© 1996 American Heart Association, Inc.
Articles |
DNA Synthesis in Adult Feline Ventricular Myocytes
Comparison of Hypoxic and Normoxic States
From the Departments of Medicine and Pharmacology (W.L.S.), University of Miami (Fla) School of Medicine.
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and Methods Results Discussion References |
|---|
Abstract Adult mammalian ventricular myocytes are terminally differentiated cells, and the prevailing perception has been that DNA synthesis and repair are not active. We tested the hypothesis that there is potential for DNA synthesis and repair by studying the ability of whole-cell extracts from adult myocytes to incorporate [
-32P]dCTP into damaged plasmids. Left
ventricular myocytes were isolated from adult cat hearts by
collagenase dissociation. Cells were maintained in room air
(control extract, CE) or made ischemic (IE) with N2
displacement of O2 and extracted for total protein. The
nicked form of the plasmid was produced by exposure to an
Fe3+/ascorbic acid free radical generating system.
Both IE and CE degraded the supercoiled form of the plasmid and
incorporated [-32P]dCTP into the nicked
(32P/DNA mass; CE=2.2, IE=3.0) and linear forms
(32P/DNA mass; CE=28.7, IE=25.2). Exposure of
plasmids to
UV light did not inhibit incorporation of label. Inhibition studies
with the cell extracts suggested a participation of polymerase 
in
myocyte DNA repair/synthesis. Myocyte extract was as active as extract
from rapidly growing COS cells at incorporating labeled
nucleotides into plasmid DNA. The ability of intact
myocytes to incorporate [-32P]dCTP into
endogenous DNA was measured in isolated cells made
permeable with saponin. Studies were done in room air or
N2. Permeable cells incorporated
[-32P]dCTP into nuclear DNA, but maximal specific
activity of DNA was observed at 15 minutes with ischemia and at
60 minutes with room air control cells (ischemia, 1.34±0.5,
0.86±0.33, 0.60±0.04; air, 1.0, 1.28±0.20,
1.87±0.38, at 15, 30,
and 60 minutes, respectively). These data indicate that mammalian adult
ventricular myocytes can actively repair and/or synthesize
both exogenous and endogenous DNA. A DNA synthetic response
to cellular damage may have important pathological and clinical
implications.
Key Words: DNA • synthesis • repair • myocyte • ischemia
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and Methods Results Discussion References |
|---|
Systems for DNA replication and repair are present in mitotic cells, but little is known about terminally differentiated nonmitotic cells. The adult mammalian ventricular myocyte is considered a terminally differentiated cell, and the prevailing perception has been that DNA synthesis becomes quiescent shortly after birth. Absence of myocardial DNA replication and/or repair synthesis in adult ventricular myocytes has been observed in models of hypertrophy,1 2 3 4 5 6 after UV irradiation,7 after healing of experimental myocardial infarction,8 and with aging.7 9 10 11 12 While freshly isolated adult ventricular myocytes apparently do not synthesize DNA,10 cultured myocytes regain some ability to replicate DNA.13 14 Cultured adult and neonatal myocytes display various degrees of differentiation, and there appears to be an inverse relationship between the degree of differentiation and DNA synthesis.15 While the preponderance of data indicates that uncultured, adult mammalian myocytes do not synthesize DNA, a few studies have reported labeled nucleotide uptake and suggested that under certain conditions the adult left ventricle has the reserve to activate limited DNA synthesis and may proceed to hyperplasia.16 Histological studies have suggested some DNA synthesis in adult ventricles after coronary artery narrowing,17 18 19 and freshly isolated adult myocyte nuclei incorporate deoxynucleotides into DNA after stimulation by cell extracts from neonatal hearts.20 Therefore, a small body of evidence indicates that adult myocytes can be induced to some DNA synthesis, but the issue of DNA repair has been largely ignored.
The importance of DNA repair in myocardial cells has recently gained attention in studies of both mitochondrial and nuclear DNA. Mitochondrial DNA from adult whole-heart tissue preparations apparently has some ability to replicate, but it is not known whether this activity is derived from the ventricular myocytes.21 22 23 24 Repair of mitochondrial DNA has not been described,25 26 and mitochondrial DNA deletions and mutations associated with defects in oxidative phosphorylation have been reported in dilated and hypertrophic cardiomyopathies,27 in autopsied hearts of patients with coronary atherosclerosis,28 and in aging.29 30 31 Mutations and deletions in nuclear genes encoding the potassium and sodium channels have recently been described and are linked to the inherited cardiac disorder long QT syndrome.32 33 It has been suggested that free radicals generated during ischemia contribute to mitochondrial DNA damage, and evidence of ischemic damage to nuclear DNA has been observed.34 35 36
Most studies of adult ventricular myocyte DNA replication or repair have been done in culture or chronic conditions of overload or infarction, but these studies have for the most part ignored the acute effects of cell manipulation, ischemia, or other stresses, times when DNA repair would be most active. The method of Wood,37 using whole-cell extracts to repair plasmids, has been used extensively to study DNA repair synthesis in various cell systems. We used this method to study the ability of extracts from normal and ischemic adult ventricular myocytes to incorporate labeled dCTP into damaged plasmids. We also studied the effects of ischemia on endogenous DNA synthesis in permeabilized adult ventricular myocytes.
Our results indicate that freshly isolated adult ventricular myocytes and their cell extracts can actively synthesize and repair DNA and incorporate labeled precursors into both endogenous and plasmid DNA. Comparative studies in a rapidly dividing transformed cell line (COS) indicate that DNA metabolism in the adult left ventricular cell is as active as in the rapidly dividing line.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and Methods Results Discussion References |
|---|
Isolation of Myocardial Cells
Adult domestic cats were surgically anesthetized with sodium pentobarbital (30 mg/kg IV) and the hearts removed. The hearts were flushed with buffer containing (mmol/L) 130 NaCl, 4.8 KCl, 1.2 MgSO4, 1.2 NaH2PO4, 4 NaHCO3, 0.5 CaCl2, 12 glucose, and 10 HEPES, pH 7.4, followed by retrograde coronary perfusion for 8 minutes with the same buffer. This procedure was followed by perfusion for 5 minutes in Ca2+-free buffer and then a 20-minute recirculating perfusion with 125 mL of Ca2+-free buffer containing 250 U/mL collagenase (type II), 25 mg trypsin inhibitor, 20 mg hyaluronidase, 19 mg protease (4.6 U/mg), and 12 mg BSA. After perfusion, the left ventricle was removed and minced in Ca2+-free buffer with 0.5% BSA but without enzymes. Cells were allowed to settle and were resuspended in the buffer containing 50 µmol/L CaCl2 and 1% BSA. After settling, the cells were again suspended in buffer with 0.5 mmol/L CaCl2 and 1% BSA and finally in buffer with 1.8 mmol/L CaCl2 and 1% BSA.
