_'I_x_s
IQ);
CDU'IG)

 

SELF - ASSEMBLED POTASSIUM 5’ - GUANOSINE
MONOPHOSPHATE. NUCLEAR MAGNETIC RESONANCE
EVIDENCE FOR STRUCTURE FORMATION IN
AQUEOUS SOLUTION

Thesis for the Degree of M. S.
MICHIGAN STATE UNIVERSITY
CLAUDIA MARIE METTLER
1977

 

 

 

LIBRARY

Michigan State
University

 

 

 

ABSTRACT
SELF-ASSEMBLED POTASSIUM 5'-GUANOSINE
MONOPHOSPHATE. NUCLEAR MAGNETIC RESONANCE
EVIDENCE FOR STRUCTURE FORMATION IN AQUEOUS SOLUTION
BY

Claudia Marie Mettler

It has been shown by variable.temperature proton
NMR studies that the spontaneous intermolecular self-
assembly of 5'-guanosine monophosphates (5'-GMP) in
aqueous solution (pH = 7.0 - 8.0) is dramatically
dependent on the nature of the alkali metal counterion.
The potassium and sodium salts of the mononucleotide
form self-structures which differ both in the number of
H(8) resonance lines and in thermal stability. More-
over, the lithium salt fails to form any NMR detectable
structure. These results show for the first time that
selective alkali metal coordination by a nucleic acid
does occur, and that the coordination interaction can
be important in the formation of regular, order structures

involving intermolecular association of nucleotides.

Claudia Marie Mettler

 

 

The proton spectra obtained at 60 MHz have limited
utility in distinguishing the number of non-equivalent
H(8) environments in self-assembled K2 (5'-GMP). How-
ever, spectra obtained at 220 MHz indicate that at least
two, and perhaps three, distinguishable structures are
present. Previously reported1 self-assembled Nag (5'-GMP)
forms only one or two self-structures.1 The chemical shift
range of the H(8) resonances is smaller for the self-
assembled K+ salt (0.50 ppm) than for the self-assembled
Na+ salt (1.3 ppm). Also, the K2 (5'-GMP) self-structures
have higher solution melting temperatures than the
Nag (5'-GMP) self-structures.

As in the case of self-structured Nag (5'-GMP), self-

structured K2 (5'-GMP) exhibits amino proton resonances.

Claudia Marie Mettler

The slow chemical exchange between amino protons and

H20 (solvent) protons indicates that hydrogen bonding as well
as specific alkali metal ion coordination is also in-

volved in the self-assembly proceSses. It is likely that
the self-structures in both salts involves formation of
associated nucleotide units through hydrogen bonding between
position 0(6) and N(7) as acceptors and N(l) and N(2)

as donors. The hydrogen-bonded nucleotide units, per-

haps tetramer units, then coordinate with selected

alkali metal ions to form a regular, ordered structure

that is slow to undergo chemical exchange. The results
obtained in this work, however, do not allow identifi-

cation of the coordination mechanism.

References

 

l. T. J. Pinnavaia, H. T. Miles, and E. D. Becker, 1.
ET: Chem. Soc., 21, 7198 (1975)

 

SELF-ASSEMBLED POTASSIUM 5'-GUANOSINE
MONOPHOSPHATE. NUCLEAR MAGNETIC RESONANCE

EVIDENCE FOR STRUCTURE FORMATION IN AQUEOUS SOLUTION

BY

Claudia Marie Mettler

A THESIS

Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of
MASTER OF SCIENCE
Department of Chemistry

1977

DEDICATION

To John Michael Duranceau

ii

ACKNOWLEDGEMENTS

I wish to thank Dr. Thomas J. Pinnavaia for suggest-
ing this thesis problem and Dow Corning Corporation for
permitting me to take a ten week leave of absence to
complete the experimental work at Michigan State

University.

iii

TABLE OF CONTENTS

Page

Dedication ii
Acknowledgements iii
Table of Contents iv
List of Tables v
List of Figures vi

