| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
|
Plant Physiology 132:1781-1789 (2003) © 2003 American Society of Plant Biologists If Homogalacturonan Were a Side Chain of Rhamnogalacturonan I. Implications for Cell Wall Architecture1Graduate School Experimental Plant Sciences, Laboratory of Plant Breeding, Wageningen University, Binnenhaven 5, 6709 PD Wageningen, The Netherlands (J.-P.V., R.J.F.J.O., R.G.F.V.); Laboratory of Food Chemistry, Wageningen University, Bomenweg 2, 6703 HD Wageningen, The Netherlands (J.-P.V., H.A.S., A.G.J.V.); Department of Biological Sciences, Purdue University, West Lafayette, Indiana 47907–1392 (M.C.M.); and Biotechnology Group, Danish Institute of Agricultural Sciences, 40 Thorvaldsensvej, DK–1871 Frederiksberg C, Denmark (P.U.)
Pectin, an important cell wall component of dicotyledonous plants, is probably the most complex macromolecule in nature. Here, we critically summarize the large amount of data on pectin structure. An alternative model for the macromolecular structure of pectin is put forward, together with ideas on how pectins are integrated into the plant cell wall.
Pectin is composed of as many as 17 different monosaccharides (Fig. 1; for review, see Ridley et al., 2001  ;
Voragen et al., 2003). These
monosaccharides are organized in a number of distinct polysaccharides, the
structures of which are schematically shown in
Figure 1. Together, these
polymers form the pectin network (Visser
and Voragen, 1996; Ridley et
al., 2001; Voragen et al.,
2003).
The first structural elements of pectin to be discussed have a backbone of
1,4-linked
HGs can contain clusters of four different (heterooligomeric) side chains
with very peculiar sugar residues (such as Api, AceA, Dha, and Kdo). These
side chains, together with the approximately nine galacturonyl residues to
which they are connected, are referred to as RG-II
(O'Neill et al., 2001
Another constituent polysaccharide of pectin, RG-I, is composed of a
repeating disaccharide unit
[
Arabinogalactan II (AG-II), is mainly associated with proteins
(arabinogalactan proteins or AGPs), and it is still unclear whether this
polysaccharide is part of the pectin complex. Pectin and AG-II often seem to
co-extract and are subsequently difficult to separate from each other,
suggesting that they can be covalently linked. AG-II is composed of 1,3-linked
Although the fine structure of the various pectic constituents is largely known, it is still unclear how these structural elements are combined into a macromolecular structure. It is likely that these constituents are covalently linked to each other (Ridley et al., 2001 ; Voragen et al.,
2003). For a long time, pectin was either depicted as is shown in
Figure 2A, in which HG and RG-I
form one continuous backbone (Visser and
Voragen, 1996), or the connection between HG and RG-I was left
undiscussed. The most important feature of the model in
Figure 2A is that the
"pectic backbone" is an extended chain comprising RG-I (including
the neutral hairs [not shown]), HG (or XGA [not shown]), and (isolated) Rha
residues interspersing the HG regions. This model was consistent with the
observed length periodicity of HG (approximately 70–100 GalA residues;
Thibault et al., 1993),
endo-polygalacturonase-resistant rhamnogalacturonan regions with a GalA to Rha
ratio of >1, and the typical RG-I structure
(McNeil et al., 1980;
Visser and Voragen, 1996).
Recently, we have summarized evidence for an alternative macromolecular
structure of pectin (Fig. 2,
B–D) that is more consistent with the most recent results in
pectin research. The main difference between the alternative models and that
in Figure 2A is that the HG
constituents are depicted as side chains of RG-I. The most important
motivations for proposing this alternative structure are briefly described
below (for a detailed discussion, see
Voragen et al., 2003). (a) It
seems unlikely that the biosynthetic machinery inserts rhamnosyl residues in
the nascent pectic backbone in an irregular pattern. (b) The length
periodicity of HG is explained in a different way. Upon treatment of pectin
with dilute acid, the
 -L-Rhap-(1 4)--D-GalpA
linkages in rhamnogalacturonans are (selectively) split. It seems likely that
similarly sized HG fragments would be obtained with both models. (c) A
thorough study in which pectin was exhaustively treated with
endo-polygalacturonase, followed by a careful fractionation of the resulting
fragments, has not provided evidence for the existence of isolated Rha
residues interspersing two HG structural elements
(Zhan et al., 1998). (d)
Studies in which pectin preparations were treated with exoPG or endo-XGA
hydrolase suggest that XGA is not an integral part of the pectic backbone
(J.-P. Vincken, G. Beldman, and H.A. Schols, unpublished data). Because XGA
and HG have the same backbone structure, it is not unreasonable to hypothesize
that this holds also for HG. Recently, Oechslin et al.
