Characterisation of a pine MYB that regulates lignification


  • Astrid Patzlaff,

    1. Department of Plant Sciences, University of Oxford, South Parks Road, Oxford OX1 3RB, UK,
    Search for more papers by this author
      These authors contributed equally to this work.
  • Stephanie McInnis,

    1. Department of Plant Sciences, University of Oxford, South Parks Road, Oxford OX1 3RB, UK,
    Search for more papers by this author
      Current address: School of Biological Sciences, University of Bristol, Woodland Road, Bristol BS8 1UG, UK. These authors contributed equally to this work.
  • Adrian Courtenay,

    1. Department of Plant Sciences, University of Oxford, South Parks Road, Oxford OX1 3RB, UK,
    Search for more papers by this author
      Current address: Schroder Investment Management Limited, 31 Gresham Street, London EC2V 7QA, UK. These authors contributed equally to this work.
  • Christine Surman,

    1. Department of Plant Sciences, University of Oxford, South Parks Road, Oxford OX1 3RB, UK,
    Search for more papers by this author
  • Lisa J. Newman,

    1. Department of Plant Sciences, University of Oxford, South Parks Road, Oxford OX1 3RB, UK,
    Search for more papers by this author
      Current address: Pioneer Hi-Bred International, Inc., 7250 NW 62nd Avenue, PO Box 552, Johnston, IA 50131-0552, USA.
  • Caroline Smith,

    1. Department of Cell & Developmental Biology, John Innes Centre, Norwich Research Park, Colney, Norwich, Norfolk NR4 7UH, UK,
    Search for more papers by this author
  • Michael W. Bevan,

    1. Department of Cell & Developmental Biology, John Innes Centre, Norwich Research Park, Colney, Norwich, Norfolk NR4 7UH, UK,
    Search for more papers by this author
  • Shawn Mansfield,

    1. Department of Wood Science, University of British Columbia, 4030-2424 Main Mall, Vancouver, BC, Canada V6T 1Z4, and
    Search for more papers by this author
  • Ross W. Whetten,

    1. Forest Biotechnology Laboratory, Department of Forestry, North Carolina State University, 2500 Partners II, 840 Main Campus Drive, Centennial Campus, Box 7247, Raleigh, NC 27695, USA
    Search for more papers by this author
  • Ronald R. Sederoff,

    1. Forest Biotechnology Laboratory, Department of Forestry, North Carolina State University, 2500 Partners II, 840 Main Campus Drive, Centennial Campus, Box 7247, Raleigh, NC 27695, USA
    Search for more papers by this author
  • Malcolm M. Campbell

    Corresponding author
    1. Department of Plant Sciences, University of Oxford, South Parks Road, Oxford OX1 3RB, UK,
      For correspondence (fax +44 1865 275074; e-mail
    Search for more papers by this author

For correspondence (fax +44 1865 275074; e-mail

Current address: School of Biological Sciences, University of Bristol, Woodland Road, Bristol BS8 1UG, UK.

Current address: Schroder Investment Management Limited, 31 Gresham Street, London EC2V 7QA, UK.

Current address: Pioneer Hi-Bred International, Inc., 7250 NW 62nd Avenue, PO Box 552, Johnston, IA 50131-0552, USA.

These authors contributed equally to this work.


A member of the R2R3-MYB family of transcription factors was cloned from a cDNA library constructed from RNA isolated from differentiating pine xylem. This MYB, Pinus taeda MYB4 (PtMYB4), is expressed in cells undergoing lignification, as revealed by in situ RT-PCR. Electrophoretic mobility shift assays (EMSAs) showed that recombinant PtMYB4 protein is able to bind to DNA motifs known as AC elements. AC elements are ubiquitous in the promoters of genes encoding lignin biosynthetic enzymes. Transcriptional activation assays using yeast showed that PtMYB4 could activate transcription in an AC-element-dependent fashion. Overexpression of PtMYB4 in transgenic tobacco plants altered the accumulation of transcripts corresponding to genes encoding lignin biosynthetic enzymes. Lignin deposition increased in transgenic tobacco plants that overexpressed PtMYB4, and extended to cell types that do not normally lignify. Taken together, these findings are consistent with the hypothesis that PtMYB4 is sufficient to induce lignification, and that it may play this role during wood formation in pine.


Lignins are complex three-dimensional phenolic polymers that are embedded in the cell walls of specialised plant cells (Boudet, 2000; Campbell and Sederoff, 1996; Douglas, 1996; Lewis and Yamamoto, 1990; Ros Barcelo, 1997). Lignins function as inter- and intramolecular glues, rigidifying the plant cell walls in which they are embedded. Lignins are hydrophobic and render cell walls impermeable to water (Hose et al., 2001; Reina et al., 2001). The complex nature of lignins makes them difficult to degrade; therefore, lignified cell walls are more resistant to enzymatic hydrolysis (Higuchi, 1971; Lewis and Yamamoto, 1990; Trojanowski, 2001). As a consequence of these different properties, lignins play important roles in mechanical support, water transport and disease resistance in plants.

The spatial and temporal control of lignin biosynthesis, known as lignification, is extremely important. Lignification is a metabolically costly process that requires large quantities of carbon skeletons and reducing equivalents. Plants do not possess a mechanism to degrade lignins (Lewis and Yamamoto, 1990); so, any carbon invested in lignin biosynthesis is not recoverable. Consequently, lignified cells represent a significant, non-recoverable carbon sink. As such, plants must carefully balance the requirement for lignification against the availability of resources to synthesise lignin polymers. Moreover, as lignin restricts the expansion of the cell wall, lignification must occur after a cell has undergone division and expansion growth. Given the metabolic cost of making the lignin polymer, coupled with its persistence and cell-rigidifying properties, the timing and localisation of lignification must be tightly regulated. Therefore, mechanisms must be in place to regulate the lignin biosynthetic pathway.

