Some regulatory and integrative aspects of purine nucleotide biosynthesis and its control: An overview

SOME REGULATORY AND INTEGRATIVE ASPECTS OF PURINE NUCLEOTIDE BIOSYNTHESIS AND ITS CONTROL: AN OVERVIEW RICHARD W. E. WATTS Division of Inherited Metabolic Diseases, Clinical Research Centre, Watford Road, Harrow, Middlesex HAl 3UJ, U.K. INTRODUCTION The regulation and integration of purine ribonucleotide and deoxyribonucleotide production can be considered from the point of view of the modifications of the activity of particular metabolic sequences and from that of the relative importance of different metabolic pathways in different tissues or at different times in the same tissue. This communication considers the control of purine biosynthesis firstly in terms of the main metabolic pathways concerned and then from the viewpoint of the relative importance of purine de n o v o synthesis and purine salvage in some special situations.

METABOLIC PATHWAYS The metabolic pathways are: (i) de novo synthesis of inosinic acid (IMP) from small molecular precursors; (ii) purine salvage reactions whereby purine bases are converted directly to nucleotides; (iii) interconversions which involve the separate addition or subtraction of carbohydrate a n d / o r phosphate moeities; (iv) interconversions which do not involve the addition of either a carbohydrate or a phosphate group; (v) the reduction of adenosine and guanosine diphosphates to the corresponding deoxy compounds: (vi) the s--adenosylhomocysteine pathway. Purine de novo Synthesis Purine de n o v o synthesis is usually represented by the linear reaction

sequence with phosphoribosylamine (PRA) formation from glutamine and phosphoribosyl diphosphate (PP-ribose-P) and ending with the cyclodehydration of 5'--phosphoribosyl-5-formamido-4-imidazole carboxamide (Figure 1). It can, however, be regarded as a cyclic process (Figure 2) with either P P - r i b o s e - P or ribose acting as a carrier on which the purine ring is assembled (1). This mechanism involves either the liberation of P P - r i b o s e - P 33

34

RICHARD W. E. WATTS

H

H I

I

PPRP

PP

H2c/NH~ "

I

GIuNI-12

H2c/NH2

--~1

o,'C\o. [o¢C\..

INH2 RIb-P

3 I~H2~/N~CHO

o,'C\..

Rib-P

CI02H CH2

0 ,.I

"i-C02H

HO~ C ~ N ~ c H

CH ¢

wm

I

H2N/ "~/

II

N

"~CH

H2N/C ~ N / I Rib-P

N I

RIb-P

0I

' II c. H2N/C~ N/ I RIb-P

0II

H~/C\c/N~ ' II - - c . OHC\N/C\ N~

9 *

I H

Hc/N~

4

H2N~ C ~ N/ I RIb-P

I Rib-P

o['

HN

Rib'P

0 II

., It

C

4 I~H2~/N\CHO

I RIb-P

HN/C\c/H~ lO,i II c. HC%N/C\Nf I Rib-P

FIG. !. The biosynthesis of inosinic acid represented as a linear reaction sequence, lnosinic acid (IMP) biosynthesis de n o v o . PPRP = PP-ribose-P = phosphoribosyl diphosphate; GIuNH2 = glutamine; Glu = glutamic acid; PP ----inorganic pyrophosphate; NH~-Rib--P= fl-phosphoribosylarnine; Rib---P= ribose 5'-phosphate. 1. amidophosphoribosyltransferase (EC 2.4.2.14); 2, phosphoribosylglycinamide synthetase (EC 6.3.4.13); 3, phosphoribosylglycinamide formyltransferase (EC 2.1.2.2); 4, phosphoribosylformylglycinamidinesynthetase (EC 6.3.3.5); 5, phosphoribosylamidazole synthetase (EC6.3.3.1); 6,phosphoribosylaminoimidazole carboxylase (EC 4.1.1.21); 7, phosphoribosylaminoimidazole-suceinoearboxamide synthetase (EC 6.3.2.6); 8, adenylo-succinate lyase (EC 4.3.2.2); 9, phosphoribosylaminoimidazole carboxamide formyltransferase (EC 2.1.2.3); 10, IMP cyclohydrolase (EC 3.5.4.10).

by the pyrophosphorylase action of hypoxanthine phosphoribosyltransferase ( H P R T ; EC 2.4.2.8) on I M P or the sequential actions of 5"nucleotidase (EC 3.1.3.5) and inosine nucleosidase (EC 3.2.2.2) to yield hypoxanthine and ribose, which is a substrate for ribokinase (EC 2.7.1.15) and hence for ribose phosphate p y r o p h o s p h o k i n a s e ( P P - r i b o s e - P synthetase; EC 2.7.6.1). The latter route is unsatisfactory in energetic terms, requiring the expenditure of three additional high energy phosphate bonds per molecule of purine synthesized. The pyrophosphorolysis of I M P is reversible, so that H R P T has a regulatory role, either channelling I M P towards P R P P production and purine excretion or conserving I M P for either A M P synthesis via adenylosuccinate or G M P synthesis via X M P . The relative importance of these actions of H P R T may vary in different tissues and at different times. The overall balance is presumably set in favor of pyrophosphorolysis in the

PURINE NUCLEOTIDE BIOSYNTHESIS

35

Urate

t

I-lypoxanthine (Hx)

[ Xanthvlate

FIG. 2. Purine

~

¢- ....

.Phosphoribosylglycinamide ...... ! /J . .

~

de novo synthesis as a cyclic process [based on (I)]. ( P P R P = PP-ribose-P =

phosphoribosyl diphosphatc)

uricotelic species which require large amounts of uric acid for nitrogen excretion. It is generally accepted that the rate of purine de n o v o synthesis is controlled by the activity of amidophosphoribosyltransferase (PP-ribose-P amidotransferase; EC 2.4.2.14). Purine nucleotides promote aggregation into a metabolically inactive monomeric form (molecular weight 270,000 Daltons) and PP-ribose-P causes disaggregation to the active form (molecular weight 133,000 Daltons) (2-4). These molecular weights are those of the human placental enzyme and the most recent work (5) shows that the low molecular weight form itself consists of 5 or 6 identical subunits each with a molecular weight of about 24,500 Daltons. Each subunit contains 3.5-4.8 molecules of iron and 3.5-4.4 molecules of sulphur. Amidophosphoribosyltransferase is readily inactivated by oxygen and it has been suggested that there are factors in cells which inhibit or regulate this inactivation process. There is also evidence that cells with the most rapid rates of growth contain the most amidophosphoribosyltransferase protein (6). In summary, this crucial enzyme appears to be subject to at least three types of regulation which depend on: (i) the relative concentration of PP-ribose-P purine nucleotides in the cell affecting the quaternary structure; (ii) the oxygen-inactivation phenomenon and the effects of possible intracellular protective factors; (iii) alteration in the rate of protein synthesis. The oxygen inactivation phenomenon may initiate amidophosphoribosyltransferase degradation (7) so that the amount of amidophosphoribosyltransferase protein present and available for regulation by the relative concentrations of PP-ribose-P and purine nucleotides depends on a balance between mechanisms (ii) and (iii). However, the mechanism which initiates new

