The plant cell wall surrounds every plant cell and is an essential component that is
involved in diverse functions including plant development, morphology, resistance towards plant pathogens etc. The plant cell wall is not only important for the plant. The cell wall has many industrial applications for instance as food additives, nutraceutical, for paper and energy production.

Pectin is a cell wall glycan that crucial for every plant growing on land. Pectin is said to be one of the most complex glycans on earth and it is hypothesized that at least 67 enzymatic reactions are involved in its biosynthesis. To date, only seven glycosyltransferase (GT) genes have been identified and characterized comprising only four biosynthetic activities within pectin biosynthesis. Therefore, increased knowledge about pectin biosynthesis is of great importance if we in the future wants to fully manipulate and exploit the diverse pectin structures for industrial, agronomic and biomedical uses. Increasing evidence suggests that complex formation is important in governing functional coordination of proteins involved in cell wall biosynthesis. In Arabidopsis thaliana, a homogalacturonan (HG) synthase core complex between GALACTURONOSYLTRANSFERASE1 (GAUT1) and GAUT7 has beesn identified and is essential for pectin biosynthesis. Interestingly, GAUT1 has been shown to be proteolytic processed from its transmembrane anchor domain and its catalytic domain is retained by GAUT7, thus ensuring biosynthesis of HG in the Golgi apparatus. Many methods exist in identifying protein-protein interaction (PPI) but many of these are developed for other organisms than plants and are most applicable for PPI detection in other organelles than the Golgi apparatus where pectin biosynthesis occurs.

In this work, different PPI detection methods are examined for their ability to detect PPI inside the Golgi lumen. The first method tested was the commercially available splitubiquitin system from Dualsystems Biotech AG. This was applied to test binary interactions between proteins involved in HG and Rhamnogalacturonan I (RG-I) biosynthesis (see Manuscript II and Manuscript V). The split-ubiquitin system was also used to perform a library interaction screen, in an unbiased way, to establish a HG biosynthetic interaction network (see Manuscript III). In addition to the split-ubiquitin system, we developed a Renilla Luciferase Protein Fragment Complementation Assay (Rluc-PCA) in Nicotiana benthamiana (See Manuscript IV) to perform binary interaction screening in a mid- to high-throughput manner. Mutants of gaut7 knockout grow normally (Manuscript V). This led us to hypothesize additional anchors of GAUT1 may exit. Based on subcellular localization and homology to GAUT7, we tested GAUT4, GAUT5, GAUT6 and GAUT9 for their ability to retain GAUT1 in the Golgi lumen. By using heterologous co-expression in N. benthamiana with fluorescence tagging and the split-ubiquitin system we found that GAUT5 and GAUT6 also formed complex with GAUT1. Double and triple mutant of gaut5, gaut6 and gaut7 were generated and revealed that the double knockout mutant of gaut6 gaut7 had skewed segregation and that a gaut5 gaut6 gaut7 homozygous triple mutant could not be obtained suggesting that no additional anchors of GAUT1 exist in Arabidopsis. This led us to hypothesize that the lack of these anchor proteins in the triple knockout mutant is gametophyte lethal. Backcrosses indicated that the male gametophyte was responsible for the lethality. Non-processing GAUT1 chimeras was generated and inserted into the gaut5 gaut6 gaut7/GAUT7 triple mutant and will be investigated if the chimeras can rescue the lethal phenotype. These results suggest that a PPI network in pectin biosynthesis may exist and plays an importance role in plant reproduction.

The work presented in this thesis will benefit future understanding of plant cell wall biosynthesis and aid the PPI discovery within cell wall biosynthesis, which in turn can benefit industrial uses of pectin.