• We are sorry, but NCBI web applications do not support your browser and may not function properly. More information
Logo of pnasPNASInfo for AuthorsSubscriptionsAboutThis Article
Proc Natl Acad Sci U S A. Sep 4, 2012; 109(36): 14699–14704.
Published online Aug 22, 2012. doi:  10.1073/pnas.1212977109
PMCID: PMC3437841
Plant Biology

Splice variant of the SND1 transcription factor is a dominant negative of SND1 members and their regulation in Populus trichocarpa


Secondary Wall-Associated NAC Domain 1s (SND1s) are transcription factors (TFs) known to activate a cascade of TF and pathway genes affecting secondary cell wall biosynthesis (xylogenesis) in Arabidopsis and poplars. Elevated SND1 transcriptional activation leads to ectopic xylogenesis and stunted growth. Nothing is known about the upstream regulators of SND1. Here we report the discovery of a stem-differentiating xylem (SDX)-specific alternative SND1 splice variant, PtrSND1-A2IR, that acts as a dominant negative of SND1 transcriptional network genes in Populus trichocarpa. PtrSND1-A2IR derives from PtrSND1-A2, one of the four fully spliced PtrSND1 gene family members (PtrSND1-A1, -A2, -B1, and -B2). Each full-size PtrSND1 activates its own gene, and all four full-size members activate a common MYB gene (PtrMYB021). PtrSND1-A2IR represses the expression of its PtrSND1 member genes and PtrMYB021. Repression of the autoregulation of a TF family by its only splice variant has not been previously reported in plants. PtrSND1-A2IR lacks DNA binding and transactivation abilities but retains dimerization capability. PtrSND1-A2IR is localized exclusively in cytoplasmic foci. In the presence of any full-size PtrSND1 member, PtrSND1-A2IR is translocated into the nucleus exclusively as a heterodimeric partner with full-size PtrSND1s. Our findings are consistent with a model in which the translocated PtrSND1-A2IR lacking DNA-binding and transactivating abilities can disrupt the function of full-size PtrSND1s, making them nonproductive through heterodimerization, and thereby modulating the SND1 transcriptional network. PtrSND1-A2IR may contribute to transcriptional homeostasis to avoid deleterious effects on xylogenesis and plant growth.

Wood is an important source of materials and energy. Wood formation is a result of the regulated accumulation of secondary xylem cells (fibers, vessels, and rays in dicots) differentiated from the vascular cambium (1). Differentiation of these cells involves wall thickening accompanied by the biosynthesis of wall components, lignin, cellulose, and hemicelluloses, and it is terminated by programmed cell death (1). Regulation of wood formation is known at the level of transcription factors (TFs). A small group of NAC TFs is implicated in wood formation (2, 3). Much of this knowledge was derived from recent work on xylogenesis in Arabidopsis (46).

Approximately 110 NAC genes are found in the Arabidopsis genome. Of these, five are named SND (Secondary Wall-Associated NAC Domain) (7), and seven are named VND (Vascular-Related NAC Domain) (8, 9). SNDs play more specific roles in fiber cell differentiation, and VNDs are activators of vessel formation. SND1 and VND6/7 can each activate the expression of the same set of 12 downstream TF genes, mostly MYBs (5, 6). SND1 and VND6/7 can also directly or indirectly activate genes associated with lignin, cellulose, and hemicellulose biosynthesis through other TFs that are also part of the secondary cell wall biosynthesis regulatory network (5, 6). In this network, higher-level (such as transacting factors) regulation of SND1 and VND6/7 is expected to prevent these NACs from activating the transcription of a cascade of TFs and pathway genes. When this transcriptional homeostasis is not maintained, stunted growth, ectopic secondary cell wall thickening, and deposition of wall components result, as demonstrated by overexpression of SND and VND genes in Arabidopsis or poplar (2). However, nothing is known about the higher-level regulation of SND or VND.

Most TFs, including NACs, dimerize to transactivate target genes in the nucleus (10). After protein synthesis in the cytoplasm, TF translocation as monomers or dimers into the nucleus therefore offers fundamental strategies for transcriptional regulation (10, 11). A few NACs, such as Arabidopsis SND1 and VND6, were shown to be in the nucleus (7, 9, 12, 13), but, surprisingly, the subcellular locations of NAC dimers have never been demonstrated. Knowledge about nucleocytoplasmic transport, including the monomer-to-dimer transition of SNDs and VNDs, is central to a more comprehensive understanding of transcriptional regulation in xylogenesis.

