 |
INTRODUCTION |
Perhaps of all the cellular
structures in plants, the cell wall has been the slowest to reveals its
secrets. In no small part, this is because the classical tools of
biochemistry have often proven to be too crude to effectively elucidate
structures as biochemically complex and ornate as the plant cell wall.
The techniques of molecular biology, however, have provided new
approaches toward understanding the plant cell wall, and the remarkable
progress that has been made in this regard in recent years was in great display at the Ninth International Cell Wall Meeting held this past
September in Toulouse, France. This convention brought together hundreds of scientists from scores of countries to discuss both formally and informally the enormous strides that have been made in
recent years in understanding plant cell wall function. Among the
highlights of the meeting were many pioneering reports concerning efforts that are under way to use micro-array analysis to study the
effects of specific physiological situations related to cell walls,
although none of these studies have yet to come to full fruition.
Studies of cell wall mutants were also prominent, as was the more
widespread application of new techniques, such as Fourier transfer
infrared resonance (FTIR) spectroscopy, for identifying such mutants.
Enormous progress has also been made in recent years in understanding
arabinogalactan-protein function. The discussion that follows is but a
small taste of the bountiful feast of plant cell wall research that was presented.
| |
SYNTHESIS OF CELL WALL POLYSACCHARIDES |
The identification of the biosynthetic enzymes involved in cell
wall biosynthesis remains one of the major unsolved problems of plant
biology. Of the four major types polysaccharides of the plant cell
wall, cellulose and callose are synthesized at the plasma membrane,
whereas pectin and hemicellulose are synthesized in the Golgi apparatus.
Cellulose is a polymer of 
-(1
4)-linked Glc residues. Despite its
apparently simple structure, many aspects of cellulose biosynthesis
remain unclear. The site of cellulose synthesis is believed to be the
rosette structures occurring in the plasma membrane. The composition of
the rosette complex changes during development, apparently because
distinct cellulose synthase A (CesA) types are required at different
stages. More than one gene product is required for the rosette to
function properly, and several papers were presented concerning the
structure and function of the presumed components of these rosettes.
Arabidopsis and maize (Zea mays) contain a family of at
least 10 CesA-encoding genes. Although genetic evidence suggests that the CesA proteins function as glycosyltransferases during cellulose synthesis, direct evidence has been lacking. Deborah P. Delmer (University of California, Davis) reported, however, that the herbicide
CGA 325'615 inhibits the synthesis of crystalline cellulose, and
promotes the accumulation of a noncrystalline form in cotton (Gossypium hirsutum) fibers. She reported that
cellulase-mediated digestion of this glucan releases CesA protein,
demonstrating that CesA is intimately associated with chain elongation.
The finding that cellulase treatment also released
sitosterol--glucan (SG) led Delmer to propose that SG might serve as
a primer for chain elongation. The ability of cotton fibers to make SG
from UDP-Glc increases substantially at the onset of secondary cell wall synthesis. Moreover, membranes derived from fibers
pretreated with the cellulose synthesis inhibitor
2,6-dichlorobenzonitrile (DCB) showed a reduced capacity to
synthesize SGs and sterol-cellodextrins, suggesting that DCB might
inhibit cellulose synthesis by inhibiting synthesis of the primer. This
idea was supported by their observation that the inhibition of
cellulose synthesis by DCB in vivo can be partially overcome by
supplying fibers with SG.
Several presentations originating from Simon Turner's laboratory
(University of Manchester, UK) brought the audience up to date on the
cellulose-deficient irx (irregular xylem) mutants of Arabidopsis. The irx mutants exhibit a phenotype in which
the xylem vessels of the plants are so weak that they collapse inwards under the negative pressures associated with the upward transport of
water. To date, four cellulose-deficient irx loci have been identified, three of which are members of the Arabidopsis CesA family.
Progress has been made using immunogold labeling to determine the exact
ultrastructural location of IRX1 and IRX3, both of which are believed
to be CesA isoforms. IRX5 apparently encodes for a member of
the CesA family that is required for secondary cell wall synthesis.
Evidence was presented that two or more distinct CesAs may be required
within a single cellulose synthase complex.
Another interesting Arabidopsis cell wall mutant is
procruste (prc). This mutant exhibits a severe
cell elongation defect in roots and dark-grown hypocotyls. The mutation
affects a gene encoding a cellulose synthase catalytic subunit (CesA6),
resulting in impaired cellulose synthesis. Soizic Rochange (Institut
National de la Recherche Agronomique, Versailles, France) reported on
the results of a screen for suppressor mutations of the
prc1-1 mutation. Three Arabidopsis mutant lines were
identified in which hypocotyl and root elongation was less impaired
than in prc1-1 mutants. The extent of restoration appeared
to be Suc dependent. Cellulose synthesis activities in these new
mutants were intermediate between wild-type and the prc
mutants, suggesting that the suppressor mutations affect a mechanism
directly linked to cellulose synthesis.
Takao Itoh (Kyoto University) reported on his laboratory's success in
demonstrating the existence of terminal complexes containing a
catalytic subunit and/or related proteins of cellulose synthase. The
catalytic region of cotton cellulose synthase was expressed in
Escherichia coli and polyclonal antibodies were produced.
Using an SDS-solubilized freeze-fracture replica labeling technique, the antibodies were specifically localized to plasma membrane complexes
on the P-fracture face of rosette complexes.
