Accelerated prenatal diagnosis of fragile X syndrome by polymerase chain reaction restriction fragment detectionкод для вставкиСкачать
American Journal of Medical Genetics 83:338–341 (1999) Accelerated Prenatal Diagnosis of Fragile X Syndrome by Polymerase Chain Reaction Restriction Fragment Detection Carl Dobkin,1* Xiao-hua Ding,1 Shu-yun Li,1,2 George Houck Jr.,1 Sarah L. Nolin,1 Anne Glicksman,1 Nan Zhong,1 Edmund C. Jenkins,2 and W. Ted Brown1 1 Department of Human Genetics, New York State Institute for Basic Research in Developmental Disabilities, Staten Island, New York 2 Department of Cytogenetics, New York State Institute for Basic Research in Developmental Disabilities, Staten Island, New York Prenatal diagnosis of fragile X syndrome requires detection of the full FMR1 mutation in chorionic villus or amniotic fluid cell samples. Although analysis of genomic DNA restriction fragment pattern is a highly reliable technique for identification of the full FMR1 mutation, standard Southern blot determination of this pattern requires significantly more genomic DNA than is initially available from a prenatal sample. To overcome this limitation we developed a method that determines the diagnostic pattern of genomic restriction fragments from a fraction of a prenatal specimen. The prenatal DNA sample is first digested with EcoRI and EagI, and after agarose gel electrophoresis, the 2- to 10-kb region of the gel is serially sectioned and amplified by polymerase chain reaction. Analysis of prenatal samples from an unaffected male and from a full mutation male showed that this approach generated a diagnostic pattern comparable with a Southern blot of 100-fold more material. This innovation enables laboratories to prenatally diagnose the full FMR1 mutation sooner than standard techniques. Am. J. Med. Genet. 83:338–341, 1999. © 1999 Wiley-Liss, Inc. KEY WORDS: fragile X syndrome; prenatal diagnosis; mental retardation Contract grant sponsor: the New York State Office of Mental Retardation and Developmental Disabilities; Contract grant sponsor: the Maternal and Child Health Program, Department of Health and Human Services; Contract grant number: MCJ360587. *Correspondence to: Carl Dobkin, Ph.D., NYS Institute for Basic Research, 1050 Forest Hill Road, Staten Island, NY 10314. Received 22 August 1997; Accepted 28 April 1998 © 1999 Wiley-Liss, Inc. INTRODUCTION The fragile X syndrome, with rare exception, is caused by expansion of a CGG triplet repeat in the 5⬘ untranslated region of the FMR1 gene [Verkerk et al., 1991; Fu et al., 1991; Eichler et al., 1993; Warren and Nelson, 1994]. FMR1 triplet repeat alleles can be separated into three size classes based on stability. Alleles with approximately 6 to 50 triplets (normal alleles) are stable both in somatic development and heritable transmission. Alleles with approximately 60 to 200 (premutation alleles) are functional and stable in somatic development but unstable upon transmission. Borderline, ‘‘grey zone,’’ alleles with 50 to 60 triplets are stable in somatic development, yet their stability in transmission in unpredictable . Alleles with >200 triplets (full-mutation alleles) are highly unstable during somatic development and in inheritance. Expansion to >200 triplet repeats is associated with methylation of a CpG island at the 5⬘ end of the FMR1 gene that prevents transcription and leads to the fragile X syndrome. Some individuals with the full mutation have a premutation (60 to 200 repeat) allele in some of their somatic cells, and they are considered ‘‘mosaics’’ [Rousseau et al., 1991]. For reasons not yet understood the full mutation is only inherited in maternal transmissions. Thus, prenatal fragile X syndrome diagnosis is only appropriate for female carriers of the pre- or full mutation. The diagnostic hallmarks of a full mutation are: 1) >200 repeats; 2) the presence of a heterogeneous array of repeat alleles in somatic cells; and 3) methylation of the 5⬘ CpG island including sites for methylation-sensitive restriction endonucleases such as EagI, BssHII, and NruI. This methylation occurs in very early development but the methylation in extra-embryonic tissue (e.g., chorionic villi) is significantly delayed [Sanford et al., 1985] and is often incomplete at the time of prenatal sampling. Diagnosis of fragile X syndrome depends primarily on the determination of the triplet repeat length in FMR1. Polymerase chain reaction (PCR) analysis of Accelerated Prenatal Fragile X Diagnosis 339 allele size is effective, accurate, and reliable for most FMR1 alleles [Brown et al., 1993]. However, some alleles with very large repeats and mosaics are difficult to characterize by PCR because of limitations intrinsic to this technique. Analysis of genomic DNA restriction