Journal of Neurochemistry Raven Press, Ltd., New York 0 1992 International Society for Neurochemistry
Relationship of Intracellular Calcium to Dependence on Nerve Growth Factor in Dorsal Root Ganglion Neurons in Cell Culture *Marc E. Eichler, ?Janet M. Dubinsky, and *?Keith M. Rich Departments of *Neurosurgery and j-Neurobiology, Washington University School of Medicine, St. Louis, Missouri, U.S.A.
Abstract: During development, neural crest-derived sensory neurons require nerve growth factor (NGF) for survival, but lose this dependency postnatally. Similarly, dissociated embryonic sensory neurons lose their NGF dependence during the first 3 weeks in cell culture. It has been hypothesized that, in sympathetic neurons, intracellular levels of calcium are related to trophic factor dependence. In vitro during the period in which embryonic-day- 15 sensory neurons become independent of NGF, intracellular calcium concentrations progressively increased in parallel to the decline in NGF dependence. This elevation of intracellular calcium was directly related to the absolute age of the neurons, not to the length of time in culture. Without NGF, immature sensory, i.e., dependent, neurons survived in the presence of high extracellular potassium, a condition that produces elevated intra-
cellular calcium. In another paradigm, measurements of intracellular calcium were determined in NGF-dependent neurons “committed to die” after NGF withdrawal. These measurements were determined prior to the time that extensive morphological changes, consistent with cell death, were noted by phase-contrast microscopy. No elevation in intracellular calcium was found in these dying neurons, but rather, a small decrease was observed prior to the disintegration of the neurons. These findings support the hypothesis that trophic factor dependence of neurons may be inversely related to levels of intracellular calcium. Key Words: Nerve growth factor-Intracellular calcium-Sensory neurons. Eichler M. E. et al. Relationship of intracellular calcium to dependence on nerve growth factor in dorsal root ganglion neurons in cell culture. J. Neurochem. 58, 263-269 (1992).
Neuronal death is a tightly regulated phenomenon that occurs in several neuronal subpopulations during development as a physiologic method of establishing the proper number of neurons that survive into the postnatal period (Oppenheim, 1989). During neuronal development, a naturally occurring cell death culls an overproduction of neurons to the population that is ultimately required to innervate specific targets. Molecular mechanisms that result in neuronal death under such circumstances are not known. Nerve growth factor (NGF) is a trophic agent crucial in the naturally occurring cell death that takes place in both dorsal root ganglion (DRG) and sympathetic neurons during embryogenesis. In vivo dependence on NGF for survival occurs as a transient phenomenon in the prenatal and early postnatal sensory neuron (Johnson et al., 1980). The mature DRG neuron is not dependent on NGF for survival (Lindsay, 1988), but does continue to ex-
press NGF receptors and requires NGF for biochemical and morphologic homeostasis (for concise review, see Johnson et al., 1986). Exogenously administered NGF is capable of preventing neuronal death after axotomy in both immature and adult rats (Yip et al., 1984;Rich et al., 1987). The intracellular events that regulate NGF dependence during maturation of DRG neurons remain unknown. Intracellular calcium concentration ([Ca2+]i) has been implicated in a final, common pathway in some forms of cell death (Schanne et al., 1979).Hypoxemicischemic neuronal injury has been linked to massive influxes of calcium into the cytosol from intracellular organelles and extracellular sources (Siesjo and Bengtsson, 1989). Activation of postsynaptic glutamate receptors on neurons has been associated with an influx of calcium intracellularly, resulting in adverse reactions leading to cell death (Choi, 1985; Rothman and Olney,
Received March 13, 1991; revised manuscript received May 24, 1991; accepted May 31, 1991. Address correspondence and reprint requests to Dr. K. M. Rich at Department of Neurological Surgery, Washington University School of Medicine, St. Louis, MO 631 10. U.S.A.
