ELSEVIER Brain Research 660 (1994) 329-332

BRAIN RESEARCH

Short communication

Purine- and pyrimidine-stimulated phosphoinositide breakdown and intracellular calcium mobilisation in astrocytes

Brian Pearce *, David Langley Department of Pharmacolo~', The School of Pharmacy, 29 / 39 Brunswt~ k Square, London WC1N lAX, UK

Accepted 26 July 1994

Abstract

PhospholnosJtide breakdown in cultured cortical astrocytes was assessed by measuring the accumulation of [~H]lnosltol phosphates (IP's) following incubations with various purines and pyrimldlnes. Dose-response relationships gave the following order of potency: 2-methylthloadenosine triphosphate (2-MeSATP) > urldine 5'-trlphosphate (UTP) > ATP = ADP > moslne 5' trlphosphate (ITP). However, 2-MeSATP and UTP were only half as effective as either ATP or ADP in stimulating [3H]IP production Astrocytes were also challenged with combined additions of maximally effective concentrations of agonlsts Responses to ADP plus UTP and 2-MeSATP plus UTP were essentially additive whilst ATP plus UTP evoked a response which was only partially additive. ATP-stimulated [3H]IP accumulation was markedly reduced in the presence of 2-MeSATP suggesting that the latter may be a partial agonlst at these receptors. We also examined the ability of ATP and UTP to increase intracellular Ca 2+ concentrations in these cells. Greater than 90% of all cells tested responded to ATP with a release from internal Ca 2+ stores but less than half of these responded similarly when challenged with UTP. Our results indicate that astrocytes possess both Pzv-purlnoceptors and a population of receptors which are also coupled to phosphomositide metabohsm and lntracellular Ca z+ mobilisatxon but recognlse ATP and the pyrlmidlne nucleotide UTP

Keywords. Astroeyte; Inositol phosphate; Ca 2+ mobihsation; Purine, Pyrlmldlne

Cell surface receptors for ATP are termed P2- purinoceptors, the receptor class being further divided into four subtypes (P2x, PzY, PzT, P2z) on the basis of criteria such as agonist potency, the response elicited upon receptor stimulation and the cell types possessing the receptors [3,6]. CNS astrocytes in culture possess receptors for ATP, activation of which leads to phos- phoinositide breakdown, increased cytosolic Ca 2+ con- centration and the release of eicosanoids [5,13,16]. By using selective agonists, two groups have provided evi- dence which indicates that these receptors are of the P2v subtype [1,9]. However, O 'Connor et al. [14] have suggested that the classification of P2v receptors should be reconsidered in the light of findings which show that certain cellular responses initiated by ATP can also be evoked by the pyrimidine uridine 5' t r iphosphate (UTP) and that some P2v agonists, such as 2-methylthioa- denosine tr iphosphate (2-MeSATP), are either unable

* Correspondmg author Fax (44) 071-753 5902

0006-8993/94/$07 00 © 1994 Elsevmr Science B V All rights reserved SSD! 0006-8993(94)00900-7

to or only partially capable of mimicking the effects of ATP and UTP. This has led to the proposal that some cells and tissues may possess so-called 'nucleotide ' or P2u receptors. In order to determine whether cortical astrocytes in culture possess such receptors, we have examined the effects of various purlnes and pyrim- idines on phosphoinositide breakdown and intracellu- lar Ca 2+ mobllisation in these cells.

Astrocyte-enriched cultures were prepared from newborn rat cerebral cortex [4], grown on poly-D- lysine-coated 24-well plates or 32 mm 2 glass coverslips and maintained in vitro for up to 28 days. Immunocyto- chemical labelling with an antibody against glial fibril- lary acidic protein revealed that > 90% of the cells in these cultures were astrocytes.

To assess [3H]IP accumulation, cultures in 24-well plates were incubated for 48 h in medium containing 1 /xCi /ml [3H]myo-inositol (15 C1/mmol, Amersham) to prelabel membrane phosphoinositides. Cultures were washed in buffer of the following composition: NaCI (116 mM), N a H C O 3 (26 mM), N a H : P O 4 (1 mM),

330 B t'carc,:, D /,anglo /Brain Resear(h 060 (1994) 3_9 ?,3_

MgSO 4 (1.5 mM), KC1 (5 mM), CaC1, (1.3 raM), glu- cose (20 mM) pregassed with 5% CO~ in 0 2 , then incubated m the same buffer containing 5 mM LICI and an agonist (all from Sigma except 2-MeSATP which was purchased from RBI Inc ). All incubations were Ior 60 mln at 37°C in a 5% C O j a l r atmosphere. Reactions were terminated by removal of the mcuba- non medium and the addition of 0.5 ml ice-cold methanol to each well. The cells were harvested and the [~H]IP's extracted by the addition of 0.75 ml of water and 0.75 ml of chloroform to each sample. Aliquots (0.75 ml) of the aqueous phases were taken, diluted to 3 ml with water then applied to columns containing 0.5 ml of a 50% slurry of AG l-X8 resin m the formate form (Biorad). Following five washes with 2 ml volumes of water, the total [~H]IP's were eluted from the columns with 1 ml of I M ammonium formate in 0.1 M formic acid.

