.. in
tL: Tb
. z.- “ .

"an. an.“ u. s ..
342E“.
3.€.s . .l.
. r .t.
.

9...}
VT 1'. 1;.ltp.l

.292“ 332?; 129.

5.; 1‘
. .0 I

 

llllllllllllllll

LIBRARY
Michigan State
University

 

 

 

 

This is to certify that the

thesis entitled

lrflCmcfi 003 cu? ’D‘MUUW Translhfin
Melcll (gnaw/Mes WNW Ammo AUC‘S)
NUCkGthS amdprgefinjted by

Jennifer 3mm H655

has been accepted towards fulfillment
of the requirements for

A degree in W

76211 /<° (Quin—a

Major professor

 

Date /32/// 7/75

0-7639 Msui, an 1743:..-5... . ' "‘1

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

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

1/96 WWW.“

 

INTERACTIONS OF DINUCLEAR TRANSITION METAL COMPLEXES WITH
AMINO ACIDS, NUCLEOTIDES, AND DNA

BY

Jennifer Simone Hess

A THESIS
Submitted to
Michigan State University

In partial fulfillment of the requirements
For the degree of

MASTER OF SCIENCE
Department of chemistry

1 998

ABSTRACT
INTERACTIONS OF DINUCLEAR TRANSITION METAL COMPLEXES WITH
AMINO ACIDS, NUCLEOTIDES, AND DNA
By

Jennifer Simone Hess

In the 1960’s the discovery that cis-platin was a potent antitumor agent
paved the way for the use of metal-based drugs in modern chemotherapeutic
applications. The severe Side effects exhibited by this drug have led to the
investigation of new types Of metal-based compounds as potential antitumor
agents. Dinuclear transition metal compounds are among these compounds.
These have been found to exhibit carcinostatic activity, and like ciS-platin, are
inhibitors of DNA replication. In order to understand potential mechanisms which
could lead to the development of more effective treatments, the interactions Of
these antitumor agents with cellular molecules must be studied. Reactions of
dinuclear compounds with thiol-dependent enzymes and/or DNA bases leading
to distortions, may be responsible for their activity. In order to develop an
understanding of the in vivo interactions these dinuclear compounds may
undergo, reactions with cellular components were studied. Presented here are
the results from the studies of dinuclear transition metal compounds with the
amino acids methionine, cysteine, and glutathione along with the nucleotides

adenosine, guanosine, and DNA.

TABLE OF CONTENTS

LIST OF TABLES ........... v

LIST OF FIGURES ..................................................................... vi

LIST OF ABBREVIATIONS ........................................................ xii

CHAPTER I. INTRODUCTION

INTRODUCTION ....................................................................... 2
Mechanism of Action of Cis-platin ........................................ 2
Dinuclear Antitumor Agents ................................................. 10

CHAPTER II. REACTIONS OF DINUCLEAR TRANSITION METAL

COMPOUNDS WITH AMINO ACIDS ............................................... 17

1. Introduction ............................................................................ 18

2. Experimental .......................................................................... 27
A. Synthesis

Preparation of Rh2(OZCCH3)4(MET)2 (1) ......................... 27

Preparation of Rh2(CYS)2(HZO)4 (2) ............................... 28

Preparation 0f Rh2(02CCH3)2(GSH)2(H20)2 (3) ................ 28

[Rh2(02CCH3)2(CH30N)5][BF4]2 + Methionine (4).............. 29

[Rn2(02CCH3)2(CH30N)5][BF4]2 + cysteine (5) ................. 29

[Rh2(02CCH3)2(CH3CN)5][BF4]2 + Glutathione (6) ............. i 30

Rh2(0200H3)4(H20)2 + O-Methyl cysteine (7) ................... 30

Rh2(02CCH3)4(H20)2 + 2-Aminothiophenol (8) .................. 30

[Rh2(DTolF)2(CH30N)5][BF4]2 +Methionine (9) .................. 31

[Rh2(OzccH3)2(CH3CN)6][BF4]2 + Cysteine (10) ............... 31

[Rh2(02CCH3)2(CH30N)5][BF4]2 + Glutathione (11) ........... 31

Re2(OchH20H3)4Cl2 + Methionine (12) ......................... 32

Re2(02CCH2CH3)4CI2 + Cysteine (13) ............................ 32

Re2(OzCCH20H3)4CI2 + Glutathione (14) ........................ 32

B. X-ray Structure Determination .......................................... 33

3. Results and Discussion ........................................................... 35

A. Dirhodium Carboxylates ................................................. 35

B. [Rh2(OchH3)2(CH30N)6]2* Reactions ............................... 48

C. Cysteine Mimic Reactions .............................................. 52

D. [Rh2(DTOIF)2(CH3CN)5][BF4]2 Reactions ............................ 57

E. Re2(OchH2CH3)4CI2 Reactions ....................................... 59

4. Conclusions ......................................................................... 61

CHAPTER III. REACTIONS OF DIRHODIUM CARBOXYLATES WITH

NUCLEOBASES ........................................................................ 63
1. Introduction .......................................................................... 64
2. Experimental ........................................................................ 70
Rh2(OchH3)4(CH3OH)2 + ATP (15) ..................................... 70
Rh2(OchH3)4(CH30H)2 + GTP (16) ..................................... 7O
Rh2(02CCH3)4(CH30H)2 + 3', 5'-TDP (17) .................. i ............ 71
Rh2(02CCH3)4(CH30H)2 4' AMP (18) ..................................... 71
Rh2(OchH3)4(CH30H)2 + GMP (19) ..................................... 72
[Rh2(02CCH3)2(CH3CN)5][BF4]2 + GMP (20) ............................ 72
[Rh2(02CCH3)2(CH30N)5][BF4]2 + AMP (21) ............................ 72
3. Results and Discussion ........................................................... 73
A. Adenosine Reactions .................................................... 73
B. Guanosine Reactions ................................................... 74
C. Thymidine .................................................................. 75
4. Conclusions ......................................................................... 76

CHAPTER IV. 1H NMR REACTIONS OF AMINO ACIDS AND DNA

PURINES WITH DIRHODIUM CARBOXYLATES .............................. 78
1. Introduction ........................................................................... 79
A. Competition Reactions ................................................... 79
B. Reactions Of Dirhodium Acetate Compounds ...................... 81
2. Experimental ......................................................................... 83
A. Competition Reactions ................................................... 83
B. Dirhodium Tetraacetate Reactions .................................... 86
3. Results and Discussion ............................................................ 90
A. Competition Reactions ................................................... ‘ 90
B. Dirhodium Tetraacetate Reactions .................................... 99
4. Conclusions .......................................................................... 116
A. Competition Reactions ................................................... 116
B. Dirhodium Tetraacetate Reactions .................................... 117

CHAPTER V. REACTIONS OF DIRHODIUM BIS-ACETATE WITH DNA 119

1. Introduction ............................................................................ 120
2. Experimental .......................................................................... 130
3. Results and Discussion ............................................................ 141
4. Conclusions ........................................................................... 148
APPENDIX ................................................................................. 149
REFERENCES ........................................................................... 176

Table 1.

Table 2.

Table 3.

Table 4.

Table 5.

Table 6.

Table 7.

Table 8.

Table 9.

LIST OF TABLES

IR stretches used in the determination of amino acid binding
SItes ...........
Proton resonances monitored during the course of the

methionine/AMP competition experiment ...........................

Proton resonances monitored throughout the course of
the glutathione/AMP competition experiment ......................

Proton resonances monitored over time for the cysteine/AMP
competition reaction ......................................................

Integration of bound:free acetate for the cysteine/dirhodium

tetraacetate reaction .......................................................

Integration of bound: free acetate for glutathione/dirhodium

tetraacetate reaction .......................................................

Integration Of bound:free acetate for the Rh2(OAc)4(ATP)2 +

cysteine reaction ............................................................

Integration of bound:free acetate for the Rh2(OAc)2(GTP)2 +
cysteine reaction ...........................................................

Integration of bound:free acetate for the reaction of dirhodium

tetraacetate with O-Methyl cysteine .................................

Table 10. DNA strands designed for crystallization studies ..................

38

92

96

99

100

102

106

108

111

142

Figure 1.

Figure 2.

Figure 3.

Figure 4.

Figure 5.

Figure 6.

Figure 7.

Figure 8.

Figure 9.

LIST OF FIGURES

The antitumor agent cis-platin ............................. _ ..........

Second generation platinum drugs (a) oxaliplatin, (b) CHIP
(iproplatin), (c) JM216, (d) diammine (1, 1-
cyclobutanedicarboxylato)platinum(lI) (carboplatin),

(e) quinoline complex, and (f) DACCP .............................

Side and top views of the three standard forms Of DNA.
B-DNA is the most common form. Major and minor grooves
are denoted by arrows .................................................

DNA base pairing model showing Watson-Crick base pairs..

Binding modes of cis-platin (a) intrastrand cross-link,
(b) interstrand cross-link, and (c) DNA-protein cross-link ......

Similarities between cis-platin and dinuclear antitumor
agents .......................................................................

Dinuclear platinum antitumor agents, (a) [{trans-
PICKNH3)2}2H2N(CH2)nNH2]2+, and (b) [{CIS-
PtCI(NH3)2}2H2N(CH2),.NH2]2* ..........................................

Dinuclear rhodium compounds that exhibit anticancer
activity (a) dirhodium (ll) carboxylates (R = CH3, CH20H3,

or CH20H2CH3), and (b) dirhodium (II) formamidinates.

(L = solvent molecule) ...................................................

Dirhenium antitumor complexes (a) rhemium (Ill) carboxylates
(R = CH3, CH20H3, or CHZCHZCHa), and (b) dirhenium bis-
carboxylates (X = halogen, L = solvent molecule) ................

Figure 10. Sulfur containing amino acids ........................................

Figure 11. Common binding modes of L-Methionine to a metal

center (a) monodentate through sulfur; and (b) bidentate
through sulfur and nitrogen forming a six membered ring.....

Figure 12. Novel cyclic trinuclear methionine compound ...................

Figure 13. Tridentate binding of methionine to a Ru center ................

vi

11

12

13

15

19

20
21

21

Figure 14.
Figure 15.

Figure 16.

Figure 17.

Figure 18.

Figure 19.

Figure 20.

Figure 21.

Figure 22.

Figure 23.

Figure 24.

Figure 25.

Figure 26.

Figure 27.

Figure 28.

Figure 29.

Postulated binding modes of Glutathine to a Pt(ll) center...
Potential binding modes Of L-Cysteine to a Pt(ll) center.....

Rhodium tetraacetate products of Methionine (a) and
Cysteine (b) .................................................. _ ...........

pKa values for the common binding Sites on (a) methionine,
(b) cysteine, and (c) glutathione ....................................

Structure of the 2:1 adduct formed in the reaction of
methionine with dirhodium tetraacetate .........................

ORTEP Of Rh2(OchH3)4(Met)2 ....................................

Possible binding modes of glutathione to dirhodium
Tetraacetate ............................................................

Possible binding modes of cysteine ..............................

UV Spectral study (a) dirhodium tetraacetate, (b) 1:1 ratio,
(0) 1:2 ratio, (d) 1:3 ratio, and (e) 1:4 ratio ......................

Possible binding modes of methionine to dirhodium
bis-acetate ..............................................................

Proposed structures for the products of the reaction
between cysteine and dirhodium bis-acetate ...................

Possible binding modes of the cysteine mimics, O-Methyl
cysteine and 2-aminothiophenol, to dirhodium tetraacetate.

(a) Proposed mechanism for the insertion reaction that
takes place with the cysteine mimics in the presence of
acetone, (b) structure of the products formed via this
mechanism ...............................................................

ORTEP of the O-Methyl cysteine product of dirhodium
tetraactate ................................................................

PLUTO Of 2-aminothiophenol derivative Of dirhodium
tetraacetate ...............................................................

Possible binding modes of methionine to the partially

solvated dirhodium ditolylformamidinate (DTOIF) complex.
(R = tolyl) ..................................................................

vii

22

23

24

39

40

42

43

45

47

49

51

53

54

55

56

58

Figure 30.

Figure 31.

Figure 32.

Figure 33.

Figure 34.

Figure 35.

Figure 36.

Figure 37.

Figure 38.

Figure 39.

Figure 40.

Figure 41.

Figure 42.

Possible binding modes of cysteine to the partially
solvated dirhodium ditolylformamidinate complex.
(R = tolyl) ............................................................... 59

Proposed structure of the product from the reaction of
glutathione with dirhenium tetrapropionate .......... . ......... 61

Binding modes of cis-platin to the DNA purines
(a) adenosine and (b) guanosine. (Where R = sugar) ...... 64

Dinuclear carboxylate compounds of (a) ruthenium,
(b) rhenium, and (c) rhodium that display antitumor
activity .................................................................. 65

Interactions Of (a) adenosine and (b) guanosine with
dirhodium tetraacetate ............................................. 66

ORTEP representation of ciS-[Rh2(DTOIF)2(9-EtAH)2(CH30N)12“
showing equatorially coordinated 9-ethylguanine ligands. 67

ORTEP representations Of Head-to-head (a) and Head-
to-tail (b) isomers of 9-Ethylguanine derivatives Of
dirhodium acetate ..................................................... 68

Binding modes of adenine and adenosine to dirhodium
forrnamidinate compounds (a) axial through N3 and (b)
through N1 and NH2. (R = tolyl and R’ = sugar) .............. 69

Equatorial bridging of AMP (a) ‘head-to-head’ and (b)
‘head-tO-tail’ isomers ................................................. 74

Possible isomers obtained in reactions of dirhodium
tetraacetate with guanosine (3) ‘head-tO-head’ and
(b) ‘head-to-tail’ ........................................................ 75

Possible product from the reaction of 3’ 5’-TDP with
dirhodium tetraacetate (a) thymidine 3’, 5’-diphosphate
(b) proposed product ................................................. 76

Intrastrand (a) , interstrand (b) and DNA-protein cross-links
(c) formed by cis-platin interactions with DNA ................. 79

Numbering scheme for the amino acids (a) methionine,
(b) cysteine, and (c) glutathione ................................... 83

viii

Figure 43.

Figure 44.

Figure 45. .

Figure 46.

Figure 47.

Figure 48.

Figure 49.

Figure 50.

Figure 51.

Figure 52.

Figure 53.

Figure 54.

Figure 55.

Figure 56.

Figure 57.

Figure 58.

Proposed product from the competition reaction of
Rh2(02CCH3)4 with methionine and AMP .....................

Proposed products formed in the competition reaction
with glutathione and AMP .........................................

Proposed products formed in the competition reaction
with cysteine and AMP ...........................................

Proposed products from the reaction of four equivalents
of cysteine with dirhodium tetraacetate ........................

Proposed products from the reaction of glutathione with
dirhodium tetraacetate ..............................................

Proposed product from the reaction of methionine with
dirhodium tetraacetate ..............................................

Proposed products from the reaction of cysteine with a
dirhodium tetraacetate/ATP derivative ..........................

Proposed products from the reaction of cysteine with a
dirhodium tetraacetate/GTP derivative ..........................

Proposed products from the reaction of cysteine with
dirhodium bis-acetate ...............................................

PrOposed products from the reaction of O-methyl cysteine
with dirhodium tetraacetate ........................................

Proposed products from the reaction of dirhodium
tetraacetate with 2-aminothiophenol .............................

Labeling scheme for 2-aminothiophenol .........................
Potentail hydrogen-bond interactions with base edges in

AT and CC base pairs. Donors and acceptors are
indicated by direction Of the arrowheads ........................

The minor groove binding drug Netropsin ........................

Crystal structure of (a) anthramycin-d(CCAACGTTGG)2

and (b) netropsin-d(CGCGAATTCGCG)2 ........................

The minor groove binding drug Anthramycin .....................

92

95

98

101

103

104

107

109

112

113

115

114

121

122

123

122

Figure 59. Structure Of cis-platin-d(CCTCTGGTCTCC)

Figure 60.

Figure 61.

Figure 62.

Figure 63.

Figure 64.

Figure 65.

(GGAGACCAGAGG) .................................................. 1 25

Crystal structure of cis-platin bound to d(CGCGAATTCGCG).
Arrows indicate the three cis-platin molecules bound to the

DNA strand .................................................................. 128
ORTEP representation of cis-[Pt(NH3)2{d(pGpG)}] ............... 129
ORTEP stereoview of the intrastrand ciS-platin d(CpGpG)

adduct ........................................................................ 129
Hanging Drop Method Of crystal growth .............................. 141
HPLC chromatograms Of (a) 10-merSC heated to 70°C for

52 h and (b) 11-mersc heated at 37°C for 64 h ................... 146
MS of TCT CTA ATC TC 11 mer ....................................... 147

A1. Methionine/AMP NMR reaction (a) methionine/AMP standard,
(b) H NMR at time= 0, (c)1 H NMR at time- - 15 min, (d)1 H NMR
at time= 2 h, (e)1 H NMR at time= 4 h, (f)1 H NMR at time= 24 h.. . 150

A2. Glutathione/AMP NMR reaction (a) glutathione/AMP standard,
(b)1 H NMR at time= 0, (c)1 H NMR at time - 15 min, (d)1 H NMR ,
at time 2 h, (e)1 H NMR at time 4 h, (f)1 H NMR at time 24 h ...... 153

A3. Cysteine/AMP NMR reaction (a) cysteine/AMP standard,
(b)1 H NMR at time= O, (c) 1H NMR at time- - 15 min,
(d)1 H NMR at time= 2 h, (e)1 H NMR at time= 4 h, (f)1 H NMR
at time= 24 h ........................................................................ 156

A4. 1H NMR reaction of Rh2(02CCH3)4 with cysteine (a) 1:1 ratio,
(b) 1:2 ratio, (c) 1:3 ratio, and (d) 1:4 ratio ..................................... 159

A5. 1H NMR reaction Of Rh2(OzCCH3)4 with glutathione (a) 1:1 ratio,
(b) 1:2 ratio, (c) 1:3 ratio, and (d) 1:4 ratio ..................................... 161

A6. 1H NMR reaction of Rh2(OchH3)4 with methionine (a) 1:1 ratio,
(b) 1:2 ratio, (0) 1:3 ratio, and (d) 1:4 ratio ..................................... 163

A7. 1H NMR reaction of Rh2(OzCCH3)4(ATP)2 with cysteine (a)
Rh2(OchH3)4(ATP)2 standard, (b) 1:1 ratio, (c) 1:2 ratio,
(d) 1:3 ratio, and (e) 1:4 ratio .................................................... 165

A8. 1H NMR reaction of Rh2(02CCH3)2(GTP)2 with cysteine (a)
Rh2(02CCH3)2(GTP)2 standard, (b) 1:1 ratio, (c) 1:2 ratio,

((1) 1:3 ratio, and (e) 1:4 ratio .................................................... 168
A9. 1H NMR reaction Of [Rh2(020CH3)2(MeCN)6][BF4]2 with cysteine
(a) 1:1 ratio, (b) 1:2 ratio, (c) 1:3 ratio, and (d) 1:4 ratio .................. 171
A10. 1H NMR reaction of Rh2(02CCH3)4 with O-methyl cysteine
(a) 1:1 ratio, (b) 1:2 ratio, (0) 1:3 ratio, and (d) 1:4 ratio ................ 173
(b)

A11. 1H NMR reaction of Rh2(OchH3)4 with 2-Aminothiophenol
(a) 1:1 ratio, (b) 1:2 ratio, (c) 1:3 ratio, and (d) 1:4 ratio ................. 175

xi

br

CITI-

°C

mol

IR

MHz

LIST OF ABBREVIATIONS

Angstrom

broad

Bohr magneton
wavenumber

degree Celcius

doublet (NMR), deuterated
parts per million (ppm)
molar extinction coefficient
wavelength

gram

mole

hour

infrared

Boltzmann constant
MegaHertz

molarity (moles per liter)

xii

mL

nm

NMR

UV

VIS

DNA

RNA

ATP

GTP

GSH
CYS
MET

9-EtGH

medium, multiplet
milliliter

bridging ligand, micro
nanometer

nuclear magnetic resonance
frequency

singlet, strong
ultraviolet

visible

weak

halide
deoxyribonucleic acid
ribonucleic acid
adenosine

guanosine

adenosine triphosphate
guanosine triphosphate
thymidine

cytidine

glutathione

cysteine

methionine

9-ethylguanine

xiii

9-EtAH 9-ethyladenine
DtOlF N, N’-p-tolylformamide
S donor solvent

L Hgand

xiv

Chapter I

Introduction

Mechanism of Action of Cis-platin

In the early 1960’s the study Of the effect Of electrical fields on Escherichia
Coli bacteria led to the discovery that cis-diamminedichloroplatinum, or cis-platin,
is a potent antitumor agent (Figure 1)1. Although the compOLInd is effective in the
treatment of lung, bladder, head and neck, cervical, testicular and ovarian
cancers, it causes severe side effects due to its extreme toxicity; these include
nephrotoxicity, myelosuppression, ototoxicity, severe nausea and vomiting,
peripheral neuropathies, and occasional cardiac abnormalities. In spite of this,
FDA approval was granted in 1979 and many cancer patients have been

receiving the drug as part of a standard chemotherapeutic regimen 2' 3.

Figure 1. The antitumor agent cis-platin.

Second generation drugs (Figure 2) are being developed with the ultimate
goal Of increased activity and reduced toxicity. Many of these potential drugs are
in clinical trials, and one Of them, carboplatin, was recently approved in the

United States for the treatment of ovarian and small lung cancers 2.

H2 HO

O O N H2 l
. t_.
/
O O \N >’N/ \Cl
H2 H2 '
OH

(a) Oxaliplatin (b) CHIP (lproplatin)
O

__<

O O
m H N O
H 3
C100, I \\\N2"’ \ /
’h. Pt"“\ /Pt \
CI/ l \NH3 H3N O O
__<0
O
(c) JM 216 (d) Carboplatin
O
Cl \ / NH3 32
Pt\ \
\ CI Pt
\ ’ °
N
0 H2 OH
0
(e) quinoline complex (f) DACCP

Figure 2. Second generation platinum drugs (a) oxaliplatin, (b) CHIP (iproplatin),
(c) JM216, (d) diammine(1,1-cyclobutanedicarboxylato)platinum(lI) (carboplatin),

(e) quinoline complex , (f) DACCP.

The mechanism of action of these platinum anticancer agents is important
to understand, as it may lead to the development Of more successful antitumor
drugs. The specific chemical reaction(s) responsible for the activity of ciS-platin
is thought to involve bifunctional attachment of the molchle to biomolecules,
which abound with possible binding Sites. The most widely studied platinum
antitumor agent, cis-platin, impairs cell division without inhibiting cell growth, as
does UV irradiation and hydroxyurea, thus the mechanism of replication inhibition
is thought to follow a Similar pathway. A culmination Of experiments performed
over the last 20 years has lead researchers to conclude that DNA synthesis is
selectively inhibited, and thus somehow responsible for the anticancer activity 3.

In spite of the large body of work on this subject, there is no direct
evidence that the actual binding of DNA by cis-platin is the cause of its
anticancer activity. Indeed some studies suggest that protein binding of cis-platin
damaged DNA may be partly responsible for its activity. Regardless of the exact
mode of action, it is irrefutable that cis-platin binds and distorts DNA. In order to
understand the mechanism of this binding reaction, the intricacies of DNA
structure must be explored.

Deoxyribonucleic acid (DNA) consists of two strands, namely a template
strand and a compliment strand. Together these strands form a double helix that
is coiled about an axis producing both major and minor grooves (Figure 3).
Repeating deoxyribose phosphodiester units comprise the backbone of each
DNA strand and serve to link together the four types of bases (Figure 4). The

DNA backbone has a polarity, the 5’ position of one pentose ring is connected to

 

Figure 3. Side and top views Of the three standard forms Of DNA.
B-DNA is the most common form. Major and minor grooves are
denoted by arrows.

’0

 

H

‘H‘N’H PL
/N N,H H-</N \N H3C ,H N/ '1'!
r< I A l y I i r
N N 1:1-H N N H N
H

OONH

Guanine Adenine Thymine Cytosine

Figure 4. DNA base pairing model Showing Watson-Crick base pairs.

the 3’ position Of the next pentose ring via 5’-3’ phosphodiester linkages. The
terminal nucleotide at one end possesses a free 5’ group, and the terminal
nucleotide on the other end possesses a free 3’ group, thus creating a type of
polarity. By convention, nucleic acid sequences are written as 5’-3’. The double
helix is stabilized by hydrogen bonds between the Watson- Crick base pairs,
adenine/thymine, and guanine/cytosine (Figure 4). Stacking interactions
between base pairs also add to the stability of the helix. DNA occurs in three
structural forms referred to as, A, B, and 2 DNA. The B form is the most stable
under physiological conditions and is the most common. A-DNA is favored in
solutions absent of water and is Shorter and has a larger diameter than B-DNA.
It is not known whether this form is actually present in cells. Z-DNA is unusual in
that it takes the form of a left handed helix unlike A- and B- DNA which are right
handed helices. Alternating C and G base pairs can readily assume the Z-DNA
structure which may play a role in the expression of certain genes ‘1.

There are a variety Of DNA binding sites for anticancer agents. The
negatively charged phosphate oxygen atoms, the phosphorous atoms, as well as
the nitrogen and oxygen atoms of the bases are all potential Sites for covalent
and noncovalent binding. An alternate method of binding is intercalation
whereby a planar molecule such as ethidium bromide, is inserted between base
pairs. Most antitumor agents bind to DNA through noncovalent interactions or
intercalation. Cis-platin is one of the few molecules to have been studied that

forms a covalent bond with DNA 5.

The DNA binding of cis-platin occurs through two guanosine or adenosine
molecules. There are three potential binding modes, namely interstrand,
intrastrand, and DNA-protein cross-links (Figure 5). The most common binding
mode is the intrastrand type. NMR and X-ray crystallography have shown that
the most common intrastrand ciS-platin cross-link is the GpG sequence. These
intrastrand cross-links alter the normal conformation of DNA by disrupting
Watson-Crick base pairing thereby causing a slight alteration in the double helix
and destacking of base pairs. lnterstrand cross-links and protein-DNA cross-

Iinks of cis-platin are rare and therefore will not be dicussed 3.

 

 

 

 

 

(a) (b) (C)

Figure 5. Binding modes Of cis-platin (a) intrastrand cross-link, (b) interstrand

cross-link, and (c) DNA-protien cross-link.

The importance of the issue Of DNA binding is that even small and

certainly large distortions to the DNA can alter protein recognition sequences.

-\v

up

’M

This, in turn, may cause replication proteins to be incapable Of recognizing their
normal binding sites. Another possibility is that structural perturbations lead to
the preferential binding of other proteins such as HMG domain proteins (High
Mobility Group proteins which may function to control gene expression and
higher-order structure, some of which are transcription factors) and histone H1.
The recognition Of cis-platin DNA adducts by HMG proteins may play a role in the
cytotoxicity of the platinum compound, indeed it has been found that HMG
proteins preferentially bind 1,2-d(GpG) Cis-platin intrastrand cross-links. Binding
of this protein spans a 14 nucleotide region with ciS-platin located in the center“.

DNA binding and unwinding are essential for the initiation Of transcription
and replication, both of which require proteins. HMG proteins can induce specific
bending in the DNA and interact with distorted DNA. The protein SRY, an HMG
transcriptional regulator, binds DNA by intercalation of its hydrophobic residues
between base pairs. This serves to unstack base pairs which leads to an
unwinding and a bending toward the major groove. These structural distortions
also serve to widen the minor groove and are similar to those caused by the
binding of cis-platin. Platinum coordination also unwinds and bends DNA,
thereby Opening and flattening the minor groove. Platination Of DNA has been
shown to block replication and transcription which is explained on the basis of
protein binding to the cis-platin binding site 6.

Other proteins besides SRY are also known to be diverted from normal
functions by cis-platin bound DNA. Many proteins in the HMG box protein family,

HMG 1 and 2, SSRP1 (human structure-specific recognition protein), lxr 1 (yeast

intrastrand cross-link recognition protein) and human UBF (rRNA transcription
factor), bind to distorted DNA, the consequence Of which is bending and
unwinding. TBP, the TATA binding protein, is another type Of protein that is
sequestered by ciS-platin bound DNA from binding at its nOrmal site. TBP is a
protein that binds the promoter region for transcription (the TATA box); without
TBP binding, transcrption and hence replication cannot occur. The histone H1
protein binds cis-platin damaged DNA more strongly than HMG; these two
proteins actually compete for binding of the cis-platin damaged DNA. Protein-
drug-DNA complexes also prevent repair of the damaged strand, for example,
HMG can inhibit nucleotide excision repair Of ciS-platin 1,2 intrastrand crosslinks.
It seems likely that the majority of cis-platin’s antitumor activity stems from

protein binding to the cis-platin damaged DNA 7'8.