Because cellulose composes the world's most abundant renewable biomass
and important industrial products such as cotton fibers and wood, the
metabolic regulation and environmental sensitivity of cellulose
synthesis are important questions. For example, when cotton fibers are
exposed to nighttime cool temperatures (15°C-18°C), their rate of
cellulose synthesis for secondary wall deposition falls to 12% to 25%
of the maximum rate at 28°C to 34°C. In temperate cotton-growing
regions, this often results in harvest of a lower yield, lower quality
fiber crop. Candace Haigler (Texas Technical University, Lubbock) has
been studying the biochemical basis of this problem. Her laboratory's
research suggests that Suc is the preferred substrate for secondary
wall cellulose synthesis, and this implicates Suc phosphate synthase
(SPS) as a possible enzyme involved in the metabolic control and
environmental sensitivity of cellulose synthesis. SPS is a primary
regulator of flux to Suc in leaves. Haigler reported that transgenic
cotton plants that constitutively overexpressed SPS produced higher
quality fibers with thicker secondary walls when the plants were grown under 34°C/5°C day/night cycling temperature. These data suggest that regulation of flux to Suc is both a point of cool temperature sensitivity in cotton fiber cellulose synthesis.
Callose is a general term for the 1,3--glucans that are widely
distributed in higher plants. During normal growth and development, callose is deposited at the cell plate of dividing cells. Callose is
also found associated with some specialized cell walls, and with sieve
plates and plasmodesmata. Callose deposition also occurs between the
plasma membrane and the cell wall in response to various types of
stresses. Although 1,3--glucan synthases have proven to be
recalcitrant to study by conventional biochemistry, Geoffrey V. Fincher
(University of Adelaide, Australia) described recent successes that his
laboratory has had in characterizing these enzymes by alternative
molecular biological techniques. A barley (Hordeum vulgare)
cDNA encoding a homolog of the FKS gene of
Saccharomyces cerevisae was cloned and heterologously
expressed in E. coli. Polyclonal antibodies raised against
fragments of the barley cDNA-encoded protein bound to a 220-kD protein
that shows 1,3--glucan synthase activity in gel.
Hemicelluloses are crosslinked with cellulose microfibrils in the cell
wall of most plant cells, forming a cellulose/hemicellulose framework
that functions as the mechanical underpinning of the cell wall. Natasha
V. Raikhel (Michigan State University, East Lansing) brought the
audience up to date on her and Ken Keegstra's (Michigan State
University) laboratories' collaborative efforts to identify enzymes
involved in the synthesis of xyloglucan, the most common hemicellulose
in plants. Xyloglucan is synthesized in the Golgi apparatus and is
composed of a linear -1,4-glucan backbone with side chains of Xyl,
Gal, Fuc, and Ara residues. A gene encoding for the glycosyltransferase
that adds the terminal Fuc to xyloglucan (AtFT1) has been
previously identified in Arabidopisis, but Raikhel pointed out that
there are no available assays for the remaining enzymes expected to be
involved in the synthesis of xyloglucan. However, their group recently
has developed a biochemical assay to analyze xylosyltransferase
activity. Raikhel outlined a bioinformatics-based approach that has led
toward the identification of a putative alpha-xylosyltransferase gene
family that is involved in xyloglucan biosynthesis. By screening mRNA
populations from cotton fibers at different developmental stages, the
group has identified several good candidates. Two of these genes are
predicted to be Golgi type II membrane proteins and glycosyltransferases.
Novel approaches toward the study of hemicellulose biosynthesis were
also the topic of Laurence Bindschedler's (Royal Holloway University
of London) talk. His laboratory has established a cell suspension
culture of tobacco (Nicotiana tabacum) transformed with the
Agrobacterium tumefaciens Tcyt gene that leads to high endogenous levels of cytokinin. In response to optimal concentrations of auxin and Suc, this cell line shows increased cell wall aggregation, elongated cells, and a 5-fold increase in wall thickness. Recovery of
wall material was 50% greater in the Tcyt culture. Several enzymes (UDP-Glu dehydrogenase, xylan synthase, and an 80-kD soluble decarboxylase) involved in hemicellulose synthesis in this system were
also described.
UDP-glucoronic acid, which provides about half of the biomass of the
cell walls of Arabidopsis, is the dominant nucleotide sugar for
hemicelluloses and pectin. Raimund Tenhaken (University of
Kaiserlautern, Germany) reported on the occurrence of three newly
discovered isoforms of UDP-Glu dehydrogenase in Arabidopsis. Using
reporter genes, they found a developmentally regulated expression of
these isoforms that suggests that the genes are expressed in cells only
when UDP-Glc-derived precursors are needed for the synthesis of new
matrix polymers. In young seedlings, an alternative pathway operates in
which inositol is oxidized directly to Glc. They also showed, contrary
to previous reports in the literature, that UDP-Glc is the only
substrate accepted by the enzyme. Moreover, they discovered that
UDP-Glc dehydrogenase is strongly inhibited by UDP-Xyl, indicating a
feedback loop for the regulation of sugar nucleotide precursors.
Antibodies provide highly specific tools with which to study the
dynamics of plant cell walls. Michael G. Hahn's (University of
Georgia, Athens) laboratory has generated several monoclonal antibodies
that recognize specific carbohydrate structures (epitopes) in plant
cell wall polysaccharides. These antibodies were used to study the
distribution of selected carbohydrate structures found in pectic and
hemicellulosic polysaccharides in tissues of wild-type and mutant
Arabidopsis plants. Their studies demonstrate that polysaccharide
structures differ among walls of different cell types, and differ even
among walls of a single cell. They also showed that the insertion of
specific polysaccharide structures into plant cell walls is under tight
developmental control. Furthermore, they provided evidence that plant
cells and tissues utilize different sets of enzymes for the
biosynthesis of wall components at different points in growth and development.
