Citation
Comparative chemical and biological studies on the fractionation of deoxyribonucleic acids

Material Information

Title:
Comparative chemical and biological studies on the fractionation of deoxyribonucleic acids
Creator:
Frankel, Fred Robert, 1934-
Publication Date:
Language:
English
Physical Description:
vii, 92, 2 leaves : ill. ; 29 cm.

Subjects

Subjects / Keywords:
Base composition ( jstor )
Chromatography ( jstor )
Ehrlich tumor carcinoma ( jstor )
Elution ( jstor )
Liver ( jstor )
Molecules ( jstor )
pH ( jstor )
Resins ( jstor )
Specimens ( jstor )
Tumors ( jstor )
Biochemistry and Molecular Biology Thesis Ph.D ( mesh )
Chemical Fractionation ( mesh )
DNA ( mesh )
Dissertations, Academic -- biochemistry and molecular biology -- UF ( mesh )
Genre:
bibliography ( marcgt )
non-fiction ( marcgt )

Notes

Thesis:
Thesis (Ph.D.)--University of Florida, 1960.
Bibliography:
Bibliography: leaves 90-92.
General Note:
Typescript.
General Note:
Vita.
Statement of Responsibility:
by Fred Robert Frankel.

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
Copyright [name of dissertation author]. Permission granted to the University of Florida to digitize, archive and distribute this item for non-profit research and educational purposes. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder.
Resource Identifier:
000896497 ( ALEPH )
25427044 ( OCLC )
AEK5126 ( NOTIS )

Downloads

This item has the following downloads:


Full Text










COMPARATIVE CHEMICAL AND BIOLOGICAL

STUDIES ON THE FRACTIONATION

OF DEOXYRIBONUCLEIC ACIDS











By
FRED ROBERT FRANKEL


A DISSERTATION PRESENTED TO THE GRADUATE COUNCIL OF
THE UNIVERSITY OF FLORIDA
IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE
DEGREE OF DOCTOR OF PHILOSOPHY









UNIVERSITY OF FLORIDA
August, 1960















ACKNOWLEDGMENT


The research attitudes and techniques acquired by the candidate

as a result of his association with Dr. Charles F. Crampton will be a

guide in all future activities of the candidate. His debt to Dr.

Crampton is truly great. The candidate would also like to acknowledge

the interest of his supervisory committee and other members of the

Department of Biochemistry, and to thank Dr. Frank W. Putnam for his

careful criticism of this manuscript during its preparation. The

interest and generosity of Dr. Joshua L. Edwards during the candidate's

two year stay in the Department of Pathology is especially appreciated.
















TABLE OF CONTENTS


Page

LIST OF TABLES v

LIST OF FIGURES vii

INTRODUCTION 1

Chemical and Biological Heterogeneity of Deoxyribonucleate -- 1

EXPERIMENTAL PROCEDURES 4

Isolation and Purification of Deoxyribonucleates -- 4
DeoxyrLbonucleates from calf thymus and rat liver -- 4
Deoxyribonucleates from Ehrlich ascites tumor cells -- 4
Chromatography of Deoxyribonucleate by Use of Mg IRC-50 -- 8
Analysis and Quantitative Estimation of Purines and Pyrimi-
dines in Hydrolysates of Deoxyribonucleates -- 13
Use of columns of Dowex 50-X4 -- 13
Paper chromatography -- 15
Radioisotope Counting Techniques -- 16
Carbon-14 labeled compounds -- 16
Tritium labeled compounds -- 17

RESULTS AND DISCUSSION 18

Studies of the Interaction Between Mg IRC-50 and Deoxyribonu-
cleate -- 18
Specificity of the interaction -- 18
Capacity of Mg IRC-50 for deoxyribonucleate -- 21
Effect of certain variables on the interaction -- 25
Temperature -- 25
pH -- 27
Ions other than Mg -- 27
Flow rate -- 32
Particle size of resin -- 34


iii















Chromatographic behavior of deoxyribonucleates from rat
liver, Ehrlich ascites tumor cells, and Pneumococcus -- 34
Rechromatography of fractions of deoxyribonucleate from
calf thymus and rat liver -- 41
Effect of sample load -- 42
Base composition of fractions of deoxyribonucleate from
calf thymus and Ehrlich ascites tumor cells -- 47
Incorporation of Precursors into Fractions of Deoxyribonucleate
from Ehrlich Ascites Tumor Cells -- 50
Preliminary experiments with C14-formate -- 50
Effect of non-radioactive format -- 56
Effect of time -- 59
Incorporation of C14 -formate into fractions of deoxyribo-
nucleate -- 61
Incorporation of H -thymidine into fractions of deoxyribo-
nucleate -- 69
In vivo study of brief duration -- 69
In vivo study of long duration -- 71
In vitro study of brief duration -- 73

CRITICISM OF THE CHROMATOGRAPHIC STUDIES 76

POSSIBLE SIGNIFICANCE OF THE BIOLOGICAL STUDIES 82

SUMMARY .. 89

REFERENCES .. 90















LIST OF TABLES


Table Page

I. Properties of some specimens of deoxyribonucleate 9

II. Base composition of successive fractions of deoxyribo-
nucleate eluted from Mg IRC-50 22

III. Estimation of maximum capacity of Mg IRC-50 for deoxy-
ribonucleate 24

IV. Distribution of constituents among chromatographic
fractions of deoxyribonucleate from Pneumococcus 38

V. Recovery of "heat peak" material from deoxyribonucleate
preparations from various sources 39

VI. Base composition of chromatographic fractions of deoxy-
ribonucleate from calf #32 thymus obtained by over-
loading columns of Mg IRC-50 46

VII. Base composition of chromatographic fractions of deoxy-
ribonucleate from calf #32 thymus .. 49

VIII. Base composition of chromatographic fractions of deoxy-
ribonucleate from Ehrlich ascites tumor cells
(EA #1) 53

IX. Base composition of chromatographic fractions of deoxy-
ribonucleate from Ehrlich ascites tumor cells
(EA #7) 54

X. In vivo incorporation of C 14-formate into fractions of
deoxyribonucleate from Ehrlich ascites tumor cells
(EA #1) .. 58

XI. Five minute in vivo incorporation of C 4-formate into
the thymine of fractions of deoxyribonucleate from
Ehrlich ascites tumor cells (EA #7) 63

XII. The specific activity of bases in fractions of deoxy-
ribonucleate from Ehrlich ascites tumor cells
(EA #7) 66















XIII. Five minute in vivo incorporation of C 4-formate into
the thymine of fractions of deoxyribonucleate from
Ehrlich ascites tumor cells (EA #9) 68

XIV. Five minute in vivo incorporation of tritiated thymi-
dine into fractions of deoxyribonucleate from
Ehrlich ascites tumor calls (EA #16) 70

XV. Twenty-three hour in vivo incorporation of tritiated
thymidine into fractions of deoxyribonucleate from
Ehrlich ascites tumor cells (EA #17) 72

XVI. Two minute in vitro incorporation of tritiated thymi-
dine into fractions of deoxyribonucleate from
Ehrlich ascites tumor cells (EA #15) 75

XVII. Relative specific activity of thymine in chromato-
graphic fractions of deoxyribonucleate from Ehrlich
ascites tumor cells 83















LIST OF FIGURES


Figure Page

1. Chromatography of purines and pyrimidines by use of
60 x 0.9 cm columns of Dowex 50-X4, "through 200" 14

2. Chromatography of deoxyribonucleate by use of (A) a
50 ml mixing chamber or (B) a 125 ml mixing cham-
ber with columns of Mg IRC-50, "through 200" 19

3. Stepwise elution of deoxyribonucleate from a 30 x 0.9 cm
column of Mg IRC-50, "through 325," at 30 C 20

4. Chromatography of deoxyribonucleate (A) at 300 C and
(B) at 500 C by use of 29 x 0.9 cm columns of
Mg IRC-50, "through 325" 26

5. The effect of certain variables on the chromatography
of deoxyribonucleate from calf #32 thymus 28

6. Chromatography of specimens of deoxyribonucleate from
Pneumococcus by use of columns of Mg IRC-50 36

7. Rechromatography of deoxyribonucleate from calf thymus
by use of columns of Mg IRC-50, "through 200" 43

8. Chromatography of deoxyribonucleate from rat liver by
use of 29 x 0.9 cm columns of Mg IRC-50, "through
325" 44

9. Preparative chromatography of deoxyribonucleate from
calf #32 thymus by use of overload conditions 48

10. Chromatography of deoxyribonucleate from Ehrlich
ascites tumor cells (EA #1) by use of Mg IRC-50 51

11. Preparative chromatography of deoxyribonucleate from
Ehrlich ascites tumor cells (EA #7) by use of over-
load conditions 52

12. Effect of time of incorporation of C 4-formate on the
specific activity of bases of deoxyribonucleate
from Ehrlich ascites tumor cells 60

13. Stepwise elution from Mg IRC-50 of deoxyribonucleate
from Ehrlich ascites tumor cells (EA #9) 67















INTRODUCTION


Chemical and Biological Heterogeneity of Deoxyribonucleate


Near the end of the nineteenth century, Miescher isolated an

acid-insoluble phosphorus- rich material from salmon sperm, and from

the nuclei of pus cells and other tissues. He termed this material

"nuclein" (1). Later workers showed that nuclein was, in fact, a

mixture of at least two substances, one containing the sugar, ribose,

and the other, the sugar, deoxyribose. This latter type of nuclein,

which was confined almost exclusively to cell nuclei, became known as

deoxyribonucleic acid.

Much of the foundation of our present knowledge of the chemical

structure of nucleic acids was established by P. A. Levene (2). More

recently, through the development and use of quantitative semi-micro

techniques for the analysis and estimation of purines and pyrimidines,

Chargaff and his colleagues showed that specimens of deoxyribonucleate

from different biological sources frequently differ in their nitrogenous

base composition (3). These workers concluded that the nuclei of dif-

ferent species may contain different molecules of deoxyribonucleate.

In agreement with this chemical evidence of species-specificity, is the

growing body of biological evidence which suggests that genetic informa-

tion may be embodied in the composition and structure of deoxyribonu-

cleate. Moreover, specimens of deoxyribonucleate from individual

sources have recently been found to consist of populations of closely

1









2

related but chemically distinct molecules. This was first shown by

Chargaff, Crampton, and Lipshitz, who extracted fractions of the nucle-

ate having decreasing contents of guanine and cytosine from histone-

chloroform gels by means of aqueous solutions of increasing salt con-

centration (4, 5). Similar findings were obtained by other workers

whose techniques also depended upon the specific, fractional dissoci-

ation of deoxyribonucleate from cationic adsorbents to which it had

been bound by electrostatic linkages (6, 7, 8). However, the elution

of deoxyribonucleate from columns of cellulose linked to triethanol-

amine residues by reaction with epichlorohydrin (ECTROLA-cellulose)

provides successive fractions of essentially unchanging composition,

but of gradually increasing average sedimentation constant (9). Thus,

another important aspect of the molecular heterogeneity of deoxyribo-

nucleate arises from variation with respect to sedimentation proper-

ties (9, 10, 11). Recent studies suggest that molecules in a single

specimen of deoxyribonucleate are also heterogeneous with respect to

their apparent density when analyzed by centrifugation in solutions of

CsCl (12). Such studies indicate that apparent density is related

directly to base composition. Evidence has also been presented for a

relationship between base composition and thermal stability of deoxy-

ribonucleate (13).

With few exceptions, the amount of deoxyribonucleate per nucle-

us in cells of a given species is a multiple of the haploid amount, and

is characteristic of the species (14). If the two daughter cells

arising from division contain the same amount of deoxyribonucleate as

the parent cell, division must be preceded or accompanied by a net









3

synthesis of deoxyribonucleate. At present, little is known regarding

those factors which initiate and govern the synthesis of the various

molecules of deoxyribonucleate present in mammalian cells or of the

mechanisms which control the constancy of their amounts per nucleus.

That the net synthesis of deoxyribonucleate is confined to interphase in

cells from several mammalian tissues, was shown in 1950 by means of cyto-

chemical techniques (15). This finding was confirmed by radioautographic

evidence (16), and has recently been extended to Ehrlich ascites tumor

cells (17). As has been shown by several groups of workers (18, 19),

the deoxyribonucleate of mammalian cells exhibits extreme metabolic

stability, suggesting that it is not degraded during the life of the

cell, or that, if degraded, its substance is efficiently reutilized.

The work to be described below stems from the paradoxical find-

ing that deoxyribonucleate, despite its polyanionic character, is bound

by columns of the magnesium form of IRC-50, a polycarboxylic acid resin.

Elution of the deoxyribonucleate yields successive fractions of gradu-

ally changing composition (20). In the present studies, attempts were

made 1) to investigate the influence of various factors on the inter-

action of deoxyribonucleate from calf thymus with Mg IRC-50, 2) to de-

termine the usefulness of this system for the chromatographic fractiona-

tion of deoxyribonucleate from rat liver, Pneumococcus, and Ehrlich asci-

tea tumor cells and 3) to compare the initial rates of in vivo incorpo-

ration of radioactive precursors into chromatographically different frac-

tions of deoxyribonucleate. The latter studies of incorporation were

undertaken with the hope of obtaining information regarding metabolic

properties of different molecules of deoxyribonucleate.















EXPERIMENTAL PROCEDURES


Isolation and Purification of Deoxyribonucleates


Deoxyribonucleates from calf thymus and rat liver

Details of the isolation of the specimens from calf thymus and

rat liver have been prepared for publication. The specimens of deoxy-

ribonucleate from the thymus of calf #30 were isolated from a nucleo-

histone which had been extracted by 0.0004 M NaHCO3. Preparations 30A

and 30B were obtained by slightly different procedures, but were indis-

tinguishable with respect to all properties which were examined. The

specimen from calf #32 thymus was obtained by use of the procedure of

Crampton et al. which depends on the insolubility of deoxyribonucleo-

proteins in a solution of 6 M urea containing 1% NaCI and 0.1% merthio-

late. Nucleoproteins of the ribose type remain soluble in this medium,

the use of which has provided specimens of deoxyribonucleate containing

little or no uracil. Deoxyribonucleate was prepared from rat #11

liver using this same method. All of these preparations were depro-

teinized by two treatments with Duponol by use of the procedure of Kay

et al. (21).


Deoxyribonucleate from Ehrlich ascites tumor cells

The line of Ehrlich ascites tumor cells2 was maintained by


1C. F. Crampton, B. Greenfield, J. Adair, and F. R. Frankel,
manuscript in preparation.

2We are indebted to Dr. C. Heidelberger for supplying the
Ehrlich ascites tumor cells used in these studies.
4









5

weekly transfers of approximately 2 x 106 cells to groups of hybrid Swiss

white mice, fifty to sixty days of age. Cells for inoculation were

aspirated by use of sterile, large bore, transfer pipettes from the

peritoneal cavities of mice which had been killed by cervical dislo-

cation. About two to four minutes were required to collect the cells

which were then centrifuged briefly at room temperature. The packed

cells were then suspended in twenty volumes of sterile saline solution.

Appropriate aliquots of such suspensions, which contained about 2 x 107

cells per ml, were used for inoculation.

Specimens of deoxyribonucleate designated EA #1 to #17 were

isolated from cells donated by the respective groups of uniform mice.

The mice of each group received 2 x 106 to 20 x 106 cells from a single

inoculum five to six days prior to sacrifice. When deoxyribonucleate

was to be extracted, the cells were collected in the manner described

above, except that the peritoneal cavities of the mice were frequently

rinsed several times with cold saline in order to increase the yield of

cells. Moreover, the cells were always placed immediately in a tube

immersed in an ice bath. In studies of the uptake of labeled compounds,

the stated intervals of incorporation correspond to the average time

which elapsed between injecting the radioactive precursor and chilling

the cells. The chilled cells were promptly centrifuged at 40 C, and

the sediment remaining after decanting the supernatant fluid was im-

mediately frozen in a mixture of dry ice and ethanol. The pellets were

occasionally stored at -20o C for twenty-four hours before being proc-

essed further.









6

In order to extract deoxyribonucleate from the packed cells,

the pellets were permitted to thaw partially in the cold room (at 4o- 6C,

at which temperature all subsequent procedures were performed). The

packed cells (2 to 5 ml) were transferred with five to ten volumes

(with respect to the volume of packed cells) of the urea medium to the

semi-micro cup of a high speed mixer equipped with cutting blades. The

mixer was then operated at full speed for two minutes. In the case of

EA #1 and EA #4, the initial homogenates were prepared by use of a

Potter-Elvehjem tissue grinder equipped with a plastic pestle. After

the addition of two drops of octyl alcohol, in order to collapse the

foam, the homogenate was transferred to a plastic centrifuge tube, and

the insoluble deoxyribonucleohistone and other cell debris were sedi-

mented by centrifugation at 20,000 x z for ten minutes. The sediment

was redispersed completely in five to ten volumes of fresh urea medium

with the aid of a mechanically driven pestle which had been machined to

fit the centrifuge tube tightly. The suspension was centrifuged as

before, and the resuspension and sedimentation were repeated using

fresh urea medium. The sediment was then washed two times in the same

manner with 0.15 M NaCl. The sediments at this point for preparations

EA #1, #4, #5, #6, and #7 were dispersed in 4 ml of distilled water and

made about 1 M in NaCl by the addition of 5 ml of 2 M NaCl. In order

to reduce the viscosity, 1 to 15 ml of I M NaCl were added to dilute the

suspensions, which were then stirred for eleven to fifteen hours in

order to effect dissociation and solubilization of the histone and

deoxyribonucleate. The suspensions in 1 M NaCl were then centrifuged

for thirty to sixty minutes at 20,000 x g in order to remove insoluble









7

constituents. Before extraction with 1 M NaCl, the sediment of EA #9

was first extracted with nine volumes, and then with one volume, of

0.2 M Ba(OAc)2 in order to dissociate histone Fraction A (22). The

sediments of BA #15, #16, and #17 were extracted with five volumes of

0.2 M Ba(OAc)2 before being dispersed in four volumes of 2 M NaCI. The

suspensions in 2 M NaCI were stirred one hour, centrifuged at 20,000 x

g for fifteen minutes, and the sediments were re-extracted with 2 M

NaCI. The combined, clear to slightly turbid, 1 M or 2 M NaC1 super-

natant fluids from each experiment were transferred to a 125 ml Erlen-

meyer flask which was shaken vigorously after the addition of one volume

of 957. ethanol. The fibers of precipitated deoxyribonucleate were

washed once with 707. ethanol.

For further deproteinization, the precipitated deoxyribonucle-

ate was suspended in 5 to 10 ml of distilled water, and an equal volume

of 0.2 M NaCl, 0.05 M Na citrate was added. Duponol was added to a

concentration of 0.457. and the solution was brought to room temperature

and stirred for three hours. Solid NaCI was added to a concentration

of 1 M, and permitted to dissolve completely, after which the solution

was centrifuged at 20,000 x g for fifteen minutes at 40 C. Deoxyribo-

nucleate was precipitated from the supernatant fluid by the addition of

an equal volume of 95% ethanol, collected, and washed with 70% ethanol.

After stirring the nucleate with 8 to 20 ml of distilled water for one

to twelve hours at 40 C, a second Duponol treatment was performed as

described above for two hours (EA #1, #4, #5, #6, and #7) or one hour

(EA #9, #15, #16, and #17). After the final precipitation by ethanol,

the deoxyribonucleate was washed three times with 70% ethanol, three









8

times with 957. ethanol, and dried overnight at room temperature in an

evacuated desiccator over anhydrous CaCl2. The product was stored at

-20o C.

Table I lists the yields and several properties of some of the

specimens of deoxyribonucleate used in these studies. The yields of

other specimens from Ehrlich ascites tumor cells, per ml of packed

cells, were as follows: EA #1, 4 mg; EA #7, 7 mg; EA #9, 3 mg; EA #15,

7 mg; EA #16, 7 mg; EA #17, 6 mg.


Chromatography of Deoxyribonucleate by Use of Mg IRC-50

Amberlite IRC-50, XE-64 (CG-50, Type 2, presently supplied by

Fisher Scientific Co.), was obtained in the hydrogen form as the

"through 200" mesh material; it was prepared for chromatography by first

converting the resin to the sodium form by adjusting to pH 8 or 9 a

6-liter suspension of the resin in distilled water by the addition of

pellets of NaOH. The finest resin particles were removed by repeated

decantations of the thoroughly stirred 6-liter aqueous suspension after

settling periods of about thirty minutes. After about ten decantations,

the settled resin was transferred in small portions either to a 200

mesh or a 325 mesh sieve and screened to yield "through 200" or "through

325" material (23). The material passing through the sieves was col-

lected and slowly washed under gravity on a Buchner funnel with succes-

sive 2 to 4 liter portions of 4 N NaOH, water, 4 N HC1, water, 4 N

NaOH, water, 4 N HCI, water, acetone, and water. The resin was con-

verted to the magnesium salt by slowly passing 1 M Mg(OAc)2 through the

resin bed until the pH of the effluent rose to that of the influent,















TABLE I
Properties of some specimens of deoxyribonucleate

Details of the methods used for the isolation and characteriza-
tion of these specimens are given in the experimental section.

Gross Loss of Per centa
Source of y(m/e weight Protein 0.1 d
specimen or on NaC
a/ml)b drying Phosphorus Biuret Ninhydrin

Calf #30A thymus 19 14.1 8.56 1.2 3.6 6930

Calf #32 thymus 24 13.9 8.28 2.9 4.5 6830

Rat #11 liver 1 14.0 7.23 18 --- 7040

Ehrlich ascites
tumor cells
EA #4 2 17.8 8.72 1.2 2.7 6430
IA #5 5 18.6 8.94 3.0 2.3 6850
IA #6 4 17.0 9.11 1.6 4.4 6450

aAs, per cent of the weight after drying.

bThe figures refer to actual recoveries per g of tissue or per
al of packed tumor cells.
CAs, per cent of weight before drying.

dAtomic extinction coefficient with respect to phosphorus, at
260 =4.









10

about pH 7 or 8. The preparation of the resin was completed by washing

with 0.05 M Mg(OAc)2, pH 7 to 8. At this point, one-to-one slurries of

the resin in 0.05 M Mg(OAc)2 were prepared and used to pour successive

10 to 15 cm segments of analytical columns (0.9 cm diameter) or prepara-

tive columns (2 cm diameter), in the manner described by Moore and

Stein (23). Unless otherwise noted, the columns were operated in con-

denser jackets through which water of 30 + 0.20 C was circulated.

Effluent from the analytical columns was collected at a rate of about

20 ml per hour, while the preparative columns were operated at a rate

of flow of about 50 ml per hour. To promote the regular flow of drops

of uniform size past the light beam of the drop counter and into the

collection tubes, a roll of silver gauze was inserted into the tip of

the column (24). The gauze made contact with the fritted glass disc of

the chromatograph tube and terminated in a point.

The zinc salt of IRC-50 was prepared by substituting solutions

of Zn(OAc)2 of pH 6 for the Mg(OAc)2 solutions in the above procedure.

Carboxymethyl-cellulose, with an exchange capacity of 0.39 meq per g

(Lot No. 59-1, supplied by Bio-Rad Laboratories) was obtained in the

hydrogen form, and was converted to the magnesium form as described

above. The sulfonated polystyrene resins (Dowex 50-X2 and Amberlite

IR-120, XE-69) were prepared as described previously (25) and converted

from the hydrogen forms to the magnesium or zinc forms as for IRC-50.

The sample to be chromatographed contained, in most cases,

about 1 mg of deoxyribonucleate per ml of the initial buffer, and was


3C. F. Crampton, F. R. Frankel, A. M. Benson, and A. E. Wade,
submitted for publication.










11
prepared by swirling the nucleate in distilled water at 4 C until the

specimen was dissolved. This required up to ten hours. Sufficient 1 M

salt solution was then added to yield a final concentration of the

starting buffer of 0.05 M. Stirring was continued for about two hours,

whereupon the solution was permitted to warm to room temperature, and

after the removal of aliquots for the purpose of estimating the total

absorbancy, the sample was added carefully to the column. The deoxy-

ribonucleate was eluted by eluents of continuously increasing concen-

tration or in a stepwise manner. Eluents of gradually changing compo-

sition were produced by use of a device constructed from a flat bottomed

flask which served as the mixing chamber and contained the solutions of

lowest concentration, and the upper ground glass joint of a wash bottle

through which was introduced from a separatory funnel, the solutions of

highest concentration. During operation of the device, the solution in

the mixing chamber was stirred by a magnetic bar and the eluent was

continuously removed through the delivery tube of the wash bottle. Most

of the analytical chromatograms were performed by use of a 500 ml mixing

chamber containing 0.05 M Mg(OAc)2, into which flowed 0.4 M Mg(OAc)2.

Exploratory experiments indicated that a mixing chamber of this size

gave satisfactory results. Smaller mixing chambers gave too few frac-

tions to cut conveniently, while larger mixing chambers gave peaks

which were impractically broad. For preparative columns, the volume of

the mixing chamber was scaled up by a factor of six, although the ratio

of the cross-sectional areas of the two columns was 4.9. The effluent

from analytical columns was collected in fractions of 2 to 5 ml while

that from preparative columns was collected in fractions of 5 to 20 ml.









12

The absorbancy of these fractions was determined at various wave lengths

in the Beckman DU spectrophotometer, using distilled water as a blank.

The light path of the quartz absorption cells was 1 cm. In an effort

to eliminate the contribution of ultraviolet absorbing constituents in

the effluent other than deoxyribonucleate, use was made of the differ-

ence between the 260 vU and 290 mn absorbancies. Amounts of deoxyribo-

nucleate, expressed as "absorbancy units," were calculated by multi-

plying the 260 mn minus 290 ig absorbancy difference by the total volume

of the solution. The observation that one mg of deoxyribonucleate per

ml of solution gives an absorbancy difference of 12.3 was uaed to con-

vert absorbancy units to mg of deoxyribonucleate. Column fractions

were stored no longer than forty-eight hours at 4 before further

treatment.

To isolate deoxyribonucleate from the column effluent, fractions

were combined where appropriate, made 1 M with respect to NaC1 by the

addition of solid salt, and the nucleate precipitated by the addition
4
of two to three volumes of cold 957. ethanol. The precipitates were

washed three times with 70% ethanol and three times with 95% ethanol.

The product was dried in an evacuated desiccator over CaCd2 and stored


4To facilitate the isolation of small amounts of pneumococcal
nucleate with transforming activity, a known quantity of a solution of
deoxyribonucleate from calf thymus was added as "carrier" prior to
precipitation with ethanol. This procedure yielded difficulty in-
terpretable data with respect to the transformation assays. An alter-
native recovery procedure, subsequently developed by Dr. R. D. Hotchkies,
is based on the co-precipitation of the nucleate with magnesium phos-
phate, and the subsequent solution of the deoxyribonucleate by use of
othylenediaminetetraacetate.









13

at -20o C. Before precipitating the deoxyribonucleate from pooled sam-

ples which contained only small amounts of material, the fractions were

first concentrated by use of a rotary evaporator, the bath temperature

of which was kept less than 36 C. In some instances, such concen-

trated solutions were dialyzed at 4 C to equilibrium against distilled

water so that the final concentration of Mg(OAc)2 was about 0.5 M.


Analysis and Quantitative Estimation of Purines and Pyrimidines in

Hydrolysates of Deoxyribonucleates


Use of columns of Dowex 50-X4

In order to determine the base composition of specimens of

deoxyribonucleate as well as of fractions of deoxyribonucleate isolated

after chromatography by use of Mg IRC-50, the procedures described by

Crampton et al..3 were used. A specimen of the material to be analyzed

was hydrolyzed with 99% formic acid in an evacuated sealed tube for

thirty minutes at 1750 C. Aliquots of the hydrolysate, after removal

of formic acid in an evacuated desiccator and solution of the residue

in 1 N sulfuric acid, were taken for phosphorus determination (26) and

chromatographic analysis. The analyses were performed by eluting the

bases from columns of Dowex 50-X4 (60 x 0.9 cm) using an eluent of

gradually increasing concentration of ammonium format of pH 4. The

column effluent was collected in 3 ml fractions, and the absorbancy of

each fraction was determined in the ultraviolet spectrophotometer, using

distilled water as a blank. A typical analysis is shown in Fig. lB.

To estimate the amount of material in each position of the chromatogram,

factors obtained by chromatographing mixtures of standard bases under
















u A u rMVI,I- STANDARD I ZAT I ON MIXTURE


0.6 THYMINE GUANINEA
I, X.G 0.6

ADENINE


Q
CYTOSINE 5-METHYL-
2 260 CYTOSINE
, 0.2 nip ,
O 2 mP 250 mp
v 260 m"
0 o .2 8 0 f. mo
Fraction No. 20 40 60 80 100 120 140 160
(2.9-3.0 ml. per fraction throughout)
AMMONIUM FORMATE 0.2N, pH4--(500 m.)---*I.ON, pH4, 44C

1.4
THYMINE FORMIC ACID HYDROLYSATE OF DNA (1.2mg)
1. t FROM MOUSE ASCITES TUMOR CELLS

1.0
S.0 ADENINE
S0.8f 120 .g.

4 06 GUANINE
97 3jg.
04-
S 260 4CYTOSINE 5-METHYL-
02 CYTOSINE
2 260 mp (3,g) 260
Fraction No. 20 40 60 80 100 120 140 160
(2.9-3.0 ml. per fraction after fraction 12)
AMMONIUM FORMATE 0.2N. pH4--(500 ml.)--I.ON. pH4. 44C



Fig. 1. Chromatography of purines and pyrimidines by use of
60 x 0..9 cm columns of Dowex 50-X4, "through 200." (A) Standardiza-
tion mixture of approximately 50 pg of each base. (B) Formic acid
hydrolysate of 1.2 mg of deoxyribonucleate from Ehrlich ascites tumor
cells.










15

the same conditions (Fig. LA) were used to convert units of absorbency

to moles of base.


Paper chromatography

Frequently, an alternative method of chromatography was used

for the isolation of bases for measurements of specific radioactivity.

The residue resulting from the formic acid hydrolysis of about 0.6 to

1.0 mg of deoxyribonucleate was dissolved in two drops of 1 N HC1 and

spotted in duplicate at the origin of a 19.5 cm x 40 cm sheet of Whatman

No. 4 filter paper, leaving three lanes to serve as blanks. This was

hung in a sealed chromatography Jar, and developed by descending

chromatography for twenty-five to thirty-five hours at room temperature

(23o-260 C) with isopropanol which was 1.96 N with respect to HC1 (27).

