Purine Analogues
George H. Hitchings and Gertrude B. Elion
I . Introduction 215 I I . Cellular Multiplication 218
A . Bacteria 218 B. Protozoa and Algae 221
C. Viruses 221 D . Cells in Culture 222 I I I . Effects in the Whole Organism 222
A . Tissue Selectivity 222
B. Tumors 223 C. Embryonic Development and Differentiation 224
I V . Effects on Function 225 A . Glycolysis 225 B. Antibody Formation 225
C. Protein Synthesis 226 V . Biochemical Effects 226
A . Purine Catabolic Enzymes 226
B. Coenzymes 227 C. Incorporation into Nucleic Acid 227
D . Purine Anabolic Enzymes 229 E. Nucleic Acid Biosynthesis 230
F. Resistance 231 V I . Conclusion 232
References 232
I. INTRODUCTION
A purine analogue may be defined as a substance which differs chem
ically from the naturally occurring purines in a limited number of struc
tural features and exhibits antagonism to one or more of the natural purines in at least one biological system. Thus, there is a dual require
ment, chemical and biological, for admission to the class of analogues, which usually is implicit in practice though sometimes unrecognized.
215
Substances with antimetabolic effects against the natural purines may be formed either by modifications of the functional groups of the natural purines or by alteration of one or more atoms of the nucleus with reten
tion of the natural functional groups, or both changes may be represented in a single molecule. Several substances of the first type have been found to have antipurine activities of interesting types and dimensions [e.g., 6-mercaptopurine (IV), 2-amino-6-mercaptopurine (V), 2,6-diaminopurine (VI)], and a few of the second type also have been studied extensively [e.g., 8-azaguanine (VII),4-aminopyrazolo[3,4-rf]-pyrimidine (VIII)] (SeeFig. 1).
However, to date no potential antipurine which possesses both an abnormal nucleus and unnatural functional groups has had outstanding biological activity.
An extensive review of structure-activity relationships among purine analogues was prepared a few years ago (1), and although updating of this material undoubtedly would be in order, it is felt that for the pur
poses of the present volume, major emphasis ought to be placed upon the biological activities per se. This problem might be approached in either of two ways: (1) detailed description of the activities of the individual analogues, one by one, or (2) an attempt to categorize the biological and biochemical activities which the various purine analogues exhibit insofar as generalizations can be made.
Different analogues do in fact have many effects which are grossly similar. In broad terms, the biological activities of the purine analogues are consistent with expectation (2, 8). They inhibit the multiplication of microorganisms and the growth of tissues, they participate in many of the metabolic reactions of the purines, and at least one has an established role as a chemotherapeutic agent (4). In a general way they have a selective action on those cells which are undergoing the most rapid multiplication.
All these effects, no doubt, are the results of specific biochemical reactions.
Many of the latter have been demonstrated with clarity, but in one respect this body of knowledge remains unsatisfactory, or at least incomplete. One would like to be able to designate, among the many enzymic reactions which a given analogue inhibits, which site is critical for the observed biological activity. Is it the incorporation into the nucleic acids? Which nucleic acid? A coenzyme? The inhibition of glycolysis? The inhibition of purine interconversions? A still more difficult problem is the designation of the critical site or sites when a selective effect, such as that which occurs in chemotherapy, is involved. Here one must explain not only why one type of cell is inhibited, but why other types presumably exposed to the same inhibitor are not, or are inhibited only at a higher concentration of the analogue.
For a chapter of this size the chief problem is the selection of truly representative material. In the preparation of this chapter some 400
Ν - Ν " " Ν ' • u
1
ribose-5'- ribose-5'- ribose-5'- Ι
phosphate phosphate phosphate (ΧΠΙ) (xiv) (XV) FI G . 1. ( I ) Adenine; ( I I ) guanine; ( I I I ) hypoxanthine; ( I V ) 6-mercaptopurine; ( V ) thioguanine; ( V I ) 2,6-diaminopurine; ( V I I ) 8-azaguanine = 5-amino-7-hydroxy-t;- triazolo[4,5-d]pyrimidine; ( V I I I ) 4-aminopyrazolo[34-d]pyrimidine; ( I X ) benzimida- zole; ( X ) 2-azaadenine = 4-airdnoimidazo[4,5-d]-î>-triazine; ( X I ) imidazo[4,5-6]- pyridine; ( X I I ) kinetin = 6-furfurylaminopurine; ( X I I I ) inosinic acid; ( X I V ) succino- adenylic acid; ( X V ) succinothioinosinic acid.
NHa O H O H
W " « " Λ Α / S A /
Η Η Η
(ι) (π) (m)
S H S H NH2
N cï N > iV > r i >
S A / - - Ά Α / v - K A /
H H H ( I V ) ( V ) ( V I )
O H N H2
M> N sX> 0 5
H H H (vn) (vm) (ix)
N H2
ώ > Qc> itO
H H H ( X ) ( X I ) ( Χ Π )
Ç H2C O O H Ç H2C O O H
O H N H C H C O O H S — C H C O O H
1 xT I ·»• I
references to the recent literature were consulted. To cite completely even the most pertinent would require many times the allotted space in bibli
ography alone, but it is hoped that the references which have been selected can provide an introduction to the whole field. Several excellent reviews are available (5-8).
Emphasis will be placed on biological effects, progressing from gross general effects to specific biochemical effects, with an attempt to cite typical articles.
II. CELLULAR MULTIPLICATION
A. Bacteria
The inhibition of the multiplication of microorganisms was, historically, the first observed biological effect of any purine analogue and remains a basic source of information for the characterization of new analogues (2, 3y 9). A considerable variety of analogues show inhibitory activity.
Several with modified purine nuclei, benzimidazoles ( I X ) (10), imidazo- triazines ( X ) (11), and imidazopyridines ( X I ) (12), represent structures in which the imidazole nucleus is intact but the pyrimidine nucleus has been modified. Others, e.g., triazolopyrimidines (VII) (13) and pyrazolo- pyrimidines (VIII) (14, 15), represent structures in which an intact pyrimidine ring complete with "natural" functional groups is fused to a modified imidazole ring. A different type of analogue, structurally, is that in which the purine nucleus is intact but which possesses modified or un
natural functional groups. 2,6-Diaminopurine, 6-mercaptopurine, 6-thio- guanine, 6-ehloro-, and 6-methylpurines are all examples of this type. In biological effects the two types of analogues are essentially indistinguish
able. Each may show a competitive antagonism toward a natural pu
rine (16).
It is possible by means of inhibition, reversal, and incorporation studies with microorganisms to deduce a great deal concerning the mechanism of action of an analogue (16-20). Thus, 2,6-diaminopurine was identified as an adenine antagonist by inhibition and reversal studies using Lactobacillus casei (16) (Fig. 2), while studies of cross resistance (16) and the incorpora
tion of labeled purines in a resistant strain of the same organism established a decreased incorporation of adenine and its analogues as the major altera
tion in metabolism which confers resistance (19). Similarly, a 6-mercapto- purine-resistant strain of L. casei was shown to grow well on xanthine and guanine, poorly on adenine and its derivatives; and scarcely at all on hypoxanthine (18) (Fig. 3). It retained sensitivity to adenine analogues the effects of which were still counteracted by adenine specifically. These
facts led to the view that resistance was primarily the result of deletion of the mechanisms for the incorporation of hypoxanthine, and the suggestion was put forward that a hypoxanthine-containing metabolite might be an intermediate in the interconversion of adenine and guanine and that it might be one of the primary sites of action of 6-mercaptopurine (18).