Galactomannans are polysaccharides with a
(14)--D-mannan backbone that is variably
(16)-
-D-galactosyl-substituted according to the
species. Galactomannans compose most of the endosperm cell walls of
many seeds where they serve as post-germinative food reserves. Those
from legume seeds are especially important in the water relations of
seed germination, and are widely exploited in the dairy industry. Golgi
membrane-bound galactomannan galactosyltransferases (GMGTs) are
important in the regulation of Gal substitution in the pathway of
galactomannan biosynthesis. Grant Reid (University of Sterling, UK)
reported on his laboratory's characterization of a GMGT from fenugreek
(Trigonella foenum-graecum). This enzyme and related ones
were overexpressed in Pichia pastoris and shown to have GMGT
activity. The ability to transgenically express GMGTs of different
types has allowed for the production of novel galactomannan varieties
that may have commercial application.
Man is a cell wall constituent found in a wide variety of
non-cellulosic polysaccharides, several of which are of agronomic importance. In Arabidopsis, Man constitutes 4% to 9% of the neutral sugars present in the leaf tissue, but the biological functions of
these Man-containing polysacchandes are as yet unknown. Non-cellulosic polysacchandes, including mannans, are synthesized in the Golgi apparatus. Synthesis occurs by the action of glycosyltransferase enzymes that use nucleotide sugars as substrates. Mannosyltransferase requires Man to be supplied as GDP-Man. The synthesis of GDP-Man and
other sugar nucleotides occurs in the cytosol. However, the catalytic
site of mannosyltransferase is likely to be luminal. Therefore, a
GDP-Man sugar nucleotide transporter could be required in the Golgi
membrane. Transporters of various nucleotide sugars are known to reside
in the Golgi membrane in animal and yeast cells, and biochemical
evidence has supported their presence in plants, but no plant
transporter has been molecularly characterized until now. Timothy
Baldwin (University of Cambridge, UK) reported upon the identification
of a gene encoding a putative Arabidopsis GDP-Man transporter
(GONST1 for Golgi nucleotide sugar transporter) that
shows extensive homology to Vrg4, a GDP-Man transporter
from S. cerevisiae and to Lpg2, a
GDP-Man transporter from Leishmania sp. His laboratory
has shown by transient gene expression in onion (Allium
cepa) epidermal cells that the GONST1 gene product is localizedin the Golgi apparatus. In addition, a GONST1
promoter::GUS reporter gene construct showed that the
gene is essentially ubiquitously expressed. GONST1 can
complement vrg4
yeast, indicating that
its product can transport GDP-Man in vivo. Moreover, in vitro transport
assays of isolated yeast membranes containing GONST1 have increased
transport of GDP-Man. Their demonstration that GONST1 is a
Golgi-localized GDP-Man transporter supports the model that such
transporters are required for Golgi glycosylation reactions, and open
up the possibility of performing experiments to alter cell wall
synthesis by altering transport of sugar nucleotides across the Golgi membrane.
Pectins, such as homogalacturonan and rhamnogalacturonans (RGs) I and
II, are major components of primary cell walls. The pectic
polysaccharides are structurally the most complex of the matrix
polymers of plant cell walls and, as a consequence, the functions of
these polymers are far from clear. Insights into the molecular biology
and biochemistry of pectin have been slow to develop. No pectin
biosynthetic enzyme has been purified to homogeneity and none of the
enzymes involved in pectin biosynthesis have been cloned and sequenced.
Homogalacturonan is involved in a range of cell wall activities that
influence cell adhesion, cell expansion, wall porosity and defense. An
important aspect of this modification is the modulation of the degree
and pattern of methylesterification of homogalacturonan domains. The
degree and pattern of methylesterification, for example, can influence
such properties as calcium binding, gelling properties, and
susceptibility to enzymatic cleavage during development and pathogenesis. The extent and pattern of methylesterification are regulated primarily by the action of wall-based pectin methylesterases that remove specific methylesters in diverse patterns. William G.T.
Willats (University of Leeds, UK) reported on his laboratory's use of
very well-characterized antibody probes, specific for certain methylester distribution patterns, to study the occurrence of populations of homogalacturonan types within cell wall microdomains. Grégory Mouille (Institut National de la Recherche Agronomique) also
employed antipectin antibodies in a whole-mount staining procedure of
Arabidopsis roots. The anti-1,4--galactan antibody LM5 showed an
easily distinguishable staining pattern. After screening the roots of
20,000 TDNA insertion lines for altered staining patterns, 16 pectin
mutant lines were isolated.
Keiko Sugimoto (John Innes Centre, Norwich, UK) reported on his
exploitation of two Arabidopsis mutants, hyp6 and
e31.10, to explore the role of pectin polysaccharides in
plant growth and development. Both mutants were initially identified in
a screening for short hypocotyls from ethyl
methanesulfonate-mutagenized populations. Further screening with FTIR
microspectroscopy showed that both hyp6 and
e3l.10 are altered in pectin compositions. From gas
chromatography-mass spectrometry analysis of cell wall neutral sugars,
they found that hyp6 and e31.10 are deficient in
Man content and rhamnose content, respectively. Both hyp6
and e3l1.10 exhibit various growth defects, including
strongly reduced hypocotyl elongation, immature trichome development,
and suppressed root hair formation. This suggests that HYP6 and E31.10
are essential for both diffuse and tip growth. Scanning electron
microscopic observations revealed that the hypocotyl epidermal cells of
both mutants have severely decreased growth anisotropy and some cells
undergo abnormal radial swelling. Transverse sections of hypocotyls
showed these mutants have incomplete cell walls, indicating that HYP6
and E31.10 are required for proper cell wall formation.
