Introduction

Plant hormone auxins, of which indole-3-acetic acid (IAA) is the most predominant, play a crucial role in controlling multiple physiological processes in plant growth and development such as cell elongation, tropic movement, vascular patterning, apical dominance and root initiation (Berleth and Sachs 2001; Ueda et al. 2016; Went and Thimann 1937). Auxin is mainly biosynthesized in the shoot apex and young leaves, and transported between cells in the basipetal direction via a combination of membrane diffusion and carrier-mediate transport systems in the plant axis, generating auxin maxima (Kramer and Bennett 2006; Muday and Murphy 2002; Roberts and Friml 2009). This unique transport is called polar auxin transport.

Polar auxin transport is considered to be regulated by several functional proteins (e.g. influx and efflux carrier proteins) located in the plasma membrane (Bandyopadhyay et al. 2007; Benjamins et al. 2005; Friml and Palme 2002; Miyamoto et al. 2011; Muday and Murphy 2002; Ueda et al. 2011). Studies regarding one of the floral mutants of Arabidopsis thaliana, a mutation of the PIN-FORMED1 (PIN1) gene, revealed that this mutation diminished polar auxin transport in the inflorescence axis (Okada et al. 1991), resulting in identification of the AtPIN1 gene encoding a 67-kDa protein with similarity to bacterial and eukaryotic carrier protein (Gälweiler et al. 1998). Since AtPIN1 proteins were detected at the basal end of auxin transport-competent cells in vascular tissue, AtPIN1 is considered to participate directly in auxin transport or assist in the assembly of other proteins with efflux activity (Friml and Palme 2002). It has also been demonstrated that PIN-dependent auxin efflux and local auxin response result in the apical-basal formation of the embryo, and thus determine the axiality of the adult plant (Friml et al. 2003). The molecular mechanisms of polar auxin transport/movement relative to such varied physiological responses as graviresponse and plant architecture determination have been studied using immunohistochamical techniques in maize and cucumber as well as Arabidopsis (Adamowski and Friml 2015; Carraro et al. 2006; Morohashi et al. 2017; Nishimura et al. 2009; Watanabe et al. 2012). Raising anti-PIN1 polyclonal antibodies for pea plants has already been reported by DeMason and Chawla (2009), but unfortunately a limited amount of immune serum was distributed upon request. This fact prompted us to produce a large amount of novel polyclonal antibody to PsPIN1, so as to clarify the mode of action of PsPIN1 in regulating auxin transport/movement in etiolated Alaska pea seedlings, and resulting in its successful and adequate raising for use in this study and future analyses.

2,3,5-Triiodobenzoic acid (TIBA) has been found to function as a polar auxin transport inhibitor as well as N-1-naphthylphthalamic acid (NPA), 9-hydroxyfluorene-9-carboxylic acid (HFCA) and others in the multiple systems of plants. TIBA has also been found capable of inducing pin-formed mutant-like plants in Arabidopsis (Okada and Shimura 1994; Oka et al. 1999) and automorphogenesis-like pea epicotyl bending (Miyamoto et al. 2005). Unfortunately, the functions of TIBA in inhibiting polar auxin transport remain unclear. There are possibilities that TIBA affects the cycling of PINs which is responsible for auxin influx (Michniewicz et al. 2007), and the amount of thick actin microfilaments and stabilized action affecting vesicle trafficking as suggested by Michniewicz et al. (2007), Dhonukshe et al. (2008) and Kojo et al. (2014). These experimental results might be sufficient to speculate that TIBA influences actin cytoskeleton, resulting in inhibition of the actin-dependent cellular trafficking including auxin transport components such as PINs. In addition, TIBA has also been shown to inhibit both recovery of AtPIN1 from the plasma membrane and re-distribution of AtPIN1 to the plasma membrane in Arabidopsis (Geldner et al. 2001). TIBA might be effective to regulate graviresponse in etiolated Alaska pea seedlings. Hoshino et al. (2007) showed that graviresponse in etiolated Alaska pea seedlings as well as an ageotropum pea mutant which showed no response of its etiolated epicotyls to gravistimulation is altered by the application of TIBA. Judging from the facts described above, it is worthwhile to study the mechanisms and functions of TIBA in graviresponse of etiolated Alaska pea seedlings in relation to altered PsPIN1 localization in plasma membrane to change the growth and development of plants. These points should be investigated in the near future to clarify the physiological role of TIBA.