fragments generated by digestion with a pair of endonucleases that includes one sensitive to methylation [Rousseau et al., 1991] reliably resolves these ambiguities. For example, digestion of genomic DNA with EcoRI (insensitive) and EagI (sensitive) generates a 2.8-kb EcoRI-EagI fragment from a normal unmethylated allele and a 5.2-kb EcoRI fragment from a normal methylated allele (representing an inactive X chromosome). Premutation alleles are detected as 3.0- to 3.4kb EcoRI-EagI fragments, whereas full-mutation alleles, which are consistently methylated and thus, not cleaved by EagI, are detected as EcoRI fragments >6 kb in size. Diagnostic analysis of genomic restriction fragments is currently done by Southern blotting. Unlike PCR, reliable genomic Southern blotting generally requires DNA from 艌106 cells. In early prenatal diagnosis, however, the initial tissue sample is usually too small to provide sufficient DNA for Southern blot analysis. Thus, in those cases in which PCR analysis is not conclusive, there is often a delay of 2–4 weeks while the prenatal sample is expanded in culture to supply adequate DNA for genomic Southern analysis. To circumvent this delay we have developed a 5- to 6-day procedure that uses PCR to detect the genomic restriction fragment pattern in the DNA from 艋50,000 cells. This number of cells is usually available for study within a few days of either amniocentesis or chorionic villus sampling. A total of 92 l was added to each tube for a final PCR volume of approximately 100 l containing 10 mM Tris-HCl, pH 8.3, 50 mM KCl, 200 M each dNTP, 2.5 U of Amplitaq polymerase (Perkin Elmer, Foster City, CA), 4 mM MgCl2, 1 M 144F (CGCTAGCAGGGCTGAAGAGAAGATG), and 1 M 632R (CTCCTCCACAACTACCCACACGAC). The primers defined a 489bp segment that began 244 nucleotides ‘‘downstream’’ from the transcription start site [Fu et al., 1991] and did not include the CGG repeat. The PCR conditions were: 94°C for 2 min; 23 cycles of 94°C for 1 min, 60°C for 1 min, 72°C for 2 min; and 72°C for 10 min. These conditions had been adjusted for the presence of the agarose gel section by increasing the Mg2+ to 4 mM. PCR products (15 l) were electrophoresed on a 2% (Nusieve 3:1, FMC) agarose gel in TAE and fractionated at 23 V/cm for 2 to 3 hr in an 11 × 14-cm horizontal gel apparatus (Gibco BRL Life Technologies, Grand Island, NY). The gel was blotted to Zeta-Probe GT membrane with 0.5 N of NaOH according to the directions of the manufacturer (BioRad), hybridized in 3× SSC at 65°C overnight with a 144F-632R PCR-probe amplified from a cloned template, pE5.1 [Fu et al., 1991], and ␣dCTP labeled with and oligolabeling kit (Amersham Pharmacia, Biotech, Piscataway, NJ). Blots were washed at 0.1× SSC at 60°C and autoradiographed for 3 to 18 hr. The DNA used as probe was synthesized as described above except that the MgCl2 was 2.5 mM, and the PCR amplification was for 30 cycles. Standard EcoRI-EagI Southern blot analyses were done with approximately 12 g of genomic DNA using previously described methods [Rousseau et al., 1991; Nolin et al., 1994]. MATERIALS AND METHODS RESULTS After amniotic fluid or chorionic villus cells were briefly cultured (2–4 days), cells from a sparsely colonized T25 flask were counted and 25,000 to 50,000 cells (approximately 20%) were harvested. DNA was isolated with a Puregene kit (Gentra Systems, Minneapolis, MN) with 300 l of lysis buffer, precipitated with freshly prepared spermine [Hoopes and McClure, 1981], redissolved in 40 l of H2O to a concentration of approximately 4 ng/l, and digested with 100 U of EcoRI and 50 U of EagI according to the directions of the supplier (New England Biolabs, Beverly, MA) in a total volume of 300 l for 18 hr at 37°C [Rousseau et al., 1991]. Digested DNA was fractionated on a 2-mm thick 1% low-gelling temperature agarose gel (NuSieve, FMC, Rockland, ME) in 40 mM Tris-acetate, 1 mM EDTA (TAE) [Sambrook et al., 1989] at 0.6 V/cm for 20 hr with 500 ng -HindIII size marker in flanking lanes. The gel was photographed, and 16–21 serial sections of the 2to 9-kb region were excised with coffee stirrer straws (approximately 3 mm in diameter) and placed into 200 l thin-walled PCR ‘‘strip’’ tubes. ‘‘Negative control’’ sections of the gel were excised outside the sample lanes and analyzed in parallel. ‘‘Positive control’’ samples were identical to negative controls except that they were spiked with an additional 100 ng of genomic DNA from an unaffected male. The procedure we developed to accelerate analysis of fragile X syndrome is shown schematically in Fig. 1. The product of 23 cycles of PCR was detected in the analytical gel by Southern blotting. The serial sections that contained the diagnostic restriction fragments yielded a PCR product and thus