Effects of Plant Cell Wall Matrix Polysaccharides on Bacterial

The crystallinity, allomorph content, and mesoscale ordering of cellulose produced by Gluconacetobacter xylinus cultured with different plant cell wal...
2 downloads 0 Views 1MB Size
Article pubs.acs.org/Biomac

Effects of Plant Cell Wall Matrix Polysaccharides on Bacterial Cellulose Structure Studied with Vibrational Sum Frequency Generation Spectroscopy and X‑ray Diffraction Yong Bum Park,*,† Christopher M. Lee,‡ Kabindra Kafle,‡ Sunkyu Park,§ Daniel J. Cosgrove,† and Seong H. Kim*,‡ †

Department of Biology and ‡Department of Chemical Engineering and Materials Research Institute, Pennsylvania State University, University Park, Pennsylvania 16802, United States § Department of Forest Biomaterials, North Carolina State University, Raleigh, North Carolina 27695, United States S Supporting Information *

ABSTRACT: The crystallinity, allomorph content, and mesoscale ordering of cellulose produced by Gluconacetobacter xylinus cultured with different plant cell wall matrix polysaccharides were studied with vibrational sum frequency generation (SFG) spectroscopy and X-ray diffraction (XRD). Crystallinity and ordering were assessed as the intensity of SFG signals in the CH/CH2 stretch vibration region (and confirmed by XRD), while Iα content was assessed by the relative intensity of the OH stretch vibration at 3240 cm−1. A key finding is that the presence of xyloglucan in the culture medium greatly reduced Iα allomorph content but with a relatively small effect on cellulose crystallinity, whereas xylan resulted in a larger decrease in crystallinity with a relatively small decrease in the Iα fraction. Arabinoxylan and various pectins had much weaker effects on cellulose structure as assessed by SFG and XRD. Homogalacturonan with calcium ion reduced the SFG signal, evidently by changing the ordering of cellulose microfibrils. We propose that the distinct effects of matrix polysaccharides on cellulose crystal structure result, at least in part, from selective interactions of the backbone and side chains of matrix polysaccharides with cellulose chains during the formation of the microfibril.



Whitney et al.19 reported that the presence of XyG in the culture medium selectively reduced the Iα fraction, whereas Bootten et al.23 reported that XyG did not affect the Iα/Iβ ratio. The effects of XyG and other cell wall matrix polymers on cellulose structure could be studied more easily and accurately with an analytical technique that is capable of selective detection of crystalline cellulose without interference from hemicellulose and pectin.26 Recently, we reported that vibrational sum frequency generation (SFG) spectroscopy can provide such advantages.27−34 SFG is a nonlinear optical process that takes place in an optical medium that lacks centrosymmetry. This makes SFG capable of selective detection of cellulose crystals imbedded in a complex cell wall containing hemicelluloses and pectins. For structural analysis of naturally produced cellulose Iα and Iβ, it is important to note that different techniques are sensitive to different structural aspects at different length scales. A brief summary of the distinctive detection rules of SFG, 13C solidstate NMR, and powder-XRD are summarized in Table S1 in

INTRODUCTION Cellulose is the most abundant biopolymer on earth and a major constituent of plant cell walls.1 Cellulose is produced as two crystalline allomorphs referred to as cellulose Iα and Iβ. Iα is the major form of cellulose synthesized by algae and bacteria, while Iβ is the dominant allomorph in land plants.2−5 In numerous studies, cell wall matrix polysaccharides have been implicated as modifiers of cellulose structure,6−14 but their effects on cellulose microfibril formation are not fully understood.15,16 This is in part due to the complexity of plant cell walls. As a simpler model system, the cellulose pellicle produced by Gluconacetobacter xylinus (G. xylinus) has been used to explore the interactions between cellulose and matrix polysaccharides.8,17−23 When G. xylinus is grown in the presence of wall matrix polymers, the bacterial cellulose becomes closely associated with the added polymers.19 These bacterial cellulose composites could provide some insights about how matrix polymers affect cellulose crystallization in plant cell walls and how cellulose interacts with matrix polysaccharides to form complex cell walls.18,24 Xyloglucan (XyG) is the main hemicellulose component of primary cell walls in dicots and nongraminaceous monocots,25 and its effects on cellulose crystal structure have been studied for the G. xylinus system, but with contradictory conclusions. © 2014 American Chemical Society

