JOURNALOF NEUROPHYSIOLOGY Vol. 67, No. 3, March 1992. Printed

in U.S.A.

Calcium Homeostasis in Dissociated Embryonic Neurons: A Flow Cytometric Analysis JEREMY

P. GRIERSON,

ROBERT

E. PETROSKI,

SEAN M. O’CONNELL,

AND

HERBERT

M. GELLER

Department of Pharmacology, UMDNJ-Robert Wood Johnson Medical School, The Graduate School, Rutgers University and Centerfor Advanced Biotechnology and Medicine, Piscataway, New Jersey 08854 SUMMARY

AND

CONCLUSIONS

1. Ca2+homeostasis in freshlydissociatedneuronsfrom embryonic rat hypothalamus,cortex, and brain stem wasinvestigated with flow cytometry. Cellsweredissociatedfrom embryonic brain by enzymatic and mechanicalmeansand wereincubatedwith the acetoxymethylesterderivative of the Ca2+-sensitivedye indo- 1. Neurons hydrolyzed and retained the dye as determined by the intensity of fluorescenceemission,whereassimilarly treated cultured astrocytesgave very low-level fluorescence. 2. The fluorescenceof the indo- 1 dye was measuredat two wavelengths(405 and 485 nm) for eachcell. Data werecollected only from thosecells(presumptive neurons) with high levels of fluorescence.Methods were developed to calibrate the level of intracellular free calcium ([ Ca2+li) asthe ratio of fluorescenceat 4 10 and 485 nm. The level of intracellular free Ca2+was then calculatedfor eachneuron. 3. A wide distribution of resting [Ca2+liwasfound, with a median of -90 nM. After addition of ionomycin to cellsin Ca2+-free medium,there wasa transientincreasein [Ca2+li, suggestingthat all embryonic neuronshad internal Ca2+stores.The presenceof active calcium extrusion mechanismswasdemonstratedwith the useof ionomycin in Ca2+-containing mediumand with metabolic inhibitors. Furthermore, incubation in sodium-freemedium resulted in a transient increasein [Ca2+li and a reducedability to eliminate elevated [Ca2+]i from the cytoplasm, suggestingthat calcium homeostasiswasdependenton the activity of the Na+Ca2+exchangemechanism. 4. Depolarization with K+ or veratrine increased[Ca2+li in ~20% of the cells.This increasewasblockedby eliminating extracellular free Ca2+or adding Co2+, nifedipine, or verapamil, suggestingmediationby voltage-sensitivecalcium channels. 5. Neuronsweresortedon the basisof high [Ca2+liand placed into dissociatedculture. After 24 h, neuronsin culture retained indo- 1fluorescence,suggesting that populationsof neuronscan be collectedon the basisof their levelsof [Ca2+]+ 6. Theseresultsdemonstratethat flow cytometric analysisallowsthe characterization of a variety of Ca2+-regulatorymechanismsin populationsof freshly dissociatedembryonic neurons. Although only a proportion of embryonic day 17neuronsexhibit voltage-sensitivecalcium channels,all neurons have developed the ability to sequesterand extrude Ca2+. INTRODUCTION

Intracellular free Ca2+ ( [ Ca” Ii) is an important regulator of many neuronal functions, including rhythmic firing (Jahnsen and Llinas 1984)) long term potentiation (Lynch et al. 1983)) and neurotransmitter release [ Cazalis et al. 1987; Von Spreckelsen et al. 1990 (see review: Kennedy 1989 ) 1. More recently, the role of Ca2+ in shaping neuronal development has become apparent: alterations in [ Ca2+li 704

appear to influence neuronal survival (Oppenheim 1988), growth cone mobility (Cohan et al. 1987 ) , and neurite elongation (Mattson and Kater 1987 ) . Thus the development of neuronal calcium homeostasis and the expression of voltage-sensitive calcium channels ( VSCCs) is of considerable interest both from the standpoint of neural physiology and that of development. Furthermore, because it is possible that the temporal expression of ion channels might be stereotypical (Gottmann et al. 1988 ) , a knowledge of channel expression during development could provide a method of identifying subpopulations of neurons. Ca2+ homeostasis is the result of a balance of several physiological processes: Ca2+ entry through VSCCs, storage and release from intracellular sites, binding to specialized proteins, and extrusion via transmembrane pumps (for reviews see Blaustein 1988; Carafoli 1987). Although virtually nothing is known about the development of intracellular sequestering mechanisms, it seems likely that the ability to regulate [ Ca2+li exists from the earliest stage of development (Blaustein 1988), because in most instances it appears that failure to regulate [ Ca2+]i results in cell death (Choi 1988). Indeed, increased [ Ca2+li may play a role in naturally occurring cell death (Oppenheim 1988 ) , as well as neuronal loss after cerebral ischemia, hypoxia, and excitotoxicity (Pauwels et al. 1989). Numerous insights, on the other hand, have been made into the development of the VSCCs and how these in turn influence development. At least three VSCC subtypes (Nowycky et al. 1985; Yaari et al. 1987) have been characterized on the basis of kinetic and pharmacological properties, and several genes sharing sequence homology with dihydropyridine-sensitive Ca 2+ channels have been identified by molecular cloning experiments (Snutch et al. 1990), suggesting that yet more subtypes exist in the rat brain. In cell culture, the temporal expression of VSCC subtypes appears to be developmentally controlled (Gottmann et al. 1988; Sontheimer et al. 1989; Yaari et al. 1987) and perhaps also differentially localized within the cell (Nowycky et al. 1985; Pemey et al. 1986; Yaari et al. 1987). In certain cells (e.g., the neuroblastoma hybrid cell line, NG108- 15), the process of differentiation appears to be correlated with an increased expression of VSCCs (Noronha-Blob et al. 1988). Finally, VSCCs appear to be particularly abundant in the growth cones of active neurites, and clusters of channels have recently been found to be associated with the most active fringe of the growth cone (Silver et al. 1990). Thus VSCCs may confer new potentiality on developing neurons ( Anglister et al. 1982).

0022-3077/92 $2.00 Copyright 0 1992 The American Physiological Society Downloaded from www.physiology.org/journal/jn at Washington Univ (128.252.067.066) on February 13, 2019.

CALCIUM

HOMEOSTASIS

Although the usual methods of choice for investigating channel development are electrophysiological, the development of calcium-sensitive probes and the techniques for utilizing them provide powerful alternatives. The detection of intracellular Ca2+ levels with the use of the dye fura- has demonstrated differences in the spatial distribution of calcium during stimulation of cultured cells (Cohan et al. 1987; Connor 1986; Lipscombe et al. 1988). Whereas these studies have focused on the calcium responses of single cells, flow cytometry has the ability to quantify calcium levels in large numbers of cells as well as to obtain selected populations of cells for subsequent testing by immunologic or molecular means. This approach also avoids a major pitfall of tissue culture; i.e., placing young neurons into culture may alter their developmental program (Gottmann et al. 1988). Flow cytometry in conjunction with the dye indo- 1 has been used extensively for the analysis of calcium regulation in lymphocytes (Alcover et al. 1986; Rabinovitch et al. 1986) and, more recently, in preparations of rat spinal cord neurons (Schieren and MacDermott 1988) and striatal synaptosomes (Wolf and Kapatos 1989). All of these studies have demonstrated that this method is particularly effective at highlighting heterogeneous responses in what might be assumed to be homogeneous populations. In this study we have examined VSCC expression and [ Ca2+li regulation in freshly dissociated cells obtained from the embryonic rat hypothalamus with the use of flow cytometry and indo-1. We show that these cells, virtually all of which are neurons, have significant stores of sequestered calcium, and that -20% of them increase their [ Ca2+li in a concentration-dependent manner after K+-induced depolarization. The K+-induced response is inhibited by calcium channel blockers and divalent cations, suggesting mediation via VSCCs. We also show that most embryonic neurons have well-developed Ca2+ extrusion mechanisms and that neurons with high [ Ca2+li after depolarization will survive if sorted and placed into culture. A preliminary report of this study has been presented (Grierson et al. 1988).

