Clinical Science and Molecular Medicine (1978) 54, 595-601

Purine biosynthesis de novo by lymphocytes in gout P. KAMOUN, J. CHANARD'», M. BRAMI (1) AND J. L. FUNCK-BRENTANO« 1 » Laboratoire de biochimte genetique and mUniteINSERM £/90, Hopital Necker, Paris, France

{Received 6 January 1977; accepted 2 December 1977)

Summary 1. A method of measurement in vitro of purine biosynthesis de novo in human circulating blood lymphocytes is proposed. The rate of early reac­ tions of purine biosynthesis de novo was deter­ mined by the incorporation of [I4C]formate into JVformyl glycinamideribonucleotidewhen the subse­ quent reactions of the metabolic pathway were completely inhibited by the antibiotic azaserine. 2. Synthesis of I4C-labelled /V-formyl glycin­ amide ribonucleotide by lymphocytes was measured in healthy control subjects and patients with primary gout or hyperuricaemia secondary to renal failure, with or without allopurinol therapy. 3. The average synthesis was higher in gouty patients without therapy than in control subjects, but the values obtained overlap the normal range. In secondary hyperuricaemia the synthesis was at same value as in control subjects. 4. These results are in agreement with the inconstant acceleration of purine biosynthesis de novo in gouty patients as seen by others with measurement of [uC]glycine incorporation into urinary uric acid.

limitantes de cette biosynthese est determinee par la mesure de l'incorporation de [uC]formate dans le ΛΓ-formyl glycinamide ribonucleotide (FGAR) quand les reactions ulterieures de cette voie metabolique sont completement et irreversiblement inhibes par un antibiotique Γ azaserine. 2. La synthese du F14C]FGAR est mesuree chez des temoins et des malades, atteints de goutte primaire ou secondaire a une insuffisance renale, traites ou non par l'allopurinol. 3. La synthese de [I4C]FGAR est plus elevee chez les goutteux non traites que chez les temoins mais les valeurs observees dans les deux groupes se chevauchent. Dans l'hyperuricemie secondaire ä une insuffisance renale, la synthese de [14C]FGAR est identique ä celle des temoins. 4. Ces resultats sont compatibles avec l'inconstante acceleration de la biosynthese de novo des purines determinee par l'incorporation de [14C]glycine dans l'acide urique urinaire. Introduction The measurement of purine biosynthesis de novo in isolated cells has been studied in gout. An insensitive method has been used in human leucocytes (Diamond, Friedland, Halberstam & Kaplan, 1969; Chang, Fam, Little & Malkin, 1974), using [14C]glycine and measuring its incor­ poration into adenine and guanine residues of DNA and 'insoluble' RNA. Alternatively, the first steps of purine biosynthesis de novo can be measured with either [14C]formate or f14C]glycine as precursor of the JV-formyl glycinamide ribo­ nucleotide and azaserine, which irreversibly inhibits the pathway at this nucleotide level (Bennett, Schabel & Skipper, 1956; French, Dawid, Day &

Key words: allopurinol, gout, lymphocytes, purines, uric acid. Abbreviation: FGAR, ΛΓ-formyl glycinamide ribo­ nucleotide. Resume 1. Une methode est proposee pour une mesure rapide de la biosynthese de novo des purines dans les lymphocytes humains. La vitesse des etapes Correspondence: Professor P. Kamoun, Hopital Necker, 75730 Paris Cedex 15, France.

595

596

P. Kamoun et al.

Buchanan, 1963). This method has been used in fibroblasts derived from skin biopsy (Kelley & Wyngaarden, 1970) and leukaemic and ascites tumour cells in culture (Henderson, 1962). We have applied this technique directly to human circulating blood lymphocytes without cell culture.

