Blue-light-induced PIN3 polarization for root negative phototropic response in Arabidopsis

Authors


For correspondence (e-mail yingtlu@whu.edu.cn).

Summary

Root negative phototropism is an important response in plants. Although blue light is known to mediate this response, the cellular and molecular mechanisms underlying root negative phototropism remain unclear. Here, we report that the auxin efflux carrier PIN-FORMED (PIN) 3 is involved in asymmetric auxin distribution and root negative phototropism. Unilateral blue-light illumination polarized PIN3 to the outer lateral membrane of columella cells at the illuminated root side, and increased auxin activity at the illuminated side of roots, where auxin promotes growth and causes roots bending away from the light source. Furthermore, root negative phototropic response and blue-light-induced PIN3 polarization were modulated by a brefeldin A-sensitive, GNOM-dependent, trafficking pathway and by phot1-regulated PINOID (PID)/PROTEIN PHOSPHATASE 2A (PP2A) activity. Our results indicate that blue-light-induced PIN3 polarization is needed for asymmetric auxin distribution during root negative phototropic response.

Introduction

Phototropism is one of the mechanisms by which plants adapt their growth and development to changing environments (Estelle, 1996; Holland et al., 2009). Plants can maximize light absorbance throughout their lifecycle by means of hypocotyl phototropism. The mechanisms underlying phototropism, especially the role of auxin in this process, have been extensively investigated in the past decade (Friml et al., 2002; Noh et al., 2003; Blakeslee et al., 2004; Esmon et al., 2006; Stone et al., 2008; Christie et al., 2011; Ding et al., 2011; Haga and Sakai, 2012).

Plants respond to differences in light quality, intensity, and direction by altering their growth and development, and processes such as phototropism, stomatal opening, and chloroplast avoidance movement are particularly sensitive to blue light (Christie, 2007). Hypocotyl phototropism includes three major steps: light perception, the asymmetric distribution of auxin, and the expression of downstream signal factors. Phototropins, a family of blue-light receptors, have two members in Arabidopsis, PHOT1 and PHOT2 (Sakamoto and Briggs, 2002). PHOT1 plays the major role in both hypocotyl and root phototropism (Liscum and Briggs, 1995; Huala et al., 1997; Motchoulski and Liscum, 1999; Sakai et al., 2000; Boccalandro et al., 2008; Ding et al., 2011; Wan et al., 2012), while PHOT2 functions only in response to the high fluence rates of blue light (Sakai et al., 2001). Interestingly, phot1 can phosphorylate the ABCB19 auxin transporter in response to blue-light-induced hypocotyl phototropism (Christie et al., 2011). The blue-light-dependent autophosphorylation of the N-terminus is required for the phototropic response (Christie et al., 1998; Inoue et al., 2008b). It has recently been demonstrated that phototropins are trafficked through the endosomal recycling pathway (Kong et al., 2006; Kaiserli et al., 2009), as are other plasma membrane-associated proteins, including PIN1 (Geldner et al., 2001).

The asymmetric auxin distribution during hypocotyl phototropism is mediated by auxin transporters of the AUXIN RESISTANT/LIKE AUXIN RESISTANT, P-GLYCOPROTEIN, and PIN families (Swarup et al., 2005; Geisler and Murphy, 2006; Petrasek et al., 2006; Yang et al., 2006). The PID-dependent phosphorylation pathway mediates PIN3 polarization, which directs auxin flow laterally towards the shaded side of hypocotyls and causes the hypocotyls to bend towards the light (Ding et al., 2011). PP2A, which act antagonistically with PID, also has been involved in the regulation of PINs polarization (Michniewicz et al., 2007).

The shift in PINs localization also requires a BFA-sensitive trafficking pathway. GNOM, a member of the ARF-GEF (exchange factors for ARF-GTPases) family, is needed for PIN1 recycling from endosomes to the plasma membrane (Geldner et al., 2003). BFA, a fungal toxin that inhibits GNOM, causes PIN1 to accumulate in endosomes called BFA compartments (Geldner et al., 2001; Dhonukshe et al., 2007; Kleine-Vehn et al., 2009). Notably, it has been shown that the BFA-sensitive, GNOM-dependent, PIN3 trafficking pathway is involved in hypocotyl phototropism, hypocotyl gravitropism and root gravitropism (Kleine-Vehn et al., 2010; Ding et al., 2011; Rakusova et al., 2011).

