The phytohormone auxin plays a critical role in plant growth and development, and its spatial distribution largely depends on the polar localization of the PIN-FORMED (PIN) auxin efflux carrier family members. In this study, we identify a putative auxin efflux carrier gene in rice, OsPIN3t, which acts in auxin polar transport but is also involved in the drought stress response in rice. We show that OsPIN3t–GFP fusion proteins are localized in plasma membranes, and this subcellular localization changes under 1-N-naphthylphthalamic acid (NPA) treatment. The tissue-specific expression patterns of OsPIN3t were also investigated using a β-glucuronidase (GUS) reporter, which showed that OsPIN3t was mainly expressed in vascular tissue. The GUS activity in OsPIN3tpro::GUS plants increased by NAA treatment and decreased by NPA treatment. Moreover, knockdown of OsPIN3t caused crown root abnormalities in the seedling stage that could be phenocopied by treatment of wild-type plants with NPA, which indicated that OsPIN3t is involved in the control of polar auxin transport. Overexpression of OsPIN3t led to improved drought tolerance, and GUS activity significantly increased when OsPIN3tpro::GUS plants were subjected to 20% polyethylene glycol stress. Taken together, these results suggest that OsPIN3t is involved in auxin transport and the drought stress response, which suggests that a polar auxin transport pathway is involved in the regulation of the response to water stress in plants.
If you can't find a tool you're looking for, please click the link at the top of the page to "Go to old article view". Alternatively, view our Knowledge Base articles for additional help. Your feedback is important to us, so please let us know if you have comments or ideas for improvement.
Plant growth and development are influenced by biotic and abiotic stresses and by endogenous phytohormones. In many cases, plants alter their endogenous hormonal levels to respond to environmental stresses. Auxin plays an essential role in regulating various aspects of plant growth and development, such as embryogenesis, cell division and elongation, vascular tissue differentiation, root patterning, shoot elongation, and embryonic patterning (Leyser, 2001, 2006). Auxin also acts as a crucial signal in responding to abiotic stresses. There are two pathways involved in stress responses: auxin signaling and auxin transport. Several groups have reported that auxin signaling mediates stress responses in plants (Fukaki et al., 1996; Wyatt et al., 2002; Seo et al., 2009; Tognetti et al., 2010), while only a few reports address the auxin transport response under abiotic stresses (Shibasaki et al., 2009; Shen et al., 2010).
Drought stress is one of the major abiotic stresses that restrict plant growth and development. Many phytohormones, such as abscisic acid (Petrasek et al., 2006), salicylic acid (SA) and jasmonic acid (JA), are known to respond to drought stress (Xiong et al., 2002). In early reports, Davenport et al. (1977) demonstrated that the basipetal transport of auxin was reduced in cotyledonary petioles, resulting in earlier leaf loss under conditions of water deficiency. The report revealed that auxin transport inhibitors and water deficits had synergistic effects on leaf abscission. Moreover, osmotic stress caused a striking increase in basipetal polar auxin transport (Sheldrake, 1979). Collectively, these results establish a link between responses to drought stress and polar auxin transport (PAT) in plants, but the molecular and cellular mechanisms involved in PAT regulating these responses under drought stress remain unclear.
Auxin is primarily synthesized in young plant tissues (Ljung et al., 2005), and PAT is mediated by a class of proteins called efflux carriers (Blakeslee et al., 2005; Zazimalova et al., 2010). The regulation of polar transport observed for auxin in plant tissues appears to be unique to this hormone because this type of regulation has not been detected for other signaling molecules (Petrasek and Friml, 2009). Auxin is actively transported through influx and efflux carriers, which are positioned asymmetrically on the plasma membrane. A previous study suggested that at least two efflux carrier protein families exhibited auxin transport activity: the PIN and ABCB (B type ATP-binding cassette super family of transporters) families. The PIN proteins act as the major auxin efflux carriers in plants and show distinct polar subcellular localization, which determines the direction of auxin flux (Okada et al., 1991; Wisniewska et al., 2006). Similar to the PIN proteins, ABCB proteins are involved in cellular auxin efflux in both plant and heterologous systems (Petrasek et al., 2006). The cooperation between the two types of auxin efflux proteins in auxin transport remains unclear (Mravec et al., 2008).
