Received for publication May 5, 1989 and in revised form September 18, 1989

Plant Physiol. (1990) 92, 327-333

0032-0889/90/92/0327/07/$01 .00/0

A Chenopod Extensin Lacks Repetitive Tetrahydroxyproline

Blocks'

Xiong-biao Li2, Marcia Kieliszewski, and Derek T. A. Lamport* D.O.E. Plant Research Laboratory and Department of Biochemistry, Michigan State University, East Lansing, Michigan 48824 Because of our interest in the phylogeny of extensin, we have begun to explore extensins of the monocots (13, 15) and also primitive dicots represented by families such as the Chenopodiaceae. Here, we report the isolation of an extensin from sugar beet cell suspension cultures. This extensin is closely homologous with other members of the extensin family, yet contains no explicit Ser-Hyp-Hyp-Hyp-Hyp pentameric repeats. However, quite remarkably, the pentamer occurs as a cryptic, interrupted or split form: Ser-Hyp-Hyp[X]-Hyp-Hyp where [X] is an insertion sequence, typically: VAL-His-Glu-Tyr-PRO or VAL-His-Lys-Tyr-PRO, resembling the insertion sequence [Y] VAL-Lys-Pro-Tyr-His-PRO of tomato extensin P1. Thus, while the two extensins share a major repeating unit, in tomato it occurs as: Ser-Hyp-HypHyp-Hyp-[Y]-Thr-Hyp-Val-Tyr-Lys, but in sugar beet as: Ser-

ABSTRACT An extensin isolated from sugar beet (Beta vulgaris) cell suspension cultures fulfills all criteria for membership of the extensin family save one, notably, lack of the 'diagnostic' pentamer SerHyp-Hyp-Hyp-Hyp. However, sequence analysis of the major tryptic peptides shows that sugar beet extensin shares a motif in common with tomato extensin P1 but differs by the position of an insertion sequence [X] or [Y] which, in sugar beet, splits the tetrahydroxyproline block: Ser-Hyp-Hyp-[X]-Hyp-Hyp-Thr-HypVal-Tyr-Lys, where [X] is [Val-His-Glu/Lys-Tyr-Pro], while in tomato the insertion sequence [Y] = [Val-Lys-Pro-Tyr-His-Pro] and, when it occurs, immediately follows the tetrahydroxyproline block: Ser-Hyp-Hyp-Hyp-Hyp-[Y]-Thr-Hyp-Val-Tyr-Lys. Based on these data we reinterpret three highly repetitive cDNA sequences, including nodulin N75 from soybean and wound-induced P33 of carrot, as extensins with split tetra(hydroxy)proline blocks.

Hyp-Hyp-[X]-Hyp-Hyp-Thr-Hyp-Val-Tyr-Lys. MATERIALS AND METHODS

Suspension Cultures Beta vulgaris, line SP-6926-O, was a gift from Dr. Joseph Saunders, from which we initiated leaf disc callus and thence cell suspension cultures on Murashige and Skoog medium (19) (minus 2,4-D) at 28°C. Inoculations were typically 10% PCV in 1 L flasks (600 mL medium) on a gyrotary shaker.

The primary cell wall of higher plants contains extensin, a small class of basic HRGPs3 generally rich in serine, lysine, tyrosine, and hydroxyproline with most ofthe hydroxyproline residues O-glycosylated by short arabinooligosaccharides (18). Although extensin is mostly insoluble, the growing wall contains a small pool of soluble extensin monomers which are precursors to the insoluble network (22). Purification and peptide mapping (23) of these monomers from tomato cell suspension cultures demonstrated multiple extensins designated P1 and P2, and we deduced the existence of a third, P3. Each is a highly periodic structure. Peptide sequences of these and other extensins from carrot (3, 4) and bean (5, 21) display a highly repeated pentameric motif: Ser-Hyp-HypHyp-Hyp (Ser-Pro-Pro-Pro-Pro in cDNA sequences), which is hence, currently considered a diagnostic sine qua non of extensin (2, 8, 11, 14, 21). However, this 'rule' is based only on dicotyledonous species, and then only in three not entirely disparate families. 'This work