36

RICHARD W. E. WATTS

amidophosphoribosyltransferase protein synthesis is not known. The biochemistry of amidophosphoribosyltransferase was fully reviewed in a recent volume in this series (8). The concentration of P P - r i b o s e - P itself depends on a balance between synthesis and utilization, and an increased intracellular concentration of P P - r i b o s e - P can be seen as driving purine d e n o v o synthesis forwards. This has been inferred at the level of the intact human subject on the basis of correlations between erythrocyte levels of P P - r i b o s e - P and urate production as well as in isolated cell and tissue systems. Thus, patients with an overactive P P - r i b o s e - P synthetase variant have an excessive rate of purine de n o v o synthesis (9) as do those with reduced H P R T activity and thereby reduced utilization of P P - r i b o s e - P for purine salvage. Apart from its role as an allosteric regulator of amidophosphoribosyltransferase and as a substrate for this enzyme, P P - r i b o s e - P participates in several other biochemical reactions so that the activity of P P - r i b o s e - P synthetase cannot be the rate-limiting step on the purine d e n o v o synthesis pathway. However, the rate of P P - r i b o s e - P production must be finely tuned to avoid over-stimulation of the pathway. The reported modulators of the activity of P P - r i b o s e - P synthetase include purine nucleotides, 2,3-diphosphoglycerate, PP-ribose-P, orthophosphate and magnesium ions. The enzymes which catalyze the reactions of the purine d e n o v o synthesis pathway are located in the cytosol fraction of the cell and the extent to which they may comprise a macromolecular complex has been an open question. Rowe (10) investigated de n o v o purine synthesis in lectin stimulated human peripheral blood lymphocytes. He copurified the enzymes of the purine de n o v o synthesis pathway and found that the peak of activity corresponded to a molecular weight of about 180,000 on sucrose density gradient separation, and SDS polyacrylamide gel electrophoresis revealed no protein of molecular weight greater than 80,000. He concluded that the enzymes do not constitute a macromolecular complex. The overall de n o v o synthesis rate was two orders of magnitude lower in the human lymphocyte system than in the avian liver. This is not surprising in view of the bird's use of uric acid as a major route for nitrogen excretion. He reported that the formylglycinamide amidotransferase (phosphoribosylformylglycinamidine synthetase; EC 6.3.5.3) catalyzed reaction was a rate-limiting step on the pathway. This enzyme is well known to be much more sensitive than amidophosphoribosyltransferase to inhibition by glutamine antagonists such as azaserine but it has not been previously proposed as a rate-limiting step, and its significance as a possible regulatory site remains to be assessed. Ammonium ions and glutamine were both used for phosphoribosylamine synthesis in the lectin stimulated lymphocyte system, and this is consistent with the evidence that amidophosphoribosyltransferase is the only enzyme catalyzing PRA synthesis (8). The anticipated inhibition of purine de n o v o synthesis by AMP and G M P was demonstrated in the

37

PURINE NUCLEOTIDE BIOSYNTHESIS

system irrespective of whether glutamine or ammonium was a nitrogen donor. The enzyme system which Rowe (10) studied also contained the purine nucleotide interconversion enzymes because the synthesis of AM P and G M P was effected when appropriate cofactors were added. Some aspects of the regulation of the purine nucleotide interconversion reactions were also demonstrable in this system in that the production of A M P from IMP was shown to be self-regulatory and G T P appeared to be the most powerful inhibitor of GM P synthesis from IMP. However, the enzymes catalyzing purine nucleotide degradation were not co-purified with those of the purine de novo synthesis pathway. Purine Interconversion, Degradation and Salvage The reactions whereby inosinic acid is converted to A M P and G M P and the purine nucleoside and bases are interconverted, degraded and salvaged are well known (Figure 3). The reactions leading from IMP to A M P and G M P are subject to complex regulations which arise from the cofactor requirements of the reactions and from product inhibition.

I

Biosynthesis de

l

lnosinic acid

novo ] 2

I

~ [Adenylosuccinate] 4

I

/'/',,TI,I ,.rl:

' Inosine

/7

11'/,

Hypoxanthine

lanthosine / 1

11'J

, Xanthine ~ to

19

G-uanosine

11'

Guanine )

3

,.TI: 1

- - Adenosine is

11' J

Adenine)

Uric acid Allantoin

FIG. 3. Interconversion, conservation and degradation pathways of the purine nucleotides, nucleosides and bases. Key to enzymes: I, IMP dehydrogenase (EC 1.2.1.14); 2, adenylosuccinate synthetase (EC 6.3.4.4); 3, adenylosuccinate lyase (EC 4.3.2.2); 4, A M P deaminase (EC 3.5.4.6); 5, 5'--nucleotidase (EC 3.1.3.5); 6, purine nucleoside phosphorylase (EC 2.4.2. I); 7, hypoxanthine phosphoribosyltransferase (EC 2.4.2.8); 8, adenine phosphoribosyltransferase (EC 2.4.2.7); 9, xanthine oxidase (ECI.2.3.2); I0, guanine deaminase (EC3.5.4.3); 11, urate oxidase (EC 1.7.3.3) (absent from human tissues); 12, adenosine deaminase (EC 3.5.4.4); 13, inosine kinase (EC 2.7.1.73); 14, adenosine kinase (EC 2.7.1.20) [adenine deaminase (EC 3.5.4.2), which catalyzes the deamination of adenine to hypoxanthine, occurs in some animal tissues but not in man]; 15, unclassified nucleoside kinase; 16, GM P synthetase (EC 6.3.4.1).