Abbreviations used: DRG, dorsal root ganglion; E. embryonic day; NGF, nerve growth factor.
263
264
M. E. EICHLER ET AL.
1986). However, cytosolic measurement of calcium shows no change in hepatocytes preceding irreversible death as determined by bleb formation on mitochondrial membranes (Lemasters et al., 1987). Although calcium appears to be a factor in cell death, its role in neuronal death during development or after trophic factor deprivation is not known. In the absence of NGF, chronic depolarization of the membrane potential by elevated levels of extracellular K+ enhances sympathetic neuronal survival in cell culture (Scott and Fisher, 1970). The beneficial effect in vitro of high extracellular Kf on NGF-deprived sympathetic neurons has been linked to influxes of calcium, via dihydropyridine-sensitive L-type Ca2+ channels, resulting in elevated [Ca2+]i(Koike et al., 1989). Koike and coworkers ( 1989) have hypothesized that calcium levels may be a factor in determining the dependency of sympathetic neurons to NGF and that, during development, alterations in [Ca2+Iimay affect this dependence. In cell culture, embryonic DRG neurons demonstrate a transient dependency on NGF for survival. This in vitro condition resembles the NGF-trophic dependence found in sensory neurons in vivo. Dissociated sensory neurons from embryonic DRG are highly NGF dependent initially, but dramatically lose their NGF dependence for survival after 3 weeks in culture (Greene, 1977; Eichler and Rich, 1989). Under such experimental conditions, neurotrophic support can be withdrawn in a standardized fashion from sensory neurons after specific lengths of time in cell culture and the death of neurons can be evaluated. The use of this trophic factor-dependence bioassay, coupled with the measurement of [Ca2'Ii, allows the correlation of [Ca2+Iiwith trophic factor dependence; also, the measurement of [Ca2+Iisubsequent to NGF withdrawal allows one to assess possible roles for the changes in calcium as the cells become "committed to die." MATERIALS AND METHODS Dissociated DRG neuronal cultures Primary dissociated cultures of embryonic-day-15 (E- 15) or day-20 (E-20) rat embryonic sensory neurons were prepared from Sprague-Dawley (Harlan, Indianapolis, IN, U.S.A.) rats. DRG were dissected and placed in L15 medium (GIBCO BRL) prior to dissociation in collagenase (1 mg/ml) (Worthington Biomedical) and trypsin (0.05%) (GIBCO BRL). Neurons were triturated and plated on a double layer of rat-tail collagen in 35-mm diameter petri dishes (Falcon), which had their centers replaced by a 25-mm diameter, 0.17mm thick, glass coverslip to facilitate the fluorescent measurements. Cultures were maintained in 90% basal medium Eagle (GIBCO BRL), 10% newborn bovine serum (Rehies), 100 pg/ml of both penicillin and streptomycin, 20 pA4 fluorodeoxyuridine, 20 p M uridine (Sigma), 1.4 nM L-glutamine, and 50 ng/ml of 2.5s mouse NGF. Cultures were kept in a 5% C 0 2 atmosphere at 37°C.
Neuronal bioassay for survival after NGF withdrawal Neurons were maintained in cell culture for periods up to 29 days. After 10 days in culture, the neurons were acutely J. Neurochem., Vol. 58, No. I , 1992
deprived on NGF by the use of pooled, guinea pig anti-mouse NGF antisera (Rich et al., 1984) that was heat-inactivated at 56°C for 20 min and that had a final titer of 4,000 against 5 ng/ml of mouse NGF as quantitated in a chick DRG explant bioassay (Fenton, 1970). Cultures were deprived of NGF by changing to standard medium as described above, but Without NGF and by adding antisera to a 2% final concentration. Neurons were routinely examined by phase-contrast microscopy for morphological changes; initial changes included perikaryal atrophy and vesicular formation within the cytoplasm. Eventually, membrane integrity was lost and neuronal somas disappeared.
