Access to all articles, new health classes, discounts in our store, and more!
Bacteriology of the Intestinal Tract in Certain Diseases: II. The Possible Inhibition of Colon Bacilli by Pathogenic Streptococci and Staphylococci
Published in The Review of Gastroenterology, Vol. 5, No. 3, September 1938, pp. 234-240. Co-Author: Clarence W. Lieb, M.A., M.D.
* * *
In the previous paper1 it was shown that there was insufficient evidence to incriminate the presence of sporulating anaerobes, Friedlander bacilli, Aerobacter aerogenes, Escherichia coli (B. coli), degraded colon bacilli (paracoli), and the absence of aciduric organisms in the etiology of “intestinal intoxication”, colitis or irritable colon.
The role of B. coli toxins was thought to be a minor one because there was considerable reduction in the number of colon bacilli in many patients with intestinal complaints. In this paper we shall try to show that reduction in the number of B. coli, suggestive of inhibition, is related to the presence of pathogenic streptococci in the feces and pathogenic staphylococci in the nasal and oral cavities.
As in the previous paper, the term “colon bacilli” will be used for the group as a whole, including Escherichia, Aerobacter and non-lactose fermenters. The term “B. coli” will be used to designate the genus Escherichia, in harmony with the trend of recent bacteriologic literature to eliminate “species” based on fermentation reactions and to consider members of the genus Escherichia as a single unit.
Inhibition of B. Coli in Feces
Felsen2 reported a case in which B. coli was absent in daily stool cultures over a period of 2 months, and mentioned a similar observation by Herter and Kendall. In our laboratory a large number of instances have been noted in which the number of B. coli was low constantly. A few examples are given in table 1.
It is difficult to establish the normal number of B. coli because of the difficulty of establishing that “normal” persons are free from bacterial infection or intoxication, and because factors affecting variations in the number of colon bacilli are but imperfectly understood. Therefore, it is impossible to make comparative studies using “normals” as controls.
Stabler and Pemberton3 observed a greater number of “coliforms” in the stools of 17 patients with arthritis than in 9 “normal” persons. However, there are several reasons for rejecting their findings. Lactose broth, which they used for initial isolation, is unacceptable from the quantitative standpoint. The number of determinations was too small for statistical analysis, particularly in view of the extreme variations in different patients and in the “controls”.
Stools which showed gross fermentation, considered evidence of small intestinal hypermotility by Owles,4 contained the smallest number of colon bacilli (table 2). The number was larger in semisolid stools and largest in formed specimens. This suggests that, the longer the specimen is in the intestinal tract, the larger the number of colon bacilli.
Clumps of fecal material may produce marked irregularities in the distribution of colon bacilli in different parts of the sample. The technic used to obtain uniform distribution was suggested by Mr. H. G. Dunham of Difco Laboratories, viz. a wad of absorbent cotton was placed in the mouth of a test tube containing the specimen and pushed to the bottom of the tube by means of a sterile pipette. The suspension filtered through the cotton and was free from clumps.
The distribution of specimens with high, medium and low numbers of total colon bacilli was not materially different in Stabler and Pemberton’s3 arthritis series, one of our earlier series, and a recent series (table 3), which suggests that the number is independent of any particular disease.
Bile did not appear to be a factor in controlling the number of colon bacilli in feces because staphylococci and the colon group grew quite well in undiluted ox bile, while streptococci were inhibited, even when it was diluted 1:1000.
The number of B. coli seemed to be inversely proportional to the severity of the intestinal disease, although this would be difficult to prove because of lack of suitable clinical methods. If true, it would suggest that factors responsible for inhibition of B. coli might be related to the toxi-infectious agent.
It had been shown by Chapman5 that the electrophoretic migration velocity (P. D.) remained constant and could be used to identify a strain. Dozois6 apparently construed this to mean that B. coli could be differentiated from A. aerogenes by this method. However, it was intended to mean that a strain with such characteristic velocity as our # 87, which was used for “implantations”, could be recognized in cultures, even though it might be mixed with other strains of B. coli. Conversely, strains with dissimilar velocities are not likely to be related. By this method, and from counts of the number of B. coli, it was possible to show on numerous occasions that the strain used for “implantation” did not remain in the intestinal tract. However, when steps were taken to overcome factors considered responsible for inhibition of B. coli (see paper III of this series), the number of B. coli returned to normal, even without “implantation”.