IN EMBRYONIC

NEURONS

705

dure was >95%. To test the contribution of the Na+-Ca2+exchanger,cells were spun down and resuspendedin a nearly sodium-free (“Na+-free”) buffer that contained (in mM): 120.1 choline chloride, 5.3 KCl, 1.5 CaCl,, 0.4 MgS04, 11.1glucose,1 sodiumpyruvate, 0.1% BSA, and 25 HEPES,pH 7.4). Resultsin Na+-free buffer were comparedwith resultsin Iscove’smodified Eagle’smedium (Sigma).

Flow cytometry and data analysis The cellswereanalyzedin a Coulter EPICS-753flow cytometer with laserexcitation at 351.1-363.8 nm at 80 mW. The fluorescent light from the two emissionpeaksof indo- 1,4 10 nm (I,,,), and 485 nm (I,,,), was collected by separatephotomultiplier tubes after transmittance through a 380-nm long-passfilter, a beam-splitting450-nmlong-passdichroic mirror, and eithera 400 * 10or 500t 20 nm bandpassfilter for I,,, and I,,, , respectively. Signalsrepresentingforward-anglelight scatter,fluorescentintensity, and time were collectedas frequency histogramsand aslist mode files. The cell suspension washeld in a heatedwaterjacket during analysisby the flow cytometer and sampledat a rate of approximately 600 cells/s. The transit time for the cellsto reach the laserwasm-30-35s.After a period of data collection from the unstimulatedsample,the flow wasinterrupted andtest substances wereaddedto the cell suspension. Test solutionswereaddedto the samplevial at 2X (KC1 in a balancedsalt solution) or 100X concentrations.The ratio remainedsteadyover the time courseof the experimentfor the population of unstimulatedcells,andtherewas no detectableautofluorescenceof thesecells at the instrument settingsusedin thesestudies.The addition of equal volumesof ethanol or water (usedassolvents)did not alter the fluorescence emission. For indo-1, [Ca2+]i is proportional to the ratio of I,,, to Id85 (Grynkiewicz et al. 1985) and estimatedfrom Eq. 1. The ratio under minimum [Ca2+]i(R,in) and the ratio under maximum [Ca2+]i(R,,) wereobtainedfrom indo- 1loadedcellsresuspended in 120mM KCl, 20 mM K-HEPES (pH 7.2), 1 mM MgSO,, 20 mM NaN,, 10 mM 2-deoxyglucose,1 PM [C

&2 a] = K deCR -&id (Rm,,-RE2

valinomycin, 2 PM carbonyl cyanine m-chlorophenylhydrazone, and 5 PM ionomycin: a solution previously described(Chusedet al. 1987) to clamp the ionic and electricalgradientsto zero. BeCell preparation and indo-l loading causeit is unlikely that zero calcium can be achievedunder these conditions(June and Rabinovitch 1988), Rminwasdefinedasthe Single-cellsuspensions were preparedfrom embryonic day 17 lowest ratio with the lowest frequency (normally -0.22). This Sprague-Dawleyrat brain (Ventimiglia and Geller 1987). Hypo- generatedan estimateof median [Ca2+]iof -30 nM under “calthalamus,cerebralcortex, and brain stem/hindbrain (referred to cium-free” conditions.R,, wastaken asthe population median just ashindbrain) tissuefragmentswere incubated for 20 min at under high-calciumconditionsand wasnormally ~2.8. The ratio 37“C in N-2-hydroxyethylpiperazine-N’-2-ethanesulfonic acid Sfl/S,,2 was calculated [as describedin (June and Rabinovitch (HEPES)-buffered Eagle’sbasal medium (pH 7.4) containing 1988)] with the useof a Perkin-Elmer MPF-66 scanningfluori0.025%trypsin and 0.1% collagenase.Once dispersedby tritura- meter and the filters from the cytometer to obtain integratedvaltion, the tissuefragmentswere passedthrough nylon mesh(40 uesof light transmittance over the excitation rangeof 35l-363 pm) and centrifuged(5 min at 300g). The cellswerethen resus- nm. This value was2.95. The Kd for indo-l wastaken as250 nM pended in Dulbecco’smodified Eagle’smedium (DMEM) with (Grynkiewicz et al. 1985) . Thesevalueswerethen usedto gener10%fetal bovine serum(FBS) at a density of 1O6cells/ml. Neu- ate a table of estimated[Ca2+lifor every ratio. The ratio for each rons were loaded with indo- 1 by adding 1 PM indo-l acetoxy- cell was calculatedas a function of time with the useof the list methylester(Molecular Probes,Eugene,OR) to the cell suspen- mode analysisprogram in Gateway (Coulter). Figure 1 showsa sionfor 30 min at 37°C. Cellswerecentrifugedfor 5 min at 300g, representativeexperiment in which the fluorescentsignalswere and the cellpelletwasresuspended in a samplebuffer consistingof collected under such conditions with the resultant ratio. In low DMEM (without phenol red), 10 mM HEPES,and 0.1% bovine calcium, most cellsexhibit low intensity of emissionat IdI0and serumalbumin (BSA, pH 7.4) at a final cell densityof 2 x 106/ml. high intensity of emissionat Ids5(Fig. 1, A and B), resultingin a Cellswerealloweda 30-min recovery period at 37°C (to hydrolyze low ratio (Fig. 1C). The gainsettingson the photomultiplier tubes the esterbond) beforebeingplacedon ice. Incubation with trypan weresetto measurecellular fluorescenceat both 14i0and Ids5under blue wasusedto demonstratethat cell viabilitv after this proce- both low and high 1Ca2+1;conditions. This resultedin the excluMETHODS

Downloaded from www.physiology.org/journal/jn at Washington Univ (128.252.067.066) on February 13, 2019.

GRIERSON

706

ET AL.

41 Onm

FIG.

0

1

2

3

RATIO

0

1

2

1. Indo-l

fluorescence histograms at high and low

1

RATIO

sion of somecells with very high Idg5,which would lead to an underestimateof the number of cellswith low [Ca2+]i. Increasing the calcium concentration elicited both an increasein I,,, intensity and a drop in I,,, (Fig. 1, D and E), resultingin a substantial increasein ratio (Fig. 1F). The valuesobtainedfor Rminand R,,, werethen appliedto Eq. I for estimationof [Ca2+]i. This method yielded ordinal data, so nonparametric methodswere used for statisticalanalysis.The analysiswas performed by PC SAS and wasrestrictedto cellsin the rangeof O-2,000 nM (i.e., 98%of the total population). The raw data and statistical resultswere exported to Sigmaplot4.0 (Jandel) for display with the useof boxand-whiskerplots (Tukey 1977) to displaythe summarystatistics. The two-parameterhistogramwasgeneratedby a Silicon Graphics Iris 4D/70GT workstation running Princeton CITGL Tools.

smaller particles (left) consisted mostly of damaged, nonviable cells; they did not show significant levels of fluorescence, nor did they survive when sorted and placed into culture. The Gaussian distribution (right) comprised intact cells (mean diameter 7 pm under the light microscope) and represented 80-90s of the total population. On the basis of these measurements, only cells falling within the second

Cell sorting Populationsof cellswere sorted into tubes containing 24-h astrocyte-conditioned medium (DMEM with 15% FBS) supplementedwith 5 pg/ ml catalaseand 1 pg/ml superoxidedismutase and then plated onto preexistingmonolayersof astrocytesin 24well culture plates(Ventimiglia and Geller 1987) . Immunocytochemistry for microtubule-associated protein 2 (MAP2) with the useof the ABC technique (Hsu et al. 1981) was subsequently performedon the cultures,which wereexaminedwith the useof a ZeissAxioplan microscopeequippedwith an epifluorescenceattachment. Fixation with acid-ethanol,necessaryfor intermediate filament immunocytochemistry, did not preserveindo- 1 fluorescence, and, therefore, labeling the indo- 1-containing cells with MAP2 and a fluorescentconjugatewasnot possible.

