Rts

Ya

Tb

c

d

a

ARRA

KASPCTB

1

rdpbeam

nLXf

0h

Industrial Crops and Products 51 (2013) 62– 69

Contents lists available at ScienceDirect

Industrial Crops and Products

journa l h om epa ge: www.elsev ier .com/ locate / indcrop

evealing the microfibrillar arrangement of the cell wall surface andhe macromolecular effects of thermochemical pretreatment inugarcane by atomic force microscopy

uri Abuda,b,1, Lilian T. Costaa,c,1, Wanderley de Souzaa,b,d, Celso Sant’Annaa,b,∗

Laboratório de Biologia Estrutural, Diretoria de Metrologia Aplicada às Ciências da Vida – Dimav, Instituto Nacional de Metrologia, Qualidade eecnologia, INMETRO, RJ, BrazilInstituto Nacional de Ciência e Tecnologia em Biologia Estrutural e Bioimagens, Universidade Federal do Rio de Janeiro, UFRJ, RJ, BrazilPolo Xerém, Universidade Federal do Rio de Janeiro, UFRJ, RJ, BrazilLaboratório de Ultraestrutura Celular Hertha Meyer, Instituto de Biofísica Carlos Chagas Filho, Universidade Federal do Rio de Janeiro, UFRJ, RJ, Brazil

r t i c l e i n f o

rticle history:eceived 4 June 2013eceived in revised form 2 August 2013ccepted 28 August 2013

eywords:tomic force microscopyugarcanelant cell wallellulose microfibrilshermochemical pretreatment

a b s t r a c t

The lignocellulosic biomass of plant cell wall is a source of sugar that can be used for second generationof bioethanol. The technology used for converting the fermentable sugar in the cell wall into bioethanolinvolves a pretreatment step that helps to decrease the recalcitrance of biomass to deconstruction. Atomicforce microscopy (AFM), due to its ability to generate high resolution images with negligible samplepreparation, is a powerful tool for studying biological samples on a nanometer scale. In this work, weelucidated the microfibrillar arrangement of the sugarcane cell wall and analyzed the effect of a ther-mochemical pretreatment at the nanoscale level using AFM images. Initially, we isolated the sugarcanecell wall to improve the AFM analysis. The cellulose microfibrils distribution in the sugarcane cell wallwas clarified by AFM. Cellulose microfibrils were organized in a polylaminated structure with a distinctorientation. The measured filaments were approximately 20 nm thick, which corresponds to the size

ioethanol of cellulose microfibrils in other crops. Additionally, we investigated the effect of the thermochemicalpretreatment on the cell wall structure. The thermochemical-pretreated cell wall had undergone lossof filaments and formation of globular structures, presumably containing lignin. The surface roughnesswas also dramatically changed. Our results provide insights that may help to understand the sugarcanecell wall organization and the mechanism by which thermochemical pretreatments modify the cell wallstructure to improve its digestibility.

. Introduction

As fossil fuels are now scarce and nonrenewable energyesources are becoming more expensive, there is an increasingemand for an alternative energy source. In terms of energyroduction, cost and environmental impact, biofuels woulde a better alternative than fossil fuels. International research

fforts strive to obtain a renewable fuel that is inexpensivend causes less pollution. As such, significant efforts have beenade to use vegetal biomass for the production of liquid fuels,

∗ Corresponding author at: Instituto Nacional de Metrologia, Qualidade e Tec-ologia – INMETRO, Diretoria de Metrologia Aplicada às Ciências da Vida – Dimav,aboratório de Biotecnologia – LABIO, Av. Nossa Senhora das Grac as 50, prédio 27 –erém, CEP: 25250-020, Duque de Caxias, RJ, Brazil. Tel.: +55 21 2145 3150;

ax: +55 21 679 1420.E-mail address: cbfilho@inmetro.gov.br (C. Sant’Anna).

1 These authors contributed equally to this work.

926-6690/$ – see front matter © 2013 Elsevier B.V. All rights reserved.ttp://dx.doi.org/10.1016/j.indcrop.2013.08.076

© 2013 Elsevier B.V. All rights reserved.

including bioethanol, which could be used to power transportation(Goldemberg, 2007).

The conversion of plant biomass (called lignocellulosicbiomass), including woody crops, herbaceous plants, grasses,starch, sugar cane, and others, into bioethanol has emerged aspotential renewable energy source to replace fossil fuels (Himmelet al., 2007; Jordan et al., 2012). A lignocellulosic biomass consistsprimarily of plant cell walls (PCW), which are mainly composed ofbiopolymers, such as cellulose, hemicellulose, and lignin. The PCWinvolves the entire cells and provides mechanical support, stressprotection, and rigidity and support to plant cells (Keegstra, 2010).The PCW is divided into 3 layers: (1) the thinner non-lignified pri-mary cell wall, (2) the thicker highly lignified secondary cell wall,and (3) the middle lamella that provides support to adjacent cells.The general architecture of the cell wall surface contains microfib-

rils embedded in an amorphous matrix, which results in a rigid andcompact mesh-like structure. The microfibril is composed of cellu-lose and the amorphous material contains hemicellulose and lignin(Kaczkowski, 2003).

