Maturation stress generation in poplar tension wood studied by synchrotron radiation microdiffraction.

Tension wood is widespread in the organs of woody plants. During its formation, it generates a large tensile mechanical stress, called maturation stress. Maturation stress performs essential biomechanical functions such as optimizing the mechanical resistance of the stem, performing adaptive movements, and ensuring long-term stability of growing plants. Although various hypotheses have recently been proposed, the mechanism generating maturation stress is not yet fully understood. In order to discriminate between these hypotheses, we investigated structural changes in cellulose microfibrils along sequences of xylem cell differentiation in tension and normal wood of poplar (Populus deltoides x Populus trichocarpa 'I45-51'). Synchrotron radiation microdiffraction was used to measure the evolution of the angle and lattice spacing of crystalline cellulose associated with the deposition of successive cell wall layers. Profiles of normal and tension wood were very similar in early development stages corresponding to the formation of the S1 and the outer part of the S2 layer. The microfibril angle in the S2 layer was found to be lower in its inner part than in its outer part, especially in tension wood. In tension wood only, this decrease occurred together with an increase in cellulose lattice spacing, and this happened before the G-layer was visible. The relative increase in lattice spacing was found close to the usual value of maturation strains, strongly suggesting that microfibrils of this layer are put into tension and contribute to the generation of maturation stress.

