Most plant glycoproteins contain substantial amounts of paucimannosidic N-glycans instead of their direct biosynthetic precursors, complex N-glycans with terminal N-acetylglucosamine residues. We now demonstrate that two β-N-acetyl-hexosaminidases (HEXO1 and HEXO3) residing in different subcellular compartments jointly account for the formation of paucimannosidic N-glycans in Arabidopsis thaliana. Total N-glycan analysis of hexo knock-out plants revealed that HEXO1 and HEXO3 contribute equally to the production of paucimannosidic N-glycans in roots, whereas N-glycan processing in leaves depends more heavily on HEXO3 than on HEXO1. Because hexo1 hexo3 double mutants do not display any obvious phenotype even upon exposure to different forms of abiotic or biotic stress, it should be feasible to improve the quality of glycoprotein therapeutics produced in plants by down-regulation of endogenous β-N-acetylhexosaminidase activities.

A common argument against using plants as a production system for therapeutic proteins is their inability to perform authentic human N-glycosylation (i.e. the presence of β1,2-xylosylation and core α1,3-fucosylation). In this study, RNA interference (RNAi) technology was used to obtain a targeted down-regulation of the endogenous β1,2-xylosyltransferase (XylT) and α1,3-fucosyltransferase (FucT) genes in Nicotiana benthamiana, a tobacco-related plant species widely used for recombinant protein expression. Three glyco-engineered lines with significantly reduced xylosylated and/or core α1,3-fucosylated glycan structures were generated. The human anti HIV monoclonal antibody 2G12 was transiently expressed in these glycosylation mutants as well as in wild-type plants. Four glycoforms of 2G12 differing in the presence/absence of xylose and core α1,3-fucose residues in their N-glycans were produced. Notably, 2G12 produced in XylT/FucT-RNAi plants was found to contain an almost homogeneous N-glycan species without detectable xylose and α1,3-fucose residues. Plant-derived glycoforms were indistinguishable from Chinese hamster ovary (CHO)-derived 2G12 with respect to electrophoretic properties, and exhibited functional properties (i.e. antigen binding and HIV neutralization activity) at least equivalent to those of the CHO counterpart. The generated RNAi lines were stable, viable and did not show any obvious phenotype, thus providing a robust tool for the production of therapeutically relevant glycoproteins in plants with a humanized N-glycan structure.

Previous studies have reported that plants contain negligible amounts of free or protein-bound N-acetylneuraminic acid (Neu5Ac). This is a major disadvantage for the use of plants as a biopharmaceutical expression system, since N-glycans with terminal Neu5Ac residues are important for the biological activities and half-lives of recombinant therapeutic glycoproteins in humans. For the synthesis of Neu5Ac-containing N-glycans, plants have to acquire the ability to synthesize Neu5Ac and its nucleotide-activated derivative, cytidine monophospho-N-acetylneuraminic acid. In this study, we have generated transgenic Arabidopsis (Arabidopsis thaliana) plants expressing three key enzymes of the mammalian Neu5Ac biosynthesis pathway: UDP-N-acetylglucosamine 2-epimerase/N-acetylmannosamine kinase, N-acetylneuraminic acid phosphate synthase, and CMP-N-acetylneuraminic acid synthetase. Simultaneous expression of UDP-N-acetylglucosamine 2-epimerase/N-acetylmannosamine kinase and N-acetylneuraminic acid phosphate synthase resulted in the generation of significant Neu5Ac amounts (1,275 nmol g−1 fresh weight in leaves) in planta, which could be further converted to cytidine monophospho-N-acetylneuraminic acid (2.4 nmol g−1 fresh weight in leaves) by coexpression of CMP-N-acetylneuraminic acid synthetase. These findings are a major step toward the production of Neu5Ac-containing glycoproteins in plants.

Analysis of the numerous possible, often isobaric structures of protein-bound oligosaccharides calls for a high-performance two-dimensional method that combines liquid chromatography's ability to separate isomers and mass spectrometry's ability to determine glycan composition. Here we investigate the usefulness of porous graphitic carbon columns coupled to ESI-MS for the separation of N-glycans with two or more sialic acids. Internal standards helped to rectify retention time fluctuations and thus allowed elution times to play an essential role in the structural assignment of peaks. For generation of a retention time library, standards representing the possible isomers of diantennary non-, mono-, and disialylated N-glycans, differing in the linkage of galactose and sialic acids as well as isobaric hybrid-type N-glycans, were produced using recombinant glycosyltransferases. Once the retention times library was established, isomers could be identified by LC-ESI-MS in the positive mode without additional MS/MS experiments. The method was applied for the detailed structural analysis of fibrin(ogen) N-glycans from various species (human, cow, pig, mouse, rat, cat, dog, Chinese hamster, horse, sheep, and chicken). All fibrins contained diantennary N-glycans. They differed in the occurrence of β1,3-linked galactose, α2,3-linked sialic acids, and N-glycolylneuraminic acid, in the mono/diantennary glycan ratio, and in the O-acetylation of neuraminic acids. The separation system's potential for analyzing tri- and tetrasialylated N-glycans was demonstrated.