Identification of hyperpolarization-activated calcium channels in apical pollen tubes of Pyrus pyrifolia


Author for correspondence: Shao-ling Zhang Tel: +86 025 84396580 Fax: +86 025 84395262 Email:


  • • The pollen tube has been widely used to study the mechanisms underlying polarized tip growth in plants. A steep tip-to-base gradient of free cytosolic calcium ([Ca2+]cyt) is essential for pollen-tube growth. Local Ca2+ influx mediated by Ca2+-permeable channels plays a key role in maintaining this [Ca2+]cyt gradient.
  • • Here, we developed a protocol for successful isolation of spheroplasts from pollen tubes of Pyrus pyrifolia and identified a hyperpolarization-activated cation channel using the patch-clamp technique.
  • • We showed that the cation channel conductance displayed a strong selectivity for divalent cations, with a relative permeability sequence of barium (Ba2+) ≈ Ca2+ > magnesium (Mg2+) > strontium (Sr2+) > manganese (Mn2+). This channel conductance was selective for Ca2+ over chlorine (Cl) (relative permeability PCa/PCl = 14 in 10 mm extracellular Ca2+). We also showed that the channel was inhibited by the Ca2+ channel blockers lanthanum (La3+) and gadolinium (Gd3+). Furthermore, channel activity depended on extracellular pH and pollen viability.
  • • We propose that the Ca2+-permeable channel is likely to play a role in mediating Ca2+ influx into the growing pollen tubes to maintain the [Ca2+]cyt gradient.


It has been clearly established that a tip-to-base free cytosolic calcium ([Ca2+]cyt) gradient is essential for pollen tube growth (Felle & Hepler, 1997). The existence of the [Ca2+]cyt gradient in growing pollen tubes has been repeatedly demonstrated by ratiometric Ca2+ imaging techniques in many plant species (Pierson et al., 1994; Franklin-Tong et al., 1997, 2002; Holdaway-Clarke et al., 1997; Messerli & Robinson, 1997; Pierson et al., 1997). A disruption or modification of the [Ca2+]cyt gradient reversibly inhibits pollen tube growth (Pierson et al., 1994; Malhó & Trewavas, 1996), indicating that the [Ca2+]cyt gradient plays a critical role in modulating polar elongation. Many studies have suggested that a Ca2+ influx through putative Ca2+ channels at the pollen tube apical plasma membrane is responsible for the formation of the [Ca2+]cyt gradient (Pierson et al., 1994; Holdaway-Clarke et al., 1997; Messerli & Robinson, 1997; Messerli et al., 1999). However, the technical difficulties in isolating spheroplasts from elongating pollen tubes, together with the problem that the spheroplasts from pollen tube apices often retain hypersecretory activity, make it difficult to characterize directly channel activities in plasma membranes of pollen tubes (Brownlee et al., 1999). Nevertheless, recent studies with the patch-clamp technique have identified stretch-activated cation channels (Dutta & Robinson, 2004) and hyperpolarization-activated potassium (K+) channels (Griessner & Obermeyer, 2003) in plasma membranes of pollen tubes of Lilium longiflorum.

Two types of voltage-dependent Ca2+-permeable cation channels have been identified in plasma membranes of higher plant cells: depolarization-activated channels (Thuleau et al., 1994; Thion et al., 1998) and hyperpolarization-activated channels (Pei et al., 2000; Véry & Davies, 2000; Shang et al., 2005). Depolarization-activated Ca2+ channels have been suggested to play a role in the transduction of Ca2+-dependent signals, while hyperpolarization-activated Ca2+ channels may be of importance in nutritional acquisition of Ca2+ (Miedema et al., 2001). Véry & Davies (2000) characterized a hyperoplarization-activated Ca2+-permeable channel in the apical plasma membrane of Arabidopsis root hairs. They suggested that Ca2+ conductance may function as a route for local Ca2+ influx into the tip of the root hair, thus contributing to its apex-high [Ca2+]cyt gradient. Like pollen tube growth, root hair elongation is a polarized process in which growth is restricted to the tip. However, whether a similar hyperpolerization-activated Ca2+-permeable channel exists at the apices of elongating pollen tubes remains unknown. In the present study, we isolated spheroplasts from apical pollen tubes of Pyrus pyrifolia cv. Housui (a pear variety that has been widely planted in Japan and China) and identified a hyperpolarization-activated Ca2+-permeable cation channel conductance using the whole-cell patch-clamp configuration.