ops an

mfclcxvpsfiite

v(aatofiampttucispmnaar2

pta(ntwsMhot

ttobsatsi(

ow

Y. Abud et al. / Industrial Cr

Cellulose, the major constituent of the PCW, is a homopoly-er arranged into crystalline and non-crystalline regions that are

ormed by �-(1,4)-linked d-glucoses (Perez and Mazeau, 2004). Theellulose is arranged into a fibril, which is surrounded by hemicel-ulose, an amorphous heteropolymer with low molecular weightompounds and a mixture of monosaccharides (glucose, mannose,ylose and arabinose) (Saha, 2003), and lignin, a polymer that pro-ides the stiffness, impermeability, and mechanical strength of thelant tissues (Donaldson, 2001). Lignin is the key molecule respon-ible for the recalcitrance of the biomass (Grabber, 2005). Cellulosebrils with a diameter of approximately 20 nm have been found

n plant cell walls, as observed using the quick-freeze, deep-etchechnique (McCann et al., 1990) and AFM (Kirby et al., 1996; Morrist al., 1997).

The physicochemical traits of the PCW make efficiently con-erting the biomass into sugars by enzymatic hydrolysis difficultMcCann et al., 1990). It has been postulated that the mechanicalnd functional properties of the cell wall provided by the molecularrchitecture cause the recalcitrance of the biomass to deconstruc-ion (Xu et al., 2011). The high degree of compaction and complexityf the lignocellulosic biomass structure makes its conversion intoermentable sugars considerably more difficult. Therefore, produc-ng ethanol from biomass is still costly and inefficient. However,dding a physical, chemical, physicochemical or biological pretreat-ent step before the enzymatic hydrolysis and sugar fermentation

rocesses can improve the production of bioethanol. These pre-reatments make the cellulose more accessible to the enzymeshat convert the carbohydrates into fermentable sugars via the sol-bilization and separation of the lignin–hemicellulose–celluloseomplex (Mosier et al., 2005; Agbor et al., 2011). Due to its chem-cal composition, plant anatomy and the variations in the PCWtructure, there is no single pretreatment method employed in theroduction of biofuels from lignocellulosic biomass. In the PCW,icron-scale traits such as lignin concentration, cell wall thick-

ess, and the cross-linkers between the cell wall polymers, as wells nanoscale traits such as the porosity of the cell wall matrix, therrangement of the cellulose fibrils and crystallinity, all convey theesistance of the plant cell walls to deconstruction (Sarkar et al.,009).

The vast structural diversity of cell walls found in differentlants, as well as in the tissues and cells within the same planthat convey the diverse functions of the cell wall, influences thebility of cellulases to penetrate the cell wall to access celluloseDonohoe et al., 2009). Thus, it is necessary to characterize the orga-ization of the PCW and to determine how this organization relateso the ability of the biomass to be hydrolyzed. Elucidating the cellall recalcitrance requires a detailed understanding of the cell wall

tructure, including the arrangement of the cellulose microfibrils.icroscopy techniques, including atomic force microscopy (AFM),

ave been widely applied to obtain detailed information of therganization and deconstruction of the PCW before and after pre-reatments (reviewed by De Souza and Sant’Anna, 2012).

AFM is an important tool for studying biological samples dueo its ability to image biological samples under atmospheric condi-ions, thus allowing measurements to be performed in their nativer near-native state (Hansma, 2001; Müller et al., 2011). AFM isased on the physical interaction between the cantilever tip and theample surface (Binnig et al., 1986). Adhesion forces between the tipnd the cell surface molecules are detected during cantilever deflec-ions, which allow one to understand the chemical properties of theample surface. This technique has been used to provide structuralnformation of the PCW at high resolution, including plant cell wall

Chundawat et al., 2011; Ding et al., 2012).

In this work, AFM was used to investigate the morphologyf the cellulose microfibrils in the sugarcane parenchyma cellall surface. Additionally, we analyzed surface changes after a

d Products 51 (2013) 62– 69 63

thermochemical pretreatment using topographic analysis and sur-face roughness measurements. In addition, we obtained a fractionof the isolated sugarcane cell wall by cell fractionation to improvethe AFM analysis.

2. Materials and methods

2.1. Plant material

Stems from the sugarcane cultivar CB45-3 were obtained fromplantations at Campos dos Goytacazes, Rio de Janeiro State, Brazil.

2.2. Isolation of the parenchyma cell wall of sugarcane

The sugarcane parenchyma cell wall was isolated according to(McCann et al., 1990). Briefly, 20 g of the central region of sugarcaneinternodes, where parenchyma cells are predominant, were mixedin 20 mL of homogenization buffer (0.3 M sucrose, 0.1 M HEPES,2 mM potassium metabisulfide and 10 mM calcium chloride) in aSorvall Omni-Mixer at 4 ◦C. The residue was filtered on filter paper,and the paste-like material was transferred to a crucible and frozenin liquid nitrogen. Subsequently, three cycles of freezing and mac-eration were performed. The sample was transferred to a beakerwith 30 mL of maceration buffer (50 mM sodium acetate (pH 5.5),50 mM NaCl, and 30 mM ascorbic acid) and subjected to shakingfor 15 min. The sample was subsequently decanted for 5 min, andthe less dense material was re-filtered on filter paper while thethicker fibers were naturally sedimented. The material retained onthe filter paper was again subjected to agitation, sedimentation andfiltration to achieve a better purification of the sugarcane cell wall.The purity of the fraction was assessed by light microscopy.

