Apical F-actin-regulated exocytic targeting of NtPPME1 is essential for construction and rigidity of the pollen tube cell wall

Authors

  • Hao Wang,

    1. School of Life Sciences, Centre for Cell and Developmental Biology and State Key Laboratory of Agrobiotechnology, The Chinese University of Hong Kong, Hong Kong, China
    Search for more papers by this author
  • Xiaohong Zhuang,

    1. School of Life Sciences, Centre for Cell and Developmental Biology and State Key Laboratory of Agrobiotechnology, The Chinese University of Hong Kong, Hong Kong, China
    Search for more papers by this author
  • Yi Cai,

    1. School of Life Sciences, Centre for Cell and Developmental Biology and State Key Laboratory of Agrobiotechnology, The Chinese University of Hong Kong, Hong Kong, China
    Search for more papers by this author
  • Alice Y. Cheung,

    1. Department of Biochemistry and Molecular Biology, University of Massachusetts, Amherst, MA, USA
    Search for more papers by this author
  • Liwen Jiang

    Corresponding author
    1. School of Life Sciences, Centre for Cell and Developmental Biology and State Key Laboratory of Agrobiotechnology, The Chinese University of Hong Kong, Hong Kong, China
    2. CUHK Shenzhen Research Institute, The Chinese University of Hong Kong, Shenzhen, China
    Search for more papers by this author

  • Accession numbers: NtPPME1(AY772945), AtVSR2 (NM_128582).
  • Upon request, all novel materials described in this publication will be made available in a timely manner for non-commercial research purposes, subject to the requisite permission from any third-party owners of all or parts of the material. Obtaining any permission will be the responsibility of the requestor.

Summary

In tip-confined growing pollen tubes, delivery of newly synthesized cell wall materials to the rapidly expanding apical surface requires spatial organization and temporal regulation of the apical F-actin filament and exocytosis. In this study, we demonstrate that apical F-actin is essential for the rigidity and construction of the pollen tube cell wall by regulating exocytosis of Nicotiana tabacum pectin methylesterase (NtPPME1). Wortmannin disrupts the spatial organization of apical F-actin in the pollen tube tip and inhibits polar targeting of NtPPME1, which subsequently alters the rigidity and pectic composition of the pollen tube cell wall, finally causing growth arrest of the pollen tube. In addition to mechanistically linking cell wall construction and apical F-actin, wortmannin can be used as a useful tool for studying endomembrane trafficking and cytoskeletal organization in pollen tubes.

Introduction

Growth of the pollen tube is guarded by vigorous polar exocytosis and endocytosis (Taylor and Hepler, 1997; Krichevsky et al., 2007; Cheung and Wu, 2008; Yang, 2008; Cai and Cresti, 2009). Numerous small secretion vesicles which carry newly synthesized plasma membrane (PM) and cell wall materials dock and fuse with the PM around the apical dome to meet the requirement of rapid expansion of the tube cell surface (Krichevsky et al., 2007; Cai and Cresti, 2009; Winship et al., 2011; Kroeger and Geitmann, 2012). However, we still do not know how exocytosis is precisely regulated to meet the demands for tube surface expansion and cell wall construction. The actin cytoskeleton is regarded as one of the most critical components involved in controlling exo/endocytic vesicle trafficking and protein targeting in the growing pollen tube (Vidali and Hepler, 2001; Vidali et al., 2001; Chen et al., 2003; Cheung et al., 2008; Fu, 2010; Bou Daher and Geitmann, 2011). According to the position in the pollen tube, the dynamics and organization of actin microfilaments in the growing pollen tube is spatially different. In the tip region, the apical F-actin filaments are short and organized into a funnel-like mesh network which is essential for exo/endocytic vesicle fusion with the PM and recycling, proper functioning of Rho-GTPase and Ca2+ signaling and gradients (Vidali and Hepler, 2001; Vidali et al., 2001; Gu et al., 2003; Cheung et al., 2008; Lee and Yang, 2008a; Fu, 2010). In contrast, long and fine actin filaments exist in the shank of pollen tubes. However, how apical F-actin regulates exocytosis to control the rigidity of the pollen tube cell wall and pollen tube growth remains largely unexplored.