Determination of intracellular calcium [Caz+Iimeasurements were determined after various periods in cell culture. Intracellular calcium was evaluated with ratio measurements ( R ) of fura-2 emissions (5 15-560 nm) when excited alternately by 340 k 10 nm and 380 +. 10 nm light (Grynkiewicz et al., 1985). DRG neurons were incubated with fura-2AM (Molecular Probes) for 1 h at 37°C and rinsed in physiological saline containing: 140 mMNaCl, 3 mMKCI, 10 mMNa HEPES, 40 mMglucose, 1.8 mMCaC12, and 0.8 mM MgCl2. Cells were warmed to 35°C and allowed to sit for 15 min before measurements were begun. Fluorescent intensity was measured with a Nikon 1.3 na 40X Fluor objective and a Nikon PI Photomultiplier with a Hammamatsu R1104 photodiode. The output was filtered at 100 Hz, digitized, and stored on a PC computer for further evaluation. All fluorescent measurements were corrected by subtracting the wavelength specific background fluorescence over collagen before computing ratios. Calcium concentrations were calculated according to the following equation (Grynkiewicz et al., 1985): [Ca2+li= KD X [(R - Rmin)/(Rmax
-
R)1 X
FdFmax
where Rminand R,, are the R values at nominally zero and maximal [Ca2'li, Fo and F,, are the emission intensities at 380 nm under these same conditions, and KD was taken to be 224 nM (Grykiewicz et al., 1985). R,,,, Rmin,Fo,and F,,, were determined from measurements of fura-2AM emissions in DRG and hippocampal neurons in the presence of 5 p A 4 ionomycin (Molecular Probes) to allow maximal influx of calcium and subsequently with addition of 17 m M EGTA to chelate calcium. Further calibrations were performed in solutions containing 4 pM fura-2 pentapotassium salt and either no added calcium or 10 mM calcium. The values obtained by these two methods were generally in agreement and were therefore combined to yield R,,,, R,i,, and FO/Fmax values of 24,0.8, and 7.6. Because the intensity of fura-2 emissions depends on the alignment of the fluorescent lamp, these constants changed after replacement of the 75W xenon bulb to 26, 0.8, and 6.9, respectively. Averaged [Ca2+Iilevels ( 18-50 cells per petri dish) were combined for all dishes sampled (n) and are reported as means k SEM.
Determination of neuronal commitment to die DRG neurons were grown in cell culture for 10 days in the presence of NGF. Neurons were then acutely deprived of NGF, by using polyclonal NGF antisera, for periods of 6, 12, 24, or 48 h before being refed with NGF-containing medium. Cells were inspected daily with phase-contrast microscopy to assess neuronal viability. Neuronal survival was assessed further by determination of protein metabolism on day 16 in culture with an [35S]methionineincorporation assay. The assay consisted of pulsing 'each well with translation-
INTRACELLULAR CALCIUM AND NGF DEPENDENCE grade [35S]methionine.The neurons were incubated for 6 h at 37°C in a 5% C 0 2 atmosphere and then lysed with NaOH; the trichloroacetic acid-precipitable counts were adsorbed on Whatman No. 3 filter paper. The radioactivity of the labeled proteins was measured on a scintillation counter. Data was corrected for nonspecific binding by subtracting counts from collagen-coated, cell-free wells (blank wells) (Eichler and Rich, 1989).
increasing to 163 f 5 nM by day 10 in culture. The [Ca2+Iireached 222 k 14 nMby day 19 in culture. No further significant increase in [Ca2+Iifor older E-20 neurons was noted. A comparison of the basal intracellular calcium level for cultured E-15 and E-20 neurons presented in Fig. 2 shows that [Ca2+Iilevels are dependent on absolute neuronal age and not the length of time in vitro.