was

Isolation of Crude Monomer

Preparation of the crude monomer from 10 flasks of 12 d old cells (about 54% PCV) involved rapid filtration of 6 L culture through a large coarsely sintered funnel, followed by a brief water wash (1.5 L), and then elution of the cell pad with 1.5 to 2 L 30 mm AIC13, for 5 min and rotary evaporation of the eluate to about 100 mL at 30C. Addition of solid TCA to a final concentration of 10% (w/v) in the concentrated eluate (overnight at 4°C) typically yielded a precipitate and, after centrifugation (1 h at l0,OOOg), dialysis (72 h at 4°C), and freeze-drying, the supemate yielded a hydroxyprolinerich (2-4%) crude monomer fraction 80-120 mg/80 g cells (dwt or 1140 g wet weight).

supported by the U. S. Department of Energy,

contract No. DE-AC02-76ERO- 1338; the U. S. Department of Agri-

culture, grant No. 88-37261-3682; and the National Science Foundation, grant No. DCB-8801713. 2 Present address: Department of Biology, Peking University, The People's Republic of China. 3Abbreviations: HRGP, hydroxyproline-rich glycoprotein; HFBA, heptafluorobutyric acid; MeCN, acetonitrile; SA, sulfoethyl aspartamide; PCV, packed cell volume; P1 and P2, glycosylated tomato extensin type 1 and 2; dw, dry weight.

Purification of Extensin Monomer We separated 70 to 200 mg of crude monomer on a BioRex 70 (100-200 mesh) column (90 x 1.5 cm) eluted (40 mL/h) with a pH and salt gradient: 250 mL of 30 mM (pH 7.6) sodium phosphate buffer and 250 mL of 30 mm (pH 6. 1) 327

328

LI ET AL.

sodium phosphate containing 0.5 M sodium chloride. The Bio-Rex-70 F2 fraction (Fig. 2) was purified further via Superose-6 fast protein liquid chromatography (Pharmacia, 300 x 10 mm i.d. 6% cross-linked agarose beads) eluted at 530 ,uL/min with 0.2 M, pH 7 sodium phosphate buffer and monitored at 220 nm. Data capture was by IBM 9001 computer with IBM CAPS software. Monomer purity after HFdeglycosylation was checked by SDS gel electrophoresis (16). Cell Wall Preparation

Cell wall preparations involved sonication of 80 g cells in 150 mL water by a Braunsonic model 1510 sonicator for 12 min at 300 W, followed by filtration on a coarsely sintered funnel and washing with 1 M NaCl (2 L) followed by 10 L distilled water and freeze drying. Transmission Electron Microscopy

Rotary-shadowed extensin precursors were prepared and examined in a JEOL 100 CX transmission electron microscope as described previously (9).

Hydroxyproline Arabinoside Profiles Hydroxyproline arabinoside profiles involved sample hydrolysis in 0.2 M Ba(OH)2 and separation of the hydroxyproline arabinosides via Technicon Chromobeads B column (75 x 0.6 cm) as described before (18). HF-Deglycosylation We deglycosylated monomeric P1 (5-30 mg) in a microapparatus (20) containing 1 to 2 mL anhydrous HF and 40 to 200 ,L anhydrous methanol for 1 h at 0°C, and quenched the reaction by slowly pouring into 20 mL stirred ice-cold water, followed by dialysis and freeze drying. We recovered 60% of the initial HRGP as the deglycosylated product, indicating a carbohydrate content of approximately 40%, confirmed by direct sugar analysis of sugars as their alditol acetates via gas chromatography (1). Amino Acid Analysis

Amino acid analyses employed a Benson cation exchange column and o-phthalaldehyde postcolumn derivatization and fluorimetric detection as previously described (23). ELISA

