M. J. Schlesinger

Carbohydrate fermentation by bacteria as influenced by the composition of the medium

The value of the fermentation test among the methods at our disposal for identification of bacteria is generally accepted. And yet, every bacteriologist must have encountered in his experience a number of instances of apparently inexplicable inconsistencies in the results obtained by this test. Not only does it often happen that a given strain producing a large amount of acid or gas will occasionally produce very little, but at times indeed it produces none at all. In fact, the amount of gas produced by a bacterium at different times varies so widely, that at present it is suggested by some bacteriologists that the amount of gas produced by a given culture has no diagnostic significance. This point of view owes its existence merely to the fact that the amount of gas produced by a given culture has no diagnostic significance. This point of view owes its existence merely to the fact that the amount of gas produced by bacteria depends on too many factors to attempt to control them. In our work we came across inconsistencies in the amount of acid and gas production, but discovered that these inconsistencies were very often due to variations in the composition of the media. The study is indeed not finished, but even in its present stage it is quite convincing. Omitting the details of the experiment, which will be published in full later, we shall state here merely the general plan and the results obtained. The experiment consisted in growing a strain of B. coli, which, in the original culture, produced very little acid or gas, upon a medium consisting of peptone-phosphate-lactose-water with the addition of an indicator permitting direct reading of hydrogen ion concentration developed in the growing culture. 1

Concerning anaphylaxis following the administration of diphtheria antitoxin

Sensitiveness of human beings to horse protein is fairly widespread as indicated by numerous reports of cases of serum sickness following administration of various therapeutic sera. The impression is, however, that in diphtheria the danger from this source is particularly slight. This comparative freedom of complications of anaphylactic nature following the administration of diphtheria antitoxin in emergency during the War has led to abandoning the preliminary skin test for sensitiveness in certain medical units. 1 We attempted to approach experimentally this question of apparent tolerance to anaphylaxis during diphtheria intoxication. We have observed that sensitized guinea pigs receiving subcutaneously large excess of diphtheria toxin withstand the intravenous injection of at least five lethal doses of the antigen to which they were previously sensitized. This apparent resistance appears about 12 to 14 hours after the administration of toxin and just about the time when the outward symptoms of intoxication begin to manifest themselves. With the view of eliciting the mechanism of this phenomenon we have made thus far the following observations: The antitryptic titer of the blood of guinea pigs injected with the toxin does not appreciably deviate from normal up to the time of death. The mechanism regulating the antitryptic titer of the blood remains unimpaired in these animals, however, since an injection of antigen to which they are sensitized is followed by a typical rise of antitrypsin. This rise of antitrypsin, incidentally can be interpreted as indication that the humoral phase of the anaphylactic response of the animals is not abolished by the previous injection of toxin. If the same dose of toxin is overneutralized with antitoxin in vitro before injection, it does not protect the sensitized guinea pigs from immediate death when even a single minimal lethal dose of antigen is introduced intravenously.

Further studies on the nature of botulinus toxin

In trying to duplicate in vitro the conditions as they exist when botulinus toxin is taken by the mouth, we have observed that the acidity equal to that of the stomach contents not only leaves the toxicity of botulinus toxin undiminished, but would actually increase its potency. It has been repeatedly stated in the literature that botulinus toxin resists action of acids, but so far as we know nobody has observed the increase in potency of this toxin resulting from the change in its hydrogen ion concentration. In attempting to establish the extent of this increase in potency we found that under the suitable conditions of the experiment the botulinus toxin which ordinarily kills mice in amounts not smaller than 3 × 10−7 cc. can be increased in potency to such an extent that 3 × 10−21 cc. occasionally and 3 × 10−18 cc. quite regularly kills mice of 18-20 grams in less than 48 hours after the intraperitoneal injection. While the total solids of such a minute dose of toxin amount to only 3 × 10−23 grams (this amount including also the inorganic portion of the medium) the toxic product thus obtained, nevertheless, possesses all the essential characteristics of bacterial toxins: it is thermolabile, it acts only after an incubation period, it reproduces in experimental animals typical symptoms of the botulinus poisoning and it exhibits strict specificity in its neutralization with the homologous antitoxin. Our studies, thus far, were limited to the toxin produced by a single strain of Bacillus botulinus, but the experiments are in progress to determine whether the observation can be extended to toxins produced by other strains of Bacillus botulinus as well as to toxins of other bacteria.

