Identification and domain analysis of ClMYBs
A total of 116 candidate sequences with MYB domains were selected from the full-length transcriptomic library, including three 4R-MYB, seven 3R-MYB, 70 R2R3-MYB, and 36 1R-MYB proteins. After the removal of redundant sequences, we identified 52 R2R3-MYB with complete ORF, which were sequentially named as ClMYB1-52 (Supplementary Table S2).
The number of amino acid residues in the 52 ClMYBs ranged from 215 to 596. The corresponding molecular weight and isoelectric points ranged between 24.3–64.1 kDa and 4.7–9.9 (Supplementary Table S2), respectively. Multiple sequence alignments showed that all the N-terminal regions of 52 ClMYBs contained conserved R2 and R3 domains. The R2 domains of ClMYBs contained three highly conserved Trp (W) residues. However, within the R3 domain, the first W residue is substituted by Leu (L), Ile (I) or Phe (F). In addition, some conserved amino acids were present around these W residues, including Gly (G2), Glu (E8), Asp (D9), Leu (L12), Gly (G20), Arg (R36), Gly (G38), Lys (K39), Cys (C41), Arg (R42), Arg (R44), Asn (N47) and Leu (L49) in the R2 domain; Glu (E8), His (H16), Gly (G20), Asn (N21), Gly (G32), Arg (R33), Thr (T34), Asp (D35), Asn (N36), Lys (K39) and Asn (N40) in the R3 domain (Fig. 1). These characteristics ensure that MYB domains fold into a stable helix-helix-turn-helix structure.
A comparative phylogenetic study of R2R3-MYBs in C. lanceolata and other plant species
To analyse the evolutionary relationship of ClMYBs and known R2R3-MYBs from A. thaliana, E. grandis, P. trichocarpa, O. sativa, P. taeda, and P. glauca, we constructed the neighbour-joining phylogenetic tree (Fig. 2; Supplementary Figure S2). According to the topological structure, all the proteins could be divided into 47 subgroups, 21 of which harboured R2R3-MYBs from C. lanceolata. Over 50% of the subgroups were supported by bootstrap value of > 70%. Lower bootstrap values were found in the subgroups with mixed sequences of gymnosperm and angiosperm origin. This can be justified because many gene duplication events in the R2R3-MYBs occurred after divergence from a common ancestor [34] and rapid diversification of the family after the division [35]. Nomenclatures of the subgroups were previously reported for A. thaliana [8, 36], later these were modified by Soler et al., (2015) [37] following a genome-wide evolutionary analysis of the R2R3-MYBs from the woody species, E. grandis. The sequences of A. thaliana and E. grandis were correctly placed in the previously assigned subgroups. Three C. lanceolata specific subgroups were identified and they were named as subgroup 13b and ClMYB15, which harboured single candidate, ClMYB24 and ClMYB15, respectively. Subgroup ClMYB17 contained two candidates, ClMYB17 and ClMYB18. Furthermore, subgroup 9c containing ClMYB28 and PgMYB11 represented a gymnosperm specific lineage.
Only seven common subgroups (S4, SAtM103, SAtM46, S13a, SAtM85, S22, S21) contained sequences from all the selected plants (Fig. 2). For the MYBs that are important in regulating secondary wall development in A. thaliana, seventeen of them were distributed across nine subgroups (Table 1; Supplementary Figure S2). AtMYB46 and AtMYB83 (subgroup SAtM46) were the second layer master switches to secondary wall developmental growth [38, 39]. AtMYB52, AtMYB54 and AtMYB69 (subgroup S21) were proposed to be the activator of lignin, xylan and cellulose biosynthesis [36, 40]. AtMYB20, AtMYB42, AtMYB43 and AtMYB85 (subgroup SAtM85) and AtMYB58 and AtMYB63 (subgroup S2 & S3) were the activators of lignin biosynthesis [41, 42]. AtMYB26 (subgroup SAtM26) was involved in secondary cell wall thickening in anther endothecium [43, 44]. AtMYB61 (subgroup SAtMYB61) was proposed to act as a mediator to allocate resources to sink tissues such as seed coat [45], root and xylem [46] by modulating genes related cell wall biosynthesis. Besides, two subgroups, namely S4 and S6, were dominated by MYBs functioning as antagonists of cell wall development. For example, AtMYB4 and AtMYB32 (subgroup S4) were the suppressors of biosynthesis of phenylpropanoid derived compounds [47, 48].
