Interleukin-10 enhances activity of ventral tegmental area dopamine neurons resulting in increased dopamine release

Abstract

Dopamine transmission from the ventral tegmental area (VTA) to the nucleus accumbens (NAc) regulates important aspects of motivation and is influenced by the neuroimmune system. The neuroimmune system is a complex network of leukocytes, microglia and astrocytes that detect and remove foreign threats like bacteria or viruses and communicate with each other to regulate non-immune (e.g neuronal) cell activity through cytokine signaling. Inflammation is a key regulator of motivational states, though the effects of specific cytokines on VTA circuitry and motivation are largely unknown. Therefore, electrophysiology, neurochemical, immunohistochemical and behavioral studies were performed to determine the effects of the anti-inflammatory cytokine interleukin-10 (IL-10) on mesolimbic activity, dopamine transmission and conditioned behavior. IL-10 enhanced VTA dopamine activity and NAc dopamine levels via decreased VTA GABA currents in dopamine neurons. The IL-10 receptor was localized on VTA dopamine and non-dopamine cells. The IL-10 effects on dopamine neurons required post-synaptic phosphoinositide 3-kinase activity, and IL-10 appeared to have little-to-no efficacy on presynaptic GABA terminals. Intracranial IL-10 enhanced NAc dopamine levels in vivo and produced conditioned place aversion. Together, these studies identify the IL-10R on VTA dopamine neurons as a potential regulator of motivational states.

INTRODUCTION

The mesolimbic dopamine system, originating in the ventral tegmental area (VTA) and projecting to the nucleus accumbens (NAc) area of the striatum, is a major network involved in motivation and behavioral reinforcement for natural and drug rewards (Berridge & Kringelbach, 2015; Eshel et al., 2015; Wise, 2008). Local VTA inhibitory γ-aminobutyric acid (GABA) neurons regulate dopamine neuronal and terminal activity, and are a sensitive substrate for drugs of abuse, such as alcohol and opioids and play a critical role in regulating rewarding and reinforcing behaviors (Bonci & Williams, 1997; Gallegos, Criado, Lee, Henriksen, & Steffensen, 1999; Theile, Morikawa, Gonzales, & Morrisett, 2011; Wadsworth et al., 2023; Yorgason et al., 2022).

The neuroimmune system consists of a complex network of cells and signaling molecules consisting of local and peripheral leukocytes that signal to other immune cells, and non-immune cells the presence of foreign threats, such as bacterial or viral infections, in order to modulate activity of the non-immune cells through cytokine signaling (Kany, Vollrath, & Relja, 2019). Indeed, mice administered lipopolysaccharide (LPS), an important outer membrane component of gram-negative bacteria that triggers immune system activation, show decreased firing in VTA dopamine neurons (Blednov et al., 2011) and increased dopamine release in the NAc in vivo (Borowski, Kokkinidis, Merali, & Anisman, 1998), highlighting that inflammatory regulation of dopamine circuitry is complex and not all the regulators are known.

The cytokine interleukin-10 (IL-10) plays a key role in downregulating peripheral and central inflammation (Kany et al., 2019; Moore, de Waal Malefyt, Coffman, & O’Garra, 2001). Despite inflammation being associated with changes in mood and mesolimbic activity, it is unknown whether this specific cytokine affects VTA dopamine activity and related circuitry. Therefore, the current study uses electrophysiology, neurochemistry, immunohistochemistry and behavioral approaches to dissect the mesolimbic neurocircuitry effects of IL-10 dopamine neurons, dopamine transmission, and related behavior. Presently, IL-10 enhances dopamine activity via disinhibition, resulting in greater NAc dopamine levels, and enhanced conditioned learning for aversive stimuli.

METHODS

Animal Subjects

C57BL/6 mice, channelrhodopsin (ChR2) expressing VGAT-ChR2-EYFP mice, and glutamate-decarboxylase-67 (GAD-67)-green fluorescent protein knock-in on a CD-1 (white albino) mice (Tamamaki et al., 2003) were bred and cared for in accordance with the National Institutes of Health Guide for the Care and Use of Laboratory Animals. Rodents were treated in strict accordance with the Brigham Young University Animal Research Committee (IACUC) guidelines, which incorporate and exceed current NIH guidelines (BYU protocol #23–0101). Once weaned at PND 21, mice were housed in maximum groups of five and given ad libitum access to water and solid food and placed on a reverse light/dark cycle with lights ON from 10 PM to 10 AM. Dopamine transmission, clearance and related behavior are all heavily dependent on diurnal rhythms, with greatest dopamine release occurring during the light cycle when animals are least active (Ferris et al., 2014; Stowe et al., 2022). Therefore, in order to avoid possible ceiling effects in dopamine transmission and related behaviors, all experiments were performed so that IL-10 was administered between ZT12–18 (12–6 pm) and effects are interpreted based off behaviorally active timepoints.

Preparation of Brain Slices

Brain slice experiments were obtained as previously described (Ronstrom et al., 2023). Experiments were performed in P28–60-day old C57BL/6-ChR2-VGAT and GAD-GFP knock-in mice. Mice were anesthetized with isoflurane anesthesia (5%) and decapitated before extracting brains. Brains were then secured onto a cutting stage using a cyanoacrylate adhesive. The brain was then sectioned in carbonated (95% O₂, 5% CO₂) artificial cerebral spinal fluid (ACSF; in mM: 126 NaCl, 11.1 Glucose, 2.5 KCl, 1.2 NaH₂PO₄, 21.4 NaHCO₃, 1.2 MgCl₂, 2.4 CaCl₂). Targeting the VTA, horizontal slices (220 μm thick) were placed in an incubation chamber containing aCSF oxygenated for at least 30 minutes before being placed in a recording tissue chamber with aCSF continuously flowing at physiological temperatures (35 °C). Cytokines (IL-10, 20 ng/ml, Peprotech) were reconstituted and frozen in aliquots until ready to use.

