CsRAV1 induces sylleptic branching in hybrid poplar


  • Alicia Moreno-Cortés,

    1. Centro de Biotecnología y Genómica de Plantas UPM/INIA, Departamento de Biotecnología, E. T. S. Ingenieros de Montes, Universidad Politécnica de Madrid, Campus de Montegancedo, Pozuelo de Alarcón, E-28223 Madrid, Spain
    Search for more papers by this author
  • Tamara Hernández-Verdeja,

    1. Centro de Biotecnología y Genómica de Plantas UPM/INIA, Departamento de Biotecnología, E. T. S. Ingenieros de Montes, Universidad Politécnica de Madrid, Campus de Montegancedo, Pozuelo de Alarcón, E-28223 Madrid, Spain
    Search for more papers by this author
  • Paloma Sánchez-Jiménez,

    1. Centro de Biotecnología y Genómica de Plantas UPM/INIA, Departamento de Biotecnología, E. T. S. Ingenieros de Montes, Universidad Politécnica de Madrid, Campus de Montegancedo, Pozuelo de Alarcón, E-28223 Madrid, Spain
    Search for more papers by this author
  • Pablo González-Melendi,

    1. Centro de Biotecnología y Genómica de Plantas UPM/INIA, Departamento de Biotecnología, E. T. S. Ingenieros de Montes, Universidad Politécnica de Madrid, Campus de Montegancedo, Pozuelo de Alarcón, E-28223 Madrid, Spain
    Search for more papers by this author
  • Cipriano Aragoncillo,

    1. Centro de Biotecnología y Genómica de Plantas UPM/INIA, Departamento de Biotecnología, E. T. S. Ingenieros de Montes, Universidad Politécnica de Madrid, Campus de Montegancedo, Pozuelo de Alarcón, E-28223 Madrid, Spain
    Search for more papers by this author
  • Isabel Allona

    1. Centro de Biotecnología y Genómica de Plantas UPM/INIA, Departamento de Biotecnología, E. T. S. Ingenieros de Montes, Universidad Politécnica de Madrid, Campus de Montegancedo, Pozuelo de Alarcón, E-28223 Madrid, Spain
    Search for more papers by this author

  • Patent P201131186. Method of increasing or reducing the development of sylleptic and/or proleptic branches in a woody plant. Filed at the Oficina Española de Patentes y Marcas on July 13, 2011 at 12:39 h.

Author for correspondence:
Isabel Allona
Tel: +34 913364543
E-mail: isabel.allona@upm.es


  • Sylleptic branching in trees may increase significantly branch number, leaf area and the general growth of the tree, particularly in its early years. Although this is a very important trait, so far little is known about the genes that control this process.
  • This article characterizes the Castanea sativa RAV1 gene, homologous to Arabidopsis TEM genes, by analyzing its circadian behavior and examining its winter expression in chestnut stems and buds. Transgenic hybrid poplars over-expressing CsRAV1 or showing RNA interference down-regulated PtaRAV1 and PtaRAV2 expression were produced and analyzed.
  • Over-expression of the CsRAV1 gene induces the early formation of sylleptic branches in hybrid poplar plantlets during the same growing season in which the lateral buds form. Only minor growth differences and no changes in wood anatomy are produced.
  • The possibility of generating trees with a greater biomass by manipulating the CsRAV1 gene makes CsRAV1 transgenic plants promising candidates for bioenergy production.


Obtaining bioenergy from plants, particularly perennial grasses and trees, could help to alleviate the effects of global warming and energy safety problems, provided that high yields can be sustained. Bioenergy crops generate power and produce liquid transport fuels (Karp & Shield, 2008). As purpose-grown wood, short-rotation woody crops, such as the fast-growing species Populus, Salix and Eucalyptus and their respective hybrids, have been proposed for high productivity and proximity to the processing plant (Hinchee et al., 2009). Fast-growing poplars can be cultivated in short-rotation 15–18-yr forestry cycles, but, if grown as short-rotation coppice (SRC), this interval is reduced by cutting back/coppicing at 3–5-yr intervals (Tschaplinski & Blake, 1989; Sennerby-Forsse, 1995). SRC crops are easy to establish and provide a multifunctional fuel source, as well as offering secondary benefits, such as low nutrient input, good energy balance, bioremediation abilities and increased biodiversity (Sims et al., 2006; Rowe et al., 2009).