Eight PIN members have been identified in the Arabidopsis genome, designated PIN1 to PIN8. PIN1, PIN2, PIN4, and PIN7 are plasma membrane (PM)-localized and strongly support PAT (Friml et al., 2002a, 2003; Xu et al., 2006). PIN5, PIN6, and PIN8 were shown to localize to the endoplasmic reticulum (ER) and were suggested to play a key role in the intracellular distribution of auxin and the regulation of cellular auxin homeostasis.
Although there are 12 putative auxin efflux carriers homologous to the AtPIN genes in the rice genome based on database searches, only OsPIN1 (Xu et al., 2005) and OsPIN2 (Chen et al., 2012) have been characterized in rice. For example, overexpression of OsPIN1 was shown to partially reverse the inhibitory effect of NPA treatment at the seedling stage, and suppression of OsPIN1 by RNA interference affected adventitious root development. Overexpression of OsPIN2 resulted in the production of more tillers and reduced the inhibitory effect of NPA treatment. These findings indicate that OsPIN1 and OsPIN2 play distinct roles in the normal growth and development of rice and act as auxin efflux carriers.
We describe the isolation and characterization of a putative auxin efflux carrier in rice that shows high amino acid sequence identity to AtPIN3. The 1857-bp full-length cDNA of this gene is designated OsPIN3t. The UniProtKB/Swiss-Prot database indicates that OsPIN3 has two transcriptional isoforms with lengths of 1857 bp [accession number D5A7J0 (618 amino acids)] and 1770 bp [accession number Q0JKX2 (589 amino acids)]. In RT-PCR experiments, only the 1857-bp full-length cDNA was found and isolated. We did not detect a 1770-bp transcript of OsPIN3 in different tissues of the japonica rice varieties Nipponbare and Zhonghua 11 under normal growth or drought-stress conditions. Overexpressing the 1857-bp isoform of this gene under the 35S promoter resulted in visible differences between plants under normal and drought stress conditions compared with controls, including more effective tillers, longer roots, a shorter shoot height and better growth. To provide evidence that OsPIN3t encodes an auxin efflux carrier, we used molecular and cellular approaches to demonstrate that OsPIN3t is an auxin-inducible gene. To observe the effect of NAA and NPA on OsPIN3t expression, we quantified GUS activity in OsPIN3tpro::GUS rice lines following NAA or NPA treatment. We also addressed the possible link between auxin transport and drought stress. We measured the OsPIN3t transcriptional level in overexpression (OE) (OsPIN3t overexpressing) and RNA interference (RNAi) (OsPIN3t knockdown) rice lines under normal and stress conditions and the GUS activity in OsPIN3tpro::GUS rice plants under polyethylene glycol (PEG) stress or NPA treatment. Based on the obtained results, we postulate that OsPIN3t is directly induced in response to auxin and it is also involved in drought stress.
OsPIN3t encodes a PIN3 family member
Bioinformatics studies suggest that OsPIN3t was presumed to be an auxin efflux carrier and a member of the PIN family. The OsPIN3t (AK063976) cDNA and genomic sequences comprise 1857 and 4307 nucleotides, respectively. The UniProtKB/Swiss-Prot database suggested that OsPIN3t has two isoforms that are produced by alternative splicing (Figure S1 in Supporting Information). We performed repeated PCR amplifications of OsPIN3t using different Taq DNA polymerases but did not detect the 1770-bp transcript. In this study, we only characterized the 1857-bp transcript, which was amplified using the primers P1 and P2 (Table S1). The OsPIN3t protein also showed conserved regions in its N- and C-termini with non-conserved regions between them. Hydropathy analyses of the OsPIN3t amino acid sequence showed that the protein contains three characteristic regions including two mem_trans superfamilies, five hydrophobic stretches in the N-terminus, a predominantly hydrophilic core, and a hydrophobic region with five transmembrane segments. These predicted results show similarity to OsPIN1b (Xu et al., 2005), and these domains are found in all PIN and PIN-like proteins. The OsPIN3t protein shares 63% sequence identity with AtPIN3 (At1g70940) and 61% sequence identity with AtPIN4 (At2g01420). In previous reports, OsPIN3t was referred to as OsPIN10a (Wang et al., 2009) or OsPIN3a (Miyashita et al., 2010).