These deductions have found support and confirmation through studies of other organisms, cells, and cell-free enzyme preparations to be discussed below.
The inhibitory effects of purine analogues are not confined to those species which have specific nutritive requirements for purines. It was sup
posed, correctly, that such organisms would have the mechanisms for the incorporation of the materials from exogenous sources (8)—mechanisms which might be induced to operate as well on substances which resembled
0.2 0.3 1.0 2.0 3.0
Purine (/ng/ml)
10
F I G . 2. Effect of purines on the inhibitory effects of 2,6-diaminopurine on the growth of Lactobacillus casei. The medium contains 0.046 m/ig/ml of folic acid and 100 Mg/ml of 2,6-diaminopurine. Curve C, the control without diaminopurine, is a composite of the curves obtained with the four natural purines. Growth is expressed in terms of acid production (ml of 0.1 Ν acid/10 ml culture). AD, adenine; GU, guanine; EX, hypoxan
thine; XA, xanthine (7).
Micromoles per ml
F I G . 3. Effects of purines and ribosides on the growth of the wild and 6-mercapto- purine-resistant ( 6- M P R ) strains of Lactobacillus casei in a medium containing 1 Mg/ml of thymine. AD, adenine; HX, hypoxanthine; GU, guanine; ADS, adenosine; HXS, hypoxanthine riboside (inosine) (4, p. 202).
the natural purines in sufficient detail. But it was scarcely to be anticipated that organisms and cells of wide variety and type which have no obvious need for exogenous purines also have such mechanisms and are similarly affected by analogues. Indeed, the first investigations on mammals, through an unlucky choice of precursor material (guanine), seemed to suggest that these organisms are unable to utilize exogenous purines (21, 22). Neverthe
less, early trials of analogues without regard to such considerations promptly demonstrated that organisms such as Escherichia coli which are inde
pendent of exogenous purines are quite as susceptible to the effects of analogues (10, 13) as those which require preformed purines. This is not to say that selectivity does not exist. Among the multitude of enzymic reactions involved in the anabolism of free purines into nucleosides, nucleotides, nucleic acids, and coenzymes and the interconversions of the purine moieties, it would be surprising if many microorganisms possessed the full complement of all possible reactions, and it is apparent that many do not; for example, the reactions necessary for the transformation of substances of the adenine series into those of the guanine series, and vice versa, may be possessed in toto by a minority of microorganisms; at least it is clear that, in many, one or more of such reactions may be rudimentary
or missing (28). All such individualities are potentially exploitable through the use of specific analogues where selective toxicity is the goal.
B. Protozoa and Algae
Purine analogues have been shown to inhibit the multiplication of a wide variety of microorganisms in addition to bacteria. Among the protozoa both pathogenic (24, 25) and nonpathogenic species (26) are represented.
Similarly, algae (27), phytoflagellates (28) f fungi (29), and higher plants (80) may show inhibition by analogues and reversal of the inhibition by specific metabolites. One analogue, kinetin (6-furfurylaminopurine, X I I ) , has rather dramatic effects on cell division in plants (81). Those effects are not obviously adenine-like or adenine-antagonistic. It seems possible that this substance, which is itself an artifact, may resemble some as yet un
discovered plant hormone.
C. Viruses
Purine antagonists are known to inhibit the multiplication of a number of viruses. 2,6-Diaminopurine was found early to inhibit the multiplica
tion of vaccinia virus (32, 33) and of spring-summer encephalitis virus (84) and Lansing polio virus (85) in tissue culture. Reversal of these effects by adenine could be demonstrated. Although there appeared to be some effect on the survival of mice infected with the encephalitis virus, this was not reflected in the virus titers of the tissues (86). Diaminopurine and 8-azaguanine were reported to inhibit the growth of psittacosis virus in the embryonic egg (37), but may have profound effects on the embryo itself (88). Benzimidazoles have been found to inhibit poliomyelitis and influenza viruses in tissue culture (89, 40), and a relationship of these to purine metabolism is suggested by reversal experiments (39). No practical therapeutic results have emerged from these studies, and in many the relationship between the effects of the analogue on the tissue and those on the virus itself is not very clear.
Clear-cut therapeutic effects against plant viruses have been obtained, however. Thus, 8-azaguanine produces "cures" when tobacco plants are inoculated mechanically with lucerne mosaic virus (41) (although failing to protect against natural transmission by the aphid) and retards infections with tobacco mosaic virus (42) and beet yellows virus (48). 2,6-Diamino
purine, 8-azaadenine, and 2-azaadenine ( X ) also were found effective against tobacco-mosaic virus in leaf culture (42). 8-Azaguanine inhibits the multiplication of phage in the lysogenic Bacillus megaterium (44)- Its effects could be correlated with its incorporation into ribonucleic acid, and are believed due to the inhibition of the synthesis of viral protein as a result (44)-
D. Cells in Culture
The effects of purine analogues on unicellular organisms are closely paralleled by their effects on mammalian cells growing in culture. In
hibition of cell division and of the outward growth of tissue expiants are accompanied by chromosome damage and mitotic aberrations (45). These effects are seen in a variety of tissues and may be caused by natural pu
rines as well as by their analogues (45). Reversal experiments often can be carried out in much the same way as with bacteria, e.g., the effects of 2,6-diaminopurine and 2-azaadenine are blocked by adenine (46, 4?) and those of 6-mercaptopurine by hypoxanthine, inosine, and inosinic acid, competitively (48). A prominent objective of a great deal of the work using tissue cultures has been to discover whether such testing could be used for the preliminary screening of antitumor agents, and thus attention has been turned to differential effects, i.e., the minimum inhibitory concen
tration acting on tumor cells as compared with that for normal cells [e.g. (49)]. It seems well established that the purine analogues which possess chemotherapeutic activity are highly cytotoxic (50), but to go further with such correlations is of dubious value in view of long experi
ence with attempts to correlate in vitro with in vivo results in many other fields of chemotherapy. The role of the host, which the study in vitro cannot assess, can be definitive (51).
III. EFFECTS IN THE WHOLE ORGANISM
A. Tissue Selectivity
In the whole animal, factors such as absorption, tissue distribution, kidney clearance, and metabolism influence the specific effects of a given analogue, often in rather unpredictable ways. Nevertheless, some common features of the action of a number of analogues can be observed, and some rather specific effects of individual substances have been reported. Philips et al. (52) have reported a careful comparison of the effects in mice, rats, and dogs of adenine and seven closely related purine analogues. Kidney damage is a feature of the action of adenine and its analogues, purine and 2-chloroadenine, due to the deposition in the renal tubules of insoluble oxidative products. Analogues such as 6-mercaptopurine also may be oxidized, but the end product [in this case 6-thiouric acid (58)] is more soluble, and kidney damage is not seen. Damage to bone marrow is a property common to most of these substances: 2,6-diaminopurine (64, 55), 6-mercaptopurine, 6-thioguanine, 6-chloropurine, purine, and 6-methyl- purine (52). Many of them also damage the intestinal epithelium and the
liver. Nevertheless, there are selective effects. Doses of thioguanine which severely depress bone marrow have only minor effects on the intestine [in animals {66), but perhaps not in man]. With both 2,6-diaminopurine (54, 55) and 6-mercaptopurine (52, 57), the first observable effect in the bone marrow is on the erythroid elements, followed later by depression of the granulocyte series.
