CHARACTERIZAUON OF A PARTICULAR UDP -GALACTURONATE: ACCEPTOR D - GAMCTURONOSYLTRANSFERASE FROM LEMNA MINOR AND STUDIES ON THE PHYSICAL PROPERTIES 9F THE PRODUCT Dissertation for the Degree of PM. D. MICHEGAN: STATE UNIVERSHY EDWIN: DeTURK LBNBACH 1975 ._..~ ”fl—w-‘VW LIBRARY Michigan State University This is to certify that the thesis entitled Characterization of a Particulate UDP-Galacturonate: Acceptor D-Galacturonosyltransferase from Lemna minor and Studies on the Physical Properties of the Product presented by Edwin DeTurk Leinbach has been accepted towards fulfillment of the requirements for Ph. D. degree in Biochemistry Major professor Date 7 I 75 0-7639 ? amomc av g "DAB 8: SUNS' 800K BINDERY INC. LIBRARY smog RS SPRIIIBPOII. mum ABSTRACT CHARACTERIZATION OF A PARTICULATE UDP-GALACTURONATE:ACCEPTOR D-GALACTURONOSYLTRANSFERASE FROM LEMNA MINOR AND STUDIES ON THE PHYSICAL PROPERTIES OF THE PRODUCT BY Edwin DeTurk Leinbach Apiogalacturonans are pectic polysaccharides which constitute about 14% of the cell wall of Lemna minor. As a part of the study of their biosynthesis, a particulate UDP-galacturonate:acceptor D-galacturonosyltransferase, capable of incorporating D-galacturonic acid from UDP-galacturonic acid into polysaccharide, was isolated from L. minor. The D-galacturonosyltransferase activity was associated with material sedimenting between 480 and 34,8009. Incorporation of D-[U-14C]galacturonic acid into polysaccharide was constant with time for 1.5 to 2.0 min at 25°C. D-Galacturonosyltransferase activity was directly proportional to the amount of the particulate transferase preparation used, and activity was optimal in citric acid-sodium phosphate buffer at pH 6.0-6.2. The particulate D- galacturonosyltransferase had an apparent Km for UDP-galacturonic acid of 8 uM. The presence of 10 mM MnCl 0.4 M sucrose, and 1% 2' bovine serum albumin in the particulate D—galacturonosyltransferase Edwin DeTurk Leinbach preparation was required for both optimum transferase activity and for maximum stability. The D—galacturonosyltransferase was stable for 30 min at 0°C. The transferase was stable for 2.0 to 2.5 min at 25°C, but only 10% of the initial activity remained after 10 min. In some experiments, greater than 50% of the initial activity could be retained for at least 60 h by storage of the transferase preparation at -20°C. None of the attempts to stabilize the D-galacturonosyltransferase activity by the addition of various stabilizing agents were successful. D-Galacturonosyltransferase activity was slightly inhibited by the addition of 3 uM UDP-apiose. In contrast, the D-apiosyltrans- ferase of L. minor is stimulated by the addition of UDP—galacturonic acid. In an attempt to solubilize the particulate D-galacturonosyl- transferase activity, the effects of various detergents, both ionic and non-ionic, were tested. All but Tween-20 caused almost complete loss of transferase activity at a concentration of 1.0%. Triton X-100, Tween-20, sodium cholate, and Emulgen—9ll all failed to cause more than 20% solubilization with an accompanying 50% loss of total transferase activity. These results suggest the importance of an intact membrane structure for D-galacturonosyltransferase activity. About 80% of the D-galacturonosyltransferase product was solu- bilized by extraction with 1% ammonium oxalate or 2% sodium hexa- metaphosphate at 50°C. However, upon chromatography of the two products on Bio-Gel P-300, the 2% sodium hexametaphosphate-solubilized product appeared to be considerably larger than the 1% ammonium Edwin DeTurk Leinbach oxalate-solubilized product, with 30% of the former product exceeding an apparent molecular weight of 1 x 106. The apparent size of the 1% ammonium oxalate-solubilized product decreased on standing or upon dialysis in water or 1% ammonium oxalate. The apparent size of the 2% sodium hexametaphosphate- solubilized product was unaffected by these treatments. The apparent size of the 1% ammonium oxalate-solubilized product was also increased by dialysis in 3% sodium hexametaphosphate and in 3% ammonium oxalate. The apparent size of both solubilized products was increased by elution with or by dialysis in 1.0 M NaCl. These results suggest that the D-galacturonosyltransferase products solubilized with sodium hexametaphosphate or with 1% ammonium oxalate probably do not represent the true size of the polysaccharide as it is synthesized in the cell-free system. The apparent size of the solubilized product is sensitive to the ionic strength of the medium in which the product is placed, with aggregation occurring as the ionic strength is increased. In addition, various ions appear to differ in their ability to cause aggregation. Thus the 2% sodium hexametaphosphate-solubilized product is likely to represent an extensively and tightly aggregated form of the D-galacturonosyl- transferase product, whereas the 1% ammonium oxalate-solubilized product represents a less-extensively and less tightly aggregated form. Whether or not 1% ammonium oxalate also causes partial degra- dation of the D-galacturonosyltransferase product is still in question. The decrease in size of the 1% ammonium oxalate-solubilized product on standing or upon dialysis in 1% ammonium oxalate could be inter— preted as either degradation or disaggregation. Edwin DeTurk Leinbach The amount of the D-galacturonosyltransferase product solubilized decreased with decreasing concentrations of both ammonium oxalate and sodium hexametaphosphate, although the latter salt is a better solu- bilizing agent than the former at low concentrations. Also, the apparent size of the sodium hexametaphosphate-solubilized product decreased as the total amount of product solubilized decreased. Whether this size decrease represents a decreased aggregation at lower sodium hexametaphosphate concentrations or whether it repre- sents solubilization of a different species of polysaccharide is unclear as yet. Preliminary results indicated that this product was not so sensitive to aggregation as was the 2% sodium hexameta- phosphate-solubilized product. The results presented demonstrate the ability of a cell—free system from L. minor to incorporate D-[U-14C]galacturonic acid from UDP-galacturonic acid into a polysaccharide which behaves like a pectic substance. The techniques developed here can be applied to further study of the pectic polysaccharides of the L. minor cell wall. CHARACTERIZATION OF A PARTICULATE UDP-GALACTURONATE:ACCEPTOR D-GALACTURONOSYLTRANSFERASE FROM LEMNA MINOR AND STUDIES ON THE PHYSICAL PROPERTIES OF THE PRODUCT BY Edwin DeTurk Leinbach A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Biochemistry 1975 The road runs ever on and on, Down from the door where it began. Now far away the road has run, And I must follow if I can; Pursuing it with eager feet Until it joins some wider way Where many paths and errands meet. And whither then, I cannot say. from The Hobbit by J. R. R. Tolkien To my parents, John and Dorothy, who first set me on that road, and to Susan, my wife and colleague, who keeps me pursuing that road when my feet are far from eager. ii ACKNOWLEDGEMENTS Above all, I wish to thank my advisor, Dr. Paul Kindel, for his patience, his guidance, and his support, both academic and financial, during the past five years. Also, I thank the members of my guidance committee for their advice and support--Drs. Richard Anderson, Robert Bandurski, Donald Farnum, and Robert Ronzio, and Deborah Delmer, who graciously agreed to serve as a substitute on very short notice. To my research colleagues, Drs. Leonard Mascaro, Jr., and Yuan-Tseng Pan, I am deeply grateful for their contributions of materials, techniques, and advice as well as for their camaraderie. And finally, a very special thanks to Linda Damos for her invaluable technical assistance and her patience with a very "green" supervisor. iii VITA Edwin DeTurk Leinbach was born in Reading, Pennsylvania, on September 24, 1948, the only child of John A. and Dorothy D. Leinbach. He grew up in the suburban Reading area and graduated from Exeter Township High School in June, 1966, as class valedictorian. In September of that same year he entered Ursinus College in Collegeville, Pennsylvania, where, under the guidance of several excellent and dedicated professors, he acquired a love for not only the science of chemistry and biology, but for their history and philosophy as well. His four years at Ursinus culminated in his receipt of the Bachelor of Science in Chemistry in June, 1970. In September, 1970, Ed began his graduate studies in the Depart- ment of Biochemistry at Michigan State University, and in April of 1971 he entered the laboratory of Dr. Paul Kindel. In September of that year, he met Susan Singley, a new graduate student in that department, and they were married in June, 1972. After having received his Ph.D. degree in September, 1975, Ed is now entering a year of postdoctoral work as an NIH fellow with Dr. Robert Barker while Sue finishes her Ph.D. studies under Dr. John Boezi. iv TABLE OF CONTENTS Page LIST OF TABLES . . . . . . . . . . . . . . . . . . . . . . . . . Vii LIST OF FIGURES. . . . . . . . . . . . . . . . . . . . . . . . . viii LIST OF ABBREVIATIONS o o o o o o o o o o o o o o o o o o o o o o x STATEMENT OF THE PROBLEM . . . . . . . . . . . . . . . . . . . . 1 LITERATURE REVIEW. . . . . . . . . . . . . . . . . . . . . . . . 3 Structure of the Plant Cell Walls . . . . . . . . . . . . 3 The Primary and Secondary Walls. . . . . . . . . . 3 Extraction of the Various Classes of Polysaccharides. . . . . . . . . . . . . . . . . 5 Cellulose. . . . . . . . . . . . . . . . . . . . . 7 Hemicelluloses . . . . . . . . . . . . . . . . . . 8 The Hydroxyproline-Rich Glycoproteins. . . . . . . 9 Pectins. . . . . . . . . . . . . . . . . . . . . . 9 Overall Organization of the Cell Wall. . . . . . . l3 Polysaccharide Biosynthesis . . . . . . . . . . . . . . . 16 Sugar Nucleotides. . . . . . . . . . . . . . . . . 16 Synthesis of Pectins . . . . . . . . . . . . . . . 19 Site of Synthesis. . . . . . . . . . . . . . . . . 20 Solubilization of Particulate Enzyme Activities. . 22 Lipid Intermediates. . . . . . . . . . . . . . . . 24 EXPERIMENTAL PROCEDURES. . . . . . . . . . . . . . . . . . . . . 27 Materials . . . . . . . . . . . . . . . . . . . . . . . . 27 General Methods . . . . . . . . . . . . . . . . . . . . . 28 Electrophoresis and Paper Chromatography. . . . . . . . . 29 Preparation of UDP[U-14C]GalA . . . . . . . . . . . . . . 30 Culture of Lemna minor. . . . . . . . . . . . . . . . . . 31 Isolation of the Particulate D-Galacturonosy1transferase. 31 Preparation. . . . . . . . . . . . . . . . . . . . 31 Standard Assay for UDPGalA:Acceptor D-Galacturono- syltransferase. . . . . . . . . . . . . . . . . . . . . 32 Solubilization of the D-Galacturonosyltransferase Product with Ammonium Oxalate and with Sodium Hexametaphosphate . . . . . . . . . . . . . . . . . . . 33 RESULTS. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34 Partial Characterization of UDPGalA:Acceptor D-Galacturonosyltransferase . . . . . . . . . . . . . . 