RG-I is an abundant pectic polymer that is highly variable in structure
and occurrence. J. Paul Knox (University of Leeds) made monoclonal
antibodies to arabinan (LM6)1 and galactan (LM5)2 epitopes that are
common components of side chains attached to rhamnose residues of RG-1.
These two antibodies promise to provide important insights into the
occurrence and function of these RGI-related structures in plant cell walls.
RG-II is only a minor component of the plant cell wall, but has a major
effect on cell wall porosity. Borate esters form cross-links to RG-II
molecules, generating a macromolecular pectic network. The formation of
these cross-links results in a pectin structure with a reduced size
exclusion limit. Low pH (1.5-3.0) buffers and neutral solutions of
calcium chelators are known to convert the borate ester cross-linked
RG-II dimer (dRG-II-B) to the RG-II monomer. These treatments also
solubilize wall-bound pectin and cause an increase in the size
exclusion limit of pectin-rich cell walls. The low pH and
chelator-mediated effects are prevented or diminished by the addition
of boric acid (1 or 10 mM). Axel Fleischer (Humboldt
University, Berlin) reported that the B-dependent increase in size
exclusion limit caused by acid buffers is tissue dependent and may
depend on the physiological state of the cells. For example, the walls
of rapidly growing suspension-cultured Chenopodium album
cells are more sensitive to the RG-II-dissociating treatments than
cells in the stationary phase. This suggests that pectin in the walls
of stationary cells contain additional (secondary?) cross-links that
render the pectin network less sensitive to hydrolysis of dRG-11-B.
Walls isolated from a cell culture of the
L-Fuc-deficient mur-1 mutant of
Arabidopsis were found to be more sensitive to low pH treatment than
the walls of wild-type cells. Walls of the mur-1 mutant and
WT cells differed in the pH dependence of boric acid release, pectin
solubilization, and size exclusion limit increase. These differences in
wall properties are consistent with the observation that
mur-1 dRG-II-B, which contains L-Gal rather than L-Fuc residues, is less stable in
vitro at low pH than wild-type dRG-II-B.
Lignins are complex phenolic polymers of plant cell walls that are
linked to wood fibers. They provide hydrophobicity to xylem vessels,
mechanical support to plant stems, and aid in plant defense against
microorganisms. Although lignins are essential for normal plant growth
and development, they are commercially undesirable because they
decrease forage digestibility, and must be removed from cellulose by chemical treatments during pulping. Almost all of
the genes of the lignin biosynthesis pathway have now been cloned and
genetically manipulated. Lignification has three steps: (a)
biosynthesis of the monolignols (e.g.
p-hydroxycinnamyl, coniferyl, and sinapyl alcohols),
(b) transport and secretion of monolignols, and (c) polymerization of
the monolignols by dehydrogenation. Several contributions concerned
various transgenic strategies for altering the
lignification process. Zara Harrison (Institute of Food Research,
Norwich, UK) reported on her group's transgenic expression in tobacco
of a bacterial gene encoding the enzyme 4-hydroxycinnamoyl-coenzyme A
(CoA) hydratase/lyase, which has been shown to convert
p-coumaryl CoA, caffeoyl CoA, and feruloyl CoA to the
corresponding benzaldehydes. Plants expressing the HCHL gene
exhibited a severely altered phenotype, including some features
(stunted growth, interveinal chlorosis, curling of leaf margins, and
red/orange coloration) characteristic of plants with altered
phenylpropanoid metabolism. A reduction in cell wall-bound phenolics
was shown to occur in the leaves and lignified tissues, and levels of
intracellular soluble phenolics, such as chlorogenic acid and rutin,
were reduced by 90% or more.
| |
CELL WALL PROTEINS |
Although the cell wall is largely composed of carbohydrate,
structural proteins also form a network that contributes to cell wall
architecture and to plant resistance to pathogen attack. Enzymes that
modify cell wall structure are also important in regulating growth and
development and response to the environment.
Pro-rich cell wall proteins (PRPs) are widely distributed in plants and
are believed to function by modeling the architecture of the primary
cell wall-surrounding specific cell types. In legumes, a subset of PRPs
is specifically expressed during symbiotic root nodule formation. Jim
B. Cooper (University of California, Santa Barbara) reported that PRPs
are deposited as a sheet surrounding the intercellular junctions
regions of cortical and vascular parenchyma tissues, and in the pit
regions of tracheary elements. The pattern of PRP deposition in xylem
walls is a mirror image of the pattern of lignin deposition. The
inhibition of phenylpropanoid biosynthesis eliminates the normal
pattern of PRP deposition, indicating that lignin deposition is
required for the spatial restriction of PRPs in xylem cell walls.
Two classes of multiple-domain PRPs have been identified in Arabidopsis
based on their primary sequence and domain organization. The expression
of one class is limited to guard cells and the early stages of lateral
root development. A second class of ATPRPs (ATPRP1 and ATPRP3) has been
localized to sites of root hair initiation. Mary L. Tierney (University
of Vermont, Burlington) described experiments in which polyclonal
antibodies raised against the PRP domain of ATPRP3 were used to show
that ATPRP1 and ATPRP3 are localized to the cell wall during seedling
growth. Immunohistochemical analysis of ATPRP3, using an epitope-tagged
version of the protein expressed in transgenic plants, showed that
AtPRP3 is localized to the base of the root hair and the growing tip.
