DNA repair and cellular aging in mammals


Material Information

DNA repair and cellular aging in mammals
Physical Description:
vii, 93 leaves : ill. ; 29 cm.
Fort, Farrel Lee, 1947-
Publication Date:


Subjects / Keywords:
DNA Repair   ( mesh )
Cell Survival   ( mesh )
Biochemistry and Molecular Biology Thesis Ph.D   ( mesh )
Dissertations, Academic -- Biochemistry and Molecular Biology -- UF   ( mesh )
bibliography   ( marcgt )
non-fiction   ( marcgt )


Thesis (Ph.D.)--University of Florida, 1978.
Bibliography: leaves 78-92.
Statement of Responsibility:
by Farrel Lee Fort.
General Note:
General Note:

Record Information

Source Institution:
University of Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
aleph - 000896493
oclc - 25427077
notis - AEK5122
System ID:

Full Text








I would like to thank my chief advisor, Dr. Peter Cerutti,

for his tolerance and advice over the years and for allowing me to

do agiLng research in his lab.

My other advisors, Dr. Gary Stein, Dr. Tom O'Brien, Dr. Carlo

Moscovici, and Dr. Joyce Remsen have given me much needed moral

support, encouragement, and advice which I appreciate tremendously.

Most importantly, without the encouragement and moral support

from, and sacrifice by, my wife, Trish, I feel I surely could not

have endured the past 4 1/2 years.


ACKNOWLEDGMENTS ------------------------------------------

LIST OF FIGURES ------------------------------------------ iv

ABSTRACT ------------------------------------------------- vi

CHAPTER 1. INTROD1UC ION ---------------------------------- 1

CHAPTER 2. EXPERIMENTAL DESIGN --------------------------- 14

CHAPTER 3. ADl.o!:)S AND MA\TERIALS -- ------------- 21

Animals ----- ---- -------- ----------------- 21

General Tissue Culture Methods --------------- 22

Isolation Of Tissue Samples ---------------------- 22

luitiation Of Fibroblast Cultures ------ --------- 23

Growth, Labeling, And Treatment Of Cells ------------- 2

Cell Lysis --------------------------------------------- 27

Sample Collection And Counting --------- --- 28

A.lkali-Labile Site Experiments ----------------------- 29

Data Reduction ------------- -------------------------- 29

Chemicals ---------------------------------------------- 32

CHAPTER 4. RESULTS .AD INTERPRETATION ------------------- 33

CIA'7.R 5. DIS'X'.------------------------------------- 4

I,IT:'TL; CALT --.---------------------------------------- 78

BH IO; RA\P.iZ CAL SF;:. ti ------------------------ 93


Figure 1. ENU; structure and DNA adducts formed ------------------- 16

Figure 2. N-AAAF; structure and DNA adducts formed ---------------- 17

Figure 3. BPADE; structure and DNA adducts formed ------------- 18

Figure 4. Protocol for labeling and growth of cell cultures ------- 25

Figure 5. Sample elution curves ----------------------------------- 31

Figure 6. Fragmentation of parental DNA after ENU treatment ------- 34

Figure 7. Alkali-labile sites produced by ENU --------------------- 37

Figure 8. Fragmentation of parental DNA after pH 6 buffer treatment 39

Figure 9. Elongation of daughter DNA after ENU treatment ---------- 40

Figure 10. Elongation of daughter DNA after pH 6 buffer treatment -- 43

Figure 11. Total DNA synthesis after ENU treatment ----------------- 44

Figure 12. Total DNA synthesis after pH 6 buffer treatment --------- 46

Figure 13. Fragmentation of parental DNA after N-AAAF treatment ---- 47

Figure 14. Alkali-labile sites produced by N-AAAF ------------------ 50

Figure 15. Fragmentation of parental DNA after DMSO treatment ------ 51

Figure 16. Elongation of daughter DNA after N-AAAF treatment ------- 52

Figure 17. Elongation of daughter DNA after DMSO treatment --------- 53

Figure 18. Total DNA synthesis after N-AAAF treatment -------------- 54

Figure 19. Total DNA synthesis after DMSO treatment ---------- 56

Figure 20. Fragmentation of parental DNA after BPADE treatment --- 57


Figure 21. Alkali-labile sites produced by BPADE -------------- 58

Figure 22. Elongation of daughter DNA after BPADE treatment ------ 60

Figure 23. Total DNA synthesis after Li'ADE treatment -------------- 61

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



Farrel Lee Fort

December 1978

Chairman: Peter A. Cerutti
Cochairman: Gary Stein
Major Department: Biochemistry and Molecular Biology

Repair of DNA damage is a process that is expected to be crucial

in setting the level of the error content of cellular components.

Errors in cellular components are expected to be deleterious to

normal cellular functions. Since maintenance of normal function

is expected to be prerequisite for longevity, DNA repair is likely

to be a determinant of the life span of an organism.

If the accumulation of errors in cellular macrormolecules plays

a role in mammalian senescence, this could also affect those compon-

ents involved in DNA repair. An impairment of DNA repair systems

with age would imply ani increasing incidence of unrepaired or mis-

repaired damage to the genome ,_ith age. The purpose of this work

was to determine, therefore, if DNA repair is altered with age in


Skin fibroblasts derived from young (3 weeks old) and aged

(2 years old) Fischcr 344 rats were treated with either 1.70 mM


ethylnitrosourea, 4 micromolar N-acetuxy-2-iacetyl-aminofluorene,

or 1.35 micromolar benzo(a)pyrene-anti-diol-epoxide. Following

treatment, the kinetics of DNA excision repair and the elongation

of newly synthesized, undamaged daughter DNA strands were followed

using the alkaline elution technique for determination of single-

strand DNA size, and the total, overall rate of DNA synthesis was


Significant differences were found between cells of young and

old rats in the formation and resealing of single-strand breaks

in parental DNA and the elongation of undamaged, daughter strand

DNA after both alkylation and arylalkylation damage. These results

are consistent with decreased efficiency of DNA repair in cells

derived from aged rats.

The relation of experimentally induced DNA damage to that

expected to occur by natural mechanisms, in vivo, is discussed as

are those aspects of DNA repair and the possible modulation of DNA

repair expected to be important for future aging research.


It is apparent that the genetic makeup of an organism is

a major determinant of the life span of that individual organ-

ism. That this is so is evidenced by the fact that the variation

among the life spans of individuals of the same species is less

than the variation among different species. It follows that in

order to determine why some species live longer than others, we

must know something about the differences in genetic make-up

between species of differing longevity. Obviously we cannot ex-

amine every gene product of even one species. The problem, then,

is to decide which functions are most likely to be critical de-

terminants of longevity.

It seems reasonable to assume.that specific functions are

responsible for maintenance of viability in all the cells of an

organism. Based on the rate of evolution of the increase in

life span for the human ancestral line, these functions may be

few in number.I With age these functions might decline in effi-

ciency, and this in turn would impair the proper functioning of

the cells. If this were true, we might expect that,.eventually,

the failure of some critical organ system would lead to the demise

of the organism. This, of course, is consistent with what we

know of death in mammals. With this in mind we can see that the

key to longevity might be processes necessary for the maintenance

of proper function in all cells.


Two questions can be asked. rirst, what are these processes,

and, second, do they deteriorate with age?

Ideas about the events leading to the dysfunction and ultimate

death of cells, and therefore of organisms, can be categorized as

either error theories or programmed death theories.

Probably the best known proponent of the error accumulation

hypothesis is Orgel.2 He originally suggested that errors pro-

duced in components of the protein synthesizing system would cause

that system to become more error-prone. Production of defective

proteins such as RNA polymerases would increase the error content

in newly made proteins. The positive feedback for error-prone-

ness would cause an irreversible build-up of mistakes until the

cell could no longer survive. In dividing cells this would also

apply to the DNA synthesizing system. Error containing DNA poly-

merases would cause the synthesis of polymerases containing more

errors until finally, the newly synthesized DNA would no longer

be functional.

This could also be applied to mitochondria since they are

known to synthesize some of their own components. However, a

large number of errors in mitochondria would be contributed by

the nuclear-cytoplasmic system. Even errors in those components

coded for by the mitochondrial genome might not be independent

of cytoplasmic influence since some of the components necessary

for mitochondrial transcription and translation are probably made

in the cytoplasm.3

Also, we must not underestimate the effect of errors in
other cellular components. If the proper functioning of the


cell is dependent upon the integrity of all its components, then

errors in any of these components would result in decreased effi-

ciency of other components. For instance, defects in the cell

membrane could be imagined to alter the availability of amino

acids and, therefore, affect protein synthesis. This might re-

sult in the synthesis of incomplete, and therefore defective,

proteins or failure to replace proteins that have become faulty.

This same idea could be extended to extracellular components.

The cells of the body are responsible for constructing and main-

taining their own environment. If this environment becomes de-

fective, then it would be expected to have an adverse effect on

the cells.

Probably the major stimulus for proposing some sort of pro-

gram or counting mechanism in normal cells that accounts for their

finite life span has been the need to account for the fact that

transformed cells in culture are capable of apparently infinite

growth while normal, diploid cells -are capable of only a limited
number of population doublings. If normal cells have a mechanism

to count off the number of cell divisions, this could be inter-

rupted or destroyed in transformed cells.

Holliday6 has pointed out, however, that there are several

lines of evidence which argue against this proposal. If untrans-

formed cells had some kind of counting mechanism that determined

the number of cell divisions before death, one might not expect

environmental influences to alter their division potential.

Nevertheless, exposure of fibroblasts to ionising radiation,

elevated temperatures or 5--fluorouracil decreases their division


potential. Also, one might expect that at any given point in the

life span of a culture of fibroblasts, all cells would be capable

of the same number of mitoses. However, cells taken from normal

cultures show heterogeneity in growth potential.
Orgel2 actually hinted at a mechanism involving the accumu-

lation of errors and cellular selection by division that would

account for the observed differences in growth potential of normal,

diploid fibroblasts and transformed cells. This idea has been

elaborated further by Holliday. Consider that for any cell type

there is a probability, P, that the accumulation of errors in the

cellular components will reach a critical level such that the

development of a higher error level is inevitable. This build-

up of errors would be irreversible, and the cell would be committed

to a sequence of events leading to death. In slowly dividing

cells, the initial accumulation of the critical level of errors

would not lead immediately to cell death because undamaged com-

ponents would be present in sufficient quantities. At any given

P, as a population of cells of this type aged, more and more

cells would accumulate errors until, eventually, virtually all of

the cells would contain an error level equal to or greater than

the critical level. All cells in the population would then be

irreversibly committed to death. In rapidly dividing cells, the

initial build-up of errors to the critical level would be lethal

because the supply of undamaged components in these cells would

be less than that required for life. Those cells that reached

the critical error level would die leaving only the uncommitted

cells in the population. In a rapidly dividing population, there


would be a continual selection for relatively undamaged cells.

Growth of the population could thus continue Indefinitely.

/ Three things would be involved in determining mortality or

immortality for a population of cells. As P decreased, or as the

mitotic rate increased, or as the size of the population increased,

the growth potential of the population would increase. In the

latter case, the more cells there were in the slowly dividing

population, the longer it would take before all the cells became

committed to death.

Since the tissues of an organism are composed of various cell

populations, these considerations apply to cellular aging which

occurs in vivo. On the basis of these arguments, then, mammalian

senescence could be explained solely by the accumulation of errors

in cellular constituents.

Orgel later concluded, however, that the error content of

cellular components should eventually reach a plateau level rather
than increase infinitely. This is because an error containing

system should be able to synthesize a new system containing fewer
errors. For instance, suppose an error containing mRNA is being

translated by a faulty translation system. Since the translation

system is making mistakes, there is a finite probability that a

message error will be translated incorrectly to insert the correct

amino acid into the newly forming peptide. This same concept

could be applied to all of the statements made above about trans-

cription and other cellular constituents. Experimental evidence

for this hypothesis has been obtained with Escherichia coli.

When E. coli were treated with streptomycin, the error level in


flagellin proteins increased within a few generations to a stable,

plateau level while essentially all rho cells remained viable.9

However, these were growing cultures, and the observed, overall,

average, error level could well have been below the critical level

for these cells. As long as the plateau level of errors is below

the critical level for any given cell, that level of errors would

be consistent with the immediate, continued viability of the cell.

Orgel's revised hypothesis can thus be seen, by this mechanism,

to fit the scheme proposed by Holliday.

Considerable evidence has been accumulated indicating that

the level of error containing macromolecules does increase in

the cells of older organisms. 0-45 Although in no case is it

known whether the observed error level approaches the critical level

for any given cell type, the fact that increased error levels are

observed indicates that the error levels in the cells of older

individuals are closer to the critical levels than in the cells

of younger individuals. This is consistent with Holliday's pro-

posal which predicts that these cell populations will eventually


The observed correlation of decreased replicative life span
of fibroblast cultures with increased donor age is also consis-

tent with Holliday's proposal. These observations have been ex-

tended to include other parameters such as rate of fibroblast

migration, cell population replication rate, cell number at con-

fluency,46 and colony size distribution.47

Day has suggested that tissue culture cells might be in a

constant state of error prone (see Results and Discussion) DNA

repair as a result of overwhelmin--n stresses induced by "fluores-

cent light, nutrient imbalances, sublethal production of peroxides

or accumulation of waste products between feedings". We now

know, for instance, that common fluorescent light induces mutations49

and chromosome damage in tissue culture cells. I would suggest

further that altered parameters observed in cultures derived from

aged donors reflect decreased capacity of in vivo aged cells to

withstand a variety of stresses whether they be damage to DNA or

to other cellular components. This could be due to a higher error

level in aged donor cells resulting in decreased ability to mount

an effective response to damaging insults.

Key processes which maintain and protect normal functioning

of cells would be those that serve to keep the error level below

the critical level. Among those expected to accomplish this are

specific removal of damaged proteins, 51-53 scavenging of harmful

radical species,54 and repair of DNA damage.55

This study has focused on DNA repair since increased capacity

fur this function has clearly been shown to be correlated with

increased life span. Hart and Setlow56 found that the ability of

fibroblasts to carry out U.V.-induced unscheduled DNA synthesis

is correlated with the log of the life span for several mammalian

species. This was confirmed by Sacher and Hart57 using fibroblasts

derived from the rodent species Mus musculus and Peromyscus leucopus.

The latter lives 2.3 times longer than the former and has 2.3 times

the repair capacity as evidenced by U.V.-induced unscheduled DNA

synthesis. There was no difference, however, in the rate of re-

joining of X-ray-induced single-strand breaks in DNA of fibroblasts


from M. musculus and P. leucopis. It is tempting to speculate,

therefore, that longevity might be correlated with increased capa-

city for repair of DNA base damage as opposed to repair of single-

strand breaks.

It is of interest to know if DNA repair capacity decreases

with age. If so, this would be reflected in an increased error

level in the DNA with age. Decreased functional capacity for DNA

repair in the face of an increased error level in DNA might indi-

cate the development cf a positive feedback for increased error

production in this system analogous to that proposed by Orgel.

Because of the central importance of the genetic information con-

tained in DNA, accumulation of unrepaired or misrepaired DNA damage

would have a pervasive, adverse effect on the functioning of each

cell and, therefore, of the entire organism. For this reason,

then, I have studied repair of DNA base damage as a function of


Evidence for accumulation of DNA damage with age in mammals

includes increased levels of single-stranded regions27'28'35 in

DNA of mice, increased levels of DNA single-strand breaks in dog

brain2 and mouse liver,3 decreased double-strand molecular

weight of rat liver DNA, increased incidence of chromosome
31 40 41 42
abnormalities in rats, mice, dogs, guinea pigs, and
32 18,39 34
humans,32 38' app-rent loss of rRNA genes in brain DNA of dogs,

and accumulation of DNA-protein crosslinks with age.37,44 Curtis and

Miller42 showed that ths. incidence of spontaneous chromosome aberrations

not only increases with age but that the rate of increase is in-

versely proportional to maximum ife span when mice, guinea pigs,


and dugs are compared. This correlates well with the studies by

Hart and Setlow and Sacher and Hart mentioned above.