34 V Linearity of the D-Galacturonosyltransferase Assay. . . . . . . . . . . . . . . . . . . . . Effect of pH on D-Galacturonosyltransferase Activity . . . . . . . . . . . . . . . . . . Effect of Various Ions on D-Galacturonosyl- transferase Activity . . . . . . . . . . . . . Effect of Sucrose on D-Galacturonosyltransferase Activity . . . . . . . . . . . . . . . . . . Determination of the Apparent Km for UDPGalA . Stability of D-Galacturonosyltransferase . . . . Effect of UDP-Apiose on D-Galacturonosyl- transferase Activity . . . . . . . . . . . . . Effects of Detergents on Activity and Solubilization . . . . . . . . . . . . . . . . Characterization of the Product of the D-Galacturono- syltransferase Reaction . . . . . . . . . . . . . . . Efficiency of Solubilization . . . . . . . . . . Size of the Solubilized Product. . . . . . . . . Stability of the Solubilized Products. . . . . . Induction of Aggregation . . . . . . . . . . . . Aggregation Induced by 1.0 M NaCl. . . . . . . . Solubilization of D-Galacturonosyltransferase Product with Various Concentrations of Sodium Hexameta- phosphate and Ammonium Oxalate. . . . . . . . . . . DISCUSSION Characteristics of UDPGalA:Acceptor D-Galacturono- syltransferase. . . . . . . . . . . . . . . . . . . . Stability, Solubilization, and the Role of the Particle Which Contains UDPGalA:Acceptor D-Galacturonosyltransferase . . . . . . . . . . . . . Similarity of the Solubilized D-Galacturonosyltrans- ferase Product to Cell Wall Apiogalacturonans . . . . Size of the Product . . . . . . . . . . . . . . . . . Relationship Between Aggregation and Galacturonan Structure . . . . . . . . . . . . . . . . . . . . . Summary . . . . . . . . . . . . . . . . . . . . . . . . LIST OF REFERENCES I O O O O I O O O O O O I O O O O O O O O I vi Page 34 34 39 41 41 41 46 49 51 51 53 60 64 76 81 93 93 95 97 98 101 103 105 Table LIST OF TABLES Effect of various ions on the incorporation of D-[U-14C]galacturonic acid into D-galacturonosyl— transferase product . . . . . . . . . . . . . . . Effect of various detergents on D-galacturonosyl- transferase activity. . . . . . . . . . . . . . . Efficiency of solubilization of D-galacturonosyl- transferase product with ammonium oxalate and sodium hexametaphosphate . . . . . . . . . . . . . . . . Effect of sodium hexametaphosphate concentration on the size of the solubilized product . . . . . . . vii Page 40 50 52 85 Figure 10 ll LI ST OF P IGURES Procedures for extraction of plant cell walls . . . . . Model for the organization of the polysaccharides of the primary cell wall of suspension-cultured sycamdre cells 0 O O O O O O O O O O O O O O O C I O O O O O O O The effect of enzyme concentration on the rate of incorporation of D-[U-14C]galacturonic acid into D-galacturonosyltransferase product . . . . . . . . . . The effect of time and temperature on the incorporation of D-[U-14C]ga1acturonic acid into D-galacturono— syltransferase product. . . . . . . . . . . . . . . . . The effect of pH on the rate of incorporation of D-[U-14C]galacturonic acid into D-galacturonosyl- tranSferase prOdUCt O O O O O O O O O O I O O O D O O O The effect of MnC12 on the rate of incorporation of D—[U-14C]galacturonic acid into D-galacturonosyl- transferase product . . . . . . . . . . . . . . . . . . The effect of sucrose on the rate of incorporation of D-[U-14C]galacturonic acid into D-galacturonosyl— transferase product . . . . . . . . . . . . . . . . . . The effect of UDPGalA concentration on the rate of incorporation of D-[U-14C]galacturonic acid into D-galacturonosyltransferase product . . . . . . . . . . The effect of UDP-apiose on the incorporation of D-[U-14C1galacturonic acid into D-galacturonosyl- transferase product . . . . . . . . . . . . . . . . . . Chromatography of 1% ammonium oxalate-solubilized product and 2% sodium hexametaphosphate-solubilized product on Bio-Gel P-300. . . . . . . . . . . . . . . . Chromatography of 2% sodium hexametaphosphate- solubilized product on Sepharose 6B and rechroma- tography of Fraction 11 on Sepharose ZB . . . . . . . . viii Page 14 36 36 38 43 43 45 48 55 57 Figure Page 12 Chromatography of 2% sodium hexametaphosphate- solubilized product on Bio-Gel P-300 and rechroma- tography of Fraction 10 on Sepharose 4B . . . . . . . . . 59 13 Decrease in size of the 1% ammonium oxalate-solubilized product on standing at 4°C and the effect of dialysis in water on this decrease . . . . . . . . . . . . . . . . 63 14 Effect of dialysis in 1% ammonium oxalate on the size of the 1% ammonium oxalate-solubilized product. . . . . . 66 15 Effect of dialysis in 1% ammonium oxalate on the size of 1% ammonium oxalate—solubilized product that was dialyzed in water . . . . . . . . . . . . . . . . . . . . 68 16 Effect of dialysis in 3% sodium hexemetaphosphate on the size of 1% ammonium oxalate-solubilized product . . . 71 17 Effect of dialysis in 3% ammonium oxalate on the size of the 1% ammonium oxalate-solubilized product which was dialyzed in 1% ammonium oxalate . . . . . . . . . . . 73 18 Effect of dialysis in 1.0 M NaCl on the size of the 1% ammonium oxalate-solubilized product . . . . . . . . . 75 19 Chromatography of 2% sodium hexametaphosphate- solubilized product on Sepharose 48 with 50 mM sodium phosphate buffer and with 1.0 M NaCl . . . . . . . 78 20 Chromatography of 1% ammonium oxalate-solubilized product on Sepharose 48 with 50 mM sodium phosphate buffer and with 1.0 M NaCl. . . . . . . . . . . . . . . . 80 21 Effect of dialysis in water for 3 h on the size of 1% ammonium oxalate-solubilized product which was aggre- gated by dialysis in 1.0 M NaCl . . . . . . . . . . . . . 83 22 Effect of sodium hexametaphosphate concentration on the size of the sodium hexametaphosphate—solubilized product . . . . . . . . . . . . . . . . . . . . . . . . . 87 23 Effect of ammonium oxalate concentration on the size of the ammonium oxalate—solubilized product . . . . . . . 89 24 Effect of 1.0 M NaCl on 2% sodium hexametaphosphate— solubilized product that elutes near Vt from Sepharose 4B. O O O O I O O O O O O I O O O O O O O O O O O O O O O 91 ix ADP BSA EDTA GDP WI SD dTDP Tris U UDP UDPGalA LIST OF ABBREVIATIONS adenosine 5'-diphosphate bovine serum albumin ethylenediaminetetraacetate, disodium salt guanosine 5'-diphosphate the Michaelis constant statistical probability weight-average degree of polymerization standard deviation thymidine 5'-diphosphate tris(hydroxymethyl)aminomethane uniformly labeled with radioactivity uridine 5'-diphosphate uridine 5'-(a-D-galactopyranosyluronic acid pyrophosphate) uridine 5'-triphosphate void volume total volume or column volume STATEMENT OF THE PROBLEM Despite the theoretical importance of the pectic polysaccharides in determining the physical properties of primary plant cell walls, little is known about their biosynthesis. No cell-free system has been described in which the incorporation of both the neutral and the acidic sugars of pectins has been studied. The cell wall of Lemna minor contains large amounts of pectic heteropolysaccharides termed apiogalacturonans, the structure and physical properties of which have been described in part. Recently, a particulate enzyme preparation was isolated from L. minor which incorporates D-apiose from UDP-apiose into polysaccharides resembling these apiogalacturonans (64-66). If the enzyme activity responsible for the incorporation of D-galacturonic acid into these same poly- saccharides could also be isolated and characterized, L. minor would become the first system in which the incorporation of both the acidic and neutral sugars of a pectin has been demonstrated. In many of the cell-free polysaccharide synthesizing systems already described, the physiological significance of the reactions studied in vitro is uncertain since the presence in the cell wall of polysaccharides of the type synthesized was not determined. By comparing the properties of the polysaccharides synthesized from UDPGalA and from UDP-apiose in vitro with those of authentic apio- galacturonans it should be possible to remove some of this uncertainty. 1 2 t The research described in this dissertation was undertaken to isolate and characterize the D-galacturonosyltransferase from L. minor, to relate its properties to those of D-apiosyltransferase, and to develop procedures for its solubilization so that the nature of its ultimate acceptor might be defined. To compare the D-galacturo- nosyltransferase product with the D-apiosyltransferase product and with authentic apiogalacturonans, the size and some of the physical properties of the D-galacturonosyltransferase product were investi- gated, and procedures were developed which can be used to study further the physical properties of the cell wall pectic substances. LITERATURE REVIEW Structure of the Plant Cell Wall The Primary and Secondary Walls The periphery of most cell types is surrounded by a coat of glycosubstances, sometimes referred to as a glycocalyx. In plants this glycocalyx has reached perhaps an extreme of development. The plant cell wall is almost entirely carbohydrate with protein consti- tuting less than 10% of the total mass. The basic plant cell wall structure is a microfibrillar array embedded in a complex heterogeneous polysaccharide matrix. In the mature cell this structure also contains a high molecular weight polyphenol, lignin. Ultrastructurally, the plant cell wall is divided into several layers based upon the content and orientation of the microfibrils. Proceeding from the middle lamella toward the cell membrane, one finds first the primary cell wall, a loose, random net of micro- fibrils in a pectin-hemicellulose matrix. This wall is laid down during cell division and elongation. It is the only cell wall layer found in soft plant tissues (1). The secondary wall is divided into three sublayers based upon the orientation of the numerous large and highly-oriented micro- fibrils. This layer, which is formed during thickening and 4 differentiation, provides greater strength and resistance than the more elastic primary wall (1). The various components of the cell wall are modified in composi- tion and ultrastructure in response to the changing conditions of growth and development. However, the basic arrangement of a fibrillar array in a complex matrix is maintained (2). This modification of the cell wall components may occur in situ or may result from deposi- tion of different polysaccharides at different stages of growth (3). The plant cell wall participates in many aspects of plant physiology: ultrastructure, growth, development, and host-pathogen interactions. For a long time, however, no clear idea of the molecular organization of this structure existed. With the development of gas- 1iquid chromatography, methylation analysis, and mass spectrometry, the various components are being resolved and defined to the point that a model for the interaction of various components can be proposed (4-6). The microfibrillar structural material of higher plants is almost invariably cellulose. The