Download Structure and function studies of plant cell wall polysaccharides

Biochemical Society Transactions
Structure and function studies of plant cell wall polysaccharides
374
Peter Albersheim*t$§, Jinhua An*, Glenn Freshour*+, Melvin S. Fuller*+, Rafael Guillen*ll, Kyung-Sik Ham*t,
Michael G. Hahn*$, ling Huang*, Malcolm O’Neill*, Andrew Whitcornbe*, Myron V. Williams*,
William S. York*t and Alan Darvill*t$
*Complex Carbohydrate Research Center, 220 Riverbend Road, +Department of Biochemistry, Life Sciences
Building, and $Department of Botany, Plant Sciences Building, University of Georgia, Athens, Georgia 30602, U.S.A.
Introduction
The plant cell wall, which is the major source of
biomass and dietary fibre, is a vital natural resource.
Primary plant cell walls, that is, the walls of growing
cells, govern many of the fundamental properties of
plant cells. The walls provide the first barrier to
pests, they physically control the rate of cell growth,
they are the organelle that ultimately controls the
shape of plant cells and consequently of organs and
whole organisms, and they are the source of an
important class of regulatory molecules called
oligosaccharins. Thus, elucidating the structures
and functions of plant cell walls is fundamental to
plant science [ 1-61.
Primary cell walls are composed of about 20%
cellulose microfibrils, 7 0 4 0 % non-cellulosic polysaccharides, and up to 10%structural glycoproteins.
Structural elucidation of primary cell wall polymers
is essential to dissecting and understanding their
roles in wall structure and other physiological functions. Although considerable progress has been
made in delineating the primary structures of the
non-cellulosic cell wall polysaccharides, many
questions remain to be answered before a reasonably accurate picture of primary cell walls can be
visualized. For example, do all cells contain all of
the wall polysaccharides? Do the polysaccharides of
the walls of different cells and different tissues have
the same or different structures? How are the
various polysaccharides distributed in the walls of
the cells of different tissues, organs and species?
What are the three-dimensional structures of these
polysaccharides? How are the polysaccharides themselves and polysaccharides and structural glycoproteins interconnected? Are all the glycosyl linkages
and glycosyl residues of which the wall polysaccharides are composed essential for wall integrity?
Can mutants be obtained that lack one or more wall
component but have normal or at least minimally
functional walls?
Abbreviations used: 2.4-D, 2,4-dichlorophenoxyacetic
acid; RG, rhamnogalacturonan; XG, xyloglucan.
§To whom correspondence should be addressed.
IlCurrent address: Consejo Superior de Investigaciones
Cientificas, Instituto de la Grasa y sus Derivados, Avda.
Padre Garcia Tejero, 4 (Heliopolis), Apartado 1078,
41012 Sevilla, Spain.
Partial answers to these questions are available, and the answers are intriguing. W e have, for
many years, been impressed by the conservation of
the structures of the wall polysaccharides. Cellulose
has the same structure in organisms as diverse as
bacteria and trees. Three galacturonic acid-rich
pectic polysaccharides, rhamnogalacturonan (RG)I, RG-I1 and homogalacturonan, and two cellulosebinding hemicelluloses, glucuronoarabinoxylan and
xyloglucan, are all present in all of the higher plants
that have been studied, and their structures, in even
the most evolutionarily diverse plants, are sufficiently conserved that there is never a problem of
knowing what polysaccharide you are examining.
Rut are their structures the same in all plants and in
all cells of the same plant?
Discussion
A structurally conserved cell wall polysaccharide
RG-I1 appears to be an extreme example of the evolutionary conservation of wall polysaccharide structure. This relatively small ( 6 kDa) polysaccharide,
which is composed of some 11 different sugars, is
an awesome example of the complexity that can be
engendered in carbohydrates by their ability to
form branched structures (Figure 1). W e have
located all of the more than 20 differently linked
glycosyl residues known to be components of RGI1 in one or more of five oligosaccharide fragments.
