Whole-bacterial cell enzyme-linked immunosorbent assay for cell-bound Moraxella bovis pili

Infectious bovine keratoconjunctivitis (IBK), caused by Moraxella bovis, is a disease of major importance in cattle industry. M. bovis has several virulence factors among which pili are crucial antigen for the protective capacity of vaccines against this disease. The production of vaccines against IBK therefore requires a reliable technique for cellular piliation level assessment on cells to be included as vaccine components. In this study we describe a specific whole-bacterial cell enzyme-linked immunosorbent assay (bact-ELISA) capable of detecting pili antigen on M. bovis cell surface. A sequential competitive bact-ELISA was developed using highly piliated M. bovis cells as antigen. Samples to be analyzed were allowed to react with anti-pilus serum prior to incubation in wells coated with piliated cells of M. bovis. This assay proved useful for the rapid, sensitive and reproducible evaluation of piliation on M. bovis cells, and represents an important tool for cellular piliation monitoring daburing M. bovis cells production in stirred bioreactors.     

Experimental production of infectious bovine keratoconjunctivitis: comparison of serological and immunological responses using pili fractions of Moraxella bovis.

The effect of vaccinating cattle and mice on the development of keratoconjunctivitis was studied. Cattle were vaccinated with whole cells, disrupted cells and pili fractions of three strains of Moraxella bovis. Mice were vaccinated with pili fractions of three strains. The resistance of all vaccinated animals was challenged with virulent cultures of M. bovis. In an attempt to correlate the response seen after vaccination and challenge with a pili fraction of M. bovis, vaccinated cattle and mice were grouped on the basis of signs of disease manifested and compared on the basis of serological responses. Serum samples were tested for antibodies by a gel diffusion precipitin test. A greater number of the sera of resistant cattle had antibodies to the homologous pili antigen than those of vaccinated nonresistant cattle. Cattle vaccinated with disrupted cells were not resistant to infectious bovine kerato-conjuctivitis and their sera lacked antibodies against the pili antigens. Vaccinated mice were more resistant to infectious bovine kerato-conjuctivitis and their sera lacked antibodies against the pili antigens. Vaccinated mice were more resistant to challenge exposure by homologous than heterologous cultures. A greater number of the sera of resistant mice had antibodies to pili antigens than nonresistant mice.     

Attachment of Moraxella bovis to calf corneal cells and inhibition by antiserum

An in vitro assay was developed using calf corneal cells to assess the importance of fimbriae in the colonisation of the bovine ocular surface by Moraxella bovis, and the role of fimbrial antibodies in the bovine immune response and resistance to infectious bovine keratoconjunctivitis (IBK). Fimbriae promoted adherence of M. bovis to calf corneal cells in culture; 15 fimbriate isolates, representative of 6 fimbrial serogroups of M. bovis, adhered to the cells whereas 4 non-fimbriate isolates failed to do so. Fimbrial antibodies in hyperimmune rabbit serum inhibited attachment of all fimbriate strains of the homologous fimbrial serogroup but not those of 5 heterologous serogroups. The relevance of these results to the use of a polyvalent fimbrial vaccine in the control of IBK is discussed.     

Adherence to bovine neutrophils and suppression of neutrophil chemiluminescence by Mycoplasma bovis

The adherence of viable and heat-treated Mycoplasma bovis to bovine peripheral blood neutrophils was studied by specific immunofluorescence staining and flow cytometry. Viable and heat-treated M. bovis cells, adhered to bovine neutrophils in dose-dependent fashion within a 30 min incubation. Fluorescence quenching using crystal violet indicated that unopsonized M. bovis cells remained on the surface of bovine neutrophils without experiencing significant ingestion. The effect of M. bovis adherence on neutrophil microbicidal function was examined by measuring luminol enhanced chemiluminescence (CL). Adherent M. bovis cells did not elicit a bovine neutrophil CL response over a 75 min incubation period. M. bovis inhibited the capacity of bovine neutrophils to mount a CL response. Inhibition occurred whether viable or heat-treated M. bovis cells were used and it occurred when neutrophils were stimulated with opsonized zymosan (OZ) or phorbol myristate acetate (PMA). Inhibition of the PMA stimulated neutrophil CL response required cytadherence by M. bovis cells. These findings suggest that activation of the bovine neutrophil respiratory burst was inhibited at or distal in the pathway to the activation of protein kinase C (PKC), the site of PMA stimulation, and that it was mediated by a direct interaction between the adhering M. bovis cells and the bovine neutrophil membrane.     