Certain facets of DNA repair are known not to change with age.
Wheeler and Lett found no change with age in the ability of in

vivo-irradiated canine cerebellar neurons to repair X-ray-induced

single-strand DNA breaks. This observation has been repeated in

rabbit retina in vivo and in primary fibroblasts derived from

rabbits of different ages.58 Ono and Okada59 have obtained similar

results in vivo with mouse cerebellum, spleen, and liver. No

changes were found in single-strand break rejoining ability in

spite of the increased levels of single-strand breaks found in

unirradiated dog cerebellum29 and mouse liver.59 Also, the aged

cells in vivo were apparently able to rejoin newly-formed, X-ray-

induced, single-strand breaks but not those that had already

accumulated with age. Accumulation of single-strand breaks with

age may represent accumulation of end groups rendering those breaks

inaccessible to the repair systems.
Yielding has suggested that accumulation of DNA damage with

age may result from inaccessibility of certain unused portions of

the genome to repair enzymes. In dividing cells all of the genome

would be used at one time or another since the entire genome must

be replicated for a cell to divide. In nondividing, terminally

differentiated celJs such as muscle cells and neurons only those

portions of the genome necessary for the function peculiar to that

cell would be used. To maintain viability for these cells only

these relatively few areas of the genome need be maintained error



Some evidence for prefere;,tial repair of different regions

of the gcnome comes from studies indicating higher levels of repair

replication in DNA that is more rapidly released by micrococcal

nuclease digestion. 61-62 These studies did not determine if a

higher level of damage was initially introduced into nuclease

accessible regions, thus accounting for the greater repair seen

there. This was ruled out, however, by Ramanathan et al. and

Wilkins and Hart. Ramanathan used labeled carcinogens and observed

preferential removal of this label from DNAase I accessible regions

after long incubation times in rat liver. Wilkins used a micrococcal

repair endonuclease extract to observe a greater percentage removal

of U.V.-induced endonuclease sites from chromatin not shielded by

protein than from those shielded by protein. Feldman et al.,6

however, found similar extents of removal of labeled benzo(a)pyrene

adducts from micrococcal nuclease accessible and inaccessible regions

of DNA in chromatin.

Further evidence for areas of "preferential repair comes from

cytological studies. Bc-linar et al.66 and Harris et al.67 both

found that the distribution of repair DNA synthesis after U.V. light

and a variety of chemical carcinogens differs from the distribution

of DNA in the nucleus. However, the two studies disagree on the

location of the greatest repair DNA synthesis; it being found at

the periphery of the nucleus in the former and in the central region

in the latter study. These differences may be due to the different

cell types used and/or to different methods used to determine DNA

distribution within the nucleus.


Yielding's scheme also predicts that as cells differentiate

to a nondividing state, their repair response to DNA damage should

decrease. This is substantiated by considerable experimental

evidence. Differentiation of myoblasts into mature myotubes is

associated with decreased repair response after treatment with

U.V. light,68 methylmethanesulfonate, or 4-nitroquinoline-l-

oxide (4-NQO).70 This is found in cultures of chick and rat

myoblasts and the L6 myoblast cell line, respectively. It is

interesting to note that rejoining of X-ray or 4-NQO-induced

single-strand breaks was not decreased upon fusion of L6 myoblasts

to form myotubes. On the other hand, the murine C-1300 neuro-

blastoma cell line does lose the capacity to rejoin X-ray-induced

single-strand breaks as well as the capacity to repair U.V.-

induced damage upon differentiation.72 Also, chick neural cells

lose, upon differentiation, the capacity to repair methylmethane-
sulfonate and N-acetoxy-2-acetylaninofluorene-induced damage.

These changes may simply reflect differences in repair capa-

city between dividing and nondividing cells. Lymphocytes stim-

ulated to divide, for instance, are known to have an increased

repair response to U.V. light and N-acetoxy-2-acetyl2amiofluorene-

induced75 damage over that found in unstiiulated cells.

The repair response in terminally differentiated cells is not

zero; just less than that in the undifferentiated, dividing stare.

It is tempting to speculate that differentiated, nondividing cells

still repair those regions of the genome necessary for proper

functioning but allow the rest to slip into disrepair. A process

such as this may be an additional factor in the accumulation of DNA


damage with age in r-ondividing celis. Additionally, it could pro-

vide a safety mechanism against cell division in tissues such as

mature muscle and nervous tissues where cell division is strictly

disallowed. Accumulation of nonrepairable breaks as observed in

the above studies29,59 might create a block to DNA replication

and therefore prevent the cell from dividing.

A more fruitful area of investigation may be changes in re-

pair of DNA base damage that occur with age. A start has been

made in this direction. Schneider and Kram have examined sister

chromatid exchange (SCE) formation after mitomycin C (MMC) damage

using human fibroblasts in culture76 and mouse and rat bone marrow

cells in vivo.77 After high doses of MMC, decreased levels of SCE's

were observed with increased donor age. Sister chromatid exchanges

are believed to be related in some way to DNA repair although the
exact nature of this involvement is not at all clear. Since MMC

is a bifunctional alkylating agent, these studies may indicate

decreased repair of DNA-DNA and/or DNA-RNA and/or DNA-protein
crosslinks. Lambert et al. have studied U.V.-induced DNA repair

synthesis in human peripheral lymphocytes and found a decrease with

donor age. This was not found, however, in a study of U.V.-induced
repair synthesis in human spermatocytes as a function of age.

Lehmann et al.81 have studied postreplication repair (see Resputs and

Discussion) of U.V.-induced damage in fibroblasts derived from aged
humans and found no alteration. Finally, Bochkov and Kuleshov and
Deknudt and Leonard have studied chromosome aberrations after treat-

ment of human lymphocytes with the alkylatir:g agents N,N',N"-tri-

ethylene thiophosphoamide or dihydrochloride-1,6-di-(chlorethyl)-


amino--l,6-desoxymannitol and X-rays, respectively, and found an

increasing, incidence of induced aberrations with age. This might

imply decreased DNA repair with increased donor age.

Until the significance of SCE's are more clearly understood

it will be difficult to interpret Schneider and Kran's results

conclusively. It is clear that more information is needed about

repair of base damage and aging in terms of different types of base

damage and in terms of different facets of the response to DNA

damage. The experiments reported here were designed to obtain

additional information of this type.



For the work reported here a technology that would provide as

much information as possible about DNA metabolism following DNA

damage was desirable. Sensitivity was a prime consideration so

that low levels of damaging agents could be used. Response to

DNA damage could then be measured at low cytotoxicity. Since

chemical agents would be used to damage DNA (see below), the use

of unlabeled chemicals was desirable so as to avoid the high cost

of labeled mutagens. The alkaline elution technique developed

by Kohn and associates83-84 was seen to provide all these advan-

tages. As described in detail in Methods and Materials, this tech-

nique.allows a sensitive measure of the degree of single-strand

fragmentation of DNA. Fragmentation of DNA below 1-1.5x10 daltons

can be detected. This is considerably higher than the upper limit

of 5x10' daltons obtainable with alkaline sucrose gradient


By labeling DNA of growing fibroblasts with 14C thymidine

before chemical treatment of the cells, fragmentation and reseal-

ing of long-term labeled, parent strand DNA, which occurs'during

prereplication excision repair, could be monitored. By giving a

pulse of 3H thymiaine to these same cells at various times after

chemical treatment, elongation of undamaged, newly synthesized

DNA strands could be followed. In these c.:-errirments this latter


synthesis occurred opposite a previously damaged, parental template.
Finally, by comparing the total amount of pulse H thymidine to
the total C thymidine taken up by the cells during the long term

pre-labeling period, a measure of the overall rate of DNA synthesis

at various times after chemical treatment could be obtained.

Since this was to be a study of DNA repair as a function

of in vivo aging, appropriate animals were needed as a source of

skin fibroblasts. A long-lived strain with no abnormal pathology

was required in order to assure that the results reflected natural
aging processes. Female, Fischer 344 rats were chosen largely

because of the availability of aged animals through the National

Institute on Aging. The youngest and oldest animals available,

three weeks old and two years old, were used in order to maximize

the chances of dEtectin- differences that might exist.

The damaging agents used were N-ethyl-N-nitrosourea (ENU)

(figure 1), il-acetoxy-2-acetylaminofluorene (N-AAAF) (figure 2),

and benzo(a)pyrene-anti-diol-epoxide (BPADE) (figure 3). A variety

of DNA adducts are produced by ENU (figure 1). The best studied

of these are N7-ethylguanine, N3-ethyladenine, 06-ethylguanine

and ethyl phosphotriesters. The N3-ethyladenine and 06-ethylgua-

nine adducts are enzymatically removed from DNA while N7-ethylgua-
nine is lost largely by spontaneous depurination. Ethyl phospho-
88 -Ok
triesters are not removed from DNA. Two adducts are produced

by N-AAAF (figure 2).90 These are thought to be enzymatically
removed from the DNA of human and rodent fibroblasts. Figure 3
lists the adducts usually found in DNA after treatment with BPADE.

0- N-N-C-N

N- Ethyl -N-Niirosourea

Derivative % of Total Ethylation

06- EtGuanine 7
02- EtThymine 6
04-EtThymine 1.4
02-EtCytosine 0.5
Ethyl Phosphotriester 70
N7-Et Guonine (14)
N3-EtAdenine (3)

Figure 1. Structure of ENU and the percentage distribution of the
adducts formed by ENU in double-stranded DNA. Parentheses
indicate an approximate division of the 17% total N-ethylguanine
adducts based on earlier work by the same author.



F.- o -

0 0 a

0 r: 41-

4C o
=u) 0)

- O a 8 0

0 a* r

0 0

0 S '0V?

c:/ Z .r.
/ tt *" ? ^s a u0
--( o *^; xx "S
/ ---\,- ii 0 c o
(f \ <>
\ t ^ l T'l
^ // z ;ss I










c .)

I -
o O
i 0

8 0

o 0


o z


o 0
or 0


I o

o- -

.- -


The major guanine adduces are Ye.woved to some extent from rodent93

and human65 calls. In addition, a similar adduct is formed at the

N7 position of guanine which is apparently very unstable and

leads to rapid depurination from the DNA.94

Alkylation and arylalkylation adducts are expected to repre-

sent different classes of DNA damage since they would create dif-

ferent degrees of distortion of the DNA double helix. For example,

Helfich has shown that treatment with N-AAAF or BPADE creates Si

nuclease sensitive sites in DNA whereas N-methyl-N'-nitro-N-nitro-
soguanidine (MNNG) treatment does not. (The adducts created by

MNNG are similar to those from ENU. A methyl group is transferred

by the former, an ethyl group by the latter. Distortion of the DNA

helix is expected to be very similar for the two.) Less helix

distortion may result from the respective N2-arylalkylgunnirie

adducts since it is the N6-arylalkyladenine96 adduct of BPADE and
the C8-arylalkylguanine adduct of N-AAAF that give rise to Sl

nuclease digestible regions.

These results do not imply that ENU-induced damage has no

effect on DNA architecture. Alterations of base pairing and second-

ary structure are expected to result from alkylation by ENU, and

this has been demonstrated experimentally for 06-ethylguanine,
and for 04-ethyluracil and 02-ethyluracil adducts. However, the

arylalkyl adducts apparently produce much greater distortion.

Repair of these two different classes of damage might be anal-

ogous to repair of base damage due to ionizing radiation versus

repair of U.V.-induced pyrimidine diners. The existence of the


genetic conditions Xerederma Pigmentosum (exhibiting excision repair

of ionizing radiation-induced base damage99 but not of pyrimidine

dimers) 100and Ataxia Telangiectasia (exhibiting normal excision

repair of pyrimidine dimers but defective excision repair of gamma

radioproducts) suggests that these two classes of DNA damage may
require different repair systems. Additionally, Xeroderma
Pigmentosum cells may be able to repair MNNG-induced damage but

not N-AAAF-induced damage. Ataxia Telangiectasia cells can
103 104
repair N-AAAF-induced damage but not MNNG-induced damage.

Therefore, the genetic definition of these two classes of DNA

damage can be extended to include the alkyl and arylalkyl adduct

damage employed in the studies reported here. Unfortunately, re-

cent evidence indicates that this classification of DNA damage

and repair is too simple. Pyrimidine dimers and N-AAAF damage may

be removed from mammalian DNA by separate repair systems having

different rate-limiting steps.103,105

As research continues it is inevitable that further distinc-

tions will be made. It is possible that one or the other of these

repair systems may contribute preferentially to senescence.



Three week old, female, pathogen free, Fischer 344 rats were

obtained from Charles River Breeding Labs. Age upon arrival was

checked by comparing the weights of the animals against a normogram

provided by the breeder. Upon arrival the rats weighed 35-40

grams each. They were housed in the animal care facilities of the

J. Hillis Miller Health Center for 1-2 days before obtaining tissues

for culture.

Two year old, pathogen free, female, Fischer 344 rats were

obtained from Charles River Breeding Labs through the National

Institute on Aging. These were from a special colony of rats main-

tained for aging studies. The two year old rats were not checked

for age upon arrival since no normogram was available for this age

group. They were simply shipped from Charles River at two years

of age. They arrived at our facility weighing approximately

270 grams each and were housed for 2-3 days before tissue samples

were obtained. The mean 50% life span for female, Fischer 344

rats is very close to that for males which is reported to be '

29 months.106

General Tis.L1ue Culture Methods

All cultures were maintained in Eagle's minimum essential

medium containing 70 mg penicillin G and 50 mg streptomycin per

liter plus 15 mM HEPES, pH 7.4, (MEM) plus additions as indicated.

All cultures were kept in a humidified 37 C incubator with a

90% air plus 10% CO2 atmosphere.

All trypsinization was done with a solution of 0.1%(w/v)

trypsin in versene which was filtered through two 0.2 micron pore

Nalgene filters before use.

All sera were heat inactivated at 560C for 45 minutes before


Phosphate buffered saline (PBS) is 0.15 M NaCl + 0.005 M

K2HPO4, pH 7.4. Versene is 0.137 M NaCl + 1.15 mM KH2PO4 + 2.68 mM

KC1 + 8.10 mM Na 2HPO4 + 0.478 mM Na 4EDTA, pH 7.4.

Isolation Of Tissue Samples

Tissue samples for the initiation of fibroblast cultures

were obtained by the following method.

After washing with 70% ethanol, the ears of an anesthetized

animal were clipped off with scissors and bathed for two'miniutes

in 70% ethanol contained in a sterile petri dish. They were then

rinsed for two minutes in sterile PBS also contained in a sterile

petri dish. The ears were then split into two pieces and allowed

to soak in the freezing medium (MEM + 10%(v/v) glycerol) for 10-

20 minutes prior to freezing in a liquid nitrogen bath. All these

procedures were carried out in a laminar flow, sterile, tissue

culture hood.

For storage the samples were held in the vapor phase of a

liquid nitrogen refrigerator.

Initiation Of Fibroblast Cultures

Rat ears were thawed, placed in a sterile petri dish contain-

ing 10-20 ml MEM + 0.1%(w/v) collagenase, and sliced with a sterile

scalpel into pieces approximately 3-4 mm on a side. The ear pieces

plus media were transferred to a sterile, square, glass bottle with

a magnetic stirrer bar in the bottom. This was then placed in a

37 C warm room and allowed to stir for 45-60 minutes. (The

square shape of the bottle seemed to aid separation of the tissues

by physical agitation during digestion.) After digestion the

contents of the bottle were filtered through sterile gauze into

a screw cap, conical centrifuge tube and centrifuged for 10 minutes

at 1000 rpm in a model PR-2, refrigerated International centrifuge.

The cell pellet was resuspended in 10 ml MEM + 10%(v/v) fetal

calf serum (FCS) and transferred to a T-25 (25 cm growth area)

plastic tissue culture flask. Due to the different sizes of three

week old and two year old rat ears, one three week old ear gave

rise to one such T-25 culture, and one two year old ear gave rise

to two such T-25 cultures. Thus all cultures were initiated with

approximately equal amounts of starting material. Cultures ini-

tiated in this manner reached confluency in 1-2 weeks and were used

within one week thereafter.

