| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Molecular Diagnostics and Genetics |
1 Unité du Métabolisme Protéino Energétique, UMR INRA 1019, 63009 Clermont-Ferrand, France.
2 Laboratoire de Biochimie, Hôpital Debrousse, Hospices Civils de Lyon, 69322 Lyon, France.
3 Laboratoire de Biolphysique Moléculaire et Cellulaire, UMR CNRS 5090, CEA Grenoble, 38054 Grenoble, France.
aAddress correspondence to this author at: UMPE, UMR INRA 1019, 58, rue Montalembert, 63009 Clermont-Ferrand, France. Fax 33-4-7360-8255; e-mail stepien{at}clermont.inra.fr.
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Methods: To determine altered mtDNA copy number in muscle from nine patients with single or multiple mtDNA deletions, we generated calibration curves from serial dilutions of cloned mtDNA probes specific to four different mitochondrial genes encoding either ribosomal (16S) or messenger (ND2, ND5, and ATPase6) RNAs, localized in different regions of the mtDNA sequence. This method was compared with quantification of radioactive signals from Southern-blot analysis. We also determined the mitochondrial-to-nuclear DNA ratio in muscle, liver, and cultured fibroblasts from a patient with mtDNA depletion and in liver from two patients with mtDNA overreplication.
Results: Both methods quantified 5–76% of deleted mtDNA in muscle, 59–97% of mtDNA depletion in the tissues, and 1.7- to 4.1-fold mtDNA overreplication in liver. The data obtained were concordant, with a linear correlation coefficient (r2) between the two methods of 0.94, and indicated that quantitative PCR has a higher sensitivity than Southern-blot analysis.
Conclusions: Real-time quantitative PCR can determine the copy number of either deleted or full-length mtDNA in patients with mitochondrial diseases and has advantages over classic Southern-blot analysis.
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
mtDNA mutations have been identified in high numbers and broad variety in both multisystem and tissue-restricted disorders. More than 200 disorders are associated with specific point mutations or single deletions in mtDNA (2). Furthermore, because of the nuclear contribution to the synthesis of respiratory chain enzymes, other clinical disorders inherited as Mendelian traits have been associated with multiple deletions or depletion of mtDNA. mtDNA depletion syndrome (MDS) is a quantitative disorder characterized by a variable tissue-specific reduction in mtDNA copy number. The clinical phenotype of MDS is highly dependent on the residual mtDNA concentrations in tissues (3). mtDNA mutations (including deletions) are frequently heteroplasmic, with mutated and wild-type mtDNA coexisting at different respective amounts in cells and subsequently in the affected tissues (4). All studies carried out to date have shown that a percentage of mutated or deleted mtDNA >60% is necessary to reveal a mitochondrial respiratory chain defect in a specific organ (5)(6). Moreover, the percentage of heteroplasmy correlates with the disease development and the severity of the clinical phenotype, as described for the T8993G mutation. This mutation can produce the lethal infantile Leigh syndrome (at least 90% mutant load) or the neurogenic muscle weakness, ataxia, and retinitis pigmentosa (NARP) syndrome with late onset (
60–80% mutant load) (7).
The lifetime accumulation of somatic mtDNA changes has been proposed as a major factor in aging (8), but the different methods of quantification have led to contradictory results: some studies showed that most cells do not contain mtDNA deletions (9)(10), whereas most others showed an accumulation of either deleted mtDNA (11)(12)(13)(14)(15) or mtDNA point mutations (16)(17)(18).
To date, the qualitative or quantitative diagnosis of mtDNA disorders, including quantification of the heteroplasmy percentage, has been carried out mainly by standard methods such as Southern-blot analysis. This method requires large tissue biopsies and difficult analyses that often produce various artifacts.
Several PCR-based methods for detecting mtDNA deletions and point mutations have been developed, although most of these assays have generated only qualitative or semiquantitative data (19)(20)(21)(22)(23)(24). Recently, a new technology, real-time quantitative PCR (q-PCR), has been proposed to quantify DNA copy number and is now an impressive tool for quantifying deleted mtDNA in tissues (25).
