Adenosine modulates hypoxia-induced responses in rat PC12 cells via the A_2A receptor

Abstract

1. The present study was undertaken to determine the role of adenosine in mediating the cellular responses to hypoxia in rat phaeochromocytoma (PC12) cells, an oxygen-sensitive clonal cell line.

2. Reverse transcriptase polymerase chain reaction studies revealed that PC12 cells express adenosine deaminase (the first catalysing enzyme of adenosine degradation) and the A_2A and A_2B adenosine receptors, but not the A₁ or A₃ adenosine receptors.

3. Whole-cell current- and voltage-clamp experiments showed that adenosine attenuated the hypoxia-induced membrane depolarization. The hypoxia-induced suppression of the voltage-sensitive potassium current (I_K(V)) was markedly reduced by adenosine. Furthermore, extracellularly applied adenosine increased the peak amplitudes of I_K(V) in a concentration-dependent manner. This increase was blocked by pretreatment not only with a non-specific adenosine receptor antagonist, 8-phenyltheophylline (8-PT), but also with a selective A_2A receptor antagonist, ZM241385.

4. Ca²⁺ imaging studies using fura-2 acetoxymethyl ester (fura-2 AM) revealed that the increase in intracellular free Ca²⁺ during hypoxic exposure was attenuated significantly by adenosine. Voltage-clamp studies showed that adenosine inhibited the voltage-dependent Ca²⁺ currents (I_Ca) in a concentration-dependent fashion. This inhibition was also abolished by both 8-PT and ZM241385.

5. The modulation of both I_K(V) and I_Ca by adenosine was prevented by intracellular application of an inhibitor of protein kinase A (PKA), PKA inhibitor fragment (6–22) amide. In addition, the effect of adenosine on either I_K(V) or I_Ca was absent in PKA-deficient PC12 cells.

6. These results indicate that the modulatory effects of adenosine on the hypoxia-induced membrane responses of PC12 cells are likely to be mediated via activation of the A_2A receptor, and that the PKA pathway is required for these modulatory actions. We propose that this modulation serves to regulate membrane excitability in PC12 cells and possibly other oxygen-sensitive cells during hypoxia.

Although the mechanism of hypoxic chemotransduction is still a matter for discussion, a generally accepted concept is that closure of oxygen-sensitive K⁺ channels could be an initial trigger for the hypoxia-induced cellular responses (Gonzalez, Almaraz, Obesco & Rigual, 1994; Lopez-Barneo, 1996). Our laboratory has previously shown that rat phaeochromocytoma (PC12) cells, the neoplastic counter-part of adrenal chromaffin cells (Greene & Tischler, 1976), depolarize during hypoxia by inhibition of an O₂-sensitive potassium (K⁺) channel (Zhu, Conforti, Czyzyk-Krzeska & Millhorn, 1996). The membrane depolarization in PC12 cells during hypoxia leads to activation of voltage-dependent calcium (Ca²⁺) channels, Ca²⁺ entry and an increase in cytosolic Ca²⁺, which are essential for regulation of a number of O₂-responsive genes and the release of neurotransmitters (Raymond & Millhorn, 1997; Zhu, Conforti & Millhorn, 1997). These responses to hypoxia in PC12 cells are essentially similar to those reported in the type I cells of the mammalian carotid body, which is a major oxygen-sensing organ that is responsible for mediating the hyperventilatory response during hypoxia (Buckler & Vaughan-Jones, 1994). Both carotid body type I cells and PC12 cells originate from neural crest and both synthesize dopamine as their major neurotransmitter.

Adenosine is an endogenous metabolite of ATP which is produced and released in various tissues in response to a number of physiological and pathological conditions, including hypoxia (Winn, Rubio & Berne, 1981). In recent years it has been shown that adenosine mediates a number of different cellular activities via specific receptors (Dalziel & Westfall, 1994; Fredholm, 1995). The four cloned adenosine receptors are classified as A₁, A_2A, A_2B and A₃ (Palmer & Stiles, 1994; Olah & Stiles, 1995). The A₁ and A₃ receptors are generally coupled to G_i protein and mediate inhibition of adenylate cyclase activity. The A₂ receptor family, A_2A and A_2B receptors, are almost always coupled to G_s protein, which stimulates adenylate cyclase activity. The A₁ and A₂ receptors appear to have distinct roles in protecting against the inadequate tissue oxygenation during hypoxia (Bruns, 1990). A₁ responses generally bring about a decrease in the amount of metabolic activity, while A₂ responses usually increase oxygen delivery by enhancing blood flow. Therefore, adenosine could be involved in the regulation of the cellular response to hypoxia. However, the mechanisms by which this occurs remain unknown.

The present study was undertaken to determine if adenosine modulates the basic ionic response to hypoxia in PC12 cells. We confirm earlier reports that PC12 cells express the A_2A and A_2B receptor subtypes but not the A₁ or A₃ subtypes (Hide, Padgett, Jacobson & Daly, 1992; van der Ploeg, Ahlberg, Parkinson, Olsson & Fredholm, 1996). Importantly, we found that extracellularly applied adenosine modulates the effect of hypoxia on membrane depolarization by increasing the voltage-sensitive K⁺ current and reducing the voltage-dependent Ca²⁺ current, which leads to decreased membrane depolarization and decreased intracellular free Ca²⁺ during hypoxia in PC12 cells. These modulatory actions of adenosine were blocked by a selective A_2A receptor antagonist, ZM241385, and by the protein kinase A inhibitor (PKI). Modulatory actions of adenosine were absent in a PKA mutant PC12 cell line. To our knowledge, this is the first report which describes the role of adenosine in modulating the cellular response during hypoxia in an O₂-sensitive cell. We propose that adenosine, via the A_2A receptor, plays a critical role in regulating membrane excitability and cytosolic Ca²⁺ levels in PC12 cells and other O₂-sensitive cells during hypoxia.