2.3. Thermochemical pretreatment

The central region of sugarcane internodes were subjected tocell wall fractionation, as described elsewhere, to obtain purifiedparenchyma cell wall fragments. The obtained cell wall fragmentswere washed three times in distilled water and were treated with2% sulfuric acid at 150 ◦C for 20 min. After extensive washing inwater, the fragments were imaged by AFM as described below.

2.4. AFM image acquisition and analysis

Freshly prepared sugarcane parenchyma cell wall fragmentsbefore and after pretreatment were deposited onto freshly cleavedmica sheets. The excess water was removed using a flow of nitro-gen to minimize possible artifacts generated by humidity andthe air-drying procedure, and the samples were imaged imme-diately. All images were acquired using a Nanowizard JPK AFM(Berlin, Germany) operating in the dynamic mode under ambientconditions, with 46% relative humidity and using AC240 can-tilevers (Veeco, Woodbury, NY) with a spring constant of 2 N/mand a resonance frequency of 70 kHz. The nominal tip radius was10 nm. AFM image analysis (n = 98) was performed using the JPKImage Processing software. The diameter of the sugarcane cellulosemicrofibrils was measured using the intermittent contact mode. Ahistogram of the diameter of the cellulose microfibrils was gener-ated, and the standard deviation was calculated.

The roughness measurements (RMS) were performed using theJPK Image Processing software. We used a non-parametric test, theMann–Whitney test, with a 95% level of significance.

2.5. Ultrastructural cytochemistry

Sugarcane internodes were hand-cut with a razor blade into2 mm × 2 mm blocks. The blocks were washed in acetone, in

64 Y. Abud et al. / Industrial Crops and Products 51 (2013) 62– 69

F nsvers s in thf

dTbicstot(1t

3

3

pbow(ioaptwc

3

wttewtr

ig. 1. Light microscopy image of sugarcane parenchyma cell wall. (a) Hand-cut trahowing the parenchyma tissue before the cell wall fractionation with intact cellractionation.

istilled water and pretreated with sulfuric acid as describe before.he samples were fixed in 2.5% glutaraldehyde in 0.08 M cacodylateuffer, pH 6.5, under vacuum, post-fixed in 1% osmium tetrox-

de, 0.8% potassium ferrocyanide, 5 mM calcium chloride in 0.1 Macodylate buffer, pH 7.2, dehydrated in an increasing acetoneeries and embedded in Spurr resin. Ultrathin sections (70 nmhick) were obtained in the LEICA ultramicrotome and depositednto copper grids. In order to selectively stain lignin, ultrathin sec-ions were treated with 1% potassium permanganate for 10 minSant’Anna et al., 2013). Samples were examined in a FEI Tecnai G2

2 Spirit LaB6 transmission electron microscope with an accelera-ion voltage of 120 kV.

. Results

.1. Sugarcane cell wall fractionation

To improve the AFM measurements, we purified the sugarcanearenchyma cell wall from the internode of the sugarcane stalky sub-cellular fractionation based on the protocol developed innion (McCann et al., 1990). The purity and integrity of the fractionas assessed by light microscopy (Fig. 1). The cell wall fragments

Fig. 1b) showed morphological characteristics as observed in thentact parenchyma tissue (Fig. 1a). Some disrupted cell walls werebserved, most likely due to the fractionation procedure. During

random observation, we did not observe the presence of otherlant structures or cellular components (Fig. 1b), indicating thathe preparation was virtually free of contaminants. This materialas utilized to locally investigate the molecular structure of the

ell wall by AFM.

.2. Microfibril organization in the sugarcane cell wall

Freshly prepared sugarcane parenchyma cell wall fragmentsere deposited onto cleaved mica to investigate the organiza-

ion of the microfibril array in the native sugarcane cell wall athe nanoscale using comprehensive AFM. Following the method

mployed by Kirby et al. (1996) the sugarcane cell wall fragmentsere imaged while moist under ambient conditions. The intermit-

ent contact mode AFM images of the parenchyma tissue clearlyevealed the assembly of the cell wall cellulose microfibrils in the

se section (2 mm thick and 5 mm long) from the periphery of sugarcane internodeseir typical organization. (b) Image of cell wall fragments obtained after cell wall

sugarcane (Fig. 2a and b). Fig. 2a and b is representative imagesfrom more than 30 random images that were taken of the cell wallsurface. The cell wall was assembled in a laminated fibrous struc-ture. In the AFM images, the cellulose microfibrils arranged in theoverlapping layers were reproducible. Each cell wall layer surfacewas composed of a highly dense network in which the elementarymicrofibrils were predominantly in a random orientation with norecognizable pattern (Fig. 2a and b). However, the oblique orienta-tion of the microfibrils appears to be predominant. The high densityof microfibrils makes visualizing the cell wall matrix, which is theamorphous material where the cellulose microfibrils are embed-ded, difficult, suggesting that there is a minimal amount of matrixmaterial. The chemical nature of the cell wall matrix componentscould not be identified in the AFM images.