The chromatogram was dried in a stream of warm air. The bases were

located by viewing the chromatogram under an ultraviolet lamp, and

were cut out, along with the adjacent blank spots, using a pattern to

assure elution of bases and blanks from equal areas of paper. The

paper segments were rolled into cylinders and deposited in 10 x 75 mm

test tubes. To each test tube was delivered 2 ml of eluting solvent

using a single calibrated volumetric pipette. Guanine and its blank

were eluted with 1 N HC1, while the other bases were eluted with


When dried and viewed under ultraviolet light, chromatograms
of bases which had been recovered from the effluent of columns of
Dowex 50 showed a single dark spot at the position of the authentic
base, as well as a bright yellow fluorescent spot which moved faster
than thymine. This fluorescent spot was seen whenever any base,
isolated from Dowex 50 effluent, was chromatographed on paper, but not
when the bases were added to the paper from stock solutions of the
bases, or directly from hydrolysates. Evidently, the fluorescent
material is derived from the sulfonated polystyrene resin.









16

0.1 N HC1. Thymine containing tritium was eluted with water. The tubes

were tightly covered with parafilm and allowed to stand twelve hours at

room temperature. At the end of this period, the absorbancy of a por-

tion of the eluate was determined in the ultraviolet spectrophotometer,

and aliquots of the remaining eluate were plated for determination of

specific radioactivity.


Radioisotope Counting Techniques


Carbon-14 labeled compounds

Bases containing carbon-14 were isolated from the pooled efflu-

ent of columns of Dovex 50 by lyophilization of the ammonium format

solution in 250 ml flasks with the aid of infra-red lamps directed at

the frozen solution. After the first lyophilization, the residue was

dissolved in 5 ml of water and relyophilized. The final residue was

taken up in 1 N HC1 (guanine) or 0.1 N HC1 (thymine, cytosine, adenine),

diluted to 2 ml with the same solvent, and an aliquot taken for deter-

mination of absorbancy. A portion of the remaining solution was

plated for determination of radioactivity. Bases labeled with carbon-14

and isolated by paper chromatography were prepared as described in the

previous section.

Compounds dissolved in 0.1 N HC1 could not be plated directly

on metal. They were therefore applied to aluminum planchets onto which

1.8 cm diameter glass cover slips were secured with the aid of a small

amount of silicone grease. Four equal aliquots containing about 5 pg

of the base were successively spread on the glass cover slip uniformly

covering the glass area, evaporated to dryness under an infra-red









17

lamp, and counted by use of a flow-window device. Counting was per-

formed at 130 volts above the starting voltage. A minimum of 3000

counts was accumulated for each sample. All counting rates were cor-

rected for background. A straight line could in most cases be drawn to

connect the points when counts per minute was plotted against total

volume plated, indicating that counting was performed under conditions

of infinitesimal thickness. The slope of the line which joined the

counting rates of the four successive aliquots was divided by the

moles of base in the sample, determined spectrophotometrically, to

obtain the specific activity of the sample.


Tritium labeled compounds

Thymine containing tritium was dissolved in water so that it

could be plated and counted on steel planchets. For this purpose, two

equal aliquots of 50 L1 containing about 0.6 ug of base were succes-

sively spread uniformly on a planchet previously washed thoroughly

with detergent, pickled by brief immersion in 6 N HC1, washed finally

with water and acetone, dried with a stream of warm air, and rimmed

with silicone grease. The planchets, with an effective diameter of

2.7 cm, were dried under an infra-red lamp on a rotating platform in a

stream of air, and counted in a windowless flow counter operated with

99.05% helium and 0.95% isobutane. Counting was performed 175 volts

above the starting voltage. The specific activities of the two ali-

quots were extrapolated to zero mass to obtain the specific activity of

the sample in the absence of self absorption.















RESULTS AND DISCUSSION


Studies of the Interaction Between Mg IRC-50 and Deoxyribonucleate


Specificity of the interaction

Under certain conditions, deoxyribonucleate is adsorbed by the

magnesium form of the polycarboxylate resin, Amberlite IRC-50. Fig. 2A

illustrates an experiment in which 1 mg of deoxyribonucleate from calf

#30B thymus was dissolved in 0.05 M Mg(OAc)2, pH 7.6 and added to a

column of Mg IRC-50 which had been equilibrated with an eluent of the

same composition as the solvent. Virtually all of the nucleate was

bound. The binding was reversible in that about 90% of the nucleate

was eluted by applying an eluent of gradually increasing Mg(OAc)2 con-

centration. The eluent was produced by adding 0.5 M Mg(OAc)2 to a 50 ml

mixing chamber initially filled with 0.05 M Mg(OAc)2. As shown in

Fig. 2B, a slower rate of increase of the salt concentration in the

eluent (achieved by use of a 125 ml mixing chamber) provided a broader

elution pattern with the same specimen. The concentrations of Mg(OAc)2

sufficient to elute most of the molecules of deoxyribonucleate were

determined by stepwise application of solutions of successively higher

molarity. As shown in Fig. 3, obtained with the specimen from calf #32

thymus, elution occurred over the relatively narrow range of 0.20 M to

0.25 M Mg(OAc)2.

In order to determine whether the elution of successive frac-

tions is specific, the composition was examined of three successive
18










































































(06-0 0
(O6Z-09Z)V o


S4

64 4

44
1-1 4







a 10
_i -t s
-. Ch M
(I*54 P
V5 4o





0 "4
0 LM g
0U to 4

CO



.5*a I $
o "o w0


0(

-s ~~ ^3 0




OS
0 U4








4.0
< 0


f44




E 'idw
.54 6 14 .4









oL 4Uow
B g *
o o -d<















0 0
i-i "4 o












"04 r,;:;4w
0 0 0
0 S4 04





CPC
.Ae "
ok4 ASl
*E S0,0

"421:

x *0


CM "C

(06Z-09Z)V '0*0























































0 %
N0
- 0 0
(7IA106Z SnfNIIAI 77AI092)

AONV80OSeV


0 0
0o





-
O
0
U

o T
0z




0










2
0






z
m
u)
U.


cr


zo
z
0
Ir-
ii.


U0


I-









0.


Cf)

z
M

U)

-1J
0


0



1-

0
I


0







u
%4





S44
5-





AS
0 '


44
0
05
*0g
-oI




4 4
0 0

0


v0

P*g


a4


00

00









21

fractions recovered from the effluent in an experiment similar to that

shown in Fig. 2B. The fractions were chosen so as to contain 20%, 40%,

and 20% of the deoxyribonucleate applied to the column. As can be

seen in Table II, the first and last fractions differed significantly

in composition from the total preparation. These results indicate that

the interaction of Mg IRC-50 with deoxyribonucleate is specific since

reducing the capacity of the resia by increasing the Mg(OAc)2 concen-

tration causes a preferential elution of molecules rich in guanine and

cytosine. That is, the affinity of the resin for molecules rich in

adenine and thymine is greater than for molecules rich in guanine and

cytosine. Accordingly, use of the interaction offers a means for

separating molecules of deoxyribonucleate of different composition.


Capacity of Mg IRC-50 for deoxyribonucleate

A fundamental consideration in studies of the interaction of

any adsorbent with a solute is the ultimate capacity of the adsorbent,

since the effects of variables may be difficult to interpret if experi-

ments are performed under conditions close to saturation. In order to

estimate the maximum capacity of Mg IRC-50 for deoxyribonucleate,

73.4 ml of a 0.1% solution in 0.05 M Mg(OAc)2 of the specimen from

calf #32 thymus were added to a 28 x 2 cm column of "through 200"

resin which had been equilibrated with 0.05 M Mg(OAc)2, pH 7.2. The

capacity of the resin was exceeded under these conditions, since

deoxyribonucleate appeared in the effluent after about 80 ml had been

collected. Unadsorbed nucleate was then removed from the column by

irrigation with 0.05 H Mg(OAc)2 until the absorbency of the effluent















TABLE II
Base composition of successive fractions of deoxyribonucleate
eluted from Mg IRC-50

The fractions were obtained from deoxyribonucleate from calf
#30B thymus as noted in the text by a chromatographic experiment simi-
lar to that shown in Fig. 2B. The percentage in parentheses which
follows each fraction number represents the portion of the deoxyribo-
nucleate of the sample recovered in the respective fraction. The
results are expressed as moles of base per 100 moles of total recovered
bases. From Reference (20).

Fraction number Unfractionated,
Base total
1(20%) 11(40%) III(20%) deoxyribonucleate

Thymine 25.4 29.0 30.0 27.6

Guanine 24.2 21.7 20.6 22.7

Cytosine 23.6 20.2 19.7 22.3

Adenine 24.5 27.1 28.5 27.5

5-Methylcytosine 2.1 1.7 1.3 (1.4)









23

approached base line. The results of replicate experiments presented

in Table III show that 1 cm of packed resin binds about 0.3 mg of deoxy-

ribonucleate.

As will be shown by the studies of base composition presented

below, molecules of nucleate which are rich in guanine and cytosine

have a low affinity for the resin and are eluted by 0.05 M Mg(OAc)2

under the experimental conditions used to estimate the capacity of

Mg IRC-50. When the extent of interaction is limited by the number of

binding sites on the resin, molecules of deoxyribonucleate with the

greatest affinity apparently compete with, and/or displace the less

strongly bound molecules. Accordingly, the absolute capacity of the

resin may well vary with the base composition or physical structure of

the molecules tested. Deoxyribonucleate degraded by heat or by enzyme

action may also interact differently with the resin. Moreover, particle

size may affect the interaction, since equivalent amounts of smaller

particles will present a greater area for the binding of macromolecules

which are unable to penetrate the surface of the resin. The estimated

maximum capacity of about 0.3 rag of deoxyribonucleate per cm3 of Mg IRC-50

may be compared with 0.5 to 2.3 mg per cm3 found for columns of

ECTEOLA-cellulose (9). The value for the capacity of ECTEOLA-cellulose

may include, however, entrained deoxyribonucleate which was eluted in

the procedure used to estimate the capacity of Mg IRC-50 as described

above.

In the experiments to be described, the amounts of deoxyribo-

nucleate chromatographed were less than one-fourth, and usually about















TABLE III
Estimation of maximum capacity of Mg IRC-50 for deoxyribonucleate
3
In both experiments 28 x 2 cm columns which contained 88 cm
of packed resin were used. The sample was added as a 0.1% solution
of deoxyribonucleate prepared in 0.05 H Mg(OAc)2, pH 7.2.

Experiment number
Conditions
743 744

Deoxyribonucleate added

Milligrams 74 75
Total absorbancy units 904 920

Deoxyribonucleate eluted by 0.05 M Mg(OAc)2

Milligrams 46 46
Total absorbency units 560 564

Capacity

Milligrams per column 28 29
Milligrams per cm3 of packed resin 0.32 0.33









25
one-seventh of the estimated maximum capacity of the columns, except

where "overload" experiments explicitly were performed.


Effect of certain variables on the interaction

Temperature -- The influence of temperature on the interaction

between deoxyribonucleate and Mg IRC-50 is illustrated in Fig. 4.

Deoxyribonucleate was eluted by a much lower salt concentration from

the column operated at 30 C than from the column equilibrated and

operated at 50 C. In contrast to the greater ease of elution custom-

arily observed when low molecular weight compounds such as amino acids

(28) and purines and pyrimidines3 are chromatographed at higher temper-

atures, the deoxyribonucleate is more strongly bound by the Mg IRC-50

at the higher temperature. With small molecules, variations in temper-

ature may affect ionizable groups of the chromatographed compound, of

the buffer, and of the resin. The contribution of secondary valence

forces to the overall interaction may also be altered. However, the

effect of temperature on the chromatography of deoxyribonucleate using

Mg IRC-50 appears to be dominated by other factors. The greater retar-

dation observed at the higher temperature might best be explained in

terms of effects on the physical structure of the resin or deoxyribo-

nucleate. Heating may increase the number of effective binding sites

per unit area of the surface of the resin by increasing the flexibility

of the cross-linked polymer chains, thereby allowing the aliphatic

matrix of the resin to swell. Higher temperatures may also reduce the

degree of hydration of the two polymers, a factor which also would be

expected to facilitate their mutual interaction. Elevated temperatures












8
0a


0
0 A

05
S.0





o .c
I o










"4 k
0 0





T -









0
S "






r LS c.4 "|
(/ M8 .40

o bo w







I Q 4 1 41



S0o N
N m .44














AONV8 0S8 4
0
'i ^ i N a
(\j 1
^ ^43









27

may induce still other structural alterations favorable to interaction

such as changes in the extent of intramolecular hydrogen bonds in cer-

tain regions of the molecules of deoxyribonucleate.

pR -- The interaction between deoxyribonucleate and Mg IRC-50

is not appreciably sensitive to the precise value of pH of solutions of

Mg(OAc)2 over the range, pH 7.2 to pH 8.1, where most of the chroma-

tographic experiments were performed. However, deoxyribonucleate passes

without apparent retardation through columns of Mg IRC-50 which are

equilibrated and operated with solutions of Mg(OAc)2 of pH 6.1. A

typical experiment performed at pH 6.1 is illustrated in Fig. 5A. It

is reasonable to ascribe the absence of interaction at 0.05 M Mg(OAc)2,

pH 6.1, largely to the protonization of negative groups of the resin or

the deoxyribonucleate or both. In this pH region, the carboxyl groups

of Amberlite IRC-50 exhibit some buffering (29); the effect on the

nucleate is probably limited to the secondary phosphoryl groups, whose

pK is about 6.5 (30). It is possible that examination of the inter-

action at values of pH between 6.1 and 7.2, as well as at different

molarities of Mg(OAc)2, would permit a decision as to whether the

observed pH effect depends more on changes in the nucleate than on

changes in the resin.

Ions other than Mg -- In the interaction between IRC-50 and

deoxyribonucleate, Mg may be regarded as a counter ion which at low

concentrations is somehow responsible for the mutual affinity between

both polymeric moieties. Higher concentrations of Mg++ destroy the

specific interaction as indicated by elution of molecules with gradually

varying composition. In order to determine whether deoxyribonucleate


















































T

0I



-.





_F


(71Vio06 SnNIWI TIN09Z)
A*ONVdOSOV


0.q 0
0-I 40





AjA





w0 'I



1 4l di4
* 0 0
Sa N 0 Q















01,10

w' 0 t-
:be a
UN UJ .0




4Kj 0 4.
-Q o-4 0












1*4 $4 4U



4* 0 S

0 0 0 N 4
| ..8 3



















0k.00
' 0 44 tO
U' 0 *S






.0 u 0.4 5
4 004


Ic,
-ro









29

can be eluted by other ions, a specimen from calf #32 thymus was ad-

sorbed to a column of Mg IRC-50 under the usual conditions, but the

eluent was produced by adding 4.7 M NaOAc, pH 7.0, to a 250 ml mixing

chamber initially filled with 0.05 M Mg(OAc)2, pH 7.3. The concen-

tration of Na+ originally chosen for the reservoir was high in order to

compensate for the lower eluting power expected for a monovalent cation.

However, 907. of the deoxyribonucleate was eluted by the time three

hold-up volumes were collected. The unexpected prompt elution by NaOAc

was confirmed by an otherwise similar experiment in which an eluent of

pH 6.4 was produced by adding 1 M NaOAc to 0.1 M NaOAc. In this case

(illustrated in Fig. 5B) 1047. of the sample was recovered in less than

two hold-up volumes. Thus, NaOAc, on a molar basis, was more than

twice as effective as Mg(OAc)2 in eluting deoxyribonucleate from Mg IRC-50.

Although it remains to be determined whether elution with NaOAc will

provide a series of fractions of gradually varying composition, it is

noteworthy that elution from Mg IRC-50 by sodium salts has been utilized

to effect a partial separation of deoxyribonucleates from E. coli and

Pneumococcus.6

Although Na IRC-50 does not bind deoxyribonucleate at 0.2 M

NaOAc, pH 6,7 exploratory experiments have shown that divalent cations

other than Mg are able to mediate the interaction between deoxyribo-
8
nucleate and IRC-50. Thus, deoxyribonucleate is bound by Ba IRC-50


6R. D. Hotchkiss and L. Mindich, personal communication (1959).

7C. F. Crampton, unpublished observations (1956).
8Unlike Mg IRC-50, the Mg form of Dowex 50-X2 (sulfonated poly-
styrene beads, "through 200" mesh) did not bind an appreciable amount












at 0.1 M Ba(OAc)2, pH 6.7, but is not eluted by 1 M Ba(OAc)2, pH 6.

Similarly, all of the deoxyribonucleate in a solution of Zn(OAc)2 was

adsorbed to a column of Zn IRC-50 which had been equilibrated with

0.05 M Zn(OAc)2, pH 6.0. It will be recalled, in contrast, that deoxy-

ribonucleate was not bound by Mg IRC-50 at pH 6.1. Moreover, none of

the deoxyribonucleate was eluted by 1 M Zn(OAc)2 of pH 6.0 as shown in

Fig. 5C. In view of the effectiveness of Na in bringing about elution

of deoxyribonucleate from Mg IRC-50, a solution of 1 M NaOAc, pH 6.0,

was next passed through the column. As can be seen, 82% of the applied

material was thereby recovered.


of deoxyribonucleate at 0.05 M Mg(OAc)2, pH 7.5. Similarly, the Mg form
of Amberlite IR-120, XE-69 (crushed particles of sulfonated polystyrene,
"through 32'5 mesh) did not bind the nucleate even when columns were
operated at 0 C or at 5 C when the molarity of the Mg(OAc)2 was re-
duced to 0.01 M. The Zn form of the XE-69 resin failed to bind deoxy-
ribonucleate at 5 C when equilibrated with 0.05 M Zn(OAc)2 of pH 6.
These experiments were performed in an effort to find an ion exchanger
for the chromatography of deoxyribonucleate which would equilibrate with
the initial eluent more rapidly than IRC-50. However, the same factor
responsible for the rapidity with which sulfonated polystyrene reaches
equilibrium, namely, the predominantly ionic character of the linkage
between sulfonate and cation, may underlie the inability of this resin
to form a stable complex with deoxyribonucleate.
The Mg form of carboxymethyl-cellulose, a resin with reactive
sites nearly identical to those of IRC-50, also failed to bind deoxyri-
bonucleate when a column was operated with 0.05 M Mg(OAc)2, pH 7.4, at
300. It is possible that the carboxyl groups of carboxymethyl-cellu-
lose are not disposed in a specific geometric configuration conducive
to the formation of stable complexes with deoxyribonucleate. Moreover,
carboxymethyl-cellulose lacks the hydrophobic matrix and methyl side
chains which in IRC-50 may reinforce linkages involving divalent cat-
ions. The possible importance of the methyl groups attached to the
hydrophobic matrix of IRC-50 is emphasized by an unpublished experiment
of Crampton, Moore, and Stein (1956) which indicated that the Mg form
of polyacrylic acid (Amberlite XE-112) did not bind deoxyribonucleate
under conditions where polymethacrylic acid did.
9C. F. Crampton, S. Moore, and W. H. Stein, unpublished obser-
vations (1956).









31

The results discussed thus far suggest that under certain con-

ditions, there is formed a ternary complex involving the negatively

charged carboxyl groups of IRC-50, a divalent cation such as Mg ,

Zn or Ba and the negatively charged phosphate groups of deoxy-

ribonucleate. The observation that Zn IRC-50 binds nucleate at pH 6,

while Mg IRC-50 does not, may be related to the fact that Zn+ forms

stronger chelation complexes than Mg (31). If at certain critical

salt concentrations, the bonds participating in complex formation are

weakened, or salt becomes plentiful enough for the formation of stable

binary complexes, the nucleate would be desorbed from the resin.

Whether a common mechanism, depending entirely on the ionic

strength of the eluent, is responsible for elution by both NaOAc and

Mg(OAc)2, or whether there are additional effects such as cation

binding, cannot as yet be decided. In any event, the desorption proc-

ess appears to be specific and the cation concentration required for

elution depends on the composition of the nucleate molecule.

It is reasonable to suggest that binding sites on deoxyribonu-

cleate containing guanylic and cytidylic nucleotides, or in the vicinity

of these residues, have a much lower affinity for Mg IRC-50 than other

locations on the deoxyribonucleate polymer. Recent theoretical (32) and

experimental (33) studies suggest that guanine-cytosine pairs have a

higher electron density than adenine-thymine pairs. This could cause a

preferential binding of cations such as Mg at regions that are


10This interaction differs from that of Mg or Ba IRC-50 with
histones (22) where the negatively charged carboxyl groups of the res-
in and the protonated side chains of these basic proteins are probably
Joined by electrostatic linkages. During the adsorption of histones,
the cations originally present are presumably exchanged for the posi-
tively charged groups of lysine and arginine.









32

abundant in guanine and cytosine. Such regions would thereby become

less negatively charged, and consequently would exhibit less affinity

for the resin than regions not bearing a Mg Similar considerations

may underlie the specific elution of deoxyribonucleate from other cat-

ionic adsorbants by sodium salts (5, 8) since Na+ may also be bound by

deoxyribonucleate (34, 35). It is worth noting in this connection that

low concentrations of MgC12 markedly reduced the concentration of NaCl

required to elute deoxyribonucleate from histone-chloroform gels (5).

Flow rate -- For optimum resolution during the analysis of

small organic compounds by ion exchange chromatography, the rate of

elution must be commensurate with continuous equilibration between the

adsorbed solute and the solute free in solution. However, most macro-

molecular substances are usually completely adsorbed to, or desorbed

from, the chromatographic adsorbent. That is, the range of molarities

(or pH, etc.) over which polymers will exhibit a finite distribution

between the free and bound forms is small, and frequently the Rf of the

compound will change abruptly from 0 to 1 (29). The suddenness of this

change is suggestive of a cooperative phenomenon, and may in fact be

viewed as deriving from the need to sever simultaneously the many bonds

between the adsorbent and the multivalent macromolecule in order for

the solute to be eluted.11


11Because of this "all or none" phenomenon, the chromatographic
behavior of many macromolecules is frequently independent of column
length (29). This was found to be the case when deoxyribonucleate was
chromatographed by use of a 72 x 0.9 cm column instead of the usual
30 x 0.9 cm column. No significant change in the chromatographic pro-.
file was observed except a slight, unexpected sharpening of the pattern.









33

Experiments were performed in order to determine the effect of

flow rate on the behavior of deoxyribonucleate when chromatographed by

use of Mg IRC-50. When deoxyribonucleate from calf thymus was chroma-

tographed at flow rates varying from 4 ml per hour to 26 ml per hour,

small but significant changes in the elution patterns were observed.

At the lowest rate, there occurred a continuous elution of small

amounts of ultraviolet absorbing material before the main portion of the

nucleate emerged. However, elution of the bulk of the nucleate always

required at least 0.2 M Mg(OAc)2. If, therefore, the material leaching

slowly from the column was deoxyribonucleate, it would represent a

small amount of material in equilibrium with the bound nucleate at con-

centrations of Mg(OAc)2 below 0.2 M. The effects of this equilibrium

would be observable only at very low rates of flow.

In a subsequent section will be discussed the observation that

a small amount of ultraviolet adsorbing material referred to as "heat

peak" can be recovered from columns, after the greatest part of the

deoxyribonucleate has been eluted, by raising the temperature of the

column from 300 to 450 or 500 C, and simultaneously changing the

eluent to 1 M Mg(OAc)2. After the column had been operated at rates of

20 to 40 ml per hour, about 2 to 37. of deoxyribonucleate from calf

thymus was recovered in this manner. However, at lower rates of flow,

more material was recovered in the "heat peak," 4% at 7 ml per hour,

and 9% at 4 ml per hour.

To determine if the length of time during which the sample

remains adsorbed to the resin results in any physical or chemical

change in the nucleate which might affect its chromatographic properties,









34

2.4 mg of the specimen from calf #32 thymus van applied to a column

which was then held thirty-six hours before administration of the elu-

ent. This delay had no apparent effect on the chromatographic pattern,

except that 3.7% of the recovered ultraviolet absorption was eluted

within one hold-up volume. It is difficult to assess the significance

of this observation since, occasionally, small amounts of ultraviolet

absorbing materials appear in this region of the chromatogram in the

absence of any known modifications with respect to the sample or to the

chromatographic procedure.

Particle size of resin -- When two chromatograms of deoxyribo-

nucleate from calf thymus were run under conditions which were identical

in all respects with the exception that one column was prepared from

resin which had passed a 200 mesh screen, while the other was prepared

from "through 325" mesh material, quite similar chromatographic pro-

files were obtained. It is possible that the higher proportion of

small particles in the "through 325" material presents a greater effec-

tive surface area resulting in a higher capacity for the nucleate.


Chromatographic behavior of deoxyribonucleates from rat liver, Ehrlich

ascites tumor cells, and Pneumococcus

The essential reproducibility of the chromatographic patterns

given by specimens of deoxyribonucleate prepared from calf thymus by the

several procedures noted in the experimental section has frequently been

verified. It was of interest to determine if specimens from other

sources likewise are adsorbed by Mg IRC-50 at low concentrations of

Mg(OAc)2, and if so, whether they can be eluted. The deoxyribonucleates









35

of two other mammalian sources were studied. Figs. 8 and 10 show that

specimens of deoxyribonucleate from rat liver and Ehrlich ascites tumor

cells are, in fact, bound by Mg IRC-50 and can subsequently be eluted

with an eluent of gradually increasing Mg(OAc)2 concentration to pro-

vide chromatographic profiles very much like those obtained from calf

thymus nucleate. The recovery of these three deoxyribonucleates, based

on ultraviolet absorption, has usually been between 80% and 1007.. When

very small samples of deoxyribonucleate from rat liver were chromato-

graphed by use of 30 x 0.9 cm columns, the recoveries were lower (62%

and 757.).

A somewhat different result was obtained when the chromato-

graphic behavior of the deoxyribonucleate from Pneumococcus was
12
studied. In the initial experiment, specimen 149 was chromatographed

on a preparative column of Mg IRC-50 (Fig. 6A). This specimen possessed

transforming activity and was estimated to contain 267. ribonucleate and

29% protein. Of the total ultraviolet absorption applied to the

column, 35% was unretarded at 0.05 M Mg(OAc)2 and only 10% was eluted

by means of the eluent of gradually increasing Mg(OAc)2 concentration.

It was found, however, that 10% of the ultraviolet absorption of the

sample could be recovered from the resin in the "heat peak" by abruptly


12We wish to express our appreciation to Dr. R. D. Hotchkiss
for supplying us with several specimens of deoxyribonucleate from this
source. The specimens were prepared in his laboratory by lysis of the
bacterial cells with 0.15% deoxycholate, followed by several treat-
ments with chloroform and isoamyl alcohol, as well as with ribonuclease.

13Performed by procedures to which reference is made in the
manuscript noted in Footnote 1.















\(L52 0\Q 046 .
03- MglRC-50 (THRU 200), 29X2 CM A

2 K1.5MG DNA(149)
0.2



FRACTION NO 40 60 140 160 60 200 220 240 26 300 320 340
0 (5ML PER FRACTION)
---- Mg(OAc)2 [O.05M-<1000 ML)--0.4M -- I I.OM], 30 -IOM Mg(Ac),, 45



M MglRC-50 (THRU 200, 30X0.9 CM B
E 0.2-
D-2.2 MGDNA(149R)
o
0.1 I -


j FRACTION NO 20 40 120 140 160 180
o (3.2ML PER FRACTION)
S( 1- FRACTK-N) Mg(OAc)2 Q05M--500 ML)--0.4M], 30* I-LOM Mg(OAc),50"


MglRC-50(THRU 200), 32XO.9CM C
Q2 -1.9MG DNA(151)

0.1-

FRACTION NO. 40 140 160 180 200 220 240 260 280
(2 ML PER FRACTION)
Mg(QAc)2 [0.05 M-(500 ML)--0.4 M] 30 [-.0 M Mg(OAc)2, 50-


Fig. 6. Chromatography of specimens of deoxyribonucleate from
Pneumococcus by use of columns of Mg IRC-50. Since slightly different
conditions of elution were used for each experiment, the chromatograms
have been plotted on different scales in order to facilitate a compari-
son of the relative distribution of ultraviolet absorbing materials.
Specimens 149, 149R and 151 were used to obtain (A), (B) and (C),
respectively.











increasing the temperature of the water jacketed column from 300 to

45 C and simultaneously applying 1 M Mg(OAc)2. The analyses presented

in Table IV suggest that the bulk of the contaminating ribonucleate was

contained in material which passed directly through the resin (Fraction

II), whereas protein was not confined to any single fraction. Results

communicated to us by Dr. R. D. Hotchkiss indicate that while Fractions

IVa, IVb, IVc, and V exhibit transforming activity, the unretarded

material was devoid of such activity. Further purification of this

same preparation by two treatments with chloroform and isoamyl alcohol

yielded a product (149R) which on chromatography by use of an analytical

column gave practically no unretarded constituents which absorbed ultra-

violet light (Fig. 6B). The distribution of eluted material in the

remaining portions of the chromatogram, however, remained unchanged.

That is, both before and after repurification, 50% of the eluted

deoxyribonucleate was found in the gradient elution region of the

chromatogram, and 50% in the "heat peak" (Table V). Two additional

pneumococcal specimens (152 and 152A), derived from the same original

preparation, differed in their physical character when precipitated

from solution in 0.85% NaCl by an equal volume of ethanol. Specimen 152

was fibrous, while 152A was jelly-like. When each of the samples was

chromatographed, 36% of the total absorbancy eluted, exclusive of

unretarded material, was present in "heat peak." Another preparation

(151), however, yielded a chromatographic pattern which, as shown in

Fig. 6C, was similar to that obtained from the mammalian nucleates,

except for the presence of unretarded material.















TABLE IV
Distribution of constituents among chromatographic
fractions of deoxyribonucleate from Pneumococcus

The sample chromatographed contained 130 absorbency units, or
about 10.6 mg of deoxyribonucleate. It gave orcinol color equivalent
to approximately 2.6 mg of ribonucleate, and gave ninhydrin color
equivalent to 2.9 mg of histone Fraction B from calf thymus. The
orcinol and ninhydrin reactions were performed by procedures referred
to in the manuscript noted in Footnote 1. The results are expressed
as per cent of the constituent in the sample recovered in the fractions
shown.

Fraction number
Constituent
II V

Absorbency units 21 10

Ribonucleate 51 4

Protein 20 24















TABLE V
Recovery of "heat peak" material from deoxyribonucleate
preparations from various sources


Preparation Per cent of added absorbancy Total
Source number units recovered in "heat peak" recolumveya


Calf thymus 32 3 86
32 3 96

Rat liver 11 5 92
11 7 62

Ehrlich ascites
tumor cells 7 3 88
9 2 97
16 3 105

Pneumococcusb 149 50 56
149R 49 24
151 6 67
152 36 27
152A 32 33

aAbsorbancy units recovered as per cent of absorbancy units


added.


bBecause of wide variation of ribonucleate content and total
recovery of absorbancy encountered with samples from Pneumococcus, the
per cent recovered in "heat peak" is expressed, for these specimens, as
per cent of the total absorbancy units eluted, exclusive of that which
was unretarded on Mg IRC-50 at 0.05 M Mg(OAc)2, pH 7.2 to 7.4.









