Search

Two-dimensional gene scanning (TDGS) is a method for analyzing multiple DNA fragments in parallel for all possible sequence variations, using extensive multiplex PCR and two-dimensional electrophoretic separation on the basis of size and melting temperature. High throughput application of TDGS is limited by the prolonged time periods necessary to complete the second-dimension electrophoretic separation step--denaturing gradient gel electrophoresis--and the current need for gel staining. To address these problems, we constructed a high-voltage, automatic, two-dimensional electrophoresis system and used this in combination with thinner gels to reduce two-dimensional electrophoresis time about 80%. Instead of gel staining, we used three different fluorophores to simultaneously analyze three samples in the same gel. These improvements greatly increase TDGS speed and throughput and make the method highly suitable for large-scale single-nucleotide polymorphism discovery and genetic testing.

An expedient, accurate, and cost-efficient test was developed to scan critical regions of the mitochondrial genome for all possible mutations by two-dimensional DNA electrophoresis. The test involves a two-step multiplex PCR amplification: a long-distance PCR to amplify almost the entire mitochondrial genome, which then serves as template for the amplification of 25 short PCR fragments in two multiplex groups corresponding to regions implicated in human diseases. The mixture of fragments was subsequently subjected to two-dimensional electrophoretic separation, first by size in a nondenaturant polyacrylamide gel and then on the basis of basepair sequence in a denaturing gradient polyacrylamide gel. This latter process of denaturing gradient gel electrophoresis is a most accurate form of mutation detection on the basis of differences in melting behavior of mutant and wildtype fragments. Evaluation of the method using samples with known homoplasmic and heteroplasmic mutations, as well as CEPH pedigrees to study segregation of polymorphic variants, indicated a very high accuracy; none of the previously identified mutations and polymorphisms escaped detection, and no erroneous segregation patterns of polymorphic variants were observed. In addition, two variants were found to be novel mutations when analyzed by sequence analysis. One of these novel mutations was a heteroplasmic mutation in the COXIII gene that was found to segregate to homoplasmy in the next generation. Heteroplasmic mutations as low as 1% of mtDNA could still be detected., (Copyright 1998 Academic Press.)

By inter-repeat PCR, multiple polymorphic loci can be targeted in parallel. To improve resolution and extend the number of detectable polymorphisms, Alu-directed inter-repeat PCR products from two large pedigrees of the Centre d'Etude du Polymorphisme Humain (CEPH) were electrophoretically resolved in non-denaturing polyacrylamide gels and, separately, on the basis of sequence content by denaturing gradient gel electrophoresis (DGGE). The resolution in DGGE gels was found to be superior to that in non-denaturing gels and a higher number of fragments was detected separately. The number of polymorphic bands detected by DGGE alone, however, was lower than that after size separation. This is ascribed to the fact that because of complete melting, small polymorphic fragments can run off the gel. With three Alu-specific primers, 18 and 16 polymorphic bands per individual were detected by size separation in pedigrees 1200 and 6600, respectively. In the same two pedigrees, seven and 15 polymorphic bands, respectively, were detected by DGGE. Segregation analysis of polymorphisms in the CEPH pedigrees indicated that most polymorphisms detected by DGGE were different from those detected by size separation.

By two-dimensional (2-D) genome typing, i.e., electrophoretic separation of restriction enzyme-digested genomic DNA on the basis of both size and sequence in denaturing gradient gels followed by hybridization analysis, several hundred alleles (spots) can be analyzed in parallel, using a micro- or minisatellite core probe. We studied the segregation of 213 and 214 spots detected by microsatellite core probe (CAC)n and minisatellite core probe 33.6, respectively, in two three-generation human pedigrees. Reproducibility of the spot patterns was such that particular spot variants could be scored in both pedigrees. Between 73 and 74% of the spots scored were variant and were transmitted in a Mendelian manner. Very little cosegregation among the 2-D spots themselves was observed, suggesting a random distribution over the genome. Several pairs of spots that appeared to contain both alleles from single loci were identified. The few spots detected by both probes (overlapping spots) showed different segregation patterns, indicating that each probe detects independent sets of genetically informative loci. These results provide a firm basis for using 2-D DNA typing to identify disease loci and for constructing a 2-D spot genetic linkage map of the human genome.