Materials and methods Subjects Three groups of subjects were studied. Informed consent for drawing blood samples was obtained from each subject and the research was carried out according to the Declaration of Helsinki. Group 1 consisted of 20 healthy volunteer blood donors (12 males and eight females) aged from 20 to 62 years. Plasma uric acid concentration measured by the standard Auto-analyzer method was 0-32 + 0-02 mmol/1 (mean ± SEM). In group 2 (18 males and six females) primary gout was diagnosed by the usual clinical criteria (Wyngaarden & Kelley, 1972). Arthritis was classified in three stages: (a) resolving acute arthritis, nine patients; (b) chronic arthritis, nine patients including one patient with uraemia; (c) destructive arthritis with extraarticular tophi (clinical joint deformity with limited movement and X-ray evidence), six patients includ­ ing three patients with uraemia. In these patients the diagnosis of primary gout was made according to the following criteria: knowledge of arthritis and hyperuricaemia 7-30 years before the onset of uraemia. Familial gout was noted in five patients of two families. Eleven patients (group 2B) were treated with allopurinol (29-48 ßmol day - 1 kg"'; 4-0-6-5 mg day - 1 kg -1 )· The plasma uric acid con­ centration (mean + SEM) was 0-39 ± 0-04 mmol/1, whereas in the remaining patients (group 2A) this concentration was 0-55 ± 0-03 mmol/1. One patient was studied before and during allopurinol therapy. Twenty-two patients in group 3 consisted of 10 males and 12 females with renal failure and secondary hyperuricaemia without any clinical criteria of gout. Eight (group 3B) were treated with allopurinol (29-48 μτηοΐ day' 1 kg - 1 ; 4·0-6·5 mg day" 1 kg -1 ). Their plasma uric acid concentration was 0-47 + 0-04 mmol/1 whereas in the remaining patients (group 3 A) this concentration was 0-64 + 0-03 mmol/1. The aetiology of their renal failure was: chronic glomerulonephritis (five cases in group 3A, two cases in group 3B); interstitial nephritis (five case in group 3A, four cases in group 3B); acute glomerulonephritis (one case in group 3A); nephrosclerosis (three cases in group 3A, one

case in group 3B); polycystic kidneys (one case in group 3B). Preparation and incubation of lymphocytes Fasting venous blood (40 ml) was collected in plastic tubes containing 2-5 mg of heparin (Liquemin, Roche). Lymphocytes were isolated by centrifugation on a Ficoll-Triosil gradient (Böyum, 1963). All centrifugations were carried out at 4°C. Lymphocytes were collected by centrifugation at 400 g for 15 min, and then washed with Hank's medium (medium 199, Grand Island Biological Co., New York). The lymphocyte pellet was sus­ pended in 2 ml of Hank's medium. Duplicate cell counts were made and the final lymphocyte con­ centration was adjusted to an average 107 cells/ml, 1 -0 ml of the lymphocyte suspension being centrifuged at 400 g for 15 min. The supernatant was discarded and the lymphocyte pellet suspended in 2 ml of the following isotonic medium: potassium dihydrogen phosphate (25 mmol/1) adjusted to pH 7-4 with NaOH solution (2-0 mol/1) and contain­ ing glutamine (2 mmol/1), glucose (5· 5 mmol/1), sodium chloride (42-2 mmol/1), azaserine (Calbiochem) (1 mmol/1), sodium [14C]formate (0-5 mmol/1; specific radioactivity 6-25 mCi/mmol;The Radiochemical Centre, Amersham, Bucks., U.K.). The lymphocyte suspensions were then incubated at 37°C for 3 h with gentle shaking in a Dubnoff metabolic incubator, after which the tubes were chilled at 4°C to inhibit further reaction. After centrifugation at 900 g for 10 min, supernatants were discarded and lymphocytes washed with 3 ml of cold isotonic sodium chloride solution. Pellets were obtained after centrifugation at 320 g for 10 min and 2-5 ml of perchloric acid (0-52 mol/1) was added to each tube. After centrifugation at 320 g for 10 min, supernatants were collected and pellets washed with 1 ml of perchloric acid (0-52 mol/1). After another centrifugation, the super­ natants were collected and mixed with the preced­ ing supernatants and saved for determination of I4 C-labelled 7V-formyl glycinamide ribonucleotide ([ 14 C]FGAR). Pellets were dissolved in 2 ml of NaOH solution (0-48 mol/1) for protein deter­ mination (Lowry, Rosebrough, Farr & Randall, 1951), bovine serum albumin (200 mg/1; Sigma, Louis, U.S.A.) being used as standard. Measurement of luC]ribonucleotide Supernatants obtained during the preceding steps were neutralized to pH 7-0 with KOH solution (1 mol/1), stored at 4°C for 30 min, and then centrifuged at 900 g for 15 min. ['"ClFGAR

597

Purine biosynthesis by lymphocytes

pH2 9

5000

5000 -

flf 0-59

IMP

I

5000

I

ff, 0 39

pHB-8 5000

V

!

