Calcium gradient dependence of Neurospora ... - Semantic Scholar
Microbiology (2003), 149, 2475–2485
DOI 10.1099/mic.0.26302-0
Calcium gradient dependence of Neurospora crassa hyphal growth Lorelei B. Silverman-Gavrila3 and Roger R. Lew3 Department of Biology, York University, 4700 Keele Street, Toronto, Ontario, Canada M3J 1P3
Correspondence Roger R. Lew [email protected]
Received 14 February 2003 Revised
9 April 2003
Accepted 8 May 2003
A tip-high cytoplasmic calcium gradient has been identified as a requirement for hyphal growth in the fungus Neurospora crassa. The Ca2+ gradient is less steep compared to wall vesicle, wall incorporation and vesicular Ca2+ gradients, but this can be explained by Ca2+ diffusion. Analysis of the relation between the rate of hyphal growth and the spatial distribution of tip-localized calcium indicates that hyphal growth rates depend upon the tip-localized calcium concentration. It is not the steepness of the calcium gradient, but tip-localized calcium and the difference in tip-localized calcium versus subapical calcium concentration which correlate closely with hyphal growth rate. A minimal concentration difference between the apex and subapical region of 30 nM is required for growth to occur. The calcium concentration dependence of growth may relate directly to biochemical functions of calcium in hyphal extension, such as vesicle fusion and enzyme activation during cellular expansion. Initiation of tip growth may rely upon random Ca2+ motions causing localized regions of elevated calcium. Continued hyphal expansion may activate a stretch-activated phospholipase C which would increase tip-localized inositol 1,4,5-trisphosphate (IP3). Hyphal expansion, induced by mild hypoosmotic treatment, does increase diacylglycerol, the other product of phospholipase C activity. This is consistent with evidence that IP3-activated Ca2+ channels generate and maintain the tip-high calcium gradient.
INTRODUCTION Polarized cell expansion, culminating in tubular extensions (tip growth), is a morphogenic process observed in all kingdoms, from bacteria to animals. In all the examples of tip-growing eukaryotic cells which have been studied so far, there is a relationship between tip extension and internal calcium gradients. For example, pollen tubes exhibit a cytoplasmic tip-high Ca2+ gradient during growth. The gradient is quite steep: [Ca2+] is about 3?0 mM at the tip, decreasing to 0?2 mM about 20 mm behind the tip (Pierson et al., 1994). Inhibition of pollen tube growth by BAPTA injection correlates with dissipation of the cytoplasmic tip-high Ca2+ gradient and inhibition of tip-localized Ca2+ influx (Pierson et al., 1994). In addition, pulsations of the tip-localized Ca2+ concentration are correlated with pulsatile growth in pollen tubes (Pierson et al., 1996; Messerli & Robinson, 1997). During pulsatile pollen tube growth, growth precedes increased Ca2+ influx and pulsatile cytoplasmic Ca2+ increases by a few seconds (Messerli et al., 1999, 2000). This suggests that Ca2+ influx ‘senses’ tip expansion during growth; the response naturally lags behind Abbreviations: BAPTA, 1,2-bis(ortho-aminophenoxy)ethane-N,N,N9,N9tetrapotassium acetate; CTC, chlortetracycline; fluo-3, 2,7-dichloro-6hydroxy-3-oxo-9-xanthenyl-49-methyl-2,29-(ethylenedioxy)dianiline-N,N,N9, N9-tetraacetic acid; IP3, inositol 1,4,5-trisphosphate. 3Both authors contributed equally to the work.
