Vol. 40 CATION EXCHANGES IN YEAST FERMENTATION 67 REFERENCES Berenblum, S. & Chain, E. (1938). Biochem. J. 32, 295. Boyle, P. J. & Conway, E. J. (1941). J. Phy8iol. 100, 1. Boyle, P. J., Conway, E. J., Kane, F. & O'Reilly, H. (1941). J. Physiol. 99, 401. Brandt, K. (1945). Acta Phys. Scand. 10, Suppl. xxx. Conway, E. J. (1939). Microdiffusion Analysis and Volu- metric Error. London: Crosby Lockwood. Conway, E. J. (1945). Biol. Rev. 20, 56. Conway, E. J. & Boyle, P. J. (1939). Nature, Lond., 144, 709. Conwav, E. J. & Breen, J. (1941). Nature, Lond., 148, 724. Conway, E. J. & Breen, J. (1945). Biochem. J. 39, 368. Conway, E. J. & Fearon, P. J. (1944). fJ. Physiol. 103, 274. Conway, E. J. & Moore, P. T. (1945). Nature, Lond., 156, 170. Conway, E. J. & O'Malley, E. (1942). Biochem. J. 36, 655. Conway, E. J. & O'Malley, E. (1943). Nature, Lond., 151, 252. Dixon, H. H. & Atkin, W. R. B. (1913). Sci. Proc. R; Dublin Soc. 14, 1. Friedemann, T. E. & Haugen, G. E. (1943). J. biol. Chem. 147, 415. Heppe!, L. A. (1939). Amer. J. Phy8iol. 127, 385. Heppel, L. A. (1940). Amer. J. Phy8iol. 128, 440. Leibowitz, J. & Kupermintz, N. (1942). Nature, Lond., 150, 253. Lu, G. D. (1939). Biochem. J. 33, 249. Mahdihassan, S. (1930). Biochem. Z. 226, 203. Pulver, R. & Verzar, F. (1940). Helv. Chim. Acta, 23, 1087. Salit, P. W. (1932). J. biol. Chem. 96, 659. Shohl, A. T. & Bennett, H. B. (1928). J. biol. Chem. 78, 643. The Metabolism and Functioning of Vitamin-like Compounds 1. AMMONIA FORMATION FROM GLUTAMINE BY HAEMOLYTIC STREPTOCOCCI; ITS RECIPROCAL CONNEXION WITH GLYCOLYSIS BY HENRY McILWAIN, Unit for Re8earch in Cel Metaboli8m (Medical Re8earch Council), Department of Biochemi8try, The Univer8ity, Sheffield (Received 5 October 1945) Nutritional studies have led to the identification of many substances which are essential for normal growth or behaviour of living organisms. In certain but not in all cases, the need for such substances has been found to be connected with their reactions in particular component systems of the organism. The commonest biochemical method of localizing vita- min action has followed Peters's (1936) observation that the metabolism of pyruvate, by intact aniimals or by tissues separated from them, was retarded during deficiency of aneurin but restored after its addition. This method, involving the successive in- duction and making good of a nutritional deficiency, has not always been found applicable. This is under- standable for, to take as an example the case quoted above, stimulation by aneurin depended on circum- stances which included: deficiency in aneurin being compatible with life of the organism, and means being found to produce deficient tissues; the de- ficient tissues being able to synthesize the added vitamin to the functioning coenzyme; and other components in the system using the coenzyme not being proportionately reduced in deficiency and remaining capable of reacting with the newly syn- thesized coenzyme. Investigations of a different type (Mcllwain & Hughes, 1944) gave an instance of a relationship which might be regarded as the inverse of that above, and used as a supplementary or comple- mentary method for localizing vitamin action. In this case, the addition of pantothenate to suspen- sions of bacteria which needed pantothenate in growth was not at first found to initiate or accelerate any process except its own inactivation; but the inactivation was connected with bacterial meta- bolism in that it occurred only during the progress of a reaction such as glycolysis. Evidence was obtained that this might be due to participation of pantothenate in glycolysis. The effect was thus in some ways comparable to the connexion found between aneurin requirement and carbohydrate intake in animals, and also to certain linked reac- tions in intermediary metabolism. The investiga- tions of Lwoff & Lwoff (1937), Morel (1941) and of Winzler, Burk & du Vigneaud (1944) are also in many ways analogous. In the present series an attempt is made to find how general and significant are such reactions and how their occurrence in non-proliferating cells may be connected with the physiological functioning of the substance, for example, in growth. In its rela- tionship to fl-haemolytic streptococci, for some strains of which it is a growth-essential (Mcllwain, Fildes, Gladstone & Knight, 1939), glutamine has now been found to offer an example of both methods of connexion betweeD a vitamin-like substance and Bioehem. 1946, 40 5
Transcript
Vol. 40 CATION EXCHANGES IN YEAST FERMENTATION 67
REFERENCES
Berenblum, S. & Chain, E. (1938). Biochem. J. 32, 295.Boyle, P. J. & Conway, E. J. (1941). J. Phy8iol. 100, 1.Boyle, P. J., Conway, E. J., Kane, F. & O'Reilly, H. (1941).
J. Physiol. 99, 401.Brandt, K. (1945). Acta Phys. Scand. 10, Suppl. xxx.Conway, E. J. (1939). Microdiffusion Analysis and Volu-
metric Error. London: Crosby Lockwood.Conway, E. J. (1945). Biol. Rev. 20, 56.Conway, E. J. & Boyle, P. J. (1939). Nature, Lond., 144, 709.Conwav, E. J. & Breen, J. (1941). Nature, Lond., 148, 724.Conway, E. J. & Breen, J. (1945). Biochem. J. 39, 368.Conway, E. J. & Fearon, P. J. (1944). fJ. Physiol. 103, 274.Conway, E. J. & Moore, P. T. (1945). Nature, Lond.,
156, 170.Conway, E. J. & O'Malley, E. (1942). Biochem. J. 36, 655.
Conway, E. J. & O'Malley, E. (1943). Nature, Lond.,151, 252.
Dixon, H. H. & Atkin, W. R. B. (1913). Sci. Proc. R;Dublin Soc. 14, 1.
Friedemann, T. E. & Haugen, G. E. (1943). J. biol. Chem.147, 415.
