| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Technical Briefs |
1
Torrey Mesa Research Institute, 3115 Merryfield Row, La Jolla, CA 92121
2
Department of Neurosciences, University of California San Diego and the UCSD Mitochondrial Disease Laboratory, CTF C103, 214 Dickinson Street, San Diego, CA 92103-8467
aauthor for correspondence: fax 858-812-1097, e-mail andras.guttman{at}syngenta.com
Gel-microchip electrophoresis is a novel combination of the well-established methods of slab-gel electrophoresis and capillary-gel electrophoresis (1). The gel-microchip format provides a multilane separation platform (a plurality of virtual channels up to 96 lanes) with excellent heat dissipation characteristics, allowing application of high voltages necessary to obtain rapid and high-performance analysis of DNA fragments. The system readily accommodates fluorescent labeling during the electrophoresis separation process (in migratio), such as intercalation with ethidium bromide or complexation with other novel, high-sensitivity DNA staining dyes, in addition to the use of conventional covalently labeled primers (i.e., before the separation process). Sample injection onto the gel microchip is accomplished by membrane-mediated loading technology (2), which also enables robotic spotting of multiple samples. The method is readily applicable for large-scale, automated, high-throughput mutation screening (3).
Mutations in mitochondrial DNA (mtDNA) (4) are recognized as important in diseases ranging from ischemic heart disease to neurodegenerative disorders (5) and diabetes (6)(7)(8)(9)(10). mtDNA mutations accumulate with age and may play an important role in aging (11).
The association of diabetes with the mtDNA tRNA Leu (UUR) A3243G point mutation responsible for the syndrome of mitochondrial encephalomyopathy with ragged red fibers and stroke-like episodes (MELAS) was reported in 1992 in a family with maternally inherited deafness and type II diabetes (7). Differences in the prevalence of the MELAS mutation among ethnic groups (9)(10)(12) highlighted the need for accurate, high-sensitivity screening for the MELAS 3243 and other tRNA mutations. Most studies searching for the 3243 mutation in diabetes have focused on non-insulin-dependent diabetes mellitus; however, many MELAS patients are insulin dependent (9)(10), a fact that suggests that all types of diabetic patients, and particularly type II patients, should be tested if the true incidence of the MELAS mutation in the diabetic population is to be determined, whether or not they are insulin dependent. Patients with mitochondrial diabetes are at risk for complications of mitochondrial disease, including cardiomyopathy, stroke, and nephropathy. The diagnosis of a maternally inherited disorder is important for genetic counseling. Surveys of the incidence of mtDNA mutations in diabetes may have missed low-percentage heteroplasmy (13) because of improper analysis methods.
We sought to evaluate a new and innovative electric field-mediated microseparation technique with the use of ultra-sensitive, real-time fluorescent detection for high- throughput detection of low-percentage mtDNA mutations.
Blood samples (3 mL) were used from controls and patients (M1–3) with suspected or diagnosed mitochondrial disease. We lysed red cells in ammonium chloride (PureGene), collected the nucleated peripheral blood leukocytes and platelets, deproteinized them in high salt, and selectively precipitated chromosomal DNA with 0.6 volumes of isopropanol. We then precipitated mtDNA in 1 volume of isopropanol at -80 °C overnight, washing in 700 mL/L ethanol, drying, and resuspension in 200 µL of Tris-EDTA (10 mmol/L Tris, 1 mmol/L EDTA, pH 7.4). Procedures regarding use of human subjects were in accordance with the current revision of the Helsinki Declaration.
We designed the primers (27mers) for PCR to produce a 501-bp amplified fragment including the ApaI restriction site produced by the MELAS mutation, near the middle of the amplicon. The sense and antisense primers were nucleotide positions 2993–3019 and nucleotide positions 3467–3493 (Sigma GenoSys). PCR amplification was carried out in a PTC-200 DNA Engine thermal cycler (MJ Research), using 2.5 ng of mtDNA-rich solution in 25 µL. The PCR mixture included 10 mM Tris, pH 8.4, 2.5 mM MgCl2, 50 mM KCl, 0.20 mM each dNTP, 2.5 U of Taq polymerase (Life Technologies), and 0.4 pmol of each primer. We used 15 PCR cycles of 45 s at 94 °C, 30 s at 55 °C, and 75 s at 72 °C. The final cycle was stopped by rapid cooling to room temperature to avoid heteroduplex formation. Amplified products were digested by ApaI (New England Biolabs) overnight at 37 °C. We spotted 0.5 µL of each reaction product onto the loading membrane tabs (The Gel Company) and analyzed by gel-microchip electrophoresis with ethidium bromide staining.