(2003) have also suggested
that XGA might be a side chain of RG-I, based on a detailed structural
investigation of pectic substances that were strongly associated with
cellulose. (5) Atomic force microscopy (AFM) of citrus pectin indicated that
HG branches might be connected to a pectic backbone
(Round et al., 2001). Although
the interpretation of the micrographs is rather complex, we think that the AFM
results could indicate that HG side chains are attached to an RG-I
backbone.
AFM will be a powerful tool to elucidate the macromolecular structure of pectin further, in particular when this technique is combined with selective (enzymic) degradation procedures. Selective degradation of pectin (enzymes and/or dilute acid) combined with chromatographic separation of the products is required to elucidate the exact nature of the point of attachment of HG to RG-I. Other challenges will be to establish how the various hairs, including HG, are distributed in the RG-I backbone. In principle, an unprecedented number of "hair styles" is possible. RG-I may carry only one kind of side chain (Fig. 2B). Other possibilities are a random distribution of hairs (Fig. 2C), cluster-like distributions of hairs of the same kind (Fig. 2D), and cluster-like distributions of different kind of hairs (not shown). It should be mentioned that it is unlikely that all hairs point into one direction. It is probably more realistic to draw the hairs in various directions, perpendicular to the RG-I backbone. However, in Figure 2 (and Fig. 4) this has not been done, because it would make the picture(s) too complicated. The fact that HRs can be partially degraded by treatment with endo-rhamnogalacturonan hydrolase or endo-rhamnogalacturonan lyase (RGL) lends some support to the existence of cluster-like distributions. The HG, arabinan, and AG-I (and AG-II) hairs are very different in nature (charge and conformation). In principle, different micro-environments may be created within the wall by combining different kind of hairs. Possibly, this plays an important role in regulating wall properties at various developmental time points and in different tissues.
In the previous section we addressed the large structural diversity of pectin macromolecules. In the plant cell wall, these macromolecules assemble in much larger networks. Three different kinds of cross-links for interconnecting pectic molecules will be discussed below.
HGs can adopt different conformations in solution of which the right-handed
2- (21) and 3-fold (31) helices seem to be most
favorable in terms of minimal energy
(Braccini et al., 1999
Two molecules of RG-II can complex with boron, forming a borate-diol ester
(Fig. 3B;
Ishii et al., 1999
Some time ago it was suggested that HG could be cross-linked to other
components by uronyl esters (Fig.
3C; Brown and Fry,
1993
Over the years, a number of monoclonal antibodies (mAbs) have been generated that recognize different pectic structures in the wall. The epitopes of these mAbs have recently been reviewed (Willats et al., 2001a ).
Microscopic analysis of walls labeled with various mAbs have shown that pectin
greatly contributes to the heterogeneity of the plant cell wall. This was
elegantly demonstrated by the work of Willats et al.
(2001b), in which three
antibodies (PAM1, JIM5, and LM7) recognizing different patterns of methoxyl
esterification were used. They showed that the wall around one cell can
contain discrete microdomains, each accumulating a particular type of HG.
Pectic polysaccharides are also deposited in a tissue-specific manner. For
example, the distribution of pectic epitopes in the periderm of potato
(Solanum tuberosum) tubers differs from that of the parenchymatous
cells (for review, see Voragen et al.,
2003, and refs. therein). Last but not least, different pectic
epitopes are deposited at different stages of development. For instance,
galactan epitopes appear later in the process of tuberization in potato than
those of arabinan.
