the need for additional alkylating agents and antimetabolites · clinical pharmacology. the...

(CANCER RESEARCH 29, 2435-2442, December 1969]

The Need for Additional Alkylating Agents and Antimetabolites

Charles Heidelberger1

McArdle Laboratory for Cancer Research, University of Wisconsin, Madison, Wisconsin 53706

The title that was assigned to me is a clear invitation topontificate. Instead, I will share with you a few modest andnot very original thoughts and speculations. These are basedon a large literature and a small experience of tilling the soilamong certain obscure and unnatural pyrimidines. This engagement with the rationality of antimetabolites followed anignoble exodus from the quagmire surrounding an even moreobscure alkylating agent. Thus, my bias is already exposed.

It is not my purpose to cover the enormous literature inthese two fields of alkylating agents and antimetabolites.Rather, I shall refer to reviews whenever possible. Nor do Iintend to explore systematically structure-activity relationships of every series of compounds, detailed screening results,comparative toxicology, clinical pharmacology, andmechanisms of action. I will try merely to focus on certainfacts, trends, and promising leads that happen to interest meand suggest points of departure for future synthetic chemicalforays. I shall emphasize drug design at the expense of ignoringthe exciting advances in cell-cycle kinetics and clinical combination chemotherapy that have been emphasized elsewherein this Symposium. I shall forbear from intoning the customary litany of our ignorance of the essential molecular lesions ofmalignancy and intend to proceed with what is known. But Ihad better proceed.

ALKYLATING AGENTS

The alkylating agents as a group are among the clinicallymost useful compounds in cancer chemotherapy. Their chemistry and attempts to design agents showing selectivity ofaction have been reviewed by Ross (46). Hirschberg (24) hasmade a very comprehensive compilation of the earlier information about their chemotherapeutic properties, and this hasbeen expanded and brought up to date by Ochoa andHirschberg (39). Various aspects of the mechanism of actionof alkylating agents in many systems have been reviewed byBrookes and Lawley (7), Lawley (32), Wheeler (57, 58), andWarwick (56). Oliverio and Zubrod (40) have discussed theirclinical pharmacology.

The alkylating agents with chemotherapeutic activity arechemically reactive and at least difunctional. The structures ofa few of the important compounds are shown in Chart 1.Compound I is sulfur mustard, the original prototype, andII-VIII are members of the nitrogen mustard (HN2) family,

CH3N'l

0

0>C ...^i-'N

f X

^

XCH2CH2CI

XCH2CH2CI

I,CH2CH2CI

HN CHSNVCH2CH2CI XCH2CH2CI

E HI

CICH,CH2X

XCH2CH2CI

^rrH

m

oCICH2CH2NH C- NCH2CH2CI

NO

â€¢yirr

CH2-CH2

CH2-CH2

X

/CH2-0XMCH2P-lN

CH,-N'H3ZECH2-CH2XCH2CH2CI1XSCH2CH2CI

CH2-N^N^Nsl

i^

IS.

0 0

CH^SOCH2CH2CH2CH2OSCH2

0 0

'American Cancer Society Professor of Oncology.

Chart 1. I, sulfur mustard; II, noi-HN2; III, nitrogen mustard, HN2;IV, Nitromin; V, phenylalanine mustard (SarcolyÃ¡n, Melphalan); VI,uracil mustard; VII, cyclophosphamide (Cytoxan, Endoxan); VIII,l,3-bis(2-chloroethyl)-l-nitrosourea, BCNU; IX, triethylenemelamine(TEM); X, Thio-TEPA; XI, Myleran.

with their characteristic 0-chloroethyl groups. Two examplesof ethyleneimines (aziridines) are IX and X, and the remainingimportant class of methane sulfonates is illustrated by Myleran(XI). Literally thousands of these compounds have beensynthesized, in which the reactive 0-chloroethyl, ethylene-imine, or methane sulfonate groups have been hung ontoinnumerable "carriers," which some mystics endow with

remarkable biologic specificity. Thus, we have phenylalaninemustard (V, Sarcolysin, Melphalan), uracil mustard (VI),antimalarial mustards, steroidal mustards, aromatic mustards,quinonoid mustards, etc.

Do these "carriers" in fact impart any biologic specificity to

the reactive groups that they embrace? In order to answer thisquestion it is necessary to have reliable comparative data onthe activities of a large number of compounds. Fortunately,

DECEMBER 1969 2435

on June 2, 2021. © 1969 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

http://cancerres.aacrjournals.org/

CharlesHeidelberger

these data have been provided by a truly heroic andmonumental study by Schmidt et al. (50). These investigatorsstudied 51 ÃŸ-chloroethylamines, 25 aziridines, 39 methanesulfonates, and 12 nonalkylating compounds for their toxicityto mice, rats, dogs, and monkeys and their activities against 18transplanted rat tumors and 7 transplanted mouse tumors.This tour de force was published in a monograph of 1528pages (50).

It has been alleged by many that the proof of thechemotherapeutic specificity of the "carrier" rests on the"fact" that the L-phenylalanine mustard has greater tumor-

inhibitory effectiveness than the D isomer or the DL mixture(V). Let us examine the most reliable information on thissubject as presented by Schmidt et al. (50). Their data on therat tumors are given in Table 1. At the bottom of the table isthe mean for 10 tumors of the ratio of the LDi0/ED90 (orED60), the therapeutic index, for the L isomer over the Disomer, which is 1.05 Â±0.19. Thus, there is no difference in thetherapeutic index of the two isomers although, admittedly, theL isomer is more toxic and is more active against the tumorsthan the D isomer. On the other hand, when the ratio of the Lisomer to the DL mixture is taken for 16 tumors, it is 0.76 Â±0.08. This finding that the DL mixture has a bettertherapeutic index than the L isomer resists rationalexplanation. However, the same comparison in mouse tumors(Table 2) shows that the L isomer is better than the D and DL,but not strikingly so. If the ratios of all the rat and mouse tumorscombined are taken, the mean L/DL ratio of the therapeuticindex for 22 tumors is 0.92 Â±0.09, and the L/D ratio for 16tumors is 1.20 Â±0.15. Neither of these ratios is significantlydifferent from 1. In addition to demonstrating biologicvariability, these data show clearly that if these optical isomersrepresent the proof for the chemotherapeutic specificity of the"carrier," then the case is lost, even though there are

differences in overall toxicity. Schmidt et al. (50) haveconcluded that "with possibly two exceptions, both in the