METHODS

Cell culture

PC12 cells were obtained from American Tissue Culture Collection (Rockville, MD, USA), and grown in Dulbecco's modified Eagle's medium-Ham's F-12 (DMEM-F-12; Gibco) containing 15 mm Hepes buffer, 2 mm L-glutamine, 10 % fetal bovine serum and penicillin-streptomycin (100 u ml⁻¹ and 100 μg ml⁻¹, respectively) in an incubator in which the environment (21 % O₂ and 5 % CO₂ (remainder N₂); 37°C) was strictly maintained. The medium was changed twice a week. PKA-deficient cells (A123.7; Ginty, Glowacka, DeFranco & Wagner, 1991) were grown in Dulbecco's modified Eagle's medium with high glucose (DMEM-H) containing 15 mm Hepes, 10 % fetal bovine serum, 5 % horse serum and 100 μg ml⁻¹ gentamicin, in an environment of 21 % O₂ and 10 % CO₂, at 37°C.

Reverse transcriptase-polymerase chain reaction (RT-PCR)

Cytoplasmic RNA was isolated from PC12 cells using TRI-REAGENT (Molecular Research Center, Cincinnati, OH, USA). RT-PCR was performed using the GeneAmpli kit (Perkin-Elmer, Norwalk, CT, USA) according to the manufacturer's instructions. Briefly, for the reverse transcriptase reaction, 1 μg of purified total RNA was incubated in the presence of: 2.5 μm oligo(dT) primer (16mer); 1 mm deoxynucleotide triphosphates; 1 U RNase inhibitor; and 2.5 U murine leukaemia virus (MuLV) reverse transcriptase, denatured at 85°C for 5 min, and incubated at 42°C for 15 min to allow the reaction to proceed. The reaction was terminated by heating to 95°C for 5 min and then the sample was maintained at 5°C for 5 min. The primers for rat A₁, A_2A, A_2B and A₃ adenosine receptors, and for mouse adenosine deaminase were as follows: A₁ receptor, 5′-CGGCAGCACCCAGACGAAGA-3′ and 5′-CCCACCATGCCGCCCTACAT-3′ (the predicted length of the amplified DNA fragment should be 579 bp); A_2A receptor, 5′-TTCAAAGTGGGAGCCACGCA-3′ and 5′-ATGGGC TCCTCGGTGTACATC-3′ (predicted fragment length, 1320 bp); A_2B receptor, 5′-GCCTCGAGTGCTTTACAGACCCCC-3′ and 5′-GAAAGTTGACTGTCCCCCGGCCTG-3′ (predicted frag-ment length, 885 bp); A₃ receptor, 5′-CACATCCTGCTGAAG AAGCAACAG-3′ and 5′-GAGCTGGCTCTTTATCTGTCA TGG-3′ (predicted fragment length, 1045 bp); adenosine deaminase, 5′-AAGGTCCGGTCCATTCTGTG-3′ and 5′-AGG GGGTCGTCTGTGTTGAG-3′ (predicted fragment length, 455 bp).

DNA amplification was accomplished in the presence of 1.5 mm MgCl₂, 1 × reaction buffer and 2.5 U AmpliTaq DNA polymerase. The PCR conditions were as follows. Denaturation for 2 min at 94°C, followed by thirty-five cycles consisting of: 90 s at 94°C; 1 min at 50°C (for the A₁ and A_2A receptors), or 60°C (for the A_2B and A₃ receptors), or 52°C (for adenosine deaminase); and 2 min at 72°C (for the A_2A receptor), or 90 s at 72°C (for the A₁ receptor, A_2B receptor, A₃ receptor and adenosine deaminase). The samples were then kept at 72°C for 7 min. The products of RT-PCR were analysed by electrophoresis on 1 % agarose gels and identity was verified by sequence analysis.

Patch-clamp recordings

The methods for patch-clamp recordings were essentially the same as those described previously (Zhu et al. 1996). The cells were plated on coverslips and placed in a perfusion chamber (volume, 200–400 μl) mounted on the stage of an inverted interference microscope (ITM-2, Olympus, Japan) and constantly perfused with the recording solution at a flow rate of 2–3 ml min⁻¹. Voltage-dependent potassium current (I_K(V)) and calcium current (I_Ca) were recorded in the conventional whole-cell voltage-clamp mode (Hamill, Marty, Neher, Sakmann & Sigworth, 1981) by using an Axopatch-200A amplifier (Axon Instruments). The patch pipettes had resistances of 4–5 MΩ when filled with an internal solution, except when otherwise mentioned All the experiments were performed at room temperature (25°C).

The I_K(V) were recorded using patch pipettes filled with (mm): 140 potassium gluconate, 1 CaCl₂, 11 EGTA, 2 MgCl₂, 3 Na-ATP, 10 Hepes and 0.2 GTP (pH adjusted to 7.2 with KOH). The composition of the external solution was (mm): 140 NaCl, 2.8 KCl, 2 CaCl₂, 2 MgCl₂, 10 Hepes and 10 glucose (pH adjusted to 7.4 with NaOH). The term I_K(V) refers to the total voltage-sensitive K⁺ current evoked by a voltage step in our experimental protocol. A previous paper from our laboratory has shown that this current (I_K(V)) contains the O₂-sensitive component termed I_K(O2) (Zhu et al. 1996). For the measurement of I_Ca, Ba²⁺ was used as the charge carrier. The pipette solution for I_Ca included (mm): 140 caesium gluconate, 1 CaCl₂, 2 MgCl₂, 10 EGTA, 10 Hepes, 3 Na-ATP and 0.2 GTP (pH adjusted to 7.2 with Tris-base), while the external solution included (mm): 122 N-methylglucamine-glutamate, 20 BaCl₂, 2 MgCl₂, 2.8 CsCl, 10 Hepes, 10 glucose (pH adjusted to 7.4 with Tris-base).