The STS-95 space experiment demonstrated that when using 14C-labelled IAA, polar auxin transport in the epicotyls of etiolated pea seedlings grown under the microgravity condition in space was reduced as compared with that under the 1 g condition on earth, and that etiolated epicotyls showed automorphogenesis (i.e. epicotyl bending and changes in root growth direction) in space (Ueda et al. 1999, 2014). These findings suggest that polar auxin transport is under gravitational control, and that a close relationship exists between automorphogenesis and the reduced-activity of polar auxin transport. Ground-based experiments using simulated weightlessness conditions on a 3-dimensional clinostat supported the results obtained in the space experiments (Miyamoto et al. 2005; Ueda et al. 2014). Furthermore, asymmetric polar auxin transport between the proximal side of the epicotyl to the cotyledon (proximal side) and distal one to the cotyledon (distal side) was found in the first internode of etiolated pea seedlings (Hoshino et al. 2006, 2007). DeMason and Chawla (2009) analyzed the transport of auxin in the roots, shoot apex, and leaf primordium using anti-PsPIN1 antiserums. However, the mechanisms by which gravity controls polar auxin transport, and by which asymmetric polar auxin transport occurs between the proximal and distal sides of epicotyls have not yet been clarified. Therefore, a study of these mechanisms requires taking an immunohistochemical approach using the anti-PsPIN1 antibody as well as a molecular biological approach of gene expression relating to polar auxin transport carrier proteins including an influx carrier PsAUX1 and an efflux carrier PsPIN1.

Here we report on raising a novel anti-PsPIN1 polyclonal antibody in detail, and then examining the localization of PsPIN1 in etiolated pea seedlings.

Materials and methods

Plant materials and growth conditions

Seeds of pea (Pisum sativum L. cv. Alaska) were purchased from Watanabe Seeds Co. (Misato, Miyagi, Japan). As the seed bed, rockwool blocks (width 87 mm × depth 42 mm × height 16 mm) cut from a large sheet of rockwool (Culture Mat; Nippon Rockwool, Tokyo, Japan) were individually placed in acrylic resin boxes (W 95 mm × D 50 mm × H 63 mm) of a precise fitting size. For ventilation, each box had four holes (1 cm in diameter) on the top, which were covered with hydrophobic fluoropore membrane (MilliSeal; Millipore, Merck, Tokyo, Japan). Twelve seeds were set so that the seeds were completely buried beneath the block surface, and with the seed axis (i.e. line to connect the plumular axis and radicle) set horizontal to the upper surface of the block. After supplying 45 mL of distilled water, each box was kept for 3 days at 23.5 °C in the dark. For treatment of an inhibitor of polar auxin transport, TIBA (Sigma-Aldrich, St. Louis, MO, USA) at 100 µM was prepared from 200 mM stock solution in dimethyl sulfoxide (DMSO) and distilled water containing 0.05% DMSO was used as the mock control. The apical hook region and subapical region of the epicotyl, and the root tips were excised from 3-day-old etiolated pea seedlings. Then the apical hook and subapical epicotyl region were divided into the proximal side and distal side of epicotyls to the cotyledon. The plant materials were immediately fixed with Carnoy’s fixative solution for immunohistochemical analyses, or frozen until use for western blotting analyses. For treatment of Brefeldin A (Sigma-Aldrich), the root tissues from 3-day-old etiolated pea seedlings were incubated for 2 h with 50 mL of 100 µM Brefeldin A in PBS buffer (137 mM NaCl, 2.7 mM KCl, 10 mM Na2HPO4, 1.76 mM KH2PO4, pH 7.4). 100 µM Brefeldin A was diluted from 200 mM stock solution in DMSO. 50 mL of PBS buffer containing 0.05% DMSO was used as the mock control. After incubation with or without Brefeldin A, the root segments were immediately fixed by Carnoy’s fixative solution.

