equation logr = P - Q.c... (1) (where r is the rate of respiration expressed as a percentage of

Transcription

1 THE ENDOGENOUS RESPIRATION OF BACILLUS CEREUS III. TEE CHANGES IN THE RATE: OF RESPIRATION CAUSED BY SODIUM CHLORIDE, IN RELATION TO HYDROGEN-ION CONCENTRATION M. INGRAM Low Temperature Research Station, Cambridge, England. (University of Cambridge and Department of Scientific and Industrial Research) Received for publication May 20, 1940 I. INTRODUCTION It has been shown that most salts stimulate respiration in low concentrations, and inhibit it in higher concentrations (Ingram, 1939b). In the latter case the rate of absorption of oxygen usually varies with salt concentration according to the equation logr = P - Q.c... (1) (where r is the rate of respiration expressed as a percentage of that in the absence of salt, c is the concentration of salt, and P and Q are constant for any one salt). Anomalous data obtained in certain cases were tentatively attributed to changes of ph during an experiment. The present paper is a more detailed study of the ph relations of the system, using a single salt, sodium chloride; this salt is typical of many, which act in a similar way (Ingram, 1939b). The paper also gives an account of a similar investigation carried out using Escherichia coli. This organism is gram-negative, and has an endogenous respiration which varies greatly with hydrogen-ion concentration. It was hoped that the alterations produced in this respiration by salts might themselves change 683

2 684 M. INGRAM considerably with hydrogen-ion concentration, and so be more easily perceptible than in the case of Bacillus cereus. II. EXPERIMENTAL METHODS The same strain of B. cereus was used in the same way as in previous investigations. Details of the procedure are given elsewhere (Ingram, 1939a and b). A similar procedure was followed with E. coli, which was grown on tryptone agar for 24 hours at The strain of E. coli used was no. 692 of the Type cultures of the Lister Institute. It was found to possess a small endogenous respiration which decayed quickly. Such behaviour is generally characteristic of gram-negative bacteria (Sevag, 1933).1 The fall in the rate of respiration was especially marked during the first hour following the preparation of the suspensions. Thus, the rates of respiration were measured over a later period, 21 to 41 hours after the preparation of the suspensions, and all the rates given below represent averages over this interval. The control and measurement of ph in the present experiments require special mention. The buffer most generally used was a mixture of KH2PO4 and KzHPO4 in 0.2 M concentration, but to obtain extreme reactions 0.1 M potassium phthalate and 0.1 M sodium borate were employed. The cells were washed and suspended in one of these solutions, and the suspension was then mixed with an equal volume of the appropriate concentration of sodium chloride solution. "Analar" reagents were used, and water distilled from Pyrex glass. The ph of each suspension, mixed with the salt solution, was measured at the end of the respiration measurements, that is, after about five hours. The determinations were made with a glass electrode, against standard buffers, and may have a uniform error of =0.02 ph-units, being accurate among themselves to ph-units. Experiments were carried out in duplicate, and each datum represents the average of a pair of determinations. 1 Some feebly gram-positive bacteria behave in the same way, e.g., Streptococcu8 lactis (Callow, 1924) and butyric acid bacilli (Sevag, 1933). In the latter case, loss of endogenous respiration is accompanied by loss of the gram-positive character, owing to oxidative autolysis.

3 EtND(O>GNOUS RESPIRATION OPF BACILLUS CEREUS 685 III. EXPERIMENTAL DATA The changes of ph which occurred in the suspension During the manipulation of suspensions in the way described above, there were factors in operation which tended to make the ph of the suspension differ from that of the buffer solution from which it was prepared. In the first place, the cells themselves exerted a buffering action about a ph approximately that which they impart to an aqueous TABLE 1 The changes of ph which occur when a suspension of bacterial cells is added to buffer solutions at 550C. BUWUR SOLVTION 3a. CURRUS 50 MGM. DRY 3. COI W=RIGT -n30paz GM. IML. DRT WRIGHT Pan UT" ph ofnmiture Cbanp of ph ph of dbxure Chanm of ph Buffer salt or hour hours hours hours hours 1hu hours hour hours 0.05 M phthalate * M phosphate M phosphate M phosphate M phosphate M phosphate M phosphate * M borate * * Solutions of low buffering power. ph of equivalent aque- ph of equivalent aqueous suspension = 6.15 ous suspension = 6.35 suspension; and this resulted in a change of ph, from that of the buffer towards that of an aqueous suspension. This is shown in table 1. The reaction of each buffer solution was moved towards a ph of about 6.2 by cells of B. cereus, and towards a ph of about 6.5 by cells of E. coli. However, the change was small after the first two hours, except in the most alkaline solutions. Table 1 shows also that the shift of ph was greater, the more the reaction of the buffer differed from that of an aqueous bacterial suspension. This was probably due, in the main, to the low buffering power of