Spherical PC12 cells of 10–15 μm in diameter were voltage clamped using the tight-seal whole-cell recording method. Seventy to seventy-five per cent of the series resistance (< 9 MΩ) was compensated electronically. All solutions containing drugs were applied extracellularly after stable recordings of I_K(_V) or I_Ca had been obtained (usually 3 min after establishing the whole-cell configuration). Current signals were electronically filtered at 1 kHz and digitally sampled. The digitized signals were analysed on a personal computer using pCLAMP 5.5 (for current signals) and Axotape (for voltage signals) analysis programs (Axon Instruments).

In the present study, hypoxia was obtained by equilibrating the perfusion medium with 10 % O₂ gas (balanced with N₂), which resulted in a P_O2 in the media of approximately 80 mmHg. In some cases the media was buffered with 100 % N₂ in the presence of 1 mm sodium dithionate (Na₂S₂O₄), an O₂ chelator. This procedure caused the P_O2 of the media to fall below 10 mmHg.

Ca²⁺ imaging

The cytosolic free Ca²⁺ concentration ([Ca²⁺]_i) was evaluated using the fluorescent Ca²⁺ indicator fura-2 acetoxymethyl ester (fura-2 AM; Molecular Probes). PC12 cells were incubated with fresh serum-free DMEM-F12 medium containing fura-2 AM (5 μm) and Pluronic F127 (0.01 %) for 40 min at 37°C. The cells were then rinsed twice with the same medium without fura-2 AM and left at 37°C for 30 min to allow for hydrolysis of the ester. The coverslips containing the cells were then placed in a Sykes-Moore chamber. Fluorescence imaging was conducted using a xenon fluorescent light source at alternating excitation wavelengths of 340 and 380 nm. Images were captured with a Hamamatsu SIT camera mounted to a Zeiss IM35 fluorescence microscope with a × 40 quartz objective. The integrated light intensities of the cells were measured at 340 and 380 nm by Image-I software (Universal Imaging Corp.) after background subtraction. The ratio of F₃₄₀/F₃₈₀ was used as an indicator of [Ca²⁺]_i, because this allows measurement of [Ca²⁺]_i independent of intracellular fura-2 AM concentration. Experimental solutions were perfused by a peristaltic pump at a flow rate of 5 ml min⁻¹.

Data analysis

The results are expressed as means ±s.e.m. (n, number of observations). An analysis of variance (ANOVA) was used for determining the significance of the measured variables. Statistical significance was accepted at the P < 0.05 level.

Materials

Adenosine, 8-phenyltheophylline, protein kinase A inhibitor fragment (6–22) amide and sodium dithionate were obtained from Sigma. ZM241385 was purchased from Tocris Cookson (Ballwin, MO, USA).

RESULTS

PC12 cells express A_2A and A_2B receptors but not A₁ or A₃ receptors

RT-PCR studies were performed to determine the expression of adenosine receptors in PC12 cells (Fig. 1). Upstream and downstream primers for rat A₁, A_2A, A_2B and A₃ receptors were constructed and used, based on the known rat cDNA sequence. Results from these experiments revealed that both the A_2A and A_2B receptors, but not the A₁ or A₃ receptors, are expressed in PC12 cells (Fig. 1A). Expression of the A₁ and A₃ receptors was detected as a single band in positive control samples (rat brain and rat lung, respectively; Fig. 1A). Analyses of the sequenced PCR products revealed 100 % homology with the GenBank cDNA sequences for both the A_2A and A_2B receptors. These findings confirm that rat PC12 cells selectively express the A_2A and A_2B receptor mRNAs. Although it could be argued that the presence of mRNAs does not always reflect the presence of protein, it is highly unlikely that the mRNAs exist but the protein is not expressed.

Figure 1. Expression of adenosine receptor and adenosine deaminase mRNA in PC12 cells detected by RT-PCR.

A, ethidium bromide visualization of PCR products obtained by RT-PCR of total RNA from PC12 cells for adenosine A₁, A_2A, A_2B and A₃ receptors. Pairs of oligonucleotides were used as primers to specifically amplify the cDNA fragments of these receptors as described in Methods. The PCR products were analysed by 1 % agarose gel electrophoresis in the presence of ethidium bromide. The left-hand lane corresponds to PCR (Promega) or pGEM markers (Promega) - the sizes of the DNA fragments for each marker are indicated on the left. RT-PCR revealed that PC12 cells express both A_2A and A_2B adenosine receptors (predicted product size, 1320 and 885 bp, respectively). On the other hand, neither the A₁ nor the A₃ adenosine receptors were detected in PC12 cells (predicted size, 579 and 1045 bp, respectively). Total RNAs from rat whole brain and rat lung were used as positive controls for A₁ and A₃ receptors. False amplification of the genomic DNA was ruled out by performing RT-PCR without reverse transcriptase, as a negative control (shown as RT-). B, RT-PCR for adenosine deaminase. PC12 cells express adenosine deaminase (predicted size, 455 bp).

We also performed RT-PCR to determine whether PC12 cells express adenosine deaminase, the enzyme that catalyses the first step in adenosine degradation. The primer pairs were constructed based on the known mouse cDNA sequence. RT-PCR revealed that PC12 cells do indeed express the mRNA for this enzyme (Fig. 1B), which is evidence that adenosine is present in PC12 cells.