0002-6302 G 2003 SGM
Printed in Great Britain
tip expansion. Such a mechanism has been proposed to be mediated by stretch-activated Ca2+ channels localized at the tip. Stretch-activated Ca2+ channels have been characterized in the oomycete Saprolegnia ferax (Garrill et al., 1993), in which there is evidence for a correlation between growth rate and the magnitude of the Ca2+ gradient measured using ratio imaging of Ca2+ and pH-sensitive fluorescent dyes (Hyde & Heath, 1997). Fungi also exhibit tip-high Ca2+ gradients during hyphal growth. Spatial cytoplasmic [Ca2+] has been measured using quantitative dual dye (fluo-3 and Fura Red) ratio imaging (Silverman-Gavrila & Lew, 2000). Analogous to pollen tubes, injection of BAPTA dissipates the gradient and stops growth (Silverman-Gavrila & Lew, 2000). Unlike pollen tubes (Pierson et al., 1994), root hairs (Felle & Hepler, 1997) or S. ferax (Lew, 1999), there is no indication that Ca2+ influx at the growing tip is responsible for generating the tip-high gradient. Although Neurospora crassa does have stretch-activated Ca2+ channels (Levina et al., 1995), there is no net Ca2+ influx during hyphal growth (Lew, 1999) and direct manipulation of the membrane potential to modify the driving force for Ca2+ influx does not affect growth rate (Silverman-Gavrila & Lew, 2000). The gradient is generated and maintained internally by the concerted action of inositol 1,4,5-trisphosphate (IP3)-activated Ca2+ release from tiplocalized vesicles (Silverman-Gavrila & Lew, 2002) and Ca2+-ATPase-mediated sequestration into the endoplasmic 2475
L. B. Silverman-Gavrila and R. R. Lew
reticulum behind the growing tip (Silverman-Gavrila & Lew, 2001). The location of the tip-localized vesicles is maintained by interaction with the actin cytoskeleton (Silverman-Gavrila & Lew, 2001). Our objective in this paper is to explore the relation between the Ca2+ gradient and growth in the ascomycete N. crassa, and to identify a possible growth sensor responsible for generating the gradient to maintain continued growth. Does growth depend upon an absolute [Ca2+] at the tip, or is it the steepness of the gradient that is required during growth? Our assessment is done in the context of spatial regulation of the Ca2+ gradient, and its relation to other aspects of the polar cytology of N. crassa hyphae. Random fluctuations of the Ca2+ distribution may generate localized regions of elevated Ca2+ to initiate tip growth. Based on stretchactivated production of diacylglycerol, we propose that activation of a tip-localized phospholipase C may ‘sense’ growth, initiating a cascade of events that maintains the Ca2+ gradient during continued hyphal growth.
METHODS Culturing. The wild-type Neurospora crassa strain RL21a (FGSC no.
2219) was grown in 35 mm tissue culture dishes on solid substrate (2 %, w/v, gellan gum) containing 2 % sucrose and Vogel’s minimal medium (Vogel, 1956), and incubated at 28–30 uC for 14 h. Prior to experiments, the culture was flooded with BS [10 mM KCl, 1 mM CaCl2, 1 mM MgCl2, 10 mM PIPES (pH adjusted to 5?8 with KOH) and the osmolality adjusted to 260 mosmol kg21 with sucrose] (Levina et al., 1995). Ratiometric fluorescence imaging of cytoplasmic calcium.
Cytosolic [Ca2+] was measured by ratio imaging the emission intensities of the Ca2+-sensitive fluorescent dyes fluo-3 and Fura Red. They were loaded ionophoretically into the hypha. The electrophysiological techniques are described in detail elsewhere (SilvermanGavrila & Lew, 2000, 2001). The micropipette was filled at the tip with 0?33 mM fluo-3 and 0?99 mM Fura Red (both as potassium salts; Molecular Probes) and backfilled with 3 M KCl. Hyphae were impaled about 35 mm behind the tip. Fluorescence imaging was performed using a Bio-Rad MRC-600 confocal apparatus with a krypton–argon mixed gas laser attached to a Nikon Optiphot 2 microscope (Silverman-Gavrila & Lew, 2000). Briefly, the dyes were excited at 488 nm using 10 % laser intensity (neutral density filter 1) and the emitted fluorescence was detected simultaneously at 522 (fluo-3) and 640 (Fura Red) nm using fast photon counting (10 scans). Ratio intensities were measured using 2?54 mm longitudinal transects within the cytoplasm of the hyphae in the software program NIH-Image (Rasband & Bright, 1995). As detailed in previous work (Silverman-Gavrila & Lew, 2000), the signal to noise ratio is high; autofluorescence contributes less than 11 and 6 % to the fluorescence intensities of fluo-3 and Fura Red, respectively. Growth measurements of hyphae microinjected with fluo-3 and Fura Red. Individual growth rates were observed with a
Nikon Optiphot microscope and a 640 water immersion objective. Growth rates were measured from thermal prints, after hyphae had resumed growth following impalement with the micropipette and dye microinjection by ionophoresis. Growth rates measured either immediately before or after the ratiometric fluorescence imaging were used to correlate growth with the cytoplasmic [Ca2+] gradient. 2476
Vesicular calcium imaging. Organellar Ca2+ fluorescence was
