Molecular Plant • Volume 2 • Number 5 • Pages 990–999 • September 2009 RESEARCH ARTICLE

Pectin May Hinder the Unfolding of XyloglucanChains during Cell Deformation: Implications ofthe Mechanical Performance of ArabidopsisHypocotyls with Pectin Alterations

Willie Abasoloa,b, Michaela Edera, Kazuchika Yamauchia,c, Nicolai Obeld, Antje Reineckea,Lutz Neumetzlerd, John W.C. Dunlopa, Gregory Mouillee, Markus Paulyd,f, Herman Hoftee andIngo Burgerta,1

a Max-Planck-Institute of Colloids and Interfaces, Department of Biomaterials, Potsdam, Germanyb College of Forestry and Natural Resources, University of the Philippines Los Banos, Philippinesc Department of Socio-Environmental Energy Science, Kyoto University, Japand Max-Planck-Institute for Molecular Plant Physiology, Potsdam, Germanye Laboratoire de Biologie Cellulaire, UR501, Institute Jean-Pierre Bourgin, INRA, Versailles, Francef Michigan State University, Plant Research Laboratory, East Lansing, Michigan, USA

ABSTRACT Plant cell walls, like a multitude of other biological materials, are natural fiber-reinforced composite materials.

Their mechanical properties are highly dependent on the interplay of the stiff fibrous phase and the soft matrix phase and

on the matrix deformation itself. Using specificArabidopsis thaliana mutants, we studied the mechanical role of the matrix

assembly in primary cell walls of hypocotyls with altered xyloglucan and pectin composition. Standard microtensile tests

and cyclic loading protocols were performed on mur1 hypocotyls with affected RGII borate diester cross-links and a hin-

dered xyloglucan fucosylation as well asqua2 exhibiting 50% less homogalacturonan in comparison to wild-type. As a con-

trol, wild-type plants (Col-0) and mur2 exhibiting a specific xyloglucan fucosylation and no differences in the pectin

network were utilized. In the standard tensile tests, the ultimate stress levels (;tensile strength) of the hypocotyls of

the mutants with pectin alterations (mur1, qua2) were rather unaffected, whereas their tensile stiffness was noticeably

reduced in comparison to Col-0. The cyclic loading tests indicated a stiffening of all hypocotyls after the first cycle and

a plastic deformation during the first straining, the degree of which, however, was much higher for mur1 and qua2 hypo-

cotyls. Based on the mechanical data and current cell wall models, it is assumed that folded xyloglucan chains between

cellulose fibrils may tend to unfold during straining of the hypocotyls. This response is probably hindered by geometrical

constraints due to pectin rigidity.

Key words: Arabidopsis thaliana; mutants; cellulose; xyloglucan; pectin; cyclic loading tests.

INTRODUCTION

The primary cell wall of plants is a unique engineering struc-

ture that combines conflicting characteristics such as rigidity as

well as plasticity and compliance (Rose and Bennett, 1999;

Whitney et al., 1999; Cosgrove, 2000). Rigidity is needed to

withstand the osmotic pressure of the living cell (Taiz, 1984)

and to cope with external loads, whereas sufficient plasticity

and compliance are needed for cell wall expansion during

growth (Baskin, 2005). Furthermore, the primary wall is specif-

ically designed to provide a rigid barrier against pathogenic

intrusions (Creelman and Mullet, 1997) and, at the same time,

performs dynamic tasks in absorption, transport, and secretion

of substances throughout plant growth and development

(Eckardt, 2003).

In dicotyledonous plants, the primary wall consists of

approximately 30% cellulose, 30% hemicelluloses, 35% pectin,

1 To whom correspondence should be addressed. E-mail ingo.burgert@

mpikg.mpg.de, fax +49 331 567 9402.

ª The Author 2009. Published by the Molecular Plant Shanghai Editorial

Office in association with Oxford University Press on behalf of CSPP and

IPPE, SIBS, CAS.

doi: 10.1093/mp/ssp065, Advance Access publication 4 September 2009

Received 5 May 2009; accepted 14 July 2009

by guest on June 3, 2013http://m

plant.oxfordjournals.org/D

ownloaded from

and 1–5% structural proteins on a dry weight basis (Vorwerk

et al., 2004). The structures of the individual polymers are well

known; however, their specific arrangement and bonding

patterns in the entire cell wall are not fully understood yet.

The complex assembly of stiff cellulose fibrils and pliable

matrix components can be characterized in the same way as

fiber-reinforced composites (Kerstens et al., 2001; Fratzl

et al., 2004). Based on deep-etching methods and NMR

spectroscopy, several cell wall models have been proposed

(Keegstra et al., 1973; Hayashi, 1989; Fry, 1989a; Talbott and

Ray, 1992; Ha et al., 1997; Cosgrove, 2000). Generally, the

cellulose microfibrils are thought to be tethered by xyloglucan

mainly through hydrogen bonding (Hayashi, 1989). Xyloglucan

attaches itself on the fibril surface as well as in between fibrils

(Pauly et al., 1999). Thereby, it coats the fibrils serving as a spacer

that prevents direct hydrogen bonding between cellulose

chains (Carpita and Gibeaut, 1993). Pectic polysaccharides

form a co-extensive network that interpenetrate this net-

work (McCann et al., 1990; Talbott and Ray, 1992) and interact

to a certain degree with hemicelluloses via both non-

covalent and covalent bonding (Fry, 1989b; Thompson and Fry,

2000).

From the 1960s onwards, mechanical measurements have

been performed on plant cell walls using extensometers to elu-

cidate factors that influence the extensibility of the cell walls as

well as the mechanical role of the cell wall components and

their interaction in the entire cell wall (Cleland, 1967, 1984;

Cleland and Rayle, 1977; Cosgrove 1988, 1989). Uniaxial tensile

tests have been also performed on variousArabidopsismutants

(Kohler and Spatz, 2002; Ryden et al., 2003; Pena et al., 2004;

Cavalier et al., 2008). Ryden et al. (2003) compared stiffness

and strength of Arabidopsis hypocotyls of GDP–fucose biosyn-

thesis mutant mur1, the xyloglucan fucosyltransferase mutant

mur2, and the xyloglucan galactosyltransferase mutant mur3.