Effect of high K+ on survival and [Ca2+Iiin E-15 DRG in cell culture E- 15 DRG were dissected, then dissociated with collage-
[Ca2+Iiafter NGF deprivation and neuronal commitment to die E- 15 neurons deprived of NGF on day 10 in cell culture appeared healthy by phase-contrast microscopy 24 h after NGF deprivation. However, by 48 h these neurons appeared vacuolated with granular-appearing neurites, changes consistent with cellular death. Concomitantly, [Ca'+Ii in 10-day-old neurons deprived of NGF for 24 h (1 1 days old) measured 140 f 3 nM. This [Ca2+Iiwas not statistically different than the [Ca2+Iiin 1 1-day-old neurons not deprived of NGF (Fig. 1, inset). However, by 48 h after NGF deprivation, [Ca"], had fallen to 101 k 4 &, which was lower than the [Ca2+Iiof 12-day-old,healthy control neurons (Fig. 1, inset). In parallel experiments, 1 0-day-old E- 15 neurons were deprived of NGF for 6, 12,24, or 48 h after which NGF was resupplied in the medium. Neurons deprived of NGF for only 6 or 12 h appeared healthy by phasecontrast microscopy at day 16 in cell culture. Yet, neurons deprived of NGF for 24 or 48 h, then refed with NGF, could not be saved and appeared dead by phasecontrast microscopy at day 16 in culture (Fig. 3). An [35S]methionineincorporation assay was performed to assess the effect of NGF deprivation on protein metabolism. By day 16 in culture, protein metabolism in E-15 neurons deprived of NGF for 24 or 48 h had decreased about 66% compared to controls; neurons deprived for only 6 or 12 h had 91% or loo%, respec-
nase, and plated on 35-mm petri dishes in the presence of NGF. On day 5 in culture, the medium was changed to either anti-NGF, anti-NGF high K+ (35 mM), or NGF (controls). To vary the K+ concentration of the medium, increased amounts of KCl were added, with the omission of corresponding amounts of NaCl, to maintain osmolarity. Neurons were examined by phase-contrast microscopy every 6 h to evaluate survival and [Ca2+Iiwas measured with fura-2AM at 24 h (6 days old). No [Ca2+Iimeasurements could be performed on the neurons exposed to anti-NGF alone because these neurons were dead by 24 h after NGF deprivation.
+
RESULTS Changes in [Ca2+]iin E-15 and E-20 DRG neurons in cell culture Five-day-old E- 15 DRG neurons grown in the presence of NGF had an [Ca2+]iof 91 & 3 nM (mean SEM, n = 6).The [Ca2+Iiof 10-day-old neurons (absolute age of 25 days) was 1 19 8 nM, and by 15 days in culture, their [Ca2+Iimeasured 171 -t 5 nM. The [Ca2+Iiprogressively increased and appeared to plateau at 201 k 4 nM after 21 days in culture. No further statistically significant increase in [Ca2+Iiwas noted, even for neurons in culture for greater than 1 month (Fig. 1). E-20 DRG neurons after 5 days in culture (absolute neuronal age, 25 days) had an [Ca2+]iof 123 k 4 nM,
*
+_
265
€15 DUG
FIG.1. lntracellular calcium measurements in E-
e
loo
1
I
# 75
I
10
I5
Days n -
6
7 5 5
B It
12
n . . m r - ~=~ ~4
5
10
5
I
ant'-NGF
--O
L
15 DRG neurons after various times in cell culture. Values are expressed as mean intracellular calcium per well k SEM. A range of 18-50 cells was measured per tissue culture well with an average of 21 ? 2 cells per well. The number of wells measured at each time was recorded as n. Inset represents intracellular calcium after NGF deprivation for 24 and 48 h in E-I5 neurons 10 days in culture. The x-axis of inset represents days in culture and nanti.NGF indicates the number of antiNGF-containing wells measured at each time point. Values of [Ca"], for 11-day-old neurons exposed to NGF or anti-NGF (inset) are not statistically different ( p > 0.05).Values of [CaZ+],for 12-day-old neurons exposed to NGF or anti-NGF are statistically different ( p
Relationship of Intracellular Calcium to Dependence on Nerve Growth Factor in Dorsal Root Ganglion Neurons in Cell Culture *Marc E. Eichler, ?Janet M. Dubinsky, and *?Keith M. Rich Departments of *Neurosurgery and j-Neurobiology, Washington University School of Medicine, St. Louis, Missouri, U.S.A.
