zyxwvut
zy
zyx
J. Med. Chem. 1982,25,952-956
952
test compound indicated above for 15 min. Experimental and
control tissues are subjected to five bath changes during the
incubation interval. Changes in bath fluid during the incubation
period are helpful in ensuring the reproducibility of tissue responses to the agonist. Control tissues are incubated with test
compound vehicle (if any). The same concentration of the agonist
is reapplied in the presence of the test compounds, and the response is registered and compared with controls. Percent inhibition produced by the test compound is calculated by subtracting the mean percentage change in control tissue from the
mean percentage change in tissues exposed to the test compound.
Additional compounds are then evaluated as long as the tissue
remains reporducibly responsive to the agonist. Six tissues ob-
tained from six animals are used simultaneously-three controls
and three experimental. Partially purified guinea pig SRS-A was
prepared and purified as described.B FPL-55712 was used as a
reference for each compound tested.
Acknowledgment. The authors are grateful to Dr. C.
Kaiser for advice and encouragement. Our thanks to Ms.
I. Uzinskas for preparing 11 and the Analytical and
Physical Chemistry staff at Smith Kline & French Laboratories for performing the mass spectra and elemental
analyses. We also thank Ms. R. Osborn, Ms. M. Graus,
and Mr. E. Poserina (deceased) for the biological screening.
zyxwvutsr
Synthesis and Antitumor Activity of New Platinum Complexes
David B. Brown,* A. R. Khokhar, M. P. Hacker, L. Lokys, J. H. Burchenal, R. A. Newman, J. J. McCormack,
and David Frost
zyxwvutsr
Vermont Regional Cancer Center and Departments of Chemistry and Pharmacology, University of Vermont, Burlington,
Vermont, and Memorial Sloan-Kettering Cancer Center, New York, New York 10021. Received June 26, 1981
A new type of antitumor platinum complex has been prepared and examined for antitumor activity against L1210
leukemia both in vitro and in vivo. The coordination environment of platinum in these complexes consists of three
anionic chloride ions and a positively charged amine. The positive charge is introduced by monoprotonation or
monoalkylation of a diamine. Platinum(1V) derivatives have been prepared for several of the complexes, and a
water-soluble sulfate derivative has been prepared for one of them. Several of these complexes exhibit significant
in vitro activity, and trichloro(3-aminoquinuclidinium)platinum(II)(QTP) exhibits significant in vivo activity as
well. An increase in life span of approximately 40% has been observed using QTP. QTP is toxic at doses slightly
in excess of effective doses.
The effectiveness of transition-metal complexes, particularly platinum complexes, as experimental antitumor
agents has been demonstrated repeatedly in recent years.l
cis-Dichlorodiammineplatinum(I1)(PDD, cisplatin) is the
prototype metal complex with antitumor activity and is
currently available for clinical use in testicular tumors,
ovarian carcinoma, and other tumor typessZ A large
number of platinum complexes have been evaluated in an
effort to identify compounds with a broader spectrum of
clinical applicability than PDD and lower toxicity than
PDD. The vast majority of platinum complexes that have
been examined are of the general form cis-A2PtXz,where
X is an anionic ligand, typically chloride, and A is an amine
(or Az a chelating diamine).
The mechanistic details of the reaction(s) of platinum
complexes with cellular macromolecules are still not established satisfactorily, but it is generally agreed that cytostatic activity results from substitution reactions involving displacement of anionic ligands from the metal
~ o m p l e x . The
~ rates of substitution reactions at squareplanar platinum(I1) are dominated by the trans e f f e ~ t . ~
All other factors being comparable, the relative rate of
displacement of a particular ligand, X, is dependent on the
nature of the ligand, Y, occupying the position trans to it
in the complex (eq 1). PDD has a leaving group, the
L
I
Y-Pt-x
IL
L
2
Y-Pt-z
I
L
chloride ion, that is situated trans to ammonia. Amines
are at the lower end of the trans effect series; consequently,
the substitution reactions of C1- in PDD are relatively
sluggish. The reactivity of PDD, and possibly its biological
activity, could be enhanced by replacing ammonia with
ligands that are higher in the trans-effect series. Replacing
ammonia with anionic ligands would produce negatively
charged complexes that might not penetrate target cells
readily. However, the desired complex neutrality could
be obtained if one ammonia were replaced by an anion and
the other ammonia were replaced by a cationic ligand. We
have chosen to introduce the chloride ion, which lies above
ammonia in the trans-effect series, as the new anionic
ligand and to use monoprotonated (or monoalkylated)
diamines as the cationic ligands. This results in a series
of compounds having the general structure I, where NLl
I
Cl-Pt-N-”Hi
I
CI
I
NH+ represents a protonated diamine. The trans effect
dictates that a chloride ion lying trans to another chloride
will be substitutionally labile, and the symmetry of the
complex provides two activated chloride ions that can serve
as the initial leaving group. This report describes the
synthesis and biological activity of a series of complexes
of type I.