The AFM images allowed for the measurement of the cellu-lose microfibrils. Elementary fibrils with slightly variable diameterswere observed in the intermittent contact mode AFM images of thesugarcane cell wall. The distribution of the microfibril diameters isshown in Fig. 2c. The histogram shows that the distribution of fiberdiameters in the AFM images (n = 98) varied in a range of 15–21 nm;the average was 18.35 nm with a standard deviation of 1.49 nm.

The first AFM images were randomly collected, and they dis-played a considerable variation in the number overlaps, orientationand arrangement among the cellulose microfibrils. To addressthis variation, we imaged different regions, selected by lightmicroscopy, of the same parenchyma cell wall using AFM (Fig. 3).Low magnification images (Fig. 3b–e) highlighted the microfibrilorientation in four distinct regions of the same cell wall fragment.Even in this case, a disorganized texture was observed, and again,the oblique microfibril orientation appeared to be predominant.

3.3. The effect of thermochemical pretreatment

The effect of thermochemical pretreatment on the cell wall cel-lulose microfibrils was investigated by examining random AFMimages of parenchyma cell wall fragments. It was clear from theAFM images that the effect on the filaments was heterogeneous,

showing structural differences among cell wall regions after thepretreatment. In some cases, the cell wall was scarcely affectedby the pretreatment (Fig. 4a and b), i.e., the distribution of themicrofibrils remained unaltered and was indistinguishable from

Y. Abud et al. / Industrial Crops and Products 51 (2013) 62– 69 65

Fig. 2. The cell wall surface of the parenchyma cells in sugarcane by AFM. Topographic (a) and height (b) AFM images of the cell wall fragments prepared by cell fractionations ofibrilm

tmccasvdsaatrsmuwcw

ctf

and switch grass (Donohoe et al., 2011). To confirm the visualimpression of the lignin-rich globular structures in sugarcane, weinvestigated the lignin content in the globular structures by trans-mission electron microscopy (TEM) analysis of ultra-thin sections

Table 1Surface roughness measurements of the sulfuric acid-pretreated cell wall. The tabledepicts the results found in a set of AFM images of the untreated and sulfuric acid-treated (low, medium and high effect) cell wall regions presented in Figs. 2 and4.

Image RMS (nm) Representative images

howing the cellulose microfibril assembly in the different layers. Elementary micricrofibrils diameter is shown in (c).

he untreated samples (Fig. 2). No change in the density of theicrofibrils or in the microfibril orientation was observed. In other

ases, we obtained direct evidence of an intermediate effect, whichorresponded to a significant change in the organization of the fil-ments, including filament loss, and the appearance of globulartructures (Fig. 4c and d). The density of the globular structuresaried at the cell wall surface (data not shown); it was abun-ant in some samples and scarce in others. While the sphericalhape of the globular structures was nearly the same, the sizeppeared to differ between structures. AFM images were used toccurately measure the globular structures that appeared after thehermochemical treatment. The globular structures had diametersanging from 30.3 nm to 60.6 nm, with an average of 43.2 nm and atandard deviation of 6.5 nm. Another feature observed after ther-ochemical pretreatment was the striking change in the overall

ltrastructure of the cell wall (Fig. 4e and f). In this case, the cellall had undergone a considerable loss of fibrils; no discernible

ellulose microfibril could be identified, indicating that in some cellall regions, the treatment completely affected its integrity.

The surface roughness was also investigated using AFM toompare the changes in the plant cell wall before and after thehermochemical pretreatment. For quantitative analysis, the sur-ace roughness is of great interest as it is directly related to the

s were predominantly in a random orientation. The distribution of the elementary

light scattering due to the grain boundary and thickness. The sur-face roughness measurements showed an increase in the disorderof microfibrils, which is correlated with the intensity of the ther-mochemical treatment. The surface roughness factor (RMS or rootmean square) is presented in Table 1. The values showed a signifi-cant difference between the control cell wall and the different levelsof thermochemical treatment for p < 0.05 (Mann–Whitney test).

Lignin-rich globular structures have been observed to formfollowing pretreatments in corn stover (Donohoe et al., 2008)

Untreated 5.679 Fig. 2aTQ (low effect) 6.454 Fig. 4aTQ (medium effect) 8.069 Fig. 4bTQ (high effect) 9.188 Fig. 4c

66 Y. Abud et al. / Industrial Crops and Products 51 (2013) 62– 69

F hyma( strated e squa

siaitrtAo

4

Sawh

Fns

ig. 3. Cellulose microfibril organization in the cell wall surface of a single parencb)–(e) Topographic AFM images of the cell wall fragments showed in (a) demonisorganized arrangement, and the oblique orientation appears to predominate. Th

tained with KMnO4 (Sant’Anna et al., 2013). Fig. 5 shows a TEMmage of the treated cell wall stained with KMnO4, in which therere clearly several droplets protruding from the cell wall surface, ass evidenced by the AFM images. Moreover, globular structures inhe inner region of the cell wall and in the middle lamella, a ligninich region, were observed. The dark staining of the globular struc-ures observed in the TEM images indicates their lignin content.ltogether, these findings indicate the re-localization and removalf lignin into globular structures after the sulfuric acid treatment.

