Introduction

Efficient vegetative propagation methods in plants include: planting stock, spore propagation and tissue culture (George et al. 2008). Plant tissue culture generally refers to the aseptic rapid culture of cells, tissues and organs under certain physical and chemical conditions in vitro (Elhiti and Stasolla 2011; Thorpe 1990). Therefore, tissue culture is a powerful tool in basic and applied researches as well as in commercial application. Plant cells used in tissue culture require the property of cell totipotency (Gautheret 2003; Vasil and Hildebrandt 1965). Cell totipotentiality is the ability of a cell to produce a new genetically identical cell and ultimately tissues, organs and complete individuals through cell division and differentiation processes (García-Gonzáles et al. 2010).

Homeobox (HB) transcription factors represent a large group of transcriptional factors, which is characterized by the presence of a conserved DNA-binding homeodomain (HD) folded into by a short stretch of 60–66 amino acids residues (Gehring et al. 1990, 1994; van der Graaff et al. 2009). HB transcription factors have also been identified in plants, and play multiple roles in plant growth and developmental processes. The plant HB superfamily can be divided into six families based on the phylogenetic relationship and domain structures, including homeodomain associated with a leucine zipper (HD-Zip), plant homeodomain associated with a finger domain (PHD finger), Bell, zinc finger associated with a homeodomain (ZF-HD), WUSCHEL-related homeobox (WOX) and knotted-related homeobox (KNOX) (Ariel et al. 2007; Chan et al. 1998).

WOX family members are involved in many key developmental processes through their activities in promoting cell division and/or preventing premature cell differentiation in plants (van der Graaff et al. 2009; Yadav et al. 2010). Besides the homeodomain, the typically conserved structure of WOX proteins is the WUS-box motif (T-L-X-L-F-P-X-X, where X represents any amino acid) (Haecker et al. 2004), which distinguishes them from other HB protein families. Plant WOX proteins have been classified into three clades based on the phylogenetic analysis: WUS, intermediate and ancient clades (Lin et al. 2013; Nardmann et al. 2009). In Arabidopsis, the members of the WUS clade include WUS and WOX1-7 (Zhang et al. 2010), containing the two-amino-acid motif (T-L) at the beginning of the WUS-box motif (van der Graaff et al. 2009).

So far, the analysis of WOX gene expression and function shows that WOX family members play a wide variety of roles in plant development, such as embryonic patterning (Haecker et al. 2004), stem-cell maintenance (Breuninger et al. 2008) and stress response (Muehe et al. 2014). WOX2 and WOX6 genes were shown to be required for embryonic patterning during embryo development in Arabidopsis (Park et al. 2005; Wu et al. 2007). In Picea abies, WOX2 expression is also tightly associated with somatic embryogenesis (Palovaara and Hakman 2008; 2009). In Arabidopsis and Antirrhinum majus, loss of WUS function promotes stem cells differentiation (Kieffer et al. 2006; Mayer et al. 1998). Root columella stem cells, which normally produce the gravity-sensing root cap cells, also undergo differentiation in the Arabidopsis wox5 loss-of-function mutant (Kamiya et al. 2003). Interestingly, WOX6 is also found to be involved in the response to cold stress in Arabidopsis (Zhu et al. 2004). However, the role of WOX genes in organogenesis during regeneration in Rosa canina remains uncharacterized.

In our previous work, we established a highly efficient regeneration system through organogenesis and somatic embryogenesis from cultured leaf explants of R. canina (Tian et al. 2008). Auxin was identified as a trigger of rhizoid formation, while cytokinin only modulates the process (Gao et al. 2013). However, the molecular mechanism of auxin-induced rhizoid formation is still unclear. Previous studies indicate that WOX family play an essential role in stem cell maintenance and lateral root formation (Breuninger et al. 2008; Liu et al. 2014; Tian et al. 2014), so they are potential key regulators during rhizoid formation. Here, we identified RcWOX1, which encoded a member of WOX family and was up-regulated rapidly and significantly by auxin. We further showed the lateral root initiation rate were remarkably increased in RcWOX1 overexpression transgenic lines.