0

RESULTS 0

[Ca 2+]i in dissociated embryonic cells The data from >30 separate embryonic dissociations and subsequent flow cytometric analysis show this technique is reproducible and reveal several consistent features of the cell populations. Freshly dissociated embryonic cells comprised two populations by forward-angle light scatter, a measure largely dependent on particle size (Fig. 2A, DiPorzio et al. 1987; Fiszman et al. 1990). The population of

50

100

FORWARD

ANGLE

150

200

LIGHT

SCATTER

250

I

I

I

1

500

1000

1500

2000

[Ca*+l (nM) 2. Resting [ Ca2+li of embryonic cells. A: histogram of forwardangle light scatter in arbitrary units vs. frequency for dissociated embryonic day 17 hypothalamic cells. Nonviable cells formed the smaller peak to the lefl, whereas intact cells formed the major peak to the righr . The bar represents the “gating window,” i.e., the range of acceptable values of forward-angle light scatter, used for data acquisition. B: representative histogram showing distribution of resting calcium ( [ Ca2+li) in dissociated embryonic day 17 hypothalamic neurons falling within the gating window. FIG.

Downloaded from www.physiology.org/journal/jn at Washington Univ (128.252.067.066) on February 13, 2019.

CALCIUM TABLE

1.

HOMEOSTASIS

IN EMBRYONIC

NEURONS

707

Summary statisticsof [Ca2’] leveZsin three dzg”erentregionsof embryonicrat brain Percentile 5

Region Hypothalamus (n = 7) Cortex@ = 3) Hindbrain (n = 3)

39.0 (22.3-66.8) 42.5 (42.0-58.5) 27.8 (27.8-46.3)

50

25 59.7 (42.0-107.7) 80.1 (71.3-84.7) 57.8 (46.0-75.7)

75

93.8 (67.1-174.5) 127.5

95

204.8 (163.4-338.7) 266.3 (25 1.8-294.1) 170.6 (170.6-227.4)

(121.5-131.7) 90.3 (75.7-112.5)

710.8 (695.4-1030.7) 950.0 (842.8-1192.4) 822.6 (689.8-885.3)

Cells were dissociated from the hypothalamus, cerebral cortex, and hindbrain/brainstem of embryonic day 17 rat brains and loaded with indo- 1 as described in experimental procedures. Samples of 10,000 cells were analyzed for [Ca2+]i by flow cytometry. Summary statistics of the [Ca2+]i distribution of the populations were obtained using PC-SAS. Values are the [Ca2+]i (in nM) for a typical experiment and the range for all experiments; range in parentheses; n = number of experiments.

peak were used for subsequent fluorescence analysis and [ Ca 2+]i estimation. We then gathered information about the range of calcium levels present in freshly dissociated embryonic cells. Freshly dissociated hypothalamic cells displayed a fairly wide range of resting [ Ca2+li, and the distribution was unimodal and skewed to the right (Fig. 2B). The median value was typically -90 nM (Table 1 ), and 75% of cells had ~200 nM [ Ca2+li. Dissociated cells from the cerebral cortex and hindbrain were qualitatively similar. The hindbrain had a median of -90 nM, and cortex was somewhat higher at - 127 nM. There were no statistically significant differences between the three different regions. If cells were kept at room temperature for the 4- to 5-h course of the experiment, the estimates of resting [ Ca2+li tended to rise slightly, probably because of cell death; this was reduced by holding the cells at 4OC. It was also noted that [ Ca2+li was higher in experiments that followed ionomycin addition. Thus extensive rinsing was necessary to remove ionomycin adsorbed into the plastic tubing. The effects of variations such as these on the interpretation of the results were minimized by including a prestimulus control in each experiment.

glycolysis and uncoupling of oxidative phosphorylation (2.81, Fig. 1 F). To test the ability of the neurons to respond to a moderate Ca2+ load, a low concentration of ionomycin (0.1 PM)

3 u

40

E u-

20 0’

I

L

TIME (mins) b

i

RATIO

B

Evidencefor Ca2+ extrusion

(f> J 1000

The elimination of [ Ca2+li from neurons is achieved primarily by two membrane proteins, a high-affinity, low( ATPase) and a capacity Ca 2+ adenosinetriphosphatase low-affinity, high-capacity Na + -Ca2+ exchanger ( Carafoli 1987). The existence of both these systems was demonstrated by experiments that used the calcium ionophore ionomycin. Under standard calcium conditions ( 1.8 mM) , all freshly dissociated indo- 1-loaded cells responded to the addition of a high concentration of ionomycin (5 PM) with a rightward shift in [ Ca2+li distribution, reaching a peak in 3-5 min (Fig. 3A). Under these conditions, the ratio approached, but remained below, R,,,. It is presumed that the cellular Ca2+ extrusion and storage mechanisms were unable to eliminate Ca2+ at this rate of influx. Even so, there was a population of cells that maintained [Ca2+li ~500 nM under these conditions (Fig. 3B). The rate of [ Ca2+li extrusion by energy-dependent means can then be appreciated by the comparing the highest ratio achieved in 5 PM ionomycin-treated cells (2.60, Fig. 3A) with the ratio, Rmax,for the same cells after ATP depletion by inhibition of

iii c) b 5

2

-

500 -

m z z 0

500

1000

1500

2000

[Ca2’ ] (nM) 3. Ionomycin-induced increases in [ Ca2+li. A : two-parameter histogram of response to 5 PM ionomycin as a function of time. Time intervals are 12 s. Ionomycin (5 PM) was added to the sample after 2 min of prestimulus data collection (+ ) . This induced a swift increase in fluorescence ratio. B: histograms showing distribution of [Ca2’li in embryonic day 17 hypothalamic neurons at rest and after ionomycin stimulation. Solid line indicates distribution of [ Ca2+]i in unstimulated cells. A histogram taken 4 min after addition of 0.1 PM ionomycin (- - -) showing a shift in the distribution of [ Ca2+li to the rig&. 2 min after addition of 5.0 PM ionomycin (. ), the majority of cells had [ Ca2+]i >2 PM [ Ca2+]i (off scale on this axis), whereas a small population of cells with relatively low [ Ca2+li remained. FIG.

l

l

Downloaded from www.physiology.org/journal/jn at Washington Univ (128.252.067.066) on February 13, 2019.

GRIERSON

708

was added to the cell suspension. This stimulus produced a modest increase in [ Ca2+li, as evidenced by a small shift to the right in the distribution (Fig. 3 B). It is likely that the elevated intracellular [ Ca2+li caused by this concentration of ionomycin was small enough for the membrane pumps to expel it fairly effectively. Thus it is clear that virtually all of the neurons have highly active Ca” extrusion mechanisms and that energy-dependent mechanisms play a significant role in homeostasis. To examine the contribution of the Na+-Ca2+ exchanger to calcium homeostasis in embryonic hypothalamus, dissociated cells were transferred to Na+-free buffer. Under these conditions, resting [ Ca2+li rose transiently, probably via the reversal of the direction of Na+-Ca2+ exchange. This was observed as an increase of 150% in the median level of [Ca2+]i to -230 nM and a broadening of the range, with the 95th percentile going from - 750 nM to 1,600 nM (Fig. 4). [ Ca2+]i gradually returned to baseline over a period of 30 min. Addition of a modest concentration of ionomycin ( 1 PM) to cells in standard conditions elicited a modest increase in [Ca’+]i (median from 50 nM to 230 nM), whereas addition of the same concentration of ionomycin to cells that had been transferred to Na+ -free buffer for 5-6 min elicited a much larger increase in [ Ca2+li (median from 230 to 1,000 nM) (Fig. 4). Thus, in Na+-free medium, the cells exhibited a reduced ability to eliminate the elevated [ Ca2+]i from the cytoplasm. Evidence for sequestered Ca2+ The treatment of cells with ethylene glycol-bis( ,&aminoethyl ether)-N,N,iV’,N’-tetraacetic acid (EGTA) reverses the concentration gradient for free Ca2+ such that intracellular free Ca2+ is rapidly lost to the extracellular milieu (Albert and Tashjian 1984). In cells that were partially calcium depleted ( 5 mM EGTA for 8 min), the addition of ionomycin induced a transient increase in calcium. This is most likely due to the rapid release of internal pools of stored calcium (Albert and Tashjian 1986 ). An example of this is shown in Fig. 5. In this experiment, pretreatment with EGTA decreased median [ Ca’+]i from 135 to 7 1 nM;

ET AL.

