. __ 1 .7 3......31.
.rx. J3v S :3.

L. .

z
55;;‘53:

«
4.13.: :,
.2: I. .

2..

..::,.I.
v . J!

. : :A:.
#9.: v.1; v1
7

Jvlv
‘ ii 3.£..4):ur x. .
:.!1..a.5:.u( r 1:...
~L:a if.) 3 5
I? 17:

3 if, .

.7 531 2 .
u: fa: a ‘

J

T. .y
$35.:
a it:

I L 24
a.

 

 

. T ..,
12.92.:

a

£11} . p . . .1.» . . . Shh. . . ,1
#fiwkfiwfi , . . éfi.‘ fig kswfiwwwm a

THEsls

’7 x .
efibgl!

This is to certify that the

thesis entitled

\ . 3“" D
f . l‘ V b
KCBLC/Plo‘\l> 5": E): hucLeécr !‘&Vl$\+|C'H

Mela! Hn-hlwwp €0quth “,er Ha
DNA MOAK" ’UWCleol/flse q"E+t\riflu2mlhe

presented by

Kemcd \/. Catalan

has been accepted towards fulfillment
of the requirements for

‘V\€ degree in M15)!

 

I f" 2
£ ”,1- t ‘l k /( _\,-—. /,. / ‘x. .
I r' k L, L /

Major professor

Date’X/EZQ/qcl

 

0-7639 MS U is an Affirmative Action/Equal Opportunity Institution

 

team
Michigan State
I University

 

 

 

PLACE IN RETURN BOX to remove this checkout from your record.
To AVOID FINES return on or before date due.
MAY BE RECALLED with earlier due date if requested.

 

DATE DUE DATE DUE DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

6/01 c:/C|FiC/DateDue.p65-p. 15

 

REACTIONS OF DINUCLEAR TRANSITION METAL
ANTITUMOR COMPLEXES WITH THE DNA MODEL
NUCLEOBASE 9-ETHYLGUANINE

Kemal Vatansever Catalan

A THESIS
Submitted to
Michigan State University
in partial. fulfillment o f therequirements'
for the degree of
MASTER OF SCIENCE

Department of Chemistry

1999

Abstract

REACTIONS OF DINUCLEAR TRANSITION METAL ANTITUMOR
COMPLEXES WITH THE DNA MODEL NUCLEOBASE 9-ETHYLGUANINE

By
Kemal V. Catalan

Mononuclear transition metal compounds such as cisplatin, iproplatin and carboplatin are
well known for their extraordinarily high carcinostatic activity. Years of research have
been devoted to the elucidation of their mechanism of action. The generally accepted
model is the inhibition of DNA replication by the covalent binding of two adjacent
guanine bases through their respective N7 atoms to the platinum metal center. Other
transition metal compounds are known to exhibit considerable anticancer activity that is
also attributed to direct metal-DNA interactions. In this vein, we are investigating the
preferred DNA binding sites as well as the exact binding modes of dinuclear transition
metal carboxylates of the type M2(02CR)4, M2X4(02CR)2 and M2(DT01F)2(02CCF3)2 (M
= Ru, Rb and Re; R = CH3, CH2,CH3, CH2CH2CH3; DTolF = di-p-tolylformamidinate; X
= halide). Recent studies in our laboratories have elucidated unprecedented bridging
modes for the model nucleobases 9-ethy1guanine and 9-ethyladenine. In addition, the
synthesis and spectroscopic characterization of the reaction products of these metal
compounds with the twelve base pair oligonucleotide, d(5’-CCT CTG GTC TCC-3’),
have been performed. Other studies involving the polymerase chain reaction (PCR)
indicate that the DNA replication process is inhibited by covalent metal binding to the
template strand. The PCR results along with 1H NMR spectroscopic, HPLC and X-ray

crystallographic results will be presented.

TABLE OF CONTENTS

page
LIST OF ABBREVIATIONS .........................................................
CHAPTER I. INTRODUCTION ............................................... 1
1. Introduction ............................................................................. 2
A. Mechanism of Action of Cisplatin ..................................... 3
B. Intrastrand Cross-links .........................................................
8
C. Interstrand Cross-links .........................................................
10
2. Proposed Design Strategies ............................................... 12
3. Dinuclear Metal Complexes ............................................... 13
A. Dinuclear Rhodium Complexes ..................................... 14
B. Dinuclear Rhenium Complexes ..................................... 16
CHAPTER II. DIRHODIUM ANTICANCER AGENTS ................. 19
1. Introduction ............................................................................. 20
2. Experimental ............................................................................. 25
A. Synthesis ............................................................................. 25

(1) Preparation of ha(u-DTolF)2(u-9-EtGH)2(OzCCF3)2 (1).. 25
(2) Preparation of ha(u-DPhF)2(u-9-EtGH)2(02CCF3)2 (2)... 26
(3) Preparation of [Rh2(u-DT01F)2(CH3CN)6][BF4]2 (3) ......... 26
(4) Preparation of [Rh2(u-DTolF)2(u-9-EtGH)2(CH3CN)6][BF4]2 (4)

........................... 27
(5) Synthesis of [Rh2(u-DTolF)2(u-9-EtGH)2(CH3CN)6][BPh4]2 (5)

........................... 27

iii

B. X-ray Crystallography ......................................................... 27

(1) [Rh2(ll-DTolF)2(CH3CN)6][BF4]2 (3) ........................... 28
(2) [ha(p-DTOIF)2(u-9-EtGH)2(CH3CN)6][BF4]2 (4) ....... 29
C. 103Rh NMR Spectroscopy ............................................... 29
3. Results ............................................................................. 30
A. Spectroscopic Properties of (1) and (2) ........................... 30
B. Spectroscopic and Crystallographic Properties of (3) ....... 35
C. Spectroscopic and Crystallographic Properties of (4) ....... 42
D. 103Rh NMR Spectroscopy ............................................... 43
(1) Theory ............................................................................. 43
(2) Results ............................................................................. 49
4. Conclusion ............................................................................. 50
CHAPTER III. DIRHENIU M ANTICANCER AGENTS ................. 55
1. Introduction ............................................................................. 56
2. Experimental ............................................................................. 57
A. Synthesis ............................................................................. 57
(1) Preparation of cis-Re2(u-9-EtG)2Br4 (6) ........................... 58
(2) Preparation of Re2(u-9-EtG)2(u-OzCCH3)4C12 (7) ....... 58

(3) Preparation of Re2(u-9-EtG)2(u-OzCCHzCH3)4C12 (8a) 59
(4) Preparation of Re2(u-9-EtG)2(u-OzCCHzCH3)4(BF4)2 (8b) 59
(5) Preparation of Re2(u-9-EtG)2(u-02CCH2CH3)4(SO4) (8c) 59
(6) Preparation of Re2(u-9-EtG)2(u-02CC6H5)2(CH3CN)2(BF4)2 (9)

........................... 60

B. X-ray Crystallography ......................................................... 6O
(1) cis-Re2(u-02CCH2CH3)2Br4(DMF)2 (10) ................. 6O

3. Results ............................................................................. 61
A. Crystallographic Determination of (10) ........................... 61

iv

B. Spectroscopic Properties of (6) ........................... 62

C. Spectroscopic Properties of (7) ........................... 70

D. Spectroscopic Properties of (8) ..................................... 71

E. Spectroscopic Properties of (9) ..................................... 73
4. Conclusion ............................................................................. 74
LIST OF REFERENCES ................................................................... 8O

mol

mmol

NMR

ppm

Abbreviations

Angstrom

broad

Bohr magneton

wavenumber

degree centigrade

doublet (NMR), day, deuterated
parts per million (ppm)

molar extinction coefficient
wavelength

gram

mole

millimole

hour

infrared

coupling constant (NMR)
Boltzmann constant
megaHertz

mole per liter

medium, multiplet
milliliter

bridging ligand
nanometer

nuclear magnetic resonance
frequency

parts per million

vi

TMS
UV

DNA
RNA

ade
d(pGpG)
9-EtGH
9-EtAH
9-EtG'
9-EtA'
DTolF'
DPhF'

form

vii
singlet, strong
tetramethylsilane
ultraviolet
weak
halide
deoxyribonucleic acid
ribonucleic acid
guanosine
adenosine
adenine or substituted adenine bases
dimer of DNA containing guanine
9-ethylguanine
9-ethyladenine
9-ethy1guanine deprotonated at the N1 position
9-ethlyadenine deprotonated at the N6 position
N,N'-p-tolyformamidinate
N,N'-diphenylformamidinate
both DTolF‘ and DPhF‘
donor solvent

ligands

vii

Table 1.
Table 2.

Table 3.

Table 4
Table 5.

Table 6

Table 7.

Equation 1.

Equation 2.

LIST OF TABLES

page
pKa values for DNA nucleobases ....................................... 5
Summary of important crystallographic data for
[Rh2(DTolF)2(CH3CN)6][BF4]2 (3)38
Selected bond distances and bond angles for
[Rh2(DTolF)2(CH3CN)6] [BF4]2 (3) .................................... 4O

Torsion angles of [Rh2(DTolF)2(CH3CN)6][BF4]2 (3) ....... 41
Summary of crystallographic data for [Rh2(DTolF)2(9-

EtGH)2(CH3CN)2][BF4]2 (4)47

Summary of crystallographic data for Re2(OzCCH2CH3)2Br4

(10) .................................................................... 62

Selected bond distances and bond angles for

Re2(OzCCH2CH3)2Br4 (10) .................................................. 64
LIST OF EQUATIONS

General mathematical expression for the NMR chemical shift in
Simplified version of Ramsey’s equation for the chemical shift

of heavy metal nuclei ....................................................... 48

viii

Figure 1

Figure 2.
Figure 3.
Figure 4.

Figure 5.
Figure 6.

Figure 7.
Figure 8.

Figure 9.

Figure 10.

Figure l 1.
Figure 12.

Figure 13.

Figure 14.
Figure 15.

Figure 16.

Figure 17.

LIST OF FIGURES

Schematic representation of Cisplatin ................................... 2
Schematic representation ofDNA4
Pathway of hydrolysis of Cisplatin ..................................... 7

Schematic representation of inter- and intrastrand

cross-links .......................................................... 9
Pathway of C2’-endo to C3’-endo conformational change.lO
Shcematic representation of other antitumor active platinum
compounds ........................................................ 11
Schematic representation of dinuclear carboxylates ........ l4
Schematic representation of “head-to-head” and “head-to-tail”

isomers ............................................................ 21
Schematic representation of Rh2(u-DTolF)2(OzCCF3)2S2 ..22

Schematic representation of Rh2(u-DTolF)2(OzCCF3)2(ade)
.......................................................................... 23
Schematic representation of Rh2(u-DTolF)2(OzCCF3)2A. . ..24
Schematic representation of the “head-to-head” and “head-to-
tail” isomers of Rh2(ll-DT0lF)2(u-9-EtGH)2(OzCCF3)2 ...32
1H NMR spectrum of Rh2(IJ-DTolF)2(u-9-
EtGH)2(OzCCF3)2..32

Shcematic representation of deprotonated 9-ethylguanine. . ...34
ORTEP diagram of [Rh2(H'DTolF)2(CH3CN)6]2+ ............. 37
PLUTO diagram depicting the torsion angles of [Rh2(u—
DTolF)2(CH3CN)6]2+ ............................................... 39
1H NMR spectrum of [Rh2(u-DTolF)2(u-9-
EtGH)2(CH3CN)2]2+ ................................................. 44

ix

Figure 18.

Figure 19.

Figure 20.
Figure 21.

Figure 22.

Figure 23.
Figure 24.

Figure 25.

Figure 26.
Figure 27.

Figure 28.

Figure 29.
Figure 30.

Figure 31.

Figure 32.

1H NMR
EtGH)2(CH3CN)2]2+ depicting the

[Rh2(H'DT01F)2(H'9'

separation of the two

spectrum of

isomers ................................................................ 45
1H NMR [Rh2(ll-DTolF)2(u-9-
EtGH)2(CH3CN)2]2+ depicting the separation of the two

spectrum of

isomers ................................................................ 46
103Rh NMR chemical shift ranges ................................. 49
103’Rh NMR spectra of Rh2(u-DTolF)2(OzCCF3)282 and Rh2(u-
DTolF)2(u-9-EtGH)2(OzCCF3)2 .................................. 52
103Rh NMR spectrum of [Rh2(u-DT01F )2(CH3CN)6]2+ ........ 53
Schematic representation of dirhenium carboxylates ............ 56
PLUTO diagram of cis-Rez(u-02CCH2CH3)2Br4(DMF)2 ...63
Schematic representation of the possible isomers of cis-Re2(u-
9-EtG)2Br4 ........................................................... 65
1H NMR spectrum of cis-Re2(p.-9-EtG)2Br4 .................... 66
Pathway depicting the cis to trans isomerization of the
dirhenium bis-carboxylates ........................................ 67
Schematic representation of the extended structure of of cis-
Re2(u-9-EtG)2Br4 .................................................. 69
Possible pathway for the deprotonation of 9-EtGH ............. 75
1H NMR spectrum of Rez(u-9-EtG)2(u-OzCCH3)4C12 ....... 76
1H NMR spectrum of Re2(p-9-EtG)2(u-OzCCHzCH3)4(BF4)2

.......................................................................... 77
1H NMR spectrum of Re2(u-9-EtG)2(u-
02CC6H5)2(CH3CN)2(BF4)2 ............................................ 79

Chapter I

Introduction

1. Introduction

The development of cisplatin (Figure 1) for the chemotherapeutic
treatment of testicular and ovarian cancers, as well as tumors of the
bladder, head and neck has stimulated investigation of the antitumor
activity of a wide range of mononuclear platinum complexes.1 This
explosion of research is due, in part, to the extremely high toxicity
observed for cisplatin. Cancer patients who are administered cisplatin are
threatened by a range of toxic side effects including nephrotoxicity,
myeolosuppression, ototoxicity, nausea, peripheral neuropathies and
cardiac abnormalities. Obviously, less dangerous treatments are sought so
that such health risks can be minimized. An equally compelling reason for
further research into cisplatin and related compounds is the possibility of
identifying compounds that may target cancers not affected by cisplatin,
e.g. deadly tumors such as lung and colon cancers. Furthermore, recent
studies suggest that cisplatin itself exhibits carcinogenic effects, although
the extent of carcinogenicity in humans is unknown.4 The challenge now
faced is to tailor and design inorganic complexes with improved biological
activity and resistance to physiological decomposition, with concomitant

reduction of toxic side effects.

Figure 1

A. Mechanism of Action of Cisplatin

Before contemplating potential candidates as antitumor drugs it is
important to first understand the accepted mechanism of action of cisplatin, .
and to give a general description of DNA. In the nucleus of a cell, DNA is
coiled around histone proteins which help to organize DNA into a
chromatin-like structure},8 This chromatin-like structure consists of two
strands coiled into a double helix. A single strand contains the nucleotide
bases guanine, adenine, cytosine, and thyrnidine linked together by a sugar-
phosphate backbone (deoxyribose). The two strands run in opposite
directions with respect to each other, and are linked by Watson—Crick
hydrogen bonding of the base pairs: guanine with cytosine and adenine with
thymine (Figure 2).198 When transcription and replication occur, the two
strands must unravel, thereby separating the hydrogen-bonding of the base
pairs.