Table 1

Tumor L DL L/DI/7 D L/D6

Table 2

Walker 256, CH,subcutaneousWalker256, CH,pulmonaryWalker256, CH,ascitesWalker256, CB,subcutaneousWalker256, CB,pulmonaryWalker256, SK,subcutaneousWalker256, SK,pulmonaryYoshida

asciteshepatomaYoshidaascitessarcomaDunningIRC-741 leukemia,subcutaneousDunningIRC-741 leukemia,ascitesLymphoma

8,subcutaneousAdenocarcinomaR-35,subcutaneousAdenocarcinomaR-35,pulmonaryMurphy-Sturm

lymphosarcoma,subcutaneousNovikoff

hepatoma, ascites178338.5331220332812.5112.51583121101012445425.5142.515.51.131.001.060.410.330.331.000.450.611.001.461.000.620.401.000.46102121101415310112.51.701.570.410.331.422.200.670.801.001.000.46

TumorSarcoma

180,subcutaneousAdenocarcinoma755,subcutaneousL1210,

subcutaneousL1210,ascitesEhrlichascitesAdenocarcinoma

SAH-I-I, subcutaneousL1.52.5.1.04.04.53.5DL1.02.51.02.53.02.0L/DLa

D11!111.5.0.0.6.5.71.02.51.02.51.7L/DÃ¨11112.5.0.0.6.6

Effects of optical isomers of phenylalanine mustard ontransplantable rat tumors (50). LDi 0 'I DC,,,or LD10/EDoo-

Â°L/DL:mean of 16 tumors, 0.76 Â±0.08.bL/D: mean of 10 tumors, 1.05 Â±0.19.

Effects of optical isomers of phenylalanine mustard on transplantablerat tumors (50). LDi0/ED90 orLD10/ED60.

Â°L/DL,mean of 6 tumors, 1.38 Â±0.11., mean of 5 tumors, 1.54 Â±0.26.

methane sulfonate series, there was no indication that specificclasses of alkylating agents or members within a class wereendowed with unique antitumor activity. Apart from theseexceptions, cyclophosphamide (VII) and the isomerie phenylalanine mustards (V) were the outstanding representatives invirtually every test system examined whether the neoplasmwas highly sensitive or relatively insensitive to alkylatingagents."

Let us now examine cyclophosphamide (VII, Cytoxan,Endoxan) a little more closely. This compound "was designed

in accordance with the principle of transversion of an inactivetransport form to an active form at special sites in the body"

(5). It was thought that tumors are rich in phosphoamidases,and hence that the P-N bond would be selectively cleaved inthe tumor to activate the drug. However, it was soon foundthat cyclophosphamide was not active against the growth oftumor cells in culture but was metabolized to an active form inthe liver (15). Hence, the properties of the compound were asdesigned, except that the activation takes place in the liverrather than in the tumor. Since that time a great deal of workhas been done, particularly by Brock (5) and Friedman (17) toelucidate the nature of this activation process. It was originallybelieved that after a series of hydrolyses, nor-HN2 (II) wasliberated and was the therapeutically active compound.However, this does not seem to be so, and the present state ofthis exceptionally complicated situation appears to be that themetabolism of cyclophosphamide in vivo does not parallel itschemical behavior. Considerably more work is needed beforean understanding is reached of the almost unique tumor-inhibitory activity of cyclophosphamide.

Another compound of particular interest is 1,3-bis(2-chlo-roethyl)-l-nitrosourea (BCNU) (VIII), which is unusual in thatit crosses the blood-brain barrier and is active againstintracerebral L1210 leukemia (49). Although it containsÃŸ-chloroethyl groups, there is some doubt as to whether it isreally an alkylating agent, since it reacts to only a slight extentwith 4-(p-nitrobenzyl)-pyridine (59). It inhibits DNA andRNA synthesis in LI210 leukemia cells in vivo and in vitro(60), and its effect on a DNA polymerase system iscomparable to that of /3-chloroethylisocyanate ; it appears toreact with the enzyme rather than with the primer (61).

It is clear that there is very little relationship between thedistribution, excretion, and metabolism of the alkylatingagents and their mechanism of action (39,40). Although theyare often considered to be "radiomimetic" compounds, this

2436 CANCER RESEARCH VOL. 29



Additional Alkylating Agents and Antimetabolites

analogy is probably overdone (56). Studies on the mechanismof resistance of experimental tumors to alkylating agents havebeen critically reviewed by Wheeler (58). The main fact is thatthere is considerable cross-resistance among these structurallydiverse compounds, but Wheeler (58) cautions against assuming from that fact that the alkylating agents act by a commonmechanism; he considered altered permeability as an important factor in resistance, and the repair of alkylated DNA (seebelow) was unknown at that time.