Heppe!, L. A. (1939). Amer. J. Phy8iol. 127, 385.Heppel, L. A. (1940). Amer. J. Phy8iol. 128, 440.Leibowitz, J. & Kupermintz, N. (1942). Nature, Lond.,
150, 253.Lu, G. D. (1939). Biochem. J. 33, 249.Mahdihassan, S. (1930). Biochem. Z. 226, 203.Pulver, R. & Verzar, F. (1940). Helv. Chim. Acta, 23, 1087.Salit, P. W. (1932). J. biol. Chem. 96, 659.Shohl, A. T. & Bennett, H. B. (1928). J. biol. Chem. 78, 643.
The Metabolism and Functioning of Vitamin-like Compounds1. AMMONIA FORMATION FROM GLUTAMINE BY HAEMOLYTIC STREPTOCOCCI;
ITS RECIPROCAL CONNEXION WITH GLYCOLYSIS
BY HENRY McILWAIN, Unit for Re8earch in Cel Metaboli8m (Medical Re8earch Council),Department of Biochemi8try, The Univer8ity, Sheffield
(Received 5 October 1945)
Nutritional studies have led to the identification ofmany substances which are essential for normalgrowth or behaviour of living organisms. In certainbut not in all cases, the need for such substances hasbeen found to be connected with their reactions inparticular component systems of the organism. Thecommonest biochemical method of localizing vita-min action has followed Peters's (1936) observationthat the metabolism of pyruvate, by intact aniimalsor by tissues separated from them, was retardedduring deficiency of aneurin but restored after itsaddition. This method, involving the successive in-duction and making good of a nutritional deficiency,has not always been found applicable. This is under-standable for, to take as an example the case quotedabove, stimulation by aneurin depended on circum-stances which included: deficiency in aneurin beingcompatible with life of the organism, and meansbeing found to produce deficient tissues; the de-ficient tissues being able to synthesize the addedvitamin to the functioning coenzyme; and othercomponents in the system using the coenzyme notbeing proportionately reduced in deficiency andremaining capable of reacting with the newly syn-thesized coenzyme.
Investigations of a different type (Mcllwain &Hughes, 1944) gave an instance of a relationshipwhich might be regarded as the inverse of that
above, and used as a supplementary or comple-mentary method for localizing vitamin action. Inthis case, the addition of pantothenate to suspen-sions of bacteria which needed pantothenate ingrowth was not at first found to initiate or accelerateany process except its own inactivation; but theinactivation was connected with bacterial meta-bolism in that it occurred only during the progressof a reaction such as glycolysis. Evidence wasobtained that this might be due to participation ofpantothenate in glycolysis. The effect was thus insome ways comparable to the connexion foundbetween aneurin requirement and carbohydrateintake in animals, and also to certain linked reac-tions in intermediary metabolism. The investiga-tions of Lwoff & Lwoff (1937), Morel (1941) and ofWinzler, Burk & du Vigneaud (1944) are also inmany ways analogous.
In the present series an attempt is made to findhow general and significant are such reactions andhow their occurrence in non-proliferating cells maybe connected with the physiological functioning ofthe substance, for example, in growth. In its rela-tionship to fl-haemolytic streptococci, for somestrains of which it is a growth-essential (Mcllwain,Fildes, Gladstone & Knight, 1939), glutamine hasnow been found to offer an example of both methodsof connexion betweeD a vitamin-like substance and
Bioehem. 1946, 40 5
H. McILWAINa metabolic process, in that it was metabolized onlyduring the course of a reaction such as glycolysis;and, also, it stimulated the reaction. The presentpaper describes the circumstances under whichmetabolism of glutamine occurs in haemolyticstreptococci; the products, kinetics, and specificityof the reaction will be described later.
EXPERIMENTAL
Organi8rn8. The cultures examined are given inTable 1; all were of human origin. ,-Haemolyticstreptococci, except the two stated as being pas-saged, were first examined within a few days oftheir isolation. For the present experiments theywere maintained for 2-3 weeks on blood-broth,during which time their reactions to glutamine did'not change. From the blood-broth, subcultures onserum-broth-agar were made for daily use. Passagewas through mice at intervals of about a month,the cultures meanwhile being maintained by weeklysubculture in blood-broth.
tants during experiments were made from side-arms.In aerobic experiments either (a) the reaction mix-ture was identical with that above but yellow phos-phorus omitted from the well and air-CO2 used inthe gas space, or (b) the reaction mixture contained0-06M-phosphate ofpH 7-6 in place of bicarbonate;the well, 2N-NaOH; and the gas space, air.
Material&. In most cases specimens of 1( + )glut-amine ofnatural origin were employed. Commercialspecimens were purified by separation of associatedmaterials as Pb salts, precipitation of glutamine asa Hg compound, and its recrystallization fromaqueous alcohol (Vickery, Pucher & Clark, 1935).Three independently prepared specimens gave indis-tinguishable biological responses. Synthetic dl-glutamine was prepared from synthetic dl-glutamicacid (McIlwain & Richardson, 1939) by Bergmann,Zervas & Salzmann's (1933) method.
Determination of NH3 and glutamine. Free NH3was determined in a Parnas apparatus by distillationfrom a borate-Na2CO3 mixture ofpH 9 and titration,warm, with N/100 H2SO4 in aqueous ethanol, using
Growth. For metabolic studies the organisms weregrown for 16-18 hr. in 100 ml. batches ofthe 'growthmedium C' of McIlwain & Hughes (1944). This con-tained a casein hydrolysate, glucose (0-05M), ar-ginine (4 x 10-4M), glutamine (4 x 1O-4M), addedammonium salts (6-8 x 10-4M), several growth-pro-moting substances, and was buffered with phosphateand NaHCO3/CO2; its initial pH of 7-6 remained > 6when the organisms were collected. This was bycentrifuging, after which the cells from-100 ml. werewashed twice with 10 ml. of 0-9% NaCl and sus-pended in salt solution. Suspensions were usedwithin 2 hr. of their preparation, but remainedcapable of glycolysis and glutamine metabolismafter at least 30 hr. storage at 20.
