievements were accomplished by scientists in the Laboratories of Bacteriology at the University of Notre Dame (LOBUND). The primary goals of gnotobiotics were not only the development of the equipment required for long-term husbandry but also phenotypic characterization of germ-free animals. The first isolators were designed by Reynolds and Gustafsson as rigid-wall stainless steel autoclave-like chambers, which were subsequently replaced by Trexler's flexible-film polyvinyl plastic isolators. Flexible-film or semi-rigid isolators are commonly used today. The long-term maintenance of gnotobiotic rodents is performed in positive-pressure isolators. However, to facilitate gnotobiotic experimental procedures, short-term husbandry systems have been developed. Gnotobiotic animal husbandry is laborious and requires experienced staff. Germ-free animals can be rederived from existing rodent colonies by hysterectomy or embryo transfer. The physiology and anatomy of germ-free rodents are different from those of specified pathogen-free (SPF) rodents. Furthermore, to guarantee gnotobiotic status, the colonies need to be regularly microbiologically monitored. Today, gnotobiotics provides a powerful tool to analyse functional effects of host-microbe interactions, especially in complex disease models. Gnotobiotic models combined with 'omics' approaches will be indispensable for future advances in microbiome research. Furthermore, these approaches will contribute to the development of novel therapeutic targets. In addition, regional or national gnotobiotic core facilities should be established in the future to support further applications of gnotobiotic models.

Generating and Analyzing Germ-Free Mice.

The normal gut microbiota has evoked many investigators' interest over the years and the pioneering work of James Reyniers in the 1920s generated the first germ-free guinea pigs. Comparing the physiology between germ-free and conventionally raised animals has provided invaluable insights on how the gut microbiota affect host biology. Today we know that the gut microbiota modulate the immune system, epithelial cell proliferation, intestinal angiogenesis, hormone production, energy absorption, and behavior. Furthermore, recent data have demonstrated that obesity is associated with an altered gut microbiota, and a direct role for the microbiota in disease development was demonstrated by the use of germ-free mice. Here we are presenting protocols for maintaining and generating germ-free mice. Curr. Protoc. Mouse Biol. 2:307-316 © 2012 by John Wiley & Sons, Inc.

Ascorbic acid catabolism by gut microflora: studies in germ-free and conventional guinea pigs

The role of gut microflora in ascorbic acid catabolism was investigated in both conventional and germ-free guinea pigs. In vitro studies demonstrated extensive degradation of the vitamin by fresh feces, cecal, and colonic contents of conventional guinea pigs. Direct injection of [1- 14C] ascorbic acid into the cecum of conventional guinea pigs in vivo yielded a 70% recovery of the label as respiratory 14CO 2 within 6 hr compared with only 5% recovery following injection into the virtually sterile peritoneum in a comparable group of conventional guinea pigs. Thus, ascorbic acid not absorbed prior to reaching the lower gastrointestinal tract stands to be extensively decarboxylated by microflora in the cecum. In a companion study of germ-free guinea pigs, 10% of an administered dose of [1- 14C] ascorbic acid was expired as 14CO 2 within 36 hr post-injection following intraperitoneal injection compared with 16% recovery in a matched group of conventional animals injected at the same site. Results of this series of studies suggest that hepatic decarboxylation and gut microflora, in tandem, contribute to ascorbic acid decarboxylation in this species.

Role of Intestinal Microflora in the Metabolism of Vitamin B-6 and 4′-Deoxypyridoxine Examined Using Germfree Guinea Pigs and Rats

In previous work identification of urinary metabolites of 4'-deoxypyridoxine which had been oxidized in the 5'-position and long-term dilution of labeled urinary metabolites with unlabeled molecules suggested possible microbial contributions. In the current studies germfree guinea pigs were able to convert 4'-deoxypyridoxine to 4'-deoxy-5-pyridoxic acid demonstrating that the ability to oxidize the 5'-position is not restricted to microorganisms. Labelling curves for urinary pyridoxic acid in rats continuously fed [14C]pyridoxine since weaning were similar in conventional and germfree animals indicating that any vitamin B-6 synthesized in the intestinal tract was not readily absorbed and metabolized. Therefore, coprophagy did not make a detectable contribution to vitamin B-6 metabolism in rats receiving a nutritionally complete diet. The difficulty in achieving comparable labeling in adult animals is probably due to very slow turnover of portions of the vitamin B-6 pool and not to microbial production of vitamin B-6. The total pool calculated from the radioactivity in the germ-free rats averaged 16.2 +/- 0.8 nmol vitamin B-6 compounds/g body wt. Only 10% of the ingested label was recovered in the feces. In addition, only about 50% of the label excreted in the urine appeared as 4-pyridoxic acid in rats. These observations suggest that it may be difficult to quantitate the total urinary and fecal excretion of ingested vitamin B-6 without using tracers.