I. Introduction 1
II. Experimental 8
A. Materials, Techniques, and Preparation 8

B. Instrumentation 10

1. Nuclear Magnetic Resonance Spectroscopy 10
2. Ultra-violet Absorption Spectroscopy 10

III. Results and Discussion 11

IV. Conclusion 32

iv

LIST OF TABLES

TABLE PAGE

I. Concentration Dependence of H(8),
H(l') Chemical Shifts for
K2 (5'-GMP) in H20 at 32°C 14

II. Temperature Dependence of H(8),
H(l') Chemical Shifts for
K2 (5'-GMP) in H20 at 0.63M 17

III. H(8) Chemical Shifts of Self-
Assembled K2 (5'-GMP) Obtained
at 220 MHz‘ 28

LIST OF FIGURES
FIGURE 4 PAGE
I. Structure of Guanylic Acid 3

II. Structure of Hydrogen-Bonded Tetramer
of Guanylic Acid Monomers 5

III. Structure of Sodium Guanosine 5'-
Monophosphate 7

IV. Proton NMR Spectra (60 MHz) of
K2 (5'-GMP) in H20 in the Concentration
Range 0.26 to 1.13M. Temperature is 32°C 15

V. Proton NMR Spectra (60 MHz) of
K2 (5'-GMP) in H20 in the Temperature
Range 5 to 70°C; Concentration is 0.63M 19

VI. Proton NMR Spectra (60 MHz) of Verifi-
cation of H(8) Signals by D20 Exchange 22

VII. Proton NMR Spectra (60 MHz) of the
Nitrogen-Bonded Protons of K2 (5'-GMP)
in H20 at 0.63M 24

VIII. H(8) Resonances at 220 MHz for K2 (5'-GMP) 30

vi

I . INTRODUCTI ON

Mononucleotides are known as the building blocks of
the nucleic acids. Nucleic acids participate in molec-
ular mechanisms which store, replicate, and transcribe
genetic information through the polynucleotide chains
of DNA and RNA. Mononucleotides contain three com-
ponents: l. a nitrogenous base, 2. a five-carbon sugar,
and 3. phosphoric acid.

The aromatic heterocyclic compounds of pyrimidine
and purine are the two types of nitrogenous bases in
nucleotides. The six-membered ring pyrimidines are
cytosine, thymine, and uracil. The purines, pyrimidines
with fused imidazole rings, are adenine and guanine.
Nitrogenous bases have limited water solubility and
exist in tautomeric forms.

Nucleosides are N-glycosides of pyrimidine and
purine bases in which the N(l) of pyrimidines and the
N(9) of purines are glycosidically linked to the C(l) of
a five carbon sugar-usually ribose or deoxyribose. The
nucleosides, cytidine, thiamine, uridine, adenosine,
and guanosine, are more Water soluble than their parent

bases.

Nucleotides are the phosphoric acid esters of the
nucleosides. Phosphoric acid is esterified to one of the
free hydroxyl groups of the pentose sugar. Although 2'
and 3' isomers exist, the nucleotides, in the free form
in cells, generally have the phosphate group in the 5'
position of the sugar. The mononucleotides are strong
acids. The two dissociable protons of the phosphoric
acid group have pK' values of about 1.0 and 6.2 At pH
7.0, the free nucleotides exist totally dissociated.1

Base-paired mononucleotides form hydrogen-bonded

11213:“ NMR evidence for

complexes in aqueous solution.
hydrogen-bonding between complementary bases was based
on the detection and measurement of chemical shifts for
the amino protons involved in hydrogen-bonding. The
amino protons involved in the hydrOgen-bonding exhibited
downfield chemical shifts from the unassociated monomers.
These downfield chemical shifts are concentration and
temperature dependent. A

Guanylic acid (and its derivatives) Figure I is the

only nucleic acid component which spontaneously forms a

regular, ordered structure with itself in aqueous solution.

 

Figure I. Structure of Guanylic Acid

In 1910, Bang reported that guanylic acid formed a
gel in concentrated solutions.5 In 1962, Gellert,
Lipsett, and Davies found that solutions of guanylic
acid (GMP) at pH 5, where one proton is dissociated,
were extremely viscous at room temperature and that
both dilute and concentrated solutions formed a clear
gel on cooling.6 The authors investigated the Optical
prOperties of the gel and the fibral structure formed
upon drying the gel. It was concluded that helix
formation by guanylic acid was occurring. A hydrogen-
bonded structure of the bases in the gel was prOposed.
The proposed structure consisted of continuous helixes
(5'-isomer) or planar tetramer units (Figure II)

stacked in a helical array (3'-isomer).

Figure II. Structure of Hydrogen-Bonded Tetramer
of Guanylic Acid Monomers

 

Figure II

In 1972, Miles and Frazier reported that guanosine
5'-mon0phosphate (5'-GMP) formed a regular, ordered
structure in neutral solution. At this pH, both protons
of guanylic acid are dissociated. The authors found
that the ordered structure was not associated with the
previously reported gel formation at pH 5. From infra-
red and chemical evidence, the proposed neutral structure
consisted of planar tetramer units (Figure II).- There
was similarity in the pattern of these planar tetramer
units to that of the acid gels. Nitrogen (l) and
nitrogen (2) were proton donors, while oxygen (6) and
nitrogen (7) were proton acceptors in the pr0posed
planar tetramer units.