molecular weight of primary wall cellulose is rather low and the degree of polymerization is more random than that of the higher molecular weight secondary wall cellulose (l). The microfibrillar array is largely responsible for giving shape and strength to the cell wall. In lower plants such as yeast and some algae the microfibrils are composed of mannans. In other algae the microfibrils are formed from helical 8-1.3-linked xylans (1). In higher plants the variability in fibril composition is replaced by a variability in the structure and composition of the 5 hemicelluloses and pectins among plant species as well as between the two cell wall layers (1). Some of these polysaccharides are linear while others are highly branched. Because they are not primary gene products they are often polydisperse with respect to both size and carbohydrate content. Also, some are present only at certain stages of growth. All these factors combine to make studies of the structure of the matrix very complex (2,3). The matrix formerly was considered to be merely an "amorphous" filler for the microfibrils. Recent studies, however, have indi- cated that its structure may be considerably more ordered (4-7). In the final stages of secondary wall formation, lignification occurs. Lignin is a high molecular weight polyphenol containing oxidized p-hydroxycinnamyl alcohols. Its polymerization and deposi— tion are purely chemical reactions with no enzymatic involvement (1,2). Several thorough reviews of the chemistry of lignin are available (8,9), and its structure and synthesis will not be dis- cussed further in this dissertation. Extraction of the Various Classes of Polysaccharides The various classes of cell wall polysaccharides are defined more by the methods used to extract them from the cell wall than by their composition. A summary of these methods is presented in Figure l. The delignification steps are eliminated for tissues containing only primary cell walls (11). Pectins are removed from the primary wall by a combination of water extraction and treatment with chelating agents such as EDTA, .mHHmz Haoo unwam no coauomuuxo MOM mousvoooum .H ousmfim :Hmuoumouaam mcmuomammocflnmum . omoHsHHwO Hamz Haoo mamcqmfioozamouomamm / \ mGMGGMEoosHm mmoHSHHooHEwm omoHsHHoond n onoflmwm manahx omoaamo somouufi: nouns Aoa mm oumuon asapOm ..m.ov flamxam ousawm afiuomm wannaom £90m oonsHHoooHom n msvfimwm P - Aflum>flumv osmflmom moumcooonn cansHom r - ouflnoHsoom>£_o:HmeHm mHocmnm Ucm mcfimfia wandaow osowmwm p . ocmncom\ao:mnum —0H"om mHHoo vofluu macs: 7 ammonium oxalate, and sodium hexametaphosphate (11). The latter has been reported to cause minimal degradation of pectins (12). The most important point to keep in mind is that each tissue studied requires its own modifications of the listed procedures in order to minimize degradation and other artifacts of isolation. Cellulose Cellulose is operationally defined to be the alkali-insoluble polysaccharide fraction from delignified cell walls. As isolated, it is seldom a pure B-l,4-glucan, but contains small amounts of other neutral sugars (3) which may result from the presence of enzymes capable of converting glucose to other sugars in situ (1,13). Alternatively, these sugars could belong to polymers which are merely adsorbed to the surface of the cellulose microfibril, or they could serve as a keying material for fibril-matrix interactions (2). From viscosity measurements, P; of total cell wall cellulose has been found to be 8,500 to 9,500. P; for primary wall cellulose is approximately 1,500 while that for secondary wall cellulose is 11,500 in cotton, flax and spruce (14). In the microfibril, the cellulose chains have long been assumed to have an antiparallel arrangement as in chitin (2). Recent evidence suggests, however, that this may be incorrect (15). X-ray diffraction has shown that the microfibrils possess both crystalline and amorphous regions. This finding has been interpreted in various ways, and numerous models of fibrillar structure have been proposed (l-3,ll,15). The precise ultrastructure of the microfibril is still in dispute. Hemicelluloses Among the matrix polysaccharides, the hemicelluloses are a group of neutral polysaccharides classified together on the basis of their method of extraction, illustrating a point made earlier. They are structurally heterogeneous, sharing none of the common properties of the pectins and cellulose (11). Several complete reviews of hemicelluloses are available (1,3,16-19), so only a brief listing will be presented here. The xylans contain B-l,4-linked main chains of D-xylose with short, single-residue side chains of L-arabinose or D-glucuronic acid and its esters. Xylans comprise the bulk of the hemicelluloses in angiosperms (3) in which they are up to 50% acetylated at the 2- or 3-position of the D-xylose residue (20). The galacto- and glucomannans are random copolymers of D- galactose or D-glucose in B-l,4-linkage with D—mannose. As in the xylans, the D-mannose residues are partially acetylated. These poly- saccharides of P; of approximately 100 to 200 are the major hemicellu- loses of the gymnosperms (3). Callose is a B-l,3-glucan elaborated in the development of phloem and in wound tissue (3). Because of its solubility properties, it is grouped with the hemicelluloses. Two types of polysaccharide fit into neither the hemicellulose nor the pectin group, although they are often elaborated with the latter (21). The arabinogalactans, which are most abundant in coni- fers, consist of 8-1,3- and B-l,6-linked galactan main chains with L-arabinose mono- or disaccharide side chains. The xyloglucans have 9 a B-D-glucan core with single D-xylose side chains which may be further substituted with L-fucose or D—galactose (16). The Hydroxyproline-Rich Glycoproteins The first report of a plant cell wall glycoprotein rich in hydroxyproline (hyPro) appeared in 1960 (22). At first it was suspected that this protein might merely be adsorbed to the cell wall polysaccharides during isolation. Now, however, the hyPro glycoprotein, termed "extensin" by Lamport, has been shown to be a definite wall component (10,23). It contains galactose in O-glycosyl linkage to serine as well as L-arabinose linked to hydroxyproline. In tomato, the glycosylated portion of the protein contains repeating units of the pentapeptide ser(hyp)4 (24,25). Extensin is viewed as being a cross-linker in the cell wall with a role in cell extension (10). Pectins Pectins are matrix polysaccharides found mainly in primary cell walls where they help determine the physical properties of the wall which are important in elongation. Characteristically they are acidic polysaccharides, although a group of water-soluble arabino- galactans is generally classified as pectins (3). With the exception of the Characeae and the Zosteraceae, both seaweeds, and one or two other sources, pectins are found exclusively in terrestrial plants (11 ,26) . The basic structure of the main chain of acidic pectins is an a-1,4—linked galacturonan interspersed with L-rhamnose. Side chains 10 may contain L-arabinose, D-galactose, D-xylose, L-fucose, or D-apiose depending upon the plant species (3). Pectins may be divided into two subgroups: those containing large "neutral blocks" of L-arabinose and D-galactose along the main chain, and those containing the neutral sugars in short, randomly- distributed side chains. These subgroups can be separated by electrophoresis, the latter subgroup being the more acidic because of its higher proportion of D-galacturonic acid residues (2,3). The pectins which contain the large neutral blocks might be char- acteristic of more mature primary cell walls (27,28). D-Galacturonic acid in pectins is often methyl esterified and acetylated. The degree of esterification and acetylation can be used to modulate the physical properties of pectins. Other changes in structure and composition may occur in response to differing demands of growth (12). Some evidence suggests that these changes may occur by transglycosylation within the matrix (29). The presence of rhamnosyl residues within the main chain causes kinks in an otherwise extended ribbon-like polysaccharide. Kinking was predicted on the basis of mathematical models (30) and was con- firmed in an analysis of the rhamnogalacturonan of sycamore cells (5,7).) The linear portions of the chain may form bundles while the kinked regions form a network which holds water molecules (31). The ionic nature of the galacturonans also allows salt forma- tion with calcium envisioned as the most important contributor to in vivo pectin structure. The physical properties of pectins in the presence of monovalent cations may be expected to differ from those observed in the presence of divalent cations (2). 11 Crystal structure analysis of calcium galacturonate has shown that the 9-coordinate calcium ion can interact with the C-6 carboxyl group and the 0-5 ring oxygen of one galacturonic acid residue, the hydroxyl groups of adjacent residues, and waters of hydration (32). By regulating its degree of hydration, a group of pectin molecules can transform itself reversibly from a fairly rigid gel to a highly viscous solution (2). Because they contain a variety of glycosidic linkages, pectins are especially susceptible to chemical modification during isolation. Acid precipitation removes neutral sugars and oligosaccharides (11, 33), while alkaline pH causes saponification of methyl esters and cleavage of the polyuronide chain (11). Trans- or B-elimination will occur even at neutral pH when polyuronides containing methyl esterified residues are heated (34). All these factors must be considered in devising extraction procedures. At pH values of 4 to 6.5, EDTA is ineffective in extracting pectins, whereas ammonium oxalate may cause decarboxylation of uronic acids (35). Sodium hexametaphosphate chelates best at pH 4 and has been shown to be a suitable extractant for apple fruit pectin (28, 36). In contrast, in lemon peel it caused deesterification and some B-elimination (l6). Barrett and Northcote separated apple fruit pectin into a low molecular weight component which was an almost-pure galacturonan, and a high molecular weight polysaccharide (27). The latter component contained about 30% D-galacturonic acid and large amounts of neutral sugar, particularly L-arabinose and D-galactose. After B-elimination of this component, L-rhamnose was isolated linked to L-arabinose and 12 D-galactose oligosaccharides. Since L-rhamnose is a component of the galacturonan chain, it seemed to be the point of attachment for the neutral sugars. This was the first demonstration that pectins contain blocks of neutral sugar attached to the main chain at specific sites. Conversely, the pectin of sycamore callus and other rapidly growing tissues is mainly a rhamnogalacturonan with short D-xylose side chains and no neutral blocks (12). This supports the hypothesis that neutral blocks are characteristic of the pectins of mature cells while galacturonans with short, randomly-distributed side chains are more typical of the pectins of rapidly growing and divid- ing cells (11). Methylation analysis has confirmed the frequent substitution of L-rhamnose by neutral sugars at the C-4 position (4,37). L-Rhamnose residues, therefore, might be recognition sites for attachment of the neutral blocks (11). The structure of the cell wall pectins of suspension cultured sycamore cells has been analyzed recently by Talmadge et a1. (4). Difficulty in derivitizing D-galacturonic acid has delayed such analyses, but the development of an improved gas-liquid chromato- graphic technique (38) and the application of methylation analysis (39) and mass spectrometry (40) now permit thorough structural studies of pectins. Talmadge et al. showed that the presence of 1,2—linked L-rhamnose units kinks the galacturonan chain as predicted by the mathematical models (4,31). In addition, attachment of neutral sugars to C-4 of about 50% of the L-rhamnose residues creates "Y-" shaped branches. 