While there is still some controversy regarding the exact order and nature of some of the enzymes involved in lignin biosynthesis, it is now thought that the monomeric precursors of lignins are synthesised through the activity of a metabolic grid (Humphreys and Chapple, 2002). The grid includes the enzymes of general phenylpropanoid metabolism, a biosynthetic pathway that is shared with many phenolic compounds. The general phenylpropanoid biosynthetic pathway is comprised of the entry-point enzyme, phenylalanine ammonia-lyase (PAL), as well as cinnamate 4-hydroxylase (C4H) and hydroxycinnamate:CoA ligase (4CL). The portion of the grid that is involved in the synthesis of the monomeric precursors of lignins, the monolignols, includes the enzymes quinate/shikimate O-hydroxycinnamoyltransferase (HCT), coumaroyl-quinate/shikimate 3-hydroxylase (C3H), hydroxycinnamoyl-CoA reductase (CCR) and hydroxycinnamyl alcohol dehydrogenase (CAD). The enzymes caffeate O-methyltransferase (COMT) and caffeoyl-CoA O-methyltransferase (CCoAOMT) are also components of the lignin biosynthetic grid, but their precise placement within the grid remains somewhat controversial (Humphreys and Chapple, 2002). Most of the genes encoding enzymes in the lignin metabolic grid have been characterised. Mechanisms underpinning the spatial and temporal control of lignification have been revealed by the investigation of the transcriptional regulation of many of these genes.

Analysis of the 5′ regulatory regions of several genes encoding lignin biosynthetic enzymes reveals a common theme. The promoters of these genes are characterised by the presence of conserved motifs, known either as AC elements, H-boxes or PAL boxes (Bell-Lelong et al., 1997; Hatton et al., 1995; Hauffe et al., 1993; Joos and Hahlbrock, 1992; Lacombe et al., 2000; Lauvergeat et al., 2002; Leyva et al., 1992; Logemann et al., 1995; Lois et al., 1989; Seguin et al., 1997). These motifs derive their name based on the fact that they are rich in cytosine and adenosine (and occasionally thymine) residues, which are found in the promoters of genes encoding enzymes including not only PAL, but also C4H, COMT, CCoAOMT, 4CL, CCR and CAD (Bell-Lelong et al., 1997; Chen et al., 2000; Hatton et al., 1995; Hauffe et al., 1993; Joos and Hahlbrock, 1992; Lacombe et al., 2000; Lauvergeat et al., 2002; Leyva et al., 1992; Logemann et al., 1995; Lois et al., 1989; Seguin et al., 1997). The ubiquity of these elements in the promoters of these genes suggests that they might function as a target for a common regulatory mechanism, which is involved in the coordinate control of the corresponding biosynthetic pathway.

At least some AC elements are important for xylem-localised gene expression. Xylem cells are amongst the best characterised of all lignified cells. Lignification is an integral and carefully coordinated component of xylem cell differentiation (Fukuda, 1996; Mellerowicz et al., 2001). Studies of the promoters of the genes encoding PAL, 4CL, CCR and CAD suggest that AC elements may be important for the transcription of these genes in lignifying xylem cells (Hatton et al., 1995; Hauffe et al., 1993; Lacombe et al., 2000; Lauvergeat et al., 2002). Disruptive point mutations, or deletion of some of these elements, result in the loss of promoter-driven transcription of reporter genes fused to the promoters for PAL, CCR or CAD genes (Hatton et al., 1995; Lacombe et al., 2000; Lauvergeat et al., 2002). These experiments suggest that these AC elements are important components in the transcriptional regulation of the lignin biosynthetic pathway.

Members of the R2R3-MYB family of transcription factors may be involved in the control of lignification via interaction with AC elements. R2R3-MYB proteins are defined by the presence of a conserved amino-terminal DNA-binding domain, known as the MYB domain, which has two imperfect helix-turn-helix repeats of approximately 50 amino acids (R2 and R3; Jin and Martin, 1999; Lipsick, 1996; Martin and Paz-Ares, 1997; Stracke et al., 2001). Several R2R3-MYB family members have been shown to bind to AC elements with a relatively high affinity and to regulate transcription from promoters containing these elements in an AC-element-dependent fashion (Grotewold et al., 1994; Jin et al., 2000; Moyano et al., 1996; Sablowski et al., 1994, 1995; Sainz et al., 1997; Sugimoto et al., 2000; Tamagnone et al., 1998a; Uimari and Strommer, 1997; Yang et al., 2001). Some R2R3-MYB proteins have been shown to regulate the biosynthesis of phenolic compounds, including lignin (Borevitz et al., 2000; Jin et al., 2000; Tamagnone et al., 1998a,b). To date, however, it remains to be determined if an MYB protein regulates lignin biosynthesis in a developmental context during the differentiation of lignified cells.

The work described herein examines the possibility that R2R3-MYB proteins play a role in regulating lignification during the differentiation of sclerified cells. As the biogenesis of wood involves a significant amount of lignification, a search was initiated for R2R3-MYB family members that were expressed during wood formation. This paper reports on the characterisation of one of the R2R3-MYB family members expressed in differentiating Pinus taeda L. (loblolly pine) wood, P. taeda MYB4 (PtMYB4). The expression pattern and functional characteristics of PtMYB4 support the hypothesis that this R2R3-MYB is involved in the regulation of lignification.