38

RICHARD W. E. WATTS

Adenylosuccinate synthetase (EC 6.3.4.4) and hence AMP synthetase requires GTP as an energy source and G M P synthetase requires ATP as an energy source. The supply of triphosphate depends on the supply of the monophosphate so this offers one mechanism whereby the production of G M P can be balanced by the corresponding AM P production and vice versa. The activity of I M P dehyd rogenase (EC 1.2.1.14) is inhibited by A M P and by GMP, the inhibition being competitive with'respect to IMP. It is also inhibited competitively by NADH with respect to both IMP and NAD ÷ (11, 12). Ullman and his collaborators (13, 14) suggested, on the basis of studies in mutant lymphoblasts in which the adenylosuccinate synthetase activity was reduced by about 80%, that adenylosuccinate synthetase deficiency might be associated with increased purine de n o v o synthesis. Willis and Seegmiller (15) studied lymphoblasts which had not been selected for an enzyme deficiency and H P R T deficient cells, and found that inhibiting either adenylosuccinate synthetase with alanosine or IMP dehydrogenase with mycophenolic acid increased the excretion of purines into the medium. They also reported accelerated de n o v o synthesis of the purine excreted into the medium as judged by [~4C] formate incorporation. This was accompanied by a decreased rate of purine synthesis for the purines remaining in the cells, but the stimulating effect on the rate of synthesis of the extruded purines was relatively large so that the overall effect was one of stimulation. The results of similar studies on liver tissue were somewhat different (Allsop and Watts, unpublished). The rate of purine de n o v o synthesis was measured in liver tissue in vitro using a [~4C] formate incorporation method, and the effect of either adenylosuccinate synthetase inhibition or IMP dehydrogenase inhibition on this parameter was studied both with respect to the purines that are retained in the tissue, and with respect to those which enter the medium. Both adenylosuccinate synthetase and IMP dehydrogenase inhibition slowed de n o v o purine synthesis with respect to the purines remaining in the tissue but did not affect the de n o v o synthesis rate of the purines secreted into the incubation medium (Table 1). Control experiments in which the distribution of isotope between the adenine and guanine nucleotides was studied showed that the redistribution of isotope expected from inhibiting either adenylosuccinate synthetase or IMP dehydrogenase occurred (Table 2). Experiments in which a sufficient concentration of azaserine was added to the system to inhibit phosphoribosylformylglycinamidine synthetase but not amidophosphoribosyltransferase confirmed that the method used was measuring purine de n o v o synthesis. The difference between the effect of inhibiting these enzymes in lymphoblasts and liver tissue may be related to the fact that the former are actively dividing cells whereas the liver cells are not. It has been argued that in the lymphoblast system the reduced level of adenine nucleotide production

39

PURINE NUCLEOTIDE BIOSYNTHESIS

TABLE 1. THE EFFECTS OF TIlE ADENYLOSUCCINATE SYNTHETASE INHIBITOR ALANOSINE (L-2-AMINO-3--HYDROXYNITROAMIDOPROPANOIC ACID) AND OF THE IMP DEHYDROGENASE INHIBITOR VIRAZOLE (I-~-R!BOFURANOSYL-I-H, 1,2,4--TRIAZOLE-3-CARBOXAMIDE) ON THE RATE OF DE N O V O SYNTHESIS OF THE PURINES RETAINED IN THE LIVER TISSUE A N D OF THOSE WHICH ARE TRANSLOCATED INTO THE INCUBATION MEDIUM Rate of purine de novo synthesis (pmol/h/mg liver protein)

System

Tissue Control Virazole (10/~M) Virazole (100/~M) Alanosine (10 #M) Alanosine (100 ~M)

6.99 + 7.40 + 3.40 + 6.41 + 4.20 ±

0.81 (72; 12) 0.82 (48; 8) 0.28 (24; 4) 0.68 (48; 8) 0.40 (24; 4)

Incubation medium 10.22 ± 11.04 + 12.20 ± 9.38 + 9.20 ±

0.64 1.09 1.09 0.60 0.72

Approximately 70 mg tissue were weighed into a glass Bijou bottle with 4 ml ice cold Krebs-Ringer phosphate (without Ca2+). Na [14C]formate (10#Ci; 100 ~1) was added, each bottle was gassed with oxygen for 30 sec and sealed. The samples were incubated at 37° for 2 hr with shaking. After incubation, the samples were chilled on ice and immediately centrifuged (1000 × g at 4° for 10 rain); the supernatant was retained, the tissue transferred to graduated 10 ml conical glass centrifuge tubes, washed 5 times in ice cold sodium chloride solution ( 154 mM) with centrifugation (I 000 × g, 4° , 5 rain) between each washing. The final pellet was drained, frozen in liquid nitrogen and stored at -20 ° . Water (0.5 ml) was added to each sample, which was allowed to thaw, sonicated for 10 sec at amplitude 8/z. A portion (10/~1) was transferred into sodium hydroxide (20m~, 0.5 ml) for protein determination. HC10~ (2 M, 0.5 ml) with [~H]AMP (0.2/~Ci; 10 ~1) was added to each tube and left on ice for 1 hr, then centrifuged ( 100 × g, at 4° for 5 min), and the supernatant heated at 100° for I hr. The vol was adjusted to 1.5 ml with water and the tubes put on ice. A solution of adenine (700 #g, 10 ~1) was added to each tube. The pH was adjusted to 11 with ammonia (SG 0.880), AgNO3 solution (20% w/v) added with frequent mixing until a precipitate just persisted, and the tubes left on ice for I hr. The precipitate was collected by centrifugation, washed twice with water, drained, suspended in HCI (100 mM, 0.5 ml) and heated for 1 hr at 100% The tubes were centrifuged (1000 × g, at 4° for 5 min) and the vol of supernatant measured. Portions (200/~1) were used for J4C measurements by liquid scintillation spectrometry. The incorporation of [14C]formate into the purines in the incubation medium was measured by the same method. The inhibitors were added as 1.0 ml of a solution to the Krebs-Ringer phosphate solution (3.0 ml). Virazole was dissolved in Krebs-Ringer phosphate solution, and alanosine was dissolved in Tris (1% w/v). The values are means + standard error. The numbers in parentheses are: first, the number of tissue samples analyzed and, second, the number of rats from which they were derived.

w h e n a d e n y l o s u c c i n a t e s y n t h e t a s e is i n h i b i t e d r e d u c e s f e e d b a c k i n h i b i t i o n o n a m i d o p h o s p h o r i b o s y l t r a n s f e r a s e a n d t h a t this l e a d s t o a n i n c r e a s e d r a t e o f d e p u r i n e s y n t h e s i s in a n a t t e m p t t o i n c r e a s e I M P s y n t h e s i s a n d h e n c e AMP p r o d u c t i o n ; s i m i l a r l y , it h a s b e e n p r o p o s e d t h a t w h e n I M P d e h y d r o g e n a s e is i n h i b i t e d t h e d e c r e a s e d a v a i l a b i l i t y o f G M P r e d u c e s t h e f e e d b a c k i n h i b i t i o n n o r m a l l y p r o d u c e d b y this n u c l e o t i d e so t h a t p u r i n e s y n t h e s i s is i n c r e a s e d . T h e r e s u l t s in t h e c a s e o f p u r i n e s r e m a i n i n g in t h e liver novo

cell c o u l d be e x p l a i n e d o n t h e h y p o t h e s i s t h a t t h e n u c l e o t i d e s t h e n a c c u m u l a t e to a d i s p r o p o r t i o n a t e e x t e n t because of the m e t a b o l i c block; g u a n i n e