Discussion
S y n t h e s i s and C h a r a c t e r i z a t i o n of Complexes.
There have been a few reports Of platinum complexes with
positively charged amine ligands, but in most instances the
complexes were prepared only as precursors for studies of
zyxwvutsrqp
zyxwvutsr
t x
(1)
(1) Cleare, M. J.; Hydes, P. C. Met. Ions Biol. Syst. 1980,11,1-62.
(2) Einhorn, L. H.; Williams, S. D.N . Engl. J. Med. 1979,300,289.
(3) Rosenberg, B. Cancer Treat. Rep. 1979, 63, 1433.
(4) Hartley, F. R. Chem. SOC.Rev. 1973, 2, 163.
0022-2623/82/1825-0952$01.25/0 0 1982 American Chemical Society
zyxwvu
zyxw
zy
zyxwvutsrqp
zyxwvutsrqpo
zyxw
Journal of Medicinal Chemistry, 1982, Vol. 25, No. 8 953
New Platinum Complexes
Table I. Chemical and Biological Data for Complexes of the Type L'HRCl,
ID,,, d m L
formula
vpt-cl, cm-,
anal. a
Ll210/0
L121OIPDD
c, H
>10 ( > l o ) >10 (5)
315, 290
C,H,N,Cl,Pt
C,H,,N,Cl,Pt
2
3-[(methylamino)methyl]pyridine
300, 290
c, H
C,H,,N2Cl,Pt
3 2-la-(met hy1amino)et hyl]pyridine
300, 290
c, H
H; Cc
C,,H,,N,Cl,Pt
300, 285
4 442-piperidinoethy1)pyridine
C,H,,N,Cl,Pt
310, 290
H; Cd
5 2-(aminoethy1)pyridine
C,,H,,N,Cl,Pt
6 l,2-bis(4-pyridy1)ethane
290
c, H
c, H
>10 (4)
>10 (5)
7
nicotine
C,,H,,N,Cl,R
310, 290
308, 290
H, N; Ce
8
1,4-diazabicyclo[2.2.2]octane(dabco) C,H,,N,Cl,Pt
9 piperazine
C,H,, N,Cl,Pt
c, H, N
310, 290
10 N-methylpiperazine
C,H,,N,Cl,R
302, 285
c, H, N
9
>10
11 3-aminoquinuclidine
C,H,,N,Cl,Pt
310, 300, 280 C, H, N
2 (>lo)
5f ( > l o )
12 N-aminopiperidine
C,H,,N,Cl,Pt
300, 285
c, H, N
4.2
> 10
13 2,5-dimethylpiperazine
C,H,,N,Cl,Pt
305, 285
C, W; Ng
10
>10
14 3-aminopyridine
C,H,N,Cl,Pt
305, 290
c, H
Analytical values were all within i0.4%, except where noted otherwise. In all cases, chloride analyses were also obtained
and confirm the formulations given.
Figures in parentheses obtained with drug solubilized in Me,SO; where no figures are
entered for a compound this signifies that the ID,, exceeds 1 0 Mg/mL. C: calcd, 29.27; found, 28.58.
C: calcd,
19.80; found, 19.14. e C: calcd, 17.39; found, 16.65. f Average of values for five independent determinations on four
different preparations. g N: calcd, 6.73; found, 6.22.
no.