Occasional cultures were checked for mycoplasma by the aridine
phosphorylase method of Levine. Although they were found to

have high levels of this enzyme, the activity was resistant to

treatment of the cells with heat 09 or chlorotetracycline.110

Therefore, the cells were judged to be free of mycoplasma contami-


Growth, L.a1Wim, Ail Trreatmc nt Of Cells

At the beginning of each experiment a confluent T-25 culture
was trypsinized and the cells transferred to a T-75 (75 cm growth

area) plastic tissue culture flask. A final volume of 20 ml MEM

+ 5% FCS + 5% calf serum (CS) was added, and 14C thymidine (Tdr)

was then added to a final level of 1.25 nCi/ml. The following day

the medium was replaced with 20 ml MEM + 5% FCS + 5% CS and 14C Tdr

added again to the same level. On the third day the medium was

pipetted from the flask, the cell monolayer washed with 10 ml

fresh medium, and 20 ml MEM + 5% FCS + 5% CS + 10-6 M Tdr was added.

On the fourth day the medium was removed, and the cells were tryp-

sinized and transferred in equal portions to fourteen T-25 flasks

in 10 ml MEM + 5% FCS + 5% CS + 10-6 M Tdr. On the fifth and sixth

days the medium was removed from the flasks and replaced with

10 ml fresh MEM + 5% FCS + 5% CS + 10-6 M Tdr. This segment of

the protocol is illustrated in the upper portion of figure 4.

Overall, during these experiments the cells were labeled with
C Tdr for two days (approximately two generation times) and

then chased with 10-6 M Tdr for four days.

The lower portion of figure 4 illustrates the timing of the

treatments given to the cells relative to the 3H Tdr pulse labeling

and harvest at the end of the treatment period. Specifically,

the chase medium was transferred from the growth flask to a new

T-25, and the treatment medium (see below) was added to the growth

flask. (All chemical treatments were given in MEM without serum

to avoid reaction with serum components which might create uncon-

trollable variation in the effective dose reaching the cells.)111

a a

1 rI



; c

* I *

I o

I.-- -


41U 49
4 :3 4-)

13 b0 0 <1

0 44 c
0- UO ,-I JJ Cfl

Ha 0 Q

4-1 >)
0 0 0- 0

AO) 00 ,C1

-4 P C 413U

0-4 W W

4J 0 4) 5 'A
4-1 # <0
0 4-1 : (3 3
a) 4 *r4 41)
4-) $40U

,O 'a 3 e
o 0 'I 0 0

ato :j r.
E-4 19 la
N.40 U*0

0 1 0

1 C rU ).

a) a> s14
< 0 4) 03
Ai ccp.a

co I3 n *a

10 CA 4
*0 E 0

GJ I 00
41 u 3l 4
a 4 1U r4

o o. _4*r
w 4-1 4 R
o x 0 4-1

0, DOS E-4 -r

c* r-4 p (
4.4 Z 3 0 0
4a) *.C .-1
0 U 04

NO 1, 1

-ri c CHi L ()
3 ,g u 'a

0 r4 0 0
p4) T 4J41
i- 4-1 *<
U- (U-(3

D 4.) 54J
0 *i- 08 4-1
0,-1 njT 4-1 Q
0.4 E -
0 .4


I ..



. c

I.) -

! -






Both flasks were then placed in a 370 C humidified incubator

(90Z air + 10% CO2) for 15 minutes. The treatment medium was then

removed from the cells, and the conditioned medium transferred

back to the growth flask for further growth and chase as needed.

freatrment media were prepared as follows. Ethylnitrosourea

was freshly dissolved before use in 0.127 M Na2HPO4 + 0.0369 M

citrate, pH 6.0 (pH 6 buffer).112 One milliliter of this solution was

added to 9 ml HEM to give 200 micrograms/ml final concentration.

Acetoxy-acetylaminofluorene was dissolved in anhydrous dimethyl-

sulfoxide (DMSO), and aliquots were stored at .-200C. Five micro-

liters of this solution was added to 10 ml MEM to give a 4 micro-

molar solution of N-AAAF. Benzo(a)pyrene-anti-diol-epoxide was

dissolved in anhydrous DMSO, and aliquots were stored at -20oC.

Ten microliters of this solution were added to 5 ml MEM to give a

1.35 micromolar solution of BPADE. For sham-treatment, control

experiments, pH 6 buffer or DMSO was added to MEM at the appro-

priate concentrations.

The first treatment was given the evening of the sixth day

and ended 16 hours before harvest on the seventh day. All treat-

ments were given in duplicate. One and a half hours before harvest

H Tdr was added to all flasks that had received treatment up to

that time and to the untreated, control flasks at a final level

of 0.2 microcuries/ial. (The levels of C Tdr and H Tdr were

chosen so as to minimize toxicity due to radiation damage but

yet provide adequate levels for accurate counting of samples.)

After 30 minutes the medium was removed from all flasks, and 10

ml of fresh MEM + 5% FCS + 5% CS + 2 micrograms/ml Tdr were added.


One hour was allotted for this final chase period. During this time

two sets of flasks that had not received H Tdr were given treatment

medium for 15 minutes as described. These treatments, ending at 0.5

and zero hours before harvest, therefore received 45 minutes chase at

the end of the experiments instead of one hour. At the end of the

treatment period all cultures were approximately 90% confluent. The

overall features of this protocol were similar to that used by

Cerutti et al..113

The harvest was accomplished as follows. The culture flasks

were taken from the 37 C incubator and placed immediately on

crushed ice (cell side down). Medium was pipetted from the flasks,

and 1 ml of ice cold trypsin solution (0.1% in versene) was added.

Trypsinization was carried out over a five minute interval on ice

by slowly rocking the flasks in order to evenly distribute the

trypsin solution over the cell monolayers. The flasks were then

tapped on the side to completely detach the cells, and 5-10 ml

ice cold PBS was added to dilute the trypsin.

Cell Lysis

The cells were immediately poured onto 2 micron pore poly-

vinylchloride filters (Millipore Corp.) that had been previously

cooled by addition of 2 ml ice cold PBS. (Each filter was con-

tained in a modified Gelman, filter, vacuum funnel apparatus)

The PBS was allowed to drain through the filter, room lights were

turned off, and the cells were lysed on the filter by addition

of 5 ml of 0.2% Sarkosyl + 2 M NaCl + 0.02 M EDTA, pH 10. After

the lysis solution drained from the filters by gravity they were

washed with 3 ml of 0.02 M EDTA, pH 10. The outflow tubes of the

filter funnels were then connected to a peristaltic pump, and the

funnels were covered to exclude all light. A solution of 0.04 M

1H.EDTA + sufficient tetrapropylammonium hydroxide to make p1l 11.85

was then pumped through the filters at 0.05 ml/min and 10 fractions

collected at 90 minute intervals each. All noprations after and

including cell lysis were carried out in subdued or excluded light

to prevent fluorescent light-induced scissions of DNA which occur

under alkaline conditions.83

Sample Collection And Counting

The 10 elution fractions were mixed with 13.5 ml Aquasol

(New England Nuclear) containing 0.3%(v/v) glacial acetic acid

for counting. (Glacial acetic acid was added to minimize alkali-

induced chemiluminescence in Aquasol.)83 The filters were trans-

ferred to scintillation vials, 0.4 ml of 1 N HC1 were added, and

the vials were capped and incubated at 70C. for one hour. After

cooling to room temperature, 2.5 ml of 0.4 N NaOH were added, and

the vials were allowed to sit at room temperature for 30 minutes

before adding 13.5 ml Aquasol containing 0.3%(v/v) glacial acetic

acid. Filter funnels together with connecting and peristaltic

pump tubing were washed by pumping through 16 ml of 0.4 N NaOl

in 4 ml fractions. To the wash fractions were added 0.5 ml

glacial acetic acid, to minimize the additional chemiluminescence

found in these samples, and 13.5 ml of Aquasol containing 0.3%(v/v)

glacial acetic acid. All samples were counted in plastic scintilla-

tion vials in a Beckman LS 8000 Liquid Scintillation System. The

cell lysis, elution, and sample preparation procedures, except

for minor modifications, were essentially those developed by

Kohn et al..8384


Alkali-Labile Site EKJperiments

To test for alkali-labile sites, cells were labeled, grown,

and lysed on filters, and the filters washed with 0.02 M EDTA,

pH 10, exactly as described above. When treatment was with ENU,

the filters were then treated with ENUI in pH 6 buffer for three

minutes. This solution was allowed to drain under gravity from

the filters which wdre then washed with 0.02 M EDTA, pH 10, and

eluted as usual. When treatment was with N-AAAF or BPADE, the

filters were washed with PBS after the EDTA wash, treated with

N-AAAF or BPADE for 3 minutes, washed with PBS, then washed again

with 0.02 M EDTA, pH 10, and then eluted as usual. The N-AAAF

and BPADE were prepared as stock solutions in anhydrous DI,1SO.

Appropriate aliquots of these stock solutions were added to PBS;

the resulting solution was applied to the filters. Appropriate

sham-treatments with pH 6 buffer or DMSO in PBS were included in

the respective experiments.

Data Reduction

After background subtraction by the scintillation counter, cor-

rection for overlap of 3H counts into the 14C channel and vice

versa and conversion of counts per minute to disintegrations per

minute was accomplished by a Wang 600 program written by the author.
14 3
The program also determined the total C and H disintegrations

per minute (DPM) contained in the 10 elution fractions plus-the

filter digest plus the four funnel wash fractions and, using this,

calculated the fraction of the total 14C or 3H DPM remaining on

the filter after the collection of each of the 10 fractions.


The fraction of the total 14C or 3H DPM remaining after each

fraction versus fraction number was plotted as in figure 5. From

plots such as this, the fraction C DPM remaining after 10.5 hours

elution (sample #7) was used to calculate the fraction 14C DPM

eluted at that time. Also, a line tangent to the 3H DPM curve

was drawn to intercept the ordinate as shown. From this intercept

the fraction of 3H DPM not rapidly eluted was determined, and
this was used to calculate the fraction of H DPM raidly eluted
from the filter. As mentioned, the C labeled DNA represents

parental strand DNA while the H labeled DNA represents newly

synthesized DNA strands.83'84

The fraction eluted DNA for each culture within an experiment

was plotted as the ratio of the experimental over the control

values for fraction eluted. As stated above, untreated, control

cultures were used in each experiment. This ratio method allows

a normalization for the baseline, systematic variation between

experiments and also produces some increase in sensitivity.

Tests for statistically significant differences employed

the Mann Whitney U statistic. The only assumptions required

for proper use of this statistic are that the sample sets being

compared should be randomly drawn from populations having the

same, continuous distribution. This is a reasonable assumption

for this application since the experiments for young and old donor

cells were all performed in identical fashion and since.the data

were collected and manipulated in the same manner.


* 0 0 e '14C DNA



2 4 6 8
Somple No. (4.5ml, 0.05 ml/min)

Figure 5. Sample elution curves. The fractions of the total 14C
or 3H DPM remaining on the filter after the elution of each
fraction are plotted on a logarithmic scale versus sample
.number. A line tangent to the 3H DPM curve is drawn to intercept
the ordinate. The intercept value is used to calculate the
fraction of the total 3H DNA that is rapidly eluted. The
fraction of 14C DPM eluted after collection of sample number 7,
after 10.5 hours of elution, is used as a measure of the degree
of fragmentation of 14C DNA.





or 199 U/mg,

type II) and


or 215 U/mg)

were purchased

from Worthington



propylammonium hydroxide

N.Y. or Eastman Kodak.

was purchased


from RSA Corp.,

was obtained


as a 30% solution

from K&K Laboratories.

N-ethyl-N-nitrosourea was purchased






fluorene was


from ICN Pharmaceuticals.


anti-diol-epoxide was


from IIT Research




All serum was purchased

from Gibco.






was purchased

from Mallinkrodt.


G was


from Sigma


Co., Streptomycin

was purchased

from ICN





(55 Ci/mmole),

and 2- 14C-thymidine

(56 mCi/mmole)

from Amersham.






was purchased


Treatment of cells with ENU leads to fragmentation of parental

DNA as seen in figure 6. As stated in the Methods section, results

are plotted as the ratio of the fraction DNA eluted in the untreated,

control cultures for each individual experiment versus time after

treatment. A ratio of one on the ordinate therefore represents the

control value. Ratios greater than one indicate fragmentation of

DNA above the level found in untreated cultures. Curves for fibro-

blasts from both three week old and two year old rats are shown here.

At earlier times after treatment parental DNA is fragmented at

over twice the level of the untreated cells and approaches control

levels as time passes after treatment. Two phases can be distin-

guished in this return towards normal elutability. An early, rapid

phase ending about 3 hours after treatment and a slower phase there-

after. Young and old donor cells did not differ in the earlier

phase, but statistically significant differences were found at 6

and 16 hours after treatment during the slow phase. The differences

between the means were 28% and 38% of the control value for 16 and

6 hnurs after the ena of treatment, respectively. As seen, the old

donor cells exhibitt less fragmentation of parental DNA at later times

after treatment than do young donor cells.

To interpret: these results, consider that the degree of frag-

mentation of parent CNA seen at any given time point is the average


3 Week Old
A 2 Year Old

2.0 -

( '0.05)

W 1.01-- 0.051

-- -

0 05 L5 3 6 16
Hours After END of Treatment (200pgj/ml ENU, 15rmin ,37C)

Figure 6. Normalized fraction of parental DNA eluted after 10.5
hours of elution versus time after treatment with ENU. The
ratio of the fraction of parental DNA eluted in treated cultures
to that eluted in untreated cultures is plotted on the ordinate.
A ratio of one on the ordinate represents the degree of frag-
mentation found in untreated, control cultures. Points represent
means standard error (n=10 for all points except the 6, 3,
1.5, and 0.5 hour time points for 2 year old rats where n=9).
At alpha=0.05, the difference is significant at the 95% con-
fidence level. All tests for significance employed the Mann
Whitney U statistic.


sum of the fragmentation and reselling processes occurring in a

culture up to that time. This occurs during the prereplication,
excision repair process which involves endonucleolytic incision

of the DNA adjacent to the damaged site, exonucleolytic removal

of the damaged region, replacement with new DNA by DNA polymerase,

and resealing by the action of DNA ligase.55 It is the overall

sum of the incision and ligation events that are represented in

these curves. A higher degree of overall fragmentation indicates

relatively greater incision activity than ligation activity.

Greater fragmentation of parental DNA at later times in young donor

cell cultures may, therefore, indicate higher residual endonucleolytic

activity. This would be consistent with greater repair activity,

at later times after treatment, in young donor cells than in old

donor cells. It is important to note that, based on published

data for human cells,115 it is expected that more than half of the

ENU-induced DNA adducts still remain in the cells used here at 16

hours after treatment.

Biphasic kinetics of DNA repair as seen here have been observed

in a number of studies. 65,103,116-121 The two phases may represent

repair of damage located in physically different regions of the

genome. For example, damage located within nucleosomes may be

less accessible to repair enzymes than that located in linker

regions. Experimental evidence for this has been observed in some

systems61-64 but not in others65,122 as mentioned in the intro-

duction. The rapid phase could represent repair of damage in the

more accessible, linker regions while the slow phase could repre-

sent repair of damage in uucleosomal core DNA. The two phases

may represent repair of dadnie in euchromatin and heterochromatin,


respectively. It may be thac J~,':'.c located in heterochromatic,

untranscribed regions of the genome are less accessible to repair

enzymes. Experimental evidence for this was mentioned earlier.

Alternatively, the two phases may represent repair of different

species of DNA damage. As seen in figures 1 and 2 both N-AAAF

and, especially, ENU create a variety of DNA adducts.

As noted earlier, 70% of the alkylation of DNA by ENU is on

phosphate groups. Phosphotriesters thus produced are alkali-
labile.123 The t1/2 for hydrolysis of phosphotriesters at pH 12.6

is one hour.123 Since elution of DNA from the filters is carried

out under alkaline conditions, alkali-catalyzed scission of phos-

photriesters could account for part of the observed fragmentation.

Figure 7 shows the results of an experiment carried out to see if

significant fragmentation of ENU-treated DNA would occur under the

conditions used here. In this experiment cells were labeled with
C Tdr, lysed on filters, and the DNA treated with ENU as described

in Methods and Materials. Significant fragmentation of DNA did

occur which increased as the concentration of ENU was increased.