These subgroups implicated for secondary cell wall development in A. thaliana also included those from woody species such as seven MYBs in P. trichocarpa (PtrMYB152, PtrMYB010, PtrMYB128, PtrMYB002, PtrMYB003, PtrMYB020 and PtrMYB021) [5, 13, 49], two E. grandis proteins (EgrMYB1 and EgrMYB2) [14, 15, 50], and three P. taeda proteins (PtMYB1, PtMYB4, and PtMYB8) [18, 19] (Table 1; Supplementary Figure S2). Not all ClMYBs were found in these secondary wall related subgroups. There were nine orthologous ClMYBs (ClMYB1, ClMYB2, ClMYB3, ClMYB4, ClMYB5, ClMYB26, ClMYB27, ClMYB49, and ClMYB51) segregated across six clades, suggesting they are the candidate regulators of secondary wall development during wood formation in C. lanceolata (Table 1). ClMYB49 belonging to subgroup SAtM46 was a homolog of AtMYB46 and AtMYB83, meanwhile ClMYB2 and ClMYB26 were clustered with subgroup S21 and SAtM103, respectively. For the subgroups associated with activation of lignin biosynthesis (subgroup SAtM85 and S2&S3), ClMYB1 and ClMYB51 belonged to SAtM85, while none were found in S2&S3. Meanwhile, ClMYB3, ClMYB4 and ClMYB5 were clustered with subgroup S4, which harbored the repressors of phenylpropanoid metabolism.
From the evolutionary analysis of R2R3-MYBs from E. grandis, MYBs of strictly woody species are detected in a total of five woody-preferential subgroups [12]. In the present study, we found that only ClMYB21 present in the woody-preferential subgroup III (Supplementary Figure S2). Subgroup 5 (S5) and AtMYB5 (SAtM5) contained a high proportion of MYBs from woody perennial plant species and were referred to as woody-expanded subgroups, with eight and six ClMYBs assigned to these two subgroups, respectively.
Table 1
Summary of R2R3-MYB transcription factors related to development of secondary cell wall.
Function
|
R2R3-MYB related to SCW development
|
Group name
|
Homologous ClMYBs
|
Biological process
|
Reference
|
Activator
|
AtMYB52, AtMYB54, AtMYB69
|
S21
|
ClMYB2
|
Lignin, xylan and cellulose biosynthesis
|
[36, 40]
|
AtMYB20, AtMYB42, AtMYB43, AtMYB85
PtrMYB152
PtMYB1
|
SAtM85
|
ClMYB1
ClMYB51
|
Lignin biosynthesis. Binds to AtMYB4 to repress flavonoid biosynthesis
|
[19, 42, 49]
|
AtMYB103
PtrMYB010, PtrMYB128
|
SAtM103
|
ClMYB26
|
Lignin biosynthesis
|
[5, 51]
|
AtMYB46, AtMYB83
PtrMYB002, PtrMYB003, PtrMYB020, PtrMYB021
EgrMYB2
PtMYB4
|
SAtM46
|
ClMYB49
|
Lignin, xylan and cellulose biosynthesis
|
[18, 39, 52–55]
|
AtMYB26
|
SAtM26
|
|
Secondary wall thickening in anther endothecium
|
[43, 44]
|
AtMYB58, AtMYB63
|
S2&S3
|
|
Lignin biosynthesis
|
[41]
|
AtMYB61
PtMYB8
|
S13a
|
ClMYB27
|
Regulates stomatal aperture closure, seed coat mucilage biosynthesis, xylem and root development
|
[45, 46, 56]
|
Repressor
|
AtMYB4, AtMYB32
EgrMYB1
|
S4
|
ClMYB3, ClMYB4, ClMYB5
|
Biosynthesis of phenylpropanoid derived compound
|
[47, 48, 50]
|
AtMYB75
|
S6
|
|
Anthocyanin biosynthesis
|
[58]
|
Motif analysis of R2R3-MYB proteins of C. lanceolata
