| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Articles |
a Address correspondence to this author at: Center for Human Genetics, Gasthuisberg O&N6, Herestr. 49, B-3000 Leuven, Belgium. Fax 32-16-345997; e-mail gert.matthijs{at}med.kuleuven.ac.be
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Key Words: indexing terms: inherited diseases • gene defects • point mutations • polymerase chain reaction • heteroplasmy
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The mitochondrial genome has some unique genetic features. First, a human cell contains on average a few hundreds of mitochondria and a thousand copies of mtDNA. In most pathological cases both normal and mutant mtDNA occur in the same cell, a phenomenon called heteroplasmy. Second, because the mtDNA codes for proteins with a role in oxidative phosphorylation, organs that rely more readily on their energy supply will be more sensitive to mtDNA defects. Finally, mtDNA is transmitted through the cytoplasm of the egg, and the inheritance of the mutations is strictly maternal. The identification of a mutation thus has implications for all maternally related members of a family.
LHON (OMIM 535000, (9)) is the most frequent cause of optic atrophy, and a good diagnostic test would supplement the clinical investigations. Currently, the assay of choice is one in which one can screen for several point mutations simultaneously, heteroplasmy can be detected, several samples can be processed in parallel, and is technically easy to handle.
Until now, LHON mutation detection has mainly been based on the gain or loss of restriction sites (5), or on allele-specific amplification (10). These methods are accurate and deal with the problem of heteroplasmy but are time consuming. Single-stranded conformation polymorphism analysis has also been used for screening of the mitochondrial genome (11). These methods have the potential to identify any mutation in a particular region of the mtDNA, although with only 60–80% efficiency (11). However, given the number of polymorphisms in the mitochondrial genome, false positives may occur often and direct sequencing becomes a requisite. Solid-phase microsequencing has thus far been the most sensitive direct assay (12). Although this method is very sensitive and amenable to automation and microchip technology (13), it is relatively time consuming in its present format.
We chose to develop a reverse dot blot (RDB) as a practical solution to mtDNA point mutation screening. The method, based on the discriminative hybridization of single-base mismatches, is simple and fast, while the range of mutations can easily be expanded. The format is suitable for commercialization.
The five most frequent, primary mutations responsible for LHON (3460, 9804, 11778, 14484, 15257 [9]) are included and the LHON assay was tested on 47 consecutive patients; for 11 patients the clinical diagnosis was confirmed.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
) was done in 50 µL of 1x buffer (Gene Amp, 10x PCR-Buffer
II; Perkin-Elmer, Norwalk, CT and Roche Molecular Systems, Branchburg,
NJ) containing 2 mmol/L MgCl2, 0.2 mmol/L dNTPs, and
2U of AmpliTaq polymerase (Perkin-Elmer). Primers were used at the
concentrations described in Table 1
. One primer of each pair was
biotinylated. The oligonucleotides were purchased from Pharmacia
Biotech (Roosendaal, The Netherlands) or Eurogentec (Seraing, Belgium).
The reactions were performed on 300 ng of genomic DNA in a GeneAmp 2400
(Perkin-Elmer) with 30 cycles of denaturation for 30 s at
94 °C, annealing for 30 s at 57 °C, extension for 30 s
at 68 °C, and a final extension of 10 min at 68 °C. After
amplification, 5 µL of the PCR mix was analyzed on a 2% agarose gel.
|
preparation of the membrane
Preparation of the Biodyne C membrane (Pall BioSupport, Portmouth,
NJ) was essentially done as described by Zhang et al.
(14). The membranes were rinsed in 0.1 mol/L HCl,
dried briefly, and activated in 200 g/L
1-ethyl-3-(3-dimethylaminopropyl)carbodiimide hydrochloride (EDAC) for
15 min. Then the filters were rinsed twice in H2O and dried
at room temperature for 10 min. Synthetic oligonucleotides with a 5'
amino linker (RDB oligonucleotides) were covalently bound to the
activated carboxyl groups of the membrane. These oligonucleotides were
purchased from Pharmacia Biotech (TFA-amino linker) or Eurogentec (C12
amino modifier). The oligonucleotides were diluted in 0.5 mol/L
NaHCO3, pH 8.4 (see Table 2
), and 2 µL of this dilution was spotted on the membrane.
