it MSU RETURNING MATERIALS: Place in book drop to LIBRARIES remove this checkout from ” your record. FINES win be charged if book is returned after the date . - ”' stamped be10w. .m 1% 7 u., n 1 “win“; \M JULIE-b "' w 3? K209 Lem; Fan-9 ,i) q;~'~ p{?; 8. . C 3’. C '1.) W " MAGIC 2 SEP g'ffi998 \\ ADHESION OF K99-POSITIVE ENTEROPATHOGENIC ESCHERICHIA COLI TO PIG INTESTINAL BRUSH BORDERS BY Sheila D. Grimes A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of ‘ MASTER OF SCIENCE Department of Pathology 1984 ABSTRACT ADHESION OF K99-POSITIVE ENTEROPATHOGENIC ESCHERICHIA COLI TO PIG INTESTINAL BRUSH BORDERS BY Sheila D. Grimes A study of 242 pigs was performed to determine the incidence of intestinal receptors for K99-positive Escherichia coli and to determine if an inheritance pattern existed for the receptors. Ileal samples, collected from pigs at three Michigan slaughterhouses, were examined using a brush border procedure modified for the identification of receptors for K99 pili. Adhesion of bacteria to brush border fragments indicated the presence of receptors. Receptors were demonstrated in 230 (95%) of the intestinal samples. From zero to greater than ten bacteria adhered per fragment. Following storage of brush borders at 4° C, bacterial aggregates lacking identifiable fragments were present in samples tested for adhesion, suggesting the release of K99 receptors from brush border membranes. Results from the brush border adhesion test indicate that most, if not all, pigs have receptors for K99 pili. Therefore, it appears that an inheritance pattern similar to that observed for K88 receptors does not exist for K99 receptors. DEDICATION To my parents ii ACKNOWLEDGEMENTS I wish to express my sincere thanks to Dr. Glenn L. Waxler, my major advisor, for his guidance, support and encouragement during the course of this study. I would also like to thank the members of my guidance committee, Dr. John P. Newman from the Department of Microbiology and Dr. Allan Trapp from the Department of Pathology, for their assistance. Special thanks are extended to Irene Brett and Paul Carlson for their technical help, Cheryl Assaff for her clerical assistance and John Allen for his work with this project. Also, I would like to thank the peOple at Michigan State University Meats Laboratory, Milliganfs Meat Packing Plant and Bainfls Packing Company for their c00peration during the course of this study. iii Table LIST OF TABLES Page Pig phenotypes based on K88 serologic variants of Escherichia coli . . . . . . . . . . . . . . 23 Source and number of intestinal samples with or without adhesion of K99-positive Escherichia c011 O O O O O O O O O O O O O O O O O O O O O 45 Comparison of results of brush border adhesion tests on fresh and glycerol-stored brush border fragments in 10 pigs . . . . . . . . . . . . . 58 Results of brush border adhesion retests on 24 Grade 1 intestinal samples stored in different manners . . . . . . . . . . . . . . . . . . . . 59 vi Figure 1 LIST OF FIGURES Phase contrast photomicrograph of a brush border fragment with Grade 1 adhesion. Notice the semilunar appearance of the microvillus surface (arrow) and the halo surrounding the fragment. (400K) . . . . . . . . . . . . . . Phase contrast photomicrograph of a brush border fragment with Grade 4 adhesion. Notice at the bottom of the micrograph the large aggregate consisting of several adjoining epithelial cell fragments. (400K) . . . . . . Electron micrograph of an E; coli bacterium with K99 pili (arrows) measuring 4.5 nm in diameter. (Phosphotungstic acid stain, 55'9BOX) O O O O O O O O O O O O O O C O O O O 0 Phase contrast photomicrograph of a brush border fragment (arrow) with Grade 2 adhesion. Notice the two bacteria adhering to the frag- ment. (400K) . . . . . . . . . . . . . . . . Phase contrast photomicrograph of a brush border fragment (arrow) with Grade 3 adhesion. Notice the 8 bacteria adhering to the frag- ment. Three small bacterial aggregates are also present. (400K) .. . .. . .. . .. . Phase contrast photomicrograph of several brush border fragments showing various grades of adhesion. All four grades of adhesion are present. The fragment in the center of the micrograph has Grade 4 adhesion. In the upper left corner, there are two fragments, one with Grade 1 adhesion and the other with Grade 2 adhesion. Three fragments are located in the lower right corner, showing Grade 1, Grade 2 and Grade 3 adhesion. An aggregate of brush border fragments is present in the upper right corner. (4OOX) . . . . . . . . . . . . . . . vii Page 44 . 47 . 49 Figure 10 11 12 Page Phase contrast photomicrograph of several brush border fragments showing various grades of adhesion. The central brush border fragment has Grade 4 adhesion. The fragment on the left has Grade 3 adhesion, and the fragment on the right has Grade 2 adhesion. Notice the adhesion of bacteria along surfaces other than the micro- villus surface of two of the fragments. Near the bottom of the photomicrograph is a bac- terial aggregate with no brush border fragment visible beneath it (arrow). Coadherence is present on the central brush border fragment. (400K) . . .. . .. . .. . .. . .. . . . . . ‘49 Phase contrast photomicrograph of a brush border fragment with no K99-positive E; coli adhering to the microvillus surface but bacteria present on the other surfaces. (400K) . . . . . . . . . . . . . . . . . . . . . 52 A series of three phase contrast photomicro- graphs (Figures 9, 10 and 11) demonstrate the detachment of a bacterial aggregate from a brush border fragment. Notice the bacterial aggregate attached to the microvillus surface of the fragment (arrow). (400K) . . . . . . . . 52 Notice that the bacterial aggregate present in Figure 9 has moved from the microvillus surface to the edge of the fragment (arrow). (400K) . . 54 Notice that the same bacterial aggregate present in Figures 9 and 10 has now moved to the posterior surface of the fragment (arrow). (400K) . . . .. . .. . .. . .. . .. . .. . 54 Distribution of the grades of adhesion of K99- positive E; coli to brush border fragments in 242 ileal samples 0 O O O O O O O O O O O O O O 57 viii TABLE OF CONTENTS Page LIST OF TABLES . . . . . . . . . . . . . . . . . . . . . vi LIST OF FIGURES . . . . . . . . . . . . . . . . . . . vii INTRODUCTION . . . . . . . . . . . . . . . . . . . . . . LITERATURE REVIEW . . . . . . . . . . . . . . . . . . . Clinical Signs Resistance to K99-Positive E; coli . . . Culture Procedures for K99-Positive E; coli . Pilus Vaccines . . . . . . . . . . . . . . . Lesions . . . . . . . . . . . . . . . . . . . . . . Pathogenesis . . . . . . . . . . . . . . . . . . . Pili . . . . . . . . . . . . . . . . . . . . . The K88 Pilus . . . . . . . . . . . . . . . . The 987P Pilus . . . . . . . . . . . . . . . The K99 Pilus . . . . . . . . . . . . . . . . 10 Receptors . . . . . . . . . . . . . . . . . . . 11 Enterotoxins . . . . . . . . . . . . . . . . . . 15 Resistance to E; coli Enteric Infections . . . . . 18 Inherited Resistance to K88-Positive E; coli . . 18 MATERIALS AND METHODS O O O O O O C O O O O O C O O O O 3 2 Specimens . . . . . . . . . . . . . . . . . . . . . 32 Brush Border Preparation . . . . . . . . . . . . . 32 E; coli Strains . . . . . . . . . . . . . . . . . . 34 Slide Agglutination . . . . . . . . . . . . . . . . 35 Brush Border Adhesion Test . . . . . . . . . . . . 36 Electron Microscopy . . . . . . . . . . . . . . . . 37 RESULTS 0 O O O O O O O O I O O O O O O O O O O O O O O 39 Brush Border Preparations . . . . . . . . . . . . . 39 Bacteria O O O O O O O O O O O O O O O O O O O O O 39 AdheSion O O O O O O O O O O O O O O O O O O O O O 42 DISCUSSION 0 I O O O O O O O O O O O O O O O O O O O O 0 60 SUMMARY 0 O O O O O O O O O O O C O O O O O O O O O O O 6 7 iv BIBLIOGRAPHY VITA . . . . 69 83 INTRODUCTION Enterotoxigenic Escherichia coli (ETEC) is one of the most important causes of neonatal diarrhea in pigs, account- ing for approximately 50% of the porcine enteric diseases present during the preweaning period. Enteric colibacillo- sis is responsible for the loss of millions of dollars in the pig raising industry yearly. Two virulence determinants are necessary for E; _c_o_l_i to be enteropathogenic: 1) enterotoxin production and 2) pilus production. Three antigenically distinct pili, designated K88, K99 and 987P, are present on separate strains of E; ggli capable of infecting pigs. The pili are filamentous organelles present on the surface of the bacterium. Pili allow the bacteria to attach to receptors specific for the pili on the mucosal surface of the pig‘s small intestine. Bacteria are then able to colo- nize the small intestine. Once attached, E; 921i produce enterotoxins which lead to hypersecretion, diarrhea, dehy- dration and often death. Current research has shown that receptors for the K88 pilus are genetically determined. Previously, it has been suggested that all pigs are susceptible to, or possess receptors for, K99-positive strains of §g_ggli. However, no substantial work has been performed to determine the incidence of receptors for the K99 pilus in pigs. Moon gt 31. (1979) stated that they had not encountered pigs con- genitally resistant to colonization by g; 92;; carrying the K99 pili. They did not indicate how this determination was made. Therefore, it is important to clarify whether or not all pigs have receptors for K99-positive strains of E; 921; or if there is a possible inheritance pattern. The objectives of this study were: 1. To adapt and modify the brush border procedure described by Sellwood gt El! (1975) for the identi- fication of intestinal receptors for K99-positive E; coli. 2. To determine the incidence of receptors for the K99 pilus in Michigan swine. 3. To determine whether or not an inheritance pattern may exist for receptors for the K99 pilus. LITERATURE REVIEW Since 1899, when Enggli was first thought to be a cause of diarrhea in neonatal pigs (Jensen, 1948), an exten- sive amount of literature has been written about numerous aspects of ETEC infections in swine, as well as calves, lambs and human beings. Several general reviews concerning colibacillosis in pigs and other animals have been published (Gaastra and de Graff, 1982; Kohler, 1972; Nielsen st 31,, 1968; Rutter, 1975). Emphasis in this literature review will be placed on the most important facts and the details relevant to the present research. Eh.£2121 a gram negative, motile or nonmotile, nonspore-forming rod, belonging to the family Enterobacteriaceae, is a normal inhabitant of the lower intestinal tract of all warm blooded animals (Bruner and Gillespie, 1973). However, when pathogenic strains of E; ggli are present in the small intestine in large numbers, diarrhea occurs. Clinical Signs Piglets from 12 hours of age to several weeks of age, with a peak incidence occurring at three days of age, are susceptible to enteric colibacillosis. Usually several or all the pigs in a litter are affected. Pigs farrowed by first-litter gilts have a greater incidence of enteric colibacillosis than do pigs farrowed by sows (Blood gt_§l,, 1979; Wilson, 1981). Infected piglets have yellowish or brownish feces which vary from a pasty to a watery consistency. The animals have rough hair coats and normal to subnormal temperatures. They are also anorectic, depressed and weak. As the condition progressively worsens, the animals become dehydrated and emaciated. Pigs commonly die within 24 hours after the onset of clinical signs (Blood gt 31., 1979; Wilson, 1981). Lesions Gross lesions are minimal. The intestines of the affected pigs may be normal or hyperemic, flaccid and dis- tended with fluid or gas. Clotted milk may be present in the stomachs. Catarrhal enteritis may also be present (Blood £21., 1979; Jubb and Kennedy, 1970). Histologically, few lesions are present in the intes- tines. Smith and Jones (1963) reported that no inflammatory changes were present in the intestinal tracts of pigs. Christie and Waxler (1972) reported that a mild neutrophilic infiltration of the intestinal villi, hydropic degeneration in villous epithelial cells and edema were present in gnoto- biotic pigs infected with E; 221;. Bacilli may be seen attached along the margins of the small intestinal villi (Wilson, 1981). Pathogenesis Following oral ingestion of Eh coli, two important factors are required for the organism to be enteropatho- genic. The necessary virulence attributes are: 1) pilus production and 2) enterotoxin production. Pili Pili, a Latin term for hairs or hair-like structures, are filamentous organelles projecting from the bacterial surface. ‘Brinton first used the term in association with these structures in 1959. Prior to that time, Duguid gt El! (1955) had used the term fimbriae, meaning fringe, threads or fibers in Latin. Both terms are presently being used in the literature. Two important functions are associated with pili: 1) they increase the active surface area which may facilitate membrane associated activities such as respiration and nutrient uptake and 2) they act as attachment organelles (Ottow, 1975). The latter function plays an important role in the pathogenesis of neonatal enteric colibacillosis. Structurally, the pili present on ETEC are composed of numerous repeating protein subunits (Isaacson, 1977: