mum; Winn Lm i" (ll ulu lzll "will! ll ‘ “I p'q‘c R” This is to certify that the thesis entitled DEGLYCOSYLATION 0F ARABINOGALACTAN PROTEINS FROM SUSPENSION-CULTURED SYCAMORE CELLS VIA ‘HYDROGEN. FLUORIDE IN PYRIDINE presented by Yukio Akiyama has been accepted towards fulfillment of the requirements for Adm—degree inhiachmany 3“ M: c!) . Major professor Date May 10, 1979 0-7639 OVERDUE FINES ARE 25¢ PER DAY PER ITEM Return to book drop to remove this checkout from your record. lvl|l.|‘ll'{ Jia‘f Isl-‘2'- DEGLYCOSYLATION OF ARABINOGALACTAN PROTEINS FROM SUSPENSION-CULTURED SYCAMORE CELLS VIA HYDROGEN FLUORIDE IN PYRIDINE By Yukio Akiyama A THESIS Submitted to ‘Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF SCIENCE Department of Biochemistry 1979 ABSTRACT DEGLYCOSYLATION or ARABINOGALACTAN PROTEINS FROM SUSPENSION-CULTURED SYCAMORE CELLS VIA HYDROGEN FLUORIDE IN PYRIDINE By Yukio Akiyama In addition to the hydroxyproline in cell wall protein "extensin", cultured plant cells secrete soluble arabinogalactan proteins (AGPs), which also contain hydroxyproline. Because the high carbohydrate content of AGPs thwarts direct attempts at amino acid sequencing, they must be deglycosylated first. Crude AGPs isolated from the medium of suspension-cultured sycamore cells were deglycosylated via 70% HF in pyridine (HF/pyr), which is easier to handle than liquid HF and found to be as efficient as HF. One hour HF/pyr treatment at room temperature removed over 90% of sugars from AGPs. HF/pyr deglycosylated AGPs were partially purified by gel-filtration on Sephadex G-100 and by ion-exchange chromatography on SP-Sephadex C-SO. The final fraction contained hydroxyproline, serine, alanine and aspartate as the predominant amino acids. This material was gel-electrophoresed. It stained poorly with coomassie blue, however, it was detected either by prelabelling the protein with fluorescein isothiocyanate or by labelling with 14C-proline. ACKNOWLEDGMENTS The advice, assistance, encouragement and collaboration of Dr. U. V. Mani is most gratefully acknowledged. Thanks are also to Fumiko, Sharon, Joan, Barbara, Jim and to my committee members Dr. D. Delmer who is my academic adviser, Dr. P. Kindel and especially Dr. D. T. A. Lamport for many useful discussions and suggestions. Finacial support from.The Japan Tobacco & Salt Public Corporation is thankfully recognised. ii TABLE OF CONTENTS Page LIST OF TuLESQQQQQOOOOOOQQQQO ........ .0............. v LIST OF FIGURES.........0.00..0.... ..... 0......0..... v1 LIST OF ABBREVIATIONS.....0.......0.................. Viii INTRODUmION....0...00....0........0..0.......000.... 1 MATERIALS AND m0D8.00.0. .......... .0.............. 14 Materials.0..........0.............0............ 14 Crude AGPs Preparation.......................... 14 From culture Med1m................0...... . 14 From Cytoplasm.0........0....0..0......0... 14 Deglycosylation with HF/pyridine................ 15 Deglycosylation With HF.....................0..0 16 Column Chromatography........................... 16 Disc-gel Electrophoresis of the Deglycosylated AGPs Fractions................... l6 HYdr01ys1s Of smleoooooooooooooooooooooooooooo 18 sugar AnaIYS18..........0..00............0...... 18 Amino Acid malys18..............0.............. 18 RESULTS....0......................................... 26 Chemical Composition of Crude AGP(M) and AGP(C). 26 Deglycosylation Time Course of AGP(M) With HF/pyridine..0............................. 26 Effect of Temperature on HF/pyr Treatment....... 33 Deglycosylation of AGP(M) via HF/pyridine....... 38 Partial Purification of Deglycosylated Hydroxyproline-containing Material......... 38 SDS Disc-gel Electrophoresis of Hydroxy- proline-rich Fractions..................... 46 HF Control Experiment........................... 50 Further Fractionation of HP/pyr InSOIuble Fraction.....................0 ........ 56 ACP(C) Deglycosylation via HF/pyridine .......... 60 DISCUSSIONS ......... . ........ ............ . ....... 66 iii TABLE OF CONTENTS--continued Page LIST OF REFERENCES.O...0..0.0.0...’..O..........0.0.0 7,0. iv LIST OF TABLES Table 10. ll. 12. 13. 14. Tomato Glycopeptides Obtained by Digestion of Tomato Cell Wall with "Cellulase"................. Extensin Tryptides................................... Sugar Analysis of Crude AGP(M) and AGP(C)............ Amino Acid Analysis of Crude AGP(M) and AGP(C)....... Hydroxyproline Recovery after Deglycosylation of AGP(M) at Different Time Intervals using Anhydrous HF/pyr plus 102 Methanol at Room Temperature......... Sugar Analysis of AGP(M) after Treatment with HF/pyr at Different Time Intervals................... Sugar Analysis of the HF/pyr Treated AGP(M) at 0°C and at Room Temperature (RT) for 1 hr................ Amin Acid Analysis of the HF/Pyr Treated AGP(M) at 0 C and at Room Temperature (RT).................. . Sugar Composition of Sephadex G-100 Fractions Of AGP(M)/HF/pyr/Super..0..00.0...................... A Comparison of the Chemical Compositions of Partially Purified Hydroxyproline—containing Fractions from Deglycosylated AGP(M)........................... Amino Acid Composition of Sycamore AGP(M) Peak III/ SP-C-SO before and after Electrophoresis (mole 2).... Amino Acid Composition of Standard Proteins (mole Z). Chemical Composition of AGP(M)/HF/pyr/Insoluble/ NaCl Extract/G-lOO III/Fractionated on SP-C-SO....... Chemical Composition of Major Peaks of HF/PYr Deglycosylated AGP(C) and ACP(M)......... ............ ‘7 Page 27 28 31 32 36 37 42 45 47 51 59 6S LIST OF FIGURES Figure Page 1. Separation of hydroxyproline-arabinosides by chromatography on chromobeads B..................... 5 2. Hydroxyproline tetraarabinoside..................... 6 3. Possible amino acid sequences in extensin........... 8 4. Gas liquid chromatography of sugars as their trimethylsilyl methyl glycosides.................... 20 5. Amino acid analysis by liquid chromatography........ 23 6. Gas liquid chromatography of amino acids as their heptafluorobutyryl isobutyl esters.................. 25 7. The kinetics of removal of sugars from AGP(M) at different tim intervals...................000... 30 8. The hydroxyproline-glycoside profile of deglycosy- lated AGP(M) at different time intervals on Biogel P2 column................................. 35 9. The flow sheet for the partial purification of hydroxyproline-rich material from AGP(M)..... ....... 39 10. Gel filtration of deglycosylated AGP(M) with HF/pyr on a Sephadex G-lOO.......................... 41 11. SP-Sephadex C-SO column chromatography of deglycosylated sycamore-maple AGP(M) with HF/pyr, G-lOO peak III fraction........... ......... . 44 12. Amount of radioactivity in gel slices.......... ..... 49 13. Gel filtration of HF-deglycosylated AGP(M) on sephadex G-loo..................0.....0...0..000. 53 14. SP-Sephadex C-50 column chromatography of HF-deglycosylated AGP(M), G-lOO peak II fraction.... 55 15. Gel filtration of HF/pyr insoluble material after extracting with 0.5 M NaCl.................... 58 vi LIST OF FIGURES--continued Figure Page 16. Gel filtration of deglycosylated AGP(C) with HF/pyr on a Sephadex G-lOO.............. ..... 62 17. SP-Sephadex C-50 column chromatography of deglycosylated AGP(C) with HF/PYr, G-lOO peak II fractionOCCC......C...............0 ...... . 64 vii AGPs, AGP(C), AGP(M) FITC HF HF/pyr Hyp Hyp-Arabs, Hyp-Aran (n;l-4) SDS SP-Sephadex C-SO (SP-C-SO) TCA LIST OF ABBREVIATIONS arabinogalactan proteins, (C, from cytoplasm, M, from medium) fluorescein isothiocyanate anhydrous hydrogen fluoride anhydrous hydrogen fluoride in pyridine hydroxyproline a short chain of arabinose (1-4) on the hydroxyl group of Hyp sodium dodecyl sulfate sulfopropyl Sephadex C-SO trichloroacetic acid viii INTRODUCTION Plant cells have a cell wall which is involved in several biological functions such as turgor and a resistance to pathogens. Thus the presence of wall is a primary barrier against disease and hence extensive research is being conducted in order to understand the complex cell wall network. Modern knowledge of the cell wall comes from chemical analysis studies, X-ray diffraction studies and work with the electron microscope. These studies led to the identification of cellulose, hemicellulose and pectic polysaccharides as the major component of the cell wall. Besides these components, Lamport (1) reported the presence of a hydroxyproline containing structural protein "extensin" in the cell wall. He suggested that extensin plays an important role in the structural rigidity of the cell wall and hence its role in the cellular extensive mechanism. By isolating and characterising hemicellulose and pectic polysaccharide fractions, Keegstra ggugl, (2) proposed a model for the structure of the primary cell wall of suspension-cultured sycamore cells. According to this somewhat hypothetical model, the cellulose microfibrils are cross-linked by the matrix polymers of the wall through a xyloglucan by means of hydrogen bonding. The xyloglucan is then linked at its reducing end to galactan which is attached to the rhamnogalacturonan of the wall. From the reducing end of rhamnogalacturonan is a 3-6 linked arabinogalactan which may be glycosidically linked to serine residues of the cell wall protein. 