Aliquots of freshly isolated cells were fixed in 3% glutaraldehyde in PBS, spotted on gel-coated slides, and air dried. The slides were dipped in H2O to remove salts and stained with hematoxylin and eosin. Dilute preparations were used for cell counts and for determination of nonmuscle cell contamination.
Model of Ischemia
Before extraction of cellular proteins,
isolated cells were
exposed to either a simulated ischemic environment or an
aerated control environment. Cells were washed in PBS and suspended in
either N2-purged or aerated buffer containing (mmol/L) 40
Tris at pH 7.6, 8 MgCl2, 167 sucrose, and 15 KCl.
For the simulated ischemic environment, inlet and outlet
cannulas (18G) were placed through rubber stoppers into 16x100-mm
tubes. Intramedic PE 190 tubing was attached to the bottom of the
N2 inlet cannula, and N2 was bubbled into the
buffer for 30 minutes before the cells were added. The venting cannula
was left open throughout the procedure. After bubbling, the cells were
quickly added and the tubing removed. The stopper was replaced and a
brisk N2 flow maintained through the inlet cannula during a
15-minute incubation at 37°C. Using this procedure, PO2
was maintained at <5 mm Hg, as measured by a biological oxygen monitor
(Yellow Springs Instrument Co). After exposure to
N2, the cells were washed in PBS and whole-cell
extract prepared, as described below. For the aerated controls, cells
were incubated in an open vessel, with PO2 maintained at
ambient values.
Whole-Cell Extracts
Whole-cell extracts were prepared from
ischemic and
control myocytes and from trypsinized cultured COS cells by the method
of Manley et al.38 All steps were performed at 4°C.
Cells were resuspended in 4 volumes of (mmol/L) 10 Tris at pH 7.9, 1
EDTA, and 5 DTT. After 20 minutes on ice, 0.5 mmol/L
phenylmethylsulfonyl fluoride and 0.5 µg/mL each of
leupeptin, pepstatin, and chymostatin was added, and the cells were
broken using a Dounce homogenizer with a "B"
pestle. Four packed-cell volumes of buffer containing (mmol/L) 50
Tris at pH 7.9, 10 MgCl2, 2 DTT, 25% sucrose, and
50% glycerol were added, and 1 packed-cell volume of saturated
(NH4)2 SO4 was added dropwise while
stirring. The mixture was stirred for 20 minutes and the extract
centrifuged at 42 000 rpm for 3 hours in an SW 50.1 rotor. The
supernatant was removed, leaving the last 2 mL, and proteins were
precipitated by addition of 0.33 g/mL of solid
(NH4)2 SO4. After the
(NH4)2 SO4 was dissolved, 10 µL
of 1 mol/L NaOH per gram of (NH4)2
SO4 was added and the suspension stirred for 30 minutes.
The precipitate was collected by centrifugation at
15 000g for 20 minutes and suspended in a small volume of
buffer with (mmol/L) 25 HEPES at pH 7.9, 100 KCl, 12
MgCl2, 1 EDTA, 2 DTT, and 17% glycerol. The extract
was dialyzed overnight against the same buffer and clarified by
centrifugation at 10 000g for 10 minutes.
The samples were quick frozen in small aliquots and stored at
-80°C. Protein was determined by the Lowry method.
Plasmid Repair Reaction
The commercial plasmids pBR322
(Sigma), Col E1
(Sigma), and pMSG (Pharmacia) were used for these studies. Plasmids
were suspended in H2O, cleaned of salts with BioRad
Prep-A-Gene DNA Miniprep kits (Catalog No. 732-6017), and eluted with
10 mmol/L phosphate buffer, pH 7.0, at 0.5 µg/µL. Plasmids were
damaged by exposure to free radicals by incubation in 10 mmol/L
phosphate buffer, pH 7.0, containing 10 µmol/L Fe2
(SO4)3 and 100 µmol/L ascorbic
acid.39 Samples were incubated at 32°C for 1 hour. After
incubation in the free radical generating system, an equal volume of
Miniprep binding buffer was added and the plasmids cleaned of
Fe3+ and ascorbic acid with the Miniprep system. Plasmids
were eluted in a small volume (usually 40 µL) of water and used for
the repair assay.
The repair reactions were done in 0.125 mL buffer
containing (mmol/L)
45 HEPES at pH 8.0, 70 KCl, 7.4 MgCl2, 0.9 DTT, 0.4 EDTA,
40 creatine phosphate, 2 ATP, 0.020 each dTTP, dGTP, and dATP, 0.005
dCTP, 3.4% glycerol, 36 µg BSA, 8 µg creatine phosphokinase, 8
µCi [-32P]dCTP (3000 Ci/mmol), 2 or 3 µg of
plasmid, and cell extract protein, as indicated.37 Where
indicated, polymerase inhibitors were added 15 minutes
before the addition of plasmid. After incubation, an equal volume of
Miniprep binding buffer was added, and samples were cleaned and
concentrated with the Miniprep system. Samples were eluted in 40 µL
of buffer with (mmol/L) 10 Tris at pH 7.9, 10 MgCl2,
100 NaCl, 1 ß-mercaptoethanol, and 0.1 mg/mL BSA. When indicated,
samples were linearized with EcoRI. DNA was resolved by
electrophoresis through 1% agarose gels impregnated with 5µg/100 mL
of ethidium bromide. Gels were photographed with Polaroid Type 55 film,
which produces a negative suitable for densitometric scanning of DNA
mass. The gels were dried on nitrocellulose and exposed to X-OMAT AR
film for 7 to 14 days. Autoradiographs were scanned with a
densitometer, and results are expressed as radiolabel in DNA per mass
of DNA.