More recently, Pinnavaia, Miles,and Becker pub-
lished nuclear magnetic resonance evidence for the
formation of a regular, ordered structure and slow
chemical exchange for self-assembled Nag (5'-GMP)
(Figure III) in neutral solution.8 Pinnavaia, et. al.
showed that the structure formed involved hydrogen?
bonding. In addition, the structure was stable enough
to observe non-equivalent proton environments. In
previous work with aqueous solution chemistry of base-
paired mononucleotides, only rapid chemical exchange
and time-average NMR spectra had been reported. The

slow chemical exchange was unprecedented.

 

N %” NH2
0

+ _ n ,.

Na O-P-O-H2C5
l O W‘

O- H

+ :
Na H H
OH OH

Figure III. Structure of Sodium Guanosine 5'-Mono-
phosphate

The authors found that the NMR lines at 0.23M

Nag (5'-GMP) were attributed to those previously re-
ported for the disordered nucleotide in dilute aqueous
solution. Specifically, the H(8) resonance was a sharp
singlet at 8.1 ppm. The self-assembly of Nag (5'-GMP) in
neutral aqueous solution was confirmed by proton NMR line
broadening with increasing concentration or decreasing
temperature. As the concentration of mononucleotide
increased (0.42M), two H(8) lines were at 7.25 and 8.55
ppm. The chemical shifts of these new lines were in-
dependent of concentration, thus being indicative of

slow chemical exchange. At 0.59M, a third H(8) line
appeared on the low-field side of the central resonance
(8.17 ppm) H(8) line (concentration dependence). The

resolution of four H(8) lines was observed at 0.83M.

With decreasing temperature, similar changes were ob-
served in the H(8) region of the Nag (5'—GMP). The
upfield shift of one H(8) self-structure line was
attributed to diamagnetic ring currents of overlapping
bases. Isoshielding contours calculated for the ring
currents in guanine by Giessner-Prettre and Pullman
quantitatively agree with this upfield shift."10 The
bases stacked above and below guanine account for a
large portion of the upfield chemical shift. The
downfield shifts of the third and fourth highest field
H(8) lines were presumed to be due to electric field
effects caused by the nearness of the phosphate group
(doubly ionized in Nag (5'-GMP))to the H(8) protons.
The H(8) lines were verified by their decrease in in-
tensity upon exchange with D20. Low field lines ob-
served in H20 but not in D20 were attributed to protons
bonded to nitrogens. If these protons, specifically
Nl-H, were unassociated, i.e., not hydrogen-bonded,
their exchange with H20 would be too rapid to observe.
The observation of these lines under complex formation
conditions, but not at high temperatures or low 5'-GMP
concentrations at which the ordered structure does not
exist, provided evidence for hydrogen-bonding of these

protons in the ordered structure of 5'-GMP.

Pinnavaia, Miles, and Becker worked exclusively

with the disodium salt of guanosine 5'-monophosphate.

The present study investigates the effect of changing
the metal ion on the self-assembly of guanosine 5'-
monophosphate in aqueous solutions. The experiments
were designed to determine whether changing the metal
ion would: 1) simply affect the chemical shift in the
H(8) region, or 2) directly alter the structure forming
process. If the metal ion were present only as a counter-
ion in solution, the former observation would be made.
If a different structure was formed with a different
metal ion, then direct metal ion coordination with the
solution structure would be indicated.
II. EXPERIMENTAL
A. Materials, Techniques, and Preparation
Sodium guanosine 5’-monophosphate monohydrate
(Na¢(§GMP))was purchased from P-L Biochemicals, Inc.
Paramagnetic ion impurities were removed from distilled
water by an electrolysis cell. The ion exchange column
was prepared with Dowex 8, a 100-200 mesh cation exchange
resin with a capacity of 1.7 meg/ml. The wet resin was
washed with: l) 3N hydrochloric acid - to remove gross
amounts of dirt, 2) a dilute solution of disodium
ethylenediaminetetraacetate— to remove paramagnetic
ion impurities, and 3) an aqueous solution of the de-
sired metal ion hydroxide or chloride (depending on
the water solubility) - to exchange the cationic resin
with the desired metal ion. An aqueous solution of