During pollen tube growth, the apical cell wall has to be strong enough to withstand the internal turgor pressure but must also be sufficiently plastic to integrate membrane and cell wall materials to support the rapidly polarized tip growth (Cheung et al., 2008; Zonia and Munnik, 2008, 2009; Kulich et al., 2010). The distribution of cell wall components in growing pollen tubes is spatially different. The wall in the tip of the pollen tube is made of a single and thin primary cell wall mainly composed of pectin. The cell wall in the pollen tube shank, however, consists of two layers – an inner secondary cell wall sheath of callose and an outer primary cell wall of mainly pectin with cellulose and hemicelluloses (Micheli, 2001; Ridley et al., 2001; Zhang et al., 2010a; Zonia and Munnik, 2011; Hill et al., 2012; Palin and Geitmann, 2012). Previous studies have found that during elongation of the pollen tube, highly methylesterified pectin is packed into and secreted via Golgi-derived vesicles. Thereafter, these are transported to the apex of the pollen tube and subsequently released into the cell wall in a methylesterified state which can be recognized by JIM7 antibodies. Following vesicle discharge, the highly methylesterified ‘soft’ pectins are gradually de-methylesterified by cell wall-associated pectin methylesterases (PMEs) by converting methoxyl groups of the polygalacturonic acid backbone chains of pectin into carboxyl groups. The de-methylesterified pectin residues, which can be detected by JIM5 and LM7 antibodies, are then cross-linked by calcium ions and form a new ‘hard’ pectin layer (Micheli et al., 2000; Micheli, 2001; Bosch and Hepler, 2005b, 2006; Tian et al., 2006; Rockel et al., 2008; Dardelle et al., 2010; Kulich et al., 2010; Zhang et al., 2010b). Recent studies suggest that the enzymatic activity of PMEs is closely associated with the formation and rigidity of the pollen tube cell wall (Bosch and Hepler, 2005b; Rockel et al., 2008). In Nicotiana tabacum, NtPPME1 is secreted into the pollen tube apical apoplast where it catalyzes de-methylesterification of pectin. Not surprisingly, silencing of NtPPME1 leads to retarded growth of tobacco pollen tubes (Bosch and Hepler, 2005b, 2006; Rockel et al., 2008).

Wortmannin, an inhibitor of phosphoinositide 3-kinase (PI3K), phosphatidylinositol 4-kinase (PI4K) and phospholipid synthesis, has been widely used to study membrane trafficking and protein dynamics in eukaryotic cells (Schu et al., 1993; Davidson, 1995; Wurmser et al., 1999; Emans et al., 2002; Wang et al., 2009; Zhang et al., 2011; Shen et al., 2013b). In plant cells, wortmannin blocks protein recycling from the pre-vacuolar compartment (PVC) to the trans-Golgi network (TGN) or Golgi apparatus and induces homotypic fusion of PVCs (Tse et al., 2004; Miao et al., 2006, 2011; Lam et al., 2007a; Wang et al., 2009, 2010a,b; Ding et al., 2012; Gao et al., 2012). In this study, we used wortmannin as a tool to investigate the possible mechanistic linkage between apical F-actin-regulated exocytosis for protein secretion and cell wall construction in tobacco pollen tubes. We demonstrate that disruption of the spatial organization of apical F-actin by wortmannin directly arrests the exocytosis-mediated targeting of NtPPME1 to the pollen tube apex as well as the tip-focused endocytosis. Further immunogold and structural transmission electron microscope (TEM) studies demonstrate that such wortmannin-induced abolition of apical targeting of NtPPME1 alters the pectic composition of the pollen tube cell wall and the wall rigidity.

Results

Wortmannin-induced abolition of NtPPME1 apical targeting alters rigidity and construction of the pollen tube cell wall

Nicotiana tabacum pollen-specific pectin methylesterase (NtPPME1) is believed to be packed together with methylesterified pectin into Golgi-derived exocytic vesicles aimed at the apical surface of the pollen tube, from which NtPPME1 is then released into the apoplast. Thus, NtPPME1 plays an essential role in controlling the rigidity of the pollen tube cell wall by converting ‘soft’ methylesterified pectin into a ‘hard’ de-methylesterified form (Bosch and Hepler, 2005a,b, 2006).

As shown in Figure 1(a) and Movie S1 in the Supporting Information, pollen tubes transformed with NtPPME1-GFP show the typical apical localization for this enzyme (Bosch and Hepler, 2005a,b; Bosch and Hepler, 2006). Quantitative measurement of accumulation of NtPPME1 at the pollen tube apex and pollen tube elongation is given in Figure 1(b). Time-series images of NtPPME1-GFP accumulation in the tube apex were analyzed and compared with pollen tube growth rates. Examination of the fluorescence intensity along the tube surface within 2 μm of the tube apex (indicated by the white box in Figure 1a) revealed that the signal intensity of NtPPME1-GFP changed in an oscillatory fashion with a periodicity similar to that of oscillation in the growth rate (Figure 1b). The concentration of wortmannin used in this study is based on both its dose-related inhibitory effects on pollen tube growth rate and the tip-focused FM4-64 dye dynamics determined in this study (Figure S1) and previous studies in lily, tobacco and Arabidopsis pollen tubes (Wang et al., 2010a, 2011b; Zhang et al., 2010b; Wang and Jiang, 2011a). After treated with wortmannin for 15 min, the fluorescent signal of apically localized NtPPME1-GFP became reduced. Meanwhile, pollen tube growth was inhibited (Figure 1c). After 30 min of wortmannin treatment, the apical localization of NtPPME1 was completely abolished (Figure 1d).

Figure 1.

Abolition by wortmannin of NtPPME1-GFP apical localization in a growing pollen tube.
(a). A representative image of growing pollen tubes expressing NtPPME1-GFP. DIC, differential interference contrast.
(b) Quantitative analysis of the signal intensity of NtPPME1-GFP on the apical surface of the pollen tube and oscillation of tube growth rate. Two hours after bombardment, time-lapse confocal images of the pollen tube were collected with minimal intervals. The GFP signal intensity (left) within 2 μm of the tube apex indicated by white box shown in (a) was compared with tip elongation rate (right). The mean growth rate was 39.51 nm sec−1. Similar results were obtained from four individual experiments.
(c), (d) Representative images of pollen tubes expressing NtPPME1-GFP treated with 8.25 μm wortmannin for 15 min (c) and 30 min (d). Scale bar: 12.5 μm.