measured after chlortetracycline (CTC; Sigma) addition. The optimal concentration of CTC which did not affect hyphal growth but still provided good fluorescence signal was determined to be about 25–50 mM. The fluorescence was detected by confocal microscopy using a Bio-Rad MRC-600 apparatus equipped with a krypton– argon laser on a Nikon Optiphot microscope with a 640 water immersion objective. A BHS filter was used (excitation at 488 nm, emission>515 nm), with no neutral density filter. The acquired images were filtered using Kalman digital filtering to improve visualization. Plots of fluorescence intensity versus distance from the tip were obtained using 2?54 mm longitudinal transects along the hyphae in the software program NIH-Image. To correct for the smaller volume elements at the hyphal apex, volume was estimated from cylindrical volume elements 0?127 mm in length, with radius calculated from an exponential best fit to hyphal diameter. The CTC fluorescence intensity for each 0?127 mm sample was divided by the volume of the corresponding volume element; i.e. for distance from the tip, n, from 0 to 25 mm in 0?127 mm steps, CTCcorrected=CTCn/volumen. Diacylglycerol measurements using HPLC. To maximize the
isolation of diacylglycerol produced at growing hyphal tips, we used conidial germlings. Large-scale conidial harvests were incubated in Vogel’s minimal medium at 37 uC for 5–6 h, at which time the germlings had grown about 100–200 mm. Prior to isolation of diacylglycerol using chloroform/methanol extraction, the germlings were treated with phospholipase C inhibitors at concentrations that inhibited growth completely (Silverman-Gavrila & Lew, 2002) (neomycin, 400 mM; 3-nitrocoumarin, 40 mg ml21; U-73122 or the inactive analogue U-73343, 400 mM), or subjected to hypoosmotic stress: either severe (a 1 : 19 dilution of Vogel’s minimal medium with distilled H2O) or mild (a 1 : 1 dilution of Vogel’s minimal medium with distilled H2O) stress. Immediately after treatment, the germlings were collected by filtration through a 0?22 mm filter, then scraped into a 1?5 ml Eppendorf tube containing 0?75 ml ice-cold chloroform/methanol (1 : 2, v/v), vortex-mixed and kept on ice for 15 min. Diacylglycerol extraction followed the protocol described in detail by Ramsdale & Lakin-Thomas (2000), adapted from Bligh & Dyer (1959). The diacylglycerol extracts were stored at 220 uC in chloroform containing 50 mg butylated hydroxytoluene ml21. Mycelial dry weight was determined by washing mycelial debris from the initial extraction in methanol, drying overnight at 60 uC, then weighing. The lipids were measured using an HPLC technique modified from Bocckino et al. (1985). We used a Betasil silica-60 (5 mm particle size) (25064?6 mm) column (ThermoHypersilKeystone). Chromatography was performed using a BioCAD Sprint chromatography system (PerSeptiva Biosystems). The solvent was hexane/2-propanol/glacial acetic acid (250 : 2?5 : 0?025) (HIA) run at 3 ml min21 at about 1200 p.s.i. (8280 kPa). Lipid samples (125 ml) were dried under N2 at 60 uC and redissolved in 500 ml HIA. After equilibration of the column with HIA, 100 ml samples were injected into the column. Lipids were detected by the A205. Diacylglycerol and ergosterol (Sigma-Aldrich) standards were used to identify HPLC peaks. All other reagents were obtained from Sigma-Aldrich and were HPLC grade. Data analysis. The experiments (77 in all) were sorted by growth
rate and mean Ca2+ gradients were calculated for subsamples (n=11 or n=7). This assured an even spread of growth rates. A statistical software package (SYSTAT, version 5.0) was used for linear and nonlinear regression analysis of the relation between growth rate and various aspects of the Ca2+ gradient. Best fits for various mathematical models were obtained by minimization of least P (observedn {predictedn )2 , with either a quasi-Newton or squares, n
Simplex method (Wilkinson, 1988). Linear or exponential models were used as described in Results. Goodness of fit was assessed quantitatively with correlation coefficients and two-tail probabilities. Microbiology 149
Calcium gradients and fungal tip growth
Fig. 1. Example of cytoplasmic free [Ca2+] imaging in a growing hypha using dual dye ratio imaging. (a) Fluo-3 fluorescence. (b) Fura Red fluorescence. (c) Pseudocolour ratiometric image of Ca2+ distribution. Quantification, based on in vitro calibrations, is shown in Fig. 2. Note that (a) and (b) have been contrast-enhanced to show the homogeneous distributions of the dyes in the cytoplasm-rich tip of the N. crassa hypha. Sequestration of the dyes is not apparent in this example, and was in fact an extremely rare event (Silverman-Gavrila & Lew, 2000). The dark halo around the pseudocolour ratio image is due to a border effect at the edge of the hypha when the ratio is calculated. Transects (2?54 mm wide) were placed longitudinally, abutted against the edge of the hyphae and completely within the cytoplasm (the haloes were not included) (SilvermanGavrila & Lew, 2000). The hypha was growing at a rate of 6?9 mm min”1 when the fluorescence images were taken. For scale, the hyphal base is 7 mm wide (a).