While the mechanisms underlying asymmetric auxin distribution in hypocotyl phototropic response have been extensively investigated, knowledge of root negative phototropism is limited. Root negative phototropism was first studied by Darwin (1880), well over a century ago. Whereas plant shoots bend towards the light to maximize light absorption, roots bend away from light to avoid the damage of light and other stressful stimuli in the upper layers of soil, and to facilitate water and nutrient absorption from the soil (Esmon et al., 2005; Monshausen and Gilroy, 2009). Progress has been made in understanding the role of blue light in the regulation of this critical process (Boonsirichai et al., 2002; Whippo and Hangarter, 2006; Holland et al., 2009; Takahashi et al., 2009). The blue-light receptor phot1 has been shown to be involved in blue-light perception, both during hypocotyl phototropism and root negative phototropism (Liscum and Briggs, 1995; Christie et al., 1998, 2011; Sakai et al., 2000, 2001; Boccalandro et al., 2008; Ding et al., 2011; Wan et al., 2012). RPT2, NPH3, and PKS1, factors that function downstream of PHOT1, are also known to be involved in root negative phototropism (Sakai et al., 2000; Boccalandro et al., 2008; Wan et al., 2012). In addition, starch appears to play a role in root negative phototropism, as evidenced by loss-of-function analyses of PGM and ADG, which are implicated in starch biosynthesis (Vitha et al., 2000). Recently, it has been reported that PIN2-based polar auxin transport is involved in root negative phototropism (Wan et al., 2012). However, the molecular and cellular mechanisms that underlie root negative phototropism, especially the role of auxin, remain to be deciphered.

In this study, we observed that the blue-light-regulated asymmetric distribution of auxin is essential for root negative phototropism. Our results suggested that this asymmetric auxin distribution is regulated by blue-light-induced PIN3 polarization. Furthermore, PIN3 polarization was found to be modulated by a BFA-sensitive, GNOM-dependent, trafficking pathway and the activity of PID/PP2A. The asymmetric distribution of PIN3 on the outer lateral membrane of columella cells is needed for auxin redistribution during blue-light-induced root negative phototropism.

Results

Asymmetric auxin distribution is modulated in blue-light-induced root negative phototropism

As both hypocotyl and root phototropic responses are mediated by blue light, blue-light receptors are expected to be involved in these processes (Huala et al., 1997; Motchoulski and Liscum, 1999; Sakai et al., 2000, 2001; Boccalandro et al., 2008; Wan et al., 2012). Cryptochrome and phototropin are blue-light receptor families in Arabidopsis. To investigate the role of these two families in the root negative phototropic response, we analyzed loss-of-function mutants of these blue-light receptor families. Two-day-old etiolated seedlings were illuminated with unilateral blue light for 2 days, and the bending angles of the roots away from the vertical direction were measured, as previously described (Figure 1a) (Sakai et al., 2000). Under low fluence rates (10 mol m−2 sec−1), the roots of cry1 and phot2 mutants exhibited normal root negative phototropic responses (Figure 1a–c), whereas the phot1 mutant showed strong defects in root phototropism (Figure 1a–d). This result suggests that blue-light-induced root negative phototropism is dependent on phot1 (Sakai et al., 2000; Boccalandro et al., 2008; Wan et al., 2012).

Figure 1.

Root negative phototropic response of mutant lines. Images of 2-day-old etiolated seedlings of the wild-type (WT) and cry1, phot2, phot1, and pin3-4 mutants and of 2-day-old WT etiolated seedlings treated with NPA (1 and 5 μm), grown on vertical plates, and then exposed to unilateral blue light (10 μmol m−2 sec−1) for another 48 h. The arrows indicate the direction of blue light (blue) and gravity (black). Bar = 1 cm. Graphical representation of the phenotypes shown in (a). Wild type, n = 138; cry1, n = 91; phot2, n = 88; phot1, n = 69; pin3-4, n = 150; and NPA-treated WT (1 and 5 μm), n = 73 and 86. The percentage of seedlings belonging to each of four different categories, i.e., 0–20°, 20–35°, 35–50°, and >50° relative to the vertical direction. (c) Root bending angles of wild-type, cry1, phot2, phot1, pin3-4 and NPA-treated WT (1 and 5 μm). The bending angles of the roots away from the vertical direction were measured after 48 h unilateral blue-light illumination and average curvatures were calculated. Values are the average of three biological replicates (n > 20 per time point on each replicate). Error bars represent standard error (SE) and indicate significant difference at **P < 0.01, ***P < 0.001, as determined by Student's t-test. (d) Root bending kinetics of WT, phot1 and pin3-4 seedlings. Root curvatures were measured every 6 h under unilateral blue-light illumination and average curvatures were calculated. Values are the average of three biological replicates (n > 10 per time point on each replicate). Error bars represent SE and indicate significant difference at *P < 0.05, **P < 0.01, ***P < 0.001, as determined by Student's t-test.

The asymmetric distribution of auxin is a key factor in hypocotyl phototropism (Friml et al., 2002; Christie et al., 2011; Ding et al., 2011). Thus, we sought to examine whether auxin is asymmetrically redistributed during blue-light-induced root negative phototropism. We first plotted a time course of the effect of blue light on DR5 activity using the auxin responsive DR5REV::GFP marker line, which reliably reveals the pattern of auxin distribution in roots (Friml et al., 2003), and found that the ratio of DR5 activity at the illuminated side versus at the shaded side of root was elevated by up to 1.3 fold after 6 h of unilateral blue-light illumination, 1.5-fold after 12 h, and two-fold after 24 h (Figure 2a–d, g). This result suggests that blue light regulates the asymmetric DR5 activity during root negative phototropism. Furthermore, we assayed auxin distribution using the DII-VENUS auxin sensor, which allows detection of dynamically relative changes in auxin distribution with cellular resolution (Brunoud et al., 2012). While no asymmetric distribution of DII-VENUS fluorescence was detected in the dark, decreased DII-VENUS fluorescence was observed at the illuminated side of roots irradiated with unilateral blue light for as little time as 0.5 h (Figure S1a, b), further demonstrating blue-light-regulated asymmetric auxin distribution. In contrast, no asymmetric DR5 activity was detected in the phot1 mutant (Figure 1e, g), which is defective in root negative phototropism. These data suggest the involvement of phot1-regulated asymmetric auxin redistribution in the root negative phototropic response.