After 10 min of incubation, the remaining active groups were quenched
with 0.1 mol/L NaOH for 10 min. Finally, the membranes were rinsed
twice in H2O and air dried for immediate use or for
storage.
|
hybridization and detection
The membranes were prehybridized in 5x saline–sodium
phosphate-EDTA (SSPE) buffer (20x SSPE is 3 mol/L NaCl, 200 mmol/L
NaH2PO4, and 20 mmol/L EDTA at pH 7.4) and 5
g/L sodium dodecyl sulfate (SDS) at 50 °C for 30 min. Forty
microliters of multiplex PCR product was denatured with 1.6 µL of 5
mol/L NaOH for 5 min at 65 °C and diluted in 2 mL of hybridization
solution (5x SSPE, 5 g/L SDS), after which the strips were hybridized
for 30 min at 50 °C. The membranes were washed in 3 mol/L
tetramethylammonium chloride (TMAC) at 50 °C for 15 min
(14). Since detection was performed with a
chemiluminescent reaction kit (Chemiluminescent kit; Tropix, Bedford,
MA), the PCR primer of the strand complementary to the RDB
oligonucleotide was biotinylated. The reaction was essentially done
according to the manufacturers' protocol. The filters were first
blocked with I-light blocking reagent and then incubated with
streptavidin–alkaline phosphatase conjugate. After the washing steps,
the membranes were incubated with the AMPPD
[3-(2'-spiroadamantane)-4-methoxy-4-(3'-phosphoryloxylphenyl-1,2-dioxetane]
substrate, activated at 42 °C for 10 min, and exposed to
ECL-hyperfilm (Amersham, Bucks, UK) for 5 to 10 min.
test samples
For optimization, short wild-type (WT) and mutant-type (MT)
fragments were generated for each mutation by PCR with the RDB
oligonucleotide that detects the mutation of interest as an
amplification primer together with the biotinylated PCR primer normally
used for amplification of the fragment (Table 1
)
(15)(16).
Total DNA samples were obtained from lymphocytes by using a salt extraction method. Samples from patients with the 3460, 11778, 14484, and 15257 mutations were a gift from J.P. Bonnefont, Hôpital Necker, Paris. In the course of the experiments, patients with these mutations and the 9804 mutation were identified (see Results).
At the time of analysis, the laboratory had also collected 47 samples from patients with optic atrophy, with either a definitive or a tentative clinical diagnosis of LHON. These samples had been collected after informed consent.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
RDB
. The RDB principle (15)(23)
was chosen as a useful method for detection of mtDNA point mutations.
The assay had to fulfill three criteria: (a) a stringent
hybridization to discriminate between oligonucleotides differing in
only one base pair, (b) the same hybridization conditions
for all WT/MT oligonucleotide pairs, and (c) a high
sensitivity to detect a low percentage of heteroplasmy.
Instead of the classical poly-T tailing (15)(23), covalent attachment of the oligonucleotides to the membrane via an amino linker was preferred. As reported by Zhang et al. (14), covalent linkage provides an increased sensitivity of the assay compared with the T-tail-based attachment. The use of a spacer between the amino group and the oligonucleotide had no influence on the binding efficiency to the membranes (data not shown).
To simplify the optimization of the hybridizations, TMAC was chosen for the posthybridization washes. This quaternary ammonium salt essentially eliminates disparities in the melting temperatures (Tms) of the oligonucleotides (24). Indeed, a better discrimination was obtained after washing with 3 mol/L TMAC compared with 0.5x saline–sodium citrate (SSC) (20x SSC is 3 mol/L NaCl, 0.3 mol/L sodium citrate at pH 7.0) and 1 g/L SDS (data not shown).
optimization of rdb
On the basis of the idea that 3 mol/L TMAC in the hybridization
reactions would eliminate the dependence of the
Tm on G-C content, all RDB oligonucleotides were
initially 18 bp long. At 50 °C, a specific hybridization was
achieved with the WT and MT primers for 3460 and 9804. By varying the
length and shifting the position of the oligonucleotides, more
mutations were detected under uniform hybridization conditions. In our
hands, TMAC was thus not as good a reagent as proposed. Also, the short
synthetic mutant fragments were an easy tool for optimization of the
RDB, although in certain cases the specific signal that was obtained
with the short fragments was not obtained with the full-size fragments
(Fig. 1
and data not shown).
|
The concentration of the oligonucleotides is a second parameter that
influences the discrimination between MT and WT. This has been tested
in the same set of experiments. The oligonucleotides were applied to
the membrane in dilution series ranging from 10 to 0.25 pmol (Fig. 1
).
