The Journal of Neurophysiology Vol. 80 No. 1 July 1998, pp. 350-364
Copyright ©1998 by the American Physiological Society

Increased Calcium Buffering in Basal Forebrain Neurons During Aging

David Murchison and William H. Griffith

Department of Medical Pharmacology and Toxicology, College of Medicine, Texas A&M University Health Science Center, College Station, Texas 77843-1114

	ABSTRACT

Abstract Introduction Methods Results Discussion References

Murchison, David and William H. Griffith. Increased calcium buffering in basal forebrain neurons during aging. J. Neurophysiol. 80: 350-364, 1998. Alterations of neuronal calcium (Ca²⁺) homeostasis are thought to underlie many age-related changes in the nervous system. Basal forebrain neurons are susceptible to changes associated with aging and to related dysfunctions such as Alzheimer's disease. It recently was shown that neurons from the medial septum and nucleus of the diagonal band (MS/nDB) of aged (24-27 mo) F344 rats have an increased current influx through voltage-gated Ca²⁺ channels (VGCCs) relative to those of young (1-4.5 mo) rats. Possible age-related changes in Ca²⁺ buffering in these neurons have been investigated using conventional whole cell and perforated-patch voltage clamp combined with fura-2 microfluorimetric techniques. Basal intracellular Ca²⁺ concentrations ([Ca²⁺]_i), Ca²⁺ influx, Ca²⁺ transients ([Ca²⁺]_i), and time course of [Ca²⁺]_i were quantitated, and rapid Ca²⁺ buffering values were calculated in MS/nDB neurons from young and aged rats. The involvement of the smooth endoplasmic reticulum (SER) was examined with the SER Ca²⁺ uptake blocker, thapsigargin. An age-related increase in rapid Ca²⁺ buffering and [Ca²⁺]_i time course was observed, although basal [Ca²⁺]_i was unchanged with age. The SER and endogenous diffusible buffering mechanisms were found to have roles in Ca²⁺ buffering, but they did not mediate the age-related changes. These findings suggest a model in which some aging central neurons could compensate for increased Ca²⁺ influx with greater Ca²⁺ buffering.

	INTRODUCTION

Abstract Introduction Methods Results Discussion References

Intracellular calcium concentrations ([Ca²⁺]_i) regulate diverse neuronal processes including development and maturation, gene expression, cell death, synaptic plasticity, transmitter release, excitability, and others (Bertolino and Llinas 1992; Choi 1992; Clapham 1995; DeCoster 1995; Ginty 1997; Holliday and Spitzer 1991; Kostyuk 1989; Malenka et al. 1989; Smith and Augustine 1988; Trump and Berezesky 1992). The pervasive involvement of Ca²⁺ in neuronal function suggested that altered Ca²⁺ homeostasis might be the fundamental mediator of age-related changes in the nervous system (Gibson and Peterson 1987; Khachaturian 1984, 1994; Landfield 1987). Despite the popularity of such an explanation for age-related neuronal dysfunction, few definitive experiments have been conducted to measure age-related changes in I_Ca, in [Ca²⁺]_i or in Ca²⁺ buffering of central neurons (Verkhratsky and Toescu 1998).

Decreases in I_Ca with age have been reported in dentate granule cells (Reynolds and Carlen 1989) and dorsal root ganglion neurons (Kostyuk et al. 1993). Elevated basal [Ca²⁺]_i and reduced [Ca²⁺]_i transients ([Ca²⁺]_i) have been described in several types of aged neurons (Kirischuk et al. 1992; Verkhratsky et al. 1994) and synaptosomes from aged rat brain (Leslie et al. 1986; Martinez et al. 1988; Martinez-Serrano et al. 1992; Vitorica and Satrústegui 1986). Increased [Ca²⁺]_i amplitude has been observed in aged adrenergic neurons (Buchholz et al. 1996; Duckles et al. 1996). Decreased Ca²⁺ influx and/or buffering with age has been proposed to explain these findings. In contrast, an age-related decrease in basal [Ca²⁺]_i and [Ca²⁺]_i in various rat and mouse neurons has implied a reduced Ca²⁺ influx and increased buffering (Hartmann et al. 1994, 1996). Age-related increases in Ca²⁺ influx have been demonstrated in hippocampal CA1 pyramidal neurons (Campbell et al. 1996; Pitler and Landfield 1990; Thibault and Landfield 1996) and in basal forebrain neurons (Murchison and Griffith 1995, 1996), suggesting impaired Ca²⁺ homeostasis with age. Altered Ca²⁺ homeostasis could explain abnormalities in Ca²⁺-dependent processes in aged animals (Barnes 1994; Baskys et al. 1990; Gibson and Peterson 1987; Landfield et al. 1978; Michaelis 1994; Norris et al. 1996).

This investigation examines the hypothesis of impaired Ca²⁺ homeostasis with age due to increased Ca²⁺ influx in the medial septum and nucleus of the diagonal band (MS/nDB) in Fisher 344 rats. The MS/nDB contains cholinergic and noncholinergic neurons that project to the neocortex and hippocampus (Dutar et al. 1995; Fibiger 1982). Functionally, these cells have been implicated in cognitive processes such as attention and some forms of memory (Bartus et al. 1982; Blokland 1996; Dutar et al. 1995; Olton et al. 1991; Zola-Morgan and Squire 1993). Changes occur in these cells with age and in Alzheimer's disease (Coyle et al. 1983; Decker 1987; Dutar et al. 1995; Fischer et al. 1989; Lamour et al. 1987; Markowska et al. 1995; Miettinen et al. 1993). Some of these changes could arise from alterations of neuronal Ca²⁺ homeostasis in these cells. Rat basal forebrain Ca²⁺ buffering has been examined in cultured (Bleakman et al. 1993) and in acutely dissociated neurons (Tatsumi and Katayama 1993). Operationally, we have divided Ca²⁺ buffering into rapid and slow systems. Rapid Ca²⁺ buffering reduces free [Ca²⁺]_i to a fraction of the amount that entered the cell and is reflected in the ratio of Ca²⁺ influx to peak [Ca²⁺]_i. Slow buffering systems enable the cell to restore basal [Ca²⁺]_i over many seconds.

Calcium homeostasis was assessed by quantitating the relationship between Ca²⁺ entry through voltage-gated Ca²⁺ channels (VGCCs) previously described in these cells (Allen et al. 1993; Murchison and Griffith 1995, 1996) and the resulting [Ca²⁺]_i measured with fura-2 microfluorimetry (Grynkiewicz et al. 1985). Similar methods have been used elsewhere to describe Ca²⁺ buffering in several cell types (Fiero and Llano 1996; Neher and Augustine 1992; Tatsumi and Katayama 1993; Tse et al. 1994; Zhou and Neher 1993). This report presents evidence for an increase in rapid Ca²⁺ buffering by aged MS/nDB neurons. Aged neurons also are shown to have unaltered basal [Ca²⁺]_i and longer [Ca²⁺]_i duration. The roles of rapidly diffusible endogenous buffers and of smooth endoplasmic reticulum (SER) also are described.

	METHODS

Abstract Introduction Methods Results Discussion References

Experimental animals

Male Fischer 344 rats were purchased from Harlan (Indianapolis, IN, NIA breeding colony). Data were collected from 166 neurons from 61 young adults (1-4.5 mo, mean = 2.0 mo) and 114 neurons from 33 aged rats (24-27 mo, mean = 24.8 mo). Animals took food and water ad libitum and were maintained on a 12-h light:dark cycle. Handling and care of the animals was in accordance with policies of the American Physiological Society and Texas A&M University.

Acutely dissociated neurons

Individual MS/nDB neurons were obtained using methods described previously (Murchison and Griffith 1996). Briefly, isoflurane (Anaquest, Liberty Corner, NJ)-anesthetized rats were decapitated. Coronal brain slices (450 µm) were microdissected to isolate the MS/nDB region and enzymatically treated (trypsin ~0.7 mg/ml; Sigma Type XI). Cells were dispersed onto the glass floor of the recording chamber after gentle trituration. The recording chamber was mounted on an inverted microscope (Axiovert 100, Zeiss) and continuously perfused at a rate of ~2 ml/min, resulting in a bath turnover of <30 s. Experiments were performed at 20-21°C.

Electrical recording

Standard whole cell patch (Hamill et al. 1981) and perforated patch (Horn and Marty 1988) voltage-clamp techniques were employed. The standard patch was used also for whole cell current-clamp recordings. Electrodes were pulled from 1.5 OD glass tubing (No. 7052 Garner Glass, Claremont, CA) on a Flaming-Brown P-87 electrode puller, heat polished to resistances of 4-10 M, and coated with wax to reduce electrode capacitance. An Axoclamp 200A amplifier and pClamp software (Axon Instruments, Foster City, CA) were used. Data acquisition and command voltage generation were converted using a Digidata 1200 interface (Axon Instruments). Data were filtered at 1 kHz and sampled at 0.2-4.0 kHz. Cell capacitance was determined by canceling the capacitance transients and reading the value from the potentiometer on the amplifier. Series resistance was compensated 70-85% for conventional whole cell recording and 50-75% for perforated-patch recording. The mean series resistance was 23.6 ± 1.1 (SE) M (n = 34) for perforated-patch recordings. Series resistance values were not always recorded for conventional whole cell patch experiments but were between 8 and 20 M. Voltage-clamped cells were held at V_h = 60 mV, and VGCCs were activated by steps to 0 mV. Steps administered to measure the Ca²⁺ buffering capacity (see further text) were 2-5 min apart to allow the [Ca²⁺]_i to return to baseline before the next depolarization. Sequences of data collection were terminated if the standing leak current increased by >100 pA, exceeded 200 pA total, if the measured [Ca²⁺]_i did not return to near baseline after an elevation, or if the baseline exceeded 250 nM. Typically, leak currents remained stable between 20 and 100 pA, and [Ca²⁺]_i baselines stayed between 50-150 nM for the duration of the experiment (1.5 h). For microfluorimetric measurements, electrodes were positioned out of the excitation light path.

