869 anhydrous niethanolic methylamine were added. The tube was sealed and heated a t 50' for 65 hr in a rocking bomb.28 The solvent was evaporated under a stream of nitrogen, and the residue was dried in vacuo over P2Oj. The residue (15.9 mg) was taken up in water and run into a column ( 5 mm i.d. X 15 em) containing 3 ml of Dowex 50R-X4 ( H + form). Xeutlal, ultraviolet-absorbing materials were removed by washing with water, Evaporaand then the product was eluted with 1.0 ,?' ",OH. tion in vucuo in a rotary evaporator in a water bath (25-50") followed by drying in vacuo over P2Oj furnished 6.6 mg of crude product (ultraviolet spectra were qualitatively indistinguishable from the standard). The prodiict was further purified o n a eoliimn ( 5 mm i.d. X 22 em) of silica gel (J. T. Baker, 3406) (2.5 g), packed in methylene chloride-methanol (97:3). The column was developed with the same solvent. Upon increasing the methanol concentration to 6%, the product was eluted in six 10-nil ciits. The four center cuts were combined on the basis of iiltraviolet and counting data and also by radiopapergram analysis. Concentration in z'acuo furnished 2.8 mg (0.12 mcurie, 11 mcuries/mmole) of amorphous product I X which was indistinguishable from MADU by ultraviolet analysis (optical density ratio comparisons) and homogeneous by radiopapergram analysis using unlabeled MADU as standard. Radiopapergram Analysis.-Whatman No. 1 filter paper strips and 2-propanol-concentrated (387,) HC1-water ( 7 5 : s :17) solution were used for descending-type papergram analysis. Sections of the paper were placed in phosphor solution and read directly in a liqriid-scintillation counter.24 MAIlT!-6-H3 was (28) 1I)U was completely destroyed using the same conditions, 80° for 18 hr, found successful for amination of BrDU. A series of 10-mg runs a i t h unlabeled I D U was performed t o determine optimum conditions.
TABLE 111 Ri values 0.22 0.32 0.63 0.73 0.75 0.80
Standards
5-Me thylaminouracil
X4DU Uracil 2-Deoxyuridine 5-Iodouracil IDU
differentiated from possible interfering contaminants (see Table 111). I n Vitro Cleavage Studies of Substituted 5-Amino-2'-deoxyuridines.-The assay for determining the cleavage of nucleosides by human spleen thymidine phosphorylase >vas carried out by Dr. Morris Zimmerman of these laboratories essentially as describedz9 for measuring the formation of 2'-deoxyribose 1phosphate. The nucleoside was incubated with the enzyme in phosphate buffer. The cleavage rate of thymidine was taken as 1007,. The rates with the 5-substituted 2'-deoxyuridines are as folloas: 5-amino (11),36%; 5-methylamino ( I ) , 277,; 5dimethylamino (111),Oyo.
Acknowledgment.-The authors wish to thank 1Ir . Richard N. Boos for microanalytical data, 11r. Alee Kalowsky for ultraviolet measurement,q, Ih.Edward A . h3acAIullan for spectrophotometric analysis (pK and isosbestic point det'erniinations), Ur. Charles Rosenbluni for assistance with radioactivity nieasurenlents, and Dr. James J. Kittick for ORD analysis. (29) G. E. Boxer and C. E. Shonk, J . B i d . Chem., 233, 535 (lY58).
Anticancer and Potential Antiviral Activity of Complex Inorganic Compounds STANLEY KIRSCHNER, YUSG-KANGWEI, DAWS FRASCIS, AND JOHNG. BEROKIV Department of Chemistry, W a y n e State C'niversity, Detroit, Mirhigan Received August 2, 1965 Recent studies indicate that viruses are associated with many types of cancerous tumors. The authors propose that complex inorganic compounds might be successfully used in the alteration of these viruses with the consequent destruction or diminution of their activity. Twenty-six complex inorganic compounds have been prepared in this study, and among them are five which are active against certain cancers, including complexes containing coordinating agents which are themselves carcinostatic, such as 6-mercaptopurine. Both the platinum(IV) and palladium( 11) complexes of 6-mercaptopurine have been found to be extremely active against Adenocarcinoma 755 and Sarcoma 180.