13 The distribution of L-rhamnose within the chain is not random; rather, it occurs as rhamnosyl-9alacturonosyl-rhamnose trisaccharides which are separated by eight D-galacturonic acid residues. Hopefully, this is only the first of many studies on the structure of a variety of pectins which will clarify the concepts of pectin structure discussed in this review. Galacturonans containing the neutral branched-chain aldopentose D-apiose have been isolated only from the aquatic angiosperm Lemna minor and three species of Zosteraceae (33,41). The apiogalacturonans of L. minor contain a galacturonan main chain with side chains of the disaccharide apiobiose and small amounts of D-apiose (41). No other neutral sugars were detected in these polysaccharides, but Beck has found D-xylose and D-galactose in apiogalacturonans (43). The L. minor apiogalacturonans can be subfractionated on the basis of solubility in NaCl. The various subfractions differ only in their D-apiose content which ranges from 7.9% to 38.1% (42). This differential solubility is probably based on the ability of D-apiose to interfere with the formation of ionic bonds between the carboxyl groups of D-galacturonic acid residues and Na+ and C1- which would tend to aggregate or precipitate the galacturonan chains. Overall Organization of the Cell Wall The data which have been gathered by analysis of the cell wall of suspension-cultured cells (4,5) have led Albersheim to propose the model for cell wall organization pictured in Figure 2 (6,7,15). Each cellulose microfibril in the primary wall is thought to be coated by a sheath of xyloglucan one molecule thick. The D-glucose a ‘5‘ 14 Figure 2. Model for the organization of the polysaccharides of the primary cell wall of suspension-cultured sycamore cells (15). 15 molecules of the xyloglucan are hydrogen—bonded to the D-glucose molecules of the cellulose on the surface of the microfibrils. Pro- truding D-galactosyl and L-fucosyl modifications of the D-xylose side chains prevent further hydrogen bonding on the side of the xyloglucan away from the microfibril. At its reducing terminus, each xyloglucan is covalently linked to a single molecule of arabinogalactan. These chains are thought to radiate from the coated microfibril like the spokes of a wheel. The arabinogalactans, in turn, are each covalently linked to a L-rhamnose residue of the rhamnogalacturonan at a point at which the latter chain kinks. By receiving numerous chains of arabinogalactan originating from different microfibrils, the rhamnogalacturonan cross-links the cell wall into a fairly rigid matrix (15). This model ignores a role for the hydroxyproline-rich glyco— proteins. Because of their L-arabinose and D—galactose content, these proteins might also cross—link the rhamnogalacturonans with other polysaccharides (10). Likewise, no role is assigned to hemi- celluloses other than xyloglucan. The model has other shortcomings. Foremost is the difficulty in rationalizing this rigid structure with a mechanism for cell elongation. Such a process would require the breaking and reforming of many covalent bonds among the cross-links. While this may seem an unnecessarily cumbersome process, the search for enzymes that could mediate such changes is in progress (15). Alternatively, while this model provides a good description of the mature primary wall, the newly-formed and elongating walls may be more plastic structures. In rapidly growing cells the physical 16 properties of the matrix polysaccharides might be more important. The most important stabilizing forces in such cases would arise from the ability of these polysaccharides to form hydrogen bonds, to form salts with ions such as calcium, and to alter the water structure of the wall by undergoing reversible gel-sol transitions. Then as the wall matured, the covalent cross-linking by proteins and polysac- charides would assume increasing importance until the rigid structure of Albersheim's model was achieved. Of course, all of this is speculation, but it reemphasizes the need to investigate carefully the structure and properties of cell walls from many different sources, from different cell types, and from cells in various stages of growth. Only then can a general model be devised which provides for both maintenance of structure and cell elongation. Polysaccharide Biosynthesis Sugar Nucleotides Sugar nucleotides appear to serve as the immediate glycosyl donors for the synthesis of oligosaccharides and polysaccharides (44). These donors, in turn, arise from the phosphorylated mono- saccharides of the photosynthetic carbon reduction cycle by the action of pyrophosphorylases and enzymes of sugar interconversion. Sugar nucleotides are favored thermodynamically as donors for poly- saccharide synthesis because their free energy of hydrolysis is higher than that of other glycosyl compounds. ’ Glycosyl derivatives of all the nucleosides have been found (11,44); however, not all sugars are linked to all nucleoside 17 diphosphates. Polysaccharide synthesizing enzymes are often specific for the nucleoside diphosphate portion as well as for the sugar portion of the sugar nucleotide. This latter point is particularly well illustrated in glucan synthesis. In cotton boll and in mung bean, synthesis of cellulose, a B-l,4—glucan is specific for GDP- glucose whereas UDP-glucose is used for synthesis of callose, a B-l,3-glucan (48). ADP-Glucose is generally the precursor to synthesis of starch, an u-l,4-glucan (11), while TDP-glucose has been shown to be involved in sucrose metabolism in sugar beet (48). The sugar nucleotides involved in pectin and hemicellulose synthesis are formed from glucose 6-phosphate by two basic pathways. Loewus has named these the glucose oxidation pathway and the myo- inositol oxidation pathway (49). In the glucose oxidation pathway, glucose 6—phosphate is converted first to glucose-l-phosphate and then to UDP-glucose which can be converted to UDP-galactose by an epimerase. The most important enzyme of the pathway is UDP-glucose dehydrogenase which, in the presence of NAD+, converts UDP-glucose to UDP-glucuronic acid (50). An epimerase converts UDP-glucuronic acid to UDP-galacturonic acid for pectin synthesis. Oxidative decarboxylation by a carboxy-lyase converts UDP-glucuronic acid to UDP-xylose which is then epimerized to UDP-arabinose (11,51). Loewus has described an alternate pathway from glucose 6- phosphate to UDP-glucuronic acid which bypasses the dehydrogenase (49). All four key enzymes of this pathway have been found in various plants. These enzymes are: l8 (1) a cycloaldolase for converting glucose 6-phosphate to the cyclitol lL-mgo-inositol l-phosphate, (2) an oxygenase for conversion of myo-inositol l-phosphate to glucuronic acid, (3) a kinase for phosphorylating glucuronic acid, and (4) a pyrophosphorylase for synthesizing UDP-glucuronic acid. From UDP-glucuronic acid on, the same enzymes are used as were described for the glucose oxidation pathway. The presence of the myo—inositol oxidation pathway has been reported in a number of plant species (52). The cell wall D-apiose of Lemna is supposedly formed by this route (53). Recent studies (54), however, have questioned the significance of this pathway. In Fraxinus pennsylvanica only 0.5% of the cell wall galacturonic acid was found to be derived from mgo-inositol, the remainder coming from UDP-glucose (54). It was speculated that the myo-inositol oxidation pathway is involved mainly in the degradation of myo- inositol which is a ubiquitous constituent of plant cells (55). UDP-Apiose is synthesized by concomitant cyclization and decarboxylation of UDP—glucuronic acid (56—58). The properties of UDP-apiose have been described by Kindel and Watson (58). Two carboxy-lyases (I and II) which convert UDP-glucuronic acid to UDP- xylose are present in L. minor together with the cyclase that synthesizes UDP-apiose. The purification and characterization of these enzymes together with the separation of carboxy-lyase II from the other two enzymes have recently been described (59). 19 Synthesis of Pectins Despite the importance of pectic polysaccharides in determining the properties of primary cell walls, little is known of their synthesis. In vitro synthesis of the main galacturonan chain has been studied only in mung bean (60,61) and in green tomato and turnip (62). Villimez et a1. (61) found UDPGalA to be the preferential donor for galacturonan synthesis in mung bean. Incorporation of D-galacturonic acid from all other sugar nucleotide donors including UDP-methyl-D- galacturonate was negligible. In tomato, some incorporation was found With dTDP-galacturonic acid (62). The basic characteristics of the mung bean D-galacturonsyl— transferase have been described (60,61). The enzyme is optimally active at pH 6.7 and 30°C, with an apparent Km for UDPGalA of 1.7 uM. The activity was increased 50% by the addition of 1.7 mM MnClz. One of the most serious problems in studying in vitro pectin synthesis is the instability of the particulate glycosyltransferase preparations. For example, the mung bean D-galacturonosyltransferase is spontaneously and totally inactivated in 5 min at 37°C, in 20 min at 30°C, and in 24 h at 0°C. Even at ~18°C, total inactivation occurs in slightly greater than three weeks (61). Almost no evidence is available on the incorporation of L- rhamnose residues or on the synthesis of the side chains or neutral blocks, because of the difficulty of analysis of components present in very small amounts. This is true in the study of the incorporation of L-arabinose (63). In these experiments it was impossible to . . . . . l4 . . . distinguish incorporation of L-[ C]arabinose into pectin from 20 incorporation into hemicellulose because the main chain sugars were unlabeled (55). The apiogalacturonans of Lemna minor represent one of the few systems in which synthesis of both the main chain and the side chains of the polysaccharide are being studied. Pan (64) has isolated and characterized a D—apiosyltransferase which, according to Mascaro (65, 66) incorporates D-apiose into