Metaboli8m. Most experiments were carried outanaerobically at 370 in Warburg vessels (conical, ofc. 20 ml.) with 1-10 mg. dry weight of bacteria and3-3-5ml. ofsolution containing 150,umol. ofNaHCO3and other reagents in equilibrium with 5% CO2in N2. Anaerobiosis was maintained by yellowphosphorus in a centre well, and additions of reac-
OriginNational Collection of Type Cultures (no. 5631'Richards.'); passaged through mice
ThroatVagina 4'ThroatThroatThroat; passaged through miceFaeces
Tashiri's indicator incorporated in the standard acidaccording to Conway (1935). Glutamine was deter-mined as the additional NH3 liberated under themild conditions of hydrolysis specified by Vickery,Pucher, Clark, Chibnall & Westall (1935); for dis-tillation of the total NH3 after such-hydrolysis, amixture ofNaOH and borate ofpH 9-5-10 was used.Both determinations were carried out on measuredportions of complete reaction mixtures, withoutseparation of bacteria. It is recognized that cautionis needed in interpreting 'labile amide-N' as glut-amine, in miscellaneous natural materials; but noinconsistencies have been found in accepting suchan interpretation in the present experiments. Theseinvolve relatively large quantities of glutamineadded as the pure compound, and relatively smallquantities of bacterial substance. The 'labile amide-N' from 5 mg. dry weight of bacteria, in a typicalexperiment in which the NH3 chanige due to theorganisms was 8,umol., was 0 5,umol.; no opinion isexpressed with respect to the origin of the bacterial' amide-N '.
68 I946
GLUTAMINE AND GLYCOLYSISLactic acid was determined through the acet-
aldehyde formed on its oxidation (Friedemann &Graeser, 1933) and suecinic acid by the dehydro-genase of muscle (Krebs, 1937).
RESULTS
A. Glutamine breakdown by streptococci(1) Its independence of growth and viability. Pre-
liminary experiments showed that when severalstrains of haemolytic streptUococci were grown over-night in media containing known quantities ofglutamine, much of this substance disappearedfrom the culture fluids and could not be extractedfrom the bacterial cells, nor detected in them inquantities sufficient to account for the amounts lostfrom the culture fluids. In attempting to reproducethis process under simpler conditions (Table 2)
glucose. The reaction, with washed streptococci, inmost cases proceeded readily in inorganic salt solu-tions containing glucose and glutamine. The reactionrequired the presence of the organisms, but thesuspensions concerned changed little, if at all, inoptical density during the reaction. Thus in anexperiment in which 8-1 ,umol. of glutamine wereinactivated in 105 min., the dry weight of strepto-cocci (Richards strain) used was 2-8 mg. This waspart of a culture yielding 56 mg. of organisms andoriginally containing 40,umol. of glutamine, whichwas a fivefold excess over the minimum quantityneeded for maximal growth of the organisms. The2-8 mg. of dry bacterial matter were thus associatedwith 0-4-2pmol. of glutamine during growth.During the period of metabolism quoted, no changein optical density of the suspensions was observed,which implied any change to be < 1-5%; the growth
Table 2. Association of streptococcal metabolism of glutamine with a particular constituent of growth nedia
(All reaction vessels initially contained bicarbonate (150,umol.), glutamine (2540,umol.) and the additional substancesgiven below in equilibrium with a gas mixture of 5% CO2 in N2, except in the aerobic experiment, when air replacedthe N2. 'Bic.-sailine' contained the inorganic salts of Krebs & Eggleston (1940) but with only 3 umol. of phosphate pervessel. Reactions were initiated by adding the bacterial suspensions to the other reactants.)
Reaction mixture
Organisms (see Table 1);dry wt. (mg.)
R; 3-8
R; 13-61,,
R; 2-5,,
R; 6-9; collected and usedat 22 hr.
R; 7-4; collected at 22 hr.and used at 46 hr.
R; 6-5; collected and usedat 46 hr.
G; 3-4,,
3081; 5-2
UC2; 3-2
Substances (,mol.) in 3-5 ml.Bic.-saline + glucose (200)
streptococcal suspensions were added to solutionswhich contained glutamine and various constituentsof the growth media, but which lacked some factorsessential for streptococcal growth. It was foundthat a reaction leading to loss of the labile amidegroup of glutamine occurred independently of themany constituents of a casein hydrolysate or of ayeast preparation, but required the presence of
of 1-5 %° of 2-8 mg. of the organisms would be asso-ciated with a change of 6 to 30m,mol. of glutamine,or 0-07-0-37 % of the change observed. Thus thechange cannot be ascribed to growth, and the pro-cess of glutamine usage is clearly dissociated fromgrowth. It was also found independent of viability,as actively glycolyzing but mainly non-viable strep-tococcal suspensions prepared as previously de-
5-2
Vol.t4O 69
H. McILWAINscribed (McIlwain & Hughes, 1944) reacted withglutamine during glycolysis.
(2) Accompanying proces8es: metaboli8m of glu-co8e. It was necessary for glutamine metabolismnot only that glucose should be present but alsothat it should itself be metabolized: partial inhibi-tion ofglycolysis by iodoacetate, fluoride, or cyanidewas found greatly to inhibit the breakdown ofglutamine (Table 3). Pantoyltaurine (10,umol./3-3 ml.) or a penicillin preparation (20 Oxfordinits/3.3 ml.) affected neither glycolysis nor reactionwith glutamine.
carried out on the following basis. If the reactionwere due to glucose or its metabolism producing alow E% and activating an independent enzyme, thismight also be achieved by other means. Neitherglutathione nor thiolacetate had this effect (Table 3);CN-, which activates some peptidases, has beenobserved above to inhibit the glutamine reaction.Concentrations of KCI up to M/10, the effects ofwhich on animal tissues have been interpreted asdue to permeability changes, did not induce reactionwith glutamnine; nor did 'urethane. Breakdown ofthe Richards and group G streptococci was obtained
Table 3. As8ociation of streptococcal metaboli8m of glutamine, with metaboliwm of gluco8e
(Experiments were performed anaerobically. 'Amino acids' comprised the group of 14 used by Gladstone (1939), inthe proportions there stated; other details were as described in Table 2.)