This pattern indicates a role for ATPRP3 in tailoring the structure of
the newly formed root hair cell wall. To further characterize AtPRP3's function in root hair growth, Tierney's group generated
AtPRP3 antisense lines. They have also identified an
AtPRP3/T-DNA insertion mutant. Seedlings homozygous for the
T-DNA knockout mutation show a root hair branching phenotype,
supporting a functional role for AtPRP3 in the determination and/or
maintenance of root hair structure. Segregation of root hair branching
is 1:1 (wild type:mutant) in F2 seedlings
generated from a backcross, suggesting that a single functional copy of
AtPRP3 may not be sufficient for proper root hair growth.
Arabinogalactan proteins (AGPs) are proteoglycans of the plant cell
extracellular matrix that have been implicated in a variety of
processes in plant growth and development, but their precise functions
are only now beginning to emerge. Most AGPs are anchored to the plasma
membrane by a glycosylphosphatidyl-inositol (GPI) anchor. Allan
Showalter (Ohio University, Athens) found that LeAGP-1, a GPI-anchored
AGP from tomato (Lycopersicon esculentum) is
immunolocalized to the surface of protoplasts and detectable in plasma
membrane preparations. LeAGP-1 was immunolocalized to developing
metaxylem of stem and petiole, the cell walls and intercellular spaces
of stylar-transmitting cells, and pollen tube tips. LeAGP-1 was also associated with cell wall thickening and lignification of particular cell types, specifically in secondary cell wall thickenings of maturing
metaxylem and secondary tracheary elements in roots and stems, and in
thickened walls of phloem sieve tube elements.
Arjon van Hengel (John Innes Centre) reported that one AGP (AtAGP30) is
not anchored to the plasma membrane. Analyses of the expression pattern
revealed that the gene transcript is only detectable in the root tip of
the seedling, and that stress conditions and hormone applications alter
AtAGP30 expression. The overexpression of AtAGP30
leads to normal root development, but poor shoot development. A
knockout mutant revealed a decrease in sensitivity to Yariv reagent.
Analysis of cell walls revealed that walls from Atagp30 mutants are chemically distinct from wild types, and that mutant root
tips contain lower amounts of arabinogalactan and protein. It is
interesting that Atagp30 mutants germinated faster on medium containing abscisic acid than did wild type.
Carolyn Schultz (University of Melbourne) identified four different
classes of AGP genes. One of these classes encodes AG peptides with protein backbones of between 10 and 13 amino acid residues. Most of these AG peptides, with the exception of AtAGP16, are
predicted to be anchored to the plasma membrane by a GPI anchor. To
determine whether AtAGP16 is processed before the addition of a GPI
anchor, Schultz's laboratory purified AGPs from Arabidopsis tissues
for deglycosylation. Protein backbones were then analyzed by matrix
assisted-laser disorption ionizing-time of flight mass spectrometry
and/or sequencing by Edman degradation. These experiments confirmed the
posttranslational processing of the AG peptides and helped determine
the cleavage site (
) residue for the addition of the GPI anchor for
each AG peptide. Gene-specific probes for several of the AG-peptide
genes are being used in in situ hybridization experiments to determine
if genes are expressed in multiple cell types within a tissue or
whether they are restricted to a single cell type. Her laboratory is
also studying the expression of the genes using the publicly available
Arabidopsis micorarray data. This has enabled them to investigate the
expression of these genes in the greater than 100 different
developmental, mutant, and environmental conditions tested so far.
Kim Johnson (University of Melbourne) reported on a new class of
Arabidopsis proteins that have -Ig-H3/fasciclin-like domains in
addition to AGP-like domains. Similar to their animal counterparts, the
fasciclin-like AGPs of plants may act as cell adhesion molecules and
function in cell-to-cell communication.
Azeddine Driouch (Université de Rouen, France) provided evidence that
alterations in AGP may underlie the altered phenotype of the
reb1-1 (root epidermal bulger) mutant of Arabidopsis. This mutant is characterized by reduced root elongation rates and by the
bulging of trichoblasts. It is interesting that a similar phenotype can
be obtained by growing Arabidopsis seedlings in the presence of
(-D-Glc3) Yariv reagent, which is known to
bind to AGPs. In addition, reb1-1 mutants contain only 30%
as much AGPs as do the wild type. In the root elongation zone,
antibodies against AGP epitopes scarcely stain the trichoblasts of the
mutant, but strongly stain the trichoblasts of wild-type plants.
Moreover, the cell walls of the trichoblasts of the mutants are only
one-half as thick as the cell walls of the wild-type. Georg Seifert
(John Innes Centre) reported that reb1 is identical to a
previously described rhd1 mutant, and determined that it is
a member of the UDP-Glc 4-epimerase family. He presented evidence that
RHD1/REB1 may act in planta in the biosynthesis of
D-Gal in roots. This suggests the involvement of
cell wall galactans and arabinogalactans in the control of cell volume
and cell elongation.
Clare Steele-King (University of Leeds) identified a Yariv
super-sensitive (yss) mutant that has a pH-dependent
phenotype. When yss mutants are grown in the presence of
Yariv reagent, their inhibition of growth is even more dramatic than
that seen in wild-type seedlings exposed to Yariv reagent.
Immunochemical studies indicate that yss has an altered AGP
composition. Yariv reagent also increases the staining of both
yss and wild-type seedlings by a monoclonal antibody that
recognizes pectin galactan, suggesting an interaction between pectin
and AGP.