The 'normalization' of germ-free guineapigs with host-specific caecal microflora.

Hysterectomy-derived germ-free guineapigs were given colonization-resistant caecal flora from mice (mCRF) or microflora obtained from the caecum of an antibiotic-decontaminated conventional guineapig (gpCRF) and compared with guineapigs raised conventionally with the sow. Body weight and the following intestinal parameters were determined for the groups: colonization resistance (CR) to Escherichia coli, relative caecal weight (RCW), beta-aspartylglycine (faeces), volatile fatty acids (caecum) and bile acids (faeces). mCRF guineapigs showed values quite different from control animals for CR and RCW, indicating the unsuitability of mouse CRF for normalizing guineapigs. In gpCRF guineapigs CR and RCW values were comparable with controls, indicating the suitability of the guineapig flora for normalizing guineapigs. mCRF guineapigs housed with gpCRF guineapigs, showed an improvement in CR and RCW, yielding values found in control animals.

Enhancement of experimental ulcerative colitis by immunization with Bacteroides vulgatus.

Previous studies with the guinea pig model for ulcerative colitis have shown that the inducing agent, carrageenan, does not provoke ulcerations in germfree guinea pigs, but animals monoassociated with Bacteroides vulgatus develop cecal ulcerations when fed carrageenan. In an effort to define the role of B. vulgatus in this model system, conventional guinea pigs were immunized with B. vulgatus before carrageenan treatment. Immunized animals developed both circulating antibody and positive skin tests for the homologous antigen. A comparison of B. vulgatus-immune and nonimmune carrageenan recipients after 30 days of carrageenan treatment revealed ulcerations of the ceca and large intestine in both groups, but the immune group developed more severe lesions than the nonimmune group. The most severe intestinal lesions were detected in B. vulgatus-immune animals fed both carrageenan and daily doses of viable B. vulgatus. Immune recipients of B. vulgatus alone did not develop ulcerations, but showed histological evidence of intestinal inflammation. Lesions observed in B. vulgatus-immune animals were most severe in the colon and rectum, in contrast to more severe cecal lesions detected in nonimmune carrageenan recipients. Similar experiments with Bacteroides fragilis as an antigen for immunization showed no difference between immune and nonimmune groups. These results suggest that immunization with B. vulgatus enhances the severity of carrageenan-induced colitis.

Toxin-induced cell membrane injury in guinea pigs given lincomycin

Guinea pigs treated with lincomycin developed colitis, acute cholecystitis and abnormalities in red blood cell morphology. The present study was designed to study the production of clostridial toxins after lincomycin treatment. Lincomycin produced abnormalities in conventional but not in germ-free guinea pigs. Clostridium difficile was cultured from cecal contents of conventional guinea pigs treated with lincomycin. Cecal filtrate from sick guinea pigs was subjected to Sepharose 4B-CL and Sephadex G-200 column chromatography, yielding a partially purified toxin. Both cecal filtrate and partially purified toxin samples contained a heat labile substance which was cytotoxic to human lung fibroblast Wl-38 cells, and which was neutralized by Clostridium sordelli antitoxin. Toxin samples given orally or intraperitoneally killed normal guinea pigs. Finally, toxin samples induced red cell membrane changes in vitro as well as producing features of acute inflammation in healthy explants of guinea pig cecum and gallbladder in organ culture. Lincomycin treated guinea pigs produced Clostridium difficile toxin(s) which in turn caused diffuse cell membrane injury.

Production of experimental ulcerative colitis in gnotobiotic guinea pigs with simplified microflora

Conventional guinea pigs provided with a solution of 5% (wt/vol) degraded carrageenan as the sole source of oral fluids developed ulcerations of their ceca and large intestines within 30 days. Similar lesions were not detected in germfree guinea pigs treated in an identical manner, suggesting that an intestinal microflora was necessary for development of intestinal lesions. To simplify the bacterial flora required for production of cecal ulcerations, 10 pools consisting of 10 bacterial strains each were isolated from the cecal microflora of carrageenan-treated animals. Groups of germfree guinea pigs were associated with 2 of the 10 pools by orogastric intubation and observed for development of disease. One-half of each group was treated with carrageenan. The two bacterial pools were characterized by the presence of cytopathic effects for WI-38 and Vero cells, increased chemotactic activity, and increased concentrations of long-chain fatty acids. The results indicated that animals associated with those two pools developed cecal ulcerations during carrageenan treatment. Preliminary results also indicated that cecal ulcerations developed in germfree animals mono-associated with a strain of Bacteroides vulgatus isolated from one of the pools, regardless of whether carrageenan was administered, suggesting a bacterial involvement in disease development in the absence of carrageenan treatment.