The results were interpreted in a way that the mechanical per-

formance of primary walls depends on both galactosylated xylo-

glucan side chains and borate-complexed rhamnogalacturonan

II. However, a mechanistic model that proposes how xyloglucan

and pectin influence the stiffness and strength of Arabidopsis

hypocotyls has not been proposed yet.

The approach reported here aims to gather further insight

into the principle deformation mechanisms of the primary cell

wall and the mechanical interactions of the polymer networks,

particularly the interactions of xyloglucan and pectin. To elu-

cidate the mechanical role and the interplay of the structural

networks in the primary cell wall, standard tensile tests and

cyclic loading experiments were carried out on mur-mutants

(mur1 and mur2) and qua2. The latter mutant contains 50%

less homogalacturonan in comparison to the wild-type

(Mouille et al., 2007; Ralet et al., 2008). Based on the mechan-

ical responses of the various hypocotyls, a simple structural

model is proposed that extends existing models on the defor-

mation of the cellulose fibril–xyloglucan network (Passioura,

1994; Passioura and Fry, 1992; Veytsman and Cosgrove,

1998) by a possible interplay of xyloglucan chains with the pec-

tin network. This model shows some analogies to the so called

‘hidden length mechanism’, which, for instance, explains the

high toughness of bone by an additional deformability of

matrix polymers due to their specific structural alignment in

the assembly and polymer interactions by means of ionic sac-

rificial bonds (Fantner et al., 2005; Gupta et al., 2007).

RESULTS

Figure 1A shows a representative stress–strain curve of a 4-day-

old Arabidopsis wild-type (Col-0) hypocotyl illustrating its me-

chanical behavior and how the mechanical parameters used in

this study were determined.

The stress–strain curve of the standard tensile test shows an

initial phase, which is followed by an almost linear phase and

a regime of non-linear deformation after yield. The curve ends

at the point of rupture. Stiffness was calculated from the slope

of the curve in segment 2 and the ultimate stress value can

be taken as an approximate measure of the strength of the

hypocotyl.

In Figure 1B, an exemplary stress–strain curve of a 4-day-old

Arabidopsis mur1 hypocotyl in a cyclic loading experiment is

shown. Cyclic loading tests can further provide important in-

formation on the deformation behavior of a sample. Stiffness

was calculated for the upward loading phases. Stiffness 1

equates to the stiffness in the standard loading experiment

presented in Figure 1A. Stiffness 2, stiffness 3, and so forth re-

flect the material response when the sample is re-loaded after

unloading in the cycles. Cyclic loading experiments also allow

the distinction between the elastic and the plastic fraction of

a material response (Cleland, 1984). In a pure elastic deforma-

tion, all energy is returned after unloading, which means that

the unloading curve should hit the abscissa in the initial point

of the experiment. The ‘plastic strain’, as indicated in Figure

1B, was calculated as a qualitative measure of irreversible de-

formation. The given example also shows that the initial slope

(stiffness 1) of a hypocotyl does not necessarily reflect a pure

elastic material response (Young’s Modulus).

Standard tensile tests according to Figure 1A were carried

out on the hypocotyls to determine their ultimate stress levels

(;strength) and the stiffness (stiffness 1). Figure 2 shows the

mechanical behavior of the 4-day-old mur hypocotyls (Fig. 2A)

and the 6-day-old qua2 hypocotyls (Fig. 2B). Ultimate stress

levels and the stiffness of Col-0 hypocotyls of both respective

ages are shown for reference.

In the ultimate stress-versus-stiffness plots (Figure 2), only

mur2 shows a significant difference in the ultimate stress level

from Col-0, whereas no significant differences between Col-

0 and the mutants with pectin alterations, mur1 and qua2,

were observed. However, in terms of stiffness, all mutants were

significantly different from Col-0. While mur2 showed only

a moderate reduction in stiffness (;13%), the stiffness of

the two mutants with pectin alterations was noticeably de-

creased. The stiffness of mur1 was reduced by ;40% and

the stiffness of qua2 by ;34% compared to the Col-0

Abasolo et al. d Cell Wall Matrix Mechanics | 991

by guest on June 3, 2013http://m

plant.oxfordjournals.org/D

ownloaded from

hypocotyls. Above the other hypocotyls, several hypocotyls of

the mur1 mutant showed a pronounced non-linear deforma-

tion phase in the beginning of the tensile tests, which made it

necessary to determine the stiffness at rather high strain levels

(probably after a certain re-stiffening). Therefore, the stiffness

shown for mur1 is likely to be the upper limit.

Although mutants were tested at different ages, one can

observe a cluster of Col-0 and mur2 and a cluster of mur1

and qua2, which differ noticeably in stiffness but not in ulti-

mate stress levels.

To better understand the deformation mechanisms of the

primary cell walls of the hypocotyls, cyclic loading tests were

carried out in axial tension according to Figure 1B. After sev-

eral cycles, the samples were stressed until failure. Figure 3A–

3E show exemplary stress–strain curves of the 4 and 6-day-old

hypocotyls, whereas, in Figure 3F–3J, the changes in stiffness

are quantified for the first, second, and third cycle.

Figure 2. Ultimate Stress versus Stiffness Plots Displaying theTensile Properties of the Hypocotyls Col-0: filled triangle, mur1:filled square, mur2: filled circle, qua2: filled diamond.

(A) 4 day-old hypocotyls of Col-0, n = 91;mur1, n = 103;mur2, n = 84.(B) 6 day-old hypocotyls of Col-0, n = 38; qua2, n = 44; Error barsshow standard deviations. In terms of ultimate stress levels t-testsrevealed no significant differences between Col-0 and the pectinaltered mutants (mur1 and qua2), but a significant differences be-tween Col-0 and mur2 at a a = 0.01 level; in terms of stiffness mur1and qua2 were significantly different from Col-0 at a a = 0.001 leveland mur2 was significantly different from Col-0 at a a = 0.05 level,respectively.

Figure 1. Exemplary Stress-Strain Curves of a Standard Tensile Testand a Cyclic Loading Test on the Hypocotyls.