The composite nature of botulinus toxin

As we will show in detail in another paper, the lethal dose of botulinus toxin by the mouth is roughly 1,000 removed from that sufficient to kill by the intraperitoneal route. This relation seems to hold true for all the laboratory animals which we investigated, including birds, and is responsible for the failure of certain investigators to kill birds by feeding even large quantities of weak toxin. In attempting to purify the toxin by precipitation, we were surprised to find that, whereas the purified toxin retained its full potency when tested by injection, it became 100 times less toxic by mouth. In general the further the purification was carried, the greater was the loss in potency of purified toxin when tested by mouth. We have been able to reëstablish the toxicity (by mouth) of such purified toxin by merely adding to it the substances removed by the process of purification. Since the potency of our purified botulinus toxin as tested by injection remains the same, whereas the toxicity by mouth varies according to the degree of purification of the toxin, it seems to us that the power of crude botulinus toxin to be absorbed through the intestine is dependent upon the presence of secondary substances mixed with the true botulinus toxin.

The state of aggregation of particles of botulinus toxin

The fact that one cubic centimeter of botulinus toxin with only 10 per cent. of solids contains at least as many as 1018 units of toxin when its hydrogen-ion concentration is adjusted to about PH = 4, 1 suggests that the molecules of this toxin must be very small and of a comparatively simple structure. On the other hand we have shown 2 that saturation of the toxic filtrate of the culture of B. botulinus with (NH4)2SO4 precipitates the toxin, thus suggesting either that toxin is adsorbed by the coarser particles of the precipitate, or that, contrary to the above assumption, the matrix of the toxin is more complex and is capable of being salted out by the (NH4)2SO4. In order to determine which is the case a number of experiments were undertaken. While they did not yield as yet any definite answer to the question of the size or the nature of the molecule of botulinus toxin, they disclosed certain interesting, though not as yet correlated facts concerning the properties of this extremely active substance. We feel that these facts are sufficiently interesting to justify our reporting them at this time. I. Crude toxin kills mice in the dose of 3 × IO-7 C.C. If its reaction is adjusted to about PH = 4 (at the temp. of 37° C.) its potency is increased so that the M.L.D. = 3 × 10-18 c.c. 1 If one adds to such an acidified toxin a solution of pepsin (at 37° C.), already 5 minutes after the addition of pepsin the potency of toxin is markedly reduced and after four hours of contact with pepsin the M.L.D. of toxin is reduced to that of the original filtrate (M.L.D. = 3 × 10-7 c.c.), but no further reduction of toxicity could be observed during further incubation for 48 hours.

The precipitation of botulinus toxin with alcohol

While attempting to purify botulinus toxin by precipitation with ethyl alcohol, we found that the alcohol causes destruction of this toxin. Even a weak solution (20-30 per cent.) of ethyl alcohol is capable of quickly destroying many thousand lethal doses of botulinus toxin in vitro. This destruction takes only five to ten minutes if the mixture is kept at 37° C. On the other hand the toxins of tetanus and diphtheria are much more resistant to the destructive effects of alcohol. It had been observed in several outbreaks of botulism that those who had partaken freely of alcoholic beverages while eating the incriminated food were not severely affected or remained entirely well, whereas others eating as much or even less of the same food were severely and sometimes fatally poisoned. We assumed that this protection from botulinus poisoning when alcoholic beverages are partaken might be due to the direct destructive action of the ethyl alcohol on the toxin in the stomach. The problem was approached experimentally. Two series of three guinea pigs each of equal weight received each per os enough botulinus toxin to kill them within 24 to 48 hours. Each guinea pig of one of the series received 6 c.c. of 30 per cent. alcohol per os immediately following the toxin. The guinea pigs of the first series died within the usual time, whereas the animals which received the alcohol survived and are apparently normal two weeks after the experiment. The effect of alcohol must be ascribed solely to its direct destructive action upon botulinus toxin. This action is quite unlike etherization 1 which delays the rate of absorption of botulinus toxin. That alcohol destroys botulinus toxin only by direct action is shown by the following experiments: Guinea pigs receiving botulinus toxin by the mouth die within the usual incubation period observed in normal controls despite the administration of large amounts of alcohol subcutaneously.

Concerning toxic byproducts of bacillus botulinus

Attempting to shorten the incubation period of botulinus toxin by injecting gradually increasing amounts of toxin into mice, one of us (Orr) has observed early that when the amount of culture filtrate injected reach 0.5 c.c. animals frequently died within a few minutes after injection. Further study of the nature of this intoxication has brought out that when B. botulinus is grown on suitable medium there are produced, in addition to the specific toxin, other poisonous products. If such a culture is filtered through a Berkfeld candle and the filtrate treated with alcohol these secondary toxic products remain in solution. The alcoholic extract equivalent to 0.5 c.c. of the original filtrate is fatal when injected intraperitoneally into mice of 17-22 grams. The first symptoms are noted immediately after injection and consist of restlessness and marked contraction of the abdominal muscles. Within a minute or two the animal shows increased response to external stimuli especially to sharp sound. Shortly the animal becomes prostrated. The increased excitability persists a few minutes longer, the respiration increases in depth and decreases in frequency, and may become as infrequent as 5 inspirations a minute. The animal goes into coma interrupted by sharp convulsive seizures with contraction of the extensor muscles throughout the body. Death occurs in 5 to 15 minutes after the injection and during one of these convulsive seizures. If less than a lethal dose is injected the animal may exhibit all the symptoms described above but will recover completely in 2 to 4 hours. The toxic substances responsible for these acute symptoms are quite distinct from the specific botulinus toxin and are not neutralized by the specific antitoxin. Besides we obtained similar products from the cultures of “atoxic” strains of B. botulinus, as well as from those of B. sporogenes, B. tetani and B. proteus when these organisms were grown on medium composed of minced meat broth. 1