We assayed the immunological relatedness of sugar beet extensin by its reaction with polyclonal antibodies raised against tomato extensin precursors P1, P2, and the HFdeglycosylated products dPi and dP2, by an ELISA method previously described ( 12). In Vitro Cross-Linking by Extensin Peroxidase

Assay for cross-linkage of sugar beet extensin monomer by a crude extensin peroxidase isolated from tomato (7) was in a 30 ,uL buffered (Mcllvaine pH 6.5) reaction system contain-

Plant Physiol. Vol. 92, 1990

ing 100 ,ug monomer, 5 ,uL crude enzyme, and 30 ItM H202 for 5 min at 23°C (7). Tryptic Peptide Mapping, Purification, and Sequencing After heating deglycosylated monomer (10 mg/mL) for 5 min in 10 mm CaCl2 we incubated it with trypsin (1:100) in a pH stat at pH 8 for 7.5 h. We generated tryptic peptide maps (from three different digests) via reverse phase HPLC on PRP-1 (150 x 4.1 mm i.d. Hamilton Co.) using mobile phase solvents: A (0.13% HFBA) and B (0.13% HFBA in 80% aqueous MeCN) at 0.5 mL/min programmed to increase B from 0 to 25% in 50 min and then held for an additional 30 min. We used two methods for further peptide purification: first, by recycling semipurified peptides obtained from poorly resolved PRP- 1 fractions; second, by further separation of PRP1 fractions on SA (sulfoethyl Aspartamide from The Nest Group) using gradient elution: buffer A = 10 mm (pH 3.0) sodium phosphate containing 10% MeCN, buffer B = A containing 1 M NaCl. We increased buffer B from 0 to 100% in 60 min at 0.5 mL/min. We desalted tryptides from SA via PRP- 1, using: buffer A = 0.1 S% TFA, buffer B = A containing 70% MeCN. Eluent B increased from 0 to 100% in 40 min and was then held 10 min (0.5 mL/min). We also separated some of the peptides from SA on a C-4 column (Baker) using the same programmed gradient elution as for the Hamilton PRP- 1. Tryptic peptides were sequenced by the MSU Macromolecular Facility using a Beckman model 890M sequencer for some peptides and an Applied Biosystems model 477A sequencer for others. RESULTS We applied the intact cell elution technique (17) at a stage of growth optimal for yield of extensin monomers (Fig. 1). Twelve d old sugar beet cell suspension cultures typically yielded 60 to 100 mg crude eluate/kg cells (wet weight). BioRex 70 fractionation of crude cell eluate (2-4% hydroxyproline by weight) gave a void and two protein fractions: Fl and a hydroxyproline-rich F2 (Fig. 2). Superose-6 gel filtration (Fig. 3, a and b) resolved F2 into two peaks: a hydroxyprolinerich, high mol wt P1 (2 x void) and a lower mol wt P2 (2.5 x void) with less hydroxyproline (Table I). The highest yield of P1 (Fig. 1) occurred at d 12 (54% PCV) corresponding to 2 mg P1/kg (wet weight) cells or 28.6 mg/kg (d wt) cells. Maximum culture growth yield was 64% PCV at d 16. Isolated cell walls contained 0.04% w/w bound hydroxyproline; after treatment with HF (1 h at 0°C), 6% of the wall remained water-insoluble (0.06% w/w hydroxyproline) while 10% was water-soluble but nondialyzable (0.18% w/w hydroxyproline). Sugar beet P1 resembled monomeric P1 extensin isolated earlier from tomato cell suspension cultures: the amino acid composition of sugar beet P1 (Table I) was typically extensinlike, as was the carbohydrate content (40% w/w) and the characteristic hydroxyproline arabinoside profile (Table II) after alkaline hydrolysis. HF-deglycosylation (which gave a

329

SUGAR BEET EXTENSIN

1:

240

Ec

E

0 4._

o

-J

160 U z

3

E

.-

3:

0

0-

40

E

0)

U

.0-0) 0

0

0

E

0 C.) 0.

.00

'a ._

0

E

0 0

10 20 30 Fraction Number(10 mL/fraction)

Figure 2. Bio-Rex 70 fractionation of the crude salt eluate from sugar beet cell suspensions. Column load was 70 mg in 30 mm, pH 7.6 sodium phosphate. V, void; Fl, fraction 1; F2, fraction 2.