(A) Wild-type hypocotyl in a standard tensile test illustrating howthe tensile stiffness and the ultimate stress level (;strength) of thehypocotyls were determined.(B) Mur1 hypocotyl in a cyclic loading test illustrating how the stiff-ness values of the different loading cycles and the ‘‘plastic deforma-tion’’ in the 1st loading cycle (eplastic) were determined.

992 | Abasolo et al. d Cell Wall Matrix Mechanics

by guest on June 3, 2013http://m

plant.oxfordjournals.org/D

ownloaded from

The exemplary stress–strain curves clearly point to a different

mechanical response of the hypocotyls with pectin alterations,

but only within the first loading phase of the experiment. The

differences in the slopes between the first and second cycles are

much more pronounced for mur1 and qua2 compared to Col-

0 andmur2. In all further cycles, the deformation pattern seems

to be rather consistent for all types of hypocotyls. The apparent

differences in slopes between the first and the second cycles

were quantified in the vertical bar graphs in Figure 3F–3J.

All samples show a pronounced increase in stiffness from

the first to the second cycle, whereas only little change in stiff-

ness can be detected from the second to the third cycle.

Normalizing the stiffness of the first cycle (stiffness 1) as

100%, Col-0 (4-day-old), Col-0 (6-day-old), and mur2 showed

an increase in stiffness from the first to the second cycle to

136, 123, and 122%, respectively, whereas the relative stiffness

to mur1 and qua2 was considerably higher, reaching 186 and

168%, respectively (Figure 4A, the normalized stiffness of Col-0

is shown only for the 4-day-old hypocotyls). Another impor-

tant parameter that reflects the different behavior of the

hypocotyls is the plastic deformation (eplastic, in Figure 1B) dur-

ing the first loading (Figure 4B).

Both hypocotyls with pectin alterations, mur1 and qua2,

show notably higher plastic strain levels compared to Col-0

and mur2. However, this finding has to be qualified by saying

that the determination of the plastic strain from the stress–

strain curves is limited by experimental uncertainties. In fact,

it is not possible to distinguish between plastic deformation

and additional elongation due to sample reorientation in the

initial phase of the experiment. Moreover, the loading and

unloading cycles were performed continuously and therefore

not all viscoelastic deformation that is characteristic for plant

cell walls (Cleland, 1984) might have been relaxed before

reloading. Hence, these data should be taken qualitatively.

In order to consider the mechanical properties of the hypo-

cotyls as being indicative of specific deformation patterns of

the cell walls, it is necessary to include the impact of the cell

wall modification on hypocotyl turgor pressure (Figure 5). Dif-

ferences in hydrostatic pressure would be of crucial relevance

on the mechanical response of the hypocotyls, because turgor

pressure can influence the longitudinal stiffness of the hypo-

cotyls mainly by its impact on the Poisson’s ratio of the hypo-

cotyl by means of increasing the stiffness of the hypocotyl in its

transverse direction.

The data indicate that the cell wall modifications did not re-

sult in pronounced differences in turgor pressure between the

hypocotyls of the mutants and the wild-type. However, since

turgor pressure was determined indirectly by the difference

of water potential and osmotic pressure, two aspects that

qualify the results need to be addressed. The indirect calcula-

tion of turgor pressure leads to rather large standard devia-

tions because of error propagation and stress relaxation of

the cell walls that is likely to occur in the psychrometer also

affects turgor pressure (Cosgrove et al., 1984). Therefore, in

terms of the latter point, it is important additional information

Figure 3. Mechanical Response of the Hypocotyls to Cyclic Loading.

(A-C) Exemplary cyclic loading curves of the three different types ofthe 4 day-old hypocotyls.(D,E) Exemplary cyclic loading curves of 6 day-old Col-0 (only therelevant segment of the full stress-strain curve is shown) andqua2 hypocotyls.(F-H) Average stiffness change in the first, second and third cycle ofthe three different types of the 4 day-old hypocotyls. Error barsshow standard deviations; Col-0, n = 30; mur1, n = 36; mur2, n = 31.(I,J) Average stiffness change in the first, second and third cycle of 6day-old Col-0 and qua2 hypocotyls. Error bars show standard devi-ations; Col-0, n = 21; qua2, n = 19.

Abasolo et al. d Cell Wall Matrix Mechanics | 993

by guest on June 3, 2013http://m

plant.oxfordjournals.org/D

ownloaded from

whether the osmotic pressures of the different hypocotyls are

in the same range (Col-0: 0.94 6 0.13 MPa, mur1: 0.92 6 0.16

MPa, mur2: 0.92 6 0.15 MPa, qua2: 0.83 6 0.04 MPa). This is

the case, even though qua2 is slightly lower and significantly

different from Col-0 (non-parametric U-test 0.05 level) but not

different from mur1 and mur2. However, values for mur1

strongly coincide with Col-0 and mur2.

DISCUSSION

The primary cell wall consists of three networks that are made

up of cellulose fibrils and xyloglucan, pectin, and structural

proteins. These networks are described as being independent

but interrelated (Schindler, 1998). For the mechanical perfor-

mance of primary cell walls, it is believed that mainly the

cellulose and xyloglucan network plays a crucial role (Carpita

and Gibeaut, 1993; Pena et al., 2004; Cosgrove, 2005); however,

also the pectin network seems to be important. Mechanical

tests by Ryden and co-workers imply that borate-complexed

rhamnogalacturonan II formation influences the mechanical

properties of the cell wall as a stiffening and strengthening

agent (Ryden et al., 2003). Although these investigations pro-

vided an indication of the mechanical relevance of the individ-

ual networks, it is not yet understood how the networks

mechanically interact in the complex cell wall assembly and

what the specific mechanical role of a certain network compo-

nent is.

The analysis of the mechanical behavior of plants with cell

walls of different structural/chemical composition may help to

get better insight into the mechanical interaction of the cell

wall macromolecules. In our study, we compared the mechan-

ical behavior of well characterized Arabidopsis hypocotyls

with alterations in the xyloglucan side chain structure and

the pectin composition, respectively. In comparison to the

wild-type, the xyloglucan fucosyltransferase mutant mur2

lacks the terminal fucose sugar unit. It has already been shown

that this alteration rather marginally affects the mechanical

performance of the hypocotyl whereas more severe xyloglu-

can side chain alterations, such as that found in the xyloglucan

galactosyltransferase mutant mur3, result in a pronounced de-

crease in strength and stiffness (Ryden et al., 2003; Pena et al.,

2004). In the hypocotyls of the GDP–fucose biosynthesis mu-

tant mur1, the amount of fucose units is 60% of that in

wild-type hypocotyls (Ryden et al., 2003). This not only leads

to the absence of terminal fucose units in the xyloglucan side

chains, but has an additional effect on pectic polysaccharides.