Abstract: During development, neural crest-derived sensory neurons require nerve growth factor (NGF) for survival, but lose this dependency postnatally. Similarly, dissociated embryonic sensory neurons lose their NGF dependence during the first 3 weeks in cell culture. It has been hypothesized that, in sympathetic neurons, intracellular levels of calcium are related to trophic factor dependence. In vitro during the period in which embryonic-day- 15 sensory neurons become independent of NGF, intracellular calcium concentrations progressively increased in parallel to the decline in NGF dependence. This elevation of intracellular calcium was directly related to the absolute age of the neurons, not to the length of time in culture. Without NGF, immature sensory, i.e., dependent, neurons survived in the presence of high extracellular potassium, a condition that produces elevated intra-
cellular calcium. In another paradigm, measurements of intracellular calcium were determined in NGF-dependent neurons “committed to die” after NGF withdrawal. These measurements were determined prior to the time that extensive morphological changes, consistent with cell death, were noted by phase-contrast microscopy. No elevation in intracellular calcium was found in these dying neurons, but rather, a small decrease was observed prior to the disintegration of the neurons. These findings support the hypothesis that trophic factor dependence of neurons may be inversely related to levels of intracellular calcium. Key Words: Nerve growth factor-Intracellular calcium-Sensory neurons. Eichler M. E. et al. Relationship of intracellular calcium to dependence on nerve growth factor in dorsal root ganglion neurons in cell culture. J. Neurochem. 58, 263-269 (1992).
Neuronal death is a tightly regulated phenomenon that occurs in several neuronal subpopulations during development as a physiologic method of establishing the proper number of neurons that survive into the postnatal period (Oppenheim, 1989). During neuronal development, a naturally occurring cell death culls an overproduction of neurons to the population that is ultimately required to innervate specific targets. Molecular mechanisms that result in neuronal death under such circumstances are not known. Nerve growth factor (NGF) is a trophic agent crucial in the naturally occurring cell death that takes place in both dorsal root ganglion (DRG) and sympathetic neurons during embryogenesis. In vivo dependence on NGF for survival occurs as a transient phenomenon in the prenatal and early postnatal sensory neuron (Johnson et al., 1980). The mature DRG neuron is not dependent on NGF for survival (Lindsay, 1988), but does continue to ex-
press NGF receptors and requires NGF for biochemical and morphologic homeostasis (for concise review, see Johnson et al., 1986). Exogenously administered NGF is capable of preventing neuronal death after axotomy in both immature and adult rats (Yip et al., 1984;Rich et al., 1987). The intracellular events that regulate NGF dependence during maturation of DRG neurons remain unknown. Intracellular calcium concentration ([Ca2+]i) has been implicated in a final, common pathway in some forms of cell death (Schanne et al., 1979).Hypoxemicischemic neuronal injury has been linked to massive influxes of calcium into the cytosol from intracellular organelles and extracellular sources (Siesjo and Bengtsson, 1989). Activation of postsynaptic glutamate receptors on neurons has been associated with an influx of calcium intracellularly, resulting in adverse reactions leading to cell death (Choi, 1985; Rothman and Olney,
Received March 13, 1991; revised manuscript received May 24, 1991; accepted May 31, 1991. Address correspondence and reprint requests to Dr. K. M. Rich at Department of Neurological Surgery, Washington University School of Medicine, St. Louis, MO 631 10. U.S.A.
Abbreviations used: DRG, dorsal root ganglion; E. embryonic day; NGF, nerve growth factor.