Cultures on 32 mm 2 coverslips were incubated for 30 rain at room temperature in the dark m buffer (142.5 mM NaCI, 3.75 mM KC1, 1 mM MgClz, 1 mM CaC12. 11 mM glucose, 5 mM HEPES, pH 7.4) con- taining 0.5 tzM of the Ca 2+ indicator Fura-2 ace- toxymethyl ester (Fura-2AM, Sigma). Coverslips were mounted on the stage of an inverted microscope (Nikon Diaphot) and fields of approximately 10 cells were viewed at 400 × using a Nikon 40 × 1.3NA phase-con- trast oil immersion fluorescence objective. The cells were placed in the lightpath of a photomultiplier, per- fused with buffer of the above composition at a rate of 0.2 ml /mln then illuminated alternately with 360 nm and 390 nm narrow band filtered UV light, the filters being changed every 250 msec. The emitted light was collected via the photomultiplier, filtered then digitised by a CED 1401 Laboratory Interface (Cambridge Elec- tronic Design) before being fed into an IBM compati- ble computer. The ratio of the emission signals at 360 nm and 390 nm were used to calculate the free mtra- cellular Ca 2+ concentration according to the method of Grynkiewicz et al. [7].

Dose-response relationships for purine- and pyriml- dine-stimulated [3H]IP accumulations are shown in Fig. 1. The result for each agonist dose used is ex- pressed as a percentage of the response to a maximally effective concentration (250 /xM) of ATP. ATP and ADP evoked the largest accumulations of [3H]IP's, the maximum responses elicited were essentially the same as were their ECs0 values (ATP, 7/zM; ADP, 10 tzM). However, the P2v-selective agonist 2-MeSATP was only 60% as effective as ATP in promoting phosphoinosl- tide breakdown although it was considerably more po- tent (ECs0 = 0.2 /zM). The pyrimldines UTP and mo- sine 5' triphosphate (ITP) also stimulated [3H]IP for- mation in these cells. In both cases the maximum response elicited was approximately 50% of that in- duced by ATP. UTP was found to be slightly more

100

8 0 / ~ / o

/ ,

0 • • ! 1 i i =

10 9 8 7 6 5 4 3

- L O G ( M ) A G O N I S T

Fig 1 Dose - r e sponse relationships for purme- and pynmtdme- snmula ted [3H]IP accumulation The response ehclted by each ago- rest concentranon used is expressed as a percentage of the response to a maximally effectwe dose (250 /zM) of ATP and are means (S.E M. < 12%) of 3 -5 determinations The following agomsts were used' ATP (e), A D P (©), 2-MeSATP ( • ). UTP ( [] ) and ITP ( A )

potent (ECs0 = 2 tzM) than ATP whereas the ECs0 value for ITP was in excess of 100/~M.

UTP is known to regulate a variety of cellular events in peripheral tissues and clonal cell lines including some of neural origin [10,14,18]. Here we show that UTP promotes phosphoinositide breakdown in cul- tured CNS astrocytes thus confirming the recent obser- vations of Bruner and Murphy [2]. However, the key findings to emerge from these experiments concern the efficacies and potencies of 2-MeSATP and UTP rela- tive to ATP. The results for 2-MeSATP contrast with those reported by Kastritsis et al. [9] who found this agonist to be equally as effective as ATP in stimulating the production of [3H]IP's in cortical astrocytes. This disparity might be explained by the ATP dose ranges used in that the maximum concentration of ATP used by these authors was 10/xM which, in our hands, was equivalent to the ECs0 value. Studies on a number of other cell types have shown that ATP concentrations in the high micromolar range are required to elicit maxi- mal phosphoinositide breakdown [8,12,15,17]. Never- theless, the results in Fig. 1 clearly show that ATP, ADP, 2-MeSATP and UTP are all capable of stimulat- ing [3H]IP accumulation in astrocytes. In this regard our data do not correspond exactly to those expected for 'nucleotide' receptor-mediated responses as out- hned by O'Connor et al. [14]. One would have expected UTP to be more potent but equally as effective as ATP with 2-MeSATP having no effect at all. However, it has been suggested that certain cells and tissues may pos- sess a mixed population of receptors where 2oMeSATP, ATP and UTP are all capable of eliciting a response but where 2-MeSATP is less effective than ATP and does not antagonise its effect [14]. In an attempt to determine whether this is the case in astrocytes, we