Transcriptional repression through repressor and corepressor TFs is also a useful mechanism for maintaining transcriptional homeostasis (14, 15). Many such TFs are dominant negatives derived from mutations or alternative splicing that disrupt TFs’ DNA-binding function, but not their protein–protein interaction or dimerization ability (16, 17). The defective TF can still heterodimerize or homodimerize with a functional TF, forming a complex where there is only one DNA-binding or activation domain instead of the required two, resulting in nonfunctional protein complexes (15, 17, 18). Both dominant-negative mutations and splice variants of TFs have been extensively studied in animals (1518). In plants, there are two very recent reports on two alternative splice forms in Arabidopsis TFs that act as regulators inhibiting their full-size gene product from activating a direct downstream target (19, 20). In animals, a dominant negative is often an antagonist of a set of targets or of multiple members of a TF family (17, 18). This more effective means of maintaining transcriptional homeostasis, which may be important for regulating hierarchical transcriptional networks, has yet to be demonstrated in plants.

Here we describe the discovery of naturally occurring alternative splicing of PtrSND1-A2, a Populus trichocarpa SND1 gene family member. The splice variant PtrSND1-A2IR is shown to be a negative regulator of multiple PtrSND1 gene family members and a MYB gene—a direct target of these members. We conducted transient transcriptional perturbation in P. trichocarpa stem-differentiating xylem (SDX) protoplasts, transactivation, and electrophoretic mobility shift assays (EMSAs). The findings were integrated with results of SDX subcellular protein colocalization and translocation, yeast two-hybridization (Y2H), and bimolecular fluorescence complementation (BiFC) to provide further evidence for the unique regulatory function of this splice variant.


Naturally Occurring Splice Variant of a PtrSND1 Gene Was Identified Through PCR Cloning, RNA Sequencing (RNA-Seq), and 3′ Rapid Amplification of cDNA Ends (RACE).

We identified 20 SND and VND homologs in P. trichocarpa and demonstrated that essentially all of them are preferentially expressed in SDX (Fig. S1 and Table S1). We focused on all four SND1 homologs, which we named PtrSND1-A1 (POPTR_0011s15640; also named PtVNS12/PtrWND1A; refs. 2 and 3), PtrSND1-A2 (POPTR_0001s45250; PtVNS11/PtrWND1B; refs. 2 and 3), PtrSND1-B1 (POPTR_0014s10060; PtVNS09/PtrWND2A; refs. 2 and 3), and PtrSND1-B2 (POPTR_0002s17950; PtVNS10/PtrWND2B; refs. 2 and 3). PtrSND1-A1 and -A2 share 90.1% protein sequence identity (Fig. S2) and are phylogenetically paired gene members, as are PtrSND1-B1 and -B2 (81.9% protein sequence identity). We then cloned the cDNAs of these four member genes to study their transcriptional functions.

These four genes have a typical NAC gene structure of three exons and two introns, encoding cDNAs of ~1.2–1.3 kb. The cDNA with the expected size for each of the four PtrSND1s was PCR-amplified from SDX (Fig. 1A) and verified by sequencing. There was a larger product (~1.7 kb) from PtrSND1-A2 (Fig. 1A) that retained the second intron from incomplete splicing of the PtrSND1-A2 gene (Version 2.0; http://www.phytozome.org). The 1.7-kb cDNA variant is more abundant in SDX than in phloem, young shoots, and roots (Fig. 1B). We readily PCR-amplified this 1.7-kb cDNA from the SDX of 12 independent P. trichocarpa plants (6–9 mo old) maintained under normal greenhouse conditions and tested at different times. Furthermore, RNA-seq of the SDX transcripts of another set of three P. trichocarpa plants confirmed the inclusion of intron 2 in the PtrSND1-A2 mRNA (Fig. 1 C and D). Essentially, no intron sequence reads were found for PtrSND1-A1, -B1, or -B2 mRNAs (Fig. S3).

Fig. 1.
Discovery of PtrSND1-A2IR. (A) PCR of PtrSND1 members PtrSND1-A1 (A1), PtrSND1-A2 (A2), PtrSND1-B1 (B1), and PtrSND1-B2 (B2). (B) qRT-PCR of PtrSND1-A2IR (A2IR) tissue-specific expression. (C and D) RNA-seq analysis of PtrSND1-A2. (C) The 20 nt in intron ...

We next tested whether the cloned 1.7-kb intron-retaining cDNA was derived from the mature mRNA. We conducted 3′ RACE PCR on SDX RNAs to amplify sequences flanking ATG and poly(A) tail (Fig. 1E, ii and iii) and obtained three products (Fig. 1F). Sequencing of these products verified that each had a 3′ poly(A) tail. The largest product (~1.9 kb; Fig. 1F) included the sequence of the second intron of the PtrSND1-A2 gene (Fig. 1E, iii)—a poly(A)-tailed version of the 1.7-kb cDNA described above. The other two products (Fig. 1F) had no introns (Fig. 1E, ii) but encoded an identical protein, and their size difference was due to the different polyadenylation sites on the completely spliced PtrSND1-A2 mRNA. For each of the other three PtrSND1s, 3′ RACE PCR resulted in amplification of only the full cDNA. These results demonstrated that this alternative splicing event occurs consistently and naturally, derived from a gene-member-specific mature mRNA that shows SDX specificity. We named this Intron Retained splice variant PtrSND1-A2IR, a unique transcript of the PtrSND1 family.