Ö

5000 r

fl(0-66

RSSSSN

5000

1 FIG. 1. Chromatographie control of l,4C]FGAR purity: solvent no. 1, butanol/acetic acid/water (70:20:10, by vol.); solvent no. 2, 95% ethanol/ammonium acetate (6 mol/l)/water (70:5:25, by vol.); solvent no. 3, propan-1-ol/aq. ammonia solution (sp.gr. 0-880)/water (60:30:10, by vol.); solvent no. 4, isobutyric acid/aq. ammonia solution (2 mol/1) (66:34, v/v). The arrow indicates the start of the chromatography. After development the paper strip was cut into 1 cm sections (showed in abscissae) for radioactivity measurements. Each section was mixed with 10 ml of Instagel. The 14C radioactivity was expressed in c.p.m. The ratio between the migration of radioactivity and the migration of solvent (Rf) is also indicated.

extraction was done according to Henderson (1962). Neutralized perchloric extracts were poured on columns (10 mm x 35 mm) of Dowex-1 X 8 formate (50-100 mesh). After adsorption, the columns were washed with 50 ml of formic acid (0-5 mol/1). Effluent and washes were discarded. FGAR was then eluted with 32-5 ml of formic acid (4 mol/1). The first 2-5 ml was discarded and six fractions of 5 ml each were collected. A portion (10 ml) of Instagel (Packard) was added to each fraction. The vials were counted for radioactivity in a liquid scintillation counter (SL 40, Intertechnique, Plaisir, France) for 20 min. The values obtained were summed and [ U C]FGAR synthesis was expressed as c.p.m./mg of protein.

FIG. 2. Electrophoretic control of 1WC|FGAR purity: highvoltage electrophoresis was performed with the following buffers: no. 1, sodium citrate (50 mmol/l), pH 2-9; no. 2, sodium borate (50 mmol/l)/EDTA (1 mmol/l), pH 8-8. Dried electrophoretic paper was cut into 1 cm sections (shown in abscissae) for radioactivity measurements. Each section was mixed with 10 ml of Instagel. The migration pattern of inosinic acid solution (IMP) on electrophoresis is also indicated. The arrow indicates the start of electrophoresis.

Control of purity

of[uC]ribonucleotide

Descending chromatography and electro­ phoresis on Whatman no. 1 paper were performed after column chromatography. However, to obtain [ I4 C1FGAR with a higher specific radioactivity we modified the technique described above by using the 114C Iformate with specific radioactivity 50 mCi/ mmol. After column chromatography, the 30 ml of formic acid (4 mol/1) extract was dried and the residue dissolved in 500 μΐ of distilled water, 50 /A being used for each paper chromatography or electrophoresis. Four solvents were used (Fig. 1). High-voltage electrophoresis (2500 V for 30 min) was performed with two buffers (Fig. 2). A portion (20 μ\) of a saturated inosinic acid solution was simultaneously electrophoresed. Radioactivity was determined as described above. Results were expressed as mean values ± 1 SEM. Results Lymphocyte purification Contamination by other blood cells was measured in 20 blood samples after isolation by the Ficoll-Triosil technique. The number of lym­ phocytes was 111 50 ± 1115//A. The contamination was 3495 ± 695 erythrocytes/μΐ, 300 ± 90 polymorphonuclear leucocytes/μΐ and 151 700 + 18 200 thrombocytes/,ul.