Results

Role of auxin in rhizoid formation

A new and efficient plant regeneration system based on the root-like organs named rhizoids has been achieved in several rose cultivars. Tian demonstrated that 2,4-D (1.5–3 µg/ml) is necessary for rhizoid formation (Tian et al. 2008). In order to study the role of auxin in detail during rhizoid initiation in R. canina, leaf explants were incubated on medium containing different concentrations of 2,4-D or NAA. We found that rhizoids can be induced by 0.1 µg/ml 2,4-D, while only callus could be distinguished at lower concentrations. However, rhizoids were not induced in the presence of 0.1 µg/ml NAA (Fig. S1). Further treatment of explants with different concentrations of NAA to identify the NAA induction threshold concentration revealed that 0.5 µg/ml NAA is sufficient to initiate rhizoid formation (Fig. 1a, b). Moreover, statistical analysis showed that rhizoid outputs induced by 2,4-D were higher than those by NAA at the same doses. Rhizoid induction with 1.5 µg/ml 2,4-D was 37.5 % higher than that with equivalent doses of NAA (Fig. 1c). The results suggest that a certain auxin concentration is required for rhizoid induction.

Fig. 1
figure 1

Effect of auxin on callus-rhizoid induction from leaf explants in R. canina. a Callus-rhizoid formation when grown on medium containing different concentrations of 2,4-D. b Callus-rhizoid formation when grown on medium containing different concentrations of NAA. Scale bar 5 mm in (a, b). c Statistical analysis of the number of rhizoids after 4 weeks of 2,4-D and NAA treatments. Rhizoid numbers were surveyed from 15 calluses. Values are presented as mean ± SD

Isolation of RcWOX1

Previous study showed that the formation of rhizoids appears to be the result of callus cell dedifferentiation. A number of WOX family transcription factors are known critical regulators of apical meristem cell division and differentiation (Haecker et al. 2004). WOX5 has been demonstrated to be mainly expressed in the root and plays an essential role in root apical meristem maintenance. To identify the potential WOX genes involved in rhizoid formation, we aligned the WOX5 gene from different species, including Arabidopsis thaliana, Theobroma cacao, Pisum sativum and Raphanus sativus and designed the degenerate primers in the conserved region (Fig. S2) and cloned the full-length cDNA of WOX gene from R. canina using RACE method. The full-length cDNA (accession number KJ699116) of RcWOX1 is 692 bp. The coding sequence encodes 186 amino acids containing a typical HD with 65 residues, a WUS box motif (T-L-Q-L-F–P-I–N) and an ERF-associated amphiphilic repression (EAR) motif (L-D-L-R-L) (Fig. 2a). Phylogenetic and domain structure comparisons of RcWOX1 with Arabidopsis WOX protein members further revealed that RcWOX1 belongs to the WUS clade of the WOX protein family, with higher sequence similarity to WOX5 than other members of the WOX family in Arabidopsis (Fig. 2a). RcWOX1 has 66.13 % sequence identity to TcWOX5, 60.57 % to PsWOX5 and 56.04 % to AtWOX5 (Fig. 2b).

Fig. 2
figure 2

Phylogenetic analysis of RcWOX1. a Phylogenetic relationships and schematic domain structures of RcWOX1 and Arabidopsis WOX proteins. Purple block represents HD domain; Cyan block represents WUS-box; Orange block represents EAR-like domain. b Sequence alignment of R. canina (RcWOX1) with its orthologues in Cicer arietinum (CaWOX5), Fragaria vesca (FvWOX5), Glycine max (GmWOX5), P. sativum (PsWOX5), Solanum lycopersicum (SlWOX5), T. cacao (TcWOX5), Vitis vinifera (VvWOX5) and A. thaliana (AtWOX5). The positions of the conserved HD, WUS-box and EAR-like domain are indicated by black lines

