Title: Synthesis and study of the interactions of polymers containing nucleic acid bases as pendant groups
Full Citation
Permanent Link: http://ufdc.ufl.edu/UF00099356/00001
 Material Information
Title: Synthesis and study of the interactions of polymers containing nucleic acid bases as pendant groups
Physical Description: ix, 112 leaves : ill. ; 28 cm.
Language: English
Creator: Aponte, Maria Amelia, 1955-
Copyright Date: 1982
Subject: Nucleic acids -- Synthesis   ( lcsh )
Polymers and polymerization   ( lcsh )
Chemistry thesis Ph. D
Dissertations, Academic -- Chemistry -- UF
Genre: bibliography   ( marcgt )
non-fiction   ( marcgt )
Statement of Responsibility: by Maria Amelia Aponte.
Thesis: Thesis (Ph. D.)--University of Florida, 1982.
Bibliography: Bibliography: leaves 107-111.
General Note: Typescript.
General Note: Vita.
 Record Information
Bibliographic ID: UF00099356
Volume ID: VID00001
Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
Resource Identifier: alephbibnum - 000352452
oclc - 09711260
notis - ABZ0419


This item has the following downloads:

synthesisstudyof00apon ( PDF )

Full Text






To my parents . and Rosi


The author would like to thank Dr. G.B. Butler as well as other

members of the chemistry faculty and her supervisory committee for

their guidance and help. Gratitude and deepest respect are expressed

to the late Dr. E.J. Gabbay, who, as first advisor to the author, pro-

vided initial research training and encouragement.

The valuable advice, assistance and friendship of every member of

the polymer research group enabled this work to become a reality and

are highly appreciated. In particular, thanks are extended to Pedro

J. Zavala for taking the solution IR spectra, to Shadpour Mallakpour

for helping with the NMR recordings and to Dr. Teng-Shau Young for his

computer assistance. The helpful discussions with Dr. Cynthia Luiggi

as well as her friendship are very worthy of mention.

Thanks are also due to Ms. Patty Hickerson for her excellent typ-

ing and to Mr. Jeff Pates for his drawings.

Financial support for this work in terms of assistantships, fel-

lowships and awards from the Department of Chemistry, the National

Science Foundation, the Procter and Gamble Company and the DuPont and

Upjohn Chemical Companies as well as the Florida Section of the Ameri-

can Chemical Society and the Graduate School of the University of

Florida is gratefully acknowledged.

Above all, the love, patience and encouragement of her parents

along every step of the way will always be remembered with deepest






ABSTRACT . . . .



General . . .

Statement of the


General .. . .


. iii





Synthesis .


Synthesis . . . .

Physical Studies. . .

Summary and Conclusions

REFERENCES . . . . .


. .


Table Page

III-1 UV Absorption Data of the Nucleic Acid Base Derivatives. 58

III-2 UV Absorption Data of Some N-Substituted Adenines. ... 59

III-3 UV Absorption Data of Some N-Substituted Thymines and
Cytosines. . . . . . . . . . ...... 61

III-4 Differences in the Chemical Shifts of the 2- and 8-H of
N-Methyladenines and Purines . . . . . . . 63

III-5 Elemental Analyses of Alternating Copolymers (29) and
(32) . . . . . . . . ... .. . . . 67

III-6 Chemical Shifts of the Nucleic Acid Base Monomer Protons
in Chloroform-d at Different Temperatures. . . . ... 76

III-7 Association Constants and Limiting Chemical Shifts of the
Thymine 3-NH Proton of (2) . . . . . . .... 78

III-8 Chemical Shifts of the Nucleic Acid Base Monomer Protons
in DMSO-d6 at Different Temperatures . . . .... 80

III-9 Chemical Shifts of the Nucleic Acid Base Polymer Protons
in DMSO-d6 at Different Temperatures . . . .... 82

III-10 Chemical Shifts of the Nucleic Acid Base Monomer and
Polymer Protons in DMSO-d6 at 600C . . . . .... 87

III-11 Job Plot Data for the Nucleic Acid Base Monomers and
Polymers Synthesized . . . . . . . .. 92

III-12 Ultraviolet Absorption Data for the Nucleic Acid Base
Monomers and Polymers Synthesized. . . . . . ... 97

I11-13 IR Absorption Data of the Nucleic Acid Base Monomers . 102


Figure Page

I-1 Structure of a section of DNA and its nucleic acid base
constituents . . . . . . . . ... . . 3

I-2 Structures of the monomers, homo- and copolymers syn-
thesized . . . . . . . ... . . . 10

II-1 Structures of several compounds prepared . . ... 45

III-1 Representative 1H NMR spectrum of the vinyl ether mono-
mers prepared. . . . . . . . . . . 53

III-2 Representative 1C NMR spectrum of the vinyl ether mono-
mers prepared. . . . . . . . . ... . 54

III-3 Representative IH NMR spectrum of the maleimide monomers
prepared . . . . . . . . ... .. . .. 55

111-4 Representative 1C NMR spectrum of the maleimide mono-
mers prepared. . . . . . . . . . .. 56

III-5 Homo- and copolymer 13C NMR chemical shifts (ppm from
TMS) . . . . . . . . ... ... . .68

III-6 Concentration dependence of the thymine 3-NH proton
chemical shift of(0) (2), (0) the 1:1 mixture of (2)
and (1) and (0) the 1:1 mixture of (2) and (19) . . 74

III-7 Temperature dependence of the adenine 6-NH2 proton chem-
ical shifts of (0) 18, (0) (24), (A) the 1:1 mixture
of (18) and (2) and (A) (32) . . . . . ... 84

III-8 Temperature dependence of the thymine 3-NH proton chem-
ical shifts of (0) (2), (A) the 1:1 mixture of (18)
and (2), (0 ) (29) and (A ) (32) . . . . ... 85

III-9 Concentration dependence of the adenine (e) 2-H and
(0) 8-H proton chemical shifts of (1) at (-) 80 and
(---) 40C in D20 and (-.-.) at 250C in CDC13. .... . 88

III-10 UV absorption curves for the binary mixture of (1) and
(2) at different molar ratios in CHC13 . . . ... 93



III-11 UV absorption curves for the binary mixture of (1) and
(2) at different molar ratios in DMSO/EG . . ... 94

III-12 UV absorption curves for the binary mixture of (1) and
(2) at different molar ratios in DMSO. . . . ... 95

III-13 Infrared spectra of (a) (1), (b) (2) and (c) the 1:1
mixture of both in CDC13 ................ 101

Abstract of Dissertation Presented to the
Graduate Council of the University of Florida
in Partial Fulfillment of the Requirements for the
Degree of Doctor of Philosophy



Marfa A. Aponte

December 1982

Chairman: Dr. George B. Butler
Major Department: Chemistry

The goal of this investigation was to synthesize nucleic acid

base monomers that could be copolymerized in alternate fashion and

study the interactions between the monomers, homo- and copolymers

prepared. Alternation was possible due to the presence of vinyl

ether and maleimide moieties that, being donors and acceptors, re-

spectively, copolymerized through a charge-transfer complex.

Homopolymerization of all maleimide monomers and of 1-(2-vinyl-

oxyethyl)thymine was accomplished. Direct copolymerization of 9-(2-

maleimidoethyl)adenine and 1-(2-vinyloxyethyl)thymine was unsuccess-

ful due to the low solubility of the former monomer, even at high

dilutions. Incorporation of a base into a preformed copolymer and

the chemical modification of another were investigated. Best results

were obtained using the latter method. Spectroscopic and elemental

analyses confirmed the alternating structure.

Concentration and temperature dependence of the chemical shifts

of several protons of the adenine and thymine monomers and polymers

were determined in water, chloroform and dimethyl sulfoxide (DMSO),

both in the presence and absence of their complementary pairs by nucle-

ar magnetic resonance (NMR) spectroscopy. Base stacking in water was

detected by an upfield shift of the aromatic protons of 9-(2-vinyloxy-

ethyl)adenine. Hydrogen bonding between the bases, evident from the

downfield shifts observed, was present in chloroform, but in DMSO the

stronger interactions of the base derivatives with the solvent led to

much weaker base-base interactions. The copolymer showed a signifi-

cantly smaller temperature effect and virtually no concentration de-

pendence of the chemical shifts, suggesting strong intramolecular hy-

drogen bonding. Charge-transfer complexation could not be detected by

this technique.

Ultraviolet spectroscopy provided little information on these sys-

tems due to the low solubility of the compounds and the apparently

small association constants for complexation. Thus, no interaction was

seen from the Job plot analysis of pairs of these systems. A small hy-

pochromism was observed for poly-9-(2-maleimidoethyl)adenine as com-

pared to the "repeating unit" model compound.

Infrared studies were used to confirm the hydrogen bonding ob-

served from the NMR data. The vinyl ether adenine and thymine monomers

were studied individually and at different concentrations of the 1:1

binary mixture. Bands that could be associated with the groups in-

volved in the bonding were seen to vary depending on concentration.




During the past two decades, interest has developed in the prep-

aration and study of functional polymers which resemble natural macro-

molecules such as the nucleic acids. Several workers1 have contrib-

uted to this area of biomimetic research with the hope of obtaining

a better understanding of such substances, of which deoxyribonucleic

acid (DNA), an important agent in the transfer of genetic information,

is an important member.

Natural polynucleotides are macromolecules whose repeating units

consist of a sugar, a phosphate and a nucleic acid base, the total

entity being known as a nucleotide. The sugar is ribose in the case

of ribonucleic acid (RNA) and deoxyribose in DNA. In neutral solu-

tion, the phosphate group is found to be ionized, thus making the

molecule charged. The bases in question could be, for DNA, the

purines adenine and guanine and the pyrimides thymine and cytosine.

The basic structure of DNA is a double helix held together by hydro-

gen bonds between the flat, aromatic bases on opposite chains and

also by hydrophobic interactions between these stacked pairs. In

the nucleotide unit, the sugar is attached to the base by an N-gly-

coside bond to the N-9 of a purine or N-1 of a pyrimidine, and the

phosphate is present as an ester on the 3' or 5' hydroxyl of the

ribose. Complementarity between the two strands is achieved by the

unique hydrogen bonding arrangement of two interactions of this type

between adenine and thymine and three between guanine and cytosine

(Figure I-1).2

The extent to which the interactions, hydrogen bonding and base

stacking, contribute to the stability of the molecule continues to

be a matter of debate among scientists, since their analysis can be

influenced by the conditions in which they are studied.

Initial attempts at attaching nucleic acid bases onto a polymeric

matrix were done by incorporating such heterocycles to cellulose de-

rivatives such as in the coupling of the nucleic acid base guanine

with a p-diazobenzoate ester of cellulose. A different approach,

whereby monomers having the desired functionalities were first syn-

thesized and then polymerized, was begun during the decade of the six-

ties. Most of the monomers thus prepared contained a vinyl, meth-

acrylate, acrylate, or any other appropriate group.4'5 Homopolymers

and copolymers of these species were successfully obtained in several

cases, and copolymers prepared with maleic anhydride, acrylamide,

methyl methacrylate, etc. were synthesized by using a radical mech-

anism. The properties of such polymers were studied, and an inter-

action among the bases on the chains on some of them and those on

natural polynucleotides was shown to exist.6 Work in this area has

also included the preparation of stereoregular polymers having pendant

nucleic acid bases, such as the poly(methacryloyloxyethyl) derivatives

of adenine, uracil, thymine and theophylline prepared by Inaki and

coworkers7 and the optically active graft polymers developed by

Overberger and Morishima.8

Nucleic Acid Bases

1 2

12N N
3 H9





3N 5

032 6

Figure I-1. Structure of a section of DNA and its nucleic acid base
constituents. (a) Schematic representation of the DNA
double helix. (b) Structure of a section of a DNA chain.
(c) Nucleic acid bases.

HN ,H3


It is noteworthy to mention that practical applications have

been found for these substances, in addition to their use as models

for polynucleotides. For example, not only have the bases themselves

been incorporated into cellulose,3 but also, vinyl polymers of these

have been employed in column chromatographic studies, exhibiting dis-

tinct separating power. They have been used in the concept of tem-

plate polymerization, in which a preformed functional polymer of one

base is expected to participate in the polymerization of a complemen-

tary monomer, thus increasing in some cases the polymerization rate0

The effect of the stereoregularity of the template polymer on the

polymerization behavior was also studied.10 The biological activity

of these compounds has also been tested. In this respect, Pitha has

extensively studied the antiviral resistance of the polyinosinic

acid-poly(1-vinylcytosine) complex and the inhibition of virus

replication by poly(vinyluracil) and poly(vinyladenine).12

The chains of natural polynucleotides are rarely of a homopoly-

meric nature. Several attempts have been made to place two bases on

the same polymeric chain, and some of these copolymers even show some

tendency toward alternation,13 but efforts to obtain a high degree of

alternation have not been properly pursued. One method involves the

initial partial incorporation of one base into poly(vinyl alcohol)

and subsequent attachment of an ethyl dihydrogen phosphate derivative

of another by coupling with dicyclohexylcarbodiimide (DCC). Using

this approach, only the percentage of the bases incorporated is known,

and little information can be obtained regarding their placement along

the chain.14 In another method, two different bases are present in

the monomer to be reacted. This monomer is prepared by opening the

epoxide ring of a base derivative by attack of another base and fur-

ther reaction of the alcohol formed with methacryloyl chloride. The

polymer eventually formed contains the two bases in a 1:1 ratio, but

of course, alternation is not achieved.15

It is expected that having a copolymeric model in which the bases

are placed in a known alternate fashion where the distances along the

chain are known would be of great help in the study of natural poly-

nucleotides. With this in mind, a series of monomers having the vinyl

moieties with proper electronic properties which would make them capa-

ble of undergoing an alternating copolymerization mechanism have been

synthesized and studied, as well as homopolymers and copolymers of

such compounds.

The discovery that certain monomers which fail to homopolymerize

readily will nevertheless copolymerize with other seemingly low reac-

tive monomers has led to an increased interest in the study of copoly-

merization, especially when such systems afford polymers in which the

monomers are found in a 1:1 ratio.

There are several ways of preparing alternating copolymers which

may be classified in two categories; one is the alternating addition

of monomers, and the other is the copolymerization via a 1:1 inter-

mediate of a stable monomer complex composed of two monomer units.16

The former is a special case of copolymerization, in which alter-

nating addition is predominant over homo-addition. Attempts to de-

scribe this process quantitatively in terms of reactivity and polarity

involve the use of the Price-Alfrey's scheme, shown below for the case

of the copolymerization of monomers M1 and M2

M1 + MH2 (M1M2)n

For M1: r1 =-Q exp[-e(el-e2)]

where rl is the reactivity ratio of M1, Q1 and Q2 are the reactivities

of monomers 1 and 2, respectively, and el and e2 are proportional to

the electrostatic interactions of the permanent charges on the substi-

tuents in polarizing the double bonds. Thus, the copolymerization

tendency is higher if there is a large difference in the e-value of

the monomers, that is, if they have sufficiently different polarity.

However, the reactivity of the monomers should be close, as given by

similar Q-values if high alternation is desired. Unfortunately, large

uncertainties in the values of Q and e which can be experimentally

determined render this scheme semiempirical.17

The polymerization of a complex intermediate composed of two

monomer units gives an alternating copolymer of high regularity. In

this respect, zwitter ions formed from two monomers are reported as

useful intermediates for alternating copolymerizations. Cyclic mono-

mers composed of two monomer units may do the same when suitable

catalysts are available. Charge-transfer complexes of electron-donor

monomers such as olefins or vinyl ethers and electron-accepting mono-

mers such as maleic anhydride or sulfur dioxide were reported as in-

termediates yielding alternate copolymers.