Effect of adenosine on the hypoxia-induced membrane depolarization

Our laboratory has previously shown that hypoxia induces membrane depolarization by suppressing an oxygen-sensitive K⁺ current in PC12 cells (Zhu et al. 1996; Conforti et al. 1997). We first examined the effect of extracellularly applied adenosine on the hypoxia-induced membrane depolarization. Figure 2A shows the effect of hypoxia (10 %) on membrane potential (V_m) before and after application of adenosine (10 μm). Hypoxia caused depolarization of the membrane potential, which returned to the baseline level after reoxygenation. The hypoxia-induced depolarization was markedly reduced in the presence of adenosine. Following recovery, a subsequent hypoxic exposure without adenosine once again led to depolarization to the same magnitude as that measured prior to adenosine treatment. Figure 2B summarizes the effect of hypoxia and adenosine on membrane potential. The results were obtained following single applications of hypoxia alone, hypoxia with adenosine, or adenosine alone on different cells. Each group represents the mean result from five separate cells. There were no significant differences in the resting membrane potential among the three groups (hypoxia, -33.7 ± 4.2 mV; hypoxia with adenosine, -31.6 ± 2.9 mV; adenosine, -29.6 ± 3.6 mV). Single application of adenosine (10 μm) itself led to membrane hyperpolarization. Taken together, the results show that the hypoxia-induced depolarization in PC12 cells is counteracted by adenosine.

Figure 2. Effect of adenosine on hypoxia-induced membrane depolarization in PC12 cells.

A, time course of the effect of hypoxia on membrane potential (V_m) with and without application of adenosine (ADO; 10 μm), measured by using the current-clamp technique. Hypoxia elicited membrane depolarization, and the V_m returned to the baseline level after reoxygenation. Hypoxia was induced by 10 % O₂. When the cell was then exposed to hypoxia in the presence of adenosine (10 μm), it showed slight, reversible hyperpolarization. A subsequent hypoxic exposure without adenosine revealed depolarization to the same extent as the first hypoxic trial. Shaded areas show the period of hypoxic exposure () and hypoxia with adenosine (10 μm; ). B, the mean data for the effect of hypoxia and/or adenosine on V_m. The results were obtained by single applications of hypoxia alone (n= 5; ), hypoxia with adenosine (10 μm, n= 5; ) or adenosine alone (10 μm, n= 5; ▪) on different cells. Hypoxia induced membrane depolarization, while it induced hyperpolarization in the presence of adenosine (10 μm). Single applications of adenosine (10 μm) led to hyperpolarization.

Activation of A₂ receptors enhances voltage-sensitive K⁺ current and reduces the hypoxia-induced inhibition of I_K(V) in PC12 cells

We performed additional experiments to characterize the modulation of the outward K⁺ current by adenosine in PC12 cells. Figure 3A shows the representative time course of the effects of adenosine on I_K(V) under normoxia. We found that exogenously applied adenosine (10 μm) enhanced the outward I_K(V). The maximal response was observed at 2.5 ± 0.5 min (n= 14) following the application of adenosine, and the increased level of the outward current persisted for almost 1 min following washout of adenosine. The current then declined gradually, and returned to the initial baseline level within 4.2 ± 0.67 min (n= 8). Figure 3B shows a representative trace for the effect of adenosine on the current-voltage relationship of I_K(V). Depolarizing ramps from −60 to +50 mV of 500 ms duration were used. The resulting currents were measured during control and after 3 min exposure to adenosine (10 μm). We measured an increased outward current and shifted reversal potential (V_rev) in the presence of adenosine. The baseline V_rev was −35.3 ± 4.2 mV (n= 5), and shifted to −46.9 ± 4.0 mV (n= 5, P < 0.05) following the application of adenosine. Since the ramps were not leak subtracted, the shift in reversal potential reflects the increased contribution of I_K(V) to total cell current. Figure 3C shows the concentration-dependent effect of adenosine on I_K(V). Adenosine (10 μm) increased the amplitude of I_K(V) by 49.6 ± 5.6 % (n= 14). A maximal increase (126 ± 16 % above baseline level) was measured with 50 μm adenosine. This effect of adenosine appears to be mediated via a receptor mechanism since the effect of adenosine on I_K(V) was blocked by a non-specific adenosine receptor antagonist, 8-phenyltheophylline (8-PT; data not shown). At a concentration of 10 μm, 8-PT markedly reduced the effect of adenosine (10 μm) on I_K(V) (5.2 ± 2.5 % change from baseline, n= 5). 8-PT itself did not induce any change in I_K(V) (0.1 ± 1.2 %, n= 5). 8-PT is known to act primarily as an adenosine receptor blocker, and the phosphodiesterase inhibitor activity of this drug is very small at the concentration used in this study (Smellie, Davis, Daly & Wells, 1979). We further found that the facilitating effect of adenosine was blocked by the selective A_2A receptor antagonist, ZM241385 (Table 1).

Figure 3. Effect of adenosine on voltage-dependent K⁺ current in PC12 cells.

A, time course of the effect of adenosine on I_K(V). I_K(V) was measured every 15 s with depolarizing steps to +50 mV from a V_h of −80 mV (800 ms duration). The current amplitude was measured at the end of the test pulse. Adenosine (ADO; 10 μm) induced a reversible enhancement in the amplitude of I_K(V). Shaded area () corresponds to the time of adenosine application. Inset, superimposed current traces recorded in the absence (a and c) and presence (b) of adenosine. B, effect of adenosine on the current-voltage relationship of I_K(V). A representative trace is shown. The study was done by ramp-depolarizing voltage pulses from -60 to +50 mV with 500 ms duration. Currents generated by ramp pulse were measured before and after the application of adenosine (10 μm). Each trace shows the mean of the currents generated by 5 repeated ramp pulses. Adenosine caused an increased outward current and also induced a shift in the reversal potential towards more negative potentials. C, concentration-dependent enhancement of I_K(V) by adenosine. The vertical line refers to the percentage increase from baseline in the amplitude of I_K(V), which is evaluated at the end of depolarizaing step pulses, similar to A. Adenosine enhanced I_K(V) in a concentration-dependent manner. A maximal increase was measured with 50 μm adenosine. In these experiments, all the measurements were done separately, in different cells. The numbers in the parentheses are the number of cells tested.