2 This model, which involves all but one of the wall polysaccharides in a covalently linked network (the exception being xyloglucan H-bonded to the cellulose microfibrils), is satisfyingly detailed except the link between extenisn and arabinogalactan. This linkage was hypothesised without direct evidence from studies on wall protein, but rather was derived from information concerning a soluble hydroxyproline-rich glycoprotein secreted into the medium. As will be discussed later, this soluble glycoprotein has been found to be an example of a new class of arabinogalactan proteins (3,4). Therefore, until more data are available from the work on the cell wall itself the method by which extensin is linked to wall polysaccharide must be regarded as undefined. Steward (5) showed the presence of bound hydroxyproline in the hydrolysates of alcohol insoluble material in the tissue cultures of carrot or potato. Later, experiments of Lamport and Northcote (6) with sycamore cells and Dougall and Shimbayashi (7) with tobacco cells from the cultured medium.showed that hydroxyproline is a major constituent in cell wall hydrolysates, accounting for about 30% of the total amino acids, the protein itself accounting for 2-102 of plant cell walls. Bound hydroxyproline was chemically characterised as 5322374-hydroxy- prroline. The distribution of hydroxyproline has been studied in various parts of plants. Thus it has been shown to be present in tissues like cotyledon, hypocotyl, leaf, coleoptile, pericap and root (1). The presence of small amounts of material in the cytoplasm has also been reported. Bound hydroxyproline is also present in the cell wall of green algae, brown algae (traces!) and in the certain forms of fungi. The red algae did not contain any hydroxyproline (8). The experiments of Lamport showed that wall protein is exceptionally resistant to proteolytic enzymes (1). But after heating the wall for 1 hr at pH 1, he observed that chymotrypsin liberated 402, and trypsin 262, of amino acids from the sycamore cell walls. He showed that a crude mixture of enzymes posissing carbohydrase and protease activities released hydroxyproline glycopeptides from the cell wall of tomato (9, Table 1). These glycopeptides contained arabinose O-glycosidically attached to hydroxyproline which is stable to alkali. Therefore alkaline hydrolysis with saturated Ba(0H)2 of the cell wall released a family of hydroxyproline arabinosides (Hyp-Arabs) (10, Figure 1). Most of the hydroxyproline residues in the tomato wall are O—substituted by tri- or tetra-saccharides of arabinose (11, Figure 2). The linkage was found to be 1&3 l-EDZ 152 134 Hyp by methylation analysis and pmr spectro- scopy (12). The molecular models showed that F-linked hydroxyproline tri- and tetra-arabinosides will conform to the type polyproline II helix which favors the formation of 3 hydrogen bonds in the peptide backbone (13). {i-Linked hydroxyproline tri- and tetra-arabinosides may provide a structural conformation to extensin what the triple helix does for collagen to produce a stable rod—like structure of high tensile strength. Although extensin has not been solubilised in an intact form, Lamport obtained high and low molecular weight peptides from the acid treated cell walls with trypsin. The low molecular weight fraction accounts for about 1/3 of the wall bound hydroxyproline and accounts for 5 tryptic peptides (Table 2) whose sequence totals 48 residues (4). These peptides probably represent either the complete extensin sequence or about 1/3 of the sequence (Figure 3). The peptides represent the complete sequence if all the high molecular weight non-sequenced material has the same sequence as the low molecular weight, but has become Table 1. Tomato Glycopeptides Obtained by Digestion of Tomato Cell Walls with "Cellulase". NHz-Terminus l. ara25 gal6 hyp10 ser3 tyr SER 2. ara14 gal3 hyplo ser3 lys2 thr val LYS 3. ara20 gal4 hyp9 ser3 lys tyr LYS 4. ara16 gal4 hyp9 ser3 tyr SER 5. ara16 gal2 hyp9 ser3 lys3 val tyr LYS .m muuonoaouso so >£nmuw0unaouno >9 novanonunmumloafiaouahxouwzn mo couumumeom .H madman m§2¥.2.m:F V n N - . . a a N I ”238208; I # mutt OI § 0.0 c n mzjommrxomorx E: own no .ovamoeanuummuumu ueaaounmxouvhm .N shaman exec/gr?” to to O :2 Table 2. Extensin Tryptides. SER-HYP-HYP-HYP-HYP-SER-HYP-SER-HYP-HYP-HYP-HYP-("TYR"-TYR)- LYS Exists with 3, 2, or 1 galactose residures. SERPHYP-HYP-HYP-HYP-SER-HYP-LYS Exists with 2, l, or 0 galactose residues; SER-HYP-HYP-HYP-HYP-THRrHYP-VAL-TYR-LYS Exists with l or 0 galactose residues. SER-HYP-HYP-HYP-HYP-LYS Exists with l or 0 galactose residues. SER-HYP-HYP-HYP-HYP-VAL~"TYR"-LYS-LYS Exists with l or 0 galactose residues. .nfiosouxm :H moosonuoo ovum onwam manwomom .m ounwwm A<> o>wum>auov ocamoumu aaoaxan I x mug mmfi I H mam MVH I M arm UJM> m "ovoo Ufio< ocafi< m=OUHme ooomfi .oo All - Iv _ < some azmmmmmno mozmaomm _ zzoze mozmacmm _ .mm xawamuouo scaamoouo b r mcouaov coon . mu - M , («ix —l < mozmoomm we mzem _ < mozmaomm m< mz smug ”mum awe»: umflaapom Hauaufieam _Mex>mmmmm_ .eemmem_ Hew>me===zmi Heemmmmzm. .ewxemmmmmmmmmmm_ ocfifioumzxouvhn Hams Hams nmm .mo you wowunnooom cocooauoo convene» we cross-linked during "maturation" similar to native cross-linked collagen or elastin. The most important notable findings in all of the extensin peptides is the occurrence of Ser-Hyp-Hyp-Hyp-Hyp (including sycamore- maple) and galactosylated serine residues (14). Lamport postulated that this is the region for the polysaccharide attachment of the cell wall. Later galactosyl serine was also found in the cell wall of carrot (15). Since methylation analysis of hydroxyproline-rich fracion of cell wall gave a lot of terminal arabinose residues, Keegstra g£_§l, (2) thought arabinogalactan is attached to protein extensin via the Ser-O-Gal linkage. Extracting cell walls with DMSO/HZOIEtOH/NaBHAINaOH, Lamport obtained some evidence for attachment of galactan (about 10 residues of galactose) to serine (4). Besides the covalently bound cell wall glycoprotein, there are at least two other kinds of macromolecules containing hydroxyproline in the plant kingdom; one is a "classical" lectin, the other being arabino- galactan proteins (AGPs) also called "all.fl-lectins". ("Classical lectins have at least two carbohydrate-binding sites. "A11.p-lectins" may have only one carbohydrate-binding site.) The first hydroxyproline containing lectin was isolated by Allen SE El- (16) from potato tubers. It is a glycoprotein consisting of 502 protein and 50% carbohydrate, the major sugar being arabinose (962) with small amounts of galactose (42). The molecular weight of the lectin was $0000. Hyp-Arabs and Ser-Gal linkages have been found in this glycoprotein and the linkage between hydroxyproline and arabinose was found to be a (17). Another lectin containing hydroxyproline occurs in the seeds of Jimson weed (18). It is a glycoprotein consisting of 72% protein and 282 carbohydrate, the major sugar is arabinose with small amount of 10 galactose and glucosamine. The hydroxyproline content of this protein is 6.3%. Besides these two lectins, the hydroxyproline containing protein isolated by Mani and Radhakrishnan from the leaves of sandal (19) is also a glycoprotein, consisting of 842 protein and 162 carbohydrate, the predominant sugars being arabinose with about 42 of galactose. The hydroxyproline content in this protein was 62. A recent observation showed that sandal protein agglutinated the trypsinised human red blood cells (20). Hyp-Arabs have‘also been found in the sandal lectin. Plants contain yet another type of macromolecule containing hydroxyproline, namely arabinogalactan proteins (AGPs) or so called "all fl-lectins". The characteristics of AGPs are l) the polymer is a glycoprotein with very high carbohydrate content (typically 80-95%), 2) the carbohydrate is an arabinogalactan (that's why they are called AGPs), 3) the protein portion is rich in hydroxyproline, serine, and alanine, and 4) the polymer precipitates with fl-glucosyl Yariv antigen. AGPs have been found in various plants from medium of suspension-cultured cells (3, 21-23), cytoplasm of suspension-cultured cells (23-25), xylem sap (4), wheat endosperm (26), seeds (27), stigma exudates (28-30), and at the surface of plant plotoplasts (31). They are ineresting macro- molecules because they are "glycoproteins in search of a function" and contenders for a role in the cell-cell recognition processes of plants. They are virtually ubiquitous in higher plants (27, 32). Jermyn and Yeow (27) isolated "all B—lectins" (same as AGPs) from various seeds using Yariv antigen and they showed the presence of this macromolecule in a variety of plants. Anderson g£_§l, (23) have isolated AGPs from suspension cultures of endosperm of Lolium multiflorum (rye 11 grass) by Yariv antigen precipitation. It has molecular weight 2.8 x 105 consisting of 842 carbohydrate with a small amount of protein (7%), rich in hydroxyproline, alanine and serine. The predominant sugars are galactose (64%) and arabinose (36%). The methylation analysis of this glycoprotein showed that branched 3-6 galactan is substituted by arabino- furanosyl residues and they also found homology in various arabinogalactan proteins from a variety of plants by methylation analysis. McNamara and Stone (33) examined the carbohydrate-peptide linkage in arabinogalactan peptide isolated from the wheat endosperm. The composition of this glycopeptide was 82 peptide which is rich in hydroxyproline and 922 carbohydrate. By successive treatment of arabinogalactan peptide with alkali, oxalic acid and enzymes, they obtained a small molecular weight compound which on further analysis was found to be 4-0-fi-E-galacto- pyranosyl-oxy-L-proline. AGPs were also shown to be present in the medium of suspension- cultured sycamore cells (3). Alkaline hydrolysis of crude AGPs from medium gave an interesting hydroxyproline-glycoside profile on Sephadex G-25. There was a hydroxyproline peak in the void volume as well as Hyp-Arabs in the inner volume. Further analysis of void volume fraction by gel-filtration and isoelectric focussing, followed by chemical analysis showed a singly residue of hydroxyproline with an arabinogalactan attached via the hydroxyl group. Further partial acid hydrolysis of this fraction with 0.1 N trifluoroacetic acid for 1 hr at 100°C yielded Hyp-Gal. Approxymately 502 of the hydroxyproline was found attached to arabino- galactan and 30% of hydroxyproline with Hyp-Arabs. The amino acid analysis of AGPs showed that hydroxyproline, alanine and serine are major amino acids. On this basis Lamport (4) suggested that the 12 polypeptide backbone of AGPs has a hydroxyproline-rich hydrophilic region and alanine-rich hydrophobic tain region. Jermyn and Yeow (27) observed high levels of hydroxyproline in all fi-lectins after proteolytic digestion and they suggested that a fraction of the hydroxyproline exists in a "core" after removal of a hydroxyproline-poor "tail". Even though both AGPs and extensin have a high hydroxyproline content, they differ in at least three ways; 1) hydroxyproline-glycoside profiles after alkaline hydrolysis; AGPs have hydroxyproline-arabino- galactan and Hyp-Arabs whereas extensin has only Hyp-Arabs, 2) alanine content; AGPs are alanine rich, however, extensin has little or no alanine, and 3) solubility; AGPs are extracted by aqueous solution, but it is hard to extract extensin from the cell wall. Despite these obvious differences there is the intriguing possibility of sequence homology between extensin and AGPs, for example, in the hypothetical hydroxyproline-rich region of AGPs. Because of its very high carbohydrate content, it is difficult to sequence AGPs directly on the sequencer. Therefore it is necessary to deglycosylate AGPs before attempts at sequencing. Mort and Lamport (34) used anhydrous HF for deglycosylation of glycoproteins without breaking the polypeptide chain. Anhydrous HF used routinely as a deprotecting reagent in protein chemistry, is an excellent protein solvent, and rapidly cleaves glycosidic linkages. For example, complete HF solvolysis of cellulose to glucose occurs within a minute at 0°C, under which conditions peptide bonds are quite stable. Anhydrous HF cleaved all the linkages of neutral and acidic sugars within 1 hr at 0°C, but the O-glycoside linkage of (N-acetylated) 0 amino sugars require somewhat more severe conditions, 3 hrs at 23 C (34). 13 However, HF did not cleave the N-glycosidic linkage between aspargine and N-acetyl glucosamine. Because HF is relatively dangerous, it is difficult to handle and it requires special apparatus. Therefore HP in pyridine (HF/pyr), which is much easier to handle than liquid BF, has been used. The purpose of this work was to test the efficiency of HF/pyr for deglycosy- lation of AGPs and to obtain the peptides for further sequencing analysis. First the appropriate conditions of HF/pyr deglycosylation was determined. Second the efficiency of HF/pyr deglycosylation was compared to liquid HF deglycosylation. And finally the deglycosylated AGPs (from medium.and cytoplasm) were partially purified. MATERIALS AND METHODS Materials. All chemicals used were of analytical reagent or the best commer- cially available grade. HF/pyridine was bought from Pierce Chemical Company, 11. The BF apparatus was bought from Peninsula Lab., Ca. (34). Sephadex G-25 (fine), G-100, blue dextran 2000, SP-Sephadex C-50, ribonuclease, and apomyoglobin were purchased from Pharmacia, Sweden. Biogel P2 was purchased from Bio-Rad Lab., Ca. Radioactive 14C-proline was bought from New England Nuclear Corp., England. Crude AGPs Preparation. From Culture Medium. Sycamore-maple suspension-cultures (Acer pseudoplatanus L.) were grown in M6E medium. After 12 days of growth, cells were harvested by filtration on a coarse sintered funnel. The filtrate was centrifuged to remove any broken cells and debri at 8000 rpm for 20 min in Sorvall RC-2. Ethanol was added to make a 702 ethanolic solution which was then allowed to settle in the coldroom overnight. The pellet obtained after centrifugation at 10000 rpm for 30 min was freeze dried. This material was called AGP(M) and contained 3-4 pg of hydroxyproline per 1 mg of sample. From Cytoplasm. Cells were washed three times with growth medium salts and then resuspended in the medium (ca. 1:1 v/v) with 0.004 M N28205, and then 14 15 sonicated at 5°C for 4 min in a Bronwill Biosonik III sonicator. The homogenate was filtered through a 20 u nylon cloth in the coldroom. The filtrate was centrifuged at 9000 rpm.for 10 min. To the super- natant solid TCA.was added to 12.52 saturation and the solution was stirred in the cbldroom till dissolved and allowed to stand overnight. The precipitate was removed by centrifugation (8000 rpm, 15 min) and the supernatant was dialysed against water for at least 2 days with 6 changes of distilled water. Dialysate was concentrated to small volume by evaporation, then dialysed again overnight with 2 changes of distilled water to remove any traces of TCA.and then freeze dried. This freeze dried material was called AGP(C), and it contained about 9 pg of hydroxyproline per 1 mg of sample. The yields of AGP(M) and AGP(C) were 7 gm and 300 mg respectively from.1000 gm of cells (wet weight). Twenty pCi of radioactive 14C-proline (specific activity 250 mCi/ nmole) was used in 600 m1 culture medium for the preparation of radio- active AGPs. The samples (1 ml) were made up to 10 ml with Aquasol (New England Nuclear) and the radioactivity was measured in a Packard Tri-carb liquid scintillation counter. Deglycosylation with HF/pyridine. Completely dried sample (500 mg) was placed in a Rel-F vessel with a stirring bar and 2 ml of anhydrous methanol was added. Then 18 ml of HF/pyr was added and reaction vesell was capped and stirred for 1 hr at room.temperature unless otherwise stated. After 1 hr 80 m1 of cold water was added to quench the reaction and the solution was dialysed against water for 2 days with 6 changes of distilled water in the coldroom. After dialysis the solution was centrifuged (12000 rpm, 30 min) and the pellet and supernatant were freeze dried. l6 Deglycosylation with HF. HF deglycosylation was performed as described by Mort and Lamport (34) using anhydrous methanol instead of anisole. After complete evaporation of HF, the deglycosylated sample was dissolved in 0.1 N NH4OH and centrifuged to remove any insoluble material. The super- natant was loaded on a Sephadex G-100 column as described next. Column Chromatoggaphy. The gel-filtration experiment was performed in 0.1 N NH4OH using a column of 500 m1 volume (2.8 x 80 cm) with void volume of 154 ml (via blue dextran) unless otherwise stated. The sample volume used in gel-filtration was 5 ml. An aliquot of the fraction was used for hydroxyproline estimation by an automated