B. Tumors
The chemotherapeutic activities of purine analogues on tumors may be regarded as a special case of tissue selectivity. Tumors, generally speaking, are rapidly dividing tissues, and it is not surprising to find them along with intestinal mucosa and bone marrow among the more sensitive tissues. Nevertheless, the rate of cell division is not the sole factor govern
ing sensitivity, as is obvious from the selective effects cited above and the wide range of effects—from strong to negligible—which an agent like 6-mercaptopurine may show on a spectrum of transplantable neoplasms growing in a single host species (58). Moreover, a successful chemothera
peutic agent, such as 6-mercaptopurine in acute leukemia, may give complete control of the disease (for a limited time) at essentially no cost in host toxicity (4). This represents a differential effect of a considerable magnitude. Many specific loci of inhibition have been discovered (these will be considered in Section V under Biochemical Effects), but despite a voluminous literature, no convincing evidence as to the source of this differential has been forthcoming.
Purine analogues have provided a great deal of stimulation for the ex
perimental cancer chemotherapists. Some 6% of such analogues show activity on one or more experimental tumors (69), something like twice the activity rate of compounds in general. One of the earliest to be sub
mitted for trial, 2,6-diaminopurine, was found active against mouse (60) and human (61) leukemias, but because of a low therapeutic index its use clinically was not pursued. Several analogues have shown considerable activity against experimental tumors, but have failed to be useful clin
ically. 8-Azaguanine (62-65) is among these, as is 4-aminopyrazolo[3,4-d]- pyrimidine (66), which produces severe liver damage and hemorrhage (67).
The only purine analogues which have proven useful clinically are 6-mer
captopurine, which accounts for the bulk of the use, its immediate con
geners, 6-thioguanine and 6-chloropurine, and some derivatives of these.
6-Mercaptopurine shows considerable selectivity among neoplastic diseases.
It is useful chiefly in the therapy of acute leukemia in childhood and chronic myelocytic leukemia (4). Thioguanine and 6-chloropurine appear to have more or less equivalent chemotherapeutic effects (68, 69).
A great many congeners of the active antipurines have been tested in tumor systems (70-73). By and large most of the modifications have led to diminished activity. The purine nucleosides are like the corresponding free purines (74), probably as a result of their rapid hydrolysis by nucle
osidases. Another group of derivatives of the thiopurines (^-heterocyclic) also have activities similar to those of the parent compounds (75-78). The latter were designed to afford protection from the catabolic disposition of the purines and thus assist in the delivery of the active substance to the tumor. Since these substances can be split both chemically and by reac
tion with —SH compounds in the tissues with release of the parent thio- purine (76, 79), it is presumed that the greater part of their activities are attributable to this release.
C. Embryonic Development and Differentiation
The developing embryo is a biological system which is highly sensitive to purine analogues. With the frog (80, 81) and chick embryo (82), the analogue may be directly applied, and with the frog, the later effects of brief exposure at specified stages of development can be observed (80).
Many abnormalities result from exposure to various analogues. However, most of these are relatively unspecific with respect to the agent applied and depend rather on the time of application; that is to say, specific tissues grow and differentiate in bursts of activity, and any of various inhibitors applied at such critical times may result in abnormalities affecting those tissues which are most actively growing (80). Tunicate (83), sea urchin (84, 85), and sand dollar (86) embryos have been used for similar studies.
In order to study separately the effects of analogues on cellular multi
plication and on differentiation per se, various regenerating systems, such as the tadpole tail (87), wound healing in the rabbit skin (88), and the aggregation and culmination of a slime mold (89), have been employed.
Perhaps somewhat akin is the prevention of the synthesis of the Kappa (killer) factor of Paramecium aurelia by 2,6-diaminopurine (90). In all these systems purine analogues have inhibited differentiation as well as multiplication, and in some (81, 90), reversal of the inhibitions by natural purines has been shown.
Purine analogues have been shown to affect mouse and rat embryos in utero at doses which may have little effect on the mother (91-94). The timing of drug administration is critical, the most effective time being just before or at the time of implantation (93). At later times various malforma
tions and anomalies may be produced (98, 94).
IV. EFFECTS ON FUNCTION
A. Glycolysis
Chronic treatment with 6-mercaptopurine causes a reduction of 50% or more in the respiration and anaerobic glycolysis of slices of sarcoma 180 without impairing the metabolism of either liver or kidney (95). This de
fect was associated with a diminution of succinic dehydrogenase activity, which did not, however, appear to be of sufficient magnitude to account for the gross defect in metabolism. Similarly, 8-azaguanine appears to stimulate the glycolysis of suspensions of leukemic cells while depressing respiration (96), whereas 6-mercaptopurine and thioguanine inhibit both glycolysis and respiration (97, 98). It is not yet possible to interpret these effects in terms of antagonism to purine metabolism.
B. Antibody Formation
Purine analogues may exert rather striking effects on the immune re
sponse. The reduction of hemolysin and precipitin antibody titers in mice by the administration of 8-azaguanine and a partial reversal of these effects by the concomitant administration of guanylic acid (but not gua
nine) was reported a number of years ago (99). However, more recent observations of the effects of 6-mercaptopurine on the immune response (100) have provided the impetus for greatly expanded interest in this field.
Therapy of rabbits with 6-mercaptopurine during the period of admin
istration of foreign serum albumin results in immune tolerance to this antigen but does not interfere when given at the height of the antibody response (101). Thus, the inhibition seems to be concerned primarily with the "tooling-up" process rather than with antibody production per se.
These observations suggest a number of practical applications but, per
haps more important, may ultimately yield information bearing on theories of the mechanisms of immunity, i.e., a decision between the "informa
tional" and "selective" (102) types of mechanism. The administration of 6-mercaptopurine influences a number of allergic-immunological phe
nomena. It produces suppression of experimental allergic encephalo
myelitis (103), small but significant extensions of skin homograft survival in rabbits (10^-106), rather striking prolongation of the survival of renal homografts in dogs (107) and rabbits (108), and significant effects in various "autoimmune" diseases (109). The effects in these more complex situations add up to a good deal less than the "immune tolerance" which was observed with a single antigenic stimulus (100) but nevertheless provide an entree into many heretofore unexplored areas.
C. Protein Synthesis
The effects of purine analogues on the immunological response may be a special case of the disturbance of protein synthesis which has been known for some years. 8-Azaguanine (110), 2,6-diaminopurine, and 6-mercaptopurine (111) interfere with the formation of adaptive enzymes by microorganisms. With 8-azaguanine in Bacillus cereus it could be shown that this inhibition occurred under conditions which did not ap
preciably affect the synthesis of certain constitutive enzymes or D N A (110). However, inhibition of general protein synthesis with the exception of cell wall proteins has been reported (112-115), and it is probable that the selective effects on the synthesis of induced enzymes may depend on the selection of particular levels of the inhibitor. Two suggestions have been put forward as to the site of this action. Since guanosine triphosphate has a cofactor role in protein synthesis, 8-azaguanosine triphosphate might be expected to have such an effect (116), but no evidence for its formation has been put forward. On the other hand, there is ample evidence for the incorporation of 8-azaguanine into RNA, as will be discussed below, and one group of workers believes the effects to be mediated through an ab
normal analogue-containing R N A . This abnormal R N A appears to be less stable in the organism than ordinary R N A (117) and behaves differently toward phenol extraction (115), which may eventually lead to a direct test of its ability to participate in protein synthesis.