While the accepted mechanism of cisplatin is not fully understood,
evidence supports the conclusion that DNA interactions are responsible for
its activity. It is not known, however, whether platination of DNA is the
sole cellular event that leads to the death of tumor cells. Experiments that
measure the rates of DNA, RNA, and protein syntheses in cisplatin-treated
cells reveal that DNA synthesis is preferentially inhibited, while RNA and
protein syntheses are only slightly affected.5 Although cisplatin has been
observed to interact with proteins, cell wall components and the sugar-
phosphate backbone, the platination of B-DNA (the classical structure of
the double helix) at the purines is thought to inhibit cell growth by creating

irreparable lesions on the DNA strands of tumor cells.6’7

Figure 2

 

o H. ,H H\ H
/ N </ N N
N N ..H N N H N o o N H

Guanine Adenine Thymjne Cytosine

Table 1 7
him; Site dial—1m
Guanine N7 2.3
Guanine N1 9.3
Guanine N3 ?
Adenine N7 41.5
Adenine Nl 3.7
Adenine N6a ?
Cytosine N3 4.4
Cytosine N4 ?
Uracil N3 9.5
Thymine N3 9.5

a = N6 is the exocyclic nitrogen after deprotonation

The observed coordination modes for Pt(H) to DNA bases are pH
dependent (Table 1), with the N7 positions of guanine and adenine and N3
of cytosine and adenine being the preferred binding sites at neutral pH
(which is approximately physiological pH). The N7 position of guanine has
been determined by X-ray crystallographic and NMR studies to be the most
likely site of coordination for platinum compounds. The guanine N1 and

N3 positions become available for binding at a pKa of 9.3 with an excess of

platinum compound. The two most common sites for binding at adenine

are the N7 and N1 positions. The pyrimidines are found to preferentially
bind at the N3 position. Cytosine readily binds at N3, unlike thymine,
which must first deprotonate before metallation can occur at this position.
The main rationale for purine N7 positions being the frequently observed
coordination sites in double helical DNA is the availability of this position
which is exposed in the major groove, and their lack of involvement in
Watson-Crick hydrogen bonding, especially in B-DNA. It has been noted
in general, that guanine N7 is the preferred site of attack by platinum
compounds for a variety of reactions.1,2»6,7 The exocyclic amino- and
keto- groups as well as the phosphate groups are possible sites for
hydrogen-bonding to the ammonia ligands of bound cisplatin, a situation
that would further disrupt the structure of the DNA coil, and aid in
stabilization of the adducts.7

One of the open questions in this field remains to be settled, namely
what form of cisplatin actually binds to DNA? In any case, it is clear that
the initial binding mode would be highly dependent on cell pH.
Unfortunately, most of the present knowledge regarding the transport of
cisplatin in living tissue is based on the behavior of the compound in vitro,
which may be difficult to extrapolate to in viva situations. For example, it
is known that depending on the Cl' concentration and pH in the medium,
cisplatin will undergo hydrolysis.6 As depicted in Figure 3, cisplatin
hydrolyzes in aqueous media and basic pH conditions to form aqua and
hydroxo species. Due to the high Cl' concentration outside of the cell
(~100 mM) hydrolysis is prevented. The C1" concentration inside the cell,
however, is considerably lower (~4 mM), a situation that facilitates the

hydrolysis process.

Figure 3

+
H20 ¥ C110,... P ,..\\\\ NH3

1
H20’ ‘ NH,

Cl
-H+ .1
H+ Cl H20
(pKa=6.3) 1
2+

C101,... P ..,.\\\ N H3 H2010,“ P ..o‘“ N H3
t t
Ho’ ‘NH, Hzo’ ‘NH3

H20 +
l' ' + (pKa=5.6)

H2010,“ Pt»... NH3 +
H0’ ‘ NH3

 

C iS -PtC12(NH3)2 ‘

 

 

O

11* -H+ (pKa=5.6)

- ' 2+
H3N/Il,h. P ..“\\\ NH3

t
Hof \OH

"” s“‘
”/x \e‘

Pt
0H2“ ‘NHQ

b

H000," ,.\\\\ NH3
P .

t ——->
Ho’ ‘NH3

 

 

B. Intrastrand Cross-links

It is known that the binding of hydrolyzed cisplatin to DNA
promotes the unwinding and shortening of the double helix with the
formation of two main binding modes; these are a short range intrastrand
cross-link that is responsible for the unwinding effect and a long range
cross-link that leads to the shortening of the strand.2:397 Extensive, studies
have confirmed that platination at the deoxyguanosine N7 atom bends the
DNA in the direction of the major groove.3

After many years of research on the subject, researchers generally
agree that the most common DNA platinum species is a bifunctional DNA
adduct of general formula (NH3)2Pt(XpX) or (NH3)2Pt(XprX) where X
is either guanosine (G) or adenosine (A), and N is any of the other DNA
nucleosides.2 Migration of bound cisplatin from one nucleobase to another
has been found to be uncommon, and is thought to be an unlikely event
under ambient conditions.1 A schematic of the four most common modes
of binding exhibited by cisplatin is depicted in Figure 4. High resolution
NMR, as well as X-ray crystallographic and chromatographic studies
support the hypothesis that GpG is the preferred binding site as opposed to
the other adducts. From the X-ray crystallographic studies performed on
cis-[Pt(NH3)2(d(pGpG))], it was observed that a phosphate group can
participate in hydrogen-bonding with one of the ammonia groups of
cisplatin. Furthermore, the 06 position on the other purine is also in the
range of hydrogen-bonding with a cisplatin ammonia ligand. The
hydrogen-bonding disrupts the Watson-Crick base pairing leading to a
subtle perturbation of the double helix.1

Figure 4

 

 

NH3 NH3 NH3
\ / /
Pt I\ NH3
‘3'.
1,2 GpG 1,2 ApG 1,3 GpoG

Interstrand Crosslink Intrastrand Crosslink
G = guanine
A = adenine

X = cytosine (C), thymine (T), adenine (A)

It is important to note that the preferred direction of binding for the
1,2-intrastrand cross-linkage is also directed from the 5‘-base to the 3'-
neighbor. Similar binding modes were found for a S'-deoxyguanosine
directing the metal to a 3'-dX (X = C,T,A where deoxyadenosine was
preferred). Studies predict that the shortening and unwinding of the
double helical DNA occurs after the platinum chelation of a G-G sequence.
This chelation causes a tilting of the G-G bases out of their parallel
alignment. The tilting instigates a kink in the helix of about 400-700.
Additional distortion occurs at the 5'—deoxyribose ring which changes from
a C2'-endo to a C3'-endo ring puckering (Figure 5) to accommodate the
strain caused by the cisplatin 1,2-intrastrand binding to a -GG-

sequence.3.7,13,14

10

Figure 5
H3N
\ DNA Base
"1
A H N
o. O I) O ”D 3

N O

15. H lift—Z4 ,0 Elf/8:148
Q Ia 5 _ & ‘ 5 lN-H

HO 5C H , N9 /43 jN H cis-Pt(NH3)22+ 1P 0‘ C13 H N9 /4 3 ‘27

2 NJN ‘ F HO 5' ~ N N
4 3. 1' H’ ‘H 4, ’ 2, 1' H' ‘H

C2'-endo C3'-endo

 

Long range 1.3-intrastrand cross-links have been attributed to a
more prominent shortening of DNA.1’3’7 Cisplatin adducts with
d(GpCpG), d(GpApG) and d(GpGpG) all have been characterized for the
1,3-intrastrand type chelation. Both 1,3- and 1,2-intrastrand cross-links
have been found for the adduct with d(GpGpG) where the 1,3-intrastrand
binding is favored in vivo, although the 1,2-intrastrand cross-link is
thought to be the more cytotoxic of the two. One of the explanations
advanced for this is that the 1,3-intrastrand cross-links create a greater
distortion in the double helix than the 1,2-type chelation, and large
perturbations to DNA are more easily recognized by DNA repair enzymes
than those that produce more subtle damage.2,15
C. Interstrand Cross-linking

Interstrand cross-links have been found to account for less than 1%
of the total amount of platinum bound to double-stranded DNA. The most
prominent cross-link was found to be cisplatin bound to two

deoxyguanosines at the guanine N7 position. It is important to note that the

11

major sequence in which the cross-link occurs is from the 5'—->3' end.”-
Although this mode of binding is not considered to be the primary DNA
adduct of cisplatin responsible for tumor cell death, it certainly cannot be
ruled out as an important binding mode for antitumor compounds.

In summary, studies have shown that intra-strand rather than inter-
strand platinum chelation predominates. Clearly, cisplatin preferentially
binds to guanine bases of DNA to form bifunctional d(GG) and d(GXG)
adducts which are thought to be easily recognized by repair enzymes.
Platination of deoxyguanosine may be preferred due'to the availability of
the exocyclic O6 to form hydrogen-bonds to cisplatin. This line of
reasoning also helps to explain the lower degree of activity observed in the
trans analogue which bends the double helix in two directions by
preferentially forming 1,3-d(GXG) adducts.16

Figure 6

 

Carboplatin CH3 OH
Iproplatin
H20 1,, \\ NH
ogso/P NH

Spiroplatin

12

2. Proposed Design Strategies
Since the presence of DNA-bound platinum that disrupts the double
helix conformation is considered cytotoxic to tumor cells, it is reasonable
to expect that other metal compounds could produce the same disruption of
the normal processes of transcription and DNA replication. Indeed
antitumor activity has been recognized for a variety of platinum
compounds including carboplatin, iproplatin, and Spiroplatin (Figure 6)
which have been extensively studied in clinical settings. In fact carboplatin
has been recently approved for clinical use.1 Carboplatin and Spiroplatin
are structurally similar to cisplatin in that they are square planar complexes
with two cis-nitrogen donors. Iproplatin, on the other hand, is an
octahedral Pt(IV) complex which is thought to undergo in viva reduction
to a square planar hydrolyzed Pt(II) complex.1 From the extensive
mechanistic studies aimed at elucidating the effect of cisplatin DNA
replication, and the collective studies of the mechanisms of similar
platinum complexes, key trends have been discerned regarding the activity
of inorganic antitumor compounds:
(1) the complexes should be designed such that two cis, rather
than trans, ligands substitute in reactions with biological
molecules.
(2) the complexes should be neutral compounds although they
may become charged after ligand exchange in the body.
(3) the geometry of the complexes should be square planar or
octahedral.
(4) the leaving groups should be ~3.4 A apart on the molecule
(theWatson-Crick ladder spacing).

 

l3

(5) the rates of exchange should fall into a narrow range of
not occuring too rapidly, as this will result in
decomposition of the compound before reaching the
tumor cells, and not too slowly, as this will result in little
or no activity.

(6) the remaining groups across from the leaving groups
should be robust, and preferably be capable of

participating in hydrogen-bonding such as amines.17

3. Dinuclear Metal Complexes

The search for transition metal complexes adhering to the criteria set
forth by the collective efforts of researchers in the field has included a
broad range of compounds. Following the discovery of the cytotoxic
behavior of cisplatin, complexes of the other platinum group metals were
among the first class of compounds tested for antitumor activity. In
addition, cyclopentadienide complexes of Mo, Ti and V were also subjected
to antitumor investigations. Other late transition metal compounds that
have been studied include cis-[Ru(NH3)4C12]+ and the trans isomer of
RuC12(DMSO)4.4 In all of these studies, new compounds have been tested
against the same tumor cell lines affected by cisplatin, which is unfortunate
since this will lead only to new drugs for which treatments already exist.1
For a completely different class of possible antitumor complexes, cancer
cell lines other than those treatable by cisplatin must also be studied.

One entirely different class of antitumor active compounds whose
structures and reactivities do not closely adhere to the cisplatin guidelines
for anticancer compounds are dinuclear transition metal complexes of

rhodium, ruthenium and rhenium. Compounds of these metals with

 

l4

bridging carboxylate ligands have been studied as to their antitumor
activity against various tumor cell lines with very promising results
ensuing. To date, however, no major advances in the development of these
compounds for pharmaceutical use have been made, most likely due to
their specialized nature and their facile decomposition under physiological
conditions. The so-called "lantem" structures of these compounds
(depicted in Figure 7) allow for two possible types of binding sites, namely

equatorial (eq) and axial (ax).

 

 

 

 

 

Figure 7
R R
0A2. 0*. R
/
A / ‘(
”so 3‘0 eq 3‘0 30
L—— M — L L—M M L
L L /T W 0/ ll
ax O
i- . >'/

 

M = Re, Rh, Ru
R = CF3, alkyl
L = C1, BI', CH3CN, CH3OH

A. Dinuclear Rhodium Complexes
Dirhodium carboxylate compounds of the type Rh2(OzCR)4L2 (R =
CH3, CH2CH3, CH2CH2CH3, CH20CH3; L = donor solvent) exhibit

considerable antitumor activity in mice bearing the Ehrlich ascites tumor.

15

These complexes were found to inhibit DNA but not RNA synthesis with
dirhodium(II) tetrabutyrate being the most potent inhibitor. Studies
indicate that these complexes bind to denatured DNA, poly-A, poly-G.
ribonuclease AA, and bovine serum albumin. The same studies also
concluded that the dirhodium tetracarboxylates do not bind to highly
polymerized native calf thymus DNA, poly-G and poly-C. It was
postulated from these preliminary observations that the observed antitumor
activity was due to reversible binding by a nitrogen donor in the
biomolecules through the axial position of the dimetal unit.13 Additional in
vitro binding to purified DNA polymerase I and RNA polymerase I was
also observed, but in vivo studies revealed only the interruption of DNA
replication. It is questionable as to whether this inhibition was due to
interactions with polymerase or to interactions with the DNA baseslgaib

Further studies of dirhodium tetracarboxylates binding to enzymes
containing sulfliydryl (-SH) groups at or near the active site revealed that
the binding occurs irreversibly to these molecule types. The rate of in vivo
enzyme inactivation was in the order butyrate > propionate > acetate >
methoxyacetate, which is correlated to the lipophilicity and solubility of the
complexes which increases in the same order. The R group identity of the
carboxylate ligands also has a direct affect on the Lewis acidity of the metal
dimer.18,20

A major question regarding the antitumor activity of dirhodium
tetracarboxylates that occurred to us early on is whether the reactions
indeed occur with only axial binding or equatorial binding to DNA
nucleobases. It was postulated by other researchers that irreversible
enzyme binding occurs by equatorial substitution of the carboxylate ligands

which may be extrapolated to DNA studies, since this binding was found to

16

be irreversible in vivo.180,d This theory was brought into serious question
in our laboratories several years ago when the feasibility of equatorial
binding of DNA purines to the dirhodium unit was verified by X-ray
crystallography for guanine (chapter 2), which adopts a previously
unknown bridging mode across the Rh-Rh bond in the neutral compound
Rh2(9-EtGH)2(02CCH3)2(CH3OH)2. Two guanine bases span the
dirhodium unit in a cis disposition and in a "head-to-tail" orientation of the
N7-O6 atoms. 1H NMR spectroscopic evidence supports a second isomer
that is assumed to be the "head-to-head" arrangement which is the more
biologically relevant adduct; attempts to characterize this adduct by X-ray
crystallography are currently underway.19 The neutral complex,
Rh2(02CCF3)4L2 (where L = CH3CN, CH3OH) yields products with 9-
EtGH that exhibit the same bridging mode for the purines as the acetate
derivative. In this case, the "head-to-head" and "head-to-tail" isomers of
the purine are observed to be formed in a 1:1 ratio as judged by 1H N MR
spectroscopy in the H8 region. A similar situation is encountered with the
partially solvated species [Rh2(OzCCH3)2(CH3CN)¢5]2+ in reactions with 9-
EtGH, from which the biologically relevant "head-to-head" isomer of the
compound [Rh2(02CCH3)2(9-EtGH)2]2+ was isolated and characterized by
X-ray crystallography. Similar bridging modes were also verified by
crystallography for diruthenium tetracarboxylate compounds in reactions
with 9-ethylguanine.19
B. Dinuclear Rhenium Complexes

The literature regarding the biological activity of rhenium is sparSe,
with no known case of rhenium poisoning having been reported. As far as
biologists are aware, rhenium is not a trace constituent in plants or animals.

