SEARCH

SEARCH BY CITATION

Keywords:

  • pollen tube;
  • spermidine;
  • Ca2+ channel;
  • H2O2;
  • polyamine oxidase

Summary

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References
  9. Supporting Information

Spermidine (Spd) has been correlated with various physiological and developmental processes in plants, including pollen tube growth. In this work, we show that Spd induces an increase in the cytosolic Ca2+ concentration that accompanies pollen tube growth. Using the whole-cell patch clamp and outside-out single-channel patch clamp configurations, we show that exogenous Spd induces a hyperpolarization-activated Ca2+ current: the addition of Spd cannot induce the channel open probability increase in excised outside-out patches, indicating that the effect of Spd in the induction of Ca2+ currents is exerted via a second messenger. This messenger is hydrogen peroxide (H2O2), and is generated during Spd oxidation, a reaction mediated by polyamine oxidase (PAO). These reactive oxygen species trigger the opening of the hyperpolarization-activated Ca2+-permeable channels in pollen. To provide further evidence that PAO is in fact responsible for the effect of Spd on the Ca2+-permeable channels, two Arabidopsis mutants lacking expression of the peroxisomal-encoding AtPAO3 gene, were isolated and characterized. Pollen from these mutants was unable to induce the opening of the Ca2+-permeable channels in the presence of Spd, resulting in reduced pollen tube growth and seed number. However, a high Spd concentration triggers a Ca2+ influx beyond the optimal, which has a deleterious effect. These findings strongly suggest that the Spd-derived H2O2 signals Ca2+ influx, thereby regulating pollen tube growth.


Introduction

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References
  9. Supporting Information

After landing on a stigma, a compatible pollen grain hydrates, germinates and grows through intracellular spaces in the pistil, while approaching the ovary for fertilization. This complex process is regulated by several factors, e.g. calcium (Franklin-Tong, 1999). The small aliphatic polycationic polyamines (PAs) putrescine (Put), spermidine (Spd) and spermine (Spm) have been linked to a plethora of physiological processes in plants, such as growth and development, and stress responses (Kusano et al., 2008). Pollen exhibits high titers of PAs, as well as high activities of biosynthetic enzymes (Song et al., 2002); PAs have also been proposed to play a role in the regulation of pollen tube growth (Bagni et al., 1981; Prakash et al., 1988; Chibi et al., 1994; Song et al., 2001, 2002; Song and Tachibana, 2007; Antognoni and Bagni, 2008). However, the functional mechanism(s) underlying the effect of PAs on pollen tube growth remains unknown.

The tip-specific cytosolic free Ca2+ ([Ca2+]cyt) gradient plays a pivotal role in controlling pollen tube elongation (Malho et al., 1995; Franklin-Tong, 1999; Iwano et al., 2009). The influx of extracellular Ca2+ is involved in the maintenance of the Ca2+ gradient (Malho et al., 1995; Holdaway-Clarke et al., 1997; Franklin-Tong, 1999; Cheung and Wu, 2008). Prior studies in this laboratory have identified a hyperpolarization-activated Ca2+-permeable channel in the pear pollen tube plasma membrane (Qu et al., 2007). PAs could increase the titers of cytosolic free Ca2+ in plants (Yamaguchi et al., 2007; An et al., 2008). In fact, Spd regulates many ion channels in animals, including Ca2+-permeable channels (Lopatin et al., 1994; Johnson, 1996; Williams, 1997; Bowie et al., 1998; Gomez and Hellstrand, 1999; Lu and Ding, 1999; Fleidervish et al., 2008). Previous work in plants has also shown that Spd inhibits the fast-activated Ca2+ channels in vacuoles (Bruggemann et al., 1998), regulates the inward K+ channels in guard cells (Liu et al., 2000), and affects inward Na+ and K+ channels in the root (Zhao et al., 2007).

In recent years, reactive oxygen species (ROS), particularly hydrogen peroxide (H2O2), have been identified as important second messengers in signal transduction networks: they regulate plant growth and developmental processes, such as cell expansion, polar growth, gravitropism, stomatal aperture and flower development, as well as stress responses (Alvarez et al., 1998; Pei et al., 2000; Coelho et al., 2002; Kwak et al., 2003; Skopelitis et al., 2006; Demidchik et al., 2007; Moschou et al., 2008a,b,c, 2009). ROS are generated by several enzymatic reactions and electron transport systems in the cells. One example of enzymatic reactions is the flavoprotein polyamine oxidase (PAO; EC 1.5.3.3). In plants, Spd can be terminally oxidized by PAO to Δ1-pyrroline, 1,3-diaminopropane (1,3-DAP) and H2O2; alternatively, Spd is converted to Put and H2O2 (Moschou et al., 2008c). Recently, it was shown that the H2O2 generated by the PAO-catalyzed oxidation of Spd regulates responses to abiotic and biotic stresses. Depending on its ‘signature’, H2O2 may signal the expression of tolerance effector genes or activate programmed cell death (PCD) (Moschou et al., 2008b). Moreover, the H2O2 cascade can result in the downstream activation of Ca2+ channels in plants, which may be the central step in many H2O2-mediated processes (Mori and Schroeder, 2004).

As Spd has been linked to the regulation of pollen tube growth, we examined whether there is a causal relationship between Spd and Ca2+ influx in the pollen tube. The results presented herein show that Spd induces an increase in the H2O2 titer in the pollen tube, causing the pollen tube hyperpolarization-activated Ca2+-permeable channel activation to induce the increase of [Ca2+]cyt. Whole-cell and single channel patch-clamp data show that Spd-derived H2O2 induces hyperpolarization-activated Ca2+ currents. To further support this finding, Arabidopsis insertion mutants of the AtPAO3 gene encoding a peroxisomal PAO (Moschou et al., 2008c) were used. These mutants showed significant insensitivity to the Spd-mediated induction of Ca2+-permeable channel activity, impaired pollen tube growth and seed set. Our data support a unique functional mechanism of Spd in regulating pollen tube growth and development.

