Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Adenosine A2A Receptor Modulates the Activity of Globus Pallidus Neurons in Rats

Adenosine A2A Receptor Modulates the Activity of Globus Pallidus Neurons in Rats ORIGINAL RESEARCH published: 07 November 2017 doi: 10.3389/fphys.2017.00897 Adenosine A Receptor Modulates 2A the Activity of Globus Pallidus Neurons in Rats 1, 2 1 1 1, 3 1 Hui-Ling Diao , Yan Xue , Xiao-Hua Han , Shuang-Yan Wang , Cui Liu , 1 1 Wen-Fang Chen and Lei Chen * 1 2 Department of Physiology, Qingdao University, Qingdao, China, Department of Physiology, Binzhou Medical University, Yantai, China, Department of Anatomy, Qingdao University, Qingdao, China The globus pallidus is a central nucleus in the basal ganglia motor control circuit. Morphological studies have revealed the expression of adenosine A receptors in the 2A globus pallidus. To determine the modulation of adenosine A receptors on the activity 2A of pallidal neurons in both normal and parkinsonian rats, in vivo electrophysiological and behavioral tests were performed in the present study. The extracellular single unit recordings showed that micro-pressure administration of adenosine A receptor 2A agonist, CGS21680, regulated the pallidal firing activity. GABAergic neurotransmission was involved in CGS21680-induced modulation of pallidal neurons via a PKA pathway. Furthermore, application of two adenosine A receptor antagonists, KW6002 or 2A SCH442416, mainly increased the spontaneous firing of pallidal neurons, suggesting Edited by: that endogenous adenosine system modulates the activity of pallidal neurons through Hansen Wang, adenosine A receptors. Finally, elevated body swing test (EBST) showed that 2A University of Toronto, Canada intrapallidal microinjection of adenosine A receptor agonist/antagonist induced 2A Reviewed by: Carla Cannizzaro, ipsilateral/contralateral-biased swing, respectively. In addition, the electrophysiological Università degli Studi di Palermo, Italy and behavioral findings also revealed that activation of dopamine D receptors by Rosario Moratalla, quinpirole strengthened KW6002/SCH442416-induced excitation of pallidal activity. Consejo Superior de Investigaciones Científicas (CSIC), Spain Co-application of quinpirole with KW6002 or SCH442416 alleviated biased swing in *Correspondence: hemi-parkinsonian rats. Based on the present findings, we concluded that pallidal Lei Chen adenosine A receptors may be potentially useful in the treatment of Parkinson’s 2A chenleiqd@163.com disease. Specialty section: Keywords: globus pallidus, adenosine A receptors, Parkinson’s disease, extracellular single unit recording, 2A This article was submitted to elevated body swing test Integrative Physiology, a section of the journal Frontiers in Physiology INTRODUCTION Received: 15 June 2017 Accepted: 24 October 2017 The rodent globus pallidus (homolog of the external segment of the primate globus pallidus) is Published: 07 November 2017 a central nucleus in the indirect pathway of the basal ganglia circuit (Jellinger, 1991), which plays Citation: important roles in movement regulation under healthy and pathological states (Raz et al., 2000; Kita Diao H-L, Xue Y, Han X-H, Wang S-Y, and Kita, 2001; Dodson et al., 2015; Hegeman et al., 2016). Based on the in vivo electrophysiological Liu C, Chen W-F and Chen L (2017) characteristics, the globus pallidus neurons are classified into three firing patterns including Adenosine A Receptor Modulates 2A high-frequency without pauses, high-frequency with pauses and low-frequency with bursts the Activity of Globus Pallidus Neurons (Benhamou et al., 2012). The spontaneous firing activities of pallidal neurons are closely associated in Rats. Front. Physiol. 8:897. doi: 10.3389/fphys.2017.00897 with movement amplitude, velocity and direction (Gage et al., 2010; Hegeman et al., 2016). Frontiers in Physiology | www.frontiersin.org 1 November 2017 | Volume 8 | Article 897 Diao et al. Pallidal Adenosine A Receptor 2A Previous studies have revealed that aberrant pallidal neuron in the globus pallidus of both intact and 6-hydroxydopamine activities appear at the onset and maintenance of motor (6-OHDA) parkinsonian rats. dysfunction in Parkinson’s diseases (Kita, 2007; Obeso et al., 2008). Furthermore, the decreased firing rate and synchronous MATERIALS AND METHODS bursting of pallidal neurons are strongly related to the motor symptoms of Parkinson’s diseases (Raz et al., 2000; Sani et al., Animals 2009; Chan et al., 2011). Adult male Wistar rats (Qingdao, China), 8–10 weeks of age and Adenosine is an endogenous purine nucleoside which weighing 220–290 g, were used in this experiment. Rats were plays a wide variety of roles in central nervous systems, ◦ housed in an environmentally controlled room at 22 ± 1 C including development, sleep, synaptic transmission, pain, with a 12 h light/dark cycle. The study was performed strictly neuroinflammation, anxiety and depression. Adenosine in accordance to the University ethics guidelines. All operations receptors (A , A , A , and A ) belong to G-protein-coupled 1 2A 2B 3 were required to lower rats’ suffering and pain. A total of 115 receptors with A and A receptors couple to Gi and Go while , 1 3 rats were used for electrophysiological experiments, with 81 A and A receptors couple to Gs proteins (Fredholm et al., 2A 2B normal rats and 34 successful parkinsonian rats. Three of the 2001, 2011; Ferre et al., 2008). It is known that adenosine A 2A 37 parkinsonian rats did not show any stable recordings and receptors are highly expressed in the striatum, globus pallidus, were excluded from the study. In addition, 78 rats were used nucleus accumbens, and olfactory tubercles of rat and human for behavioral study and 8 rats were used for immunoflurescence brain, as compared to other adenosine receptor subtypes with staining. widespread brain distribution (Jarvis and Williams, 1989; Martinez-Mir et al., 1991; Rosin et al., 2003). The selective and Establishment of 6-Hydroxydopamine specific location of adenosine A receptors suggests that it may 2A (6-OHDA) Hemi-Parkinsonian Rat Model become a potential therapeutic target for basal ganglia diseases, Rats were injected with chloral hydrate (400 mg/kg, i.p.) and fixed particularly Parkinson’s disease (Preti et al., 2015; Cunha, 2016). in the stereotaxic instrument (NarishigeSN-3, Tokyo, Japan). Considerable preclinical studies have shown that blockade of 6-hydroxydopamine (6-OHDA, 4 μg/μl in saline with 0.01% adenosine A receptors could symptomatically relieve from 2A ascorbic acid, 4 μl) was injected into the left medial forebrain parkinsonian motor deficits without L-DOPA-related motor bundle (AP −4.3, ML +1.7, DV −8.4 mm from Bregma) using side effects (Bibbiani et al., 2003; Jenner, 2014; Preti et al., 2015; microsyringe at a rate of 1.0 μl/min. Animals that exhibited at Cunha, 2016). least 210 net contralateral rotations in 30 min after delivered Anatomical and morphological studies suggest that the apomorphine (0.2 mg/kg, s.c.) were considered as successful adenosine A receptors are principally expressed on 2A hemi-parkinsonism rats. striatopallidal terminals in the globus pallidus (Rosin et al., 2003; Shindou et al., 2003). Previous in vitro electrophysiological studies show that stimulation of adenosine A receptors Electrophysiological Recordings in Vivo 2A enhances GABA release and therefore augments the inhibitory One of the advantages of in vivo extracellular recording used postsynaptic currents (IPSCs) in the globus pallidus (Mori and in this study is to investigate the direct effects of drugs on the Shindou, 2003; Floran et al., 2005). Other studies reveal that single pallidal neurons in both physiological and pathological adenosine A receptor activation exerts a dual effect on the conditions. The other advantage of the present multibarrel 2A release of GABA (Mayfield et al., 1993; Dayne Mayfield et al., microelectrodes is to apply drugs directly to the neurons 1996; Morales-Figueroa et al., 2014). However, Querejeta et al. recorded. According to our previous studies (Xue et al., 2010; (2010) demonstrate that intrapallidal infusion of adenosine Chen et al., 2015), extracellular recordings were performed in the A receptor agonist and antagonist have no effects on the globus pallidus of normal urethane (1 g/kg, i.p.) anesthetized rats. 2A spontaneous firing rate of the globus pallidus neurons in both Anaesthesia levels were monitored constantly by testing reflexes sham and ipsilaterally dopamine-denervated rats. As adenosine to a cutaneous pinch and maintained by giving supplement A receptor antagonist is a potential therapeutic target for dose of chloral hydrate (0.1 g/kg, i.p.) when necessary. Rats 2A parkinsonian motor deficits (Bibbiani et al., 2003; Jenner, 2014; were fixed in the stereotaxic instrument. Two holes were Preti et al., 2015; Cunha, 2016), the study of pallidal adenosine drilled bilaterally over the globus pallidus according to the A receptors will provide an insight into the movement stereotaxic atlas (0.8–1.2 mm posterior to bregma and 2.5– 2A regulation in both normal and parkinsonian states. Up to 3.5 mm lateral to the midline, Paxinos and Watson, 1986). All present, little is known about the in vivo electrophysiological exposed cortex were covered with a thin layer of saline to and behavioral effects and the possible mechanisms of adenosine prevent drying. Recordings of neuronal spike and micro-pressure A receptors in the globus pallidus under both normal ejection of drugs were used with fined 3-barrelled micropipettes 2A and abnormal states. By using multibarrel microelectrode (tip diameter 3–10 μm, resistance10–20 M). The recording extracellular recordings and elevated body swing test (EBST), microelectrode contained 2% pontamine sky blue dissolved in we therefore investigated the effects of adenosine A receptors 0.5 M sodium acetate. The other two microelectrodes were filled 2A with vehicle (saline) and various drugs, including CGS21680, KW6002, SCH442416, quinpirole, gabazine, nipecotic acid, or Abbreviations: 6-OHDA, 6-hydroxydopamine; GABA, γ-aminobutyric acid; PKA, protein kinase A; EBST, elevated body swing test. H-89. Drugs were delivered using 4-channel pressure injector Frontiers in Physiology | www.frontiersin.org 2 November 2017 | Volume 8 | Article 897 Diao et al. Pallidal Adenosine A Receptor 2A (PM2000B, Micro Data Instrument, Inc., USA) with short- globus pallidus sections were blocked with 5% donkey serum pulse gas pressure (1,500 ms, 5.0–15.0 psi). The recording (Solaribo) and 0.3%Triton X-100 in PBS for 60 min at 4 C. electrical signal was amplified by a micro-electrode amplifier And the sections were subsequently incubated with a rabbit (MEZ-8201, Nihon Kohden, Tokyo, Japan), low- and high- polyclonal anti-adenosine A receptor antibody (sc-13937, 2A pass filtered at 0.3 and 3 kHz, monitored with a memory 1:100; Santa Cruz) and a goat polyclonal anti-parvalbumin oscilloscope (VC-11, Nihon Kohden, Tokyo Japan) and an audio antibody (ab3289, 1:500; Abcam) for 48 h at 4 C. After washing monitor. The electrical signal was transferred into bio-electricity three times with PBS (5 min), the sections were incubated signal analyzer and computer. The pallidal spiking data capture with the following secondary antibodies, Alexa 488-conjugated and analysis were available via spike 2 software (Cambridge donkey anti-rabbit (1:1,000; Abcam) and Alexa 647-conjugated Electronic Design, UK). donkey anti-goat (1:1,000; Abcam), for 2 h at room temperature. Drug application was initiated after basal spontaneous firing Lastly, the sections were washed, mounted, coverslipped and of pallidal neurons remained stable for at least 10 min. The examined under fluorescent microscope or laser scanning 120 s average firing before drug application was regarded as the confocal microscopy (Leica, Wetzlar, Germany). basal firing, and the 50 s maximally changing discharge after The results were analyzed by counting the number of drug administration was considered as the effect. The increase adenosine A receptor-positive cells in the globus pallidus. 2A or decrease in firing frequency was considered statistically The adenosine A receptor-positive cells were examined and 2A significant when the firing rate was higher or lower than the mean quantified using Image J software (NIH, Bethesda, MD). We ±2 SD of the baseline (SD = standard deviation). Firing pattern selected six sections (at the levels of 0.8–1.2 mm posterior was assessed via the coefficient of variation (CV) which referred to bregma) from the globus pallidus from each rat for cell to the standard deviation of the interspike intervals (ISI) divided counting. Results were expressed as the average number of by mean. positive cells obtained from the six sections (cells visualized at 400× magnifications). Elevated Body Swing Test (EBST) EBST was performed in awake and freely-moving rats according Histological Controls to previously described methods (Borlongan et al., 1995; After electrophysiological and behavioral experiments, we Baluchnejadmojarad and Roghani, 2004). The rat was placed in needed to verify the position of electrophysiological recording a transparent plastic cage (40 × 40 × 35 cm). After a 10 min- sites and cannula placements. Rats were deeply anesthetized habituation, the rat was elevated to ∼2 cm above the bottom of and perfused transcardially with a saline solution containing the cage by holding 2 cm above the bottom of tail. The body 4% paraformaldehyde. Recording/microinjection sites were swing referred to the rat head with deviation of vertical axis more confirmed using camera (Figure 1). To further confirm the than 10 . Before every swing was recorded, the rat head must 6-OHDA parkinsonian rat model, coronal substantia nigra be suspended at a vertical axis. If the rat did not bend its head sections were incubated with a rabbit anti-tyrosine hydroxylase after being elevated over 5 s, the tail was gently pinched. Each test antibody (T8700, 1:1000; sigma) following secondary antibodies rat was recorded for 60 s. Initially, only rats showing unbiased (1:1,000; Abcam) of Alexa 488-conjugated donkey anti-rabbit. behavior were chosen for the study. The number of tyrosine hydroxylase-positive neurons on the All experimental rats were anesthetized with chloral hydrate lesioned side of the substantia nigra pars compacta decreased to (400 mg/kg, i.p.), and implanted with the stainless steel guide 17.08 ± 1.84%, which was significantly lower than that of normal cannula in the globus pallidus on either side (o.d., 0.4 mm; rats (Figure 5A). i.d., 0.3 mm). The cannula was secured to the skull with screws and dental acrylic. The vehicle (saline) or drugs (CGS21680, KW6002, and SCH442416) was unilaterally microinjected into Drugs and Statistical Analysis the globus pallidus with 1 μl microsyringe. The injection rate was CGS21680, KW6002, SCH44216 and quinpirole were purchased maintained at 0.2 μl/min with a total volume of 0.5 μl and the from Tocris (Bristol, UK). DMSO, 6-OHDA hydrochloride, microsyringe was kept in position for an additional 3 min before apomorphine hydrochloride, gabazine, and H-89 were obtained retraction. from Sigma (St Louis, Mo, USA). Nipecotic acid was purchased The number of biased body swing was counted per minute. from Abcam (Cambridge, MA, USA). The percentage score was calculated for each rat, i.e., the number All data were presented as mean ± S.E.M. Paired-samples of biased swings was divided by the total number of swings and t-test was used for comparing the difference of spontaneous multiplied by 100%. Over 70% of biased swings were defined as firing before and after drug application. Statistical comparisons the criterion for biased swing behavior. between or among groups were determined with independent- samples t-test and one-way ANOVA. Chi-square test was applied Double Immunofluorescence Staining to compare the fractions of CGS21680 responsive neurons and The double immunofluorescence technique was used to observe different concentrations of drugs. Before t-tests, Levene’s test the expression of adenosine A receptors and parvalbumin in F-test was performed to determine if the data is normally 2A rat globus pallidus. Brain tissue was obtained from normal and distributed. Wilcoxon–Mann–Whitney test was used for data parkinsonian rats, and then was fixed in 4% paraformaldehyde that are not normally distributed. P < 0.05 was considered as overnight. Tissues containing the globus pallidus were sectioned the level of significance. Data analysis was performed using IBM at 40 μm after dehydration in 20 and 30% sucrose. Coronal SPSS 22.0 software. Frontiers in Physiology | www.frontiersin.org 3 November 2017 | Volume 8 | Article 897 Diao et al. Pallidal Adenosine A Receptor 2A FIGURE 1 | Confirmation of the recording and microinjection location within the globus pallidus. (A) A coronal brain section illustrating the trace of microelectrodes in the globus pallidus. (B) A coronal brain section showing the injector tip for microinjection. GP, globus pallidus; Str, striatum; scale bars = 1 mm. RESULTS P < 0.05, paired-samples t-test), with the average increase of 33.04 ± 1.60% (t = 10.99, df = 15, P < 0.001 compared to Effects of Adenosine A Receptor 2A that of vehicle administration, independent-samples t-test). In Activation on the Spontaneous Firing of the remaining 20 pallidal neurons, 10 μM CGS21680 did not alter the firing rate significantly. The highest concentration of Globus Pallidus Neurons in Normal Rats We first observed the effects of adenosine A receptor selective CGS21680 (100 μM) only increased the pallidal firing rate in 2A 3 out of the 19 neurons, while the lowest concentration of the agonist CGS21680 on the spontaneous firing rate of 47 pallidal neurons tested. The average neuronal firing rate was 14.02 drug (0.1 μM) only had decreasing effects in 1 out of the 13 neurons. Further analysis revealed that the percentages of pallidal ± 1.44 Hz. Micro-pressure administration of 1 μM CGS21680 neurons which were inhibited by different concentrations of significantly decreased the frequency of spontaneous firing from 14.89 ± 2.48 Hz to 9.03 ± 2.14 Hz in 21 out of the 47 pallidal CGS21680 (0.1, 1, 10, and 100 μM) were significantly different based on Chi-square test (x = 28.26, df = 6, P < 0.001). The neurons (n = 21, P < 0.001, t = 5.78, df = 20, paired- samples t-test, Figures 2A,C). The average decrease was 46.08 percentage of reactive neurons (44.68%) with 1 μM CGS21680- induced inhibition of firing rate was strongest compared to that ± 5.71%, which was significantly different (Z = 4.72, P < 0.001, Mann-Whitney test) compared to that of vehicle administration of other drug concentrations (Figure 2D), while there was no significant change in the percentage of reactive neurons with (basal: 15.31 ± 3.24 Hz; vehicle: 15.58 ± 3.21 Hz; n = 12, t = 1.61, df = 11, P > 0.05, paired-samples t-test). In only CGS21680-induced excitatory effects of all drug concentrations. Therefore, we chose the concentration of 1 μM to further explore 7 out of the 47 pallidal neurons, 1 μM CGS21680 increased the firing frequency from 10.05 ± 3.40 to 12.66 ± 3.77 Hz the effects of adenosine A receptors on pallidal neurons in the 2A subsequent experiments. (n = 7, t = 4.34, df = 6, P < 0.01, paired-samples t-test, Next, we analyzed the possible changes of firing pattern Figures 2B,C), with the average increase of 50.80 ± 20.68% (Z = 3.47, P < 0.01 compared to that of vehicle administration, by calculating the CV before and after CGS21680 application. Micro-pressure ejection of 1 μM CGS21680 didn’t change the CV Mann-Whitney test). In the remaining 19 pallidal neurons, 1 μM CGS21680 did not alter the firing rate significantly (t = 1.27, significantly (basal: 1.30 ± 0.32; CGS21680: 1.77 ± 0.39; n = 21, t =1.90, df = 20, P > 0.05, paired-samples t-test, Figure 2E) in the df = 18, P > 0.05, paired-samples t-test). Similar to that of 1 μM CGS21680, the higher concentration of CGS21680 (10 μM) also 21 neurons with CGS21680-induced decrease in firing rate. There was also no change of CV in the neurons with 1 μM CGS21680- modulated the spontaneous firing of globus pallidus neurons. The average spontaneous firing rate of all the 35 pallidal neurons induced increase in firing rate (basal: 0.59 ± 0.15; CGS21680: 0.56 ± 0.14; n = 7, t = 0.71, df = 6, P > 0.05, paired-samples tested was 17.12 ± 2.24 Hz. Application of 10 μM CGS21680 decreased the frequency of spontaneous firing from 23.35 ± t-test, Figure 2E). Similar to that of 1 μM CGS21680, there was no significant change in CV following micro-pressure ejection of 5.90 to 13.93 ± 4.97 Hz in 9 out of the 35 pallidal neurons CGS21680 at the concentrations of 0.1, 10, and 100 μM (data not (n = 9, t = 6.94, df = 8, P < 0.001, paired-samples t-test). shown). The average decrease was 47.70 ± 5.97%, which was significantly According to our previous study (Chen et al., 2015), we different (t = 8.73, df = 18, P < 0.001, independent-samples t-test) compared to that of vehicle administration (basal: 16.03 classified pallidal neurons into three types (i.e., higher frequency without pauses, higher frequency with pauses and low frequency ± 2.83 Hz; vehicle: 16.42 ± 2.90 Hz; n = 11, t = 1.88, df = 10, P > 0.05, paired-samples t-test). CGS21680 at 10 μM increased with bursts, Figures 2F–H). Based on this classification, the proportion of 36.17% high frequency without pauses, 48.94% the firing rate from 6.78 ± 2.82 to 9.91 ± 3.83 Hz in other 6 out of the 35 pallidal neurons (n = 6, t = 2.63, df = 5, high frequency with pauses and 14.89% low frequency with Frontiers in Physiology | www.frontiersin.org 4 November 2017 | Volume 8 | Article 897 Diao et al. Pallidal Adenosine A Receptor 2A FIGURE 2 | Effects of micro-pressure ejection of adenosine A receptor selective agonist, CGS21680, on the spontaneous firing of pallidal neurons in normal rats. 2A (A) Typical frequency histograms showing that 1 μM CGS21680 decreased the firing rate of a pallidal neuron. (B) In this neuron, 1 μM CGS21680 increased the firing rate slightly. (C) Pooled data summarizing the effects of 1 μM CGS21680 on the firing rate of pallidal neurons in normal rats. The black bars represent the basal firing (Continued) Frontiers in Physiology | www.frontiersin.org 5 November 2017 | Volume 8 | Article 897 Diao et al. Pallidal Adenosine A Receptor 2A FIGURE 2 | Continued rate before CGS21680 application, while the gray bars represent the firing rate after CGS21680 application. **P < 0.01, ***P < 0.001, paired-samples t-test. (D) The graph summarizing the percentages of pallidal neurons with different concentrations (0.1, 1, 10, and 100 μM) of CGS21680-induced decrease in firing rate. (E) Pooled data summarizing the effects of CGS21680 on the CV of pallidal neurons. (F–H) Typical traces showing three firing patterns of globus pallidus neurons, high frequency without pauses (F), high frequency with pauses (G), and low frequency with bursts (H). (I) Pooled data showing the proportion of the three firing pattern types in total neurons and CGS21680-induced inhibitory or excitatory neurons. bursts were observed within the present 47 pallidal neurons. Furthermore, within 21 out of the 47 pallidal neurons with CGS21680-induced decrease in firing rate, these firing pattern types represented a proportion of 28.57, 47.62, and 23.81%, respectively. In another 7 CGS21680-induced excitatory neurons, 42.86, 42.86, and 14.28% exhibited above three firing patterns, respectively. There was no significant difference among the proportions of firing pattern types of the neurons in CGS21680- induced inhibitory or excitatory neurons and the total 47 neurons (x = 1.15, df = 4, P > 0.05, Chi-square test, Figure 2I). Endogenous Adenosine Modulated the Spontaneous Firing of Globus Pallidus Neurons through Adenosine A Receptors 2A in Normal Rats To elucidate the possible modulation of spontaneous pallidal neuronal activity by endogenous adenosine through adenosine A receptors, we explored the effects of selective adenosine A 2A 2A receptor antagonists on the spontaneous firing activity of pallidal neurons. Micro-pressure administration of selective adenosine A receptor antagonist, KW6002 (1 μM), significantly increased 2A the spontaneous firing rate from 12.92 ± 3.98 to 15.95 ± 4.70 Hz in 11 out of the 27 pallidal neurons (n = 11, t = 3.05, df = 10, P < 0.05, paired-samples t-test, Figures 3A,C), with the average increase of 29.48 ± 10.76% (Z = 2.97, P < 0.05 compared to that of vehicle administration, Mann-Whitney test). In 4 neurons, 1 μM KW6002 decreased the firing rate from 9.81 ± 4.50 to 6.82 ± 3.67 Hz. In the remaining 12 neurons the firing rate was not affected (data not shown). Based upon above results, the present study first demonstrated that endogenous adenosine modulated the spontaneous firing of pallidal neurons in vivo. To further provide more evidence, we FIGURE 3 | Effects of adenosine A receptor antagonists, KW6002 and 2A continued to observe the effects of another selective antagonist SCH442416, on the spontaneous firing of pallidal neurons in normal rats. (A) SCH442416 on the pallidal firing. Similar to that of KW6002, Typical frequency histograms showing that 1 μM KW6002 increased the firing rate of a pallidal neuron. (B) In this neuron, 1 μM SCH442416 increased the the predominant effect of SCH442416 (1 μM) was an increase firing rate. (C) Pooled data summarizing the excitatory effects of both KW6002 in the firing rate of globus pallidus neurons. Micro-pressure and SCH442416 on the firing rate of pallidal neurons. The black bars represent administration of 1 μM SCH442416 significantly increased the the basal firing rate before drugs application, while the gray bars represent the spontaneous firing rate from 11.46 ± 2.19 Hz to 15.03 ± 2.58 Hz firing rate after drugs application. *P < 0.05, ***P < 0.001, paired-samples in 13 out of the 22 pallidal neurons (n = 13, t = 6.03, t-test. df = 12, P < 0.001, paired-samples t-test, Figures 3B,C). The average increase was 38.78 ± 8.56%, which was significantly different (Z = 3.91, P < 0.001, Mann-Whitney test) from that of Blockade of Adenosine A Receptors was 2A vehicle ejection. Moreover, we compared the increasing effects of Involved in Exogenous CGS21680-Induced two antagonists, SCH442416 and KW6002, on the spontaneous firing rate of pallidal neurons. The excitatory action of 1 μM Modulation of Pallidal Firing SCH442416 on firing rate was slightly stronger than that of 1 μM Next two adenosine A receptor antagonists KW6002 and 2A KW6002 (t = 2.18, df = 22, P < 0.05, independent-samples SCH442416 were used to further study whether adenosine A 2A t-test). receptors are involved in CGS21680-induced modulation of Frontiers in Physiology | www.frontiersin.org 6 November 2017 | Volume 8 | Article 897 Diao et al. Pallidal Adenosine A Receptor 2A pallidal firing rate. As shown in Figure 4A, 1 μM CGS21680 did not change the firing rate significantly in the presence of 1 μM KW6002. The second time application of 1 μM CGS21680 alone decreased the firing rate which indicated that CGS21680- induced inhibitory effect of firing rate was prevented by KW6002. In total 17 pallidal neurons, CGS21680 only decreased the spontaneous firing rate in 5 out of the 17 pallidal neurons in the presence of 1 μM KW6002. The average decrease (13.93 ± 2.96%) of the firing rate was significantly weaker than that in the absence of 1 μM KW6002 (46.08 ± 5.71%, Z = 2.83, P < 0.01, Mann-Whitney test, Figure 4C). Moreover, SCH442416 also blocked CGS21680-induced decrease of pallidal firing rate (Figures 4B,C). Similar to that of CGS21680-induced inhibitory effects, KW6002 and SCH442416 also blocked the increasing effects of CGS21680 on pallidal firing rate (data not shown). Effects of Adenosine A Receptor 2A Activation on the Spontaneous Firing of Globus Pallidus Neurons in 6-OHDA Hemi-Parkinsonian Rats In this study, we further explored the direct modulation of adenosine A receptors on the spontaneous pallidal firing in 6- 2A OHDA hemi-parkinsonian rats (Figure 5A). The average basal firing rate of pallidal neurons on lesioned sides of 6-OHDA parkinsonian rats was significantly different from that of normal rats or unlesioned sides of parkinsonian rats (F = 8.16, df = 2, P < 0.01, one-way ANOVA). The average basal firing rate of pallidal neurons on the lesioned side of parkinsonian rats (7.94 ± 0.98 Hz, n = 44) was significantly lower than that of normal rats (14.02 ± 1.46 Hz, n = 47, t = 3.48, P < 0.01), as well as that on the unlesioned side of parkinsonian rats (14.33 ± 1.26 Hz, n FIGURE 4 | Both KW6002 and SCH442416 blocked CGS21680-induced = 40, t = 3.51, P < 0.01). Concerning the firing patterns, 34.09% decrease in firing rate. (A) Typical frequency histograms showing that (15/44) of the pallidal neurons recorded exhibited low frequency application of 1 μM CGS21680 in the presence of 1 μM KW6002 did not with bursts on the lesioned side of parkinsonian rats, which was change the firing rate. However, the second time application of CGS21680 more than that of normal rats (7/47, 14.89%, x = 4.57, df = 1, P alone significantly decreased the firing rate in this neuron. (B) Typical frequency histograms showing that 1 μM SCH442416 blocked 1 μM < 0.05, Chi-square test). CGS21680-induced decrease in firing rate. (C) Pooled data summarizing the Furthermore, we observed the effects of micro-pressure effects of both KW6002 and SCH442416 on CGS21680-induced decrease of ejection of CGS21680 (1 μM) on the spontaneous firing activity firing rate. The black bars represent the effects of CGS21680 alone, while the of pallidal neurons in 6-OHDA hemi-parkinsonian rats. On the dark gray bars and light gray bars represent effects of CGS21680 together lesioned side, 1 μM CGS21680 significantly decreased the firing with KW6002 or SCH442416, respectively. *P < 0.05, **P < 0.01, compared to CGS21680 alone group, Mann-Whitney test. rate from 8.14 ± 2.04 to 6.20 ± 1.81 Hz in 7 out of the 19 neurons (n = 7, t = 6.92, df = 6, P < 0.001, paired-samples t-test). The average decrease was 33.04 ± 5.63% (t = 5.99, df = 13, P < 0.01 compared with that of vehicle ejection, independent- Effects of Endogenous Adenosine on the samples t-test). In 2 neurons the frequency increased by 40.18 Spontaneous Firing of Globus Pallidus ± 18.74% and in the left 10 neurons the firing rate was not Neurons through Adenosine A Receptors affected (t = 1.46, df = 16, P > 0.05, independent-samples t- 2A test). On the unlesioned side, 1 μM CGS21680 also exhibited in 6-OHDA Hemi-Parkinsonian Rats bidirectional effects on spontaneous discharge of pallidal neurons Previous studies have found that blockade of adenosine A 2A (data not shown). The CGS21680-induced inhibitory effects on receptors exerts therapeutic potential in Parkinson’s disease. We both lesioned and unlesioned sides of parkinsonian rats were further observed the direct regulation of the spontaneous pallidal not different from that in normal rats (F = 0.68, df = 2, P > firing by two selective adenosine A receptor antagonists, 2A 0.05, one-way ANOVA, Figure 5B). Moreover, similar to that KW6002 and SCH442416, in 6-OHDA hemi-parkinsonian of normal rats, 1 μM CGS21680 didn’t significantly change the rats. pallidal firing pattern on both lesioned and unlesioned sides of We firstly observed the effects of 1 μM KW6002 on the pallidal parkinsonian rats (data not shown). firing rate in 6-OHDA hemi-parkinsonian rats. On the lesioned Frontiers in Physiology | www.frontiersin.org 7 November 2017 | Volume 8 | Article 897 Diao et al. Pallidal Adenosine A Receptor 2A side, the average basal firing rate of 14 neurons recorded was the relationship of adenosine A receptors and dopamine D 2A 2 10.04 ± 2.29 Hz. KW6002 tended to increase the spontaneous receptors in hemi-parkinsonian rats. Similar to that of normal firing rate in 3 out of the 14 pallidal neurons (basal: 15.42 ± rats, the excitatory effects induced by combination of quinpirole 5.51 Hz; KW6002: 18.57 ± 6.34 Hz, n = 3, t = 3.66, df = 2, with KW6002 or SCH442416 tended to be stronger than that of P = 0.07, paired-samples t-test), but there was no statistical adenosine A receptor antagonist alone in hemi-parkinsonian 2A difference. The average increase was 23.72 ± 4.59% (Figure 5B). rats (Figure 6D). In only one neuron, KW6002 decreased the firing rate with Gabaergic Neurotransmission was average decrease of 49.75% and in the left 10 neurons, KW6002 did not significantly change the firing rate (t = 0.71, df = 9, P > Involved in CGS21680-Induced Modulation 0.05, paired-samples t-test). On the unlesioned side, the average of the Spontaneous Firing of Globus basal firing rate of 16 neurons recorded was 14.98 ± 2.19. 1 μM Pallidus Neurons KW6002 significantly increased the firing rate by 26.08 ± 9.41% The competitive GABA receptor antagonist (gabazine) and the in 8 out of the 16 pallidal neurons (t = 2.52, df = 14, P < selective GABA transporter-1 inhibitor (nipecotic acid) were 0.05 compared with that of vehicle group, independent-samples used to further identify the possible involvement of GABAergic t-test), while did not change the firing rate in the remaining neurotransmission in CGS21680-induced decrease of firing 8 neurons (t = 0.34, df = 14, P > 0.05 compared with that rate. As shown in Figures 7A,C, in the presence of gabazine, of vehicle group, independent-samples t-test). In addition, the CGS21680 only decreased the firing rate by 8.04 ± 2.35% in 4 percentage of KW6002 responsive neurons on the lesioned side out of the 13 globus pallidus neurons, which was significantly (28.57%, 4 out of 14) tended to be lower than that in normal weaker than that of the second time application of CGS21680 rats (55.56%, 15 out of 27), and that on the unlesioned side in the absence of gabazine (decrease: 33.59 ± 9.59%, n = 6, (50.00%, 8 out of 16), although there was no statistic difference. Z = 2.35, P < 0.05, Mann-Whitney test). In another set of Further studies revealed that SCH442416 increased the firing rate experiment, nipecotic acid (10 μM) significantly decreased the from 11.31 ± 3.10 Hz to 14.86 ± 3.92 Hz in 9 out of the 21 firing rate from 14.73 ± 2.73 Hz to 6.12 ± 2.28 Hz in 17 out neurons on the lesioned side (n = 9, t = 2.74, df = 8, P < 0.05, of 23 pallidal neurons tested (n = 17, t = 5.45, df = 16, P paired-samples t-test). The average increase was 36.47 ± 10.30% < 0.01, paired-samples t-test, Figure 7B). In the presence of (t = 3.42, df = 13, P < 0.01 compared with that of vehicle group, nipecotic acid, CGS21680 decreased the firing rate by 10.59 ± independent-samples t-test). Furthermore, 1 μM SCH442416- 1.15% in only 2 out of the 6 neurons in which the second induced excitatory effects of pallidal neurons on the lesioned side time application of CGS21680 alone inhibited the firing rate of parkinsonian rats was not significantly different from that of by 44.38 ± 12.62% (Z = 2.00, P < 0.05, Mann-Whitney test, normal rats (t = 0.39, df = 20, P > 0.05, independent-samples Figure 7C). t-test, Figure 5B). CGS21680 Modulated the Spontaneous Effects of Application of Both Adenosine Firing Activity of Globus Pallidus Neurons A Receptor Antagonist and Dopamine D 2A 2 through PKA Pathway Receptor Agonist on the Spontaneous It is known that activation of adenosine A receptors stimulates 2A Firing of Globus Pallidus Neurons in G-protein/AC/cAMP/PKA signaling pathway. We determined Normal and Hemi-Parkinsonian Rats whether selective PKA inhibitor, H-89, blocked adenosine A 2A Next, extracellular recordings were used to further explore the receptor-induced modulation of firing activity of globus pallidus interactions between adenosine A receptors and dopamine neurons. In one set of experiment, micro-pressure administration 2A D receptors in normal and hemi-parkinsonian rats. Firstly, we of H-89 (10 μM) significantly increased the spontaneous firing observed the relationship between adenosine A receptors and rate from 7.05 ± 1.43 Hz to 13.36 ± 2.65 Hz in 12 out of 2A dopamine D receptors in normal rats. Adenosine A receptor 15 pallidal neurons tested (n = 12, t = 3.74, df = 11, P < 2 2A antagonist was applied first. In the same neuron, this antagonist 0.01, paired-samples t-test, Figure 8A). The average increase was was applied again in the presence of selective dopamine D 107.29 ± 26.29%. This effect occurred within 15 min after H- receptor agonist quinpirole. As shown in Figures 6A,C, the first- 89 injection and lasted over 30 min. In the remaining 3 pallidal time application of adenosine A receptor antagonist KW6002 neurons, H-89 did not alter the firing rate significantly (t = 2A (1 μM) alone slightly increased the pallidal firing rate by 19.14 ± 3.03, df = 2, P > 0.05, paired-samples t-test). In another 9.51% in 9 out of 15 pallidal neurons. However, in the presence set of experiments, CGS21680 (1 μM) was applied first to of quinpirole (1 mM), the second time application of KW6002 pallidal neurons. In neurons which were inhibited by CGS21680, significantly increased the spontaneous firing rate by 67.33 ± CGS21680 was applied again in the presence of H-89 (10 μM). 