In addition to the R2 and R3 domains, 17 conserved motifs were identified via MEME; these motifs were designated as motifs 1 to 17 in increasing order of their corresponding E-values (Fig. 3; Supplementary Data S1). Motif 1 was upstream of the R2 domain, and the other motifs were downstream of the R3 domain (Fig. 3B). Among the different subgroups, the amount and type of motifs were significantly different. Most of the members belonging to the same subgroup shared one or more motifs (Fig. 3A). For example, ClMYB8, ClMYB9, ClMYB11 and ClMYB29 belonging to S22 contained motif 5, motif 8, and motif 17. Motif 15 carrying the core sequence of (E/I)LIHMADFFQ was unique to ClMYB3, ClMYB4 and ClMYB5 in the subgroup S4. Whereas, motif 17 with the sequence of (D/E)(I/V)NLDL(C/S)ISLP was the ERF-associated amphiphilic repression (EAR) motif that common between angiosperm and gymnosperm in subgroup S4 and dictated the repressive role of the members on the phenylpropanoid metabolism [59].
Expression profiles of ClMYBs in different organs and tissues
To explore the involvement of ClMYB in the secondary wall development, expression profiles of the 52 ClMYB encoding genes in the six organs and tissues were analyzed using the available transcriptome data. Forty-three ClMYB were expressed in all the organs and tissues tested (FPKM > 0), eight of which (ClMYB5, 8, 9, 11, 17, 23, 25, and 47) were constitutively expressed (FPKM > 2 in all samples) (Fig. 4; Supplementary Table S3). Compared with the other ClMYBs, ClMYB9 showed relatively higher expression levels in the organs and tissues examined, with a corresponding mean FPKM value of 100.4. In some cases, the expression profiles of closely related members were similar. For example, the expression patterns of ClMYB22, ClMYB23, and ClMYB25 in the subgroup SAtM5, one of the woody-expanded subgroups, they were clustered in two sub-clades that were highly expressed in female cone. However, in most cases, the expression profiles of the genes present in the same subgroup differed, suggesting their specific roles in different organs. This notion was manifested by ClMYB14 and ClMYB45, though they also belonged to subgroup SAtM5, they were highly expressed in the male cone. Likewise, the expression of another SAtM5 ortholog, ClMYB20 was highly expressed in root and bark tissues.
The expression profiles of the nine candidates that might involve in regulating secondary wall development in lignified tissues could be divided into three categories: ClMYB1, ClMYB26, ClMYB27, ClMYB49, and ClMYB51 were dominantly expressed in the stem xylem. (ClMYB3 and ClMYB5) and (ClMYB2 and ClMYB4) were segregated in sub-clades that exhibited a high expression in the root and female cone, respectively. Since the R2R3-MYBs may act as an activator or repressor to secondary wall development, we followed their expression profile in different stem segments with maturation increasing from S1 to S5 (Fig. 5, Supplementary Table S3). For the xylem specific regulators (ClMYB1, ClMYB26, ClMYB49, and ClMYB51), their expression increased congruent with stem maturation; except ClMYB27 that its expression was indifferent in the developmental stem tissues. The expression of ClMYB3, ClMYB4, and ClMYB5 in the stems tended to decrease with stem maturation, suggesting they may play negative regulatory roles in wood formation. Interestingly, we found that although ClMYB21 belonged to the woody-preferential subgroup III, its gene expression was not significantly different in the developmental stems. The RNA-Seq data were further verified by qRT-PCR experiments on eight representative samples for 12 selected MYB genes (Supplementary Figure S3). Correlation coefficients of 0.845–0.956 (p < 0.01) indicated that the gene expression patterns obtained from the RNA-Seq data were consistent with the qRT-PCR results.
