A developmentally regulated hydroxyproline-rich glycoprotein in maize pericarp cell walls

We have studied the accumulation of peptidyl hydroxyproline in the pericarp of developing maize (Zea mays L., Golden cross Bantam sweet corn) kernels. Although this hydroxyproline accumulates throughout development, it is most soluble and its content per milligram dry weight greatest at midmaturation stages of development. Salt-soluble proteins containing this hydroxyproline from isolated cell walls of developing kernels were fractionated on a CsCl density gradient and on a Chromatofocusing column, resulting in the purification of an hydroxyproline-rich glycoprotein, PC-1. PC-1 is a basic protein of approximately 65 to 70 kilodaltons in molecular weight with an isoelectric point of at least 10.2 and a density of 1.38 to 1.39 in CsCl. Amino acid composition data indicate that it is rich in hydroxyproline, threonine, proline, lysine, and glycine. Its relation to dicot extensin is discussed.     

Biochemical and tissue print analyses of hydroxyproline-rich glycoproteins in cell walls of sporophytic maize tissues

In an effort to understand the role of hydroxyproline-rich glycoproteins (HRGPs) in plant cell wall structure, we studied the distribution and physical properties of PC-1-like proteins (PC-1 being the major pericarp HRGP) throughout sporophytic tissues of two maize (Zea mays L.) varieties. We determined total amounts of hydroxyproline, an indicator of HRGPs, and did tissue print and Western blot analysis. We found hydroxyproline in cell walls of stems, leaves, roots, tassels, and silks. We also observed reactivity of anti-PC-1 monoclonal antibodies with anatomical prints of these tissues on nitrocellulose paper. Stem nodes and silks contained the most hydroxyproline and exhibited the strongest reaction with the antibody. PC-1 was localized in vascular bundles and the epidermis of stem tissue. However, localization to a specific cell type in the silk could not be determined at the resolution of the tissue print. The stem node protein had the same electrophoretic mobility as the pericarp protein as determined on Western blots prepared from cationic neutral gels. Protein extracts from silk tissues of both varieties studied contained one protein of the same size/charge as that found in pericarp, as well as some minor variant bands. The data presented here document that cell wall proteins are present in many tissues of the maize plant, although they are primarily in cell types contributing to support.     

Peptides isolated from cell walls of Medicago truncatula nodules and uninfected root

The hydroxyproline-rich root nodules of legumes provide a microaerobic niche for symbiotic nitrogen-fixing Rhizobacteria. The contributions of the cell wall and associated structural proteins, particularly the hydroxyproline-rich glycoproteins (HRGPs), are therefore of interest. Our approach involved identification of the protein components by direct chemical analysis of the insoluble wall. Chymotryptic peptide mapping showed a "P3-type" extensin containing the highly arabinosylated Ser-Hyp4-Ser-Hyp-Ser-Hyp4-Tyr3-Lys motif as a major component. Cell wall amino acid analyses and quantitative hydroxyproline arabinoside profiles, predominantly of tri- and tetraarabinosides, confirmed this extensin as the major structural protein in the cell walls of both root nodules and uninfected roots. On the other hand, judging from the Pro, Glu and non-glycosylated Hyp content, the nodule-specific proline-rich glycoproteins, such as the early nodulins (ENOD-PRPs), are present in much lesser amounts. Although we isolated no PRP peptides from nodule cell walls, a single PRP peptide from root cell walls confirmed the presence of a PRP in roots and represented the first direct evidence for a crosslinked PRP in muro. Compared with root cell walls (approximately 7% protein dry weight) nodule cell walls contained significantly more protein (approximately 13% dry weight) with an overall amino acid and peptide composition indicating the presence of structural protein unrelated to the HRGPs.     

A second extensin-like hydroxyproline-rich glycoprotein from carrot cell walls

The insoluble extensin matrix of dicot cell walls has been studied most fruitfully by examining the salt-extractable precursors to this matrix. Multiple extensin-like hydroxyproline-rich glycoproteins (HRGPs) have been isolated, or their existence inferred, from tomato, potato, bean, soybean, melon, carrot, and other plants. We and others previously have studied a carrot extensin which we call extensin-1. Here we report on the properties of extensin-2, a second salt-extractable hydroxyproline-rich glycoprotein from carrot. Like extensin-1, extensin-2 contains large amounts of hydroxyproline, serine, histidine, and lysine. In contrast, its tyrosine content is only about one-third that of extensin-1. Arabinose and galactose are the most abundant neutral sugars in both proteins, and nearly identical buoyant densities in CsCl suggest a similar proportion of carbohydrate in each. The size of extensin-2 is about half the size of extensin-1 based on: (a) the measured lengths of shadowed molecules (about 40 versus 84 nanometers); (b) the migration of extensin-2 in acid-urea gels relative to monomers, dimers, and trimers of extensin-1; and (c) the Stokes' radii of these molecules as determined by gel filtration chromatography. Electron microscopy of shadowed extensin-2 molecules indicates that they contain kinks, which may indicate the presence of intramolecular isodityrosine cross-links, but intermolecular cross-links, either with other extensin-2 molecules or extensin-1 molecules, are observed rarely if ever.     