N32 (5'-GMP) was passed through the ion exchange

10

column packed with the metal ion exchanged resin.
Essentially total exchange occurred. The exchanged
5'-GMP salt solution was 1y0phi1ized (freeze-dried).
Then, the exchanged 5'-GMP salt was recovered by sub-
liming the water. Sodium 2,2,3,3,-dg-3-trimethylsilyl-
propionate (TSP) was purchased from Merck and Company,
Inc. for the nuclear magnetic resonance reference
standard. A 0.003M TSP solution had a pH of 7.90. A
0.63M solution of K2 (S'GMP) made with the TSP solution
had a pH of 8.85. I
B. Instrumentation
1. Nuclear Magnetic Resonance Spectroscopy

Proton magnetic resonance spectra were obtained with
a 60 MHz Varian A56/60D spectrometer. Probe temperature
was regulated with a Varian V-6040 temperature controller.
The probe temperatures were measured by the chemical
shift differences between the proton resonances of
methanol (0-40°C) and ethylene glycol (40-90°C). The
instrument was calibrated by the chemical shift differ-
ence between tetramethylsilane and chloroform of 436 Hz.
Maximum radio—frequency field strengths and spectrum
amplitudes were used to maximize spectra resolution.

2. Ultra-violet Absorption Spectroscopy

Guanosine 5'-monophosphate solution U-V absorption
spectra were determined with a Cary 17 Ultraviolet-Visible
Absorption Spectrometer. Solution concentrations were

obtained from Beer's Law:

11

log (Io/I)=elc=A
I - intensity of the incident light

g - molar absorptivity (or decadic molar ex-
tinction coefficient)

1 - cell length in centimeters

c - concentration in moles per liter

A - absorbance or Optical density11
The molar absorptivity (e) for Nag (5'-GMP) was taken
to be 1.41 X 10“ at 252 nm, as determined by P-L
Biochemicals, Inc. This molar absorptivity value was
assumed to remain constant while the metal ion was
changed. The cell length was one centimeter.

III. RESULTS AND DISCUSSION

Preliminary experiments were carried out with the
lithium, potassium, barium, and magnesium salts of
guanosine 5'-monophosphate. The magnesium and barium
salts have limited water solubility. The low solubility
is attributed to the divalent charge and low ionic
radius of the metal. No H(8) line is observed at
60 MHz for a saturated solution of magnesium or barium
guanosine 5'-monophosphate at 5°C and 42°C.

The lithium salt of guanosine 5'-mon0phosphate is
water soluble. At approximately 0.5M, a single H(8)
line of the mononucleotide is observed at 7.1 ppm.
Varying the temperature did not change the H(8) chemical
shift, appreciably broaden the line, or cause extra

H(8) lines to form. The lithium salt behaves differently

12

from the sodium salt; no evidence of self-assembly is
apparent.

This observation was the first evidence for metal
ion coordination in the formation of a self-structure
by guanosine 5'-mon0phosphate. The lithium cations
act only as counterions in solution for the mono-
nucleotide. The sodium cations appear to be specifically
coordinated with the guanosine 5'-mon0phosphate. I

The potassium salt of guanosine 5'-monophosphate
is water soluble. At approximately, 0.5M and 32°C
two H(8) lines are observed at 7.8 and 7.5 ppm, respect-
ively. Lowering the temperature alters these chemical
shifts, appreciable line broadening is observed at 5°C.
In H20 solution, but not in D20, low-field lines are
observed at 11.2, 10.8, 10.5 and 9.9 ppm. These latter
are attributed to nitrogen-bonded protons. Two H(8)
lines are observed with 60°C with 0.63M of the potassium
salt. In contrast, the four H(8) lines of the sodium
salt are melted out below 40°C.

These initial observations confirm: l) the absence
of a solution self-structure with the lithium salt of
guanosine 5'-monophosphate, 2) the formation of a
solution self-structure with the potassium salt, 3)

a difference from the structure previously reported
for the sodium salt, and 4) metal ion coordination in the
formation of a solution self-structure of guanosine

S'Hmonophosphate. The solution self-assembly of

l3

potassium guanosine 5'-monophosphate is investigated
further to determine the behavior and the effect on
structure of the potassium metal ion.