Expression patterns of ClMYBs involved in compression wood development
Trees develop reaction wood in response to a gravistimulus. In gymnosperms, reaction wood usually develops on the lower side of leaning stems or branches and is referred to as compression wood. Typical features of C. lanceolate compression wood could be observed when plants were mechanically bent at a 45° angle from the vertical axis. The tracheid walls of compression wood were significantly thicker than those of normal wood (Fig. 6A, Supplementary Data S2) and contained more lignin but less cellulose [60]. By analysing the expression of ClMYBs encoding genes associated with secondary wall development in the compression wood, we could follow how they were coordinated or reprogrammed that might cause perturbation of anatomical and chemical features of woody tissues in Chinese fir (Fig. 6B, Supplementary Data S3).
Xylem specific ClMYBs generally exhibited increased expression level in compression wood as the bending period prolonged albeit with minor differences. The expression level of ClMYB26, ClMYB27 and ClMYB49 exhibited similar expression patterns, as the expression level in the compression wood was significantly greater than that in the upright wood and reached its maximum after 60 days of bending. The expression level of ClMYB1 in the compression wood was first increased but then decreased with the progression of the bending; also, these levels were significantly higher than those in the upright wood. No significant difference in expression levels between the opposite wood and upright wood, except that ClMYB49 in opposite wood had shown a significant higher expression level than that in the upright wood. The expression level of ClMYB51 in both the compression wood and opposite wood fluctuated and did not exhibit a distinct increment in the compression wood as that shown by other xylem specific MYBs.
The responses of the potential negative regulators ClMYB3, ClMYB4, and ClMYB5 to bending were also different from each other during compression wood induction. The expression level of ClMYB3 in the compression wood was significantly lower than that in the control, but the expression level in the opposite wood was not significantly different from that of upright wood. The expression level of ClMYB4 in the compression wood decreased significantly during the first 10 days and then returned to values similar to that in the upright wood, while the corresponding values of the opposite wood were significantly lower than those of the control and were 0.45 and 0.44 at 30 and 60 days after bending, respectively. Additionally, the expression of ClMYB5 in the compression wood gradually increased, reaching a maximum (2.11) after 60 days of bending. Only after ten days of bending, the expression of this gene in the opposite wood was significantly different from that in the control.
The phylogenetic and gene expression analysis suggested that the function of the xylem specific ClMYBs related to the regulation of secondary wall development might be conserved in C. lanceolata. To verify the protein function, we examined the functionality of ClMYB1 (GenBank accession number: JQ904045) by subcellular localization and overexpression in tobacco plant. ClMYB1 was highly homologous to PtMYB1 and PtrMYB152 (Fig. 7A) and was expressed predominantly in the developing stem, suggesting that this gene participated in secondary wall development in C. lanceolata. The subcellular localization showed that the control GFP was distributed throughout the whole epidermal cell, while the ClMYB1:GFP fusion protein accumulated exclusively in the nucleus (Fig. 7B). The results indicated that ClMYB1 is localised in the nucleus, which is consistent with its function as a transcription factor. As shown in Fig. 7C, the yeast transformants harboring the negative control plasmid pGBKT7 and the recombinant plasmid pGBKT7-ClMYB1 were both able to grow on SD/-Trp media, but only the yeast containing the recombinant plasmid grew well on the selection media and induced the expression of the α-galactosidase reporter gene, indicating that ClMYB1 is a transcriptional activator. Taken together, these results indicated that ClMYB1 functions as a transcriptional activator in the nucleus.