Immunohistochemistry

Animals were anesthetized in 2–4% isoflurane and perfused with 4% paraformaldehyde (PFA). Brains were then extracted and held in 4% PFA for 24–48 hours after which brains were transferred to a 30% sucrose solution. Brains were sliced at 30 μm sections. The immunohistochemistry and fluorescence imaging of dopaminergic neurons in the VTA of C57 mice were visualized using an anti-Tyrosine hydroxylase (TH) antibody (Millipore, CAT# AB9702). Antibody labeling protocol was used to label IL-10R (alpha) (Novus Biologicals, Cat # AF-474-NA) and TH. The anti-IL-10R (alpha) antibody was conjugated using a CF488 labeling kit (Biotium, CAT#92211). The slices were incubated in 1x PBS (in mM: 137 NaCl, 3 KCl, 10.4 NaH2PO4, 1.8 NaHCO3). Slices were then incubated with 1:100 dilution of primary antibodies in PBS-T (1xPBS with 0.4% Triton X-100) for 2 hours and subsequently washed with 1 × PBS three times for 5 min each. Goat Anti-chicken Alexa Fluor (AF) 405 antibody (Thermo Fisher Cat # A48260) was applied 1:100 with PBS-T to the slices for 2 hours. Unbound secondary antibody was removed by washing three times for 5 min each with PBS-T. Coverslips were mounted onto glass microscope slides using mounting media (Fluoromount G, ThermoFisher). Images were captured using an Olympus Fluoview FV1000 microscope using an Olympus UPlanAPO 60x oil HR 1.5 NA objective (Olympus). The AF405 and CF488 fluorophores were excited with 405 and 488 nm lasers, respectively.

Electrophysiology Recordings in Brain Slices

Electrophysiology studies utilized borosilicate glass capillary electrodes (2.5–6 MΩ) filled with one of four solutions. The solutions used were: 1. ACSF solution (described above) 2. KCl internal solution, consisting of 145 mM KCl, 8 mM NaCl, 0.2 mM MgCl₂, 2 mM EGTA, 10 mM HEPES, and 2 mM Mg-ATP, for studying optogenetically-evoked inhibitory post-synaptic currents (oIPSC) or miniature IPSCs (mIPSC). 3. KCl internal solution containing the phosphoinositide 3-kinase (PI3K) inhibitor (LY294002; 10 μM), Cayman Chemical, Item #: 70920). 4. KCl internal solution containing a negative control structural homolog of the LY294002 PI3K inhibitor (LY303511; 10 μM; Cayman Chemical, Item #: 15514). This pharmacological condition was used to control for some of the off-target effects of LY294002.

For cell-attached recordings of neuronal firing (aCSF internal), a seal (10MΩ – 1GΩ) was created between the cell membrane and the recording pipette. Spontaneous spike activity was then recorded in voltage-clamp mode with an Axon Instruments Multiclamp 700B (Molecular Devices, Sunnyvale, CA, USA) amplifier and sampled at 10 kHz using an Axon 1440A digitizer (Molecular Devices), and collected and analyzed using pClamp10 (Molecular Devices, San Jose, CA, USA) or Axograph 10 (Axograph, Sydney, Australia). A stable baseline recording of current activity was obtained for 5–10 min before adding any substances. Whole-cell voltage clamp recordings were filtered at 2 kHz with an Axon Instruments Multiclamp 700B amplifier and digitized at 5 to 20 kHz respectively using an Axon 1440A digitizer. Axon Instruments pClamp ver10, Axograph10, Mini Analysis (Synaptsoft: Decatur, GA), and Igor Pro (Wavemetrics: Oswego, OR) software packages were utilized for data collection and analysis. Blue light (~470 nm) was used to evoke oIPSCs. Using a paired-pulse stimulus, the stimulation light intensity was adjusted to a half-max level in order to allow the measurement of changes that increase or decrease oIPSC levels. For experiments where boiled IL-10 (20 ng/ml) was used, the solution containing the IL-10 was placed in a microcentrifuge tube and submerged in boiling water for at least one minute. Recordings of mIPSCs occurred in aCSF with 3 mM kynurenic acid (Sigma, Cat. No. K3375) and 100–500 μM lidocaine (Sigma, Cat. No. L5647) to block glutamatergic EPSCs followed by IL-10 (10–40 ng/ml; Suryanarayanan et al., 2016).

Ex Vivo Voltammetry

Slices were prepared as described above, transferred to the recording chamber, and perfused with aCSF (34°C) at a rate of ~1.8 ml/min. Fast scan cyclic voltammetry recordings were performed and analyzed using Demon Voltammetry and Analysis software (Yorgason, Zeppenfeld, & Williams, 2017)(RRID:SCR_014468). The carbon fiber electrodes used for voltammetry experiments were manufactured in-house. Briefly, a carbon fiber (7 μm diameter, Thornel T-650, Cytec) was aspirated into a borosilicate glass capillary tube (TW150, World Precision Instruments (WPI), Sarasota, FL). Electrodes were pulled on a P-87 Horizontal pipette puller (Sutter Instruments, Novato, CA) and cut so that 100–150 μm of carbon fiber protruded from the tip of the glass. The electrode potential was linearly scanned as a triangular waveform from −0.4 to 1.2 V and back to −0.4 V (Ag vs AgCl) with a scan rate of 400 V/s. Carbon fibers were advanced completely into the tissue at a 20° angle with the tip positioned ~85 μm below the slice surface. Dopamine release was evoked through electrical stimulation (1 pulse/min) via a glass micropipette (30 μA, monophasic⁺, 0.5 ms). Paired pulse stimulations were performed with interstimulus intervals of 0.5, 1, 2, 4, 8, 12, and 16 Hz.