Figure 2.

The asymmetric auxin distribution during root negative phototropism. (a–f) DR5 activity was monitored in the DR5REV::GFP marker line exposed to unilateral blue-light illumination (10 μmol m−2 sec−1) for 0 h (a), 6 h (b), 12 h (c), or 24 h (d), and in DR5REV::GFP phot1 (e) and DR5REV::GFP pin3-4 (f) seedlings exposed to unilateral blue-light illumination (10 μmol m−2 sec−1) for 24 h. Arrows indicate blue-light direction. Bars = 50 μm. (g) GFP signal intensities in (a–f) were quantified and their ratios at the illuminated side versus the shaded side are presented in (g). At least 12 seedlings were imaged per line for each of three replicates. Error bars represent standard deviation and indicate significant difference at **P < 0.01, as determined by Student's t-test.

PIN3 is involved in asymmetric auxin distribution and root negative phototropism

As auxin was redistributed during blue-light-induced root negative phototropism, we used the polar auxin transport inhibitor N-1-naphthylphthalamic acid (NPA) to further examine the role of auxin polar transport in this response. When Arabidopsis plants were treated with NPA, the roots showed severe defects in blue-light-induced negative phototropism (Figure 1a–c), consistent with a previous report (Wan et al., 2012). This finding suggests that polar auxin transport plays an important role in root phototropism. When the hypocotyls were removed from the plants and their roots were illuminated with unilateral blue light, the roots still bent away from the light (Figure S2), indicating that auxin from the shoot is not necessary for the blue-light-induced redistribution of auxin in the root, consistent with a previous report showing that removal of the shoot does not impair root gravitropism (Rashotte et al., 2000).

The directional flow of auxin is mediated by auxin polar transporters of the AUX and PIN families (Geisler and Murphy, 2006; Petrasek et al., 2006; Yang et al., 2006), and the auxin efflux carrier PIN3 is key factor in the asymmetric distribution of auxin during hypocotyl phototropism and hypocotyl gravitropism (Friml et al., 2002; Blakeslee et al., 2004; Ding et al., 2011; Rakusova et al., 2011). To test whether PIN3 is involved in root negative phototropism, the pin3-4 mutant was treated with unilateral blue light and the root phototropic response was monitored. We found that pin3-4 had reduced root negative phototropic response (Figure 1a–d), and less asymmetry in DR5 activity in unilateral blue light as compared with the wild-type plants (Figure 2f, g). These results reveal a role of PIN3 in root negative phototropism, and show that PIN3 activity is involved in the generation of asymmetric auxin distribution underlying root negative phototropic response.

Unilateral blue-light illumination induced the asymmetric distribution of PIN3 in columella cells

It has been reported that PIN3 polarization is needed for the generation of asymmetric auxin distribution during hypocotyl phototropism and root gravitropism (Friml et al., 2002; Harrison and Masson, 2008; Kleine-Vehn et al., 2010; Ding et al., 2011; Rakusova et al., 2011). Thus, we examined the effect of blue light on PIN3–GFP distribution in the PIN3::PIN3–GFP marker line. As a result, we made an interesting observation regarding asymmetric distribution of PIN3–GFP in columella cells during blue-light-induced root negative phototropic response: in the dark, no obvious polarity of PIN3–GFP was observed on the outer lateral membrane of columella cells (Figure 3a, e); however, the PIN3–GFP signal was stronger on the outer lateral membrane of columella cells at the illuminated side of roots exposed to unilateral blue light (Figure 3b–d, f–h), in contrast to higher PIN3 activity on outer lateral membrane of endodermis cells at the shaded side of hypocotyls exposed to unilateral blue light (Ding et al., 2011), suggesting a role of this increased PIN3–GFP accumulation on the outer lateral membrane of columella cells for higher auxin activity at the illuminated side of the roots. A time course analysis revealed that PIN3 can be distributed asymmetrically to the outer lateral membrane of columella cells on the illuminated side within 30 min of unilateral blue-light illumination in one single root (Figure S3a–d). This asymmetry in PIN3–GFP distribution following the unilateral blue-light illumination, hereafter called ‘blue-light-induced PIN3 polarization’, is involved in the generation of asymmetric auxin distribution during root negative phototropic response. The ratio of PIN3–GFP signal intensity on the outer lateral membrane of columella cells at the illuminated side versus the shaded side of roots increased up to about 1.3-fold after 0.5 h of unilateral blue-light illumination, 1.5-fold after 1 h, and 1.8-fold after 2 h (Figure 3m). While the cry1 and phot2 mutants showed blue-light-induced PIN3 polarization (Figure 3n, o), no visible blue-light-induced PIN3 polarization was detected in phot1 plants under unilateral blue-light illumination (Figure 3i–m). These results demonstrate that blue-light-induced PIN3 polarization is mediated by the blue-light receptor phot1.