Expression dynamics of RcWOX1 during rhizoid induction

Given the high sequence similarity of RcWOX1 to WOX5 from other species, we further investigated whether RcWOX1 is involved in rhizoid formation. Rhizoid formation occurred one month after callus initiation upon cultivation on medium containing auxin. We first examined the RcWOX1 expression dynamics within the first month after callus initiation. As shown in Fig. 3a, the expression level of RcWOX1 gradually increased during the first 3 weeks after induction as compared with the control, and then slowly decreased (Fig. 3a), suggesting significant correlation with rhizoid formation. To determine whether auxin affects RcWOX1 expression within a narrower time scale, 2-week callus were incubated in MS liquid medium containing 0.5 μg/ml 2,4-D for 0, 4, 8, 16 and 24 h. Our results showed that RcWOX1 expression was significantly increased from 4 h after auxin treatment compared with the control (Fig. 3b). Given that auxin is required for rhizoid induction and the RcWOX1 level is correlated with the rhizoid formation process, we speculate that auxin induces rhizoid formation of R. canina at least partially by regulating RcWOX1 expression.

Fig. 3
figure 3

Quantitative real-time PCR showing RcWOX1 expression. a qRT-PCR analysis of RcWOX1 expression during rhizoid formation 0, 1, 2, 3 and 4 weeks after callus initiation. b qRT-PCR analysis of RcWOX1 expression in callus treated with 0.5 μg/ml 2,4-D for various time-periods

Root-specific expressed RcWOX1 is localized to nuclear

The efficient rose regeneration system is mainly divided into five stages: leaf explants, callus, rhizoids, protocorm-like bodies (PLBs) and regenerated seedlings. For analysis of the tissue expression patterns of RcWOX1, RT-PCR was performed to examine levels of the transcript in root, stem, leaf, seedlings, callus and rhizoid. Highest RcWOX1 expression was evident in roots, rhizoids, callus and seedlings, but not the other tissues tested (Fig. 4a).

Fig. 4
figure 4

Tissue expression patterns and subcellular localization of RcWOX1. a Analysis of the RcWOX1 transcript in different organs using RT-PCR. Top target sequence was amplified by 25 cycles of semi-quantitative PCR. Center target sequence was amplified by 30 cycles of semi-quantitative PCR. Bottom TUBULIN (TUB) transcripts were detected as controls. (bd) Histochemical localization of GUS activity in the 8-day-old RcWOX1pro::GUS transgenic Arabidopsis seedlings, primary root tip (b), root hair zone (c) and lateral root (d). e Subcellular localization of RcWOX1 in Arabidopsis root tip cells. Scale bar 50 µm in (b, e) and 500 µm in (c, d)

To further investigate the expression patterns of RcWOX1 in plants, we cloned a 1,257 bp genomic fragment upstream of the gene and analyzed its cis-elements using PLACE (Higo et al. 1999). Among these, an auxin- and several cytokinin-responsive elements were located at positions −1,230/−1,225, −656/−652, −611/−607, −206/−202 and −147/−143. We additionally identified a HD-Zip (TAATGATTA) motif at positions −150/−142 (Fig. S3). Next, a RcWOX1 promoter::β-glucuronidase (GUS) recombinant plasmid was constructed and transformed into Arabidopsis plants via an Agrobacterium-mediated method. As shown in Fig. S4, overall GUS staining revealed specific expression of RcWOX1 in roots. Further detailed analysis disclosed strong GUS activity in the primary root tip (Fig. 4b), root hair zone (Fig. 4c) and lateral root tip (Fig. 4d) in transgenic plants, which was consistent with RT-PCR findings in Fig. 4a. To ascertain the subcellular localization of RcWOX1, the ORF was translationally fused to the C-terminus of the GFP reporter under control of the CaMV 35S promoter and transformed into Arabidopsis. The GFP signal was localized in the nuclei of Arabidopsis root cells, clearly suggesting that RcWOX1 is a nuclear transcription factor (Fig. 4e).