Mooi and de Graaf, 1979). The pili on ETEC were originally thought to be capsular (K) antigens and were designated K88 and K99. However, K antigens are polysaccharide in nature and pili are protein in nature. Hence their designation as K antigens is not entirely correct. Pili acting as adhesive organelles are peritrichously arranged on the baCterial surface (Ottow, 1975). The pili present on ETEC are distinctly different from the common or type I pili usually associated with §;.EQli° Several prOperties of the two types of pili differ. One major distinguishing prOperty is that ETEC pili have mannose-resistant hemagglutinating activity, compared to the mannose-sensitive hemagglutinating activity of type I pili (Duguid gt a” 1955; Evans e_t 11., 1979). Following entry of bacteria into the small intestine, there are several non-immunological defense mechanisms, such as gut peristalsis, villous pumping and flowing ingesta and mucus, which tend to rapidly wash bacteria out of the small intestine (Dixon, 1960; Moon, 1980). Pili, however, enable ETEC to overcome these natural defense mechanisms by allowing bacteria to attach and rapidly proliferate in the small intestine to numbers comparable to those normally found in the sluggishly motile, avillous large intestine (Bertschinger gt g},, 1972; Jones and Rutter, 1972). Evidence supporting the fact that pili on ETEC facilitate adhesion has been demonstrated tggytgg and £2 ytttg. Microscopically, several early reports demonstrated the close association of ETEC with small intestinal villi (Arbuckle, 1970, 1971; Drees and Waxler, 1970; Moon gt gt" 1971; Staley fl g_l_., 1969). Bertschinger _e_t_gl. (1972) showed that enteropathogenic gt gglt_(EEC) are more likely to be associated with the intestinal epithelium than nonenteropathogenic _Et._ coli (NEEC). The EEC were present adjacent to the brush border along the entire villus, but the NEEC were randomly distributed in the central lumen. In conventionally reared pigs, Jones and Rutter (1972) demonstrated that the K88-positive strain of'gt.gglt was able to adhere to the mucosa of the small intestine and cause diarrhea, compared to the K88-negative mutant which was unattached, was distributed throughout the lumen and did not cause diarrhea. The results indicated that the K88 pilus is responsible for adhesion of K88-positive bacteria to the small intestinal mucosa and that adhesion is required for the virulence of K88-positive bacteria in pigs. Sup- portive evidence for pili facilitating adhesion was also provided by Nagy gt g1. (1976, 1977) when strain 987?, a K88-negative piliated strain of ETEC, was also shown to adhere to the small intestine. The EEHXEEEQ adhesion of piliated ETEC has been demon- strated with the brush border technique (Sellwood gt gl,, 1975), intestinal epithelial cell technique (Isaacson gt _l,, 1978a; Nagy gt gl,, 1977; Wilson and Hohman, 1974), the intestinal villous technique (Girardeau, 1980) and hemagglu— tination studies (Evans gt gl., 1979). The pili most commonly present on strains of _E_._gglt capable of causing diarrhea in neonatal pigs are K88, K99 and 987P. Recently, adhesive factors distinct from the former three, with characteristics of mannose-resistant pili, have been discovered (Aning and Thomlinson, 1983; Awad-Masalmeh gt gl,, 1982; Moon gt g},, 1980). Morris gt gt. (1983) have recently discovered the F41 pilus on piglet strains of ETEC. The F41 pilus appears to be the same pilus that was described by Moon.gt_gt. (1980) and Awad-Masalmeh gt gt. (1982). Two strains of ETEC have been shown to produce both K88 and K99 pili (Schneider and To, 1982). The K88 Pilus. Of the three most commonly found pili, the K88 pilus was the first to be discovered and conse- quently has had the most written about it. Urskov gt gt. first described the K88 pilus in 1961. The extrachromosom- ally transferred K88 pilus is encoded for by a 50 megadalton plasmid (Orskov ££.Elrr 1966; Shipley 2£.El:r 1978). Chemi- cally, the K88 pilus has been identified as a protein, containing all the common amino acids except cysteine- cystine, with little or no carbohydrate present (Anderson gt gt., 1980; Stirm gt_a_l., 1967b). The lack of sulfur- containing amino acids suggests that no disulfide bonds are present in the pilus. Structurally, Stirm gt gt. (1967a) described it as a filamentous antigen with a diameter of 7- 11 nanometers (nm) and a length of 100 to 150 nm. ‘Wadstrflm gt gt. (1979) estimated the K88 pilus's diameter to be 2.1 nm. The K88 pilus is a polymer of noncovalently linked subunits whose monomeric molecular weight ranges from 23,000 to 26,000, depending upon the serologic variant (Anderson gt g_l_., 1980; Mooi and de Graaf, 1979). Serologically, there are three variants of the K88 antigen, K88ab, K88ac and K88ad (Guinée and Jansen, 1977a; Drskov gt gt,, 1964). The common antigenic determinant, a, is combined with a variable antigenic factor designated b, c or d. The existence of the K88ab and K88ac variants has been known since 1964, but the K88ad variant was just discovered recently. The variants are antigenically different due to amino acid sequence differences in the pilus proteins. Amino acid sequences corresponding to the antigenic variables a, b, c and d have been predicted (Gaastra and de Graaf, 1982; Gaastra gt gt,, 1979, 1983; Klemm and Mikkelsen, 1982). The K88 pilus is species specific. Strains of gt coli that are K88-positive are able to colonize pig intestine tg tttg and cause diarrhea, but K88-positive ETEC have not been reported as enteropathogenic in other species (Moon gt gt., 1977; Smith and Balls, 1967). The 987P Pilus. Present knowledge about the 987P pilus is limited. The 987P pilus was first determined to facili- tate intestinal adhesion and colonization in 1976 (Nagy gt gg,, 1976). The pilus is composed of protein and has a diameter of 7 nm (Fusco gt_gt,, 1978; Isaacson and Richter, 1981). The molecular weight of the 987P pilus's subunit was first reported as 18,900 by Fusco gt gt. (1978) and has since then been reported to be 20,000 by Isaacson and Richter (1981). Presently, it is not known whether or not the 987P pilus is genetically encoded for by a plasmid or a chromosome (Moon gt gt., 1979). The 987P pilus is also considered to be species specific as a virulence factor in pigs, even though it has been found on calf strains of E; 10 ggtt. In calves, 987P piliated strains of g; 221i appear to be harmless (Moon, 1978b). The K99 Pilus. Drskov gt gt. first described the Kco antigen present on calf and lamb enteropathogenic strains of _E_._ggtt as K99 in 1975. In 1977, the K99 pilus was dis- covered on pig strains of ETEC (Moon gt gt,, 1977). Isaacson (1977) concluded that the K99 antigen was a pilus based on its subunit structure, external surface location and rod-like shape (1977). The K99 pilus is encoded for by a transmissible plasmid (Smith and Linggood, 1972). The pilus's length was reported to be 130 to 160 nm and its diameter was reported to be 7.41.6 nm (Altmann gt gt., 1982; Isaacson, 1977). Recently, Isaacson gt gt. (1981) have reported the pilus's diameter to be 4.8 nm. The K99 pilus is protein in nature and lacks the sulfur-containing amino acids, cysteine and methionine (Altmann gt_gt,, 1982). In the past, there has been a great deal of controversy concerning the molecular weight and characterization of the K99 pilus with disparate reports based on different isolation procedures originating from several laboratories (Altmann gt gt,, 1982; Chantner, 1982; de Graaf _e_t_gt., 1981; Isaacson, 1977, 1978; Morris gtgt., 1977, 1978a,b). Recently, Morris gt_gt, (1980) have reported that there are two pilus subunits present in sero- groups 09 and 0101 with different molecular weights and ionic charges. The cationic and lower molecular weight component is the K99 pilus and the high molecular weight, anionic component is F41 which has been shown to be a pilus 11 with adhesive abilities (Morris gt gt,, 1982). The NOIGCU- lar weight of the K99 pilus subunit was reported by de Graaf (1981) to be 18,500 and by Chanter (1982) to be 19,000. The K99 pilus is not species specific. Calf, lamb and pig strains of gt ggtt bearing K99 pili are pathogenic (Moon gt gt,, 1977; Drskov gt gt,, 1975; Smith and Linggood, 1972). Receptors Pili let bacteria attach to special receptors present on the small intestinal mucosa. Present knowledge about the actual nature of the receptors to which K88, K99 and 987P attach is limited. Research is currently ongoing to determine the chemical composition of the receptors. Reports indicate that chemically the receptor is composed of glycolipids or glycoproteins (Gibbons gt gt,, 1975; Kearns and Gibbons, 1979). Gibbons gt gt. (1975) determined that glycoproteins with a terminal B-D-galactosyl structure were able to inhibit the hemagglutination of guinea pig erythro- cytes by K88 pili, suggesting that the glycoprotein might bind with the K88 pilus and inhibit hemagglutination due to its resemblance to the K88 receptor. The B-galactosyl residues have also been shown to play an important role in the binding of K88 pili to brush borders (Sellwood, 1980). Binding of K88 pili to brush borders occurs from 4° C to 37° C, while binding to the guinea pig erythrocytes only occurs at 40 C, indicating that the fit for the natural receptors is better than that of the guinea pig erythrocyte receptors 12 (Sellwood, 1980; Gibbons gt gt,, 1975). Glycolipids have also been shown to have K88 receptor activity (Sellwood and Kearns, 1979; Kearns and Gibbons, 1979). The brush border receptor for the K88 pilus reacts chemically in a way simi- lar to that of the cholera toxin receptor which is a GMl ganglioside; however, according to Sellwood and Kearns (1979), it doesn“t really appear to be a ganglioside. Sugars and glyc0proteins inhibit binding to brush borders. Glycoproteins with N-acetylglucosamine and N- acetylgalactosamine as terminal sugars inhibit binding of the K88 antigen and, to a lesser degree, sugars alone, such as N-acetylglucosamine, N-acetylgalactosamine and N-acetyl- mannosamine, can also inhibit binding (Anderson gt gt., 1980). Hemagglutination tests have been performed on erythro— cytes from several species in an attempt to identify speci- fic receptor sites. Parry and Porter (1978) demonstrated that K88ab adhered strongly to chicken erythrocytes but not to the extent of that seen on the brush borders of intes- tinal epithelial cells. Guinea pig erythrocytes were agglu- tinated by both K88ab and K88ac pilus preparations and live organisms. They concluded that both the a and the b or c determinants of the K88 pilus for K88ab and K88ac are involved in adhesion to intestinal epithelial cells and guinea pig erythrocytes. However, in the chicken erythro- cyte, adhesion of the pilus appears to involve only the K88b determinantfs receptor on the erythrocyte (Parry and Porter, 13 1978). The hemagglutination reaction is easier to study, but the adhesion is not as specific as that seen with the natural receptors on epithelial cells. Intestinal epithelial cells are believed to have dif- ferent receptors for different pili. Using strains of gt ggtt with K99 and 987P pili, binding of piliated E; ggtt to brush borders was inhibited by homologous pili, but not heterologous pili which indicated that a difference in receptors existed (Isaacson gt gt,, 1978a). There may not be a separate receptor for each of the serologic variants of K88-positive Eh ggtt, Bijlsma gt gt. (1982) showed that blocking the receptor site with one serologic variant of the K88 antigen also inhibited the adhesion of other K88 variants. For example, when brush borders from pigs susceptible to all three serologic variants of E; 2211 were exposed to the K88ab antigen and the receptor sites were blocked, then K88ad- and K88ac- positive strains of E; ggtt were no longer able to adhere. However, adhesion of K88ad to brush borders from one pig susceptible to all three serologic variants of K88 was not inhibited by K88ab or K88ac antigen. This suggests that the receptor site for K88ad in brush borders from pigs suscep- tible to all three serologic variants may have two config- urations which can only be detected by receptor blocking. Bijlsma gt gt, consider that only one receptor, depending upon its modifications, will allow attachment of one, two or all three of the K88 serologic variants (Bijlsma gt gt., 1982). 14 Wilson and Hohmann (1974) believe that the K88a anti- genic component doesnft affect the ability of bacteria to attach to intestinal mucosal cells. They found that the antigen necessary for blocking adhesion was the c or b component. Bijlsma gt gt. (1982) say that it is unlikely that the antigenic components b, c or d are exactly the same as the adhesion component of the K88 pilus because K88ac antigen is able to block K88ab and K88ad receptors of pheno- type A brush borders, yet K88c antibody doesn't react with K88b and K88d. Conversely, Parry and Porter (1978) were able to block the adhesion of K88ab- and K88ac-positive strains of E; ggtt to brush borders by using antisera speci- fic for the K88a determinant. Antibodies against K88b and K88c antigens only inhibited homologous strains of K88- positive E_. £32.