The observations above suggest that the wall acquires a set of pectic
molecules fitted to specific developmental stages or at a given location. In
very general terms, the following rules may apply: (a) various HG epitopes can
surround one cell with that of Ca2+-pectate confined to
the middle lamella regions, cell corners, and pit fields containing
plasmodesmata; (b) arabinan is generally present throughout the wall of
dividing cells, whereas galactan is mainly present close to the plasma
membrane of expanding cells; (c) RG-II seems to be present throughout the
wall, except in the middle lamella region; (d) AG-II epitopes tend to be
localized close to the plasma membrane or in walls of particular cell types,
for example, tracheary elements (Ridley et
al., 2001
The precisely regulated deposition of pectic polysaccharides suggests that the various constituents of pectin each have their specific functions in determining the properties of the wall (e.g. porosity and water-holding capacity). At this moment, our picture of whether or how individual pectic components contribute to particular properties of the wall is very incomplete. The main reason for this is that our knowledge on the biosynthesis of cell wall polysaccharides, pectin in particular, is still in its infancy (Ridley et al., 2001 ). The
first glycosyltransferase gene involved in polymerizing the backbone of a
pectic structural element remains to be identified, although it seems as if
important progress has been made very recently (see below). The first enzyme
involved in decorating one of the pectic backbones has been identified only a
few months ago (Iwai et al.,
2002). Because of this, no attempts have been made to interfere
with pectin biosynthesis using a reverse genetics approach.
A number of mutants with an altered pectin structure/deposition have been
generated (Penfield et al.,
2001
An alternative approach for determining the biological significance of the
various structural elements of pectin is to express fungal pectinases in muro.
Such experiments have been done with three enzymes derived from
Aspergillus aculeatus: endogalactanase (EGAL;
Sørensen et al., 2000 The wild-type and transgenic potato tuber cell walls were also investigated using microscopy in combination with antibody labeling. The LM5 antibody was used to probe the presence of galactan hairs and as a (indirect) diagnostic tool for RG-I. Compared with wild-type potato tubers, the EGAL-expressing tubers had far fewer galactan epitopes recognized by LM5 in their walls, demonstrating that the EGAL effectively removed the galactan from the cell wall. In addition to the complete absence of LM5 epitopes in the primary walls of RGL transformants, a redistribution of the LM5 label to the middle lamella region was observed in some cases. The in muro cleavage of the RG-I backbone with the concomitant redistribution of the LM5 epitope suggests that the galactan hairs do not interact with other cell wall components.
Typically, potato tuber cell walls seem to tolerate a low level of either
galactan or arabinan hairs under the growth conditions applied, because both
the tubers and the appearance of the cells of the EGAL and EARA transgenic
potato tubers look normal (Sørensen
et al., 2000
In the previous sections, we have seen that pectin comprises a large collection of pectic molecules, which can differ considerably in their hair style. In the wall, these molecules are cross-linked in different ways forming a larger network. By tuning the biosynthesis/deposition of various pectic molecules and the secretion of wall-modifying enzymes, different microdomains can be formed in the wall, which are expected to provide site-specific properties for facilitating various processes. This last section will deal with how the pectic network might be formed in the wall. In current wall models, a cellulose-xyloglucan network co-exists with a pectin network (McCann and Roberts, 1991 ).
The former is thought to be the load-bearing structure of the wall, whereas
the latter determines its porosity. Cosgrove
(2000) hypothesized three
possible models for the plant cell wall: (a) the "sticky network",
(b) the "multi-coat", and (c) the "stratified wall"
model. The first model is characterized by xyloglucan tethering microfibrils
in one layer and between layers. In the second model, cellulose microfibrils
act as a kind of nucleus around which hemicellulose and subsequently pectin
are deposited; there is no direct interlinking of microfibrils. The third
model shows alternating layers of cellulose/xyloglucan and pectin; xyloglucan
tethers microfibrils within one layer. In all of these models, pectin fills
the gaps in the cellulose-xyloglucan network, and little consideration is
given to the occurrence of microdomains and the formation of the middle
lamella region. Our new ideas concerning pectin macromolecular structure, in
which pectic molecules are regarded as molecular brushes with different
hairstyles (Fig. 2, B–D),
have implications for the assembly of the plant cell wall. We propose a new
model (Fig. 4), featuring
explanations for (a) the formation of the middle lamella and (b) of
microdomains, and (c) control of wall thickness. We assume that the molecular
brushes are formed as such in Golgi, and that they are not assembled outside
the cell by the action of transferases. Contrary to the three models discussed
above, pectin has a much more important role than filling the interstices of
the cellulose/xyloglucan network. In our view, the wall consists of a
cellulose/xyloglucan framework, which is embedded in a matrix of
interconnected modular pectins. This will be elaborated below; for a more
detailed discussion of the cellulose/xyloglucan network, we refer to Cosgrove
(2000). To this end, it is not
entirely clear which polysaccharide is deposited first in the cell plate:
pectin, xyloglucan, or cellulose. Some papers argue for codeposition of pectin
and cellulose in the cell plate (Matar and
Catesson, 1988), whereas others argue for pectin first
(Shea et al., 1989). For our
model, we have assumed that pectin is deposited before cellulose.