C

5

10

15

Days After Subculture Figure 1. Growth curve of sugar beet cell suspensions and P1 yield as a function of culture age.

40% weight loss in agreement with sugar analysis) followed by gel electrophoresis (Fig. 4) gave one band (nominally 40 kD) confirming P1 purity. Four sets of polyclonal antibodies previously raised against tomato extensins (P1 and P2, and deglycosylated dP1 and dP2) all cross-reacted strongly with sugar beet P1 (Table III). Electron microscopic visualization of P1 as flexuous rods (data not shown) and its reactivity as a substrate for tomato extensin peroxidase also implied that sugar beet P1 was monomeric extensin (Fig. 5). However, P1 peptide mapping and amino acid sequences of the peptides provided an unexpected twist: HF-deglycosylation and trypsinolysis gave a characteristically simple peptide map (Fig. 6) dominated by a few major peptides (Table IV), none of which contained the repetitive pentapeptide Ser-HypHyp-Hyp-Hyp motif characteristic of all extensins known to data. However, the major peptides sequenced did contain the seven-residue sequence: Hyp-Hyp-Thr-Hyp-Val-Tyr-Lys in common with H5 of tomato P1 extensin (Table IV). Sugar beet peptides are, therefore, closely related to the major repetitive tryptic peptides H5 and H20 of tomato extensin P1, where H20 can be viewed as H5 interrupted by a short

Minutes

Minutes

Figure 3. Superose-6 chromatography of F2 from Bio-Rex 70. A, 1.74 mg F2; B, 75 jig P1 after purification.

insertion sequence, whose position within the tetrahydroxyproline block is not excluded (see "Discussion"). DISCUSSION

Judging from its location at the cell surface and its size, shape, composition, cross-linkability, and immunoreactivity, sugar beet P1 is an extensin monomer, yet it lacks the repetitive Ser-(Hyp)4 pentapeptide motif currently considered a diagnostic sine qua non of extensin. However, definition of extensin based exclusively on Ser-(Hyp)4 ignores other criteria of sequence homology. Thus, most of sugar beet P1 occurs as the 15-mer repeat H5 and, assuming H1 and H2 are contiguous, a closely related 15-mer which differs only by the substitution of Lys for Glu at residue No. 6 (Table IV). Yet,

330

LI ET AL.

Table I. Amino Acid Compositions of F1, F2, P1, P2, dP1, and Cell Wall from Sugar Beet (mol %/6) P2 Cell Wall F2 P1 dP1 Amino Acid F1 0.4 16.0 24.9 33.8 16.9 37.7 Hyp 4.8 0.4 12.0 6.5 3.8 0.6 Asp 5.1 6.7 6.9 Thr 5.7 6.3 6.7 12.2 10.2 7.3 Ser 9.6 16.2 8.3 7.1 4.7 Glu 3.4 6.6 2.6 12.0 5.1 3.9 6.7 6.3 7.4 6.1 Pro 1.4 9.5 8.6 4.9 8.8 0.5 Gly 8.7 5.2 0.6 7.3 2.8 1.5 Ala 1.2 0.7 0.0 0.0 0.0 0.0 Cys 6.5 8.7 10.8 8.7 9.8 7.5 Val 1.6 0.3 0.4 1.3 0.8 0.3 Met 4.8 1.1 2.3 0.5 2.9 0.5 lle 9.7 3.1 1.9 0.4 1.9 0.6 Leu 9.4 1.3 5.8 8.8 3.1 6.5 Tyr 3.8 1.4 0.3 2.0 0.9 0.0 Phe 7.1 7.4 9.5 7.0 9.2 9.6 Lys 2.2 1.8 5.9 2.5 4.5 5.6 His 3.3 1.3 0.7 3.0 1.4 0.1 Arg