Although the range of concentrations used here includes that used

in treating cells (200 micrograms/ml), no direct comparison can

be made as to the levels of damage incurred in the two types of

experiments. In the experiment of figure 7 naked DNA was treated

with ENU, while in the experiments of figure 6 whole cells were

treated. The effective concentration of ENU within the whole

cells is expected to be less than that when naked DNA was treated

with the same concentration of ENU. As mentioned earlier, ethyl-

phosphotriesters are not known to be removed from cellular DNA.8889



pH 6 Buffer Only
20p.g//ri ENU

-" 0 lOCt0/ml ENU

2 "A OOp ml ENU

" 0*-0 400g/mi ENU

0 2 4 6 8 10
Soample No. (4.5 ml, 0.05ml/min)

Figure 7. Log of the fraction DNA remaining on the filters after
collection of each sample versus sample number. Naked DNA on
the filters was treated with ENU as described in Methods and
Materials. Points represent the means of duplicates.


Therefcire, alkali-induced scission cannot accou'.t for the decreas-

ing fragmentation seen in figure 6. However, if the experiments

were carried out over a longer post-treatment incubation period,

one might predict that the curves for both young and old donor cells

would eventually level off at the same degree of fragmentation.

This level would be higher than that for the untreated controls

and would be due to persistent ethyl phosphotriesters.

Since ENU was given to the cells in medium devoid of serum

it was necessary to do an experiment in which the cells were treated

with medium only. Since the ENU was dissolved in a pH 6 buffer,

as described earlier, the appropriate amount of this buffer was

added in this experiment. The results for parent strand fragmenta-

tion are shown in figure 8. As can be seen, the effect is negli-

gible compared to that obtained with ENU.

The effect of ENU on daughter strand elongation can be seen

in figure 9. The old donor cells exhibited greater elongation

of daughter strand DNA which essentially reached the length found-

in untreated, control cultures by six hours after treatment.

Daughter strand DNA in young donor cells did not reach control

levels during the course of the experiments. Statistically signi-

ficant differences were found between all pairs of time points

as indicated in the figure. The differences were 16%, 31%,- 23%,

and 21% of control values at 10, 6, 3, and 1.5 hours after treatment,


The rate of elongation of DNA synthesized within a 1.5 hour

time block which commences at various times after treatment is

being measured here. This DNA is synthesized opposite a previously


00.5 1.5 3 6 16
Hours After END of Treatment (pH6 Buffer [0I% V/v ,15 min, 37*C)

Figure 8. Relative fragmentation of parental DNA after sham-treatment
with the pH 6 buffer (see Methods and Materials) used in the ENU
experiments. Results are plotted as in figure 6. Points repre-
sent the means of duplicates.

F344 FEMALt-" c'-iROR1 ASTS
3 Week Old
A 2 Yeor Old


ooi1 -----

(cC. O.CA)

0 Q5 1.5 3 6 16
Hours After END of Treolment (200/.g/ml ENU, 15min., 370C)

Figure 9. Normalized fraction of undamaged, newly synthesized,
daughter strand DNA rapidly eluted versus time after treatment
with ENU. The ratio of the fraction of daughter strand DNA
rapidly eluted in treated cultures to that rapidly eluted in
untreated, control cultures is plotted on the ordinate. A
ratio of one on the ordinate thus represents the amount of
rapidly eluted, daughter DNA found in untreated, control
cultures. Points represent means- standard error (n=10 for
all points except the 6, 3, and 1.5 hour time points for 2
year old rats where n=9). Statistical analysis was as des-
cribed for figure 6.



-- --f


damaged template. If damage remains in the I'NA it could create

a block to the elongation of daughter strands. If the parental

strand daiiage were removed or if the DNA replication machinery

were able to bypass the damaged region of the template, no block

to daughter strand elongation would occur.

Replication of DNA opposite a damaged, parental template is
known to occur, and several possible mechanisms have been pro-

posed.125 One is a recombination type mechanism where a segment

of the daughter strand of one duplex and a correspondiiig section

of the damaged strand exchange places so that replication may

continue. The result of this would be the appearance of damaged

residues in untreated, daughter strand DNA. Experimental evidence

for this in mammalian cells has been presented. 126127 However,

it has been argued that those results are also consistent with by-

pass of template damage without the need to postulate recombinational

This could be accomplished if bases were inserted in random

fashion opposite the damaged region merely to get on with repli-

cation. This would result in the introduction of inappropriate

bases in the daughter strand and would, therefore, increase the

error level of the cell. It would be analogous to "SOS," error
prone postreplication repair in bacteria. Some evidence-for an

inducible, error prone DNA repair in mammalian cells has also been

presented.130 Another bypass mechanism would be for the daughter

strand of one duplex, having replicated past the region on its

parent strand corresi.onding to the darmge.d region on the other

parent strand, to fold back at the replication fork and base pair

with the daughter strand of the dor4acp.ed duplex. Replication could

thus continue on both daughter strands as far as it had on the first
one thereby bypassing the damaged region. These three mechanisms

are all possible pathways of postrepilication repair (see Discussion).

If one accepts that the young donor cells are more active at

removing parent strand damage (figure 6), then one possible inter-

pretation would be that the old donor cell replication machinery

is more adept at bypassing template damage than that of young donor

cells. An alternative interpretation is that single-stranded

regions formed during the excision repair process create blocks to

the elongation of daughter strands. Since the young donor cells

incise damaged DNA more actively (as seen in figure 6), there

would be more blocks to daughter strand elongation in these cells.

Figure 10 shows the effect of sham-treatment on daughter

strand DNA elongation. As can be seen there is hardly any effect

at all.

Treatment of cells with DNA damaging agents in known to cause"

suppression of DNA synthesis.132 It was of interest to know if the

overall rate of DNA synthesis would be less in old donor cells after

treatment. The ratio of the total 3H DPM/total 14C DPM relative to

that of the untreated controls after ENU treatment is plotted on

the ordinate in figure 11. (Again, a ratio of one on the ordinate

indicates the untreated, control value.) Total DNA synthesis was

suppressed equally in both young and old donor cells. Although

there was some recovery after six hours, DNA synthesis was suppress-

ed slightly more after 16 hours than at earlier time points.


i* I I_.__..... .... ._ _
005 15 3 6 -- 16
Hours After END of Treatment (pH6 Buffer [lO% V/v, 15minm 37-C)

Figure 10. Relative amount of rapidly eluted, newly synthesized,
daughter strand DNA found after sham-treatment with the pH 6
buffer (see Methods and Materials) used in the ENU experiments.
Results are plotted as in figure 9. Points represent the
means of duplicates.

3.0 I




. (

3.0 -

S005 1.5 3 6 1E
Hours After END of Treotment (200g/ml ENU, 15 min., 376C)

Figure 11. The ratio of the total 311 DPM/total 14C DPM in ENU-
treated cultures divided by that in untreated, control cultures
versus time after treatment. A ratio of one on the ordinate
represents the overall rate of DNA synthesis, during the last
1.5 hours of the experiments, in untreated, control cultures.
Points represent means standard error (n=10 for all points
except the 6 and 3 hour time points for 2 year old rats where
n=9). Statistical analysis was as described for figure 6.

a 3 Week Old
A 2 Year Old



2.0 -


,' .. !



Figure 12 shows the suppression of DNA synthesis found in

sham-treated cultures. Significant suppression is seen early after

treatment, but DNA synthesis returns to the level found in untreated

cultures by 16 hours. This obviously was not the case with ENU-

treated cultures.

Experiments using N-AAAF and BPADE were performed in essentially

the same manner as those using ENU. The only difference being

that N-AAAF and 32ADS were dissolved in DMSO instead of pH 6 buffer

before addition to MEM in the culture flasks.

Fragmentation of parental DNA after N-AAAF is seen in figure

13. The response obtained with the dose used was much greater than

seen in the ENU experiments. This could partially account for the

greater variance found for the young donor cell curve in figure 13.

However, this variance was largely accounted for, as was the variance

in all these experiments, by variation between experiments rather

than variation between duplicates within a given experiment. A

different primary culture was utilized to derive the cells used for

each experiment. This fact alone could account for the observed

variation. However, it was necessary during the course of the

experiments to use different batches of media and sera which may

have contributed even further variation. Nevertheless, when analyzed

on a statistical basis, the differences between means were found

to be significant for all time points at the levels indicated in

figure 13. Differences between the means were 153%, 235%, 298%,

342%, 288%, and 461% of the untreated, control values for 16, 6, 3,

1.5, 0.5 and zero hours after the end of treatment, respectively.

Greater fragmentation of parental DNA was seen at all time points



S--L_ I I I
00.5 1.5 3 6 16
Hours After END of Treatment ( pH6 Buffcr[l0% % 15 mmin., 37C)

Figure 12. Relative rate of DNA synthesis during a 0.5 hour time
interval at various times after sham-treatment with the pH 6
buffer (see Methods and Materials) used in the ENU experiments.
Results are plotted as in figure 11. Points represent means
of duplicates.

2.0 1-


o10 -

9.0* 3 Week Old
9. .025) A 2 Yeor Old


(- -0.025)

S\ (co0.025)

(5c 0.025)




005 15 3 6 16
Hours After END of Treatment (4;Lm N-AAAF, 15 min.,37C)

Figure 13. Normalized fraction of parental DNA eluted after 10.5
hours of elution versus time after treatment with N-AAAF.
Results are plotted as in figure 6. Points represent 'leans
standard error (n=4). Statistical analysis was as described
for figure 6.


in the young donor cells; a result very similar to that found after

treatment with ENU. In these experiments the initial degree of

fragmentation in young donor cells at zero hours after the end of

treatment was considerably greater than in old donor cells. This

was not seen in the ENU experiments. In addition, at 1.5 hours

after the end of treatment there appears to be a resurgent increase

in fradgne, tation over that seen at 0.5 hours; also seen to a lesser

extent in the old donor cells. Although the magnitude of the In-

crease is less than the standard error in either case, this trend

could possibly have arisen from the greater disruption of chromatin

structure expected with N-AAAF and/or, secondarily, from the much

greater repair response obtained. A greater challenge to the repair

systems may have allowed the finer aspects of frr'w.Mentation kinetics

during the early phase to become manifest. With ENU which would

create less distortion of the helix, the young and old repair

systems were apparently functionally equivalent during the initial

phase. Calculations based on data reported by Goth-Goldstein,

Singer, and Amacher et al. indicate the level of total adducts

formed per mole of DNA phosphate by ENU to be approximately twice

that formed by N-AAAF at the concentrations used in these experiments.

These are only rough estLmates and depend upon the assumption of a

linear relation between concentration and number of adducts formed.

They suggest that a much greater repair response is elicited by

relatively few N-AAAF adducts when compared to ENG.

Although no reports of N-AAAF-induced alkali-labile bonds

have appeared, this possibility was examined.. The results of treating


naked DNA on filters with N-AAAF, as described in Methods and

Materials, is shown in figure 14. Increasing rate of elution

of DNA from the filters is seen with increasing concentration

of N-AAAF. Judging from these results, the degree of alkali-

induced fragmentation of DNA after N-AAAF treatment in the experi-

ments reported here would be much less than after ENU treatment.

The effect of this fragmentation on the results obtained would

be essentially as discussed for the ENU experiments.

Although some fragmentation of DNA after DMSO sham-treat-

ment is apparent (figure 15), it is not greater than 21% of the

control value and certainly does not account for the results

seen with N-AAAF treatment.

Elongation of daughter strands after N-AAAF treatment (figure

16) was less than after ENIU treatment; however, the results were

qualitatively the same. Daughter DNA was elongated to a greater

degree in old donor cells, and statistically significant differences,

as indicated in figure 16, were found between all pairs of time

points., Differences between the means were 81%, 51%, 102%, and

99% of control values at 16, 6, 3, and 1.5 hours after the end of

treatment, respectively. Sham-treatment had essentially no effect

on daughter strand elongation as seen in figure 17.

Suppression of total DNA synthesis was greater after N-AAAF

(figure 18) than after ENU treatment. Statistically significant

differences between the means were found for the 16 and .3 hour

time points. However, these differences, 18% and 6% of the control


-- --- DMSO (0.4% V/v)

"A 2pm N-AAAF

"** 4p.m N-AAAF

2 4 6 8 10
Simple No.(4.5ml, 0.05mW/min)

Figure 14. Log of the fraction of DNA remaining on the filters
after collection of each sample versus sample number. Naked
DNA on the filters was treated with N-AAAF as described in
Methods and Materials. Points represent the means of duplicates.




i I .. .
00.5 1.5 3 6 16
Hours After END of Treatment (D/-O [0.1% /V] 15 min., 37 C)



Collagenase (190 or 199 U/mg, type .II) and trypsin (205, 212,

or 215 U/mg) were purchased from Worthington Biochemicals. Tetra-

propylammuonium hydroxide was purchased from RSA Corp., Ardsley,

N.Y. or Eastman Kodak. Sarkosyl was obtained as a 30% solution

from K&K Laboratories. N-ethyl-N-nitrosourea was purchased from

Biochemical Laboratories, Bohemia, N.Y.. N-acetoxy-2-acetylamino-

fluorene was purchased from ICN Pharmaceuticals. Benzo(a)pyrene-

anti-diol-epoxide was obtained from IIT Research Institute,

Chicago, Illinois. All serum was purchased from Gibco. Dimethyl--

sulfoxide prepared under N2 was purchased from Mallinkrodt. Pen-

icillin G was purchased from Sigma Chemical Co., Streptomycin

was purchased from ICN Pharmaceuticals. Schwarz-Mann supplied

3 Week Old
A 2 Year Old

(.c 0025)

( .OD25)



I I I I --

After END of Treatment (4/.m N-AAAF, 15min.. 370C)

005 15

Figure 16. Normalized fraction of undamaged, newly synthesized,
daughter strand DNA rapidly eluted versus time after treatment
.with N-AAAF. Results are plotted as in figure 9. Points
represent means standard error (n=4). Statistical analysis
was as described for figure 6.






LO --------

0 05 1.5 3 6 16
Hours After END of Treatment (DMSO [0.1% %]v, 15min., 37C)

Figure 17. Relative amount of rapidly eluted, newly synthesized,
daughter strand DNA found after sham-treatment with DMSO.
Results are plotted as in figure 9. Points represent the
means of triplicates.

3 Week Old
A 2 Year Old

( c-0j025)
(,c =0.025)

0 0.5 1.5 3 6 16
Hours After END of Treatment (4pm N-AAAF, 15 min., 37C)

Figure 18. The ratio of total 3H DPM/total 14C DPM in N-AAAF-treated
cultures divided by that in untreated, control cultures versus
time after treatment. Results are plotted as in figure 11.
Points represent means standard error (n=4). Statistical
analysis was as described for figure 6.






values at 16 and 3 hours, respectively, were numerically small

when compared to the other differences found after N-AAAF

treatment. Sham-treatment with DMSO (figure 19) yielded a slight

suppression of DNA synthesis seen during the first three hours

after treatment. This result is consistent with that seen after

pH 6 buffer sham-treatment (figure 12).

The total number of experiments using BPADE was not sufficient

to allow statistical analysis using the Mann Whitney U statistic

as in the other experiments. However, certain aspects of the res-

ponse obtained after BPADE treatment appear similar to the responses

after ENU and N-AAAF treatment. As seen in figure 20, parent strand

fragmentation was greatest in young donor cells and virtually non-

existent in old donor cells. Fragmentation in young donor cells

returned to the levels found in sham-treated control experiments

(figure 15) by 6 hours after the end of treatment. A trend towards

resurgent fragmentation of parent strand DNA is apparent at 1.5

hours as was the case for N-AAAF.

Benzo(a)pyrene-diol-epoxides are known to form phosphotriesters

in DNA with low frequency, which are expected to lead to

fragn.entaLion of DNA under the alkaline conditions in these experi-

ments. Figure 21 provides evidence for this. Naked DNA on filters

was treated with four levels of BPADE in this experiment, the highest

of which, 2.7 micromolar, led to increased fragmentation of DNA.

Although a small amount of nicking of DNA at BPADE-induced phos-
photriesters is expected at physiological pH, the overall con-

sequences of BPADE-induced phosphotriesters for the results reported


3.0 -

2.0 -



it ,~ --

1.5 3

Hours After END of Treoatment (DMSO [O.I%1/ 15 min., 37C)

Figure 19. Relative rate of DNA synthesis during a 0.5 hour time
interval at various times after sham-treatment with DMSO.
Results are plotted as in figure 11. Points represent the
.means of triplicates.