- It appears that receptors for pili are capable of releasing EEHXEEEQ from brush borders or intestinal cell membranes. Kearns and Gibbons (1979) reported that plasma membranes prepared from positive brush borders lost 98% of their receptor activity; They were able to demonstrate that a supernatant fraction, obtained during the preparation of K88-positive brush borders, was able to enhance the adhesion of K88-positive gt 2211 to both positive and negative brush borders. Supernatant fractions obtained from negative brush borders did not enhance adhesion. Utilizing pig mucosal organ cultures, Staley and Wilson (1983) have demonstrated the release of receptors for K88-positive §;.Egli into culture media. Dean and Isaacson (1982) have identified a 15 soluble 987P pilus receptor-containing fraction which was released from brush borders stored at 4° C. They have shown that receptor activity is heat stable and that the receptor- containing fraction consists of over 40 proteins. Enterotoxins Once attachment and colonization have occurred, ETEC are able to produce enterotoxins. Enterotoxin production is the second important virulence determinant (Smith and Hall, 1967). Enterotoxins are factors that cause the net movement of fluid and electrolytes from plasma to the intestinal lumen when given by the intraluminal route (Dorne gt gt., 1976). Two types of enterotoxins, based on their thermo- lability, antigenicity and molecular weight, are produced by ETEC: 1) a heat stable (ST), nonantigenic enterotoxin with a low molecular weight (Smith and Halls, 1967; Smith and Gyles, 1970) and 2) a heat labile (LT), antigenic entero- toxin with a high molecular weight (Gyles, 1974; Gyles and Barnum, 1969). The heat labile toxin is inactivated by heat at 60° C for 15 minutes and the heat stable toxin is resis- tant to temperatures of 100° C for 15 minutes. lBoth types of enterotoxin production are controlled by transmissible plasmids (Skerman gt gt,, 1972; Smith and Halls, 1968). Strains of ETEC may produce only one or both types of enterotoxin. Strains of ETEC that possess the K88 pilus may be associated with the production of LT, ST or both (Guirée and Jansen, 1979; Gyles and Barnum, 1969; Smith and Gyles, 1970; Saderlind and Mdllby, 1979). The 987P pilus is 16 associated with the production of ST (Guinée and Jansen, 1979b; Moon gtht,, 1980). The K99 pilus is also associated with ST production (Guinée and Jansen, 1979b; Isaacson gt g" 1978b; Moon gtg_l., 1976). Although many different strains of ETEC produce LT, it is a highly homologous substance in all the strains. The LT was considered by Finkelstein gt gt. (1976) to be a hetero- genous molecule whose molecular weight ranged from 35,000 to 100,000 Daltons, but Dorner (1976) and Evans gt gt. (1976) reported a molecular weight of approximately 100,000 Daltons. Finkelstein gt g9, (1976) suggested that the heterogeneity was due to proteolytic splitting of the mole- cule during purification. The LT is similar, antigenically and functionally, to the cholera toxin of Vibrio cholera (Carpenter, 1972; Clements and Finkelstein, 1978; Gyles, 1974; Gyles and Barnum, 1969; Moon, 1971). Both toxins activate membrane bound adenyl cyclase which consequently leads to an increase in intracellular cyclic adenosine 3',5'-mon0phosphate (cAMP) in epithelial cells of the small intestine (Evans gt gt,, 1972; Hewlett, 1974). Subsequently, there is a net secre- tion of an isotonic, alkaline fluid rich in electrolytes as a result of increased sodium secretion and decreased chlo- ride absorption. The increased secretion by the crypt cells leads to hypersecretion and a secretory diarrhea (Moon gt _a_l_., 1978a). Hamilton gt gt. (1978) have tested LT in pig ligated loops and stated that activation of adenyl cyclase 17 by bacterial enterotoxins may not be essential for intes- tinal secretion. The second type of enterotoxin is ST. The ST is con- sidered to be nonantigenic, although it has been shown to act as a hapten. It may be made antigenic by coupling it to a carrier protein (Alderete and Robertson, 1978; Frantz and Robertson, 1981). However, according to Moon gt gt, (1983) when pregnant swine are immunized with STA coupled to a carrier protein, they are not protected against ETEC pro- ducing STA alone. The vaccine stimulated the production of antibodies with high binding activity but low neutralizing activity and thus low protective activity for suckling pigs. The ST is a monomeric octadecapeptide (Lallier gt_gt,, 1982) with a molecular weight of 4,400 (Alderete and Robinson, 1978). There is evidence that suggests that STs from different strains of ETEC represent a heterogenous group of gt ggtt (Guerrant gtgl_., 1975; Steiner gt g_l_., 1972). At least two distinct types of ST are known: 1) STl or STA which is methanol soluble and active in the infant mouse model and 2) STII or STB which is methanol insoluble and inactive in the infant mouse model (Newsome gt gt,, 1978). Utilizing various assay systems, ST doesn't appear to activate adenyl cyclase (Giannella, 1977; Hamilton gt gt., 1978; Sack, 1975; Smith and Gyles, 1970). The mechanism of action of ST involves the activation of guanylate cyclase in the small intestinal epithelial cells (Hughes gt gt., 1978; Newsome gt g_l_., 1978). 18 Resistance EE.§; coli Enteric Infections Neonatal pigs are highly susceptible to diarrhea caused by ETEC. Diarrhea caused by K88-positive ETEC may occur during the neonatal and postweaning period, but diarrhea due to K99-positive ETEC is not known to occur after weaning in pigs three to eight weeks old (Moon, 1978b; Sojka, 1965). As pigs age, they become resistant to ETEC (Moon and Whipp, 1970; Nielsen gt_gt,, 1968; Smith and Halls, 1967). The mechanism of age resistance is unknown. British researchers have shown that pigs may be born resistant to diarrhea caused by K88-positive ETEC (Gibbons gt gt,, 1977; Sellwood gt a1., 1975). Inherited Resistance tg K88-Positive E; coli British researchers have determined that the presence or absence of receptors for the K88 pilus of E; ggtt is inherited. Two alleles are present at a single locus and are inherited in a simple Mendelian manner; One allele, the allele coding for the receptor, is dominant over the other allele. Three genotypes occur: homozygous dominant (SS, susceptible, adherent), heterozygous (Ss, susceptible, adherent) and homozygous recessive (ss, resistant, non- adherent). Two phenotypes occur, susceptible and resistant. The susceptible or positive phenotype is the expression of the dominant allele (Gibbons gt gt,, 1977; Sellwood gt gt., 1975). Phenotypic expression is independent of the age of the pig and is fully established at birth (Gibbons gt gt., 1977). 19 It is difficult to study the genetic aspects of El ggtt resistance since the phenotype usually cannot be observed until the animal is dead (Gibbons gt_gt,, 1977). Recently, though, biopsies by enterotomy have been performed on living animals (Snodgrass _e_t gt., 1981). The simple tr; Egg technique used by Sellwood gt gt. (1975) demonstrates the adhesion of K88-positive E; 221$ to brush borders from pig intestinal cells. Adhesion of bacteria to brush borders occurs in pigs which have the receptors (positive pigs) and no adhesion occurs in pigs lacking the receptors (negative pigs) (Sellwood _e_t_ gt., 1975). With the previous technique, pig litters are divided into three types: 1) non-segregating susceptible (entire litter susceptible), 2) non-segregating resistant (entire litter resistant) and 3) segregating (litter contains piglets of both phenotypes). Gibbons gt gt. (1977) considered that matings which gave a non-segregating resistant litter of more than 6 piglets were derived from homozygous recessive parents. Once the pigs are condition- ally identified as homozygous recessive animals, they are mated to unknown animals and the resulting litter is pheno- typed. The unknown sire or dam is then assigned a provi- sional genotype and phenotype. If the litter segregates, then the unknown parent is a heterozygote. If the litter doesnflt segregate and all piglets are susceptible, then the parent is homozygous dominant. Or, if the litter doesn‘t segregate and all piglets are resistant, then the parent is 20 homozygous recessive. Therefore, genotypes and phenotypes can be assigned to the parents (Gibbons gt gt,, 1977). Rutter gt gt. (1973) have shown experimentally that negative pigs are resistant to infection by K88-positive gt ggtt. K88-positive bacteria colonized the gut of positive pigs more readily than the gut of negative pigs, and the positive piglets were more likely to be susceptible to diarrhea caused by §;.2211- In negative piglets, the organisms were unable to attach and rapidly disappeared from the intestines (Rutter gt gfl,, 1973). Sellwood (1979) reported that in a natural outbreak of scours, negative piglets were resistant to diarrhea caused by E; ggtt. However, susceptibility to infection is a little more complicated in an $2.!$!2 situation. Positive piglets receiving colostrum or milk from the dam that contains K88 antibodies are passively protected (Rutter and Jones, 1973). It is assumed that positive piglets are susceptible to infection, but passive protection due to K88 antibodies in the colostrum prevents enteropathogenic:gthgtt_from attaching and multiplying to high numbers in the small intestines The negative phenotype primarily determines whether an animal will be resistant to infection with K88- positive strains of 3;.2213' but if the positive animals receive antibodies, they may also be resistant to infection (Rutter gt gt., 1973). In herds examined by Gibbons gt gt. (1977), both genes were present with S predominating. The S gene, or suscep- tible gene, is at a disadvantage to the s gene or resistant 21 gene since the environment contains virulent K88-positive E; ggtt. There are many explanations for the persistence of the S gene, some are based on compensatory selection against the homozygous recessive genotype. However, there is really no substantial evidence for compensatory selection. No statistically significant differences between resistant and susceptible animals with respect to the economically impor- tant parameters of growth rate, food conversion efficiency and carcass quality can be shown. There is also no reason to suspect artificial selection by the breeder. The expla- nation proposed is that under normal conditions, piglets are protected from neonatal colibacillosis by antibodies in the colostrum (Gibbons gt gt,, 1977). In the course of an epidemic, the dam becomes immune, and the selective advantage of the 35 genotype becomes reduced. In the initial stages of the disease, when viru- lent K88-positive §;.EQll are introduced into a herd not previously or recently exposed to the organism, there is a high selection in favor of the resistant gene. The suscep- tible breeding sows begin to supply antibodies in their colostrum, but the resistant sows donfit recognize K88- positive bacteria as a pathogen and hence antibody synthesis is not stimulated. Once immunity is established in a herd, the offspring of both homozygous dominants and heterozygotes will be well protected against colibacillosis. The off- spring of homozygous recessives are protected only if they are themselves homozygous recessives, so the selective 22 advantage of the recessive homozygote is reduced. The heterozygous offspring of homozygous recessive dams are at a considerable disadvantage compared to the rest of the pig pOpulation. 'They are genetically susceptible and minimally protected by antibodies in the colostrum or milk. Thus, there is a selection against a particular class of heterozy- gotes, which tends to eliminate the less common gene (Gibbons gt gt., 1977). Most of the studies on the genetic aspects of receptors for K88-positive gt ggtt have been done on pigs in England and Australia. Only a few studies have been performed in the United States. In Michigan, Sher and Waxler (1981a,b) randomly collected intestinal samples from the Michigan State University Department of Animal Science swine herd at the time of slaughter and examined them for K88 receptors by the brush border adhesion technique. Using a strain of gt ggtt carrying the K88ac pilus, 23% of the 109 pigs tested had receptors. When a similar study was carried out by Waxler gt gt, (1983), using 14 litters sired by four boars from the same herd, 38% of the pigs had receptors. The results found by Sher and Waxler (1981a,b) and Waxler gt gt. (1983) suggest that the percentage of pigs with K88 receptors in Michigan is lower than that reported in England (Sellwood gt gt,, 1975) and Australia (Snodgrass gt gt,, 1981). Recently, Bijlsma gt gt. (1982) in the Netherlands have found that there are five phenotypes in swine based on the three serologic variants of the K88 antigen. 