Skip to main content
NCBI home page
Plant Physiology logoLink to Plant Physiology
. 2012 Sep 14;160(3):1204–1217. doi: 10.1104/pp.112.203364

Impact of the Absence of Stem-Specific β-Glucosidases on Lignin and Monolignols1,[W]

Aurélie Chapelle 1, Kris Morreel 1, Ruben Vanholme 1, Philippe Le-Bris 1, Halima Morin 1, Catherine Lapierre 1, Wout Boerjan 1, Lise Jouanin 1, Nathalie Demont-Caulet 1,*
PMCID: PMC3490608  PMID: 22984124

Abstract

Monolignol glucosides are thought to be implicated in the lignin biosynthesis pathway as storage and/or transportation forms of cinnamyl alcohols between the cytosol and the lignifying cell walls. The hydrolysis of these monolignol glucosides would involve β-glucosidase activities. In Arabidopsis (Arabidopsis thaliana), in vitro studies have shown the affinity of β-GLUCOSIDASE45 (BGLU45) and BGLU46 for monolignol glucosides. BGLU45 and BGLU46 genes are expressed in stems. Immunolocalization experiments showed that BGLU45 and BGLU46 proteins are mainly located in the interfascicular fibers and in the protoxylem, respectively. Knockout mutants for BGLU45 or BGLU46 do not have a lignin-deficient phenotype. Coniferin and syringin could be detected by ultra-performance liquid chromatography-mass spectrometry in Arabidopsis stems. Stems from BGLU45 and BGLU46 mutant lines displayed a significant increase in coniferin content without any change in coniferyl alcohol, whereas no change in syringin content was observed. Other glucosylated compounds of the phenylpropanoid pathway were also deregulated in these mutants, but to a lower extent. By contrast, BGLU47, which is closely related to BGLU45 and BGLU46, is not implicated in either the general phenylpropanoid pathway or in the lignification of stems and roots. These results confirm that the major in vivo substrate of BGLU45 and BGLU46 is coniferin and suggest that monolignol glucosides are the storage form of monolignols in Arabidopsis, but not the direct precursors of lignin.

β-Glucosidases belong to Glycosyl Hydrolase Family 1 (GH1). These enzymes catalyze the hydrolysis of a Glc linked to an aglycone moiety in the β-position. Most of them have a molecular mass ranging between 53 and 68 kD and are active at acidic pH. In dicotyledonous angiosperms, they are generally localized in the cell walls and are implicated in various processes like defense against pathogens, regulation of phytohormone activity, or phenylpropanoid biosynthesis (Xu et al., 2004).

The Arabidopsis (Arabidopsis thaliana) genome contains 47 β-glucosidase genes and pseudogenes and one β-glucosidase-like gene belonging to the GH1 family. Among them, β-GLUCOSIDASE45 (BGLU45 [At1g61810]), BGLU46 (At1g61820), and BGLU47 (At4g21760) belong to group 10 of Arabidopsis GH1 (Xu et al., 2004). These three genes are composed of 12 exons and share 50% sequence identity between them and the Pinus contorta gene (Dharmawardhana et al., 1995, 1999) that encodes a β-glucosidase specific for coniferin, the glucoside of the monolignol coniferyl alcohol (Xu et al., 2004). BGLU45 and BGLU46 are located in tandem on chromosome 1 and share about 80% sequence identity. The BGLU45 and BGLU46 proteins are secreted to the cell wall. The BGLU47 gene is located on chromosome 4 and is predicted to be directed to the peroxisome (Xu et al., 2004).

Monolignol glucosides accumulate in the cambium of gymnosperm wood. They have also been detected to a lower extent in some woody angiosperms, principally of the Magnoliaceae and the Oleaceae families (Terazawa et al., 1984a, 1984b), but they were detected more recently also in other angiosperms like flax (Linum usitatissimum; Huis et al., 2012). Coniferin was detected for the first time in the growing cambial sap and in the green stems of young spruce (Picea abies; Freundenberg and Harkin, 1963) and then repeatedly studied in conifer samples (Higuchi et al., 1977; Leinhos and Savidge, 1993, 1994). Syringin, the glucoside of sinapyl alcohol, is mostly present in angiosperms and has been less often found than coniferin in plant tissues (Sticher and Lahloub, 1982; Malarz et al., 2005; Kim et al., 2007; Yang et al., 2007). Recently, it has been shown that both coniferin and syringin can accumulate in roots of Arabidopsis grown in continuous light conditions. This accumulation was assigned to a “high-radiance response” (Hemm et al., 2004; Bednarek et al., 2005).

The slow turnover and the small pool of coniferin in spruce seedlings suggest that, if coniferin is involved in lignification, it would mediate only a part of lignin biosynthesis (Marcinowski and Grisebach, 1977; Grisebach, 1981). Coniferin seems to be produced and to accumulate in young cells to supply coniferyl alcohol during lignification, which could explain its slow turnover (Fukushima et al., 1997). Some experiments with radiolabeled monolignol glucosides have shown that they can be incorporated in wood lignins (Terashima et al., 1986, 1988, 1992, 1993; Terashima and Fukushima, 1988, 1989; Fukushima and Terashima, 1990, 1991a, 1991b; Rolando et al., 2004). However, tracer experiments with radiolabeled coniferin fed to stem segments of Ginkgo biloba revealed that coniferin is incorporated into lignin less efficiently than coniferyl alcohol and that it could be transiently oxidized to coniferaldehyde, which joins the monolignol pathway before its conversion into coniferyl alcohol and incorporation into lignins (Tsuji et al., 2004, 2005; Tsuji and Fukushima, 2004). In P. contorta, [3H]Phe was incorporated into polymeric lignin and lignin precursors (coniferyl alcohol and p-coumaryl alcohol) but was not incorporated into coniferin, suggesting that monolignol glucosides were not directly related to lignification (Kaneda et al., 2008). Recent studies in Arabidopsis proposed that monolignol glucosides would be the storage form of monolignols in vacuolar vesicles, whereas they would be transported to the cell wall as aglycones in plasma membrane vesicles (Miao and Liu, 2010; Liu et al., 2011).

Many β-glucosidases specific for monolignol glucosides have been identified in different gymnosperm and angiosperm species. In gymnosperms, the first β-glucosidases capable of hydrolyzing monolignol glucosides have been found in cell wall fractions of spruce hypocotyls and roots (Marcinowski and Grisebach, 1978). In another study, β-glucosidases specific for monolignol glucosides were located in the inner layer of the secondary cell walls, particularly in the epidermal layer and in the vascular bundles, of spruce seedlings (Marcinowski et al., 1979). In P. contorta, one β-glucosidase, located in the cell walls of developing xylem, was found to be in vitro specific for coniferin, syringin, and coniferin synthetic analogs (Dharmawardhana et al., 1995, 1999). In Pinus sylvestris and spruce, the hydrolysis of coniferin by β-glucosidases seems to be correlated with radial growth and xylem lignification. However, the β-glucosidase activity is low in these two conifers, and there is no clear relationship between lignification and coniferin hydrolysis (Marjamaa et al., 2003).

In angiosperms, the first isolated β-glucosidase specific for monolignol glucosides was found in the cell walls of chickpea (Cicer arietinum) cell suspension cultures (Hösel et al., 1978) and in the cell walls of xylem, epidermis, endodermis, and exodermis of roots and stems of chickpea seedlings (Burmeister and Hösel, 1981). Its best in vitro substrate is coniferin, but it can also hydrolyze some other aromatic β-glucosides (Hösel et al., 1978). In soybean (Glycine max), a β-glucosidase in vitro specific for coniferin, syringin, and ferulic acid 4-O-β-d-glucoside has been isolated (Hosel and Todenhagen, 1980). In cell suspension cultures of Petroselinum hortense and wheat (Triticum aestivum), the appearance of β-glucosidase activity is correlated with the initiation of lignification and, sometimes, with the appearance of other enzymes involved in lignin biosynthesis (Hosel et al., 1982).

The heterologous expression of Arabidopsis BGLU45 and BGLU46 genes in Pichia pastoris revealed that BGLU45 is highly specific for the three monolignol glucosides (coniferin, syringin, and p-coumaryl alcohol 4-O-β-d-glucoside). By contrast, BGLU46 has p-coumaryl alcohol glucoside as the preferred substrate, but nevertheless it displays a broad specificity toward various substrates such as coniferin, syringin, salicilin, arbutin, and phenyl β-d-glucosides (Xu et al., 2004; Escamilla-Treviño et al., 2006). The in vitro substrate specificity of BGLU45 and BGLU46 and the expression of their genes in stems (Escamilla-Treviño et al., 2006) support the hypothesis that these enzymes are involved in the lignification of Arabidopsis. To (1) verify whether coniferin is a true substrate for β-glucosidase in vivo and (2) investigate the potential role of β-glucosidase in lignification, we have studied the expression of the BGLU45, BGLU46, and BGLU47 genes and the impact of their silencing on lignin and soluble phenolics of Arabidopsis stems and roots.