We developed experimental conditions that allowed us to quantify precisely, by real-time q-PCR, different qualitative and quantitative mtDNA defects, including deletion, depletion, and overreplication of mtDNA, specifically identified in several mitochondrial pathologies. In this study, we describe this new method, simultaneously compare real-time q-PCR with classic Southern-blot analysis, demonstrate the multiple advantages provided by this new q-PCR technology, and detail the major variables to be controlled for optimum quantification, such as the quality of extracted mtDNA, oligonucleotide design, and PCR conditions.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Patients harboring single or multiple large-scale mtDNA deletions (patients 1–9).
The clinical and molecular features of patients harboring single or multiple large-scale mtDNA deletions (patients 1–9) are summarized in Table 1
.
|
Patient with MDS (patient 10).
This infant died at 12 months of age from a MDS with total intestinal villous atrophy and malabsorption syndrome as the early clinical onset, followed by hypotonia, mental regression, and progressive liver failure. Measured enzymatic activities of the respiratory chain complexes that contained mtDNA-encoded subunits (complexes I, III, and IV) were severely decreased in muscle and liver biopsies.
Patients with mtDNA overreplication (patients 11 and 12).
Patient 11 presented with a hepatocerebral form of severe neonatal lactic acidosis and liver failure. Citrate synthase activity was greatly increased in a liver biopsy (224 U mg of protein; reference interval, 43–94 U mg of protein), and measured enzymatic activities of the enzymatic respiratory chain complexes containing mtDNA-encoded subunits were severely decreased (9%, 19%, and 10% of the control means for complexes I, III, and IV activities, respectively). In contrast, the activities of succinate dehydrogenase and complex II, which are encoded by nuclear DNA (nDNA) only, were within the appropriate reference intervals. Southern-blot analysis of a liver biopsy failed to detect either mtDNA deletion or a reduction in the mtDNA copy number.
Patient 12 presented with a hepatocerebral form of severe neonatal lactic acidosis associated with tubulopathy, muscle wasting, and liver failure. Enzymatic measurements of the respiratory chain complexes in muscle and liver biopsies revealed that the activities of complexes I and IV were severely decreased (10% and 30% of the control means for complexes I and IV, respectively in muscle; 6% of the control means for complexes I and IV in liver). Citrate synthase activity was greatly increased in liver (214 U; reference interval, 43–94 U). Southern-blot analysis of the liver biopsy failed to detect either large-scale mtDNA deletion or copy number reduction. Furthermore, no mutation in the 22 mitochondrial tRNA genes of liver mtDNA was found. A sibling, a girl, died of the same syndrome with similar conditions.
detection and mapping of MTDNA deletions
mtDNA deletions were detected by PCR as described previously (26)(27). To characterize the nucleotide sequences at the breakpoints of deleted mtDNA, we performed direct sequencing of the PCR fragments in both directions on an automated sequencer (373A Applied Biosystems) with the DYEnamicTM terminator Cycle Sequencing Kit (Amersham Biosciences).
dna quantification by southern-blot analysis
For mtDNA and nDNA quantification by Southern-blot analysis, the blots were hybridized with a mitochondrial (ND1) probe and a nuclear (18S rRNA) probe as described previously (3). The multicopy 18S rRNA nuclear gene was selected to obtain optimum sensitivity for nDNA signals. The radioactive signals from the mtDNA and nDNA bands were quantified both by scanning the autoradiographic exposures with a Scanning Densitometer (CS 9000; Shimadzu) and by direct radiographic quantification with a PhosphorImager apparatus (Bio-Rad) with Molecular Analyst Software.
dna quantification by real-time Q-PCR
For mtDNA and nDNA quantification by PCR analysis, we generated calibration curves with serial dilutions of the target gene PCR products. We selected the single-copy nuclear ATP synthase ß gene (ATPsynß) to determine the real ratio of mtDNA to nDNA copy number. PCR products of mitochondrial genes (16S, ND2, ATPase6, and ND5) and the nuclear ATPsynß gene (Table 2
) were subcloned into the plasmid vector pGEM®-T Easy (pGEM-T and pGEM-T Easy Vector Systems; Promega) according to the manufacturer’s instructions. Plasmid DNA was isolated and purified with Nucleobond® AX cartridges (Macherey-Nagel). Mitochondrial and nuclear insert length was verified by restriction analysis with EcoRI, and each plasmid concentration was determined by triplicate measurements of the absorbance at 260 nm.