hydroxyproline analyser (10) after prior hydrolysis of the sample in 5 N NaOH. SP-Sephadex C-50 ion-exchange chromatography was performed as follows. The material to be fractionated was dissolved in 0.01 N HCl (pH 2) and centrifuged to remove any insoluble material. Then the sample was applied to a column (1.2 x 26 cm, bed volume 30 m1) and washed with 1.5 bed volume of 0.01 N HCl after which the column was eluted with a gradient between 0.01 N HCl and l M.NaCl in 0.01 N HCl (5 bed volumes). An aliquot of the fraction was used for hydroxyproline determination. Biogel P2 column (-400 mesh, 2 columns of 0.13 x 90 cm) was used for the separation of hydroxyproline-glycosides. The column was equilibrated and eluted with water containing 0.012 sodium azide. Column eluate was monitored continuously with the hydroxyproline analyser. Disc-gel Electrophoresis of the Deglycosylated AGPs Fractions. l7 Deglycosylated AGPs fractions were electrophoresed in 122 polyacryl- amide gels (35). Prior to electrophoresis the sample was prelabelled by the procedure of Muramoto g£_§l, (36). About 500 ug of deglycosylated AGPs were dissolved in 30 ul of carbonate buffer (pH 9.5) and labelled with 5 ul of fluorescein isothiocyanate (FITC, 10 mg/ml solution in acetone) for 10 min at 50°C followed by gel-filtration on a Sephadex G-25 (fine) to remove excess FITC. The standard proteins used were ribonuclease, apomyoglobin, and a soluble hydroxyproline-containing lectin from leaves of Santalum ElEEE.L° (19). After electrophoresis the sample was eluted from gel slices by a procedure of Drescher and Lee (37) modified by omitting fixation and staining. Three to 6 mm fluorescent bands were cut and placed in a 3 m1 microflex tube. To this 0.2 ml of 12 SDS was added and homogenised gently for 5 min using a Kontes pestle. The vial was sealed and incubated at 40°C for overnight. Then 0.2 ml of 0.12 SDS was added to the mixture which was resuspended on a vortex mixer, then centrifuged for 5 min at 1000 x g. The supernatant was collected. The 0.12 SDS extraction of the pellet was repeated three more times. The volume of the pooled extracts was reduced to 100 ul by blowing down with nitrogen, and the extract applied to a Sephadex G-25 (fine) column (0.9 x 14 cm) equilibrated with water. The column was eluted with water. Using a long wave U. V. lamp, the separation of labelled protein from unreacted dye was observed. The fluorescent eluate which voided the column was collected. The material was freeze dried and analysed for amino acids either by conventional amino acid analyser or gas chromatography. After electrophoresis of 14C-labelled material, 2 mm gel slices starting from the origin were taken and transfered to a counting vial, 18 and then the Aquasol added, homogenised and counted in a scintillation counter. Hydrolysis of Sample. For amino acid or quantitative hydroxyproline analysis, sample (1-5 mg) was hydrolysed with 200 ul of 6 N HCl in 1 m1 microflex tube for 18 hrs at 110°C. After evaporation of HCl in a stream of nitrogen, 500 ul of 0.001 N HCl was added and an aliquot was used for analysis. For qualitative hydroxyproline analysis of the column eluate, an aliquot (max. 200 pl, if the sample was more than 200 pl, first freeze dry then add NaOH) was hydrolysed with 800 p1 of 5 N NaOH for 1 hr at 121°C in a polyethylene tube, then neutralised with 850 pl of 5 N HCl and subjected to automated hydroxyproline analyser. For hydroxyproline-glycoside profile determination, the sample was hydrolysed with saturated (0.22 M) Ba(OH)2 (10 mg/ml) in a 5 m1 micro- flex vial for 18 hrs at 105°C, neutralised with concentrated H2804 (8 pl/ml Ba(0H)2), and then centrifuged for 15 min at 10000 rpm. The supernatant was freeze dried and redissolved in 200 ul of water, then loaded on a Biogel P2 column as described earlier. §ggar Analysis. Sugar analysis was performed on the trimethyl silylated methyl glycosides as described by Bhatti g£_gl, (38) using a Perkin-Elmer 900 gas chromatograph fitted with dual columns, with the output connected to a Spectra Physics System IV Autolab integrator (Figure 4). The support Gas-Chrom Q and stationary phases SE-30 and SP-2100 were bought from Supelco Co., Bellefonte, Pa. Amino Acid Analysis. Most amino acid analyses were performed by liquid chromatography l9 .HoueeeeH Amfl .Houeeeez Asa .eHu Aea .eHo Ana .Heo AeH .aaeo and .Heo AuH .eez HHH .Hee Hoe .eez as .efleo Am .Hax Ha .Hax Ae .eee Am .oee He .eee an .ee< am .eee 1H “oxmoo mo aOHumonHucooH .cHE q no oHon HmHanH so nuHs .cfla\oo H um u ommH on u oomH aouw ooaamumoun A:w\H x .NHV casHoo ooHNImm Nm m ouao oouooncH ems coHumHom oSu mo H1 039 .ousuouonamu soon as coHuouHuo>Huoo can you oesoHHm mos poo: co MHmm .AH.H.mV oneNoHHuHlonuoamxon\oanHoouoHnoH%:uoaHuu\o:HoHuaa mo HaeomN wcHoom %o oouuHhHHlonuoEHuu mos uH coca .un H ammoH as now ounuuuooaou Boon um ON m uo>o uoumoonoo annom> :H mouon one unnumuoaaou Eoou um aOHuouonm>o cowouuHo uH> ooHuo um: oHanm one .:HE om How xooHo onumo: u omm :H voooHn mos HmH> one .anuaoo ocoomnu one moo nuHs ooHooo zHoumHooEEH some one one mH pom uopo c3oHn mos oomouuH: hum one women was Ho: oHHocmSuoE z m. H uuo «o H: ounces: one .eee on eeeee.ee eee oooeaom ee none Mono noumooHooo anooo> m :H menu onmouoHa o cH ooHuo one: venomous HocuouoH mm HouHcoma mo moHoaonma ooH man AHouHmosH wcHonHoch Human some «0 w: >uao39 .moonooaHm H>nuoa HhHHmthquHuu uHonu mm mumwno mo hnnmuwoumaouno oHowHH com .o musmfim 20 .V ”HSMfih 33:33 as; on On cc on cc om OH MH OH «H H A m h m N m 1 9H NH c w o H m NH wH m 21 (Figure 5), others were performed by gas liquid chromatography on the heptafluorobutyryl isobutyl ester derivatives as described by Mackenzie and Tenaschek (39, Figure 6). Hydroxyproline was determined on an automated hydroxyproline analyser (10) after hydrolysis of the sample. 22 e .924 8N .2: AS .eea Ann . ez Ann .oee Aen .eee Ann .eeneez ASH .een Ann .enn ANH .eez Ann .eno non .Hee Aw .ene Am .ene an .eee Ae .ene Am .eem As .een Am .ee< AN .eee AH “oxoon mo noHuooHMHunooH .nn Cum um ooHoo onHao nosuo onu .an own um oonmonomom uHonu an oouoouoo onus onHHonn ono onHHounhxonozm .uoumuwounH ooHounn >H aouozm oonhnm onuoonm a he oouoHnoHoo one: woman Home onu one noozHono anomnan ooumnouno no ounH oom was nEnHoo moo aonm unonHo one .Amv unonEoH ho oooHuoooo on unoHooum sumnouum oHnoH .mn noHuonHoEoo m :uHs oounHo one nEnHoo u oomonoEouno mo oou no ooooHn onus oaoonmuo HonuounH mo onHonoHnon mo moHoEonmn OOH one onHHounzxonozn «.0 ml om .oHoo onHam come no ooHononon .3me .mnnmuwoumsouno oHnoHH ho onhHmno oHoo onHa< .m ouanm 23 m ouanm Amuse cane m J J NH ON q OH NH HH «H g fl « ‘ H H . 0‘) asuodssu zonoanaq 24 .eeo Ame .me< Ann .enn Aen .ene Ann .eno Asa .eee Ann .ee< Ame .uaz Ann .e»: Aon .enee uaHoomenm Am .oee Am .eHH an .eme fie .uem an .uen He .He> Am .nno Au .eH< AH unseen mo nOHueonHuneoH .nHE e «o oHon HeHanH ne nuHs .nHa\ooc ue ooomu ou comm noun erEeuwoun .eHexHene nemne you owe: me nEnHoo oaem ouno oeuoeHnH ones noHunHoe eHnu «0 H1 039 .A>\> .Hqu ooHuohnne oHueoe\eueueoe Hanue mo H1 cw nH oe>HoeeHo one eunuenenaeu noon ue Eeeuue nemouan nH eeenzno uneHnHonH ou oeueuone>o eea uonooun onu .onnueuenaeu Boon cu ooHooo .nHa OH now u ooHH ou oeuee: mes HeH> ecu one ooooe ones eoHnomnne 0HumunoouonHmeunen mo H1 om one oueueue Hanna «0 H1_om .eunueuenneu aoou ue aeenue nowouan nH oeHno nouw< .eunueuenaeu noon ou ooHooo one nHE ow now ooowH ue xooHo wnHuee: nH oeoeHn men HeH> any .nueo oHnoe nH one on now oeueoHnoe one oeooe ewes HoneunooeH nH Ho: 2 m mo H1 oeuonnn ono .eoHsu ooHuoou mes eHnnee ecu one oeooe one: eoHnoHno oneHznuon mo H1 oenonnn one .eunueuenaou Boon ue aeonue nemonan neon: HeH> ermouoHe Ha H e nH oeHno ones oueoneue HenuounH ee oHoe oHHooean mo eeHoaonen on ean oHoe onHae noee mo eeHoaonen oononn: eno .euoueo qunooeH HmumunoouonHmeuno: uHenu ee eoHoe onHae mo hnneuwoueaouno oHDUHH new .0 enanm a \3 3t 1") 3%' 3 J _____..