Ca2+ random walks. A computer program was written in C to produce a 64 by 64 array in which each array element contained 50 calcium ions initially. A uniform [0,1] random number generator based on a combination of two linear congruential sequences (L. Devroye; http://www-cgrl.cs.mcgill.ca/~luc/rng.html) was used to move 25 % of the calcium molecules to one of the four bounding array elements, depending on whether the random number fell in the range 0<x¡0?25, 0?25<x¡0?5, 0?5<x¡0?75 or 0?75<x¡ 1?0. The process was iterated 128 times. For visualization, the arrays were converted into images that were processed with a Gaussian filter and linear contrast stretch in NIH-Image (Rasband & Bright, 1995).
RESULTS An example of cytoplasmic [Ca2+] ratio imaging is shown in Fig. 1; quantification is shown in Fig. 2. The dyes
co-localized in the cytoplasm-rich regions of the hypha. Occasional zones of exclusion, probably nuclei, were observed behind the hyphal apex. Dye sequestration was very rare; those experiments were not used in the analysis. The pseudocolour image shows the spatial distribution of the tip-high Ca2+ gradient. Longitudinal transects were used to produce quantitative measurements for analysis of the Ca2+ gradient dependence of the growth rate (Fig. 2). Individual experiments were sorted according to growth rate and compiled into samples of 11 experiments each (Table 1). Thus there is a mean growth rate for each mean gradient. Cytoplasmic [Ca2+] is compiled for regions of the hyphae extending from the tip to 20 mm, where subapical [Ca2+] approaches its basal level, [Ca2+]basal. To test the relation between the gradient and growth rate, we determined the correlations between growth rate and the steepness of the [Ca2+] gradient, tip-localized [Ca2+], and the difference between tip-localized [Ca2+] and basal [Ca2+]. Hyphal growth dependence on Ca2+ gradient steepness and magnitude The steepness of the gradient can be quantified by using an exponential fit of [Ca2+] versus distance from the tip. We used an exponential equation of the form ½Ca2z ~½Ca2z basal z½Ca2z max .e{d=tau , where [Ca2+]basal is the basal [Ca2+], and is summed with [Ca2+]max to approximate tip-localized [Ca2+], d is the distance from the tip, and tau is a measure of the steepness of the gradient; a small tau corresponds to a steep gradient (Table 1). Growth rate was poorly correlated with the steepness of the gradient (Fig. 3); the Pearson correlation coefficient was very small, r2=0?000, P=0?462. Rather than steepness per se, it is possible that growth rate depends upon the magnitude of the Ca2+ gradient, either tip-localized [Ca2+], or the difference between tip-localized and basal free cytoplasmic [Ca2+] (Fig. 3). Similar correlations were found for both measurements of the [Ca2+] gradient; the correlations, while small, were statistically significant. To assure that the choice of sample size (n=11) did not cause a fortuitous correlation, the 77 individual experiments, sorted by growth rate, were recompiled into sample sizes of 7. No correlation was observed for growth versus gradient steepness (r 2= 0?078, P=0?172). The correlation coefficients for growth Fig. 2. Quantification of cytoplasmic free [Ca2+] imaging using dual dye ratio imaging. The ratio image in Fig. 1 was quantified by taking a longitudinal transect along the hypha. An in vitro calibration was used to convert the ratio to cytoplasmic [Ca2+]. The best fit is to an exponential function of the form ½Ca2z ~½Ca2z basal z½Ca2z max .e{d=tau , where [Ca2+]basal is the subapical [Ca2+], summed with [Ca2+]max to approximate tiplocalized [Ca2+]. The growth rate of the hypha was 6?9 mm min”1; the steepness (tau) was 6?6 mm.