The charge-transfer (CT) complex exists in resonance form between

the non-bond and the dative states of the donor monomer (D) and the

acceptor monomer (A).

A + D _z [(A,D) + (A-.,D +)]

non-bond dative

Thus the wave functions of the CT-complex [ECT] can be expressed as a

linear combination of wave functions of the non-bond state [w(A,D)]

and the dative state [l(A--,D +)],

ICT = a'(A,D) + b'(A'-.,-D).

The equilibrium constant K may be calculated from ultraviolet (UV) and

nuclear magnetic resonance (NMR) spectroscopic measurements.

In the UV, the CT-complex shows its own characteristic absorp-

tion; the equilibrium constant is measured by using the Scottl8 and

Benesi-Hildebrand19 equation from the molar extinction coefficient, e,

of the CT-complex,

[A][D]l/d = 1/EK + [D]/E
where [A] is acceptor concentration, [D] is donor concentration, d is

absorbancy, and 1 is path length.

In the NMR, the acceptor (or donor) proton gives a chemical shift

in an amount, ACT = 6CT 6o, where 6CT and 6 are chemical shifts of

the CT-complex and the acceptor, respectively. The chemical shift of

the system, Aobs = 6obs is expressed by the Hanna-Ashbaugh20


i/Aobs = 1/KACT[D] + I/ACT-
Because the CT-complex is in general more polar than the component

monomers, the equilibrium constant and, therefore, the rate of copoly-

merization decreases with increasing dielectric constant and elevating


In the case of alternating copolymerization via a donor-acceptor

monomer complex, the rate of copolymerization is proportional to the

concentration of the complex [C]. The latter is also proportional

to the product of the concentration of both monomers, when the equi-

librium constant K is small;

R a[C]aK[D][A]

It follows that the rate of copolymerization should maximize at a

monomer feed molar ratio of 1:1. This is observed in the case of the

alternating copolymerization of styrene-maleic anhydride, 2-chloro-

ethyl vinyl ether-maleic anhydride, N-vinyl carbazole-fumaronitrile,

N-vinyl carbazole-dimethyl fumarate, and ethyl vinyl ether-maleic

anhydride. However, there may be some exceptions, such as in the

system vinyl acetate-maleic anhydride. In this case, the concentra-

tion of the CT-complex attains a maximum at the 1:1 monomer ratio,

but the monomer composition giving the maximum rate of alternation is

not 1:1; it is at a maleic anhydride composition of 0.2-0.5, depending

on the monomer concentration.

The system N-phenylmaleimide (NPM) and 2-chloroethyl vinyl ether

(CEVE), recently studied by Olson,21 has been demonstrated to give an

alternating copolymer. It would be expected, therefore, that connect-

ing the nucleic acid bases to a maleimide or a vinyl ether moiety and

reacting pairs of this kind should lead to polymers with alternating


Statement of the Problem

Goal of Research

With all of the above stated in mind, it was thought that a new

series of monomers could be synthesized having polymerizable double

bonds of two types, either of a donor or an acceptor nature. These

would be capable of undergoing an alternating copolymerization process,

leading to a completely new approach in the synthesis of functional

polymers containing nucleic acid bases. These monomers would contain

a nucleic acid base in their structure with the linkage attaching it

to the polymerizable moiety at the same position as in the natural

polynucleotides. Homopolymers as well as the proposed copolymers of

these monomers could be prepared, and studies of the monomers and poly-

mers could be conducted in order to investigate the interactions among

the bases. Moreover, since their actual location along an alternating

structure would be known, new information could be obtained with re-

spect to such interactions, especially since the complicating factors

of the sugar and phosphate backbone would be eliminated. Spectro-

scopic analysis of the interactions of such species could be pursued

using techniques such as NMR, UV and infrared (IR) spectroscopy.

Strategy of Design

For the purpose mentioned above, monomers 1-3 and 18-21, shown

in Figure 1-2, were designed. In general, these monomers are classi-

fied into two major categories; one is the vinyl ethers (1-3) and the

other the maleimides (18-21). In addition, modifications of some mono-

mers were performed leading to compounds 4 and 22 in order to render

these more soluble in the appropriate solvents for the polymerization.







(1) A
(2) T
(3) C
(4) AcC



(24) A
(25) C1P
(26) T
(27) C

2 2





(18) A
(19) C1P
(20) T
(21) C
(22) AcC

r n 4
2 2
(28) T



B5 B6 X R
(29) C1P T (30) N CH2CH2Br
(32) A T (31) 0

Figure 1-2. Structures of the monomers, homo- and copolymers


Compound 23, in which the unsaturation of the maleimide moiety was

eliminated by hydrogenation, was prepared as a model compound.

Polymerization of these species afforded homopolymers 24-28 and

copolymers 29-32. Three approaches were tried in order to obtain

copoly[9-(2-maleimidoethyl)adenine/1-(2-vinyloxyethyl)thymine] (32),

and one of them chosen as the adequate route to this polymer (Figure 1-2).

Physical studies of these species were conducted and will be dis-

cussed in later chapters of this work.




Melting points, given in degrees centigrade, were taken in a

Thomas-Hoover Capillary Melting Point Apparatus and are reported


Elemental analyses were performed by Atlantic Microlab, Inc.,

Atlanta, Georgia.

The proton (1H) and carbon (13C) NMR spectra were recorded using

a Jeol JNM-FX-100 Nuclear Magnetic Resonance Spectrometer. Tetra-

methylsilane (TMS) or solvent peaks [dimethyl sulfoxide-d6 (DMSO-d6) =

2.49, 1H; 39.5, 13C or chloroform-d (CODC3) = 7.24, 1H; 77.0, 13C]22

were used as internal standards, and the chemical shifts (6) are

given in parts per million (ppm). The multiplicities of the peaks

observed are described as singlet (s), doublet (d), doublet of doublet

(dd), triplet (t), quartet (q) or multiple (m).

Infrared spectra were obtained as potassium bromide (KBr) pel-

lets using a Perkin Elmer 281 Infrared Spectrophotometer, having the

1601 cm-1 line of a polystyrene film as a standard. Solution spectra

were recorded on a Nicolet 7199 FT-IR instrument. The observed fre-

quencies are reported in wavenumbers (cm- ), with the intensity of

the bands described as strong (s), medium (m) or weak (w).

The ultraviolet absorption studies were run on a Beckman Acta V

UV-Visible or a Cary 17 Spectrophotometer.

Reagents and Solvents

Reagents were obtained from Aldrich Chemical Co., Mallinckrodt,

Inc. or Fisher Scientific Co., unless otherwise noted. Deuterated

solvents were purchased from Merck & Co., Inc.

Solvents were purified as needed by following procedures reported

in the literature.23 Thus, dimethylformamide (DMF) was shaken and

stored overnight over potassium hydroxide pellets and distilled from

calcium oxide at 50C under reduced pressure; dichloromethane (CH2C12)

was distilled from phosporous pentoxide (P4010) immediately prior to



9-(2-Vinyloxyethyl)adenine (1)

All three nucleic acid base vinyl ether derivatives (1-3) were

prepared by employing alkylation methods reported earlier.24,25

Into a previously flamed 300 ml, three-necked, round-bottomed

flask fitted with a gas inlet tube, mechanical stirrer, thermometer

and drying tube were weighed 1.2 g (24 mmole) of a 50% oil dispersion

of sodium hydride (PCR Research Chemicals, Inc.). After rinsing four

times with hexane under nitrogen flow, 200 ml of freshly distilled

DMF were added as well as 3.0 g (22 mmole) of adenine. The mixture

was stirred for 1 hour at 600C to produce the white, thick suspension

of the adenine sodium salt. To this was added 3.4 ml (33 mmole) of

2-chloroethyl vinyl ether. The suspension was stirred at 55-600C for

24 hours. The precipitate present after this period was filtered,

and the yellow OMF solution was evaporated at reduced pressure and

low heat to yield a yellowish solid. Hot 1,2-dichloroethane was added

to the solid, and the mixture filtered while hot. After slow cooling,

white crystals were formed. These were filtered, recrystallized from

the same solvent and dried. A total of 2.95 g (64.7%) of product

was obtained, melting point 140-1450C.

Analysis, calculated for C9H11N50: C, 52.67; H, 5.40; N. 34.13.

Found: C, 52.46; H, 5.42; N, 34.25.

H NMR (DMSO-d6) 6 4.18 (m, 6H), 6.45 (dd, 1H), 7.23 (s, 2H),

8.10 (s, 1H), 8.15 (s, 1H).

C NMR (DMSO-d6) 6 42.30, 65.55, 87.58, 118.67, 141.05, 149.43,

151.18, 152.35, 155.81.

IR 3310 (s), 3145 (s), 1665 (s), 1600 (s), 1575 (s), 1474 (m),

1417 (m), 1358 (m), 1323 (s), 1310 (m), 1255 (m), 1239 (m), 1208 (s),

1065 (w), 1015 (m), 950 (w), 820 (w), 794 (w), 718 (w), 650 (w) cm-1

UV max (nm) 0.1N HC1 (258), 0.1N KOH (260), H20 (260); max (H20)
max L max 2
= 1.51 x 10

1-(2-Vinyloxyethyl)thymine (2)

The sodium salt of thymine was prepared as previously described

for the case of adenine by reacting 5.0 g (40 mmole) of thymine and

2.1 g (44 mmole) of a 50% oil dispersion of sodium hydride in 250 ml

of DMF for 1 hour at 600C. At this point, 6.0 ml (59 mmole) of 2-

chloroethyl vinyl ether was added, and after increasing the tempera-

ture of the suspension to about 900C, it was left to stir for 46 hours.

The small amount of solid present was filtered, and the DMF evapo-

rated to yield a beige paste. To this paste were added 100 ml of

water and 100 ml of chloroform. After separation, two 100 ml por-

tions of chloroform were added to the aqueous layer. The organic

layers were collected and washed with 100 ml of water and left to

dry over magnesium sulfate. After reduction of the solvent volume,

white needles separated and were eventually recrystallized from ben-

zene/petroleum ether to afford 2.31 g (29.6%) of compound, melting

point 110-1120C.

Analysis, calculated for C9H12N203: C, 55.09; H, 6.17; N, 14.28.

Found: C, 55.14; H, 6.19; N, 14.27.

H NMR (DMSO-d6) 5 1.75 (s, 3H), 3.88 (s, 4H), 4.0 (dd, 1H),

4.23 (dd, 1H),6.49 (dd, 1H), 7.48 (s, 1H), 11.27 (s, 1H).
1C NMR (DMSO-d6) 5 11.79, 46.49, 65.11, 87.43, 108.10, 141.78,

150.79, 151.33, 164.15.

IR 3170 (m), 3045 (s), 2820 (w), 1690 (s), 1624 (s), 1462 (s),

1416 (m), 1385 (m), 1362 (m), 1352 (m), 1322 (m), 1270 (w), 1236 (s),

1196 (s), 1144 (w), 1082 (m), 1030 (m), 992 (w), 965 (s), 910 (w),

852 (s), 778 (w), 760 (m), 700 (w), 698 (w) cm-1.

UV Xmax (nm) 0.1N HC1 (271), 0.1N KOH (269), H20 (270); max

(H20) = 1.01 x 104
1-(2-Vinyloxyethyl)cytosine (3)

Using the same procedure as reported previously, preparation of

this compound was accomplished by first reacting 2.4 g (50 mmole) of

a 50% sodium hydride oil dispersion and 5.0 g (45 mmole) of cytosine

in 250 ml of freshly distilled DMF for 2 hours at 600C. To the white

suspension was then added 14 ml (1.4 x 102 mmole) of 2-chloroethyl

vinyl ether in two equal portions, one at the beginning of the

reaction and the other after 36 hours. The suspension was stirred

at 60-70C for a total reaction time of 60 hours. The precipitate

obtained was filtered, and evaporation of the solvent led to a yellow

solid. A 1:1 solution of methanol/l,2-dichloroethane was added to

this solid, and after filtering while hot, the solid formed was re-

crystallized from the same solvent system to afford 3.45 g (42.3%)

of white flakes, melting point 185-1870C.

Analysis, calculated for C8H11N302: C, 53.03; H, 6.12; N, 23.19.

Found: C, 52.98; H, 6.15; N, 23.16.

H NMR (DMSO-d6) 6 3.86 (s, 4H), 3.98 (dd, 1H), 4.21 (dd, 1H),

5.62 (d, 1H), 6.46 (dd, 1H), 7.05 (s, 2H), 7.49 (d, 1H).

1C NMR (DMSO-d6) 6 48.05, 65.26, 87.34, 92.89, 146.60, 151.38,

155.62, 166.00.

IR 3360 (s), 3110 (m), 1660 (s), 1615 (s), 1518 (m), 1480 (s),

1416 (m), 1382 (s), 1350 (m), 1322 (m), 1275 (m), 1250 (w), 1210 (s),

1190 (m), 1128 (w), 1112 (w), 1018 (w), 970 (w), 950 (w), 830 (m),

784 (m), 705 (m), 643 (w), 610 (w) cm-I

UV max (nm) 0.1N HC1 (280), 0.1N KOH (272), H20 (272); emax

(H20) = 8.68 x 103

N-Acetyl-l-(2-vinyloxyethyl)cytosine (4)

The amino group of 1-(2-vinyloxyethyl)cytosine was acetylated,

using a modification of Otter and Fox's procedure,26 by refluxing 1.Og

(5.5 mmole) of (3), 5 ml of acetic anhydride and 0.75 ml of pyridine

in 50 ml of absolute ethanol for 1 hour. The solvent was then re-

moved to leave a white paste that was recrystallized from ethyl ace-

tate. A total of 0.69 g (56%) of white needles were obtained, melt-

ing point 161-1620C.

Analysis, calculated for C10H13N303: C, 53.80; H, 5.87; N, 18.82.

Found: C, 53.83; H, 5.93; N, 18.90.

H NMR (DMSO-d6, int. ref. DMSO-d6) 6 2.08 (s, 3H), 3.98 (m, 5H),

4.21 (dd, 1H), 6.45 (dd, 1H), 7.12 (d, 1H), 7.98 (d, 1H), 10.80 (s,

13C NMR (DMSO-d6, int. ref. DMSO-d6) 6 24.25, 48.95, 64.65, 87.60,

94.77, 150.86. 151.35, 155.10, 162.46, 170.85.

IR 3230 (m), 3020 (w), 2960 (w), 1716 (s), 1660 (s), 1615 (s),

1552 (s), 1488 (s), 1425 (m), 1355 (s), 1310 (s), 1230 (s), 1192 (s),

1126 (w), 1065 (w), 1038 (w), 1000 (w), 980 (w), 960 (m), 830 (m),

815 (s), 788 (s), 724 (w), 670 (w) cm-1

N-(2-Chloroethyl)maleamic acid (5)

To a solution of 2.6 g (31 mmole) of sodium bicarbonate in 30 ml

of water placed in a 125 ml Erlenmeyer flask and magnetically stirred

was added 3.6 g (31 mmole) of 2-chloroethylamine hydrochloride. All

dissolved and some effervescence was observed. To the clear solution

was added, slowly and with stirring, 3.0 g (31 mmole) of freshly

crushed maleic anhydride. After an increase in the amount of effer-

vescence, it dissolved almost completely, and a white precipitate

started to form. The mixture was stirred at room temperature for 2

hours. The white precipitate was filtered and dried in vacuo over

P4010. Recrystallization could be performed using chloroform as the
solvent, but it was not essential to obtain pure product. A total of

2.59 g (47.7%) of compound was obtained, melting point 95-970C.