Figure 3.

Unilateral blue-light illumination induced the asymmetric distribution of PIN3 in columella cells. (a–d, i–l, n, o) PIN3 localization, as revealed by GFP fluorescence, in the columella cells of PIN3::PIN3–GFP and PIN3::PIN3–GFP phot1 plants grown in the dark (a, i) or exposed to unilateral blue light (10 μmol m−2 sec−1) for 0.5 h (b, j), 1 h (c, k), and 2 h (d, l), and in the columella cells of PIN3::PIN3–GFP cry1 (n) and PIN3::PIN3–GFP phot2 plants (o) exposed to unilateral blue light (10 μmol m−2 sec−1) for 2 h. Arrowheads show PIN3–GFP localization on the outer lateral membrane of columella cells. Arrows indicate the blue-light direction. Bars = 10 μm. (e–h) Details of blue-light-induced PIN3 polarization. The fluorescence intensities of PIN3–GFP in (a–d) are presented and enlarged in (e–h). Arrowheads show PIN3–GFP signal on the outer lateral membrane of columella cells. Arrows indicate the blue-light direction. Bars = 20 μm. (m) PIN3–GFP fluorescence intensities in (a–d, i–l) were quantified and the ratios of PIN3–GFP signal on the outer lateral membrane of columella cells at the illuminated side versus the shaded side are presented in (m). At least twelve seedlings were imaged for each of three replicates. Error bars represent standard deviation and indicate significant difference at **P < 0.01 (Student's t-test).

Blue-light-induced PIN3 polarization is regulated by a BFA-sensitive, GNOM-dependent trafficking pathway

The polar targeting of PINs can be regulated by de novo protein synthesis or degradation. To investigate the possible role of protein synthesis in blue-light-induced PIN3 polarization in root negative phototropism, we analyzed whether PIN3 polarization requires de novo protein synthesis using the protein synthesis inhibitor, cycloheximide (CHX). Our results showed that blue-light-induced PIN3 polarization occurred normally in PIN3::PIN3–GFP plants treated with CHX (Figure S4b, e, g), a finding that suggested that de novo protein synthesis is not the major role in the blue-light-induced PIN3 polarization. If 2-day-old etiolated seedlings transferred onto the ½MS plate with CHX were exposed to unilateral blue light for 48 h, the root negative phototropic response was reduced (Figure S4j). The asymmetric auxin distribution was also altered by longer CHX treatment (Figure S4h, i). These results suggest that the asymmetry of PIN3 does not require protein synthesis initially, but later steps in the process do so. In addition, the possible role of proteolytic protein degradation in this process was analyzed using MG132 (Wang et al., 2013), an inhibitor of the 26S proteasome. No visible difference was observed in blue-light-induced PIN3 polarization, asymmetric auxin distribution and root negative phototropic response between roots subjected to MG132 treatment and roots lacking the drug treatment (Figure S4c, f, g–j), a finding that indicated that protein degradation does not participate in blue-light-induced PIN3 polarization in root negative phototropism.

PIN proteins are recycled constitutively between endosomes and the plasma membrane, and the recycling is sensitive to the vesicle trafficking inhibitor BFA (Geldner et al., 2001; Dhonukshe et al., 2007). Thus, the BFA-sensitive, vesicle trafficking pathway may be involved in blue-light-dependent polarization of PIN3 in root negative phototropism. Then, BFA was used to test whether the BFA-sensitive trafficking pathway is involved in blue-light-induced PIN3 polarization in root negative phototropism. BFA treatment strongly inhibited the blue-light-induced PIN3 polarization and normal asymmetric auxin distribution (Figure 4a, c, e, f, h, j). The root negative phototropic response was also affected by BFA treatment (Figures 4k and S5), consistent with a previous report (Wan et al., 2012). These results demonstrate that the BFA-sensitive trafficking pathway is involved in blue-light-induced PIN3 polarization and root negative phototropism.

Figure 4.