. Discussion

The chemical composition of the PCW, including sugarcane (De

ouza et al., 2012), is well-known. However, the actual chemicalnalysis does not consider the structural diversity of the cellall. The organization of the PCW investigated by microscopyas recently gained additional importance because of its ability

ig. 4. AFM analysis of the effect of sulfuric acid pretreatment on the sugarcane cell wall. (ao effect after pretreatment. (c) and (d) Region of the cell wall showing filament loss anignificantly affected by the cellulose microfibril organization following the sulfuric acid

fragment. (a) Light microscopy image of a cell wall isolated by cell fractionation. the distribution of the elementary microfibrils in the different layers. There is ares annotated as 1–4 in (a) correspond to the local AFM image in (b)–(e).

to investigate at the macro-, micro-, and nanoscale the cell walldeconstruction after pretreatment and enzymatic hydrolysis(McCann and Carpita, 2008; Chundawat et al., 2011; De Souza andSant’Anna, 2012). While there are a large number of microscopyreports describing the cell wall arrangement in other plants (forexample, Chundawat et al., 2011; Ding et al., 2012; Xu et al., 2006;Abdul Khalil et al., 2008, 2010; Ma et al., 2011), in sugarcane, thisstructure is well described mainly at the macroscopic level (DeSouza et al., 2012; Moore, 1987). While the understanding of theintegrity of the cell wall constituents at the microscopic level hasbeen reported (Sant’Anna et al., 2013), the nanoscopic level hasnever been investigated in detail.

AFM is a powerful tool for investigating the fibrillar archi-

tecture of the cell wall surface (Yarbrough et al., 2011). In thisstudy, the internodal sugarcane cell wall fragments were depositedonto mica, and the organization of the filamentous array innative (non-extracted) parenchyma cell wall was elucidated at the

) and (b) AFM topographic image demonstrating the cell wall regions with virtuallyd the formation of globular structures. (e) and (f) Region of the cell wall that waspretreatment. Note the absence of filaments.

Y. Abud et al. / Industrial Crops an

Fig. 5. Lignin localization in the sulfuric acid pretreated cell wall by ultrastructuralcytochemistry. (a) TEM image of untreated cell wall showing deposition of lignin bydark KMnO4 staining. (b) TEM image of the ultra-thin sections stained with KMnO4

showing lignin content of globular structures extruding from the cell wall surface.The dark lignin globules can also be observed in the inner region of cell wall. Lignin-rich middle lamella presenting heavy KMnO4 dark staining is indicated by arrows.(t

ntta(eaodta(rBaplmdt

2012). Similar structures were recently reported in corn stover

c) Higher magnification of the region of the cell wall surface depicted in (b) showinghe lignin globular structures extruding from the cell wall.

anoscale level using comprehensive AFM analysis. Using the pro-ocol described by McCann et al. (1990) we succeeded in isolatinghe sugarcane parenchyma cell wall, for the first time. The puritynd integrity of the fraction was assessed by light microscopyFig. 1b). AFM tapping mode images were acquired under moistnvironment conditions and we used parenchyma tissue of sug-rcane as a representative plant tissue because it is composedf large and relatively uniform cells. The sugarcane cell wall wasetermined to be a laminated fibrous structure in which elemen-ary microfibrils were randomly orientated (Fig. 2a and b), event the level of a single cell wall (Fig. 3). Using AFM, Kirby et al.1996) clearly demonstrated a distinct variation in the microfib-illar texture of the water chestnut, Amsterdamse bak carrot andintje potato. In comparison to such crops, we could not identify

similar pattern of cellulose microfibril organization in sugarcanearenchyma cells. Considering the general organization of the cel-

ulose microfibrils, these findings are indicative of variable cell wall

orphology between feedstocks. In terms of recalcitrance, such

ifference may contribute to the different levels of natural resis-ance that a cell wall has to deconstruction.

d Products 51 (2013) 62– 69 67

In the tapping mode AFM images, the diameter of the cellu-lose microfibrils was observed to vary slightly in the sugarcaneparenchyma cell wall surface. The measured elementary microfib-ril diameter varied from 15 nm to 21 nm, with an average diameterof about 19 nm (Fig. 2c). Our microfibril diameter results are consis-tent with cellulose fibril diameters (20–25 nm) found in other plantcell walls observed using the quick-freeze, deep-etch technique(McCann et al., 1990) or by AFM (Kirby et al., 1996; Morris et al.,1997). We cannot discard that the slight discrepancies between ourwork and other studies using AFM could be an effect of using dif-ferent crops, may be related to the tip size and shape, or may dueto differences in the scanning speed and cell wall features. In thisstudy, individual cellulose microfibrils with a diameter of approxi-mately 3 nm were not resolved (Davies and Harris, 2003; Xu et al.,2007; Ding et al., 2012).

An effective pretreatment method is one that enhances theaccessibility to cellulose and increases the solubility of the poly-mers to generate sugar monomers without forming inhibitorydegradation products of fermentation process (Hendriks andZeeman, 2009; Brodeur et al., 2011). A single pretreatment is notsuitable due to the structural diversity of biomass.