acidic polysaccharides which appear to be apiogalacturonans of the type isolated from L. minor cell walls. Apart from these studies, the only reports of incorporation of any unit other than D-galacturonic acid into pectin have concerned the methyl esterification of polygalacturonate. Using a particulate preparation from mung bean, Kauss showed incorporation of the methyl group from S-adenosylmethionine into pectin (55). Only endogenous pectin or a galacturonan prepared in situ could be esterified. Exogenous acceptor had no effect on esterification, apparently because it lacked access to the methyltransferase. The newly esterified pectin was also protected from the action of both exogenous and endogenous pectin methylesterase (67). Agents that disrupt lipid membranes, such as Triton X-100, sodium dodecyl- sulfate and phospholipase A, removed this protection. Thus, it appears that all the enzymes needed for polymerization and esterifi- cation are contained in the same membrane-bound particle (55). Site of Synthesis The close proximity of the cell wall and cell membrane suggests that the biosynthesis and development of each should be closely coordinated. This has been found to be true. 21 Studies in Lolium longiflorum have shown that the polysaccharides found in Golgi vesicles are similar in composition to the polyuronides (pectins) of the cell wall matrix (68,69). These findings agree with previous reports of a vesicular origin for the matrix polysaccharides (70-72). In addition, a wide range of ultrastructural, histochemical and biochemical studies point to participation of the Golgi apparatus in hemicellulose and pectin synthesis. The membranes of the Golgi vesicles fuse with and become a part of the plasma membrane while the contents of the vesicles are deposited in the cell wall as precursors to the matrix phase (69). Despite the evidence presented by Ray et a1. (73) that Golgi vesicles are able to synthesize a B-l,4-glucan, the higher degree of order of the cellulose microfibrils makes it more likely that the enzymes responsible for cellulose synthesis are present in or on the surface of the plasma membrane or even within the wall itself (3,72). Vesicle fusion with the plasma membrane is almost never observed during late primary and secondary wall formation (69), further supporting this argument. Also, cell elongation in pollen tubes can be inhibited by CaCl When this is done, the number of 2. vesicles is reduced as is their sugar content and the labeling of the matrix polysaccharides. Cellulose synthesis, however, is unaffected, strongly suggesting that it proceeds independently of the Golgi vesicles (69). Reexamining Ray's results (73), vander Woude et a1. (74) found that both Golgi fractions and plasma membrane fractions synthesized B-l,3- and B-1,4-glucans. However, which synthesis predominated depended on the concentration of UDP-glucose. At low concentrations 22 (1.5 uM), glucan synthesis in the Golgi fractions predominated, but at high UDP-glucose concentrations (1 mM), the plasma membrane frac- tions were most active in glucan synthesis. Thus, they attributed Ray's findings to the low concentrations of UDP-glucose which his group used. In vivo UDP-glucose concentrations in excess of 30 uM have been reported (75). This concentration is intermediate between the concentrations used by vander Woude (74). The significance of the Golgi glucan synthetase is unclear. One possible explanation is that it represents the cellulose synthe- sizing enzyme which has been synthesized on the endoplasmic reticulum and within the smooth membrane system for transport in an inactivated form to the plasma membrane where it is activated. Recent brief reports suggest that the Golgi glucan synthetase activity may be involved in xyloglucan synthesis (76) or that it may be a precursor to cell surface glucan synthetase (77). There is also evidence that the plasma membrane glucan synthe- tase may be a mobile enzyme complex, the movements of which are directed by cellulose microfilaments (78). Hopefully, the morphology of cellulose synthesis may be clarified in the near future. Solubilization of Particulate'Enzyme Activities One reasonable approach to the problem of the instability of particulate polysaccharide synthesizing systems is to try to solu- bilize the enzymes. This involves treating the particulate prepara- tion with some agent which disrupts the lipid membrane of the particle. In theory, the solubilized preparation will either be more stable than the particulate form or, should solubilization 23 cause loss of activity, reconstitution of the enzyme in an artificial membrane may cause reactivation and stabilization. Numerous bacterial or animal systems have been treated in this way, and the techniques involved have been reviewed thoroughly (79-81). Unfortunately, little success has been reported in plant systems. The greatest success has been achieved with the glucan synthe— tases. Two studies indicated that digitonin treatment of particulate mung bean preparations "solubilized" both B-l,3- and 8-1,4-synthesizing activities (47,82). These preparations could be precipitated by centrifugation at 30,0009. Such preparations have been shown to contain particulate matter. A truly "soluble" activity remains in the supernatant after centrifugation at 100,000g (83). Nonetheless, these digitonin-treated preparations were more stable than their particulate counterparts (82). Sonication and treatment with trypsin and lipase or phospholipase did not cause solubilization. Subsequently, Tsai and Hassid (84) showed that following treat- ment of the particulate glucan synthetase from Avena coleoptiles with 8% digitonin, 40 to 50% of the initial activity remained soluble after centrifugation at 100,0009. This soluble preparation contained both B-l,3- and B-l,4-glucan synthetase activities which could be separated by column chromatography on hydroxyapatite. The only report of the successful solubilization of a hetero- polysaccharide synthesizing system concerns the glucomannan of mung bean (85). Treatment of the particulate mannosyltransferase with 0.5% Triton X-100 produced an enzyme preparation which remained soluble after centrifugation at 300,0009. Unfortunately, this preparation, which also could utilize GDP-glucose, was still 24 relatively unstable, losing two-thirds of its activity in 5 h at 0°C. Thus, solubilization does not automatically guarantee stabilization. This last study is an important advance in attacking one of the biggest problems of studying polysaccharide synthesis in vitro. Particulate enzyme preparations contain endogenous polysaccharides which are presumably used as acceptors by the transferase under study. In order to relate in vitro observations of polysaccharide synthesis to in vivo mechanisms, systems must be devised in which the nature of the acceptor can be completely defined. Solubilization of the particulate glycosyltransferases represents a critical step in this direction. Lipid Intermediates Related to the acceptor problem is the possibility that other intermediates are synthesized that function between the sugar nucleo- tides and the completed glycan chain. A critical advance in under- standing the synthesis of bacterial cell wall glycans was the discovery of sugars bound to polyisoprenoids. These polyisoprenoid lipids are likely intermediates in polysaccharide and glycoprotein synthesis in plant and animal systems as well. These studies have been reviewed recently (86) and only the highlights of the plant studies will be discussed. Preliminary studies on a cell-free system from mung bean showed that a lipid-like material served as a precursor to mannan synthesis from GDP-mannose (87). The intermediate had the properties of a complex glycolipid. Significantly, it could not be isolated by common lipid solvents until the particulate system was treated with 25 detergent which also destroyed mannosyltransferase activity. Thus, the intermediate seems to be held in the same compartment as the transferase. The sugar moiety of the intermediate was identified as a monosaccharide (88). An oligosaccharide derivative of a low molecular weight membrane protein was suggested as a second inter- mediate. The protein might act as a primer for polysaccharide synthesis or serve to anchor the completed polysaccharide in the cell wall (88). Unfortunately, no further reports on these studies have appeared. The reactions by which the bacterial lipid intermediates are formed from sugar nucleotides are freely reversible, indicating that the intermediates have almost the same group transfer potential as sugar nucleotides (86). Kauss (89) confirmed this reversibility for mannose transfer in the mung bean system. However, he cautioned that the kinetics of transfer are such that the labeled sugar could be transferred to polysaccharide via back-donation from glycosyl- lipid to nucleotide. This argument is supported by further studies which failed to show transfer of label from partially purified 14C- mannosyl-lipid to polysaccharide (90). These same studies confirmed the exchange between endogenous 14C-mannosyl-lipid and GDP and showed that in the absence of GDP or GDP-mannose, mannose is hydrolyzed from the lipid rather than transferred to polysaccharide (90). Clark and Villemez (91) have shown that phytanol, a saturated commercial isoprenol, can substitute for endogenous mannosyl acceptor in mung bean. The authors hope that the availability of this commer— cial substrate will speed the solubilization and purification of the transferases involved. 26 The only other plant system in which glycosyl-polyisoprenol synthesis has been described is cotton fibers (92). In this system both D-mannose from GDP-mannose and D-glucose from GDP-glucose are incorporated into acidic lipids of the type described in bacteria. Extraction of cotton bolls with chloroform-methanol has demonstrated the presence of significant amounts of L-arabinose and D-galactose in the lipid phase (92). By analogy with the bacterial systems, the structure and transfer potential of the lipids described in these studies suggests that they may participate in the synthesis of at least the terminal branches of polysaccharides or in glycoprotein synthesis. Until transfer of the sugar moieties from these glycosyl—lipids to such macromolecular acceptors has been conclusively documented, however, the lipids cannot be regarded as obligatory intermediates in in vivo systems in plants. EXPERIMENTAL PROCEDURES Materials D—Glucuronic acid, D-galactose l-phosphate, D-galacturonic acid l-phosphate, UDPGalA, inorganic pyrophosphatase (pyrophosphate phosphohydrolase, E.C. 3.6.1.1) and Sepharose 28, 4B, and 6B were purchased from Sigma Chemical Co. UTP was supplied by P—L Biochemicals, Inc. Platinum oxide (Adams' catalyst) was obtained from MC/B Manufacturing Chemists. Bio-Gel P-30 (100-200 mesh), Bio-Gel P-lOO (100-200 mesh), Bio-Gel P-200 (100-200 mesh) and Bio-Gel P-300 (50- 100 mesh) were purchased from Bio-Rad Laboratories. Ultrogel AcA-22 and Ultrogel AcA-34 were supplied by LKB Instruments, Inc. p-Hydroxy- diphenyl was purchased from ICN/K&K Laboratories, Inc. Triton X-100 for solubilization of membranes was supplied by Research Products International Corp., Elk Grove Village, Illinois. a-D—[U-14CJGalactose 1-phosphate (214 mCi/mmol) and UDP- [U-14C]glucuronic acid (190 mCi/mmol) were purchased from New England Nuclear Corp. UDP-[U-14C]apiose (2.8 mCi/mmol) was enzymatically synthesized from UDP-[U-14Cngucuronic acid (58, and Pan and Kindel, unpublished experiments). Butylated hydroxytoluene, Tween-20, Emulgen-911, Emulgen—9l3, sodium cholate, and sodium deoxycholate were gifts from Dr. S. D. Aust. Phenylmethylsulfonyl fluoride was a gift from Dr. W. C. Deal. Nupercaine hydrochloride was generously provided by Dr. L. L. Bieber. 