Microdialysis and High-Performance Liquid Chromatography

A microdialysis probe (CMA 7, Harvard Apparatus, Holliston, MA, USA) was stereotaxically inserted into the NAc at the following coordinates (from bregma): +1.5 AP, ±0.6 ML, −5.0 DV (16 mice, 8/condition). The aCSF was perfused through the probe at a rate of 2.0 μl/min. Samples were collected every 20 minutes for a baseline period of at least 2 hours. Once a stable baseline was established, an intracerebroventricular (ICV) microinjection of 60 ng of IL-10 in 1 μl of aCSF (or 1 μl aCSF alone) was administered. The microinjection was carried out using a 10 μL Hamilton syringe (Reno, NV, USA) with a 25-gauge needle in conjunction with a microsyringe pump injector (UMP3, World Precision Instruments (WPI), Sarasota, FL, USA) attached to a microdrive controller (Micro4, WPI, Sarasota, FL, USA). The coordinates for the microinjection were as follows (from bregma): −0.5 AP, ±1.1 ML, −2.4 DV. Microdialysis sampling continued for an additional 2 hours following the microinjection. Statistical analysis was performed over the first hour in order to facilitate comparison across electrophysiology and microdialysis experiments. Mice were anesthetized for the duration of the procedure using isoflurane (1 – 2%).

Determination of dopamine concentration in the microdialysis samples was performed using a high-performance liquid chromatography (HPLC) system (Ultimate 3000, Thermo Fisher Scientific, Waltham, MA, USA) coupled to an electrochemical detector (Coulochem III, ESA, Paway, CA). The electrochemical detector included a guard cell (5020, ESA) set at +275 mV, a screen electrode (5014B, ESA) set at −100 mV, and a detection electrode (5014B, ESA) set at +220 mV. Dopamine was separated using a C18 reverse phase column (HR-80, Thermo Fisher Scientific, Waltham, MA). Mobile phase containing 75 mM H₂NaO₄P, 1.7 mM sodium octane sulfonate, 25 μM ethylenediaminetetraacetic acid, 0.71 mM trimethylamine, and 10% acetonitrile was pumped through the system at a flow rate of 0.5 ml/min. Dopamine levels following the ICV microinjection were expressed as percentage of baseline, with the baseline being determined by the average dopamine concentrations of 3 consecutive stable collections occurring prior to the microinjection. The first hour of microdialysis data on dopamine release in the NAc was analyzed using a mixed model analysis of variance (ANOVA) with time as a within-subjects factor, treatment condition as a between-subjects factor, and a time by treatment condition interaction term. A Greenhouse-Geisser correction for sphericity was applied to the results of the ANOVA. Prior to analysis, the data were checked for outliers (classified as a datum falling ±2 interquartile ranges (IQR) beyond the median). Two data points were identified as outliers using this method. These outliers were subsequently fenced to the upper limit of the outlier test (median + 2 IQR) as we did not find any evidence of data input error or variation from experimental protocol that would have justified the removal of these data points. The data were also found to be normal using the Shapiro-Wilk test.

Place Conditioning

Place conditioning occurred in a 16” × 16” × 16” apparatus constructed out of Plexiglass. The floor of the apparatus was divided into an 8” × 16” rough compartment and an 8” × 16” smooth compartment on testing days by the insertion of a 1/8” thick piece of textured acrylic. During conditioning days, the insertion or removal of a 16” × 16” piece of textured acrylic allowed pairing of the different treatment conditions with the different contexts.

The place conditioning procedure consisted of 3 phases: pre-test, conditioning, and post-test. The pre-test and post-test phases occurred over one day each: The mice (10 mice per condition) were individually placed into the apparatus and allowed to choose between the two contexts for 30 minutes on these days. The pre-test and post-test sessions occurred prior to beginning and after completion of the conditioning trials, respectively. The conditioning trials occurred over the course of three days. Each day, the mice were lightly anesthetized with isoflurane (3%) and given an intra-cisterna magna (ICM) injection of aCSF (1 μl) in the morning and an ICM injection of IL-10 (60 ng in 1 μl aCSF) or vehicle (1 μl aCSF) in the afternoon. Immediately following the injections, the mice were transported to the place conditioning apparatus and placed in the smooth compartment (after an aCSF injection) or the rough compartment (after an IL-10 injection) after waking up from the anesthesia. The mice were left in the compartment for 10 minutes for each conditioning trial. The apparatus has been found to be biased, with mice showing an initial preference for the smooth compartment. Thus, the rough compartment is an aversive conditioner.

The time spent in the IL-10 or vehicle paired chamber was analyzed using a mixed models ANOVA with treatment condition as a between-subjects factor, test (pre- or post-) as a within-subjects factor, and treatment condition by test as an interaction term. A Greenhouse-Geisser correction for sphericity was applied to the results of the ANOVA. Data were cleaned as described in the microdialysis and high-performance liquid chromatography section. One datum was identified as an outlier and bounded to the median + 2 IQR limit for outlier detection. The data were found to be relatively normal using the Shapiro-Wilk test.

Statistical Analyses

All results are presented as raw mean values and percent control ± SEM. Results before and after drug exposure were compared using a two-tailed paired t-test, or Kolmogorov-Smirnov test for frequency distribution comparisons. Experiments relying on variance in frequency or current amplitude were analyzed using a one-way or two-way repeated measures ANOVA with Bonferroni or Tukey’s HSD post-hoc tests for comparisons between individual points. Statistical significance required ≥ 95% level of confidence (p≤0.05). Analysis software included Microsoft Excel, Igor Pro (Wavemetrics, Oswego, OR), Prism 5 (GraphPad, San Diego, CA) and NCSS 8 (NCSS LLC, Kaysville UT). Significance levels are indicated on graphs with asterisks *,**,*** and correspond to significance levels p<0.05, 0.01 and 0.001, respectively. Figures were constructed with Igor Pro, Prism 5 (GraphPad, San Diego, CA), and Adobe Illustrator software.