|
The molecular concentrations of the plasmid stock solutions were calculated, and the stock solutions were diluted from 2 x 108 copies to 2 x 104 copies/µL for mtDNA and from 2 x 106 copies to 2 x 102 copies/µL for nDNA to generate the calibration curves. PCRs were performed in a LightCycler instrument with 2 µL of master mixture (LightCycler-FastStart DNA Master SYBR Green I; Roche Diagnostics) containing Taq DNA polymerase, deoxynucleotide triphosphates, and SYBR Green I; 3 mM MgCl2, 0.5 µM each of the forward (for) and reverse (rev) primers (Table 2
), 5 µL of the DNA template (2 ng/µL) or calibrator (with a known copy number); and water in a final volume of 20 µL. The primers were designed with use of Oligo 6.0 software (MedProbe). The same DNA dilutions were used for both nDNA and mtDNA amplification.
The reaction was conducted as follows: 1 cycle of Taq DNA polymerase activation for 8 min at 95 °C, followed by 45 cycles (temperature transition of 20 °C/s) at specific conditions for each fragment (Table 2
). At the end of the amplification process, melting curves were analyzed between 65 and 95 °C (temperature transition of 0.1 °C/s) with continuous fluorescence monitoring to control for the absence of nonspecific products. Calibration curves were used to quantify mtDNA and nDNA copy numbers, which were based on the linear relationship between the crossing point cycle values and the logarithm of the starting copy number. The slope was determined, and the corresponding real-time PCR efficiencies for each mitochondrial and nuclear gene amplification were calculated according to the equation: E = 10(-1/slope). The data are means of three serial measurements. The run-to-run variability, verified for several controls and patients, was 2%. This variability is usually reported to be 0–5% (28). However, to avoid any experimental error, we included controls and patients for each mtDNA defect in the same run.
nucleotide polymorphism detection in mitochondrial ATPASE6 and ND5 genes
For patients and a control population (100 individuals, including the controls used in this study), mtDNA fragments encompassing the PCR products of the ATPase6 and ND5 genes (Table 2
) were amplified by PCR from positions 8648 to 9990 and 12572 to 13510, respectively. For patients 1 and 3, direct sequencing of these mtDNA fragments was performed in both directions as described above.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
). Single mtDNA deletions had been detected previously in the patient samples by a PCR-based screening procedure (27), and the deletion boundaries had been sequenced (Table 1
). The deletion size ranged from 2950 bp (patient 5) to 7436 bp (patients 1 and 3). The amount of deleted mtDNA in samples was quantified by both Southern blotting and real-time q-PCR analysis. In the q-PCR analysis, the mtDNA copy number (total and wild-type) was quantified by amplification of four randomly distributed mtDNA fragments: two fragments outside (16S and ND2 gene fragments) and two others within the deleted region (ND5 and ATPase6 fragments); for patients 4–7, only the ND5 fragment was amplified within deleted regions (Table 2
). For patients 8 and 9, who harbored multiple mtDNA deletions, both ND5 and ATPase6 fragments were included in most previously sequenced deletions (87–100%) (26)(29). Several remaining deletions with very low percentages of heteroplasmy, localized from primers ND5 rev to 16S for, ND2 rev to ATPase6 for, and ATPase6 rev to ND5 for, could not be quantified regardless of the primers used. The copy numbers of the four mtDNA fragments were quantified in control individuals (without deleted mtDNA) to correct for the variability of q-PCR efficiency between fragments and thus to relatively quantify the mtDNA.
The results were compared with those obtained with Southern-blot analysis (Table 1
). Higher percentages of deleted mtDNA were observed by q-PCR compared with the values found by Southern-blot analysis. This was true whether the Southern blots were analyzed by scanning densitometry of autoradiographs or by a PhosphorImager (patient 4). The linear correlation coefficient between the two data sets (r2) was 0.94 (P <0.0001; Fig. 1
), but the shift of the slope toward the q-PCR axis showed that this method is more sensitive than the Southern-blot approach. One possible reason for this could be either incomplete or nonhomogeneous mtDNA loading on the Southern-blot transfer membrane.
