• We are sorry, but NCBI web applications do not support your browser and may not function properly. More information
Logo of jphysiolThe Journal of Physiology SiteMembershipSubmissionJ Physiol
J Physiol. Sep 15, 1999; 519(Pt 3): 723–736.
PMCID: PMC2269531

Two different ionotropic receptors are activated by ATP in rat microglia


  1. Our aim was to assess whether ATP-induced inward currents in microglia are due to a single or more than one purinergic receptor. The ATP dose-response curve showed two components, whose presence might be due to the activation of high and low affinity receptors.
  2. The P2Z/P2X7 specific receptor agonist benzoylbenzoyl-ATP (Bz-ATP) and some P2 receptor agonists were tested. The rank order of potency was Bz-ATP >> ATP = 2-methylthio-ATP (2-MeSATP) > α,β-methylene ATP (α,β-meATP) ≥ ADP. β,γ-MethyleneATP (β,γ-meATP), UTP and adenosine were ineffective.
  3. The non-specific P2 receptor antagonist suramin antagonized by 92 ± 2 % the inward current induced by 100 μm ATP, and by 51 ± 8 and 68 ± 6 % those induced by 3 mm ATP and 100 μm Bz-ATP, respectively. The P2Z/P2X7 antagonist oxidized ATP (oATP) almost abolished the inward current induced by 3 mm ATP or Bz-ATP, but was ineffective against 100 μm ATP.
  4. Inward currents induced by low ATP concentrations (≤ 100 μm) were generally followed by an almost complete and irreversible desensitization, while those elicited by ATP ≥ 1 mm showed only a partial decline. Interestingly, the inward current induced by 100 μm 2-MeSATP showed a large desensitization, while that induced by Bz-ATP did not.
  5. In voltage-ramp experiments, the 100 μm ATP-induced current exhibited a slight inward rectification more visible at negative potentials, while the 3 mm ATP-induced current did not.
  6. ATP induced a fast and large increase in [Ca2+] that promptly recovered in the continuous presence of low ATP doses, but did not recover in high ATP doses. As with desensitization, the response to Bz-ATP mimicked that of high doses of ATP.
  7. When Ca2+ mobilization due to P2Y receptors was blocked by thapsigargin-induced Ca2+ depletion or by pertussis toxin treatment, 10 μm ATP was still able to induce a Ca2+ transient, which represented the contribution of the Ca2+ influx induced by P2X receptors
  8. In conclusion, the inward currents and a fraction of the Ca2+ transients induced by ATP in microglia are due to at least two ATP-sensitive receptor channel types, whose different properties and sensitivity to ATP may be associated with different functional roles.

Microglial cells, the resident macrophages in the brain, are involved in the protection of the integrity of the CNS but, in some instances, are also involved in establishing or exacerbating a number of pathological conditions (Kreutzberg, 1996; Minghetti & Levi, 1998). They are capable of performing a variety of functions such as secretion of cytokines, eicosanoids and free radicals, presentation of the antigen to T lymphocytes, phagocytosis, migration and proliferation. They are highly responsive to environmental stimuli, one of which may be extracellular ATP.

ATP, first considered as ‘energetic money’ for the intracellular processes, is now known to function also as an extracellular messenger. The following observations support this view: (a) it can be secreted at specific sites by vesicular or granular release or by transport systems, (b) specific purinergic receptors are widely distributed on several mammalian cells, and (c) its extracellular concentration is normally kept low by an efficient system of hydrolytic enzymes (Dubyak & El-Moatassim, 1993). Nevertheless, the cytoplasmic concentration of ATP is in the millimolar range, and cell lysis is believed to cause a sudden and strong build-up of the interstitial ATP concentration in certain pathological conditions (White & Hoehn, 1991).

ATP-sensitive receptor channels form the P2X subfamily comprising, up to now, seven members. The seventh member, only recently recognized as a member of this family and called P2Z/P2X7, shares some features with the other members, but is peculiar in that it allows the passage through the channel or pore of molecules as large as 900 Da. This property is conferred by a long intracellular carboxy-terminal tail, which is responsible for the membrane permeabilization observed at high ATP concentrations (Surprenant et al. 1996). First described in transformed mouse fibroblasts (Rozengurt et al. 1977), the presence of this receptor has been documented in mast cells (Cockcroft & Gomperts, 1979), macrophages (Steinberg et al. 1987), and lymphocytes (Markwardt et al. 1997). In the brain P2Z/P2X7 receptors are present in microglia (Ferrari et al. 1996) and astrocytes (Ballerini et al. 1996). Metabotropic G protein-linked receptors, once named P2T, P2U and P2Y receptors, are now grouped in the P2Y subfamily, which comprises eight receptors.

Microglial cells are known to bear both ligand-gated (P2X) and metabotropic (P2Y) nucleotide receptors (Walz et al. 1993; Haas et al. 1996; Illes et al. 1996). The former are responsible for ATP-induced depolarization and Ca2+ influx, the latter for events such as Ca2+ release from intracellular stores. Studies on microglial P2X receptors have not presented a consistent and conclusive picture, probably due to the heterogeneity of the species and preparations used, the different experimental approaches adopted, the poor specificity of the pharmacological tools available, and, until recently, the lack of an unequivocal P2 receptor classification (Walz et al. 1993; Langosch et al. 1994; Nörenberg et al. 1994; Ferrari et al. 1996; Haas et al. 1996; Chessell et al. 1997). In particular, studies carried out on mouse cell lines using intracellular Ca2+ recording (Ferrari et al. 1996) and patch-clamp methods (Chessell et al. 1997) emphasized the presence of P2Z/P2X7 receptor channels, while patch-clamp studies on rat microglial cells concluded that the fast depolarizing response to ATP was due to P2X receptor channels (Illes et al. 1996). Finally, studies on mouse slice preparations envisaged the presence of both a ‘normal’ P2X receptor channel, and a P2Z/P2X7 ‘pore’ responsible for the ATP-induced permeabilization, but detectable only after a sustained treatment with high doses of ATP (Haas et al. 1996).