Dinuclear Antitumor Agents

In the search for alternatives to cis-platin, dinuclear transition metal
complexes that may exhibit similar DNA binding modes are being explored.
These complexes are composed of two square planar metal units that are joined
by a metal-metal bond (Figure 6). Among the complexes of this structural type
found to exhibit anticancer activity are dinuclear ruthenium, rhodium, and

rhenium complexes.

10

Equatorial Sites

 

HsN/n,,,,Pt,,,tn\NHs \ s \
Cl‘ QCI L M. M
90° GIT/’9’ I I? Kaxial sites
90° with weakly
bound solvent

Figure 6. Similarities between cis-platin and dinuclear antitumor agents.

Dinuclear Platinum Compounds

The serious side effects of cis-platin led to the exploration of dinuclear
platinum complexes (Figure 7). These compounds, while sharing a similar
square planar motif, are active in cells that show resistance to cis-platin. These
dinuclear platinum compounds have been found to cause more damage than the
parent compound in lung cancer and other experimental cancer cells. Both inter-
and intra-strand complexes are formed with the dinuclear platinum but severe
DNA bending does not occur. The complexes bind at the GpG sites, and can
also span across bases without difficulty to bind GCGC Sites. Unlike the cis-

platin isomers, both the cis- and trans- forms are active 9'10.

11

CI NH3 H3N Cl H3N Cl Cl NH3

\ / \ / \ / \ /
Pt Pt Pt Pt
/ ‘ / \ / \ / \
H3N NH2(CH2)nH2N NHs H3N NH2(CH2)nH2N NH3
(8) (b)
Figure 7. Dinuclear platinum antitumor agents, (a) [{trans-

PtCKNH3)2}2H2N(CH2)nNH2]2+, and (D) [{CiS-PtCKNH3)2}2H2N(CH2)nNH2]2+.

Platinum dinuclear compounds unwind DNA in a manner analogous to cis-
platin, but do not bend DNA as cis-platin does. The bending of DNA by the
parent compound is a likely signal for HMG family protein binding. HMG proteins
recognize the DNA damage caused by the dinuclear compounds, but the
recognition is not as effective. Thus it is unlikely that HMG proteins are
responsible for the altered activity Of the platinum dinuclear compounds. These
new anticancer agents are being subjected to further research to determine what

is responsible for their increased activity in cis-platin resistant cells.

Dinuclear Rhodium Compounds

In the 1970’s dirhodium (ll) carboxylate compounds (depicted in Figure
8a) were found to exhibit considerable antitumor activity in mice against Erlich
ascites tumors and L121Otumors11. These compounds exhibit equivalent activity
and lower toxicity than that of cis-platin. The anticancer activity Of
Rh2(02CCR)4L2 (where R = CH3, CH20H3, or CH20H20H3) was studied, and the

butyrate derivative was found to be the most potent inhibitor of replication,

12

followed by dirhodium propionate and finally dirhodium acetate. DNA binding
experiments indicated that dirhodium tetracarboxylate compounds bind to double
and singe stranded DNA, polyA, polyG and ribonuclease AA. It has been
speculated that the anticancer activity of these compounds arises from DNA

13,14

binding

I I<R T T
,O’ .O /\
L-Rh-—/Rh—L L~Rh=£1—Rh—NL
O>/!O I 0&0/

R O O O O

(O/\

O =CFchO')

(a) (b)
Figure 8. Dinuclear rhodium compounds that exhibit anticancer activity (a)
dirhodium (ll) carboxylates (R = CH3, CH20H3, or CH2CHzCH3), and (b)

dirhodium (ll) formamidinates. (L = Solvent molecule).

It is known that dirhodium (ll) carboxylate compounds inhibit DNA
synthesis in vivo. A suggested mechanism of action is via enzyme binding
through the —SH groups at or near the active sites. Protein binding has been
found to deactivate the protein’s function and could be related to the observed
antitumor activity. Another possibility is the binding of an adenine residue

through axial interactions to a dirhodium tetracarboxylate molecule 12.

13

A variation on the dirhodium (ll) carboxylate compounds mentioned above
is the dirhodium (ll) formamidinate family Of compounds.
Rh2(form)2(02CCF3)2H20 (form = N, N’-di-p-tolylformamidinate, DTOIF) exhibits
the same ‘Iantern’ structure as the carboxylate derivatiVes (Figure 8b), and
readily provides binding sites for biomolecules, since both the axial H20 and
equatorial trifluoroacetate ligands are easily labilized. The dirhodium (ll)
formamidinate compounds have been tested for antitumor activity against
Yoshida ascites sarcoma, and T8 sarcoma. It was found that these compounds
exhibit higher antitumor activity and reduced toxicity as compared to dirhodium

(ll) tetracarboxylate compounds “'15.

Dinuclear Rhenium Compounds

Rhenium is notable in that it is the least toxic metallic element. There
have been no documented cases Of rhenium poisoning thus making rhenium
compounds the preferred choice for metal—based drugs such as antitumor
compounds. Dirhenium carboxylate complexes, which exhibit the same lantern
structure as the dirhodium carboxylates (Figure 9a), exhibit a bacteriostatic effect
on Escherichia COIi strain W3350 (thy). These cells become elongated as in the
case Of cis-platin treated cells. The dirhenium compounds seem to selectively

inhibit bacterial DNA synthesis with little or no effect on RNA synthesis 16.

14

 

 

I /O/I€O I /O/'I</O
L— Re Re—‘L L— Re Re’l-
0/ I /| X/ /
>T"° I X I

O O X X

(a) (b)
Figure 9. Dirhenium antitumor complexes (a) rhenium (Ill) carboxylates (R =
CH3, CH20H3, or CH20H2CH3), and (b) dirhenium bis-carboxylates (X = halogen

L = solvent molecule).

Antitumor studies were performed with the dirhenium tetracarboxylate
compounds. It was found that dirhenium propionate is the most effective against
replication, dirhenium tetraacetate is not as active and dirhenium tetrabutyrate
Shows higher toxicity without the benefit Of much greater activity. These
compounds are active against sarcoma S-180, leukemia P-388, and melanoma
B-16. Unfortunately, the compounds are susceptible to decomposition in
aqueous solutions, which requires the use of high doses to be effective. The
decomposition products are thought to be inert insoluble rhenium oxides, which
are deposited subcutaneously as a brownish material in the mice used in the
studies.

The instability of the dirhenium tetracarboxylate compounds prompted the

investigation of derivatives that would be much more stable in water. Dirhenium

15

bis-carboxylates (Figure 9b) were found to be interesting alternatives, as for
example, bis (u-propionato) diaquotetrabromodirhenium (III) which shows the
highest activity and lowest toxicity. High doses Were still needed to effect
carcinostatic activity, however, and further studies have been hampered by this

realization.

l6

Chapter II

Reactions of Dinuclear Transition Metal
Compounds with Amino Acids

l7

1. Introduction

The effectiveness of metal-based chemotherapeutics in the treatment of
cancer cannot be understood without further investigation Of chemical reactions
between metal complexes and biomolecules. In the past, Studies have focused
on the interactions of cis-platin with DNA. Since DNA is thought to be the main
target of cis-platin, much time has been devoted to studying the interactions
between this compound and nucleotides. Molecules, such as proteins, also play
a central role in modulating the activity of antitumor agents. Evidence has
indicated that reactions with thiOI-dependent enzymes are linked to cis-platin’s

I 17. Sulfur-bound Pt(ll) is still chemically reactive and

anticancer activity as wel
may be able to go on to react with DNA.

Sulfur containing amino acids and proteins are thought to serve as
possible binding Sites for molecules such as cis-platin. Platinum is a soft acid,
and sulfur is a soft base, thus a natural affinity exists. Oxygen and nitrogen are
hard bases, SO sulfur poses the largest ‘threat’ to decomposition of platinum
compounds. To understand the nature of sulfur binding, reactions of antitumor
compounds with amino acids such as cysteine and methionine, along with the
tripeptide glutathione, have been extensively studied. (Figure 10) 3'4.

This Chemistry is important, as methionine and glutathione-platinum
adducts have been found in the urine of patients undergoing treatment with cis-

18

platin An interesting consequence of protein binding is that dose limiting

nephrotoxicity is due to the binding of Pt(Il) to sulfhydryl groups in enzymes.

+ +
Mes NH3+ HS £3 HS NHCO NH3
R4,. _ : n 5 H
OOC OOC ‘QQCVNHOC .
L-Methionine L-Cystiene Glutathione (reduced form)

Figure 10. Sulfur containing amino acids.

Since sulfur containing amino acids and proteins play a vital role in the
antitumor activity of agents such as cis-platin, it is important to discuss the
biological importance Of these molecules. The main point is that their functions
can be disrupted if their concentrations are depleted by the binding of an
antitumor agent, such as cis-platin or dirhodium tetraacetate. Both methionine
and cysteine are used as precursors to the citric acid cycle, which is important in
the production of ATP. Methionine is also an important factor in the Synthesis Of
several biomolecules such as cysteine, creatine, proteins and polyamines.
Cysteine is a key building block in the synthesis Of glutathione. Glutathione acts
as [a peroxide scavenger which detoxifies the cell, and also protects red cells
from oxidative damage and helps maintain their structure. Furthermore,
glutathione acts as a redox buffer which maintains the heme in its ferrous state.
A decrease in glutathione concentration can lead tO hemolysis, or hemolytic
anemia in which the cells are depleted of oxygen ‘1. The disruption of normal cell
function and the inhibition of DNA synthesis are an important part Of the

antitumor activity of transition metal complexes.

19

Metabolites Of cis-platin include platinum complexed to both methionine

and glutathione. As mentioned earlier, platinum-methionine adducts were found

18. L-Methionine

in the urine of patients undergoing treatment with ciS-platin
contains three potential binding sites, namely a carboxylate, an amine, and a
thioether group. The usual modes of binding are monodentate through the sulfur
(Figure 11a), bidentate through the sulfur and the nitrogen (Figure 11b), and at

low pH’S (_<. 0.5) bidentate through the sulfur and oxygen 192°.

/—\ f‘
H C _
NH NH 3. /Pt\ Cl
C \.< LAW C'\ 8 °'
/ P
H T ”1131 Cl/ \
2 CH3 N COOH

(a) (b)
Figure 11. Common binding modes of L-Methionine to a metal—center. (a)
monodentate through sulfur; and (b) bidentate through sulfur and nitrogen

forming a six membered ring.

The crystal structure of a novel cyclic trinuclear structure, [Pt3(L-
Met)3]-H20, was recently obtained by Davidson, et al. (Figure 12) 18. Crystal
structures or [Pt(L-Met)Cl2], [Pt(Gly-L-Met)Cl], [Pt(L-H3MetO)CI2], and [Pt(L-
MetO)2] have also been obtained. (L-MetO = L-Methionine S-Oxide). An example

of tridentate binding of L-Methionine to a metal center has been observed in the

20

complex [(n5-Cp*)Ru(L-Met)]. a ruthenocene radiopharmaceutical compound that

coordinates methionine through the nitrogen, sulfur and oxygen 22 (Figure 13).

/Pt Pt
° .1 l
0 \Pt/
I \
S O 0
CH3

 

Figure 12. Novel cyclic trinuclear methionine compound.

Figure 13. Tridentate binding of Methionine to a Ru center.

Glutathione plays a complex role in the metabolism of metal-based drugs.
This tripeptide is the single most important source of intracellular non-protein
sulfl'Iydryls, with concentrations ranging from 1-10 mM depending on the cell type
20

Levels of glutathione present in the cell have important effects on the

metabolism Of cis-platin. Depletion of glutathione levels results in the increased

21

toxicity Of kidney cells by the platinum drug. Pre-treatment of patients
undergoing cis-platin therapy with glutathione results in reduced renal toxicity
without depleting antitumor activity. It was found that cis-platin resistant tumor
cells have increased levels Of glutathione, and that depletion of the tripeptide
levels can reverse this resistance. Intracellular glutathione appears to be
involved in attenuating cis-platin induced nephrotoxicity, and pre-treatment with
glutathione may be useful in protecting the kidneys from ciS-platin toxicity. With
six potential metal binding sites (sulfur, nitrogen from the glutamate amino, two
carboxyl oxygens, and two peptide nitrogens) a number Of adducts are possible,

including bridging structures through the sulfur atom 2023”” (Figure 14).

NH3

H N S HO O
3 \P/ H o m S
/ K N\/ILOH HOW/\N \ / O
H3N N O 0 O H P\ ”J
W s/ N N 0”
O NH OH O O
3 OWOH

NH3
S O
R>P< )NWOH
R N H NH2
0
0¥
OH

Figure 14. Postulated binding modes of Glutathione to a Pt(ll) center.

22

In attempts to model protein binding Of the sulfhydryl group to platinum(ll),
studies were carried outwith the amino acid cysteine. Platination Of proteins can
occur through cysteine residues. It is thought to bind through the amino nitrogen
and the sulfhydryl group to form a N, S- metallacycle. This type of complex has
been found to be more thermodynamically stable than a dimetallic bis(u-thiolato)

complex. Possible chelating structures are depicted in Figure 15 21128.

'OOC

Me3P S H2 900 :2 S
\P/ \P/5 t \P/ :L
/ K H2N/ \S S/ Kuz COO'

Me3P N COOH
H2 >—/

‘OOC +
+I'I3N H3N
COO COO
H N 8 NH MeaP S PMe3
3¥/\/3 \<><
/ ‘\ /
H3N \S/P NH3 M9313 S PMe3
K/coo' R/COO'
NH 3* NH3"

Figure 15. Potential binding modes of L-Cysteine to a Pt(ll) center.

Elucidation of the mechanism by which activity occurs is fundamental in

the study of any potential antitumor drug. Reactivity studies of the antitumor

active compound, dirhodium tetraacetate, Rh2(OZCCH3)4-2HZO, with amino acids

23

 

 

have not been studied in detail, and the results that have been reported are not
particularly compelling. In early studies with methionine and cysteine, it was
thought that methionine reacted by replacing the axial water molecules (Figure
16a). Cysteine, on the other hand, was postulated to reSult in cleavage of the
29—31

dirhodium carboxylate core to yield a paramagnetic bis-cysteine complex

(Figure 16b).

HO O O O
% IL/Iéfls/ NH2}_D;.1:1h\"/(\SR11N/
O OH O

(a) (b)

Figure 16. Rhodium tetraacetate products of Methionine (a) and Cysteine (b).

The destruction of the rhodium-rhodium bond was postulated based on
the results from experiments aimed at understanding the metabolism Of the
antitumor compound. In a study involving mice implanted with Erlich ascites
tumors, breakdown of the carboxylate cage was determined by the amount of
14002 exhaled by the mice. Data Obtained from injecting mice with 14C labeled
dirhodium tetraacetate indicated that rhodium and 14C rapidly disappear during
the first hour after dose administration and continue to decline for 6 h. It was
reported that levels of dirhodium(ll) acetate as measured by rhodium content

were higher than levels Obtained from 1‘10 content. The conclusions drawn from

24

these results was that the dirhodium carboxylate cage was breaking down, since
the 14C label was being removed more quickly than the rhodium. It was
reasoned that decomposition of dirhodium tetraacetate was generating acetate
ions, which would be measured by the amount of 14CO; exhaled. The rhodium
would form a nonabsorbable metabolite, thus escaping detection. Based on this
study dirhodium(ll) tetraacetate was claimed to react in vivo with cysteine to form
an unstable rhodium complex which breaks down yielding acetate ions, H1, and
an insoluble precipitate consisting of rhodium(lll) and cysteine 33. The reaction of
dirhodium tetraacetate with proteins containing an -SH moiety are thought to
follow the same general pathway. It is important to point out, however, that the
researchers failed to recognize that if the dirhodium tetraacetate led to production
of a bis-acetate analog, Rh2(OchH3)2, and 2 02CCH3', the same results would
be obtained from their experiment. We maintain that the evidence Cited from this
experiment does not irrefutably prove the breakdown of the cage structure.
Reaction of dirhodium tetraacetate with enzymes containing -SH groups
lead to inactivation Of the enzyme if the -SH group is at or near an active site.
The reaction of cysteine with the compound is dramatically different than any
other amino acid. This difference stems from the lack of reversible binding which
is Observed for the other amino acids. The conclusion that cysteine disrupts the
carboxylate cage was demonstrated by 1H NMR spectroscopy at pH 7.5.
Addition of cysteine to dirhodium(ll) acetate causes a shift in the acetate peak
from bound acetate to free acetate. Integration indicates that the reaction is

complete at a 1:4 ratio of dirhodium(ll) acetate to cysteine. The hypothesis

25

posed by the researchers is that enzyme inhibition is due to initial interactions
between one or more —SH groups and dirhodium(ll) carboxylate, followed by
cage breakdown and tight binding Of rhodium(ll) to the active site of the
enzyme”. The tentative conclusion was that the antitumor activity of
dirhodium(ll) carboxylates is due to the binding Of enzymes or proteins containing
-SH groups.

The -SH containing tripeptide, glutathione, was used for detoxification in
P388 ascites tumors. An increased dose Of glutathione was injected after
administration Of a toxic dose Of dirhodium(ll) tetrapropionate. This experiment
resulted in detoxification of the metal drug along with an enhancement of
antitumor activity. It was suggested that decomposition products could be the
reason for this increased activity 13. It is necessary to point out, however, that
other possibilities exist, such as the formation of a bis-acetate adduct, or, as
observed in the case of the platinum drugs, an amino acid-metaI-DNA cross-link.

Due to the lack of solid evidence for the identity of products in
dirhodium(ll) tetraacetate reactions with amino acids and glutathione, it was
deemed necessary to undertake detailed investigations Of this chemistry to aid in
elucidation of the antitumor action of the dirhodium carcinostatic compounds.
The promising anticancer activity of other dimetal carboxylate compounds, such
as the dirhodium formamidinate and dirhenium carboxylates prompted the
investigation of these compounds as well. Each reaction was performed in bulk

as well as in the NMR tube, which were monitored by 1H NMR spectroscopy

26

(Chapter 4). The products were characterized by UV-VIS, IR, and NMR

spectroscopies, mass spectrometry and elemental analysis when possible.

2. Experimental
A. Synthesis

All reactions were carried out under inert conditions via standard Schlenk line
techniques. The dirhodium complexes, Rh2(OzccH3)2-L2 and
[Rh2(OchH3)2(CH3CN)3][BF4]2 were prepared by literature procedures. The
amino acids methionine, glutathione, and cysteine as well as the cysteine
mimics, O-Methyl Cysteine and 2-Aminithiophenol, were purchased from Sigma
and Aldrich and used without further purification.
Preparation of Rh2(02CCH3)4(MET)2 (1)

TO a solution of Rh2(OchH3)4~2MeOH (0.100 g, 0.2 mmol) in 50 mL DD

H20, methionine (0.0596 g, 0.4 mmol) was added, which led to a color change
from blue-green to dark violet. The reaction was stirred for 24 h at room
temperature after which time the volume was reduced to ~ 10 mL on a
rotoevaporator and excess acetone was added to precipitate the violet
compound. IR (KBr) v cm'1: 1588 (NH31), 1433,1600(COO‘). 1H NMR (020) 8
ppm: 1.71 (s, bound acetate), 1.83 (CH3), 2.36 (YCHz), 2.94 (NH2), 3.79 (aCH).
UV-vis spectroscopy (H20): Amax = 562 nm (e = 245 Lmol’1cm'1). Elemental
analysis calc’d for Rh2013012N282H34: C, 26.99; H, 4.25; N, 3,93. Found: C,

26.88; H, 4.29; N, 3,67. Electrospray MS: m/z+ 740 [2 Met + 2 Rh + 4 acetate]",

27

681 [2 Met + 2Rh + 3 acetate]*, 621 [2 Met + 2Rh + 2 acetate]*, 592 [1 Met +
2Rh + 4 acetate]*.
Preparation of Rh2(CYS)4(H2O)4 (2)

To a solution of Rh2(O2CCH3)4-2MeOH (0.100 g, 0.2 mmol) in 10 mL of
DDH2O, was added 0.8 mmol cysteine, which led to the immediate formation Of a
rust-brown solution. The solution was stirred for 2 h at room temperature and the
volume was reduced to 5 mL on a rotoevaporator. The rust colored product was
precipitated by the addition Of excess acetone. IR (KBr) v cm'1: 1615, 1196
(NH2), 1624, 1390 (C00). 1H NMR (D20) 8 ppm: 3.24 (m, BCH2), 3.95 (ctCH),.
UV-vis spectroscopy (H2O): Ame. = 586 nm (a = 70 Lmol'1cm'1). Elemental
Analysis calc’d for Rh2C12010N2S2H14: C, 18.91; H, 4.20; N, 7.30. Found: C,
18.95; H, 3.80; N, 7.36. Electrospray MS: m/z+ 718 [2Rh + 4 cys + 2 H2O - 8H],
599 [2Rh + 3 cys + 2 H2O — 6H], 480 [2Rh + 2 cys + 2 H2O - 4H], 342.93 [2Rh +
1 cys + 2 H2O - 2H], 241 [Rh + cys + H2O — H].

Preparation of Rh2(O2CCH3)2(GSH)2(H2O)2 (3)

Glutathione (0.246 g, 0.8 mmol) was added to a solution of
Rh2(O2CCH3)4-2MeOH (0.100 g, 0.2 mmol) in 10-15 mL Of DD H2O. After 5
minutes the solution became dark green, and 20 minutes later it became brown
with the appearance of a red crystalline precipitate. The solution was stirred
overnight at room temperature with no further color change. The precipitate was
removed by filtration, and a yellow-orange solid was Obtained from the filtrate by

the addition of excess acetone. IR (KBr) v cm’1: 3067, 1610, 1234 (NH2), 1540

(NH31), 1648, 1412 (C00). 1H NMR (D20) 8 ppm: 1.74 (bound acetate), 2.00

28

(CH2) 2.37, 3.15 (CH2), 3.63 (CH), 3.77 (CH2),4.58 (otCH). UV-vis spectroscopy:
max = 562 nm (e = 62 LmOl'1cm‘1). Elemental Analysis calc’d for
Rh2C24013N682H24: C, 29.56; H, 4.50; N, 8.60. Found: C, 29.48; H, 4.51; N, 8.86.
Electrospray MS: m/z" 1054 [2 gsh + 2 Rh + 2 acetate + 51H2O —4H], 884 [2 gsh
+ 2 Rh + acetate], 825 [2 gsh + 2 Rh -— 4H].

[Rh2(02CCH3)2(CH3CN)3][BF4]2 + Methionine (4)

Methionine (0.226 g, 0.10 mmol) was added to a solution of dirhodium bis-
acetate (0.050 g, 0.050 mmol) in 10 mL of DD H2O at 25 °C. Upon addition, the
color changed to a reddish-orange. The reaction stirred overnight after which the
volume was reduced to 5 mL and acetone was added to precipitate the red
product. IR (KBr) v cm'1: 3053, 1452 (NH31), 1620, 1410 (C00). UV-vis
spectroscopy: max = 529 nm (e = 154 Lmol'1cm’1). 1H NMR 8 ppm: 2.06
(acetate), 2.10 (BCH2), 2.47 (CH3), 2.49 (yCH2), 3.70 (t, ctCH). Elemental
Analysis cale’d for Rh2C14013N2S2H31: C, 22.8; H, 4.1; N, 4.7. Found: C, 21.7;
H, 4.4; N, 4.73.

[Rh2(O2CCH3)2(CH3CN)3][BF4]2 + Cysteine (5)

Cysteine (0.0398 g, 0.20 mmol) was added to a solution of dirhodium bis-
acetate (0.050 g, 0.050 mmol) in 10 mL of DD H2O at 25 °C. The color
immediately changed to dark red and then brown within a few minutes. The
reaction was stirred for 2 h, reduced to ~5 mL and treated with acetone to
precipitate a dark yellow solid. IR (KBr) v cm'1: 3075, 1539 (NH31), 1649, 1410
(C00). UV-vis spectroscopy: kmax = 517 nm (e = 130 Lmol'1cm'1). 1H NMR 5

ppm: 2.05 (S, acetate), 3.15-2.76 (BCH2), 3.63 (otCH) 3.77, (s, unk). Elemental

29

Analysis calcd for Rh2C16012N4S4H23: C, 24.8; H, 3.4; N, 7.2. Found: C, 24.0; H,
4.1; N, 6.9.
[Rh2(02CCH3)2(CH3CN)3][BF4]2 + Glutathione (6)

Glutathione (0.0967 g, 0.20 mmol) was added to a Solution of rhodium bis-
acetate (0.050 g, 0.050 mmol) in 10 mL Of DD H2O at 25 °C. Upon mixing, the
color changed immediately to a dark red-orange. The reaction was stirred
overnight, the volume was reduced, and acetone was used to precipitate a brown
solid. IR (KBr) v cm'1: 3100, 1586 (NH3*), 1623, 1490 (000'). UV-vis
spectroscopy (H20): 1...... = 515 nm (e = 457 Lmol'1cm'1). 1H NMR (500 MHz,
D20) 5 ppm: 2.02 (acetate), 2.14 (CH2), 2.51 (CH2), 2.90-3.22 (BCH2), 3.88 (CH),
3.94 (CH2).

Rh2(O2CCH3)4 + O-MeCYS (7)

O-Methyl cysteine (0.1357 g, 0.79 mmol) was added to a solution of
Rh2(O2CCH3)4 (100 mg, 0.197 mmol) in 20 mL of DD H2O under N2. Immediately
upon addition Of the O-methyl cysteine, the solution became yellow; the color
gradually changed to orange, and finally became a dark reddish orange. The
solution was stirred for 2 h at room temperature, reduced in volume and treated
with an excess of acetone to precipitate the product. UV-Vis spectroscopy (H2O):
max = 525 nm. 1H NMR (D20) 8 ppm: 1.82 (s, acetate), 2.15 (acetone adduct),
3.33 (aCH), 3.81 (CH3), 4.50 (8CH2).

Rh2(O2CCH3)4 + 2-Aminothiophenol (8)
A solution Of Rh2(O2CCH3)4.2H2O (100 mg, 0.197 mmol) in 15 mL CH3OH,

was treated with 0.08 mL of 2-aminothiophenol (0.79 mmol) under an N2

30

atmosphere. Immediately upon addition, the solution became a reddish-orange.
The reaction was stirred for 2 h at room temperature, the volume was reduced,
and excess acetone was added to precipitate the product. Uv-vis spectroscopy:
1...... = 537 nm. 1H NMR (CD3OD), 8 ppm: 1.85 (s, acetate), 6.89 (t, phenyl), 7.09
(d, phenyi), 7.53 (d, phenyi), 7.65 (t, phenyl).

[Rh2(DToIF)2(CH3CN)3][BF4]2 + Methionine (9)

A solution of [Rh2(DTolF)2(CH3CN)3][BF4]2 (0.100 g, 0.10 mmol) in 15 mL of
CH3OH was added to a solution of methionine (0.0274 g, 0.2 mmol) in 5 mL of
DD H2O. The resulting solution gradually changed from green to dark grey. EPR
spectroscopy Shows a broad signal indicative of one unpaired electron g = 1.87.
[Rh2(DTolF)2(CH3CN)5][BF4]2 + Cysteine (10)

A solution of [Rh2(DTOIF)2(CH3CN)5][BF4]2 (.1009, 0.1 mmol) in 15 mL of
CH3OH was added to a solution of cysteine (0.446 g, 0.2 mmol) in 15 mL of DD
H2O. The reaction was stirred overnight during which time, the solution became
a yellow-orange color. The solution was filtered under N2 and excess solvent
reduced in vacuo. IR (KBr) v cm'1: 1580 (N-C-N), 3025, 1052 (NH31), 1622, 1390
(000'). 1H NMR (CD3OD) 8 ppm: 2.03 (CH3CN), 2.45 (CH3CN), 2.95 (BCH2),
3.30 (CH3), 3.63 (t, aCH), 7.27 (tolyl), 7.32 (tolyl).