OsPIN3t is responsive to auxin and auxin transport inhibitors
The promoter region of the OsPIN3t gene contains auxin-responsive elements (SURECOREATSULTR11 and CATATGGMSAUR) (Xu et al., 1997; Maruyama-Nakashita et al., 2005). Specifically, there are two GAGACA sequences at positions −573 and −745 and two CATATG sequences at positions −167 and −758 in the promoter region (Figure S2). To test whether OsPIN3t is auxin inducible, several assessments were performed. First, we used quantitative RT-PCR to determine OsPIN3t transcriptional levels with NAA or NPA treatment at the seedling stage. Untreated seedlings showed an expected expression level, and the expression level was increased in NAA-treated plants and decreased in NPA-treated plants (Figure 1a; Table S1). We further analyzed the effect of NPA or NAA on OsPIN3tpro::GUS rice lines, in which the 1846-bp upstream sequence (from ATG) was used to drive the GUS reporter gene. As expected, GUS activity increased strikingly with NAA treatment and decreased significantly with NPA treatment (Figure 1b). The GUS staining in the coleoptile of OsPIN3tpro::GUS rice lines strongly increased after treatment with 50 nm NAA and decreased distinctly after treatment with 10 μm NPA (Figure 1c). Similar results were observed in Arabidopsis by another group (Tsuda et al., 2011). These results provide evidences that OsPIN3t is directly induced in response to auxin.
To assess the functional role of OsPIN3t in auxin transport, a genetic approach was used. Homozygous OE, RNAi and WT seeds were germinated in MS medium or MS medium supplemented with 5 μm NPA. After 5 days, more adventitious crown roots emerged from the WT and OE stems in MS medium, and fewer crown roots emerged from the plants subjected to NPA treatment (Figure 2a,b). We counted the numbers of seminal and crown roots of 1- to 9-day-old seedlings (Figure 2c–f). The results suggested that knockdown of OsPIN3t suppressed the development of seminal and crown roots. To compare the root phenotypes and development of adventitious roots of OE, RNAi and WT plants, we used the T2 generation in the following experiment. Two-day-old seedlings were transferred from MS medium to MS medium supplemented with 5 μm NPA or 0.2 μm NAA. On the seventh day after transfer, the numbers of adventitious roots were counted. As shown in Table 1, overexpression of OsPIN3t improved adventitious root growth, whereas knockdown of OsPIN3t suppressed adventitious root growth. The decrease in adventitious roots observed in RNAi plants was phenocopied by NPA treatment of WT plants. This finding shows that OsPIN3t is involved in root growth and development in rice (Inukai et al., 2005; Liu et al., 2005).
Table 1. Number of roots in wild type (WT), overexpression (OE) and RNA interference (RNAi) seedlings with NAA or 1-N-naphthylphthalamic acid (NPA) treatment
Number of adventitious roots
Means in the same column followed by the same letter are not significantly different (P <0.05, least significant difference test).
aNumbers present the mean ± SE of three separate experiments each performed on a population of 15 < n <20 seedlings per genotype and per treatment.
bSeedlings grew for 9 days under 0.2 μm NAA treatment after a 2-day germination.
cSeedlings grew for 9 days under 0.5 μm NPA treatment after a 2-day germination.
Relative root number of WT, OE and RNAi seedlings with NAA or NPA treatment.
8.75 ± 1.30b
10.25 ± 0.43a,b
6.75 ± 0.83c
9.25 ± 0.83b
11 ± 0.71a
9 ± 0.71b
5.5 ± 0.50c
5.75 ± 0.43c
5.5 ± 1.11c
OsPIN3t is localized to the plasma membrane, and altered localization is observed under NPA treatment
As determined by the TMpred (Hofmann and Stoffel, 1993) program, OsPIN3t contains 10 transmembrane domains. In Arabidopsis, PIN1, PIN2, PIN3, PIN4, PIN6, and PIN7 share a similar molecular structure; these proteins exhibit a long central loop and are primarily localized to the plasma membrane. PIN5 and PIN8 possess a short central loop and localize to the internal cellular and plasma membranes. Therefore, the different PIN proteins show different subcellular localizations, which might be attributable to PIN-specific molecular properties (Ganguly et al., 2010). OsPIN3t is expected to localize to the plasma membrane because it has a molecular structure similar to AtPIN3. To investigate the subcellular localization of OsPIN3t, three constructs, CaMV35S::OsPIN3t-GFP, OsPIN3tpro::OsPIN3t-GFP, and an empty CaMV35S::GFP vector, were individually introduced into EHA105 cells, which were then transformed into tobacco and rice plants. We used confocal microscopy to examine the expression and subcellular localization of OsPIN3t–GFP fusion protein in the epidermal cells of tobacco and root tip cells of rice. Fluorescence microscopy showed that the OsPIN3t–GFP fusion protein was distributed only within the plasma membrane (Figure 3a,b) of tobacco epidermal cells and transgenic rice root cells. This result contrasted with what was observed for the GFP control, which showed fluorescence throughout tobacco epidermal cells and rice root tip cells (Figure 3c). To further examine the expression and subcellular localization of OsPIN3t–GFP fusion protein in root tip cells of rice, plasmolysis of rice root epidermal cells expressing OsPIN3t–GFP fusion proteins was performed. Images also supported that the OsPIN3t localized to the plasma membrane (Figure 3d).