Splice Variant PtrSND1-A2IR Encodes a Unique NAC-Domain Protein in P. trichocarpa SDX cells.

We next analyzed whether the PtrSND1-A2IR mRNA is processed for protein production. The full-size PtrSND1-A2 cDNA was predicted to produce a protein of 418 amino acids (aa) with a conserved N-terminal NAC domain (180 aa) and a C-terminal activation domain (238 aa) (refs. 12 and 22; Fig. 1E, iv). The NAC domain is encoded by exons 1 and 2 and the β6 subdomain from exon 3 (22). The activation domain is encoded by the exon 3 portion without the β6 sequence. The PtrSND1-A2IR cDNA (1.9 kb; Fig. 1F) encodes a predicted protein of only 166 aa because of a premature termination codon (PTC) (Fig. 1E, iii) in the retained intron 2. As a result, PtrSND1-A2IR would be a NAC-domain protein that has no activation domain but has a protein dimerization domain (β′ and α1a/b), β1–β5 subdomains, and a unique C terminus of 12 aa translated from the retained intron 2 portion upstream of the PTC (Fig. 1E, v).

We then performed Western blot analysis to test for the presence of PtrSND1-A2IR protein. Protein-specific polypeptides (Fig. 1E, iv and v) were selected as immunogens to make antibodies that would distinguish PtrSND1-A2IR and -A2 and discriminate these from the other three full-size PtrSND1 proteins. Antibody specificity was validated against Escherichia coli recombinant proteins from the five PtrSND1 genes (Fig. 1G, i). The SDX proteins gave two bands with sizes corresponding to the predicted molecular masses of PtrSND1-A2 (47.2 kDa) and PtrSND1-A2IR (19.5 kDa), respectively, by using antibodies that would recognize the NAC domain of these two proteins (Fig. 1H, i) but not of the other three PtrSND1s (Fig. 1G, i). The identity of these two proteins was further discriminated by anti–PtrSND1-A2–specific antibody (Fig. 1H, ii). These results demonstrate the presence of a unique NAC protein, PtrSND1-A2IR, in P. trichocarpa SDX. We then characterized the function of this protein and its regulatory relationship with the other four full-size PtrSND1 members.

PtrSND1-A2IR Inhibits PtrMYB021 Gene Expression.

In Arabidopsis, SND1 activates directly the expression of AtMYB46 and AtMYB83 (23). We tested whether all five P. trichocarpa SND1 members (including PtrSND1-A2IR) can directly transactivate MYB gene expression in P. trichocarpa. We focused on PtrMYB021 (POPTR_0009s05860), the ortholog of AtMYB46 (24). We overexpressed each of the five PtrSND1 genes in P. trichocarpa SDX protoplasts, using protocols that we recently developed (25). All four full-size PtrSND1s could induce a twofold to fourfold increase in abundance of endogenous PtrMYB021 transcripts in SDX protoplasts (Fig. 2A), consistent with the known SND1-mediated transactivation of MYB targets. In contrast, overexpression of PtrSND1-A2IR significantly reduced the PtrMYB021 transcript level (Fig. 2A). A repression of MYB expression by any fully spliced SND1 has not been previously detected. Effector–reporter-based gene transactivation assays further suggested that the observed induction of PtrMYB021 expression was a result of the activation of the PtrMYB021 promoter by the PtrSND1 member (Fig. 2B). The assays also showed that PtrSND1-A2IR could not activate the PtrMYB021 promoter, consistent with the lack of an activation domain in PtrSND1-A2IR and suggesting that the observed PtrSND1-A2IR–mediated suppression of PtrMYB021 expression (Fig. 2A) operates by other repression mechanisms.

Fig. 2.
PtrSND1-A2IR inhibits expression of the PtrMYB021 (MYB021) gene and the PtrSND1 gene members. (A) qRT-PCR analysis of endogenous MYB021 transcript abundance in P. trichocarpa SDX protoplasts overexpressing individually the five PtrSND1 members. pUC19-35S-sGFP ...