1
L
3-(aminomethy1)pyridine
zy
Table 11. Chemical Data for Platinum(1V) Complexes of the Type L+PtCl,X,
method
of prepano.
L
X
formula
cm
ration
15
N-methyldabco
OH
C'7H17N202a3R
300
D
10
3-aminoquinuclidine
OH
C7H1,N202a3R
312
D
17
3-aminoquinuclidine
c1
C,H, ,N, c1,R
310, 285
E
18
nicotine
OH
C10H1,N202a3R
310, 280
D
19
3-[(methylamino )methyl]pyridine
OH
C,H13N202C13R
310, 280
D
20
3-[(methylamino )methyl]pyridine
c1
c,HI, N, a ,R
310, 290
E
21
1,1,4-trimethylpiperazine
OH
C,H19N202a3R
D
a All of the compounds in this table were inactive when evaluated against L1210 in vitro, except for compounds 16 and
17 (see Table 111).
VPt$l,
the kinetics of ring closure upon deprotonation. T o our
knowledge, only three complexes having structure I have
been characterized. Terzis6 prepared and determined the
crystal structure of trichloro(9-methy1adeninium)platinum(II), Maresca et aL6 prepared trichloro(tetramethy1ethylenediammonium)platinum(II),and Adeyemo et alS7
prepared trichloro(4-amino-2,6-dimethylpyrimidine)platinum(I1). We find that a large number of complexes of
type I can be prepared using the two general reactions
outlined in eq 2. The composition of each complex was
t
KZPtCI4
t
HCI
'
(2)
HN-N--bt-Cl
N-N'ZHCI
t
KzPtC14
t NaOH
/
I
CI
verified by C, H, and (generally) N microanalysis (Table
I). Under the preparative conditions employed, a variety
of products or impurities [e.g., (N-N),PtCl,, (NNH+)2PtC14,etc.] are plausible and might not be readily
detected by C, H, and N analysis. However, these alternative formulations differ significantly in chloride ion
content, and we have consequently obtained C1 analyses
for all complexes prepared. These data, combined with
the other analytical data, demonstrate unequivocally a
ratio of 3Cl/diamine/Pt. Although the analytical data for
a few of the compounds do not fall within generally accepted limits, purification was generally impossible due
( 5 ) Terzis, A. Znorg. Chem. 1976,15, 793.
(6) Maresca, L.;Natile, G.; Rizzardi, G. Znorg. Chim.Acta 1980,
38,137.
(7) Adeyemo, A.; Teklu, Y.; Williams, T. Znorg. Chim. Acta 1981,
51, 19.
reactions.
Several derivatives of structure I, desirable because of
enhanced aqueous solubility and because they possessed
characteristics useful for possible structure-activity studies,
were prepared by the reactions summarized in eq 3. These
LPtC13 + Ag2S04
LPtCl(SOJ.Hz0 + 2AgC1
--
zyxwvutsrqpon
CI
N-N
to both low solubility and potentially interfering hydrolysis
LPtC13 + 2H202
LPtC1,
LPtIVC13(0H)2+ 2H20
+ H202+ 2HC1-
+
LPtIVC15 2H20
(3)
last two reactions, which represent oxidation to Pt(IV), are
difficult to control. Furthermore, the Pt(1V) complexes
are somewhat unstable and subject to an apparent hydrolysis reaction. Consequently, analytical results for some
of them are not totally satisfactory. Again, however, the
combination of C, H, N, and C1 analyses (obtained for all
of these complexes) are adequate to establish the ratio of
3(or 5)Cl/diamine/Pt/2(or 0)OH.
Certain diamines could not be induced to form complexes of type I. Powerful chelators, such as ethylenediamine, apparently do not provide a stable complex that
is coordinated a t only one nitrogen. Amines containing
two equivalent but electronically remote nitrogen atoms,
such as 1,2-bis(4-pyridyl)ethylene,generally protonate
simultaneously and lead to tetrachloroplatinate salts,
(LH22+)PtC14.Finally, some amines fail to produce the
desired complexes for reasons that are not readily identified.