We further analyzed the effect of NPA treatment on the subcellular localization of OsPIN3t–GFP fusion protein in root tip cells of rice. Five-day-old rice root tips were treated with 10 μm NPA; as shown in Figure 3(e), the plasma membrane localization of OsPIN3t-GFP fusion proteins was altered, and the proteins shifted to internal positions within cells. This finding suggests that NPA had an effect on the subcellular localization of OsPIN3t–GFP.
Genetic analysis suggest that OsPIN3t confers tolerance to drought stress
The 1846-bp promoter sequence of OsPIN3t was analyzed using the PLACE database program (http://www.dna.affrc.go.jp/PLACE/) (Higo et al., 1999). The promoter sequence contains many putative stress response-related cis-acting elements, such as ABRE, CRT/DRE, and MYBRS (MYB recognition sites) elements (Figure S3).
To examine the expression patterns of OsPIN3t and observe whether the OsPIN3t promoter is responsive to drought stress in rice, the OsPIN3tpro::GUS rice lines were used in the following experiment. Histochemical analysis of T1 rice plants showed that the strongest signals were found in the coleoptile and vascular bundles of leaves, roots, and shoots, whereas weak signals were found in lamina joints. No GUS staining was observed in transgenic rice seeds or in floral parts, except for the vascular bundles of the lemma and filament (Figure 4a–h). To assess whether OsPIN3t is involved in the drought stress response, GUS staining and GUS activity were monitored in OsPIN3tpro::GUS lines under PEG-induced osmotic stress (Lagerwerff et al., 1961). The GUS staining in coleoptiles and GUS activity in whole transgenic rice plants were greatly increased following exposure to 20% PEG stress (Figure 4i,j).
To further confirm the change in the OsPIN3t transcriptional level in response to drought stress, we analyzed the OsPIN3t expression level in WT, OE and RNAi rice plants under drought stress conditions. We also tested the effect of 20% PEG treatment on seed germination in OsPIN3t OE and RNAi plants. Under normal conditions, there were no significant differences in the seed germination rates or visible phenotypes of transgenic lines or WT plants; however, the OE plants exhibited longer roots (Figure 5a; seedlings in 1/2 MS medium). We next placed the T2 generation of OE, RNAi and WT seeds onto 1/2 MS medium containing 20% PEG (or medium without PEG as a control) and allowed the seeds to germinate and grow into seedlings. There was a noticeable difference between the seedlings grown under the PEG treatment (Figure 5b). The OE seedlings exhibited better shoot growth, longer roots, and a greater number of adventitious roots than the RNAi and WT plants. The growth of RNAi and WT plants was also inhibited by PEG treatment compared with the growth of OE plants. Using OsACTIN1 as a reference gene, the OE lines showed higher OsPIN3t expression levels than the RNAi and WT plants (Figure 5c). These data suggested that PEG-induced osmotic stress induced OsPIN3t expression. Moreover, the expression levels of two drought-responsive genes, OsDREB2A (Figure 5d) and OsAP37, were higher in OE plants than in WT plants under stress conditions. Therefore, the overexpression of OsPIN3t led to up-regulation of OsDREB2A and OsAP37 and increased tolerance of rice plants to drought stress (Kim and Kim, 2009; Oh et al., 2009).