14.09% in the same 9 neurons, which was stronger than that As shown in Figures 8B,C, CGS21680 alone decreased the in the absence of quinpirole (Z = 2.67, P < 0.01, Wilcoxon spontaneous firing rate from 11.40 ± 2.15 to 7.78 ± 1.87 Hz signed-rank test). In another set of experiment, quinpirole in 7 out of 13 pallidal neurons (n = 7, t = 3.11, df = 6, enhanced SCH442416-induced increase of pallidal firing rate P < 0.05, paired-samples t-test). The average decrease was from 23.00 ± 12.12% to 57.06 ± 23.90% (Z = 2.20, P < 0.05, 31.22 ± 7.11%. In the presence of H-89, the second time Wilcoxon signed-rank test, Figures 6B,C). Secondly, we explored application of CGS21680 did not cause significant change in Frontiers in Physiology | www.frontiersin.org 8 November 2017 | Volume 8 | Article 897 Diao et al. Pallidal Adenosine A Receptor 2A FIGURE 5 | Comparison of the electrophysiological effects of adenosine A receptor agonist and antagonist in both normal and parkinsonian rats. (A) Confirmation 2A of 6-OHDA hemi-parkinsonian rat model. Fluorescent images revealing tyrosine hydroxylase (TH) immunostaining in the substantia nigra pars compacta (SNc) of both normal (a) and 6-OHDA hemi-parkinsonian (b) rats. Scale bars = 100 μm. (B) Comparison of the effects of 1 μM CGS21680, 1 μM KW6002, and 1 μM SCH442416 on the spontaneous firing rate of pallidal neurons between normal (black bars) and 6-OHDA parkinsonian rats (gray bars). ns, not significant. spontaneous firing rate (1.97 ± 1.53%, Z = 3.23, P < 0.01 (72.50 ± 3.13%, n = 8, Z = 2.81, P < 0.01 compared to compared with that of CGS21680 alone, Wilcoxon signed-rank that of vehicle administration and 71.67 ± 6.00%, n = 6, test). Z = 2.06, P < 0.05, respectively, Mann-Whitney test, Figure 9A). Secondly, the percentages of swing responses in awake parkinsonian rats were studied. In line with previous Asymmetrical Motor Behavior Induced by finding (Abrous et al., 1998), a strong lesion-induced ipsilateral Pallidal Adenosine A Receptors in Awake 2A bias was observed in present unilateral 6-OHDA-lesioned Normal and 6-OHDA Hemi-Parkinsonian rats (98.95 ± 0.72%, n = 19). Unilateral microinjection of vehicle into the globus pallidus of lesioned side did not Rats As adenosine A receptors could modulate the firing rate of alter lesion-induced ipsilateral bias (98.33 ± 1.67%, n = 6). 2A The percentage of lesion-induced bias was not affected pallidal neurons at single cellular level, we further explored the behavioral effects of modulating adenosine A receptors in by administration of KW6002 (93.75 ± 2.63%, n = 8, 2A Z = 1.07, P > 0.05, Mann-Whitney test) or SCH442416 the globus pallidus by EBST at the integral level. Firstly, the swing responses induced by adenosine A receptor agonist (93.33 ± 2.36%, n = 9, Z = 1.11, P > 0.05, Mann-Whitney 2A test). Furthermore, the rats were intrapallidally injected with and antagonists in awake normal rats were observed. The rats receiving unilateral vehicle administration displayed unbiased KW6002 (1 μM) or SCH442416 (1 μM) 30 min after quinpirole administration (0.05 mg/kg, s.c.). Quinpirole alone did not swings (53.33 ± 3.33%, n = 6). Unilateral microinjection of CGS21680 (1 μM) into the globus pallidus significantly induced alter lesion-induced ipsilateral bias (98.00 ± 1.49%, n = 5). strong ipsilateral bias (85.00 ± 4.23%, n = 8, Z = 3.13, P < Co-administration of quinpirole with KW6002 or SCH442416 significantly decreased lesion-induced ipsilateral biased swing 0.01 compared to that of vehicle administration, Mann-Whitney test, Figure 9A). In contrast to that of CGS21680, unilateral (63.75 ± 5.32%, n = 8, Z = 2.99, P < 0.01 and 64.44 ± 4.12%, n = 9, Z = 3.27, P < 0.01, respectively, Mann-Whitney test, microinjection of 1 μM KW6002 and 1 μM SCH442416 exhibited biased swings contralateral to the drug-injection side Figure 9B). Frontiers in Physiology | www.frontiersin.org 9 November 2017 | Volume 8 | Article 897 Diao et al. Pallidal Adenosine A Receptor 2A FIGURE 7 | GABAergic transmission was involved in CGS21680-induced inhibition of the firing activity of globus pallidus neurons in normal rats. (A) Typical frequency histogram showing that 0.1 mM gabazine blocked CGS21680-induced decrease of firing rate of globus pallidus neuron. (B) Typical frequency histogram showing that in the presence of 10 μM nipecotic acid, CGS21680 did not induce clear decrease of firing rate. However, after long time recovery, the second time application of CGS21680 alone decreased the firing rate in the same neuron. (C) Pooled data summarizing CGS21680-induced decrease of firing rate in the presence (black bars) and absence (gray bars) of gabazine or nipecotic acid. *P < 0.05, Mann-Whitney test. Expression of Adenosine A Receptors 2A FIGURE 6 | Quinpirole enhanced KW6002 and SCH44216-induced increase and Parvalbumin in the Globus Pallidus in firing rate of pallidal neurons. (A) The first time application of 1 μM KW6002 Neurons alone increased the pallidal firing rate by 48.86% in this neuron of normal rat. In the presence of 1 μM quinpirole, the second time application of KW6002 The present immunostaining showed that adenosine A 2A significantly increased the firing rate by 78.07% in this neuron. (B) Typical receptors were expressed in the globus pallidus of both frequency histograms showing that quinpirole enhanced 1 μM normal and 6-OHDA parkinsonian rats. Average number of SCH44216-induced excitation of firing rate in this neuron. (C) Pooled data pallidal adenosine A receptor-positive neurons per slice in 2A summarizing the effects of quinpirole on KW6002 and SCH44216-induced normal rats was 34.96 ± 1.88, which was not significantly increase of firing rate in normal rats. (D) Pooled data summarizing the effects of quinpirole on KW6002 and SCH44216-induced increase of firing rate in different from that in 6-OHDA parkinsonian rats (36.04 parkinsonian rats. The black bars represent the effects of KW6002 or ± 2.39, Z = 0.73, P > 0.05, Wilcoxon signed-rank test, SCH44216 alone, while the gray bars represent the effects of KW6002 or Figures 10A–F). Moreover, the cellular location of adenosine SCH44216 together with quinpirole. *P < 0.05, **P < 0.01, ns, not significant, A receptors and parvalbumin in the globus pallidus was 2A paired-samples t-test. studied by using double immunofluorescence labeling. The Frontiers in Physiology | www.frontiersin.org 10 November 2017 | Volume 8 | Article 897 Diao et al. Pallidal Adenosine A Receptor 2A FIGURE 8 | PKA pathway was involved in CGS21680-induced change of pallidal firing activity in normal rats. (A) Application of H-89 (10 μM) significantly increased the firing rate. (B) The first time application of 1 μM CGS21680 decreased the firing rate. In the presence of H-89, the second time application FIGURE 9 | Evaluation of modulatory effects of pallidal adenosine A 2A of CGS21680 did not alter the firing rate in this cell. (C) Pooled data showing receptors on body asymmetry in awake normal and 6-OHDA that H-89 significantly blocked CGS21680-induced decrease of firing rate in hemi-parkinsonian rats using elevated body swing test. (A) In normal rats, pallidal neurons tested. The black bars and gray bars represent the effects of intrapallidal microinjection of CGS21680 (1 μM) induced ipsilateral-biased CGS21680 in the absence and presence of H89, respectively. **P < 0.01, swing, while KW6002 (1 μM) and SCH44216 (1 μM) induced Wilcoxon signed-rank test. contralateral-biased swing. (B) In 6-OHDA hemi-parkinsonian rats, unilateral microinjection of KW6002 or SCH44216 into the globus pallidus of lesioned side did not alter 6-OHDA-induced ipsilateral biased swing significantly, while co-application of KW6002 or SCH44216 together with quinpirole (0.05 mg/kg, s.c.) decreased the ipsilateral biased swing. *P < 0.05, **P < 0.01 compared result showed that adenosine A receptors were expressed 2A to vehicle (saline) group, Mann-Whitney test. in both parvabumin-positive and parvabumin-negative neurons (Figures 10G–I). pallidal firing through GABA receptors in both rats and mice DISCUSSION (Xue et al., 2010; Chen et al., 2013). The bulk of evidence has indicated that presynaptic stimulation of adenosine A 2A Activation of Adenosine A Receptors 2A receptors bidirectionally regulates GABA release from globus Inhibits the Spontaneous Firing of Pallidal pallidus slices in rats. For example, Floran et al. (2005) show Neurons through Facilitation of Gabaergic that CGS21680 evokes GABA release from rat globus pallidus Neurotransmission slices at a wide range of concentration (10 nM to 10 μM). It The present electrophysiological results showed that activation has been reported that adenosine A receptor agonist exerts 2A of adenosine A receptors by CGS21680 mainly exerted facilitatory effects on GABA release at lower concentrations, 2A inhibitory effects on the spontaneous firing of the globus but inhibitory effects at higher concentrations (Mayfield et al., pallidus neurons which were blocked by adenosine A receptor 1993; Dayne Mayfield et al., 1996; Morales-Figueroa et al., 2A antagonists. It is well-known that the globus pallidus receives 2014). The present in vivo extracellular recordings demonstrated dense GABAergic innervation originating from the striatum. Our that pre-application of GABA receptor antagonist, gabazine, previous studies have shown that endogenous GABA modulates blocked CGS21680-induced decrease of firing rate suggesting Frontiers in Physiology | www.frontiersin.org 11 November 2017 | Volume 8 | Article 897 Diao et al. Pallidal Adenosine A Receptor 2A FIGURE 10 | Double immunostaining for adenosine A receptors and parvalbumin in the globus pallidus. Fluorescence photomicrographs showing the expression 2A of adenosine A receptors in the pallidal neurons in both normal (A–C) and 6-OHDA parkinsonian (D–F) rats. Confocal laser scanning photomicrographs (G,H,I) 2A showing the expression of adenosine A receptors (green) and parvalbumin (red), and the overlapping expression (yellow) in normal rats. The arrow indicated 2A non-overlap of parvalbumin and adenosine A receptors in this cell. Scale bars = 100 μm in (A,D); 50 μm in (B–I); 25 μm in (C,F). 2A that enhancement of GABAergic neurotransmission may be of pallidal neurons, which was consistent with previous study involved in activation of adenosine A receptor-induced (Shindou et al., 2002). 2A inhibition of firing activity in the globus pallidus. GABA transporter-1 (GAT-1) is the prominent transporter in rat Endogenous Adenosine Modulates the globus pallidus. Gonzalez et al. (2006) found that stimulation Activity of Pallidal Neurons through A 2A of adenosine A receptors increases GABA level through 2A Receptors, Which May Be Associated with inhibition of GAT-mediated GABA uptake. Therefore, the high level of extracellular GABA in the globus pallidus may be the Heteromers with Dopamine D Receptors major possible mechanism of adenosine A receptor-induced Present in vivo electrophysiological and behavioral studies 2A inhibition of spontaneous discharge of pallidal neurons. Our first demonstrated that application of adenosine A receptor 2A in vivo electrophysiological studies further revealed that in antagonists increased the spontaneous firing rate of pallidal the presence of the selective GAT-1 inhibitor, nipecotic acid, neurons and induced contralateral biased swing of rats, CGS21680 could not induce inhibitory effect on the spontaneous suggesting that endogenous adenosinergic system is involved firing of the globus pallidus neurons. These findings enable in the regulation of the firing activity of the globus pallidus us to further identify that blockade of GABA transporter- and motor behavior in normal rats. Adenosine A receptors 2A induced enhancement of GABAergic neurotransmission could be divided into two proposed populations based on may be involved in activation of adenosine A receptor- whether forming heteromers with dopamine D receptors (Ferre 2A 2 mediated inhibition of pallidal neurons. Moreover, the present et al., 2008; Orru et al., 2011). The different pharmacological results suggested that activation of cAMP/PKA pathway was properties of the two antagonists, KW6002 and SCH442416, involved in CGS21680-induced modulation of firing activity depend on their affinities to different population of adenosine Frontiers in Physiology | www.frontiersin.org 12 November 2017 | Volume 8 | Article 897 Diao et al. Pallidal Adenosine A Receptor 2A A receptors. KW6002 has high affinity for adenosine A and the expression of molecular markers (Cooper and Stanford, 2A 2A receptors forming heteromers with dopamine D receptors, 2000; Benhamou et al., 2012; Mallet et al., 2012; Hernandez while SCH442416 shows very low affinity for adenosine A et al., 2015; Karain et al., 2015). The CGS21680-induced 2A receptors co-expressed with dopamine D receptors. The present bidirectional effects (inhibition or excitation) may be associated electrophysiological study revealed that two populations of with the diversity of pallidal neurons. The GABAergic globus adenosine A receptors both are involved in endogenous pallidus neurons are classificated into parvalbumin-positive 2A adenosine-induced modulation of pallidal neuronal activity. and parvalbumin -negative neurons based on the expression Moreover, the present electrophysiological results also illustrated of the calcium-binding protein parvalbumin (Saunders et al., that the effects of SCH442416 on pallidal firing (38.78 ± 8.56%) 2016). Parvalbumin-positive neurons represent the majority of were slightly stronger than that of KW6002 (29.48 ± 10.76%). “prototypic” pallidal neurons which exhibit fast and regular firing The percentage of SCH442416-induced excitatory neurons (13 spontaneous activity and innervate primarily the subthalamic out of 22, 59.09%) was a bit higher than that of KW6002 (11 out nucleus. Most parvalbumin-negative neurons are “arkypallidal” of 27, 40.74%). Early studies have demonstrated that anesthesia pallidal neurons which exhibit slower and more irregular especially urethane reduces dopamine release (Kelland et al., spontaneous activity and project strongly back to dorsal 1989; Hamilton et al., 1992). Therefore, a possible explanation striatum. Based on the classification of firing patterns, the for the different intensity of the two antagonists may be that present electrophysiological study showed that CGS21680- anesthetic reduces dopamine release and then inhibits the activity induced decrease or increase of firing activity was observed in of endogenous adenosine A receptor/dopamine D receptor all the three types of pallidal neurons. Additionally, the present 2A 2 complex. Moreover, the present study showed that the percentage double immunostaining showed that adenosine A receptors 2A of KW6002-responsive neurons (4 out of 14, 28.57%) on the are located on both parvalbumin-positive and parvalbumin- lesioned side of hemiparkinsonian rats tended to be lower than negative pallidal neurons. Unfortunately, for the technical that of normal rats (15 out of 27, 55.56%), as well as that on limitation, we could not label the recorded neurons and further the unlesioned side (8 out of 16, 50.00%). The weaker effect identify the types of pallidal neurons precisely. of KW6002 in dopamine-denervated side may be related to the lower extent of dopamine binding to D receptors. Additionally, Adenosine A Receptors are Functional in Suarez et al. (2016) demonstrates that dopamine depletion 2A 6-OHDA Parkinsonian Rats reduces dendritic spines of striatal medium spiny neurons expressing dopamine D receptors and the loss is accompanied The present extracellular recordings displayed lower basal with a decrease in synaptic strength. Adenosine A receptor- firing rate of pallidal neurons after 6-OHDA lesions. However, 2A dopamine D receptor heteromers are selectively localized on the some studies show inconsistent changes of firing rates in terminals of striatal medium spiny neurons in the globus pallidus parkinsonian models (Zold et al., 2007; Ellens and Leventhal, (Floran et al., 2005). Whether the weaker effect of KW6002 is 2013). According to Leblois et al. (2006), basal ganglia neurons related to the morphology changes of synaptic spine remains to would not exhibit abnormal firing activity unless extensive be explored. dopamine depletion is produced. The present immunostaining indicated that the number of hydroxylase-immunoreactive cells in lesioned substantia nigra pars compacta of hemiparkinsonian Activation of Adenosine A Receptors 2A rats decreased to about 17%. Therefore, the decrease of pallidal Induces a Weak Excitation in Partial firing rates may be associated with severe dopamine depletion Pallidal Neurons in substantia nigra of hemiparkinsonian rats. Inhibition of The present study also showed that activation of adenosine A the globus pallidus may contribute to impeding voluntary 2A receptors produced a weak excitation in a few parts of pallidal movement in Parkinson’s disease through indirect pathway neurons. In addition to GABAergic innervation, the globus of the basal ganglia circuits, and ultimately inhibiting the pallidus receives cholinergic innervation from the brainstem activities of thalamic motor nuclei and motor cortex. The pedunculopontine tegmental nucleus (Eid et al., 2016). Nicotine present adenosine A receptor antagonists-induced increase 2A acetylcholine increases spiking rate of pallidal neurons (Rios of discharge frequency indicated that adenosine A receptor 2A et al., 2016). It has been reported that activation of adenosine antagonists may contribute to alleviating motor symptoms in A receptors by CGS 21680 facilitates acetylcholine release Parkinson’s disease by normalizing the firing rate of pallidal 2A in some brain regions including hippocampus and striatum neurons. Furthermore, the CGS21680-induced inhibition of (Kurokawa et al., 1994; Ribeiro et al., 1996; Rebola et al., pallidal firing rate in present parkinsonian rats is not significantly 2002). Although, no morphological study shows the expression different from that in normal rats. Previous studies have revealed of adenosine A receptors on cholinergic nerve terminals in changes of adenosine A receptor expression in discrete brain 2A 2A the globus pallidus, one may hypothesize that adenosine A regions of patients dying with Parkinson’s disease as well as 2A receptor-mediated possible modulation of acetylcholine release Parkinson’s animal models (Hurley et al., 2000; Villar-Menendez may be responsible for CGS21680-induced weak excitatory et al., 2014). For example, Hurley et al. (2000) reported the effects. decreasing level of adenosine A receptor mRNA in some 2A It is well-known that the globus pallidus neurons are diverse striatal regions, increasing in the substantia nigra and no change in electrophysiology, axonal projections, dendritic morphology, in any other brain regions examined. Similarly, no clear change of Frontiers in Physiology | www.frontiersin.org 13 November 2017 | Volume 8 | Article 897 Diao et al. Pallidal Adenosine A Receptor 2A the expression of pallidal adenosine A