40

As shown in Table V, 2 to 37. of the nucleate from calf thymus

and ascites tumor cells, and 5 to 77 of that from rat liver is eluted

in the "heat peak" position. Further studies should clarify whether

the unusual chromatographic behavior of the pneumococcal deoxyribonu-

cleates can be attributed to the use of preparative procedures very

different from those employed in the isolation and purification of the

nucleates from the other sources studied. In this connection, however,

a specimen of calf thymus deoxyribonucleate which was prepared in the

laboratory of Dr. R. D. Hotchkiss by two deproteinizations with chloro-

form and two treatments with Duponol, yielded a chromatographic pattern

in all ways like that given by the specimens of deoxyribonucleate from

calf #30 and #32 thymus glands. Additional evidence that "heat peak"

material is not entirely an artifactitious product arising by action of

the resin on the nucleate is provided by the observation that such

material was absent from rechromatographed fractions of a specimen from

rat liver (see Fig. 8B).14 It must be recalled, however, that the

amount of material in the "heat peak" increased simply by running the

column at a very low rate of flow. Moreover, rechromatography of "heat

peak" material from the nucleate of ascites tumor cells proved disap-

pointing, since when a sample of 0.1 mg was applied to a fresh column


141n one experiment, two fractions isolated by ethanol precipi-
tation from an "overload" chromatogram of deoxyribonucleate from
Ehrlich ascites tumor cells were combined and rechromatographed under
non-overload conditions. One of the fractions contained material that
originally emerged unretarded, while the other fraction contained
material that was originally eluted gradually between 0.2 M and 0.4 M
Mg(OAc)2. On rechromatography, very little material emerged unretarded.
However, about as much material was eluted in the "heat peak" as was
eluted during gradient elution.









41

of Mg IRC-50, nothing was eluted by the original conditions, and less

than 107. was eluted by 1 M NaCl at 45 C. In this experiment, a sig-

nificant amount of material was recovered only by passing 0.5 M NaOH

through the column. Thus, while a part of the material in "heat peak"

may be a genuine component of the nucleate specimen, a part may also be

an artifact. In a subsequent section, which is concerned with studies

of the incorporation of radioactive precursors into the deoxyribonu-

cleate of Ehrlich ascites tumor cells, further observations regarding

"heat peak" and its possible biological significance will be presented.

The "heat peak" material described here may be related to the

fractions of deoxyribonucleate obtained from various sources by Brown

and Brown (8). The fractions comprised 2 to 10% of the preparations,

and were removed from columns of histone-cellulose by an eluent of high

molarity and pH. These authors concluded that the material, which had

a base composition similar to the original deoxyribonucleate, was an

artifact produced by the column.

A characteristic feature of the chromatography of deoxyribonu-

cleate from Pneumococcus is the low recovery (24 to 677.) of ultra-

violet absorbing material from columns of Mg IRC-50.


Rechromatography of fractions of deoxyribonucleate from calf thymus and

rat liver

In order to study factors which determine the points of elution

of molecules of deoxyribonucleate, material eluted at different regions

of the Mg(OAc)2 concentration gradient was adjusted to 0.05 M Mg(OAc)2$

combined, and rechromatographed on a fresh column of Mg IRC-50. Fig. 7B









42
shows an analytical chromatogram of two fractions of deoxyribonucleate

which had been eluted from the preparative column illustrated in Fig. 7A

and recovered by precipitation by ethanol. Similar experiments in

which analytical columns were used both for isolation of fractions of

deoxyribonucleate from rat liver and for their rechromatography are

shown in Figs. 8A and 8B. In this case, the fractions had not been

recovered for rechromatography by precipitation with ethanol, but were

adjusted to 0.05 M Mg(OAc)2 by dialysis. The results of the rechroma-

tography experiments indicate that although there is some displacement

in absolute position, the relative positions of elution are inherent

properties characteristic of the molecules of deoxyribonucleate. That

is, the gradual elution of successive fractions appears not to be

governed by a heterogeneity of binding sites in the resin, or by a

reduction in the capacity of the adsorbent to which all molecules of

deoxyribonucleate are equally sensitive. It would be desirable to ex-

tend these experiments by rechromatographing mixtures of fractions

isolated from radioactive and from non-radioactive specimens of deoxy-

ribonucleate.


Effect of sample load

Although the relative positions of elution of the rechromato-

graphed fractions of deoxyribonucleate from rat liver are similar to

those of the original material (Fig. 8), the shift in absolute positions

indicates that competition and displacement effects play a role in the

fractionation process. The molecules in Fraction III of Fig. 8A must

have the greatest affinity for the resin, since they require the



























0
0
ro





-J


0


0 fl
UI



cin
o





0

CD <
0
c'a
0--3


O


(f'll 06' SnNIIV rMl 09 ) A0ONV88OS9V


4
"0
ott'

40


'H Q
OMA
o- *-
4M
10
G~l


04

0 41
a 0w
"4 U- 41



14j 000
0 H
0 *








.244 0 to
V4 U
A 41
u *1

4 1 6 1
| a











V 00
000.DO8
o 9.4
M1 o0












A"40
101
0 to % m
0 ^
e 44 *






0 o











148
4 C O
N *g

0 **

M M 0


*,08


0 *

.K alo

,{: n0 rM
MMS^
: 4









44



; ,-iv
.. *4

0







0 0
0






"%4J
V4
{ 0B
S..,4 ',







8 8 -



4 2


00
u0
So A 44
<1 10 0
-i ? S

J I I a s












S :- 8 T
(7MO62 SNIIN 77N09) I)
AONVSHOSSV









45

highest salt concentration for their elution. Molecules having a

strong affinity for Mg IRC-50 may be able to compete effectively for

binding sites on the resin, and thereby to displace the less strongly

bound molecules to positions further down the column. The positions of

elution of different molecules of deoxyribonucleate may therefore be

related to their initial positions on the column. When large samples

are chromatographed, deoxyribonucleate may be adsorbed to the resin

sites throughout most of the length of the column. Under such condi-

tions, material might be expected to emerge earlier than when smaller

samples are chromatographed. For example, in Fig. 8C, where 0.43 mg of

deoxyribonucleate from rat #11 liver was chromatographed, elution did

not begin until a higher Mg(OAc)2 molarity was reached than in the

experiment shown in Fig. 8A, where a 2 mg sample of the same specimen

of deoxyribonucleate was chromatographed. Whether the concentration of

deoxyribonucleate in the sample added to the column has an effect on

the elution pattern has not been studied systematically. It must be

noted that the sample solution used in Fig. 8A (1 mg per ml) was more

concentrated than that used in Fig. 8C (0.2 mg per ml).

The possibility that deoxyribonucleate molecules having a low

affinity for the resin are displaced by more strongly bound molecules

was tested further as follows. Calf #32 thymus nucleate was added to

columns of Mg IRC-50 in amounts sufficient to exceed the capacity of the

resin. In the two experiments presented in Table VI, 62% and 367. of the

specimens added were eluted under the starting conditions (0.05 M

Mg(OAc)2, pH 7.2). As shown, the material which emerged abruptly

(Fractions I, II, and III of experiment 743 and Fractions I and II of















TABLE VI
Base composition of chromatographic fractions of deoxy-
ribonucleate from calf #32 thymus obtained by overloading
columns of Mg IRC-50

To the 28 x 2 cm column used for experiment 743 was added 74 mg
of deoxyribonucleate, an amount which exceeded the capacity of the res-
in by 62%. Experiment 766 was performed by use of a 31 x 0.9 cm column
to which was added 22 mg of deoxyribonucleate, an amount which exceeded
the capacity of the resin by 367..

Experiment number
Description of fractions
743 766


Fraction number I II III IV I II III

Per cent of deoxyribonu- 10 11 8 28 18 18 52
cleate in fraction

Molarity of Mg(OAc)2 0.05 0.05 0.05 0.05 0.05 0.05 0.4
used for elution

Basesa

Thymine 25.2 25.9 -- 27.3 25.7 27.0 28.7
Guanine 25.6 24.5 -- 23.2 23.5 23.1 20.4
Cytosine 22.4 22.0 -- 22.4 23.5 21.8 20.7
Adenine 24.5 26.2 -- 25.5 25.1 26.1 29.2
5-Methylcytosine 2.5 1.5 -- 1.6 2.2 2.0 0.9

aAs moles of base per 100 moles of total recovered bases.









47

experiment 766) contained more guanine and cytosine than total deoxyri-

bonucleate (see Tables II and VII), and thereby resembled deoxyribonu-

cleate which is eluted first by the lower concentrations of Mg(OAc)2

under "non-overload" conditions, such as Fraction I in Table II. On the

other hand, the material not eluted by 0.05 M Mg(OAc)2 but subsequently

eluted as in experiment 766 by means of 0.4 M Mg(OAc)2 has a composition

like that of fractions eluted by the higher concentrations of Mg(OAc)2

under "non-overload" conditions, such as Fractions II and III in Table IL

It therefore seems likely that the same molecules of deoxyribonucleate

which are eluted early by increasing concentrations of Mg(OAc)2 from a

column bearing a small load of deoxyribonucleate, are among the first

to emerge when the capacity of a column is deliberately exceeded. Even

when the sample load is excessive, interaction of Mg IRC-50 with deoxy-

ribonucleate remains demonstrably specific.


Base composition of fractions of deoxyribonucleate from calf thymus and

Ehrlich ascites tumor cells

As discussed above, the interaction of deoxyribonucleate with

Mg IRC-50 has provided a means for eluting fractions of different com-

position from columns to which specimens of total deoxyribonucleate

were adsorbed. The studies with overloaded columns suggested an

equally effective alternative fractionation technique, as discussed in

connection with Table VI. A more complete chromatographic fractionation

of deoxyribonucleate from calf #32 thymus is illustrated in Fig. 9. The

base compositionsof fractions of the unadsorbed deoxyribonucleate and of

the fractions eluted gradually by increasing concentrations of Mg(OAc)2

are shown in Table VII.















* I


0
*
0 0

1S Qg



0

m m1 A +I


u +

00 4 a t
C I *





a0 00
004 U
0

A



44


0 a M 0 0






SN J 0






IU J 'a
4 93
*M "4






0U w
OMa@o
><9oM>


o o ci 3
6z SnNIVI 771V093)

A)NV8OSBV
















e4

MS|


WW 0


0u"
0
4e 0
'- 4



44
0 4J 0
U41-4.
v 0








00
^a4.
.4 A
Me



e*d






0: 5A<
'14 S
0 0

^ 9




4 x
1 04




tdo to
14 01







41~ w
440
p014




# 0 H




-14 0 &
W .













141.
d K




S4 4
41







ol i
"I .4*










0




404
*94JJ









x K
1 u
4JU .0

(4 '4 4.


41l UN.


$. 0
SO>
M43 14


41
4..
S
*u





1'4
OS

















0-4
1-4




00%
9








0-
Po











>


I
5













H






H
H
at


r4















H





H

H

M
H^
H

H^


0
N




0
N




0'.
H



H
(u.'



041


4.10
414.5

'I.'


V-4
M
N
*







N

1-4







*

in


*
P4










*
141








0
1-4
1-0
H





















.4














0
U()
*

H-


H-


H-








0
*


s0










+U.






14+

U+









50

The successive fractions show variations in composition similar

to fractions previously obtained by Chargaff, Crampton, and Lipshitz

(4, 5) and by Brown and Brown (8), whose methods were based on the

interaction of deoxyribonucleate with basic proteins. Chargaff et al.

noted that the ratio of 5-methylcytosine to cytosine in the successive

fractions varied markedly, and concluded that the replacement of cyto-

sine by the 5-methyl derivative does not occur at random. As shown in

Table VII, a 5-fold change in the 5-methylcytosine content of the

chromatographic fractions occurs with less than a doubling of cytosine

or guanine, which confirms the results of Chargaff et al.

Chromatograms of deoxyribonucleate from Khrlich ascites tumor

cells obtained under "non-overload" conditions (EA #1) or overload

conditions (EA #7) are shown in Figs. 10 and 11, respectively. The

base compositions of the isolated fractions are presented in Tables VIII

and IX. Both techniques yielded fractions which show the same trends

in composition as the fractions from calf thymus deoxyribonucleate.

However, the extreme fractions of the deoxyribonucleate from tumor

cells differ far less than the corresponding fractions shown in

Tables II, VI, and VII.


Incorporation of Precursors into Fractions of Deoxyribonucleate

from Ehrlich Ascites Tumor Cells


Preliminary experiments with C14-formate

The ability of Mg IRC-50 to separate chemically different mole-

cules of deoxyribonucleate suggested that information regarding the

synthesis of these molecules might be provided by studies of the initial




















3
.8


















X
0
-4

















O
u
U







,.
S
0
'u
"4


U
U






0






U










O









'-
'o
0
J3.







14















.1


0
0
"4
U0
"4


4.
0^


mow nxw )aw
A39X3SBVW






































































U-,




v(wO6Z SnNIIN /7 O9Z)
AONV8uOSSv


I A
-- ~ ~ -- *
v-4 0- 1"

& ~ 29 aS


0M 3



40



T u >%41c
Kf- So 0 U


4 U 41 24
6a an J

S'*
S



041 0.
Q u >>4J a
od aa
Ai ft4 ofi^ e



00 % 4< i


i 1. i
liP








1sJ1
S44 Ve 06
QM
.S a








4 1 .4
S40





04* 1.4








ba 04
09
S0;
rV f t3
v 10 0 a^








sb w
.-.M 0 J3
rO 0 "0
M a
S 'M -) 0 4



04 B

r "0 4


ck. o *< *
ff) n u u
<0 u O S


f W 3."2
Q O B




V-4
I r *SO f
,- t4 U .


Ir


"4M
;#4
0 0s
D Q a B u
IA. 0 "4*

944 -.4.
ons 0- &














TABLE VIII
Base composition of chromatographic fractions of
deoxyribonucleate from Ehrlich ascites tumor cells (RA #1)

The fractions were derived from the experiment illustrated in
Fig. 10. The percentage in parentheses which follows each fraction
number represents the portion of the deoxyribonucleate of the sample
recovered in the respective fraction. The results are expressed as
moles of base per 100 moles of total recovered bases.

Fraction number Unfractionated,
Base total
1(19%) 11(24%) 111(34%) deoxyribonucleate

Thymine 27.7 28.1 29.2 28.8

Guanine 22.2 21.1 20.6 21.3

Cytosine 21.0 21.7 19.6 19.5

Adenine 28.8 28.8 30.0 29.5

5-Methylcytosine (0.4) (0.3) 0.7 0.9

A + Ta
G + CHC 1.30 1.32 1.45 1.40


aThe ratio of the molar quantities of adenine plus thymine to
guanine plus cytosine plus 5-methylcytosine.















*
14





14 4fl 0
74
0.8
400
P -4
0044

.0 4.
* 00
"4.






0







"4,
4.14
a 44









1-4
MM*








4-0
*4o n








S60
IiS
0
























"4,
* 0












'44 '44
00 0
0a










u o44
Mae 0
4 WO










U U
j.0


Me1
8. 8 4J












*U 0-I
M 3.











* 0.44
We
*r*





0 44


60
u4 e
Mo *



MOd
o ^


> *U *
QU au


4 0 M


0. e p


4 "4
*







S
0
"4
P
>

0 P
"4 4J

C tt


**'+
y


+ v
4 +
0









55

in vivo rates of incorporation of radioactive precursors. Cultures of

Ehrlich ascites tumor cells growing in the peritoneal cavities of mice

were chosen for these studies because they present the following ad-

vantages. 1) During the logarithmic phase of growth under natural in

vivo conditions, the cells are less likely to lose viability than cells

grown in vitro. Consequently, the possibility of an abnormal metabolism

of their deoxyribonucleate is minimized and most findings can be con-

strued with reasonable confidence to have a bearing on the synthesis of

authentic deoxyribonucleate. 2) Populations of ascites tumor cells are

relatively homogeneous in contrast to alternative biological systems in

which there is rapid synthesis of deoxyribonucleate such as regenerating

liver. The interpretation of any chemical or metabolic properties of

fractions of the deoxyribonucleate of liver would be confused by the

presence of several cell types which respond differently to hepatectomy,

as was recently found by Edwards et al. (36). To be sure, in vivo

cultures of Ehrlich ascites tumor cells gradually become hemorrhagic.

Although absent from erythrocytes, deoxyribonucleate is present in

leucocytes, which may also infiltrate the cultures, but which rarely,

if ever, exceed about 1% of the cell population. 3) Finally, of great

practical importance, is the fact that deoxyribonucleate is readily

obtained from these cells in good yield and in a highly polymerized

form by use of strong urea solutions.1 The denatured deoxyribonucleate

provided by certain alternative extraction procedures (37) would not be

suitable for fractionation by use of Mg IRC-50, although they could con-

tain constituents absent from the specimens examined in the present

studies.









56

Effect of non-radioactive format -- After the administration

of C 4-formate, it was necessary to allow time enough for a measurable

amount of radioactivity to be incorporated into deoxyribonucleate. It

was desired, however, to limit the duration of the incorporation period,

so that only a small proportion of the total deoxyribonucleate would

have the opportunity to replicate in the presence of the precursor. A

brief period would tend to exaggerate any preferential uptake of the

precursor into chromatographically different fractions. An incorpora-

tion period of ten minutes was chosen for the first experiment. This

corresponds to about 1% of the time required for the cells to double in

number (eighteen hours). The C -formate injected is an efficient

precursor of the 2 and 8 positions of the purine ring, and of the

5-methyl group of thymine. With these considerations in mind, the ten

minute incorporation period was followed by an interval of 110 minutes

in order to allow time for incipiently Labeled molecules of deoxyribo-

nucleate to be completed, to undergo secondary structural alterations,

to combine with histone, or to acquire whatever other characteristics

may be essential for insolubility in the urea medium used during iso-

lation. In an effort to minimize further incorporation during the

110 minute interval, a 100-fold excess of unlabeled format was injected

ten minutes after the C 4-formate. It was anticipated that this amount

of unlabeled format would dilute the labeled pools of the cells so

that the radioactivity incorporated during the 110 minute interval

would be insignificant compared with that incorporated in the preceding

ten minutes.









57

Accordingly, each of two tumor-bearing mice (EA #1) was injected

with 0.1 ml of buffered saline solution containing 5 pcuries of C 4-for-

mate having a specific activity of 1 curie per p mole. The mice were

reinjected after ten minutes with 1 ml of buffered saline solution

containing 500 p.moles of unlabeled format. The tumor cells were har-

vested 120 minutes after the injection of C -formate, and were proc-

essed as described in the experimental section. The twice-deproteinized

deoxyribonucleate was found to contain appreciable amounts of carbon-14

in thymine, guanine, and adenine. Cytosine contained no appreciable

radioactivity, while 5-methylcytosine was found to contain some counts,

which, however, could have resulted from contamination by the highly

labeled adenine as a result of the incomplete separation of these two

bases (see Fig. 1). Therefore, it is still not known whether format

is a direct precursor of the methyl group of 5-methylcytosine. A 3.9 mg

specimen of the deoxyribonucleate was eluted from an analytical column

of Mg IRC-50 (30 x 0.9 cm) by use of an eluent of gradually increasing

Mg(OAc)2 concentration (0.05 M to 0.4 M) to obtain the chromatogram

shown in Fig. 10. The effluent was combined to provide three succes-

sive fractions from which the nucleate was precipitated by ethanol and

hydrolyzed. The hydrolysates were analyzed by use of columns of

Dowex 50 in order to estimate the base composition of the fractions and

to separate the bases for the subsequent determination of specific

activities. The base compositions have already been presented in

Table VIII. Specific activities are given in Table X. Under the con-

ditions of incorporation used in this experiment, it is apparent that

all of the fractions attained similar specific activities.















TABLE X
In vivo incorporation of C14-formate into fractions of
deoxyribonucleate from Ihrlich ascites tumor cells (EA #1)

The fractions were derived from the experiment illustrated in
Fig. 10. The percentage in parentheses which follow each fraction
number represents the portion of the deoxyribonucleate of the sample
recovered in the respective fraction. The results are expressed as
counts per minute per ipmole of base.

Fraction number Unfractionated,
Base total
1(19%) 11(24%) 111(34%) deoxyribonucleate

Thymine 372 400 404 422

Guanine 286 236 281 256

Adenine 226 191 195 204









59

Effect of time -- In view of the results of the preceding ex-

periment, three additional experiments were performed in order to

evaluate the efficiency with which the 500 moles of unlabeled format

prevent further incorporation of C -formate into the tumor cell

nucleate. In all three experiments, the results of which are presented

in Fig. 12, an interval of seven minutes was allowed to elapse after

C 4-formate. However, the period of exposure to unlabeled format was

varied. In the first experimental group of three mice (EA #4), this

period was 113 minutes. That is, 500 Amoles of unlabeled format were

injected to each mouse seven minutes after 54curies of C 4-formate had

been injected, and the tumor cells were isolated 120 minutes after the

injection of C 14-formate. In the second group of four mice (IA #5),

C14-formate and the unlabeled compound were injected as for EA #4, but

the cells were isolated twenty-five minutes after the injection of

C 4-formate. In the third group of three mice (EA #6), unlabeled for-

mate was not injected at all. That is, the tumor cells were isolated

promptly seven minutes after injection of C 14-formate. If the admin-

istration of unlabeled format completely and immediately prevents

further uptake of the labeled precursor, the specific activity of the

bases in the deoxyribonucleate isolated from all three groups of mice

would be the same or possibly would decrease with time because of the

continued synthesis of deoxyribonucleate from diluted pools of pre-

cursors. As shown in Fig. 12, however, the incorporation of radio-

activity into the purines continued slowly for 120 minutes despite the

injection of a 100-fold molar excess of unlabeled format. On the

other hand, the specific activity of thymine did remain the same over









60


o
0
--------___ ------------- .'3
*6
o A:
0 0
U


u .


90 o.4


O Jo


0 0 r 0



< 0J
C 2 1

0 z Z
\ \ z rr )5
\ \ 0
\- \

"O o
0


z \.4










o o o o o:
() 0 0'4J- o
\0 \ r4 -0 o

49O
44








-834









61

the intervals studied, suggesting that the non-radioactive format in-

jected did succeed in diluting pools of precursors of this base. The

differences among the curves obtained for the purines and for thymine

could result from differences in the size of the pools of the respec-

tive precursors. Thus, the purines may be formed from large pools

with which format equilibrates slowly and incompletely, while thymine

may be formed from small pools with which format equilibrates rapidly

and more completely.


Incorporation of C 4-formate into fractions of deoxyribonucleate

As noted above, deoxyribonucleate containing reasonably large

amounts of radioactive thymine is recoverable from tumor cells seven

minutes after the injection of C 4-formate. Since deproteinized deoxy-

ribonucleate is soluble in the urea medium, this observation suggests

the possibility that deoxyribonucleate may be insoluble in the urea

medium even during the initial stages of replication. This observa-

tion is perhaps related to the recent finding that histones are con-

served as well as deoxyribonucleate during replication of the tumor

cells (38). In this system, therefore, deoxyribonucleate at all stages

of replication may be insoluble in the urea medium because it occurs

combined with histone. At any rate, the finding indicated that

studies of the heterogeneity in the uptake of precursors into fractions

of deoxyribonucleate during even shorter intervals were feasible.

Under these conditions, preferential labeling would be accentuated by

the rapidity with which the specific activity of the pools of precur-

sors of thymine changes.









62

In the next experiment, therefore, tumor cells (BA #7) were col-

lected from each of ten mice five minutes after the injection of 0.1 ml

of buffered saline containing 3.7 moles of labeled format with a

specific activity of 13.7 gcuries per pmole. A fifty-seven mg sample

of deoxyribonucleate isolated from the cells was chromatographed by use

of a preparative column of Mg IRC-50 (19 x 2 cm), with the results

shown in Fig. 11. Under these overload conditions, 327. of the sample

was eluted by 0.05 M Mg(OAc)2. Most of the remaining deoxyribonucleate

was removed from the column by gradient elution with Mg(OAc)2 (0.05 M

to 0.4 M). A further small portion, referred to as "heat peak," was

eluted by I M Mg(OAc)2 when the temperature of the column was raised to

45 C.

The column effluent was combined to provide seven fractions.

The deoxyribonucleate which was recovered from the fractions by pre-

cipitation with ethanol was found to have the base compositions shown

in Table IX. As discussed earlier, the extreme fractions differ far

less than extreme fractions which have been obtained by similar methods

from deoxyribonucleate from calf thymus.

Despite the similar base compositions, however, significant

differences were found with respect to the specific activity of the

thymine in these fractions, as shown in Table XI. The first three

fractions comprise the nucleate eluted by 0.05 M Mg(OAc)2 under the

overload conditions. The specific activity of the thymine in these

three fractions is higher than that in the next three fractions which

contain material recovered from the column by gradient elution. The

thymine present in Fraction VII, the "heat peak," was found to have the
















44


0
GD

o 0
"A
0

0M
04
4.,
GD


+1

00


.0
e
eu
bJ G41
41 0 4j

60 0 9
("4 4

B 93



4) w
US0



GD$4&
kgg





.4
00 o


M4 00
6000

40
CS -
*0 *J $4









Cis
0"
1 0 t


C.56D
0e0













)40
P4 .







'5 0-
0w 4







"w4 "4
3 .A U
GD,

Iei












*"4 4)
Sa 0.



8 0













4 0 0.
4G 0











004 G
a 0 ."
0. V4



O M :
QU 2U
g ^
SN R
4 OU


U U O )
aj -^


ico
1-4


. 0

0
". 4


'4J 4
0 0
0 $
A
j3 o
0 .fl


S '44
o
3 N
U 4


& 1
3 o
0 0
0M u


jog. 0
1 0 to
4 4
90
90



(5
GD


^00




czo







0
"4









+1
"P4
. -, c* C0 -












1. (n 0
1-1
'0
/-s *"4



^ +1




-% 00

1-4 00. 0^
"4 r. "4 V
5%


0





0

a
0







3
u
-.4






M
GD

'5










M
S





M
4
-4J
"4
0.

0
GD



4.
0


r









64

highest specific activity. While the fractions shown account for 88%

of the absorbancy applied to the column, they account only for 60% of

the radioactivity. The discrepancy would be accounted for if material

having a specific activity of about 3000 CPM/mole (a value similar to

that of Fraction VII) remained on the column.

The observed differences in specific activity could result from

contamination of the thymine by constituents arising from the destruc-

tion of deoxyribose by formic acid, or from hydrolysis to amino acids

of variable amounts of protein or other contaminants in the successive

fractions of deoxyribonucleate. Acidic decomposition products would

probably emerge near thymine from columns of Dowex 50.15 In order to

explore this possibility, the thymine residues were eluted from the

planchets with 0.1 N HCl and rechromatographed on filter paper by use

of isopropanol-HCl. The specific activities of the thymine after

rechromatography on paper, also presented in Table XI, were quite simi-

lar to those determined directly on the Dowex 50 effluent. The slight

increase observed after rechromatography may have resulted from the

elimination of materials which contributed ultraviolet absorption.5 In

order to determine whether reproducible values for specific activities

could be obtained by use of paper chromatography alone, independent

hydrolysates of aliquots of Fractions II and V, and of the total deoxy-

ribonucleate were chromatographed directly, and analyzed as before. As

shown in Table XI, the duplicate determinations agreed within less than


15The sulfuric acid in which the hydrolysate of the deoxyribo-
nucleate was dissolved prior to chromatography would presumably be
excluded from the polysulfonate resin; in fact, it was found to be
eluted in about 24 ml, while thymine emerged after about 60 ml.









65

2.5%, and were similar to the values originally found. Appropriate

analyses of other regions of such paper chromatograms provided values

for the specific activities of adenine and guanine of some of the frac-

tions. As shown in Table XII, the specific activity of guanine was

found to vary among the fractions in the same way as that of thymine.

The evidence strongly suggests that there are present in the

specimen of deoxyribonucleate described above, molecules which vary not

only in base composition but also in the extent to which they incorpo-

rated radioactivity from C -formate during the five minute in vivo

incubation period. That such variations in the extent of labeling can

be obtained reproducibly was confirmed by isolating deoxyribonucleate

from tumor cells removed from each of five of a second group of mice

(EA #9) five minutes after the injection of 50 curies of C -formate.

A solution containing 12.6 mg of the specimen was added to an analytical

column of Mg IRC-50 (28 x 0.9 cm). Under these overload conditions,

457. of the sample was eluted from the resin by 0.05 M Mg(OAc)2. As

shown in Fig. 13, stepvise changes of the eluent provided 477. of the

sample at 0.4 M Mg(OAc)2, and 2% at 1 M Mg(OAc)2 when the temperature

was raised to 500 C. A small amount of additional material was eluted

finally by 0.5 M NaOH at room temperature. The specific activities of

the thymine of the four fractions were determined after paper chroma-

tography and are shown in Table XIII. The pattern of differential

labeling is very similar to that found in the previous experiment,

despite the examination of fewer, larger fractions in the present case.















TABLE XII
The specific activity of bases in fractions of
deoxyribonucleate from Ehrlich ascites tumor cells (IA #7)

The fractions were derived from the chromatogram illustrated in
Fig. 11. To obtain the results shown here, the bases were isolated
from hydrolysates by chromatography on paper, either directly, or after
initial purification by use of columns of Dowex 50.

Specific activity (CPM/Lole)
Fraction
Adenine Guanine Thymine

1 18 46 818

V -- 12 388

VII -- 202 2960

Unfractionated, total
deoxyribonucleate 28 63 938









67













0.60

o | so MgIRC-50 (THRU 200)Q9X28CM
S0.40 L 126 MG DNA
zu
w z
it 0.30 -
0
0.20 -

0,0 -.


TUBE NUMBER (5ML PER TUBE) 50 100
S 0SGM Mg(OAc. 3C oM MgOAc)s.30-C IMM(A MIj O SMNto,24C


Fig. 13. Stepvise elution from Mg IRC-50 of deoxyribonucleate
from Ehrlich ascites tumor cells (KA #9).











68


4.4
0 9

>>


4 +1
4.4 SS o N N a 0 0 0D
0 ON



Sto


0 4 9 0

A C .01
.4 0
$4.