Reaction mixture
Substances (pmol.) in 3-5 ml.Bic.-saline + glucose (200)
functionally related compounds did not permit thereaction with glutamine (Table 3). Reactions com-
parable to glycolysis did not take place. Underanaerobic conditions lactate, the main product (seebelow) of the organisms' reaction with glucose, per-
mitted 1/50 of the glutamine decomposition asso-
ciated with glucose; the ratio with pyruvate was
1/15.In attempting to induce the glutamine reaction
without metabolism of glucose, experiments were
by rubbing with powdered glass by Wiggert, Silver-man, Utter & Werkman's (1940) technique and a
few variations of it. The resulting preparations were
almost inactive towards glutamine.(3) Accompanying processes: procuction of NH3.
The method used in determining glutamine involvedestimation of NH3 in the same solutions, and it wasimmediately found that the reaction of streptococciwith glucose and glutamine liberated a volatile base,characterized (see later) as NH.. The anaerobic pro-duction per mol. of glutamine undergoing reaction
Organisms;dry wt. (mg.)
0R; 5-4
R
P,,,,
R; 4-9VP
.,
,,
,,
R; 5-0,,
R; 2-6VP
Pt
Pt
F; 5-4
lgR,,9
70 &946
Vol. 40 GLUTAMINE AND GLYCOLYSIS
Table 4. Yield of ammonia from streptococcal metabolism of glutamine during glycoly8si
71
(Glutamine was the natural 1( +) compound except in the instance specified. Details of suspending fluids are given inTable 2; all contained glucose, 200pmol. per 3-5 ml. Experiments were carried out anaerobically except where indicated.Casein hydrolysate was that of McIlwain & Hughes (1944); 20 mg. were added per 3-5 ml. in the experiment indicated.)
approximated to 1 mol., the commonest yield beingbetween 0-85 and 0-97 mol./mol. of glutamine(Table 4). Lower ratios were found occasionally,especially aerobically (in the phosphate mediumdescribed in the experimental part), but ratiossignificantly greater than unity were not observed.Ratios determined anaerobically were not markedlydependent on the strain of organism, its Qco,, orthe suspending fluid.
B. The anaerobic reaction of streptococciuith glucose
For examining its products, the reaction wasstudied under the conditions employed in following
the organisms' reaction with glutamine: anaero-bically, with washed suspensions in saline solutionscontaining inorganic salts and bicarbonate which,with the CO2 of the gas phase, formed a buffer ofpH 7-6. The effects of additional substances, andthe course of the reaction, are considered later.In the simple medium, with excess glucose, themajority of the strains caused a gas evolution of200-400u1I./mg. dry weight/hr. (Qco.; Tables 4, 7, 9).WVith limited quantities of glucose, gas evolution bythe two strains examined was found to proceed atrates almost identical with those given in the pre-sence of excess glucose, but to reach sharply amaximum value corresponding to a little less than2 mol. of gas per mol. of glucose added (Table 5).
Table 5. Producs of streptococcal glycolysis(Reactions were carried out anaerobically in the bicarbonate saline of Table 2.)
Table 6. Source of gas evolved by streptococci during glycolysis in the presence of bicarbonate
(Vessels contained in their main portion the sline of Table 2 (but with the lesser quantities of bicarbonate specifiedbelow), glucose, 200,umol., and other substances indicated. Organisms were contained in one side-arm and 0-2 ml. 5N-H2SO4 in another. Experiments, except where indicated, were anaerobic, and were continued until gas production bythe organisms themselves was maximal.)
reaction and before addingby the later acid toaddition 2nd vesselof acid (min.)325162280298329
6280859562
Difference ingas evolution
between(a) and (b),as % of (a)
-1-8+2-5+2-2<1-1-8
By determining the total C02 which could beliberated by excess acid from the complete systemof organisms and reagents, before and after a periodof metabolism (Table 6), the gas production was
shown to be almost entirely due to displacement ofC02 from bicarbonate by acids formed from thebacteria. The samlte was foulnd true aerobically. Itwas anticipated (cf. Hewitt, 1932) that the anaerobicproduct from glucose would be lactic acid, and thiswas determined (Table 5). Results showed thatlactic acid appeared in the ratio of about 1-9 moL/mol. of glucose supplied, i.e. that all but 5% of theglucose which reacted and 1 or 2% 6f the gas whichwas produced could be accounted for by the glyco-lytic reaction: C6H12O6 2C2H603 . Of other possibleproducts, succinic acid has been determined audany production found to be < 0-5 mol. % of theglucose reacting. Evolution of C02 from bicar-bonate- and glucose-containing solutions by thepresent streptococci is subsequently referred to as
a measure of glycolysis.
C. Stimulation of glycolysis by glutamtine
(1) Suboptimal glycolysis of streptococcal supen-
sions. The rates of glycolysis by ,-haemolyticstreptococci, in the relatively simple media of thepreceding experiments, were not always constantthroughout the course of the reaction. On its initia-tion by addition of streptococcal suspensions to theglucose-containing solutions, the reaction began ata rate which, in the course of some 30 min., fell toabout 70-80% of its initial value; after this itremained approximately constant or fell at a slowrate. Fig. 1 illustrates this with respect to theRichards strain; Table 7 includes other strains ofstreptococci, of which the majority behaved simi-larly. The rate of glycolysis was little affected bythe addition of many pure substances (see below);it was increased some 5-15% by a casein hydro-lysate or a yeast preparation (Table 7), and was
considerably increased by the presence of broth.Suspensions which had not been washed in salinealso afforded higher values.
Glutamine (.&mol.)