V. BIOCHEMICAL EFFECTS
An understanding of the mechanism (s) of action of an antimetabolite rests ultimately on the delineation of its effects on specific biochemical reactions. There is now available a great deal of information concerning the effects of a number of purine analogues; their incorporation into nucleic acids and a variety of metabolites has been demonstrated. In a number of instances inhibitory effects of these abnormal metabolites on specific enzymic reactions have been shown. Nevertheless, there remains to be discovered, in any instance, the reason for the selective effects which some analogues indubitably show.
A. Purine Catabolic Enzymes
Purine analogues participate in many of the enzymic reactions of the purines. 8-Azaguanine is deaminated by guanase (118) in vitro, and this reaction appears to be a major pathway of elimination of the analogue in
the mouse and monkey (119). Its antitumor effects may depend to a con
siderable extent on the relatively low guanase activities of the sensitive tumors (120). 2,6-Diaminopurine riboside is deaminated by adenosine deaminase (121). Κ host of purine (and pyrimidine) analogues are attacked by xanthine oxidase (61, 122, 123). A number of these compete with natural substrates for the enzyme and show inhibitory effects (122, 124, 126) which apparently can be demonstrated in vivo (126) as well as in vitro. 6-Chlorouric acid (formed by the action of xanthine oxidase on 6-chloropurine) (124) and 8-azaguanine and 8-azaxanthine (127) are inhibitors of uricase. 6-Mercaptopurine and thioguanine are converted in vivo to 6-thiouric acid, which is excreted even by species which catabolize uric acid to allantoin (128, 129). It is clear that the analogues are catab- olized and excreted extensively and that these properties in many in
stances limit the physiological activities (130).
B. Coenzymes
The possibility that purine analogues interfere with the formation or functions of essential coenzymes or become involved via the synthesis of a fraudulent coenzyme has been explored to a limited extent. Both thio
guanine and 6-mercaptopurine affect the mitosis of cells in tissue culture, and these effects are preventable by the addition of coenzyme A (131).
6-Mercaptopurine interferes with the acetylation of sulfanilamide both in vivo and in vitro and appears to inactivate coenzyme A (132), but it is not clear that these observations are necessarily an explanation of the effects observed in tissue culture. When nicotinamide is administered to mice, the D P N content of the liver rises markedly over the next several hours and subsequently falls to pretreatment levels. Both the rise and the fall are inhibited in 6-mercaptopurine-treated mice (138).
C. Incorporation into Nucleic Acid
1. RIBONUCLEIC ACID ( R N A )
The incorporation of 8-azaguanine into the R N A of tobacco mosaic virus ( T M V ) and B. cereus is well established. T M V was isolated from plants treated with 8-azaguanine; the R N A was prepared from it and subjected to alkaline hydrolysis. The presence of the isomeric 8-aza- guanosine-2'- and -3'-phosphates in these hydrolyzates was demonstrated
(134). With B. cereus, up to 40% of the RNA-guanine may be replaced by 8-azaguanine (135). An examination of such nucleic acid revealed that the cyclic nucleotides (2',3'-cyclic phosphates), which appear to be ter-
minai groups, contained a high proportion of 8-azaguanine. This suggests the possibility that the presence of an 8-azaguanine residue blocks or inhibits the progression of the chain. Guanine is replaced by 8-azaguanine in the RNA's of a variety of systems, both plant and animal; in most of these, the replacement is of the order of 1-3% of the guanine. The in
corporation into animal nucleic acids usually is small and the identification circumstantial (136, 187). Thus, when C
14
-labeled 8-azaguanine was given to tumor-bearing mice, it could be shown that C
14
was present in the nucleotide fraction which was prepared by alkaline hydrolysis of the R N A (187), but the amount was too small for further identification. Similarly, there is presumptive evidence for the incorporation of 6-mercaptopurine-S
35 into the R N A of tumors (188), but the radioactivity found in the R N A fraction was equivalent to only one mercaptopurine unit in 8000 of the purine fraction, an amount too small for unequivocal characterization.
There appears to be a somewhat larger incorporation of thioguanine into the R N A of rat and mouse tissues and tumors (189). The nucleic acids were extracted by sodium chloride solution, the R N A was degraded by alkaline hydrolysis, and the nucleotide fraction so obtained was chromato- graphed on Dowex-1. A peak of radioactivity in the nucleotide fractions was observed which occupied the position expected for the nucleoside-2'- and -3'-phosphates (139). This all adds up to a presumption of incorpora
tion of the abnormal bases into the RNA's of mammalian as well as bac
terial cells, but a definitive characterization of these materials has not yet been possible with mammalian tissue.
2. DEOXYRIBONUCLEIC ACID ( D N A )
The incorporation of purine analogues into D N A is generally small or negligible. B. cereus, which replaces up to 40% of its RNA-guanine with 8-azaguanine, replaces less than 1% of the DNA-guanine (136). Further
more, a number of bacterial strains and species which are capable of the replacement of major fractions of the DNA-thymine by its analogues show little or no abnormal DNA-purine when exposed to 8-azapurines under similar conditions (186). There is presumptive evidence for the in
corporation of small quantities of 6-mercaptopurine into mammalian DNA's (128). In fact, a somewhat larger amount of this analogue appears to enter D N A than R N A (138). However, the identification of this mate
rial has proceeded only so far as to show that S 35
is present in purified D N A (Kirby technique) after the administration of 6-mercaptopurine-S
35 to tumor-bearing animals. Similarly, the presence of thioguanine-C
14 in the D N A of Ehrlich ascites tumor cells can be inferred from the finding of a radioactive fraction in the deoxynucleotide fraction from a Dowex-1 column (189).
The importance as a mechanism of growth inhibition of the incorpora
tion of an unnatural purine into either or both nucleic acids is largely con
jectural. With tobacco mosaic virus, the replacement of guanine by 8-aza
guanine appeared to reduce the infectivity of the virus (184). This case comes closest to a demonstration that incorporation into a nucleic acid (in this case R N A ) is per se responsible for the inhibitory effect of the analogue, for a major and perhaps the most important structure of the virus is the nucleic acid. But even this example falls short of an unequivocal demonstration of the point, for various constituent proteins are essential to the infective mechanisms of the virus, and 8-azaguanine is known to interfere with protein synthesis (140) and might therefore be inhibiting the formation of some essential structure of the virus other than its nucleic acid. The incorporation of 8-azaguanine into the R N A of B. cereus is ac
companied by the formation of equivalent "extra" quantities of adenylic, cytidylic, and uridylic acids, and therefore the 8-azaguanine-containing R N A appears to be formed over and above the usual cellular complement of RNA. One might infer from this that it is metabolically inert, but it has not yet been possible to design a direct test of this assumption. Finally, the finding that 6-mercaptopurine is incorporated equally well into the DNA's of sensitive and nonsensitive tumors (188) suggests that incor
poration into nucleic acids per se may not provide a final identification of the locus of the inhibitory effects of the analogue.
The finding of purine analogues in the nucleic acids implies their par
ticipation in a number of the anabolic reactions of the normal purines, since purines apparently enter the nucleic acids only by condensation reac
tions involving the nucleoside di- [for R N A (141, 14®)]
or
triphosphate [for R N A (142a, 142b) and for D N A (142, 148)]. (A theoretically possible alternative, direct base exchange, has found no experimental support to date.)