RESULTS

IL-10 Increases the Firing Rate of VTA Dopamine Neurons

The activity of VTA dopamine and GABA neurons was measured before and after exposure to the anti-inflammatory cytokine IL-10 (Figure 1). Both dopamine and GABA neurons were examined since VTA GABA neurons are upstream from and known to regulate VTA dopamine neuron firing (Steffensen, Svingos, Pickel, & Henriksen, 1998). Bath application of IL-10 (20 ng/ml) increased VTA dopamine neuron firing rate by 70.63 ± 20.87% (Figure 1AB, paired t-test; t₍₁₂₎=4.99 n=13, p<0.001), but had no effect on the firing rate of VTA GABA neurons 0.63 ± 5.32% (Figure 1CD, paired t-test; t₍₁₂₎=0.59, n=13, p=0.56). Thus, anti-inflammation appears to enhance mesolimbic activity through effects on dopamine neurons.

Figure 1. IL-10 enhances VTA dopamine and not GABA neuron activity.

(A) Representative traces of the effects of IL-10 (20 ng/ml) on the firing rate of a dopamine neuron. (B) IL-10 increased VTA dopamine neuron firing rate. (C) Representative traces of the effects of IL-10 on the firing rate of a GABA neuron. (D) IL-10 had no effect on VTA GABA neuron firing rate. ***, p<0.001

The IL-10Rα subunit is responsible for binding selectively to IL-10 (Josephson, Logsdon, & Walter, 2001). Therefore, IL-10Rα expression was examined using confocal microscopy in the VTA to determine IL-10R localization in relation to dopamine neurons. Immunohistochemistry was used to label dopamine neurons and IL-10Rα in the VTA using fluorescent conjugated antibodies against tyrosine hydroxylase (TH; 405 nm) and IL-10Rα (488 nm). IL-10 receptors were observed on dopamine and non-dopamine cells (Figure 2), suggesting that IL-10 will produce effects on dopamine activity that could involve pre- and post-synaptic effects.

Figure 2. Localization of IL-10Rα in the Ventral Tegmental Area.

Immunohistochemistry was used to investigate the distribution of IL-10Rα subunits in the VTA of C57 mice. (A) Tyrosine hydroxylase (TH) antibodies were used to label dopamine neurons (blue) and antibodies for (B) IL-10Rα (green). (C) The merged channel for TH and IL10Rα indicates expression of IL-10 receptors localized on TH positive dopamine neurons (small arrows), and non-dopamine cells (large arrows) of the VTA. (D) Enlarged merge image from dashed box depicted in C.

Effects of IL-10 on Optogenetically-Evoked IPSCs on VTA Dopamine Neurons

VTA GABA neuron firing was unaffected by IL-10 (Figure 1CD). However, non-dopamine cells also express IL-10Rα (Figure 2), and IL-10 induced increases in dopamine firing could potentially still be explained by changes in GABA release onto dopamine neurons. This is especially important since GABA inputs to VTA dopamine neurons come from local GABA neurons, as well as distal GABAergic inputs (Matsui, Jarvie, Robinson, Hentges, & Williams, 2014). Previous studies have shown that IL-10 inhibits GABAergic synaptic inhibition in the hippocampus via a postsynaptic mechanism (Suryanarayanan et al., 2016). Therefore, GABA release onto dopamine neurons was examined, and the effects of IL-10 measured on VTA dopamine neurons. Dopamine neurons were visually identified and verified by recording an Ih-current (Figure 3A). Transgenic ChR2 vesicular GABA transporter (VGAT) mice were used to optically stimulate GABA terminals (oIPSC; Figure 3B) while recording from dopamine neurons. IL-10 perfusion (20 ng/ml) decreased oIPSC amplitude on dopamine neurons by 21.69 ± 6.31 % (Figure 3C, paired t-test; t₍₉₎=2.40, n=10, p<0.05), with no effect on the paired pulse ratio (Figure 3D, paired t-test; t₍₉₎=1.39, n=10, p=0.25). Therefore, IL-10 reduces GABA currents through a postsynaptic mechanism, which likely leads to disinhibition of dopamine neurons.

Figure 3. IL-10 reduces evoked GABA IPSCs onto VTA dopamine neurons.

(A) Dopamine neurons were identified by large size and presence of Ih current. (B) Representative trace of the effects of IL-10 (20 ng/ml) on the optogenetically-evoked IPSCs (oIPSC) in a VTA dopamine neuron. This neuron had an IPSC that was decreased from 340pA to 190pA during IL-10 perfusion. (C) IL-10 decreased the amplitude of oIPSCs in 9 out of 10 neurons. (D) oIPSC paired-pulse ratio did not show any difference between IL-10 and baseline conditions. *, p<0.05

Effects of IL-10 on mIPSCs on VTA Dopamine Neurons

To corroborate the paired-pulse postsynaptic IL-10 effects on stimulated IPSCs, miniature IPSC (100 μM lidocaine) amplitude and frequency changes were measured in VTA dopamine neurons before and after IL-10 exposure. Changes in mIPSC amplitude and frequency are thought to reflect postsynaptic and presynaptic changes respectively (Auger & Marty, 1997). The mIPSCs were measured in the presence of increasing concentrations of IL-10 (Figure 4; 10 ng/ml, 20 ng/ml and 40 ng/ml). IL-10 significantly decreased the amplitude of mIPSCs in a dose-dependent manner (Figure 4B, one-way ANOVA, F_{(3, 32)}=4.35, n=9, p<0.05 with a Tukey’s post hoc test showing significance for Baseline versus 20 and 40 ng/ml, q=3.90, q=4.11 respectively p<0.05). However, IL-10 did not affect the frequency of mIPSCs (Figure 4C, one-way ANOVA, F_{(3, 32)}=1.4, n=9, p=0.26 and a Tukey’s post hoc test revealed no differences). Furthermore, application of heat-denatured IL-10 resulted in a loss of this inhibitory effect (Figure 4D, paired t-test; t₍₂₎=1.94, n=3, p=0.19). These results confirm that IL-10 reduces GABA input onto VTA dopamine neurons via a postsynaptic mechanism.