263
264
M. E. EICHLER ET AL.
1986). However, cytosolic measurement of calcium shows no change in hepatocytes preceding irreversible death as determined by bleb formation on mitochondrial membranes (Lemasters et al., 1987). Although calcium appears to be a factor in cell death, its role in neuronal death during development or after trophic factor deprivation is not known. In the absence of NGF, chronic depolarization of the membrane potential by elevated levels of extracellular K+ enhances sympathetic neuronal survival in cell culture (Scott and Fisher, 1970). The beneficial effect in vitro of high extracellular Kf on NGF-deprived sympathetic neurons has been linked to influxes of calcium, via dihydropyridine-sensitive L-type Ca2+ channels, resulting in elevated [Ca2+]i(Koike et al., 1989). Koike and coworkers ( 1989) have hypothesized that calcium levels may be a factor in determining the dependency of sympathetic neurons to NGF and that, during development, alterations in [Ca2+Iimay affect this dependence. In cell culture, embryonic DRG neurons demonstrate a transient dependency on NGF for survival. This in vitro condition resembles the NGF-trophic dependence found in sensory neurons in vivo. Dissociated sensory neurons from embryonic DRG are highly NGF dependent initially, but dramatically lose their NGF dependence for survival after 3 weeks in culture (Greene, 1977; Eichler and Rich, 1989). Under such experimental conditions, neurotrophic support can be withdrawn in a standardized fashion from sensory neurons after specific lengths of time in cell culture and the death of neurons can be evaluated. The use of this trophic factor-dependence bioassay, coupled with the measurement of [Ca2'Ii, allows the correlation of [Ca2+Iiwith trophic factor dependence; also, the measurement of [Ca2+Iisubsequent to NGF withdrawal allows one to assess possible roles for the changes in calcium as the cells become "committed to die." MATERIALS AND METHODS Dissociated DRG neuronal cultures Primary dissociated cultures of embryonic-day-15 (E- 15) or day-20 (E-20) rat embryonic sensory neurons were prepared from Sprague-Dawley (Harlan, Indianapolis, IN, U.S.A.) rats. DRG were dissected and placed in L15 medium (GIBCO BRL) prior to dissociation in collagenase (1 mg/ml) (Worthington Biomedical) and trypsin (0.05%) (GIBCO BRL). Neurons were triturated and plated on a double layer of rat-tail collagen in 35-mm diameter petri dishes (Falcon), which had their centers replaced by a 25-mm diameter, 0.17mm thick, glass coverslip to facilitate the fluorescent measurements. Cultures were maintained in 90% basal medium Eagle (GIBCO BRL), 10% newborn bovine serum (Rehies), 100 pg/ml of both penicillin and streptomycin, 20 pA4 fluorodeoxyuridine, 20 p M uridine (Sigma), 1.4 nM L-glutamine, and 50 ng/ml of 2.5s mouse NGF. Cultures were kept in a 5% C 0 2 atmosphere at 37°C.
Neuronal bioassay for survival after NGF withdrawal Neurons were maintained in cell culture for periods up to 29 days. After 10 days in culture, the neurons were acutely J. Neurochem., Vol. 58, No. I , 1992
deprived on NGF by the use of pooled, guinea pig anti-mouse NGF antisera (Rich et al., 1984) that was heat-inactivated at 56°C for 20 min and that had a final titer of 4,000 against 5 ng/ml of mouse NGF as quantitated in a chick DRG explant bioassay (Fenton, 1970). Cultures were deprived of NGF by changing to standard medium as described above, but Without NGF and by adding antisera to a 2% final concentration. Neurons were routinely examined by phase-contrast microscopy for morphological changes; initial changes included perikaryal atrophy and vesicular formation within the cytoplasm. Eventually, membrane integrity was lost and neuronal somas disappeared.