Table II. Hydroxyproline Arabinoside Profiles of Sugar Beet Extensin P1 and Cell Wall P1 Cell Wall Hyp-Ara,,8 32.4 27.4 Hyp 31.2 13.7 Hyp-Ara 18.4 16.7 Hyp-Ara2 34.9 15.7 Hyp-Ara3 5.6 4.0 Hyp-Ara4 a Expressed as % of total Hyp.

despite their lack of Ser-(Hyp)4, these repetitive 1 5-mers are homologous with the repetitive 16-mer tryptic peptide H20 from tomato P1 extensin, judged by two criteria: 1. Both the sugar beet 15-mers and tomato H20 share an identical seven residue C-terminal region: Hyp-Hyp-Thr-HypVal-Tyr-Lys. 2. We can relate the sugar beet 15-mers to the simpler repetitive peptide decamer: Ser-Hyp-Hyp-Hyp-Hyp-ThrHyp-Val-Tyr-Lys, which occurs as tomato H5 and which also occurs in carrot and petunia cDNA sequences (21) and tobacco peptide sequences (M Kieliszewski, D Lamport, unpublished data). Tomato H20 differs from tomato H5 only by an insertion sequence [Val-Lys-Pro-Tyr-His-Pro] immediately following the tetrahydroxyproline block. (In the following discussion square brackets denote insertion sequences.) Although the sugar beet extensin 15-mer contains a similar insertion sequence [Val-His-Glu-Tyr-Pro], its insertion splits the rigid tetrahydroxyproline block. Facile recognition of these insertion sequences is possible only at the peptide level because their proline residues remain unhydroxylated, a distinction not apparent at the cDNA or genomic level where proline and hydroxyproline share the same four codons. We conclude that extensin structure is not quite so monolithic as previously supposed and that the tetrahydroxyproline theme also occurs as a variation split by an insertion sequence.

Plant Physiol. Vol. 92, 1990

A

B

C

200,900097,400 68,000

43900 dP1-7

25,700--

Figure 4. SDS-PAGE of HF-deglycosylated sugar beet P1. Lane A, protein molecular mass standards. Lysozyme (14.3 kD), 13-Lactoglobulin (18.4 kD), a-chymotrypsinogen (25.7 kD), ovalbumin (43 kD), bovine serum albumin (68 kD), phosphorylase b (97.4 kD), and myosin (H-chain, 200 kD). Lanes B and C, 3 and 6 ,g sugar beet dP1, respectively.

Table Ill. Antibodies Raised against Tomato Extensins P1, P2, dP1 and dP2a Cross-reactivity with sugar beet extensin P1. Antibody Raised

a 0.2

ELISA

Anti Raisb against Tomatob

% Cross-Reactivity with Sugar Beet P1

P1 P2 dP1 dP2

60.0

dg/well.

b

69.2 102.0 87.6 Dilution 1/100.

A change in prolyl hydroxylase specificity leading to incom-

plete hydroxylation of the tetra(hydroxy)proline blocks would add further subtlety to the variation, without changing extensin peptide backbone homology. Thus cDNA or genomic sequences of three putative cell wall proteins, previously difficult to understand, we now interpret as extensins in which the tetra(hydroxy)proline block is split by an insertion sequence: 1. The cDNA sequence of soybean nodulin N75, putatively involved as a structural protein in nodule morphogenesis (8), contains the repetitive motif: Pro-Pro-[His-Glu-Lys-Pro]-ProPro similar to the sugarbeet 1 5-mer core: Hyp-Hyp-[Val-HisGlu-Tyr-Pro]-Hyp-Hyp, except for a valine deletion, a lysine for tyrosine substitution, and unknown extent of proline

hydroxylation.

331

SUGAR BEET EXTENSIN

E

c 0

cm

E

. .... .