D. Purine Anabolic Enzymes
The synthesis of abnormal nucleosides by the reaction of 8-azaguanine (144)y 6-mercaptopurine (145), and 2,6-diaminopurine (146) with pen
tose-1-phosphates in the presence of nucleoside phosphorylases has been demonstrated. The resulting nucleosides can give rise to nucleotides by phosphorylation reactions (121, 147). Alternatively, and perhaps more important, there exists the reaction of the analogue with phosphoribosyl- pyrophosphate (PRPP) under the influence of a nucleotide pyrophos- phorylase to form the riboside-5'-phosphate directly. This has been shown for 6-mercaptopurine with inosinic acid pyrophosphorylase from both beef liver (148) and E. coli (149), and for this and several other analogues using
hog liver enzyme (150). 2-Fluoroadenine, the 2- and 8-azaadenines, and 4-aminopyrazolo[3,4-d]pyrimidine have been shown to be substrates for adenylic acid pyrophosphorylase (151). The conversion of the monophos
phate derivatives to di- and triphosphates by adenosine triphosphate (ATP) under the influence of a hog kidney preparation (152) has been shown. The participation of an abnormal base in these reactions may be entirely comparable to that of a normal base; for example, the reaction of 6-mercaptopurine and PRPP as catalyzed by the inosinic acid pyro
phosphorylase from E. coli is as rapid as that of hypoxanthine and greater than that of guanine (14$),
a n
d the synthesis of thioinosinic acid by ascites tumor cells is rapid in vivo (153). It seems probable that this an- abolism is an essential prerequisite to many important biological effects;
indeed, as will be discussed below, drug resistance is often accompanied by deletion (or impairment) of an essential nucleotide phosphorylase, i.e., in the absence of the nucleotide-forming enzyme, the analogue loses its activity. The possibility that an analogue may compete with normal purines for PRPP or a phosphorylase or pyrophosphorylase has received little attention, although these appear to be potential sites of inhibition.
E. Nucleic Acid Biosynthesis
There is a good deal of evidence that purine analogues not only compete with free purines for incorporation, but that they also, and perhaps more significantly, interfere with the synthesis de novo of the purine moieties of the nucleic acids. This can be documented for a variety of analogues, several precursors, and many types of tissue. Thus, 6-mercaptopurine, 2,6-diaminopurine and benzimidazole (164), 6-chloropurine (155) and thioguanine (156) all have been reported to inhibit the biosynthesis of nucleic acids when glycine-C
14
is used as precursor. Similarly, the in
corporation of formate-C 14
into the nucleic acids of Flexner-Jobling sar
coma and spleen is inhibited by 2,6-diaminopurine and 8-azaguanine (157), and into the nucleic acids of adenocarcinoma 755 by 4-amino- pyrazolo[3,4-d]pyrimidine (158). The uptake of Ρ
3 2 θ4
Ξ
into the R N A of tumors likewise is reduced by a number of purine analogues (169, 160).
Despite all this evidence of interference with the de novo pathways, it has been reported that neither 6-mercaptopurine nor its ribonucleotide inter
feres with the synthesis of inosinic acid ( X I I I ) by soluble pigeon liver enzymes (148). It is pertinent, therefore, to look beyond inosinic acid, to its transformation into adenylic and guanylic acids and their utilizations, for a site of action of 6-mercaptopurine (ribonucleotide). Moreover, such a locus of action would provide an explanation of its interference both with synthesis de novo and the utilization of exogenous purines. In fact, 6-mer-
captopurineriboside-5 phosphate (thioinosinic acid) inhibits in cell-free systems the conversion of inosinic acid both to succinoadenylic acid ( X I V )
(161, 162) and to xanthylic acid (161), thus interfering with the formation of both adenylic and guanylic acids. Furthermore, both thioinosinic acid and succinothioinosinic acid ( X V ) (162) inhibit the cleavage of succino
adenylic acid by the enzyme adenylosuccinase (162), and the succino analogue also blocks the cleavage of 5-aminoimidazole-iV'-succino-car- boxamide ribotide by this same enzyme (163).
An antimetabolite may not only compete with a metabolite but also interfere with its formation. Thus, a number of purine analogues have been shown to exert feedback control of purine biosynthesis in much the same manner as the "normal" purines. 6-Mercaptopurine, thioguanine, and 2,6-diaminopurine strongly repressed synthesis de novo by a strain of E. coli, while 8-azaadenine, 8-azaguanine, 8-azahypoxanthine, and 4-amino- pyrazolopyrimidine exerted moderate effects (163a). Thus, the antipurines, by inhibition of both the formation and the utilization of the metabolites which they affect, may act at several points sequentially on the same pathway.
F. Resistance
Drug resistance is a feature of the action of antimetabolites which limits their usefulness as chemotherapeutic agents, but provides, through bio
chemical studies of the resistant cells, some important clues concerning their mechanisms of action. Thus, a prominent, though perhaps not uni
versal, feature of purine analogue-resistant cells is their failure to convert the analogue (and the corresponding metabolite) to the nucleotide (16If- 172). Cells of leukemia LI 210 which are resistant to 6-mercaptopurine are deficient in the ability to convert either it or hypoxanthine to the corre
sponding nucleotide but retain the ability to convert adenine to adenylic acid (170). Similarly, resistance to 2,6-diaminopurine by cells in tissue culture involves loss of adenylic acid pyrophosphorylase (171), and neo
plasms resistant to 8-azaguanine are deficient in the ability to form both guanylic and 8-azaguanylic acids (172). Such studies are of value not only in defining a mechanism of resistance, but in their support for the identi
fication of the analogue ribonucleotide as the primary inhibitor.
Deletion or impairment of the nucleotide pyrophosphorylase reaction has been demonstrated so widely that there is a tendency to regard this as an all-encompassing mechanism of resistance to purine analogues. How
ever, several different 6-mercaptopurine-resistant strains of Streptococcus faecalis involve a variety of mechanisms (173), one of which is desulfur-
ization. The resistance of a strain of Ehrlich ascites cells to thioguanine has been attributed to increased catabolism of the analogue (174). Failure
of a cell to form the nucleotide may not depend on a lack of the appropriate pyrophosphorylase (175), although this conclusion is often drawn without an examination of the enzymic activities of cell-free preparations. Finally, the apparent incorporation of isotopically labeled 6-mercaptopurine into the D N A of a resistant line of adenocarcinoma 755, to an extent at least equal to that of the sensitive line, suggests that in this instance a different mechanism of resistance must be sought (176).
VI. CONCLUSION
Finally, it seems permissible to philosophize a bit on the subject of enzymic specificity and the revelations which studies of antimetabolites have brought forth. These investigations suggest a radical revision of the concept of the enzyme as a highly specific catalyst which accepts only its
"natural" substrate and in all-or-none fashion rejects all others. Enzymic specificity probably rests in a major way on opportunity; it is determined by the integration of all the various imperfect specificities which decide the ultimate fates of natural materials and in the end usually allow only the "natural" substrate to reach its enzyme in significant quantities.