We noted with interest that several dinuclear complexes of rhenium have

17

been subjected to antitumor testing; these are the compounds
Rez(OzCR)4SO4 (R=CH3, CH2CH3, CH2CH2CH3) which were tested
against sarcoma,~melanoma, and leukemia cell lines in mice. The
tetrapropionate complex was observed to be far more effective than the
tetraacetate and the more toxic tetrabutyrate against all three cell lines. It
was found that high consecutive doses of [Re2(OzCCH2CH3)4]2+ were
necessary for inhibition of the melanoma and sarcoma tumors in mice.
Leukopenia (decrease of white blood cell production) was observed in the
leukemia bearing mice, but the same high doses needed for the inhibition of
the tumors significantly pronounced the myelosuppressive effects.17 Since
it is known that the Re-Re quadruple bond of the cationic tetracarboxylate
compounds is susceptible to decomposition to rhenium oxides in the

”’26 it has been postulated that this could account for the

presence of water,
necessity of high doses in order to observe antitumor activity. In other
words, the compound is effective only if the rate of cytotoxicity is greater
than the rate of hydrolysis.

More recently, a class of more stable carboxylate compounds of
rhenium, the cis biscarboxylates, was observed to exhibit greater antitumor
activity and even lower toxicities than the cationic tetracarboxylates.
Although the complex cis-Rez(02CCH2CH3)2X4(H20)2 (X = halogen)
undergoes hydrolysis much slower than its tetracarboxylate analogue, the
dose levels were still considered to be too high for further testing.
Nevertheless, these results support the inhibition of DNA synthesis with no
interactions occurring with proteins or RNA, which is a promising starting
point for future generations of rhenium-based drugs.”b

Clearly, the collective studies of dinuclear transition metal complexes

as antitumor agents is a topic worthy of scientific investigation. Attempts

18

to develop transition metal complexes with greater potency and reduced
toxicity has led researchers into the studies of dinuclear metal complexes.
Consideration of the currently accepted mechanism of action of cisplatin as
well as structure-activity requirements are used as the basis for the search
for new and improved transition metal complexes in the fight against

cancer and other illnesses.

Chapter II

Dirhodium Anticancer Agents

19

2O

1. Introduction

The discovery of the inhibition of DNA replication by dirhodium(H)
u-tetra-carboxylates led us to investigate their interactions with 9-
ethylguanine (9-EtGH) in hopes of developing model compounds that
would lend insight into the process.19 Previous research efforts did not
elucidate the exact binding site of the dirhodium compounds in tumor cell
DNA, but studies indicated that covalent interactions with single-stranded
or unwound DNA was a likely mode of action?1 Work in our laboratories
established the remarkable finding that the u-tetra-carboxylates of
dirhodium(H) form products with two bridging 9-EtGH ligands cis to each
other, contrary to previous investigations which suggested that no
significant binding occurred between this purine and the dinuclear metal
center.21 It was further discovered that the complexes formed by 9-EtGH,
[Rh2(u-9-EIGH)2(u-02CCH3)2((CH3)2C0)212+, and Rh2(u-9-EtG)2(u-
OzCCH3)2(CH3OH)2 exhibit both "head-to-head" and "head-to-tail"
isomers (Figure 8).19 Since the bases in any given strand of DNA are
stacked in a "head-to-head" fashion, the "head-to-head" isomers are the
primary targets as model compounds in the context .of the mechanism of
metal complex attack on DNA. Inter-strand "head-to-tail" products are
also possible with these dinuclear metal systems, but this type of bridging
in the double helix would most likely create large distortions that are
recognized as damaged by DNA repair enzymes and are therefore

repaired.

21

Figure 8

Geometrical Isomers

 

 

A A

N o T 0

M'-—M " Mi—M

’ l ’ — ’ l ’ l
N 0 O N
v V

Head to head Head to tail

Formation of a product containing a bridging 9-EtGH ligand relies
on the lability of the equatorial ligands in the lantern structure. One such
lantern structure of dinuclear rhodium(II) that strongly retains two of its

ligands in a cis configuration as well as possessing two labile equatorial

ligands is the complex first synthesized by Piraino and co-workers, Rh2(u-
DTolF)2(u-02CCF3)2(H20)2 (DTolF = N,N'-p-tolylformamidinate)
(Figure 9). As compared to u-tetrakisformamidinate derivatives, e.g.
Rh2(ll-DTolF)4, which do not allow access to the equatorial positions, the
mixed-ligand derivative exhibits both axial (two) and equatorial (four)
coordination sites for binding to Lewis bases. The antitumor activity of
this complex was tested against various tumor cell lines and it was found to
exhibit equal antitumor activity and a lower toxicity than dinuclear
rhodium(II) complexes of the type Rh2(u-02CR)4L2. The researchers
postulate that the lability of the bridging trifluoroacetates and the axial

water molecules in solution lead to the formation of the charged solvated

22

cation, [Rh2(i.t-DT01F)2S(5]2+ (S = solvent) which is suspected to be the

important physiological form of the active compound.22

 

H3C H CH3
/\ \Q I
N N ' N/rg\

0A0 = '(02CCF3)

S = donor solvent

Although their preliminary work did not include verification of
structures by X-ray crystallography, Piraino concluded that adenine and
N6,N6-dimethyladenine did not effect the substitution of the equatorial
(CF3COz)' groups, but instead were axially coordinated through the N3

positions. It was further suggested that the N9 position participates in

23

hydrogen bonding with an oxygen atom of the equatorial trifluoroacetate

groups (Figure 10). It is important to point out that the coordination mode

Figure 10

 

Proposed structure of ha(u—DTolF)2(u-OZCCF3)2(ade)n,n-1
n = 2

of the more biologically relevant adenosine is required to be different than
that of adenine, since the N3 position in adenosine is sterically hindered
from coordination by the ribose ring. Conductivity measurements as well

as 1H-NMR and EPR spectroscopies support the formulation of the
adenosine product as the mixed-valence complex, [Rh2(p.-DTolF)2(u-

ado)2(02CCF3)](02CCF3) (where ado = adenosine), with bridging

adenosine moieties. The presence of the strongly basic forrnamidinate
groups lowers the oxidation potential of the Rh24+ fragment, thereby
allowing for the facile formation of mixed valence Rh25+ complexes. It
was proposed that the purine binds to the ha core through the N1 and
deprotonated N6 positions rather than via the usual N6-N7 positions
(Figure 11).

24

Figure 11

A
N N
l /’\
..‘\“N ..“‘\\N ’CF3
’%—Rh— O—q
HN l 0

EV
N N | NH
ribose/ Ni] 6]}
N
N J

1‘1
ribose

 

Proposed structure of
Rh2(u'DT01F)2(u'OZCCF3)2(A)n,n- l
A = adenosine; n = 2

Similar results with cytosine, l-methylcytosine and deoxycytidine
derivatives were obtained, but no X—ray structures were determined.
Unlike the u-tetracarboxylate derivatives that are reported to be unreactive
towards cytosine, the mixed-ligand complex is described as containing the
pyrimidine bound through the N3 and deprotonated exocyclic N4 positions.
In addition, Piraino reports that Rh2(u-DTolF)2(u-02CCF3)2(H20)2 does
not react with guanine, deoxyguanosine and uracil”?l This chapter reports
complementary work carried out in our laboratories regarding the
reactivity of this dinuclear complex with guanine and its substituted .

derivatives. Along with these results are reported the synthesis, structural

25

characterization and similar reactions of the novel, partially solvated
complex [Rh2(u-DTolF)2(CH3CN)6](BF4)2.
2. Experimental
A. Synthesis
‘ Unless otherwise stated, all reactions were carried out under an

argon atmosphere by the use of standard Schlenk-line techniques. The
mixed ligand complexes, [Rh(cod)(form)]2 and Rh2(u-form)2(u-

02CCF3)2(H20)2 (form = N,N'-p-tolylformamidinate, DTolF‘; N,N'-
diphenylformamidinate, DPhF'), were prepared as described in the
literature.22d The purine derivatives 9-EtGH and guanosine were

purchased from Sigma and were used without further purification.
Preparation of Rh2(u-DTolF)2(u-9-EtGH)2(OzCCF3)2 (1)

An amount of Rh2(u-DTolF)2(u-02CCF3)2(H20)2 ( 26 mg, 3.61 x
10:5 mol) was dissolved in 4 mL of CH3CN and 9-EtGH (12.7 mg, 7.08 x
10'5 mol) was added to the red solution, after which time the mixture was
refluxed for approximately 1 h. The olive-green reaction mixture was
allowed to cool to room temperature, filtered and treated with EtzO to
effect precipitation of the product. The green solid was washed with 20
mL of EtzO and dried under reduced pressure (32 mg, 83% yield). 1H
NMR (CD3CN) 5 ppm: 1.35-1.37 (m, CH3), 1.95 (s, CH3CN), 2.15 broad
(8, H20), 2.23 (s, CH3), 4.05 (m, CH2), 6.80 (d, phenyl), 6.89 (d, phenyl),
7.03-7.19 broad (m, phenyl), 7.55 (m, NCHN), 7.99 (3, H8), 8.02 (3, H8).
19F NMR (both CD3CN and acetone-d6) 8 ppm: '12.49 referenced to
C6H5CF3 (s, OZCCF3). IR (Nujol, NaCl) cm'1 1647 mb, 1564 s, 1487 s,
1360 m, 1261 s, 1199 s, 1134 br, 1084 br, 1028 br.

26

Preparation of Rh2(u-DPhF)2(u-9-EtGH)2(02CCF3)2 (2)

Rh2(u-DPhF)2(u-02CCF3)2(H2O)2 (16.8 mg, 2.34 x 10'5 mol) was
dissolved in 5 mL of distilled CH3CN and 9-EtGH (9.0 mg, 5.0 x 10‘5 mol)
was added to the red solution. After ~1 h of constant reflux the solvent
was evaporated under reduced pressure to yield a dark green residue which
was dried and redissolved in acetone (~3 mL). The green acetone solution
was treated with ~5 mL of hexanes and cooled to ~10 0C. The bright
green precipitate was removed by filtration and dried in vacuo (19 mg,
79% yield). 1H NMR (CD3CN) 5 ppm: 1.35-1.38 (m, CH3), 1.95 (s,
CH3CN), 2.21 broad (3, H20), 4.04-4.07 broad (m, CH2), 6.82 (d, phenyl),
6.89 (d, phenyl), 7.04-7.20 broad (m, phenyl), 7.55 (m, NCHN), 7.93 (8.
H8), 8.03 (5, H8). 19F NMR (both CD3CN, acetone-d6) 8 ppm: 1247
referenced to C6H5CF3 (s, OZCCF3). 103Rh NMR (CD3CN) 5 ppm:
5551.98.
Preparation of [Rh2(u-DT01F)2(CH3CN)6](BF4)2 (3)

A quantity of [Rh(cod)(DTolF)]2 (97 mg, 1.50 x 10'4 mol) and
AgBF4 (120 mg, 6.16 x 10'4 mol) was dissolved in 20 mL of a 1:1 mixture
of CH2C12 and CH3CN in the absence of light. After 2-3 h the yellow-

orange mixture had changed to a clear, pale green solution. The mixture
was stirred for an additional 2 days during which time silver metal was
observed to precipitate. The mixture was filtered through a Celite plug,
and the orange-red filtrate was concentrated under reduced pressure.
Following the addition of 3 mL of EtzO, the mixture was cooled to ~10
0C. An orange-red microcrystalline precipitate was collected by filtration,
washed with 3 x 5 mL EtzO and dried in vacuo (138 mg, 93% yield). 1H
NMR (CD3CN) 5 ppm: 1.96 (s, CH3CN), 2.25 (s, CH3), 2.49 (s, CH3CN),
6.99 (m, phenyl), 7.51 (t, NCHN). 103Rh NMR (CD3CN) 8 ppm: 4648.73.

27

Preparation of [Rh2(u-DTolF)2(u-9-EtGH)2(CH3CN)2](BF4)2 (4)

(a) A quantity of 9-ethylguanine (41.7 mg, 0.233 mmol) was added
to a stirring solution of [Rh2(u-DT01F)2(CH3CN)6](BF4)2 (115.4 mg, i
0.116 mmol) dissolved in of 5 mL of CH3OH and 10 mL of CH3CN. The
mixture was gently refluxed for ~2 h, during which time the orange-red
mixture 'with the suspended 9-ethylguanine became a clear green solution.
The reaction mixture was filtered, and the filtrate was evaporated in vacuo
to yield a green solid (60 mg, 47% yield). 1H N MR (CD3OD) 8 ppm: 1.36
(t, CH3), 1.40 (t, CH3), 1.92 (s, CH3CN), 2.23 (s, CH3), 3.98 (q, CH2),
4.05 (q, CH2), 6.80 (m, phenyl), 6.89 (m, phenyl), 7.00 (m, phenyl), 7.50
(t, NCHN), 7.55 (t, NCHN), 8.32 (s, H8), 8.35 (s, H8). IR (Nujol, NaCl)
cm'1 2311 s, 2305 s, 1641 m,br, 1608 m, 1577 s, 1508 s, 1027 s, 1178 s,
1057 br, 821 s, 790 s, 763 s, 721 s.

(b) The above reaction was repeated with an acetonitrile and acetone
solvent mixture. Two equivalents of 9-EtGH (46 mg, 2.57 x 10'4 mol)
were added to a stirring acetonitrile (5 mL) and acetone mixture (1 mL) of
(4) (138.5 mg, 1.29 x 10'4 mol), and the mixture was stirred at constant
reflux for ~3 h. The reaction mixture was filtered in air through a Celite
plug and the solvent was removed in vacuo (110 mg, 78% yield).
Synthesis of [Rh2(u-DTolF)2(u-9-EtGH)2(CH3CN)2](BPh4)2 (5)

The metathesis of (4a) with two equivalents of AgBPh4 in acetone
yields (5) in nearly quantitative yields. IR (Nujol, NaCl) cm:1 1585 b,
1504 s, 1261 s, 1084 m,br, 1028 s, 804 m.