The great challenge for biomass production is to develop crops with a suite of desirable physical and chemical traits, whilst increasing biomass yields by a factor of two or more (Ragauskas et al., 2006; Vanholme et al., 2010). In addition to targeting photosynthesis to increase the initial capture of light energy, the manipulation of genes involved in nitrogen metabolism has also been a successful approach to increase biomass (Jing et al., 2004; Van Camp, 2005). So far, however, breeding and research efforts have focused on the single-stem growth of poplars, and there is a need to identify traits and genomic loci to create improved SRC biomass-yielding genotypes. Woody biomass yield is a highly complex trait as it represents the combined outcome of many other complex traits, each under separate polygenic control (Novaes et al., 2009; Rae et al., 2009).

The lateral buds of most temperate woody species do not grow out during the season in which they form. These proleptic buds overwinter and grow out during the following spring. However, in poplar and a few other temperate species, as well as many tropical species, some lateral buds grow out sylleptically, that is, they grow out during the same season in which they form without an intervening rest period and usually without bud scale formation (Späth, 1912). Sylleptic branching in poplar may increase significantly branch number, leaf area and the general growth of the tree, particularly in its early years (Ceulemans et al., 1990; Wu et al., 2000). Rae et al. (2009) examined genetic correlations between yield-related traits to identify ‘early diagnostic’ indicators of yield. These authors observed that early biomass was a reasonable predictor of coppice yield, and that leaf size, cell number, and stem and sylleptic branch number were also valuable traits. In a study by Dillen et al. (2009), designed to assess the environmental, temporal and genetic stability of the link between growth and a series of tree architecture, leaf and phenological traits, it was found that the number of sylleptic branches and area of the largest leaf on the main stem were the best predictors of growth, irrespective of the experimental site or hybrid family.

Related to ABI3 and Viviparous 1 (RAV) is a DNA-binding protein with two distinct types of DNA-binding domain: AP2 and B3. The RAV proteins bind as monomers to bipartite recognition sequences through these two distinct DNA-binding domains solely found in higher plants (Kagaya et al., 1999; Yamasaki et al., 2004). RAV1 has been implicated in several aspects of plant physiology and development. Previous studies have shown that RAV gene expression is induced by low temperature, darkness, wounding, drought and salt stress, touch and pathogen attack (Gutiérrez et al., 2002; Fowler et al., 2005; Lee et al., 2005; Sohn et al., 2006; Kagaya & Hattori, 2009; Li et al., 2011). In addition, RAV gene expression has been attributed a role in mediating plant hormone responses (Hu et al., 2004; Zhao et al., 2008), RAV1 and RAV2 have also been ascribed to a novel group of transcriptional repressors in Arabidopsis (Ikeda & Ohme-Takagi, 2009). and TEMPRANILLO (TEM1), a transcription factor belonging to the RAV family, has been reported to bind and repress the promoter region of the FLOWERING LOCUS T (FT) gene (Castillejo & Pelaz, 2008).

Here, we describe the cloning and characterization of the chestnut RAV1 gene, and show that the over-expression of this gene in hybrid poplar induces sylleptic branching as a profitable trait to increase biomass yield.