The deficiency of fucose in rhamnogalacturonan II reduces its

ability to form borate diesters through the apiosyl residue

(Kobayashi et al., 1996). Qua2 shows an exclusive alteration

in the pectin network by means of 50% less homogalactur-

onan in comparison to the wild-type (Mouille et al., 2007; Ralet

et al., 2008).

From the standard tensile tests, two clusters of material

properties were observed, one consisting of Col-0 and mur2

and one of mur1 and qua2. In terms of tensile stiffness of

the mur-mutants, a similar trend was reported by Ryden et al.

(2003), although their values were somewhat higher (;50%).

This is likely due to differences in the nutrient concentration of

the medium, the growth temperature (25�C (Ryden et al.,

2003) versus 22�C here), and the test-setup (in particular test-

ing velocity). Indeed, when Col-0 was grown at 25�C, stiffness

was ; 42% higher than the ones grown at 22�C (data not

shown). For ultimate stress, however, the results were partly

inconsistent with the Ryden data. They found Col-0 to have

the highest strength and mur2 to be stronger than mur1,

whereas, in our study, Col-0 and the two mur-mutants as well

as the qua2 showed rather similar ultimate stress levels. These

Figure 4. Relative Stiffness and ‘‘Plastic Deformation’’ of the Hypo-cotyls in Cyclic Loading Tests.

(A) Comparison of the relative stiffness in the first three loadingcycles of the 4 day-old hypocotyls (Col-0, mur1, mur2) and the 6day-old hypocotyls (qua2).(B) Percent strain of plastic deformation (eplastic) during the first cy-cle of the 4 day-old hypocotyls (Col-0, mur1, mur2) and the 6 day-old hypocotyls (qua2).

Figure 5. Turgor Pressures of the Hypocotyls.

Turgor pressure of the 4 day-old hypocotyls (Col-0,mur1, mur2) andthe 6 day-old hypocotyls (qua2) were calculated from water poten-tial and osmotic pressure measurements. Error bars show standarddeviations based on calculations of error propagation (Col-0, n = 37;mur1, n = 12, mur2, n = 36, qua2, n = 4).

994 | Abasolo et al. d Cell Wall Matrix Mechanics

by guest on June 3, 2013http://m

plant.oxfordjournals.org/D

ownloaded from

inconsistencies should be taken into account here, although

they cannot be explained by different hypocotyl processing

at this stage.

One limitation of the mechanical characterization of hypo-

cotyls is that the system is fairly complex and cell wall proper-

ties are not measured directly. However, our main interest was

not in stiffness and ultimate stress values of the cell walls, but

in the change of deformation patterns of the hypocotyls and

relative changes in stiffness and ultimate stress in the course of

cell wall modifications. In order to calculate cell wall proper-

ties, several structural parameters (e.g. tissue density, cell

length) and water interactions would need to be measured.

One important parameter that can be expected to influence

also the mechanical deformation pattern of the hypocotyls

is the turgor pressure. However, as we could not detect consid-

erable differences in turgor pressure between the hypocotyls

of the mutants and the wild-type, the different mechanical

behaviors of the mutants with pectin alterations (mur1,

qua2) are not likely to be due to changes in turgor pressure.

Another aspect to be considered is that water may be pressed

out of the cells during mechanical testing and that pectin alter-

ations may facilitate this process by increasing the permeabil-

ity of the cell wall (Fleischer et al., 1999). However, such

a process could not explain the more pronounced re-stiffening

in the loading cycles of hypocotyls with pectin alterations com-

pared to the wild-type and the mur2 mutant. Therefore, we

conclude that differences in the mechanical behaviors are

mainly indicative for changes in structure–property relation-

ships due to the cell wall modifications.

Although cell wall properties could have not been directly

measured, our finding that stiffness rather than ultimate stress

of the hypocotyls is more affected by the pectin alterations

(see Figure 2) has several implications for the possible spatial

organization of pectin in the cell wall. The observations rule

out that pectin is the matrix component that prominently con-

tributes to a direct tethering of cellulose fibrils, as both param-

eters (stiffness and ultimate stress) would be affected in this

configuration. On the contrary, in terms of mur2, stiffness

and ultimate stress are reduced by almost the same relative

amount compared to the wild-type. However, the data clearly

point to the mechanical relevance of pectin as a structural el-

ement, and, although mur1 and qua2 possess alterations of

different pectin components, they showed a comparable me-

chanical response.

To specify the possible mechanical role of pectin, the specific

deformation patterns of the hypocotyls under cyclic regimes

have to be considered. All mutants and the wild-type displayed

a stiffening effect from the first to the second cycle during

cyclic loading but the increase in stiffness was more pro-

nounced for mur1 and qua2. Moreover, while Col-0 and

mur2 showed only little plastic deformation, the irreversible

deformation of mur1 and qua2 in the first cycle was noticeably

higher (see Figure 4).

The increase in stiffness from the first to the second cycle

should be generally associated with polymer reorientation

in the cell wall towards the stress axis (Richmond et al.,

1980). According to several authors (Preston, 1974; Reiterer

et al., 1999), the stiffness of a plant cell wall is a function of

its cellulose microfibril orientation. However, assuming almost

transverse-oriented fibrils, an increase in stiffness due to mi-

crofibril reorientation is unlikely, since much higher strains

than applied are needed to induce a noticeable passive reor-

ientation of cellulose fibrils (Burgert and Fratzl, 2009).

Although different deformation mechanisms should be dom-

inant at various strain rates, it is interesting to note that also

Marga et al. (2005) could not detect a change in cellulose fibril

orientation even after up to 30% straining of cucumber hypo-

cotyls in a creep experiment.