W e know to which position of galactosyluronic acid
residues in the backbone each of the oligosaccharide side chains is attached, although we have yet to
identify which residues in the backbone contain
which side chain. W e have even begun to identify
the sites of 0-acetylation and methyl esterification.
The amazing thing is that, even with all this structural complexity, we have never seen a single difference in the structure of RG-I1 obtained from plants
as diverse as dicots, monocots (cereals), and gymnosperms. Furthermore, this fascinating molecule
has no known function!
-
Cytochemical studies demonstrate the structural
diversity of individual wall polysaccharides
If the extreme structural conservation of RG-I1 were
the rule for all wall polysaccharides, the answer to
the question ‘Do the polysaccharides of the walls of
Advances in Structural Glycobiology
The partial structure of RG-II in plant cell wall pectic polysaccharides. This
figure shows the a- I ,4-linked galactosyluronic acid backbone of RG-II
together with four structurally characterized side chains, A-D.
v
2
2
PDW
t
t
a-o-F&(1+3)-a~-Fucp
2
4
I
t
Me
1
P~Glcp4
2
t
1
aDGalp
PL-Rhap
3
t
KDOp
5
1
a-L-Fnap
P~-Aral
t
1
1
a-~GalpA-(1+2)-P~-Rhp(3tl).P~Galp4
4
3
2
&D.OHAp
5
2
3’
3’
1
t
1
BDAplf
1
t
3
t
t
t
1
t
1
PL-AceAl
2
C
D
t
1
a-L-Fucp(l-+2)-a.~Galp
2
4
I
t
Me
1
a-L-Anp
2
t
A
1
a 4 - W
B
different cells and different tissues have the same
structure?’ would be straightforward. W e and
others have been attempting to answer this important question by generating monoclonal antibodies
and recombinant Fab fragments to cell wall polysaccharide epitopes and using these carbohydratebinding proteins for cytochemical localization
studies. The results are striking and clearly establish
that at least some of the structures of the non-cellulosic polysaccharides vary from cell to cell and even
among the different wall faces of a single cell.
Our results include the following observations: a binding protein may only bind its epitope in
selected regions of cell walls; a single cell generates
walls, on different sides of the cell, with different
polysaccharide epitopes or with epitopes that have
distinctly different availability to the binding proteins; the walls of a cell located beside a cell that is
heavily stained with a binding protein may not be
stained at all (this includes differential staining of
the two halves of a wall separating two cells, where
one half is made by one cell and the other half by
the other cell); and the presence or availability of
epitopes is developmentally regulated. For example,
an epitope is not detected in the walls of cells for
several days after the walls are formed, but the
epitope is then added to and distributed throughout
the wall. (We recognize that it is also possible the
epitope becomes ‘unmasked but, if that were true, it
would still mean that the structure of the wall
changes in a developmentally regulated manner.)
There are enough examples of developmental
regulation to conclude that the structures of primary
cell wall polysaccharides are not static and that the
‘same’ polysaccharide in different cells has a structure that differs in important ways. Thus, although a
generalized model of the primary cell walls of plants
might be constructed that gives a fair representation
of the locations and physical functions of the six
major wall polysaccharides, it would require many
different models, with a level of structural knowledge not yet available, to give informative pictures
of the structures of the walls of different cell types.
A cell wall polysaccharide whose structure varies
from tissue t o tissue
Xyloglucans (XGs) are found in nature as widely
distributed as RG-I1 but their structures are not as
highly conserved. All XGs have a p- 1,4-1 )-glucan
backbone and I)-xylosyl residues a- 1&linked to
many glucosyl residues of the backbone. XGs bind
strongly to cellulose microfibrils via multiple hydrogen bonds connecting the backbone of XG to the
surface of the cellulose microfibrils. This cross-linking of cellulose microfibrils is thought to give
strength to the wall as well as limit the rate of elongation growth of the wall. An attractive, currently
in-vogue hypothesis is that the rate of elongation
growth is governed by the rate of XG transglycanase-catalysed exchange of XG chains. It is hypothesized that the rate at which XG chains can change
partners determines the rate at which cellulose
375
Biochemical Society Transactions
376
microfibrils move relative to one another and that
this movement limits the rate of cell wall elongation.