receptors was observed parkinsonian rats which therefore induces imbalance of bilateral 2A in present morphological study, which may be one of the possible movement output. It was reported that tail pinch may further reasons for the present similar electrophysiological results under cause dopamine release in the unlesioned striatum (Sindhu both normal and parkinsonian states. et al., 2005). Similar to the above supposition, application of adenosine A receptor antagonists in the presence of 2A Pallidal Adenosine A Receptors quinpirole could alleviate 6-OHDA-induced biased swing, which 2A further verified that pallidal adenosine A receptors play Modulate Motor Behavior in Awake Rats 2A important roles in the therapy of motor symptoms in Parkinson’s As adenosine A receptors modulated pallidal firing, we 2A disease. hypothesized that pharmacological manipulation of adenosine In conclusion, the present study indicated that pallidal A receptors in the globus pallidus maybe participate in 2A adenosine A receptors play prominent roles in motor motor modulation in awake rats. EBST is a simple, sensitive 2A modulation under both healthy and parkinsonian states, which and accurate behavioral test used to evaluate asymmetrical further verified that pallidal adenosine A receptor is potentially behavior in animals with a unilateral cerebral lesion such as 2A useful in the treatment of Parkinson’s disease. Therefore, more Parkinson’s disease, Huntington’s disease and ischemic stroke studies will be needed to explore the functions of pallidal (Borlongan et al., 1995; Baluchnejadmojarad and Roghani, 2004; adenosine A receptors in both heath and disease. Tabuse et al., 2010; Ingberg et al., 2015). The asymmetric 2A swing behavior has been attributed to the imbalance of motor control in the basal ganglia circuit. In present study, ETHICS STATEMENT we observed that unilateral microinjection of adenosine A 2A This study was carried out in accordance with the receptor agonist or antagonist into the globus pallidus displayed recommendations of the University guidelines on animal significant biased swing behavior ipsilaterally or contralaterally, ethics. The protocol was approved by an Animal Ethics respectively. Unilateral microinjection of adenosine A receptor 2A Committee of Qingdao University. agonist produces hypoactivity of the globus pallidus and then disinhibits the subthalamic nucleus. The thalamocortical activity is suppressed by enhanced GABAergic inhibition from the output AUTHOR CONTRIBUTIONS nucleus of the basal ganglia. Subsequently, the hypoactivity of ipsilateral motor cortex leads to the imbalance of the activity H-LD performed experiments and wrote the draft. H-LD, YX, X- HH, S-YW, CL, and W-FC analyzed the data. H-LD wrote the of bilateral limb muscles. The mechanism of adenosine A 2A manuscript. LC designed and supervised the project. receptor antagonist-induced contralateral swing is just contrary to that of adenosine A receptor agonist. Thus, the present 2A behavioral test suggested that pallidal adenosine A receptors ACKNOWLEDGMENTS 2A are involved in motor regulation. Moreover, the strong ipsilateral This work was supported by the grants from the National Natural bias was observed in present unilateral 6-OHDA-lesioned rats. Science Foundation of China (31671076), the Bureau of Science The possible mechanism may be that dopamine degeneration and Technology of Qingdao (14-2-3-1-nsh). decreases the activity of motor cortex on the lesioned side of REFERENCES Chan, C. S., Glajch, K. E., Gertler, T. S., Guzman, J. N., Mercer, J. N., Lewis, A. S., et al. (2011). HCN channelopathy in external globus pallidus neurons in models Abrous, D. N., Rodriguez, J. J., Montaron, M. F., Aurousseau, C., Le Moal, of Parkinson’s disease. Nat. Neurosci. 14, 85–92. doi: 10.1038/nn.2692 M., and Barneoud, P. (1998). Behavioural recovery after unilateral Chen, L., Xu, R., Sun, F. J., Xue, Y., Hao, X. M., Liu, H. X., et al. (2015). lesion of the dopaminergic mesotelencephalic pathway: effect of Hyperpolarization-activated cyclic nucleotide-gated (HCN) channels regulate repeated testing. Neuroscience 84, 213–221. doi: 10.1016/S0306-4522(97) firing of globus pallidus neurons in vivo. Mol. Cell. Neurosci. 68, 46–55. 00498-3 doi: 10.1016/j.mcn.2015.04.001 Baluchnejadmojarad, T., and Roghani, M. (2004). Evaluation of functional Chen, X. Y., Xue, Y., Wang, H., Zhu, S. H., Hao, X. M., and Chen, L. (2013). asymmetry in rats with dose-dependent lesions of dopaminergic nigrostriatal Modulation of firing activity by endogenous GABAA receptors in the globus system using elevated body swing test. Physiol. Behav. 82, 369–373. pallidus of MPTP-treated parkinsonian mice. Neurosci. Bull. 29, 701–707. doi: 10.1016/j.physbeh.2004.04.005 doi: 10.1007/s12264-013-1351-x Benhamou, L., Bronfeld, M., Bar-Gad, I., and Cohen, D. (2012). Globus Cooper, A. J., and Stanford, I. M. (2000). Electrophysiological and morphological pallidus external segment neuron classification in freely moving rats: a characteristics of three subtypes of rat globus pallidus neurone in vitro. J. comparison to primates. PLoS ONE 9:e45421. doi: 10.1371/journal.pone.00 Physiol. 527(Pt 2), 291–304. doi: 10.1111/j.1469-7793.2000.t01-1-00291.x 45421. Cunha, R. A. (2016). How does adenosine control neuronal dysfunction and Bibbiani, F., Oh, J. D., Petzer, J. P., Castagnoli, N. Jr., Chen, J. F., Schwarzschild, neurodegeneration? J. Neurochem. 139, 1019–1055. doi: 10.1111/jnc.13724 M. A., et al. (2003). A2A antagonist prevents dopamine agonist-induced motor Dayne Mayfield, R., Larson, G., Orona, R. A., and Zahniser, N. R. (1996). complications in animal models of Parkinson’s disease. Exp. Neurol. 184, Opposing actions of adenosine A2a and dopamine D2 receptor activation 285–294. doi: 10.1016/S0014-4886(03)00250-4 on GABA release in the basal ganglia: evidence for an A2a/D2 receptor Borlongan, C. V., Randall, T. S., Cahill, D. W., and Sanberg, P. R. (1995). interaction in globus pallidus. Synapse 22, 132–138. doi: 10.1002/(SICI)1098- Asymmetrical motor behavior in rats with unilateral striatal excitotoxic 2396(199602)22:2<132::AID-SYN6>3.0.CO2-E lesions as revealed by the elevated body swing test. Brain Res. 676, 231–234. Dodson, P. D., Larvin, J. T., Duffell, J. M., Garas, F. N., Doig, N. M., Kessaris, doi: 10.1016/0006-8993(95)00150-O N., et al. (2015). Distinct developmental origins manifest in the specialized Frontiers in Physiology | www.frontiersin.org 14 November 2017 | Volume 8 | Article 897 Diao et al. Pallidal Adenosine A Receptor 2A encoding of movement by adult neurons of the external globus pallidus. Neuron Kita, H., and Kita, T. (2001). Number, origins, and chemical types of 86, 501–513. doi: 10.1016/j.neuron.2015.03.007 rat pallidostriatal projection neurons. J. Comp. Neurol. 437, 438–448. Eid, L., Parent, A., and Parent, M. (2016). Asynaptic feature and heterogeneous doi: 10.1002/cne.1294 distribution of the cholinergic innervation of the globus pallidus in Kurokawa, M., Kirk, I. P. Kirkpatrick, K. A., Kase, H., and Richardson, primates. Brain Struct. Funct. 221, 1139–1155. doi: 10.1007/s00429-014- P. J. (1994). Inhibition by KF17837 of adenosine A2A receptor-mediated 0960-0 modulation of striatal GABA and ACh release. Br. J. Pharmacol. 113, 43–48. Ellens, D. J., and Leventhal, D. K. (2013). Review: electrophysiology of basal doi: 10.1111/j.1476-5381.1994.tb16171.x ganglia and cortex in models of Parkinson disease. J. Parkinsons Dis. 3, 241–254. Leblois, A., Boraud, T., Meissner, W., Bergman, H., and Hansel, D. doi: 10.3233/jpd-130204 (2006). Competition between feedback loops underlies normal and Ferre, S., Quiroz, C., Woods, A. S., Cunha, R., Popoli, P., Ciruela, F., pathological dynamics in the basal ganglia. J. Neurosci. 26, 3567–3583. et al. (2008). An update on adenosine A2A-dopamine D2 receptor doi: 10.1523/JNEUROSCI.5050-05.2006 interactions: implications for the function of G protein-coupled Mallet, N., Micklem, B. R., Henny, P., Brown, M. T., Williams, C., Bolam, J. P., et al. receptors. Curr. Pharm. Des. 14, 1468–1474. doi: 10.2174/1381612087844 (2012). Dichotomous organization of the external globus pallidus. Neuron 74, 80108 1075–1086. doi: 10.1016/j.neuron.2012.04.027 Floran, B., Gonzalez, B., Floran, L., Erlij, D., and Aceves, J. (2005). Interactions Martinez-Mir, M. I., Probst, A., and Palacios, J. M. (1991). Adenosine A2 receptors: between adenosine A(2a) and dopamine D2 receptors in the control of selective localization in the human basal ganglia and alterations with disease. [(3)H]GABA release in the globus pallidus of the rat. Eur. J. Pharmacol. 520, Neuroscience 42, 697–706. doi: 10.1016/0306-4522(91)90038-P 43–50. doi: 10.1016/j.ejphar.2005.06.035 Mayfield, R. D., Suzuki, F., and Zahniser, N. R. (1993). Adenosine A2a Fredholm, B. B., IJzerman, A. P., Jacobson, K. A., Klotz, K. N., and Linden, receptor modulation of electrically evoked endogenous GABA release J. (2001). International Union of Pharmacology. XXV. Nomenclature and from slices of rat globus pallidus. J. Neurochem. 60, 2334–2337. classification of adenosine receptors. Pharmacol. Rev. 53, 527–552. doi: 10.1111/j.1471-4159.1993.tb03526.x Fredholm, B. B., IJzerman, A. P., Jacobson, K. A., Linden, J., and Muller, C. Morales-Figueroa, G. E., Marquez-Gomez, R., Gonzalez-Pantoja, R., Escamilla- E. (2011). International Union of Basic and Clinical Pharmacology. LXXXI. Sanchez, J., and Arias-Montano, J. A. (2014). Histamine H3 receptor activation Nomenclature and classification of adenosine receptors–an update. Pharmacol. counteracts adenosine A2A receptor-mediated enhancement of depolarization- Rev. 63, 1–34. doi: 10.1124/pr.110.003285 evoked [3H]-GABA release from rat globus pallidus synaptosomes. ACS Chem. Gage, G. J., Stoetzner, C. R., Wiltschko, A. B., and Berke, J. D. (2010). Selective Neurosci. 5, 637–645. doi: 10.1021/cn500001m activation of striatal fast-spiking interneurons during choice execution. Neuron Mori, A., and Shindou, T. (2003). Modulation of GABAergic transmission in 67, 466–479. doi: 10.1016/j.neuron.2010.06.034 the striatopallidal system by adenosine A2A receptors: a potential mechanism Gonzalez, B., Paz, F., Floran, L., Aceves, J., Erlij, D., and Floran, B. (2006). for the antiparkinsonian effects of A2A antagonists. Neurology 61, S44–S48. Adenosine A2A receptor stimulation decreases GAT-1-mediated GABA doi: 10.1212/01.WNL.0000095211.71092.A0 uptake in the globus pallidus of the rat. Neuropharmacology 51, 154–159. Obeso, J. A., Rodriguez-Oroz, M. C., Benitez-Temino, B., Blesa, F. J., Guridi, J., doi: 10.1016/j.neuropharm.2006.03.011 Marin, C., et al. (2008). Functional organization of the basal ganglia: therapeutic Hamilton, M. E., Mele, A., and Pert, A. (1992). Striatal extracellular implications for Parkinson’s disease. Mov. Disord. 23(Suppl. 3), S548–S559. dopamine in conscious vs. anesthetized rats: effects of chloral hydrate doi: 10.1002/mds.22062 anesthetic on responses to drugs of different classes. Brain Res. 597, 1–7. Orru, M., Quiroz, C., Guitart, X., and Ferre, S. (2011). Pharmacological doi: 10.1016/0006-8993(92)91498-4 evidence for different populations of postsynaptic adenosine A2A Hegeman, D. J., Hong, E. S., Hernandez, V. M., and Chan, C. S. (2016). The receptors in the rat striatum. Neuropharmacology 61, 967–974. external globus pallidus: progress and perspectives. Eur. J. Neurosci. 43, doi: 10.1016/j.neuropharm.2011.06.025 1239–1265. doi: 10.1111/ejn.13196 Paxinos, G., and Watson, C. (1986). The Rat Brain in Stereotaxic Coordinates. New Hernandez, V. M., Hegeman, D. J., Cui, Q., Kelver, D. A., Fiske, M. P., Glajch, K. E., York, NY: Academic Press. et al. (2015). Parvalbumin+ neurons and npas1+ neurons are distinct neuron Preti, D., Baraldi, P. G., Moorman, A. R., Borea, P. A., and Varani, K. (2015). classes in the mouse external globus pallidus. J. Neurosci. 35, 11830–11847. History and perspectives of A2A adenosine receptor antagonists as potential doi: 10.1523/JNEUROSCI.4672-14.2015 therapeutic agents. Med. Res. Rev. 35, 790–848. doi: 10.1002/med.21344 Hurley, M. J., Mash, D. C., and Jenner, P. (2000). Adenosine A(2A) receptor Querejeta, E., Martinez-Romero, B., Miranda, J. E., and Delgado, A. (2010). mRNA expression in Parkinson’s disease. Neurosci. Lett. 291, 54–58. Modulation of the striato-pallidal pathway by adenosine A2a receptors doi: 10.1016/S0304-3940(00)01371-9 depends on dopaminergic striatal input. Brain Res. 1349, 137–142. Ingberg, E., Gudjonsdottir, J., Theodorsson, E., Theodorsson, A., and Strom, doi: 10.1016/j.brainres.2010.06.040 J. O. (2015). Elevated body swing test after focal cerebral ischemia Raz, A., Vaadia, E., and Bergman, H. (2000). Firing patterns and correlations of in rodents: methodological considerations. BMC. Neurosci. 16, 50. spontaneous discharge of pallidal neurons in the normal and the tremulous 1- doi: 10.1186/s12868-015-0189-8 methyl-4-phenyl-1,2,3,6-tetrahydropyridine vervet model of parkinsonism. J. Jarvis, M. F., and Williams, M. (1989). Direct autoradiographic localization of Neurosci. 20, 8559–8571. adenosine A2 receptors in the rat brain using the A2-selective agonist, [3H]CGS Rebola, N., Oliveira, C. R., and Cunha, R. A. (2002). Transducing system 21680. Eur. J. Pharmacol. 168, 243–246. doi: 10.1016/0014-2999(89)90571-2 operated by adenosine A(2A) receptors to facilitate acetylcholine Jellinger, K. A. (1991). Pathology of Parkinson’s disease. Changes other release in the rat hippocampus. Eur. J. Pharmacol. 454, 31–38. than the nigrostriatal pathway. Mol. Chem. Neuropathol. 14, 153–197. doi: 10.1016/S0014-2999(02)02475-5 doi: 10.1007/BF03159935 Ribeiro, J. A., Cunha, R. A., Correia-de-Sa, P., and Sebastiao, A. M. (1996). Jenner, P. (2014). An overview of adenosine A2A receptor Purinergic regulation of acetylcholine release. Prog. Brain. Res. 109, 231–241. antagonists in Parkinson’s disease. Int. Rev. Neurobiol. 119, 71–86. doi: 10.1016/S0079-6123(08)62107-X doi: 10.1016/B978-0-12-801022-8.00003-9 Rios, A., Barrientos, R., Alatorre, A., Delgado, A., Perez-Capistran, T., Chuc- Karain, B., Xu, D., Bellone, J. A., Hartman, R. E., and Shi, W. X. (2015). Rat globus Meza, E., et al. (2016). Dopamine-dependent modulation of rat globus pallidus pallidus neurons: functional classification and effects of dopamine depletion. excitation by nicotine acetylcholine receptors. Exp. Brain. Res. 234, 605–616. Synapse 69, 41–51. doi: 10.1002/syn.21783 doi: 10.1007/s00221-015-4491-6 Kelland, M. D., Freeman, A. S., and Chiodo, L. A. (1989). Chloral hydrate Rosin, D. L., Hettinger, B. D., Lee, A., and Linden, J. (2003). Anatomy of anesthesia alters the responsiveness of identified midbrain dopamine adenosine A2A receptors in brain: morphological substrates for integration neurons to dopamine agonist administration. Synapse 3, 30–37. of striatal function. Neurology 61, S12–S18. doi: 10.1212/01.WNL.0000095205. doi: 10.1002/syn.890030105 33940.99 Kita, H. (2007). Globus pallidus external segment. Prog. Brain. Res. 160, 111–133. Sani, S., Ostrem, J. L., Shimamoto, S., Levesque, N., and Starr, P. A. (2009). Single doi: 10.1016/S0079-6123(06)60007-1 unit “pauser” characteristics of the globus pallidus pars externa distinguish Frontiers in Physiology | www.frontiersin.org 15 November 2017 | Volume 8 | Article 897 Diao et al. Pallidal Adenosine A Receptor 2A primary dystonia from secondary dystonia and Parkinson’s disease. Exp. in mice. J. Clin. Neurosci. 17, 1412–1416. doi: 10.1016/j.jocn.2010. Neurol. 216, 295–299. doi: 10.1016/j.expneurol.2008.12.006 01.056 Saunders, A., Huang, K. W., and Sabatini, B. L. (2016). Globus Villar-Menendez, I., Porta, S., Buira, S. P., Pereira-Veiga, T., Diaz-Sanchez, pallidus externus neurons expressing parvalbumin interconnect the S., Albasanz, J. L., et al. (2014). Increased striatal adenosine A2A subthalamic nucleus and striatal interneurons. PLoS ONE 11:e0149798. receptor levels is an early event in Parkinson’s disease-related pathology doi: 10.1371/journal.pone.0149798 and it is potentially regulated by miR-34b. Neurobiol. Dis. 69, 206–214. Shindou, T., Nonaka, H., Richardson, P. J., Mori, A., Kase, H., and Ichimura, doi: 10.1016/j.nbd.2014.05.030 M. (2002). Presynaptic adenosine A2A receptors enhance GABAergic Xue, Y., Han, X. H., and Chen, L. (2010). Effects of pharmacological block of synaptic transmission via a cyclic AMP dependent mechanism in the GABA(A) receptors on pallidal neurons in normal and Parkinsonian State. rat globus pallidus. Br. J. Pharmacol. 136, 296–302. doi: 10.1038/sj.bjp. Front. Cell. Neurosci. 4, 2. doi: 10.3389/neuro.03.002.2010 0704702 Zold, C. L., Ballion, B., Riquelme, L. A., Gonon, F., and Murer, M. Shindou, T., Richardson, P. J., Mori, A., Kase, H., and Ichimura, M. G. (2007). Nigrostriatal lesion induces D2-modulated phase-locked (2003). Adenosine modulates the striatal GABAergic inputs to the globus activity in the basal ganglia of rats. Eur. J. Neurosci. 25, 2131–2144. pallidus via adenosine A2A receptors in rats. Neurosci. Lett. 352, 167–170. doi: 10.1111/j.1460-9568.2007.05475.x doi: 10.1016/j.neulet.2003.08.059 Sindhu, K. M., Saravanan, K. S., and Mohanakumar, K. P. (2005). Conflict of Interest Statement: The authors declare that the research was Behavioral differences in a rotenone-induced hemiparkinsonian rat conducted in the absence of any commercial or financial relationships that could model developed following intranigral or median forebrain bundle be construed as a potential conflict of interest. infusion. Brain. Res. 1051, 25–34. doi: 10.1016/j.brainres.2005. 05.051 Copyright © 2017 Diao, Xue, Han, Wang, Liu, Chen and Chen. This is an open- Suarez, L. M., Solis, O., Aguado, C., Lujan, R., and Moratalla, R. (2016). L- access article distributed under the terms of the Creative Commons Attribution DOPA oppositely regulates synaptic strength and spine morphology in D1 and License (CC BY). The use, distribution or reproduction in other forums is permitted, D2 striatal projection neurons in dyskinesia. Cereb. Cortex. 26, 4253–4264. provided the original author(s) or licensor are credited and that the original doi: 10.1093/cercor/bhw263 publication in this journal is cited, in accordance with accepted academic practice. Tabuse, M., Yaguchi, M., Ohta, S., Kawase, T., and Toda, M. (2010). No use, distribution or reproduction is permitted which does not comply with these A simple behavioral test for locomotor function after brain injury terms. Frontiers in Physiology | www.frontiersin.org 16 November 2017 | Volume 8 | Article 897 http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Frontiers in Physiology Pubmed Central

Adenosine A2A Receptor Modulates the Activity of Globus Pallidus Neurons in Rats

Frontiers in Physiology , Volume 8 – Nov 7, 2017

Loading next page...