In a network of transversely oriented cellulose fibrils inter-

connected by xyloglucan chains, the latter are likely to be the

main load-bearing part. The length of the xyloglucan chains is

much longer than the spacing between cellulose fibrils

(McCann and Roberts, 1991; McCann et al., 1992), which makes

it likely that there is some folding of the xyloglucan chains.

During straining of the hypocotyls, the xyloglucan chains

may unfold. Hence, a possible explanation for the general stiff-

ening effect during the first cycle is that the xyloglucan chains

are straightening and thereby stiffening the entire wall. Thus,

the relative increase in stiffness would depend on the absolute

axial extension, since it dictates the degree of straightening of

the xyloglucan backbone.

The hypocotyls with pectin alterations show an entirely dif-

ferent deformation pattern compared to the wild-type by

meansofstiffeningbehaviorandplasticdeformation.Thesedif-

ferences seem to disappear after the second cycle. If pectins hin-

dertheunfoldingofthexyloglucanchain,thentheywouldhave

the capability to stiffen the network and reduce the amount of

chain flexibility in the first loading cycle. Although Thompson

and Fry (2000) could show that xyloglucan and pectin are able

to bind together covalently, an extensive cross-linking and, in

particular, an inter- and intra-chain connection between xylo-

glucan chains and pectin seems unlikely. Therefore, besides

some cross-linking, mainly geometrical constraints in pectin–

xyloglucan interactions may influence the flexibilityof the xylo-

glucan chains. If pectin, surrounding xyloglucan chains, has

a high rigidity, the unfolding process is hindered or limited.

The rigidity of the pectin should depend on the amount of pec-

tin (qua2)andthenumberof ion-mediatedcross-links(mur1),as

cross-linking in general stiffens polymer networks (Boyd and

Phillips, 1993). By modifying the pectin in the cell wall (mur1

andqua2), theamountofpectinandion-mediatedcross-linking

decreases, which softens the system and allows the xyloglucan

chains to unfold during deformation.

In Figure 6, the influence of pectin rigidity on xyloglucan

chain unfolding and thereby cell wall deformability is illus-

trated schematically. Figure 6A shows the initial state as as-

sumed for the two clusters, with and without pectin

alteration, respectively. The xyloglucan chains are laterally

bonded to the cellulose fibrils and folded in the space between

the fibrils. The xyloglucan chains are embedded in a dense

Abasolo et al. d Cell Wall Matrix Mechanics | 995

by guest on June 3, 2013http://m

plant.oxfordjournals.org/D

ownloaded from

mesh of pectin. The pectin alterations are simply illustrated by

a wider pectin mesh, since the model does not distinguish be-

tween the different pectin components, the number of ion-

mediated bonds in pectin, and cross-links to the xyloglucan.

As the alterations of two different pectin polymers lead to sim-

ilar mechanical responses and the nature of the interconnec-

tions between the xyloglucan and pectin networks are not too

clear, the mechanical tests alone are not sufficient to draw

a more specific model. Figure 6B illustrates the deformation

status of the cell wall after the first loading cycle. The cell walls

with pectin alteration—lower amount of pectin and ion-

mediated cross-links—show larger plastic deformation.

The larger unfolding and further alignment of the xyloglu-

can chains would result in a stiffening of the cell wall, which is

consistent with theories on polymer network processing

(Ward, 1997). Hence, the higher stiffening observed in the

hypocotyls with pectin alterations may be explained by a lower

pectin rigidity, which allows more unfolding and xyloglucan

chain straightening.

An alternative hypothesis or a parallel mechanism could be

that the pectin may contribute directly to the stiffness of the

cellulose–matrix composite. As shown by Zykwinska et al.

(2005), neutral sugar side chains of pectin are able to bind

in vitro to the cellulose fibril surface. According to Proseus

and Boyer (2007), the number of calcium-mediated cross-links

determines the stiffness and strength of pectin, which influen-

ces the cell wall deformability during growth. Therefore,

pectin alterations may also affect the capacity of pectin in teth-

ering cellulose fibrils. However, Cleland and Rayle (1977) could

not find a direct influence of calcium ion concentrations on the

stiffness of cell walls.

Our study showed that by changes in the pectin network,

the hypocotyl stiffness is far more affected than its ultimate

stress. Thus, it seems unlikely that the cell wall properties

are strongly dependent on a cellulose tethering function

of pectin, although failure of the cell wall is rather con-

trolled by the bonding pattern of the cellulose–xyloglucan

network.

The model of the xyloglucan and pectin interactions in some

ways resembles the mechanical effect of unfolding of macro-

molecules seen for many biological systems (e.g. DNA, titin,

etc.), which has been investigated by many authors both

experimentally and theoretically (e.g. Rief et al., 1997; Lu

et al., 1998). However, in contrast to ‘sticky chain models’

(Jager, 2001) exhibiting sacrificial bonds that inter-connect

a folded chain, here, it is probably the rigidity of the embed-

ding medium that may control the unfolding process. The pro-

posed model, but also a model for a direct tethering of

cellulose fibrils by pectin, point to an interesting analogy to

the mechanics of bone of which the high toughness has been

explained by a hidden length mechanism and sacrificial bonds

(Fantner et al., 2005; Gupta et al., 2007). As in bone, ion-

mediated bonds are likely to control the pectin rigidity either

by Ca2+ ions in homogalacturonan or by borate in RG II and

thereby influence the cell wall deformability.

In consequence, by ion-mediated links in pectin, plants

would not only be able to modulate cell wall structure, perme-

ability, swelling ability, and porosity (Zehirov and Georgiev,

2003; Jarvis, 1992; Zwieniecki et al., 2001; Fleischer et al.,

1999), but also influence the plastic deformability of the cell

wall at small strains. This finding may have implications on

the current understanding of the underlying mechanisms that

facilitate cell wall elongation during cell growth (reviewed by

Cosgrove, 2005). Proseus and Boyer (2007) showed that the

growth rate of Chara could be controlled by the deformability

and strength of pectin adjusted by the number of calcium-

mediated cross-links. In addition, our model is consistent with

a theory that cells might be able to initiate cell wall loosening

by regulating the ion-mediated cross-linking in the pectin.

The advantage in the process of cell wall elongation driven

by turgor pressure would be that it allows increasing

deformability of the cell wall without severely affecting its

strength.

