STUDIES ON THE RELATIONSHIP OF POTASSIUM TO METABOLISM AND PURINE BIOSYNTHESIS IN ESCHERICHIA COLI1 SAMUEL FRIEDMAN AND CHARLES L. FOX, JR. New York Medical CoUege, Flower and Fifth Avenue Hospitals, New York, New York

Received for publication February 15, 1954

During inhibition of the growth of Escherichia coli by sulfonamides, a diazotizable amine is formed (Fox, 1942). Following its isolation (Stetten and Fox, 1945) and identification as 4-amino5-imidiazolecarboxamide (Shive et al., 1947), its important roles in purine (Buchanan, 1952) and nucleic acid metabolism (Brown et al., 1952) were disclosed. Accordingly the relationships of the chief intracellular cation, potasium to growth, metabolism and the bacterial synthesis of this metabolite were investigated. The importance of potassium in cellular physiology has received an increasing appreciation (McQuarrie, 1953; Sheppard, 1951). In contrast to sodium, potassium functions in numerous metabolic reactions. Its role in phosphorylations, i.e., in fructokinase, phosphofructokinase, and pyruvate phosphopherase activity, as well as stimulation of oxidative

phosphorylation, carbohydrate metabolism, respiration, fatty acid oxidation, and decarboxylation of malate has been reviewed (Lardy, 1951; Lehninger, 1950). Potassium also stimulates metaphosphate (Schmidt et al., 1949), acetoin (Nossal, 1952), and pantothenate formation (Maas, 1952); aldehyde dehydrogenase (Black, 1951) and lactase activity (Cohen and Monod, 1951); acetate activation by coenzyme A (Stadtman, 1952); various oxidations (Quastel and Webley, 1942); and this cation is required for the growth of several microorganisms (MacLeod and Snell, 1947; Rahn, 1936; Falk, 1923) and other forms of life. The present study illustrates some aspects of the role that potassium exerts on the anabolic metabolism of E8cherichia coli. Data are presented indicating that within defined limits the extent of uptake of assimilable carbon, phos'This investigation was supported by a grant E-352 from the National Microbiological Institute, National Institutes of Health, U. S. Public Health Service. A preliminary report of portions -of this data has appeared (Friedman and Fox, 1953).

phorus, and nitrogen substrates and the resultant synthesis of cell mass, as well as purine synthesis by one pathway, are dependent quantitatively on potasium concentration. MATERIALS AND METHODS

Chemicals. The inorganic salts used were either cp or reagent grade. Other substances were obtained commercially and where necessary were converted to the neutral sodium salts. Several were found to be contaminated with significant amounts of K+ and were treated with the cation-exchange resin IR-120(Na+). Media and organism. The growth medium designated SGH contained per liter Na2HPO4, 6 g; NaH2PO4, 3 g; NH4Cl, 1 g; NaCl, 1 g; MgSO.-7H20, 0.1 g; glucose, 4 g and histidine, 80 mg; the latter two being autoclaved and added separately. Such media on analysis contained 125 mEq Na+, 18.7 mEq NH+, 0.08 mEq Mg++, and less than 0.03 mEq K+. All additions were prepared in medium SGH. In this medium histidine enhances the growth response of E. coli, strain B, and the K+ effects reported are independent of its presence. E. coli, strain B, and the auxotrophs, B96 and M48A-33, were used. B96 is a purine auxotroph from E. coli, strain B, M48A-33 is a p-aminobenzoic acid (PABA) auxotroph from E. coli, strain W; both accumulate 4-amino-5-imidazolecarboxamide (AIC) (Gots, 1950; Gots and Chu, 1952). The organisms were maintained on agar slants of similar media containing 5.1 mEq K+ per L as KCl, supplemented with 5 jig per ml of xanthine for growth of B96 and 5 mpAg per ml of p-aminobenzoic acid for M48A-33, and stored at 5 C. To prepare inocula for experiments, about 105 organisms per ml taken from week old slants were grown 17 to 19 hours in media containing 0.13 ,Eq K+ per ml. These were harvested and washed twice by centrifugation with either 0.05 M tris-(hydroxy-methyl) aminomethane buffer, pH 7.4, or distilled water and adjusted to suitable concentration. Unless 186