Wood cells are produced in the cambium at the periphery of the stem. The formation of the secondary wall occurs at the end of cell elongation by the deposition of successive layers made of cellulose microfibrils bounded by an amorphous polymeric matrix. Each layer has a specific chemical composition and is characterized by a particular orientation of the microfibrils relative to the cell axis (Mellerowicz and Sundberg, 2008). Microfibrils are made of crystalline cellulose and are by far the stiffest constituent of the cell wall. The microfibril angle (MFA) in each layer is determinant for cell wall architecture and wood mechanical properties.
During the formation of wood cells, a mechanical stress of a large magnitude, known as "maturation stress" or "growth stress" (Archer, 1986;Fournier et al., 1991), occurs in the cell walls. This stress fulfills essential biomechanical functions for the tree. It compensates for the comparatively low compressive strength of wood and thus improves the stem resistance against bending loads. It also provides the tree with a motor system (Moulia et al., 2006), necessary to maintain the stem at a constant angle during growth (Alméras and Fournier, 2009) or to achieve adaptive reorientations. In angiosperms, a large tensile maturation stress is generated by a specialized tissue called "tension wood." In poplar (Populus deltoides 3 Populus trichocarpa), as in most temperate tree species, tension wood fibers are characterized by the presence of a specific layer, called the G-layer (Jourez et al., 2001;Fang et al., 2008), where the matrix is almost devoid of lignin (Pilate et al., 2004) and the microfibrils are oriented parallel to the fiber axis (Fujita et al., 1974). This type of reaction cell is common in plant organs whose function involves the bending or contraction of axes, such as tendrils, twining vines (Bowling and Vaughn, 2009), or roots (Fisher, 2008).
The mechanism at the origin of tensile maturation stress has been the subject of a lot of controversy and is still not fully understood. However, several recent publications have greatly improved our knowledge about the ultrastructure, chemical composition, molecular activity, mechanical state, and behavior of tension wood. Different models have been proposed and discussed to explain the origin of maturation stress (Boyd, 1972;Bamber, 1987Bamber, , 2001Okuyama et al., 1 These authors contributed equally to the article. * Corresponding author; e-mail clair@lmgc.univ-montp2.fr. The author responsible for distribution of materials integral to the findings presented in this article in accordance with the policy described in the Instructions for Authors (www.plantphysiol.org) is: Bruno ).
[C] Some figures in this article are displayed in color online but in black and white in the print edition.
[W] The online version of this article contains Web-only data. [OA] Open Access articles can be viewed online without a subscription.
www.plantphysiol.org/cgi/doi/10.1104/pp. 109.149542 1994Yamamoto, 1998Yamamoto, , 2004Alméras et al., 2005Alméras et al., , 2006Bowling and Vaughn, 2008;Goswami et al., 2008;. The specific organization of the G-layer suggests a tensile force induced in the microfibrils during the maturation process. Different hypotheses have been proposed to explain this mechanism, such as the contraction of amorphous zones within the cellulose microfibrils (Yamamoto, 2004), the action of xyloglucans during the formation of microfibril aggregates (Nishikubo et al., 2007;, and the effect of changes in moisture content stimulated by pectin-like substances (Bowling and Vaughn, 2008). A recent work (Goswami et al., 2008) argued an alternative model, initially proposed by Mü nch (1938), which proposed that the maturation stress originates in the swelling of the G-layer during cell maturation and is transmitted to the adjacent secondary layers, where the larger MFAs allow an efficient conversion of lateral stress into axial tensile stress. Although the proposed mechanism is not consistent with the known hygroscopic behavior of tension wood, which shrinks when it dries and not when it takes up water (Clair and Thibaut, 2001;Fang et al., 2007;Clair et al., 2008), this hypothesis focused attention on the possible role of cell wall layers other than the G-layer. As a matter of fact, many types of wood fibers lacking a G-layer are known to produce axial tensile stress, such as normal wood of angiosperms and conifers (Archer, 1986) and the tension wood of many tropical species (Onaka, 1949;Clair et al., 2006b;Ruelle et al., 2007), so that mechanisms strictly based on an action of the G-layer cannot provide a general explanation for the origin of tensile maturation stress in wood. In order to further understanding, direct observations of the mechanical state of the different cell wall layers and their evolution during the formation of the tension wood fibers are needed. X-ray diffraction can be used to investigate the orientation of microfibrils (Cave, 1966(Cave, , 1997a(Cave, , 1997bPeura et al., 2007Peura et al., , 2008aPeura et al., , 2008b and the lattice spacing of crystalline cellulose. The axial lattice spacing d 004 is the distance between successive monomers along a cellulose microfibril and reflects its state of mechanical stress (Clair et al., 2006a;Peura et al., 2007). If cellulose microfibrils indeed support a tensile stress, they should be found in an extended state of deformation. Under this assumption, the progressive development of maturation stress during the cell wall formation should be accompanied by an increase in cellulose lattice spacing. Synchrotron radiation allows a reduction in the size of the x-ray beam to some micrometers while retaining a strong signal, whereby diffraction analysis can be performed at a very local scale (Riekel, 2000). This technique has been used to study sequences of wood cell development (Hori et al., 2000;Mü ller et al., 2002). In this study, we report an experiment where a microbeam was used to analyze the structural changes of cellulose in the cell wall layers of tension wood and normal wood fibers along the sequence of xylem cell differentiation extending from the cambium to mature wood (Fig. 1). The experiment was designed to make this measurement in planta, in order to minimize sources of mechanical disturbance and be as close as possible to the native mechanical state (Clair et al., 2006a). The 200 and 004 diffraction patterns of cellulose were analyzed to investigate the process of maturation stress generation in tension wood.

RESULTS
Typical diffraction patterns obtained from one tension wood sample and one normal wood sample are shown in Figures   Anatomy of the Studied Samples Figures 2D and 3D show transverse sections of the studied sample strips that are focused on a 1,200-mmlong radial strip extending from the phloem to the mature wood. A 50-to 100-mm-thick tangential strip of periphloem fibers is visible in the bark, centered around 200 to 250 mm from the cambial area. These thick-walled cells are easily detected on the x-ray intensity patterns of both 004 and 200 cellulose crystal planes and were used to match the diffraction patterns and the microscopic observations.
In the first 400 mm after the cambium, the progressive thickening of the fiber secondary wall (formation of the S1 and S2 layers) is clearly visible both in normal wood and tension wood (D1-D3 in Figs. 2 and 3). No difference in cell wall appearance between tension wood and normal wood could be seen in this area. After 400 mm, the thickening of the cell walls is complete for the normal wood sample (D4 and D5 in Fig. 2). In the tension wood sample, G-layers (D5 in Fig. 3) start to become visible in fibers 600 mm after the cambium (thin dotted line in Fig. 3). No G-layer was observed in the normal wood sample.