40

RICHARD W. E. WATTS

TABLE 2. EFFECT OF THE ADENYLOSUCCINATE SYNTHETASE INHIBITOR ALANOS1NE AND OF THE IMP DEHYDROGENASE INHIBITOR VIRAZOLE ON THE DISTRIBUTION OF [14C]FORMATE BETWEEN THE PURINE BASES SEPARATED BY CHROMATOGRAPHY ON PLASTIC BACKED CELLULOSE-300 THIN LAYERS WITH BUTAN-I-OL:METHANOL:WATER:AMMON1A(25% w/v) (60:20:20:1) [14C]formate incorporation % distribution

System Adenine No inhibitors + Alanosine (100 ~M) + Virazole (100/~M)

51 3 86

Hypoxanthine 3 34 10

Guanine 46 63 4

nucleotides when adenylosuccinate synthetase is inhibited and adenine nucleotides when I M P dehydrogenase is inhibited actually increase feedback inhibition on amidophosphoribosyltransferase. The de novo synthesis rate of purines or purine derivatives which were extruded into the incubation medium appears to be independent of these influences. The differential effects of the inhibitors on the rates of synthesis of the purines being extruded from the liver cells, and those which were not, suggest that purine de novo synthesis in the liver may be functionally compartmentalized so that the portion of the total synthesis which is destined for export from the cell is not subject to the same regulatory influences as that which leads to the production of purine nucleotides for the tissue's own internal use, and this might apply in vivo. Evidence has been presented for the compartmentalization of uridine nucleotide pools in rat hepatoma cells (16). These interpretations may imply that adenine and guanine nucleotides are equally effective modulators of amidophosphoribosyltransferase activity and therefore of the rate of purine de novo synthesis. Comparison of the effects of H P R T and adenine phosphoribosyltransferase (APRT; EC 2.4.2.7) deficiencies suggests that this may not be so at least in vivo. Thus, in H P R T deficiency where guanine salvage is deficient, purine de novo synthesis is accelerated whereas it is not in A P R T deficiency where adenine salvage is deficient. The Interconversion o f Ribo- and Deoxyribonucleotides

The ribose nucleotides are converted to deoxyribonucleotides at the nucleoside diphosphate level and this is catalyzed by ribonucleoside diphosphate reductase (ribonucleotide reductase; EC1.17.4.1). The deoxyribonucleoside diphosphates are catalytically phosphorylated to the deoxynucleoside triphosphates which are the substrates for D N A nucleotidyltransferase ( D N A polymerase; EC 2.7.7.7). The mechanism and regula-

PURINE NUCLEOTIDEBIOSYNTHESIS

41

tion of this enzyme were fully reviewed by Thelander and Reichard (l 7). In summary: deoxynucleotide synthesis begins with the reduction of CDP to dCDP and UDP to dUDP by an ATP-activated enzyme, proceeds to GDP reduction via a dTTP-dependent enzyme and finally reaches ADP reduction by a dGTP-activated enzyme. Accumulation of dATP inhibits the whole reaction sequence. Accumulation of dT T P shuts off the reduction of CDP and UDP and accumulation ofdGTP turns off GDP reduction as well as CDP and UDP reduction. This mechanism provides a balanced supply ofdCTP, dUDP (and hence dTTP), dGTP and dATP for DNA synthesis; it also explains the biochemical pathology of ADA and PNP deficiencies. In the former, deoxyadenosine accumulates and is converted to dATP by the (deoxy)nucleoside and (deoxy)nucleotide kinases which are present in large amounts in lymphoid tissue. The dATP produced inhibits all aspects of ribonucleotide diphosphate reductase action, and the cells are unable to respond to infection. In PNP deficiency, deoxyguanosine accumulates and is converted to dGTP by the actions of (deoxy)guanosine and (deoxy)guanylate kinases which are present in relatively large concentrations in T-lymphocytes, The dGTP inhibits the production of dCTP and dTTP so that DNA production in response to an immunogenic challenge fails because of lack of pyrimidine deoxyribonucleotides. ADA deficiency affects both B- and T-cell function, and PNP deficiency affects only the function of T-cells. The reason for this particular specificity is still not completely understood.

The s-Adenosylhomocysteine Pathway s-Adenosylmethionine is the labile methyl group donor for l-carbon transfer reactions including the methylation of d U M P to dTM P. The reaction product s-adenosylhomocysteine is normally hydrolyzed to homocysteine under the catalytic influence of s-adenosylhomocysteine hydrolase (EC 3.3.1.1) as part of the cyclic mechanism shown in Figure 4. Adenosine and deoxyadenosine inhibit this enzyme, but deoxyadenosine is the much more powerful inhibitor, and the irreversible binding of deoxyadenosine with consequent failure to metabolize s-adenosylhoinocysteine, which in turn inhibits the methylation of dUMP, has been proposed as an alternative explanation for the immunoparesis in ADA deficiency (18-20). s--adenosylhomocysteine hydrolase is also inhibited by inosine, and this has been proposed as the basis of the impaired T-cell function in purine nucleoside phosphorylase deficiency (21). The reduced activity of s--adenosylmethionine hydrolase in HPRT deficiency is attributed to the potentiating effect of hypoxanthine on the inosine inhibition of this enzyme. This might be related to the small degree of B cell dysfunction which some workers have found in HPRT deficiency. It is also of interest that inosine accumulates in the plasma

42

RICHARD W. E. WATTS Ornithine

~f

-'- ATP ~

C02

"

4,

S-Adenosylmethionine (SAM) T

Decarboxylated SAM ~

Putrescine N'-~l~yt~rmidine

]

S-hdenosylhomc~ysteine (SAIl) Homoc Adenosine(Ado)