A detailed study on the sugarcane cell wall components afterdiluted acid pretreatment was conducted by Canilha et al. (2011).In the current study, we applied AFM to explore the morphologicalaspects of the pretreated parenchyma sugarcane cell wall surfaceto understand the nanoscale traits involved in the cell wall recalci-trance. Thermochemical pretreatment is generally more effective insolubilizing a large fraction of lignin and hemicelluloses, releasingan insoluble form and opening the crystalline cellulosic substrate(Mosier et al., 2005). The AFM image of the pretreated parenchymasugarcane cell wall showed that some of the measured cell wallareas, no effect could be found; the general morphology of thecell wall appeared undisturbed (Fig. 4a and b). In other cell wallregions, the loss of cellulose microfibrils and the formation of sev-eral globular structures were observed, indicating the removal ofmatrix components in the form of globular structures (Fig. 4c andd). The loss of the cell wall matrix content, as well as the increasein the matrix porosity following pretreatment, are crucial changesthat improve the ability of the cell wall to be digested by cellulases(Chundawat et al., 2011; Corrales et al., 2012). In other cell wallareas, the pretreatment had a substantial impact on the microfibrilstructure and arrangement (Fig. 4e and f). In conclusion, the fibrillarstructure of the cell wall was completely lost, confirming our pre-vious results using AFM analysis (De Souza and Sant’Anna, 2012).The sulfuric acid pretreatment has been used to efficiently removehemicelluloses and break down the cellulose into fermentable sug-ars (Manzoor et al., 2012). This characteristic may explain thecomplete loss of the cellulose microfibrils found in some regionsof cell wall. Our results point to the presence of cell wall regionswith more or less recalcitrance, even when the analysis is focusedon the cell wall surface of the same cell type. It may be possiblethat the cellulose microfibril orientation and lignin concentrationare related to the observed difference in the structural effect of thethermochemical pretreatment on the sugarcane cell wall surface.Further experiments should be conducted to determine whetherthe heterogeneity found after pretreatment is due to the cellu-lose microfibril organization and orientation, local matrix chemicalcomposition, or lignin complexity and concentration.

The presence of several globules spread all over the cell wallsurface after sulfuric acid pretreatment observed by AFM (Fig. 4cand d) is in agreement with our previous result obtained using highresolution scanning electron microscopy (De Souza and Sant’Anna,

after sulfuric acid and hot water pretreatment (Donohoe et al.,2008) and switch grass upon diluted acid, ammonia fiber expan-sion and liquid hot water pretreatments (Donohoe et al., 2011).

6 ops an

Tcer4g2wtsl

drltDmsst

fTtsah(

mrD2swstwui

A

ffiTNFR

R

A

A

A

B

B

8 Y. Abud et al. / Industrial Cr

he lignin content in the corn stover globular structures wasonfirmed by FTIR, NMR and microscopy techniques (Donohoet al., 2008). In sugarcane, the diameters of the globular structuresanged from 30.3 nm to 60.6 nm, with an average diameter of3.2 nm, as evidenced by AFM. In corn stover, the diameters of thelobular structures ranged from 5 nm to 10 �m (Donohoe et al.,008). This difference may be due to the lignin concentration, cellall structure and/or the pretreatment protocol employed. Note

hat while we examined the thin-walled parenchyma cells, thetudy performed in corn stover focused on the thick-walled highlyignified xylem and sclerenchyma.

The chemical nature of the globular structures was not fullyetermined in this study; however, by using morphological crite-ia and KMnO4 staining, we propose that they are composed ofignin. Potassium permanganate staining has been widely usedo visualize lignin in different cell wall types (Bland et al., 1971;onaldson, 1994) before (Sant’Anna et al., 2013) and after pretreat-ent (Donohoe et al., 2008). TEM observations (Fig. 5) provided

ome additional information that could not be obtained by AFM,pecifically the chemical nature of the globules formed after pre-reatment.

The surface roughness, shown by AFM, was also significantly dif-erent in the untreated and pretreated parenchyma cells (Table 1).he surface roughness factor increased with the severity of the pre-reatment. The small effect on the cell wall upon pretreatment isimilar to the untreated cell walls. However, in the more severelyffected sugarcane cell wall regions, the RMS is almost 2-foldigher, as was demonstrated in corn stover after AFEX treatmentChundawat et al., 2011).

Although AFM studies describing the arrangement of celluloseicrofibrils in plant cell walls of different feedstock have been

eported (Kirby et al., 1996; Morris et al., 1997; Ding et al., 2012;avies and Harris, 2003; Thimm et al., 2000; Ding and Himmel,006; Kirby, 2011), the current study is the first to show the detailedtructural information of the native and pretreated sugarcane cellall surfaces. Our findings regarding the macromolecular ultra-

tructure of the sugarcane cell wall provide insights that may helpo comprehend the organization and deconstruction of the cellall after pretreatment. This information is crucial to establish annderstanding of the molecular mechanisms of pretreatment to

mprove the production, research and technology of biofuel.

cknowledgments

This work was performed in collaboration and with supportrom the Centro de Pesquisas da Petrobras/Cenpes. In addition,nancial support was provided by the Ministério de Ciência eecnologia/Financiadora de Estudos e Projetos (Finep), Conselhoacional de Desenvolvimento Científico e Tecnológico (CNPq), andundac ão Carlos Chagas Filho de Amparo à Pesquisa do Estado doio de Janeiro (FAPERJ).