What then is the mechanism of action of the alkylating agents,if, indeed, a single mechanism exists? It appears from anoverwhelming body of data that the primary action of thesecompounds is a direct attack on DNA. The chemistry of thishas been worked out in a brilliant series of investigations byBrookes and Lawley, which has been reviewed by Lawley (32).They showed that sulfur mustard (I) exerts a nucleophilicattack on the N-7 of guanine residues in the DNA, whichlabilizes the glycosidic bond and causes the release of guaninefrom the DNA, which can thus give rise to mutations. Theyalso showed that sulfur mustard could react with two guaninesto produce a cross-linking, which they postulate to be theprimary cytotoxic action, and which would prevent DNAreplication (7). They have demonstrated with sensitive andresistant strains of bacteria that in the latter there is an activerepair mechanism that excises the cross-linked guanines fromthe DNA (33). They also found in bacteria that similarcross-linking was obtained with triethylenemelamine (TEM)(IX) and butadiene diepoxide, and produced additional evidence that the cross-linking is interstrand (34). Crathorn andRoberts extended this work to mammalian cells (HeLa) andhave demonstrated that with sulfur mustard there is a directrelationship between lethality and attack on DNA; they alsofound repair-excision of alkylated guanines from the DNA(11). Ruddon has reported that DNA reacted with HN2 (HI)has a decreased template activity for RNA polymerase, and toa lesser extent for DNA polymerase (48), which is opposite tothe inhibitory effects of nucleic acid biosynthesis found inintact cells.

However, the view that the primary chemotherapeutic effectof alkylating agents results from their attack on DNA is notuniversally accepted, largely upon the basis of work done inmice with tumors. Golder et al. (19) found that the extent ofalkylation by HN2 was very slight and that there was someevidence of crosslinking DNA to protein. Wheeler and Alexander (59) studied the in vivo alkylation of sensitive andresistant tumors implanted bilaterally into the same mouse,and found no correlation between the extent of alkylation ofthe DNA and chemotherapeutic effect. However, it is evidentto all those who are concerned with the binding of carcinogensto DNA that there is considerable specificity in binding, and infact Doskocil and Sormova (13) have found that not allguanines in DNA are equally reactive to sulfur mustard. Untilthe nature of this specificity is understood, the mechanism ofaction of the alkylating agents will remain somewhat obscure.

What then is the need for additional alkylating agents?Schmidt et al. (50) have stated, "In one sense the lack of

specificity of the various alkylating agents in both thetherapeutic activity and toxicity spheres is disappointing. Onthe surface at least, it offers little hope that any agent of this

chemical type will be found which will affect some neoplasmsin a unique way ...." Therefore, I feel that continuation of a

vast effort of synthesis aimed at the random attachment ofalkylating groups to a fanciful collection of "carriers" can no

longer be justified. However, there appear to me to be threeareas in which specific and rational syntheses could be justifiedat this time: (a) preparation of possible metabolites ofcyclophosphamide and its derivatives, (b) synthesis of furtherderivatives of BCNU, and (c) once we have a much moresophisticated knowledge of the parameters regulating thespecificity of the attack on DNA, this information might beput to use in the area of drug design.

ANTIMETABOLITES

Let us herewith define a metabolite as some naturallyoccurring compound produced during metabolism and anantimetabolite as a compound related structurally to themetabolite which prevents its further utilization by competingwith it for an enzyme. This definition and field came out ofthe pioneering efforts of D. D. Woods (64), who in 1940recognized the metabolite-antimetabolite relationship ofp-aminobenzoic acid and sulfanilamide. The role of thesecompounds in pharmacology and biochemistry was explainedwhen the structure and function of folie acid were elucidated.Reviews of various aspects of cancer chemotherapy withantimetabolites of many sorts have been provided by Stock(53), Langen (30), Baker (2), and Timmis (54).

Antifolics

The antifolics in cancer chemotherapy have three distinctions: (a) they were the first compounds to be effective againstacute leukemia in children; (b) they cure a high percentage offemale patients with choriocarcinoma; and (c) they are nowwhere the action is.

Folie acid (XII) is reduced to tetrahydrofolate by theenzyme dihydrofoÃ-ate reducÃ-ase,which is powerfully inhibitedby antifolates such as Amethopterin (XIII, Methotrexate).Tetrahydrofolate reacts with formaldehyde (or other compounds at the same oxidation level) to give isomerie formyl ormÃ©thylÃ¨netetrahydrofolates, which are the coenzymes of allone-carbon metabolism. Consequently, these coenzymes arerequired for two steps of purine synthesis and for thymidylatesynthetase. A great deal of work has been done on themechanism of resistance to antifolics, and suffice it to say thatthis is not fully understood at present. Although Amethopterin is a very powerful inhibitor of dihydrofolate reductase,there is still a need to obtain more selective and specificcompounds.

A simple modification, by the addition of a mÃ©thylÃ¨negroupto give homofolic acid (XIV), has been studied by Friedkin etal. (43). They have found some antimalarial activity with thiscompound and have further demonstrated that tetrahydro-homofolates are inhibitors of thymidylate synthetase. Veryrecently, Hutchison (27) has reported that a series ofquinazoline analogs, of which XV is the most active, exertgreater inhibitions of bacteria and tumor cell growth than doesAmethopterin. Obviously, further work along these two linesis indicated.

DECEMBER 1969 2437



CharlesHeidelberger

A Vc-NHCHCH2CH2COOH

COOH

xn

^-C-NHCHCH2CH2COOH

COOHXIII

_ O~VcONHCHCH2CH2COOH

COOH

k ^-C-NHCHCH2COOH

COOHzz

impermeable to the compound (3). Hopefully, this difficultymay not be insurmountable. If so, we can look forward toinhibitors of dihydrofolate reductase with unsurpassedspecificity to cancer tissue. Here, then, is a supreme exampleof carrying out enzymatic structure activity studies combinedwith skillful and imaginative chemistry.

These latter studies of Hitchings (25) and Baker (3) gave riseto the third category of distinction listed above.