Since this organism could not be implanted by rectal instillation, the relief from symptoms attributed to bacterial infection suggested an antigenic effect, rather than to the ability of B. coli to dominate the intestinal flora. This was supported by the observation that best results were obtained after the culture had passed the logarithmic, or most vigorous, phase of growth.
Relationship Between Pathogenic Streptococci and Staphylococci and the Number of Colon Bacilli
It was noted that the administration of streptococcal vaccines often was followed by considerable increase in the number of B. coli. This suggested that streptococci might be implicated in some way in the inhibition of B. coli in the feces. The experiments to be described offer evidence in favor of this hypothesis. Since there was a possibility that staphylococci, which are known to affect the gastrointestinal tract, also might affect the number of colon bacilli, search was made for them also.
Cultures were obtained from patients with different types of chronic disease; most of whom had clinical evidence of focal infection. Since they are rarely found in levels below the duodenum probable pathogenic staphylococci were isolated from nose and throat cultures, using the methods of Chapman, Lieb, Berens and Curcio,7 and Chapman, Lieb and Curcio,8 and tested for pathogenicity by the methods of Chapman, Berens, Nilson and Curcio.9 Cultures also were made of the feces of these patients, counting the number of B. coli, A. aerogenes and non-lactose fermenters (paracoli) by plating serial dilutions on Bacto Levine’s eosin methylene blue medium. Streptococci were isolated from blood agar plates, differentiated from enterococci,10 and tested for probable pathogenicity by the method of Chapman and Curcio.11
When the results were arranged according to the presence of pathogenic streptococci in the feces and pathogenic staphylococci in the nose and throat (table 4), there was an inverse relationship between the average number of total colon bacilli, the presence of pathogenic streptococci and, to a certain extent, pathogenic staphylococci. The lowest count of total colon bacilli was associated with the presence of pathogenic streptococci in the feces and pathogenic staphylococci in the nose and throat. In the second group, which was similar to the first except that pathogenic staphylococci were not found, there were more than twice as many B. coli as in the first group. The number of B. coli was highest in the third and fourth groups, with no pathogenic streptococci in the feces.
The degradation of Aerobacter to paracoli was discussed in the previous paper,1 and was attributed to the harmful influence of pathogenic streptococci and, to a slight extent, pathogenic staphylococci. The failure of staphylococci to reach lower levels of the gastrointestinal tract may account for the minor influence of this organism. The smallest number of paracoli and, consequently, the least degeneration of colon bacilli, occurred in the group from which no pathogenic streptococci were isolated from the feces and no pathogenic staphylococci from the nose and throat. The highest number of paracoli, representing the greatest degradation, was found in the group from which pathogenic streptococci were found in the feces and pathogenic staphylococci in the nose and throat (table 4).
Out of 34 specimens with pathogenic streptococci in the feces, 83.3 per cent had less than 100,000 million total colon bacilli. In the absence of pathogenic streptococci only 53.6 per cent of 85 specimens had less than 100,000 million total colon bacilli per 100 grams dry feces.
These results suggest that the average number of total colon bacilli bears an inverse relationship, and the average number of paracoli a direct relationship, to the presence of pathogenic streptococci in feces, and to a lesser extent to the presence of pathogenic staphylococci in the nose and throat. However, the correlation did not hold in individual specimens or small series, as shown in table 5. This may be explained by errors due to random sampling, differences in consistency of the specimens, difficulties in isolating pathogenic streptococci, differences in oxidation-reduction potential of the feces, etc.
Excessive Numbers of Colon Bacilli
In some instances, a stool from a patient with intestinal symptoms showed a considerable increase in the number of B. coli. It could not be demonstrated that the strains were capable of growing more luxuriantly than other strains in streptococcal filtrates. Therefore, the excess numbers in these specimens were probably not due to greater resistance of the strains to streptococcal inhibiting substance, and must have been due to other factors.