I

Figure 6. Tryptic peptide map of HF-deglycosylated sugar beet P1.

0

io

20 30 Minutes

40

50

0

10

20 30 Minutes

40

50

Figure 5. Superose-6 gel filtration profiles of extensin monomers before (A, C) and after (B, D) cross-linking for 5 min at 230C by extensin peroxidase. A and B represent sugar beet extensin P1. C and D are tomato P1 a. Three size classes of cross-linked extensins appear progressively with time: a void peak (Vo) (note that Superose6 has a nominal exclusion limit of 40 MD for globular proteins), a large-oligomer fraction (I), and a small-oligomer fraction (II). Peak IlIl is monomeric extensin. Peak IV is crude enzymic protein. Peak V, VI, and VIl are buffer components and 2-mercaptoethanol of the stop reagent.

2. The soybean genomic repeat (Pro-Pro-Val-Tyr-Lys). (2, 10) is homologous with the repeating octapeptide Ser-HypHyp-Hyp-Hyp-Val-Tyr-Lys (23) of tomato extensin P2 and is a very simple extensin consisting entirely of 50% hydroxylated (2) tetra(hydroxy)proline blocks split by the simple insertion sequence [Val-Tyr-Lys], with Ser and Thr-Pro sequences deleted. 3. Similarly, we also deduce that the wound-induced protein P33 of carrot (3, 4, 25) which has numerous repeating units of Pro-Pro-[Val-x-y]-Pro-Pro, deduced from the cDNA clone

(3) is also an extensin, similar to the soybean extensin above, and may therefore be similarly hydroxylated. A possibly related repetitive motif: Pro-Pro-[Pro-Val-His-Leu]-Pro-Pro (6, 10) also occurs as the N-terminal domain of y-zein (26) suggesting its origin by gene fusion (27) of an extensin with another protein. Why should some extensins contain more contiguous hydroxyproline than others? Consider extensin as a block copolymer (12, 23) of alternating rigid and flexible microdomains. Then splitting the relatively rigid tetrahydroxyproline block inevitably increases the overall molecular flexibility. The reason for increased flexibility is an open question. The answer must involve the wall as a micro-composite whose mechanical properties, 'tailored to the tissue' (23), reflect interactions among all the components. Possibly, extensins which contain numerous blocks of tetrahydroxyproline are the exception rather than the rule in the plant kingdom, and may be more representative of advanced herbaceous dicot families than primitive dicots or advanced monocots. On the other hand, judging from the single C-terminal Ser-(Hyp)4 of maize THRGP extensin (15), tetrahydroxyproline blocks may be a primitive feature. If so, we predict their presence in the gymnosperms. Clearly, some features of extensin are more highly conserved than others. For a structural protein it has an unusually high interspecific variability that results not only from single substitutions or deletions, but from insertion or deletion of entire microdomains, typically [Val-(His ...Tyr...Lys...)Pro]. While these may often be of similar structure in different dicot species they occupy different positions relative to blocks of hydroxyproline, and this might be related to their suggested role as cross-link domains (23). Hence the evolution of the extensins has probably involved extensive mini-gene shuffling. This tends to obscure direct interspecific genomic comparison and may partly explain why a heterologous probe from a dicot (carrot genomic clone pDC5A 1) (4) has not detected monocot extensins. Considering the wide detection range of such probes one might have inferred the absence of extensins from graminaceous monocots were it not for the recent isolation of the glycoproteins per se from Zea mays (1 1, 13, 15) and corroborative cDNA sequences (24).

Li ET AL.