Analysis, calculated for C6H8NO3C1: C, 40.58; H, 4.54; N, 7.89;

C1, 19.96. Found: C, 40.44; H, 4.58; N, 7.82; C1, 19.95.


H NMR (DMSO-d6, int. ref. DMSO-d6) 6 3.66 (m, 4H), 6.31 (q, 2H),

9.05 (s, 1H), CO H not observed.
C NMR (DMSO-d6, int. ref. DMSO-d6) 6 42.76, 131.13, 131.95,

165.39, 165.88, NHCH2CH2C1 overlapped with solvent.

IR 3260 (s), 3110 (m), 2980 (w), 2850 (w), 1918 (w), 1860 (w),

1710 (s), 1632 (s), 1550 (s), 1424 (s), 1360 (m), 1310 (s), 1250 (s),

1225 (s), 1185 (m), 1086 (w), 1032 (m), 980 (m), 886 (w), 845 (s),

735 (m), 655 (m), 622 (m) cm-1.

N-(2-Bromoethyl)maleamic acid (6)

Using a similar procedure as reported for the chloro derivative,

preparation of this compound was accomplished by adding slowly and

with stirring 3.0 g (31 mmole) of maleic anhydride to a solution of

2.6 g (31 mmole) of sodium bicarbonate and 6.3 g (31 mmole) of 2-

bromoethylamine hydrobromide in 30 ml of water. Abundant efferves-

cence was observed again, and the solid that precipitated after stir-

ring for 2 hours at room temperature was filtered. It was then dried

over P4010 in vacuo to finally yield 5.17 g (76.1%) of white crystals,

melting point 108-1100C.

Analysis, calculated for C6H8NO3Br: C, 32.45; H, 3.63; N, 6.31;

Br, 35.99. Found: C, 32.38; H, 3.67; N, 6.27; Br, 35.79.
1H NMR (DMSO-d6, int. ref. DMSO-d6) 6 3.55 (m, 4H), 6.31 (q, 2H),

9.02 (s, 1H), C2 H not observed.
C NMR (DMSO-d6, int. ref. DMSO-d6) 6 31.41, 131.13, 132.00,

165.34, 165.88, NHCH2CH2Br overlapped with solvent.

IR 3260 (m), 3105 (m), 2980 (w), 2845 (w), 1924 (w), 1850 (w),

1705 (s), 1628 (s), 1545 (s), 1500 (s), 1420 (m), 1358 (m), 1310 (m),

1290 (m), 1222 (s), 1170 (m), 1085 (w), 1018 (m), 972 (m), 876 (w),

845 (s), 730 (w), 630 (w) cm-1

3,6-Endoxo-1,2,3,6-tetrahydrophthalic anhydride (7)

This compound, originally synthesized by Diels and Alder,27 was

prepared by following the procedure reported by Narita et al.28 To

80.0 g (0.816 mole) of finely-crushed maleic anhydride and 300 ml of

benzene placed in a 500 ml, three-necked, round-bottomed flask

equipped with a mechanical stirrer and a condenser was added 80 ml

(1.1 mole) of freshly distilled furan. The mixture was warmed
slightly to dissolve most of the solid and stirred at room tempera-

ture for about 24 hours. The solid that precipitated was filtered

and dried in a dessicator overnight. The filtrate that remained had

its volume reduced and afforded more crystals upon standing. A total
of 124.2 g (91.63%) of adduct were obtained, melting point 115-1170C

(dec) [literature, 1180C (dec)29].
H NMR (DMSO-d6, int. ref. DMSO-d6) 6 3.30 (s, 2H), 5.33 (s, 2H),

6.56 (s, 2H).
1C NMR (DMSO-d6, int. ref. DMSO-d6) 6 49.05, 81.66, 136.83,


IR 3600 (w), 3090 (w), 2990 (w), 1858 (s), 1790 (s), 1595 (w),
1310 (m), 1280 (m), 1230 (s), 1144 (m), 1084 (s), 1020 (s), 948 (s),

920 (s), 900 (s), 878 (s), 848 (s), 730 (m), 690 (m), 672 (m), 632

(m) cm-1

N-(2-Chloroethyl)-3,6-endoxo-1,2,3,6-tetrahydrophthalic acid (8)

A total of 6.1 g (73 mmole) of sodium bicarbonate, 8.4 g (73

mmole) of 2-chloroethylamine hydrochloride and 12.0 g (72.2 mmole)

of (7) were dissolved, in that order, in 100 ml of water by mag-

netically stirring the reactants in a 250 ml Erlenmeyer flask.

Abundant foaming occurred along with the precipitation of a white

solid which was filtered and dried over P4010 in vacuo. No recrys-

tallization was needed for the purification of the 14.09 g (79.44%)

of product formed, melting point 125-127C.

Analysis, calculated for C10H12NO4C1: C, 48.89; H, 4.92; N,

5.70; C1, 14.43. Found: C, 48.75; H, 4.92; N, 5.64; C1, 14.68.

H NMR (DMSO-d6, int. ref. DMSO-d6) 6 2.60 (s, 2H), 3.44 (m,

4H), 4.88 (s, 1H), 5.07 (s, 1H), 6.42 (s, 2H), 7.85 (s, 1H),

CO2H not observed.

C NMR (DMSO-d6, int. ref. DMSO-d6) 6 40.77, 42.91, 46.61,

79.07, 80.49, 136.44, 136.78, 171.29, 172.85.

IR 3338 (m), 3038 (m), 1695 (s), 1655 (s), 1530 (s), 1420 (m),

1388 (m), 1302 (m), 1248 (m), 1218 (m), 1100 (s), 1040 (m), 995 (w),
925 (m), 902 (m), 820 (m), 742 (w), 715 (m), 670 (w) cm-

N-(2-Bromoethyl)-3,6-endoxo-1,2,3,6-tetrahydrophthalic acid (9)

Into a 2,000 ml, three-necked, round-bottomed flask equipped

with an efficient mechanical stirrer was placed 120.1 g (0.723 mole)

of (7) along with 300 ml of acetone. To the mixture, a saturated

solution of 148.1 g (0.723 mole) of 2-bromoethylamine hydrobromide

and 60.7 g (0.723 mole) of sodium bicarbonate in 400 ml of water was

added slowly through an addition funnel with efficient stirring.

Abundant effervescence developed after about 10 minutes, and the al-

most clear solution became frothy and very thick. The precipitate

thus formed was stirred for 1 hour at room temperature and filtered

after this period. After washing with cold water, the white solid

was dried in a dessicator over P4010 to yield 154.9 g (73.86%) of

material, melting point (dec) 113-1150C.

Analysis, calculated for C10H12N04Br: C, 41.40; H, 4.17; N,

4.83; Br, 27.54. Found: C, 41.45; H, 4.17; N, 4.79; Br, 27.42.

1H NMR (CDC13) 6 2.80 (s, 2H), 3.49 (m, 4H), 5.08 (s, 1H), 5.33

(s, 1H), 6.44 (m, 2H), 6.93 (s, 1H), C02H not observed.

C NMR (CDC13) 6 30.80, 41.04, 47.96, 79.44, 81.05, 135.49,

136.81, 172.04, 173.41.

IR 3320 (s), 3050 (m), 1690 (s), 1655 (s), 1530 (s), 1415 (m),

1388 (m), 1300 (m), 1260 (s), 1210 (s), 1173 (m), 1100 (w), 1040 (m),

993 (w), 904 (s), 820 (m), 712 (m), 670 (m) cm-1

N-(2-Chloroethyl)-3,6-endoxo-1,2,3,6-tetrahydrophthalimide (10)

The general method for the cyclization of maleamic acids to

maleimides reported by Barrales-Rienda et al.30 was used to obtain

both haloethyl derivatives of 3,6-endoxo-1,2,3,6-tetrahydrophthalimide.

A suspension of 15 ml of acetic anhydride and 0.67 g (8.2 mmole)

of sodium acetate placed in a 25 ml, three-necked, round-bottomed

flask fitted with a mechanical stirrer and thermometer was warmed to

about 80C by means of an oil bath. This had 4.0 g (mmole) of (8)

slowly added, and the mixture was stirred at 80-900C for 15 minutes.

It was then poured, slowly and with much stirring, into 20 ml of an


ice-water mixture. The oil that formed was extracted twice with 20

ml of chloroform. The organic layers were washed twice with 20 ml

of water and left to dry over anhydrous magnesium sulfate. The sol-

vent was then reduced, and upon cooling, the crystals formed were

filtered. These were obtained in 35.3% (1.31 g), melting point 134-


Analysis, calculated for C10H1oN03C1: C, 52.76; H, 4.43; N,

6.15; C1, 15.57. Found: C, 52.56, H, 4.49; H, 6.12; C1, 15.56.

H NMR (CDC13) 6 2.90 (s, 2H), 3.74 (m, 4H), 5.28 (s, 2H), 6.53

(s, 2H).

C NMR (CDC13) 6 39.77, 40.16, 47.42, 80.90, 136.51, 175.80.

IR 3460 (w), 3100 (w), 1772 (m), 1700 (s), 1428 (m), 1400 (s),

1364 (m), 1340 (s), 1320 (m), 1215 (m), 1160 (s), 1105 (w), 1015 (m),

960 (m), 925 (m), 890 (m), 873 (s), 850 (m), 823 (w), 718 (m), 642

(m) cm-1

N-(2-Bromoethyl)-3,6-endoxo-1,2,3,6-tetrahydrophthalimide (11)

As for the previous case, a suspension was formed by adding 5.7 g

(69 mmole) of sodium acetate to 50 ml of acetic anhydride and heating

to approximately 800C. To this suspension was slowly added 30.4 g

(105 mmole) of (9), keeping the temperature at about 800C at all

times. After complete addition, the suspension was left to stir at

850C for 30 minutes. The flask was then cooled with an ice bath, and

a mass of crystals precipitated at around 500C. It was poured very

slowly and with efficient stirring into 300 ml of an ice-water mix-

ture and left stirring for 1 hour. After this period, the initially

oily reaction product became a fine, sandy solid, which was easily

filtered and dried. Recrystallization was performed from chloro-

form/cyclohexane to afford 14.67 g (51.45%) of crystals, melting

point 139-1420C.

Analysis, calculated for C10H10NO3Br: C, 44.14; H, 3.70; N,

5.15; Br, 29.37. Found: C, 44.11; H, 3.73; N, 5.12; Br, 29.34.

1H NMR (DMSO-d6) 6 2.98 (s, 2H), 3.57 (t, 2H), 3.74 (t, 2H),

5.15 (s, 2H), 6.55 (s, 2H).

C NMR (CDC13) 6 26.85, 40.11, 47.47, 80.95, 136.56, 175.65.

IR 3455 (w), 3092 (w), 1772 (m), 1700 (s), 1428 (m), 1398 (s),

1360 (m), 1335 (s), 1320 (m), 1258 (m), 1200 (s), 1145 (s), 1104 (m),

1058 (w), 1010 (s), 956 (m), 920 (m), 870 (s), 850 (m), 820 (m),
804 (w), 716 (m), 645 (m) cm-1

N-(2-Chloroethyl)maleimide (12)

The two new alternative routes described below were found to

lead to this maleimide derivative, for which other methods of prepara-

tion had been reported earlier in the literature.31

Approach A. In the first case, dehydration of acid (5) was em-

ployed to prepare this compound. Initially, 34 ml of acetic anhy-

dride and 1.38 g (16.8 mmole) of sodium acetate were heated to 800C.

After achieving this temperature, 6.0 g (34 mmole) of (5) was added

slowly and mechanically stirred at 80C for 10 minutes. Heat was then

removed and the temperature permitted to drop to room temperature dur-

ing the next 10 minutes. The reaction mixture turned brownish beige,

and a solid mass of crystals formed upon further cooling with an ice

bath. The mixture was poured into 50 ml of an ice-water mixture, and

two 50 ml portions of chloroform were added in order to extract the

product. The organic layers were combined, washed with 50 ml of water

and dried over magnesium sulfate. After reducing the volume of the

solvent and allowing the residue to stand, a thick orange paste formed.

The paste had chloroform added, was warmed, and cyclohexane addition

with subsequent cooling led to precipitate formation. These crystals

were dissolved in a minimum amount of chloroform and percolated through

a silica gel/chloroform column. Reduction of the volume of the eluate

produced 0.76 g (14%) of white crystals, melting point 60-630C.

Approach B. The retro-Diels Alder approach to maleimides reported

by Narita et al.28 also produced N-(2-chloroethyl)maleimide as well as

maleimides 13 and 18-21 as follows. Into a 100 ml round-bottomed

flask equipped with a reflux condenserwere placed 2.0 g (8.8 mmole) of

(10) after previous warming of 75 ml of bromobenzene and 0.15 g (0.88

mmole) of 4-tert-butylcatechol. The reaction mixture was refluxed

for 30 minutes and was filtered while hot, after which the solvent

evaporated under reduced pressure. A yellowish oil was obtained, and

50 ml of hot cyclohexane was added. The solution was heated, and

after filtration was allowed to slowly cool. After this period, very

fine needles were formed which could be sublimed at 55C/0.2-0.3 mm Hg

to afford 0.82 g (59%) of slightly yellow crystals, melting point 60-630C.

Analysis, calculated for C6H6NO2C1: C, 45.16; H, 3.79; N, 8.78;

C1, 22.21. Found: 45.06; H, 3.82; N, 8.73; Cl, 22.14.

1H NMR (CDC13, int. ref. CDC13) 6 3.73 (m, 4H), 6.71 (s, 2H).

1C NMR (CDC13, int. ref. CDC13) 6 39.13, 40.69, 134.12, 170.14.

IR 3458 (w), 3100 (w), 2940 (w), 1700 (s), 1584 (m), 1442 (s),
1412 (s), 1374 (m), 1332 (m), 1286 (w), 1222 (w), 1146 (m), 1128 (m),

1072 (w), 1050 (w), 982 (w), 930 (w), 894 (w), 832 (m), 720 (w), 695

(m), 660 (m) cm-1.
N-(2-Bromoethyl)maleimide (13)

A solution of 2.0 g (7.3 mmole) of (11), 0.12 g (0.73 mmole) of

4-tert-butylcatechol and 100 ml of bromobenzene was refluxed for 30

minutes. The bromobenzene was subsequently removed to afford a yellow

oil that on cooling and scratching produced needles. These were re-

crystallized from cyclohexane to give 1.12 g (75%) of white needles

that could be sublimed at 500C/0.2-0.3 mm Hg, melting point 59-62C.

Analysis, calculated for C6H6NO2Br: C, 35.32; H, 2.96; N, 6.86;
Br, 39.16. Found: C, 35.57; H, 3.02; N, 6.84; Br, 39.01.

H NMR (CDC13) 6 3.53 (t, 2H), 3.95 (t, 2H), 6.75 (s, 2H).

C NMR (CDCI3) 6 28.17, 39.14, 134.22, 170.09.

IR 3450 (w), 3090 (m), 2924 (w), 1770 (m), 1700 (s), 1580 (m),

1440 (s), 1410 (s), 1372 (m), 1322 (m), 1255 (m), 1202 (m), 1138 (m),
1122 (m), 1060 (m), 1042 (w), 975 (w), 920 (m), 874 (w), 830 (s), 718

(w), 692 (s) cm-1
9-(2-3,6-Endoxo-1,2,3,6-tetrahydrophthalimidoethyl)adenine (14)

The same alkylation method previously described for the case of
the vinyl ether monomers 1-3 was applied for the synthesis of compounds
14, 15 and 17.