P. Kamoun et al.

598

_

10 000

E

5000

0

10 20 106 x lymphocyte count

FIG. 3. Variation of | 1 4 ClFGAR synthesis with number of lymphocytes in healthy control subjects. Each symbol refers to the same blood donor in whom synthesis was measured with three or four different quantities of lymphocytes. The time of incubation was 3 h in the presence of l' 4 Clformate and azaserine (1 mmol/l). Regression equation: y = 520(+70)x + 1670(±620).

Synthesis (c.p.m./106 lymphocytes) FIG. 4. Correlation between [ M C]FGAR synthesis (c.p.m.) expressed per mg of protein and per 106 lymphocytes. Results are for 6 x 106 lymphocytes per incubation tube as indicated in Fig. 3. Regression equation: y = 5·9(±2·5) x + 4890(±1490); r -= 0-49, P < 0-05.

Purity of I uC\ribonucleotide Only one peak of radioactivity was detected on chromatography with solvents no. 2 and no. 4 (Fig. 1). The corresponding RF values were 0-39 and 0-38 respectively, differing from those obtained by Kelley & Wyngaarden (1970) with ascending chromatography. On the other hand, on chromato­ graphy with solvents no. 1 and no. 3 Ä F values were 0-05 and 0-17 respectively, identical with those found by Kelley & Wyngaarden (1970); another peak of radioactivity was also seen but this second peak contained less than 9% of the total radioactivity. In the absence of azaserine the same main peaks were found but at a very low level. With high-voltage electrophoresis only one peak was obtained (Fig. 2). The comparative migration of the labelled material and inosinic acid solution confirmed the results of Kelley & Wyngaarden (1970). Therefore we considered that more than 90% of the radioactivity measured on eluates obtained by ion-exchange chromatography was [I4C1FGAR (as determined by paper chromato­ graphy with solvents no. 1 and no. 3). Biosynthesis of

[uC\ribonucleotide

Control subjects. 1'4C|FGAR synthesis was proportional to the number of lymphocytes in­ cubated with azaserine (1 mmol/l)) (Fig. 3) (r = 0-91, P < 0-001). However, it seems that this

relationship is only linear between 2 and 16 x 106 lymphocytes. We also compared [ 14 ClFGAR synthesis expressed per mg of protein and per 106 lymphocytes (Fig. 4). [ l4 C]FGAR synthesis was proportional to the incubation time (Table 1) in control subjects, in gout and in renal failure with and without allopurinol therapy. A 3 h incubation time was thus selected after checking that more than 85% of the cells were alive by Trypan Blue exclusion. Fig. 5 shows the t 14 C]FGAR values obtained when the azaserine concentration in the incubation medium ranged from 0·01 to 1 mmol/l. At higher concentrations [ I4 C]FGAR synthesis decreased and was suppressed by azaserine (10 mmol/l). In 20 healthy control subjects [ 14 C]FGAR biosynthesis measured after 3 h incubation at 37°C with azaserine (1 mmol/l) was 8530 ± 350, ranging from 5890 to 10 390 c.p.m./mg of protein. Patients with primary gout. In 13 patients with primary gout without allopurinol therapy (group 2A) ['"ClFGAR biosynthesis was 11750 ± 840 c.p.m./mg of protein, higher than the control value (Fig. 6), but did not achieve statistical significance. Seven gouty patients overlap with the control value. In eight of 11 treated patients ['"ClFGAR biosynthesis decreased below the control value. Among the three non-responders two belonged to the same family.

Purine biosynthesis by lymphocytes

599

TABLE 1. Rate of synthesis of "C-labelled N-formyt glycinamide ribonucleotide from 107 lymphocytes In the presence ofazaserine (1 mmolll) and [ "Clformate Values are in c.p.m./mg of protein (mean + SEM) for the incubation times shown. Numbers in parentheses indicate the number of determinations. Values were selected in order to take into account experiments in which lymphocytes were used for at least two incubation-time measurements. Subjects

Allopurinol therapy

14

C radioactivity (c.p.m./mg of protein)

Regression equations

60 min

90 min

180 min

Healthy (control)

None

3 7 1 0 + 140 (7)

5430 + 270 (7)

9450 + 410 (9)

Primary gout

None

3890 + 660 (4)

6470 + 700 (10)