3 Week Old
A 2 Yeor Old

I I 1
005 1.5 3 6 16
Hours After END of Treatment (1.35 fm 8PADE, 15 rmin., 37C)

Figure 20. Normalized fraction of parental DNA eluted after 10.5
hours of elution versus time after treatment with BPADE. Results
are plotted as in figure 6. Points represent means of
-quadruplicates (3 week old rats). or duplicates (2 year old rats).



DMSO (0.4% v/v)
A O.68,/m BPADE
a i.351/m BPADE

"o o 2.7pm BPADE

2 4 6 8 10
Sample No.(4.5 ml, 0.05 ml/,'min)

Figure 21. Log of the fraction of DNA remaining on the filters
after collection of each sample versus sample number. Naked
DNA on the filters was treated with BPADE as described in
Methods and Materials. Points represent the means of duplicates.


here would be essentially as discussed earlier for the ENU and

N-AAAF experiments.

While the curves for daughter strand elongation after BPADE

treatment (figure 22) of young and old donor cells are very similar

at later times after the end of treatment, a divergence is evident

at the 1.5 and 3 hour time points. Less elongation is evident in

the old donor cells at early times after treatment; a result

converse to those after ENU and N-AAAF treatments.

Although.there is suppression of DNA synthesis early after

BIADL treatment (figure 23) with no recovery, as was seen after

ENU and N--AAAF treatment, more suppression is apparent in the

old donor cells. This was not seen in the other experiments.

Certain aspects of the response to BPADE treatment may differ

from those obtained after ENU and N-AAAF treatment, but a greater

number of experiments would be required in order to verify this.

The differences between young and old donor cells seen here

could possibly be accounted for if there were differences in the

-rate of cell division and/or cell survival after treatment. If

the old donor cells divided more rapidly than the young donor cells,

the damaged residues might be diluted out. This would result in

a lower repair response observed at the time of harvest. When

the cultures were first initiated the old donor cells appeared to

grow slightly slower than the young donor cells. This has been

observed in other systems,46 but is the opposite situation to that

required to produce the observed difference in repair response.


3 Week O;d
A 2 Year Old


0 5 L5 3 6 16
Hours After END of Treatment (I.35).m BPADE, 15 min., 37C)

Figure 22. Normalized fraction of undamaged, newly synthesized,
daughter strand DNA rapidly eluted versus time after treatment
with BPADE. Results are plotted as in figure 9. Points
represent means of nuadruplicates (3 week old rats) or
duplicates (2 year old rats).

2.0 -


3 Week Old
A 2 Year O10



A ,-.



O05 15

Hours After END of Treotment (1.35p.m BPADE, 15 min.,37 C)

Figure 23. The ratio of total 3H DPM/total 14C DPM in BPADE-treated
cultures divided by that in untreated, control cultures versus
time after treatment. Results are plotted as in figure 11.
Points represent means of quadruplicates (3 week old rats) or
duplicates (2 year old rats).

- -~-------


At the end of the experiments all cultures were approximately

90% confluent for both young and old donors; therefore, any

differences in growth rate were not apparent during the treat-

ment and repair interval. Also, the overall rate of DNA synthesis

at the end of the experiments in the untreated, control cultures,

as measured by the ratio of total 3H DNA/total 14C DNA, was not

significantly different between young and old cultures.

Cytotoxicity in these experiments was estimated by deternin-
ing the total C DPM in cultures incubated for 16 hours after treat-

ment relative to that in untreated, control cultures. In all cases
the loss of C DPM was less than 10%, and there was no statistically

significant difference in this loss between young and old donor

cultures. However, cell killing might occur without detachment

from the growth surface, and this cannot be ruled out. Greater cell

killing in the old donor cultures might account for the results

presented here. If this were the case it would mean that the old

donor cells are less resistant to the damaging insults presented

to them. Decreased overall resistance to DNA damaging agents would

be an expected result of decreased DNA repair efficiency.

Differences between cultures may have resulted if equal concen-

trations of chemical agent in the media created different levels of

DNA damage in young and old donor cultures. This cannot.be'entirely

ruled out but is unlikely. The three chemicals used act directly,

requiring no metabolic activation by the cell.91'112'135 Differences

ih cell volume or macromolecular content might lead to differences

in reactivity with cellular contents and, therefore, to differences

in the level of adducts formed in the DNA. However, Schneider and Mitsu

found that fibroblasts derived from human donors show no

differences in cell volume or macromolecular content as a function
of donor age. If this can be assumed to be true for the rat

fibroblasts used here, there is no reason to expect unequal initial

DNA adduct levels between young and old donor cells.

Differential inactivation of repair or metabolic systems

between young and old donor cells could have been a factor in these

experiments. If equal amounts of alkylation or arylalkylation

of components other than DNA created unequal impairments in meta-

bolism and/or DNA repair, this might account for the results obtained.

If substantiated, this would be a useful finding in itself. However,

the fact that total DNA synthesis was suppressed essentially to the

same extent by ENU and N-AAAF treatment argues against this. If

preferential inactivation of old donor cell enzymes occurred, we

might expect it to be manifest in the overall level of DNA syn-

thesis as well.

One final possibility is that increased levels of protein-DNA

or DNA-DNA crosslinks in old donor cells could account for the

decreased rate of elution of old donor cell DNA from the filters.

Apparently, DNA-protein crosslinks do increase with age in

mammals.37 44 It is also known that DNA-protein crosslinks inhibit

elution of DNA from polyvinylchloride filters under alkaline .
io 136-139 "
conditions. If this was involved in these experiments, one

might expect DNA from untreated, control, old donor cultures to be

eluted at a slower rate than that from untreated, control, young

donor cultures. This was not observed.


The results reported here provide further evidence for de-

clining DNA repair capacity with in vivo aging in mammals. However,

I feel this work should be extended and complemented by studies

of the removal of specific adducts from young and old donor cells.

Virtually every agent known to damage DNA produces several species

of damage, each of which may be repaired with different efficiency

and kinetics. This has been demonstrated for ENU 15 and N-AAAF.140

In the case of ENU, 06-ethylguanine has been shown to be removed

much more slowly from rat brain DNA than the other base adducts

formed.87 This correlates with preferential production of brain

tumors in those rats. When compared to the overall level of

adducts, 06-ethylguanine is a relatively minor product, 86 yet it
has the highest correlation with tumor production. Likewise,

repair of only certain species of DNA damage produced by a single

agent may be deficient in aged animals or in animals endowed with

short life spans. Such information would be crucial to a complete

understanding of those processes leading to increased longevity.

This concept could be applied to the work of Hart and Setlow

and Sacher and Hart57 as well. Their studies of repair of U.V.-

induced damage as a function of maximum life span capacity did not ex-

amine removal of specific lesions. Ultraviolet light-induced lesions
include monomeric saturation of pyrimidine rings and DNA-protein



crosslinks146 as well as pyrinidine dimners.141 Differences in

repair of any one of these might have produced the results they


That DNA repair is a multifaceted process is emphasized by the

discovery of the base excision and postreplication repair pathways.

Either or both of these may be critically important to an under-

standing of aging.

Excision repair begins with endonucleolytic incision of DNA

adjacent to a damaged region. In nucleotide excision repair, the

incision is accomplished in one step. Base excision repair requires

two steps for incision; first, cleavage of the base from its deoxy-

ribose and second, incision of the DNA at or adjacent to the result-

ing apurinic site.148

Friedberg et al. have shown that mammalian cells contain uracil

glycosidase activity. Such an activity could initiate the excision

repair process by cleaving the glycosidic bond between a uracil in DNA

and' its corresponding deoxyribose. This process would leave an

apyrimidinic/apurinic site in the DNA. Endonucleases specifically

recognizing apurinic sites have been isolated from rabbit and calf

thymus,149 rat liver,150 and calf liver. The repair process

would continue by exonuclease removal of damaged residues, resynthesis

of lost DNA by DNA polymerase, and ligation of the new and old DNA

regions. These remaining processes are presumably indentical for

nucleotide and base excision repair.

Note, as mentioned, that N7-ethylguanine, a product of ENU-

induced DNA damage, and the BPADE-induced N7-guanine adduct are

lost from the DNA primarily by depurination. 15294 Repair of the


resulting apurinic sites is expected to play a significant role

in the repair processes monitored after treatment in this study.

Repair of apurinic sites in DNA may be of prime importance

for determining the cellular error level and, therefore, the rate

of natural aging of an organism Lindahl153 has estimated, from the

rate constants for depurination and depyrimidation in native DNA

in aqueous solution,-that the average mammalian cell at 37 C is

expected to lose 10,000 purines and 500 pyrimidines over a 20 hour

period. These numbers are small compared to the total numbers of

purines and pyrimidines in the DNA of a cell. However, they indi-

cate that the DNA repair systems are called upon every day in every

cell to perform thousands of repair operations. This occurs in

,addition to repair of damage from environmental sources. Also,

in order to prevent errors from accumulating in the genome, the

repair systems would need to have virtually 100% efficiency. Con-

,sider, for example, that 99% efficiency is routinely achieved.

This encompasses both recognition of damage and fidelity of resyn-

\thesis. At 10,000 depurinations per day, 99% efficiency would

-leave an average of 100 errors in the genome of every cell in the

entire organism. Over a year this would become 37,000 errors in

,the genome of every cell. Thus errors could easily accumulate

in DNA, starting from birth, simply due to the aqueous and thermal

.environment of the DNA and to the lack of absolute perfection of

the DNA repair process. A study of the efficiency of repair of

Apurinic sites as a function of the capacity for increased longe-

vity may therefore be very appropriate.

Postreplication repair is usually defined operate ionally as

the chasing of newly synthesized DNA into high molecular weight


DNA, as measured by alkaline sucrose gradient sedimentation, in

spite of the presence of damaged bases in parental, template DNA
during this process. This is similar to the process measured in

the experiments reported here. The major difference is in the DNA

single-strand length being measured. In alkaline sucrose gradients,

,only DNA pieces below an upper limit of 5x108 daltons are detect-

-able.154 Any changes in length above this limit are not observed.

With the alkaline elution technique, the upper size limit is extended
to 1.5x10 daltons; increases in length can be followed up to this

limit.83 It has been estimated that rejoining of DNA to this length

represents the joining of at least three replicon regions in the
DNA. Therefore, the elongation measurable in these experiments

may not be entirely analogous to that usually referred to as post-

replication repair.

,Recent evidence suggests that a higher level of postreplication

-repair activity can be induced in mammalian cells by low levels of

DNA damage.155-157 In bacteria the inducible component of post-
replication repair has been found to be error prone, and, more

recently, an inducible, error prone repair process has been report-
ed in mammalian cells. Specifically, it may be that damage not

removed by the prereplication, excision repair processes is left

to be dealt with by the postreplication repair mechanism'., This

may be processed without error by either a recombination mechanism

or by hybridizing of daughter strands on homologous duplexes as

discussed in Results. However, if the original level of damage

is such that neither the prereplication nor the error-free, post-

replication repair systems are able to deal with it effectively,

the postreplication repair system may be forced to carry out an
error producing process in order to replicate the DNA. This latter

process would correspond to the insertion of random bases opposite

damaged regions of parental DNA as discussed in Results. The

studies reported here may suggest increased capacity for postrepli-

cation repair in aged rats since there was greater daughter strand

elongation in old donor cells in the presence of damaged, parental

DNA. It is possible that cells from aged rats rely more heavily on

postreplication repair to deal with damaged DNA and, therefore, are

more likely to call on the error producing process than are cells from

young rats. Note that these results seem to contradict the findings of

Lehmann et al. of no alteration of postreplication repair with age in
humans. The techniques used here are expected to be much more

sensitive than those used by Lehmann et al., and, as mentioned, the

phenomena observed may not be precisely analogous.

That DNA repair is a multifaceted process is emphasized further

by the finding that extracts of excision-deficient Xeroderma Pigmento-

sum fibroblasts are able to excise pyrimidine dimers from purified

DNA but not from DNA in chromatin. Apparently some component

required for accessibility of repair enzymes to the DNA is missing

or defective in these cells. It is possible that factors such as

this or those responsible for subtle functions yet to be discovered

could be involved in the aging process.

Another problem that deserves more study is that of mito-

chondrial DNA repair. Very little is known about this subject.

In Tetrahymena there are two DNA polymerase activities.159-160 One

is induced by U.V. irradiation, thymine starvation and treatment


with ethidium bromide. The other appears to be maintained at con-

stant levels, and, under normal conditions, constitutes the major

DNA polymerase activity. The minor, inducible activity is associated

with the mitochondrial fraction. Since it is induced by agents

that cause DNA damage, it is tempting to speculate that this enzyme

is involved in repair of mitochondrial DNA. Also, the induction

is inhibited by cycloheximide and actinomycin D but not by chlor-

amphenicol. Therefore, the inducible, mitochondrial DNA polymerase

is synthesized by the nuclear-cytoplasmic system.

Repair of mitochondrial DNA is, apparently, difficult to

demonstrate. Clayton et al. searched for removal of pyrimidine dimer

specific T4 endonuclease sites from human cells and found none. In

their experiments, mitochondrial DNA was isolated, incubated with

T4 endonuclease and then analyzed for nicking by the endonuclease

using ethidium bromide-CsCl equilibrium centrifugation. Mito-

chondrial DNA was still nicked by the endonuclease 48 hours after

U.V. irradiation of the cells. These results can be criticized,

however, since the assay used could not distinguish between one

and 50, for example, dimers in the closed, circular DNA. The

actual number of diners may have decreased from 50 to one during

post-irradiation incubation, but the assay would have only shown

that the DNA was still being nicked. On the other hand,.re air of

,gamma-ray-induced single-strand breaks has been clearly demonstrat-

ed in Tetrahymena pyriformis. Some evidence for repair of U.V.-

induced damage in yeast mitochondria has also been obtained. After

observing an increase in survival of normal colonies upon dark

holding after high U.V. doses and a decrease in petite colonies upon

dark holding after low U.V. doses, Hixon et al. concluded that

mitochondrial DNA was repaired in this system.

/ Mitochondrial DNA repair could be a key factor in mammalian

aging. A buildup of errors in this system would lead to cellular

senescence irregardless of the efficiency of repair in the nucleus.

To assure the relevance of a study of DNA repair to the natural

.aging process, it is important to understand how the damage created

experimentally is related to that which occurs in vivo. Natural pro-

duction of apurinic sites was discussed above. Unrepaired apurinic

sites can lead to single-strand chain breakage with a t,/2 of about 100

hours under physiological conditions. They can also lead, albeit

with low frequency, to the formation of cross links in DNA. This

results since the ribose at an apurinic/apyrimidinic site is in

equilibrium between the aldehyde and furanose form. It is the reactive

aldehyde which can lead to cross link formation.164

Other pathways for damage production under physiological condi-
165 di166
tions include deamination of cytosine and adenine to yield

uracil and hypoxanthine, respectively, and opening of the imidazole

ring of adenine. The isolation of a mammalian endonuclease

specific for uracil in DNA points out the significance of cytosine

deamination for cell viability.148

Another principal source of damage to DNA, and other cellular com-

ponents as well, is from the action of free radicals and peroxides pro-
duced by normal metabolic processes and the metabolism of certain

chemicals. These species, especially hydroxyl free radicals

which are produced upon reaction of superoxide anion with

peroxides, would be expected to produce the same types


of damage produced by ionizing radiation which yields hydroxyl
radicals upon interaction with water.171 These include DNA-protein
138 172
crosslinks, and DNA base damage and single-strand breaks.

Double-strand breaks could be effectively accomplished by produc-

tion of single-strand breaks in close proximity on opposite strands.

Besides metabolic processes, another important source of free

radical damage is background radiation from radionuclides contained

within the cell. Neutrons produced by interaction of cosmic radia-

tion with our upper atmosphere react with N to produce H and C
on a steady, continuous basis. These radionuclides mix with

the biosphere to maintain a steady-state level of each in all systems

that are in equilibrium with the environment; including man.

The level of 14C, for example, is maintained at about 15 DPM per

gram of carbon in all equilibrating systems. When an animal

dies, it no longer equilibrates with the environment. The 14C

that was in the tissues at death decays with time, and, therefore,

the decrease in DPM per gram of carbon is one way of determining

the time'of death of an organism.