Ado kina.~

Inosine

~

Cysteine

PNP HPRT

HypOxanthine

[

II, AMP•

ImIMF -

I

Adenylosuccinate

-

FIG. 4. The s-adenosylhomocysteine pathway and its linkage to the synthesis and interconversion of putrescine and spermidine by the participation of decarboxylated s--adenosylmethionine in the spermidine synthetase (EC 2.5.1.16) reaction. The dotted lines indicate the stimulation of s--adenosylmethionine decarboxylation by putrescine and the reduction in ornithine decarboxylase activity brought about by putrescine. SAH = s-adenosylhomocysteine. SAH hydrolase (EC 3.3.1. I) is inhibited by deoxyadenosine even more powerfully than by adenosine. Adokinase, adenosine kinase (EC 2.7.1.20); ADA, adenosine deaminase (EC3.5.4.4); PNP, purine nucleoside phosphorylase (EC2.4.2.1); HPRT, hypoxanthine phosphoribosyltransferase (EC 2.4.2.8).

of patients with severe renal failure and there is also a poorly defined immunodeficiency in this condition. S-Adenosylmethionine is also a key intermediate in the production of the polyamines, spermidine and putrescine (Figure 4), PURINE

SYNTHESIS AND SALVAGE SPECIAL SITUATIONS

DE NOVO

IN S O M E

The concept of the liver as a major site of purine d e n o v o biosynthesis for export to other organs has had wide credence (22). It has also been suggested that some tissues have an inherently limited purine d e n o v o synthesis capacity and that this may be insufficient to meet their needs at certain stages of development so that, under these conditions, purine salvage becomes particularly important. Purine d e n o v o synthesis has been reported to be relatively inefficient in bone marrow (23) and brain (24) and these organs have therefore been regarded as being particularly dependent on purine phosphoribosyltransferase catalyzed purine salvage. The problem of assessing the relative activities of the purine de n o v o synthesis pathway and the purine salvage pathway in different tissues was approached by measuring the activities of amidophosphoribosyltransferase, which is the rate-limiting enzyme on the purine de n o v o synthesis pathway and

43

PURINE NUCLEOTIDE BIOSYNTHESIS

of hypoxanthine phosphoribosyltransferase, respectively, and calculating the ratio of the two enzyme activities (25). This is theoretically imperfect, because there is no proof that measurements of enzyme activity under saturating substrate conditions in vitro reflects the rates at which metabolites flow along the individual pathways in vivo. However, it was shown experimentally that the age-related changes in purine de novo synthesis as measured by the incorporation of [~4C]-formate into the cellular purines of brain tissue were similar to age-related changes of amidophosphoribosyltransferase activity and very different from those of hypoxanthine phosphoribosyltransferase activity (Figure 5). The results in Table 3 show that purine de novo synthesis activity is widely distributed and that there is only an approximately 5-fold difference between the lowest (testis) and highest (liver) values observed. There is no support for the view that the brain has particularly low purine de novo synthetic activity. The activities of APRT and H P R T did not parallel one another. Purine de novo synthesis is more active in skeletal muscle than in liver (26). This could be a major component of the total purine de novo synthesis in vivo, but the extent to which muscle might manufacture purines for export to other tissues has not been assessed. Purine Synthesis and Salvage at Different Stages o f Brain Development

Interest in the possible changes in the relative activities of the purine de novo synthesis and purine salvage in the brain during development arose from the observation that deficient activity of H P R T is associated with a characteristic syndrome of neurological dysfunction in the Lesch-Nyhan syndrome. The reportedly limited capacity of the central nervous system to synthesize purine de novo (23, 24) suggested that if H P R T is deficient the

',-7

600.

"~

500.

,

4o0.

Birth

-60 7 -50 -30

300_

200-

E

-~

-6

o-

# I !

#

-io 8._~ -o ~

Days

FIG. 5. Comparison of the rat.- of purine de n o v o synthesis (A) measured by the incorporation of [14C] formate into the purines of brain tissue (as in Table I except that 100 mg oftissuewas used) in vitro with the activities of amidophosphoribosyltransferase (n) and hypoxanthine phosphoribosyltransferase (O) at different stages in the development of the rat brain [see (25) for the enzyme assays].

aNot analyzed.

Central nervous system cerebral cortex basal ganglia cerebellum medulla/pons spinal cord Liver Spleen Kidney Testis

Tissue

101 5:6.67 134 5:6.24 126 5:9.06 116 5:5.16 148 5:9.70 26 5:3.01 203 5:32.9 42 5:2.9 82 4- 6.2

APRT

(14) (14) (14) (14) (12) (10) (10) (10) (10)

374 5:26.3 417 5:33.8 422 + 20.1 361 5:23.2 247 5:18.2 373 5:21.2 306 5:13.2 62-1- 4.9 78 5:2.5

HPRT

(14) (14) (14) (14) (14) (10) (10) (10) (10)

(16) (16) (16) (16) (16) (9) (9) (9) (9)

PRPP-At 11.5 5:1.42 12.7 + 1.42 9.9 + 1.24 9.1 5:1.3 6.0 + 0.8 21.1 ± 1.1 6.3 5:0.4 13.3 5:0.7 3.9 5:1.3

Enzyme activity ( n m o l / h r / m g protein) [mean + SEM (no. animals)]

33 5:3.4 33 5:3.1 45 5:4.8 40 5:5.0 40 5:6.2 18 5:0.9 49 + 4.91 5 5:0.34 20 5:5.86

PRPP-At

Ratio H P RT

5.5 5.36 4.69 1.37 3.04

:J: 0.41 4- 0.73 ± 0.29 ± 0.34 5:0.67 n.a. a n.a. n.a. n.a.

(14) (14) (14) (14) (14)

(nmol/mg protein)

PRPP content

TABLE 3. ADENINE P H O S P H O R I B O S Y L T R A N S F E R A S E (APRT), HYPOXANTHINE PHOSPHORIBOSYLTRANSFERASE (HPRT) AND A M 1 D O P H O S P H O R I B O S Y L T R A N S F E R A S E ACTIVITIES AND THE P H O S P H O R I B O S Y L P Y R O P H O S P H A T E ( P P - R I B O S E - P ) CONTENTS OF THE CENTRAL NERVOUS SYSTEM AND SOME OTHER TISSUES O F 8 - W E E K - O L D RATS (25)