Determination of intracellular calcium [Caz+Iimeasurements were determined after various periods in cell culture. Intracellular calcium was evaluated with ratio measurements ( R ) of fura-2 emissions (5 15-560 nm) when excited alternately by 340 k 10 nm and 380 +. 10 nm light (Grynkiewicz et al., 1985). DRG neurons were incubated with fura-2AM (Molecular Probes) for 1 h at 37°C and rinsed in physiological saline containing: 140 mMNaCl, 3 mMKCI, 10 mMNa HEPES, 40 mMglucose, 1.8 mMCaC12, and 0.8 mM MgCl2. Cells were warmed to 35°C and allowed to sit for 15 min before measurements were begun. Fluorescent intensity was measured with a Nikon 1.3 na 40X Fluor objective and a Nikon PI Photomultiplier with a Hammamatsu R1104 photodiode. The output was filtered at 100 Hz, digitized, and stored on a PC computer for further evaluation. All fluorescent measurements were corrected by subtracting the wavelength specific background fluorescence over collagen before computing ratios. Calcium concentrations were calculated according to the following equation (Grynkiewicz et al., 1985): [Ca2+li= KD X [(R - Rmin)/(Rmax
-
R)1 X
FdFmax
where Rminand R,, are the R values at nominally zero and maximal [Ca2'li, Fo and F,, are the emission intensities at 380 nm under these same conditions, and KD was taken to be 224 nM (Grykiewicz et al., 1985). R,,,, Rmin,Fo,and F,,, were determined from measurements of fura-2AM emissions in DRG and hippocampal neurons in the presence of 5 p A 4 ionomycin (Molecular Probes) to allow maximal influx of calcium and subsequently with addition of 17 m M EGTA to chelate calcium. Further calibrations were performed in solutions containing 4 pM fura-2 pentapotassium salt and either no added calcium or 10 mM calcium. The values obtained by these two methods were generally in agreement and were therefore combined to yield R,,,, R,i,, and FO/Fmax values of 24,0.8, and 7.6. Because the intensity of fura-2 emissions depends on the alignment of the fluorescent lamp, these constants changed after replacement of the 75W xenon bulb to 26, 0.8, and 6.9, respectively. Averaged [Ca2+Iilevels ( 18-50 cells per petri dish) were combined for all dishes sampled (n) and are reported as means k SEM.
Determination of neuronal commitment to die DRG neurons were grown in cell culture for 10 days in the presence of NGF. Neurons were then acutely deprived of NGF, by using polyclonal NGF antisera, for periods of 6, 12, 24, or 48 h before being refed with NGF-containing medium. Cells were inspected daily with phase-contrast microscopy to assess neuronal viability. Neuronal survival was assessed further by determination of protein metabolism on day 16 in culture with an [35S]methionineincorporation assay. The assay consisted of pulsing 'each well with translation-
INTRACELLULAR CALCIUM AND NGF DEPENDENCE grade [35S]methionine.The neurons were incubated for 6 h at 37°C in a 5% C 0 2 atmosphere and then lysed with NaOH; the trichloroacetic acid-precipitable counts were adsorbed on Whatman No. 3 filter paper. The radioactivity of the labeled proteins was measured on a scintillation counter. Data was corrected for nonspecific binding by subtracting counts from collagen-coated, cell-free wells (blank wells) (Eichler and Rich, 1989).
increasing to 163 f 5 nM by day 10 in culture. The [Ca2+Iireached 222 k 14 nMby day 19 in culture. No further significant increase in [Ca2+Iifor older E-20 neurons was noted. A comparison of the basal intracellular calcium level for cultured E-15 and E-20 neurons presented in Fig. 2 shows that [Ca2+Iilevels are dependent on absolute neuronal age and not the length of time in vitro.