INot only do H and 14C produce ionizing radiation in the form

of beta particles, but they are expected to produce damage due to

the nature and location of their decay products. For example,

single-strand breaks arise from decay of (2- 3H) adenine contained

in DNA due to emission of beta particles while DNA-DNA crosslinks

are caused by local effects of the decay other than beta emission.175

Decay of (5-3H) cytosine in DNA leads to formation of uracil.176

This is thought to occur by decay of H to yield (5- 3He) cytosine

which would carry a positive charge. Subsequent elimination of

3He and attack of the positively charged pyrimidine ring by

hydroxyl anion would yield (5-OH) cytosine which would spontaneous-
ly deaminate to yield uracil. Local effects due to transforma-

tion of 14C to 14N might be expected to lead to ring fragmentation

and/or abnormal base pairing relationships. Consider, for example,

(6- 14C) guanine. When this decays the resulting (6- 14N) guanine

would probably be found only as the enol tautomer, thereby disrupt-

ing normal base pairing.

Another major source of natural, intracellular radiation is

K. It is from 40K that we receive our highest dose of background

radiation. The combined dose rate of beta and gamma radiation
40 14
from 40K is about 30 times higher than that from C.

Free radical damage to other macromolecules, such as membrane

lipids, may also lead to DNA damage. Lipid autoxidation has been
shown to damage DNA in artificial systems.178 In the cell this

could occur by the juxtaposition of DNA and lipid as is found at

the nuclear envelope. Free radical damage to lipids initiates a -

chain reaction whereby the free radical is transferred from one

unsaturated lipid side chain to another in continuous layers of

lipid as found in cell membranes. It has been estimated that
chain reactions such as this have a range up to 0.001 cm. If

a strand of DNA is adjacent to the lipid bilayer of the nuclear

envelope when a free radical is "passing by", the free radical

cold attact the DNA rather than another lipid side chain.

Spontaneous depurination, metabolic production of free radicals,

and internal sources of ionizing radiation could thus constitute a

significant, natural source of damage to DNA and other cellular


components. It may have been these agents and the damage they

produce that had to be dealt with in order for our ancestral line

to effect an increase in life span.

As mentioned, ionizing radiation produces, via hydroxyl radicals,

several types of DNA damage. Monomeric saturation of pyrimidine rings

would lead to minimal helix distortion while DNA-DNA and DNA-protein

crosslinks could create maximal helix distortion.172 Free radical-

induced damage may thus be compared to ENU and N-AAAF-induced damage,

respectively. Similarly, ultraviolet light, as used in the Hart

and Setlow and Sacher and Hart studies, produces monomeric

saturation of pyrimidine rings and DNA-protein cross-

links46,139 in addition to pyrimidine dimers, as mentioned earlier.

The former would produce a low degree and the latter two a high

degree of helix distortion. Therefore, it can be appreciated that,

in fact, experimentally induced damages are seemingly appropriate

models of naturally occurring DNA damage.

-Evidence of decreased DNA repair in aged animals leads naturally

to the following question. Does DNA repair capacity decline at a

constant rate throughout life from birth or is there a point in

life where the decline is initiated or hastened? Discovery of

some event in development which hastens the decline of processes

for repair and/or protection of cellular components could lead to

our intervention with the mechanism involved so as to decrease the

rate of aging thereafter.
As Cutler has pointed out, there are probably two levels of

processes determining the rate of aging. At the cellular level

are those processes serving to protect the cell from natural,


internal sources of damage and those for repairing dij'aage after

it occurs. At the systemic level are those factors which modulate

the protective and/or repair processes within the cells.

Calow180 has argued that accumulation of error containing

macromolecules in aging organisms is likely to be due to declining

activity of repair systems which would be expected to accumulate

errors as well. However, he argues that the decline in repair

activity is the by-product result of some process necessary during

the development of the organism. Bullough has discussed evidence

suggesting that as cells mature and differentiate, they are committed

to a pathway leading to senescence and death. This can be seen,

for instance, in the differentiation and death of epidermal cells

after division of the basal, stem cells. While this system is

subject to rather rapid turnover, analogy can be drawn to other

systems with long or infinite turnover times such as nervous tissue

and skeletal muscle. It may be that the events leading to matura-

tion, with its attendent slowing down of cell growth and turnover,.

somehow produce, as a side effect, an inhibition of repair and

protection processes so as to insure the accumulation of error-

containing components and, therefore, the senescence of the organism.

Inhibition of DNA repair upon differentiation of nerve and

muscle cells was discussed earlier. Additional evidence points

to likely sources of further inhibition of repair and/or protective

processes during development. Denckla, for instance, has

demonstrated the existence of a pituitary factor that lowers the

minimal oxygen consumption (MOC) of rats. This factor is apparently
produced beginning at puberty and blocks the stimulation of the


NOC by thyroxin. The MOC in rats increases until puperty and

declines thereafter. In humans, even though the serum protein

bound iodine (PBI) remains constant with increasing age, the basal
metabolic rate (BMR) decreases with age in adults. This

situation is exactly what would be expected in humans if, as has

been suggested,184 the pituitary-thyroid axis serves only as a

servomechanism to maintain the level of circulating thyroid hormone

at a constant level. If a factor is produced by the human pituitary

similar to that in rats, the PBI would be expected to remain constant,

but the effect of thyrcxin on the tissues, reflected in the BMR,

would be expected to decrease at puberty. To offer an explanation

for the continued decrease in BMR with advancing age, we must look

at the action of thyroxin at the cellular level.

Although the cellular effects of thyroxine are not limited to

mitochondria, Tata et al. have shown that thyroxin at physiological

doses, stimulates both mitochondrial respiration and oxidative

phosphorylation in rats. Clearly then, thyroxine would have an affect

on the availability of useable energy in a cell. Let us assume for

the sake of argument that a certain level of stimulation by thyroxin

is necessary to maintain the proper energy level in a cell. We

know that energy is required by all of the DNA repair processes

either directly in the process of repair or indirectly in the

synthesis of the repair systems. If the energy supply of a cell

is diminished by something that blocks the required stimulation

by thyroxin, we can see that the error level in that cell might

increase. In fact, it is not hard to imagine that, in this situa-

tion, the error level might snowball. The normal processes that

maintain the error level at an acceptable level would be hampered

in their function and so themselves become more error laden. This

would fit in very nicely with Holliday's concept discussed in the

Introduction. The decrease in energy availability would cause the

probability of a given cell becoming committed to an irreversible

sequence of events leading to death to increase to an unacceptable

level. This, in turn, would guarantee the senescence and ultimate

death of the organism.

By removing the pituitary from rats and giving replacement

thyroid therapy, Denckla was able to restore the MOC of adult rats

to levels found in young, prepubertal rats. Further, Bilder

and Denckla were able to reverse age-associated declines in graft

rejection and reticuloendothelial function of.rats similarly


Recent evidence also indicates that a substance from the pitui-

tary inhibits peripheral responsiveness to growth hormone. When

the pituitary is removed from rats, the peripheral response to growth

hormone returns. Both the thyroid hormone blocking factor and the

growth hormone blocking factor have long biological half lives of
about six months. It is tempting to speculate that they may be

the same or similar factors.

Study of the relation between DNA repair and aging will,not

be complete, in my opinion, until the possible role of hormonal

events occurring at puberty in modulating DNA repair have been

assessed. Also, as discussed, there are many facets of DNA repair

which remain to be studied. Certain of these may be more important

in determining the rate of aging than others, and certain aspects


may be subject to a greater degree of modulation by hormonal events.

Resistance of cells to DNA damaging insults is only one part of

the overall scheme for protection and repair of cellular damage.

All of these mechanisms need to be studied in relation to maximum

life span capacity, as well as possible modulation of these functions

during development and senescence.


1. Cutler, R.G. Evolution of Human Longevity and the Genetic
Complexity Governing Aging Rate. Proc. Natl. Acad. Sci.
USA 72:4664 (1975).

2. Orgel, L.E. Ageing Clones of Mammalian Cells. Nature
243:441 (1973).

3. Schatz, G., and Mason, T.L. The Biosynthesis of Mitochon-
drial Proteins. Ann. Rev. Biochem. 43:51 (1974).

4. Sheldrake, A.R. The Ageing, Growth and Death of Cells.
Nature 250:381 (1974).

5. Hayflick, L. The Limited In Vitro Lifetime of Human Diploid
Cell Strains. Exp. Cell Res. 37:614 (1964).

6. Holliday, R. Growth and Death of Diploid and Transformed
Human Fibroblasts. Fed. Proc. 34:51 (1975).

7. Lima, L., Malaise, E., and Macieira-Coelho, A. Aging In Vitro:
Effect of Low Dose-Rate Irradiation on the Division
Potential of Chick Embryonic Fibroblasts. Exp. Cell Res.
73:345 (1972).

8. Smith, J.R., and Hayflick, L. Variation In the Life Span of
Clones Derived From Human Diploid Cell Strains. J. Cell
Biol. 62:48 (1974).

9. Edelmann, P., and Gallant, J. On the Translational Error
Theory of Aging. Proc. Natl. Acad. Sci. USA 74:3396

10. Sharma, H.K., and Rothstein, M. Age-Related Changes in the
Properties of Enolase From Turbatrix aceti. Biochemistry
17:2869 (1978). .

11. Patel, M.S. Age-Dependent Changes in the Oxidative Metabolism
in Rat Brain. J. Gerontol. 32:643 (1977).

12. Chung, W. Enzyme P.egulbtion During Development and Aging.
Biochem. Biophys. Res. Commun. 75:879 (1977).

13. Pranama, H.R., and Lane, R.S. Impaired Protein Synthesis in
Aged Turbatrix aceti. Fed. Proc. 36:292 (1977).

14. Carter, D.B. Age-Related Change in Uptake of N-Ethyl
Maleimide by F3 Histone in Rat Brain and Liver Chromatin.
Fed. Proc. 36:293 (1977).

15. Liew, C.C., and Gornall, A.G. Covalent Modification of
Nuclear Proteins During Aging. Fed. Proc. 34:186

16. Kellogg, E.W., and Fridovich, I. Superoxide Dismutase in
the Rat and Mouse as a Function of Age and Longevity.
J. Gerontol. 31:405 (1976).

17. James, T.C., and Kanungo, M.S. Alterations in Atropine Sites
of the Brain of Rats as a Function of Age. Biochem.
Biophys. Res. Commun. 72:170 (1976).

18. Reiss, U., and Gershon, D. Comparison of Cytoplasmic Super-
oxide Dismutase in Liver, Heart and Brain of Aging Rats
and Mice. Biochem. Biophys. Res. Commun. 73:255 (1976).

19. Uchiyama, M., and Kawashima, Y. Age Related Alterations In
Chain Elongation and Mono-Unsaturation of Fatty Acyl
CoA In Rat Liver Microsomes. Exp. Gerontol. 13:57 (1978).

20. Takeuchi, N., Koga, M., Yamamura Y., Tanaka, F., and Yamaguchi,
Y. Impairment of Hepatic Cholesterol Synthesis From
Squalene and the Function of Hepatic Sterol Carrier
Protein System By Aging. Exp. Gerontol. 13:1 (1978).

21. Adelman, R.C., Stein, G., Roth, G.S., and Englander, D.
Age-Dependent Regulation of Mammalian DNA Synthesis and
Cell Proliferation In Vivo. Mech. Age. Dev. 1:49 (1972).

22. Shocken, D.D., and Roth, G.S. Age-Associated Loss of Beta
Adrenergic Receptors From Human Lymphocytes In Vivo.
Adv. Exp. Med. Biol. 97:273 (1978).

23. Klug, T.L., and Adelman, R.C. Age-Dependent Accumulation of
an Immunoreactive Species of Thyrotropin (TSH) Which
Inhibits Production of Thyroid Hormones. Adv. Exp.
Med. Biol. 97:259 (1978).

24. Gershon, H., and Gershon, D. Inactive Enzyme Molecules in
Aging Mice: Liver Aldolase. Proc. Natl. Acad.' Sci.
USA 70:909 (1973).

25. Zeelon, P., Gershon, H., and Gershon, D. Inactive Enzyme
Molecules in Aging Organisms. Nematode Fructose-1,6-
diphosphate Aldolase. Biochemistry 12:1743 (1973).

26. Gershon, H., and Gershon, D. Detection of Inactive Enzyme
Molecules in Aging Organisms. Nature 227:1214 (1970).


27. Price, G.B., Modak, S.P., and Makinod.:n, T. Age-Associated
Changes in the DNA of Mouse Tissue. Science 171:917

28. Chetuang-, C.J., Boyd, V., Peterson, L., and Rushlow, K.
Single-Stranded Region-, in DNA of Old Mice. Nature
253:130 (1975).

29. Wheeler, K.T., and Lett, J.T. On the Possibility that DNA
Repair is Related to Age in Non-Dividing Cells. Proc.
Natl. Acad. Sci. USA 71:1862 (1974).

30. Massie, H.R., Baird, M.B., Nicolosi, R.J., and Samis, H.V.
Changes in the Structure of Rat Liver DNA in Relation
to Age. Arch. Biochem. Biophys. 153:736 (1972).

31. Curtis, H.J. Biological Mechanisms Underlying the Aging
Process. Science 141:686 (1963).

32. Jarvik, L.F., Yen, F.S., and Moralishvili, E. Chromosome
Examinations in Aging Institutionalized Women. J.
Gerontol. 29:249 (1974).

33. Barton, R.W.,'Waters, L.C., and Yang, W.K. In Vitro DNA
Synthesis by Low-Molecular-Weight DNA Polymerase --
Increased Infidelity Associated with Aging. Fed. Proc.
33:1419 (1974).

34. Johnson, R., and Strehler, B.L. Loss of Genes for Ribosomal
RNA in Ageing Brain Cells. Nature 240:412 (1972).

35. Chetsanga, C.J., Tuttle, M., Jacoboni, A., and Johnson, C.
Age-Associated Structural Alterations in Senescent Mouse
Brain DNA. Biochim. Biophys. Acta 474:180 (1977).

36. Barton, R.W., and Yang, W.K. Low Molecular Weight DNA Poly-
merase: Decreased Activity In Spleens of Old Balb/c
Mice. Mech. Age. Dev. 4:123 (1975).

37. Acharya, P.V.N. The Isolation and Partial Characterization
of Age-Correlated Oligo-Deoxyribo-Ribonucleotides With
Covalently Linked Aspartyl-Glutamyl Polypeptides. In:
Beers, R.F., Herriott, R.M., and Tilghman, R.C. (eds.)
Molecular and Cellular Repair Processes. p. 13. '
Baltimore: Johns Hopkins University Press (1972).

38. Galloway, S.M., and Buckton, K.E. Aneuploidy and Ageing:
Chromosome Studies on a Random Sample of the Population
Using G-Banding. Cytogenet. Cell Genet. 20:78 (1978).

39. Deknudt, G., and Leonard, A. Aging and Radiosensitivity of
Human Somatic Chromosomes. Exp.-Gerontol. 12:237 (1977).

40. Crol-icy, C., and Curtis, H.J. The icvelopment of Somatic
Mutations in Mice With Age. Proc. Natl. Acad. Sci. USA
49:626 (1963).

41. Curtis, H.J., Leith, J., and Tilley, J. Chromosome
Aberrations in Liver Cells of Dogs of Different Ages.
J. Gerontol. 21:268 (196U).

42. Curtis, H.J., and Miller, K. Chromosome Aberrations in Liver
Cells of Guinea Pigs. J. Gerontol. 26:292 (1971).

43. Ono, T., and Okada, S. Comparative Studies of DNA Size in
Various Tissues of Mice During the Aging Process. Exp.
Gerontol. 11:127 (1976).

44. Gaubatz, S.W., and Cutler, R.G. Age-Related Differences in
the Number of Ribosomal RNA Genes of Mouse Tissues.
Gerontology 24:179 (1978).

45. Bolla, R., and Brot, N. Age Dependent Changes in Enzymes
Involved in Macromolecular Synthesis in Turbatrix aceti.
Arch. Biochem. Biophys. 169:227 (1975).

46. Schneider, E.L., and Mitsu, Y. The Relationship Between In
Vitro Cellular Aging and In ,Vivo Human Age. Proc. Natl.
Acad. Sci. USA 73:3584 (1976).

47. Smith, J.R., Pereiro-Smith, O.M., and Schneider, E.L. Colony
Size Distributions as a Measure of In Vivo and In Vitro
Aging. Proc. Natl. Acad. Sci. USA 75:1353 (1978).