Hybridization fragments were made using linearized plasmid as a
template, a random primers kit (GIBCO-BRL, No. 18187-013), and
[-32P]dCTP as the label. After Southern blotting and
hybridization, films were exposed for 30 minutes.
To compare DNA synthetic activity in myocytes with that in a rapidly growing cell line, we prepared whole-cell extracts, as described above, from cultured COS cells, a fibroblast-like, SV40-transformed African Green monkey kidney cell line. Cells were harvested by trypsinization at 70% confluence and produced a 0.5-mL packed-cell pellet. After the reaction incubation with damaged plasmid Col E1, the DNA was precipitated with 10% cold TCA and 50 µg of salmon sperm DNA. The samples were filtered onto Whatman GF/C filters and washed with 30 mL of cold TCA. After a wash of 10 mL 95% ethanol, the filters were dried and counted for radioactivity.
DNA Repair/Synthesis in Permeable Whole Cells
Freshly
isolated cells were washed with (mmol/L) 30 HEPES at pH
7.0, 100 KCl, 20 NaCl, and 1 EGTA and suspended in 50 mL of the same
buffer with 0.01% saponin to permeabilize the
cells.40 After 5 to 8 minutes on ice, the cells were
pelleted at 50g and washed in buffer containing (mmol/L) 40
Tris at pH 7.6, 8 MgCl2, 167 sucrose, and 15
KCl.
The ischemia surrogate was produced as described above except that the stopper contained one additional cannula to withdraw samples. This cannula was fitted with Intramedic PE tubing that extended into the sample on the bottom and had a Luer-Lok hub stopcock on the top. Samples were withdrawn with a syringe attached to the stopcock that allowed sampling without exposing the remaining sample to air. Each tube contained between 1.5 and 3.0 mL, depending on the experimental design, and aliquots of 0.3 to 0.6 mL were withdrawn.
The medium for
the repair reaction for permeable cells contained
(mmol/L) 40 Tris at pH 7.6, 8 MgCl2, 167 sucrose, 15
KCl, 5 ATP, 40 creatine phosphate, and 5 µmol/L each dATP, dGTP, and
dTTP, 1 µmol/L dCTP, 12 to 15 µCi/mL
[-32P]dCTP
(3000 Ci/mmol), and 250 µg/mL creatine phosphokinase.41
EDTA was omitted from the reaction mixture because it is a scavenger of
free radicals. Samples, ischemic and nonischemic,
were incubated at 36°C for the indicated times.
Samples were washed twice in cold PBS and 0.5 mL of buffer with (mmol/L) 10 Tris at pH 7.4, 10 EDTA, 150 NaCl, and 0.4% SDS, and 2 mg/mL proteinase K was added to the pellet. Samples were heated at 65°C for 15 minutes and incubated with gentle shaking at 37°C overnight. The samples were extracted with 1:1 chloroform/phenol and precipitated with ethanol. The pellet was dissolved in 0.5 mL of buffer containing 10 mmol/L Tris at pH 7.4 and 0.1 mmol/L EDTA and incubated with 30 µg of DNase-free RNase A for 1 hour at 37°C. The sample was again extracted with chloroform/phenol, ethanol precipitated, and dissolved in a small volume of the Tris/EDTA buffer. Half of the sample was diluted, the OD at 260 nm was determined, and 5 to 10 µg of DNA was electrophoresed through 1% agarose gels as described above.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and Methods Results Discussion References |
|---|
A photograph of a typical preparation at low-power magnification (100x) is shown in Fig 1
. The inset shows
the preparation at a magnification of 400x. The cells show typical
myocyte morphology with rod shapes, organized striations, and
well-defined sarcolemmas. Some rounded, contracted, and densely
stained myocytes are also present.
|
Three separate preparations were read for percentage of rods and
nonmuscle cell contamination. Analyses of over 800 cells in the
final preparations did not reveal any nonmuscle cells. Nonmuscle cells
were also not detected in any of the three fresh preparations or after
each of the three settling steps (over 3000 cells analyzed).
The percentage of rod-shaped myocytes in the final preparations
ranged from 71% to 92% and averaged 81%. Thus, our isolated cell
preparations contained exclusively myocytes, 
81% of which were rods
and 19% of which were rounded.
We tested the ability of cellular extracts from control and
N2-exposed ventricular myocytes to incorporate
[-32P]dCTP into plasmids that had been exposed to
reactive oxygen. pBR322 was exposed to the
Fe3+/ascorbic acid free radical generating system
for 1 hour, and plasmid Col E1 was used as a control plasmid
not exposed to Fe3+/ascorbic acid. Plasmids were
linearized with EcoRI digestion before agarose gel
electrophoresis. Lane 1 of Fig 2A shows an ethidium
bromide–stained gel of pBR322 (4365 bp) and plasmid Col
E1 (6600 bp) before treatment with Fe3+/ascorbic
acid or repair assay. Exposure of pBR322 to
Fe3+/ascorbic acid and subsequent 3-hour incubation
with untreated plasmid Col E1 in repair buffer without cell
extract resulted in two distinct plasmid bands with the same mobility
as the unincubated plasmids (Fig 2A, Lanes 2 and 3) and no
incorporation of [-32P]dCTP (Fig 2B,
Lanes 2 and 3).
Addition of 46 µg of protein from CE resulted in greatly reduced mass
of both plasmids (Fig 2A, Lanes 4 and 5) but clear
incorporation of
[-32P]dCTP into both the
non–Fe3+/ascorbic acid–treated Col
E1 and the treated pBR322 (Fig 2B, Lanes 4 and 5). When 46
µg of
protein from IE was added to the repair incubation, the mass of both
plasmids was also reduced, but in contrast to incubation with CE,
smaller DNA fragments were observed (Fig 2A, Lanes 6 and 7).
Heavy
incorporation of [-32P]dCTP accompanied the
reduction
in plasmid mass, and radiolabel was apparent over a wide range of the
smaller fragments (Fig 2A, Lanes 6 and 7). Lane 8 contains a
DNA ladder
of a HindIII
digest from 23 130 to 2027 bp.
|
Incorporation of label into the untreated plasmid was unexpected.