In addition to calcium, a high proton concentration (low pH, 4.5–6.5) also facilitates both pollen germination and tube growth (Hepler et al., 2001). Whereas there is considerable consensus concerning the presence of [Ca2+]cyt gradients and their relationship to growth, the role of protons remains debatable (Holdaway-Clarke et al., 2003). It has been held that protons may be a more fundamental polarizing regulator than [Ca2+]cyt, but many issues remain to be clarified. Increasing evidence has suggested that proper control of local pH, both apoplastically and symplastically, contributes significantly to polarized growth (see e.g. Hepler et al., 2001). In this study, we observed distinct effects of extracellular pH on pollen tube apical plasma membrane Ca2+ currents.

Materials and Methods

Pyrus pyrifolia Nakai cv. Housui pollen was collected annually from GaoYou Fruit Experimental Yard (JiangSu Province, China). Pollen grains were preserved by drying at air temperature for 12 h and were stored in silica gel at −20°C.

Isolation of spheroplasts from pollen tubes

Pollen was grown at 24°C for 3 h on modified Brewbaker & Kwack medium comprising (in mm): 0.55 Ca(NO3)2, 1.60 H3BO3, 1.60 MgSO4, 1.00 KNO3, 440 sucrose and 5 2-(N-Morpholino)ethanesulfonic acid hydrate (MES)/Tris (pH 6.0–6.2; pH was adjusted by Tris) (Brewbaker & Kwack, 1963). Pollen tubes were washed twice with de-ionized water and incubated in an enzyme solution for 2.5 h at 32°C to release spheroplasts. The enzyme solution was composed of 1% (weight/volume (w/v)) macerozyme R-10, 2.0% (w/v) Cellulase RS-10 (Onozuka, Tokyo, Japan), 0.7% (w/v) Pectolyase Y-23 (Seishin, Tokyo, Japan) and 1% (w/v) bovine serum albumin (BSA) (Sigma, Mexico City, Mexico). The enzyme solution was then exchanged with the control bathing solution.

Experimental solutions

The standard bath solution contained (in mm) 0.2 glucose, 10 CaCl2 and 5 MES, adjusted to an osmolality of 800 mOsm and a pH of 5.8 with d-sorbitol and Tris, respectively. The pipette solution comprised (in mm) 1 MgCl2, 0.1 CaCl2, 4 Ca(OH)2, 10 ethyleneglycoltetraacetic acid (EGTA), 2 MgATP, 10 HEPES, 100 CsCl and 0.1 GTP, adjusted to a pH of 7.3 and an osmolality of 1100 mOsm by Tris and d-sorbitol, respectively. ATP was incorporated to delay rundown of currents (Forscher & Oxford, 1985) and GTP was incorporated to sustain possible G-protein-related activity (Edwards et al., 1989; Yawo & Momiyama, 1993). The free calcium concentration in the pipette solution was approx. 10 nm, calculated with the chemical speciation program geochem (Parker et al., 1987). Changes to bath and pipette solutions are given in the figure legends.

Electrophysiology and data analysis

The pipettes were pulled from borosilicate glass blanks and coated with Sylgard (184 silicone elastomer kit; Dow Corning, Midland, MI, USA). Pipette resistance ranged from 15 to 35 Ω in 10 mm CaCl2. Whole-cell plasma membrane currents were measured using an Axon 200B amplifier (Axon Instrument, Foster City, CA, USA). 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. Series resistance and capacitance were compensated accordingly. The membrane was held at a holding potential of 0 mV and then the voltage was either clamped at discrete values for 2.5 s or changed rapidly and continuously in a ‘ramp’. Voltage protocols for tail current analysis are described in the figure legends. Data were sampled at 2 kHz and filtered at 0.5 kHz, then analyzed using pClamp 9.0 (Axon Instrument). Junction potentials were corrected according to Amtmann & Sanders (1997). All experiments were conducted at room temperature (20–22°C). The permeability of the channels to Ca2+ relative to chlorine (Cl) (PCa/PCl) was estimated using the equation derived from the Goldman–Hodgkin–Katz (GHK) equation (Goldman, 1943; Hodgkin & Katz, 1949):


(ZCl and Zca, the valency of Cl and Ca2+ ions, respectively; [Ca]i and [Cl]i, the activities of intracellular Ca2+ and Cl, respectively; [Ca]o and [Cl]o, the activities of extracellular Ca2+ and Cl, respectively; F, R and T have their usual values (R, 8.14J·mol−1·K−1; F, 96500J·V−1·mol−1; T, 298K). Erev, the measured reversal potential (in volts).)