>

>

PURINE NUCLEOTIDE BIOSYNTHESIS

45

developing nervous system might be subject to an especial constraint, when the rates of cell replication are highest and this might restrict brain development in a characteristic way. The observation that granulocyte/ macrophage progenitor cells of bone marrow from patients with H P R T deficiency had less proliferative potential in vitro than the corresponding normal cells (27) gave this idea some general support. It was also shown that although H P R T deficient lymphocytes synthesized DNA, RNA and protein normally in response to phytohemagglutinin stimulation, these processes were impaired if the mitogenic stimulus was applied to lymphocytes in which purine de n o v o synthesis was also inhibited by selective azaserine inhibition of phosphoribosylformylglycinamidine synthetase (28). These results appeared to support the concept that cell division in brain would be impaired by hypoxanthine phosphoribosyltransferase deficiency. A study of the relationship between age and the levels of activity of the phosphoribosyltransferases and amidophosphoribosyltransferase in the rat showed that the activities of amidophosphoribosyltransferase and A P R T remain steady or decrease slightly between birth and maturity, but that H P R T activity increased (Figure 6). Plotting the ratio of H P R T activity to amidophosphoribosyltransferase activity with respect to age and comparing this with the time course of the rates of brain cell accretion extrapolated from human data shows that the highest values for this ratio were after the main Cerebral

E

cortex

Cer~llum

o

f

Medulla/Pons

-1

i

t

t

i

i

I T

I

I

I

I

l

i

l

12345678

Spinal cord

Age

l

( weeks )

200

100 i i r I I i 1 2 3 4 5 6 7 8

t

I

I

l

=

w

I

t

i

t

1 2 3 4 5 6 7 8 Age ( weeks )

FIG. 6. Amidophosphoribosyltransferase(e), hypoxanthinephosphoribosyltransferase(O) and phosphoribosyltransferase(Q) activities of rat central nervous system in rats age I-8 weeks (25).

adenine

46

R I C H A R D W. E. W A T T S

bursts of neuroglia and neuroblast proliferation (Figure 7). Contrary to the previous suggestions, this indicates that purine de n o v o synthesis meets the needs of growth and differentiation but that purine salvage becomes important later possibly in relation to a neuropharmacological function (25). The evidence that cyclic nucleotides may be post-synaptic neurotransmitters (29, 30) led to attempts to correlate the changes in purine d e n o v o synthesis and H P R T activities with alterations in cAMP and cGMP content of the brain. The results (Figure 8) show that although cGMP tends to parallel the total purine synthesis rate after about the seventh day of postnatal life, the enzyme levels do not correlate closely with the levels of these second messengers in the brain and in particular the time courses for the cGMP level and H P R T activity are entirely different. This suggests that the effects of H P R T deficiency are not mediated directly through an effect on cGMP level although it is impossible to test this directly in the HPRT-deficient subject's brain. The general conclusion that purine de n o v o synthesis appears to be sufficient to meet the need of those phases of brain growth when neuroblast and neuroglia proliferation are occurring most rapidly, at least in the rat, is compatible with the normal morphology of the human brain in H P R T deficiency and with its normal level of amidophosphoribosyltransferase activity (31).

Rat

10-

0~-

~

o

i wl,

4

i i,

8

iT,

l

12 16 20

~

1

,

3

I I,II

5

7

*--- Gestation (days) : --

9

,I

~

I I

//---1

11 13 15 17 19 21

56

Postnatal age I days )

.

u~

-I Human

Birth

I

I

I

0 t

1

2 3 4 5 6 7 Gestation (months)

I

I

t

I

I

I

I

_ Brain cell (DNA) accretion rate

I

I

I

I

I

!

I

I

I

8 9 1 2 3 4 5 6 7 8 9 -~: Postnatal age (months)

I/---7

24 .

FIG. 7. Comparison between the enzyme activity ratio [(hypoxanthine phosphoribosyltransferase)/(amidophosphoribosyltransferase)] for rat brain and the brain cell accretion rate in m a n with an estimate of the corresponding developmental stage in the rat. Whole brain (*), cerebral cortex (O), basal ganglia (D), cerebellum (A), medulla/pons (<>).

PURINE NUCLEOTIDE BIOSYNTHESIS

47 -4

12oloo--2

-1 ~ T .E

o"

I

7

II

I

I

|

I

Birth

-0 -60

c

-50

4110-

-dO -30 -20

E

~

c:

o-

i

I I

I

w

-10 -0

Days

FIG. 8. The cAM P (e) and cGM P (11)concentrations in rat brain together with the corresponding activities of hypoxanthine phosphoribosyltransferase (O), amidophosphoribosyltransferase (13) and total purine synthesis measured by [14C]formate incorporation into the tissue purines (A). (The cyclic nucleotide determinations were made with assay kits supplied by the Radiochemical Centre, Amersham, Bucks., U.K. The rats were decapitated, the brain freeze clamped in liquid nitrogen, homogenized in EDTA (4 raM, 1 ml), the homogenate heated to 100° , centrifuged and the supernatant used for the assay).

Purine Synthesis and Salvage in Mitogen Stimulated Lymphocytes The initiation o f purine de novo synthesis early in cell t r a n s f o r m a t i o n involves an event at the cell surface and a n unidentified m e c h a n i s m which t r a n s m i t s the i n f o r m a t i o n t h a t the cell is being e x p o s e d to an antigen, o r to a mitogen, a n d will need to t r a n s f o r m . L y m p h o c y t e s m u s t be e x p o s e d to a m i t o g e n for a b o u t 12 hr in o r d e r to p r o d u c e t r a n s f o r m a t i o n . R N A synthesis begins after a b o u t 5 hr a n d D N A synthesis after a b o u t 24 hr, but there are visible changes in the cell surface, referred to as c a p p i n g , within a few min o f e x p o s u r e to the mitogen. The cap is f o r m e d by the m i g r a t i o n of r e c e p t o r - b o u n d ligand (mitogen) particles to one pole o f the cell (32). The cap is shed f r o m the surface o f the cell, a l t h o u g h some m a y be internalized. This leaves the cell surface d e n u d e d o f ligand b i n d i n g c o m p o n e n t s for 3-8 hr. C a p p i n g is an energy requiring process which d e p e n d s on the density o f the receptors, and on the ' s p a n ' a n d frequency of r e i t e r a t i o n of the b i n d i n g sites o f the ligand (33). Evidence has been a d v a n c e d for the