Figure 4. IL-10 reduces GABA miniature IPSCs (mIPSCs) onto VTA dopamine neurons.

(A) Representative traces of the effects of IL-10 on mIPSC in a VTA dopamine neuron (B) IL-10 (20 and 40 ng/ml) decreased the amplitude of mIPSCs on VTA dopamine neurons. (C) IL-10 showed a trend in decreased frequency but showed no significant effect of mIPSCs in VTA dopamine neurons. (D) Boiled IL-10 was used as a control for mIPSC recordings and showed no effect on amplitude. *, p<0.05

PI3K Inhibitor Blocks IL-10 Effects on VTA Dopamine neurons.

To evaluate the postsynaptic mechanism by which IL-10 inhibits GABA signaling, mIPSCs were measured with a PI3K inhibitor dialyzed into VTA dopamine neurons via the internal patching solution. The PI3K inhibitor LY294002 (10 μM) and its structural homologue LY303511 (10 μM) were tested alongside each other, with LY303511 acting as the negative control. In control LY303511 experiments, IL-10 reduced mIPSC amplitude (Figure 5ALeft; LY303511; paired t-test: t₍₇₎=2.40, n=8, p<0.05), which is consistent with postsynaptic results in Figure 4B. In contrast, inclusion of PI3K inhibitor LY294002 in the recording pipette prevented the inhibitory effects of IL-10 on mIPSC amplitude (Figure 5ARight; LY294002; paired t-test: t₍₆₎=0.06, n=7, p=0.96). Therefore, IL-10 postsynaptic effects on dopamine neurons include activation of PI3K in dopamine neurons and downstream effects on GABAA receptors.

Figure 5. IL-10 postsynaptic effects are through PI3 kinase activity.

(A) In control conditions (10 μM LY303511 in patch pipette), IL-10 (20 ng/ml) reduced mean mIPSC amplitude. This effect was precluded in the presence of the PI3K inhibitor (10 μM LY294002 in patch pipette). (B) IL-10 had no effect on mIPSC frequency under control conditions, but reduced mIPSC frequency in the presence of LY294002. (C) Cumulative amplitude and (D) inter-event interval distribution plots where IL-10 reduced amplitude and increased frequency in control conditions (LY303511). In contrast, with the PI3K inhibitor, IL-10 had no effect on amplitude (C) but decreased mIPSC frequency. *,*** p<0.05, p<0.001

Importantly, the effects of IL-10 on mIPSC frequency in control conditions (LY303511) were similar in nature to our results in Figure 4C, in that IL-10 had no statistical effect on mean mIPSC frequency (Figure 5BLeft; LY303511; paired t-test: t₍₇₎=0.02, n=8, p=0.99). However, in the presence of PI3K inhibitor LY294002, IL-10 reduced mean mIPSC frequency (Figure 5BRight; LY294002; paired t-test: t₍₆₎=2.54 n=7, p<0.05). Frequency distribution plots were made of mIPSC events and corroborate these data, showing that in control conditions (LY303511), mIPSC amplitude is reduced (Figure 5CLeft; LY303511; Kolmogorov-Smirnov test: Baseline: Mean=0.07 nA ± 0.001 nA SEM and IL-10: Mean=0.06 nA ± 0.001 nA, D-value=0.09, p<0.001), but is significantly enhanced by IL-10 in the presence of LY294002 (Figure 5CRight; LY294002; Kolmogorov-Smirnov test: Baseline: Mean=0.04 nA ± 0.001 SEM and IL-10: Mean=0.05 nA ± 0.001 nA SEM, D-value=0.08, p<0.001). Interestingly, instantaneous inter-event intervals for these data show clear increases in the amount of smaller intervals in control conditions (Figure 5DLeft; LY303511; Kolmogorov-Smirnov test: Mean=1205.54 ms ± 47.26 ms SEM and IL-10: Mean=1217.61 ms ± 78.88 ms SEM, D-value=0.27, p<0.001). In contrast, LY294002 treated neurons show a significant increase in the probability of observing longer intervals after IL-10 (Figure 5DRight; LY294002; Kolmogorov-Smirnov test: Baseline: Mean=676.13 ms ± 22.11 ms SEM and IL-10: Mean=799.72 ms ± 28.71 ms SEM, D-value=0.05, p<0.001). These results confirm our other findings that IL-10 effects are predominantly postsynaptic, and extend our findings that postsynaptic effects are mediated through the PI3K signaling pathway. The additional frequency distribution analysis suggests that other non-PI3K targets may also be involved in IL-10 presynaptic regulating effects on dopamine neurons. However, these effects were not potent enough to manifest in a simple means test as shown in Figure 5BLeft.

Effects of IL-10 on Dopamine Release in the NAc

Based on our findings of IL-10 induced increases in dopamine neuron firing, the effects of IL-10 on NAc dopamine release were examined. First, fast scan cyclic voltammetry was used to investigate direct effects of IL-10 on evoked dopamine release from NAc dopamine terminals in brain slices where dopamine cell bodies are no longer connected (Figure 6). A paired-pulse protocol of evoked dopamine release was used to determine if IL-10 effects on dopamine release were sensitive to additional intrinsic or synaptic inputs that may be present during high and low frequency stimulations. Paired pulse inhibition was observed for the second pulse (Figure 6AB). Paired-pulse inhibition was reduced (peaks were larger) at longer intervals across control and IL-10 conditions. IL-10 had minimal effects on evoked dopamine release or interaction between these terms, except the slowest (33.3 seconds) interval (Figure 6C, repeated measures Two-Way ANOVA within-subject comparison for both factors; IL-10: F_(1,69)=1.58, p=0.28; interstimulus interval: F_(6,64)=49.87, p<0.001; interaction: F_(6,64)=2.62, p<0.05, with a Tukey’s post hoc analysis test, baseline vs IL-10 at 33.3 seconds, p<0.05). Therefore, IL-10 effects on dopamine terminals in a reduced preparation are minimal, and mostly manifest in lower frequency stimulations.