Similarly, the over-all effects of an antimetabolite may have to be regarded as an integrative function of all the many reactions in which it participates;
in this one it behaves quite like a natural substrate, in that it is only loosely bound to the enzyme and thus easily displaced by the natural metab
olite, while in a third it may be tightly bound and highly inhibitory. Thus, for example, it seems essential to the action of 6-mercaptopurine that it be a good substrate for inosinic acid pyrophosphorylase. The resulting nucleo
tide, in all probability, not only represses the biosynthesis of inosinic acid through feedback control but also interferes with its conversion to succino- adenylic acid, hydrolysis of the latter, and oxidation of inosinic acid to xanthylic acid. Quite apart then from the possibility of the incorporation of thioinosinic acid into nucleic acids and essential coenzymes, it is clear that 6-mercaptopurine may act sequentially on several steps in the bio
synthesis of the natural nucleotides. It is probable that it is this multi
plicity of loci and their sequential arrangement which are responsible for the over-all effects of the analogue. It is tempting to speculate that powerful effects are exhibited only by those analogues which become involved in a number of biosynthetic reactions.
R E F E R E N C E S
1. G.H.HitchingsandG.B.Elion,Proc.3rdIntern.Congr.Biochem.,BrusseU, 1966,p.55.
2. G. H . Hitchings, G. B . Elion, E. A . Falco, P. B . Russell, M . B . Sherwood, and H . VanderWerff, J. Biol. Chem. 183,1 (1950).
3. G. H . Hitchings, G. B. Elion, Ε. Α . Falco, P. Β. Russell, and H . Vander Werff, Ann. Ν. Y. Acad. Sci. 52, 1318 (1950).
4. G. H . Hitchings and C. P. Rhoads, Ann. Ν. Y. Acad. Sci. 60, 183 (1954).
5. R . E. Handschumacher and A . D . Welch, in ' T h e Nucleic Acids" (E. Chargaff and J. N . Davidson, eds.), Vol. I l l , p. 453. Academic Press, N e w York, 1960.
6. W . Shive and C. G. Skinner, Ann. Rev. Biochem. 27, 643 (1958).
7. S. C. Hartman and J. M . Buchanan, Ann. Rev. Biochem. 28, 365 (1959).
8. E. P. Anderson and L . W . Law, Ann. Rev. Biochem. 29, 577 (1960).
9. G. B . Elion, G. H . Hitchings, and H . VanderWerff, / . Biol. Chem. 192, 505 (1951).
10. D . W . Woolley, J. Biol. Chem. 152, 225 (1944).
11. D . W . Woolley and E. Shaw, J. Biol. Chem. 189, 401 (1951).
12. G. W . Kidder and V . C. Dewey, Arch. Biochem. Biophys. 66, 486 (1957).
13. R . O. Roblin, Jr., J. O. Lampen, J. P. English, Q. P. Cole, and J. R . Vaughan, Jr., J. Am. Chem. Soc. 67, 290 (1945).
14. R . K . Robins, / . Am. Chem. Soc. 78, 784 (1956).
15. E. A . Falco and G. H . Hitchings, / . Am. Chem. Soc. 78, 3143 (1956).
16. G. B . Elion, S. Singer, G. H . Hitchings, M . E. Balis, and G. B . Brown, Biol.
Chem. 202, 647 (1953).
17. G. B . Elion and G. H . Hitchings, J. Biol. Chem. 187, 511 (1950).
18. G. B . Elion, S. Singer, and G. H . Hitchings, J. Biol. Chem. 204, 35 (1953).
19. G. B . Elion, H . VanderWerff, G. H . Hitchings, M . E. Balis, D . H . Levin, and G. B . Brown, Biol. Chem. 200, 7 (1953).
20. D . J. Hutchison, Ann. Ν. Y. Acad. Sci. 60, 212 (1954).
21. A . A . Plentl and R . Schoenheimer, J. Biol. Chem. 153, 203 (1944).
22. M . E. Balis, D . H . Marrian, and G. B . Brown, J. Am. Chem. Soc. 73, 3319 (1951).
23. G. B . Brown and P. M . Roll, in "The Nucleic Acids" (E. Chargaff and J. N . Davidson, eds.), Vol. I I , p. 341. Academic Press, N e w York, 1955.
24. M . Nakamura and M . B . James, Exptl. Parasitol. 2,19 (1953).
25. M . Nakamura and S. Jonsson, Arch. Biochem. Biophys. 66, 183 (1957).
26. G. W . Kidder and V . C. Dewey, J. Biol. Chem. 179, 181 (1949).
27. P . Arnow, A . C. Sampath, J. J. Oleson, and J. H . Williams, Am. J. Botany 39, 151 (1951).
28. F. Moewus, Science 130, 921 (1959).
29. R . Fuerst, C. E. Somers, and T . C. Hsu, Bacteriol. 72, 387 (1956).
30. R . S. de Ropp, Cancer Research 11, 663 (1951).
31. F. M . Strong, "Topics in Microbial Chemistry," p. 125. Wiley, N e w York, 1958.
32. R . L . Thompson, M . L . Wilkin, G. H . Hitchings, G. B . Elion, E. A . Falco, and P. B. Russell, Science 110, 454 (1949).
33. R . L . Thompson, M . L . Price, S. A . Minton, G. B . Elion, and G. H . Hitchings, / . Immunol. 65, 529 (1950).
34. C. Friend, Proc. Soc. Exptl. Biol. Med. 78, 150 (1951).
35. G. C. Brown, / . Immunol. 69, 441 (1952).
36. A . E . Moore and C. Friend, Proc. Soc. Exptl. Biol. Med. 78, 153 (1951).
37. H . R . Morgan, Exptl. Med. 95, 277 (1952).
38. J. S. Younger and J. E. Salk, Cancer Research 10, 250 (1950).
39. A . C. Hollinshead and P . K . Smith, J. Pharmacol. Exptl. Therap. 123, 54 (1958).
40. I . Tamm, J. Bacteriol. 72, 42 (1956).
41. R . E. F. Matthews, J. Gen. Microbiol. 8, 277 (1953)..
42. I . R . Schneider, Phytopathology 44, 273 (1954).
43. G. E. Russell and À . R . Trim, Nature 179, 151 (1957).
44. R . Jeener, C. Hamer-Casterman, and N . Mairesse, Biochim et Biophys. Acta 35, 166 (1959).
45. J. J. Biesele, R . E. Berger, M . Clarke, and L . Weiss, Exptl. Cell Research Suppl. 2, 279 (1952).
46. J. J. Biesele, R . E. Berger, A . Y . Wilson, G. H . Hitchings, and G. B . Elion, Cancer 4, 186 (1951).
47. J. J. Biesele, Cancer 5, 787 (1952).
48. M . T . Hakala and C. A . Nichol, Biol. Chem. 234, 3224 (1959).
49. J. J. Biesele, R . E. Berger, and M . Clarke, Cancer Research 12, 399 (1952).
50. G. J. Dixon, H . E. Skipper, and F. M . Schabel, Jr., Federation Proc. 19,142 (1960).
51. G. B . Elion, S. W . Callahan, G. H . Hitchings, and R . W . Rundles, Proc. Am. Assoc.
Cancer Research 4, 222 (1961).
52. F. S. Philips, S. S. Sternberg, L . Hamilton, and D . A . Clarke, Ann. N. Y. Acad.
Sci. 60, 283 (1954).
53. G. B. Elion, S. Mueller, and G. H . Hitchings, Am. Chem. Soc. 81, 3042 (1959).
54. G. E. Cartwright, J. G. Palmer, G. H . Hitchings, G. B . Elion, F. D . Gunn, and M . M . Wintrobe, / . Lab. Clin. Med. 35, 518 (1950).