 
/lp/pubmed-central/adenosine-a2a-receptor-modulates-the-activity-of-globus-pallidus-qBkjbkw4l6

References

References for this paper are not available at this time. We will be adding them shortly, thank you for your patience.

Publisher
Pubmed Central
Copyright
Copyright © 2017 Diao, Xue, Han, Wang, Liu, Chen and Chen.
ISSN
1664-042X
eISSN
1664-042X
DOI
10.3389/fphys.2017.00897
Publisher site
See Article on Publisher Site

Abstract

ORIGINAL RESEARCH published: 07 November 2017 doi: 10.3389/fphys.2017.00897 Adenosine A Receptor Modulates 2A the Activity of Globus Pallidus Neurons in Rats 1, 2 1 1 1, 3 1 Hui-Ling Diao , Yan Xue , Xiao-Hua Han , Shuang-Yan Wang , Cui Liu , 1 1 Wen-Fang Chen and Lei Chen * 1 2 Department of Physiology, Qingdao University, Qingdao, China, Department of Physiology, Binzhou Medical University, Yantai, China, Department of Anatomy, Qingdao University, Qingdao, China The globus pallidus is a central nucleus in the basal ganglia motor control circuit. Morphological studies have revealed the expression of adenosine A receptors in the 2A globus pallidus. To determine the modulation of adenosine A receptors on the activity 2A of pallidal neurons in both normal and parkinsonian rats, in vivo electrophysiological and behavioral tests were performed in the present study. The extracellular single unit recordings showed that micro-pressure administration of adenosine A receptor 2A agonist, CGS21680, regulated the pallidal firing activity. GABAergic neurotransmission was involved in CGS21680-induced modulation of pallidal neurons via a PKA pathway. Furthermore, application of two adenosine A receptor antagonists, KW6002 or 2A SCH442416, mainly increased the spontaneous firing of pallidal neurons, suggesting Edited by: that endogenous adenosine system modulates the activity of pallidal neurons through Hansen Wang, adenosine A receptors. Finally, elevated body swing test (EBST) showed that 2A University of Toronto, Canada intrapallidal microinjection of adenosine A receptor agonist/antagonist induced 2A Reviewed by: Carla Cannizzaro, ipsilateral/contralateral-biased swing, respectively. In addition, the electrophysiological Università degli Studi di Palermo, Italy and behavioral findings also revealed that activation of dopamine D receptors by Rosario Moratalla, quinpirole strengthened KW6002/SCH442416-induced excitation of pallidal activity. Consejo Superior de Investigaciones Científicas (CSIC), Spain Co-application of quinpirole with KW6002 or SCH442416 alleviated biased swing in *Correspondence: hemi-parkinsonian rats. Based on the present findings, we concluded that pallidal Lei Chen adenosine A receptors may be potentially useful in the treatment of Parkinson’s 2A chenleiqd@163.com disease. Specialty section: Keywords: globus pallidus, adenosine A receptors, Parkinson’s disease, extracellular single unit recording, 2A This article was submitted to elevated body swing test Integrative Physiology, a section of the journal Frontiers in Physiology INTRODUCTION Received: 15 June 2017 Accepted: 24 October 2017 The rodent globus pallidus (homolog of the external segment of the primate globus pallidus) is Published: 07 November 2017 a central nucleus in the indirect pathway of the basal ganglia circuit (Jellinger, 1991), which plays Citation: important roles in movement regulation under healthy and pathological states (Raz et al., 2000; Kita Diao H-L, Xue Y, Han X-H, Wang S-Y, and Kita, 2001; Dodson et al., 2015; Hegeman et al., 2016). Based on the in vivo electrophysiological Liu C, Chen W-F and Chen L (2017) characteristics, the globus pallidus neurons are classified into three firing patterns including Adenosine A Receptor Modulates 2A high-frequency without pauses, high-frequency with pauses and low-frequency with bursts the Activity of Globus Pallidus Neurons (Benhamou et al., 2012). The spontaneous firing activities of pallidal neurons are closely associated in Rats. Front. Physiol. 8:897. doi: 10.3389/fphys.2017.00897 with movement amplitude, velocity and direction (Gage et al., 2010; Hegeman et al., 2016). Frontiers in Physiology | www.frontiersin.org 1 November 2017 | Volume 8 | Article 897 Diao et al. Pallidal Adenosine A Receptor 2A Previous studies have revealed that aberrant pallidal neuron in the globus pallidus of both intact and 6-hydroxydopamine activities appear at the onset and maintenance of motor (6-OHDA) parkinsonian rats. dysfunction in Parkinson’s diseases (Kita, 2007; Obeso et al., 2008). Furthermore, the decreased firing rate and synchronous MATERIALS AND METHODS bursting of pallidal neurons are strongly related to the motor symptoms of Parkinson’s diseases (Raz et al., 2000; Sani et al., Animals 2009; Chan et al., 2011). Adult male Wistar rats (Qingdao, China), 8–10 weeks of age and Adenosine is an endogenous purine nucleoside which weighing 220–290 g, were used in this experiment. Rats were plays a wide variety of roles in central nervous systems, ◦ housed in an environmentally controlled room at 22 ± 1 C including development, sleep, synaptic transmission, pain, with a 12 h light/dark cycle. The study was performed strictly neuroinflammation, anxiety and depression. Adenosine in accordance to the University ethics guidelines. All operations receptors (A , A , A , and A ) belong to G-protein-coupled 1 2A 2B 3 were required to lower rats’ suffering and pain. A total of 115 receptors with A and A receptors couple to Gi and Go while , 1 3 rats were used for electrophysiological experiments, with 81 A and A receptors couple to Gs proteins (Fredholm et al., 2A 2B normal rats and 34 successful parkinsonian rats. Three of the 2001, 2011; Ferre et al., 2008). It is known that adenosine A 2A 37 parkinsonian rats did not show any stable recordings and receptors are highly expressed in the striatum, globus pallidus, were excluded from the study. In addition, 78 rats were used nucleus accumbens, and olfactory tubercles of rat and human for behavioral study and 8 rats were used for immunoflurescence brain, as compared to other adenosine receptor subtypes with staining. widespread brain distribution (Jarvis and Williams, 1989; Martinez-Mir et al., 1991; Rosin et al., 2003). The selective and Establishment of 6-Hydroxydopamine specific location of adenosine A receptors suggests that it may 2A (6-OHDA) Hemi-Parkinsonian Rat Model become a potential therapeutic target for basal ganglia diseases, Rats were injected with chloral hydrate (400 mg/kg, i.p.) and fixed particularly Parkinson’s disease (Preti et al., 2015; Cunha, 2016). in the stereotaxic instrument (NarishigeSN-3, Tokyo, Japan). Considerable preclinical studies have shown that blockade of 6-hydroxydopamine (6-OHDA, 4 μg/μl in saline with 0.01% adenosine A receptors could symptomatically relieve from 2A ascorbic acid, 4 μl) was injected into the left medial forebrain parkinsonian motor deficits without L-DOPA-related motor bundle (AP −4.3, ML +1.7, DV −8.4 mm from Bregma) using side effects (Bibbiani et al., 2003; Jenner, 2014; Preti et al., 2015; microsyringe at a rate of 1.0 μl/min. Animals that exhibited at Cunha, 2016). least 210 net contralateral rotations in 30 min after delivered Anatomical and morphological studies suggest that the apomorphine (0.2 mg/kg, s.c.) were considered as successful adenosine A receptors are principally expressed on 2A hemi-parkinsonism rats. striatopallidal terminals in the globus pallidus (Rosin et al., 2003; Shindou et al., 2003). Previous in vitro electrophysiological studies show that stimulation of adenosine A receptors Electrophysiological Recordings in Vivo 2A enhances GABA release and therefore augments the inhibitory One of the advantages of in vivo extracellular recording used postsynaptic currents (IPSCs) in the globus pallidus (Mori and in this study is to investigate the direct effects of drugs on the Shindou, 2003; Floran et al., 2005). Other studies reveal that single pallidal neurons in both physiological and pathological adenosine A receptor activation exerts a dual effect on the conditions. The other advantage of the present multibarrel 2A release of GABA (Mayfield et al., 1993; Dayne Mayfield et al., microelectrodes is to apply drugs directly to the neurons 1996; Morales-Figueroa et al., 2014). However, Querejeta et al. recorded. According to our previous studies (Xue et al., 2010; (2010) demonstrate that intrapallidal infusion of adenosine Chen et al., 2015), extracellular recordings were performed in the A receptor agonist and antagonist have no effects on the globus pallidus of normal urethane (1 g/kg, i.p.) anesthetized rats. 2A spontaneous firing rate of the globus pallidus neurons in both Anaesthesia levels were monitored constantly by testing reflexes sham and ipsilaterally dopamine-denervated rats. As adenosine to a cutaneous pinch and maintained by giving supplement A receptor antagonist is a potential therapeutic target for dose of chloral hydrate (0.1 g/kg, i.p.) when necessary. Rats 2A parkinsonian motor deficits (Bibbiani et al., 2003; Jenner, 2014; were fixed in the stereotaxic instrument. Two holes were Preti et al., 2015; Cunha, 2016), the study of pallidal adenosine drilled bilaterally over the globus pallidus according to the A receptors will provide an insight into the movement stereotaxic atlas (0.8–1.2 mm posterior to bregma and 2.5– 2A regulation in both normal and parkinsonian states. Up to 3.5 mm lateral to the midline, Paxinos and Watson, 1986). All present, little is known about the in vivo electrophysiological exposed cortex were covered with a thin layer of saline to and behavioral effects and the possible mechanisms of adenosine prevent drying. Recordings of neuronal spike and micro-pressure A receptors in the globus pallidus under both normal ejection of drugs were used with fined 3-barrelled micropipettes 2A and abnormal states. By using multibarrel microelectrode (tip diameter 3–10 μm, resistance10–20 M). The recording extracellular recordings and elevated body swing test (EBST), microelectrode contained 2% pontamine sky blue dissolved in we therefore investigated the effects of adenosine A receptors 0.5 M sodium acetate. The other two microelectrodes were filled 2A with vehicle (saline) and various drugs, including CGS21680, KW6002, SCH442416, quinpirole, gabazine, nipecotic acid, or Abbreviations: 6-OHDA, 6-hydroxydopamine; GABA, γ-aminobutyric acid; PKA, protein kinase A; EBST, elevated body swing test. H-89. Drugs were delivered using 4-channel pressure injector Frontiers in Physiology | www.frontiersin.org 2 November 2017 | Volume 8 | Article 897 Diao et al. Pallidal Adenosine A Receptor 2A (PM2000B, Micro Data Instrument, Inc., USA) with short- globus pallidus sections were blocked with 5% donkey serum pulse gas pressure (1,500 ms, 5.0–15.0 psi). The recording (Solaribo) and 0.3%Triton X-100 in PBS for 60 min at 4 C. electrical signal was amplified by a micro-electrode amplifier And the sections were subsequently incubated with a rabbit (MEZ-8201, Nihon Kohden, Tokyo, Japan), low- and high- polyclonal anti-adenosine A receptor antibody (sc-13937, 2A pass filtered at 0.3 and 3 kHz, monitored with a memory 1:100; Santa Cruz) and a goat polyclonal anti-parvalbumin oscilloscope (VC-11, Nihon Kohden, Tokyo Japan) and an audio antibody (ab3289, 1:500; Abcam) for 48 h at 4 C. After washing monitor. The electrical signal was transferred into bio-electricity three times with PBS (5 min), the sections were incubated signal analyzer and computer. The pallidal spiking data capture with the following secondary antibodies, Alexa 488-conjugated and analysis were available via spike 2 software (Cambridge donkey anti-rabbit (1:1,000; Abcam) and Alexa 647-conjugated Electronic Design, UK). donkey anti-goat (1:1,000; Abcam), for 2 h at room temperature. Drug application was initiated after basal spontaneous firing Lastly, the sections were washed, mounted, coverslipped and of pallidal neurons remained stable for at least 10 min. The examined under fluorescent microscope or laser scanning 120 s average firing before drug application was regarded as the confocal microscopy (Leica, Wetzlar, Germany). basal firing, and the 50 s maximally changing discharge after The results were analyzed by counting the number of drug administration was considered as the effect. The increase adenosine A receptor-positive cells in the globus pallidus. 2A or decrease in firing frequency was considered statistically The adenosine A receptor-positive cells were examined and 2A significant when the firing rate was higher or lower than the mean quantified using Image J software (NIH, Bethesda, MD). We ±2 SD of the baseline (SD = standard deviation). Firing pattern selected six sections (at the levels of 0.8–1.2 mm posterior was assessed via the coefficient of variation (CV) which referred to bregma) from the globus pallidus from each rat for cell to the standard deviation of the interspike intervals (ISI) divided counting. Results were expressed as the average number of by mean. positive cells obtained from the six sections (cells visualized at 400× magnifications). Elevated Body Swing Test (EBST) EBST was performed in awake and freely-moving rats according Histological Controls to previously described methods (Borlongan et al., 1995; After electrophysiological and behavioral experiments, we Baluchnejadmojarad and Roghani, 2004). The rat was placed in needed to verify the position of electrophysiological recording a transparent plastic cage (40 × 40 × 35 cm). After a 10 min- sites and cannula placements. Rats were deeply anesthetized habituation, the rat was elevated to ∼2 cm above the bottom of and perfused transcardially with a saline solution containing the cage by holding 2 cm above the bottom of tail. The body 4% paraformaldehyde. Recording/microinjection sites were swing referred to the rat head with deviation of vertical axis more confirmed using camera (Figure 1). To further confirm the than 10 . Before every swing was recorded, the rat head must 6-OHDA parkinsonian rat model, coronal substantia nigra be suspended at a vertical axis. If the rat did not bend its head sections were incubated with a rabbit anti-tyrosine hydroxylase after being elevated over 5 s, the tail was gently pinched. Each test antibody (T8700, 1:1000; sigma) following secondary antibodies rat was recorded for 60 s. Initially, only rats showing unbiased (1:1,000; Abcam) of Alexa 488-conjugated donkey anti-rabbit. behavior were chosen for the study. The number of tyrosine hydroxylase-positive neurons on the All experimental rats were anesthetized with chloral hydrate lesioned side of the substantia nigra pars compacta decreased to (400 mg/kg, i.p.), and implanted with the stainless steel guide 17.08 ± 1.84%, which was significantly lower than that of normal cannula in the globus pallidus on either side (o.d., 0.4 mm; rats (Figure 5A). i.d., 0.3 mm). The cannula was secured to the skull with screws and dental acrylic. The vehicle (saline) or drugs (CGS21680, KW6002, and SCH442416) was unilaterally microinjected into Drugs and Statistical Analysis the globus pallidus with 1 μl microsyringe. The injection rate was CGS21680, KW6002, SCH44216 and quinpirole were purchased maintained at 0.2 μl/min with a total volume of 0.5 μl and the from Tocris (Bristol, UK). DMSO, 6-OHDA hydrochloride, microsyringe was kept in position for an additional 3 min before apomorphine hydrochloride, gabazine, and H-89 were obtained retraction. from Sigma (St Louis, Mo, USA). Nipecotic acid was purchased The number of biased body swing was counted per minute. from Abcam (Cambridge, MA, USA). The percentage score was calculated for each rat, i.e., the number All data were presented as mean ± S.E.M. Paired-samples of biased swings was divided by the total number of swings and t-test was used for comparing the difference of spontaneous multiplied by 100%. Over 70% of biased swings were defined as firing before and after drug application. Statistical comparisons the criterion for biased swing behavior. between or among groups were determined with independent- samples t-test and one-way ANOVA. Chi-square test was applied Double Immunofluorescence Staining to compare the fractions of CGS21680 responsive neurons and The double immunofluorescence technique was used to observe different concentrations of drugs. Before t-tests, Levene’s test the expression of adenosine A receptors and parvalbumin in F-test was performed to determine if the data is normally 2A rat globus pallidus. Brain tissue was obtained from normal and distributed. Wilcoxon–Mann–Whitney test was used for data parkinsonian rats, and then was fixed in 4% paraformaldehyde that are not normally distributed. P < 0.05 was considered as overnight. Tissues containing the globus pallidus were sectioned the level of significance. Data analysis was performed using IBM at 40 μm after dehydration in 20 and 30% sucrose. Coronal SPSS 22.0 software. Frontiers in Physiology | www.frontiersin.org 3 November 2017 | Volume 8 | Article 897 Diao et al. Pallidal Adenosine A Receptor 2A FIGURE 1 | Confirmation of the recording and microinjection location within the globus pallidus. (A) A coronal brain section illustrating the trace of microelectrodes in the globus pallidus. (B) A coronal brain section showing the injector tip for microinjection. GP, globus pallidus; Str, striatum; scale bars = 1 mm. RESULTS P < 0.05, paired-samples t-test), with the average increase of 33.04 ± 1.60% (t = 10.99, df = 15, P < 0.001 compared to Effects of Adenosine A Receptor 2A that of vehicle administration, independent-samples t-test). In Activation on the Spontaneous Firing of the remaining 20 pallidal neurons, 10 μM CGS21680 did not alter the firing rate significantly. The highest concentration of Globus Pallidus Neurons in Normal Rats We first observed the effects of adenosine A receptor selective CGS21680 (100 μM) only increased the pallidal firing rate in 2A 3 out of the 19 neurons, while the lowest concentration of the agonist CGS21680 on the spontaneous firing rate of 47 pallidal neurons tested. The average neuronal firing rate was 14.02 drug (0.1 μM) only had decreasing effects in 1 out of the 13 neurons. Further analysis revealed that the percentages of pallidal ± 1.44 Hz. Micro-pressure administration of 1 μM CGS21680 neurons which were inhibited by different concentrations of significantly decreased the frequency of spontaneous firing from 14.89 ± 2.48 Hz to 9.03 ± 2.14 Hz in 21 out of the 47 pallidal CGS21680 (0.1, 1, 10, and 100 μM) were significantly different based on Chi-square test (x = 28.26, df = 6, P < 0.001). The neurons (n = 21, P < 0.001, t = 5.78, df = 20, paired- samples t-test, Figures 2A,C). The average decrease was 46.08 percentage of reactive neurons (44.68%) with 1 μM CGS21680- induced