B. X-ray Crystallography

The structure of [Rh2(u-DTolF)2(CH3CN)6](BF4)2 (3) was

determined by the application of general procedures that have been fully

described elsewhere.24 Geometric and intensity data were collected on a

28

Rigaku AFC6S diffractometer with graphite-monochromated MoKa (2.0. =

0.71069 A) radiation and were corrected for Lorentz and polarization
effects. A unit cell for [Rh2(u-DTolF)2(u-9-EtGH)2(CH3CN)2](BF4)2 (4)

was determined on a Nicolet P3/V diffractometer upgraded to a Siemens
P3/F with graphite monochromated MoKa (la = 0.71073 A) radiation; the
reflections were corrected for Lorentz and polarization effects. All
calculations were performed with VAX computers on a cluster network
within the Department of Chemistry at Michigan State University using the
Texsan software package of the Molecular Structure Corporation. 23
[Rh2(u-DT01F)2(CH3CN)6I(BF4)2 (3)

(i) Data Collection and Reduction. Large single crystals of (2) were
obtained from the slow diffusion of a concentrated acetonitrile solution into
toluene. The crystals grew as long orange-red crystals at the interface of
the two solvents after 12 days. A suitable single crystal, with the
approximate dimensions of 0.78 x 0.26 x 0.21 mm3, was mounted on the
tip of a glass fiber with Dow Corning grease. Cell constants were obtained
from a least squares refinement using 24 carefully centered reflections in
the range 290 < 20 < 370. Data were collected at '100 i 1 0C in the range
4 g 20 g 470, by using the (1)-scan method. Weak reflections (those with F
< 100(F)) were rescanned at a maximum of 3 rescans and the counts were
accumulated to ensure good counting statistics. A total of 8005 reflections
were collected; of these 3865 reflections with F02 > 30(F0)2 were used in
the measurement. Periodic measurement of three representative reflections
at regular intervals revealed that no loss of diffraction intensity had
occurred during data collection. An empirical absorption correction was

applied on the basis of azimuthal scans of 3 reflections with x near 90°.

29

(ii) Structure Solution and Refinement. The space group was
determined to be Pbca based on the observed systematic absences. The
positions of all non-hydrogen atoms were obtained by application of the
direct methods programs MITHRIL and DIRDIF followed by successive
full-matrix least-squares cycles.24 All non-hydrogen atoms were refined
with anisotropic thermal parameters. Hydrogen atoms were treated as
fixed contributors at idealized positions and were not refined. Final least
squares refinement of 567 parameters resulted in residuals of R = 0.046

and Rw = 0.063 and a goodness-of—fit = 2.01. A final difference Fourier

map revealed the highest peak in the difference map to be 0.87 e'/A3.
[Rh2(u-DT01F )2(u-9-EtGH)2(CH3CN)2](BF4)2 (4)

(i) Cell Determination. Single crystals of (4) were obtained from
slow diffusion of a concentrated CH3OH solution of the compound into
toluene. The crystals grew as prismatic rectangles over a period of 3-4
weeks. A single crystal was mounted on the tip of a glass fiber with Dow
Corning grease. Geometric and intensity data were collected at '100 i 2
0C. Indexing and refinement of 15 reflections in the range 4 g 20 g 120
selected from a rotational photograph gave preliminary unit cell
parameters for an orthorhombic C-centered crystal system.
C. 103Rh NMR Spectroscopy

The 103Rh NMR spectra were recorded on a Varian 500-MHz
spectrometer equipped with a 5 mm Nalorac 103Rh probe with an
acquisition time of 12-16 h. The pulse width was 10 us at a temperature of
~35 0C. 103Rh NMR shifts are referenced to [E](103Rh) = 3.16 MHz
(Equation 1).25 The term 5 is the chemical shift, 8’ is chemical shift with
respect to the published reference (TMS), E. is the frequency of the desired

reference (103Rh), E’ is the frequency of the published reference.

30

Equation 1

5 = 5’(:'1'-:,ref/~"'3ref) + [(E,ref ' Eref) / Eref] X 106

3. Results

A. Spectroscopic Properties of (1) and (2)
The solid-state structures of Rh2(u-form)2(u-02CCF3)2(H20)2

complexes lend insight into the possible structures of (1) and (2). Rh2(u-
form)2(u-02CCF3)2(H20)2 exhibits two cis-DTolF and two cis-02CCF 3
groups spanning the metal centers. Equatorial substitution of the
(CF3C02)' ligands is expected due to the weak binding interactions of these
ligands in comparison to the basic forrnamidinate ligands. Drawings of the
proposed structures for (l) and (2) (including different isomers) are
depicted in Figure 12.

The characterization of dinuclear metal centers ligated by purines by
1H NMR spectroscopy is useful because of the characteristic downfield shift
of the H8 signal by 0.5 to 1 ppm (8 = 7.63 ppm for free 9-EtGH in
CD3OD).4 The 1H NMR spectra of (l) and (2) exhibit two signals
assignable to the H8 positions of bridging 9-ethylguanines with H8 shifts
observed at a 5 > 7.9 ppm (Figure 13). The presence of signals
attributable to 9-EtGH in CD3CN is an important observation since the
purine is insoluble in this solvent.

Another important point that should be addressed is the color change
that occurs during the reaction. The parent dirhodium(H) complexes,
Rh2(u-form)2(u-02CCF3)2(H20)2, are green solids that yield red solutions

upon addition of CH3CN which displaces the axial water molecules. Green

colors for dirhodium(H) complexes are usually associated with axial

31

oxygen donors, while red colors are associated with axial nitrogen donors.
The color change back to green that occurs with coordination of 9-EtGH is
attributed to the presence of an oxygen atom from CF3COz' ligands bound
to the axial positions.

The two H8 signals observed in the spectra for (1) and (2) indicate
the presence of two isomers in a roughly 1:1 ratio, based on integration.
The two isomers are proposed as being the polar "head-to-head" and less
polar "head-to-tail" isomers (Figure 8). One would expect the less polar
isomer to be of lower energy than the "head-to-head" isomer by analogy to
complexes of the type M2(x-hp)4 (x-hp = 6-substituted hydroxypyridine)
where the non-polar isomers are preferred over the polar ones.26 The
presence of both isomers in the 1H NMR spectrum indicates a small energy
difference in the formation of the two compounds.

The previously characterized products of reactions between
Rh2(OzCR)4L2 and 9-EtGH have demonstrated that two types of bridging
guanine ligands are possible: (i) 9-EtGH, the neutral ligand form and (ii) 9-

EtG' the anionic or deprotonated form, where the purine is deprotonated
at the N1 position (Figure 14). It has been documented that the pKa of the

N1 position is lowered by l.5-2.0 pKa units upon coordination to platinum
complexes (from ~94 to 8.0).7 The anionic ligand, 9-EtG' has also been
observed to form in situ in acidic media.7 The deprotonation of Nl-H in

guanine is observed in reactions with metal complexes containing basic

leaving groups such as CH3C02' (pr = 9.25). The much lower basicity
of the trifluoroacetate leaving groups of (1) and (2) (pr > 13) are
expected to allow the coordination of the neutral 9-EtGH species to the

dinuclear metal center.

 

 

32

Figure 12

 

F3C~c—o—Rhi—Rn—o—c:
rgN O l o
l l
N} N O N
,N—l :7
H3CH2C HN

head-to-head isomer

A
N N

(i i NhN ,0?3

1:30 C— o— Rh"— Rh=i o—c‘

FN/Ol/ o
H3CH2C’ NW 0
NVNH N‘}
HN N\
\§ CHZCH3

N
head-to—tail isomer

33

39:6 5 5 225 2805:: o». §~€bfio5~€bm5ENAONOQJX
A: 5 OUwOZ. .

mbw

/ s8

 

 

 

 

 

 

m m w m m A. w m n 0 EH:

34

Figure 14
0 o
N 3 N
1 . \> #11 \>
HzN N I.“ N I?
R R
neutral guanine N 1 deprotonated guanine

The green color of compounds (1) and (2) supports the presence of
axial CF3C02' groups in the solid state.26 The 19F NMR spectra of
complexes (1) and (2) ('12.49 and '12.47 ppm, respectively) are
indicative of dissociated CF3C02' groups in solution.27 19F NMR signals
for bridging equatorial CF3C02' groups are commonly observed 5060
ppm upfield (8 ~ '77 ppm) from the free acid referenced to C6H5CF3.
Axially coordinated CF3COz‘ groups have been observed approximately in
the same region.27 The singlets at '12.49 ppm for (1) and “12.47 ppm for
(2) are indicative of outer-sphere CF3C02'. Displacement of these groups
by acetonitrile is conceivable, but solvent displacement by a weak donating
solvent like acetone (the solvent used for the 19F NMR experiment) is
unlikely. The observed signal indicates the relatively weak binding
strength of the trifluoroacetate ligand to the [Rh2(form)2]2+ species in
solution. Similar 19F NMR spectra are observed for the starting
compounds Rh2(u-form)2(u-02CCF3)2(H20)2 which support a rapid
exchange of the labile equatorial ligands at room temperature. The
extreme lability of the CF3C02' ligands can be explained by the strong 0-
donation of the formamidinate ligands which weaken the equatorial Rh-O

interactions. The inductive and resonance effects from the tolyl groups

35

that are trans to the CF3C02‘ groups donate electron density to the N
atoms, which in turn direct the added electron density toward the dinuclear
center, thereby weakening the CF3COz' binding interactions.

B. Spectroscopic and Crystallographic Properties of 4(3)

The novel, partially solvated complex, (3) was prepared after our

recent discovery of the related complex, [Ir2(u -
DTolF)2(CH3CN)6](BF4)2.28 Early studies of Rh2(u-DTolF)2(u-
OzCCF3)2(HzO)2 suggested the formation of the. solvated species in
solution with trifluoroacetate being liberated as an anion.22b We hoped
that slight variations such as replacing the counterions would increase the
chances of obtaining single crystals suitable for X-ray diffraction studies.
The labile acetonitrile ligands in (3) render this complex ideal for the
synthesis of other interesting [Rh2(u-form)2]2+ derivatives that are not
available by other routes. Crystal parameters and basic information
pertaining to data collection are summarized in Table 2. The ORTEP
representation of (3) is depicted in Figure 15, and bond distances and
angles are presented in Table 3. From this information, insight into the
structures of [Rh2(u-form)2(u-9-EtGH)2]2+ moieties may be gained.
Compound (3) crystallizes in an orthorhombic-C crystal system.
The equatorial bridging DTolF' groups are cis to each other with average
torsional twist angles of 180 (Table 4). This degree of non-ideality for the
Rh24+ moiety is not uncommon and is observed with most of the metal-
metal single bonded structures of this type.26 The angles formed by the N -
C-N fragments contained in the five-membered ring (123 (1)0 and 124
(1)0) are within normal ranges observed for similar compounds. The C-N
distances of the forrnamidinate fragments (~1.33 (1) A) are sufficiently

short so as to suggest significant double bond character for the N-C-N

36

groups. The Rh—N (~2.04 (9) A) bond distances to the formamidinate
groups are longer than the corresponding distances in the bi s-
trifluoroacetate derivative (1.99 (4) A). The average equatorial
acetonitrile M-N distance is approximately 0.04 A shorter than the metal-
forrnamidinate distances, supporting a strong o-donation to the dinuclear
center. The axial acetonitrile interactions, Rhl-N8 (2.188 (8) A) and Rh2-
N4 (2.211 (8) A), are ~0.1 A shorter than the axial acetonitrile interaction
observed in the crystal structure of the tetra-formamidinate derivative,
Rh2(u-DPhF)4(NCCH3) (2.106(4) A). This implies the presence of fairly
weak axial interactions.5

The Rh-Rh bond distance of (3), 2.560(1) A, is slightly longer than
that found in cis-ha(H'DTolF)2(tt-02CCF3)2(H20)2, which exhibits a Rh-
Rh bond distance of 2.43(1) A22 The nitrile ligands appear to be stronger
electron donors than the trifluoroacetate ligands based on 103Rh NMR
chemical shifts. It is difficult to ascertain whether the addition of a
bridging unit such as 9-EtGH will lengthen or shorten the Rh-Rh bond
compared to the trifluoroacetate derviative, but it is safe to assume that the
electron donation from the neutral purine would allow for similar metal-
metal bonding.

A 1H NMR spectrum of (3) is in accord with the structural data
from the crystallographic results. The singlet at 5 = 1.96 ppm is assignable
to the axial acetonitrile ligands whereas the singlet at 5 = 2.49 ppm can be
attributed to the equatorial acetonitrile groups. The singlet at 8 = 2.25
ppm is due to the four equivalent methyl groups of the tolyl substituent,
and a multiplet centered at 5 ~69 ppm is assigned to the phenyl rings. The
triplet centered at 8 = 7.51 ppm is observed for the H atom of the N-CH-N
fragment of the formamidinate groups with 103Rh-1H coupling of 30 Hz..

37

Figure 15

  

  

L
C39 C38

ORTEP Representation of [Rh2(DTolF)2(CH3CN)6]2+.

 

 

 

38

Table 2. Summary of important crystallographic data for
[Rh2(u-DT01F)2(CH3CN)61[BF412 (3)

 

 

formula Rh2C42H60N10B2F8
formula weight 1084.42
space group Pbca

a, A 21.646 (3)
b, A 31.272 (3)
c, A 14.561 (4)
01, deg 90

[3, deg 90

7, deg 90

v, A3 9857 (3)
Z 8

dcalc, g/cm3 ' 1.461

11 (MoKor), cm:1 7.27
temp, 0C ‘100 ;1-_ 2
no. of unique reflections, 3865

no. of variables 567
reflection/parameter ratio 6.82

Ra 0.046
wa 0.063

 

a R: ZIIFol - IFclI/ZIFo'. b Rw = [XWUFQI - ch|)2/ZWIF0|2]1/2; W =
1/62(|F0|)

39

Figure 16
a
C7
C4; ”C6
C2 5
C36 C19’18%221 Cl; ”C14 3 13
a M ’02
”a. ”"05 N1»? ”11
a a ’ ”a135,, a
C35 (:31 ' N9 . N2 ” a C10
’3 ,2 C42 5“” C9
,1.
a a N7 N10, 9
a C23 5 C27 5
C24 ’29 C40 C28
5
a C41
C30 -

PLUTO Representation of [Rh2(DTolF)6(CH3CN)6]2+
Depicting Torsion Angles.