Determination of polar auxin transport

The measurement of polar auxin transport was determined according to the method reported previously (Ueda et al. 1999) with some modifications. Epicotyl segments (15-mm long) were prepared from 3-day-old etiolated seedlings grown in the presence or absence of TIBA. Approximately 30 µL of aqueous lanolin (30% water, w/w) containing 1.75 µM (1 µM Ci mL− 1) of [1-14C]-indole-3-acetic acid (American Radiolabeled Chemical, St. Louis, MO, USA) was introduced to the bottom of a 1.5-mL Eppendorf tube. The apical side of epicotyl segments was put into the lanolin, then incubated for 18 h at 23.5 °C in the dark. After incubation, the epicotyl segments were divided into the proximal and distal sides, and then 2-mm pieces of the opposite (basal) side of the segments were excised and put into scintillation cocktails. Radioactivity was measured using a liquid scintillation counter (model 2000CA; Packard Instrument, Meriden, CT, USA).

Production of PsPIN1 antibody

To produce the polyclonal antibodies of PsPIN1, an oligo-peptide fragment with the hydrophilic region of PsPIN1 from amino acids 387–400, ‘VDGHRETQEDYLEK’ was synthesized according to the solid-phase peptide synthesis method (Eurofins Genomics, Tokyo, Japan) (Fig. 1). Then the peptides were conjugated to keyhole limpet hemocyanin protein (KLH) for the antigen. After immunization to rabbits for 77 days, the polyclonal anti-PsPIN1 antiserum obtained was affinity-purified against the PsPIN1 specific oligo-peptides using the manufacturer’s antibody production protocols.

Fig. 1
figure 1

Amino acid sequence alignment of pea and Arabidopsis PIN proteins. PsPIN1 (accession no. AY222857), PsPIN2 (accession no. AB112364), PsPIN3 (accession no. AB488678) and AtPIN1 (accession no. AF089084) were aligned using GENETYX ver. 10.1.5 software (Genetyx, Tokyo, Japan). Numbers to the left and right of each sequence denote the positions of amino acid residues in the corresponding proteins. Identical and similar amino acid residues are shaded with black and gray colors, respectively. The oligo-peptide region for targeting specific anti-PsPIN1 antibody is underlined in bold under multiple sequence alignments

Full size image

Extraction of membrane proteins and related western blotting analysis

The 12 frozen segments from pea seedlings were homogenized with 0.5 ml of the extraction buffer [8 M urea, 2% (w/v) CHAPS, 50 mM dithiothreitol (DTT) containing proteinase inhibitor cocktail tablets (Roche, Germany)]. Then an equal amount of 5% SDS solution was added to the homogenate, which was shaken for 1 h at 4 °C for extracting membrane proteins. After centrifugation at 13,000×g for 5 min at 4 °C, the supernatants as the membrane protein fraction were collected. Protein quantification in the membrane protein fraction was performed using the 2-D Quant Kit as per the manufacturer’s instructions (GE Healthcare Japan, Tokyo, Japan). Next, 20 µg of total proteins were applied on Ready Gel 5–20% (Wako, Osaka, Japan), and then subjected to SDS-PAGE. After SDS-PAGE, the proteins were transferred onto a polyvinylidene difluoride (PVDF) membrane (Bio-Rad, Tokyo, Japan) with Trans-Blot Turbo (Bio-Rad) using the manufacturer’s protocols. The blotted membrane was immersed in wash buffer [TBS buffer (20 mM Tris–HCl pH 7.6, 137 mM NaCl) containing 0.1% (v/v) Tween20], and incubated for 1 h with TBS containing 5% skim milk. Then the blotted membrane was incubated with the anti-PsPIN1 antibody at a 2000-fold dilution in blocking buffer for 2 h at room temperature. After primary antibody incubation, the hybridization membrane was washed three times for 10 min with the wash buffer and incubated with Anti-rabbit Ig, HRP-Linked in Whole Ab (GE Healthcare Japan) at a 10,000-fold dilution for 2 h at room temperature. Then the hybridization membranes were washed three times and incubated in ECL Select western blotting detection reagent (GE Healthcare Japan). Chemiluminescent signals were detected with a VersaDoc system (Bio-Rad). For total protein staining, the membrane was treated with Ponceau-S staining solution as per the manufacturer’s instructions (Beacle, Inc., Kyoto, Japan). The primary antibody of anti-β-actin (clone C4; Santa Cruz Biotechnology, CA, USA) and the secondary antibody of Anti-mouse Ig, HRP-Linked in Whole Ab (GE Healthcare Japan) were used as loading controls.