Yolanda Gaspar (University of Melbourne) reported on her identification
of a number of Arabidopsis mutants with T-DNA insertions in or around
the coding region of the classical AGP proteins backbone using a PCR
based approach. One of these mutants was identified independently based
on its resistance to A. tumefaciens rat1 (resistant to
A. tumefaciens transformation).
Hyp-rich glycoproteins, such as extensin, are small peptide repeats
that are extensively O-glycosylated at their Hyp residues. The Hyp contiguity hypothesis is based on correlations between Hyp
oligoarabinosylation and blocks of contiguous Hyp residues, and
predicts Hyp galactosylation and subsequent arabinogalactan polysaccharide addition to clustered noncontiguous Hyp. One of the
highlights of the meeting was Marcia Kieliszewski's (Ohio University)
presentation in which she reported results that confirmed that amino
acid sequences can be used to predict the glycosylation profiles of
cell wall proteins such as AGPs and extensins. She described how using
a series of simple Hyp-rich glycoproteins, her laboratory was able to
identify the first O-glycosylation code of plants and
provided several examples illustrating its use for predicting
glycoprotein structure from a gene.
Expansins are a large multigenic class of proteins that are thought to
aid in cell wall extension during growth by disrupting hydrogen bonds
at the interfaces between individual cell wall polymers. Previous
studies have implicated expansins in diverse processes, including cell
growth and polarity, root hair formation and growth, organogenesis and
fruit ripening. Research from Hans G. Edelmann's laboratory
(University of Bonn) provided new insights into the subcellular and
tissue-specific localization of expansins in maize. The authors noted
great diversity in the labeling patterns depending upon the type of
expansin antibody used. One type labeled cell plates in the meristem
and pitfields in the elongation region, whereas another recognized
longitudinal cell walls throughout growing root apices and tips of
young root hairs. A third type produced strong cytoplasmic labeling
that accumulates predominantly with the bulging areas associated with
emergent root hairs. These findings indicate that different expansins
are found at unique subcellular locations, suggesting that they may
have different physiological functions.
Endoxyloglucan transferases (XETs) are a class of enzymes that cut and
rejoin xyloglucan chains. For expansion to occur in plant cells,
cellulose microfibrils need to move past one another. It is believed
that wall loosening results when xyloglucan, the major hemicellulose in
the cell wall and capable of tethering adjacent microfibrils, is
modified. During this process, XETs play a central role in the
construction and modification of cell wall architecture. XETs are
encoded by a large multigene family termed xyloglucan-related proteins
and are classified into three or four subfamilies with respect to their
deduced amino acid sequences. The completion of the Arabidopsis genome
project allowed Kazuhiko Nishitani (Tohoku University, Japan) and his
coworkers to characterize this family in detail. They identified 33 open reading frames that putatively encode for XETs. Each member of
this family exhibits a distinct expression profile in terms of both
organ specificity and responsiveness to hormonal signals. In situ
hybridization and promoter::GFP gene constructs of select
members of these 33 open reading frames revealed strict and specific
cell-type expression patterns within respective organs. These facts
imply that each member of this gene family is individually committed to
a certain specific process that is regulated by different hormones in a specific tissue at a specific point in development.
Ryusuke Yokoyama (Tohoku University) presented data concerning the
secretion pathways of XET in the tobacco Bright-Yellow 2 cell system.
In the cell enlargement process, which occurs mainly during the
interphase of the cell cycle, the protein was extensively secreted into
the apoplast via the ER-Golgi apparatus network. However, as the cell
progressed to mitosis, the apoplast-directed transport was diminished,
and the protein was then exclusively located in the phragmoplast and
eventually transported to the cell plate. These results indicate a role
for XET in both the construction of the cell plate and the cell wall,
and also suggest that the existence of a cell-cycle dependent switching
mechanism that precisely controls the direction of the delivery of
secretory vesicles containing the same protein.
Kris Vissenberg (University of Antwerp, Belgium) developed a
fluorescent technique where active XET incorporates fluorescent substrates into the cell wall. In Arabidopsis roots, high XET activity
was confined to the elongation zone and is up-regulated at the site of
future root hair emergence. Closer examination revealed that the
fluorescence in the cell wall exhibited a fibrillar pattern,
reminiscent of the pattern of cortical microtubules and of cellulose
microfibrils in the cell wall. Interference with cellulose deposition
greatly reduced the fluorescence and the fibrillar pattern, whereas
application of oryzalin resulted in the loss of the parallellism in the
fibrillar pattern. Latrunculin B on the other hand decreased the
fluorescence, but failed to abolish the fibrillar pattern. From these
results, the authors concluded that XET probably has different roles in
the cell wall. Besides a potential role in cell wall loosening, XETs
are important in the buildup and restructuring of xyloglucans in the
cell wall. It is most likely that expansin is the primary cell
wall-loosening factor, but needs the concerted action of active XETs
and endoglucanases to accomplish its task.
| |
MECHANISMS OF CELL GROWTH |
Both Peter Schopfer (University of Freiburg, Germany) and Stephen
C. Fry (University of Edinburgh, UK) presented papers relating to the
role of hydroxyl radicals in cell wall loosening and cell growth.