Solutions and drugs

BATH SOLUTIONS. Cells in the recording chamber were perfused with "normal" physiological solution, consisting of (in mM) 140 NaCl, 3 KCl, 2 CaCl₂, 1.2 MgCl₂, 33 D-glucose, and 10 N-2-hydroxyethylpiperazine-N'-2-ethanesulfonic acid (HEPES) (pH 7.4 with NaOH, osmolarity 310-330 mOsm). This solution was used for recording during current-clamp experiments and when recording fura-2 signals from unclamped cells. For voltage-clamp experiments isolating I_Ca, the recording solution was switched to (in mM) 132 NaCl, 2 CaCl₂, 2 MgCl₂, 33 D-glucose, 10 tetraethylammonium (TEA) chloride, 0.0005 tetrodotoxin (Calbiochem, La Jolla, CA), and 10 HEPES, (pH, 7.4; osmolarity 310-330 mOsm). Thapsigargin (400 nM, Alomone Labs, Jerusalem, Israel) was dissolved in ethanol (0.08% final concentration) and added to the bath. Cadmium (100 µM) was added to the recording solution to block voltage-gated Ca²⁺ channels for calculation of the Cd²⁺ sensitive current. All chemicals were obtained from Sigma (St. Louis, MO) unless otherwise indicated.

INTRACELLULAR SOLUTIONS. For conventional whole cell current clamp recording, the internal pipette solution contained (in mM) 136 K-gluconate, 25 NaCl, 2 MgCl₂, 0.05 ethylene glycol-bis(-aminoethyl ether)-N,N,N',N'-tetraacetic acid (EGTA; Fisher Scientific, Pittsburgh, PA), 0.1 GTP, 4 ATP, and 10 HEPES, (pH 7.2 with KOH; osmolarity 280-300 mOsm). The pipette solution for conventional whole cell voltage-clamp recording contained (in mM) 140 CsCl, 20 TEA-Cl, 0.1 GTP, 4 ATP, and 10 HEPES, (pH 7.2 with CsOH; osmolarity 280-300 mOsm). In a few experiments, an alternative low chloride solution was used containing (in mM) 122 CsAcetate, 15 CsCl, 20 HEPES, 10 TEA, 4 ATP, and 0.1 GTP (pH 7.2 with CsOH; osmolarity 280-300 mOsm). The pipette solutions for perforated-patch recording contained either (in mM) 20 N-methyl-D-glucamine, 40 CsCl, 30 TEA-Cl, 140 HEPES, 5 EGTA, and 2 MgCl₂ or 122 CsAcetate, 15 CsCl, 20 HEPES, 10 TEA, and 5 EGTA (both pH 7.2 with CsOH; osmolarity 280-300 mOsm). The perforated-patch solution contained nystatin (400-500 µg/ml dissolved in 0.5% dimethylsulfoxide, DMSO).

FURA-2 LOADING. Cells used for perforated-patch or unclamped recordings were loaded with the permeable ester of fura-2 (fura-2 AM, 0.5 µM) from the bath using a 3- to 3.5-min application followed by 25-50 min washout to allow for fura de-esterification. Fura-2 AM was dissolved in 0.05% DMSO with 12.5 µg/ml pluronic. For conventional whole cell recording, cells were loaded with the impermeable salt of fura-2 (pentapotassium salt, fura-2 K⁺₅) via the patch pipette for 4-6 min before data acquisition. Both forms of fura-2 and pluronic were obtained from Molecular Probes (Eugene, OR).

Intracellular [Ca²⁺] measurements

A dual wavelength ratiometric fura-2 microfluorimetry system was used to measure mean [Ca²⁺]_i in the somata of selected neurons. Cells were positioned so that the somata maximally occupied the excitation field, and the focal point was centered in the cytoplasm. The excitation field had a diameter of 10 µm, which is smaller than the smallest dimension of most somata. Illumination was provided by a xenon arc lamp (Zeiss), and fluorescence was excited alternately at wavelengths of 340 and 380 nm by means of a rotating (40 Hz) filter wheel that generated a signal synchronized to the excitation wavelength. The fura-2 fluorescence signal was monitored by a photomultiplier tube (Hamamatsu) with a 510- to 560-nm band-pass filter. The output of the photomultiplier (340 and 380 wavelength samples) was directed to separate sample and hold circuits, where an analog divider circuit continuously computed the ratio of the sampled 340 and 380 nm signals after analog subtraction of background fluorescence at each wavelength. To minimize bleaching of the dye and potential photo damage, neurons were illuminated only during data acquisition episodes. Ambient light levels were minimized, and neurons were visualized under low intensity near infra-red light.

Both in vivo and in vitro calibrations were used to estimate [Ca²⁺]_i. Calibrated solutions of known [Ca²⁺] (Molecular Probes) and calibrated pipette solutions containing known [Ca²⁺] and [EGTA] corrected for pH, temperature, and ionic strength were prepared with 1-50 µM fura-2 K⁺₅. The fluorescent intensities associated with 340 and 380 nm excitation of droplets of these solutions (in vitro) and of cells loaded with the calibrated solutions via whole cell patch (in vivo) were measured and curves relating the 340/380 ratio to [Ca²⁺] were constructed. The values obtained in these calibrations were used to convert the experimental fluorescent intensity ratios to [Ca²⁺] over the physiological range of [Ca²⁺]_i using the equation: [Ca²⁺] = K_dB(R  R_min)/(R_max  R), where K_d is the dissociation constant of fura-2, B equals 380_min/380_max (a property of the fluorescent system), R_min = 340/380 in the absence of Ca²⁺, R_max = 340/380 in high Ca²⁺, and R = 340/380 measured experimentally (Grynkiewicz et al. 1985). The calibration values obtained were remarkably similar for R_max and B no matter what method was employed, whereas the in vivo R_min was consistently lower than that measured in vitro. During the 2 yr that these experiments were conducted, there was about a 13% increase in the [Ca²⁺] corresponding to a given ratio over the range of 0-600 nM attributable to the increase in the B value with age of the system. For this report, the [Ca²⁺]_i was estimated using averaged values for R_max (6.22) and B (9.0) and in vivo values for R_min (0.20) and K_d. R_max was measured at calculated [Ca²⁺]s between 1.35 and 39.4 µM. In vivo K_d values were calculated to be 242 nM in young cells and 257 nM in aged cells, thus separate calibration curves were used for the two age groups. We furthermore modified the calibration curves for the fura-2 AM loaded cells to reflect a presumed elevated viscosity of cytoplasm relative to pipette solution as described by Poenie (1990). Ambient background fluorescence and cellular autofluorescence were subtracted. Groups of experiments were conducted concurrently in young and aged neurons.

There are several potential sources of error associated with the use of fluorescent Ca²⁺-sensitive probes that have been taken into consideration (Kao 1994; Moore et al. 1990; Sturek et al. 1991). The accuracy of the [Ca²⁺] corresponding to the fluorescence ratios is subject to a large number of variables regardless of the methods used for calibration, and attempts to provide corrections may themselves be liable to introduce errors (Neher 1995). For the purpose of comparing Ca²⁺ buffering between young and aged cells, small errors in the reported [Ca²⁺]_i are less important than the relative differences observed between the two age groups. In this regard, the [Ca²⁺]_i values we report are intended to be recognized as estimates. Comparative evidence indicates that these are good estimates: the resting [Ca²⁺]_i, the [Ca²⁺]_i and the buffering values reported here are well within the accepted physiological ranges for such values in neurons and some other cells (Neher 1995). Additionally, there appeared to be little evidence of incomplete hydrolysis or compartmentalization with the fura-2 AM (Zhou and Neher 1993). The short loading time and low concentration of fura-2 AM combined with the low temperature and long de-esterification time minimizes these possible sources of error. Possible compartmentalization was examined by rupturing the membrane of a cell loaded with fura-2 AM and recorded with perforated patch and checking for residual fluorescence after washout of the dye.

CALCULATING RAPID BUFFERING CAPACITY. A modification of the method of Hille and colleagues (Tse et al. 1994) was used to calculate buffering strength , which is a partition coefficient describing the ratio of buffer-bound to free ion. By this method

(1)

where: Ca²⁺_int is the integral of Ca²⁺ influx (charge from measured I_Ca), v is the cell volume, [Ca²⁺]_i is the measured change in concentration of intracellular free Ca²⁺, [SCa] is the change in concentration of Ca²⁺ bound to endogenous buffers, and [BCa] is the change in concentration of Ca²⁺ bound to exogenous buffers (fura-2). Equation 1 can be rewritten

(2)

 is the sum of the endogenous buffering strength (_S) and exogenous buffering strength (_B). Note that in recordings made in conventional whole cell configuration (i.e., not perforated) _S represents the nondiffusible endogenous buffering strength. The slope of the [Ca²⁺]_i versus Ca²⁺ entry plot (Figs. 3, 4, 7, and 8) is therefore the quantity 1/(1 + ). The cellular Ca²⁺ buffering value () is the reciprocal of the slope of the linear portion of the plot minus one. This buffering value represents the ratio of buffer bound to free Ca²⁺ ions. Only those cells having at least three data points in the linear portion of the plot were included in the analysis.

View larger version (22K): [in this window] [in a new window]   FIG. 3. Calculation of rapid Ca2+ buffering values in individual cells. A: representative ICa records and corresponding fluorescent ratio [Ca2+]i records from a young (2 mo) neuron. This cell was recorded in conventional whole cell mode with 50 µM fura-2 K+5 in the patch pipette. The voltage steps (60 to 0 mV) shown were of 75-, 200-, and 400-ms durations. B: example of superimposed ICa and corresponding fluorescent ratio [Ca2+]i records from an aged (25 mo) neuron. Recording as in A except with 100 µM fura-2 K+5. Shown are responses to voltage steps of 25-, 50-, 100-, 200-, and 500-ms duration. C: data from the cells shown in A and B plotted in the [Ca2+]i vs. Ca2+ entry form used for calculation of the rapid buffering slopes. , data from the young cell; , data from aged cell. Ca2+ entry was calculated from the integrals of the ICa and the estimated cell volume as described in METHODS. , linear portions of the curve from which the slopes were measured; - - -, nonlinear regions that are not used to calculate the rapid buffering slope. These slopes reflect the extent to which free Ca2+ entering the cell is rapidly buffered. In this example, only 1 of every 346 Ca2+ ions entering the young cell remained free, thus the smaller the slope, the greater the buffering.