U oU




0 4 41 f


0 x 0
a* *











"04 0 0
:4 3 a h
; 4 %4 4





s 0*
A1 4.1 400 in"
Uu
g ^ "4






0 e
4 9 u4 o w



S m **









93 0 Z kn
66 33

















%4O
"41
5.0 0 0"4
0 00%
*.0 0J -


M. 54 "0 0 03









0 to'
A1 *W v
1.4 S 0S 4 4.1 *
41" 0 e'S 0 U ^ 6
1. 0' u~ fli t0 > 4 0 0

Q"4 "tj "






4"4 0
00 6 "4 Z^ 0 S
"4* S 60
0.0 H 0 .








ug 0 4 "4 0
-41 Hg H -4







00 H
* U ..U U
H 0 "



"4 "4 B'
4 "4 "4 "4









69

Incorporation of H 3-thymidine into fractions of deoxyribonucleate

In vivo study of brief duration -- The possibility that the

differential labeling of the thymine discussed above results from a

methyl group exchange involving intact deoxyribonucleate molecules was

excluded by an experiment in which the incorporation of H -thymidine

was studied. Cells were removed from each of seven mice five minutes

after the injection of 10 pcuries of H 3-thymidine having a specific

activity of 360 pcuries per mole. A 13.1 mg sample of the deoxyribo-

nucleate isolated from the cells (EA #16) was chromatographed on a

28 x 0.9 cm column of Mg IRC-50 as in the previous experiment, except

that 1 M NaCl at 500, rather than 1 M Mg(OAc)2, was used to elute Frac-

tion III. This modification minimized the precipitation of Mg(OH)2

when 0.5 M NaOH was passed through the column. Because of its asym-

metry, Fraction IV was recovered in two portions, IVa and IVb. Thymine

was isolated from hydrolysates of the five fractions by paper chroma-

tography and found to have the specific activities given in Table XIV.

Although an entirely different precursor was used, the extent of incor-

poration into chromatographic fractions varied in the same manner as

was found originally with C 4-formate. That is, the specific activity

of thymine in the fraction eluted by 0.4 M Mg(OAc)2 was less than that

found in the fractions eluted by 0.05 M Mg(OAc)2 or by the more vigor-

ous conditions.

These data suggest that the observed heterogeneity of incorpo-

ration occurs at points which lie beyond the completion of the nitroge-

nous base in the sequence of reactions leading to the synthesis of

deoxyribonucleate. The evidence does not exclude the possibility of an















144 .
0 V4
s|


S +1 +1 +1
"CA
0 N NO E 0 4
(4 pin Ef -Q N 0' '0
0 N M4 0 0 N N
444
0 "
o M


44a
0'oF
a ?* .
o AM





0 0
*00





P4
S{ S'S






Ma
0 1 0







W4.
*00
e M <(
we QMO 00 0
*0 U 4 M S 0





,40 00


a ex
41 Q V-4

MSa


-4 "40
w ai
0 (0 0














In '
a a 4




04 4^ M 3n








0 00



0.P4
00
a to g on













31- 4448
*< 00* o
5 Qa U O o













1-4 #4 a 4 3
H W H H :
OM B MN -- 3' '
0 4 44 0 M Ot ***

6M o < u o an
Q u *n 0 K B


S S g o o
*^ D4M


>~ 44 *
M e
li o
3 "13
a *a d
I -3 Q
u u

a 0 4
> u a 4
*S ^ -FD









71

exchange of thymine residues with otherwise intact molecules. However,

as was shown in Table XII, the incorporation of C -formate into

adenine and guanine paralleled the incorporation into thymine. This

suggests that if exchange occurs, it would involve all bases, and would

therefore be tantamount to overall synthesis. It is likely, therefore,

that the observed heterogeneity of incorporation of radioactivity is a

property associated with the biosynthesis of whole molecules of deoxy-

ribonucleate.

In vivo study of long duration -- As discussed above, differ-

ential labeling among chromatographically purified fractions of deoxy-

ribonucleate was reproducibly observed when tumor cells were harvested

five to seven minutes after the injection of C14-formate or H 3-thymi-

dine. The specific activities of thymine in fractions of deoxyribonu-

cleate from tumor cells harvested from each of five mice (EA #17)

twenty-three hours after injection with 10 4curies of the H 3-thymidine

are given in Table XV. The specific activity of the total deoxyribo-

nucleate was only about 2.7 times that found five minutes after the

injection of H 3-thymidine (EA #16, Table XIV). Since the generation

time of the tumor cells is eighteen hours, the deoxyribonucleate iso-

lated twenty-three hours after injection would be expected to have a

specific activity about one-half as great as that isolated five hours

after injection. If it is assumed that the specific activity was

close to maximal five hours after injection, then the amount of radio-

activity that is incorporated into deoxyribonucleate during the first

five minutes after injection is about one-fifth of the maximum amount

that can be incorporated under these experimental conditions. Evidently,












72


U14
0 04
> a

4 0 0% P-
.4 U.)c
S +1 +l +

4* CO r CO c in
44 ,.4
1 "4 E o r o r
0 u
"4.
0 to
JJ P. u
U CO

toi

044
0
S4 hw


.5.

e *In 0
Fv 4
A U s 4 V
9 V 'O
pq




4 %4
*^ S 0 44
*4



1O U -

'J SS
a a
0
A U 4.'
0 8r fl y

^ 50 0 .

V4 ft Q
u M -^ o o


n ~ c N "
O 3 e0 0 0
.5.4- 0.4 a
0A i n 0

"4 Q inh I
$dbl I

0O4 0M 0 0 SP
G 0 0O46 ) -4*J
S41 ) l 0 X



3 o e6 mn *o
((3 003 0 40a 54
4 0 0.
00 00 O O..10

s8g o o -
O 4
04 x


0
0 $4

H 0
0 4


00 A



V4
$41
>00 "





0
fa c i a

M~ w4 De 0
U 0
1-4S
"4 44
1-4 "-4 H H











any thymidine which is not incorporated in deoxyribonucleate shortly

after injection rapidly becomes unavailable for incorporation.

After H 3-thymidine is injected, it is possible that the specif-

ic activities of pools of precursors at and above the level of thymi-

dine fall almost as abruptly as they rise after the injection of

C 14-formate (see Fig. 12). Nevertheless, the chromatographic fractions

provided by deoxyribonucleate isolated twenty-three hours after injec-

tion (Table XV) did not show differences in specific activities, in

contrast to the fractions provided by specimens which were isolated

five minutes after injection (Table XIV). As studied under these con-

ditions, therefore, the differential labeling of chromatographic frac-

tions is a transient phenomenon. Thus, there is no evidence that

molecules in fractions having a high initial specific activity gradu-

ally undergo discrete changes in chromatographic behavior. Instead,

the specific activity of all fractions approaches that of the total

deoxyribonucleate. That is, the specific activities of the fractions

initially less radioactive increase at the expense of the "heat peak"

fraction, the specific activity of which, however, does not fall below

that of the other fractions. The comparatively low specific activity

given in Table XV for the thymine of EA #17 total deoxyribonucleate

would indicate that the average specific activity of thymine in the

material that remained on the column was about 500 CPM/tikole.

In vitro study of brief duration -- As discussed in the pre-

ceding section, about one-fifth of the maximum possible incorporation

of thymidine occurs during the first five minutes after injection. In

an effort to reduce the extent of incorporation still further, an









74

in vitro experiment was performed. Cells were removed from the perito-

neal cavities of six mice with the aid of about 2 ml of saline per

mouse, and transferred to a glass vial at 35 o-370 C, a process which

required about sixty minutes. After adding 8 pcuries of the tritiated

thymidine, the cell suspension was incubated for two minutes with

occasional stirring. The deoxyribonucleate (IA #15) was isolated and

chromatographed to provide the data which are given in Table XVI.

Although the total amount of precursor incorporated was small, the

distribution of specific activities is similar to that seen previously

in the five minute in vivo experiments. It is difficult to compare the

extent of incorporation with previous experiments since an equivalent

of fewer curies per mouse was provided in this in vitro experiment.

In view of the loss of the capacity for synthesis of deoxyribonucleate

by cells maintained under in vitro conditions, it is possible that

incorporation occurred only in the last cells transferred to the incu-

bation vessel. Nevertheless, preferential incorporation of thymidine

into "heat peak" material was even more pronounced in this two minute

in vitro experiment than in the five minute in vivo experiments

described above. It will be of interest to study even shorter inter-

vals of incorporation.
















'04










0








M 0
.4





00

.1

*O4








0


0





-A

@0
0 *



"94


0










t4s


0
0
(M
g9\

U
4




-t
0


0 td
"4
00
4

ii




U 0

94 44
> 0
1-4 1.4 | M >


0o
/"*


0
O
M

^N
49

U

eo



0
0


.94




I
'5


0
'5
0
U
.94
"4
p.






.IZ
0
0
.94
4.5
0
.94
0















CRITICISM OF THE CHROMATOGRAPHIC STUDIES


The results of the chemical and biological studies presented

in the preceding sections emphasize the need for further information

regarding the molecular parameters which govern the specific inter-

action of deoxyribonucleate with Mg IRC-50. The chemical studies

indicate that under certain conditions Mg IRC-50 has an affinity for

all molecules of deoxyribonucleate, but that in general the affinity

of the resin for molecules rich in adenine and thymine is stronger than

for molecules rich in guanine and cytosine. Accordingly, the rank in

the hierarchy of affinity of molecules of deoxyribonucleate for

Mg IRC-50 appears to be governed to some extent by total base compo-

sition. The possibility that relatively minor chemical factors, in

addition to total base composition, may modify this affinity is illus-

trated by a consideration of the distribution of 5-methylcytosine

among the successive chromatographic fractions. While the mole per

cent of cytosine of extreme fractions changes by factors of 1.3 and

1.2 in the specimens from calf thymus and Ehrlich ascites tumor cells,

respectively, the mole per cent of 5-methylcytosine changes by factors

of 5.2 and 2.0. In time, detailed structural studies may reveal

whether the affinity of deoxyribonucleate for Mg IRC-50 is lessened by

the relative preponderance of the satellite base per se, or whether

this affinity depends on exaggerated base sequences which are correlated

with enrichment by 5-methylcytosine. During the course of this work,









77

improved chromatographic procedures were developed for the analysis of

pyrimidine nucleoside 31,5 -diphosphates, dinucleoside triphosphates,

and higher homologues present in acid (39) and diphenylamine (40)

hydrolysates of deoxyribonucleate. By providing evidence with regard

to the relative clustering of purines and pyrimidines, these techniques

may lead to a clearer definition of distinctive features of the struc-

ture of deoxyribonucleate to which the interaction with Mg IRC-50 is

sensitive. When more discriminating criteria for the characterization

of deoxyribonucleates are applied, the differences between the chroma-

tographic fractions may prove to be more profound than presently

suspected. That is, "the differentiation between nucleic acids on the

basis of differences in the contents of their several nitrogenous

constituents [has been] a comparatively crude expedient, since it does

not permit a distinction between sequence variations that are not

accompanied by a change in total composition" (5). The extent to which

particle weight influences the chromatographic behavior of molecules

of deoxyribonucleate on columns of Mg IRC-50 has not as yet been

determined. It would be of interest to apply techniques such as

have been used by Schumaker and Schachman (11) in order to determine

the distribution of sedimentation coefficients within the chromato-

graphic fractions. As was noted in the Introduction, the chromato-

graphic procedures of Bendich and co-workers provide molecules having

progressively increasing sedimentation coefficients but essentially

unchanging base composition (9). The last fractions eluted from

columns of ECTEOLA-cellulose have an average sedimentation coefficient

higher than that of the total, unfractionated material. However, the









78

successive fractions also exhibit a progressively greater heterogeneity

with respect to distribution of sedimentation coefficients. For this

reason it is difficult to compare fractions obtained by use of Mg IRC-50

with fractions obtained using columns of ECTEOLA-cellulose. Different

principles appear to underlie the two fractionation methods. In con-

trast to the results obtained by Bendich and co-workers, the procedures

of Chargaff (4), of Butler (6), and of Brown (8) and their co-workers

provide fractions of deoxyribonucleate which exhibit the same trends in

base composition as the fractions eluted from columns of Mg IRC-50.

The use of Mg IRC-50 for the isolation of such fractions of deoxyribo-

nucleate presents a number of advantages not offered by the previous

procedures. There is little doubt about the reproducibility of the

Mg IRC-50 chromatograms. The resin is readily available commercially

and several lots have provided essentially identical results. The

earlier procedures are difficult to standardize and to reproduce

because they depend on the use of histones of uncertain homogeneity.

Moreover, "denatured" histone and undenatured histone yield considerably

different chromatographic results (8). Although the histone-chloroform

gel procedure causes serious losses of transforming activity, passage

through columns of Mg IRC-50 or through certain histone-cellulose

columns (8) has been found to be innocuous in this respect.

The chromatographic studies have not as yet been concerned

directly with the problem of whether each specimen of deoxyribonucleate

consists of a spectrum of differently constituted molecules or rather,


16C. F. Crampton, E. Chargaff, and R. D. Hotchkiss, unpublished
observations (1953).









79

consists of a mixture of a relatively few different molecules. The use

of Mg IRC-50 by itself or in conjunction with alternative fractionation

methods may eventually provide information pertinent to this problem.

In this regard, determination of the distribution of sedimentation

coefficients (10, 11) as well as density gradient analysis in CsCl

solutions (12) and studies of the temperature dependence of ultraviolet

absorbancy (13) may prove to be convenient techniques for evaluating

the extent of heterogeneity in fractions obtained from columns of

Mg IRC-50. The interpretation of the chromatographic studies would also

be aided by definitive knowledge of the chromatographic behavior of

specimens of deoxyribonucleate which are, in effect, initially more

homogeneous than those examined thus far. Such specimens could be

obtained by proper choice of biological origin, by the preliminary use

of alternative fractionation techniques, or by the application of con-

ditions, such as controlled heating, which preferentially destroy the

native structure of certain molecules without altering others.

The evaluation of molecular parameters which underlie the

chromatographic fractionations in the biological studies is even more

complicated. Unlike the "crude expedient" of measurements of total

base composition, measurements of transforming activity need not be the

result of a positive contribution by all of the molecules present in a

given fraction. It will be recalled that the transforming activity of

specimens of deoxyribonucleate from Pneumococcus was distributed

throughout most of the chromatographic fractions. This could result

either from the existence of the same biological activity in molecules

of different composition or from the association of all of the









80

biological activity with molecules of discrete composition which, for

some reason, interacted with Mg IRC-50 to an equivalent degree over a

diffuse range of concentrations of Mg(OAc)2. Similar uncertainties

pertain to the results of the incorporation experiments in which the

specific activities of thymine changed considerably throughout the suc-

cessive fractions. An incorporation period of five minutes corresponds

to 17. of the average time of 8.4 hours which each cell devotes to the

synthesis of deoxyribonucleate during the growth cycle (17). Since the

generation time of the cells is about eighteen hours (17), roughly half

of all the cells present in an exponentially growing culture are engaged

in synthesizing deoxyribonucleate. If utilization of the precursors

began immediately after injection in all cells which were synthesizing

deoxyribonucleate, the isolated specimens would be about "0.5% labeled."

However, it is not known whether many molecules are slightly labeled,

or a small number of molecules are intensively labeled. Moreover, the

occurrence of radioactivity in all chromatographic fractions poses the

same problem noted above with respect to the ubiquity of transforming

activity among the fractions from pneumococcal specimens.

The characterization of fractions by measurements of activities

is hazardous, since the distribution of activity may be governed by

minor structural peculiarities not shared by all of the molecules present

in the fractions. That is, incipiently labeled deoxyribonucleate might

be expected to behave abnormally upon chromatography because a double-

stranded structure had not been completed when the cells were chilled

prior to isolation of the specimen. Moreover, the chromatographic

behavior of the small absolute quantities of labeled deoxyribonucleate









81

could also be dictated chiefly by associated contaminants, such as pro-

tein, polysaccharide, or lipid, which had been attached during the

process of replication. In this connection, it will be recalled that

the specimen from rat liver contained a large amount of protein but

gave chromatographic patterns similar to those provided by calf thymus.

On the other hand, the preparations from Pneumococcus, which also con-

tained large amounts of protein (and ribonucleate), gave quite differ-

ent patterns. A systematic study of the protein content of chromato-

graphic fractions was not attempted routinely because of the presence

of only trace amounts of protein in most of the total specimens.

The fact that columns of Mg IRC-50 are able to separate differ-

ent molecules of deoxyribonucleate rests not upon the biological

studies but rather upon the chemical studies of composition and upon

the results of the rechromatography experiments. Nevertheless, the

biological studies have shown that, under certain experimental condi-

tions, affinity for Mg IRC-50 is in some way related to the metabolic

origin of molecules of deoxyribonucleate.















POSSIBLE SIGNIFICANCE OF THE BIOLOGICAL STUDIES


The results of the incorporation experiments are summarized in

Table XVII, which lists the relative specific activities of thymine in

the two chromatographic fractions between which the greatest differ-

ences were usually obtained. As can be seen, preferential labeling of

chromatographically different fractions of deoxyribonucleate from

Ehrlich ascites tumor cells was observed following brief incorporation

periods. The data again direct attention to the "heat peak" fraction,

the unusual properties of which were discussed in earlier sections. It

will be recalled that material in this position may be partly artifact

and partly an authentic component present in the deoxyribonucleate

submitted to fractionation. The base composition of the material

resembles that of total deoxyribonucleate and therefore is not in line

with the progressive increase in the ratio,(adenine + thymine)/(gua-

nine + cytosine), otherwise noted throughout the chromatogram. Of

particular significance is the fact that the specimens of deoxyribonu-

cleate from Pneumococcus provided relatively large amounts of "heat

peak" material in which specific transforming activity was comparable

to that of the total specimen.

Had the incorporation experiments compared in Table XVII dealt

with synchronously growing cells, the preferential labeling could be

explained on the premise that chromatographically different molecules

are synthesized sequentially. Radioautographic studies following


















O~ N 4
c~ ~e u.~


"4

0
0 1
w1-
m51
0 4J


41

4 U

S00
de a

SM









lw
0
41
Se
















$S
* j




















'.4
41
v



0s
a &


4
oMA




Si


S0
41
J 1



5'2





0 0
.04
**




'0
e4 1
W O
US'








w0
a).0








e44



I-'
1*4 1-4
X ~4


d
4.
41

0

la

'I



N
U


0

A
B*


u *

0


?1 d1 ?1



a C a
0o 1 t
V V t
I14 V4 V





WN s I nU*


0
I.'
'1
-4




0~~


N





41
0
'.4



'1







it~
'-I
4h
4


> >











administration of H3-thymidine suggest that sequential synthesis of de-

oxyribonucleate actually occurs in chromosomes of grasshopper testes (41)

as well as in plant tissues (42). However, in the present studies,

ascites tumor cells at all stages of replication were examined, thereby

excluding specific temporal sequence as a possible explanation of pref-

erential labeling.

The synthesis of deoxyribonucleate by Ehrlich ascites tumor

cells occurs during 8.4 hours of the interphase period of the life

cycle of these cells (17). It is possible that the biosynthesis of

individual molecules of deoxyribonucleate requires a large portion of

this time. If so, radioactivity would be confined to limited regions

of large numbers of incomplete molecules after brief incorporation

periods. Accordingly, the observed chromatographic distribution of

radioactivity could depend largely upon heterogeneity with respect to

degree of completion.

On the other hand, the complete synthesis of molecules of

deoxyribonucleate may require only a few seconds. That is, the

8.4 hour period of synthesis might encompass the successive syntheses

of many different molecules which are started and finished at differ-

ent times in individual cells. In this case different molecules could

attain different specific activities if there were a more rapid

17It is also possible that some molecules of deoxyribonucleate
are synthesized more rapidly than others because of restrictions imposed
by the statistical ease with which available precursors can be aligned
and polymerized in order to complete characteristic sequences of certain
of the molecules. Moreover, molecules containing large amounts of
5-methylcytosine might be assembled less rapidly than molecules which do
not contain this satellite base.









85

incorporation of the labeled precursor into one group of molecules than

into another. Such preferential incorporation could occur in at least

three ways.

1) The specimens of deoxyribonucleate may contain a small group

of macromolecular precursors from which all molecules of deoxyribonu-

cleate are derived. Such precursors would exhibit a more rapid increase

in specific activity than would the product molecules after the injec-

tion of isotopic compounds. Moreover, such precursor molecules would

subsequently lose their activity more rapidly than would the product

molecules. It is possible that deoxyribonucleate in the "heat peak"

corresponds to such precursor material. The "heat peak" fraction

differs from the bulk of the nucleate not only by its very high rate of

incorporation of radioactive precursors, but also by its abnormal af-

finity for Mg IRC-50. Moreover, the composition of "heat peak" deoxy-

ribonucleate resembles that of the whole specimen, a property that

would be expected for a fraction from which all the molecules of a

preparation arise. Precursor, or incomplete deoxyribonucleate, might

well be expected to differ structurally from completed nucleate in a

manner which could cause a greater affinity for the resin. Precursor

deoxyribonucleate could be single-stranded (43), multi-stranded (44,45),

or associated with materials which are not attached to completed mole-

cules. On the other hand, the specific activity of precursor deoxy-

ribonucleate should approach zero a long time after administration of

isotope, a requirement which is not met by the "heat peak" material.

The specific activity of all fractions, including "heat peak," ap-

proached a similar value twenty-three hours after injection. It is









86

quite possible, however, that precursor deoxyribonucleate in "heat peak"

is contaminated by a large amount of ordinary nucleate. In such an

event, the large amount of radioactive product deoxyribonucleate which

is eluted non-specifically in "heat peak" would obscure the low specific

activity of the precursor material. The differences in the specific

activities of the chromatographic fractions other than "heat peak"

could be attributable to a form of precursor deoxyribonucleate which

emerges together with the early chromatographic fractions from

Mg IRC-50.

2) If the different molecules of deoxyribonucleate in a single

cell are synthesized from independent pools of low molecular weight

precursors they would become labeled at unequal rates if the pools

differed in size. Those molecules formed from the smallest pools of

precursors would incorporate the added label most rapidly.

3) If molecules of deoxyribonucleate differed with respect to

the rate at which they were synthesized from a single pool of precur-

sors, differences in specific activity could result if there were

rapid turnover in a portion of the molecules, or possibly, if differ-

ent molecules were synthesized at specifically related times not

related to their chromatographic behavior.

The experimental observation of preferential labeling of differ-

ent fractions of deoxyribonucleate is not without precedent in the

literature. For example, Bendich, Russel, and Brown (46) found that

deoxyribonucleate extracted from rat tissues with 10% NaCl at 850 C

yielded fractions which differed in their solubility in saline. The

two fractions of deoxyribonucleate isolated from various organs several









87

days after injection of C 14-formate contained different amounts of

radioactivity. However, the specific activities of different bases

from the fractions of a single organ did not parallel one another, nor

did different organs consistently show a greater incorporation into one

of the fractions than into the other. The authors conclude, "the

results not only show that [deoxyribonucleatej of any single organ is not

metabolically homogeneous, but that the individual bases of the nucleic

acids of a given organ are renewed at dissimilar rates, and that the

pattern of renewal varies from organ to organ." However, the differ-

ences reported for fractions from a single organ were smallest with

respect to thymine, the base least likely to be contaminated by

material derived from highly radioactive ribonucleate. The greatest

difference in specific activity of thymine occurred in a specimen from

small intestine, where specific activities of 614 CPM/4mole and

779 CPM/imole were reported for Fractions 1 and 2, respectively. In

contrast to the experiments of Bendich et al., the studies summarized

in Table XVII have revealed large specific activity differences among

fractions with respect to a base which is not present in ribonucleate

from most sources. As has been noted previously (Table XII), parallel

results were obtained with one series of fractions by specific activity

measurements of guanine, which is present in both ribo- and deoxyribo-

nucleates. In addition, deoxyribonucleate was isolated by use of the

urea medium which has been found to provide specimens free from appre-

ciable amounts of uracil, which is a characteristic constituent of

ribonucleate. Finally, extremely similar patterns of incorporation

were given both by H 3-thymidine and C -formate, precursors which are









88

utilized for the synthesis of the two types of nucleate with consider-

ably different efficiencies.

A second report of metabolic heterogeneity of deoxyribonu-

cleate is that of Friedkin and Wood (47), who incubated chicken bone

marrow cells in vitro for two hours with C14-thymidine. The incubation

mixtures were then frozen, lyophilized, and extracted with organic

solvents. After stirring in 0.9% NaCl, the material was treated with

sodium dodecyl sulfate by the procedure of Kay et al. (21). Three

successive extractions provided three fractions of deoxyribonucleate.

The specific activity of the thymidine isolated from enzymatic hydroly-

sates varied by as much as 3.3 from one fraction to another. The

interesting results of Friedkin and Wood merit further study. They

are, however, somewhat difficult to interpret since the fractions were

obtained from unpurified deoxyribonucleate by ill-defined techniques

not known to give fractions which differ in other respects. Moreover,

the use of a system consisting of an initially heterogeneous population

of cells, in which there is not known to be a net synthesis of deoxy-

ribonucleate, invites uncertainties which were avoided as far as

possible in the present work.















SUMMARY


Variables affecting the specific interaction between Mg IRC-50

and highly purified specimens of deoxyribonucleate were examined.

Conditions optimal for the analysis of molecules of different chemical

composition were developed and applied to specimens of deoxyribonucleate

isolated from Ehrlich ascites tumor cells which had incorporated

C 14-formate or H3-thymidine for very brief intervals of time. Large

differences between the fractions with respect to the specific activity

of their thymine residues were found, suggesting that the columns of

Mg IRC-50 may be used to separate molecules of different biological

origin as well as of different chemical composition. The basis of the

interaction of Mg IRC-50 and deoxyribonucleate is discussed critically

and explanations are offered for the preferential incorporation of

precursors into chromatographically different fractions.















REFERENCES


1. Miescher, F., Hoppe-Seyler's Med. chem. Unters., 4, 441 (1871);
translated by Gabriel, M. L., in M. L. Gabriel and S. Fogel
(Editors), Great experiments in biology, Prentice-Hall, Inc.,
Englewood Cliffs, New Jersey, 1955, p. 233.

2. Levene, P. A., and Bass, L. W., Nucleic acids, The Chemical
Catalog Company, New York, 1931.

3. Chargaff, E., J. Cellular Comp. Physiol., 38, 41 (1950).

4. Chargaff, E., Crampton, C. F., and Lipshitz, R., Nature, 172, 289
(1953).

5. Crampton, C. F., Lipshitz, R., and Chargaff, E., J. Biol. Chem.,
211, 125 (1954).

6. Lucy, J. A., and Butler, J. A. V., Nature, 174, 32 (1954).

7. Brown, G. L., and Watson, M., Nature, 172, 339 (1953).

8. Brown, G. L., and Brown, A. V., Symposia Soc. Exptl. Biol., 12,
6 (1958).

9. Rosenkranz, H. S., and Bendich, A., J. Am. Chem. Soc., 81, 902
(1959).

10. Butler, J. A. V., and Shooter, K. V., in W. D. McElroy and B.
Glass (Editors), The chemical basis of heredity, Johns Hopkins
Press, Baltimore, 1957, p. 540.

11. Schumaker, V. N., and Schachman, H. K., Biochim. et Biophys. Acta,
23, 628 (1957).

12. Sueoka, N., Marmur, J., and Doty, P., Nature, 183, 1429 (1959).

13. Marmur, J., and Doty, P., Nature, 183, 1427 (1959).

14. Vendrely, R., in E. Chargaff and J. N. Davidson (Editors), The
nucleic acids. Vol. II, Academic Press, Inc., New York, 1955,
p. 155

15. Swift, H., Physiol. Zool., 23, 169 (1950).

16. Howard, A., and Pelc, S. R., Heredity, 6, 261 (1953).
90









91

17. Edwards, J. L., Koch, A. L., Youcis, P., Freese, H. L., Laite, M.
B., and Donalson, J. T., J. Biophys. Biochem. Cytol., 7,
273 (1960).

18. Swick, R. W., Koch, A. L., and Handa, D. T., Arch. Biochem.
Biophys., 63, 226 (1956).

19. Fresco, J. R., and Bendich, A., J. Biol. Chem., 235, 1124 (1960).

20. Crampton, C. F., Benson, A. M., Rodeheaver, J. L., and Wade, A. E.,
Federation Proc., 17, 206 (1958).

21. Kay, E. R. M., Simmons, N. S., and Dounce, A. L., J. Am. Chem.
Soc., 74, 1724 (1952).

22. Crampton, C. F., Stein, W. H., and Moore, S., J. Biol. Chem.,
225. 363 (1957).

23. Moore, S., and Stein, W. H., J. Biol. Chem., 211, 893 (1954).

24. Hirs, C. H. W., Moore, S., and Stein, W. H., J. Biol. Chem., 200,
492 (1953).

25. Crampton, C. F., and Petermann, M. L., J. Biol. Chem., 234, 2642
(1959).

26. King, E. J., Biochem. J., 26, 292 (1932).

27. Wyatt, G. R., Biochem. J., 48, 584 (1951).

28. Moore, S., and Stein, W. H., J. Biol. Chen., 192, 663 (1951).

29. Moore, S., and Stein, W. H., Advances in Protein Chem., XI, 191
(1956).

30. Lee, W. A., and Peacocke, A. R., J. Chem. Soc., 3361 (1951).

31. Gurd, F. R. N., and Wilcox, P. E., Advances in Protein Chem., XI,
312 (1956).

32. Pullman, B., and Pullman, A., Biochim. et Biophys. Acta, 36 343
(1959).

33. Reiner, B., and Zamenhof, S., J. Biol. Chem., 228, 475 (1957).

34. Shack, J., and Bynum, B. S., Nature, 184, 635 (1959).

35. Wiberg, J. S., and Neuman, W. F., Arch. Biochem. Biophys., 72, 66
(1957).









92

36. Edwards, J. L., Smith, S. W., Westmark, E. R., and Youcis, P. M.,
Federation Proc., 18. 475 (1959).

37. Schmidt, G., and Thannhauser, S. J., J. Biol. Chem., 161, 83 (1945).

38. Frankel, F. R., Knapp, J., and Crampton, C. F., Federation Proc.,
19, 306 (1960).

39. Shapiro, H. S., and Chargaff, E., Biochim. et Biophys. Acta, 39,
68 (1960).

40. Burton, K., and Petersen, G. B., Biochem. J., 75, 17 (1960).

41. Lima-de-Faria, A., J. Biophys. Biochem. Cytol., 6, 457 (1959).

42. Taylor, J. H., Expt Cell Research, 15, 350 (1958).

43. Sinsheimer, R. L., J. Mol. Biol., 1, 43 (1959).

44. Stent, G., Advances in Virus Research, 5, 138 (1958).

45. Rich, A., Nature, 181, 521 (1958).

46. Bendich, A., Russel, P. J., Jr., and Brown, G. B., JI. Biol. Chem.,
203, 305 (1953).

47. Friedkin, M., and Wood, H., IV, J. Biol. Chem., 220, 639 (1956).















VITA


The candidate was born in Baltimore, Maryland on July 6, 1934.

