J. Bronfenbrenner

True Polyvalence of Pure Bacteriophages.

As found in their natural environment, bacteriophages often exhibit lytic activity against a variety of organisms sometimes quite unrelated to each other. This polyvalence is usually due to the presence of several independent phages which can be separated from each other by special procedures. Frequently, however, even such isolated single phages, satisfying all known criteria of purity, may exhibit activity against several more or less closely related species of bacteria. In such cases the characteristics of these bacteriophages usually remain immutable no matter which one of the organisms susceptible to their activity is used for their propagation. 1 Since the active principle in such instances is regenerated equally well on either one of the several susceptible organisms, it is perhaps possible to suppose that the active principle is represented by a very similar if not an identical substance in each case, and that the bacteria in turn have a common element in their composition (“receptor”) rendering them susceptible to it. If this could be proven, the nature of this true polyvalence of pure phages would be satisfactorily explained. We have recently made some observations which bring supporting evidence for such a view. One of the coli-phages in our collection, used in the various experiments in our laboratory for nearly 6 years and invariably found to be pure according to all accepted criteria, exhibits the same degree of lytic activity against our type strains of B. coli, B. dysenteriae “Shiga”, and B. dysenteriae “Flexner”, irrespective of whether it is propagated with one or another of these organisms. When this phage is subjected to heating, all 3 valencies disappear at the same time. However, if the speed of inactivation by heat is delayed by the addition of glycerin or of saccharose, 2 the Shiga and Flexner valencies disappear slightly in advance of that against B. coli (see protocol).

Uniformity of size of bacteriophage particles.

Summary Neither fractional ultrafiltration nor fractional centrifugation have revealed any heterogeneity of size among the particles of the bacteriophage T2K (formerly called PC). The methods used for the measurement of size, namely, diffusion through agar and sedimentation in an angle centrifuge, were essentially qualitative. but the experimental arrangements were such as to yield unequivocal results. It must be concluded therefore that this bacteriophage has a minimal particle diameter of 60 mμ or more, as has been found previously for unfractionated phages of the same group by a variety of methods. It is necessary also to withdraw the conclusion previously reached that this bacteriophage has a low infectivity for susceptihle bacteria. An explanation is suggested for some anomalous results previously obtained in attempts to measure the diffusion of bacteriophage through the porous glass boundary. The use of bacteriophage to measure the rate of flow of fluid as a means of control in the application of this method to other materials is described.

Evaluation of Germicides by a Manometric Method.

The need for a method capable of evaluating germicides for chemotherapeutic use is universally recognized. Obviously, the first requirement is that the method should simulate as nearly as possible the conditions met in the animal body. In addition it should be relatively simple and should yield information regarding the action of the test substance on bacteria, and on animal tissues, in terms which would permit a direct comparison of the respective effects. The method of Lambert, 1 employed more recently by Salle, 2 does not seem to fulfill these requirements, since the parasitotropic and organotropic properties of the disinfectant are tested under profoundly different conditions. Moreover, it is based on a tissue-culture technic which is time consuming and laborious. The manometric method proposed here has proved simple, rapid, and remarkably flexible. It permits the conditions of testing to be modified at will, and at the same time allows wide latitude in the choice of animal and bacterial test-subjects. Our experience with it has been limited so far to comparison of the effect of certain disinfectants on the adult mouses liver, and on suspensions of Bacterium coli to which comparable amounts of protein in the form of horse serum were added. The depressant effect on the rate of oxygen-consumption in glucose-succinate buffer was taken as a measure of destructive action in either case.

On Factors Limiting Bacterial Growth. I.