We have analyzed tumor and lymphocyte DNA from six breast cancer patients by one- and two-dimensional DNA fingerprinting using micro- and minisatellite core probes to estimate the extent and nature of DNA alterations in tumors. Both approaches were compared regarding sensitivity in genome analysis. We find that the number of deletions and amplifications increases linearly with the number of restriction fragments analyzed using the two-dimensional approach, as compared with the number found using the more traditional one-dimensional method. A set of four micro- and minisatellite core probes resulted in a total number of approximately 70 bands per patient using one-dimensional analysis of RsaI-digested DNA. When the same DNA was analyzed with the two-dimensional approach about 300 analyzable spots were resolved. In one patient, the tumor DNA contained 11 amplified spots and 14 deleted spots when compared to the patients lymphocyte DNA. Using HaeIII-digested DNA, a maximum of 845 spots could be observed, with only three probes.

We have explored the usefulness of a cloned low-repetitive DNA element, termed RTVL-H, for the region-specific isolation of DNA cosmids from chromosome 21. Hybridization of the RTVL-H element as a probe to genomic Southern blots of DNAs from a chromosome 21-specific somatic cell hybrid panel demonstrated the presence of at least seven RTVL-H-related sequences in defined regions dispersed over the chromosome. Some of the RTVL-H sequences detected by using the chromosome 21-specific somatic cell hybrid panel were subsequently isolated from a chromosome 21-specific cosmid library and mapped using somatic cell hybrids. In general, close correspondence was found between mapping results obtained using the consensus RTVL-H probe on the chromosome 21-specific somatic cell hybrid panel and the more precise mapping obtained using cosmids. The results demonstrate that sequences such as the RTVL-H element can be used for isolating region-specific sequences.

The induction and disappearance of DNA single-strand breaks (SSB) in human peripheral blood lymphocytes (PBL) and fibroblasts exposed to methyl methanesulfonate (MMS) were investigated by using the alkaline filter elution assay. In the two cell types, identical amounts of SSB were induced during a 45-min treatment with a given dose of MMS. In quiescent PBL only 9 +/- 4% (mean +/- SD) of the induced SSB had disappeared at 1 h after exposure, whereas in phytohemagglutinin-stimulated PBL, 23 +/- 12% disappeared within the same repair period. The percentage SSB disappearance in confluent fibroblasts was 25 +/- 2% at 1 h after exposure. As in PBL, the percentage SSB disappearance in fibroblasts appeared to be proliferation-dependent; actively dividing fibroblasts removed 50 +/- 12% of the MMS-induced SSB during the 1-h repair period. The accumulation of SSB in PBL, but not in fibroblasts, during MMS exposure in the presence of the excision-repair inhibitor 1-beta-D-arabinofuranosylcytosine indicated the utilization of different repair pathways in these two cell types. The generally lower rate of disappearance of MMS-induced SSB in PBL as compared to fibroblasts correlated with an increased loss of cell viability, measured by determining the incorporation of [3H]thymidine.

Sedimentation of nucleoids through neutral sucrose density gradients has shown that nucleoids isolated from phytohemagglutinin (PHA)-stimulated human peripheral blood lymphocytes (PBL) sediment faster than nucleoids derived from quiescent lymphocytes, which was attributed to rejoining of DNA single-strand breaks (SSB) present in the resting cells (A.P. Johnstone, and G.T. Williams (1982) Nature (London) 300, 368). We isolated PBL from donors and determined the amount of SSB in nonradiolabeled, untreated resting and PHA-stimulated cells by applying the alkaline filter elution technique. Calibration was based on dose-dependent induction of SSB by 60Co-gamma-radiation. Quiescent cells did not contain a sizable amount of SSB. Mitogen-stimulated cells showed equally low amounts of SSB per cell. The present study indicates that the interpretation of the results obtained with the nucleoid sedimentation technique concerning the supposed rejoining of SSB in PHA-stimulated human lymphocytes is incorrect. Other, equally sensitive, techniques such as alkaline filter elution appear to be preferable for studies on DNA damage and repair.

UV-induced DNA excision repair was studied as DNA repair synthesis and dimer removal in rat fibroblast cultures, initiated from either dense or sparse inocula of primary cells grown from skin biopsies. During passaging in vitro an initial increase in DNA repair synthesis, determined both autoradiographically as unscheduled DNA synthesis (UDS) and by means of the BrdU photolysis assay as the number and average size of repair patches, was found to be associated with a morphological shift from small spindle-shaped to large pleiomorphic cells observed over the first twenty generations. In cell populations in growth crisis, a situation exclusively associated with thin-inoculum cultures in which the population predominantly consisted of large pleiomorphic cells, UDS was found to occur at a low level. After development of secondary cultures into immortal cell lines, both repair synthesis and morphology appeared to be the same as in the original primary spindle-shaped cells. At all passages the capacity to remove UV-induced pyrimidine dimers was found to be low, as indicated by the persistence of Micrococcus luteus UV endonuclease-sensitive sites. These results are discussed in the context of terminal differentiation and immortalization of rat fibroblasts upon establishment in vitro.