1954]

187

RELATIONSHIP OF POTASSIUM TO METABOLISM IN E. COLI

otherwise indicated, an inoculum of about

(Is 160~I60-

4 X 104 organisms per 5 ml was used in the

actual experiments. Methods. Studies were done in acid cleaned test tubes, suitably rinsed and covered with aluminum caps. Turbidity was read on a KlettSummerson photoelectric colorimeter with a 54 filter. Reducing sugar was determined by the method of Somogyi (1952) and Nelson (1944), 4-amino-5-imidazolecarboxamide as diazotizable amine (Brattan and Marshall, 1939) following acetylation with acetic anhydride at pH 7 for 30 minutes (Stetten and Fox, 1945) using a synthetic hydrochloride salt as standard, phosphorus by the methods of Fiske-SubbaRow (1925) or of Lowry and Lopez (1946), adenosine triphosphate pyrophosphate as 7 minute hydrolyzable phosphate (LePage and Umbreit, 1945), and ammonia by a micro-nesslerization technique (Johnson, 1941). Pentose and desoxypentose nucleic acids were determined by the methods of Drury (1948) and Stumpf (1947) following hydrolysis at 100 C for 30 minutes in 5 per cent trichloracetic acid, using commercially obtained nucleic acids with known phosphorus contents as standards. For analysis of intracellular potassium, 19 to 22 hour old cells were harvested by centrifugation and washed twice with distilled water which was retained for analysis of the K+ released during this manipulation. The washed cells were dried to constant weight in vacuo over P205 and defatted according to Lowry and Hastings (1942). The defatted cells were extracted with 0.75 N nitric acid (Lowry and Hastings, 1942) using approximately 1 ml of acid per 10 mg defatted cells. The wash water and ether extract used in defatting were evaporated, and each residue extracted with acid. The K+ concentrations of all the extracts were determined by an internal standard flame photometer (Fox, 1951). RESULTS

(1) Potassium requirement in medium SGH. In view of the metabolic importance of K+, the optimal concentrations required for cellular synthesis, expressed as growth in medium SGH, were determined and found to range from 2.6 to 40 ;iEq per ml. Half maximal growth occurred with less than 0.13 ,uEq per ml (figure la). This stimulation was exerted early in the growth

120. i00.

° 80.

I.-

60.

6hrs 40.

20A I

0 4

I

I

I

I,

8 12 16 20

I

_

40

uEq K*/ml Figure la. The stimulation by K+ of growth of Escherichia coli in medium SGH. Turbidity plotted as optical density of individual cultures containing increasing concentrations of K+ in medium SGH. (Trace contamination of 0.03 pEq/ ml K+ found in the medium, plus intracellular K+ contributed by the inoculum, could be responsible for growth in the absence of added cation.)

cycle as evidenced by the increased number of cells in exponential growth at 6 hours; neither anaerobiosis nor vigorous aeration altered this effect. Similar type curves were obtained with sulfate or acetate as the anion except that there was less growth at 24 hours, and the inhibitory effects with higher concentrations were more marked. At 10.2 pEq K+ per ml which is optimal, the equivalent ratios of cations to K+ in medium SGH are Na+ 12.2, NH+ 1.8, and Mg++ 0.08. Apparently K+ deficient cells are not inhibited strongly by Na+ since the Na+:K+ ratio can be increased to 16.9 (ionic strength 0.23) with no inhibition of growth (figure lb). In contrast increases in NH+ to 5.6 (ionic strength 0.235) or Mg++ to 8.2 (ionic strength 0.23) caused marked inhibition of growth. Increasing concentratioins of K+ up to 250 ,uEq per ml did not reverse

188

SAMUEL FRIEDMAN AND CHARLES L. FOX, JR.