There is as yet no agreement among bacteriologists as to the specific physical and chemical influences responsible for changes observed in the rate of growth during the life of the bacterial culture. 1 In Fig. 1 typical curves are presented, illustrating these changes as they occurred during the development of a broth culture of Bacterium coli. The plotted values were calculated by us from the data of Martin, 2 by means of the formula B-A/T for the rate of population-increase, and log B-log A/T log 2 for the rate of growth, where A represents the population at the beginning and B at the end of the observed time-interval T. It will be seen that during the period of rapid increase in population, the rate of growth shows a sharp rise, followed by a continuous decrease. It has been the purpose of our experiments to reinvestigate the causes for these changes. The possibility had presented itself that measurements of rate of oxygen-consumption per cell might be used as a convenient index of comparative growth-rates. The observations of Martin on the oxygen-uptake of Bad. coli cultures seemed to support this inference. In Fig. 1 we have plotted the rates of oxygen-consumption which he observed, using a single time-scale for these and for the corresponding growth-rates. A comparison of these curves shows that a close relation exists between the rate of growth, and the rate of oxygen-uptake per cell, throughout the duration of the experiment. The present paper deals with the influence of certain factors on the rate of oxygen-consumption per cell. Future publications will present fuller justification for extending conclusions based on these data to the question of actual growth-rates. The methods we have used were essentially the same in all experiments. Bacterial cells to be examined were centrifuged from the culture fluid, resuspended in sterile salt solution, and standardized both by photoelectric estimation of turbidity using solutions of copper sulfate as standard for comparison, and by the dilution and pourplate method to obtain the viable count. The medium used was ordinary 0.5% beef extract—1.0% peptone broth, containing sufficient phosphate buffer of pH 6.6 to give a final concentration of M/20. Whenever the culture-fluids were to be examined, they were sterilized by heat or by filtration, with or without readjustment of pH. The required amount of suspension, buffer, and saline were pipetted onto the floor of the respirometer-vessel, and peptone broth of 5 times the desired concentration was placed in a side arm. When necessary, as in the examination of culture-filtrates, this procedure was reversed, the cells being placed in the side arm and the culture-fluid in the vessel proper. Rolls of starch-free filter paper moistened with 10% KOH served to adsorb the carbondioxide liberated. The vessel was then attached to its differential manometer of the Barcroft type, together with a compensator containing the same amounts of nutrient broth, and the apparatus was allowed to reach equilibrium in a waterbath held at 37 ± .001°C, by means of a thyratronrelay. Manometric readings were begun and continued at 5-minute intervals following the introduction of the substrate from the side-arm. With the corrected values of average oxygen-uptake for the successive 5-minute intervals, curves were laid down, and extrapolated graphically to the vertical axis, where the hypothetical quantity of oxygen consumed per 5 minutes at zero time could be read off. In this way it was possible to obtain consistent values for oxygen-uptake per cell to the exclusion of complicating factors of cellular multiplication and changing composition of medium. It is evident that changes in the rate of growth may be dependent either on the physiology of the cell, on the nature of its environment, or on both. Our first experiments were directed toward separate study of changes in cellular activity, and alterations in the medium of growth in relation to the multiplication-rate observed in broth cultures of Bad. coli. The cultures for examination were prepared by inoculation of 100 ml. of infusion-broth in 500 ml. Erlenmeyer flasks with 0.1 ml of an 18-hour test-tube culture in the same medium and incubated at 37.5°C. without resort to special methods of aeration. The results obtained by the methods outlined above may be summarized very briefly for our present purpose by saying that neither the changes detected in the rate of oxygen-uptake of cells taken from cultures of different age, nor differences in the effect of the corresponding culture-filtrates upon growth and respiration, were sufficiently great to explain the divergence of growth-rates observed at different times in the parent-cultures. Some additional environmental factor apparently was involved.

Respiration of So-Called Filterable Viruses.

Some time ago one of us described a simple micro-respirometer which permits detection of very small amounts of CO2. 1 By the use of this method we were able to detect production of one cubic millimeter of CO2 by 15 million staphylococci in less than ten minutes. However, this is not the limit of the sensitiveness of this method, since the time during which measurable amounts of CO2 may be permitted to accumulate in the respirometer may be extended for days, so that the negative findings obtained by this method are significant. Having failed to detect by this method evidence of respiration in filtrates of so-called bacteriophage, 1 , 2 we thought it of interest to inquire into the question of respiration (CO2 production) by other so-called filterable viruses, particularly those of rabies and herpes. Since these viruses cannot be secured free from tissue cells, it was necessary, at the outset, to differentiate, on the one hand, between the oxygen uptake and the respiration proper with production of CO2, and, on the other hand, between the respiration of tissues as against that of the viruses themselves. The parallel use of Warburgs respirometer to determine oxygen uptake 3 and of our closed cell respirometer 1 gave us the means of fulfilling the first requirement. In order to eliminate the effect of respiration of surviving tissue cells, fresh autopsy material was placed into 50 per cent glycerine and preserved in this condition for months. From time to time samples of these tissues were emulsified and subjected to test. In all, 6 experiments were made with fixed virus and 15 with virus of herpes. Emulsions, made from preserved tissues during the first 8 weeks of ageing in glycerin, showed oxygen uptake and production of CO2 at approximately the same rate as did normal brain tissue preserved for the same length of time and under similar conditions.

Serum Protection Tests with the Lansing Strain of Murine Poliomyelitis Virus

Summary Serum protection tests in mice have been done and the results are significant as tested by a simple statistical method. The experiments are in favor of the view that the Lansing strain is immunologically related to human poliomyelitis. We acknowledge, with thanks, assistance from the following: Mr. Louis Mihelich and Miss Frances Love (technical assistance), and Doctor Martin Bronfenbrenner (advice on statistical method).

Deleterious Effects of Local Application of Staphylococcus Bacteriophage.