We have examined infrared spectra of all of these complexes. The spectra are, in general, complex and are not
open to simple interpretation. In general, however, these
compounds show two absorption bands in the platinum-
zyxwvutsrqpo
zyxwvutsrq
zyxwvutsrq
954 Journal of Medicinal Chemistry, 1982, Vol. 25, No. 8
chlorine stretching region (Tables I and 11). Three infrared active modes (2A1 + BJ are expected on the basis
of symmetry considerations, but only two distinct absorptions are usually observed in the related ions
(LPtC13)-.8
It is a difficult problem to determine which of the two
nitrogen sites of the diamine is coordinated to platinum
and, consequently, which nitrogen site is protonated. It
is probable that the specific isomeric form that is produced
is a consequence of some complex interplay of both kinetic
and thermodynamic factors. In particular, its identity will
depend upon the Lewis basicity of each nitrogen donor site
toward the Lewis acid platinum in the form PtClC. Since
knowledge of that Lewis basicity does not exist, we have
no predictive ability. In the case of compound 11,we have
tried to resolve the structural question by examination of
infrared spectra in the N-H stretching region. Comparison
of spectra for 3-aminoquinuclidine, its mono- and diprotonated hydrochlorides, and compound ll suggests that
in the platinum complex it is the NH2 group which is
protonated. Specifically, a broad and intense band a t ca.
2800 cm-l, which can be assigned to the NH+ group in the
mono- and dihydrochloride, is absent from compound 11.
Infrared spectra thus suggest structure I1 for compound
Brown et al.
Table 111. Activity of 3-Aminoquinuclidinium
Complexes against L1210 Cells in Vitro
compd
ID,,, M/mL
L1210/
L1210/
0
PDD
zyxwvu
H3&4t
--cl
CI
I1
11. Although we were able to prepare single crystals of
compound 11, the crystals proved to be twinned and,
consequently,not usable for X-ray structure determination
purposes.
Biological Studies. All compounds reported have been
evaluated as inhibitors of the growth of murine leukemic
cells (L1210) and a PDD-resistant subline (L121OIPDD).
We have chosen a 50% inhibitory dose (IDm)of 10 pg/mL
as our upper-limit criterion for a significant level of activity. Compounds that satisfy the criterion are evaluated
further for ability to prolong the life span of mice inoculated with L1210 cells.
It must be stressed that there is no direct evidence
concerning the forms of the compounds in Table I under
conditions of biological evaluation. (Indeed, to some degree
the same statement applies to PDD.) The acid-base (eq
4) and hydrolysis (eq 5) equilibria will be established in
(HNWN)PtC13 + H+
[(N-N)PtCl,]-
+ H2O
+ [(N-N)PtCl,]-
(N-N)PtCl,(H,O)
LPtC1, ( 1 1 )
LPtCl, ( 1 7 )
LPtCl,(OH), (16)
LPtCl(SO,)
2
4.4
4.5
4.6
5
>10
> 10
> 10
Table IV. Activity of
Trichloro( 3-aminoquinuclidinium)platinum(11)
against L1210 Cells in Vivon
no. dose, mg/kg
schedule
% T/C
lb
5
1
123
1
125
10
1
toxic
50
100
1
1, 2, 5, 6, 9
1, 2, 5, 6, 9
138
5
1, 5 , 9
110
2c
0.62
1, 5, 9
122
1.25
1, 5, 9
126
2.50
1, 5, 9
120
5
1, 5, 9
130
10
1, 5, 9
133
20
1, 5, 9
119
3
2.5
1. 5. 9
122
5
10
1; 5; 9
139
20
1, 5, 9
140
40
1, 5, 9
148
~
a Because of the low solubility of this compound, data
from different laboratories may not be directly comparable due to different sonification/suspension/solubilization
procedures.
Data obtained at the University of
Vermont.
Experiments 2 and 3 represent data obtained
from the NCI utilizing two different screening contractors.
*
solubility rather than inherent inactivity of the molecule.