50
16-76-7
0
22~
200
0
5 15 25 33-3Time (min.) after addition of organisms
Fig. 1. Effect of glutamine on streptococcal glycolysis.Ordinate: Anaerobic gas evolution in successive periodsof 10 min., expressed as QC02 (pl./mg. dry wt./hr.) fromsuspensions of the Richards strain after addition to saline-bicarbonate-glucose containing the quantities of glut-amine indicated (except the dotted line A, which refersto a mixture containing no glutamine but NH4C1,50I&mol.). When the experiment was stopped 35 min.after addition of the organisms, the following quantities(!&mol.) of glutamine remained: 43-3 (of 50), 10-2 (of 16.7),0-1 (of 6.7) and <0-1 (of 2.2).The organismis were thus capable of glycolysis at
rates some 60-100% higher than those observed insimple salt solutions, and substances inducing higher
OrganismsG
9,,
P,,
R
EH. MaILWAIN I94672
GLUTAMINE AND GLYCOLYSIS
Table 7. Effects of materials on the rate ofstreptococcal glycolysis
(Saline suspensions of organisms were added from side-arms, to vessels containing bic&rbonate-saline with glucose(200,umol.) and the addenda specified below, at times takenas zero. Details of constituents of the reaction-mixturesare given in Table 2. In cases marked (*) observation was
prevented by growth of thie bacteria under examination.No allowance has been made for C02 retention and thestimulation recorded as being given by broth and serum
solutions is therefore minimal.)Qco2 as % of the
5-25 min. valuewithout addenda(which is givenitalicized, in
Casein hydrolysate(10 mg.) and yeast pre-paration (10 mg.)Serum, 29%Broth, 29%
,,30 Glutamine, 0-007M0
Serum, 29%, 5.7%Broth, 29%Glutamine, 0-007M
UC2,,t
,,.
9,,
149
145(223)105
129
158155
(200)118
103163124(460)93,95163100
rates were present hii broth and perhaps removedfrom the cells by washing.
Glutamine made good a large part, of the organ-
isms' deficiency in this respect. Glycolysis was
increased to up to 100% of its original value, theincrease varying with different strains and prepara-
tions ofthe streptococci. With the strains ofTable 7,the increase in glycolysis due to glutamine was notadditive to that caused by broth, though greaterstimnulation was found to be caused by casein and
yeast preparations with glutamine, than by eitheralone. The present broth permitted growth ofexacting streptococci in glutamine-deficient media(cf. Mcfwain et al. 1939). It was concluded thatthe stimulation caused by broth was due in part toglutamine, and in part to substances of the sortoccurring in the casein hydrolysate or yeast pre-paration. Fig. 1 shows 0-003M-glutamine to beadequate for its maximal effect with the presentexperimental arrangement; the stimulation givenby 0-00063M-solutions, though initially not muchless than that given by 0-015m, fell much morerapidly.As the experiments of Fig. 1 were carried out with
organisms whose glycolysis was falling in the initialphase of the reaction, a different experimental ar-rangement was chosen to show more. clearly whetherthe effect of glutamine was a stimulation of glyco-lysis rather than its maintenance by the substanceat a level fromwhich it fell in its absence. In Fig. 2Aglutamine was added to already glycolyzing organ-isms at different times during the reaction. In eachcase a large and prompt increase in glycolyais fol-lowed. The substance is thus capable of increasing,and not only of maintaining, glycolysis. Its effectis manifested within 2 min. of its addition and isthus not likely to be due to growth of the bacteria;absence of growth was confirmed by optical densitymeasurements.
(2) Possible impurities in glutamine specimens.The majority of the present experiments had beencarried out with specimens of natural l-glutamine,which was shown by isolation (McIlwain et al. 1939)to be the form encountered by the organisms inanimal tissues. As the glycolysis experiments werecarried out in simple media, it was offirst importanceto find whether the stimulation caused by the speci-mens was due to glutamine rather than to bio-logically active impurities which can always beassumed to be associated with such preparations.Carefully purified natural glutamine was found togive the same response as ordinary specimens, and,further, glutamine synthesized from synthetic glut-amic acid gave a similar response. The quantitativeconsiderations of the following paragraphs givefurtheAr reasons for associating the effect on glyco-lysis with glutamine itself.
(3) Responses to varying concentrations of glut-amine. In the experiment of Fig. 2B, the quantitiesof glutamine added increased the total glycolysis inthe first hour by amounts between 67 and 42% ofthat ofthe control without glutamine. Nevertheless,the maximum increases in Qc,02 which occurred inthe first 5 or 10 min. after glutamine hadbeen added,were approximately equal, independently of thequantity of glutamine added (0.9-451pmol.) andwere all of about 90%. The effect of falling quan-tities of glutamine was to produce a more rapid fall
VoI. 40 73
H. McILWAINin the rate of glycolysis during the hour after theiraddition.A reason for this behaviour is to be seen in the
finding of § A, that glutamine is decomposed duringglycolysis. At the end of the experiment of Fig. 2 B,little or no glutamine remained of the smallestquantity added and only about one-third of the4-5,mol. A similar decomposition is shown to haveoccurred during the differently arranged experimentof Fig. 1. It will be seen that in both these cases,when all added glutamine had been decomposed,
Time (min.)
CO2 evolution in the presence of glutamine was dueto the same processes as occurred in its absence, itsmain product, lactic acid, was estimated in solutionsin which glycolysis had occurred with and withoutglutamine. It was found (Table 8) that the increasedgas formation could largely be accounted for byincreased lactic acid production. Succinic acidformation was found to remain at <0- 5 mol. %.Anaerobic CO2 evolution by streptococci with glut-amine as their only substrate, remained at the lowlevel associated with the 'endogenous' metabolism
Fig. 2. Promptness of response by glycolyzing streptococci to additions of glutamine. Uni-ts: as Fig. 1. Vessels initiacontained saline-bicarbonate-glucose and organisms in 3 ml., except the controls, of which the volume was 3-5 ml.In A, glutamine (45,mol. in 0-5 ml.) was added at the times indicated; in B, the varying quantities of glutamineshown (in 0-5 ml. water) were added at 60 min. After the latter experiment, the quantities of glutamine (,umol.)remaining were: 36-2 (of 45), 1-6 (of 4-5) and <0-1 (of 0-9).
the level of glycolysis had not fallen to that of thecontrol without added glutamine; this is consideredbelow.
(4) Products of glucose metabolism in the presenceof glutamine. To see whether the increased rate of
Table 8. Comparison of increase in gas evolutionfrom bicarbonate-glucose, following addition of glut-amine or ammonia, and increase in lactic acidformation
(Vessels contained excess glucose (200,umol.) and thereagents in the saline-bicarbonate of Table 2. Suspensionsof the Richards streptococci were added from side-arms,and the reaction occupied 110 min.)
of the organism (Tables 2 and 3). The final total ofgas evolved from limited quantities of glucose inexperiments of the type of Table 6 but with addedglutamine, differed from controls without glutamineby <2%.