NH OH OH

XVIII

SH

XVII

Chart 2. XII, folie acid; XIII, amethopterin (Methotrexate); XIV,homofolic acid; XV, quinazoline analog; XVI, pyrimethamine(Daraprim); XVII, Bakei and Meyer's analog.

A classical study of dihydrofolate reducÃ-ase inhibition byanother series of inhibitors, the 2,4-diaminopyrimidines, illustrated by XVI, Daraprim, an active antimateria!, has been carriedout by Hitchings (25). He has studied the kinetics of interactionof a series of these compounds with dihydrofolate reductasesisolated from bacteria and various protozoa and has foundstriking differences. This makes it possible to achieve aremarkable specificity in the chemotherapy of infectiousdisease by taking advantage of these species differences inenzymes.

Baker has for many years been working on "active-site-directed irreversible inhibitors" of various enzymes, including

dihydrofolate reducÃ-ase. This productive research has beendescribed in his book (2) and in innumerable papers since. Bystudying the kinetics and nature of the inhibition of thisenzyme with a large number of structurally related compounds, he has "mapped" the region of the active site in terms

of ionic and hydrophobic regions. As the near-culmination ofthis approach, Baker and Meyer (3) have recently reportedthat compound XVII has the fantastic specificity of being apowerful irreversible inhibitor of the enzyme isolated fromLI210 cells, but not from mouse liver, spleen, and intestine!This finding has remarkable connotations as to the nature ofthe carcinogenic change, but it cannot yet be interpreted.Surely, this compound must be the "magic bullet" that we

have all been searching for, with apparent specificity forirreversible inhibition of only the tumor enzyme. Unfortunately, however, this compound is inactive in vitro atinhibiting the growth of LI210 cells which are presumably

SH ' 0

k CH3

xn XXtt XXTT1NHo NH2 NH2 ,oc>

HO

XXVI" ~ "~" CHNHCH2CH=CC >!>

HO OH

Chart 3. XVIII, adenine; XIX, guanine; XX, 8-azaguanine; XXI,6-mercaptopurine; XXII, thioguanine; XXIII, dimethyltriazenoirnid-azole carboxamide; XXIV, psicofuranine; XXV, arabinosyladenine;XXVI, Cordycepin; XXVII, R = H, Tubercidin, R = CM, Toyocamycin,R = CONH2, Sangivamycin; XXVIII, Formycin; XXIX, isopentenyl-adenosine.

Furine Ant Â¡metabolites

Various aspects relating to these compounds have beenreviewed by Brockman (6), Cohen (10), Fox (16), andHeidelberger (23). The naturally occurring purine bases,adenine (XVIII) and guanine (XIX), have a large number ofanalogs related to them in various ways. A number of theserepresent isosteric replacements in the imidazole ring: 8-azaguanine (XX), Tubercidin (XXVII), and Formycin (XXVIII).Two very important purine base analogs are 6-mercaptopurine(XXI) and thioguanine (XXII), which are effective in cancerchemotherapy. Their nucleosides are known and have beenstudied extensively, and it is of great interest that thea-anomer of 2'-deoxythioguanosine, in contrast to all others





studied, is biologically active and is incorporated into DNA(36). It has also been found that 6-methylmercaptopurineribonucleoside has quite a different mechanism of action than6-mercaptopurine (6-MP) and its ribonucleoside (4), and thatthe periodate oxidation product of its ribonucleoside inhibitstumor growth and has other interesting biologic actions (28).

Another interesting compound is 4(5X3,3-dimethyl-triazene)-imidazole-5(4)-carboxamide (XXIII), which is ananalog of an intermediate in purine biosynthesis and hasconsiderable clinical activity against melonomas (51). Itsbiochemical mode of action is not yet known (to this author),but it represents an important new series of compounds that isworthy of further chemical and biochemical exploration.

Arabinosyladenine (XXV) has considerable activity againstsome tumors, and studies of its mechanism of action have beencarried out (cf. 10, 23). Several purine nucleoside antibioticswith interesting structures and tumor-inhibiting activity havebeen isolated, characterized, and studied. These includeCordycepin (XXVI), Psicofuranine (XXIV), Tubercidin, Toyo-camycin, Sangivamycin (XXVII), and Formycin (XXVIII)(cf. 16). This latter compound is of particular interest because anumber of its biologic properties have been explained by Wardand Reich (55) as being due to the fact that, in polynucleotideform, its individual residues exist in the syn, rather than in thecommon anti, conformation. This poses interesting possibilities for other C-nucleosides.

Isopentenyl adenosine (XXIX) was isolated from transfer RNAand its structure determined by Hall et al. (20). It was foundto be a powerful cytokinin in plants and also had inhibitoryactivity against tumor cells and possibly against acute leukemiain children. Because of this interesting activity, a number ofrelated compounds have been synthesized and tested byFleysher et al. (14).

Although this presentation has been necessarily brief andsuperficial, it is clear that there are a number of interestingstructural modifications of purine nucleosides with biologicactivity. There are also several other different types ofstructural modifications that I have not listed here. Therefore,it is quite clear that further synthetic work with purine andpurine nucleoside analogs should be biologically, and probablychemotherapeutically, rewarding.

Pyrimidine Antimetabolites

Some general reviews of this subject are those of Stock (53),Langen (30), Baker (2), Timmis (54), and Brockman (6).

