Proc. Nati. Acad. Sci.
USA
Vol. 74, No. 12, pp. 5677-5681, December 1977
Immunology
Lymphospecific toxicity in adenosine deaminase deficiency and
purine nucleoside phosphorylase deficiency: Possible role
of nucleoside kinase(s)
(immunodeficiency/lymphocyte/purine deoxyribonucleoside kinase/purine deoxyribonucleotides)
DENNIS A. CARSON*, JONATHAN KAYE*, AND J. E. SEEGMILLERt
*Division of Rheumatology, Department of Clinical Research, Scripps Clinic and Research Foundation, La Jolla, California 92037; and t Department of
Medicine, University of California, San Diego, La Jolla, California 92037
Contributed by J. Edwin Seegmiller, September 26, 1977
ABSTRACT Inherited deficiencies of the enzymes adeno-
sine deaminase (adenosine aminohydrolase; EC 3.5.4.4) and
purine nucleoside phosphorylase (purine-nucleoside:ortho-
phosphate ribosyltransferase; EC 2.4.2.1) preferentially interfere
with lymphocyte development while sparing most other organ
systems. Previous experiments have shown that through the
action of specific kinases, nucleosides can be "trapped" intra-
cellularly in the form of 5'-phosphates. We therefore measured
the ability of newborn human tissues to phosphorylate adeno-
sine and deoxyadenosine, the substrate of adenosine deaminase,
and also inosine, deoxyinosine, guanosine, and deoxyguanosine,
the substrates of purine nucleoside phosphorylase. Substantial
activities of adenosine kinase were found in all tissues studied,
while guanosine and inosine kinases were detected in none.
However, the ability to phosphorylate deoxyadenosine, deoxy-
inosine, and deoxyguanosine was largely confined to lympho-
cytes. Adenosine deaminase, but not purine nucleoside phos-
phorylase, showed a similar lymphoid predominance. Other
experiments showed that deoxyadenosine, deoxyinosine, and
deoxyguanosine were toxic to human lymphoid cells. The tox-
icity of deoxyadenosine was reversed by the addition of de-
oxycytidine, but not uridine, to the culture medium. Based upon
these and other experiments, we propose that in adenosine
deaminase and purine nucleoside phosphorylase deficiency,
toxic deoxyribonucleosides produced by many tissues are se-
lectively trapped in lymphocytes by phosphorylating en-
zyme(s).
During the past 5 years, Giblett and her colleagues have dem-
onstrated an association between severe deficiencies of either
adenosine deaminase (adenosine aminohydrolase; EC 3.5.4.4),
or purine nucleoside phosphorylase (purine-nucleoside:ortho-
phosphate ribosyltransferase; EC 2.4.2.1) and inherited forms
of human immunodeficiency disease (1, 2). Although the
clinical pictures overlap, children with adenosine deaminase
deficiency usually suffer from a combined immunodeficiency
syndrome, with impairment of T cell development and in most
cases of B cell function as well, while those with purine nucle-
oside phosphorylase deficiency have primarily a deficit in T
cell development and the associated cellular immune functions.
Both diseases are accompanied by severe lymphopenia. Al-
though enzyme is virtually absent from all tissues examined,
apparently only the growth and development of the lymphoid
system is severely retarded (3). It is therefore likely that the
immune defect in deaminase and phosphorylase deficiency is
not due to a generalized disorder of growth, but rather to a
primary lymphocyte abnormality and/or circulating toxins that
are specifically lymphocytoxic. The latter concept is in accord
with the reversal of the immunodeficient state accompanying
The costs of publication of this article were defrayed in part by the
payment of page charges. This article must therefore be hereby marked
"advertisement" in accordance with 18 U. S. C. §1734 solely to indicate
this fact.
5677
deaminase deficiency by enzyme replacement in the form of
erythrocyte transfusions (4).
Three biochemical mechanisms have been proposed to ex-
plain the association of deaminase deficiency with immuno-
deficiency disease, i.e., adenosine-induced pyrimidine star-
vation (5), hypoxanthine deficiency (6), and adenosine-me-
diated elevations in cyclic AMP concentrations (7). In the ab-
sence of further information, these hypotheses do not explain
the preferential impairment of lymphoid development seen
in both phosphorylase and deaminase deficiency. In the present
studies, we suggest that lymphospecific toxicity in deaminase
and phosphorylase deficiency may result from the selective
accumulation in lymphoid tissues, particularly the thymus, of
toxic deoxyribonucleotides, mediated by nucleoside
kinase(s).