48

RICHARD W. E. WATTS

involvement of specific proteases in the earliest stages of mitogenesis, for an inward migration of calcium ions, for the activation of membrane ATPases, for alteration in membrane phospholipid metabolism and for activation of glucose and amino acid transport. However, it is unclear how these different phenomena fit together to form an ordered sequence, or how the operation of the sequence is triggered. It has been suggested (34) that the inward migration of calcium ions activates guanylate cyclase (EC 4.6.1.2), producing a pulse of c G M P which activates P R P P synthetase leading to a pulse of P P - r i b o s e - P and that this finally initiates de n o v o purine synthesis by the allosteric activation ofamidophosphoribosyltransferase. The most recent work (35) has failed to confirm the previous observation (36) of a short pulse of PP-ribose-P production during the first hr after phytohaemagglutinin stimulation. The P P - r i b o s e - P concentrations as well as the activities of PP-ribose-P synthetase and amidophosphoribosyltransferase were unaltered during the whole period from 0-96 hr after phytohaemagglutinin stimulation. These enzyme assays were performed under saturating substrate conditions so that allosteric activation at low substrate concentrations would have been obscured. However, the failure to confirm the very early peak of PP-ribose-P production calls into question any theory which postulates this compound as a metabolic signal between the application of a stimulus to the cell surface and the initiation of transformation. Therefore, the nature of the second messenger in lymphocyte transformation remains obscure. As in other contexts, c A M P has been proposed, and changes in the ratio of c A M P to c G M P may be more important than the absolute level of either cyclic nucleotide. Changes in the ratio of ATP to ADP concentrations also occur, but whether they precede or follow a burst of purine de n o v o synthesis is not known. Recent studies (37) have shown that the transformation of lymphocytes, in which adenosine deaminase is blocked by 2'-deoxycoformycin, is maximally inhibited by deoxyadenosine when the deoxyribonucleoside is added to the cell culture simultaneously with the mitogen (Con A). Its effect diminishes with time after the application of the mitogen and it has no effect if applied 12 hr or more later. Ribonucleotide reductase activity is most important during the s-phase of the cell cycle when DNA synthesis is occurring most rapidly. Therefore, deoxyadenosine appears to interfere with the transformation process at a very early stage, either, rather than, or as well as, during the s-phase. This could reflect the inhibitory action of deoxyadenosine on s-adenosylhomocysteine hydrolase and indicate that either a transmethylation reaction or spermidine synthesis is necessary at an early stage of transformation. Alternatively, deoxyadenosine might be converted to cyclic dATP and this could interfere with either a second messenger function of c A M P or the cyclic nucleotide ratio.

PURINE NUCLEOTIDEBIOSYNTHESIS

49

SUMMARY The regulation and integration of purine nucleotide biosynthesis is considered from the viewpoint of the main groups of reaction sequences involved and with respect to some specific organs and tissues. Inhibiting either I M P dehydrogenase or adenylosuccinate synthetase in rat liver in vitro reduced the rate of purine d o n o v o synthesis with respect to the purine remaining in the tissue and did not materially affect the rate with respect to the purines extruded into the incubation medium. These results are considered in contrast to the results of previous studies in cultured lymphoblasts. The relative activities of purine de n o v o synthesis and of purine salvage have been assessed in different tissues by the activities of amidophosphoribosyltransferase and hypoxanthine phosphoribosyltransferase (HPRT), respectively. Changes in purine de n o v o synthesis as measured by [~4C]formate incorporation into cellular purines were reflected in the amidophosphoribosyltransferase activities. The capacity of different tissues to synthesize purines d e n o v o is widespread and the role of the liver as the main site of purine de n o v o synthesis in v i v o and exporting purines to other tissues appears questionable. Regulatory mechanisms may well be tissue specific. The age-related changes in the activity of the purine d e n o v o synthesis and purine salvage pathways, respectively, in the brain suggest that it is physiological or neuropharmacological functions of the developed brain rather than cell division and organogenesis which require a high level of purine salvage relative to purine de n o v o synthesis. This is compatible with the observation that purine d e n o v o synthesis alone can meet the needs for additional purine nucleotides which lectin induced lymphocyte transformation involves. The mechanism whereby purine d e n o v o synthesis is initiated during lectin induced lymphoblast transformation remains obscure.

I. 2.

3, 4. 5. 6.

REFERENCES J. P. MAPES and H. A. KREBS, Rate-limiting factors in urate synthesis and gluconeogenesis in avian liver, Biochem. J. 172, 193-203 (1978). E. W. HOLMES, J. B. WYNGAARDENand W. N. KELLEY, Human glutamine phosphoribosylpyrophosphate and amidotransferase: Two molecular forms interconvertible by purine ribonucleotides and phosphoribosylpyrophosphate,J. Biol. Chem, 248, 6035-6040 (1973). G. KING, J. C. MEADE, G. C. BOUNOUS and E. W. HOLMES, Demonstration of ammonia utilization for purine biosynthesis by the intact cell and characterization of the enzymatic activity catalyzing this reaction, Metabolism 28, 348-357 (1979). M. ITAKURA, T. SABINA, P. HEALD and E. W. HOLMES, Basis for the control of purine biosynthesis by purine nucleotides, J. Clin. Investig. 67, 994-1002 (1981). A. UDOM and E, W. HOLMES, Purification and characterization of human amidophosphoribosyltransferase, Z. Kiln. Chem. Kiln. Biochem. 20, 428 (1982), M. TSUDA, N. KATUNUMA, H. P. MORRISand G. WEBER, Purification, properties and immunotitration of hepatoma glutamine phosphoribosylpyrophosphateamidotransferase, Cancer Res. 39, 305-311 (1979).

50 7.

RICHARD W. E. WATTS R.L. SWITZER, Selective inactivation and degradation of enzymes during sporulation, Federation Proc. 38, 475 (1979).

8. 9. 10. 11. 12. 13.

14.

15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25.

26. 27.

28.

E. W. HOLMES, Kinetic, physical and regulatory properties of amidophosphoribosyltransferase, Advances in Enzyme Regulation 19, 215-231 ( 1981). M.A. BECKER, K. O. RAIVIO, B. BAKAY, W. B. ADAMS and W. L. NYHAN, Variant human phosphoribosylpyrophsphate synthetase altered in regulatory and catalytic functions, J. Clin. lnvestig. 65, 109-120 (1980). P.B. ROWE, E. McCAIRNS, D. SAUER and D. FAHEY, De novo purine synthesis in human lymphocytes, Z. Klin. Chem. Klin. Biochem. 20, 411-412 (1982). E.W. HOLMES, M. D. PEHLKE, A. LEYVA and W. N. KELLEY, Human inosinic dehydrogenase: regulatory properties in vitro and in vivo, Israel J. Med. Sci. 9, 21-22 (1973). R.C. JACKSON, G. WEBER and H. P. MORRIS, IMP dehydrogenase, an enzyme linked with proliteration and malignancy, Nature 256, 331-333 (1975). B. U L L M A N , M . A . W O R M S T E D , B.B. L E V I N S O N , L.J. G U D A S , A. C O H E N , S . M .