 

40

 

 

 

 

Table 3. Selected bond distances and bond angles for
[Rh2(ll-DT01F)2(CH3CN)61[BF412 (3)
Bond Distances
A B A-B (A) A B A-B (A)
Rhl Rh2 2.560(1) Rhl N 1 2.026(8)
Rhl N6 2.00(l) Rhl N7 2.016(9)
Rhl N8 2.188(8) Rhl N9 2.030(8)
Rh2 N2 2.047(8) Rh2 N3 2.01(1)
Rh2 N4 2.211(8) Rh2 N5 2.052(8)
Rh2 N10 2.020(9) N1 C22 1.31(1)
N2 C22 1.33(1) N5 C42 1.32(1)
N9 C42 1.33(1)
Bond Angles

A B C A-B-C (0)
Rh2 Rhl N8 178.0(3)
N1 Rhl N7 177.4(4)
N6 Rhl N9 178.2(4)
N2 Rh2 N3 177.6(4)
N5 C42 N9 124.0(1)
N 1 C22 N2 123.0(1)
Rhl Rh2 N4 176.4(3)
N1 Rhl N6 88.7(4)
N1 Rhl N8 96.4(4)
N2 Rh2 N4 95.8(4)
N2 Rh2 N10 91.5

41

Table 4. Torsion angles of [Rh2(l.1-DTolF)2(CH3CN)6][BF4]2 (3)

 

 

A B C D A-B-C (0)
Rhl Rh2 N2 C22 -22.9(8)
Rhl Rh2 N5 C42 -25.3(8)
Rh2 Rhl N1 C22 ' -20.7(8)
Rh2 Rhl N9 C42 -17.9(8)
N1 Rhl Rh2 N2 18.6(4)

N5 Rh2 Rhl N9 18.5(4)

42

C. Spectroscopic and Crystallographic Properties of (4)

The reaction of (3) with two equivalents of 9—EtGH was carried out
in three different solvent systems. The use of CH3CN proved to be
unsuccessful due to the extreme insolubility of the purine in this solvent.
Upon addition of CH3OH to the CH3CN suspension, however, the reaction
mixture became green which signified that a reaction was occurring.
Prolonged heating in CH3OH should be avoided, as it reduces the dimetal
species to rhodium metal, which is not unusual in light of the fact that
methanol reduction of Rh2(OzCCH3)4(CH3OH)2 occurs readily under
conditions of heating.26’29 As found for the products formed from the
reaction with the u-tetracarboxylates and 9-EtGH, (4) exhibits two H8 1H
NMR resonances in a 1:1 ratio both in CD3OD and CD3CN at 8 = 8.32 and
8.35 ppm (Figure 17). The tWo triplets observed for the N-CH-N signals
at 8 = 7.50 and 7.55 ppm are also a direct indication of two types of
formamidinate environments with 103Rh-1H coupling of 30 Hz. The broad
resonances in the aromatic region of the formamidinate ligand as well as
the broadened ethyl regions of the purine further support this assumption.
The spectrum in deuterated methanol reveals a chemical shift for the H8
region of ~0.7 ppm.

Since the two isomers proposed for the structure of (4) differ in
polarities, it is conceivable that their solubilities will also differ. A slight
variation in solubility was observed when the mixture was subjected to the
slow addition of toluene. A 1H NMR spectrum of the precipitate obtained
from the separation exhibited two signals in the H8 region at 8 = 8.32 and
8.35 ppm in a 1:2 ratio, respectively (Figure 17a). The 1H NMR spectrum
of the filtrate reinforced this observation, as two signals in a reverse 2:1 '

ratio, were observed (Figure 17b).

 

43

Single crystals of (4) were obtained by slow diffusion of a
concentrated methanol solution layered with toluene at room temperature.
Unfortunately, X-ray data collection for this crystal had to be aborted due
to instrument failure. The preliminary unit cell for (4) supports the
conclusion that it is a new compound. Attempts to reproduce single
crystals of (4) are currently underway. The results for the unit cell are
presented in Table 5. The isolation of (5) was carried out in an attempt to
obtain better quality crystals of [Rh2(u-DTolF)2(u-9-EtGH)2L2]2+. A
concentrated solution of (5) in acetone was allowed to slowly diffuse into
toluene to afford long green fibers which were deemed unsuitable for
single crystal X-ray diffraction. Further anionic substituions are currently
underway.

D. 103Rh NMR Spectroscopy
1. Theory of 103Rh NMR Chemical Shifts

Although 103Rh is 100% abundant, 103Rh NMR spectroscopy is
difficult to obtain. The problems arise from the inherent sensitivity of this
nucleus, as well as the relatively long spin-lattice relaxation time and very
low frequency (5 = 3.16 MHz).2521

103Rh NMR spectroscopy allows for the direct spectroscopic probing
of the local symmetry at the metal center. The chemical shifts of 103Rh
nuclei are interpreted in terms of the Ramsey theory of nuclear magnetic

shielding. Ramsey's equation is comprised of two parts: (i) the diamagnetic
term (0d), produced by the charge distribution or the ground state electron

configuration and (ii) the paramagnetic term (op), produced by the
electronic transition of the ground state electrons and the hindrance of this
transition caused by other electrons and nuclei in the molecule. Although

the diamagnetic term is a dominant factor in determining NMR shifts for

44

Ewe—d 3 .: 23w $6928 em HENAFUHSENQbmHOIE
AOIuOZvuzwmfiw 3v 5 OUGOU.

 

 

 

 

 

 

1 m c

 

11 ill!

 

J—lfiiu—dfifiq-JJ—fid—-«qq_lql—d——qfiqq—<q1<44fiqddddqdflquqjq.dfiddl—

Ho 0 m u m m. n w m g 0 DUB

45

Ema—d 5

.I 27.5 .6825: cm _§~At-0._.o_3~€-c-m~05~

AOIwOZvfl—wmtm A3 3 ODuOU. $89388 mozoimbm
90 39:0: 0». 850:9

 

Pow

46

1556 E E 225 2503:: 04 _”HAPUHEENAtbmnOIX

AOIuOZVnzwmflu 3V 5 OUuOU. 3:38 mo=o¢<m=m
Bo 2530: on. 8:550.

 

 

 

mam mow
. . . _ _ . . . _ _ . . . _ . . . _ .
who ohm who «bu
{ {
Pom

 

Table 5. Summary of crystallographic data for

[Rh2(u-DTolF)2(u-9-EtGH)2(CH3CN )2] [BF412 (4)

 

 

formula
formula weight
a, A

b, A

c, A

01, deg

13, deg

11, deg
V, A3

Z
temp, °C

Rh2C58H54N16B2F8
1368.59
21.98 (3)
10.98 (1)
23.21 (2)
90

105.32 (9)
90

5402 (10)
4

'85 i 2

 

48

nuclei such as 1H, the paramagnetic term takes on a dominant role for the
variations of the chemical shifts for heavier nuclei such as 103Rh.
Blomberg et 01. has published a simplified version of Ramsey's equation

(Equation 2), the first term of which is the diamagnetic contribution.25b

The second term consists of the following factors: AEd-d, the lowest d-d
excitation energY; <rnd-3>, the d-orbital radii; Di, the imbalance of the d-
orbital electron population; and B, a constant. Hence the paramagnetic
shielding term is dependent upon d-d electron transition energies (AB). A
direct correlation between ligand field of the d7-d7 Rh211L4 complexes and
the 103Rh chemical shifts can be expected. Previous studies have revealed
that, although certain ranges of chemical shifts for the different oxidation
states can be discerned, the overlap of these shifts (Figure 18) prevents the
absolute assignment of oxidatiOn state based on 103Rh chemical shifts. The
reliable trend in 103Rh NMR spectroscopy is the influence of the ligand
field on the 103Rh chemical shift within a series of related compounds. A
higher AE created by a strong ligand field should induce an upfield
chemical shift. Conversely, a lower AE should generate a downfield
chemical shift.2521 A review article by Mann further elaborates that when
the paramagnetic term is small, the ligand field contribution would have
only a minimal effect on AE; a small change in the ligand field would then
be expected to have a large effect on the chemical shift when the
paramagnetic term is large. This is ill support of the findings reported by
Blomberg, et 01.25

Equation 2
5 = 'Apelectrons '1' (13(1'nd'3 )Di) / (AEd-d)

49

It is of importance to note that 103Rh NMR spectroscopy is highly
temperature dependent. Large shifts in the 103Rh NMR spectra of many
compounds occur with temperature changes (~2 ppm/K). Thus, the precise
temperature of the experiment must be known in order to accurately report
the 103Rh chemical shift within 1 ppm.

In addition, the dependence of the geometry of Rh compounds has
also been observed to have dramatic effects on the chemical shift. A
change from cis to trans geometry has been correlated with a downfield
shift in all documented cases. Similarly, downfield shifts are observed

when going from afac- to a mer geometry.25a

 

 

 

Figure 20
‘1573 - 19311:] Rh (V)
l 9931 - '1839 l Rh (III)
I 7644 -1395 l Rh (11)

 

 

2344 - ‘1224 Rh (1)

-1360. Rh (metal)
_|__|_|__l_l_l_l_l_
10000 8000 6000 4000 2000 0 ~2000 ppm

103Rh chemical shift ranges.

2. 103Rh NMR Spectral Studies of Dirhodium Antitumor Agents

The 103Rh signal observed for (2) at 5 = 5551.98 ppm represents a
noticeable upfield shift from the signal observed for Rh2(u-DTolF)2(ll-

OZCCF3)2(H20)2 at 5 = 5886.60 ppm. This suggests that the ligand field
strength of u-9-EtGH is somewhat stronger than that of u-trifluoroacetate.
The shift observed for (2) can also be compared to the 103Rh NMR shift

50

found for the comparable compound Rh2(mhp)4, which is a singlet at 5 =

5745 ppm.26 Although the singlet observed for (2) is a good indication of
a single of 103Rh environment, recall that the 1H NMR spectrum for this
compound suggests the presence of two isomers. The 103Rh NMR
spectrum expected for the "head-to-tail" isomer should resemble the
spectrum of Rh2(mhp)4, in which each Rh atom is equivalent. On the other
hand, the 103Rh NMR spectrum expected for the "head-to-head" isomer
should contain a two doublet signal for the two magnetically inequivalent
103Rh nuclei, akin to the 103Rh NMR spectrum of the “dimer of dimers
[Rh2(mhp)4]2”. Note that the signal-to-noise ratio in the 103Rh NMR
spectrum of (2) is very poor, such that only a singlet for the presumably
"head-to-tail" isomer is observed (Figure 19). The noisy baseline that is
observed in the spectrum is attributed to severe problems in tuning the
probe

The 103Rh NMR spectrum of the partially solvated complex, (3),
exhibits a singlet for the equivalent 103Rh nuclei at 5 = 4648.73 ppm. The
signal is observed ~1200 ppm upfield from the signal observed for Rh2(u—
DPhF)2(u-02CCF3)2(H20)2 in CD3CN. This indicates an increase of
electron density on the dimetal core increasing the ligand field (AE),
thereby increasing the shielding, and suggesting that the acetonitrile ligands
are somewhat stronger donors than the trifluoroacetate ligands. The 103Rh
spectrum observed for (3) is in accord with the results of the X-ray
diffraction studies.
4. Conclusion

Recent developments in our laboratories in the chemistry of
dirhodium tetra-carboxylate compounds with DNA purines opens the

question of the possible involvement of the 06 position in the observed

51

anticancer activity of this class of compounds. With the ultimate goal of
increasing the antitumor activity displayed by dirhodium complexes,

investigations of purine reactions of Rh2(u-form)2(u-OZCCF3)2(HZO)2 and A
[Rh2(u-DTolF)2(CH3CN)6]2+ have been explored. The products with 9-
EtGH formed from the reactions with Rh2(u-form)2(u-OZCCF3)2(HZO)2
and [Rh2(u-DT01F)2(CH3CN)6]2+ show little or no preference for the
formation of the "head-to-head" isomer over the "head-to-tail" isomer.
These compounds, then, appear to react similarly to the dirhodium
tetracarboxylate compounds that were previously studied in our
laboratories, and will most likely exhibit similar and possibly more
effective behavior in vivo. From this information, it is hoped that the
possible binding modes responsible for the antitumor activity of the
dirhodium complexes may be elucidated and compared to the growing

database of chemistry for dinuclear anticancer agents.

52

ans... 2 sea. 2:» 868... a. Extendiefoeonncefi:doze

:63 ea BEFencfieeebamioennax 96:65.

ammo

 

       

.1...

      

Fr

     

4 ..tfil

  

... «.4 4.1.4.1114! 44adqdqfi

_E __. ...r.-.

 

.. .E...=.:Z.. .__.: .. _ __ 55‘.

395 _ Goo _ ~80 :38 58c 88 CO8 38 «08 .38 ESE

mum _

 

 

«1fl11114—1d<<d41<W—Jquld444d—iiu4i141114dldii

88 88 88 $8 38 $8 88

«444daqdqd4qd-qdidi—

muoo mooo 38 2::

53

Ems—.6 3 Sum: 23% $835. 9. _§~At-UH05~AOIwOZVa :wmnn A8.

poem

 

 

 

«q—UA-q—d-qq—dqdqfidddddqadqflqqqdii—

mooo muoo mooo muoo aooo wuoc Mull

Chapter III

Dirhenium Anticancer Agents

54

55

1. Introduction

Our discoveries of unique bridging modes for guanine and adenine
bound to dinuclear transition metal compounds of rhodium and ruthenium
directed our interest to yet another class of dinuclear transition metal
complexes possessing the same lantern structure, namely Re2(III) bis- and
tetra-carboxylates. The cis-bis- and tetra-carboxylates of dirhenium(III)
have been known to exhibit carcinostatic activity, with results of in vivo

tests supporting the inhibition of DNA synthesis as the primary target for
antitumor activity. Previous antitumor studies of Re2(u-OZCCH3)4(SO4)

and cis-Re2(u-OZCR)2Br4 (R = CH3, CHZCH3, CHZCH2CH3) (Figure 20)
revealed inhibition of DNA akin to analogous dirhodium complexes, but
decomposition of the compounds led to the administering of doses far too
high for efficient antitumor activity. Interestingly, the researchers
observed what appeared to be rhenium oxide decomposition deposits
during the necropsies of their test animals.17

This research was initiated with the hypothesis that the N7, 06
bridging mode adopted by 9-EtGH in the dirhodium chemistry would also
be observed in the dirhenium systems. The requirement of the dinuclear
complex to easily undergo equatorial ligand exchange is the foundation on
which the likelihood of observing a bridging mode for the purines is based.
Although it has been previously stated that the bis- and tetra-carboxylates
of dirhenium(III) were observed to slowly undergo decomposition with
destruction of the dinuclear core, the complexes were nonetheless found to
inhibit DNA in vivo.17 Attempts in our laboratories to understand the
nature of DNA interactions has focused on reactions with guanine
derivatives such as 9-EtGH with dirhodium and diruthenium compounds.

Once an understanding of the coordination chemistry that may be involved

56

in DNA inhibition is obtained, modification of the complexes to improve
their ability to withstand the destruction of the quadruple bond will be
investigated. Such tailoring is expected to improve the antitumor

properties of the dirhenium compounds.

Figure 23

 

 

 

— CH3 _ 2+
A A R
0 CH3 0 O
.A A
.30 [To [’0 I‘O
R7—R' -> '- '°<-S
/ e I e S /Re /Re
O /O l BI’ 131'
y
H3C 0Y0 Br Br
CH3

 

 

R =' alkyl, 3 = solvent

2. Experimental
A. Synthesis

All reactions were carried out under an argon atmosphere by using
proper Schlenk-line techniques. The starting compounds,
(“BU4N)2(R62C13), (“BU4N)2(R62BI8), Cis-R62(u-02CR)zBr4L2, trans-
Re2(tt-OZCR)2Br4, Re2(u-02CR)4C12, and Re2(u-OZCEt)4(SO4)-nHZO

were prepared by literature methods.30

57

Preparation of cis-Re2(u-9-EtG)zBr4 (6)

An amount of cis-Re2(u-02CCH2CH3)2Br4(CH3OH)2 (60 mg, 6.65 x
10-5 mol) was dissolved in methanol (10 mL) and 9-EtGH (25 mg, 1.39 x
10'4 mol) was added to the stirring solution. The mixture was refluxed for
24 h during which time the color changed from green to dark purple-
brown. The solution was concentrated under reduced pressure and diethyl
ether (7 mL) was added to afford a dark purple-brown precipitate. The
reaction mixture was filtered in air, the purple-brown solid was washed
with 3 x 15 mL of hexanes followed by 2 x 10 mL of diethyl ether, and
dried in vacuo. The solid was reprecipitated from a methanol (10 mL) and
diethyl ether (5 mL) mixture cooled at “100C. The product was collected
by filtration in air and washed with 2 x 5 mL of diethyl ether, and dried in
vacuo (35 mg, 39 % yield). 1H NMR (CD3OD) 5 ppm: 8.94 (s, H8), 4.17
(q, CH2), 1.44 (t, CH3). Electronic absorption spectrum (CH3OH) Amax ,
nm (e): 511 (3021), 326 (26,660), 318 (26,190).