Immunohistochemical analyses of localization of PsPIN1

The immunolocalization of PsPIN1 was analyzed according to the methods of Nishimura et al. (2009) and Watanabe et al. (2012) with some minor modifications. Etiolated pea shoots and roots excised from the seedlings were immediately fixed in Carnoy’s fixative solution (60% ethanol, 30% chloroform and 10% acetic acid) (Yoneyama et al. 2003) for 3 h at 4 °C, along with maintenance of their orientation to gravity. Samples of the proximal and distal sides of the subapical epicotyl region, those of the apical hook, and root cap were excised, then further fixed for 90 min. After dehydration with an ethanol and tert-butyl alcohol series, the samples were embedded in Paraplast Plus (Sigma-Aldrich). Sections (10 µm) cut with a microtome (model RM2135; Leica Biosystems, Germany) were placed on silicon-coated glass slides. The slides were deparaffinized, treated with detergent solution {10% (v/v) DMSO and 3% (v/v) Nonidet P-40 in 0.5-fold PME [50 mM piperazine-1,4-bis(2-ethanesulfonic acid) (PIPES) pH 7.2, 5 mM MgSO4, 5 mM EGTA]} for 30 min, and then washed twice in 0.5-fold PME for 10 min. The slides were further treated with blocking buffer [onefold PME with 1% (w/v) bovine serum albumin and 0.5% (v/v) TritonX-100] for 1 h, and incubated overnight at 4 °C with 100-fold diluted anti-PsPIN1 antibody in onefold PME with 0.1% (w/v) bovine serum albumin. The slides were rinsed three times for 10 min with 0.5-fold PME containing 0.1% (v/v) Tween20, and then incubated with 300-fold diluted Alexa Fluor 488-conjugated goat anti-rabbit IgG (Molecular Probes, OR, USA) for 3 h at room temperature. After immunostaining, the nucleus was stained with 2 µg mL− 1 Propidium Iodide (Sigma-Aldrich) in 0.5-fold PME for 5 min. After washing with 0.5-fold PME containing 0.1% Tween20 three times for 10 min, the sample sections were sealed with ProLong Gold Anti slow fade reagent (Molecular Probes). The prepared samples were observed using an epi-fluorescence microscope (model BX51; Olympus, Tokyo, Japan), with images being acquired by a DP72 CCD camera (Olympus). For detecting the fluorescent images of Alexa488 and Propidium Iodide, U-FBNA and U-FGW filter sets (Olympus) were used, respectively. After immunohistochemical staining quantification analyses of PsPIN1 accumulation were conducted with image J software (Wayne Rasband, National Institutes of Health, USA) using five tissue sections. The total amount of green signal of PsPIN1 was estimated based on the independent experiments with five repeats. More than 100 cells in the hook region of etiolated Alaska pea seedlings were analyzed with at least five repeats. Student’s t test was used for statistical analyses.

Results and discussion

Effect of TIBA, an inhibitor of polar auxin transport, on morphogenesis and auxin transport in etiolated pea seedlings

As shown in Fig. 2a, etiolated epicotyls (the first internode) of the seedlings grown with water bent near the cotyledonary node due to greater elongation in the proximal side to the cotyledons (proximal side) than that in the distal one (distal side), thus grew upward due to graviresponse. The application of TIBA from the roots inhibited the graviresponse, resulting in epicotyl bending with ca. 60° from the horizontal position, whereas elongation growth of epicotyls was hardly affected by TIBA (Fig. 2b, c). As shown in Fig. 2d, polar auxin transport in the proximal side was significantly higher than that in the distal one as already reported (Hoshino et al. 2006, 2007). However, TIBA greatly reduced polar auxin transport not only in the proximal side but also in the distal one, thus confirming the facts regarding polar auxin transport in the first internode of etiolated pea seedlings. These results led us to provide additional evidence for such facts on polar auxin transport in the first internode of etiolated pea epicotyls using immunohistchemical analyses with the antibody of PsPIN1 protein, which is responsible and essential for auxin efflux from etiolated pea epicotyls cells.