Results

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References
  9. Supporting Information

Effect of Spd on pollen tube growth

As mentioned previously, PAs have been proposed to play a role in the regulation of pollen tube growth (Bagni et al., 1981, Chibi et al., 1994; Prakash et al., 1988; Song et al., 2001, 2002; Song and Tachibana, 2007). To elucidate the potential mechanism of Spd action on pollen tube growth, we performed in vivo and in vitro experiments using pollen derived from the plants Purys pyrifolia and Arabidopsis thaliana. S-Adenosylmethionine decarboxylase (SAMDC; EC 4.1.1.50) is the rate-limiting enzyme in Spd synthesis (Antognoni and Bagni, 2008). Thus, we treated pollen of the pear variety Imamuraaki with the SAMDC inhibitor mitoguazone (methylglyoxal bis-guanylhydrazone, MGBG, 1 mm) and used it to pollinate styles of the variety Hosui, using untreated pollen as a control. At 72 h post-pollination, most of the pollen tubes of the untreated control had reached the ovary (Figure 1a), whereas nearly all MGBG-treated pollen tubes (n = 9 of 10 pistils) reached about one-third of the length of the corresponding controls (Figure 1b). In accordance with the results from the in vivo experiments, pollen treated with 1 mm MGBG in vitro resulted in a decrease (52 ± 4%) of pollen tube length 6 h post-treatment compared with the control (Figure 1c).

image

Figure 1.  Effect of methylglyoxal bis-guanylhydrazone (MGBG) and spermidine on the growth of pear pollen tubes. (a) and (b) Growth of Imamuraaki pollen tubes in Hosui styles observed by fluorescence microscopy. Control (a) and treated (b) pollen tubes subjected to 1 mm MGBG in vivo. Pollen tube growth was assayed 72 h post-pollination. Scale bar: 50 μm. The white arrow points to the end of the pollen tubes. (c) Pollen tube length 6 h post-culture in vitro with (MGBG) or without (Col) the application of 1 mm MGBG. The values are the means of six independent experiments, each consisting of at least 50 determinations. The error bars represent the SEMs. *Statistical significance (P < 0.05, Student’s t-test). (d) Effect of spermidine on pollen tube growth. The values are the means of six independent experiments, each consisting of at least 50 determinations. Error bars represent the SEMs. #No statistical difference (P > 0.05, Student’s t-test).

Download figure to PowerPoint

We further examined the effect of exogenous Spd on in vitro pollen tube growth by adding it to the culture medium. The addition of 0.1 mm Spd to the culture medium only slightly affected pollen tube length 6 h post-treatment, whereas 1 mm Spd markedly reduced tube length, by 52 ± 5%, at 6 h post-treatment (Figure 1d). We further tested the effect of MGBG and Spd on Arabidopsis pollen tube growth. MGBG of 1-mm concentration significantly decreased pollen tube length at 6 h post-treatment (Figure 1a). In contrast, when applied in the concentration range of 0.01–1.0 mm, Spd exhibited a dose-dependent effect on Arabidopsis pollen tube growth: a lower Spd concentration promoted pollen tube growth, whereas a higher concentration of Spd inhibited pollen tube elongation (Figure S1b). These results can be explained by the effect of Spd on the intracellular titers of [Ca2+]cyt (see below).

Spermidine increases cytosolic-free calcium by stimulating Ca2+-permeable channels

To explore the mechanism by which Spd affects pollen tube growth, we tested whether Spd affects the [Ca2+]cyt in the pollen tube, as the Ca2+ influx plays a pivotal role in the regulation of germination and tube growth. Yellow cameleon 3.1 (YC3.1) is a low-affinity indicator that has been used to monitor [Ca2+]cyt during pollen germination and in papilla cells of Arabidopsis (Iwano et al., 2004, 2009). The Arabidopsis pollen tube [Ca2+]cyt increased 2.04 ± 0.39-fold post-treatment with 1 mm Spd within 8 ± 3 min (Figure 2, n = 24).

image

Figure 2.  Exogenous spermidine increases cytosolic-free calcium in Arabidopsis pollen tubes. The intensity of yellow cameleon fluorescence is displayed with pseudocolors, representing the relative [Ca2+]cyt, as illustrated by the pseudocolors bar. (a) Control (top) and 1 mm spermidine treatment (bottom) in Arabidopsis pollen tubes. Scale bar: 10 μm. (b) Column representation of -fold increase in [Ca2+]cyt induced by 1 mm spermidine in Arabidopsis pollen tubes (P < 0.001, Student’s t-test; n = 24).

Download figure to PowerPoint

To further establish the source of the increased [Ca2+]cyt in pollen tube protoplasts, we studied the activity of hyperpolarization-activated Ca2+-permeable channels in both pear and Arabidopsis pollen following the application of Spd. Typical Ca2+ currents recorded for the plasma membrane of pear pollen tube using the whole-cell patch-clamp configuration are shown in Figure 3a (n = 10). The membrane potential was clamped at 0 mV and stepped to values from −200 to 0 mV at intervals of 20 mV. When 1 mm Spd was perfused to the bath solution for periods of up to 30 min, the currents were not affected (Figure 3b,n = 5). In contrast, when the same concentration of Spd was added to the pipette solution, the magnitude of the Ca2+ currents increased within 10 min of the whole-cell configuration being established (Figure 3c, n = 6). Figure 3d presents a summary of the current–voltage (IV) relationship under control conditions and in the presence of 1 mm Spd in the pipette solution. At −200 mV, 1 mm Spd increased the average whole-cell current from 913 to 1651 pA, an 81 ± 22% increase, when compared with the control level. Using 1 mm Spd in the pipette solution also significantly increased the Ca2+ current in Arabidopsis pollen protoplasts (Figure S2a,b). Furthermore, the Spd-induced Ca2+ current in the Arabidopsis pollen protoplasts showed a dose-dependent response (Figure S2c).

image

Figure 3.  Spermidine induces Ca2+ currents in pear pollen tube protoplasts. (a) Control whole-cell currents of pear pollen tube protoplasts in response to voltage steps from −200 to 0 mV, with 20-mV increments with 10 mm CaCl2 in the bath solution. (b) Ca2+ currents with 1 mm spermidine in the bath solution. (c) Ca2+ currents with 1 mm spermidine in the pipette solution. (d) Current–voltage (IV) relationships from the control (= 10) and 1 mm spermidine in the pipette solution (n = 6). The values are given as means ± SEMs. The difference in Ca2+ currents between the control and the pipette spermidine treatment was statistically significant when the voltage was at −200 mV (P < 0.05, Student’s t-test).