Trichophyton mentagrophytes Dermatophytosis in Germfree Guinea Pigs

Primary and secondary Trichophyton mentagrophytes dermatophytosis was studied in germfree and conventionally-reared Strain 2 guinea pigs. Although the onset and early development of the primary cutaneous lesions appeared similar in both groups, the T. mentagrophytes monoassociated guinea pigs exhibited more severe skin ulcerations and took twice as long to heal as their conventionally-reared counterparts. Cutaneous reinfection of T. mentagrophytes monoassociated guinea pigs was also protracted; however, these lesions healed in about the same time as a primary infection on conventionally-reared guinea pigs. Germfree guinea pigs, sensitized by cutaneous infection with T. mentagrophytes, manifested 3 correlates of systemic cell-mediated immunity: (1) delayed-type hypersensitivity to intracutaneous injection of trichophytin antigen, (ii) in vitro blastogenesis of spleen and lymph node cells to polyclonal mitogens and Trichophyton antigens, and (iii) allergic contact dermatitis 48 hr following cutaneous reinfection. In an additional experiment, we observed that the time course and severity of a primary T. mentagrophytes infection of germfree guinea pigs was reduced by prior feedings with heat-killed T. mentagrophytes culture. These experiments confirm that the normal microbial skin flora is not required for initiation, development or clearance of T. mentagrophytes dermatophytosis. The disease was protracted and severe in T. mentagrophytes monoassociated guinea pigs, but was abbreviated following either active infection or feeding heat killed mentagrophytes culture. These results support the hypothesis that cutaneous T. mentagrophytes infection is a cell-mediated hypersensitivity disease similar to contact dermatitis.

Interactions between Entamoeba histolytica and complement

THE pathogenic mechanisms in amoebiasis are not fully understood. Histopathological studies of infected tissues have shown lysis of host cells near the parasites, particularly in the liver where abscesses are formed. In the intestine, inflammatory reactions are often slight, unless there is secondary bacterial involvement. The infiltrating cells belong mainly to the lymphocyte and macrophage cell series, and neutrophil leukocytes are sparse1, suggesting that there is a low-grade chronic inflammation. Tacheuchi and Phillips2 have made systematic studies of experimentally infected germ-free guinea pigs, which have revealed the early events in invasive amoebiasis. When the amoebae had not yet made contact with the colonic epithelial cells, the latter showed degenerative changes such as loss of brush borders, swelling of mitochondria and dilatation of the endoplasmic reticulum. At the same time polymorphonuclear leukocytes (PMN) as well as some macrophages were seen to infiltrate the lamina propria, and extravasated erythrocytes were observed in the same areas. A high proportion of the PMN showed signs of degeneration, including disruption of the cell membrane. As changes were observed some distance from the parasites, Tacheuchi and Phillips2 concluded that the parasites exert their effects by means of a soluble mediator(s), possibly a toxin elaborated by the parasites. Another possibility is that enzymes released from amoebae damage both epithelium and PMN. A third explanation would be that the amoebae can interact with host material to generate factors which are chemotactic for leukocytes and can also exert toxic effects. A likely candidate would be the complement system. For this reason, the experiments described here were undertaken to analyse interactions between amoebae and the complement system. IgG antibody fractions were prepared from sera of individuals suffering from amoebiasis, and unheated sera from normal humans or experimental animals were used as a source of heat-labile complement components. We found that Entamoeba histolytica activates complement by the alternative pathway, and can be lysed by the reaction products.