Fig. 2
figure 2

Effects of 2,3,5-triiodobenzoic acid (TIBA), an auxin transport inhibitor, on growth and polar auxin transport in epicotyls of etiolated pea seedlings. Pea seeds were allowed to germinate and grow for 3 days at 23.5 °C in the dark. a 3-day-old etiolated pea seedlings grown in the absence (left) and presence (right) of TIBA at 10− 4 M. The grid line in the backboard of the chamber is 1 cm apart from each other. b, c Length and bending of epicotyls of 3-day-old etiolated pea seedlings, respectively. The epicotyl bending was expressed as the angle away from the cotyledons from the horizontal angle as illustrated in Fig. 2c. Statistical analyses were carried out using 20 pea seedlings. Values represent mean ± SE, **P < 0.01 by a Student’s t test. d Polar auxin transport determined using 14C-labeled IAA. Five independent experiments were measured for statistical analysis. Values represent mean ± SE, *P < 0.05 by a Student’s t test. D distal side, P proximal side

Full size image

Production of anti-PsPIN1 antibody and distribution of PsPIN1 in etiolated pea seedlings

The cDNA of PsPIN1, a putative auxin efflux carrier has been isolated from pea (Chawla and DeMason 2004), and demonstrated the closest similarity to AtPIN1 (Bai and DeMason 2006; Chawla and DeMason 2004). Until now, 3 cDNA containing the open reading frames of PsPIN1, PsPIN2 and PsPIN3 (with accession numbers AY222857, AB112364 and AB488678, respectively) have been isolated from etiolated pea epicotyls (Chawla and DeMason 2004; Hoshino et al. 2005; Ueda et al. 2012). PsPIN1 specific to 14 amino-acid peptide sequences from amino acids 387–400, ‘VDGHRETQEDYLEK’ with a hydrophilic region was used as the antigen against PsPIN1, because this peptide fragment is not completely homologous to other PsPINs (Fig. 1). After immunization to rabbits, the polyclonal anti-PsPIN1 antiserum obtained was affinity-purified against the PsPIN1-specific oligo-peptides. The affinity-purified antiserum specifically recognized approximately 67.8-kDa polypeptide in the membrane protein fraction extracted from 3-day-old etiolated pea seedlings (Fig. 3). Since the molecular size of ca. 67.8-kDa coincided with that of the putative PsPIN1 protein (Chawla and DeMason 2004), this purified-antiserum is recognized as being valuable as a PsPIN1 protein-specific antibody.

Fig. 3
figure 3

Western blot analysis using a novel anti-PsPIN1 antibody. After SDS-PAGE, the gel was blotted onto the PVDF membrane. Detection was conducted with Ponceau-S solution for total protein staining, pre-immune antiserum, and an affinity-purified novel anti-PsPIN1 antibody. The arrow indicates a molecular size of approximately 67.8-kDa corresponding to that of a putative PsPIN1

Full size image

Accumulation levels of PsPIN1 protein in each tissue of etiolated pea seedlings as determined by western blotting analysis demonstrated that PsPIN1 protein was clearly observed in the apical hook (hook), subapical region of the epicotyl (epicotyl), and the root tip, with the most abundant bending shown in the hook (Fig. 4), thus suggesting an important role of PsPIN1 in polar auxin transport relative to apical-basal formation in the axis.

Fig. 4
figure 4

Accumulation levels of PsPIN1 proteins in etiolated pea seedlings. The affinity-purified anti-PsPIN1 antibodies detected specifically ca. 67.8-kDa protein in microsomal membrane fractions of various tissues of etiolated pea seedlings. Actin was used as a protein loading control and detected by anti-β-Actin antibody. The parenthesized number below each band represents the relative amount of PsPIN1 protein normalized by the amount of actin in three independent experiments (mean ± SE)

Full size image

Hoshino et al. (2006, 2007) found asymmetric polar auxin transport between the epicotyl side proximal to the cotyledon (proximal side) and the distal one to the cotyledons (distal side) in the first internode of etiolated pea seedlings using 14C-labeled IAA, with polar auxin transport in the proximal side being much higher than that in the distal side. Similar asymmetric localization of mRNA of PsPIN1, and an asymmetric distribution of auxin response in the proximal and distal sides of epicotyls were also demonstrated by northern blot analyses using PsPIN1 and auxin-inducible gene PsIAA4/5, respectively (Hoshino et al. 2006, 2007).