Wall-localized reactions may lead to the production in the apoplast of
(·OH) hydroxyl radicals, a highly reactive species that
can cause nonenzymic scission of polysaccharides. Such scission may
loosen the wall and thus promote cell expansion or fruit softening. To provide a method for detecting the proposed action of endogenous ·OH radicals in vivo, Fry's laboratory devised a method
for "fingerprinting" the products formed when plant cell wall
polysaccharides are attacked by ·OH. Treatment of wall
polysaccharides with ·OH radicals prompted oxidative
scission and simultaneously increased the number of
NaB3H4-reacting groups
present. The latter are proposed to be glycosulose and
glycosulosuronate residues that are
NaB3H4-reducible back to
the original sugar residue and to their more unusual epimers.
Driselase- or acid-catalyzed hydrolysis of the ·OH-treated polymers gave characteristic
"fingerprints" of radiolabeled products seen after
chromatography and electrophoresis. Products from tamarind
xyloglucan included [3H]isoprimeverose,
[3H]Gal, [3H]Glc,
[3H]Ara, [3H]Rib, and
[3H]Man. Those from citrus pectin
included[3H]Gal,
[3H]GalA, and at least seven unidentified
acidic products, probably including those derived from
[3H]TalA and [3H]GulA.
The patterns of 3H-products are useful
fingerprints by which polysaccharides that have been ·OH
attacked during wall loosening in vivo may be recognized. Applied to
maize coleoptile cell walls, the method showed no evidence for
·OH radical attack on polysaccharides during
auxin-induced wall loosening. However, applied to the walls of ripening
pear (Pyrus communis) fruit, the method demonstrated a
steady increase in NaB3H4-reducible groups
during the softening process. In contrast to Fry's findings, Schopfer
presented evidence that oxygen radical chemistry may also play a role
in polymer breakdown during auxin-induced extension growth. Backbone
cleavage of cell wall polysacchandes can be accomplished in vitro by
·OH produced from hydrogen peroxide in a Fenton reaction
or in a reaction catalyzed by peroxidase supplied with
O2 and NADH. They presented evidence that
coleoptile growth of maize seedlings is accompanied by the release of
reactive oxygen intermediates in the cell wall. Auxin promotes release
of O·2 and subsequent
generation of ·OH when inducing elongation growth.
Experimental generation of ·OH in the wall causes an
increase in wall extensibility in vitro and replaces auxin in inducing
growth. Auxin-induced growth can be inhibited by scavengers of
O·2, hydrogen peroxide,
or ·OH, or inhibitors interfering with the formation of
these molecules in the cell wall. These results support the idea that
·OH produced by peroxidase acts as a wall-loosening agent
during auxin-induced growth.
| |
METHODS AND MODELS: HIGH-VOLTAGE ELECTRON TOMOGRAPHY |
Until recently, the analysis of cellular structures at the
electron microscopical level was limited by both specimen preparation methods and by image analysis problems. For example, when cells are
preserved by chemical fixatives, the preservation of different types of
cellular structures in their native state is limited both by the slow
rate of cross-linking reactions and by the selective nature of the
cross-links that are formed. On the other hand, the deciphering of
three-dimensional cellular structures is limited by the thickness of
the serial thin sections used for the reconstructions. L. Andrew
Staehelin's (University of Colorado, Boulder) research laboratory has
overcome both of these limitations by combining cryofixation/freeze-substitution methods in conjunction with dual-axis high-voltage electron tomography of serial thick sections (0.25-0.4 pm). This new methodology produces tomographic slices that look like
electron micrographs but are only 2.3 nm thick versus 60 to 80 nm for
normal thin sections. Overall, the three-dimensional resolution in
these reconstructed specimens is about 6 nm, which enables them to see
and identify large molecules such as clathrin triskelions, dynamin
spirals, and kinesin motor proteins within the thick sections. By
tracing the outlines of cellular structures in the individual 2.3-nm
slices they can also produce high resolution, three-dimensional models
of membrane compartments and cytoskeletal systems.
NMR studies, mostly from the laboratory of Mike Jarvis (Glasgow
University, UK), are providing new details of the submolecular organization of cell walls. For example, at least four polymorphs of
cellulose exist, namely cellulose I to IV. The native form, cellulose
I, and the mercerized or regenerated form, cellulose II, are the two
most common polymorphs, cellulose II being the most stable crystalline
form. Native cellulose contains two allomorphs, I and I, the
proportions of which are dependent on the organism in which it is
synthesized. There is evidence that the cellulose microfibrils of
higher plants have a crystalline core of cellulose I, I or both.
Microfibrils from the textile celluloses cotton, flax (Linum
usitatissimum) and ramie (Boehmeria nivea) are larger than those from other higher plants, and contain only 20% to 30% of
surface chains rather than the normal figure of 60% to 70%. Taking
linear combinations of the solid-state 13C NMR
spectra of flax and celery (Apium graveolens) cellulose, Jarvis and his coworkers extracted the complete
13C spectrum of the surface chains, which
differed considerably from the spectra of the I and l forms. They
also extracted the spectrum of the surface chains in an independent way
by 13C spin relaxation editing, taking advantage
of their greater mobility than the chains in the crystalline core of
the microfibrils. A good correspondence was observed between the
spectra obtained by these two approaches. They showed that the surface
chains had a mixture of two conformations at C-6, both different from
the trans-gauche conformation found in crystalline cellulose
I and I. Neither of the C-6 conformations in the surface chains
permits the formation of the intramolecular hydrogen bond from O-2
that stabilizes the chain conformation in crystalline celluloses. This implies that H bonding to other polysaccharides or water is possible for both O6-H and O2-H in the surface chains. The spin relaxation editing experiments also provided information on the degree to which
each C atom in the surface chains was freed from the rigid network of H
bonds that hold the core chains together. More direct information on H
bonding was obtained through an
1H-13C heteronuclear
correlation experiment. This allowed the 1H NMR
spectrum of the hydroxyl protons in cellulose to be extracted. Because
the 1H spectra of hydroxyl protons depend
principally on the strength of H bonding, this experiment provides a
novel way to elucidate H bonding patterns in the solid state.