[Rh2(DTolF)2(CH3CN)5][BF4]2 + Glutathione (11)

A solution of [Rh2(DTOIF)2(CH30N)6][BF4]2 (0.050 g, 0.05 mmol) in 15 mL

of CH3OH was added to a solution of glutathione (0.57 g, 0.2 mmol) in 15 mL Of

DD H2O. Upon mixing, the solution became Olive green and was allowed to stir

31

overnight. The brown precipitate that formed was collected and washed with
CH3OH. IR (KBr) v cm'1: 1583 (N-C-N), 3070, 1504 (NH31), 1616, 1405 (000').
Re2(O2CCH2CH3)4CI2 + Methionine (12)

A solution of Re2(O2CCH2CH3)4CI2 (0.025 g, 0.35 mmol) in 2.5 mL of
CH3OH was added to a solution of methionine (0.1012 g, 0.7 mmol) in 2 mL of
DD H2O. The two solutions were mixed and an additional 2 mL of DD H2O was
added to increase the solubility Of methionine. The color Changed from orange to
bright green and then gradually darkened. The reaction was stirred overnight,
after which time a brown solid was collected and washed with DD H2O.
Re2(O2CCH2CH3)4CI2 + Cysteine (13) .

A solution of Re2(O2CCH2CH3)4Cl2 (0.025 g, 0.35 mmol) in 2.5 mL of
CH3OH was added to a solution of cysteine (0.1644 g, 1.4 mmol) in 2 mL of
CH3OH and 0.5 mL Of DD H2O. Upon mixing, the solution turned green, but
within a few seconds the solution became orange, and finally dark brown. The
reaction was stirred for 3 h, after which time a brown precipitate was removed by
filtration and washed with H2O.

Re2(O2CCH2CH3)4CI2 + Glutathione (14)

A solution of Re2(O2CCH2CH3)4Cl2 (0.025 g, 0.35 mmol) in 2.5 mL of
CH3OH was added to a solution of glutathione (0.4174 g, 1.4 mmol) in 3.5 mL of
CH3OH and 0.5 mL Of DD H2O. The resulting mixture became a cloudy green
color, then progressed to a light brown color which darkened and took on a
reddish hue. After a few minutes, the solution turned dark purple and a

precipitate was Observed to form. The reaction was stirred for 4 h, after which

32

time a purple solid was collected. 1H NMR (D20) 8 ppm: 2.01 (CH2), 2.38 (CH2),
3.14-2.76 (8 CH2), 3.66 (a CH), 4.59 (CH2). 13C NMR (D20) 5 ppm: 171.64,
170.46, 169.30 (carboxylate), 50.57 (or CH), 49.37 (CH), 38.50 (CH2), 35.43
(CH2), 28.05 ([3 CH2), 22.24 (CH2). Elemental 1 Analysis calc’d for
ReC20012N2S2H30: C, 33.8; N, 11.9; H, 4.7. Found: C, 33.4; N, 12.08; H, 5.50.
B. Xray Structure Determination

A hemisphere of data was collected for compounds (1) and ome, a sphere
of data was collected for athph on a Siemens SMART diffractometer at —100 i
2°C with graphite monochromated Mo K01 radiation (A... = 0.71073 A) and were
corrected for Lorentz and polarization effects. The frames were integrated with
the Siemens SAINT software package, and the data was corrected for absorption
using the SADABS program. The structures were solved using the SHELXS
program and refinements were carried out by full matrix least-squares
calculations on F2 using the SHELXL-97 program.

Crystals Of Rh2(O2CCH3)4(MET)2 were Obtained from the Slow diffusion of
a dirhodium tetraacetate solution in acetone into a solution of methionine in
water. Several days later, a violet rectangular platelet Of dimensions 0.57 x 0.13
x 0.16 mm was obtained. A total Of 1321 frames were collected at —100 i 2 °C
with a scan width of 0.3 ° in (I) and an exposure time of 30 sec/frame from which
45 reflections were selected for indexing. The reflections were refined to give
unit cell parameters for an orthorhombic cell. The data collected were integrated
using an orthorhombic cell to yield a total of 3983 unique reflections. After data

reduction, 2445 unique reflections remained. The Rim and Rsigma were 0.1198

33

and 0.1738 respectively. The positions of the rhodium atoms were located by
direct methods. The remaining non-hydrogen atoms were located through
successive cycles Of least-squares refinements and difference Fourier maps. All
non-hydrogen atoms were refined anisotropically. The’final full matrix least
squares reveal maximum and minimum peak heights of 4.86 and —3.06,
respectively with residuals of R1 = 0.1777 and wR2 = 0.4222 for 140 parameters
and 1 restraints. The final goodness of fit was 2.609.

Crystals of the dirhodium tetraacetate O-methyl cysteine derivative (7)
were obtained from the slow diffusion of a dirhodium tetraacetate solution in
acetone into a solution of O-methy; cysteine in water. Several days later, a red
rectangle of dimensions 0.18 x 0.05 x 0.08 mm was obtained. A total Of 1321
frames were collected at —100 i 2 °C with a scan width of 0.3 ° in c0 and an
exposure time of 30 sec/frame from which 57 reflections were selected for
indexing. The reflections were refined to give unit cell parameters for a
tetragonal cell. The data collected were integrated using a tetragonal cell to yield
a total Of 5165 unique reflections. After data reduction, 2727 unique reflections
remained. The Rim and Rstgma were 0.1087 and 0.3332 respectively. The
positions of the rhodium atoms were located by direct methods. The remaining
non-hydrogen atoms were located through successive cycles (of least-squares
refinements and difference Fourier maps. All non-hydrogen atoms were refined
anisotropically. The final full matrix least squares reveal maximum and minimum

peak heights of 0.72 and -0.84, respectively with residuals of R1 = 0.0509 and

34

wR2 = 0.1412 for 127 parameters and 3 restraints. The final goodness Of fit was
0.578.

Crystals Of the dirhodium tetraacetate 2-aminothiophenol derivative (8)
were obtained from the slow diffusion of a dirhodium tetraacetate solution in
acetone into a solution of 2-aminothiophenol in methanol. Several days later, a
red rectangle of dimensions 0.518 x 0.104 x 0.181 mm was Obtained. A total of
6345 frames were collected at —100 i 2 °C with a scan width of 0.3 ° in (D and an
exposure time Of 30 sec/frame from which 63 reflections were selected for
indexing. The reflections were refined to give unit cell parameters for a triclinic
cell. The data collected were integrated using a triclinic cell to yield a total Of
16335 unique reflections. After data reduction, 7810 unique reflections
remained. The Rim and Rsigma were 0.0657 and 0.1511 respectively. The
positions of the rhodium atoms were located by direct methods. The remaining
non-hydrogen atoms were located through successive cycles Of least-squares
refinements and difference Fourier maps. All non-hydrogen atoms were refined
anisotropically. The final full matrix least squares reveal maximum and minimum
peak heights of 5.72 and —4.21, respectively with residuals of R1 = 0.3532 and

wR2 = 0.5761 for 1137 parameters and 2 restraints.

3. Results and Discussion
A. Dirhodium Tetraacetate Reactions
The reactions Of dirhodium carboxylate compounds with different types of

axial donors lead to color changes that are useful for the identification of

35

derivative compounds. With oxygen donors such as water or methanol, the
complex takes on a blue or green color. The axial binding of nitrogen or sulfur
donors causes the compound to be red to orange in color, whereas with
phosphorous donors, a red to yellow color ensues. There are two transitions that
are important for these compounds, one at ~600 nm and one at ~450 nm. The
600 nm band is a transition assigned to Rh-Rhivr - Rh-Rh 0*. This band is most
sensitive to changes in the axial ligands. It remains essentially constant for
oxygen donors and becomes blue—shifted with nitrogen or sulfur donors. The
band at 450 nm is due to more than one transition, one Of which is the Rh-Rh m
- 0 0* transition. Obviously, it is sensitive to changes in the equatorial ligands,
but it shows little or no change with different axial donors 33. The color of the
dirhodium carboxylate derivatives, which is due to the visible transition at ~600
nm is a convenient characterization aid when amino acids are used as ligands.
Infrared spectroscopy is also a useful tool for the charaCterization Of
dirhodium tetraacetato-amino acid complexes. The amino acids contain three
groups that can be used as probes to determine the binding site(s) Of the amino
acid to the dirhodium compound (Table 1)“. Uncoordinated amino groups
exhibit peaks in the range of 1585-1683 cm'1 and 1492-1534 cm'1, whereas
coordinated amino groups are shifted to 1610-1620 cm'1 and 3090-3340 cm'1.
This indicates that the amino group is participating in bonding with the metal.
The S-C stretch is present around 2500 cm'1, which is dramatically shifted to
higher frequencies upon metal binding to the sulfur. The carboxylate group Shifts

are also useful for determining whether or not the carboxylate group is

36

participating in binding. For the amino acid complexes with dirhodium
tetraacetate, however, the carboxylate modes from the dirhodium compound
(1589 and 1440 cm'1) and the amino acid carboxylate peaks (ranging from 1597-
1610 cm'1 and 1410-1415 cm'1) overlap with each other, thus making it difficult to
ascertain if carboxylate binding has occurred.

In order to fully characterize these compounds, numerous factors must be
taken into account. The electronic spectral properties are important, but they do
not always give the entire picture. Along with these tools, 1H NMR spectroscopy
is used to account for peak shifting near the proposed binding site. Elemental
analyses and mass spectrometry were performed in some cases to help
determine the nature of the product formed. When possible, X-ray analyses
were used to determine the structures of the compounds.

Rh2(O2CCH3)4 + Methionine (1)

The compound formed in this reaction is thought to be a simple 1:2 axial
adduct Of Rh2(O2CCH3)4 29. An instantaneous reaction takes place upon addition
Of the amino acid to dirhodium tetraacetate, accompanied by a color change from
blue-green to violet. This color change indicates that either a nitrogen or sulfur
donor is bound in the axial position. The electronic spectral data are in accord
with axial binding of a nitrogen or sulfur donor as the visible transition is blue-
Shifted from Ame. = 585 for dirhodium tetraacetate 33. Based on the pKa’s for the
possible binding sites of the amino acid (Figure 17), the most likely donor is the
nitrogen, however, the sulfur atom has the highest affinity toward the rhodium

atom .

37

Table 1. IR stretches used in the determination Of amino acid binding sites.

 

 

 

 

Functional Group IR Stretch (cm'1 )
NH2
Coordinated 3230-3340
3090-3280
1610-1620
1 023-1060
Uncoordinated 1 585-1683
1492-1 534
1 1 10-1237
COO‘
Coordinated 1589-1643
1 347-141 1
Uncoordinated 1597-1610
1410-1415
Tetraacetate 1 589
1440

38

(a) MeS @1132 (b) HS H3
_ § H \14
OOC -005 H
213 1.92

2.12
(c) 8-75 M000"
HS NHCO NH3“
\4 9.65
3.59 g H
'ooc NHOC
\/

Figure 17. pKa values for the common binding sites on (a) methionine, (b)

cysteine, and (c) glutathione.

The infrared spectra of the compound Shows that the amino acid is
coordinated through only one functional group. A stretch at 1588 cm'1 indicates
the presence of an uncoordinated amino group. With this binding site ruled out,
the sulfur and carboxylate groups are the best prospects for binding. The SC
stretch has shifted, an indication that the sulfur is involved in binding.
Carboxylate peaks occur at 1433 and 1600 cm'1, which are located at the
positions for uncoordinated carboxylates. These data point to a Simple axial
adduct bound through the sulfur atom.

The characterization of dirhodium-amino acid derivatives by 1H NMR

spectroscopy is useful for predicting the binding site of the amino acid. The

39

amino acid groups nearest the dirhodium binding should exhibit the greatest shift
in the spectrum. Methionine peaks indicate a change in the electronic
environment near the sulfur atom. The protons next to the sulfur, yCH2 and CH3,
Show the greatest downfield shift. These data indicates that the sulfur, rather
than the amino or carboxylate groups, participates in binding to the metal.

Once the binding site is determined, the next step is to identify the exact
structure of the compound formed in the reaction. Elemental analysis and
electrospray mass spectrometry were used to confirm the formulation as a 2:1
axial adduct. The electrospray MS data Show a peak at m/z+ 740, which
corresponds to two rhodium atoms, two methionine molecules, and four acetate
ions. Further support is lent by the elemental data, which corresponds to the

formula Rh2(O2CCH3)4(Met)2 (Figure 18).

+10% 0A0
Im/ / s/ NH2
I
I
O

HZN /SO;

——Rh—-
/ \42
2"0 I
V0 O 0”
Figure 18. Structure Of the 2:1 adduct formed in the reaction Of methionine with

dirhodium tetraacetate.

An ORTEP representation Of Rh2(O2CCH3)4(Met)2 which crystallizes in the

orthorhombic space group P21221 is depicted in Figure 19. The structure depicts

40

a dirhodium tetraacetate core with two axially bound methionine ligands. The
rhodium-rhodium bond length is 2.44 Which is longer/ than the corresponding
dirhodium tetraacetate bis water adduct which has a rhodium-rhodium bond
length of 2.386(1) A. This structure was not completely sOlved due to a twinning
problem. Looking at the systematic absences, it is apparent that the space group
is P21221, however, further refinement in a twinning program is needed to fully
solve the structure. The final R was 17.7%.

Rh2(O2CCH3)4 + Glutathione (3)

If —SH containing biomolecules destroy the dinuclear unit Of Rh2(O2CCH3)4
as has been suggested in the early literature, then glutathione should be capable
of effecting this type of decomposition as well. Upon reaction of glutathione with
dirhodium tetraacetate, a yellow—orange compound is formed. The yellow-orange
color of this compound indicates a possible sulfur or nitrogen binding to the metal
center. Ultraviolet-visible spectroscopy revealed the presence Of aband at 582
nm and a shoulder at ~440 nm. This suggests that the nature Of the equatorial
ligand has changed while the axial ligand has remained the same.

Infrared analysis shows that the glutathione molecule is coordinating
through more than one binding site. The spectra reveal the presence of both
coordinated (3067 and 1610 cm'1) and uncoordinated (1540 cm'1) amino groups.
The S-C stretch has shifted such that it overlaps with another peak, indicating
that sulfur is involved in binding as well. Carboxylate peaks appear at 1648 and
1412 cm‘1, not clearly indicating the presence of carboxylate binding, or lack

thereof. It appears that both amino groups and the sulfur are involved in binding

41

v

3‘

I

i

27.. w

85341100632. Lo 55:828. DEC .2 2:9".

.23..

 

42

to the metal center. The most probable binding modes are indicated in Figure

 

 

20.
0A0 O/‘\O 0A0 A0
I ,O’I‘p I ,O’Ifo |,0’l‘.o |,0’|‘,o
L;R Rh-L L;R ,Rh—L L;Rh—Rh-L L-;Rh-——Rh-L
NI,S/i SIN; MIN/l le’l
N N K K, KN
\_/

(C)
/s /s H 0
Where \N] = \N "JO“

Figure 20. Possible binding modes of glutathione to dirhodium tetraacetate.

1H NMR data indicate shifts in the proton resonances closest to the
sulfhydryl group. The yCH2 peak shifts by the greatest amount and splits into a
quartet of doublets. This indicates a change in the environment of the protons
nearest the sulfur atom. The data point to a type Of bis-chelating/bidentate
interaction between glutathione and dirhodium tetraacetate. The presence of an
electronic transition at 582 nm suggests that the Rh-Rh bond is still intact,
however further analysis is needed to fully understand the nature of binding that
had occurred.

Elemental analysis and mass spectrometry were performed in order to aid
in the identification of the new compound. Electrospray MS shows the presence

Of a peak at 1054 m/z“, which corresponds to two rhodium atoms, two

43

glutathione molecules, two acetate ions, five water molecules and four H' ions.
Elemental analysis further supports the MS data, and leads to a formulation of
Rh2(O2CCH3)2(GSh)2(H2O)4. These findings are not in accordance with the
accepted notion that all —SH containing biomolecules destroy the dirhodium unit.
As can be ascertained from the above data, the rhodium-rhodium bond is indeed

intact, with two acetates being substituted by two glutathione molecules.

Rh2(O2CCH3)4 + Cysteine (2)

ln attempts to debunk early reports that cysteine, possessing the -SH
moiety, leads to mononuclear compounds 11'12'29'32, the reaction with dirhodium
tetraacetate was performed. This reaction is not as biologically relevant as the
glutathione reaction, since both the amino and carboxylate groups are free
instead Of bound, as would be the case in a protein. The formation of a rust
colored product does not fit into the color scheme for a simple aXiaI donor of
nitrogen or sulfur based on earlier electronic spectral reports, leaving Open the
possible equatorial binding mode of either a nitrogen or sulfur donor (Figure 21).
Electronic specral data revealed a transition at Amax = 586 nm, suggesting that

the dirhodium bond is still intact and the axial donor is an oxygen atom.

44

Nb Is
A :.. U
2 3
SI IrN

/N
S/
NRI%IN
é . U
N S
/
SI%I#-S
/N
S
NIB IS
, _/ u
N S
SIV%IHIN
/N
S/
SlxI IS
A , _/. U
N
NI/%Ing

(d)

(C)

(b)

(a)

 

 

(f)

(e)

an... m
fire/4.
2 SI%IS
mH/f_/o@m.
O

2
w
H SS
OH/ /_ H
2 OI%IO O
m
H/ s M.
H
m oImIo o
CA
SS
N
2
H

Figure 21. Possible binding modes of cysteine.

45

In an attempt to understand the nature of binding in the compound, an
electronic spectral study was undertaken (Figure 22). Up to eight equivalents Of
cysteine were added to a solution of dirhoidum tetraacetate. The presence of the
band at ~580-600 was unaltered, leading one to conclude. that the dinuclear core
was still intact. With the addition Of each successive equivalent, the ~450 nm
transition became more blue shifted until it became a shoulder on another peak.
These results suggest that the equatorial ligands are being replaced by cysteine.

Infrared data reveal the presence Of two types Of binding for cysteine.
Amino peaks appear in the spectra at 1615 and 1196 cm'1, which could indicate
either coordinated or uncoordinated amino groups. The S-C stretch has shifted
into a neighboring peak. As usual, the carboxylate modes do not clearly indicate
the presence or absence of binding. Stretches at 1624 and 1390 cm’1 could
indicate that the carboxylate group is coordinated, but, overlap of peaks in this
region renders a definitive assignment difficult. What is clear, hoWever, is that
the sulfhydryl group is involved in some type Of binding to the metal center.

Due to the difficulty in deciding the nature Of this compound, elemental
analysis and electrospray mass specrtometry were performed. Elemental data
support the formulation Rh2(Cys)4(H2O)2. Electrspray MS results Show a peak at
m/z+ 718 that can be assigned to two rhodium atoms, four cysteine molecules,
two water molecules and eight H' ions. These data indicate that cysteine does

not destroy the dinuclear unit, but that it replaces all four acetate ligands.

46

.238 E 5 cam
.25 3 § .er 3 8V 6:9 E § 6588.63 E2855 E 3628. .6868 >2 Na 2:9“.

 

 

 

 

 

 

 

 

 

 

 

 

 

.... l... 8...
/ 88
(A ..
E
8.:
.. ...» ...“... .. .. ...“...
/\/ .. \ ...
av Q
83 8.:
8o 03 Sue...

 

 

 

 

 

 

3v /\/\. E

g— 08.-

 

 

47

B. [Rh2(O2CCH3)2(CH3CN)3]2* Reactions

Since the dirhodium compounds previously studied in our labs reveal that
DNA purines are capable of replacing two bridging carboxylate ligands 36, the
dirhodium bis-acetate reactions were explored with. amino acids. The
replacement Of two acetate groups signifies that the active form Of the compound
may be the bis-acetate adduct. If the -SH containing amino acids are going to
destroy the carboxylate cage, the breakdown should occur more quickly with two
of the four acetates already released.

[Rh2(O2CCH3)2(CH3CN)3]2* + Methionine (4)

Methionine was added to a solution of dirhodium bis-acetate in a 2:1 ratio,
as in the case for the tetraacetate reaction. We anticipated that the product
would not be a simple axial adduct, since solvated equatorial binding sites are
available in addition to the axial positions. The product resulting from this
reaction is red in color. The electronic absorption data Show a band at 529 nm,
which represents a shift of ~38 nm from the parent bis-acetate compound.

Infrared data reveal more about the nature of the product. The presence
of uncoordinated amino groups at 1452 cm'1 lead to the conclusion that the
amino group does not participate in binding. Carboxylate peaks at 1620 and
1410 cm'1 are not particularly helpful about lending insight into binding modes,
but they are in the range for coordinated carboxylate groups. The parent
compound, dirhodium bis-acetate, ShOWS an intense stretch due to the [BF4]'

anion, which is absent in the methionine complex. This would suggest the

48

formation of a neutral compound, since the counterions are no longer present.
The net charge on the methionine ligands would then be 1'.

1H NMR spectroscopy measured on the dirhodium bis-acetate products do
not aid in the determination of the binding site(s) of the methionine ligand. Each
of the proton resonances exhibits a shift as compared to the free amino acid. It
would be expected that the CH3 and yCH2 would shift the most, however, since
the amino acid is presumably binding through two functional groups this would
result in shifting Of the aCH and BCH2 resonances as well.

Due to the lack of hard evidence for the exact nature of this product,
elemental analysis was obtained to determine the number of methionine ligands
bound to the bis-acetate core. The results are in accord with the formula,
Rh2(O2CCH3)2(Met)2(H2O)2. Based on the pKa values (Figure 17) for the free
amino acid, the most likely binding sites are through the nitrogen and sulfur
atoms, but IR data does not support this conclusion. It appears that the

carboxylate oxygen and sulfur atoms participate in binding instead (Figure 23).

 

I /O’”I§O I /O/ 60
H20-Rh—Rh-H2 2O—Rh Rh—H20
,s/ / l ,S/ I \S/ I

O O O
Q/(HOL/S\ (X0

OH OH
”2'" HZN ”211 HZN 01"

Figure 23. Possible binding modes of methionine to dirhodium bis-acetate.

49

[Rh2(O2CCH3)2(CH3CN)3]2’ + Cysteine (5)

If cysteine destroys the dirhodium unit, it should break down the bis-
acetate cage as well. The product from this reaction is dark yellow, a slightly
different color than the dirhodium tetraacetate product. The visible spectrum
shows a band at 517 nm, shifted from the 567 nm band of the parent compound.

IR data Show the usual shifting of the S-C stretch, indicating that sulfur is
binding to the metal center. Amino stretches appear at 3975 and 1539 cm'1
indicating possible coordination to the metal. Carboxylate peaks are at 1649 and
1410 cm'1, indicating that this functional group is a possible binding site as well.
As was the case for the methionine derivative, the [BF4]' anions are absent in the
cysteine product, leading to the conclusion that the cysteine groups have a net
charge of 1'.

1H NMR spectroscopy was performed in order to determine the binding
modes of the amino acid. Both the otCH and BCH2 resonances shift upon
coordination to the bis-acetate compound. As was the case for the dirhodium
tetraacetate cysteine derivative, the BCH2 protons are present in a quartet of
doublets which appears to be a common occurrence in NMR spectra of
dirhodium complexes bound to both glutathione and cysteine.

Elemental analysis was performed to determine the number of cysteine
molecules bound to the dirhodium core. The data indicate that the likely formula
for the product of this reaction is Rh2(O2CCH3)2(CyS)2 (Figure 24). It appears
that the reaction Of cysteine with dirhodium bis-acetate does not break the

dinuclear core as was reported for the corresponding tetraacetate reaction.

50

O O O O
S Rh/ Rh/ES S Rh/O /OS
8/ / I 0/

Figure 24. Proposed structures for the products Of the reaction between

cysteine and dirhodium bis-acetate (S = solvent).

[Rh2(O2CCH3)2(CH3CN)3]2* + Glutathione (6)

Glutathione was added to rhodium tetraacetate in a 4:1 ratio, and the
resulting solid was brownish-red, unlike the product Obtained from dirhodium
tetraacetate. Since glutathione does not destroy the metal-metal bond in
dirhodium tetraacetate reactions, it seemed unlikely that it would cause the bis-
acetate to break down. A transition at 515 nm is Observed in the visible
spectrum, which is Shifted from 567 nm of the bis-acetate precursor complex.

Infrared analysis Show sulfur binding has occurred according to the Shift of
the S-C stretch. Amino stretches appear at 3100 and 1586 cm'1, indicating the
possibility of coordination. Once again, the stretches of the carboxylate groups
at 1623 and 1490 cm'1, do not provide definitive evidence for or against

coordination. Unlike the methionine and cysteine cases, the B-F stretch in the

51

[BFij anion is present at 1085 cm'1. This indicates that the glutathione
molecules do not neutralize the charge on the rhodium atoms.

The binding sites appear to be through the sulfur and the amino group,
which would give a five membered ring. By 1H NMR spectral analysis, the peak
shifts correspond to the proposed product (Figure 20). The proton resonances
shift most for the gsh1 and gsh2 peaks. The quartet of doublets appears as
expected and is assigned to gsh3.

C. Cysteine Mimic Reactions

AS a result of the controversy surrounding the stability of the dirhodium
tetraacetate core in the presence Of —SH containing molecules, cysteine mimics
were explored. These molecues, O-methylcysteine and 2-aminothiophenol, were
chosen because they contain the SH-C-C-NH2 moiety. Since the mimics contain
the same reactive Sites as cysteine, the reactions should proceed in a similar
manner. These molecules are soluble in a variety Of organic solvents, therefore
the chemistry is not limited to reactions in water.

The reactions appeared to follow that of cysteine, with the solutions
becoming dark upon addition of the ligand. The compounds are nearly the same
color as the cysteine derivatives, but the similarities end there. Electronic
spectral analyses of the compounds reveal transitions at 525 nm for the O-
MeCys product and 537 nm for the 2-AThPh compound. 1H NMR data show the
presence of bound acetate as well as aromatic peaks for the phenol ring of 2-
AThPh. The O-MeCys product exhibits an acetate peak and the O-MeCys

peaks. The presence of bound acetate in the NMR spectra suggest that these

52

mimics do not behave in the same manner as cysteine with dirhodium
tetraacetate. This could be due to a chelation through the O and S functional
groups, rather than the N and 8 functional groups as previously thought for

cysteine (Figure 25).

 

O O O O 0 0< O O<
’O/f I’O/f L L FIIh/O/Ffli/OL L II't’h/O’Rh’ L
L;Rh——Rh—L L-Rh—Rh— ; — _ _ _
/ I / / ’ /
N l ,S l S I .N N I N I N i s I
N N Ks S K KN
L] L]

Figure 25. Possible binding modes of the cysteine mimics, O-Methyl cyseine

and 2-aminothiophenol, to dirhodium tetraacetate. N-C-C-S = cysteine mimic.

Compound (7) crystallizes in the tetragonal space group P4(3) with the
asymmetric unit consisting of a dirhodium tetraacetate unit and One O-methyl
cysteine ligand (Figure 27). The rhodium-rhodium bond length is 2.41(3) A which
is longer than the corresponding parent compound bond length, this could be due
to stronger interactions of the axial sulfur and nitrogen donors which are
coordinated at Rh-N distance Of 240(2) A and a Rh-S distance of 253(5) A.

A PLUTO representation of compound (8) which crystallizes in the triclinic
space group P-1 is presented in Figure 28. The structure consists of two
dirhodium tetraacetate units linked together by a 2-aminothiophenol ligand

through the sulfur and the nitrogen atoms. From the unit cell, it is apparent that

53

:NH SH _> :N’ ‘S; I M‘N S
\_/
O O
O O
/ /

(b)

O< O O

/ /

I /O I /O / \ I /O /O
Rh——-Rh—N.~ ,-S-;Rh-——Rh

O/I O/I :x: O I O/l

>1” >T" 0

Figure 26. (a) Proposed mechanism for the insertion reaction that takes place
with the cysteine mimics in the presence Of acetone, (b) structure Of the products

formed via this mechanism.

54

 

55

Figure 27. ORTEP of the O-methyl cysteine product of dirhodium tetraacetate.

998352 EEBOEE .6 c>zm>tcu BaccaoEBEEm-

 

N Lo 05E .8 659“.

56

the structure is a twin therefore, additional twinning parameters will need

to be used in order to fully solve this structure.

D. [Rh2(DTOIF)2(CH3CN)3][BF4]2 Reactions

The partially solvated complex, [Rh2(DTolF)2(CH3CN)6][BF4]2 was used in
place Of its parent complex, Rh2(DTOIF)2(O2CCF3)2. Since the
ditolylformamidinate trifluoroacetate compound has been found to lose two
triflouroacetate groups, it seemed logical to begin with the solvated compound
that would be present in vivo. Naturally, this connotes that the CH3CN lignads
will be replaced by H2O in aqueous media.