Immunogenicity of a Moraxella bovis bacterin containing attachment and cornea-degrading enzyme antigens

An adjuvanted Moraxella bovis bacterin containing attachment antigens and cornea-degrading enzyme antigens protected cattle from infectious bovine keratoconjunctivitis (IBK) when experimentally challenged with homologous and heterologous challenge cultures of M. bovis. This bacterin also protected cattle against field exposure to M. bovis. Transmission electron microscopy and fluorescein labeled anti-M. bovis pili antiserum showed pili on the M. bovis bacterin strain. Scanning electron microscopy demonstrated a fibrillar glycocalyx. The bacterin strain of M. bovis, but not all strains of M. bovis, destroyed bovine corneal cell monolayers in vitro. Bovine corneal cells began to separate from each other within 5 min after M. bovis organisms were added and adhered to the cell monolayers. Moraxella bovis organisms remained attached to the disintegrating cells as the cell membrane separated and was digested. Vaccination stimulated bacterial agglutination antibodies. However, protection against experimental challenge was more closely related to the cornea-degrading enzyme content of the experimental bacterins. Twenty-two of 29 cattle (76%) vaccinated with bacterins containing a relative enzyme activity (REA) greater than 0.4 were protected in a rigorous challenge of immunity test. Only 1 of 21 non-vaccinated calves (5%) was free of IBK. Ninety-two percent (24/26) of calves vaccinated with a bacterin containing a REA greater than 0.29 remained free of IBK following field exposure, whereas 47% (8/17) non-vaccinated calves developed IBK. Only 8 of 12 calves (67%) vaccinated with a bacterin containing a REA of 0.09 remained free of IBK. In a larger field efficacy test consisting of 32 herds in six states, the incidence of IBK in individual herds ranged from 0% to 55%. The overall rate of infection was 11.2%. Vaccination of calves with an M. bovis bacterin that contained a REA of 0.63 reduced the incidence of IBK from 11.2% (217/1931) in the non-vaccinated controls to 4.3% (66/1520) in cattle vaccinated once and to 3.1% (48/1536) in cattle vaccinated twice.     

The protective efficacy of pili from different strains of Moraxella bovis within the same serogroup against infectious bovine keratoconjunctivitis.

Three groups of ten calves were each immunised with a total of 400 micrograms pili prepared from three separate strains of Moraxella bovis in Alhydrogel-oil adjuvant as two divided, equal doses 21 days apart. Groups 1 and 2 each received a monovalent vaccine made from strain 4L and S276R respectively, which belonged to pili serogroup A. Group 3 received vaccine made from pili of strain Maff1, belonging to serogroup F. A further group of ten calves served as non-vaccinated controls. Calves in groups 1 and 2 had developed serogroup A-specific antibody and those in group 3 developed serogroup F-specific antibody, and some evidence of cross-reacting antibody was also detected when measured by an agglutination test using formalin-killed piliated cells of serogroup A strain 4L. Although antibody titres measured against purified pili by ELISA were highest with homologous serogroup antigens, cross-reactive titres to shared epitopes of M. bovis pili were also detected by this method. Ocular challenge of the 40 calves with virulent M. bovis of serogroup A strain S276R was carried out 14 days after the second vaccine dose. All non-vaccinated calves developed infectious bovine keratoconjunctivitis (IBK). The percentage protection in groups 1 (strain 4L) and 2 (strain S276R) was 60% and 80% respectively (P less than 0.05), with mean lesion scores of 0.7 and 0.3 out of a possible 6.0. The percentage protection of calves in group 3 (strain Maff1) was only 30%, with a mean lesion score of 1.4 compared with 2.2 for non-vaccinated controls. The present findings, together with other evidence indicating that immunity to IBK is serogroup-specific, suggest that inclusion of pili from one representative strain from each of the seven Australian and British serogroups in a polyvalent, subunit vaccine should effectively protect the majority of cattle against IBK caused by most field strains of M. bovis encountered in Australia and the United Kingdom.     

