Functional identification of an Arabidopsis pectin biosynthetic homogalacturonan galacturonosyltransferase

Abstract

Galacturonosyltransferases (GalATs) are required for the synthesis of pectin, a family of complex polysaccharides present in the cell walls of all land plants. We report the identification of a pectin GalAT (GAUT1) using peptide sequences obtained from Arabidopsis thaliana proteins partially purified for homogalacturonan (HG) alpha-1,4-GalAT activity. Transient expression of GAUT1 cDNA in the human embryonic kidney cell line HEK293 yielded uridine diphosphogalacturonic acid:GalAT activity. Polyclonal antibodies generated against GAUT1 immunoabsorbed HG alpha-1,4-GalAT activity from Arabidopsis solubilized membrane proteins. blast analysis of the Arabidopsis genome identified a family of 25 genes with high sequence similarity to GAUT1 and homologous genes in other dicots, in rice, and in Physcomitrella. Sequence alignment and phylogenetic Bayesian analysis of the Arabidopsis GAUT1-related gene family separates them into four related clades of GAUT and GAUT-like genes that are distinct from the other Arabidopsis members of glycosyltransferase family 8. The identification of GAUT1 as a HG GalAT and of the GAUT1-related gene family provides the genetic and biochemical tools required to study the function of these genes in pectin synthesis.

Fig. 1.

GAUT1 has GalAT activity. (A) GalAT activity in equal amounts of media from HEK293 cells transiently transfected with the JS33Δ1-43 (JS33) or JS36Δ1-41 (JS36) cDNA constructs or empty vector control and immunoabsorbed with anti-HA antibodies conjugated to protein A-Sepharose. Solubilized Arabidopsis protein with GalAT activity (Inset) was used as a positive control. The average of time-0 reactions (background) is shown. Data are the average (±SE) of duplicate samples from 60-min reactions. Data were analyzed for significance compared with empty vector controls by using a two-sample, one-tailed Student's t test and an a priori α of 0.05 [95% confidence; t_JS36 = 10.284 (P ≈ 0.005); t_JS33 = 0.180 (P > 0.25)]. Comparable results were obtained in three separate experiments. (B) Immunoabsorption of GalAT activity from Arabidopsis solubilized membrane proteins by using GAUT1 antiserum. Solubilized Arabidopsis membrane proteins partially purified by SP-Sepharose chromatography were incubated with increasing amounts of GAUT1 antiserum-coated Dynabeads, and proteins immunoabsorbed by anti-GAUT1 antibodies were magnetically separated from the SP-Sepharose fraction. GalAT activity was measured in both anti-GAUT1 immunoabsorbed (black boxes) and depleted (open diamonds) fractions and compared with similar fractions obtained by using preimmune serum (Inset). Open bar, preimmune serum immunodepleted fraction; filled bar, preimmune serum immunoabsorbed fraction.

Fig. 2.

Characterization of products made by anti-GAUT1 immunoabsorbed protein. (A) Sensitivity of synthesized product to cleavage by EPG. Products synthesized during 2-h reactions containing buffer, UDP-[¹⁴C]GalA, OGA acceptors, and the Arabidopsis SP-Sepharose fraction, the SP-Sepharose fraction after immunodepletion of GAUT1 (GAUT1 depleted), or the anti-GAUT1 immunoabsorbed material (GAUT1 enriched). Each fraction was incubated overnight with water, boiled EPG, or native EPG. Radiolabeled products recovered by using the filter assay are shown. Note that small oligomers (e.g., monomer and dimer) do not bind to the filters. (B) Separation of nonradioactive GalAT reaction products by electrophoresis on a 30% polyacrylamide gel. Lanes 2–4 show the products recovered after incubation of SP-Sepharose purified solubilized Arabidopsis membrane proteins with UDP-GalA and OGAs enriched for a DP 13 for 0, 3, and 22 h at 30°C. Lanes 6–8 show the same series of incubation times with the SP-Sepharose fraction after immunodepletion with GAUT1 antiserum. Lanes 10–12 show similar series with the product synthesized by GAUT1-immunoabsorbed protein from the SP-Sepharose fraction. Lanes 5, 9, and 13 show product recovered after digestion of the respective 22-h reaction products with EPG. Lane 1 shows 0.1 μg of OGA standard of DP 7–23.

Fig. 3.

Characterization of the Arabidopsis GAUT1-related gene superfamily. (A) Schematic representation of domain structure and conserved amino acid motifs in the Arabidopsis GAUT and GATL proteins. Positions of conserved residues found in the GAUT family are numbered relative to GAUT1. Residues 22–44 represent a predicted transmembrane region. The positions of conserved residues in the GATL family are numbered relative to GATL1. The conserved amino acid residues within the motifs diagnostic for the GAUT1-related superfamily and the GAUT and GATL subfamilies are shown. (B) Phylogenetic analysis of the GAUT1-related superfamily in A. thaliana. Alignment of the complete sequences of all 25 members of the GAUT1-related superfamily was carried out with clustalx (27) using parameters suggested by Hall (44). Bayesian analysis employing the program mrbayes (45, 46) was used to infer phylogenetic relationships among the members of the superfamily and to group the protein sequences into related clades. The analysis was carried out for 100,000 generations using a mixture of amino acid transition parameter models. The phylogram presented is the majority rule tree. Percentage clade credibility values for each branch are given in parentheses.