505
J . Med. Chem. 1982,25, 505-518
In Vitro Growth Assays. A. Melanotic Melanoma (MzR) and Amelanotic Melanoma (Az). Toxicity was assayed by growth inhibition of two melanoma cell lines, M2R and Az. MzR is a melanotic cell line developed from B16 murine melanoma and was obtained from J. P. Mather.Ig Az is an amelanotic cell line cloned from MzR and has a 27-fold lower tyrosinase specific activity than MZRlsas assayed by the tyrosine hydroxylation method of Pomerantz.zO For growth studies, 1 X lo4 cells suspended in growth medium (a 1:l mixture of Dulbecco's modified Eagle's MEM and Ham's F-12 with 1.2 g/L of NaHC03, 15 mM N-2-hydroxyethylpiperazine-N'-2-ethanesulfonicacid, and 10% fetal bovine serum) were seeded into 60-mm tissue-culture dishes. Incubations were at 37 "C in a humidified atmosphere of 5% COz in air. After 24 h, the medium was aspirated off from above the cell monolayer and replaced with 4 mL of fresh medium containing various concentrations of the test substances. Growth was followed for at least 3 additional days by detaching the cells with trypsin and determining the cell number on a Coulter cell counter. Growth NH),3.06(2H,t,CHz,J=3.0H~),4.15(4H,q,OCHz,J=3.5 comparisons were made at a time when the control cultures had Hz), 4.95 (2 H, s, OCHJ, 6.18 (3 H, m, aromatic), 6.95 (1H, m, aromatic), 7.30 (5 H, s, aromatic). Anal. (Cz3H30Nz05~2HC1~ increased %fold, according to the following formula: O.5HzO) C, H, N, C1. growth (% of control) = N*-(m-Hydroxyphenyl)ornithine (IC). Amine 19b (9.5 g, net increase in cell number of test culture 22.9 mmol) was refluxed in 100 mL of 0.5 N KOH for 4 h. After x 100 net increase in cell number of control culture the solution cooled, the precipitates were collected and washed with a small amount of ethanol. The potassium salt was dissolved B. Human Epidermoid Carcinoma of the Nasopharynx in 20 mL of water, acidified with concentrated HCI, and refluxed and P-388 Murine Leukemia. These assays were carried out for 1.5 h. The solvent was evaporated to dryness, and the residue by the screening program of the Developmental Therapeutic was suspended in 30 mL of MeOH and filtered to remove the Program of NCI. potassium chloride. The filtrate was again evaporated to dryness, In Vivo Antitumor Testing. Antitumor activity was deand the residue was dissolved in 30 mL of concentrated HC1 and termined as percent T / C values, with T / C I 125% defined as stirred at room temperature for 3 h. The reddish solution was statistically significant. Dose-response studies were carried out extracted with CHCl, twice, and the aqueous layer was evaporated for each compound according to published National Cancer Into dryness. The crude product was purified on a CISreverse-phase stitute protocols.18 Treatment begins 24 h after intraperitoneal column, using HzO as eluent. The front band, R, 0.55 (HzO), (ip) tumor implant on days 1-9 with intraperitoneal doses of the fractions were pooled and lyophilized to give 0.7 g (21%) of the compound under investigation. Normal saline (0.9% NaCl) was final product, IC. This compound is very hygroscopic and gives used as a vehicle. no specific melting point: mass spectrum (as 4Me3Si), m / z (relative intensity) 512 (M+., 22), 497 (M - CH3, 1.7), 395 (M C02Me3Si, 3.1), 266 (33), 253 (17), 142 (66), 73 (100). Anal. (19) J. P. Mather