MATERIALS AND METHODS
Tissue Extracts. Newborn human tissues obtained at autopsy
within 24 hr of death were frozen at -20°. Extracts were pre-
pared by mincing the specimens in 10 mM Tris buffer (pH 7.4),
followed by five cycles of freeze-thawing, and ultracentrifu-
gation of the particulate material.
Peripheral blood lymphocytes and granulocytes were isolated
by dextran sedimentation of heparinized whole blood from an
adult volunteer, followed by centrifugation through Ficoll-
Hypaque (8). Red cells were lysed with Tris-buffered ammo-
nium chloride and the cells were washed and frozen (9).
The protein content of all tissue extracts was determined by
Lowry's method, with bovine serum albumin as a standard
(10).
Enzyme Assays. Kinase activities in cell extracts and column
fractions were determined by a modification of the method of
Ives et al. (11). For the measurement of deoxyguanosine, gua-
nosine, deoxyinosine, and inosine kinases, the final concentra-
tions of the reactants were: 50 mM Tris (final pH 7.4), 10 mM
ATP, 10 mM MgCI2, 15 mM NaF, 1 mg of protein per ml, and
0.4-2 ,uCi of substrate at a concentration of 300 ,M in a total
volume of 100 X. Adenosine and deoxyadenosine kinases were
similarly measured, except that the ATP and magnesium
concentrations were 5 and 2.5 mM, respectively. In addition,
to each sample was added the deaminase inhibitor erythro-
9-(2-hydroxy-3-nonyl)adenine hydrochloride (EHNA) to a final
concentration of 5 ,M.
The reactions were initiated by the addition of labeled sub-
strate. After 30 min at 370 in a shaking water bath, the samples
Abbreviations: deaminase, adenosine deaminase; phosphorylase, purine
nucleoside phosphorylase; EHNA, erythro-9-(2 hydroxy-3-nonyl)-
adenine hydrochloride.
5678 Immunology: Carson et al. Proc. Natl. Acad. Sci. USA 74 (1977)

Table 1. Enzyme activities in human tissues

Purine
Adenosine Deoxyadenosine Deoxyguanosine Deoxyinosine Adenosine nucleoside
kinase kinase kinase kinase deaminase phosphorylase
I II I II I II I II II II
Thymus 0.79 0.86 1.35 0.78 1.92 1.39 0.34 0.31 982.8 23.3
Spleen NA 0.53 NA 0.20 NA 0.33 NA 0.07 12.4 54.0
Brain NA 1.01 NA 0.14 NA 0.16 NA 0.05 5.0 10.3
Kidney NA 1.15 NA 0.07 NA 0.08 NA 0.10 1.8 100.0
Liver 0.81 2.26 0.12 0.07 0.04 0.07 0.05 0.04 1.1 36.2
Lung 1.32 0.81 0.11 0.06 0.03 0.08 0.03 0.02 0.8 38.0
Small
intestine 0.41 0.52 0.13 0.08 0.03 0.11 0.09 0.07 14.2 63.9
Heart 0.48 0.51 0.13 0.08 0.03 0.11 0.07 0.06 2.1 32.2
Peripheral
lymphocytes 1.00 0.32 0.21 0.09 20.7 114.7
Peripheral
granulocytes 0.83 0.05 <0.02 0.03 11.9 121.4
Human tissues were obtained from two babies (I and II) who died during parturition, while peripheral lymphocytes and granulocytes were
isolated from the blood of a normal adult. Activities are expressed as nmol of product per min/mg of protein at a substrate concentration of 300
PM. NA, not available for study.