Received: April 15, 2014 Revised: May 19, 2014 Published: May 20, 2014 2718

dx.doi.org/10.1021/bm500567v | Biomacromolecules 2014, 15, 2718−2724

Biomacromolecules

Article

Figure 1. Hemicellulose (a) and pectin (b) contents of bacterial composites formed with polysaccharides in the culture medium. All pectin concentrations in the culture medium are 0.5% (w/v). Error bars are ±standard errors of the means (SEM, n = 3). XyG = xyloglucan, AX = arabinoxylan, Ara = arabinan, Gal = galactan, PGA = polygalacturonic acid. mass fraction of tamarind XyG or citrus PGA from 0 to 100% (Figure S1). Sugar Analysis. To extract noncellulosic polysaccharides bound to bacterial cellulose, 2 mg of a dry sample was treated twice with 1.5 mL of 4 M NaOH containing 20 mM NaBH4 for 6 h at room temperature (RT) with agitation. The collected supernatant was neutralized with diluted acetic acid, filtered (0.45 μm), dialyzed, and dried under vacuum. The alkali-extracted matrix polysaccharides were hydrolyzed in 1 mL of 72% sulfuric acid at RT for 40 min with shaking, diluted 10 times with ddH2O, and total sugar contents were measured by the phenol-sulfuric acid method.38 Figure 1 summarizes the actual amount of hemicellulose and pectin incorporated into the bacterial cellulose pellicle at their given concentrations in the culture media. Note that pectins were incorporated to a lower extent than were hemicelluloses. SFG Spectroscopy Measurements. SFG spectroscopy was performed for freeze-dried and pressed pellets of bacterial cellulose in reflection mode in ambient conditions as described.27,28 A modelocked Nd:YAG laser (EKSPLA, Vilnius, Lithuania) generated a < 30 ps pulse of 1064 nm at a 10 Hz repetition rate. The 532 nm visible excitation pulse was generated by frequency doubling. The 2.6−3.7 μm (2700−3800 cm−1) infrared pulse with 90%) was from Sigma-Aldrich. After 4 days fermentation, bacterial pellicles were scraped with a razor blade to remove bacteria and nonpellicle residues, washed with 2.6% sodium hypochlorite for 5 min, washed thoroughly with distilled deionized water (ddH2O), ground in liquid nitrogen, freeze-dried, and pressed into pellets for SFG and X-ray diffraction (XRD) measurements. To construct reference calibration curves, ground and dried bacterial cellulose was physically mixed with varying 2719

dx.doi.org/10.1021/bm500567v | Biomacromolecules 2014, 15, 2718−2724

Biomacromolecules

Article

Figure 2. SFG spectra of bacterial cellulose/hemicellulose composites. Bacteria were cultured in the presence of (a) xyloglucan (XyG), (b) xylan, and (c) arabinoxylan (AX). Percentage values indicate hemicellulose contents in the bacteria culture medium; a.u.: arbitrary unit. the volume fraction of crystalline cellulose in the probe volume.28 It is difficult to analyze theoretically the cellulose SFG intensity due to light scattering and nonlinear nature of the SFG process.29 Thus, we prepared calibration samples by physically mixing bacterial cellulose with known amounts of XyG or PGA, measured their SFG intensities, and created calibration curves (Figure S1). In this analysis, the sum of the intensities of the alkyl SFG peaks at 2920, 2944, and 2968 cm−1 was plotted as a function of the amount of cellulose in the reference sample using the calibration curve. The calibration curves of Figure S1a,b were used to estimate the amount of crystalline cellulose in cellulose/hemicellulose and cellulose/pectin composites. The hydroxyl/alkyl ratio of fitted peak area was calculated with the equation: [{(A3240 + A3270 + A3300 + A3320 + A3370)/5}/{(A2920 + A2944 + A2968)/ 3}] (Figures S1−S3). XRD Measurements. Crystalline structure and crystallinity index of cellulose were assessed by XRD using a Rigaku (Tokyo, Japan) Ultima IV diffractometer with CuKα radiation having a wavelength λ(Kα1) = 0.15406 nm generated at 40 kV and 44 mA. The diffraction intensities of freeze-dried samples placed on a quartz substrate were measured in the 2θ range of 8 to 42° using a step size of 0.02° at a rate of 2°/min. Cellulose crystallinity was estimated based on the amorphous subtraction method.39 An amorphous standard, prepared by dissolving ball-milled cellulose in dimethyl sulfoxide with paraformaldehyde followed by precipitation in an alkoxide solution,39,40 was used to subtract the amorphous portion from the diffraction profiles. A scale factor was applied to the amorphous standard spectrum, so that the baseline between the crystalline XRD peaks was zero without any negative values after subtraction of the amorphous portion. Cellulose crystal size (D) in the perpendicular to the (200) plane (∼22.8°) was determined using the Scherrer equation with the assumption that peak broadening is caused by a decrease in crystal size:

vibration region, three main peaks were detected at 2920, 2944, and 2968 cm−1.31,33 The 2944 cm−1 peak is putatively assigned to the CH2 stretching vibration of the hydroxymethylene (C6H2OH) group of cellulose in the tg conformation.29 The 2920 and 2968 cm−1 peaks are more prominent in the randomly arranged structure of cellulose microfibrils observed in algal cell walls and tunicate, while the 2944 cm−1 peak is dominant in the antiparallel-packed cellulose Iβ samples shown in plant secondary cell walls.33 There were some small variations in the relative intensities of these alkyl stretch peaks from different cell cultures that were initiated at different days, even when identical culture conditions were used (Figure S2). This variation of the alkyl stretch region was thought to be due to local differences in the size, packing, bundling, or interfibril spacing of cellulose microfibrils in bacterial pellicles.33 More details are currently being studied. In the OH stretch region, six components at 3240, 3270, 3300, 3320, 3370, and 3450 cm−1 can be identified in peak deconvolution (Figure S3).33 Except the 3450 cm−1 peak, all others are believed to originate from the crystalline cellulose. The origin of the peak at 3450 cm−1 is not clear at this point, but it is likely to come from weakly hydrogen-bonded hydroxyl groups. We speculate that these hydroxyl groups are present at the surface region or between crystalline phases where cellulose chains may have interactions with hemicellulose or water molecules. Unlike the peaks in the CH/CH2 region, the relative intensities of the crystalline OH components (3240, 3270, 3300, 3320, and 3370 cm−1) did not show significant changes among the control samples (Figure S2). The 3240 and 3270 cm−1 peaks are characteristic of cellulose Iα and Iβ, respectively.13,33 Thus, deconvolution of these two peaks (Figure S3) can be used to estimate the relative abundance of cellulose Iα and Iβ allomorphs in the sample.33 The intensity of the 3370 cm−1 peak varies concomitantly with the 3240 cm−1 peak; thus, it is considered to be specific cellulose Iα.33 However, this peak was not included in the polymorphic abundance analysis since cellulose Iβ does not have any peak equivalent to this additional peak. The intensity ratio of the alkyl peaks to the OH peaks was nearly constant at 1:2 (Figure S4). 2. Changes in Bacterial Cellulose Structure by Hemicelluloses. The SFG spectra of the bacterial pellicles obtained after 4 days culture in the presence of hemicelluloses (XyG, xylan, and arabinoxylan (AX)) are shown in Figure 2. The SFG signal intensity decreased with increasing hemi-

D = k · λ /(β · cos θ ) where k is the Scherrer constant (0.84), λ is the X-ray wavelength (0.154 nm), β is the full-width at half-maximum (fwhm) at (200) plane, and θ is the Bragg angle.41



RESULTS AND DISCUSSION 1. SFG Spectra of Bacterial Cellulose Produced without Hemicellulose or Pectin in the Culture Medium. We first characterized the SFG spectra of pellicles produced without hemicellulose or pectin. The spectral regions of interest in SFG analyses were the alkyl stretch (2800−3000 cm−1) and hydroxyl stretch (3000−3500 cm−1) regions. A representative spectrum of bacterial cellulose produced without any interactions with matrix polysaccharides is shown as the control in Figure 2. In the CH and CH2 (alkyl) stretch 2720

dx.doi.org/10.1021/bm500567v | Biomacromolecules 2014, 15, 2718−2724

Biomacromolecules

Article

Figure 3. SFG peak intensity of crystalline cellulose structure of bacterial cellulose/hemicellulose composites. (a) Alkyl asymmetric vibration region (2920, 2944, and 2968 cm−1) of bacterial cellulose/hemicellulose. (b) Cellulose Iα fraction was calculated using SFG peak intensity ratio of 3240 cm−1/(3240 + 3270 cm−1). Error bar is SEM (3 ≤ n ≤ 4). Some error bars are too small to be visible.