Immunoprobing with JIM5, JIM7 and LM7 antibodies allows one to recognize different patterns of methyl esterification on pectic homogalacturonans (HG) in the plant cell wall (Knox et al., 1990; Willats et al., 2000a,b). JIM7 recognized highly methylesterified pectin in tobacco pollen tube tips (Figure 2a). JIM5 and LM7, which probe for de-methylesterified pectin, were only detected distal to the tip of the pollen tube (Figure 2c,e). After 30-min wortmannin treatment, JIM7 could not be detected in the pollen tube tip (Figure 2b). In addition, JIM5 and LM7 epitopes were found throughout the whole tube including the tube apical region (Figure 2d,f). Hypothetical models of the spatial distribution of methylesterified and de-methylesterified pectin in untreated and wortmannin-treated pollen tube cell walls are shown in Figure 2(g,h).

Figure 2.

Immunocytological characterization of pectin distribution of wortmannin-treated and non-treated pollen tubes.
Germinating tobacco pollen tubes were fixed and immunolabeled with JIM7 (a), JIM5 (c) and LM7 (e) antibodies respectively. Thirty-minute 8.25 μm wortmannin-treated pollen tubes were immunostained with JIM7 (b), JIM5 (d) and LM7 (f) antibodies. (g, h) Hypothetical models of esterified and de-esterified pectin distributions in untreated and wortmannin-treated pollen tube cell walls. The light shade represents highly methylesterified pectin. The dark shade indicates de-methylesterified pectin in the shank of the pollen tube. Scale bar: 12.5 μm.

To further study the possible changes at the ultrastructural level caused by the wortmannin-induced mis-localization of NtPPME1-GFP, we performed transmission electron microscopy (TEM) studies on ultra-thin sections prepared from high-pressure frozen/freeze-substituted tobacco pollen tubes treated or not treated with wortmannin. An overview of the untreated tobacco pollen tube is shown in Figure 3(a). An enlarged area of the tip of the pollen tube is presented in Figure 3(b). Compared with the shank region shown in Figure 3(c), the cell wall in the tip of the pollen tube consists of only a thin layer of methylesterified pectic primary cell wall (mpw) whereas the shank tube cell wall is composed of de-methylesterified pectic primary cell wall (dpw) and secondary cell wall (sw). In contrast, wortmannin-treated pollen tubes showed different structures. As shown in Figure 3(d) (an overview of a typical wortmannin-treated pollen tube), formation of large vacuoles was evident and no obvious clear zone with secretory vesicles was observed. Enlarged areas from a wortmannin-treated pollen tube tip and shank are given in Figure 3(e,f). The tip region and shank showed the same two cell wall layers. Thus, the thin single layer of the pectic primary cell wall in the tube tip was completely abolished after wortmannin treatment, which is consistent with the wortmannin-induced abolition of tip-localized NtPPME1 (Figure 1). Taken together, these results indicate that the exocytic trafficking of NtPPME1 to the apical region of the pollen tube is responsible and essential for maintaining the proper pectic cell wall composition and cell wall rigidity in growing pollen tubes.

Figure 3.

Ultrastructure analysis of wortmannin-induced alteration of pollen tube cell wall construction.
(a) Overview of an untreated tobacco pollen tube.
(b) Enlarged apical region of the pollen tube from (a). A thin and single primary cell wall mainly containing methylesterified pectin (mpw) from the pollen tube tip region is shown.
(c) Distal region back from pollen tube tip region. Two layers containing demethylesterified primary cell wall (dpw) and secondary cell walls (sw) are shown. (d) Overview of wortmannin-treated tobacco pollen tube.
(e, f) Enlarged apical and sub-apical region of the pollen tube from (d). The similar two layers of cell wall (primary and secondary cell walls) are shown in both (e, f). M, mitochondria; SV, secretory vesicle. Scale bars in (a, b) are 10 μm; in (b, c, e) and (f) they are 500 nm.

Identification of NtPPME1 apical targeting dynamics by fluorescence recovery after photobleaching analysis

To monitor the secretion dynamics of NtPPME1 to the pollen tube apex, we employed fluorescence recovery after photobleaching (FRAP). We photobleached a median section of the tip region and traced the NtPPME1-GFP recovery caused by exocytic transportation and multiple rounds of recycling of NtPPME1 to and from the apical surface of the pollen tube. A representative example of the FRAP analysis of a pollen tube expressing NtPPME1-GFP is shown in Figure 4 and Movie S2. Before photobleaching, NtPPME1-GFP was concentrated at the apical surface of the growing pollen tube (Figure 4a). The bleached area is indicated by the white circle in Figure 4(b). After photobleaching, the fluorescent signal in the apical area was reduced to about 8% of its original level (Figure 4b). The GFP signals started to gradually reappear in the apical dome within about 101 sec (= 5) after bleaching. The recovery first occurred in the center of the apical dome and then gradually moved laterally, as indicated by the numbers 1–3 (Figure 4c–f). The fluorescent signal recovery rates of the photobleached tip and non-bleached shank cytosol regions were also analyzed (Figure 4i). These results indicate that the dynamics of NtPPME1 apical targeting in the growing pollen tube is via a exocytosis pathway.