were boiled for 2 min and insoluble material was centrifuged Reagents. [8-14C[Adenosine and [8- 4C[deoxyadenosine
at 4°. Control experiments with [14C]inosine monophosphate were purchased from New England Nuclear (Boston, MA) and
showed no nucleotide breakdown under these conditions. used to synthesize inosine and deoxyinosine by treatment with
Nucleotides were separated from the nucleosides and bases calf deaminase (Calbiochem, San Diego, CA). [8-14C]Guanosine
by chromatography on PEI-cellulose thin-layer plates (E. was also obtained from New England Nuclear, while [8-3H]-
Merck, Darmstadt) affixed with a paper wick and developed deoxyguanosine came from Amersham/Searle (Arlington
overnight in methanol/water (1:1) (12). In representative ex- Heights, IL). All isotopes were tested for purity by thin-layer
periments the nucleotides that remained at the origin were chromatography and appropriately diluted with unlabeled
further fractionated into the mono-, di-, and triphosphates by nucleoside before use. EHNA was kindly provided by the
a second development in sodium formate (pH 3.4) (13) or by Burroughs Wellcome Co. (Research Triangle Park, NC). All
two-dimensional chromatography with a discontinuous buffer other reagents were of the highest grade obtainable from
system (14). commercial sources.
When tritiated isotopes were used, the nucleotide spots were
cut out and extracted with 1 ml of 1 M TrisIHCl/0.7 M MgCl2 RESULTS
at pH 7.4 before the addition of scintillation fluid (11). When Kinase Activities in Human Tissues. Table 1 shows the
14C isotopes were used, extraction of the product was unrtec- activities of adenosine kinase, deoxyadenosine kinase, deoxy-
essary. inosine kinase, and deoxyguanosine kinase in newborn human
Deaminase and phosphorylase activities were also deter- tissues as well as in adult human Jymphocytes and granulocytes.
mined radiochemically, as previously described (8, 15). Guanosine and inosine kinase activities were undetectable (less
Ion-Exchange Chromatography. Twenty milligrams of than 0.02 nmol/min per mg of protein) in any tissue, and are
thymic cell extracts was dialyzed against 5 mM sodium phos- not shown. As can be seen, the ability to phosphorylate adeno-
phate (pH 7.9) and applied to a 1-ml column of DE52-cellulose sine was widespread among human organs. On the contrary the
(Whatman Ltd., Maidstone, Kent) equilibrated at 40 with the ability to phosphorylate deoxyadenosine and deoxyguanosine
same buffer. After unbound material was washed with the was largely confined to the thymus and peripheral blood
above buffer, the column was eluted with a linear gradient lymphocytes. Deoxyinosine kinase also showed greatest activity
made from 6 ml of the above buffer and 6 ml of the same buffer in lymphoid cells, although not as marked as that of deoxy-
with 0.3 M NaCl. One-milliliter fractions were collected and adenosine and deoxyguanosine kinases.
assayed for enzymatic activity as described above. From 25-75% of the phosphorylated deoxyadenosine and
Lymphocyte Growth Studies. Normal human peripheral deoxyguanosine in thymic extracts was recovered as the de-
blood lymphocytes were cultured as described in RPMI 1640 oxyribonucleoside triphosphate. The product of deoxyinosine
medium supplemented with 2 mM glutamine, 10% heat-in- kinase, on the other hand, was exclusively deoxyinosine
activated human AB Rh-positive serum, and 10 ,tg of phyto- monophosphate.
hemagglutinin-M per ml (DIFCO, Detroit, MI) as well as var- Deaminase and Phosphorylase Levels in Human Tissues.
ious nucleosides (8). After 3 days, blastogenesis was measured Confirming the results of Adams and Harkness, we found that
by the linear uptake of tritiated leucine into protein over a 4-hr deaminase, like deoxyadenosine kinase, was present in higher
period. Thymidine uptake was not measured, insofar as pre- concentration in lymphoid tissues than in other organs (Table
vious experiments showed that nucleosides can increase or 1) (17). Phosphorylase, however, showed no lymphoid pre-
decrease thymidine uptake independent of any effects on ponderance.
growth (8). Chromatographic Properties of Nucleoside Kinases. The
A clone (Cl-35) of the human lymphoblastoid cell line WIL-2, chromatography of a human thymic extract on DE52-cellulose
deficient in hypoxanthine phosphoribosyltransferase (EC clearly separated adenosine and deoxyadenosine kinase ac-
2.4.2.8), was selected as described and grown in minimum es- tivities (Fig. 1). Deoxyguanosine and deoxyinosine kinase ac-
sential medium with 10% dialyzed fetal calf serum (16). tivities cochromatographed with deoxyadenosine kinase (results
 Immunology: Carson et al. Proc. Natl. Acad. Sci. USA 74 (1977) 5679

30.