2000

1500

ii

1000

x $

500

E

N

,+ I

/

I

I

I

I

0

1

2

3

4

5

TIME

I

0

2

(minutes)

5. Role of extracellular calcium vs. intracelhtlar calcium stores in response to ionomycin. EGTA (5 mM) was applied to the sample 8 minutes before start of the experiment and its continued presence is indicated by the open bar. Closed bar indicates addition of 5 PM ionomycin. The ionophore transiently increased median [Ca2’], (0) from 71 to 110 nM, followed by a decrease to 53 nM. Solid line represents 95th percentile (95 PCTL) of the [ Ca2+], distribution (note different scale). This increased from 600 to 1,200 nM after ionomycin addition and remained at this level for the duration of the experiment. LOG.

after addition of ionomycin, a transient increase in [ Ca2+li to a median value of - 100 nM was observed. The magnitude of the increase immediately after addition of ionomytin is not measurable in these experiments, because Ca2+ extrusion is occurring during the transit time from the sample vial to the laser, - 30 s. The increase in median [ Ca2+li was transient, with the median [ Ca’+]i decreasing to 53 nM within several minutes. However, this treatment also resulted in a sustained increase in the percentage of cells with high (>500 nM) [Ca2+li (Fig. 5); the percentage of cells with [ Ca’+]i >500 nM actually rose from 7.6% before ionomycin to 18.4% afterward. Thus it appears that the cells retain significant levels of sequestered calcium after dissociation and that a small fraction contain either large amounts or are able to release it more gradually.

L

2000

1500 g A

-

1000

G 500

0

T-

I

I

Pre Control

Post

-

Pre Na+-

Po$ 8 free

4. Elimination of extracellular Na+ increases [Ca2’],. Box-andwhisker plots of [Ca”], under standard conditions and after 5-6 min. in Na’-free medium (see METHODS). Control situation (Pre), in each case, is followed by stimulation with 1 PM ionomycin (Post). Boxes enclose 25th and 75th percentiles of [Ca2+], distribution and are bisected by median; whiskers indicate 5th and 95th percentiles. HG.

FTG. 6. KC1 induced a rise in [Ca”], in a subpopulation of cells. This two-parameter histogram represents cell number vs. [ CaZC],as a function oftime. Arrow illustrates distribution ofintracellular [ Cazf 1,in cell population before addition of 50 mM KC1 to the sample vial at the beginning of the histogram. KC1 addition resulted in an increase in intracellular [Ca*‘], to the 800 nM range in a subpopulation of cells. Duration of the histogram was 8 min, and maximum [ Ca2+], was 2,000 nM.

Downloaded from www.physiology.org/journal/jn at Washington Univ (128.252.067.066) on February 13, 2019.

CALCIUM

HOMEOSTASIS

B 0i

IN EMBRYONIC

NEURONS

709

50 25

(ii)

)3OmM

k

j

1

g 0

/

k LL

FIG. 7. KC1 induced a concentration-dependent increase in [ Ca2+]i. A: box-and-whisker plots of [ Ca2+li distribution for hypothalamic cells 4-5 min after addition of KC1 at indicated concentrations (percentiles as in Fig. 4). B: positive shift in population histograms of [ Ca2+li for data in A. In these figures, only positive values are displayed after prestimulus histograms have been subtracted from poststimulus histograms. i: response to 10 mM KCl; area under curve was ~6% of total population, ii: 20 mM KCl; area equals 13%, iii: 30 mM KCl; area equals 1 l%, iv: 50 mM KCl; area equals 15% v: 70 mM KCl; area equals 27%.

(iii) 5o 25ih!!q+y+e, 5i1

(iv>

tv>

i0

200

400

600

[Ca’+]

800

1000

1200

1400

0

500

(hi)

1000

[Ca2+]

1500

2000

(nM)

[Ca2+]i changesin responseto depolarization

Participation of VSCCs

Depolarization of the cell suspension with KC1 induced an immediate increase in [ Ca2+li. Figure 6 is a two-parameter histogram that illustrates the time course of the response to depolarization. After addition of KCl, the distribution of [ Ca2+li in the population became bimodal, with the appearance of a smaller subpopulation with [ Ca2+li in the range of 800 nM. Analysis of histograms determined that the responsive subpopulation composed 20% of the total cell population. This response was sustained for the duration of the experiment, although in some experiments the level of [ Ca2+li gradually diminished (data not shown). In parallel experiments, potassium acetate produced a calcium response equivalent to that of KCl, whereas sodium chloride had no effect on [ Ca2+li (data not shown). The response to KC1 was concentration-dependent over the range of lo-70 mM (Fig. 7A). The heterogeneity in response is reflected in the shift of the parameters of the [Ca2+]i distribution. The percentage of cells with high [ Ca2+li increased much more with increasing KC1 concentration than did the median (95th percentile from 450 to 1190 nM, median from 7 1.3 to 150 nM, at 70 mM K+). The shift in the distribution of [ Ca2+li in response to KC1 is more clearly illustrated in the “difference histograms” in Fig. 7 B, in which the distribution of [ Ca2+li in unstimulated cells has been subtracted from the distribution after stimulation (only the positive values are displayed). Note that in this experiment the distribution is bimodal: the first mode represents a population of cells with a small shift in [ Ca2+ji in the range of 150-400 nM, and the second represents a larger population in which [Ca2+li shifted to the range of 400-2,000 nM. Membrane depolarization can also be evoked by opening the voltage-sensitive sodium channels. This was achieved with the use of the alkaloid mixture veratrine, which contains the sodium channel agonist veratridine. This compound induced a dose-dependent rise in [Ca2+li; a concentration of 35 pg/ml veratrine produced a 1Ca2+1; eauivalent to 20 mM KC1 (data not shown).

The source of Ca2+ for the increase in [ Ca2+li after KC1 depolarization could have been Ca2+ entry through VSCCs or Ca2+ release from intracellular stores. Several experiments were designed to test these alternatives (Fig. 8). After incubation of cells with 50 mM K+ in the absence of free extracellular Ca 2+ (i.e., 2 mM EGTA), no increase in [Ca2+li was observed. The response to depolarization in Ca2+-containing medium was also attenuated by the divalent cations Cd2+ (2 mM) and Co2+ (2 mM). The K+-induced [ Ca2+li increase was also attenuated by the dihydropyridine calcium channel antagonist nifedipine, and, to a lesser extent, by the organic compound verapamil. In addi-

KC1

+

verapamil

t-i

i

:

i

:

I

I. I 0

I

I

I

I

I

200

400

600

800

1000

1200

1400

[Ca2+] (nM) FIG. 8. KCl-induced increases in [ Ca2+li are attenuated by calciumchannel antagonists. Box-and-whisker plots of [ Ca2’J for samples embryonic day 17 hypothalamic cells, 4-5 min after stimulation by 50 mM KC1 or veratrine (35 pg/ml). EGTA (2 mM), Co2+ (2 mM), verapamil ( 1 PM), and nifedipine ( 1 PM) were added simultaneously with KC1 (percentiles as in Fig. 4).