55. F. S. Philips and J. B. Thiersch, Proc. Soc. Exptl. Biol. Med. 72, 401 (1949).
56. F. S. Philips, S. S. Sternberg, L . D . Hamilton, and D . A . Clarke, Cancer 9, 1092 (1956).
57. J. S. Latta and R . P. Gentry, Anat. Record 132, 1 (1958).
58. K . Sugiura, Cancer Research Suppl. 3, 18 (1955).
59. G. B. Brown, "Antimetabolites and Cancer," p. 285. A m . Assoc. Advance. Sci., Washington, 1955.
60. J. H . Burchenal, A . Bendich, G. B. Brown, G. B. Elion, G. H . Hitchings, C. P.
Rhoads, and C. C. Stock, Cancer 2, 119 (1949).
61. J. H . Burchenal, D . A . Karnofsky, E. M . Kingsley-Pillers, C. H . Southam, W . P . L . Myers, G. C. Escher, L . F. Craver, H . W . Dargeon, and C. P . Rhoads, Cancer 4, 549 (1951).
62. G. W . Kidder, V . C. Dewey, R . E. Parks, Jr., and G. L . Woodside, Science 109, 511 (1949).
63. C. C. Stock, L . F. Cavalieri, G. H . Hitchings, and S. M . Buckley, Proc. Soc. Exptl.
Biol. Med. 72,565 (1949).
64. K . Sugiura, G. H . Hitchings, L . F. Cavalieri, and C. C. Stock, Cancer Research 10, 178 (1950).
65. A . Gellhorn, Cancer 6, 1030 (1953).
66. H . E. Skipper, R . K . Robins, J. R . Thomson, C. C. Cheng, R . W . Brockman, and F. M . Schabel, Jr., Cancer Research 17, 579 (1957).
67. R . K . Shaw, R . N . Shulman, J. D . Davidson, D . P. Rail, and E. Frei, Cancer 13, 482 (1960).
68. D . R . Hales, R . W . Jerner, B. E. Hall, F. M . Willett, J. Franco, and T . V . Feicht- meir, Clin. Research Proc. 5, 32 (1957).
69. R . R . Ellison and J. H . Burchenal, Clin. Pharmacol. Therap. 1, 631 (1960).
70. G. B. Elion, Proc. Roy. Soc. Med. 50, 7 (1957).
71. D . A . Clarke, G. B. Elion, G. H . Hitchings, and C. C. Stock, Cancer Research 18, 445 (1958).
72. H . E. Skipper, J. A . Montgomery, J. R . Thomson, and F. M . Schabel, Jr., Cancer Research 19, 425 (1959).
73. J. A . Montgomery, Cancer Research 19, 447 (1959).
74. Η . Ε . Skipper, J. R . Thomson, D . J. Hutchison, F. M . Schabel, Jr., and J. J.
Johnson, Jr., Proc. Soc. Exptl. Biol Med. 95, 135 (1957).
75. G. H . Hitchings and G. B . Elion, Proc. Am. Assoc, Cancer Research 3, 27 (1959).
76. G. B . Elion, S. Bieber, and G. H . Hitchings, Cancer Chemotherapy Repts. No. 8, 36 (1960).
77. O. S. Selawry et al, Cancer Chemotherapy Repts. No. 8, 56 (1960).
78. R . W . Rundles, T . E. Fulmer, R . T . Doyle, and T . W . Gore, Cancer Chemotherapy Repts. No. 8, 66 (1960).
79. G. B. Elion, S. W . Callahan, G. H . Hitchings and R . W . Rundles, Cancer Chemo
therapy Repts. No. 8, 47 (1960).
80. S. Bieber, Cellular Comp. Physiol. 44, 11 (1954).
81. S. Bieber, R . Bieber, and G. H . Hitchings, Ann. N. Y. Acad. Soi. 60, 207 (1954).
82. D . A . Karnofsky, Cancer Research Suppl. 3, 83 (1955).
83. C. L . Markert, Anal Record 113, 554 (1952).
84. Ε. E. Palincsar, Biol. Bull. 117, 436 (1959).
85. T . Gustafson and S. Hôrstadius, Exptl. Cell Research Suppl. 3, 170 (1955).
86. D . A . Karnofsky, Proc. Am. Assoc. Cancer Research 3, 124 (1960).
87. S. Bieber and G. H . Hitchings, Cancer Research 19, 112 (1959).
88. B. Cooperman and E. L . Howes, Surg. Gynecol. Obstet. 92, 105 (1951).
89. E. Hirschberg and G. Merson, Cancer Research Suppl. 3, 76 (1955).
90. M . Williamson, W . Jacobson, and C. C. Stock, J. Biol. Chem. 197, 763 (1952).
91. G. L . Woodside, Anal Record 113, 551 (1952).
92. J. B . Thiersch, Ann. Ν. Y. Acad. Sci. 60, 220 (1954).
93. J. B . Thiersch, Proc. Soc. Exptl. Biol. Med. 94, 40 (1957).
94. H . Tuchmann-Duplessis and L . Mercier-Parot, Compt. rend. soc. biol. 153, 1133 (1959).
95. E. Mihich, D . A . Clarke, and F. S. Philips, Proc. Soc. Exptl. Biol. Med. 92, 758 (1956).
96. D . Burk, J. Laszlo, J. Hunter, K . Wight, and M . Woods, J. Natl. Cancer Inst. 24, 57 (1960).
97. J. Laszlo, J. Stengle, K . Wight, and D . Burk, Proc. Soc. Exptl Biol. Med. 97, 127 (1958).
98. J. Laszlo, C. Ellis, R . W . Rundles, and G. B . Elion, Proc. Am. Assoc. Cancer Re
search 4, 243 (1961).
99. R . A . Malmgren, Β . E. Bennison, and T . W . McKinley, J. Natl. Cancer Inst. 12, 807 (1952).
100. R . Schwartz and W . Dameshek, Nature 183, 1682 (1959).
101. R . Schwartz, J. Stack, and W . Dameshek, Proc. Soc. Exptl. Biol. Med. 99, 164 (1958).
102. F. M . Burnet, Perspectives in Biol. Med. 3, 447 (1960).
103. L . W . Hoyer, R . M . Condie, and R . A . Good, Proc. Soc. Exptl. Biol. Med. 103, 205 (1960).
104. R . Schwartz and W . Dameshek, J. Clin. Invest. 39, 952 (1960).
105. W . Meeker, R . Condie, D . Weiner, R . L . Varco, and R . A . Good, Proc. Soc. Exptl Biol.. Med. 102, 459 (1959).
106. J. L . Robinson and C. L . Christian, Nature 187, 796 (1960).
107. R . Y . Calne, Lancet I , 417 (1960).
108. P . Nathan, E. Gonzalez, H . Pescovitz, R . Fowler, Jr., and B . F. Miller, Federation Proc. 19, 216 (1960).
109. W . Dameshek, Abstr. 8th Intern. Soc. Hematol, Tokyo, I960, p. 336.
110. E. H . Creaser, Biochem. J. 64, 539 (1956).
111. L . Ottey, J. Pharmacol. Exptl. Therap. 115, 339 (1955).
112. H . Chantrenne and S. Devreux, Biochim. et Biophys. Acta 39, 486 (1960).
113. M . H . Richmond, Biochim. et Biophys. Acta 34, 325 (1959).
114. D . B. Roodyn and H . G. Mandel, J. Biol. Chem. 235, 2036 (1960).
115. E. Otaka, Exptl. Cell Research 21, 229 (1960).
116. H . G. Mandel, Arch. Biochem. Biophys. 76, 230 (1958).
117. H . Chantrenne, Biochem. Pharmacol. 1, 233 (1959).
118. A . Roush and E. R . Norris, Arch. Biochem. 29, 124 (1950).
119. H . G. Mandel, E. L . Alpen, W . D . Winters, and P. K . Smith, J. Biol. Chem. 193, 63 (1951).
120. E. Hirschberg, J. Kream, and A . Gellhorn, Cancer Research 12, 524 (1952).
121. G. B. Brown, Texas Repts. Biol. and Med. 10, 961 (1952).
122. D . C. Lorz and G. H . Hitchings, Federation Proc. 9, 197 (1950).
123. F. Bergmann, G. Levin, and H . Kwietny, Arch. Biochem. Biophys. 80, 318 (1959) 124. D . E. Duggan and E. Titus, Biol. Chem. 234, 2100 (1959).