B Pearce, D Langley/Brain Research 060 (1994) 329-332 331

0 O - -

0 O - -

0 O-- LO

0 O -

0 O - e~

0 0 - OI

0 O -

et I--

0

I

o .

_ L

!

0 . I-.

@

I ,- I , ¢~

._L

O.

÷ 4- O. I,-

1

. . 1 . . e l el

• ®

I I

+ ÷

O. 0 . I ' - I--

Fig. 2 The effect of combined agomst addmons on [3H]IP accumula- tion Cultures were incubated with ATP, ADP, UTP (all at 250 g M )

and 2-MeSATP (10 p.M) either alone or m combination Results are

m e a n s ± S E M from 3-5 determinat ions and are expressed as a percent stimulation over basal (260 +_ 38 dpm, n = 5) [3H]IP accumu-

lation

examined the effect of combined agonist additions on phosphomosit ide breakdown.

Accumulations of [3H]IP's in response to maximally effective agonlst concentrations used either alone or in combination are shown in Fig. 2. As m the experiments described in Fig. 1, we found ATP and ADP to be the most effective agonists with 2-MeSATP and UTP being approximately half as effecttve. A combined addition of ATP and ADP produced an accumulation of [3H]IP's whtch was essentially the same as that evoked by either agonist alone. This confirms our previous conclusion that ATP and ADP act on the same receptors on these cells [16]. Incubations with UTP in combination with either ATP or ADP gave contrasting results in that [3H]IP accumulation following the addition of ADP and UTP together was 92% of that predicted for a fully additive response whereas that for ATP and UTP together was just 76% of the predicted value. One conclusion to be drawn from the first of these experi- ments is that ADP and UTP act on distinct populations of receptors on these cells, however, the nature of the interaction between ATP and UTP is not so clear cut. The fact that UTP did not antagonise the effect of

ATP, as has been shown in other cells [12], but pro- duced a partial additivity ts again suggestive of a mixed population of receptors but with ATP bemg recognised by both. Combined additions of 2-MeSATP plus both ATP and UTP were also tested. The results in Fig 2 suggest that 2-MeSATP may act as a parttal agonist in that it reduced ATP-st imulated [3H]IP accumulation to a level below that elicited by either agonist alone In contrast, when UTP and 2-MeSATP were added to- gether their effects were almost completely additive (87% of the predicted response), thus adding further evidence to our proposal that astrocytes possess recep- tors for both purines and pynmidines. It would appear, however, that pyrtmidine receptors are not present on all astrocytes m these cultures

2000

1000

0 1500

1 0 0 0

=E

"t 0 500

0

2 0 0 0

lOOO

\

A

\ Fig 3 Changes m =ntracellular Ca 2+ concentration in response to ATP and UTP Astrocytes were challenged for 30 s (indicated by the bar above each trace) with 10 ~,M concentrations of either ATP or UTP In A and B, cells were exposed sequentially to ATP then UTP, m C, UTP was apphed before ATP The traces are representative of

approximately 30 cell fields examined over a number of experiments on different batches of cultures

3:t2 B Pear, e, D Langle~ ~Brain Research 660 (1994) 329-332

Investigations into intracellular Ca 2 + mobdtsatlon in Fura-2AM-loaded ceils revealed that the majority ( > 95%) of astrocytes respond to ATP, a findmg which is in accordance with previously published observations [11]. Characteristically, ATP was found to evoke a large mcrease in cytosohc Ca 2+ concentration which in many cases was as much as 20-fold greater than the basal levels. The response was rapid in onset but decayed within 10-15 s to a level some 5-fold above resting Ca 2+ concentrations and remained there for the duration of the agonist application (Ftg. 3). Of all the cell fields examined approximately 60% failed to show a response to UTP (Fig. 3C), of those that did, the majority (70%) evoked a response which bore a striking resemblance to that of ATP (Fig. 3A) the remainder giving only the mitial raptd phase of the response (Fig. 3B).