RESULTS

Expression Profiles of the BGLU45, BGLU46, and BGLU47 Genes

The expression profiles of BGLU45, BGLU46, and BGLU47 were studied by quantitative (q)-reverse transcription (RT)-PCR with primers specific for each of these genes (Supplemental Table S1). BGLU45 was exclusively expressed in the stems (stages 6.1, 6.5, and 6.9 according to Boyes et al., 2001), BGLU46 was mainly expressed in seedlings, rosette leaves, and stems (stage 6.5), and BGLU47 was mainly expressed in rosette leaves but was barely detectable in the stems (Fig. 1). These results are mainly in accordance for stems with those of the Web-based GeneCAT expression tool (http://genecat.mpg.de; Supplemental Fig. S1).

Figure 1.

Figure 1.

Open in a new tab

qRT-PCR relative transcript abundance study of BGLU45, BGLU46, and BGLU47 in various Arabidopsis organs. Their expression was compared with the housekeeping EF1 gene. S, Seedlings; RL, rosette leaves; FL, flowers; St1, stem stage 6.1; St2, stem stage 6.5; St3, stem stage 6.9 (according to Boyes et al., 2001).

To further study the expression of BGLU45 and BGLU46 in the basal part of the stem, in situ hybridization experiments were performed using specific digoxigenin (DIG)-labeled probes for each of these genes. In spite of its poor sensitivity, which gave no conclusive result for BGLU45, this method revealed that BGLU46 was expressed in the protoxylem of the mature stems (stage 6.5) of the Columbia (Col-0; Supplemental Fig. S2) and Wassilewskija (WS; data not shown) accessions.

Both the qRT-PCR and in situ hybridization experiments consistently revealed that BGLU45 and/or BGLU46 are expressed in lignifying organs and tissues (i.e. in stems and in xylem). Moreover, in silico studies showed coexpression with many genes involved or potentially involved in the lignification process, like the 4-coumarate CoA ligase (4CL1) and 4CL2, the cinnamate-4-hydroxylase (C4H), and the cinnamoyl CoA reductase (CCR2) genes (Supplemental Table S2). To further evaluate whether BGLU45 and BGLU46 are involved in lignification, we studied various mutant lines affected in the expression of these genes and the impact of the mutation on lignin and soluble phenolics. BGLU47 is weakly expressed in stem, but as this gene shares 50% identity with BGLU45 and BGLU46, one knockout mutant for BGLU47 was characterized.

Isolation and Characterization of BGLU45, BGLU46, and BGLU47 Mutant Lines

T-DNA insertion mutant lines were selected from the Arabidopsis T-DNA insertion collections of Versailles (Bouché and Bouchez, 2001), the Salk Institute (Alonso et al., 2003), and GABI-Kat (Li et al., 2003, 2007; Rosso et al., 2003; Table I; Fig. 2). For BGLU45, we characterized three null mutants, bglu45-1, bglu45-2, and bglu45-3. The mutant bglu45-1 (FLAG_337E02), obtained in the WS genetic background, has a T-DNA inserted in the seventh exon of the At1g61810 gene. blgu45-2 (Salk_117269) and bglu45-3 (Salk_104069), obtained in the Col-0 accession, possess a T-DNA inserted in the fourth exon and in the seventh exon of the At1g61810 gene, respectively. For BGLU46, we selected three insertions, bglu46-1, bglu46-2, and bglu46-3. The bglu46-1 line (FLAG_267A12 in the WS accession) has a T-DNA insertion 5 bp before the ATG codon of the At1g61820 gene. Both the bglu46-2 (FLAG_395C03; WS accession) and bglu46-3 (GABI 295C07; Col-0 accession) lines have a T-DNA inserted in the fifth intron of the gene. Finally, only one bglu47 (FLAG_153A11; WS accession) mutant line could be selected for the At4g21760 gene, with a T-DNA insertion located in the second exon of the BGLU47 gene. RT-PCR performed on homozygous lines demonstrated that all the lines were null mutants for the target genes except bglu46-1, which is an overexpressor at the RNA level (Fig. 3). None of these T-DNA insertion mutants displayed any visible phenotype when grown in long-day or short-day conditions.

Table I. List of single mutants used in this study.

The different mutant lines at the homozygous stage were shown by RT-PCR to be knockout except bglu46-1, which was an overexpressor.

Gene Arabidopsis Genome Initiative No. Flanking Sequence Tag Mutant Name T-DNA Insertion RT-PCR Accession
GLU 45 At1g61810 FLAG_337E02 bglu45-1 Exon 7 Null WS
Salk_117269 bglu45-2 Exon 4 Null Col-0
Salk_104069 bglu45-3 Exon 7 Null Col-0
GLU 46 At1g61820 FLAG_267A12 bglu46-1 5′ untranslated region Overexpressor WS
FLAG_395C03 bglu46-2 Intron 5 Null WS
GABI 295C07 bglu46-3 Intron 5 Null Col-0
GLU47 At4g21760 FLAG_153A11 bglu47 Exon 2 Null WS
Open in a new tab

Figure 2.

Figure 2.

Open in a new tab

Schematic representation of the T-DNA insertions present in BGLU45 (A), BGLU46 (B), and BGLU47 (C) mutant lines. The T-DNA insertion sites are shown by the gray flags. The gray and black arrows indicate the positions of the primers used for genotyping and of the primers used for the expression studies, respectively. The black dashed arrows indicate the primers used for qRT-PCR.

Figure 3.

Figure 3.

Open in a new tab

RT-PCR analyses of β-glucosidase transcripts from bglu45 (A), bglu46 (B), and bglu47 (C) mutant lines, as compared with the corresponding wild-type (WT) sample (stems from plants grown in long-day conditions). Lane 1, expression of the β-glucosidase gene; lane 2, expression of the β-tubulin gene.

Biochemical Analyses

Further analyses were performed to check the absence of the BGLU45 and BGLU46 proteins in the corresponding two null mutants, bglu45-1 and bglu46-2. Western-blot analyses were performed with corresponding antibodies after a concanavalin A purification step of the protein extracts, as both BGLU45 and BGLU46 were predicted to be glycosylated (http://psort.ims.u-tokyo.ac.jp). Not unexpectedly, the purified extracts from the bglu45-1 and bglu46-2 null mutants were found to be deficient in BGLU45 and BGLU46 proteins, respectively (Fig. 4A). However, the signals detected on the western blots from the mutant lines were not reduced to the zero level. As BGLU45 shares 76% structural identity with BGLU46, this residual signal was most likely linked to cross reactions between the antibody specific for BGLU45 and the BGLU46 protein (or, conversely, between the antibody raised against BGLU46 and the BGLU45 protein). Global β-glucosidase activity was determined on the same protein extracts and with the synthetic substrate p-nitrophenyl-β-d-glucopyranoside. In these conditions, the global β-glucosidase activity was reduced by about 8.5% and 6.5% in bglu45-1 and bglu46-2, respectively (Fig. 4B).

Figure 4.

Figure 4.

Open in a new tab

Biochemical characterization of the bglu45-1 and bglu46-2 mutants. A, Immunoblot analysis with antibodies against Arabidopsis BGLU45 (a) and BGLU46 (b) proteins. B, Quantification of β-glucosidase activity in partially purified protein extracts in percentage of the wild type (WT). Error bars correspond to the se.

To investigate the localization of BGLU45 and BGLU46 in planta, immunolocalization experiments were performed in the same mutant lines, bglu45-1 and bglu46-2, as compared with the corresponding control (Fig. 5). In the control sample, the antibody specific for BGLU45 gave a positive signal both in the xylem and in the interfascicular fibers. In the bglu45-1 mutant, this signal disappeared in the interfascicular fibers (Fig. 5A), which strongly supports the occurrence of BGLU45 in this tissue. No conclusion could be drawn about the occurrence of BGLU45 in the xylem, because of the strong autofluorescence of the xylem and the possibility of cross reaction between BGLU46 (previously shown to occur in the xylem; Supplemental Fig. S2) and the BGLU45-specific antibody. In agreement with the experiments involving DIG-labeled probes against BGLU46 (Supplemental Fig. S2), the specific signal for the antibody directed against BGLU46 was mostly detected in the protoxylem of the wild type, whereas this signal disappeared in the bglu46-2 mutant. This suggests that BGLU46 is mainly located in the protoxylem (Fig. 5B).

Figure 5.

Figure 5.