We next used EMSA to test whether the PtrSND1-mediated transactivation of PtrMYB021 is a result of direct binding of PtrSND1 to the PtrMYB021 promoter. Retardation of DNA probe mobility and probe competition demonstrated that each of the four full-size PtrSND1 members can directly bind to a conserved sequence motif in the PtrMYB021 promoter (Fig. 2C and Fig. S4). However, PtrSND1-A2IR did not bind to the PtrMYB021 promoter (Fig. 2C). The lack of DNA-binding ability suggests that the β6 subdomain (22, 26), which is missing in PtrSND1-A2IR (Fig. 1E, v), plays an important role in DNA binding. These results confirm that PtrMYB021 is a common and direct transactivation target of the four full-size PtrSND1 members. Transactivation and EMSA results revealed that the splice variant PtrSND1-A2IR negatively regulates PtrMYB021 gene expression through a mechanism that is independent of an activation domain and direct PtrMYB021 DNA-binding ability in PtrSND1-A2IR.

PtrSND1-A2IR Inhibits the Expression of PtrSND1 Gene Members.

We investigated further the regulatory role of PtrSND1-A2IR by testing whether PtrSND1-A2IR can also affect the expression of the other four PtrSND1 members. Overexpression of PtrSND1-A2IR in P. trichocarpa SDX protoplasts resulted in drastically reduced transcript abundances of the endogenous PtrSND1-A1, -B1, and -B2, but had essentially no effect on the expression of its full-size isoform, PtrSND1-A2 (Fig. 2D). Because the transcript level of a gene is a function of the abundance of the TF that regulates the expression of the gene, the reduced expression of PtrSND1-A1, -B1, and -B2 by PtrSND1-A2IR suggests that PtrSND1-A2IR inhibits the expression of the TF that regulates these genes. To test this suggestion, we first verified whether each of the four full-size PtrSND1 members is a TF activating its own gene. Many TFs in animals and plants regulate their own expression (self-activation or -repression) in addition to responding to other input signals (13, 27).

Transactivation assays confirmed that each of the four full-size PtrSND1 members could activate its own promoter (self-activation), indicated by induced β-glucuronidase (GUS) activity (Fig. 2E). Using EMSA, we further demonstrated that such self-activation is a result of direct binding of PtrSND1 to its own promoter (Fig. 2F and Fig. S4). These results strongly suggest that PtrSND1-A2IR can suppress PtrSND1-A1, -B1, and -B2 gene expression (Fig. 2D) through inhibiting the self-activation (Fig. 2E) of these PtrSND1s. The inability of PtrSND1-A2IR to suppress PtrSND1-A2 expression (Fig. 2D, ii) suggests that PtrSND1-A2 is transactivated not only by its own protein (Fig. 2E, ii) but also by other unknown TFs or input signals. PtrSND1-A2IR may attenuate the self-activation of PtrSND1-A2 but have no effect on PtrSND1-A2 expression induced by other factors, resulting in an unchanged PtrSND1-A2 transcript level (Fig. 2D, ii).

To further investigate the PtrSND1-A2IR–mediated suppression of PtrSND1 genes, we performed transactivation assays of PtrSND1-A2IR on PtrSND1 promoters. The results demonstrated that PtrSND1-A2IR could not activate the promoters of PtrSND1 member genes (Fig. S5), just as PtrSND1-A2IR could not activate PtrMYB021 (Fig. 2B). EMSA showed that the lack of such activation is because PtrSND1-A2IR cannot bind to the promoters of PtrSND1 or PtrMYB021 (Fig. 2 C and F and Fig. S4). These results suggest that the PtrSND1-A2IR–mediated attenuation of PtrSND1 and PtrMYB021 gene expression (Fig. 2 A and D) follows a similar mechanism where a direct binding of PtrSND1-A2IR to the promoter of its target gene is not essential. By having a dimerization domain intact near the N terminus (Fig. 1E, v) (22, 26), PtrSND1-A2IR may use its dimerization ability to prevent full-size PtrSND1 proteins from accomplishing their normal activation functions, leading to attenuated expression of PtrSND1 and PtrMYB021 (Fig. 2 A and D).

PtrSND1-A2IR acts as a unique gene repression mediator, regulating the SND1 transcriptional network in P. trichocarpa. PtrSND1-A2IR is neither an active nor a passive repressor, such as VNI2, which reduces VND7 transactivation activity as a regulator of VND7 (28). However, our results strongly suggest the involvement of specific protein–protein interactions in the PtrSND1-A2IR transcriptional regulation. We then investigated the subcellular localization of all five PtrSND1 members, as well as specific dimerization and subcellular locations of these dimers.

PtrSND1-A2IR Is in Cytoplasmic Foci and the Four Full-Size PtrSND1s Are in the Nucleus of P. trichocarpa SDX Cells.