|
For patient 1, quantification of ATPase6 and ND5 (both localized in the deleted region) revealed that 17.9% (Table 1
) and 74.8% (data not shown), respectively, of the mtDNA had been deleted. To explain this discrepancy, we hypothesized that a point mutation in the ND5 gene could be responsible for reduced amplification of this fragment from full-length mtDNA molecules. Regions flanking the q-PCR ATPase6 and ND5 fragments (Table 2
) were sequenced, revealing the presence of a new transition at nucleotide (nt) 13020 of the Cambridge mtDNA sequence (2)(30). This position corresponds to the second nucleotide from the 3' end of the ND5 forward primer and thus should lead to lowered amplification efficiency. This T13020C transition does not affect the amino acid sequence of the ND5 protein (G228G), but creates a new HaeIII restriction site. HaeIII restriction fragment length polymorphism analysis of the PCR product (nt 12572–13510) revealed that this polymorphism was present in all mtDNA molecules from the patient’s muscle, from the patient’s mother, and from 2 of 100 control individuals. Moreover, none of the control individuals used in this study harbored this polymorphism (data not shown).
MTDNA depletion or overreplication
Total DNA from patients with mtDNA depletion or overreplication was analyzed simultaneously by Southern blotting and real-time q-PCR.
mtDNA depletion (patient 10).
Southern blots were hybridized with 32P-labeled mitochondrial (ND1) and nuclear (18S rRNA) probes, and the radioactive signals were quantified with a PhosphorImager (Fig. 2
). q-PCR was performed by amplification of mitochondrial ND2 and nuclear ATPsynß (Fig. 3
). The quantification was confirmed by amplification of a second mtDNA fragment (ATPase6) to avoid any mtDNA polymorphisms (data not shown). The same controls were used for both assays for each tissue or cell type. The two methods produced consistent data: 97.1%, 78.3%, and 71.0% mtDNA depletion in muscle, liver, and fibroblasts, respectively, by Southern-blot analysis (Fig. 2
and Table 3
), and 96.7% (Fig. 3
and Table 3
), 73.7%, and 58.5% mtDNA depletion (Table 3
) by q-PCR.
|
|
|
mtDNA overreplication (patients 11 and 12).
mtDNA in liver from patients 11 and 12 was assayed by Southern-blot and q-PCR analysis (Table 3
). q-PCR analysis revealed significantly increased mtDNA in the samples from both patients (2.3- and 4.1-fold increase, respectively). Southern-blot analysis also revealed increases of 1.7- and 2.3-fold. The higher mtDNA/nDNA ratios obtained by q-PCR were probably attributable to more precise quantification of the high mtDNA copy numbers compared with Southern blotting, in which signal saturation could not be avoided.
efficiency of Q-PCR
Calibration curves were generated from duplicate measurements of three independent serial dilutions of the target gene PCR products (Fig. 3
). The crossing point cycle and the starting copy number logarithm were well correlated, with variability <10%. The corresponding amplification efficiencies, 94–100%, were in the optimum range, but even a slight variation in efficiency could lead to nonnegligible differences in the absolute copy number (31). Thus, results obtained from the four different mtDNA probes were normalized by amplification of control full-length mtDNA. To evaluate the variability of the q-PCR results, we quantified the copy numbers in muscle biopsies from 10 control individuals for the four amplified mtDNA fragments, with at least three determinations per control (not shown). The maximum SD obtained (0.06) between the values from the four mtDNA fragments attests to the reliability of the q-PCR method.
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Until recently, Southern-blot analysis was the standard method for assessing quantitative mtDNA alterations or deletions when mitochondrial diseases were diagnosed (27)(34)(35). However, this method requires a large amount of DNA from tissue biopsies and is time-consuming because it requires several serial steps, which are not always carried out with well-defined quality controls. In addition, this method is susceptible to artifacts, such as membrane saturation for DNA binding; incomplete DNA transfer; hybridization background, which could lead to inaccurate determinations; and signal saturation when detection is by autoradiography. Thus, this method is less suitable for rapid and accurate pre- and postnatal diagnosis or for family inheritance studies. In the present work, we studied several patients presenting three types of mtDNA alterations: single or multiple heteroplasmic large-scale mtDNA deletions, mtDNA depletion, and significant increases in mtDNA copy number (overreplication). Each type of mtDNA alteration was assessed in the patient samples and appropriate controls by both Southern blotting and real-time q-PCR. We conclude that real-time q-PCR is an accurate method that could be applied to the diagnosis of diverse mitochondrial diseases linked to quantitative mtDNA alterations.
quantification of deleted MTDNA
Samples exhibiting single or multiple mtDNA deletions (Table 1
) were analyzed by both methods, and the percentage of heteroplasmy was assessed in all cases. Real-time q-PCR consistently produced higher deletion percentages, especially for patient 4, and the correlation between the results obtained from both methods was linear and showed that PCR quantification is more sensitive than Southern blotting.