The PtMYB4 gene encodes a stereotypical R2R3-MYB protein

The PtMYB4 cDNA (GenBank Accession no. AY356371) was cloned from a cDNA library constructed from mRNA from differentiating P. taeda xylem. The library was probed using a pool of PCR products that had been generated using degenerate primers based on consensus R2R3-MYB amino acid sequences. PtMYB4 was one of the two cDNAs isolated in this screen and is the focus of this manuscript, while the other, PtMYB1, was examined in a separate study.

The PtMYB4 cDNA is predicted to encode a 35.1-kDa protein of 314 amino acids (Figure 1). The predicted protein is a stereotypical R2R3-MYB protein, containing two MYB repeats at the amino terminus and a putative transcriptional regulation domain at the carboxy terminus of the protein. The two MYB repeats, R2 and R3, are comprised of 54 and 51 amino acids, respectively, and have the predicted helix-loop-helix structures containing conserved tryptophan residues implicated in DNA binding (Figure 1). The predicted PtMYB4 protein is closely related to MYB proteins from other plant species, particularly the Arabidopsis R2R3-MYB proteins, AtMYB46 and AtMYB83 (Figure 1).

Figure 1.

Sequence analysis of the PtMYB4 cDNA.

(a) Nucleotide sequence of PtMYB4 with amino acid translation, GenBank accession number. The conserved stereotypical MYB DNA-binding domains, R2 and R3, are highlighted in light and dark grey, respectively. The putative C-terminal activation domain is located between position 122 and 314.

(b) Sequence similarity of the N-terminal region of predicted PtMYB4 including the conserved R2R3 DNA-binding domains with other MYB transcription factors. Amino acid residues highlighted in light grey are identical in more than 50% of the sequences; residues highlighted in dark grey indicate a conservative amino acid substitution. The palest grey amino acids represent the conserved tryptophan residues (W) predicted to be involved in DNA binding. The rectangles denote predicted helix-loop-helix structures.

(c) Neighbour-joining (NJ) tree based on amino acid sequence similarity of PtMYB4 in relation to the Arabidopsis thaliana MYB sequences with which it shares the greatest similarity, as well as AtGl1 and the human c-MYB (Hs c-MYB) sequence, which are included as outgroups. Amino acid sequence alignment and subsequent NJ tree were generated using clustalx (1.8). The tree was rooted on Hs c-MYB. GenBank Accession nos.: PtMYB4 (XXX), Hs c-MYB (M15024), AtGL1 (M79448), AtMYB20 (AF062869), AtMYB35 (AF062877), AtMYB40 (7573446), AtMYB46 (7630039), AtMYB50 (AC079733), AtMYB61 (AC003970), AtMYB86 (AF058914), AmMYB308 (JQ0960), AmMYB330 (JQ0957), HvMYB33 (X70881), LeMYB1 (X95297).

The PtMYB4 gene maps to a unique locus on the loblolly pine linkage map

In the course of sequencing the two cDNA clones corresponding to PtMYB4, potential allelic polymorphism was uncovered. One of the two clones had a sequence polymorphism in the 3′ non-coding region. This sequence polymorphism was because of allelic variation, which was used to place PtMYB4 on the randomly amplified polymorphic DNA (RAPD) linkage map for P. taeda clone 7-56 (see Experimental procedures). PtMYB4 was mapped to an interval that was less than 0.04 cM from the B12_1020 RAPD marker and 1.4 cM from the H12_700 RAPD marker, which corresponds to linkage group 14 on the P. taeda clone 7-56 RAPD map (

The PtMYB4 gene is preferentially transcribed in lignifying cells

The tissue-localised expression of PtMYB4 in the stems of pine seedlings was determined by in situ RT-PCR. PtMYB4 transcript accumulation was limited to specific cells of the stem, including the epidermal layer, early differentiating xylem cells, phloem fibres and pith cells (Figure 2a). The accumulation of PtMYB4 transcripts was limited to cells that were destined to lignify, as determined by co-localisation of transcript accumulation with cells adjacent to those staining with a common histochemical stain for lignin, phloroglucinol–HCl (Figure 2b). In contrast, the expression pattern of the positive control, the RUBISCO large subunit, was uniformly distributed throughout the stem (Figure 2d).

Figure 2.

PtMYB4 expression in pine stems coincides with lignifying cells as determined by in situ RT-PCR.

(a) Detection of PtMYB4 transcripts in pine stems as detected by in situ RT-PCR (blue stain). Cells that are staining positive for PtMYB4 transcripts are indicated by arrowheads in the regions indicated by the bars.

(b) As per above, but showing the coincidence of PtMYB4 transcripts as detected by in situ RT-PCR (blue) and lignifying cells as detected by phloroglucinol–HCl staining (red).

(c) A negative control, where the detection was carried out following RNAse treatment.

(d) A positive control, reporting on the presence of RBCL transcripts. Note that the staining is not restricted to lignifying cells and that it is spread throughout the stem section.

Bars = 100 µm.

Recombinant PtMYB4 protein binds to AC elements

Electrophoretic mobility shift assays (EMSAs) showed that purified recombinant PtMYB4 protein bound to three different AC elements (Figure 3). The three different AC elements, AC-I, AC-II and AC-III, were originally identified during the course of a functional dissection of the bean PAL2 promoter (Hatton et al., 1995), and differ by single nucleotide changes (Figure 3). The PtMYB4 protein did not bind to mutated versions of these elements (Figure 3).

Figure 3.

Analysis of PtMYB4 binding to AC elements.