http://mic.sgmjournals.org
2477
L. B. Silverman-Gavrila and R. R. Lew
Table 1. Data summary for cytoplasmic [Ca2+] gradients at the hyphal tip Mean [Ca2+] from ratio images from different regions of the hyphae are tabulated for hyphae growing at the mean growth rates as shown. The difference between tip-localized and basal [Ca2+] and the gradient steepness (tau) are shown for the compiled growth rates. Cytoplasmic free [Ca2+] behind the hyphal tip (nM)
Growth rate (mm min21) Mean
SD
n
0–2?5 mm
2?5–5 mm
5–10 mm
10–20 mm
0 1?57 3?81 5?94 7?39 10?75 14?25 19?58
0 0?71 0?82 0?45 0?601 1?76 0?98 2?81
19 11 11 11 11 11 11 11
190 272 302 196 428 387 420 387
183 256 295 182 343 356 329 349
178 239 255 159 320 328 293 320
162 203 193 134 268 256 273 243
versus either tip-localized [Ca2+], or the difference between tip-localized and basal free cytoplasmic [Ca2+] were smaller due to increased variability with the smaller sample sizes (r2=0?23), but were still statistically significant (Pr2>0?52, P
DOI 10.1099/mic.0.26302-0
Calcium gradient dependence of Neurospora crassa hyphal growth Lorelei B. Silverman-Gavrila3 and Roger R. Lew3 Department of Biology, York University, 4700 Keele Street, Toronto, Ontario, Canada M3J 1P3
Correspondence Roger R. Lew [email protected]
Received 14 February 2003 Revised
9 April 2003
Accepted 8 May 2003
A tip-high cytoplasmic calcium gradient has been identified as a requirement for hyphal growth in the fungus Neurospora crassa. The Ca2+ gradient is less steep compared to wall vesicle, wall incorporation and vesicular Ca2+ gradients, but this can be explained by Ca2+ diffusion. Analysis of the relation between the rate of hyphal growth and the spatial distribution of tip-localized calcium indicates that hyphal growth rates depend upon the tip-localized calcium concentration. It is not the steepness of the calcium gradient, but tip-localized calcium and the difference in tip-localized calcium versus subapical calcium concentration which correlate closely with hyphal growth rate. A minimal concentration difference between the apex and subapical region of 30 nM is required for growth to occur. The calcium concentration dependence of growth may relate directly to biochemical functions of calcium in hyphal extension, such as vesicle fusion and enzyme activation during cellular expansion. Initiation of tip growth may rely upon random Ca2+ motions causing localized regions of elevated calcium. Continued hyphal expansion may activate a stretch-activated phospholipase C which would increase tip-localized inositol 1,4,5-trisphosphate (IP3). Hyphal expansion, induced by mild hypoosmotic treatment, does increase diacylglycerol, the other product of phospholipase C activity. This is consistent with evidence that IP3-activated Ca2+ channels generate and maintain the tip-high calcium gradient.
INTRODUCTION Polarized cell expansion, culminating in tubular extensions (tip growth), is a morphogenic process observed in all kingdoms, from bacteria to animals. In all the examples of tip-growing eukaryotic cells which have been studied so far, there is a relationship between tip extension and internal calcium gradients. For example, pollen tubes exhibit a cytoplasmic tip-high Ca2+ gradient during growth. The gradient is quite steep: [Ca2+] is about 3?0 mM at the tip, decreasing to 0?2 mM about 20 mm behind the tip (Pierson et al., 1994). Inhibition of pollen tube growth by BAPTA injection correlates with dissipation of the cytoplasmic tip-high Ca2+ gradient and inhibition of tip-localized Ca2+ influx (Pierson et al., 1994). In addition, pulsations of the tip-localized Ca2+ concentration are correlated with pulsatile growth in pollen tubes (Pierson et al., 1996; Messerli & Robinson, 1997). During pulsatile pollen tube growth, growth precedes increased Ca2+ influx and pulsatile cytoplasmic Ca2+ increases by a few seconds (Messerli et al., 1999, 2000). This suggests that Ca2+ influx ‘senses’ tip expansion during growth; the response naturally lags behind Abbreviations: BAPTA, 1,2-bis(ortho-aminophenoxy)ethane-N,N,N9,N9tetrapotassium acetate; CTC, chlortetracycline; fluo-3, 2,7-dichloro-6hydroxy-3-oxo-9-xanthenyl-49-methyl-2,29-(ethylenedioxy)dianiline-N,N,N9, N9-tetraacetic acid; IP3, inositol 1,4,5-trisphosphate. 3Both authors contributed equally to the work.