Accordingly, we assessed the plasmid forms in the untreated preparation
and the effect of the Fe3+/ascorbic acid system on
the plasmid forms. An ethidium bromide–stained gel of undigested
pMSG (7626 bp) showed that the commercial plasmid preparation contained
significant amounts of both nicked (N) and supercoiled (S) forms in
addition to a small amount of the linear form (L) (Fig 3A, Lane
1). Exposure to
Fe3+/ascorbic acid for 1 hour at 32°C, but without
the repair reaction, increased the ratio of nicked/supercoiled plasmid
threefold, indicating DNA nicking by free radicals (Fig 3A,
Lane 2). No
increase in the linear form was detected after treatment with
Fe3+/ascorbic acid.
|
The time dependence of cell extract–mediated DNA synthesis in
plasmids was analyzed by incubating
Fe3+/ascorbic acid–treated pMSG with or without
40 µg of protein from IE in repair buffer for 15, 30, and 60 minutes
at 30°C. Fig 3A
(Lane 3) shows one band of the linearized
plasmid
before Fe3+/ascorbic acid and (Lane 4) the same
plasmid after free radical treatment. There was a small decrease in the
mass of the Fe3+/ascorbic acid–treated plasmid
relative to the untreated plasmid, probably due to some loss during
extraction. Incubation for 15, 30, and 60 minutes in repair buffer
without cell extract (Lanes 5, 6, and 7, respectively) did not change
the pattern. When IE was added to the 15-, 30-, and 60-minute
incubations, there was a sharp decrease in the mass of the plasmid (Fig
3A, Lanes 8, 9, and 10, respectively) similar to that observed
with
plasmids Col E1 and pBR322 in Fig 2. Again, fragments
of DNA
were apparent by 60 minutes of incubation with IE. Radiolabel increased
with time of incubation in repair buffer with IE (Fig 3B, Lanes
8
through 10), and at 60 minutes a smear of label was observed
consistent with the appearance of fragments in the ethidium
bromide–stained gel. No radiolabel was observed in the absence of
cell extract (Fig 3B, Lanes 3 through 7).
Because of the technical difficulty of making accurate measurements
from gels and autoradiograms containing smears of DNA
fragments, we used the extract from control cells, which produced fewer
fragments than extract from ischemic cells, to measure the
relationship between extract concentration and incorporation of
[-3P]dCTP into damaged plasmids. Incorporation of
radiolabel into Fe3+/ascorbic acid–treated
pBR322 was dependent on the concentration of cellular extract. Addition
of as little as 5 µg of CE protein caused a decrease in plasmid mass
(Fig 4A, Lane 2) relative to 0 time (Lane 1) and
incorporation of radiolabel (Fig 4B, Lane 2). Increasing
amounts of CE,
up to 40 µg of protein, resulted in progressively decreasing pBR322
mass but increasing radiolabel (Fig 4A and 4B,
Lanes 2 through 6).
Addition of CE from 5 to 40 µg protein did not produce the fragments
observed previously with IE despite the loss of plasmid mass.
Densitometric scans of the gel negative and the autoradiograph showed
that the relationship between CE concentration and
[-32P]dCTP incorporation was linear and increased
with
increasing cell extract protein (Fig 4C).
|
When plasmids were linearized before gel electrophoresis, it was not
possible to determine the forms that were digested or labeled.
Therefore, we repeated the repair reaction and omitted the
EcoRI step. After incubation in repair buffer without cell
extract for 1 hour, 75% of the plasmid mass was present in the
supercoiled form, 25% was in the nicked form, and a small amount,
<1%, was in the linear form (Fig 5A, Lane 1).
Incubation with 10 µg of CE protein resulted in complete loss of the
supercoiled plasmid and a 24% decrease in the nicked form (Lane 2).
Thus, addition of CE produced an overall loss of 80% of the total
plasmid mass. Incubation with 10 µg of IE protein also resulted in
loss of the supercoiled form, but in contrast to the CE sample, the
nicked form doubled (Lane 3). Despite the increase in the nicked form,
there was still an overall loss of 45% of the plasmid mass. In the
presence of either cell extract, [-32P]dCTP was
incorporated into both the nicked and linear forms (Fig 5B).
The DNA
specific activity was greatest in the linear form (17.5 and 13.8 for CE
and IE, respectively) and was approximately 10 times greater than in
the nicked form (1.3 for both CE and IE). Therefore, despite greater
plasmid loss in the presence of CE relative to IE, the similar specific
activities suggest that degradation of the plasmid is independent of
the DNA repair/synthesis mechanism.
|
Loss of plasmid mass in the presence of cell extract was apparent as
early as 15 minutes, but there was a surprising lack of detectable DNA
fragments, particularly when CE was added. Some possibilities to
explain this apparent discrepancy could be (1) very small fragments
were running off the gel, (2) fragments were lost during the
Prep-A-Gene Miniprep purification to remove unbound label, and (3) the
fragment sizes were so heterogenous that they escaped
detection. To resolve this issue, we incubated pMSG with 30 µg of
protein from either CE or IE and processed the samples for Southern
blot analysis with and without the Miniprep cleanup. Blots were
hybridized with labeled probes against pMSG generated from random
primers. In the absence of cell extract, there was little difference in
mass or label between samples without or with cleanup (Fig 6A,
Lanes 1 and 2, respectively). All three forms of the
nonlinearized plasmid hybridized with the probes (Fig 6B, Lanes
1 and
2). As observed previously, incubation for 1 hour with 30 µg of
protein from CE resulted in reduction of plasmid mass (Lanes 3 and 4).
In the absence of Miniprep cleanup, both mass and label were detected
at the origin of the gel (Lane 3). In the sample treated with the
Miniprep system (Lane 4), neither mass nor label could be detected at
the origin, and a small amount of the labeled nicked and linear forms
migrated through the gel. Incubation of plasmid with 30 µg of protein
from IE produced a pattern similar to the CE samples except that
without cleanup (Lane 5), less material remained at the origin and more
fragments, which appeared as a smear, entered the gel. After cleanup,
no material remained at the origin, and most of the fragments were
removed (Lane 6). Therefore, use of the Miniprep system removed DNA at
the origin, reduced the fragments that entered the gel, and increased
the amount of the nicked and linear forms of the plasmid that entered
the gel.
|
We considered the possibility that DNA removed by the Miniprep system
or retained at the origin without cleanup was bound to protein.