Open in a new tab

Immunolocalization of BGLU45 (A) and BGLU46 (B) in Arabidopsis stems. Alexa 568 signal corresponding to the BGLU45 antibody and to the BGLU46 antibody is shown in green. Lignin autofluorescence is in red. BGLU45 is mainly located in the interfascicular fibers, and BGLU46 is mainly located in the protoxylem (white arrows).

Lignin Analyses in the Mature Stems of the Control and of the bglu45 and bglu46 Mutants

Lignin stainings were performed on sections of mature Arabidopsis stems (stage 6.5) of the mutant and control lines. No differences could be seen between the sections of the control and mutant lines, whatever their growth conditions (short-day or long-day regime), except for the bglu45-1 mutant grown in long-day conditions. In these conditions, this mutant systematically displayed supplementary cell layers rich in G units (according to the Maüle staining) in the interfascicular fibers of the basal part of mature stems (Fig. 6A and Supplemental Fig. S3). These supplementary cell layers were not present in the middle of the stem (Fig. 6B). This particular phenotype was observed in the two other bglu45 mutants (Col-0 accession), but not systematically, depending on the culture conditions.

Figure 6.

Figure 6.

Open in a new tab

Maüle staining of stem cross sections from WS and bglu45-1 plants grown in long-day conditions. A, Basal part of the stem. B, Medium part of the stem. Supplementary cell layers rich in G units can be observed at the basal part of the stem in the bglu45-1 mutant (black arrows). Bar = 100 μm.

The lignin content of extractive-free mature stems was measured for the mutant and corresponding control lines. These analyses were performed on several biological replicates for all the homozygous mutant lines. For each growth, the appropriate control was grown simultaneously in the same conditions. In spite of this between-culture variability and interestingly enough, the mature inflorescence stems of the bglu45-1 mutant systematically displayed a slight but significant increase in Klason lignin content when grown in the long-day regime (Table II). No difference was observed when plants were grown in the short-day regime (data not shown). In the same way, bglu45-3 displayed a slightly higher lignin content than its control for two independent cultures (Table II). In contrast, the stem lignin content of the BGLU46 and BGLU47 mutant lines (bglu46-2, bglu46-3, and bglu47) was not significantly different from that of the corresponding control (Table II).

Table II. Klason lignin content and lignin composition in mature inflorescence stems of the control (WS and Col-0) and β-glucosidase mutants grown together in the long-day regime.

The Klason analyses were expressed as weight percentages of the extractive-free samples. The syringyl-to-guaiacyl (S/G) ratio was measured by thioacidolysis. The data are mean values and se of two independent duplicates. KL, Klason lignin; CWR, cell wall residue. The asterisk indicates the overexpressor line.

Sample KL/CWR Yield (H + G + S) S/G
% µmol g−1 KL S/G molar ratio
WS 19.96 ± 0.18 1,101 ± 59 0.34
bglu45-1 21.87 ± 0.17 1,046 ± 33 0.43
Col-0 17.53 ± 0.13 1,056 ± 69 0.39
bglu45-2 18.08 ± 0.16 1,083 ± 9 0.39
Col-0 16.03 ± 0.1 1,353 ± 32 0.38
bglu45-3 16.99 ± 0.2 1,402 ± 25 0.40
WS 19.96 ± 0.18 1,101 ± 59 0.34
bglu46-1* 19.26 ± 0.09 986 ± 36 0.36
WS 18.26 ± 0.09 1,499 ± 11 0.38
bglu46-2 18.59 ± 0.07 1,344 ± 1 0.37
Col-0 17.43 ± 0.13 1,331 ± 12 0.40
bglu46-3 16.82 ± 0.01 1,373 ± 38 0.38
WS 16.78 ± 0.14 1,510 ± 12 0.37
bglu47 16.98 ± 0.16 1,468 ± 15 0.37
Open in a new tab

Thioacidolysis did not reveal any significant effect of the mutation on lignin structure. Neither the thioacidolysis yield (calculated on the basis of the Klason lignin content), which is the signature of the frequency of units only linked by β-O-4 bonds in lignins, nor the relative abundance of the H, G, and S monomers allowed us to discriminate the mutant lines from the corresponding control (Table II).

Metabolome Analyses

Stem Analyses

To further decipher the roles of the BGLU45 and BGLU46 genes, metabolome analyses were performed on mature Arabidopsis stems grown under short-day conditions. The method, in which ultra-performance liquid chromatography (UPLC) is coupled to mass spectrometry (MS), has been optimized for the detection of phenolic compounds (Morreel et al., 2004).

Around 100 compounds were profiled in the stem. Statistical analyses were performed on these remaining data to determine which compounds were significantly different between the mutant lines and their respective wild type. Nearly all of the differentials were significantly increased in the mutant lines, except one unknown compound, which was reduced in one mutant line (bglu46-2).

In the wild type (WS and Col-0 ecotypes), coniferyl alcohol, sinapyl alcohol, coniferin, and syringin were detected in the stems (Fig. 7A). Whereas syringin was difficult to detect because of its low content in the two ecotypes, coniferin could be easily detected in Arabidopsis stems. In contrast, the p-coumaryl alcohol 4-O-β-d-glucoside was not detected, either in WS or in Col-0. The content of coniferyl alcohol and sinapyl alcohol was about 1.5 and two times higher in the Col-0 ecotype than in the WS ecotype, respectively, whereas the coniferin and the syringin contents were nearly 25 times and 11 times higher in Col-0 than in WS, respectively (Fig. 7B). The ratios between the contents of monolignols and their respective glucosides in the stem were extremely different between the two ecotypes. In Col-0 stems, the intensity of the coniferin-associated mass-to-charge ratio (m/z) peak was 175 times higher than that of the m/z peak of free coniferyl alcohol, whereas this ratio was only 10 in the WS ecotype. In contrast, the intensity of the m/z peak for syringin was only twice higher than that of the free sinapyl alcohol-associated m/z peak in the Col-0 ecotype, whereas it was 13 times higher in WS (Fig. 7B).

Figure 7.

Figure 7.

Open in a new tab

Contents of coniferyl alcohol, sinapyl alcohol, coniferin, and syringin in Arabidopsis mature stems. A, Mean content of coniferyl alcohol, sinapyl alcohol, coniferin, and syringin in Arabidopsis mature stems (Col-0 and WS ecotypes) after UPLC-MS analysis. B, Relative content of coniferyl alcohol, sinapyl alcohol, coniferin, and syringin in the Col-0 ecotype with the WS ecotype reduced to 100%. Values correspond to relative intensities of each compound based on mass spectra.

Concerning the BGLU45 mutants, the amounts of 10 compounds were found to be significantly increased between mutant lines and the wild type (Col-0 and WS; Table III). Among them, five were identified or partially identified. All these compounds were glycosylated and derived from the G unit. The major compound affected by the down-regulation of BGLU45 was coniferin, both in WS and in Col-0. The coniferin content was more than 20 times higher in the bglu45-1 mutant line than in the WS line (Table III; Fig. 8). For the Col-0 ecotype, the coniferin level was increased three or four times in the mutant lines. In addition, vanillin 4-O-hexoside (a benzenoid with the same aromatic substituents as coniferin) was increased in bglu45-2 and bglu45-3, a compound identified as lariciresinol hexoside or isodihydrodehydrodiconiferyl alcohol hexoside in bglu45-3, dehydrodiconiferyl alcohol (DDC) hexoside in bglu45-1, and isodehydrodiconiferyl alcohol (IDDC) hexoside in bglu45-1, bglu45-2, and bglu45-3 mutants (Table III; Fig. 8).

Table III. Main compounds occurring in lower (boldface) or higher (lightface) amounts in the stems of the bglu45-1, bglu46-1, bglu46-2, and bglu47 mutants (WS background) and of the bglu45-2, bglu45-3, and bglu46-3 mutants (Col-0 background) relative to the control samples.

These compounds were separated, identified by UPLC-MS, and quantified from a specific ion selected on their mass spectra (deprotonated molecule [M−H]− or its acetate adduct [M−H+60]− or aglycone ion). IDDDC, Isodihydrodehydrodiconiferyl alcohol. The percentage of increase or decrease of the different compounds in the mutants relative to the wild type is shown in parentheses. The asterisk indicates the overexpressor line.

WS
 Col-0
Identification Retention Time Ion bglu45-1 bglu46-1* bglu46-2 bglu47 bglu45-2 bglu45-3 bglu46-3
min m/z %
4-O-Hexosylvanilin 6.5 373 (M−H)− +)206) +)219)
Coniferin 6.8 401 (M−H+60)− 160 +++)2238) +)189) +)383) +)285) +)239)
Unknown 7.3 − (57) − (34)
Unknown 9.7 597 +)266) +)181) +)160)
Unknown 11.5 535 +)262) +)197)
Unknown 12.1 743 +)251) +)178) +)175)
Lariciresinol hexose or IDDDC hexose 13.1 521 (M−H)− +)164)
Unknown 14.3 503 +)256)
DDC hexoside 14.3 579 (M−H+60)− +)178) +)145)
IDDC hexoside 14.5 581 (M−H+60)− +)517) +)276) +)216) +)173)
Unknown 15.3 549 +)175)
Open in a new tab
Figure 8.