Dipolar aprotic solvents, such as dimethyl sulfoxide, are
frequently used to solubilize compounds for biological
studies, and, as indicated in Table I, parallel experiments
were carried out, in vitro, on some of the compounds
dissolved in Me2S0. In several instances, the complexes
solubilized in Me2S0 exhibited enhanced activity over that
observed when water was used as the administration vehicle; in the case of the trichloro(3-aminoquinuclidinium)platinum(II) (compound 1l),activity was
lower when the complex was solubilized with Me2S0. It
is difficult to interpret the results of studies obtained in
the presence of Me2S0,since Me2S0 can react with complexes of this typee by sequential substitution for chloride
ions. Thus, a mixture of Me2SO-containing species may
be present in the cell culture experiments.
Most of the compounds that we have prepared have
relatively little activity as inhibitors of the growth of L1210
and L1210/PDD cells in vitro (Table I). The most interesting compound in terms of biological activity is compound 11 with 3-aminoquinuclidine as the ligand, especially in view of its ability to produce roughly equivalent
inhibition of the growth of both L1210 and L121OIPDD
cells. Several compounds related to compound 11 were
evaluated in vitro. These compounds (Table 111)had activity against the “standard” L1210 cells comparable to
that of compound 11and were less effective against L1210
cells resistant to PDD.
Compound 11 was administered intraperitoneally to
mice with L1210 leukemia. Studies in our laboratory re-
zyxwvu
+ C1-
(4)
(5)
solution, and the dominant species a t physiological pH will
be a function of the pK of the coordinated amine, the
chloride ion concentration, and time. For the majority of
the diamines employed in our studies, the pK‘s are such
that the protonated form should predominate. (For example, we have measured the two pK, values for protonated 3-aminoquinuclidineas 8.6 and ca. 11.) It is of course
possible that coordination of one nitrogen center to platinum will alter the pK of the other nitrogen center, although we do not expect this effect to be large.
We also must emphasize that the water solubility of the
compounds in Table I is not high; the poor activity observed for several of the compounds may reflect poor
(8) Goodfellow, R. J.; Goggin, P. L.; Duddell, D. A. J. Chem. SOC.
A 1968, 504.
zyxw
(9) Tobe, M. L.; Khokhar, A. R. J. Clin. Hematol. Oncol. 1977, 7,
114.
New Platinum Complexes
zyxwvu
zyxw
zy
zyxwvut
Journal of Medicinal Chemistry, 1982, Vol. 25, No. 8 955
vealed that the compound has significant antileukemic
activity. For example, when given as a single dose (10
mg/kg; day 1) the compound gave a percent T / C value of
125%,and the percent T/C value was increased somewhat
(to 138) when a more intensive schedule of administration
was employed (5 mg/kg; days 1,2,5,6,and 9). Our resulta
were confirmed by independent studies carried out under
the auspices of the Developmental Therapeutics Program,
National Cancer Institute (Table IV). We also evaluated
the soluble sulfato derivative of compound 11 in vivo, but
we found it to be more toxic than compound 11 and to be
ineffective against murine leukemia a t a nontoxic dose (5
mg/kg). The biological activity of compound 11 is not
shared by the ligand per se since 3-aminoquinuclidine
hydrochloride was found to exhibit no cytotoxicity in vitro
under conditions in which the complex was quite active.
We have examined several other compounds for in vivo
activity, A complex which satisfies our cell culture criterion
for activity (compound lo), one which fails to satisfy our
criterion (compound 8), and an N-alkylated diamine complex [trichloro(N,N,iV-trimethy1piperazinium)platinum(II)] all have been studied in mice, but in none of these
cases was significant activity observed a t any dose.
The activity of compound 11 in the experimental leukemia system prompted us to carry out toxicological
studies on this complex. When administered as a single
intraperitoneal injection, compound 11 caused no immediate signs of toxicity, but within 48 h, mice became
lethargic, developed piloerection, and displayed signs of
tetany. The toxic effects progressed until paralysis of the
hind limbs was observed. Deaths occurred 3 to 8 days
following treatment. Gross necropsy revealed signs of renal
toxicity, including pale renal cortex, friability, and exaggerated corticomedullary border. Other grossly observable
signs of toxicity were sharply reduced spleen size and
distention of the small intestine. The calculated LDlo,
LDW, and LDw values were 22, 42, and 83 mg/kg, respectively. Toxic signs and gross necropsy findings were
similar in mice given compound 11 as five daily intraperitoneal injections. The LDlo, LDm, and LDw values for
the latter regimen were 10,25, and 48 mg/kg, respectively.