In addition to the whole-cell recording configuration, the outside-out excised patch configuration was used to record single-channel currents. The resistance of the pipettes ranged from 100 to 120 Ω in 10 mm CaCl2. Data were sampled at 1 kHz and the recording time was 51 s. Bessel filtering was at 2 kHz. Smooth lines were obtained from best fits of data to the Boltzmann equation:

PO/PO(max) = {1 + exp[(Vmax − V1/2)/k]}1

(PO/PO(max), the relative open frequency; Vmax, the membrane potential (the voltage applied); V1/2, the half-maximal voltage of activation; k, the slope factor.) V1/2 and k were determined for each current obtained from individual spheroplasts (Salapatek et al., 2002; Misonou et al., 2004).

Growth analysis

Photomicrographs (Biological Microscopes Motic BA200; Motic, Amoy, FuJian Province, China) were analyzed using motic images advanced 3.2 (Motic) to determine germination rates and pollen tube lengths.

Statistical analysis

Tests were conducted using Student's t-test.


Isolation of spheroplasts from pollen tubes

After incubation of the pollen tubes in the enzymatic solution, spheroplasts were released from the apical regions (Fig. 1). To ensure that spheroplasts were derived from the pollen tube apex, pollen tubes were observed every 10 min during incubation in the enzyme solution and observed during solution exchange. The spheroplasts of the pollen tubes were distinguished from pollen grain protoplasts by their smaller size. The mean diameters of the spheroplasts isolated from the pollen tubes and pollen grain protoplasts were 43.5 ± 1.7 µm (n = 60) and 75.1 ± 2.9 µm (n = 50), respectively. Few pollen grain protoplasts were obtained as it is difficult to remove pollen walls by general enzymolysis, largely because of the many low-molecular-weight compounds (carotenes and flavonoids), lipids and proteins in the exine (Stanley & Linskens, 1985; Chay et al., 1992). Those that were released often remained securely attached to their cell wall (see Fig. 1, protoplasts 1–3) even if enzymolysis was prolonged.

Figure 1.

Enzymatic release of pollen tube spheroplasts and pollen grain protoplasts of Pyrus pyrifolia cv. Housui. Note the extrusion of a spheroplast from a pollen tube apex. Protoplasts from pollen grains (1–4) were greater in diameter than spheroplasts released from pollen tube apices.

Identification and characterization of Ca2+ channels

The patch-clamp whole-cell configuration was used to characterize ionic channels in the plasma membrane of apical spheroplasts derived from the pollen tubes of P. pyrifolia (Fig. 2a). For pollen harvested in 2006, seal resistance was typically 1 GΩ and such seals were obtained in approx. 40% of attempts. With Ca2+ being the only permeant cation in the bath (10 mm), hyperpolarizing pulses of > −100 mV from the holding potential of 0 mV elicited large, time-dependent inward currents (Fig. 2b,c). Application of 15 successive hyperpolarizing voltage ramps (with 2-min intervals between ramps) revealed that the inward Ca2+ currents were stable (Fig. 2d; n = 8). Reversal of the osmotic gradient across the membrane (bath 1200 mOsm, pipette 1100 mOsm) had no effect on the conductance (data not shown).

Figure 2.