20a

30

0~~~~~~

40_ I30~-

10o

30 40 0 CATION: K* (#u Eqs /ml1)

50

Figure lb. Antagonism by other cationis in medium SGH of the growth stimulation by K+. Inhibition is plotted against the ratio of-equivalents of other cations to optimal K+ concentration (10.2 j&Eq per ml). growth inhibitions of 15 or 40 per cent caused to medium SGH 'of up to 300 2 by the addition ;&Eq per ml of Na+, NH4, or Mg++ ions. This inability of increae K+ to reverse inhibitions by these cations at optimal ratios makes unlikely the possibility that the K+ effect is primarily due to a competitive ion altago i r. -(2) Poltasium partitfion in E. coli. IntraceRulr K+ in E. coli is partitioned between a freely diffusible fraction and a retained or "4bound" fraction (Cowie et al., 1949) so that accurate measurement of the internal concentration of this cation is difficult. For example, 19 hour celLs grown on an optimal concentration of 10.2 ;&Eq K+ per ml contained an estimated total of 0.378 ,uEq K+ per mg fat-free dry weight (1.47 per cent) when cell equiivalent to 61.2 mg were analyzed.2 Under the conditions of these experi2 A primary grown dried beer yeast powder intended for pharmaceutical purposes, Fleischmann Laboratories, Type 200B, Saccharomyces, was al80 analyzed. The value found for several portions was 0.582 IAEq potassium per mg (2.3 per cent).-

[VOL. 68

\ ts, 35 per cent (0.135 ,uEq) was retained in the acterial cells and 65 per cent (0.242 gEq) was diffusible and recovered in the wash water (0.234 uEq) and the ether extract (0.008 ;&Eq). The diffusible fraction was extractable also at 5 C. No difference in K+ concentration could be found in the retained fraction of 19 to 22 hour cells grown on median and optimal concentrations of cation (table 1, column a). (3) Potassium as a metabolic regulator. The available supply of K+ regulates the uptakes of glucose, amonia, and inorganic phosphate and the resultant synthesis of cell mass in medium SGH. This effect with 19 hour dense cell suspensions is shown in figure 2. Total uptake was greatest with 10.2 ,uEq K+ per ml, but maxmal rate of activity occurred below 3 ;sEq. In contrast to the uptakes of glucose and phosphate, less ammonium ion disappeared at each concentration of K+. This relationship was similar when cell suspensions three times as dense (2.8 mg dry weight) were tested between 0 and 102 ;uEq K+. Total uptake was maximal at 40 uEq after which it levelled off. Turbidity increased only 3 per cent, mainly between one and 15 ,uEq K+, although at least 25 per cent of the initial concentrations of glucose, ammonia, and inorganic phosphate remained. In an attempt to replace the K+ requirement for growth, various metabolites or cofactors were tested in SGH media to which was added 0.002 ,Eq K+ per ml, an amount insufficient for growth. The following were unable to replace the need for this cation when added singly, or in a wide variety of combinations, and through a range of concentrations of which the highest tried per 5 ml mediumSGH is indicated: 2 um adenine, 2.2 um adenosine, 6 ,M adenosine triphosphate (ATP), 3 AM diphosphopyridine nu¢leotide (DPN), 400 jg desoxypentose nucleic acid hydrolyzate, 200 pg pentose nucleic acid hydrolyzate, 5 mg of casein hydrolyzate, 2.1 JAM 1+ glutamic acid, 3.9 sim glutamine, a multivitamin preparation,3 10 pAg p-aminobenzoic acid, 30 pg p-hydroxybenzoic acid, 30 pug biotin, 30 pg folic acid, 1.3 pg vitamin B12, 200 JAM citrate, 10 jAM isocitrte, 200 JAm pyruvate, 200 ' A 1:50 dilution of "pancebrin" (Lilly) injectable containing 40 pg thiamin, 88 pg riboflavin, 12 pg pyridoxine, 12 pg pantothenate, 80 pg nicotinamide, 240 pg ascorbic acid, 8 pug a-tocopherol, 4 units of vitamin A, and 4 units of vitamin D was used.