(a) The sequence of the oligonucleotides for AC-I, AC-II and AC-III used as probes for the EMSAs. Each AC element, or the mutated AC versions thereof (mAC) is underlined.

(b) EMSAs showing binding of purified, recombinant PtMYB4 to AC elements or their mutated counterparts (mAC). The free probe without added proteins is designated as control (AC).

Recombinant PtMYB4 protein can activate transcription from AC elements in yeast

The PtMYB4 protein was expressed in yeast cells under the control of a galactose-inducible promoter to determine if it had the capability to bind and activate transcription from AC elements (Figure 4). The yeast strains harboured a reporter construct comprised of a sequence containing a triple repeat of a specific AC element, the minimal yeast CYC1 promoter and the β-galactosidase coding sequence (Figure 4). When expression of recombinant PtMYB4 protein was induced by galactose, the protein induced transcription from any of the three AC elements, as reported by increased β-galactosidase activity (Figure 4).

Figure 4.

PtMYB4-mediated activation of promoter activity in an AC-element-dependent fashion as determined by transcriptional activation assays in the yeast, Saccharomyces cerevisiae.

(a) The sequence of the oligonucleotides AC-I, AC-II and AC-III cloned into the reporter vector. Each AC element (underlined) is triplicated within the segment.

(b) Schematic representation of the effector (pYES2::PtMYB4) and reporter (pLacZi::AC) constructs used in this study. The effector construct drives the expression of PtMYB4 under the control of a galactose-inducible promoter. The reporter construct drives the expression of the lacZ gene from the minimal yeast CYC1 promoter fused to AC elements. Expression of lacZ, indicated by an increase in β-galactosidase activity, only occurs when the AC elements recruit a functional transcription factor.

(c) Quantitative analysis of β-galactosidase activity in yeast after treatment with the non-inducer, glucose (white bars), or the inducer, galactose (grey bars). The measurements in liquid assay were made from at least three independent replicates. Error bars represent SDs. Activation of promoter fused to the AC elements by PtMYB4 protein, following growth of the yeast in galactose, gave rise to β-galactosidase activity that was significantly different from the corresponding controls, as determined by anova (P < 0.005), including the reporter alone with no effector (AC) or reporter with empty effector (AC + pYES2) after growth on non-inducing glucose.

Overexpression of PtMYB4 in transgenic plants alters the accumulation of transcripts corresponding to genes encoding lignin biosynthetic enzymes

The cDNA encoding the PtMYB4 protein was overexpressed in transgenic tobacco plants under the control of the double CaMV 35S promoter. Transgenic plants harbouring the T-DNA from an empty pBINPLUS vector served as a control for the transformed lines overexpressing the PtMYB4 cDNA. The ability of the recombinant PtMYB4 protein to alter the transcription of genes related to lignin biosynthesis was assessed by Northern blot analysis of RNA from the transgenic plants. RNA was extracted from three sibling plants for three different independently transformed lines per construct.

On the basis of Northern blot assessment of transcript accumulation, three trends in gene expression were observed in the plants overexpressing the PtMYB4 cDNA relative to controls (Figure 5): first, the expression of the gene encoding PAL was decreased; second, the expression of the genes encoding C4H and 4CL was unchanged; and third, the expression of C3H, CCoAOMT, COMT, CCR and CAD was increased.

Figure 5.

Northern blot analysis of transcript accumulation in transgenic tobacco overexpressing the PtMYB4 transgene (PtMYB4-OE, right column) or harbouring an ‘empty’ T-DNA lacking the MYB transgene (pBIN controls, left column).

DNA probes hybridised to 10 µg total RNA isolated from individual T2 transgenic tobacco plants. Three independently derived control lines (C1, C2 and C3) and three independently derived PtMYB4-OE lines (OE1, OE2 and OE3) were used. RNA was hybridised with probes homologous to the tobacco genes for PAL (sequence similar to X78269; unpublished from C. Martin), C3H (trans-cinnamate 4-monooxygenase; unpublished from Dr C. Martin), 4-coumarate 3-hydroxylase (C3H; unpublished from Dr C. Chapple), 4-coumarate:CoA ligase (4CL; unpublished from Dr C. Martin), CCoAOMT (GenBank Accession no. U62734), COMT (GenBank Accession no. X74453), CCR (unpublished from Dr C. Halpin), CAD (GenBank Accession no. X62344).

Transgenic plants overexpressing PtMYB4 accumulate greater quantities of lignin and exhibit ectopic lignification

To determine the quantity of lignins present in the plants overexpressing the PtMYB4 cDNA, the amount of extractable lignins was assessed using the Klason method, which provides an accurate means by which to quantify lignins (Jung et al., 1999) and has been widely used as an indicator of lignin content (Chen et al., 2002; Franke et al., 2000; Marita et al., 2003; Meyermans et al., 2000; Pincon et al., 2001; Piquemal et al., 2002; Zhong et al., 2000). Seven independently transformed lines of tobacco overexpressing the PtMYB4 cDNA (PtMYB4-OE) were compared with four independently transformed control lines that had been transformed with an empty T-DNA vector. The quantities of both acid-insoluble and acid-soluble lignin contents were assessed. These analyses revealed that the tobacco lines overexpressing the PtMYB4 cDNA had more lignin than the corresponding controls (Figure 6).

Figure 6.

Accumulation of lignin in transgenic tobacco overexpressing the PtMYB4 transgene (PtMYB4-OE) or harbouring an ‘empty’ T-DNA lacking the MYB transgene (pBIN controls) as determined by Klason extraction.