23 Table 1. Pig phenotypes based on K88 serologic variants of Escherichia coli. Adhesion of E; coli strains producing: Phenotype K88ab K88ac K88ad A + + + B + + - C + - + D — - + E - _ - The phenotypes differ depending upon whether or not a pig is susceptible to adhesion in the brush border test to three (phenotype A), two (phenotypes B and C) or one (phenotype D) of the serologic variants of K88 or resistant to all three variants (phenotype E). Sellwoodfs adhesion negative phenotype corresponds to the D and E phenotypes, because the K88ad serologic variant of §;.22li was not discovered when he performed his study; The existence of five phenotypes complicates the rather simple genetic model proposed by British researchers and additional work is needed to determine the inheritance pattern of the five phenotypes. In the future, it may be possible to reduce the level of neonatal scours caused by K88-positive gt ggtt in a herd which has not been genotyped by using only resistant boars when breeding sows. Thus, susceptible offspring can only be 24 produced by susceptible dams capable of protecting them by antibody production. IProgeny born of resistant dams will produce resistant offspring which will not require maternal protection. 'There should be an immediate decrease in neonatal scours. With the continued use of resistant boars, the incidence of the resistant genotype should become increasingly common. Genetic protection will then be possible as compared to passive protection (Sellwood, 1979). Biopsy by enterotomy may allow the pig farmer who is reluctant to buy boars of the negative phenotype, which may not meet the standards of conformation or performance of his herd, with the alternative of identifying negative animals within his existing breeding stock (Snodgrass gt gt., 1981). Resistance tg K99-Positive E; coli Moon gt gt. (1979) have stated that they have not encountered pigs congenitally resistant to colonization by E; 2212 carrying the K99 or 987P pili. However, using isolated small intestinal epithelial cells from pigs one day, three weeks and six weeks old, Runnels gt gt, (1980) demonstrated resistance with age to K99-positive E; ggtt. From 8.8 to 14.5% more K99-positive ETEC were shown to adhere to one-day-old piglet epithelial cells than to six- week-old pig epithelial cells. The resistance to adhesion wasn't demonstrable with 3-week-old pigs. 25 Culture Procedures for K99-Positive gt coli $2.!iEEQ studies with K99-positive and 987P-positive ETEC have been hampered by the extreme difficulty in growing and detecting piliated organisms in culture media (Guinée gt __l_., 1976; Isaacsongt_a_l.., 1978a; Drskove_tgt., 1975). In contrast, K88-positive ETEC become piliated relatively easily in several types of ordinary media, such as trypti- case soy broth, under routine growth conditions. Recently, progress has been made in the development of new media and culture procedures which enhance the production and detec- tion of piliated gt ggtt'tg XiEEQo Several factors, such as temperature, oxygen concentration, glucose concentration and amino acid concentration, are capable of influencing the expression of the K99 pilus. The expression of the K99 pilus, and other pili, is temperature dependent. The optimum temperature for pilus production is 37° C (Brinton, 1965; de Graaf gt gt,, 1980b; Isaacson _e_t gt., 1977; Drskov gtgt., 1975). Below 30° C, no significant levels of pili are detectable (de Graaf gt 5g,, 1980b). .At 18° C, pilus production does not occur (Brinton, 1965; Isaacson e_t__a_t., 1977; Drskov gtgt., 1975). Both the subunit synthesis and pilus assembly of the K99 pilus are inhibited at 18° C (Isaacson, 1983). Pilus production is also affected by the oxygen concentration. Under anaerobic conditions, the production of K99 is severely inhibited. Isaacson (1980) noted that when aeration is enhanced due to mechanical shaking, there 26 is a six-fold increase in K99 production, while nonaerated bacteria produce essentially no K99. Generally, the intes- tine is considered to have a low reduction-oxidation poten- tial, so, as postulated by Girardeau gt_gt, (1982b), either special growth conditions are present in the intestine which enable E; ggtt to produce K99 pili at a low reduction- oxidation potential or those bacteria which are attached to the intestinal mucosa are in a relatively aerobic environ— ment in comparison to that of the central lumen. Several ideas concerning why K99 is so difficult to detect have been proposed. Guinée gt_gt, (1976) suggested that the K99 antigen was masked by the abundant capsular antigen or that the enzymes involved in K99 synthesis were inhibited by catabolites of glucose. According to Isaacson gt_gt. (1978b), passing K99-positive ETEC in liquid medium several times leads to the development of acapsular mutants and a corresponding increase in the detectability of K99. However, Girardeau gt gt. (1982b) has observed no associa- tion between K99 detection and the absence or presence of capsular antigens. Isaacson (1980) reported that glucose repressed K99 synthesis, especially at a concentration of 0.5% glucose. However, when cyclic adenosine BEER-monophosphate was added to the medium, the glucose mediated repression of K99 synthesis was overcome. The greatest expression of the K99 pilus occurred at a 5 mM concentration of cAMP. In contrast to what Guinée gt_gt. (1976) had found previously, Isaacson (1980) determined that glucose repression was indeed subject 27 to cAMP-dependent catabolite repression. The difference in results was probably due to the much lower concentration of cAMP used by Guinée gt gt, (1976) compared to the concentration used by Isaacson (1980). Using M2 Minca medium with ammonium sulfate as the nitrogen source, the expression of the K99 pilus in two ETEC strains was not repressed by a high (2%) concentration of glucose. Twelve strains of bacteria were tested by Girardeau gt gt, (1982b). Two different groups, based on their glucose dependency, were found: one group where K99 pilus production occurs on Minca medium without glucose and a second group where K99 is produced only in the presence of glucose. Isaacson (1983) has recently reported that the addition of glucose to the growth media doesn”t‘affect K99 subunit synthesis or pilus assembly. The glucose-dependent expression of K99 is sensitive to the reduction-oxidation potential. Glucose shows a greater inhibitory effect on K99 production when the bacteria are grown under anaerobic conditions instead of aerobic conditions (Girardeau gt gt,, 1982b). The amino acid L-alanine has an inhibitory effect on K99 synthesis (de Graaf gt gt,, 1980a). Concentrations of L-alanine greater than 8 mM lead to more than 80% inhibition of K99 synthesis in media rich in amino acids (Girardeau gt 1., 1982a). Adding L-threonine or L-isoleucine to the media decreases the effects of L-alanine (Isaacson, 1983). 28 Girardeau gt gt, (1982a) believe that L-alanine is inhibitory to K99 synthesis due to its involvement in either enzyme repression or the inhibition of plasmid expression. The reduction-oxidation potential is also important since there was a greater inhibitory effect by L-alanine on K99 synthesis when bacteria are grown under anaerobic rather than aerobic conditions (Girardeau gt gt,, 1982a). Since K99 is usually not detectable on ETEC grown at 37° C on routine culture media designed for the isolation of Enterobacteriaceae (Guinée gt gt,, 1976), several types of synthetic media have been developed to enhance the produc- tion of the K99 pilus antigen. In complex media, the pro- duction of K99 is strongly decreased. Minca, a buffered semi-synthetic medium at pH 7.5, improves the detectability of the K99 antigen, especially when the cultures are subcul- tured two or three times in liquid Minca. Cultures strongly agglutinate with K99 antiserum on primary isolation, but there is only weak agglutination following subculturing on solid Minca (Guinée gt g_l_., 1976). When cultures are grown on solid Minca at 37° C for only 6-8 hours, the detection of K99 is improved. IsoVitalex added to Minca medium helps to improve the detection of K99, possibly by stabilizing the K99 pilus on.§t.ggtt, In solid Minca, the pili may become stuck to the bacterial surface, but in liquid Minca, the pili are more stretched or rigid which enhances their detectability. Therefore, it is beneficial to pass g; ggtt through liquid medium. Detectability of the K99 antigen is improved when cultures are grown for 6-8 hours or 20-24 29 hours on Minca-IsoVitaleX (M1) medium also (Guinée gt gt,, 1977). Frances gt gt, (1982) have found that E medium, a synthetic medium containing essential salts, citric acid and 1% dextrose, is better than blood agar, MI agar, MI broth and T88 (without glucose) for increasing the detectability of K99-positive g; ggtt from diarrheic calves and piglets. They have recommended that several media, such as blood agar, E agar and MI broth, be used when culturing ETEC for pilus typing to enhance the chance of detecting pili. The E agar, according to Frances gt gt, (1982), is less expensive, is less chemically complex and has a much more luxuriant growth pattern for K99-positive Eh ggtt than MI agar. E broth is nearly as effective as MI broth in promoting 987P production. It is not as effective as blood agar, though, in promoting the expression of K88 (Frances gt gt,, 1982). Blood agar promotes the expression of many K88- and 987P—positive strains (Moon gt gt,, 1980). Isaacson (1983) reported that the synthesis of K99 subunits occurs throughout the life cycle of the bacterium and assembled K99 pili appear during the logarithmic phase. .Approximately 92% of all cellular K99 is associated with the outer mem- brane, and 4% is associated with the inner membrane. No K99 is detected in the cytoplasm. 30 Pilus vaccines Currently, several pilus vaccines have been developed for the protection of pigs against neonatal enteric coli- bacillosis. Vaccines are administered parenterally or orally to sows prior to farrowing (Moon, 1981). The vaccine induces protective levels of antibody in the sow‘s colostrum and milk which provide passive immunity to the newborn pig. Research has demonstrated that protection correlates with antipilus antibodies in the colostrum and milk.(Acres gt _a_l., 1978; Morgan gtgt., 1978; Nagy gtgt., 1978; Rutter and Jones, 1973; Rutter gt gt,, 1976). Several researchers believe that colostral antibodies prevent the colonization of ETEC by blocking adhesion (Nagy gt gt,, 1978; Rutter, 1975; Rutter gt gt,, 1973, 1976). Piglets suckling vac- cinated dams are more resistant to infection by ETEC than control pigs when challenged by ETEC with homologous pili (Moon, 1981). According to Nagy gt gt, (1978), mortality, morbidity and the duration of diarrhea are decreased in piglets suckling vaccinated dams. Also, the number of E; ggtt attached to the villous epithelium of the small intes- tine is decreased. Several vaccines which use either purified pili or whole bacterial cell preparations have been developed. Pigs may be infected by strains of ETEC with different pilus antigens, so multivalent vaccines are necessary to provide protection from enteric colibacillosis. Pitman-Moore has developed a trivalent bacterin called Porcimune which 31 contains K88, K99 and 987P piliated strains of gt coli (Anonymous, 1981). Recently, Pitman-Moore has also devel- Oped an gt ggtt bacterin containing 4 piliated strains of gt gt, K88, K99, 987P and F41 (To, 1983). Subunit vaccines, consisting of pili only, have also been developed. A trivalent subunit vaccine which utilizes recombinant DNA gene splicing techniques has been made by Salsbury. The plasmids responsible for pilus production are introduced into a laboratory strain of gt ggtt. The labora- tory strain of gt 221$ then produces many more pili than are normally produced by a wild strain of E;.E£ll- The pili are sheared off from the bacterial surface by mechanical means and incorporated into the vaccine. The endotoxins asso- ciated with bacterins, which cause adverse effects such as allergies, shock and abortions, are excluded from subunit vaccines (Anonymous, 1983). MATERIALS AND METHODS Specimens Samples of ileum, each measuring six to eight inches in length, were obtained from 242 freshly killed pigs at local slaughterhouses. The slaughterhouses were Michigan State University Meats Laboratory, East Lansing, MI (1/20/83- 8/22/83, 60 specimens), Milliganfls Meat Packing Plant, Jackson, MI (7/15/83-8/24/83, 148 specimens) and BainTs Packing Company, Howell, MI (9/7/83-9/14/83, 34 specimens). Ileal samples were carefully flushed with cold 0.15 M NaCl solution to remove mucus and ingesta. They were then placed in plastic bags and packed in crushed ice until further procedures were carried out, usually within two to three hours of the time of slaughter. Brush Border Preparations Intestinal specimens were processed using a modifica- tion of the procedure described by Sellwood gt gt, (1975). In the laboratory, the intestinal specimen was ligated on one end, the lumen was filled with an EDTA buffer solution containing 0.096 M NaCl, 0.008 M KH2P04, 0.0056 M NaZHPO4, (L0015 M KCl and 0.01 M ethylenediaminetetracetate (EDTA), pH 6.8 and the other end of the intestine was clamped with a 32 33 hemostat. The buffer was at room temperature. The intes- tinal specimen was filled with the buffer until slight distention occurred. To prevent drying of the external surface, sections of ileum were immersed in a similar solu- tion containing 0.3 M sucrose instead of EDTA, pH 6.8, for 15-20 minutes at room temperature. ZNext, the EDTA buffer in the intestinal lumen was discarded and replaced with the sucrose buffer until the lumen was half-filled. The epithe- lial cells were detached by gently massaging each intestinal section between the