11710+ 1020 (10)

>> = 61(±10)x + 750(±1230)

29-48 Mtaol day"'kg-1

2670 + 260 (3)

5020 + 890 (7)

8860 + 1990 (7)

y = 47(± Π)χ + 470(±2070)

None

3030 + 350 (3)

4950 + 310 (12)

8670 + 590 (12)

>> = 43(±6)* + 940(±760)

29-48 μιτιοΐ day"'kg~'

—

4830 ± 5 1 0 (4)

7320 + 950 (4)

y = 30(±9)x+

Renal failure with hyperuricaemia

Patients with renal failure. In secondary gout [I4C]FGAR biosynthesis in lymphocytes was 8610 ± 560 and 8570 + c.p.m./mg of protein respectively, in 14 patients without allopurinol therapy (group 3A) and eight patients with therapy (group 3B). Thus in each group [14C]FGAR synthesis was at the same level as in control subjects.

>> = 4 7 ( + 3 ) Λ : + 1020(+350)

1950(±1140)

»-Λ

|> 10«

Correlation between [HC]ribonucleotide and plasma uric acid We did not find any correlation between [14C]FGAR values and plasma uric acid con­ centration in these patients; gouty patients with uraemia could not be separated from gouty patients without chronic renal failure. In one gouty patient, [UC]FGAR biosynthesis was measured before and after 10 weeks of allopurinol therapy (18 μιηοΐ day -1 kg -1 ; 2-5 mg day -1 kg-1)· Plasma uric acid concentration fell from 0·60 to 0-33 mmol/1 and [14C]FGAR biosynthesis fell from 9650 to 6150 c.p.m./mg of protein. Discussion The lymphocyte suspension obtained by Böyum's (1963) technique is contaminated by a significant number of other blood cells, especially thrombocytes, but as human platelets do not synthesize nucleotides de novo (Jerushalmy, Sperling, Pinkhas, Krynska & de Vries, 1974) the values of [14C]FGAR radioactivity expressed per mg of protein or per 106 lymphocytes are not significantly different (Fig. 3). We are unable to measure

001

0-1 0-25 0-50 1 2 Concn. ofazaserine (mmol/1)

FIG. 5. Relationship between [14C|FGAR synthesis in circula­ ting lymphocytes and various concentrations of azaserine in the incubation medium. The vertical bars indicate mean ± SD for seven healthy control subjects ( o ) and five patients with renal failure ( · ) . Background radioactivity without azaserine was 2950 ± 440 (mean ± SEM) and 2480 ± 260 c.p.m./mg of protein respectively for control and patients with renal failure. The curves are the values observed for two gouty patients. Back­ ground radioactivity without azaserine was respectively 2710 (continuous line) and 6010 (broken line) c.p.m./mg of protein.

Chromatographie and electrophoretic migration of pure [14C]FGAR. However, Le Page & Jones (1961) and Henderson (1962) have shown that the Chromatographie technique used in our study could measure [14C]FGAR. When human epidermoid carcinoma cells are incubated with [I4C]formate and azaserine, four radioactive products can be isolated: serine, formyl glycinamide ribonucleotide

P. Kamoun et al.

600 4

·. 10 r

•

t

A%

^ F• ~

·· • ·· • : • ··

·· ··

1

•

• •

• •

M •

1

1

•

s

··

·1· • • ·· ··

• •

1

3A

1 3B

FIG. 6. Individual values of |I4C]FGAR biosynthesis in circulating lymphocytes from healthy control subjects (1), patients with primary gout without (2A) or with allopurinol therapy (2B), and patients with renal failure without (3A) or with (3B) allopurinol therapy. The values are provided by 10' lymphocytes incubated for 3 h in the presence of I l4C|formate and azaserine (1 mmol/l).