Table 1.

Effect of adenosine A_2A receptor antagonist ZM241385 on the response of I_K(V) and I_Ca to adenosine

	Percentage change from baseline
	IKV	ICa
50 μm adenosine	112.4 ± 12.5 (4)	−48.9 ± 3.5 (5)
10 nm ZM241385	−1.6 ± 0.8 (4)	2.2 ± 0.7 (4)
10 nm ZM241385 +
50 μm adenosine	5.0 ± 0.3 (4) *	−4.9 ± 1.2 (4) *

Values are means ±s.e.m. The numbers of cells tested are shown in parentheses.

^*Significant difference (P < 0.01) from the value in the presence of 50 μm adenosine.

We next examined the effect of A₂ receptor activation on the hypoxia-induced change of I_K(V) in PC12 cells. Depolarizing step pulses from a holding potential (V_h) of −80 mV to +50 mV (800 ms in duration) were used to evoke I_K(V). Current amplitudes were measured at the end of the test pulse. Figure 4A shows the effect of adenosine (10 μm) on I_K(V) during hypoxia. Hypoxia was obtained by exposing cells to 10 % O₂, which led to a 20 % reduction in I_K(V). Adenosine counteracted this effect by increasing I_K(V) to a level that was slightly higher than the original baseline level. Figure 4B shows the mean results from this series of experiments. The enhancement of I_K(V) by adenosine during hypoxia could account for the hyperpolarization that was reported in the previous section.

Figure 4. Effect of adenosine on hypoxia-induced I_K(V) inhibition in PC12 cells.

A, time course of the change in I_K(V) during hypoxia before and during application of adenosine (ADO; 10 μm). I_K(V) was measured every 15 s with depolarizing steps to +50 mV from a V_h of −80 mV (800 ms duration). Current amplitude was measured at the end of the test pulse. Hypoxia was obtained with 10 % O₂ in this experiment. Hypoxia-induced inhibition of the I_K(V) was counteracted by subsequent administration of adenosine. Shaded areas correspond to the period of hypoxic exposure () and the application of adenosine (). Inset, I_K(V) currents corresponding to control period (a), exposure to hypoxia (b) and after application of adenosine (c). B, mean data for this experiment (±s.e.m., n= 8). The response was evaluated as percentage change from baseline outward current. Columns show hypoxia () and hypoxia in the presence of adenosine (10 μm; ).

Effect of adenosine on intracellular free Ca²⁺ during hypoxia

Our laboratory reported previously that hypoxia increases intracellular free Ca²⁺ levels ([Ca²⁺]_i) in PC12 cells (Zhu et al. 1996). In the current study, we performed experiments to determine if adenosine modulates the hypoxia-induced enhancement of [Ca²⁺]_i in PC12 cells. Figure 5A shows a representative recording of [Ca²⁺]_i in the absence and presence of adenosine (50 μm). Hypoxia (100 % N₂ and 1 mm sodium dithionate) produced a gradual rise in [Ca²⁺]_i which reached a peak within 5 min and returned to the baseline levels following reoxygenation. The peak levels of [Ca²⁺]_i were similar during the first and second exposures to hypoxia. The hypoxia-induced enhancement of [Ca²⁺]_i was reduced in the presence of adenosine (50 μm). Upon removal of adenosine from the perfusate, an increase in [Ca²⁺]_i was elicited when the cells were re-exposed to hypoxia alone. Figure 5B summarizes the results from the Ca²⁺-imaging studies. These results show clearly that adenosine inhibits the hypoxia-induced enhancement of [Ca²⁺]_i in PC12 cells.

Figure 5. Effect of adenosine on hypoxia-induced enhancement of intracellular free Ca²⁺ in PC12 cells.

A, representative measurement of cytosolic free Ca²⁺ concentration ([Ca²⁺]_i) during hypoxia in the presence and absence of adenosine. [Ca²⁺]_i was estimated by using the fluorescent Ca²⁺ indicator fura-2 AM. Imaging was conducted using a xenon fluorescent light source at alternating excitation wavelengths of 340 and 380 nm. The ratio of F₃₄₀/F₃₈₀ was used to reflect [Ca²⁺]_i. Solutions were perfused by a peristaltic pump at a flow rate of 5 ml min⁻¹. Hypoxia was induced by saturating the 1 mm sodium dithionate-containing perfusate with 100 % N₂. Hypoxia induced the elevation of [Ca²⁺]_i, which returned to the baseline level upon reoxygenation. The response to hypoxia was similar for two repeated exposures. The response was reduced in the presence of adenosine (ADO; 50 μm). The final increase in [Ca²⁺]_i is due to depolarization induced by high K⁺ (30 mm). Shaded areas correspond to the period of hypoxic exposure () and hypoxia with adenosine (). B, averaged data from Ca²⁺ imaging study. The cells were first exposed to hypoxic perfusate, and then to hypoxia alone again or to hypoxia with 50 μm adenosine. These two protocols were conducted in separate groups of cells (n= 12 and 14, respectively). There were no differences in the peak levels of [Ca²⁺]_i between the first and second hypoxic exposure (left). On the other hand, the enhancement of [Ca²⁺]_i during hypoxia was significantly inhibited in the presence of adenosine (right).