thumb and fingers. The intestinal con- tents were collected in 50 ml plastic centrifuge tubes. The intestine was again half-filled with sucrose buffer and massaged. The latter procedure was repeated until 45-50 m1 of the epithelial cell suspension were obtained. The following steps were performed at 4° C. The epi- thelial cell suspension was centrifuged at 1200 x g for 15 minutesa. The supernatant was removed from each tube, and the pellet was resuspended to a volume of 45-50 ml with cold 0.005 M EDTA, pH 7.4 (adjusted with 0.5 M Na2C03). Next, the suspension was homogenized with a Teflon serrated-tipped tissue grinder (clearance - 0.15-0.23 mm)b by moving the pestle up and down six times while it rotated at approxi- mately 900 rpm. The suspension was centrifuged at 300 x g for 5 minutes, and the supernatant was discarded. The aInternational Refrigerated Centrifuge, Model PR-6, International Equipment Co., Needham, MA. bThomas Co., Philadelphia, PA. 34 pellet was resuspended with the EDTA buffer, and homogeniza— tion and centrifugation were repeated three to five times until the supernatant appeared clear. Next, the pellet was washed three times with Krebs-Henseleit (KH) buffer, pH 7.4, composed of 0.12 M NaCl, 0.014 M KCl, 0.025 M NaHCO3 and 0.001 M KHZPO4. The pellet was then resuspended to eight times its volume with KH buffer and filtered through glass wool. The KH solution containing the brush borders was allowed to stand for 15 minutes to let any mucus present rise to the surface. The mucus was discarded, and the brush border suspension was held overnight at 4° C. For prolonged storage, part of the brush border suspension was mixed with an equal volume of glycerol and stored at -70° C (Bijlsma gt 1., 1982). E. coli Strains The K99-positive gt ggtt used in the brush border adhesion test was strain 431 (0101:K30). The culture, obtained from Dr. R. E. Isaacson, School of Public Health, University of Michigan, Ann Arbor, was maintained in Minca broth (Guinea gt _a_1,, 1977) with IsoVitaleXc (MIB). Cul- tures of gt ggtt 431 were aerobically incubated at 37° C in a shaking water bath5 for 12-16 hours prior to testing for cBaltimore Biological Laboratories, Cockeysville, MD. dPrecision Scientific Co., Chicago, IL. 35 adhesion. The culture was transferred daily from MIB to MIB to insure piliation on the day of testing. The K99-negative strain was isolated from a feline urinary culture, and stock cultures were maintained in tryp- ticase soy brothe (TSB). The cultures of gt ggtt in TSB were incubated aerobically at 37° C for 12-16 hours prior to testing. Following incubation, both cultures were centrifuged at 1500 x g for 20 minutes. The supernatant was discarded, and the bacteria were washed in KH and centrifuged three times. The pellets were then resuspended in 5 ml of KH buffer to a concentration of approximately 1 x 109 colony-forming units per ml. Slide Agglutination A standard slide agglutination test was carried out to verify piliation using K99 antiserum. Prior to this study, the antiserum was produced in Dr. G. L. Waxler's laboratory, Department of Pathology, Michigan State University, East Lansing, MI. The antiserum was produced by hyperimmunizing rabbits with purified pili obtained from Dr. R. E. Isaacson. A 1:10 dilution of K99 antiserum in KH was used for the tests. One drop of each of the bacterial suspensions was placed on a glass slide, a drop of antiserum was added, and the two drops were mixed with a wooden applicator stick. eBaltimore Biological Laboratories, Cockeysville, MD. 36 The glass slide was held near a light source with a magni- fying mirror and slowly rocked and tilted in a circular motion for approximately one minute. Degrees of agglutina- tion were recorded as strong, intermediate, weak and nega- tive. Only those gt ggtt 431 cultures showing strong or intermediate agglutination were used in the brush border adhesion test. Feline cultures which tested negative for agglutination were used. A control slide using KH instead of antiserum was used to check each of the bacterial suspen- sions for autoagglutination. Brush Border Adhesion Test The procedure described by Sellwood gt gt, (1975) was adapted for the identification of intestinal receptors for K99-positive gt ggtt. A 0.2 ml volume of the brush border suspension was mixed with 0.1 m1 of each of the bacterial suspensions in small shell vials. The brush border- bacterial suspension was incubated at 37° C with continuous gentle mixing for 30 minutes. Each §;.Egl$ culture was also incubated with 0.1 ml of KH to determine if autoagglutina- tion was occurring. Initially, a brush border suspension with receptors for K88 pili was mixed with a K88-positive strain of gt ggtt (strain 1248, 0157:K? Hly) as a positive control. In the latter part of the study, a known K99- positive brush border suspension was used as a positive control. Following incubation, a drop of each of the sus- pensions was placed on a glass slide and coverslipped. The 37 slide was then viewed by phase contrast microscopyf with a 40X objective. Brush borders were assigned grades depending upon the numbers of bacteria adhering to the brush border fragments. The grades were as follows: Grade 1 (G1), 0-1 bacterium per fragment; Grade 2 (G2), 2-5 bacteria per fragment; Grade 3 (G3), 6-10 bacteria per fragment; and Grade 4 (G4), >10 bacteria per fragment. Brush border fragments with bacteria covering the entire microvillus surface, regardless of the number of bacteria adhering, were also considered to be Grade 4. Electron Microsc0py To further verify piliation of the K99-positive strain of §£.22121 a few selected cultures of gt ggtt 431 which were piliated according to the slide agglutination test were examined by transmission electron microsc0py. The method used was similar to that described by Chantner (1982). The gt ggtt 431 culture was washed and suspended in distilled water. A drop of bacterial suspension was placed on Formvar carbon-coated gridsg. After 30 seconds, the excess fluid was removed with filter paper. Next, the grids were stained for 15 seconds with 1% sodium phosphotungstateu The excess fZeiss Photomicroscope III, Carl Zeiss, Oberkocken, West Germany. gPelco, Tustin, CA. 38 stain was removed, and the grids were air dried before being examined by transmission electron microscopyh. hZeiss Electron Microscope II, Carl Zeiss, Oberkocken, West Germany. RESULTS Brush Border Preparations The brush border fragments prepared were easily dis- cernible when viewed by phase contrast microscopy. Varying in size, the fragments were semilunar to round in shape with a distinct microvillus border. Characteristically, a halo of bright light was present around each of the brush border fragments. Cell remnants other than the microvillus border also comprised the fragment (Figure 1). Occasionally, within a brush border preparation, the attachment between adjoining epithelial cells would not be disrupted, and large fragments would be present consisting of several cells (Figure 2L. Only small amounts of mucus and debris were present in most brush border preparations. However, in some preparations there was an excessive amount of mucus present. Fragments trapped in mucus or present in large numbers in close association with each other were not used for the determination of adhesion. Bacteria The growth conditions used to enhance piliation of the K99-positive strain of gt coli were successful and fairly 39 40 Figure 1. Phase contrast photomicrograph of a brush border fragment with Grade 1 adhesion. Notice the semilunar appearance of the microvillus surface (arrow) and the halo surrounding the fragment. (400K) Figure 2. Phase contrast photomicrograph of a brush border fragment with Grade 4 adhesion. Notice at the bottom of the micrograph the large aggregate consisting of several adjoining epithelial cell fragments. (400K) Figure 1 Figure 2 42 reliable. On the desired day of testing, piliation was confirmed by slide agglutination on 21 of 29 attempts (72%). Pili were also demonstrated by electron microscopy on the surface of K99-positive bacteria. The pili, measuring 4.5 nm in width (Figure 3), were filamentous structures peritrichously arranged on the bacterial surface. Adhesion Of the 242 ileal samples examined, adhesion of K99- positive gt ggtt was demonstrated in 230 (95%) of the pig intestinal samples. Initially, 24 of the intestinal samples were negative for adhesion. However, adhesion was later demonstrable in 12 of the 24 samples (Table 2). The number of K99-positive gt ggtt adhering to brush border fragments was less than that seen with brush border fragments with receptors for K88 pili but more than that seen with the feline negative control strain. In pigs with receptors for K99 pili, the number of bacteria adhering per brush border fragment varied from 0 to greater than 10 bacteria per brush border (Figures 1, 2, 4 and 5). Brush border fragments were present in the same preparation with no bacteria adhering to some brush border fragments and varying numbers of bacteria adhering to other fragments (Figures 6 and 7). Essentially no feline Eh.£21l adhered to the brush border fragments. However, occasionally, one or two feline bacteria would be trapped adjacent to a brush border fragment. Hence, if only one K99-positive bacterium 43 Figure 3. Electron micrograph of an.§; coli bacterium with K99 pili (arrows) measuring 4.5 nm in diameter. (Phos- photungstic acid stain, 55,980X) Figure 3 45 .ucaHm ochxomm paw: m.:mmfifisflzw .huoumuonmq mumps wuamum>flco oumum :mmfinofiz+ .mcmmsou ucaxomm 992mm,. OH mma 0H NMH mvH HTm H mm o vm ow +N a mm N mm «m «H :ofimoncc oz cofimmnn< :ofimmncd oz :oflmonoa .oz Hmuoa omsonuounmsmHm muasmmm Hmcflm muasmwm HmwuacH o>fiuamomnmmx mo :oflmoapm usonufi3 uo nufi3 memEmm Hmcwummucfi .aHoo mflcofiuonomm mo Hones: new mousom .m ofinme 46 Figure 4. Phase contrast photomicrograph of a brush border fragment (arrow) with Grade 2 adhesion. Notice the two bacteria adhering to the fragment. (400K) Figure 5. Phase contrast photomicrograph of a brush border fragment (arrow) with Grade 3 adhesion. Notice the 8 bacteria adhering to the fragment. Three small bacterial aggregates are also present. (400K) 47 Figure 4 Figure 5 48 Figure 6. Phase contrast photomicrograph of several brush border fragments showing various grades of adhesion. All four grades of adhesion are present. The fragment in the center of the micrograph has Grade 4 adhesion. In the upper left corner, there are two fragments, one with Grade 1 adhesion and the other with Grade 2 adhesion. Three frag- ments are located in the lower right corner, showing Grade 1, Grade 2 and Grade 3 adhesion. An aggregate of brush border fragments is present in the upper right corner. (400K) Figure 7. Phase contrast photomicrograph of several brush border fragments showing various grades of adhesion. The central brush border fragment has Grade 4 adhesion. The fragment on the left has Grade 3 adhesion. Notice the adhesion of bacteria along surfaces other than the micro- villus surface of two of the fragments. Near the bottom of the photomicrograph is a bacterial aggregate with no brush border fragment visible beneath it (arrow). Coadherence is present on the central brush border fragment. (400K) 49 Figure 6 Figure 7 50 was noticed adhering per brush border fragment in a prepara- tion, the sample was considered negative for receptors for K99 pili. Adherence to the brush border fragment was not limited to the microvillus surface. Adhesion also occurred on other surfaces of the brush border fragment (Figure 7). Often, when adhesion wasn't present on the microvillus surface, K99-positive gt ggtt were present on other surfaces (Figure 8). Such attachment did not occur with the feline strain of gt ggtt. Coadherence occurred with the K99-positive bacteria. The bacteria adhered not only to the brush border membrane, but also concurrently adhered to adjacent bacteria which in turn, were adherent to the fragment (Figure 7). Large aggregates of bacteria were present in some preparations containing K99-positive gt ggtt and brush border fragments, especially in preparations refrigerated overnight (Figures 5 and 7). No brush border fragments could be identified beneath the aggregates. Sometimes small bacterial aggregates were seen in preparations with feline bacteria or preparations with K99-positive bacteria alone, but not to the extent of those aggregates seen in brush border preparations with K99-positive bacteria. While the brush border preparations were being examined by phase contrast microscopy, occasionally bacterial aggregates could be seen to detach from brush border fragments and float free in the preparation (Figures 9, 10 and 11). 51 Figure 8. Phase contrast photomicrograph of a brush border fragment with no K99-positive g_._ coli adhering to the microvillus surface but bacteria present on the other surfaces. (400K) Figure 9. A series of three phase contrast photomicro- graphs (Figures 9, 10 and 11) demonstrate the detachment of a bacterial aggregate from a brush border fragment. ‘Notice the bacterial aggregate attached to the microvillus surface of the fragment (arrow). (400X) 52 - ' 'J a . 0 p a u. '1’. ' c' . ' ' t I | o o .l . . 0. C .. ~ 0. .' I . O . . 