Whole-cell recordings of current obtained from the apical spheroplast plasma membrane of Pyrus pyrifolia cv. Housui. (a) A pipette forming a high-resistance electrical seal with the spheroplast plasma membrane. Seal resistance always exceeded 1 GΩ. (b) Whole-cell currents from an individual spheroplast were elicited by sequential step-wise hyperpolariziation of the membrane to −200 mV from a holding potential of 0 mV. The inset depicts the voltage clamp protocol. The bathing solution contained 10 mm calcium (Ca2+). Polarity convention: a downward current deflection is the entry of positive charge into the spheroplast or the exit of negative charge. (c) Ca2+ currents were recorded in response to a hyperpolarizing voltage ramp from 0 to −200 mV (ramp speed 9.84 mV s−1); 10 mm Ca2+ (n = 8). (d) Ca2+ currents elicited by 15 successive voltage ramps (ramp speed 15.19 mV s−1) recorded in 10 mm Ca2+ from one spheroplast. The interval between each ramp was 2 min. Currents were superimposed. (e) Effect of extracellular CaCl2 on the current–voltage (I/V) relationship. Each data point shows mean ± standard error (closed circles, 1 mm, n = 12; open circles, 5 mm, n = 9; triangles, 10 mm, n = 44). The difference in Ca2+ currents between different Ca2+ concentrations was statistically significant when voltage ranged from −200 to −180 mV (**, P < 0.01), and from −160 to −140 mV (*, P < 0.05).

To verify the nature of the currents, the concentration of external CaCl2 was varied. As shown in Fig. 2(e), the hyperpolarization-activated inward currents were markedly reduced when the CaCl2 concentration was reduced from 10 to 1 mm (n = 9–44), indicating that Ca2+ was the charge carrier. The mean (± standard error (SE)) reversal potential (Erev) of the inward current (determined from the ‘tail-currents’) was 22.5 ± 8.7 mV in 10 mm CaCl2 (Fig. 3a,c; n = 9) and decreased to 12.6 ± 4.7 mV when the external CaCl2 was reduced to 1 mm (Fig. 3b,c; n = 7). As the estimated equilibrium potentials for Ca2+ decreased from 174 to 145 mV but those for Cl increased from 41 to 99 mV as external CaCl2 was reduced from 10 to 1 mm, the variational trend in Erev suggests that the inward currents were likely to have been carried by Ca2+ influx rather than Cl efflux. The relative permeability of the channels to Ca2+ and Cl (PCa/PCl) was estimated according to the modified GHK equation using tail-current Erev values (Fig. 3c). A permeability ratio of PCa/PCl of 14 was determined, which suggests that this conductance is selective for Ca2+ over Cl.

Figure 3.

Tail-current analysis reveals the reversal potential of the whole-cell conductance of Pyrus pyrifolia cv. Housui. (a, b) Evidence for a calcium (Ca2+) component in the inward current with 10 mm (a) and 1 mm (b) extracellular CaCl2. After activation of the conductance by a single voltage step from 0 to −200 mV, the voltage was increased in a single step to depolarizing voltages ranging from −10 or −20 mV to 60 mV (see insets). The arrow indicates the reversal current (i.e. no net current is observed); this permits estimation of the reversal potential, the voltage at which no net current passes. (c) Current–voltage (I/V) relationship of the tail currents. The upper and lower arrows indicate the reversal potentials in 1 mm (closed circles, n = 7) and 10 mm (open circles, n = 9) extracellular CaCl2, respectively.

Selectivity of the hyperpolarization-activated inward conductance

To examine whether the hyperpolarization-activated inward conductance was selective for divalent cations, the 10 mm CaCl2 in the bath was substituted with equimolar MgCl2, BaCl2, SrCl2 or MnCl2. By comparing the magnitude of the inward currents from −120 to −200 mV, a permeability sequence of barium (Ba2+) ≈ Ca2+ > magnesium (Mg2+) > strontium (Sr2+) > manganese (Mn2+) was established (Fig. 4; n = 10–16). Inward currents also were observed when 10 mm KCl replaced 10 mm CaCl2 in the bath solution. However, currents were approx. 10 times smaller at −200 mV than those recorded with extracellular CaCl2 or BaCl2 (Fig. 5; n = 9). Therefore, it can be concluded that the hyperpolariszation-activated inward conductance was strongly selective for Ca2+.

Figure 4.

The effect of different extracellular divalent cations on the whole-cell current–voltage (I/V) relationship of apical spheroplasts of Pyrus pyrifolia cv. Housui. Each curve was based on 10 pollen tube spheroplasts except for calcium (Ca2+) (n = 18) and barium (Ba2+) (n = 16). Each data point represents the mean ± standard error recorded in 10 mm extracellular divalent cation. The inward currents were significantly different between Ca2+ and magnesium (Mg2+); Ca2+ and manganese (Mn2+); and Ca2+ and strontium (Sr2+) when the test voltage was between −180 and −200 mV (**, P < 0.01), and between −160 and −120 mV (*, P < 0.05); but they were not significantly different (P > 0.05) between Ba2+ and Ca2+.

