Peroxidase activity can dictate the in vitro lignin dehydrogenative polymer structure Vale ´rie Me ´chin * , Ste ´phanie Baumberger, Brigitte Pollet, Catherine Lapierre UMR 206 Chimie Biologique, INRA/INA PG, F-78850 Thiverval Grignon, France Received 5 October 2006; received in revised form 9 November 2006 Available online 21 December 2006 Abstract The objective of this study was to assess the influence of the peroxidase/coniferyl alcohol (CA) ratio on the dehydrogenation polymer (DHP) synthesis. The soluble and unsoluble fractions of horseradish peroxidase (HRP)-catalyzed CA dehydrogenation mixtures were recovered in various proportions, depending on the polymerization mode (Zutropf ZT/Zulauf ZL) and HRP/CA ratio (1.6–1100 purp- urogallin U mmol À1 ). The ZL mode yielded 0–57%/initial CA of unsoluble condensed DHPs (thioacidolysis yields <200 lmol g À1 ) with a proportion of uncondensed CA end groups increasing with the HRP/CA ratio (7.2–55.5%/total uncondensed CA). Systematically lower polymer yields (0–49%/initial CA) were obtained for the ZT mode. In that mode, a negative correlation was established between the b-O- 4 content (thioacidolysis yields: 222–660 lmol g À1 ) and the HRP/CA ratio. In both modes, decreasing the HRP/CA ratio below 18 U mmol À1 favoured an end-wise polymerization process evidenced by the occurrence of tri-, tetra- and pentamers involving at least one b-O-4 bond. At low ratio, the unsoluble ZT DHP was found to better approximate natural lignins than DHPs previously synthesized with traditional methods. Besides its possible implication in lignin biosynthesis, peroxidase activity is a crucial parameter accounting for the structural variations of in vitro DHPs. Ó 2006 Elsevier Ltd. All rights reserved. Keywords: Dehydrogenation polymers; Coniferyl alcohol; Horseradish peroxidase; Lignin structure; Glycerol b-aryl ether bonds 1. Introduction Lignin plays an essential role in plant growth, develop- ment and uses. It improves water conduction through tra- cheary elements, limits pathogen attacks but also restricts the degradation of cell wall polysaccharides by enzymes, thus decreasing feeding value. Although studies on lignin biosynthesis have started in the early-50s (Higuchi, 1990), important findings occurred within the last few years. Facilitated by the combination of molecular biology, genetics, bioinformatics, biochemistry and physiology, these findings underlie the successive updated schemes pro- posed for the monolignol biosynthetic pathways (Boerjan et al., 2003; Barriere et al., 2004; Ralph, 2005; Sibout et al., 2005; Chiang, 2006). After their synthesis, the mono- lignols are thought to be transported under a glycosylated form to the cell wall where oxidative polymerization takes place (Steeves et al., 2001). This last step in lignin synthesis still raises questions but the involvement of peroxidases (among others proteins) to catalyze the production of monolignol radicals is quite widely admitted (Onnerud et al., 2002; Boerjan et al., 2003; Ros Barcelo et al., 2004). Formation of the polymer would result from radical couplings between two dehydrogenated compounds, the oxidized monomer and the growing polymer phenoxy rad- ical (endwise polymerization) or two oligomer phenoxy radicals (bulk polymerization) (Brunow, 1998). Given the colossal complexity of lignin in terms of bio- synthesis, structure and interactions with the cell wall poly- saccharidic network, the investigation of normal, mutants or transformant plants altered in their lignification profile is not enough for plainly deciphering lignin structure and properties. A variety of model systems have consequently 0031-9422/$ - see front matter Ó 2006 Elsevier Ltd. All rights reserved. doi:10.1016/j.phytochem.2006.11.024 * Corresponding author. Tel.: +33 1 30 81 54 62; fax: +33 1 30 81 53 73. E-mail address: mechin@grignon.inra.fr (V. Me ´chin). www.elsevier.com/locate/phytochem Phytochemistry 68 (2007) 571–579 PHYTOCHEMISTRY