Thus, 5.0 g (37 mmole) of adenine was added to 1.9 g (41 mmole)
of a 50% sodium hydride oil dispersion weighed into a three-necked,
round-bottomed flask equipped with a gas inlet tube, mechanical

stirrer, thermometer and drying tube. A total of 300 ml of DMF was

poured in, and after stirring for 3 hours at 60C, 11.1 g (40.7 mmole)

of (11) was added. The suspension became almost clear after about 4

hours of reaction. It was left for 24 hours at 50-60C, a precipitate

forming by then and the liquid becoming dark brown. The precipitate

was filtered after cooling the flask and was washed repeatedly by

stirring vigorously in water and filtering until the pH of the fil-

trate became neutral. The DMF portion had the solvent reduced, and

upon standing afforded a solid which was treated as the first portion.

The fine white solid obtained was dried in vacuo at 500C, to finally

yield 6.9 g (57%) of a product that decomposed above 2050C.

Analysis, calculated for C15H14N603: C, 55.21; H, 4.32; N, 25.75.

Found: C, 55.26; H, 4.34; N, 25.77.

1H NMR (DMSO-d6) 6 2.78 (s, 2H), 3.73 (m, 2H), 4.29 (m, 2H), 5.05

(s, 2H), 6.51 (s, 2H), 7.15 (s, 2H), 8.00 (s, 1H), 8.10 (s, 1H).

In the case of this and compounds 15 and 19, 10% chromium(III)

2,4-pentanedionate (Cr acac) was added in order to enhance the 13C

NMR signals of the purine ring carbons.

C NMR (DMSO-d6/Cr acac) 6 46.81, 80.00, 118.45, 136.00, 140.34,

149.63, 151.99, 155.54, 175.48, -NCH 2H2N-, overlapped with solvent.

IR 3380 (m), 3320 (m), 3190 (m), 1765 (m), 1695 (s), 1646 (s),

1598 (s), 1472 (m), 1430 (m), 1412 (m), 1399 (m), 1368 (m), 1320 (m),

1295 (m), 1250 (m), 1188 (m), 1164 (m), 1096 (w), 1064 (w), 1015 (m),

930 (m), 880 (s), 845 (m), 798 (w), 722 (m), 698 (m) cm-1

6-Chloro-9-(2,3-endoxo-1,2,3,6-tetrahydrophthalimidoethyl)purine (15)

The sodium salt of 6-chloropurine was prepared as previously de-

scribed for the case of adenine by warming 5.0 g (32 mmole) of this

compound and 1.7 g (35 mmole) of a 50% sodium hydride oil dispersion

in 200 ml of DMF at 60C for 30 minutes. This time, however, the

purine dissolved completely and a yellow solution, rather than a white

suspension, was obtained. Addition of 9.68 g (35.6 mmole) of (11)

caused the solution to become brown. It was reacted for 24 hours at

60-650C, very little solid being filtered after this period. The DMF

was then removed to obtain a brown mass of solid to which 100 ml of

water and 100 ml of chloroform were added and the layers separated

after brief warming. The water layer was extracted twice with 100 ml

of chloroform, and the organic fractions were collected, washed with

100 ml of water and dried over anhydrous sodium sulfate. Reduction

of the solvent volume eventually led to the precipitation of a beige

solid that was later recrystallized from chloroform to give 5.20 g

(46.4%) of compound, melting point 175-177C.

Analysis, calculated for C15H12N503C1: C, 52.11; H, 3.50; N,

20.25; Cl, 10.25. Found: C, 52.19; H, 3.52; N, 20.20; C1, 10.19.
H NMR (CDC3) 6 2.77 (s, 2H), 3.98 (m, 2H), 4.48 (m, 2H), 5.20

(m, 2H), 6.49 (m, 2H), 8.04 (s, 1H), 8.75 (s, 1H).

C NMR (CDC13, int. ref. CDC13/Cr acac) 6 37.91, 42.15, 47.27,

80.85, 121.84, 131.59, 136.31, 145.23, 150.98, 151.81, 175.50.

IR 3110 (w), 3024 (m), 1770 (m), 1700 (s), 1595 (s), 1560 (s),

1500 (m), 1428 (m), 1390 (s), 1370 (s), 1330 (s), 1276 (m), 1200 (m),

1176 (m), 1138 (s), 1090 (w), 1012 (s), 950 (m), 920 (m), 870 (s),

850 (m), 790 (w), 725 (m), 700 (w), 648 (m), 634 (m) cm-1

1-(2-3,6-Endoxo-1,2,3,6-tetrahydrophthalimidoethyl)thymine (16)

Into a 200 ml three-necked, round-bottomed flask fitted with a

gas inlet, a drying tube and a thermometer were placed 100 ml of DMSO,

5.0 g (40 mmole) of thymine, 3.5 g (13 mmole) of (11) and 6.0 g (43

mmole) of anhydrous potassium carbonate. The suspension was magneti-

cally stirred at 50-600C under nitrogen flow for 24 hours. The solid

present was removed, and the filtered DMSO solution was evaporated at

reduced pressure. To the remaining solid was added 200 ml of water,

and after transferring to a separatory funnel, four 100 ml portions

of chloroform were added to extract the product. These portions were

collected and dried over anhydrous sodium sulfate overnight. At this

point, the solvent was evaporated, and the solid that remained was

recrystallized from water to afford 1.75 g (42.9%) of a white material

that decomposed above 2000C.

Analysis, calculated for C15H15N305: C, 56.78; H, 4.76; N, 13.24.

Found: C, 56.53; H, 4.82; N, 13.17.

H NMR (DMSO-d6) 6 1.66 (s, 3H), 2.89 (s, 2H), 4.16 (m, 4H),

5.07 (s, 2H), 6.54 (s, 2H), 7.17 (s, 1H), 11.18 (s, 1H).

C NMR (DMSO-d6, int. ref. DMSO-d6) 6 11.67, 36.96, 46.13, 47.10,

80.29, 107.98, 136.39, 141.51, 150.86, 164.32, 176.21.

IR 3420 (w), 3170 (m), 3040 (m), 1770 (m), 1698 (s), 1690 (s),

1460 (s), 1430 (s), 1400 (s), 1358 (m), 1335 (m), 1252 (m), 1230 (m),

1198 (m), 1152 (m), 1140 (s), 1028 (m), 950 (m), 930 (m), 875 (s),

850 (m), 820 (m), 754 (m), 720 (m), 640 (m) cm-1

1-(2-3,6-Endoxo-1,2,3,6-tetrahydrophthalimidoethyl)cytosine (17)
Preparation of the sodium salt of cytosine was achieved as pre-

viously described by reacting 5.0 g (45 mmole) of cytosine and 2.4 g

(49 mmole) of a 50% sodium hydride oil dispersion in 250 ml of DMF

for 4 hours at 60-700C. Addition of 12.3 g (45.0 mmole) of (11)

caused the white suspension to turn pink. The reaction was allowed

to proceed for 21 hours at 60-700C. The white solid that was present

was filtered, rinsed with acetone and washed three times by vigorously
stirring it in about 20 ml of water and filtering until the pH of the

filtrate became neutral. After drying in the vacuum oven, 5.16 g

(37.9%) of a product that decomposed above 200C was obtained.

Analysis, calculated for C14H14N404: C, 55.62; H, 4.67; N, 18.53.

Found: C, 55.67; H, 4.71; N, 18.52.

H NMR (DMSO-d6, int. ref. DMSO-d6) 6 2.86 (s, 2H), 3.62 (m, 4H),

5.07 (s, 2H), 5.49 (d, 1H), 6.52 (s, 2H), 7.09 (s, 2H), 7.25 (d, 1H).
1C NMR (DMSO-d6, int. ref. DMSO-d6) 6 47.05, 80.29, 93.06, 136.39,

145.89, 155.69, 166.02, 176.11, -NCH2CH2N- overlapped with solvent.

IR 3360 (s), 3090 (m), 1770 (m), 1700 (s), 1662 (s), 1620 (m),
1520 (m), 1490 (s), 1440 (s), 1394 (s), 1360 (s), 1334 (m), 1275 (m),

1222 (m), 1173 (m), 1150 (m), 1120 (m), 1092 (w), 1015 (m), 925 (m),

872 (m), 850 (w), 785 (s), 710 (m), 650 (m) cm-1

9-(2-Maleimidoethyl)adenine (18)

To 500 ml of bromobenzene placed in a 1,000 ml round-bottomed
flask equipped with a reflux condenser and magnetic stirrer was added

0.32 g (1.9 mmole) of 4-tert-butylcatechol. The solution was warmed

to just below its boiling point, and 6.23 g (19.1 mmole) of (14) was

slowly added to it. The suspension which was formed was refluxed for

1 hour. After this period, the mixture was orange-yellow. The bromo-

benzene was then evaporated under reduced pressure to afford a yellow

solid. In this manner, 4.39 g (89.0%) of a product that decomposed

at 220-224oC was obtained after repeatedly washing the solid with hot

methanol and acetone.

Analysis, calculated for C11H10N602: C, 51.16; N, 3.90; N, 32.54.

Found: C, 50.93; H, 3.96; N, 32.37.

H NMR (DMSO-d6) 6 3.81 (m, 2H), 4.29 (m, 2H), 6.92 (s, 2H), 7.16

(s, 2H), 8.04 (s, 1H), 8.09 (s, 1H).
1C NMR (DMSO-d6) 6 37.25, 41.17, 118.53, 134.37, 140.75, 149.62,

152.26, 155.72, 170.34.

IR 3340 (m), 3150 (m), 1698 (s), 1660 (s), 1600 (s), 1580 (s),

1488 (m), 1436 (m), 1420 (m), 1405 (m), 1358 (m), 1330 (m), 1305 (m),

1252 (m), 1140 (m), 1060 (w), 900 (w), 830 (m), 795 (w), 734 (w),
695 (m) cm-1

UV max (nm) 0.1N HC1 (258), 0.1N KOH (260), H20 (260); emax (H20)

= 1.63 x 104

6-Chloro-9-(2-maleimidoethyl)purine (19)

This monomer was synthesized by refluxing 5.19 g (15.0 mmole) of

(15) and 0.25 g (1.5 mmole) of 4-tert-butylcatechol in 200 ml of bromo-

benzene for 30 minutes. The solid present eventually dissolved, and

the solution became yellow. The bromobenzene was subsequently removed,

and the solid that remained was recrystallized from 1,2-dichloroethane

to yield 3.08 g (74.0%) of beige needles, melting point 213-2140C.

Analysis, calculated for C11H8N502CI: C, 47.58; H, 2.90; N,

25.22; C1, 12.77. Found: C, 47.61; H, 2.92; N, 25.18; C1, 12.71.
1H NMR (CDC13) 6 4.04 (m, 2H), 4.55 (m, 2H), 6.65 (s, 2H), 8.02

(s, 1H), 8.72 (s, 1H).

1C NMR (CDCl3/Cr acac) 6 37.18, 42.59, 124.09, 131.44, 134.22,

144.74, 151.18, 152.01, 169.75.
IR 3110 (w), 3070 (m), 1770 (w), 1710 (s), 1598 (m), 1568 (m),

1494 (w), 1438 (m), 1408 (s), 1338 (m), 1254 (w), 1145 (w), 1122 (m),
1048 (w), 962 (w), 938 (m), 833 (m), 692 (m), 635 (w) cm-

UV Amax (nm) 0.1N HC1 (266), 0.1N KOH (266), H20 (266); Emax H20=
8.64 x 10

l-(2-Maleimidoethyl)thymine (20)

A solution of 1.75 g (5.51 mmole) of (16), 0.092 g (0.55 mmole)

of 4-tert-butylcatechol and 100 ml of bromobenzene was refluxed for

30 minutes. The clear solution had the solvent removed, and the re-

maining white solid was washed with hot methanol. After cooling, it

was filtered and dried to afford 1.13 g (82.2%) of product that decom-

posed above 2400C.
Analysis, calculated for C11H11N304: C, 53.01; H, 4.45; N, 16.86.
Found: C, 52.89; H, 4.51; N, 16.77.
1H NMR (DMSO-d6) 6 1.70 (s, 3H), 3.70 (m, 4H), 7.02 (s, 2H), 7.45

(s, 1H), 11.16 (s, 1H).
1C NMR (DMSO-d6) 6 11.70, 35.92, 46.06, 108.59, 134.47, 141.05,

150.94, 164.10, 170.63.
IR 3150 (m), 3085 (m), 3000 (m), 2830 (m), 1700 (s), 1675 (s),

1482 (m), 1430 (s), 1408 (s), 1350 (s), 1256 (m), 1230 (w), 1120 (m),

1076 (w), 1040 (w), 1020 (w), 920 (m), 900 (m), 838 (m), 760 (m), 718

(w), 693 (s), 650 (w) cm-1

UV max (nm) 0.1N HCl (270), 0.1N KOH (266), H20 (270); emax
(H20) = 9.40 x 103
1-(2-Maleimidoethyl)cytosine (21)

In order to prepare this monomer, 3.0 g (9.9 mmole) of (17) and
0.16 g (0.99 mmole) of 4-tert-butylcatechol in 300 ml of bromobenzene
were refluxed for 2 hours without achieving complete dissolution.
After this period, the solvent was evaporated under diminished pres-
sure, and the remaining solid was washed by stirring in hot methanol,

filtered and dried. A total of 2.20 g (94.8%) of a solid that could

be recrystallized from methanol and decomposed above 270C was ob-

Analysis, calculated for C10H10N403: C, 51.28; H, 4.30; N, 23.92.
Found: C, 51.17; H, 4.34; N, 23.86.
1H NMR (DMSO-d6, int. ref. DMSO-d6) 6 3.67 (m, 4H), 5.51 (d,

1H), 6.98 (s, 4H), 7.38 (d 1H).
1C NMR (DMSO-d6, int. ref. DMSO-d6) 6 36.38, 47.05, 93.45,
134.49, 145.65, 155.74, 165.83, 170.65.

IR 3370 (m), 3100 (m), 1700 (s), 1665 (s), 1620 (s), 1520 (m),
1480 (s), 1438 (m), 1405 (m), 1385 (m), 1349 (m), 1270 (m), 1200 (w),
1160 (m), 1110 (w), 1070 (w), 904 (w), 835 (m), 800 (w), 780 (w),
750 (w), 720 (w), 690 (m) cm-1
UV max (nm) 0.1N HC1 (281), 0.1N KOH (273), H20 (271); Emax

(H20) = 8.91 x 103

N-Acetyl-1-(2-maleimidoethyl)cytosine (22)

Acetylation of this monomer was accomplished as described earlier
for the case of (4) by allowing 1.0 g (4.3 mmole) of (21), 10 ml of

acetic anhydride, 0.4 ml of pyridine and 150 ml of absolute ethanol

to reflux for 1 hour. Filtration while hot was performed, and the

solvent removed from the filtrate to yield a slightly orange solid.

Recrystallization from ethanol/water produced 0.25 g (21%) of a solid

that decomposed at 230-2350C.

Analysis, calculated for C12H12N404: C, 52.17; H, 4.38; N, 20.28.

Found: C, 52.20; H, 4.42; N, 20.35.

H NMR (DMSO-d6, int. ref. DMSO-d6) 6 2.07 (s, 3H), 3.75 (m, 2H),

3.94 (m, 2H), 6.99 (s, 2H), 7.03 (d, 1H), 7.96 (d, 1H), 10.74 (s,1H).

1C NMR (DMSO-d6, int. ref. DMSO-d6) 6 24.22, 35.82, 48.25, 94.99,

134.47, 150.01, 155.23, 162.25, 170.63, 170.78.