In searching for possible explanations it was recalled that one of the characteristics of the intestinal flora of patients with pernicious anemia, where the number of fecal bacteria is characteristically high, is the ascendency of the colonic flora into the duodenum, possibly connected with the gastric anacidity which is a constant finding in this disease. Thus, an abnormally high count of colon bacilli in the feces could be due to proliferation of this organism in the duodenum. This would necessitate postulating that the inhibiting agent was inactivated or rendered ineffective in the duodenum. When insufficient acid is produced in the stomach, it is likely that the alkalinity of the duodenal fluid will be only slightly reduced. Such an excessively alkaline fluid might inactivate streptococci, which tolerate a narrower range of pH than does B. coli.
In certain patients, there was an excessive number of B. coli, although the patients were ill for several years and had streptococcal foci of infection. When the illness became more severe, and the foci showed evidence of increased activity, the number of B. coli in the feces was materially reduced. This suggests that originally the foci were either closed or mild in character.
Thus, an abnormally large number of colon bacilli in the feces could be attributed to: (1) Absence or diminution of pathogenic streptococci, or a metabolite produced by them, in the gastrointestinal tract; or (2) ascendency of the colonic flora into the duodenum, as a possible result of gastric anacidity. The subject is being studied further.
Summary
The relationships between streptococci, staphylococci and colon bacilli can be summed up as follows. Except in certain conditions dependent upon little understood changes in the upper respiratory or gastrointestinal tracts, pathogenic streptococci in the gastrointestinal tract, and to a certain extent pathogenic staphylococci in the upper respiratory tract, produce marked inhibition of colon bacilli. Along with this, colon bacilli, particularly Aerobacter, are degraded to paracoli.
The excessive number of colon bacilli in certain stools was attributed to phenomena connected with the upper gastrointestinal tract.
These findings suggest that pathogenic streptococci and staphylococci may be responsible for changes in the number and biochemical properties of commensal intestinal organisms (colon bacilli). It would be logical to expect fundamental changes in gastrointestinal physiology as a result of these influences. Thus, pathogenic streptococci and staphylococci may play a significant role in gastrointestinal symptomatology. This possibility will be considered in the following paper.
References Cited:
- Lieb, C. W., Chapman, G. H., Rawls. W. B. and Stiles, M. H.: “Bacteriology of the intestinal tract in certain chronic diseases. I. Sporulating anaerobes, aciduric organisms and the colon group.” Rev. Gastroent. 5: 142.
- Felsen, J.: “Intestinal oxygenation in idiopathic ulcerative colitis; preliminary report.” Arch. Int. Med. 48: 786-792, Nov. 1931.
- Stabler, S. H. and Pemberton, R.: “A note on the intestinal bacteria in chronic arthritis.” Am. J. Med. Sci. 193: 233-241, Feb. 1937.
- Owles, W. H.: “The etiology of intestinal carbohydrate dyspepsia.” Guy’s Hosp. Rep. 86: 411-9, Oct. 1936.
- Chapman, G. H.: “Electrophoretic potential as an aid in identifying strains of the B. coli group.” J. Bact. 18: 339-342, Nov. 1929.
- Dozois, K. P.: “Variations in electrophoretic mobilities of Escherichia, Aerobacter and ‘intermediate’ strains.” J. Bact. 31: 211-5, Feb. 1936.
- Chapman, G. H., Lieb, C. W., Berens, C. and Curcio, L.: “The isolation of probable pathogenic staphylococci.” J. Bact. 33: 533-543, May 1937.
- Chapman, G. H., Lieb, C. W. and Curcio, L. G.: “The use of bromothymol blue agar and phenol red mannitol agar for the isolation of pathogenic types of staphylococci.” Am. J. Clin. Path., Tech. Suppl. 2: 3-11, Jan. 1938.
- Chapman, G. H., Berens, C., Nilson, E. L. and Curcio, L. G.: “The differentiation of pathogenic staphylococci from non-pathogenic types.” J. Bact. 35: 311-334, March, 1938.
- Chapman, G. H.: “Studies of streptococci. IV. Resistance of enterococci.” J. Bact. 32: 41-6, July 1936.
- Chapman, G. H. and Curcio, L.: “Studies of streptococci. lI. Quantitative differences in resistance to sodium bicarbonate and hexylresorcinol.” J. Bact. 31: 333-7, April 1936.