During the study of the therapeutic value of specific bacteriophage in experimental staphylococcus skin infections in rabbits it was observed that certain strains of bacteriophage, instead of promoting the healing, have caused a temporary intensification of the local reaction. Suspecting that this effect might have been due to the presence of the “Reynals Spreading Factor” in the bacteriophage lysate, the following experiments were performed. An 18-hour agar slant culture of non-invasive staphylococci was suspended in 10 cc. of sterile distilled water, diluted with an equal volume of broth, and 1 cc. of the resulting suspension was injected intracutaneously into two sites on each flank of each of the 5 rabbits. The hair on both flanks was removed by means of barium sulphide 2 days previous to injection. The sites of the injections in 4 of the rabbits were covered by wet dressings consisting of a piece of folded gauze on top of which was placed a small pad of absorbent cotton which in turn was covered with gauze. A circular piece of adhesive tape with a hole in the center was used to hold the dressing in place. Through the hole in the adhesive tape the dressings were saturated twice daily with (1) broth (Rabbit No. 356) (2) bacteriophage propagated upon the homologous non-invasive strain of staphylococci (Rabbit No. 358), (3) bacteriophage propagated upon a highly invasive strain (Rabbit No. 357) and (4) autolysate of the invasive strain prepared according to Duran-Reynals, 1 (Rabbit No. 359). The fifth rabbit (No. 354) receiving the injection of the culture served as control, and was left without any dressings. The dressings were removed once every day and the size of the lesions measured in each case as follows: The outline of the entire area involved in the reaction at each site of infection was traced with India ink on a strip of cellophane laid over the lesion, and subsequently measured with a planimeter. The area of the individual lesions in each rabbit was remarkably constant.

On the Filtrability of Bacteria.

The statement has been made that filtrates, presumably sterile, prepared from lysed cultures of various bacteria sometimes yield, on standing, a visible growth. The growth so obtained has been interpreted as pointing to a filtrable stage in the life cycle of the bacteria employed. 1 , 2 We have followed for a period of three years lysed cultures of different bacteria, and have watched for the appearances mentioned. On the basis of our studies we have concluded that all such secondary growths are the results of accidents due to imperfection of the filter itself, or to faulty technique employed in carrying out the filtration process. We have recently subjected comparatively large amounts of lysed cultures to fractional filtration through new candles, and noted that while the first several fractions of the filtrate came through wholly sterile, the later fractions often yielded growths. The number of bacteria which may pass the filter in this way is too small to be detected by the ordinary subculture control; but when entire fractions of the filtrates are incubated, the gradual lessening of the efficiency of the filter, which now permits organisms to pass, is shown. Probably the breakdown of the filter is attributable to the coating of the surface with the colloids of the culture medium and distribution of the charge on its particles.

Changes in viscosity during lysis of bacteria by bacteriophage.

The fact has been observed that prior to lysis in the presence of the bacteriophage, bacteria usually undergo more or less marked swelling. The extent of swelling and the relative proportion of swollen bacteria, as well as the actual relation between swelling and lysis are difficult to establish by direct microscopic examination, because the swelling and lysis of bacteria go on simultaneously and continuously, and because the degree of swelling of individual bacteria varies to such an extent that results of such an analysis must, of necessity, be highly subjective. Moreover, in the case of cocci it is very difficult to decide whether or not swelling takes place at all. It seemed, therefore, that since swelling of bacteria increases the relative volume occupied by the solids in the medium, the swelling of the total bacterial mass (with proper correction for growth) should bring about an alteration in viscosity of the solution. Measurements were malde both by means of a capillary viscosimeter of Ostwald and in the torsion viscosimeter of du Noüy. It was found that in general the viscosity of the mixture of bacteria with a corresponding bacteriophage increases steadily up to the time when visible lysis sets in, at which time the viscosity begins to diminish until it gradually reaches the originallevel. In general, the greater the relative concentration of phage, the sooner the maximum viscosity is attained. The greater the number of bacteria (within the limits compatible with the concentration of phage present) the greater is the percentage of change in viscosity. At its maximum, the increase in viscosity of the mixture has varied in the experiments thus far performed, between 14 and 50 per cent, depending on the relative concentration of bacteria and bacteriophage.

Stimulation of Bacterial Metabolism by Bacteriophage.

Examination of an automatic cinematographic record of the progress of lysis of B. coli in the presence of bacteriophage suggested that addition of phage to pure cultures of B. coli causes an increase in the rate of cell division as compared with the rate of division maintained by the same cultures in the absence of bacteriophage. 1 This growth stimulating effect of bacteriophage was again observed in a very striking way, when the phage and bacteria were brought together on solid medium under circumstances which precluded the free lysis of bacteria. Under these conditions the density of bacterial growth was markedly increased under the drops of bacteriophage.2, 3 These and subsequent observations lead us to conclude that the effect exerted by phage is primarily that of acceleration of the metabolic activities of susceptible bacteria (including the increased rate of multiplication) and that actual lysis, while the most striking feature of the phenomenon, is in fact only secondary in importance and occurs only as a result of this change in the metabolic rate and the consequent osmotic changes in the cells taking place if free water happens to be available.4, 5 The increase in the rate of multiplication of bacteria in the presence of phage has been observed by many investigators. However, in most instances the evidence was more or less indirect. More recently there appeared two accounts of experiments in which the authors attempted to count bacteria directly and concluded that the addition of phage does not cause acceleration of growth.6, 7 In view of the manifest importance of this question in connection with the proper understanding of the mechanism of the transmissible lysis we decided to carry out further experiments. Two flasks, each containing 200 cc. of broth (pH 7.4), received a suspension of bacteria so diluted that the initial count in the flask approximated 105 organisms per cc.