Our studies of the biological activity of compound 11
indicate that complexes of the type LPtC13 represent a new
class of antitumor platinum coordination compounds. We
are currently engaged in studies of the preparation and
evaluation of new examples of this class of compound that
may be more effective and less toxic than compound 11.
the reaction mixture was stirred at room temperature for 3 to 4
h. The orange-yellow product was filtered off and then washed
with water, ethanol, acetone, and finally ether. The product was
dried over Pz05under vacuum: yield 65%.
Method B. Trichloro(3-aminoquinuclidinium)platinum(11) (11). 3-Aminoquinuclidinedihydrochloride (0.199g, 1 mmol)
was dissolved in water (5 mL) and NaOH (0.04g, 1 mmol) was
added. The resulting solution was added to a filtered aqueous
solution of K2PtC14(0.415g, 1 mmol), and the reaction mixture
was stirred at room temperature for 24 h. The pale orange-yellow
product was filtered off, washed with water, ethanol, and then
acetone, and finally dried over P206under vacuum: yield 72% .
Method C. Trichloro(N-methyldabconium)platinum(II).
K2PtC14(0.415g, 1 mmol) was dissolved in water (5mL), and the
filtered solution was treated with N-methyldabconium chloride
(0.162g, 1 mmol) in water (5mL). The reaction mixture was
stirred at room temperature for 4 to 5 h. The orange-colored
product was filtered off and washed with an excess of water,
ethanol, and acetone. Finally, the product was dried under
vacuum: yield 72%. Trichloro(l,l,4-trimethylpiperazinium)platinum(I1) was also prepared by this method. Neither of these
compounds exhibited significant cytotoxicity.
Method D.
Trichlorodihydroxo(3-aminoquinuclidinium)platinum(IV)(16). Hydrogen peroxide (10
mL, 30% aqueous) was added to a heated and stirred suspension
of trichloro(3-aminoquinuclidium)platinum(II) (0.05g) in water.
There was a vigorous effervescence, and the product was obtained
in a quantitative yield by evaporating the resulting solution to
dryness. The brown product was dried over Pz05under vacuum.
Method E. Pentachloro(3-aminoquinuclidinium)p1atinum(1V) (17). PtC13(0H)z(3-aminoquinuclidinium)
was prepared
as in method D but not separated from the reaction mixture.
When effervescence had ceased, concentrated hydrochloric acid
was added, and the mixture was heated further for a few minutes.
On cooling, a nearly quantitative yield of the yellow precipitate
was filtered off and washed with water, then acetone, and finally
ether. The compound was dried under vacuum.
Method F. (3-Aminoquinuclidinium)monochlorosulfatoplatinum(I1). Trichloro(3-aminoquinuclidinium)platinum(II)
(1.0g, 2.3 mmol) was suspended in water (20 mL), and a solution
of AgzS04(0.725g, 2.3 mmol) in water (100mL) was added. The
reaction mixture was stirred at room temperature for several hours
in the dark. The precipitated AgCl was filtered off, and the yellow
filtrate was evaporated to dryness using a rotary evaporator. A
quantitative yield of brown-yellowproduct was obtained and dried
over PzO5 under vacuum: IR Y 1140 (SO,) cm-'.
Biological Studies. Mouse leukemic L1210 cell lines studied
at Vermont were obtained from Dr. Burchenal's laboratory at
Sloan-Kettering Institute (Rye, NY) and were maintained in
McCoy's 5A medium containing 10% fetal calf serum (Grand
Island Biological Co., Grand Island, NY). L1210/0 cells were
highly sensitive (IDrn= 0.5pg/mL) to PDD.12 PDD controh were
periodically run to verify the continued viability of the screen.
