Stefano Canevascini, Krisztina Kozma, Marianne Grob, Jasmin Althaus, Dominique Klein, Ruth Chiquet-Ehrismann and Jan Hofsteenge. Protein glycosylation is crucial for the development and growth, as well as the normal functioning and survival of an organism. However, in most instances we do not know the exact function and the mechanism of action of glycan modifications at the molecular level. We want to elucidate the function of glycosylations that occur on thrombospondin type 1 repeats (TSRs). TSRs are cysteine- and arginine-rich modules of about 60 amino residues in length that are part of many proteins with important physiological functions, like e.g. inhibition of angiogenesis, cell- and axonal guidance and regulation of complement in higher organisms. TSRs undergo two different types of glycosylation: i) C-mannosylation of tryptophan residues and (ii) the addition of O-linked fucose to serine and threonine residues, followed by extension with a single glucose to form Glc-beta1,3-Fuc-O-Ser/Thr. We have begun to study the latter process in C. elegans, in order to reveal its possible function. The C. elegans genome encodes 29 TSR-containing proteins, some of which have been found to be important for normal development, like UNC-5, GON-1, ADT-1 and NAS-36.. We have demonstrated that C. elegans
pad-2 encodes the fucosyltransferase (POFUT-2) that modifies TSRs. Extracts from HEK293T cells overexpressing PAD-2 show a 10-fold increase in transferase activity towards a TSR. Furthermore, product analysis by mass spectrometry of isolated peptides showed the fucose to be linked to the correct Thr residue. In contrast, no increase in TSR fucosylation was observed by overexpressing C. elegans POFUT1, the enzyme that fucosylates EGF-like repeats (encoded by C15C7.7). We are characterizing the
pad-2(
tm1756) mutant that we received from the NBP, Japan. The
tm1756 strain is homozygous viable and lacks all fucosyltransferase activity towards a TSR. The mutant does not show embryonic lethality nor morphogenetic defects (Menzel et al., Genomics 84: 320, 2004), but a precocious dorsal migration of the anterior gonad arm is observed. The latter is reminiscent of the phenotype resulting from the precocious expression of UNC-5 in DTCs (Su et al., Dev. 127: 585, 2000). Genetic experiments show that
unc-6,
unc-5 and
unc-40 genes are epistatic to
pad-2, suggesting that
pad-2 is acting as an inhibitor of the netrin pathway in DTCs. We are now testing whether the TSRs of C. elegans UNC-5 are fucosylated in vivo and whether this modification is necessary for the
pad-2-mediated inhibition of the DTC ventro-dorsal migration. Preliminary evidence indicates that C. elegans also contains the glucosyltransferase activity that attaches glucose to the fucose It will be of interest to examine its possible function.