He received a degree of Bachelor of Science in June, 1955 from the

College of Chemistry and Physics at the Pennsylvania State University,

and in June, 1957, a degree of Master of Science from the College of

Agricultural and Biological Chemistry at the same institution. His

thesis was titled, "Electrophoretic and Bactericidal Studies on the

Sera of Normal and Scouring Calves." He began his studies for the

degree of Doctor of Philosophy at the College of Medicine, University

of Florida in September, 1957. He has held teaching assistantships and

research fellowships at these two institutions.




Full Text

PAGE 1

COMPARATIVE CHEMICAL AND BIOLOGICAL STUDIES ON THE FRACTIONATION OF DEOXYRIBONUCLEIC ACIDS By FRED ROBERT FRANKEL A DISSERTATION PRESENTED TO THE GRADUATE COUNOL OF THE U NIVER S ITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGR EE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLOR IDA August, 19 6 0

PAGE 2

ACKNOWLEDGMENT The research attitudes and techniques acquired by the candidate a1 a result of his association with Dr. Charle F. Crampton will be a guide in all future activities of the candidate. His debt to Dr. Crampton 11 truly great. The candidate would also like to acknowledge the interest of his 1upervisory coimnittee and other members of the Department of Biochemistry, and to thank Dr. Frank W. Putnam for hia careful criticism of this manuscript during its preparation. The interest and generoaity of Dr. Joshua L. Edwards during the candidate's two year stay in the Department of Pathology ii especially appreciated. -, ii

PAGE 3

TABLE OF CONTENTS LIST OF TABLES LIST OP FIGURES INTRODUCTION Chemical and Biological Heterogeneity of Deoxyribonucleate --l EXPERIMENTAL PROCEDURES Isolation and Purification of Deoxyribonucleates --4 Deoxyribonucleates from calf thymus and rat liver --4 Deoxyribonucleates from Ehrlich ascites tumor cells --4 Chromatography of Deoxyribonucleate by Use of Mg IRC-5O --8 Analysis and Quantitative Estimation of Purines and Pyrimidines in Hydrolysates of Deoxyribonucleates --13 Use of columns of Dowex 5O-X4 --13 Paper chromatography --15 Radioisotope Counting Techniques --16 Carbon-14 labeled compounds --16 Tritium labeled compounds --17 RESULTS AND DISCUSSION . . . . . . . Studies of the Interaction Between Mg IRC-5O and Deoxyribonucleate --18 Specificity of the interaction --18 Capacity of Mg IRC-5O for deoxyribonucleate 21 Effect of certain variables on the interaction 25 Temperature --25 pH --27 i+ Ions other than Mg --27 Plow rate --32 Particle size of resin --34 iii Page V vii 1 4 18

PAGE 4

Chromatographic behavior of deoxyribonucleates from rat liver, Ehrlich aacites tumor cells, and Pneumococcus --34 Rechromatograpby of fractions of deoxyribonucleate from calf thymus and rat liver --41 Effect of sample load --42 Base composition of fractions of deoxyribonucleate from calf thymus and Ehrlich aacitea tumor cells --47 Incorporation of Precursors into Fractions of Deoxyribonucleate from Ehrlich Ascites Tumor Cells --50 Preliminary experiments with cl4.formate --50 Effect of non-radioactive formate --56 Effect of time --59 Incorporation of c14-formate into fractions of deoxyribonucleate --61 Incorporation of H 3-thymidine into fractions of deoxyribo nucleate --69 In vivo study of brief duration --69 In vivo study of long duration --71 In vitro study of brief duration --73 CRITICISM OF mE CHROMATOGRAPHIC STUDIES . POSSIBLE SIGNIFICANCE O F THE BIOLOGICAL STUDIES SUMMARY REFERENCES . . . . . iv 76 8 2 89 90

PAGE 5

Table I. II. III. IV. v. VI. VII. VIII. IX. x. XI. XII. LIST OP TABLES Properties of some specimens of deoxyribonucleate Base composition of successive fractions of deoxyribo-nucleate eluted from Mg IRC-50 Estimation of maximum capacity of Mg IRC-50 for deoxy-Yibonucleate ............. Distribution of constituents among chromatographic fractions of deoxyribonucleate from Pneumococcus Recovery of "heat peak" material from deoxyribonucleate preparations from various sources Base composition of chromatographic fractions of deoxyribonucleate from calf #32 thymus obtained by overloading columns of Mg IRC-50 Base composition of chromatographic fractions of deoxyribonucleate from calf #32 thymus Base composition of chromatographic fractions of deoxyribonucleate from Ehrlich ascites tumor cells (E.A #1) . . . . . . . . . . . Base composition of chromatographic fractions of deoxyribonucleate from Ehrlich ascites tumor cells (EA 117) 14 In vivo incorporation of C -formate into fractions of deoxyribonucleate from Ehrlich ascites tumor cells (EA #1) . . . . . . . . . . 14 Five minute in vivo incorporation of C -formate into the thymine of fractions of deoxyribonucleate from Ehrlich ascites tumor cells (EA #7) The specific activity of bases in fractions of deoxyribonucleate from Ehrlich ascitea tumor cells (EA #7) V Page 9 22 24 38 39 46 49 53 54 58 63 66

PAGE 6

XIII. XIV. xv. XVI. XVII. 14 Five minute in vivo incorporation of C -formate into the thymine of fractions of deoxyribonucleate from Ehrlich ascites tumor cells (EA #9) Five minute in vivo incorporation of tritiated thymi dine intofractions of deoxyribonucleate from Ehrlich ascites tumor cells (EA #16) .. ~nty-three hour in vivo incorporation of tritiated thymidine into fra'ctI'ons of deoxyribonucleate from Ehrlich ascites tumor cells (EA #17) Two minute in vitro incorporation of tritiated thymi dine intofractions of deoxyribonucleate from Ehrlich ascites tumor cells (EA 115) ... Relative specific activity of thymine in chromato-graphic fractions of deoxyribonucleate from Ehrlich ascites tumor cells vi 68 70 72 75 83

PAGE 7

Figure l. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. LIST OF FIGURES Page Chromatography of purines and pyrimidines by use of 60 x 0.9 cm columns of Dowex 50-X4, "through 200" 14 Chromatography of deoxyribonucleate by use of (A) a 50 ml mixing chamber or(!) a 125 ml mixing chamber with columns of Mg IRC-50, "through 200" 19 Stepwise elution of deoxyribonucleate from a 30 x 0.9 cm coll.Dim of Mg IRC-50, "through 325," at 30 C 20 Chromatography of deoxyribonucleate (A) at 30 C and (!) at 50 C by use of 29 x 0.9 cm-columns of Mg IRC-50, "through 32511 26 The effect of certain variables on the chromatography of deoxyribonucleate from calf #32 thymus 28 Chromatography of specimens of deoxyribonucleate from Pneumococcus by use of colunms of Mg IRC-50 36 Rechromatography of deoxyribonucleate from calf thymus by use of columns of Mg IRC-50, "through 200" 43 Chromatography of deoxyribonucleate from rat liver by use of 29 x 0.9 cm columns of Mg IRC-50, "through 325'' . . . . . . . . . . 44, Preparative chromatography of deoxyribonucleate from calf #32 thymus by use of overload conditions 48 Chromatography of deoxyribonucleate from Ehrlich ascites tumor cells (EA #1) by use of Mg IRC-50 51 Preparative chromatography of deoxyribonucleate from Ehrlich ascites twnor cells (EA 17) by use of over-load conditions 52 14 Effect of time of incorporation of C -formate on the specific activity of bases of deoxyribonucleate from Ehrlich ascites tumor cells . 60 Stepwise elution from Mg IRC-50 of deoxyribonucleate from Ehrlich aacitea tumor cells (EA 19) 67 vii

PAGE 8

INTRODUCTION Chemical and Biological Heterogeneity of Deoxyribonucleate Near the end of the nineteenth century, Miescher isolated an acid-insoluble phosphorus-rich material from salmon sperm, and from the nuclei of pus cells and other tissues. He termed this material 11nuclein11 (1). Later workers showed that nuclein was, in fact, a mixture of at least two substances, one containing the sugar, riboae, and the other, the sugar, deoxyribose. This latter t y pe of nuclein, which was confined almost exclusively to cell nuclei, became known as deoxyribonucleic acid. Much of the foundation of our present knowledge of the chemical structure of nucleic acids was established by P.A. Levene (2). More recently, through the development and use of quantitative semi-micro techniques for the analysis and estimation of purines and pyrimidines, Chargaff and his colleagues showed that specimens of deoxyribonucleate from different biological sources frequently differ in their nitrogenous base composition (3). These workers concluded that the nuclei of different species may contain different molecules of deoxyribonucleate. In a greement with this chemical evidence of species-specificity, is the growing body of biological evidence which suggests that genetic information may be embodied in the composition and structure of de o x yribonucleate. Moreover, specimens of deoxyribonucleate from individual sources have recently been found to consist of populations of closely l

PAGE 9

2 related but chemically distinct molecules. Thie was fir1t shown by Chargaff, Crampton, and Lipshitz, who extracted fractions of the nucleate having decreasing contents of guanine and cytosine from hietone chloroform gels by means of aqueous solutions of increasing salt concentration (4, 5). Similar findings were obtained by other workers whose technique& also depended upon the specific, fractional di1sociation of deoxyribonucleate from cationic adsorbents to which it had been bound by electrostatic linkages (6, 7, 8). However, the elution of deoxyribonucleate from columna of cellulose linked to triethanol amine reaidues by reaction with epichlorohydrin (ECTEOLAcelluloae) provides successive fractions of e1sentially unchanging compo1ition, but of gradually increasing average sedimentation constant (9). Thus, another important aspect of the molecular heterogeneity of deoxyribonucleate arises from variation with respect to sedimentation properties (9, 10, 11). Recent studies suggest that molecule in a single specimen of deoxyribonucleate are also heterogeneous with reapect to their apparent density when analyzed by centrifugation in solution of CsCl (12). Such studies indicate that apparent density is related directly to base composition. Evidence has also been presented for a relationehip between base composition and thermal stability of deoxyribonucleate (13). With few exceptions, the amount of deoxyribonucleate per nucleus in cells of a given species is a multiple of the haploid amount, and is characteristic of the species (14). If the two daughter cells arising from division contain the same amount of deoxyribonucleate as the parent cell, division must be preceded or accompanied b y a net

PAGE 10

3 synthesis of deoxyribonucleate. At present, little is known regarding thoae factors which initiate and govern the synthesis of the various molecule of deoxyribonucleate present in ma.umalian cells or of the mechaniama which control the constancy of their amounts per nucleus. That the net synthesis of deoxyribonucleate is confined to interphase in cell from several mauaalian tiaaues, was shown in 1950 by means of cytochemical techniques (15). This finding was confirmed by radioautographic evidence (16), and has recently been extended to Ehrlich ascites tumor cells (17). As has been shown by several groups of workers (18, 19), the deoxyribonucleate of manwalian cells exhibits extreme metabolic stability, suggesting that it is not degraded during the life of the cell, or that, if degraded, its substance is efficiently reutilized. The work to be described below stems from the paradoxical finding that deoxyribonucleate, despite its polyanionic character, is bound by columns of the magneaium form of IRC-50, a polycarboxylic acid reain. Elution of the deoxyribonucleate yields successive fractions of gradually changing composition (20). In the present studies, attempts were made!) to inveatigate the influence of various factors on the interaction of deoxyribonucleate from calf thymus with Mg IRC-50, 2) to de termine the usefulness of this system for the chromatographic fractionation of deoxyribonucleate from rat liver, Pneumococcus, and Ehrlich aaci tes tumor cells and 3) to compare the initial rates of in~ incorporation of radioactive precursors into cbromatographically different frac tions of deoxyribonucleate. The latter studies of incorporation were undertaken with the hope of obtaining information regarding metabolic propertiea of different molecules of deoxyribonucleate.

PAGE 11

EXPERIMENTAL PROCEDURES Isolation and Purification of Deoxyribonucleates Deoxyribonucleates from calf thymus and rat liver Details of the isolation of the specimens from calf thymus and 1 rat liver have been prepared for publication. The specimens of deoxy-ribonucleate from the thymus of calf #30 were isolated from a nucleo histone which had been extracted by 0.0004 M NaHC03 Preparations 30A and 30B were obtained by slightly different procedures, but were indistinguishable with respect to all properties which were examined. The specimen from calf #32 thymus was obtained by use of the procedure of 1 Crampton il al. which depends on the insolubility of deoxyribonucleo-proteins in a solution of 6 M urea containing lo/. NaCl and 0.1% merthiolate. Nucleoproteins of the ribose type remain soluble in this medium, the use of which has provided specimen of deoxyribonucleate containing little or no uracil. Deoxyribonucleate was prepared from rat #11 liver using this same method. All of these preparations were deproteinized by two treatments with Duponol by use of the procedure of Kay il al. (21). Deoxyribonucleate from Ehrlich ascite1 tumor cells 2 The line of Ehrlich ascites tumor cells was maintained by 1c. F. Crampton, B. Greenfield, J. Adair, and F. R. Frankel, manuscript in preparation. 2we are indebted to Dr. C. Heidelberger for supplying the Ehrlich ascites tumor cells used in these studies. 4

PAGE 12

5 weekly transfers of approximately 2 x 10 6 cells to groups of hybrid Swias white mice, fifty to sixty days of age. Cells for inoculation were aspirated by use of sterile, large bore, transfer pipettes from the peritoneal cavities of mice which had been killed by cervical dislo cation. About two to four minutes were required to collect the cells which were then centrifuged briefly at room temperature. The packed cells were then suspended in twenty volumes of sterile saline solution. 7 Appropriate aliquots of such suspensions, which contained about 2 x 10 cells per ml, were used for inoculation. Specimens of deoxyribonucleate designated EA #1 to #17 were isolated from cells donated by the respective groups of uniform mice. The mice of each group received 2 x 10 6 to 20 x 10 6 cells from a single inoculum five to six days prior to sacrifice. When deoxyribonucleate was to be extracted, the cells were collected in the manner described above, except that the peritoneal cavities of the mice were frequently rinsed several times with cold saline in order to increase the yield of cells. Moreover, the cells were always placed immediately in a tube iumeraed in an ice bath. In studies of the uptake of labeled compounds, the stated intervals of incorporation correspond to the average time which elapsed between injecting the radioactive precursor and chilling the cells. 0 The chilled cells were promptly centrifuged at 4 C, and the sediment remaining after decanting the supernatant fluid was immediately frozen in a mixture of dry ice and ethanol. The pellets were 0 occasionally stored at -20 C for twenty-four hours before being proc-essed further.

PAGE 13

6 In order to extract deoxyribonucleate from the packed cells, the pellets were permitted to thaw partially in the cold room (at 4-ff'c, at which temperature all subsequent procedures were performed). The packed cells (2 to 5 ml) were transferred with five to ten volumes (with respect to the volume of packed cells) of the urea medium to the semi-micro cup of a high speed mixer equipped with cutting blades. The mixer was then operated at full speed for two minutes. In the case of EA #1 and EA #4, the initial homogenates were prepared by use of a Potter-Elvehjem tissue grinder equipped with a plastic pestle. After the addition of two drops of octyl alcohol, in order to collapse the foam, the homogenate was transferred to a plastic centrifuge tube, and the insoluble deoxyribonucleohistone and other cell debris were sedi mented by centrifugation at 20,000 x A for ten minutes. The sediment was redispersed completely in five to ten volumes of fresh urea medium with the aid of a mechanically driven pestle which had been machined to fit the centrifuge tube tightly. The suspension was centrifuged as before, and the resuspension and sedimentation were repeated using fresh urea medium. The sediment was then washed two times in the same manner with 0.15 M NaCl. The sediments at this point for preparations EA #1, #4, #5, #6, and #7 were dispersed in 4 ml of d istilled water and made about l Min NaCl by the addition of 5 ml of 2 M NaCl. In order to reduce the viscosity, 1 to 15 ml of l M NaCl were added to dilute the suspensions, which were then stirred for eleven to fifteen houri in order to effect dissociation and solubilization of the histone and deoxyribonucleate. The suspensions in 1 M NaCl were then centrifuged for thirty to sixty minutes at 20,000 x A in order to remove insoluble

PAGE 14

7 conatituents. Before extraction with 1 M NaCl, the sediment of EA #9 was first extracted with nine volumes, and then with one volume, of 0.2 M Ba(OAc) 2 in order to dissociate histone Fraction A (22). The 1ediment8 of BA #15, #16, and #17 were extracted with five volumes of 0.2 M Ba(OAc) 2 before being dispersed in four volumes of 2 M NaCl. The suspensions in 2 M NaCl were stirred one hour, centrifuged at 20,000 x g for fifteen minutes, and the sediments were re-extracted with 2 M NaCl. The combined, clear to slightly turbid, 1 Mor 2 M NaCl supernatant fluids from each experiment were transferred to a 125 ml Erlen meyer flask which was shaken vigorously after the addition of one volume of 95% ethanol. The fibers of precipitated deoxyribonucleate were wa1hed once with 70% ethanol. For further deproteinization, the precipitated deoxyribonucleate was suspended in 5 to 10 ml of distilled water,and an equal volume of 0.2 M NaCl, 0.05 M Na citrate was added. Duponol was added to a concentration of 0.45% and the solution was brought to room temperature and stirred for three hours. Solid NaCl was added to a concentration of l M, and permitted to dissolve completely, after which the solution was centrifuged at 20,000 x g for fifteen minutes at 4 C. Deoxyribonucleate was precipitated from the supernatant fluid by the addition of an equal volume of 95% ethanol, collected, and washed with 7(1% ethanol. After stirring the nucleate with 8 to 20 ml of distilled water for one 0 to twelve hours at 4 C, a second Duponol treatment was performed aa described above for two hours (EA #1, #4, #5, #6, and #7) or one hour (EA #9, #15, #16, and 117). After the final precipitation by ethanol, the deoxyribonucleate was washed three times with 7(1% ethanol, three

PAGE 15

8 times with 95% ethanol, and dried overnight at room temperature in an evacuated desiccator over anhydrous CaC12 The product was stored at -20 c. Table I lists the yields and several properties of some of the specimens of deoxyribonucleate used in these studies. The yields of other specimens from Ehrlich ascites tumor cells, per ml of packed cells, were as follows: EA #1, 4 mg; EA #7, 7 mg; RA 19, 3 mg; EA 115, 7 mg; EA #16, 7 mg; EA 117, 6 mg. Chromatography of Deoxyribonucleate by Use of Mg IRC-50 Amberlite IRC-50, XE-64 (CG-SO, Type 2, presently supplied by Fisher Scientific Co.), was obtained in the hydrogen form as the "through 20~' mesh material; it was prepared for chromatography by first converting the resin to the sodium form by adjusting to pH 8 or 9 a 6-liter suspension of the resin in distilled water by the addition of pellets of NaOH. The finest resin particles were removed by repeated decantations of the thoroughly stirred 6-liter aqueous suspension after settling periods of about thirty minutes. After about ten decantations, the settled resin was transferred in small portions either to a 200 mesh or a 325 mesh sieve and screened to yield "through 200" or "through 325" material (23). The material passing through the sieves was collected and slowly washed under gravity on a Buchner funnel with successive 2 to 4 liter portions of 4 N NaOH, water, 4 N RCl, water, 4 N NaOH, water, 4 N HCl, water, acetone, and water. The resin was con verted to the magnesium salt by slowly passing l M Mg(OAc)2 through the resin bed until the pH of the effluent rose to that of the influent,

PAGE 16

9 TABLE I Properties of some specimens of deoxyribonucleate Details of the methods used for the isolation and characterization of theae specimens are given in the experimental aection. Groas Loss of Per cent8 Source of yield weight (P)!60d 1pecimen (mg/g Protein O.l M on NaCl or dryingc Phosphoru1 Biuret Ninhydrin mg/ml) b Calf #JOA thymus 19 14.1 8.56 1.2 3.6 6930 Calf #32 thymus 24 13.9 8.28 2.9 4.5 6830 Rat #11 liver 1 14.0 7.23 18 ---7040 Ehrlich ascites tumor cella EA #4 2 17.8 8.72 1.2 2.7 6430 BA #5 5 18.6 8.94 3.0 2.3 6850 EA #6 4 17.0 9.11 1.6 4.4 6450 8As, per cent of the weight after drying. bThe figures refer to actual recoveries per g of tissue or per ml of packed tumor cells. cAs, per cent of weight before drying. dAtomic extinction coefficient with respect to phosphorus, at 260 ~-

PAGE 17

10 about pH 7 or 8. The preparation of the resin was completed by washing with 0.05 M Mg(OAc)2 pH 7 to 8. At this point, one-to-one slurries of the resin in 0.05 M Mg(OAc)2 were prepared and used to pour successive 10 to 15 cm segments of analytical columns (0.9 cm diameter) or preparative columns (2 cm diameter), in the manner described by Moore and Stein (23). Unless otherwise noted, the columns were operated in con-o 0 denser jackets through which water of 30 0.2 C was circulated. Effluent from the analytical columns was collected at a rate of about 20 ml per hour, while the preparative columns were operated at a rate of flow of about 50 ml per hour. To promote the regular flow of drops of uniform size past the light beam of the drop counter and into the collection tubes, a roll of silver gauze was inserted into the tip of the column (24). The gauze made contact with the fritted glass disc of the chromatograph tube and terminated in a point. The zinc salt of IRC-50 was prepared by substituting solutions of Zn(OAc) 2 of pH 6 for the Mg(OAc)2 solutions in the above procedure. Carboxymethyl-cellulose, with an exchange capacity of 0.39 meq per g (Lot No. 59-1, supplied by Bio-Rad Laboratories) was obtained in the hydrogen form, and was converted to the magnesium form as described above. The sulfonated polystyrene resins (Dowex 50-X2 and Amberlite 3 IR-120, XE-69) were prepared as described previously (25) and converted from the hydrogen forms to the magnesium or zinc formt as for IRC-50. The sample to be chromatographed contained, in most cases, about 1 mg of deoxyribonucleate per ml of the initial buffer, and was Jc. F. Crampton, P. R. Frankel, A. M. Benson, and A. E. Wade, submitted for publication.

PAGE 18

11 0 prepared by swirling the nucleate in distilled water at 4 C until the specimen was dissolved. This required up to ten hours. Sufficient 1 M salt aolution was then added to yield a final concentration of the starting buffer of 0.05 M. Stirring was continued for about two hours, whereupon the solution was permitted to warm to room temperature, and after the removal of aliquots for the purpose of estimating the total absorbancy, the sample was added carefully to the column. The deoxy ribonucleate was eluted by eluents of continuously increasing concen tration or in a stepwise manner. Eluents of gradually changing compo sition were produced by use of a device constructed from a flat bottomed flask which served as the mixing chamber and contained the solutions of lowest concentration, and the upper ground glass joint of a wash bottle through which was introduced from a separatory funnel, the solutions of highest concentration. During operation of the device, the solution in the mixing chamber was stirred by a magnetic bar and the eluent was continuously removed through the delivery tube of the wash bottle. Most of the analytical chromatograms were performed by use of a 500 ml mixing chamber containing 0.05 M Mg(OAc)2 into which flowed 0.4 M Mg(OAc)2 Exploratory experiments indicated that a mixing chamber of this size gave satisfactory results. Smaller mixing chambers gave too few frac tions to cut conveniently, while larger mixing chambers gave peaks which were impractically broad. For preparative columns, the volume of the mixing chamber was scaled up by a factor of six, although the ratio of the crosssectional areas of the two columns was 4.9. The effluent from analytical columns was collected in fractions of 2 to 5 ml wile that from preparative columns was collected in fractions of 5 to 20 ml.

PAGE 19

12 The absorbancy of these fractions was determined at various wave lengths in the Beckman DU spectrophotometer, using distilled water as a blank. The light path of the quartz absorption cells was l cm. In an effort to eliminate the contribution of ultraviolet absorbing constituents in the effluent other than deoxyribonucleate, use was made of the difference between the 260 n and 290 qi absorbancies. Amounts of deoxyribonucleate, expressed as "absorbancy units," were calculated by multiplying the 260 n minus 290 n absorbancy difference by the total volume of the 1olution. The observation that one mg of deoxyribonucleate per ml of solution gives an absorbancy difference of 12.3 was uaed to convert absorbancy units to mg of deoxyribonucleate. Column fractions 0 were stored no longer than forty-eight hours at 4 before further treatment. To isolate deoxyribonucleate from the column effluent, fractions were combined where appropriate, made 1 M with respect to NaCl by the addition of eolid salt, and the nucleate precipitated by the addition of two to three volumes of cold 95% ethanol.4 The precipitates were washed three times with 70% ethanol and three times with 95% ethanol. The product was dried in an evacuated desiccator over CaC12 and etored ""ro facilitate the isolation of small amounts of pneumococcal nucleate with transforming activity, a known quantity of a solution of deoxyribonucleate from calf thymus was added as "carrier" prior to precipitation with ethanol. This procedure yielded difficultly interpretable data with respect to the transformation assays. An alter native recovery procedure, subsequently developed by Dr. R. D. Hotchkiss, is based on the co-precipitation of the nucleate with magnesium phosphate, and the subsequent solution of the deoxyribonucleate by use of ethylenediaminetetraacetate.

PAGE 20

13 0 at -20 C. Before precipitating the deoxyribonucleate from pooled sam-ples which contained only small amounts of material, the fractions were first concentrated by use of a rotary evaporator, the bath temperature 0 of which was kept less than 36 C. In some instances, such concen 0 trated solutions were dialyzed at 4 C to equilibrium against distilled water so that the final concentration of Mg(OAc)2 waa about 0.5 M. Analysis and Quantitative Estimation of Purines and Pyrimidines in Hydrolysates of Deoxyribonucleates Use of colwnns of Dowex 50-X4 In order to determine the base composition of specimens of deoxyribonucleate as well as of fractions of deoxyribonucleate isolated after chromatography by use of Mg IRC-50, the procedures described by 3 Crampton!!_ al. were used. A specimen of the material to be analyzed was hydrolyzed with 99% formic acid in an evacuated sealed tube for thirty minutes at 175 C. Aliquots of the hydrolysate, after removal of formic acid in an evacuated desiccator and solution of the residue in 1 N sulfuric acid, were taken for phosphorus determination (26) and chromatographic analysis. The analyses were performed by eluting the bases from columns of Dowex 50-X4 (60 x 0.9 cm) using an eluent of gradually increasing concentration of ammonium formate of pH 4. The column effluent was collected in 3 ml fractions, and the absorbancy of each fraction was determined in the ultraviolet spectrophotometer, using distilled water as a blank. A typical analysis is shown in Fig. 1~. To estimate the amount of material in each position of the chromatogram, factors obtained by chromatographing mixtures of standard bases under

PAGE 21

0 8 260 m URACIL THYMINE 14 STANDARDIZATION MIXTURE A GUANINE ADENINE CYTOSINE CYTOSINE 250 m 280 m __, -~--~~ ----'---'----'~~--' Fract ion No. 20 40 60 80 100 120 140 160 (2.9-3. 0 ml. per fraction throuohout) AMMONII.M FORMATE O.2N, pH4--{5OO ml.}--1.ON, pH4, 44C 1.4.--------.r---------------------------1 2 :l,...~ I--.. 1.0 ~I--.. ~;?:: 0 8 -.J ...... -.J 0 6 ~THYMINE 110,..0 FORMIC ACID HYDROLYSATE OF DNA (1.2mg) FROM MOUSE ASCITES TUMOR CELLS GUANINE 97. 3,-.o B ADENINE 120,-.0 8 0.4 & -:::: 260 0 2 m CYTOSINE 5-METHYL65.5,-,.o. CYT OSINE ( \ 260m,-,. C3,-.ol 260 m,-. Fract Ion No. 20 40 60 80 100 120 140 160 (2.9-3. 0 ml. per fraction ofter.fraction 12) AMMONIUM FORMATE O .2N, pH4---{5OO ml.r--1.ON, pH4, 44 C Fig. l Chr o matog r aphy of purines and pyrimidines by use of 60 x 0 9 cm columns of Dowex 50-X4, through 200. 11 (~) Standardiza-tion mixture of approximately 50 g of each baae. (~) Formic acid hydrolyaate of 1.2 mg of deoxyribonucleate from Ehrlich ascites tumor cells.

PAGE 22

15 the same conditions (Fig. l!) were used to convert unite of abaorbancy to moles of base. Paper chromatography Frequently, an alternative method of chromatography waa used for the isolation of bases for measurements of apecific radioactivity. The residue resulting from the formic acid hydrolysis of about 0.6 to 1.0 mg of deoxyribonucleate was dissolved in two drops of 1 N HCl and spotted in duplicate at the origin of a 19.5 cm x 40 cm sheet of Whatman No. 4 filter paper, leaving three lanes to serve as blanks. This was hung in a sealed chromatography jar, and developed by descending chromatography for twenty-five to thirty-five hours at room temperature 0 0 (23 C) with isopropanol which was 1.96 N with respect to HCl (27). The chromatogram was dried in a stream of warm air. The bases were 5 located by viewing the chromatogram under an ultraviolet lamp, and were cut out, along with the adjacent blank spots, using a pattern to assure elution of bases and blanks from equal areas of paper. The paper segments Vf::re rolled into cylinders and dep osited in 10 x 75 till1 teat tubes. To each test tube was delivered 2 ml of eluting solvent using a 1ingle calibrated volumetric pipette. Guanine and its blank were eluted with 1 N HCl, while the other bases were eluted with hen dried and viewed under ultraviolet light, chromatograms of bases which had been recovered from the effluent of columns of Dowex 50 showed a single dark spot at the position of the authentic base, aa well as a bright yellow fluorescent spot which moved faster than thymine. This fluorescent spot wa1 1een whenever any base, i1olated from Dowex 50 effluent, was chromatographed on paper, but not when the bases were added to the paper from stock solutions of the ba1e1, or directly from hydrolysatea. Evidently, the fluorescent material is derived from the aulfonated polystyrene resin.