Materials and Methods

Plant materials and growth conditions

For the four-season experiments, stem and bud specimens were harvested from European chestnut trees (Castanea sativa Mill.) in Madrid, Spain (Zarzalejo: 4º11′W, 40º35′N). In the continuous light (LL), long-day (LD) and low temperature (4ºC, LD) experiments, leaf and stem samples were harvested from 16–24-wk-old chestnut seedlings placed in growth chambers under the conditions described in Ramos et al. (2005). Unless otherwise stated, the hybrid poplar Populus tremula × P. alba INRA clone 717 1B4 was used. In wild-type (WT) and transgenic lines, leaf samples were harvested from 6-wk-old plantlets grown in vitro in conditions of 22°C and LD. The PtaRAV1 and PtaRAV2 induction experiment was conducted by exposing these plantlets to 4°C for 3 h. Poplars were cultured in vitro and transferred to 3.5-l pots containing blond peat, pH 4.5. Histochemical and initial phenotype assessments were conducted with plants grown in a glasshouse for 8 wk under conditions of 20 ± 2°C and natural light supplemented with artificial illumination to complete an LD photoperiod. The plants were fertilized once every 2 wk with a solution of 1 g l−1 Peters Professional 20-20-20 (Everris International, Geldermalsen, the Netherlands) and 1 g l−1 Welgro Micromix trace elements (Comercial Química Massó, Barcelona, Spain). For detailed phenotyping, plants were grown in a growth chamber under controlled conditions (21°C, LD, 65% relative humidity and 150 μmol light intensity) and fertilized with the same solution just once, 2 wk after transplantation.

CsRAV1 cDNA isolation

A 661-bp-long fragment of CsRAV1 was initially isolated by enriching for winter-specific transcripts using a PCR-Select cDNA Subtraction Kit (Clontech Laboratories, Mountain View, CA, USA). The full-length cDNA clone was obtained using the SMARTer RACE cDNA Amplification Kit (Clontech Laboratories).

Generation of binary vector constructs and plant transformation

The CsRAV1 coding region (CDS) was amplified from chestnut winter stem cDNA using gene-specific primers with attB sites CsRAV1 CDS forward (5′-GGGGACAAGTTTGTACAAAAAAGCAGGCTTCATGGATGGAAGTTGCATAGATG-3′) and CsRAV1 CDS reverse (5′-GGGGACCACTTTGTACAAGAAAGCTGGGTCTTACAAAGCTCCAACTATCCTTTG-3′). A 552-bp-long fragment between the AP2 and B3 domains of poplar RAV1 was amplified from P. alba winter stem cDNA using gene-specific primers with attB sites PtRAV1-RNAi forward (5′-GGGGACAAGTTTGTACAAAAAAGCAGGCTTTGCTGCTCAGAGATTCCGTG-3′) and PtRAV1-RNAi reverse (5′-GGGGACCACTTTGTACAAGAAAGCTGGGTCGAAAGCAAACAATGTCACCAGC-3′). For PCR, PfuUltra Hotstart High-Fidelity DNA Polymerase (Agilent Technologies, La Jolla, CA, USA) was used, and the PCR products were purified and inserted into pDONR222 (Life Technologies, Carlsbad, CA, USA). Insertions in the resulting entry clones were sequence verified. The Gateway cassettes carrying CsRAV1 CDS or the PaRAV1 RNA interference (RNAi) fragment were then transferred into the destination binary vectors pGWB15 (Nakagawa et al., 2007) and pHELLSGATE12 (Wesley et al., 2001), respectively. These constructs were transferred into Agrobacterium tumefaciens strain GV3101/pMP90 (Koncz & Schell, 1986). Hybrid poplar was transformed via an Agrobacterium-mediated protocol described previously (Gallardo et al., 1999). Selection was conducted in kanamycin-containing medium and, once whole plantlets had been regenerated, independent transformed individuals (lines) were screened by reverse transcription-polymerase chain reaction (RT-PCR) for CsRAV1 over-expression or for down-regulated PtaRAV1 and PtaRAV2 expression.