The pyrimidine bases that occur in the nucleic acids areuracil (XXX), thymine (XXXI), and cytosine (XXXII). Theonly pyrimidine antimetabolite that is chemotherapeuticallyactive as the free base is 5-fluorouracil (FU) (XXXIII), whichwas originally synthesized on the basis of a very definiterationale. With all other pyrimidine analogs, the nucleosidesare required for biologic activity. A derivative more active thanFU is 5-fluoro-2'-deoxyuridine (FUDR) (XXXIV); and trifluo-

rothymidine (XXXV), in addition to being a potent tumorinhibitor, is also the most active compound known to inhibitthe replication of DNA viruses. These compounds have beenthoroughly reviewed (22, 23). 5-Iodo-2 '-deoxyuridine (IUDR),

(XXXVI) is clinically active against herpes simplex keratitis

and is incorporated into DNA instead of thymine, but it hasbeen disappointing in cancer chemotherapy (cf. 23). Aninteresting approach to increasing the chemotherapeutic effectof a close relative, 5-iodo-2'-deoxycytidine (ICDR), was made

by Woodman (63), who complexed the 5 -phosphate of ICDRwith various polycations and found that it was incorporatedinto mouse tumor DNA in vivo to a considerably greaterextent than was the nucleoside. Such an approach may findapplication to other nucleotide analogs.

Chart 4. XXX, uracU; XXXI, thymine, XXXII, cytosine; XXXIII,5-fluorouracil; XXXIV, 5-fluoro-2'-deoxyuridine, FUDR; XXXV,trifluorothymidine; XXXVI, 5-iodo-2'-deoxyuridine, IUDR; XXXVII,

arabinosyl cytosine; XXXVIII, 6-azauridine; XXXIX, 5-azauridine; XL,3-deazacytidine; XLI, S'-fluorothymidine; XLII, Showdomycin.

Arabinosylcytosine (ara-C) (XXXVII) is a compound that isvery active against tumors, human acute leukemias, and DNAviruses (cf. 10, 23). Its mechanism of action is also underintensive study. It is rapidly deactivated by deamination toarabinosyluracil (ara-U), and interesting research has beencarried out to find inhibitors of this deaminase in the hope ofincreasing the chemotherapeutic effects of ara-C (8).

Isosteric replacements have also been fruitful with pyrimidine nucleosides. 6-Azauridine (XXXVIII) and 5-azauridine(XXXIX) have shown moderate activity against varioustumors, and this field has been reviewed by Skoda (52). It is ofinterest that although trifluorothymidine and 6-azathymidine

DECEMBER 1969 2439



CharlesHeidelberger

have biologic activity, a compound, 5-trifluoromethyl-6-aza-2'-deoxyundine, that combines both substitutions, was

biologically inert (12). Another compound with an isostericreplacement that has activity against bacteria and tumor cellsin culture is 3-deazacytidine (XL) as very recently reported byRobins et al. (45). In our laboratory we are synthesizingcomparable compounds in the deoxyribonucleoside series, aswell as other pyrimidine nucleoside analogs with isostericreplacements. Langen and Kowollik (31) have recently prepared 5 '-fluorothymidine (XLI), which is active against thy-

midylate kinase, and hence is a nucleotide, rather than anucleoside, analog.

A compound that appears superficially to resemble apyrimidine nucleoside in the antibiotic Showdomycin (XLII)(cf. 16). However, biochemical work by Roy-Burman et al.(47) indicates that it does not act as a pyrimidine analog, butrather it inhibits various enzymes as a consequence of thealkylating properties of the maleiimide moiety.

As in the case of the purine antimetabolites, the pyrimidinenucleosides are continuing to provide interesting new pharmacologically active compounds, including many that are notspecifically mentioned here. Further chemical efforts in thisfield seem to be fully justified.

Poly imdeot Â¡des

There is now abundant evidence that mammalian cells cantake up intact polynucleotides. This has been reviewed forDNA by Ledoux (35), and Click has shown that DNAobtained from mouse thymus can actually inhibit the growthof L1210 tumors in vivo (18). In our own studies, we madeoligonucleotides of FUDR-5'-phosphate (FUDRP) in 1961

(44), but these did not show any greater activity than FUDRat inhibiting the incorporation of formate into the DNAthymine of Ehrlich ascites cells in vitro.

An important recent discovery by Park and Baron (41)

H2 N C CH2CH2Ã‡H COOH

NH,

XL-Ill

HgH-C-CH2CH COOH

NH2

TTCV

HgN-C-NHOH

yo/ir

CH,CH-CH CHO3 i i

OH OH

NEN-CH2CO CH2CH2CH-COOH

NH2

XLIV

N5NCH2CO CH2CH COOHNH2

XLVI

CH3C-CHO CH2-CH-CHO

O OH Ã’H

XLVIM XLIX

demonstrated that double-stranded poly-inosinic:poly-cytidylic acid preparation (poly-IC) is capable of preventingor inhibiting viral replication in vitro and in vivo by stimulating interferon production. Very recently, Levy et al. (37) andHoman et al. (26) have reported that poly-IC inhibits thegrowth of several tumors in mice, and they have done somepharamacologic studies with the polymer. This opens up anexciting new field of macromolecular chemotherapy. Particu-lary when more is known about the mechanisms of theseeffects, it appears possible that major advances in specificcancer chemotherapy may be in the offing. In these days ofwild speculations about genetic engineering, it may not betotally inappropriate to prophesy that one day cancer may bereversed by appropriate polynucleotide chemistry [for areview of the chemistry of polynucleotides, see Michelson etal. (38)]. For example, if as proposed by Pitot andHeidelberger (42), carcinogenesis is the result of theperpetuated deletion of a represser, then if the messengerRNA that codes for this represser were synthesized, it couldcause the revision of that cancer to normal. Impossible? Whoknows?

Other Miscellaneous Antimetabolites

Again, the listing of these compounds will not be complete,but are selected arbitrarily according to my o Â».ncurrentinterests. Glutamine (XLIII) is an important biosyntheticintermediate, and diazo-oxo-norleucine (DON, XLIV) is itsantimetabolite. In view of the interesting therapeutic effects ofL-asparaginase, Handschumacher et al. (21) designed ananalogous inhibitor of the substrate asparagine (XLV), namelydiazo-oxo-nor-L-valine (XLVI), which they showed to haveenzymatic specificity. Such an approach may be applied toother amino acids, should other enzymes of amino acidmetabolism be found with tumor-inhibitory activities.