Download figure to PowerPoint

We further examined the nature of ion selectivity of the Spd-stimulated Ca2+ channels in pear by replacing the 10 mm CaCl2 in the bath solution with 10 mM BaCl2 or ZnCl2. The Spd-stimulated currents showed that Ba2+ was a larger charge carrier than Ca2+ (Figure S3). The Spd-stimulated current was poorly permeable to Zn2+ (Figure S3), which acts as a cation channel blocker, and was similar to what the hyperpolarization-activated Ca2+-permeable channels in root cells showed (Demidchik et al., 2007). Gd3+, a broad-spectrum Ca2+ channel blocker, when supplied at a concentration of 100 μm in the bath solution, inhibited the Spd-induced Ca2+ current within 3 min, whereas currents decreased by about 95 ± 1% at −200 mV (Figure S3). In other words, these results indicated that the Spd-induced currents were mediated by specific Ca2+-permeable channels.

Exogenous spermidine indirectly stimulates Ca2+-permeable channel activity

To further investigate the molecular mechanism by which Spd regulates the Ca2+-permeable channels, we used outside-out single patches to test Spd action directly on channel activity. Outside-out patches are used to test the direct action of an agent on channel activity to exclude the effect of the intracellular compartment. In control cells, when holding the voltage at −180 mV, the channel open probability (Po) was approximately 0.30 ± 0.06 (Figure 4a,b, n = 6). Compared with the control, both the internal and external addition of 1 mm Spd could not activate Po (Figure 4a,b, n = 3 and 3, respectively). Therefore, Spd cannot directly stimulate the hyperpolarization-activated Ca2+-permeable channels, suggesting an indirect mode of action that probably requires the coordinated action of one or several cellular factors.

image

Figure 4.  Spermidine indirectly stimulates Ca2+ channel activity in pear pollen tube protoplasts. (a) Outside-out single channel current recorded for the control experiment (n = 6); outside-out single channel current recorded for 1 mm intracellular spermidine treatment (n = 3) and 1 mm extracellular spermidine treatment (n = 3), respectively. All data were recorded at −180 mV. (b) Relative opens probability (Po) of the outside-out single channel under 1 mm intracellular spermidine and extracellular spermidine treatments normalized to control. #No significant difference (P > 0.05, Student’s t-test). The values are given as means ± SEMs.

Download figure to PowerPoint

Exogenous spermidine increases ROS

As the effect of Spd on the Ca2+ channels was indirect, we speculated whether its action could be exerted via H2O2 generated by PAO-mediated Spd oxidation. We therefore tested whether exogenous Spd could increase the H2O2 concentration in pollen tubes and pollen tube protoplasts, and used the ROS fluorescent dye H2DCF-DA to monitor changes in H2O2. The treatment with Spd induced an increase in fluorescence both in the pear and Arabidopsis pollen tubes (Figures 5a and S4a). The relative fluorescence emission increased by 85 ± 25 and 93 ± 22% after treatment with 1 mm Spd for 5 min in the pear and Arabidopsis pollen tubes, respectively (Figures 5c and S4b, n = 15 and 13, respectively): this suggested an enhanced ROS accumulation in the pollen tubes. However, no significant differences were observed between the control and pollen tube protoplasts treated with 1 mm Spd within the first 5 min (Figure 5b,d, n = 17); however, this was expected because protoplasts lack a cell-wall localized PAO (Cona et al., 2006).

image

Figure 5.  Spermidine increases the titers of H2O2 in pear pollen tubes. (a) Accumulation of H2O2 in intact pear pollen tubes and (b) accumulation of H2O2 in pear pollen tube protoplasts detected by H2DCF-DA fluorescence in the absence (left) or presence (right) of 1 mm spermidine. Scale bars: 10 μm. (c) Relative fluorescence of H2O2 after treatment with 1 mm spermidine for 5 min in pear pollen tubes and (d) in pear pollen tube protoplasts (n = 15 and 17; P < 0.05 and P > 0.05, Student’s t-test, respectively).

Download figure to PowerPoint

Hydrogen peroxide activates Ca2+ currents

We further wished to confirm that the key targets of H2O2 signal transduction in pollen are the Ca2+-permeable channels. Previous studies in roots (Demidchik et al., 2007) and guard cells (Pei et al., 2000) demonstrated that H2O2 stimulated the hyperpolarization-activated Ca2+-permeable channels. In pear pollen tube protoplasts, we observed that the currents responded to extracellular 10 mm H2O2 and increased within 5 ± 2 min of establishing the whole-cell patch-clamp configuration (Figure 6a). At −200 mV, 10 mm H2O2 increased the whole-cell current 2.72 ± 0.31-fold compared with the corresponding control levels (Figure 6a, n = 6). Furthermore, the influx current amplitudes triggered by H2O2 increased in a dose-dependent manner (Figure 6b, n = 6), and decreased after longer time periods, suggesting reversibility (Figure 6c).

image

Figure 6.  Hydrogen peroxide induces Ca2+ currents in pear pollen tube protoplasts. (a) Current–voltage (IV) relationship in control pollen and pollen treated with 10 mm H2O2 (n = 6). The values are given as means ± SEMs. The difference in Ca2+ currents between the control and H2O2 treatment was statistically significant when the voltage was −200 mV (P < 0.05, Student’s t-test). (b) Dose-dependence of the H2O2-induced current in pear pollen tube protoplasts (the values are given as means ± SEMs, n = 6). (c) Temporal response of H2O2-activated Ca2+ current in pear pollen tube protoplasts; 0 min, black; 2 min, green; 6 min, purple; 10 min, blue; 16 min, red.