PAGE 23

16 0.1 N HCl. Thymine containing tritium was eluted with water. The tubes were tightly covered with parafilm and allowed to stand twelve hours at room temperature. At the end of this period, the absorbancy of a portion of the eluate was determined in the ultraviolet spectrophotometer, and aliquots of the remaining eluate were plated for determination of specific radioactivity. Radioisotope Counting Techniques Carbon-14 labeled compounds Bases containing carbon-14 were isolated from the pooled effluent of columns of Dowex 50 by lyophilization of the amnonium formate solution in 250 ml flasks with the aid of infra-red lamps directed at the frozen solution. After the first lyophilization, the residue was dissolved in 5 ml of water and relyophilized. The final residue was taken up in l N HCl (guanine) or O.l N HCl (thymine, cytosine, adenine), diluted to 2 ml with the ame solvent, and an aliquot taken for determination of absorbancy. A portion of the remaining solution was plated for determination of radioactivity. Bases labeled with carbon-14 and isolated by paper chromatography were prepared as described in the previous section. Compounds dissolved in 0.1 N HCl could not be plated directly on metal. They were therefore applied to aluminum planchets onto which 1.8 cm diameter glass cover slips were secured with the aid of a small amount of silicone grease. Four equal aliquots containing about 5 g of the base were successively spread on the glass cover slip uniformly covering the glass area, evaporated to dryness under an infra-red

PAGE 24

17 lamp, and counted by use of a flow-window device. formed at 130 volts above the starting voltage. Counting was perA minimum of 3000 counts was accumulated for each sample. All counting rates were corrected for background. A straight line could in most cases be drawn to connect the points when count1 per minute was plotted against total volume plated, indicating that counting was performed under conditions of infinitesimal thickness. The slope of the line which joined the counting rates of the four auccessive aliquots was divided by the moles of base in the sample, determined spectrophotometrically, to obtain the specific activity of the sample. Tritium labeled compounds Thymine containing tritium was dissolved in water so that it could be plated and counted on steel planchets. For this purpose, two equal aliquots of 50 l containing about 0.6 g of base were successively spread uniformly on a planchet previously washed thoroughly with detergent, pickled by brief iDWDersion in 6 N HCl, washed finally with water and acetone, dried with a stream of warm air, and rimmed with silicone grease. The planchets, with an effective diameter of 2.7 cm, were dried under an infra-red lamp on a rotating platform in a stream of air, and counted in a windowless flow counter operated with 99.05% helium and 0.95% isobutane. Counting was performed 175 volts above the starting voltage. The specific activities of the two aliquots were extrapolated to zero masa to obtain the specific activity of the sample in the absence of self absorption.

PAGE 25

RESULTS AND DISCUSSION Studies of the Interaction Between Mg IRC-50 and Deoxyribonucleate Specificity of the interaction Under certain conditions, deoxyribonucleate is adsorbed by the magnesium form of the polycarboxylate resin, Amberlite IRC-50. Fig. 2! illustrates an experiment in which 1 mg of deoxyribonucleate from calf 130B thymus was dissolved in 0.05 M Mg(0Ac) 2 pH 7.6 and added to a column of Mg IRC-50 which had been equilibrated with an eluent of the same composition as the solvent. Virtually all of the nucleate wa bound. The binding was reversible in that about 90% of the nucleate was eluted by applying an eluent of gradually increasing Mg(OAc)2 concentration. The eluent was produced by adding 0.5 M Mg(OAc)2 to a 50ml mixing chamber initially filled with 0.05 M Mg(OAc)2 As shown in Fig. 2~, a slower rate of increase of the salt concentration in the eluent (achieved by use of a 125 ml mixing chamber) provided a broader elution pattern with the same specimen. The concentrations of Mg(OAc)2 sufficient to elute most of the molecules of deoxyribonucleate were determined by stepwise application of solutions of successively higher molarity. As shown in Fig. 3, obtained with the specimen from calf 132 thymus, elutlon occurred over the relatively narrow range of 0.20 M to 0.25 M Mg(OAc)2 In order to determine whether the elution of successive fractions is specific, the composition was examined of three successive 18

PAGE 26

-0 f 1.2 ] A (\ 11mg. DNA, 0.9 x 28cm. 0 U) 0.8 (\J
PAGE 27

-i 0 (j) 1.20 >-(\J u Cl) 0 .80 ::) z <( en 0:: 0 (/) en <( z ~0.40 0 <.O (\J 2.5 MG DNA (91.4% RECOVERY) FRACTION NUMBER (2 .0ML FRACTIONS) 50 100 200 t t t t t t 005M O.IOM 0.15M 0.20M 0.25M 0.25M--{250ML}---0.40M SOLUTIONS OF Mg(0Ac} 2 pH 7. 25 Fig. 3. Stepwise elution of deoxyribonucleate from a 30 x 0.9 cm column of Mg IRC-50, "through 325," at 30 C. The sample of deoxyribonucleate was from calf #32 thymus. N 0

PAGE 28

21 fractions recovered from the effluent in an experiment similar to that shown in Fig. 2~. The fractions were chosen so as to contain 20%, 4C1'1o, and 20% of the deoxyribonucleate applied to the column. Ae can be seen in Table II, the first and last fractions differed significantly in composition from the total preparation. These results indicate that the interaction of Mg IRC-50 with deoxyribonucleate is specific since reducing the capacity of the reaiG by increasing the Mg(OAc)2 concentration causes a preferential elution of molecules rich in guanine and cytosine. That is, the affinity of the resin for molecules rich in adenine and thymine is greater than for molecules rich in guanine and cytosine. Accordingly, use of the interaction offers a means for separating molecules of deoxyribonucleate of different composition. Capacity of Mg IRC-50 for deoxyribonucleate A fundamental consideration in studies of the interaction of any adsorbent with a solute is the ultimate capacity of the adsorbent, since the effects of variables may be difficult to interpret if experiments are performed under conditions close to saturation. In order to estimate the maximum capacity of Mg IRC-50 for deoxyribonucleate, 73.4 ml of a 0.1% solution in 0.05 M Mg(OAc)2 of the specimen from calf #32 thymus were added to a 28 x 2 cm column of "through 200" resin which had been equilibrated with 0.05 M Mg(OAc)2 pH 7.2. The capacity of the resin was exceeded under these conditions, since deoxyribonucleate appeared in the effluent after about 80 ml had been collected. Unadsorbed nucleate was then removed from the column by irrigation with 0.05 M Mg(OAc)2 until the absorbancy of the effluent

PAGE 29

22 TABLE II Base composition of successive fractions of deoxyribonucleate eluted from Mg IRC-50 The fractions -were obtained from deoxyribonucleate from calf #JOB thymus as noted in the text by a chromatographic experiment simi lar to that shown in Fig. 2B. The percentage in parentheses which follows each fraction number represents the portion of the deoxyribonucleate of the sample recovered in the respective fraction. The results are expreased as moles of base per 100 moles of total recovered bases. From Reference (20). Fraction number Unfraction.ated, Base total I ( 20%) II( 407.) 111(20%) deoxyribonucleate Thymine 25.4 29.0 30.0 27.6 Guanine 24.2 21.7 20.6 22.7 Cytosine 23.6 20.2 19.7 22.3 Adenine 24.5 27.l 28.5 27.5 5-Methylcytosine 2.1 1.7 1.3 (1. 4)

PAGE 30

23 approached base line. The results of replicate experiments presented 3 in Table III show that 1 cm of packed resin binds about 0.3 mg of deoxy-ribonucleate. As will be shown by the studies of base composition presented below, molecules of nucleate which are rich in guanine and cytosine have a low affinity for the resin and are eluted by 0.05 M Mg(OAc)2 under the experimental conditions used to estimate the capacity of Mg IRC-50. When the extent of interaction is limited by the number of binding sites on the resin, molecules of deoxyribonucleate with the greatest affinity apparently compete with, and/or displace the less strongly bound molecules. Accordingly, the absolute capacity of the resin may well vary with the base composition or physical structure of the molecules tested. Deoxyribonucleate degraded by heat or by enzyme action may also interact differently with the resin. Moreover, particle size may affect the interaction, since equivalent amounts of smaller particles will present a greater area for the binding of macromolecules which are unable to penetrate the surface of the resin. The estimated 3 maximum capacity of about 0,3 mg of deoxyribonucleate per cm of Mg IRC-50 3 may be compared with 0.5 to 2.3 mg per cm found for columns of EC'l'EOLA-cellulose (9). The value for the capacity of ECTEOLA-cellulose may include, however, entrained deoxyribonucleate which was eluted in the procedure used to estimate the capacity of Mg IRC-50 as described above. In the experiments to be described, the amounts of deoxyribonucleate chromatographed were less than one-fourth, and usually about

PAGE 31

24 TABLE III Estimation of maximum capacity of Mg IRC-50 for deoxyribonucleate 3 In both _experiments 28 x 2 cm columns which contained 88 cm of packed resin were used. The sample was added as a 0.17. solution of deoxyribonucleate prepared in 0.05 M Mg(OAc)2 pH 7.2. Conditions Deoxyribonucleate added Milligrams Total absorbancy units Deoxyribonucleate eluted by 0.05 M Mg(OAc)2 Milligrams Total absorbancy units Capacity Milligrams per column Milligrams per cm3 of packed resin Experiment number 743 744 74 904 46 560 28 0.32 15 920 46 564 29 0.33

PAGE 32

25 one-seventh of the estimated maximwn capacity of the columns, except where "overload'' experiments explicitly were performed. Effect of certain variables on the interaction Temperature --The influence of temperature on the interaction between deoxyribonucleate and Mg IRC-50 is illustrated in Fig. 4. Deoxyribonucleate was eluted by a much lower salt concentration from the column operated at 30 C than from the column equilibrated and 0 operated at 50 C. In contrast to the greater ease of elution custom-arily observed when low molecular weight compounds such as amino acids 3 (28) and purines and pyrimidines are chromatographed at higher temperatures, the deoxyribonucleate is more strongly bound by the Mg IRCS0 at the higher temperature. With small molecules, variations in temperature may affect ionizable groups of the chromatographed compound, of the buffer, and of the resin. The contribution of secondary valence forces to the overall interaction may also be altered. However, the effect of temperature on the chromatography of deoxyribonucleate using Mg IRC-50 appears to be dominated by other factors. The greater retardation observed at the higher temperature might best be explained in terms of effects on the physical structure of the resin or deoxyribonucleate. Heating may increase the number of effective binding site per unit area of the surface of the resin by increasing the flexibility of the cross-linked polymer chains, thereby allowing the aliphatic matrix of the resin to swell.-Higher temperatures may also reduce the degree of hydration of the two polymers, a factor which also would be expected to facilitate their mutual interaction. Elevated temperatures

PAGE 33

i "'~ 30c ;---en t; N OIO 2 6 MG DNA (856-X. RECOVERY) ,' a) z a:: 50 IOO l50 200 0 j Q21 8 50C ct 0
PAGE 34

2 7 may induce still other structural alterations favorable to interaction such as changes in the extent of intramolecular hydrogen bonds in certain regions of the molecules of deoxyribonucleate. !! --The interaction between deoxyribonucleate and Mg IRC-50 is not appreciably sensitive to the preciae value of pH of solutions of Mg(OAc)2 over the range, pH 7.2 to pH 8.1, where most of the chromatographic experiments were performed. However, deoxyribonucleate passes without apparent retardation through columns of Mg IRC-50 which are equilibrated and operated with solutions of Mg(OAc)2 of pH 6.1. A typical experiment performed at pH 6.1 is illustrated in Fig. 5!. It ie reasonable to ascribe the absence of interaction at 0.05 M Mg(0Ac) 2 pH 6.1, largely to the protonization of negative groups of the resin or the deoxyribonucleate or both. In this pH region, the carboxyl groups of Amberlite IRC-50 exhibit some buffering (29); the effect on the nucleate is probably limited to the secondary phosphoryl groups, whose pK is about 6.5 (30). It is possible that examination of the interaction at values of pH between 6.1 and 7.2, as well as at different molarities of Mg(OAc)2 would permit a decision as to whether the observed pH effect depend.a more on changes in the nucleate than on changes in the resin. -++ Ions other than Mg --In the interaction between IRC-50 and -++ deoxyribonucleate, Mg may be regarded as a counter ion which at low concentrations is somehow responsible for the mutual affinity between both polymeric moieties. -++ Higher concentrations of Mg destroy the specific interaction as indicated by elution of molecules with gradually varying composition. In order to determine whether deoxyribonucleate

PAGE 35

0 0) N >-(f) u :::) Zz <1 CD 0::: O:::j_ <1 0 (!) (\J ----.601A .4 I l Mg IRC-50 0.43 MG DNA 20f--IJ ~I I FRACT~ NUMBER (2.0 ML) 20 0.05M l'v1g(0Ac)2.pH6.l-j .20L C .10 B Mg IRC-50 2.6 MG DNA 10 20 30 40 1-NaOAc. pH6.4 [0.1M-i500 ML}--1.0MH Zn IRC -50 0.95 MG DNA (82% RECOVERY) FRACTION NUMBER (LS ML) 20 30 40 50 60 1-IM Zn(0Ac) 2 ,pH6.0 + IM Na0Ac. pH 6.0-----------i 70 80 Fig 5. The effect of certain variables on the chromatography of deoxyribonucleate from calf 132 thymus. The 29 x 0.9 cm columns of IRC-50, 11throug h 200," uaed for the experiments were operated at 30 C. (A) shows that deoxyribonucleate was not adsorbed to a column of Mg IRC-50 equilibrated and operated-with 0.05 M Mg(OAc)2 of pH 6.1. (~) illustrates that deoxyribonucleate is eluted from columns of Mg IRC-50 by low concentrations of NaOAc. () shows that 1 M NaOAc, but not 1 M Zn(OAc) 2 elutea deoxyribonucleate from columns of Zn IRC-50. N 00

PAGE 36

2 9 can be eluted by other ions, a specimen from calf #32 thymus was adsorbed to a column of Mg IRC-50 under the usual conditions, but the eluent was produced by adding 4.7 M NaOAc, pH 7.0, to a 250 ml mixing chamber initially filled with 0.05 M Mg(OAc)2 pH 7.3. The concentration of Na+ originally chosen for the reservoir was high in order to compensate for the lower eluting power expected for a monovalent cation. However, 90% of the deoxyribonucleate was eluted by the time three hold-up volumes were collected. The unexpected prompt elution by NaOAc was confirmed by an otherwise similar experiment in which an eluent of pH 6.4 was produced by adding 1 M NaOAc to O.l M NaOAc. In this case (illustrated in Fig. 5~) 1047. of the sample was recovered in less than two hold-up volumes. Thus, NaOAc, on a molar basis, was more than twice as effective as Mg(OAc)2 in eluting deoxyribonucleate from Mg IRC-50. Although it remains to be determined whether elution with NaOAc will provide a series of fractions of gradually varying composition, it is noteworthy that elution from Mg IRC-50 by sodium salts has been utilized to effect a partial separation of deoxyribonucleates from!-coli and 6 Pneumococcus. Although Na IRC-50 does not bind deoxyribonucleate at 0.2 M 7 NaOAc, pH 6, exploratory experiments have shown that divalent cations ++ other than Mg are able to mediate the interaction between deoxyribo-nucleate and IRC-50. 8 Thus, deoxyribonucleate is bound by Ba IRC-50 6R. D. Hotchkiss and L. Mindich, personal conmrunication (1959). 7c. P. Crampton, unpublished observations (1956). 8unlike Mg IRC-50, the Mg form of Dowex 50-X2 (sulfonated poly styrene beads, "through 20011 mesh) did not bind an appreciable amount

PAGE 37

30 9 at O.l M Ba(OAc) 2 pH 6.7, but is not eluted by l M Ba(OAc) 2 pH 6. Similarly, all of the deoxyribonucleate in a solution of Zn(OAc) 2 was adsorbed to a column of Zn IRC-50 which had been equilibrated with 0.05 M Zn(OAc) 2 pH 6.0. It will be recalled, in contrast, that deoxyribonucleate was not bound by Mg IRC-50 at pH 6.1. Moreover, none of the deoxyribonucleate was eluted by l M Zn(OAc) 2 of pH 6.0 as shown in Fig. 5. + In view of the effectiveness of Na in bringing about elution of deoxyribonucleate from Mg IRC-501 a solution of l M NaOAc, pH 6.0, was next pas1ed through the column. As can be seen, 827. of the applied material was thereby recovered. of deoxyribonucleate at 0.05 M Mg(OAc)2, pH 7.5. Similarly, the Mg form of Amberlite IR-120, XE-69 (crushed particles of sulfonated polystyrene, "through 325" mesh) did not bind the nucleate even when columns were operated at o C or at 5 C when the molarity of the Mg(OAc)2 was reduced to 0.01 M. The Zn form of the XE-69 resin failed to bind deoxyribonucleate at 5 C when equilibrated with 0.05 M Zn(OAc)2 of pH 6 These experiments were performed in an effort to find an ion exchanger for the chromatography of deoxyribonucleate which would equilibrate with the initial eluent more rapidly than IRC-50. However, the same factor responsible for the rapidity with which sulfona.ted polystyrene reaches equilibril.Ull, namely, the predominantly ionic character of the linkage between aulfonate and cation, may underlie the inability of this resin to form a stable complex with deoxyribonucleate. The Mg form of carboxymethyl-cellulose, a resin with reactive sites nearly identical to those of IRC-50, also failed to bind deoxyribonucleate when a column ws operated with 0.05 M Mg(OAc)2, pH 7.4, at 30. It is possible that the carboxyl groups of carboxymethyl-cellulose are not disposed in a specific geometric configuration conducive to the formation of stable complexes with deoxyribonucleate. Moreover, carboxymethyl-cellulose lacks the hydrophobic matrix and methyl side chains which in IRC-50 may reinforce linkages involving divalent cat ions. The poaaible importance of the methyl groups attached to the hydrophobic matrix of IRC-50 is emphasized by an unpublished experiment of Crampton, Moore, and Stein (1956) which indicated that the Mg form of polyacrylic acid (Amberlite XE-112) did not bind deoxyribonucleate under conditions where polymethacrylic acid did. 9c. F. Crampton, s. Moore, and W. H. Stein, unpublished observations (1956).

PAGE 38

31 The results discussed thus far suggest that under certain condition, there is formed a ternary complex involving the negatively ++ charged carboxyl groups of IRC-50, a divalent cation such as Mg ++ -t+ Zn or Ba and the negatively charged phosphate groups of deoxy-10 ribonucleate. The observation that Zn IRC-50 binds nucleate at pH 6, while Mg IRC-50 does not, may be related to the fact that Zn* forms -t+ stronger chelation complexes than Mg (31). If at certain critical salt concentrations, the bonds participating in complex formation are weakened, or salt becomes plentiful enough for the formation of stable binary complexes, the nucleate would be desorbed from the resin. Whether a coIIIIIOn mechanism, depending entirely on the ionic strength of the eluent, is responsible for elution by both NaOAc and Mg(OAc)2 or whether there are additional effects such as cation binding, cannot as yet be decided. In any event, the desorption process appears to be specific and the cation concentration required for elution depends on the composition of the nucleate molecule. It is reasonable to suggest that binding sites on deoxyribonucleate containing guanylic and cytidylic nucleotides, or in the vicinity of these residues, have a much lower affinity for Mg IRC-50 than other locations on the deoxyribonucleate polymer. Recent theoretical (32) and experimental (33) studies suggest that guanine-cytosine pairs have a higher electron density than adenine-thymine pairs. This could cause a i+ preferential binding of cations such as Mg at regions that are lOThie interaction differs from that of Mg or Ba IRC-50 with biatones (22) where the negatively charged carboxyl groups of the resin and the protonated side chains of these basic proteins are probably joined by electrostatic linkages. During the adsorption of histones, the cations originally present are presumably exchanged for the positively charged groups of lysine and arginine.

PAGE 39

32 abundant in guanine and cytosine. Such regions would thereby become less negatively charged, and consequently would exhibit less affinity ++ for the resin than regions not bearing a Mg Similar considerations may underlie the specific elution of deoxyribonucleate from other cat-+ ionic adsorbants by sodium salts (5, 8) since Na may also be bound by deoxyribonucleate (34, 35). It is worth noting in this connection that low concentrations of MgC12 markedly reduced the concentration of NaCl required to elute deoxyribonucleate from histonechloroform gels (5). Flow rate --For optimum resolution during the analysis of small organic compounds by ion exchange chromatography, the rate of elution must be cot1111ensurate with continuous equilibration between the adsorbed solute and the solute free in solution. However, moat macromolecular substances are usually completely adsorbed to, or desorbed from, the chromatographic adsorbent. That is, the range of molarities (or pH, etc.) over which polymers will exhibit a finite distribution between the free and bound forms is small, and frequently the Rf of the compound will change abruptly from Oto l (29). The suddenness of thia change is suggestive of a cooperative phenomenon, and may in fact be viewed as deriving from the need to sever simultaneously the many bonds between the adsorbent and the multivalent macromolecule in order for the solute to be eluted. 11 11Becauae of this "all or none'' phenomenon, the chromatographic behavior of many macromolecules is frequently independent of column length (29). This was found to be the case when deoxyribonucleate was chromatographed by uae of a 72 x 0.9 cm column inatead of the usual 30 x 0.9 cm colwm1. No significant change in the chromatographic pro-. file was observed except a slight, unexpected sharpening of the pattern.

PAGE 40

33 Experiments were performed in order to determine the effect of flow rate on the behavior of deoxyribonucleate when chromatographed by use of Mg IRC-50. When deoxyribonucleate from calf thymus was chromatographed at flow rates varying from 4 ml per hour to 26 ml per hour, small but significant changes in the elution patterns were observed. At the lowest rate, there occurred a continuous elution of small amounts of ultraviolet absorbing material before the main portion of the nucleate emerged. However, elution of the bulk of the nucleate always required at least 0.2 M Mg(OAc)2 If, therefore, the material leaching slowly from the column was deoxyribonucleate, it would represent a small amount of material in equilibrium with the bound nucleate at concentrations of Mg(OAc)2 below 0.2 M. The effects of this equilibrium would be observable only at very low rates of flow. In a subsequent section will be discussed the observation that a small amount of ultraviolet adsorbing material referred to as 11heat peak" can be recovered from columns, after the greatest part of the deoxyribonucleate has been eluted, by raising the temperature of the 0 0 0 column from 30 to 45 or 50 C, and simultaneously changing the eluent to l M Mg(OAc)2 After the column had been operated at rates of 20 to 40 ml per hour, about 2 to 3% of deoxyribonucleate from calf thymua was recovered in this manner. However, at lower rates of flow, more material was recovered in the 11heat peak," 4% at 7 ml per hour, and 9% at 4 ml per hour. To determine if the length of time during which the sample remains adsorbed to the resin results in any physical or chemical change in the nucleate which might affect its chromatographic properties,

PAGE 41

34 2.4 mg of the apecimen from calf #32 thymus wem applied to a column which was then held thirty-six hours before administration of the eluent. This delay had no apparent effect on the chromatographic pattern, except that 3.n. of the recovered ultraviolet absorption was eluted within one bold-up volume. It is difficult to assess the aignificance of this observation since, occasionally, emall amounts of ultraviolet absorbing materials appear in this region of the chromatogram in the absence of any known modifications with respect to the sample or to the chromatographic procedure. Particle size of resin --When two chromatograms of deoxyribonucleate from calf thymus were run under conditions which were identical in all respects with the exception that one column was prepared from resin which had passed a 200 mesh screen, while the other waa prepared from "through 325" mesh material, quite similar chromatographic profiles were obtained. It is possible that the higher proportion of small particles in the "through 325" material presents a greater effective surface area resulting in a higher capacity for the nucleate. Chromatographic behavior of deoxyribonucleates from rat liver, !hrlich ascites tumor cells, and Pneumococcua The essential reproducibility of the chromatographic patterns given by specimens of deoxyribonucleate prepared fran calf thymus by the several procedures noted in the experimental section has frequently been verified. It was of interest to determine if specimens from other sources likewise are adsorbed by Mg IRC-50 at low concentrations of Mg(OAc)2 and if so, whether they can be eluted. The deoxyribonucleates

PAGE 42

35 of two other mammalian sources were studied. Figs. 8 and 10 show that specimens of deoxyribonucleate from rat liver and Ehrlich aacites tumor cells are, in fact, bound by Mg IRC-50 and can subsequently be eluted with an eluent of gradually increasing Mg(OAc)2 concentration to provide chromatographic profiles very much like those obtained from calf thymus nucleate. The recovery of these three deoxyribonucleates, based on ultraviolet absorption, has usually been between 80% and 100%. When very small samples of deoxyribonucleate from rat liver were chromatographed by use of 30 x 0.9 cm columns, the recoveries were lower (62% and 757.). A somewhat different result was obtained when the chromatographic behavior of the deoxyribonucleate from Pneumococcus was studied. 12 In the initial experiment, specimen 149 was chromatographed on a preparative column of Mg IRC-50 (Fig. ~). This specimen possessed transforming activity and was estimated to contain 26% ribonucleate and 13 29% protein. Of the total ultraviolet absorption applied to the column, 35% was unretarded at 0.05 M Mg(OAc)2 and only 10% was eluted by means of the eluent of gradually increasing Mg(OAc)2 concentration. It was found, however, that 10"/o of the ultraviolet absorption of the sample could be recovered from the resin in the "beat peak" by abruptly 12we wish to express our appreciation to Dr. R. D. Hotchkiss for supplying us with several specimens of deoxyribonucleate from this source. The specimens were prepared in his laboratory by lysis of the bacterial cells with 0.15% deoxycholate, followed by several treatments with chloroform and isoamyl alcohol, as well as with ribonuclease. 13Performed by procedures to which reference is made in the manuscript noted in Footnote 1.

PAGE 43

36 0 3 Q'L52 MglRC-50 (THRU 200), 29X2 CM A .46 0 2 l0 5 MG DNA ( 149) ll \ 0 1 :::i FRACTIO\/ NO 40 60 140 o (5 ML PER FRACTION) l 220 240 26 j--Mg(OAc)2 [ 0 05M--{IOOOML}-0.4M--\---I.OM].30 -I-I.OM Mg(OAc~,45 en ::, z :::i 0 ID N > u 0 2 0 1 MglRC-50 (THRU 200), 30X0. 9 CM 2.2 MG DNA(l49 R) B FRACTION NQ 20 (3.2ML FfER FRACTIO\/) 40 120 140 l60 Mg(OAcl.! g).05M---{500 ML}-0.4M], 30 -----1-LQM Mg(OAc)2,50 0 en CD ex 0.2 Q 0.34 MglRC-50(THRU 200), 32X0.9CM -1.9 MG DNA (151) C 0.1 FRACTIO\J NO. 40 (2 ML PER FRACTIO\/) 60 140 160 180 200 220 240 260 280 I-----Mg(~c) 2 [ 0 .05 M-{500 ML}-0.4 M]. 30 ---~LO M Mg(OAc~, 50 Fig. 6. Chromatography of specimens of deoxyribonucleate from Pneumococcus by use of columns of Mg IRC-50. Since slightly different conditions of elution were used for each experiment, the chromatograms have been plotted on different scales in order to facilitate a comparison of the relative distribution of ultraviolet absorbing materials. Specimen 149, 149R and 151 were used to obtain (A), (B) and (C), respectively. -

PAGE 44

37 increasing the temperature of the water jacketed column from 30 to 0 45 C and simultaneously applying l M Mg(OAc)2 The analyses presented in Table IV suggest that the bulk of the contaminating ribonucleate was contained in material which passed directly through the resin (Fraction II), whereas protein was not confined to any single fraction. Results conmunicated to us by Dr. R. D. Hotchkiss indicate that while Fractions IVa, IVb, IVc, and V exhibit transforming activity, the unretarded material was devoid of such activity. Further purification of this same preparation by two treatments with chloroform and isoamyl alcohol yielded a product (l49R) which on chromatography by use of an analytical column gave practically no unretarded constituents which absorbed ultraviolet light (Fig. 6!). The distribution of eluted material in the remaining portions of the chromatogram, however, remained unchanged. That is, both before and after repurification, 50% of the eluted deoxyribonucleate was found in the gradient elution region of the chromatogram, and 50% in the "heat peak" (Table V). Two additional pneumococcal specimens (152 and 152A), derived from the same original preparation, differed in their physical character when precipitated from solution in 0.85% NaCl by an equal volume of ethanol. Specimen 152 was fibrous, while 152A was jelly-like. When each of the samples was chromatographed, 36% of the total absorbancy eluted, exclusive of unretarded material, was present in "heat peak." Another preparation (151), however, yielded a chromatographic pattern which, as shown in Fig. 6, was similar to that obtained from the mammalian nucleates, except for the presence of unretarded material.

PAGE 45

38 TABLE IV Distribution of constituents among chromatographic fraction of deoxyribonucleate from Pneumococcus The sample chromatographed contained 130 abeorbancy unit1, or about 10.6 mg of deoxyribonucleate. It gave orcinol color equivalent to approximately 2.6 mg of ribonucleate, and gave ninhydrin color equivalent to 2.9 mg of hiatone Fraction B from calf thymus. '11le orcinol and ninhydrin reactions were performed by procedure referred to in the manuscript noted in Footnote 1. The result, are expresaed as per cent of the constituent in the sample recovered in the fractions shown. Constituent Abeorbancy units Ribonucleate Protein Fraction number II V 21 51 20 10 4 24

PAGE 46

39 TABLE V Recovery of "heat peak" material from deoxyribonucleate preparations from various sources Source Calf thymus Rat liver Ehrlich ascites tumor cells Pneumococcusb Preparation number 32 32 11 11 7 9 16 149 149R 151 152 152A Per cent of added absorbancy units recovered in "heat peak" 3 3 5 7 3 2 3 50 49 6 36 32 Total column recover ya 86 96 92 62 88 97 105 56 24 67 27 33 8Absorbancy units recovered as per cent of abaorbancy units added. bBecause of wide variation of ribonucleate content and total recovery of absorbancy encountered with samples from Pneumococcus, the per cent recovered in "heat peak" is expressed, for these specimens, as per cent of the total absorbancy units eluted, exclusive of that which was unretarded on Mg IRC-50 at 0.05 M Mg(OAc)2 pH 7.2 to 7.4.

PAGE 47

40 As shown in Table V, 2 to 370 of the nucleate from calf thymus and ascites tumor cells, and 5 ton. of that from rat liver is eluted in the "heat peak" position. Further studies should clarify whether the unusual chromatographic behavior of the pneumococcal deoxyribonucleates can be attributed to the use of preparative procedures very different from those employed in the isolation and purification of the nucleates from the other sources studied. In this connection, however, a specimen of calf thymus deoxyribonucleate which was prepared in the laboratory of Dr. R. D. Hotchkiss by two deproteinizations with chloroform and two treatments with Duponol, yielded a chromatographic pattern in all ways like that given by the specimens of deoxyribonucleate from calf #30 and #32 thymus glands. Additional evidence that "heat peak" material is not entirely an artifactitious product arising by action of the resin on the nucleate is provided by the observation that such material was absent from rechromatographed fractions of a specimen from rat liver (see Fig. 81!).14 It must be recalled, however, that the amount of material in the "heat peak" increased simply by running the column at a very low rate of flow. Moreover, rechromatography of "heat peak" material from the nucleate of ascites tumor cells proved disappointing, since when a sample of 0.1 mg was applied to a fresh column 14rn one experiment, two fractions isolated by ethanol precipitation from an "overload" chromatogram of deoxyribonucleate from Ehrlich ascites tumor cells were combined and rechromatographed under non-overload conditions. One of the fractions contained material that originally emerged unretarded, while the other fraction contained material that was originally eluted gradually between 0.2 Mand 0.4 M Mg(OAc)2 On rechromatography, very little material emerged unretarded. However, about as much material was eluted in the "heat peak" as was eluted during gradient elution.