We used fluorescent proteins to reveal the subcellular location of PtrSND1 members in P. trichocarpa SDX protoplasts. We cotransfected the protoplasts with 35S-PtrSND1:sGFP and 35S-H2A-1:mCherry nuclear marker plasmids (29) and demonstrated that each of the four full-size PtrSND1s colocalizes with the marker in the nucleus (Fig. 3 AD). Exclusive nuclear location of these four PtrSND1s was observed for 85–95% of the transfected protoplasts examined, whereas in the remaining protoplasts, PtrSND1s were found in both nucleus and cytoplasmic foci. PtrMYB021 was also predominantly in the nucleus (Fig. 3F).

Fig. 3.
Protein colocalization in P. trichocarpa SDX protoplasts demonstrate that PtrSND1-A2IR is translocated from cytoplasmic foci to the nucleus by full-size PtrSND1 members. (AF) Subcellular localization of five PtrSND1 fusion proteins: A1 (A), A2 ...

In sharp contrast to the nuclear location of the full-size PtrSND1s, cotransfection of 35S-PtrSND1-A2IR:sGFP and 35S-H2A-1:mCherry plasmids demonstrated that the splice variant PtrSND1-A2IR was located exclusively in small punctate structures in the cytoplasm of the protoplasts (Fig. 3E). The exclusive location of PtrSND1-A2IR in cytoplasmic foci and sporadically abnormal residence of PtrSND1s in these foci suggest that PtrSND1-A2IR may retain these PtrSND1s in the cytoplasm through protein–protein interactions. If PtrSND1-A2IR interacts with any PtrSND1, these proteins should be colocalized. We then performed protein subcellular colocalization experiments.

Cytoplasmic PtrSND1-A2IR Can Be Translocated into the Nucleus by Full-Size PtrSND1s.

To colocalize PtrSND1-A2IR with each of the five PtrSND1 members, we prepared a 35S-PtrSND1-A2IR:mCherry fusion gene construct and cotransferred it with each of the 35S-PtrSND1:sGFP constructs into P. trichocarpa SDX protoplasts. Unexpectedly, in the presence of any one of the four full-size PtrSND1 members, PtrSND1-A2IR was translocated from the cytoplasmic foci (Fig. 3E) to the nucleus, demonstrated by the nuclear colocalization of PtrSND1-A2IR (mCherry) with each of the four full-size PtrSND1s (sGFP) (Fig. 3 GJ). Such translocation of PtrSND1-A2IR was exclusive to all cotransfected protoplasts.

Cotransfection of SDX protoplasts with 35S-PtrSND1-A2IR:mCherry and 35S-PtrSND1-A2IR:sGFP showed that sGFP and mCherry signals, both representing PtrSND1-A2IR, remained entirely colocalized in cytoplasmic foci (Fig. 3K). As a control experiment, we cotransfected the protoplasts with 35S-PtrSND1-A2IR:sGFP and 35S-PtrMYB021:mCherry and demonstrated that PtrSND1-A2IR remained exclusively in cytoplasmic foci, whereas PtrMYB021 was in the nucleus (Fig. 3L). Therefore, PtrSND1-A2IR can only be translocated into the nucleus by a full-size PtrSND1 carrier, not just any TF proteins.

Y2H Demonstrates That PtrSND1-A2IR Dimerizes with Five PtrSND1 Members.

We next performed Y2H assays to test whether PtrSND1-A2IR can dimerize with the carrier PtrSND1 for translocation into the nucleus. Each of the four full-size PtrSND1s heterodimerized with PtrSND1-A2IR (Fig. 4 AD), and PtrSND1-A2IR self-interacted (Fig. 4E). The interactions for the PtrSND1-A2IR/PtrSND1-A2 (A2IR/A2; Fig. 4B) and the A2IR/A2IR (Fig. 4E) dimers were stronger than those for the A2IR/A1, A2IR/B1, and A2IR/B2 heterodimers. The heterodimerization in yeast and the nuclear colocalization in protoplasts (Fig. 3 GJ) of two proteins [(A2IR and A1), (A2IR and A2), (A2IR and B1), and (A2IR and B2)] are strong evidence that the pairs of proteins exist as heterodimers in plant nuclei. To further test this inference, we performed BiFC assays in P. trichocarpa SDX protoplasts.

Fig. 4.
Y2H demonstrates that PtrSND1-A2IR interacts with each of the five PtrSND1s. The full-size A2IR and the NAC-domain from A1, A2, B1, and B2 were each fused to the Gal4 binding domain (BD) to make BD:A2IR, and BD:A1, BD:A2, BD:B1, and BD:B2 fusions as the ...

BiFC Identifies PtrSND1/PtrSND1-A2IR Heterodimers in the Nucleus of SDX Cells.