and G. H. Sato, Exp. Cell Res., 120, 191 (1979). (20) S. H. Pomerantz, Science, 164, 838 (1969). ( C I ~ H ~ ~ N Z ~ ~ C ~C, Z H, . ~ /N,, HC1.Z ~ ) with ether 3 times, and the ether extracts were combined, dried over NazS04,and evaporated to dryness. The crude product was purified by chromatography on a silica gel column and eluted with CHCl,-EtOAc (9:1, v/v) mixed solvent. Fractions with R f 0.65 on silica gel plate (same solvent system) were pooled and evaporated to dryness to yield 11g (64%) of yellow oil. Samples for elemental analyses were purified by preparative TLC: NMR (CDC13)6 1.22 (6 H, t, CH3,J = 3.5 Hz), 1.75 (2 H, CHz, m), 2.60 (2 H, CHz, m), 3.08 (2 H, t, NHz, J = 3.0 Hz), 4.24 (4 H, q, OCHz, J = 3.5 Hz), 4.92 (2 H, s, OCHz), 6.15 (3 H, m, aromatic), 6.98 (1 H, m, aromatic), 7.25 (5 H, s, aromatic), 7.65 (4 H, m, aromatic); mass spectrum, m / z (relative intensity) 544 (M'., 56), 212 (loo), 122 (12), 104 (8.0),91 (75). Anal. (C31H32N207) C, H, N. N-(y,yDicarbethoxy-y-aminobuty1)-m-(benzyloxy)aniline (19b). This compound, a yellow oil, was prepared on a 5.5-g scale (95%) from 18b by the same procedure used for 19a: mp for the dihydrochloride 170 "C (decomposed); NMR (CDC13) 6 1.22 (6 H, t, CH3,J = 3.5 Hz), 1.90 (4 H, m, CHzCHz),2.65 (3 H, s, NHz,
Antitumor Agents. 2.' Bisguanylhydrazones of Anthracene-9,lO-dicarboxaldehydes K. C. Murdock,*pt R. G. Child,? Yang-i Lin,? J. D. Warren,? P. F. Fabio,? Ving J. Lee,? P. T. IZZO,?S. A. Lang, Jr.,+ Robert B. Angier,*it R. V. Citarella,* Roslyn E. Wallace,* and Frederick E. Durr* Department of Chemical Research and Department of Chemotherapy Research, Infectious Disease Research Section, Medical Research Division, American Cyanamid Company, Lederle Laboratories, Pearl River, New York 10965. Received July 14, 1981 9,lO-Anthracenedicarboxaldehyde bis[ (4,5-dihydro-lH-imidazol-2-yl)hydrazone] (bisantrene, VI-1) showed anticancer activity in mice vs. both leukemias and solid tumors. Increases in life span vs. the following neoplasms were: P-388 leukemia, 137%; B-16 melanoma, 122%; Lieberman plasma cell tumor, >85%; colon tumor 26, 150%; Ridgway osteogenic sarcoma, 85%. There were significant numbers of long-term survivors. Both DNA and RNA synthesis were strongly inhibited. The drug was resistant to biodegradation and was bound strongly to tissues; in monkeys the half-life for disappearance from serum was 6 days. Related hydrazones were synthesized, and structure-activity relationships are discussed. Two routes to ring-substituted anthracene-9,lO-dicarboxaldehydeintermediates were developed. T h e compounds which have been shown to bind t o DNA by intercalation2 have generally been condensed tricyclic aromatics with at least one basic function. Since several of these compounds are active as antitumor agent^,^?^ we synthesized and tested widely differing polycyclic aromatics with various basic side chains. We repently reported t h e synthesis and antitumor properties of one of these Department of Chemical Research. t Department of Chemotherapy Research.
corn pound^.^^^ It has been named mitoxantrone (1) and is now undergoing clinical trialse5 NHCHzCHzNHCHzCHzOH
-2HCI
~6
'd
NHCH,CH,NHCH,CH,OH 1
0022-2623/82/1825-0505$01.25/0 0 1982 American Chemical Society
506 Journal of Medicinal Chemistry, 1982, Vol. 25, No. 5
Murdock et al.