inhibition of firing rate was strongest compared to that ± 5.71%, which was significantly different (Z = 4.72, P < 0.001, Mann-Whitney test) compared to that of vehicle administration of other drug concentrations (Figure 2D), while there was no significant change in the percentage of reactive neurons with (basal: 15.31 ± 3.24 Hz; vehicle: 15.58 ± 3.21 Hz; n = 12, t = 1.61, df = 11, P > 0.05, paired-samples t-test). In only CGS21680-induced excitatory effects of all drug concentrations. Therefore, we chose the concentration of 1 μM to further explore 7 out of the 47 pallidal neurons, 1 μM CGS21680 increased the firing frequency from 10.05 ± 3.40 to 12.66 ± 3.77 Hz the effects of adenosine A receptors on pallidal neurons in the 2A subsequent experiments. (n = 7, t = 4.34, df = 6, P < 0.01, paired-samples t-test, Next, we analyzed the possible changes of firing pattern Figures 2B,C), with the average increase of 50.80 ± 20.68% (Z = 3.47, P < 0.01 compared to that of vehicle administration, by calculating the CV before and after CGS21680 application. Micro-pressure ejection of 1 μM CGS21680 didn’t change the CV Mann-Whitney test). In the remaining 19 pallidal neurons, 1 μM CGS21680 did not alter the firing rate significantly (t = 1.27, significantly (basal: 1.30 ± 0.32; CGS21680: 1.77 ± 0.39; n = 21, t =1.90, df = 20, P > 0.05, paired-samples t-test, Figure 2E) in the df = 18, P > 0.05, paired-samples t-test). Similar to that of 1 μM CGS21680, the higher concentration of CGS21680 (10 μM) also 21 neurons with CGS21680-induced decrease in firing rate. There was also no change of CV in the neurons with 1 μM CGS21680- modulated the spontaneous firing of globus pallidus neurons. The average spontaneous firing rate of all the 35 pallidal neurons induced increase in firing rate (basal: 0.59 ± 0.15; CGS21680: 0.56 ± 0.14; n = 7, t = 0.71, df = 6, P > 0.05, paired-samples tested was 17.12 ± 2.24 Hz. Application of 10 μM CGS21680 decreased the frequency of spontaneous firing from 23.35 ± t-test, Figure 2E). Similar to that of 1 μM CGS21680, there was no significant change in CV following micro-pressure ejection of 5.90 to 13.93 ± 4.97 Hz in 9 out of the 35 pallidal neurons CGS21680 at the concentrations of 0.1, 10, and 100 μM (data not (n = 9, t = 6.94, df = 8, P < 0.001, paired-samples t-test). shown). The average decrease was 47.70 ± 5.97%, which was significantly According to our previous study (Chen et al., 2015), we different (t = 8.73, df = 18, P < 0.001, independent-samples t-test) compared to that of vehicle administration (basal: 16.03 classified pallidal neurons into three types (i.e., higher frequency without pauses, higher frequency with pauses and low frequency ± 2.83 Hz; vehicle: 16.42 ± 2.90 Hz; n = 11, t = 1.88, df = 10, P > 0.05, paired-samples t-test). CGS21680 at 10 μM increased with bursts, Figures 2F–H). Based on this classification, the proportion of 36.17% high frequency without pauses, 48.94% the firing rate from 6.78 ± 2.82 to 9.91 ± 3.83 Hz in other 6 out of the 35 pallidal neurons (n = 6, t = 2.63, df = 5, high frequency with pauses and 14.89% low frequency with Frontiers in Physiology | www.frontiersin.org 4 November 2017 | Volume 8 | Article 897 Diao et al. Pallidal Adenosine A Receptor 2A FIGURE 2 | Effects of micro-pressure ejection of adenosine A receptor selective agonist, CGS21680, on the spontaneous firing of pallidal neurons in normal rats. 2A (A) Typical frequency histograms showing that 1 μM CGS21680 decreased the firing rate of a pallidal neuron. (B) In this neuron, 1 μM CGS21680 increased the firing rate slightly. (C) Pooled data summarizing the effects of 1 μM CGS21680 on the firing rate of pallidal neurons in normal rats. The black bars represent the basal firing (Continued) Frontiers in Physiology | www.frontiersin.org 5 November 2017 | Volume 8 | Article 897 Diao et al. Pallidal Adenosine A Receptor 2A FIGURE 2 | Continued rate before CGS21680 application, while the gray bars represent the firing rate after CGS21680 application. **P < 0.01, ***P < 0.001, paired-samples t-test. (D) The graph summarizing the percentages of pallidal neurons with different concentrations (0.1, 1, 10, and 100 μM) of CGS21680-induced decrease in firing rate. (E) Pooled data summarizing the effects of CGS21680 on the CV of pallidal neurons. (F–H) Typical traces showing three firing patterns of globus pallidus neurons, high frequency without pauses (F), high frequency with pauses (G), and low frequency with bursts (H). (I) Pooled data showing the proportion of the three firing pattern types in total neurons and CGS21680-induced inhibitory or excitatory neurons. bursts were observed within the present 47 pallidal neurons. Furthermore, within 21 out of the 47 pallidal neurons with CGS21680-induced decrease in firing rate, these firing pattern types represented a proportion of 28.57, 47.62, and 23.81%, respectively. In another 7 CGS21680-induced excitatory neurons, 42.86, 42.86, and 14.28% exhibited above three firing patterns, respectively. There was no significant difference among the proportions of firing pattern types of the neurons in CGS21680- induced inhibitory or excitatory neurons and the total 47 neurons (x = 1.15, df = 4, P > 0.05, Chi-square test, Figure 2I). Endogenous Adenosine Modulated the Spontaneous Firing of Globus Pallidus Neurons through Adenosine A Receptors 2A in Normal Rats To elucidate the possible modulation of spontaneous pallidal neuronal activity by endogenous adenosine through adenosine A receptors, we explored the effects of selective adenosine A 2A 2A receptor antagonists on the spontaneous firing activity of pallidal neurons. Micro-pressure administration of selective adenosine A receptor antagonist, KW6002 (1 μM), significantly increased 2A the spontaneous firing rate from 12.92 ± 3.98 to 15.95 ± 4.70 Hz in 11 out of the 27 pallidal neurons (n = 11, t = 3.05, df = 10, P < 0.05, paired-samples t-test, Figures 3A,C), with the average increase of 29.48 ± 10.76% (Z = 2.97, P < 0.05 compared to that of vehicle administration, Mann-Whitney test). In 4 neurons, 1 μM KW6002 decreased the firing rate from 9.81 ± 4.50 to 6.82 ± 3.67 Hz. In the remaining 12 neurons the firing rate was not affected (data not shown). Based upon above results, the present study first demonstrated that endogenous adenosine modulated the spontaneous firing of pallidal neurons in vivo. To further provide more evidence, we FIGURE 3 | Effects of adenosine A receptor antagonists, KW6002 and 2A continued to observe the effects of another selective antagonist SCH442416, on the spontaneous firing of pallidal neurons in normal rats. (A) SCH442416 on the pallidal firing. Similar to that of KW6002, Typical frequency histograms showing that 1 μM KW6002 increased the firing rate of a pallidal neuron. (B) In this neuron, 1 μM SCH442416 increased the the predominant effect of SCH442416 (1 μM) was an increase firing rate. (C) Pooled data summarizing the excitatory effects of both KW6002 in the firing rate of globus pallidus neurons. Micro-pressure and SCH442416 on the firing rate of pallidal neurons. The black bars represent administration of 1 μM SCH442416 significantly increased the the basal firing rate before drugs application, while the gray bars represent the spontaneous firing rate from 11.46 ± 2.19 Hz to 15.03 ± 2.58 Hz firing rate after drugs application. *P < 0.05, ***P < 0.001, paired-samples in 13 out of the 22 pallidal neurons (n = 13, t = 6.03, t-test. df = 12, P < 0.001, paired-samples t-test, Figures 3B,C). The average increase was 38.78 ± 8.56%, which was significantly different (Z = 3.91, P < 0.001, Mann-Whitney test) from that of Blockade of Adenosine A Receptors was 2A vehicle ejection. Moreover, we compared the increasing effects of Involved in Exogenous CGS21680-Induced two antagonists, SCH442416 and KW6002, on the spontaneous firing rate of pallidal neurons. The excitatory action of 1 μM Modulation of Pallidal Firing SCH442416 on firing rate was slightly stronger than that of 1 μM Next two adenosine A receptor antagonists KW6002 and 2A KW6002 (t = 2.18, df = 22, P < 0.05, independent-samples SCH442416 were used to further study whether adenosine A 2A t-test). receptors are involved in CGS21680-induced modulation of Frontiers in Physiology | www.frontiersin.org 6 November 2017 | Volume 8 | Article 897 Diao et al. Pallidal Adenosine A Receptor 2A pallidal firing rate. As shown in Figure 4A, 1 μM CGS21680 did not change the firing rate significantly in the presence of 1 μM KW6002. The second time application of 1 μM CGS21680 alone decreased the firing rate which indicated that CGS21680- induced inhibitory effect of firing rate was prevented by KW6002. In total 17 pallidal neurons, CGS21680 only decreased the spontaneous firing rate in 5 out of the 17 pallidal neurons in the presence of 1 μM KW6002. The average decrease (13.93 ± 2.96%) of the firing rate was significantly weaker than that in the absence of 1 μM KW6002 (46.08 ± 5.71%, Z = 2.83, P < 0.01, Mann-Whitney test, Figure 4C). Moreover, SCH442416 also blocked CGS21680-induced decrease of pallidal firing rate (Figures 4B,C). Similar to that of CGS21680-induced inhibitory effects, KW6002 and SCH442416 also blocked the increasing effects of CGS21680 on pallidal firing rate (data not shown). Effects of Adenosine A Receptor 2A Activation on the Spontaneous Firing of Globus Pallidus Neurons in 6-OHDA Hemi-Parkinsonian Rats In this study, we further explored the direct modulation of adenosine A receptors on the spontaneous pallidal firing in 6- 2A OHDA hemi-parkinsonian rats (Figure 5A). The average basal firing rate of pallidal neurons on lesioned sides of 6-OHDA parkinsonian rats was significantly different from that of normal rats or unlesioned sides of parkinsonian rats (F = 8.16, df = 2, P < 0.01, one-way ANOVA). The average basal firing rate of pallidal neurons on the lesioned side of parkinsonian rats (7.94 ± 0.98 Hz, n = 44) was significantly lower than that of normal rats (14.02 ± 1.46 Hz, n = 47, t = 3.48, P < 0.01), as well as that on the unlesioned side of parkinsonian rats (14.33 ± 1.26 Hz, n FIGURE 4 | Both KW6002 and SCH442416 blocked CGS21680-induced = 40, t = 3.51, P < 0.01). Concerning the firing patterns, 34.09% decrease in firing rate. (A) Typical frequency histograms showing that (15/44) of the pallidal neurons recorded exhibited low frequency application of 1 μM CGS21680 in the presence of 1 μM KW6002 did not with bursts on the lesioned side of parkinsonian rats, which was change the firing rate. However, the second time application of CGS21680 more than that of normal rats (7/47, 14.89%, x = 4.57, df = 1, P alone significantly decreased the firing rate in this neuron. (B) Typical frequency histograms showing that 1 μM SCH442416 blocked 1 μM < 0.05, Chi-square test). CGS21680-induced decrease in firing rate. (C) Pooled data summarizing the Furthermore, we observed the effects of micro-pressure effects of both KW6002 and SCH442416 on CGS21680-induced decrease of ejection of CGS21680 (1 μM) on the spontaneous firing activity firing rate. The black bars represent the effects of CGS21680 alone, while the of pallidal neurons in 6-OHDA hemi-parkinsonian rats. On the dark gray bars and light gray bars represent effects of CGS21680 together lesioned side, 1 μM CGS21680 significantly decreased the firing with KW6002 or SCH442416, respectively. *P < 0.05, **P < 0.01, compared to CGS21680 alone group, Mann-Whitney test. rate from 8.14 ± 2.04 to 6.20 ± 1.81 Hz in 7 out of the 19 neurons (n = 7, t = 6.92, df = 6, P < 0.001, paired-samples t-test). The average decrease was 33.04 ± 5.63% (t = 5.99, df = 13, P < 0.01 compared with that of vehicle ejection, independent- Effects of Endogenous Adenosine on the samples t-test). In 2 neurons the frequency increased by 40.18 Spontaneous Firing of Globus Pallidus ± 18.74% and in the left 10 neurons the firing rate was not Neurons through Adenosine A Receptors affected (t = 1.46, df = 16, P > 0.05, independent-samples t- 2A test). On the unlesioned side, 1 μM CGS21680 also exhibited in 6-OHDA Hemi-Parkinsonian Rats bidirectional effects on spontaneous discharge of pallidal neurons Previous studies have found that blockade of adenosine A 2A (data not shown). The CGS21680-induced inhibitory effects on receptors exerts therapeutic potential in Parkinson’s disease. We both lesioned and unlesioned sides of parkinsonian rats were further observed the direct regulation of the spontaneous pallidal not different from that in normal rats (F = 0.68, df = 2, P > firing by two selective adenosine A receptor antagonists, 2A 0.05, one-way ANOVA, Figure 5B). Moreover, similar to that KW6002 and SCH442416, in 6-OHDA hemi-parkinsonian of normal rats, 1 μM CGS21680 didn’t significantly change the rats. pallidal firing pattern on both lesioned and unlesioned sides of We firstly observed the effects of 1 μM KW6002 on the pallidal parkinsonian rats (data not shown). firing rate in 6-OHDA hemi-parkinsonian rats. On the lesioned Frontiers in Physiology | www.frontiersin.org 7 November 2017 | Volume 8 | Article 897 Diao et al. Pallidal Adenosine A Receptor 2A side, the average basal firing rate of 14 neurons recorded was the relationship of adenosine A receptors and dopamine D 2A 2 10.04 ± 2.29 Hz. KW6002 tended to increase the spontaneous receptors in hemi-parkinsonian rats. Similar to that of normal firing rate in 3 out of the 14 pallidal neurons (basal: 15.42 ± rats, the excitatory effects induced by combination of quinpirole 5.51 Hz; KW6002: 18.57 ± 6.34 Hz, n = 3, t = 3.66, df = 2, with KW6002 or SCH442416 tended to be stronger than that of P = 0.07, paired-samples t-test), but there was no statistical adenosine A receptor antagonist alone in hemi-parkinsonian 2A difference. The average increase was 23.72 ± 4.59% (Figure 5B). rats (Figure 6D). In only one neuron, KW6002 decreased the firing rate with Gabaergic Neurotransmission was average decrease of 49.75% and in the left 10 neurons, KW6002 did not significantly change the firing rate (t = 0.71, df = 9, P > Involved in CGS21680-Induced Modulation 0.05, paired-samples t-test). On the unlesioned side, the average of the Spontaneous Firing of Globus basal firing rate of 16 neurons recorded was 14.98 ± 2.19. 1 μM Pallidus Neurons KW6002 significantly increased the firing rate by 26.08 ± 9.41% The competitive GABA receptor antagonist (gabazine) and the in 8 out of the 16 pallidal neurons (t = 2.52, df = 14, P < selective GABA transporter-1 inhibitor (nipecotic acid) were 0.05 compared with that of vehicle group, independent-samples used to further identify the possible involvement of GABAergic t-test), while did not change the firing rate in the remaining neurotransmission in CGS21680-induced decrease of firing 8 neurons (t = 0.34, df = 14, P > 0.05 compared with that rate. As shown in Figures 7A,C, in the presence of gabazine, of vehicle group, independent-samples t-test). In addition, the CGS21680 only decreased the firing rate by 8.04 ± 2.35% in 4 percentage of KW6002 responsive neurons on the lesioned side out of the 13 globus pallidus neurons, which was significantly (28.57%, 4 out of 14) tended to be lower than that in normal weaker than that of the second time application of CGS21680 rats (55.56%, 15 out of 27), and that on the unlesioned side in the absence of gabazine (decrease: 33.59 ± 9.59%, n = 6, (50.00%, 8 out of 16), although there was no statistic difference. Z = 2.35, P < 0.05, Mann-Whitney test). In another set of Further studies revealed that SCH442416 increased the firing rate experiment, nipecotic acid (10 μM) significantly decreased the from 11.31 ± 3.10 Hz to 14.86 ± 3.92 Hz in 9 out of the 21 firing rate from 14.73 ± 2.73 Hz to 6.12 ± 2.28 Hz in 17 out neurons on the lesioned side (n = 9, t = 2.74, df = 8, P < 0.05, of 23 pallidal neurons tested (n = 17, t = 5.45, df = 16, P paired-samples t-test). The average increase was 36.47 ± 10.30% < 0.01, paired-samples t-test, Figure 7B). In the presence of (t = 3.42, df = 13, P < 0.01 compared with that of vehicle group, nipecotic acid, CGS21680 decreased the firing rate by 10.59 ± independent-samples t-test). Furthermore, 1 μM SCH442416- 1.15% in only 2 out of the 6 neurons in which the second induced excitatory effects of pallidal neurons on the lesioned side time application of CGS21680 alone inhibited the firing rate of parkinsonian rats was not significantly different from that of by 44.38 ± 12.62% (Z = 2.00, P < 0.05, Mann-Whitney test, normal rats (t = 0.39, df = 20, P > 0.05, independent-samples Figure 7C). t-test, Figure 5B). CGS21680 Modulated the Spontaneous Effects of Application of Both Adenosine Firing Activity of Globus Pallidus Neurons A Receptor Antagonist and Dopamine D 2A 2 through PKA Pathway Receptor Agonist on the Spontaneous It is known that activation of adenosine A receptors stimulates 2A Firing of Globus Pallidus Neurons in G-protein/AC/cAMP/PKA signaling pathway. We determined Normal and Hemi-Parkinsonian Rats whether selective PKA inhibitor, H-89, blocked adenosine A 2A Next, extracellular recordings were used to further explore the receptor-induced modulation of firing activity of globus pallidus interactions between adenosine A receptors and dopamine neurons. In one set of experiment, micro-pressure administration 2A D receptors in normal and hemi-parkinsonian rats. Firstly, we of H-89 (10 μM) significantly increased the spontaneous firing observed the relationship between adenosine A receptors and rate from 7.05 ± 1.43 Hz to 13.36 ± 2.65 Hz in 12 out of 2A dopamine D receptors in normal rats. Adenosine A receptor 15 pallidal neurons tested (n = 12, t = 3.74, df = 11, P < 2 2A antagonist was applied first. In the same neuron, this antagonist 0.01, paired-samples t-test, Figure 8A). The average increase was was applied again in the presence of selective dopamine D 107.29 ± 26.29%. This effect occurred within 15 min after H- receptor agonist quinpirole. As shown in Figures 6A,C, the first- 89 injection and lasted over 30 min. In the remaining 3 pallidal time application of adenosine A receptor antagonist KW6002 neurons, H-89 did not alter the firing rate significantly (t = 2A (1 μM) alone slightly increased the pallidal firing rate by 19.14 ± 3.03, df = 2, P > 0.05, paired-samples t-test). In another 9.51% in 9 out of 15 pallidal neurons. However, in the presence set of experiments, CGS21680 (1 μM) was applied first to of quinpirole (1 mM), the second time application of KW6002 pallidal neurons. In neurons which were inhibited by CGS21680, significantly increased the spontaneous firing rate by 67.33 ± CGS21680 was applied again in the presence of H-89 (10 μM). 