| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Articles |
The Mitochondrial and Metabolic Disease Center, University of California, San Diego, School of Medicine, Departments of
1
Medicine,
2
Neurosciences, and
3
Pediatrics, 200 West Arbor Dr., San Diego, CA 92103-8467.
a Author for correspondence. Fax 619-543-7868; e-mail naviaux{at}ucsd.edu
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
is the
principal polymerase required for mitochondrial DNA replication.
Primary or secondary deficiencies in the activity of DNA polymerase
may lead to mitochondrial DNA depletion. We describe a sensitive and
robust clinical assay for this enzyme. Methods: The assay was performed on mitochondria isolated from skeletal muscle biopsies. High-molecular weight polynucleotide reaction products were captured on ion-exchange paper, examined qualitatively by autoradiography, and quantified by scintillation counting.
Results: Kinetic analysis of DNA polymerase by this method
showed a Km for dTTP of 1.43 µmol/L and a
Ki for azidothymidine triphosphate of 0.861
µmol/L. The assay was linear from 0.1 to 2 µg of mitochondrial
protein. The detection limit was 30 units (30 fmol dTMP
incorporated in 30 min). The linear dynamic range was three orders of
magnitude; 30–30 000 units. Imprecision (CV) was 6.4% within
day and 12% between days. Application of this assay to a mixed
population of 38 patients referred for evaluation of mitochondrial
disease revealed a distribution with a range of 0–2506 U/µg,
reflecting extensive biologic variation among patients with
neuromuscular disease.
Conclusion: This assay provides a useful adjunct to current laboratory methods for the evaluation of patients with suspected mitochondrial DNA depletion syndromes.
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
. This
enzyme is found as an 
,ß heterodimer and as an monomer
associated with four other unidentified cellular proteins
(4). Both and ß subunits have been cloned
(5)(6)(7). The subunit is catalytic and contains both the
polymerase and the 3'-to-5' proofreading exonuclease activities. It is
140 000 daltons in size. The ß subunit is 43 000 daltons and of
unknown function.
The substrate virtuosity displayed by DNA polymerase is
unique among eukaryotic polymerases. The mitochondrial enzyme is an
effective reverse transcriptase (RNA-directed DNA polymerase) when
measured on homopolymeric RNA templates. It is also an effective DNA
polymerase when directed by single-stranded homopolymeric DNA,
single-stranded heteropolymeric DNA, or activated double-stranded DNA
templates (8). The experimental choice of template radically
influences both the measured kinetic performance of the enzyme and its
response to specific inhibitors, such as azidothymidine (AZT)
(9). The most sensitive assays developed to date have
exploited the RNA-directed DNA polymerase activity of the enzyme. These
assays are typically ~10- to 50-fold more sensitive than conventional
assays, which are performed with activated double-stranded DNA
templates (9).
Mitochondrial disorders are among the most recently recognized groups
of diseases known to medicine. They affect both children and adults and
are particularly difficult to diagnose because of a broad spectrum of
phenotypic presentations (10)(11). Muscle biopsy
is often obtained in the course of the diagnostic evaluation of
patients with suspected mitochondrial disease. Several inherited forms
of mtDNA depletion have been described (12)(13)(14). Depletion
of mtDNA may also result from AZT (15) or fialuridine
(16) toxicity. Deficiency in any one of the 20 or so
components of the mtDNA replisome could theoretically lead to mtDNA
depletion, and we have reported a patient in whom mtDNA depletion was
associated with deficient activity of mtDNA polymerase
(17). The development of a sensitive and reliable clinical
assay for DNA polymerase may be useful in clinical practice and in
the elucidation of mechanisms underlying the mtDNA depletion syndromes.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
protein and citrate synthase determinations
Mitochondrial protein was determined from thawed samples by a
small-scale modification of the Lowry method (Bio-Rad kit no.
500-0112). This required ~20 µg of mitochondrial protein.
Mitochondrial yield was quantified by measurement of the mitochondrial
matrix protein citrate synthase (CS) by the method of Sheperd and
Garland (19), which required ~5 µg of mitochondrial
protein. One unit of CS was defined as 1 pmol/min of acetyl CoA
consumed.
dna polymerase assays
Approximately 5 µg of thawed mitochondria was lysed and
stabilized in an equal volume of glycerol lysis buffer (25 mmol/L Tris,
pH 8.0, 50 mmol/L KCl, 20 mL/L Triton X-100, 500 mL/L glycerol),
mixed by trituration in a 10- to 20-µL micropipette, brought to a
concentration of at most 1 g/L with additional glycerol lysis buffer,
and placed on ice. The RNA-directed DNA polymerase assay was a
modification of methods described by Goff et al. (20).