Over 40 different oligosaccharide substructures of XGs, containing up t o 20 glycosyl residues
per oligosaccharide, have been isolated and structurally characterized (Figure 2). Newly isolated XG
oligosaccharides can be identified, if they have been
characterized before, simply by examining selected
reporter group signals in their one-dimensional
n.m.r. spectra or by using artificial neural network
pattern recognition technology developed in our
centre. Even XG oligosaccharides that have never
been seen before can be identified from their structural reporter groups and, eventually we anticipate,
by neural network technology.
The XG subunits present in different plants
show large differences in their structures. Methods
are now being perfected to isolate and characterize
the X C oligosaccharides generated from milligram
quantities of tissue. This will make it possible to
determine whether and how the subunit compositions of XGs vary from cell type to cell type. In
other words, some of the cell-specific epitopes
observed with monoclonal antibodies may well be
caused by differences in XG structures. Cell-specific epitopes are probably also caused by structural
variations in KG-I and glucuronoarabinoxylan.
-
Oligosaccharins oligosaccharides with
regulatory functions
XG appears not only to have an important physical
function in primary cell walls, but considerable
evidence obtained in several laboratories suggests
that XG-derived oligosaccharides regulate the rate
of elongation growth. Oligosaccharides with regulatory functions are commonly referred to as oligosaccharins.
The most studied oligosaccharins are those
that regulate defence responses to pests. The ability
of oligosaccharins to regulate physiological functions other than those involved in defence against
pests became apparent when it was shown that the
XG-derived nonasaccharide XXFG (Figure 2) inhibits, at a concentration of 10
10 - " M,that portion
of the growth of pea stem segments that is stimulated by the addition, to the incubation medium, of
the auxin (phytohormone) analogue 2,4-dichlorophenoxyacetic acid (2,4-D) at a concentration of
10-"M. The i,-fucosyl residue of XXFG is essential
for its oligosaccharin activity. Removal of the
fucosyl residue with a fucosidase abolishes the
oligosaccharide's growth-inhibitor activity. This
conclusion is substantiated by the inability of either
the octasaccharide XXLG or the heptasaccharide
XXXG (Figure 2) to inhibit the 2,4-I>-stimulated
growth of the pea stem segments.
~
'-
*
Structures of some of the smaller XG oligosaccharide fragments
XXGol
GXXGol
v
XLGol
XLXGol
XXXGol
XXLGol
XLLGol
4r
7
GXFGol
XXFGol
XLFGol
[I P-~-Glcp-(I -4)-
A
a-D-Xylp-( 1 - 6 ) -
B-D-Gatp-( I
&
FGol
XFFGol
-.2)-
u-L-Fuc~-(
I + 2)-
Advances in Structural Glycobiology
A variety of XG-derived oligosaccharides,
whether or not they contain the fucosyl residue
required for them to be growth inhibitors, slightly
stimulate the growth of pea stem segments grown in
the absence of added 2,4-D. The growth stimulation
is only observed when the oligosaccharides are at
concentrations of 10-"M and above, i.e. at oligosaccharide concentrations at least 100-fold greater than
are required to inhibit the growth of the same tissue.
All of the XC, oligosaccharides that promote
growth possess a free a-1)-xylosyl-1,6-/3-1)-glucosyl
moiety located at their non-reducing terminus, providing evidence that these two glycosyl residues
participate in the growth-promoting action of these
oligosaccharides. It is not a coincidence that exactly
the same two glycosyl residues are required at the
non-reducing termini of XG oligosaccharides in
order for the oligosaccharides to function as acceptor substrates for XG transglycanase, the putative
growth-promoting enzyme. The a-i)-xylosyl-l ,fip-i)-glucosyl disaccharide of the XG oligosaccharides is apparently required for XG transglycanase
to stimulate growth of pea stems.
The inhibitory effect of XXFG on 2,4-Dstimulated growth of pea stems exhibits a concentration optimum ( 10-X-lO-"M), a phenomenon
which is sometimes exhibited by the classical
phytohormones. Higher and lower concentrations
o f XXFG have little or no inhibitory effect on the
growth rate, although XXFG, at much higher concentrations, has a small stimulatory effect on the
growth rate of the pea stems, described above.
