Glycosylphosphatidylinositol anchoring: control through modification.

Since their discovery (Low and Saltiel, 1988), glycosylphosphatidylinositol-anchored proteins (GPI-APs) have provoked intense interest as crucial regulators for growth, morphogenesis, reproduction, and disease pathogenesis in organisms ranging from yeast and trypanosomes to animals and plants. The lipid moiety, the glycosylphosphatidylinositol (GPI) anchor, is synthesized in the endoplasmic reticulum (ER); the protein component is cotranslationally inserted into the ER and posttranslationally modified by the addition of a GPI anchor (Kinoshita et al., 2013; Fig. 1). GPI-APs are then transported via the Golgi to the outer surface of the plasma membrane. The lipid anchor mediates stable attachment of these proteins to the cell surface, where some play important roles as signaling regulators from sphingolipidand sterol-enriched membrane microdomains (Simons and Gerl, 2010). Some GPI-APs are released from the cell membrane by phosphatidylinositol-specific phospholipases to the extracellular matrix, where they might engage in processes such as cell adhesion and cell-cell communication. In Arabidopsis (Arabidopsis thaliana), there are about 250 predicted GPI-APs (Borner et al., 2003), a relatively large number compared with about 150 in mammals and 50 in the budding yeast (Saccharomyces cerevisiae). Important functions for plant GPI-APs have been elucidated through the study of individual proteins, such as the COBRA family in cell expansion and cell wall biosynthesis (Brady et al., 2007), ARABINOGALACTAN PROTEIN18 in megagametogenesis (Demesa-Arévalo and Vielle-Calzada, 2013), and LORELEI in the pollen tube-female gametophyte interaction (Capron et al., 2008; Tsukamoto et al., 2010; Duan et al., 2014). However, it is the studies of mutants defective in GPI biosynthesis that underscore the general importance of GPI-APs as a class: lacking the capacity to assemble the anchor is lethal. The GPI anchor is synthesized by an elaborate biosynthetic pathway, starting on the cytoplasmic side of the ER and ending with a completely assembled core anchor on the lumenal surface of the ER (Fig. 1). Prior to their transport out of the ER, proteins destined for GPI modification are cleaved at a C-terminal signature sequence by a GPI transamidase complex that in two enzymatic steps concomitantly attaches a GPI anchor to the C terminus of processed proteins (Kinoshita, 2014). Most of the knowledge on GPI biosynthesis and GPI-AP modification is derived from studies in mammals and yeast, but the pathway is likely to be conserved in plants (Ellis et al., 2010). In a recent article in Plant Physiology, Dai et al. (2014) reported that a GPI anchor biosynthesis mutant, abnormal pollen tube guidance1 (atpg1), displays both embryo lethality and severely depressed male fertility. They determined that APTG1 is homologous to the yeast GPI10 and human PIG-B (for phosphatidylinositol glycan anchor biosynthesis) proteins, the last of three distinct mannosyltransferases that modify the precursor anchor (Fig. 1), and showed that APTG1 can functionally substitute for GPI10 in a conditionally lethal gpi10 yeast mutant. Previous studies have demonstrated that Arabidopsis SETH1 (a male fertility god in Egyptian mythology), SETH2, and PEANUT1 (PNT1), encoding homologs of mammalian PIG-C, PIG-A, and PIG-M (Fig. 1) and their corresponding yeast counterparts, respectively, are important for male fertility (Lalanne et al., 2004; Gillmor et al., 2005). In addition, loss of the first mannosyltransferase in the pathway in pnt1 results in early seedling lethality. pnt1 embryos are severely defective, displaying various cell division anomalies and exhibiting altered levels and ectopic deposition of cell wall polymers. The results reported by Dai et al. (2014), therefore, further demonstrate