19541

189

RELATIONSHIP OF POTASSIUM TO METABOLISM IN E. COLI TABLE 1 Intracellular contents of cells grown in median and optimal concentrations of K+ K+ ,PEQ RETAIND (a)

pG 7 YIN p P ~~HYDROLYZIAL (b)

(a)

jq/mI

(c)

G DSOXYPENTOS IU NUCLEIC ACID

(d)

0.1

2.6

0.1

2.6

0.1

2.6

4.5 4.0

5.0 4.4 5.2

69.7 69.3 86.8

80.5 89.4 91.7

31.8 33.8 36.4

32.5 31.6 34.9

0.130 0.132 4.4 4.9 75.3 87.2 34.0 .0.030 4;0.043 +0.36 40.36 -49.9 d45.9 4d2.3

33.0 4-1.7

0.1

2.6

AG PENTOSE NUCLEIC ACID

Experiment 1 2 3

Mean ........................ SD ..........................

0.113 0.112 0.164

0.112 0.108 0.176

4.7

All analytical values are expressed as per mg dry weight and retained K+ as fat-free dry weight. Turbidity of cells at optimal K+ was 1.6 times greater than median cells, i.e., at 19 hours turbidities were 98 and 62. &m a-ketoglutarate, 200 ;Am fumarate, 200 jum succinate, and 200 jAm aspartate. All combinations were tested in the presence of 3 juM adenosine triphosphate and 1 JAm diphosphopyridine. In view of K+ involvement in synthesis and transfer of high energy phosphate (Lardy, 1951) and the alleged impermeability of cells to extracellular adenosine triphosphate, the adenine nucleotide pyrophosphate contents of dried cells grown on median and optimal concentrations of K+ were determined. Small but comparatively greater amounts of nucleotide pyrophosphate were found in celLs grown on optimal K+ concentrations (table 1, column b). (4) Potassium and purine biosJnthe8si. The relationship of K+ to the synthesis of intracellular components was studied in a representative process: accumulation of bacterial derived 4-amino-5-imidazolecarboxamide (Fox 1942; Stetten and Fox, 1945; Shive et al., 1947) as an index of the purine synthesizing process by two auxotrophs. The K+ requirement for their growth was similar to that for the wild type when tested at concentrations of auxotrophic substance giving maimum accumulation (table 2). Stimulation by K+ of 4-amino-5-imidazolecarboxamide accumulation, however, and of growth is different. With 0.03 juEq per ml, 50 per cent of total growth but only 3 per cent of total 4-amino-5-imidazolecarboxamide are obtained. With lOX as much K+ (0.3), growth is only 1.6x greater, but 4-amino-5-imidazolecarboxamide accumulation increased about 20 X. Maximal accumulation of 4-amino-5-imidazolecarboxamide by the p-aminobenzoic acid auxotroph was but 4 that of the purine mutant and

only H as much K+ was required (0.3 to 1.3 pEq K+ per ml versus more than 10.2 for B96). A similar dependence on K+ for 4-amino-5imidazolecarboxamide formation also was observed in sulfadiazine inhibited cultures. Evidence that K+ participates in cellular

,pEq K5/mI Figure 2. K+ as a metabolic regulator. Washed cells (0.5 ml equivalent to 0.86 mg dry wt) were added to 4.5 ml of medium SGH containing increasing K+ concentrations. At zero time, the turbidity was 106 and 22.4 pM of glucose, 18.6 pM of NH+ and 63 Am of inorganic phosphate (Pi) were present. The amount of substrate that disappeared and the turbidity increment, at 3 hours, are plotted against the K+ concentration of the medium.

190

[VOL. ff

SAMUEL FRIEDMAN AND CHARLES L. FOX, JR.