Download figure to PowerPoint

The outside-out single patch clamping results further showed that H2O2 directly stimulates the Ca2+ channels. The channel open probability was increased 2.92 ± 0.33-fold when the patches were treated with 10 mm H2O2 compared with the corresponding untreated controls (Figure 7a,b, n = 3). Thus, H2O2 directly stimulates the hyperpolarization-activated Ca2+-permeable channel activity in pear pollen tubes.

image

Figure 7.  Hydrogen peroxide directly activates a Ca2+ channel in pear pollen tube protoplasts. (a) Outside-out single channel current recorded for control pollen tubes and pollen tubes treated with 10 mm H2O2, respectively. All data were recorded at −180 mV (n = 3). (b) Relative open probability (nPo) of outside-out single channel under 10 mm H2O2 treatments normalized to the control. *Significant difference (P < 0.05, Student’s t-test). The values are given as means ± SEMs.

Download figure to PowerPoint

Regulation of Ca2+-permeable channels by Spd is mediated by H2O2

To further determine whether the regulation of Ca2+-permeable channels by Spd is mediated by H2O2, we used ascorbate, an H2O2 scavenger. The response of the increased currents to intracellular Spd was strongly inhibited by 2 mm ascorbate (Figure 8a, n = 6). Furthermore, when protoplasts were treated with 2 mm ascorbate, the addition of Spd in the pipette solution up to 10 mm did not increase the Ca2+ current (Figure 8a, n = 4). The presence of 1 mm MGBG in the pipette solution also inhibited the Ca2+ current (Figure 8b, n = 7), but this inhibition could be reversed by exogenously supplied 10 mm H2O2 (Figure 8b, n = 6). These data confirm that H2O2 mediates the Spd-induced Ca2+ current.

image

Figure 8.  Effect of ascorbate and mitoguazone on the Ca2+ currents of pear pollen tube protoplasts. (a) Current–voltage (IV) relationship for treatments with 1 mm spermidine (n = 10), 1 mm spermidine plus 2 mm ascorbate (ASA) (n = 6) and 10 mm spermidine plus 2 mm ascorbate (n = 4), respectively. The values are given as means ± SEMs. The difference in Ca2+ currents between the 1 mm spermidine treatment and the 10 mm spermidine plus 2 mm ascorbate treatment was statistically significant at −200 mV (P < 0.01, Student’s t-test). (b) Current–voltage (IV) relationships for treatment with 1 mm mitoguazone (MGBG) (n = 7), and 1 mm MGBG plus 10 mm H2O2 (n = 6), respectively. The values are given as means ± SEMs. The difference in Ca2+ currents between 1 mm MGBG treatment and 1 mm MGBG plus 10 mm H2O2 treatment was statistically significant at −200 mV (P < 0.01, Student’s t-test).

Download figure to PowerPoint

Loss-of-function of AtPAO3 inhibits pollen tube growth and decreases seed set by disruption of the spd-induced Ca2+ current

To further establish that the Spd-activated Ca2+ currents were mediated by PAO-derived H2O2, we analyzed two Arabidopsis PAO mutants of the gene AtPAO3 (At3g59050; Figure 9a,b). The T-DNA or Ds inserts were located in the sixth and eighth exons of the gene AtPAO3, respectively. We previously showed that the AtPAO3 protein is localized to peroxisomes, and catalyzes the sequential oxidation of Spm to Spd and of Spd to Put, to produce H2O2; we also demonstrated that the best substrate for this pathway is Spd (Moschou et al., 2008c). Analysis of the mutant lines showed that plants homozygous for the mutation (Atpao3−/−) do not accumulate the corresponding mRNA (Figure 9c,d). For the following analyses, we named the Atpao3 SALK collection T-DNA insertion mutant line 19 and the Atpao3 Ds element insertion mutant line 44 Atpao3-1 and Atpao3-2, respectively. The homozygous Atpao3-1 and Atpao3-2 plants produced less seeds compared with wild-type plants (Figure S5a,b). We used the Atpao3-1 mutant plant as the male or female in crosses with wild-type plants to investigate whether the Atpao3 mutation affected the male or female function, adopting the protocol published by Jiang et al. (2005). When a homozygous Atpao3-1 plant was used as a male in a cross with a wild-type plant, only a few ovules were fertilized in the silique and developed into seeds (Figure 5a,b). In a reciprocal cross, however, nearly all of the ovules (n = 83 out of 89 siliques) were able to produce viable seeds, as observed in the wild type, indicating that the Atpao3 mutation did not affect female fertility (Figure S5a,b). When we used a heterozygous Atpao3 plant as a male to cross with wild-type plants, the F1 seedlings segregated in a ratio of 1:25.5 (38:968, kanamycin resistant:kanamycin sensitive) instead of the expected 1:1 segregation. If we used the heterozygous Atpao3 plant as a female to cross with wild-type plant, the F1 seedlings showed an average segregation ratio of approximately 1:1 (172:181, kanamycin resistant:kanamycin sensitive). These results indicated the Atpao3 mutation resulted in defects of male gametophytic function, but the female gametophytic function appeared to be normal.

image

Figure 9.  Characterization of two Atpao3 insertion mutant lines. (a) Site of T-DNA and Ds insertion for the Atpao3-1 and Atpao3-2 mutants, respectively. (b) PCR-based confirmation of the insertion sites. (c) RT-PCR detection of the AtPAO3 transcript from flowers of the wild-type, Atpao3-1 (SALK) and Atpao3-2 (WDL) plants. The signals were normalized to actin. (d) RNA gel-blot analysis of AtPAO3 abundance from leaves of wild-type, Atpao3-1 (homozygous lines 19 and 30; SALK) and Atpao3-2 (homozygous lines 40 and 44; WDL) plants.