Hydroxyurea (XLVII) has been shown at high concentrationsto be a reversible inhibitor of DNA synthesis (65), and it hassome cancer chemotherapeutic activity which has stimulatedthe synthesis of related analogs. It has been shown by Krakoffet al. (29) to be an inhibitor of ribonucleoside diphosphatereducÃ-ase.

Lastly, I would like to consider three aldehydes. Veryrecently Apple and Greenberg (1) reported that the combination of two metabolites in minor branches of the 3-carbonstage of glycolysis, 2-oxopropanal (XLVIII) and 2,3-dihy-droxypropanal (XLIX), cured some mice with transplantedtumors. It has been demonstrated by Ciaranfi et al. (9) thatL-eryrfiro-2,3-dihydroxybutyraldehyde, an inhibitor of protein synthesis and not of glycolysis, has considerableantitumor activity. Thus, it appears that more synthesis ofaldehydes may be in order.

CONCLUSIONS

Chart 5. XLIII,glutamine; XLIV, diazo-oxo-norleucine (DON); XLV,asparagine; XLVI, diazo-oxo-norvaline; XLVII, hydroxyurea; XLVIII,2-oxopropanal; XLIX, 2,3-dihydroxypropanal; L, L-erythro-2,3-duiy-droxybutyraldehy de.

What is the need for additional alkylating agents andantimetabolites? In my opinion there is no need for additionalalkylating agents dreamed up ad hoc, but possibly there is aneed for compounds specifically designed relative to the





interest in the metabolism and modes of action of cyclophos-phamide and BCNU. I hope that I have made a convincing casefor the need for the design and preparation of additionalantimetabolites. The limits for these should only be theimagination and vision of the investigator and his chemicalskill.

REFERENCES

1. Apple, M. A., and Greenberg, D. M. Arrest of Cancer in Mice byTherapy with Normal Metabolites. II. Indefinite Survivors AmongMice Treated with Mixtures of 2-Oxopropanal (NSC-79019) and2,3-Dihydroxypropanal (NSC-67934). Cancer ChemotherapyKept., 52: 687-696, 1969.

2. Baker, B. R. Design of Active-Site-Directed Irreversible EnzymeInhibitors. New York: John Wiley & Sons, 1967.

3. Baker, B. R., and Meyer, R. B., Jr. Irreversible Enzyme Inhibitors.CXLIII. Active-site Directed Irreversible Inhibitors of DihydrofolicReducÃ-aseDerived from 5-(p-Aminophenoxypropyl)-2,4-diamino-6-methylpyrimidine with a Terminal Sulfonyl Fluoride, J. Med.Chem., 12: 108-111,1969.

4. Bennett, L. L., Jr., Brockman, R. W., Schnebli, H. A., Chumley, S.,Dixon, G. J., Schabel, F. M., Dulmadge, E. A., Skipper, H. E.,Montgomery, J. A., and Thomas, H. J. Activity and Mechanism ofAction of 6-Methylthiopurine Ribonucleoside in Cancer CellsResistant to 6-Mercaptopurine. Nature, 205: 1276-1279, 1965.

5. Brock, N. Pharmacological Characterization of Cyclophosphamideand Cyclophosphamide Metabolites. Cancer Chemotherapy Rept.,51: 315-325, 1967.

6. Brockman, R. W. Mechanisms of Resistance to Anticancer Agents.Advan. Cancer Res., 7: 130-235,1963

7. Brookes, P., and Lawley, P. D. Reaction of Some Mutagenic andCarcinogenic Compounds with Nucleic Acids. J. Cellular Comp.Physiol., 64 (Suppl. 1): 111-128,1964.

8. Camiener, G. W., and Tao, R. V. Studies on the EnzymaticDeamination of Ara-Cytidine. IV. Inhibition by an Acridine Analogand Organic Solvents. Biochem. Parmacol., 17: 1411-1423, 1968.

9. Ciaranfi, E., Loreti, L., Borghetti, A., and Guidotti, G. Studies onthe Anti-tumor Activity of Aliphatic Aldehydes. II. Effects onSurvival of Yoshida Ascites Hepatoma-bearing Rats. European J.Cancer,/: 147-151,1965.

10. Cohen, S. S. Introduction to the Biochemistry of D-ArabinosylNucleosides. Prog. Nucleic Acid Res. Mol. BioL, 5: 1-88, 1966.

11. Crathorn, A. R., and Roberts, J. J. Mechanism of the CytotoxicAction of Alkylating Agents in Mammalian Cells and the Evidencefor the Removal of Alkylated Groups from Deoxyribonucleic Acid.Nature, 211: 150-153, 1966

12. Dipple, A., and Heidelberger, C. Fluorinated Pyrimidines. XXVIII.The Synthesis of 5-Trifluoromethyl-6-azauracil and 5-Trifluoro-methyl-6-aza-2 -deoxyuridine. J. Med. Chem., 9: 715-718,1966.

13. Doskocil, J., and Sormova, Z. The Reaction of DNA withMustards. II. The Reaction Kinetics. Collection Czech. Chem.Commun. 30: 492-506, 1965.

14. Fleysher, M. H., Hakala, M. T., Bloch, A., and Hall, R. H. Synthesisand Biological Activity of Some N-6-Alkyladenosines. J. Med.Chem., 11: 717-720, 1968.

15. Foley, G. E., Friedman, O. M., and Drolet, B. D. Studies on theMechanism of Action of Cytoxan. Evidence for Activation in Vivoand in Vitro. Cancer Res., 21: 57-63, 1961.