48. Day, R.S. Inducible Error-Prone Repair and Cellular Senescence.
In: Nichols, W.W., and Murphy, D.G. (eds.) Cellular
Senescence and Somatic Cell Genetics. DNA Repair Processes.
p. 217. Miami: Symposia Specialists (1977).

49. Bradley, M.O., and Sharkley, N.A. Mutagenicity and Toxicity
of Visible Fluorescent Light to Cultured Mammalian Cells.
Nature 266:724 (1977).

50. Parshad, R., Sanford, K.K., Jones, G.M., and Tarone, R.E.
Fluorescent Light-Induced Chromosome Damage and Its
Prevention in Mouse Cells in Culture. Proc. Natl,. Acad.
Sci. USA 75:1830 (1978).

51. Goldberg, A.L. Degradation of Abnormal Proteins in E. Coli.
Proc. Natl. Acad. Sci. USA 69:422 (1972).

52. Bozcuk, A.N. Testing the Protein Error Hypothesis of Aging
in Drosophila. Exp. Gerontol. 11:103 (1976).

53. Yushok, W.D. ATP-Dependent Turnover of Tumor Cell Protein
Modified by Amino Acid Analogs. Fed. Proc. 33:1545

54. Fridovich, 1. The Biology of Oxygen Radicals. Science
201:875 (1978).

55. Cerutti, P.A. Excision Repair of DNA Base Damage. Life
Sciences 15:1567 (1975).

56. Hart, R.W., and Setlow, R.B. Correlation Between Deoxyribo-
nucleic Acid Excision-Repair and Life-Span in a Number
of Mammalian Species. Proc. Natl. Acad. Sci. USA 71:
2169 (1974).

57. Sacher, G.A., and Hart, R.W. Longevity, Aging and Comparative
Cellular and Molecular Biology of the House Mouse, Mus
musculus and the White-Footed Mouse, Peromyscus leucopus.
Birth Defects: Orig. Art. Ser. 14:71 (1978).

58. Lett, J.T. Cellular Senescence and the Capacity for Rejoining
DNA Strand Breaks. In: Nichols, W.W., and Murphy, D.G.
(eds.) Cellular Senescence and Somatic Cell Genetics.
DNA Repair Processes. p. 89 Miami: Symposia Specialists

59. Ono, T., and Okada, S. Does the Capacity to Rejoin Radiation-
Induced DNA Breaks Decline in Senescent Mice? Int. J.
Radiat. Biol. 33:403 (1978).

60. Yielding, K.L. A Model For Aging Based On Differential Repair
of Somatic Mutational Damage. Perspect. Biol. Med.
17:201 (1974).

61. Cleaver, J.E. Nucleosome Structure Controls Rates of Excision
Repair in DNA of Human Cells. Nature 270:451 (1977).

62. Bodell, W.J. Nonuniform Distribution of DNA Repair in Chromatin
After Treatment With Methyl Methanesulfonate. Nucl.
Acids Res. 4:2619 (1977).

63. Ramanathan, R., Rajalakshmi, S., and Sarma, D.S.R. Non-
Random Nature of In Vivo Interaction of 311-N-Hydroxy-
2-Acetylaminofluorene and Its Subsequent Removal From
Rat Liver Chromatin-DNA. Chem.-Biol. Interact. 14:375

64. Wilkins, R.J., and Hart, R.W. Preferential DNA Repair in
Human Cells. Nature 247:35 (1974).

65. Feldman, G., Remsen, J., Shinohara, K., and Cerutti, P.
Excisability and Persistence of Benzo(a)pyrene DNA
Adducts in Epithelioid Human Lung Cells. Nature
274:796 (1978).

66. Berliner, J., Himes, S.W., Aoki, C.T., and Norman, A.
The Sites of Unscheduled DNA Synthesis Within Irradiated
Human Lymphocytes. Radiat. Res. 63:544 (1975).

67. Harris, C.C., Connor, R.J., Jackson, F.E., and Lieberman,
M.W. Intranuclear Distribution of DNA Repair Synthesis
Induced by Chemical Carcinogerni or Ultraviolet Light in
/ Human Diploid Fibroblasts. Cancer Res. 34:3461 (1974).

68. Stockdale, F.E. DNA Synthesis in Differentiating Skeletal
Muscle Cells. Initiation by Ultraviolet Light. Science
171:1145 (1971).

69. Hahn, G.M., King, D., and Yang, S.J. .Quantitative Changes
in Unscheduled DNA Synthesis in Rat Muscle Cells After
Differentiation. Nature New Biology 230:242 (1971).

70. Chan, A.C., NG, S.K.C., and Walker, I.G. Reduced DNA Repair
During Differentiation of a Myogenic Cell Line. J. Cell
Biol. 70:685 (1976).

71. Byfield, J.E., Lee, Y.C., Klisak, I., and Finklestein, J.Z.
Effect of Differentiation on the Repair of DNA Single
Strand Breaks in Neuroblastoma Cells. Biochema. Biophys.
Res. Commun. 63:730 (1975).

72. McCombe, P., Lavin, M., and Kidson, C. Control of DNA Repair
Linked to Neuroblastoma Differentiation. Int. J. Radiat.
Biol. 29:523 (1976).

73. Karran, P., Moscona, A., and Strauss, B. Developmental
Decline in DNA Repair in Neural Retina Cells of Chick
Embryos. J. Cell Biol. 74:274 (1977).

74. Spiegler, P., and Norman, A. Kinetics of Unscheduled DNA
Synthesis Induced by Ionizing Radiation in Human Lympho-
cytes. Radiat. Res. 39:400 (1969).

75. Scudiero, D., Norin, A., Karran, D., and Strauss, B. DNA
Excision-Repair Deficiency of Human Peripheral Blood
Lymphocytes Treated With Chemical Carcinogens. Cancer
Res. 36:1397 (1976).

76. Schneider, E.L., and Kram, D. Examination of the Effect of
Aging on Cell Replication and Sister Chromatid Exchange..
In: Nichols, W.W., and Murphy, D.G. (eds.) Cellular
Senescence and Somatic Cell Genetics. DNA Repair,Processes.
p. 177 Miami: Symposia Specialists (1977).

77. Kram, D., Schneider, E.C., Tice, R.R., and Gianas, P. Aging
and Sister Chromatid Exchange. Exp. Cell Res. 114:471

78. Galloway, S.M. What Are Sister Chromatid Exchanges? In:
Nichols, W.W., and Murphy, D.G. (eds.) Cellular Senes-
cence and Somatic Cell Genetics. DNA Repair Processes.
p. 191 Miami: Symposia Specialists (1977).

79. Lambert, B., Rengborg, U., and Swanbeck, G. Repair of U.V.
Induced DNA Lesions in Peripheral Lymphocytes From
Healthy Subjects of Various Ages, Individuals With
Down's Syndrome and Patients Wich Actinic Keratosis.
Mutat. Res. 46:133 (1977).

80. Chandley, A.C., and Kofman-Alfaro, S. Unscheduled DNA
Synthesis in Human Germ Cells Following U.V. Irradiation.
Exp. Cell Res. 69:45 (1971).

81; Lehmann, A.R., Kirk-Bell, S., and Jaspers, N.G.J. Postreplica-
tion Repair in Normal and Abnormal Human Fibroblasts.
In: Nichols, W.W., and Murphy, D.G. (eds.) Cellular
Senescence and Somatic Cell Genetics. DNA Repair Processes.
p. 203 Miami: Symposia Specialists (1977).

82. Bochkov, N.P., and Kuleshov, N.P. Age Sensitivity of Human
Chromosomes to Alkylation Agents. Mutat. Res. 14:345

83. Kohn, K.W., Erickson, L.C., Ewig, R.A.G., and Friedman,
C.A. Fractionation of DNA From Mammalian Cells By
Alkaline Elution. Biochemistry 15:4629 (1976).

84. Kohn, K.W., Friedman, C.A., .wi.g, R.A.G., and Iqbal, Z.M.
DNA Chain Growth During Replication of Asynchronous
L1210 Cells. Alkaline Flution of Large DNA Segments
From Cells Lysed On Filters. Biochemistry 13:4134 (1974).

85. Cutler, R.G. Alterations With Age in the Informational
Storage and Flow Systems of the Mammalian Cell. Birth
Defects: Orig. Art. Ser.. 14:463 (1978).

86. Singer, B. All Oxygens in Nucleic Acids React With Carcino-
genic Ethylating Agents. Nature 264:333 (1976).

87. Goth, R., and Rajewsky, M.F. Persistence of 06-Ethylguanine
in Rat Brain DNA: Correlation With Nervous System-
Specific Carcinogenesis by Ethylnitrosourea. Proc.
Natl. Acad. Sci. USA 71:639 (1974).

88. Shooter, K.V. DNA Phosphotriesters As Indicators of Cumulative
Carciinoen-Induced Damage. Nature 274:612 (19.78).

89. Shooter, K.V., and Merrifield, R.K. An Assay For Phospho-
triester Formation In the Reaction Of Alkylating Agents
With Deoxyribosenucleic Acid In Vitro and In Vivo.
Chem.-Biol. Interact. 13:223 (1976).

90. Yamasaki, H., Pulkrabek, P., Grunberger, D., and Weinstein,
1 B. Differential Excision From DNA of the C-8 and
N Guanosine Adducts of N-Acetyl-2-Aminofluorene by
Single Strand-Specific Endonucleases. Cancer Res.
37:3756 (1977).

91. Amacher, D.E., Elliott, J.A., Lieberman, M.W. Differences
in Removal of Acetylamdiofluorene and Pyrimidine Dimers
From the DNA of Cultured Marmialian Cells. Proc. Natl. Acad.
Sci. USA 74:1553 (1977).

92. Osborne, M.R., Beland, F.A., Harvey, R.G., and Brookes, P.
The Reaction of 7alpha, 8beta-Dihydroxy-9beta, 10beta-
epoxy-7,8,9,10-Tetrahydrobenzo(a)pyrene With DNA. Int.
J. Cancer 18:362 (1976).

93. Shinohara, K., and Cerutti, P.A. Excision Repair of Benzo(a)-
pyrene-deoxyguanosine Adducts in Baby Hamster Kidney
21/C13 Cells and in Secondary Mouse Embryo Fibroblast
C56Bl/6J. Proc. Natl. Acad. Sci. USA 74:979 (1977).

94. Osborne, M.R., Harvey, R.G., and Brookes, P. The Reaction
of Trans-7,8-Dihydroxy-anti-9,10-epoxy-7,8,9,10-Tetra7
hydrobenzo(a)pyrene With DNA Involves Attack at the N -
Position of Guanine Moieties. Chem.-Biol. Interact.
20:123 (1978).

95. Helfich, R.H., Dorney, D.J., Maher, V.M., and McCormick, J.J.
Reactive Derivatives of Benzo(a)pyrene and 7,12-Dimethyl-
benz(a)anthracene Cause Sl Nuclease Sensitive Sites in
DNA and "U.V. Like" Repair. Biochem. Biophys. Res.
Commun. 77:634 (1977).

96. Kakefuda, T., Yamamoto, H.A. Modification of DNA by the
Benzo(a)pyrene Metabolite Diol-Epoxide r-7,t- 8-Dihydroxy-
t-9,10-oxy-7,8,9,10-Tetrahydtobenzo(a)pyrene. Proc. Natl.
Acad. Sci. USA 75:415 (1978).

97. 'Mehta, J.R., and Ludlum, D.B. Synthesis and Properties of
Poly(06-methylguanylic acid) and Poly(06-ethylguanylic
Acid). Biochemistry 15:4329 (1976).

98. Singer, B., Fraenkel-Conrat, H., and Kusmierek, J.T. Prepara-
tion and Template Activities of Polynucleotides Containing
02- and 04-alkyluridine. Proc. Natl. Acad. Sci. USA
75:1722 (1978).

99. Hariharan, P.V., and Cerutti, P.A. Excision of Ultraviolet
and Gamma Ray Products of the 5,6, Dihydroxydihydrothymine
Type by N-'clear Preparations of Xeroderma Pigmentosum
Cells. Biochim. Biophys. Acta 447:375 (1976).

100. Cook, K., Friedberg, E.C., and Cleaver, J.E. Excision of
Thymine Dimers From Specifically Incised DNA by Extracts
of Xeroderma Pigmentosum Cells. Nature 256:235 (1975).

101. Paterson, M.C., Smith B.D., Lohman, D.H.M., Anderson, A.K.,
and Fishman, L. Defective Excision Repair of Gamma-Ray
Damaged DNA in Human (Ataxia Telangiectasia) Fibroblasts.
Nature 260:444 (1976).

102. Woff, S., Rodin, B., and Cleaver, J.E. Sister Chromatid
Exchanges Induced By Mutagenic Carcinogens in Normal and
Xeroderma Pigmentosum Cells. Nature 265:347 (1977).

103. Amacher, D.E., and Lieburman, M.W. Removal of Acetylamino-
fluorene From the Dt:A of Control and Repair-Deficient
Human Fibroblasts. Biochem. Biophys. Res. Commun.
74:285 (1977).

104. Kraemer, K.H. Progressive Degenerative Diseases Associated
With Defective DNA Repair: Xeroderma Pigmentosum and
Ataxia Telangiectasia. In: Nichols, W.W., and Murphy,
D.G. (eds.) Cellular Senescence and Somatic Cell Genetics.
DNA Repair Processes. p. 37 Miami: Symposia Specialists

105. Ahmed, F.E., and Setlow, R.B. Different Rate-Limiting Steps
in Excision Repair of Ultraviolet and N-acetoxy-2-acetyl-
aminofluorene-Damaged DNA in Normal Human Fibroblasts.
Proc. Natl. Acad. Sci. USA 74:1548 (1977).

106. Coleman, G.L., Barhold, W.S., Osbaldiston, G.W., Foster, S.J.,
and Jonas, A.M. Pathological Changes During Aging In
Barrier ReAred Fischer 344 Male Rats. J. Gerontol.
32:258 (1977).

107. Barile, M.F., Hopps, H.E., Grabowski, M.W., and Riggs, D.B.
The Identification and Sources of Mycoplasmas Isolated
From Contaminated Cell Cultures. Ann. N.Y. Acad. Sci.
225:251 (1973).

108. Levine, E.M. Mycoplasma Contamination of Animal Cell.-Cultures:
A Simple, Rapid Detection Method. Exp. Cell Res. 74:99

109. Hayflick, L. Decontaminating Tissue Cultures Infected With
Pleuropneumonia-Like Organisms. Nature 185:783 (1960).

110. Hayflick, L. The Limited In Vitro Lifetime of Human Diploid
Cell Strains. Exp. Cell Res. 37:614 (1964).

111. Maher, V.M., Birch, N., Otto, J.R., and NcCormick, J.J.
Cytoxicity of Carcinogenic Aromatic Amides in Normal
and Xeruderma Pigmentosum Fibroblasts With Different'
DNA Repair Capabilities. J. Natl. Cancer Inst. 54:1287

112. Goth, R., and Rajewsky, M.F. Ethylation of Nucleic Acids by
Ethylnitrosourea-1-14C in the Fetal and Adult Rat.
Cancer Res. 32:1501 (1972).

113. Cerutti, P.A., Sessions, F., Hariharan, P.V., and Lusby, A.
Repair Of DNA Damage Induced By Benzo(a)pyrcne Diol-
Epoxides I and II In Human Alveolar Tumor Cells.
Cancer Res. 38:2118 (1978).

114. Snedecor, G.W., and Cochran, W.G. Statistical Methods. 6th
Edition. p. 130 Ames: Iowa State University Press (1973).

115. Goth-Goldstein, R. Repair of DNA Damaged by Alkylating Carcino-
gens is Defective in Xeroderma Piginentosum-Derived
Fibroblasts. Nature 267:8i (1977).

116. Spiegler, P., and Norman, A. Kinetics of Unscheduled DNA
Synthesis Induced by Ionizing Radiation in Human Lympho-
cytes. Radiat. Res. 39:400 (1969).

117. Clarkson, J.M., and Evans, H.J. Unscheduled DNA Synthesis
in Human Leukocytes After Exposure to U.V. Light, Gamma-
Rays and Chemical Mutagens. Mutat. Res. 14:413 (1972).