[Rh2(DTolF)2(CH3CN)3][BF4]2 + Methionine (9)

Two equivalents of methionine were added to the partially solvated
dirhodium bis-formamidinate compound. The solution changed from green to
dark grey. This grey color for dirhodium complexes is typically an indication of a
paramagnetic species. EPR spectroscopy was performed, and indeed a broad
Signal at g = 1.87 was found. Elemental analysis was performed on the product

and the results point to the formula Rh2(DTOIF)2(Met)2(H2O)2 (Figure 29).

57

 

 

~ /-\ ,R I 'N/l-\ 4R

HO rlth/N Flth/NHO HO Rh/ Rh/NHO

2 " '— 2 2 " _ 2
,s/ O ,S/| \S/

O O
S O
QHOIL/ \ CK M
HN OH OH

2 NH2 H2N H2N OH

Figure 29. Possible binding modes of methionine to the partially solvated

dirhodium ditolylformamidinate (DTOIF) complex. (R = tolyl)

[Rh2(DTolF)2(CH3CN)3][BF4]2 + Cysteine (10)

Four equivalents Of cysteine were added to the partially solvated
dirhodium formamidinate compound. The color of the compound is yellow-
orange, which is similar to the colors Of the other cysteine products. Infrared
spectroscopy shows the absence of an S-C stretch, as is expeCted for sulfur
binding. The N-C-N bend is Observed at 1580 cm'1, which indicates that the
bridging ligands are still in place. Uncoordinated amino peaks appear at 1052
cm'1. Carboxylate peaks appear at 1622 and 1390 cm'1, indicating that the
carboxylate groups are coordinated. 1H NMR spectroscopy shows shifts in both
the otCH and BCH2 proton resonances, as is expected upon cysteine

coordination to a dirhodium center. Proposed products are depicted in Figure 30.

58

 

 

‘N R N’ ‘N R N
HO Rh/ Rh/H HO Rh/ ran/Ho
2 - "’ 2 2 - — 2
S/ S/ O/l S/
O l O HO/L/S ( O
H2N OHHzN on H2N HZN OH

Figure 30. Possible binding modes of cysteine to the partially solvated

dirhodium ditolylformamidinate complex. (R = tolyl)

[Rh2(DTOIF)2(CH3CN)5][BF4]2 + Glutathione (11)

Four equivalents of glutathione were added to the partially solvated
dirhodium bis-formamidinate compound. The brown compound precipitated from
solution as an insoluble product. Infrared data show that the sulfur is
participating in binding to the metal center. The presence of a bend at 1583 cm'1
due to N-C-N illustrates the rhodium-rhodium bond is still intact. Amino peaks at
1504 cm‘1 correspond to uncoordinated amino stretches. Carboxylate peaks at
1616 and 1405 cm’1 may correspond to a coordinated carboxylate group. These
derivatives appear to resemble the proposed products of

[Rh2(DTOIF)2(CH3CN)5][BF4]2 with cysteine in Figure 30.
E. R82(02CCH2CH3)4C|2 Reactions

Due to the anticancer activity and low toxicity of the dirhenium

tetracarboxylate compounds”, reactions of amino acids with

59

Re2(02CCH20H3)4Cl2 were explored. Since this compound is known to
eventually decompose in aqueous solutions, the reactions were performed in
methanol with only 5% water to aid the solubility of the amino acids.
Re2(OchH2CH3)4Cl2 + Methionine (12) and Re2(02CCH2CH3)4CI2 + Cysteine
(13)

These reactions undergo many color changes until they reach a final
brown color with the deposition of brown precipitates. The precipitates are
unidentifiable by IR spectroscopy and elemental analyses, but are most likely
some form of rhenium oxide decomposition products, as was suggested in the
mouse studies.

Re2(OchH2CH3)4CI2 + Glutathione (14)

Four equivalents of glutathione were added to a solution of dirhenium
tetrapropionate. The color changes are essentially the same as for the
methionine and cysteine complexes, except that the color progresses from brown
to a purplish brown color with a purple solid being obtained. This solid which is
highly water soluble is thought to be Re(Gsh)2 by elemental analysis. Both 1H
and 130 NMR spectroscopy were used to discern the nature of ligand binding.
Shifts occur for the BCH2 and the aCH peaks along with a CH2 peak indicating
that glutathione is coordinated through the sulfur and an amino group (Figure

31).

6O

HO
#0
NHz H0 N\ /s o o
HOMN‘CS/Rg )uWOH
0 0 N o .2
.4

OH

Figure 31. Proposed structure of the product from the reaction of glutathione

with dirhenium tetrapropionate.

4. Conclusions

The reactions of sulfur containing amino acids with dirhodium tetraacetate
produced interesting results. The methionine reaction proceeded according to
the earlier literature proposal of axial binding to the dirhodium core. The —SH
containing reactions, however, do not proceed with breakdown of the dinuclear
unit as was postulated in the early literature. The reaction of dirhodium
tetraacetate with glutathione leads to the displacement of two acetate ligands by
two glutathione molecules. Obviously the dinuclear cage is still intact. It would
appear from these results that not all —SH containing biomolecules serve to
destroy the dinuclear unit. The cysteine reaction shows the displacement of all
four acetates by four cysteine molecules, but according to the electronic spectral
data, the rhodium-rhodium bond is still intact. Conclusions in the early literature
about the stability of the dirhodium carboxylate core are false, as can be

concluded from the evidence of the NMR studies presented in Chapter 4.

61

Dirhodium bis-acetate was used in reactions with the sulfur containing
amino acids to determine if the amino acids would break down the core in the
absence of two bridging ligands. Instead of a simple axial adduct, the reaction
with methionine leads to a bis-chelating product. From "IR spectral analysis, the
binding sites appear to be the oxygen and sulfur atoms. The cysteine derivative
also appears to contain two anionic chelating cysteines that bind through the
oxygen and sulfur atoms as well. Glutathione, which chelates through the
nitrogen and sulfur atoms, is neutral as verified by the presence of [BF4]’
stretches in the infrared.

Reactions of the antitumor active dirhodium formamidinate compound
were explored with the sulfur containing amino acids. Methionine forms a
paramagnetic compound with the partially solvated ditolylformamidinate
compound, with evidence pointing towards one unpaired electron. Cysteine
appears to form an oxygen-sulfur chelate with the formamidinate compound.
Glutathione also forms a bis-chelating interaction, most likely through the oxygen
and sulfur as well.

The reactions with dirhenium tetrapropionate result in different types of
products. Cysteine and methionine derivatives both result in the formation of a
brown precipitate that is most likely the decomposition product mentioned in the
mice studies. The glutathione product, however, is a soluble purple compound

that consists of two glutathione ligands chelated to one rhenium atom.

62

Chapter III

Reactions of Dirhodium Carboxylates with Nucleobases

63

1. Introduction

Studies involving the carcinostatic activity of transition metal antitumor
agents point to the conclusion that this behavior is a consequence of the
interactions of these compounds with DNA. The antitumor agent cis-platin is
known to react with both guanine and adenine residues (Figure 32) on DNA by
forming 1, 2-intrastrand cross-links. These conformations cause the DNA to
bend and distort, possibly leading to the sequestering of proteins to this site.
Ultimately this could be responsible for the inhibition of replication. It is not
known whether the actual DNA binding event is the cause of antitumor activity,

however, or if the activity stems from the binding of proteins to the damaged

Site34,36,37
Cl Cl
|
N HN N
N \
k I \> \ ' >
\N R H2N N a

(a) (b)
Figure 32. Binding modes of cis-platin to the DNA purines (a) adenosine and (b)

guanosine. Where R = sugar.

The search for antitumor agents that are active against cis-platin resistant

cancers has led to the investigation of dimetal carboxylate compounds as

alternatives. Among these dimetal carboxylates are compounds containing the

64

transition metals ruthenium, rhenium, and rhodium (Figure 33). The cause for
the carcinostatic activity displayed by the dinuclear compounds may be a result
of enzyme deactivation, replication inhibition from nucleotide binding (thus
depleting the substrate pool), binding to a single stranded DNA template, or a

combination of these events 38““.

O O<R o o R o o R
Cl—Ru—Ru-CI H o—Re’—-—Re’—0H L—Rh’—Rh—L
/ / 2 / / 2 / I /
R O O X X R O O

Y Y

R R
(a) (b) (C)
Figure 33. Dinuclear carboxylate compounds of (a) ruthenium, (b) rhenium, and

(c) rhodium that display antitumor activity.

Antitumor agents such as the dirhodium carboxylate complexes exhibit
anticancer activity that is thought to be a result of binding to the DNA template.
Early studies involving the binding of dirhodium carboxylates to nucleosides and
nucleotides indicated that the compounds are reactive with poly A sequences,
but not poly G 38. Reactions of dirhodium tetraacetate with DNA purines revealed
that adenine binds axially through N7, with the exocyclic amino group
participating in hydrogen binding with a carboxylate oxygen (Figure 34a). Steric

repulsions between the guanine 06 and the carboxylate oxygen, however, would

65

be expected to prevent axial binding of this nucleotide (Figure 34b). It has also

been noted that adenine is able to interact axially through it bonding

interactions‘z'".
H- bond
Her/H \ N H c://l\o
H8
O< SUGAR
(a) _|Rh,/°/lo 51./N“
SUGAR/N/\/N O/ /| /
H8 O O IN
4
V . \ ...
H2N N repulsron
b } /
( ) N/ —-0. /'\
.——— b 0< H8
/
.N___Rh/O Rh/__ NAN—esueAR
SUGAR’NV o/l o/
.. T J. --
Y "o— /N
NJ

Figure 34. Interactions of (a) adenosine and (b) guanosine with dirhodium

tetraacetate.

Contrary to earlier literature reports that suggested only axial binding was
possible 5"53, studies in our laboratories show that dirhodium carboxylate and
dirhodium formamidinate react with purines via substitution of two equatorial

40,48-50

carboxylate ligands 9-ethlyadenine reacts with

66

[Rh2(DToIF)2(MeCN)6][BF4]2 to form a bidentate bridge across the dimetal center
through the N7 and exocyclic NH2 sites (Figure 35) 5°. The 9-ethylguanine
molecule reacts with [Rh2(DTolF)2(0200F3)2(MeCN)2][BF4]2 by replacing the two
trifluoroacetate ligands to form a bidentate bridge through the N7 and 06 atoms.
The 9-ethylguanine ligand can bind in a ‘head-to-head’ or ‘head-to-tail’ fashion as
indicated in Figure 36 4°. This same type of isomer formation was observed in
reactions of dirhodium compounds with 9-ethyladenine. Since DNA bases are
arranged in a ‘head-to-head’ fashion, these are the best models for the

mechanism of action on the template strand.

/‘ _ ‘0
3" (i) S
“kW-Kw . ,
s v i
3"
N(4
’\
S09
Q)

 

Figure 35. ORTEP representation of cis-[Rh2(DTo|F)2(9-EtAH)2(CH3CN)]2+

showing equatorially coordinated 9-ethylguanine ligands.

67

(a) 4.»

 

C(16)J .1 ‘ /|R\
4‘
0(16’ ’4’ Mom)” ~ ~.-\ (I) (|)<R
‘ /Rh’-9-;Rh’
' Rh(2) O I
‘Qa‘ "‘7’ ~ 0
P: '3 3 S / \N
"3'5. 1‘ ‘6 ,
‘7 Vs. .
H3CHZC N=(
I
:7“ 5 I} NH2
\.
- ,._ o?
\ 67
(b) ,, . R
- 03 -. '
‘ 01' A ‘ ' ‘04 ’ . O R
'----'~ g I /I<
th-Q—Rh‘

Figure 36. ORTEP representations of Head-to-head (a) and Head-to-tail (b)
isomers of 9-Ethylguanine derivatives of dirhodium acetate.

68

The antitumor activity of the dimetal compound
[Rh2(DtolF)2(OchF3)2(MeCN)2][BF4]2 is a result of the open axial positions, as
well as the easily labilized trifluoroacetate ligands. The lability of these groups
allows for mono and bidentate neutral ligands to bind, e.g. 9-ethylguanine and 9-
ethyladenine. Studies performed with adenine and adenosine show that two
different types of binding can occur. Adenine was found to coordinate axially
through the N3 atom (Figure 37a), whereas adenosine was observed to
coordinate equatorially through the N1 and exocyclic NH2 groups (Figure 37b).
Coordination of adenosine in this fashion leads to a mixed valence product. The
blue color of the complex indicates the presence of Rh“, the result of a single
electron oxidation. These findings suggest that the adenosine ligands are acting
as bidentate mono-anionic ligands. In reactions of the
[Rh2(DTolF)2(OZCCF3)2(MeCN)2][BF4]2 complex, the antitumor activity may be a

result of axial coordination, equatorial coordination, or merely redox processes 15.

 

 

R R
R\ /\ /R \N/R\N/
N R. N-\ R n N/ \N’R CF
I/N /N /\N Rh Rh/—O’C 3
Rh Flh—N \ /| / “D
FaC>’|——’O NH2 HN N
O D \ /
\/ HN E
CFa VN \ N
N
\ NR.

(3) (b)
Figure 37. Binding modes of adenine and adenosine to dirhodium

formamidinate compounds (3) axial through N3 and (b) equatorially through N1

and NH2. (R = tolyl and R’ = sugar).

69

In light of these findings, the reactions of dirhodium tetraacetate and the
dirhodium bis-acetate derivative with different nucleobases were explored.
Adenosine and guanosine mono- and tri-phosphates were reacted with both
dinuclear carboxylate compounds in order to confirm the early literature reports
about the reactivity of these compounds with nucleobases. The reaction of 3,5-

thymidine diphosphate was also explored.

2. Experimental

Rh2(02CCH3)4-2CH30H + ATP (15)

Adenosine triphosphate (0.108 g, 0.1 mmol) was added to a solution of
Rh2(OchH3)4-2CH30H (0.0509, 0.1 mmol) in 20 mL of DDHZO. The solution
immediately changed from blue-green to purple. After stirring for 24 h, the
solution was concentrated on a rotoevaporator, and a purple solid was obtained
by the addition of excess ethanol. 1H NMR (D20) 8 ppm: 1.78 (acetate), 4-5
(sugar), 6.5 (H1’), 8.45 (H2), 8.75 (H8). 3‘P NMR (Na3P04 reference) 8 ppm: -
10.8 (d), -11.3 (d), -23.0 (t).

Rh2(02CCH3)4-2CH30H + GTP (16)

Guanosine triphosphate (0.0519, 0.1 mmol) was added to a solution of
Rh2(OchH3)4-ZCH30H (0.0509, 0.1 mmol) in 15 mL of DDH20. The solution
was stirred for 2 days with no noticeable color change. The mixture was then
heated to 65°C and stirred overnight which led to the production of an emerald
green solution. The solution was concentrated on a rotoevaporator and excess

ethanol was added to precipitate a green solid. 1H NMR (D20) 8 ppm: 1.8

70

(acetate), 3.5-5 (sugar), 5.7 (H1’), 8.6 (H8). 31P NMR (Na3PO4 reference) 8 ppm:
-9.1, 9.8, -10.3.

Rh2(02CCH3)4-2CH30H + 3’,5’-TDP (17)

Thymidine diphosphate (0.026 g, 0.05 mmol) was added to a solution of
Rh2(02CCH3)4-ZCH30H in 10 mL of DDH20. The reaction mixture was stirred for
several days without noticeable color change. The solution was then heated to
65°C and stirred for four days. After this time the solution had become green.
The solution was concentrated on a rotoevaporator and excess ethanol was
added to precipitate a yellow solid. 1H NMR (020) 8 ppm: 1.79 (CH3), 3.6-5
(sugar), 6.12 (NH), 7.65 (H4). 31P NMR (NagPO4 reference) 8 ppm: 0.510, 0.390.
Rh2(02CCH3)4-2CH3OH + AMP (18)

Adenosine monophosphate (0.1449, 0.395 mmol) was added to a solution of
Rh2(0200H3)4-20H30H (100 mg, 0.197 mmol) in 10 mL of NaHzPO4 buffer at a
pH of 7.0. Immediately upon addition of the nucleotide the solution changed from
blue green to purple. The solution was stirred at 37°C for 24 h and a purple solid
was obtained from the addition of excess methanol. 1H NMR (D20) 8 ppm: 1.86
(acetate), 3.5-5 (sugar), 6.02 (H1’), 8.55 (H2), 8.69 (H8). Elemental Analysis
calculated for Rh2024026N10P2H41. Calc: C, 25.4; N, 9.3; H, 4.9. Found: C, 24.6;
N, 8.5; H, 4.0.

Rh2(02CCH3)4-ZCH30H + GMP (19)

Guanosine monophosphate (0.1604 g, 0.395 mmol) was added to a solution of
Rh2(02CCH3)4-2CH30H (100 mg, 0.197 mmol) in 10 mL of NaH2P04 buffer at a

pH of 7.0. The reaction was heated to 60°C for 24 h, after which time the

71

solution was green. A green solid was obtained by the addition of excess
methanol. 1H NMR (020) 8 ppm: 1.79 (acetate), 3.5-5 (sugar), 5.75 (H1'), 8.64
(H8).
[Rh2(OchH3)2(CH3CN)5][BF4]2 + GMP (20)
Guanosine monophosphate (0.12319, 0.3 mmol) was added to a solution of
[Rh2(OchH3)2(CH30N)6][BF4]2 (100 mg, 0.15 mmol) in 10 mL of DDH20.
Immediately upon addition, the solution became a pinkish-red. The solution was
evaporated to yield a purplish solid. . 1H NMR (D20) 8 ppm: 1.933 (acetate), 3.8-
4.2 (sugar), 5.78 (H1’), 8.041 (H8). 31P NMR (Na3P04 reference) 8 ppm: -7.8.
Elemental analysis calculated for Rh2014024N11BzF8PH27. Calc: C, 21.6; N, 10.3;
H, 3.7. Found: C, 22.6; N, 9.7; H, 3.9.
[Rh2(02CCH3)2(CH3CN)5][BF4]2 + AMP (21)
Adenosine monophosphate (0.1096 9, 0.3 mmol) was added to a solution of
[Rh2(02CCH3)2(CH3CN)5][BF4]2 (100 mg, 0.15 mmol) in 10 mL of DDH20. The
solution changed from purple to pink upon addition of the nucleotide. 1H NMR
(D20) 8 ppm: 1.963 (acetate), 3.5-4.5 (sugar), 6.042 (H1’), 8.245 (H2), 8.470
(H8). 31P NMR (Na3PO4 reference) 8 ppm: -15.2. Elemental analysis calculated
for Rh2022N1oC2282F8P2H28. Calc: C, 21.6; N, 11.4; H, 3.2. Found: C, 22.2; N,
11.1; H, 3.5.
3. Results and Discussion
A. Adenosine reactions

Reactions of dirhodium tetraacetate with AMP and ATP lead to products

with two types of ligands. 1H and 31P NMR spectroscopy were used to

72

characterize the nature of the products. In the 1H NMR spectra, certain peaks
were observed to shift downfield from the free nucleotide resonances. The H8
proton in the product shifts by 0.5 ppm, the H2 shifts by 0.3 ppm, and the H1’
shifts by 0.36 ppm. 31P NMR data show only slight chemical shifts, with two of
the phosphorus peaks having shifted by ~0.27 ppm, and the third having
changed position by 0.5 ppm. Preliminary studies indicate that two adenosine
ligands are bound axially to the dirhodium tetraacetate core (Figure 34a) as was
suggested in the early literature.

Reactions of the partially solvated complex
[Rh2(020CH3)2(CH30N)6][BF4]2 led to different results. Upon addition of two
equivalents of AMP, the solution becomes red, instead of the purple color
obtained with the dirhodium tetraacetate. The 1H NMR spectrum shows the
same general pattern as the dirhodium tetraacetate derivative, i.e. the H8 and H2
protons shift by 0.27 and 0.15 ppm respectively. 31P NMR spectroscopy shows a
negligible shift in the phosphorous resonance. These data taken together with
elemental analysis results lead to the conclusion that the adenine ligands are

equatorially bound to the dirhodium core (Figure 38).

73

 

 

O O
/O /O / /
OHz—Rh Rh—Hzo 0H2;Rh / Rh — H20
m /| ~ ~|
N
z N N
I N N H
H \ H8 \7 H8
/ /
N \ N \
\
H2\N SUGAR H2 N SUGAR

(a) (b)
Figure 38. Equatorial bridging of AMP (a) ‘head-to-head’ and (b) ‘head-to-tail’

isomers.

B. Guanosine Reactions

Despite early literature reports, we have found that dirhodium tetraacetate does
indeed react with guanine bases. Upon heating the reactions to 60°C, the
solution becomes green, indicating that a reaction has taken place. 1H NMR
spectroscopy shows that two guanosine ligands have substituted for two bridging
acetate ligands. The H8 proton shifts by 0.75 ppm indicating that binding at the
N7 site has occurred. 31P NMR spectroscopy shows minimal shifts for each of
the three phosphate resonances. The same type of reaction occurs with the
dirhodium bis-acetate to give what we formulate as

[Rh2(OchH3)2(GMP)2(H20)2][BF4]2 (Figure 39).

74

O O 0 0
/ /O OH Rh/ Rh/
OH§;Th /Rh~H20 6/ I / H20
, .o , I , |
N 0 N 0
F 6
RN / NH RN / NH
N4 N/
NH2 NH2

Figure 39. Possible isomers obtained in reactions of dirhodium tetra-and bis-

acetate with guanosine (a) ‘head-to—head’ and (b) ‘head-to-tail’.

C. Thymidine Reaction

The reaction of thymidine with dirhodium tetraacetate was not expected to occur,
since the pyrimidine nucleobases are not as basic as purines. Upon heating the
reaction for 2 weeks at 60°C, however, a change in color ensues. A yellow solid
was obtained from the green solution by precipitation with acetone. The 1H NMR
spectrum shows minimal shifts for the proton resonances, but there is a shift of a
sugar resonance. Only a slight shift of 0.05 ppm occurs for the remaining sugar
protons. According to integration, there are two TDP ligands and two acetate
ligands on the dirhodium tetraacetate core. The nature of this binding is
uncertain, as this particular reaction has not been performed with dirhodium
tetraacetate or any of the other carcinostatic transition metal complexes under

investigation. 31P NMR spectroscopy shows a shift for the phosphorous

75

resonances of 0.39 and 1.24 ppm. From the data it appears that TDP has

replaced two equatorial acetate ligands and is binding through the phpsphorous

atoms (Figure 40).
0A0
o I ——-'
CH OH Rh/O Rh/O
- — —H
HN 3 3/ / 20
(I) o
O
N O C/o\ ,/o \P/
’/P\ \
o o /0

(a) (b)
Figure 40. Possible product from the reaction of 3’, 5’ TDP 'with dirhodium

tetraacetate (a) thymidine 3’,5’-diphosphate (b) proposed product.

4. Conclusions

Despite early literature reports on the reactivity of the purine nucleobases with
dirhodium carboxylates that indicated only axial adenine binding was possible,
we have found unprecedented binding modes for these ligands in our
laboratories. Adenine and its derivatives were found to axially bind the dirhodium
tetraacetate core as was suspected, but it is also capable of forming a bidentate

bridge across the rhodium-rhodium bond via N7/NH2 interactions with

76

[Rh2(DTOIF)2(CH30N)5][BF4]2 and [Rh2(OZCCH3)2(CH3CN)6][BF4]2. The guanine
and guanosine bases were reported to be unreactive with respect to the
dirhodium carboxylate compounds due to a lack of observed color change during
the reaction. Our studies have shown that guanine does, in fact, bind to the
dirhodium core in a unique bridging mode involving the N7 and 06 atoms. 1H
NMR studies indicate that the purines form 1:2 adducts with dirhodium
tetraacetate and dirhodium bis-acetate. These results may shed light on the
possible interactions of dirhodium carboxylate compounds with both double-
stranded and single-stranded DNA

DNA pyrimidines do not possess the favorable Lewis base properties
necessary to react with dirhodium carboxylates. Thymidine, 3’, 5’-diphosphate,
however, was found to react with dirhodium tetraacetate under forcing conditions.
The reaction required high temperatures and a long period of time to ensue.
Forcing the reaction to occur resulted in the binding of the dirhodium complex to
the phosphate groups rather than to the base itself. Due to the extreme
conditions of this reaction, it is not likely to be representative of cellular

conditions.

77

Chapter IV
1H NMR Studies of amino acids and DNA purines with Dirhodium

Tetraacetate Derivatives

78

1. INTRODUCTION
A. Competition Reactions

Cis-platin’s antitumor activity is thought to be a result of the formation of
inter— and intra-strand cross-links with DNA (Figure 41a, b) or DNA-protein cross-
links (Figure 41c). Although the biological role is not clear, it has been postulated
that platinum mediated DNA-protein cross-links may be involved in the
mechanism of action of cis-platin 4. Direct platination of replication enzymes is
another possible mechanism for the antineoplastic activity displayed by cis-platin.
Platination of DNA polymerase is known to inhibit replication via binding of the
cysteine residues of the zinc finger domains. This causes the release of zinc

thereby inhibiting the enzyme’s ability to function in replication 26.

 

Figure 41. Intrastrand (a), interstrand (b) and DNA-protein cross-links (c) formed

by cis-platin interactions with DNA.

79

Metal ions are reactive towards both peptides and proteins. Furthermore,
the close association of DNA and proteins, especially in nucleosome cores (DNA
wound around an octamer of proteins to form a more compact structure) would
help facilitate these cross-linking interactions. It has been suggested that
formation of stable platinum mediated DNA-protein crosslinks in vivo may involve
sulfur containing amino acids 4. Model studies were performed with cis-platin
using amino acids with a high affinity to platinum. These amino acids were
constrained to bind the metal atom through their side chains, since this is most
likely to occur with a protein residue. The platinum-amino acid complexes were
then reacted with nucleotides. In competitive binding studies, the platinum
complexes were found to bind the amino acid first, and subsequently to the
nucleotide 54.

In a related study of platinum-methionine adducts found in the urine of
patientsungergoing treatment with cis-platin, it was found that GMP selectively
displaces the sulfur bound methionine. The bis-chelate, [Pt(MET-H-S,N)2], a
byproduct of cis-platin metabolism, is unreactive towards nucleobases at neutral
pH. By monitoring the reactions by 1H NMR spectroscopy, it was found that
intermolecular displacement of the sulfur bound methionine occurs by N7 bound
5’-GMP 5556.

Platinum has a higher affinity for sulfur ligands than it does for nitrogen
ligands, thus sulfur biomolecules diminish the antitumor activity of platinum
complexes. Sulfur nucleophiles have even been used as rescue agents to

remove excess platinum from the body during cis-platin therapy 25. Displacement

80

reactions will most likely be facilitated if the methionine adduct is formed by an
accessible met residue on a DNA binding protein 57.

Competitive binding studies of dirhodium(ll) acetate with the sulfur containing
amino acids, cysteine and methionine, as well as with glutathione were
performed with the nucleotide AMP. These experiments were modeled after the

1H NMR spectral studies mentioned above 565859.

B. Reactions of Dirhodium Acetate Compounds

According to reports by Bear, et al., the interaction of dirhodium(ll)
carboxylates with —SH containing biomolecules results in deactivation of the
carboxylate compound. These studies focused primarily on reactions of
dirhodium(ll) tetraacetate with the amino acid cysteine, and led to claims that
interactions of these compounds results in the breakdown of the dirhodium
carboxylate cage. This purportedly yields an insoluble rhodium(lll) cysteine
species, acetate ions, and H+ ‘2'”.

Additional experiments were performed in an attempt to support the
conclusion that breakdown of the carboxylate cage occurs upon introduction of
dirhodium tetraacetate into tumor cells. Studies involving mice implanted with
Erlich ascites tumors and injected with dirhodium(ll) acetate were performed, and
metabolism of the compound was followed. It was found that dirhodium
tetraacetate breaks down into its components within 2 h of injection. This claim

is based on analyses using 14C labeled acetate coupled with the tracking of

rhodium being achieved by atomic absorption spectroscopy. Results indicated

81

bil

 

that the 1“C label was being removed more rapidly than the rhodium, which was
attributed to the decomposition of dirhodium tetraacetate into easily mobile OAc'
and insoluble rhodium residues. If the dirhodium tetraacetate compound lost
only two acetate ions (and not four), however, the resUlts from tracking studies
would still be consistent with the fact that acetate ions are removed at a faster
rate than rhodium. The researchers suggest complete decomposition as an
explanation for the results, without weighing any alternatives 32. The loss of two
bridging ligands is common for the paddlewheel structures that exhibit antitumor
activity; a good example of this behavior is exhibited by the mixed ligand
complex, Rh2(OZCCF3)2(DTolF)2(H20)2, which loses the two trifluoroacetate
ligands upon reaction with DNA ‘1. The researchers had no definitive proof for
their claims, they only pose the decomposition hypothesis as a possible
explanation for their data.