Figure 6. Effects of IL-10 on Evoked dopamine Release in the NAc Ex Vivo Using Fast Scan Cyclic Voltammetry and Microdialysis.

(A) Example trace protocol for paired-pulse ratio and cyclic voltammogram. (B) Example of FSCV color plot indicating a dopamine signal. (C) A paired-pulse ratio protocol was tested and then repeated with the superfusion of IL-10 (20 ng/ml). There was no significant difference between the control and IL-10 except at the very last time point of the stimulation intervals. (D) Microinjections of IL-10 (60 ng, ICV) moderately enhance dopamine release in the NAc at 20 and 40 minutes after the microinjection. Microinjections of vehicle (1 μl aCSF, ICV) failed to enhance DA release in the NAc. *, p<0.05

In vivo intra-NAc dopamine release was also measured via microdialysis following an ICV injection of IL-10 (60 ng/mouse). ICV microinjections of IL-10 modestly enhanced NAc dopamine release, an effect that was greatest during the first hour following the microinjection (Figure 6D, repeated measures two-way ANOVA, between-subject factor of IL-10 exposure; IL-10: F_(1,41) = 11.05, η2p=0.439 p<0.01). Bonferroni post-test comparisons of individual timepoints during the first hour revealed that dopamine release in the NAc was enhanced specifically at 20 (t₍₇₎=3.12, p<0.05) and 40 (t₍₇₎=3.10, p<0.05) minutes post IL-10 microinjection. There was not a significant effect of time (F_(3,39)=1.23, p=0.31) or a time by treatment condition interaction on dopamine release (F_{(3, 39)}=1.48, p=0.16). Thus, IL-10 induced increases in NAc dopamine levels are likely due to increased VTA dopamine firing.

Effects of IL-10 on Place Conditioning

Microdialysis and electrophysiology results indicate that IL-10 increases NAc dopamine likely through disinhibition and subsequent increases in dopamine cell firing. NAc dopamine elevations are heavily implicated in learning for aversive and rewarding stimuli (Bromberg-Martin, Matsumoto, & Hikosaka, 2010). Therefore, behavioral experiments were performed to test the effects of intracranial IL-10 infusions on conditioned learning. Intracranial IL-10 ICM injections were performed in mice during a place conditioning task (Figure 7). In Figure 7A is a diagram of the conditioning procedures, which is described in detail in the methods section. Briefly, the place conditioning procedure involved three phases: pre-test, conditioning, and post-test. During the conditioning trials, mice received ICM injections of aCSF (1 μl) or IL-10 (60 ng), followed by placement in either a smooth or rough compartment for 10 minutes. The rough compartment served as an aversive conditioner due to the mice’s initial bias for the smooth compartment. Place conditioning was then measured during the post-test day. IL-10 produced conditioned place aversion (Figure 7B; Two-way mixed ANOVA with IL-10 as the between subject factor; IL-10: F_(1,18)=6.40, p<0.05; Time effect: F_(1,18)=2.14, p=0.16; Interaction: F_(1,18)=1.02, p=0.33) as indicated by decreased time spent in the IL-10 paired rough chamber compared with vehicle chamber controls (Bonferroni’s post hoc test: t₍₁₇₎=2.39, p<0.05). Importantly, there was no difference between IL-10 and vehicle chamber time during the pre-test phase (t₍₁₈₎=0.82, p>0.05). The CS− time is also included to assess magnitude of conditioning (Figure 7C). IL-10 increased time spent in CS− during the post-test (Figure 7C; Two-way repeated measures ANOVA with IL-10 exposure and conditioning as within subject factors; IL-10: F_(1,18)=6.39, p<0.05; Time effect: F_(1,18) =2.14, p=0.16; Interaction: F_(1,18) =1.02, p=0.33; Bonferroni’s posthoc analysis revealed a difference between IL-10 and vehicle during post-conditioning: t₍₁₇₎=2.39, p<0.05). Thus, intracranial IL-10 enhances conditioned learning, likely through induced dopamine changes.

Figure 7. IL-10 increases learned aversion during place conditioning paradigm.

(A) Place conditioning IL-10 administration protocol. (B) Interleukin-10 administration (60 ng, ICM) produced a conditioned place aversion. Avoidance of a chamber paired with IL-10 administration (CS+) was increased following the pairings as compared with a group that received only vehicle pairings (1 μl aCSF, ICV) in the same chamber. (C) Place conditioning CS− chamber results are also shown to indicate the magnitude of conditioning. *, p<0.05

DISCUSSION

The present study investigated the effects of the cytokine IL-10 on VTA circuit activity and downstream dopamine release. VTA dopamine neuron firing increases from IL-10 activation due in part to decreases in GABAergic synaptic inhibition onto dopamine neurons. The IL-10Rα subunit was observed on dopamine soma, supporting direct postsynaptic effects. The receptor was also observed on some non-dopaminergic cells in the midbrain, suggesting that other regional effects may exist. In terminal regions, in vivo microdialysis measures of NAc dopamine were also increased by IL-10, though evoked dopamine release (as measured by voltammetry) was relatively insensitive to IL-10. Thus, the majority of IL-10 dopamine excitatory effects appear to be through somatic interactions. Remarkably, IL-10 administration resulted in conditioned place aversion, indicating a multifaceted role of this anti-inflammatory cytokine in modulating dopamine transmission and motivation.