Antibody localization of extensin in cell walls of carrot storage roots

The accumulation and cross-linking of hydroxyproline-rich glycoproteins (HRGPs) in cell walls of dicotyledonous plants has been correlated with a number of wall-strengthening phenomena. Polyclonal antibodies raised against glycosylated extensin-1, the most abundant HRGP in carrot (Daucus carota L.) cell walls, recognize this antigen on gel and dot blots and on thin sections of epoxy-embedded carrot-root cell walls. Since wall labeling can be largely reduced by preincubating the antibodies with purified extensin-1, most labeling can be attributed to recognition of this antigen. The remaining label may be the result of recognition of extensin-2, a second carrot HRGP, or other wall components (cellulose, hemicellulose and pectin are not recognized). Extensin-1 label was distributed quite uniformly across the cell wall but was absent from the expanded middle lamella at the intersection of three or more cells and was reduced in the narrow middle lamella between two cells. This distribution is essentially the same as that of cellulose. Because of limitations of this labeling technique, it is not possible to construct a complete model of the structure of the cross-linked extensin matrix. Nonetheless, short, linear arrays of gold particles may represent small portions of the extensin matrix or of individual extensin molecules as they are exposed on the surface of sections. These and other results presented here indicate that: a) newly synthesized extensin is added to the wall by intussusception; b) extensin cannot cross the middle lamella separating the walls of adjacent cells; and c) incorporation of extensin is a late event in the development of phloem-parenchyma cell walls in carrot.Abbreviations dE-1 antibodies antibodies raised against deglycosylated extensin 1 - ELISA enzyme-linked immunosorbant assay - gE-1 antibodies antibodies raised against glycosylated extensin 1 - HRGP hydroxyproline-rich glycoprotein - PAGE polyacrylamide gel electrophoresis - RG-1 rhamnogalacturonan I - SDS sodium dodecyl sulfate     

Solubilization of covalently bound extensin from capsicum cell walls

Acidified sodium chlorite cleaves isodityrosine and solubilizes covalently bound hydroxyproline-rich material from cell walls. This has been taken as evidence that isodityrosine acts as a cross-link holding the hydroxyproline-rich glycoprotein extensin in the cell wall. However, acidified chlorite was found to cleave peptide bonds in salt-soluble extensin and in bovine serum albumin (BSA). This invalidates the use of conventional acidified chlorite treatment to provide evidence for isodityrosine cross-links. The ratio of BSA:chlorite was important in determining peptidyl cleavage. At a ratio of 0.75:1.00 (mole amino acid residues/mole chlorite), or higher, peptidyl cleavage was not detected. Furthermore, in samples where a low concentration of radioactive extensin was present, BSA substantially protected the peptide bonds of the extensin against peptidyl cleavage during treatment with acidified chlorite, while not preventing the cleavage of isodityrosine. Therefore, acidified sodium chlorite plus BSA was a more specific reagent for the cleavage of isodityrosine than was acidified chlorite alone. This modified treatment solubilized in intact form the `covalently bound' extensin from cell walls of Capsicum frutescens (chili pepper) suspension cultures, providing new evidence compatible with the view that extensin molecules are held in the cell wall by isodityrosine cross-links.     