0 ~ . l‘ ‘ a.” . ‘ ‘ ' ' ' '0 > . - : , . . - ‘ D . o o .0 . § .. ‘ v .. I . . . ’ o . a . o a: . O 0 Q 0 O . a- 0 ..' .- o o . \ . . g } ‘ . _ . 0 o ‘I” ° . O I ° ‘ '5‘?" on ' ' -: v . . ~ v r. ’0 ' ' . a. ’* 'f :3‘. I a s II S ‘ . . . . o . o .0 h . . o 0.0 . ' . .u 0 ‘ 1'. s . . . o I O. . . 0 '° - .:. ‘0 ~ c , ‘ .- . o . u ‘ 0 ' . :. . ' . . 1 O . ‘ - 0'. '. ‘ . Q ’6 ’ . ' ' o . a. . .’ ~ 0 o . a V ' : . t I‘ u a Figure 8 Figure 9 53 Figure 10. Notice that the bacterial aggregate present in Figure 9 has moved from the microvillus surface to the edge of the fragment (arrow). (400X) Figure 11. Notice that the same bacterial aggregate present in Figures 9 and 10 has now moved to the posterior surface of the fragment (arrow). (400K) Figure 10 Figure 11 55 The distribution of the grades of adhesion of bacteria to brush border fragments from the 242 ileal samples examined is illustrated in Figure 12. Brush border fragments stored in glycerol were well preserved. Ten different brush border preparations positive for adhesion when initially tested were reexamined for adhe- sion following their storage in glycerol. The adhesion results obtained from the brush borders stored in glycerol were similar to those seen in fresh preparations. The results of the brush border adhesion test are recorded in Table 3. Using brush border preparations refrigerated for an extended period of time (> 48 hours) or stored in glycerol, the brush border adhesion test was repeated on the 24 ileal samples initially negative for adhesion. The brush border adhesion test was repeated as many as three times on the samples (Table 4). ' When adhesion was demonstrated in a brush border prep- aration, the brush border adhesion test was not repeated on the preparation. Following the use of glycerol stored brush border preparations in the brush border adhesion test, adhe- sion was demonstrated in 12 of the samples which were ini- tially negative for adhesion. 56 Figure 12. Distribution of the grades of adhesion of K99-positive gt coli to brush border fragments in 242 ileal samples. NUMBER OF PIGS 60 50 4O 3O 20 57 Gl- O-l bacterium/brush border fragment GZ‘ 2'5 bacteria/brush border fragment 63' 6-IO bacteria/brush border fragment 64‘ >10 bacteria/brush border fragment fl ,— — ‘1 GI GI GI GI 62’ 62 62 Ga 63 G4 + + + + + + 62 62 62 63 63 G4 1’ f + G3 G3 G4 + G4 BRUSH BORDER GRADES OF ADHESION Figure 12 58 Table 3. Comparison of results of brush border adhesion tests on fresh and glycerol-stored brush border fragments in 10 pigs. Brush Borders Differences in Pig No. Fresha Glycerol Grades of Adhesion# 6 63.64 G2,G3,G4 1 16 20 62 62 = 48 63,64 G2,G3,G4 1 16 68 61,62 61.62 = 98 G2,G3,G4 G1,G2,G3,G4 1 16 113 Gl,G2 61.62.63 1 16 116 61.62.63 61.62.63 = 130 Gl,G2 61.62.63 1 16 131 G3 62.63 1 16 194 Gl,G2,G3 61.62.63 = *The fresh brush borders were stored overnight in the refrigerator prior to testing. += — no change; | 1G - increased by one grade of adhesion; | 16 - decreased by one grade of adhesion. 59 Table 4. Results of brush border adhesion retests on 24 Grade 1 intestinal samples stored in different manners. No. of Type of Storage* Samples Test Results Refrigeration 24 Large bacterial aggre- gates, no visible brush border fragments Glycerol (I) 12 G1 4 61,62 2 G1,G2,G3 2 G2,G3 4 63,64 Glycerol (IIH- 12 G1 *Brush border fragments were refrigerated for an extended period of time (> 48 hours) or brush border fragments were stored in glycerol. +Using new preparations of glycerol stored brush borders, the brush border adhesion test was repeated on the 12 samples stored in glycerol which were Grade 1. DISCUSSION The adaptation of the brush border technique for the detection of intestinal receptors for K99 pili was success- ful. The primary modification consisted of growing the K99- positive strain of 22.221; in liquid Minca medium with IsoVitaleX to enhance piliation. Also to insure piliation, the bacteria were transferred daily in MIB and incubated at 37° C with continuous shaking. On eight test days the bacteria were not piliated according to the slide agglutina- tion test. Four of those times, IsoVitaleX was inadvert- ently excluded from the media. The reason for the lack of piliation in the other four cases is unknown. Minor varia- tions in growth conditions, such as incubation temperature, incubation time, and rate of mechanical shaking may have influenced the state of piliation. Also, the number of broth transfers of the culture may have been a factor. The K99-positive strain of g._ ggl_i_ was also demon- strated to be piliated with electron microscopy; The fila- mentous structures on the bacterial surface were thinner, straighter and more numerous than flagella. The pili also differed from flagella because they did not display sinous waves with a characteristic wavelength. The pili present on the K99-positive strain of gt coli were peritrichously 60 61 arranged and measured 4.5 nm in width. According to Isaacson gt gt. (1981), K99 pili measure < 5 nm in width. The number of K99-positive gt ggtt adhering to brush borders was less than that seen with K88-positive gt_ggtt. Isaacson gt gt. (1978a) observed an average of 22.5 K99- positive bacteria per epithelial cell in one day old pigs, compared to a study by Awad-Masalmeh gt gt, (1982) where 27.2165 adherent K88-positive bacteria were present per epithelial cell. The limited and varied adhesion present in the brush border preparations from slaughter age pigs (4-6 months of age) in this study tends to concur with the results of Runnels gt gt, (1980) who demonstrated an age related resistance to adhesion. The adherence of piliated gt ggtt to surfaces besides the microvillus surface has also been observed by other researchers. Using epithelial cells, Parry and Porter (1978) noticed the attachment of a few piliated gt ggtt along the basement membrane and other cellular membranes. However, most of the bacteria were tightly packed along the brush border. Sellwood (1980) called the attachment of K88- positive bacteria to components of the cell other than the microvilli nonspecific attachment. He was able to destroy the nonspecific attachment by exposing the brush borders to 1% formaldehyde. Aning and Thomlinson (1983) observed that with fresh brush borders, attachment could be seen only along the microvillus surface, However, with older brush borders, there was also attachment to the cytoplasm. Dean 62 and Isaacson (1982), using brush borders and epithelial cells with a 987P-positive strain of gt_ggtt, also noted that adhesion was not restricted to the microvillus surface but occurred on the basolateral membrane as well. They considered that receptors were distributed over the entire cell surface. Some of the brush borders used by Dean and Isaacson were stored for 24 hours. In this study, the brush borders used were usually stored overnight. However, when some fresh brush border preparations were used, there was minor nonspecific adhesion. Nonspecific adhesion did not occur with the feline strain of gt £211; Receptors may be present on the entire epithelial surface and with modifica- tions in the cell surface due to decomposition, the recep- tors for the pili may be exposed. The coadherence seen in this study has also been observed by Dean and Isaacson (1982). They noticed that coadherence was not observed when brush borders were absent or when the nonpiliated mutant of the 987P strain was used. Large bacterial aggregates, lacking identifiable brush borders, were present in this study. Dean and Isaacson (1982), using a 987P-positive strain of §;.Egl£r also noticed large bacterial aggregates in their brush border preparations. They did not see the aggregates when fresh brush borders were used or when the nonpiliated 987P mutant strain was used. .Also, when brush borders were extensively washed, bacterial aggregates were not seen. In this study, large bacterial aggregates were not seen with the feline 63 strain of gt ggtt or in preparations containing only K99- positive bacteria and no brush borders. The presence of the aggregates suggests that by some mechanism, receptors with bacteria attached to them are released from the brush border membrane. The bacteria-receptor complex then forms a clump or an aggregate. Coadherence may occur by a similar mecha- nism. Further evidence for the detachment of bacteria with their receptors was presented when bacterial aggregates were seen detaching from brush borders and floating free in the preparation. Recently, Dean and Isaacson (1982) identified a soluble 987P pilus receptor-containing fraction which was released from brush borders stored at 4° C. Utilizing pig mucosal organ cultures, Staley and Wilson (1983) demon- strated the release of receptors for K88-positive gt ggtt_ into culture media. It appears that receptors for pili may be released tg_!tttg from brush borders and intestinal cell membranes. The release of the receptors could lead to their isolation, purification and chemical identification. From the results obtained with the brush border adhe- sion test, it appears that most, if not all, pigs have receptors for K99 pili. Hence, the results tend to confirm the assumption by Moon gt gt. (1979) that all pigs are congenitally susceptible to colonization by gt ggtt. How- ever, in 12 of the 242 ileal samples tested, no adhesion could be demonstrated. The lack of adhesion in the intes- tinal samples occurred on days when adhesion was present in 64 other brush border preparations, as well as the control, so there is no reason to doubt that the bacteria were piliated. Initially, adhesion could not be demonstrated in 24 of the ileal samples. The brush border adhesion test was repeated on the samples. The first time the test was repeated, brush border preparations which had been refrigerated for an extended period of time (> 48 hours) were used. ‘No discernible brush borders were present in any of the 24 samples tested, but numerous large aggregates consisting of K99-positive bacteria were seen. As stated earlier, the presence of the aggregates suggests the possibility of receptor release with storage. Since bacterial aggregates were present in all 24 intestinal samples, it may be possible that all the samples had receptors for K99 pili. When glycerol-stored brush border preparations were used, adhesion was demonstrable in only 12 of the 24 sam- ples. Eight of the samples had only partial adhesion, G1, GZ and G3, perhaps indicating the presence of only a few receptors. Storage in glycerol may limit receptor release and hence, adhesion could be demonstrated in 12 samples. The 12 G1 samples which had been stored in glycerol were tested again, and no adhesion could be demonstrated. Only a very few or no receptors may have been present due to extreme age-related resistance. Storage in glycerol may not limit receptor release completely and hence, brush border 65 fragments with a minimal number of receptors may lose all of their receptors into solution. Brush border preparations were routinely stored at 4° C overnight. If receptor release does occur, receptors were probably continuously lost into solution prior to testing. If brush border preparations had been tested immediately, adhesion may have been demonstrated in the 24 samples. The retesting of 10 ileal samples initially positive before storage in glycerol, demonstrated that the results obtained from fresh and glycerol stored brush borders were similar. No differences in grades of adhesion were recorded for four samples. In four others, the adhesion results were decreased by one grade which helps to substantiate the fact that minor receptor release may occur with glycerol stored brush borders. Two samples showed an increase of one grade. Possibly, only a few fragments had the higher grade of adhesion and were not noticed in the initial test. The possibility does exist that the 12 pigs were born without receptors for K99 pili. However, it seems unlikely because of the overwhelming number of pigs in which adhesion was demonstrated. Results indicate that most, if not all, pigs have receptors for K99 pili. Therefore, it appears that an inheritance pattern similar to that observed for K88 recep— tors does not exist for K99 receptors. Additional work is needed to clarify the nature of the adhesion and how it may differ from that with K88-positive gt_coli. Also, work is 66 needed on the isolation of a K99-positive receptor- containing fraction which would help characterize the receptor for the K99 pilus antigen. Dean and Isaacson (1982) have partially characterized the 987P receptor and have shown that the receptor is heat stable and that the 987P receptor-containing fraction consists of over 40 proteins. SUMMARY A study of 242 pigs was performed to determine if all pigs have receptors for K99-positive strains of Escherichia ggtt and to determine if an inheritance pattern exists for the receptors. Ileal samples, collected from pigs at three Michigan slaughterhouses, were examined using a brush border procedure modified for the identification of receptors for K99 pili. The primary modification of the brush border procedure consisted of growing the K99-positive strain of gt ggtt in liquid Minca medium with IsoVitaleX to enhance piliation. Adhesion of K99-positive gt ggtt to brush border frag- ments indicated the presence of receptors for K99 pili. Of the 242 samples examined, receptors were demonstrated in 230 (95%) of the intestinal samples. From zero to greater than ten bacteria adhered per brush border fragment. The limited adhesion suggested an age-related resistance. Adhesion of bacteria to surfaces other than the microvillus surface and coadherence also was observed. Following storage of brush border preparations at 4° C, bacterial aggregates lacking identifiable brush border fragments were present in samples tested for adhesion, suggesting the release of K99 receptors from the brush border membrane. 