Figure 5.

Extracellular potassium (K+) weakly permeates the hyperpolarization-activated conductance. (a) K+ currents were recorded with the whole-cell configuration (10 mm KCl replaced CaCl2 as the charge carrier; the pipette solution contained 100 mm CsCl; n = 21). (b) Comparison of mean (± standard error) current–voltage (I/V) relationships with K+ or calcium (Ca2+) as the charge carrier (closed circles, 10 mm KCl; open circles, 10 mm CaCl2; n = 9). Insets are schematics of the voltage clamping protocol.

Single-channel Ca2+ currents

Further characterization of the Ca2+ channels was conducted at the single-channel level using outside-out excised membrane patches with high-resistance seals (> 5 GΩ) (Fig. 6a). As the voltage was hyperpolarized from −100 to −200 mV, the mean (± SE) magnitude of the inward current was significantly increased from −0.67 ± 0.95 pA to −8.87 ± 3.10 pA (n = 8). The threshold for channel opening was approx. −100 mV (Fig. 6b). In addition to the change in currents, the mean (± SE) opening frequency (PO) was also increased from 0.0413 ± 0.007 to 0.425 ± 0.0360 (Fig. 6c; n = 8). These observations indicate that the activity of the Ca2+ channel depended on voltage. The PO curve was fitted by the Boltzmann equation (Eqn 6), where V1/2 was −163.79145 ± 3.54645 mV, and K was 17.08056 ± 3.46271 (n = 8).

Figure 6.

Single-channel recordings of calcium (Ca2+) currents. (a) Outside-out patches of the apical pollen tube spheroplast plasma membrane of Pyrus pyrifolia cv. Housui were polarized to different voltages (10 mm extracellular Ca2+). Step-like current events – an indication of the opening (O) and closing (C) of single channels – were seen only when the voltage was between −100 and −200 mV. Different channel open states (O1, O2) occurred at the most hyperpolarized voltage (−200 mV). No activity was observed between −80 and +100 mV. An addition of 100 µm lanthanum (La3+) or gadolinium (Gd3+) to the bath solution inhibited channel activity, as indicated by very brief openings. (b) The current–voltage (I/V) relationship of single-channel inward currents when voltage was varied from −100 to −200 mV. (c) Open probability (PO) as a function of applied voltage. The PO curve was fitted by the Boltzmann equation PO/PO(max) = 1/[1 − exp(V − V1/2)/K], where V1/2 is −163.79145 ± 3.54645 mV and K is 17.08056 ± 3.46271 (n = 8). Closed circles, 10 mm Ca2+.

Effects of gadolinium (Gd3+) and lanthanum (La3+)

Gd3+ and La3+ are two widely used Ca2+ channel blockers effective in both animal and plant cells. As shown in Fig. 7, 100 µm La3+ or Gd3+ markedly inhibited the hyperpolarization-activated inward currents (n = 7). Application of this concentration of Gd3+ or La3+ abolished germination and markedly inhibited tube elongation in P. pyrifolia (Table 1). Taken together, these results suggest that the Ca2+-permeable cation channel identified here is involved in pollen germination and tube elongation.

Figure 7.

Effects of lanthanum (La3+) and gadolinium (Gd3+) on inward currents of the pollen tube apical spheroplast plasma membrane of Pyrus pyrifolia cv. Housui. (a) Normal calcium (Ca2+) currents recorded in 10 mm extracellular Ca2+. (b) Ca2+ currents with 100 µm La3+ added to the bath solution (representative of seven independent trials). (c) Normal barium (Ba2+) currents recorded in 10 mm extracellular Ba2+. (d) Ba2+ currents with 100 µm Gd3+ added to the bath solution (representative of seven independent trials). (e) Comparison of the mean (± standard error) current–voltage (I/V) relationships from the four conditions shown in (a)–(d) (P < 0.01 for the data points marked **). Note that La3+ and Gd3+ completely blocked the channels (n = 7).