CLIFT and D. W. MARTIN, JR., Abnormal regulation of purine metabolism in a cultured mouse T-cell lymphoma mutant partially deficient in adenylosuccinate synthetase, Advanc. Exptl. Med. Biol. 122A, 375-386 (1980). D.W. MARTIN, JR., B. ULLMAN, M. A. WORMSTED and M. B. COHEN, Purine oversecretion in cultured murine lymphoma cells deficient in adenylosuccinate synthetase: a genetic model for inherited hyperuricaemia and gout, Z. Klin. Chem. Klin. Biochem. 20, 394 (1982). R.C. WILLIS and J. E. SEEGMILLER, lncreasesin purine excretion and rateofsynthesis by drugs inhibiting IMP dehydrogenase and adenylosuccinate activities, Advanc. Exptl. Med. BioL 122B, 237-241 (1980). M. J. LOSMAN and E. H. HARLEY, Evidence for compartmentation of uridine nucleotide pools in rat hepatoma cells, Biochim. Biophys. Acta 521,762-769 (1978). L. THELANDER and P. REICHARD, Reduction ofribonucleotides, Ann. Rev. Biochem. 48, 133-158 (1979). N. M. KRED1TCH, The methylation hypothesis of adenosine toxicity, pp. 153-156 in Enzyme Defects and I m m u n e Dysfunction (K. ELLIOTT and J. WHELAN, eds.) Ciba Foundation Symposium 68 (New series). Elsevier/North Holland, Amsterdam (1979). M.S. HERSHFIELD, Apparent suicide inactivation of human lymphoblast s-adenosylmethionine hydrolase by 2-deoxy-adenosine and adenosine arabinoside, J. Biol. Chem. 254, 22-45 (1979). M.S. HERSHFIELD, N. M. KREDITCH, D. R. OWNDY and R. BUCKLEY, In vivo inactivation of erythrocyte s--adenosylhomocysteine hydrolase by 2'--deoxyadenosine in adenosine deaminase-deficient patients, J. Clin. lnvestig. 63, 807-811 (1979). M. S. HERSHFIELD, Proposed explanation for s--adenosylhomocysteine hydrolase deficiency in purine nucleoside phosphorylase and hypoxanthine-guanine phosphoribosyltransferase-deficient patients, J. Clin. lnvestig. 67, 696-701 ( 198 I). A.W. MURRAY, The biological significance of purine salvage, Ann. Rev. Biochem. 40, 811-826 (1971). L.G. LAJTHA and J. R. VANE, Dependence of bone marrow cells on the liver for purine, Nature 182 191-192 (1958). W.J. HOWARD, L. A. KERSON and S. H. APPEL, Synthesis de novo of purine in slices of rat brain and liver, d. Neurochem. 17, 121-123 (1970). J. ALLSOP and R. W. E. WATTS, Activities of amidophosphoribosyltransferase (EC 2.4.2.14) and the purine phosphoribosyltransferases (EC 2.4.2.7 and 2.4.2.8) and the phosphoribosylpyrophosphate content of rat central nervous system at different stages of development, J. Neurol. Sci. 46, 221-233 (1980). P. BROSH, P. BOER, E. ZOREF-SHANI and O. SPERLING, De novo purine synthesis in skeletal muscle, Biochim. Biophys, Acta 714, 181-183 (1982). R. O. McKERAN, A. HOWELL, T. M. ANDREWS, R. W. E. WATTS and C. F. ARLETT, Observations on the growth in vitro of myeloid progenitor cells and fibroblasts from hemizygotes and heterozygotes of 'complete' and 'partial" hypoxanthine guanine phosphoribosyl transferase (HGPRT) deficiency, J. Neurol. Sci. 22, 183-195 (1974). R. O. McKERAN and R. W. E. WATTS, Use of phytohaemagglutinin stimulated

PURINE NUCLEOTIDE BIOSYNTHESIS

29 30. 31.

32. 33. 34.

35.

36. 37. 38.

AI~R~C

51

lymphocytes to study effects of hypoxanthine-guanine phosphoribosyltransferase (HGPRT) deficiency on polynucleotide and protein synthesis in the Lesch-Nyhan syndrome, £ Med. Genet. 13, 91-95 (1976). F.E. BLOOM, The role of cyclic nucleotides in central synaptic function, Rev. Physiol. Biochem. Pharmacol. 74, 1-103 (1975). R. O. McKERAN and R. W. E. WATTS, Purine metabolism and cell physiology, pp. 219-252 in Recent Advances in Endocrinology and Metabolism (J. L. H. O'RIORDAN, ed.), Churchill Livingstone, Edinburgh (1978). R . W . E . WATTS, E. SPELLACY, D.A. GIBBS, J. ALLSOP, R.O. McKERANandG. E. SLAVIN, Clinical, postmortem, biochemical and therapeutic observations on the Lesch-Nyhan syndrome with particular reference to the neurological manifestations, Quart. £ Med. N.S. 51, 43-78 (1982). G. F. SCHREINER and E. R. UNANUE, Capping and the lymphocyte: models for membrane reorganisation, £ ImmunoL 119, 1549-1551 (1977). D.A. HUME and M..I. WlEDEMANN, Mitogenic lymphocyte transformation, Research Monographs in Immunology, Volume 2, pp 91-92. Elsevier/North Holland. Biomedical Press, Amsterdam (1980). T. HOVI, A. C. ALLISON, K. RAIVIO and A. VAHERI, Purine metabolism and control of cell proliferation, pp. 225--242 in Purine and l)vrimidine Metabolism. Ciba Foundation Symposium 48 (new series) (K. ELLIOTT and D. W. FITZSIMONS, eds.), Elsevier/ Excerpta Medica / North Holland, Amsterdam (1977). J. ALLSOP and R. W. E. WATTS, The amidophosphoribosyltransferase (EC 2.4.2.14) and ribosephosphate pyrophosphokinase (EC 2.7.6.1) activities, and phosphoribosylpyrophosphate content of unstimulated and stimulated human lymphocytes, J. Mol. Med. 3, 105-110 (I 978). T. HOVI, J. ALLSOP and A. C. ALLISON, Rapid increase of phosphoribosylpyrophosphate concentration after mitogenic stimulation of lymphocytes, FEBS Lett. 55, 291-293 (1975). L. THUILLIER, F. GARREAU and P. H. CARTIER, Consequences of adenosine deaminase deficiency on thymocyte metabolism, Europ. J. Immunol. 11,788-794 (1981). A. E. PEGG, H. HIBASAMI, 1. MATSUI and D. R. BETHELL, Formation and interconversion of putrescine and spermidine in mammalian cells, Advances in Enzyme Regulation 19, 427-451 (1980).