Figure 4.

Fluorescence recovery after photobleaching (FRAP) analysis of exocytic targeting of NtPPME1-GFP to the apical surface of pollen tube.
Fluorescence recovery after photobleaching analysis of NtPPME1-GFP in growing tobacco pollen tubes. Photobleaching was performed and recovery was analyzed in the apical surface area of the pollen tube. A series of representative FRAP analysis time-lapse images of a pollen tube expressing NtPPME1-GFP was recorded every 6 sec for 5 min (a–h). The bleached area is marked by the white circle. The labeling numbers in the images indicate the GFP signal recovery direction (from 1 to 3). Note that photobleaching did not affect pollen tube growth. The mean growth rate was 38.37 ± 3.12 nm sec−1 (n = 5) after photobleaching. This rate were similar to what was observed in the unbleached NtPPME1-GFP pollen tubes. (i) Quantitative analysis of FRAP. Fluorescence recovery was measured by calculating the mean GFP signal intensity in the region of interest on the apex surface. Fluorescence on the apex surface area almost recovered in about 185 sec (■), whereas, little fluorescence recovery occurred in the shank cytosol (●). Similar results were obtained from five individual experiments. Scale bar: 12.5 μm.

Apical F-actin regulates exocytic targeting of NtPPME1 to the apical surface of the pollen tube

Spatial organization of the actin cytoskeleton is critical for controlled exo/endocytic vesicle trafficking and protein targeting during pollen tube growth (Vidali and Hepler, 2001; Vidali et al., 2001; Cheung et al., 2008; Fu, 2010; Bou Daher and Geitmann, 2011; Winship et al., 2011). We therefore next assessed the roles of actin microfilaments in exocytic targeting of NtPPME1 and examined the effects of wortmannin on actin microfilaments in tobacco pollen tubes. Pollen tubes expressing NtPPME1-GFP were treated with 2 nm latrunculin B or 0.15 nm jasplakinolide. The spatial targeting of NtPPME1-GFP to the apex of the pollen tube was disrupted 15 30 min after treatment with latrunculin B or jasplakinolide (Figure 5c,e,g,i) when compared with untreated pollen tube (Figure 5a). Meanwhile, pollen tubes expressing GFP-mTalin, an actin microfilament reporter, were treated with latrunculin B and jasplakinolide, respectively. Pollen tubes expressing GFP-mTalin showed long actin bundles in the untreated shank, with a short apical F-actin funnel-like mesh network in the tip region (indicated by the arrow) of the pollen tube (Figure 5b, Movie S3). All the actin microfilaments in the pollen tube were rapidly de-polymerized or fragmented 15–30 min after latrunculin B or jasplakinolide treatment (Figure 5d,f,h,j). Taken together, these results indicated that actin microfilaments regulate the exocytic targeting of NtPPME1 in pollen tubes.

Figure 5.

Exocytic trafficking and apex targeting of NtPPME1 in the pollen tube is dependent on the apical F-actin cytoskeleton.
Pollen tubes expressing NtPPME1-GFP (a) were treated with 2 nm latrunculin B or 0.15 nm jasplakinolide. Time-lapse images of pollen tubes treated with latrunculin B for 10 and 20 min (c and e, respectively) and pollen tubes treated with jasplakinolide for 10 and 20 min (g and i, respectively) are shown.
Pollen tubes expressing the actin cytoskeleton marker mTalin-GFP (b) were treated with 2 nm latrunculin B/0.15 nm jasplakinolide. Time-lapse images of pollen tubes treated with latrunculin B for 10 and 20 min (d and f, respectively) and pollen tubes treated with jasplakinolide for 10 and 20 min (h and j, respectively) are shown. Scale bar: 12.5 μm.

Wortmannin disrupted the spatial structure and organization of F-actin in the tube apex

The effects of wortmannin on the actin cytoskeleton were also examined. Figure 6(a) shows a pollen tube expressing GFP-mTalin. The arrow points to the apical funnel-like F-actin mesh network in the tip region of the pollen tube. After 15–30 min of wortmannin treatment, apical F-actin organization was gradually disrupted and the mesh moved from the pollen tube tip into the more distal sub-apical region, where the structure became punctate like (Figure 6c,e). Time-lapse images monitoring the dynamic changes in the apical F-actin mesh network after wortmannin treatment are shown in Figure S2 and Movie S4. We used this information to generate hypothetical models of apical F-actin cytoskeleton dynamics and exocytic vesicles at the pollen tube tip after wortmannin treatment or no treatment (Figure 6b,d).

Figure 6.