In the present investigation carried out on cultured rat microglial cells, we combined patch-clamp and intracellular Ca2+ measurements to clarify whether the fast inward current and the Ca2+ increase caused by ATP are due to the presence of a single or of more than one purinergic receptor channel.


Cell culture

Experiments were carried out on rat microglial cells obtained from the cerebral cortex of 1- to 2-day-old rats as described elsewhere (Levi et al. 1993). All experiments were carried out in accordance with the directives of the Council of the European Communities N. 86/609/CEE. Briefly, after pups were rendered hypothermic and decapitated, the cerebral cortex was dissected out and the meninges were carefully removed. Then the tissue was subjected to mechanical dissociation and enzymatic digestion. Mixed primary cultures were grown for 7-10 days on poly-L-lysine-coated culture flasks in Basal Eagle's Medium (BME) supplemented with 10 % endotoxin-free fetal calf serum (FCS), 2 mM glutamine and 100 μg ml−1 gentamicin (37°C, 5 % CO2). After mild shaking, microglial cells were harvested and plated on uncoated glass coverslips (105 cells cm2). To further improve the purity of microglial cultures, non-adhering cells were removed after 20 min by changing the medium. Cells were used for recording from the first to the third day in secondary culture. Data obtained on different days were pooled, since the ATP-induced inward current did not vary with the age of the secondary culture.


In order to record whole-cell currents the patch-clamp method was employed (Hamill et al. 1981). Glass coverslips bearing microglial cells were placed in a recording chamber on the stage of an Axiovert 35 inverted microscope (Zeiss, Germany). Unless otherwise stated, the bulk solution had the following composition (mM): 140 NaCl, 5 KCl, 2.5 CaCl2, 1 MgCl2, 10 D-glucose, 10 Hepes-NaOH (room temperature, pH 7.4, 290 mosmol l−1). Divalent cation-free extracellular solutions were obtained by omitting Ca2+ and Mg2+. The electrodes were filled with the following solution (mM): 130 KCl, 1 CaCl2, 2 MgCl2, 10 EGTA, 10 Hepes-KOH (pH 7.2, 290 mosmol l−1). In some cases Cs+ replaced K+ in the internal electrode solution. When filled with these solutions, the input resistance of the electrodes ranged from 2 to 4 MΩ. A multi-barrelled apparatus connected to an electronically driven rotary motor (Biologic, France) performed local perfusion of the cell under study. The single barrels had an internal diameter of 300 μm and the localized stream of solution was allowed to flow at a low speed (100 μl min−1) in order to avoid damage to the cell during recording. The rate of change of solution was measured as the shift of the electrode current caused by switching from a normal to a diluted external solution. The 10-90 % change of the external solution measured in that way occurred in 80 ± 4 ms. A detailed description of the patch-clamp set-up has been provided elsewhere (Visentin et al. 1995). The results were expressed as means ±s.e.m.

Intracellular Ca2+ measurement

Optical fluorimetric recordings with fura-2 acetoxymethyl ester (fura-2) were utilized to evaluate the intracellular Ca2+ concentration. Fura-2 stock solutions were obtained by adding 50 μg of fura-2 to 50 μl of 75 % DMSO + 25 % pluronic acid. Cells were bathed for 50 min at room temperature with 5 μl stock solution diluted in 1 ml of extracellular solution, giving a final concentration of 5 μm fura-2. This solution was then removed, replaced with extracellular solution, and the dishes were quickly placed on the microscope stage.

During the experimental session the solution in the dish was continuously replaced by fresh external solution (for the composition see ‘Electrophysiology’). The Ca2+-free external solution had the following composition (mM): 140 NaCl, 5 KCl, 2 MgCl2, 0.5 EGTA, 10 D-glucose, 10 Hepes-NaOH (pH 7.4). The area containing the cells under study was bathed by a control or test solution provided by a local perfusion system consisting of a double-barrelled apparatus. From 7 to 21 cells were analysed at the same time. The sample interval was 2 s. The Ca2+ concentration was expressed as the ratio of fura-2 fluorescence at 340 and 380 nm excitation wavelengths. A Leica DM IRB microscope (Leica, Germany) and a Hamamatsu Argus 50 system (Hamamatsu, Japan) were used to measure fluorescence, and the software package Origin (Microcal, MA, USA) was used for the graphical treatment of data.

Materials and solutions

2-Methylthio-ATP (2-MeSATP), α,β-methylene-ATP (α,β-meATP), β,γ-methylene-ATP (β,γ-meATP), adenosine, ATP, 2′-3′-O-4-benzoylbenzoyl-ATP (Bz-ATP), oxidized ATP (oATP), suramin, apyrase (grade VI), poly-l-lysine and gentamicin were purchased from Sigma. BME and FCS were supplied by Gibco BRL, pertussis toxin by List Biological Labs and fura-2 by Molecular Probes. Thapsigargin was a kind gift from Alomone Labs. A fresh ATP stock solution was made daily by diluting ATP (Na+ salt) in the external solution. All the other chemicals were dissolved in external solution and stored in aliquots at -80°C. In the light of the strongly acidic nature of the phosphate groups and of the possible sensitivity to pH of the P2X receptors, great care was taken to control the pH.


The membrane potential of microglial cells sojourns at two levels, -73 mV and -33 mV, with the most probable level being the most hyperpolarized (Nörenberg et al. 1994; Visentin et al. 1995). When the membrane potential was clamped at a holding level of -70 mV and ATP was applied through a local and fast delivery system, it elicited an inward current in the majority of cells studied. The amplitude of the current varied from cell to cell, independently of cell size and cell membrane capacitance. In a minority of cells, an outward membrane current appeared as ATP was removed. Such outward current was more visible at more positive holding potentials. At -20 mV an outward current followed the inward current, while at 0 mV only the outward component was present. The outward current was never observed when Cs+ replaced K+ in the pipette solution, indicating that K+ was the main charge carrier. We did not further characterize the outward current and focused our attention on the inward current.