—J J 25 .1:— N H "H" H O H Chang—— 4) ”‘1: H "-——-—=: :3... so" e,_} -H... ‘--< asuodseu 10309390 20 30 40 Time (minutes) 10 Figure 6 RESULTS Chemical Composition of Crude AGP(M) and AGP(C). The sugar composition of crude AGP(M) and AGP(C) is in Table 3 and the amino acid analysis is in Table 4. The major sugars are Ara, Xyl, Gal, and Glc and the predominant amino acids are Hyp, Ser, Ala, Glu, Val, and Lys. From the amino acid and sugar analysis, AGP(M) consists of 902 carbohydrate and 102 protein and AGP(C) consists of 802 carbo- hydrate and 202 protein. Deglycosylation Time Course of AGP(M) with HFprridine. In order to find the optimum time required to remove most of the sugars from AGPs, 100 mg of AGP(M) was treated at different time intervals (0 to 240 min) with anhydrous HF in pyridine (9 ml) in the presence of anhydrous methanol (1 ml) at room temperature and the reaction was quenched by the addition of 40 m1 of water. Then the solu- tion was dialysed against water for 2 days in the coldroom and the dialysate was freeze dried. Figure 7 shows the kinetics of removal of sugars at different time intervals. It is clear from the Table 5 that there was an appreciable decrease in the total weight after deglycosylation with a complete recovery of hydroxyproline at different time intervals. Table 6 shows mole 2 of sugar and moles of sugar per mole of hydroxyproline remaining after the treatment of AGP(M) with HF/PYr at different time intervals. More than 902 of sugars were removed after 45 min treatment with HF/pyr (140 moles to 14 moles sugar 26 27 Table 3. Sugar Analysis of Crude AGP (M) and AGP(C)a 'AGP(M) AGP(C) Ara 23.4 (32.8) 41.5 (13.2) Rha 2.7 ( 3.9) 5.2 ( 1.7) Fuc 5.4 ( 7.6) 2.3 ( 0.7) Xyl 19.2 (26.8) 5.2 ( 1.6) GalU 8.2 (11.5) 2.2 ( 0.7) Man 1.1 ( 1.5) 5.1 ( 1.6) Gal 25.4 (35.6) 35.4 (11.2) Glc 14.6 (20.4) 3.0 ( 1.0) aData expressed as mole 2 of total sugars. Numbers in parenthesis indicates the moles of sugar per 1 mole of Hyp. 28 Table 4. Amino Acid Analysis of Crude AGP(M) and AGP(C)a AGP (M) AGP (c) Hyp 6.1 9.1 Asp 10.6 13.2 Thr 5.8 7.3 Ser 7.8 ' 10.7 Glu 9.2 14.3 Pro n.d. n.d. Gly 8.3 8.3 Ala 7.1 11.3 Val 7.4 6.3 Cys 0.2 0.6 Met 1.2 0.6 Ile 3.0 2.6 Leu 5.3 4.3 Tyr 2.2 0.8 Phe 4.1 1.5 Lys 6.6 6.3 His 4.1 1.2 Arg 2.3 1.4 aData expressed as mole 2 of total amino acids. n.d. not determined. Figure 7. 29 The kinetics of removal of sugars from AGP(M) at different time intervals. One hundred mg of AGP(M) were treated with 9 ml HF/PYr and 1 ml MeOH at different time intervals at room temperature. After adding water to quench the reaction, the solution was dialysed and freeze dried. One to 2 mg of each material were analysed for sugar. 15 H 0 Sugar, moles per 1 mole Hyp A 3O r—o Ara #-—-;a Fuc o—---o GalU o---—o Man A—---A Gal Ar---A Glc -— ----A.‘- '-a------ -~- Time (hrs) Figure 7 31 «a mOH mm mm ooH mHH mm OOH oone>ooem N can men emm NeN . nmm «an new omm An: .Hnuoao enHHounhxouomm uneaueous «.me m.me a.me m.am m.em ~.om m.me o uan\em an eu>eEue N m.em a.em e.em m.oe e.me m.ae ~.em one Anne mane euneaanee unamue use no oceans can own on oe .me on an o “ease uane eunueuenaea Boom ue Honenue: NOH ean uhn\m= enouohnne wnHon eHe>ueunH oaHB uneneMMHn ue szmu< mo noHueHmeoohHmen neuwe >ne>ooem onHHounhxonozm .m eHoeH Table 6. 32 Sugar Analysis of AGP(M) after Treatment with HF/pyr at Different Time Intervals Time Course of HF/pyr Treatment mm). .. . .9. s .90 .45... , 60 90 180 240 "Mole 2'of Total Sugg Ara 23.4 2.5 3.1 2.2 3.0 2.4 2.5 1.9 Rha 2.7 1.9 1.9 2.0 2.4 1.9 1.3 1.0 Fuc 5.4 0 0.6 0 0 0 0 0 Xyl 19.2 22.0 13.0 8.4 8.6 4.0 2.9 2.2 GalU 8.2 25.0 43.4 55.0 50.5 62.2 60.0 67.5 Man 1.1 2.3 2.1 3.4 5.0 5.8 6.3 6.3 Gal 25.4 31.9 16.4 11.7 ‘10.7 4.4 4.2 3.2 Glc 14.6 14.3 19.4 17.3 19.8 19.3 22.7 17.8 Molesgper l Mole of Hyp Ara 32.8 1.3 0.5 0.3 0.4 0.3 0.3 0.2 Rha 3.9 1.0 0.3 0.3 0.3 0.2 0.1 0.1 Fuc 7.6 0 0.1 0 0 0 0 0 Xyl 26.8 11.7 1.9 1.2 1.2 0.4 0.3 0.2 GalU 11.5 13.3 6.2 7.8 7.1 6.5 6.0 6.2 Man 1.5 1.2 0.3 0.5 0.7 0.6 0.6 0.6 Gal 35.6 17.0 2.4 1.7 1.5 0.5 0.4 0.3 Glc 20.4 7.6 2.8 2.4 2.8 2.0 2.3 1.6 140 ’53 V15 14 14 ll 10 9 33 per mole of hydroxyproline). Galacturonic acid was the major sugar remaining after HF/pyr treatment (6-7 moles per mole of hydroxyproline). Figure 8 shows the hydroxyproline-glycoside profile of deglycosy- lated AGP(M) at different time intervals on Biogel P2 column. Six peaks corresponding to Hyp-arabinogalactan (void), Hyp-Ara4, Hyp-Ara3, Hyp-Araz, Hyp-Ara and free Hyp were observed with AGP(M) after alkaline hydrolysis 1 with Ba(0H)2. There was a decrease in the amount of Hyp-arabinogalactan and Hyp-Arabs whereas free Hyp increased as the time of HF/pyr treatment increases. However, there was a peak at the Hyp-Ara region (before 1 free Hyp) appearing after 15 min of treatment of AGP(M) with HF/pyr and was at a maximum.at 45 min after treatment. In 90 min deglycosy- lated AGP(M), there was 162 of Hyp in the Hyp-Ara region and 842 was 1 in the free Hyp region. The peak in the Hyp-Ara position did not 1 contain any arabinose and therefore it is not Hyp-Aral. But, galactose and glucose were found in HF/pyr "resistant" fraction. The molar ratio of total sugar to Hyp was about 1. These results suggest the possibility of Hyp-Gal and Hyp-Glc in the HF/pyr "resistant" fraction. This corroborated well with earlier results from this laboratory (3) which raised the possibility of glucosyl hydroxyproline. Effect of Temperature on HF/pyr Treatment. Mort and Lamport (34) observed that the efficiency of HF deglycosy- lation is different at 0°C and at room.temperature. Therefore degly- cosylation of AGP(M) with HF/PYr was performed at 0°C and at room temperature for 1 hr. After treatment and dialysis, the dialysate was centrifuged at 12000 rpm for 20 min to separate the soluble and insoluble material. Amino acid and sugar analyses of these fractions are given in Table 7 and 8. The efficiency of deglycosylation was 1.4 34 Figure 8. The hydroxyproline-glycoside profile of deglycosylated AGP(M) at different time intervals on Biogel P2 column. About 20'mg of deglycosylated AGP(M) at different time intervals were treated with 2 m1 of 0.22 M Ba(0H) in a 5 ml microflex vial for 18 hrs at 105°C. Then it was neutralised with concentrated H 804, and then centrifuged. The supernatant was freeze dried and redissolved in 200 p1 of water, then loaded on a Biogel P2 column. - Void Reaction Hyp-arabi o 3 Time (min) A galactan HA 2 Free Hy 0 J p l l ,l flél l 15 M L. !_ I 1 J , g l 2 l J l 1 4 .0 m l L 1 L I 180 ' Al L, __4_ l L 1. , l 240 ) l a. _._ I _1 _1 1— I 0 1 2 3 4 5 Time (hrs) Figure 8 36 Table 7. Sugar Analysis of the HF/pyr Treated AGP(M) at 0°C and at Room Temperature (RT) for 1 hr Fraction Soluble after Insoluble after HF/pyr HF/pyr Temperature RT 0°C RT 0°C Hyp (mole) 1 1 1 1 Sugar (moles per 1 mole of Hyp) Ara 0.4 1.9 0.1 0 Rha 0.2 0.7 0 0 Fuc 0 0.1 0 0 Xyl 0.4 10.3 0.2 1.0 GalU 22.6 19.5 5.8 8.7 Man 0.8 0.7 0.2 0.6 Gal 1.2 13.1 0.3 1.8 Glc 17.0 5.5 0.8 0.8 Total 43 52 7 13 37 Table 8. Amino Acid Analysis of the HF/PYr Treated AGP(M) at 0°C and at Room Temperature (RT) Fraction Soluble after Insoluble after HF/pyr HF/pyr Temperature RT 0°C RT 0°C Amino Acids (mole 2) Hyp 11.5 7.1 6.1 8.6 Asp 12.0 12.8 11.0 10.9 Thr 7.3 8.3 6.2 5.8 Ser 11.7 12.7 7.8 7.9 Glu 9.9 10.0 8.8 9.6 Pro n.d.a n.d. 8.3 9.3 Gly 8.0 8.7 7.4 7.3 Ala 11.4 15.2 6.4 6.9 Val 5.0 7.4 6.3 5.9 Cys 0 0 1.3 0.7 Met 1.7 0 1.0 1.1 Ile 3.0 3.1 2.9 3.0 Leu 5.0 4.2 5.2 4.9 Tyr 1.4 0.7 1.9 2.0 Phe 2.9 0.9 4.3 3.4 Lys 4.8 5.3 7.5 6.4 His 3.5 1.7 4.9 4.0 Arg 1.0 1.9 2.5 2.3 n.d. not determined. 38 times greater at room temperature than at 0°C. Therefore the HF/pyr deglycosylation has been performed routinely at room temperature for 1 hr. Deglycosylation of AGP(M) via HF/pyridine. Partial Purification of Deglycosylated Hydroxyprolin-containing Material. Figure 9 shows the flow sheet for the partial purification of deglycosylated hydroxyproline-containing material. Most of the degly- cosylated material (ca. 752) remained insoluble after dialysis against water. Therefore the solution was centrifuged and the further fractio- nation was done with the supernatant fraction. This material was freeze dried and loaded on a Sephadex G-100 column (500 m1 bed volume) with 0.1 N NH OH (642 of the hydroxyproline dessolved). There were four 4 peaks of hydroxyprolinc-containing material (Figure 10). Peak I (void) contained 102 of the total hydroxyproline, while the percentage of hydroxyproline in peaks II, III and IV is 25, 40 and 25, reapectively. Sugar analyses of these four fractions are in Table 9. Peak III was chosen for further purification. This fraction was freeze dried and dissolved in 0.01 N HCl and loaded on an SP-Sephadex C-50 column. The hydroxyproline-containing material was retarded and appeared in the gradient elution (Figure 11). The amino acid analysis and sugar analysis of this fraction are given in Table 10. The hydroxyproline content of this fraction was about 17 mole 2. In this fraction the ratio of sugars to protein (w/w) was 15:85. Peak II was also purified in the same way. The final fraction was obtained from SP-Sephadex C-50 at the 0.35 M NaCl region. The amino acid and sugar composition of this fraction is given in Table 10. The ratio of sugars to protein in this fraction (wfiw) 39 .szmu< noum HeHueueE :oHulenHHonnmxouohn mo noHueonHunn HeHunen one now ueese 3OHm may .m euanm an»: we ammo Ann: me name Honz : ma.o Honz 2 nm.o ue oeunHov nOHuoeum ue oounHev noHuoenm noHuIenHHounmxouoh: soHnIonHHonnmxouohm onuonem omuouem Anne ea came an»: m: omev Anne a: cane Hen: we oenv >H gene HHH anon HH anon Heao>o H gene a a a e canto xueeanum no eeaueuunae Hue Anne we ooeno moemz z H.o :uHs oouoenuxm an»: ma oommv an»: ml ooosv oeHuo museum .uneuenuennm ooHuo museum .ueHHom nae on see ooo~a oewDMHnuneo .eHehHeHo .neues oo< moo: enouohnne NOH :qu .naeu Soon as H uzn\m= Anne n1 oceans szmu< eonnu 4O .enHHounhxouomn you oeneeeeeues.enoHuoeuu euenneuHe mo euonUHHe H1 com one oeuoeHHooeuosienOHuoeum HS :09 .moemz z H.o :uHa oeunHe one oeueuoHHane mes nanHoo may .Aao m.