Wortmannin disrupts the spatial organization of apical F-actin in the pollen tube.
(a) GFP-mTalin was transformed and expressed in the pollen tube. The arrow indicates the presence of an apical F-actin mesh network (A) in the tip of the pollen tube. DIC, differential interference contrast.
(b) A hypothetical model of an untreated pollen tube shows the dynamics of apical F-actin-dependent exocytic vesicles carrying NtPPME1 and methylesterified pectins.
(c, e) Pollen tubes expressing GFP-mTalin were treated with 8.25 μm wortmannin for 15 min (c) or 30 min (e). Arrowheads indicate the disruption of the spatial organization of apical F-actin mesh network after wortmannin treatment.
(d) A hypothetical model of disruption of apical F-actin organization by wortmannin in the pollen tube tip. Wortmannin disrupted the spatial organization of apical F-actin from the mesh network in the tip region. This gradually moved to the sub-apical region and became a punctate like structure in the pollen tube. Exocytosis of NtPPME1 is inhibited. Scale bar: 12.5 μm.

The effects of wortmannin in pollen tubes

In yeast and mammalian cells, wortmannin has been extensively used as a PI3K inhibitor (Davidson, 1995; Wurmser et al., 1999; Arighi et al., 2004). To test the function of wortmannin as a specific PI3K inhibitor, the phosphatidylinositol 3-phosphate (PI3P)-specific biosensor GFP-2xFYVE, a fluorescent fusion protein of two FYVE domains which binds to PI3P with high affinity, was used. Stable expressions in tobacco suspension culture BY-2 cells and Arabidopsis plant lines both show full co-localization of GFP-2XFYVE with the late endosome/pre-vacuolar marker AtRABF2 (Wurmser et al., 1999; Vermeer et al., 2006; Lee et al., 2008a). In this study, expression of GFP-2xFYVE in tobacco pollen tubes showed a pattern similar to that observed in Arabidopsis root hair cells (Lee et al., 2008a). Punctate dots of various sizes were distributed throughout the whole growing pollen tube and co-localized with the PVC reporter RFP-AtVSR2. They moved along with the cytoplasmic streaming except in the tip of the tube (Figure 7a, panel 1, Movie S5). After treatment with wortmannin for 15 min, GFP-2xFYVE punctate dots formed enlarged ring-like structures co-localizing with RFP-AtVSR2 (Figure 7a, panel 2). Thirty minutes after the addition of wortmannin the majority of the GFP-2xFYVE fluorescent signal relocated from the enlarged PVCs to the cytosol, while the VSR-labeled enlarged PVCs remained the same (Figure 7a, panel 3). In contrast, LY294002, another commonly used PI3K inhibitor in yeast and mammalian cells, did not cause PVC enlargement in identical experiments (Figure 7a, panel 4), albeit the GFP-2xFYVE fluorescent signal also became more diffuse and the pollen tube clear zone was abolished after treatment for 15 and 30 min (Figure 7a, panel 5). Moreover, wortmannin also induced PVCs to enlarge in Arabidopsis thaliana root cells (Figure 7b) and other plant cell types (Tse et al., 2004; Miao et al., 2006, 2011; Lam et al., 2007a; Wang et al., 2009, 2010b). Different from the previous observed effects of wortmannin on the tip localization of NtPPME1 (Figure 1) and apical F-actin organization (Figure 6), treatment with LY294002 did not alter the polar exocytic targeting of NtPPME1 in tobacco pollen tubes (Figure S3) or disrupt the dynamic organization of F-actin (Figure S4, Movie S6).

Figure 7.

Distinct effects of wortmannin and LY294002 on the dynamics of phosphatidylinositol 3-phosphate (PI3P) and pre-vacuolar compartments (PVCs).
(a) Growing tobacco pollen tubes expressing a PI3P-specific biosensor (GFP-2 × FYVE) or PVC reporter (RFP-AtVSR2) (panel 1) were treated with wortmannin (panels 2 and 3) and LY294002 (panels 4 and 5). GFP-2 × FYVE-labeled vesicles fully co-localized with RFP-AtVSR2 in the pollen tube (panel 1). Pollen tubes expressing both GFP-2 × FYVE and RFP-AtVSR2 were treated with 8.25 μm wortmannin for 15 min. GFP-2 × FYVE-labeled vesicles formed enlarged ring-like vesicles and co-localized with RFP-AtVSR2 (panel 2).
After 30-min of wortmannin treatment, GFP-2 × FYVE failed to label PVCs and became more diffuse, appearing almost cytosolic, while the RFP-AtVSR2-labeled PVCs remained enlarged (panel 3). Pollen tubes expressing both GFP-2 × FYVE and RFP-AtVSR2 were treated with 20 μm LY294002 for 15 min. GFP-2 × FYVE-labeled vesicles remain punctate and co-localize with RFP-AtVSR2 (panel 4). After 30 min of LY294002 treatment, GFP-2 × FYVE failed to label PVCs and became more diffuse, appearing almost cytosolic, but the PVC reporter RFP-AtVSR2 remains punctate (panel 5). DIC, differential interference contrast. Scale bar: 12.5 μm.
(b) Arabidopsis thaliana root cells transiently expressing the PVC reporter (GFP-AtVSR2) via particle bombardment were treated by wortmannin and LY294002 (right panels). The PVCs in the root cell became enlarged and formed ring-like structures after 30 min of wortmannin treatment but remained punctate after 30-min of LY294002 treatment. Scale bar: 50 μm.