Briefly, four twofold serial dilutions of lysed mitochondria in
glycerol dilution buffer (25 mmol/L Tris pH 8.0, 50 mmol/L KCl, 500
mL/L glycerol) containing ~1.0, 0.5, 0.25, and 0.125 g/L of
mitochondrial protein were prepared for each sample. One microliter of
each dilution was pipetted into the bottom of 25-µL GeNunc module
(48-well, 4 x 12 tray; Nunc cat. no. 2-32549) reaction trays at
room temperature. An equal volume of glycerol lysis buffer was pipetted
into four separate wells and used as the blank. One tube of beef heart
or human muscle mitochondria was thawed and used as an internal
standard that was included in every assay as a control for day-to-day
variation. Nine microliters of 1.11x reaction buffer was added
to each well, and the full 10-µL reaction volume was mixed by rapid
micropipetting six times. The GeNunc reaction trays were sealed with an
adhesive top (supplied by the manufacturer) and floated in a 37 °C
water bath for 30 min. Reactions were stopped by the addition of 5 µL
of each reaction onto 1.3-cm squares drawn and numbered with a
pencil (conveniently done during the 30-min incubation period) on DE81
anion-exchange paper (Whatman; cat. no. 3658 915). The DE81 paper was
washed three times for 5 min each in 250 mL of 2x standard saline
citrate (0.3 mol/L NaCl, 0.03 mol/L sodium citrate, pH 7.0),
once in 250 mL of 950-1000 mL/L ethanol, air-dried, and then analyzed
by autoradiography for 1 h at room temperature using BioMax
MSTM film and a matched intensifying screen
(Kodak). One microliter of 1.11x reaction buffer was spotted in
quadruplicate on DE81 paper and used directly, without washing, as an
internal control for radioisotopic specific activity and decay. The
DE81 paper was then cut into squares containing the high-molecular
weight polymerase reaction products and quantified by scintillation
counting in 5 mL of Ecolume (ICN). The mean background (determined from
quadruplicate blanks) was subtracted from each reaction, and the total
dTMP incorporation into high-molecular weight products (radioactive and
nonradioactive) was calculated.
Optimized conditions were as follows: 50 mmol/L Tris-HCl, pH 8.0, 100
mmol/L NaCl, 2.5 mmol/L KCl, 5 mmol/L MgCl2 and
0.5 mmol/L MnCl2, 10 mmol/L dithiothreitol (DTT),
1 mL/L Triton X-100, 25 mL/L glycerol, 5 µmol/L dTTP, 10 mg/L
poly(rA) (Pharmacia; cat. no. 27-4110-02; mean length, 500 nucleotides;
Mr = 1.85 x
105; 54 nmol/L, yielding a nucleotide
concentration of 27 µmol/L), 5 mg/L
oligo(dT)12–18 (Pharmacia; cat. no. 27-7858-02;
Mr = 4905; 1.0 µmol/L, yielding a
nucleotide concentration of 15 µmol/L and a primer-to-template molar
ratio of 18.5:1), 500 µCi/mL
32P--dTTP (3000
Ci/mmol), 37 °C for 30 min. The final specific activity of the
radionucleotide was typically 150–220 counts/min (cpm) per fmol.
In practice, 0.45 mL of 1.11x reaction buffer (enough for 45
reactions) was prepared by combining 100 µL of 5x buffer stock (0.25
mol/L Tris, pH 8.0, 25 mmol/L MgCl2, 5 mL/L
Triton X-100, 50 mmol/L DTT, and 25 mmol/L KCl), 100 µL of 5x
template stock [50 mg/L poly(rA), 25 mg/L
oligo(dT12–18), and 25 µmol/L dTTP], 5 µL
of 50 mmol/L MnCl2, 25 µL of
32P--dTTP (10
µCi/µL, 3000 Ci/mmol, 0.167 µmol/L final radionuclide
concentration), and 220 µL of diethylpyrocarbonate-treated
water. The 5x buffer and template stocks could be stored for up
to 6 months at -20 °C. MnCl2 could not be
added until the day of the assay because of precipitation that occurred
in the concentrated stocks.