Changes in the Amount of Cellulose along a Development Sequence
The intensity of the 004 reflection depends on the amount of diffracting microfibrils crossed by the beam. Therefore, local variations in density (e.g. related to the presence of vessels) create slight fluctuations of the intensity profiles (but this does not affect the position of the peak, i.e. the lattice spacing of cellulose). The contributions of small MFA (,16°) and large MFA (.16°) to the intensity profiles are shown separately in Figures 2C and 3C. Intensity is repre-sented on a log scale in order that variations at low intensity can be seen despite the high intensity reached in mature wood. Comparison between these profiles indicates the relative abundance of cellulose oriented with a large and a small angle at each position.
After the first peak in the periphloem fibers, the cambial zone is characterized by a very low signal due to the absence of secondary walls. The intensity increases along 400 mm after the cambium, parallel to the thickening of the cell wall. During this phase, similar intensity profiles are observed for small MFA and large MFA in both normal wood and tension wood samples. After this phase, the signals of normal wood and tension wood begin to clearly differ. In normal wood, large and small angles have similar constantintensity patterns, consistent with the fact that the cell walls have reached their final thickness. In tension wood, the signal emerging from small MFA deviates from the large MFA, where the former becomes three to five time higher than the latter. This steep increase in intensity for small MFA starts approximately 420 mm after the cambium, long before any G-layer has become visible.

Changes in MFA along a Development Sequence
The change in intensity of the 200 reflection is shown in Figures 2A and 3A (on a log scale) for different azimuthal sectors. The diffraction patterns are similar to those obtained for the 004 reflection for equivalent angle classes, consistent with the fact that both reflect the variations in cellulose quantity. The 200 reflections, however, were less scattered and therefore more suitable for the analysis of a wider range of MFAs.
Insets show the full 200 reflections at some specific locations. In the first 50 to 100 mm after the cambium, the signal is very low (data not shown). Cell walls at that stage are composed only of the primary walls, and the very low quality of the signal does not allow the analysis of microfibril orientation (Mü ller et al., 2002). At the end of this stage (represented by the vertical red dotted line farthest to the left), a signal at large angles can be detected (A1 in Figs. 2 and 3), showing the dominant contribution of large MFA (estimated to be approximately 45°) of the S1 layer. In the next 150 to 200 mm, the signal intensity continues to increase together with the thickening of the cell wall. Clearly, this increase occurs at different rates for different angle classes of MFA. The contribution of MFA , 28°increases much quicker than that of larger MFA. This indicates that the developing layer has lower MFA than the previous one and probably corresponds to the development of the S2 layer. The peak angle at that stage is approximately 22°(A2 in Fig. 2), indicating that the mean MFA is about 25°(Supplemental Material S1). Normal wood and tension wood have very similar patterns until a point where the signal from MFA , 16°diverges from larger MFAs (second red dotted line from the left, located at +250 mm in normal wood and +350 mm in tension wood). This divergence indicates that the cell wall layers deposited from that point have a MFA , 16°, both in normal wood and tension wood. In normal wood, the contribution of this class of angles stops increasing near +400 mm (A3 and A4 in Fig. 2) and remains modest. In tension wood, the increase is much larger (A3 and A4 in Fig. 3) and continues until +600 mm, parallel to the increase in small MFA 004 reflection intensity previously noted. Analysis of the 200 reflection intensity for tension wood reveals a very large specific contribution at small angles, with intensity five times higher than for large angles, starting long before any G-layer is visible. A further increase of this contribution is observed during the development of the G-layer.