Table I. Cyclic Carbonates of cis-9,l O-Dihydro-9,10-ethanoanthracene-ll,l2-diols
anal.a or lit. no. R method % yield mp, "C formula mp, "C 1-1 H A 94 C1'7H17,03 259-260 260-262 I-2 1,2-benzo A 80 C21H1403 219-224' 220-224 I-3 2-methyl A 24 225-227 C18H1403 c, H 1-4 2-methyl A 51 183-185 C18H1403 c, H 1-5 9-cyano A 83 203-205 C 1 8 H l l No 3 C, H, N I- 6 9-methyl A 75 253-254 C18H140S c, H 1-7 1,4-dimethoxy d A 50 283-285 C,,H,,O,. 0.2 5H,O c, €3 1-8 1,4-dimethoxy A 25 255-260 C,,H1,0;0.5H,0 c, H 1-9 2,3-benzo A 11 250-251 '21 c, H 1-10 1,4-diacetoxy A 63 275-277 C,lH,,O, c, €3 1-11 2,3-dimethyl A 60 207-212 C19H1603 c, H 1-12 1,4-dimethyl A 94 225-245 C ,,H 16 0 3' 0.2 5H,O c, H 1-13 2-tert-butyl A 47 250-252 C H,, 0,.0.2 5 H,O c, H 1-14 2,6-difluoro A 45 240-245 C17H10F203 c, H, F 1-15 1-chloro A 90 242-250 H, C1; Ce 16 'lo 3 1-16 9,lO-dimethyl A 85 280-281 C19H1603 c, €3 1-17 2-chloro A 83 200-230 CI,H,,C103 C, H, C1 1-18 2-acetamido A 66 268-271 C19H15N04 c, H, N a Analytical results were within 20.4% of the theoretical values for all elements listed, except as shown in subsequent footnotes. Reference 7. Reference 8. Syn (or anti) racemic mixtures; the racemic mixtures I-3,I-4, 1-7, and 1-8 were obtained by fractional crystallization of the crude reaction mixtures from CH,CI,-CH,OH (1:3). e C: calcd, 68.3; found, 67.7.
Another compound synthesized as a potential DNA intercalator was 9,lO-anthracenedicarboxaldehydebis[ (43dihydro-lH-imidazo1-2-yl)hydrazone] dihydrochloride (VI-1, bisantrene hydrochloride; Scheme I). It inhibited splenomagaly in mice with Rauscher leukemia and showed a high degree of anticancer activity against both P-388 leukemia and B-16 melanoma in mice. The corresponding monosubstituted compound, 9-anthracenecarboxaldehyde (4,5-dihydro-lH-imidaz01-2-yl)hydrazone hydrochloride, was inactive. Therefore, a broad synthetic program was initiated to prepare "two-armed compounds analogous to
Scheme I CHO
I
I
CHO
\ -
N ''
VI-1.
This report describes the synthesis of a variety of -C= N- compounds derived from 9,lO-anthracenedicarboxaldehyde and ring-substituted analogues and includes a discussion of the structure-activity relationships among these compounds and some older antitumor agents. Chemistry. Anthracenedicarboxaldehydes formed bisguanylhydrazones (e.g., VI-1, Scheme I) readily and completely when the requisite aminoguanidines were used as their dihydrohalide salts. However, many aminoguanidines were available as monohydrohalides. With these salts, reaction was incomplete unless a 2nd equiv of acid was added. Methylation of VI-1 with excess methyl iodide in DMF gave a single product, VI-4. The indicated exocyclic N-methylation in VI-4 was established by an For part 1of this series, see K. C. Murdock, R. G. Child, P. F. Fabio, R. B. Angier, R. E. Wallace, F. E. Durr, and R. V. Citarella, J.Med. Chem., 22,1024 (1979);K. C. Murdock, and F. E. Durr, U S . Patent 4 197 249 (1980). S. Neidle, Progr. Med. Chem. 16, 151 (1979);H. S. Schwartz, Adu. Cancer Chemother., 1, l(1979). B. F. Cain and G. J. Atwell, Eur. J. Cancer, 10,539 (1974);M. Rozencweig, D. D. Von Hoff, R. L. Cysyk, and F. M. Muggia, Cancer Chemother. Pharmacol., 3, 135 (1979). R. E. Wallace, K. C. Murdock, R. B. Angier, and F. E. Durr, Cancer Res., 39, 1570 (1979). D. D. Von Hoff, E. Pollard, J. Kuhn, E. Murray, and C. A. Coltman, Cancer Res., 40, 1516 (1980); D. S. Alberts, K. S. Griffith, G. E. Goodman, T. S. Herman, and E. Murray, Cancer Chemother. Pharmacol., 5, 11 (1980).