27 28 NP-40 and Sarkosyl-97 were gifts from Drs. J. A. Boezi and L. F. Lee. General Methods Paper chromatograms were scanned for radioactivity with a Packard radiochromatogram scanner, Model 7201 (Packard Instrument Co., Inc.). All other radioactivity measurements were made with a Packard Tri-Carb liquid scintillation spectrometer, Model 3310. The following solutions were used for liquid scintillation counting: (A) Bray's solution (93), (B) 5.5 g of 2,5-diphenyloxazole and 0.1 g of l,4-bis[2—(4- methyl-SSphenyloxazolyl)]-benzene in 667 m1 of reagent grade toluene and 333 ml of Triton X-100, (C) 4.0 g of 2,5-bis-2-(S-t-butylbenzoxazolyl)-thiophene in 1.0 l of reagent grade toluene. Aqueous radioactive samples (1.0—1.3 ml) were counted in 10 m1 of solutions A or B. Radioactive samples on paper were counted by complete immersion of the paper in solution C. The counting effi- ciencies of solutions A, B and C were 77, 81, and 63%, respectively. Gel chromatography columns were prepared according to the manufacturers' recommendations. All columns were eluted at 24-27°C by downward flow. V6 and Vt for each column were determined by measuring the elution volumes of solutions of blue dextran (V0) and phenol red (Vt)° The dimensions of all columns were 1.0 cm i.d. x 40 cm unless otherwise noted. 29 The total uronic acid content of solubilized products was determined colorimetrically using p-hydroxydiphenyl according to the method of Blumenkranz and Abou-Hansen (94). The mean, the standard deviation, and the standard error were calculated for results based on two or more observations. The Student's t test was used for comparing sample means. The levels of probability upon which decisions of significance were based are given at the appropriate place in the text. Electrophoresis and Paper Chromatography Electrophoresis of D-galactose l-phosphate, D-galacturonic acid 1-phosphate, and UDPGalA was performed using a Pherograph Original Frankfurt, Type 64 electrophoresis apparatus (distributed by Brinkmann Instruments, Inc.). Samples were applied to Schleicher and Schuell 589C paper and were electrophesed in 0.2 M anmoniumformate (pH 3.6) at 4°C at constant current. Descending paper chromatography was performed with Schleicher and Schuell 589C paper at 4°C or at 24-27°C. The solvents were: (A) 95% ethanol—1.0 M ammonium acetate (pH 3.8) (7:3, v/v), and (B) 95% ethanol-1.0 M ammonium acetate (pH 7.5) (7:3, v/v).. Non-radioactive sugar l-phosphates were detected by the pro- cedure of Bandurski and Axelrod (95). Non-radioactive sugar nucleo- tides were visualized in the dark using a Mineralight Model UVS-ll ultraviolet lamp (Amax = 2735 A.). Samples were eluted from paper chromatograms with cold water by centrifugal extraction in plastic spin thimbles (Reeve-Angel, Clifton, N.J.). 30 Preparation of UDPlU-14CIGa1A 4 . . D-[U-1 ClGalacturonic aCid l-phosphate was prepared by catalytic oxidation of D-[U-14C]galactose l—phosphate by modification of the method of Marsh (97). PtO2 (Adams' catalyst - 50 mg) was activated by bubbling H gas through an aqueous suspension of the catalyst at 2 1 atm for 2 h. Activated catalyst was resuspended in 0.1 N NaHCO3 and combined with about 4.5 uCi of D-[U-14clgalactose l-phosphate in 7% aqueous ethanol. 02 gas was bubbled through this suspension at 25°C and 1 atm for 7-8 h. The catalyst was then removed by centri- fugation and the clear supernatant was chromatographed in solvent A or electrophoresed. Yields of D—[U—14C]galacturonic acid l-phosphate averaged 50-55%. - UDP-[U-14C1GalA was prepared from D-[U-14C]galacturonic acid and UTP by the method of Feingold et a1. (97) using a nonspecific uronic acid pyrophosphorylase (uronic acid l-phosphate uridyltrans- ferase) isolated from mung beans. The standard reaction mixture contained: 2 mM UTP, 40 pH MgCl 5 ug inorganic pyrophosphatase, 2: 0.2 uCi D-[U-14clgalacturonic acid l-phosphate and 0.5 mg of crude pyrophosphorylase preparation in 0.1 M Tris-HC1 (pH 7.3) containing 50 mM 2-mercaptoethanol. Total assay volume was 100 pl. Incubations were performed at 30°C for 1 h. Samples were immediately chromatographed in solvents A or B or were electrophoresed. Yields averaged 60-70% from D-galacturonic acid 1-phosphate. In the synthesis of both D—[U-14clgalacturonic acid l-phosphate and UDP[U-14C]GalA, the radioactive products were identified by their co-migration with non-radioactive standards or with radioactive 31 compounds previously identified by their co—migration with these non-radioactive standards upon chromatography or electrophoresis. Culture of Lemna minor Lemna minor was cultured according to previously described methods (58). For these studies, plants were grown on 4 l of inorganic culture medium in uncovered white plastic pans (29 x 32 cm). Fronds were harvested from 1-2 d after their growth had just covered the surface of the pan, usually 18-22 d after inoculation of the culture. Isolation of the Particulate D-Galacturonosyltransferase Preparation Lemna minor was harvested on a double layer of cheesecloth, washed thoroughly with distilled water, blotted to remove excess water, and weighed. All remaining steps were performed at 4°C. The fronds were chilled for 5 min, then ground gently with a mortar and pestle for l min. The standard grinding medium was 50 mM sodium phosphate buffer (pH 7.3) containing 1% BSA (w/v). Two milliliters (2.0 m1) of this buffer were used for each gram of wet plant material. The crude homogenate was filtered through four layers of cheesecloth, and the filtrate was centrifuged at 4809 for 10 min. The precipitate was discarded, and the supernatant was recentrifuged at 34,8009 for 10 min. The supernatant was discarded. The 34,800g precipitate was resuspended by gentle trituration in 0.1 M citric acid-50 mM sodium phosphate buffer (pH 5.7) containing 10 mM MnClZ, 0.4 M sucrose, and 1% BSA (w/v). Five-tenths milliliter 32 (0.5 m1) of buffer was used for each gram of plant material. Any modifications of this buffer or of the grinding medium will be described where appropriate in the text. This resuspended 34,8009 precipitate is termed the particulate D-galacturonosyltransferase preparation. Standard Assay for UDPGalA:Acceptor D-Galacturonosyltransferase For characterization of the D-galacturonosyltransferase, the standard reaction mixture contained 20 to 40 pmol of UDP[U-14C]GalA (10,000-20,000 dpm), 1 or 5 nmol of UDPGalA, and 50 ul of the par— ticulate D-galacturonosyltransferase preparation. The mixture was incubated for 1.0 min at 25°C. For characterization of the D-galacturonosyltransferase product, 14c] the standard reaction mixture contained 0.10 to 0.20 nmol of UDP[U- GalA (50,000—100,000 dpm), 1.0 nmol of UDPGalA and 100 pl of par- ticulate D—galacturonosyltransferase preparation. These assay mixtures were incubated at 25°C for 20 min. Variations in standard assay procedures are noted at the appro- priate places in the text. Each reaction mixture was extracted with 1 ml of 1% KCl in 75% (v/v) aqueous methanol. Following brief centrifugation, the supernatant was discarded, and each precipitate was then extracted in turn with two 1 m1 portions of absolute methanol, with one more 1 ml portion of KCl-methanol, and with two 1 m1 portions of water. All extractions were performed at 24-27°C. The material which remained insoluble after these extractions was termed the D—galacturonosyltransferase product. The radioactivity 33 in this material was assayed in scintillation solution C as a measure of D-galacturonosyltransferase activity. Solubilization of the D-Galacturonosyltransferase Product with Ammonium Oxalate and with Sodium Hexametaphosphate In product characterization studies, the D-galacturonosyltrans- ferase product was extracted at 50°C with freshly prepared solutions of sodium hexametaphosphate or ammonium oxalate. Each sample of D-galacturonosyltransferase product was extracted in turn with one 0.3 ml portion and three 0.1 m1 portions of the appropriate extractant. Each extraction was performed for 15 min, after which the samples were centrifuged and the supernatants were removed. The supernatants from the four extractions were pooled and portions were withdrawn and assayed for radioactivity in scintillation solution B. The residue remaining after the four extractions was resuspended by boiling for 10 min in 2% NaOH. This material was then streaked on Whatman 3MM paper and counted in scintillation solution C. The material solubilized by these extractions was named for the particular concentration of extractant used. Thus, the material solubilized by extraction with 1% ammonium oxalate was termed the 1% ammonium oxalate—solubilized product. RESULTS Partial Characterization of UDPGalA:Acceptor D-Galacturonosyltransferase Linearity of the D—Galacturonosyl— transferase Assay The amount of D-[U-l4clgalacturonic acid incorporated into water- and methanol-insoluble product is a linear function of the amount of particulate D-galacturonosyltransferase preparation used between 0 and 100 pl (Figure 3). At 25°C, the amount of D-[U—14C]galacturonic acid incorporation is a linear function of time for about 1.5 to 2 min, as shown in Figure 4. Incorporation is almost complete by 10 min. At 0°C, the rate of incorporation is much lower, but it remains linear for at least 15 min. Addition of more UDPGalA at 1.5 min failed to increase the rate of incorporation, suggesting that the decreased rate of D-[U-14CJ- galacturonic acid incorporation after 1.5 min is not due to depletion of substrate. Effect of pH on D-Galacturonosyl- ' transferase Activity As shown in Figure 5, the D-galacturonosyltransferase is optimally active between pH 6.0 and 6.2. Activities were approximately equal at the same pH in citric acid-sodium phosphate and Tris-maleate 34 35 Figure 3. The effect of enzyme concentration on the rate of incorporation of D-[U—14C]galacturonic acid into D-galacturonosyl— transferase product. Assays were performed as described in the Experimental Procedures, but the amount of D-galacturonosyltrans— ferase preparation was varied as indicated. Assays were performed for 5 min. Figure 4. The effect of time and temperature on the incor- poration of D-[U-14C]galacturonic acid into D-galacturonosyl- transferase product. Assays were performed as described in the Experimental Procedures, but the length of the incubation period and the temperature of the incubation were varied as indicated. Incubations were performed at 25°C (0) or at 0°C (0). The concentration of UDPGalA in the assay was 10 uM. 36 N O-.