Concentration dependence of the ATP-induced inward current

The amplitude of the ATP-induced inward current was concentration dependent (Fig. 1). The lowest effective ATP dose was 1 μm, at which only 40 % of the cells (n = 17) responded with a visible current. Even at 10 mM ATP the response did not reach a maximum; hence we could not calculate the EC50. Interestingly, the curve showed a composite shape, as if the inward current were due to the presence of two receptors with different affinities for ATP. Although we could not calculate the EC50 for each of the two components, we estimated that the inward currents induced by 100 μm and 3 mM ATP could well represent the activity of the high and low affinity receptors, respectively. Therefore, we utilized these concentrations to characterize the different responses to ATP. The inset of Fig. 1 shows the dose-response curve for the range of ATP concentrations at which a putative high affinity receptor would be detectable. We also recorded the effect of ATP in the absence of divalent cations (Mg2+ and Ca2+), in order to increase the concentration of the form ATP4-, which is the agonist of the P2Z/P2X7 receptor. As predicted, under these conditions the amplitude of the ATP-induced inward current increased at all the concentrations and the lowest effective ATP concentration became 100 nM (n = 4), but no tendency to saturation was observed, even at 1 mM ATP.

Figure 1
The dose-response curve of the ATP-induced inward current does not show a simple sigmoidal shape and shifts leftward in the absence of divalent cations

Rank order of potency of P2X receptor agonists

In order to characterize pharmacologically the receptor(s) linked to the inward current response, a number of P2 receptor agonists were tested. Each of them was applied at a concentration of 100 μm, and only one application of an agonist was given to the cell under study (Fig. 2). The P2Z/P2X7 receptor agonist Bz-ATP was the most effective and induced a mean inward current peak of 310 ± 98 pA (n = 22). ATP (n = 149) and 2-MeSATP (n = 12) were much less effective than Bz-ATP and induced similar inward currents (65 ± 6 and 50 ± 6 pA, respectively). Among the methylene-substituted analogues, α,β-meATP caused a small but detectable current (7 ± 2 pA; n = 9), while β,γ-meATP was completely without effect (n = 5). Among the degradation products of ATP, ADP was slightly effective at 100 μm (n = 6), while adenosine was completely ineffective up to a concentration of 1 mM (n = 5). UTP was also ineffective in inducing inward currents (n = 4).

Figure 2
Rank order of potency of P2 receptor agonists

Different antagonistic effects of suramin and oATP on the ATP- and Bz-ATP-induced inward current

In an attempt to discriminate between different inward current components, the P2Z/P2X7 receptor antagonist oATP and the P2 receptor antagonist suramin were also used. Since the oATP interaction with the receptor is a slow process, taking place over hours (Murgia et al. 1993), we treated microglial cultures with 300 μm oATP for 2 h and then exposed them to the potent and specific P2Z/P2X7 agonist Bz-ATP. As expected, the amplitude of the Bz-ATP-induced inward current (310 ± 98 pA) was greatly reduced in cells pretreated with oATP (36 ± 13 pA; n = 6; Fig. 3A).

Figure 3
Antagonistic effect of oATP and suramin on the inward current induced by ATP agonists

The inward current induced by 3 mM ATP was almost totally abolished by the P2Z/P2X7 receptor antagonist oATP, the mean amplitude decreasing from 550 ± 65 pA (n = 59) to 53 ± 10 pA (n = 21), respectively, in untreated and oATP-treated cells. In contrast, the inward current induced by 100 μm ATP (65 ± 6 pA) was not significantly affected by the P2Z/P2X7 antagonist (50 ± 10 pA; n = 8).

The P2 receptor antagonist suramin has been shown not to block the P2Z/P2X7 receptors in the NTW8 mouse microglial cell line (Chessell et al. 1997), and to block them only partially (up to 38 %) in transfected HEK 293 cells (Surprenant et al. 1996) and almost completely in human macrophages (Rassendren et al. 1997). When acutely applied for a few minutes to our cells, 300 μm suramin inhibited by 92 ± 2 % (n = 6) the amplitude of the inward current induced by 100 μm ATP, while with 3 mM ATP the antagonism was only 51 ± 8 % (n = 6). This high concentration of suramin was also able to antagonize by 68 ± 6 % (n = 5) the Bz-ATP-induced inward current (Fig. 3B). In order to avoid the occurrence of desensitization, the agonists were applied for 2 s only, which, in control experiments, did not cause a decrease in the current amplitude.

Desensitization of the ATP-induced inward current

The onset of the inward current was complete within a few hundred milliseconds of the start of the agonist application, and was independent of the concentration and nature of the agonist used. In contrast, long-lasting (30 s) applications of ATP showed the presence of a slow desensitization with a single exponential time constant around 5 s, whose magnitude depended on the concentration (Fig. 4A) and nature of the agonist (see next section). With ATP concentrations up to 300 μm, the mean percentage of desensitizing current was around 75 % of the total current, while with higher ATP concentrations the mean percentage of desensitization decreased to 40 % of the total current (Fig. 4B). In order to study in more detail this different behaviour, we used again the two ATP concentrations of 100 μm and 3 mM. The probability distribution of the percentage of desensitization in different cells was plotted for both ATP concentrations (Fig. 4C). With 3 mM ATP (n = 20) the percentage was equally distributed, while with 100 μm ATP (n = 43) two sub-populations of cells could be detected: a large one, showing almost complete desensitization, and a less abundant population that did not show any desensitization.