Therefore, we used phenol/chloroform extraction to remove protein
before electrophoresis. Fig 7 shows that DNA and label
remaining at the origin were removed by protein extraction (Lane 2, CE;
Lane 3, IE). Furthermore, a smear of smaller fragments was apparent in
both samples, but some discrete fragments were also present in the
IE sample (Lane 3).
|
UV light has been reported to cause DNA breaks, thymine dimers, and
other base
modifications25 37 42 43 44 45
that could potentially
inhibit DNA synthesis if the bases are not excised. We tested the
ability of our cell extract to incorporate
[-32P]dCTP
into pBR322 that had been exposed to UV light. Fig 8
shows an ethidium bromide–stained gel of
nonlinearized pBR322 that had been untreated (Lane 1) and treated (Lane
2) with UV light. As noted previously, a portion of the commercial
plasmid contained the nicked form. Exposure to UV increased the ratio
of the nicked form to the supercoiled form only 15%, compared with the
threefold increase we observed in the presence of
Fe3+/ascorbic acid. Lanes 3 and 4 (Fig 8) each
contain both the EcoRI digested plasmid pBR322 after
treatment with 450 J/m2 UV light (pBR322) and the untreated
pMSG. Lane 3 contains the plasmid mixture after 1.5 hours of incubation
without CE and Lane 4 with 40 µg of protein from CE. Addition of CE
caused a decrease in the DNA mass and resulted in incorporation of
label into both the treated and untreated plasmids, suggesting that DNA
synthesis was not inhibited by prior treatment of pBR322 with UV light.
No decrease in mass or incorporation of label was detected without
addition of CE (Lane 3).
|
To determine whether the cell extracts were sensitive to some common
polymerase inhibitors, we measured DNA synthetic activity
in the presence of NaCl, NEM, BuphdGTP, and ddTTP. The mass and label
are shown in Fig 9A and 9B, and the calculated
specific
activities in the linear and nicked forms in Fig 10A and
10B, respectively. Incubation of pMSG with 10 µg of
protein from
CE (Lane 1) or IE (Lane 2) produced the nicked and linear forms (Fig
9A). Radiolabel was incorporated into both nicked and linear
forms (Fig 9B), with CE and IE giving similar specific
activities (Fig 10A and 10B). Addition of 150
mmol/L NaCl, which inhibits , , and 
polymerases, resulted in inhibition of incorporation of
[-32P]dCTP into both the nicked and linear forms
(Fig 9B, Lane 3, CE; Lane 4, IE; and Fig 10A
and 10B). In contrast to
extracts without inhibitors, a small amount of the
supercoiled form was detected in the ethidium bromide–stained gel.
In the presence of 20 mmol/L NEM, which inhibits polymerases , ,
, and 
, incorporation of radiolabel was also suppressed (Fig
9B,
Lane 5, CE; Lane 6, IE; and Fig 10A and 10B).
Similar to NaCl, a small
amount of the supercoiled form was detected. BuphdGTP (200 µmol/L),
which inhibits polymerases , ß, , and , reduced
incorporation of radiolabel into the linear form by 67% in the
presence of CE (Fig 9B, Lane 7, and Fig 10A)
and 79% in the presence
of IE (Fig 9B, Lane 8, and Fig 10A).
Incorporation into the nicked form
was reduced by 91% in both the CE and IE samples (Fig 10B).
The
smallest inhibition was observed in the presence of 120 µmol/L ddTTP,
which has been shown to strongly inhibit polymerases ß and and
partially inhibit polymerases and . Addition of ddTTP produced a
46% inhibition into the linear form (CE and IE) and a 52% and 63%
inhibition into the nicked form in the presence of IE and CE,
respectively (Fig 9B, Lanes 9 and 10; Fig 10A
and 10B).
|
|
To compare the DNA synthetic activity in adult myocytes with that of an
actively mitotic cell line, we compared the ability of myocyte extract
with COS cell extract to repair damaged plasmids. Free
radical–treated plasmid Col E1 was incubated with 15
µg of protein from control myocyte extract, COS cell extract, or a
mixture of the two for 1 hour. Aliquots were taken, and DNA was
precipitated with TCA. The TCA-precipitable counts for two experiments
are shown in the Table. The TCA-precipitable counts were
almost identical in the presence of either myocyte or COS cell
extracts, indicating similar activity. After subtraction of background
counts in the absence of cell extract, the activity of the combined
sample was 80% of that expected from the addition of CE and COS alone.
These results indicate that under similar conditions, cell extract from
adult myocytes has DNA synthetic activity comparable to that of a
rapidly growing cell line.
|
Our observations with cell extracts and plasmid DNA confirmed that in
this system, adult myocardial cells are capable of active DNA repair.
To investigate whether adult ventricular myocytes can
incorporate label into endogenous DNA, we used a permeable
cell system. Fig 11 shows a time course (in triplicate)
for DNA mass (A) and incorporation of [-32P]dCTP (B)
into whole-cell DNA. After 15 minutes of incubation (Fig 11,
Lanes
2, 3, and 4), the pattern of mass in the ethidium bromide–stained
gel was similar to the 0-time sample in Lane 1 and revealed DNA at the
origin and a smear that entered the gel. After 60 minutes (A, Lanes 5,
6, and 7), smaller fragments of DNA were observed as a larger smear,
and this increased at 120 (Lanes 8, 9, and 10) and 150 (Lanes 11, 12,
and 13) minutes. Autoradiography of the gel (Fig 11B) revealed
intense labeling at all time points, but at 120 and 150
minutes, the label was smeared the length of the gel.
|
To assess the effects of ischemia on the specific activity of
whole-cell DNA, we incubated permeable cells in an N2
atmosphere and withdrew samples at 15, 30, and 60 minutes. Longer
incubations were not done because of the difficulty in obtaining
accurate scans of large smears of DNA that were observed at times
greater than 60 minutes. Fig 12A and 12B show
a
representative ethidium bromide–stained gel and
autoradiograph. One predominant band of mass (A) and label (B) was
observed at all time points. In the N2-treated samples, the
label decreased with increasing time (Lanes 3, 5, and 7; 15, 30, and 60
minutes, respectively). However, label increased with time of
incubation in the control cells (Lanes 2, 4, and 6; 15, 30, and 60
minutes, respectively). A plot of the specific activities of the
primary band in 12A and 12B is shown in Fig 12C, and the
inverse
relationship between aerobic and ischemic samples is
apparent.
|
), Cells incubated in air; (
), cells
incubated in N2.