Download figure to PowerPoint

To investigate the cause of the abnormal seed set, we compared the in vitro and in vivo growth rate of Atpao3 pollen tubes with that of wild-type pollen tubes. The mean pollen tube length of both Atpao3-1 and Atpao3-2 was significantly reduced compared with that of the wild type at 6 h post-culture in vitro (Figure 10a). Furthermore, both the Atpao3-1 and Atpao3-2 pollen tubes were able to germinate and grow on stigmatic cells like the wild-type in vivo (Figure 10b). However, the growth rate of Atpao3 pollen was reduced significantly compared with the wild type, a finding showing similarities to the in vitro result. At 48 h post-pollination, all wild-type pollen tubes reached the bottom of the ovary, whereas both the Atpao3-1 and Atpao3-2 pollen tubes reached only less than half of the transmitting tract length (Figure 10b).

image

Figure 10.  Pollen tube growth of Arabidopsis wild-type and AtPAO3 mutants in vitro and in vivo. (a) In vitro pollen tube growth of Arabidopsis wild-type and Atpao3 mutants 6 h post-culture. All experiments were repeated three times, and each treatment had four replicates. One hundred pollen tubes were measured to determine the pollen tube length for each replicate. Each data point is expressed as the mean ± SEM. The difference in pollen tube length at 6 h post-culture between the wild-type and Atpao3 was statistically significant (P < 0.01, Student’s t-test), but there was no significant difference between Atpao3-1 and Atpao3-2 (P > 0.05, Student’s t-test). (b) Arabidopsis wild-type (24 h post-fertilization) and Atpao3 mutant (48 h post-fertilization) pollen tube growth in vivo: 1, wild type × wild type; 2, Atpao3-1 × Atpao3-1; 3, Atpao3-2 × Atpao3-2; 4, Atpao3-1 × wild type; 5, wild type × Atpao3-1. Scale bar: 500 μm.

Download figure to PowerPoint

We further studied the effect of Spd on the Ca2+ current of Arabidopsis pollen in these Arabidopsis mutants. Activation of the Ca2+ currents by 1 mm Spd was significantly disrupted in both Atpao3-1 and Atpao3-2 mutants (Figure 11a,b, n = 11; Figure S6a, n = 5). However, in both mutants the Ca2+ channel could still be activated by the application of H2O2 (Figure 11c,d, n = 9; Figure S6b, n = 4). These data reveal that the mode of action of Spd in the regulation of pollen tube growth is exerted by the PAO-generated H2O2.

image

Figure 11.  The spermidine-induced Ca2+ current is mediated by AtPAO. (a) Ca2+ current of the Atpao3-1 mutant pollen protoplasts in the absence and presence of 1 mm intracellular spermidine. (b) Current–voltage (IV) relationships from the cells in (a) (P > 0.05, Student’s t-test; n = 11). (c) Ca2+ current of Atpao3 mutant pollen protoplasts in the absence and presence of 5 mm H2O2. (d) Current–voltage (IV) relationships from cells in (c) (P < 0.001, Student’s t-test; n = 9).

Download figure to PowerPoint

Discussion

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References
  9. Supporting Information

The PAs have been linked to numerous plant processes, such as growth and development, as well as responses to stresses, but the mechanism(s) of their action remain(s) largely unknown. Recently, Moschou et al. (2008a,b) showed that salinity induces the release of Spd into the apoplast of tobacco, where it is oxidized by PAO. Depending on their concentration, the H2O2-generated signals activate the expression of tolerance effector genes or induce PCD (Yoda et al., 2003, 2006, 2009). In this work, we show that Spd is also highly involved in the elongation of pollen tubes.

The first line of evidence of the implication of Spd in pollen tube elongation came from the observation that in both pear and Arabidopsis, pollen tube growth was inhibited by the SAMDC inhibitor, MGBG (Figures 1 and S1). SAMDC is the rate-limiting enzyme in Spd synthesis (Antognoni and Bagni, 2008). Overexpression (Wi et al., 2006) or downregulation (Moschou et al., 2008b) of the SAMDC gene in tobacco confirmed the importance of this enzyme in Spd homeostasis and stress tolerance adaptation.

Calcium is a central second messenger modulating pollen tube growth (Malho et al., 1995; Franklin-Tong, 1999; Iwano et al., 2009). Thus, the involvement of Spd in the regulation and fine-tuning of the intracellular Ca2+ titers could result in crosstalk between Spd and pollen tube elongation. Consequently, we strived to identify the potential mechanism by which Spd affects pollen tube growth, by testing whether or not Spd affects the [Ca2+]cyt in pollen tubes. Interestingly, the addition of 1 mm Spd for 8 ± 3 min resulted in a significantly increased [Ca2+]cyt in Arabidopsis pollen tubes (Figure 2); more importantly, Spd was associated with the regulation of pollen tube growth in a dose-dependent manner (Figures 1d and S1b). Lower concentrations of Spd enhanced Ca2+ to levels that promote pollen tube growth, whereas higher Spd led to a further increase in Ca2+ concentration, which seems to exert an inhibitory effect on pollen tube elongation. Similarly, in Arabidopsis pollen tubes, the pharmacological depolymerization of actin controls Ca2+-permeable channels, and results in the supra-optimal increase of endogenous Ca2+ titers in the pollen tube tip, leading to pollen tube growth arrest (Wang et al., 2004).

Ion channels are a newly identified target of PAs in planta (Kusano et al., 2007), they are also an important downstream target of H2O2 (Mori and Schroeder, 2004). Hyperpolarization-activated Ca2+-permeable channels are the predominant Ca2+ influx systems in plants (Foreman et al., 2003), and have been identified in guard cells (Hamilton et al., 2000; Pei et al., 2000), in roots (Kiegle et al., 2000; Very and Davies, 2000; Demidchik et al., 2007) and in pollen (Wang et al., 2004; Shang et al., 2005; Qu et al., 2007; Wu et al., 2007). Interestingly, exogenous Spd significantly induces pollen protoplast Ca2+ currents (Figures 3 and S2). In accordance with the regulatory role of Spd on K+ currents in guard cells (Liu et al., 2000), the Ca2+ currents in pollen increased rapidly when Spd was added to the pipette solution, but the currents were not changed when Spd was added to the bath solution (Figure 3). This suggests that Spd interacts with the Ca2+-permeable channel in the intracellular side, or that the Spd-induced Ca2+ currents are mediated by intracellular component(s).