In order to determine the validity of the results obtained from the
aforementioned experiments, 1H NMR studies were repeated under the same
conditions as previously reported ‘1. Reactions were monitored with cysteine,
glutathione (which contains a cysteine moiety), methionine, as well as the
cysteine mimics, O-Methyl cysteine and 2-Aminothiophenol. All the above
mentioned molecules contain the -SH moiety in question, except for methionine,
which contains —S-CH3, which is expected to bind axially, with no perturbation to
the cage structure. This reaction was performed to determine what effect axial

binding has on the acetate shifts in the NMR. Reactions of the —SH containing

82

amino acids were performed in an effort to test the notion that all —SH containing
amino acids cause the breakdown of the dirhodium carboxylate cage.

Another type of experiment being pursued involves the reaction of
dirhodium tetraacetate derivatives with cysteine. For these reactions, dirhodium
tetraacetate is reacted first with ATP or GTP, then cysteine is added; the reaction
is monitored over time by 1H NMR spectroscopy. The reactions were carried out
in order to determine whether or not cysteine would react with the nucleotide

bound rhodium and lead to breakdown of the dirhodium carboxylate unit.

2. EXPERIMENTAL
A. Competition Reactions

The reactions were carried out in a 50 mM NaHzPO4 buffer solution at pH 7 in
D20. 1H NMR spectral measurements were performed on a Varian 300 MHz
instrument at time = 0, 15 minutes, 2, 4, and 24 hours. Standard 1H NMR
spectra of the amino acid/nucleotide mixtures were measured, and the reaction

progress was monitored.

H H“ (I? 1 2 ("34 5 6
Hooc—c-NH2 HOOC-C-NH2 HOOCCNHCHz-C-NHCCHZCHZCHCOOH
)4ch p HZC 8 H29 3 NH2
2%
CH3

(8) (b) (C)
Figure 42. Numbering scheme for the amino acids. (a) methionine, (b) cysteine,

and (c) glutahtione.

83

Methionine/AMP NMR tube reaction (1)

A 1H NMR spectrum of a 1:1 methioninezAMP mixture was recorded for
use as a reference. 1H NMR (D20) 8: 1.96 (met CH3), 2.00 (m, met BCHz), 2.47
(t, met yCHz), 3.69 (met aCH), 3.88, 4.20, 4.33 (AMP sugar), 5.97 (d, ade H1’),
8.07 (ade H2), 8.39 (ade H8). A 4 mM solution of dirhodium tetraacetate in
deuterated buffer, 1H NMR (020) 8: 1.74 (s,CH3), was added to an 8 mM solution
of 1:1 methioninezAMP. At time = 0 minutes 1H NMR (D20) 8: 1.76 (s, CH3), 2.01
(m, met BCHz), 2.52 (t, met yCHz), 3.71 (met aCH), 6.01 (ade H1”), 8.18 (ade
H2), 8.46 (ade H8). At time = 15 min 1H NMR (020) 8: 1.76 (s, CH3), 2.00 (s,
acetate, met BCHZ), 2.52 (t, met 'YCHz), 3.71 (met otCH), 6.01 (ade H1’), 8.18
(ade H2), 8.45 (ade H8). At time = 2 h 1H NMR (020) 8: 1.76 (s, CH3), 2.01
(acetate, met BCHz), 2.52 (t, met yCHz), 3.71 (met aCH), 6.02 (ade H1’), 8.20
(ade H2), 8.45 (ade H8). At time = 4h 1H NMR (020) 8: 1.76 (s, CH3), 2.02
(acetate, met BCHZ), 2.54 (t, met yCHz), 3.71 (met aCH), 6.03 (ade H1’), 8.21
(ade H2), 8.47 (ade H8). At time = 24 h 1H NMR (020) 8: 1.76 (s, CH3), 2.02
(acetate, met BCHz), 2.54 (t, met yCHz), 3.71 (met aCH), 6.02, 6.03 (ade H1’),
8.21 (ade H2), 8.47 (ade H8).
Glutathione/AMP NMR tube reaction (2)

A 1H NMR spectrum of a 1:1 glutathionezAMP mixture was recorded for
use as a reference. 1H NMR (D20) 8: 2.01 (gsh4), 2.39 (gsh5), 2.78 (gsh3), 3.61

(gsh6), 3.61 (93111), 4.40 (gsh2), 5.95, 5.97 (d, ade H1’), 8.07 (ade H2), 8.37

(ade H8). A 4 mM solution of dirhodium tetraacetate in deuterated buffer, 1H

84

NMR (D20) 8: 1.74 (s,CH3), was added to an 8 mM solution of 1:1
glutathionezAMP. At time = 0 1H NMR (020) 8: 1.75, 1.76 (CH3), 2.00 (gsh4),
2.39 (gsh5), 2.79 (gsh3), 3.61 (gsh6), 3.61 (gsh1), 4.40 (gsh2), 6.02 (d, ade H1”),
8.18 (ade H2), 8.45 (ade H8). At time = 15 min 1H NMR (D20) 8: 1.74, 1.77
(CH3), 2.00 (gsh4), 2.39 (gsh5), 3.61 (gsh6), 3.61 (gsh1), 5.98, 6.00 (d, ade H1’),
8.14 (ade H2), 8.42 (ade H8). At time = 2 h 1H NMR (020) 8: 1.74, 1.77 (CH3),
2.00 (gsh4), 2.37 (gsh5), 3.61 (gsh6), 3.61 (gsh1), 5.98, 6.00 (d, ade H1’), 8.12
(ade H2), 8.42 (ade H8). At time = 4 h 1H NMR (020) 8: 1.74, 1.77 (CH3), 2.00
(gsh4), 2.36 (gsh5), 3.61 (gsh6), 3.61 (gsh1), 5.98, 6.00 (d, ade H1’), 8.14 (ade
H2), 8.42 (ade H8). At time = 24 h 1H NMR (020) 8: 1.75, 1.77 (CH3), 2.00
(gsh4), 2.37 (gsh5), 3.61 (gsh6), 3.61 (gsh1), 5.98, 6.00 (d, ade H1’), 8.13 (ade
H2), 8.42 (ade H8).
Cysteine/AMP NMR tube reaction (3)

A 1H NMR spectrum of a 1:1 cysteinezAMP mixture was recorded for use
as a reference. 1H NMR (D20) 8: 2.88 ( cys BCHz), 3.82 (cys aCH), 5.95, 5.97
(d, ade H1’), 8.07 (ade H2), 8.38 (ade H8). A 4 mM solution of dirhodium
tetraacetate in deuterated buffer, 1H NMR (020) 8: 1.74 (s,CH3), was added to an
8 mM solution of 1:1 cysteinezAMP. At time = 0 1H NMR (020) 8: 1.75 (s, CH3),
2.93 ( cys BCHZ), 3.82 (cys otCH), 5.96, 5.98 (d, ade H1’), 8.10 (ade H2), 8.40
(ade H8). At time = 15 min 1H NMR (D20) 8: 1.74 (s, CH3), 2.92 ( cys BCHz), 3.82
(cys aCH), 5.96, 5.98 (d, ade H1’), 8.08 (ade H2), 8.39 (ade H8). At time = 2 h 1H
NMR (D20) 8: 1.74 (s, CH3), 2.86 ( cys BCHz), 3.82 (cys aCH), 5.96, 5.98 (d, ade

H1’), 8.08 (ade H2), 8.39 (ade H8). At time = 4 h 1H NMR (D20) 8: 1.74 (s, CH3),

85

2.88 ( cys BCHz), 5.96, 5.98 (d, ade H1’), 8.08 (ade H2), 8.39 (ade H8). At time =
24 h 1H NMR (020) 8: 1.74 (s, CH3), 2.98 ( cys BCHz), 3.95 (cys dCH), 5.96, 5.98

(d, ade H1’), 8.08 (ade H2), 8.39 (ade H8).

3. Dirhodium Tetraacetate Reactions

Each reaction was carried out in a 50 mM NaHzPO4 buffer solution at a pH
of 7.5 in D20. 1H NMR spectra were performed on a Varian 300 mHz instrument
for each of the experiments, except for the 2-aminothiophenol reaction which was
measured with a Varian 500 MHz instrument. Standard NMR spectra of the
starting materials, Rh2(OchH3)4(CH30H)2, Rh2(OchH3)4(ATP)2,
Rh2(02CCH3)2(GTP)2(H20)2, and [Rh2(02CCH3)2(CH30N)6][BF4]2 were obtained,
and reaction progress was monitored according to these spectra.
Rh2(02CCH3)4(CH30H)2 + cysteine (4)

A 0.4 M solution of Rh2(02CCH3)4(CH30H)2 was placed in an NMR tube
and titrated with a 0.4 M solution of cysteine in increments of one equivalent up
to four equivalents. 1H NMR spectra were recorded at a 1:1 ratio, 1:2 ratio, 1:3
ratio, and 1:4 ratio of dirhodium tetraacetate to cysteine. Reaction progress was
monitored by shifts in the acetate peak. 1H NMR (020), 1:1 ratio, 8 ppm: 1.62 (s,
bound acetate) 1.79 (s, free acetate). 1H NMR (D20), 1:2 ratio, 8 ppm: 1.62 (s,
bound acetate), 1.79 (s, free acetate). 1H NMR (020), 1:3 ratio, 8 ppm: 1.62 (s,
bound acetate), 1.79 (s, free acetate). 1H NMR (D20), 1:4 ratio, 8 ppm: 1.62 (s,

bound acetate), 1.79 (s, free acetate).

86

Rh2(02CCH3)4(CH30H)2 + glutathione (5)

A 0.4 M solution of Rh2(OZCCH3)4(CH30H)2 was placed in an NMR tube
and titrated with a 0.4 M solution of glutathione in increments of one equivalent
up to four equivalents. 1H NMR spectra were recorded at a 1:1 ratio, 1:2 ratio,
1:3 ratio, and 1:4 ratio of dirhodium tetraacetate to glutathione. Reaction
progress was monitored by shifts in the acetate peak. 1H NMR (D20), 1:1 ratio, 8
ppm: 1.67 (s, bound acetate) 1.84 (s, free acetate). 1H NMR (D20), 1:2 ratio, 8
ppm: 1.67 (s, bound acetate) 1.84 (s, free acetate). 1H NMR (D20), 1:3 ratio, 8
ppm: 1.67 (s, bound acetate) 1.84 (s, free acetate). 1H NMR (D20), 1:4 ratio, 8
ppm: 1.67 (s, bound acetate) 1.84 (s, free acetate).

Rh2(02CCH3)4(CH3OH)2 + methionine (6)

A 0.4 M solution of Rh2(OchH3)4(CH30H)2 was placed in an NMR tube
and titrated with a 0.4 M solution of methionine in increments of one equivalent
up to four equivalents. 1H NMR spectra were recorded at a 1:1 ratio, 1:2 ratio,
1:3 ratio, and 1:4 ratio of dirhodium tetraacetate to methionine. Reaction
progress was monitored by shifts in the acetate peak. 1H NMR (D20), 1:1 ratio, 8
ppm: 1.65 (s, bound acetate), 2.02 (s, new acetate). 1H NMR (020), 1:2 ratio, 8
ppm: 1.65 (s, bound acetate), 2.00 (s, new acetate). 1H NMR (020), 1:3 ratio, 8
ppm: 1.65 (s, bound acetate), 1.97 (s, new acetate). 1H NMR (D20), 1:4 ratio, 8
ppm: 1.65 (s, bound acetate), 1.95 (bs, new acetate).

Rh2(OchH3)4(ATP)2 + cysteine (7)
A 0.4 M solution of Rh2(OZCCH3)4(ATP)2 was placed in an NMR tube and

titrated with a 0.4 M solution of cysteine in increments of one equivalent up to

87

four equivalents. 1H NMR spectra were recorded at a 1:1 ratio, 1:2 ratio, 1:3
ratio, and 1:4 ratio of Rh2(02CCH3)4(ATP)2 to cysteine. Reaction progress was
monitored by shifts in the acetate resonance. 1H NMR (020), standard, 8 ppm:
1.73 (s, bound acetate), 8.28 (s, ade H2), 8.49 (s, ade H8). 1H NMR (D20), 1:1
ratio, 8 ppm: 1.73 (s, bound acetate), 1.88 (s, free acetate), 6.03 (ade H1”), 8.27
(s, ade H2), 8.47 (s, ade H8). 1H NMR (020), 1:2 ratio, 8 ppm: 1.73 (s, bound
acetate), 1.88 (s, free acetate), 6.03 (d, ade H1”), 8.27 (s, ade H2), 8.47 (s, ade
H8). 1H NMR (020), 1:3 ratio, 8 ppm: 1.73 (s, bound acetate), 1.88 (s, free
acetate), 6.02 (d, ade H1”), 8.27 (s, ade H2), 8.46 (s, ade H8). 1H NMR (D20),
1:4 ratio, 8 ppm: 1.73 (s, bound acetate), 1.88 (s, free acetate), 6.03 (d, ade H1”),
8.26 (s, ade H2), 8.46 (s, ade H8).

Rh2(02CCH3)2(GTP)2(H20)2 + cysteine (8)

A 0.4 M solution of Rh2(020CH3)2(GTP)2(H20)2 was placed in an NMR
tube and titrated with a 0.4 M solution of cysteine in increments of one equivalent
up to four equivalents. 1H NMR spectra were taken at a 1:1 ratio, 1:2 ratio, 1:3
ratio, and 1:4 ratio of Rh2(OchH3)2(GTP)2(H20)2 to cysteine. Reaction progress
was monitored by shifts in the acetate peak. 1H NMR (D20), standard, 8 ppm:
1.98 (s, bound acetate), 8.63 (s, gua H8). 1H NMR (020), 1:1 ratio, 8 ppm: 1.81
(s, free acetate), 1.98 (s, bound acetate), 5.73 (d, gua H1”), 8.60 (s, gua H8). 1H
NMR (D20), 1:2 ratio, 8 ppm: 1.81 (s, free acetate), 1.98 (s, bound acetate), 5.71
(d, gua H1”), 7.90 (s, gua H2), 8.63 (s, gua H8). 1H NMR (D20), 1:3 ratio, 8 ppm:
1.81 (s, free acetate), 1.98 (s, bound acetate), 5.69, 5.71 (d, gua H1”), 7.90 (s,

gua H2), 8.62 (s, gua H8). 1H NMR (020), 1:4 ratio, 8 ppm: 1.81 (s, free acetate),

88

1.98 (s, bound acetate), 5.69, 5.71 (d, gua H1”), 7.90 (s, gua H2), 8.61 (s, gua
H8).
[Rh2(02CCH3)2(CH3CN)5][BF4]; + cysteine (9)

A 0.4 M solution of [Rh2(02CCH3)2(CH3CN)6][BF4]2 was placed in an NMR
tube and titrated with a 0.4 M solution of cysteine in increments of one equivalent
up to four equivalents. 1H NMR spectra were taken at a 1:1 ratio, 1:2 ratio, 1:3
ratio, and 1:4 ratio of rhodium bisacetate to cysteine. Reaction progress was
monitored by shifts in the acetate peak. 1H NMR (D20), standard, 8 ppm: 1.79
(s, bound acetate). 1H NMR (020), 1:1 ratio, 8 ppm: 1.79 (s, bound acetate). 1H
NMR (020), 1:1 ratio, 8 ppm: 1.79 (s, bound acetate). 1H NMR (020), 1:2 ratio, 8
ppm: 1.79 (s, bound acetate). 1H NMR (D20), 1:3 ratio, 8 ppm: 1.79 (s, bound
acetate). 1H NMR (020), 1:4 ratio, 8 ppm: 1.79 (s, bound acetate).
Rh2(OZCCH3)4(CH30H)2 + O-Methyl cysteine (10)

A 0.4 M solution of Rh2(OZCCH3)4(CH30H)2 was placed in an NMR tube
and titrated with a 0.4 M solution of O-Methyl cysteine in increments of one
equivalent up to four equivalents. 1H NMR spectra were recorded at a 1:1 ratio,
1:2 ratio, 1:3 ratio, and 1:4 ratio of rhodium tetraacetate to O-Methyl cysteine.
Reaction progress was monitored by shifts in the acetate peak. 1H NMR (D20),
1:1 ratio, 8 ppm: 1.74 (s, bound acetate), 1.92 (s, acetate). 1H NMR (D20), 1:2
ratio, 8 ppm: 1.74 (s, bound acetate), 1.92 (s, acetate). 1H NMR (020), 1:3 ratio,
8 ppm: 1.74 (s, bound acetate), 1.92 (s, acetate). 1H NMR (D20), 1:4 ratio, 8

ppm: 1.74 (s, bound acetate), 1.92 (s, acetate).

89

Rh2(02CCH3)4(CH30H)2 + 2-aminothiophenol (11)

A 0.4 M solution of Rh2(02CCH3)4(CH30H)2 was placed in an NMR tube
and titrated with a 0.4 M solution of 2-aminothi0phenol in increments of one
equivalent up to four equivalents. 1H NMR spectra were recorded at a 1:1 ratio,
1:2 ratio, 1:3 ratio, and 1:4 ratio of rhodium tetraacetate to 2-aminothiophenol.
Reaction progress was monitored by shifts in the acetate peak. 1H NMR
(CD300), 1:1 ratio, 8 ppm: 1.82 (s, bound acetate). 1H NMR (c0300), 1:2 ratio,
8 ppm: 1.87 (s, bound acetate), 1.87 (s, acetate). 1H NMR (CD30D), 1:3 ratio, 8
ppm: 1.86 (s, bound acetate), 1.88 (s, acetate). 1H NMR (cosoo), 1:4 ratio, 8

ppm: 1.90 (s, bound acetate), 1.93 (s, acetate).

3. Results and Discussion
A. Competition Reactions
Methionine/AMP NMR tube reaction (1)

This reaction was performed to determine whether one type of cellular
biomolecule is preferred over the other. Both methionine and AMP are capable
of behaving as axial ligands, so a preference for one should be possible to
observe. Since sulfur is a stronger donor than nitrogen, it would seem likely that
the methionine would be preferred over the AMP. In order to ascertain whether
this is the case, there are three sets of resonances that are important to monitor
in the NMR spectra. These are the acetate, methionine and adenine 1H NMR

peaks which shift upon their coordination with dirhodium tetraacetate. By

90

monitoring these shifts, the course of the reaction can be determined (Table 2,
Figure A1).

Upon addition of dirhodium tetraacetate to the methionine/AMP solution,
the acetate peak shifts from 1.74 to 2.01 ppm for a Change of 0.27 ppm. This
downfield shift is in accordance with the spectrum of the bulk product and
signifies axial binding of the amino acid. Throughout the rest of the experiment,
a gradual shift downfield occurs until after 24 h the final acetate peak position is
at 2.02 ppm.

Methionine peaks shift as expected for an axially bound sulfur donor
ligand. The —SCH3 shifts upfield from 1.96 ppm in the standard to 1.76 ppm after
24 h. The aCH resonance begins to shift at time zero, moving 0.02 ppm, and
remaining constant for the duration of the experiment. The 'YCHz peak shifts until
it reaches 2.04 ppm for a total shift of 0.08 ppm. The BCHz shift coincides with
the acetate peak. Changes in the resonances of these protons are expected,
with the ‘YCHQ closest to the sulfur binding site, showing the greatest shift upon
binding, followed by the BCHz then the aCH resonances.

The adenine region reveals binding of the nucleotide as well. The H8
proton shifts throughout the entire experiment by a total of 0.09 ppm, H2 shifts
0.14 ppm, and H1” shifts 0.06 ppm. This indicates adenine coordination to the
metal as well. In the bulk product, significant shifts are observed in the H8/H2
region of the spectrum as well.

From the aforementioned data, it is apparent that dirhodium tetraacetate

reacts with both methionine and AMP (Figure 43) and that when both reagents

91

are present, neither inhibits binding of the other. This is an encouraging result,
as it shows that the amino acid does not sequester the dirhodium tetraacetate
and keep it from reacting with the nucleotide. It is therefore possible that, in the
cell, proteins that contain methionine residues bind to the dirhodium molecule

which then goes on to form an adduct with DNA and inhibit replication.

 

H2
/ \ \N-H‘ A
N ‘o o
/ /‘l—<
1/0 /O /
/N —l|Rh /Rh—'S NH2
N o
SUGAR/ \H/B )T’0 1 p:
o 0 OH

Figure 43. Proposed product from the competition reaction of Rh2(02CCH3)4

with methionine and AMP.

Table 2. Proton resonances monitored during the course of the methionine/AMP

competition experiment.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Standard t=0 min t=15 min t=2 h t=4 h t=24 h
Ade H8 8 8.34 8 8.46 8 8.45 8 8.45 8 8.47 8 8.47
Ade H2 8 8.07 8 8.18 8 8.18 8 8.20 8 8.21 8 8.21
Ade H1” 8 5.97 8 6.01 8 6.01 8 6.02 8 6.03 8 6.03

8 5.95 8 6.02
Met CH3 81.96 8 1.76 8 1.76 8 1.76 8 1.76 8 1.76
Met BCHZ 8 2.00 8 2.01 8 2.00 8 2.01 8 2.02 8 2.02
Met ‘YCHz 8 2.47 8 2.52 8 2.52 8 2.52 8 2.54 8 2.54
Met otCH 8 3.69 8 3.71 8 3.71 8 3.71 8 3.71 8 3.71
Acetate 8 1.74 8 2.01 8 2.00 8 2.01 8 2.02 8 2.02

 

92

 

Glutathione/AMP NMR tube reaction (2)

Since glutathione resembles a protein with a cysteine moiety, this is an
important reaction to monitor by NMR spectroscopy. Previous results of
reactions of Rh2(OchH3)4 with glutathione support the conclusion that the
molecule binds equatorially in a 2:1 fashion by replacing two of the acetates on
the tetraacetate core. The question remains, however, whether this dirhodium-
glutathione adduct is able to react further with DNA, or whether the dirhodium
complex has been deactivated with respect to further substitution, as is the case
with cis—platin. The answer is obtained by monitoring changes in the acetate,
glutathione (gsh2 and gsh3 protons, Figure A2c), as well as adenine regions of
the spectrum (Table 3, Figure A2).

Upon addition of dirhodium tetraacetate to the glutahtione/AMP mixture,
the bound acetate shifts to 1.76 ppm, and a free acetate peak appears at 1.75
ppm. As time progresses, further shifts in the acetate peak are negligible with
the final values being 1.75 ppm for free and 1.77 ppm for bound acetate. The
overall shift in these peaks is 0.03 ppm for bound acetate and 0.00 ppm for the
free acetate. These results support the previous conclusion that two acetates
are released in the glutathione reaction with rhodium tetraacetate.

The binding of glutathione to the dirhodium tetraacetate core can be
observed by the loss of the free gsh3 peak as well as the gsh2 peak. After 15
minutes, the gsh3 peak has diminished in intensity, and less intense features
downfield begin to appear. These new resonances are identical to the ones that

comprise the quartet of doublets in the bulk Rh2(OAc)2(GSH)2(H20)4 product.

93

The gsh2 peak has shifted into the HDO peak and is no longer visible.
Reduction of the gsh3 peak continues through 2 hours into the reaction after
which time it is no longer visible. All other GSH peaks exhibit negligible shifts.
The peaks used to monitor AMP reactions are the H8, H2 and the H1”
groups. Each of these peaks shifts at the time of mixing. The overall shifts for
H8, H2, and the H1’ protons are 0.05, 0.06, and 0.03 ppm, respectively. These
data indicate that the AMP is binding as well, most likely in the axial positions
based on the bulk reaction (Figure 44).
The results of this experiment support the conclusion that when faced with both
glutathione (an —SH containing amino acid) and a nucleotide, dirhodium
tetraacetate reacts with both simultaneously. In the presence of the —SH
containing tripeptide, glutathione, dirhodium tetraacetate is able to bind the
nucleotide adenine. These results refute the earlier conclusion that all —SH
containing biolmolecules deactivate the tetraacetate by breaking down the core.
From the spectra it is clear that both biomolecules are binding. On a more
general level, the data suggest that the binding of a cysteine moiety on a protein

will not disable the ability of dirhodium acetate to bind DNA.

94

 

 

N 0 0 H8
/N—Rh /Rh'—N
SUGAR’NfiBN ' S L /
”H I
\_/ NH \N/le
H2 N H.
(if ”'4'vo H8
/N—,R /Rh—N
‘—’ HN- \ JHZ
H2 N H. H N
7// \ N'Ha A
N o 0 H8
/N —Rh—-;R —N
SUGAR’N\/ N N /
H8 K K’ N
HN \ J
H N H2
H
H27l/ \ .N'H. A
N 0 0 H8
Nth-—/Rh—N
SUGAR’NJ N S 1 /
H8 K KN N
H N \ J
H N H2

Figure 44. Proposed products formed in the competition reaction with glutathione
and AMP.

95

its

at

Table 3.

Proton resonances monitored throughout the course of the

glutathione/AMP competition experiment.

 

 

 

 

 

 

 

 

 

 

 

 

Standard t=0 min t=15 min t=2 h =4 h t=24 h
Ade H8 8 8.37 8 8.45 8 8.42 8 8.42 , 8 8.42 8 8.42
Ade H2 88.07 88.18 88.14 88.12 88.14 88.13
Ade H1” 8 5.97 8 6.02 8 6.00 8 6.00 8 6.00 8 6.00
8 5.95 8 5.98 8 5.98 8 5.98 8 5.98
Gsh 4 8 2.01 8 2.00 8 2.00 8 2.00 8 2.00 8 2.00
Gsh 5 8 2.30 8 2.39 8 2.39 8 2.37 8 2.36 8 2.37
Gsh 3 8 2.78 8 2.79 8 sm 8 sm 8 2.83 8 2.84
Gsh 6 8 3.61 8 3.61 8 3.61 8 3.61 8 3.61 8 3.61
Gsh 1 8 3.61 8 3.61 8 3.61 8 3.61 8 3.61 8 3.61
Gsh 2 8 4.40 8 4.40
Acetate 81.74 8 1.76 8 1.77 8 1.77 8 1.77 8 1.77
81.75 8 1.74 8 1.74 8 1.74 8 1.75

 

 

 

 

 

 

 

 

Cysteine/AMP NMR tube reaction (3)

Since cysteine concentrations in the cell are much lower than the
concentrations of methionine and glutathione, this reaction is not as indicative of
what may happen in a cell treated with dirhodium tetraacetate. The reaction was
performed because previous researchers had reported some surprising results
about this reaction that we found difficult to accept. Furthermore cysteine is an
important moiety in a large number of proteins, although the glutathione reaction
is a better model for reactions of proteins with antitumor agents.

Reactions of Rh2(020CH3)4 with cysteine proceed differently than the
previous two reactions (Table 4, Figure A3) in that the acetate is immediately

labilized within the time of mixing (time zero) as indicated by the appearance of

96

free acetate resonances. Based on other NMR experiments and on the bulk
reaction, this is not an unexpected result. Displacement of all the acetate ligands
appears to be the trend for cysteine reactions with dirhodium tetraacetate.
Cysteine peak shifts occur for both the otCH and BCHz groups. At time
zero, the BCHz experiences the greatest shift (0.05 ppm) whereas the aCH does
not appear to have shifted. After 15 minutes, the BCHz remains essentially
unchanged, but it decreases in intensity and the familiar quartet of doublets
begins to appear in the expected position. The aCH peak decreases in intensity
and begins to appear downfield of a sugar peak. This trend continues until the
original peak has disappeared and the new peak remains unchanged after 24 h.
An unusual shifting of the adenine resonances occurs throughout the
course of the reaction. At the time of mixing, H8 has shifted by 0.02 ppm, H2 by
0.03 ppm and H1’ by 0.01 ppm, a clear indication that the nucleotide is reacting.
After 5 minutes has elapsed, the peaks begin to shift upfield again. H8 moves by
0.01, H2 by 0.02, and H1” by 0.01 for shifts of 0.01 for H8, 0.01 for H2, and 0.00
for H1”. These values remain essentially constant throughout the rest of the
experiment. It appears as though the AMP reacts initially, but then is released
after about fifteen minutes. One possible explanation for this is that axial binding
of AMP to the core would no longer be stabilized by H-bonding of the exocyclic
amine group in the absence of acetate oxygen atoms (Figure 45). The lack of
stabilization via hydrogen bonding is the reason that guanine was originally

thought to be unreactive towards dirhodium tetraacetate. Indeed guanine purines

97

 

 

SUGAR/

Figure 45. Proposed products formed in the competition reaction with cysteine

and AMP.