Single-stranded DNA-directed assays were identical to the RNA-directed
assays except that 10 mg/L of poly(dA) was substituted for poly(rA).
Activated DNA-directed assays were performed with 2 µg of
mitochondrial protein, 100 mg/L activated calf thymus DNA
(Worthington), 50 mmol/L Tris-HCl, pH 8.0, 50 mmol/L KCl, 5 mmol/L
MgCl2, 0.5 mmol/L MnCl2, 10
mmol/L DTT, 1 mL/L Triton X-100, 5 µmol/L dTTP, 50 µmol/L each of
the three dNTPs (dGTP, dATP, and dCTP), and 500 µCi/mL
32P--dTTP (3000
Ci/mmol), and incubation at 37 °C for 30 min.
Specific activities were expressed in units per microgram of mitochondrial protein. One unit was defined as 1 fmol of nucleotides (total) incorporated into high-molecular weight products in a 30 min reaction at 37 °C. Moloney murine leukemia virus reverse transcriptase was obtained from Life Technologies. The specific activity was 350 000 BRL units/mg (1 BRL unit = 1 nmol/10 min). This was equivalent to 1.05 x 109 fmol/µg incorporated in a 30 min reaction. AZT-triphosphate (AZT-TP) was obtained from Sierra Bioresearch. Dideoxythymidine triphosphate (ddTTP) was supplied as a 5 mmol/L stabilized solution from Pharmacia (cat. no. 27-2085-01).
kinetic analysis
Michaelis-Menton kinetics were analyzed graphically using the
methods of Eisenthal-Cornish-Bowden direct and Hanes-Wolf ([S]/v vs
[S]) plots (21). AZT-TP and ddTTP inhibition kinetics were
analyzed graphically using double reciprocal and secondary
Kapp plots (22).
statistics
Statistical analyses were performed with Stata (Stata Corporation)
and Excel 98 (Microsoft) software packages.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
) in the range of 0.1–2 µg of mitochondrial protein (data not
shown). The optimum temperature and pH were 40 °C and 8.0,
respectively. Thirty-minute reactions at 37 °C and a pH of 8.0 were
adopted as standard conditions for the assay.
|
To verify that the only DNA polymerase activity detectable in the
mitochondrial isolates was DNA polymerase and not DNA polymerase
, ß, 
, or 
from the nucleus, we tested preparations with a
variety of inhibitors. Table 1
shows the results of mtDNA polymerase activity measured on
three different templates in the presence of several inhibitors.
Aphidicolin (0.1 mmol/L) did not inhibit the mitochondrial polymerase
on DNA templates but produced a 22% inhibition on RNA templates.
Ethidium bromide was a potent inhibitor on both RNA and DNA templates.
Actinomycin D inhibited activity only on double-stranded DNA templates.
ddTTP and phosphonoformate (foscarnet) were potent inhibitors on all
substrates, whereas AZT-TP was a significantly more potent inhibitor
when measured on RNA templates.
|
The effects of divalent cations are summarized in Table 2
. Equivalent catalytic activity was observed on poly(rA) and
poly(dA) templates. The mitochondrial enzyme was eightfold more
efficient on these single-stranded templates than on activated
double-stranded DNA (1105 vs 138 U). Under conditions of limiting dTTP
(1 µmol/L), polymerase reaction velocity on an RNA template was
greatest with MnCl2 alone (145% = 864 U). The
addition of MgCl2 with
MnCl2, or MgCl2 alone,
reduced net polymerization under these conditions. However, when dTTP
was raised to 5 µmol/L or higher, the highest activities were
observed when both MnCl2 (0.5 mmol/L) and
MgCl2 (5 mmol/L) were present. The effect of
variations in other reaction conditions is summarized in Table 3
. NaCl at 100 mmol/L produced maximum activity. Oligo(dT) primer
length also affected the reaction rates. Primer sizes less than 9-mer
were clearly inferior, whereas 12- to 18-mers and 30-mers were
equivalent. We adopted oligo(dT12–18) in the
routine assay.
|
|
We typically isolated 250 µg of mitochondrial protein for every 100 mg of skeletal muscle over 250 mg processed (n = 10). The efficiency of mitochondrial isolation by the method of Hatefi et al. (18) dropped with decreasing biopsy size and could not be used for specimens much smaller than ~250 mg. A typical open muscle biopsy obtained from a child with suspected mitochondrial disease was 600 mg and yielded 1000 µg of mitochondrial protein.