Figure 8.

Open in a new tab

Main compounds deregulated in the stems of the bglu45-1 and bglu46-2 mutants (WS background) and of the bglu45-2, bglu45-3, and bglu46-3 mutants (Col-0 background) relative to the control samples after UPLC-MS analysis. Values correspond to relative intensities of each compound based on mass spectra. Asterisks represent statistically different values.

Six compounds were significantly increased in the BGLU46 knockout mutants compared with the wild type, three of which were identified or partially identified. The coniferin content was 1.9 and 2.9 times higher in bglu46-2 and bglu46-3, respectively, than in the wild type. The DDC hexoside was increased in bglu46-2, and the IDDC hexoside was increased in bglu46-3 (Table III; Fig. 8).

In bglu47, no identified compound was significantly different from the wild type (data not shown).

It is interesting that the contents of coniferyl alcohol, sinapyl alcohol, coniferaldehyde, sinapaldehyde, and syringin were not significantly different in the mutant lines compared with the wild type, either for BGLU45 or for BGLU46 and BGLU47 mutant lines. Moreover, no increase of these compounds was observed in the potential BGLU46-overexpressing line (Table III), but one unknown compound was decreased.

Root Analyses

As monolignol glucosides accumulate in light-grown roots of Arabidopsis (Hemm et al., 2004), the same analyses were performed on 15-d-old light-grown roots of the BGLU45 and BGLU46 mutants. As for the stems, around 100 compounds were profiled in the roots.

The compounds detected in the roots were different from those detected in stems. In WS and Col-0 light-grown roots, coniferin and syringin could be detected whereas coniferyl and sinapyl alcohols were not detected. In the two ecotypes, the syringin level was around four times lower than the coniferin level (Fig. 9). As for the stems, a difference between the two ecotypes was observed, but to a lower extent. The monolignol glucoside level was 6-fold lower in the WS than in the Col-0 light-grown roots (Table IV).

Figure 9.

Figure 9.

Open in a new tab

Mean content of coniferin and syringin in Arabidopsis light-grown roots (Col-0 and WS ecotypes) after UPLC-MS analysis. Values correspond to relative intensities of each compound based on mass spectra.

Table IV. Main compounds occurring in lower (boldface) or higher (lightface) amounts in the roots of the bglu45-2 and bglu45-3 mutants (Col-0 background) relative to the control samples.

These compounds were separated, identified by UPLC-MS, and quantified from a specific ion selected on their mass spectra (deprotonated molecule [M−H]− or its acetate adduct [M−H+60]− or aglycone ion). IDDDC, Isodihydrodehydrodiconiferyl alcohol. The percentage of increase or decrease of the different compounds in the mutants relative to the wild type is shown in parentheses.

Col-0
Identification Retention Time Ion bglu45-3 bglu45-2
min m/z %
Unknown 3.1 419 +)282)
Scopoletin derivate 3.7 191 (aglycone ion) +)184) +)158)
Ferulic acid hexoside 3.9 193 (aglycone ion) +)175) +)166)
Unknown 5.7 209 −(29)
Coniferin 6.3 179 (aglycone ion) +)151)
Sinapyl alcohol + hexose? 8.9 209 (aglycone ion) +)146)
Coniferaldehyde derivative? 9 177 (M−H)− +)188)
Lariciresinol hexose or IDDDC hexose 14.4 521 (M−H)− +)134)
Unknown 14.7 327 −(61)
S(8-8)S-Syringaresinol 16.4 417 (M−H)− −(48)
Open in a new tab

Globally, these roots seemed less affected in their phenolic metabolome than stems by the repression of these β-glucosidase genes. However, it must be taken into account that around three times fewer compounds were detected in light-grown roots compared with mature stems (Table IV; Fig. 10).

Figure 10.

Figure 10.

Open in a new tab

Main compounds deregulated in the roots of the bglu45-2 and bglu45-3 mutants (Col-0 background) relative to the control samples after UPLC-MS analysis. Values correspond to relative intensities of each compound based on mass spectra. Asterisks represent statistically different values.

Significant differences between the BGLU45 mutant lines and the wild type were only observed in the Col-0 ecotype. Ten compounds were significantly different in BGLU45 mutant roots when compared with the wild type, of which seven were identified or partially identified. Except for coniferin, these compounds were different from those deregulated in the mutant stems. Coniferin was significantly increased in the bglu45-3 mutant. Two hexosylated compounds, a scopoletin derivative, a ferulic acid derivative, and a coniferaldehyde derivative were significantly increased in bglu45-2 and bglu45-3, and the S(8-8)S-syringaresinol was decreased in bglu45-2. No differences were detected between BGLU46 and BGLU47 mutant roots and the wild type (data not shown).

DISCUSSION

In previous studies, the implication of monolignol glucosides in the lignification process has been repeatedly suggested (Boerjan et al., 2003). The hypothesis was that monolignol glucosides would be stored in the vacuole or transported from the cytosol to the cell wall and then deglucosylated by β-glucosidases before their polymerization in muro (Freundenberg et al., 1952; Sarkanen, 1971). However, recent studies suggest that the monolignol glucosides could be, at least in angiosperms, a storage form of the monolignols and not their direct transportation form (Miao and Liu, 2010; Kaneda et al., 2011; Liu et al., 2011; Liu, 2012).

Three Arabidopsis β-glucosidase genes have been identified (Xu et al., 2004) as closely related to the P. contorta β-glucosidase gene (Dharmawardhana et al., 1995, 1999). Two of these enzymes (BGLU45 and BGLU46) display an in vitro affinity for the monolignol glucosides (Xu et al., 2004; Escamilla-Treviño et al., 2006).

BGLU47 is closely related to BGLU45 and BGLU46. However, this gene is expressed at a lower level than BGLU45 and BGLU46 in stems, and a knockout mutant line presents no significant differences in lignin and phenolic metabolome. The expression profile of BGLU45 and BGLU46 showed that these genes are expressed in lignifying organs and tissues (i.e. in fibers and in xylem). The relative transcript abundance of BGLU45 was higher in Arabidopsis mature stems (stage 6.9) than BGLU46. BGLU45 seems to play a more important role in the stem than BGLU46, as there are more metabolic perturbations in the bglu45 mutant lines than in the bglu46 mutant lines. Within the stem, these two proteins seem to be mainly located in the interfascicular fibers and in the protoxylem, respectively, as determined by immunolocalization. No compensation of the lack of expression of one gene on the expression of the other β-glucosidase gene could be detected in the respective mutant lines (data not shown). This explains why mutations in BGLU45 or BGLU46 induced an increase in coniferin level.

However, the absence of the BGLU45 and BGLU46 proteins led only to a decrease of about 8% of the global glucosidase activity in Arabidopsis mature stem. This moderate decrease of the global glucosidase activity may be accounted for by the low specificity of the employed enzymatic test and/or by the redundancy between various stem-specific β-glucosidases in Arabidopsis. Indeed, according to the GeneCAT expression tool (http://genecat.mpg.de), at least four (BGLU18, BGLU26, and BGLU37/BGLU38) of the total β-glucosidase genes present in Arabidopsis are expressed in the stem (data not shown). Some of those β-glucosidases are expressed at a higher level than BGLU45 and BGLU46 in the stem, which could explain why the total β-glucosidase activity was nearly unchanged in the mutant lines.

The absence of BGLU45 and BGLU46 is not linked to any visible growth phenotype under either short-day or long-day conditions. However, lignin histochemistry revealed a particular phenotype in the mature stems of the bglu45-1 mutant grown in the long-day regime, as compared with the corresponding control (WS accession). In the basal region, the interfascicular fibers of this mutant displayed supplementary cell layers rich in G units. This result was corroborated by a weak but systematic increase in the global Klason lignin level of the extract-free stems, whereas no difference in lignin structure could be seen by thioacidolysis. These additional cell layers suggest that the bglu45-1 mutant is somehow affected in its developmental program in the WS accession. However, these additional lignified cells and the corresponding higher Klason lignin level do not argue for the participation of BGLU45 in lignification. In the bglu45 mutants, an increase in DDC glucoside and IDDC glucoside was observed. The DDC glucoside is a promoter of cell division (Binns et al., 1987), which could be related to the presence of the supplementary cell layers observed in the bglu45-1 mutant line. This phenotype was sometimes observed in the other bglu45 null mutants obtained in the Col-0 background, with a systematic tendency to a slight increase in the Klason lignin level together with a small increase of IDDC hexoside. This observation suggests that the impact of BGLU45 absence is lower in the Col-0 accession than in WS.

