ORGANIZATION OF A HUMAN UDP-GALNAC-POLYPEPTIDE, N-ACETYLGALACTOSAMINYLTRANSFERASE GENE AND A RELATED PROCESSED PSEUDOGENE

We have previously characterized a cDNA that encodes a full length hum an UDP-GalNAc:polypeptide, N-acetylgalactosaminy/transferase (GalNAc-t ransferase) (J.A. Meurer et al., J. Biochem., 118, 568-574, 1995). The present report describes the characterization of the corresponding hu man GalNAc-transferase gene and a related pseudogene. Two human genomi c libraries, lambda and P1, were screened with probes derived from the human GalNAc-transferase nucleotide sequence, resulting in the isolat ion of four genomic clones. Southern blotting, PCR analysis, and seque ncing revealed that three clones, lambda.HG-5, P1.GALN-A, and P1.GALN- B, contained overlapping genomic sequences that encompass over 55 kilo base pairs (kb) of genomic DNA and comprise a portion of the human Gal NAc-transferase 5'-and 3'-untranslated regions and the entire coding r egion. The human GalNAc-transferase gene structure consists of at leas t 11 exons ranging in size from 99 to > 620 nucleotides which are sepa rated by 10 introns ranging in size from 0.7 to similar to 12.5 kb. Th e fourth genomic clone, P1-GALN-psi, contained a similar to 2.4 kb seq uence region which shares an overall 78.6% nucleotide identity with co ding region exons 1 and 3 through 11 of the human GalNAc-transferase g ene. However, a lack of intron sequences, as well as the presence of m ultiple nucleotide mutations, insertions, and deletions that disrupt t he potential GalNAc-transferase reading frame, suggest that P1.GALN-ps i contains a processed pseudogene. Screening of a human/rodent somatic cell hybrid panel with a P1.GALN-psi probe localized the GalNAc-trans ferase pseudogene to chromosome 3. Hence, the human genome contains at least two related GalNAc-transferase genes that are located on separa te chromosomes.