Plant material and RNA isolation
Several tissues were isolated from two 33-year-old P. glauca trees felled in July 2003, from a progeny trial established near Quebec City (Canada). All tissues were frozen in liquid nitrogen immediately upon removal from the tree and stored at -80°C until further use. We collected newly formed needles from the upper crown. Differentiating secondary xylem and phloem, as well as bark tissues were collected from three 30–40 cm bolts taken from the lower third of the main stem. These vascular tissues were scraped with a scalpel immediately after peeling the bark. Tissues scrapped from the exposed inner side of the bark and from surface of the exposed wood were labelled as differentiating secondary phloem and xylem, respectively. Similarly, differentiating xylem and bark (including phloem) were collected from large roots located in a one-meter radius from the base of the stem. Samples from each tree and each tissue were kept separate for RNA extraction and gene expression studies.
A gravitropic treatment to induce compression wood formation was performed on 3-year-old spruce seedling stock. The seedlings were transferred to 3 L pots one month before the experiment, grown in a greenhouse with 16 hours light per day, and fertilised weekly with 20 g/L N-P-K. A randomised design of 24 young trees was established in which 12 trees were maintained at 45° angle by leaning the pots and tying the plants to stakes (also at 45°); 12 seedlings were grown in the normal vertical position. Destructive tissue samplings were carried out 4, 28 and 76 hours after the beginning of the treatment. The average diameter of the plants near the base was 7.2 +/- 0.61 mm, their average height was 60.63 +/- 7.25 cm, and the terminal leader was 19.83 +/- 2.91 cm. For each time point, four vertical and four leaning trees were harvested mid-morning in a randomised order, and three randomly selected trees were used for gene expression analyses. Secondary xylem was collected as described above from the two sides of the main stem of the seedlings; the lower and upper sides representing compression wood and opposite side wood, respectively, for the leaning trees, or left and right side for vertical tree. The whole terminal leader was also collected from each plant. Total RNAs were isolated from tissues described above and ground in liquid nitrogen with a pestle and mortar, except for the gravitropic treatment where a Mixer Mill MM300 engine (Retsch) was used to grind in Microtubes (Eppendorf). The RNAs were extracted from each tissue sample and each tree or seedling separately, following the procedure of Chang et al. , in Oakridge tubes or in Microtubes. RNA concentration and quality were determined with a bioAnalyser (model 2100, Agilent Technologies; RNA 6000 Nano Assay kit).
DNA cloning and accession numbers
Previous reports described two complete coding sequences of R2R3-MYB genes from pine (PtMYB1 and PtMYB4) [10, 11] and one from spruce  as well as several partial sequences  expressed in xylem tissues of pine. We isolated partial spruce cDNA clones representing putative orthologues of the pine MYB genes by PCR amplification, and identified several partial and putative full-length sequences among the spruce EST sequences data of the ARBOREA project derived from 17 different cDNA libraries For each of the partial spruce and pine gene sequences, we obtained complete coding sequence and UTRs by using 3' RACE, 5' RACE or both cloning methods on spruce or pine mRNA from needles or xylem (SMART RACE cDNA Amplification Kit, Invitrogen, Carlsbad, CA). DNA was cloned in pCR2.1 with the TA cloning Kit (Invitrogen, Carlsbad, CA) and sequenced. The sequence analyses presented hereafter are based upon cDNA clones containing the complete coding sequences of the MYB, which were isolated as a single fragment from reverse transcribed RNA, with gene specific primer pairs for each of the 13 sequences from spruce and five sequences from pine. The numbering of pine MYB genes (PtMYB1-8 and PtMYB14; no PtMYB5 and 6 reported) is in accordance with Patzlaff et al., [10, 11]; the numbering of spruce genes was established on the basis of putative orthology with the pine sequences.
In addition, we isolated the 13 corresponding sequences from spruce genomic DNA (gDNA). Genomic DNA was extracted from needles of white spruce using the Genomic-Tip Kit (Qiagen, Mississauga, Ontario). The entire coding region with introns was isolated by PCR amplification with gene specific primer pairs spanning each gene's coding region (Additional file 1) and cloned in pCR2.1 with the TA cloning Kit (Invitrogen, Carlsbad, CA). The gDNA was from Picea glauca genotype Pg653 and so did most of the cDNA clones (although a few came from wild Picea glauca genotypes). Each clone was sequenced at least through the MYB DBD in order to determine the number of introns present in this region. Some nucleotide differences were observed between cDNA and gDNA sequences due to the genotypic variation, but no non-sense mutation were detected. The genomic sequences of PgMYB 3, 6, 7 and 11 showed 1 to 3 non synonymous substitutions giving no less than 99.2% amino acids identity; however, we do not find nucleotide mismatches in spruce MYB 2, 4, 8, 9 and 12.