Table 1.  Gadolinium (Gd3+) and lanthanum (La3+) affect germination and growth of Pyrus pyrifolia cv. Housui pollen
  1. GdCl3 or LaCl3 was added to a final concentration of 100 µm before germination and then pollen tube length was determined at 2 and 4 h after germination. Alternatively, pollen grains were germinated under control conditions and inhibitor at a concentration of 100 µm was added 2 h after germination (treatment). At 4 h, the lengths of control pollen tubes were compared with those in the treatments, and the difference was found to be significant (**, P < 0.01). Experiments were repeated three times and data are mean (± standard error) values from at least 300 observations.

Germination rate (%) 73  0  0
Length of pollen tube at 2 h (µm)226 ± 21  0  0
Length of pollen tube at 4 h (µm)451 ± 28  0  0
Length of pollen tube (µm) 2 h after451 ± 28**230 ± 20228 ± 18
addition of inhibitor to growing pollen tubes

Viability of pollen influences Ca2+ influx

There was an extremely significant difference (P < 0.01) in germination rate and pollen tube length (determined at 3 h of incubation) between the pollen that was collected in 2004 and then stored for 2 yr and that collected in 2006 and then stored for 6 months. The former exhibited a germination rate of only 4.3% (n = 276) and mean (± SE) pollen tube length of 151.68 ± 86.41 µm (n = 62), while the latter exhibited a germination rate of > 70% (n = 369) a pollen tube length of > 400 µm (n = 304). These data suggested that prolonged storage deleteriously affected pollen viability. Additionally, the Ca2+ current magnitude of apical spheroplasts from the 2004 pollen was approx. 12 times smaller than that of apical spheroplasts from the 2006 pollen (Fig. 8; n = 16). Thus, Ca2+ channel activity at tip of the pollen tube was affected by pollen storage. The low germination and growth rates of the 2004 pollen coupled with the low success rate in obtaining high-electrical-resistance seals (5%) curtailed further investigation of this phenomenon. Seal resistances were in the range of 400–700 Ω and were not improved by coating pipette tips with poly l-lysine.

Figure 8.

Pollen viability affects the magnitude of the calcium (Ca2+) currents. (a) Representative whole-cell recording from an apical pollen tube spheroplast from Pyrus pyrifolia cv. Housui pollen collected in 2004 (10 mm extracellular Ca2+). (b) Comparison of mean (± standard error) whole-cell Ca2+ currents in apical pollen tube spheroplasts obtained from pollen collected in 2004 (open circles, n = 16) and 2006 (closed circles, n = 29). Recording conditions were as in (a).

pH dependence of Ca2+ currents

The pH of the medium was found to influence both pollen germination rate and pollen tube length. The medium pH was adjusted from 4.5 to 7.0 using MES and/or Tris. Pollen germination differed significantly with medium pH (Fig. 9a). It was highest at pH 5.5 (80.23%; n = 316) and lowest at pH 7.0 (8.95%; n = 410) but there was no significant difference between pH 5.5 and 6.5. Low germination rates were observed at pH 4.5 (n = 493) and pollen tube growth was arrested at this pH (Fig. 9b). Despite showing the lowest germination rates, pollen tube length after 1 h at pH 7.0 was significantly greater than at pH 4.5. During the first 2 h of growth, pollen tube lengths were greatest at pH 5.5 and 6.5, but at 3 h those at pH 6.5 were significantly longer (P < 0.01) than those at pH 5.5 (Fig. 9b). External pH also affected the Ca2+ currents; these decreased gradually as the pH was increased from 4.5 to 6.5 (Fig. 10a,b). For instance, relative to the currents at pH 6.5 (measured at −200 mV) the currents at pH 4.5 and 5.5 were approx. 3 and 2 times higher, respectively. These results show that Ca2+ currents were pH dependent.

Figure 9.

Effect of extracellular pH on germination and tube growth of Pyrus pyrifolia cv. Housui pollen. (a) Germination rates of pollen at pH 4.5 (n = 493), pH 5.5 (n = 316), pH 6.5 (n = 721), and pH 7.0 (n = 410). (b) Pollen tube lengths were measured at 1, 2, and 3 h after germination (pH 4.5, n = 93; pH 5.5, n = 235; pH 6.5, n = 307; pH 7.0, n = 71). Growth was arrested at pH 4.5 and greatly inhibited at pH 7.0. Growth rate was fastest at pH 6.5, particularly after 2 h.

Figure 10.