67 68 In this study, the brush border procedure was successfully modified for the identification of K99 receptors. The results obtained from the brush border adhesion test indicate that most, if not all, pigs have receptors for K99 pili. Therefore, it appears that an inheritance pattern similar to that observed for K88 receptors does not exist for K99 receptors. BIBLIOGRAPHY BI BLI OGRAPHY Acres, S.D., Isaacson, R.E., Babiuk, L.A. and Kapitany, IuA.: Immunization of calves against enterotoxi- genic colibacillosis by vaccinating dams with purified K99 antigen and whole cell bacterins. Infect. Immun. 25: 121-126, 1979. Alderete, 03F. and Robertson, ILC.: Purification and chemical characterization of the heat stable entero- toxin produced by porcine strains of enterotoxigenic Escherichia coli. Infect. Immun. 19: 1021-1030, 1978. Altmann, K., Pyliotis, N.A. and Mukkur, T.K.S.: A new method for the extraction and purification of K99 pili from enterotoxigenic Escherichia coli and their characterization. Biochem. J. 201: 505-513, 1982. Anderson, M.J., Whitehead, J.S. and Kim, Y.S.: Interaction of Escherichia coli K88 antigen with porcine intes— tinal brush border membranes. Infect. Immun. 29: 897-901, 1980. Aning, K.G. and Thomlinson, J.R.: Adhesion factor distinct from K88, K99, 987P, CFA I and CFA II in porcine Escherichia coli. Vet. Rec. 112: 251, 1983. Anonymous: Porcimune. Pitman-Moore Inc., Washington Crossing, New Jersey, 1981 (a technical report). Anonymous: Magic bullet for baby pigs. Anim. Nutr. Health 38:22-26, 1983. Arbuckle, J.B.R.: The location of Escherichia coli in the pig intestine. J. Med. Microbiol. 3: 333-340, 1970. Arbuckle, JJLR.: EnterOpathogenic Escherichia coli on the intestinal muc0polysaccharide layer of pigs. Vet. Path. 104: 93-97, 1971. Awad-Masalmeh, M., Moon, H.W., Runnels, P.L. and Schneider, ILA.: Pilus production, hemagglutination, and adhesion by porcine strains of enterotoxigenic Escherichia coli lacking K88, K99 and 987? antigens. Infect. Immun. 35: 305-313, 1982. 69 70 Bertschinger, H.U., Moon, H.W. and Whipp, S.C.: Association of Escherichia coli with the small intestinal epithelium. I. Comparison of enteropathogenic and nonenteropathogenic porcine strains in pigs. Infect. Immun. 5: 595-605, 1972. Bijlsma, I.G.W., de Nijs, A., van der Meer, C. and Frik, CLF.: Different pig phenotypes affect adherence of Escherichia coli to jejunal brush borders by K88ab, K88ac or K88ad antigen. Infect. Immun. 37: 891-894, 1982. Blood, D.C., Henderson, J.A. and Radostitts, O.M.: Veterinary Medicine, 5th ed., Lea and Febiger, Brinton, CKL, Jr.: Non-flagellar appendages of bacteria. Brinton, CAL, Jr.: The structure, function, synthesis and genetic control of bacterial pili and a molecular model for DNA and RNA transport in Gram negative bacteria. Trans. N.Y. Acad. Sci. 27: 1003-1054, 1965. Bruner, D.S. and Gillespie, J.H.: Hagan's Infectious Diseases gt Domestic Animals, Cornell, Ithaca, New York, University Press, p. 139, 1973. Carpenter, ChC., Jr.: Cholera and other enterotoxin related diseases. J. Infect. Dis. 126: 551-564, 1972. Chantner, N.: Structural and functional differences of the anionic and cationic antigens in K99 extracts of Escherichia coli B41. J. Gen. Microbiol. 128: 1585- 1589, 1982. Christie, B.R. and Waxler, G.L.: Experimental colibacil- losis in gnotobiotic baby pigs. II. Pathology. Can. J. Comp. Med. 37: 271-280, 1973. Clements, J.D. and Finkelstein, R.A.: Demonstration of shared and unique immunological determinants in enterotoxins for Vibrio cholera and Escherichia coli. Infect. Immun. 22: 709-713, 1978. Dean, E.A. and Isaacson, R.E.: _I__r_1_ vitro adhesion of piliated Escherichia coli to small intestinal villous epithelial cells from rabbits and the identification of a soluble 987P pilus receptor- containing fraction. Infect. Immun. 36: 1192-1198, 1982. 71 de Graaf, F.K., Klaasen-Boor, P. and van Hees, J.E.: Biosynthesis of the K99 surface antigens repressed by alanine. Infect. Immun. 30: 125-128, 1980a. de Graaf, F.K., Klemm, P. and Gaastra, W.: Purification, characterization and partial covalent structure of Escherichia coli adhesive antigen K99. Infect. Immun. 33: 877-883, 1981. de Graaf, F.K., Wientjes, F.B. and Klaasen-Boor, P.: Production of K99 antigen by enterotoxigenic Escherichia coli strains of antigen groups 08, 09, 020 and 0101 grown at different conditions. Infect. Immun. 27: 216-221, 1980b. Dixon, J.M.S.: The fate of bacteria in the small intestine. J. Pathol. Bacteriol. 79: 131-139, 1960. Dorner, F., Jaksche, H. and Stokl, W.: Escherichia coli enterotoxin: Purification, partial characterization and immunological observations. J. Inf. Dis. 133: 5142-8156, 1976. Drees, D.T. and Waxler, G.L.: Enteric colibacillosis in gnotobiotic swine: A fluorescence microscopic study. Am. J. Vet. Res. 31: 1147-1157, 1970. Duguid, J.P., Smith, I.W., Dempster, G. and Edmunds, P.N.: Non-flagellar filamentous appendages ('fimbriae") and hemagglutinating activity in Bacterium coli. J. Pathol. Bacteriol. 70: 335-348, 1955. Evans, D.J., Jr., Chen, L.C., Curlin, G.T. and Evans, D.G.: Stimulation of adenyl cyclase by Escherichia coli enterotoxin. Nature 236: 137-138, 1972. Evans, D.J., Jr., Evans, D.G. and Dupont, H.L.: Hemaggluti- nation patterns of enterotoxigenic and enteropatho- genic Escherichia coli determined with human bovine, chicken and guinea pig erythrocytes in the presence and absence of mannose. Infect. Immun. 23: 336-346, 1979. Evans, D.J., Evans, D.G., Richardson, S.H. and Gorbach, SJL: Purification of polymixin released, heat labile enterotoxin of Escherichia coli. J. Inf. Dis.133: 397-5102, 1976. Finkelstein, R.A., Larue, M.K., Johnston, D.W., Vasil, M.C., Cho, G.J. and Jones, J.R.: Isolation and preperties of heat labile enterotoxin from enterotoxigenic Escherichia coli. J. Inf. Dis. 133: 8120-8137, 1976. 72 Francis, D.H., Remmers, G.A. and De Zeeuw, P.S.: Production of K88, K99 and 987P antigens by Escherichia coli cultured on synthetic and complex media. J. Clin. Microbiol. 15: 181-183, 1982. Frantz, J.C. and Robertson, D.C.: Immunological properties of Escherichia coli heat-stable enterotoxins: Development of a radioimmunoassay specific for heat stable enterotoxins with suckling mouse activity. Infect. Immun. 33: 193-198, 1981. Fusco, P., To, A., To, S. and Brinton, C., Jr.: The purifi- cation and characterization of four types of Escherichia coli pili and the specificity of Ei coli pili for immunity, colonization and adhesion. In XIIIth UJL-Japan Conference gt Cholera, Atlanta, Georgia, pp. 60-70, 1978. Gaastra, W. and de Graaf, FJL: Host specific fimbrial adhesions of noninvasive enterotoxigenic Escherichia coli strains. Microbiol. Rev. 46: 129-161, 1982. Gaastra, W., Klemm, P. and de Graaf, F.K.: The nucleotide sequence of the K88ad protein subunit of porcine enterotoxigenic Escherichia coli. FEMS Microbiol. Lett. 18: 177-183, 1983. Gaastra, W., Klemm, P., Walker, J.M. and de Graaf, F.K.: K88 fimbrial proteins amino and carboxyl terminal sequences of intact proteins and cyanogen bromide fragments. FEMS Microbiol. Lett. 6: 15-18, 1979. Giannella, R.A.: Specificity of the suckling mouse assay (SMA) for heat stable Escherichia coli enterotoxin (ST) and mode of action of the toxin. Gastroenterology 72: A-39/1062, 1977 (Abstract). Gibbons, R.A., Jones, G.W. and Sellwood, R.: An attempt to identify the intestinal receptor for the K88 adhesin by means of a haemagglutination inhibition test using glyc0proteins and fractions from sow colostrum. J. Gen. Microbiol. 86: 228-240, 1975. Gibbons, R.A., Sellwood, R., Burrows, M.R. and Hunter, P.A.: Inheritance of resistance to neonatal Escherichia coli diarrhoea in the pig: Examination of the genetic system. Theor. Appl. Genet. 51: 65-70, 1977. Girardeau, JJL: A new tg vitro technique for attachment to intestinal villi using enteropathogenic Escherichia coli. Ann. Microbiol. (Inst. Pasteur) 131: 31-37, I980. 73 Girardeau, J.P., Dubourguier, H.C. and Gouet, P.H.: Inhibi- tion of K99 antigen synthesis by L-alanine entero- toxigenic Escherichia coli. J. Gen. Microbiol. 128: 463-470, 1982a. Girardeau, J.P., Dubourguier, H.C. and Gouet, P.H.: Effect of glucose and amino acids on expression of K99 antigen in Escherichia coli. J. Gen. Microbiol. 128: 2243-2249, 1982b. Guerrant, R.L., Moore, R.A., Kirschenfeld, D.M. and Sands, FLA.: Role of toxigenic and invasive bacteria in acute diarrhea of childhood. New England J. Med. 293:567-573, 1975. Guinée, P.A.M. and Jansen, W.H.: Behavior of Escherichia coli K antigens K88ab, K88ac, and K88ad in immuno- electrophoresis, double diffusion and hemagglutina- tion. Infect. Immun. 23: 700-705, 1979a.. Guinée, P.A.M. and Jansen, W.H.: Detection of enterotoxi- genicity and attachment factors in Escherichia coli strains of human, porcine and bovine origin; a com- parative study. Zentralbl. Bacterial. Parasitenkd. Infection Skr. Hyg. Abt. Orig. Reihe A 243: 245-257, 1979b. Guinée, P.A.M., Jansen, W.H. and Agterberg, C.M.: Detec- tion of the K99 antigen by means of agglutination and immunoelectrOphoresis in Escherichia coli isolates from calves and its correlation with enterotoxigenicity. Infect. Immun. 13: 1369-1377, 1976. Guinee, P.A.M., Veldkamp, M.J. and Jansen, W.H.: Improved Minca medium for the detection of K99 antigen in calf enterotoxigenic strains of Escherichia coli. Infect. Immun. 15: 676-678, 1977. Gyles, CJL: Relationship among heat-labile enterotoxins of Escherichia coli and Vibrio cholerae. J. Infect. Dis. 129: 277-283, 1974. Gyles, C.L. and Barnum, D.A.: A heat labile enterotoxin from strains of Escherichia coli enteropathogenic for pigs. J. Infect. Dis. 120: 419-426, 1969. Hamilton, D.L., Johnson, M.R., Forsyth, G.W., Roe, W.E. and Nielsen, NJL: The effect of cholera toxin and heat labile and heat stable Escherichia coli enterotoxin on cyclic AMP concentrations in small intestinal mucosa of pig and rabbit. Can. J. Comp. Med. 42: 327-331, 1978. 74 Hewlett, E.L., Guerrant, R.L., Evans, D.J., Jr. and Greenough, W.E., III: Toxins of Vibrio cholerae and Escherichia coli stimulate adenyl cyclase in rat fat cells. Nature 249: 371-373, 1974. Hughes, J.M., Murad, F., Chan, B. and Guerrant, R.L.: Role of cyclic GMP in the action of heat stable entero- toxin of Escherichia coli. Nature 217: 755-756, 1978. Isaacson, ILE.: K99 surface antigen of Escherichia coli: Purification and partial characterization. Infect. Immun. 15: 272-279, 1977. Isaacson. ILE.: K99 surface antigen of Escherichia coli: Antigenic characterization. Infect. Immun. 22: 555- 559, 1978. Isaacson, ILE.: Factors affecting expression of the Escherichia coli pilus K99. Infect. Immun. 28: 190- 194, 1980. Isaacson, ILE.: Regulation of expression of Escherichia coli pilus K99. Infect. Immun. 40: 633-639, 1983. Isaacson. ILE., Colmenero J. and Richter, P.: Escherichia coli K99 pili are composed of one subunit species. FEMS Microbiol. Lett. 12: 229-232, 1981. Isaacson, R.E., Dean, E.A., Morgan, R.L. and Moon, H.W.: Immunization of suckling pigs against enterotoxi- genic Escherichia coli induced diarrheal disease by vaccinating dams with purified K99 or 987P pili: Antibody production in response to vaccination. Infect. Immun. 29: 824-826, 1980. Isaacson, R.E., Fusco, P.C., Brinton, C.C. and Moon, H.W.: tg vitro adhesion of Escherichia coli to porcine small intestinal epithelial cells: Pili as adhesive factor. Infect. Immun. 21: 392-397, 1978a. Isaacson, R.E., Moon, H.W. and Schneider, R.A.: Distribu- tion and virulence of Echerichia coli in the small intestine of calves with and without diarrhea. Am. J. Vet. Res. 39: 1750-1755, 1978b. Isaacson, R.E., Nagy, B. and Moon, H.W.: Colonization of porcine small intestine by Escherichia coli: Colonization and adhesion factors of pig entero- pathogens that lack K88. J. Infect. Dis. 135: 531- 539, 1977. 75 Isaacson, REE. and Richter, P.: Escherichia coli 987P pilus purification and partial characterization. J. Bacteriol. 146: 784-789, 1981. Jensen, C.C.: C.O. Jensen's Selected Papers, 1886-1908, Vol. 1, Ejaar Munksgaard, Copenhagen, 290, 1948. Cited by Kohler (1972). Jones, G.W. and Rutter, J.M.: Role of the K88 antigen in the pathogenesis of neonatal diarrhea caused by Escherichia coli in piglets. Infect. Immun. 6: 918- 927, 1972. Jubb, K.V.F. and Kennedy, P.C.: Pathology gf_ Domestic Animals, 2nd ed., Vol. 2, Academic Press, New YOrk, pp. 105-109, 1970. Kearns, M. and Gibbons, RJL: The possible nature of the pig intestinal receptor for the K88 antigen of Escherichia coli. FEMS Microbiol. Lett. 6: 165-168, 1979. Klemm, P. and Mikkelsen. In: Prediction of antigenic determinants and secondary structure of the K88 and CFA I fimbrial proteins from enteropathogenic Escherichia coli. Infect. Immun. 