Biomacromolecules 2008, 9, 767–771
767
Communications Tracing Cellulose Elements Adsorbed on Composite Cellulose Biomaterials by a New Labeling Method Jean-Paul Joseleau,* Valérie Chevalier-Billosta,* and Katia Ruel Centre de Recherches sur les Macromolécules Végétales (CERMAV, UPR 5301, CNRS) BP 53, 38041 Grenoble Cedex 9, France Received October 12, 2007; Revised Manuscript Received November 21, 2007
In view of tracing the fate of cellulose fine elements added to a suspension of cellulose fibers, a novel method for specific labeling of polysaccharides in a composite material was developed. The purpose was to visualize a given cellulose material within a cellulose mixture. The method consists of generating aldehyde groups in the chain by mild periodic acid oxidation followed by biotinylation of the carbonyls. Once added to the composite, the biotinylated molecules are labeled with streptavidin conjugated to a fluorescent probe for confocal microscopy, or streptavidin-gold for electron microscopy observations. In the present work, the fate of fresh fines (neverdried) and dead fines (dried) when they were added to a purified suspension of fibers was followed by observation of the labeling in confocal and electron microscopy. The differential mode of interaction of fresh fines and dead fines with the fibers was correlated to the mechanical characteristics measured on the corresponding papers. The versatility of the new labeling method and its possible generalization to other polysaccharides incorporated to a polysaccharide or nonpolysaccharide material should be of potential interest for the study of composite microstructure.
Introduction In the composite fiber wall, cellulose microfibrils in which cellulose chains are highly organized provide the mechanical strength and hemicelluloses and lignin contribute by their ability to form strong hydrogen bonded interactions and cross-links with the microfibrils.1 Paper manufacturing is one of the major industries of products derived from wood fibers. The process involves dissociation of the fibers from within the wood tissues and removal of a part of the lignin. Another important operation is the mechanical refining, whose purpose is to enhance the interfiber surface bonding capacity. All these structural changes in fiber integrity2 greatly influence the pulp behavior. In these processes that drastically affect the fibers, cellulose crystallinity is modified.3 The mechanical treatment of refining causes other important effects due to the release of fines, defined as elements of small size (ca. 75 µm) compared to the fiber size (several mm)4 with various hydrodynamic specific volumes and different particle characteristics.5 Among these, microfibrillar fines originating from the mechanical peeling of fibers have a crucial importance because they interact tightly with the fibers, influencing their aggregation and thereby the physical properties. The correlation between the interaction of fines with the fibers and their effects on the mechanical properties are not clearly understood. In the recycling of paperboard, a stage of high intensity refining is intended to restore the properties of recovered fibers. This operation generates an excess of two kinds of fibrillar fines that may adversely affect the pulp behavior.6,7 These are the so* Authors to whom correspondence should be addressed. E-mail:
[email protected] (J.-P.J.);
[email protected] (V.C.-B.). Telephone: 33-476-03-76-61. Fax: 33-476-54-72-03 .
called fresh fines, newly produced from the S2 layer of fibers, and the dead fines, which correspond to fibrillar fines that have undergone irreversible hornification due to drying during the previous manufacturing of paper. The respective roles of these fines on paper qualities are becoming increasingly important with the growing rate of recovery of fibers and their recycling.4 In this work, to get an enhanced understanding of the role of cellulose fines, it was necessary to visualize the way the cellulose fragments interact with the larger cellulose fibrils. Methods for staining cellulose in light microscopy and fluorescence microscopy are available. An approach had been implemented in scanning electron microscopy (SEM) using back-scattered electrons, allowing the visualization of halogenated fines when they were added to a long fibers fraction.8 However, methods allowing specific labeling of a cellulose microfibril fraction with a heavy atom for its visualization in transmission electron microscopy (TEM) are scarce. Here, to overcome the difficulty of differentiating the smaller cellulose objects from the cellulose network, we developed a novel method of labeling that allowed tracing of the distribution of the cellulose fines within the cellulose composite at the scale of electron microscopy. The method involves mild periodic acid oxidation of the β-1,4-linked glucosyl residues of the cellulose chains to generate a few dialdehyde groups followed by reductive amination coupling to biotin and labeling by formation of the strong affinity complex with a streptavidin labeled derivative.