Compound (6) was also prepared from the bis-carboxylate
compounds cis-Re2(u-02CCH3)2Br4(CH3OH)2 and cis-Re2(u-
OZCCHZCHZCH3)2Br4(CH3OH)2 in nearly quantitative yields. Reactions
involving the trans isomers also yield (6) in quantitative yields from the
same synthetic approach.

Preparation of Re2(u-9-EtG)2(u-OZCCH3)4C12 (7)

Two equivalents of 9-EtGH (23 mg, 1.28 x 10'3 mol) were added to
a suspension of Re2(u-02CCH3)4C12 (52 mg, 6.40 x 10'5 mol) and CH3CN
(15 mL) and CH3 OH (2 mL) and the mixture was stirred at room
temperature for 24 h. A red-brown solid precipitated which was collected
by filtration in air. The compound was redissolved in CH3CN and

precipitated by addition of Eth and hexanes. The resulting precipitate

58

was washed with 2 x 5 mL of diethyl ether and dried in vacuo (23 mg, 34
% yield). 1H NMR (CD3OD) 5 ppm: 8.43(s, H8), 4.12 (q, CH2), 1.48 (m,
CH3).
Preparation of Re2(u-9-EtG)2(u-OZCCHZCH3)2Cl2 (8)

(a) A mixture of Re2(u-02CCH2CH3)4C12 (103 mg, 1.18 x 10‘4
mol) and CH3CN (10 mL) was treated with two equivalents of 9-EtGH (56
mg, 3.13 x 10'4 mol). Upon addition of methanol (~2 mL) the reaction
color immediately changed from salmon to orange. The mixture was
allowed to stir for an additional 2.5 h during which time a finely divided

red-brown precipitate was observed to form. The solid was filtered ill air
and dried in vacuo (10 mg 7.8 % yield). 1H NMR (CD3OD) 5 ppm: 8.45

(8, H8), 4.11 (q, CH2), 3.81 (q, CH2), 1.48 (m, CH3).

(b) A solution of Re2(u-OZCCH2CH3)4C12 (52 mg, 7.07 x 10'5 mol)
and CH3CN (15 mL) was combined with two equivalents of AgBF4 (30
mg, 1.54 x 10'4 mol) and the salmon colored mixture was allowed to stir
for 12-18 h. During this time the mixture became yellow-brown in color.
The reaction mixture was filtered through a Celite plug, two equivalents of
9-EtGH (27 mg, 1.51 x 10‘4 mol) were added to the filtrate, and the
mixture was stirred for 2 h, after which time CH2C12 (5 mL) was added to
afford the precipitation of excess Ag salts. The mixture was filtered in air
and the filtrate was concentrated under reduced pressure to yield a brown
residue (22 mg, 32 % yield). 1H NMR (CD3OD) 5 ppm: 8.80 (3, H8), 4.16
(q, CH2), 3.8, (m, CH2), 1.44 (t, CH3), 1.27 (t, CH3).

(c) A solution of Re2(n-02CCH2CH3)4(so4)nHzo (30 mg, 3.56 x
10'5 mol) and CH3CN (15 mL) was combined with two equivalents of 9-
EtGH (12.7 mg, 7.12 x 10'5 mol) and the mixture was stirred for 24 h.

The mixture, which turned red-brown during this time, was filtered in air

59

and concentrated under pressure to yield a red-brown solid ( 19 mg, 59 %
yield). 1H NMR (CD3OD) 5 ppm: 8.02 (5, H8), 4.15 (q, CH2), 4.09 (q,
CH2), 1.35 (t, CH3), 1.28 (t, CH3).
Preparation of Re2(u-9-EtG)2(u-OZCC6H6)2(CH3CN)2(BF4)2 (9)

The tetra-benzoate complex (63 mg, 6.79 x 10’5 mol) was suspended
in CH3CN (25 mL), treated with AgBF4 (26 mg, 1.34 x 10‘4 mol) and
stirred for 24 h. The green-brown mixture was filtered through a Celite
plug and reacted with two equivalents of 9-EtGH (20 mg, 1.11 x 10'4 mol)
for 12-16 h. The brown mixture was filtered in air, and the filtrate was
concentrated under reduced pressure to afford a brown residue. The solid
was dissolved in CH3CN (8 mL), precipitated with toluene (15 mL) and
collected by filtration. The solid was washed with 2 x 10 mL of toluene
and hexanes (35 mg, 45 % yield). 1H NMR (CD3OD) 5 ppm: 8.02 (s, H8),
7.43-8.00 (m, phenyl), 4.24 (m, CH2), 1.94 (s, ax-CH3CN), 1.50 (t, CH3).
B. X-ray Crystallography

The structure of cis-Re2(u-02CH2CH3)2Br4(DMF)2 (10) (DMF =
dimethylforrnamide) was determined using general applications that have
been described elsewhere. Geometric and intensity data were collected on
a Rigaku AFC6S diffractometer with graphite-monochromated MoKa (Au
= 0.71069 A) radiation and were corrected for Lorentz and polarization
effects. The calculations used for the structure determination were
performed with VAX computers on a cluster network within the
Department of Chemistry at Michigan State University using the Texsan
software package of the Molecular Structure Corporation.23,24
cis-Re2(u-OZCH2CH3)2Br4(DMF)2 (10)

(I) Data Collection and Reduction. Large single crystals of (10)
were obtained from a DMF solution of cis-(Re2(u -

60

OZCHZCH3)2Br4(CH3OH)2 that had been cooled to ~10 °C for 2-3 days.

A suitable single crystal with the approximate dimensions of 0.518 x 0.181
x 0.415 mm3 was mounted on the tip of a glass fiber with Dow Corning
grease. Cell constants were obtained from a least squares refinement using
24 carefully centered reflections in the range 29° < 20 < 37°. Data were
collected at 1'100 :1; 2 °C by using the (1)-scan method in the range 4 _<_ 20 _<_
47°. Weak reflections (those with F < lOo(F)) were rescanned at a
maximum of 2 rescans and the counts were accumulated to ensure good
counting statistics. A total of 3918 reflections were collected. Of the total
data collected, 1922 reflections were unique data with F02 > 36(F0)2.
Periodic measurement of three representative reflections at regular
intervals revealed that a 13% loss of diffraction intensity had occurred
during‘data collection. A decay correction was applied to account for the
phenomenon. All empirical absorption correction was applied on the basis
of azimuthal scans of 3 reflections with x near 90°.

(ii) Structure Solution and Refinement. The positions of all non-
hydrogen atoms were obtained by application of the direct methods. With
the exception of two atoms, all atoms were refined with anisotropic
thermal parameters. Hydrogen atoms were treated as fixed contributors at
idealized positions and were not refined. Disorder in the structure caused
certain atoms to refine as non-positive definite. The current least squares
refinement of 235 parameters gave residuals of R = 0.077 and RW = 0.093
and a goodness-of-fit = 2.93. A final difference Fourier map revealed the

highest peak in the difference map to be 2.6 e'/A3.

61

3. Results
A. Crystallographic Properties of (10)

The crystal structure of (10), the reported antitumor active
dirhenium(III) bis-carboxylate, was determined to verify the cis
arrangement of the propionate ligands.2°’3O Recall that one of the criteria
for antitumor activity observed for cisplatin and its related compounds is
the presence of cis labile groups. Crystal parameters and basic information
pertaining to data collection are summarized in Table 6. The Pluto
representation of (10) is depicted in Figure 21 and selected bond distances
and angles are summarized in Table 7.

Compound (10) crystallizes in the monoclinic-P crystal system.
Although a certain amount of disorder that involves the carboxylate ligands
and the bromide ligands is evident, a cisoid arrangement of the propionate
ligands was confirmed. The Re-Re bond distance (2.248 (2) A) is
consistent with other quadruply bonded compounds of this type (Re-Re
bond distances ~2.25 A). The compound conforms with the previously
established pattern of the bis-carboxylates which will be discussed in the
next section.7
B. Spectroscopic Properties of (6)

The structure of (6) was deduced from the 1H NMR and UV-visible
spectral studies and from extrapolation of the structural and chemical
properties of the cis-bis-carboxylates.26»3O Unlike the results found for the
analogous the dirhodium chemistry, the 1H NMR spectrum of (6) (Figure
22) in CD3OD exhibits only one singlet in the H8 region (8.94 ppm), which
suggests that only one product is formed in the reaction. This is a dramatic

shift of the H8 resonance compared to free 9-EtGH which is commonly
observed at ~7.6 ppm in CD3OD. The quartet centered at 4.17 ppm is

62

Table 6. Summary of crystallographic data for
Rez(ll’02CCH2CH3)2Br4(DMF)4 (10)

 

 

formula RezC12H24N2Br4O6
formula weight 984.36
space group P21 /n

a, A 9.825 (7)
b, A 13.258 (3)
c, A 18.433 (4)
01, deg 90

[3, deg 93.13 (9)
7, deg 90(0)

v, A3 2397 (2)
Z 4

dcalc, g/cm3 ' 2.727

p (MoKa), cm‘1 168.59
temp, °C '100 i 2
Ra 0.077
wa 0.093

 

a R: 2:111:01 - chll/ZIFOI. b Rw = [Zw(|Fo| - IFCI)2/2wlFo|2]1/2; w =
1/62(|Fo|)

63

Figure 24

 

 

PLUTO Representation of Re2(OzCCH2CH3)2Br4(DMF)2

64

Table 7. Selected bond distances and bond angles
for R62(u-02CCH2CH3)2Br4(DMF)2 (10)

Bond Distances

 

 

A B A-B (A) A B A-B (A)
Rel Re2 2.248(2) Rel 01 234(3)
Rel 02 202(3) Rel 05 204(3)
Re2 03 207(3) Re2 06 199(3)
Re2 04 247(3) Rel Brl 2.455(5)
Rel Br2 2.452(5) Re2 Br3 2.457(5)
Re2 Br4 2.467(4)

Bond Angles

 

 

A B C A-B-C (0)
Rel R62 Brl 105.1(1)
Re2 Rel Br2 104.7(1)
Rel Re2 Br3 104.3(1)
Rel Re2 Br4 104.6(1)
02 C4 O3 116(4)

OS C7 O6 127(5)

65

attributed to the CH2 protons of the 9-ethy1 group. The triplet centered at

1.43 ppm is the resonance expected for the shifted methyl protons of the 9-
ethyl group. No signals are observed for bound carboxylates, thus
implying that a substitution of these ligands for the purines has occurred.
Another feature discerned from the spectrum is that no signal for axial
solvent molecules was observed.

The presence of one H8 signal does not provide sufficient
information as to whether the 9-EtGH ligands are bound to the dirhenium
center in the “head-to-head” or “head-to-tail” fashion since a center of
symmetry exists in both proposed isomers relating the bound purines
(Figure 23). The indication of one isomer being formed counters the trend
observed in the chemistry of the dirhodium systems, for which both
isomers were observed in reactions of the cis- and tetra—acetate compounds.
It appears that the substitution of trifluoroacetate proceeds with specific
formation of only one product whereas the acetate displacement leads to

both possible isomers.

 

Figure 25
/\ /\
N O 0
”Br ’Br ”Br Br
[Re-7Re 0T ’RéjRe
N\/IO Nv-o I

 

66

Ema—d Ma

 

_: 23w £8233 om ”Bushmavnwfi A3 5 OUwOU.

 

 

67

The most likely structure for compound (6) is a cis orientation of
the purines, since the compounds of the type Re2(u-L)2X482 (L = bridging
ligand, X = halide, S = solvent) tend to adopt this configuration with the
presence of at least two solvent molecules in the axial positions.26,3O For
example, except for a few rare cases, the cis-bis-carboxylates of
dirhenium(IH) all contain the carboxylate ligands in the cis positions as
long as axial solvent molecules are present.26,31 It has been proven that
removal of the axial ligands occurs when heating the bis—carboxylate
compounds above 250 0C. At these temperatures, the dirhenium(IH) bis-
carboxylates sublime and isomerize to the trans derivatives in the vapor
phase (Figure 24). The trans-bis-carboxylates adopt a polymeric structure
with the halides assuming pseudo axial-equatorial positions which bridge
the dimetal centers. The cis-trans isomerization is reversible in solution

upon addition of neutral donor solvents.26

 

 

 

Figure 27
5x
0 O O
|,.X fix 250°C I,X |.,..~X
S—* Re Re‘—S ~ ‘ Re—Re
I I I + 28 / / I
DVD I X X
R X X 0Y0
R
X = halide
S = solvent

R = alkyl

 

68

Unless axial solvent molecules are displaced in the NMR solvent,
there is no spectrosc0pic evidence for the presence of axial solvent
molecules bound to (6). Compound (6) is suspected to contain the
deprotonated 9—EtG' ligand bridging the dimetal center in a cisoid fashion
through the N6 and 07 positions similar to other dimetal systems with this
purine. One possibility is that (6) is anextended structure with the
equatorial Br' atoms of one dinuclear center bridging a second dinuclear
center through its axial positions (Figure 25). It is unlikely that a cis-trans
isomerization occurred in the reaction, since the temperature was
maintained at the boiling point of methanol (~65 0C), and it has not been
shown that cis-trans isomerization occurs except at extremely high
temperatures and in the solid state. It cannot be ruled out, however, that
the purine could be bound in a transoid configuration since a center of
symmetry relating the bound purines would also exist, but only X-ray
crystallography can confirm the arrangement of the ligands.

The purine ligand present in compound (6) is expected to be the

anionic form of the ligand since the pr’s of the carboxylates present
(CH3C02', 9.25; CH3CH2COZ', 9.13; CH3CH2CH2C02', 9.19) in the
starting compounds are within the range of the well documented pKa of the
N1 position of bound 9-EtGH which is lowered upon metal binding from
~94 to ~8.0.7 The topic of the deprotonation of the N1 position was
treated in the previous chapter.