38: 41-45, 1982. Kohler, ELM.: Pathogenesis of neonatal enteric colibacil- losis of pigs. J. Am. Vet. Med. Assoc. 160: 574- 582, 1972. Lallier, R., Bernard, F., Gendreau, M., Lazure, C., Seidan, ~ N.G., Chretien, M. and St.-Pierre, S.A.: Isolation and purification of Escherichia coli heat stable enterotoxin of porcine origin. Anal. Biochem. 127: 267-275, 1982. Mooi, F.R. and de Graaf, F.K.: Isolation and characteriza- tion of the K88 antigens. FEMS Microbiol. Lett. 5: Moon, H:W.: Mechanisms in the pathogenesis of diarrhea: A review. J. Am. Vet. Med. Assoc. 172: 443-448, 1978a. Moon, H.W.: Pili as protective antigens in vaccines for the control of enterotoxigenic Escherichia coli infec- tions. In Proceedings, Second International §ympo- sium gg Neonatal Diarrhea, University of Saskatche- wan, Saskatoon, Saskatchewan, Canada, pp. 393-405, 1978b. Moon, Moon, Moon, MOOD, Moon, Moon, Moon, Moon, Moon, 76 HAL; Luminal and mucosal factors of small intestine affecting pathogenic colonization. In Secretory Diarrhea, eds. Field, M., Fordtran, J.S. and Schultz, S.G., American Physiological Society, Bethesda, Maryland, pp. 127-139, 1980. ILWk: Protection against enteric colibacillosis in pigs suckling orally vaccinated dams: Evidence for pili as protective antigens. Am. J. Vet. Res. 42:173-177, 1981. H.W., Baetz, A.L. and Giannella, R.A.: Immunization of swine with heat stable gt coli enterotoxin coupled to a carrier protein does not protect suckling pigs against an gt coli strain that produces heat stable enterotoxin. Infect. Immun. 39: 990-992, 1983. H.W., Isaacson, R.E. and Pohlenz, J.: Mechanism of association of enterOpathogenic Escherichia coli with intestinal epithelium. Am. J. Clin. Nutr. 32: 119-127, 1979. H.W., Kohler, E.M., Schneider, R.A. and Whipp, S.C.: Prevalence of pilus antigens, enterotoxin types and enteropathogenicity among K88 negative enterotoxi- genic Escherichia coli from neonatal pigs. Infect. Immun. 27: 222-230, 1980. H.W., Nagy, B., Isaacson, R.E. and Orskov, 1.: Occur- rence of K99 antigen on Escherichia coli isolated from pigs and colonization of pig ileum by K99 positive enterotoxigenic gt coli from calves and pigs. Infect. Immun. 15: 614-620, 1977. lLW: and Whipp, SJL: Development of resistance with age by swine intestine to effects of enteropatho- genic ggcherichia coli. J. Infect. Dis. 122: 220- 223, 1970. H.W., Whipp, S.C. and Baetz, A.L.: Comparative effects of enterotoxins from Escherichia coli and Vibrio cholera on rabbit and swine small intestine. Lab. Invest. 25: 133-140, 1971. H.W., Whipp, S.C. and Skarvedt, S.M.: Etiologic diag- nosis of diarrheal diseases of calves: Frequency and methods for detecting enterotoxin and K99 antigen production by Escherichia coli. Am. J. Vet. Res. 37: 1025-1029, 1976. 77 Morgan, R.L., Isaacson, R.E., Moon, H.W., Brinton, C.C. and To, CAL: Immunization of suckling pigs against enterotoxigenic Escherichia coli-induced diarrheal disease by vaccinating dams with 987P or K99 pili: Protection correlates with pilus homology of vaccine and challenge. Infect. Immun. 22: 771-777, 1978. Morris, J.A., Stevens, A.E. and Sojka, W.J.: Preliminary characterization of cell free K99 antigen isolated from Escherichia coli B41. J. Gen. Microbiol. 99: 353-357, 1977. Morris, J.A., Stevens, A.E. and Sojka, W.J.: Anionic and cationic components of the K99 surface antigen from Escherichia coli B41. J. Gen. Microbiol. 107: 173- 175, 1978a. Morris, J.A., Stevens, A.E. and Sojka, W.J.: The isoelec- tric point of cell free K99 antigen exhibiting hemagglutinating properties. Infect. Immun. 19: 1097-1098, 1978b. Morris, JtA., Thorns, CLJ., Scott, A.C., Sojka, W.J. and Wells, G.A.: Adhesion tg vitro and tr; vivo associated with an adhesive antigen (F41) produced by a K99 mutant of the reference strain Escherichia coli B41. Infect. Immun. 36: 1146-1153, 1982. Morris, J.A., Thorns, C.J. and Sojka, W.J.: Evidence for two adhesive antigens on the K99 reference strain Escherichia coli B41. J. Gen. Microbiol. 118: 107- 113, 1980. Morris, J.A., Thorns, C.J., Wells, G.A.H., Scott, A.C. and Sojka,‘WJL: The production of F41 fimbriae by piglet strains of enterotoxigenic Escherichia coli that lack K88-fimbriae, K99-fimbriae and 987P- fimbriae. J. Gen. Microbiol. 129: 2753-2759, 1983. Nagy, 8., Moon, H.W. and Isaacson, R.E.: Colonization of porcine small intestine by Escherichia coli: Ileal colonization and adhesion by pig enteropathogens that lack K88 antigen and by some acapsular mutants. Infect. Immun. 13: 1214-1220, 1976. Nagy, 8., Moon, H.W. and Isaacson, R.E.: Colonization of porcine intestine by enterotoxigenic Escherichia coli: Selection of piliated forms tg vivo, adhesion of piliated forms to epithelial cells tg vitro and incidence of a pilus antigen among porcine entero- pathogenic gt coli. Infect. Immun. 16: 344-352, 1977. 78 Nagy, 8., Moon, H.W., Isaacson, R.E., To, C.C. and Brinton, (LC.: Immunization of suckling pigs against entero- toxigenic Escherichia coli infection by vaccinating dams with purified pili. Infect. Immun. 21: 269- 274, 1978. Newsome, P.M., Burgess, M.N., Bywater, R.J., Cowley, C.M. and Mullan, N.A.: Studies of the biological activities and mechanism of action of heat stable E; coli enterotoxins. In Proceedings, Second International §ymposium gt Neonatal Diarrhea, University of Saskatchewan, Saskatoon, Saskatchewan, Canada, pp. 119-133, 1978. Nielsen, N.O., Moon, H.W. and Roe, W.E.: Enteric colibacil- losis in swine. J. Am. Vet. Med. Assoc. 153: 1590- 1606, 1968. Drskov, I. and Drskov, F.: Episome carried surface antigen K88 of Escherichia coli: 1. Transmission of the determinant of the K88 antigen and influence on the transfer of chromosomal markers. J. Bacteriol. 91: 69-75, 1966. Orskov, I., Drskov, F., Smith, H.W. and Sojka, W.J.: The establishment of K99, a thermolabile, transmissible Escherichia coli K antigen previously called “Kco", possessed by calf and lamb enteropathogenic strains. Acta Pathol. Microbiol. Scand. 83: 31-36, 1975. Drskov, I., Orskov, F., Sojka, W.J. and Leach, J.M.: Simul- taneous occurrence of gt coli B and L antigens in strains from diseased swine. Acta Pathol. Micro- biol. Scand. 53: 404-422, 1961. Orskov, I., Drskov, F., Sojka, W.J. and Wittig, W.: K antigens K88ab (L) and K88ac (L) in g._ coli. Acta Pathol. Microbiol. Scand. 62: 439-447, 1964. Ottow, JJLG.: Ecology, physiology and genetics of fimbriae and pili. Ann. Rev. Microbiol. 29: 79-108, 1975. Parry, S.H. and Porter, P.: Immunological aspects of cell membrane adhesion demonstrated by porcine entero- pathogenic Escherichia coli. Immunology 34: 41-49, 1978. Runnels, P.L., Moon, H.W. and Schneider, R.A.: Development of resistance with host age to adhesion of K99 positive Escherichia coli to isolated intestinal epithelial cells. Infect. Immun. 28: 298-300, 1980. 79 Rutter, J}M.: Escherichia coli infections in piglets: Pathogenesis, virulence and vaccination. Vet. Rec. 96: 171-175, 1975. Rutter, J.M., Burrows, M.R., Sellwood, R. and Gibbons, R.A.: A genetic basis for resistance to enteric disease caused by Escherichia coli. Nature 257: 135-136, 1973. Rutter, J.M. and Jones, G.W.: Protection against enteric diseases caused by Escherichia coli: A model for vaccination with a virulence determinant? Nature 242: 531-532, 1973. Rutter, J.M., Jones, G.W., Brown, G.T.H., Burrows, M.R. and Luther, PJL: Antibacterial activity in colostrum and milk associated with protection of piglets against enteric disease caused by K88-positive Escherichia coli. Infect. Immun. 13: 667-676, 1976. Sack, RJL: Human diarrheal disease caused by enterotoxi- genic Escherichia coli. Ann. Rev. Microbiol. 29: 333-351, 1975. Schneider, R.A. and To, S.C.M.: Enterotoxigenic Escherichia coli strains that express K88 and 987P pilus antigens. Infect. Immun. 36:417-418, 1982. Sellwood, R.: Escherichia coli diarrhoea in pigs with or without the K88 receptor. Vet. Rec. 105: 228-230, 1979. Sellwood, R.: The interaction of the K88 antigen with porcine intestinal epithelial cell brush borders. Biochem. et Biophys. Acta 632: 326-335, 1980. Sellwood, R., Gibbons, R.A., Jones, G.W. and Rutter, J.M.: Adhesion of enteropathogenic Escherichia coli to pig intestinal brush borders: The existence of two pig phenotypes. J. Med. Microbiol. 8: 405-411, 1975. Sellwood, R. and Kearns, M.: Inherited resistance to Escherichia coli diarrhea in pigs: The genetics and nature of the intestinal receptor. In Frontiers gt Knowledge tg the Diarrhoeal Disease, eds. Janowitz, H.D., Sachar, D.B., Projects in Health, Inc., Upper Monclair, New Jersey, pp. 113-122, 1979. Sher, 3.8. and Waxler, G.L.: Inheritance of ggcherichia coli K88 receptors in the intestine of the pig. In Sixty-Second Annual Meeting Conference gt Research Workers tg Animal Disease, Chicago, Illinois, p. 1, 1981 (Abstract). 80 Sher, S.B. and Waxler, G.L.: Inherited resistance to Escherichia coli diarrhea in the pig. Report of Swine Research, 1981. Research Report 437, Michigan Agricultural Experiment Station, East Lansing, pp. 68-69. 1981. Shipley, P.L., Gyles, C.L. and Falkow, 8.: Characterization of plasmids that encode for the K88 colonization antigen. Infect. Immun. 20: 559-566, 1978. Skerman, F.J., Formal, S.B. and Falkow, S.: Plasmid associated enterotoxin production in a strain of Escherichia coli isolated from humans. Infect. Immun. 5: 622-624, 1972. Smith, H.W. and Gyles, C.L.: The relationship between two apparently different enterotoxins produced by enteropathogenic strains of Escherichia coli of porcine origin. J. Med. Microbiol. 3: 387-401, 1970. Smith, H.W. and Halls, 5.: Observations by the ligated intestinal segment and oral inoculation methods on Escherichia coli infections in pigs, calves, lambs and rabbits. J. Path. Bact. 93: 499-529, 1967. Smith, H.W. and Halls, 8.: The transmissible nature of the genetic factor in Escherichia coli that controls enterotoxin production. J. Gen. Microbiol. 52: 319- 334, 1968. Smith, H.W. and Jones, J.E.T.: Observations on the alimen- tary tract and its bacterial flora in healthy and diseased pigs. J. Path. Bact. 86: 387-412, 1963. Smith, H.W. and Linggood, M.A.: Further observations on Escherichia coli enterotoxins with particular regard to those produced by atypical piglet strains used by calf and lamb strains: The transmissible nature of these enterotoxins and of a K antigen possessed by calf and lamb strains. J. Med. Microbiol. 5: 243- 250, 1972. Snodgrass, D.R., Chandler, D.S. and Makin, T.J.: Inheri- tance of Escherichia coli K88 adhesion in pigs: Identification of a nonadhesive phenotype in a com- mercial herd. Vet. Rec. 109: 461-463, 1981. deerlind, D. and Mdllby, R.: Enterotoxins, O-groups and K88 antigen in Escherichia coli from neonatal pigs with and without diarrhea. Infect. Immun. 24: 611- 616, 1979. 81 Sojka, WALa Escherichia coli in domestic animals and poultry. Commonwealth Agricultural Bureau, Weybridge, England, 1965. Staley, T.E., Jones, E.W. and Corley, L.D.: Attachment and penetration of Escherichia coli into intestinal epithelium of the ileum in newborn pigs. Amer. J. Pathol. 56: 371-392, 1969. Staley, T. E. and Wilson, T. B.: Recovery of intestinal membrane binding sites for K88 E_._ c__g_li from pig mucosal organ cultures. Fed. Proc. 42: 507, 1983 (abstract). Steiner, A.L., Parker, C.W. and Kipnis, D.M.: Radioimmuno- assay for cyclic nucleotides. J. Biol. Chem. 247: 1106-1113, 1972. Stirm, S., Drskov, F., Orskov, I. and Birch-Anderson, A.: Episome-carried surface antigen K88 of Escherichia coli. III. Morphology. J. Bacteriol. 93: 740-748, 1967a. Stirm, S., Drskov, F., Orskov, I. and Mansa, B.: Episome- carried surface antigen K88 of Escherichia coli. II. Isolation and chemical analysis. J. Bacteriol. 93: 731-739, 1967b. To, SAL: Prevention of colibacillosis in neonatal swine with a 4 pilus E. coli bacterin. Mod. Vet. Practice 65: 39-41,1984. Wadstrbm, T., Smyth, C.J., Faris, A., Johnson, P. and Freer, JJL: Hydrophobic adsorptive and hemagglutinating properties of enterotoxigenic Escherichia coli with different colonizing factors: K88, K99 and coloni- zation factor antigens and adherence factors. In Proceedings of the International Symposium on Neonatal Diarrhea, Stuart Brandle Publishing Service, Edmonton, Canada, pp. 29- -55, 1979. Waxler, G.L., Grimes, S.D., Sher, S.B. and Hogberg, M.G.: Genetic resistance to Escherichia coli diarrhea in the pig. Report of Swine Research, 1983. Research Report 456, Michigan Agricultural Experiment Station, East Lansing, pp. 62-64, 1983. Wilson, M.:R“: Enteric colibacillosis. In Diseases of _S__wi__ne, eds. Leman, A.D., Glock, R. D., Mengeling, W. —L., Penny, R. H. C., Scholl, E. and Straw, 8., 5th ed., Iowa State University Press, Ames, Iowa, pp. 471-477, 1981. 82 Wilson, M.R. and Hohmann, A.W.: Immunity to Escherichia coli in pigs: Adhesion of enteropathogenic Escherichia coli to isolated intestinal epithelial cells. Infect. Immun. 10: 776-782, 1974. VITA The author was born in Tuskegee Institute, Alabama on February 24, 1958. Her primary and secondary education was completed in several schools located throughout the United States and Germany; She graduated from Tuskegee Institute's School of Veterinary Medicine in the spring of 1982. In the summer of 1982, the author entered a graduate program at Michigan State University. She plans to complete the Ph.D. degree at the same institution. 83