Generalized infection in syphilitic rabbits resulting from the inadequate salvarsan therapy

In the course of study of the spirocheticidal action of salvarsan in vitro it was observed that in very low concentrations, instead of exerting inhibiting action, salvarsan markedly stimulated the growth of spirochetes. 1 Later, on the bases of this observation one of us recommended the use of minute amounts of salvarsan in the medium for isolation of spirochetes in vitro. 2 At the same time it was suspected that introducing minute amounts of salvarsan might also stimulate the growth of spirochetes in vivo. Accordingly a number of male rabbits with experimental syphilitic orchitis as well as two infected females were treated with varying amounts of salvarsan (from 0.004 gm. per kilo down to 0.000004 gm. per kilo) and several of them developed generalized infection, whereas controls treated with large amounts (0.03 gm. per kilo to 0.005 gm. per kilo) or those left untreated showed no tendency to generalized lues during at least 14 months following the date of the experiment. In addition to the involvement of mucous membranes, skin and bones two rabbits developed keratitis. In one rabbit at the autopsy was found a gumma of the liver which was confirmed as such by several pathologists. One of the infected females lost her young twice and the progeny of the third pregnancy was distinctly inferior and all of the four young died within a month after birth. The work reported above was carried out at the Laboratories of the Western Pennsylvania Hospital, Pittsburgh, in 1914-1916. While a brief verbal mention of these findings was made at the time 3 we have not thus far communicated our observations in print desiring first to confirm them on a larger number of animals. In view of the growing tendency to decrease the dosage of salvarsan in the treatment of syphilis, we feel it useful to call attention to the facts observed by us.

Some suggestions for rational auto-serum therapy

Assuming that certain dermatoses as well as intestinal and respiratory disturbances may be due to hypersensitiveness of certain individuals to various proteins, a number of investigators suggested that the onset of acute pathologic phenomena in such cases may be due to the appearance in the circulation of specific protein, causing anaphylactic reaction. On the basis of this assumption various authors have successfully applied the method of immunization in the treatment of such conditions. In cases where the identity of protein in question was obvious from the history of the case, or where the different methods have permitted to disclose its nature by a special investigation, the treatment consists in immunization of patient by repeated injections of the specific protein. In those cases where the nature of protein could not be determined, it was suggested to use the blood of the patient, as a carrier of antigen. If the presence of antigen in the blood of a sensitized individual causes anaphylactic phenomena, it is evident that during the periods of freedom from symptoms the patients blood is probably free from circulating antigen. This fact is lost sight of by many authors who in applying the auto-serum therapy presumably on the basis of the above described theoretical considerations collect the blood at regular intervals without any regard as to whether such blood contains free circulating antigen or not. Much has been written of late on therapy by parenteral introduction of nonspecific proteins 1 and it is possible, if the observations of the clinicians are correct, that the therapeutic effects of promiscuous injections of patients own serum are due to some other phenomena than that of immunization by the circulating antigen. 2 Though in such cases apparently injections of normal horse serum can be expected to give just as good results, human serum is evidently to be preferred in order to avoid an additional sensitization to a horse protein.

Serologic method for detecting infection in foods

The isolation and identification of the infecting organism in contaminated foods is usually beset with difficulties and successful results are not the usual outcome. Negative findings often are due to the fact that it is impossible to examine the whole sample of the incriminated food bacteriologically. Since the pathogenic bacteria are usually few in number, and are not distributed throughout the food, it is more or less a matter of luck if one succeeds in isolating an organism which might justify the suspicions. If one attempts to increase the number of the specific bacteria by enrichment through incubation, he at the same time increases the number of saphrophytes, and thus adds to his difficulties. We find that the entire sample of suspected food can be advantageously analyzed for the presence of any suspected organism, or their split products (in addition to an attempt to isolate individual bacteria), by the following procedure. The whole of the sample of food is chopped up and an extract made from it. This extract is concentrated so that all the specific bacterial protein is collected in a very small volume of liquid. 1 This concentrated solution is then tested against a set of specific immune sera. We have been able to detect by this method the presence of B. botulinus protein in 20 gram samples of artificially inoculated food, where the concentration of toxin was so small that it would have required giving at least 7 grams by mouth or 1.3 grams by injection into a mouse of 15-20 grams to obtain a positive result. This method enables one to determine the presence of a suspected organism in contaminated food within 24 hours after receiving the specimen.