Although xylan reduced the crystallinity more than XyG, it did not alter the Iα/Iβ ratio as substantially as the XyG case. This result suggests that the selective disruption of the cellulose Iα allomorph during cellulose crystallization is a distinctive effect of XyG. Hemicelluloses with different backbone structure (XyG vs xylan) and side-chain substitution (xylan vs AX) showed distinctive effects on the crystallization of the cellulose produced by G. xylinus. Consistent with previous work,42 XyG and xylan reduced the crystallization of bacterial cellulose. In contrast, AX (xylan with arabinofucosyl side chains) as well as neutral pectins, and PGA (without Ca2+ ions; as shown in section 3) hardly reduced SFG intensities of the cellulose produced by bacteria. These results suggest that XyG and xylan have greater abilities to modulate cellulose microfibril structure than do the other cell wall matrix polysaccharides tested in this study. Although the amounts of XyG and xylan incorporated into the cellulose pellicle are similar, their effects on cellulose crystallization differ. XyG reduced the Iα/Iβ ratio more significantly than xylan, while xylan reduced the crystallinity more than XyG. XyG consists of a backbone of β-(1,4)-linked glucopyranose residues, similar to the cellulose backbone; the xylan backbone is very similar, the principal difference being the absence of the exocyclic C6 group. As both polymers are able to bind to cellulose,42 our data suggest that the ability to modulate the cellulose allomorph during crystallization requires binding of the hemicellulose to the nascent microfibril. Consistent with these results, other C6 polysaccharides such as mannan and carboxymethyl cellulose (CMC) also greatly reduce the Iα/Iβ ratio of bacteria cellulose.43,44 In contrast to xylan, AX in the culture medium did not significantly decrease bacterial cellulose crystallinity. Similar to our data (Figure 2), Iwata et al.42 reported that arabinoglucuronoxylan in the culture medium showed a lower binding affinity to cellulose in a bacteria composite than glucuronoxylan. Likewise, a highly substituted arabinoxylan showed a decrease in adsorption to cellulose.45−47 These results support our observation that the high degree of side-chain substitution to xylan lowers the binding to cellulose, weakening its impact on the crystallization of bacterial cellulose. Both XyG and AX are heteropolymers containing side-chain sugars, but XyG more greatly reduced the crystal size of bacterial cellulose than did AX (Figure S8e). Smith et al.48 reported that XyG-rich cell walls (onion and cabbage) had

cellulose concentration in the culture medium (Figure 2), with substantial changes for xylan and XyG and only marginal effects of AX. The hydroxyl/alkyl SFG peak intensity ratio remained constant (Figure S4), but the 3240 cm−1 (Iα) component was more reduced than the 3270 cm−1 (Iβ) component, especially for the XyG case. Figure 3a shows the relative cellulose crystallinity estimated by comparing the SFG intensity in the alkyl stretch region (2920, 2944, 2968 cm−1) with a standard calibration curve made with physical mixtures of bacterial cellulose and XyG (Figure S1a). The calibration curve accounts for the attenuation of the IR beam by hemicellulose and the nonlinear dependence of SFG intensity on cellulose concentration. This method assumes that the physical density of the compressed pellets used for the measurements is the same for all samples. The incorporation of hemicellulose reduced the SFG peak intensities of cellulose, which we interpret as largely due to a reduction in crystallinity (see also Results and Discussion of this point later). Although the amount of XyG and xylan bound to cellulose was similar (Figure 1a), the cellulose crystallinity reduction effect was greater for xylan than XyG (Figure 3a). At ∼20% (w/w) binding, xylan reduced the crystallinity by ∼55% compared to the control sample, while XyG reduced it by ∼35% at the same binding. In contrast, AX had only a minor effect, implying that AX does not have strong affinity to cellulose (Figure 3a). These results indicate that xylan and XyG affect the cellulose crystallization process more strongly than does AX. The significant reduction of cellulose crystallinity (crystalline content) by XyG and xylan but not by AX was confirmed by the XRD analyses of the same samples (Figure S8). Figure 3b shows the cellulose Iα fraction in these composites, calculated from the deconvolution of the 3240 (Iα) and 3270 cm−1 (Iβ) peaks (Figures S5−S7).33 Bacterial cellulose typically consists of ∼70% Iα and ∼30% Iβ allomorphs.19 The values determined from the SFG data of the control samples are slightly lower than the literature values, but they do not deviate significantly.33 Previous work reported that the Iα fraction was reduced by XyG in the culture medium,17,19 but a more recent study reported that the effect of XyG on the Iα/Iβ ratio was negligible.23 Our SFG analysis showed that the Iα fraction indeed decreased from ∼60 to ∼20% (Figure 3b) at a binding of ∼20% (w/w) XyG into the pellicle (Figure 2b). It is notable that the effects of xylan and AX on the selective reduction of cellulose Iα are not as strong as the XyG case (Figure 3b). 2721