RT-PCR expression analysis

The samples analyzed in these experiments were pools of two or three individuals. Total RNA extraction, single-stranded cDNA synthesis, primer design, real-time PCRs and data analysis were performed as described previously (Ibáñez et al., 2008). The gene-specific primers used were as follows:









Western blot protein expression analysis

Soluble protein extracts were obtained by homogenization of c. 250 mg of ground leaf material in 250 μl of extraction buffer (Rubio et al., 2005), followed by incubation for 30 min on ice, sonication for 10 s in an ultrasonic processor UP100H (Hielscher Ultrasonics, Teltow, Germany; cycle 1, amplitude 100%) and centrifugation for 10 min at 12 000 g and 4°C; 25 μl of supernatants were boiled in sample buffer and run on a 10% sodium dodecylsulfate-polyacrylamide gel electrophoresis (SDS-PAGE) gel. Immunoblotting was conducted as described previously (Berrocal-Lobo et al., 2011) using a 1 : 1000 dilution of anti-hemagglutinin (anti-HA) High Affinity clone 3F1C antibody (Roche Diagnostics, Indianapolis, IN, USA) and a 1 : 5000 dilution of anti-rat horseradish peroxidase (HRP)-conjugated antibody (Sigma-Aldrich, St. Louis, MO, USA).


The 20th internode was cut into 3-mm-long pieces and the pieces were fixed under vacuum in a freshly made up solution of 4% formaldehyde (Sigma-Aldrich) in phosphate-buffered saline (PBS) (Sigma-Aldrich), kept overnight at 4°C and then stored in a solution of 0.1% formaldehyde in PBS at 4°C; 40–50-μm sections were obtained on a Vibratome 1000 Plus (The Vibratome Company, St. Louis, MO, USA) under water. The sections were stained with a 1% solution of toluidine blue O (Panreac Química, Castellar del Vallès, Spain), 2.5% phloroglucinol-HCl (Sigma-Aldrich) or calcofluor white (Sigma-Aldrich). Toluidine blue- and phloroglucinol-stained sections were examined using a Zeiss Axiophot microscope (Carl Zeiss, Oberkochen, Germany) under bright field, and photographed with a Leica DFC300 FX DCC (Leica Microsystems, Wetzlar, Germany). Stacks of calcofluor-stained sections were collected on a Zeiss 710 confocal microscope (Carl Zeiss) under the 405-nm excitation line.

Results and Discussion

Castanea sativa RAV1 is homologous to Arabidopsis TEM genes

To identify the genes specifically expressed during winter dormancy in a deciduous tree, a cDNA subtraction was conducted. Among the clones preferentially expressed in winter, a CsRAV1 fragment was isolated from a winter-enriched SSH library prepared using stem tissue (2-yr-old branch internodes) collected from eight time points every 3 h throughout 1 day, in July and December, from adult European chestnut. The corresponding full-length cDNA clone was obtained using the rapid amplification of cDNA ends-polymerase chain reaction (RACE-PCR) technique on the same tissue. This clone encoded a polypeptide containing the two domains characteristic of the RAV protein family: an AP2 and a B3 DNA-binding domain (Fig. 1a) (Kagaya et al., 1999; Swaminathan et al., 2008). Fig. 1(b) provides the results of a phylogenetic analysis comparing CsRAV1 with Arabidopsis and poplar RAV proteins. The closest relatives to CsRAV1 are PtRAV1 and PtRAV2, and all three group together with the Arabidopsis RAV proteins TEM1 and TEM2/AtRAV2. TEM1 and TEM2 genes encode proteins with functions in the control of flowering time. The correct balance between CONSTANS (CO) and TEM activities determines FT expression to trigger flowering (Castillejo & Pelaz, 2008). Moreover, the module CO/FT is known to control the timing of flowering and seasonal growth cessation in poplar (Böhlenius et al., 2006; Hsu et al., 2011), and so we speculated that CsRAV1, PtRAV1 and PtRAV2 could also play a role in controlling these developmental processes in trees. The CsRAV1 protein, like poplar PtRAV1 and PtRAV2 and Arabidopsis TEM1 and TEM2, features the RLFGV motif in its C-terminal region (Fig. 1a), indicating the possible involvement of chestnut and poplar proteins in repressive functions, as this motif has been identified in several transcriptional repressors (Castillejo & Pelaz, 2008; Ikeda & Ohme-Takagi, 2009). CsRAV1 also contains specific stretches comprising eight histidines (His) or seven serines (Ser). PolyHis-containing proteins seem to be fundamental for developmental processes (Salichs et al., 2009), whereas the presence of a Ser stretch suggests that CsRAV1 function could be regulated by phosphorylation processes. As there is 91.9% identity between PtRAV1 and PtRAV2, we hypothesized that these poplar proteins may be functionally redundant.