CH31
* DMF
H
VI-1 VI-4
alternative synthesis from methylhydrazinoimidazoline, 2. The preparation of ring-substituted analogues of lead compound VI-1 required as intermediates the corresponding substituted 9,lO-anthracenedicarboxaldehydes. Some previously described syntheses6of these dialdehydes were not satisfactory, so two other methods were developed as outlined in Scheme 11. Newman and c o - w o r k e r ~treated ~ ~ ~ anthracene and sev(6) G. Rio and B. Sillion, C. R. Hebd. Seances Acad. Sci., 244,623 (1957); B. H. Klanderman, J. Org. Chem., 31, 2618 (1966). (7) M. S. Newman and R. W. Addor, J.Am. Chem. Soc., 77,3789 (1955).
Journal of Medicinal Chemistry, 1982, Vol. 25, No. 5 507
Bisguanylhydrazones of Anthracene-9,lO-dicarboxaldehydes
Table 11. cis-9,1O-Dihydro-9,10-ethanoanthracene-l l,12-diols R
no. 11-1 11-2 11-3 11-4 11-5 11-6 11-7 11-8 11-9 11-10 11-11 11-12 11-13 11-14 11-15 11-16
R H 1,2-benzo 2-methyl 2-methyld 9-methyl 1,4-dimethoxy 1,4-dimethoxy 1,4-dihydroxy 2,3-benzo 2,3-dimethyl 1,4-dimethyl 2-tert-butyl 2,6-difluoro 1-chloro 9,lO-dimethyl 2-chloro
R
method
% yield
B B B B B B B B B B B B B B B B
89 .90 95 76 92 81 84 80 58 69 96 97 81 85 96 75
formula
anal.a or lit. mp, "C
202-204 196-198 227-228 153-1 56 138-140 187-188 189-192 280-285 238-241 225-230 158-1 60 198-200 139-141 180-1 82 201-203 195-210
See footnotes a-d in Table I. e C: calcd, 81.2; found, 80.5.
f
F: calcd, 13.9; found, 13.2
Table 111. 9,1O-Dihydro-9,1O-anthracenecarboxaldehydes H
no.
R
method
%yield
CHO
mp, "C
formula
ana1.O:
111-1 H (cis) Cb& D 94 144-146 Cl,HlZO* c, H H (trans) 111-2 E 77 132-1 3 5 Ci6HizOa c, H 111-3 l,2-benzo (cis) C 50 170-172 CzoHi,Oa c, €3 111-4 2-methyl (cis) D 96 1 25-1 26 CI,H,,OZ c, €3 111-5 1,4-dimethoxy (cis) C 28 130-140 C18H1604 c, H 111-6 2,3-benzo (cis) C 39 170-1 7 2 C,OHl,OZ c, H 111-7 2,3-dimethyl (cis) H; Cc D 66 113-11 7 CiaHi6Oa 111-8 1,4-dimethyl (cis) D 95 159-160 C18H160Z c, H 111-9 1-chloro (cis) C 34 144-146 C16H,,C10,~0.25H~0 C, H, C! 111-10 9J0-dimethyl (cis) C 96 193-195 C18H160Z c, H 111-11 2-chloro (cis) C 33 113-115 C,6Hl,C10~~0. 25Hz0 C, H, C1 1,5-dichloro 111-12 E Ci6HioClzOa C, H, C1 55 179-188 a Analytical results were within +0.4% of the theoretical values for all elements listed, except as shown in subsequent footnotes. See ref 9. C: calcd, 81.8; found, 81.0. Reference 40.
era1 derivatives, 3, with vinylene carbonate. Subsequent hydrolysis gave cyclic diols 5, which were then oxidized to dicarboxylic acids7 or dihydrodialdehydes 6.8 We found that a slight modification of their lead tetraacetate oxidation procedure smoothly converted cyclic diols 5 to aromatic dialdehydes 7 in generally satisfactory yields. In addition, it was found that aqueous sodium periodate under controlled conditions oxidized cyclic diols 5 exclusively to the dihydro dialdehydes 6, which could then be oxidized to the aromatic dialdehydes 7 by mild oxidizing agents. In two cases, as shown in Tables I and 11, the cyclic carbonates 4 and cyclic diols 5 were separated into syn and anti isomers. However, such separations were not required, since mixtures of isomers were also converted to the symmetrical aromatic dialdehydes 7. A second dialdehyde synthesis utilized the more readily available anthraquinones rather than anthracenes and has been described in detail by Lin and co-workers.* Reaction
(8) M. S. Newman and Z. U. Din, J. Org. Chem., 36, 966 (1971).