WT oligonucleotide concentrations, which gave a strong signal when
hybridized with WT probe and no signal with the MT probe, were chosen
and vice versa. The sequence and concentration of the oligonucleotides
used in the final assay are given in Table 2
.
All target regions, i.e., five fragments of 210 to 833 bp, encompassing
the different mutations, were amplified in one multiplex PCR (Fig. 2
). The fragment lengths and primer sets are given in Table 1
.
|
screening for lhon
The LHON strip was tested on 47 samples. Several outcomes were
assembled in Fig. 3
. For 11 patients, the diagnosis was confirmed: Five showed the
mutation at 11778 (an example is given in lane 4), three the
mutation at 14484 (lane 5), one had the 3460 mutation (lane 2), one the
rare 9804 mutation (lane 3), and one was positive for 15257 (lane 6).
All these mutations were virtually homoplasmic. Fig. 3
shows an
additional feature of the RDB assay: A polymorphism in the region
covered by the RDB oligonucleotide results in the absence of a signal
for both the WT and MT oligonucleotides (lane 7). Sequencing showed the
presence of a polymorphism at position 3459 in this patient.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Good results were obtained with the covalent attachment of the amino-linked oligonucleotides to the negatively charged membranes. Compared with T-tailed oligonucleotides they are easier to use, the covalent binding is more reproducible, and a higher sensitivity can be achieved (14). TMAC was used in the washing buffer. However, the exact conditions for a dependence of the Tm purely on length were not achieved. Less than half (4 of 10) of the original oligonucleotides of 18 bp were retained in the final assay. For the remaining RDB oligonucleotides, both length and position still had to be empirically determined.
The detection criteria for LHON mutations are less stringent than for mtDNA mutations associated with mitochondrial encephalo(myo)pathies, since LHON mutations are almost always homoplasmic in affected individuals. The RDB assay easily detects mutations in two-allelic systems (15)(23). Moreover, control experiments have shown that low percentages of heteroplasmy (<1%) are readily detected (unpublished data).
Obermaier-Kusser et al. (18) have described 12 different combinations of mutations in 21 of 29 LHON patients or families (72%). With this assay, we would have detected 81% of their confirmed LHON patients, missing those patients that harbor only secondary mutations. It would therefore be interesting to add the 4216 and 13708 mutations to the assay. On the other hand, the low number of positives in the present screening is due to the fact that the samples were tested as they presented in the diagnostic laboratory, without stringent clinical review of the symptomatology.
The sample included five unrelated cases of 11778. These had previously been identified by SfaNI digestion. The assay detected four different mutations in six other, unrelated patients. Thus, the 11778 mutation is responsible for almost 50% of LHON cases, as was previously noted. It is interesting to note that a patient with the mutation at 9804 was identified. Only three patients with this mutation were described thus far (21).
In conclusion, this RDB assay covers the most frequently observed mutations and greatly facilitates the molecular diagnosis of LHON.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
1 Nonstandard abbreviations: mtDNA, mitochondrial DNA;
LHON, Leber hereditary optic neuropathy; RDB, reverse dot blot; SSPE,
saline–sodium phosphate–EDTA; SDS, sodium dodecyl sulfate; TMAC,
tetramethylammonium chloride; SSC, saline–sodium citrate; WT, wild
type; and MT, mutant type. 
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The following articles in journals at HighWire Press have cited this article:
|
|
 |
|
  S. Lappin, J. Cahlik, and B. Gold Robot Printing of Reverse Dot Blot Arrays for Human Mutation Detection J. Mol. Diagn., November 1, 2001; 3(4): 178 - 188. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L.-J. C. Wong and D. Senadheera Direct detection of multiple point mutations in mitochondrial DNA Clin. Chem., October 1, 1997; 43(10): 1857 - 1861. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