Downloaded from www.physiology.org/journal/jn at Washington Univ (128.252.067.066) on February 13, 2019.

GRIERSONET AL.

710

tion, the dihydropyridine agonist BayK 8644 was found to potentiate the response in two of eight experiments (data not shown). These data, coupled with the observation that the magnitude of Ca2+ influx was correlated with the concentration of KCl, would suggest that the increases in [ Ca2+li after KC1 addition were due to Ca*+ influx through vsccs.

Cells with high [Ca’l, survive in culture To affirm that responsive cells were neurons, and to determine that the neurons were viable, we collected live cells on the basis of high [Ca*+]i and placed them into culture. The sorting criterion was relative position on a plot of 4 lonm versus 485nm fluorescence, because the slope of a line with these axes is proportional to [Ca*+]i (Chused et al. 1987). The sort gate was set to collect cells >80th percentile of resting [ Ca2+li, before and after 50 mM KC1 addition, into tubes containing DMEM and 15% FBS (see METHODS). The number of neurons that entered the sort gate increased by 20% after KC1 addition. This increase was re-

fleeted as a small increase in cell number as compared with the unstimulated sample after 2 days in culture. The cells were collected and plated onto confluent monolayers of cerebral cortical astrocytes, a method that promotes cell survival and neurite outgrowth (Ventimiglia and Geller 1987). After 24 h in culture, there were many round cells attached to the monolayer, some with short processes. Virtually all of these exhibited some indo-l fluorescence under ultraviolet illumination (Fig. 9, A and B). No fluorescent cells were found to have a glial-like (type- 1 astrocyte; Raff et al. 1983) morphology, and all cells that could be defined morphologically as astrocytes were nonfluorescent and likely to be from the preexisting monolayer. Immunocytochemistry for MAP2, a neuronal marker (Fischer et al. 1986 ) , revealed numerous process-bearing cells (Fig. 9C). All morphological types normally found in cultures of hypothalamic neurons (Ventimiglia and Geller 1987) were observed in these cultures, and no single type appeared to be overrepresented. In agreement with other studies (Chused et al. 1987; Rabinovitch et al. 1986), we found that indo-l is not toxic to embryonic brain cells (data not shown), but it may slightly reduce neurite outgrowth (control: 2 16 * 34 pm/cell, n = 12; indo-l loaded: 172 + 27 pm/cell, 72= 11,4 d.i.v.) . The sorted neurons survived in culture for 2 1 wk. DISCUSSION

In the present study, we have used flow cytometry and the calcium-sensitive probe indo- 1 to investigate aspects of calcium homeostasis and the expression of VSCCs in developing mammalian neurons. This work suggests that I) there are significant internal stores of Ca*+ in these cells; 2) immature CNS neurons exhibit a prodigious ability to expel Ca*+ from the cytoplasm; 3) in a large number of cells, the Na+-Ca*+ exchanger plays a prominent role in maintaining Ca*+ homeostasis; and 4) there is a voltage-dependent uptake of Ca*+ in a proportion of cells, which appears to be mediated via the VSCCs. It was also noted that embryonic cells seldom responded to the experimental manipulations as a single population, and this indicates that the ability to regulate [Ca*+]i in immature neurons is heterogeneous.

Flow cytometry and [Ca”]i estimation

FIG. 9. lndo-l loaded cells, sorted on the basis of high [ Ca*+],, survive in culture. A : phase-contrast photomicrograph of an indo-l-loaded cell after 24 h in culture. This live cell was from a sample sorted by flow cytometer on the basis of resting [ Ca2+], in the range 300- 1,000 nM. Bar: 10 MM. B: ultraviolet fluorescence photomicrograph of the same cell as in A. C: immunocytochemistry using an antibody to MAP2 of sorted cells maintained in culture for 4 days. Cells were sorted on the basis of high Ca’+ after a 50-mM KCl-induced depolarization. Bar: 20 PM.

Flow cytometry was used in the present experiments because it has the advantage of incredible analytic speed, which allows one to acquire data from large numbers of cells in a very short time. Thus cells can be analyzed almost immediately after removal from the brain. It could be argued that these cells are more representative of the in vivo state, because possible differentiation by placing them into culture and possible cell selection through differential survival are avoided. Although this technique has been used extensively in immunologic studies, only recently have its analytic advantages been utilized in neurobiology. It has been used, for example, to sort populations of retrogradely labeled (Schaffner et al. 1987) and antibody-labeled (DiPorzio et al. 1987) neurons. A possible drawback of this methodology is the potential loss of physiological response caused by the dissociation procedure. Although one must

Downloaded from www.physiology.org/journal/jn at Washington Univ (128.252.067.066) on February 13, 2019.

CALCIUM

HOMEOSTASIS

be aware of this possibility, there is a large body of data indicating that, in terms of channel response and secretory activity, acutely isolated cells can be remarkably healthy. For example, physiological responses have been analyzed with the use of voltage-sensitive dyes in spinal cord (Mandler et al. 1988 ) and hippocampal neurons (Fiszman et al. 1990) and with the use of indo- 1 to analyze substance P responsiveness in spinal cord neurons and clonal cells ( Schieren and MacDermott 198 8 ) . [ Ca*+]i was estimated with the use of indo- 1, a structural analog of fura-2, with a similar sensitivity to Ca*+ ( Grynkiewicz et al. 1985) but with fluorescent characteristics more suited to use in flow cytometry (Grynkiewicz et al. 1985; Rabinovitch et al. 1986). In the present study, it appears that the freshly dissociated embryonic neurons hydrolyze the dye efficiently, because most cells exhibited significantly lower I485 in the presence of high ionomycin and more than 95% exhibit a ratio >2.0 under conditions used for R,,,. The neurons also retained the dye for a long period of time, as shown by cellular dye fluorescence 1 day after loading. The calcium calibration of this technique, as with any ionic dye, can only provide a rough estimate of the true level of intracellular Ca*+ . In this study, we have imposed a stringent calibration procedure, i.e., the determination of was made after a period of ATP depletion &in and 4nax and in the presence of Na+ and Ca*+ ionophores. These precautions decreased the likelihood of the cells maintaining an ionic differential during the determination (Chused et al. 1987). The range of [ Ca*+]i that was found in the freshly dissociated embryonic neurons was fairly wide and could have arisen from the inevitable disruption of cell membranes during the dissociation procedure. However, median levels of [ Ca*+]i obtained in different preparations were consistently low ( 67- 175 nM) . These levels are comparable to those found in other preparations: a range of ‘I resting [ Ca*+]i of 40-200 nM has been reported for mouse dorsal root ganglion neurons (Duchen et al. 1990)) whereas cerebellar explant cultures ( >5 d.i.v.) exhibited a range of 50-80 nM (Connor et al. 1987). Perhaps embryonic neurons possess levels of calcium appropriate to their developmental age.