The reproducibility of mitochondrial enzyme assays was limited
initially by incomplete solubilization of organelles after a
freeze-thaw cycle. This led to particulate behavior that arose from
mitochondrial fragments, aggregates, and clumps that were not
distributed homogeneously in solution. To address this problem, we
solubilized the thawed mitochondrial preparations in an equal volume of
glycerol lysis buffer consisting of 25 mmol/L Tris, pH 8.0, 500 mL/L
glycerol, 50 mmol/L KCl, and 20 mL/L Triton X-100. Once solubilized,
66% of the polymerase activity was present in the soluble fraction of
the mitochondrial lysates. This activity remained in solution and was
not pelleted by centrifugation for 10 min at 10 000g (Table 3
). This activity was stable in glycerol lysis buffer for at least
6 h on ice (Table 3
). The effect of multiple freeze-thaw cycles
was also tested. There was no significant effect on polymerase activity
frozen in 0.25 mol/L sucrose and subjected to three freeze-thaw cycles
(data not shown).
Total 32P-dTMP incorporation (as cpm) increased
with decreasing total dTTP down to 0.167 µmol/L (100% of dTTP
supplied as 32P--dTTP,
500 µCi/mL, 3000 Ci/mmol; Fig. 2
A). This produced more intense spots during autoradiography and
more counts during scintillation counting, thus increasing assay
sensitivity, but also led to an overall decrease in the calculated
polymerase reaction velocity, measured as the total (radioactive plus
nonradioactive) dTMP incorporated per unit of time. Increasing the
concentration of 32P--dTTP beyond 500 µCi/mL
(1.1 x 107 dpm and 1.67 pmol per 10-µL
reaction, 0.167 µmol/L) was costly and unnecessary because assay
sensitivity was not a limiting factor. Increasing the total dTTP to 10
µmol/L (100 pmol per 10-µL reaction) decreased radioactive
incorporation by competition with unlabeled nucleotide (molar ratio of
dTTP to 32P-dTTP was 10 µmol/L to 0.167
µmol/L, or 60:1) but produced an overall increase in polymerase
reaction velocity. Formal kinetic analysis of the mitochondrial
polymerase in isolates from human skeletal muscle revealed conventional
hyperbolic initial velocity curves (Fig. 2A
). The
Km for dTTP was determined graphically
by two methods. A Hanes-Wolf plot produced a result of 1.48 µmol/L
for the Km (Fig. 2B
). An
Eisenthal-Cornish-Bowden direct plot produced a result of 1.43 µmol/L
(Fig. 2C
). We adopted a dTTP concentration of 5 µmol/L. This was
3.5-fold higher than the Km for dTTP
in the assay and produced a final radionucleotide specific activity of
150–220 cpm/fmol. Under these conditions, the assay had a detection
limit of 15–30 units (1 unit equals 1 fmol of dTMP incorporated
in 30 min). The intrinsic, linear dynamic range was three orders of
magnitude, permitting quantification of polymerase activities up to
30 000 U (2.2 x 106 incorporated cpm, 20%
of the total radioactivity) per reaction. These higher limits were
tested using purified Moloney murine leukemia virus reverse
transcriptase (data not shown). Specific activities of mtDNA
polymerase in muscle isolates did not exceed 1800 U/µg. In
practice, the linear dynamic range for mitochondrial isolates from
muscle was 20-fold (0.1–2 µg) because of deviations from linearity
that occurred as mitochondrial protein input exceeded 2 µg (data not
shown).
|
The mitochondrial polymerase was exquisitely sensitive to inhibition by
AZT-TP (Ki = 0.861 µmol/L; Fig. 3
). The Ki for ddTTP was
determined in a similar fashion and was found to be 0.0456 µmol/L
(data not shown). In the lower ranges of inhibitor concentrations, both
AZT-TP and ddTTP were competitive inhibitors. However, at higher
concentrations of AZT-TP (>10 µmol/L), the inhibition was mixed and
included both competitive and noncompetitive components (data not
shown).
|
We routinely performed four serial twofold dilutions for each sample
and analyzed the results by both autoradiography and scintillation
counting (Fig. 4
A). These data were used to calculate a best fit, four-point
regression line for each sample. The within-day CV was 6.4% (n =
5; Fig. 4A
). The between-day CV was 12% (n = 11) using
replicate samples frozen in small aliquots, thawed and assayed on
different days over a period of 6 months.