View larger version (26K): [in this window] [in a new window]   FIG. 4. Comparison of rapid Ca2+ buffering in young and aged neurons recorded with conventional whole cell patch clamp. A: superimposed fura-2 fluorescent ratio records and calibrated [Ca2+]i for a young (1 mo) and an aged (24.5 mo) neuron recorded as in Fig. 3 with 50 µM fura-2 K+5. , prolonged return to baseline seen in aged cells. B: [Ca2+]i/Ca2+ entry plot of the data in A. Slope of the aged neuron is smaller, and so it has a greater buffering value than the young cell. Despite greater Ca2+ entry in the aged cell, the [Ca2+]i never reaches the levels observed in the young cell. Plots like these were used to measure the buffering slopes for the comparison graphed below. Only cells with 3 data points in the linear () portion of the plot were included in the analysis. C: histogram showing greater rapid buffering in aged neurons at 3 concentrations of fura-2 K+5. Values shown are the reciprocals of the calculated slopes, which are equal to the rapid buffering values plus one (see METHODS). Greater values indicate greater buffering. Values are shown as means ± SE. For each concentration of fura, the buffering values of the aged neurons were significantly (P < 0.03) greater than those of the young cells.

View larger version (25K): [in this window] [in a new window]   FIG. 7. Effects of thapsigargin on rapid buffering in a young neuron. This example from a 1.5-mo-old animal recorded with perforated patch shows the superimposed ICa and fluorescent ratio records with the corresponding [Ca2+]i/Ca2+ entry plot for the control (A) and after thapsigargin has been used to block the SER Ca2+ pump (B). Thapsigargin treatment has almost no effect on the slope of the linear portion of the plot, but the deviation from linearity at higher levels of Ca2+ entry is reduced. Straight line was fit to the 1st 5 data points (records depicted above plots), while the curve was fit to all the points (longer step durations not shown in top records).

View larger version (25K): [in this window] [in a new window]   FIG. 8. Effects of thapsigargin on rapid buffering in an aged neuron. This example from a 24-mo-old animal recorded with perforated patch displays the superimposed ICa and fluorescent ratio records with the corresponding [Ca2+]i/Ca2+ entry plot for the control (A) and after thapsigargin has been used to block the SER Ca2+ pump (B). Details as in Fig. 7.

A [Ca²⁺]_i versus Ca²⁺ entry plot was constructed for each cell tested by activating I_Ca for several durations (thus generating several levels of Ca²⁺ entry) and measuring the resulting [Ca²⁺]_i. The total Ca²⁺ entry was calculated by integrating the Cd²⁺-sensitive I_Ca over time and normalizing for cell volume (v). Cell volume was calculated from the membrane capacitance (pF) assuming the capacitance of biological membranes to be 1 µF/cm² and the cell to be spherical. It is recognized that this is an overestimate of the accessible cell volume (Neher 1995) but it defines the minimum [Ca²⁺] expected from the measured Ca²⁺ charge and is constant between the age groups assuming a similar morphological distribution (see Table 1). The data also were analyzed by an alternative method that quantitates the Ca²⁺ influx by normalizing the Ca²⁺ charge to the cell capacitance (Q_n = pC/pF), assuming equivalent accessible cell volumes (see Fig. 5).

View larger version (15K): [in this window] [in a new window]   FIG. 5. Buffering values improve with increasing voltage step duration and Ca2+ influx. Ratio of the [Ca2+]i to the normalized charge (Qn) is plotted against the voltage step duration. Qn is the integrated ICa (pC) divided by the cell capacitance (pF). In this method, the smaller the [Ca2+]i/Qn ratio, the better the buffering. A: data from the conventional whole cell recordings with fura-2 K+5 are pooled by age and plotted against step duration on a log scale. For young cells, n = 10-39, and for aged cells, n = 12-29. B: similar relationship is seen in the data from perforated patch recordings. For young cells, n = 9-13, and for aged cells, n = 5-8. * Significant difference between the buffering relationships in young and aged neurons (P < 0.01, 2-way ANOVA).

Data analysis

PARAMETERS FOR [Ca²⁺]_i MEASUREMENTS. The resting [Ca²⁺]_i for cells loaded with fura-2 AM was determined before contact with the patch electrode or any other manipulations to the cell and was taken as the mean ratio in a long (several seconds) segment of the fluorescence ratio signal that was uninterrupted by spontaneous events. The resting [Ca²⁺]_i for cells loaded with fura-2 K⁺₅ was measured in cells clamped at V_h = 60 mV before any test depolarizations and 4-5 min after patch rupture. The peak [Ca²⁺]_i was measured by meaning the ratio values in the region between the two largest excursions in the 340/380 fluorescence ratio signal, converting the mean peak ratio to a [Ca²⁺]_i and subtracting the baseline [Ca²⁺]_i. The baseline [Ca²⁺]_i was measured as the mean value during the 300-600 ms period immediately before the onset of the [Ca²⁺]_i transient. The time course of recovery to baseline [Ca²⁺]_i was assessed relative to the peak [Ca²⁺]_i (s/nM) for each transient and was measured from the onset of the transient to the point where the ratio signal first equaled the previous baseline. All fluorescence ratio signals were low-pass filtered post hoc (7 point boxcar averaging).

I_Ca measurements

All I_Ca measurements were converted to charges by integrating the cadmium-sensitive currents.

Statistical analysis

Comparisons were performed using either independent or paired two-tailed t-tests and one- and two-way analyses of variance (ANOVA) with significance determined by P < 0.05. Values are supplied as means ± SE.

	RESULTS

Abstract Introduction Methods Results Discussion References

Detailed investigations of age-related changes in VGCCs from the MS/nDB of F344 rats have been conducted in this lab (Murchison and Griffith 1995, 1996). In those studies, we demonstrated that our techniques permit recording from similar neuronal populations in both young and aged animals. In Fig. 1, we show that both types of electrophysiologically identifiable neurons [slow and fast afterhyperpolarization (sAHP and fAHP)] (Gorelova and Reiner 1996; Griffith and Matthews 1986; Markram and Segal 1990) are present in our young and aged samples. Furthermore, our lab has demonstrated comparable physiology between acutely dissociated cells from young and aged animals in -aminobutyric acid systems (Griffith and Murchison 1995) and excitatory amino acid systems (Jasek and Griffith 1998). These findings strongly support the conclusion that the acute dissociation procedure does not introduce a sampling artifact or compromise data for young and aged neuronal populations. Table 1 shows the morphological distribution of young and aged cells used to calculate buffering values in this investigation. In our previous studies of VGCCs, Ba²⁺ was used as the charge carrier through the Ca²⁺ channel. The principal results of Murchison and Griffith (1996) regarding age-related changes in high-voltage-activated currents (HVA) were confirmed in this study in which Ca²⁺ carried the charge. Specifically, there was no difference in initial HVA charge density (25-ms duration, Table 1), and there was a decrease in fast and slow current inactivation with age (reflected as a decrease in the charge density of young relative to aged at longer step durations in Table 1, P < 0.01, 2-way ANOVA).

View larger version (30K): [in this window] [in a new window]   FIG. 1. Current clamp with conventional whole cell patch in acutely dissociated young and aged neurons. Different cell types present in the medial septum and nucleus of the diagonal band (MS/nDB) can be identified by their physiological properties (Gorelova and Reiner 1996; Griffith and Matthews 1986). These neuronal cell types can be recorded from both young and aged acutely dissociated preparations in proportions similar to those observed in slice preparations. A: examples of the slow afterhyperpolarization inward rectification (sAHP-IR) cholinergic cell type from young (2 mo) and aged (24 mo) animals. , diagnostic features: bottom left, inward rectification; bottom right, slow afterhyperpolarization; top, prominent afterdepolarization. Slow firing rate of these cells is illustrated below. B: examples of the rapidly firing fast afterhyperpolarization (fAHP) cell type from young (2 mo) and aged (26 mo) animals. Comparable current steps were applied to the young and aged cells.

Despite the probability of increased Ca²⁺ influx with age, due to an increased low-voltage-activated current density (Murchison and Griffith 1995) and decreased HVA current inactivation (Murchison and Griffith 1996), there was no age-related difference in the resting [Ca²⁺]_i in either unclamped cells loaded with fura-2 AM or in voltage-clamped cells loaded with fura-2 K⁺₅ (Table 1). This result suggests that there may be a compensatory increase in buffering capacity of aged neurons to offset the increased Ca²⁺ influx and maintain [Ca²⁺]_i at normal levels. The 80-100 nM basal [Ca²⁺]_i reported here for MS/nDB neurons is comparable with that observed in other brain regions (Miller 1991) and in acutely dissociated neurons from another rat forebrain region, the nucleus basalis (Tatsumi and Katayama 1993). Basal [Ca²⁺]_i of ~100 nM has been reported also for rat dorsolateral septal neurons impaled with sharp microelectrodes in a thin slice (Zheng et al. 1996) and for cultured septal neurons (Bleakman et al. 1993).

More evidence of a possible compensatory Ca²⁺ buffering increase with age was observed from experiments such as those shown in Fig. 2. Depolarization-induced Ca²⁺ influx consistently produced a greater [Ca²⁺]_i in young cells. This was true for unclamped cells depolarized by 50 mM K⁺ (not shown) and for cells depolarized in current clamp (Fig. 2A) or voltage clamp (Fig. 2B). Even when aged cells showed greater numbers of spikes (Fig. 2A) or larger Ca²⁺ influx (Fig. 2B), the peak [Ca²⁺]_i was smaller.