XXFG possesses the fucosyl-containing side chain
that is required to inhibit growth of the pea stems,
and the oligosaccharide also possesses the free a+xylosyl- 1,h-P-i)-gIucosyl disaccharide at its nonreducing terminus that is required, at lO-"M, to
stimulate growth. The abilities of this oligosaccharin
to inhibit growth at lower concentrations and to
stimulate growth at higher concentrations can
explain the observed concentration optimum exhibited by XXFG when measuring its ability to inhibit
growth.
Octasaccharide XXI,G and heptasaccharide
XXXG, on the other hand, which show no ability to
inhibit the growth of pea stems at any concentration
tested, are nevertheless, at the higher concentrations, able to slightly stimulate the growth of pea
stems. This result is expected, as neither of these
oligosaccharides possesses the fucosyl residue
required for inhibiting growth and both these oligosaccharides possess the free a-i)-xylosyl- 1,6-p-i)glucosyl disaccharide at their non-reducing termini,
which is required to stimulate growth. Thus, the
biological activities of XXLG and XXXG, in agreement with experimental observation, are not
expected to show a concentration optimum.
Pentasaccharide FG inhibits the 2,4-I>stimulated growth of pea stems at 1 0 - X Mand above
but shows no ability, at any concentration, to
stimulate the growth of pea stems. Again, this result
is expected, as FG possesses the fucosyl-containing
tetrasaccharide responsible for growth inhibition,
but FG does not possess the free a-i)-xylosyl-1,6P-i)-gIucosyl disaccharide at its non-reducing
terminus required to stimulate growth. Thus, once
again, in agreement with experimental observation,
the biological activity of FG does not show a concentration optimum.
Conclusion
There are now six well-characterized oligosaccharins. Their activities are plant- and tissue-specific,
and they elicit such diverse phenomena as the formation of flowers or vegetative shoots (depending
on the tissue), the inhibition of root growth, the
synthesis of antimicrobial compounds, lignification,
activation of ion transport, depolarization of the
plasma membrane, stimulation of an oxidative
burst, root hair deformation, cortical cell divisions,
and pseudo-nodule formation, in addition to the
inhibition and stimulation of elongation growth
described in this paper.
Exciting areas of ongoing oligosaccharin
research include studies of the families of enzymes
that generate and degrade oligosaccharins and the
proteins that specifically bind to and inhibit those
enzymes. Studies of the physiological receptors of
oligosaccharins also show great promise. The
enzymes that form and degrade oligosaccharins and
their inhibitors, together with the physiological
receptors of oligosaccharins, are almost certainly
largely responsible for when and where oligosaccharins are active in plant tissues. These proteins
undoubtedly determine the developmental expression of oligosaccharins as well as whether oligosaccharins are at effective levels such as during
attempted pathogenesis. These proteins may even
be involved in determining race/cultivar specificity
of plant-microbe interactions. It is fair to conclude
that the studies of plant cell wall polysaccharide
structures, which led to this research, have had
unexpected rewards.
This work was supported in part by the LJ.S. Lkpartment
Energy (DOE:)-funded (UE-FG09-93EK20097)
Center for Plant and Microbial Complex Carbohydrates
of
377
Biochemical Society Transactions
(to P.A. and
A.Lj.) and DOE grants LE-FG0593EK20114 (to P.A.) and LlE-FG05-93ER20115 (to
A.D.).
378
1 O’Neill, M., Albersheim, 1’. and Darvill, A. (1990) in
Methods in Plant Biochemistry, pp. 41.5-441,
Academic I’ress. New York
2 Meyer, H., Hansen, T., Nute, D., Albersheim. P.,
L>arvill,A,, York, W. and Sellers.J. (1991) Science 251.