0002-6302 G 2003 SGM
Printed in Great Britain
tip expansion. Such a mechanism has been proposed to be mediated by stretch-activated Ca2+ channels localized at the tip. Stretch-activated Ca2+ channels have been characterized in the oomycete Saprolegnia ferax (Garrill et al., 1993), in which there is evidence for a correlation between growth rate and the magnitude of the Ca2+ gradient measured using ratio imaging of Ca2+ and pH-sensitive fluorescent dyes (Hyde & Heath, 1997). Fungi also exhibit tip-high Ca2+ gradients during hyphal growth. Spatial cytoplasmic [Ca2+] has been measured using quantitative dual dye (fluo-3 and Fura Red) ratio imaging (Silverman-Gavrila & Lew, 2000). Analogous to pollen tubes, injection of BAPTA dissipates the gradient and stops growth (Silverman-Gavrila & Lew, 2000). Unlike pollen tubes (Pierson et al., 1994), root hairs (Felle & Hepler, 1997) or S. ferax (Lew, 1999), there is no indication that Ca2+ influx at the growing tip is responsible for generating the tip-high gradient. Although Neurospora crassa does have stretch-activated Ca2+ channels (Levina et al., 1995), there is no net Ca2+ influx during hyphal growth (Lew, 1999) and direct manipulation of the membrane potential to modify the driving force for Ca2+ influx does not affect growth rate (Silverman-Gavrila & Lew, 2000). The gradient is generated and maintained internally by the concerted action of inositol 1,4,5-trisphosphate (IP3)-activated Ca2+ release from tiplocalized vesicles (Silverman-Gavrila & Lew, 2002) and Ca2+-ATPase-mediated sequestration into the endoplasmic 2475
L. B. Silverman-Gavrila and R. R. Lew
reticulum behind the growing tip (Silverman-Gavrila & Lew, 2001). The location of the tip-localized vesicles is maintained by interaction with the actin cytoskeleton (Silverman-Gavrila & Lew, 2001). Our objective in this paper is to explore the relation between the Ca2+ gradient and growth in the ascomycete N. crassa, and to identify a possible growth sensor responsible for generating the gradient to maintain continued growth. Does growth depend upon an absolute [Ca2+] at the tip, or is it the steepness of the gradient that is required during growth? Our assessment is done in the context of spatial regulation of the Ca2+ gradient, and its relation to other aspects of the polar cytology of N. crassa hyphae. Random fluctuations of the Ca2+ distribution may generate localized regions of elevated Ca2+ to initiate tip growth. Based on stretchactivated production of diacylglycerol, we propose that activation of a tip-localized phospholipase C may ‘sense’ growth, initiating a cascade of events that maintains the Ca2+ gradient during continued hyphal growth.
METHODS Culturing. The wild-type Neurospora crassa strain RL21a (FGSC no.
2219) was grown in 35 mm tissue culture dishes on solid substrate (2 %, w/v, gellan gum) containing 2 % sucrose and Vogel’s minimal medium (Vogel, 1956), and incubated at 28–30 uC for 14 h. Prior to experiments, the culture was flooded with BS [10 mM KCl, 1 mM CaCl2, 1 mM MgCl2, 10 mM PIPES (pH adjusted to 5?8 with KOH) and the osmolality adjusted to 260 mosmol kg21 with sucrose] (Levina et al., 1995). Ratiometric fluorescence imaging of cytoplasmic calcium.
Cytosolic [Ca2+] was measured by ratio imaging the emission intensities of the Ca2+-sensitive fluorescent dyes fluo-3 and Fura Red. They were loaded ionophoretically into the hypha. The electrophysiological techniques are described in detail elsewhere (SilvermanGavrila & Lew, 2000, 2001). The micropipette was filled at the tip with 0?33 mM fluo-3 and 0?99 mM Fura Red (both as potassium salts; Molecular Probes) and backfilled with 3 M KCl. Hyphae were impaled about 35 mm behind the tip. Fluorescence imaging was performed using a Bio-Rad MRC-600 confocal apparatus with a krypton–argon mixed gas laser attached to a Nikon Optiphot 2 microscope (Silverman-Gavrila & Lew, 2000). Briefly, the dyes were excited at 488 nm using 10 % laser intensity (neutral density filter 1) and the emitted fluorescence was detected simultaneously at 522 (fluo-3) and 640 (Fura Red) nm using fast photon counting (10 scans). Ratio intensities were measured using 2?54 mm longitudinal transects within the cytoplasm of the hyphae in the software program NIH-Image (Rasband & Bright, 1995). As detailed in previous work (Silverman-Gavrila & Lew, 2000), the signal to noise ratio is high; autofluorescence contributes less than 11 and 6 % to the fluorescence intensities of fluo-3 and Fura Red, respectively. Growth measurements of hyphae microinjected with fluo-3 and Fura Red. Individual growth rates were observed with a
Nikon Optiphot microscope and a 640 water immersion objective. Growth rates were measured from thermal prints, after hyphae had resumed growth following impalement with the micropipette and dye microinjection by ionophoresis. Growth rates measured either immediately before or after the ratiometric fluorescence imaging were used to correlate growth with the cytoplasmic [Ca2+] gradient. 2476
Vesicular calcium imaging. Organellar Ca2+ fluorescence was