Figure 4
Low ATP concentrations induce inward currents showing a more pronounced desensitization compared to high ATP concentrations

We then tested whether we could reproduce such behaviour using the available purinergic agonists. Figure 5A shows that both 100 μm ATP and 2-MeSATP caused a similar desensitization (83 ± 3 %, in the case of 2-MeSATP; n = 12). Figure 5B shows that the P2Z/P2X7 receptor agonist Bz-ATP(middle trace) induced an inward current showing little desinsitization, similar to that induced by 3 mM ATP.

Figure 5
P2X agonists induce inward currents showing a more pronounced desensitization than those induced by the P2Z/P2X7 agonist Bz-ATP

We also wondered if the P2Z/P2X7-specific antagonist oATP could unmask the desensitizing current by blocking the non-desensitizing component carried by P2Z/P2X7 receptor channels. In support of this idea, when 3 mM ATP was applied to cells previously treated with oATP, not only was the inward current amplitude smaller, but the percentage of desensitizing current also changed. This effect was not particularly evident from the mean percentage of desensitization due to 100 μm ATP and 3 mM ATP in oATP-treated cultures (Fig. 5C). However, when the probability distributions of inward currents induced by 3 mM ATP in untreated and oATP-treated cells were plotted, a sub-population of oATP-treated cells showing a large desensitization became clearly visible (Fig. 5D; n = 20). The current shown in Fig. 5B (lower trace) does not represent the mean response to oATP, but only the population of cells in which oATP caused the appearance of desensitization.

Lack of recovery from desensitization and cross-desensitization

When the duration of the application of 100 μm ATP was long enough to provoke nearly full desensitization of the inward current, recovery, if any, from such desensitization was very small (Fig. 6A, left traces). The same happened when the agonist was 2-MeSATP (Fig. 6A, middle traces), α,β-meATP (data not shown) and, in the cells treated with the P2Z/P2X7 antagonist oATP, 3 mM ATP (Fig. 6A, right traces). Within the first 3 min after the end of the first ATP application recovery from desensitization reached a maximum of 20 %, and the current induced by a second ATP application was completely unable to recover, even after a 10 min interval (data not shown; n = 6). In the light of this observation we wanted to ascertain whether the ATP-induced current that we recorded was not underestimated because of desensitization due to ATP released by cells in culture. We therefore treated cells with the ATPase apyrase (30 U ml−1) for 2 h before recording. The mean amplitude of the current induced by 100 μm ATP after the treatment (86 ± 19 pA; n = 13) did not differ significantly from that obtained during normal conditions (65 ± 6 pA; see Fig. 1).

Figure 6
Inward currents induced by P2X receptor activation do not recover from desensitization

To further support our hypothesis of the co-existence of different P2X receptors, we looked for the presence of cross-desensitization. For this purpose a 20 s long application of 100 μm ATP was followed by the application of either 100 μm 2-MeSATP or 3 mM ATP. It was clearly evident that the current elicited by 2-MeSATP was already desensitized, while 3 mM ATP induced an inward current whose amplitude was unaffected by the previous application of 100 μm ATP (Fig. 6B).

Voltage dependence: reversal potential and inward rectification

In order to confirm further the co-existence of two different P2X receptors in microglia, we analysed the voltage dependence of the inward current induced by different concentrations of ATP (100 μm and 3 mM) and by the specific P2Z/P2X7 receptor agonist Bz-ATP. Voltage ramps were applied 20-30 s before switching to ATP, at the peak and steady state of the inward current. Digital subtraction of the current recorded before application of 100 μm ATP from the peak and steady-state (20 s after peak) inward currents revealed the ‘pure’ ATP current at the two application time points (Fig. 7A). Both traces crossed the voltage axis near 0 mV, showing that the reversal potential was close to 0 mV and did not change throughout the ATP application. Moreover, a slight but consistent inward rectification, more evident at negative potentials, was observed in the peak current trace. The whole-cell conductance induced by 100 μm ATP was 1.54 ± 0.2 nS at -70 mV, decreasing to 60 % of this value at +50 mV (n = 8).

Figure 7
Only inward currents induced by low ATP concentrations show inward rectification

The same experimental procedure was used to study the voltage dependence of the 3 mM ATP-induced inward current. In this case also, the reversal potential was close to 0 mV at the peak (Fig. 7B) and steady state (not shown). On the other hand, almost no inward rectification was caused by this ATP concentration. The whole-cell conductance induced by 3 mM ATP was 4.3 ± 0.8 nS at -70 mV, decreasing to only 90 % of this value at +50 mV (n = 4).

Membrane potential: correlation with desensitization

We then proceeded to characterize the influence of ATP on the membrane potential and to draw a correlation between the effect of ATP on this parameter and the different desensitization observed at low and high ATP concentrations. In order to do so we employed the current-clamp mode of the patch-clamp technique.

ATP application from 1 μm to 3 mM caused a fast depolarization of the membrane potential. With up to 10 μm ATP, the depolarization was followed by a slow hyperpolarization in the continuous presence of the agonist (Fig. 8A; n = 18). With an ATP concentration higher than 100 μm, the membrane potential remained close to 0 mV until the agonist was removed (Fig. 8C; n = 5). With 100 μm ATP, the membrane potential change could follow either of the two patterns described above (n = 5). Such behaviour correlated closely with the different desensitization properties demonstrated during the inward current recording. Finally, we sought to test whether the lack of recovery from desensitization of the inward current could influence the effect of ATP on membrane potential. In keeping with this hypothesis, when a series of pulses of ATP at a low concentration was applied to a cell, the depolarization decreased with each pulse (Fig. 8B; n = 16). In contrast, when a high dose of ATP was used, the amplitude of the effect on the membrane potential remained unchanged throughout the series of ATP pulses (Fig. 8D; n = 5).