4 686 M. INGRAM solutions with extreme reactions; although Shaughnessy and Winslow (1927) have shown that cells of B. cereus and E. coli give off carbon dioxide in response to an alkaline menstrum, and ammonia in response to an acid menstrum, which would enable the cells to bring about relatively large ph changes in solutions of reaction widely different from their own. The addition of sodium chloride introduced a further complication. There was an immediate increase in acidity due to the increased ionic strength of the solution. This change has been discussed 'by Green (1933); it was of serious magnitude with solutions containing more than 0.2 M sodium chloride. Moreover, the presence of the sodium chloride altered the buffering power of cells of B. cereus (Shaughnessy and Winslow, 1927), but the changes arising from this cause were small enough to be neglected. The complexity of the system made it essential to rely on actual observation of the ph of each suspension. The observation was made at the end of the period of observed respiration. This gave a just estimate of the average ph of the suspension over the period during which the respiration was measured because, as has been shown above, the change in reaction following the mixture of the salt with the buffer, and the smaller changes brought about by the buffering action of the cells themselves, were practically completed before measurements of the respiration were begun. The relation between ph and respiration In unbuffered suspensions, low concentrations of sodium chloride increase respiration while high concentrations diminish it (Ingram, 1939b). Similar changes were found to occur in buffered suspensions, over the whole of the experimental range of hydrogen-ion concentration. The concentrations at which the salt was stimulatory were always less than 0.2 M. This concentration of salt never produced a ph shift greater than 0.2 unit, as is shown by a representative series of determinations in table 2. Thus, this table gives the remainder of the data pertaining to the stimulation of respiration, without regard to the shift of ph brought about by the

5 ENDOGENOUS RESPIRATION OF. BACILLUS CEREUS salt. Examination of the data for E. coli reveals the following two points: 1. The salt concentration which gave the maximum rate of respiration was least at the ph of optimum respiration. Thus 687 TABLE 2 The rates of respiration by cells of E. coli and B. cereus suspended in dilute solutions of sodium chloride buffered at different phs E. coli: Molar concentration of sodium chloride ph of suspension Molar concentration of so- Rate of respiration expressed as: dium chloride (a) Qo0 (b) a percentage of that in the absence of salt ph OF SUSPENSION IN BAUITB SOLUTION QE2 Percent Qo, Per cent Qo2 Per cent Per cent E. coli Qot Per cent Q02 Per cent Qo2 Per cent Qo2 Per cent B. cereus Qo2 = cu. mm. 02 per mgm. dry weight per hour. at ph 7.5 the rate of respiration was greatest with M sodium chloride, whereas at ph 5.6 or 9 the maximum respiration occurred in a solution with 0.1 M sodium chloride. 2. The proportional increase in respiration brought about by

6 688 M. INGRAM addition of sodium chloride was also least at the ph of optimum respiration. Here, the highest rate of respiration was 10 per cent greater than that in the absence of salt. At ph 5.6 the rate of respiration in buffer alone was about one-fifth of that at ph 7.5, but by the addition of 0.1 M sodium chloride the rate was increased some 80 per cent; conditions were similar at ph 9. The stimulation of respiration by salt was thus much greater at ph FIG. 1. TE RATE OF RESPIRATION BY CULLS OF B. CEREUS AND E. COLI, SUS- PENDED IN BUFFERED SOLUTIONS OF SODIUM CHLORIDE The increase in acidity, which results from the addition of salt to a buffered suspension, is clearly shown. extreme ph than at the optimum; but it was inadequate to counteract the depressant effect of ph, so that the absolute values of the rate of respiration were never as great as that at the optimum ph of 7.5. B. cereus behaves in the same way, but to a lesser degree. This is probably the result of the relative indifference of B. cereuo to changes of hydrogen-ion concentration within the range investigated.