dx.doi.org/10.1021/bm500567v | Biomacromolecules 2014, 15, 2718−2724

Biomacromolecules

Article

Figure 4. SFG spectra of bacterial cellulose/pectin composites. (a) Bacteria were cultured in the presence of 0.5% PGA and a variety of CaCl2 concentrations. (b) Bacteria were cultured in the presence of 0.5% neutral pectins. Percentage values indicate pectin contents in the bacteria culture medium; a.u.: arbitrary unit.

Figure 5. SFG peak intensity of crystalline cellulose structure of bacterial cellulose/pectin composites. (a) SFG peak intensity of C6H2 asymmetric vibration region (2920, 2944, and 2968 cm−1) of bacterial cellulose/pectin composites. (b) Cellulose Iα fraction was calculated using SFG peak intensity ratio of 3240 cm−1/(3240 + 3270 cm−1). Values in parentheses indicate CaCl concentration (mM) in the culture medium. Error bar is SEM (3 ≤ n ≤ 4). Some error bars are too small to be visible.

Table 1. Summary of the Changes of SFG, NMR, and XRD Signals for Bacterial Composites Produced in the Presence of XyG (Hemicellulose) and PGA (Pectin)a techniques wall polysaccharides signal changes for crystalline cellulose a

SFG XyG large reduction

NMR PGA small reduction

XyG19 large reduction

XRD PGA22 no reduction

XyG large reduction

PGA small reduction

NMR data were published results of Whitney et al. (1995) for XyG and Chanliaud and Gidley (1999) for PGA.

of the bacterial cellulose (Figures 4a and 5a). Also, the changes in the SFG intensity as well as the Iα/Iβ ratio were insignificant (Figures 5b and S9). In the presence of Ca2+ and PGA in the culture medium, the SFG intensities of the bacterial cellulose were substantially reduced (Figures 4a and 5a). This effect could arise from a reduction of cellulose crystallinity or mesoscale ordering. Ca2+ alone (without PGA) did not change the bacterial cellulose SFG spectrum (Figure S10). As Ca2+ concentration increased from 0.1 to 2 mM, the amount of PGA incorporated into the bacterial cellulose pellicle slightly increased from ∼1 to ∼3.5% (Figure 1b). The reduced SFG intensity of bacterial cellulose by PGA + Ca2+ was accompanied by a reduction of XRD

smaller cellulose crystal size than GAX-rich walls (ryegrass and pineapple). These results are consistent with our observations of a greater reduction of cellulose crystal size by XyG, when compared with AX. This effect may be a common process in both plant cell walls and bacterial composites. 3. Changes in Bacterial Cellulose Structure by Pectins. To examine the effect of pectins on the bacterial cellulose structure, G. xylinus was cultured in the presence of pectins. Polygalacturonic acid (PGA) is the largest component of pectins in plant primary cell walls.49 In the absence of CaCl2 in the culture medium, the PGA binding to the bacterial cellulose was less than 0.1% (w/w; Figure 1b); thus, the presence of PGA in the culture medium hardly altered the SFG intensities 2722