Using BiFC, we tested for heterodimers between PtrSND1-A2IR and each of the four full-size PtrSND1s, and all possible homodimers from the five PtrSND1s. Different combinations of two plasmids, each containing a target protein fused at the N terminus to one of two complementing segments of CFP—CFPN (aa 1–173) and CFPC (aa 174–329)—were cotransformed together with the H2A-1:mCherry nuclear marker plasmid into SDX protoplasts. A2IR heterodimerized with each of the full-size PtrSND1s, and these heterodimers (A2IR/A1, A2IR/A2, A2IR/B1, and A2IR/B2) colocalized with the H2A-1:mCherry marker nearly exclusively in the nucleus (Fig. 5 AI). The four full-size PtrSND1s homodimerized, and these dimers (A1/A1, A2/A2, B1/B1, and B2/B2) were also nucleus specific (Fig. 5 JQ). A2IR/A2IR homodimers were exclusively in cytoplasmic foci (Fig. 5 R, S, and I), validating the PtrSND1-A2IR colocalization results (Fig. 3K) and the finding that the translocation of PtrSND1-A2IR into the nucleus requires a full-size PtrSND1.

Fig. 5.
BiFC in P. trichocarpa SDX protoplasts demonstrates that A2IR heterodimerizes with the four full-size PtrSND1s and that all five PtrSND1s homodimerize. (A, C, E, and G) A2IR:CFPC cotransferred with A1:CFPN (A), A2:CFPN (C), B1:CFPN (E), or B2:CFPN (G ...

The results with RNA-Seq, protein gel blotting, transgene transcriptional regulation, gene transactivation, EMSA, protein subcellular localization and colocalization, Y2H, and BiFC provide strong evidence for unique NAC transcriptional regulation, in which a cytoplasmic splice variant, PtrSND1-A2IR, is translocated by its full-size family members into the nucleus for repressing the expression of these members and their direct target genes.


Complex biological processes in growth and development, such as secondary cell wall or wood formation, are regulated at many levels (a hierarchy) by transacting elements (30). SND1s are an essential part of such a hierarchical network. Although SND1s’ downstream targets have been studied extensively, their upstream regulators (for activation or repression) have not previously been identified (6). Repression is as necessary as activation in transcriptional control for growth and development, and it is frequently achieved through dominant-negative TFs (10, 14, 17). In plants, alternative splicing-generated dominant-negative TFs that can act as an individual antagonist against multiple target gene products have not been reported. A dominant-negative TF that can antagonize its own family members on their self-activation is also previously unknown in plants.

In this study, we showed that a naturally occurring, SDX-specific splice variant, PtrSND1-A2IR, is a dominant-negative regulator. It antagonizes the autoregulation of its family members (Fig. 2D) as well as the activation of their common target PtrMYB021 (Fig. 2A). We reported here several distinct mechanisms of PtrSND1 regulation. Typically, a dominant-negative TF lacks a DNA-binding domain (10, 16, 17). PtrSND1-A2IR has the β1 and β2 subdomains (Fig. 1E, v) believed necessary for DNA binding, but lacks the β6 subdomain. However, EMSA demonstrated that PtrSND1-A2IR does not bind to promoters of PtrSND1 member and PtrMYB021 genes (Fig. 2 C and F and Fig. S4). The inclusion of the β6 region, in addition to β1 and β2, in the β-sheet structure may be important for DNA binding. Although PtrSND1-A2IR lacks DNA-binding ability, it dimerizes (Fig. 5) because it retains the N-terminal dimerization domain (Fig. 1E, v).

Although homodimers of full-size PtrSND1 members are nucleus specific (Fig. 5 JQ), PtrSND1-A2IR remained exclusively in the cytoplasm as homodimers or monomers (Figs. 3E and and5R).5R). The cytoplasmic foci location of PtrSND1-A2IR suggests that this protein is involved in posttranscriptional degradation of its target RNAs (31). However, there is no RNA-binding motif in PtrSND1-A2IR that would indicate a degradation mechanism. Instead, PtrSND1-A2IR is translocated into the nucleus as a heterodimeric partner with any of its full-size PtrSND1 family members (Figs. 3 G–J and 5 A–I). The formation of these heterodimers must occur in the cytoplasm because PtrSND1-A2IR monomers or homodimers are not available in the nucleus. This subcellular translocation of a splice variant is unique and may represent an integral part of the nucleocytoplasmic transport required for the SND1 transcriptional regulation.

In the nucleus, the PtrSND1-A2IR/PtrSND1 heterodimer has only one functional DNA-binding structure (from the full-size PtrSND1), instead of the required two. Consequently, the defective DNA-binding system may disrupt subsequent gene transactivation, resulting in suppressed self-activation of PtrSND1s (Fig. 2D). The suppressed level of PtrSND1s would then lead to reduced transactivation of their target genes, such as PtrMYB021, resulting in attenuated transcript levels of PtrMYB021, as observed in SDX protoplasts in which PtrSND1-A2IR is overexpressed (Fig. 2A). It is also possible that the defective DNA-binding system created by the PtrSND1-A2IR/PtrSND1 heterodimer directly disrupted the normal PtrSND1-mediated transactivation of PtrMYB021, causing a reduced level of PtrMYB021 transcripts (Fig. 2A).

