Search

Numerous putative glycosyltransferases (GTs) have been identified using bioinformatic approaches. However, demonstrating the activity of these GTs remains a challenge. Here, we describe the development of a rapid in vitro GT-array screening platform for activity of GTs. GT-arrays are generated by cell-free in vitro protein synthesis and binding using microplates precoated with a N-terminal Halo- or a C-terminal GST-tagged GT-encoding plasmid DNA and a capture antibody. These arrays are then used for screening of transferase activities and the reactions are monitored by a luminescence GLO assay. The products formed by these reactions can be analyzed directly from the microplates by mass spectrometry. Using this platform, a total of 280 assays were performed to screen 22 putative fucosyltransferases (FUTs) from family GT37 (seven from Arabidopsis and 15 from rice) for activity toward five acceptors: non-fucosylated tamarind xyloglucan (TXyG), arabinotriose (Ara3), non-fucosylated rhamnogalacturonan I (RG-I), and RG-II from the mur1-1 Arabidopsis mutant, and the celery RG-II monomer lacking Arap and MeFuc of chain B and L-Gal of chain A. Our screen showed that AtFUT2, AtFUT5, and AtFUT10 have activity toward RG-I, while AtFUT8 was active on RG-II. Five rice OsFUTs have XyG-FUT activity and four rice OsFUTs have activity toward Ara3. None of the putative OsFUTs were active on the RG-I and RG-II. However, promiscuity toward acceptors was observed for several FUTs. These findings extend our knowledge of cell wall polysaccharide fucosylation in plants. We believe that in vitro GT-array platform provides a valuable tool for cell wall biochemistry and other research fields. [ABSTRACT FROM AUTHOR]

Although plants contain substantial amounts of arabinogalactan proteins (AGPs), the enzymes responsible for AGP glycosylation are largely unknown. Bioinformatics indicated that AGP galactosyltransferases (GALTs) are members of the carbohydrate-active enzyme glycosyltransferase (GT) 31 family (CAZy GT31) involved in N- and O-glycosylation. Six Arabidopsis GT31 members were expressed in Pichia pastoris and tested for enzyme activity. The At4g21060 gene (named AtGALT2) was found to encode activity for adding galactose (Gal) to hydroxyproline (Hyp) in AGP protein backbones. AtGALT2 specifically catalyzed incorporation of [14C]Gal from UDP-[14C]Gal to Hyp of model substrate acceptors having AGP peptide sequences, consisting of non-contiguous Hyp residues, such as (Ala-Hyp) repetitive units exemplified by chemically synthesized (AO)7 and anhydrous hydrogen fluoride-deglycosylated d(AO)51. Microsomal preparations from Pichia cells expressing AtGALT2 incorporated [14C]Gal to (AO)7, and the resulting product co-eluted with (AO)7 by reverse-phase HPLC. Acid hydrolysis of the [14C]Gal-(AO)7 product released 14C-radiolabel as Gal only. Base hydrolysis of the [14C]Gal-(AO)7 product released a 14C-radiolabeled fragment that co-eluted with a Hyp-Gal standard after high performance anion-exchange chromatography fractionation. AtGALT2 is specific for AGPs because substrates lacking AGP peptide sequences did not act as acceptors. Moreover, AtGALT2 uses only UDP-Gal as the substrate donor and requires Mg2+ or Mn2+ for high activity. Additional support that AtGALT2 encodes an AGP GALT was provided by two allelic AtGALT2 knock-out mutants, which demonstrated lower GALT activities and reductions in β-Yariv-precipitated AGPs compared with wild type plants. Confocal microscopic analysis of fluorescently tagged AtGALT2 in tobacco epidermal cells indicated that AtGALT2 is probably localized in the endomembrane system consistent with its function. [ABSTRACT FROM AUTHOR]

Virtually nothing is known about the mechanisms and enzymes responsible for the glycosylation of arabinogalactan proteins (AGPs). The glycosyltransferase 37 family contains plant-specific enzymes, which suggests involvement in plantspecific organs such as the cell wall. Our working hypothesis is that AtFUT4 and AtFUT6 genes encode α(1,2)fucosyltransferases (FUTs) for AGPs. Multiple lines of evidence support this hypothesis. First, overexpression of the two genes in tobacco BY2 cells, known to contain nonfucosylated AGPs, resulted in a staining of transgenic cells with eel lectin, which specifically binds to terminal a-linked fucose. Second, monosaccharide analysis by high pH anion exchange chromatography and electrospray ionization mass spectrometry indicated the presence of fucose in AGPs from transgenic cell lines but not in AGPs from wild type cells. Third, detergent extracts from microsomal membranes prepared from transgenic lines were able to fucosylate, in vitro, purified AGPs from BY2 wild type cells. Susceptibility of [14C]fucosylated AGPs to α(1,2)fucosidase, and not to α(1,3/4)fucosidase, indicated that an α(1,2) linkage is formed. Furthermore, dearabinosylated AGPs were not substrate acceptors for these enzymes, indicating that arabinosyl residues represent the fucosylation sites on these molecules. Testing of several polysaccharides, oligosaccharides, and glycoproteins as potential substrate acceptors in the fucosyl transfer reactions indicated that the two enzymes are specific for AGPs but are not functionally redundant because they differentially fucosylate certain AGPs. AtFUT4 and AtFUT6 are the first enzymes to be characterized for AGP glycosylation and further our understanding of cell wall biosynthesis. [ABSTRACT FROM AUTHOR]