Compounds to be tested were dissolved in distilled water or
dimethyl sulfoxide (Me2SO);compoundsthat were poorly soluble
in water were also prepared in suspension. Stock solutions were
prepared at constant ratios up to 500 times that required in the
growth medium, so that 10 pL of stock solution could be added
to 5 mL of inoculated growth medium; the maximum concentration of Me@O in incubation media was 0.2%, a concentration
that had no effect on cell growth or viability. Cells were inoculated
into media at a concentration of approximately 5 x IO4 cells/mL
and allowed to grow for 96 h in 5% COzat 37 "C. Control cells
grew to a density of 1.2 X lo6cells/mL; cell density was measured
by use of an electroniccell counter (Coulter Counter, Model ZF).
To evaluate the activity of compounds against L1210 in vivo,
we employed the following procedure, L1210 cells (1 X lo6
suspended in 0.1 mL of physiological saline solution) were inoculated intraperitoneally into BDF, mice (20-22g) and drug
treatment (ip) initiated 24 h after inoculation of leukemic cells.
Drugs were dissolved or suspended in 0.3% hydroxypropylcellulose
in saline ("Klucel", obtained from the National Cancer Institute,
zyxwvut
zyxwvutsrqp
Experimental Section
zyx
zyxwvu
zyxwv
Chemistry. Most of the complexes were obtained by the
representative methods described in the following paragraphs;
those that required special procedures are discussed individually.
All the ligand (diamines) were purchased from Aldrich Chemical
Co. and used without further purification. KzPtC14was obtained
from Matthey Bishop, Inc. 1,1,4-Trimethylpiperazinumchloride
and N-methyldabconium chloride were prepared by established
rnethods.l0J1 Infrared spectra of the complexes (asKBr pellets)
were recorded on a Beckman IR-BOAspectrophotometer (4000-250
cm-'1; Pt-Cl stretching frequencies are listed in Tables I and 11.
Elemental analyses were performed by Integral Microanalytical
Laboratories, Inc., Raleigh, NC.
Preparation of Complexes. Method A. [Trichloro(Ndabconium)platinum(II)] (8). 1,4-Diazabicyclo[2.2.2]octane
(Dabco; 0.112 g, 1 mmol) was dissolved in water (5mL), and 3
M HCl (0.33 mL) was added. The resulting solution was added
to a filtered aqueous solution of K,PtCI, (0.415g, 1 mmol), and
(10) Quagliano, J. V.;Banerjee, A. K.; Goedken, V. L.; Vallarino,
L. M.J. Am. Chem. SOC.1970,92,482.
(11) Murthy, A. S. N.; Quagliano, J. V.; Vallarino, L. M. Inorg.
Chim. Acta 1972,6, 49.
(12) Burchenal, J. H.;Kalaher, K.; Dew, K.; Lokys, L. Cancer
Treat. Rep. 1979,16, 1493.
956
zyxwvutsrq
zyxwvutsrq
zyxwv
zy
zyxwvutsrq
J.Med. Chem. 1982,25,956-960
NIH, Bethesda, MD). Animals that received no drug treatment
died between 7 and 9 days after inoculation of L1210 cells. All
animals were housed in central animal facilities having controlled
temperature, relative humidity, and photoperiods.
Toxicological studies were done in male CD1albino mice (18-25
g). Compound 11 was administered intraperitoneally as a suspension in Klucel such that 0.1 mL of suspension/lO g of body
weight delivered the desired dose. Two treatment schedules were
evaluated, i.e., a single intraperitoneal injection or five daily
intraperitonealinjections. Mice were observed daily for 14 days
after the final injection. The LDlo, LDm,and LDgovalues were
calculated for each treatment schedule using the probit analysis
method of Finney.13 Gross necropsy examination was performed
on all mice that died during the observation period as well as those
mice sacrificed at the completion of the study.
Acknowledgment. This work was supported by Grants
CA 24543 and CA 22435 from the National Cancer Institute (Vermont) and grants from the American Cancer
Society (27V) and the Hearst Foundation (SKI). We are
grateful t o Linda Mathews for exemplary experimental
assistance.