Figure 8
ATP-induced depolarization: correlation with the inward current desensitization

Intracellular Ca2+ changes induced by P2 receptor activation

P2X receptor activation is known to cause the opening of channels that are also permeable to Ca2+ and thus to provoke cytosolic Ca2+ build-up. In the light of the important implications of this event, we examined whether the activation of possibly different P2X receptors could induce different Ca2+ responses. Microglial cells studied in a normal Ca2+-containing solution showed great sensitivity to fast applications of ATP. A concentration as low as 1 μm was able to elicit a Ca2+ response in 64 % of the cells studied (n = 28). The rate of change of Ca2+ was fairly fast. The duration of the raising phase of the Ca2+ accumulation ranged from 5 to 10 s, and did not depend markedly on the ATP concentration. In order to confirm the sensitivity of the microglial P2X receptors to agonists shown to be able to induce an inward current, we tested their ability to elicit a Ca2+ increase. In accordance with patch-clamp studies, the ATP analogues 2-MeSATP (n = 27) and α,β-meATP (n = 49), and the P2Z/P2X7 agonist Bz-ATP (n = 11) were able to induce a Ca2+ build-up (Fig. 9). However, while the effect of Bz-ATP can be considered as confirmation of the presence of P2Z/P2X7 receptors, we cannot assume that the other agonists only cause Ca2+ influx induced by P2X receptors; they may also induce Ca2+ release via P2Y receptor activation. In fact, an increase in Ca2+ was also induced by β,γ-meATP (n = 51) and UTP (n = 98), both of which were unable to elicit inward currents in patch-clamp experiments. The contribution of P2Y receptors was also shown in experiments carried out in Ca2+-free solution, in which ATP (Fig. 10A), α,β-meATP and UTP were all able to elicit Ca2+ transients (data not shown).

Figure 10
Contributions of Ca2+ influx and Ca2+ release to the Ca2+ transient induced by P2 receptor activation
Figure 9
Rise of intracellular Ca2+ induced by ATP and ATP analogues

In order to ascertain the contribution of P2X receptors, two strategies were employed. The first consisted in recording the effect of 10 μm ATP after having depleted endoplasmic reticulum Ca2+ stores with the Ca2+-ATPase inhibitor thapsigargin (Fig. 10B). After treatment sufficient to empty the Ca2+ stores, as judged by the lack of effect of UTP (100 μm), ATP was still able to elicit an increase in intracellular Ca2+ (n = 34). The second strategy was based on the possible sensitivity to pertussis toxin of the P2Y response (Fig. 10C). When microglial cultures were treated with the toxin, 80 % of the cells responded to 10 μm ATP in a normal Ca2+-containing solution (n = 76), while none of them responded to ATP in a Ca2+-free solution (n = 54). The Ca2+ increase observed in these conditions represents the contribution of the Ca2+ influx induced by P2X receptor activation.

Intracellular Ca2+ changes: correlation with desensitization

As for membrane voltage, a correlation with the inward current desensitization could also be found for the Ca2+ transient. Applications of low ATP concentrations (1-10 μm) lasting 30 s caused a Ca2+ enhancement followed, in the continuous presence of ATP, by a recovery toward the basal Ca2+ level. In contrast, 30 s applications of high concentrations of ATP (> 100 μm) brought about an intracellular Ca2+ build-up that reached a plateau lasting longer than the ATP application (Fig. 11A and C). As for membrane voltage, a certain amount of variability in the time course of the Ca2+ response to ATP was also present, especially at the intermediate concentration of 100 μm (not shown).

Figure 11
ATP-induced Ca2+ rise: correlation with desensitization of the inward current

In order to test whether lack of recovery from desensitization could influence the effect of repeated applications of ATP on intracellular Ca2+, 30 s pulses of ATP were applied to the cells every 3 min. In this case also, ATP concentration dependence was evident. The amplitude of the Ca2+ transients induced by low ATP concentrations (1-10 μm) decreased throughout the series of ATP pulses, while that induced by high ATP concentrations did not (Fig. 11B and D). When 100 μm Bz-ATP was used, it induced Ca2+ transients that lasted longer than the agonist application and did not decrease in amplitude throughout a series of repeated stimulations (data not shown).


The aim of the present study was to understand whether the inward current due to stimulation of P2X receptors in rat microglia was due to one or more types of such receptors. In order to do so we performed pharmacological and kinetic analyses and a combination of the two. We also correlated the results obtained using whole-cell voltage clamp with those obtained using current clamp and intracellular Ca2+ measurement.

Pharmacological evidence for the existence of two types of P2X receptors

Our pharmacological approach led to the following observations: (a) ATP induced inward currents with a dose-response relationship exhibiting a composite shape; (b) the omission of divalent cations from the external solution greatly augmented the amplitude of the inward current; (c) the rank order of potency was:

equation image

while β,γ-meATP and UTP were unable to induce inward currents; (d) the P2Z/P2X7 antagonist oATP was more efficient in blocking Bz-ATP- and 3 mM ATP-induced inward current than 100 μm ATP-induced inward current, while the P2 antagonist suramin blocked 100 μm ATP- induced responses more efficiently than the 3 mM ATP- and Bz-ATP-induced responses.

The composite shape of the ATP dose-response relationship obtained in the presence of divalent cations suggests the existence of high- and low-affinity P2X receptors in microglia. The same type of dose-response relationship was obtained when measuring ATP-induced intracellular Ca2+ transients in the microglial cell lines N9 and N13 (Ferrari et al. 1996). In that case, the authors interpreted the dose-response relationship as being due to both Ca2+ release from intracellular stores and Ca2+ influx, and considered P2Z/P2X7 receptors as the main mediators of Ca2+ influx. Yet they did not discard the possibility that ionotropic ATP receptors other than the P2Z/P2X7 are present on microglia.

Our pharmacological studies support the idea that the low-affinity receptor, which responds to millimolar concentrations of ATP, can be identified with the P2Z/P2X7 subtype. This receptor is known to be sensitive only to ATP4-, the concentration of which is increased by lowering the concentration of divalent cations. Hence, the large increase in the response to ATP when divalent cations were omitted from the extracellular solution is an indication of the presence of P2Z/P2X7 receptors. As a further confirmation of the presence of these receptors we showed that the P2Z/P2X7 receptor agonist Bz-ATP induced large inward currents and that the P2Z/P2X7-specific antagonist oATP was more effective with Bz-ATP or a high dose of ATP (3 mM) than with a low ATP dose (100 μm).