Proton nuclear magnetic resonance spectra at
60 MHz of potassium guanosine 5'-monophosphate (K2(5'-GMP))
at various concentrations in water at 32°C are shown
in Table I and Figure IV.‘ The lines observed in dilute
solution at 0.26M are assigned to the disordered mono-
nucleotide. The single H(8) line is at 8.0 ppm. As the
concentration increases to 0.30M, the lines broaden,
spin-spin coupling of the C(l') ribose proton is ob-
scured (5.6 ppm) and a second H(8) line grows out of
the base line at 7.5 ppm. At 0.4M, both H(8) lines
increased in intensity. At 0.50M and 0.63M, the H(8)
lines increased little, if any, in intensity from
0.40M. At 1.13M, a saturated solution, there is ex-
tensive line broadening and upfield shifting of the
H(8) lines. Four unresolved H(8) lines are observed
at 7.6, 7.3, 7.1, and 6.5 ppm, respectively. This
indicates rapid chemical exchange and time - averaging
of two or more classes of protons.

Different changes in the H(8) region are observed
vvith decreasing temperature in Table II and Figure V.
The lines observed at 70°C at 0.63M in aqueous solution
are attributed to the disordered mononucleotide. The

Single H(8) line is at 7.9 ppm. As the temperature

14

mmcfla Amvm ©m>Home

In: HDOM mm: Amzol.mv «M .sofiuoaom pmumu5DMm GHQ

own pam>aom mo amaze Hanasmaum

 

 

mm.q I I I mo.m I saa.o
mo.¢ I I I mo.m I zma.o
ss.e I I I mo.m I spa.o
mm.¢ I I I qo.m I zam.o
H>.¢ I I I Ho.m I zm~.o
~>.q o».m I om.n mm.n I amm.o
¢>.¢ om.m I w¢.n mm.b I zoe.o
ob.s H>.m I >¢.k 85.5 I zam.o
mn.e mk.m I A¢.k as.“ I smo.o
I I om.o oH.h am.b oo.s nxcmumuspmmc zmH.H
HOGHDOSm . . GOHDMHHSOUGOU

moms I.va namfimmo Amvm imcm

mma Eonm Ema .m

oomm um oam ca AmzwI.mV am now mumaam Hmanmno

H mflmfla

A.va .Amvm mo oozonsmmmo coaumnusoocou

15

Figure IV. Proton NMR Spectra (60 MHz) of K2 (5'-GMP)
in H20 in the Concentration Range - 0.26M
to 1.13M. Temperature is 32°C

16

   

 
 

U3M.
(so? d)

  

 

n 063M

/ 0.50 M

0.40 M

 

 

 

‘ 0.32 M

0.26 M

8.0l 5.56
Sppm

Figure IV

17

ooom um mafiumma umumau

com on mafiaooo mnomomo

 

hm.b

 

mmB EOHM Ema .w

2mm.o um our a“ xmzoI.mv

as you mumwnm Hmoflsmgo

A.va .Amvm mo wosmoammoo masumuwmfiwa

HH fldmda

on.v an.m he.“ amm
He.¢ sm.m I um.> om
ne.v em.m I ma.» om
mm.¢ mm.m I «a.» on
Hm.¢ pn.m I mm.» om
em.w m>.m we.“ mm.“ om
su.e on.m he.“ us.b ow
mn.q mo.m 54.5 mu.» mm
mm.v I mv.> mm.~ om
nm.v I me.“ mp.“ ma
Hm.¢ I ev.h HA.“ oH
om.e I I mm.k m
mn.q on.m me.n Hp.h 0mm
. oum A.va umwasoam camflmms xmvm Amcm no.1 musumummsms

18

Figure V. Proton NMR Spectra (60 MHz) of K2 (5'-GMP)
in H20 in the Temperature Range - 5° to 70°C.
Concentration is 0.63M

 

 

 

/.

20

decreases to 60°C, the lines broaden, and a second
H(8) line grows out of the base line at 7.5 ppm. At
50°C, the H(8) line at 7.5 ppm grows in intensity and
the spin-spin coupling of the C(l') ribose proton is
obscured. Between 23°C and 10°C, the intensity of the
H(8) peak attributed to the disordered nucleotide
decreases (8.0 ppm). The H(8) line at 7.5 ppm does
not change appreciably in intensity but moves slightly
downfield as the temperature decreases. At 5°C, the
two H(8) peaks merge into one unresolved peak, centered
around 8.0 ppm.

Confirmation of the assignment of the H(8) lines is
obtained by a decrease in their absolute intensities
after exchange with deuterium oxide (D20). (Figure VI).
The proton exchange conditions (heating at 70°C for
two hours) are such that the H(8) protons (7.9-8.0 ppm)
are replaced by deuterium but the ribose protons
(specifically the H(l') proton at 5.7 ppm) are not.