Despite considerable controversy in recent years regarding the possible role of viruses as cancer-producing agents, there is no longer any question that certain types of cancer are virus caused and some scientists' feel that most, if not all cancers, have one or more viruses associated with them. Since viruses contain proteins and nucleic acid units, some atoms of which are excellent coordinating agents for metal ions, it was decided to study the possibility of altering the virus via its coordination and/or chelation to metal ions with the expectation that this alteration will result in the elimination or diminution of its viral activity.
suffice because of their almost certain coordination to one or more of the many nonviral proteins, amino acids, and other coordinating groups found in most parts of the animal system. Consequently, this investigation is concerned with the introduction of nietal ions in the form of moderately stable complex inorganic compounds (coordination compounds), the ligands of which might possibly be displaced by viruses, thus forming new complexes with the metal ion, for example
Discussion
where Vi represents a virus; 111, a metal ion; L, a ligand; and where nz and n represent the number of molecules of virus and ligand, and z and y represent the charges on the metal ion and ligand, respectively. ilnother important aspect of this problem is finding and capitalizing upon some difference between viral systems which may cause abnormal growth and the many benign coordinating systems in animals, because
Perhaps the most important aspect of this problem involves "appropriate" introduction of a metal ion into the vicinity of a virus so that complexation or coordination niay occur. Nere introduction of simple or hydrated metal ions into living organisms mill not (1)
S. E. Lima, Cancer Res., 20, 669 (1960).
m(Vi)
+ [ M + z ( L - ~ ) n ] Z -=n ~ [lI+z(Vi),]+z
+ nL-11
(1)
the latter will conipetc for any metal ion introduced into a system in the form of a complex. Although thr reasons for the marked nntirancer activiiy of thcw cwmplexes :ire riot c.ompletely untl(~qtoot1. furt1ic.i. study of the effccts of thew cowdiiiatioii (~)1n1)oii11(1~ niay lcad to a delineation of mnir of thc tlifferriic~s, (veil though they may bc slight. I:or cxaniplc, i f viral and nonviral systems tend to form rings oj different numbers o j atmis with metal ions, the fornier could Iiavc a much higher stability than the latter as 1)ai.tof :I (Loordination compound (e.y., six-meniberrtl I s. ~ ( ~ 1 1 1 n~embcredrings). In this casc it noiiltl alnio.it wrtninly mean the difference between a reaction of tlic initial complex with the viral ten] and no rcwtioii with the Iioriviral one. Ariothcr slight difference in structure which niight bc detectable by the use of complexes is an atom differ( w e . If the molecules of benevolent and nialevolent coordinating ligands are similar in structure and in chelatc ring size but differ in that they hare a different coordinating atom in a particular location (e.g., one contains a sulfur where the other has an oxygen o r nitrogen), then the selection of the appropriate metal ion niay help t o utilize this difference t o alter t h e physiological properties of the nialevolcnt coordinating agent. Platinum ions, for exani~)le,form very \t:thlr complexes with sulfur as ii coordinating atoni of' ii ligand and relatively weak complexes with oxygeii ab n donor atom. The situation is ~)ractirallywverwtl, however, with magnesium or bcrylliuni ions. Metal ions were chosen from 1ho.e areab of t h e periodic table which mould be reprcwitative of threc major types of atom coordiriation. Platinum metals were chosen as examples of good sulfur coordinatorz, and complexes of these metal ions are reported hew. l\lagnesiurn, beryllium, and aluminrini metals arc. currently being studied a b exaniplcls of good oxygen coordinators, a5 are iron ancl cohalt as example:, of good nitrogen coordinators. Because the exact nature of the difference between coordinating sites on cancer-producing viruses and noncarcinogenic proteins, nucleic acids, etc., is riot yet known, design of coordination compounds having ligands which are easily displaced by viruses but not tiy iioriviral proteins or nucleic. acids is extreinely difficult. For example, proteins usually h a m inany possible and different coordination sites2zdsuch aLs :mino, imidazole, sulfhydryl, phenoxy1 groups, etc., and each reacts differently toward different nietal ions. Several metal-protein complexes and metal :imino acid complexes have already been prepared by othcr investigators4-l 3 and these have tiem studied All cwnpouiids were noiitoxic; except were noted otherwise. with regard t o other types of physiologicn1 activity. Ir gg = @ycylglycinato anion, hi = histidine, im = imidamle, me ( 2 ) S. W. P u t n a m , Prolezns, l B , 893 (1953). (3) '\V L Koltun, J . -4m. Chem. Soc., 79, 6681 (1957). (4) G. Yelsenfeld a n d M. P. Printz, zbzd., 81, 6259 (1959). (5) U. L. Vallee, Proc. 4th Intern. Conor. Bzochem. V z e n n ~8, , 138 (1960). (6) 0. 9. Shields, H. Markowitz, G E. Cartwnght, and ,M. Wintrobe, "bletsl Binding in Medicine." X. J. Seven a n d L. i, Johnson, Ed., J. I3. Lippincott Co., Montreal, 1960, p 259 ( 7 ) F. R . N Gurd and D. S. Goodman, J . A m Chem. Soc., 74, 670 (1952). (8) C Tanford, zbzd., 74, 211 (1952) (9) T. Peters, Jr., Bzorham. Bzophys Acta, 99, 546 (1960). ( I O ) I. A I Klotz and H. G . Curme, J . Am. Chem. Soc., 70, 939 (1948). (11) I. h l Mota, J. M , Urquhart, T. 4. Klota, a n d 1919 (1955). (12) E. Breslow a n d F. K. N. Gurd, J. Bzol. Chem., 298, 1332 (1963). (13) F'. 11. N. Guard and P. E. Wllcou, Aduan. Protean Chem., 11, 311 (19j6).