Data are shown for acid-insoluble lignins (light grey portion of each bar) and acid-soluble lignins (dark grey), as a percentage of total DW. The total lignins are represented by the combined value of each bar. Each data point represents the mean obtained by the analysis of a minimum of five plants per line, measured in duplicate, and the standard errors are indicated by error bars.

Relative to the controls, plants overexpressing the PtMYB4 cDNA lignified ectopically as determined by UV autofluorescence and phloroglucinol–HCl staining of petioles and stems (Figure 7). While UV autofluorescence derived from the lignin polymer was limited to the xylem cells of the control plants, cells in the phloem and the pith autofluoresced in the stems of plants overexpressing the PtMYB4 cDNA. Similarly, where phloroglucinol–HCl staining was limited to specific xylem cell files of the control plants, staining included additional xylem cell files, the phloem and the pith in plants overexpressing PtMYB4 (Figure 7). These results were consistent across the independently transformed lines.

Figure 7.

Histochemical detection of lignin in transgenic tobacco plants harbouring an ‘empty’ T-DNA lacking the MYB transgene (a,c,e,g) or in transgenic tobacco plants overexpressing the PtMYB4 transgene (b,d,f,h).

Lignin was visible either using UV autofluorescence (a,b) or following staining with phloroglucinol–HCl (c–f). Accumulation was observed in the stems (a–d) or in petioles (e,f). Schematic diagrams show the regions of lignin accumulation in wild-type control petioles (g) or in the petioles of transgenic tobacco plants overexpressing the PtMYB4 transgene (h).


The appropriate control of metabolic flux is necessary for the survival of all organisms. The timing and localisation of resource allocation, particularly in autotrophic organisms such as plants, must be appropriately balanced between supply and demand. Resource allocation can be regulated at a number of levels, including the level of transcription of metabolic enzymes. To date, there are a limited number of examples of transcriptional regulators that function to regulate resource allocation in plants. In this paper, data are presented that suggest that PtMYB4 may function as a transcriptional regulator that controls resource allocation into the lignin biosynthetic pathway.

The PtMYB4 locus encodes a stereotypical R2R3-MYB transcription factor (Figure 1). On the basis of screening an unamplified cDNA library constructed from the mRNA from differentiating pine xylem, the PtMYB4 transcript was of low abundance, representing only 1 out of every 100 000 sequences that were screened. Consistent with this finding, we were only able to detect PtMYB4 expression in a limited number of cell types, using in situ RT-PCR (Figure 2). PtMYB4 transcripts accumulated in cells that appeared to be lignifying on the basis of the co-occurrence of PtMYB4 transcripts and the presence of lignin as detected by histochemical staining (Figure 2). These findings support a role of PtMYB4 in lignifying cell types.

Normally, lignification occurs in only a subset of plant cells. This includes xylem cells, interfascicular fibres and phloem fibres (Esau, 1965a,b). In some plants, the pith and some peridermal cells may also undergo lignification (Esau, 1965a,b). The overexpression of the PtMYB4 cDNA in transgenic tobacco resulted in a net increase in the accumulation of lignin relative to control plants (Figure 6). In the plants overexpressing PtMYB4, lignification extended beyond the normal number of xylem cell files to include additional xylem cell files, as well as phloem cells and pith cells (Figure 7). Thus, PtMYB4 overexpression increases the number of cell types that are competent to lignify. However, PtMYB4 overexpression was only sufficient to induce ectopic lignification in a limited number of cell types, despite the fact that it was overexpressed in all cells. It may be that PtMYB4 activity is contingent on the presence of a co-factor, or on post-translational modification, which is not present in some cells; therefore, these cells do not lignify even when PtMYB4 is overexpressed. PtMYB4 provides a useful starting point to identify factors that control the competency to lignify.

The induction of lignification by PtMYB4 is likely to a result of the activation of lignin biosynthetic genes containing AC elements, as recombinant PtMYB4 protein was able to bind to AC elements in vitro (Figure 3) and activate gene expression from AC elements in yeast (Figure 4). Northern blot analyses showed that PtMYB4 overexpression could increase the transcription of some of the genes encoding lignin biosynthetic enzymes (Figure 5). Notably, the genes that showed increased transcript abundance encoded enzymes that are believed to be involved in the ‘monolignol-specific’ portion of the lignin biosynthetic pathway: CCoAOMT, CCR, COMT and CAD (Humphreys and Chapple, 2002). The transcription of the gene encoding C3H was also enhanced within this set of genes, even though it is currently associated with the ‘early’ portion of the lignin biosynthetic grid. Previous studies have suggested that the level of C3H transcription might be important in directing the flux of metabolites towards lignin production, and that the transcription of CCoAOMT, CCR and CAD is modulated according to the metabolic demand (Anterola et al., 2002). The net result of the increased transcription of these genes may result in an increase in the flux through the monolignol portion of the lignin biosynthetic pathway, thus increasing the sink strength of lignin biosynthesis.

In contrast to the transcription of the monolignol-specific genes, there was a decrease in the quantity of transcripts corresponding to PAL in transgenic plants overexpressing PtMYB4 (Figure 5). In this respect, PtMYB4 may be acting in an analogous fashion to AtMYB4, which represses the transcription of genes encoding enzymes involved in phenylpropanoid metabolism (Jin et al., 2000). While it remains to be determined if PtMYB4 truly functions as a transcriptional repressor, the possibility that it may function as both an activator and a repressor is intriguing as such dual functionality has been reported for animal MYB proteins (Masselink et al., 2001).