Pathways of penetration ofEscherichia coli 055 through the intestinal wall in germfree and ordinary animals

Permeability of the intestinal barrier in germfree animals for bacteria of the enteropathogenic strainEscherichiacoli 055 was studied. Germfree and ordinary rats and guinea pigs were used. During the first 24 h after peroral infection withE.coli 055 an increase in the bacteriemia was observed in the germfree animals, whereas in the ordinary animals the bacteriemia was transient in character. Electron-microscopic investigation of the intestinal mucosa of the germfree animals revealed disturbances of intercellular contacts and the formation of spaces containing numerous microorganisms between the enterocytes. In the germfree animals more marked changes also were found in the microvessels of the mucosa. In ordinary animals ingestion and digestion of theE.coli cells by enterocytes and leukocytes was observed. Disturbances of the intestinal barrier discovered in the germfree animals explain the increased bacteriemia in these animals in the absence of a microflora. The results point to the important role of the microbial factor in the formation of the intestinal barrier.

Barrier-fixing function in germfree animals

The state of barrier-fixative function was studied in germfree and conventional guinea pigs and rats. Under conditions of conventional animals contamination with E. coli 055 (in doses of 500 million and 10 milliard microbial bodies for subcutaneous and oral inoculation, respectively) only an early transitory bacteremia developed at the early postinfection periods. As to bacteriemia in gnotobiotes, it increased progressively leading to the animal death in the course of 2 to 3 days. A decreased fixative and bactericidal capacity of the regional lymphatic apparatus and deep structures of the mononuclear-phagocytic system was revealed in germfree animals. An experimental confirmation of the participance of antibodies in the manifestation of the barrier-fixative function to E. coli was obtained. These studies demonstrated an important role of the microbial factor in the formation of the macroorganism barrier-fixative function.

Opsonic activity of the serum of germfree guinea pigs contaminated by single strains of the intestinal microflora

A bone marrow fragment transplanted under the kidney capsule created a focus of ectopic hemopoiesis, whose isze, measured by the number of hemopoietic cells, was proportional to the implant size. Dimensions of the focus proved to be 11/2--21/2 greater in the irradiated than in the intact recipients. Cells building up the focus of heterotopic hemopoiesis had a different radiosensitivity in the intact and irradiated recipients--their Do constituted about 160 and 350 rad, respectively. In this connection it is supposed that two cell populations of precursors took part in the creation of the focus. Their possible relations with the determined and inducible osteogenic precursor cells are discussed.

Electron microscope studies of experimental Entamoeba histolytica infection in the guinea pig. II. Early cellular and vascular changes accompanying invasion of the lamina propria.

Early cellular and vascular changes in response to invasion of lamina propria by Entamoeba histolytica were studied sequentially, at the ultrastructural level, in germfree guinea pigs inoculated intracecally with amebae and enteric flora derived from patients with acute amebic colitis. Approximately one week post-inoculation the animals developed acute colitis with mucosal invasion by trophic amebae. Although epithelial cells at the sites of amebic invasion showed progressive cytoplasmic changes and desquamation resulting in microerosions, most mesenchymal elements in the lamina propria appeared normal without cytopathic changes even when in direct contact with invading amebae. Only the polymorpho-nuclear leukocytes (PMN) apposed or topographically close to amebae exhibited degenerative changes which were characterized by condensation of nucleoplasm and cytoplasm, extra-cellular release of cytoplasmic components including granules, and, finally, lysis of cell membranes. Capillaries and venules in the lamina propria showed a variety of changes such as swelling and gap formation at the intercellular endothelial junctions and more rarely at the fenestrae. Blood vessels physically close to amebae showed formation of endothelial cytoplasmic blebs which pinched off into the vascular or extravascular space. Platelet and fibrin thromboses were common in the more severely damaged capillaries and venules. Fragments or clumps of fibrin-like material were found also in the extracellular spaces. Amebic invasion of the lamina propria, then, is accompanied by continued epithelial shedding, PMN degeneration, and changes in both capillaries and venules consisting of endothelial damage and occlusive thrombosis. The vascular changes appeared to be closely related to PMN degeneration resulting from interaction of PMN with invading amebae.

Effect of contamination of germfree guinea pigs by individual members of the intestinal microflora on antibody and complement levels

On the 3rd day after birth germfree guinea pigs were contaminated by one of the following representatives of the normal intestinal microflora:Bacillusmesentericus,Bacillussubtilis,Staphylococousalbus, andStreptococcusfaecalis. The levels of antibodies against the microorganisms used for monocontamination and also againstEscherichiacoli 055, which is pathogenic for guinea pigs, and the serum complement levels were studied in the animals at the age of 2 weeks. Contamination of the guinea pigs byB.mesentericus andB.subtilis did not significantly change the antibody levels against these microorganisms, whereasS.albus andS.faecalis appreciably stimulated antibody formation. Similar results were obtained with respect toE.coli 055. The complement level was significantly increased by the spore-bearing aerobes and byS.albus.

