A, Arrows point at tapetum within the anther. developing microspores showed that the earliest detectable defect was in primexine formation. Furthermore, wild-type microspores contained primexine-localized epitopes indicative of the presence of xylan, but these were absent in phenotype of a mutant in (((((At1g33430) is usually coexpressed with genes in the sporopollenin synthesis pathway (de Azevedo Souza et al., 2009; Dobritsa et al., 2011; Grienenberger et al., 2010; Kim et al., 2010) and encodes a carbohydrate-active enzyme GT31 family protein (Campbell et al., 1997; Qu et al., 2008). UPEX1 is positioned in group A with Arabidopsis galactosyltransferase (GalT) proteins (Qu et al., 2008), some of which have been characterized as arabinogalactan protein (AGP) GalTs (Geshi et al., 2013; Ogawa-Ohnishi and Matsubayashi, 2015), suggesting that UPEX1 could have a similar GalT activity in AGP galactosylation. The ((At1g27600) encodes the previously characterized IRREGULAR XYLEM9-LIKE (IRX9L) GT family 43 protein (Wu et al., 2010). is usually a paralog of double mutants (Wu et al., 2010). Here, we tested the hypothesis that and function in exine patterning in developing Arabidopsis microspores by modulating the synthesis or modification of components of the primexine wall layer. Detailed analysis of the and mutant phenotypes over the course of anther development indicated that both mutants are defective in primexine formation at the early tetrad stage and that exine patterning defects are evident immediately thereafter. We used immunolabeling to show that xylan epitopes are present at the surface of developing microspores in wild-type anthers, suggesting that this primexine contains this polysaccharide. While mutants lack detectable xylan in the walls of developing microspores, mutants exhibit enhanced cross reactivity to an antixylan antibody, suggesting a change in primexine cell wall structure with increased abundance or enhanced accessibility of the xylan epitope. Promoter-reporter gene assays and genetic analyses were consistent with functions of and in sporophytic tapetal cells. These data support a model in which xylan and AGP cell wall components synthesized in the tapetum are incorporated into the primexine wall and play roles in primexine development and anchoring of sporopollenin to the microspore surface early in microspore development. RESULTS Loss-of-Function Alleles of GTs At1g33430 and At1g27600 A large-scale screen for Arabidopsis mutants affecting pollen exine formation identified (At1g33430) and (At1g27600) as Arabidopsis genes encoding GTs involved in exine patterning (Dobritsa et al., 2011). We obtained the previously identified SNJ-1945 loss-of-function T-DNA insertion allele (Salk_037323; Wu et al., 2010), and two previously described T-DNA insertion alleles (Salk_091466 insertion in exon 1; Sail_544-C02, insertion in exon 3; Dobritsa et al., 2011), which we renamed and and lack wild-type transcripts and are loss-of-function alleles, and that has a flower-specific expression pattern (Supplemental Figs. S1 and S2). Two transcripts, and (Supplemental Fig. S3). The and alleles were used for all subsequent analyses. Homozygous and plants grew normally (Supplemental Figs. S1 and S2) and showed no reductions in fertility (Supplemental Table S1). We initially analyzed and pollen phenotypes by staining mature pollen grains with Auramine O and viewing them with fluorescence microscopy (Grienenberger et al., 2010; Kim et al., 2010; Dobritsa et al., 2011) to reveal any differences in exine structure and patterning. Using this method, was found to produce pollen without the regular reticulate and net-like exine pattern common of wild-type pollen (Fig. 1). This confirms the phenotype of pollen reported by Dobritsa et al. (2011). Light microscopy of toluidine blue-stained wild-type and microspores at the free microspore Mouse monoclonal to STAT3 stage after exine deposition (Fig. 1) indicated that this microspores had comparable morphology to wild type at this stage. We also SNJ-1945 imaged pollen using two-photon (2P) microscopy, which allows live cell imaging of microspores in intact anthers using the intrinsic autofluorescence of exine under UV excitation (Quilichini et al., 2014b, 2015b). As seen in Physique 1, microspores appeared to have a thinner pollen wall compared to wild type with less intense emission, but the exine showed similar autofluorescence. No changes in the autofluorescence of tapetal cells or the locular matrix were observed. Open in a separate window Physique 1. and pollen phenotypes. A to C, Wild-type pollen; D to F, pollen; G to I, pollen. SNJ-1945 A, C, and D, Auramine-O-stained mature pollen released from anthers viewed by fluorescence microscopy. B, E, and H, Light micrographs of images of chemically.