|
When quadruplicate determinations were used to measure DNA polymerase
activity instead of a regression, the within-day imprecision (CV)
was 2.5% (n = 5; Fig. 4B
). However, we found that the accuracy of
this method was inferior to regression. The magnitude of this error was
greater for more dilute samples of mitochondrial protein. When
quadruplicate determinations of samples containing 0.1–0.4 µg of
protein were analyzed, this method systematically overestimated the
true activity in some mitochondrial isolates because of the more
significant effect that a non-zero y-intercept had on
samples containing a lower absolute amount of polymerase activity. The
consequences of the intercept problem are illustrated in Fig. 4C
. In
this experiment, a single sample of muscle mitochondria was analyzed
both by serial, twofold, four-point regression and by quadruplicate
determinations of four different inputs of mitochondrial protein. Fig. 4C
illustrates the same data, analyzed by two different methods.
Graphical analysis of the results produced a regression line that had a
positive y-intercept that was not removed by subtraction of
the experimental (glycerol lysis buffer) blanks. For example, in the
sample illustrated on the left in Fig. 4C
, the y-intercept
was 40 117 cpm. The best fit regression line was y
(cpm) = 116 904x + 40 117. The polymerase activity
for this sample was calculated by dividing the slope by the specific
activity of 32P-dTTP in the reaction, which was
188 cpm/fmol in the experiment illustrated. The result was 622 U/µg
(116 904 cpm/µg divided by 188 cpm/fmol). The apparent specific
activity of the sample was also calculated on the basis of
quadruplicate determinations (shown on the right in Fig. 4C
). These
data converged on the true value of 622 U/µg as the absolute activity
in the sample increased, but each of the quadruplicate values
systematically overestimated the true activity because the intercept
was not taken into account. We adopted regression analysis for the
standard assay.
We applied the assay to the measurement of mtDNA polymerase in 38
patients (age range, 3 months to 38 years) with neuromuscular disease
in whom muscle biopsies were carried out for the evaluation of a
suspected mitochondrial cause. This population was chosen because
consent for anterior quadriceps muscle biopsies is not easily obtained
from age-matched, healthy controls. Twenty-four patients (63%) had
confirmed mitochondrial disease. Fourteen patients (37%) had
encephalomyopathies or myopathies of unknown cause. The results were
normalized for mitochondrial content by expressing them as the ratio of
polymerase (in U/µg) and CS (in U/µg) specific activities These
were then arrayed as frequency distribution in Fig. 5
. The mean, median, mode, and SD were 668, 551, and 490, and
506, respectively. The range was 0–2 506. The patient with no
detectable activity had Alpers syndrome and mtDNA depletion. The case
report describing this patient has been published (17). The
patient with the highest mitochondrial polymerase activity normalized
to CS was an 11-year-old boy with profound lactic acidemia of abrupt
onset at puberty.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
activity in mitochondrial isolates obtained from human skeletal muscle.
This assay had excellent sensitivity (30 U/µg mitochondrial protein),
specificity (absence of nuclear DNA polymerase contamination), and
reproducibility (within- and between-day CVs, 6.4% and 12%,
respectively). It could be performed on isolated mitochondria that had
been frozen and stored in liquid nitrogen or at -80 °C until
sufficient numbers of samples had been accumulated for convenient
assay. Typically, <30 µg of total mitochondrial protein was
sufficient for all three required assays: protein concentration (20
µg); CS activity (5 µg); and mitochondrial polymerase activity (5
µg).