In contrast, according to the western blot analyses using whole extractable proteins accumulation levels of PsPIN1 protein in the hook and epicotyl were similar between the proximal and distal sides (Fig. 4). This discrepancy leads us to analyze the cellular localization of PsPIN1 proteins that are important for physiological function for polar auxin transport using histochemical analysis.

Cellular localization of PsPIN1 in shoots and root tissues of etiolated pea seedlings

The cellular localization of PsPIN1 proteins in the roots, subapical region of epicotyl, and apical hook region of 3-day-old etiolated pea seedlings was determined with an anti-PsPIN1 antibody.

In the roots, PsPIN1 proteins are asymmetrically distributed in the plasma membrane, specifically in the basal side in central cylinder cells (Fig. 5a, b). The localization pattern of PsPIN1 proteins in the roots of etiolated pea seedlings was similar to that of PsPIN1 in pea roots and AtPIN1 in Arabidopsis roots as reported by DeMason and Chawla (2009) and Gälweiler et al. (1998), respectively. Conversely, Brefeldin A known as a vesicle trafficking inhibitor has been demonstrated to induce AtPIN1 accumulation in large intracellular compartments, mimicking the physiological effects of auxin transport inhibitors (Geldner et al. 2001). In addition, the subfamily of guanine-nucleotide exchange factor on ADP-ribosylation factor G protein (ARF GEF) to which GNOM belongs is thought to be involved in vesicular transport of the secretory pathway from the endoplasmic reticulum (ER) through the Golgi to the plasma membrane in animals and yeasts. On the other hand, in Arabidopsis thaliana, GNOM did not work on the secretory pathway but on endosomal protein recovery from the plasma membrane (Geldner et al. 2001; Tanaka et al. 2013). Indeed, the phenotype of gnom was not similar to the secretory pathway mutant, and the elongation of the pollen tube in which the secretory pathway actively worked was normal (Geldner et al. 2001). A similar effect of Brefeldin A on PINs accumulation in the endosomal compartment was also reported in maize seedlings (Boutté et al. 2006; Nishimura et al. 2009). As shown in Fig. 5c, d, when Brefeldin A was treated with pea root tissues, an accumulation of PsPIN1 in the large intracellular compartment was observed, but not localization in the basal side of the plasma membrane. This suggests that the membrane localization of PsPIN1 proteins is regulated by vesicular transport as well as AtPIN1 and ZmPIN1s. Thus, observations of the asymmetric distribution of PsPIN1 in the plasma membrane and the Brefeldin A-induced accumulation in the endosomal compartment in pea roots strongly suggest that PsPIN1 is the same as Arabidopsis AtPIN1 in terms of functioning as an auxin efflux carrier protein. Using the novel anti-PsPIN1 antibody described here is considered useful for immunohistochemical analyses of the cellular localization of efflux carriers for polar auxin transport.

Fig. 5
figure 5

PsPIN1 localization in roots of etiolated pea seedlings. An affinity-purified anti-PsPIN1 antibody was used in immunohistochemical analysis as a primary antibody. Signals for PsPIN1 and the nucleus appeared green and red in color, respectively. a, b Localization of PsPIN1 in central cylinder cells in roots. c, d Localization of PsPIN1 in root tissues incubated absence (c) and presence (d) of Brefeldin A at 10− 4 M. Arrowheads indicate the localization or accumulation of PsPIN1 proteins. cc central cylinder, en endodermal cells. Bars are 200 µm in a, b, and 50 µm in c, d