To assay survival rates of rice plants under PEG-induced osmotic stress or water stress conditions, we did the followed experiments. First, 10-day-old OE, RNAi and WT seedlings were transferred to 1/2 MS liquid medium containing 20% PEG for 7 days and allowed to recover under normal conditions for another 7 days. The plants that did not grow after the recovery period were considered to be dead. At the end of the assay, more OE seedlings survived than RNAi and WT seedlings, which were greatly withered (Figure 5e). Second, when OE, RNAi, and WT seeds germinated in soil, similar germination rates and almost the same rice plant seedling growth was shown for all three lines (Figure S4a). Last, the 5-day-old seedlings were subjected to soil without the provision of water for 6 days, and the survival rates of the three lines were very different (Figure S4b,c). Our data suggest that the OsPIN3t gene plays inter-related roles under drought stress conditions.
Phenotypes of rice overexpressing OsPIN3t
To better understand the function of OsPIN3t in rice, quantitative RT-PCR was performed using WT, OE, and RNAi plants. The results showed that OsPIN3t expression was strongly increased in the shoot apex, shoot, root, and panicle in the independent OE lines but was significantly decreased in the RNAi lines compared with the WT lines (Figure S5). Three independent homozygous RNAi lines (RNAi plants R3, R6, and R7) and three homozygous OE lines (OE plants OE3, OE8, and OE9) were used for further analysis. The results indicated that the OsPIN3t transcriptional levels were similar for the individuals of each of the three kinds of transgenic rice line (Figure S6). In the OE plants, several agronomic traits were different from in the WT plants, including significantly increased root lengths and numbers of adventitious roots. In 10 days, the RNAi seedlings exhibited shorter shoots and roots than the WT and OE plants (Figure 6a–c). On the 14th day after germination, the shoots and roots of RNAi seedlings were slightly longer and shorter, respectively, than those of WT plants, and the shoots of the OE plants were the shortest of the three types (Figure 6d–f). These differences were maintained into the mature stage in the rice plants (Figure 6g). The OE lines also presented more effective tillers and a higher average seed set percentage than the WT and RNAi plants (Figure 6h,i).
Although the OsPIN3t OE plants exhibited shorter shoots and slightly lower thousand kernel weights (TKWs), these plants presented more effective tillers and higher average rates of seed setting (Figure 6h,i). The OE and WT plants showed similar seed and panicle phenotypes to WT and transgenic rice (Figure S7), and similar yields per plant under normal conditions; the RNAi plants showed fewer effective tillers, lower seed setting rates, lower TKWs, and a lower yield per plant compared with the WT and OE plants.
To confirm the visible phenotypic differences between OE and RNAi plants resulting from OsPIN3t gene overexpression or knockdown, quantitative RT-PCR was performed using different primers (Table S1). We also repeatedly attempted semi-quantitative and quantitative RT-PCR with different cDNA templates, but we could not detect the 1770-bp isoform of OsPIN3t.
The auxin response plays a crucial role in plant growth and development by forming local concentration gradients. Previous studies have revealed that the distribution of PAT efflux carriers is directional, allowing active flow of auxin through plant tissues (Okada et al., 1991; Wisniewska et al., 2006). Auxin signaling has recently been reported to be mediated by environmental stress responses in plants (Ganguly et al., 2010; Jung and Park, 2011; Rahman, 2012; Tognetti et al., 2012). However, comparatively little insight has been obtained regarding the auxin transport response to drought stress, especially in rice. Our findings provide evidence for the involvement of an auxin transport efflux carrier in the drought stress response.
OsPIN3t encodes a member of the auxin efflux carrier protein family. By identifying the involvement of OsPIN3t in rice growth through PAT, we showed that: (i) overexpression of OsPIN3t improved drought stress tolerance in transgenic rice; (ii) when OsPIN3tpro::GUS plants were subjected to drought stress, GUS expression was responsive to drought stress, and GUS activity significantly increased under NAA treatment and decreased under NPA treatment in the coleoptiles of OsPIN3tpro::GUS rice plants; and (iii) OsPIN3t–GFP fusion proteins localized to the plasma membrane, and this subcellular localization was altered by NPA treatment.