Z
0-
CL80
w
NC

30% 60
P-A
0
15
qcs
40-
9
w
w
Q 20
106

0.1 Am L.OAM 10sM lOO1M 1mM

Nucleoside Concentration
FIG. 2. Potentiation of deoxyadenosine toxicity by the deaminase
Fraction Number inhibitor EHNA. Phytohemagglutinin-stimulated human peripheral
FIG. 1. Chromatography of nucleoside kinases on DE52-cellulose. lymphocytes were cultured with varying concentrations of deoxy-
A human thymic-cell extract was fractionated on DE52-cellulose with adenosine, either (0) with or (0) without 5,uM EHNA. After 3 days,
an NaCl gradient in phosphate buffer (pH 7.9). Individual fractions uptake of tritiated leucine into acid-precipitable material was de-
were assayed for (0) A28o, (0) adenosine kinase, and (-) deoxyade- termined. Percent control leucine uptake = (cpm with nucleoside/cpm
nosine kinase activity. Deoxyguanosine kinase and deoxyinosine ki- without nucleoside) X 100. Each point represents the mean percent
nase, which were also assayed, cochromatographed with the major uptake ±SEM of six replicate cultures. In the presence of 5 ,M EHNA
deoxyadenosine kinase peak. alone, leucine uptake was 82 + 4% of control values.
not shown). The approximate Km of the enzyme purified on osides can be acted upon by specific kinases and "trapped"
DE52-cellulose for all the purine deoxyribonucleosides was 400 intracellularly in the form of 5'-phosphates which do not readily
,uM at 37°C. This value represents an upper limit, since the traverse the plasma membrane. Thus the uptake of nucleosides
preparation was contaminated with phosphorylase. The relative by tissues is limited by the availability of specific phosphor-
V'max of deoxyinosine phosphorylation was only one-third that ylating enzymes.
of deoxyadenosine and deoxyguanosine. There was no evidence Because human adenosine kinase is distributed among many
of substrate inhibition at deoxyribonucleoside concentrations organs, plasma adenosine is not likely to be selectively taken up
up to 1 mM. by lymphoid tissues. In fact, under physiologic conditions most
Lymphocyte Growth Studies. The addition of deoxyade- adenosine produced intracellularly via nucleotidases and
nosine to peripheral lymphocyte cultures depressed their re- phosphatases may be rephosphorylated by the highly avid
sponse to phytohemagglutinin (Fig. 2). The toxicity of deoxy- adenosine kinase without reaching the extracellular environ-
adenosine was potentiated 1000-fold by inhibition of deaminase ment, since at low substrate concentrations the metabolism of
with EHNA. adenosine is a function of the relative Km of adenosine kinase
Not having a potent phosphorylase inhibitor or a phospho- (3 AM) and deaminase (25-40 AM) for adenosine (14, 21, 22).
rylase-deficient cell line, we could not study in isolation the There is no good evidence that deaminase functions normally
effects of deoxyguanosine and deoxyinosine on human lym- as a major pathway in the production of uric acid. In support
phocytes. However, the growth of the hypoxanthine phos- of this interpretation, deaminase-deficient children have normal
phoribosyltransferase-deficient human lymphoblastoid cell line uric acid levels in plasma and urine (23).
C1-35, although unaffected by either guanosine or inosine, was The ability to phosphorylate and trap deoxyadenosine, unlike
inhibited by deoxyguanosine and deoxyinosine (Table 2). Other adenosine, is largely confined to lymphoid tissues, whether or
studies showed that this cell line, as expected, had the ability not they are actively dividing, as is the newborn thymus, or are
to phosphorylate the purine deoxvribonucleosides but not the in a quiescent state, as is the adult peripheral blood lymphocyte.
ribonucleosides guanosine or inosine.