125. P. Feigelson, J. D . Davidson, and R . K . Robins, J. Biol. Chem. 226, 993 (1957).
126. P. Feigelson and J. E. Ultmann, Acta. Unio Intern. Contra Cancrum. 16,609 (1960).
127. E. R . Norris and A . Roush, Arch. Biochem. 28, 465 (1950).
128. G. B . Elion, S. Bieber, and G. H . Hitchings, Ann. N. Y. Acad. Sci. 60, 297 (154).
129. L . Hamilton and G. B . Elion, Ann. N. Y. Acad. Sci. 60,304 (1954).
130. H . G. Mandel, Pharmacol. Revs. 11, 743 (1959).
131. J. J. Biesele, Ann. N. Y. Acad. Sci. 71, 1054 (1958).
132. S. Garattini and R . Paoletti, Giorn. ital. chemioterap. 3, 55 (1956).
133. N . O. Kaplan, A . Goldin, S. R . Humphreys, M . M . Ciotti, and F. E. Stolzenbach, J. Biol. Chem. 219, 287 (1956).
134. R . E. Matthews, / . Gen. Microbiol. 10, 521 (1954).
135. J. D . Smith and R . E. F. Matthews, Biochem. J. 66, 323 (1957).
136. J. H . Mitchell, Jr., H . E. Skipper, and L . L . Bennett, Jr., Cancer Research 10, 647 (1950).
137. H . G. Mandel, P . E. Carlo, and P. K . Smith, J. Biol. Chem. 206, 181 (1954).
138. S. Bieber, L . S. Dietrich, G. B . Elion, G. H . Hitchings, and D . S. Martin, Cancer Research 21, 228 (1961).
139. LePage, G. Α . , Cancer Research 20, 403 (1960).
140. H . Chantrenne and S. Devreux, Nature 181, 1737 (1958).
141. M . Grunberg-Manago and S. Ochoa, Am. Chem. Soc. 77, 3165 (1955).
142. L . A . Heppel and J. C. Rabinowitz, Ann. Rev. Biochem. 27, 613 (1958).
142a. S. B. Weiss and L . Gladstone, Am. Chem. Soc. 81, 4118 (1959).
142b. S. B . Weiss, Proc. Natl. Acad. Sci. U. S. 46, 1020 (1960).
143. M . J. Bessman, I . R . Lehman, J. Adler, S. B . Zimmerman, E. S. Simms, and A . Kornberg, Proc. Natl. Acad. Sci. U. S. 44, 633 (1958).
144. M . Friedkin, J. Biol. Chem. 209, 295 (1954).
145. M . Friedkin, Biochim. et Biophys. Acta 18, 447 (1955).
146. E. D . Korn and J. M . Buchanan, Biol. Chem. 217, 183 (1955).
147. A . Kornberg and W . E . Pricer, Jr., J. Biol. Chem. 193, 481 (1951).
148. L . N . Lukens and K . A . Herrington, Biochim. et Biophys. Acta 24, 432 (1957).
149. C. E. Carter, Biochem. Pharmacol. 2, 105 (1959).
150. J. L . W a y and R . E. Parks, Jr., Biol. Chem. 231, 467 (1958).
151. J. K . Roy, C. A . Haavik, and R . E. Parks, Jr., Proc. Am. Assoc. Cancer Research 3, 146 (1960).
152. J. L . W a y , J. L . Dahl, and R . E. Parks, Jr., J. Biol. Chem. 234, 1241 (1959).
153. A . R . P . Paterson, Can. J. Biochem. Physiol. 37, 1011 (1959).
154. G. A . LePage and J. L . Greenlees, Cancer Research Suppl. 3, 102 (1955).
155. A . C. Sartorelli and B . A . Booth, Arch. Biochem. Biophys. 89,118 (I960).
156. A . C. Sartorelli and G. A . LePage, Cancer Research 18, 1329 (1958).
157. C. Heidelberger and R . A . Keller, Cancer Research Suppl. 3, 106 (1955).
158. L . L . Bennett, Jr., R . W . Brockmann, and D . Smithers, Proc. Am. Assoc. Cancer Research 3, 94 (1960).
159. W . C. Werkheiser, R . J. Winzler, and D . W . Visser, Cancer Research 15, 641 (1955).
160. J. D . Davidson and Β . B . Freeman, Cancer Research 15, 31 (1955).
161. J. S. Salser, D . J. Hutchison, and M . E. Balis, J. Biol. Chem. 235, 429 (1960).
162. A . Hampton, Federation Proc. 19, 310 (1960).
163. R . W . Miller, L . N . Lukens, and J. M . Buchanan, J. Biol. Chem. 234,1806 (1959).
163a. J. S. Gots and E. G. Gollub, Proc. Soc. Exptl. Biol. Med. 101, 641 (1959).
164. R . W . Brockman, M . C. Sparks, and M . S. Simpson, Biochim. et Biophys. Acta 26, 671 (1957).
165. R . W . Brockman, C. Sparks, D . J. Hutchison, and H . E. Skipper, Cancer Research 19, 177 (1959).
166. R . W . Brockman and P . Stutts, Federation Proc. 19, 313 (1960).
167. A . R . P. Paterson, Can. J. Biochem. Physiol. 38, 1117 (1960).
168. A . R . P . Paterson, Can. J. Biochem. Physiol. 38, 1129 (I960).
169. J. Tomizawa and L . Aronow, J. Pharmacol. Exptl. Therap. 128, 107 (1960).
170. R . W . Brockman, Cancer Research 20, 643 (1960).
171. I . Lieberman and P . Ove, / . Biol. Chem. 235, 1765 (1960).
172. R . W . Brockman, L . L . Bennett, Jr., M . S. Simpson, A . R . Wilson, J. T . Thomson, and H . E. Skipper, Cancer Research 19, 856 (1959).
173. M . E. Balis, V . Hylin, M . K . Coultas, and D . J. Hutchison, Cancer Research 18, 440 (1958).
174. A . C. Sartorelli, G. A . LePage, and E. C. Moore, Cancer Research 18, 1232 (1958).
175. A . R . P. Paterson, Proc. Am. Assoc. Cancer Research 3, 141 (1960).
176. S. Bieber, R . Orsini, and G. H . Hitchings, Proc. Am. Assoc. Cancer Research 4, 210 (1961).