eferences

bdul Khalil, H.P.S., Siti Alwani, M., Ridzuan, R., Kamarudin, H., Khairul, A., 2008.Chemical composition, morphological characteristics, and cell wall structure ofMalaysian oil palm fibers. Polym. Plast. Technol. Eng. 47, 273–280.

bdul Khalil, H.P.S., Ireana Yusra, A.F., Bhat, A.H., Jawaid, M., 2010. Cell wall ultra-structure, anatomy, lignin distribution, and chemical composition of Malaysiancultivated kenaf fiber. Ind. Crop. Prod. 31, 113–121.

gbor, V.B., Cicek, N., Sparling, R., Berlin, A., Levin, D.B., 2011. Biomass pretreatment:fundamentals toward application. Biotechnol. Adv. 29, 675–685.

innig, G., Quate, C.F., Gerber, C.H., 1986. Atomic force microscope. Phys. Rev. Lett.56, 930–933.

land, D.E., Foster, R.C., Logan, A.F., 1971. The mechanism of permanganate andosmium tetroxide fixation and the distribution of the lignin in the cell wall ofPinus radiata. Holzforschung 25, 137–143.

d Products 51 (2013) 62– 69

Brodeur, G., Yau, E., Kimberly, B., Collier, J., Ramachandran, K.B., Ramakrishnan, S.,2011. Chemical and physicochemical pretreatment of lignocellulosic biomass:a review. Enzyme Res., http://dx.doi.org/10.4061/2011/787532.

Canilha, L., Santos, V.T., Rocha, G.J., Almeida e Silva, J.B., Giulietti, M., Silva, S.S., Felipe,M.G., Ferraz, A., Milagres, A.M., Carvalho, W.J.A., 2011. Study on the pretreat-ment of a sugarcane bagasse sample with dilute sulfuric acid. Ind. Microbiol.Biotechnol. 38, 1467–1475.

Chundawat, S.P.S., Donohoe, B.S., Sousa, L.C., Elder, T., Agarwal, U.P., Lu, F., Ralph, J.,Himmel, M.E., Balan, V., Dale, B.E., 2011. Multi-scale visualization and character-ization of lignocellulosic plant cell wall deconstruction during thermochemicalpretreatment. Energy Environ. Sci. 4, 973–984.

Corrales, R.C.N.R., Teixeira, M.F.M., Perrone, C.C., Sant’Anna, C., De Souza, W., Abud, Y.,Bon, E.P.S., Ferreira-Leitão, V., 2012. Structural evaluation of sugar cane bagassesteam pretreated in the presence of CO2 and SO2. Biotechnol. Biofuels 5, 36.

Davies, L.M., Harris, P.J., 2003. Atomic force microscopy of microfibrils in primarycell walls. Planta 217, 283–289.

De Souza, A.P., Leite, D.C.C., Pattathil, S., Hahn, M.G., Buckeridge, M.S.,2012. Composition and structure of sugarcane cell wall polysaccharides:implications for second-generation bioethanol production. Bioenerg. Res.,http://dx.doi.org/10.1007/s12155-012-9268-1.

De Souza, W., Sant’Anna, C., 2012. Microscopy as a tool to follow deconstruction oflignocellulosic biomass. In: Méndez-Vilas, A. (Ed.), Current Microscopy Contrib-utions to Advances in Science Technology. Formatex Research Center, Espanha,pp. 639–645.

Ding, S.Y., Himmel, M.E., 2006. The maize primary cell wall microfibril: a new modelderived from direct visualization. J. Agric. Food Chem. 54, 597–606.

Ding, S.Y., Liu, Y.S., Zeng, Y., Himmel, M.E., Baker, J.O., Bayer, E.A., 2012. How doesplant cell wall nanoscale architecture correlate with enzymatic digestibility?Science 338, 1055–1060.

Donaldson, L.A., 1994. Mechanical constraints on lignin deposition during lignifica-tion. Wood Sci. Technol. 28, 111–118.

Donaldson, L.A., 2001. Lignification and lignin topochemistry—an ultrastructuralview. Phytochemistry 57, 859–873.

Donohoe, B.S., Decker, S.R., Tucker, M.P., Himmel, M.E., Vinzant, T.B., 2008. Visu-alizing lignin coalescence and migration through maize cell walls followingthermochemical pretreatment. Biotechnol. Bioeng. 101, 913–925.

Donohoe, B.S., Selig, M.J., Viamajala, S., Vinzant, T.B., Adney, W.S., Himmel, M.E., 2009.Detecting cellulase penetration into corn stover cell walls by immuno-electronmicroscopy. Biotechnol. Bioeng. 103, 480–489.

Donohoe, B.S., Vinzant, T.B., Elander, R.T., et al., 2011. Surface and ultrastruc-tural characterization of raw and pretreated switchgrass. Bioenerg. Res. 102,11097–11104.

Goldemberg, S., 2007. Science 315, 808–810.Grabber, J.H., 2005. How do lignin composition, structure, and cross-linking affect

degradability? A review of cell wall model studies. Crop Sci. 45, 820–831.Hansma, H.G., 2001. Surface biology of DNA by atomic force microscopy. Annu. Rev.