dx.doi.org/10.1021/bm500567v | Biomacromolecules 2014, 15, 2718−2724

Biomacromolecules



crystallinity and crystal size (Figure S11), whereas a change in crystallinity was not detected by 13C NMR (Table 1).22 From this comparison, it appears that SFG and XRD detect structural features of cellulose not apparent by NMR, probably from differences in packing (bundling) of cellulose microfibrils at a length scale beyond the range of NMR measurements. It is well-known that Ca2+ ions can cross-link PGA molecules in the solution, forming a gel-like network. When cellulose microfibrils are extruded into this solution, the Ca2+ cross-linked PGA network may disrupt cellulose microfibril aggregation, resulting in changes of SFG signal intensity. This inference is consistent with the fact that the SFG intensity is sensitive to cellulose crystallinity as well as microfibril packing and aggregation (Table S1).33,34,50 As discussed earlier, hemicelluloses can indeed change the crystalline structure of cellulose microfibrils,19 resulting in changes of the Iα/Iβ ratio (Figure 3b). Changes in the packing and alignment of cellulose microfibrils can be monitored with the SFG hydroxyl/alkyl intensity ratio in plant cell walls.34 However, we did not observe a substantial difference in the hydroxyl/alkyl ratio of the bacterial composites cultured with PGA (with Ca2+) or hemicelluloses (Figure S4). If we understand how cellulose crystallinity and intermicrofibril aggregation in a bacterial composite differently contribute to SFG spectra, this could provide insightful information about how hemicelluloses and pectins differently influence cellulose structure and formation in plant cell walls. Neutral pectins (e.g., arabinan and galactan) bind to cellulose, although not as strongly as do XyG and xylan.51−54 In our experiments, arabinan and galactan were incorporated up to ∼1% (w/w) in the bacterial composite (Figure 1b), yet the associated reductions of SFG intensity and Iα/Iβ ratio were negligible (Figures 5b and S12). These results indicate that the interaction of neutral pectins with cellulose is not strong enough to change microfibril crystallization, aggregation, or packing, at least in the bacterial pellicles. This contrasts with the stronger binding of XyG and xylan, leading to distinctive effects on cellulose crystallization and packing. It is likely that similar physical effects occur in the plant cell wall, leading to distinctive hemicellulose−cellulose interactions in primary cell walls (which contain XyG)55,56 and secondary cell walls (which contain xylan).57



Article

ASSOCIATED CONTENT

S Supporting Information *

Summary of the detection phenomenon and obtainable information on SFG, NMR, and XRD, SFG calibration curves of bacterial cellulose, SFG hydroxyl/alkyl ratio, peak fitting of SFG spectra, and XRD measurements of cellulose/pectin composites. This material is available free of charge via the Internet at http://pubs.acs.org.



AUTHOR INFORMATION

Corresponding Authors

*Phone: 814-865-3752. E-mail: [email protected]. *Phone: 814-863-4809. E-mail: [email protected]. Notes

The authors declare no competing financial interest.



ACKNOWLEDGMENTS This work was supported by The Center for Lignocellulose Structure and Formation, an Energy Frontier Research Center funded by the U.S. Department of Energy, Office of Science, and Office of Basic Energy Sciences under Award Number DESC0001090. The authors acknowledged Dr. Bon-Wook Koo (North Carolina State University) for XRD analysis of bacterial composite samples.