332

Plant Physiol. Vol. 92, 1990

Table IV. Amino Acid Sequences of Tryptic Peptides from Deglycosylated Sugar Beet Extensin P1, Compared with Peptides H5 and H20 of Tomato Extensin P1 All peptides were fractioned initially on a Hamilton PRP-1 column (Fig. 6). SB Hi (6-mer): Ser-Hyp-Hyp-Val-His-Lys (sequenced once) SB H2 (9-mer): Tyr-Pro-Hyp-Hyp-Thr-Hyp-Val-Tyr-Lys (sequenced twice) SB H3 (9-mer): Tyr-Pro-Hyp-Hyp-Thr-Hyp-Val-Tyr-Lys (sequenced once) (Ser-Hyp-Hyp-Hyp-Thr/Tyr?-Hyp-Ser-Pro minor sequence) SB H4 (1 5-mer): Ser-Hyp-Hyp-Val-His-Glu-Tyr-Pro-Hyp-Hyp-Thr-?-Val-Tyr-Lys (sequenced once) SB H5 (1 5-mer): Ser-Hyp-Hyp-[Val-His-Glu-Tyr-Pro]-Hyp-Hyp-Thr-Hyp-Val-Tyr-Lys (sequenced twice) Tom H5 (1 0-mer): Ser-Hyp-Hyp-Hyp-Hyp-Thr-Hyp-Val-Tyr-Lys (from ref 23) Tom H20 (1 6-mer): Ser-Hyp-Hyp-Hyp-Hyp-[Val-Lys-Pro-Tyr-His-Pro]-Thr-Hyp-Val-Tyr-Lys (from ref 23) Amino acid compositions of peptides sequenced Amino acid

Hi

H2a d

H3b d

H2b

H4b.d

H5c.d

molar ratios 5.0 4.3 3.2 3.6 3.4 2.1 Hyp 1.1 1.1 1.0 1.1 1.3 0.1 Thr 1.4 0.9 0.7 0.6 0.2 1.0 Ser 1.1 0.9 0.5 0.8 1.3 0.0 Pro 1.5 1.1 1.0 1.0 1.0 0.7 Val 1.7 1.1 1.4 1.4 1.6 0.0 Tyr 1.0 1.0 0.7 0.8 1.3 1.0 Lys 0.7 0.7 0.2 0.3 0.1 0.8 His b a These peptides were sequenced after further purification Sequenced after further purfication on sulfoethyl aspartamide and C4 columns. c Sequenced after a final cleanup on a sulfoethyl aspartamide column followed by desalting by recycling three times on the PRP-1 column. d The peptide pairs H2/H3 and H4/H5 probably differ only by minor residual glycosylation. on the PRP-1 column. ACKNOWLEDGMENTS We thank Pat Muldoon for amino acid analyses, Dr. Joseph F. Leykam and Ms. Melanie M. Corlew for peptide sequencing, and Dr. John Heckman for the electron microscopy.

11.

12.

LITERATURE CITED 1. Albersheim P, Nevins DJ, English PD, Karr A (1 967) A method for analysis of sugars in plant cell-wall polysaccharides by gasliquid chromatography. Carbohydr Res 5: 340-345 2. Averyhart-Fullard V, Datta K, Marcus A (1988) A hydroxyproline-rich protein in the soybean cell wall. Proc Natl Acad Sci USA 85: 1082-1085 3. Chen J, Varner JE (1985) Isolation and characterization of cDNA clones for carrot extensin and a proline-rich 33-kDa protein. Proc Natl Acad Sci USA 82: 4399-4403 4. Chen J, Varner JE (1985) An extra cellular matrix protein in plants: characterization of a genomic clone for carrot extensin. EMBO J 4: 2145-2151 5. Corbin DR, Sauer N, Lamb CJ (1987) Differential regulation of a hydroxyproline-rich glycoprotein gene family in wounded and infected plants. Mol Cell Biol 7: 4337-4344 6. Esen A, Bietz JA, Paulis JW, Wall JS (1982) Tandem repeats in the N-terminal sequence of a proline-rich protein from corn endosperm. Nature 296: 678-679 7. Everdeen DS, Kiefer S, Willard JJ, Muldoon EP, Dey PM, Li XB, Lamport DTA (1988) Enzymic cross-linkage of monomeric extensin precursors in vitro. Plant Physiol 87: 616-621 8. Franssen HJ, Nap J-P, Gloudemans T, Stiekema W, van Dam H, Govers F, Louwerse J, van Kammen A, Bisseling T (1987) Characterization of cDNA for nodulin-75 of soybean: A gene product involved in early stages of root nodule development. Proc Natl Acad Sci USA 84: 4495-4499 9. Heckman JW, Terhune BT, Lamport DTA (1988) Characterization of native and modified extensin monomers and oligomers by electron microscopy and gel filtration. Plant Physiol 86: 848-856 10. Hong JC, Nagao RT, Key JL (1987) Characterization and sequence analysis of a developmentally regulated putative cell

13. 14.