Low electroendosmosis (-mr = 0.06) agarose III (Amresco) was dissolved in 45 mmol/L Tris, 45 mmol/L borate, 1 mmol/L EDTA buffer, pH 8.3 [0.5x Tris-borate-EDTA buffer (TBE)]. We obtained electrophoresis-grade Tris, boric acid, EDTA, and ethidium bromide from ICN. Linear polyacrylamide [(LPA); Mr 700 000–1 000 000] was from Sigma. All buffer and sample solutions were filtered through a 0.2 µm nylon membrane syringe filter (Fisher Scientific).
We suspended 1 g of agarose powder in 50 mL of 1x TBE buffer and boiled the mixture repetitively in a microwave oven until clear. We then added 50 mL of 40 mL/L LPA in water and 100 µL of 150 nmol/L ethidium bromide, and the mixture was kept at 60 °C for 15 min. A preheated (35–45 °C) float glass microchip cassette (14) was filled with 2 mL of the melted composite agarose–LPA gel. After several minutes of cooling and solidification, the gel microchip was ready for use. Samples were injected by membrane-mediated loading. The injection end of the gel microchip had a straight edge with no individual wells. We spotted 0.5 µL of samples onto the tips of the membrane loader and inserted the tips in close proximity to the straight edge of the gel. The migrating DNA fragments were imaged in real time by the scanning laser-induced fluorescence detector. The effective separation length of the gel microchip was 60–65 mm. The applied electric field strength was 75 V/cm, generating 7–9 mA. Integrated relative fluorescence units were computed by ImageQuant Software (Molecular Dynamics) and by Caesar Workstation (SciBridge Software).
The principle of mutation analysis derived from PCR–restriction fragment length polymorphism (RFLP) is depicted in Fig. 1A
, showing that the ApaI restriction enzyme cuts the mutant site, but not the wild-type DNA. Generating digestion products that are similar in size (250 and 251 bp) effectively provides a twofold improvement in sensitivity. Each sample was run in triplicate using one lane for the restriction digest and an adjacent control lane for the uncut DNA. When the analysis was complete, the used gel was replaced by pumping fresh melted agarose-based gel composition into the microchip.
|
The method could detect <1% pathogenic A3243G mtDNA point mutation in heteroplasmic blood samples. With ethidium bromide staining we could routinely detect as little as 1–2% mutation in a mixture with 98–99% wild type. Sensitivity and dynamic range were explored with clonal mixtures and with patient samples in which the percentage of the A3243G mutation was quantified by radioisotopic end-cycle labeling (13). Mixtures of mtDNA clones of wild type and the A3243G mutant (percentages of mutant DNA, 100%, 75%, 50%, 25%, 15%, 10%, 5%, 2%, and 1%) were loaded (
1 ng of digested PCR product) and analyzed on the gel microchip in a multilane format (Fig. 1B
). The amount of DNA was quantified by spectrophotometric quantification of PCR products (at 260 nm) after column purification to remove the primers. The amount of DNA was also quantified by comparing band intensity to a known standard of the 50-bp ladder. The 501-bp fragment was the uncut target DNA, and the 250- and 251-bp fragments were the products of the ApaI restriction digestion. Separation and visualization required <15 min. The detection signal corresponded to the amount of DNA. The limit of detection of the integrated laser induced fluorescence–avalanche photodiode system was also demonstrated by revealing 0.5% mutation, corresponding to 5 pg of DNA (not shown).
Blood-based mtDNA analysis of three mothers (M1, M2, and M3) of patients with clinical MELAS is depicted in Fig. 1C
. Traces A and B depict analyses of sample M1 without and with ApaI digestion, revealing 1.0% mutation burden. Similarly, traces C and E and D and F (Fig. 1C
) show the analysis of the undigested and digested samples of patients M2 and M3, respectively. In these, 5.2% (Fig. 1C
, trace D) and 5.4% (Fig. 1C
, trace F) mutation burdens were found. The results were validated by quantitative radioactive PCR (13). Although these individuals had low-percentage heteroplasmy in the blood, they were still at risk from the complications of mitochondrial disease from possibly higher percentages of mutation in other tissues. Samples M2 and M3 were from patients with diabetes.
In summary, these studies suggest that the PCR-RFLP-based diagnostic test in conjunction with rapid gel-microchip electrophoresis can greatly facilitate large-scale analysis of mtDNA heteroplasmy of the specific A3243G point mutation.
Acknowledgments
This work was supported by Enterprise Partners; Syngenta, Genomics Research and Technology, the UCSD Mitochondrial and Metabolic Disease Center, and, in part, by NIH Grant M01 RR00827.
References
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