Figure 1.

Sequence alignment and phylogenetic analysis of CsRAV1 and related proteins. (a) Amino acid sequence comparisons of CsRAV1 and related proteins in Populus thricocarpa and Arabidopsis thaliana. Lines indicate the conserved domains, AP2 and B3, found in RAV proteins. Asterisks denote the conserved motif found in repressor RAV members. Identical residues in more than half of the proteins are boxed in black, whereas similar residues are boxed in gray. (b) Unrooted neighbor-joining tree built with CsRAV1 and all known RAV proteins in P. thricocarpa and A. thaliana. Reference codes for the proteins used in both the sequence alignment and the phylogram are JN595888 (CsRAV1), B9HWL7 (PtRAV1), B9HIZ2 (PtRAV2), B9GYV2 (PtRAV3), POPTR_0006s20040 (PtRAV4) and B9IMF3 (PtRAV5) for P. thricocarpa, and P82280 (TEM2/AtRAV2), Q9C6M5 (TEM1), Q9ZWM9 (AtRAV1), Q9LS06 (AT3G25730), Q9C6P5 (AT1G50680) and Q9C688 (AT1G51120) for A. thaliana. The sequence alignment and phylogram were both generated using CLUSTALX2 software (Larkin et al., 2007) and plotted using BioEdit (Hall, 1999) and TreeView (Page, 1996) programs, respectively.

CsRAV1 gene circadian expression peaks at noon

Given that Arabidopsis TEM1 and TEM2 genes are circadian-regulated (Castillejo & Pelaz, 2008), we examined the possibility that the CsRAV1 gene was rhythmically expressed in chestnut leaves all day long. Through RT-PCR, we analyzed its expression in leaves collected from 16–24-wk-old seedlings grown under LL conditions at 22°C over a 48-h period. mRNA levels of the CsRAV1 gene showed a strong circadian rhythm (Fig. 2a).We also examined the expression of CsRAV1 in the stems of 16–24-wk-old chestnut seedlings grown under the conditions 22°C and LD (Fig. 2b). RT-PCR analysis revealed an oscillating CsRAV1 expression pattern consistent with that described for leaves of plantlets exposed to LL (Fig. 2a,b). The CsRAV1 transcript peaked at noon, whereas TEM1 and TEM2 peaked at dusk (Castillejo & Pelaz, 2008). This different time of expression in the day suggests that, although the CsRAV1 gene shows closest homology to Arabidopsis TEM genes, in woody plants it may play another role. As occurs with other circadian genes in chestnut (Ramos et al., 2005; Ibáñez et al., 2008), CsRAV1 circadian expression is disrupted when seedlings are transferred to conditions of 4°C and LD (Fig. 2c).

Figure 2.

Quantitative reverse transcription-polymerase chain reaction (RT-PCR) analysis of CsRAV1 gene expression. (a) Circadian expression of CsRAV1 in chestnut plantlet leaves grown under conditions of continuous light (LL) and 22°C. Light gray areas indicate 8-h-long subjective nights. Plotted values and error bars are the fold-change means ± SD recorded in three biological replicates. (b) Circadian expression of CsRAV1 in chestnut plantlet stems grown under conditions of long day (LD) and 22°C. Dark gray area indicates the 8-h-long nights. Plotted values and error bars are the fold-change means ± SD recorded in two biological replicates. (c) Disruption of circadian CsRAV1 expression in chestnut plantlet stems grown under conditions of LD and 4°C. Plotted values and error bars are the fold-change means ± SD recorded in two biological replicates. Seasonal expression of CsRAV1 in stems (d) and buds (e) of adult chestnut plants sampled in the morning in spring (March), summer (June), autumn (September) and winter (December). Plotted values and error bars are the fold-change means ± SD recorded in three technical replicates.