Data collection was conJined to neurons Several recent studies have demonstrated voltage-sensitive calcium conductances in differentiated astrocytes in vitro (Barres et al. 1989; Corvalan et al. 1990), although their presence in vivo is uncertain. Therefore the possibility that astrocytes contributed to the present data must be considered. Several lines of evidence would suggest that by far the greatest majority of cells analyzed in these experiments were neurons. Firstly, attempts to load cultured astrocytes with indo-l in this laboratory have found only very low levels of fluorescence; it appears that the dye is rapidly expelled from the cytoplasm of astrocytes, which was not the case for neurons (see above). It is possible that the prominent peaks of low-fluorescence events were composed of astrocytes, and these dimly fluorescing cells were not used in the statistical analyses. Secondly, it is widely held that the majority of astrocytes are born postnatally, e.g., in the em-

IN EMBRYONIC

NEURONS

711

bryonic day 17 hippocampus only 5% of cells are astrocytes (Gasser and Hatten 1990), whereas in the hamster hypothalamus, glial fibrillary acidic protein immunoreactivity is not detectable until the end of the first postnatal week ( Suarez et al. 1987). In our investigations ( Petroski and Geller, unpublished data), we have also been unable to detect a significant number of glial cells in embryonic day 17 dissociates with the use of immunocytochemical methods. Thus the likelihood of astrocytes contributing significantly to the present results is negligible.

Neurons are heterogeneousin their ability to regulate [Ca 2+]i The ability to regulate [ Ca*+]i is likely to exist from the earliest stage of neuronal development, because failure to regulate [ Ca*+]i seems to result in cell death (Pauwels et al. 1989 ) . However, as neurons mature and extend processes, the requirement for Ca*+ probably changes. A recent study of neuronal development in Helisoma has indicated that cells can be classified into strong or weak regulators of [ Ca*+]i, and that these states are dynamic and correlate with the active growth phase (Mills and Kater 1990). In dissociated embryonic cells, low concentrations of ionomytin (0. l- 1 FM) induced only a small increase in [ Ca*+]i. Although it is possible that nonspecific binding to protein reduces the efficacy of ionomycin as a Ca*+ ionophore (DeLorme et al. 1988), it is more likely that the activity of neuronal Ca*+ pumps extrudes this ion as fast as its enters (Albert and Tashjian 1986). The unimodal distribution of Ca*+ under these conditions suggests that all cells have a roughly equal ability to expel this ion. However, with higher doses of ionomycin (25 PM), the distribution of [ Ca*+]i was bimodal, with most cells exhibiting very high [ Ca*+]i . Thus, in spite of the high ionomycin concentration and associated Ca *+ influx, some cells can continue to maintain relatively low [ Ca*+]i. A heterogeneity in the efficiency of calcium elimination was only demonstrated when the Ca*+ load was excessive and only a subpopulation was able to maintain low [ Ca*+]i. This result suggests that perhaps embryonic hypothalamic neurons, as with Helisoma neurons, might be classified into strong and weak regulators. The heterogeneous distribution could have been an artifact of compartmentalization of indo- 1 or the differential actions of ionomycin in different subcellular compartments. However, the homogeneous distribution in metabolically poisoned cells would suggest that the bimodal distribution in normal cells was due to a differential ability of cells to store or extrude [ Ca*+ Ii. The data also suggest that both energy-dependent and ion-exchange mechanisms are involved in reducing the Ca*+ load and maintaining [Ca*‘]i in these cells. The contribution of energy-dependent mechanisms in reducing an intracellular Ca*+ load was clearly evident during stimulation with high concentrations of ionomycin. The maximum fluorescence ratio was found to increase significantly after ATP rundown, i.e., more Ca*+ accumulated in the cells after energy depletion. The identity of the Ca*+ pump involved is not clear; although extrusion via the Ca*+-ATPase is the most likely explanation, uptake by mitochondria may also be important (Carafoli 1987; Duchen et al. 1990).

Downloaded from www.physiology.org/journal/jn at Washington Univ (128.252.067.066) on February 13, 2019.

712

GRIERSON

Inhibition of mitochondrial function and hypoglycemia have been shown to be important in maintaining resting [ Ca*+]i in cultured mouse sensory neurons (Duchen et al. 1990), and although this was not tested in the present study, it is probably also true of rat hypothalamic neurons. When neurons were transferred to Na+-free medium, resting [ Ca*+]i increased transiently. The increase in resting Ca*+ is probably due to the low extracellular Na+. Mouse sensory neurons exhibited a similar response after the substitution of Li+ for Na+ (Duchen et al. 1990). On the other hand, an increase in resting [ Ca*+]i was not observed in Helisoma neurons under the same conditions (Mills and Kater 1990). Under a greater Ca*+ load, as delivered by a low dose of ionomycin or A23 187, both hypothalamic neurons and Helisoma neurons exhibit a reduced ability to expel Ca*+. Thus the Na+-Ca*+ exchanger appears to be important for the elimination of cytosolic Ca*+ after ionophore stimulation. Intracellular pools of stored Ca*+ have been reported in some neuronal preparations (Andrews et al. 1988; Lipscombe et al. 1988 ) , and ionomycin treatment after a brief period of Ca*+ depletion has been used to demonstrate the existence of a thyrotropin-releasing hormone-sensitive store in the pituitary-derived cell line, GH& (Albert and Tashjian 1986). Using the combination of EGTA and ionomycin, the present study provides evidence for intracellular Ca*+ pools in embryonic hypothalamic neurons. Ionomycin induced a transient intracellular Ca*+ increase despite the absence of extracellular free Ca*+ ; this Ca*+ must have been liberated from intracellular pools. One pool is probably identical to that in frog sympathetic neurons, where Ca*+ can be released by caffeine and by Ca*+ (i.e., Ca *+ -induced Ca*+ release, Lipscombe et al. 198 8 ) . An inositol- 1,4,5=trisphosphate=sensitive pool may also be present (Berridge and Irvine 1989). It is likely that these pools are ionomycin sensitive. In the continued presence of EGTA and ionomycin, some cells in the present study maintained high [ Ca*+]i for a prolonged period. Because the calciumbinding protein calbindin appears to be particularly abundant in the embryonic hypothalamus (Enderlin et al. 1987), it is possible that the presence of calcium-binding proteins in this population slowed the release of Ca*+ from the internal pool.

VSCCs are present in a subpopulation of embryonic neurons It seems clear that Ca*+ plays an important role in growth cone mobility and neurite extension (Cohan et al. 1987; Mattson and Kater 1987). A growing body of evidence also indicates that differentiation is correlated with increased expression of VSCCs (Noronha-Blob et al. 1988; Gottmann et al. 1988 ) and that VSCCs are clustered within the growth cone (Anglister et al. 1982; Silver et al. 1990). We have found that in freshly dissociated embryonic neural tissue, consisting of neuronal somata, depolarization by KC1 or veratrine increases [ Ca*+]i. The effect of K+ was concentration dependent over the range of lo-70 mM, blocked by Co*+ or Cd*+, and abolished in the absence of extracellular free Ca*+ . The response of embryonic neurons to KC1 was also attenuated by the calcium-channel antago-

ET AL.