Interleukin-10 Signaling and Inflammation

During inflammatory states of infection, immune cells of the leukocytic network respond by first releasing pro-inflammatory and second releasing anti-inflammatory cytokines in a balanced and orchestrated manner. The initial inflammatory response is thought to contain and/or remove the threat, which is followed by the slower anti-inflammatory response that prevents unnecessary tissue damage and restore normal function (For review see (Kany et al., 2019)). As an anti-inflammatory cytokine, IL-10 is transcriptionally upregulated and expressed/released by leukocytes in general (Kany et al., 2019), which maintain IL-10 serum levels constitutively at 4.8–9.8 pg/mL in humans (Sarris et al., 1999) and mice (Hillyer & Woodward, 2003). Importantly, these values are likely underestimates, due to nuances with ELISA collections, and some measures of serum IL-10 levels are in the ng range (Hillyer & Woodward, 2003). While the extent of IL-10 cytokine penetration across the blood brain barrier at the VTA is unknown, this constitutive activity is at least suggestive of a baseline peripheral immune activity level that may influence CNS dopamine activity. Peripheral leukocytic cytokines control CNS cytokine release from local monocytes (i.e. microglia; (Lobo-Silva, Carriche, Castro, Roque, & Saraiva, 2016)) and astrocytes (Hulshof, Montagne, De Groot, & Van Der Valk, 2002). Furthermore, IL-10 production is thought to be mostly limited to local glia cells (Hillyer & Woodward, 2003) or peripheral sources (Moore et al., 2001). Since IL-10R is also localized on non-dopamine cells, an important consideration is that IL-10R are likely expressed on local glial cells, and that activation may influence secondary cytokine release, an implication that was not presently explored. The present work was performed in adolescents and it is currently unknown how mesolimbic IL-10 levels vary across development. However, dopamine release is developmentally regulated and decreases as animals age (Walker et al., 2010), which could be related to IL-10 levels or receptor activity. Interestingly, NAc IL-10 gene expression and levels are upregulated following early life environmental stimuli and/or drug exposure, which continues into adulthood (Schwarz, Hutchinson, & Bilbo, 2011). Early life stress and trauma are also associated with changes in inflammatory responses (Cattaneo et al., 2015), and stress is a known regulator of dopamine release (Yorgason et al., 2016; Yorgason, Espana, Konstantopoulos, Weiner, & Jones, 2013). Therefore, IL-10 expression and receptor mediated effects on dopamine release across developmental periods may uncover important interactions between these two signaling molecules. The role of development on systemic and brain specific IL-10 expression remains an area of active investigation (Almanan et al., 2020; Porcher et al., 2021; Porro, Cianciulli, & Panaro, 2020; Zhang, Bailey, Braun, & Gensel, 2015) and is something that is important to address in the future.

The IL-10R typically consists of a dimer between IL-10Rα and IL-10Rβ subunits. The dimerization of these subunits is essential for signaling since IL-10Rα has a higher affinity for IL-10 (K_d = 50–200 pM) compared to IL-10Rβ (K_d = 2–4 mM range) (Tan, Indelicato, Narula, Zavodny, & Chou, 1993), and IL-10Rβ enhances the binding of IL-10 to IL-10Rα, and recruits the downstream signaling effectors (Walter, 2014). Importantly, the IL-10Rβ subunit is also able to dimerize with other cytokine receptors, including IL-22, IL-26, IL-28 and IL-29, and receptor specificity for IL-10 is thought to be mostly due to the IL-10Rα subunit (Donnelly, Sheikh, Kotenko, & Dickensheets, 2004; Jones, Logsdon, & Walter, 2008). Thus, these other cytokine receptors containing the IL-10Rβ subunit may have similar effects on dopamine function, though this will need to be explored explicitly.

Activation of the IL-10R activates the Jak/STAT or tyrosine kinase (TYK) pathway (Walter, 2014), reducing inflammatory cytokine production and decreasing inflammation (Murray, 2006). For instance, inhibition of the JAK1 pathway prevents IL-10 induced reductions in LPS mediated superoxide production (Qian & Flood, 2008). Additionally, conditional knockout of the JAK/STAT effector STAT3 reduces macrophages IL-10 sensitivity, suggesting that this pathway is involved in IL-10R effects for macrophages in general (Riley, Takeda, Akira, & Schreiber, 1999; Takeda et al., 1999). Furthermore, IL-10Rβ activates the TYK2 pathway which signals through the PI3K and subsequently the AKT pathway, and these effectors appear responsible for IL-10’s neuroprotective effects (Kwilasz, Grace, Serbedzija, Maier, & Watkins, 2015; Strle et al., 2002).

The present findings identified a postsynaptic role for PI-3K in IL-10 mediated increases in dopamine neuron activity. However, since IL-10R activates many different signaling pathways, these other pathways may also influence dopaminergic function, particularly during prolonged inflammatory and subsequent IL-10 anti-inflammatory states (Antoniv & Ivashkiv, 2011; Verma et al., 2016). Notably, PI3K/AKT activity has been implicated in both increasing (Pribiag & Stellwagen, 2013; Serantes et al., 2006; Vetiska et al., 2007) and decreasing (Suryanarayanan et al., 2016) GABA_A receptor membrane levels, an effect that likely depends on the GABA subunit actively being phosphorylated (Pribiag & Stellwagen, 2013; Vetiska et al., 2007). Although present mechanisms were predominantly postsynaptic, in instances where IL-10 induced changes are presynaptic, these same kinases are likely involved, albeit through different presynaptic mechanisms (Patel et al., 2021; Suryanarayanan et al., 2016).