of anthraquinones 8 with dimethylsulfonium methylide gave dioxiranes 9, which were then rearranged to 10-(hydroxymethyl)anthracene-9-carboxaldehydes710, with LiBr in acetonitrile. Oxidation of 10 with Me,SO then gave the desired aromatic dialdehyde 7. Rearrangement of oxiranes 9 with BF3 gave dihydrodialdehydes 6. Structure-Activity Relationships. Thirty different aminoguanidine derivatives of 9,lO-anthracenedicarboxaldehyde are listed in Table VI. All but four (VI(9) (a) Y.4. Lin, S. A. Lang, Jr., C. M. Seifert, R. G. Child, G. 0. Morton, and P. F. Fabio, J. Org. Chem., 44, 4701 (1979). (b) Dimethyl derivative i was prepared from VI-4 by method X
(93% yield), mp 240-245 (Cz~HS~Ng2HI.0.5HzO) C, H,N. CH=R
"C
dec.
Anal.
508 Journal of Medicinal Chemistry, 1982, Vol. 25, No. 5
Murdock et al.
Table IV. 9.1 0-Anthracenedicarboxaldehvdes CHO
I
CHO
anal.' or lit. mp, "C formula mp, "C H 245-247 C16H1602 244-245' 1,2-benzo 193-198 C,OHI,O2 c, H 2-methyl 162-164 Cl7H12OZ c, H 1,4-dimethoxy 5 208-21 2 C,,H,,O;O. 25H20 c, H 2,3-benzo 20 215-217 C,OHI,O, c, H 2,3-dimethyl 47 203-204 C,,H,,OZ c, H 1,4-dimethyl 10 158-1 6 2 C1,H,,O;0.7 5H,O c, H 2-tert-butyl 31 125-1 26 C,OHl,O2 c, H 2,6-difluoro 32 240-242 C,,H,F,O;l. 5H,O C, H; Fd 2-ethyl 25 Cl8H140, H; Ce 99-100 1-chloro-2-methyl G 90 175-177 CI, H 11 c10 2 f 1-chloro F 96 186-1 89 C,,H,CIO, c, H 2-chlor0 F 94 193-196 CI6H,C10;0.25H,O C, H 1-fluor0 G 21 226-238 CI6H,F0;0.7 5Hz0 c, H, F 2-fluor0 G 82 214-222 C,6H9F02 f 1,5-difluoro G 61 24 8-2 50 CI,H,FZO, c, H, N, F a Analytical results were within i 0.4% of the theoretical values for all elements listed, except as noted in subsequent footnotes. Also prepared as in ref 40. Reference 6. F: calcd, 12.8; found, 11.7. e C: calcd, 82.4; found, 81.9. f The crude product was used in the next step. no. IV-1 IV-2 IV-3 IV-4 IV-5 IV-6 IV-7 IV-8 IV-9 IV-10 IV-11 IV-12 IV-13 IV-14 IV-15 IV-16
R
method Fb F F F F F F F F G
%yield 85 75 76
10,11,13,22) were accepted as active. None was clearly superior to the lead compound, VI-1; however, based on "percent increase in life span" (ILS) and "cures" in the P-388 leukemia and B-16 melanoma tests, the following compounds show activities approximately equivalent to VI-1: hydrazinoimidazoline derivatives VI-4 and -7, hydrazinotetrahydropyrimidine derivatives VI-2,5,8,9, hydrazinoazepine derivative VI-3, and substituted aminoguanidine derivatives VI-15 (2-methyl), VI-17 (2,2-dimethyl), VI-18 (2,3-dimethyl), and VI-25 (2-furfuryl). These results do not suggest a clear structure-activity relationship. However, a steric factor is suggested, since the most active compounds are those in which the aminoguanidine moiety is substituted with small groups or is involved in a cyclic structure. Conversely, larger substituents cause a decrease in activity that is moderate for most monosubstituted aminoguanidine moieties but very large for the marginally active 2,3-diisopropylaminoguanidine derivative VI-20 and the bulky cyclic compounds VI-10 and -11. Complete methylation of the imidazoline nitrogen atoms of VI-1, as in VI-13 or its isomer b:i eliminates all activity. Further substitution of the anthracene nucleus (Tables VI1 and IX) also has not produced any compound clearly superior to lead compound