View larger version (29K): [in this window] [in a new window]   FIG. 2. Whole cell current clamp and voltage clamp with fura-2 ratiometric measurements of [Ca2+]i in MS/nDB neurons from young and aged rats. A: current-clamp recordings from a young (1.5 mo) and an aged (26 mo) neuron with a pipette containing 50 µM fura-2 K+5. Top: spiking in response to a small depolarizing current step. Bottom: fura-2 fluorescence ratio and calibrated [Ca2+]i. Peak [Ca2+]i was smaller in the aged cell despite the greater number of spikes. B: voltage-clamp recordings from a young (1 mo) and an aged (24.5 mo) neuron with a pipette containing 50 µM fura-2 K+5. Top: ICa evoked by 25-ms steps to 0 mV from Vh = 60. Bottom: corresponding [Ca2+]i. Charge and Ca2+ entry were calculated as detailed in METHODS. Although more Ca2+ entered the aged cell, the peak [Ca2+]i was smaller. In this and all figures shown, the Cd2+-sensitive ICa is depicted.

The relationship between Ca²⁺ influx and [Ca²⁺]_i was quantitated for young and aged cells in a series of conventional whole cell voltage-clamp experiments using different concentrations of fura-2 K⁺₅. These experiments involve the generation of different levels of Ca²⁺ influx by activating VGCCs for different durations and measuring the resulting Ca²⁺ entry and corresponding [Ca²⁺]_i, as detailed in the methods. Examples of this type of experiment are shown in Fig. 3. A sequence of current records and the resulting fluorescence ratio signals are shown for a young neuron in Fig. 3A, and superimposed records are shown for an aged neuron in Fig. 3B. When the calculated Ca²⁺ entry is plotted against the measured [Ca²⁺]_i, as in Fig. 3C, the slopes of the linear portions of the plots () approximate the reciprocal of the Ca²⁺ buffering capacity of the cell under these recording conditions. Thus the smaller the slope, the greater the buffering value. These slopes reflect the extent to which free Ca²⁺ entering the cell is rapidly buffered. The nonlinear portions that are not used for calculating the slope are shown as - - -.

Because the indicator dye (fura-2) acts as an exogenous buffer, thus influencing the process of interest (endogenous buffer), we attempted to use as low a concentration of indicator as practical. However, too little indicator does not provide a sufficient signal/noise ratio and does not preclude saturation by a large Ca²⁺ influx. Therefore, three concentrations (10, 50, and 100 µM) of fura-2 K⁺₅ were tested. The data from these experiments are presented in Fig. 4. Superimposed fluorescence ratio records from a young and an aged neuron are displayed in Fig. 4A and the corresponding [Ca²⁺]_i/Ca²⁺ entry plot is shown in Fig. 4B. Despite the larger levels of Ca²⁺ entry attained in the aged cell, the [Ca²⁺]_i never reaches the levels achieved by the young neuron, indicating a greater rapid Ca²⁺ buffering capacity of the aged cell. Also note the delay in the recovery to baseline of the fluorescent ratio signal in the aged neuron ().

Figure 4C graphs the mean rapid buffering values for young and aged neurons at each of the three concentrations of fura-2 K⁺₅. In each case, the buffering values in the aged neurons are significantly greater (P < 0.03) than those in the young neurons. The values for 10 µM fura-2 K⁺₅ were 311 ± 33 (n = 10) in young and 501 ± 76 (n = 8) in aged neurons. The respective values at 50 µM were 195 ± 15 (n = 13) and 421 ± 103 (n = 10), and at 100 µM they were 162 ± 26 (n = 8) and 440 ± 81 (n = 6). Judging from a few experiments using high concentrations of fura-2 K⁺₅ (data not shown), it appears that 200-300 µM fura-2 K⁺₅ is required to outcompete the nondiffusible endogenous buffer in MS/nDB neurons. This agrees well with the findings in rat nucleus basalis neurons (Tatsumi and Katayama 1994) and compares with an estimate of 70 µM fura-2 to equal the endogenous buffer of bovine adrenal chromaffin cells (Zhou and Neher 1993).

As the 10 µM fura-2 K⁺₅ buffering values are the most likely to be in error due to the low fluorescent intensity and possible saturation, the 50 and 100 µM values probably represent conditions in which the fura-2 acts primarily as an indicator without contributing buffering capacity much in excess of the endogenous capacity. This conclusion is supported by the evidence that the buffering values are not different for the 50 and 100 µM fura-2 concentrations. This implies that most of the buffering is occurring through endogenous processes at these concentrations of indicator. Thus the endogenous rapid Ca²⁺ buffering strengths of intact MS/nDB neurons are estimated to be near the buffering values calculated from the 50 and 100 µM data. These values are in the middle to high end of those reviewed by Neher (1995) and suggest that rapid Ca²⁺ buffering is relatively strong in rat MS/nDB neurons. The only other study providing a quantitative assessment of the endogenous Ca²⁺ buffering strength in basal forebrain neurons gave an estimated buffering value of ~150 for acutely dissociated nucleus basalis neurons from 6- to 18-day-old Wistar rats (Tatsumi and Katayama 1994).

Another way to analyze the buffering value data is by plotting the ratio of the [Ca²⁺]_i to the normalized charge (Q_n = pC/pF) against the voltage step duration, as in Fig. 5. In this method, the smaller the [Ca²⁺]_i/Q_n ratio, the better the buffering. When the data from the conventional whole cell recordings with fura-2 K⁺₅ are pooled by age and plotted against step duration on a log scale (Fig. 5A), there is an approximately linear relationship between the degree of buffering (represented by the [Ca²⁺]_i/Q_n ratio) and the Ca²⁺ influx (which increases with step duration). Thus larger Ca²⁺ loads are buffered better in both young and aged cells. Greater Ca²⁺ influx and duration of influx apparently permits more complete activation of the available cellular buffering mechanisms, while fewer mechanisms operate on smaller and briefer Ca²⁺ influx. The buffering relationship measured in this way was significantly better for the aged cells (P < 0.01 2-way ANOVA). The findings for cells recorded with perforated patch and loaded with fura-2 AM are parallel and are presented in Fig. 5B.

To begin to explore the possible Ca²⁺ buffering mechanisms that may mediate the observed age-related changes, perforated-patch methods were used. The perforated patch with fura-2 AM allows recording of Ca²⁺ influx and [Ca²⁺]_i with minimal disruption of the intracellular buffering mechanisms. With this type of recording, diffusible buffer is not lost as it may be with conventional whole cell patch recording. A further advantage of the perforated method is the stability of I_Ca over time (e.g., lack of rundown), which permits a uniform Ca²⁺ influx throughout the experiment. Fura-2 AM loaded by our protocol gave a fluorescent intensity similar to that of 10 µM fura-2 K⁺₅, in agreement with Tatsumi and Katayama (1993).

The perforated-patch data confirm the results of the previous experiments. The Ca²⁺ entry/[Ca²⁺]_i buffering values were significantly greater in the aged neurons (young: 370 ± 59, n = 13; aged: 740 ± 185, n = 7; Fig. 6). The perforated buffering values were not significantly different from those determined from the comparable conventional whole cell data with 10 µM fura-2 K⁺₅, although the young values were significantly greater than those from the 50 and 100 µM fura-2 K⁺₅. However, the parallel shift in the buffering values for young and aged cells with the perforated patch suggests that the age-related increase in rapid Ca²⁺ buffering is mediated primarily by some mechanism other than diffusible buffer. The role of diffusible buffer in Ca²⁺ homeostasis seldom has been examined. The present results suggest a rather limited role for diffusible buffers in both rapid and slow Ca²⁺ buffering in agreement with Zhou and Neher (1993) and Stuenkel (1994).

View larger version (28K): [in this window] [in a new window]   FIG. 6. Ca2+ entry/ [Ca2+]i histograms for young and aged cells recorded with perforated patch and fura-2 AM. Protocol for measurement of buffering slopes was identical to that in Figs. 3 and 4. Left: buffering values for aged neurons are significantly greater (P < 0.03, 2-tailed t-test) than those of young neurons with the minimally invasive perforated patch method. Right: values obtained for controls and after thapsigargin (Thap) treatment (see RESULTS) in the same cells. Rapid buffering values did not change when the SER Ca2+ pump was blocked in either young or aged neurons. Values are given as means ± SE.

To assess the role of smooth endoplasmic reticulum (SER) Ca²⁺ sequestration in rapid Ca²⁺ buffering, thapsigargin was used to block the SER Ca²⁺ pump (Pozzan et al. 1994). The paradigm for determination of Ca²⁺ buffering values was performed on young and aged neurons before and after a 6-min bath application of 400 nM thapsigargin with subsequent 1-min washout. The efficacy of this application protocol was confirmed by observed blockade of caffeine induced Ca²⁺ release (not shown). Neurons were subjected to one to three depolarizing voltage steps after the washout of thapsigargin and before the test steps to deplete any previously sequestered Ca²⁺ that might be released. There was no difference in the rapid buffering values for either young or aged cells after the SER Ca²⁺ pumps were blocked, as shown in Fig. 6. Typical examples of perforated-patch recordings and Ca²⁺ entry/[Ca²⁺]_i plots from a young and an aged neuron are presented in Figs. 7 and 8. Note that although thapsigargin does little to alter the slope of the linear portion of the Ca²⁺ entry/[Ca²⁺]_i plot, the deviation from linearity at larger levels of Ca²⁺ entry is reduced. Qualitatively similar results were obtained from unclamped cells loaded with fura-2 AM and depolarized by 50 mM K⁺. In these latter experiments, the [Ca²⁺]_i after thapsigargin was (expressed as percent of control) +4.9 ± 8.1% in young (n = 9) and 6.0 ± 5.1% in aged neurons (n = 11).