~ x owv nanHoo OCHIw xeoennem e no oeonH mes uneuennenne on» one oewnMHnuneo .moemz z H.o Ha m :qu oeuoenuxe eea HeHneuea eHne .Awa omHv oeHno museum mes uneuennenne one .oem:MHuuneo one oeeaHeHo eez noHunHoe on» .noHuoeen one nonenv ou neuez Ha om mnHooe Heum< .ennuenenaeu Boon ue a: H now woe: Ha N ean u>n\m= Ha wH :uHB oeueeuu ones szmw< mo we oeuonn: e>Hm .ooHlu neoesnem e no u%n\mm nuHa szmu< oeueHmeoomHmeo mo noHueuuHHw How .oH unawae OH ouanm :5 O5 Hooaoz nOHuoeum Ow 41 H .u m0.0 lOH.O lmHé lON.O HHH an coma. En: 42 Table 9. Sugar Composition of Sephadex G-100 Fractions of AGP(M)/HF/pyr/Super F...t1.n._ M. , _ A I H,“ I .II n , III . Iv Mole 2 of'Total'Sugar Ara 1.2 0 0 4.5 Rha 0 0 0 2.8 Fuc 0 0 0 0 Xyl 0 0 0 6.4 GalU 87.0 89.4 82.9 36.6 Man 4.9 3.9 3.9 2.0 Gal 0.9 1.4 1.8 11.7 Glc 6.0 5.4 11.4 36.0 Molesgper 1 Mole of Hyp Ara 1.96 0 0 1.79 Rha 0 0 0 1.11 Fuc 0 0 0 0 Xyl 0 0 0 2.54 GalU 142 33.1 20.5 14.6 Man 7.95 1.43 0.95 0.82 Gal 1.50 0.50 0.45 4.66 Glc 9.78 2.01 2.80 14.3 43 .Hoez 2 H wanHeunoo Ho: 2 H0.0 one.40= z H0.0 neeauen uneHoenw :uHa .H: on HH enoHuoenm "Hum z HO.O :uHB .OH ou H enoHuoeum uenoHHom me mes eHnoesoe nOHunHo any .enHHounbnonoxn now oemeeee eyes enOHuoenw .euenueuHe mo euoncHHe H1 OON one oeuoeHHoo mes noHuoeuu Ha e>Hm .eBnHoo Omlo xeoennemlmm ne ou.oeHHone mes uneuenuenne esp .oewnMHuuneo one Ho: 2 HO.O nH oe>HoeeHo mes HeHneuea eH£H .oeHno eueenm one oeHoon eeB AOH .mHm eeev HHH xeen OOHIO Neoennem .eoauunue HHH xeue conuo .une\ee :33 96 m3. eHnealeuoaeome oeuthfleoomeeo mo Anneumoueaouno EnnHoo Onto neoennemlmm .HH unseen 44 no Honz HH euanm HHS my noonnz nOHuoeum He mm mm NH O H - 1 a a o l no.0 an Oom< a»: l OH.O L mH.O 45 Table 10. A Comparison of the.0hemica1 Compositions of Partially Purified Hydroxyproline-containing Fractions from Deglycosylated AGP(M) Arg Fraction HF/pyr/G-lOO II/ HF/PYt/G-lOO III/ HF/G-100 II/ SP-C-50 SP-C-SO SP-C-50 Concentration of - NaCl 0!) 0.35 0.15 0.28 Final Yield (2, Hyp basis) 2 5 6 Sugar, mole Z'(mo1es 'per 1 mole Hyp) Ara 4.6(0.08) 0 (0) 14.9(0.05) Rha 0 (0) 0 (0) 0 (0) Fuc 0 (0) 0 (0) 0 (0) Xyl 7.0(0.05) 6.4(0.03) 19.1(0.07) GalU 0 (0) 0 (0) 0 (0) Man 10.7(0.15) l8.7(0.11) 0 (0) Gal 50.0(0.24) 47.1(0.26) 8.7(0.04) Glc 27.6(0.3l) 27.9(0.15) 46.7(0.20) Amino Acids, mole 2 Hyp 22.8 17.4 17.7 Asp 6.9 11.5' 11.1 Thr 9.6 8.3 9.5 Ser 15.0 16.3 14.1 Glu 4.4 6.9 6.8 Pro 2.1 n.d. n.d. Gly 5.6 5.0 5.8 Ala 15.3 11.9 14.8 Val 6.2 5.9 6.7 Cys 0 0.6‘ 0 Met 0 1.4 0 Ile 2.0 5.0 2.8 Leu 3.4 2.9 4.2 Tyr 0 0.4 0 Phe 1.5 1.9 1.4 Lys 3.3 3.1 3.3 His 0.7 0.6 0.6 1.2 1.0 1.1 n.d. not determined. 46 was 13:87. For radioactive AGP(M), the purification procedure described in the flow sheet (Figure 9) was used for the fractionation. The radio- activity present in the SP-Sephadex C-50 fraction (peak III) was 45000 cpm. SDS Disc-gel Electrophoresis of Hydroxyproline-rich Fractions. So far there are no reports of the gel-electrophoresis of AGPs either intact or deglycosylated. This may be due to their poor reacti- vity with dyes such as coomassie blue. For example, the major fraction of the deglycosylated sycamore-maple AGPs stained very poorly, if at all, with coomassie blue R250 (40). Therefore the combined coomassie blue 6250 staining/fixation method of Blakesly and Boezi (41) was tried to avoid the possibility that a hydroxyproline-rich protein might be eluted from the gel during the lengthy fixation/destaining procedures commonly in use. However, the deglycosylated AGPs fractions again stained poorly. Therefore AGPs were prelabelled by two methods, first by growth in 14C-proline and second by reaction with FITC. The prelabelled fractions were subjected to polyacrylamide gel-electro- phoresis and appropriate gel slices were eluted, desalted, and analysed for amino acids via gas chromatography and conventional amino acid analyser, all as described in Materials and Methods. Electrophoresis of the (FITC-labelled) major fraction of deglycosylated AGPs showed one minor and two major but somewhat diffuse fluorescent bands which were eluted and analysed (Table 11). After electrophoresis of the (14C- 1abelled) major fraction of deglycosylated AGPs, 2 mm gel slices were prepared for 14C assay in a liquid scintillation spectrometer. Most of the 14C-labelled material was present in slice 9, 16-18 mm from the 47 Table 11. Amino Acid Composition of Sycamore AGP(M) Peak III/SP-C-SO before .ané .aft¢I._EIs¢II°IIh9I¢eIS (mole Z). - v Slice 1 Slice 2 Slice 3 AAl GC AA GC AA GC AA Hyp 17.4 19.9 26.5 19.0 15.7 5.0 n.d. Asp 11.5 4.5 5.8 9.9 10.8 16.3 23.5 Thr 8.3 6.8 8.2 9.5 8.0 10.7 13.7 Ser 16.3 19.0 21.7 20.5 17.9 14.7 17.1 Glu 6.9 6.7 7.5 3.0 7.4 9.1 9.2 Pro n.da n.d. n.d. n.d. n.d. n.d. n.d. Gly 5.0 17.0 7.8 n.d. 7.0 n.d. 3.5 Ala 11.9 11.7 15.4 12.2 14-7 12.0 n.d. Val 5.9 2.4 n.d. 5.7 4.7 7.4 10.9 Cys 0.6 n.d. n.d. n.d. n.d. n.d. n.d. Met 1.4 n.d. n.d. n.d. n.d. n.d. n.d. Ile 2.9 3.0 n.d. 5.3 3.4 5.8 7.3 Leu 5.0 1.8 n.d. 6.4 5.8 8.6 14.2 Tyr 0.4 0.8 2.9 n.d. n.d. 1.8 n.d. Phe 1.9 2.4 n.d. 4.1 2.2 5.4 n.d. Lys 2.1 1.5 3.8 n.d. 2.5 2.7 n.d. His 0.6‘ n.d. n.d. n.d. rid. n.d. n.d. Arg 1.0 1.8 n.d. n.d. n d. n.d. n.d. n.d. not determined. 121% [21 f1 1m AAl Data obtained from amino acid analyser before labelling with FITC (i.e. starting material before electrophoresis). 00 Data obtained from gas chromatography after labelling with FITC. AA Data obtained from amino acid analyser after labelling with FITC. Six hundred pg-of protein (100 pg hydroxyproline) was labelled with FITC amd applied to gel. The regions with intense fluorescent bands (slice 1 and 3), faint fluorescent band (slice 4), fluorescent smear (slice 2) and no fluore- scent region (slice 5, control) were cut from.the top of the gel and the material was eluted as described under methods. The amino acid analysis was performed with this material after hydrolysis. Slices 4 and 5 gave no amino acids. The hydroxyproline recovery from the gel was 602. 48 Figure 12. Amount of radioactivity in gel slices. Radioactive AGP(M) was deglycosylated with HF/pyr and the deglycosylated AGPs were purified according to Figure 9. The final fraction after AP-C-SO was gEl-electrophoresed and 2 mm gel slices were prepared for C assay in a liquid scintillation spectrometer. cpm top 800 600 400 200 49 122 acrylamide gel 2 mm slices 10 L I 20 30 40 Slice Number (2 mm slices) Figure 12 trackin . lbqrtom 50 50 origin (Figure 12). In a separate experiment, a portion of the major 14C-labelled fraction was hydrolysed with 6 N HCl and the hydrolysate was separated on an amino acid analyser, and successive 1 ml fractions were assayed for 14C-label in a liquid scintillation spectrometer. The results showed an exclusive distribution of label between hydroxyproline (802) and proline (202). As a check on the methodology used above, control proteins were also analysed (ribonuclease, apomyoglobin, and the soluble hydroxyproline-rich glycoprotein from.Santalum 31293). Table 12 shows the reasonably good agreement for the amino acid analysis (by gas chromatography and conventional amino acid analyser) of control proteins before and after SDS gel-electrophoresis. HF Control Experiment. To compare the efficiency of HF/pyr with anhydrous HF, the control experiment with HF was performed. Five hundred mg of AGP(M) was deglycosylated with 18 m1 of HF and 2 ml of anhydrous methanol for 1 hr at room temperature. Further fractionation was as for the HF/pyr experiment. After evaporation of HF, the deglycosylated material was extracted with 0.1 N NH 0H and the extract was loaded on a Sephadex 4 G-100 column (bed volumn 500 ml). Figure 13 shows the elution profile, which differs slightly from.the HF/pyr experiment (see Figure 10): there were only three peaks containing hydroxyproline. The major peak (II) was further purified with SP-Sephadex C-50 column chromato- graphy. A single hydroxyproline peak appeared in a gradient region (Figure 14). The chemical composition of this material and the final yield in terms of hydroxyproline were compared with the results obtained from HF/pyr experiment (Table 10). The amino acid compositions of the major peaks using HF/pyr (peak III/SP-C-SO) and HF (peak II/SP-C-SO) .eHeenonnonnooHe nonwe How man Bonn oennHe HeHneuea uneoeenonHm enn mo eHexHene oHsnenmoueaonno new 0 .eHeenocnonnoeHe enomeo annenwoneaonno new aonm oenHenoo nOHuHeonaoo oHoe onHE< a .neehHene oHoe onHae aonw oenHenoo euen .e .:>nnenaonoon mo xoooonem: xoon eon aonu nexen enen e .Hew eon nonm enHHonnnxonohn mo ane>ooen NON me: onenn .nHeuonn Heonee mo eeeo men nH .Hew emu no oeHHnne we: HeHneuea eHonn one OHHm :nHa oeHHeneH one? Ami ONHV nHeuonn Heonee one .Ama OONV nHAOHwozaone .Awa OONV eeeeHonnoon .oenHEneneo non .o.n 51 o.o o.m .o.n m.e o.N m.e m.H ~.m One .o.n H.H .o.n .o.n w.s m.e .o.n N.m eHm m.m m.H .o.n ¢.m w.HH e.HH m.o H.O can N.e m.