A composite graph of DNA specific activity after incubation of cells in
air or N2 is shown in Fig 13 (mean±SEM;
air, n=6; N2, n=4). Whole,
permeabilized myocardial cells that were incubated in
air increased endogenous DNA specific activity for up to 60
minutes relative to the respective 15-minute sample. Cells exposed to
N2 had maximal specific activity at 15 minutes, and label
decreased at 30 and 60 minutes relative to the 15-minute sample
incubated in air.
|
To determine whether the fragmented DNA pattern we observed in whole
permeable cells was a consequence of the experimental procedure (ie,
cell isolation, permeabilization, and/or nuclease activity) or was
indicative of DNA damage during the DNA extraction procedure, we
extracted DNA from fresh left ventricular tissue and
compared the gel patterns. Fresh tissue was pulverized in liquid
N2 with a mortar and pestle, digested in proteinase K,
extracted with chloroform/phenol, and precipitated in ethanol. After
RNase treatment, the DNA was electrophoresed through 0.4% agarose to
allow better detection of larger DNA fragments. Fig 14
shows DNA from isolated, permeabilized
cells (Lane 1) and from fresh tissue (Lanes 2 and 3). Lane 4 contains a
DNA standard with sizes ranging from 48 502 to 8271 bp. DNA from both
preparations contained fragments of variable sizes that appeared as
a smear on the gel. The detectable fragments encompassed a wide size
distribution and ranged from >48 000 to <8271 bp. In addition, some
DNA was apparent at the origin. Therefore, the DNA fragments we
observed in the isolated, whole-cell preparation were not a result
of cell manipulation or nuclease activity during isolation and
permeabilization but probably due to breakage during DNA
extraction.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and Methods Results Discussion References |
|---|
Adult ventricular myocytes are terminally differentiated cells, and the preponderance of published data supports the long-standing concept that these postmitotic cells have no capacity for nuclear DNA synthesis or repair activity. Little is known about the ability of myocardial mitochondria to synthesize DNA, but the perception is that it is also largely inactive and lacks repair enzymes. Recent investigations into some forms of cardiomyopathy have uncovered deletions and mutations in mitochondrial DNA that have been attributed to exposure of mitochondria to high levels of free radicals and to the lack of DNA repair enzymes. Damage to myocardial nuclear DNA has received less attention, but histological studies have shown that ischemia causes chromatin clumping.34 35 Free radicals have been shown to damage DNA in nonmyocyte systems through strand breaks and formation of modified bases,39 45 46 47 48 49 50 and hypoxia induces fragmentation of DNA in cultured neonatal rat heart cells.36 Since the myocardium is highly aerobic and at risk for exposure to damaging oxidative free radicals, it is conceivable that there would exist mechanisms for DNA repair in the adult myocardial cell under conditions of ischemia or hypoxia. Indeed, it would be desirable for myocytes that are present throughout a normal life span to have some mechanism for DNA repair. Indirect evidence that the adult myocardium maintains the ability to synthesize and/or repair DNA has been provided by reports that polymerases
, ß, and
and the polymerase cofactor, proliferating cell nuclear
antigen, are
present.12 18 51 52 53 54 55 56
In addition, it has
been reported that dedifferentiated cultured adult
ventricular myocytes regain polymerase
activity.14 The technique of Wood37 measures the activity of whole-cell extracts to repair damaged plasmid DNA. This method has been used with a variety of cell types, but not with adult ventricular myocytes. Preparation of cell extracts from whole tissues, as opposed to cultured cells, has the potential for mixed cell types. This is particularly important for myocardial tissue because small amounts of mitotically active nonmuscle cells could confound the results. Cutilletta et al57 reported that nonmuscle cells adhere to myocytes during cell isolation and can be observed by nuclear staining. We could not detect any nonmuscle cell contamination in stained preparations of our isolated cells, either before, during, or after the settling steps. Therefore, we are confident that the cell extracts used for these studies are exclusively from myocytes.
During the course of these studies, we observed a decrease in the
plasmid mass in the presence of extract from either control or
ischemic cells. However, the pattern of degradation was
different. Extract from control cells reduced the mass of both control
and free radical–treated plasmids without the appearance of
discrete fragments. In contrast, extract from ischemic cells
produced some discrete fragments, suggesting a nonrandom cleavage or a
less active nuclease. The degradation of supercoiled plasmid that we
observed is compatible with an endonuclease with potent nicking
activity.49 58 However, loss of the supercoiled form
in
the presence of ischemic extract was accompanied by an increase
in the nicked form, whereas both forms were degraded with the control
cell extract. These data suggest the presence of different nuclease(s)
or differences in the concentration of the nuclease(s), with greater
concentration in the control extract producing greater degradation. The
different patterns of plasmid degradation do not appear to reflect a
greater concentration in the control extract, because reducing the
concentration from 40 to 5 µg of protein did not alter the pattern.
We propose that different nucleases are active in ischemic and
control preparations. Nucleases can be of mitochondrial, nuclear, or
cytoplasmic origin,58 59 60 and results
obtained with
whole-cell extracts make it difficult to determine the source.
Regardless of the mode of DNA degradation, label appeared in nicked and
linear plasmids and plasmid fragments as early as 15 minutes of
incubation, and the specific activity of the label increased with
increasing concentration of extract and with time. It is noteworthy
that the specific activities of both the linear and nicked forms and
the response to polymerase inhibitors were similar with the
two myocyte extracts, suggesting that despite the differences in
degradation patterns, they share a common mechanism for DNA repair
synthesis. Whereas incorporation of label was linear with extract
concentration, it was not linear with time. There was enhanced
incorporation of label between 30 and 60 minutes compared with the
first 30 minutes (Fig 3). The increase in label coincided with
the
appearance of fragments, suggesting that additional sites are provided
by nuclease(s).