Activation of A₂ receptors inhibits a voltage-dependent Ca²⁺ current in PC12 cells

To further investigate the mechanisms by which adenosine attenuated the hypoxia-induced enhancement of intracellular Ca²⁺, the direct effects of adenosine on voltage-dependent Ca²⁺ currents were studied in PC12 cells. Figure 6A shows the peak current-voltage relationship before and after the application of adenosine (10 μm). I_Ca was measured with test pulses of 100 ms duration from a V_h of −80 mV to test potentials from −50 to +60 mV. Adenosine (10 μm) reduced the amplitude of I_Ca by 39.3 ± 7.8 % at +20 mV (n= 6). The effect of adenosine on I_Ca was determined as a percentage change by comparing the peak amplitude of I_Ca in the presence of adenosine to that of I_Ca prior to the exposure to adenosine. The time course of adenosine-induced inhibition of I_Ca is shown in Fig. 6B. Ca²⁺ current was elicited from a V_h of −80 mV with depolarizing steps to +20 mV (pulses of 160 ms duration). The effect of adenosine on I_Ca reached a peak in 1.0 ± 0.3 min (n= 6) and recovered to the original control level within 2.5 ± 0.5 min after the start of washout. We found that adenosine inhibited I_Ca in a concentration-dependent manner (Fig. 6C). The inhibitory effect of adenosine (10 μm) on I_Ca was blocked by an adenosine receptor antagonist, 8-PT (10 μm; −5.7 ± 4.4 % change from baseline, n= 5). 8-PT itself did not induce any change in I_Ca (−1.4 ± 5.1 %, n= 5; data not shown). In addition, the effect of adenosine on I_Ca was also blocked by a selective A_2A receptor antagonist, ZM241385 (Table 1).

Figure 6. Effect of adenosine on voltage-dependent Ca²⁺ current in PC12 cells.

A, peak current-voltage relationship before and after the application of adenosine (ADO; 10 μm). I_Ca was measured with 100 ms-long test pulses from a V_h of −80 mV to test potentials ranging from −50 to +60 mV (10 mV increments). Adenosine decreased I_Ca at the voltage range examined. B, time course of the effect of adenosine on I_Ca. I_Ca was measured at room temperature every 30 s by 160 ms test pulses from a V_h of −80 to +20 mV. Peak current amplitude was measured for evaluation. The charge carrier was 20 mm Ba²⁺. Adenosine (10 μm) elicited a decrease in the amplitude of I_Ca, which fully returned to baseline level after washing. Inset, superimposed current traces recorded before (a), during (b) and after (c) application of adenosine. C, concentration-response relationship of the effect of adenosine on I_Ca. Adenosine reduced the peak I_Ca in a concentration-dependent manner. The response was evaluated as percentage decrease from baseline inward currents. The numbers in parentheses indicate the number of cells for each data point.

Protein kinase A (PKA) mediates I_K(V) and I_Ca responses to adenosine

Activation of the A₂ receptor is thought to be associated with stimulation of adenylate cyclase and activation of PKA (Dalziel et al. 1994; Fredholm, 1995). We hypothesized, therefore, that the effects of adenosine on I_KV and I_Ca are mediated by PKA. To test this possibility, we examined the effect on I_KV and I_Ca in cells in which PKA was blocked by the PKA inhibitor fragment (6–22) amide (PKI). This small, heat-stable inhibitory peptide binds to the catalytic subunit of PKA and inactivates the enzyme (Glass, Lundquist, Katz & Walsh, 1989). Dialysis of cells with a pipette solution containing PKI has been shown to effectively inhibit the PKA-mediated current events (Schackow & Ten Eick, 1994). In the present study, PC12 cells were dialysed with PKI (100 μm) using large-tip pipettes of 1–2 MΩ resistance for at least 6 min prior to the start of an experiment. We found that the intracellularly applied PKI abolished the effect of adenosine on I_K(V) and I_Ca (Fig. 7A).

Figure 7. Involvement of protein kinase A in the response of I_K(V) and I_Ca to adenosine.

A, abolition of the effects of adenosine on I_K(V) and I_Ca by intracellular PKA inhibitor fragment (6–22) amide (PKI). PC12 cells were dialysed with a pipette solution including 100 μm PKI using large-opening pipettes of 1–2 MΩ resistance for at least 6 min. The experiments were started after steady-state levels of I_K(V) or I_Ca currents were achieved. Voltage protocols were the same as those described in the legends of Figs 3 and 6. Stimulatory effect of adenosine (ADO; 10 μm) on I_K(V) (left; ) was significantly abolished in the presence of PKI in a pipette solution (left; ▪). Similarly, inhibitory action of adenosine (10 μm) on I_Ca (right; ) was significantly reduced by PKI (right; ▪). B, absence of the effects of adenosine on I_K(V) and I_Ca in PKA-deficient PC12 cells (A123.7), which lack PKA activity by genetic mutation. In PKA-deficient PC12 cells, the effect of adenosine (10 μm) on I_K(V) and I_Ca was not observed (▪), while wild-type PC12 cells cultured under the same conditions as A123.7 cells showed the usual response (). The numbers in the parentheses are the number of cells tested.

These results were confirmed by our finding that adenosine failed to induce significant changes in either I_K(V) or I_Ca in mutant PC12 cells which are deficient in PKA (A123.7; Ginty et al. 1991). The results from these experiments are shown in Fig. 7B. To rule out the possibility that these results were due to the absence of adenosine receptors in A123.7 cells, we examined the expression of adenosine receptors in these cells by RT-PCR. We found that the A123.7 cells express both the A_2A and A_2B receptors at approximately the same level as wild-type PC12 cells (Fig. 8). Taken together, these results show that the effects of adenosine on I_KV and I_Ca are mediated via the PKA pathway.

Figure 8. Expression of adenosine receptors in PKA-deficient PC12 cells.