To further understand the mechanism by which Spd activates the Ca2+-permeable channels, we used outside-out single channels. In this patch-clamp configuration, Spd did not directly activate Ca2+-permeable channels (Figure 4). These results reinforce the view of the existence of cellular component(s), linking Spd to the modulation of ion channel activity, as previously hypothesized (Liu et al., 2000). Spd can be terminally oxidized by PAO to Δ1-pyrroline, 1,3-diaminopropane (1,3-DAP) and H2O2, or Spd can be converted to putrescine with simultaneous H2O2 production (Moschou et al., 2008c) (Figures 5 and S4). Furthermore, the exogenous application of H2O2 directly activates the pollen tube hyperpolarization-activated Ca2+-permeable channels (Figures 6 and 7). A large number of reports have shown that the activation of the plasma membrane hyperpolarization-activated Ca2+-permeable channels is a central step in many H2O2-mediated processes. In guard cells, the ABA-triggered increase in cytosolic Ca2+ resulting in guard cell closure was mediated by H2O2 (Pei et al., 2000; Kwak et al., 2003). In roots, the endogenous ROS generated by NADPH-oxidase (Foreman et al., 2003) and exogenous H2O2 (Demidchik et al., 2007) also increased the net Ca2+ influx through the hyperpolarization-activated Ca2+-permeable channels.

In protoplasts lacking cell walls, the apoplastic PAO (terminal catabolism of Spd) is also lacking, and Spd could not be oxidized to produce H2O2 in the bath solution (Figure 5b). Therefore, the Ca2+ channels could not be activated by Spd added to the bath solution (Figure 3b). However, PAOs are also present intracellularly (Cona et al., 2005; Moschou et al., 2008c), and could oxidize the intracellular Spd to generate H2O2, which could in turn activate the plasma membrane-localized Ca2+-permeable channels. Moreover, the increased apoplastic oxidation of Spd could lead to deleterious effects, as reported previously (Moschou et al., 2008b). Consistent with the previous statement, the exogenous application of Spd at high concentrations halted pollen tube growth. That H2O2 is the signalling molecule in Spd-responsive Ca2+ conductance is further supported by the use of ascorbate, a scavenger of H2O2, which abolishes the Spd-induced activation (Figure 8a).

In Atpao3 mutants lacking AtPAO3, a peroxisomal PAO protein with a profound preference for Spd oxidation (Moschou et al., 2008c), Spd-induced Ca2+ currents are significantly disrupted (Figures 11a,b and S6a) when compared with the wild type, resulting in decreased pollen tube growth rate (Figure 10) and less seed production (Figure S5).

We propose that Spd-induced Ca2+ current is mediated by Spd/PAO-derived H2O2 because: (i) Spd increases [Ca2+]cyt by indirectly activating the hyperpolarization-activated Ca2+-permeable channels; (ii) Spd is oxidized by PAO, generating H2O2; and (iii) H2O2 directly affects the hyperpolarization-activated Ca2+-permeable channel activity in the pollen tube. The Spd-induced activation of the plasma membrane Ca2+-permeable channels in pollen is regulated by the H2O2 derived from the PAO-mediated oxidation of Spd, and results in Ca2+ influx. Taken together, the results presented here indicate that the mechanism of Spd action on the regulation of pollen tube growth probably involves downstream signaling by H2O2, which is generated by the PAO-mediated oxidation of Spd, to activate Ca2+-permeable channels; these in turn determine the intracellular Ca2+ titers, and the latter control pollen tube elongation (Figure S7).

Experimental procedures

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References
  9. Supporting Information

Plant material

Pollen from P. pyrifolia Nakai cv. Hosui and Imamuraaki was collected annually from the Nanjing Agricultural Fruit Experimental Yard and preserved by drying in air at ambient temperature for 12 h, and then stored in silica gel at −20°C. Newly opened Arabidopsis flowers were also used. Inorganic salts, Spd, MGBG, ascorbate and H2O2 were purchased from Sigma-Aldrich (http://www.sigmaaldrich.com).

Isolation and characterization of the AtPAO3 mutant

Two AtPAO3 mutant lines, the Salk line 121288 (Atpao3-1) and the WiscDsLox line 477-480A12 (Atpao3-2), were purchased from the Salk Institute (http://www.salk.edu). The Salk lines are kanamycin resistant. The lines with an insertion in both alleles were analyzed by PCR, using the following primers: FW 5′-GGGTAGTGGTGAAGACAGAGGATG-3′ and RV 5′-GAGTCAATATGTGCCACGTTTCG-3′ for Atpao3-1; and FW 5′-TATCGCCCTGTAACAACACTCTC-3′ and RV 5′-CATCCGTAGGAAGTTTCTGCAAC-3′ or 5′-GTAGTTTGGAGCAGTACCGGTGA-3′ for Atpao3-2.

The mRNA abundance of the AtPAO3 gene was confirmed as previously described (Moschou et al., 2008a–c). For RT-PCR, the primers described above were used, whereas for Atpao3-2 analysis, the primers for Atpao3-1 were used and vice versa. In wild-type plants, both primer pairs produced the same result.

Pollen culture conditions

Mature pear pollen was incubated in liquid culture for germination and growth. The solution contained the following components: 0.55 mm Ca(NO3)2, 1.60 mm H3BO3, 1.60 mm MgSO4, 1.00 mm KNO3, 440 mm sucrose and 5 mm 2-(N-morpholino)ethanesulfonic acid hydrate (MES), pH 6.0–6.2 (the pH was adjusted with Tris). MGBG or Spd was added directly into the medium as required. The pollen was incubated in small Petri dishes at 24 ± 1°C for 6 h. The pollen tube length was measured under a light microscope with Image-Pro (Media Cybernetics, http://www.mediacy.com).

Wild-type and Atpao3 mutant pollen was cultured as described by Boavida and McCormick (2007). Briefly, fresh Arabidopsis pollen was added in liquid germination medium containing 1.60 mm boric acid, 5 mm CaCl2, 5 mm KCl, 1 mm MgSO4 and 440 mm sucrose, pH 7.5 (the latter was adjusted with NaOH). Pollen was incubated in small Petri dishes at 24 ± 1°C for 6 h. The pollen tube length was measured under a light microscope with Image-Pro (Media Cybernetics).