The difference between the two methods was constant (6%) as shown by the linear correlation analysis. This could be attributable to the different types of detection. For Southern-blot quantification, two radioactive bands corresponding to deleted and full-size mtDNA molecules must be simultaneously quantified. When the percentage of the deletion is lowest, the difference in band intensity is the highest, which provides the least accurate measurement. Indeed, when the deletion percentage is low, the signal for the deleted mtDNA is weak and is often more difficult to distinguish from the background signal than is the signal for full-size mtDNA. In contrast, with q-PCR, the undeleted and deleted sequences are assayed separately and quantified accurately with calibration curves, leading to higher precision.
Our study of patient 1 revealed a discrepancy in assaying the deletion because of a polymorphic site. This observation underscores the fact that at least two different primer pairs should be used outside and inside the deleted region to properly quantify heteroplasmic deletions. Furthermore, if a single primer pair is used inside the deleted region, as proposed previously for the ND4 gene (25), a deleted mtDNA region could be missed. This is the case for patient 5 with deletion of nt 12439–15390, which encompass the ND5, ND6, and part of the cytochrome b genes.
quantification of MTDNA depletion or overreplication
Southern blotting and q-PCR were used to determine the mtDNA copy number in different tissues from patients either with mtDNA depletion or overreplication. The mtDNA copy number was normalized to the nuclear ATPsynß gene copy number to take into account any variation of total cellular DNA in the samples. The results obtained by the two methods agreed for both muscle and liver from the patient with mtDNA depletion, but some discrepancy was observed in fibroblasts. We have previously performed such a quantification in two patients with specific liver mtDNA depletion, and the two methods led to similar values (36). Our present results show that CVs of 20% for q-PCR and 37% for Southern blot could not be avoided in age-matched controls, regardless of the tissue studied. This variation is a consequence of the variability in mtDNA copy number among individuals, even in biopsies from the same tissue. The much lower CV for real-time q-PCR confirms the accuracy of this method.
We also studied two patients with increased mtDNA copy numbers (mtDNA overreplication) associated with severe respiratory chain enzymatic defects. These diseases, characterized only biochemically, are related to an abnormally large increase in the amount of mtDNA, which is unable to provide normal respiratory chain enzymatic activities. The mtDNA copy number determined by real-time q-PCR showed a more than fourfold increase in mtDNA copy number compared with controls (for patient 12). These increased values are up to twofold higher than those obtained by Southern-blot analysis. This discrepancy is attributable to two aspects of the Southern-blot method that could lead to underestimation of the mtDNA-to-nDNA ratio. The first aspect is that the efficiency of DNA transfer to the membrane is dependent on DNA size and concentration: when high concentrations of mtDNA are blotted for mtDNA overreplication analysis, the transfer may not be as complete as for the smaller nDNA fragments at much lower concentrations, and the blotting membrane could become saturated. The second aspect is that for Southern blotting performed for patients with mtDNA overreplication, nDNA and mtDNA were quantified by simultaneous hybridization of the blot with nuclear and mitochondrial probes. The mtDNA radioactive signal was so strong that it generated a background signal that contaminated and overestimated the nDNA signal. The mtDNA/nDNA ratio was thus lower in the Southern-blot analysis than in q-PCR. For q-PCR, however, nDNA and mtDNA are amplified separately with specific primers allowing precise quantification of each DNA species.
In conclusion, the real-time q-PCR method we described is a faster and more accurate tool than Southern blotting for quantification of mtDNA alterations, including deletions. This method could be used to improve pre- and postnatal molecular diagnosis linked to qualitative or quantitative mtDNA defects. Moreover, these PCR conditions could be also used to quantify mtDNA random point mutations in muscle from elderly individuals. Such a study could help to determine whether mtDNA mutations are a major factor in aging.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