14.09% in the same 9 neurons, which was stronger than that As shown in Figures 8B,C, CGS21680 alone decreased the in the absence of quinpirole (Z = 2.67, P < 0.01, Wilcoxon spontaneous firing rate from 11.40 ± 2.15 to 7.78 ± 1.87 Hz signed-rank test). In another set of experiment, quinpirole in 7 out of 13 pallidal neurons (n = 7, t = 3.11, df = 6, enhanced SCH442416-induced increase of pallidal firing rate P < 0.05, paired-samples t-test). The average decrease was from 23.00 ± 12.12% to 57.06 ± 23.90% (Z = 2.20, P < 0.05, 31.22 ± 7.11%. In the presence of H-89, the second time Wilcoxon signed-rank test, Figures 6B,C). Secondly, we explored application of CGS21680 did not cause significant change in Frontiers in Physiology | www.frontiersin.org 8 November 2017 | Volume 8 | Article 897 Diao et al. Pallidal Adenosine A Receptor 2A FIGURE 5 | Comparison of the electrophysiological effects of adenosine A receptor agonist and antagonist in both normal and parkinsonian rats. (A) Confirmation 2A of 6-OHDA hemi-parkinsonian rat model. Fluorescent images revealing tyrosine hydroxylase (TH) immunostaining in the substantia nigra pars compacta (SNc) of both normal (a) and 6-OHDA hemi-parkinsonian (b) rats. Scale bars = 100 μm. (B) Comparison of the effects of 1 μM CGS21680, 1 μM KW6002, and 1 μM SCH442416 on the spontaneous firing rate of pallidal neurons between normal (black bars) and 6-OHDA parkinsonian rats (gray bars). ns, not significant. spontaneous firing rate (1.97 ± 1.53%, Z = 3.23, P < 0.01 (72.50 ± 3.13%, n = 8, Z = 2.81, P < 0.01 compared to compared with that of CGS21680 alone, Wilcoxon signed-rank that of vehicle administration and 71.67 ± 6.00%, n = 6, test). Z = 2.06, P < 0.05, respectively, Mann-Whitney test, Figure 9A). Secondly, the percentages of swing responses in awake parkinsonian rats were studied. In line with previous Asymmetrical Motor Behavior Induced by finding (Abrous et al., 1998), a strong lesion-induced ipsilateral Pallidal Adenosine A Receptors in Awake 2A bias was observed in present unilateral 6-OHDA-lesioned Normal and 6-OHDA Hemi-Parkinsonian rats (98.95 ± 0.72%, n = 19). Unilateral microinjection of vehicle into the globus pallidus of lesioned side did not Rats As adenosine A receptors could modulate the firing rate of alter lesion-induced ipsilateral bias (98.33 ± 1.67%, n = 6). 2A The percentage of lesion-induced bias was not affected pallidal neurons at single cellular level, we further explored the behavioral effects of modulating adenosine A receptors in by administration of KW6002 (93.75 ± 2.63%, n = 8, 2A Z = 1.07, P > 0.05, Mann-Whitney test) or SCH442416 the globus pallidus by EBST at the integral level. Firstly, the swing responses induced by adenosine A receptor agonist (93.33 ± 2.36%, n = 9, Z = 1.11, P > 0.05, Mann-Whitney 2A test). Furthermore, the rats were intrapallidally injected with and antagonists in awake normal rats were observed. The rats receiving unilateral vehicle administration displayed unbiased KW6002 (1 μM) or SCH442416 (1 μM) 30 min after quinpirole administration (0.05 mg/kg, s.c.). Quinpirole alone did not swings (53.33 ± 3.33%, n = 6). Unilateral microinjection of CGS21680 (1 μM) into the globus pallidus significantly induced alter lesion-induced ipsilateral bias (98.00 ± 1.49%, n = 5). strong ipsilateral bias (85.00 ± 4.23%, n = 8, Z = 3.13, P < Co-administration of quinpirole with KW6002 or SCH442416 significantly decreased lesion-induced ipsilateral biased swing 0.01 compared to that of vehicle administration, Mann-Whitney test, Figure 9A). In contrast to that of CGS21680, unilateral (63.75 ± 5.32%, n = 8, Z = 2.99, P < 0.01 and 64.44 ± 4.12%, n = 9, Z = 3.27, P < 0.01, respectively, Mann-Whitney test, microinjection of 1 μM KW6002 and 1 μM SCH442416 exhibited biased swings contralateral to the drug-injection side Figure 9B). Frontiers in Physiology | www.frontiersin.org 9 November 2017 | Volume 8 | Article 897 Diao et al. Pallidal Adenosine A Receptor 2A FIGURE 7 | GABAergic transmission was involved in CGS21680-induced inhibition of the firing activity of globus pallidus neurons in normal rats. (A) Typical frequency histogram showing that 0.1 mM gabazine blocked CGS21680-induced decrease of firing rate of globus pallidus neuron. (B) Typical frequency histogram showing that in the presence of 10 μM nipecotic acid, CGS21680 did not induce clear decrease of firing rate. However, after long time recovery, the second time application of CGS21680 alone decreased the firing rate in the same neuron. (C) Pooled data summarizing CGS21680-induced decrease of firing rate in the presence (black bars) and absence (gray bars) of gabazine or nipecotic acid. *P < 0.05, Mann-Whitney test. Expression of Adenosine A Receptors 2A FIGURE 6 | Quinpirole enhanced KW6002 and SCH44216-induced increase and Parvalbumin in the Globus Pallidus in firing rate of pallidal neurons. (A) The first time application of 1 μM KW6002 Neurons alone increased the pallidal firing rate by 48.86% in this neuron of normal rat. In the presence of 1 μM quinpirole, the second time application of KW6002 The present immunostaining showed that adenosine A 2A significantly increased the firing rate by 78.07% in this neuron. (B) Typical receptors were expressed in the globus pallidus of both frequency histograms showing that quinpirole enhanced 1 μM normal and 6-OHDA parkinsonian rats. Average number of SCH44216-induced excitation of firing rate in this neuron. (C) Pooled data pallidal adenosine A receptor-positive neurons per slice in 2A summarizing the effects of quinpirole on KW6002 and SCH44216-induced normal rats was 34.96 ± 1.88, which was not significantly increase of firing rate in normal rats. (D) Pooled data summarizing the effects of quinpirole on KW6002 and SCH44216-induced increase of firing rate in different from that in 6-OHDA parkinsonian rats (36.04 parkinsonian rats. The black bars represent the effects of KW6002 or ± 2.39, Z = 0.73, P > 0.05, Wilcoxon signed-rank test, SCH44216 alone, while the gray bars represent the effects of KW6002 or Figures 10A–F). Moreover, the cellular location of adenosine SCH44216 together with quinpirole. *P < 0.05, **P < 0.01, ns, not significant, A receptors and parvalbumin in the globus pallidus was 2A paired-samples t-test. studied by using double immunofluorescence labeling. The Frontiers in Physiology | www.frontiersin.org 10 November 2017 | Volume 8 | Article 897 Diao et al. Pallidal Adenosine A Receptor 2A FIGURE 8 | PKA pathway was involved in CGS21680-induced change of pallidal firing activity in normal rats. (A) Application of H-89 (10 μM) significantly increased the firing rate. (B) The first time application of 1 μM CGS21680 decreased the firing rate. In the presence of H-89, the second time application FIGURE 9 | Evaluation of modulatory effects of pallidal adenosine A 2A of CGS21680 did not alter the firing rate in this cell. (C) Pooled data showing receptors on body asymmetry in awake normal and 6-OHDA that H-89 significantly blocked CGS21680-induced decrease of firing rate in hemi-parkinsonian rats using elevated body swing test. (A) In normal rats, pallidal neurons tested. The black bars and gray bars represent the effects of intrapallidal microinjection of CGS21680 (1 μM) induced ipsilateral-biased CGS21680 in the absence and presence of H89, respectively. **P < 0.01, swing, while KW6002 (1 μM) and SCH44216 (1 μM) induced Wilcoxon signed-rank test. contralateral-biased swing. (B) In 6-OHDA hemi-parkinsonian rats, unilateral microinjection of KW6002 or SCH44216 into the globus pallidus of lesioned side did not alter 6-OHDA-induced ipsilateral biased swing significantly, while co-application of KW6002 or SCH44216 together with quinpirole (0.05 mg/kg, s.c.) decreased the ipsilateral biased swing. *P < 0.05, **P < 0.01 compared result showed that adenosine A receptors were expressed 2A to vehicle (saline) group, Mann-Whitney test. in both parvabumin-positive and parvabumin-negative neurons (Figures 10G–I). pallidal firing through GABA receptors in both rats and mice DISCUSSION (Xue et al., 2010; Chen et al., 2013). The bulk of evidence has indicated that presynaptic stimulation of adenosine A 2A Activation of Adenosine A Receptors 2A receptors bidirectionally regulates GABA release from globus Inhibits the Spontaneous Firing of Pallidal pallidus slices in rats. For example, Floran et al. (2005) show Neurons through Facilitation of Gabaergic that CGS21680 evokes GABA release from rat globus pallidus Neurotransmission slices at a wide range of concentration (10 nM to 10 μM). It The present electrophysiological results showed that activation has been reported that adenosine A receptor agonist exerts 2A of adenosine A receptors by CGS21680 mainly exerted facilitatory effects on GABA release at lower concentrations, 2A inhibitory effects on the spontaneous firing of the globus but inhibitory effects at higher concentrations (Mayfield et al., pallidus neurons which were blocked by adenosine A receptor 1993; Dayne Mayfield et al., 1996; Morales-Figueroa et al., 2A antagonists. It is well-known that the globus pallidus receives 2014). The present in vivo extracellular recordings demonstrated dense GABAergic innervation originating from the striatum. Our that pre-application of GABA receptor antagonist, gabazine, previous studies have shown that endogenous GABA modulates blocked CGS21680-induced decrease of firing rate suggesting Frontiers in Physiology | www.frontiersin.org 11 November 2017 | Volume 8 | Article 897 Diao et al. Pallidal Adenosine A Receptor 2A FIGURE 10 | Double immunostaining for adenosine A receptors and parvalbumin in the globus pallidus. Fluorescence photomicrographs showing the expression 2A of adenosine A receptors in the pallidal neurons in both normal (A–C) and 6-OHDA parkinsonian (D–F) rats. Confocal laser scanning photomicrographs (G,H,I) 2A showing the expression of adenosine A receptors (green) and parvalbumin (red), and the overlapping expression (yellow) in normal rats. The arrow indicated 2A non-overlap of parvalbumin and adenosine A receptors in this cell. Scale bars = 100 μm in (A,D); 50 μm in (B–I); 25 μm in (C,F). 2A that enhancement of GABAergic neurotransmission may be of pallidal neurons, which was consistent with previous study involved in activation of adenosine A receptor-induced (Shindou et al., 2002). 2A inhibition of firing activity in the globus pallidus. GABA transporter-1 (GAT-1) is the prominent transporter in rat Endogenous Adenosine Modulates the globus pallidus. Gonzalez et al. (2006) found that stimulation Activity of Pallidal Neurons through A 2A of adenosine A receptors increases GABA level through 2A Receptors, Which May Be Associated with inhibition of GAT-mediated GABA uptake. Therefore, the high level of extracellular GABA in the globus pallidus may be the Heteromers with Dopamine D Receptors major possible mechanism of adenosine A receptor-induced Present in vivo electrophysiological and behavioral studies 2A inhibition of spontaneous discharge of pallidal neurons. Our first demonstrated that application of adenosine A receptor 2A in vivo electrophysiological studies further revealed that in antagonists increased the spontaneous firing rate of pallidal the presence of the selective GAT-1 inhibitor, nipecotic acid, neurons and induced contralateral biased swing of rats, CGS21680 could not induce inhibitory effect on the spontaneous suggesting that endogenous adenosinergic system is involved firing of the globus pallidus neurons. These findings enable in the regulation of the firing activity of the globus pallidus us to further identify that blockade of GABA transporter- and motor behavior in normal rats. Adenosine A receptors 2A induced enhancement of GABAergic neurotransmission could be divided into two proposed populations based on may be involved in activation of adenosine A receptor- whether forming heteromers with dopamine D receptors (Ferre 2A 2 mediated inhibition of pallidal neurons. Moreover, the present et al., 2008; Orru et al., 2011). The different pharmacological results suggested that activation of cAMP/PKA pathway was properties of the two antagonists, KW6002 and SCH442416, involved in CGS21680-induced modulation of firing activity depend on their affinities to different population of adenosine Frontiers in Physiology | www.frontiersin.org 12 November 2017 | Volume 8 | Article 897 Diao et al. Pallidal Adenosine A Receptor 2A A receptors. KW6002 has high affinity for adenosine A and the expression of molecular markers (Cooper and Stanford, 2A 2A receptors forming heteromers with dopamine D receptors, 2000; Benhamou et al., 2012; Mallet et al., 2012; Hernandez while SCH442416 shows very low affinity for adenosine A et al., 2015; Karain et al., 2015). The CGS21680-induced 2A receptors co-expressed with dopamine D receptors. The present bidirectional effects (inhibition or excitation) may be associated electrophysiological study revealed that two populations of with the diversity of pallidal neurons. The GABAergic globus adenosine A receptors both are involved in endogenous pallidus neurons are classificated into parvalbumin-positive 2A adenosine-induced modulation of pallidal neuronal activity. and parvalbumin -negative neurons based on the expression Moreover, the present electrophysiological results also illustrated of the calcium-binding protein parvalbumin (Saunders et al., that the effects of SCH442416 on pallidal firing (38.78 ± 8.56%) 2016). Parvalbumin-positive neurons represent the majority of were slightly stronger than that of KW6002 (29.48 ± 10.76%). “prototypic” pallidal neurons which exhibit fast and regular firing The percentage of SCH442416-induced excitatory neurons (13 spontaneous activity and innervate primarily the subthalamic out of 22, 59.09%) was a bit higher than that of KW6002 (11 out nucleus. Most parvalbumin-negative neurons are “arkypallidal” of 27, 40.74%). Early studies have demonstrated that anesthesia pallidal neurons which exhibit slower and more irregular especially urethane reduces dopamine release (Kelland et al., spontaneous activity and project strongly back to dorsal 1989; Hamilton et al., 1992). Therefore, a possible explanation striatum. Based on the classification of firing patterns, the for the different intensity of the two antagonists may be that present electrophysiological study showed that CGS21680- anesthetic reduces dopamine release and then inhibits the activity induced decrease or increase of firing activity was observed in of endogenous adenosine A receptor/dopamine D receptor all the three types of pallidal neurons. Additionally, the present 2A 2 complex. Moreover, the present study showed that the percentage double immunostaining showed that adenosine A receptors 2A of KW6002-responsive neurons (4 out of 14, 28.57%) on the are located on both parvalbumin-positive and parvalbumin- lesioned side of hemiparkinsonian rats tended to be lower than negative pallidal neurons. Unfortunately, for the technical that of normal rats (15 out of 27, 55.56%), as well as that on limitation, we could not label the recorded neurons and further the unlesioned side (8 out of 16, 50.00%). The weaker effect identify the types of pallidal neurons precisely. of KW6002 in dopamine-denervated side may be related to the lower extent of dopamine binding to D receptors. Additionally, Adenosine A Receptors are Functional in Suarez et al. (2016) demonstrates that dopamine depletion 2A 6-OHDA Parkinsonian Rats reduces dendritic spines of striatal medium spiny neurons expressing dopamine D receptors and the loss is accompanied The present extracellular recordings displayed lower basal with a decrease in synaptic strength. Adenosine A receptor- firing rate of pallidal neurons after 6-OHDA lesions. However, 2A dopamine D receptor heteromers are selectively localized on the some studies show inconsistent changes of firing rates in terminals of striatal medium spiny neurons in the globus pallidus parkinsonian models (Zold et al., 2007; Ellens and Leventhal, (Floran et al., 2005). Whether the weaker effect of KW6002 is 2013). According to Leblois et al. (2006), basal ganglia neurons related to the morphology changes of synaptic spine remains to would not exhibit abnormal firing activity unless extensive be explored. dopamine depletion is produced. The present immunostaining indicated that the number of hydroxylase-immunoreactive cells in lesioned substantia nigra pars compacta of hemiparkinsonian Activation of Adenosine A Receptors 2A rats decreased to about 17%. Therefore, the decrease of pallidal Induces a Weak Excitation in Partial firing rates may be associated with severe dopamine depletion Pallidal Neurons in substantia nigra of hemiparkinsonian rats. Inhibition of The present study also showed that activation of adenosine A the globus pallidus may contribute to impeding voluntary 2A receptors produced a weak excitation in a few parts of pallidal movement in Parkinson’s disease through indirect pathway neurons. In addition to GABAergic innervation, the globus of the basal ganglia circuits, and ultimately inhibiting the pallidus receives cholinergic innervation from the brainstem activities of thalamic motor nuclei and motor cortex. The pedunculopontine tegmental nucleus (Eid et al., 2016). Nicotine present adenosine A receptor antagonists-induced increase 2A acetylcholine increases spiking rate of pallidal neurons (Rios of discharge frequency indicated that adenosine A receptor 2A et al., 2016). It has been reported that activation of adenosine antagonists may contribute to alleviating motor symptoms in A receptors by CGS 21680 facilitates acetylcholine release Parkinson’s disease by normalizing the firing rate of pallidal 2A in some brain regions including hippocampus and striatum neurons. Furthermore, the CGS21680-induced inhibition of (Kurokawa et al., 1994; Ribeiro et al., 1996; Rebola et al., pallidal firing rate in present parkinsonian rats is not significantly 2002). Although, no morphological study shows the expression different from that in normal rats. Previous studies have revealed of adenosine A receptors on cholinergic nerve terminals in changes of adenosine A receptor expression in discrete brain 2A 2A the globus pallidus, one may hypothesize that adenosine A regions of patients dying with Parkinson’s disease as well as 2A receptor-mediated possible modulation of acetylcholine release Parkinson’s animal models (Hurley et al., 2000; Villar-Menendez may be responsible for CGS21680-induced weak excitatory et al., 2014). For example, Hurley et al. (2000) reported the effects. decreasing level of adenosine A receptor mRNA in some 2A It is well-known that the globus pallidus neurons are diverse striatal regions, increasing in the substantia nigra and no change in electrophysiology, axonal projections, dendritic morphology, in any other brain regions examined. Similarly, no clear change