Effect of ectopic expression of 35S::RcWOX1 on Arabidopsis root development

Gene expression data suggested that RcWOX1 might be involved in root organogenesis. Accordingly, we further analyzed the phenotypes of RcWOX1 overexpression in transgenic Arabidopsis plants. Two lines of transgenic plants with high RcWOX1 expression were selected using RT-PCR for subsequent phenotypic analysis (Fig. 5a). Five days after germination, we observed no obvious differences between RcWOX1-overexpressing and wild-type lines (Fig. 5b). However, ten days after germination, the lateral root number of the OW7 and OW12 transgenic plants increased by 46 and 39.3 % as compared with the wild type, respectively (Fig. 5c; Fig. S5), indicating that RcWOX1 significantly promotes organogenesis of lateral roots.

Fig. 5
figure 5

Overexpression of RcWOX1 in Arabidopsis. a Analysis of RcWOX1 transcripts using RT-PCR in six transgenic lines. TUBULIN (TUB) was used as an internal control. b Comparison of 5-day-old seedlings of the wild type (WT) with the transgenic plants (OW7 and OW12). c Comparison of 10-day-old seedlings of the wild type (WT) with the transgenic plants (OW7 and OW12). Scale bar 5 mm in (b, c)

Expression of auxin-related genes in RcWOX1 transgenic plants

The analysis of the promoter cis-elements of RcWOX1 revealed that this gene may involve in auxin signal transduction. Moreover, PIN1 expression was reduced in the embryos of the wox8-1wox9-1 double mutant (Breuninger et al. 2008). To check whether auxin signal transduction is affected in the RcWOX1 transgenic Arabidopsis plants, we analyzed the expression of the auxin efflux carrier genes, PIN1, 2, 3, 4 and 7, which is involved in root development (Blilou et al. 2005), in 1-week-old seedlings using qRT-PCR. We found that PIN1 and PIN7 genes were slightly up-regulated in 35S::RcWOX1 transgenic seedlings as compared with the control group (Fig. 6). Based on these results, we propose that RcWOX1 participates in lateral root organogenesis via regulating the auxin-signaling pathway.

Fig. 6
figure 6

Expression analysis of auxin efflux carrier genes in the RcWOX1-OX Arabidopsis. qRT-PCR analysis of PIN1, 2, 3, 4 and 7 genes in the wild-type and transgenic plants. Ubiquitin (UBQ) was used as an internal control. Asterisks indicate statistically significant differences (Student’s t test; P < 0.01) between wide type and RcWOX1-OX. Values are presented as mean ± SD

Discussion

It has been established that WOX protein family is one member of plant HB transcription factor superfamily (van der Graaff et al. 2009) and is involved in various aspects of plant development. In the present study, we isolated the RcWOX1 gene from R. canina and characterized its expression patterns during rhizoid formation. We also found that overexpression of RcWOX1 had a significant effect on the formation of lateral root. Collectively, our results suggested that RcWOX1 plays an essential role in rhizoid formation, a root-like structure, from leaf explants of R. canina.

Auxin is one of the earliest discovered plant growth promoting hormones (Davies 2010) that mainly regulates embryogenesis (Müller and Sheen 2008; Nolan et al. 2003), vascular differentiation (Aloni et al. 2006), organogenesis (Cheng et al. 2007; Pernisová et al. 2009) and tropic growth (Esmon et al. 2006; Takahashi et al. 2009). In R. canina, threshold concentrations of 2,4-D and NAA required to induce rhizoid formation were determined as 0.1 and 0.5 µg/ml, respectively, in the present study. Below the threshold, only callus could be distinguished and any organogenic response was not detected, while both callus and rhizoids were induced above these concentrations. This finding is consistent with data from a previous study on Arabidopsis which reported that induction of new root-like organs requires a specific concentration of auxin (Pernisová et al. 2009). Moreover, at the same dose levels, outputs induced by 2,4-D were significantly higher than that of NAA. Taken together, these results suggest that auxin is essential for the formation of callus and rhizoids, which is a prerequisite for the development of PLB.