the PCR results indicated the pattern of the otherwise undetectable EcoRIEagI genomic fragments in the original gel. The PCR product did not include the CGG repeat and thus avoided the difficulties inherent in amplifying that region. Fig. 2 shows the application of this procedure to a prenatal sample of 艋5 × 104 cells from an unaffected male fetus as compared with standard Southern blot analysis of DNA derived from >2 × 106 cells expanded from the same sample. The normal 2.8-kb EcoRI-EagI fragment was detected similarly in both the PCR analysis and the standard Southern blot. In this particular 23-cycle PCR analysis the 2.8-kb fragment was detected primarily in serial section 18, although a small amount was also found in sections 15–17 and 19–21. Increased number of PCR amplification cycles, however, yielded a uniform amount of PCR product from all serial sections tested, whereas sections from regions outside of the sample lane did not yield a product (data not shown). This procedure was applied to an amniocentesis 340 Dobkin et al. Fig. 1. PCR detection procedure. The location of FMR1 restriction fragments in the preparative gel is indicated by the serial sections that yield a PCR product. In the box on the right, the lines representing the diagnostic restriction fragments show the positions of the EcoRI and EagI sites, the CGG repeat (filled circles), and the PCR primers (horizontal arrows). The PCR product does not include the CGG repeat. sample of 艋5 × 104 cells from a full mutation male fetus. Figure 3 compares the PCR analysis to a standard Southern blot analysis of DNA derived from >2 × 106 cells expanded from the same sample. The fullmutation allele (approximately 8 kb) was detected primarily in section 3 and a small amount was also found in sections 1 and 2. The standard Southern blot also showed a majority of the fragments as a full mutation band of approximately 8 kb. In addition the standard Southern blot showed a minor full mutation band at approximately 6 kb and a premutation band at approximately 3 kb. These minor bands were not seen in the PCR detection analysis. DISCUSSION Standard Southern blot analysis of the FMR1 mutation is an extremely reliable technique that is particularly effective as a secondary screen subordinate to PCR analysis for fragile X syndrome diagnosis. However, it is often inappropriate for prenatal analysis where the tissue sample is limited and the speed of the analysis is critical. To accelerate the analysis of the genomic restriction fragment pattern we have developed a procedure that applies PCR amplification of isolated restriction fragments to increase sensitivity of FMR1 mutation detection at least 100-fold compared with standard Southern analysis. PCR amplification of the diagnostic EcoRI-EagI restriction fragments in the agarose gel reveals their location and provides a sen- Fig. 2. Unaffected male. Detection of 2.8-kb EcoRI-EagI restriction fragment from DNA extracted from a chorionic villus sample. (a) Standard Southern blot detection (艌2 × 106 cells, 12 g DNA) with probe StB12.3 [Rousseau et al., 1991]. (b) PCR detection (艋5 × 104 cells, 艋0.3 g DNA). Twenty-one serial sections of the preparative gel were amplified by PCR with primers 3⬘ to the CGG repeat. The 489-bp 144F-632R PCR product was detected after agarose gel fractionation and Southern blotting. Positive control (+) shows product from 0.1 g of added genomic DNA template. Negative control (−) shows product from a ‘‘control’’ agarose section taken from another region of the gel. The positions of -HindIII DNA size markers, 9.4, 6.5, 4.3, and 2.2 kb are indicated. sitive representation of the restriction fragment pattern seen in a standard Southern blot. The PCR amplification was directed to a specific sequence within the diagnostic fragments that reflects the methylation pattern of the DNA and did not include the CGG triplet repeat. When 艌30 cycles of PCR amplification were used, however, we detected this specific sequence in all of the serial gel sections (data not shown). This indicated a low level of this specific sequence in all of the serial gel sections presumably originating from partial digestion products, degraded DNA, diffusion, and interactions with the gel matrix. To avoid this problem, the PCR amplification was limited to the exponential phase where it was semiquantitative and only detected amplification from the most abundant restriction fragments. In practice this limited the number of PCR amplification cycles that could be used for the serial sections. The procedure was also sensitive to the sample size and yielded consistent results only when the num- Accelerated Prenatal Fragile X Diagnosis 341 blot probably exaggerated the level of the premutation allele. The described