__‘. 0"“) it 2 a p OSIVUOJUODNI VIVO IOWd t at l l 100 l. o. «2 '- O OBIVUOJUODNI V'IVO IOWN TIME (MIN) pl TRANSFERASE PREPARATION Figure 3 Figure 4 37 Figure 5. The effect of pH on the rate of incorporation of D-[U-14C]galacturonic acid into D-galacturonosyltransferase product. The values depicted represent data from four separate experiments. In each experiment the rate of D-[U-14C]galacturonic acid incorporation at the pH measured which was closest to 6.0 was set equal to 100. All other rates measured within that experiment were reported relative to this rate. Rates of incor- poration were measured in 0.1 M citric acid-50 mM sodium phosphate buffer (0), in 50 mM Tris-maleate buffer (0), or in 50 mM Tris— HCl buffer (A). All assays were performed as described in the Experimental Procedures, but the composition and pH of the resuspension buffer were varied. The pH values reported in Figure 5 represent the final pH of the reaction mixture. The concentration of UDPGalA in the assay was 10-12 uM. 38 N Ail/\IJOV BAIIV'IBH Figure 5 8.0 7.0 6.0 5.0 4.0 i o 39 buffers; however, at pH values greater than 6.5, the D-galacturono- syltransferase activity in Tris-HCl buffer was slightly higher than in citric-acid sodium phosphate buffer at the same pH values. The high buffering capacity of the transferase preparation required use of a resuspension buffer of pH 5.7 to achieve the optimal pH in the assay. Effect of Various Ions on D-Galacturono- syltransferase Activity Of all the cations tested, only Mn+2 caused significant stimu- lation of D-galacturonosyltransferase activity (Table l). The apparent slight stimulations caused by Co+2, Zn+2, and Fe+2 are not significant at the P;0.05 level. The inhibitions caused by NiCl2 and by all the salts listed below it in Table l are significant. Mg+2 had no significant effect. The chlorides seem to be the most effective anions. Although all the results presented in Table 1 were obtained at salt concentrations of 5 mM, the same trends were observed at 10 mM. MnCl caused maximum stimulation of the transferase at a concen- 2 tration of 10 mM whether the MnCl2 was added to the resuspension buffer or to the assay (Figure 6). The sharp decrease in activity at MnCl2 concentrations greater than 10 mM may be a general ionic strength effect since similar decreases were caused by MgCl2 and NaCl. Although only MnCl caused significant stimulation, the presence 2 0f 2.5 mM EDTA in the reaction mixture decreased D-galacturonosyl- transferase activity by about 40%. Addition of equal concentrations 0f MnCl2 or MgCl2 with the EDTA did not reverse EDTA inhibition. 40 Table 1. Effect of various ions on the incorporation of D-[U-14C]- galacturonic acid into D-galacturonosyltransferase product nmol D-galacturonic acid Addition incorporated per min :_SD None 0.161 :_0.036 MnC12 0.186 :_0.037 MnSO4 0.164 :_0.025 MgCl2 0.155 :_0.027 MgSO4 0.159 :_0.041 Mg(C2 0.165 :_0.055 CoCl2 0.178 :_0.028 FeCl2 0.173 :_0.022 ZnCl2 0.171 :_0.058 NiCl2 0.136 :_0.018 NH4C1 0.133 :_0.022 NaCl 0.130 :_0.013 CaCl2 0.114 :_0.018 CuCl2 0.100 : 0.012 HgCl2 0.016 : 0.006 The D-galacturonosyltransferase preparation was isolated as described in the Experimental Procedures except that the resuspension buffer contained no metal ions. The standard D-galacturonosyltransferase assay was modified to contain the ion being tested at a final concen- tration of 5 mM. The concentration of UDPGalA was 50 uM in all assays. 41 The effects of adding higher metal ion concentrations were not studied. Effect of Sucrose on D—Galacturono- syltransferase Activity The addition of sucrose to the resuspension buffer caused a marked increase in D-galacturonosyltransferase activity. As shown in Figure 7, optimal stimulation occurred at 0.4 to 0.5 M sucrose. 14 . . However, sucrose decreased D-[U- C]galacturonic acid incorporation when it was added to the grinding buffer. Determination of the Apparent Km for UDPGalA The effect of UDPGalA concentration on D-galacturonosyltransferase activity is shown in Figure 8, which is a typical Lineweaver-Burk plot of the results. From these plots, an apparent K.m for UDPGalA of 8 uM was determined. The highest rate of incorporation shown in Figure 8 is 0.108 nmol D-galacturonic acid per min, but this value varied considerably depending upon the particular D-galacturonosyl- transferase preparation. This must be accepted in a crude particulate enzyme system. In contrast, the apparent Km value varied little. Stability of D-Galacturonosyltransferase Inactivation of the D-galacturonosyltransferase is both spontaneous and temperature-dependent. The enzyme is stable for 2.0 to 2.5 min at 25°C, but less than 10% of the initial activity remains after 10 min. At 0°C, 10% of the initial transferase activity is lost in 30 min. About 60% remains after 3 h, and less than 45% after 5 h. The presence of 10 mM MnCl and 0.4 M sucrose provided only slight 2 stabilization at 0°C. 42 Figure 6. The effect of MnC12 on the rate of incorporation of D—[U-14C]galacturonic acid into D-galacturonosy1transferase product. Assays were performed as described in the Experimental Procedures. MnC12 was added to the resuspension buffer at the concentration indicated in Figure 6 (MnC12 was also added to the reaction mixture so that the concentrations of MnC12 in the resuspension buffer and in the reaction mixture were the same). The concentration of UDPGalA was 10 uM. Figure 7. The effect of sucrose on the rate of incorpora- tion of D-[U-14C]galacturonic acid into D-galacturonosyltransferase product. Assays were performed as described in the Experimental Procedures, but the concentration of sucrose in the resuspension buffer was varied as indicated. (The final concentration of sucrose in the reaction mixture was one-half that shown in Figure 7). The grinding medium and the resuspension buffer both con- tained 5 mM MnClz. The final concentration of UDPGalA was 50 uM. 43 h mummwm :5 @353”— n._ 0.. nd fl 4 q f‘o How 100— om— 31VII 'IOIIlNOD S 6 $5on :25 «6.2”— oo. mm on 3 o u u q .4 o I .. mm o l Lon T .. nu fit _ o I: l “Np _ b. _ p 31V! 'IOEIlNOD 1 44 Figure 8. The effect of UDPGalA concentration on the rate of incorporation of D—[U-14C]galacturonic acid into D-galacturono- syltransferase product. Assays were performed as described in the Experimental Procedures, except that the concentration of UDPGalA was varied as indicated. 45 123E 0V0 00.0 0N0 9.0 0 9.0. \. o. JON > .00 I_. 10¢ — _ - Figure 8 46 The stability at -20°C is variable. In three experiments, 50% of the initial transferase activity was lost in 24-36 h, whereas in two other experiments, greater than 50% of the initial activity remained after 60 h. The activity seemed to be very sensitive to the method of thawing the frozen transferase preparation. Quick- thawing at 25°C produced greater recovery of activity than slow- thawing in ice water. This may explain some of the variability. Regardless of the variability in the rate of inactivation, storage in 50% glycerol always caused a 10% to 20% increase in the recovery of transferase activity up to 60 h at -20°C. A survey of a number of reagents which stabilize other enzymes failed to find any that would significantly stabilize the D-galacturono- syltransferase at 0°C. These reagents included butylated hydroxy- toluene, dithiothreitol, 3-mercapto-l,2-propanediol, phenylmethyl- sulfonyl fluoride and Nupercaine hydrochloride. Storage of the enzyme in high ionic strength solutions of NaCl and MgCl also failed to provide any significant stabilization. 2 As shown previously, high ionic strength also decreased initial D-galacturonosyltransferase activity. Effect of UDP-Apiose on D-Galacturonosyl- transferase Activity The major known galacturonan of L. minor cell walls is a hetero— polysaccharide containing D-apiose. Therefore, the effect of UDP- apiose on D-[U-14C1galacturonic acid incorporation was studied. The incorporation of D-[U-l4clgalacturonic acid into polysaccharide with time in the absence of UDP-apiose and in the presence of 3.0 uM UDP-apiose is shown in Figure 9. The presence of UDP-apiose did not 47 Figure 9. The effect of UDP—apiose on the incorporation of D-[U-14C]galacturonic acid into D—galacturonosyltransferase product. Assays were performed as described in the Experimental Procedures, except that 50 mM Tris-maleate buffer (pH 5.7) was used for the resuspension buffer. Assays contained 10 uM UDPGalA (O) or 10 uM UDPGalA + 3 uM UDP-apiose (O). 48 I I - 9 - 9. "‘ l0 . '~~ K.3‘0, ‘~‘d‘. a s a are OBIVHOdtIODNI V'IVS) 'lOINd. Figure 9 (MIN) TIME 49 . 14 . . . . increase D-[U- CJgalacturonic aCid incorporation. In fact, UDP- apiose slightly decreased both the rate and the extent of D-[U-l4cl- galacturonic acid incorporation. These results are in marked contrast . . 1 . . . to the increase in D-[U- 4C]apiose incorporation caused by UDPGalA (64). Effects of Detergents on Activity and Solubilization As a prelude to attempting to solubilize the particulate D— galacturonosyltransferase, the effects of a number of ionic and non- ionic detergents on incorporation of D-[U-14C]galacturonic acid were studied (Table 2). At a concentration of 0.1% most detergents decreased incorporation by no more than 20% when they were added to the D— galacturonosyltransferase preparation just before assay. At this concentration sodium cholate, Emulgen-9ll and Tween-20 had no signifi- cant effect at the P§O.l level. However, all but Tween-20 caused almost complete loss of activity by the time their concentration was increased to 1.0%. Mixing Tween-20 with any other detergent caused significant inhibition at all concentrations (E;O.1). Tween-20, Triton X-100, Emulgen-9ll, and sodium cholate were tested for their ability to solubilize the particulate D-galacturono- syltransferase. The transferase was prepared by the usual procedure (see Experimental Procedures) except that the 34,800g precipitate was resuspended in buffer containing the appropriate detergent at concen- trations from 0.1 to 1.0%. The resuspended preparation was then centrifuged for 10 min and the supernatant and pellet were assayed for D-galacturonosyltransferase. In none of these studies was more than 20% solubilization achieved. The presence of detergent caused at least 50% loss of 50 Table 2. Effect of various detergents on D-galacturonosyltransferase activity Amount of original D- Concentration in galacturonosyltransferase Addition % (v/v) activity remaining (%) :_SD None -- 100 i_ 6 Triton X-100 0.01 103 :_16 0.10 91 :_15 1.0 24 :_14 NP-40 0.01 74 :_ 3 0.10 76 :_ 6 1.0 7 i_ 0 Emulgen-9ll 0.10 102 i_ 5 1.0 18 :_ 3 Emulgen-9l3 0.10 80 :_ 0 1.0 31 :_ 1 Tween-20 0.10 97 :_ 6 ' 1.0 92 :_ 6 Sarkosyl-97 0.01 80 :_ 3 0.10 75 :_ O 1.0 12 :_ 4 Sodium cholate 0.10 108 i l 1.0 6 :_ 1 Sodium deoxycholate 0.10 84 :_ 1 1.0 11 :_ 0 The D-galacturonosyltransferase preparation was isolated as described in the Experimental Procedures except that the detergent being tested was added to the resuspension buffer at the appropriate concentra- tion. All samples were assayed within five minutes of resuspension. The standard D-galacturonosyltransferase assay contained 10 mM UDPGalA. 