The fact that PtMYB4 overexpression decreases the accumulation of PAL transcripts, and given that PAL is the entry-point enzyme in the general phenylpropanoid pathway, the net result of PAL downregulation might be to decrease overall flux into the biosynthesis of phenylpropanoid-derived compounds, including lignin. Contrary to this, lignin accumulation actually increased in these plants. These results are consistent with previous observations of flux control in phenolic metabolism. Transgenic tobacco plants that had varying levels of co-suppression of PAL transcription uncovered a hierarchy of flux into phenylpropanoid-derived metabolites (Bate et al., 1994). While relatively modest changes in PAL activity had a profound effect on the accumulation of soluble phenylpropanoids, such as chlorogenic acid, lignin biosynthesis was relatively unaffected until PAL levels were very low (Bate et al., 1994). These findings suggest that the sink strength for the lignin biosynthetic pathway is greater than that for other phenylpropanoids, so that it is preferentially synthesised when phenylpropane precursors are limiting. Thus, in transgenic plants overexpressing PtMYB4, the PtMYB4 protein may function as a switch to direct the flux of phenylpropanoids into the biosynthesis of lignins. By limiting flux into general phenylpropanoid metabolism and by increasing the sink strength of the monolignol-specific portion of the pathway, PtMYB4 might divert metabolism away from the synthesis of other phenylpropanoids and into the lignin biosynthesis specifically.

While PtMYB4 overexpression is sufficient to induce lignification, it remains to be determined if it is necessary for lignification. Assessment of the necessity of PtMYB4 in the control of lignification in pine is problematic. Perhaps the greatest problem is redundancy in the R2R3-MYB family. In Arabidopsis, there are at least two R2R3-MYB family members that could be the functional orthologues of PtMYB4. Given the extent of genetic redundancy uncovered in the pine genome (Kinlaw and Neale, 1997), we would predict that pine might possess even more putative paralogues than the two found in the Arabidopsis genome. Such functional redundancy confounds loss-of-function experiments aimed at determining necessity. In gene families that have functional redundancy, related family members often must be suppressed before a phenotype is observed (Fitter et al., 2002; Liljegren et al., 2000). Aside from this, reproducible genetic engineering has not yet become a reality for pine; therefore, directed modification of PtMYB4 in a homologous background is likely to prove difficult. In order to address these problems in future experiments aimed at testing necessity of the PtMYB4-like genes to regulate lignification, it may be necessary to suppress several MYB family members simultaneously, focusing on the Arabidopsis PtMYB4 orthologues.

PtMYB4 is one of the few examples of a transcription factor that is sufficient to increase lignification. Two functionally redundant Arabidopsis MADS transcription factors, SHATTERPROOF (SHP)1 and SHP2, were found to be necessary for lignification of the valve margin of the mature silique, and were also shown to be sufficient to increase lignification when overexpressed in transgenic plants (Liljegren et al., 2000). The Arabidopsis PAP1 gene, which encodes an R2R3-MYB protein, increases all facets of phenylpropanoid metabolism, including lignification, when overexpressed (Borevitz et al., 2000). PtMYB4 appears to be unique relative to PAP1 in that it induces lignin biosynthesis without the visible increase in anthocyanin biosynthesis that was observed when PAP1 was overexpressed. If anything, PtMYB4 is more likely to divert carbon from the biosynthesis of other phenylpropanoids, such as anthocyanins, towards lignin biosynthesis. Given the relationship between lignin content and energy production (Demirbas, 2000, 2001), PtMYB4 may be useful in strategies aimed at specifically increasing lignin content in fibre crops grown for the conversion of biomass into energy.

Experimental procedures

Plant growth conditions

The mature P. taeda tree was cultivated in Hell Pocosin, Purgatory, North Carolina, USA (Latitude, 34°58′ 20.586N; Longitude, 77°35′ 28.889W). The tree was a clone of P. taeda tree 7-56, which had been generated by grafting a scion from 7-56 onto a rootstock. The resulting clone was grown for approximately 35 years prior to harvesting.

Pinus taeda seeds, derived from seed parent 7-56 (kindly provided by Dr Barry Goldfarb, North Carolina State University), were stratified at 4°C, germinated in vermiculite and grown in a greenhouse with supplemental fluorescent lighting to provide a 16-h photoperiod.

Transgenic tobacco plants (Nicotiana tabacum cv. Samsum) were cultivated in potting mix in a greenhouse with supplemental fluorescent lighting to provide a 16-h photoperiod.

cDNA cloning

Xylem was harvested from a mature P. taeda tree and ground to a fine powder under liquid nitrogen using a coffee grinder. RNA was isolated from differentiating xylem according to an published protocol (Chang et al., 1993). Poly(A)+ RNA was purified from the isolated RNA using PolyAttract as per manufacturer's instructions (Promega Corp., Madison, WI, USA). A LambdaZap (Stratagene, La Jolla, CA, USA) cDNA library was constructed and screened according to standard protocols.

To generate the probe to screen the xylem cDNA library for clones encoding MYB proteins, a degenerate PCR approach was taken. Degenerate primers corresponding to conserved regions of the MYB coding sequence (G[A/P]WTNNED and PGRTDN) were used to amplify PCR products using first-strand cDNA from differentiating xylem as a template. The resultant products were cloned and sequenced to verify that they encoded MYB proteins. These clones were pooled and used to probe 200 000 clones from an unamplified pine xylem cDNA library. Fourteen positive clones were identified, purified and sequenced according to standard protocols. Twelve of these clones had the same sequence, coded for by PtMYB1 (Newman et al., submitted), and the other two corresponded to a completely different sequence, PtMYB4, which is the subject of this paper.