The effects of 16 factors on measured mtDNA polymerase activity in
skeletal muscle were evaluated: reaction time, total protein,
temperature, pH, detergent, Na+,
K+, Mg2+,
Mn2+,
PO42-, DTT, RNA template,
single-stranded DNA template, activated double-stranded DNA template,
nucleotide (dTTP) concentration, and primer length. Although it was not
possible to explicitly test all combinations
(>216) of these factors, a few important
interactions were noted. For example, the presence of
Mg2+ under limiting dTTP concentrations, at or
below the Km (1 µmol/L dTTP; Table 2
), produced a strong reduction in net polymerase activity. However,
when substrate dTTP was increased to 5 µmol/L, the addition of
Mg2+ stimulated polymerase activity 15% (85%
with Mn2+ alone, 100% with
Mn2+ and Mg2+; 5 µmol/L
dTTP; Table 2
). The measured activity in this assay represents a sum of
the forward polymerase and "backward" 3'-to-5' proofreading
exonuclease activities of DNA polymerase . The exonuclease activity
of DNA polymerase is known to be Mg2+
dependent (4), whereas the polymerase activity may use
either Mn2+ or Mg2+ (with a
preference for Mn2+ on RNA templates and
Mg2+ on DNA templates; Table 2
). This combination
of catalytic characteristics leads to kinetic stalling of the enzyme
under conditions of limiting dTTP (<2–5 µmol/L), during which the
backward exonuclease activity predominated. The details of
divalent cation preference were also influenced by monovalent cation
concentration and ionic strength. Lower ionic strengths (0–50 mmol/L
NaCl) favor exonuclease activity, whereas higher ionic strengths
(100–150 mmol/L NaCl) favor polymerase and inhibit the exonuclease
(4)(23).
The assay was highly specific for mtDNA polymerase . Cellular nuclei
were removed by differential centrifugation before assay, and an RNA
template was used to measure polymerase activity. Because cellular
nuclei were removed, little or no detectable nuclear DNA polymerase
activity was present. This was evidenced by the ethidium bromide-,
AZT-TP-, and ddTTP-sensitive activity observed with the assay (Table 1
), and the classical Michaelis-Menten kinetics of the enzyme (Fig. 2
). The Km for dTTP was 1.43
µmol/L, as determined by the Eisenthal-Cornish-Bowden method
(21). This value, obtained with relatively crude
mitochondrial isolates, corresponded well to published values of 0.4–2
µmol/L using enzyme preparations that were purified as much as
10 000-fold (24)(25)(26). In addition, RNA was used as the
template. Under these conditions, even if nuclear polymerase
contamination did occur, virtually no DNA polymerase , ,
or activity would be detected because these nuclear polymerases are
unable to use RNA templates
(9)(24)(27). DNA polymerase ß has
good RNA-directed activity, roughly equal to its DNA-directed activity,
but it is insensitive to AZT-TP [Ki
>1000 µmol/L, (28)]. The enzyme measured in this assay
was exquisitely sensitive to AZT-TP
(Ki = 0.861 µmol/L; Fig. 3
) and
therefore was not DNA polymerase ß. The constellation of catalytic
properties exhibited by the enzyme measured in this assay showed that
our methodology is highly specific for mtDNA polymerase and is
devoid of detectable contamination with DNA polymerases , ß, ,
or .
The physical removal of cellular nuclei and cytoplasm (containing
soluble proteins, nucleotides, and mRNA, rRNA, and tRNA) before assay
increased the sensitivity, specificity, and reliability of the assay.
However, it also led to the unavoidable loss of an unknown fraction of
DNA polymerase that occurs naturally outside of mitochondria in
postmitotic cells, such as skeletal muscle. Bona fide DNA polymerase
occurs both in the nucleus (29) and in the cytoplasm
outside of mitochondria as well as in mitochondria in
nonproliferating cells (30). In contrast, dividing cells
contain DNA polymerase only in mitochondria (31). The
physical isolation of mitochondria before assay also reduced the
concentration of cellular nucleotides introduced nonspecifically into
the reactions, permitting tighter control of dTTP concentrations for
kinetic analysis (Fig. 2
).
The capture of high-molecular weight polynucleotide reaction products on anion-exchange paper (DE81) in this assay helped to avoid the tedious steps of trichloroacetic acid precipitation, collection, and washing of individual spun glass filters that are used routinely in most other polymerase assays (9)(24)(28). The simplicity of DE81 paper contributed to improved replicate precision and dramatically reduced the time and labor required to assay four-point dilutions and quadruplicate patient samples.
We found that measurement of polymerase activity by regression analysis
was significantly more accurate than quadruplicate determinations (Fig. 4
). The accuracy of quadruplicate determinations was reduced because of
the problem of assay background (non-zero y-intercept) in
some samples that was not eliminated by subtraction of the sample
buffer blank (Fig. 4C
). When results were calculated by regression, the
effects of a non-zero y-intercept were easily identified and
accounted for. The slope of the regression line was the true function
of the polymerase activity in the sample. The y-intercept
was a function of non-polymerase components in samples that produced a
fixed background of radioactivity in the assay. Not all samples
displayed non-zero y-intercepts, and the magnitude of this
effect varied from sample to sample. It was small in certain
mitochondrial isolates (data not shown) but large in others (Fig. 4C
).