measured after chlortetracycline (CTC; Sigma) addition. The optimal concentration of CTC which did not affect hyphal growth but still provided good fluorescence signal was determined to be about 25–50 mM. The fluorescence was detected by confocal microscopy using a Bio-Rad MRC-600 apparatus equipped with a krypton– argon laser on a Nikon Optiphot microscope with a 640 water immersion objective. A BHS filter was used (excitation at 488 nm, emission>515 nm), with no neutral density filter. The acquired images were filtered using Kalman digital filtering to improve visualization. Plots of fluorescence intensity versus distance from the tip were obtained using 2?54 mm longitudinal transects along the hyphae in the software program NIH-Image. To correct for the smaller volume elements at the hyphal apex, volume was estimated from cylindrical volume elements 0?127 mm in length, with radius calculated from an exponential best fit to hyphal diameter. The CTC fluorescence intensity for each 0?127 mm sample was divided by the volume of the corresponding volume element; i.e. for distance from the tip, n, from 0 to 25 mm in 0?127 mm steps, CTCcorrected=CTCn/volumen. Diacylglycerol measurements using HPLC. To maximize the
isolation of diacylglycerol produced at growing hyphal tips, we used conidial germlings. Large-scale conidial harvests were incubated in Vogel’s minimal medium at 37 uC for 5–6 h, at which time the germlings had grown about 100–200 mm. Prior to isolation of diacylglycerol using chloroform/methanol extraction, the germlings were treated with phospholipase C inhibitors at concentrations that inhibited growth completely (Silverman-Gavrila & Lew, 2002) (neomycin, 400 mM; 3-nitrocoumarin, 40 mg ml21; U-73122 or the inactive analogue U-73343, 400 mM), or subjected to hypoosmotic stress: either severe (a 1 : 19 dilution of Vogel’s minimal medium with distilled H2O) or mild (a 1 : 1 dilution of Vogel’s minimal medium with distilled H2O) stress. Immediately after treatment, the germlings were collected by filtration through a 0?22 mm filter, then scraped into a 1?5 ml Eppendorf tube containing 0?75 ml ice-cold chloroform/methanol (1 : 2, v/v), vortex-mixed and kept on ice for 15 min. Diacylglycerol extraction followed the protocol described in detail by Ramsdale & Lakin-Thomas (2000), adapted from Bligh & Dyer (1959). The diacylglycerol extracts were stored at 220 uC in chloroform containing 50 mg butylated hydroxytoluene ml21. Mycelial dry weight was determined by washing mycelial debris from the initial extraction in methanol, drying overnight at 60 uC, then weighing. The lipids were measured using an HPLC technique modified from Bocckino et al. (1985). We used a Betasil silica-60 (5 mm particle size) (25064?6 mm) column (ThermoHypersilKeystone). Chromatography was performed using a BioCAD Sprint chromatography system (PerSeptiva Biosystems). The solvent was hexane/2-propanol/glacial acetic acid (250 : 2?5 : 0?025) (HIA) run at 3 ml min21 at about 1200 p.s.i. (8280 kPa). Lipid samples (125 ml) were dried under N2 at 60 uC and redissolved in 500 ml HIA. After equilibration of the column with HIA, 100 ml samples were injected into the column. Lipids were detected by the A205. Diacylglycerol and ergosterol (Sigma-Aldrich) standards were used to identify HPLC peaks. All other reagents were obtained from Sigma-Aldrich and were HPLC grade. Data analysis. The experiments (77 in all) were sorted by growth
rate and mean Ca2+ gradients were calculated for subsamples (n=11 or n=7). This assured an even spread of growth rates. A statistical software package (SYSTAT, version 5.0) was used for linear and nonlinear regression analysis of the relation between growth rate and various aspects of the Ca2+ gradient. Best fits for various mathematical models were obtained by minimization of least P (observedn {predictedn )2 , with either a quasi-Newton or squares, n
Simplex method (Wilkinson, 1988). Linear or exponential models were used as described in Results. Goodness of fit was assessed quantitatively with correlation coefficients and two-tail probabilities. Microbiology 149
Calcium gradients and fungal tip growth
Fig. 1. Example of cytoplasmic free [Ca2+] imaging in a growing hypha using dual dye ratio imaging. (a) Fluo-3 fluorescence. (b) Fura Red fluorescence. (c) Pseudocolour ratiometric image of Ca2+ distribution. Quantification, based on in vitro calibrations, is shown in Fig. 2. Note that (a) and (b) have been contrast-enhanced to show the homogeneous distributions of the dyes in the cytoplasm-rich tip of the N. crassa hypha. Sequestration of the dyes is not apparent in this example, and was in fact an extremely rare event (Silverman-Gavrila & Lew, 2000). The dark halo around the pseudocolour ratio image is due to a border effect at the edge of the hypha when the ratio is calculated. Transects (2?54 mm wide) were placed longitudinally, abutted against the edge of the hyphae and completely within the cytoplasm (the haloes were not included) (SilvermanGavrila & Lew, 2000). The hypha was growing at a rate of 6?9 mm min”1 when the fluorescence images were taken. For scale, the hyphal base is 7 mm wide (a).
