Susan A. Ledbetter

Current efforts to map the human genome are focused on individual chromosomes or smaller regions and frequently rely on the use of somatic cell hybrids. We report the application of the polymerase chain reaction to direct amplification of human DNA from hybrid cells containing regions of the human genome in rodent cell backgrounds using primers directed to the human Alu repeat element. We demonstrate Alu-directed amplification of a fragment of the human HPRT gene from both hybrid cell and cloned DNA and identify through sequence analysis the Alu repeats involved in this amplification. We also demonstrate the application of this technique to identify the chromosomal locations of large fragments of the human X chromosome cloned in a yeast artificial chromosome and the general applicability of the method to the preparation of DNA probes from cloned human sequences. The technique allows rapid gene mapping and provides a simple method for the isolation and analysis of specific chromosomal regions.

Human-rodent hybrid cell lines have been analyzed with regard to their human DNA content by using various DNA probe sets, derived from the hybrids, for in situ hybridization to normal human metaphase chromosome spreads. Total genomic hybrid DNA was compared with probe sets of hybrid DNA that were highly enriched in human sequences. The latter probes were obtained by amplification through the polymerase chain reaction (PCR) using oligonucleotide primers directed to human specific subsequences of the interspersed repetitive sequences Alu and L1. Previously unidentified chromosomal material within hybrid lines was characterized with speed and precision. It is demonstrated that the complete human complement of hybrid lines can be rapidly assessed by comparing the data obtained with the Alu-PCR products with the results from the L1-PCR products or from the genomic hybrid DNA. This approach using interspersed repetitive sequence-PCR products is simple and fast and also provides an alternative way of generating complex DNA probe sets for the specific delineation of entire chromosomes or subchromosomal regions by in situ hybridization.

Fluorescence in situ hybridization (FISH) using two cosmid probes (41A and P13) from the Miller-Dieker syndrome (MDS) critical region in 17p13.3 was performed in a blinded comparison of three MDS patients with submicroscopic deletions and in four normal relatives used as controls. The controls showed both chromosome 17 homologues labeled in 85%-95% of cells, while each patient showed only one homologue labeled in 75%-80% of cells. Two MDS patients with cryptic translocations were also studied. In one case, a patient and her mother had the same der(17) (p+), but the reciprocal product of the translocation could not be identified in the mother by G-banding (i.e., it was a "half-cryptic" translocation). FISH revealed a 3q;17p translocation. The other case involved a patient with apparently normal karyotype. Because a large molecular deletion was found, a translocation involving two G-negative telomeres (i.e., a "full-cryptic" translocation) was postulated. FISH studies on her father and normal brother showed an 8q;17p translocation. These studies demonstrate that in situ hybridization is an efficient method for deletion detection in Miller-Dieker syndrome. More important, parental studies by FISH on patients demonstrating molecular deletions and a normal karyotype may identify cryptic translocation events, which cannot be detected by other molecular genetic strategies. Similar in situ strategies for deletion detection can be developed for other microdeletion syndromes, such as Prader-Willi/Angelman syndrome or DiGeorge syndrome.

The human genome contains approximately 50,000 copies of an interspersed repeat with the sequence (dT.dG/dA.dC)n, where n = approximately 10-60. We and others have found that several of these repeats have variable lengths in different individuals, with allelic fragments varying in size by multiples of 2 bp. These "microsatellite" variable number of tandem repeats (VNTRs) may be scored by PCR, using unique flanking primers to amplify the repeat-containing regions and resolving the products on DNA sequencing gels. Since few VNTRs have been found on the X chromosome, we screened a flow-sorted X chromosome-specific genomic library for microsatellites. Approximately 25% of the phage clones hybridized to a poly (dT-dG).poly(dA-dC) probe. Of seven X-linked microsatellites present in positive phages, five are polymorphic and three have both eight or more alleles and heterozygosities exceeding 75%. Using PCR to amplify genomic DNAs from hybrid cell panels, we confirmed the X localization of these VNTRs and regionally mapped four of them. The fifth VNTR was regionally mapped by virtue of its tight linkage to DXS87 in Centre du Polymorphisme Humain families. We conclude that whatever factors limit the occurrence of "classical" VNTRs and RFLPs on the X chromosome do not appear to operate in the case of microsatellite VNTRs.