Structure and Function of UDP-GalNAc:polypeptide alpha-GalNAc transferases

National Institute Of Diabetes And Digestive And Kidney Diseases

Investigators

Linked publications & trials

Abstract

Further characterization of the ppGalNAcT family:[unreadable] [unreadable] To date, 16 human ppGalNAcT isoforms have been described in the literature. Jaya Raman has conducted an extensive search of human and mouse nucleotide sequence databases that revealed the existence of an additional 4 human isoforms which she is characterizing (together with other members of the lab) with respect to expression pattern, and relatedness to known family members by sequence analysis and enzymatic activity. In a manuscript that is being prepared, we will assign unique names to each of these novel isoforms and propose names for several previously described mammalian ppGalNAcTs to eliminate confusion present within the nomenclature of current sequence databases and the ppGalNAcT literature.[unreadable] [unreadable] Molecular mechanisms of ppGalNAcT catalysis:[unreadable] [unreadable] Previous studies have suggested that the lectin domain modulates the glycosylation of glycopeptide substrates and may underlie the strict glycopeptide specificity of some isoforms (ppGalNAcT-7, -10). Jaya Raman, in collaboration with Tim Fritz, Tom Gerkin and David Live, examined the activity and glycosylation site preference of lectin domain deletion and exchange constructs of the peptide/glycopeptide transferase ppGalNAcT-2 (hT2) and the glycopeptide transferase ppGalNAcT-10 (hT10). They demonstrated that the lectin domain of hT2 directs glycosylation site selection for glycopeptide substrates. Pre-steady state kinetic measurements showed that this effect is attributable to two mechanisms, either lectin domain-aided substrate binding or lectin domain-aided product release following glycosylation. They found that glycosylation of peptide substrates by hT10 requires binding of existing GalNAcs on the substrate to either its catalytic or lectin domain, thereby resulting in its apparent strict glycopeptide specificity. These results highlight the existence of two modes of site selection used by these ppGalNAcTs: local sequence recognition by the catalytic domain and the concerted recognition of distal sites of prior glycosylation together with local sequence binding mediated, respectively, by the lectin and catalytic domains. The latter mode may facilitate the glycosylation of serine or threonine residues which occur in sequence contexts that would not be efficiently glycosylated by the catalytic domain alone. Local sequence recognition by the catalytic domain differs between hT2 and hT10 in that hT10 requires a preexisting GalNAc residue while hT2 does not. This work was recently published.[unreadable] [unreadable] Biological consequences of ablating expression/activity of ppGalNAcTs:[unreadable] [unreadable] In collaboration with Jamey Marth (HHMI, UCSD) and Stasia Anderson, Daryl Despres and M Starost from NIH, Yu Guan has studied the phenotypes resulting from the deletion of ppGalNAcT-1. The genotypic ratio of offspring from heterzygote crosses was +/+, +/-, -/-=1.48: 2.8: 1 from a total of 265 animals suggesting 33% of the ppGalNAcT-1 KO animals die before the age of 1 month (the age at which animals were genotyped). She found that the heart/body weight ratio was increased in 4-month-old KO animals compared to heterozygotes and wild-type littermate controls. Histological analysis revealed right or left ventricle hypertrophy, enlarged cardiomyocytes, increased interstitial fibrosis, and mitochondria swelling and dissolution in knockout animals. She is currently analyzing the underlying causes of this change. [unreadable] [unreadable] Hazuki Miwa has demonstrated by real-time PCR, that ppGalNAcT-1 (T1) transcripts are highly expressed in bone suggesting the involvement of T1 in bone protein glycosylation. The apparent molecular masses of bone proteins extracted from wild-type and T1-knockout mice were determined by SDS-PAGE and western blot analysis to reveal that osteopontin (OPN) and bone sialoprotein (BSP) from T1-knockout migrated further than did those from the wild-type. After mucin-type O-glycans were enzymatically removed, the apparent molecular masses of OPN and BSP from the wild-type and T1KO were reduced to the same size, suggesting that the observed molecular mass differences were due to incomplete glycosylation in T1-knockout mice. While T1-knockout mice lack an obvious bone phenotype, OPN and BSP play a variety of functional roles. Thus, the effects of T1-mediated glycosylation on their functions will be investigated. [unreadable] [unreadable] In collaboration with Michael Econs (Univ. Indiana), Tim Fritz has helped to characterize a ppGalNAcT-3 null mouse. In humans, mutation of the isoform results in humoral calcinosis which, among other things, yield high levels of serum phosphate and ectopic calcifications. A manuscript has been submitted for publication.[unreadable] [unreadable] We are collaborating with two different chemistry groups to identify small molecule inhibitors of ppGalNAcTs (Carolyn Bertozzi, HHMI and UC Berkeley and Suzzane Walker, Harvard Medical School). Yu Guan is performing the biological assays on cell lines with candidate inhibitors.[unreadable] [unreadable] We are collaborating with Dr. Danielle Dube, Bowdoin College on the characterization of a novel yeast-two hybrid system that she developed which challenges protein interactions within the Golgi rather than the nucleus.