No lignin content and composition changes could be shown in the bglu46 mutants, either in the knockout mutants or in the overexpressor line. However, as both BGLU45 and BGLU46 are expressed in stems, although with a differential pattern of expression (BGLU45 mainly in fiber and BGLU46 in xylem), it might be of interest to silence both of them in order to observe a higher perturbation of the phenolic metabolism. These two genes are located in tandem on the first chromosome of Arabidopsis; therefore, it was not possible to obtain a double mutant by crossing the single mutants for each gene. However, the two β-glucosidases have complementary expression patterns, no compensation was observed at the transcriptional level, and the increase in coniferin and other G-derived glucoside contents was detected in bglu45 and bglu46 mutants, suggesting an equal importance of the two genes.

Coniferyl alcohol and sinapyl alcohol could be detected, but their m/z traces were much lower than those of their glucosides in Arabidopsis stems. The coniferin content of the stems was much higher than the syringin content, which was not easily detectable. As Arabidopsis lignin is mainly composed of G units (70%), the higher content of coniferin compared with syringin in stems is in accordance with a more important transport/storage of G units than of S units in the cell wall. However, the low amount of the aglycone forms of the monolignols is in favor of their direct incorporation in lignins.

The metabolome analyses showed important differences between the WS and Col-0 accessions for these compounds. The coniferin content was about 25 times higher in Col-0 than in WS. Furthermore, the ratio between the m/z traces of coniferin and coniferyl alcohol was 17.5-fold higher in Col-0 than in WS. The β-glucosidase activity was determined in stems of Col-0 and WS but was nearly the same in the two accessions (data not shown). These low differences in β-glucosidase activity detected between the two ecotypes cannot explain the variation in the content of these compounds. The content of coniferin being higher in Col-0 than in WS, the increase of coniferin induced by the absence of the BGLU45 and BGLU46 genes probably has a lower impact in Col-0 than in WS. Thus, an increase of two or three times the amount of coniferin in Col-0 mutant lines will have less impact than a 20-fold increase in the amount of coniferin in WS mutant lines, while the WS wild type already accepts much less coniferin than the Col-0 wild type. This could explain the “lignin deposition phenotype” observed for the bglu45-1 mutant line in WS and at a lower level in the bglu45 mutant lines in the Col-0 background.

Previous studies of BGLU45 and BGLU46 have shown the in vitro affinity of BGLU45 for the three monolignol glucosides (coniferin, syringin, and p-coumaryl alcohol 4-O-β-d-glucoside), with a high affinity for coniferin and syringin, whereas BGLU46 displays a larger in vitro substrate specificity, with a high affinity for salicilin (a phenolic glucoside), p-coumaryl alcohol 4-O-β-d-glucoside, and phenyl β-d-glucoside and a lower affinity for coniferin and syringin (Xu et al., 2004; Escamilla-Treviño et al., 2006). In vivo, studies of the mutant lines show a significant increase of coniferin in the stems of the bglu45 and bglu46 knockout mutants, whereas no difference could be detected for syringin. This could be due to the small amount of syringin in the Col-0 and WS wild types. This suggests that, unlike the results obtained in vitro, the favorite substrate in vivo for BGLU45 and BGLU46 would be coniferin in the stems. However, it must be taken into consideration that BGLU46 is specific to xylem tissues rich in G units. The β-glucosidase enzymes are also able to deglucosylate different other substrates in planta, but to a lower extent. These other substrates are all glycosylated compounds derived from guaiacyl derivatives. The absence of expression of BGLU46 leads to similar results to the absence of expression of BGLU45, with fewer compounds affected. This is consistent with the lower relative transcript abundance of BGLU46 compared with BGLU45 and its localization in xylem tissues. No significant difference was observed between the bglu46-1 overexpressor line and the wild type, suggesting that these enzymes are not limiting for the deglucosylation process.

In light-grown roots, as for mature stems, coniferin and syringin levels were higher in the Col-0 ecotype than in the WS ecotype. However, in the two ecotypes, the level of coniferin was three to five times higher than the level of syringin, which would be expected, as the root lignin is mainly composed of G units.

Only the bglu45 mutants in the Col-0 ecotype presented deregulations in monolignol derivatives in light-grown roots. As for the stem analyses, the metabolome analyses of the roots revealed an ecotype effect. In contrast to the results obtained in the stems, the coniferin was not deregulated in all the mutant lines by the absence of the β-glucosidase in the roots. Other glycosylated metabolites were also deregulated. This suggests that the impact of these enzymes is less important in light-grown roots than in stems. However, light-grown roots constitute an artificial system and may not reflect the normal root metabolism.

CONCLUSION

The two monolignol glucosides, coniferin and, to a lower extent, syringin, were detected in Arabidopsis stems. In contrast, the third monolignol glucoside (p-coumaryl alcohol glucoside) and also the coniferaldehyde or the sinapaldehyde glucosides could not be detected. The contents of monolignol glucosides and to a lesser extent of coniferyl and sinapyl alcohols were variable in the stems depending of the studied Arabidopsis ecotype.

The study of BGLU45 and BGLU46 mutants showed that the down-regulation of these genes, and particularly of BGLU45, increased the coniferin content of the stems (particularly in the WS ecotype), whereas it had no impact on the levels of syringin and of coniferyl and sinapyl alcohols. The fact that, in planta, BGLU45 and BGLU46 are able to deglucosylate coniferin is consistent with previous in vitro results (Xu et al., 2004; Escamilla-Treviño et al., 2006). The higher impact of the down-regulation of BGLU45 in the WS ecotype could be due to the lower content of coniferin in WS compared with Col-0 in normal growth conditions.

BGLU45 and BGLU46 mutations induced a large increase in coniferin (and not in syringin), whereas no impact on the corresponding alcohols was observed. These results are in favor of a role of the monolignol glucosides in the storage of monolignols and not as direct precursors of lignin. This work presents, to our knowledge, the first relative quantification of monolignol glucosides in Arabidopsis stems. It suggests that, at least in angiosperms, the monolignol glucosides could be a storage form of monolignols. The monolignols may be directly incorporated under their free form, as suggested by recent studies (Miao and Liu, 2010; Liu et al., 2011) in normal growth conditions. However, this storage pool could be used when lignins need to be synthesized de novo in response to various stresses. This possibility is suggested by the deregulation of the BGLU45 and BGLU46 genes in biotic stress conditions (http://www.genevestigator.com). This possibility needs future investigation.

MATERIALS AND METHODS

Plant Material and Growth Conditions

The Arabidopsis (Arabidopsis thaliana) accessions WS and Col-0 were used in this work. Mutants were identified in the Arabidopsis T-DNA insertion collection of Versailles (Bouché and Bouchez, 2001), in the Salk Institute collection (Alonso et al., 2003), and in the GABI-Kat collection (Li et al., 2003, 2007; Rosso et al., 2003).

For long-day conditions, transgenic and wild-type Arabidopsis plants were grown together in the same greenhouse under standard conditions (25°C day/15°C night, 16 h of light, 60% relative humidity) during 10 weeks.

For short-day conditions (conditions used for metabolic studies mainly), transgenic and wild-type Arabidopsis plants were grown together in growth chambers. Seeds were sown on soil (Saniflor ref. 252020) with 10% vermiculite. Each tray carried 10 replicates of the wild type and 10 replicates of each mutant line in a random way. After 2 d of vernalization at 4°C, plants were grown under short-days conditions (9 h of light) at 22°C with 55% relative humidity and a light intensity of 120 µE m−2 s−1. Conditions were changed to long days (16 h of light) after 2 months.

For root analysis in in vitro conditions, transgenic and wild-type Arabidopsis plants were grown together on vertical square plates. Seeds were sown, after sterilization, on Murashige and Skoog medium with Suc. Each plate carried 10 replicates of the wild type and 10 replicates of a mutant line. After 4 d of vernalization at 4°C, plantlets were grown in continuous light conditions at 21°C to 22°C with a light intensity of 110 µE m−2 s−1 (cool-white fluorescent tungsten tubes; OSRAM) during 15 d.

RNA Extraction and Semiquantitative RT-PCR

Total RNAs from different plant organs and seedlings were extracted using the Qiagen RNeasy plant kit and treated by the DNase I kit (Sigma-Aldrich). RNAs were quantified by measuring their A260 with a spectrophotometer.

Complementary DNAs (cDNAs) were synthesized using 1 µg of RNA in a 20-µL volume with the SuperScript First Strand Synthesis System for RT-PCR (Invitrogen) according to the manufacturer’s instructions. The cDNA were quantified using β-TUBULIN4 (At5g44340)-specific primers. The relative transcript abundance was studied by RT-PCR or qRT-PCR using gene-specific primers (Supplemental Table S1) according to standard protocols using a Mastercycler Realplex (Eppendorf).