Mapping the PtMYB4 locus in Pinus taeda

Placement of the PtMYB4 locus on the RAPD map for P. taeda clone 7-56 utilised a combination of bulk segregant analysis and selective genotyping with pine megagametophyte DNA. This approach has been described previously by Wilcox et al. (1996). One of the two PtMYB4 cDNA clones had a sequence polymorphism in the 3′ non-coding region, which generated a BstE1 restriction site that was not found in the other clone. In order to determine if this polymorphism was because of allelic variation or an error in cDNA synthesis, megagametophytes from open-pollinated 7-56 seeds were genotyped using a cleavable amplifiable polymorphic sequence (CAPS) approach. Primers were designed that flanked the BstE1 restriction site, and these were used to PCR-amplify the alleles from DNA template derived from individual megagametophytes. The amplified DNA was used in a restriction digest with BstE1, to determine the identity of the allele. The resulting CAPS marker for the PtMYB4 locus was used to identify megagametophytes that possessed one allele or the other. A total of 96 megagametophytes were genotyped in this manner, and they showed perfect Mendelian (1 : 1) segregation of the CAPS alleles. ‘Bulks’ of each allele were constructed by pooling DNA from 12 individuals that had one allele or the other. Three bulks were constructed in this fashion for each allele and they were used as template in a bulk segregant analysis approach (Wilcox et al., 1996) to identify RAPD markers that were linked to the PtMYB4 locus. Two hundred different RAPD primers were screened, and two generated PCR products that co-segregated with the PtMYB4 alleles, B12 (PCR product: 1020 bp) and H12 (PCR product: 700 bp). Selective genotyping of 96 individual randomly chosen megagametophytes with the PtMYB4 CAPS marker and the two RAPD markers, B12_1020 and H12_700, allowed a linkage relationship to be established between the three loci, which was determined using MAPMAKER MACINTOSH.

In situ RT-PCR

Hundred-micrometre serial sections were prepared from 21-day-old P. taeda seedlings. In situ RT-PCR was performed on the sections according to a published protocol (Koltai and Bird, 2000). The method used M-MLV RNAse H minus Point Mutant (Promega) as the reverse transcriptase and Tth DNA polymerase (Roche Ltd., Hertfordshire, UK) as the thermal stable polymerase, and incorporated digoxigenin-11-dUTP (Roche) into the PCR product. The gene-specific primer pairs were 413EX51 (5′-GCG GAT CCA AAT GAG CTG CAC AAC AGG AGG-3′) and 413EX31 (5′-ACG CGT CGA CAT ACT TCC CAC CTG ATC ATG-3′) for amplification of PtMYB4, or, for amplification of PtRBCL, PtRBCL-5′ (5′-ATT GGG AGT TCC TAT CGT TAT GC-3′) paired with PtRBCL-3′ (5′-GCT TCA CGG ATC ACT TCA TTA CC-3′). The PtRBCL primers were based on sequence for P. taeda ribulose-1,5-bisphosphate carboxylase/oxygenase, large subunit (rbcL) gene (GenBank Accession no. AF119177). RT-PCR on purified P. taeda total RNA resulted in single products being generated for both primer sets. Following RT-PCR, the sections were developed as per Koltai and Bird (2000). Sections were mounted on 90% glycerol, viewed under a microscope and digitally photographed. Sections were counter-stained with 1% phloroglucinol in 6N HCl to identify cells with lignified cell walls. As a negative control, a subset of the sections were washed with sterile de-ionised water, and was then treated with 20 µl of 2 mg ml−1 RNAse A (Sigma) in sterile de-ionised water at 37°C for 30 min. The RNAse-treated sections were washed with sterile de-ionised water before continuing through the in situ RT-PCR protocol.

Production of recombinant protein in Escherichia coli

The gene-specific primers 413EX51 and 413EX31 were used in a PCR reaction to amplify the full-length PtMYB4 coding sequence. The PCR product was subcloned into the BamHI/SalI sites of the expression vector pET30c (Novagen Inc., Madison, WI, USA). The E. coli strain BL21 (DE3) was used for the expression of the recombinant proteins. Bacterial growth and protein induction were performed as described by the manufacturer (Novagen). Recombinant protein was purified by affinity chromatography using a Ni–NTA column (Qiagen Ltd., Sussex, UK) under denaturing conditions. The purification was verified by SDS–PAGE analysis (Laemmli, 1970) and visualisation by Coomassie blue staining. The protein was quantified, and protein concentrations were then adjusted by dilution so that 0.6 µg of the purified protein was used in the EMSAs described below.

Electrophoretic mobility shift assays with pine MYB proteins

Purified recombinant PtMYB4 protein was diluted 10-fold and allowed to re-nature prior to the binding experiment. Each binding assay contained 0.6 µg of the purified MYB protein, and was conducted according to established EMSA protocols (Hatton et al., 1995; Sablowski et al., 1994). EMSA reactions were pre-incubated for 25 min at 4°C without the DNA radioactive probe to allow non-specific binding, and were then incubated with the radiolabelled probe (30 000 c.p.m. per reaction) for 30 min. The reactions were then loaded into 5% (w/v) polyacrylamide, 2% glycerol native gel and electrophoresed for 2–3 h at 180 V at 4°C. Following electrophoresis, the gel was dried under vacuum on Whatman 3 MM paper, and exposed overnight to Kodak X-ray film with two intensifying screens.

