Molecular basis of loss -of -function and gain -of -function CHO glycosylation mutants

Abstract

Glycosylation mutants provide access to structure/function relationships of carbohydrates. Delineation of precise molecular bases of the glycosylation defect in each mutant is essential to maximize their use in glycobiology. The molecular basis of loss-of-function and gain-of-function glycosylation mutants was sought in this study. Lec1 and Lec1A Chinese Hamster Ovary (CHO) cells are loss-of-function mutants. Lec1 cells lack GlcNAc-TI activity. While Lec1A mutants have weak GlcNAc-TI activity due to the reduced affinity of GlcNAc-TI for both substrates. The molecular origins of six independent Lec1 mutants were determined. One had GlcNAc-TI gene (Mgat1) transcripts of reduced size while each of five others possess a unique insertion or transition mutation in the Mgat1 gene coding region leading to a truncated GlcNAc-TI protein. Two point mutations were identified that gave rise to the Lec1A phenotype in three independent Lec1A mutants. The mutation D212N resides in a conserved DXD motif that makes critical interactions with bound UDP-GlcNAc and Mn2+ ion and the R303W is predicted to alter interactions important in stabilizing a critical structural element in rabbit GlcNAc-TI.;LEC11B gain-of-function mutants express an alpha(1,3)fucosyltransferase cgFUT6B gene, which is not expressed in parental CHO cells. Consequently, LEC11B cells make Lewis X and sialyl-Lewis X glycan determinants that are absent from parent cells. cgFUT6B gene is expressed in LEC11B cells due to the loss of a trans-acting negative regulator. A functional expression cloning approach using a retrovirus system was developed to isolate the negative regulator. The most promising candidate among the rescued cDNAs for a negative regulator was a partial cDNA encoding the C-terminal region of Histone Deacetylase 5 (HDAC5). Both partial and full-length cDNAs of CHO HDAC5 inhibited Lewis X expression and suppressed cgFUT6B transcripts when expressed in the retroviral vector, but neither functioned as a suppressor when expressed from the CMV promoter. No differences in HDAC5 transcripts or protein levels between LEC11B and CHO cells were found. Therefore, while HDAC5 could not be the basis for the mutation that gave rise to LEC11B, it is a candidate for a second site revertant of the mutation and HDAC5 may regulate FUT6 gene expression in vivo.