VI-1. Mono- and dihalogen substitution in various positions caused no significant change in activity (VII-1 to -6 and IX-1). This was also true for substitution with a 2-methyl group (VII-7). However, dimethyl derivatives VII-8 and -9, dimethoxy derivative VII-13, and 2-ethyl derivative VII-10 showed decreased activities, while 2-tert-butyl derivative VII-11 was completely inactive, indicating again a steric effect. If intercalation is indeed involved in the mechanism of action of these compounds, the inactivity of compounds such as VII-11 would be understandable, since the requisite sandwiching of the flat anthracene nucleus between adjacent base pairs of a DNA chain would be expected to be much diminished by a side chain bulky in three dimensions. The addition of another coplanar aromatic ring to give tetracyclic analogues VII-14 and -15 had only a slight
effect on activity. Equilibrium dialysis studieslO showed binding of VI-1 to DNA, which is probably intercalative. Further studies are in progress. An examination of the structures of other derivatives of 9,10-anthracenedicarboxaldehyde,listed in Table VI, shows that most of the derivatives with significant antitumor activity were hydrazones derived from aminoguanidines. Inactive compounds were derived from substituted hydrazines (VI-33-35,39,44), acylhydrazines (VI-37,38), semicarbazides and thiosemicarbazides (VI41-43), tosylhydrazide (V1-32), and dimethylethylenediamine (VI-36). The exceptions were aminoamidine derivative VI-31, which was moderately active against P-388 leukemia and marginally active against B-16 melanoma, and bisthiosemicarbazone VI-40, which has modest activity only against P-388 leukemia. The finding of activity in the nonbasic bisthiosemicarbazone VI-40 led us to synthesize the bis(dimethylaminopropy1) derivative VIII-1 with the hope of improving the activity of VI-40. This derivative (VIII-1) did indeed show considerably greater activity and potency than VI-40, indicating again the need for basic side chains to produce highly active compounds. Additional analogues of VIII-1 were prepared (Table VIII), but none was as active as the lead compound VI-1. A number of 9,lO-dihydro derivatives of lead compound VI-1 and its analogues were prepared and tested (Table V). None was superior to VI-1, and most of them were either less active or less potent. Activity may depend on aromatization by dehydrogenation in vivo since the 9,lOdimethyl derivative V-6 was inactive. (10) G. Nicolau, W. H. Wu, S. Gordon, D. Cosulich, and W. McWilliams, Antimicrob. Agents Chemother., abstr 26 (1980). Dialysis of 14C-labeledVI-1 at 2 X M against calf thymus
DNA at lo4 M (based on nucleotide content) required 72 h to reach equilibrium, when 66% of the radioactivity in the DNA compartment was bound to the DNA. The membrane used was Spectra/por #2 natural cellulose in 0.1 M NaCl with 0.01 M sodium cacodylate buffer, pH 6.00. We thank Dr. Gabriela Nicolau for these data. Further DNA binding studies are in progress.
Journal of Medicinal Chemistry, 1982, Vol. 25, No. 5 509
Bisguanylhydrazones of Anthracene-9,lO-dicarboxaldehydes
Table V. 9,1O-Dihydro-B,1 0-anthracenedicarboxaldeh yde Bis( imidazolin- 2-ylh ydrazones)
y,
,cH="HfJ
H
% I n c r e a s e i n Median L i f e Span No.