The low field lines which are observed in water but not
in deuterium oxide are attributed to N-bonded protons
(11.2, 10.8, 10.5, and 9.9 ppm; Figure VII). The lowest
field lines are probably Nl-H, as was reported by
Pinnavaia, et. al.8 The N1 proton exchange of un-
associated guanylic acid is too rapid in water to permit
observation of an NMR line..2r13 However, Nl-H is
observed in hydrogen-bonded complexes.1311“v’5 The

N-bonded proton lines are observed under conditions

21

Figure VI. Proton NMR Spectra (60 MHz) of Verification
of H(8) Signals by D20 Exchange

23

Figure VII. Proton NMR Spectra (60 MHz) of the
Nitrogen-Bonded Protons of K2 (5'-GMP)
in H20 at 0.63M

24

 

 

HH> whomfim

Fan. .m
mm 90. m6.

 

<

25

favorable for complex formation of K2 (5'-GMP). These
lines are not observed at temperatures and mononucleotide
concentrations at which a solution structure is not

found to exist. The appearance of the Nl-H protons

lines under conditions of complex formation provides
evidence for the hydrogen-bonding of these protons in

the ordered solution structure of K2 (5'-GMP).

The H(8) resonance at 8.0 ppm is attributed to the
unassociated mononucleotide. This line shifts somewhat
to lower field with decreasing concentration and de-
creasing temperature (Figures IV and V). This is
indicative of rapid exchange and time—averaging of two
or more non-equivalent protons. This observation
has previously denoted formation of non-regular stacked
aggregates, perhaps in equilibrium with a regular
structure containing rapidly exchanging classes of
protons.°r12r16

The H(8) line upfield (7.5 ppm) from the unassociated
mononucleotide signal did not shift with changes in
concentration or temperature. This may have resulted
from the formation of a regular structure in slow ex-
change with unassociated monomers or stacked aggregates.
A limiting structure appears in both the concentration
and temperature series (Figures VI and VII). There
are small changes in the absolute intensities and
chemical shifts of the H(8) protons between 0.40M

and 0.63M and between 23°C and 50°C respectively. When

26

extreme conditions of either concentration (saturated
solution) or temperature (less than 10°C) are applied,
a new solution structure or structures may be formed.
The structures formed at low temperatures and formed in
saturated solutions of the mononucleotide are in rapid
equilibrium - indicated by the broad, unresolved lines
observed. The differences in these 60 MHz spectra at
the extremes of temperature and concentration show
that under different conditions, more than one solution
structure is forming. This is in contrast with the
similarity found between the conCentration and temper-
ature spectra of Na2 (5'-GMP). (Pinnavaia, et. a1.°)
Since only one H(8) line is observed besides the
monomer line at 60 MHz under structure forming conditions
for K2 (5'-GMP) (excluding extremes of temperature and
concentration), a single solution structure with
equivalent H(8) protons is hypothesized. The bonding
scheme shown in Figure II - a tetramer of hydrogen-
bonded guanine bases - is favored for the K2 (5'-GMP)
self-structure. The tetramer is formed by two hydrogen
bonds per guanine base unit. All of the guanines and
their H(8) regions are equivalent. Previous work shows
that alternative bonding schemes are not reasonable,
because none produce more than two hydrogen bonds per
base.“17 This considerably stable tetramer could be
the limiting K2 (5'-GMP) structure indicated in the

concentration range of 0.40M-0.63M and in the temperature

27

range of 32-50°C.

In the communication published by Pinnavaia, Miles
and Becker, the H(8) non-equivalence (four H(8) lines
were observed) in the solution structure formed by
Na2 (5'-GMP) was attributed to limited head-to-tail
stacking of tetramer units. The new K2 (5'—GMP) self-
structures which form under extremes of concentration
and temperature, could be stacking of the tetramer
units. Further self-assembling would account for the
upfield shift of H(8) lines in the concentration spectra.
Such a variety of stacked tetramer units would account
for the different H(8) environments observed with
saturated solutions of K2 (5'-GMP). It would not
necessarily account for the single broad H(8) signal
observed at 5°C with 0.63M. This could just have been
a rapid equilibrium between the tetramer and the monomer.