To compare the slow buffering systems, we chose to measure the time course of the [Ca²⁺]_i recovery to baseline. This measure accounts for the activities of the several mechanisms involved in Ca²⁺ clearance. Although there is an increased rapid buffering strength in aged neurons, the recovery to baseline [Ca²⁺]_i after Ca²⁺ influx is delayed relative to the [Ca²⁺]_i. When recovery times are normalized to the [Ca²⁺]_i (s/nM), there are significant age-related differences. The mean recovery values for individual unclamped cells loaded with fura-2 AM and each depolarized by three or more 50 mM K⁺ pulses of different duration were 0.108 ± 0.010 s/nM (n = 27) for young neurons and 0.161 ± 0.015 s/nM (n = 20, P < 0.01, 1-way ANOVA) for aged. Because there were no systematic differences in the recovery values for the different concentrations of fura-2 K⁺₅, the conventional whole cell data were pooled. The mean recovery values for all fura concentrations and step durations from the experiments graphed in Fig. 4C were: 0.135 ± 0.005 s/nM (n = 39) for young cells and 0.159 ± 0.005 s/nM (n = 29, P < 0.01) for aged. When these data are pooled by step durations (Table 1), the recovery values increase with increasing step durations and correspondingly increased [Ca²⁺]_i. The mean recovery values are longer at all step durations in the aged neurons (P < 0.01, 2-way ANOVA). Recovery values obtained in perforated-patch recordings were not significantly different from those recorded with conventional whole cell patch. Because of the high affinity of fura-2 for Ca²⁺, the recovery to basal [Ca²⁺]_i is delayed slightly due to the dissociation of Ca²⁺ from the dye. Therefore the physiological recovery from elevated [Ca²⁺]_i is somewhat more rapid than that observed under these experimental conditions. This effect is present in both young and aged neurons and does not influence the comparative results.

Although the thapsigargin sensitive stores do not appear to be principal mediators of rapid Ca²⁺ buffering, the block of the SER Ca²⁺ pumps significantly delayed the recovery to baseline after a [Ca²⁺]_i in both young and aged perforated-patch voltage-clamped cells. Examples of the control and test fluorescent ratio records are shown in Fig. 9. Note that thapsigargin does not alter the peak or rise of the [Ca²⁺]_i but dramatically slows the return to baseline (). The effects of thapsigargin were parallel in young and aged cells. The recovery values were: young control, 0.098 ± 0.006 s/nM; young test, 0.168 ± 0.011 s/nM (P < 0.01 paired 2-tailed t-test, n = 23 steps in 6 cells) and aged control, 0.107 ± 0.009 s/nM; aged test, 0.169 ± 0.015 s/nM (P < 0.01 n = 25 steps in 5 cells). Interestingly, this effect of thapsigargin on recovery time was not observed in unclamped cells, suggesting that there is a voltage-dependent aspect to slow buffering. Although thapsigargin did not initially alter the basal [Ca²⁺]_i; after a series of voltage steps, the basal [Ca²⁺]_i appeared to reset at a greater concentration after a large [Ca²⁺]_i.

View larger version (27K): [in this window] [in a new window]   FIG. 9. Effects of thapsigargin on the slow buffering recovery to basal [Ca2+]i in young and aged neurons. Superimposed fluorescent ratio records and [Ca2+]i in response to 2 different duration voltage steps in control conditions and after thapsigargin treatment. A: data from a young (1 mo) neuron. B: data from an aged (24 mo) cell. In each case, the thapsigargin treatment delays the recovery () of the transient to basal [Ca2+]i. There was no age-related difference in the effect of thapsigargin.

	DISCUSSION

Abstract Introduction Methods Results Discussion References

These results support the idea of major age-related alterations in Ca²⁺ homeostasis in central mammalian neurons. Aged MS/nDB neurons have an increased rapid Ca²⁺ buffering ability via a mechanism that was not mediated by rapidly diffusible buffer or by SER Ca²⁺ sequestration. Aged neurons also display a prolonged Ca²⁺ transient that was not mediated by diffusible buffer or by SER Ca²⁺ pumps, although SER Ca²⁺ pumps were involved in slow Ca²⁺ buffering. Rapidly diffusible endogenous buffers had a limited involvement in both rapid and slow buffering.

Age-related changes in Ca²⁺ homeostasis discovered here contrast somewhat with other published results. Age-related increases in basal [Ca²⁺]_i with associated declines in [Ca²⁺]_i in neurons are reported from rat dorsal root ganglia, hippocampus, and neocortex (Kirischuk et al. 1992; Verkhratsky et al. 1994). These findings are attributed to a reduced Ca²⁺ influx through VGCCs and reduced Ca²⁺ buffering (Kostyuk et al. 1993; Verkhratsky et al. 1994). A similar explanation has been offered for age-related changes in Ca²⁺ homeostasis in rat brain synaptosomes involving reduced mitochondrial Ca²⁺ buffering (Martinez-Serrano et al. 1992; Vitorica and Satrustegui 1986). Aged peripheral adrenergic neurons have increased Ca²⁺ transient amplitude supposedly mediated by a reduction in SER buffering (Buchholz et al. 1996; Duckles et al. 1996; Tsai et al. 1996). Hartmann et al. (1994, 1996) have presented evidence of age-related decreases in basal [Ca²⁺]_i and high K⁺ induced [Ca²⁺]_i in some rat and mouse neurons and suggested a mechanism of reduced VGCC function and increased Na⁺/Ca²⁺ exchange. Diminished Ca²⁺ buffering with age has been proposed also from evidence of reduced levels of Ca²⁺ binding proteins in aged rat and rabbit hippocampal neurons (De Jong et al. 1996; Villa et al. 1994) and in aged rat MS and other brain regions (Iacopino and Christakos 1990; Krzywkowski et al. 1996). In aged rat MS/nDB neurons, there is an increased Ca²⁺ influx through VGCCs (Murchison and Griffith 1995, 1996). This investigation shows that reduction of Ca²⁺ transients with age is due to a nondiffusible, non-SER mediated increase in rapid Ca²⁺ buffering and that basal [Ca²⁺]_i is maintained during aging by a nondiffusible mechanism. These findings underscore the cautionary note voiced by Verkhratsky and Toescu (1998) that age-related increases in [Ca²⁺]_i and reduced buffering should not be assumed.

Block of the SER Ca²⁺ pump by thapsigargin demonstrated the involvement of SER in slow buffering restoration of basal [Ca²⁺]_i and in the establishment of the rapid buffering [Ca²⁺]_i plateau level. SER uptake was not a mechanism of rapid buffering per se, but limited the amount of Ca²⁺ needing to be buffered by the rapid mechanisms. This finding contrasts with the situation found in rat adrenal chromaffin cells (Park et al. 1996) where SER uptake is not important for Ca²⁺ clearance. That the SER uptake is not a mechanism of rapid buffering is generally accepted in basal forebrain neurons (Bleakman et al. 1993; Tatsumi and Katayama 1993; Zheng et al. 1996) and other cells (Blaustein 1988), and it is recognized as an important slow buffering mechanism in hippocampal neurons (Mironov 1995), dorsal root ganglion neurons (Shmigol et al. 1994), neocortical pyramidal neurons (Markram et al. 1995), and pituitary gonadotrophs (Tse et al. 1994). This is the first report of the involvement of the SER in slow Ca²⁺ buffering in forebrain neurons. Tatsumi and Katayama (1993) found no effect when cyclopiazonic acid was used to block the SER, and Bleakman et al. (1993) saw no effect with thapsigargin on unclamped cells tested with high K⁺ depolarizations. A similar lack of effect was observed by Stuenkel (1994). There was also no effect of thapsigargin on unclamped cells tested with high K⁺ in this study. Effects were evident only in voltage-clamped neurons. It is probable that non-SER slow buffering mechanisms can compensate for the loss of SER Ca²⁺ uptake when the cell is allowed a prolonged depolarization but are inhibited when the cell is clamped at V_h = 60 mV. The Ca²⁺-ATPase has a voltage dependence (Allen and Baker 1986) that could allow it to mediate such a compensatory response.

The nonlinear portions of Ca²⁺entry/[Ca²⁺]_i curves are probably due to a combination of fast current inactivation reducing the rate of Ca²⁺ influx and the operation of a low-affinity, high-capacity buffering system (Thayer and Miller 1990) activated when the Ca²⁺ entry exceeds 75-150 µM. The levels of Ca²⁺ entry at which the Ca²⁺entry/[Ca²⁺]_i plots deviate from linearity were not measured definitively but were not apparently different in young and aged cells. The influence of thapsigargin on the linearity of the Ca²⁺entry/[Ca²⁺]_i curves suggests that the SER store may sequester Ca²⁺ and remove it from the cytoplasm once a certain elevated level of [Ca²⁺]_i is reached. The mitochondria have been considered the principal mediators of the low-affinity, high-capacity buffering that is triggered in other cells at [Ca²⁺]_i of 500-600 nM (Hehl et al. 1996; Herrington et al. 1996; Stuenkel 1994; White and Reynolds 1995). In MS/nDB cells, it appears that a similar SER-mediated system is triggered at lower [Ca²⁺]_i and may operate to limit [Ca²⁺]_i to levels below those which activate mitochondrial uptake. Such a system takes on an added significance as evidence accumulates that mitochondrial involvement in buffering of greater Ca²⁺ loads is a key event in excitotoxic and ischemic neuronal injury (Budd and Nichols 1996; Kiedrowski and Costa 1996; Schinder et al. 1996; Wang and Thayer 1996; Wenk et al. 1996a; White and Reynolds 1996). The extended linear range of the Ca²⁺entry/[Ca²⁺]_i curves in thapsigargin also indicates that high affinity rapid buffering capacity of cells may not be near saturation at [Ca²⁺]_i levels that activate the low-affinity, high-capacity systems. Supralinearity of the Ca²⁺entry/[Ca²⁺]_i curves (indicative of Ca²⁺ induced Ca²⁺ release, CICR) (Verkhratsky and Shmigol 1996) was not observed in any MS/nDB neurons, suggesting that CICR is not a prominent feature in these cells (Murchison and Griffith, unpublished data).