Usually, the middle lamella is represented as a separate layer between
cells: a Ca2+-pectate gel (high Mr
HG with a low degree of esterification). The questions of why these polymers
accumulate here and whether these middle lamella HGs are connected to other
wall polymers have not really been addressed. We propose that
(methyl-esterified) HG-rich molecular brushes (with virtually no arabinan or
galactan) are deposited first in the cell plate of two daughter cells. After
this, subsequent layers of pectin molecules are deposited. HG hairs can be
de-esterified by PMEs, (some of) which are suggested to act in a block-wise or
processive fashion. The unesterified regions of HGs (from different cells)
become sensitive to Ca2+ and can form a gel (indicated
with 1 in Fig. 4A).
Interestingly, microscopic studies with labeled antibodies have shown that PME
and Ca2+-pectate can colocalize to areas of cell-cell
contact such as the middle lamella (Morvan
et al., 1998 After the first layer of pectin has been laid down, two processes are expected to run in parallel, leading to the formation of the primary cell wall and the middle lamella. (a) Newly synthesized (in Golgi) cell wall polymers (pectin, but also xyloglucans) are continuously being added on both sides of the Ca2+-pectate gel (or its precursor), which is pushed away from the cell membrane. Eventually, the middle lamella is formed. (b) Cellulose microfibrils are synthesized at the cell membrane and extruded into the extracellular matrix, where they can be cross-linked by xyloglucans. This is also a continuous process, which can be regarded as slowly winding threads around a cell. The cellulose microfibrils are displaced from the plasmalemma by deposition of material synthesized in Golgi, explaining the fact that different lamellae of cellulose microfibrils are encountered in the wall. In Figure 4A, the addition of two extra pectin/cellulose layers is shown, but it should be realized that more layers can be inserted in a similar way. For simplicity, xyloglucans (and other hemicelluloses) are not indicated in the figure.
We hypothesize that the pectin molecules in the second (and following
layers) are secured in the wall by two predominant mechanisms. First, there is
the physical constraint in which the RG-I backbone of the pectic molecules are
pushed against the cellulose microfibres (indicated with 3 in
Fig. 4A; see also
Fig. 4B), because new material
is continuously being added to the wall at the plasmalemma. The hairs fill
spaces between the cellulose microfibrils (indicated with 5 in
Fig. 4B), thereby restricting
the lateral freedom of movement of the pectic brushes. There is some evidence
that mainly the pectin molecules highly substituted with side chains associate
in one way or another with cellulose (or cellulose/xyloglucan), because
high-strength alkaline solutions (in combination with a subsequent enzyme
treatment) were required to extract these polysaccharides
(Orfila et al., 2002
In summary, we propose that the pectin network is assembled in a roof
tile-like fashion. The RG-I backbone probably runs parallel with the cell
membrane, whereas the hairs have a more perpendicular orientation with respect
to the RG-I backbone. The HG hairs are presumably longer than the neutral ones
and are postulated to overlap partially (as shown in
Fig. 4A) or completely with
those of the flanking pectic layers to facilitate cross-linking thru RG-II. If
we consider the observed length periodicity of HG of approximately 70 to 100
GalA residues (Thibault et al.,
1993
The model presented in Figure