Purification and Partial Characterization of Tomato Extensin Peroxidase

Early plant defense response is characterized by elevation of activity of peroxidases and enhanced insolubilization of hydroxyproline-rich glycoproteins, such as extensin, in the cell wall. The insolubilization process (cross-linking between soluble extensin precursor molecules) is catalyzed by extensin peroxidases. We have ionically eluted extensin peroxidases from intact water-washed suspension-cultured tomato (hybrid of Lycopersicon esculentum Mill. and Lycopersicon peruvianum L. [Mill.]) cells and purified them to homogeneity by molecular sieve and cation-exchange chromatography. Four ionic forms of peroxidase (PI,PII,EPIII, and EPIV) were resolved; only the latter two cross-linked tomato soluble extensin. The molecular weight (34,000-37,000), amino acid composition, and isoelectric point (9.0) of the extensin peroxidases were determined. Substrate specificities of the enzymes were investigated: soluble extensin and potato lectin (a hydroxyproline-rich glycoprotein with a domain that strongly resembles extensin) were cross-linked by only two forms of the enzyme, whereas bovine serum albumin, aldolase, insulin, a number of other marker proteins, and proteins eluted from tomato cells (except extensin) could not be cross-linked. We have also isolated a yeast elicitor that enhances total peroxidase activity and extensin insolubilization within 1 h of challenge in cultured cells of tomato. A highly sensitive enzyme-linked immunosorbent assay technique using polyclonal antiserum raised against soluble tomato extensin was used to demonstrate extensin insolubilization in vivo. A tomato cell-wall peroxidase that cross-links extensin has been purified and may have a role in plant defense.     

A repetitive proline-rich protein from the gymnosperm douglas fir is a hydroxyproline-rich glycoprotein

Intact cell elution of suspension cultures derived from Douglas fir, Pseudotsuga menziesii (Mirbel) Franco, yielded two extensin monomers, the first hydroxyproline-rich glycoproteins (HRGPs) to be isolated from a gymnosperm. These HRGPs resolved on Superose-6 gel filtration. The smaller monomer was compositionally similar to angiosperm extensins like tomato P1. The larger monomer had a simple composition reminiscent of repetitive proline-rich proteins (RPRPs) from soybean cell walls and contained proline, hydroxyproline, and sugar; hence designated a proline-hydroxyproline-rich glycoprotein (PHRGP). The simple composition of the PHRGP implied a periodic structure which was confirmed by the simple chymotryptic map and 45-residue partial sequence of the major proline-hydroxyproline-rich glycoprotein chymotryptide 5: Lys-Pro-Hyp-Val-Hyp-Val-Ile-Pro-Pro-Hyp-Val-Val-Lys-Pro-Hyp-Hyp-Val- Tyr-Lys-Pro-Hyp-Val-Hyp-Val-Ile-Pro-Pro-Hyp-Val-Val-Lys-Pro-Hyp-Hyp- Val-Tyr-Lys-Ile-Pro-Pro(Hyp)-Val-Ile-Lys-Pro. Proline-hydroxyproline-rich glycoprotein chymotryptide 5 contained an 18-residue tandem repeat devoid of tetra(hydroxy)-proline or serine; it also contained two instances of the five-residue motif Hyp-Hyp-Val-Tyr-Lys and five of the general Pro-Pro-X-X-Lys motif, thereby establishing its homology with typical angiosperm RPRPs and extensins from tomato, petunia, carrot, tobacco, sugar beet, and Phaseolus. Unlike the nonglycosylated soybean RPRP, the highly purified Douglas fir PHRGP was lightly glycosylated, confirmed by a quantitative hydroxyproline glycoside profile, indicating that extensins can range from highly glycosylated hydroxyproline to little or no glycosylated hydroxyproline. Comparison of extensin sequence data strongly indicates that a major determinant of hydroxyproline glycosylation specificity is hydroxyproline contiguity: extensins with tetrahydroxyproline blocks are very highly arabinosylated (>90% hydroxyproline glycosylated), tri- and dihydroxyproline are less so, and single hydroxyproline residues perhaps not at all. Despite high yields of extensins eluted from intact cells, the Douglas fir cell wall itself was hydroxyproline poor yet remarkably rich in protein (>20%), again emphasizing the existence of other structural cell wall proteins that are neither HRGPs nor glycine-rich proteins.     