The age-related increase in recovery time may be explained by gradual release of Ca²⁺ from rapid buffers. If rapid Ca²⁺ buffering is greater in aged neurons, then more Ca²⁺ must be bound or sequestered by the rapid buffers. This Ca²⁺ then must be released so it can be cleared from the cell by the slow buffering mechanisms. It follows that a greater amount of Ca²⁺ rapidly buffered and then gradually released would delay the return to basal [Ca²⁺]_i. Such a mechanism of delayed recovery to basal [Ca²⁺]_i due to release of buffered Ca²⁺ has been shown to occur for mitochondrial Ca²⁺ buffering (Babcock et al. 1997; Herrington et al. 1996). Increased mitochondrial Ca²⁺ buffering in the aged cells could well explain the delayed return to basal [Ca²⁺]_i after a large Ca²⁺ load. Alternatively, it may be that slow Ca²⁺ buffering is decreased with age, as has been implied for aged dorsal root ganglion cells (Kirischuk et al. 1992). The fact that thapsigargin delays the recovery to baseline [Ca²⁺]_i without influencing the rapid Ca²⁺ buffering values indicates that the effect of thapsigargin is on a slow buffering mechanism and implies that this mechanism represents uptake into the SER. However, [Ca²⁺]_i eventually returns to baseline with the SER Ca²⁺ pump blocked, so there must be additional mechanisms of slow buffering. The parallel shift in young and aged recovery values in thapsigargin suggests that an alteration in the SER is not the mechanism of the age-related difference. The activation of calcium release activated current (I_crac) (Fasolato et al. 1994) by the thapsigargin depletion of the SER store could have contributed to increased transient durations observed in thapsigargin, however, it seems more likely that I_crac would show a sustained activation after store depletion that would be manifest as an elevated [Ca²⁺]_i baseline (Thomas and Hanley 1994). In fact, an eventual increase in the resting [Ca²⁺]_i was observed after thapsigargin application in about half of the cells tested.

These findings support a model in which an increased rapid buffering capacity is generated in aged MS/nDB neurons as a compensatory response to increased Ca²⁺ influx, although the reverse relationship cannot be ruled out at this time. The mechanism of this increase is unknown but does not involve the SER or a readily diffusible buffer. This does not exclude an increase in Ca²⁺ binding proteins, as many of these proteins are probably membrane bound or sequestered (Pozzan et al. 1994) and are likely to diffuse very slowly; however, available evidence (see earlier text) suggests that Ca²⁺ binding proteins are lost with age. An increase in mitochondrial Ca²⁺ buffering with age could explain both the age-related increase in rapid Ca²⁺ buffering values and the delayed return to basal [Ca²⁺]_i. Although the mitochondria traditionally have been considered to participate only in the slow buffering of large Ca²⁺ loads, recent investigations have shown that mitochondrial buffering can begin rapidly enough to influence peak [Ca²⁺]_i (Babcock et al. 1997). A compensatory upregulation of slow buffering in response to increased Ca²⁺ influx has been shown in chick sensory neurons (Bolsover et al. 1996).

The augmented buffering capacity in aged MS/nDB neurons reduces the peak [Ca²⁺]_i while possibly delaying the return to baseline [Ca²⁺]_i. If this buffering change extends into the synaptic terminals, it could influence Ca²⁺ dependent processes such as transmitter release and synaptic plasticity. It recently has been proposed that reduced Ca²⁺ buffering could explain age-related changes in hippocampal synaptic plasticity (Foster and Norris 1997). Many of these changes also could be explained by increased Ca²⁺ buffering with age. However, buffering in the somatic cytoplasm may not reflect that in restricted dendritic and synaptic regions. Instead, our findings may relate more to Ca²⁺ regulation of gene expression (Ginty 1997) or cell death. Because elevated [Ca²⁺]_i may or may not be associated with various types of neurodegeneration (Rothman and Olney 1995), it is unknown whether these age-related changes in Ca²⁺ homeostasis tend to enhance neuronal survival or to contribute to age-related cell death. Given the important role of elevated [Ca²⁺]_i in mediating excitotoxicity (Choi 1992, 1995; Lu et al. 1996), CNS trauma (McIntosh et al. 1997), and effects of Alzheimer's disease (Mattson 1994, Mattson et al. 1997; Wolozin et al. 1996) and in preventing apoptosis (Franklin and Johnson 1992), any alterations in Ca²⁺ buffering could be important determinants of differential cell survival in the MS/nDB where there is disease- and age-related loss of neurons (Dutar et al.1995; Fischer et al. 1989; Miettinen et al. 1993; Smith and Booze 1995) and an increased response to excitatory amino acid (EAA) applications (Jasek and Griffith 1998). For instance, the recently reported age-related resistance of F344 rat basal forebrain neurons to EAA toxicity (Wenk et al. 1996b) could be related to increased Ca²⁺ buffering in these cells. If correlations could be made between changes in Ca²⁺ buffering and cell survival, then potential therapies to moderate age-related or disease-induced cognitive deficits caused by neuronal cell death may be formulated.

	ACKNOWLEDGEMENTS

  The authors thank Dr. Michael J. Davis for help in developing the microfluorimetric recording system.

  This work was supported by National Institute of Aging Grant AG-07805.

	FOOTNOTES

  Address reprint requests to W. H. Griffith.

  Received 29 September 1997; accepted in final form 30 March 1998.