METHODS

Arabidopsis thaliana (L.) Heynh. hypocotyls of wild-type

(Col-0), mur1, mur2, and qua2 were utilized for this study.

Figure 6. Simple Structural Model of the Influence of the Geomet-rical Interactions of Folded Xyloglucan Chains with Pectin.

Cell walls of hypocotyls with pectin alteration are illustrated witha wider mesh (CF, cellulose fibril; XG, xyloglucan chain; PE, pectin).(A) Initial state before straining with a given space between thefibrils D0.(B) Cell walls with pectin alterations (wider mesh) show largerplastic deformation after the first loading cycle than cell walls with-out pectin alteration (D0 , D1 , D2).

996 | Abasolo et al. d Cell Wall Matrix Mechanics

by guest on June 3, 2013http://m

plant.oxfordjournals.org/D

ownloaded from

Mur1 is defective in the GDP-D-mannose-4,6-dehydratase

(Bonin et al., 1997), leading to an altered structure of both

xyloglucan and rhamnogalacturonan II (RG II) (Reiter et al.,

1993), whilemur2 shows a specifically altered xyloglucan struc-

ture particularly in xyloglucan fucosyltransferase (Vanzin et al.,

2002). Qua2 possesses 50% less homogalacturonan in compar-

ison to wild-type (Mouille et al., 2007). Sterilized seeds were

plated in 8.8 g l�1 Murashige and Skoog basal medium in

0.8% agar, incubated in the dark at 22�C for 4 d in terms of

the mur mutants and 6 d in terms of qua2. Hypocotyls of

Col-0 were tested after both growth periods. Tests on older

qua2 hypocotyls were necessary because the 4-day-old hypo-

cotyls were very short and showed yielding at stress levels that

were too low for running consistent cyclic loading tests.

Wild-type and the mutant hypocotyls were individually

glued onto foliar frames, fixing them by a stepwise combina-

tion of rapid cyanoacrylate adhesive and Glass Ionomer Luting

Cement (3M ESPE Ketac� Cem l). The free length of the hypo-

cotyl was located between the basal and the middle part. The

foliar frames were specifically designed to be mounted onto

a microtensile apparatus via pin-hole assembly. This is of crucial

relevance, since the highly sensitive load cell with a maximum

capacity of 500 mN forbids other mechanical clamping. The

microtensile apparatus consists of a linear table driven by a step

motor that allows feed rates of between 0.5 and 30 lm s�1.

Black markers on the foliar frame allow following elongation

more precisely compared to the machine path of the testing

device (Burgert et al., 2003).

Hypocotyls were tested at room temperature at strain rates

of either 10 or 15 lm s�1; the gauge lengths of the 4-day-old

hypocotyls were, on average: Col-0 ;2.5 mm; mur1 ;2.2 mm;

mur2;2.4 mm; the gauge lengths of the 6-day-old hypocotyls

were, on average: Col-0 ;2.6 mm; qua2 ;2.7 mm. The dura-

tion of a simple loading test was ;30 s; cyclic loading tests

until completing the third cycles took ;60 s. Vapor was con-

stantly applied to the specimens to inhibit sample drying dur-

ing the measurement. In the successive loading–unloading

cycles, hypocotyls were subjected to several cycles before the

sample was stressed until failure.

For transferring force–elongation curves into stress–strain

curves, the cross-sections of the hypocotyls were calculated

on the basis of diameter measurements under the microscope,

assuming a circular outline of the hypocotyls. From the stress–

strain curve generated in the standard tensile tests (see

Figure 1A), sample stiffness (slope of the curve at a linear

segment) and the ultimate stress level (curve’s peak) of the

hypocotyls were derived. In terms of cyclic loading (see

Figure 1B), stiffness was determined in the upward loading

phase for the first three cycles.

Turgor Pressure Determination

Turgor pressure was determined by the difference of water

potential and osmotic pressure. Both parameters were mea-

sured with a C-52 sample chamber (Wescor, Logan, UT) in

a sample holder with a diameter of 9.5 mm and a depth of

4.5 mm. The psychrometer was read by an automated datalog-

ger (Wescor, Logan, UT). The system was calibrated with a so-

dium chloride solution (1000 mmol kg�1) used as a water

potential standard of 2500 kPa (25�C). Psychrometers were

allowed to equilibrate under each set of conditions. Cooling

time to initiate condensation on the psychrometric junction

was 30 s and the temperature depression from evaporative

cooling was measured 30 s after active cooling ceased for a

period of 30 s. The measured values were averaged. For

the determination of the water potential, 15 seedlings were

measured after at least 20 h of equilibration in one sample

chamber.

The osmotic pressure was determined as described by

Kutschera(1991).Thehypocotylswereexcisedfromtheseedlings.

Thereafter, they were frozen in liquid nitrogen, homogenized,

and centrifuged for 2 min at 10 000 rpm. After centrifugation,

theosmoticconcentrationofthesupernatantwasmeasuredafter

an equilibration time of at least 2 h in the sample chamber.

Statistical Evaluation

Statistical evaluation of the mechanical behavior of the differ-

ent types of hypocotyls was performed by t-tests at a = 0.05,

a = 0.01, and a = 0.001 confidence levels.

FUNDING

The financial support from the Max Planck Society and the EU grant

028974, project CASPIC, is gratefully acknowledged.

ACKNOWLEDGMENTS

We would like to thank Norma Funke, MPI-MP, as well as

Annemarie Martins, Petra Leibner, and Susann Weichold, MPI-

KG, for excellent technical support. No conflict of interest declared.

REFERENCES

Baskin, T.I. (2005). Anisotropic expansion of the plant cell wall.

Annu. Rev. Cell Dev. Biol. 21, 203–222.

Bonin, C.P., Potter, I., Vanzin, G.F., and Reiter, W.-D. (1997). The

MUR1 gene of Arabidopsis thaliana encodes an isoform of

GDP-D-mannose-4,6-dehydratase, catalyzing the first step in

the de novo synthesis of GDP-L-fucose. Proc. Natl Acad. Sci.

U S A. 94, 2085–2090.

Boyd, R.H., and Phillips, P.J. (1993). The Science of Polymer Mole-

cules (Cambridge: Cambridge University Press).