7 ENDOGENOUS RESPIRATION OF BACILLUS CEREUS 689 With concentrations of sodium chloride greater than 0.2 M, the shift of ph had to be taken into account, and its magnitude may be seen from figure 1, where the ph of each suspension is plotted against the rate of uptake of oxygen, for suspensions with various concentrations of sodium chloride. It is clear that the addition of sodium chloride did not change the ph of optimum respiration to any significant degree. By the erection of ordinates at any ph on figure 1, one may calculate by interpolation the rates of respiration in solutions Molar Concentration Of Sodium Dilorice FIG. 2. THE RATES OF RESPIRATION BY CELLS OF B. CEREUS AND E. COLI, WHICH WOULD BE OBSERVED IN SALT SOLUTIONS BUFFERED AT CONSTANT ph, CALCULATED FROM FIG. 1 with differing salt concentrations but constant ph. The results may then be used to test the validity of the relation logr = P-Qx... (1) r = rate of respiration c = salt concentration P and Q constant which has only been shown to hold in unbuffered suspensions (Ingram, 1939b). The results of this calculation are set out in figure 2. The linear relation between log r and c is apparent,

8 690 M. INGRAM and establishes equation (1) for both B. cereus and B. coli. This is true at any ph and moreover, the slope of the line relating log r to c, and hence the value of Q in this equation, is the same whatever the ph. (The logarithmic ordinate in figure 2 gives undue prominence to errors in the determination of the lower rates of respiration.) Q is about 0.75 for B. cereus and 0.60 for E. coli. IV. DISCUSSION One might be considerably misled, if the relation between respiration in buffers and concentration of added salt were expressed without regard to ph. For example, if suspensions of B. cereus in phosphate buffer of ph 5.5 mixed with different concentrations of sodium chloride are treated as a homogeneous series of constant ph, a curve such as that shown in figure 3 is obtained instead of the linear relation of figure 2. This happens because the rate of respiration is depressed not only by the salt, but also by the increased acidity which it causes; conversely, starting with a suspension on the alkaline side of the optimum ph, the rates of respiration in the more concentrated salt solutions appear high, because the shift of ph is in the direction of greater respiration in this case. Thus one might imagine that addition of salt caused a shift of the optimum ph in an alkaline direction, and that salt shows its greatest potency as an inhibitor of bacterial activity in acid solutions. However, with an organism such as B. cereus which has a broad respiration/ph optimum, an intermediate state is possible, in which the change of ph brought about by the addition of a salt is not great enough to alter the rate of respiration appreciably, and under such circumstances the relation expressed by equation (1) is not obscured if the ph is neglected (e.g., the curve for ph 7.5 on figure 3). The curve relating to suspensions in buffer of ph 5.5, shown on figure 3, is exactly similar to that presented by Nicolai (1926) in his study of the respiration of E. coli in phosphate buffers, and to data obtained with buffered suspensions of B. cereuwsby the present writer (Ingram, 1939b). It is clear that these earlier data have no significance because of the neglect of the latent changes of ph.

9 ENDOGENOUS RESPIRATION OF BACILLUS CEREUS It was suggested (Ingram, 1939b) that the stimulating action of dilute salt solutions on the respiration of B. cereus was probably greater in buffered solutions of ph about 6.0 than in unbuffered solutions of ph about The data of table 2, in which a greater range of hydrogen-ion concentration is included, confirm this observation. The increased stimulatory effect of dilute salt solutions with increasing acidity was easily perceptible at V 0Op2 7s5 CherENTRATa:Ion of with toe Onferea spnsion. mdt, after mixture FIG. 3. THE RATES OF RESPIATION ACTUALLY OBSERVED WHEN INCREASING CHLORIDE ARE ADDED TO A SUSPENSION CONC3ENTRATIONS OF 90DIU OF B. ceneus, IN Bun= SOLUTION OF INITAL ph 5.5, 7.5 OR 9.2 ph 5.6 with E. coli, as the divergence from the ph optimum was considerable. In addition, a similar increase was apparent at ph 9. Thus the stimulation of respiration by sodium chloride, and the concentration of the salt necessary to produce it, were both least at the ph of optimum respiration and both increased on either side of this ph. This is borne out by the behaviour of the curves on figure 2, at low concentrations of sodium chloride. Moreover, in this figure the whole of each