Ethidium bromide visualization of PCR products obtained by RT-PCR of total RNA for adenosine receptors in PKA-deficient PC12 cells. Size markers, pGEM and PCR marker, are shown in the left-hand lanes (the sizes of the DNA fragments are indicated on the left). Pairs of oligonucleotides were used for RT-PCR as described in Methods. RT-PCR was performed according to the same protocols as those shown in Fig. 1. This cell line expresses both A_2A and A_2B receptors. Neither A₁ nor A₃ receptors were detected in PKA-deficient cells. The results were the same as the wild-type PC12 cells shown in Fig. 1. Negative controls without reverse transcriptase (labelled as RT-) were performed for each gene in order to assess any genomic DNA contamination.

DISCUSSION

The present study was undertaken to investigate the role of adenosine in modulating the cellular responses to hypoxia in PC12 cells. We confirmed that these cells express both the A_2A and A_2B receptors, but not the A₁ or A₃ receptors. An important observation was that membrane depolarization and enhancement of intracellular free Ca²⁺ during hypoxia are markedly attenuated by adenosine in PC12 cells. We also found that the voltage-sensitive K⁺ current (I_K(V)) and hypoxia-induced suppression of the I_K(V) are greatly attenuated by adenosine. Moreover, our results show that activation of the A₂ receptor inhibited the voltage-activated Ca²⁺ current. Our data revealed that the effects of adenosine on these currents are mediated via A_2A receptors and that PKA is necessary for adenosine action. These results provide strong evidence that adenosine modulates membrane excitability during hypoxia in PC12 cells, and that these effects are mediated by the action of adenosine on both K⁺ and Ca²⁺ currents through the A_2A receptor and a PKA-dependent intracellular signalling pathway.

We examined the expression of adenosine receptor subtypes in PC12 cells. Adenosine receptors were identified in PC12 cells in an early study using a ligand-binding technique (Williams, Abreu, Jarvis & Noronha-Blob, 1987). More recent studies using ligand binding (Hide et al. 1992) and RT-PCR (van der Ploeg et al. 1996) revealed that PC12 cells express both A_2A and A_2B receptors. However, the presence of A₁ or A₃ receptors was not investigated. Our experiments using RT-PCR confirm these earlier findings, and in addition, we report that neither A₁ nor A₃ receptor mRNAs are present in these cells.

An important finding in the current study was that membrane depolarization and the increase in intracellular free Ca²⁺ evoked by hypoxia were significantly attenuated by adenosine in PC12 cells. This is important because adenosine is one of the major mediators for many physiological and pathophysiological responses that are associated with hypoxia in different tissues (Bruns, 1990; Dalziel & Westfall, 1994; Fredholm, 1995). Since the effects of adenosine on I_K(V) and I_Ca were blocked by the selective A_2A receptor antagonist ZM241385, it is likely that these effects of adenosine on the hypoxia-induced cellular responses are mediated via A_2A receptors. ZM241385 is selective for A_2A receptors and blocks only this A₂ subtype at the concentration used in this study (Poucher et al. 1994). It has been shown that the A_2A and A_2B receptors have different tissue distributions and ligand affinities (Olah et al. 1995). Both receptors are thought to be coupled to the G_s protein which stimulates adenylate cyclase activity. However, it takes a 100-fold higher concentration of the specific A₂ receptor agonist NECA (5′-N-ethyl-carboxamidoadenosine) to stimulate the A_2B receptor than the A_2A receptor (Chern, Lai, Fong & Liang, 1993). Since PC12 cells only express the A₂ receptors, the possible cross-reactivity with A₁ and A₃ receptors is not a problem.

Our results suggest that the enhancement of the voltage-sensitive K⁺ current (I_K(V)) by adenosine is the possible mechanism for the reduced membrane depolarization during hypoxia in PC12 cells. This is because the activation of A₂ receptors with adenosine enhances I_K(V), and this effect, which leads to membrane hyperpolarization, was abolished by pretreatment of the cells with the adenosine receptor antagonist. Furthermore we showed that the hypoxia-induced inhibition of I_K(V) is significantly counteracted by adenosine. Adenosine was previously shown to increase resting K⁺ conductance via the A₁ receptor in CNS neurons, which, in turn, decreased cellular excitability (Gerber, Greene, Haas & Stevens, 1989). However, very little information exists concerning the role of the A₂ receptor in modulating K⁺ currents. It would also be important to establish whether adenosine acts on the hypoxia-sensitive K⁺ channel (I_K(O2)) and other subtypes of the K⁺ channel. A previous paper from our laboratory has revealed that the I_K(O2) in PC12 cells has characteristics similar to those of the delayed rectifier-like I_K (Zhu et al. 1996). A recent single-channel experiment has shown the type of K⁺ channel that is present in PC12 cells (Conforti & Millhorn, 1997). It was revealed that only the 20 pS delayed rectifier K⁺ channel is inhibited by hypoxia. There have been several reports which indicate that the A₂ receptors modulate the ATP-sensitive K⁺ channel (I_K(ATP)), which is believed to be involved in hypoxia-induced coronary artery dilatation (Kleppisch & Nelson, 1995). The modulation of K⁺ currents in response to adenosine cannot be ascribed to I_K(ATP) activation, since the concentration of ATP in our recording pipette was sufficient to inactivate I_K(ATP). In the current study, we are not able to distinguish whether adenosine and hypoxia act on the same set of K⁺ channels. Single-channel experiments will have to be conducted to determine whether adenosine-responsive K⁺ channels are the same as the I_K(O2).