IR 3300 (m), 3110 (w), 1770 (w), 1710 (s), 1695 (s), 1660 (s),

1632 (m), 1552 (m), 1495 (s), 1435 (m), 1410 (m), 1372 (m), 1345 (m),
1305 (m), 1230 (m), 1150 (m), 1108 (m), 1020 (w), 902 (w), 830 (m),

790 (m), 750 (m), 693 (m) cm-1

9-(2-Succinimidoethyl)adenine (23)

The model compound in question was synthesized according to
Olson21 by reacting 0.20 g (0.77 mmole) of monomer (18) with 0.40 g
(7.2 mmole) of iron powder suspended in 15 ml of a 2:1 (v/v) solution

of water/glacial acetic acid placed in an Erlenmeyer flask. The ini-
tial suspension was magnetically stirred for 1 hour while being heated

to 900C on a hot plate. The mixture became a brown solution after

vigorous effervescence and dissolution of the metal. Upon cooling,

a red-brown precipitate formed which was removed, and to the filtrate

was added enough sodium bicarbonate to bring the solution to neutral-

ity. Extraction was performed with chloroform, and after drying over

magnesium sulfate and evaporation of the solvent, a gummy beige solid

was obtained. Recrystallization from ethyl acetate led to the isola-

tion of 0.075 g (37%) of white flakes that decomposed above 2200C.

Analysis, calculated for C11H12N602: C, 50.76; N, 4.65; N, 32.29.

Found: C, 50.66; H, 4.69; N, 32.28.
H NMR (DMSO-d6, int. ref. DMSO-d6) 6 2.45 (s, 4H), 3.56 (m, 2H),

3.73 (m, 2H), 7.16 (s, 2H), 8.09 (s, 2H).
C NMR (DMSO-d6, int. ref. DMSO-d6) 6 27.85. 118.55, 140.78,

149.84, 152.38, 155.84, 177.28, -NCH2CH2N- overlapped with solvent.

IR 3460 (m), 3290 (m), 3150 (m), 2980 (w), 1780 (m), 1700 (s),

1645 (s), 1580 (m), 1510 (w), 1474 (m), 1436 (m), 1416 (m), 1400 (s),

1355 (m), 1320 (m), 1299 (s), 1242 (m), 1188 (m), 1150 (s), 1080 (m),
1000 (w), 922 (w), 905 (w), 890 (w), 862 (w), 815 (w), 795 (w), 710

(w), 660 (m) cm-'
UV max (nm) 0.1N HC1 (258), 0.1N KOH (262), H20 (262); emax

(H20) = 1.50 x 104
Poly[9-(2-maleimidoethyl)adenine] (24)

All maleimide homopolymers and copolymers were synthesized by

reacting the dissolved monomers placed in a sealed polymerization tube
using azobisisobutyronitrile (AIBN) as the initiator.

Thus, to prepare poly[9-(2-maleimidoethyl)adenine], 1.5 g (5.8

mole) of (18) and 9.5 x 10-3 g (0.058 mmole) of AIBN were dissolved

in 20 ml of a 2:1 DMF/0.4 N HC1 solution. The solution was placed in

a polymerization tube, sealed under high vacuum and left in a 85-900C

oil bath for four days, after which the originally yellow solution

turned red. The solid obtained by precipitation of the red solution

into acetone was filtered and purified by dissolving it in the minimum

amount of DMSO and reprecipitation into acetone. Further purification

was achieved by first dissolving the polymer in 0.4 N HC1, pouring

into 100 ml of 0.1 N NaOH and dialyzing against water for 3 days.

Precipitation of the polymer occurred, and after evaporating the solv-

ent, the pink solid was washed with acetone and filtered. After dry-

ing in the vacuum oven, a total of 1.3 g (89%) of a pink solid was


Analysis, calculated for C11H10N602 + 0.25 H20: C, 50.28; H,

4.03; N, 31.99. Found: C, 50.45; H, 3.99; N, 30.44.
IH NMR spectra of polymers are characteristically difficult to

describe due to the broadness of their peaks. Therefore, they will be

reported by indicating the chemical shift for the maximum of the curves

observed along with the amount of protons that appear in that region.

Since high temperature was required for minimizing line broadening of

the signals,21 the temperature at which each 1H and 13C NMR spectrum

was taken will also be indicated.

H NMR (DMSO-d6, 250C) 6 3.39 (2H), 3.56 (2H), 4.09 (2H), 7.13

(2H), 8.09 (2H).

1C NMR (DMSO-d6, int. ref. DMSO-d6, 1100C) 6 118.74, 140.97,

149.79, 152.42, 155.69, 174.55, -NCH2CH2N- and -CH-CH- of

succinimide unit overlapped with solvent.

IR 3350 (s), 3220 (m), 1780 (w), 1710 (s), 1640 (s), 1600 (s),

1480 (m), 1405 (s), 1360 (m), 1330 (m), 1304 (m), 1250 (m), 1200 (w),

1160 (m), 1020 (w), 800 (w), 910 (w), 650 (m) cm-1

Intrinsic viscosity ([n]) (DMF, 300C) = 0.0421 dl/g.

UV "max (nm) 0.1N HCl (258), emax (0.1N HC1) = 9.60 x 103
Poly[6-chloro-9-(2-maleimidoethyl)purine] (25)

Into a polymer tube was placed a solution of 0.50 g (1.8 mmole)

of (19) and 5.9 x 10-3 g (0.036 mmole) of AIBN in 4 ml of DMF. After

sealing under vacuum, the tube was immersed in a 40C water bath which

was irradiated with a 100 watts UV quartz lamp (Hanovia) for 6 days.

Some precipitate formed by then, and the solution was orange. The

product was precipitated into methanol and later dissolved in DMSO

and precipitated again into methanol to yield, after filtration and

drying, 0.16 g (32%) of a pale-pink solid.

Analysis, calculated for C11H8N502CI: C, 47.58; H, 2.90; N,

25.22; C1, 12.77. Found: C, 47.45; H, 2.94; N, 25.06; C1, 12.48.
H NMR (DMSO-d6, int. ref. DMSO-d6, 900C) 6 3.92 (2H), 4.47 (4H),

8.50 (2H).

C NMR (DMSO-d6, 400C) 6 104.31, 139.36, 146.31, 148.01, 150.82,

175.77, -NCH2CH2N- and -CHCH- of succinimide unit overlapped with


[n] (DMF, 300C) = 0.0254 dl/g.

IR 3440 (m), 3100 (w), 2950 (w), 1775 (w), 1700 (s), 1590 (s),

1560 (s), 1496 (m), 1435 (m), 1400 (s), 1332 (s), 1260 (w), 1210 (m),
1170 (m), 1140 (m), 940 (w), 856 (w), 790 (w), 640 (w) cm-1.
Poly[1-(2-maleimidoethyl)thymine] (26)
This polymer was prepared by placing 1.0 g (4.0 mmole) of (20)
and 0.033 g (0.20 mmole) of AIBN in 12 ml of a 5:1 DMF/0.4N HC1 solu-
tion. After pouring the mixture into a polymerization tube which was
subsequently evacuated and sealed, it was reacted at 80-90C for 5
days. A small amount of solid remained undissolved, and the liquid
was red. Precipitation was performed in ether and purification accom-
plished by dissolution of the polymer in the minimum amount of DMF,
reprecipitation into ether, and subsequent dialysis as in the case of
polymer (24). A total of 0.30 g (30%) of pink polymer was obtained.
Analysis, calculated for C11H11N304 + 0.25 H20: C, 52.07; H,

4.57; N, 16.56. Found: C, 51.77; H, 5.06; N, 16.34.
1H NMR (DMSO-d6, int. ref. DMSO-d6, 100C) 6 1.72 (3H), 3.05 (2H),

3.68 (4H), 7.28 (1H), 10.51 (1H).
13C NMR (DMSO-d6, int. ref. DMSO-d6, 100C) 6 10.99, 44.91,

108.51, 140.43, 150.52, 163.48, 172.69, -NCH2CH2N[C(0)]2 and -CHCH- of
succinimide unit overlapped with solvent.
IR 3450 (m), 3180 (m), 3040 (m), 2810 (w), 1770 (m), 1700 (s),

1465 (m), 1430 (s), 1400 (s), 1340 (s), 1250 (m), 1200 (m), 1160 (m),
1135 (m), 1010 (w), 905 (w), 770 (m), 680 (w) cm-1
[n] (DMF, 30C) = 0.0632 dl/g.

Poly[l-(2-maleimidoethyl)cytosine] (27)

Synthesis of this polymer was accomplished by placing 0.57 g

(2.4 mmole) of (21) and 4.0 x 10-3 g (0.024 mmole) of AIBN dissolved

in 12 ml of a 2:1 DMF/0.4N HC1 solution in a polymer tube, sealing

under high vacuum and letting it react for 5 days at 80-900C. The

solution was red after this period. Precipitation of the product was

done by using acetone. After redissolving in DMSO, the polymer was

reprecipitated into acetone. Further purification involved dialyzing

the polymer as described before for (24). The precipitated polymer

was then filtered and dried to afford 0.48 g (83%) of pink solid.

Analysis, calculated for C10H10N403"H20: C, 47.62; H, 4.80; N,

22.21. Found: C, 48.10; H, 4.80; N, 21.17.
1H NMR (DMSO-d6, int. ref. DMSO-d6, 1000C) 6 2.96 (2H), 4.20 (4H),

6.37 (3H), 8.10 (1H).
13C NMR (DMSO-d6, int. ref. DMSO-d6, 1000C) 6 37.01, 46.47, 93.74,

146.62, 154.75, 164.12, 175.38, -CHCH- of succinimide unit overlapped

with solvent.

IR 3250 (m), 3200 (m), 1770 (w), 1700 (s), 1642 (s), 1520 (m),

1490 (s), 1432 (m), 1388 (s), 1355 (m), 1280 (w), 1175 (m), 1130 (w),

785 (m), 610 (w) cm-1

[n] (0.1N HC1, 30C) = 0.0042 dl/g.
Poly[l-(2-vinyloxyethyl)thymine] (28)
A total of 1.0 g (5.1 mmole) of monomer (2) was dissolved in 10
ml of chloroform. After cooling the solution to 00C by means of an

ice/salt bath under nitrogen flow, a solution of 0.013 g (0.051 mmole)
of iodine dissolved in 1 ml of chloroform was added. The initially

purple iodine solution turned brownish yellow. It was stirred at 0C

for 22 hours, after which 1 ml of methanol and 1 ml of a saturated

sodium thiosulfate solution were added, the yellow color disappear-

ing. The layers were eventually separated and the organic one dried

over anhydrous sodium sulfate. After filtration, precipitation of the

polymer was obtained in ether to afford 0.23 g (22%) of product.

Analysis, calculated for C9H12N203: C, 55.09; H, 6.17.

Found: C, 50.40; H, 6.60.
H NMR (CDC13, 50C) 6 1.75 (3H), 3.08 (4H), 3.64 (3H), 7.41

(1H), 9.83 (1H).
C NMR (CDCI3, 500C) 6 12.13, 40.60, 48.54, 63.11, 72.78,

110.05, 141.63, 151.18, 164.54.

IR 3500 (w), 3190 (w), 3050 (w), 2940 (w), 1695 (s), 1670 (s),

1468 (s), 1350 (m), 1240 (m), 1110 (s), 900 (m), 765 (w), 685 (w) cm-1

[n] (DMF, 300C) = 0.0227 dl/g.
mine] (29)

Preparation of this copolymer was accomplished by placing 1.3 g

of (19) and 0.94 g of (2) in equimolar (4.8 mmole) amounts in 25 ml

of dichloromethane along with 7.9 x 10-3 g (0.048 mmole) of AIBN and

pouring the mixture into a polymer tube which, after evacuation and

sealing, was left to react in a 600C bath for 5 days. All the solid

initially present eventually dissolved, and after about 1 hour, a

white precipitate started to form. This was eventually filtered and

dried in vacuo. A total of 1.9 g (83%) of a white, powdery solid was

obtained after redissolving it in DMF and reprecipitation in acetone.

Analysis, calculated for C20H20N705C1: C, 50.69; H, 4.25; N,

20.69; Cl, 7.48. Found: C, 50.59; H, 4.30; N, 20.57; C1, 7.37.
H NMR (DMSO-d6, int. ref. DMSO-d6, 1000C) 6 1.69 (3H), 3.02

(2H), 3.79 (7H), 4.50 (4H), 7.13 (1H), 8.51 (1H), 8.66 (1H), 10.68 (1H).
C NMR (DMSO-d6, int. ref. DMSO-d6, 1000C) 6 11.14, 32.96,

46.86. 66.78, 75.02, 108.17, 130.49, 140.77, 146.67, 148.81, 150.37,

150.96, 151.79, 163.53, 174.25, 177.62, -NCH2CH2N- and -CHCH- of

succinimide unit overlapped with solvent.

IR 3470 (m), 3060 (w), 2950 (w), 1774 (m), 1698 (s), 1592 (s),

1562 (s), 1500 (w), 1465 (m), 1436 (m), 1404 (s), 1335 (s), 1276 (w),

1260 (m), 1220 (m), 1172 (m), 1142 (m), 1100 (m), 940 (m), 858 (w),
790 (w), 636 (m) cm-

[n] (30C, DMF) = 0.147 dl/g.
Copoly[N-(2-bromoethyl)maleimide/1-(2-vinyloxyethyl)thymine] (30)

This copolymer was prepared by reacting 0.62 g (3.0 mmole) of

(13), 0.60 g (3.0 mmole) of (2) and 5.0 x 10-3 g (0.030 mmole) of
AIBN dissolved in 10 ml of dichloromethane. Polymerization was ac-

complished after placing the reagents in a polymerization tube, seal-
ing and placing it in a 60C bath for 3 days. The slightly yellow

solution eventually produced a white precipitate that was filtered,

redissolved in DMSO and reprecipitated into acetone. A total of 0.78

g (64%) of copolymer was obtained.
Analysis, calculated for C15H18N305Br: C, 45.01; H, 4.53; N,

10.50; Br, 19.96. Found: C, 44.76; H, 4.64; N, 10.35; Br, 20.05.
1H NMR (DMSO-d6, int. ref. DMSO-d6, 1100C) 6 1.74 (3H), 2.89 (2H),

3.47 (7H), 3.70 (4H), 7.23 (1H), 9.28 (1H).

C NMR (DMSO-d6, int. ref. DMSO-d6, 110C) 6 11.09, 27.71, 32.70,

39.50, 47.00, 67.11, 75.35, 108.12, 140.87, 150.37, 163.48, 174.60,

177.76, -CHCH- of succinimide unit overlapped with solvent.

IR 3470 (w), 3200 (w), 3040 (w), 2920 (w), 1770 (m), 1700 (s),

1462 (m), 1430 (m), 1396 (s), 1350 (m), 1240 (m), 1150 (m), 1100 (m),
900 (w), 766 (w), 684 (w) cm-1.

[n] (DMF, 30C) = 0.0927 dl/g.
Copoly[maleic anhydride/1-(2-vinyloxyethyl)thymine] (31)

A solution of 0.25 g (2.5 mmole) of maleic anhydride, 0.50 g (2.5

mmole) of 1-(2-vinyloxyethyl)thymine and 4.2 x 10-3 g (0.025 mmole) of

AIBN in 8 ml of dichloromethane was placed in a polymerization tube,

evacuated and left at 65C for 1 day. The precipitate formed was

dropped into ether, filtered, dissolved in DMF and reprecipitated into

an excess of acetone. The polymer was filtered and dried in vacuo to

afford 0.52 g (70%) of a white powder.