Wortmannin and apical exo/endocytosis in pollen tubes

In rapidly growing pollen tubes, active apical endocytosis is coordinated with exocytosis and plays an important role in the control of pollen tube polarity (Camacho and Malho, 2003; Cai and Cresti, 2009; McKenna et al., 2009; Zonia, 2010). The dye FM4-64 is widely used as a tracer to study pollen tube endocytosis. Labeling of living pollen tube cells by FM4-64 dye resulted in a distinct inverted cone-shaped staining pattern in the tube apex, containing both endocytic and exocytic vesicles (Parton et al., 2001, 2003; Camacho and Malho, 2003; Zonia and Munnik, 2008; Wang et al., 2010a; Zonia, 2010).

We found that re-orientation of pollen tube polarity during growth is closely associated with the ‘V’-shaped accumulation of exo/endocytic vesicles in the apex of the growing pollen tube (Figure S5, Movie S7). Therefore, we also used wortmannin to investigate the dynamics of apical exo/endocytosis in tobacco pollen tubes. As shown in Figure 8(a), after adding FM4-64 to growing tobacco pollen tubes, active tip-focused vesicle accumulation was stained by FM dye. After wortmannin treatment of the FM dye-labeled pollen tubes for 10 min, the apical inverted endo/exocytic zone was disrupted (Figure 8b). After wortmannin treatment for 20 min, the ‘comet tail’ of the apical vesicle accumulation region was strongly inhibited (Figure 8d). Finally, the typical ‘V’-shaped apical exo/endocytosis was completely abolished by wortmannin after 30 min of treatment (Figure 8f).

Figure 8.

Wortmannin and LY294002 have distinct inhibitory effects on endocytosis in the pollen tube. Germinating tobacco pollen tubes were stained with FM4-64 (a). Time-lapse confocal images of pollen tubes labeled with FM4-64 were subsequently treated with wortmannin (b, d and f) or LY294002 (c, e and g) for 10–30 min. (h, i) Pollen tubes were treated with 8.25 μm wortmannin or 20 μm LY294002 for 30 min prior to subsequent FM4-64 uptake. DIC, differential interference contrast. Scale bar: 12.5 μm.

Although LY294002 shared strong inhibitory effects on pollen tube apical exo/endocytosis, its action appeared to be different from that of wortmannin. LY294002 caused the FM dye to form a clustered compartment in the sub-apical region of the pollen tube after 10 min of treatment (Figure 8c). By 20 min, the majority of the apical vesicle accumulation region was already abolished and the sub-apical FM dye cluster disappeared (Figure 8e). LY294002 completely inhibited apical endo/exocytosis after 30 min of drug treatment (Figure 8g). In addition, when we first treated growing pollen tubes with wortmannin or LY294002 for 30 min prior to adding FM dye, wortmannin could inhibit endocytosis in the apex of the pollen tube but not in tube shank because FM4-64 still could be taken up into the pollen tube (Figure 8h). On the other hand, LY294002 blocked endocytic processes in the pollen tube and uptake of the FM dye ceased completely (Figure 8i).

Discussion

Spatial organization of apical F-actin is crucial for exocytosis-mediated pectic cell wall construction in the pollen tube

The actin cytoskeleton is critical for pollen tube growth because organelle movement, vesicle trafficking and signaling networks are all dependent on the spatial organization and dynamics of actin (Vidali and Hepler, 2001; Vidali et al., 2001; Chen et al., 2003; Cheung et al., 2008; Lee et al., 2008b; Fu, 2010; Bou Daher and Geitmann, 2011). In the tip of the pollen tube the actin microfilaments are much shorter and organized as a funnel-like mesh network which mediates exocytic/recycling vesicle fusion with the PM and endocytosis (Chen et al., 2003; Gu et al., 2003; Ren and Xiang, 2007; Cheung et al., 2010; Fu, 2010; Bou Daher and Geitmann, 2011). In this study, we demonstrated that wortmannin disrupted the spatial organization of apical F-actin, resulting in the failure of exocytic trafficking and targeting of NtPPME1 to the surface of the pollen tube apex. This subsequently resulted in the alteration of cell wall construction and rigidity in pollen tubes.

In the tobacco pollen tube, NtPPME1 is packed and exocytosed together with highly methylesterified pectins coming from the Golgi apparatus. The NtPPME1 is inactivated during intracellular trafficking to prevent de-esterification of pectin prior to it reaching the tip cell wall (Micheli, 2001; Bosch and Hepler, 2005b; Bosch et al., 2005a). A different distribution of NtPPME1 activity in the pollen tube can be monitored by using different antibodies to detect methylesterified pectins (JIM7) in the tip region and de-methylesterified ones (JIM5 and LM7) in the shank. After wortmannin has disrupted apical F-actin organization and subsequently abolished exocytic transport of NtPPME1 and methylesterified pectins, the pectic cell wall composition of the pollen tube was modified (Figures 2 and 3). The remaining NtPPME1s in the apical apoplasm might change the remaining methylesterified pectins into the de-methylesterified form, causing the JIM7 epitopes to be missing from tip of the pollen tube, whereas JIM5 and LM7 still label the whole tube.