(FGAR) and nucleotides identified as formyl glycinamide ribonucleoside polyphosphates. But most of the radioactivity is found in FGAR (Brockman & Chumley, 1965). Identical studies in cultured human fibroblasts (Kelley & Wyngaarden, 1970; Rosenbloom, Henderson, Caldwell, Kelley & Seegmiller, 1968) have also shown a maximal incorporation of 14C in FGAR. We have checked that free [l4Clformate and serine were discarded by the Dowex-1 X 8 formate chromatography. Two of the four solvents used for paper chromatography gave the same RF values as those obtained by Kelley & Wyngaarden (1970). However, with the two ammoniacal solvents, the R¥ values were different in ascending and descending chromato­ graphy. Taking into account the migration of inosinic acid, the electrophoretic results were identical with those obtained by these investi­ gators. The dose-response relationship between [MC1FGAR synthesis and azaserine concentration also suggests that the main radioactivity peak was [ 14 C]FGAR. With azaserine concentrations equal to or higher than 10 mmol/l, the amido-phosphoribosyl transferase (EC.2.4.2.14) is inhibited and [14C]FGAR synthesis inhibited. Reem (1972) showed that the enzymes required for the first two steps of the purine biosynthesis de novo are present in human spleen cells. The presence of at least the two first enzymes in circulating blood lymphocytes is confirmed by our results, in contrast to Scott (1962), who studied [14C|glycine incorporation into soluble nucleotides

and nucleic acids in human leucocytes and con­ cluded that normal and leukaemic human leuco­ cytes lacked the enzyme system which catalyses the first and the second steps of purine biosyn­ thesis de novo. Scott (1962) was also unable to find [14Clformate incorporation into the purines of the nucleic acids extracted from normal mature granulocytes in chronic myelocytic leukaemia, possibly because of the low sensitivity of the tech­ nique with low specific radioactivity formate, the concentration of which when distributed within the formate pool might decrease to background values. Nevertheless Diamond et al. (1969) and Chang et al. (1974) showed that [14C]glycine is incorporated into leucocyte nucleic acids. There is little variation in [ 14 C]FGAR bio­ synthesis in control subjects (Fig. 6). On the contrary, in patients with primary gout, these values are widely scattered and eight of 11 patients treated with allopurinol were in the control range. On the other hand, [ 14 C]FGAR biosynthesis was not modified in renal failure associated with hyperuricaemia. The incorporation of [14C]glycine into urinary uric acid was not determined in our subjects so we have no evidence that the rate of FGAR synthesis in lymphocytes correlated with the rate of purine synthesis in vivo. The results in the two families of gouty patients were reasonably homogeneous. The first family included two twin brothers with plasma uric acid concentrations of 0-57 and 0·61 mmol/l respec-

Purine biosynthesis by lymphocytes U

tively. The corresponding [ C]FGAR values were 7900 and 8080 c.p.m./mg of protein, both within our control values. The second family included two men and one woman whose plasma uric acid concentrations were 0-42, 0-26 and 0·35 mmol/1 respectively, and the corresponding [14C]FGAR values were 12 660, 19 170 and 12 970 c.p.m./mg of protein, the last being higher than the control values, in spite of the fact thefirsttwo men were treated with allopurinol for more than 1 year (38 and 30 /imol day -1 kg -1 ; 5-2 and 4· 1 mg day -1 kg -1 respectively). Values for the incorporation of glycine into urinary uric acid are indicative rather than definitive measurements of the rates of purine production. No independent quantitative method for assessing purine production is always reliable (Wyngaarden & Kelley, 1972). The 24 h urinary uric acid value generally represents about twothirds of the turnover in normal man, but in gout this may be a smaller fraction (Seegmiller, Grayzel, Laster & Liddle, 1961). The glycine-incorporation studies suggest that subjects with primary gout show a spectrum of rates of production of uric acid ranging from values within the normal range to much higher values in patients with hypoxanthine phosphoribosyl transferase (EC 2.4.2.8) deficiency (Wyngaarden & Kelley, 1972). Our results in vitro support these observations. Our technique allows rapid and accurate measurement in vitro of purine biosynthesis de novo by human lymphocytes. Its sensitivity is high: [14C]FGAR concentration was always higher than 1800 c.p.m./mg of protein whereas a lower concentration of 45 d.p.m./mg of protein was achieved by the technique of Diamond et al. (1969), as modified by Chang et al. (1974).