OsPIN3t expression is responsive to auxin or auxin transport inhibitor treatment
There are four OsPIN genes on chromosome 1: Os01g0643300, Os01g0919800, Os01g0715600, and Os01g0802700. The close proximity of the chromosomal locations of the OsPIN genes suggests that the PIN gene family was generated by duplication of chromosomal segments. The OsPIN proteins share conserved regions within their N-termini and C-termini, but their amino acid sequences diverge in central regions. These characteristics imply that different PINs mediate various functions in plant growth and development. In Arabidopsis, the divergent structures of PIN proteins are responsible for their different subcellular localizations and indicate their differential catalytic activities when involved in PAT (Ganguly et al., 2010). The subcellular localizations of rice PIN proteins have not yet been reported. In this study, we showed that OsPIN3t–GFP fusion proteins localize to the plasma membrane, and this localization is altered by NPA treatment (Figure 3e), which suggests that the flow and redistribution of auxin in rice root tips are mediated by OsPIN3t.
Our results indicated that OsPIN3t was mainly expressed in vascular tissue (Figure 4a–h), and OsPIN3t expression was up-regulated by NAA, without de novo synthesis of other proteins. The formation of vascular elements has been proposed to be regulated by PAT (Fukuda, 2004). Previous studies indicated that OsPIN1b and OsPIN1d were expressed in presumptive vascular tissue areas and were associated with vascular bundle development (Wang et al., 2009). Moreover, overexpression of OsPIN3t in rice plant affected seedling growth rate in the roots and shoots. As per the Cholodny–Went hypothesis (1926), differential distribution of auxin in lateral directions due to gravity or light stimulation caused differential growth rates, which could ultimately lead to unbalanced growth or a loss of gravitropism in the shoot or root (Friml et al., 2002b; Ottenschlager et al., 2003; Brunoud et al., 2012). In addition, the finding that GUS activity in OsPIN3tpro::GUS rice lines was promoted by NAA treatment and decreased by NPA treatment also strongly suggests that OsPIN3t is directly induced in response to auxin.
OsPIN3t is involved in root growth at the seedling stage
OsPIN3t participates in rice root development and plays a key role at the vegetative growth stage. Overexpression of OsPIN3t led to the development of longer roots and more adventitious roots in OE plants, whereas knockdown of OsPIN3t in rice plants resulted in slightly shorter adventitious roots. These data are consistent with the root phenotypes of Arabidopsis plants in which AtPIN3 was knocked down (Friml et al., 2002b). Many studies have shown that lateral root formation in Arabidopsis and crown root formation in rice are regulated by auxin signaling and auxin transport (Marchant et al., 2002; Inukai et al., 2005; Liu et al., 2005). In this study, we monitored the seminal and crown root development of OE, RNAi, and WT plants under treatment with auxin or auxin transport inhibitors. The results indicated that the initiation of crown roots was altered by NPA and NAA treatments in OE and RNAi rice plants, which suggests that crown root development is controlled by auxin signaling through PIN proteins (Figure 2c). In Arabidopsis, five PIN genes have been found to be associated with the control of cell division and cell expansion during root outgrowth (Blilou et al., 2005). Although the phenotypes associated with AtPIN and OsPIN3t overexpression cannot be compared because of the differences in root development between rice and Arabidopsis, these similarities suggest that OsPIN3t and AtPIN play similar roles in these two plant species.
In this study, NPA treatment had little effect on root growth in the OE plants but had a greater impact in the RNAi plants. We deduced that higher levels of OsPIN3t expression could partially overcome the inhibition of adventitious root growth and development caused by NPA treatment (Table 1). Taken together, these observations strongly imply that OsPIN3t is involved in root development. However, knockdown of OsPIN1 resulted in significant inhibition of the emergence and development of adventitious roots (Liu et al., 2005; Xu et al., 2005; Chen et al., 2012), and over-expression of OsPIN2 did not result in induction of root differentiation in transgenic rice plants (Chen et al., 2012). These findings indicate that different OsPIN proteins might mediate diverse functions in root growth. It is therefore likely that extensive signaling regulates root development and PAT.