15.

16. 17.

18. 19.

20.

21.

22.

wall protein gene isolated from soybean. J Biol Chem 262: 8367-8376 Hood EE, Shen QX, Varner JE (1988) A developmentally regulated hydroxyproline-rich glycoprotein in maize pericarp cell walls. Plant Physiol 87: 138-142 Kieliszewski M, Lamport DTA (1986) Cross-reactivities of polyclonal antibodies against extensin precursors determined via ELISA techniques. Phytochemistry 25: 673-677 Kieliszewski M, Lamport DTA (1987) Purification and partial characterization of a hydroxyproline-rich glycoprotein in a graminaceous monocot, Zea mays. Plant Physiol 85: 823-827 Kieliszewski M, Lamport DTA (1988) Tying the knots in the extensin network. In JE Varner, ed, Self-Assembling Architecture. 46th Symposium of the Society for Experimental Biology, 1987, St. Paul, MN, pp 61-77 Kieliszewski M, Leykam JF, Lamport DTA (1989) Structure of the threonine-rich extensin from Zea mays. Plant Physiol 92: 316-326 Laemmli UK (1970) Cleavage of structural proteins during assembly of the head of bacteriophage T4. Nature 227: 680-685 Lamport DTA (1965) The protein component of primary cell walls. Adv Bot Res 2: 151-218 Lamport DTA (1967) Hydroxyproline-O-glycosidic linkage of the plant cell wall glycoprotein extensin. Nature 216: 13221324 Murashige T, Skoog F (1962) A revised medium for rapid growth and bioassay with tobacco tissue cultures. Physiol Plant 15: 473-497 Sanger MP, Lamport DTA (1983) A microapparatus for liquid hydrogen fluoride solvolysis. Sugar and amino sugar composition of Erysiphe graminis and Triticum aestivum cell walls. Anal Biochem 128: 66-70 Showalter AM, Varner JE (1989) The biology and molecular biology of plant hydroxyproline-rich glycoproteins. In A Marcus, ed, The Biochemistry of Plants: A Comprehensive Treatise, Vol 15, Molecular Biology. Academic Press, New York, pp 485-520 Smith JJ, Muldoon EP, Lamport DTA (1984) Isolation of extensin precursors by direct elution of intact tomato cell suspension cultures. Phytochemistry 23: 1233-1239

SUGAR BEET EXTENSIN

23. Smith JJ, Muldoon EP, Willard JJ, Lamport DTA (1986) Tomato extensin precursors P1 and P2 are highly periodic structures. Phytochemistry 25: 1021-1030 24. Stiefel V, Perez-Grau L, Albericio F, Giralt E, Ruiz-Avila L, Dolors Ludevid M, Puigdomenech P (1988) Molecular cloning of cDNAs encoding a putative cell wall protein from Zea mays and immunological identification of related polypeptides. Plant Mol Biol 11: 483-493

333

25. Tierney ML, Wiechert J, Pluymers D (1988) Analysis of the expression of extensin and p33-related cell wall proteins in carrot and soybean. Mol Gen Genet 211: 393-399 26. Wang S-Z, Esen A (1986) Primary structure of a proline-rich zein and its cDNA. Plant Physiol 81: 70-74 27. Yourno J, Kohno T, Roth JR (1970) Enzyme evolution: generation of a bifunctional enzyme by fusion of adjacent genes. Nature 228: 820-824