A gymnosperm extensin contains the serine-tetrahydroxyproline motif

The extensin family is a diverse group of hydroxyproline-rich glycoproteins located in the cell wall and characterized by repetitive peptide motifs glycosylated to various degrees. The origin of this diversity and its relationship to function led us earlier to compare extensins of the two major groups of angiosperms from which we concluded that the highly glycosylated Ser-Hyp₄ motif was characteristic of advanced herbaceous dicots, occurring rarely or not at all in a representative graminaceous monocot (Zea mays) and a chenopod (Beta vulgaris) representative of primitive dicots. Because these results could arise either from loss or acquisition of a characteristic feature, we chose a typical gymnosperm representing seed-bearing plants more primitive than the angiosperms. Thus, salt eluates of Douglas fir (Pseudotsuga menziesii) cell suspension cultures yielded two monomeric extensins differing in size and composition. The larger extensin reported earlier lacked the Ser-Hyp₄ motif, was rich in proline and hydroxyproline, and contained peptide motifs similar to the dicot repetitive proline-rich proteins. The smaller extensin monomer reported here (Superose-6 peak 2 [SP2]) was compositionally similar to typical dicot extensins such as tomato P1, mainly consisting of Hyp, Thr, Ser, Pro, Val, Tyr, Lys, His, abundant arabinose, and a small but significant galactose content. A chymotryptic peptide map (on Hamilton PRP-1) of anhydrous hydrogen fluoride-deglycosylated SP2 yielded eight peptides sequenced after further purification on a high-resolution fast-sizing column (polyhydroxyethyl aspartamide; Poly LC). Significantly, two of the eight peptides contained the Ser-Hyp₄ motif, consistent both with the SP2 amino acid composition as well as the presence of hydroxyproline tetraarabinoside as a small (4% of total Hyp) component of the hydroxyproline arabinoside profile; thus, hydroxyproline tetraarabinoside corroborates the presence of Ser-Hyp₄, in agreement with our earlier observation that Hyp contiguity and Hyp glycosylation are positively correlated. Interestingly, other peptide sequences indicate that SP2 contains motifs such as Ser-Hyp₃-Thr-Hyp-Tyr, Ser-Hyp₄-Lys, and (Ala-Hyp)_n repeats that are related to and typify dicot extensins P1, P3, and arabinogalactan proteins, respectively. Overall, these peptide sequences confirm our previous prediction that Ser-Hyp₄ is indeed an ancient motif and also strongly support our suggestion that the extensins comprise an extraordinarily diverse, but nevertheless phylogenetically related, family of cell wall hydroxyproline-rich glycoproteins.     

Effect of a Fungal Disease on Extensin, the Plant Cell Wall Glycoprotein

The hydroxyproline-rich plant cell wall glycoprotein, extensin,shows a parallel increase of its arabinose and hydroxyprolinecontents in melon plants infected with anthraonose fungus. Thesevariations are located in the hydroxyproline arabinosides whichcharacterize this macromoleeule, particularly in the hydroxyprolinetetra-arabinosides and tri-arabinosides.     

Serine-O-galactosyl linkages in glycopeptides from carrot cell walls

Cell walls obtained from carrot disks aged for 6 days were treated with mild acid to remove the arabinosyl sidechains from the hydroxyproline residues of extensin, and subsequently digested with trypsin. The peptides in the tryptic digest were fractionated according to MW by gel filtration and further purified with Dowex 5OX2. The peptides were rich in hydroxyproline and contained small amounts of carbohydrate, especially galactose. Treatment of the glycopeptides with NaOH in the presence of Na₂SO₃ resulted in a considerable loss of serine residues (up to half in some fractions) and the formation of cysteic acid. Free carbohydrate, consisting mostly of galactose was released by this treatment. Treatment with NaOH in the presence of NaBH₄, resulted in the release of carbohydrate sidechains which primarily contained galactitol and galactose. The data indicate that the serine-_O-galactosyl linkage occurs in glycopeptides of different sizes and is most abundant in the hydroxyproline-rich glycopeptide fractions.     

Al-induced cell wall hydroxyproline-rich glycoprotein accumulation is involved in alleviating Al toxicity in rice

Cell wall components such as pectin and hemicelluloses have been proposed to be involved in aluminum resistance mechanisms in plants. However, whether hydroxyproline-rich glycoproteins (HRGPs), one of the most abundant proteins of the cell walls, are involved in Al resistance mechanisms remains elusive. In this study, two rice cultivars Xiushui 03 (Al resistant) and Xiushui 128 (Al sensitive) significantly differing in Al resistance were identified. In the absence of Al, no significant difference was observed in contents of glycoproteins and hydroxyproline in cell wall fractions of these two cultivars. At the early stage of Al toxicity, glycoproteins and hydroxyproline were significantly induced in these two cultivars, but levels of their accumulation in cell walls were much higher in cv. Xiushui 03 than in cv. Xiushui 128. At the late stage of Al toxicity, their accumulation in cell walls dramatically decreased in cv. Xiushui 128 and, however, still kept a high level in cv. Xiushui 03. The finding that Al-induced changes of glycoproteins and hydroxyproline were completely consistent indicates that Al-induced glycoproteins are HRGPs. Further observation utilizing transmission electron microscope showed that HRGPs were greatly accumulated in cell walls leading to thickening of cell walls in cv. Xiushui 03, however, HRGPs and cell walls greatly decreased in cv. Xiushui 128. These data suggest that Al-induced HRGP accumulation in cell walls is involved in alleviating Al toxicity in rice.