< m.n e.n m.m N.m N.m e.~ one H.H ~.m o.OH o.< O.~ m.e 0.0 w.e nhs ~.n m.o m.mH N.MH «.mH O.H e.H o.H noH m.m m.m m.w n.o O.m e.H m.H ¢.~ eHH .o.n m.H .o.n 0.0 m.H m.H H.O ~.m no: out: ooN 0‘0: on.“ 0‘0: 0‘0: oven “0° 08 m.m O.e e.m N.m N.m m.m H.o m.m He> o.OH ~.a H.eH O.HH m.HH O.NH «.5 h.m eH< m.oH m.e .o.n m.o «.5 .o.n .o.n e.N hHO O.e m.o m.n m.m o.~ .o.n H.m ~.m onm O.n e.m n.oH o.eH m.~H n.m e.MH h.m nHO 0.x m.m H.e O.e a.m o.mH «.mH H.~H new O.m O.m m.q m.m m.m O.~H e.m H.O nee n.o O.m m.“ m.o N.m m.OH 0.0H H.NH nee O.m n.m O O O O O O on: o .e o o e o o e nHenonm Heonem nHAOHwomaon< eeeeHonnoon AN oHoav enHenonm oneoneum mo nOHuHeonaou once onHa< .NH eHneH 52 .enHHonnhxono%: now oezeeee en 3 enoHnoenm enennenHe mo euoanHe H1 OON one H8 OH me: eanHo> noHnoenm .mo :2 z H.O nan oenenoHHHnoe n.enHoo OOHIO xeoennem no O.~ x cm no no enonneoHHnne onenenee oBu nan oeuenoHuoenw mes HHS OH unooev unenennenne eon one oewnmnnuneo men noennxe ena .mOemz z H.O :nHa oenuennxe mes HeHnenea on» .mm wanenone>e nonme .onnnenonaen Soon en n: n new mouz.de N mane me He en runs euueuuu uuus Arcane eo we rupees; u>ne .oonno xueuenum en szeu< euuennnounnmueuee no eonunnunae nuo .mn unseen 53 OO On me anaeee neonnz noHuoenm On On ON OH q- 1 d ‘1 \ lb P) a ' HHH HH H oHo> .1. m0.0 OH.O an Oom< a»: mH.O ON.O 54 .mcaaoummxouwhn you um>mmmw muma mcowuomum mumauouam mo muoavaam H1 ooN new as N was masfio> coauumum .aaaaou omnoumm am so vmaamam mm3 uamumnumasm any .cmmSmfiuuawu can Hon 2 HO.O ca vm>aommfiu was HmuHMumE mane .vowuu mummum can vmaoom was Ana .wam mmmv HH 3mma coalw xmumnmom .aowuuwuw HH xwmn ooHIo .93 m3. wou¢a>moo§moclmm mo znawuwouwaounu Suzaou omno xmvwzmmmlmm .qH muswfim 55 «H whamah umnaaz aouuomum mm ae ¢m . 1 cc Houz an ma v no.o oa.o mH.o a: ocm< mm: o~.o mN.o were similar. The major difference was that there was more glucose left in the HF experiment than HF/pyr, whereas there was more galactose in HF/pyr than in HF. The efficiency of deglycosylation of other sugars by both reagents was almost same. The final yield of SP-C-SO fractions was also almost same (Table 10). So HF/pyr is thought to be as efficient as anhydrous HF for the deglycosylation of AGP(M). Further Fractionation of HF/PYr Insoluble Fraction. After HF/pyr treatment of AGP(M) and dialysis, there was insoluble material which was freeze dried, and attempts made to extract hydroxy- proline because it had a lot of hydroxyproline (752 of original AGP(M), see Figure 9). Some solvents (water, 0.01 N HCl, 0.5 M NaCl and 0.1 N NHAOH) were tried to extract hydroxyproline. As 0.5 M NaCl gave the most hydroxyproline extraction (60% extracted), the insoluble material was extracted with 0.5 M NaCl twice (10 ml/gm). After centrifuging to remove the insoluble material, the supernatant was gel-filtered on Sephadex G-100 (2.6 x 80 cm, bed volume 400 m1) equilibrated with 0.5 M NaCl. The elution profile is given in Figure 15. Three peaks contained hydroxyproline. One of these three peaks (III) was chosen for further fractionation. This fraction was dialysed against 0.01 N HCl for 2 days in the coldroom with 6 changes of 0.01 N HCl, the dialysate was loaded on SP-Sephadex C-SO column as described earlier. Three peaks were obtained, one at washing, and two within the gradient (0.2 M and 0.5 M NaCl region). The hydroxyproline distribution in these three peaks were 182, 332 and 492. The two major peaks which appeared at 0.2 M and 0.5 M NaCl were analysed for sugar and amino acid (Table 13). Amino acid analyses suggested that first peak in the gradient may come from AGPs because of its high alanine content and the second peak may come 57 .mc«aoummxouv>s you commons mums mnofiuomum mumEmuHm mo 329:0 H1 com com nouooaaoo 0.33 mnoauomum Ha m>am .Homz z m.o sues umumunfiafisvm canaou coauu xocmsamm Bo c.~ x cm no no mcoaumowamam monumamm unom up woumcoauomuw mos Aae omv uomuuxm was .mofiau Homz z m.o mo Ha 0H suwa wouomuuxm mm3 Aaw av szmo< aoum HmHumuwa mannaomnw u%a\mm .Humz z m.o no“: waauomuuxo nmumm Hmaumums manoaomaa u>m\m= mo soaumuuafim Hmo .mH shaman 58 on -}O VD mH munwam wonsnz cowuomwm om co on ON 1 HHH HH H vao> a \ o mo.o OH.o an oom< a»: nH.o o~.o 59 Table 13. Chemical Composition of AGP(M)/HF/pyr/Insolub1e/NaCl Extract/G~100 III/ Fra¢t19nated on S?-C+50 . Fraction Gradient I Gradient II Sandal Lectina NaCl Concentration - (M) 0.2 0.5 - 'Sugar, mole 2 (moles per 1 mole HyP) Ara 0 (0) 0 (0) Rha 0 (0) O (O) Fuc 0 (O) O (0) Xyl 0 (0) 0 (0) GalU 0- (0) 0 (0) Man 49-1 (1.54) 0 (0) Gal 29.7 (0.93) 82.1 (0.17) Glc 21.2 (0.66) 17.9 (0.04) ‘Amino Acids, mole 2 Hyp 9.0 7.2 5.9 Asp 13.0 10.0 10.6 Thr 8.7 4.7 4.8 Ser 14.6 5.9 5.9 Glu 5.9 11.0 12.1 Pro n.d. 6.6 6.3 Gly 7.4 13.4 13.4 Ala 12.2 5.6 8.6 Val 5.8 3.9 4.1 Cys 4.3 0.3 0.9 Met 0 2.0 0.7 Ile 3.3 1.7 3.3 Leu 7.0 3.47 6.1 Tyr 0.8 1.5 3.7 Phe 1.9 4.5 3.5 Lys 1.9 8.6 4.3 His 0.6 7.6 1.3 1.6 2.1 5.8 . AIS. n.d. not determined. 8Taken from ref. (19). 60 from a lectin-like protein because the amino acid composition (especia- lly acidic amino acids) were similar to those from hydroxyproline- containing sandal lectin (Table 13). AGP(C) Deglycosylation via HF/PYridine. Jermyn and Yeow (27) suggested that secreted AGPs into the medium of suspension-cultured cells would have been exposed to the full effect of any degradation enzymes present. If so, AGP(C), which could be the precursor of AGP(M), may have less hydroxyproline than AGP(M), if enzyme preferentially degraded the hydroxyproline poor region of AGPs. A 280 mg of AGP(C) (2550 pg Hyp) was deglycosylated with 4.5 ml of HF/pyr and 0.5 ml of anhydrous methanol at room temperature for 1 hr. Further fractionation was according to Figure 9. The supernatant fraction (2400 pg Hyp) was freeze dried and extracted with 0.1 N NH40H (1480 pg Hyp extracted) and loaded on Sephadex G-100 column (bed volume 500 ml). Figure 16 shows the elution profile. The major peak (II, 830 pg Hyp) was loaded on SP-Sephadex C-SO column (30 m1 bed volume). The single hydroxyproline peak appeared at 0.39 M NaCl (Figure 17). This fraction (660 pg Hyp) was analysed for sugar and amino acid composition and the result is shown in Table 14 together with AGP(M)/ HF/pyr/Super/G-IOO III/SP-C-SO. Both sugar and amino acid compositions were similar in these two fractions, suggesting that AGP(M) would not be an extensive degradation product of AGP(C) in sycamore suspension cultures. 61 .mnnoumhxouvzz How @9933 9:5 mnofiuuoum 3253:. no muonwzm H1 com wow cmuomaaoo who: anowuooum Ha one .moqmz z H.O sue: vmumunaawovo Aao ¢.H x anv caoaou coalu Xavmsaom o no commoa on: ads NV uomuuxo any can mocmz z H.O cows wouomuuxo mos Hmauoume.mana .Awe may coauv museum was unnumaumanm one .vomnwauuamo pom vom>Hmap mm: coauoflom can .sofiuommu onu Second Ou “mums mnfivvm umuu< .munumwmnaou Eoou on a: H you AHE m.ov moo: cam Aaa m.qv u>a\m= suds woumouu mmz Ame owuv onmw< .ooHIu Noumaaom o no u>a\m= :uH3 onmo< woumH%moo%Hmov mo aoaumuuawm How .oH shaman 62 ea shaman ponenz noauomum me mm nu ma d d a a full y; »n_ >H HHH HH H vHo> fie mo.o ca.o E: oon¢ chm nH.o c~.o 63 .mcfiaouazxouczn u0w commons who? mnoauomum mumoumuam mo muonvfiam H1 com com nmuooafioo muo3 mnoauomuw HE 039 .casaoo omIUImm am no uoaammm mm3 uomuuxm one .Hum z HO.O sues vmuomuuxo was .vofiuu museum and umaoon «ma AcH .wfim momv HH xmom OOHIU xovmsamm .noauomuu HH some ooHIu .u>m\mm cues onmu< umuMHhmoohawmc mo xsmmuwoumEounu :Esaoo onto xmcmnammlmm .NH enemas 64 NH shaman Honenz coauooum o m¢.o o~.o an oom< ah: mH.o o~.o 65 Table 14. Chemical Composition of Major Peaks of HF/pyr Deglycosylated AGP(C) . ”HAFSHH..H_M and AGP(M) Fraction AGP(C)/HF/pyr/Super/ AGP(M)/HF/pyr/Super/ 0-100 II/SP-C-SO 04100 III/SP-C-SO Concentration ~ of NaCl CM) 0.39 0.15 '§ggar, mole z (moleS'per'l mole Hyp) Ara 10.2 (0.12) 0 (0) Rha 0 (0) O (0) Fuc 0- (0) 0 (0) Xyl 5.8 (0.07) 6.4 £0.03) GalU 0 (0) 0 (0) Man 18.5 (0.23) 18.7 (0.11) Gal 28.5 (0.36) 47.1 (0.26) Glc .37.0 (0.47) 27.9 (0.15) ' Amino Acids, mole Z Hyp 14.0 17.4 Asp 13.4 11.5 Thr 7.6 8.3 Ser 11.0' 16.3 Glu 10.8 6.9 Pro 5.0 n.d. Gly 5.7 5.0 Ala 12.7 11.9 Val 7.1 5.9 Cys O 0.6 Met 0 1.4 Ile 2.1 5.0 Leu 3.6 2.9 Tyr 0.5 0.4 Phe 0.7 1.9 Lys 4.4 3.1 His 0.6 0.6 0.9 1.0 n.d. not determined. DISCUSSIONS The major conclusions in this work are as follows. 