Involvement of the brefeldin A (BFA)-sensitive, GNOM-dependent trafficking pathway in blue-light-induced PIN3 polarization. (a–d) PIN3 localization, as revealed by GFP fluorescence, in columella cells of PIN3::PIN3–GFP seedlings (a, c), and PIN3::PIN3–GFP GNOMM696L seedlings (b, d), treated or not with BFA and exposed to unilateral blue light for 2 h. Four-day-old etiolated seedlings were pretreated with DMSO as a control or BFA (50 μm) in the dark for 1 h, and subsequently exposed to unilateral blue-light illumination (10 μmol m−2 sec−1) for 2 h. Arrowheads show PIN3–GFP localization on outer lateral membrane of columella cells. Arrows indicate blue-light direction. (e) PIN3–GFP fluorescence intensities in (a–d) were quantified and the ratios at the illuminated side versus shaded side are summarized in (e). Error bars represent standard deviation and indicate significant difference at **P < 0.01 (Student's t-test). At least 12 seedlings were imaged per line for each of three replicates. Bars = 10 μm. (f–i) DR5 activity monitored in the DR5REV::GFP (f, h) and DR5REV::GFP GNOMM696L (g, i) seedlings treated or not with BFA (20 μm) and exposed to unilateral blue-light illumination (10 μmol m−2 sec−1) for 24 h. Arrows indicate blue-light direction. Bars = 50 μm. (j) GFP signal intensities in (f–i) were quantified and their ratios at the illuminated side versus the shaded side are presented in (j). At least 12 seedlings were imaged per line for each of three replicates. Error bars represent standard deviation and indicate significant difference at **P < 0.01, as determined by Student's t-test. (k) Root negative phototropism phenotypes were examined for 2-day-old etiolated wild-type (n = 177 or n = 424) and GNOMM696L (n = 104 or n = 98) seedlings treated or not with BFA (20 μm) and illuminated with unilateral blue light (10 μmol m−2 sec−1) for 2 days. gnomR5 (n = 61) showed strong defects in root negative phototropism when exposed to unilateral blue light (10 μmol m−2 sec−1) for 2 days. Each unilateral light-illuminated root was assigned to one of twelve 30° sectors and the length of each bar represents the percentage of seedlings showing the direction of root growth within that sector.

GNOM, an important molecular target of BFA, has been reported to be involved in BFA-sensitive trafficking (Geldner et al., 2003). A partial loss-of-function gnomR5 mutant (Geldner et al., 2004) was assayed for blue-light-induced root negative phototropism. The reduced blue-light-induced root negative phototropic response again indicated the importance of a GNOM-dependent PIN trafficking pathway in this process (Figures 4k and S5). The differences in the root negative phototropic response of gnomR5 mutant and BFA-treated wild-type plants are probably that gnomR5 mutant has the reduction of GNOM function (Geldner et al., 2004). In support of this observation, the roots of GNOMM696L lines, which express a genetically engineered BFA-resistant version of GNOM (Geldner et al., 2003), exhibited normal blue-light-induced PIN3 polarization, asymmetric auxin distribution and root negative phototropic responses in both the presence and absence of BFA (Figures 4b, d, e, g, i–k and S5).

PID and PP2A antagonistically mediate blue-light-induced PIN3 polarization

PID, a member of membrane-associated AGC3 kinase, is involved in PIN3 polarization in hypocotyl phototropism and hypocotyl gravitropism (Ding et al., 2011; Rakusova et al., 2011). The polar distribution of PIN3 in blue-light-induced root negative phototropism may also be modulated by this pathway. Thus, Pro35S:PID seedlings constitutively expressing PID (Benjamins et al., 2001) and pid-14 seedlings lacking PID activity (Huang et al., 2010) were used. The Pro35S:PID plants, which had severe defects in root negative phototropism (Figures 5q and S5), showed the abnormal blue-light-induced PIN3 polarization in response to unilateral blue-light illumination (Figure 5b, g, l, p), suggesting that the PID-mediated pathway may be involved in blue-light-induced PIN3 polarization and root negative phototropism. However, Upon unilateral blue-light illumination, the pid-14 mutant exhibited normal blue-light-induced PIN3 polarization and root negative phototropism (Figures 5c, h, m, p, q and S5). Considering the existence of WAG1 and WAG2, the closest homologues of PID, wag1 wag2 pid triple mutant (Cheng et al., 2008; Dhonukshe et al., 2010) was employed. The reduced blue-light-induced PIN3 polarization and root negative phototropism in this triple mutant further reveal a role of PID in this response (Figures 5d, i, n, p, q and S5).

Figure 5.

PID and PP2A antagonistically modulate blue-light-induced PIN3 polarization. (a–j) PIN3 localization, as revealed by GFP fluorescence, in columella cells of PIN3::PIN3–GFP (a, f), PIN3::PIN3–GFP Pro35S:PID (b, g), PIN3::PIN3–GFP pid-14 (c, h), PIN3::PIN3–GFP wag1 wag2 pid (d, i), and PIN3::PIN3–GFP pp2aa1 (e, j) seedlings in the dark or exposed to unilateral blue light (10 μmol m−2 sec−1) for 2 h. Bars = 10 μm. (k–o) Details of blue-light-induced PIN3 polarization. The fluorescence intensities of PIN3–GFP in (f–j) are presented and enlarged in (k–o). Arrowheads show PIN3–GFP signal on the outer lateral membrane of columella cells. Arrows indicate the blue-light direction. Bars = 20 μm. (p) PIN3–GFP fluorescence intensities in (a–j) were quantified and the ratios at outer lateral membrane at the illuminated side versus shaded side are summarized in (p). At least 12 seedlings were imaged per line for each of three replicates. Error bars represent standard deviation and indicate significant difference at **P < 0.01 (Student's t-test). Arrowheads show PIN3–GFP localization on outer lateral membrane of columella cells. Arrows indicate blue-light direction. (q) The phenotypes of root negative phototropism in 2-day-old etiolated wild-type (n = 101), Pro35S:PID (n = 84), pid-14 (n = 65), wag1 wag2 pid (n = 73), and pp2aa1 (n = 146) seedlings subjected to unilateral blue-light illumination (10 μmol m−2 sec−1) for 2 days. Each unilateral light-illuminated root was assigned to one of twelve 30° sectors and the length of each bar represents the percentage of seedlings showing the direction of root growth within that sector.