98

are found to bind equatorially instead of axially, most likely due to repulsive

interactions between the 06 ketone and acetates.

equatorial ligand than AMP, it may not allow AMP to compete for the equatorial

sites.

Table 4.

Proton resonances monitored over time for the cysteine/AMP

competition reaction.

lf cysteine is a better

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Standard t=0 min t=15 min =2 h =4 h t=24 h
Ade H8 8 8.38 8 8.40 8 8.39 8 8.39 8 8.39 8 8.39
Ade H2 8 8.07 8 8.10 8 8.08 8 8.08 8 8.08 8 8.08
Ade H1” 8 5.97 8 5.98 8 5.98 8 5.98 8 5.98 8 5.98

8 5.95 8 5.96 8 5.96 8 5.96 8 5.96 8 5.96
Cys BCH2 8 2.88 8 2.93 8 2.923 8 2.86s 8 2.88 8 2.98
Cys 0tCH 8 3.82 8 3.82 8 3.82 8 3.82 8 sh 8 3.95
Acetate 81.74 8 1.75 8 1.74 8 1.74 8 1.74 8 1.74

 

 

B. Dirhodium Tetraacetate Reactions
Rh2(02CCH3)4(CH30H)2 + cysteine (4)

This reaction was performed in order to verify the results of previous
experiments that claimed cysteine disrupts the cage of the dirhodium tetraacetate
molecule. One equivalent of cysteine was titrated into a solution of dirhodium
tetraacetate for a total of four equivalents. As can be clearly seen in the 1H NMR
spectra after addition of each equivalent (Figure A4), the original acetate peak
decreases and a new peak appears. Integration of these two peaks (Table 5)
shows that after the addition of one equivalent of cysteine, one acetate has been

lost from the dirhodium tetraacetate core. The addition of a second equivalent of

99

cysteine gives a 1:1 ratio of bound acetate to new acetate, resulting from the
displacement of two acetates. Three equivalents of cysteine result in the
displacement of yet another acetate, leaving one bound acetate. The addition of
the fourth and final equivalent of cysteine, results 'in the replacement of all
original acetate ligands. To determine which type of acetate the peak at 1.79
ppm corresponded to (free or bound), a solution of N302CCH3 was added to the
reaction. This resulted in the increase of the new acetate resonance, leading to
the conclusion that cysteine causes the displacement of all bound acetate

ligands from the dirhodium core (Figure 46).

Table 5. Integration of bound:free acetate for the cysteine/dirhodium

tetraacetate reaction.

 

Ratio Bound Free Integration
Tetraacetate 1 .62

1 :1 1.62 1.79 1.0:0.32

1 :2 1.62 1.79 1.0:0.65

1 :3 1.62 1.79 1.0:0.33

1 :4 1.62 1.79 1 020.13

 

100

o s S O S O S O
’— /—x ’— f\
1.4.78 4.87.1.8 .L/°_..?s 1.4:...»
o/ I .s/ I o, I ,s I o/ I ,s/ I o/ I ,s/I
g s o s s o 6

HO NH OH HO
0 T o s/ij
o
s Flth/s Rh/SS s flan/s in s
s/ I / I s/ I / I
o S o O
0 IL/3
OH HO
H2N HzN OH *1sz NH;

Figure 46. Proposed products from the reaction of four equivalents of cysteine

with dirhodium tetraacetate.

Rh2(02CCH3)4(CH30H)2 + glutathione (5)

This experiment was carried out in the same manner as the cysteine
experiment. Since free —SH containing biomolecules are claimed to be capable
of destroying the tetraacetate cage, it is reasonable to assume that glutathione
should cause the same type of reaction as cysteine. As can be observed from
the NMR data (Figure A5, Table 6), this is not the case. The addition of one
equivalent of glutathione causes free acetate to appear in the NMR spectrum

(close to a 1:3 ratio of freezbound). Following the addition of a second

101

equivalent, the free acetate peak increases in intensity, and the ratio is close to a
1:1 ratio of bound:free acetate. The addition of the third and fourth equivalent
causes little additional change in the spectrum. It appears from these data that
glutathione does not react with dirhodium tetraacetate in a manner akin to
cysteine. Instead of replacing all of the acetate groups, two remain attached to
the dirhodium core (Figure 47). This is further supported by elemental analyses
carried out on the bulk product which support the formulation of the product as

Rh2(OzCCH3)4(GS)2(H20)4.

Table 6. Integration of bound:free acetate for the glutathione/dirhodium

tetraacetate reaction.

 

Ratio Bound Free Integration
Tetraacetate 1 .62

1:1 1.67 1.84 1.02020

1 :2 1.67 1.84 1020.44
1:3 1.67 1.84 1.0:0.72
1:4 1.67 1.84 10:10

 

102

 

 

O O
.— ,<
I/0 |<o0 I/O '00
Lth Rh—L L—Rh /Rh-L
N l ,s I N | .s I
N S N

o o o o
L-llah’: ll??? L L-FIRh/g: Flag? L
“’1 ~1 “’1 81
K K. K KN

Figure 47. Proposed products from the reaction of glutathione with
dirhodium tetraacetate

103

Rh2(OchH3)4(CH30H)2 + methionine (6)

Following the same procedures as above, the methionine reaction was
performed in an effort to document the type of reaction that takes place with an
-SCH3 containing molecule. If methionine reacts with dirhodium tetraacetate to
form an axial bond through the sulfur atom, the acetate peak should shift
accordingly. The addition of one equivalent of methionine leads to the
appearance of a broad resonance at 2.02 ppm (Figure A6). This is thought to be
an acetate peak associated with an axially bound methionine. Integration is
inconclusive due to the presence of an overlapping BCH peak. The addition of a
second equivalent of methionine causes the peak to increase in intensity. This
trend continues with the addition of the third and fourth equivalents as well. This
type of behavior is different than that observed for reactions of the —SH
containing biomolecules. This is obviously due to differences in the lability of
acetate ligands in the presence of axial methionine versus the equatorial

glutathione and cysteine molecules (Figure 48).

Figure 48. Proposed product from the reaction of methionine with dirhodium

tetraacetate.

104

Rh2(02CCH3)4(ATP)2 + cysteine (7)

This reaction was performed under the same conditions as the dirhodium
tetraacetate reactions. The purpose of this experiment is to ascertain the
influence of a nucleotide co-ligand on the reaction with cysteine. The reaction
was followed by NMR spectroscopy through the addition of four equivalents of
cysteine (Figure A7, Table 7). The standard (A7a) shows an acetate peak at
1.73 ppm. The addition of one equivalent of cysteine (A7b) leads to the
appearance of a peak at 1.88 in a one to three ratio with the bound acetate
resonance. Also of note is the shift of the H8 resonance from 8.49 to 8.47 ppm.
The addition of a second equivalent (A7c) leads to an increase in intensity of the
free acetate peak, which now integrates at a 1:1 ratio with the bound acetate.
The H8 peak continues to shift upfield and is now at 8.47 ppm. Three
equivalents lead to a further increase in the intensity of the free acetate peak, but
the integration is still at a 1:1 ratio. The H8 resonance shifts to 8.46 ppm. The
final addition of cysteine (A7e) does not seem to affect the spectrum. The ratio
of bound to free acetate remains at 1:1 and the position of the H8 proton remains

constant.

105

Table 7. Integration of bound:free acetate for the Rh2(02CCH3)4(ATP)2 +

cysteine reaction.

 

Ratio Bound Free Integration
Standard 1.73

1 :1 1.73 1.88 1.02024

1 :2 1.73 1.88 100.43

1 :3 1.73 1.88 1.02081

1 :4 1.73 1.88 100.95

 

Shifts are observed in the H8 proton since it is closest to the site of
metallation and would be most sensitive to any changes in the electronic
environment around the dirhodium core. From this experiment it is apparent that
the nucleotide, in this case ATP, bound to the dirhodium core prevents the
displacement of all bound acetates as opposed to the dirhodium tetraacetate
experiment with cysteine. Extrapolated to a cellular setting, one may speculate
that DNA bound dirhodium acetate would not be readily destroyed by an amino
acid such as cysteine. The amino acid could complex itself with the rhodium-

DNA adduct, but would not lead to loss of DNA interactions (Figure 49).

106

H27//\N HN.H'00)\

l /O’|<O ;'8\N’SUGAR
N— N

s/
SUGAR’N\Z8N N ._1/ I“ /
N\ / H-NH \ N/ZIHZ

           

H27l/: \ HN H 00*
/N—Rh/-—/Rh—N/
sucAR’N\/ N I 3| /

H8
8\ /N I'lNH \N/FH2
H2 H-
T: ‘ H H MA
H8
I AR
1/0 I<O //\N,SUG
N—Rh-—/-Rh—N

SUGAR’N\/ NKI N

/
S k8,. IN
\ /l
H N H2

H2
YI/N/ \ HN- H 0*
0 H8
< SUGAR
1/0’1/0 FN’
N-;Rh—Rh-N

NJ N I S/I
SUGAR’ H8 Ks N

v. 1,)
H-N' \ H2

H N

/S
\HN]\|(OH

0

Figure 49. Proposed products from the reaction of cysteine with a
dirhodium tetraacetate/ATP derivative.

107

Rh2(02CCH3)2(GTP)2 + cysteine (8)

If the trend as in the same as the ATP reaction, one may expect that
reaction of the GTP-rhodium adduct with cysteine will lead to an incomplete
substitution of acetate ligands. The bound acetate peak appears at 1.98 ppm in
the standard (Figure A8, Table 8). The addition of one equivalent of cysteine
lead to the appearance of a small peak at 1.81 ppm. Two equivalents of cysteine
cause an increase in intensity of this new peak, and a slight shift in the H8 proton
resonance. The addition of three equivalents does not lead to an increase in the
intensity of the new acetate peak. The fourth equivalent shows no change in the
acetate region, but the H8 peak shifts by ~0.02 ppm upfield from its original
position. This experiment reveals that at most only one of the acetate ligands is

replaced by cysteine, but that the parent molecule is still intact (Figure 50).

Table 8. Integration of bound:free acetate for the Rh2(OZCCH3)2(GTP)2 +

cysteine reaction.

 

Ratio Bound Free Integration
Standard 1.98

1:1 1.98 1.81 1.0:O.10

1 :2 1.98 1.81 1.0:0.44
1:3 1.98 1.81 1.0:0.43

1 :4 1.98 1.81 1.0:0.49

 

108

A A

0 O O O

 

 

 

COOH I 0/I<o HOOC COOH I 0" <0 HOOC
/ /
O I N O I
o N N
1*? TAT
r>s /NH COOH /S/— “N HOOC
HN;Rh /Rh—s> HZN’K/S7Rh /Rh’—s NH;
O I N I 0 I N I
o N o N
o
HN N
Nj I \>H8
o H2N \N T
SUGAR
N HOOC
J NH,
3 s

Figure 50. Proposed products from the reaction of cysteine with a dirhodium

tetraacetate/GTP derivative

109

[Rh2(OchH3)2(CH3CN)5][BF4]2 + cysteine (9)

This reaction was performed to determine whether or not cysteine has the
same effect on the bis-acetate as it does on the tetraacetate compound.
Following the titration by 1H NMR spectroscopy (FigUre A9), it is observed that
the reaction that takes place is different than what was anticipated, namely that
the two acetate ligands would be replaced by two cysteine molecules. With the
addition of each equivalent of cysteine, only one acetate peak is present. Since
the appearance of another acetate peak, namely the free acetate peak, does not
occur, any shifts in the peak resonance would be due to bound acetate
influenced by the changes introduced by the presence of a new ligand (Figure
51). The acetate peak is at 1.79 ppm for the standard spectrum, and shifts to
1.80 ppm during the course of the reaction.

Rh2(OchH3)4(CH30H)2 + O-Methyl cysteine (10)

Since the O-Methyl cysteine molecule possesses the same active binding
site as cysteine, namely the HS-C-C-NHZ moiety, it was expected that the NMR
reaction would show the displacement of all the acetate ligands (Figure A10,
Table 9). With the addition of one equivalent of O-Mecys, the appearance of a
new acetate peak at 1.92 ppm is apparent. Integration against the initial acetate
peak shows less than a 1 :3 ratio. The addition of a second equivalent shows an
increase in the intensity of the new peak which now integrates as one acetate to
three acetate ligands. The addition of three and four equivalents of O-Methyl
cysteine lead to an integration of 1:1 for initialznew acetate. This reaction was

expected to result in the displacement of all four acteates, but instead, only two

110

were substituted (Figure 52). This reaction seems to follow the trend of the

glutathione rather than the cysteine chemistry.

Table 9. Integration of bound:free acetate for the reaction of O-methyl cysteine

with dirhodium tetraacetate.

 

Ratio Bound Free Integration
Standard 1.67

1:1 1.67 1.72 1.0:0.21

1 :2 1.67 1.72 1.0:0.35

1 :3 1.67 1.72 1.0:0.47

1 :4 1.67 1.72 1.0:0.49

 

Rh2(OchH3)4(CH3OH)2 + 2-aminothiophenol (11)

The 1H NMR tube reaction of 2-aminothiophenol proved to be more
difficult than the previous reactions as 00300 was required to solubilize the 2-
aminothiophenol. The tetraacetate was dissolved in the 020 buffer used for all
the other experiments. With the addition of one equivalent of 2-aminothiophenol,
a shoulder appears on the downfield side of the acetate peak (Figure A11).
Integration is not possible due to the close proximity of these peaks. Two
equivalents lead to an increase in the shoulder, and a slight shifting of the phenol

protons. Addition of a third and fourth equivalent serve to increase the intensity

111

 

o o o o
L FLh/O/FIREPL L IIRh/O/REOL
O’I ,s/I 0’ s I)

0

U U

0A0 0A0
L-th/:Rh’9L L lth/O/llR<l-9L
0’1 0 I 0/l S/I
L/ L/S LO

Figure 51. Proposed products from the reaction of cysteine with dirhodium bis-

acetate.

112

 

O O
| ,0’|<.0 l ,0’|<.0
L;Rh—;'Rh"L L'—’Rh /Rh-L
N l ,s I N l ,s I
N S N
\_/
0A0 0A0
L FIQh/O’IIRh’OL L IIRh/O’Rh’OL
”’i “/1 ”’1 S/'
K k K KN
\N \N OCH3
0

Figure 52. Proposed products from the reaction of O-methyl cysteine with

dirhodium tetraacetate.

113

of the shoulder and to further increase the shifting of the phenol protons (Figure
54), for total shifts of 0.06 (b), 0.10 (a), 0.05 (d), and 0.04 (c) ppm. These data
support the conclusion that this ligand exhibits a binding behavior different from

that of cysteine (Figure 53) in that it does not replace the acetate ligands.

H(b) H(c)

H(a) H(d)

HS NH2

Figure 54. Labeling scheme for 2-aminothiophenol.

114

>

 

L

71 l,
C}

Figure 53. Proposed products from the reaction of dirhodium tetraacetate with

2-aminothiophenol.

115

4. Conclusions
A. Competition reactions

Reactions of dirhodium tetraacetate in which both amino acids and
nucleotides are present have led to interesting conclUsions about the reactivity of
antitumor agents such as cis-platin and dirhodium tetraacetate. This type of
reaction allows one to determine which biomolecule is preferred, if a preference
indeed exists. The possibility of protein-metal-DNA crosslinking can be
envisioned due to the outcome of these reactions. The competition reactions
with methionine and glutathione reveal that, although binding of an amino acid
occurs, this does not prevent binding of the nucloeotide and vice versa. The
reactions show that when faced with both an amino acid and a nucleotide, the
dirhodium compounds react with both types of molecules. One of the more
important outcomes of this work is the realization that binding of an -SH
containing peptide, glutathione, does not result in a deactivation of the dirhodium
acetate compound as previously postulated. The cysteine reaction is different in
that both the amino acid and the nucleotide react at the start of the reaction, but
with time, the AMP seems to be selectively removed from the dirhodium
compound. This could be the result of steric interactions between the amino
group on the adenine and the bridging cysteine. The stabilizing influence of
hydrogen bonds is no longer possible without the acetate ligands, and the
adenine may not be a good axial lignad in the absence of these interactions. In

the final analysis, the most important reactions are those with glutathione and

116

methionine, as these indicate that, even in the presence of sulfur containing

amino acids, dirhodium tetraacetate is still capable of reacting with DNA purines.

B. Dirhodium Tetraacetate Reactions

Titration of the sulfur containing amino acids cysteine, methionine, and
glutathione into solutions of dirhodium tetraacetate and its derivatives has led to
some interesting findings. In the case of methionine, a simple axial addition is
observed, whereas with glutathione, displacement of two acetates is preferred.
Finally, cysteine displaces all four acetate ligands to give a single, diamagnetic
product. This latter result is in accordance with the earlier experiments that
suggest cysteine breaks the carboxylate cage. The claims of a Rh(ll)
mononuclear product have been refuted, however, as we have incontrovertible
evidence that the Rh-Rh bond has not been broken. Earlier literature had
claimed that reactions of dirhodium tetraacetate with —SH containing
biomolecules causes this breakdown to occur. Results from the glutathione
experiment have proven this claim to be unfounded. In the case of cysteine,
mass spectrometry and elemental analyses on the bulk product (Chapter 2) point
to the conclusion that the acetates are replaced by cysteine ligands, but that the
dinuclear rhodium core is still intact.

Cysteine mimics, such as O-Methyl cysteine and 2-Aminothiophenol, were
studied to determine if molecules with the same HS-C-C-NHz moiety as cysteine
would lead to substitution of the acetate ligands. The reaction of O-Methyl

cysteine with dirhodium tetraacetate did not follow the expected pattern for a

117

nitrogen-sulfur bridging molecule. Instead of substituting all four of the acetate
ligands, it appears that only two of the acetates are replaced during the course of
the reaction. 2-Aminothiophenol gave similar results. Although integration is
hindered by the close proximity of the two acetate peaks, it appears that
incomplete substitution of the bound acetate is occurring. Due to these findings,
the question is posed about the potential binding sites of the amino acid cysteine.
The mimics were chosen for their HzN-C-C-SH moiety, which was considered to
be the reactive group on the cysteine molecule. Since these molecules do not
behave in the same manner as cysteine, it seems logical to conclude that the
carboxylate group is playing a role in the metal reactivity of the complex.
Experiments wherein dirhodium tetraacetate that had been previously
reacted with a nucleotide was followed by treatment with cysteine supports the
conclusion that the reaction with cysteine is not destructive to the dinuclear
structure. Both the ATP and GTP reactions indicate displaCement of acetate
ligands is occurring, but, complete substitution does not occur. After the fourth
equivalent of cysteine has been added, the core structure is still intact. It is
possible to predict from these model studies that dirhodium tetraacetate/DNA
adducts would not be disrupted due to the presence of a strong amino acid, such

as cysteine.

118

Chapter V

Reactions of Dirhodium Bis-acetate with DNA

119

1. Introduction

Inhibition of DNA replication can be achieved by the interaction of DNA
with an agent that binds either by covalent or non-covalent interactions.
Noncovalent antitumor compounds include intercalators and minor groove
binding molecules. lntercalators exhibit a high degree of sequence specificity
and are composed of functional groups attached to a planar aromatic
chromophore. Anthracyclines, which demonstrate anticancer activity, consist of
a sugar group that can position itself in the minor groove by hydrogen bonding to
base pair edges (Figure 55). Likely targets include CpG and TpG, as a result of
the favorable binding site on the exocyclic N2 of guanosine. lntercalation of
these types of compounds causes the minor groove to widen, resulting in an
overall distortion of the helix 5.

The minor groove of DNA is an optimal target for antitumor compounds. A
molecule that binds to the minor groove is able to evade the major groove repair
enzymes. A majority of these agents inhibit transcription and/or replication and
have been shown to interfere with DNA topoisomerase II (introduces negative
supercoils in DNA, and is necessary for DNA synthesis to occur). Although
minimal structural distortion occurs, more subtle alterations to recognition signals
for regulatory proteins may occur, causing a lack of sequence recognition by the
protein, thereby inhibiting DNA replication and/or repair. Two examples of the
drugs that cause this activity to occur are netropsin and anthramycin 605‘.

Netropsin (Figure 56) is a naturally occurring antibiotic that exhibits both

120

anticancer and antiviral activity. This activity stems from its ability to bind the

minor groove of

x N\
‘ _o .
H3C o -------- NH; \\ T-A
/ / N
S ar
/ NH ....... N / ”9
\/N
/N “
Sugar ’0
\ ,v \
r... ”fl
2 \ \ N C'G
/ \ Sugar
N ------- NH
/N
/N \ \.‘
Sugar ’10 ......... NH'{

Figure 55. Potential hydrogen-bond interactions with base edges in AT and GO

base pairs. Donors and acceptors are indicated by direction of the arrowheads.

DNA, specifically AT rich regions. A crystal structure of a netropsin adduct has
been obtained with the decamer d(CGCAATTGCG) (Figure 57b). Netropsin has
also been crystallized with several other strands of DNA, e. g.
d(CGCGAATTCGCG)2 and d(CGCAAAT'l'l'GCG)2. Information obtained from
crystal structures of these drugs is important in the design and synthesis of
analogues with altered sequence specificity, for example preferential GC base

pair binding 60.

121

O .
CH3 0 W NH2
NM
H NH2

Figure 56. The minor groove binding drug Netropsin.

Anthramycin (Figure 58) is a naturally occurring antiutmor antibiotic that
has been crystallized with the decamer d(CCAACGTTGG) (Figure 57a). The
drug was found to covalently bind in a 2:1 ratio with DNA and to target the 5’-
GGT sequences at each end of the duplex. Binding causes an overall decrease
in the width of the minor groove. Like netropsin, amide tails participate in
hydrogen bonding to base pair edges. From these structures, and others like
them, it is possible to identify two important factors in the binding of minor groove

drugs, these are groove width alterations and hydrogen bonding to base edges“.

 

Figure 58. The minor groove binding drug Anthramycin.

122

.«Aooootioooovuaaaeac 3 Em «Aootoofioovnégsescm a so 2385 .995

3V

0
u\
l
\(l \ft'
alto. o v ... "It
C
/‘

/
\
/

a‘\/ \., L

\
I r
..J. t ! Fw-U RU ~/ u

.3 2:9".

/ his o\.ul~ o. -09 o
— \J\ . . ‘k ‘g/ \.
/o\ _

 

123

Macroscopic DNA properties, such as bending of the helix upon drug
binding, remain to be assessed and understood at the molecular level. This is
important for minor groove binding drugs since these interactions may be
connected to the drugs’ ability to interfere with gene regulatory proteins. In some
cases, these interactions induce large structural changes in DNA.

Unlike the non-covalent binding compounds, very few covalent complexes
have been crystallized with DNA. Covalent complexes that bind DNA include cis-
platin, mitomycin nitrogen mustards, and anthramycin. In this work, we hope to
add dirhodium bis-acetate to this relatively short list. One reason for the paucity
of these structures is that crystallization of covalent complexes has proven to be
exceedingly difficult. To date, there are only two known structures involving a
covalently bound transition metal complex with DNA 5. The most recognized
structure consists of the antitumor active cis-platin bound to the DNA dodecamer
d(CCTCTGGTCTCC)-(GGAGACCAGAGG) (Figure 59). Upon binding to this
double stranded oligonucleotide, cis-platin forms 1, 2 intrastrand cross-links
between two neighboring guanine bases on the same strand. The formation of
this cross-link causes the DNA to become severely distorted, resulting in a ~40°
bend in the DNA. This structure is the first example of a covalently bound
transition metal compound with double stranded DNA 63.

The ability to crystallize this type of complex is important for future Pt drug
design, and for a general understanding of the mechanism by which these
compounds kill tumor cells. Metallation of the DNA strand causes the duplex to

bend towards the major groove and leads to the formation of a type of N8

124

 

Figure 59. Structure of cis-platin-d(CCTCTGGTCTCC)(GGAGACCAGAGG).

125

hybrid. This type of hybrid may serve as a cellular recognition signal for certain
proteins. This phenomena is thought to occur in the case of other A/B hybrid
DNA strands that contain recognition sequences for certain proteins.

Part of cis-platin’s potent activity is linked to its ability to distort DNA which
causes proteins, such as HMG1, to bind these sites thereby diverting the protein
from normal activities. In addition, these proteins serve to shield the cis-platin
adducts from excision repair. The binding of cis-platin to DNA causes the helix to
bend, thereby exposing the minor groove. This exposure allows proteins, such
as HMG/SRY, to bind via their hydrophobic residues which further disrupts base
pair stacking. HMG binding to the distorted DNA causes the bend angle to
increase by 30-50° which alleviates some of the strain placed on the cis-platin
molecule 6.

Although the modes of binding are different, the covalent and non-
covalent drugs seem to share the same general properties. Coordination of cis-
platin to the GG nucleotides results in hydrogen bonding between one of the
platinum ammine ligands and the terminal oxygen of a guanosine terminal
phosphate. Metallation of the DNA also resluts in destacking of the bases and a
distortion of the platinum coordination geometry. Another consequence of cis-
platin binding is major groove compression and minor groove widening to
resemble A DNA. It is clear that both classes of antitumor compounds possess
the ability to hydrogen bond and cause groove width distortion 5'63.

Another type of cis-platin binding to DNA was observed when the drug

was diffused into pregrown crystals of d(CGCGAATTCGCG)2. Crystals were

126

grown of the sequence, and cis-platin was added to the crystals which led to its
incorporation into the structure. The X-ray data that were obtained (Figure 60)
indicate that DNA did not become distorted, but there was a clear indication of
cis-platin’s affinity for guanine 64.

To understand the nature of cis-platin binding on a smaller scale, smaller
di- and tri-nucleotides were used. Crystal structures were obtained with the
dinucleotide d(pGpG) (Figure 61) and the trinucleotide d(CpGpG) (Figure 62).
Lippard’s d(pGpG) structure shows the formation of two Pt-N bonds to the N(7)
atoms of the two adjacent guanine nucleotides 65. Base stacking is disrupted by
the binding of cis-platin as indicated by the dodecamer crystal structure and NMR
modeling studies 66. Since cis-platin was known to preferentially bind guanine,
Reedin and coworkers opted to study the effect of platinum binding on the
stacking interactions of neighboring bases. Tripeptides were designed that
contain the GpG moiety and in the case of d(CpGpG), the oligonucleotide was
crystallized with bound cis-platin. It was found that hydrogen bonding between
the amino groups and 06 occurs. The cytosine forms hydrogen bonds with a
cytosine on a neighboring molecule 67. These studies revealed that even simple
oligonucleotides exhibit the disruption of base pair stacking and hydrogen

bonding that occur with the larger pieces of DNA.

127

 

Figure 60. Crystal structure of cis-platin bound to d(CGCGAATTCGCG).
Arrows indicate the three cis-platin molecules bound to the DNA strand.

128

 

Figure 62. ORTEP stereoview of the intrastrand cis-platin
d(CpGpG) adduct.

129

2. Experimental

Each DNA strand was obtained from Yale and purified by anion exchange
HPLC on a Perkin Elmer Diode array HPLC equipped with a Source Q15 Anion
exchange column. After purification, the DNA was concentrated on a DEAE
(Diethylaminoethyl) Cellulose column, then desalted via either G-25 Size
Exclusion Sephadex column or by using Centricon SR-3. The resulting strands
were then annealed to their complements and set up in boxes for crystallization.
A. Reactions
[Rh2(O2CCH3)2(CH3CN)5][BF4]2 + TGC GTT AAC GC (1)

A solution of [Rh2(O2CCH3)2(CH3CN)6][BF4]2 (1.9 mg, 3 pmol) in 50 pL of
DD H20 was added to a solution of 11-mer (3 pmol) in 150 pL DD H20. The
resulting solution immediately turned bright orange upon addition of the
dirhodium bis-acetate precursor. The mixture was heated to 70°C for 52 h. At
the end of the reaction time, the solution was greenish-brown color.
[Rh2(O2CCH3)2(CH3CN)5][BF4]2 + CGC AAT TGC G (2)

A solution of [Rh2(O2CCH3)2(CH30N)6][BF4]2 (1.9 mg, 3 pmol) in 50 pL of
DD H20 was added to a solution of 10-mer (3 pmol) in 150 uL DD H20. The
resulting solution immediately turned a bright reddish-orange color upon addition
of the dirhodium bis-acetate. The mixture was heated to 70°C for 52 h during
which time the color turned greenish-brown.
[Rh2(O2CCH3)2(CH3CN)5][BF4]2 + CCT CTG GTC TCC (3)

A solution of [Rh2(O2CCH3)2(CH3CN)6][BF4]2 (3.8 mg, 6 pmol) in 50 pL of

DD H20 was added to a solution of the GG 12-mer (3 pmol) in 200 p.L DD H20.