In addition to the dual DNA-binding specificity, a set of two transactivation or transrepression domains through dimerization is also necessary for TFs to be fully functional in regulating target gene expression (10, 14, 16, 17). The PtrSND1-A2IR/PtrSND1 heterodimer has only one transactivation domain, further disrupting the normal PtrSND1 functions to result in inhibited expression of PtrSND1 and PtrMYB021 genes (Fig. 2 A and D). This mechanism may also explain the repressed vessel differentiation due to inhibited VND7 function in Arabidopsis in which an artificially truncated VND7 lacking the transactivation domain was overexpressed (9). These inhibition mechanisms have been well characterized for many dominant-negative repressors. Among the best-studied repressors are the members of the Id (inhibitor of DNA-binding) subfamily of basic helix–loop–helix TFs involved in cell cycle and tumorigenesis in animals (18). Ids lack a DNA-binding region, and, consequently, they disrupt the function of other bHLH TFs, making them nonproductive through dimerization (18).

The formation of nonproductive PtrSND1-A2IR/PtrSND1 heterodimers would also prevent full-size PtrSND1s from dimerizing to generate functional complexes. This sequestration of functional PtrSND1s together with the dominant-negative effects of PtrSND1-A2IR may efficiently maintain transcriptional homeostasis in the SND- and VND-regulated networks. A few such dominant negatives may have evolved from genome duplication in woody plants to dampen adverse effects due to transcriptional redundancy. PtrSND1-A2IR has provided an entry to discovering more about the upstream regulation of SND and VND and to reveal the comprehensive hierarchical network regulating wood formation.

Materials and Methods

Plant materials, PCR cloning, RNA-Seq, and 3′RACE, qRT-PCR, SDS/PAGE, Western blotting, P. trichocarpa SDX protoplast preparation/transfection, gene transactivation assays, plasmid constructions, EMSA, Y2H, and BiFC are described in detail in SI Materials and Methods. They are described in the sequence shown above. All primer sequences are listed in Table S1.

Supplementary Material

Supporting Information:


We thank Kyung-Hwan Han and Won-Chan Kim for analyzing the PtrSND1-A2IR for RNA binding motifs. This work was supported by the Office of Science (Biological and Environmental Research), Department of Energy Grant DE-SC000691 (to V.L.C.). X.-H.Z. thanks the sabbatical support of the North Carolina State University Jordan Family Endowment.


The authors declare no conflict of interest.

This article contains supporting information online at www.pnas.org/lookup/suppl/doi:10.1073/pnas.1212977109/-/DCSupplemental.