Phys. Chem. 52, 71–92.Hendriks, A.T., Zeeman, G., 2009. Pretreatments to enhance the digestibility of lig-

nocellulosic biomass. Bioresour. Technol. 100, 10–18.Himmel, M.E., Ding, S.Y., Johnson, D.K., Adney, W.S., Nimlos, M.R., Brady, J.W., Foust,

T.D., 2007. Biomass recalcitrance: engineering plants and enzymes for biofuelsproduction. Science 315, 804.

Jordan, D.B., Bowman, M.J., Braker, J.D., Dien, B.S., Hector, R.E., Lee, C.C., Mertens, J.A.,Wagschal, K., 2012. Plant cell walls to ethanol. Biochem. J. 442, 241–252.

Kaczkowski, J., 2003. Structure, function and metabolism of plant cell wall. ActaPhysiol. Plant. 25, 287–305.

Keegstra, K., 2010. Plant cell walls. Plant Physiol. 154, 483–486.Kirby, A.R., 2011. Atomic force microscopy of plant cell walls. Methods Mol. Biol.

715, 169–178.Kirby, A.R., Gunning, A.P., Waldron, K.W., Morris, V.J., Ng, A., 1996. Visualization of

plant cell walls by atomic force microscopy. Biophys. J. 70, 1138–1143.Ma, J.F., Yang, G.H., Mao, J.Z., Xu, F., 2011. Characterization of anatomy, ultrastructure

and lignin microdistribution in Forsythia suspensa. Ind. Crop. Prod. 33, 358–363.Manzoor, A., Khokhar, Z.U., Uzma, A.H., Ahma, S.A., Syed, Q.A., Baig, S., 2012. Dilute

sulfuric acid: a cheap acid for optimization of bagasse pretreatment. Sci. Int.(Lahore) 24, 41–45.

McCann, M.C., Carpita, N.C., 2008. Designing the deconstruction of plant cell walls.Curr. Opin. Plant Biol. 11, 314–320.

McCann, M.C., Wells, B., Roberts, K., 1990. Direct visualization of cross-links in theprimary plant cell wall. J. Cell Sci. 96, 323–334.

Moore, P.H., 1987. Anatomy and morphology. In: Heinz, D.J. (Ed.), SugarcaneImprovement Through Breeding. Elsevier, Amsterdam, pp. 273–311.

Morris, V.J., Gunning, A.P., Kirby, A.R., Round, A., Waldron, K., Ng, A., 1997. Atomicforce microscopy of plant cell walls, plant cell wall polysaccharides and gels. Int.J. Biol. Macromol. 21, 61–66.

Mosier, N., Wyman, C.E., Dale, B.E., Elander, R.T., Lee, Y.Y., Holtzapple, M., Ladisch,M., 2005. Features of promising technologies for pretreatment of lignocellulosicbiomass. Bioresour. Technol. 96, 673–686.

Müller, S.A., Müller, D.J., Engel, A., 2011. Assessing the structure and function of singlebiomolecules with scanning transmission electron and atomic force micro-scopes. Micron 42, 186–195.

Perez, S., Mazeau, K., 2004. Conformation, structure and morphologies of cellu-lose. In: Dimitriu, S. (Ed.), Polysaccharides: Structural Diversity and FunctionalVersatility. Marcel Dekker, New York, pp. 41–68.

Saha, B.C., 2003. Hemicellulose bioconversion. J. Ind. Microbiol. Biotechnol. 30,279–291.

ops an

S

S

T

Y. Abud et al. / Industrial Cr

arkar, P., Bosneaga, E., Auer, M., 2009. Plant cell walls throughout evolution:towards a molecular understanding of their design principles. J. Exp. Bot. 13,3615–3635.

ant’Anna, C., Cost, L.T., Abud, Y., Biancatto, L., Miguens, F.C., De Souza, W., 2013.

Sugarcane cell wall structure and lignin distribution investigated by confocaland electron microscopy. Microsc. Res. Tech. 76, 829–834.

himm, J.C., Burritt, D.J., Ducker, W.A., Melton, L.D., 2000. Celery (Apium graveolens L.)parenchyma cells examined by atomic force microscopy: effect of dehydrationon cellulose microfibrils. Planta 212, 25–32.

d Products 51 (2013) 62– 69 69

Xu, F., Zhong, X.C., Sun, R.C., Lu, Q., 2006. Anatomy, ultrastructure and lignin distri-bution in cell wall of Caragana Korshinskii. Ind. Crop. Prod. 24, 186–193.

Xu, P., Donaldson, L.A., Gergely, Z.R., Staehelin, L.A., 2007. Dual-axis electron tomo-graphy: a new approach for investigating the spatial organization of wood

cellulose microfibrils. Wood Sci. Technol. 41, 101–116.

Xu., P., Liu, H., Donaldson, L.A., 2011. Mechanical performance and cellulosemicrofibrils in wood with high S2 microfibril angles. J. Mater. Sci. 46, 534–540.

Yarbrough, J.M., Himmel, M.E., Ding, S.Y., 2011. Plant cell wall characterization usingscanning probe microscopy techniques. Bioenerg. Res. 102, 11097–11104.