D. Actions on glycolysis of some compounds func-tionally or structurally related to glutamine
The effects of the following compounds are, forpresent purposes, selected from a fuller investigationofthe structural and species specificity ofglutamine-action.
(1) Ammonia. It had been noticed in severalinstances, of which Figs. 1 and 2 give examples,that although optimal stimulation of glycolysis inthe presence of small quantities of glutamine wastransitory only, the rate of glycolysis did not, whenpractically all added glutamine had been decom-posed, fall to the level of controls without addedglutamine. As ammonia was known to be formed
74 I946
30
GLUTAMINE AND GLYCOLYSIS
from glutamine during such experiments (§ A) andknown to stimulate reactions in yeast which areanalogous to the present one (Winzler et al. 1944),the effect of ammonium salts on streptococcalglycolysis was examined. An increase was foundwhich was sufficient to account for the level ofglycolysis after decomposition of glutamine.
Table 9. Stimulation of streptococcal glycoly8is byammonia; compared with, and in addition to, theeffect of glutamine
(Experimental arrangement as described in Table 7.)
Reaction mixture
Organism; Addendabatch (f&mol.)
R; I and II 0,,91 NH4C1, 0 05,, NH4CJ, 0-5,,9 NH4Cl, 59,, NH4C1, 509,, Glutamine, 25
R; III 0,, NH4C1, 2
NH4CI, 40Glutamine, 10Glutamine, 10 and
,,
k; IV
1,,
Strain F
9,,
UC2
UB,,
,,~
NH4CJ, 2Glutamine, 10 andNH4CI, 40
0
NH4C1, 25KCI, 25KCI, 3000
NH4Cl, 25Glutamine, 25Glutamine, 25 andNH4Cl, 25
0
NH4 glutamate, 10Glutamine, 100
NH4 glutamate, 25Glutamine, 25
QC02 as % of thevalues for the same
period withoutaddenda (italicized,in parentheses; as,ul./mg. dry wt./hr.)
The increase in glycolysis caused by ammoniumsalts (Table 9) was not sufficiently great to suggestthat the major effect of glutamine could be duemerely to the ammonia derived from it. The increaseby ammonium salts was less in magnitude even
when concentrations of the salts were used whichwere much greater than could be produced in themedium as a whole, from the glutamine concentra-tions having maximal effect. Thus, during the ex-
periment of Table 9 with batch I of the Richardsstreptococci, the glutamine decomposed after a
period of 50 min. glycolysis was 3 8 t/mol. and the
ammonia of this reaction mixture, 3 0jmol.; yetthe stimulation of glycolysis caused by 50,umol. ofadded ammonium chloride, was just half of thatcaused by the glutamine. Other results quoted inTable 9 indicate that such considerations are notvitiated by zone effects in the actions of differentconcentrations of ammonium salts; the possibilityof ammonia exerting a greater or different effectwhen produced locally by the organism itself isconsidered below.
Nevertheless, the effect of ammonium salts onglycolysis did not appear to be distinct in naturefrom that of glutamine; the actions of the twocompounds were additive if at all, to a small extentonly (Table 9) and like glutamine, ammonium saltsincreased lactic acid formation from glucose in pro-portion to the increase caused in CO2 displacementfrom bicarbonate (Table 8).
Certain effects of K salts on glycolysis in animaltissues have been found similar to those of NH3(Weil-Malherbe, 1936). Streptococcal glycolysis wasfound to be stimulated by high concentrations ofKCl; a 0- 1 M-solution had less effect than 0-005 or0-001M-NH3 (Table 9).
(2) Biotin. As Winzler et al. (1944) found anaction of biotin on yeast respiration and fermenta-tion to be intimately connected with the action ofammonium salts on the same process, the effect ofbiotin on streptococcal glycolysis in the presenceand absence of ammonia, was examined. In neithercase did biotin have an observable effect. In com-paring this result with the action on yeast, it is tobe noted that although the reagents of the presentexperiments were biotin-free, no attempt was madeto obtain streptococcal cells deficient in biotin.
(3) Arginine and urea. Of the compounds relat6dto glutamine which have been studied, many arepotential sources of ammonia to various bacteria;but during glycolysis the present streptococci pro-duced ammonia at a considerable rate only fromarginine (Table 10). This reaction differed notablyfrom that with glutamine in being independent ofglycolysis; it was presumably due to the action ofarginine dihydrolase which Hills (1940) found tobe exhibited by suspensions of streptococci, in-cluding the Richards strain.Comparison of the effects of arginine and glut-
amine on glycolysis as measured by CO2 evolution,required knowledge of the effects of the compoundsand of changes occurring in them, on the CO2-NaHCO3 equilibrium during manometry. The buffer-ing action of the concentrations of arginine, glut-amine and a-ketoglutarate of Table 10 resulted ina small binding of CO2,. This was assessed by themethod of Dickens & Simer (1932) over the rangeof acid formation of the experiments of Table 10,and would affect the values there quoted by <5 %.In addition, however, 1 mol. of C02 is fixed by each
75VoI. 40
Table 10. Stimulation of streptococcal glycolysis by arginine and urea;absence of comparable effect by c-ketoglutarate
(Experimental axrangement as described in Table 7, using the Richards streptococci (different batches in A, B and C).)
ACompound added (/Lmol.)
0NH4Cl (10)Arginine (10)
,,9 corrected as described in textUrea (10)
,, (2)B 0
a-Ketoglutarate (10)19.AI
C
NH4C1 (§5),, and oc-ketoglutarate (10),, ,, 1(2.5)
Glutamine (10)
O
0NH4Cl (10)
, (2-5)Arginine (10)
(2.5)Glutamine (10)
mol. of arginine decomposing according to thereaction (Hills, 1940):
As the increase in bicarbonate ion following thisreaction is small in comparison with the quantityadded as NaHCO3, an approximate correction forthe C02 fixed could be made by adding to the ob-served gas evolution during arginine decomposition,1 mol. of C02 for each 2 mol. of NH3-formed. Inthis way the corrected values of Table 10 were
obtained. They indicated that the effect of arginineon glycolysis was similar to that ofammonium salts.The validity of this conclusion was confirmed bydeternining iodometrically the lactic acid formedin similar experiments, with the results given in C,Table 10.The rate of ammonia production from arginine
by the streptococci, expressed as a mean QNH inTable 10, was equal to or greater than their rate ofproduction of ammonia from glutamine duringglycolysis. If the effect of glutamine on glycolysiswere due only to ammonia, produced in proximityto the organisms, its effect would not be expectedto be greater than that of arginine.