nists nifedipine and verapamil. The source of Ca*+ thus appeared to be extracellular, although the potential contribution of Ca*+-induced Ca*+ release has not been eliminated. Although most of this work was conducted on hypothalamic neurons, similar responses were observed in neurons from the cerebral cortex and hindbrain. Interestingly, the increase in the median [ Ca*+]i in embryonic hypothalamic neurons in response to depolarization appeared to be attributable largely to a rise in [ Ca2+li, in -20% of the population, to a level of - 800 nM with 50 mM K+ . A K+-induced increase in [ Ca*+]i of this magnitude has been reported in other neuronal preparations where optical methods were used. Lipscombe and colleagues ( 1988) have shown [ Ca*+]i approaching 1 PM in depolarized frog sympathetic neurons, and Connor et al. ( 1987) reported that some K+-induced [ Ca*+]i peaks in cerebellar granule cells were -2 PM. Veratradine has been shown to increase [ Ca*+ Ii in freshly dissociated immature Purkinje cells to 1.6 PM (Sorimachi et al. 1990). Thus it would appear that these estimates of neuronal [ Ca*+]i from flow cytometry are consistent with those from other optical methods. The identity of the VSCC subtype(s) activated in these cells remains uncertain. Most previous studies suggest that the transient, or low-voltage activated, channel develops first (Gottmann et al. 1988; McCobb et al. 1989; Yaari et al. 1987)) but it seems unlikely that these would permit the large and long-lasting changes in [ Ca*+]i seen in the present work. Rather, the presence of non-inactivating L channels is suggested by fact that the Ca*+ flux was virtually eliminated by nifedipine. Nonetheless, the data support the hypothesis that the response belies Ca*+-specific channels displaying voltage-dependent activation and pharmacological properties analogous to neuronal VSCCs (Gottmann et al. 1988; Nowycky et al. 1985; Yaari et al. 1987) and that these are present in -20% of cells at the embryonic age examined. In summary, we have demonstrated that flow cytometric analysis allows the characterization of a variety of Ca*+ -regulatory mechanisms in populations of dissociated embryonic neurons. Plow cytometry detected subpopulations of cells that exhibited differences in Ca*+ homeostasis, both in terms of channel expression and calcium storage and extrusion. These cells were successfully sorted on the basis of their intracellular calcium concentration and maintained in tissue culture for further study. Clearly, much more information is needed to completely characterize calcium regulation during this stage of neural development. These methods could then be used to obtain and further analyze populations of neurons on the basis of their response to these and other pharmacological agents, such as neurotransmitters, which interact with VSCCs. We thank Dr. E. Yurkow for valuable assistance with the flow cytometer and Drs. J. and D. La&in for their many helpful suggestions. This research was supported by National Institute of Neurological Disorders and Stroke Grant NS-25 168 (to H. Geller). Present addresses: J. P. Grierson, AFRC Unit of Molecular Signalling, Dept. of Zoology, Downing Street, Cambridge CB2 3EJ, UK; S. M. O’Connell, Dept. of Surgery, UMDNJ-Robert Wood Johnson Medical School, New Brunswick, NJ.

Downloaded from www.physiology.org/journal/jn at Washington Univ (128.252.067.066) on February 13, 2019.

CALCIUM

HOMEOSTASIS

Address for reprint requests: H. M. Geller, Dept. of Pharmacology, Robert Wood Johnson Medical School, 675 Hoes Lane, Piscataway, NJ, 08854.

Received 14 March 199 1; accepted in final form 5 November 199 1. REFERENCES P. R. AND TASHJIAN, A. H., JR. Relationship of thyrotropin-releasing hormone-induced spike and plateau phase in cytosolic free Ca2’ concentrations to hormone secretion. Selective blockade using ionomytin and nifedipine. J. Biol. Chem. 259: 15350-15363, 1984. ALBERT, P. R. AND TASHJIAN, A. H., JR. Ionomycin acts as an ionophore to release TRH-regulated Ca2+ stores from GH& cells. Am. J. Physiol. 25 1 (Cell Physiol. 20): C887-C89 1, 1986. ALCOVER, A., WEISS, M. J., DALEY, J. F., AND REINHERZ, E. L. The Tl 1 glycoprotein is functionally linked to a calcium channel in precursor and mature T-lineage cells. Proc. Natl. Acad. Sci. USA 83: 26 14-26 18,

ALBERT,

1986. ANDREW&

S. B., LEAPMAN, R. D., LANDIS, D. M. D., AND REESE, T. S. Activity-dependent accumulation of calcium in Purkinje cell dendritic spines. Proc. Natl. Acad. Sci. USA 85: 1682-1685, 1988. ANGLISTER, L., FARBER, I. C., SHAHAR, A., AND GRINVALD, A. Localization of voltage-sensitive calcium channels along developing neurites: their possible role in regulating neurite elongation. Dev. Biol. 94: 35 l365,1982. BARRES, B.

A., CHUN, L. L. Y., AND COREY, D. P. Calcium current in cortical astrocytes: induction by CAMP and neurotransmitters and permissive effects of serum factors. J. Neurosci. 9: 3 169-3 175, 1989. BERRIDGE, M. J. AND IRVINE, R. F. Inositol phosphates and cell signalling. NatureLond. 341: 197-205, 1989. BLAUSTEIN, M. P. Calcium transport and buffering in neurons. Trends Neurosci. 11: 438-443, 1988. CARAFOLI, E. Intracellular calcium homeostasis. Annu. Rev. Biochem. 56: 395-433,1987. CAZALIS, M., DAYANITHI,

G., AND NORDMANN, J. J. Hormone release from isolated nerve endings of the rat neurohypophysis. J. Physiol. Lond. 390: 55-70, 1987. CHOI, D. W. Calcium-mediated neurotoxicity: relationship to specific channels types and roles in ischemic damage. Trends Neurosci. 11: 465469,1988. CHUSED, T. M., WILSON, A., GREENBLATT, D., ISHIDA, Y., EDISON, L. J., TSIEN, R. Y., AND FINKELMAN, F. D. Flow cytometric analysis of mu-

rine splenic B lymphocyte cytosolic free calcium response to anti-IgM and anti-IgD. Cytometry 8: 396-404, 1987. COHAN, C. S., CONNOR, J. A., AND KATER, S. B. Electrically and chemically mediated increases in intracellular calcium in neuronal growth cones. J. Neurosci. 7: 3588-3599, 1987. CONNOR, J. A. Digital imaging of free calcium changes and of spatial gradients in growing processes in single mammalian central nervous system cells. Proc. Natl. Acad. Sci. USA 83: 6 179-6 183, 1986. CONNOR, J. A., TSENG, H.-Y., AND HOCKBERGER, P. E. Depolarizationand transmitter-induced changes in intracellular Ca2+ of rat cerebellar granule cells in explant cultures. J. Neurosci. 7: 1384-l 400, 1987. CORVALAN, V., COLE, R., DE VELLIS, J., AND HAGIWARA, S. Neuronal modulation of calcium channel activity in cultured rat astrocytes. Proc. Natl. Acad. Sci. USA 87: 4345-4348, 1990. DELORME, E. M., RABE, C. S., AND MCGEE, R., JR. Regulation of the number of functional voltage-sensitive Ca++ channels on PC 12 cells by chronic changes in membrane potential. J. Pharmacol. Exp. Ther. 244: 838-843,1988. DIPORZIO, U., ROUGON,

G., NOVOTNY, E. A., AND BARKER, J. L. Dopaminergic neurons from embryonic mouse mesencephalon are enriched in culture through immunoreaction with monoclonal antibody to neural specific protein 4 and flow cytometry. Proc. Natl. Acad. Sci. USA

84: 7334-7338, 1987. DUCHEN, M. R., VALDEOLMILLOS,

M., O’NEILL, S. C., AND EISNER, D. A. Effects of metabolic blockade on the regulation of intracellular calcium in dissociated mouse sensory neurones. J. Physiol. Lond. 424: 4 1 l-426,

1990. ENDERLIN,

S., NORMAN, A. W., AND CELIO, M. R. Ontogeny of the calcium binding protein calbindin D-28k in the rat nervous system. Anat. Embryol. 177: 15-28, 1987.