In Situ Hybridization

Segments of Arabidopsis stems and roots were fixed in 4% formaldehyde in phosphate-buffered saline (PBS) under vacuum for 1 h and left in new fixative overnight. After fixation, the sections were washed, dehydrated in a series of ethanol concentrations (10%–95% ethanol), and then embedded in paraffin (Jackson, 1991). Paraffin transverse sections (8–10 µm) were cut on a microtome (Leica) and mounted on slides (Dako). Sense and antisense probes of cDNA parts of BGLU45 and BGLU46 were synthesized using DIG-UTP (Roche) according to the manufacturer’s instructions. In situ hybridizations were carried out as described by Jackson (1991) with some changes: after dehydration of the tissues, a prehybridization step was performed on the slide (1 h at hybridization temperature in 50% formamide, 5× saline-sodium citrate, 100 µg mL−1 transfer RNA, 50 µg mL−1 heparin, and 0.1% Tween). Immunodetection of the DIG-labeled probe was performed with an anti-DIG antibody coupled with alkaline phosphatase using the nitroblue tetrazolium-5-bromo-4-chloro-3-indolyl phosphate kit according to the manufacturer’s instructions (Roche). Primer sequences used for the DIG-labeled probe amplification are shown in Supplemental Table S1.

Lignin Study

Lignin stainings were performed on the base of mature stems of Arabidopsis cross sections (60 µm thickness) obtained with a Vibratome (Leica VT1000S). Stem sections from wild-type and transgenic lines were observed after Wiesner and Maüle reactions according to standard protocols.

Lignin analyses were carried out on the inflorescence stems collected at full maturity and ground to 0.5 mm. Extractive-free samples were prepared using a Soxhlet apparatus by sequentially extracting the ground material with toluene:ethanol (2:1, v/v), ethanol, and water. Lignin content of dried mature stems was estimated on the extractive-free samples using the standard Klason procedure (Dence, 1992). Lignin structure was studied on the extract-free material using the thioacidolysis procedure (Lapierre et al., 1995, 1999). The lignin-derived thioacidolysis monomers were identified by gas chromatography-MS as their trimethylsilylated derivatives. The quantitative analysis of guaiacyl (G) and syringyl (S) thioacidolysis monomers, analyzed as their trimethylsilylated derivatives, was carried out from specific ion chromatograms reconstructed at m/z 269 for G monomers and 299 for S monomers.

One to three biological replicates (corresponding to a pool of 24 plants at each time) were analyzed two times separately.

Protein Extraction

Total protein extracts were obtained by homogenization of fresh tissues in extraction buffer consisting of 25 mm BisTris (pH 7.0), 200 mm CaCl2, 10% (v/v) glycerol, 4 µm sodium-cacodylate, and 1:200 (v/v) protease inhibitor cocktail (P-9599; Sigma Chemical). The material was centrifuged twice at 8°C for 10 min at 10,000g. The supernatant was collected and centrifuged again for 45 min at 15,000g.

Protein purification was performed by affinity chromatography on a 0.5- × 3-cm column filled with 1 mL of concanavalin A-Sepharose (Sigma) and washed with 3 mL of extraction buffer. The soluble protein extract was loaded, and the column was washed with 10 mL of 20 mm Tris-HCl, 0.5 m NaCl buffer (pH 7.4). The proteins were eluted with 0.3 m methyl-α-glucopyranoside in the same buffer. The eluates were collected (1 mL per fraction) and tested for enzymatic activity. Pooled fractions showing enzymatic activity were equilibrated in Tris-HCl buffer (pH 7.4) containing 5% glycerol (v/v) and 0.015% Triton X-100 (v/v). Glycerol was added to the buffer to prevent partial inactivation of the enzyme. Protein content in the extracts was quantified according to Bradford (1976).

Antibodies

Specific antibodies were generated by the Eurogenetech Society from peptide sequences of BGLU45 and BGLU46 (Supplemental Fig. S4).

Western-Blot Analysis

Protein samples (15 µg) were heated at 100°C for 5 min in charging buffer. The resulting samples were cooled, briefly centrifuged, and the supernatant was loaded on a 10% acrylamide SDS-PAGE device with a 5% resolving gel using a Bio-Rad Protein II apparatus. SDS-PAGE was run at 60 V for 1 h and then at 80 V for 2 h.

Proteins were transferred onto a 0.45-mm Hybond ECL membrane (Hybond C+; Amersham Bioscience) using an electroblotting apparatus (Bio-Rad Laboratories). Protein detection was performed using primary monoclonal antibodies raised against BGLU45 or BGLU46 (3.6 × 10−7 m) and alkaline phosphatase-conjugated secondary antibodies. Blots were developed using the nitroblue tetrazolium-5-bromo-4-chloro-3-indolyl phosphate kit (Roche).

β-Glucosidase Activities

Crude extracts were purified by passage on a concanavalin A chromatography column, as BGLU45 and BGLU46 were predicted to be glycosylated proteins. The β-glucosidase activity was determined at 30°C toward the synthetic substrate p-nitrophenyl-β-d-glucopyranoside at 15 mm. The reaction mixture contained 200 mm acetate buffer (pH 5). The reaction product was measured spectrophotometrically at 405 nm.

Immunoblot analyses and β-glucosidase activity quantification were performed on the same two biological replicates (each biological replicate corresponding to a pool of 24 plants). Each experiment was performed three times.

Immunolocalization

Samples for immunofluorescence microscopy were prepared according to the method of Baluska et al. (1992). They were fixed in 4% formaldehyde in PBS for 1 h under vacuum at room temperature. The stems were dehydrated in a series of ethanol concentrations (10%–95% ethanol in PBS), then transferred to ethanol:histoclear (2:1, 1:1, and 1:2 steps), histoclear, histoclear:paraffin (1:1), and finally three different paraffin baths.

Transverse sections of 12 µm were obtained with a microtome (Leica) and mounted on slides. The paraffin was eliminated with histoclear, and the sections were rehydrated in a series of ethanol concentrations (100%–30% ethanol in PBS).

The sections for immunostaining were blocked for 30 min with bovine serum albumin to saturate the nonspecific binding sites. The primary antibody raised against BGLU45 or BGLU46 was diluted in bovine serum albumin/PBS, and the samples were incubated for 2 h at 37°C. The sections were washed in PBS and then left at room temperature for 1 h with the secondary antibody (goat anti-rabbit IgG linked to Alexa Fluor 568 [dilution, 1:500]; Molecular Probes, Invitrogen). The labeled sections were viewed with a confocal laser-scanning microscope (Leica SP2 AOBS).

UPLC-MS Analyses

Data Acquisition

Frozen samples of Arabidopsis were ground in liquid nitrogen and extracted in 90:10 methanol:water. A second extraction was then performed in cyclohexane:water, and the liquid phase was used for reverse-phase liquid chromatography-MS analysis. A total of 15 µL of each sample was injected on a Waters Acquity UPLC system equipped with an Acquity UPLC BEH C18 (2.1 × 100 mm, 1.7 µm) column.

A gradient of two buffers was used: buffer A (100:1:0.1, water:acetonitrile:ammonium acetate [2 mol L−1]; pH 5) and an organic buffer B (100:1:0.1, acetonitrile:water:ammonium acetate [2 mol L−1]; pH 5). The gradient applied started at 95% A for 0.1 min and decreased to 55% A in 36.9 min (flow, 200 µL min−1).

The column temperature was set at 40°C. A UV/visible absorption spectrum was measured between 190 and 600 nm. An Atmospheric Pressure Chemical Ionization device, operated in the negative ion mode, was used as an ion source to couple UPLC with an ion trap MS instrument (LCQ Classic; Thermoquest), with vaporizer temperature of 450°C, capillary temperature of 150°C, source current of 5 mA, sheath gas flow of 21, and auxiliary gas flow of 3). Ions in the mass range between 135 and 1,000 m/z were measured. At least eight replicates of mutant and wild-type lines were analyzed randomly in order to perform statistical analyses.

Data Analysis

After acquisition of the data (corresponding to 15,000 m/z for the stems and 5,000 m/z for the roots), the nonlinear retention time and alignment were corrected using the Metalign software (http://www.metalign.wur.nl/UK/). The corrected results were imported into Microsoft Excel and then filtered. The early-eluted (between 1 and 3 min) samples were removed. Merely consistent signals were analyzed: only the signals that were present at least in six replicates for a single genotype were taken into account, the other ones being rejected as noise. Then, the remaining data were normalized to the sample dry weight. A correlation test in R (http://www.r-project.org/) permitted us to regroup all the correlated signals and to keep only the most important ones for further statistical analyses. Then, Principal Component Analysis, ANOVA, and Student’s t test analyses were performed on the remaining data. The ANOVA results were corrected with the multiple testing of Benjamini and Hochberg, in R, at a significance level of P < 0.01.