118. Paterson, M.C., Lohman, P.H.M., and Sluyter, M.L. Use of a
U.V. Endonuclease From Micrococcus luteus To Monitor
the Progress of DNA Repair In U.V.-Irradiated Human
Cells. Mutat. Res. 19:245 (1973).

119. Cleaver, J.E. Repair Replication in Chinese Hamster Cells
After Damage From Ultraviolet Light. Photochem. Photobiol.
23:17 (1970).

120. Cleaver, J.E., Thomas, G.H., Trosko, J.E., and Lett, J.T.
Excision Repair (Dimer Excision, Strand Breakage and
Repair Replication) In Primary Cultures of Eukaryotic
(Bovine) Cells. Exp. Cell Res. 74:67 (1972).

121. Frei, J.V., and Lawley, P.D. Methylation of DNA in Various
Organs of C57'1i Mice By a Carcinogenic Dose of N-Methyl-
N-Nitrosourea and Stability of Some Methylation Products
Up to 18 Hours. Chem.-Biol. Interact. 10:413 (1975).

122. Meltz, M.L., and Painter, R.B. Distribution of Repair
Replication in the HeLa Cell Genome. Int. J. Radiat.
Biol. 23:637 (1973).

123. Lawley, P.D. Reaction of N-Methyl-N-Nitrosourea (MNUA)
With 32P-Labelled DNA: Evidence for Formation of
Phosphotriester. Chem.-Biol. Interact. 7:127 (1973).

124. Meyn, R.E., Vizard, D.L., Hewitt, R.R., and Humphrey, R,M.
The Fate of Pyrimidine Dimers in the DNA of Ultraviolet
Irradiated Chinese Hamster Cells. Photochem. Photobiol.
20:221 (1974).

125. Hanawalt, P.C. DNA Repair Processes: An Overview. In:
Nichols, W.S., and Murphy, D.G. (eds.) Cellular Senescence
and Somatic Cell Genetics. DNA Repair Processes. p. 1
Miami: Symposia Specialists (1977).

126. Waters, R., and Regan, J.D. Recombination of U.V.-Induced
Pyrimidine Dimers in Human Fibroblasts. Biochem. Biophys.
Res. Commun. 72:803 (1976).

127. Buhl, S.N., and Regan, J.D. Repair Endonuclease-Sensitive
Sites in Daughter DNA of Ultraviolet Irradiated Human
Cells. Nature 246:484 (1973).

128. Lehmann, A.R., and Kirk-Bell, S. Pyrimidine Dimer Sites
Associated With the Daughter DNA Strand In U.V. Irradiated
Human Fibroblasts. Photochem. Photobiol. 27:297 (1978).

129. Witkin, E.M. Ultraviolet Mutagenesis and Inducible DNA Repair
in Escherichia coli. Bacteriol. Rev. 40:869 (1976).

130. DasGupta, V.B., and Summers, S.C. Ultraviolet Reactivation of
Herpes Simplex Virus is Mutagenic and Inducible in
Mammalian Cells. Proc. Natl. Acad. Sci. USA 75:2378

131. Lehmann, A.R., and Bridges, B.A. DNA Repair. Essays in
Biochem. 13:71 (1977).

132. Painter, R.B. Rapid Test to Detect Agents That Damage DNA.
Nature 265:650 (1977).

133. GCamp-er, H.B., Tung, A.S.-C., Straub, K., Bartholomew, J.C.,
and Calvin, M. DNA Strand Scission by Benzo(a)pyrene
Diol Epoxides. Science 197:671 (1977).

134. Koreeda, M., Moore, P.D., Yagi, H., Yeh, H.J.C., and Jerina,
D.M. Alkylation of Polyguanilic Acid at the 2- Amino
Group and Phosphate by the Potent Mutagen (-)-7beta,
8alpha-Dihydroxy-9beta, lO0beta-epoxy-7,8,9,10-tetra-
hydrobenzo(a)pyrene. J. Am. Chem. Soc. 98:6720 (1976).

135. King, H.W.S., Osborne, M.R., Beland, F.A., Harvey, R.G., and
Brookes, P. -7alpha, 8beta-Dihydroxy-9beta, 10beta-
epoxy-7,8,9,10-Tetrahydrobenzo(a)pyrene is an Intermediate
in the Metabolism and Binding to DNA of Benzo(a)pyrene.
Proc. Natl. Acad. Sci. USA 73:2679 (1976).

136. Fornace, A.J., Kohn, K.W., and Kann, H.E. DNA Single-Strand
Breaks During Repair of U.V. Damage in Human Fibroblasts
and Abnormalities of Repair in Xeroderma Pigmentosum.
Proc. Natl. Acad. Sci. USA 73:39 (1976).

137. Ewig, R.A.C., and Kohn, K.W. DNA Damage and Repair in Mouse
Leukemia L1210 Cells Treated With Nitrogen Mustard, 1,3-
Bis-(2-Chloroethyl)-l-Nitrosourea, and Other Nitrosoureas.
Cancer Res. 37:2114 (1977).

138. Fornace, A.J., and Little, J.B. .DNA Crosslinking Induced
By X-Rays and Chemical Agents. Biochem. Biophys. Acta
477:343 (1977).

139. Fornane, A.J., and Kohn, K.W. D!IA-Prorein Cross-Linking
By Ultraviolet Radiation in Normal Human and Xeroderma
Pigmentosum Fibroblasts. Biochem. Biophys. Acta 435:95

140. Westra, J.G., Kriek, E., and Hittenhausen, H. Identification
of the Persistently Bound Form of the Carcinogen N-Acetyl-
2-Aminofluorene to Rat Liver DNA In Vivo. Chem.-Biol.
Interact. 15:149 (1976).

141. Swenberg, J.H., Koestner, A., Weschler, W., and Denlinger,
R.H. Quantitative Aspects of Transplacental Tumor
Induction With Ethylnitrosourea in Rats. Cancer Res.
32:2656 (1972).

142. Goth, R., and Rajewsky, M.F. Molecular and Cellular Mechan-
isms Associated With Pulse-Carcinogenesis in the Rat
Nervous System By Ethylnitrosourea: Ethylation of Nucleic
Acids and Elimination Rates of Ethylated Bases from the
DNA of Different Tissues. Z. Krebsforsch. 82:37 (1974).

143. Yamane, R., Wyluda, B.J., and Shelman, R.G. Dihydrothymine
From U.V.-Irradiated DNA. Proc. Natl. Acad. Sci. USA
58:439 (1967).

144. Hariharan, P.V., and Cerutti, P.A. Formation of Products of
the 5,6 Dihydroxydihydrothymine Type by Ultraviolet Light
In HeLa Cells. Biochemistry 16:2791 (1977).

145. Vanderhoek, J.Y., and Cerutti, P.A. The Stability of Deoxy-
cytidine Photohydrates in the Mononucleotide, Oligonucleo-
tides and DNA. Biochem. Biophys. Res. Commun. 52:1156

146. Habazin, V., and Han, A. Ultra-violet-Light-Induced DNA-to
Protein Cross-Linking in HeLa Cells. Int. J. Radiat.
Biol. 17:659 (1970).

147. Serlow, R.B., and Carrier, W.L. Pyrimidine Dimers inUltra-
violet-Irradiated DNA's. J. Mol. Biol. 17:237 (1966).

148. Friedberg, E.C., Rude', J.M., Cook, K.H., Ehrmann, U.K.,
Mortelmans, K., Cleaver, J.E., and Slor, H. Excision
Repair in Mammalian Cells and the Current Status of
Xeroderma Pigmentosum. In: Nichols, W.W., and Murphy,
D.G. (eds.) Cellular Senescence and Somatic Cell Genetics.
DNA Repair Processes. p.21 Miami: Symposia Specialists

149. Lindahl, R. Mammalian Deoxyribonucleasus Acting on DWmaged
DNA. In: Beers, R.F.. Herriott, R.M., and Tilghman, R.C.
(eds.) Molecular and Cellular Repair Froceo !.-. p. 13
Baltimore: Johns Hopkins University Press (1972).

150. Verly, W.S., and Paquette, Y. An Endonuclease for Depurinated
DNA in Rat Liver. Can. J. Biochem. 51:1003 (1973).

151. Kirtikar, DM., Kuebler, J.P., Dipple, A., and Goldthwait,
D.A. Enzymes Involved in Repair of DNA Dn'aaged By
Chemical Carcinogens and Gamma-Irradiation. In:
Schultz, J., and Ahmed, F. (eds.) Cancer Enzymology.
p. 139 New York: Academic Press (1976).

152. Lawley, P.D., and Brookes, P. Further Studies on the Alkylation
of Nucleic Acids and Their Constituent Nucleotides.
Biochem. J. 89:127 (1963).

153. Lindahl, T. DNA Repair Enzymes Acting on Spontaneous Lesions in
DNA. In: Nichols, W.W., and Murphy, D.G. (eds.) Cellular
Senescence and Somatic Cell Genetics: DNA Repair Processes.
p. 225 Miami: Symposia Specialists (1977).

154. Lehmann, A.R. Postreplication Repair of DNA in Mammalian
Cells. Life Sciences 15:2005 (1974).

155. D'Ambrosio, S.M., and Setlow, R.B. Enhancement of Post-
replication Repair in Chinese Hamster Cells. Proc.
Natl, Acad. Sci. USA 73:2396 (1976).

156. Setlow, R.B., and Grist, E. Enhancement of the Rate of
Postreplication Repair in Mammalian Cells. Biophys.
J. 16:183a (1976).

157. D'Ambrosio, S.M., Aebersold, P.M., and Setlow, R.B. Enhance-
ment of Postreplication Repair In Ultraviolet-Light-
Irradiated Chinese Hamster Cells By Irradiation In G2
or S-Phase. Biophys. J. 23:71 (1978).

158. Mortelmans, K., Friedberg, E.C., Slor, H., Thomas, F., and
Cleaver, J.E. Defective Thymine Dimer Excision by Cell-
Free Extracts of Xeroderma Pigmentosum Cells. Proc.
Natl. Acad. Sci. USA 73:2757 (1976).

159. Uestergaard, 0., Marcker, K.A., and Keiding, J. Induction
of Mitochondrial DNA Polymerase in Tetrahymena. Nature
227:708 (1970).

160. Keiding, J., and Westergaard, Q. Induction of DNA Polymerase
Activity in Irradiated Tetrahymena Cells. Exp. Cell Res.
64:317 (1971).

161. Clayton, D.A., Doda, J.N., and Friedberg, E.C. The Absence
of a Pyrimidine Dimer Repair Mechanism In Mammalian
Mitochondria. Proc. Natl. Acad. Sci. USA 71:2777 (1974).

162. Pacupathy, K., Netrawali, H.S., Pradhan, D.S., ad'u Sreerniivasan,
A. Repair of Radiation-Induced Strand Scissions in Nuclear
and Mitochondrial DNAs in Tet rl. .v,. :'a pyriformis. Radiat.
Res. 66:147 (1976).

163. Hixon, S.C., Gaudin, D., and Yielding, K.L. Evidence For
the Dark Repair of Ultraviolet Damage in Saccharomyces
cerevisiae Mitochondrial DNA. Proc. Soc. Exp. Biol.
Med. 150:503 (1975).

164. Lindahl, T., and Andersson, A. .Rate of Chain Breakage at
Apurinic Sites in Double-Stranded DNA. Biochemistry
11:3618 (1972).

165. Lindahl, T., and Nyberg, B. Heat-Induced Deamination of
Cytosine Residues in DNA. Biochemistry 13:3405 (1974).

166. Jones, A.S., Mian, A.M., and Walker, R.T. The Action of Alkali
on Some Purines and Their Derivatives. J. Chem. Soc.
C:692 (1966).

167. Garrett, E.R., and Mehta, P.J. Solvolysis of Adenine Nucleo-
sides. II. Effects of Sugars and Adenine Substituents
on Alkaline Solvolysis. J. Am. Chem. Soc. 94:8542

168. Boveris, A. Mitochondrial Production of Superoxide Radical
and Hydrogen Peroxide. Adv. Exp. Med. Biol. 78:67 (1977).

169. Cerutti, P.A., and Remsen, J.F. Formation and Repair of DNA
Damage Induced by Oxygen Radical Species In Human Cells.
In: Nichols, W.W., and Murphy, D.G. (eds.) Cellular
Senescence and Somatic Cell Genetics. DNA Repair Processes.
p. 147 Miami: Symposia Specialists (1977).

170. Fridovich, I. The Biology of Oxygen Radicals. Science
201:875 (1978).

171. Weiss, J.J. Chemical Effects of Ionizing Radiations on
Nucleic Acids and Related Compounds. Prog. Nucleic
Acid Res. Mol. Biol. 3:103 (1964).

172. Cerutti, P.A. Repairable Damage in DNA: Overview. In:
Han-walt, P.C., and Setlow, R.B. (eds.) Moleculat
Mechanisms For Repair of DNA, Part A. p. 3 New York:
Plenum Publishing Corp. (1975).

173. Wang, C.H., Willis, D.L., and Loveland, W.D. Radiotracer
Methodology in the Biological, Environmental, and Physical
Sciences. p. 12 Englewood Cliffs: Prentice-Hall,
Inc. (1975).

174. Lal, D., and Suess, H. The Radioactivity of the Atmosphere
and Hydroshpere. Ann. Rev. Nucl. Sci. 18:407 (1968).

175. Krasin, F., and Person, S. ]i!tA Crosslinks, Single-Strand
Breaks and Effects on Bacteriopihage T4 Survival From
Tritium Decay of (2-311) Adenine, (8-3H) Adenine and
S(8-3H) Guanine. J. Mol. Biol. 101:197 (1976).

176. Krasin, F., Person, S., Snipes, W., Benson, B. Local Effect
for (5-3H) Cytosine Decays: Production of a Chemical
Product With Possible Mutagenic Consequences. J. Mol.
Biol. 105:445 (1976).

177. Casarett, A.P. Radiation Biology, p. 328 Englewood Cliffs:
Prentice-Hall, Inc. (1968).

178. Pietronigro, D.D., Jones, W.B.G., Kalty, K., and Demopoulos,
H,B. Interaction of DNA and Liposomes as a Model for
Membrane-Mediated DNA Damage. Nature 267:78 (1977).

179. Casarett, A.P. Radiation Biology. p. 87 Englewood Cliffs:
Prentice-Hall, Inc. (1968).

180. Calow, P. Bidder's Hypothesis Revisited: Solution to Some
Key Problems Associated With General Molecular Theory
of Ageing. Gerontol. 24:448 (1978).

181. Bullough, W.S. Mitotic Control in Adult Mammalian Tissue.
Biol. Rev. 50:99 (1975).

182. Denckla, W.D. Role
the Decline of
Clin. Invest.

of the Pituitary and Thyroid Glands in
Minimal 0 Consumption With Age. J.
53:572 (1974).

183. Pittman, J.A. The Thyroid and Aging. J. Am. Geriatrics
Soc. 10:10 (1962).

184. Rall, J.E. The Role of the Thyroid in Endocrine Control
Mechanisms. Perspect. Biol. Med. 17:218 (1974).

185. Tata, J.R., Ernster,
S., and Hedman,
the Cell Level.

L., Lindberg, 0., Arrhenius, E., Pederson,
R. The Action of Thyroid Hormones at
Biochem. J. 86:408 (1963).

186. Bilder, G.E., and Denckla, W.D. Restoration of Ability to
Reject Xenografts and Clear Carbon After Hypophvsectomy
of Adult Rats. Mech. Age. Dev. 6:153 (1977).'*

187. Denckla, W.D. Interactions Between Age and the Neuroendo-
crine and Immune Systems. Fed. Proc. 37:1263 (1978).


The author was born December 28, 1947, in Ocala, Florida. He

grew up and went to high school in Wildwood, Florida. After high

school he attended Lake-Sumter Junior College in Leesburg, Florida,

where he received the Math Award for his studies in calculus and the

A.A. degree in 1967. He then attended the University of Florida as

a preprofessional student. He was elected to Phi Beta Kappa and

received the B.S. in 1969 with a major in zoology and a minor in

chemistry. He attended the University of Florida Medical School

off and on from 1969 to 1972 completing two years of the medical

curriculum. After working as a laboratory technologist for two years,

he entered the graduate program in biochemistry at the University

of Florida. During his first year of graduate study he was supported

in part by a Selby Foundation Scholarship and during the subsequent

three years by a NCI Training Grant.