R
Method
% Yield
Mp,"C
Formula
A n a l ysesa
(Optimum Dose, mg/kg) B - 1 6 Melanoma'
'C u r e 9 I
P-388 Leukemiab
v- I
H
H
13
258-262
C22H24N8'2HCl
C,H,N,CI
72 (12.5)
C,H,N,CI
0 (ZOO)
C,H,N,CI
105 (225)
NT
C , H , C I ;N
73 (50)
NT
B-16
100 (12.5)
0
.0.5H20
V-2
1,b-dimethoxy
H
IO
250-255
CZ4Hz8N8O2.2HC1
NT (not tested)
. I .0H20
V-3
2,3-benzo
H
8
290-295
CZ6Hz6N8.2HCI ,0.5H20
V-4
2,j-dimethyl
H
35
285-290
Cz4Hz8N8'2HCl ,0.5P20
1
V-5
I-chloro
H
51
200 (de c )
C22H23N8Cl.2HC1
C;H,N,Cle
100 ( 5 0 )
3/ M
1 5 3 (50) t
.0.5H?O
V-6
9, IO-dimethyl
H
44
275-277
CZ4Hz8N8.2HC1
C,H,N;CI
0 (200)
0 (50)
' 1 .25H20
V-7
2-chloro
H
77
190 (dec)
C22H25N8C1.2HC1
C,H,CI;Na
V-8
I ,5-dichloro'
H
15
ZO5-2lO
C22H22N8C12.2HC1
C,H,Cl;N
100 (50)
0 (200)
230 (12.5)
5/ a
0 (50)
. C H OH
3 7
Although none of the analogues in Tables V-IX was considered to be clearly superior to VI-1, there was the possibility that some of the compounds might prove to be superior if tested further. Therefore, eight of the most interesting compounds were compared against L-1210 leukemia, colon tumor 26, Madison lung carcinoma 109, and the advanced disease forms of P-388 leukemia, B-16 melanoma, and L-1210 leukemia. The results of these tests (Table X)also indicate that none of these compounds is broadly superior to the lead compound VI-1. Details of a more extensive pharmacological testing of VI-1 will be reported separately.ll It was found that it gave an ILS of >85% in mice with the Lieberman plasma cell tumor, and 150% and 85% in mice with colon tumor 26 and Ridgway osteogenic sarcoma, respectively. It was active by intraperitoneal, intravenous, or subcutaneous injection, but was inactive orally. An adriamycin-resistant subline of P-388 leukemia was completely cross-resistant to VI-1. DNA and RNA syntheses were strongly inhibited (11) R. V. Citarella, R. E. Wallace, K. C. Murdock, R. B. Angier, and F. E. Durr, Antimicrob. Agents Chemother., abstr 23 (1980); Cancer Res., 42, 440 (1982).
in L-5178Y lymphoma cells in vitro.ll In cultures of human diploid fibroblasts (WI-38) and colon carcinoma cells (WiDr), it was lethal in both rapidly dividing and nondividing phases of the cell cycle, suggesting that it might be effective against slow-growing tumors.12 It was mutagenic in the Ames test with Salmonella typhimurium strain TA 98, but the mutagenic potency was much less than that of daunorubicin. Thus, there were 0.3 to 0.4 reverse mutations per nanomole for VI-1 as compared with 0.2 to 0.3 for mitoxantrone (1)13 and 100 f 30 for daunorubicin.14 The drug is surprisingly stable and persistent in vivo. In the rat, dog, and monkey, the half-lives for disappearance from serum of a single intravenous injection of 14Clabeled VI-1 were 1.5, 3, and 6 days, respectively. It was well distributed throughout the organs and glands studied. (12) R. E. Wallace, R. V. Citarella, and F. E. Durr. Antimicrob. Agents Chemother., abstr 28 (1980). (13) For these data we thank Dr. Jane S. Allen, American Cyan-
amid Co. Agricultural Research Center, Princeton, N.J. (14) G. L. Tong, H. Y. Wu, T. H. Smith, and D. W. Henry, J. Med. Chem., 22, 912 (1979).
Murdock et al.
510 Journal of Medicinal Chemistry, 1982, Vol. 25, No. 5
Table VI. 9,lO-AnthracenedicarboxaldehydeBishydrazones FH=R
@I@ I
CH4 R
1 " 1
Mp,'C
97
J
"H-$)'
VI-I
% Yieldb
Methoda
Formula
288-289 C22H22N8.2HC1 ,0.5H20
CHNCI
CHNC!