Serologic cross-reactivity of Australian Moraxella bovis to vaccinal bacterin strains as determined by competitive ELISA

OBJECTIVE: To conduct a serologic survey and define pili antigenic variability via the serologic cross-reactivity of Moraxella bovis isolates from naturally occurring infectious bovine keratoconjunctivitis (IBK) outbreaks in Australia. This project applies to the development of an M bovis pili-based vaccine targeting Australian strains originating from intensive cattle producing regions. PROCEDURE: Ocular swabs were collected from cattle affected with clinical signs of IBK from 25 veterinary practices. Standard criteria were used to identify 70 M bovis. Pure, piliated isolates were evaluated with a modified competitive enzyme-linked immunosorbent assay (ELISA) for cell-bound M bovis pili to determine their serologic cross-reactivity with pili of vaccinal bacterin strains EPP63, FLA64, and SAH 38. RESULTS: Sixty-four percent (45/70) of M bovis isolates demonstrated homologous pili antigens to a vaccinal strain. M bovis isolates homologous to one of the three vaccinal strains were obtained in 77% (34/44) of IBK outbreaks sampled. No IBK outbreak had isolates homologous to more than one vaccinal strain; however, 29% (10/34) of outbreaks with a cross-reacting strain had non-cross-reacting strains as well. CONCLUSION: The similar prevalence of pilus antigen homology to strain FLA64 was observed with isolates derived from NSW, Tasmania, and Victoria, compared with results of prior smaller serologic studies, suggests that the common pilus antigens in M bovis within Australia have been relatively stable over the last 20 years. The prevalence of a limited number of pilus antigens in M bovis suggest that the application of a vaccine containing the bacterial strains EPP63, FLA64, and SAH38 may provide a useful management tool for reducing production losses associated with IBK in Australia.     

Adherence to various host cell lines of Mycoplasma bovis strains differing in pathogenic and cultural features

Mycoplasma bovis is known to be responsible for pneumonia and arthritis in calves, as well as mastitis in dairy cows. Despite clear evidence of its pathogenic potential, little is known about mechanisms of cytadherence and the molecular factors involved. The purpose of this work was to compare adherence rates of M. bovis field strains to different host cell lines and study the effects of cloning and sub-culturing M. bovis strains on their adherence properties. Eighteen metabolically labeled M. bovis strains isolated from different pathological backgrounds were examined in adherence trials using four different host cell lines, i.e. embryonic bovine lung (EBL), embryonic bovine trachea (EBTr), Madin Darby bovine kidney (MDBK) and rabbit kidney (RK) cells. Although large interstrain variations in adherence rates (3.4-19.1%) were measured they could not be correlated to the pathological background (pneumonia, arthritis or mastitis). Adherence rates to the fibroblast cell line (EBTr) were significantly lower than those to the three epithelial cell lines (EBL, MDBK and RK). The only non-pathogenic strain (221/89) exhibited lower adherence rates than three isolates from clinical mastitis. Interestingly, adherence rates were significantly reduced after in vitro passaging. In contrast, no effect of single cloning of strains on adherence was observed. There was no general correlation between expression of variable surface proteins (Vsps) as monitored by immunoblotting and adherence rates, although alterations in Vsp expression profiles were seen as a consequence of passaging. As there is probably a large number of adhesins, variable and non-variable, on the surface of M. bovis cells the issue is very complex, and the most active components have yet to be identified.     

Detection of shared magnetic antigenic determinants on whole Moraxella bovis pili by use of antisera to cyanogen bromide-cleaved M. bovis pilus protein

OBJECTIVE: To determine the ability of antisera against cyanogen bromide-cleaved pili from 4 strains of Moraxella bovis to react with whole or nondenatured pili. SAMPLE POPULATION: Antisera to 4 strains of M. bovis produced by New Zealand White rabbits. PROCEDURE: Pili from 4 strains of M. bovis were collected and purified. Pilus proteins (pilin) were cleaved, using cyanogen bromide. Whole pilus and cyanogen bromide-cleaved pilin were injected into rabbits. Antisera were serially diluted, reacted with 4 strains of M. bovis, and examined by immunoelectron microscopy and indirect immunofluorescence. RESULTS: Antisera to whole pili aggregated and distorted pili from homologous strains, but pili from heterologous strains were unaffected. Antisera to cleaved pilin fragments resulted in partial aggregation and thickening of homologous and heterologous pili, suggestive of heterospecific antibodies. Attachment of antibodies to pili was detected by indirect immunofluorescence, indicating a strong reaction of antisera to whole pili with homologous pili. Weak cross-reactions were evident with certain heterologous strains. In contrast, antisera to cleaved pilin fragments reacted strongly with pili from homologous and heterologous strains. CONCLUSIONS AND CLINICAL RELEVANCE: We detected shared antigenic determinants on pili from various strains of M. bovis that were not immunogenic in intact pili. These sites were immunogenic after cleavage of pilus protein with cyanogen bromide, and antisera produced to protein fragments reacted with whole pili from heterologous strains of the organism. Vaccines produced from cyanogen bromide-treated pili may induce broader immunity against infectious bovine keratoconjuctivitis than that provided by currently available vaccines.     

