Changes in the serologic relationships of strains of escherichia coli isolated from humans

Thesis (M.A.)--Boston UniversityAlthough Escherichia coli is a normal inhabitant of the intestinal tract of man and animal, a great deal of interest has been shown in this organism since it was first described in 1886. Early investigations of the significance of E. coli in man and animals were hampered by the lack of adequate serological methods. The significance of E. coli in scours, a diarrheal condition of calves, was shown by Jensen (Handbuch der pathogenen Mikroorganismen von Kolle-Wasserman VI:121, 1913). He found two types of E. coli present in calf feces, one of which, type A, was present in the diseased animal and not in the healthy one. The pathogenic strains, however, could not be determined on the basis of fermentation reactions. Calves appeared to be the only experimental animal suitable for differentiating between the pathogenic and non-pathogenic strains. With the decline of the incidence of specific diarrhea caused by the salmonellae and shigellae groups, considerable interest has been shown in the role of E. coli as the etiologic agent of diarrhea of infants. The isolation of strains of E. coli from infants with diarrhea again showed that pathogenic and non-pathogenic strains could not be differentiated by fermentation reactions. During the 1940's Kauffman (J. Immunol., 57: 71, 1947) established an antigenic scheme of classifying E. coli. All strains possessed a thermostable 0 antigen which is enveloped by a K antigen. The K antigens constitute a group which occurs as envelopes, capsules or sheaths. This group is further divided into L, A and B antigens, depending upon their biochemical and serologic properties. The motile strains possess an H antigen. At the present time 14 different serotypes associated with E. coli diarrheal disease have been reported. The antigenic diversity of E. coli has proved helpful in determining the persistence of intestinal tract of man. In a study of two individuals over a period of of four months (Kauffman and Perch, Acta path. microbiol. Scand 20: 201, 1943) one to three different antigenic groups were found in each specimen from one subject. In the other subject the same antigenic strain was found in ten successive specimens in addition to one or more transient strains. In a study of one individual for 15 months (Wallick and Stuart, J. Bact., 45: 121, 1943) 650 E. coli strains were isolated, 85.3 per cent being antigenically identical. Sears et al. (J. Bact., 59: 293, 1950) reported on four subjects studied for a period from three months to 2 1/2 years, and presented additional evidence that both resident and transient strains of E. coli are found in the normal intestinal flora. Attempts to change the resident strains by feeding of large numbers of organisms were unsuccessful. The ingested organisms recovered for short periods of time and then disappeared, which placed them in the category of transient strains. The changes in the serologic relationships of E. coli isolated from two normal individuals were studied in our investigation by the agar-diffusion precipitin technique of Ouchterlony (Acta path. microbiol. Scand., 25: 186, 1948). The expected continuity of a dominant antigenic structure was found throughout the 10 months' period of study of subject P. In subject B a dominant strain was found in the first four specimens which disappeared and was not re-isolated throughout the study period. The period in which a dominant strain of E. coli was present was followed by on in which the majority of the isolated strains were related to the original strain. Many unrelated strains were also present. The agar-diffusion preipitin technique was found to be a sensitive method of determining antigenic relationships of strains of E. coli even when the conventional tube agglutination method indicated no relationship at significant titer levels