OsPIN3t is involved in drought tolerance
The function of AtPIN2 in basipetal auxin transport was dramatically reduced by cold, and the function of AtPIN3 in the root gravitropic response was also inhibited by cold stress. These data imply that cold stress affects auxin transport, rather than auxin signaling in Arabidopsis (Shibasaki et al., 2009; Shen et al., 2010). Drought stress affects the hormonal balance of a plant during growth and development by reducing cytokine synthesis and activating ABA biosynthesis (Haberer and Kieber, 2002). Although the function of auxin in response to drought stress is relatively uncharacterized, water deficiency was shown in early 1977 to reduce the basipetal transport of auxin in cotton (Davenport et al., 1977). In recent years, genomic and physiological data have revealed that plant hormones perform important actions within a complex network in which there is significant signaling (Nemhauser et al., 2006). The concentration of plant hormones is altered when rice plants experience water deficits. Drought stress treatments significantly reduced concentrations of indole-3-acetic acid in rice grains during the grain filling stage (Yang et al., 2001). As the primary mediators of auxin transport in plants, PIN proteins were presumed to participate in the drought stress response either directly or indirectly. Phototropin 1 (phot1) (Wan et al., 2012) is an Arabidopsis ortholog of the Ser/Thr protein kinase PINOID (PID), which catalyzes PIN phosphorylation, contributes crucially to the regulation of apical-basal PIN polarity (Kleine-Vehn et al., 2009), and can improve drought tolerance at the seedling stage (Galen et al., 2007).
There is currently no evidence for the direct involvement of other OsPIN proteins in the plant response to drought stress, probably because OsPIN family members share responsibilities related to plant growth, and most OsPIN genes are uncharacterized. Previous studies suggested that PEG osmotic stress could induce water deficit stress in a relatively controlled manner (Lagerwerff et al., 1961; Yu et al., 2008; Huang et al., 2009), and our results showed OE OsPIN3t improving drought tolerance (Figures 5 and S4). In the drought stress experiments using transgenic rice plants, knockdown of OsPIN3t to lead to growth inhibition following 20% PEG treatment or water deficiency. In contrast, overexpression of OsPIN3t resulted in better transgenic rice growth. To investigate whether OsPIN3t was involved in the drought stress response, we used OsPIN3tpro::GUS lines to examine GUS staining and activity when these lines were subjected to drought stress. As expected, GUS staining and activity significantly increased under drought stress. To further confirm that OsPIN3t affected the expression of other drought stress genes, we discovered that the rice drought-responsive genes OsDREB2A and OsAP37 were significantly up-regulated in OE plants and slightly up-regulated in RNAi plants (Figure 5c,d). Importantly, studies have shown OsDREB2A expression to be induced by drought and salt stress (Dubouzet et al., 2003) and that the overexpression of OsAP37 increased grain yield under drought stress conditions (Okada et al., 1991; Kim and Kim, 2009; Oh et al., 2009). Moreover, overexpression of OsPIN3t in rice plants improved drought tolerance, which also suggests that OsPIN3t possesses distinct functions compared with OsPIN1 (Xu et al., 2005) and OsPIN2 (Chen et al., 2012). Although we are uncertain about the mechanism of involvement of OsPIN3t in the drought stress response, OsPIN3t mediates auxin transport, plays an important role in rice root growth, and improves drought tolerance.
In conclusion, OsPIN3t plays a key role in rice shoot and root development, is responsible for PAT, and is involved in drought stress responses.
Cloning of OsPIN3t and vector construction
The rice (Oryza sativa L.) cultivar Zhonghua 11 was used for all experiments. The rice plants were grown in culture solution (Yoshida et al., 1976) in a growth room at temperatures of 28/22°C (day/night) with 70% humidity under 14/10 h light and dark cycles. Total RNA was extracted from 7-day-old seedlings, and a 1857-bp open reading frame of OsPIN3t was amplified with primers P1 and P2 via reverse transcription (RT)-PCR and ligated into a T-vector to form pMDPIN3t for sequencing and sub-cloning. For construction of the 35S::OsPIN3t-GFP vector, the full-length OsPIN3t cDNA without the stop codon was amplified from pMDPIN3t. After verification of the correct sequence, the fragment was cloned into the KpnI and XbaI sites in the sense orientation, driven by a CaMV35S promoter in the pCAMBIA2300OCS expression vector. For the OsPIN3tpro::GUS vectors, the 1846-bp DNA sequence upstream of the OsPIN3t ATG (start codon) was amplified using primers P3 and P4. After sequencing, the fragment was digested with HindIII and XbaI, and ligated into the pCAMBIA1300-GUS vector. The 1857-bp coding sequence with the stop codon was amplified and digested with BamHI and XbaI to construct the 35S promoter::OsPIN3t::NOS overexpression vector. To construct the RNAi vector, the 424-bp coding sequence of OsPIN3t was amplified with primers P5 and P6, digested with SacI and SpeI followed by BamHI and KpnI, and subsequently ligated into the pTCK303 vector (Wang et al., 2004). All of the primer sequences used in these experiments are listed in Table S1. All constructs were introduced into the Agrobacterium tumefaciens EHA105 strain to be used for rice transformation according to a published protocol (Hiei et al., 1994; Toki et al., 2006) with some modifications. Transformed calli were selected using hygromycin, and the regenerated transgenic rice plants were grown in a greenhouse under 16/8 h of light/dark, temperatures of 28/25°C (day/night), and 80% relative humidity. The constructs containing GFP were also transformed into tobacco leaves via injection methods.