REFERENCES

(1) Pauly, M.; Keegstra, K. Plant J. 2008, 54, 559−568. (2) Atalla, R. H.; Vanderhart, D. L. Science 1984, 223, 283−285. (3) Jarvis, M. Nature 2003, 426, 611−612. (4) Nishiyama, Y.; Langan, P.; Chanzy, H. J. Am. Chem. Soc. 2002, 124, 9074−9082. (5) Nishiyama, Y.; Sugiyama, J.; Chanzy, H.; Langan, P. J. Am. Chem. Soc. 2003, 125, 14300−14306. (6) Agarwal, U. P.; Reiner, R. S.; Ralph, S. A. Cellulose 2010, 17, 721−733. (7) Chen, L.; Carpita, N. C.; Reiter, W. D.; Wilson, R. H.; Jeffries, C.; McCann, M. C. Plant J. 1998, 16, 385−392. (8) Bootten, T. J.; Harris, P. J.; Melton, L. D.; Newman, R. H. Biomacromolecules 2009, 10, 2961−2967. (9) Bootten, T. J.; Harris, P. J.; Melton, L. D.; Newman, R. H. J. Exp. Bot. 2004, 55, 571−583. (10) Dick-Pérez, M.; Zhang, Y.; Hayes, J.; Salazar, A.; Zabotina, O. A.; Hong, M. Biochemistry 2011, 50, 989−1000. (11) Fernandes, A. N.; Thomas, L. H.; Altaner, C. M.; Callow, P.; Forsyth, V. T.; Apperley, D. C.; Kennedy, C. J.; Jarvis, M. C. Proc. Natl. Acad. Sci. U.S.A. 2011, 108, E1195−203. (12) Sturcova, A.; His, I.; Apperley, D. C.; Sugiyama, J.; Jarvis, M. C. Biomacromolecules 2004, 5, 1333−1339. (13) Sugiyama, J.; Persson, J.; Chanzy, H. Macromolecules 1991, 24, 2461−2466. (14) Sugiyama, J.; Vuong, R.; Chanzy, H. Macromolecules 1991, 24, 4168−4175. (15) Uhlin, K. I.; Atalla, R. H.; Thompson, N. S. Cellulose 1995, 2, 129−144. (16) Atalla, R. H.; Hackney, J. M. Int. J. Biol. Macromol. 1993, 15, 109−112. (17) Whitney, S. E.; Willson, E.; Webster, J.; Bacic, A.; Reid, J. S. G.; Gidley, M. J. Am. J. Bot. 2006, 93, 1402−1414. (18) Whitney, S. E.; Gidley, M. J.; McQueen-Mason, S. J. Plant J. 2000, 22, 327−334. (19) Whitney, S. E.; Brigham, J. E.; Darke, A. H.; Reid, J. S. G.; Gidley, M. J. Plant J. 1995, 8, 491−504. (20) Chanliaud, E.; Burrows, K. M.; Jeronimidis, G.; Gidley, M. J. Planta 2002, 215, 989−996. (21) Chanliaud, E.; De Silva, J.; Strongitharm, B.; Jeronimidis, G.; Gidley, M. J. Plant J. 2004, 38, 27−37.

CONCLUSION

The noncentrosymmetry constraint and phase synchronizing condition of SFG process allows detection of crystalline cellulose and its distribution in bacterial cellulose composites containing hemicelluloses or pectins. Since the crystallinity and intermicrofibril packing of cellulose change SFG signal intensities, we used SFG spectra to study how hemicelluloses and pectins influence cellulose crystallinity and mesoscale assembly. Our results showed that XyG decreased the cellulose Iα/Iβ allomorph ratio, whereas neither xylan nor AX had this effect. We also found that PGA with Ca2+ decreased the intensities of SFG and XRD without changing the Iα/Iβ ratio, an effect attributed to reduced intermicrofibril aggregation and packing of cellulose. These results offer insights into the mechanisms by which cell wall matrix polysaccharides modulate cellulose structure and assembly in plant cell walls. 2723