The resolution obtained with the 60 MHz nuclear
magnetic resonance spectrometer compared with the
resolution obtained with the 220 MHz instrument used by
Pinnavaia, Miles and Becker is mediocre. The 220 MHz
spectra of K2 (5'-GMP) were obtained from the National
Institute of Health, compliments of Dr. C. Fisk. The
220 MHz spectra of the H(8) region of a 0.8M solution
of K2 (5'-GMP) are shown in Table III and Figure VIII.
At 18°C, two lines are prominent at 7.85 and 7.53 ppm.
The peak at 7.85 ppm has a low field shoulder. The peak

at 7.53 ppm has an upfield shoulder. Another peak appears

28

TABLE III

H(8) Chemical Shifts of Self-Assembled
K2 (5'-GMP)e-Obtained at 220 MHz

6, ppm from TSP

 

Temperature (°C) H(8)
18 7.85 7.53
6.5 7.86 7.51
3 7.88 7.52

 

e0.8M K2 (5'-GMP)

29

Figure VIII. H(8) Resonances at 220 MHz for
K2 (5'-GMP)

30

 

 

 

i 'kosm, I8°

 

 

7.85 7.53
. ' 0.8 M, 65°
7.86 7.5!
0.8M, 3°
I
7.88

Figure VIII

31

at about 7.3 ppm. At 6.5°C, the H(8) region of 0.80M
K2 (5'-GMP) is broadened. The main H(8) peaks are at
7.86 and 7.51 ppm, respectively. All the shoulders, and
the new peak at about 7.3 ppm are more pronounced. At
3°C, the H(8) region is broader than at 6.5°C. The
main peaks are at 7.88 and 7.52 ppm. The shoulders are
more prominent. Two new shoulders also appear - one on
the high field side of the 7.88 ppm line and one on the
low field side of the 7.52 ppm line. Both of the two
main peaks move downfield with decreasing temperature
(Table III).

Even though the resolution with the 220 MHz
spectrometer is not complete for the K2 (5'-GMP) solution,
it is improved over the 60 MHz spectra. With the 220
MHz evidence, it is apparent that the K2 (5'-GMP)
solution structure is not as simple as the 60 MHz spectra
portrays. In addition, the K2 (5'-GMP) solution
structures appear much more complicated than the previously
reported Naz (5'-GMP) structures; with two prominent
peaks, there are at least seven unresolved peaks.

The seven or more H(8) lines observed in the 220
MHz spectra indicate that more than one solution structure
or more than one H(8) environment is present. The lack
of resolution in part of the spectra indicates rapid
equilibrium between H(8) environments. The pre-
dominance of a few peaks indicate that some structures

(H(8) environments) are more stable than others. The

32

220 MHz spectra do not prove or disprove the presence

of stacking of the tetramer units. The presence of
tetramer units has been assumed. The existence of a
stable upfield H(8) line upfield from the K2 (5'-GMP)
monomer line indicated a structure with equivalent H(8)
environments was present. As mentioned previously, only
the tetramer structure has been found to account for such
H(8) equivalence. The tetramer is definitely not the
only solution structure formed with K2 (5'-GMP) but it
certainly accounts for a major portion of the solution
structure.
IV. CONCLUSION

Metal ion dependence has been proven in the solution

self-assembly of guanosine 5'-mon0phosphate. In com-
parison, Li2 (5'-GMP), Na2 (5'-GMP), and K2 (5'-GMP)

each exhibit unique self-assembly behavior. The extent
of self-association appears to increase with: 1) in-
creasing ionic radius: Li+- 0.682, Na+- 0.972, and

K+- 1.333, or 2) decreasing charge-to-mass ratio.1°

Potassium guanosine 5'-mon0phosphate forms different

self-structures from those reported for Na2 (5'-GMP);

a limiting structure in slow exchange with the unassociated
monomer forms. Under extremes of concentration or
temperature, a new structure may form which is in rapid
equilibrium with the monomer. If a new structure is

not formed, the equilibrium state is shifted from that

exhibited for the limiting case. The K2 (5'-GMP)

33

self-structure exhibits more stability than Na2 (5'-GMP).
Self-structure H(8) lines appear in more dilute solutions
of K2 (5'-GMP) compared with Na2 (5'-GMP). The self-
structure of K2 (5'-GMP) also forms at higher temperatures
than Na2 (5'-GMP).

The metal ion plays a structure forming role in the
solution self-assembly of 5'—guanosine monophosphate.
The location of the metal ion such that it can influence
the solution self-assembly of 5'-GMP has yet to be dis-
covered. The metal ion could be accommodated in one or
more positions.