The expression of a high affinity receptor, in addition to the P2Z/P2X7 receptor, is consistent with the ability of low doses of ATP (in some cases as low as 1 μm) to elicit a response. Such concentrations would not be expected to be active on P2Z/P2X7 receptors.

More insidious is the issue regarding the specificity of the other P2 receptor agonists tested. 2-MeSATP and α,β-meATP are known to elicit both P2X and P2Y receptor-related effects, and within the family of P2X receptors their efficacy depends on the specific receptor subtype involved (Bhagwat & Williams, 1997). Interestingly, studies on P2Z/P2X7 receptors showed the lack of effect of α,β-meATP and the greater potency of ATP than 2-MeSATP (Soltoff et al. 1992; Surprenant et al. 1996; Chessell et al. 1997). The low but consistent response of our cells to α,β-meATP and the similar effectiveness of ATP and 2-MeSATP are consistent with the presence of receptors other than the P2Z/P2X7 receptors in rat microglia. This is also in keeping with a previous study (Illes et al. 1996) in which the authors interpreted the 2-MeSATP-induced inward current in microglial cells as being due to the presence of P2X receptors. The low effectiveness of oATP with 100 μm ATP also suggests that at this ATP concentration the response is due to P2X receptors other than the P2X7/P2Z type. Finally, the (slightly) higher sensitivity to suramin of the responses elicited by a low concentration of ATP (100 μm) than that of the responses elicited by 3 mM ATP and 100 μm Bz-ATP is also consistent with the presence of receptors other than P2X7/P2Z.

Kinetic evidence for the existence of two types of P2X receptors

According to the literature, the presence and the rate of desensitization of the ATP-induced inward currents differ in different cells and sometimes also within the same cell type (Zhou & Galligan, 1996; Cook & McCleskey, 1997). This behaviour is due to the expression of different subunits, which can participate in the assembling of homomeric or heteromultimeric receptor channels with different features (Lewis et al. 1995; Thomas et al. 1998). In particular, the P2Z/P2X7 receptor is known to desensitize very little and very slowly.

In our study, the percentage of the inward current undergoing desensitization was inversely related to the concentration of ATP used (Fig. 5A and B), being minimal with 3 mM ATP. Moreover, the desensitization induced by the two ATP analogues 2-MeSATP and Bz-ATP resembled that elicited by low (100 μm) and high (3 mM) ATP concentrations, respectively. Considering that the former analogue has been described as a P2X agonist in microglia (Illes et al. 1996), and the latter as a specific P2Z/P2X7 agonist, this observation also indicates the presence of two different P2X receptors in rat microglia.

Although following a sufficiently long application of a low concentration of ATP (100 μm) most cells were fully desensitized and were unable to recover, a sub-population of cells showing no desensitization was also detected. Two possible interpretations can be proposed: these cells expressed only non-desensitizing receptors such as P2Z/P2X7 receptors, or they were already desensitized by previous contact with the nucleotide (applied to or released by a neighbouring cell). The presence of a highly desensitizing current during the application of 3 mM ATP in oATP-treated cells (Fig. 5B) was most probably due to the unmasking of the desensitizing component, once the P2Z/P2X7 receptors were blocked by oATP. But even if, in a sub-population of cells, the desensitization was equal to or above 70 % (Fig. 5D), the mean percentage of desensitization did not change significantly after treatment with oATP (Fig. 5C): this is because of the presence of a sub-population of cells which did not show significant desensitization, perhaps due to lack of expression of P2X-desensitizing channels. Finally, the observation that no recovery from desensitization was present in cells treated with 2-MeSATP and in cells pre-treated with oATP when 3 mM ATP was used as an agonist (Fig. 6A) supports the concept that desensitization has to be ascribed in all cases to P2X channels. This concept is strengthened by cross-desensitization studies.

Functional aspects of the ATP response

After obtaining pharmacological and kinetic evidence in favour of the presence of two different P2X receptors in microglia, P2Z/P2X7 receptors, sensitive to high doses of ATP, and undetermined P2X receptors, sensitive to low doses of ATP, we analysed two functional aspects of ATP stimulation, membrane depolarization and an increase in intracellular Ca2+ concentration. These experiments must be interpreted with caution, because the response of microglia to ATP also involves metabotropic P2Y receptors, responsible for Ca2+ release from intracellular stores (Greenberg et al. 1988) and for the slow appearance of a K+ conductance induced through mechanisms involving intracellular Ca2+ and G proteins (Nörenberg et al. 1997). Both these effects can interfere with the effects on membrane potential and intracellular Ca2+ levels.

Our study suggests that at low ATP concentrations the desensitization of the inward current and the rise of a slow K+ conductance could combine to repolarize the membrane after the fast depolarization induced by P2X receptor activation.

As to changes in intracellular Ca2+, our results point to a dual nature of the Ca2+ transient induced by ATP. A P2Y receptor-mediated Ca2+ release from intracellular stores is supported by the persistence of Ca+2 transients in experiments in Ca2+-free media, and further confirmed by the observation that agonists unable to elicit inward currents, such as UTP and β,γ-meATP, were able to induce Ca2+ transients. On the other hand, a Ca2+ influx due to P2X receptors was shown by the persistence of an increase in intracellular Ca2+ when exposure to a low ATP concentration occurred after depleting endoplasmic reticulum Ca2+ stores with thapsigargin or after blocking the G protein-dependent Ca2+ release with pertussis toxin.

As it is clear from the above considerations, our data on membrane potential and intracellular Ca2+ cannot be used as additional evidence in favour of the existence of more than one P2X receptor in microglia. However, they help clarify the overall picture of the differential effects of low and high ATP concentrations in microglia.