A previous study showed that WOX5 is specifically expressed in quiescent center cells and loss of WOX5 function leads to terminal differentiation in distal stem cells, supporting an essential role of this gene in regulation of root stem cell differentiation and lateral root initiation (Sarkar et al. 2007). Our sequence analysis showed that RcWOX1 has high sequence similarity with WOX5. Moreover, qRT-PCR analysis revealed that RcWOX1 is expressed predominantly at the rhizoid initiation stage, with high transcript abundance at the beginning of induction followed by a decrease during auxin-induced rhizoid formation. The induction effect of auxin on RcWOX1 expression was also evident within a shorter time frame of 0–24 h of treatment. These data indicate that auxin-induced expression of RcWOX1 is a possible prerequisite for initiation of rhizoid formation.

In Arabidopsis, WOX genes are involved in the regulation of PIN1 and DR5rev:GFP expression during embryo development, indicating that WOX transcriptional machinery is essential for the establishment of localized auxin response (Breuninger et al. 2008). Thus, we speculated that RcWOX1 plays a role in the regulation of auxin signal transduction. Our results showed altered expression of polar auxin transport genes, PIN1 and PIN7, in RcWOX1 overexpression transgenic plants. Moreover, analysis of the effect of RcWOX1 overexpression on Arabidopsis lateral root formation revealed that lateral root densities are significantly increased in RcWOX1-overexpressing lines as compared with the control. Taken together, these data presented here suggested that RcWOX1 is likely an important player in auxin-induced rhizoid formation in R. canina.

Rose is one of the most popular cut flowers,which has broad markets and high sales (Zlesak 2006). Due to dependence on the genotype, induction of somatic embryogenesis and regeneration in vitro are difficult, which seriously restricts commercial production and the international trade of roses. Therefore, establishment of an efficient regeneration system is a prerequisite for rose molecular breeding. Studies on the mechanism of regeneration will provide a theoretical basis for rapid propagation of rose in vitro. In this study, we found that RcWOX1 appears to play a pivotal role in auxin-induced rhizoid formation, which is a critical step for R. canina regeneration through PLB. Our results reveal new insights into the molecular mechanism of R. canina regeneration system and provide a theoretical basis for efficient regeneration in vitro, which might have potential application in rapid propagation and engineering of rose.

Materials and methods

Plant materials

Stem segments from 1-year-old shoots of R. canina were harvested from the field and surface-sterilized. Single node segments were germinated in shoot multiplication medium (SMM) according to the protocol of Tian et al. (2008). Leaf explants from 6-week-old plantlets were used for all analyses in this study.

Auxin treatments

For long-term auxin treatment, leaf explants were cultured on rhizoid formation medium containing 0.01, 0.05, 0.1, 0.5, 1.0 and 1.5 μg/ml 2,4-D or NAA for 4 weeks. For short-term auxin treatment, callus cultured in the dark for 2 weeks were incubated in half-strength MS liquid medium supplemented with 0.5 μg/ml 2,4-D for 0, 4, 8, 16 and 24 h.

Cloning of the RcWOX1 gene

Total RNA of rhizoids was extracted using the RN09-EASY spin Kit (Biomed, China). cDNA was synthesized using Superscript Ш reverse transcriptase (Invitrogen, China). Degenerate primers P_for (5′-AAGTGYGGKCGDTGGAAYCCAAC-3′) and P_rev (5′-CCARTAGAAVACRTTCTTGCTC-3′) were designed for RcWOX1 cloning based on the conserved regions of WOX5 genes from A. thaliana, T. cacao, P. sativum and R. sativus. The amplification product was used to design 3′ RACE primers 3′-race_for (5′-TACTGAGCAGGTCAAAGTTTTGACG-3′) and 3′-race_rev (5′-GGCCACGCGTCGACTAGTAC-3′). The 5′ fragment was obtained using the Genome Walking Kit (Takara, Japan). The full-length of RcWOX1 was amplified using PrimeSTAR HS DNA Polymerase (Takara, Japan) and cloned into the pEASY-Blunt Simple Cloning Vector (Trans, China) for sequencing. Sequence alignments were generated using ClustalW and ESPript program. Phylogenetic analysis was performed using the MEGA4 program. Accession numbers for the public sequences used in the phylogenetic analysis are listed in Table S1.