Isolation and characterization of monoclonal antibodies against pili of Corynebacterium renale and Corynebacterium pilosum

Five monoclonal antibodies against pili of Corynebacterium renale 115 P+ (piliated clone) and two monoclonal antibodies against pili of C. pilosum 92 P+ (piliated clone) were produced. These antibodies bound to pili of the homologous strain in in enzyme-linked immunosorbent assay (ELISA) and agglutinated P+ but not P- (non-piliated clone) of each homologous strain. The five monoclonal antibodies against C. renale 115 P+ pili were divided into 2 groups, comprising 16/5, 160/1 and 32/6 and 13/4 and B20/3, based on the results of a competitive binding assay. The results may indicate the presence of at least 2 distinct antigenic areas on the pilus of C. renale 115 P+. The monoclonal antibodies of the first group inhibited adhesion of C. renale 115 P+ bacteria to the epithelial cells of bovine vulva, while the second group did not. Two monoclonal antibodies against C. pilosum 92 P+ pili recognized the same area on the pilus of C. pilosum 92 P+, and inhibited the adhesion of C. pilosum 92 P+ bacteria to the epithelial cells of bovine vulva. The adhesion of these bacteria was inhibited by the monoclonal antibodies in the form of IgG as well as by the Fab fragment. The strains of C. renale and C. pilosum which reacted with each of the anti-C. renale 115 P+ pili and anti-C. pilosum 92 P+ pili monoclonal antibodies were small in number and of restricted distribution.     

Pathogenicity and immunogenicity of piliated and nonpiliated phases of Moraxella bovis in calves

Pathogenicity evaluations of Moraxella bovis strain EPP 63 grown in the piliated and nonpiliated phase indicated that the organism grown in the piliated phase induced clinical keratoconjunctivitis, whereas the organism grown in the nonpiliated phase did not induce disease under identical challenge conditions. Calves inoculated with culture grown in the piliated phase developed significantly (P less than 0.01) higher lesion scores than did calves inoculated with culture grown in the nonpiliated phase. Vaccination/challenge exposure evaluations indicated that calves immunized with vaccine prepared from piliated culture had significantly lower lesion scores than did control calves (P less than 0.001) or calves immunized with vaccine prepared from nonpiliated culture (P less than 0.05). Similarly, calves immunized with piliated vaccine developed a significantly higher antibody titer against purified pili. Nonvaccinated controls and calves immunized with nonpiliated vaccine did not develop a significant increase in antibody titer. The results indicate that M bovis pili constitute an important factor in the pathogenicity of keratoconjunctivitis and may be used as a vaccine in the control of keratoconjunctivitis in cattle.     

Serologic and protective characterization of Moraxella bovis pili

This study was conducted to determine the protective nature of purified M. bovis EPP 63 pili in controlling experimentally induced Infectious Bovine Keratoconjunctivitis, and to determine antigenic similarity of pili isolated from various M. bovis isolates. Ten calves were vaccinated twice, 28 days apart, with 5.0 mg (protein) EPP 63 purified pili. Ten calves were maintained as non-vaccinated controls. All calves were exposed to ultraviolet light prior to challenge. The calves were challenged by instilling approximately 2.0 X 10(8) CFU of EPP 63 piliated organisms into the conjunctival sac. Antisera to respective pili types were prepared by immunizing the rabbits with purified pili from M. bovis strains EPP 63, FLA 64, IBH 68, MED 72 and ATCC 10900. Rabbit serum was evaluated for cross reactivity by enzyme-linked immunosorbent assay (ELISA). Purity of pili preparations was demonstrated on SDS-PAGE gels. Molecular weight of pili subunit was determined to be approximately 20,000 for EPP 63, 19,500 for IBH 68 and ATCC 10900, and 17,500 for FLA 64 and MED 72. One of 10 (10%) calves vaccinated with EPP 63 purified pili, and 6 of 10 (60%) nonvaccinated controls developed IBK, respectively. Average eye scores for vaccinates and controls were 0.05 and 0.85, respectively. Significant cross-reaction was found between EPP 63 and MED 72 pili. FLA 64 and ATCC 10900 were similar; however, antiserum to IBH 68 pili showed some degree of cross reaction with other pili.     