PAGE 48

41 of Mg IRC-50, nothing was eluted by the original conditions, and lees 0 than 10% was eluted by l M NaCl at 45 C. In this experiment, a sig nificant amount of material was recovered only by passing 0.5 M NaOH through the colwnn. Thus, while a part of the material in "beat peak" may be a genuine component of the nucleate specimen, a part may also be an artifact. In a subsequent section, which is concerned with studies of the incorporation of radioactive precursors into the deoxyribonu cleate of Ehrlich ascites tumor cells, further observations regarding "heat peak'' and its possible biological significance will be presented. The "heat peak" material described here may be related to the fractions of deoxyribonucleate obtained from various sources by Brown and Brown (8). The fractions comprised 2 to 10% of the preparations, and were removed from columns of histone-cellulose by an eluent of high molarity and pH. These authors concluded that the material, which had a base composition similar to the original deoxyribonucleate, was an artifact produced by the column. A characteristic feature of the chromatography of deoxyribonu cleate from Pneumococcus is the low recovery (24 to 67%) of ultraviolet absorbing material from columns of Mg IRC-50. Rechromatography of fractions of deoxyribonucleate from calf thymus and rat liver In order to study factors which determine the points of elution of molecules of deoxyribonucleate, material eluted at different regions of the Mg(OAc)2 concentration gradient was adjusted to 0.05 M Mg(OAc)2 combined, a nd rechromatographed on a fresh column of Mg IRC-50. Fig. 7B

PAGE 49

42 shows an analytical chromatogram of two fractions of deoxyribonucleate which had been eluted from the preparative column illustrated in Pig. 7A and recovered by precipitation by ethanol. Similar experiments in which analytical columns were used both for isolation of fractions of deoxyribonucleate from rat liver and for their rechromatography are shown in Figs. 8! and 8!. In this case, the fractions had not been recovered for rechromatography by precipitation with ethanol, but were adjusted to 0.05 M Mg(OAc)2 by dialysis. The results of the rechromatography experiments indicate that although there is some displacement in absolute position, the relative positions of elution are inherent properties characteristic of the molecules of deoxyribonucleate. That is, the gradual elution of successive fractions appears not to be governed by a heterogeneity of binding sites in the resin, or by a reduction in the capacity of the adsorbent to which all molecules of deoxyribonucleate are equally sensitive. It would be desirable to extend these experiments by recbromatographing mixtures of fractions isolated from radioactive and from non-radioactive specimens of deoxyribonucleate. Effect of sample load Although the relative positions of elution of the rechromatographed fractions of deoxyribonucleate from rat liver are similar to those of the original material (Pig. 8), the shift in absolute positions indicates that competition and displacement effects play a role in the fractionation process. The molecules in Fraction III of Fig. 8! must have the greatest affinity for the resin, since they require the

PAGE 50

:::t 0 (1) (\J en :::::> z :::t 0 (0 >u z <( CD en CD <( 1.0 0.8 0.6 0 .4 0 2 MglRC-50(THRU 200), 30X2 CM 30MG DNA EFFWENT ML 25 0 5 0 0 750 1000 0.4 0.3 0.2 0 1 ------Mg(0Ac)2 [ 0 .05 M--{1000 M L}-0.4 M], 30 M g lRC-50 (THRU 200),29X0.9 CM FRACTIONS na-sz:m: FROM ABOVE CHROMATOGRAM I .......-:::--:= ., ----------=-------~--------~~-------~ EFFLUENT ML 40 80 120 160 200 A B 11------Mg(0Ac)2 [ 0.05 M--{250 ML}--0. 4 M]. 30 --1 Fig. 7. Rechromatography of deoxyribonucleate from calf thymus by use of columns of Mg IRC-50, "through 200." 'n>.e sample added to the 30 x 2 cm preparative column used to obtain (A) was specimen JOA from calf thymus. About 0.6 mg of F'Yaction II and 1.2 mg of Fraction VIII were mixed-to provide the sample added to the 29 x 0.9 cm analytical chromatogram illustrated ln (_!!). ""'

PAGE 51

020 015 QIO A 2.03 MG OF TOTAL DNA 0 b../ \.,,., ..., >-(1) N QI() ~---;---;~5o~~~~~~~~IO~O~__./"~;;;;==~=====:~~~==~====:~===~==~ a:: z Q0S 0~ (/) CIIO~--;-------;~~~;;:--;;;;~:::::=~~==~:==~ 150 zoo 340 8 (9l8 .. RECOVERY) FRACTIONS I 8 ill FROM ABOVE CHROMATOGRAM, 0.35 MG DNA (74.8,. RECOVERY) 100 150 2SO oost! C 0.43 MG OF TOTAL DNA (6Z.O .. RECOVERY) -~-/~--~" J\ FRACTION NUMBER (197-205 IU IOO 150 zoo 290 Mg(OAc~. pH7.2 ~.05M------{500ML)----0.4M], 30C IM Mg(OA<, 509C AT I Fig. 8. Chromatography of deoxyribonucleate from rat liver by uae of 29 x 0.9 cm coluuma of Mg IRC-50, "through 325." The total specimen from rat #11 liver was used to obtain(!) and(), while a mixture of Fractions I and III from(~) served as the sample for(!!_). f:

PAGE 52

45 highest salt concentration for their elution. Molecules having a strong affinity for Mg IRC-50 may be able to compete effectively for binding sites on the resin, and thereby to displace the less strongly bound molecules to positions further down the column. The positions of elution of different molecules of deoxyribonucleate may therefore be related to their initial positions on the column. When large samples are chromatographed, deoxyribonucleate may be adsorbed to the resin sites throughout most of the length of the column. Under such conditions, material might be expected to emerge earlier than when smaller samples are chromatographed. For example, in Fig. 8, where 0.43 mg of deoxyribonucleate from rat Ill liver was chromatographed, elution did not begin until a higher Mg(OAc)2 molarity was reached than in the experiment shown in Fig. 8!, where a 2 mg sample of the same specimen of deoxyribonucleate was chromatographed. Whether the concentration of deoxyribonucleate in the sample added to the column bas an effect on the elution pattern has not been studied systematically. It must be noted that the ,ample solution used in Pig. 8! (l mg per ml) was more concentrated than that used in Fig. 8 (0.2 mg per ml). Toe possibility that deoxyribonucleate molecules having a low affinity for the resin are displaced by more strongly bound molecules wa1 te1ted further as follows. Calf #32 thymus nucleate was added to columns of Mg IRC-50 in amounts sufficient to exceed the capacity of the resin. In the two experiments presented in Table VI, 62% and 36% of the specimens added were eluted under the starting conditions (0.05 M Mg(OAc)2 pH 7.2). As shown, the material which emerged abruptly (Fractions I, II, and III of experiment 743 and Fractions I and II of

PAGE 53

46 TABLE VI Base composition of chromatographic fractions of deoxy ribonucleate from calf 132 thymus obtained by overloading columns of Mg IRC-50 To the 28 x 2 cm column used for experiment 743 was added 74 mg of deoxyribonucleate, an amount which exceeded the capacity of the resin by 621 Experiment 766 was performed by use of a 31 x 0.9 cm column to which was added 22 mg of deox yribonucleate, an a mount which exceeded the capacity of the resin by 36 % Experiment nwuber Description of fractions 743 766 Fraction number I II Ill IV I II III Per cent of deoxyribonu-10 11 8 28 18 18 52 cleate in fraction Molarity of Mg(OAc)2 0.05 0.05 0.05 0.05 0.05 0.05 0.4 used for elution Basesa Thymine 25.2 25.9 --27.3 25.7 27.0 28.7 Guanine 25 6 24.5 --23.2 23.5 23.l 20.4 C ytosine 22.4 22.0 --22.4 23.5 21.8 20.7 Adenine 24 5 26.2 --25,5 25.l 26.l 29.2 5-Methylcytosine 2.5 1.5 --1.6 2.2 2.0 0.9 aAs moles of base per 100 moles of total recovered bases.

PAGE 54

47 experiment 766) contained more guanine and cytosine than total deoxyribonucleate (see Tables II and VII), and thereby resembled deoxyribonucleate which is eluted first by the lower concentrations of Mg(OAc)2 under "non-overload" conditions,such as Fraction I in Table II. On the other hand, the material not eluted by 0.05 M Mg(OAc)2 but subsequently eluted as in experiment 766 by means of 0.4 M Mg(OAc)2 has a composition like that of fractions eluted by the higher concentrations of Mg(OAc)2 under ''non-overload" conditions, such as Fractions II and III in Table IL It therefore seems likely that the same molecules of deoxyribonucleate which are eluted early by increasing concentrations of Mg(OAc)2 from a column bearing a small load of deoxyribonucleate, are among the first to emerge when the capacity of a column is deliberately exceeded. Even when the sample load is excessive, interaction of Mg IRC-50 with deoxyribonucleate remains demonstrably specific. Base composition of fractions of deoxyribonucleate from calf thymus and Ehrlich ascites tumor cells As discussed above, the interaction of deoxyribonucleate with Mg IRC-50 has provided a means for eluting fractions of different composition from columns to which specimens of total deoxyribonucleate were adsorbed. The studies with overloaded columns suggested an equally effective alternative fractionation technique, as discussed in connection with Table VI. A more complete chromatographic fractionation of deoxyribonucleate from calf 132 thymus is illustrated in Fig. 9. The base compositionsof fractions of the unadsorbed deoxyribonucleate and of the fractions eluted gradually by increasing concentrations of Mg(OAc)2 are shown in Table VII.

PAGE 55

0 r U N 0!!O z Cf) d5 ::) 040 a:: z 0 030 Cf) :;:j 020 Mg IRC-50 (THRU 200) 2 X 28 CM, 30C 75 MG DNA ( 89.6% RECOVERY) @ OJOlb_~l f-m: ll-j ,__.. __/,i ------=t: .... _.,,.-~ --EI L DI I ix I x ::::::::J-_ I TUBE NUMBER (!Or.a. PER TUBO 100 I ~ 200 2!10 :500 f-005 M Mg(0Aclz, pH 7.2---,-----Mg(0Ad-2 pH7.2 [0.05M----{ 3000 ML)-----0.4 Mt-----------, Fig. 9. Preparative chromatography of deoxyribonucleate from calf #32 thymus by use of overload conditions. In the experiment shown, the capacity of the 28 x 2 cm column of Mg IRC-50 was exceeded so that 59% of the added deoxyribonucleate was eluted by 0.05 M Mg(OAc)2 The nominal temperature at which the column was operated was 30 C. The unusual fluctuations in the elution pattern are probably the result of irregular variations which were greater than+ 0.2 C. g;

PAGE 56

TABLE VII Base com;e2sition of chromatogra2hic fractions of deoxyribonucleate from calf #32 thymus The fractions were derived from the experiment illustrated in Fig. 9. The percentage in paren-theses which follows each fraction nwnber represents the portion of the deoxyribonucleate of the sample recovered in the respective fraction. The results are expressed as moles of base per 100 moles of total recovered bases. Fraction number Unfractionated total Base 1(3%) 11(12%) 111(11% ) IV(29"/4) V(4%) VI(5% ) VII(9%) VIII(lOo/ o ) IX(4% ) X(2% ) deoxyribonucleate Thymine 23.3 26.3 27.4 28.3 25.3 30.3 30.7 30.3 30.4 32.l 27.9 Guanine 27.l 24.2 23.3 22.0 25.0 20.4 20.1 20.0 19.4 19.6 22.8 \D Cytosine 23.5 22.6 22.l 21.3 22.l 19.8 18.9 19.0 19.2 18.l 21.l Adenine 22.9 25.0 25.3 26.4 25.3 28. 6 29.4 29.9 30.3 29.7 26.8 5-Methylcy-tosine 3.1 1.9 2.0 2.0 2.3 0.9 0.9 0.8 0.7 0.6 1.5 A+ T8 0.86 1.05 1.11 1.20 1.02 1.44 1.51 1.51 1. 54 1. 62 1.21 G + C + M C ~e ratio of the molar quantities of adenine plus thymine to guanine plus cytosine plus 5-methylcytosine.

PAGE 57

50 The successive fractions show variations in composition 1imilar to fractions previously obtained by Chargaff, Crampton, and Lipshitz (4, 5) and by Brown and Brown (8), whose methods were based on the interaction of deoxyribonucleate with basic proteins. Cbargaff .!!. al. noted that the ratio of 5-methylcytosine to cytosine in the successive fractions varied markedly, and concluded that the replacement of cytosine by the 5-methyl derivative does not occur at random. As shown in Table VII, a 5-fold change in the 5-methylcytosine content of the chromatographic fractions occurs with less than a doubling of cytosine or guanine, which confirms the results of Chargaff !!_ al, Chromatograms of deoxyribonucleate from Ehrlich ascites ttuDOr cells obtained under "non-overload" conditions (EA fl) or overload conditions (EA 17) are shown in Figs. 10 and 11, respectively. The base compositions of the isolated fractions are presented in Tables VIII and IX. Both techniques yielded fractions which show the same trends in composition as the fractions from calf thymus deoxyribonucleate. However, the extreme fractions of the deoxyribonucleate from tumor cells differ far less than the corresponding fractions shown in T ables II, VI, and VII. Incorporation of Precursors into Fractions of Deoxyribonucleate from Ehrlich Ascites Tumor Cells 14 Preliminary experiments with C -formate The ability of Mg IRC-50 to separate chemically different molecules of deoxyribonucleate suggested that information regarding the synthesis of these molecules might be provided by studies of the initial

PAGE 58

~ I 1 1 i ~KJ:=!:~ L I f [\, I :n:::71 m--b 50 IOO l50 200 I M;J(OAcl2 [0.05M,pH7.4 (500ML}-0.4M,pH7.9], 30C I I Fig. 10. Chromatography of deoxyribonucleate from Ehrlich ascites tumor cells (EA #1) by use of M g IRC-50. V,

PAGE 59

I >-(\J u (/) z :::>
PAGE 60

53 TABLE VIII Base composition of chromatographic fractions of deoxyribonucleate from Ehrlich ascites tumor cells (EA #1) 'nle fractions were derived from the experiment illustrated in Fig. 10. The percentage in parentheses which follows each fraction number represents the portion of the deoxyribonucleate of the sample recovered in the respective fraction. The results are expressed as moles of base per 100 moles of total recovered bases. Fraction number Unfractionated, Base total 1(197.) II(247.) III ( 341.) deoxyribonucleate Thymine 27.7 28.l 29.2 28.8 Guanine 22.2 21.1 20.6 21.3 Cytosine 21.0 21.7 19.6 19.5 Adenine 28.8 28.8 30.0 29.5 5-Methylcytoaine (0. 4) (O. 3) 0.7 0.9 A+ T8 1.30 1. 32 1.45 1. 40 G+C+MC aThe ratio of the molar quantitiea of adenine plus thymine to guanine plus cytosine plus Smethylcytosine.

PAGE 61

TABLE IX Base composition of chromatographic fractions of deoxyribonucleate from Ehrlich ascites tumor cells (EA #7) The fractions were derived from the experiment illustrated in Fig. 11. The percentage in parentheses following each fraction number represents the portion of the deoxyribonucleate of the sample recovered in the respective fraction. The results are expressed as moles of base per 100 moles of total recovered bases. Fraction number Unfractionated, Base total 1(9%) II(l2%) III(ll% ) IV(l7%) V(l9%) VI(l0%) VII(3%) deoxyribonucleate Thymine 27.2 28.2 27.9 28.8 29.6 30.7 29.0 28.4 Guanine 22.7 22.l 22.2 21.2 20.4 20.7 21.3 20.5 Cytosine 21.5 21. l 21.6 20.2 19.3 17.3 19.3 20.9 Adenine 27.4 27.9 27.6 29.2 29.8 30.2 29.4 29.3 5-Methylcytosine 1.1 0.7 0.7 0.6 0.9 1.2 1.0 0.9 -A+ Ta 1.21 1.28 1.25 l. 38 1.46 1.55 1.40 1.36 G+C+MC ~he ratio of the molar quantities of adenine plus thymine to guanine plus cytosine plus 5-methylcytosine. V1

PAGE 62

55 vivo rates of incorporation of radioactive precursors. Cultures of Ehrlich ascites tumor cells growing in the peritoneal cavities of mice were chosen for these studies because they present the following advantages. !) During the logarithmic phase of growth under natural in vivo conditions, the cells are less likely to lose viability than cells grown in vitro. Consequently, the possibility of an abnormal metabolism of their deoxyribonucleate is minimized and most findings can be construed with reasonable confidence to have a bearing on the synthesis of authentic deoxyribonucleate. l:) Populations of ascites tumor cells are relatively homogeneous in contrast to alternative biological systems in which there is rapid synthesis of deoxyribonucleate such as regenerating liver. The interpretation of any chemical or metabolic properties of fractions of the deoxyribonucleate of liver would be confused by the presence of several cell types which respond differently to hepatectomy, as was recently fowtd by Edwards~ al. (36). To be sure,~~ cultures of Ehrlich ascites tumor cells gradually become hemorrhagic. Although absent from erythrocytes, deoxyribonucleate is present in leucocytes, which may also infiltrate the cultures, but which rarely, if ever, exceed about 1% of the cell population. ~) Finally, of great practical importance, is the fact that deoxyribonucleate is readily obtained from these cells in good yield and in a highly polymerized 1 form by use of strong urea solutions. The denatured deoxyribonucleate provided by certain alternative extraction procedures (37) would not be suitable for fractionation by use of Mg IRC-50, although they could con tain constituents absent from the specimens examined in the present studies.

PAGE 63

56 Effect of non-radioactive formate --After the administration 14 of C -formate, it was necessary to allow time enough for a measurable amount of radioactivity to be incorporated into deoxyribonucleate. It was desired, however, to limit the duration of the incorporation period, so that only a small proportion of the total deoxyribonucleate would have the opportunity to replicate in the presence of the precursor. A brief period would tend to exaggerate any preferential uptake of the precursor into chromatographically different fractions. An incorporation period of ten minutes was chosen for the first experiment. Thia corresponds to about 1% of the time required for the cells to double in number (eighteen hours). 14 The C -formate injected is an efficient precursor of the 2 and 8 positions of the purine ring, and of the 5-methyl group of thymine. With these considerations in mind, the ten minute incorporation period was followed by an interval of 110 minutes in order to allow time for incipiently labeled molecules of deoxyribonucleate to be completed, to undergo secondary structural alterations, to combine with histone, or to acquire whatever other characteristics may be essential for insolubility in the urea medium used during isolation. In an effort to minimize further incorporation during the 110 minute interval, a 100-fold excess of unlabeled formate was injected ten minutes after the c14-formate. It was anticipated that this amount of unlabeled formate would dilute the labeled pools of the cells so that the radioactivity incorporated during the 110 minute interval would be insignificant compared with that incorporated in the preceding ten minutes.

PAGE 64

57 Accordingly, each of two tumor-bearing mice (EA #1) was injected 14 with O.l ml of buffered saline solution containing 5 curies of C -for-mate having a specific activity of 1 curie per mole. The mice were reinjected after ten minutes with 1 ml of buffered saline solution containing 500 moles of unlabeled formate. The tumor cells were har-14 vested 120 minutes after the injection of C -formate, and were proc-essed aa described in the experimental section. The twice-deproteinized deoxyribonucleate was found to contain appreciable amounts of carbon-14 in thymine, guanine, and adenine. Cytosine contained no appreciable radioactivity, while 5-methylcytosine was found to contain some counts, which, however, could have resulted from contamination by the highly labeled adenine as a result of the incomplete separation of these two bases (see Fig. 1). Therefore, it is still not known whether formate is a direct precursor of the methyl group of 5-methylcytosine. A 3.9 mg specimen of the deoxyribonucleate was eluted from an analytical column of Mg IRC-50 (30 x 0.9 cm) by use of an eluent of gradually increasing Mg(OAc)2 concentration (0.05 M to 0.4 M) to obtain the chromatogram shown in Fig. 10. The effluent was combined to provide three successive fractions from which the nucleate was precipitated by ethanol and hydrolyzed. The hydrolysates were analyzed by use of columns of Dowex 50 in order to estimate the base composition of the fractions and to separate the bases for the subsequent determination of specific activities. The base compositions have already been presented in Table VIII. Specific activities are given in Table X. Under the conditions of incorporation used in this experiment, it is apparent that all of the fractions attained similar specific activities.

PAGE 65

5 8 TABLE X In vivo incorporation of cl4_formate into fractions of deoxyribonucleate from lhrlich ascitea tumor cells (!A #1) The fractions were derived from the experiment illustrated in Pig. 10. The percentage in parentheses which follows each fraction number represents the portion of the deoxyribonucleate of the sample recovered in the respective fraction. The results are expressed as counts per minute per mole of base. Fraction number Unfractionated, Base total I(l9t) II (24% ) III(34'7.) deoxyribonucleate Thymine 372 400 404 422 Guanine 286 236 281 256 Adenine 226 191 195 204

PAGE 66

59 Effect of time --In view of the results of the preceding experiment, three additional experiments were performed in order to evaluate the efficiency with which the 500 moles of unlabeled formate 14 prevent further incorporation of C -formate into the tumor cell nucleate. In all three experiments, the results of which are presented in Fig. 12, an interval of seven minutes was allowed to elapse after 14 C -formate. However, the period of exposure to unlabeled formate was varied. In the first experimental group of three mice (EA #4), this period was 113 minutes. That is, 500 moles of unlabeled formate were 14 injected to each mouse seven minutes after 5curies of C -formate had been injected, and the tumor cells were isolated 120 minutes after the 14 injection of C -formate. In the second group of four mice (EA #5), 14 C -formate and the unlabeled compound were injected as for EA #4, but the cells were isolated twenty-five minutes after the injection of c14-formate. In the third group of three mice (EA #6), unlabeled for mate was not injected at all. That is, the tumor cells were isolated 14 promptly seven minutes after injection of C -formate. If the admin-istration of unlabeled formate completely and inmediately prevents further uptake of the labeled precursor, the specific activity of the bases in the deoxyribonucleate isolated from all three groups of mice would be the same or possibly would decrease with time because of the continued synthesis of deoxyribonucleate from diluted pools of precursors. As shown in Fig. 12, however, the incorporation of radioactivity into the purines continued slowly for 120 minutes despite the injection of a 100-fold molar excess of unlabeled formate. On the other hand, the specific activity of thymine did remain the same over

PAGE 67

w 5001 / e e THYMINE./ -----i 4001 I I I ~'->300 I II GUANINE""" > I II u 200 I --------\...ADENINE
PAGE 68

61 the intervals studied, suggesting that the non-radioactive formate injected did succeed in diluting pools of precursors of thia base. The differences among the curves obtained for the purines and for thymine could result from differences in the size of the pools of the respective precursors. Thus, the purines may be formed from large pools with which formate equilibrates slowly and incompletely, while thymine may be formed from small pools with which formate equilibrates rapidly and more completely. 14 Incorporation of C -formate into fractions of deoxyribonucleate As noted above, deoxyribonucleate containing reasonably large amounts of radioactive thymine is recoverable from tumor cells seven 14 minutes after the injection of C -formate. Since deproteinized deoxy-ribonucleate is soluble in the urea medium, this observation suggests the possibility that deoxyribonucleate may be insoluble in the urea medium even during the initial stages of replication. This observation is perhaps related to the recent finding that histones are con served as well as deoxyribonucleate during replication of the tumor cells (38). In this system, therefore, deoxyribonucleate at all stages of replication may be insoluble in the urea medium because it occurs combined with histone. At any rate, the finding indicated that studies of the heterogeneity in the uptake of precursors into fractions of deoxyribonucleate during even shorter intervals were feasible. Under these conditions, preferential labeling would be accentuated by the rapidity with which the specific activity of the pools of precursors of thymine changes.

PAGE 69

62 In the next experiment, therefore, tumor cells (EA #7) were collected from each of ten mice five minutes after the injection of O.l ml of buffered saline containing 3.7 moles of labeled formate with a specific activity of 13.7 curies per mole. A fifty-seven mg sample of deoxyribonucleate isolated from the cells was chromatographed by use of a preparative column of Mg IRC-50 (19 x 2 cm), with the results shown in Fig. 11. Under these overload conditions, 32% of the sample was eluted by 0.05 M Mg(OAc)2 Most of the remaining deoxyribonucleate was removed from the column by gradient elution with Mg(OAc)2 (0.05 M to O. 4 M). A further small portion, referred to as 11heat peak, 11 was eluted by 1 M Mg(OAc)2 when the temperature of the column was raised to 45 c. The column effluent was combined to provide seven fractions. The deoxyribonucleate which was recovered from the fractions by precipitation with ethanol was found to have the base compositions shown in Table IX. As discussed earlier, the extreme fractions differ far less than extreme fractions which have been obtained by similar methods from deoxyribonucleate from calf thymus. Despite the similar base compositions, however, significant differences were found with respect to the specific activity of the thymine in these fractions, as shown in Table XI. 'nle first three fractions comprise the nucleate eluted by 0.05 M Mg(OAc)2 under the overload conditions. The specific activity of the thymine in these three fractions is higher than that in the next three fractions which contain material recovered from the column by gradient elution. 'nle thymine present in Fraction VII, the "heat peak," was found to have the -

PAGE 70

TABLE XI Five m.inute in vivo incorporation of cl4_formate into the thymine of fractions of deoxyribonucleate from Ehrlich ascites tumor cells (KA #7) The fractions were derived from the experiment illustrated in Fig. 11. The percentage in parentheses which follows each fraction number represents the portion of the deoxyribonucleate of the sample recovered in the respective fraction. The results are expressed as counts per minute per mole of thymine. Method of isolation Fraction number Unfractionated, of thymine total from hydrolysate 1(9%) II(l2%) III( 11%) IV(l7%) V(l9%) VI(l0%) VII( 3%) deoxyribonucleate Column chromatography 697 773 651 488 332 639 2870 857 Rechromatography on pa-per following column 741 16a 938 9a chromatography 844 818 516 388 578 2960 Direct chromatography 830 108 434 la 926 oa on paper a The variation between dupl.icate determinations. ac...,

PAGE 71

64 highest specific activity. While the fractions shown account for 88% of the absorbancy applied to the column, they account only for 60% of the radioactivity. The discrepancy would be accounted for if material having a specific activity of about 3000 CPM/mole (a value similar to that of Fraction VII) remained on the column. The observed differences in specific activity could result from contamination of the thymine by constituents arising from the destruc tion of deoxyribose by formic acid, or from hydrolysis to amino acids of variable amounts of protein or other contaminants in the successive fractions of deoxyribonucleate. Acidic decomposition products would probably emerge near thymine from columns of Dowex 50. 15 In order to explore this possibility, the thymine residues were eluted from the planchets with O.l N HCl and rechromatographed on filter paper by use of isopropanolHCl. The specific activities of the thymine after rechromatography on paper, also presented in Table XI, were quite simi lar to those determined directly on the Dowex 50 effluent. The slight increase observed after rechromatography may have resulted from the elimination of materials which contributed ultraviolet absorption.5 In order to determine whether reproducible values for specific activities could be obtained by use of paper chromatography alone, independent hydrolysates of aliquots of Fractions II and V, and of the total deoxy ribonucleate were chromatographed directly, and analyzed as before. As shown in Table XI, the duplicate determinations agreed within less than 15 The sulfuric acid in which the hydrolysate of the deoxyribo nucleate was dissolved prior to chromatography would presumably be excluded from the polysulfonate resin; in fact, it was found to be eluted in about 24 ml, wile thymine emerged after about 60 ml.

PAGE 72

65 2.5%, and were similar to the values originally found. Appropriate analyses of other regions of such paper chromatograms provided values for the specific activities of adenine and guanine of some of the fractions. As shown in Table XII, the specific activity of guanine was found to vary among the fractions in the same way as that of thymine. The evidence strongly suggests that there are present in the specimen of deoxyribonucleate described above, molecules which vary not only in base composition but also in the extent to which they incorpo-14 rated radioactivity from C -formate during the five minute in vivo incubation period. That such variations in the extent of labeling can be obtained reproducibly was confirmed by isolating deoxyribonucleate from tumor cells removed from each of five of a second group of mice 14 (EA 19) five minutes after the injection of 50 curies of C -formate. A solution containing 12.6 mg of the specimen was added to an analytical column of Mg IRC-50 (28 x 0,9 cm). Under these overload conditions, 45% of the sample was eluted from the resin by 0.05 M Mg(OAc)2 As shown in Fig. 13, stepwise changes of the eluent provided 47% of the sample at 0.4 M Mg(OAc)2 and 2% at 1 M Mg(OAc)2 when the temperature was raised to 50 C. A small amount of additional material was eluted finally by 0.5 M NaOH at room temperature. The specific activities of the thymine of the four fractions were determined after paper chroma tography and are shown in Table XIII. The pattern of differential labeling is very similar to that found in the previous experiment, despite the examination of fewer, larger fractions in the present case.