The toxicity of deoxyadenosine alone or deoxyadenosine plus Table 2. Effect of nucleosides upon the growth of a human
EHNA toward phytohemagglutinin-stimulated lymphocytes lymphoblastoid cell line deficient in hypoxanthine
was reversed by the addition of deoxycytidine but not uridine phosphoribosyltransferase
to the culture medium (Table 3), consistent with previous ex- Cells/ml X 10-4
periments by Reichard et al. (18) and Klenow (19). The in- Nucleoside 0 hr 24 hr 48 hr 72 hr
hibitory effects of deoxyguanosine and deoxyinosine on the
growth of human lymphoblastoid cell lines, however, were not None 5 9 27 72
reversed by deoxycytidine. Inosine, 2 mM 5 7 20 52
Inosine, 1 mM 5 10 25 65
DISCUSSION Deoxyinosine, 2 mM 5 5 7 8
Deoxyinosine, 1 mM 5 5 10 19
In the studies reported here, we have tried to explain not the Guanosine, 1 mM 5 7 16 40
detailed biochemical mechanisms by which deficiencies of Guanosine, 0.5 mM 5 8 38 94
adenosine deaminase and purine nucleoside phosphorylase Deoxyguanosine, 1 mM 5 4 5 8
produce immunodeficiency, but rather why both defects might Deoxyguanosine, 0.5 mM 5 3 12 25
preferentially impair lymphocyte function. Patterson has Cells were grown in minimum essential medium with 10% dialyzed
summarized evidence which suggests that purines may be ex- fetal calf serum. Nucleosides at the concentrations indicated were
changed between cells and tissues at the nucleoside level, via added at time zero. Comparable concentrations of hypoxanthine or
a process of facilitated diffusion (20). Subsequently the nucle- 0.1 mM guanine had no effect on cell growth (32).
5680 Immunology: Carson et al.
Table 3. Deoxycytidine reversal of deoxyadenosine toxicity during phytohemagglutinin-induced human lymphocyte transformation
Inhibitor
Deoxyadenosine, 1 mM
Deoxyadenosine, 10 AM + EHNA, 5 AM
No phytohemagglutinin
* Human peripheral blood lymphocytes were stimulated with phytohemagglutinin in the presence or absence of the indicated nucleosides. After
3 days the percent control leucine uptake into acid-precipitable material was determined as described in the legend ot Fig. 2. Six cultures were
examined in all cases.
t P values compare the means of the cultures with uridine or deoxycytidine with those containing deoxyadenosine alone by Students' t test.
NS, not significant.
However, the Km of 400 gAM for deoxyadenosine phosphoryl-
ation far exceeds the reported Km of 7 AiM for deamination (22).
Hence, when deaminase is present, deoxyadenosine is likely to
be deaminated rather than phosphorylated at low substrate
concentrations. A low rate of conversion of deoxyadenosine to
adenine ribonucleotides has also been reported (14). That the
tissue distribution of deaminase in general parallels that of
deoxyadenosine kinase may indicate a role for deaminase in the
prevention of deoxyadenosine phosphorylation. In deaminase
deficiency, on the other hand, deoxyadenosine may be excreted
by many tissues impaired in their ability to deaminate or
phosphorylate the nucleoside, only to become trapped in
lymphoid tissues in spite of a low plasma concentration.
Polmar et al. have reported a 10-fold increase in ATP content
of peripheral lymphocytes from a deaminase-deficient patient
as compared to those of normal controls, and that transfusion
of frozen irradiated erythrocytes decreased ATP concentrations
(4). This result conflicts with our hypothesis, which would not
predict the selective accumulation of adenine ribonucleotides
by lymphocytes. However, the enzymatic methods used by
Polmar et al. would not distinguish ATP from dATP. It is
therefore possible that, in fact, deoxyadenosine nucleotides and
not adenine ribonucleotides are elevated in deaminase-deficient
lymphocytes. Further investigations with more specific methods
should resolve this question.