1. Evert RF. Esau's Plant Anatomy: Meristems, Cells, and Tissues of the Plant Body: Their Structure, Function, and Development. 3rd Ed. Hoboken, NJ: Wiley & Sons; 2006.
2. Ohtani M, et al. A NAC domain protein family contributing to the regulation of wood formation in poplar. Plant J. 2011;67:499–512. [PubMed]
3. Zhong R, McCarthy RL, Lee C, Ye ZH. Dissection of the transcriptional program regulating secondary wall biosynthesis during wood formation in poplar. Plant Physiol. 2011;157:1452–1468. [PMC free article] [PubMed]
4. Ko JH, Yang SH, Park AH, Lerouxel O, Han KH. ANAC012, a member of the plant-specific NAC transcription factor family, negatively regulates xylary fiber development in Arabidopsis thaliana. Plant J. 2007;50:1035–1048. [PubMed]
5. Demura T, Ye ZH. Regulation of plant biomass production. Curr Opin Plant Biol. 2010;13:299–304. [PubMed]
6. Wang HZ, Dixon RA. On-off switches for secondary cell wall biosynthesis. Mol Plant. 2012;5:297–303. [PubMed]
7. Zhong R, Demura T, Ye ZH. SND1, a NAC domain transcription factor, is a key regulator of secondary wall synthesis in fibers of Arabidopsis. Plant Cell. 2006;18:3158–3170. [PMC free article] [PubMed]
8. Kubo M, et al. Transcription switches for protoxylem and metaxylem vessel formation. Genes Dev. 2005;19:1855–1860. [PMC free article] [PubMed]
9. Yamaguchi M, Kubo M, Fukuda H, Demura T. Vascular-related NAC-DOMAIN7 is involved in the differentiation of all types of xylem vessels in Arabidopsis roots and shoots. Plant J. 2008;55:652–664. [PubMed]
10. Klemm JD, Schreiber SL, Crabtree GR. Dimerization as a regulatory mechanism in signal transduction. Annu Rev Immunol. 1998;16:569–592. [PubMed]
11. Nigg EA. Nucleocytoplasmic transport: Signals, mechanisms and regulation. Nature. 1997;386:779–787. [PubMed]
12. Xie Q, Frugis G, Colgan D, Chua NH. Arabidopsis NAC1 transduces auxin signal downstream of TIR1 to promote lateral root development. Genes Dev. 2000;14:3024–3036. [PMC free article] [PubMed]
13. Wang H, Zhao Q, Chen F, Wang M, Dixon RA. NAC domain function and transcriptional control of a secondary cell wall master switch. Plant J. 2011;68:1104–1114. [PubMed]
14. Gaston K, Jayaraman P-S. Transcriptional repression in eukaryotes: Repressors and repression mechanisms. Cell Mol Life Sci. 2003;60:721–741. [PubMed]
15. Seo PJ, Hong SY, Kim SG, Park CM. Competitive inhibition of transcription factors by small interfering peptides. Trends Plant Sci. 2011;16:541–549. [PubMed]
16. Müller CW. Transcription factors: Global and detailed views. Curr Opin Struct Biol. 2001;11:26–32. [PubMed]
17. Amoutzias GD, Robertson DL, Van de Peer Y, Oliver SG. Choose your partners: Dimerization in eukaryotic transcription factors. Trends Biochem Sci. 2008;33:220–229. [PubMed]
18. Ruzinova MB, Benezra R. Id proteins in development, cell cycle and cancer. Trends Cell Biol. 2003;13:410–418. [PubMed]
19. Seo PJ, Kim MJ, Ryu JY, Jeong EY, Park CM. Two splice variants of the IDD14 transcription factor competitively form nonfunctional heterodimers which may regulate starch metabolism. Nat Commun. 2011;2:303. [PubMed]
20. Seo PJ, et al. A self-regulatory circuit of CIRCADIAN CLOCK-ASSOCIATED1 underlies the circadian clock regulation of temperature responses in Arabidopsis. Plant Cell. 2012;24:2427–2442. [PMC free article] [PubMed]
21. Zhong R, Lee C, Ye ZH. Functional characterization of poplar wood-associated NAC domain transcription factors. Plant Physiol. 2010;152:1044–1055. [PMC free article] [PubMed]
22. Ernst HA, Olsen AN, Larsen S, Lo Leggio L, Lo Leggio L. Structure of the conserved domain of ANAC, a member of the NAC family of transcription factors. EMBO Rep. 2004;5:297–303. [PMC free article] [PubMed]
23. Zhong R, Richardson EA, Ye ZH. The MYB46 transcription factor is a direct target of SND1 and regulates secondary wall biosynthesis in Arabidopsis. Plant Cell. 2007;19:2776–2792. [PMC free article] [PubMed]
24. Ko JH, Kim WC, Han KH. Ectopic expression of MYB46 identifies transcriptional regulatory genes involved in secondary wall biosynthesis in Arabidopsis. Plant J. 2009;60:649–665. [PubMed]
25. Chen HC, et al. Membrane protein complexes catalyze both 4- and 3-hydroxylation of cinnamic acid derivatives in monolignol biosynthesis. Proc Natl Acad Sci USA. 2011;108:21253–21258. [PMC free article] [PubMed]
26. Puranik S, Sahu PP, Srivastava PS, Prasad M. NAC proteins: Regulation and role in stress tolerance. Trends Plant Sci. 2012;17:369–381. [PubMed]
27. Becskei A, Serrano L. Engineering stability in gene networks by autoregulation. Nature. 2000;405:590–593. [PubMed]
28. Yamaguchi M, et al. VND-INTERACTING2, a NAC domain transcription factor, negatively regulates xylem vessel formation in Arabidopsis. Plant Cell. 2010;22:1249–1263. [PMC free article] [PubMed]
29. Tenea GN, et al. Overexpression of several Arabidopsis histone genes increases agrobacterium-mediated transformation and transgene expression in plants. Plant Cell. 2009;21:3350–3367. [PMC free article] [PubMed]
30. Yu H, Gerstein M. Genomic analysis of the hierarchical structure of regulatory networks. Proc Natl Acad Sci USA. 2006;103:14724–14731. [PMC free article] [PubMed]
31. Macara IG. Transport into and out of the nucleus. Microbiol Mol Biol Rev. 2001;65:570–594. [PMC free article] [PubMed]

Articles from Proceedings of the National Academy of Sciences of the United States of America are provided here courtesy of National Academy of Sciences
PubReader format: click here to try


Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...


Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...