Plant material and growth conditions
The rice (O. sativa subsp. japonica) cultivar Zhonghua 11 was used as the plant material for the experiments described herein. Rice seedlings were grown in a growth chamber or greenhouse at temperatures of 28/25°C (day/night) with a 16/8 h light/dark cycle under 50% relative humidity (Kawasaki et al., 2001). To assay primary agricultural traits, rice plants were also grown in field under normal conditions. To perform a drought tolerance assay, rice seeds were germinated in 1/2 MS medium containing 20% PEG. For survival analysis, 10-day-old rice seedlings were treated with 20% PEG in 1/2 MS liquid medium for 1 week, and the seedlings were then transferred to new 1/2 MS medium for another week. The rice plants that did not grow were considered dead. For the NAA and NPA treatments, germinated rice seeds were sown in 1/2 MS solid medium containing 0.5 μm NAA or NPA and grown for 5 days.
Total RNA was extracted using TRIzol reagent according to the manufacturer’s instructions. For first-strand cDNA synthesis, 500 ng of total RNA and M-MLV reverse transcriptase (TaKaRa, DRR037A, http://www.takara.com.cn/) was used. Quantitative RT-PCR was performed using the Applied Biosystems 7000 thermocycler (http://www.appliedbiosystems.com/) with SYBR Premix Ex Taq following the reagent specifications and using primers specific for the examined genes (Table S1). OsACTIN1 was used as an internal reference gene for normalization of all data in this experiment.
Histological analysis of OsPIN3t promoter::GUS activity
Histochemical GUS activity analysis was carried out according to Jefferson’s method (Jefferson et al., 1987). Transgenic rice plant samples were incubated in 5-bromo-4-chloro-3-indolyl-b-glucuronic acid buffer in a dark 37°C water bath. After staining, the tissues were rinsed several times with 70% ethanol to remove chlorophyll and surface dyes. The GUS activity was quantified by monitoring the cleavage of the GUS substrate 4-methylumbelliferyl β-d-glucuronide (MUG). The results were averaged from three independent experiments.
Subcellular localization of OsPIN3t
To examine the subcellular localization of OsPIN3t, the CaMV35S::OsPIN3t–GFP, OsPIN3tpro::OsPIN3t–GFP and CaMV35S::GFP vectors were constructed. All vectors were first introduced into the EHA105 bacterial strain, followed by transient transformation into tobacco leaf epidermal cells; after 3 days of infiltration, GFP expression was examined using a confocal laser-scanning microscope (Zeiss LSM510, http://www.zeiss.com/). All plasmids were also transformed into rice calli and transgenic rice plants to further analyze the subcellular localization of OsPIN3t fusion protein. Prior to imaging, roots were stained for 30 s in 10 mg ml−1 propidium iodide (PI) in water. For the plasmolysis experiments, the roots were treated with 0.8 m mannitol for 2 h with shaking and next staining with 10 mg ml−1 PI in 0.8 m mannitol.
We thank Dr Kang Chong of the Institute of Botany, Chinese Academy of Science, for providing the pTCK303 vector. Dr Xiangdong Fu and Dr Yonghong Wang of the Institute of Genetics and Developmental Biology, Chinese Academy of Science, for providing the pCAMBIA2300OCS vector and helping to analyze polar auxin transport. This work was supported by the National Natural Science Foundation of China (30870214), China National Major Scientific Project (2011ZX08009-003, 2009ZX08009-017B), and the Natural Science Foundation of Hebei Province in China (C2010000385).