16. Fox, J. J., Watanabe, K. A., and Bloch, A. Nucleoside Antibiotics.Prog. Nucleic Acid Res. Mol. BioL, 5: 252-314,1966.

17. Friedman, O. M. Recent Biologic and Chemical Studies ofCyclophosphamide. Cancer Chemotherapy Rept., 57: 327-333,

1967.

18. Click, J. L. The Specificity of Inhibition of L1210 Tumor Growthby Mouse Thymus Deoxyribonucleic Acid. Cancer Res., 27:175-180, 1967.

19. Golder, R. H., Martin-Guzman, G., Jones, J., Goldstein, N. O.,Rotenberg, S., and Rutman, R. J. Experimental ChemotherapyStudies. III. Properties of DNA from Ascites Cells Treated in Vivowith Nitrogen Mustard. Cancer Res., 24: 964-968, 1964.

20. Hall, R. H., Robins, M. J., Stasiuk, L., and Tredford, R. Isolationof N6-(ry-Dimethylallyl)adenosine from Soluble Ribonucleic Acid.

J. Am. Chem. Soc., 88: 2614-2615,1966.21. Handschumacher, R. E., Bates, C. J., Chang, P. K., Andrews, A. T.,

and Fischer, G. A. 5-Diazo-4-oxo-L-Norvaline: ReactiveAsparagine Analog with Biological Specificity. Science, 161:62-63, 1968.

22. Heidelberger, C. Fluorinated Pyrimidines. Prog. Nucleic Acid Res.Mol. Biol., 4: 1-50, 1965.

23. Heidelberger, C. Cancer Chemotherapy with Furine and PyrimidineAnalogues. Ann. Rev. Pharmacol., 7: 101-124, 1967;

24. Hirschberg, E. Patterns of Response of Animal Tumors toAnticancer Agents. A Systematic Analysis of the Literature inExperimental Cancer Chemotherapy, 1945-1958. Cancer Res.,23 (ft. 2): 521-980, 1963.

25. Hitchings, G. H. Chemotherapy and Comparative Biochemistry: G.H.A. Clowes Memorial Lecture. Cancer Res., 29: 1895-1903, 1969.

26. Homan, E. R., Zendzian, R. P., and Adamson, R. H. Some Aspectsof the Pharmacology and Toxicology of Polyinosinic: PolycytidylicAcid. Proc. Am. Assoc. Cancer Res., 10: 40, 1969.

27. Hutchison, D. J. Quinazoline Antifolates: Biological Activities.Cancer Chemotherapy Rept., 52: 697-705, 1969

28. Kimball, A. P., Wilson, M. J., Bell, J. P., and LePage, G. A.Inhibition of Thymidylate Kinase and DNA Polymerase by thePeriodate Oxidation Product of ÃŸ-D-Ribosyl-o-methylthiopurine.Cancer Res., 28: 661-665, 1968.

29. Krakoff, I. H., Brown, N. C., and Reichard, P. Inhibition ofRibonucleoside Diphosphate ReducÃ-aseby Hydroxyurea. CancerRes., 28: 1559-1565, 1968.

30. Langen, P. Antimetabolite des Nucleinsaure-Stoffwechsels. EastBerlin: Akademie-Verlag, 1968.

31. Langen, P., and Kowollik, G. 5'-Deoxy-5 -fluorothymidine, aBiochemical Analogue of Thymidine-5 '-Monophosphate Selectivity

Inhibiting DNA Synthesis. European J. Biochem., 6: 344-351,1968.

32. Lawley, P. D. Effects of Some Chemical Mutagens and Carcinogenson Nucleic Acids. Prog. Nucleic Acid Res. Mol. Biol., 5: 89-132,1966.

33. Lawley, P. D., and Brookes, P. Molecular Mechanism of theCytotoxic Action of Difunctional Alkylating Agents and ofResistance to this Action. Nature, 206: 480-483, 1965.

34. Lawley, P. D., and Brookes, P. Interstrand Cross-linking of DNA byDifunctional Alkylating Agents. J. Mol. Biol., 25: 143-160, 1967.

35. Ledoux, L. Uptake of DNA by Living Cells. Prog. Nucleic AcidRes. Mol. Biol., 4: 231-267, 1965.

36. LePage, G. A. The Metabolism of a-2 -Deoxythioguanosine inMurine Tumor Cells. Can. J. Biochem., 46: 655-661, 1968.

37. Levy, H. B., Law, L. W., and Rabson, A. Inhibition of TumorGrowth by Polyinosinic-Polycytidylic Acid. Proc. Am. Assoc.Cancer Res., 10: 49, 1969.

38. Michelson, A. M., Massoulie, J., and Guschlbauer, W. SyntheticPolynucleotides. Prog. Nucleic Acid Res. Mol. Biol., 6: 83-141,

1967.39. Ochoa, M., Jr., and Hirschberg, E. Alkylating Agents. In: R. J.

Schnitzer and F. Hawking (eds.), Experimental Chemotherapy,Vol. 5, pp 1-132. New York: Academic Press, 1967.

40. Oliverio, V. T., and Zubrod, C. G. Clinical Pharmacology of the

DECEMBER 1969 2441



Charles Heidelberger

Effective Antitumor Drugs. Ann. Rev. Pharmacol., 5: 335-356,

1965.41. Park, J. H., and Baron, S. Herpetic Keratoconjunctivitis: Therapy

with Synthetic Double-Stranded RNA. Science, 162: 811-813,

1968.42. Pilot, H. C., and Heidelberger, C. Metabolic Regulatory Circuits

and Carcinogenesis. Cancer Res., 23: 1694-1700,1963.43. Plante, L. T., Crawford, E. J., and Friedkin, M. Enzyme Studies

with New Analogs of Folie Acid and Homofolic Acid. J. Biol.Chem., 242: 1466-1476, 1967.