In these studies, we used an in vitro free radical generating system that increased the nicked/relaxed form of the plasmid threefold before incubation with the cell extract. We do not know whether the DNA synthesis we observed was a result of initiation of DNA synthesis on the nicked template or removal of oxidized bases and subsequent repair of the plasmid. As mentioned above, it is probable that nuclease(s) present in the extract produced additional sites for DNA synthesis and repair. Therefore, these observations may represent a combination of repair synthesis and replication synthesis. According to conventional concepts, extract from control myocardial cells should represent a quiescent state and would not label DNA. However, it must be considered that the process of enzymatic dissociation, associated changes in calcium homeostasis, and the extraction of cellular protein may cause stress and cellular changes that activate DNA synthetic enzymes. Indeed, in our whole-cell preparations, incorporation of label into endogenous DNA was observed in nonischemic cells and may also be related to cell isolation. The DNA synthetic capacity of cell extracts from control myocytes was equivalent to that of a rapidly growing transformed cell line and suggests that myocytes are much more active than previously thought.
Using the cell extracts and plasmid DNA, we were able to obtain some
information about which polymerase(s) could be responsible for our
observations. Both high NaCl and NEM inhibited incorporation of label,
indicating that polymerases ß and are probably not involved. We
did observe 25% to 30% residual activity in the presence of BuphdGTP,
which could be due to polymerase ; however, this is unlikely, since
no activity was observed in the presence of NaCl, which stimulates
polymerase . Whereas polymerase is potently inhibited by
BuphdGTP, polymerase is only partially inhibited at the
concentration we used.61 Additionally, ddTTP caused a 50%
inhibition, which also indicates that polymerase is probably not
involved, since it is unaffected by ddTTP. The partial inhibition by
ddTTP is suggestive of polymerases and . Since polymerase
has been reported in adult rat heart, it is likely that it plays a role
in our observations.51 Polymerase functions in both
DNA replication and repair, both of which could be active in our
preparations.62 We are aware of the problems of using
inhibitors in whole cells and cellular extracts and
recognize that they can display anomalous behavior.63
Given these considerations, our data suggest that polymerase plays
a major role in DNA metabolism in the adult
ventricular myocyte. The preservation of small amounts of
the supercoiled form of the plasmid in the presence of NaCl and NEM is
probably due to inhibition of nuclease, but this inhibition did not
prevent extensive degradation of the plasmid.
During the course of these studies, we observed that plasmids not
exposed to free radicals also incorporated
[-32P]dCTP.
Analysis of the uncut plasmids revealed that some of the native
plasmids contained the relaxed/nicked form, which would serve as a
substrate. Other investigators have observed incorporation into
nontreated plasmids64 65 even when the plasmid
preparation
was highly purified.39 45
While free radical exposure produces primarily DNA strand breaks, UV
radiation can cause base modifications, such as pyrimidine dimers, that
prevent DNA repair or synthesis if there is no ability to excise the
damaged base. The presence of pyrimidine dimers in repair-deficient
mitochondria has been shown to slow or arrest DNA
replication,42 and extracts from excision
repair-deficient cell lines, such as xeroderma pigmentosum, are
unable to incorporate labeled nucleotides into UV-damaged
plasmids.45 64 To determine whether our cell extracts
could incorporate [-32P]dCTP into UV-treated
plasmid,
we performed the repair assay using extract from control cells on
plasmid that had been exposed to UV light. We detected DNA synthesis in
the UV-treated plasmid analogous to that for the same plasmid treated
with the free radical system (compare Figs 4 and
8). While it is
probable that some of the activity we observed was due to the presence
of nicked forms in the native plasmid and nuclease activity,
interruption of DNA synthesis by pyrimidine dimers would have resulted
in reduced labeling. Therefore, it appears that cell extracts from
adult myocytes are able to incorporate nucleotides into DNA
after exposure to UV, which suggests that they have the ability
to excise damaged bases.
Isolated ventricular myocytes made permeable with saponin
and incubated in the presence of [-32P]dCTP
incorporated label into endogenous DNA. At early time
points 
60 minutes, it appeared that there was one predominant DNA
band, suggesting a nonrandom cleavage. However, extraction of DNA from
fresh, whole cardiac tissue indicated the presence of multiple
fragments of various sizes in both the isolated cells and fresh tissue.
Therefore, it is not possible for us to conclude that the fragmented
DNA we observed from permeable cells was due to specific cleavage
characteristic of apoptosis, as has been reported for hypoxic
cultured neonatal rat heart cells.36 It is clear from the
large size of some of the labeled DNA fragments (>58 000 bp) that it
is of nuclear rather than mitochondrial origin (16 500 bp).
The major difference we observed between control and ischemic
cells was in the rate of incorporation of [-32P]dCTP
into endogenous DNA, with incorporation of label being
accelerated in the presence of N2. It is difficult to
determine why there was faster incorporation of label in the
ischemic preparations, but it may be related to enhanced free
radical production under ischemic conditions. Since
there was no external source of free radicals in the whole-cell
preparations, any free radical–mediated damage to DNA had to
result from endogenous metabolic processes that
would be accelerated in the presence of N2. Free radical
generation has been shown to be greatly enhanced in ischemic
myocardium.35 66 67 68 69 70
Additionally, cell
deterioration would be enhanced during ischemia, and activation
of various nucleases would contribute to degradation and reduced label
at later time points.
Although a few studies have suggested possible low-level DNA
synthesis in adult ventricular myocytes, most reports have
concluded that there is no active DNA synthesis or repair. Three
histological studies in models of experimental
infarction have reported limited nuclear DNA synthesis. Approximately
9% of left ventricular myocyte nuclei (versus 2% in
control cells) were labeled after 10 successive injections, at 12-hour
intervals, of [3H]thymidine between 5 and 10 days after
extensive myocardial infarction in rats. The label was confined to the
subepicardial layer of the surviving myocytes.17 Another
long-term study reported that 2% of left
ventricular myocyte nuclei (versus 0.25% in
sham-operated rats) were labeled with BrdU 7 days after
experimental coronary artery stenosis. Additionally,
there was enhanced expression of proliferating cell nuclear antigen RNA
and proliferating cell nuclear antigen protein at 2 and 7 days after
surgery.18 Labeling of left ventricular
myocyte nuclei with BrdU was increased 1% 1 week after
coronary artery narrowing and production of
hypertrophy.19 One short-term
study20 reported DNA synthesis in nuclei from freshly
isolated adult rat ventricular myocytes. These nuclei
incorporated approximately half of the [-32P]dCTP
into
DNA as nuclei from neonatal myocytes, which maintain active DNA
synthesis. Labeling was increased in the nuclei from adult cells
by including myocardial cell extract from neonates. Through in vivo
injections of BrdU and [3H]thymidine, the author
concluded that the observed labeling was due to DNA replication rather
than repair.20 However, the BrdU studies were done in
1-week-old rats, when DNA replication is still active.