dx.doi.org/10.1021/bm500567v | Biomacromolecules 2014, 15, 2718−2724

Biomacromolecules

Article

(22) Chanliaud, E.; Gidley, M. J. Plant J. 1999, 20, 25−35. (23) Bootten, T. J.; Harris, P. J.; Melton, L. D.; Newman, R. H. Carbohydr. Res. 2008, 343, 221−229. (24) Whitney, S. E.; Gothard, M. G.; Mitchell, J. T.; Gidley, M. J. Plant Physiol. 1999, 121, 657−664. (25) Carpita, N. C.; Gibeaut, D. M. Plant J. 1993, 3, 1−30. (26) Kim, S. H.; Lee, C. M.; Kafle, K. Korean J. Chem. Eng. 2013, 30, 2127−2141. (27) Barnette, A. L.; Bradley, L. C.; Veres, B. D.; Schreiner, E. P.; Park, Y. B.; Park, J.; Park, S.; Kim, S. H. Biomacromolecules 2011, 12, 2434−2439. (28) Barnette, A. L.; Lee, C. M.; Bradley, L. C.; Schreiner, E. P.; Park, Y. B.; Shin, H.; Cosgrove, D. J.; Park, S.; Kim, S. H. Carbohydr. Polym. 2012, 89, 802−809. (29) Lee, C. M.; Mohamed, N. M.; Watts, H. D.; Kubicki, J. D.; Kim, S. H. J. Phys. Chem. B 2013, 117, 6681−6692. (30) Handakumbura, P. P.; Matos, D. A.; Osmont, K. S.; Harrington, M. J.; Heo, K.; Kafle, K.; Kim, S. H.; Baskin, T. I.; Hazen, S. P. BMC Plant Biol. 2013, 13, 1−16. (31) Kafle, K.; Xi, X.; Lee, C. M.; Tittmann, B. R.; Cosgrove, D. J.; Park, Y. B.; Kim, S. H. Cellulose 2014, 21, 1075−1086. (32) Kafle, K.; Greeson, K.; Lee, C. M.; Kim, S. H. Text. Res. J. 2014, DOI: 10.1177/0040517514527379. (33) Lee, C. M.; Kafle, K.; Park, Y. B.; Kim, S. H. Phys. Chem. Chem. Phys. 2014, 16, 10844−10853. (34) Park, Y. B.; Lee, C. M.; Koo, B.-W.; Park, S.; Cosgrove, D. J.; Kim, S. H. Plant Physiol. 2013, 163, 907−913. (35) Kim, S. H.; Lee, C. M.; Kafle, K.; Park, Y. B.; Xi, X. SPIE Opt. Eng. Appl. 2013, 884508−884508. (36) Park, S.; Baker, J. O.; Himmel, M. E.; Parilla, P. A.; Johnson, D. K. Biotechnol. Biofuels 2010, 3, 10. (37) Hestrin, S.; Schramm, M. Biochem. J. 1954, 58, 345−352. (38) Dubois, M.; Gilles, K.; Hamilton, J. K.; Rebers, P. A.; Smith, F. Anal. Chem. 1956, 28, 350−356. (39) Park, S.; Johnson, D. K.; Ishizawa, C. I.; Parilla, P. A.; Davis, M. F. Cellulose 2009, 16, 641−647. (40) Schroeder, L. R.; Gentile, V. M.; Atalla, R. H. J. Wood Chem. Technol. 1986, 6, 1−14. (41) Nishiyama, Y.; Johnson, G. P.; French, A. D. Cellulose 2012, 19, 319−336. (42) Iwata, T.; Indrarti, L.; Azuma, J.-I. Cellulose 1998, 5, 215−228. (43) Whitney, S. E. C.; Brighama, J. E.; Darkea, A. H.; Reidb, J. S. G.; Gidley, M. J. Carbohydr. Res. 1998, 307, 299−309. (44) Yamamoto, H.; Horii, F. Cellulose 1994, 1, 57−66. (45) Köhnke, T.; Brelid, H.; Westman, G. Cellulose 2009, 16, 1109− 1121. (46) Köhnke, T.; Ostlund, A.; Brelid, H. Biomacromolecules 2011, 12, 2633−2641. (47) Köhnke, T.; Pujolras, C.; Roubroeks, J. P.; Gatenholm, P. Cellulose 2008, 15, 537−546. (48) Smith, B. G.; Harris, P. J.; Melton, L. D.; Newman, R. H. Plant Cell Physiol. 1998, 39, 711−720. (49) Mohnen, D. Curr. Opin. Plant Biol. 2008, 11, 266−277. (50) Hieu, H. C.; Tuan, N. A.; Li, H.; Miyauchi, Y.; Mizutani, G. Appl. Spectrosc. 2011, 65, 1254−1259. (51) Zykwinska, A.; Thibault, J. F.; Ralet, M. C. Carbohydr. Polym. 2008, 74, 957−961. (52) Zykwinska, A.; Thibault, J. F.; Ralet, M. C. J. Exp. Bot. 2007, 58, 1795−1802. (53) Zykwinska, A. W.; Ralet, M. C.; Garnier, C. D.; Thibault, J. F. Plant Physiol. 2005, 139, 397−407. (54) Zykwinska, A.; Gaillard, C.; Buleon, A.; Pontoire, B.; Garnier, C.; Thibault, J. F.; Ralet, M. C. Biomacromolecules 2007, 8, 223−232. (55) Zhao, H.; Baker, G. A.; Cowins, J. V. Biotechnol. Prog. 2010, 26, 127−133. (56) Pauly, M.; Albersheim, P.; Darvill, A.; York, W. S. Plant J. 1999, 20, 629−639. (57) Imai, T.; Terashima, N.; Yasuda, S. J. Wood Sci. 1998, 44, 106− 110. 2724

dx.doi.org/10.1021/bm500567v | Biomacromolecules 2014, 15, 2718−2724