Analysis, calculated for C13H14N206: C, 53.06; H, 4.79; N, 9.52.
Found: C, 50.97; H, 5.04; N, 9.91.
H NMR (DMSO-d6, int. ref. DMSO-d6, 70C) 5 1.73 (3H), 2.73 (2H),

2.89 (2H), 3.30 (2H), 3.76 (3H), 7.21 (1H), 10.65 (1H).
1C NMR (DMSO-d6, int. ref. DMSO-d6, 1100C) 6 11.04, 46.35, 66.54,

75.07, 108.02, 141.46, 150.38, 163.54, 172.07, -CHCH- of succinimide

unit and -CH2CHO- overlapped with solvent.

IR 3480 (w), 3200 (w), 3060 (w), 2930 (w), 1855 (m), 1780 (s),

1670 (s), 1470 (m), 1350 (m), 1235 (m), 1100 (m), 924 (m), 770 (w),
682 (w) cm-1

[n] = 0.332 dl/g.

Copoly[9-(2-maleimidoethyl)adenine/1-(2-vinyloxyethyl)thymine] (32)

Three approaches were investigated in order to obtain this copoly-


Approach A: The direct copolymerization of 9-(2-maleimidoethyl)-

adenine and l-(2-vinyloxyethyl)thymine under dilute conditions was

carried out as follows.

Into a polymerization tube were placed 0.79 g (3.0 mmole) of (18),

0.60 g (3.0 mmole) of (2), 5.0 x 10-3 g (0.030 mmole) of AIBN and 50

ml of chloroform. The tube was sealed under vacuum and left at 60-700C

for 4 days. After this period, it was opened and the solvent evapo-

rated. The residue which remained was dissolved in the minimum amount

of DMSO and precipitated into acetone. The solid that formed was fil-

tered and dried to yield 0.34 g (24%) of product after redissolving it

in DMSO and reprecipitation in acetone.

Approach B: Incorporation of adenine into preformed copolymer

(30) was accomplished by following Inaki's method.7

Thus, a suspension of sodium adenide in DMF was prepared by

reacting 0.51 g (3.7 mmole) of adenine and 0.18 g (3.7 mmole) of

sodium hydride in 200 ml of DMF for 1 hour at 60C. The sodium salt

dissolved completely, and after cooling in an ice bath, 0.50 g (1.2

mmole) of (30) were slowly added. The solution obtained turned from

initially pink to bluish over a period of 1 week at 40C. The suspen-

sion formed by then had the solvent removed, and the residue was

stirred in water. Enough 0.1N HC1 was added to bring the pH to neu-

tral, and repeated washings of the solid were performed by stirring

it in water followed by filtration. The product obtained after these


steps was dissolved in DMSO and precipitated into acetone. After

filtration and drying, a total of 0.32 g (57%) of copolymer was ob-


Approach C: Pitha's and Pitha's33 strategy for the transforma-

tion of the chloro substituent into an amino group in the synthesis

of poly(9-vinyladenine) was used in this case.

Into a pressure bottle were placed 0.85 g (1.8 mmole) of (29)

and 100 ml of methanol. The suspension was treated with ammonia gas

until saturation was achieved by bubbling the gas while cooling the

reaction vessel by means of an isopropanol/dry ice bath. The pres-

sure bottle was stoppered and left in a 50C oil bath for 1 week.

The suspension initially turned pink and eventually became grayish.

After removal of the solvent, the residue was stirred in 25 ml of

water, filtered and dried. Purification was carried out by dissolv-

ing the polymer in DMSO and reprecipitation into acetone. After

filtration and drying, 0.78 g (96%) of the desired copolymer were


Analysis, calculated for C20H22N8C10: C, 52.85; H, 4.88; N,

24.66; C1, 0.00. Found: C, 52.57; H, 4.94; N, 24.50; C1, 0.00.

H NMR (DMSO-d6, int. ref. DMSO-d6, 90C) 6 1.69 (3H), 3.13 (5H),

3.76 (6H), 4.31 (2H), 6.78 (2H), 7.15 (1H), 7.99 (1H), 8.11 (1H),

10.96 (1H).

13C NMR (DMSO-d6, int. ref. DMSO-d6, 900C) 6 11.18, 32.42,

47.05, 65.96, 72.11, 108.27, 118.45, 140.48, 149.60, 150.52, 152.08, 155.49

163.73, 175.28, 177.63, -NCH2CH2- and -CHCH- of succinimide unit over-

lapped by solvent.


IR 3330 (m), 3200 (m), 2950 (w), 1772 (w), 1700 (s), 1635 (s),

1595 (m), 1472 (m), 1434 (m), 1400 (m), 1355 (m), 1330 (m), 1300 (m),

1245 (m), 1152 (m), 1100 (w), 1070 (w), 920 (w), 795 (w), 768 (w),
645 (w) cm-I

Figure II-1 shows the structures of additional compounds prepared

which were not shown in previous Figure I-1.


N ,


(6) Br


HN0 CH 3





II-1. Structures of several compounds prepared.


0 C












Synthesis of the desired vinyl ether and maleimide monomers could

be accomplished, in principle, by direct alkylation of the preformed

nucleic acid bases or through a series of steps involving formation of

these bases by ring closure reactions. The former approach, that is,

the direct alkylation method, was utilized successfully for the prep-

aration of both series of monomers due to its greater simplicity. The

principal disadvantage of this approach arises from the fact that each

of the bases has several sites susceptible to alkylation. It was

therefore necessary to use procedures that gave alkylation predomi-

nantly at the desired site and to confirm the structures of the prod-

ucts obtained by UV spectroscopy.

Since the alkylation of a suspension of sodium adenide in DMF has

been shown24 to give primarily 9-substituted product, it was the ap-

proach chosen for the preparation of the adenine derivatives. In this

manner, 9-(2-vinyloxyethyl)adenine was obtained using 2-chloroethyl

vinyl ether as the alkylating species. Likewise, treatment of cyto-

sine and thymine with sodium hydride led to their sodium salts, and

alkylation at the desired position of the ring was accomplished by

using the same alkylating reagent (Scheme III-I).

Scheme III-1:
B = adenine, thymine or cytosine

1. NaH

It should be noted that two other routes to this alkylation were

also investigated. The first involved the preparation of hydroxyethyl

derivatives of the bases (33)by treatment with ethylene carbonate34 fol-

lowed by transetherification using a vinyl ether and mercuric acetate

as the catalyst.336 Even though there was no difficulty in obtain-

ing the hydroxyethylated bases, the transetherification step did not

afford any vinyl ether product. The second route would have taken

advantage of the increased reactivity toward alkyl halides possessed

by trimethylsilyl derivatives (34)of the nucleic acid bases37'38 which

could be prepared by reaction with hexamethyldisilazane or trimethyl-

silyl chloride.39 In this case, no alkylation was evident, and this

approach was also abandoned (Scheme 111-2).

Scheme 111-2:


B > BCH2CH2 Hg(OAc)2

B -- -- > B(SiMe3)x----- --------

x = 1, 2

In order to obtain the maleimide monomers, the first approach

attempted was to prepare a suitably derivatized maleimide ring which

could be attached later to the nucleic acid bases using a similar al-

kylation procedure as in the vinyl ether series. Thus, N-(2-chloro-

methyl)maleimide (35) was synthesized according to the method described

by Tawney et al.40 Preparation of N-(2-chloroethyl)maleimide (12), a

compound previously reported in the literature,31 by first treating

maleimide with ethylene carbonate and eventual chlorination proved un-

successful. This compound was obtained, however, by the reaction of

maleic anhydride with 2-chloroethylamine and subsequent closure of the

ring, although in a very low yield. The same route was utilized in

the synthesis of N-(2-bromoethyl)maleimide (13) with poor results. An

easier synthetic approach which led to moderate yields of (12) and (13)

was the retro Diels-Alder reaction performed on (10) and (11), respec-

tively. The reason and method for preparing the two latter bicyclic

compounds will be discussed later.

The last step in this synthetic strategy, that is, the attachment

of the maleimide moiety to the nucleic acid base, was tried employing

a variety of basic catalysts including potassium carbonate, sodium hy-

droxide and triethylamine. In all cases, either no reaction occurred

or polymerization of the N-(2-haloethyl)maleimides was immediate.

In addition, alkylation to the trimethylsilyl derivatives of ade-

nine and thymine was also investigated as in the vinyl ether deriva-

tives case, but, again, no alkylation took place.

Maleamic acids (5) and (6) were also successfully prepared. Al-

kylation of the nucleic acid base at this step with hope of eventual

ring closing was attempted, but without success (Scheme 111-3).

Scheme 111-3:

0 0 0

0 (35) B


0 0
(12) C1
(13) Br


X' X"
(5) C1 (10) Cl
(6) Br (11) Br

SNCH 2CH2 X> (12), (13)

N-Carboxyalkylmaleimides (n = 1 and 2) (36) were prepared by re-
ported procedures,41-43 with better results by Rich's et al. method43
which involved the preparation of (37). No desired compound (38) was
obtained when (36) was reacted with the nucleic acid bases in the pre-
sence of coupling reagents such as DCC or isobutyl chloroformate.

Other routes which were contemplated included the alkylation of
maleimide44 with 2-chloroethyl derivatives of the nucleic acid bases
(39),45 and the preparation of 2-aminoethyl base derivatives24 (40),
followed by closure of their addition products to maleic anhydride.
The second route had as an obvious disadvantage the reactivity of the
amino groups of some of the bases toward maleic anhydride, thus sug-
gesting the need for initial protection of such groups to prevent the
formation of undesired products (Scheme 111-4).
Scheme 111-4:

0 \ NH2(CH2)nCO2H


>l H



DB > N(CH2)n CB


[ 0 NCH2CH2B

(40) 0

It was finally realized that direct alkylation employing a male-
imide ring was not the best approach, and a synthetic route was de-
signed in which the double bond of the maleimide moiety was protected46
in the form of Diels-Alder adduct (11). This key alkylation reagent
was prepared using the same approach that had yielded maleamic acids
(5) and (6). After alkylation with the bases, heating in bromobenzene
led to deprotection of the maleimide double bond as the last step.
This route proved successful for the preparation of the adenine, 6-
chloropurine, thymine and cytosine derivatives. It also afforded (13)
in reasonable yield from (11) as mentioned before (Scheme 111-5).
Scheme 111-5:


0 / 0 0

CH r AcO2/NaAc OH
(11) NHCH2CH2Br

B/NaH 0



In case further purification of the compounds prepared was needed
as indicated by thin layer chromatography, column chromatography or
soxhlet extraction was carried out using the appropriate solvents.

The solubility of all maleimides, except for the chlorine contain-

ing derivative (19), was very low in common organic solvents such as

benzene, chloroform, ethyl acetate, 1,2-dichloroethane and ethanol.

However, the adenine and cytosine maleimide derivatives were soluble

in acidic solution. The vinyl ether derivatives exhibited, on the

other hand, rather good solubilities in polar organic solvents, the

thymine and N-acetylcytosine derivatives being especially easy to dis-

solve in most.

Representative spectra for the vinyl ether and maleimide monomers

are shown in Figures III-1 4 for the case of monomers (2) and (18).

The spectrum of (2) exhibits a very clear 1H NMR ABX pattern for the

vinyl protons labeled below.47
T-- 2-CH2-2--0-C = C

46.49 65.11 151.33 HA

The calculated chemical shifts and coupling constants are as follows:

vX = 6.49, vA = 4.23, vB = 4.01 ppm; JAX = 14.2, JBX = 6.8 and JAB =

2.0 Hz. The dimethylene chain protons appear as a singlet at 6 3.88.

The 13C NMR resonances for the carbons composing the vinyloxyethyl

group are indicated above.

The 1H NMR spectrum of maleimide (18) shows a singlet for the

maleimide vinyl protons at 6 6.92. The dimethylene group appears as

an AA'XX' pattern with the protons on the carbon next to the adenine

ring nitrogen resonating at 6 4.29 and those on the adjacent chain

carbon at 6 3.81. The resonances for the carbons of the maleimido-

ethyl group are shown below.

0 -








4* -



r z LL
3: -
*-< +-
u i
CO *r
+-> )



















01 0








C- .
C o


C. 0


4L 0


0 0
* C)

C-1 Ll




















i Q)


37.25 41.17

0 170.34

From the above discussion it can be seen that the polymerization

process could be followed by monitoring the disappearance of the vinyl

peaks, since the regions in which they are found are not obscured by

any other resonances and are easily interpreted.

Ultraviolet spectroscopy has been used successfully in the differ-

entiation of adenine derivatives substituted at various nitrogen atoms.

For this, several substituted adenines have been prepared by unambigu-

ous, independent routes and their absorption curves compared under

acidic, neutral and basic conditions. This analysis has shown definite

trends in the absorption maxima observed for sets of comparably sub-

stituted adenines.48 Table III-1 presents spectral data obtained for

the compounds prepared and Table III-2 those of several 1-, 3-, 7- and

9-substituted adenine derivatives reported in the literature. It can

be seen that 9-substituted adenines exhibit comparatively low absorp-

tion maxima (257-263 nm) under all conditions, while 7- and 3-substi-

tuted derivatives absorb at much higher values (270-274 nm), and 1-

substituted adenine shows a large difference between its acidic and

basic maxima (259 and 270 nm, respectively). Analysis of this informa-

tion served to confirm the 9-substitution assigned to the purine deriva-

tives prepared in this study.

UV Absorption

Table III-1

Data of the Nucleic Acid Base


Compound 0.1N HC1 0.1N KOH Neutral

( 1) 258 260 260

(18) 258 260 260

(14) 258 260 258

(23) 258 262 262

(19) 266 266 266

( 2) 271 269 270

(20) 270 266 270

(3) 280 272 272

(21) 281 273 271

a Measurements taken in water at room temperature.
bmax = wavelength at which maximum absorbance occurs
given in nanometers.

Table III-2

UV Absorption Data of Some N-Substituted Adeninesa


3 9

Compound max
R Nc H+ OH- Neutral Reference

CH2CH20C(O)C=CH2 9 263 45
CH2CH20P(0)(OH)2 9 261.5 49

CH2CH(OH)CH20H 9 261 50

CH2CH(NH2)CO2H 9 257 262 51

CH=CH2 9 261 52

CH2CH2CH3 9 259 261 261 24

CH3 9 260 260 53

CH2CH(OH)CH20H 3 274 272 273 50

CH3 3 274 273 53

CH3 1 259 270 53

CH3 7 272 270 53

a Measurements taken in water at room temperature.
bmax = wavelength at which maximum absorbance occurs given in
c N = 1,3,7 or 9 and represents the site of substitution on the
adenine ring.

Similar trends can be observed for the cytosine series, in which

the maximum absorbances for the 1-substituted derivatives (272-283.5

nm) exhibit marked differences from those of 3-substituted cytosine

(274-294 nm). The same rationale was used to confirm the 1-substitu-

tion at the thymine ring (Table 111-3).

Correct assignment for the 9-substituted chloropurine derivatives

(15) and (19) was confirmed by reacting (15) with methanolic ammonia

for 3 days at 40C and comparing the spectra of the compound obtained

with that of previously prepared (14). The absorption maxima obtained



(15) (14)

under acidic, basic and neutral conditions for (14) are shown in Table

III-1 and are identical to those of the product of the reaction shown

above. In addition, H NMR and IR data also proved that the product

obtained from the ammonolysis is authentic (14).

6-Chloro-9-(2-maleimidoethyl)purine (19) would also be expected

to have the substitution assigned following this rationale (even though

all its absorption maxima occur at 266 nm), since the retro Diels-Alder

deprotection step would not alter the alkylation position.