Compounds with a statistically different signal were identified by their retention times, their mass spectral data, and their UV spectra. For the oligolignols, data were compared with previously identified compounds (Morreel et al., 2004).

Supplemental Data

The following materials are available in the online version of this article.

  • Supplemental Figure S1. BGLU45, BGLU46, and BGLU47 expression profiles obtained from the GeneCAT database.

  • Supplemental Figure S2. In situ hybridization experiments in inflorescence stem cross sections from Col-0 plants grown under long-day conditions.

  • Supplemental Figure S3. Maüle staining of stem cross sections from wild-type (WS accession) and bglu45-1 plants grown in short-day conditions.

  • Supplemental Figure S4. Peptide sequences used for the generation of BGLU45 and BGLU46 antibodies.

  • Supplemental Table S1. Sequence of primers used for PCR.

  • Supplemental Table S2. Coexpression of the BGLU45 and BGLU46 genes with genes implicated or potentially implicated in the secondary cell wall formation in Arabidopsis.

Supplementary Material

Supplemental Data
supp_160_3_1204__index.html (979B, html)

Acknowledgments

We thank the greenhouse management team (Unité Mixte de Recherche 1318 Institut National de la Recherche Agronomique-AgroParisTech) for growing the plants.

Glossary

GH1

Glycosyl Hydrolase Family 1

qRT

quantitative reverse transcription

q

quantitative

RT

reverse transcription

DIG

digoxigenin

Col-0

Columbia

WS

Wassilewskija

UPLC

ultra-performance liquid chromatography

MS

mass spectrometry

m/z

mass-to-charge ratio

DDC

dehydrodiconiferyl alcohol

IDDC

isodehydrodiconiferyl alcohol

PBS

phosphate-buffered saline

cDNA

complementary DNA

References

  1. Alonso JM, Stepanova AN, Leisse TJ, Kim CJ, Chen H, Shinn P, Stevenson DK, Zimmerman J, Barajas P, Cheuk R, et al. (2003) Genome-wide insertional mutagenesis of Arabidopsis thaliana. Science 301: 653–657 [DOI] [PubMed] [Google Scholar]
  2. Baluska F, Parker JS, Barlow PW. (1992) Specific patterns of cortical and endoplasmic microtubules associated with cell growth rearrangements of F-actin arrays in growing cells of intact maize root apex tissues: a major developmental switch occurs in the postmitotic transition region. Eur J Cell Biol 72: 113–121 [PubMed] [Google Scholar]
  3. Bednarek P, Schneider B, Svatos A, Oldham NJ, Hahlbrock K. (2005) Structural complexity, differential response to infection, and tissue specificity of indolic and phenylpropanoid secondary metabolism in Arabidopsis roots. Plant Physiol 138: 1058–1070 [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Binns AN, Chen RH, Wood HN, Lynn DG. (1987) Cell division promoting activity of naturally occurring dehydrodiconiferyl glucosides: do cell wall components control cell division? Proc Natl Acad Sci USA 84: 980–984 [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Boerjan W, Ralph J, Baucher M. (2003) Lignin biosynthesis. Annu Rev Plant Biol 54: 519–546 [DOI] [PubMed] [Google Scholar]
  6. Bouché N, Bouchez D. (2001) Arabidopsis gene knockout: phenotypes wanted. Curr Opin Plant Biol 4: 111–117 [DOI] [PubMed] [Google Scholar]
  7. Boyes DC, Zayed AM, Ascenzi R, McCaskill AJ, Hoffman NE, Davis KR, Görlach J. (2001) Growth stage-based phenotypic analysis of Arabidopsis: a model for high throughput functional genomics in plants. Plant Cell 13: 1499–1510 [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Bradford MM. (1976) A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal Biochem 72: 248–254 [DOI] [PubMed] [Google Scholar]
  9. Burmeister G, Hösel W. (1981) Immunohistochemical localization of β-glucosidases in lignin and isoflavone metabolism in Cicer arietinum L. seedlings. Planta 152: 578–586 [DOI] [PubMed] [Google Scholar]
  10. Dence C. (1992) Lignin determination. In C Dence, S Lin, eds, Methods in Lignin Biochemistry. Springer-Verlag, Berlin, pp 33–61
  11. Dharmawardhana DP, Ellis BE, Carlson JE. (1995) A β-glucosidase from lodgepole pine xylem specific for the lignin precursor coniferin. Plant Physiol 107: 331–339 [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Dharmawardhana DP, Ellis BE, Carlson JE. (1999) cDNA cloning and heterologous expression of coniferin β-glucosidase. Plant Mol Biol 40: 365–372 [DOI] [PubMed] [Google Scholar]
  13. Escamilla-Treviño LL, Chen W, Card ML, Shih MC, Cheng CL, Poulton JE. (2006) Arabidopsis thaliana β-glucosidases BGLU45 and BGLU46 hydrolyse monolignol glucosides. Phytochemistry 67: 1651–1660 [DOI] [PubMed] [Google Scholar]
  14. Freundenberg K, Harkin JM. (1963) The glucosides of cambial sap of spruce. Phytochemistry 2: 189–193 [Google Scholar]
  15. Freundenberg K, Reznik H, Boesenberg H, Rasenack D. (1952) Das an der verholzung beteiligte Fermentsystem. Chem Ber 85: 641–647 [Google Scholar]
  16. Fukushima K, Taguchi S, Matsui N, Yasuda S. (1997) Distribution and seasonal changes of monolignol glucosides in Pinus thunbergii. Mokuzai Gakkaishi 43: 254–259 [Google Scholar]
  17. Fukushima K, Terashima N. (1990) Heterogeneity in formation of lignin. XIII. Formation and structure of p-hydroxyphenyl lignin in various hardwoods visualized by microautoradiography. J Wood Chem Technol 10: 413–433 [Google Scholar]
  18. Fukushima K, Terashima N. (1991a) Heterogeneity in formation of lignin. XIV. Formation and structure of lignin in differentiating xylem of Ginkgo biloba. Holzforschung 45: 87–94 [Google Scholar]
  19. Fukushima K, Terashima N. (1991b) Heterogeneity in formation of lignin. XV. Formation and structure of lignin in compression wood of Pinus thunbergii studied by microautoradiography. Wood Sci Technol 25: 371–381 [Google Scholar]
  20. Grisebach H. (1981) Lignins. In PK Stumpf, MD Hatch, eds, The Biochemistry of Plants, Vol 7. New York, pp 457–478
  21. Hemm MR, Rider SD, Ogas J, Murry DJ, Chapple C. (2004) Light induces phenylpropanoid metabolism in Arabidopsis roots. Plant J 38: 765–778 [DOI] [PubMed] [Google Scholar]
  22. Higuchi R, Aritomi M, Donnelly DMX. (1977) Monolignol and dilignol glycosides from Pinus contorta leaves. Phytochemistry 16: 1007–1011 [Google Scholar]
  23. Hosel W, Fiedler-Preiss A, Borgmann E. (1982) Relationship of coniferin β-glucosidase to lignification in various plant cell suspension cultures. Plant Cell Tissue Organ Cult 1: 137–148 [Google Scholar]
  24. Hösel W, Surholt E, Borgmann E. (1978) Characterization of β-glucosidase isoenzymes possibly involved in lignification from chick pea (Cicer arietinum L.) cell suspension cultures. Eur J Biochem 84: 487–492 [DOI] [PubMed] [Google Scholar]
  25. Hosel W, Todenhagen R. (1980) Characterization of a β-glucosidase from Glycine max which hydrolyses coniferin and syringin. Phytochemistry 19: 1349–1353 [Google Scholar]
  26. Huis R, Morreel K, Fliniaux O, Lucau-Danila A, Fénart S, Grec S, Neutelings G, Chabbert B, Mesnard F, Boerjan W, et al. (2012) Natural hypolignification is associated with extensive oligolignol accumulation in flax stems. Plant Physiol 158: 1893–1915 [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Jackson D. (1991) In situ hybridization in plants. In D Bowles, S Gurr, M McPherson, eds, Molecular Plant Pathology: A Practical Approach. Oxford University Press, Oxford, pp 163–174