"
H
o
N
I
87
301-302
1
M e d i a n % I n c r e a s e i n L i f e Span (Optimum Dpses mg/kg) P-388 L e u k e r n l a d "Cures"e 8-16 melanomar "Cures"5
inalyses'
17/42
137 ( 1 2 . 5 )
I
82 ( 1 2 . 5 )
I
122 ( 6 )
I
1/12
I06 ( 1 2 )
2/10
1
0
C26H30N8'2HI
VI-4
r r
I
__ CHNI
I
35 ( 2 5 )
1/12
97 (25)
0
CHCl ;Nh
120 ( 1 2 . 5 )
L-
CHCl;Ni 205 (100)
CHCl,Nj
316
143 ( 2 5 )
I H
23
285-290
1/6
4/10 .____
~
130 ( 1 2 . 5 )
5/10
242 (6)
_ I _
N N H - m h :
1
0
127 ( 1 . 5 )
I
I
4/10 I
C40H58N8.2HCI.0.5H20
VI-IO
H .
49
VI-11
VI-I2
I VI-I3
VI-15
NH,
1,
I
w
N "H C :
VI-I4
N N "H C : NHCH,
I
1
N/,Cy~~
80
44
1 1 298-300
1
252-254
~
Cl8Hl8N8'2HCl.0.25H
C20H22N8'2HI'H20
20
CHNCI
1 3 2 (25)
CHN I
157 ( 2 5 )
CH
VI-16
CHNBr
84
U
"2 YNHC
VI-17
H " \N(cH,),
NNHCENCH3 NHCH,
VI-18
1
N(H20)
1
a7 61
1
320-322
281-283
CZ2Hz6N8.2HI
~
CHNI
j
1
N N H C"
VI-21
._
\"-@
1
p
Ether
1
53
>300"
1
91 ( 1 2 . 5 )
1
0
1 I03 (6.25)
1
8/20
1 4 5 (12)
1 I
3/18
50 ( 6 )
125
(3)
1
0
12/10
CZ2Hz6N8'2HI
CHNCI
1
1
6/24
C
8N8'2HI
90 ( 2 5 )
0
80 (12)
1/10
0
0
0
TABLE V I ( c o n t ' d ) Methoda % Y i e l d b
R
No.
Mp,"C
Formula
Analyses'
M e d i a n % I n c r e a s e i n L i f e Span ( O p t i m a l Doses mg/kg) P-388 Leukemiaa
VI-22
NH NHNC~NH(CH2)i5CH3
VI-23
VI-27
vI-28 VI-29
"Cures"'
0
0 (50)
0
I2
235-240
C50H82N8'2HC1
CHCl ;N1
70
279-282
C32H30N8.2H1
CHNl
68 ( 2 5 )
0
4 7 (12)
0
J
69
239-240
C 30 H2 6 N8 .2HCI,2H20
CHNCI
67
(SO)
0
5 0 (6)
0
0
167 ( 1 2 . 5 )
6/10
90
130-140
CHNCI
91 ( 5 0 )
J
C28H26N8S2.2HC1'0.5H20
CHN; S C l m
9 0 (50)
0
75 (12)
0
C30H28N10~4HC1~0.5H20
CHNCI
9 0 (12.5)
0
78 (25)
0
114 (25)
116
7 7 (25)
0
\"cH,@
+"
NNHC h H - 0
Vi-25
VI-26
5-16 melanomat
4"
"HC
V 1-24
0 (200)
'lCures"r
I ,NHCH2~v
NN H cdN
90
160-200
93
299-302
C
H
N 0 ,2HC1'0.5H20
26
NNHC"" \"cH,@
V
N.NHGzCH2&)
V
92
298-3021 C30H28N10.4HC1
CHN; C I n
NNHC
v
97
285-2901 C30H28N10'4HCl
CHNCl
95 (50)
0
74 (25)
0
CHN i
80 (200)
0
62 ( 2 5 )
0
CHNCI
95 (25)
0
35 ( 1 2 )
9
~CHNS o ~
0 (200)
0
NT
I
I
VI-30
NNHC C -