NMR analysis has also proved to be of value for differentiating

between isomerically substituted adenines and purines. In this case,

Table III-3

UV Absorption Data of Some N-Substituted Thymines and Cytosinesa

HN C 3 N3

1 1'

Compound Xmaxc
R Nb H+ OH- Neutral Reference

CH2CH2OC(O)C=CH2 1 273 45
CH2CH2OP(0)(OH)2 1 272 51

CH2CH(NH2)CO2H 1 269 271 51

CH2CH2CH3 1 272.5 270.0 272 24

CH2CH(OH)CH20H 1' 272 50

CH2CH(NH2)CO2H 1' 279 275 51

CH2CH2CH3 1' 283.5 274 274 24

CH3 1' 283 274 54

CH3 3 274 293 294 54

a Measurements taken in water at room temperature.
N = 1,1' or 3 and represents the site of substitution on the
pyrimidine rings.
c max = wavelength at which maximum absorbance occurs given in

a larger chemical shift difference (As) between the signals correspond-

ing to the 2- and 8-aromatic protons of the purine ring is observed for

the 1- or 3-substituted derivatives [22-32 Hertz (Hz)] than for the

same protons of the 7- or 9-substituted compounds (2-11 Hz) in DMSO.48

Table III-4 shows the differences in chemical shifts of such protons

for the compounds prepared and for several purines and adenines re-

ported. The A6 values of the purines synthesized thus seem to agree

only with 7- or 9-substitution at the purine ring. Moreover, the A6

for (15) decreases on treatment with ammonia to that expected for the

9-substituted adenine.


Homopolymerization of maleimide derivatives (18-21) was conducted

via a radical mechanism using AIBN as the initiator. The homopolymer

of l-(2-vinyloxyethyl)thymine was obtained by a cationic mechanism us-

ing iodine as the initiator.

In the case of the synthesis of copolymer (32), the first attempts

at polymerizing 9-(2-maleimidoethyl)adenine (18) and l-(2-vinyloxy-

ethyl)thymine (2) via a radical mechanism were unsuccessful. This was

due to the very low solubility of the maleimide monomer, which pre-

vented the use of non-polar solvents such as dichloromethane, that are

known to increase the association constant for the formation of a

charge-transfer complex leading to the copolymerization.16 Thus, very

little, if any copolymer was obtained when using a high concentration

of both monomers in dipolar aprotic solvents such as DMF or DMSO, in

which mostly homopolymers of the maleimide monomers were produced.

Therefore, after several trials using this approach which nevertheless

Table III-4

Differences in the Chemical Shifts
of the 2- and 8-H of N-Methyladenines and Purinesa

CH3 N H8

Compound X Nb 6 C Reference
(1) 5
(18) 5
(14) 12
(23) 0
(19) 68
(15) 20
H 1 22 48
H 3 32 48
NH2 1 26 48
NH2 3 28 48
H 7 2 48
H 9 11 48
NH2 7 6 48
NH2 9 3 48

a Spectra taken in DMSO-d6.
b N = 1, 3, 7 or 9 and represents the site of substitution on the
purine ring.
c A6 = 16H8 6H21 = chemical shift differences given in Hertz.

afforded a small conversion of copolymer using chloroform at high dilu-

tion, other approaches were sought.

Incorporation of bases into preformed, stereoregular polymers such

as l-(2-bromoethoxycarbonyl)-1-methylethylene7 was chosen as an alter-

native route. Thus, alternating copolymer (30), prepared from N-(2-

bromoethyl)maleimide (13) and (2), was synthesized in dichloromethane.

A suspension of sodium adenide was reacted with this copolymer and in-

corporation, as evidenced by spectral analysis, took place. However,

rather low yields of copolymer were obtained by this method.

Another approach, successfully applied by Pitha and Pitha33 in the

preparation of poly(9-vinyladenine) by treatment of poly(6-chloro-9-

vinylpurine) with ammonia, was also investigated. For this, purine (19)

was synthesized, and its high solubility in non-polar solvents allowed

it to copolymerize with (2) to give copolymer (29). After treatment of

the latter with ammonia, the desired transformation of the chloro group

into the amino functionality took place, leading to desired copolymer

(32) containing the bases adenine and thymine (Scheme 111-6).

Use of the same strategy was attempted for the copolymerization of

N-acetyl-1-(2-vinyloxyethyl)cytosine (4) and (19). The product ob-

tained would have been eventually converted into a copolymer having cy-

tosine and adenine as the pendant bases. This approach seemed very

convenient since both the acetyl and chloro functionalities could be

removed using the ammonolysis treatment. However, the desired copoly-

mer could not be obtained even though a variety of reaction conditions,

which included varying the kind and amount of initiator, the total mono-

mer concentration, different temperatures and radical initiation tech-

niques such as heat, UV irradiation and redox systems were tried.

Copolymerization Approaches:
Scheme 111-6:

Br HC:N N i N >

1 CH2CH2 Br
'C 2 (30) 0 E

p C CH2 CHN N j N
cH3 (18) HN ,CH
; yN --- 0 0
(2) CH2CH20CH=CH2 2 (32)


(19) 0 0

N,> 0 C
N C3
Copolymer (31), prepared from maleic anhydride and (2), was also
synthesized to aid in the study of other similar polymers containing
the thymine vinyl ether derivative. It was found to be very helpful
in the assignment of both the 1H and 1C NMR resonances of polymers
(29) and (30) and in analyzing the data from UV and NMR experiments,
as will be seen later on.

It had been shown before that vinyl ether and maleimide monomers

copolymerize in an alternate fashion.21'55 The experimental results in

the present case indicate that again this is true. In the first place,

elemental analyses setting both monomers as the repeating unit in a

1:1 ratio agree well with the calculated values for both (29) and (32),

as seen from Table 111-5. The 1H NMR analysis of the aromatic region

indicates that the areas are roughly in accord with the amount of pro-

tons expected for a 1:1 ratio of monomers. Also, a comparison of the

1C NMR chemical shifts of similar carbons in the homopolymers with

those of the copolymer shows marked differences (Figure 111-5).

Of considerable importance is the fact that the copolymers formed

contain three chiral centers along the chain which were not present in

the monomers. This leads to differences in several carbons which are

stereochemically sensitive and were used in the study of the mechanism

of copolymerization of the NPM/CEVE system21 whose backbone structure

is similar to those of the copolymers under present consideration. It

should be pointed out that in this study, however, no stereochemical

analysis was carried out.

Since the groups on the monomers are rather bulky and the polymeri-

zation was performed via a radical mechanism, it may be suggested that

a random placement of the pendant bases along the chain should be the

most favorable arrangement. This stereoregularity should exert a large

influence on the way the bases interact in the same and between dif-

ferent chains.

The molecular weights of some maleimide homopolymers were esti-

mated from their intrinsic viscosities by using the expression shown



C )

co r- C) CD
r- : o

O u r_ LO C:) c: L
a, C' CJ C) C'

4- C ) :-
,o L

0 0
}r -
7 Ln ID 2 D

L S V __ +- 3n
5~a \u /=\ a3 i
^ )-''-^^ =0 3
c ^ / ^-
1 / ^=
5 % s
o i / o 3
\ra o -z> M r
\ c^ c



0 1

[n] = 2.61 x 10-4 M 0.701

where [n] is the intrinsic viscosity in dl/g determined in DMF at 300C,

and Mn is the number-average molecular weight. This equation was ini-

tially established for the polymaleimide case by Bamford, Bingham and

Block56 and has been used frequently558 in the calculation of molecu-

lar weights of other polymaleimides. Thus, M was found to be about

1,400 and 2,500 for (24) and (26), respectively. This indicates that

the degree of polymerization of the polymaleimides was very low (~ 5-

10), leading only to oligomers of these species. The intrinsic vis-

cosity of copolymer (29) amounted to 0.147 dl/g. This is appreciably

larger than the values obtained for the polymaleimides and may suggest

a longer chain length for this copolymer and copolymer (32).

All polymaleimides, in accord with literature reports,57 were pink

powders insoluble in most organic solvents except DMSO and DMF, in the

latter only to a limited extent. Polymers (24), (26) and (27) are

somewhat hygroscopic, as evidenced by the elemental analyses and ap-

pearance on handling. Polyvinyl ether (28), on the contrary, was

readily soluble in many solvents. The copolymers prepared also exhib-

ited very low solubility in most organic solvents except DMSO and DMF.

The properties of the polymers described above will serve to ex-

plain the results obtained from the physical studies discussed in the

next section.

Physical Studies

Nuclear Magnetic Resonance Spectroscopy

The 1H and 13C NMR spectra59'60 of nucleic acid bases provide a

sensitive tool for investigating the interactions among them, since

the changes in the local environment of the atoms involved are mani-

fested by shifts in their resonant frequencies. Several environmental

parameters can be varied in order to obtain information using this

method, such as the solvent, the concentration of the bases and the

temperature. Thus, base stacking, hydrogen bonding and even the forma-

tion of charge-transfer complexes can be detected by this method, since

they manifest themselves differently depending on the parameters set.

Significant upfield shifts in the resonant frequencies of the aro-

matic ring protons of purines on increasing their concentration in

water have been observed by Chan and coworkers.61 These shifts are

not due to a simple dilution effect and are seen to decrease in mag-

nitude upon increasing the temperature. Upfield shifts are known to

occur in aromatic systems where the induced diamagnetic currents in

stacked heterocyclic rings result in an increased magnetic shielding

of the ring protons and a shift of resonances to higher applied

fields.62 Vertical base stacking with average partial ring overlap

was thus postulated to occur in an aqueous environment where the

strong water-water hydrogen bonds would prevent facile interaction of

the solute with the solvent and encourage preferential molecular asso-

ciation of the bases. Furthermore, when the same purines are placed

in a solvent that interacts with them readily, such as DMSO, a hydro-

gen bond acceptor, the shifts become essentially independent of con-

centration, this yielding further evidence for the vertical stacking

interaction scheme in aqueous media through hydrophobic association.61

On the other hand, DMSO has served as a medium in which hydrogen

bonding can be detected for certain pairs of nucleic acid base

derivatives. In this case, downfield shifts of the resonant frequencies

of the interacting protons of purines and pyrimidines are observed upon

increasing the concentration of these species.6364 These shifts,

which become even larger on successively increasing the concentration

of a complementary derivative, are comparatively smaller when a non-

complementary compound is added and are displaced upfield when the

temperature of the system is increased. Hydrogen bonding is known to

decrease the magnetic shielding of the hydrogen atoms directly involved

in the bonding, causing appreciable downfield shifts.65 Therefore, the

downfield shifts observed for these systems are consistent with an in-

teraction of this kind. It is important to point out that in DMSO,

these effects are of small absolute value. Moreover, they are much

more difficult to detect for adenine/thymine than for cytosine/guanine

pairs. In addition, no upfield shifts of the aromatic ring protons can

be detected in this solvent, so vertical stacking for these particular

systems is completely ruled out in DMSO.

A solvent in which hydrogen bonding between complementary nucleic

acid base derivatives can be detected much easier is chloroform, a

relatively inert solvent. The only difficulty is that many nucleic

acid bases and their derivatives are only sparingly soluble in it.

However, for those compounds that do dissolve, large downfield shifts

are observed for the groups that are expected to participate in hydro-

gen bonding. For example, Akashi and coworkers66'67 have reported

that in the case of the methacryloyloxyethyl monomer derivatives of

adenine (MAOA) and thymine (MAOT), the concentration dependence of the

thymine 3-NH proton chemical shift for the 1:1 mixture of MAOA and

MAOT and that of MAOT in DMSO show large differences. An appreciably

3-NH 0 2 NH2

HN 3 N



larger downfield shift is observed for the complementary pair, which
was attributed to an increase in hydrogen bonding with increasing con-
centration. Since MAOT solutions also exhibited downfield shifts on
increasing concentration, although of smaller magnitude, self-associa-
tion was also postulated to occur. Association constants for both
self- and complementary association were calculated from the NMR data
and larger values obtained for the interaction between MAOA and MAOT.
This mode of interaction was further confirmed by noticing the upfield
shifts of the 3-NH peak with an increase in the temperature, which
would-tend to disrupt the hydrogen bonded complex. These results seem
to explain the higher polymerization rates observed for the MAOA/MAOT
monomer pair system.
Interestingly, polymers that contain nucleic acid bases as pendant
groups have also been studied by this technique.66 In connection with
this, the chemical shifts of the monomers, homopolymers and copolymers
having adenine and thymine as the base moieties were determined at dif-
ferent temperatures. Larger chemical shifts were observed for those
groups expected to be involved in hydrogen bonding. Furthermore, even

though in DMSO the hydrogen bonding interaction between pyrimidine and

purine bases is weaker than in chloroform, intramolecular hydrogen

bonding associations could be detected by this method.67

Another complexation phenomenon that could theoretically be stud-

ied by NMR is the formation of 1:1 molecular complexes due to charge-

transfer associations. In this respect, Hanna and Ashbaugh20 have used

this technique to determine equilibrium quotients and shifts of pure

complex for the association of 7,7,8,8-tetracyanoquinodimethane (TCNQ)

and various methyl substituted benzenes. The values obtained agree

well with previous data determined spectroscopically from the Benesi-

Hildebrandl9 method. In their study, upfield shifts of the involved

protons of the acceptor, even if of a small magnitude, are observed

when increasing the concentration of donor at a fixed concentration of


Thus, NMR studies were conducted under different conditions in

order to understand the behavior of the monomers and polymers prepared.

Figure III-6 shows the effect of increasing the total concentration of

a 1:1 mixture of nucleic acid base monomers in chloroform on the chem-

ical shift of the thymine derivative 3-NH proton. A pronounced down-

field shift in the frequency of this proton is noticed, which is of

larger magnitude for the adenine/thymine vinyl ether derivative pair

than for the thymine vinyl ether derivative by itself at the same (and

even larger) concentrations. This would rule out a solvent concentra-

tion effect as the only cause of the observed downfield shift.

Figure III-6 also shows the effect of increasing the total con-

centration of a 1:1 mixture of (2) and (19) on the same ring proton.

CDC13, 250C

12.00 -


10.00 -

9.00 -

2.0 4.0 6.0
molar concentration (x 10-2M)

Figure 111-6.

Concentration dependence of the thymine 3-NH proton
chemical shift of (0) (2), (0) the 1:1 mixture of
(2) and (1) and (S) the 1:1 mixture of (2) and (19).


This pair is structurally similar to the one previously discussed, ex-

cept that the amino group of the adenine derivative has been replaced

by a chloro functionality. This substituent, which is incapable of

hydrogen bonding to the thymine monomer in the same fashion as the

adenine derivative, causes the downfield shift to coincide with that

observed for the thymine derivative by itself.

Table III-6 presents the chemical shifts of the 6-NH2 and 8-H pro-

tons of adenine and the 3-NH and 6-H protons of thymine derivatives

when equimolar solutions in CDC13 at fixed total concentrations of

these species were heated to different temperatures. An upfield shift

is observed on heating, indicating that some complexation is being dis-

rupted. The effect is more pronounced for the 3-NH and the 6-NH2 pro-

tons of thymine and adenine, respectively, than for the 8-H and 6-H

protons of the same pair. The chemical shifts of the latter were in-

cluded for comparison and exhibit a very small temperature effect on

their resonant frequencies.

All this evidence tends to indicate that the mode of interaction

for the adenine/thymine vinyl ether derivatives in chloroform is mainly

of a hydrogen bonding nature. No upfield shift was observed for the

aromatic ring protons, ruling out a base stacking interaction. Besides,

the thymine vinyl ether monomer is probably associating,since a notice-

able effect on increasing the concentration of its chloroform solution

took place.