Consequently, it is not possible to conclude with certainty whether the
activity in adult nuclei was due to replication or to repair. Our
results are consistent with those in nuclei from freshly
isolated cells in that adult myocytes can incorporate label into
nuclear DNA. However, our results suggest that a repair mechanism is
present and that adult ventricular myocytes can be as
active as a mitotic cell line. Collectively, these studies indicate
that under certain conditions DNA synthesis can be observed in adult
ventricular myocyte nuclei, and our results suggest that
DNA repair is also present.
Our studies, which were performed in cells isolated from the adult cat ventricle, support active DNA repair synthesis of both endogenous nuclear and exogenous plasmid DNA. Furthermore, there are differences between cells exposed to N2 and cells maintained in a normoxic environment, and these studies should provide a background for further delineation of the active components in the cell extracts. A DNA synthetic response to cellular damage may have important pathophysiological and clinical implications. It may signal changes in cytosolic and membrane processes in cells surviving ischemic injury and thus influence their behavior after "healing" or during a subsequent ischemic event.
|
Selected Abbreviations and Acronyms |
|---|
|
|
Acknowledgments |
|---|
This work was supported by grants to Dr Kozlovskis from the American Heart Association, Florida Affiliate (No. 9401215), and the Applebaum Foundation and to Dr Myerburg from the National Institutes of Health (grant HL-21735) and the American Heart Association, Florida Affiliate (No. 9401204). We gratefully acknowledge Dr George Wright for generously providing butylphenyl dGTP, Dr Kathleen Downey for her critical analysis and helpful discussions, and Shirley Delgado for the preparation of this manuscript.
|
Footnotes |
|---|
Reprint requests to Patricia L. Kozlovskis, PhD, University of Miami Medical School, Department of Medicine (R-94), PO Box 016960, Miami, FL 33101.
Received May 22, 1995; accepted October 25, 1995.
|
References |
|---|
|
TopAbstractIntroductionMaterials and Methods Results Discussion References |
|---|
- Morkin E, Ashford TP. Myocardial DNA synthesis in experimental cardiac hypertrophy. Am J Physiol. 1968;215:1409-1413.
-
Grove D, Zak R, Nair KG, Aschenbrenner V.
Biochemical correlates of cardiac hypertrophy, IV:
observations on the cellular organization of growth during myocardial
hypertrophy in the rat. Circ
Res. 1969;25:473-485.
[Abstract/Free Full Text] - Rumyantsev PP. DNA synthesis in atrial myocytes of rats with aortic stenosis. Adv Myocardiol. 1983;4:147-162. [Medline] [Order article via Infotrieve]
-
Mann DL, Kent RL, Cooper G IV. Load regulation
of the properties of adult feline cardiocytes: growth induction
by cellular deformation. Circ Res. 1989;64:1079-1090.
[Abstract/Free Full Text] - Kellerman S, Moore JA, Zierhut W, Zimmer H-G, Campbell J, Gerdes AM. Nuclear DNA content and nucleation patterns in rat cardiac myocytes from different models of cardiac hypertrophy. J Mol Cell Cardiol. 1992;24:497-505. [Medline] [Order article via Infotrieve]
-
Soonpaa MH, Field LJ. Assessment of
cardiomyocyte DNA synthesis during hypertrophy
in adult mice. Am J Physiol. 1994;266:H1439-H1445.
[Abstract/Free Full Text] - Lampidis TJ, Schaiberger GE. Age-related loss of DNA repair synthesis in isolated rat myocardial cells. Exp Cell Res. 1975;96:412-416. [Medline] [Order article via Infotrieve]
- Van Krimpen C, Smits JFM, Cleutjens JPM, Debets JJM, Schoemaker RG, Boudier Hajs, Bosman FT, Daemen MJAP. DNA synthesis in the non-infarcted cardiac interstitium after left coronary artery ligation in the rat: effects of captopril. J Mol Cell Cardiol. 1991;23:1245-1253. [Medline] [Order article via Infotrieve]
- Rumyantsev PP. Interrelations of the proliferation and differentiation processes during cardiac myogenesis and regeneration. Int Rev Cytol. 1977;51:187-273.
- Claycomb WC. DNA synthesis and DNA enzymes in terminally differentiating cardiac muscle cells. Exp Cell Res. 1979;118:111-114. [Medline] [Order article via Infotrieve]
- Nag AC, Carey TR, Cheng M. DNA synthesis in rat heart cells after injury and the regeneration of myocardia. Tissue Cell. 1983;15:597-613. [Medline] [Order article via Infotrieve]
-
Marino TA, Haldar S, Williamson EC, Beaverson K, Walter
RA, Marino DR, Beatty C, Lipson KE. Proliferating cell nuclear
antigen in developing and adult rat cardiac muscle cells.
Circ Res. 1991;69:1353-1360.
[Abstract/Free Full Text] - Nag AC, Cheng M. DNA synthesis of adult mammalian cardiac muscle cells in long-term culture. Tissue Cell. 1986;18:491-497. [Medline] [Order article via Infotrieve]
-
Claycomb WC, Bradshaw HD Jr. Acquisition of multiple
nuclei and the activity of DNA polymerase and reinitiation of DNA
replication in terminally differentiated adult cardiac muscle cells in
culture. Dev Biol. 1983;99:331-337. [Medline]
[Order article via Infotrieve]
- Claycomb WC, Moses RL. Growth factors and TPA stimulate DNA synthesis and alter the morphology of cultured terminally differentiated adult rat cardiac muscle cells. Dev Biol. 1988;127:257-265. [Medline] [Order article via Infotrieve]
- Linzbach AJ. Heart failure from the point of view of quantitative anatomy. Am J Cardiol. 1960;5:370-382.