Experimental Procedures Pulp Material. The pulp used in this study was prepared in the laboratory by kraft cooking of spruce chips. The cooking was carried out with 22% active alkali and 35% sulfidity. The steaming temperature of 170 °C was reached in 90 min and maintained for 90 min. The chips
10.1021/bm7011339 CCC: $40.75 2008 American Chemical Society Published on Web 01/31/2008
768 Biomacromolecules, Vol. 9, No. 3, 2008 were disintegrated in pulper for 15 min and then screened in a fine slot screener (30/100 mm). The final unbleached pulp was centrifuged and had a κ number of 30. Refining and Separation of Fines. The pulp was submitted to refining in a Valley pile, giving an extent of refining of 45° SR (Schopper-Riegler degree) according to the norm ISO 5264. Fines were separated from fibers by filtration in a Bauer MacNett apparatus using a 200 mesh filter and recovered by decantation. These constituted the “fresh fines”. A part of the fines was subjected to drying for one hour at 105 °C, resuspended in water, then disintegrated in a PFI mill at 30 000 rotation and constituted the “dead fines”. On the other hand, pure fibers devoid of contamination with fines were collected. For the evaluation of the association capacity of fresh and dead fines with fibers, varying proportions of fines were added to a suspension of pure fibers and the mixture was filtered on a 200 mesh filtered as before. PARA-Gold Labeling. Samples of fines and fibers suspension were placed in nylon cloth bags (20 µm) and kept in a 1% solution of NaBH4 for 3 h with occasional stirring. The bags were transferred in water and washed several times to neutral pH. Oxidation with periodic acid solutions (2% and 5%) was carried out for various times (30, 60, and 120 min) at 20 °C in the dark to prevent overoxidations. After thorough washing in water, the nylon bags were placed in 2.5 mL absolute methanol and acetic acid (1.5 mL) containing a 3 Å molecular sieve. Biotinylation and reductive amination were performed by addition of 5 mg of biotinamidocaproyl hydrazide (BACH) and 0.5 mg of cyanoborohydride (NaBH3CN)12 and the suspension kept at 50 °C for 16 h in the presence of a 3Å sieve. The nylon bags were transferred into methanol and washed exhaustively with absolute methanol and then distilled water. At this stage, for investigating the way fines interact with the fibers, the fines having undergone biotinylation were reintroduced into a suspension of fibers and the mixture fixed for electron microscopy preparation. Transmission Electron Microscopy (TEM). Samples were fixed in 0.2% (v/v) glutaraldehyde (Electron Microscopy Sciences, PA) and 2% (w/v) paraformaldehyde (Fluka, Buchs, Switzerland) in 0.05 M phosphate buffer (pH 7.0) for 2 h at room temperature and overnight at 4 °C. Samples were then washed in the 0.05 M phosphate buffer. Dehydration was performed in a series of aqueous solutions of increasing ethanol concentration. Progressive infiltration with LR White resin (Electron Microscopy Sciences, PA) was carried out by serial incubation in ethanol solutions of increasing LR White resin concentration followed by several incubations in 100% LR White. The pulp samples were then embedded in gelatin capsules and the resin allowed to polymerize for 24 h at 50 °C. Ultrathin sections (ca. 50 nm in thickness) were prepared (Ultratome III, LKB, Bromma, Sweden) and collected on carbon coated gold grids or plastic rings. Labeling for Transmission Electron Microscopy. Ultrathin sections (50 nm) collected on plastic rings were floated on 0.01 M PBS 2% fish gelatin, then on streptavidin-gold conjugate (Sigma-Aldrich) at various concentrations between 1/50 to 1/200 for 1 h at 37 °C. After incubation, the sections were rinsed with distilled water and transferred on carbon coated gold grids and poststained in 2.5% aqueous uranyl acetate. Observations were performed at 80 kV with a Philips CM 200 cryo-electron microscope. Labeling for Scanning Electron Microscopy and Confocal Microscopy. The biotinylated materials either alone or mixed into a suspension of fibers were treated for labeling with streptavidin-gold for SEM in the same conditions as for TEM. For laser confocal microscopy, the coupling was done with the fluorescent probe streptavidin-FluoProbes 630 (Interchim).