Winter expression of the CsRAV1 gene in chestnut stems and buds

We next examined CsRAV1 expression patterns during the four seasons in 2-yr-old chestnut branches and buds collected in the morning under natural conditions (Fig. 2d,e). In this experiment, in both stems and buds (Fig. 2d,e, respectively), RT-PCR revealed higher CsRAV1 mRNA levels in September compared with March and June, and peaking in December, suggesting important roles for this gene in both tissues during the winter. As CsRAV1 is a circadian gene, such differences can be explained by the disruption of the chestnut circadian clock during winter or in response to low temperature (Ramos et al., 2005; Ibáñez et al., 2008).

CsRAV1 over-expression induces sylleptic branching in hybrid poplar

To gain further insight into CsRAV1 function, the hybrid poplar clone INRA 717-1B4 was transformed to over-express CsRAV1 (3xHA:CsRAV1 OX) under the control of the cauliflower mosaic virus (CaMV) 35S promoter. Through RT-PCR, we analyzed the expression of CsRAV1 and, in Fig. 3(a), the seven independently transformed lines with the highest transgene expression levels are shown. To follow the expression of the protein, we detected the fusion protein by immunoblotting using an anti-HA antibody (Fig. 3b). In all lines, the protein was detected, showing that not only is the transgene expressed, but it is also translated to the expected protein. The overall appearance of the transgenic plants was similar to WT plants when grown under in vitro culture conditions. However, when plants over-expressing CsRAV1 were transferred to soil in the glasshouse, they surprisingly developed sylleptic branches c. 1 month later, when they were c. 25 cm tall. By contrast, sylleptic branching was seldom observed in the control untransformed WT poplars. Branch development was further examined in 3xHA:CsRAV1 OX lines #11, #37, #60 and #88 under controlled environmental conditions. Under these conditions, sylleptic branching started much earlier than in the glasshouse, only 2 wk after transfer, when the poplars were c. 10 cm tall. These lines exhibited different frequencies of sylleptic branch number per node (Fig. 3c). Sylleptic branching in representative defoliated WT and 3xHA:CsRAV1 OX #60 individuals is shown in Fig. 3(d). Thus, over-expression of the CsRAV1 gene induces the early formation of sylleptic branches in poplar transgenics without overwintering. This supports the idea that, in natural conditions, when days become longer and temperatures rise, CsRAV1 accumulated during winter in the lateral buds could trigger the development of the branches (proleptic branching).

Figure 3.

Sylleptic branching in CsRAV1 over-expressor (3xHA:CsRAV1 OX) hybrid poplars. (a) Quantitative reverse transcription-polymerase chain reaction (RT-PCR) analysis of 3xHA:CsRAV1 transgene levels in plantlet leaves from seven lines and the wild-type (WT). Plotted values and error bars are the fold-change means ± SD recorded in three technical replicates. (b) Western blot analysis of 3xHA:CsRAV1 protein levels in the same lines as analyzed in (a). Soluble protein extracts were immunoblotted with an anti-hemagglutinin (anti-HA) antibody (top panel). Similar protein loading and transfer in each lane were verified by staining with red Ponceau S (bottom panel). (c) Sylleptic branch number per node in 8-wk-old 3xHA:CsRAV1 OX (n = 9 per line) and WT (n = 11) plants grown in a growth chamber. Plotted values and error bars are the frequency means ± SE. (d) Sylleptic branching in representative defoliated WT and 3xHA:CsRAV1 OX #60 individuals (scale bar corresponds to 5 cm).