TABLE 2 Potassium stimulation of growth and 4-amino-5imidazolecarboxamide (AIC) accumulation by auxotrophs P-MINOBENZOIC ACID AUXTO

aEQ K+

M48A-33 Tur-

bidity

0 0.03 0.3 1.3 2.6 10.2 10 .2-p-aminobenzoic acid or xanthine

18 74 103 118 125 140 0

AIC

0.04 0.06 1.29 1.30 1.04 0.84 trace

1.8.

1596

I00

000 010

00,

.0*

.01

.01

..x

PUXNE

AuoioHB96 AUXOTROPB Tur-

bidity

22 86 124 132 144 144 0

AIC

0.04 0.24 4.2 5.1 5.7 6.2 trace

System: 19 hour culture in 5.0 ml of medium SGH containing 20 pg glutamic acid and glycine per ml (Ravel et al., 1948). M48A-33 supplemented with 5 mpug p-aminobenzoic acid per ml, B96 with 10 pg xanthine per ml. Data pg per ml except turbidity.

syntheses independent of growth may be obtained by utilizing auxotrophs which accumulate precursor but do not grow in the absence of auxotrophic substance.4 In this way the K+ dependence of 4-amino-5-imidazolecarboxamide accumulation in the absence of growth was demonstrated with dense cell suspensions of the purine and p-aminobenzoic acid auxotroph (figure 3). Furthermore, in experiments with metabolic inhibitors which block 4-amino-5imidazolecarboxamide accumulation by B96 in such a system, K+ in high concentrations was able to lessen the extent of inhibition of 4-amino-5imidazolecarboxamide accumulation by azide and fluoride (table 3). Inhibition by 1 X 10-' mi iodoacetic acid, 1 X 10-2 M sodium cyanide, and 4 X 104 M dinitrophenol created blocks in metabolism not reversible by similar concentrations of K+. In keeping with this data, wild type E. coli, strain B, grown in optimal K+ concentrations had relatively higher pentose nucleic acid contents than cells grown in median K+ concentrations; desoxypentose nucleic acid contents were almost equal (table 1, columns c and d). 4Dr. J. S. Gots kindly made available his data with the auxotroph B96 (Gots and Love, 1954). The p-amino-benzoic acid auxotroph was similarly found to accumulate 4-amino-5-imidazolecarboxamide in the absence of growth.

0

4

1.6 2.0

4.0

5.2

,aEq K*/mI Figure 3. The effect of K+ on 4-asnino-5-imidazolecarboxamide accumulation by purine (B96) and p-aminobenzoic acid (M48A-33) mutants in absence of auxotrophic substance. Washed cells (0.5 ml equivalent to 0.28 mg dry weight B96 and 0.78 mg M48A-33) were added to 4.5 ml medium SGH supplemented with 20,pg per ml of glutamic acid and glycine (Ravel et al., 1948) containing increasing concentrations of K+. In absence of auxotrophic substance turbidity remained constant and the histidine in medium SGH did not relieve the purine requirement of B96. Nonacetylatable diazotizable amine per ml determined at 3 hours is plotted against K+ concentration.

TABLE 3

The effect of K+ on inhibitors of 4-amino-5-imidazolecarboxamide (AIC) accumulation by purine auxotroph B96 AUC ACCUNULATED IN INHIBITOR ADDD

SCE 01:

X 10-4

0.0001

0.13

0.64

2.9 2.1 1.0 0.1

3.4 2.3 1.2 0.1

3.3 2.9 1.5 0.2

mEq K' mEq K' mEq K+

None ............... Na fluoride .......... Na azide ............ 2,4-Dinitrophenol ...

-

100 10 4

System: 0.5 ml cells (0.8 mg dry wt), 0.1 ml inhibitor, K+ as KCI, 20 pg/ml glutamic acid and glycine (Ravel et al., 1948), plus SGH media containing no auxotrophic substance to 5 ml volume. At 3.5 hours samples centrifuged 6,500 rpm for 15 minutes, and nonacetylatable diazotizable amine determined. Data gg per ml.