MGBG inhibitor studies and immunofluorescence microscopy

Pear pollen was incubated for 30 min in BK solution containing 1 mm MGBG. Pollen incubated in BK solution without MGBG served as a control. Styles of the Housui variety were pollinated by the treated Imamuraaki pollen, harvested after 72 h, then stained with decolorized aniline blue and examined by epifluorescence.

Pollen protoplast isolation

The preparation of pear pollen tube protoplasts was as described by Qu et al. (2007). Briefly, pollen tubes were cultured for 3 h, washed twice and incubated in enzyme solution for 2.5 h at 32°C to release the spheroplasts. The enzyme solution was pear germination solution containing 1% (w/v) macerozyme R-10 (Onozuka, http://www.yakult.co.jp), 2.0% (w/v) cellulase R-10 (Onozuka), 0.7% (w/v) pectolyase Y-23 (Seishin, Tokyo, Japan) and 1% (w/v) bovine serum albumin (BSA) (Sigma-Aldrich). The enzyme solution was exchanged with a bath solution (0.2 mm glucose, 10 mm CaCl2 and 5 mm MES, adjusted to an osmolality of 800 mosmol L−1 and a pH of 6.0 with d-sorbitol and Tris, respectively).

The preparation of Arabidopsis pollen protoplasts was performed as described previously (Wu et al., 2007). More specifically, flowers were placed in 1 ml of Arabidopsis pollen germination solution in a 1.5 ml tube to release pollen. The cell walls were digested in germination medium containing 1.5% (w/v) cellulase R-10 (Onozuka) and 0.5% (w/v) Pectolyase Y-23 (Seishin) at 25°C for 2 h. The isolated protoplasts were washed twice with bath solution (10 mm CaCl2 and 5 mm MES/Tris, adjusted to 350 mOsm with d-sorbitol, pH 6.0).

H2O2 detection in pollen tube and pollen tube spheroplasts

The production of H2O2 in pollen tubes was examined by loading with 2,7-dichlorofluorescin diacetate (H2DCF-DA; Molecular Probes/Invitrogen, http://www.invitrogen.com), as described by Pei et al. (2000). Pollen was incubated in the liquid germination solution for ∼3 h until the pollen tubes had germinated; then 50 μm H2DCF-DA was added to the floating solution for 20–30 min. Spd was added after loading the dye. The control experiments showed no increase in H2O2 concentration. The pollen tubes were observed under a Leica confocal microscope (http://www.leica.com) (excitation, 485 ± 10 nm; emission, 535 ± 10 nm); images were acquired and the fluorescence of the pollen tubes was analyzed using the Leica confocal software.

Cytosolic-free Ca2+ detection in pollen

YC3.1-expressing Arabidopsis pollen was cultured as described above. After 2–6 h at 22°C, the pollen tube [Ca2+]cyt was monitored using a Zeiss confocal microscope (http://www.zeiss.com). Spd was added during recording. The samples were scanned by an argon laser (excitation 458 nm); images were acquired and the fluorescence of the pollen tube was analyzed using the Zeiss confocal software ZEN 2009 Light Edition.

Characterization of the Atpao3 mutant phenotype

Aniline blue staining of pollen tubes in pistils was performed as described by Ishiguro et al. (2001). The pre-emasculated mature wild-type flowers were pollinated either with wild-type or Atpao3 pollen. The pollinated pistils were collected 48 h after pollination and briefly fixed in a fixing solution of ethanol:acetic acid (3:1) for 4 h at room temperature. The fixed pistils were washed three times (25°C) with distilled water and treated in a softening solution of 8 m NaOH overnight. Then, the pistil tissues were washed in distilled water and stained in aniline blue solution (0.1% aniline blue in 0.1 m K2HPO4-KOH buffers) overnight in the dark. The stained pistils were observed and photographed with an Olympus fluorescence microscope (http://www.olympus-global.com).

Morphological analysis of siliques

For morphological analysis, Arabidopsis ovules from dissected siliques were imaged with a dissecting microscope.

Patch-clamp experiments

The patch-clamp method used was as described by Qu et al. (2007). The standard bath solution contained the following components: 0.2 mm glucose, 10 mm CaCl2 and 5 mm MES, adjusted to an osmolality of 800 and 300 mosmol L−1 for pear and Arabidopsis with d-sorbitol, respectively; the final pH of 6.0 was adjusted with Tris. The pipette solution was prepared using the following reagents: 1 mm MgCl2, 0.1 mm CaCl2, 4 mm Ca(OH)2, 10 mm ethyleneglycol tetraacetic acid (EGTA), 2 mm MgATP, 10 mm Hepes, 100 mm CsCl and 0.1 mm GTP, adjusted to a pH of 7.1 by Tris and an osmolality of 1100 and 350 mosmol L−1 for pear and Arabidopsis with d-sorbitol, respectively. Changes made to the bath and pipette solutions are given in the figure legends.

The patch-clamp pipettes were pulled from borosilicate glass capillaries (WPI, http://www.wpiinc.com) using a microelectrode puller (PB-7; Narishige, http://narishige-group.com), to a resistance of 15–40 MΩ in 10 mm CaCl2 in the bath solution when filled with intracellular solution. Protoplasts were patch clamped with an Axopatch-200B amplifier and digitized using the DigiData 1200 interface and pCLAMP9 software (Axon Instruments, http://www.moleculardevices.com.). The whole-cell configuration was obtained using a short burst of suction applied to the pipette interior to rupture the membrane, resulting in a substantial increase in capacitance. The series resistance and capacitance were compensated accordingly; seal resistances were >1.5 GΩ. The membrane was held at a holding potential of 0 mV, and the voltage clamp protocols either comprised a series of steps of 2.5 sec or were changed rapidly and continuously in a ramp (ramp speed 9.84 mV sec−1). Data were sampled at 2 kHz and filtered at 0.5 kHz. Junction potentials were corrected as described by Qu et al. (2007). The outside-out patch configuration was used to record single-channels currents; the resistance of the pipettes ranged from 10 to 12 GΩ, and data were acquired at 20 kHz and low-pass filtered at 5 kHz. During post analysis, the data were further filtered at 200 Hz. Single-channel events were listed and analyzed by pCLAMP9 (Axon Instruments). We used the 50% threshold cross-method to determine valid channel opening. When multiple channel events were observed in patch, the total number of functional channels (n) in the patch was determined by observing the number of peaks detected on all point-amplitude histograms. The product of the number of channels (nPo) and the open probability, or the open probability (Po) itself, was used to measure the channel activity within a patch. These data were used to confirm the effects of the chemicals. Data are presented as means ± SEMs, and statistical differences were compared by using a Student’s t-test. All experiments were performed at 20–22°C.