44. Remy, D. C., Sunthanker, A. V., and Heidelberger, C. Studies onFluorinated Pyrimidines. XIV. The Synthesis of Derivatives of5-Fluoro-2 -deoxyuridine-5 '-phosphate and Related Compounds. J.

Org. Chem., 27: 2491-2500, 1961.45. Robins, M. J., Currie, B. L., Robins, R. K., and Bloch, A. The

Biological Activity of 3-Deazapyrimidine Nucleosides. Proc. Am.Assoc. Cancer Res., 10: 73,1969.

46. Ross, W. C. J. Biological Alkylating Agents: Fundamental Chemistry and the Design of Compounds for Selective Toxicity.London: Butterworths, 1962.

47. Roy-Burman, S., Roy-Burman, P., and Visser, D. W. Showdomycin,A New Nucleoside Antibiotic. Cancer Res., 28: 1605-1610, 1968.

48. Ruddon, R. W., and Johnson, J. M. The Effects of NitrogenMustard on DNA Template Activity in Purified DNA and RNAPolymerase Systems. Mol. Pharmacol., 4: 258-273, 1968.

49. Schabet, F. M., Jr., Johnston, T. P., McCaleb, G. S., Montogomery,J. A., Laster, W. R., and Skipper, H. E. Experimental Evaluation ofPotential Anticancer Agents. VIII. Effects of Certain Nitrosoureason Intracerebral L1210Leukemia,Cancer Res., 25:725-733,1963.

50. Schmidt, L. H., Fradkin, R., Sullivan, R., and Flowers, A.Comparative Pharmacology of Alkylating Agents. Cancer Chemotherapy Rept., Supplement 2, 1-1528, 1965.

51. Shealy, Y. F., Montgomery, J. A., and Laster, W. R., Jr. AntitumorActivity of Triazenoimidazoles. Biochem. Pharmacol., 11:674-676,1962.

52. Skoda, J. Mechanism of Action and Application of Azapy-rimidines. Prog. Nucleic Acid Res. Mol. Biol., 2: 197-219, 1963.

53. Stock, J. A. Antimetabolites. In: R. J. Schnitzer and F. Hawking

(eds.), Experimental Chemotherapy, Vol. 4, pp. 79-237. NewYork: Academic Press, 1966.

54. Timmis, G. M., Antagonists of Furine and Pyrimidine Metabolitesand of Folie Acid. Advan. Cancer Res., 6: 369-402, 1961.

55. Ward, D. C., and Reich, E. Conformational Properties of Poly-formycin: A Polyribonucleotide with Individual Residues in theSyn Conformation. Proc. Nati. Acad. Sei. U. S., 61: 1494-1501,

1968.56. Warwick, G. P. The Mechanism of Action of Alkylating Agents.

Cancer Res., 23: 1315-1333, 1963.57. Wheeler, G. P. Studies Related to the Mechanism of Action of

Cytotoxic Alkylating Agents: A Review. Cancer Res., 22:651-688, 1962.

58. Wheeler G. P. Studies Related to Mechanisms of Resistance toBiological Alkylating Agents. Cancer Res., 23: 1334-1349, 1963.

59. Wheeler, G. P., and Alexander, J. A. Studies with Mustards. V. InVivo Fixation of C1 of Labeled Alkylating Agents by Bilaterally

Grown Sensitive and Resistant Tumors. Cancer Res., 24:1331-1337,1964.

60. Wheeler, G. P., and Bowdon, B. J. Effects of l,3-Bis(2-chloro-ethyl)-l-nitrosourea upon the Synthesis of Proteins and NucleicAcids in Vivo and m Vitro. Cancer Res., 25: 1770-1778,1965.

61. Wheeler, G. P., and Bowdon, B. J. Effects of l,3-Bis(2-chloro-ethyl)-l-nitrosourea and Related Compounds upon the Synthesisof DNA by Cell-free Systems. Cancer Res., 28: 52-59, 1968.

62. Wheeler, G. P., and Chumley, S. Alkylating Activity ofl,3-Bis(2-chloroethyl)-l-nitrosourea and Related Compounds. J.Med. Chem., 10: 259-261,1967.

63. Woodman, R. J. Localized Incorporation of lododeoxyuridinefrom Polycation-complexed lododeoxycytidylic acid into DNA ofSeveral Murine and Hamster Tumors. Cancer Res., 28: 2007-2016,1968.

64. Woods, D. D. Relation of p-Aminobenzoic Acid to the Mechanismof Action of Sulfanilamide. Brit. J. Exptl. Path., 21: 74-90,1940.

65. Young, C. W., Schochetman, G., Hodas, S., and Bab's, M. E.

Inhibition of DNA Synthesis by Hydroxyurea: Structure-ActivityRelationships. Cancer Res., 27: 535-540, 1967.




1969;29:2435-2442. Cancer Res Charles Heidelberger The Need for Additional Alkylating Agents and Antimetabolites

Updated version

http://cancerres.aacrjournals.org/content/29/12/2435.citation

Access the most recent version of this article at:

E-mail alerts related to this article or journal.Sign up to receive free email-alerts

Subscriptions

Reprints and

[email protected] at

To order reprints of this article or to subscribe to the journal, contact the AACR Publications

Permissions

Rightslink site. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC)

.http://cancerres.aacrjournals.org/content/29/12/2435.citationTo request permission to re-use all or part of this article, use this link

http://cancerres.aacrjournals.org/content/29/12/2435.citationhttp://cancerres.aacrjournals.org/cgi/alertsmailto:[email protected]://cancerres.aacrjournals.org/content/29/12/2435.citationhttp://cancerres.aacrjournals.org/

the need for additional alkylating agents and antimetabolites · clinical pharmacology. the...

Documents