	REFERENCES

Abstract Introduction Methods Results Discussion References

• ALLEN, T.G.J., SIM, J. A., BROWN, D. A. The whole cell calcium current in acutely dissociated magnocellular cholinergic basal forebrain neurones of the rat. J. Physiol. (Lond.) 460: 91-116, 1993.[Abstract/Free Full Text]
• ALLEN, T.J.A., BAKER, P. F. Influence of membrane potential on calcium efflux from giant axons of loligo. J. Physiol. (Lond.) 378: 77-96, 1986.[Abstract/Free Full Text]
• BABCOCK, D. F., HERRINGTON, J., GOODWIN, P. C., PARK, Y. B., HILLE, B. Mitochondrial participation in the intracellular Ca²⁺ network. J. Cell Biol. 136: 833-844, 1997.[Abstract/Free Full Text]
• BARNES, C. A. Normal aging: regionally specific changes in hippocampal synaptic transmission. Trends Neurosci. 17: 13-18, 1994.[Medline]
• BARTUS, R. T., DEAN, R. L., BEER, B., LIPPA, A. S. The cholinergic hypothesis of geriatric memory dysfunction. Science 217: 408-417, 1982.[Abstract/Free Full Text]
• BASKYS, A., REYNOLDS, J. N., CARLEN, P. L. NMDA depolarizations and long-term potentiation are reduced in the aged rat neocortex. Brain Res. 530: 142-146, 1990.[Medline]
• BERTOLINO, M., LLINAS, R. R. The central role of voltage-activated calcium channels in neuronal cells. Annu. Rev. Pharmacol. Toxicol. 32: 399-421, 1992.[Medline]
• BLAUSTEIN, M. P. Calcium transport and buffering in neurons. Trends Neurosci. 11: 438-443, 1988.[Medline]
• BLEAKMAN, D., ROBACK, J. D., WAINER, B., MILLER, R., HARRISON, N. L. Calcium homeostasis in rat septal neurons in tissue culture. Brain Res. 600: 257-267, 1993.[Medline]
• BLOKLAND, A. Acetylcholine: a neurotransmitter for learning and memory? Brain Res. Rev. 21: 285-300, 1996.
• BOLSOVER, S. R., KATER, S. B., GUTHRIE, P. B. Spatial gradients of cytosolic calcium concentration in neurones during paradoxical activation by calcium. Cell Calcium 20: 373-379, 1996.[Medline]
• BUCHHOLZ, J., TSAI, H., FOUCART, S., DUCKLES, S. P. Advancing age alters intracellular calcium buffering in rat adrenergic nerves. Neurobiol. Aging 17: 885-892, 1996.[Medline]
• BUDD, S. L., NICHOLLS, D. G. Mitochondria, calcium regulation, and acute glutamate excitotoxicity in cultured cerebellar granule cells. J. Neurochem. 67: 2282-2291, 1996.[Medline]
• CAMPBELL, L. W., HAO, S. Y., THIBAULT, O., BLALOCK, E. M., LANDFIELD, P. W. Aging changes in voltage-gated calcium currents in hippocampal CA1 neurons. J. Neurosci. 16: 6286-6295, 1996.[Abstract/Free Full Text]
• CHOI, D. W. Excitotoxic cell death. J. Neurobiol. 23: 1261-1276, 1992.[Medline]
• CHOI, D. W. Calcium: still center-stage in hypoxic-ischemic neuronal death. Trends Neurosci. 18: 58-60, 1995.[Medline]
• CLAPHAM, D. E. Calcium signaling. Cell 80: 259-268, 1995.[Medline]
• COYLE, J. T., PRICE, D. L., DELONG, M. R. Alzheimer's Disease: a disorder of cortical cholinergic innervation. Science 219: 1184-1190, 1983.[Abstract/Free Full Text]
• DECKER, M. W. The effects of aging on hippocampal and cortical projections of the forebrain cholinergic system. Brain Res. Rev. 12: 423-438, 1987.
• DECOSTER, M. A. Calcium dynamics in the central nervous system. Adv. Neuroimmunol. 5: 233-239, 1995.[Medline]
• DEJONG, G. I., NABER, P. A., VAN DER ZEE, E. A., THOMPSON, L. T., DISTERHOFT, J. F., LUITEN, P.G.M. Age-related loss of calcium binding proteins in rabbit hippocampus. Neurobiol. Aging 17: 459-465, 1996.[Medline]
• DUCKLES, S. P., TSAI, H., BUCHOLZ, J. N. Evidence for decline in intracellular calcium buffering in adrenergic nerves of aged rats. Life Sci. 58: 2029-2035, 1996.[Medline]
• DUTAR, P., BASSANT, M. H., SENUT, M. C., LAMOUR, Y. The septohippocampal pathway: structure and function of a central cholinergic system. Physiol. Rev. 75: 393-427, 1995.[Free Full Text]
• FASOLATO, C., INNOCENTI, B., POZZAN, T. Receptor-activated Ca²⁺ influx: how many mechanisms for how many channels? Trends Pharmacol. Sci. 15: 77-83, 1994.[Medline]
• FIBIGER, H. C. The organization and some projections of cholinergic neurons of the mammalian forebrain. Brain Res. Rev. 4: 327-388, 1982.
• FIERRO, L., LLANO, I. High endogenous calcium buffering in Purkinje cells from rat cerebellar slices. J. Physiol. (Lond.) 496: 617-625, 1996.[Abstract/Free Full Text]
• FISCHER, W., GAGE, F. H., BJORKLUND, A. Degenerative changes in forebrain cholinergic nuclei correlate with cognitive impairments in aged rats. Eur. J. Neurosci. 1: 34-45, 1985.
• FOSTER, T. C., NORRIS, C. M. Age-associated changes in Ca²⁺-dependent processes: relation to hippocampal synaptic plasticity. Hippocampus 7: 602-612, 1997.[Medline]
• FRANKLIN, J. L., JOHNSON, E. M. JR Suppression of programmed neuronal death by sustained elevation of cytoplasmic calcium. Trends Neurosci. 15: 501-508, 1992.[Medline]
• GIBSON, G. E., PETERSON, C. Calcium and the aging nervous system. Neurobiol. Aging 8: 329-343, 1987.[Medline]
• GINTY, D. D. Calcium regulation of gene expression: isn't that spatial? Neuron 18: 183-186, 1997.[Medline]
• GORELOVA, N., REINER, P. B. Role of the afterhyperpolarization in control of discharge properties is septal cholinergic neurons in virto. J. Neurophysiol. 75: 695-706, 1996.[Abstract/Free Full Text]
• GRIFFITH, W. H., MATTHEWS, R. T. Electrophysiology of ACHE-positive neurons in basal forebrain slices. Neurosci. Lett. 71: 169-174, 1986.[Medline]
• GRIFFITH, W. H., MURCHISON, D. Enhancement of GABA-activated membrane currents in aged Fischer 344 rat basal forebrain neurons. J. Neurosci. 15: 2407-2416, 1995.[Abstract]
• GRYNKIEWICZ, G., POENIE, M., TSIEN, R. Y. A new generation of Ca²⁺ indicators with greatly improved fluorescence properties. J. Biol. Chem. 260: 3440-3450, 1985.[Abstract/Free Full Text]
• HAMILL, O. P., MARTY, A., NEHER, E., SAKMANN, B., SIGWORTH, F. J. Improved patch-clamp techniques for high-resolution current recording from cells and cell-free membrane patches. Pflügers Arch. 391: 85-100, 1981.[Medline]
• HARTMANN, H., ECKERT, A., MÜLLER, W. E. Disturbances of the neuronal calcium homeostasis in the aging nervous system. Life Sci. 55: 2011-2018, 1994.[Medline]
• HARTMANN, H., VELBINGER, K., ECKERT, A., MÜLLER, W. E. Region-specific downregulation of free intracellular calcium in the aged rat brain. Neurobiol. Aging 17: 557-563, 1996.[Medline]
• HEHL, S., GOLARD, A., HILLE, B. Involvement of mitochondria in intracellular calcium sequestration by rat gonadotropes. Cell Calcium 20: 515-524, 1996.[Medline]
• HERRINGTON, J., PARK, Y. B., BABCOCK, D. F., HILLE, B. Dominant role of mitochondria in clearance of large Ca²⁺ loads from rat adrenal chromaffin cells. Neuron 16: 219-228, 1996.[Medline]
• HOLLIDAY, J., SPITZER, N. C. Calcium channels in the regulation of cell development and cellular interactions. In: Calcium Channels: Their Properties, Functions, Regulations and Clinical Relevance, edited by L. Hurwitz, L. D. Partridge, and J. K. Leach Boca Raton, FL: CRC, 1991, p. 137-155.
• HORN, R., MARTY, A. Muscarinic activation of ionic currents measured by a new whole-cell recording method. J. Gen. Physiol. 92: 145-159, 1988.[Abstract/Free Full Text]
• IACOPINO, A. M., CHRISTAKOS, S. Specific reduction of calcium-binding protein (28-kilodalton calbindin-D) gene expression in aging and neurodegenerative diseases. Proc. Natl. Acad. Sci. USA 87: 4078-4082, 1990.[Abstract/Free Full Text]
• JASEK, M. C., GRIFFITH, W. H. Pharmacological characterization of ionotropic excitatory amino acid receptors in young and aged rat basal forebrain. Neuroscience 82: 1179-1194, 1998.[Medline]
• KAO, J.P.Y. Practical aspects of measuring [Ca²⁺] with fluorescent indicators. Methods Cell Biol. 40: 155-180, 1994.[Medline]
• KHACHATURIAN, Z. Towards theories of brain aging. In: Handbook of Studies on Psychiatry and Old Age, edited by D. W. Kay, and G. D Burrows Amsterdam: Elsevier, 1984, p. 7-30.
• KHACHATURIAN, Z. S. Calcium hypothesis of Alzheimer's disease and brain aging. NY Acad. Sci. 747: 1-11, 1994.
• KIEDROWSKI, L., COSTA, E. Glutamate-induced destabilization of intracellular calcium concentration homeostasis in cultured cerebellar granule cells: role of mitochondria in calcium buffering. Mol. Pharmacol. 47: 140-147, 1995.[Abstract]
• KIRISCHUK, S., PRONCHUK, N., VERKHRATSKY, A. Measurements of intracellular calcium in sensory neurons of adult and old rats. Neuroscience 50: 947-951, 1992.[Medline]
• KOSTYUK, P. G. Diversity of calcium ion channels in cellular membranes. Neuroscience 28: 253-261, 1989.[Medline]
• KOSTYUK, P., PRONCHUK, N., SAVCHENKO, A., VERKHRATSKY, A. Calcium currents in aged rat dorsal root ganglion neurones. J. Physiol. (Lond.) 461: 467-483, 1993.[Abstract/Free Full Text]
• KRZYWKOWSKI, P., POTIER, B., BILLARD, J. M., DUTAR, P., LAMOUR, Y. Synaptic mechanisms and calcium binding proteins in the aged rat brain. Life Sci. 59: 421-428, 1996.[Medline]
• LAMOUR, Y., DUTAR, P., JOBERT, A. Septo-hippocampal neurons: altered properties in the aged rat. Brain Res. 416: 277-282, 1987.[Medline]
• LANDFIELD, P. W. Increased calcium-current hypothesis of brain aging. Neurobiol. Aging 8: 346-347, 1987.[Medline]
• LANDFIELD, P. W., MCCAUGH, J. L., LYNCH, G. Impaired synaptic potentiation in the hippocampus of aged, memory-deficient rats. Brain Res. 150: 85-101, 1978.[Medline]
• LESLIE, S. W., CHANDLER, L. J., BARR, E. M., FARRAR, R. P. Reduced calcium uptake by rat brain mitochondria and synaptosomes in response to aging. Brain Res. 329: 177-183, 1986.
• LU, Y. M., YIN, H. Z., CHIANG, J., WEISS, J. H. Ca²⁺-permeable AMPA/kainate and NMDA channels: high rate of Ca²⁺ influx underlies potent induction of injury. J. Neurosci. 16: 5457-5465, 1996.[Abstract/Free Full Text]
• MALENKA, R. C., KAUER, J. A., PERKEL, D. J., NICOLL, R. A. The impact of postsynaptic calcium on synaptic transmission. Trends Neurosci. 12: 444-450, 1989.[Medline]