Detection of cell detachment activity induced by Moraxella bovis

OBJECTIVE: To characterize the effect that filtrate obtained from cultures of Moraxella bovis has on cultured corneal epithelial cells and other types of cultured mammalian cells. SAMPLE POPULATION: Cultured hamster corneal epithelial cells, bovine epithelial cells, and several transformed cell lines exposed to culture filtrate from a pathogenic isolate of M bovis. PROCEDURE: Moraxella bovis was cultured, and bacteria were removed by filtration. The resulting bacterial culture filtrate was incubated with various types of cultured cells, and effects of the filtrate on detachment of various mammalian cell types was quantified by the use of neutral red dye. Additionally, bacterial culture filtrate was treated with protease inhibitors as well as trypsin and heat prior to incubation with cultured mammalian cells. RESULTS: Bacterial culture filtrate of M bovis caused all types of cultured cells to detach from each other and from the substrate, with the maximal effect evident 2 hours after initiating incubation. Detached cells were alive, and detachment was reversible. Serine protease inhibitors (phenylmethylsulfonylfluoride and alpha2-macroglobulin) inhibited cell detachment attributable to bacterial culture filtrate. Heating and treatment with trypsin also inhibited cell detachment. CONCLUSIONS AND CLINICAL RELEVANCE: Moraxella bovis produces a soluble factor that causes reversible detachment of cultured cells. This activity may play a role in the pathogenesis of infectious bovine keratoconjunctivitis.     

Cytopathic effects of Moraxella bovis on cultured bovine neutrophils and corneal epithelial cells

The effects of Moraxella bovis on the morphologic features of purified bovine neutrophils and bovine corneal epithelial cells were examined, using transmission and scanning electron microscopy and light microscopy. Within 2 minutes after incubation of bovine neutrophils with living M bovis, electron microscopic cellular changes included vacuolation, swelling, and loss of microplicae. Most of the neutrophils were lysed by 10 minutes of incubation. Human neutrophils phagocytosed the M bovis and remained intact, even after 30 minutes of incubation with the bacteria. Living M bovis killed bovine corneal epithelial cells in vitro. Sterile filtrates prepared from 6-hour shaker cultures of M bovis also killed bovine corneal epithelial cells, but the cytotoxic activity was less than that produced by the living bacteria. Cellular changes were first observed in specimens collected 1 hour after corneal cell monolayers were inoculated with sterile culture filtrates. The changes in these cells included pit-like lesions on the cellular surface, cellular separation, and vacuolation.     

Infectivity and virulence of Australian strains of Moraxella bovis for the murine and bovine eye in relation to pilus serogroup sub-unit size and degree of piliation

The degree of piliation of 29 haemolytic and 4 non-haemolytic Australian strains of Moraxella bovis representing 7 different pilus antigen groups was determined. The infectivity and virulence for the eye was measured in steroid-treated mice and in cattle. Non-piliated strains failed to infect the murine eye. Most moderately or heavily piliated strains reproducibly produced the highest infectivity and virulence scores in mice when compared with lightly or very lightly piliated strains (p less than 0.05). Non-haemolytic, piliated strains were infective and in one instance virulent for mice. Almost similar levels of infectivity and virulence were observed for 7 representative haemolytic strains tested in both cattle and mice. The relative molecular weight of pilin sub-units was compared using sodium dodecyl-sulphate polyacrylamide gel electrophoresis. Three classes of pili, alpha, beta and gamma of ascending sub-unit size were identified among the 7 pilus antigen serogroups. Pilin sub-unit size bore no relationship to the degree of piliation but most strains that were highly virulent in mice and cattle expressed alpha and gamma sub-units. Some strains appeared capable of switching from alpha to beta or form beta to gamma sub-unit production.     

Adherence as a Prerequisite for Infection of the Bovine Mammary Gland by Bacteria

Using a simple in vitro test it was demonstrated that staphylococci, Streptococcus agalactiae, and micrococci, the species of bacteria which are commonly isolated from udder infections, adhered to mammary gland epithelial cells readily and in large numbers. Some strains of organisms which are associated with sporadic outbreaks or occur less commonly, like Str. dysgalactiae and Str. uberis, adhered moderately. Escherichia coli, Klebsiella spp., Corynebacterium pyogenes, C. bovis, Str. bovis, and Str. faecalis, species which are isolated occasionally, adhered poorly. From these studies, it appears that selective adherence of bacteria to the epithelial cells is a factor contributing to the ability of organisms to infect the mammary gland and may, therefore, be considered an important stage in the pathogenesis of bovine mastitis.     