As mentioned in the Introduction, there is no agreement in the literature as to the type(s) of P2X receptor(s) expressed by microglial cells. The matter is not irrelevant. On one hand the P2Z/P2X7 receptor shows the peculiar ability to provoke membrane permeabilization to molecules up to 900 Da, to induce cell fusion (Chiozzi et al. 1995), interleukin-1β release (Ferrari et al. 1997b) and apoptotic cell death (Ferrari et al. 1997a). On the other hand, the features of the other P2X receptor that we propose is expressed in microglia would allow other functions to be accomplished. The small amplitude of the response, the fairly fast decay due to desensitization, the failure to respond to successive applications of ATP and the Ca2+ permeability of the channel make this receptor a good candidate for a fine signalling mechanism in which the specific ion channel pattern of a given cell can also play an important role by influencing the effect of ATP.

It is known that the sensitivity of microglia to ATP varies in the course of development and can be altered by cytokines. Interestingly, interferon-γ was shown to increase the sensitivity of macrophages to ATP by enhancing the expression of P2Z/P2X7 receptors (Blanchard et al. 1991; Falzoni et al. 1995) and prostaglandin E2 to attenuate this effect via a cAMP-mediated mechanism (Humphreys & Dubyak, 1998).

In conclusion, the presence of receptors with different affinities would allow microglial cells to switch on completely different responses depending on the concentration of ATP in the vicinity of the cell bearing both receptor channels. Moreover, developmental and environmental conditions could influence the reactivity of microglia to ATP by differentially turning on the expression of one of the P2X receptors.

Finally, can the subtype of P2X receptor described in the present study be identified with one of the known P2X receptor subtypes? Based on kinetic and pharmacological considerations, the microglial receptor could not belong to the homomeric P2X1 or P2X3 group of receptors, which exhibit quite fast desensitization (in the order of hundreds of milliseconds) and respond to both ATP and α,β-meATP with an EC50 in the 1 μm range. Among the other receptor subtypes, P2X4 is the most probable candidate, because of the slow desensitization, and the small but measurable sensitivity to α,β-meATP. According to the observations of Le and colleagues (Le et al. 1998), even a heteromeric P2X4-P2X6 receptor could be a candidate, in view of its increased sensitivity to α,β-meATP, 2-MeSATP and suramin, which would make it functionally similar to the microglial receptor channel described here.


This work was supported by the Research Project on Multiple Sclerosis of the Istituto Superiore di Sanità and by the AIDS Project of the Italian Ministry of Health (grant number 10/A/H to G.L.).