of Frontiers in Physiology | www.frontiersin.org 13 November 2017 | Volume 8 | Article 897 Diao et al. Pallidal Adenosine A Receptor 2A the expression of pallidal adenosine A receptors was observed parkinsonian rats which therefore induces imbalance of bilateral 2A in present morphological study, which may be one of the possible movement output. It was reported that tail pinch may further reasons for the present similar electrophysiological results under cause dopamine release in the unlesioned striatum (Sindhu both normal and parkinsonian states. et al., 2005). Similar to the above supposition, application of adenosine A receptor antagonists in the presence of 2A Pallidal Adenosine A Receptors quinpirole could alleviate 6-OHDA-induced biased swing, which 2A further verified that pallidal adenosine A receptors play Modulate Motor Behavior in Awake Rats 2A important roles in the therapy of motor symptoms in Parkinson’s As adenosine A receptors modulated pallidal firing, we 2A disease. hypothesized that pharmacological manipulation of adenosine In conclusion, the present study indicated that pallidal A receptors in the globus pallidus maybe participate in 2A adenosine A receptors play prominent roles in motor motor modulation in awake rats. EBST is a simple, sensitive 2A modulation under both healthy and parkinsonian states, which and accurate behavioral test used to evaluate asymmetrical further verified that pallidal adenosine A receptor is potentially behavior in animals with a unilateral cerebral lesion such as 2A useful in the treatment of Parkinson’s disease. Therefore, more Parkinson’s disease, Huntington’s disease and ischemic stroke studies will be needed to explore the functions of pallidal (Borlongan et al., 1995; Baluchnejadmojarad and Roghani, 2004; adenosine A receptors in both heath and disease. Tabuse et al., 2010; Ingberg et al., 2015). The asymmetric 2A swing behavior has been attributed to the imbalance of motor control in the basal ganglia circuit. In present study, ETHICS STATEMENT we observed that unilateral microinjection of adenosine A 2A This study was carried out in accordance with the receptor agonist or antagonist into the globus pallidus displayed recommendations of the University guidelines on animal significant biased swing behavior ipsilaterally or contralaterally, ethics. The protocol was approved by an Animal Ethics respectively. Unilateral microinjection of adenosine A receptor 2A Committee of Qingdao University. agonist produces hypoactivity of the globus pallidus and then disinhibits the subthalamic nucleus. The thalamocortical activity is suppressed by enhanced GABAergic inhibition from the output AUTHOR CONTRIBUTIONS nucleus of the basal ganglia. Subsequently, the hypoactivity of ipsilateral motor cortex leads to the imbalance of the activity H-LD performed experiments and wrote the draft. H-LD, YX, X- HH, S-YW, CL, and W-FC analyzed the data. H-LD wrote the of bilateral limb muscles. The mechanism of adenosine A 2A manuscript. LC designed and supervised the project. receptor antagonist-induced contralateral swing is just contrary to that of adenosine A receptor agonist. Thus, the present 2A behavioral test suggested that pallidal adenosine A receptors ACKNOWLEDGMENTS 2A are involved in motor regulation. Moreover, the strong ipsilateral This work was supported by the grants from the National Natural bias was observed in present unilateral 6-OHDA-lesioned rats. Science Foundation of China (31671076), the Bureau of Science The possible mechanism may be that dopamine degeneration and Technology of Qingdao (14-2-3-1-nsh). decreases the activity of motor cortex on the lesioned side of REFERENCES Chan, C. S., Glajch, K. E., Gertler, T. S., Guzman, J. N., Mercer, J. N., Lewis, A. S., et al. (2011). HCN channelopathy in external globus pallidus neurons in models Abrous, D. N., Rodriguez, J. J., Montaron, M. F., Aurousseau, C., Le Moal, of Parkinson’s disease. Nat. Neurosci. 14, 85–92. doi: 10.1038/nn.2692 M., and Barneoud, P. (1998). Behavioural recovery after unilateral Chen, L., Xu, R., Sun, F. J., Xue, Y., Hao, X. M., Liu, H. X., et al. (2015). lesion of the dopaminergic mesotelencephalic pathway: effect of Hyperpolarization-activated cyclic nucleotide-gated (HCN) channels regulate repeated testing. Neuroscience 84, 213–221. doi: 10.1016/S0306-4522(97) firing of globus pallidus neurons in vivo. Mol. Cell. Neurosci. 68, 46–55. 00498-3 doi: 10.1016/j.mcn.2015.04.001 Baluchnejadmojarad, T., and Roghani, M. (2004). Evaluation of functional Chen, X. Y., Xue, Y., Wang, H., Zhu, S. H., Hao, X. M., and Chen, L. (2013). asymmetry in rats with dose-dependent lesions of dopaminergic nigrostriatal Modulation of firing activity by endogenous GABAA receptors in the globus system using elevated body swing test. Physiol. Behav. 82, 369–373. pallidus of MPTP-treated parkinsonian mice. Neurosci. Bull. 29, 701–707. doi: 10.1016/j.physbeh.2004.04.005 doi: 10.1007/s12264-013-1351-x Benhamou, L., Bronfeld, M., Bar-Gad, I., and Cohen, D. (2012). Globus Cooper, A. J., and Stanford, I. M. (2000). Electrophysiological and morphological pallidus external segment neuron classification in freely moving rats: a characteristics of three subtypes of rat globus pallidus neurone in vitro. J. comparison to primates. PLoS ONE 9:e45421. doi: 10.1371/journal.pone.00 Physiol. 527(Pt 2), 291–304. doi: 10.1111/j.1469-7793.2000.t01-1-00291.x 45421. Cunha, R. A. (2016). How does adenosine control neuronal dysfunction and Bibbiani, F., Oh, J. D., Petzer, J. P., Castagnoli, N. Jr., Chen, J. F., Schwarzschild, neurodegeneration? J. Neurochem. 139, 1019–1055. doi: 10.1111/jnc.13724 M. A., et al. (2003). A2A antagonist prevents dopamine agonist-induced motor Dayne Mayfield, R., Larson, G., Orona, R. A., and Zahniser, N. R. (1996). complications in animal models of Parkinson’s disease. Exp. Neurol. 184, Opposing actions of adenosine A2a and dopamine D2 receptor activation 285–294. doi: 10.1016/S0014-4886(03)00250-4 on GABA release in the basal ganglia: evidence for an A2a/D2 receptor Borlongan, C. V., Randall, T. S., Cahill, D. W., and Sanberg, P. R. (1995). interaction in globus pallidus. Synapse 22, 132–138. doi: 10.1002/(SICI)1098- Asymmetrical motor behavior in rats with unilateral striatal excitotoxic 2396(199602)22:2<132::AID-SYN6>3.0.CO2-E lesions as revealed by the elevated body swing test. Brain Res. 676, 231–234. Dodson, P. D., Larvin, J. T., Duffell, J. M., Garas, F. N., Doig, N. M., Kessaris, doi: 10.1016/0006-8993(95)00150-O N., et al. (2015). Distinct developmental origins manifest in the specialized Frontiers in Physiology | www.frontiersin.org 14 November 2017 | Volume 8 | Article 897 Diao et al. Pallidal Adenosine A Receptor 2A encoding of movement by adult neurons of the external globus pallidus. Neuron Kita, H., and Kita, T. (2001). Number, origins, and chemical types of 86, 501–513. doi: 10.1016/j.neuron.2015.03.007 rat pallidostriatal projection neurons. J. Comp. Neurol. 437, 438–448. Eid, L., Parent, A., and Parent, M. (2016). Asynaptic feature and heterogeneous doi: 10.1002/cne.1294 distribution of the cholinergic innervation of the globus pallidus in Kurokawa, M., Kirk, I. P. Kirkpatrick, K. A., Kase, H., and Richardson, primates. Brain Struct. Funct. 221, 1139–1155. doi: 10.1007/s00429-014- P. J. (1994). Inhibition by KF17837 of adenosine A2A receptor-mediated 0960-0 modulation of striatal GABA and ACh release. Br. J. Pharmacol. 113, 43–48. Ellens, D. J., and Leventhal, D. K. (2013). Review: electrophysiology of basal doi: 10.1111/j.1476-5381.1994.tb16171.x ganglia and cortex in models of Parkinson disease. J. Parkinsons Dis. 3, 241–254. Leblois, A., Boraud, T., Meissner, W., Bergman, H., and Hansel, D. doi: 10.3233/jpd-130204 (2006). Competition between feedback loops underlies normal and Ferre, S., Quiroz, C., Woods, A. S., Cunha, R., Popoli, P., Ciruela, F., pathological dynamics in the basal ganglia. J. Neurosci. 26, 3567–3583. et al. (2008). An update on adenosine A2A-dopamine D2 receptor doi: 10.1523/JNEUROSCI.5050-05.2006 interactions: implications for the function of G protein-coupled Mallet, N., Micklem, B. R., Henny, P., Brown, M. T., Williams, C., Bolam, J. P., et al. receptors. Curr. Pharm. Des. 14, 1468–1474. doi: 10.2174/1381612087844 (2012). Dichotomous organization of the external globus pallidus. Neuron 74, 80108 1075–1086. doi: 10.1016/j.neuron.2012.04.027 Floran, B., Gonzalez, B., Floran, L., Erlij, D., and Aceves, J. (2005). Interactions Martinez-Mir, M. I., Probst, A., and Palacios, J. M. (1991). Adenosine A2 receptors: between adenosine A(2a) and dopamine D2 receptors in the control of selective localization in the human basal ganglia and alterations with disease. [(3)H]GABA release in the globus pallidus of the rat. Eur. J. Pharmacol. 520, Neuroscience 42, 697–706. doi: 10.1016/0306-4522(91)90038-P 43–50. doi: 10.1016/j.ejphar.2005.06.035 Mayfield, R. D., Suzuki, F., and Zahniser, N. R. (1993). Adenosine A2a Fredholm, B. B., IJzerman, A. P., Jacobson, K. A., Klotz, K. N., and Linden, receptor modulation of electrically evoked endogenous GABA release J. (2001). International Union of Pharmacology. XXV. Nomenclature and from slices of rat globus pallidus. J. Neurochem. 60, 2334–2337. classification of adenosine receptors. Pharmacol. Rev. 53, 527–552. doi: 10.1111/j.1471-4159.1993.tb03526.x Fredholm, B. B., IJzerman, A. P., Jacobson, K. A., Linden, J., and Muller, C. Morales-Figueroa, G. E., Marquez-Gomez, R., Gonzalez-Pantoja, R., Escamilla- E. (2011). International Union of Basic and Clinical Pharmacology. LXXXI. Sanchez, J., and Arias-Montano, J. A. (2014). Histamine H3 receptor activation Nomenclature and classification of adenosine receptors–an update. Pharmacol. counteracts adenosine A2A receptor-mediated enhancement of depolarization- Rev. 63, 1–34. doi: 10.1124/pr.110.003285 evoked [3H]-GABA release from rat globus pallidus synaptosomes. ACS Chem. Gage, G. J., Stoetzner, C. R., Wiltschko, A. B., and Berke, J. D. (2010). Selective Neurosci. 5, 637–645. doi: 10.1021/cn500001m activation of striatal fast-spiking interneurons during choice execution. Neuron Mori, A., and Shindou, T. (2003). Modulation of GABAergic transmission in 67, 466–479. doi: 10.1016/j.neuron.2010.06.034 the striatopallidal system by adenosine A2A receptors: a potential mechanism Gonzalez, B., Paz, F., Floran, L., Aceves, J., Erlij, D., and Floran, B. (2006). for the antiparkinsonian effects of A2A antagonists. Neurology 61, S44–S48. Adenosine A2A receptor stimulation decreases GAT-1-mediated GABA doi: 10.1212/01.WNL.0000095211.71092.A0 uptake in the globus pallidus of the rat. Neuropharmacology 51, 154–159. Obeso, J. A., Rodriguez-Oroz, M. C., Benitez-Temino, B., Blesa, F. J., Guridi, J., doi: 10.1016/j.neuropharm.2006.03.011 Marin, C., et al. (2008). Functional organization of the basal ganglia: therapeutic Hamilton, M. E., Mele, A., and Pert, A. (1992). Striatal extracellular implications for Parkinson’s disease. Mov. Disord. 23(Suppl. 3), S548–S559. dopamine in conscious vs. anesthetized rats: effects of chloral hydrate doi: 10.1002/mds.22062 anesthetic on responses to drugs of different classes. Brain Res. 597, 1–7. Orru, M., Quiroz, C., Guitart, X., and Ferre, S. (2011). Pharmacological doi: 10.1016/0006-8993(92)91498-4 evidence for different populations of postsynaptic adenosine A2A Hegeman, D. J., Hong, E. S., Hernandez, V. M., and Chan, C. S. (2016). The receptors in the rat striatum. Neuropharmacology 61, 967–974. external globus pallidus: progress and perspectives. Eur. J. Neurosci. 43, doi: 10.1016/j.neuropharm.2011.06.025 1239–1265. doi: 10.1111/ejn.13196 Paxinos, G., and Watson, C. (1986). The Rat Brain in Stereotaxic Coordinates. New Hernandez, V. M., Hegeman, D. J., Cui, Q., Kelver, D. A., Fiske, M. P., Glajch, K. E., York, NY: Academic Press. et al. (2015). Parvalbumin+ neurons and npas1+ neurons are distinct neuron Preti, D., Baraldi, P. G., Moorman, A. R., Borea, P. A., and Varani, K. (2015). classes in the mouse external globus pallidus. J. Neurosci. 35, 11830–11847. History and perspectives of A2A adenosine receptor antagonists as potential doi: 10.1523/JNEUROSCI.4672-14.2015 therapeutic agents. Med. Res. Rev. 35, 790–848. doi: 10.1002/med.21344 Hurley, M. J., Mash, D. C., and Jenner, P. (2000). Adenosine A(2A) receptor Querejeta, E., Martinez-Romero, B., Miranda, J. E., and Delgado, A. (2010). mRNA expression in Parkinson’s disease. Neurosci. Lett. 291, 54–58. Modulation of the striato-pallidal pathway by adenosine A2a receptors doi: 10.1016/S0304-3940(00)01371-9 depends on dopaminergic striatal input. Brain Res. 1349, 137–142. Ingberg, E., Gudjonsdottir, J., Theodorsson, E., Theodorsson, A., and Strom, doi: 10.1016/j.brainres.2010.06.040 J. O. (2015). Elevated body swing test after focal cerebral ischemia Raz, A., Vaadia, E., and Bergman, H. (2000). Firing patterns and correlations of in rodents: methodological considerations. BMC. Neurosci. 16, 50. spontaneous discharge of pallidal neurons in the normal and the tremulous 1- doi: 10.1186/s12868-015-0189-8 methyl-4-phenyl-1,2,3,6-tetrahydropyridine vervet model of parkinsonism. J. Jarvis, M. F., and Williams, M. (1989). Direct autoradiographic localization of Neurosci. 20, 8559–8571. adenosine A2 receptors in the rat brain using the A2-selective agonist, [3H]CGS Rebola, N., Oliveira, C. R., and Cunha, R. A. (2002). Transducing system 21680. Eur. J. Pharmacol. 168, 243–246. doi: 10.1016/0014-2999(89)90571-2 operated by adenosine A(2A) receptors to facilitate acetylcholine Jellinger, K. A. (1991). Pathology of Parkinson’s disease. Changes other release in the rat hippocampus. Eur. J. Pharmacol. 454, 31–38. than the nigrostriatal pathway. Mol. Chem. Neuropathol. 14, 153–197. doi: 10.1016/S0014-2999(02)02475-5 doi: 10.1007/BF03159935 Ribeiro, J. A., Cunha, R. A., Correia-de-Sa, P., and Sebastiao, A. M. (1996). Jenner, P. (2014). An overview of adenosine A2A receptor Purinergic regulation of acetylcholine release. Prog. Brain. Res. 109, 231–241. antagonists in Parkinson’s disease. Int. Rev. Neurobiol. 119, 71–86. doi: 10.1016/S0079-6123(08)62107-X doi: 10.1016/B978-0-12-801022-8.00003-9 Rios, A., Barrientos, R., Alatorre, A., Delgado, A., Perez-Capistran, T., Chuc- Karain, B., Xu, D., Bellone, J. A., Hartman, R. E., and Shi, W. X. (2015). Rat globus Meza, E., et al. (2016). Dopamine-dependent modulation of rat globus pallidus pallidus neurons: functional classification and effects of dopamine depletion. excitation by nicotine acetylcholine receptors. Exp. Brain. Res. 234, 605–616. Synapse 69, 41–51. doi: 10.1002/syn.21783 doi: 10.1007/s00221-015-4491-6 Kelland, M. D., Freeman, A. S., and Chiodo, L. A. (1989). Chloral hydrate Rosin, D. L., Hettinger, B. D., Lee, A., and Linden, J. (2003). Anatomy of anesthesia alters the responsiveness of identified midbrain dopamine adenosine A2A receptors in brain: morphological substrates for integration neurons to dopamine agonist administration. Synapse 3, 30–37. of striatal function. Neurology 61, S12–S18. doi: 10.1212/01.WNL.0000095205. doi: 10.1002/syn.890030105 33940.99 Kita, H. (2007). Globus pallidus external segment. Prog. Brain. Res. 160, 111–133. Sani, S., Ostrem, J. L., Shimamoto, S., Levesque, N., and Starr, P. A. (2009). Single doi: 10.1016/S0079-6123(06)60007-1 unit “pauser” characteristics of the globus pallidus pars externa distinguish Frontiers in Physiology | www.frontiersin.org 15 November 2017 | Volume 8 | Article 897 Diao et al. Pallidal Adenosine A Receptor 2A primary dystonia from secondary dystonia and Parkinson’s disease. Exp. in mice. J. Clin. Neurosci. 17, 1412–1416. doi: 10.1016/j.jocn.2010. Neurol. 216, 295–299. doi: 10.1016/j.expneurol.2008.12.006 01.056 Saunders, A., Huang, K. W., and Sabatini, B. L. (2016). Globus Villar-Menendez, I., Porta, S., Buira, S. P., Pereira-Veiga, T., Diaz-Sanchez, pallidus externus neurons expressing parvalbumin interconnect the S., Albasanz, J. L., et al. (2014). Increased striatal adenosine A2A subthalamic nucleus and striatal interneurons. PLoS ONE 11:e0149798. receptor levels is an early event in Parkinson’s disease-related pathology doi: 10.1371/journal.pone.0149798 and it is potentially regulated by miR-34b. Neurobiol. Dis. 69, 206–214. Shindou, T., Nonaka, H., Richardson, P. J., Mori, A., Kase, H., and Ichimura, doi: 10.1016/j.nbd.2014.05.030 M. (2002). Presynaptic adenosine A2A receptors enhance GABAergic Xue, Y., Han, X. H., and Chen, L. (2010). Effects of pharmacological block of synaptic transmission via a cyclic AMP dependent mechanism in the GABA(A) receptors on pallidal neurons in normal and Parkinsonian State. rat globus pallidus. Br. J. Pharmacol. 136, 296–302. doi: 10.1038/sj.bjp. Front. Cell. Neurosci. 4, 2. doi: 10.3389/neuro.03.002.2010 0704702 Zold, C. L., Ballion, B., Riquelme, L. A., Gonon, F., and Murer, M. Shindou, T., Richardson, P. J., Mori, A., Kase, H., and Ichimura, M. G. (2007). Nigrostriatal lesion induces D2-modulated phase-locked (2003). Adenosine modulates the striatal GABAergic inputs to the globus activity in the basal ganglia of rats. Eur. J. Neurosci. 25, 2131–2144. pallidus via adenosine A2A receptors in rats. Neurosci. Lett. 352, 167–170. doi: 10.1111/j.1460-9568.2007.05475.x doi: 10.1016/j.neulet.2003.08.059 Sindhu, K. M., Saravanan, K. S., and Mohanakumar, K. P. (2005). Conflict of Interest Statement: The authors declare that the research was Behavioral differences in a rotenone-induced hemiparkinsonian rat conducted in the absence of any commercial or financial relationships that could model developed following intranigral or median forebrain bundle be construed as a potential conflict of interest. infusion. Brain. Res. 1051, 25–34. doi: 10.1016/j.brainres.2005. 05.051 Copyright © 2017 Diao, Xue, Han, Wang, Liu, Chen and Chen. This is an open- Suarez, L. M., Solis, O., Aguado, C., Lujan, R., and Moratalla, R. (2016). L- access article distributed under the terms of the Creative Commons Attribution DOPA oppositely regulates synaptic strength and spine morphology in D1 and License (CC BY). The use, distribution or reproduction in other forums is permitted, D2 striatal projection neurons in dyskinesia. Cereb. Cortex. 26, 4253–4264. provided the original author(s) or licensor are credited and that the original doi: 10.1093/cercor/bhw263 publication in this journal is cited, in accordance with accepted academic practice. Tabuse, M., Yaguchi, M., Ohta, S., Kawase, T., and Toda, M. (2010). No use, distribution or reproduction is permitted which does not comply with these A simple behavioral test for locomotor function after brain injury terms. Frontiers in Physiology | www.frontiersin.org 16 November 2017 | Volume 8 | Article 897

Journal

Frontiers in PhysiologyPubmed Central

Published: Nov 7, 2017

References