X-ray crystallographic studies to verify the spectroscopic
assignments of (6) are in order. A crystal structure would ascertain which

isomer is the preferred isomer formed in the reaction and would confirm

 

69

 

 

Figure 28
/\
N 0 T3” :20
Br Br """" Re—RC ------
..o‘ / I /
-----/-Re—/Re- ------ Br IBr I
N\/'O Br Br -----
----- Br Br
"head-to-head"
/\ OANe\
o N ,N’ ...0

Br ------- ' """
I fBr I o” (RC-7R6"
--""/'Rle—/Re' """" Br Br
N \/l() I Br Br """"
--------- Br Br

 

"head-to-tail"

the possible extended structure previously described. Attempts to

crystallize (6) are underway. The solubility of (6) is limited to H20 and

CH3OH which leaves very few choices of solvent combinations. All of the

combinations that were tried led to the conclusion that this compound tends
to form an amorphous solid. Derivatization of (6) by substituting the
coordinated Br' ligands for solvents groups in the presence of non-
coordinating anions such as [PF6]' in hopes of obtaining a compound with
better crystal packing properties were tried with little success.
C. Spectroscopic Properties of (7) _

As is the case of (6), compound (7) is only slightly soluble in water
and methanol. The 1H NMR spectrum of (6) was contaminated by

 

70

unreacted starting materials and unknown impurities. Attempts to purify
the product by reprecipitation resulted in major loss of the sample. No
further attempts to repeat this reaction were made since the biological
studies of Re2(u-02CCH3)4C12 indicated that the propionate analog of this
complex exhibited far greater antitumor activity.

D. Spectroscopic Properties of (8)

(a) One of the problems associated with investigating the properties
of the tetra—acetate of dirhenium(IH) is the relatively strong binding
interactions of the axial chlorides, which result in low solubility and slow
substitution reactions. The more biologically relevant tetra-carboxylate of
dirhenium(IH), the tetra-propionate derivative, is far more soluble than its
acetate analog simply because of the presence of the more soluble propyl
substituents. The presence of strongly bound axial halides in the dirhenium
tetra-carboxylates, in general, require that substitution reactions take place
not axially, but via dissociation of the equatorial ligands, thus causing
reactions to be very sluggish. Equatorial attack of the Re26+ core is
rationalized by the necessity for deprotonation and displacement of the
propionate ligands to occur simultaneously (Figure 26).

Although the poor yield of this reaction is discouraging, the 1H N MR
spectrum of (8a) in CD3OD exhibits a broad singlet in the H8 region of 9-
EtGH at 8.45 ppm which verifies that a reaction did indeed take place
(Figure 27). The H8 region of this spectrum also shows a broad signal of
multiplets centered at approximately 8.0 ppm. This suggests that several
products are being formed in this reaction. Reprecipitation of the

compound from CH3CN with EtzO and hexanes allowed for the separation

of the major product from the impurities. The 1H NMR spectrum in

71

CD3CN indicates that unreacted dirhenium tetra-propionate was also
present.

(b) The [BF4'] salt of Re2(u-OZCCH2CH3)4C12 is generated in situ
by the addition of two equivalents of AgBF4 in order to help facilitate the
reaction with 9-EtGH. Without the presence of the axial chloride ligands,
the Re26+ unit has available the axial and equatorial sites for attack by the
purine. The 1H NMR spectrum of (8b) in deuterated methanol exhibits
only one signal in the H8 region of 9-EtGH at 8.80 ppm. This chemical

shift is analogous to the corresponding value observed in the spectrum of
(6). The quartet observed at 4.15 ppm is attributed to the CH2 protons of

9-ethyl substituent while the quartet at 4.09 ppm can be assigned to the CH2
protons of the bound propionate ligands. The CH3 protons resonate at 1.28

and 1.35 ppm for the bound propionate and 9-ethyl group on the guanine,
respectively.

Analogous to the reactions of the tetra-carboxylates of dirhodium
and diruthenium with 9-EtGH, cis substitution of two acetates is expected to
occur for the compounds Re2(u-OZCR)4X2.26 The observed trend for
these molecules to undergo cis substitution as the more energetically
favored pathway has been established.26 Thus the proposed structure of
(8b) would be similar to the structure of Rh2(u-9-EtG(H))2(u-OZCR)2L2
wherein the purine ligands are arranged in a cis configuration with respect
to each other.19 Furthermore, the bound 9-EtGH is expected to be
deprotonated at the N1 position due to the basicity of the propionate
leaving groups (pr ~9.13).

It was also hoped that the presence of different ligands such as the
propionate ligands might aid in the isolation of single crystals suitable for

X-ray crystallography. Recall that compound (6) is suspected to exhibit

 

72

poor crystal packing arrangements leading to amorphous solids.
Unfortunately, similar properties were also observed for (8b).

(c) The [SO4]2' salt of the tetra-propionate complex was found to
decompose in water, but is quite stable in CH3CN. The 1H N MR spectrum
of this compound shows that the reaction does not proceed cleanly, as
evidenced by the 1H N MR spectrum depicted in Figure 28. The resonance
for the H8 region of 9-EtGH is assigned to the signal observed at 8.00 ppm.
Both the quartet and triplet expected for the methylene and methyl protons
of the purine are observed at 4.15 and 1.35 ppm, respectively. The
methylene and methyl protons of bound propionate are found at 4.09 and
1.28 ppm, respectively. The spectrum also exhibits signals for unreacted
starting material. The compound (8c) is proposed to exhibit a structure
similar to (8b) as evidenced by 1H NMR spectroscopy which supports the
assignment of two bound 9-EtG' and two bound propionate ligands.

E. Spectroscopic Properties of (9)

The difficulty in crystallizing the compounds (6)-(8) led us to
investigate possible model compounds that might possess enhanced crystal
packing properties. Unfortunately, compound (9) precipitates from
solution to form amorphous solids. The pr of the benzoate ligands
(~9.81) is well within the range for deprotonation of the N1 position of 9-
EtGH to occur, therefore it is expected that the deprotonated form of the
purine is present. It would also seem to appear that the trend of only one
isomer being formed in the dirhenium(IH) chemistry is upheld. The 1H
NMR spectrum of (9) indicates only one H8 proton at 8.02 ppm, which is
shifted downfield from the H8 signal from ~7.6 ppm for the free ligand.

The aromatic protons of the benzoate ligands resonate in the range of 7.43-

8.00 ppm. The CH2 protons of the purines are observed in the expected

73

region at 4.24 ppm in a 1:1 ratio with the benzoate protons. The triplet
antipicipated for the CH3 protons of 9-EtG' are assigned to the signal at
1.50 ppm. As determined from the integration, two of the benzoate groups
are bound to the metal in a 1:1 ratio with 9-EtG'. By analogy to the
compounds formed from dirhenium(IH) tetra-carboxylates and 9-EtGH,
(9) is expected to contain a cis configuration of its equatorial ligands. The
presence of the 1H NMR signal at 1.94 ppm suggests that acetonitrile
groups occupy the axial positions of the dinuclear unit and are exchanged
in the NMR solvent.2

4. Conclusion

The investigation of dirhenium compounds as antitumor agents has
been largely ignored by the medical community due to the large amounts,
of compound required to effectively inhibit tumor growth. It has been
noted that the reason for high dosage rates of dirhenium tetra-carboxylates
is that compounds of this type readily decompose in vivo.3 No prior
investigations of the coordination chemistry of these systems with DNA
nucleic acids has been investigated. Furthermore, no attempts to tailor the
ligand environment to prevent facile decomposition under physiological
conditions have been reported.

It has been found that compounds of dinuclear rhenium(III) bind to
the model DNA nucleobase 9-ethylguanine. The reactions proceed via
substitution of two cis carboxylate ligands for two deprotonated 9-
ethylguanine ligands. It has been observed that the products form only one
of the two possible isomers. The establishment of this product as the
“head—to-head” or “head-to-tail” isomer must be provided by X-ray

crystallography. Regardless of the isomer question, it is expected that the

74

X-ray study will confirm that two purines bridge the dinuclear center

through the O6 and N7 positions in all cases.

 

75‘

 

 

 

3min we

 

 

 

r u 3:225:85 333838 cm :6 08.605458 :meW 3 PEG:

76

Enid u: _I 2?.» $0923 cm weNAt-o-anv~€-O~OOINOIw505 33
E OUuOZ.

 

 

 

 

MTG

 

 

 

I» I r 7);? P ¥
I 144 (I JII I

4 4 44
._.__._2.qq____.__..._..q.__a.u—a.2___2q.__42.___.__q2_._1«.fi—1

am up Ho 0 m V m m A w m » o

77

fluid u— 5* 23% $5055. a». Nowere-mHQvmfitbunnmunmuvflwmtN
$3 5 OUmO—u.

 

 

 

was

-r e- r L :

 

 

 

 

 

 

 

 

 

‘ I! I 74 I

q ‘

 

u—_dqq—dd-4—_-q_-144—4qqd—_ddqdqdH—fifiddddquad—dq—«_1fi14—dqdu—qd

pm 2 90 m m V m m A w m p o

78

3%.; S E zzw €855 a weXe-e-m49~€-o~no:~n:uEmoe
88 m: OUwOU.

 

 

mbo

 

 

 

 

 

_—_—q—q—_-———qq—_——-—u_—q_——_qq———_—u——‘—“—q—q—__q__d_.__~I._l

S m m V m m R w m u 0 use

79

Ens-d um _I 225 £62qu o». woNAtbanfiAFONOOa:uvnAwEv N
GV E 98.00.

mbw

 

 

 

 

 

 

L 1 ‘EFILE _
_2_4___44.._..5._._4.2_._._........___.4.«4

m u m m e w m H 0 e2:

 

 

80

List of References
1. (a) Rosenberg, B.; VanCamp, L. Nature 1969, 222, 385. (b) Thomson,
A. Proceedings of the Second International Symposium on Platinum
Coordination Complexes in Cancer Chemotherapy, Connors and Roberts,
Ed.; Springer-Verlag; Berlin, 1974. (c) Eastman, A. Pharmac. Ther.
1987, 34, 155. (d) Pasini, A.; Zunino, F. Angew. Chem. Int. Ed. Engl.
1987, 26, 615 and references therein. (e) Frey, U.; Ranford, J. D.; Sadler,
P. J. Inorg. Chem. 1993, 32, 1333 and references herein. (f) Howell, S.
B., Ed. Platinum and Other Metal Complexes in Cancer Chemotherapy
Plenum Press, New York, 1991. (g) Baker, S. A.-B.; Perez-Soler, R,;
Khokhar, A. R. J. Coord. Chem. 1993, 29, 1. (g) McAuliffe, C. A.;
Shanna, H. L.; Tinker, N. D. Chemistry of the Platinum Group Metals
Recent Developments, Studies in Inorganic Chemistry 11, Hartley, F. R.,
Ed.; Oxford, New York, 1991, Ch.16. (h) Keppler, B. K. New J. Chem.
1990, 14, 389.
2. (a) Umapathy, P. Coord. Chem. Rev. 1989, 95, 129. (b) Lippard, S. J .,
Ed. Prog. Inorg. Chem. Wiley, New York, 1987, 37 ,1 and references
therein. (c) Sundquist, W. 1.; Lippard, S. J. J. Coord. Chem. Rev. 1990,
100, 293 and references therein.
3. Lippard, S. J .; Berg, J. M. Principles of Bioinorganic Chemistry, Univ.
Sci. Books; Mill Valley, CA; 1994.
4. (a) Burgess, J. Trans. Met. Chem. 1993, 18, 439. (b) Hollis, L. S.;
Amundsen, A. R.; Stern, E. W. J. Med. Chem. 1989, 32, 136. (c) Allesio,
E.; Balducci, G.; Calligaris, M.; Costa, G.; Attia, W. M.; Mestroni, G.
Inorg. Chem. 1991, 30, 609.
5. (a) Harder, H. C.; Rosenberg, B. Int. J. Cancer. 1970, 6, 207. (b)
Howle, J. A.; Gale, G. R. Biochem. Pharmocol. 1970, 19, 2757. (c)

 

81

Sherman, S. E. Lippard, S. J. Chem. Rev. 1987 , 87, 1153.

6. Reedijk, J. Pure & Appl. Chem. 1987, 59, 181.

7. Lippard, S. J. Pure & Appl. Chem. 1987, 59, 731.

8. Prescott, L. M.; Harley, J. R; Klein, D. A. Microbiology, Wm. C.
Brown Pub. New York, 1990, Ch. 10.

10. HoWe-Grant, M. E., Lippard, S. J. Metal Ions in Biological Systems ;
Sigel, H., Ed.; Marcel Dekker; New York, 1980; Vol. 2, p.63.

12. (a) Pascoe, J.; Roberts, J Biochem. Pharmacol. 1974, 23, 1345. (b)
Zwelling, L. Cancer Research 1978, 38, 1762 and references therein.

13. (a) den Hartog, J.; Altona, C.; van Boom, J.; van dere Marcel, A.;
Reedijk, J. J. Am. Chem. Soc. 1984, 106, 1528. (b) den Hartog, J.; Altona,
C.; Chottard, J .; Girault, J.; Lallemand, J.; DeLeeuw, F.; Marcelis, A.;
Reedijk, J. Nucl. Acids Res. 1982, 10, 4715.

14. (a) Sherman, S. E.; Gibson, D.; Wang, A. H. -J.; Lippard, S. J. J. Am.
Chem. Soc. 1988, 110, 7368. (b) Sherman, 8.; Gibson, D.; Wang, A.;
Lippard, S. Science, 1985, 230, 412 and references therein.

15. (a) van der Veer, J .;Ligtvoet, G.; van der Elst, H.; Reedijk, J. J. Am
Chem. Soc. 1986, 108, 3860. (b) van der Veer, J.; van der Elst, H.; den
Hartog, J .; Reedijk, J. Inorg. Chem. 1986, 25, 4657.

16. (3) Beck, D.; Brubaker, R. J. Bacteriol. 1973, 116, 1247. (b)
Ciccarelli, R.; Solomon, M.; Varshavsky, A.; Lippard, S. Biochemistry
1985, 24 , 7533. (c) Sundquist, W.; Lippard, S.; Stollar, B. Biochemistry
1986, 25, 1520.

17. (a) Dmitrov, N. V.; Eastwood, G. W. Current Chemother. Proc. Int.
Cong. Chemother. 10th 1977. 1978, 2, 1319. (b) Eastland, G. W.; Yang,
G.; Thompson, T. Meth. and Find. Exptl. Clin. Biochem. 1989, 10, 41.
18. (a) Erck, A.; Sherwood, E.; Bear, J. L.; Kimball, A. P. Cancer Res.

 

82

1976, 36, 2204. (b) Bear, J. L.; Gray, Jr., H. B.; Rainen, L., et. al. Cancer
Chemotherapy Reports 1975, 59, 611. (c) Howard, R. A.; Spring, T. G.;
Bear, J. L. Cancer Res. 1976, 36, 4402. ((1) Chen, J .; Kostic, N. M. Inorg.
Chem. 1988, 27, 2862.

19. Dunbar, K. R.; Matonic, J. H.; Saharan, V. P.; Crawford, C. A.;
Christou, G. J. Am. Chem. Soc. 1994, 116, 2201.

20. (a) Kopf-Maier, P.; Kopf, H. Struct. Bond. 1988, 70, 104. (b) Kopf-
Maier, P.; Kopf, H. Drugs of the Future 1986, 11, 297. (b) Kuo, L. Y.;
Kanatzidis, M. G.; Sabat, M.; Tipton, A. L.; Marks, T. J. J.Am. Chem.
Soc. 1991, 113, 9027 and references therein. (c) Heim, M. E.; Flechtner,
H.; Keppler, B. K. Prog. Clin. Biochem. Med. 1989, 10, 217.