542-544
3 Darvill, A., Augur, C., Hergmann, C., Carlson, R. W.,
Cheong, J.-J., Eberhard, S., Hahn. M. G., Lo, V.-M.,
Marfa, V., Meyer, H.. Mohnen, D., O’Neill. M. A,, Spiro,
M. D., Van Halbeek, H., York, W. S. and Albersheim,
1’. (1 992) Glycobiology 2, 181- 198
4 York, W. S., Harvey, I,. K., Guillen, K.,Albersheim, P.
and Darvill, A. G. (1993) Carbohyd. Kes. 248,
285-301
5 Kormelink, F. J. M., Hoffmann, K. A,, Gruppen, H.,
Voragen, A. G. J., Kamerling, J. P. and Vliegenthart, J.
F. G. (1993) Carbohyd. Kes. 249, 369-382
6 Fry?S. C. (1993)Curr. Biol. 3, 355-357
Keceived 24 December 1993
Structure and biosynthesis of carbohydrate ligands for animal lectins involved in
cell adhesion
Moon Jae Cho and Richard D. Cummings*
The Department of Biochemistry and Molecular Biology, University of Oklahoma Health Sciences Center,
P.O. Box 2690 I , BSEB 325, Oklahoma City, Oklahoma 73 190, U.S.A.
Introduction
Well over 50 different carbohydrate-binding proteins or lectins have been identified in animal tissues, and many of these are now known to be
important in cell-cell adhesion and cell-matrix
adhesion [l]. Each of the known lectins interacts
with different carbohydrate ligands, and there is
now great interest in identifying the ligands and
gaining more insight into the biosynthesis of both
the ligands and the lectins and the functional consequences of their interactions. In this regard, our laboratory has been studying a number of the
Ca‘+ -dependent (C-type) lectins, such as P-selectin,
and the soluble (S-type) Calf -independent lectins,
such as I,- 14.
I,-14 is a 14 kDa non-glycosylated protein
produced in the cytoplasm of many different types
of cells, including epithelial cells, neuronal cells, fibroblasts and muscle cells. Original experiments on
I,- 14 had indicated that it recognizes terminal blinked galactosyl residues on glycoconjugates, and
it was therefore defined as a b-galactoside-binding
lectin [ 21. Hut results from several laboratories indicated that I,- 14 interacted with high-molecularmass glycoconjugates [ 3-51. Our laboratory
initiated studies on the carbohydrate-binding specificity of I,-14 by using an affinity-chromatography
approach. We examined the interaction of many
Abbreviations used: GNT-V, N-acetylglucosaminyltransferase V; LAMP, lysosome-associated membrane glycoprotein.
‘To whom correspondence should be addressed.
types of oligosaccharides with immobilized I,- 14,
and our results demonstrated that immobilized
L-14, purified from either bovine or porcine heart
tissue, binds with high affinity to Asn-linked oligosaccharides containing the repeating disaccharide
unit [3Gal/?1-4GlcNAc/31] or poly-N-acetyllactosamine sequence [ 6 ] . The presence of terminal bgalactosyl residues was not required for binding of
these oligosaccharides, as they contain internal Bgalactosyl residues [ 7 ] .
Subsequently, we used the affinity-chromatography approaches and direct binding assays to
identify intact glycoproteins that bind avidly to I,-14
[8,9]. Total cell extracts were passed over affinity
columns of immobilized bovine I,-14 and the
bound glycoproteins analysed. Only a few glycoproteins bound to I,-14 and they were identified as
lysosome-associated
membrane
glycoproteins
(LAMPS) 1 and 2 and laminin [7-91. Hoth of these
glycoproteins were shown to contain abundant
amounts of poly-N-acetyllactosamine, which was
responsible for their interaction with I,- 14.
These results caused us to focus attention
on the biosynthesis of poly-N-acetyllactosamine
sequences and their distribution on glycoproteins.
A number of different glycoproteins are known to
contain Asn-linked oligosaccharides with poly-Nacetyllactosamine sequences, but in many examples
the poly-N-acetyllactosamine chains occur on complex-type tri- and tetra-antennary Asn-linked oligosaccharides containing 2,6-branched a-mannose.
The enzyme responsible for adding the %-branch’
to Asn-linked oligosaccharides is N-acetylglucos-