Transcriptional activation assays in yeast

Plasmids and yeast strains used in the transcriptional activation assays were prepared using standard techniques, and details are available upon request. The pYES2 vector (Invitrogen Ltd., Paisley, UK) was used to construct the effector plasmid, which drove the PtMYB4 cDNA under the control of the galactose-inducible promoter, GAL1 (Figure 4). Reporter constructs contained triple repeats of a given AC element upstream of the yeast minimal CYC1 promoter and the coding sequences for β-galactosidase (Figure 4) based on the pLacZi vector (Clontech, Palo Alto, CA, USA). The yeast reporter strains were constructed by integrating the reporter construct into the yeast genome using established protocols (Fink et al., 1990).

Yeast transformation of reporter strains with pYES2::PtMYB4 was based on a published protocol (Agatep et al., 1998). The β-galactosidase liquid assay was carried out using O-nitrophenyl-β-d-galactopyranoside (ONPG) as the substrate (Fink et al., 1990). The β-galactosidase activities obtained in the assays were given as a β-galactosidase unit (1 unit of β-galactosidase is defined as the amount that hydrolyses 1 mmol of ONPG to o-nitrophenol and d-galactose per minute per cell).

Overexpression of PtMYB4 in transgenic tobacco

Plasmids for the overexpression of PtMYB4 in transgenic plants were prepared using standard techniques, and details are available upon request. To produce a construct for constitutive high-level expression of PtMYB4, the full-length cDNA was cloned into pJIT60 (Guerineau and Mullineaux, 1993) between the double CaMV 35S enhancer and the CaMV terminator to create pJITMYB4. The entire expression cassette was cloned into the binary vector pBINPLUS (van Engelen et al., 1995) for Agrobacterium-mediated transformation of plants.

Tobacco (N. tabacum cv. Samsun NN) leaf discs were transformed using Agrobacterium tumefaciens strain GV3101 (Fraley et al., 1983). For selection of transgenic individuals, seeds were surface-sterilised and distributed on plates containing MS medium (Murashige and Skoog, 1962) supplemented with 200 mg l−1 kanamycin sulphate and 200 mg l−1 timentin.

Tobacco RNA isolation and Northern analysis

RNA was extracted from tobacco leaf material (Goldsbrough and Cullis, 1981), electrophoretically separated, transferred to Biodyne Membrane B (Pall Corp., Hampshire, UK) and probed with radiolabelled probes prepared from cloned cDNAs according to standard protocols. The cDNAs that were used as probes corresponded to PAL (EC; similar to GenBank Accession no. X78269, kindly provided by Dr C. Martin, John Innes Centre); cinnamic acid 4-hydroxylase (C4H; trans-cinnamate 4-monooxygenase), EC; unpublished sequence, kindly provided by Dr C. Martin); COMT (EC; GenBank Accession no. X74453); 4CL (EC; unpublished sequence, kindly provided by Dr C. Martin); CCR (EC; unpublished sequence, kindly provided by Dr C. Halpin, University of Dundee); cinnamyl-alcohol dehydrogenase (CAD; EC; GenBank Accession no. X62344, kindly provided by Dr C. Halpin); coumarate 3-hydroxylase (C3H; EC; unpublished sequence, kindly provided by Prof. C.C.S. Chapple, Purdue University); and CCoAOMT (EC; GenBank Accession no. U62734).


Tobacco petioles and stems were sectioned using a vibrating microtome to generate 100-µm serial sections. For histochemical analysis of lignification, sections were stained with 1% (w/v) phloroglucinol in 6 m HCl, mounted in 3 m HCl/50% glycerol, viewed under the microscope and digitally photographed.

Cell wall analysis

The chemical composition of the different tobacco lines was determined using a modified Klason analysis. In brief, freeze-dried tobacco was ground to pass a 40-mesh screen using a Wiley mill. The ground tobacco (1 g) was then soxhlet-extracted with 100 ml acetone for 8 h to remove extractable components and to minimise the formation of ‘pseudolignin’ during Klason analysis. The total weight of extractable components was determined gravimetrically by rotary evaporation. The extracted lignocellulosic material was air-dried to remove solvent and was then analysed in triplicate for sugar and lignin composition as follows.

A 0.2-g sample of extracted tobacco was transferred to a 15-ml reaction vial in an ice bath. A 3-ml aliquot of 72% (w/w) H2SO4 was added to the sample and was thoroughly mixed for 1 min. The test tube was immediately transferred to a water bath maintained at 20°C and was subsequently mixed for 1 min after every 10 min. After 2 h of hydrolysis, the contents of each test tube were transferred to a 125-ml serum bottle, using 112 ml nanopure H2O, to rinse all residue and acid from the reaction vial. The serum bottles (containing 115 ml H2SO4 at 4% (w/w) plus tobacco) were sealed with septa and autoclaved at 121°C for 60 min. Samples were allowed to cool, and the hydrolysates were vacuum-filtered through pre-weighed medium coarseness sintered-glass crucibles, washed with 200 ml warm (approximately 50°C) nanopure H2O to remove residual acid and sugars and dried overnight at 105°C. The dry crucibles were weighed to determine Klason (acid-insoluble lignin) lignin gravimetrically. The filtrate was also analysed for acid-soluble lignin by absorbance at 205 nm according to TAPPI Useful Method UM250.


We are very grateful to members of the Campbell lab for their kind assistance in various aspects of this work. S. Mansfield is a Canada Research Chair in Wood and Fibre Quality. This work was generously supported by funding from the US Department of Energy to R.W.W., R.R.S. and M.M.C., by funding from the Natural Science and Engineering Research Council of Canada (NSERC) and the Canadian Fund for Innovation (CFI) to S. Mansfield and by funds from the UK Biotechnology and Biological Sciences Research Council (BBSRC) to M.W.B and M.M.C.