The impact of extracellular pH on inward calcium (Ca2+) currents. (a) The inward currents (mean ± standard error) were compared at three bathing solution pH values (closed circles, pH 4.5, n = 15; triangles, pH 5.5, n = 44; open circles, pH 6.5, n = 12). The difference in Ca2+ currents was statistically significant when the voltage was held between −160 and −200 mV (**, P < 0.01), and between −140 and −120 mV (*, P < 0.05). (b) Ca2+ currents at different pH values recorded from a single spheroplast of Pyrus pyrifolia cv. Housui. The dark gray area indicates that the current at pH 5.5 was greater than that at pH 6.5 (n = 3). The light gray area indicates that the current at pH 4.5 was higher than that at pH 5.5. The inset shows the voltage protocol.


Hyperpolarization-activated Ca2+-permeable channels have been identified in a number of plant cells (Miedema et al., 2001), such as in the root hairs (Véry & Davies, 2000) and pollen grains (Wang et al., 2004) of Arabidopsis, and in the pollen grains of lily (Lilium davidii Duchartre; Shang et al., 2005). Previous studies have also isolated spheroplasts from the pollen tubes of lily, identifying K+ inwardly rectifying channels (Griessner & Obermeyer, 2003) and stretch-activated cation channels (Dutta & Robinson, 2004). The present study revealed the existence of hyperpolarization-activated Ca2+-permeable channels in the plasma membrane of pollen tubes of P. pyrifolia. The hyperpolarization-activated Ca2+ conductance exhibited similar characteristics to those reported in the literature (Pei et al., 2000; Véry & Davies, 2000; Wang et al., 2004; Shang et al., 2005) in terms of activation kinetics, selectivity and pharmacology.

It is well established that tip growth, as seen in pollen tubes, algal rhizoids, fungal hyphae and root hairs, is associated with an apex-high [Ca2+]cyt gradient (e.g. Messerli & Robinson, 1997; Felle & Hepler, 1997; Wymer et al., 1997; Messerli et al., 1999). The [Ca2+]cyt gradient has been suggested to result from a localized Ca2+ influx through Ca2+-permeable channels at the apical plasma membrane (Wymer et al., 1997; Holdaway-Clarke et al., 2003). However, evidence of the presence of functional Ca2+-permeable channels in the apices of pollen tubes has been limited. In the present study, we successfully isolated spheroplasts from the apical regions of elongating pollen tubes in P. pyrifolia and routinely obtained the high-resistance seals (> 1GΩ) that are essential for characterizing the activity of ionic channels by patch-clamp electrophysiology. A hyperpolarization-activated inwardly rectifying current was found to dominate the conductance of the pollen tube spheroplast plasma membrane. The inward current was carried by a Ca2+ influx (PCa/PCl of 14), but the channel was also permeable to other divalent cations with a permeability sequence of Ba2+≈ Ca2+ > Mg2+ > Sr2+ > Mn2+. This is comparable to the cation channel identified in the plasma membranes of root hair tips in Arabidopsis (Véry & Davies, 2000). In Arabidopsis pollen grains, however, a cation channel with a different permeability sequence of Mg2+ > Ca2+ > Ba2+ > Sr2+ > Mn2+ was observed (Wang et al., 2004).

Similar to the Ca2+-permable channels of Arabidopsis root hairs and pollen grains (Véry & Davies, 2000; Wang et al., 2004), the channel identified here in P. pyrifolia pollen tubes was sensitive to La3+ and Gd3+, two broad-spectrum Ca2+-channel blockers effective in both animal and plant systems. These two inhibitors also suppressed pollen germination and tube elongation. Moreover, the low viability of pollen corresponded to low channel activity at the pollen tube apex. Therefore, we suggest that the Ca2+-permeable channels identified in the present study have physiological significance in regulating pollen germination and tube elongation in P. pyrifolia.