Acknowledgments We acknowledge the excellent technical assistance provided by Miss Marie-Helene Gaillard. This investigation was supported by Contrat INSERM no. 74.5.105.0.

601

References BENNET, L.L., JR, SCHABEL, F.M., JR & SKIPPER, H.E. (1956)

Studies on the mode of action of azaserine. Archives of Biochemistry and Biophysics, 64,423-436. BOYÜM, A. (1963) Separation of leucocytes from blood and bone marrow. Scandinavian Journal of Clinical and Laboratory Investigations, 12, Suppl. 97. BROCKMAN, R.W. & CHUMLEY, S. (1965) Inhibition of formylglyein amide ribonucleotide synthesis in neoplastic cells by purine analogs. Biochimica et Biophysica Acta, 95,365-379. CHANG, G.N., FAM, A., LITTLE, A.M. & MALKIN, A. (1974)

The uptake of glycine U C into the adenine and guanine of DNA and insoluble RNA of human leukocytes. Advances in Experimental Medicine and Biology, 41B, 407-416. DIAMOND, H.S., FRIEDLAND, M., HALBERSTAM, D. & KAPLAN,

D. (1969) Glycine C 14 incorporation into nucleic acid purine by leucocytes obtained from normal and gouty subjects. Annals of the Rheumatic Diseases, 28,275-280. FRENCH, T.C., DAWID, I.B., DAY, R.A. & BUCHANAN, J.M.

(1963) Azaserine reactive sulfydryl group of 2-formamido iV-ribrosylacetamide 5'-phosphate:L-glutamate amidoligase (adenosine diphosphate). 1. Purification and properties of the enzyme from Salmonella typhimurium and the synthesis of Lazaserine- 14 C. Journal of Biological Chemistry, 238, 2 1 7 1 2177. HENDERSON, J.F. (1962) Feedback inhibition of purine biosyn­ thesis in ascites tumor cells. Journal of Biological Chemistry, 237,2631-2635. JERUSHALMY, Z., SPERLING, O., PINKHAS, J., KRYNSKA, M. &

DB VRIES, A. (1974) Enzymes of purine metabolism in platelets: phosphoribosyl pyrophosphate synthetase and purine phosphoribosyl transferase. Advances in Experi­ mental Medicine and Biology, 41,159-162. KELLEY, W.N. & WYNGAARDEN, J.B. (1970) Effects of allo­ purinol and oxipurinol on purine synthesis in cultured human cells. Journal of Clinical Investigation, 49,602-609. LE PAGE, G.A. & JONES, M. (1961) Purine thiols as feedback inhibitors of purine synthesis in ascites tumor cells. Cancer Research, 21,624-649. LOWRY, O.H., ROSEBROUGH, N.J., FARR, A.L. & RANDALL,

R.J. (1951) Protein measurement with Folin-phenol reagent. Journal of Biological Chemistry, 193,265-275. REEM, G.H. (1972) De novo purine biosynthesis by two path­ ways in Burkitt lymphoma cells and in human spleen. Journal of Clinical Investigation, 51,1058-1062. ROSENBLOOM,

F.M.,

HENDERSON,

J.F.,

CALDWELL,

I.C.,

KELLEY, W.N. & SEEGMILLER, J.E. (1968) Biochemical basis of accelerated purine biosynthesis de novo in human fibroblasts lacking hypoxanthine-guanine phosphoribosyl trans­ ferase. Journal of Biological Chemistry, 243,1166-1173. SCOTT, J.L. (1962) Human leukocytes metabolism in vitro. I. Incorporation of adenine 8- u C and formate-"C into the nucleic acids of leukemic leukocytes. Journal of Clinical Investigation, 41,67-79. SEEGMILLER, J.E., GRAYZEL, A.I., LASTER, L. & LIDDLE, L.

(1961) Uric acid production in gout. Journal of Clinical Investigation, 40,1304-1314. WYNGAARDEN, J.B. & KELLEY, W.N. (1971) Gout. In: The Metabolic Basis of Inherited Diseases, pp. 889-968. Stanbury, J.B., Wyngaarden, J.B. & Fredrickson, D.S. McGrawHill, New York.