4 is consistent with the presence of microdomains in the wall. In
our view, the various pectic molecules are anchored in the wall and,
consequently, they cannot move around freely. Pectic molecules with galactan
hairs, which are usually deposited later in development, would be expected to
localize in the pectin layer closest to the cell membrane. More importantly,
the model can explain that pectin can build a wall of normal thickness when a
cellulose-xyloglucan network is virtually lacking
(Wells et al., 1994
In this paper, we have provided for the first time, to our knowledge, a detailed picture on how the pectin network may be present in the plant cell wall. Our model is rooted in a putative structure in which HG, XGA, (arabino) galactan, and arabinan occur as side chains to RG-I, forming a kind of molecular brush. By combining different hairs in various fashions, a large number of hairstyles can be made. The forthcoming pectin modules may be deposited in the wall in a time-specific manner, enabling the plant cell to create different micro-environments in the wall and to adapt to varying conditions and needs at a given time point or location. It is clear that a number of things need to be substantiated further. First, the cross-link between RG-I and HG needs to be further established, which will not be an easy task. Second, the distribution of hairs in the RG-I backbone will be an enormous challenge, i.e. do the hairs occur in an at random or in block-wise fashion in RG-I? Third, a number of questions concerning RG-II remain to be answered: How many RG-II elements do HG hairs contain? How are the RG-II elements distributed in the HG backbone? Besides structural details, these investigations are expected to provide clues on the nature of the priming molecules for the biosynthesis of the various hairs. In parallel, a more detailed characterization of glycosyltransferases (when discovered) will provide a clearer picture on the range of different polysaccharide structures that can be expected. With the identification of the first enzymes belonging to the pectin biosynthetic machinery, exciting times in pectin research seem to be ahead of us. Experiments aimed at modifying the expression of these enzymes in plants will help in understanding the biological significance of the various polysaccharides. These studies will no doubt provide important spin-off to industry, striving to understand/predict pectin functionality in different applications. Received February 19, 2003; returned for revision March 20, 2003; accepted April 29, 2003.
www.plantphysiol.org/cgi/doi/10.1104/pp.103.022350.
1 This work was supported by the European Union (CT97 2224 "Remodelling
pectin structure in plants"). M.C.M. is the recipient of a University
Research Fellowship from the Royal Society. * Corresponding author: e-mail Jean-Paul.Vincken{at}wur.nl; fax 31–317–483457.
Bouton S, Leboeuf E, Mouille G, Leydecker M-T, Talbotec J, Granier F, Lahaye M, Höfte H, Truong H-N (2002) Plant Cell 14: 2577–2590 Braccini I, Grasso RP, Pérez S (1999) Carbohydr Res 317: 119–130[CrossRef][Web of Science][Medline] Brown JA, Fry SC (1993) Plant Physiol 103: 993–999[Abstract] Cosgrove DJ (2000) Plant Physiol Biochem 38: 109–124[CrossRef][Web of Science][Medline] Daas PJH, Boxma B, Hopman AMCP, Voragen AGJ, Schols HA (2001) Biopolymers 58: 1–8[Medline]
Fleischer A, O'Neill MA, Ehwald R (1999)
Plant Physiol 121:
829–838 Gaspar Y, Johnson KL, McKenna JA, Bacic A, Schultz CJ (2001) Plant Mol Biol 47: 161–176[CrossRef][Web of Science][Medline] Gelineo-Albersheim I, Darvill A, Albersheim P (2001) Book of Abstracts of the Ninth International Cell Wall Meeting, September 2–7, Toulouse, France. CNRS/INRA, Toulouse, p 183 Henrissat B, Coutinho PM, Davies GJ (2001) Plant Mol Biol 47: 55–72[CrossRef][Web of Science][Medline] His I, Driouich A, Nicol F, Jauneau A, Höfte H (2001) Planta 212: 348–358[CrossRef][Web of Science][Medline] Ishii T, Matsunaga T (2001) Phytochemistry 57: 969–974[CrossRef][Web of Science][Medline]
Ishii T, Matsunaga T, Hayashi N (2001)
Plant Physiol 126:
1698–1705
Ishii T, Matsunaga T, Pellerin P, O'Neill MA, Darvill A,
Albersheim P (1999) J Biol Chem
274:
13098–13104 Iwai H, Ishii T, Satoh S (2001) Planta 213: 907–915[Web of Science][Medline]
Iwai H, Masaoka N, Ishii T, Satoh S (2002)
Proc Natl Acad Sci USA 99:
16319–16324 Lau JM, McNeil M, Darvill AG, Albersheim P (1987) Carbohydr Res 168: 245–274[CrossRef]
Liners F, Thibault J-F, van Cutsem P (1992)
Plant Physiol 99:
1099–1104 Matar D, Catesson AM (1988) Protoplasma 146: 10–17 McCann MC, Roberts K (1991) In CW Lloyd, ed, The Cytoskeletal Basis of Plant Growth and Form. Academic Press, New York, pp 109–129
McNeil M, Darvill AG, Albersheim P (1980)
Plant Physiol 66:
1128–1134 Morvan O, Quentin M, Jauneau A, Mareck A, Morvan C (1998) Protoplasma 202: 175–184[CrossRef] Oechslin R, Lutz MV, Amado R (2003) Carbohydr Polym 51: 301–310[CrossRef]
O'Neill MA, Eberhard S, Albersheim P, Darvill AG
(2001) Science 294:
846–849 Oomen RJFJ, Doeswijk-Voragen CHL, Bush MS, Vincken J-P, Borkhardt B, van den Broek LAM, Corsar J, Ulvskov P, Voragen AGJ, McCann MC et al. (2002) Plant J 30: 403–413[CrossRef][Web of Science][Medline] Orfila C, Huisman MMH, Willats WGT, van Alebeek G-JWM, Schols HA, Seymour GB, Knox JP (2002) Planta 215: 440–447[CrossRef][Web of Science][Medline]
Penfield S, Meissner RC, Shoue DA, Carpita NC, Bevan MW
(2001) Plant Cell
13:
2777–2791 Ridley BL, O'Neill MA, Mohnen D (2001) Phytochemistry 57: 929–967[CrossRef][Web of Science][Medline] Round AN, Rigby NM, MacDougall AJ, Ring SG, Morris VJ (2001) Carbohydr Res 331: 337–342[CrossRef][Medline] Sabba RP, Durso NA, Vaughn KC (1999) Int J Plant Sci 160: 275–290[CrossRef] Schols HA, Bakx EJ, Schipper D, Voragen AGJ (1995) Carbohydr Res 279: 265–279[CrossRef] Shea EM, Gibeaut DM, Carpita NC (1989) Planta 179: 293–308
Shedletzky E, Shmuel M, Trainin T, Kalman S, Delmer D
(1992) Plant Physiol
100:
120–130
Skjøt M, Pauly M, Bush MS, Borkhardt B, McCann MC,
Ulvskov P (2002) Plant Physiol
129:
95–102
Sørensen S, Pauly M, Bush M, Skjøt M, McCann MC,
Borkhardt B, Ulvskov P (2000) Proc Natl Acad
Sci USA 97:
7639–7644 Thibault J-F, Renard CMGC, Axelos MAV, Roger P, Crépeau M-J (1993) Carbohydr Res 238: 271–286[CrossRef] Visser J, Voragen AGJ (1996) Progress in Biotechnology 14: Pectins and Pectinases. Elsevier, Amsterdam Voragen AGJ, Schols HA, Visser RGF (2003) Advances in Pectin and Pectinase Research. Kluwer Academic Publishers, Dordrecht, The Netherlands Wells B, McCann MC, Shedletzky E, Delmer D, Roberts K (1994) J Microsc 173: 155–164 Willats WGT, McCartney L, MacKie W, Knox JP (2001a) Plant Mol Biol 47: 9–27[CrossRef][Web of Science][Medline]
Willats WGT, Orfila C, Limberg G, Buchholt HC, van Alebeek
G-JWM, Voragen AGJ, Marcus SE, Christensen TMIE, Mikkelsen JD, Murray
BS et al. (2001b) J Biol Chem
276:
19404–19413 Zhan D, Janssen P, Mort AJ (1998) Carbohydr Res 308: 373–380[CrossRef][Web of Science][Medline] This article has been cited by other articles:
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ASPB Publications | PLANT PHYSIOLOGY® | THE PLANT CELL | |
|---|---|---|---|
TOP
CONSTITUENT POLYMERS, PRIMARY...
MODELS FOR THE MACROMOLECULAR...
-D-Xylp-(1