= methionate anion, bz = benzimidazole, da = daraprim [2,4diarnino-5-( 4-chlorophenyl)-6-ethylpyrimidine],dcda = tlichlorodaraprim [2,4-diamino-5-( 3,4-dichlorophenyl)-6-ethylpyrimidine]. cS180 = Sarcoma 180, 1,1210 = L1210 lymphoid leukemia, Ca755 = hdenocarcirionia i 3 5 , LL = Lewis lung carcinoma. d Toxic.
FurstI4 has studied thtl possible role of metals and has concluded that they play a significant and important role in cancer. Anticancer Ligands.-Another aspect of this research on ihe physiologicd activity of complex inorganic com(14) .A. l'urst, ref b, CLapter i b , p 339.
May 1966
ANTICANCER AND ANTIVIRAL ACTIVITYOF COMPLEX IKOHGANIC COMPOUNDS ACTIVITY OF
TABLE I1 FOL-ND TO BE
h l E T A L COMPLEX COYPOCNDS
371
ACTIVE .\S A4KTITUMOR AGENTS
-Animal tumor activity--
% Cornpd
I
Test system
S180'
Ca755"
Ca7.55'
I1
S180'
Ca755'
I11
S180'
Ca755"
Daily dose, mdkg
Survivors
150 100 63.6 41.4
616 616 616 616
-3.2 -2.0 -0.8 -1.7
9211176 12211176 29211176 34911176
7 10 24 29
28.0 28.0 28.0 28.0 28.0 28.0 32.0 16.0 8.0 4.0
10/10 10/10 9/10 9/10 9/10 9/10 loll0 loll0 loll0 loll0
-3.7 -2.5 -6.6 -3.5 -5.3 -2.3 -5.0 -2.6 -3.4 -2.5
231930 01984 011439 011544 22 11629 411436 1811699 011699 7111699 8911699
2 0 0 0 1 0 1 0
5/6 416 416 5 16
-1.2 -0.3 -1.6 -3.2
15011134 1191714 1061901 31811740
13 16 11 18
10110 10110 9/10 8/10 10110 7/10 10/10 9/10 loll0
-0.9 -1.6 -1.5 -4.1 -4.4 -2.8 -5.3 -4.0 -3.1
6811427 24711557 56911706 8411106 19711684 6811603 2411150 211150 122/1150
4 15 33 7 11 2 0 10
112 56 28 14
616 616 616 616
-3.7 -2.8 -0.9 -1.2
17011028 24911028 35811028 75011028
16 24 34 72
46 23 11 46 23 11 10 8 6
7/10 loll0 10/10 9/10 10110 10/10 loll0 10110 10110 loll0 loll0 loll0 loll0 loll0 10/10 10110 10/10 loll0
-3.6 -1.9 -2.3 -4.6 -3.5 -1.9 -1.0 -1.2 -1.6 -1.6 -0.3 -1.7 -0.9 -1.2 -1.9 -0.1 -1.1 $0.1
2111379 011379 3511379 1711803 511803 5711803 01485 51485 01485 101485 1081485 1711021 1611021 63/ I 02 1 35/ I 02 1 42911021 40611021 61911021
10110 10/10 10/10 loll0 loll0 10/10 919 10110 loll0 loll0
-0.1 - 113 -2.7 1.3 0.8 -2.8 -0.9 -1.2 -0.7 -1.8
1211042 1411042 4311042 497/1042 41611042 1621842 2411842 7 181842 8171842 331706
100 100 100 100 9.00 9.00 9.00 9.00 9.00 9.00 36.00 18,00 9.00
4 2 10 8 6 4 2 1
0.5
IV
Ca755"
tumor w t decrease
240 120 60 30 15 30 15 7.5 3.25 1000
Curesa
-Specificity Confidence
testIndex
99.7
2.1
99.7
4.!)
99.7
2.6
99.7
3.9
95.0
2.4
99.7
7.S
4 5
4
5 10 7 6 9 4 10 9 10 8
8 8 6 5
0 0 3
1 1 6 3 42 39 60 1 1
4 47 39 19 28 85 07
4