The 13 MYB genes from spruce and five MYB genes from pine have the following accession numbers: PgMYB1 [GenBank: DQ399073], PgMYB2 [GenBank: DQ399072], PgMYB3 [GenBank: DQ399071], PgMYB4 [GenBank: DQ399070], PgMYB5 [GenBank: DQ399069], PgMYB6 [GenBank: DQ399068], PgMYB7 [GenBank: DQ399067], PgMYB8 [GenBank: DQ399066], PgMYB9 [GenBank: DQ399065], PgMYB10 [GenBank: DQ399064], PgMYB11 [GenBank: DQ399063], PgMYB12 [GenBank: DQ399062], PgMYB13 [GenBank: DQ399061] and PtMYB2 [GenBank: DQ399060], PtMYB3 [GenBank: DQ399059], PtMYB7 [GenBank: DQ399058], PtMYB8 [GenBank: DQ399057], PtMYB14 [GenBank: DQ399056].
The nucleotides sequences of candidate genes involved in wood formation come from the spruce EST assembly directory number 8 (dir8) of the ARBOREA project Their percentage amino acid sequence similarity with other species is given in brackets. They are:
phenylalanine ammonia lyase (PAL) [dir8: contig10199] partial coding sequence (cds), 85% to Pinus taeda [GenBank: U39792];
4-coumarate: CoA ligase (4CL) [dir8: contig10433] partial cds, 86% to Pinus taeda [GenBank: U39405];
caffeoyl-CoA 3-O-methyltransferase (CCOaOMT) [dir8: contig5884] complete cds, 92 % to Pinus taeda [GenBank: AF036095];
arabinogalactan protein (AGP) [dir8: contig10745] complete cds, 75 % to Pinus taeda [GenBank: U09556];
cinnamyl alcohol dehydrogenase (CAD) [dir8: contig9065] complete cds, 95 % to Pinus radiata [GenBank: AF060491],
and the housekeeping gene elongation factor alpha (EF1-α) [dir8: contig10829] complete cds, 99% to Picea abies [GenBank: AJ132534].
Sequence analyses and phylogenetic studies
Nucleotide and amino acid sequence alignments were obtained with Clustal W  using BioEdit software version 6.0.7 and BoxShade 3.21 to highlight sequences. Delineation of introns was achieved by aligning the cDNA and genomic nucleotide sequences of the PgMYB from the start codon to stop codon on the basis that introns begin with GT and end with AG dinucleotides. Phylogenetic studies were performed with the 13 predicted MYB protein sequences from white spruce, one sequence from black spruce (PmMBF1, ), and seven loblolly pine sequences including previously reported PtMYB1 and 4 [10, 11]. We also included 11 diverse Arabidopsis MYB sequences and two R1R2R3-MYB genes from human and mouse as landmarks to classify the MYBs according to previous reports [18, 19, 30]. We constructed a neighbour-joining tree based on a Clustal W amino acid alignment generated with the Mega 2.0 method  and using 1000 bootstraps to estimate the node strength (parameters are Poisson correction and pair-wise deletion as described in ).
The accession numbers of the Arabidopsis genes analysed are: AtMYB13 [GenBank: At1g06180], PtMYB1 [GenBank: AY356372], AtMYB4 [GenBank: At4g38620], AtMYB103 [GenBank: At1g63910], AtMYB46 [GenBank: At5g12870], PtMYB4 [GenBank: AY356371], AtMYB61 [GenBank: At1g09540], AtMYB52 [GenBank: At1g17950], AtMYB44 [GenBank: At5g67300], AtMYB20 [GenBank: At1g66230], AtMYB101 [GenBank: At2g32460], AtMYB106 [GenBank: At3g01140], AtMYB33 [GenBank: At5g06100], PmMBF1 [GenBank: U39448]. The accession numbers of the conifer MYB genes are as above.
MEME analysis software  was used to identify amino acid regions conserved between several members of a subgroup of sequences (according to Kranz et al. ) containing one or more spruce MYBs. The parameters setting was the number of motifs to find: 5; minimum width of motif: 5 and maximum: 15. We used complete protein sequences of MYBs from 12 conifer species and from 14 angiosperm species (Additional file 2). We also included 10 partial conifer sequences that encompassed the C-terminal region  and were closely related to PgMYB6, 7 and 9 (Additional file 3).