19541

RELATIONSHIP OF POTASSIUM TO METABOLISM IN E. COLI DISCUSSION

The data presented illustrate the importance of K+ in the synthetic reactions of E. coli. This is expressed as a K+ requirement for optimal exponential growth and attainment of maximal cell yield during the growth cycle and for the proper uptake and assimilation of carbon, phosphorus, and nitrogen substrates by the cell. The relationship found in dense suspensions of 19 hour celLs between K+ concentrations and the rates of asimilation of the substrates tested may provide a biochemical explanation for the well known dependence of normal growth of E. coli upon the K+ concentration in the medium (Hotchkiss, 1923; Winslow and Dolloff, 1928; Roberts et al., 1949). The free outward diffusion of K+ which occurs during harvesting and washing of the cells in preparation for analysis may account for the wide range of values of K+ reported in microorganisms. The higher total value of K+ in E. coli, strain B, reported here is based on a more critical accounting for the freely diffusible K+ fraction and supersedes values previously reported by this laboratory (Friedman and Fox, 1953). The retained K+ fraction would appear essential for metabolic competence but insufficient for optimal functioning. As evidence for this, no difference was found in the retained fraction of median and optimal grown cells, indicating that quantitative intracellular incorporation of K+ is not the sole basis for its stimulating effect. Washed dense cell suspensions containing only such a store of K+ function at suboptimal levels; furthermore, the addition of various substrates including K+ mediated reaction products does not relieve this requirement. The addition of small quantities of K+ to such cells, however, causes a marked rise in activity, suggesting a catalytic function and attesting to the importance of the freely diffusible fraction for optimal metabolism. The comparatively increased adenine nucleotide pyrophosphate content of cells grown with optimal K+ and the failure of K+ mediated reaction products to relieve this K+ requirement suggest that this cation may function at several levels: in energy synthesis or its transport, or in coupled reactions between energy and substrate levels leading to assimilation. Involvement of K+ in such functioning would explain adequately the metabolic role of this cation in the processes illustrated in this paper: in assimilation of sub-

191

strates during and in the absence of growth. Some extension is required also of the common concept that K+ transport is dependent on metabolism since optimal metabolism itself is dependent on K+. A requirement for K+ in purine synthesis as measured by bacterial 4-amino-5-imidazolecarboxamide accumulation which is independent of growth and occurs prior to closure of the purine ring also has been demonstrated. This is reflected in the comparatively increased syntheses of more actively metabolized pentose nucleic acid in cells grown in optimal K+ concentrations and in the ability of K+ to relieve partially the inhibition of 4-amino-5-imidazolecarboxamide accumulation by azide or fluoride. Ability of high concentrations of K+ to influence these two inhibitors may further imply a role for K+ in phosphorylation reactions involved in purine synthesis at sites not influenced by

dinitrophenol. ACKNOWLEDGMENT

The authors wish to thank Dr. J. S. Gots of the Department of Microbiology, School of Medicine, University of Pennsylvania, Philadelphia, Pa., for supplying the auxotroph B96; Dr. B. D. Davis of the Tuberculosis Research Laboratory, U. S. Public Health Service, New York, N. Y., for the auxotroph M48A-33; Dr. A. S. Schultz of the Fleishmann Laboratories, Stamford, Connecticut, for the beer yeast 200B; and Dr. C. S. Miller of the Sharpe and Dohme Laboratories, West Point, Pa., for a generous sample of 4-amino-5-imidazolecarboxamide hydrochloride. The flame spectrophotometric determinations were kindly done by the Misses Helene S. Pandelakis and Eloise L. Rhoades of this laboratory. SUMMARY