Interleukin-10 Receptor Expression and Functional Activity in VTA Dopamine neurons

Expression of IL-10R has been shown previously throughout the brain, including in cells of the hippocampus, cortex and spinal cord (Lim et al., 2013). The present work extends these findings by showing IL-10R expression on VTA dopamine and non-dopamine cells. The functional physiological results herein are confirmatory of postsynaptic effects of IL-10 on GABA currents within dopamine neurons. The presence of IL-10R on non-dopamine cells suggests that additional VTA cells may be targeted by IL-10. Therefore, some of the IL-10R expression observed on non-dopamine cells may be interpreted as follows: 1, The non-dopamine cells may be nonGABAergic neurons connected presynaptic to dopamine cells (the connectivity of non-DA IL-10R+ cells is unknown) since presynaptic GABA effects were minor. 2, IL-10R in non-dopamine cells may be recruiting different effectors with no direct presynaptic tuning effects. 3, IL-10R effects in non-dopamine cells may exert effects on a faster/slower temporal scale (and thus not detected in our assays). 4, Constitutive IL-10 activity may have resulted in a ceiling effect response in some cells, so that IL-10 bath application effects on non-dopamine cells are not apparent in the current state, but may become apparent during inflamed or infected states where IL-10 levels are decreased. The identity of the non-dopamine VTA cells is unknown, but may include local glutamate neurons, non-innervating GABA neurons, and/or glia (astrocytes, microglial, etc) to name a few (Gomez et al., 2019; Phillips et al., 2022; Shaerzadeh et al., 2020; Steffensen et al., 1998). Since our present studies focused on cell firing and GABA transmission, and not on glutamate release or excitatory postsynaptic currents, future studies may investigate a direct role for IL-10 on local excitatory inputs.

IL-10 Increases Dopamine Release

Increased VTA dopamine neuron firing translates to increased dopamine levels in the NAc. Increases in NAc dopamine levels are typically associated with increases in reward and motivated behavior (Balfour, 2009; Deitrich, Dunwiddie, Harris, & Erwin, 1989; Fields & Margolis, 2015; Salamone & Correa, 2002; Wise, 2008). However, dopamine increases have also been observed in instances of stress and associated with aversive behavior (Kutlu et al., 2021; McCutcheon, Ebner, Loriaux, & Roitman, 2012; Wenzel, Rauscher, Cheer, & Oleson, 2015). IL-10, and anti-inflammatory compounds in general, have previously been implicated as a modulator for drug conditioning (Kohno et al., 2019). For instance, morphine administered rats had upregulated NAc IL-10, and neonatal handling blocked morphine CPP reinstatement, which was associated with decreases in IL-10 gene expression in microglia (Schwarz et al., 2011). Furthermore, nicotine administered rats exhibited CPP and reduced striatal IL-10 levels, thus increased levels of IL-10 may explain the aversive behavior seen in our study (Rosa et al., 2021).Additionally, IL-10 in the basolateral amygdala decreased ethanol consumption, but not sucrose consumption or performance in an open-field test (Marshall, McKnight, Blose, Lysle, & Thiele, 2017)-inflammatory IL-1 receptor in the basolateral amygdala decreases ethanol consumption (Marshall et al., 2016) These two studies indicate the anti-inflammatory mechanisms can reduce drug seeking, potentially by producing aversion. In clinical trials, individuals were injected with IL-10 and this was well tolerated at low doses (≤ 25 μg/kg) but individuals exhibited flu-like symptoms at higher doses (100 μg/kg)(Chernoff et al., 1995; Huhn et al., 1997; Huhn et al., 1996) Thus, the aversion observed in rodents may be due to non-specific inflammation due to ICM IL-10 infusion, including sickness behavior that may counter any rewarding aspect of the increased dopamine levels. However, an important consideration is that even with highly controlled microinfusions, it is difficult to determine the spread of the infusion. Likewise, the neuroimmune network is highly complex and it is unclear what other immune effects occur with the IL-10 infusions.

IL-10 Effects on Conditioned Learning

The finding of decreased time spent in the CS+ chamber following pairing with IL-10 are indicative of an IL-10 mediated enhancement in aversive learning or possibly increased preference for the smooth CS− chamber. However, several other factors may also contribute to this outcome (Bardo & Bevins, 2000; Huston, Silva, Topic, & Muller, 2013; Tzschentke, 1998). Previous literature has indicated that IL-10 modulates both habituation and memory for previously encountered objects (Harvey, Smith, English, Mahon, & Commins, 2006; Worthen, Garzon Zighelboim, Torres Jaramillo, & Beurel, 2020), and depression- and anxiety-like behaviors (Bluthe et al., 1999; Harvey et al., 2006; Mesquita et al., 2008; Munshi, Parrilli, & Rosenkranz, 2019; Patel et al., 2021; Yang et al., 2021). The relationship between IL-10 and anxiety may be of particular relevance to the current findings. Some studies have reported reduced IL-10 expression accompanied by increased depression and anxiety-like behavior, an effect which can be reversed by repeated injection or viral overexpression of IL-10 (Patel et al., 2021; Worthen et al., 2020; Yang et al., 2021). However, others have noted that acute injections of IL-10 induced anxiety-like behavior in non-pathological animals (Harvey et al., 2006; Munshi et al., 2019). Therefore, the increased aversion for the IL-10 paired chamber may be related to a short-lived anxiogenic effect of IL-10, or possibly a longer lasting and somewhat delayed anxiolytic effect, which may have contributed to increased preference for the CS− chamber, depending on its onset and duration. However, this interpretation is highly speculative and timing effects will need to be explored further for a definitive explanation.

Conclusions

The results herein implicate IL-10 as a key mediator in modulating mesolimbic activity and subsequent dopamine release and related behavior. Future studies are needed to investigate the effects of other cytokines, particularly anti-inflammatory cytokines, in the mesolimbic dopamine system in order to more fully understand how cytokines are involved in motivated behavior. Additionally, since IL-10R was found on non-dopamine neurons there is a great need for studying the effects of IL-10 in non-dopaminergic neurons within the VTA, particularly glutamatergic inputs which were not currently investigated but may represent one of the additional IL-10R expressing non-dopaminergic cells. Further research about IL-10’s modulation of dopamine release will clarify the role of IL-10 in the neural circuitry of reward and could potentially lead to novel pharmacological and behavioral treatments.

Interleukin-10 (IL-10) increases ventral tegmental area (VTA) dopamine firing.

• IL-10 receptor expression is present in dopamine and non-dopamine cells of the VTA.

• IL-10 reduces postsynaptic GABA currents on dopamine neurons through PI3 kinase.

• IL-10 increases accumbens dopamine release to facilitate conditioned learning.