The increase in [Ca²⁺]_i during hypoxia in oxygen-sensitive cells occurs as the result of membrane depolarization and activation of voltage-dependent Ca²⁺ channels (Gonzalez et al. 1994; Lopez-Barneo, 1996). In the carotid body cells, the hypoxia-induced increase in [Ca²⁺]_i is inhibited by pharmacological blockade of voltage-gated Ca²⁺ channels (Buckler & Vaughan-Jones, 1994). Ca²⁺ channel blockers have also been shown to inhibit the hypoxia-induced enhancement of [Ca²⁺]_i in PC12 cells (Raymond & Millhorn, 1997). Therefore it seems most likely that the increase in [Ca²⁺]_i during hypoxia in PC12 cells is mediated by the activation of voltage-gated Ca²⁺ channels. The mechanism by which adenosine causes the attenuation of hypoxia-evoked enhancement of [Ca²⁺]_i is unknown, but could be due to the effect of adenosine on the voltage-dependent Ca²⁺ channels. Support for this comes from the finding that extracellularly applied adenosine inhibited the voltage-dependent Ca²⁺ currents in PC12 cells, and that this effect was blocked by the adenosine receptor antagonist. To our knowledge, this is the first evidence that a voltage-gated I_Ca is modulated by activation of the A₂ receptor in PC12 cells. It was recently reported that adenosine inhibits Ca²⁺ currents in guinea-pig adrenal chromaffin cells, but the specific receptor that mediates this response was not identified (Otsuguro, Ohta, Ito & Nakazato, 1996). In PC12 cells, adenosine has been shown to modulate the ATP-activated non-selective cation channels, which are related to the ATP-evoked Ca²⁺ influx (Inoue, Watano, Koizumi, Nakazawa & Burnstock, 1994). Adenosine has been shown to reduce voltage-activated Ca²⁺ currents via the A₁ receptor in neuronal cells (Scholz & Miller, 1991). Our findings indicate that the A₂ receptor regulates the I_Ca and the level of intracellular free Ca²⁺ in a manner that is similar to that reported for A₁ receptor activation in the nervous system (Fredholm & Dunwiddie, 1988).

Our findings reveal that PKA is required for the A₂ receptor modulation of both I_K(V) and I_Ca in PC12 cells. This conclusion is based on our results which showed that the effects were blocked by PKI and absent in PKA-deficient PC12 cells. These results are in agreement with the traditional notion that the A₂ receptor is associated with stimulation of adenylate cyclase via G_s protein (Dalziel et al. 1994; Fredholm, 1995). Although the regulation of ion channels by G proteins has been intensively investigated (Wickman & Claphan, 1995), relatively little is known about the coupling of the A₂ receptor to either the K⁺ or Ca²⁺ channel in PC12 cells. Our study suggests that activation of the cAMP-PKA pathway leads to inhibition of Ca²⁺ channels and stimulation of K⁺ channels. It is generally known that cAMP stimulates the L-type Ca²⁺ channel in cardiac muscles and skeletal muscles (Wickman & Claphan, 1995) and modulates the gating properties of L-type Ca²⁺ channels in bovine adrenal chromaffin cells, which are phenotypically similar to PC12 cells (Doupnik & Pun, 1992). However, it is possible that the action of cAMP on Ca²⁺ channel activities is tissue dependent, since cAMP acts to inhibit the opening of L-type Ca²⁺ channels at depolarized potentials in vascular smooth muscle cells (Sperelakis, Xiong, Haddad & Masuda, 1994). The inhibition of I_Ca and the stimulation of I_K(V), induced by the activation of PKA by adenosine, would suppress membrane depolarization elicited by hypoxia and thereby diminish the Ca²⁺ influx during hypoxia. Our laboratory has shown that intracellular free Ca²⁺ is required for the regulation of tyrosine hydroxylase gene expression during hypoxia in PC12 cells (Raymond & Millhorn, 1997). We have also shown that dopamine is released from PC12 cells during hypoxia (Zhu et al. 1997). Therefore it is possible that A₂ receptor-mediated regulation of the membrane potential and the intracellular free Ca²⁺ levels plays an important role in the modulation of hypoxia-induced gene expression and dopamine release.

We examined the expression of adenosine deaminase in PC12 cells by RT-PCR, and found that the cells express this enzyme. Since adenosine deaminase catalyses the first step of adenosine degradation, our result suggests that PC12 cells produce adenosine. If adenosine were released from PC12 cells during hypoxia, it seems possible that adenosine, in turn, acts in an autocrine- or paracrine-like fashion to regulate the cellular response to hypoxia via the A₂ receptors. There is growing evidence that this might indeed be the case. For example, it has been reported that the A₂ receptor in PC12 cells is activated by endogenously released adenosine (Erny, Berezo & Perlman, 1981). In addition, it has been reported that there is a direct linkage between adenosine production and the breakdown of ATP during hypoxia (Winn et al. 1981), since the major source of adenosine is from the breakdown of intracellular ATP (Dubyak & El-Moatassim, 1993). If adenosine is released and mediates feedback regulation of ionic currents in PC12 cells during hypoxia, we would expect to find that the hypoxia-induced inhibition of I_K(V) is enhanced in the presence of an adenosine receptor antagonist. We examined this possibility; however, we failed to measure any difference in these experiments (data not shown). The lack of effect of the antagonist might be due to the fact that the concentration of adenosine failed to reach a high enough level to produce any effect. This is because (1) the cells were plated sparsely and any adenosine released may have been diluted to such an extent that the effective concentration was never reached, and (2) the cells were continuously superfused and thus any adenosine released would have been washed away. However, under in vivo conditions, where the cells are contained in a small volume, the situation is very different. A recent study demonstrated that increasing the intracellular concentration of adenosine applied through the patch-clamp pipette in a single neuron selectively inhibits the excitatory postsynaptic potentials in that cell, which is consistent with the hypothesis that adenosine acts as a feedback to regulate the cellular response to various stimuli (Brundege & Dunwiddie, 1996). Further studies are required to confirm the hypothesis that the cellular membrane exicitability of PC12 cells during hypoxia is regulated by endogenous adenosine in a feedback manner.