PAGE 73

66 TABLE XII The specific activity of bases in fractions of deoxyribonucleate from Ehrlich ascites tumor cells (IA #7) The fractions were derived from the chromatogram illustrated in Fig. 11. To obtain the results shown here, the ba1e1 were isolated from hydrolysates by chromatography on paper, either directly, or after initial purification by use of columns of I>owex 50. II V VII Fraction Unfractionated, total deoxyribonucleate Specific activity (CPM/ mole) Adenine 18 28 Guanine 46 12 202 63 Thymine 818 388 2960 938

PAGE 74

0 O'l (\j u(/) Z:,
PAGE 75

I II III IV TABLE XIII Five minute in vivo incorporation of c14-formate into the thymine of fractions of deoxyribonucleate from Ehrlich ascites tumor cells (EA #9) The fractions were derived from the chromatogram illustrated in Fig. 13. Fraction Conditions of elution from Mg IRC-50, 2 8 x 0.9 cm 0.05 M Mg(OAc)2 30 0 0.4 M Mg(OAc)2 30 0 l M Mg(OAc)2 50 0 0 0.5 M NaOH, 24 Per cent of added sample (13 mg) recovered in fraction 45 47 2 3 Specific activity of thymine, CPM/mole 429 220 1,360 1,600 Unfractionated, total 396 7a deoxyribonucleate (100) 8Tbe variation bet-ween duplicate determinations. (1\ 00

PAGE 76

69 3 Incorporation of R -thymidine into fractions of deoxyribonucleate In vivo study of brief duration -The possibility that the differential labeling of the thymine discussed above results from a methyl group exchange involving intact deoxyribonucleate molecules was 3 excluded by an experiment in which the incorporation of H thymidine was studied. Cells were removed from each of seven mice five minutes 3 after the injection of 10 curies of H -thymidine having a specific activity of 360 curies per mole. A 13.l mg sample of the deoxyribonucleate isolated from the cells (EA #16) was chromatographed on a 28 x 0.9 cm column of Mg IRC-50 as in the previous experiment, except 0 that 1 M NaCl at 50, rather than 1 M Mg(OAc)2 was used to elute Frac-tion III. This modification minimized the precipitation of Mg(OH)2 when 0.5 M NaOH was passed through the column. Because of its asymmetry, Fraction IV was recovered in two portions, IVa and !Vb. Thymine was isolated from hydrolysates of the five fractions by paper chromatography and found to have the specific activities given in Table XIV. Although an entirely different precursor was used, the extent of incorporation into chromatographic fractions varied in the same manner as 14 was found originally with C -formate. That is, the specific activity of thymine in the fraction eluted by 0.4 M Mg(OAc)2 was leas than that found in the fractions eluted by 0.05 M Mg(OAc)2 or by the more vigorous conditions. These data suggest that the observed heterogeneity of incorporation occurs at points which lie beyond the completion of the nitroge nous base in the sequence of reactions leading to the synthesis of deoxyribonucleate. The evidence does not exclude the possibility of an

PAGE 77

TABLE XIV Five minute in vivo incorporation of tritiated thymidine into fractions of deoxyribonucleate from lhrlich ascites tumor cells (IA #16) I II III IVa IVb Fraction Unfractionated, total deoxyribonucleate Conditions of elution from Mg IRC-50, 28 x 0.9 cm 0 0.05 M Mg(OAc)2 30 0 0.4 M Mg(OAc)2 30 1 M NaCl, 50 0 0 0.5 M NaOH, 24 0 0.5 M NaOH, 24 Per cent of added sample (13 mg) recovered in fraction 50 43 3 <3 <3 (100) 8nie variation between duplicate determinations. Specific activity of thymine, CPM-'ximole 252 3a 132 14 1,040 623 299 266 98 ""-1 0

PAGE 78

71 exchange of thymine residues with otherwise intact molecules. However, 14 as was shown in Table XII, the incorporation of C -formate into adenine and guanine paralleled the incorporation into thymine. This suggests that if exchange occurs, it would involve all bases, and would therefore be tantamount to overall synthesis. It is likely, therefore, that the observed heterogeneity of incorporation of radioactivity is a property associated with the bioaynthesis of whole molecules of deoxyribonucleate. In vivo study of long duration --Aa discussed above, differential labeling among chromatographically purified fractions of deoxyribonucleate was reproducibly observed when tumor cells were harvested 14 3 five to seven minutes after the injection of C -formate or H -thymidine, The specific activities of thymine in fractions of deoxyribonucleate from tumor cells harvested from each of five mice (EA 117) 3 twenty-three hours after injection with 10 curies of the H -thymidine are given in Table XV. The specific activity of the total deoxyribonucleate was only about 2.7 times that found five minutes after the 3 injection of H -thymidine (EA 116, Table XIV). Since the generation time of the tumor cells is eighteen hours, the deoxyribonucleate isolated twenty-three hours after injection would be expected to have a specific activity about one-half as great as that isolated five hours after injection. If it is assumed that the specific activity was close to maximal five hours after injection, then the amount of radioactivity that is incorporated into deoxyribonucleate during the first five minutes after injection is about one-fifth of the maximum amount that can be incorporated under these experimental conditions. Evidently,

PAGE 79

I II III IV TABLE XV Twenty-three hour in vivo incorporation of tritiated thymidine into fractions of deoxyribonucleate from Ehrlich ascites tumor cells (EA #17) Fraction Conditions of elution from Mg IRC-50, 30 x 0.9 cm 0.05 M Mg(OAc)2 30 0 0.4 M Mg(OAc)2 30 0 0 1 M NaCl, 50 0 O. 5 M NaOH, 24 Per cent of added sample (11 mg) recovered in fraction 22 50 3 < 3 Specific activity of thymine, CPM/mmole 806 40 768 598 837 788 Unfractionated, total 705 31 deoxyribonucleate (100) 8Toe variation between duplicate determinations. ""' N

PAGE 80

73 any thymidine which is not incorporated in deoxyribonucleate shortly after injection rapidly becomes unavailable for incorporation. After H 3-thymidine is injected, it is possible that the apecific activities of pools of precursors at and above the level of thymidine fall almost as abruptly as they rise after the injection of 14 C -formate (see Fig. 12). Nevertheless, the chromatographic fractions provided by deoxyribonucleate isolated twenty-three hours after injection (Table XV) did not show differences in specific activities, in contrast to the fractions provided by specimens which were isolated five minutes after injection (Table XIV). As studied under these conditions, therefore, the differential labeling of chromatographic fractions is a transient phenomenon. Thus, there is no evidence that molecules in fractions having a high initial specific activity gradually undergo discrete changes in chromatographic behavior. Instead, the specific activity of all fractions approaches that of the total deoxyribonucleate. That is, the specific activities of the fractions initially less radioactive increase at the expen.se of the "heat peak'' fraction, the specific activity of which, however, does not fall below that of the other fractions. The comparatively low specific activity given in Table XV for the thymine of EA 117 total deoxyribonucleate would indicate that the average specific activity of thymine in the material that remained on the column was about 500 CPM/ri\lmole. In vitro study of brief duration -As discussed in the preceding section, about one-fifth of the maximum possible incorporation of thymidine occurs during the first five minutes after injection. In an effort to reduce the extent of incorporation still further, an

PAGE 81

74 in vitro experiment was performed. Cells were removed from the peritoneal cavities of six mice with the aid of about 2 ml of saline per d f d 1 vi l 35-37 C whi h mouse, an trans erre to a g ass a at a proceas c required about sixty minutes. After adding 8 curies of the tritiated thymidine, the cell suspension was incubated for two minutes with occasional stirring. The deoxyribonucleate (EA #15) was isolated and chromatographed to provide the data which are given in Table XVI. Although the total amount of precursor incorporated was small, the distribution of specific activities is similar to that seen previously in the five minute in vivo experiments. It is difficult to compare the extent of incorporation with previoua experiments since an equivalent of fewer curies per mouse was provided in this in vitro experiment. In view of the loss of the capacity for synthesis of deoxyribonucleate by cells maintained under in vitro conditions, it is possible that incorporation occurred only in the last cells transferred to the incubation vessel. Nevertheless, preferential incorporation of thymidine into "heat peak" material was even more pronounced in thia two minute in vitro experiment than in the five minute in~ experiments described above. It will be of interest to study even shorter intervals of incorporation.

PAGE 82

I II III IV TABLE XVI Two minute in vitro incorporation of tritiated thymidine into fractions of deoxyribonucleate from Ehrlich ascitea tumor cells (EA #15) Fraction Conditions of elution from Mg IRC-50, 30 x 0.9 cm 0.05 M Mg(OAc)2 30 0 0.4 M Mg(OAc)2 30 0 l M NaCl, 50 0 0 O. 5 M NaOH, 24 Per cent of added sample (11 mg) recovered in fraction 33 44 2 <2 Specific activity of thymine, CPM/mmole 2.75 0.25 a 1.75 0.058 16.00 12.00 Unfractionated, total deoxyribonucleate (100) 2.95 o.osa 4The variation between duplicate determinations. ..... V1

PAGE 83

CRITICISM OF THE CHROMATOGRAPHIC STUDIES The results of the chemical and biological studies presented in the preceding sections emphasize the need for further information regarding the molecular parameters which govern the specific interaction of deoxyribonucleate with Mg IRC-50. The chemical studies indicate that under certain conditions Mg IRC-50 has an affinity for all molecules of deoxyribonucleate, but that in general the affinity of the resin for molecules rich in adenine and thymine is stronger than for molecules rich in guanine and cytosine. Accordingly, the rank in the hierarchy of affinity of molecules of deoxyribonucleate for Mg IRC-50 appears to be governed to some extent by total base compo sition. The possibility that relatively minor chemical factor, in addition to total base composition, may modify this affinity is illustrated by a consideration of the distribution of 5-methylcytosine among the successive chromatographic fractions. While the mole per cent of cytosine of extreme fractions changes by factors of 1.3 and 1.2 in the specimens from calf thymus and Ehrlich ascites tumor cells, respectively, the mole per cent of 5-metbylcytosine changes by factors of 5 2 and 2.0. In time, detailed structural studies may reveal whether the affinity of deoxyribonucleate for Mg IRC-50 is lessened by the relative preponderance of the satellite base per se, or whether this affinity depends on exaggerated base sequences which are correlated with enrichment by 5-methylcytosine. During the course of this work, 76

PAGE 84

77 improved chromatographic procedures were developed for the analysis of I I pyrimidine nucleoside 3 ,5 -diphosphates, dinucleoside triphosphatea, and higher homologue& present in acid (39) and diphenylamine (40) hydrolysates of deoxyribonucleate. By providing evidence with regard to the relative clustering of purines and pyrimidines, these technique may lead to a clearer definition of distinctive features of the structure of deoxyribonucleate to which the interaction with Mg IRC-50 is sensitive. When more discriminating criteria for the characterization of deoxyribonucleates are applied, the differences between the chromatographic fractions may prove to be more profound than presently suspected. That is, "the differentiation between nucleic acids on the basis of difference, in the contents of their several nitrogenous constituents [has beeaja comparatively crude expedient, since it does not permit a distinction between sequence variations that are not accompanied by a change in total composition" (5). The extent to which particle weight influences the chromatographic behavior of molecules of deoxyribonucleate on columns of Mg IRC-50 has not as yet been determined. It would be of interest to apply techniques such aa have been used by Schumaker and Schachman (11) in order to determine the distribution of sedimentation coefficients within the chromatographic fractions. As was noted in the Introduction, the chromatographic procedures of Bendich and co-workers provide molecules having progressively increasing sedimentation coefficients but essentially unchanging base composition (9). The last fractions eluted from columns of ECTBOLA-cellulose have an average sedimentation coefficient higher than that of the total, unfractionated material. However, the

PAGE 85

78 successive fractions also exhibit a progressively greater heterogeneity with respect to distribution of sedimentation coefficients. For this reason it is difficult to compare fractions obtained by use of Mg IRC-50 with fractions obtained using columns of ECTBOLA-cellulose. Different principles appear to underlie the two fractionation methods. In contrast to the results obtained by Bendich and co-workers, the procedures of Chargaff (4), of Butler (6), and of Brown (8) and their co-worker provide fractions of deoxyribonucleate which exhibit the same trends in base compoaition aa the fractions eluted from columns of Mg IRC-50. The use of Mg IRC-50 for the isolation of such fraction of deoxyribonucleate presents a number of advantage not offered by the previous procedures. There is little doubt about the reproducibility of the Mg IRC-50 chromatograms. The resin is readily available cotllllercially and several lots have provided essentially identical results. The earlier procedures are difficult to standardize and to reproduce because they depend on the use of histones of uncertain homogeneity. Moreover, "denatured" histone and undenatured hiltone yield considerably different chromatographic results (8). Although the histone-chloroform 16 gel procedure causes serious losses of transforming activity, passage through columns of Mg IRC-50 or through certain histone-cellulose columns (8) has been found to be innocuous in this respect. The chromatographic studies have not as yet been concerned directly with the problem of whether each specimen of deoxyribonucleate consists of a spectrum of differently constituted molecules or rather, 16c. F. Crampton, E. Chargaff, and R. D. Hotchkiss, unpublished observations (1953).

PAGE 86

79 consists of a mixture of a relatively few different molecules. The use of Mg IRC-50 by itself or in conjunction with alternative fractionatioo methods may eventually provide information pertinent to this problem. In this regard, determination of the distribution of sedimentation coefficients (10, 11) as well as density gradient analysis in CsCl solutions (12) and studies of the temperature dependence of ultraviolet absorbancy (13) may prove to be convenient techniques for evaluating the extent of heterogeneity in fractions obtained from columns of Mg IRC-50. The interpretation of the chromatographic studie1 would also be aided by definitive knowledge of the chromatographic behavior of specimens of deoxyribonucleate which are, in effect, initially more homogeneous than those examined thus far. Such specimens could be obtained by proper choice of biological origin, by the preliminary use of alternative fractionation techniques, or by the application of conditions, such as controlled heating, which preferentially de1troy the native structure of certain molecules without altering others. 'nte evaluation of molecular parameters which underlie the chromatographic fractionations in the biological studies is even more complicated. Unlike the "crude expedient" of measurements of total base composition, measurements of transforming activity need not be the result of a positive contribution by all of the molecules present in a given fraction. It will be recalled that the tran1forming activity of specimens of deoxyribonucleate from Pneumococcus was distributed throughout most of the chromatographic fractions. This could result either from the existence of the same biological activity in molecules of different composition or from the association of all of the

PAGE 87

80 biological activity with molecules of discrete composition which, for some reason, interacted with Mg IRC-50 to an equivalent degree over a diffuse range of concentrations of Mg(OAc)2 Similar uncertainties pertain to the results of the incorporation experiments in which the specific activities of thymine changed considerably throughout the suc cessive fractions. An incorporation period of five minutes corresponds to 1% of the average time of 8,4 hours which each cell devotes to the synthesis of deoxyribonucleate during the growth cycle (17), Since the generation time of the cells is about eighteen hours (17), roughly half of all the cells present in an exponentially growing culture are engaged in synthesizing deoxyribonucleate. If utilization of the precursors began inmediately after injection in all cells which were synthesizing deoxyribonucleate, the isolated specimens would be about "0.5% labeled," However, it is not known whether many molecules are slightly labeled, or a small number of molecules are intensively labeled. Moreover, the occurrence of radioactivity in all chromatographic fraction poses the same problem noted above with respect to the ubiquity of transforming activity among the fractions from pneumococcal specimens. The characterization of fractions by measurements of activities is hazardous, since the distribution of activity may be governed by minor structural peculiarities not shared by all of the molecules present in the fractions. That is, incipiently labeled deoxyribonucleate might be expected to behave abnormally upon chromatography because a double stranded structure had not been completed when the cells were chilled prior to isolation of the specimen. Moreover, the chromatographic behavior of the small absolute quantities of labeled deoxyribonucleate

PAGE 88

81 could also be dictated chiefly by associated contaminants, such as protein, polysaccharide, or lipid, which had been attached during the process of replication. In this connection, it will be recalled that the specimen from rat liver contained a large amount of protein but gave chromatographic patterns similar to those provided by calf thymus. On the other hand, the preparations from Pneumococcus, which also contained large amounts of protein (and ribonucleate), gave quite different patterns. A systematic study of the protein content of chromatographic fractions was not attempted routinely because of the presence of only trace amounts of protein in most of the total specimens. The fact that columns of Mg IRC-50 are able to separate different molecules of deoxyribonucleate rests not upon the biological studies but rather upon the chemical studies of composition and upon the results of the rechromatography experiments. Nevertheless, the biological studies have shown that, under certain experimental conditions, affinity for Mg IRC-50 is in some way related to the metabolic origin of molecules of deoxyribonucleate.

PAGE 89

POSSIBLE SIGNIFICANCE OF 'niE BIOLOGICAL STUDIES The results of the incorporation experiments are summarized in Table XVII, which lists the relative specific activities of thymine in the two chromatographic fractions between which the greateat differ ences were usually obtained. As can be seen, preferential labeling of chromatographically different fractions of deoxyribonucleate from Ehrlich ascites tumor cells was observed following brief incorporation periods. The data again direct attention to the "heat peak" fraction, the unusual properties of which were discussed in earlier sections. It will be recalled that material in this position may be partly artifact and partly an authentic component present in the deoxyribonucleate submitted to fractionation. The base composition of the material resembles that of total deoxyribonucleate and therefore is not in line with the progressive increase in the ratio, (adenine+ thymine)/(guanine + cytosine), otherwise noted throughout the chromatogram. Of particular significance is the fact that the specimens of deoxyribonucleate from PneWDOcoccus provided relatively large amounts of "heat peak" material in which specific transforming activity was comparable to that of the total specimen. Had the incorporation experiments compared in Table XVII dealt with synchronously growing cells, the preferential labeling could be explained on the premise that chromatographically different molecules are synthesized sequentially. Radioautographic studies following 82

PAGE 90

TABLE XVII Relative specific activity of thymine in chromatographic fractions of deoxyribonucleate from Ehrlich ascites tumor cells The specific activities are relative to that of thymine in the respective unfractionated specimens of total deoxyribonucleate. The figures are based on results presented in greater detail in the tables indicated. Table XI XIII XIV xv XVI Specimen EA 17 EA 19 EA #16 EA #17 EA #15 Mode of labeling Precursor Conditions of incubation 14 C -fonnate 5 min, in vivo 14 C -fonnate 5 min, in vivo 3 H -thymidine 5 min, in vivo 3 H -thymidine 23 hr, in vivo 3 H -thymidine 2 min, in vitro Relative specific activity of thymine in chromatographic fractions Fraction eluted by 0.4 M Mg(OAc)2 (0.51) 0 56 0.50 1.1 0.59 "Heat peak" 3.2 3 4 3.9 1.2 5.4 CX> ""'

PAGE 91

84 3 administration of H -thymidine suggest that sequential synthesis of de oxyribonucleate actually occurs in chromosomes of grasshopper testes (41) as well as in plant tissues (42). However, in the present studies, ascites tumor cells at all stages of replication were examined, thereby excluding specific temporal sequence as a possible explanation of pref erential labeling. The synthesis of deoxyribonucleate by Ehrlich a&cites tumor cells occurs during 8.4 hours of the interphase period of the life cycle of these cells (17). It is possible that the biosynthesis of individual molecules of deoxyribonucleate requires a large portion of this time. If so, radioactivity would be confined to limited regions of large numbers of incomplete molecules after brief incorporation periods. Accordingly, the observed chromatographic distribution of radioactivity could depend largely upon heterogeneity with respect to degree of completion. On the other hand, the complete synthesis of molecules of 17 deoxyribonucleate may require only a few seconds. That is, the 8.4 hour period of synthesis might encompass the successive syntheses of many different molecules which are started and finished at different times in individual cells. In this case different molecules could attain different specific activities if there were a more rapid 17rt is also possible that some molecules of deoxyribonucleate are synthesized more rapidly than others because of restrictions imposed by the statistical ease with which available precursors can be aligned and polymerized in order to complete characteristic sequences of certain of the molecules. Moreover, molecules containing large amounts of 5methylcytosine might be assembled less rapidly than molecules which cb not contain this satellite base.

PAGE 92

85 incorporation of the labeled precursor into one group of molecules than into another. Such preferential incorporation could occur in at least three ways. !) The specimens of deoxyribonucleate may contain a small group of macromolecular precursors from which all molecules of deoxyribonu cleate are derived. Such precuraors would exhibit a more rapid increase in specific activity than would the product molecules after the injection of isotopic compounds. Moreover, such precursor molecules would subsequently lose their activity more rapidly than would the product molecules. It is possible that deoxyribonucleate in the "heat peak" corresponds to such precursor material. The "heat peak" fraction differs from the bulk of the nucleate not only by its very high rate of incorporation of radioactive precursors, but also by its abnormal affinity for Mg IRC-50 Moreover, the composition of "heat peak" deoxyribonucleate resembles that of the whole specimen, a property that would be expected for a fraction fran which all the molecules of a preparation arise. Precursor, or incomplete deoxyribonucleate, might well be expected to differ structurally from completed nucleate in a manner which could cause a greater affinity for the resin. Precur1or deoxyribonucleate could be single-stranded (43), multi-stranded (44,45), or associated with materials which are not attached to completed molecules. On the other hand, the specific activity of precursor deoxyribonucleate should approach zero a long time after administration of isotope, a requirement which is not met by the "heat peak" material. The specific activity of all fractions, including "heat peak," approached a similar value twenty-three hours after injection. It is

PAGE 93

86 quite possible, however, that precursor deoxyribonucleate in "heat peak" is contaminated by a large amount of ordinary nucleate. In such an event, the large amount of radioactive product deoxyribonucleate which is eluted non-specifically in "heat peak" would obscure the low specific activity of the precursor material. The differences in the specific activities of the chromatographic fractions other than "heat peak" could be attributable to a form of precursor deoxyribonucleate which emerges together with the early chromatographic fractions from Mg IRC-50. 1) If the different molecules of deoxyribonucleate in a single cell are synthesized from independent pools of low molecular weight precursors they would become labeled at unequal rates if the pools differed in size. Those nx>lecules formed from the smallest pools of precursors would incorporate the added label most rapidly. 1) If molecules of deoxyribonucleate differed with respect to the rate at which they were synthesized from a single pool of precursors, differences in specific activity could result if there were rapid turnover in a portion of the molecules, or possibly, if different molecules were synthesized at specifically related times not related to their chromatographic behavior. The experimental observation of preferential labeling of different fractions of deoxyribonucleate is not without precedent in the literature. For example, Bendich, Russel, and Brown (46) found that 0 deoxyribonucleate extracted from rat tissues with 10 % NaCl at 85 C yielded fractions "1hich differed in their solubility in saline. The two fractions of deoxyribonucleate isolated from various organs several

PAGE 94

87 14 days after injection of C -formate contained different amounts of radioactivity. However, the specific activities of different bases from the fractions of a single organ did not parallel one another, nor did different organs consistently show a greater incorporation into one of the fractions than into the other. The authors conclude, "the results not only show that [deoxyribonucleataj of any single organ is not metabolically homogeneous, but that the individual bases of the nucleic acids of a given organ are renewed at dissimilar rates, and that the pattern of renewal varies from organ to organ." However, the differences reported for fractions from a single organ were smallest with respect to thymine, the base least likely to be contaminated by material derived from highly radioactive ribonucleate. The greatest difference in specific activity of thymine occurred in a specimen from small intestine, where specific activities of 614 CPM/ mole and 779 CPM/mole were reported for Fractions 1 and 2, respectively. In contrast to the experiments of Bendich al., the studies sunmarized in Table XVII have revealed large specific activity differences among fractions with respect to a base which is not present in ribonucleate from most sources. As has been noted previously (Table XII), parallel results were obtained with one series of fractions by specific activity measurements of guanine, which is present in both ribo-and deoxyribonucleates. In addition, deoxyribonucleate waa isolated by use of the urea medium which has been found to provide specimens free from appreciable amounts of uracil, which is a characteristic constituent of ribonucleate. Finally, extremely similar patterns of incorporation 3 14 were given both by H -thymidine and C -formate, precursors which are

PAGE 95

88 utilized for the synthesis of the two types of nucleate with considerably different efficiencies. A second report of metabolic heterogeneity of deoxyribonucleate is that of Friedkin and Wood (47), who incubated chicken bone 14 marrow cells,!!! vitro for two hours with C -thymidine. The incubation mixtures were then frozen, lyophilized, and extracted with organic solvents. After stirring in 0.9% NaCl, the material was treated with sodium dodecyl sulfate by the procedure of Kay!!_ al. (21). Three successive extractions provided three fractions of deoxyribonucleate. The specific activity of the thymidine isolated from enzymatic hydrolysates varied by as much as 3.3 from one fraction to another. The interesting results of Friedkin and Wood merit further study. They are, however, somewhat difficult to interpret since the fractions were obtained from unpurified deoxyribonucleate by ill-defined techniques not known to give fractions which differ in other respects. Moreover, the use of a system consisting of an initially heterogeneous population of cells, in which there is not known to be a net synthesis of deoxyribonucleate, invites uncertainties which were avoided as far as possible in the present work.

PAGE 96

SUMMARY Variables affecting the specific interaction between Mg IRC-50 and highly purified specimens of deoxyribonucleate were examined. Conditions optimal for the analysis of molecules of different chemical composition were developed and applied to specimens of deoxyribonucleate isolated from Ehrlich ascites tumor cells which had incorporated 14 3 C -formate or H -thymidine for very brief intervals of time. Large differences between the fractions with respect to the specific activity of their thymine residues were found, suggesting that the columns of Mg IRC-50 may be used to separate molecules of different biological origin as well as of different chemical composition. The basis of the interaction of Mg IRC-50 and deoxyribonucleate is discussed critically and explanations are offered for the preferential incorporation of precursors into chromatographically different fractions. 8 9

PAGE 97

\ REFERENCES 1. Miescher, F., Hoppe-Seyler's Med. chem. Unters., ~, 441 (1871); translated by Gabriel, M. L., in M. L. Gabriel and S. Fogel (Editors), Great experiments _!!l biology, Prentice-Hall, Inc., Englewood Cliffs, New Jersey, 1955, p. 233. 2. Levene, P.A., and Bass, L. w., Nucleic acids, The Chemical Catalog Company, New York, 1931. 3. Chargaff, E., :!,. Cellular Comp. Physiol., 38, 41 (1950). 4. Chargaff, E., Crampton, C. F., and Lipshitz, R., Nature, 172, 289 (1953). 5. Crampton, C. F., Lipshitz, R., and Chargaff, E., :!,. Biol. Chem., 211, 125 (1954). 6. Lucy, J. A., and Butler, J. A. v., Nature, 174, 32 (1954). 7. Brown, G. L., and Watson, M., Nature, 172, 339 (1953). 8. Brown, G. L., and Brown, A. V., Symposia Soc. Exptl. Biol., 12, 6 (1958). 9. Rosenkranz, H. S., and Bendich, A., :!,. ~Chem. ~-, fil, 902 (1959). 10. Butler, J. A. v., and Shooter, K. V., in W. D. McBlroy and B. Glass (Editors), The chemical basis of heredity, Johns Hopkins Press, Baltimore, 1957, p. 540. 11. Schumaker, V. N., and Schachman, H.K., Biochlm. !!_ Biophye. Acta, 23, 628 (1957). 12, Sueoka, N., Marmur, J., and Doty, P., Nature, 183, 1429 (1959). 13. Marmur, J., and Doty, P., Nature, 183, 1427 (1959). 14. Vendrely, R., in E. Chargaff and J. N. Davidson (Editors), 11le nucleic acids, Vol. II, Academic Press, Inc,, New York, 1955, p. 155 15. Swift, H., Physiol. Zool,, 23, 169 (1950). 16. Howard, A., and Pelc, S. R., Heredity,,, 261 (1953). 90

PAGE 98

91 17. Edwards, J. L., Koch, A. L., Youcis, P., Freese, H, L., Laite, M. B., and Donalson, J. T., J. Biophys. Biochem. Cytol., l, 273 (1960). 18. Swick, R. W., Koch, A. L., and Randa, D. T., Arch. Biochem. Biophys., 63, 226 (1956). 19. Fresco, J. R., and Bendich, A., J. Biol. Chem., 235, 1124 (1960). 20. Crampton, C. F., Benson, A. M., Rodeheaver, J. L., and Wade, A. B., Federation Proc., !l, 206 (1958). 21. Kay, E. R. M., Simmons, N. s., and Dounce, A. L., J. ~Chem. Soc., 74, 1724 (1952). 22. Crampton, C. F., Stein, W. H., and Moore, s., J. Biol. Chem., 225, 363 (1957). 23. Moore, s., and Stein, W H., J. Biol.~-, 211, 893 (1954). 24. Hirs, C.H. W., Moore, s., and Stein, W. H., J. Biol. Chem., 200, 492 (1953). 25. Crampton, C. F., and Petermann, M. L., IBiol. Chem., 234, 2642 (1959). 26. King, E. J., Biochem. J., 26, 292 (1932). 27. Wyatt, G. R., Biochem. l-, 48, 584 (1951). 28. Moore, s., and Stein, W. H., J. Biol. Che!,, 192, 663 (1951). 29, Moore, s., and Stein, W. H., Advances in Protein Chem., XI, 191 (1 95 6). 30. Lee, W. A., and Peacocke, A. R., J. Chem. Soc., 3361 (1951). 31. Gurd, F. R. N., and Wilcox, P. E., Advances!!! Protein Chem., XI, 312 (1956). 32. Pullman, B., and Pullman, A,, Biochim. ~t Biophys. ~, 36, 343 (1959). 33. Reiner, B., and Zamenhof, s., J. Biol. Chem., 228, 475 (1957). 34. Shack, J., and Bynum, B. s., Nature, 1 84, 635 (1959). 35. Wiberg, J. s., and Neuman, W. F., Arch. Biochem. Biophys., 72, 66 (1957).

PAGE 99

92 36. Edwards, J. L., Smith, s. w., Westmark, E. R., and Youcis, P. M., Federation Proc.,!!!, 475 (1959). 37. Schmidt, G., and l1lannhauaer, S. J., l Biol. Chem., 161, 83 (1945). 38. Frankel, F. R., Knapp, J., and Crampton, C. F., Federation Proc., 19, 306 (1960). 39. Shapiro, H. S., and Chargaff, E., Biochim. !!_ Biophys. ~, 39, 68 (1960). 40. Burton, K., and Petersen, G. B., Biocbem. l-, 75, 17 (1960). 41. Lima-de-Faria, A., l Biophys. Biochem. Cytol., E_, 457 (1959). 42. Taylor, J. H., Exptl. Cell Research, 15, 350 (1958). 43. Sinsheimer, R. L., l ~Biol.,!, 43 (1959). 44. Stent, G., Advances in Virus Research, 1, 138 (1958). 45. Rich, A., Nature, 181, 521 (1958). 46. Bendich, A., Russel, P. J., Jr., and Brown, G. B., l Biol. Chem., 203, 305 (1953). 47. Friedkin, M., and Wood, H., IV, l Biol.~-, 220, 639 (1956).

PAGE 100

VITA The candidate was born in Baltimore, Maryland on July 6, 1934. He received a degree of Bachelor of Science in June, 1955 from the College of Chemistry and Physic at the Pennsylvania State University, and in June, 1957, a degree of Master of Science from the College of Agricultural and Biological Chemistry at the same institution. His theais was titled, "Electrophoretic and Bactericidal Studies on the Sera of Normal and Scouring Calves." He began his studies for the degree of Doctor of Philosophy at the College of Medicine, University of Florida in September, 1957. He has held teaching assistantship and research fellowships at these two institutions.

PAGE 101

This dissertation was prepared under the direction of the chair man of the candidate's supervisory COUElittee and has been approved by all members of that coumittee. It was submitted to the Dean of the College of Medicine and to the Graduate Council, and was approved as partial fulfillment of the requirements for the degree of Doctor of Philosophy. August 13, 1960 Dean, Graduate School

PAGE 102

\ 0 UNIVERSITY OF FLORIDA lllllllllllllllllllllllllllll l l]IIIIIIIIIIII IIIIIII I I IIIIIIIIII 3 1262 08554 3212


xml version 1.0 encoding UTF-8
REPORT xmlns http:www.fcla.edudlsmddaitss xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.fcla.edudlsmddaitssdaitssReport.xsd
INGEST IEID EJCU1UQ5O_KQV53A INGEST_TIME 2012-02-29T18:45:20Z PACKAGE AA00009134_00001
AGREEMENT_INFO ACCOUNT UF PROJECT UFDC
FILES