The NMR concentration dependence data shown in Figure III-6 was

also used for computing the association constants for the binding of

monomers (1) and (2). For this, the general procedure described by

Table III-6

Chemical Shifts of the Nucleic Acid Base Monomer Protons
in Chloroform-d at Different Temperatures
Chemical Shifta
Temperature Adenine Thymine
Monomer (OC) 6-NH2 8-H 3-NH 6-H

(2)b 25 9.00 7.06







(19)c 40
(19)/(2)c 25



3-NH 6-H



a Chemical shifts given in ppm from TMS.
b Concentration of each monomer = 6.0 x 10 2M.
SConcentration of each monomer = 5.6 x 10 M.
d Monomer concentration = 6.0 x 10 3M.
e Concentration of (19) = 6.0 x 10-3M; concentration of (2) = 1.0 M.
The chemical shift shown was not strictly assigned to either 2-
or 8-H.

Iwahashi and Kyogoku60 and applied to monomer systems by Akashi and

collaborators66 was used. The association constant K is defined as
Cd/C2 where Cd and C are the dimer and monomer concentrations, re-
spectively. The association constant KTT for the self-association of

(2) is given by

KTT (o T)(6TT 6T)/2C(6TT 6)2

where 60 is the observed chemical shift, 6T is the limiting chemical

shift of free (2), 6TT is the limiting chemical shift of bonded (2),
and C' is the total concentration of the solute. The association con-
stant KTA was evaluated by the expression:

KTA = T)(TA T)/C(TA 6)2

where 6TA is the limiting chemical shift of the thymine 3-NH proton of
the (1)/(2) dimer. The values of KTT, 6T, 6TT, KTA and 6TA were cal-

culated by the curve fitting method using a computer. These values
are all summarized in Table III-7.

The results suggest that the (1)/(2) complex is quite stable at
25C in chloroform solution when compared to the dimer formed by the

self-association of (2). The present data were similar to those ob-
tained for the system studied by Akashi and coworkers under similar

The monomer pair formed by (2) and (19) was also studied in order
to determine whether a charge-transfer complex between the vinyl ether
and the maleimide moieties was being formed. Thus, a solution in
chloroform of the monomers in which a large excess of the vinyl ether

was present in order to increase the association constant was analyzed


Table III-7

Association Constants and Limiting Chemical Shifts
of the Thymine 3-NH Proton of (2)a

Monomer 6T 6TT 6TA K(1/mole)

(2) 7.92 11.10 7.41

(1)/(2) 8.20 14.79 61.81

a Chemical shifts given in ppm from TMS.
Measurements taken in CDC13 at 250C.

by monitoring any change in the frequency of the maleimide vinyl pro-

tons. No upfield shift was observed, which would have been indicative

of a complex of this kind.

The results mentioned above are not completely unexpected, since

Olson21 also failed to detect the formation of a charge-transfer com-

plex between the pair N-phenylmaleimide and 2-chloroethyl vinyl ether

by this method. This could be explained by a small association constant

for the formation of such a complex or if the protons involved are only

very slightly altered in their environment by an interaction of this


Unfortunately, the low solubility of the maleimide monomers in

chloroform hindered their analysis in this solvent. They were studied,

however, in DMSO at similar concentrations as those used for the vinyl

ether derivatives in chloroform. A comparison of the upfield shifts

observed on increasing the temperature of solutions of adenine and thy-

mine maleimide derivatives individually and for the 1:1 mixture of both

is shown in Table 111-8. In the first place, all the resonant fre-

quencies of the 6-NH2 and 8-H protons of adenine and the 3-NH and 6-H

protons of the thymine monomers appear downfield shifted as compared to

their positions in chloroform solution. This effect is more pronounced

for the 3-NH and 6-NH2 protons. More important is the fact that almost

no difference in the upfield shifts on increasing the temperature can

be observed for the binary solutions versus the individual compounds.

Besides, the net changes in the chemical shifts are smaller in DMSO

than in chloroform. Similar results were obtained for the pairs 9-(2-

maleimidoethyl)adenine/l-(2-vinyloxyethyl)thymine and

Table III-8

Chemical Shifts of the Nucleic Acid Base Monomer Protons
in DMSO-d6 at Different Temperatures

b Temperature
Monomer (oC)

(20) 25

(18) 25

(20)/(18) 25

(2) 25

(2)/(18) 25

(1) 25

(2)/(1) 25
a Chemical shifts given
bConcentration of each

6-NH2 8-H

7.14 8.04
6.92 8.04
6.61 8.06

7.14 8.04
6.96 8.04
6.62 8.06

7.13 8.04
6.93 8.04
6.67 8.06

7.19 8.09
7.00 8.06

7.21 8.10
6.98 8.05

in ppm from TMS.
nomer 4.0 x 102.
monomer = 4.0 x 10 M.

3-NH 6-H

11.16 7.44
10.97 7.38

11.18 7.44
10.99 7.38

11.25 7.47
11.00 7.41
10.74 7.36

11.28 7.47
11.08 7.41
10.68 7.34

11.31 7.48
11.07 7.42

9-(2-vinyloxyethyl)adenine/1-(2-vinyloxyethyl)thymine. All these re-

sults are consistent with the monomers interacting more strongly with

the solvent than with their complementary base derivatives or with

themselves. This is not completely unexpected, considering previous

results reported in the literature.64 Thus, the adenine/thymine mono-

mer pairs do not exhibit detectable hydrogen bonding when DMSO is used

as the solvent.

Homopolymers and copolymers containing the nucleic acid bases were

also studied by this technique. Again, their very low solubility in

chloroform and other organic solvents required the analysis to be per-

formed in DMSO. Table 111-9 shows the chemical shifts observed for the

same protons as in the monomer case on varying the temperature. Com-

paring the data obtained for the maleimide monomers versus those of

their homopolymers as shown in Tables III-8 and III-9, a definite trend

can be observed for the chemical shift temperature dependence of the

6-NH2 protons of the adenine moieties. These values increase in the

order (18)<(18)/(20)<(24)<(24)/(26). The total difference in chemical

shift between monomer (18) and the mixture of homopolymers at 90C

amounts to 0.16 ppm, much higher than the values observed for protons

6-H and 8-H at the same temperature (0.04 ppm in both cases). The sig-

nal of the 3-NH protons, unfortunately, broadened upon heating, so the

lack of data at higher temperatures prevents any comparison as for the

6-NH2 case. A clear pattern thus emerges where the lower chemical

shifts exhibited by the monomers and their binary solutions as compared

with the homopolymers suggest that a larger amount of hydrogen bonding

interaction exists for solutions where pairs of complementary polymers

Table III-9

Chemical Shifts of the Nucleic Acid Base Polymer Protons
in DMSO-d6 at Different Temperatures

Chemical Shifta
b Temperature Adenine Thymine
Polymer (OC) 6-NH2 8-H 3-NH 6-H

(26) 25 11.19 7.41







7.19 8.11
7.10 8.09
6.80 8.12


11.21 7.41
11.01 7.35
S 7.33

11.13 7.27
10.95 7.25
10.67 7.25


3-NH 6-H



Chemical shifts given
Concentration of each
4.0 x 10-2M.

in ppm from TMS.
polymer (based on repeating unit) =





are present. This interaction, however, is rather small and can be

attributed to several factors. First, the chain length of these homo-

polymers is rather short, as evidenced by the low molecular weight

estimated from their intrinsic viscosities. Second, these polymers

have random stereoregularity along their chains, so pairing along them

may not be very efficient and may occur in a disorganized manner.

These interactions are also dependent on the flexibility of the back-

bone and the distance of the bases from the same. Even though they

are farther away than in vinyl polymers containing them, which never-

theless exhibit significant pairing in some cases,1 the flexibility of

the polymaleimide chain is not expected to be very large, due to the

rigidity of the succinimide ring. With all of this in mind, it is

still rather surprising that some type of complexation could be de-

tected under the present conditions, and this reflects the importance

of the polymeric structure in interactions between nucleic acid bases.

Figures III-7 and III-8 present the 6-NH2 and 3-NH chemical shift

dependence on temperature of the monomers, homo- and copolymers con-

taining adenine or 6-chloropurine and thymine. The most drastic dif-

ference in both cases is observed for copolymer (32), having the bases

adenine and thymine along the chain in alternate fashion. Farther

downfield shifts are maintained even at high temperatures, and the

effect is more pronounced at 600C. Very interesting is the fact that

copolymer (29), in which the amino group is replaced by a chloro func-

tionality, exhibits the farthest upfield shifts of all the species in

question. The same trend is also shown by (31), where the repeating

unit contains a succinic anhydride moiety (see Table III-9). The

DMSO-d6, 4.0 x 10-2M

Temperature (OC)

Figure 111-7.

Temperature dependence of the adenine 6-NH2 proton
chemical shifts of (0) (18), ( ) (24), (A)
the 1:1 mixture of (18) and (2) and (A ) (32).

7.10 -

7.00 r


6.80 H

6.70 1-

1 0 0


Figure III-8.

DMSO-d6, 4.0 x 10-2M

S40 60 80 100
Temperature (C)

Temperature dependence of the thymine 3-NH proton
chemical shifts of (0) (2), (A) the 1:1 mixture
of (18) and (2), (0) (29) and (A) (32).





10.80 -

10.70 r

values of the chemical shifts at 60C increase in the order (18) <

(18)/(2) < (24) < (32) for the 6-NH2, and (29) < (2) < (18)/(2) < (32)

for the 3-NH protons, as summarized in Table III-10. These data would

indicate that a hydrogen bonding interaction more difficult to disrupt

by heat is present in the case of copolymer (32), since the chemical

shifts are less affected by an increase in temperature. Concentration

studies performed on monomers (2) and (18) and copolymers (29) and (32)

showed no dependence on this parameter. Such phenomenon could be ex-

plained by invoking an absence of intermolecular interaction or some

kind of intramolecular association. This observation as well as the

results obtained from the temperature dependence studies may be ex-

plained as follows: For the copolymers, the increased flexibility and

the longer chain length would allow preferential intramolecular inter-

action of the bases as shown below. For (32), this would lead to ab-

A- -T

A A- -T T CP

(32) (29)

normally higher shifts which are less affected by temperature and sug-

gest intramolecular complementary pairing. In the case of polymer (29),

the lower values may indicate that the interaction with DMSO is reduced

in favor of intramolecular self-association of the thymine bases in the

chain. On the other hand, the behavior of the monomer pair may be ra-

tionalized in terms of monomer-solvent association, as discussed before.

Table III-10

Chemical Shifts of the Nucleic Acid Base
Monomer and Polymer Protons in DMSO-d6 at 600C

Chemical Shifta
b Adenine Thymine
Compound 6-NH2 8-H 3-NH 6-H

(2) 11.00 7.41

(18) 6.92 8.04 -

(2)/(18) 6.93 8.04 11.08 7.41

(24) 6.96 8.11 -

(29) 10.97 7.19

(32) 7.10 8.09 11.17 7.10

a Chemical shifts given in ppm from TMS.
bConcentration of each monomer and polymer
unit) = 4.0 x 10-2 M.

(based on repeating

--0 0

-- 0'--","""
S-. *

2.0 4.0 6.0 8.
molar concentration (x 10-2M

Figure III-9.

Concentration dependence of the adenine (0)
2-H and (0) 8-H proton chemical shifts of
(1) at (-) 80 and (---) 40C in D20 and
(-.-*) at 250C in CDC13.






The purine monomer 9-(2-vinyloxyethyl)adenine was also studied in

water at different concentrations and two temperatures. Figure III-9

represents the data obtained and shows that upon increasing the concen-

tration of this monomer, an upfield shift of the 8- and 2-H aromatic

protons takes place. This shift is larger in magnitude at lower temp-

eratures and more pronounced for one of these protons. Since these re-

sults are comparable to those found by Chan et al.61 in an analysis of

purine and 6-methylpurine, assignment of the 8- and 2-H protons was

done by proposing the partial overlap stacking interaction of this base

derivative shown below. This mode of stacking would explain the smal-


(a) Partial overlap (b) Hydrogen bonding coplanar
stacking arrangement arrangement

ler effect noticed for the 8-H proton which would be less affected by

the purine ring anisotropy than the 2-H proton, directly over the

electron cloud.

For comparison, Figure III-9 also contains the data for the same

monomer when in the presence of its complementary thymine vinyl ether

derivative in chloroform at 250C. In this case, a slight downfield

shift occurs on increasing the concentration which might be related to

the hydrogen bonding process.

Ultraviolet Absorption Spectroscopy

Interactions among nucleic acid bases and their derivatives have

been studied extensively by UV spectroscopy. The most common tech-

nique employed is the analysis of Job plots68 of solutions containing

pairs of such species. In this procedure, solutions having a fixed

total molar solute concentration but varying in the mole fraction of

each constituent have their absorption curves taken over the range

where their chromophores absorb the most. The absorbance observed at

a particular wavelength is plotted versus the mole fraction of one con-

stituent, and in most cases, a negative deviation from Beer's law is

observed which is more pronounced for the equimolar solution of pairs

of complementary base derivatives such as 9-ethyladenine and 1-cyclo-

hexyluracil. This hypochromism has been explained by invoking the

formation of a 1:1 dimer complex between the species in question.

Hypochromism in polynucleotides, which is the decrease in absorb-

ance per chromophore in the polymer compared to that of the monomer,

has been attributed by Tinoco69 and Rhodes70 to dipole-dipole interac-

tions between transition moments in neighboring oscillators. Large

hypochromic effects are predicted for this model when the base planes

are arranged parallel and one above another. Alternatively, Nesbet71

has attributed hypochromism to a local field effect and calculated

that this effect by itself can lead to a large hypochromism for hydro-

gen bonded base pairs even in the absence of vertical stacking.

Thomas and Kyogoku72 have provided some insight into this controversy

by performing UV studies on base pair derivatives which were previously

investigated by IR spectroscopy and found to pair via hydrogen bonding,

but not to base stack at the concentrations and solvents used. Since

these systems have exhibited significant hypochromicity, they conclude

that vertical stacking of bases as occurs in synthetic polynucleotides

is not the only condition for UV absorption band hypochromism as had

been commonly assumed. Thus, hydrogen bonding in polymers containing

nucleic acid bases may also be detected under the appropriate condi-

tions by UV spectroscopy because of the hypochromicity induced.

The nucleic acid base monomers and polymers synthesized were stud-

ied by this method in order to detect any interaction among complemen-

tary derivatives. Chloroform was the solvent of first choice since it

had been widely used for these purposes, due to its low interaction

with the base derivatives and rather good solvation properties. How-

ever, it was found that the low solubility of the maleimide deriva-

tives in particular prevented their analysis in this solvent, even

though low concentrations were used. Several pairs were nevertheless

analyzed and are reported in Table III-11. It seems evident from the

correlation coefficients calculated from the Job plots that very lit-

tle deviation from a straight line is exhibited by any pair. However,

quite a large effect is observed on changing the solvents in which the

measurements were taken, as seen when comparing Figures III-10 III-

12. Cross-over points vary from none in DMSO to one in chloroform

and two in DMSO/ethylene glycol (EG). (The reason for using the last

solvent system mentioned will be given later on.) Since the presence

of such isosbestic points has been taken as evidence for the existence

of a molecular complex, these results suggest the possibility of one

forming more readily in the DMSO/EG system and in chloroform. The

University of Florida Home Page
© 2004 - 2010 University of Florida George A. Smathers Libraries.
All rights reserved.

Acceptable Use, Copyright, and Disclaimer Statement
Last updated October 10, 2010 - - mvs