Full size image

Figures 6 and 7 show the cellular localization of PsPIN1 in the proximal and distal sides of the subapical region and apical hook region of etiolated epicotyls, respectively. In the subapical region of epicotyls, PsPIN1 localized in the basal side of the plasma membrane of endodermal cells where sedimentary amyloplast acts for perception of the direction of gravity in stems (Fukaki et al. 1998; Miyamoto et al. 2014) (Fig. 6a–c). An asymmetric distribution of PsPIN1 was observed, and quantitative analysis of PsPIN1 demonstrated that the amount of PsPIN1 localized in the proximal side of epicotyls was much higher than that in the distal side (Fig. 6g). Hoshino et al. (2007) have reported that the accumulation of PsPIN1 mRNA analyzed by in situ hybridization was much higher in the proximal side of epicotyls than that in the distal side of epicotyls in etiolated pea seedlings as polar auxin transport, resulting in an asymmetric endogenous level of auxin in these regions as previously reported (Ueda et al. 2014). These results suggest that the asymmetric expression of the PsPIN1 gene results in asymmetric polar auxin transport between the proximal and distal sides of the subapical region of etiolated pea epicotyls. When pea seeds were germinated and grown with TIBA instead of water, the graviresponse of epicotyl was slightly inhibited due to the inhibition of polar auxin transport (Fig. 2) and the amount of PsPIN1 in the proximal side of epicotyls was substantially reduced (Fig. 6d–g). An asymmetric distribution of PsPIN1 between the proximal and distal sides of the subapical region of epicotyls was also disturbed (Fig. 6g). These results also suggest that PsPIN1 gene expression and distribution of PsPIN1 leading to asymmetric polar auxin transport are required for graviresponse in etiolated pea seedlings. The fact that an agravitropic pea mutant ageotropum, in which gravitropism of the epicotyl is inhibited, showed reduced polar auxin transport compared with that of Alaska pea (Hoshino et al. 2007; Miyamoto et al. 2014), strongly supports the suggestion mentioned above.

Fig. 6
figure 6

PsPIN1 localization in the subapical region of epicotyls of etiolated pea seedlings grown in the presence or absence of TIBA. Etiolated pea seedlings were grown in the absence (a–c) or presence of TIBA (d–f). Sample preparations are the same as in Fig. 5. b, e Enlarged views of localization of PsPIN1 in the proximal side of the subapical region of the epicotyl. c, f Enlarged views of localization of PsPIN1 in the distal side of the subapical region of the epicotyl. g Relative amount of PsPIN1 in the proximal and distal sides of the subapical region of the epicotyl. Values are expressed as the relative value to the amount of PsPIN1 in the proximal side of non-treated seedlings. Fluorescent values of PsPIN1 in five independent experiments were measured for statistical analysis. Values represent mean ± SE, **P < 0.01 by a Student’s t test. en endodermal cells, vb vascular bundle. Bars are 200 µm in a, d, and 20 µm in b, c, e, f

Full size image
Fig. 7
figure 7

PsPIN1 localization in the apical hook region of etiolated pea seedlings. Etiolated pea seedlings were grown in the absence (a–c) or presence of TIBA (d–f). Sample preparations are the same as in Fig. 5. The boxes in a and d correspond to b, c, e and f. b, e Enlarged views of localization of PsPIN1 in the proximal or inner side of the apical hook. c, f Enlarged view of localization of PsPIN1 in the distal or outer side of the apical hook. g Frequency of PsPIN1 localization in endodermal cells of the proximal and distal sides of the apical hook of etiolated pea seedlings grown in the presence or absence of TIBA. Three patterns of PsPIN1 localization were classified: that in the basal side only, that in the basal and lower sides, and other localization. More than 100 cells of pea seedlings in five independent experiments were measured for statistical analysis. Values represent mean ± SE. Different letters within the same localization mean a significant difference at P < 0.01 by a Student’s t test. en endodermal cells, vb vascular bundle. Bars are 200 µm in a, d, and 20 µm in b, c, e, f

Full size image

In addition, we found interesting localizations of PsPIN1 induced by TIBA; TIBA reduced PsPIN1 level in epicotyl, but not in apical hook region (Figs. 6, 7). Whereas the mechanism is unclear, the difference is possible due to differences in growth feature of these organs; epicotyl has already risen upright and will only elongate straight in the future, but the apical hook must rise from bending and turn the leaf straight up. It is known that TIBA stabilizes actin filaments, inhibiting auxin transport (Dhonukshe et al. 2008; Kojo et al. 2014). Actin filaments also act as a rail along which PIN protein is transported to the plasma membrane via vesicle trafficking (Geldner et al. 2001). Thus studies on actin dynamics in apical and elongation regions will be required in relation to PIN localization as well as polar auxin transport. When TIBA is applied to pea seedlings in the absence of gravitational stimulus like 3-dimensional clinostat, the seedlings show unique morphogenesis like response to gravity (Hoshino et al. 2007). As described in Introduction, it is worthwhile to study the mechanisms and functions of TIBA not only in relation to altered localization of PsPIN1 in plasma membrane of etiolated Alaska pea seedlings but also relationship between auxin transport/movement and morphogenesis in plants as already reported by Okada et al. (1991).