The conclusions drawn from thts study are that cortical astrocytes in culture possess phospholnosltide- linked receptors for both purines and pyrimidines. The purinoceptors are of the PzY subtype where ATP, ADP and 2-MeSATP are agonists but where the latter may be a partial agonist. Receptors for UTP which also recognise ATP but not ADP and 2-MeSATP are also present but only exist on discrete populations of astro- cytes

This work is supported by The Wellcome Trust, D.L. is in receipt of an MRC studentship.

[1] Bruner, G and Murphy, S, ATP-evoked arachldomc aod mo- bd~sat~on m astrocytes is vm a P2.~-purmergic receptor, J Neu- rochem, 55 (1990) 1569-1575

[2] Bruner. G and Murphy, S, UTP actwates multiple second messenger systems in cultured rat astrocytes, Neurosct Lett , 162 (1993) 105-108

[3] Burnstock, G. and Kennedy, C, Is there a bas~s for distinguish- mg two types of P2-purlnoceptorg, Gen. Pharmacol, 16 (1985) 433-440

[4] Dutton, G R, Cume, D N and Tear, K, An Improved method for the bulk isolation of vmble perikarya from postnatal cerebel- lum, J Neurosct. Methods, 3 (1981) 421-427

[5] Geb~cke-Haerter. P J, Wurster, S. Schobert, A and Herrtmg.

G, P2-purmoceptor reduced prostaglandm ",3,nthcsla m prlmar~ rat astrocyte cultures, Naltnyn-Scmetdebe1~, ~ 4~h Phatma{~.l 338 (1988) 704-707

[6] Gordon, J L., Extracellular ATP ellect, sources and late, Btochem J, 233 (1986) 309-319

[7] Grynklewlcz, G, Poenle, M and Tslen, R Y, A new generation of Ca :+ mdlcators w~th greatly ~mproved fluorescence propel- ties, J Btol Chem, 260 (1985) 3440-3450

[8] Huang, H M and Sun, G Y, Effects ot ATP on phosphatldyh- nosltol-phosphohpase C and mosltol 1-phosphate accumulation ,n rat brain synaptosomes, J Neurochem, 50 (1988) 366-374

[9] Kastntsls, C H.C, Salm, A K. and McCarthy, K D., Stimulation of the Pc', purmerg~c receptor on type I astrogha results m mos~tol phosphate formation and calcmm mobdlsat~on, J IVeu- rochem, 58 (1992) 1277-1284

[10] Lm, T-A, Lustlg, K D, Sportlello, M G. Welsman, G A and Sun, G Y, Signal transductlon pathways coupled to a P2u re- ceptor m neuroblastoma>,ghoma (NG108-15) cells, J Neu- rochem, 60 (1993) 1115-1125

[11] McCarthy, K D and Salm, A.K, Pharmacologically-distract sub- sets of astrogha can be ~dentlfied by their calcmm response to neurohgands, Neuroscwnce, 41 (1991) 325-333

[12] Murrm, R.J and Boarder, M R., Neuronal 'nucleotlde' receptor hnked to phosphohpase C and phosphohpase D? Stlmulatmn of PC12 cells by ATP analogues and UTP, Mol Pharmacol, 41 (1992) 561-568

[13] Neary, J T , van Breeman, C, Forster, E, Norenberg, L O B and Norenberg, M.D, ATP stimulates calcium reflux m primary astrocyte cultures, Btochem Btophys Re,s Commun, 157 (1988) 1410-1416

[14] O'Connor, S E, Dainty, I.A and Left, P, Further subclassfflca- tlon of ATP receptors based on agonlst studies, Trend~ Pharma- col Sct, 12 (1991) 137-141

[15] Pavenstadt, H., Spath, M, Schlunck, G, Nauk, M, Fischer, R., Wanner, C. and Scholtmeyer, P, Effects of nucleotides on the cytosohc free calcmm actlwt~ and mos~tol phosphate formation m human glomerular epxthehal cells, Br J Pharmacol, 107 (1902) 189-195

[16] Pearce, B, Murphy, S, Jeremy, J , Morrow, C and Dandona, P, ATP-evoked Ca 2 + mobdlsatlon and prostanold release from astrocytes P2-purmerglc receptors linked to phosphomosittde hydrolysis, J Neurochem, 52 (1989) 971-977

[17] Sato, K., Okajima, F and Kondo, Y., Extracellular ATP stimu- lates three different receptor-signal transduction systems m FRTL-5 thyroid cells, Btochem J, 283 (1992) 281-287

[18] Seffert, R. and Schultz, G., Involvement of pynmldmoceptors an the regulation of cell functions by urldme and uracil derivatives, Trends Pharmacol Sct., 10 (1989) 365-369