As described in the Introduction, sylleptic branches grow out from the lateral buds during the same growing season in which the buds have formed. This leads to the production of more branches and more leaves, and thus increases the plant’s photosynthetic capacity, and is also thought to assist in increasing the overall growth and biomass of the tree at a young age (Cline & Dong-Il, 2002). Although this is a very important trait, so far little is known about the genes implicated in the regulation of this process. In recent years, a few research groups have addressed this issue by identifying the genomic regions driving the control of this trait in woody species by mapping quantitative trait loci (QTLs). Rae et al. (2009) identified five QTLs responsible for 20% variation between genotypes in terms of biomass yields in poplar, and found that, in a number of cases, the QTL for sylleptic branches co-located with the QTL for stem number, suggesting that these two traits may be under the same genetic control. It should be mentioned that the gene PtRAV1 occurs on linkage group (LG) X (position 12806696) close to poplar biomass loci co-localizing with the marker PMGC2571 (position 12223650) (Rae et al., 2009) and the QTL for sylleptic branching near the marker P2855 (position 11966383) (Ma et al., 2008; Novaes et al., 2009). Further, gene PtRAV2 (position 7464203) appears close to a QTL which is responsible for the sylleptic branching trait, localized on LG VIII in the region of marker ORPM381 (position 7292412) (Rae et al., 2009).

Several articles and reviews have examined the hormonal control of shoot branching in Arabidopsis, discussing the roles of auxin, cytokinin and strigolactone in regulating shoot branching and the genes involved in this process (Aguilar-Martínez et al., 2007; Leyser, 2009; Beveridge & Kyozuka, 2010). Similarly, at the physiological level, the balance between auxin and cytokinin has been implicated in the promotion of sylleptic branching in hybrid poplar (Cline & Dong-Il, 2002; Cline et al., 2006).

Poplars showing altered CsRAV1 expression show normal wood anatomy and normal growth

Given that cytokinin and auxin play important roles in both the induction of sylleptic branching and regulation of wood formation (Nieminen et al., 2008; Nilsson et al., 2008; Dettmer et al., 2009), we analyzed whether changes in CsRAV1 gene expression might alter the wood anatomy of the plants. For this experiment, in addition to 3xHA:CsRAV1 OX plants, PtaRAV1 and PtaRAV2 knock-down (PtaRAV1&2 KD) plants were produced. As hypothesized above, RAV1 and RAV2 poplar genes could have redundant functions, and so we designed an RNAi transgene to down-regulate both endogenous hybrid poplar RAV1 and RAV2 genes. We have established previously that PtaRAV1 and PtaRAV2 are highly induced after 3 h at 4°C in WT (Supporting Information Fig. S1a); therefore, we assessed the reduced endogenous expression of RAV1 and RAV2 under these conditions. We selected seven PtaRAV1&2 KD lines showing expression levels of both genes under 40% WT expression (Fig. S1b). The overall appearance of the knock-down transgenic plants was similar to WT plants when grown under in vitro culture or glasshouse conditions. As for the control untransformed WT poplars, sylleptic branching was also seldom observed. The anatomy of stem cross-sections at the 20th internode was examined in 3-month-old representative individuals 3xHA:CsRAV1 OX #60, PtaRAV1&2 KD #1 and WT plants. Toluidine blue staining showed the overall structure of the stem sections. Xylem and phloem areas were clearly identified at the inner and outer sides of the cambium meristematic cell layers, respectively, and sclerenchyma-supporting tissue appeared blue inside a purple-stained region in which the collenchyma and phloem were localized. Phloroglucinol was used to stain the lignin-containing cell walls of sclerenchyma fibers and xylem vessels. Calcofluor white was used to identify the presence of cellulose in the cell walls, where a higher intensity of staining was seen in the xylem areas closer to the cambium meristem. The presence of Evans blue in the calcofluor-staining solution quenches lignin autofluorescence under the excitation of the 405-nm laser line, thus preventing unspecific cross-talk in the sclerenchyma, which does not appear stained. No anatomical differences in the organization and structure of the stems were observed in sections from transgenic and WT plants (Fig. 4a). Moreover, no evident changes among genotypes in the overall appearance of 8-wk-old glasshouse-grown plants were detected, apart from the presence of sylleptic branches in the over-expressor plants (Fig. 4b). Growth characteristics were further assessed in detail under controlled environmental conditions. WT plants, 3xHA:CsRAV1 OX lines #11, #37, #60 and #88 and PtaRAV1&2 KD lines #1, #8, #22 and #54 were grown in a growth chamber for 8 wk (Fig. 4c). Under these conditions, over-expressor plants showed a slight reduction in height ranging from 4% (#60) to 16% (#11), whereas knock-down plants were not significantly different from WT plants. In general, WT and transgenic lines had the same leaf number on the main stem, except for over-expressor lines #11 (− 6%) and #60 (4%). These differences, in both height and leaf number, translated to a lower height to node ratio (HNR) for the 3xHA:CsRAV1 OX plants. Moreover, only over-expressor line #60 displayed a slight increase in stem diameter (9%) compared with the WT. Although some authors have reported that changes in regulatory protein expression involved in growth and/or development usually cause wood anatomy and growth alterations in transgenic hybrid poplars (Groover et al., 2006; Du et al., 2009; Yordanov et al., 2010; Zawaski et al., 2011), unexpectedly, we detected no such dramatic changes. Longer experiments in field-grown transgenic poplars will be required for an accurate phenotypic assessment of the gene’s effects when suppressed or over-expressed, considering that sylleptic branching in poplar is highly sensitive to the growth environment.

