Nucleic Acids Research, Vol. 18, No. 18 5407 Direct detection and automated sequencing of individual alleles after electrophoretic strand separation: identification of a common nonsense mutation in exon 9 of the human lipoprotein lipase gene Akira Hata, Margaret Robertson, Mitsuru Emi and Jean-Marc Lalouel* Howard Hughes Medical Institute and Department of Human Genetics, University of Utah Health Sciences Center, Salt Lake City, UT 84132, USA Received June 26, 1990; Accepted August 15, 1990 ABSTRACT Large-scale screening by direct sequencing of DNA to detect molecular variants remains a laborious endeavor whose difficulty is compounded by heterozygosity. We show that mobility shifts of single-stranded DNA electrophoresed under nondenaturing conditions can be used not only to detect variants (Orita,M. et al., 1989, Genomics, 5, 874 - 879), but also to separate and sequence directly individual alleles. In this manner, we have identified a common variant of human lipoprotein lipase resulting from a nonsense mutation in exon 9 of the gene. Whether this variant is of functional significance remains to be determined. INTRODUCTION A variety of methods have been proposed to identify directly mutations in genomic DNA, whether enzymatic (1,2), chemical (3-5) or physical (6- 8). Their scope is further enhanced when they are used to analyze enzymatically amplified DNA (9). These methods can complement direct sequencing in large-scale screening of subjects, by identifying that subset of DNA segments to be sequenced. The method proposed by Orita et al. (7,8), referred to by the authors as 'single-strand conformation polymorphism (SSCP) analysis', has the merit of great experimental simplicity. It relies on the differential electrophoretic mobility of single-stranded DNA under nondenaturing conditions, as even a single nucleotide substitution may induce conformational changes leading to detectable mobility shifts (10). Because the detection of point mutations by direct sequencing of enzymatically amplified DNA from heterozygous individuals can present experimental difficulties, we investigated whether the method proposed by Orita et al. (7,8) could also be used to purify single-stranded molecules harboring mutations, which could then be submitted to sequencing. We report that separation of single-stranded DNA in nondenaturing polyacrylamide gels, followed by elution, enzymatic amplification, and direct sequencing according to a thermocycling protocol on an automated sequencer, led to the identification and characterization of a common variant of human lipoprotein lipase (LPL), a key enzyme in the hydrolysis of dietary and endogenous fat (11). MATERIALS AND METHODS Direct sequencing of asymmetrically amplified single-stranded DNA In all instances, genomic DNA was prepared from peripheral blood leukocytes following standard procedures. A 248 bp segment of exon 29 of the apolipoprotein B gene (apo B) was amplified by the polymerase chain reaction using oligonucleotide primers (5'-GAGTGGAAGGACAAGGCCCAG-3', 5'-CCTA- CCTCCCTTATGAACATAGT-3') derived from the normal apoB sequence (12,13), and the resulting product was purified on a Centricon 100 (Amicon, Danvers, MA). An aliquot (1 yd) was used to generate single-stranded DNA by asymmetric amplification (14), one primer being augmented with sequence complementary to the universal M 13 priming site (5'-TGTAAAACGACGGCCAGT-3'). Sequencing was performed by the dideoxy chain termination method (15) with Taq polymerase by means of an ABI 373A DNA sequencer (Applied Biosystems, Foster City, CA). Direct search for mutations in genomic DNA For each of the 9 translated exons of human LPL (16), 0.1 ,^g genomic DNA was enzymatically amplified (17) on a DNA Thermal Cycler (Perkin Elmer Cetus, Norwalk, CT) basically following the method of Orita et al. (1). The reaction mixture contained 10 pmol of each primer, 2 nmol of each dNTP, 0.1 jig of genomic DNA, 10 MCi of [a-32P]dCTP (3000 Ci/mmol, 10 mCi/ml, NEN) and 0.25 units of Taq polymerase (Perkin- Elmer Cetus, Norwalk, CT) in 10 Iul of amplification buffer. The oligonucleotides used as primers and the cycling parameters are given elsewhere (18). The reaction mixture was diluted with 50 M1 of 0.1 % NaDodSO4 and 10 mM EDTA. An aliquot was * To whom correspondence should be addressed