Some aspects of the importance of potassium

to the anabolic metabolism of Escherichia coli have been examined. Narrow concentration ranges of potassium were required for optimal exponential growth, attainment of maximum

cell yield and efficient uptakes of ammonia, glucose and inorganic phosphate from saltsglucose media by suspensions of wild type E. coli, strain B, as well as for 4-amino-5-imidazolecarboxamide accumulation by a purine and paminobenzoic acid auxotroph. Analysis of 19 hour cells showed that a greater portion of

192

SAMUEL FRIMAN AND C

intracellular potassium is freely diffusible and a lesser fraction retained. No quantitative difference could be found in the retained fraction of potamsium in median and optimal grown cells. Optimal grown cells, however, had comparatively increased pentose nucleic acid and adenine nucleotide pyrophosphate contents. A variety of metabolites or cofactors including potasium mediated reaction products was unable to replace the potassium requirement of low potassium cells. The general metabolic significance of these findings is discu. REFERENCES BLAcK, S. 1961 Yeast aldehyde dehydrogenase. Arch. Biochem. and Biophys., 34, 86-97. BRATTAN, A. C., AND MARSHALL, E., JR. 1939 A new coupling component for sulfanilamide determination. J. Biol. Chem., 128, 537-550. BROWN, G. B., RoLL, P. M., AND WEINFELD, H. 1952 Biosynthesis of nucleic acids, pp. 385406. In Phosphorus metabolism. Vol. II. Edited by McElroy, W. D., and Glass, B. The Johns Hopkins Press, Baltimore, Md. BUCHANAN, J. M. 1952 Studies on the biosynthesis of purines in vitro,pp. 406-422. In Phosphorus metabolim, idem. COHEN, M., AND MONOD, J. 1951 Purification et propri6t6s de la 0-galactosidase (lactase) d'Escherichia coli. Biochim. et Biophys. Acta, 7, 153-174. COWIE, D. B., ROBERTS, R. B., AND ROBERTS, I. Z. 1949 Potassium metabolism in Escherichia coli. 1. Permeability to sodium and potassium ions. J. Cellular Comp. Physiol., 34, 243-257. DRURY, H. F. 1948 Identification and estimation of pentoses in presence of glucose. Arch. Biochem., 19, 456-466. FALK, I. S. 1923 The role of certain ions in bacterial physiology. A review (Studies on salt action VII). Abstr. Bacteriol., 7, 33-50, 87-105, 133-147. FISKE, C. H., AND SUBBAROW, Y. 1925 The colorimetric determination of phosphorus. J. Biol. Chem., 66, 375-400. Fox, C. L., JR. 1942 Production of a diazotizable substance by E. coli during sulfonamide bacteriostasis. Proc. Soc. Exptl. Biol. and Med., U, 102-104. Fox, C. L., JR. 1951 Stable internal standard flame photometer for potassium and sodium analysis. Anal. Chem., 23, 137-142. FRIEDMAN, S., AND Fox, C. L., JR. 1953 Effects of potassium on purine synthesis and growth

S L. FOX, JR.