Analysis of transcript accumulation by Q-RTPCR
To analyse the transcript abundance of PgMYBs, first-strand cDNA was synthesised starting from one microgram of RNA treated with amplification grade DNAse I (Invitrogen) and purified on an RNeasy column (Qiagen), using oligo(dT) primers and SuperScript II RT (Invitrogen) according to the manufacturer's instructions. The resulting cDNA was diluted 1:5 with sterile RNAse-free water and stored at -20°C; the same cDNA was used for all Q-RTPCR (quantitative reverse transcription PCR) assays for any given tissue sample. Steady-state mRNA levels were determined on cDNA by quantitative RT-PCR using Opticon Monitor 2 fluorescence detection system (MJ research) and DyNAmo SYBR Green QPCR kit (Finnzymes Oy, Espoo, Finland). Gene primer pairs were designed using the Primer 3 software  to anneal near the 3' end of each transcript (usually in 3' UTR) to ensure primer specificity. The forward and reverse primers were as follows (amplicon length indicated in brackets):
PgMYB1: 5'-gattgtacattaacccagtaa-3' and 5'-taaaccatgtggtatctgtta-3' (148 bp);
PgMYB2: 5'-tgggtattctaggtatttcc-3' and 5'-attaggtaagtatgcaggg-3' (99 bp);
PgMYB3: 5'-agatcacggacccagatcaac-3' and 5'-gagcgaacgacctccttcag-3' (147 bp);
PgMYB4: 5'-gcagtttgagtttgagtgtg-3' and 5'-ctggagcatagatttgatga-3' (162 bp);
PgMYB5: 5'-aattctggcagcgaactg-3' and 5'-aatgcttcgtggtggaatc-3' (173 bp);
PgMYB6: 5'-tttccttccttcatttcaac-3' and 5'-taaatttgggtttctgttgc-3' (108 bp);
PgMYB7: 5'-tcgagttgcacatcaggag-3' and 5'-gagtgtggatggcaaacag-3' (156 bp);
PgMYB8: 5'-ggtggactcagttgtaataa-3' and 5'-gtatctcacctatttacagatca-3' (101 bp);
PgMYB9: 5'-gaaattcgagaaacatggtg-3' and 5'-aaacgacagaaatcgagaac-3' (149 bp);
PgMYB10: 5'-gctgtattttaacatttcatgg-3' and 5'-acaacaatctttctttttctcc-3' (133 bp);
PgMYB11: 5'-cccagcttatgactggaag-3' and 5'-tacagaacaaccatgcagac-3' (157 bp);
PgMYB12: 5'-caggtgacttaactctattccag-3' and 5'-tcacatagaacaggcatgg-3' (143 bp);
PgMYB13: 5'-aaattacagctagagtgagagg-3' and 5'-aacttgaaccgtacacgac-3' (86 bp) and
EF1-α (elongation factor alpha): 5'-aactggagaaggaacccaag-3' and 5'-aacgacccaatggaggatac-3', (114 bp). Forward and reverse primer pairs used for the Q-RTPCR of the wood formation-related genes are:
PAL: 5'-tggatttgcatcctactg-3' and 5'-tccatcttcaactataggac-3' (103 bp);
4CL: 5'-cattcctcaaaagcatgaagag-3' and 5'-atcgcatccacaaagtacac-3' (150 bp);
CCOaOMT: 5'-attgagatcagccaaatcc-3' and 5'-gcgctctccctataatcag-3' (124 bp);
AGP: 5'-gcgtccattgttttaatgtag-3' and 5'-tgtatttatccctctgtctgc-3' (181 bp) and
CAD: 5'-ctggactacatcaatactgc-3' and 5'-gatttactcattctgcacg-3' (141 bp).
The PCR reaction mixture (20 μL) consisted of 10 μL DyNAmo SYBR Green qPCR mix, 1 μL primers (0.25 μM forward and 0.25 μM reverse), 1 μL of cDNA and 8 μL of RNase-free water. The cycling conditions were 95°C for 2 min, then 35 cycles of 95°C for 10 sec, 55°C for 10 sec, 72°C for 8 sec, followed by a plate reading. The melting curve readings were carried out every 0.2°C from 65 to 95°C, holding 1 sec at each temperature. Standard curves were established for each spruce MYB and cell-wall-related cDNAs and were used to determine RNA abundance in each sample. The standards consisted of serial dilutions of PCR amplicons prepared from each cDNA, cloned in pCR2.1, with M13 reverse and forward primers. Amplicon standards were gel purified (Qiagen), and product length and concentration were verified using a bioAnalyser (model 2100 Agilent Technologies, DNA 1000 LabChip kit). The standard curves were determined from duplicate reactions from the dilutions series of each amplicon. Raw data were converted with the following parameters: no blank subtraction, subtract baseline and average over cycle range according each case. We calculated the number of transcript molecules per ng of total RNA [(DNA quantity quantified in g * DNA base pair mass per gram of DNA)/M13 Reverse-Forward amplified fragment length in bp]. For within tissue comparisons that were carried on differentiation xylem (including compression wood induction) and apical shoots young trees, the RNA abundance was normalised using the abundance of RNA of the elongation factor alpha (EF1-α), as an endogenous control (calculated as the following ratio: target gene (ng)/EF1-α (ng)).