51 initial D-galacturonosyltransferase activity during this second 34,8009 centrifugation. Addition of various amounts of glycerol up to a final concentration of 75% had varying effects on the recovery of transferase activity. As a further complication, the presence of glycerol made separation of the supernatant and pellet fractions more difficult after the second 34,8009 centrifugation, so that determination of "solubilized" activity was much more uncertain. Attempts to restore activity by adding boiled transferase preparation to the second 34,8009 supernatant or by recombining supernatant and pellet fractions were also unsuccessful. Apparently, any treatment that is capable of severely disrupting the membrane system which encloses the transferase will cause significant loss of activity. Further attempts at solubilization will require procedures more elaborate than those described here. Characterization of the Product of the D-Galacturonosyltransferase Reaction Efficiency of Solubilization The efficiency of solubilization of the D-galacturonosyltrans- ferase product by ammonium oxalate and by sodium hexametaphosphate was determined at 50°C (Table 3). Sodium hexametaphosphate is the more effective solubilizing agent (extractant). The amount of product solubilized remains unchanged at hexametaphosphate concentrations as low as 0.10%. The efficiency of solubilization decreases rapidly at ammonium oxalate concentrations below 0.5%. Fifty percent extraction requires at least a five-fold higher concentration of ammonium oxalate than of sodium hexametaphosphate. 52 Table 3. Efficiency of solubilization of D-galacturonosyltransferase product with ammonium oxalate and sodium hexametaphosphate Amount of product solubilized by: Concentration of Ammonium Sodium hexa- extractant oxalate metaphosphate (%) (w/v) (%) (%) 2.0 ND3 78 1.0 86 83 0.50 84 80 0.25 55 78 0.10 33 82 0.050 ND 68 0.025 ND 23 0.010 ND 11 aND: value not determined. D-Galacturonosyltransferase product was prepared and was extracted with ammonium oxalate or with sodium hexametaphosphate as described in the Experimental Procedures, except that the concentration of extractant was varied as indicated. 53 Size of the Solubilized Product The sizes of the D-galacturonosyltransferase products solubilized with the two extractants were compared by chromatography on Bio-Gel P-300 with 50 mM sodium phosphate buffer (Figure 10). Both the 1% ammonium oxalate-solubilized product and the 2% sodium hexametaphosphate- solubilized product eluted throughout the fractionation range of Bio-Gel P-300. This gel reportedly separates dextrans with weight- average molecular weights between 5,000 and 125,000 (99). About 40% of the 2% sodium hexametaphosphate-solubilized product eluted from the P-300 column in V0 (fractions 8-11) versus less than 10% of the 1% ammonium oxalate-solubilized product. Thus, the 2% sodium hexametaphosphate-solubilized product apparently contained more material of apparent high molecular weight than did the 1% ammonium oxalate-solubilized product. This was surprising since these two extractants solubilize approximately equal amounts of D—galacturono- syltransferase product. The size of the 2% sodium hexametaphosphate—solubi1ized product was studied further by chromatography on agarose gels with higher exclusion limits. A molecular weight of l x 106 is the upper exclusion limit of Sepharose 6B for polysaccharides (99). When the 2% sodium hexametaphosphate-solubi1ized product was chromatographed on this gel with 50 mM sodium phosphate buffer, about 30% of the radioactive product eluted in V0 (fractions 10-13) (Figure 11A). This product is larger than any previously reported galacturonan. To study the size of this product further, material eluting from Bio-Gel P-300 in V0 was rechromatographed on Sepharose 4B (Figure 12B) with 50 mM sodium phosphate buffer. The 54 Figure 10. Chromatography of 1% ammonium oxalate—solubilized product and 2% sodium hexametaphosphate-solubi1ized product on Bio-Gel P-300. D-Galacturonosyltransferase product was prepared and was extracted with 1% ammonium oxalate (O) or 2% sodium hexametaphosphate (O) as described in the Experimental Procedures. The two products were chromatographed on a Bio-Gel P-300 column which was equili- brated and eluted with 50 mM sodium phosphate buffer (pH 6.8) at a rate of 3-4 ml per h. The two products contained 25,000 and 19,500 dpm, respectively. Fractions (1.0 ml) were collected and were assayed for radioactivity in scintillation solution B. Values of radioactivity (%) represent the percent of total eluted radio- activity which was collected in each fraction. Total recoveries of the two products from the columns were 85% and 90%, respectively. 55 mum—232 ZO_._.o._._>_._.U._._ >_ko._._>_._.U°-> 0. ON on F' (%) Ail/\ILOVOIOV‘d Figure 21 84 sodium phosphate buffer. The results are summarized in Table 4 and in Figure 22. The size of the product did not decrease significantly until the total amount solubilized began to decrease. This could mean either that concentrations of sodium hexametaphosphate below 0.05% failed to extract larger molecular weight products, or that the product solubilized simply appeared to be of lower molecular weight because the lower sodium hexametaphosphate concentration caused less aggregation. In contrast, preliminarystudies indicated that lowering the concentration of ammonium oxalate used for extraction to 0.1% did not affect the Ultrogel AcA 34 elution profile substantially (Figure 23). Both ammonium oxalate and sodium hexametaphosphate solubilized a species of polysaccharide which could not be aggregated by NaCl. This material eluted as a symmetrical peak near Vt on Sepharose 48 with 1.0 M NaCl. Depending on the experiment, it contained up to one-third of the total radioactive polysaccharide eluted from the column. When 2% sodium hexametaphosphate-solubilized product was chromatographed on Sepharose 4B with 50 mM sodium phosphate buffer and the material eluting near Vt was rechromatographed on Sepharose 48 with 1.0 M NaCl, no significant aggregation occurred (Figure 24). Rechromatography in 1.0 M NaCl of a fraction of NaCl-aggregated, 1% ammonium oxalate-solubilized product which eluted near Vt also failed to cause aggregation. Product which was solubilized with 0.025% sodium hexametaphosphate and which represented only 20-25% of the total incorporated 85 Table 4. Effect of sodium hexametaphosphate concentration on the size of the solubilized product Concentration of sodium Amount Amount eluted hexametaphosphate solubilized in V0 3313- n0- (%) (W/v) (%) (%) A 2.0 77 34 0.50 79 28 0.10 81 27 B 2.0 75 40 0.050 68 35 0.025 23 14 0.010 11 9 D-Galacturonosyltransferase product was prepared and was solubilized with sodium hexametaphosphate as described in the Experimental Pro- cedures except that the concentration of sodium hexametaphosphate was varied. Fifty microliter (50 ul) portions of solubilized products were assayed for radioactivity in scintillation solution B for determining the percent of the total D-galacturonosyltransferase product solubilized. The remaining solubilized product was chromato- graphed on an Ultrogel AcA 34 column that was equilibrated and eluted with 50 mM sodium phosphate buffer (pH 6.8). Fractions (1.0 ml) were collected and were assayed for radioacti-ity in scintillation solu- tion 8. Values of "Amount eluted in V0" represent the percent of total eluted radioactivity collected in the V0 fractions. 86 Figure 22. Effect of sodium hexametaphosphate concentration on the size of the sodium hexametaphosphate—solubilized product. D-Galacturonosyltransferase product was prepared and was extracted with sodium hexametaphosphate as described in the Experimental Procedures except that the concentration of sodium hexametaphosphate was varied. Products solubilized with 0.01% (O), 0.025% (O), 0.025% (A), or 2.0% (I) sodium hexametaphosphate contained 3,000, 4,500, 15,000 and 16,500 dpm, respectively. Each sample was chromatographed on an Ultrogel AcA 34 column which was equilibrated and eluted with 50 mM sodium phosphate buffer (pH 6.8) at a rate of 5 ml per h. Fractions (1.0 ml) were collected and were assayed for radioactivity in scintillation solution 8. Values of radio- activity (%) represent the percent of total eluted radio- activity collected in each fraction. Total recoveries for the four products were 58%, 90%, 91%, and 100%, respectively. 87 0' a? ,."" l/ 0 3°" 44 \%\\.\L.: ‘0~.‘~ «\Q ~o 45.\ ‘4‘: 2 e l \. _ O r m9 3 N f. ,3; \. I / IO IO- 5 O (%) Ail/\IiOVOICIV/‘d Figure 22 FRACTION NUMBER 88 Figure 23. Effect of ammonium oxalate concentration on the size of the ammonium oxalate-solubilized product. D-Galacturonosyltransferase product was prepared and was extracted with ammonium oxalate as described in the Experi- mental Procedures, except that the concentration of ammonium oxalate was varied. Products solubilized with 0.1% (O) or with 1.0% (O) ammonium oxalate contained 10,000 dpm and 22,000 dpm, respectively. Both products were chromatographed on an Ultrogel AcA 34 column which was equilibrated and eluted with 50 mM sodium phosphate buffer (pH 6.8) at a rate of 5 ml per h. Fractions (1.0 ml) were collected and were assayed for radioactivity in scintillation solution B. Values of radio- activity (%) represent the percent of total eluted radioactivity collected in each fraction. Total recoveries for the two samples were 98 and 100%, respectively. 89 mum—232 ZO_._.0<«_.._ O_ (%) Ail/\IlOVOICthI Figure 23 90 Figure 24. Effect of 1.0 M NaCl on 2% sodium hexameta- phosphate-solubilized product that elutes near Vt from Sepharose 4B. D-Galacturonosyltransferase product was prepared and was extracted with 2% sodium hexametaphosphate as described in the Experimental Procedures. Solubilized product, containing 23,000 dpm, was chromatographed on a Sepharose 48 column that was equilibrated and eluted with 50 mM sodium phosphate buffer (pH 6.8) (A) at a rate of 5 ml per h. Fractions (1.0 ml) were collected, and a 100 ul portion of each fraction was assayed for radioactivity in scintillation solution B. Material eluting in Fraction 25, which contained 6% of the total eluted radioactivity, was rechromatographed on a Sepharose 48 column that was equilibrated and eluted with 1.0 M NaCl (B) at a rate of 5 ml per h. Fractions (1.0 ml) were collected and were assayed for radioactivity in scintillation solution B. Values of radioactivity (%) represent the percent of total eluted radioactivity collected in each fraction. Total recoveries from columns A and B were 95 and 98%, respectively. 91 IV. 0 VII. I. ”V O‘\\\\O\\\\ ‘1 MW mm nU ./ . o [0’ Q]: IIIOnlllIIIIO .0 / O O u v .. O 0 .. 2 2 ‘0‘ \\\Q\\\0 A 0 Il'.\ . O VOI' .H. O 1 MI - P V F P n n p v m 5 O m. B m 5 0 loss >t>fio