Burgert, I., and Fratzl, P. (2009). Plants control the properties and

actuation of their organs through the orientation of cellulose

fibrils in their cell walls. Integrative and Comparative Biology.

49, 69–79.

Burgert, I., Fruhmann, K., Keckes, J., Fratzl, P., and Stanzl-

Tschegg, S.E. (2003). Microtensile testing of wood fibers com-

bined with video extensometry for efficient strain detection.

Holzforschung. 57, 661–664.

Abasolo et al. d Cell Wall Matrix Mechanics | 997

by guest on June 3, 2013http://m

plant.oxfordjournals.org/D

ownloaded from

Carpita, N.C., and Gibeaut, D.M. (1993). Structural models of

primary cell walls in flowering plants: consistency of molecular

structure with the physical properties of the walls during

growth. Plant J. 3, 1–30.

Cavalier, D.M., et al. (2008). Disruption of two Arabidopsis thaliana

xylosyltransferase genes results in plants deficient in xyloglucan,

a major primary cell wall component. Plant Cell. 20, 1519–1537.

Cleland, R.E. (1967). Extensibility of isolated cell walls: measure-

ment and changes during cell elongation. Planta. 74, 197–209.

Cleland,R.E. (1984).The Instrontechniqueasameasureofimmediate-

past wall extensibility. Planta. 160, 514–520.

Cleland, R.E., and Rayle, D.L. (1977). Reevaluation of the effect of

calcium ions on auxin-induced elongation. Plant Physiol. 60,

709–712.

Cosgrove, D.J. (1988). Mechanism of rapid suppression of cell

expansion in cucumber hypocotyls after blue-light irradiation.

Planta. 176, 109–116.

Cosgrove, D.J. (1989). Characterization of long-term extension of

isolated cell walls from growing cucumber hypocotyls. Planta.

177, 121–130.

Cosgrove, D.J. (2000). Expansive growth of plant cell walls. Plant

Physiol. Biochem. 38, 109–124.

Cosgrove, D.J. (2005). Growth of the plant cell wall. Nature

Reviews—Molecular Cell Biology. 6, 850–861.

Cosgrove, D.J., Van Volkenburgh, E., and Cleland, R.E. (1984). Stress

relaxation of cell walls and the yield threshold for growth.

Planta. 162, 46–54.

Creelman, R.A., and Mullet, J.E. (1997). Oligosaccharins, brasshino-

lides and jasmonates: nontraditional regulators of plant growth,

and development and gene expression. Plant Cell. 9, 1211–1223.

Eckardt, N.A. (2003). Cellulose synthesis takes the CesA train. Plant

Cell. 15, 1685–1688.

Fantner, G.E., et al. (2005). Sacrificial bonds and hidden length

dissipate energy as mineralized fibrils separate during bone

fracture. Nature Materials. 4, 612–616.

Fleischer, A., O’Neill, M.A., and Ehwald, R. (1999). The pore size of

non-graminaceous plant cell walls is rapidly decreased by borate

ester cross-linking of the pectic polysaccharide rhamnogalactur-

onan II. Plant Physiol. 121, 829–838.

Fratzl, P., Burgert, I., and Gupta, H.S. (2004). On the role of interface

polymers for the mechanics of natural polymeric composites.

Phys. Chem. Chem. Phys. 6, 5575–5579.

Fry, S.C. (1989a). Cellulases, hemicelluloses and auxin-stimulated

growth: a possible relationship. Physiol. Plant. 75, 532–536.

Fry, S.C. (1989b). The structure and functions of xyloglucan. J. Exp.

Bot. 40, 1–11.

Gupta, H.S., Fratzl, P., Kerschnitzki, M., Benecke, G.,

Wagermaier, W., and Kirchner, H.O.K. (2007). Evidence for an

elementary process in bone plasticity with an activation enthalpy

of 1eV. J. Royal Society Interface. 4, 277–282.

Ha, M.-A., Apperley, D.C., and Jarvis, M.C. (1997). Molecular rigidity

in dry and hydrated onion cell walls. Plant Physiol. 115, 593–598.

Hayashi, T. (1989). Xyloglucans in the primary cell wall. Annu. Rev.

Plant Physiol. Plant Mol. Biol. 40, 139–168.

Jager, I. (2001). The ‘sticky chain’: a kinetic model for the deforma-

tion of biological macromolecules. Biophysical J. 81, 1897–1906.

Jarvis, M.C. (1992). Control of thickness of collenchyma cell walls by

pectins. Planta. 187, 218–220.

Keegstra, K., Talmadge, K.W., Bauer, W.D., and Albersheim, P.

(1973). The structure of the plant cell walls. III. A model of the

walls of suspension-cultured sycamore cells based on the inter-

connections of the macromolecular components. Plant Physiol.

51, 188–196.

Kerstens, S., Decraemer, W.F., and Verbelen, J.P. (2001). Cell walls at

the plant surface behave mechanically like fiber-reinforced com-

posite materials. Plant Physiol. 127, 381–385.

Kobayashi, M., Matoh, T., and Azuma, J. (1996). Two chains of

rhamnogalacturonan II are cross-linked by borate–diol ester

bonds in higher plant cell walls. Plant Physiol. 110, 1017–1020.

Kohler, L., and Spatz, H.-C.H. (2002). Micromechanics of plant tis-

sues beyond the linear–elastic range. Planta. 215, 33–40.

Kutschera, U. (1991). Osmotic relations during elongation growth

in hypocotyls of Helianthus annuus L. Planta. 184, 61–66.

Lu, H., Isralewitz, B., Krammer, A., Vogel, V., and Schulten, K. (1998).

Unfolding of titin immunoglobulin domains by steered molecu-

lar dynamics simulation. Biophysical J. 75, 662–671.

Marga, F., Grandbois, M., Cosgrove, D.J., and Baskin, T.I. (2005). Cell

wall extension results in the coordinate separation of parallel

microfibrils: evidence from scanning electron microscopy and

atomic force microscopy. Plant J. 43, 181–190.

McCann, M.C., and Roberts, K. (1991). Architecture of the primary

cell wall. In The Cytoskeletal Basis of Plant Growth and Form,

Lloyd, C.W., ed. (New York: Academic Press), pp. 109–129.