  • Ballerini P, Rathbone MP, Di Iorio P, Renzetti A, Giuliani P, D'Alimonte I, Trubiani O, Caciagli F, Ciccarelli R. Rat astroglial P2Z (P2X7) receptors regulate intracellular calcium and purine release. NeuroReport. 1996;7:2533–2537. [PubMed]
  • Bhagwat SS, Williams M. P2 purine and pyrimidine receptors: emerging superfamilies of G-protein-coupled and ligand-gated ion channel receptors. European Journal of Medicinal Chemistry. 1997;32:183–193.
  • Blanchard DK, McMillen S, Djeu JY. IFN-γ enhances sensitivity of human macrophages to extracellular ATP-mediated lysis. Journal of Immunology. 1991;147:2579–2585. [PubMed]
  • Chessell IP, Michel AD, Humphrey PPA. Properties of the pore-forming P2X7 purinoceptor in mouse NTW8 microglial cells. British Journal of Pharmacology. 1997;121:1429–1437. [PMC free article] [PubMed]
  • Chiozzi P, Sanz JM, Ferrari D, Falzoni S, Aleotti A, Buell GN, Collo G, Di Virgilio F. Spontaneous cell fusion in macrophage cultures expressing high levels of the P2Z/P2X7 receptor. Journal of Cell Biology. 1995;138:697–706. [PMC free article] [PubMed]
  • Cockcroft S, Gomperts BD. ATP induces nucleotide permeability in rat mast cells. Nature. 1979;279:541–542. [PubMed]
  • Cook SP, McCleskey EW. Desensitization, recovery and Ca2+-dependent modulation of ATP-gated P2X receptors in nociceptors. Neuropharmacology. 1997;36:1303–1308. [PubMed]
  • Dubyak GR, El-Moatassim C. Signal transduction via P2-purinergic receptors for extracellular ATP and other nucleotides. American Journal of Physiology. 1993;265:C577–606. [PubMed]
  • Falzoni S, Munerati M, Ferrari D, Spisani S, Moretti S, Di Virgilio F. The purinergic P2Z receptor of human macrophage cells. Characterization and possible physiological role. Journal of Clinical Investigation. 1995;95:1207–1216. [PMC free article] [PubMed]
  • Ferrari D, Chiozzi P, Falzoni S, Dal Susino M, Collo G, Buell G, Di Virgilio F. ATP-mediated cytotoxicity in microglial cells. Neuropharmacology. 1997a;36:1295–1301. [PubMed]
  • Ferrari D, Chiozzi P, Falzoni S, Hanau S, Di Virgilio F. Purinergic modulation of interleukin-1β release from microglial cells stimulated with bacterial endotoxin. Journal of Experimental Medicine. 1997b;185:579–582. [PMC free article] [PubMed]
  • Ferrari D, Villalba M, Chiozzi P, Falzoni S, Ricciardi-Castagnoli P, Di Virgilio F. Mouse microglial cells express a plasma membrane pore gated by extracellular ATP. Journal of Immunology. 1996;156:1531–1539. [PubMed]
  • Greenberg S, Di Virgilio F, Steinberg TH, Silverstein SC. Extracellular nucleotides mediate Ca2+ fluxes in J774 macrophages by two distinct mechanisms. Journal of Biological Chemistry. 1988;263:10337–10343. [PubMed]
  • Haas S, Brockhaus J, Verkhratsky A, Kettenmann H. ATP-induced membrane currents in ameboid microglia acutely isolated from mouse brain slices. Neuroscience. 1996;75:257–261. [PubMed]
  • Hamill OP, Marty A, Neher E, Sakmann B, Sigworth FJ. Improved patch-clamp techniques for high-resolution current recording from cells and cell-free membrane patches. Pflügers Archiv. 1981;391:85–100. [PubMed]
  • Humphreys BD, Dubyak GR. Modulation of P2X7 nucleotide receptor expression by pro- and anti-inflammatory stimuli in THP-1 monocytes. Journal of Leukocyte Biology. 1998;64:265–273. [PubMed]
  • Illes P, Nörenberg W, Gebicke-Haerter PJ. Molecular mechanisms of microglial activation. B. Voltage- and purinoceptor-operated channels in microglia. Neurochemistry International. 1996;29:13–24. [PubMed]
  • Kreutzberg GW. Microglia: a sensor for pathological events in the CNS. Trends in Neurosciences. 1996;19:312–318. [PubMed]
  • Langosch JM, Gebicke-Haerter PJ, Nörenberg W, Illes P. Characterization and transduction mechanisms of purinoceptors in activated rat microglia. British Journal of Pharmacology. 1994;113:29–34. [PMC free article] [PubMed]
  • Le KT, Babinski K, Seguela P. Central P2X4 and P2X6 channel subunits coassemble into a novel heteromeric ATP receptor. Journal of Neuroscience. 1998;18:7152–7159. [PubMed]
  • Levi G, Patrizio M, Bernardo A, Petrucci T, Agresti C. Human immunodeficiency virus coat protein gp120 inhibits the β-adrenergic regulation of astroglial and microglial functions. Proceedings of the National Academy of Sciences of the USA. 1993;90:1541–1545. [PMC free article] [PubMed]
  • Lewis C, Neidhart S, Holy C, North RA, Buel G, Surprenant A. Coexpression of P2X2 and P2X3 receptor subunits can account for ATP-gated currents in sensory neurons. Nature. 1995;377:432–435. [PubMed]
  • Markwardt F, Löhn M, Böhm T, Klapperstück M. Purinoceptor-operated cationic channels in human B lymphocytes. The Journal of Physiology. 1997;498:143–151. [PMC free article] [PubMed]
  • Minghetti L, Levi G. Microglia as effector cells in brain damage and repair: focus on prostanoids and nitric oxide. Progress in Neurobiology. 1998;54:99–125. [PubMed]
  • Murgia M, Hanau S, Pizzo P, Rippa M, Di Virgilio F. Oxidized ATP. An irreversible inhibitor of the macrophage purinergic P2Z receptor. Journal of Biological Chemistry. 1993;268:8199–8203. [PubMed]
  • Nörenberg W, Cordes A, Blöhbaum G, Fröhlich R, Illes P. Coexistance of purino- and pyrimidinoceptors on activated rat microglial cells. British Journal of Pharmacology. 1997;121:1087–1098. [PMC free article] [PubMed]
  • Nörenberg W, Gebicke-Haerter PJ, Illes P. Voltage-dependent potassium channels in activated rat microglia. The Journal of Physiology. 1994;475:15–32. [PMC free article] [PubMed]
  • Nörenberg W, Langosch JM, Gebicke-Haerter PJ, Illes P. Characterization and possible function of adenosine 5′-triphosphate receptors in activated rat microglia. British Journal of Pharmacology. 1994;111:942–950. [PMC free article] [PubMed]
  • Rassendren F, Buell GN, Virginio C, Collo G, North RA, Surprenant A. The permeabilizing ATP receptor, P2X7. Cloning and expression of a human cDNA. Journal of Biological Chemistry. 1997;272:5482–5486. [PubMed]
  • Rozengurt E, Heppel LA, Friedberg I. Effect of exogenous ATP on the permeability properties of transformed cultures of mouse cell lines. Journal of Biological Chemistry. 1977;252:4584–4590. [PubMed]
  • Soltoff SP, McMillian MK, Talamo BR. ATP activates a cation-permeable pathway in rat parotid acinar cells. American Journal of Physiology. 1992;262:C934–940. [PubMed]
  • Steinberg TH, Newman AS, Swanson JA, Silverstein SC. ATP4- permeabilizes the plasma membrane of mouse macrophages to fluorescent dyes. Journal of Biological Chemistry. 1987;262:8884–8888. [PubMed]
  • Surprenant A, Rassendren F, Kawashima E, North RA, Buell G. The cytolytic P2Z receptor for extracellular ATP identified as a P2X receptor (P2X7) Science. 1996;272:735–738. [PubMed]
  • Thomas S, Virginio C, North RA, Surprenant A. The antagonist trinitrophenyl-ATP reveals co-existance of distinct P2X receptor channels in rat nodose neurones. The Journal of Physiology. 1998;509:411–417. [PMC free article] [PubMed]
  • Visentin S, Agresti C, Patrizio M, Levi G. Ion channels in rat microglia and their different sensitivity to lipopolysaccharide and interferon-γ Journal of Neuroscience Research. 1995;42:439–451. [PubMed]
  • Walz W, Ilschner S, Ohlemeyer C, Banati R, Kettenmann H. Extracellular ATP activates a cation conductance and a K+ conductance in cultured microglial cells from mouse brain. Journal of Neuroscience. 1993;13:4403–4411. [PubMed]
  • White JS, Hoehn K. Release of adenosine and ATP from nervous tissue. In: Stone TW, editor. Adenosine in the Nervous System. London: Academic Press; 1991. pp. 173–195.
  • Zhou X, Galligan JJ. P2X purinoceptors in cultured myenteric neurons of guinea-pig small intestine. The Journal of Physiology. 1996;496:719–729. [PMC free article] [PubMed]

Articles from The Journal of Physiology are provided here courtesy of The Physiological Society
PubReader format: click here to try


Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...


Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...