Adhesion of Corynebacterium pilosum by pili to epithelial cells of bovine vulva

Piliated (P+) and nonpiliated (P-) clones of Corynebacterium pilosum were selected, and their adhesion to the epithelial cells of the bovine vulva and vaginal vestibule was examined. The number of P+ bacteria of C pilosum that adhered to vulval epithelial cells was greater than that of P- bacteria. The adhesion of P+ bacteria, but not P- bacteria, to the epithelial cells was inhibited by the antipilus antiserum; therefore, the adhesion of C pilosum to the epithelial cells of the vulva was primarily dependent on the pili. The number of C pilosum that adhered to the epithelial cells of the bovine vulva and of the vaginal vestibule increased by decreasing the pH.     

Studies about the mechanism of internalization by mammary epithelial cells of Escherichia coli isolated from persistent bovine mastitis

The purpose of this study was to investigate the interaction between Escherichia coli and primary mammary epithelial cell cultures derived from cows with persistent intramammary infection (IMI). Two strains of E. coli, isolated from the milk of two different cows suffering from persistent E. coli IMI were tested for adhesion to and invasion of three primary mammary epithelial cell cultures derived from mammary biopsies of the two infected cows. Intracellular E. coli were detected during five days post infection in vitro. Both strains of E. coli adhered to and invaded monolayers of all three primary mammary epithelial cell cultures. One strain adhered less but invaded more than the other. Comparison with other mammary pathogens indicated that E. coli invaded the cells less efficiently than Staphylococcus aureus, about as efficiently as Streptococcus dysgalactiae and more efficiently than Streptococcus uberis. The mechanism of E. coli invasion was studied using the cytoskeleton disrupting agents colchicine and cytochalasin D. These compounds inhibited the invasion of E. coli. Invasion of E. coli could also be inhibited by the phosphokinase inhibitors genistein and staurosporin in a dose-dependent fashion. Phorbol-myristyl-acetate (PMA) had no effect on the invasion of E. coli. Histology of mammary tissue revealed chronic inflammatory changes in quarters that were persistently infected by E. coli. Intracellular bacteria were not detected in mammary tissue sections. Polymerase chain reaction (PCR) analysis suggested that the two strains of E. coli lacked genes encoding for bundle-forming pili (bfpA), intimin (eae) and translocated intimin receptor (tir), which are characteristic for enteropathogenic E. coli (EPEC).     

Plasmid content of piliated and nonpiliated forms of Moraxella bovis

Plasmid profiles were compared between nonpiliated and piliated forms of Moraxella bovis isolates. The piliated form of M bovis isolate IBH64 contained 1 fewer plasmid than did the nonpiliated form. Piliated and nonpiliated cells of IBH64 contained plasmids having molecular size of 45, 32.8, 4.9, and 4.6 kilobases (kb). Single- and double-restriction endonuclease digestion by Ava I and Nde I indicated that the size of the additional plasmid carried by the nonpiliated form of IBH64 was approximately 43.6 kb. The M bovis isolates, Newport and GRS, contained the same number of plasmids in either their piliated or nonpiliated form.     

Expression of type 1 pili by Escherichia coli strains of high and low virulence in the intestinal tract of gnotobiotic turkeys

Highly virulent (strain 1) and weakly virulent (strain 3) Escherichia coli were examined using immunofluorescent and electron microscopic techniques to determine their ability to express type 1 pili in the intestinal tract of 3-week-old gnotobiotic turkeys. Turkeys were necropsied on postinoculation day (PID) 1, 2, 5, 8, and 12. Nonpiliated forms of strains 1 and 3 were more numerous than piliated forms in cecal and colonic contents examined by negative staining electron microscopy. A piliated form of strain 1 was seen in intestinal contents on each PID and was more numerous in cecal contents than in colonic contents. The mucus blanket of the cecum and colon contained large numbers of bacteria, although organisms were rarely intimately associated with the intestinal epithelium. Immunofluorescent staining indicated large numbers of piliated forms of strains 1 and 3 within the mucus blanket of the cecum and colon on PID 2, 5, 8, and 12. Piliated bacteria were infrequently seen in the ileal mucus blanket. Serum antibody titers to type 1 pili increased markedly by PID 5 and persisted in turkeys inoculated with strain 1. In contrast, antibody titers in turkeys exposed to strain 3 increased gradually and varied markedly among birds at each PID. Type 1 pili may not be important for adherence of pathogenic E coli to intestinal epithelium of turkeys.