IN EMBRYONIC

NEURONS

713

I., SHEA, T. B., SAPIRSTEIN, V. S., AND KOSIK, K. S. Expression and distribution of microtubule-associated protein 2 (MAP2) in neuroblastoma and primary neuronal cells. Brain Res. 390: 99- 109, 1986. FISZMAN, M. L., NOVOTNY, E. A., LANGE, G. D., AND BARKER, J. L. Embryonic and early postnatal hippocampal cells respond to nanomolar concentrations of muscimol. Dev. Brain Res. 53: 186-l 93, 1990. GASSER, U. E. AND HATTEN, M. E. Neuron&a interactions of rat hippocampus in vitro: glial-guided neuronal migration and neuronal regulation of glial differentiation. J. Neurosci. 10: 1276-1285, 1990. GOTTMANN, IS., DIETZEL, I. D., Lux, H. D., HUCK, S., AND ROHRER, H. Development of inward currents in chick sensory and autonomic neuronal precursor cells in culture. J. Neurosci. 8: 3722-3732, 1988. GRIERSON, J. P., PETROSIU, R. E., O’CONNELL, S. M., AND GELLER, H. M. Flow cytometric analysis of calcium fluxes in dissociated embryonic rat hypothalamic neurons. Sot. Neurosci. Abstr. 14: 135, 1988. GRYNKIEWICZ, G., POENIE, M., AND TSIEN, R. Y. A new generation of Ca2+ indicators with greatly improved fluorescence properties. J. Biol. Chem. 260: 3440-3450, 1985. Hsu, S.-M., RAINE, L., AND FANGER, H. Use of avidin-biotin-peroxidase complex (ABC) in immunoperoxidase techniques: a comparison between ABC and unlabeled antibody (PAP) procedures. J. Histochem. Cytochem. 29: 577-580, 198 1. JAHNSEN, H. AND LLINAS, R. Ionic basis for the electroresponsiveness and ocillatory properties of guinea-pig thalamic neurones in vitro. J. Physiol. Lond. 349: 227-247, 1984. JUNE, C. H. AND RABINOVITCH, P. S. Flow cytometric measurement of cellular ionized calcium concentration. Pathol. Immunopathol. Res. 7: FISCHER,

409-432,1988. KENNEDY, M. B. Regulation of neuronal function by calcium. Trends Neurosci. 12: 4 17-420, 1989. LIPSCOMBE, D., MADISON, D. V., POENIE, M., REUTER, H., TSIEN, R. W., AND TSIEN, R. Y. Imaging of cytosolic Ca2+ transients arising from Ca2+

stores and Ca2+ channels in sympathetic neurons. Neuron 1: 355-365, 1988. LYNCH, G., LARSON, J., KELSO, S., BARRIONUEVO, G., AND SCHOTTLER, F. Intracellular injections of EGTA block induction of hippocampal long-term potentiation. Nature Lond. 305: 7 19-72 1, 1983. MANDLER, R. N., SCHAFFNER, A. E., NOVOTNY, E. A., LANGE, G. D., AND BARKER, J. L. Flow cytometric analysis of membrane potential in embryonic rat spinal cord cells. J. Neurosci. Methods 22: 203-2 13, 1988. MATTSON, M. P. AND RATER, S. B. Calcium regulation of neurite elongation and growth cone motility. J. Neurosci. 7: 4034-4043, 1987. MCCOBB, D. P., BEST, P. M., AND BEAM, IS. G. Development alters the expression of calcium currents in chick limb motoneurons. Neuron 2: 1633-1643,1989. MILLS, L. R. AND KATER,

S. B. Neuron-specific and state-specific differences in calcium homeostasis regulate the generation and degeneration of neuronal architecture. Neuron 4: 149- 163, 1990. NORONHA-BLOB, L., RICHARD, C., AND UPRICHARD, D. C. Voltage-sensitive calcium channels in differentiated neuroblastoma X glioma hybrid (NG 108- 15) cells: characterization by Quin 2 fluorescence. J. Neurochem.50: 1381~1390,1988. NOWYCKY, M. C., Fox, A. P., AND TSIEN, R. W. Three types of neuronal calcium channel with different calcium agonist sensitivity. Nature Lond. 316:440-443,1985. OPPENHEIM, R. W.

Naturally occurring cell death during neural development. Trends Neurosci. 11: 487-493, 1988. PAUWELS, P. J., VAN Assouw, H. P., LEYSEN, J. E., AND JANSSEN, P. A. J. Ca2+-mediated neuronal death in rat brain neuronal cultures by veratridine: protection by flunarizine. Mol. Pharmacol. 36: 525-53 1, 1989. PERNEY, T. M., HIRNING, L. D., LEEMAN, S. E., AND MILLER, R. J. Multiple calcium channels mediate neurotransmitter release from peripheral neurons. Proc. Natl. Acad. Sci. USA 83: 6656-6659, 1986. RABINOVITCH, P. S., JUNE, C. H., GROSSMAN, A., AND LEDBETTER, J. A. Heterogeneity among T cells in intracellular free calcium responses after mitogen stimulation with PHA or anti-cd3. Simultaneous use of INDO1 and immunofluorescence with flow cytometry. J. Immunol. 137: 952961, 1986. RAFF, M. C., ABNEY, E. R., COHEN, J., LINDSAY, R., AND NOBLE, M. Two types of astrocytes in cultures of developing rat white matter: differences in morphology, surface gangliosides, and growth characteristics. J. Neurosci. 3: 1289-1300, 1983. SCHAFFNER, A. E., ST. JOHN, P. A., AND BARKER, J. L. Fluorescence-acti-

Downloaded from www.physiology.org/journal/jn at Washington Univ (128.252.067.066) on February 13, 2019.

714

GRIERSON

vated cell sorting of embryonic mouse and rat motoneurons and their long-term survival in vitro. J. Neurosci. 7: 3088-3 104, 1987. SCHIEREN, I. AND MACDERMOTT, A. Plow cytometric identification and purification of cells by ligand-induced changes in intracellular calcium. J. Neurosci. ikfethods 26: 35-44, 1988. SILVER, R. A., LAMB, A. G., AND BOLSOVER, S. R. Calcium hotspots caused by L-channel clustering promote morphological changes in neuronal growth cones. Nature Lond. 343: 75 I-754, 1990. SNUTCH, T. P., LEONARD, J. P., GILBERT, M. M., LESTER, H. A., AND DAVIDSON, N. Rat brain expresses a heterogeneous family of calcium channels. Proc. Nat/. Acad. Sci. USA 87: 3391-3395, 1990. SONTHEIMER, H., TROTTER, J., SCHACHNER, M., AND KETTENMANN, H. Channel expression correlates with differentiation stage during the development of oligodendrocytes from their precursor cells in culture. Neuron 2: 1135-l 145, 1989. SORIMACHI, M., MORITA, Y., AND NAKAMURA, H. Possible regulation of the cytosolic-free calcium concentration by Na+ spikes in immature cerebellar Purkinje cells. Neurosci. Lett. 111: 333-338, 1990.

ET AL. SUAREZ, I., FERNANDEZ, GARCIA-SEGURA, L.

B., BODEGA, G., TRANQUE, P., OLMOS, G., AND M. Postnatal development of glial fibrillary acidic protein immunoreactivity in the hamster arcuate nucleus. Dev. Brain Res. 37: 89-95, 1987. TUKEY, J. W. Exploratory Data Analysis. Reading, MA: Addison-Wesley, 1977. VENTIMIGLIA, R. AND GELLER, H. M. Cell types and cell-substrate interactions in serum-free dissociated cultures of rat hypothalamus. Brain Res. 436: 339-35 1, 1987. VON SPRECKELSEN, S., LOLLIKE, K., AND TREIMAN, M. Ca2+ and vasopressin release in isolated rat neurohypophysis: differential effects of four classes of Ca2+ channel ligands. Brain Res. 5 14: 68-76, 1990. WOLF, M. E. AND KAPATOS, G. Plow cytometric analysis of rat striatal nerve terminals. J. Neurosci. 9: 94- 105, 1989. YAARI, Y., HAMON, B., AND Lux, H. D. Development of two types of calcium channels in cultured mammalian hippocampal neurons. Science Wash. DC 235: 680-682, 1987.

Downloaded from www.physiology.org/journal/jn at Washington Univ (128.252.067.066) on February 13, 2019.

Calcium homeostasis in dissociated embryonic neurons: a flow cytometric analysis.

1. Ca2+ homeostasis in freshly dissociated neurons from embryonic rat hypothalamus, cortex, and brain stem was investigated with flow cytometry. Cells...
3MB Sizes 0 Downloads 0 Views