Results and Discussion In the present work, a chemical pulp from spruce was refined and the fine fraction generated during the mechanical treatment was separated from the bulk of the fibers. To reproduce the
Communications
Figure 1. Retention of fresh and dead fines into the fibers suspension.
occurrence of dead fines and fresh fines in the recycled papers, a part of the fines was never-dried (fresh fines) and another part was dried at 105 °C (dead fines) before being remixed in increasing proportions with pure fibers. The high proportion of both fines retained in the fiber network (Figure 1), as much as 85% ( 2 for fresh fines and 81% ( 2 for dead fines, indicates that fines constitute important integral components of the pulp. However, their influence on the mechanical properties of the papers made from the pulps showed that the fresh fines contributed better to the mechanical properties than the dead fines.9 It was, therefore, important in order to get a better knowledge of how the two types of fines influence the formation of the paper sheet and its physical properties to visualize the fines and their fate relative to the fibers. Because of their identical chemical nature, differentiating the added cellulose microfibrillar fines from the cellulose fiber framework necessitated specifically labeling the fines before they were added to the fiber suspension. Principle of the PARA-Gold Method. For our novel procedure to specifically label cellulose microfibrils in a cellulose mixture and consisting of the sequence of reactions: Periodic Acid oxidation, Reductive Amination and Gold labeling, we propose the designation PARA-Gold method. The method takes advantage of the exceptionally high affinity between avidin and biotin (dissociation constant of about 10-15), which ensures a strong complexation of the avidin probe to the biotinylated substrate. In the past years, we used selective oxidation with galactose oxidase followed by biotinylation and complexation with streptavidin-gold for a specific ELISA-plate assay for xyloglucans.10 Complexation to streptavidin-gold was recently implemented to label the reducing end of chitin11 and cellulose microcrystals12 for TEM investigation. In the present work, the first step of the sequence of reactions leading to the labeling of the chains of cellulose consisted of periodic acid oxidation of the glucose rings to produce a dialdehyde group resulting from the oxidative cleavage of the carbon-carbon bond between C2 and C3 (Figure 2). In the second step, the aldehyde groups were biotinylated by reductive amination procedure13 with NaBH3CN as described by Imai et al.12 The biotinylation reagent used in the reaction was the biotinamidocaproyl hydrazide (BACH), which underwent nucleophilic addition at the aldehyde carbon atom with formation of a Schiff base iminium intermediate that was subsequently reduced by NaBH3CN. In the second part of the sequence of reactions that uses the affinity of avidin for biotin, the complex was formed between the cellulose-BACH derivative and commercial streptavidin conjugated to different dyes or contrasting probes for optical, fluorescence, or electron microscopy. Thus, streptavidin-FluoProbe and streptavidin-gold
Communications
Figure 2. Sequence of the reactions leading to PARA-gold labeling of cellulose. The sequence involves the generation of a few randomly distributed dialdehyde groups by mild periodic acid oxidation. The carbonyl groups are then coupled to biotin by reductive amination with a hydrazide derivative of biotin (BACH) and NaBH3CN. In the last step corresponding to the visualization of the labelled molecules, biotin is complexed to a streptavidin conjugate carrying gold for electron microscopy, or the FluoProbe Alexa 630 for laser confocal microscopy.