130

The resulting solution immediately turned a bright orange upon addition of the
dirhodium bis-acetate. The volume was increased to 400 pL, and the resulting
solution was heated to 70°C for 72 h. At the end of the reaction time, the solution
was a greenish-brown color.
[Rh2(O2CCH3)2(CH3CN)5][BF4]2 + CCT CTA ATC TCC (4)

A solution of [Rh2(O2CCH3)2(CH3CN)6][BF4]2 (3.8 mg, 3 pmol) in 50 pL of
DD H20 was added to a solution of the AA 12-mer (3 pmol) in 150 pL DD H20.
The resulting solution immediately turned a bright reddish—orange upon addition
of the dirhodium bis-acetate. The volume was brought to 400 pL, and the
resulting solution was heated to 70°C for 72 h. At the end of the reaction time,
the solution was a greenish-brown color.
[Rh2(O2CCH3)2(CH3CN)3][BF4]2 + TCT AAT CT (5)

A solution of [Rh2(O2CCH3)2(CH30N)6][BF4]2 (3.0 mg, 4 pmol) in 50 pL of
DD H20 was added to a solution of the 8-mer (2 pmol) in 150 pL DD H20. The
resulting solution immediately turned bright orange upon addition of the
dirhodium bisacetate. The volume was increased to 400 pL, and the resulting
solution was heated to 37°C for 72 h. At the end of the reaction time, the solution
was a reddish-purple color.
[Rh2(O2CCH3)2(CH3CN)5][BF4]2 + CTC TAA TCT T (6)

A solution of [Rh2(O2CCH3)2(CH3CN)6][BF4]2 (3.0 mg, 4 pmol) in 50 pL of
DD H20 was added to a solution of the 10-mer (2 pmol) in 150 pL DD H20. The
resulting solution immediately turned bright orange upon addition of the

dirhodium bisacetate. The volume was increased to 400 pL, and the resulting

131

solution was heated to 37°C for 72 h. At the end of the reaction time, the solution
was a reddish-purple color.
[Rh2(O2CCH3)2(CHacN)5][BF4]2 + GAG ATT AGA GA (7)

A 2 pmol quantity of the 11-mer was dissolved in 400 pL of degassed DD
H20. To this was added 65 pL of a 0.04 M degassed
[Rh2(O2CCH3)2(CH3CN)6][BF4]2 solution (1:1.3 ratio). The resulting pink solution
was heated at 37°C for 72 h. At the end of the reaction time, the solution was a
reddish-purple color.
[Rh2(O2CCH3)2(CH3CN)5][BF4]2 + CCT CTA ATC TCC (8)

A 4 pmol aliquot of the 12-mer was dissolved in 400 pL of degassed DD
H20. To this was added 65 pL of a 0.04 M degassed
[Rh2(O2CCH3)2(CH30N)5][BF4]2 solution (1:1.3 ratio). The solution, which
immediately turned orange, was heated at 37°C for 72 h. At the end of the
reaction time, the solution was a reddish-purple color.
B. Anion Exchange HPLC Purification.
1. Unmetallated Strands

Each unmetallated strand was purified using a NaOH/NaCl eluant system.
The DNA was dissolved in a 10 mM NaOH solution and purified one micromole
at a time using a gradient. Eluent A: 10 mM NaOH, 0.2 M NaCl, eluant B: 10 mM
NaOH, 1 M NaCl.
GGA GAC CAG AGG was purified by the following gradient: 5 minutes gradient
36% B, 60 minutes gradient 41% B, 5 minutes gradient 100% B, 5 minutes

isocratic 100% B, 5 minutes gradient 0% B.

132

GGA GAT TAG AGG was purified by the following gradient: 5 minutes gradient
40% B, 60 minutes gradient 45% B, 5 minutes gradient 100% B, 5 minutes
isocratic 100% B, 5 minutes gradient 0% B.
GAG ATT AGA A was purified by the following gradient: 5 minutes gradient 30%
B, 60 minutes gradient 35% B, 5 minutes gradient 100% B, 5 minutes isocratic
100% B, 5 minutes gradient 0% B.
GAG TI'A GA was purified by the following gradient: 5 minutes gradient 30% B,
60 minutes gradient 35% B, 5 minutes gradient 100% B, 5 minutes isocratic
100% B, 5 minutes gradient 0% B.
GAG ATI' AGA GA was purified by the following gradient: 5 minutes gradient
35% B, 60 minutes gradient 40% B, 5 minutes gradient 100% B, 5 minutes
isocratic 100% B, 5 minutes gradient 0% B.
TGG AGA TTA GAG G was purified by the following gradient: 5 minutes gradient
66% B, 60 minutes gradient 71% B, 5 minutes gradient 100% B, 5 minutes
isocratic 100% B, 5 minutes gradient 0% B.
AGG AGA TI'A GAG G was purified by the following gradient: 5 minutes
gradient 66% B, 60 minutes gradient 71% B, 5 minutes gradient 100% B, 5
minutes isocratic 100% B, 5 minutes gradient 0% B.
2. Metallated Strands

Each reacted strand was purified using an acetate buffer system. The
DNA was purified one micromole at a time using a gradient. Eluent A: 0.02 M

N802CCH3, 20% CHgCN, eluant B: 0.02 M N302CCH3, 1.2 M KCI, 20% CH30N.

133

TGC GTT AAC GC was purified by the following gradient: 5 minutes gradient
40% B, 60 minutes gradient 45% B, 5 minutes gradient 100% B, 5 minutes
isocratic 100% B, 5 minutes gradient 0% B.

GCG TI'A ACG C was purified by the following gradient: 5 minutes gradient
40% B, 60 minutes gradient 45% B, 5 minutes gradient 100% B, 5 minutes
isocratic 100% B, 5 minutes gradient 0% B.

CCT CTG GTC TCC was purified by the following gradient: 5 minutes gradient
30% B, 30 minutes gradient 40% B, 5 minutes gradient 100% B, 5 minutes
isocratic 100% B, 5 minutes gradient 0% B.

CCT CTA ATC TCC was purified by the following gradient: 5 minutes gradient
30% B, 30 minutes gradient 40% B, 5 minutes gradient 100% B, 5 minutes
isocratic 100% B, 5 minutes gradient 0% B.

CTC TAA TCT T was purified by the following gradient: 5 minutes gradient 35%
B, 30 minutes gradient 40% B, 5 minutes gradient 100% B, 5 minutes isocratic
100% B, 5 minutes gradient 0% B.

CTC AAT CT was purified by the following gradient: 5 minutes gradient 30% B,
30 minutes gradient 35% B, 5 minutes gradient 100% B, 5 minutes isocratic
100% B, 5 minutes gradient 0% B.

TCT CTA ATC TC was purified by the following gradient: 5 minutes gradient 28%
B, 30 minutes gradient 33% B, 5 minutes gradient 100% B, 5 minutes isocratic

100% B, 5 minutes gradient 0% B.

134

CCT CTA ATC TCC was purified by the following gradient: 5 minutes gradient
30% B, 30 minutes gradient 37% B, 5 minutes gradient 100% B, 5 minutes
isocratic 100% B, 5 minutes gradient 0% B.
C. DEAE column
The purpose of this column is to concentrate a large volume of sample

into a few milliliters. Diethylaminoethyl cellulose (DEAE) resin is equilibrated with
10 mM TRIS buffer. The DNA sample is diluted up to four times its original
concentration with 10 mM TRIS, and loaded onto a 10 cm x 1.5 cm column. The
DNA is then eluted from the column by a 10 mM TRIS, 1 M NaCl solution and
collected in 1 mL fractions that are screened by UV at 265 nm.
D. G-25 Sephadex column

The G-25 Sephadex column is used to desalt DNA fractions after the DEAE
column has been run. Salty DNA is loaded onto the column, eluted with DD H20
and collected in 1 mL fractions that are screened by UV absorbance at 265 nm.
The samples are then dried in a centrivap to concentrate.
E. Centricon SR-3

Centricon SR-3 were used to simultaneously concentrate and desalt larger
strands of DNA (12bp and larger). Samples were placed in the Centricons and
centrifuged at 6500 RPM until minimal volume was achieved, usually after 2h.

The samples were washed with two portions of 2 mL DD H20 to desalt. The final

volume was between 40-100 pL.

135

F. Concentrations

The concentration of the samples was determined by Beer’s law: A=bec.
Molar absorptivity coefficients for each DNA strand were obtained from the
biopolymer calculator on the Yale web site. The mOIar absorptivity coefficient for
dirhodium bis-acetate transitions were determined by UV analysis at 265 nm
using a known concentration of the sample. The samples were dissolved in 200-
400)). L of DD H20 for a final concentration of 10 mg/mL.
G. Annealing

Annealing the reacted strands to their complements was achieved by
heating the samples to 100°C followed by a gradual cooling to room temperature.
H. Crystallization screens

The following crystallization screens were obtained from the Hampton

Research Catalog:

Nucleic Acid Mini-Screen (NAM)

 

 

1. 10% MPD, pH 5.5, 20 mM Cobalt Hexamine, 20 mM MgCI2

2. 10% MPD, pH 5.5, 20 mM Cobalt Hexamine, 80 mM NaCl, 20 mM MgCl2
3. 10% MPD, pH 5.5, 20 mM Cobalt Hexamine, 12 mM NaCl, 20 mM MgCl2
4. 10% MPD, pH 5.5, 20 mM Cobalt Hexamine, 40 mM LiCl, 20 mM MgCl2

5. 10% MPD, pH 6.0, 12 mM Spermine tetra-HCI, 80 mM KCI, 20 mM MgCl2
6. 10% MPD, pH 6.0, 12 mM Spermine tetra-HCI, 80 mM KCI

7. 10% MPD, pH 6.0, 12 mM Spermine tetra-HCI, 80 mM NaCl, 20 mM MgCl2

8. 10% MPD, pH 6.0, 12 mM Spermine tetra-HCI, 80 mM NaCl

136

9. 10% MPD, pH 6.0, 12 mM Spermine tetra-HCI, 80 mM NaCl, 12 mM KCI,
20 mM MgCl2

10.10% MPD, pH 6.0, 12 mM Spermine tetra-HCI, 12 mM NaCl, 80 mM KCI

11.10% MPD, pH 6.0, 12 mM Spermine tetra-HCI, 80 mM NaCl, 20 mM BaCI2

12.10% MPD, pH 6.0, 12 mM Spermine tetra-HCI, 80 mM KCI, 20 mM BaCI2

13.10% MPD, pH 6.0, 12 mM Spermine tetra-HCI, 80 mM SrCI2

14.10% MPD, pH 7.0, 12 mM Spermine tetra-HCI, 80 mM KCI, 20 mM MgCl2

15.10% MPD, pH 7.0, 12 mM Spermine tetra-HCI, 80 mM KCI

16.10% MPD, pH 7.0, 12 mM Spermine tetra-HCI, 80 mM NaCl, 20 mM MgCI2

17.10% MPD, pH 7.0, 12 mM Spermine tetra-HCI, 80 mM NaCI

18.10% MPD, pH 7.0, 12 mM Spermine tetra-HCI, 80 mM NaCl, 12 mM KCI,
20 mM MgCI2

19.10% MPD, pH 7.0, 12 mM Spermine tetra-HCI, 12 mM NaCl, 80 mM KCI

20.10% MPD, pH 7.0, 12 mM Spermine tetra-HCI, 80 mM NaCl, 20 mM BaCI2

21.10% MPD, pH 7.0, 12 mM Spermine tetra-HCI, 80 mM KCI, 20 mM BaCI2

22.10% MPD, pH 7.0, 12 mM Spermine tetra-HCI, 40 mM LiCl, 80 mM SrCl2,
20 mM MgCl2

23.10% MPD, pH 7.0, 12 mM Spermine tetra-HCI, 40 mM LiCl, 80 mM SrCI2

24.10% MPD, pH 7.0, 12 mM Spermine tetra-HCI, 80 mM SrCl2, 20 mM MgCl2

Where MPD= 2-methyl-2,4-pentane diol, the buffer consists of 40 mM sodium

cacodylate, and the dehydrant is 35% v/v MPD.

137

Natrix Formulation (with slight modification)

1.

2.

0.01 M Mg Chloride, 0.05 M Sodium Citrate pH 5.6, 2.0 M Lithium Sulfate
0.01 M Mg Acetate, 0.05 M Sodium Citrate pH 5.6, 2.5 M Ammonium Sulfate
0.1 M Mg Acetate, 0.05 M Sodium Citrate pH 5.620% MPD

0.2 M KCI, 0.01 M Ma Sulfate, 0.05 M Sodium Citrate pH 5.6, 10% PEG 400
0.2 M KCI, 0.01 M Mg Chloride, 0.05 M Sodium Citrate pH 5.6, 5% PEG 8000
0.1 M Ammonium Sulfate, 0.01 M Mg Chloride. 0.05 M Sodium Citrate pH
5.6, 20% PEG 2000

0.02 Mg Chloride, 0.05 M Sodium Cacodylate pH 6.0, 15% lsopropanol

0.005 M Mg Sulfate, 0.1 M Ammonium Acetate, 0.05 M Sodium Cacodylate
pH 6.0, 0.6 M NaCl

0.1 M KCI, 0.01 M Mg Chloride, 0.05 M Sodium Cacodylate pH 6.0, 10% PEG
400

10.0.005 M MgSulfate, 0.05 M Sodium Cacodylate pH 6.0, 5% PEG 4000

11.0.01 M Mg Chloride, 0.05 M Sodium Cacodylate pH 6.0, 1.0 M Lithium Sulfate

12.0.01 M Mg Sulfate, 0.05 M Sodium Cacodylate pH 6.0, 1.8 M Lithium Sulfate

13.0.015 M Mg Acetate, 0.05 M Sodium Cacodylate pH 6.0, 1.7 M Ammonium

Acetate

14.0.1 M K Chloride, 0.025 M Mg Chloride, 0.05 M Sodium Cacodylate pH 6.0,

15% lsopropanol

15.0.04 M Mg Chloride, 0.05 M Sodium Cacodylate pH 6.0, 5% MPD

16.0.04 M Mg Acetate, 0.05 M Sodium Cacodylate pH 6.0, 30% MPD

138

 

17.0.2 M KCI, 0.01 M Ca Chloride, 0.05 M Sodium Cacodylate pH 6.0, 10% PEG
4000

18.0.01 M Mg Acetate, 0.05 M PIPES pH 6.5, 1.3 M Lithium Sulfate

19.0.01 M Mg Sulfate, 0.05 M PIPES pH 6.5, 2.0 M Ammonium Sulfate

20.0.1 M Ammonium Acetate, 0.015 M Mg Acetate, 0.05 M PIPES pH 6.5, 10%
lsopropanol

21.0.2 M KCI, 0.005 M Mg Chloride, 0.05 M PIPES pH 6.5, 10% 1,6 hexanediol

22.0.08 M Mg Acetate, 0.05 M PIPES pH 6.5, 15% PEG 400

23.0.2 M KCI, 0.01 M Mg Chloride, 0.05 M PIPES pH 6.5, 10% PEG 4000

24.0.2 M Ammonium Acetate, 0.01 M Ca Chloride, 0.05 M PIPES pH 6.5, 10%
PEG 4000

25.0.08 M Mg Acetate, 0.05 M PIPES pH 6.5, 30% PEG 4000

26.0.2 M KCI, 0.1 M Mg Acetate, 0.05 M PIPES pH 6.5, 10% PEG 8000

27.0.2 M Ammonium Acetate, 0.01 M Mg Acetate, 0.05 M PIPES pH 6.5, 30%
PEG 8000

28.0.05 M Mg Sulfate, 0.05 M HEPES pH 7.0, 1.6 M Lithium Sulfate

29.0.01 M Mg Chloride, 0.05 M HEPES pH 7.0, 4.0 M Lithium Chloride

30.0.01 M Mg Chloride, 0.05 M HEPES pH 7.0, 1.6 M Ammonium Sulfate

31.0.005 M Mg Chloride, 0.05 M HEPES pH 7.0, 25% PEG 400

32.0.2 M KCI, 0.01 M Mg Chloride, 0.05 M HEPES pH 7.0, 20% 1,6 hexanediol

33.0.2 M Ammonium Chloride, 0.01 M Mg Chloride, 0.05 M HEPES pH 7.0, 30%
1,6 hexanediol

34.0.1 M KCI, 0.005 M Mg Sulfate, 0.05 M HEPES pH 7.0, 15% MPD

139

 

 

35.0.1 M KCI, 0.01 M Mg Chloride, 0.05 M HEPES pH 7.0, 5% PEG 400
36.0.1 M KCI, 0.01 M Ca Chloride, 0.05 M HEPES pH 7.0, 10% PEG 400
37.0.2 M KCI, 0.025 M Mg Sulfate, 0.05 M HEPES pH 7.0, 20% PEG 400
38.0.2 M Ammonium Acetate, 0.15 M Mg Acetate, 0.05 M HEPES pH 7.0, 5%
Peg 4000
39.0.1 M Ammmonium Acetate, 0.02 M Mg Chloride, 0.05 M HEPES pH 7.0, 5%
PEG 8000
40.0.01 M Mg Chloride, 0.05 M TRIS HCI pH 7.5, 1.6 M Ammonium Sulfate
41.0.1 M KCI, 0.015 M Mg Chloride, 0.05 M TRIS HCI pH 7.5, 10% PEG 400
42.0.01 M Mg Chloride, 0.05 M TRIS HCI pH 7.5, 5% lsopropanol
43.0.01 M Mg Chloride, 0.05 M Ammonium Acetate, 0.05 M TRIS HCI pH 7.5,
10% MPD
44.0.2 M KCI, 0.05 M Mg Chloride, 0.05 M TRIS HCI pH 7.5, 10% PEG 4000
45.0.025 M Mg Sulfate, 0.05 M TRIS HCI pH 8.5, 1.8 M Ammonium Sulfate
46.0.005 M Mg Sulfate, 0.05 M TRIS HCI pH 8.5, 35% 1,6 hexanediol
47.0.1 M KCI, 0.01 M Mg Chloride, 0.05 M TRIS HCI pH 8.5, 30% PEG 400
48.0.01 M Ca Chloride, 0.2 M Ammonium Chloride, 0.05 M TRIS HCI pH 8.5,
30% PEG 4000
Biological Buffers used:
PIPES-1,4-Piperazinediethanesulfonic acid]
HEPES- N-[2—Hydroxyethyllpiperazine-N’-[2-ethanesulfonic acid]

TRIS-tris[Hydroxymethyllaminomethane

140

These screens were used in the Hanging Drop method of crystal growth. 4x6
well-plates were used and 500 pL of buffer or dehydrant was placed each well.
On a plastic microscope cover slip, 2 pL of a 10 mg/mL DNA solution and 2 pL
buffer were placed in a drop. The cover slip was then suspended over the well
(Figure 63). Slow diffusion then takes place which can eventually lead to crystal

growth.

Figure 63. Hanging Drop Method of crystal growth

\7 ’
U +—— Dehydrant

Each of the DNA strands was specifically designed to have a single

 

3. Results and Discussion

binding site, namely AA or G6 (Table 10). The 10 mersc and 11 mersc were
created to be self complementary so that each strand in the duplex would have a
potential binding site for the rhodium bis-acetate. Several of the DNA strands
have overhangs of one base pair, which is useful as an anchor for packing
interactions in crystal growth. The AA binding site is preferred due to higher

product yields and shorter reaction times with [Rh2(O2CCH3)2(CH30N)612”.

141

 

Table 10. DNA strands designed for crystallization studies.

 

 

Strand Binding site Name
TGC GTT AAC GC AA 11 mersc
CG CAA TTG CGT
GCG TTA ACG C AA 10 mersc
CGC AAT TGC G
CCT CTG GTC TCC GG GG 12mer
GGA GAC CAG AGG
CCT CTA ATC TCC AA AA 12 mer
GGA GAT TAG AGG
CTC TAA TCT T AA 10 mer
GAG ATT AGA A
CTC AAT CT AA 8 mer
GAG TTA GA
TCT CTA ATC TC AA 11 mer
GA GAT TAG AGA
CC TCT AAT CTC C AA 2A-13 mer
TGG AGA TTA GAG G
CC TCT AAT CTC C AA 3A-13 mer

AGG AGA TTA GAG G

 

142

At the outset of this research, it was believed that the DNA product could
only be formed in good yields by heating the reaction to fairly high temperatures,
i.e. 70°C for 72 hours. The product of the high temperature reaction eluted from
the HPLC column only at the highest salt concentrations, and was obtained in
very low yields (around 30%). The use of buffers and different reaction times did
not improve the yields. It seemed unusual that the reaction would proceed only
under the harshest of conditions; indeed this suspicion was confirmed by the
results of a MALDI mass spectrometry experiment performed on the reacted 12
mer GG product. The results showed that the high temperature product is an
adduct bound to several dirhodium units.

The antitumor active compound, Rh2(0Ac)4(H20)2, inhibits DNA replication
under normal cellular conditions, i.e. 37°C. Rethinking the strategy resulted in a
radical change in protocol. Obviously, the products that form at biological
temperature should be the adducts that cause a halt in cancer cellular replication,
thus the reaCtion was performed at 37°C with the 8- and 10-mers in hopes of
obtaining the biologically relevant product. HPLC purification revealed a large
metallated peak eluting at about 30-40 % B along with a small peak eluting at
high salt concentrations (Figure 64). The new peak was collected and desalted
according to the usual protocols then submitted for mass spectrometric analysis.
The results showed that the rhodium bis-acetate is bound to the DNA in the
expected 1:1 ratio (Figure 65). This is the first definitive proof that the dirhodium
bis-acetate is intact and bound to a single site on the DNA strand, most likely the

target site. Based on charge considerations, the phosphate backbone seems to

143

 

 

be a likely target as well, however, there are no structures in the NADB where a
charged group is in close proximity with the DNA backbone. Due to the lack of
binding of charged antitumor complexes to the phosphate groups on DNA, it
seems likely that the dirhodium bis-acetate is bound to purine sites on the strand.

The MALDI mass spectroscopic data shows that the adduct
Rh2(O2CCH3)2-d(TCTCTAATCTC) was formed in the milder reaction. Four
assignable peaks appear in the spectrum. The peak at 3567 corresponds to the
[DNA + Rh2(O2CCH3)2 — 3H]' peak. Another peak appears at 3506 which
corresponds to [DNA + Rh2(O2CCH3) - 4H]’. [DNA + Rh2 — 5H]' occurs at 3446,
and [DNA — H]' occurs at 3245. These results are clear evidence that the
reaction proceeds as expected at 37°C.

Attempts to crystallize these Rh-DNA complexes has proven to be difficult.
Each DNA strand has been set up in boxes with the crystallization screens
previously mentioned. Most of the DNA strands produce a precipitate upon slow
diffusion with the dehydrant. The 70°C reactions produced unusual looking
precipitates, which appeared to be brown in color. Small microcrystalline
precipitates of the AA-12 mer appeared, but attempts to maximize their size were
unsuccessful. The 10-mersc formed multiple, thin platelets in the NAM as did the
11-mersc. No single crystals have been observed to date. At 4°C, the 10-mer
formed very tiny crystals with the NAM but produced nothing at room
temperature. The 11-mer formed tiny crystals that were larger than the 10-mer,

so the conditions were optimized in attempts of growing larger crystals. The

144

 

 

 

other DNA strands formed precipitates in some of the conditions, but no

crystalline materials are visible at this time.

145

 

7.";

.c .8 .2

0.5 a. B8; 55-: 3v 95 5 mm .2 0.2 2 .328; 855-2 E a 3985526 od: .3 2:9“.

3V

2 J

 

 

 

 

E

 

146

 

E: mom

 

 

 

 

 

E: mom

 

 

 

......m- 3m + <29 Qaflm firm- ~8<ovmzm + <on 8.88
...E- ~8<ovam + <29 9 8882.8 38..” a same . as 2 o» 02 <5 Be a ms. .3 2:9“.

 

 

 

.N‘EV «an:
8.3 8m” 8.8 8.3 8mm 8.8 8.8 38
f
11.2.51? Eff/(J; ._ 15!
x $356.. H1 .11.!) .. E (J ISON
_ 2}) V...) \g\. ,7. 2 x
. if...
8 8 I89.
6
m as
.0
g
a .88
h
Ioooow
....
my
.9
l I l I - I - I v -88?

 

 

 

 

munoo

147

4. Conclusions

It is now known that upon reaction with DNA,
[Rh2(O2CCH3)2(CH3CN)6][BF4]2 forms a 1 :1 adduct as expected based upon
model studies with nucleotides. The presence of two acetates and two rhodium
atoms as confirmed by MALDI-MS revealed that the reaction does proceed at
37°C. The most likely site of bindning is to the AA target site, however, without a
crystal structure or definitive 2D-NMR data, it is not possible to draw a firm
conclusion. Crystallization studies look promising, with the formation of tiny
crystals of the AA 11 mer as a starting point. Regrettably, large single x-ray
quality crystals have not been obtained to date with the DNA strands mentioned

in this chapter.

148

 

 

thin-m!

 

APPENDIX

149

 

 

 

 

 

 

 

 

 

 

 

 

 

 

I f I *r—r T I f f I T I r '
C 7 5 5 C 3 3 9’
b)
1
1
l‘ I
l ‘ l
. I i g
L] f A fiL’J/V 411inij ML...
"f g I ‘—T **r "‘ r ‘ r I r 1 I 1" ‘- " T "’”[—T—-Y_"" 7' ‘] Y " T V I V" 7 r. l ff ' Y 1 '
9 8 7 6 5 G 3 2 pp

Figure A1. Methionine/AMP NMR reaction. a) methionine/AMP standard,

b) 1H NMR at time = 0, c) 1H NMR at time =15 min, d) 1H NMR at time = 2 h,

e)1H NMR at time = 4 h, f) 1H NMR at time = 24 h.

150

“I

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

it.--

 

l
I I .
l [I 1I
I... .. J UM.) fJUL“
9 ..... ..fi ; fig Wig;
l
d)
l
l
[A J EMU-.10.
SJ“: 2 .4. "fia' ...

151

6)

 

 

 

 

 

 

 

 

 

 

I 1 IL} 1 ;' / i
"le AV [I \Wfifv’ .‘ IHJ ' / \_J* / \_“J K1
"T—T_.~T_-'—r —Y—‘ W "Y—'T—_'—1—"T_ f j Y_' Y T i Ta I fl 1 fl
9 I 7 6 5 O 3 2 p;-

152

 

l .

l
1

“f‘t’fi’t"f"‘r 1 r fir T '**T v
8 7 O 5 4 3 2 ppm

 

 

 

 

 

 

 

l
l
l
i
b) t
I
".
l
.11 I .
......)L. J.“ Jr ‘\”'JUL_J' L4, WLJUJ ¥_____..
ffi—‘Tfi—r—r—vzq—H'n'qn—r" "1—7" 1" 'I‘ v I I"r T r I l v I ' T I ..
I 8 7 6 5 4 3 2 ppm

Figure A2. Glutathione/Amp NMR reaction. a) glutathione/AMP standard,
b) 1H NMR at time = 0, c) 1H NMR at time =15 min, d) 1H NMR at time = 2 h,
e) 1H NMR at time = 4 h, f) 1H NMR at time = 24 h.

153

d)

r T T Y
i
I

 

”..ka -

154

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

l1I 1k 1
T "I “T— ‘ Ifl —“—T—‘r ' fi— 1 Tfi 1 1 T r I 1 T
9 O 7 6 5 3 2 n-

155

 

 

 

 

 

 

 

 

 

i
1 I ’1 M I“ If
......___.J'_..z- _J ##2/ K..‘.JL___/'1L JU‘LL
"‘z'fi * m I ......Wfi._. ‘ "'r—T— I
9 I 7 6 5 4 3 p:-
b) it
1
1
11
1
-..—.....JL 1' UUJ‘M
1 fiJIL J ' It ‘ A. Q“;
""T --_._.,_--T , _._,__._.. . _. , 1—’“*"*—*—E , ...—T.., . , , f. . f. r— r . v , I
! O 7 G 5 C 3 3 D:-

Figure A3. Cysteine/AMP NMR reaction. a) cysteine/AMP standard, b) 1H NMR
at time = 0 c) 1H NMR at time =15 min, d) 1H NMR at time = 2 h, e) 1H NMR at
time = 4h, f) 1H NMR at time = 24 h.