Patients with phosphorylase deficiency show elevated con-
centrations of inosine, deoxyinosine, guanosine, and deoxy-
guanosine in their plasma and urine, thus providing further
evidence for the absence of guanosine kinase or inosine kinase
activity in human tissues (24). Consequently, phosphorylase
(and hypoxanthine phosphoribosyltransferase), may well be
required for the normal recycling of inosine and guanosine, a
fact that is reflected in the necessarily widespread tissue dis-
tribution of phosphorylase and the reported overproduction of
purines by phosphorylase-deficient patients (24).
Deoxyguanosine and deoxyinosine, unlike their ribonucle-
oside analogues, can be directly phosphorylated and trapped
by lymphocytes enriched in the appropriate kinase. As was true
for deoxyadenosine, here too the K111 of 400 ,M for phospho-
rylation compared to approximately 50 ,uM for phosphorolysis
(25) suggests that at physiologic substrate concentrations
phosphorylation is unlikely to be significant. In phosphorylase
deficiency on the other hand, deoxyguanosine and deoxyinosine
should be taken up by lymphocytes at an increased rate as well
as excreted into the urine unchanged.
Durham and Ives have previously shown that the murine
enzyme deoxycytidine kinase is confined to lymphoid tissues,
especially the thymus (26). These authors subsequently purified
Proc. Natl. Acad. Sci. USA 74 (1977)
% control leucine
Reversor, 100,uM uptake i SEM* Pt
None .35 i 1
Uridine 36 + 2 NS
Deoxycytidine 59 + :3 <0.05
None 19 ± :3
Uridine 26 i 2 NS
Deoxycytidine 70 + 5 <0.05
5+1
the calf thymus enzyme and showed that it could also phos-
phorylate deoxyadenosine and deoxyguanosine (27). Krygier
and Momparler, on the other hand, purified an activity from
calf thymus that phosphorylated deoxyadenosine and deoxy-
guanosine but not deoxycytidine (28). Recently, Krenitsky et
al. have studied in detail the substrate specificity of a partially
purified calf deoxycvtidine kinase that could phosphorylate
deoxyadenosine, deoxyinosine, and deoxyguanosine as well as
many synthetic substrates (29). Not yet having purified the
human enzyme to homogeneity, we cannot conclusively de-
termine its substrate specificity. However, in preliminary ex-
periments we found that human deoxycytidine phosphorylating
activity cochromatographed on DE52-cellulose with deoxy-
adenosine kinase.
The mechanisms by which trapped deoxyribonucleotides
might inhibit lymphocyte growth are unclear. Under certain
conditions deoxyribonucleotide triphosphates can be inhibitors
of ribonucleotide reductase (30, 31). Our studies showed that
deoxyadenosine and deoxyguanosine were indeed converted
by lymphocyte extracts to the respective triphosphates. How-
ever, deoxyinosine, which was converted only to the mono-
phosphate form, still inhibited cell growth, albeit at higher
concentrations than did deoxyguanosine.
Before enzyme replacement in the form of erthyrocyte
transfusion was used, the immunodeficient state associated with
deaminase deficiency had been successfully treated with bone
marrow or fetal liver transplantation (4, 33). When these
preferential methods fail or are unavailable, an attempt might
be made to modify the disease state at the molecular and cel-
lular level. As we have shown, under certain conditions in vitro,
deoxycytidine can partially reverse deoxyadenosine toxicity,
perhaps through competitive inhibition of deoxyadenosine
phosphorylation and "trapping". Whether or not deoxycytidine
is of value in the treatment of adenosine deaminase or purine
nucleoside phosphorylase deficiency disease remains to be es-
tablished. In any case, further investigations on the mechanisms
of deoxypurine toxicity seems warranted.
This study was supported by National Institutes of Health Grants
GM 23200, GM 17702, and AM 13622, grants from the Kroc Foun-
dation, National Foundation, and by Special Grant 797 from the
California Division of the American Cancer Society. D.A.C. is a Special
Fellow of the Leukemia Society of America.
1. Giblett, E. R., Anderson, J. E., Cohen, F., Pollara, B. & Meuwissen,
H. J. (1972) Lancet ii, 1067-1069.
2. Giblett; E. R., Amman, A.J., Sandman, R., Wara, D. W. & Dia-
mond, L. K. (1975) Lancet ii, 1010-1013.
3. Hirschhorn, R., Levvtska, V., Pollara, B. & Meuwissen, H. J.
(1973) Nature New Biol. 246, 200-202.