Qco3 as % of the valuewithout addendum, duringthe first hour (italicized,
Lactic acid produced during90 min., as % of the valuewithout addition (italicized,in parentheses; as ,umol.)
(43.5)125120128121171
Mean QNH3(,umol./mg.dry wt./hr.)
<0-03
2-91
<0-03<0-03
<0-03
3-21-81-65
Urea also stimulated glycolysis by the Richardsstreptococci (Table 10), but ammonia was not pro-duced from it in notable quantities; production ata rate 1/100 that of the liberation from argininecould have been detected. The stimulation was lessthan that by ammonia.The possible production of glutamine from urea
and arginine was also examined, but < 0- 3 /tmol. wasfound to be produced by streptococci from 25 ,umol.of either substance, with or without glucose, duringa time in which 8-3 jumol. ofNH3 were liberated fromthe arginine. A possible interpretation of the or-ganisms' response to urea is that small quantities ofammonia were produced from it and underwentfurther reactions with the organisms; small quan-tities of added ammonia have been observed todisappear from certain reaction-mixtures duringglycolysis.
(4) Glutamate and o-ketoglutarate. Sodium gluta-mate (5 or 501tmol./3-5 ml. of glucose-bicarbonatesaline) was found to be without effect on theanaerobic glycolysis of the Richards and Group Gstreptococci. The stimulation given by ammoniumglutamate was not greater than that given byammonium chloride; Table 9 includes results withother organisms which showed that ammonium
76 I946H. MCILWAIN
GLUTAMINE AND GLYCOLYSIS
glutamate did not have the effect of glutamine.Ammonium glutamate was also present in thecasein hydrolysate, whose stimulation of glycolysis(Table 7) was much less than that of glutamine.
oc-Ketoglutarate alone had a small effect only onglycolysis. A range of concentrations, in the pre-sence of ammonium salts, had little effect otherthan that of the salts themselves; in addition to theresults given in Table 10, this has been found to bethe case with other preparations ofRichards strepto-cocci of Qco, 377 and 264 p1./mg. dry weight/hr.and mean QHN. (glutamine) of 2-04 and 1.16 umol./mg. dry weight/hr. respectively, whose glycolysiswas stimulated by glutamine to 147 and 139% ofthe control values, during the first 30 mi.
DISCUSSION
The present findings appear to offer the first instancein which relatively small quantities of glutaminespecifically (as distinct from oc-ketoglutarate orglutamate) have been observed to stimulate a majormetabolic process which proceeded, though moreslowly, in the absence of added glutamine. A con-clusion of the type commonly drawn from suchevidence would be that glutamine normally func-tioned in the process of glycolysis which it stimu-lated. This is feasible in the present instance, as theincreased glycolysis was not beyond ithat obtainablein mixtures of naturally occurring substances whichthe organisms ordinarily encounter; the strepto-coccal cells normally contained material with someof the chemical properties of glutamine, and itsbehaviour as a growth essential showed glutamineto subserve, specifically, some function of impor-tance to the cells. The effect on glycoltsis was thusless likely to be an artefact than that given, forexample, by arsenites or nitrophenols.
There is in addition the less common and inde-pendent type ofobservation, that glutamine reacted,with loss of NH3, during its service as growth-factorand also during the course of glycolysis in nor-proliferating streptococcal suspensions. Strepto-cocci have not been found to cause changes inglutamine under other circumstances (cf. also Hills,1940). Processes have been examined which mightliberate, activate or facilitate access to an inde-pendent enzymee. These included physical breakageof the bacterial cells; the addition of benzene,K salts, thiol compounds and miscellaneous naturalmaterials. Of such treatments, those which pre-vented glycolysis prevented also the glutaminereaction. Inhibitors showed similar effects, but itwas noticeable (Table 3) that the proportional inhi-bitions caused in glycolysis and glutamine break-down were usually not the same. This suggests thatone may expect to separate the glutamine reactionfrom the over-all process of glycolysis, possibly
through study of the separate reactions which con-stitute it; but that metabolism of glutamine isfirmly associated with that ofglucose in the ordinarybehaviour of streptococci.Thus two lines of evidence associate glutamine
with glycolysis in streptococci. Previous studies ofthe compound's metabolic significance have con-cerned organisms more complex than bacteria. Incertain plants, where it constitutes a much greaterproportion of the dry weight than it does of strepto-cocci, glutamine has been considered to be a non-toxic reserve of NH3 or oc-ketoglutarate. The knownimportance of the latter compound in carbohydratemetabolism has been offered in explanation ofinterrelations observed in plants between glutamineand carbohydrates (Chibnall, 1939). Glutamine wasnot in this case allocated a catalytic role, exceptin so far as the oc-ketoglutarate formed from it mightfunction in cycles of the iaocitrate type. The con-clusion thus differs from that derived above withrespect to streptococci.Uncombined glutamine is widely distributed in
animal tissues (McIlwain et al. 1939; Hamilton,1942; Harris, Roth & Harris, 1943; Archibald, 1944)and hydrolysis to glutamic acid and ammonia hasbeen found to occur in several tissues. The reactionin liver and kidney (Krebs, 1935; Archibald, 1944)differed from that in streptococci in being inde-pendent of added carbohydrate and in not beinginhibited, but sometimes stimulated, by CN-. Otheraspects of the behaviour of glutamine in animaltissues showed some points of resemblance-pos-sibly superficial-to the streptococcal reaction. Thusthe concentration of a glutamine-like substance inhuman blood fell after administration of glucose(Harris et al. 1943). Respiration of -certain tissueswhich synthesized glutamine from added glutamicacid was accelerated during the synthesis (Krebs,1935; Weil-Malherbe, 1936); but in this case glut-amic acid itself was oxidized. Glutamine synthesiswas found in these investigations to take placeunder certain limited conditions: only aerobicallyin kidney, whose reaction with carbohydrate wasaerobic only; in brain and retina, anaerobically also,as could glycolysis; CN- inhibited the synthesis andrespiration in kidney but neither glycolysis nor thesynthesis in retina. These limitations were inter-preted (Krebs, 1935) as due to requirement of energyfor glutamine synthesis; but they are very similarto the circumstances under which NH3 is producedfrom glutamine in the present experiments.