156

 

 

 

 

 

 

 

C)

d)

L...

 

 

 

,1

 

 

157

 

 

 

 

 

 

 

 

"*‘V r-I-r-z‘T—Y—T—V—r—r‘ 7 '

11 10

 

 

 

 

 

 

 

 

 

 

 

 

a) 1

 

1, I
11 1 1 1 1
_I 1 1 1 1
111 1 1 11 1 1.
1 l 11 '1 l 1
1‘ l 1111-111 11 1 1
1 1 1 1! 11
I I . ,- I
_, 1 1 .
1 1 1 1| 1
1 1 ,7'11 1a
.‘ I ’ K.
1 // 1H1»! Xxx-A»- -JK . ,w,
\ 4 A‘ ~~ r‘nrw r— JK VJ - 1 fl‘ 1N \I—W‘AA
h l 3 H 311 3b 24 H 2.0 lb lb 14 1. ppm
b)
I
I 1
1
1 1 11
i1 11
‘ 1 . 1 11
1 ‘ 1.4 I
1,1 1 I 1 1 1 1
1 1‘11 I 1 I 1
1 ' .1 1 1 1 1 1
1 1 11 1 1 11 ‘1 1
I l . I 1 1 1 11
1‘ 1 1 11 1' 1
1111 i I. I ‘1
I . I 2 x
1 1 ‘\ \
.7 X \ \M \n
1‘ .--“ 11\,[I~‘1"1VJ1 'K/‘K'xv‘x'xxf A . {(1 r\\1~\/1‘~’\‘,x~eft JVW-‘g
‘wk ., . ’ J” V ' ‘"‘«~\ “yfvx__.,,.,.,u/\—~v~.l~.f'
3.4 3.3 3.0 2.8 3.11 2.4 2.2 3.0 III 1.0 1.4 ppm

Figure A4. 1H NMR reaction of Rh2(02CCH3)4 with cysteine. a) 1:1 ratio,
b) 1:2 ratio, c) 1:3 ratio, and d) 1:4 ratio.

159

 

 

C)

 

 

Arw
..
f,
..

“A,“ v *w'v"* .

\11

PW"

1.4

1.6

1.3

III

(0

 

 

 

 

I ) .....
/.
fl
\
I
.
,.
4
.
y
w
‘
5.
a.
_.
_
I
v
.
._
..
.
\
x.
x
.\
¢
.,
K.»
x
I.\
'1’ 1 1 1.
1 1 1 I I I1: I
‘11 l
..
.
x.
x
11 1 1 1 Jr. 1 11

ppm

1.4

1.6

1.8

3.11

3;.
«Vb

3.8

3.0

‘._
1‘.

3.1)

 

160

1
1 I
‘ 1 r I a x h; 1'
1 I ‘1 4 I I I 1 I~ k
1 1 1. l J" ‘1 F ‘1 “($1 1‘! f .1 V.\,\ D 1‘ ’
11 ‘ ' 1 ‘1 ‘ " ""3k1‘4 ' d' “‘bg/‘xk‘ NM1WV‘N'f‘fl/ MW

3h 34 3,3 3,11 3,8 2,6 3.4 3.3 2,1) 18 1.6 1.4 ppm

 

b)

 

1
1
1
1
1
1
11
1 1
1 1
1 .
1 1
I I 1
I I . 1.1 ,
11 1 1 1 " I I
1 1 1 I1 ‘1 1
I :1 1 111, 1 '1 1 1
I I1 11 1 1,1 1 1 I
1 i I
. 11 1 1 .' 1 1 1
s1 1 , 1“ v. 1
1 1 ,1 1 1 1 11 11
1 1 1 1 1 1 11 11 1‘
I 1 I 1
| I 1 1 1V1 1 l
1 1.71 ’1
1 ,1/1'1 5] I
a K '1 "\N H ~I‘“\ 2~ M! ff- um!” vflxfi
“f; \1 hr! 1
3,1» 3.4 3.2 31'» 2x 3.1 3.4 3.3 3.0 1.8 1.6 1.4 ppm

Figure A5. 1H NMR reaction of Rh2(OZCCH3)4 with glutathione. a) 1:1 ratio,
b) 1:2 ratio, c) 1:3 ratio, and d) 1:4 ratio.

161

 

 

VR-‘/\~~\ ,,

.-.x, ~_,— ,

fi111w 11 1 1
.1 /
f;
.V
w
t
.\
I 1 || Mlx|1
It,“ 1 ..I.6
IIIII Ider II
,2
fi
A
,o
~
\
p
R
I“
\1
I1 1.1 I1
, 1I l~1
IVE
.r
a
w
.
5
a
"
.\.
.R.
n
,v
,,

pp“)

1.4

1.6

1.3

.11

fl

wm

36

.‘
h

3.8

1.11

1 .

31

H
‘I‘A

u‘ /’ _

‘,__/I

ppm

.‘

1.6

1,8

.11

3

.‘I ,0

1L
«3

‘\/

162

 

 

1"
r' “' r ~ V8” 'andfi”
1 r’ ’~ ' H
I _\t H,
V“ - r 1 1 ,,,,,,
31, 7:4 3; 3.0 3,3 in 2.4 22 2.11 1.8 1.6 1.4 ppm
1
11
1 1
I 1 11
1 1
1 I
.' 1
I 1
11 11. 1
I \1 , 1
11 1 1
.1\, 1,1 1 1
t1 \1 1 1
I 11 1|
., A I ”A; x “A ‘f‘ - M
_/ W 7 ‘-W."‘-"’--v‘\~_ 4 ,
In 1.4 3,: 3n :3 :.h 2.4 3.: :1» Is 16 1.4 I: ppm

Figure A6. 1H NMR reaction of Rh2(OZCCH3)4 with methionine. a) 1:1 ratio,
b) 1:2 ratio, 0) 1:3 ratio, and d) 1:4 raito.

163

 

 

 

 

 

_,.. . ‘nrv 'v

MIR «JV—.A

‘7... VH

\tl‘vv'yxdr'”

C)

P111"

1.4

1.6

1,8

.11

‘i

«I .

3.6

3.11

«o.

fi\r.

3 .4

w

—— ’.v~‘_,~‘_~ ~_-

(1)

 

ppm

3

1.4

1.0

1.8

w.
1,-

-\ _

Q‘—

164

1 ,III
I 1 1 111 I
1 1 1 1 "I '11
I 1 I I 1
f '1 ,1 1
1 , 1 1 1‘
~ I I. \ \X‘ \1"
rd. 1, 1 , *x‘\ V\ 1.1 \ ' 4.; ffl “\f‘y‘cm’vd‘
v ” JNIA / ,' I‘_.‘M"n"\ J 'flwlj" " v
- V _ , . ,1 ‘ .r'\_ , A- " "
w .”‘».‘/ fr ’v\4 ‘1vr \f"¢ V~wuqd M. [V
'I ~ ~ I w w w 1 ' , 3
I4 \ _ I II - I (I -.«1 33 _11 1 8 1.0 1.4 1“ 1.11 ppm
P
.I
I
I 1
y 1
I'I 1 I
, .‘J‘ 1 'I
1 1 1
I J 1 f 141‘
I
1 1 ' l , I. p a
I 1' I '1 1- I i : l
L '1‘ l‘l ’1 t l l I I V 17%}, I . . "A II 19",” 1‘V-' 111.1)1 1‘
I a 1 1 ‘ I 1 ‘ ' ‘ 1 1 lb 4 V 111 1 “'1 1" I ‘ "F 1112111“ 1 r fl 1 I" , .I i. ‘- 3 I 1‘94 1325.3 ‘I A v
I 1 " '7 1.1 1“ ‘ .“~ 1‘ ~- I ’ ‘1 1~\‘ 1 a" ‘;‘ I
' 1 1 P "1 1" 1 '1‘ W1 1 1‘ "‘ 1" "It“ '1‘?" ‘1‘ I ".‘fi‘h‘v’ ‘._ ’1 I .1": I ~ 1A .IN 1I' v 1 1' 1‘1" ’1' 1
1- l U ' I1 ’ i f ‘ 11
I I I , o 1‘.

6.2 ppm

b)

,_ 1.11 ppm

I
1 I1 I. 1 1
1 1’ ,1" 1‘: i l.

I , II 1 #11 1k 1.

I ‘ o. 1 ,4 I
1 n \1 I'll," I l 1 1", ‘ . 1} H ,1 I I 1.1 11! 3‘1"," 5* 1' I“ 'V U
1 V ‘ . ' I . 1 11‘."“'_~“ ,:1 run ,4. .
‘ 1 m1 IuI :11 1" "14V“ . 1:1. 1' a”: 11 W ' r 1'“ ..

‘ O

6.5 ppm

Figure A7. 1H NMR reaction of Rh2(020CH3)4(ATP)2 with cysteine.

a) Rh2(OZCCH3)4(ATP)2 standard, b) 1:1 ratio, c) 1:2 ratio, d) 1:3 ratio, and
e) 1:4 ratio.

165

 

I11

 

 

 

 

 

V
I
I I I L 1
I L I;
I '9
‘ L I I 1 I
. ‘ , I l I I ‘I I
I I ,
I I .
I II , " / K
I ' ‘ *4, | r \ I- L .
J -I I I! I II. J I h h! I II\ f1” A A
. I ‘I I -I v I ,II ‘ A . a ' f ”" " .»' V. ‘ A
‘7 . 5 VI, I “Vt , b! - ’1 ¥'V/ ‘- \.I \l J' ‘_.«‘\‘,J\'II‘L‘,\ (Ix/«(r " ‘I v IV
"“n ‘ - I '\~’-' I ‘ I I , I-I - x, \ A on .‘ I '
I . -- \4 ~ -’ v v 1
‘.r I',-\~x..: I ’
, , , WT - ,,,,, ,fi - 7‘-
~ ~ 1 ‘ x a x x a 1 'I 3
I4 9- 3H -8 Mb -i .- .0 I3 10 I4 I. IU US ppm
J
I"
II . J I
'I
I.‘ I ‘
I II
I J
k‘ I ‘l I
u I
I I / R; g
r; ,I a I ' I
I I , .1 l I ..I I
I" kfi 2%: I 1’ a A J I NI . fgII ‘III IL‘ "k: 5 ‘L
‘l I .1 ' ~‘ J ‘ , ' .1 .‘ t, ‘ "1‘ I “IV“ ‘ nu, . ‘r ’1 ,I ‘. I" ['4 ..A'N. I‘x‘tJ‘III 'fv‘IJ'IININ J. 'n l' ‘4 a" IIJI'II I 5’ , I: LIV. 2' u . I _, \IJ .1
,FII . .1": .‘e iI 7 - n a i ' ' I
r i J
a, ........ -I - - 7-‘ . A _ . ?'

6o 64 a: pwn

d)
I I I

t
I I I
I~I I t I I
I. I I ‘
I I
I I
I I! p I I I H
I p I‘ I I‘ i
I ‘ I
I l I’ 'I
' ’ I 3 I I I
_ I I | I,‘
I i p I' r .‘ ‘I I U
‘ 1 {I IK J , I
I r I I. I ,
"5 (x J ‘.’\ r‘ J‘ :I' Y “Iv"
I I I I f. ‘ ‘ I x
A. I I H ,1 j I / I a [I ‘J.‘»,' VN’I
II I , I‘ If.“ I II ‘17‘ I III -‘p~/‘:~\~.I
" ,‘ M A '.',,~I \k If ‘ I '1 *A'I‘I‘ l a ’ p ‘ Nrr- r. _'r VA‘I‘III.‘ “ L A1 I ’1‘" I'fl/JI-‘I‘ s‘ aw V 1 ‘
.~ ‘ ' a ‘ u' r ‘ " r‘ J' H
I w VIIIilr~v-’ ,, . I __ _ . ........ . ._-__._.. - v
‘ ‘ 1 ‘ ‘ ‘t 1 a n 1
I4 1- r” -X -6 -4 ;. -U IS in 14 13 I” pwn
E
I“
‘ .
,I II II
I I ‘. I
I I I I II II
| I I II IL
I i! 1 ' if 1 I_F\"1
[I I .‘ v\‘ I -U U H"\_, . ‘
II‘I \ l / I' . A ‘ A .\ . fi‘ 1‘ .~‘.‘. III“ .‘AI/I'fi-v‘x '-
I I ,‘I II.) 'J, r; ; '4 . f. I ‘ , . ., I ,- ‘ A ,‘ L’ f ‘N” ’ ' I’ - "2|, \J‘IM‘I “f" \ {I Ad.” I‘.‘\ ’.’ I: - I! 'tI'
I , 7 V I e 4 . I. I I L . . I ‘ '. I ‘ ' I , 9 . ‘ ,
‘I '4 ‘L 1' ‘ is l I -1!“ Y ~ I! II ”RE”! Krfifir III III“! \irfi" ‘J.’ IIPI’I‘.’\”“.I~V 1'1"; “‘ I‘P“ ‘Ib' I “hiya.“ ”. NI M I; 1} I
1 I -\
7.3 7.0 H 7- 70 0.8 6.6 6.4 6.. ppm

VII“; (III: .4 III-V _. I
ix ’ III EH 3.2 so

166

 

XX

'o:
l1

3!)

3.”

84

X“

711

167

\«J

Sq

I8 I!)

6.8 hh 0.4 H

mm

ppm

 

3.“ :x :p 2.4 32 2.4.) l 3 m 1.4 1.2 m 0.8 ppm

.. t
\’ ‘ ‘ a ‘ v

a ,"‘..AJ\‘.V\ ' *4 1‘. v/ ‘4

x '4“ i

3.5 X.“ ‘5 Tu 0.5 0.0 ppm

b)

Inwv, .fit’ ‘7 fix . ,‘J ‘~« _ V y ,
. .../'x ,. .,_ V «.W _ . . ... ww- " \m _ ... ...
~_\ ‘\ “<f‘ M'v“..r flv‘ A'~'\.~«/\’\VV' ”" w\7\“.".AWPd\A—fi ‘./ W“J.‘r‘v‘ "\/.4 V I ~~

34 1,: H1 38 “go 2.4 2.2 3.“ is If) 1.4 .2 ppm

-7-,,.fla ”NH ..V/NHH..-

‘- _
M-.~. ,
-—\ _._.\

8.5 81) 7.5 70 (15 0.0 ppm

Figure A8. 1H NMR reaction of Rh2(OZ,CCH3)2(GTP)2 + cysteine.
a) Rh2(02CCH3)2(GTP)2 standard, b) 1:1 ratio, 0) 1:2 ratio, d) 1:3 ratio, and
e) 1:4 ratio

168

 

 

 

(0

XS

4 f)

' __-‘/

in ‘ '
- .5
ju '
I.)
ppm
1 “
A- 71 “A -‘ 'J¥k "F r—v ‘-7* — N -._ 7- ‘~‘ H N‘ F.-- fifik “A- S ‘ ‘ “~ “4‘”_-, _ _/ \— v-L H
7.5
70 '
6.)
(Hi
rm
f
III
\
I{ I/
n n H
u ‘I ' I
1' a)" p I I! z "
¥ a a ’ ‘ ‘ ’ ’L . ~qu 1 1‘ I Ni
1 < '
fl.“ 1 ‘ k
2,0 '
I)
ppm
i U
/ KN

ppm

169

 

 

 

 

6)

 

 

‘ ' ._._ HM}..—
H-,.__A I“

3 .0

.\.
au—

ppm

l.5

A~mfr

,.,. -..-\(4 _
“a“: .W--.-.—-—__, .
*-I—I¢V~

“\n.

. “VAM.‘

\'

-

XII

Wm

M)

(15

170

 

 

\_fi
~— \J“
3.” 25 3,0 |.5 1.0 0.5 ppm
l
l.
H
‘i
b) i
I?
.
.1
I i
‘ 1
I l
i
i ‘ r” K.
N \.‘ A H A... A M A— HA', 3‘“ “V M“. .M_
f 4 n x h J i i H LS l 0 l 4 ppm

Figure A9. 1H NMR reaction of [Rh2(OzCCH3)2(MeCN)6][BF4]2 with cysteine.
A) 1:1 ratio, b) 1:2 ratio, c) 1:3 ratio, d) 1:4 ratio.

171

 

d)

Tm

‘8 3.6 2.4 ‘." TN [.8 1.6 H ppm

3.8 ”.6

172

“3*

 

 

- —'\_, ‘-‘\,_,_'\. /.,'

 

a)

 

 

 

’ , J‘ J
I /
# W 4 k L i.“ w _ ..-- d, \.\W ”Wyn/"W _ J. kw...“
p S 4 3 2 pm
i
'
b) ‘
1|
}
i
M
p
‘.
H
i
e» w
,‘w‘ L ‘ ’

ti 5 i i‘

. I, 't 1’! II!

61 L " i (M :4

.4 i i I a, H i,

AM, (LAW/1:1“ . _fi, “H a, \ \. 7%, Wm \_ _________________ _j \_W, m w
h 5 4 3 3 ppm

Figure A10. 1H NMR reaction of Rh2(OZCCH3)4 + 2-Aminothiophenol. a) 1:1 ratio,
b) 1:2 ratio, c) 1:3 ratio, and d) 1:4 ratio.

173

 

 

d) J

174

‘v-‘w- ......— ”fiihmf

ppm

I
l
\ w‘fw... -.....

ppm

 

 

 

 

 

h 5 «i 3 2 ppm

b)

 

 

4 3 3 ppm

Figure A11. 1H NMR reaction of Rh2(02CCH3)4 + 2-Aminothiophenol. a) 1:1 ratio,
b) 1:2 ratio, c) 1:3 ratio, and d) 1:4 ratio.

175

 

 

 

 

REFERENCES

177

 

References

1 ..Rosenberg, B.; VanCamp, L. Nature 1969, 222, 385.
2. Hambley, T. W. Coordination Chemistry Reviews 1997, 166, 181-223.
3. Sherman, S. E.; Lippard, S. J. Chem. Rev. 1987, 87, 1153-1181.

4. Lehninger, A. L.; Nelson, D. L.; Cox, M. M Principles of Biochemistry 1993,
Worth Pub. 111-130, 334-341.

5. Neidle, S. Biopoly. 1997, 44, 105-221.

6. Ohndorf, U.-M.; Witehead, J. P.; Raju, N. L.; Lippard, S. J. Biochemistry 1997,
36, 14807-14815.

 

7. Yaneva, J.; Leuba, S. H.; van Holde, K.; Zlatanova, J. Proc. Natl. Acad. Sci.
USA 1997, 94, 13448-13451.

8. Vichi, P.; Coin, F.; Renaud, J.-P.; Vermeulen, W.; Hoeijmakers, J. H. J.;
Moras, D.; Egly, J. —M. The EMBO Journal 1997, 16, 24, 7444-7456.

9. Farrell, N.; Kelland, L. R.; Roberts, J. D.; Vanbeusichem, M. Canc. Res. 1992,
52, 5065. .

10. Farrell, N.; Qu, Y.; Feng, L.; Vanhouten, 8. Biochemistry, 1990, 29, 9522.

11. Bear, J. L.; Howard, R. A.; Dennis, A. M. Curr. Chemother. 1978, 1321-
1323.

12. Howard, R. A.; Spring, T. 6.; Bear, J. L. Cancer Research 1976, 36, 4402-
4405.

13. Kitchens, J.; Bear, J. L. Inorg. Nucl. Chem. 1969, 31, 2415-2421.

14. Fimiani, V.; Ainis, T.; Cavallaro, A.; Piraino, P. Journal of Chemotherapy
1990, 2. 5, 319-326.

15. Piraino. P.; Tresoldi, 6.; Lo Schiavo, S. Inorg. Chim. Acta 1993, 203, 101-
105

16. (a) Eastland, G. W. Jr.; Yang, G.; Thompson, T. Math. and Find. Exptl. Clin.
Pharmacol. 1983, 5, 7, 435-438. (b) Dimitrov, N. V.; Eastland, G. Curr.

178

Chemother. Proc. Int. Congress. Chemother. 10‘" Meeting 1977, 2, 1319-
1321. (c) Cotton, F. A.; DeCanio, E. C.; Kibala, P. A.; Vidyasagar, K. Inorg.
Chim. Acta. 1991, 184, 221-228. (d) Chatterjee, M.; Achari, B.; Banerjee, R.;
Chakrabarti, C.; Dattagupta, J. K.; Banerjee, 8. Inorganic Chemistry 1998,
37, 21, 5424-5430.

17. Van Boom, S. S. G. E.; Reedijk, J. J. Chem-Soc. Chem. Commun. 1993,
1397-1398.

18. Davidson, Y. Y.; Chang, S.-C.; Norman, R. E. J. Chem. Soc. Dalton Trans.
1995, 77-81.

19. Guo, Z.; Casellato, U.; Faraglia, G.; Graziani, R.; Sitran, S. J. Chem. Soc.
Dalton Trans. 1994, 3231-3235.

20. Bednarski, P. J. Journal of Inorganic Biochemistry 1995, 80, 1-19.

20. Sheldrick, W. S.; Gleichmann, A. Journal of Organometallic Chemistry 1994,
470, 183-187.

22. Berners-Price, S. J.; Kuchel, P. W. Journal of Inorganic Biochemistry 1990,
38, 305-326.

23. Djuran, M. l.; Lempers, E. L. M.; Reedijk, J. Inorg. Chem. 1991, 30, 2648-
2652.

24. Odenheimer, B.; Wolf, W. lnorganica Chimica Acta 1982, 66, L41-L43.

25. Perez-Benito, J. F .; Arias, C.; Amat, E. New Journal of Chemistry 1995, 19,
1 089-1094.

26. Bose, R. N.; Ghosh, S. K.; Moghaddas, S. Journal of Inorganic Biochemistry
1997, 199-205.

27.Wyatt, K. S.; Harrison, K. N.; Jensen, C. M. Inorg. Chem. 1992, 31 , 3867-
3868.

28. Bose, R. N.; Li, D. ; Kennedy, M.; Basu, S. J. Chem. Soc. Chem. Commun.
1995, 1731-1732.

29. Pneumatakakis, G.; Psaroulis, P. lnorganica Chimica Acta 1980, 46, 97-
100.

30. Levason, W.; McAuliffe, C. A. Inorg. Chem. Nucl. Lett. 1977, 13, 123-127.

31. Bukanova, A. E., et. al. J. Inorg. Chem (Russ) 1990, 35, 2003-2006.

179

 

 

 

32. Erck, A.; Sherwood, E.; Bear, J. L.; Kimball, A. P. Cancer Research 1976,
36, 2204-2209.
33. Felthouse, T, R. Prog. Inorg. Chem. 1982, 29, 72-97.

34. Nakamoto, K. Infrared and Raman Spectra of Inorganic and Coordination
Compounds 1978, Wiley and Sons, 305-307.-

35.Boyar, E. 8.; Robinson, S. D. Coord. Chem. Rev. 1983, 50, 109-208.

36.Catalan, K.; Hess, J.; Maloney, M.; Mindiola, D.; Ward, 0.; and Dunbar,
K. submitted to Inorg. Chem.

36. Pasini, A.; Zunino, F. Angew. Chem. Int. Ed. Engl. 1987, 26, 615-624.

37. Bamham, K. J.; Bemers-Price, S. J.; Frenkiel, T. A.; Frey, U.; Sadler, P. J.
Angew. Chem. Int. Ed. Engl. 1995, 34, 17, 1874-1877.

38. Erck, A.; Rainen, L.; Whileyman, J.; Chang, l.-M.; Kimball, A. P.; Bear, J.
Proceedings of the Society for Experimental Biology and Medicine 1974,
145, 1278-1283.

39.Bear, J. L.; Gray, Jr., H. B.; Rainen, L.; Chang, I. M.; Howard, R.; Serio, G.;
Kimball, A. P. Cancer Chemotherapy Reports 1975, 59, 3, 611-620.

40. Dunbar, K. R.; Matonic, J. H.; Saharan, V. P.; Crawford, C. A.; Cristou, G. J.
Am. Chem. Soc. 1994, 116, 2201-2202.

41 . Dennis, A. M.; Howard, R. A.; Bear, J. L. Inorganica Chimica Acta 1982, 66,
L31-L34.

42. Waysbort, D.; Eichhorn, G. L. Inorg. Chem. 1993, 32, 4774, 4779.
43. Aoki, K.; Yamazaki, H. J. C. 8. Chem. Comm. 1980, 186-188.

44. Aoki, K.; Yamazaki, H. J. Am. Chem. Soc. 1984, 106, 3691-3692.
45. Farrell, N. J. C. 8. Chem. Comm. 1980, 1014-1016.

46. Rainen, L.; Howard, R. A.; Kimball, A. P.; Bear, J. L. Inorganic Chemistry
1975, 14, 11, 2752-2754.

47. Oriskovich, T. A.; White, P. S.; Thorp, H. H. Inorganic Chemistry 1995, 34, 7,
1629-1631.

48. Day, E. F.; Crawford, C. A.; Folting, K.; Dunbar, K. R.; Cristou, G. J. Am.

180

 

 

 

Chem. Soc. 1994, 116, 9339-9340.

49. Crawford, C. A.; Day. E. F.; Saharan, V. P.; Folting, K.; Huffman, J. C.;
Dunbar, K. R.; Cristou, G. Chem. Commun. 1996, 1113-1114.

50. Catalan, K. V, mindiola, D. J. Ward, D. L; Dunbar, K. R. Inorg. Chem.
1997, 36, 2458-2460.

51. Farrell, N. Journal of Inorganic Biochemistry 1981, 14, 261-265.

52.Chifotides, H. T.; Dunbar, K. R.; Matonic, J. H. Inorg. Chem. 1992, 31, 4628-
4634.

53. Pneumatakakis, G.; Hajilidis, N. J. C. S. Dalton 1979, 596-599.

54.Gibson, D.; Arvanitis, G. M.; Berman, H. M. Inorg. Chim. Acta 1994, 218, 11-
19.

55. Davidson, Y. Y.; Chang, S.-C.; Norman, R. E. J. Chem. Soc. Dalton Trans.
1995, 77-81

56.Djuran, M. l.; Lempers, E. L. M.; Reedijk, J. Inorg. Chem. 1991, 30, 2648-
2652.

57.Barnham, K. J.; Djuran, M. l.; del Socorro Murdoch, P.; Sadler, P. J. J. Chem.
Soc. Chem. Commun. 1994, 721-722.

58. Bose, R. N.; Moghaddas, S.; Weaver, E. L.; Cox, E. H. Inorg. Chem. 1995,
34, 5878-5883.

59.Appleton, T. G.; Connor, J. W.; Hall, J. R.; Prenzler, P. D. Inorg. Chem. 1989,
28, 2030-2037.

60. Nunn, C. M.; Garman, E.; Neidle, S. Biochemistry 1997, 36, 4792-4799.

61.Kopka, M. L.; Goodsell, D. S.; Baikalov, l.; Grzeskowiak, K.; Casico. D.;
Dickerson, R. E. Biochemistry 1994, 33, 13593-13610.

62.(a) Werner, M. H.; Huth, J. R.; Gronenborn, A. M.; Clore, G. M. Cell 1995, 81,
705-714. (b) Love, J. J.; Li, X.; Case, D. A.; Geise, K.; Grosschedl, R.;
705-715. Wright, P. E. Nature 1995, 376, 791-795.

63.Takahara, P. M.; Rosenzweig, A. C.; Frederick, C. A.; Lippard, S. J. Nature
1995, 377, 649-652.

64.Wing, R. M.; Pjura, P.; Drew, H. R.; Dickerson, R. E. The EMBO Journal

181

 

1984, 3, 5, 1201-1206.

65. (a) Sherman, S. E.; Gibson, D.; Wang, A. H.-J.; Lippard, S. J. J. Am. Chem.
Soc. 1988, 110, 7368-7381. (b) Sherman, S. E.; Gibson, D.; Wang, A. H.-J.;
Lippard, S. J. Science 1985, 230, 412-417.

66. (a) Girault, J.-P.; Chottard, G.; Lallemand, J.-Y.; Chottard, J.-C. Biochemistry
1982, 21, 1352-1356. (b)Yang, D.; van Boom, S. S. G. E.; Reedijk, J.; van
Boom, J. H.; Wang, A. H.-J. Biochemistry 1995, 34, 12912-12920.

67.Admiraal, G.; van der Veer, J. L.; de Graff, R. A. G.; den Hartog, J. H. J.;
Reedijk, J. J. Am. Chem. Soc, 1987, 109, 2, 592-594.

182

 

 

UNIV. LIBRRRIES

““11 I11!" Ill

7893 36

HICHIGRN STRTE
lllllllllllllllllllllllllllll
3129301