1. HF in pyridine deglycosylates arabinogalactan proteins (AGPs, from the medium of suspension-cultured sycamore cells) as efficiently as does anhydrous HF. Considering the ease of handling, HF/pyr is a better reagent for the deglycosylation of AGPs than anhydrous HF. 2. The efficiency of deglycosylation of HF/pyr depends on the tempera— ture and reaction time. The best result was obtained by treating for 1.5 hrs at room temperature (more than 902 of sugars were removed). One hr solvolysis at room temperature was almost as efficient as that of 1.5 hrs at room.temperature. 3. The major sugars which remained undialysable after HF/pyr deglycosy- lation were galacturonic acid and glucose which were almost completely removed from the protein portion after gel-filtration on Sephadex G-100 and ion-exchange chromatography on SP-Sephadex C-50. 4. After SP-Sephadex C-SO column chromatography, the partially purified AGPs contained less than 1 residue of total sugars per 1 mole of hydroxyproline. The amino acid analysis of the final fraction showed that hydroxyproline, serine, alanine and aspartic acid are predominant amino acids. These four accounted for 572 of the total amino acids. 5. This final material entered the polyacrylamide gel but gave a rather diffuse band. This was confirmed by prelabelling the protein with FITC and 14C-proline and by eluting from the gel followed by 66 67 amino acid analysis. 6. Arabinogalactan proteins isolated from the cytoplasm of sycamore cells were also deglycosylated with HF/pyr. The final fraction after SP-C-SO showed almost the same chemical composition as those of AGPs from the culture medium, which suggested that they may be the same and the AGPs of the culture medium was ng£_an extensive degradation product of cytoplasmic AGPs. 7. Hydroxyproline-glucose may be present in AGPs. This was obtained as an HF/pyr "semi-resistant" hydroxyproline-glycoside together with hydroxyproline-galactose by Biogel P2 column chromatography after Ba(0H)2 hydrolysis of partially deglycosylated AGPs. The identification of Hyp-Glc is in progress. In order to examine the protein portion of AGPs, they need to be deglycosylated first because of their high carbohydrate content. Chemical deglycosylation was tried by using anhydrous HF, which is examined by Mort and Lamport (34), but here it was modified by the use of 70% HF in pyridine. First the optimum condition was determined. Time course experiment (Table 6) and temperature experiment (Table 7) showed that HF/pyr treatment for 1 hr at room temperature removed over 902 of sugars from AGPs. The control experiment by use of anhydrous HF showed no significant difference from the use of HF/pyr. At room temperature HF/pyr removed almost all of Ara, Rha, Fuc and Man from AGPs within 30 min, and Xyl and Gal were removed within 90 min almost completely. Certain amounts of GalU and Glc were present undialysable even after 4 hrs HF/pyr treatment. But the polyuronide was completely removed from the protein by SP-C-SO column chromatography, indicating the 68 polyuronide does not bound covalently to the protein portion. Almost all of glucose was removed from the protein after partial purification by Sephadex G-100 and SP-C-SO. At this stage more than 992 of sugars were removed. The great advantage of HF/pyr is that it is easier to handle than liquid HF and it does not require special apparatus. If HF/pyr degly- cosylates other glycoprotein as it does for AGPs, it will become a good method for chemical deglycosylation. The only problem will arise if the polyuronide is going to be deglycosylated. In this case anhydrous HP will be better than HF/pyr. The deglycosylated AGPs were partially purified. The final fraction contained less than 1 redidue of sugar per 1 mole of hydroxyproline. Cytoplasmic AGPs were also isolated and deglycosylated. The partially purified fraction showed the similar chemical compositions to those from medium AGPs. Both AGPs may be the same and the medium AGPs would not be an extensive degradation product of cytoplasmic AGPs. The sequen- cing analysis of this partially purified deglycosylated AGPs is now in progress. There were HF/pyr "resistant" hydroxyproline-glycosides in AGPs. They were obtained by gel-filtration on Biogel P2 column and appeared at the region of Hyp-Ara after HF/pyr treatment of AGPs followed by 1 Ba(OH)2 hydrolysis. Sugar analysis of this fraction showed the presence of galactose (802) and glucose (20%), suggesting the possibili- ty of Hyp-Glc as well as Hyp-Gal as HF/pyr "resistant" hydroxyproline- glycosides. If Hyp-Glc is identified, that will be the first finding of this linkage in plant. What is the function of AGPs? There are at least three possibilities. 69 First is that it would be the precursor of cell wall protein "extensin". If an enzyme removes hydroxyproline-poor alanine-rich tail portion from AGPs, the hydroxyproline-rich region could become extensin. Amino acid sequence of AGPs will give some information. Second is that it could play a role in the cell-cell recognition processed in plants, because they are extracellular macromolecules, they are glycoproteins, and they are ubiquitous in higher plants. Third possibility is that they could be the plasticiser in cell wall network to cross-link polysaccharides. In any case the further structural anaysis of AGPs and extensin will give more information about their functions in plants. LIST OF REFERENCES 10. ll. 12. 13. 14. 15. 16. 17. 18. 19. LIST OF REFERENCES Lamport, D. T. A. Adv. in Bot. Res. 2, 151 (1965). . Keegstra, K., Talmadge, K. W., Bauer, W. D. and Albersheim, P. Plant Physiol. 21, 188 (1973). Pope, D. G. Plant Physiol. 52, 894 (1977). Lamport, D. T. A.: in "Recent Advances in Phytochemistry" (eds. F. A. Loewus and V. C. Runeckles), New York: Plenum Press (1977), Vol. II, P. 79. . Steward, F. C. and Thompson, J. F. Ann. Rev. Plant Physiol. 1, 233 (1950). Lamport, D. T. A. and Northcote, D. H. Nature 188, 685 (1960). Dougall, D. K. and Shimbayashi, K. Plant Physiol. 32, 396 (1960). Gotelli, I. B. and Cleland, R. Amer. J. Bot. §§, 907 (1968). Lamport, D. T. A. Biochemistry 8, 1155 (1969). Lamport, D. T. A. and Miller, D. H. Plant Physiol. 48, 454 (1971). Lamport, D. T. A. Nature 216, 1322 (1967). Akiyama, Y. and Kata, K. Agric. 13101. Chem. 31, 79 (1977). Lamport, D. T. A. in "Biochemistry of Plant III", New York: Academic Press (1979, in press). Lamport, D. T. A., Katona, L. and Roerig, S. Biochem. J. 1 3, 125 (1973). Cho, Y. P. and Chrispeels, M. J. Phytochemistry 12, 165 (1976). Allen, A. K. and Neuberger, A. Biochem. J. 135, 307 (1973). Allen, A. K., Desai, N. N., Neuberger, A. and Greeth, J. M. Biochem. J. 171, 665 (1978). Horejsi, V. and Kocourek, J. Biochem. Biophys. Acta 532, 92 (1978). Mani, U. V. and Radhakrishnan, A. N. Biochem. J. 141, 147 (1974). 70 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. 34. 35. 36. 37. 38. 39. 40. 41. 71 Mani, U. V. Personal Communication. Hori, H. and Sato, S. Phytochemistry 12, 1485 (1977). Kata, K., Watanabe, F. and Eda, s. Agric. Biol. Chem. Q, 533 (1977). Anderson, R. L., Clarke, A. E., Jermyn, M. A., Knox, R. B. and Stone, B. A. Aust. J. Plant Physiol. 4, 143 (1977). Lamport, D. T. A. Ann. Rev. Plant Physiol. 21, 235 (1970). Hori, H. and Fujii, T. Plant and Cell Physiol. 12, 1271 (1978). Fincher, G. B. and Stone, B. A. Aust. J. Biol. Sci. 21, 117 (1974). Jermyn, M. A. and Yeow, Y. M. Aust. J. Plant Physiol. 2, 501 (1975). Knox, R. B., Clarke, A., Harrison, 8., Smith, P. and Marchalonis, J. J. Proc. Natl. Acad. Sci., USA 12, 2788 (1976). Loewus, F. and Labarca, C.: in "Biogesnesis of Plant Cell wall Polysaccharides" (eds. F. Loewus), Academic Press. P. 175 (1973). Aspinall, G. O. and Rosell, K. G. Phytochemistry $1, 919 (1978). Larkin, P. J. J. Cell. Sci. 29, 283 (1978). Clarke, A. E. and Knox, R. B. Quart. Rev. Biol. 22, 3 (1978). McNamara, M. K. and Stone, B. A. AGP News 1, 16 (1977). Mort, A. J. and Lamport, D. T. A. Anal. Biochem. 22, 289 (1977). Davis, B. J. Ann. N. Y. Acad. Sci. 121, 404 (1964). Muramoto, K., Meguro, H. and Tuzimura, K. Agric. Biol. Chem. 41, 2059 (1977). Drescher, D. G. and Lee, K. 8. Anal. Biochem. 24, 559 (1978). Bhatti, T., Chambers, L. E. and Clamp, J. R. Biochem. Biophys. Acta 222, 339 (1970). Mackenzie, S. L. and Tenaschuk, D. J. Chromatography 111, 413 (1975). Wéber, K. and Osborn, M. J. Biol. Chem. 244, 4406 (1969). Blakesley, R. W. and Boezi, J. A. Anal. Biochem. 22, 580 (1977). I1‘111111111111“