Figure 4.

Phenotypic analysis of over-expressor (3xHA:CsRAV1 OX) and PtaRAV1 and PtaRAV2 knock-down (PtaRAV1&2 KD) lines. (a) Histochemical analysis of cross-sections of stem (20th internode) obtained from 3-month-old 3xHA:CsRAV1 OX #60 (left column), wild-type (WT, middle column) and PtaRAV1&2 KD #1 (right column). After toluidine blue O staining (top row), the different stem tissues were identified: ca, cambium; co, collenchyma; ph, phloem; sc, sclerenchyma; xy, xylem (xy). Phloroglucinol detects lignin in the cell walls of the sclerenchyma and xylem vessels (middle row). Calcofluor stains cellulose (bottom row). (b) Eight-week-old 3xHA:CsRAV1 OX #60 (left), WT (middle) and PtaRAV1&2 KD #1 (right) plants grown in a glasshouse (scale bar corresponds to 5 cm). (c) Growth performance of 8-wk-old 3xHA:CsRAV1 OX (gray bars; n = 9 per line), PtaRAV1&2 KD (red bars; n = 7 per line) and WT (green bars; n = 11) plants grown in a growth chamber. Diameters were measured 10 cm above the ground. The height to node ratio (HNR) was calculated by dividing the height by the node number. Plotted values and error bars are means ± SE. Data points denoted with asterisks were significantly different from the WT according to a t-test at α = 0.5 (*< 0.05, **< 0.01, ***< 0.001).


Chestnut CsRAV1 is a circadian gene, homologous to TEM genes in Arabidopsis, which peaks at noon under vegetative growing conditions and is highly expressed during winter dormancy and in response to low temperature. Its possible role as a transcriptional repressor in modulating hormone responses in tree stem and buds remains to be elucidated. Surprisingly, our results show that, when over-expressed in hybrid poplar, the CsRAV1 protein induces sylleptic branching in the same growing season in which the lateral buds form, whereas no changes in wood anatomy and only minor growth differences are produced. This makes CsRAV1 transgenic plants promising candidates for biomass production. Long-term field trials are needed to confirm the possibility that the manipulation of this gene will lead to the production of trees with a greater biomass.


The authors are grateful to the plant culture facility of the Servicios de Investigación de la Universidad de Málaga, J. C. del Pozo and V. Jorge for technical advice and help. We thank M. Ayllon and M. M. Salmean for allowing us to collect field samples in their estate. This work was supported by grants (AGL-2008-00168 and PIM2010PKB-00702) and a Predoctoral Fellowship from the Spanish Ministerio de Ciencia e Innovación (to P.S.-J.). A.M.-C. holds a postdoctoral fellowship Juan de la Cierva-UPM.