PP2A phosphatase activity is also required for auxin transport-dependent plant development (Michniewicz et al., 2007). Thus, the pp2aa1 (rcn1) mutant, which losses the phosphatase activity of PP2AA1 (Garbers et al., 1996) was used to explore the possible role of PP2A in blue-light-induced PIN3 polarization. Upon illumination, the asymmetric distribution of PIN3 was less pronounced in pp2aa1 mutants exposed to unilateral blue light than in wild-type plants illuminated in such a manner (Figure 5e, j, o, p). Combined with our observation that pp2aa1 had a reduced root negative phototropic response (Figures 5q and S5), these results demonstrate that PP2A activity is involved in blue-light-induced PIN3 polarization and root negative phototropism.

The expression of both PID and PP2A is regulated by blue-light receptor phot1

PID expression is known to be regulated by blue light during hypocotyl phototropism (Ding et al., 2011; Haga and Sakai, 2012). To investigate whether the expression of PID and PP2A are modulated by blue light, we used the PID::PID–YFP and PP2AA1::PP2AA1–GFP marker lines (Michniewicz et al., 2007). Our results indicate that the fluorescence signal intensity resulting from PID–YFP was reduced in roots exposed to blue light compared with that of plants grown in the dark (Figure 6a, e). In contrast, while the PP2AA1–GFP signal was weak in the dark (Figure 6b), blue-light illumination can increase the PP2AA1–GFP signal intensity (Figure 6b, f). These data demonstrate that the expression of both PID and PP2AA1 is regulated by blue light.

Figure 6.

Blue-light-regulated expression of PID and PP2A. (a, c) PID localization, as revealed by YFP fluorescence, in the root cells of PID::PID–YFP (a) and PID::PID–YFP phot1 (c) seedlings grown in the dark or subjected to different times of blue light (10 μmol m−2 sec−1). Bars = 30 μm. (b, d) PP2AA1 localization, as revealed by GFP fluorescence, in the root cells of PP2AA1::PP2AA1–GFP (b) and PP2AA1::PP2AA1–GFP phot1 (d) seedlings grown in the dark or subjected to different times of blue light (10 μmol m−2 sec−1). Bars = 30 μm. (e, f) The ratio of PID (e) and PP2AA1 (f) fluorescence intensities in the roots of seedlings grown in different times of blue light (10 μmol m−2 sec−1) illumination versus in darkness are presented in (e, f) for the wild-type or phot1. At least twelve seedlings were imaged per line for each of three replicates. Error bars represent standard deviation and indicate significant difference at **P < 0.01 (Student's t-test). (g, h) The relative expression of PID (g) and PP2AA1 (h) was measured by quantitative real-time PCR relative to an internal tubulin control (TUB4). RNA was isolated from 4-day-old seedlings of wild-type and phot1 seedlings grown in the dark or exposed to different times of blue light (10 μmol m−2 sec−1). The relative expression ratios of PID and PP2AA1 were obtained by normalizing to the expression in dark-grown wild-type or phot1 in (g, h). The experiments were repeated three times. Error bars represent standard deviations and indicate significant difference at *P < 0.05 and **P < 0.01 (Student's t-test).

We also tested whether the expression of both PID and PP2AA1 is regulated by blue-light receptor phot1. As expected, no visible change in PID–YFP and PP2AA1–GFP fluorescence was observed in the roots of the phot1 progeny, either when grown in the dark or exposed to unilateral blue light (Figure 6c–f). These observations were further supported by quantitative real-time PCR (qRT-PCR) analysis of PID and PP2AA1 expression in plants grown first in the dark and then exposed to blue light (Figure 6g, h). Overall, these results demonstrate that the expression of both PID and PP2AA1 is regulated by blue-light receptor phot1.

Discussion

Our data indicated that blue-light-induced PIN3 polarization, which is regulated by a BFA-sensitive, GNOM-dependent, trafficking pathway and the activity of PID and PP2A, is involved in root negative phototropism. The PID- and PP2A-modulated switching of PINs localization has been documented in numerous other studies (Friml et al., 2004; Wisniewska et al., 2006; Michniewicz et al., 2007; Kleine-Vehn et al., 2009; Huang et al., 2010; Zhang et al., 2010; Ding et al., 2011; Rakusova et al., 2011). In these reports, The PID-dependent PINs polarization is regulated by a BFA-sensitive, GNOM-dependent, trafficking pathway during organogenesis and development (Kleine-Vehn et al., 2009; Sorefan et al., 2009; Dhonukshe et al., 2010). Recently, PID-dependent PIN3 polarization is shown to be involved in asymmetric distribution of auxin during hypocotyl phototropism and gravitropism, and this asymmetric distribution of PIN3 is also regulated by a BFA-sensitive, GNOM-dependent, trafficking pathway (Ding et al., 2011; Rakusova et al., 2011). Furthermore, blue-light-induced PIN2 recycling is regulated by a BFA-sensitive trafficking pathway during root negative phototropism (Wan et al., 2012). Taken together, these data suggest that PID-mediated PINs polarization via a BFA-sensitive, GNOM-dependent, trafficking pathway may be a universal mechanism to direct polar auxin transport in response to environmental and endogenous cues.