McCann, M.C., Wells, B., and Roberts, K. (1990). Direct visualization

of cross-links in the primary cell wall. J. Cell Sci. 96, 323–334.

McCann, M.C., Wells, B., and Roberts, K. (1992). Complexity in the

spatial localization and length distribution of plant cell-wall

matrix polysaccharides. J. Microscopy. 166, 123–136.

Mouille, G., et al. (2007). Homogalacturonan synthesis in Arabidop-

sis thaliana requires a Golgi-localized protein with a putative

methyltransferase domain. Plant J. 50, 605–614.

Passioura, J.B. (1994). The physical chemistry of the primary cell

wall: implications for the control of expansion rate. J. Exp.

Bot. 45, 1675–1682.

Passioura, J.B., and Fry, S.C. (1992). Turgor and cell expansion: be-

yond the Lockhart equation. Aust. J. Plant Physiol. 19, 565–576.

Pauly, M., Albersheim, P., Darvill, A., and York, W.S. (1999). Molec-

ular domains of the cellulose/xyloglucan network in the cell walls

of higher plants. Plant J. 20, 629–639.

Pena, M.J., Ryden, P., Madson, M., Smith, A.C., and Carpita, N.C.

(2004). The galactose residues of xyloglucan are essential to

maintain mechanical strength of primary cell walls in Arabidop-

sis during growth. Plant Physiol. 134, 443–451.

Preston, R.D. (1974). The Physical Biology of Plant Cell Walls (London:

Chapman and Hall).

Proseus, T.E., and Boyer, J.S. (2007). Tension required for pectate

chemistry to control growth in Chara corallina. J. Exp. Bot. 58,

4283–4292.

Ralet, M.C., Crepeau, M.J., Lefebvre, J., Mouille, G., Hofte, H., and

Thibault, J.F. (2008). Reduced number of homogalacturonan

domains in pectins of an Arabidopsis mutant enhances the

flexibility of the polymer. Biomacromolecules. 9, 1454–1460.

998 | Abasolo et al. d Cell Wall Matrix Mechanics

by guest on June 3, 2013http://m

plant.oxfordjournals.org/D

ownloaded from

Reiter, W.D., Chapple, C.C.S., and Somerville, C.R. (1993). Altered

growth and cell-walls in a fucose-deficient mutant of Arabidop-

sis. Science. 261, 1032–1035.

Reiterer, A., Lichtenegger, H., Tschegg, S., and Fratzl, P. (1999).

Experimental evidence for a mechanical function of the cellu-

lose microfibril angle in wood cell walls. Phil. Mag. A. 79,

2173–2184.

Richmond, P.A., Metraux, J.-P., and Taiz, L. (1980). Cell expansion

patterns and directionality of wall mechanical properties in

Nitella. Plant Physiol. 65, 211–217.

Rief, M., Gautel, M., Oesterhelt, F., Fernandez, J.M., and Gaub, H.E.

(1997). Reversible unfolding of individual titin immunoglobulin

domains by AFM. Science. 276, 1109–1112.

Rose, J.K.C., and Bennett, A.B. (1999). Cooperative disassembly of

the cellulose-xyloglucan network of plant cell walls: parallels

between cell expansion and fruit ripening. Trends Plant Sci. 4,

176–183.

Ryden, P., Sugimoto-Shirasu, K., Smith, A.C., Findlay, K., Reiter, W.-

D., and McCann, M. (2003). Tensile properties of Arabidopsis cell

walls depend on both a xylogucan cross-linked microfibrillar

network and rhamnogalacturonan II-borate complexes. Plant

Physiol. 132, 1033–1040.

Schindler, T.M. (1998). The new view of the primary cell wall. Z.

Pflanzenernahr. Bodenk. 161, 499–508.

Taiz, L. (1984). Plant cell wall expansion: regulation of cell wall

mechanical properties. Annu. Rev. Plant Physiol. 35, 585–657.

Talbott, L.D., and Ray, P.M. (1992). Molecular size and separability

features of pea cell wall polysaccharides: implications for models

of primary wall structure. Plant Physiol. 98, 357–368.

Thompson, J.E., and Fry, S.C. (2000). Evidence for covalent linkage

between xyloglucan and acidic pectins in suspension-cultured

rose cells. Planta. 211, 275–286.

Vanzin, G.F., Madson, M., Carpita, N.C., Raikhel, N.V., Keegstra, K.,

and Reiter, W.-D. (2002). The mur2 mutant of Arabidopsis thali-

ana lacks fucosylated xyloglucan because of a lesion in fucosyl-

transferase AtFUT1. Proc. Natl Acad. Sci. U S A. 99, 3340–3345.

Veytsman, B.A., and Cosgrove, D.J. (1998). A model of cell wall

expansion based on thermodynamics of polymer networks. Bio-

phys. J. 75, 2240–2250.

Vorwerk, S., Somerville, S., and Somerville, C. (2004). The role of

plant cell wall polysaccharide composition in disease resistance.

Trends Plant Sci. 9, 203–209.

Ward, I.M. (1997). Structure and properties of oriented polymers

(London: Chapman and Hall).

Whitney, S.E.C., Gothard, M.G.E., Mitchell, J.T., and Gidley, M.J.

(1999). Roles of cellulose and xyloglucan in determining the

mechanical properties of primary plant cell walls. Plant Physiol.

121, 657–663.

Zehirov, G.T., and Georgiev, G.I. (2003). Effects of boron starvation

on the apoplastic and total solute concentrations influencing

nodule growth and acetylene reduction rate. Bulg. J. Plant Phys-

iol. 367–373.

Zwieniecki, M.A., Melcher, P.J., and Holbrook, N.M. (2001). Hydro-

gel control of xylem hydraulic resistance in plants. Science. 291,

1059–1062.

Zykwinska, A.W., Ralet, M.C., Garnier, C.D., and Thobault, J.F.J.

(2005). Evidence for in vitro binding of pectin side chains to

cellulose. Plant Physiol. 139, 397–407.

Abasolo et al. d Cell Wall Matrix Mechanics | 999

by guest on June 3, 2013http://m

plant.oxfordjournals.org/D

ownloaded from