conjugates were used to visualize the labeled fines in laser confocal microscopy and electron microscopy, respectively. As for any other colloidal gold conjugate the electron-opaque gold particles could be observed in scanning (SEM) as well as transmission electron microscopy (TEM). To create minimal alteration of the original structure of cellulose, it was important to carry out the periodic acid oxidation in such a way that only a limited number of glucosyl residues would be oxidized. Thus conditions largely below the stoichiometry of the reaction were used. The amount of periodic acid (2%, 2 h, 20 °C in the dark) was calculated to be about 1/100 of the theoretical amount necessary to oxidize all the glucosyl rings contained in the sample of fines. Biotinylation was carried out with BACH. Because a few critical factors had been reported to limit the reductive amination, a temperature of 50 °C and a reaction time of 14 h were carefully observed.12 Fate of PARA-Gold Labeled Cellulose Fine Elements in the Suspension of Cellulose Fibers. To follow the fate of the reintroduced fines relative to the fibers by the PARA-gold method, the fines, which were submitted to a pretreatment of reduction with NaBH4 to eliminate the potentially existing aldehyde groups, were submitted to mild controlled periodic acid oxidation and then biotinylated before being added to the suspension of pure fibers. Although it is difficult to ascertain that the added groups did not create conformational modification nor alteration in the H-bonding capacity of the fines, the largely understoichiometric conditions of the periodic acid oxidation warrant a very limited change of cellulose fine surface. Two kinds of labeling were performed depending of the type of microscopy. For laser confocal microscopy, the biotinylated mixture was coupled to the fluorescent streptavidin-FluoProbe 630. The images of fibers suspension into which fines were
Biomacromolecules, Vol. 9, No. 3, 2008 769
incorporated showed that the fresh fines were largely associated to the fibers and appeared more homogeneously distributed within the fiber wall than the dead fines that accumulated in patches (Figure 3). For electron microscopy, the pulps were labeled by coupling with streptavidin-gold conjugate. SEM examination demonstrated the capacity of the fines to interact at the surface of the fibers (Figure 4). The well-known heterogeneity of the pulps necessitated examination of a large number of images in electron microscopy in order to get a significant appreciation of the most representative status of the fibers and fines and their labeling. It thus appeared that the pulps with dead fines that aggregated against the fibers were more compact than those with fresh fines. With the dead fines, patches of higher density of gold particles were observed. By opposition, the fresh fines appeared more individualized and their distribution was looser, a characteristic that allows them to penetrate more evenly within the fibers network. Furthermore, at the higher scale of resolution of TEM and because TEM operates in transmission on ultrathin sections of the samples, it was possible to compare the fate of the two types of fines and their respective capacity to penetrate within the fibers walls (Figure 5). With this technique, it could be shown that fresh fines had a higher capacity to interact into the fiber wall than dead fines. In the case of fresh fines, the gold particles appeared largely dispersed whereas, once again, the dead fines tended to accumulate into local clumps. The respective behavior of fresh and dead fines is consistent with the fact that fresh fines are made of a limited number of cellulose microfibrils characterized by their flexibility whereas, in the dead fines, as an effect of drying, the microfibrils are aggregated into bulkier and stiffer complexes. This is also consistent with the observation that a higher proportion of the fresh fines were retained by hydrogen bonding interactions than the dead fines (data not shown), suggesting a more intimate association to cellulose fibers. Because of their morphological difference, the flexible fresh fines can more easily fill the open spaces between the cellulose fibrils created by the chemical and mechanical pulping treatments and contribute to a more bonded fibril network. Conversely, the dead fines, which have undergone the phenomenon of hornification14,15 due to drying, are less flexible, and this together with their larger size impairs their capacity of penetration between the lamellar structures of the fiber wall. As a result, they accumulate in morphological cracks and defects16 along the fibers and are heterogeneously located. The differential way of the two types of fines to interact within the fibers network explains the differences of mechanical properties of the pulps whether they were enriched in fresh fines or dead fines.8 The possibility to specifically visualize the fate of fines within the fiber network substantiates the fiber-fiber bonding by the fresh fines. On the other hand, the distribution of dead fines in batches of aggregated cellulose fines explains the conformability and aptitude to swelling of the pulp with dead fines.
Conclusion In conclusion, the new PARA-gold method allowed the selective visualization, at different scales of resolution and by different techniques of microscopy, of cellulose elements within a cellulose material environment. SEM observations revealed the fate of fresh and dead fines, respectively, relative to the surface of the fibers, confirmed by laser confocal
770 Biomacromolecules, Vol. 9, No. 3, 2008
Communications
Figure 3. Interaction between fines and fibers observed by laser confocal microscopy. The fines were labeled with the PARA-FluoProbe 630 method. The fresh fines interact in an apparently homogeneous manner with the fiber (a). The dead fines (b) bind with more difficulty and tend to accumulate as superficial patches (circles).