10 692 6M. INGRAM curve falls lower the greater the divergence from the optimum ph, whether this divergence is on the acid or the alkaline side. In fact the effects of salt, both stimulatory and inhibitory, were symmetrical about the ph of optimum respiration. The value of Q given above for B. cereus (0.75) is appreciably greater than that reported earlier (0.68-Ingram, 1938b). This organism was originally isolated from a medium containing roughly 1.0 M sodium chloride, and has since been cultured in one with only 0.2 M sodium chloride. The decline in resistance TABLE 3 A comparison of the effects of salt and ph on the growth and respiration of E. coli. The rates of growth and of respiration are calculated as percentages of those in the absence of salt ph Growth:* Peptone Peptone M sodium chloride Peptone M sodium citrate Respiration:t 0.1 M phosphate M phosphate M sodium chloride M phosphate M sodium chloride * Calculated from the data of Sherman and Holm, The ph of optimum growth was 6.8. t The ph of optimum respiration is about 7.5. to salt during the nine months elapsing between the two measurements of Q suggests that the halotolerant character may be "adaptive." It is interesting to compare, at various hydrogen-ion concentrations, the data expressing the action of sodium chloride on the respiration of E. coli with those of Sherman and Holm (1922) for the action of this salt on growth by E. coli. Sherman and Holin arrayed their data against actual measurements (by indicators) of the ph of the medium; the data of table 3 were obtained by interpolation. Table 3 shows that the stimulation of growth by sodium chloride was least at the optimum

11 ENDOGENOUS RESPIRATION OF BACILLUS CEREUS ph, just as is the case for respiration; so that low concentrations of salt have a similar broadening effect on the ph-optima of growth and respiration. The inhibition of growth caused by 0.2 M sodium citrate can be appreciated by the use of an analogy. It has been shown that, in its influence on the rate of respiration by B. cereus, 0.2 M sodium citrate corresponds to 0.6 M sodium chloride (Ingram, 1939b). Thus, one may compare the action of 0.2 M citrate on the growth of E. coli with that of 0.6 M chloride on respiration. When this is done there is a fair degree of correspondence between the two sets of data (table 3). This suggests two things: firstly, that it is legitimate to suppose that sodium citrate is roughly equivalent to three times its concentration of sodium chloride; secondly, that salts act in a similar manner on growth and on respiration. The work described above was carried out as part of the programme of the Food Investigation Board and is published by permission of the Department of Scientific and Industrial Research. V. SUMMARY Manometric measurements have been made of the rate of absorption of oxygen by cells of Bacillus cereu= and Escherichia coli, suspended in buffer solutions containing sodium chloride. The ph of each suspension was measured immediately after the rate of respiration had been determined. The following conclusions may be drawn from the data: (a) The ph of optimum respiration, about 7.3 for E. coli and 6.3 for B. cereus, is not altered when sodium chloride is added. (b) If small concentrations of the salt are added the rate of respiration is increased; larger concentrations diminish it. This is 693 true whatever the ph. (c) The effect of the salt in increasing respiration is greater the more the ph differs from that of optimum respiration, be it more acid or more alkaline. (d) The concentration of sodium chloride needed to bring about the maximum stimulation of respiration is also least at the ph of optimum respiration.

12 694A M. INGRAM (e) The inhibition of the respiation of either orgasm is expressed by an equation. log r = P - Q.c The equation relates rate of respiration and concentration of sodium chloride whatever the ph. (f) The values of Q in this equation are independent of ph, but differ for the two species. REFERENCES CAL1,OW, A. B The oxygen uptake of bacteria. Biochem. Z., 18, GREEN, A. A The preparation of acetate and phosphate buffers of known ph and ionic strength. J. Am. Chem. Soc., 68, INGRAM, M. 1939a The endogenous respiration of Bacillus cereua. I. Changes in the rate of respiration with the passage of time. J. Bact., 38, INGRAM, M. 1939b II. The effect of salts on the rate of absorption of oxygen. J. Bact., 38, NicoLAI, H. W t.ber den Fermentstoffwechsel der Bakteriei III. Biochem. Z., 179, SEVAG, M. G tvber die Beziehungen zwischen enzymatischen AktivitAt, Morphologie, und Firbbarkeit von Butters!Lurebakterien und fiber den Mechanismus der Restatmung. Z. Hyg. Infektionskrankh., 114, HAUGHNESSY, H. J., AND WINSLOW, C.-E. A The diffusion products of bacterial cells as influenced by the presence of various electrolytes. J. Bact., 14, SHERMAN, J. M., AND HOLM, G. E The growth of Bact. coli in relation to hydrogen-ion concentration. J. Bact., 7,