Cyclic AMP formation during phagocytosis

The content of cyclic adenosine-3′, 5′-monophosphate (AMP) in phagocytic macrophages was shown to be increased especially during phagocytosis of the living microbes. The cyclic AMP formed during phagocytosis could be detected in the incubation medium, but in the cells it remained at almost the same level. The cyclic AMP concentration in cells of the intestinal mucosa and in the blood serum of germfree guinea pigs also was increased after injection ofEscherichia coli 055 cells; this points to the participation of the adenylate cyclase system in interaction between microorganisms and the epithelium of the small intestine.

Blastocystis hominis: Pathogenic potential in human patients and in gnotobiotes

Germfree guinea pigs were inoculated orally, in some experiments, and intracecally, in others, with Blastocystis hominis and the enteric flora from symptomatic patients. Other germfree guinea pigs received the parasite from axenic culture and still others from monoxenic culture with Proteus vulgaris. Fourteen of 43 animals inoculated orally with B. hominis and patient's enteric flora developed B. hominis infections and those with particularly heavy infections developed watery diarrhea of more than 1 week's duration immediately prior to sacrifice. Similar results were obtained from intracecal inoculations in that 13 of 28 animals developed infections and those with the greatest numbers of B. hominis had watery diarrhea for more than 1 week prior to sacrifice. Gross pathologic changes in these animals were mostly unremarkable, with only a slight hyperemia observed in several of the symptomatic animals. Microscopic examination, however, revealed frequent penetration of intestinal epithelium by B. hominis and the parasites in significant numbers were observed within the epithelium. There was a slight increase in cellularity in the lamina propria but parasites were not observed therein and their presence in the epithelium did not provoke an inflammatory response. Only one of eight animals inoculated from monoxenic cultures developed B. hominis infection (asymptomatic), and infections were not produced in animals inoculated from axenic culture. As a result of our observations of diarrhea in patients with particularly heavy infections with B. hominis together with the demonstration of similar symptoms in animals heavily infected with this parasite, we believe B. hominis may occasionally be related causally to the production of such symptoms by a mechanism not completely understandable.

Nature of the toxic factors in burns studied in germ-free guinea pigs

Toxicity (action on migration of leukocytes in culture) of the serum and saline extracts of the internal organs after burns was studied in experiments on germ-free and contaminated guinea pigs. Despite the total absence of a microbial flora, the toxic properties of the serum and extracts of internal organs of the germ-free animals were the same as those of the contaminated guinea pigs. The results point to the important role of histiogenic factors in the development of burn toxemia.

Local Schwartzmann's phenomenon in germfree guinea pigs

To reproduce the local Schwartzmann phenomenon, heat-killed cells ofCitrobacter orEscherichiacoli were injected intradermally and intravenously into 11 germfree guinea pigs, two guinea pigs monoassociated withCitrobacter, and 25 control guinea pigs. All the germfree animals were refractory to the phenomenon, whereas 20 of the 25 control animals gave a positive reaction. A zone of infiltration at the site of intradermal injection was present in all the control guinea pigs but in none of the germfree animals. The possible primary and secondary mechanisms of suppression of the Schwartzmann phenomenon in the absence of the normal microflora are discussed.

Electron Microscope Studies of Experimental Entamoeba Histolytica Infection in the Guinea Pig

Germ-free guinea pigs were inoculated intracecally with Entamoeba histolytica and the enteric flora derived from a human patient with acute amebic colitis. Aanimals were killed at post-inoculation intervals of 7 to 12 days. The mode of penetration of cecal epithelium by the ameba was examined by light and electron microscopy. The following sequence was reconstructed from numerous individual observations. When the amebae were in moderately close proximity to the brush, border, the microvilli became shortened, irregular, and sometimes disappeared. Dense material was observed between the amebae and microvilli. When the ameba was very close to the epithelium the apical portion of epithelial cytoplasm projected into the lumen contacting the organism, thus becoming detached from adjoining cells. This produced spaces between epithelial cells through which amebae invaded interepithelial spaces. Initially the ameba penetrated the interglandular epithelium. Later, it penetrated equally the glandular and interglandular epithelium barrier. There were marked alterations of cytoplasmic components of epithelial cells. Polymorphonuclear leukocytes migrated into the epithelium filling these spaces; these often showed a variety of degenerative processes. Amebae, utilizing their pseudopodia, moved further through the intercellular spaces and reached the lamina propria.