Changes in Axial Lattice Spacing of Cellulose along a Development Sequence
The evolution of the 004 planes mean lattice spacing (d 004 ) is shown separately for the small MFA (,16°) and large MFA (.16°) classes (Figs. 2C and 3C). The graph displays only the points with a signal clear enough to accurately determine the lattice spacing.
In normal wood, d 004 remains constant along the profile for both small and large angle microfibrils. In tension wood, the d 004 profile for large angles is also roughly constant, although more scattered because of the comparatively lower intensity of the signal for this category of angles. For small MFA in tension wood, one can clearly see a progressive increase in lattice spacing in the first 550 mm after the cambium. This increase occurs before the G-layer becomes visible. It is particularly steep in the area where the innermost part of the S2 layer, dominated by small MFA, diverges on the intensity signal (Fig. 3A). Similar patterns were observed in almost all studied profiles of normal wood and tension wood, except one tension wood profile out of six, for which the d 004 signal was very scattered (Supplemental Material S2).
The axial strain of cellulose was computed as the relative change in d 004 observed in this area. For small angle microfibrils in tension wood, it ranged between +0.1% to +0.3% with a mean of +0.18% for the five tension wood profiles. In most of the tension wood profiles, no or very little delay was observable between the deposition of cellulose (detected by intensity) and the increase in mean lattice spacing. DISCUSSION X-ray microdiffraction allowed the exploration of changes in cellulose ultrastructure along sequences of cell wall development in normal and tension wood. Variations in intensity were consistent with the progressive thickening of the cell walls. Variations of the 200 and 004 reflection intensities were observed along the sequences and interpreted in terms of MFA of the developing layer and change in cellulose lattice spacing. The different patterns observed in normal and tension wood revealed new information about the particular ultrastructure of tension wood secondary walls and their mechanical behavior during fiber development.

MFA of Cell Wall Layers
Interpretation of the 200 reflection profiles, in terms of MFA distribution, is not straightforward because walls that are not perpendicular to the beam create some artificial increase of the signal at small azimuth angles (Cave, 1997b). Therefore, the signal intensity at small azimuth does not always reflect the amount of cellulose with small MFA and must be interpreted with caution. We used a numerical model to assess our results (Supplemental Material S1). Based on this analysis, we found that the development of the secondary wall started with the deposition of a layer having a large mean MFA and assumed that this corresponds to the S1 layer. This stage was followed by an abrupt decrease in MFA of the deposited layers, assumed to correspond to the early development of the S2 layer. The MFA of this layer was evaluated to be approximately 25°, which is in the upper range of MFA usually reported for poplar. After this stage, we observed that further increase in wall thickness occurred along with a steep decrease in MFA of the deposited wall layers in both normal and tension wood. Accurate determination of this MFA was not possible, but simulations clearly show that the mean MFA is less than 10°in normal wood and close to 0°in tension wood. Microscopic observations showed that this occurs before any G-layer is visible in tension wood and corresponds to the development of the inner part of the S2 layer.
These observations indicate that the MFA is not uniform in the S2 layer of the studied samples. Donaldson and Xu (2005), observing individual cell walls with polarized light, reported that the S2 layer of Pinus radiata had a relatively uniform microfibril angle but noticed a trend of increasing MFA toward the S1 layer. Using x-ray microdiffraction across single walls of Norway spruce (Picea abies), Peura et al. (2005) showed that the dominant MFA had a broad distribution ranging over approximately 20°to 30°. Our observations suggest that there is a decrease in MFA of the S2 layer toward the lumen in poplar wood.
Our results also show that the MFA in the inner part of the S2 layer is much lower in tension wood than in normal wood. This result was compared with previous reports about the MFA of the S2 layer in poplar tension wood. Mü ller et al. (2006) analyzed the 200 diffraction patterns of mature tension wood. These patterns were dominated by a very strong central peak, but the authors detected "shoulders" beside this peak, very similar to what we observed (A3 and A4 in Fig. 3), and ascribed them to the reflection of the S2 layer. The two layers could also be clearly distinguished by a different apparent crystal size as estimated from the peak broadening. They concluded that the MFA of the S2 layer was between 20°and 25°. This is consistent with the MFA that we determined for the outer part of the S2 layer. They ascribed the presence of a strong central peak to the diffraction of the G-layer, but it is not possible to determine whether or not the inner S2 layer also contributed to the peak of small MFA in that study, because the signal emerging from the G-layer has a much larger magnitude that would therefore overshadow the signal emerging from the inner S2 layer. Goswami et al. (2008) used selective enzymes to remove the G-layer from a tension wood block and measured the MFA using x-ray diffraction. They found a value of approximately 36°, larger than the usual values of MFA in poplar normal wood, and assumed that this was representative of the S2 layer of tension wood. However, careful observation of their micrographs ( Fig. 1 in Goswami et al. 2008) reveals that the enzyme treatment probably removed not only the G-layer but also part of the S2 layer. In that case, the large observed MFA would be representative of the S1 layer and the outer part of the S2 layer, consistent with what we found here.

