Xyloglucan (XyG) represents the most abundant hemicellulose in the primary cell wall of many dicot and non-graminaceous monocot plant species. Found throughout the land plant lineage, the structure of this polysaccharide varies by species, tissue, and developmental stage. Some proteins responsible for the biosynthesis and metabolism of XyG have been previously identified but additional genes remain to be uncovered. In the work presented here, several genetic and genomic approaches were used to identify previously unknown genes involved in XyG biosynthesis and metabolism. A forward genetic screen of mutagenized Arabidopsis thaliana resulted in the identification of the Altered Xyloglucan 9 (AXY9) gene, coding for a protein of unknown function, as part of an apparently plant-specific pathway for the acetylation of polysaccharides. The AXY9 protein may be part of a system for shuttling activated acetyl groups across the Golgi membrane for use by polysaccharide-specific acetyltransferases. A transcriptional profiling approach used on developing XyG-rich Tropaeolum majus seeds revealed a galactosyltransferase that acts on XyG, as demonstrated by mutant analysis in A. thaliana. By utilizing comparative genomics, two putative XyG arabinofuranosyltransferase genes were discovered from tomato, which are able to act in vivo to arabinosylate XyG. Expression of these genes rescued mechanical and morphological phenotypes of an A. thaliana mutant deficiency for XyG galactosylation, providing insight into the function of XyG substitution. To provide a testing platform for candidate genes involved in XyG biosynthesis and to investigate the activities of these genes independently from endogenous plant proteins, an attempt was made to reconstruct the XyG biosynthetic pathway in a heterologous host. This effort resulted in the production of a glucan backbone and the UDP-xylose donor substrate required for the xylosyltransferases, but a XyG polysaccharide was not detected.

Most non-cellulosic plant cell wall polysaccharides including the hemicellulose xyloglucan and the pectic polysaccharides can be O-acetylated. This feature has direct significance in the use of these polymers in the food and biofuel industry. For example, increased pectin acetylation can reduce its gelling abilities and is hence detrimental in its application as a food thickener or emulsifier. In general, plant biomass with wall polymers with high acetate content can negatively influence biomass hydrolysis by fungal enzymes and interfere with downstream fermentation by yeasts.

Genetic and biochemical approaches were used to identify genes involved in wall polymer acetylation in the model species Arabidopsis. A forward genetics approach was able to identify wall acetylation differences within a population of Arabidopsis ecotypes in both total wall acetate and xyloglucan O-acetylation. One of these naturally occurring mutant alleles of a recently identified putative xyloglucan acetyltransferase (AXY4) was found in the Ty-0 ecotype. Ty-0 completely lacks xyloglucan acetyl-substituents in roots and leaves, suggesting that the lack of XyG acetylation does not compromise the fitness of the ecotype in its ecological niche. In an attempt to further characterize the AXY4 protein, and characterize peptide domains involved in cell wall polymer recognition, a functional domain swap approach was utilized without success. A biochemical approach to identify and purify a putative xyloglucan acetylesterase was attempted also without success. Another way to modulate wall acetylation is by apoplastic plant acetylesterases, which naturally remove acetyl-substituents on polymers. To characterize genes in the Pectin acetyl esterase gene family in Arabidopsis thaliana, reverse genetics was used and lead to the identification of two genes responsible for altering wall acetate levels of pectins. Further pectin fractionation experiments suggest that PAE8 and PAE9 act predominantly but distinctively on pectic rhamnogalacturonan I. Pectin acetylesterase activity was demonstrated in vitro for the proteins encoded by PAE8 and PAE9. Through genetic means it is possible to increase wall acetate content, but this modification lead to a reduction in Arabidopsis inflorescence size giving insights about the function of acetylation of pectins.

Glycosyltransferases catalyze the transfer of sugars from activated phosphate sugar substrates to various acceptor molecules ranging from primary metabolites to glycolipids to plant cell walls. The model plant Arabidopsis encodes about 500 glycosyltransferase proteins. Forty-one of these are members of Glycosyltransferase Family 8. These enzymes are thought to function in synthesis of diverse compounds including the disaccharide galactinol, and the cell wall polymer pectin, and the energy storage polysaccharide starch. This thesis will describe evidence that allowed us to assign additional functions of members of this family: five of these enzymes synthesize the cell wall polymer glucuronoxylan, while one synthesizes glycosyl inositol phosphoryl-ceramide sphingolipids.

The eight Plant Glycogenin-like Starch Initiation Proteins (PGSIPs) were named based on their homology to glycogenin, a protein in fungi and animals that initiates glycogen synthesis by glucosylating itself. The PGSIP proteins were proposed to initiate starch synthesis in a similar manner. However, two of these proteins, PGSIP1 and PGSIP3, are highly coexpressed with glucuronoxylan biosynthetic proteins. Mutations in PGSIP1 cause a decrease in cell wall glucuronoxylan content and microsomal xylan:glucuronosyltransferase activity, leading to the re-annotation of five of the PGSIP proteins as Glucuronic Acid Substitution of Xylan (GUX). Heterologously expressed and purified GUX1, GUX2, and GUX4 show xylan glucuronosyltransferase activity in vitro. These proteins transferred glucuronic acid to a xylooligomer acceptor with different patterns, showing that these related proteins have subtle differences in specificity.

Another protein, PGSIP6, is related to both the GUX proteins and the galactinol synthase (GolS) proteins, which transfer galactose onto the acceptor inositol. The phylogenetic placement of PGSIP6 between the GUX and GolS proteins led us to suspect that this enzyme catalyzes the transfer of glucuronic acid to the inositol moiety of glycosyl inositol phosphorylceramides, a class of glycosphingolipids that are abundant in plant plasma membranes. Expression of PGSIP6 in yeast led to production of glucuronic acid linked to inositol phosphorylceramide. Overexpressing PGSIP6 in Nicotiana benthamiana plants led to an increase in inositol phosphorylceramide glucuronosyltransferase activity, while silencing expression of PGSIP6 led to a decrease in activity. PGSIP6 has now been renamed Inositol Phosphorylceramide Glucuronosyltransferase 1 (IPGT1).