  28. Kaneda M, Rensing KH, Wong JC, Banno B, Mansfield SD, Samuels AL. (2008) Tracking monolignols during wood development in lodgepole pine. Plant Physiol 147: 1750–1760 [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Kaneda M, Schuetz M, Lin BS, Chanis C, Hamberger B, Western TL, Ehlting J, Samuels AL. (2011) ABC transporters coordinately expressed during lignification of Arabidopsis stems include a set of ABCBs associated with auxin transport. J Exp Bot 62: 2063–2077 [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Kim MR, Moon HT, Lee DG, Woo ER. (2007) A new lignan glycoside from the stem bark of Styrax japonica S. et Z. Arch Pharm Res 30: 425–430 [DOI] [PubMed] [Google Scholar]
  31. Lapierre C, Pollet B, Petit-Conil M, Toval G, Romero J, Pilate G, Leple JC, Boerjan W, Ferret VV, Jouanin L, et al. (1999) Structural alterations of lignins in transgenic poplars with depressed cinnamyl alcohol dehydrogenase or caffeic acid O-methyltransferase activity have an opposite impact on the efficiency of industrial kraft pulping. Plant Physiol 119: 153–164 [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Lapierre C, Pollet B, Rolando R. (1995) New insights into the molecular architecture of hardwood lignins by chemical degradation methods. Res Chem Intermed 21: 397–412 [Google Scholar]
  33. Leinhos V, Savidge RA. (1993) Isolation of protoplasts from developing xylem of Pinus banksiana and Pinus strobus. Can J For Res 23: 343–348 [Google Scholar]
  34. Leinhos V, Savidge RA. (1994) Investigation of coniferin compartmentation in developing xylem of conifers during lignification. Acta Hortic 381: 97–103 [Google Scholar]
  35. Li Y, Rosso MG, Strizhov N, Viehoever P, Weisshaar B. (2003) GABI-Kat SimpleSearch: a flanking sequence tag (FST) database for the identification of T-DNA insertion mutants in Arabidopsis thaliana. Bioinformatics 19: 1441–1442 [DOI] [PubMed] [Google Scholar]
  36. Li Y, Rosso MG, Viehoever P, Weisshaar B. (2007) GABI-Kat SimpleSearch: an Arabidopsis thaliana T-DNA mutant database with detailed information for confirmed insertions. Nucleic Acids Res 35: D874–D878 [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Liu CJ. (2012) Deciphering the enigma of lignification: precursor transport, oxidation, and the topochemistry of lignin assembly. Mol Plant 5: 304–317 [DOI] [PubMed] [Google Scholar]
  38. Liu CJ, Miao YC, Zhang KW. (2011) Sequestration and transport of lignin monomeric precursors. Molecules 16: 710–727 [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Malarz J, Stojakowska A, Szneler E, Kisiel W. (2005) Furofuran lignans from a callus culture of Cichorium intybus. Plant Cell Rep 24: 246–249 [DOI] [PubMed] [Google Scholar]
  40. Marcinowski S, Falk H, Hammer DK, Hoyer B, Grisebach H. (1979) Appearance and localization of a β-glucosidase hydrolyzing coniferin in spruce (Picea abies) seedlings. Planta 144: 161–165 [DOI] [PubMed] [Google Scholar]
  41. Marcinowski S, Grisebach H. (1977) Turnover of coniferin in pine seedlings. Phytochemistry 16: 1665–1667 [Google Scholar]
  42. Marcinowski S, Grisebach H. (1978) Enzymology of lignification: cell-wall bound β-glucosidase for coniferin from spruce (Picea abies) seedlings. Eur J Biochem 87: 37–44 [DOI] [PubMed] [Google Scholar]
  43. Marjamaa K, Lehtonen M, Lundell T, Toikka M, Saranpää P, Fagerstedt KV. (2003) Developmental lignification and seasonal variation in β-glucosidase and peroxidase activities in xylem of Scots pine, Norway spruce and silver birch. Tree Physiol 23: 977–986 [DOI] [PubMed] [Google Scholar]
  44. Miao YC, Liu CJ. (2010) ATP-binding cassette-like transporters are involved in the transport of lignin precursors across plasma and vacuolar membranes. Proc Natl Acad Sci USA 107: 22728–22733 [DOI] [PMC free article] [PubMed] [Google Scholar]
  45. Morreel K, Ralph J, Kim H, Lu F, Goeminne G, Ralph S, Messens E, Boerjan W. (2004) Profiling of oligolignols reveals monolignol coupling conditions in lignifying poplar xylem. Plant Physiol 136: 3537–3549 [DOI] [PMC free article] [PubMed] [Google Scholar]
  46. Rolando C, Daubresse N, Pollet B, Jouanin L, Lapierre C. (2004) Lignification in poplar plantlets fed with deuterium-labelled lignin precursors. C R Biol 327: 799–807 [DOI] [PubMed] [Google Scholar]
  47. Rosso MG, Li Y, Strizhov N, Reiss B, Dekker K, Weisshaar B. (2003) An Arabidopsis thaliana T-DNA mutagenized population (GABI-Kat) for flanking sequence tag-based reverse genetics. Plant Mol Biol 53: 247–259 [DOI] [PubMed] [Google Scholar]
  48. Sarkanen KV. (1971) Precursors and their polymerization. In KV Sarkanen, CH Ludwig, eds, Lignins: Occurrence, Formation, Structure and Reactions. Wiley Interscience, New York, pp 95–163
  49. Sticher O, Lahloub MF. (1982) Phenolic glycosides of Paulownia tomentosa bark. Planta Med 46: 145–148 [DOI] [PubMed] [Google Scholar]
  50. Terashima N, Fukushima K. (1988) Homogeneity in formation of lignin. XI. An autoradiographic study of the heterogeneous formation and structure of pine lignin. Wood Sci Technol 22: 259–270 [Google Scholar]
  51. Terashima N, Fukushima K. (1989) Biogenesis and structure of macromolecular lignin in the cell wall of tree xylem as studied by microautoradiography. In NG Lewis, MG Paise, eds, Plant Cell Wall Polymers, Biogenesis and Biodegradation. American Chemical Society, Washington, DC, pp 160–168
  52. Terashima N, Fukushima K, He LF, Takabe K. (1993) Comprehensive model of the lignified plant cell wall. In HG Jung, DR Buxton, RD Hatfield, J Ralph, eds, Forage Cell Wall Structure and Digestibility. American Society of Agronomy, Madison, WI, pp 247–270
  53. Terashima N, Fukushima K, Imai T. (1992) Morphological origin of milled wood lignin studied by radiotracer method. Holzforschung 46: 271–275 [Google Scholar]
  54. Terashima N, Fukushima K, Sano Y, Takabe K. (1988) Heterogeneity in formation of lignin. X. Visualization of lignification process in differentiating xylem of pine by microautoradiography. Holzforschung 42: 347–350 [Google Scholar]
  55. Terashima N, Fukushima K, Takabe K. (1986) Heterogeneity in formation of lignin. VIII. An autoradiographic study on the formation of guaiacyl and syringyl lignin in Magnolia kobus DC. Holzforschung 40: 101–105 [Google Scholar]
  56. Terazawa M, Okuyama H, Miyake M. (1984a) Phenolic compounds in living tissues of woods. I. Phenolic β-glucosides of 4-hydroxycinnamyl alcohol derivatives in the cambial sap of woods. Mokuzai Gakkaishi 30: 322–328 [Google Scholar]
  57. Terazawa M, Okuyama H, Miyake M. (1984b) Isolation of coniferin and syringin from cambial tissue and inner-bark of some angiospermous woods. Mokuzai Gakkaishi 30: 409–412 [Google Scholar]
  58. Tsuji Y, Chen F, Yasuda S, Fukushima K. (2004) The behavior of deuterium-labeled monolignol and monolignol glucosides in lignin biosynthesis in angiosperms. J Agric Food Chem 52: 131–134 [DOI] [PubMed] [Google Scholar]
  59. Tsuji Y, Chen F, Yasuda S, Fukushima K. (2005) Unexpected behavior of coniferin in lignin biosynthesis of Ginkgo biloba L. Planta 222: 58–69 [DOI] [PubMed] [Google Scholar]
  60. Tsuji Y, Fukushima K. (2004) Behavior of monolignol glucosides in angiosperms. J Agric Food Chem 52: 7651–7659 [DOI] [PubMed] [Google Scholar]
  61. Xu Z, Escamilla-Treviño L, Zeng L, Lalgondar M, Bevan D, Winkel B, Mohamed A, Cheng CL, Shih MC, Poulton J, et al. (2004) Functional genomic analysis of Arabidopsis thaliana glycoside hydrolase family 1. Plant Mol Biol 55: 343–367 [DOI] [PubMed] [Google Scholar]
  62. Yang MC, Lee KH, Kim KH, Choi SU, Lee KR. (2007) Lignan and terpene constituents from the aerial parts of Saussurea pulchella. Arch Pharm Res 30: 1067–1074 [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplemental Data
supp_160_3_1204__index.html (979B, html)
supp_pp.112.203364_203364Supplemental_data_pp.pdf (4.6MB, pdf)

Articles from Plant Physiology are provided here courtesy of Oxford University Press

RESOURCES