• MARKOWSKA, A. L., OLTON, D. S., GIVENS, B. Cholinergic manipulations in the medial septal area; age-related effects on working memory and hippocampal electrophysiology. J. Neurosci. 15: 2063-2073, 1995.[Abstract]
• MARKRAM, H., HELM, P. J., SAKMANN, B. Dendritic calcium transients evoked by single back-propagating action potentials in rat neocortical pyramidal neurons. J. Physiol. (Lond.) 485: 1-20, 1995.[Abstract/Free Full Text]
• MARKRAM, H., SEGAL, M. Electrophysiological characteristics of cholinergic and non-cholinergic neurons in the rat medial septum-diagonal band complex. Brain Res. 513: 171-174, 1990.[Medline]
• MARTINEZ, A., VITORICA, J., SATRUSTEGUI, J. Cytosolic free calcium levels increase with age in rat brain synaptosomes. Neurosci. Lett. 88: 336-342, 1988.[Medline]
• MARTINEZ-SERRANO, A., BLANCO, P., SATRUSTEGUI, J. Calcium binding to the cytosol and calcium extrusion mechanism in intact synaptosomes and their alterations with aging. J. Biol. Chem. 267: 4672-4679, 1992.[Abstract/Free Full Text]
• MATTSON, M. P. Calcium and neuronal injury in Alzheimer's disease. NY Acad. Sci. 747: 50-76, 1994.
• MATTSON, M. P., FURUKAWA, K., BRUCE, A. J., MARK, R. J., BLANC, E. Calcium homeostasis and free radical metabolism as convergence points in the pathophysiology of dementia. In: Molecular Mechanisms of Dementia, edited by W. Wasco, and R. E. Tanzi Totowa, NJ: Humana Press, 1997, p. 103-143.
• MCINTOSH, T. K., SAATMAN, K. E., RAGHUPATHI, R. Calcium and the pathogenesis of traumatic CNS injury: cellular and molecular mechanisms. Neuroscientist 3: 169-175, 1997. [Abstract/Free Full Text]
• MICHAELIS, M. L. Ion transport systems and Ca²⁺ regulation in aging neurons. NY Acad. Sci. 747: 407-418, 1994.
• MIETTINEN, R., SIRVIO, J., RIEKKINEN, P. SR, LAAKSO, M. P., RIEKKINEN, M., RIEKKINEN, P. JR Neocortical, hippocampal and septal parvalbumin- and somatostatin-containing neurons in young and aged rats; correlation with passive avoidance and water maze performance. Neuroscience 53: 367-378, 1993.[Medline]
• MILLER, R. J. The control of neuronal Ca²⁺ homeostasis. Prog. Neurobiol. 37: 255-285, 1991.[Medline]
• MIRONOV, S. L. Plasmalemmal and intracellular Ca²⁺ pumps as main determinants of slow Ca²⁺ buffering in rat hippocampal neurones. Neuropharmacology 34: 1123-1132, 1995.[Medline]
• MOORE, E.D.W., BECKER, P. L., FOGARTY, K. E., WILLIAMS, D. A., FAY, F. S. Ca²⁺ imaging in single living cells: theoretical and practical issues. Cell Calcium 11: 157-179, 1990.[Medline]
• MURCHISON, D., GRIFFITH, W. H. Low-voltage activated calcium currents increase in basal forebrain neurons from aged rats. J. Neurophysiol. 74: 876-887, 1995.[Abstract/Free Full Text]
• MURCHISON, D., GRIFFITH, W. H. High-voltage activated calcium currents in basal forebrain neurons during aging. J. Neurophysiol. 76: 158-174, 1996.[Abstract/Free Full Text]
• NEHER, E. The use of fura-2 for estimating Ca buffers and Ca fluxes. Neuropharmacology 34: 1423-1442, 1995.[Medline]
• NEHER, E., AUGUSTINE, G. J. Calcium gradients and buffers in bovine chromaffin cells. J. Physiol. (Lond.) 450: 273-301, 1992.[Abstract/Free Full Text]
• NORRIS, C. M., KOROL, D. L., FOSTER, T. C. Increased susceptibility to induction of long-term depression and long-term potentiation reversal during aging. J. Neurosci. 16: 5382-5392, 1996.[Abstract/Free Full Text]
• OLTON, D. S., WENK, G. L., MARKOWSKA, A. M. Basal forebrain, memory and attention. In: Activation to Acquisition: Functional Aspects of the Basal Forebrain Cholinergic System, edited by and R. T. Richardson Boston: Birkhäuser, 1991, p. 247-262.
• PARK, Y. B., HERRINGTON, J., BABCOCK, D. F., HILLE, B. Ca²⁺ clearance mechanisms in isolated rat adrenal chromaffin cells. J. Physiol. (Lond.) 492: 329-346, 1996.[Abstract/Free Full Text]
• PITLER, T. A., LANDFIELD, P. W. Aging-related prolongation of calcium spike duration in rat hippocampal slice neurons. Brain Res. 508: 1-6, 1990.[Medline]
• POENIE, M. Alteration of intracellular fura-2 fluorescence by viscosity: a simple correction. Cell Calcium 11: 85-91, 1990.[Medline]
• POZZAN, T., RIZZUTO, R., VOLPE, P., MELDOLESI, J. Molecular and cellular physiology of intracellular calcium stores. Physiol. Rev. 74: 595-636, 1994.[Free Full Text]
• REYNOLDS, J. N., CARLEN, P. L. Diminished calcium currents in aged hippocampal dentate gyrus granule neurones. Brain Res. 479: 384-390, 1989.[Medline]
• ROTHMAN, S. M., OLNEY, J. W. Excitotoxicity and the NMDA receptorstill lethal after eight years. Trends Neurosci. 18: 57-58, 1995.[Medline]
• SCHINDER, A. F., OLSON, E. C., SPITZER, N. C., MONTAL, M. Mitochondrial dysfunction is a primary event in glutamate neurotoxicity. J. Neurosci. 16: 6125-6133, 1996.[Abstract/Free Full Text]
• SHMIGOL, A., KOSTYUK, P., VERKHRATSKY, A. Role of caffeine-sensitive Ca²⁺ stores in Ca²⁺ signal termination in adult mouse DRG neurones. Neuroreport 5: 2073-2076, 1994.[Medline]
• SMITH, M. L., BOOZE, R. M. Cholinergic and GABAergic neurons in the nucleus basalis region of young and aged rats. Neuroscience 67: 679-688, 1995.[Medline]
• SMITH, S. J., AUGUSTINE, G. J. Calcium ions, active zones and synaptic transmitter release. Trends Neurosci. 11: 458-464, 1988.[Medline]
• STUENKEL, E. L. Regulation of intracellular calcium and calcium buffering properties of rat isolated neurohypophysial nerve endings. J. Physiol. (Lond.) 481.2: 251-271, 1994.[Abstract/Free Full Text]
• STUREK, M., CALDWELL, W. M., HUMPHREY, D. A., WAGNER-MANN, C. Methods for simultaneous voltage-clamp, microfluorimetry, and video of cells. I. Electronic and optical instrumentation. In: Ion Channels Vascular Smooth Muscle Cells and Endothelial Cells, edited by N. Sperelakis, and H. Kuriyama New York: Elsevier, 1991, p 239-267.
• TATSUMI, H., KATAYAMA, Y. Regulation of the intracellular free calcium concentration in acutely dissociated neurones from rat nucleus basalis. J. Physiol. (Lond.) 464: 165-181, 1993.[Abstract/Free Full Text]
• TATSUMI, H., KATAYAMA, Y. Calcium homeostasis in the presence of fura-2 in neurons dissociated from rat nucleus basalis: theoretical and experimental analysis of chelating action of fura-2. J. Neurosci. Methods 53: 209-215, 1994.[Medline]
• THAYER, S. A., MILLER, R. J. Regulation of the intracellular free calcium concentration in single rat dorsal root ganglion neurones in vitro. J. Physiol. (Lond.) 425: 85-115, 1990.[Abstract/Free Full Text]
• THIBAULT, O., LANDFIELD, P. W. Increase in single L-type calcium channels in hippocampal neurons during aging. Science 272: 1017-1020, 1996.[Abstract]
• THOMAS, D., HANLEY, M. R. Pharmacological tools for perturbing intracellular calcium storage. Methods in Cell Biol. 40: 65-89, 1994.[Medline]
• TRUMP, B. F., BEREZESKY, I. K. The role of cytosolic Ca²⁺ in cell injury, necrosis and apoptosis. Curr. Opin. Cell Biol. 4: 227-232, 1992.[Medline]
• TSAI, H., POTTORF, B. J., FOUCART, S., BUCHHOLZ, J., DUCKLES, S. P. Evidence for decline in smooth endoplasmic reticulum calcium buffering with age in rat adrenergic nerves. Soc. Neurosci. Abstr. 22: 770, 1996.
• TSE, A., TSE, F. W., HILLE, B. Calcium homeostasis in identified rat gonadotrophs. J. Physiol. (Lond.) 477: 511-525, 1994.[Abstract/Free Full Text]
• VERKHRATSKY, A., SHMIGOL, A. Calcium-induced calcium release in neurones. Cell Calcium 19: 1-14, 1996.[Medline]
• VERKHRATSKY, A., SHMIGOL, A., KIRISCHUK, S., PRONCHUK, N., KOSTYUK, P. Age-dependent changes in calcium currents and calcium homeostasis in mammalian neurons. Ann. NY Acad. Sci. 747: 365-381, 1994.[Medline]
• VERKHRATSKY, A., TOESCU, E. C. Calcium and neuronal aging. Trends Neurosci. 21: 2-7, 1998.[Medline]
• VILLA, A., PODINI, P., PANZERI, M. C., RACCHETTI, G., MELDOLESI, J. Cytosolic Ca²⁺ binding proteins during rat brain ageing: loss of calbindin and calretinin in the hippocampus, with no change in the cerebellum. Eur. J. Neurosci. 6: 1491-1499, 1994.[Medline]
• VITORICA, J., SATRUSTEGUI, J. Involvement of mitochondria in the age-dependent decrease in calcium uptake of rat brain synaptosomes. Brain Res. 378: 36-48, 1986.[Medline]
• WANG, G. J., THAYER, S. A. Sequestration of glutamate-induced Ca²⁺ Loads by mitochondria in cultured rat hippocampal neurons. J. Neurophysiol. 76: 1611-1621, 1996.[Abstract/Free Full Text]
• WENK, G. L., DANYSZ, W., ROICE, D. D. The effects of mitochondrial failure upon cholinergic toxicity in the nucleus basalis. Neuroreport 7: 1453-1456, 1996a.[Medline]
• WENK, G. L., STOEHR, J. D., MOBLEY, S. L., GURNEY, J., MORRIS, R. J. Age-related decrease in vulnerability to excitatory amino acids in the nucleus basalis. Neurobiol. Aging 17: 1-7, 1996b.[Medline]
• WHITE, R. J., REYNOLDS, I. J. Mitochondria and Na⁺/Ca²⁺ exchange buffer glutamate-induced calcium loads in cultured cortical neurons. J. Neurosci. 15: 1318-1328, 1995.[Abstract]
• WHITE, R. J., REYNOLDS, I. J. Mitochondrial depolarization in glutamate-stimulated neurons: an early signal specific to excitotoxin exposure. J. Neurosci. 16: 5688-5697, 1996.[Abstract/Free Full Text]
• WOLOZIN, B., LUO, Y., WOOD, K. Neuronal loss in aging and disease. In: Cellular Aging and Cell Death, edited by N. J. Holbrook, G. R. Martin, and R. A. Lockshin New York: Wiley-Liss, 1996, p. 283-302.
• ZHENG, F., GALLAGHER, J. P., CONNOR, J. A. Activation of a metabotropic excitatory amino acid receptor potentiates spike-driven calcium increases in neurons of the dorsolateral septum. J. Neurosci. 16: 6079-6088, 1996.[Abstract/Free Full Text]
• ZHOU, Z., NEHER, E. Mobile and immobile calcium buffers in bovine adrenal chromaffin cells. J. Physiol. (Lond.) 469: 245-273, 1993.[Abstract/Free Full Text]
• ZOLA-MORGAN, S., SQUIRE, L. R. Neuroanatomy of memory. Annu. Rev. Neurosci. 16: 547-63, 1993.[Medline]

0022-3077/98 $5.00 Copyright ©1998 The American Physiological Society