 Immunology: Carson et al.
4. Polmar, S. H., Wetzler, E. M., Stern, R. C. & Hirschhorn, R. C.
(1975) Lancet ii, 743-746.
5. Green, H. & Chan, T.-S. (1973) Science 182, 836-837.
6. Benke, P. J. & Dittman, D. (1976) Pediatr. Res. 10, 642-646.
7. Wolberg, G., Zimmerman, T. P., Hiemstra, K., Winston, M. &
Chu, L.-C. (1975) Science 187, 957-959.
8. Carson, D. A. & Seegmiller, J. E. (1976) J. Clin. Invest. 57,
274-282.
9. Boyle, W. (1968) Transplantation 6, 761-772.
10. Lowry, 0. H., Rosebrough, N. J., Farr, A. L. & Randall, R. J.
(1951) J. Biol. Chem. 193,265-271.
11. Ives, D. H., Durham, J. P. & Tucker, V. J. (1969) Anal. Biochem.
28, 192-205.
12. Crabtree, G. W. & Henderson, J. F. (1971) Cancer Res. 31,
985-991.
13. Randerath, K. & Randerath, E. (1964) J. Chromatogr. 16,
111-125.
14. Snyder, F. F. & Henderson, J. F. (1973) J. Biol. Chem. 248,
5899-5904.
15. Snyder, F. F., Mendelsohn, J. & Seegmiller, J. E. (1976) J. Clin.
Invest. 58, 654-666.
16. Lever, J. E., Nuki, G. & Seegmiller, J. (1974) Proc. Natl. Acad.
Sci. USA 71, 2679-2683.
17. Adams, A. & Harkness, R. A. (1976) Clin. Exp. Immunol. 26,
647-649.
18. Reichard, P., Canellakis, Z. N. & Canellakis, E. S. (1960) Biochim.
Biophys. Acta 41, 558-565.
19. Klenow, H. (1962) Biochim. Biophys. Acta 61, 885-890.
Proc. Natl. Acad. Sci. USA 74 (1977) 5681
20. Patterson, A. R. P., Kim, S. C., Barnard, 0. & Cass, C. E. (1975)
Ann. N.Y. Acad. Sci. 255,402-410.
21. Schnebli, H. P., Hill, D. L. & Bennet, L. L., Jr. (1967) 1. Biol.
Chem. 242, 1997-2004.
22. Agrawal, R. P., Sagar, S. M. & Parks, R. E., Jr. (1975) Biochem.
Pharrmacol. 24, 693-701.
23. Mills, G. C., Schmalsteig, F. C., Trimmer, K. B., Goldman, A. S.
& Golblum, R. M. (1976) Proc. Natl. Acad. Sci. USA 73,
2867-2871.
24. Cohen, A., Doyle, D., Martin, D. W., Jr. & Amman, A. J. (1976)
N. Engl. J. Med. 95, 1449-1454.
25. Kim, B. K., Chu, S. & Parks, R. E., Jr. (1968) J. Biol. Chem. 243,
1771-1776.
26. Durham, J. P. & Ives, D. H. (1969) Mol. Pharmacol. 5, 358-
375.
27. Durham, J. P. & Ives, D. H. (1970) J. Biol. Chem. 245, 2276-
2284.
28. Krygier, V. & Momparler, R. L. (1971) J. Biol. Chem. 246,
2745-2751.
29. Krenitsky, T. A., Tuttle, J. V., Koszalka, G. W., Chen, I. S.,
Beacham, L. M., III, Rideout, J. L. & Elion, G. B. (1976) J. Biol.
Chem. 251, 4055-4061.
30. Reichard, P. (1968) Eur. J. Biochem. 3, 259-266.
31. Moore, E. C. & Hurlbert, R. B. (1966) J. Biol. Chem. 241,
4802-4809.
32. Brenton, D. P., Astrin, K. H., Cruikshank, M. K. & Seegmiller,
J. E. (1977) Biochem. Med. 17,231-247.
33. Keightley, R. G., Lawton, A. R., Cooper, M. D. & Yunis, E. J.
(1975) Lancet ii, 850-883.