Acknowledgements

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References
  9. Supporting Information

We thank Megumi Iwano from the Graduate School of Biological Sciences, Nara Institute of Science and Technology, for providing us with the YC3.1 Arabidopsis seeds. We thank Drs V.E. Frankling-Tong and Natalie Poulter of the School of Bioscience, as well as Drs Yuchun Gu and Su Wang of the School of Medicine, University of Birmingham, UK, for their valuable advice and help in improving the manuscript. We also thank International Science Editing for improving our manuscript. This work was supported by the earmarked fund for Modern Agro-industry Technology Research System (3-38).

References

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References
  9. Supporting Information

Supporting Information

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References
  9. Supporting Information

Figure S1. Effect of MGBG and spermidine on Arabidopsis pollen tube growth. (a) Pollen tube length 6 h post-culture. MGBG of 1 mm inhibited pollen tube growth in vitro. The values are the means of six independent experiments, each consisting of at least 50 determinations. The error bar represents the SEM, (< 0.01, Student’s t-test). (b) Effect of Spd on Arabidopsis pollen tube growth. The values are the means of six independent experiments, each consisting of at least 50 determinations. The error bar represents the SEM.

Figure S2. Spermidine-induced Arabidopsis pollen protoplasts Ca2+ currents. (a) Ca2+ current of Arabidopsis pollen protoplasts in the absence and presence of 1 mm intracellular spermidine. (b) Current–voltage (I–V) relationships from the cells in (a). (< 0.001, Student’s t-test; n = 5). (c) Average whole-cell currents at −200 mV against Spd concentration in Arabidopsis pollen protoplasts. The number of cells averaged was n = 20, 6, 6, 4, 6, 5, 7, 5 at a Spd concentration of 0, 0.01, 0.1, 0.2, 0.4, 0.6, 0.8, 1 mm, respectively.

Figure S3. Spermidine-activated channels are Ca2+-permeable and non-selective in pear pollen tube protoplasts. Currents at −200 mV recorded in 10 mm BaCl2 (n = 4), 10 mm CaCl2 (n = 10), 10 mm MgCl2 (n = 8), 10 mm ZnCl2 (n = 7) and 10 mm CaCl2 plus the Ca2+ channel blocker 100 μm Gd3+ (n = 4) in the bath solution. Spermidine of 1 mm was added in all treatments in the pipette solution.

Figure S4. Spermidine increased H2O2 content in Arabidopsis pollen tubes. (a) ROS production in intact Arabidopsis pollen tubes detected by (H2DCF-AC) fluorescence in the absence (left) or presence (right) of 1 mm Spd. Bar = 10 μm. (b) Column representation of the fold increase in [H2O2] induced by 1 mm Spd recorded in Arabidopsis pollen tubes (< 0.001, Student’s t-test; n = 13).

Figure S5. Seed production from different crosses. (a) 1, Atpao3-1 silique showing less seed production; 2, Atpao3-2 silique showing less seed production as Atpao3-1; 3, Atpao3-1 × wild-type silique showing nearly full seed set. 4, Wild-type × Atpao3-1 silique showing less seed production as Atpao3-1; 5, wild-type silique showing full seed set. (b) Seed ratio of different crosses Atpao3-1× (n = 32); Atpao3-2× (n = 24); Atpao3-1 × Wild-type (n = 17); Wild-type × Atpao3-1 (n = 21); Wild-type × (n = 10).

Figure S6. Spermidine-induced Ca2+ current was mediated by the AtPAO. (a) Current–voltage (I–V) relationships of Atpao3-2 mutant pollen protoplasts in the absence and presence of 1 mm intracellular Spd (= 5). (b) Current-voltage (I–V) relationships of Atpao3-2 mutant pollen protoplasts in the absence and presence of 5 mm H2O2 (n = 4).

Figure S7. Proposed model for spermidine action on pollen tube growth. (a) Inhibition of spermidine (Spd) synthesis abolishes tube elongation, because the Spd-derived H2O2 is not generated and the Ca2+ channels are not activated, thus preventing the necessary Ca2+ influx. (b) Under normal conditions, Spd is efficiently oxidized by PAO generating physiological titers of H2O2 that regulate the activation of Ca2+ channels; this results in increased intracellular Ca2+, which induces elongation of the pollen tube tip. (c) Concentrations above the optimal Spd titers in pollen tubes result in the generation of an exceedingly higher amount of H2O2 that activates Ca2+ channels. However, either the Ca2+ influx is not restricted to the tip, or the influx is so high that deleterious effects are observed. Exogenous Spd application results in the accumulation of H2O2 generated by the action of the PAO, which in turn further increases the endogenous H2O2 content. Greater oxidation of Spd can lead to even more deleterious effects, such as H2O2-induced PCD (Moschou et al., 2008b). All the previous results reinforce the view that the homeostatic regulation of Spd is indispensable for normal pollen tube growth.

As a service to our authors and readers, this journal provides supporting information supplied by the authors. Such materials are peer-reviewed and may be re-organized for online delivery, but are not copy-edited or typeset. Technical support issues arising from supporting information (other than missing files) should be addressed to the authors.

FilenameFormatSizeDescription
TPJ_4301_sm_Legends.doc34KSupporting info item
TPJ_4301_sm_Figures.pdf263KSupporting info item

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.