Localization of PsPIN1 in the proximal side of the apical hook was almost observed in the basal side of the plasma membrane of endodermal cells, and this localization pattern was the same as in the epicotyls (Fig. 7b). In contrast, PsPIN1 proteins in the distal side of the apical hook were localized in the basal side and lower side (inside the bending) of the plasma membrane (Fig. 7c). The lateral localization pattern of PsPIN1 in the distal side was clearly reduced by TIBA treatment, whereas the basal localization pattern was significantly increased (Fig. 7e–g). This localization of PsPIN1 in the distal side of the apical hook is apparently adequate to enhance auxin transport/movement to the proximal (concave/inner) side from the distal (convex/outer) side, and thus enhances auxin transport/movement from the site of auxin synthesis to the basal parts. These results together with the fact that TIBA did not affect the expression of gene encoding PsPIN1 protein (Hoshino et al. 2006, 2007) strongly suggest that TIBA at least affects translation, but not transcription and/or actin dependent recycling of PsPIN1 in the endodermal cells of etiolated pea seedlings, resulting in automorphogenesis-like growth and development in etiolated pea seedlings by reducing polar auxin transport. As TIBA might affect vesicular trafficking related to PsPIN1 recycling specific to lateral localization, the lateral localization of PsPIN1 significantly decrease in the distal side as well (Fig. 7g). Also, these localization patterns of PsPIN1 in the distal side in TIBA treated Alaska pea seedlings might be related to the altered angles of apical hook in the seedlings (Fig. 2a), resulted in the slight opening of apical hook in TIBA treatment (Fig. 2a). Although NPA did not directly affect PIN proteins or PIN cycling (Kim et al. 2010; Petrásek et al. 2002), TIBA has been shown to interfere with both steps of PIN protein cycling, which entail constitutive cycling between the endosomes and plasma membrane that depends on actin cytoskeleton and the internalization of PIN proteins into so-called “Brefeldin-A compartments” (Dhonukshe et al. 2008; Michniewicz et al. 2007).

As a result, the asymmetric distribution of auxin in the proximal and distal sides of epicotyls may be caused by asymmetric localization of PsPIN1 upstream of auxin transport in etiolated pea seedlings. Change in auxin transport dependent on the localization of PIN proteins has been shown in Arabidopsis and cucumber seedlings. Localization of AtPIN3 in the root columella cells of Arabidopsis changed in response to the direction of gravity (Friml et al. 2002, 2003). The localization pattern of CsPIN1 changed in the upper and lower endodermal cells in the transition zone between the hypocotyl and roots in cucumber seedlings during gravimorphogenesis, and CsPIN5 changed the localization pattern in the epidermal cells for hydrotropism of roots (Morohashi et al. 2017; Yamazaki et al. 2016). It is thus possible that the amount of auxin in the proximal side of epicotyls is increased via unique localization of PsPIN1 in the distal side of the apical hook.

We have already carried out the space experiments using the Japanese experimental module “Kibo” in the International Space Station (ISS) to clarify the molecular mechanisms of gravity-controlled auxin transport/movement and its relation to growth and development in pea seedlings. Novel interesting findings of this study will contribute to elucidate the practices and procedures for an appropriate posture of land plants, efficient plant cultivation, and food production in space as well as on the earth. Raising an anti-PsPIN1 antibody became an invaluable tool in this study. More detailed findings of this space experiment will be reported elsewhere in the near future.

In conclusion, this is the first report concerning the relationship between morphogenesis and auxin efflux facilitator protein using a novel pea specific anti-PsPIN1 antibody. Our findings suggest that asymmetric and/or lateral auxin transport/movement and the accumulation of endogenous auxin in the proximal and distal sides of epicotyls in etiolated pea seedlings could be controlled by an asymmetric distribution of PsPIN1 auxin efflux carrier protein.