Apical F-actin-dependent net balance of exocytosis and endocytosis is essential for pollen tube growth

Rapid pollen tube growth and cell polarization require a balance between apical endocytosis and exocytosis (Krichevsky et al., 2007; Lee and Yang, 2008a; Zonia and Munnik, 2011). When only apical endocytosis was inhibited without affecting exocytosis and hydrodynamic streaming, the shape of the pollen tube tip became swollen or balloon-like (Zonia and Munnik, 2008). Precise regulation of exocytosis and endocytosis by apical F-actin guards polar pollen tube growth. Disruption of the spatial organization of apical F-actin by wortmannin subsequently inhibited apical exo/endocytic vesicle accumulation and motion (Figures 6 and 8). Recent studies of plant ROP superfamily proteins have demonstrated that ROP1 is mainly localized to the apical surface of the pollen tube where tube growth takes place and controls exocytic vesicle targeting via regulation of apical F-actin at the tube tip (Gu et al., 2003; Kost, 2008; Lee et al., 2008a,b). Moreover, FRAP analysis of NtPPME1 (Figure 4 in this study) and receptor like kinase (RLK) which is targeted to the PM via exocytosis in the growing pollen tube (Lee et al., 2008b) reveals that the exocytosis targeting site is in the middle of the tube apex. However, studies using pulse-chase of styryl FM dyes and time-lapse imaging with differential interference contrast microscopy show that apex is the site for endocytosis while exocytosis occurs in the sub-apical region (Zonia and Munnik, 2008, 2011). Future studies are needed to identify the sites for exo/endocytosis and characterize distinct populations of trafficking vesicles at the tube tip. These studies may require identification of new markers and the use of advanced image analysis tools (e.g. super-resolution optical microscopy or variable-angle epifluorescence microscopy).

Tip-focused and shank endocytosis in the pollen tube show different sensitivities to wortmannin

FM dye has been commonly used as a reliable marker to monitor endocytosis in plant cells (Camacho and Malho, 2003; Cheung and Wu, 2007, 2008; Lee and Yang, 2008a; Yang, 2008; Zonia, 2010; Zonia and Munnik, 2011). In pollen tubes, the apical V-shaped high-fluorescent staining pattern of FM dye indicates that a high rate of membrane endocytosis and recycling is occurring in the tip (Figure 8). Apical exo/endocytosis is closely associated with the re-orientation of the pollen tube growth axis and accomplishment of pollen tube polarization (Figure S5). Recent studies suggest that endocytosis at the apical cell surface is coupled to ROP GTPases, exocytosis, hydrodynamic fluxes and Ca2+ gradients to establish, maintain and re-orientate pollen tube polarity (Camacho and Malho, 2003; Chen et al., 2003; Gu et al., 2003; Holdaway-Clarke et al., 2003; Kost, 2008; Lee and Yang, 2008a; Cheung and Wu, 2011; Hill et al., 2012). Apical endocytosis is one of the determining factors controlling the polarization and growth of pollen tubes. However, in addition to active apical endocytosis, endocytosis also occurs in the distal region behind the apical dome. Along the shank of the pollen tube, endocytic vesicles are considerably larger than the apical ones under the electron microscope. Studies following the take-up of charged nanogold particles into tobacco pollen tubes demonstrated the existence of two distinct endocytic pathways (Moscatelli et al., 2007). The different sensitivities of endocytosis in the pollen tube tip and shank to wortmannin further supports the notion that there are two distinct endocytic pathways in the pollen tube. Compared with the inhibitory effects of wortmannin on endocytosis in tobacco BY-2 cells, the pollen tube provides an ideal cell system for separating at least two different types of endocytosis of which only one – V-shaped apical endocytosis at the tip – is inhibited by wortmannin (Matsuoka et al., 1995; Emans et al., 2002; Tse et al., 2004; Moscatelli et al., 2007).

Wortmannin is a useful tool for studying endomembrane dynamics in pollen tubes

In yeast and mammalian cells, wortmannin has been extensively used as a PI3K inhibitor that affects multiple steps in membrane trafficking (Davidson, 1995; Wurmser et al., 1999; Arighi et al., 2004). Wortmannin is used to inhibit both protein recycling to the PM and protein delivery to the lysosome/vacuole. Interestingly, low concentrations of wortmannin (e.g. 10–25 nm) inhibited the endocytic trafficking of transferrin receptors in mammalian cells but did not affect Vps34p endocytosis in yeast (Shpetner et al., 1996; Spiro et al., 1996). In addition, high concentrations of wortmannin (e.g. 50–100 μm) also inhibited the endocytosis of various multiple enzymes including PIP4 kinase, myosin light chain kinase (MLCK) and mitogen-activated protein kinase (MAPK) in mammalian cells (Nakanishi et al., 1992; Sasaki et al., 1995; Carnero and Lacal, 1998; Xie et al., 1999; Zhao and Herness, 2009).

In plant cells, wortmannin has commonly been used to block protein recycling from the PVC to the TGN or Golgi apparatus, to inhibit endocytosis and to induce enlargement of the PVC (Jiang and Rogers, 1998; Tse et al., 2004; Miao et al., 2006; Lam et al., 2007a,b; Wang et al., 2009, 2010a,b; Cai et al., 2011). However, unlike the observation in mammalian cells, wortmannin treatment in plant cells fails to reduce the binding of sorting nexin SNX1 to endosomes (Jaillais et al., 2008).