Variations in Cellulose Lattice Spacing
Our procedure was designed to measure the evolution of cellulose axial lattice spacing (d 004 ) along a sequence of cell wall development. Two effects can produce a change in d 004 : either a mechanical strain in the crystal, as was previously demonstrated in several studies (Ishikawa et al., 1997;Nakai et al., 2005;Clair et al., 2006a;Peura et al., 2007), or a change in equilibrium lattice spacing, for example due to a change in crystal size (Davidson et al., 2004). Indeed, several studies reported differences in apparent cellulose crystal size measured from the broadening of the 200 peak and the application of the Sherrer equation (Washusen and Evans, 2001;Mü ller et al., 2002Mü ller et al., , 2006Washusen et al., 2005;Ruelle et al., 2007), but this method is criticized since broadening is not affected only by crystal size but also by other factors such as crystal disorder (Kennedy et al., 2007;Nishiyama, 2009; for more details, see Supplemental Material S3). Moreover, since there has not been any report, so far, on the occurrence of different cellulose synthase complexes within a secondary cell wall, we believe that a change in crystal size is unlikely to occur within a single wall layer. Therefore, we assume that the d 004 of cellulose at mechanical equilibrium is a constant, so that its variations are reflecting mechanical strain and stress in the cellulose crystal (for a more detailed discussion about the equilibrium lattice spacing, see Supplemental Material S3). However, the observed diffraction patterns should be interpreted with caution, as there are at least two possible sources of artifacts. The first is a possible artifact in data processing: a change of the 004 peak position on the screen may result either from a change in d 004 or from a change in the distance between the sample and the detector. This possibility of artifact has been minimized due to the special care taken with the control of sample vertical alignment during the measurement. The flatness of the profiles obtained for normal wood confirms the quality of this vertical alignment. The second possible artifact is related to the mechanical perturbation of wood during sample preparation, such as the suppression of the tree self-weight when cutting the segment or the release of maturation stress when preparing the sample. Again, we designed our procedure in order to minimize these effects, and we subsequently checked numerically that these effects were negligible. The overall reliability of our procedure is again attested by the flatness of the profiles obtained for normal wood.
The major result we obtained is a clear difference between the patterns of changes in d 004 between normal wood and tension wood. We have provided evidence that the cellulose lattice spacing increases in the inner part of the S2 layer only in tension wood. In addition, we could demonstrate that this increase was correlated to the deposition of microfibrils with very small MFA before G-layer formation. This may indicate that, inside the inner part of the S2 layer of tension wood, cellulose is put under tension just after its deposition. Furthermore, the magnitude of the strain deduced from our experiments (between 0.1% and 0.3%) is very close to the macroscopic released strain of maturation stress usually reported for poplar tension wood (Fang et al., 2008).