Previous experiments have shown that loss-of-function mutations in the PP2AA1 (RCN1) gene result in plants defective in root and shoot gravitropism by altering auxin transport (Garbers et al., 1996; Deruere et al., 1999; Rashotte et al., 2001; Shin et al., 2005; Muday et al., 2006; Michniewicz et al., 2007; Blakeslee et al., 2008; Rahman et al., 2010). Further evidence indicates that this phosphatase can act antagonistically to membrane-associated AGC3 kinases, including PID, WAG1 and WAG2 (Michniewicz et al., 2007; Dhonukshe et al., 2010; Huang et al., 2010; Zhang et al., 2010). If PID is over-expressed, the transgenic lines exhibit an apical–basal shift in the polarity of several PIN proteins and result in their roots to be agravitropic (Christensen et al., 2000; Benjamins et al., 2001; Friml et al., 2004; Rahman et al., 2010). While loss-of-function pid alleles do not show profound defects in root growth (Christensen et al., 2000; Benjamins et al., 2001), Sukumar et al. (2009) demonstrated that pid-9 exhibits delayed rates of root gravitropism. The triple mutant wag1 wag2 pid showed the strong defective in root development (Dhonukshe et al., 2010). In our studies, we also observed antagonistic effects of PID and PP2AA1 in root negative phototropic response. Both PID over-expression lines and pp2aa1 mutant can change PIN3 distribution and result in abnormal root negative phototropism. Similarly, while pid-14 mutant did not affect root negative phototropism, the wag1 wag2 pid triple mutant exhibited reduced root negative phototropism.

Our results also showed that while the expression of PP2AA1 mediated by phot1 was increased with blue-light irradiation, the inhibition of de novo protein synthesis resulted from CHX treatment for 2 h did not change blue-light-induced PIN3 polarization. The explanation for this event could be that the light-increased PP2AA1 expression is not needed for normal blue-light-induced PIN3 polarization. But, PP2AA1 is essential because our mutant analyses indicated a role of PP2AA1 in both PIN3 polarization and root negative phototropism. Thus, de novo protein synthesis is not a major role in root negative gravitropism. It is reported that PP2AA1 phosphatase activity is required for proper polar PIN1 localization (Michniewicz et al., 2007). It is possible that PP2AA1-dependent dephosphorylation could be important in blue-light-induced PIN3 polarization.

Unilateral blue-light illumination can polarize PIN3 to the outer lateral membrane of endodermis cells at the shaded side of hypocotyls (Ding et al., 2011). Similarly, increased PIN3 activity was detected on the outer lateral membrane of columella cells at the illuminated side of roots exposed to unilateral blue light (Figure 3a–h). The mechanism by which blue light differentially adapts the polarization of PIN3 in different organs is unclear.

Unlike the phot1 mutant, the pin3-4 mutant just only exhibited reduced root negative phototropism compared to the wild-type seedlings (Figure 1a–d), implying that other auxin transport proteins may be also involved in this process. PIN2 has been shown to be involved in root negative phototropism (Wan et al., 2012). PIN7, which act with PINs together in a functionally redundant manner for the redistribution of auxin in root gravitropic response (Kleine-Vehn et al., 2010), may also participate in this process. Recently, it has been revealed that ABCB19 is involved in auxin redistribution in hypocotyl phototropism (Christie et al., 2011). Thus, the other auxin transporters could be further analyzed for their possible roles in root negative phototropism.

Our results indicated that the phot1-regulated PIN3 polarization is involved in root negative phototropism (Figure 3). The blue-light-induced PIN3 polarization presumably redirects auxin flow to the illuminated side of roots, where auxin accumulates and causes roots bending away from the light source. However, it has been reported that phot1 does not directly phosphorylate PIN3 and appears to act on PIN3 through intermediate processes in hypocotyl phototropism (Ding et al., 2011). Another report also indicates that PIN auxin efflux carriers are necessary for plus-induced but not continuous light-induced phototropism (Haga and Sakai, 2012). Upon these results in hypocotyl phototropism, the PIN3-mediated mechanism underlying root negative phototropic response remains to be fully determined.

Phototropins can respond to the changing environments induced by blue light, and utilize specific proteins as signal transducers in these responses. A previous report indicated that PKS1, a phot1-interacting protein, plays a role in root negative phototropic response (Boccalandro et al., 2008). Also, NPH3, another phot1-interacting protein, is required for many physiological processes induced by blue light (Motchoulski and Liscum, 1999; Inoue et al., 2008a, 2010). MAB4/ENP, an NPH3-like protein, also regulates the polarity of PIN1 in cotyledon primordia and functions as a signal transducer in hypocotyl phototropism (Furutani et al., 2007, 2011; Matsuda et al., 2011). RPT2, like NPH3, has a BTB domain and functions as an essential signal transducer associated with phot1 during root negative phototropism (Sakai et al., 2000; Inada et al., 2004). Recently, NPH3 was shown to regulate PIN2 recycling during root negative phototropic response (Wan et al., 2012). Thus, other NPH3-related proteins, including MAB4/ENP and RPT2, are strong candidates for linking phototropins with polar auxin transporters in root negative phototropism.