Voltage and membrane tension are anticipated to be significant factors in the regulation of plasma membrane Ca2+ channels in polar growth and signalling. There is evidence that reorienting treatments applied during pollen tube growth induce a depolarization of the plasma membrane, suggesting that voltage-gated Ca2+ channels might be activated (Malhóet al., 1995, 2000). Dutta & Robinson (2004) have identified a stretch-activated Ca2+ channel in pollen-tip spheroplasts of Lilium longiflorum. The selectivity of this channel was not determined but it was found to be blocked by spider venom; application of this toxin to growing pollen tubes blocked Ca2+ influx, implicating the stretch-activated channel in growth. A similar stretch-activated Ca2+ channel (which was Gd3+-sensitive and open at negative voltages and showed a very strong selectivity for Ca2+ over K+) was observed after 1 h or more of incubation in the germination medium, shortly before germination occurred; it was not present in a functional form in newly hydrated pollen grains (Dutta & Robinson, 2004). There is substantial evidence for a role of mechanosensitive ion channels in detecting and eliciting responses to animal cell volume changes (McCartey & O’Neil, 1992; Chen et al., 1996). In plants, stretch-activated Ca2+-permeable channels have been found in guard cell plasma membranes (Cosgrove & Hedrich, 1991; Grabov & Blatt, 1998). In tip growing cells, such as pollen tubes, turgor pressure could regulate stretch-activated Ca2+ channels only within narrow limits despite variations in cell elongation rate (Benkert et al., 1997). The finding here that the hyperpolarization-activated Ca2+-permeable channel was present in both pollen grains and pollen tubes suggests that the hyperpolarization-activated Ca2+ channels may play an important role in germination, [Ca2+]cyt oscillations and maintenance of the [Ca2+]cyt gradient at the tip of the pollen tube.

Pollen tubes possess an oscillatory pH gradient (6.8 at the tip and 7.5 in the clear zone, detected with BCECF-dextran; Feijóet al., 1999). Tip acidity follows growth by ∼20° (while ionic activities and fluxes show the same period as growth, they usually do not show the same phase and mostly lag behind) whereas the alkaline band anticipates growth by ∼125° (Holdaway-Clarke & Hepler, 2003; Hepler et al., 2005, 2006). These results indicate the alkaline band as a central player in growth control. In addition to the internal pH gradient, there are marked currents of protons in the extracellular medium both in root hairs and in pollen tubes. Apoplastic, apical H+ may promote a more plastic and extensible wall; for example, acidic pH decreases the activity of pectin methylesterase (PME; Moustacas et al., 1986), thus reducing the number of carboxyl residues and the amount of Ca2+ cross-linking. Low pH may also enhance the activity of acidic isoforms of PME (Li et al., 1996, 2002), which, together with pectin hydrolyases, cause the degradation of pectin gels (Bordenave, 1996). However, in lily, the germination rate and the average growth rate of short tubes are more sensitive to changes in pH while for longer tubes the pH has less effect, except at pH 7.0, at which growth is completely stopped (Holdaway-Clarke et al., 2003). In this study, we found that there were no differences in germination and initial growth between pH 5.5 and 6.5. However, growth rate was significantly increased at pH 6.5 after 2 h, suggesting that relatively long pollen tubes could be more sensitive to changes of pH. High extracellular H+ (low pH) promoted Ca2+ influx through Ca2+ channels in the membrane of apical pollen tubes. The lower the pH of the bath solution, the greater the Ca2+ currents. As extremes of low and high pH inhibited germination and pollen tube growth, we reason that pH regulation of Ca2+ influx mediated by the hyperpolarization-activated Ca2+ channel must be finely tuned to allow growth to proceed. Too great or too low an influx would be inhibitory.

In summary, apical spheroplasts from elongating pollen tubes of P. pyrifolia were isolated and a hyperpolarization-activated Ca2+-permeable inward conductance was identified in the plasma membrane. The Ca2+-permeable channel was also permeable to other divalent cations including Ba2+, Ca2+, Mg2+, Sr2+, and Mn2+. The calcium-channel blockers La3+ and Gd3+ inhibited the hyperpolarization-activated channel as well as pollen germination and tube elongation. Moreover, the channel was regulated by extracellular pH and activity varied with pollen viability. We conclude that the Ca2+-permeable channel characterized in the present study is likely to play a role in pollen germination and tube elongation.


We thank all the scientists who helped to improve our manuscript. Special thanks to Dr Julia Davies (Department of Plant Sciences, University of Cambridge) and Dr Wu Li (The Rockefeller University). We are indebted to the two anonymous reviewers, whose insightful comments resulted in a much improved final version of the manuscript. This work was supported by the Science Foundation of Doctoral Subject Point of the Chinese Ministry of Education (No: B200523) and the scientific research fund of HYIT (No: HG0606).