CONCLUSION
Under the assumption that cellulose equilibrium lattice spacing does not vary within the S2 layer, our results strongly suggest that the inner part of the S2 layer significantly contributes to the generation of maturation stress in poplar tension wood and that this occurs before the appearance of the G-layer. This conclusion is in sharp contrast with current conceptions of maturation stress generation as exclusively based on the action of the G-layer (Bowling and Vaughn, 2008;Goswami et al., 2008;. There is evidence that cell walls without G-layers can produce significant axial tensile stress, for example in normal wood or in the tension wood of many tropical species (Onaka, 1949;Clair et al., 2006b;Ruelle et al., 2007). This does not obviate a contribution of the G-layer to the development of tensile stress in G-layer tension woods. Indeed, a higher tensile stress in the G-layer compared with adjacent layers is most likely responsible for the contraction of the G-layer observed on longitudinal sections of poplar and beech (Fagus species) tension wood with light microscopy (Clair et al., 2005) and on transverse sections using atomic force microscopy (Clair and Thibaut, 2001). Likewise, a positive correlation between the released strain of maturation stress and the amount of G-layer in the tissue has been reported for several different wood materials (Clair et al., 2003;Yamamoto et al., 2005;Fang et al., 2008). In addition, this study demonstrated that the inner part of the S2 layer of tension wood had a MFA close to that of the G-layer. It has been already shown in species lacking the G-layer that tension wood has a lower MFA than normal wood (Clair et al., 2006b;Ruelle et al., 2007). We may speculate that there is a continuum in mechanical design and function between normal wood, tension wood without the G-layer, and tension wood with the G-layer, as recently suggested by a study of the physical structure of the G-layer in several tropical species (Chang et al., 2009). Such a common mechanism, yet to be identified at the molecular level, may create the tension in the cellulose microfibrils, while the small MFA function may be restricted to an efficient stress transmission in the direction of the fiber axis.

Plant Material
Experiments were performed on young poplar trees (Populus deltoides 3 Populus trichocarpa 'I45-51'). In early spring 2007, trees were grown from cuttings in the INRA center at Orleans, France. They were tilted 35°from vertical to induce the formation of tension wood with high tensile maturation stress on the upper side of the stems. On June 24, trees were moved to the European Synchrotron Radiation Facility (ESRF) in Grenoble, France. They were kept tilted, outdoors, and were regularly watered until the start of the experiments. After preliminary experiments, four trees, approximately 2 m high, were selected and studied successively between June 27 and June 28. Sample preparation and measurements of each tree were carried out within 8 h.

Sample Preparation
Sample preparation was designed to prevent the release of mechanical stress within the segment in order to keep the sample as close as possible to its in planta mechanical state (Clair et al., 2006a). On each tree, a 40-cm-long stem segment (approximately 3 cm in diameter) was taken from the curved basal part of the stem. This length was chosen so that the central part of the segment was far enough from its ends to minimize mechanical disturbance of the studied area. The segment was then prepared to leave a thin strip of peripheral wood on the mid part of one side (Fig. 1), using a homemade device made of two parallel razor blades inserted radially into the segment. Special care was taken to remain parallel to the fibers and thus to avoid chopping them. Bark and wood were removed on both sides of the strip in order to leave a window for the x-ray beam. The strip of approximately 400 mm thick in the tangential direction included the bark, the cambial area, and the first 2 mm of wood. The samples were kept wet during the whole preparation process.
Three stem segments were sampled on the upper side of the tilted stems, and one was sampled on the opposite side. However, later anatomical observations revealed that one of the segments sampled on the upper side had not produced G-fibers and had a MFA typical of normal wood, so that our sample material finally included two strips representative of tension wood and two strips representative of normal wood.