FTIR spectroscopy is becoming more commonplace as a means to identify
mutations that affect plant cell wall polysaccharide components and
wall architecture. Nick Carpita's laboratory (Purdue University, West
Lafayette, IN), for example, has identified a broad range of genes
involved in the biogenesis and dynamic alteration of plant cell wall
architecture during growth and development. Among the mutations his
group has been able to identify so far are detects in formation of
nucleotide-sugar substrates, in polysaccharide synthases that make
polymer backbones, in glycosyl-transferases that add side groups, in
secretion and targeting, in cytoskeletal orientation that directs wall
architecture, in assembly of wall polysaccharide components, in wall
dynamics during growth, and in wall disassembly and recycling of
hydrolyzed monosacchandes. The FTIR "forward" screen can be used to
detect specific alterations in wall structure and architecture caused
by chemical mutagens or by the insertion of DNA "tags," regardless
of the genetic basis of the alteration. In this respect, the screen is
powerfully selective at the end-product level. They are also using
reverse genetic approaches to identify cell wall biogenesis-related
genes in Arabidopsis and maize. It is anticipated that this approach
will provide a broad set of cell wall mutants that will serve to
further the development of libraries of FTIR "spectrotypes" that
will be useful for determining gene function. They have devised a
systematic protocol, which employs biochemical, spectroscopic, and
imaging methods, to categorize defects in wall structure and
architecture into one of six stages of wall biogenesis or disassembly.
A major practical goal is to generate plants with genetically defined variation in composition and architecture to permit assessment of
modifications on wall properties and plant development. Several of
these genes, as well as several of the plants with genetically defined
alterations, may be of economic importance. Examples include the
modification of pectin-cross-linking or cell-cell adhesion to increase
shelf life of fruits and vegetables, the enhancement of dietary fiber
contents of cereals, the improvement of yield and quality of fibers,
and the relative allocation of carbon to wall biomass for use as biofuels.
FTIR spectroscopy was also used by Mike Jarvis (Glasgow University) to
obtain ordinary and polarized spectra in the hydroxyl-stretching region
of flax cellulose I and cellulose II samples. At the same time, spectra
obtained on surface-deuterated samples in the polarized IR beam enabled
them to distinguish the behavior of internal chains from the behavior
of surface chains. It is possible, by comparison of the experimental
spectra with modeled FTIR spectra, to gain information about the
distribution of orientations of the intrachain and interchain hydrogen
bonds. This information is relevant to the H-bonding arrangements that
attach surface cellulose to interior cellulose.
The mechanical properties of cell walls are relevant for understanding
how cells and tissues grow and function. Their stiffness properties
generally include elastic and viscous components, and are non-linear
with strain. David M. Bruce and his coworkers (Silsoe Research
Institute, Bedford, UK) have taken a modeling approach toward
understanding the mechanics of plant tissue and cell walls, based on
describing the wall as a structure rather than a material. Potato tuber
tissue was employed as a model system. These fluid-filled cells with
high adhesion, have been shown to be incompressible up to 22% tissue
strain over time scales of up to 15 s. Given an imposed
deformation in one direction, the constancy of volume of each cell
allows its dimensional changes in the other directions to be
calculated. Within the tissue, the planar wall faces have been found to
remain planar during deformation, but their shape has been observed to
change as a function of the orientation of the walls relative to
applied strain. Therefore, within the plane of a wall, the strain must
be a function of direction. Key assumptions of their model are that
cellulose microfibrils, the tensile elements in the wall, resist only
tension and offer insignificant resistance to re-orientation or
shortening. As a result of deformation imposed on the tissue,
microfibrils, whatever their initial orientation in the wall, will
become reoriented toward alignment with wall tension. Their results
suggest that microfibrils initially aligned with wall tension fail at a
calculated strain of some 8%. This allows wall extension that brings
other microfibrils more into alignment, and hence into play to resist
wall extension. Using measured values of cell wall size and shape, and
microfibril dimensions from electron micrographs, they deduced the
shape of a stress-strain curve for a microfibril up to incipient
failure of the tissue. Microfibril modulus (a maximum of 130 MPa) and
strength (7.5 GPa) align well with values for cellulose in the literature.
Laurence Melton (University of Auckland, New Zealand) discussed the
value of celery cell walls as a model for cell wall architecture. The
parenchyma cell walls of celery were investigated using solid-state 13C NMR, x-ray diffraction, and atomic force
microscopy. Analyses indicated that the walls consist mainly of
cellulose (42 mol %) and pectic polysaccharides (51 mol %), and
contain a remarkably low level of xyloglucan (2 mol %) and xylan (2 mol %); hence, this is a good model system for studying the physical
relations of cellulose microfibrils and pectic polysaccharicles.
Solid-state 13C NMR results indicate the walls
exist in three domains: rigid (cellulose microfibrils), semirigid
(galacturonan), and mobile (arabinan and galactan). There was no
evidence for non-cellulosic polysaccharides adhering to the surface of
the cellulose microfibrils. Their results support a cell wall model in
which cross-linking of cellulose microfibrils by individual
non-cellulosic polysaccharides, such as xyloglucan, is neither expected
nor necessary.
TOP
INTRODUCTION
SYNTHESIS OF CELL WALL...