Quantitative real-time PCR

Total RNAs were extracted and 2 μl cDNAs were used for quantitative real-time PCR analysis on an ABI 7500 Real-Time PCR System (Applied Biosystems, USA) using SYBR® Premix Ex Taq™ II (Sangon Biotech, China). The PCR procedure was as follows: denature at 95 °C for 30 s, followed by 40 cycles of 5 s at 95 °C, 30 s at 58 °C. The primers used for this quantitative real-time PCR were RcWOX1_for (5′-ACTTACACTGTTGGAGCAGAAATGG-3′) and RcWOX1_rev (5′-ACAAGTTTCATTTCTCTCCCCCTAC-3′); PIN1_for (5′-TACTCCGAGACCTTCCAACTACG-3′) and PIN1_rev (5′-TCCACCGCCACCACTTCC-3′); PIN2_for (5′-GGCGAAGAAAGCAGGAAGA-3′) and PIN2_rev (5′-GGTGGGTACGACGGAACA-3′); PIN3_for (5′-GAGGGAGAAGGAAGAAAGGGAAC-3′) and PIN3_rev (5′-CTTGGCTTGTAATGTTGGCATCAG-3′); PIN4_for (5′-CTGAACGATGGCTATACGGAGAAG-3′) and PIN4_rev (5′-CTTGGCTTGTAATGTTGGCATCAG-3′); PIN7_for (5′-CGGCTGATATTGATAATGGTGTGG-3′) and PIN7_rev (5′-GCAATGCAGCTTGAACAATGG-3′); 18S rRNA_for (5′-ACTTACACTGTTGGAGCAGAAATGG-3′) and 18S rRNA_ rev (5′- ACAAGTTTCATTTCTCTCCCCCTAC-3′); UBQ10_for (5′-CACACTCCACTTGGTCTTGCGT-3′) and UBQ10_rev (5′-TGGTCTTTCCGGTGAGAGTCTTCA-3′) were used as internal controls for normalization. The expression levels were analyzed using the relative 2−ΔΔCT method. All experiments were performed on three biological replicates.

Promoter analysis

Genomic DNA from leaf explants was isolated using the Plant Genomic DNA Purification Kit (Tiangen Biotech). The promoter sequence of RcWOX1 was obtained using specific primers SP2 (5′-CATGATCTCGACGTATGAACTCCTTG-3′) and SP3 (5′-ACCACAAGGACATGGGAGTAAGCC-3′) coupled with primers (AP1–AP4) provided by the Genome Walking Kit. The promoter fragment was amplified using the specific primers HindIII_for (5′-GTAAAGCTTAGCCACATGACATAACAAACTCC-3′) and XbaI_rev (5′-CGCTCTAGATTCAGCTTCAAATGCCTAGAAAG-3′), and inserted into pBI121-GUS to produce the RcWOX1pro::GUS expression vector, which was transformed into Arabidopsis using the floral-dip method.

For RcWOX1pro::GUS activity analysis, the 6-day-old seedlings were incubated with GUS staining solution at 37 °C overnight and decolorized using 70 % ethanol. The GUS signal was visualized under a Zeiss LSM 510 META microscope.

Subcellular localization

The open reading frame (ORF) of RcWOX1 was amplified with primers PstI_for (5′-GTAGGATCCATGGATGAAAATATGTCAGGGTT-3′) and BamHI_rev (5′-CGCCTGCAGTCAAAGGAAGCTTAAGCGTAGATC-3′). After digestion with BamHI and PstI, the fragment was inserted into pCambia 2300-GFP to obtain the pCambia 2300-GFP-RcWOX1 expression vector for Arabidopsis transformation. GFP fluorescence was visualized using confocal microscopy. Images were processed using ImageJ software and Adobe Photoshop CS3.

Phenotypic analysis

For lateral root density measurements, Arabidopsis seedlings were grown vertically on 1/2 MS medium. Seedlings were photographed 5 and 10 days after germination, respectively. For 10-day-old seedlings, the number of lateral roots was counted and subjected to statistical analysis using Origin8.0 software.