21. (a) Bear, J. L.; Howard, R. A.; Dennis, A. M. Curr. Chemother., Proc.
Int. Cong. Chemother., 10th. 1978, 1977, 1321. (b) Rao, P. N.; Smith, M.
L.; Pathak, S.; Howard, R. A.; Bear, J. L. J. Natl. Cancer Inst. 1980, 64,
905. (c) Bear, J. L.; Gray, H. B.; Rainen, L.; et al. Cancer Chemother.
Rep. 1975, 59, 611. ((1) Howard, R. A.; Spring, T. G.; Bear, J. L. Cancer
Res. 1976, 36, 4402.

22. (a) Piraino, P.; Tresoldi, G.; Schiavo, S. L. Inorg. Chim. Acta 1993,
203, 101. (b) Schiavo, S. L.; Sinicropi, M. S.; Tresoldi, G.; Arena, C. G.;
Piraino, P. J. Chem. Soc. Dalton Trans. 1994, 1517. (c) Fimiani, V.;
Ainis, T.; Cavallaro, A.; Piraino, P. J. Chemother. 1990, 2, 319. (d)
Piraino, P.; Bruno, G.; Tresoldi, G.; et al. Inorg. Chem. 1987, 26, 91.

23. TEXSAN-TEXRAY Structure Analysis package, Molecular Structure
Corporation, 1985. f

24. (a) MITHRIL: Integrated Direct Methods Computer Program,
Gihnore, C. J. J. Appl. Cryst. 1984, 17, 42. (b) DIRDIF: Direct Methods

for Difference Structures, An Automatic Procedure for Phase Extension;

 

83

Refinement of Difference Structure Factors. Beurskens, P. T. Technical
Report, 1984.

25. (a) Mann, B. E. Annual Reports on NMR Spectroscopy 1991, 23, 141
and references therein. (b) Akermark, B.; Blomberg, M. R. A.; et al. J.
Amer. Chem. Soc. 1986, 116, 3405. (c) Brown, T. H.; Green, P. J. J.
Amer. Chem. Soc. 1970, 92, 2539. (d) Juranic', N. Coord. Chem. Rev.
1989, 96, 253. (e) Brevard, C.; Schirnpf, R. J. Mag. Res. 1982, 47, 528.
(f) Bonnaire, R.; Davoust, D.; Platzer, N. Org. Mag. Res. 1984, 22, 80.
(g) Boyar, E. 13.; Robinson, 3. D. Inorg. Chim. Acta 1982,64, 1.193. (h)
Carr, C.; Glaser, J.; Sandstrom, M. Inorg. Chim. Acta 1987, 131, 153.

26. Cotton, F. A. and Walton, R. A. Multiple Bonds Between Metal Atoms
Oxford Press, 2nd ed. 1993, and references therein.

27. (a) Webb, G. A. Annu. Rep. on NMR Spect. 1983, 14. (b) Garner, C.
D.; Hughes, B. Adv. in Inorg. and Radiochem. 1975, 17, 1. (c) Matonic J.
H.; Chen, S. J.; Pence, L. E.; Dunbar, K. R. Polyhedron 1992, 11, 541.
28. Dunbar, K. R.; Majors, S. 0.; Sun, J. S. Inorg. Chim. Acta 1995, 229,
373.

29. Rempel, G. A.; Legzdins, P.; Smith, H.; Wilkinson, G. Inorg. Synth.
13, 90.

30. (a) Walton, R. A.; Srinivasan, V. Inorg. Chem. 1980, 19, 1635. (b)
Cotton, F. A.; Oldham, C.; Walton, R. A. Inorg. Chem. 1967, 11, 214. (0)
Cotton, F. A.; Oldham, C.; Walton, R. A. Inorg. Chem. 1966, 10, 1798.
(d) Misailova, T. V.; Kotel'nikova, A. S. et al. Russ. J. Inorg. Chem.
1981, 26, 343 and references therein.

31. Cotton, F. A.; DeCanio, E. C.; Kibala, P. A.; Vidyasagar, K. Inorg.
Chim. Acta 1991, 184, 221.

 

84

APPENDIX

TABLES OF ATOMIC POSITIONAL PARAMETERS AND
EQUIVALENT ISOTROPIC DISPLACEMENT PARAMETERS

85

TABLE A.l. Atomic positional parameters and equivalent isotropic

displacement parameters (A2) and their estimated standard deviations for
[Rh2(u-DT01F)2(CH3CN)6][BF4]2 (3).

 

 

atom x y z B(eq)
Rh(1) 0.63807(4) 0.14889(3) 0.10314(5) 1.71(3)
Rh(2) O.56796(4) O.l4524(3) —0.03820(5) 1.70(3)
N (1) 0.5862(4) 0.0990(3) 0.1470(5) 1.8(4)
N(2) 0.5077(4) 0.1109(3) 0.0413(5) 1.9(4)
N(3) 0.6278(4) 0.1766(3) -0.1195(6) 3.0(5)
N(4) 0.5125(4) 0.1412(3) -0.1654(6) 2.4(4)
N (5) 0.6158(4) 0.0904(3) -0.0683(5) 2.0(4)
N(6) 0.5830(4) 0.1882(3) 0.1743(6) 2.4(4)
N(7) 0.6864(4) 0.1997(3) 0.0572(6) 2.3(4)
N(8) 0.7004(4) 0.1503(3) 0.2215(6) 2.3(4)
N(9) 0.6943(4) 0.1080(3) 0.0344(6) 1.9(4)
N(10) 0.5240(4) 0.2005(3) -0.0081(6) 2.2(4)
C(1) 0.6088(5) 0.0694(3) 0.2157(7) 2.1(5)
C(2) 0.6234(5) 0.0276(4) 0.191 1(8) 2.7(5)
C(3) 0.6138(5) 0.0821(3) 0.3075(7) 2.1(5)
C(4) 0.6452(5) -0.0003(4) 0.258(1) 3.4(6)
C(S) 0.6345(6) 0.0530(4) 0.3722(7) 3.5(6)
C(6) 0.6492(5) 0.01606(4) 0.3487(9) 3.6(6)
C(7) 0.6721(7) -0.0202(5) 0.420(1) 6.4(9)
C(8) 0.4467(5) 0.0997(4) 0.0147(6) 2.1(5)
C(9) 0.4101(5) 0.1300(3) -0.0285(7) 2.6(5)
C(10) 0.3508(5) 0.1198(4) -0.0602(7) 2.9(6)
C(l 1) 0.3253(5) 0.0789(4) -0.0453(8) 3.1(6)
C(12) 0.2606(5) 0.0678(5) -0.078(1) 4.7(7)
C(13) 0.3618(5) 0.0494(4) 0.0003(8) 2.9(5)
C(14) 0.4219(5) 0.0589(3) 0.0288(8) 2.8(5)
C(15) 0.5994(5) 0.0680(3) -0.1498(7) 1.8(5)
C(16) 0.6327(5) 0.0738(4) -0.2302(8) 3.9(6)
C(17) 0.6112(6) 0.0558(5) -0.3111(8) 4.6(7)
C(18) 0.5571(5) 0.0316(4) -0.3l38(7) 3.0(6)
C(19) 0.5331(6) 0.0151(5) -0.4040(9) 4.6(7)
C(20) 0.5268(5) 0.0258(3) -0.2332(8) 2.6(5)
C(21) 0.5466(5) 0.0432(3) -0.1515(7) 1.9(5)
C(22) 0.5310(5) 0.0907(3) - 0.1 139(7) 2.0(5)

C(23)

0.6557(6)

0.1925(5)

-0.178(1)

5.3(8)

TABLE A.1. continued.

atom

C(24)
C(25)
C(26)
C(27)
C(28)
C(29)
C(30)
C(31)
C(32)
C(33)
C(34)
C(35)
C(36)
C(37)
C(38)
C(39)
C(40)
C(41)
C(42)
B(l)
B(2)
F( 1)
F0.)
W)
W)
F(S)
F(6)
F(7)
F(8)
H(l)
H(2)
H(3)
H(4)
H(S)
H(6)
H(7)
H(8)
H(9)
H(lO)

X

0.6902(8)
0.4863(6)
0.4532(6)
0.5590(6)
0.5275(7)
0.7178(5)
0.7568(6)
0.7585(5)
0.8019(5)
0.8629(6)
0.8798(6)
0.9452(7)
0.8355(6)
0.7748(5)
0.7372(6)
0.7862(7)
0.4940(6)
0.4568(6)
0.6722(5)
0.356(1)
0.604(1)
0.5908(5)
0.6595(4)
0.5588(5)
0.61 1 1(6)
0.3365(6)
0.3072(6)
0.3847(7)
0.3923(6)
0.6181
0.6035
0.6575
0.6383
0.6454
0.7126
0.6717
0.4254
0.3271
0.2431

86

3’

0.2139(8)
0.1358(3)
0.1284(4)

Z

-0'.251(1)

-0.2323(8)
-0.3223(8)

0.21 18(4)
0.2416(6)
0.2273(4)
0.2635(4)
0.1026(3)
0.1293(4)
0.1269(4)
0.0986(4)
0.0987(5)
0.0712(4)
0.0728(3)
0.1476(4)
0.1446(5)
0.2299(4)
0.2690(4)
0.0859(3)
0.2409(8)
0.1716(6)
0.1937(3)
0.1501(3)
0.1439(4)
0.1982(4)
0.2131(4)
0.2631(4)
0.2232(4)
0.2715(4)
0.0181
0.1 104
-0.0281
0.0616
-0.0444
-0.0289
-0.0067

0.2213(8)
0.283(1)
0.0347(8)
0.007(1)
0.0605(6)
0.0233(8)
0.056(1)
0.1251(9)
0.163(1)
0.1574(8)
0.1271(7)
0.2762(7)
0.3475(8)
0.0008(7)
0.015(1)

-0.361(7)

0.746(2)
0.516(1)
0.5897(7)
0.5290(8)
0.4974(8)
0.4455(8)
0.6836(8)
0.771(1)
0.810(1)
0.704(1)
0.1297
0.3254
0.2396
0.4343
0.4212
0.4052
0.4784

0.1580
0.1405
0.0919

-0.0380
-0.0920
-0.1084

B(eq)

12(1)
3.1(6)
4.7(7)
3.9(7)
8(1)
2.9(6)
5.1(8)
2.0(5)
3.1(6)
3.9(6)
3.6(6)
7(1)
3.5(6)
2.4(5)
3.0(5)
5.5(7)
2.8(6)
5.3(8)
2.2(5)
7.1(5)
5.3(4)

10.5(7)
9.5(6)

11.2(7)

11.5(8)

13(1)

15(1)

15(1)

13(1)
3.3
2.6
4.1
4.3
7.7
7.7
7.7
3.2
3.4
5.7

87

TABLE A.1. continued.

atom x y z B(eq)
H(l 1) 0.2630 0.0446 -0.1207 5 .7
H( 12) 0.2538 0.0599 -0.0278 5.7
H(13) 0.3455 0.0218 0.0120 3.5
H(l4) 0.4456 0.0375 0.0582 3.3
H(15) ' 0.6699 0.0894 -0.2297 4.7
H(l6) 0.6338 0.0598 . -0.3666 5.5
H(17) 0.4970 -0.0009 -0.3936 5.6
H(l8) 0.5246 0.0386 -0.4431 5.6
H(19) 0.5639 -0.0025 -0.4315 5.6
H(20) 0.4904 0.0089 -0.2335 3.2
H(21) 0.5240 0.0381 -0.0969 2.4
H(22) 0.5068 0.0694 0.1428 2.5
H(23) 0.7314 0.2018 -0.2525 14.2
H(24) 0.6704 0.2075 -0.3070 14.2
H(25) 0.6924 0.2430 -0.2401 14.2
H(26) 0.4130 0.1399 -0.3166 5.8
H(27) 0.4747 0.1424 -0.3705 5.8
H(28) 0.4514 0.0988 -0.3336 5.8
H(29) 0.5569 0.2546 0.3209 9.4
H(30) 0.5069 0.2626 0.2465 9.4
H(31) 0.4974 0.2270 0.3194 9.4
H(32) 0.7815 0.2718 0.0578 6.2
H(33) 0.7821 0.2542 -0.0418 6.2
H(34) 0.7316 0.2865 -0.0119 6.2
H(35) 0.791 1 0.1490 -0.0237 3.7
H(36) 0.8938 0.1451 0.0301 4.6
H(37) 0.9479 0.774 0.2090 7.9
H(38) 0.9279 0.0927 0.1 147 7.9
H(39) 0.9534 0.1259 0.1883 7.9
H(40) 0.8469 0.0504 0.2021 4.3
H(41) 0.7446 0.0539 0.1515 2.9
H(42) 0.7670 0.1433 0.4058 6.7
H(43) 0.8094 0.1197 0.3367 6.7
H(44) 0.8123 0.1691 0.3446 6.7
H(45) 0.4829 0.2922 0.0292 6.3
H(46) 0.4358 0.2746 -0.0416 6.3
H(47) 0.4283 0.2639 0.0619 6.3
H(48) 0.6987 0.0659 -0.0649 2.6

H(49)

0.6638

0.2172

-0.1386

5.8

88

TABLE A.1. continued.

atom x y z B(eq)
H(50) 0.6895 0.1730 -0.1681 5.8
H(51) 0.4907 0.1654 -0.2486 3.1
H(52) 0.5222 0.1211 -0.2585 3.1
H(53) 0.5191 0.2049 0.1928 4.4
H(54) 0.5627 0.1917 0.2724 4.4
H(55) 0.6819 0.2442 0.0184 3.0
H(56) 0.7270 0.2338 0.0981 3.0
H(57) 0.7685 0.1573 0.23 30 3.3
H(58) 0.7219 0.1731 0.3064 3.3
H(59) 0.5065 0.2323 0.0648 2.8
H(60) 0.5232 0.2470 -0.0339 2.8

 

 

 

 

.1...

z... .3? 8.3.6.

 

124:1. ....... j . L.
y _ 1...... . I
.. 2...... . .. t...

.._..:.: . ...z . . . . 2......

 

. 91:15::
,1 .2 1.5..

5:2 .2,
: 52.3....
1......)
5;:
1:}...

a: .
.. lawfua
..mcr gram.

. r.
L: .
.z....;..:..,.....?
5...... ....r.:.......
. : r 1

r
.r .5
1%.

e .a v2... .
a... , .r... .
551:

5 l 1 . . .
. — m— t .1 v3.0 .517.— .. 1.. 72:.
”Sizes... 5... ..y :2 .7 1 e
.9: A'J’C—II- 1.3-wan!— .1»; I
._ . 3.1.51 rut... . .m... ..n: . E. .9“
.... . . . . 15.5.... 3157......
... . .. : ..r. .3. a.) : a.
re . :7 . .u e
2.... 2. . .. 3.1....
I... 17.3....
.
..

.1 m
,
.9.

.. .. ..
r... 53..

31.1.8.2.
.. :2
1.

.
.r.

.
“.9932... it
.. "Em. . a. ..
. a"...

..u

_
afghan. 55:5,... . 55.1.5.

t...

 

.a

...i...

3.... :r.

4
2.”; a}!

a.
...F

 

 

 

 

 

 

53:6.

..N x

t. a...

..t e;

m
a...
a.

a...

. . .
$2.1

£3,
a?

.
.

1"

._

”fir