Experimental procedures

Plant material

The following published transgenic and mutant lines were used in this study: DR5REV::GFP (Friml et al., 2003); DII-VENUS (Brunoud et al., 2012); PIN3::PIN3–GFP (Ding et al., 2011); PID::PID–YFP (Michniewicz et al., 2007); PP2AA1::PP2AA1–GFP (Michniewicz et al., 2007); GNOMM696L (Geldner et al., 2003); gnomR5 (Geldner et al., 2004); pin3-4 (Friml et al., 2003); pid-14 (Huang et al., 2010); wag1 wag2 pid (Dhonukshe et al., 2010); Pro35S:PID (Benjamins et al., 2001); pp2aa1 (Garbers et al., 1996); cry1 (SALK_069292); phot1 (SALK_146058); and phot2 (SALK_142275). All of above mutants were verified by PCR and RT-PCR. The double or triple mutants and/or transgenic lines were obtained by crossing the respective lines above and confirmed by PCR. All PCR primers used for genotyping are listed in Table S1.

Plant growth and light conditions

Arabidopsis seeds were surface sterilized with 5% bleach for 5 min, washed three times with sterile water, and plated on agar medium containing half-strength Murashige and Skoog medium (1962) and 0.8% agar (w/v). Seedlings were grown in darkness for 2–4 days at 22°C and stimulated by blue light for further analysis. The blue light (nm = 475) was provided by an R30 LED Light (enLux) and LH-100SP-LED (NK system). Light fluence rates were measured by a Li250 quantum photometer (Li-Cor, www.licor.com). All experiments in darkness were carried out under a dim green safe light.

Measurement of root negative phototropism

Arabidopsis seeds were grown in the dark for 2 days and then transferred to unilateral blue-light illumination for 2 days as described previously (Sakai et al., 2000). The bending angles of the roots away from the vertical direction were measured and analyzed using image j software (http://rsb.info.nih.gov/ij/) and plotted using prism 5.0 software (GraphPad, www.graphpad.com). At least three independent experiments were carried out.

Pharmacological treatments

Four-day-old etiolated seedlings on half-strength Murashige and Skoog medium were treated with BFA (50 μm), CHX (50 μm), or MG132 (50 μm) for 1 h in the dark. The seedlings were then exposed to unilateral blue light and imaged. In experiments involving the phototropic response, 2-day-old etiolated seedlings grown on plates without drug treatment were transferred to solid half-strength Murashige and Skoog medium containing BFA (20 μm), NPA (1 or 5 μm), CHX (50 μm), or MG132 (50 μm) and exposed to unilateral blue light for 2 days. In control experiments, seedlings were treated with an equal amount of solvent dimethyl sulphoxide (DMSO). Propidium iodide (PI; 0.05%) was dissolved in distilled water. Each experiment was performed at least three times.

Quantitative real-time PCR analysis

RNA was extracted from 4-day-old seedlings using TRIzol Reagent (Invitrogen, www.invitrogen.com). After 1 μg of total RNA was treated with RQ1 RNase-free DNase (Promega, www.promega.com), first-strand cDNA synthesis was carried out using Superscript II Reverse transcriptase (Invitrogen) according to the manufacturer's instructions. TUB4 was used as an internal control. qRT-PCR analysis was performed on a Bio-rad CFX96 apparatus with the dye SYBR Green (Invitrogen). All individual reactions were performed in triplicate. The primers used for qRT-PCR analysis are listed in Table S1.

Confocal microscopy

An Olympus (www.olympus.com) FV1000 ASW confocal scanning microscope was used. Emission wavelengths were as follows: PI, 600 to 640 nm; GFP, 500 to 540 nm; and YFP, 525 to 565 nm. Four-day-old etiolated seedlings were grown in the dark and stimulated with unilateral blue light. Etiolated seedlings were placed on slides and shoots were removed with a blade, leaving only the roots for quick confocal observations. The signal intensity was measured using Photoshop CS4 and image j software (http://rsb.info.nih.gov/ij/). Based on the previous report (Ding et al., 2011; Rakusova et al., 2011), the fluorescence intensity ratios were obtained by comparing DR5–GFP fluorescence intensities between the illuminated side and shaded side of the root in the responsive part, PIN3–GFP fluorescence intensities between on the outer lateral membrane of columella cells at the illuminated side and shaded side of the roots. For fluorescence intensity ratios of PID–YFP and PP22A1–GFP were obtained by comparing fluorescence intensities between light-illuminated and dark-grown roots from the whole image by confocal microscope. At least 12 seedlings were imaged per line for each of three replicates.

Acknowledgements

We thank Jiri Friml, Gerd Jürgens, and Jian Xu for sharing their published materials. This work was supported by the National Natural Science Foundation of China (#90917001) and Key Project of Chinese Ministry of Education (#311026) to Y-T Lu.

Conflicts of Interest

The authors declare no competing financial interests.

Ancillary