~~~~
~~~
~
~
~
(13) Finney, D. J. "Probit Analysis",3rd ed.; Cambridge University
Press: New York, 1971; pp 20-87.
Affinity Therapeutics. 1. Selective Incorporation of 2-Thiouracil Derivatives in
Murine Melanomas. Cytostatic Activity of 2-Thiouracil Arotinoids, 2-Thiouracil
Retinoids, Arotinoids, and Retinoids
zyxwvutsrqpon
zyxwvut
Frank Watjen,+ Ole Buchardt,*J and Eyvind Langvadt
Chemical Laboratory 11, The H . C. 0rsted Institute, University of Copenhagen, Universitetsparken 5, and The Fibiger
Laboratory, Nordre Frihavnsgade 70, DK-2100 Copenhagen, Denmark. Received November 16, 1981
The incorporation of 2-[3SS]thiouracil
and two of its derivatives into murine melanomas, in vivo, was studied. It
was confirmed [J. R. Whittaker, J.Biol. Chem., 246,6217-6226 (1971)] that 2-thiouracil has a marked affinity for
melanin-producing tissue and that an affinity for such tissue could be sustained by 5-substituted 2-thiouracils. A
series of derivatives of arotinoids and retinoids, with or without a 2-thiouracil group as a potential carrier to obtain
affinity for melanomas, was examined for cytostatic activity, in vitro. None of these showed significant activity
against murine melanomas.
The lack of tissue selectivity of the presently used cancer
chemotherapeutic drugs constitutes a major problem.
Consequently, the construction of chemotherapeutics that
show specific affinity for, in casu, melanoma tissue would
constitute an important improvement in such drugs.
It is known that 2-thiouracil (1) and 6-propyl-2-thiouracil
Scheme I
Scheme I1
' ' NAS
sJ?J
5&
H
H
I
2
(2) exhibit marked affinities for melanin-producing tissue
in vitro,l and 1 a similar affinity in vivo,2 where they
presumably act as false precursors for melanin.1*2
This affinity for melanin-producing tissue offers a
possibility for preparing new potent drugs against malignant melanoma, where primary tumors and metastases
often show a very high rate of melanin synthesis.
The necessary prerequisites for the development of an
anticancer drug based on a carrier capacity of 2-thiouracil
are (1)that 2-thiouracil can act as a carrier of substituents
into the target tissue, i.e., that the affinity of the thiouracil
moiety is of such a character that it is sustained in variously substituted derivatives, and (2) that substituents
with, for example cytostatic properties or substituents
capable of releasing, for example cytostatic drugs can be
inserted into the thiouracil nucleus, etc.
We are attempting to develop such drugs, and our initial
strategy was prompted by early reports that indicated that
retinoids might cause regression of prenoplastic lesions and
malignant skin lesion^."^ Furthermore, studies, in vitro,
have shown that retinoids can inhibit cell proliferation and,
+Universityof Copenhagen.
The Fibiger Laboratory.
0022-2623/82/1825-0956$01.25/0
idependently of this, stimulate melanogenesis.6 This,
combined with the fact that the incorporation of thiouracil
was known to be related to the rate of melanin synthesis,
constituted the basis for the present work, which indicates
that it is possible to design derivatives of 2-thiouracil with
affinity for murine melanomas.
Results
Chemistry. From a biochemical point of view, it appears most obvious to introduce substitutents in the 5- or
(1) J. R. Whittaker, J. Biol. Chem., 246, 6217-6226 (1971).
(2) L. Dencker, B. Larsson, K. Olander, S. Ullberg, and M. Yokota,
Br. J. Cancer, 39, 449-452 (1979).
(3) W. Bollag, Eur. J. Cancer, 10, 731-737 (1974).
(4) R. Lotan, G . Neumann, and D. Lotan, Cancer Res., 40,
1097-1102 (1980).
( 5 ) W. Bollag and A. Matter, Ann. N.Y. Acad. Sci., 359, 9-23
(1981).
(6) R. Lotan and D. Lotan, Cancer Res., 40, 3345-3350 (1980).
0 1982 American Chemical Society