1) The tetramer structure of hydrogen-bonded guanine
bases has a hole in the center (Figure II). The size
of the hole would determine whether a metal ion could
be complexed. 2) If stacking of the tetramers occurs,
there may be complexing between the tetramer sheets. It

is known that there is 3.43 between stacked sheets.1I1’

3) Each guanosine mononucleotide has a phosphate group
containing two negative charges. In a tetramer structure,
there are a total of eight negative charges. The
positioning of these phosphates on one side of the tetramer
would generate a region with high electron density which
could attract positive metal ions in solution.

Since crystallization of 5'-guanosine monophosphate
has not been achieved, X-ray crystallography can not
be done to find out where the metal ions are. Metal

ion specific nuclear magnetic resonance (NaT for example)

34

could be done to find out how many metal ion environ-
ments are present. .Other metal ions with similar radii
and/or charges, such as Tl+, Ag+, Rb+, NHI+, and
(CH3)IN+, could also be exchanged for Na+ in order to
see if self—structures form. Perhaps a self-structure
could be forced to form with the lithium salt (or other
salt which forms no apparent self-structure) of 5'-
.guanosine monophosphate by the addition of an excess of
a lithium halide to thelithiUm 5'-GMP solution. C-13
nuclear magnetic resonance could be done to observe
which carbons are involved in the hydrogen-bonding or
metal ion bonding.

In conclusion, the metal ion plays a structure
forming role in the solution self-assembly of 5'-guanosine
monophosphate. The metal ion dependence of a nucleic
acid component in solution has significant biological
relevance. However, the self-assembly occurs under
conditions of such high mononucleotide concentration or
at low temperatures, that the biological relevance has,

at most, limited practical significance.

BIBLIOGRAPHY

10.

11.

12.

13.

14.

'U. S. A., 69, 2025 (1972).

-Ԥ_1_, 1089 (1974).

BIBLIOGRAPHY

L. Lehninger, Bicehemistry: The Molecular Basis

‘of Cell Structure an Function; CE. I2, 25, 28;

 

 

WOrIdP Puinshers, Inc., N. Y., N. Y., 1970.

Raszka and N. 0. Kaplan, Proc. Nat. Acad. Sci.

 

 

R. Shoup, H. T. Miles, and E. D. Becker, Biochem.

'Bithys.'Res.'Commun.,‘2, 194 (1966).

Katz and S. Penman,‘J,’Mol.'Biol.,'l§, 220 (1966).

Bang, Bioch. Ztschr., 26, 293 (1910).

 

Gellert, M. N. Lipsett, and D. R. Davies, Proc.

‘Nat. Acad. Sci., 48, 2013 (1962).

 

T. Miles and J. Frazier, Biochem. Biophys. Res.
Commun., 49, 199 (1972).

 

J. Pinnavaia, H. T. Miles, and E. D. Becker,

‘ I1.'1_&1_9_.'Chem.‘Soc., 9_7_, 7198 (1975).

Giessner-Prettre and B. Pullman, J. Theor.

'Biol, 27, 87 (1970)-

Giessner-Prettre and B. Pullman, Biochem. Biophys.

'Res.‘Commun.,'ZQ, 578 (1976).

 

L. Jolly, The Synthesis and Characterization of

 

 

‘Inorganic Compounds; Prentice-Hall, Inc.,

Englewood C i s, N J., p. 317.

T. Miles, F. B. Howard, and J. Frazier, Science,

“142, 1458 (1963).

M. Crothers, C. W. Hilbers, and R. G. Shulman,
Proc. Nat. Acad.'Sci.'U,'§,'A,,'ZQJ 2899 (1973).

 

 

B. Arter, G. C. Walker, 0. C. Uhlenbeck, and
P. G. Schmidt, Biochem. Biophys. Res. Commun.,

 

 

35

15.

16.

17.

18.‘

19.

BIBLIOGRAPHY
(Cont'd.)

D. J. Patel and C. W. Hi1bers,‘Biochemistry,‘14,
2651 (1975).

 

M. P. Schweizer, A. D. Broom, P. O. P. T‘so, and
D. P. Hollis,‘g,‘§m,'Chem.'Soc.,20, 1042 (1968).

 

M. Hattori, J. Frazier and H. T. Miles, Biopolymers,
‘14, 2095 (1975).

 

 

Handbook of Chemistr and'Physics, ed. Robert C.
Weast,_—. I974, 1975.

J. D. Watson and F. H. C. Crick, Nature, 171,
738 (1958).

 

36

AV'

7.-

C
\1

177 42

HI”!

‘ l
3

.h
B
L
v:
T
.b
D.
E
V
N
U
:.
Tl

.',1.(‘H.ajf~f.' S 2'