Effect of high K+ on survival and [Ca2+Iiin E-15 DRG in cell culture E- 15 DRG were dissected, then dissociated with collage-
[Ca2+Iiafter NGF deprivation and neuronal commitment to die E- 15 neurons deprived of NGF on day 10 in cell culture appeared healthy by phase-contrast microscopy 24 h after NGF deprivation. However, by 48 h these neurons appeared vacuolated with granular-appearing neurites, changes consistent with cellular death. Concomitantly, [Ca'+Ii in 10-day-old neurons deprived of NGF for 24 h (1 1 days old) measured 140 f 3 nM. This [Ca2+Iiwas not statistically different than the [Ca2+Iiin 1 1-day-old neurons not deprived of NGF (Fig. 1, inset). However, by 48 h after NGF deprivation, [Ca"], had fallen to 101 k 4 &, which was lower than the [Ca2+Iiof 12-day-old,healthy control neurons (Fig. 1, inset). In parallel experiments, 1 0-day-old E- 15 neurons were deprived of NGF for 6, 12,24, or 48 h after which NGF was resupplied in the medium. Neurons deprived of NGF for only 6 or 12 h appeared healthy by phasecontrast microscopy at day 16 in cell culture. Yet, neurons deprived of NGF for 24 or 48 h, then refed with NGF, could not be saved and appeared dead by phasecontrast microscopy at day 16 in culture (Fig. 3). An [35S]methionineincorporation assay was performed to assess the effect of NGF deprivation on protein metabolism. By day 16 in culture, protein metabolism in E-15 neurons deprived of NGF for 24 or 48 h had decreased about 66% compared to controls; neurons deprived for only 6 or 12 h had 91% or loo%, respec-
nase, and plated on 35-mm petri dishes in the presence of NGF. On day 5 in culture, the medium was changed to either anti-NGF, anti-NGF high K+ (35 mM), or NGF (controls). To vary the K+ concentration of the medium, increased amounts of KCl were added, with the omission of corresponding amounts of NaCl, to maintain osmolarity. Neurons were examined by phase-contrast microscopy every 6 h to evaluate survival and [Ca2+Iiwas measured with fura-2AM at 24 h (6 days old). No [Ca2+Iimeasurements could be performed on the neurons exposed to anti-NGF alone because these neurons were dead by 24 h after NGF deprivation.
+
RESULTS Changes in [Ca2+]iin E-15 and E-20 DRG neurons in cell culture Five-day-old E- 15 DRG neurons grown in the presence of NGF had an [Ca2+]iof 91 & 3 nM (mean SEM, n = 6).The [Ca2+Iiof 10-day-old neurons (absolute age of 25 days) was 1 19 8 nM, and by 15 days in culture, their [Ca2+Iimeasured 171 -t 5 nM. The [Ca2+Iiprogressively increased and appeared to plateau at 201 k 4 nM after 21 days in culture. No further statistically significant increase in [Ca2+Iiwas noted, even for neurons in culture for greater than 1 month (Fig. 1). E-20 DRG neurons after 5 days in culture (absolute neuronal age, 25 days) had an [Ca2+]iof 123 k 4 nM,
*
+_
265
€15 DUG
FIG.1. lntracellular calcium measurements in E-
e
loo
1
I
# 75
I
10
I5
Days n -
6
7 5 5
B It
12
n . . m r - ~=~ ~4
5
10
5
I
ant'-NGF
--O
L
15 DRG neurons after various times in cell culture. Values are expressed as mean intracellular calcium per well k SEM. A range of 18-50 cells was measured per tissue culture well with an average of 21 ? 2 cells per well. The number of wells measured at each time was recorded as n. Inset represents intracellular calcium after NGF deprivation for 24 and 48 h in E-I5 neurons 10 days in culture. The x-axis of inset represents days in culture and nanti.NGF indicates the number of antiNGF-containing wells measured at each time point. Values of [Ca"], for 11-day-old neurons exposed to NGF or anti-NGF (inset) are not statistically different ( p > 0.05).Values of [CaZ+],for 12-day-old neurons exposed to NGF or anti-NGF are statistically different ( p