[VOL. 68

in Escherichia coli. Federation Proc., 12, 206. GoTs, J. S. 1960 The accumulation of 4-amino5-imidazolecarboxamide by a purine-requiring mutant of Eacherichia coli. Arch. Biochem., 29, 222-223. GOTS, J. S., AND CHU, E. C. 1952 Studies on purine metabolism in bacteria. 1. The role of p-aminobenzoic acid. J. Bact., 64, 537546. GoTs, J. S. AND LOVE, 5. 196 Purine metabolism in Bacteria. II. J. Biol. Chem. (in press). HoTcxIsSs, M. 1923 Studies on salt action. VI. The stimulating and inhibiting effect of certain cations on bacterial growth. J. Bact., 8, 141-162. JOHNSON, M. J. 1941 Isolation and properties of a pure yeast polypeptidase. J. Biol. Chem., 137, 575-686. LARDY, H. A. 1951 The influence of inorganic ions on phosphorylation reactions, pp. 477499. In Phosphorus metabolism. Vol. I. Edited by McElroy, W. D., and Glass, B. The Johns Hopkins Press, Baltimore, Md. LEHNINGER, A. L. 1950 Role of metal ions in enzyme systems. Physiol. Revs., 30, 393-429. LEPAGE, G. A., AND UMBREIT, W. W. 1945 Manometric techniques and related methods for study of ti8sue nmtabolism, pp. 160-172. 1st Ed. Edited by Umbreit, W. W., Burris, R. H., and Stauffer, J. F. Burgess Publishing Co., Minneapolis, Minn. LOWRY, 0. H., AND HASTINGS, A. B. 1942 Histochemical changes associated with aging. 1. Methods and calculations. J. Biol. Chem., 143, 257-269. LOWRY, 0. H., AND LOPEZ, J. A. 1946 The determination of inorganic phosphate in the presence of labile phosphate esters. J. Biol. Chem., 162, 421-428. MAAs, W. K. 1952 Pantothenate studies. III. Description of the extracted pantothenatesynthesizing enzyme of Eacherichia coli. J. Biol. Chem., 198, 23-32. MACLEOD, R. A., AND SNELL, E. E. 1947 Some mineral requirements of the lactic acid bacteria. J. Biol. Chem., 170, 351-3. MCQUARRIE, I. 1953 A symposium on potassium metabolism. J. Lancet, 73, 159-256. NELSON, N. 1944 A photometric adaptation of the Somogyi method for the determination of glucose. J. Biol. Chem., 153, 375-380. NOS8AL, P. M. 1952 The effects of glucose and potassium on the metabolism of pyruvate in Lactobacillus arabinosus. Biochem. J. (London), 50, 591-595. QUASTEL, J. H., AND WEBLET, D. M. 1942 Vita-

19541

RELATIONSHIP OF POTASSIUM TO METABOLISM IN E. COLI

min B1 and bacterial oxidations. 2. The effects of magnesium, potassium, and hexosediphosphate ions. Biochem. J. (London), 36, 8-33. RAHN, 0. 1936 Substitutes for potassium in the metabolism of the lowest fungi. J. Bact., 32, 393-399. RAVEL, J. M., EAKIN, R. E., AND SHIVE, W. 1948 Glycine, a precursor of 5(4)-amino-4(5)imidazolecarboxamide. J. Biol. Chem., 172, 67-70.

ROBERTS, R. B., ROBERTS, I. Z., AND COWIE, D. B. 1949 Potassium metabolism in Escherichia coli. II. Metabolism in the presence of carbohydrates and their metabolic derivatives. J. Cellular Comp. Physiol., 34, 259291. SCHMIDT, G., HECHT, L., AND THANNHAUSER, S. J. 1949 The effect of potassium ions on the absorption of orthophosphate and the formation of metaphosphate by bakers yeast. J. Biol. Chem., 178, 733-742. SHEPPARD, C. W. 1951 New developments in

193

potassium and cell physiology: 1940-1950.

Science, 114, 85-91. SHIVE, W., ACKERMANN, W. W., GORDON, M., GETZENDANER, M. E., AND EAKIN, R. E. 1947 5(4)-amino-4(5)-imidazolecarboxamide, a precursor of purines. J. Am. Chem. Soc., 69, 725. SOMOGYI, M. 1952 Notes on sugar determination. J. Biol. Chem., 195, 19-23. STADTMAN, E. R. 1952 The purification and properties of phosphotransacetylase. J. Biol. Chem., 196, 527-534. STETTEN, M. R., AND Fox, C. L., JR. 1945 An amine formed by bacteria during sulfonamide bacteriostasis. J. Biol. Chem., 161, 333-349. STUMPF, P. K. 1947 A colorimetric method for the determination of desoxyribonucleic acid. J. Biol. Chem., 169, 367-371. WINSLOW, C.-E. A., AND DOLLOFF, A. F. 1928 Relative importance of additive and antagonistic effects of cations upon bacterial growth. J. Bact., 15, 67-92.