Ca2+ random walks. A computer program was written in C to produce a 64 by 64 array in which each array element contained 50 calcium ions initially. A uniform [0,1] random number generator based on a combination of two linear congruential sequences (L. Devroye; http://www-cgrl.cs.mcgill.ca/~luc/rng.html) was used to move 25 % of the calcium molecules to one of the four bounding array elements, depending on whether the random number fell in the range 0<x¡0?25, 0?25<x¡0?5, 0?5<x¡0?75 or 0?75<x¡ 1?0. The process was iterated 128 times. For visualization, the arrays were converted into images that were processed with a Gaussian filter and linear contrast stretch in NIH-Image (Rasband & Bright, 1995).
RESULTS An example of cytoplasmic [Ca2+] ratio imaging is shown in Fig. 1; quantification is shown in Fig. 2. The dyes
co-localized in the cytoplasm-rich regions of the hypha. Occasional zones of exclusion, probably nuclei, were observed behind the hyphal apex. Dye sequestration was very rare; those experiments were not used in the analysis. The pseudocolour image shows the spatial distribution of the tip-high Ca2+ gradient. Longitudinal transects were used to produce quantitative measurements for analysis of the Ca2+ gradient dependence of the growth rate (Fig. 2). Individual experiments were sorted according to growth rate and compiled into samples of 11 experiments each (Table 1). Thus there is a mean growth rate for each mean gradient. Cytoplasmic [Ca2+] is compiled for regions of the hyphae extending from the tip to 20 mm, where subapical [Ca2+] approaches its basal level, [Ca2+]basal. To test the relation between the gradient and growth rate, we determined the correlations between growth rate and the steepness of the [Ca2+] gradient, tip-localized [Ca2+], and the difference between tip-localized [Ca2+] and basal [Ca2+]. Hyphal growth dependence on Ca2+ gradient steepness and magnitude The steepness of the gradient can be quantified by using an exponential fit of [Ca2+] versus distance from the tip. We used an exponential equation of the form ½Ca2z ~½Ca2z basal z½Ca2z max .e{d=tau , where [Ca2+]basal is the basal [Ca2+], and is summed with [Ca2+]max to approximate tip-localized [Ca2+], d is the distance from the tip, and tau is a measure of the steepness of the gradient; a small tau corresponds to a steep gradient (Table 1). Growth rate was poorly correlated with the steepness of the gradient (Fig. 3); the Pearson correlation coefficient was very small, r2=0?000, P=0?462. Rather than steepness per se, it is possible that growth rate depends upon the magnitude of the Ca2+ gradient, either tip-localized [Ca2+], or the difference between tip-localized and basal free cytoplasmic [Ca2+] (Fig. 3). Similar correlations were found for both measurements of the [Ca2+] gradient; the correlations, while small, were statistically significant. To assure that the choice of sample size (n=11) did not cause a fortuitous correlation, the 77 individual experiments, sorted by growth rate, were recompiled into sample sizes of 7. No correlation was observed for growth versus gradient steepness (r 2= 0?078, P=0?172). The correlation coefficients for growth Fig. 2. Quantification of cytoplasmic free [Ca2+] imaging using dual dye ratio imaging. The ratio image in Fig. 1 was quantified by taking a longitudinal transect along the hypha. An in vitro calibration was used to convert the ratio to cytoplasmic [Ca2+]. The best fit is to an exponential function of the form ½Ca2z ~½Ca2z basal z½Ca2z max .e{d=tau , where [Ca2+]basal is the subapical [Ca2+], summed with [Ca2+]max to approximate tiplocalized [Ca2+]. The growth rate of the hypha was 6?9 mm min”1; the steepness (tau) was 6?6 mm.
http://mic.sgmjournals.org
2477
L. B. Silverman-Gavrila and R. R. Lew
Table 1. Data summary for cytoplasmic [Ca2+] gradients at the hyphal tip Mean [Ca2+] from ratio images from different regions of the hyphae are tabulated for hyphae growing at the mean growth rates as shown. The difference between tip-localized and basal [Ca2+] and the gradient steepness (tau) are shown for the compiled growth rates. Cytoplasmic free [Ca2+] behind the hyphal tip (nM)
Growth rate (mm min21) Mean
SD
n
0–2?5 mm
2?5–5 mm
5–10 mm
10–20 mm
0 1?57 3?81 5?94 7?39 10?75 14?25 19?58
0 0?71 0?82 0?45 0?601 1?76 0?98 2?81
19 11 11 11 11 11 11 11
190 272 302 196 428 387 420 387
183 256 295 182 343 356 329 349
178 239 255 159 320 328 293 320
162 203 193 134 268 256 273 243
versus either tip-localized [Ca2+], or the difference between tip-localized and basal free cytoplasmic [Ca2+] were smaller due to increased variability with the smaller sample sizes (r2=0?23), but were still statistically significant (Pr2>0?52, P