Because it was not possible to predict which samples would show a
significant non-zero y-intercept without performing
regression analysis, we routinely adopted regression as a standard part
of this assay.
Analysis of the frequency histogram shown in Fig. 5
showed that
activities of mitochondrial polymerase in this heterogeneous population
did not follow a gaussian distribution. The distribution was
skewed to the right and revealed gaps that suggested that the
distributions were sampled from two or more distinct populations. This
was biologically plausible because the patients we studied represented
a complex mixture of mitochondrial (63%) and nonmitochondrial (37%)
encephalomyopathies and myopathies. The standard deviation was
calculated (SD = 506) but was not particularly meaningful because
of the strong departure from a gaussian distribution. The large
biologic range revealed by this assay was reflected by the large ratios
of population standard deviations to the population means (
/µ
= 506/668 = 76%). This biologic variation was 6- to 12-fold
greater than the experimental variation of the assay (CV = 6–12%),
and therefore was real and not an artifact of the assay methodology.
When the clinical presentations were correlated with DNA polymerase
activity represented at the extremes of the frequency histograms, we
found that patients on both tails of the distribution had very strong,
high-mortality phenotypes. The patient with no detectable DNA
polymerase had Alpers syndrome and mtDNA depletion (17).
The patient with the highest polymerase/CS activity was a
well-nourished 11-year-old boy with precocious puberty and abrupt onset
of profound lactic acidemia (17 mmol/L), diabetes insipidus,
thiamine-resistant wet beriberi (encephalopathy, polyneuropathy,
cardiomyopathy), seizure disorder, myopathy, temporal hair graying, and
treatment-resistant megakaryocytic leukemia. Both patients died.
The assay that we describe combines features from classical methods in
virology that are used to measure reverse transcriptase in retroviruses
(20) with conventional RNA- and DNA-directed DNA polymerase
assays used to measure purified fractions of mtDNA polymerase
(4)(23). The combination of these methods
enabled us to produce a more sensitive and reliable assay for the
measurement of mtDNA polymerase . This was done in clinically
relevant samples (0.3–0.6 g) of human skeletal muscle biopsies
submitted as part of the systematic histochemical, ultrastructural,
molecular, and biochemical evaluation of patients with suspected mtDNA
depletion.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
: identification of residues critical for exonuclease activity and dideoxynucleotide sensitivity. Biochemistry 1998;37:10529-10539.
[Medline]
[Order article via Infotrieve]
. Genomics 1996;36:449-458.
[ISI][Medline]
[Order article via Infotrieve]
, ß, , , incorporate fialuridine (FIAU) monophosphate into DNA and are inhibited competitively by FIAU triphosphate. Biochemistry 1994;33:14620-14624.
[Medline]
[Order article via Infotrieve]
deficiency and mitochondrial DNA depletion in a child with Alpers syndrome. Ann Neurol 1999;45:54-58.
[ISI][Medline]
[Order article via Infotrieve]
at sites of multiple adjacent analog incorporation, decrease mtDNA abundance, and cause mitochondrial structural defects in cultured hepatoblasts. Proc Natl Acad Sci U S A 1996;93:3592-3597.
: utilization of dUTP and dTTP during in vitro DNA synthesis. Nucleic Acids Res 1988;16:5645-5659.
is inhibited competitively and noncompetitively by phosphorylated zidovudine. Circ Res 1994;74:344-348.
by different mechanisms. Biochem Biophys Res Commun 1991;179:776-783.
[ISI][Medline]
[Order article via Infotrieve]
, ß are able to incorporate anti-HIV deoxynucleotides into DNA. J Biol Chem 1992;267:21459-21464.
from synaptosomal mitochondria and rat-brain nuclei. Eur J Biochem 1977;81:249-258.
[ISI][Medline]
[Order article via Infotrieve]
-polymerase. J Biol Chem 1977;252:3351-3356.
is expressed and translated in the absence of mitochondrial DNA maintenance and replication. Nucleic Acids Res 1996;24:2753-2759.
The following articles in journals at HighWire Press have cited this article:
|
|
 |
|
  E. Fosslien Mitochondrial Medicine - Molecular Pathology of Defective Oxidative Phosphorylation Ann. Clin. Lab. Sci., January 1, 2001; 31(1): 25 - 67. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