X-Ray Setup and Experimental Method
The experiment was carried out at the ID13 beamline of the ESRF using a 5-mm beam (Riekel, 2000). The wavelength of the monochromatic x-ray beam was l = 0.961176 Å . A MAR165 CCD detector (16-bit readout with 2,048 3 2,048 pixels) was placed at approximately 177 mm from the sample, so that it recorded both the 200 and 004 reflections. The distance between the sample and the detector was determined using an silver behenate calibration powder (Blanton et al., 1995), which was directly applied on the wood strip after the measurements.
The sample holder was designed to maintain the stem segment perpendicular to the beam, while allowing precisely monitored vertical and lateral displacements. On each wood strip, three radial profiles (1 mm apart from each other along the fiber direction) were recorded in the transition zone between bark and mature wood (Fig. 1). Each profile contains 150 measurement points at successive radial positions separated by 10 mm, extending from outer bark to mature wood. Samples were kept wet during the measurements using a microdrop system fixed to the wood segment.

Data Processing
A total of 1,800 diffraction patterns were recorded for this study. They were processed using the Fit2D software (http://www.esrf.eu/computing/ scientific/FIT2D) and homemade software procedures programmed with Microsoft Excel/Visual Basic. Despite the very small amount of wood material crossed by the x-ray beam (low thickness and small beam size), the data yielded a high-quality signal with very low noise, allowing a detailed analysis of the 004 and 200 reflections.
The 200 reflection in each pattern (centered at u 200 = 7.31°) was radially integrated across its profile in order to obtain a plot of the azimuthal intensity distribution of 200 planes. The azimuthal plot was used to obtain information about the orientation of microfibrils. The diffraction pattern was interpreted in terms of MFA distribution. For cell walls oriented perpendicular to the beam, the variation of intensity as a function of azimuthal position is a direct image of the MFA distribution (Cave, 1997b). The contribution of cell walls that are not perpendicular to the beam creates some distortion of the signal, but this distortion can be taken into account in the interpretation (Supplemental Material S1).
The 004 reflection was used to obtain information about cellulose lattice spacing. The diffraction patterns were integrated on a defined azimuthal sector (see details below), and the intensity was plotted as a function of the radial position on the image. The radial position of maximal intensity (r max ) was determined using local polynomial fitting. Any radial position r on the pattern corresponds to a value of the Bragg angle: u 004 = ½atan(r/L), where L is the distance between the sample and the camera. The Bragg angle of the cellulose 004 reflection is approximately equal to 10.73°at the experimental wavelength of l = 0.961176 Å , but variations in cellulose lattice spacing d 004 induce slight changes according to the equation of diffraction: d 004 = l/(2sinu 004 ). The mean lattice spacing, therefore, was computed as d 004 = l/(2sin[(atan(r max /L))/2]).
A given microfibril contributes to the 004 reflection at an azimuthal position that depends on the microfibril orientation. Therefore, the lattice spacing of cellulose can be computed separately for different classes of MFA by analyzing the signal in specific azimuthal sectors. The above-mentioned calculations were carried out for two distinct azimuthal integration sectors (60°-12°and 612°-24°), yielding two distinct sets of lattice spacings. The first is representative of the fraction of cellulose with MFA , 16°, and the second is representative of the fraction with MFA . 16°(Supplemental Material S1).

Microscopic Observations
The studied wood strips were taken from the segment just after the x-ray diffraction experiment and kept in water at 4°C until further processing. First, samples were dehydrated in ethanol, then in propylene oxide, and finally they were embedded in Epon after polymerization at 60°C. Transverse sections of 1 mm thickness were stained with methylene blue/AzurII mix and observed with a light microscope (Leica Microsystems).

Supplemental Data
The following materials are available in the online version of this article.
Supplemental Material S1. Interpretation of diffraction patterns.
Supplemental Material S2. Changes in lattice spacing in all studied profiles.
Supplemental Material S3. Discussion about the equilibrium lattice spacing.

ACKNOWLEDGMENTS
We gratefully acknowledge Richard Davies (ID13-ESRF) for technical support, Françoise Laurans and Alain Moreau (UR588 Amélioration, Génétique, et Physiologie Forestières, INRA Orléans) for excellent assistance in preparing the histological sections, and J. Paul MacLean (Laboratoire de Mécanique et Génie Civil, CNRS, Université Montpellier 2) for checking the English and making improvements to the manuscript.