procedure can significantly accelerate the resolution of prenatal diagnoses that are ambiguous by PCR allele size determination. It is particularly useful for amniocentesis samples because of the time constraints placed on these analyses. We have incorporated this procedure into our prenatal analysis protocol. Although it is more costly and technically demanding than a standard Southern blot, this method can reduce the time required for complete prenatal FMR1 mutation analysis to as short as 1 week. ACKNOWLEDGMENTS We thank the families who participated in this study. This study was supported in part by funds from the New York State Office of Mental Retardation and Developmental Disabilities and in part by grant MCJ360587 (to E.C. Jenkins) from the Maternal and Child Health Program, Department of Health and Human Services. REFERENCES Fig. 3. Full-mutation male. Detection of 7- to 9-kb EcoRI-EagI restriction fragments from amniocyte DNA. (a) Standard Southern blot detection (艌2 × 106 cells, 12 g) with probe StB12.3 [Rousseau et al., 1991]. (b) PCR detection (5 × 104 cells, 艋0.3 g). Sixteen serial section of the preparative gel were amplified by PCR and analyzed by Southern blotting (as in Fig. 2). The positions of -HindIII DNA size markers are indicated. ber of cells analyzed was 20 to 50,000. It is likely that a much smaller sample (5 × 103 cells) would be effective for diagnosis if the number of amplification cycles were increased to 25 or 26. However, a smaller amount of DNA template would increase the risks posed by contamination and might not specifically represent the fetus. It is also possible that a fraction (<0.5 mg) of the original sample could be used directly but we have not attempted this because of the difficulty in determining the exact cell number in an original tissue sample. There were differences between the patterns of EcoRI-EagI restriction fragments detected by PCR and the standard Southern blot analyses. For example, the premutation band apparent in the Southern blot in Fig. 3a was not detected in the PCR analysis as shown in Fig. 3b. Analyses of standard Southern blots [Dobkin et al., 1996] suggested that premutation alleles in mosaics were over represented, presumably because the transfer of large, full mutation fragments was relatively inefficient. In contrast the size of the fragment in an agarose gel section should not affect PCR amplification of an internal sequence. Thus, although mosaicism was apparent in the Southern blot (Fig. 3a), the pattern in Fig. 3b suggests that the standard Southern Brown WT, Houck Jr GE, Jeziorowska A, Levinson F, Ding X-H, Dobkin C, Zhong N, Henderson J, Sklower-Brooks S, Jenkins EC. 1993. Rapid fragile X carrier screening and prenatal diagnosis using a nonradioactive PCR test. JAMA 270:1569–1575. Dobkin C, Nolin SL, Cohen I, Sudhalter V, Bialer MG, Ding H, Jenkins EC, Zhong N, Brown WT. 1996. Tissue differences in fragile X mosaics: Mosaicism in blood cells may differ greatly from skin. Am J Med Genet 58:641–642. Eichler EE, Richards S, Gibbs RA, Nelson DL. 1993. Fine structure of the human FMR-1 gene. Hum Mol Genet 2:1147–1153. Fu Y-H, Kuhl DPA, Pizzuti A, Pieretti M, Sutcliffe JS, Richards S, Verkerk AJMH, Holden JJA, Fenwick RG Jr, Warren ST, Oostra BA, Nelson DL, Caskey CT. 1991. Variation of the CGG repeat at the fragile X site results in genetic instability: Resolution of the Sherman Paradox. Cell 67:1047–1058. Hoopes BC, McClure WR. 1981. Studies on the selectivity of DNA precipitation by spermine. Nucleic Acid Res 9:5493–5504. Nolin SL, Glicksman A, Houck Jr GE, Brown WT, Dobkin CS. 1994. Mosaicism in fragile X affected males. Am J Med Genet 51:509–512. Rousseau F, Heitz D, Biancalana V, Blumenfield S, Kretz C, Boué J, Tommerup N, van der Hagen C, DeLozier-Blanchet C, Croquette M-F, Gilgenkrantz S, Jalbert P, Voelckel M-A, Oberlé I, Mandel J-L. 1991. Direct diagnosis by DNA analysis of the fragile X syndrome of mental retardation. N Engl J Med 325:1673–1681. Sambrook J, Fritsch EF, Maniatis T. 1989. Molecular Cloning. A Laboratory Manual, 2nd ed. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory Press. Sanford J, Chapman V, Rossant J. 1985. DNA methylation in extraembryonic lineages of mammals. Trends Genet. 1:89–93. Verkerk AJMH, Pieretti M, Sutcliffe JS, Fu Y-H, Kuhl DPA, Pizzuti A, Reiner O, Richards S, Victoria MF, Zhang F, Eussen BE, van Ommen G-J B, Blonden LAJ, Riggins GJ, Chastain JL, Kunst CB, Galjaard H, Caskey CT, Nelson DL, Oostra BA, Warren ST. 1991. Identification of a gene (FMR-1) containing a CGG repeat coincident with a breakpoint cluster region exhibiting length variation in fragile X syndrome. Cell 65:905–914. Warren ST, Nelson DL. 1994. Advances in molecular analysis of fragile X syndrome. JAMA 271:536–542.