In this study, both wortmannin and LY294002 were shown to inhibit PI3K activity and change the expression patterns of GFP-2xFYVE, a specific biosensor for PI3P, from a punctate pattern into a diffuse cytosolic pattern in growing pollen tubes (Figure 7). However, in pollen tubes expressing the PVC marker RFP-AtVSR2, the dynamic responses of PVCs to wortmannin and LY294002 treatments were different – only wortmannin caused PVC vacuolation. In addition, wortmannin was also shown to disrupt the spatial organization and proper function of the apical F-actin mesh network in the tip of the pollen tube, thus blocking the correct exocytic targeting of NtPPME1 and resulting in alteration of cell wall construction and inhibition of apical endocytosis. However, the mechanistic link between wortmannin and apical F-actin in plant cells remains elusive. In mammalian cells, through a Cdc42/Rac1-independent mechanism, PI3K associated with the family of p21-activated kinase 1 (PAK1) to regulate PAK1 kinase activity, leading to reorganization of the actin cytoskeleton (Papakonstanti and Stournaras, 2002). To further illustrate the underlying mechanism of wortmannin-disrupted apical F-actin in the growing pollen tube, identification of PI3K-associated/interacting proteins involved in the regulation of actin organization might be one of the areas for future studies.

Experimental Procedures

Plant materials and pollen tube germination

Tobacco (N. tabacum) plants were grown in the greenhouse at 22°C under a light cycle of 12 h light and 12 h darkness. Fresh pollen grains were harvested before use. For pollen tube germination, pollen grains were suspended in tobacco pollen-specific germination medium containing 10% sucrose, 0.01% boric acid, 1 mm CaCl2, 1 mm Ca(NO4) 2H2O, 1 mm MgSO4 7H2O, pH 6.5 at 27.5°C for 2 h.

Drug treatment and antibodies

Stock solutions of wortmannin (1 mm in DMSO; Sigma-Aldrich, http://www.sigmaaldrich.com), LY294002 (4 mm in DMSO; Sigma-Aldrich), brefeldin A (1 mm in DMSO; Sigma-Aldrich), FM4-64 (4 mm in DMSO; Invitrogen, http://www.invitrogen.com), latrunculin B (10 μm in DMSO; Calbiochem,http://www.millipore.com), jasplakinolide (4 μm in DMSO, Sigma-Aldrich) were prepared and stored at −20°C. These drugs were diluted in germination medium to appropriate working concentrations before incubation with germinating pollen tubes. For each drug treatment, germinating pollen tubes were mixed with drugs in working solutions in germination medium at a 1:1 ratio to minimize sample variation. The JIM5, JIM7 and LM7 antibodies were purchased from the Paul Knox Cell Wall Lab at the University of Leeds (PlantProbes, http://www.plantprobes.net/index.php). Monoclonal anti-rat antibodies (Sigma-Aldrich) were used for immunofluorescent detection as secondary antibodies.

Particle bombardment of tobacco pollen

Fresh anthers were harvested from 10–12 tobacco flowers and transferred into 20 ml of pollen germination medium. They were vortexed vigorously to release pollen grains into the germination medium. Pollen grain preparation and subsequent transient expression of proteins in the growing pollen tube via particle bombardment were carried out as previously described (Kost et al., 1998; Wang et al., 2010a, 2011b; Wang and Jiang, 2011a; Zhang et al., 2011).

Electron microscopy of the resin-embedded pollen tube

The general procedures for TEM sample preparation and ultra-thin sectioning of samples were as previously described (Tse et al., 2004, 2006; Lam et al., 2007a; Wang et al., 2010a; Gao et al., 2012; Shen et al., 2013a). For pollen tubes, high-pressure freezing and substitution were carried out as previously described (Wang et al., 2010a). London Resin White sections were prepared followed by aqueous uranyl acetate/lead citrate post-staining and examination in a Hitachi H-7650 transmission electron microscope with a CCD camera (Hitachi High Technologies, http://www.hitachi-hitec.com/global/) operating at 80 kV.

Immunofluorescence labeling and confocal imaging

Fixation and preparation of tobacco pollen tubes and subsequent labeling with antibodies and analysis by immunofluorescence were performed as previously described (Tse et al., 2004; Lam et al., 2007a; Wang et al., 2010a, 2011b; Miao et al., 2011). Confocal observation and image collection were performed as previously described (Jiang and Rogers, 1998; Wang et al., 2010a, 2011b, 2012; Cai et al., 2011, 2012; Miao et al., 2011).

Acknowledgements

We thank Dr Sheila McCormick (Plant Gene Expression Center and University of California, Berkeley) for sharing the Lat52 promoter. This work was supported by grants from the Research Grants Council of Hong Kong (CUHK466309, 466610, 466011, 465112, CUHK2/CRF/11G, HKUST10/CRF/12R and HKBU1/CRF/10), NSFC/RGC (N_CUHK406/12), NSFC (31270226), the Shenzhen Peacock Project (KQTD201101) and CUHK Schemes to LJ, and the US National Foundation of Science (MCB0618339) to AYC. The authors declare that they have no conflicts of interest.

Ancillary

Advertisement