Further experiments are required to specify theassociation between glutamine and streptococcalglycolysis, but the following observations are sug-gestive. Weil-Malherbe (1936) confirmed Krebs's(1935) finding that the NH3 reacting in animaltissues with glutamic acid not only formed glutaminebut was transferred further, and considered such
VoI. 40 77
78 H. McILWAIN I946reactions tQ be offunctional importance. In strepto-cocci the greater part ofthe labile amide ofglutamineappeared as NH3. The general view regarding suchlinked metabolism of vitamin-like compounds, de-rived from previous studies of the behaviour ofpantothenate, has been that they represent anunbalanced performance of a normal process, dis-turbed possibly through presentation of a relativelylarge quantity of the substance concerned. Thetransference of NH3 rather than its liberation maythus be the significant reaction of glutamine whichis linked with streptococcal glycolysis. The stagesof conversion of glucose to lactic acid in certainbacteria and in animal tissues are closely similarand the possible participation of NH3 transferencein reactions associated with glycolysis in animaltissues has been suspected but not found to beassociated with glutamine (cf. Kleinzeller, 1942).Reasons can, however, be seen for streptococciaffording a specially suitable material for showingsuch association: intact organisms can be used;their requirement for glutamine as growth factorindicates a relevant deficiency, and their high rateof glycolysis is associated with little endogenouschange in carbohydrate or NH3.
SUMMARY1. Glutamine (determined as labile amide-am-
monia) disappeared during growth of several strainsof haemolytic streptococci in a complex medium.The process could be reproduced in mixtures ofwashed non-proliferating streptococci with mediaconstituents, when it occurred at rates of - 0-5 to-2 ,umol. glutamine/mg. dryweight oforganisms/hr.
2. Reaction did not occur between glutamineand streptococci in salt solutions; the medium con-stituent necessary to the reaction was identified asglucose. It was necessary for glutamine metabolism
under a wide variety of experimental arrangements,not only that glucose should be present but alsothat it should itself be metabolized.
3. The over-all reaction of streptococci withglucose, in the presence or absence of glutamine,was a glycolysis with conversion of 95% of theglucose to lactic acid. The rate of this process inwashed suspensions was increased by various mediaconstituents, and taken to up to 190% of its originalrate, by the addition ofglutamine. Such stimulationby glutamine began within 2 min. of its additionand fell, as glutamine was decomposed, to a levelwhich, when glutamine could no longer be detected,nevertheless remained above that of a controlwithout added glutamine.
4. The organisms' reaction with glutamine pro-duced a maximum of 1 mol. of a volatile base permol. of glutamine reacting. Their glycolysis wasstimulated also by NH3, but to a lesser degree thanby glutamine. Of related compounds, none gavea stimullation equal to or greater than NH3, but theeffect of arginine approached that of NH.; NH3 wasproduced from arginine by the organisms, inde-pendently of glycolysis. Stimulation of glycolysisby NH3 appeared adequate to account for the effectsof arginine and of urea (which also acceleratedglycolysis) and also for the after-effect of glutamine.
5. It is considered that glutamine itself, as dis-tinct from a-ketoglutarate or glutamate, plays arole in streptococcal glycolysis; with evidence fromthe behaviour of glutamine and NH3 in other typesof organism, it is suggested that this participationmay take the form of an NH3 transference.
I am greatly indebted to Mr D. E. Hughes for assistanceduring these investigations; to Prof. H. A. Krebs andDr F. Dickens for advice; and to members of the Depart-ment of Bacteriology of this University for cultures ofstreptococci, and for their passage.
REFERENCES
Archibald, R. M. (1944). J. bid. Chem. 154, 643.Bergman, M., Zervas, L. & Salzmann, L. (1933). Ber. dt8ch.
chem. Ge8. B, 66, 1288.Chibnall, A. C. (1939). Protein Metaboliem in the Plant.New Haven: Yale University Press.
Conway, E. J. (1935). Biochem. J. 29, 2755.Dickens, F. & Simer, F. (1932). Biochem. J. 26, 90.Friedemann, T. E. & Graeser, J. B. (1933). J. biol. Chem.
100, 291.Gladstone, G. P. (1939). Brit. J. exp. Path. 20, 189.Hamilton, P. (1942). J. biol. Chem. 145, 711.Harris, M. M., Roth, R. T. & Harris, R. S. (1943). J. clin.
Inve8t. 22, 569, 577.Hewitt, L. F. (1932). Biochem. J. 26, 208, 464.Hills, G. M. (1940). Biochem. J. 34, 1057.Kleinzeller, A. (1942). Biochem. J. 36, 729.Krebs, H. A. (1935). Biochem. J. 29, 1951.Krebs, H. A. (1937). Biochem. J. 81, 2095.
Krebs, H. A. & Eggleston, L. V. (1940). Biochem. J. 34,442.Lwoff,A. &Lwoff, M. (1937). Proc.Roy.Soc.-B,122, 352,373.McIlwain, H. (1944). Biochem. J. 38, 97.Mcllwain, H., Fildes, P., Gladstone, G. P. & Knight,
B. C. J. G. (1939). Biochem. J. 33, 223.Mcllwain, H. & Hughes, D. E. (1944). Biochem. J. 38, 187.Mcllwain, H. & Richardson, G. M. (1939). 6iochem. J, 33,44.Morel, M. (1941). Ann. In8t. Pasteur, 67, 285, 449.Peters, R. A. (1936). Lancet, 1, 1161.Vickery, H. B., Pucher, G. W. & Clark, H. E. (1935).
J. biol. Chem. 109, 39.Vickery, H. B., Pucher, G. W., Clark, H. E., Chibnall, A. C.& Westall, R. G. (1935). Biochem. J. 29, 2710.
Weil-Malherbe, H. (1936). Biochem. J. 30, 665.Wiggert, W., Silverman, M., Utter, M. F. & Werkman, C. H.
(1940). Iowa State Coil. J. Sci. 14, 179.Winzler, R. J., Burk, D. & du Vigneaud, V. (1944). Arch.