Figure 4. Scanning electron microscopy of PARA-gold labeled fresh fines (a) and dead fines (b) showing the light distribution of gold particles in the former and the patchy aspect (circles) corresponding to the aggregated dead fines. Bar: 5 µm.
Figure 5. Transmission electron microscopy of PARA-gold labeled fines interacting with fibers. (a) The fresh fines are seen in the transverse section adsorbed at the surface of the fiber (circle, dotted arrows) and also inside the delaminated fiber wall (arrow heads). (b) The dead fines penetrate in lower amount inside the fiber wall and form patches in the fiber morphological defects of the surface (circles).
microscopy. TEM examination showed that the fresh fines could better incorporate within the lamellar structure of the chemical pulp. The observations at the ultrastructural scale substantiate the improved bonding characteristics observed with the fresh fines and the lower conformability and lower aptitude to swelling of the pulps with dead fines. Our results underscore the importance of controlling both the amount and the quality of fines in the papers. This is of particular interest during recycling in which the refining stages should
be carried out in conditions yielding optimal proportions of fresh microfibrilar fines to reach better product specifications. It is worth noting that the present method may be useful as a selective labeling approach in the study of any composite and polymer blends containing carbohydrate moieties because it is applicable to any polysaccharide susceptible to periodate oxidation. By changing the type of complexation of streptavidin with colloidal gold or fluorescent probe or any other dye, the method offers large versatility of utilization in various microscopy modes. The versatility of the new labeling method and its possible generalization to other polysaccharide containing biomacromolecule incorporated to a polysaccharide or nonpolysaccharide material is of potential interest for the study of the microstructure of composites and biomaterials. Acknowledgment. Financial support for this work was provided by Région Rhône-Alpes (contract no. 0301 4672 01) and by the Association Nationale de la Recherche Technique (convention CIFRE no. 376/2003 allocated to V. ChevalierBillosta).
References and Notes (1) Atalla, R. H.; Hackney, J. M.; Ulhin, I; Thompson, N. S. Int. J. Biol. Macromol. 1993, 15, 109–112. (2) Brändström, J.; Joseleau, J-P.; Cochaux, A.; Giraud-Telme, N.; Ruel, K. Holzforschung 2005, 59, 675–680. (3) Hult, E-L.; Larsson, T.; Iversen, T. Cellulose 2000, 7, 35–55. (4) Luukko, K.; Maloney, T. C. Cellulose 1999, 6, 123–135. (5) Luukko, K.; Paulapuro, H. J. Pulp Paper Sci. 1999, 25, 273–277. (6) Zhang, S.; Peterson, D.; Qi, D. Effects of fines concentration on mechanical,properties of recycled paper. TAPPI Recycling symposium. Proceeding 2000, 653–661.
Communications (7) Laivins, J. V.; Scallan, A.M. J. Pulp Paper Sci. 1996, 22, 178–183. (8) De Silveira, G.; Zhang, X.; Berry, R.; Wood, J. R. J. Pulp Paper Sci. 1996, 22, 315–320. (9) Wistara, N.; Zhang, X.; Young, R. A. Cellulose 1999, 6, 325–348. (10) Priem, B.; Chambat, G.; Ruel, K.; Joseleau, J.-P. J. Carbohydr. Chem. 1997, 16, 625–633. (11) Imai, T.; Watanabe, T.; Yui, T.; Sugiyama, J. FEBS Lett. 2002, 510, 201–205. (12) Kim, N-H.; Imai, T.; Wada, M.; Sugiyama, J. Biomacromolecules 2006, 7, 274–280.
Biomacromolecules, Vol. 9, No. 3, 2008 771 (13) Che, F. Y.; Song, J. F.; Wang, K. Y.; Xia, Q. C. BioTechniques 2001, 30, 1271–1281. (14) Oksanen, T.; Buchert, J.; Viikari, L. Holzforschung 1997, 51, 355– 360. (15) Newman, R. H. Cellulose 2004, 11, 45–52. (16) Chevalier-Billosta, V.; Joseleau, J-P.; Cochaux, A.; Ruel, K. Cellulose 2007, 14, 141–152.
BM7011339
