Projections from the Rostral Zona Incerta to the Thalamic Paraventricular Nucleus Mediate Nociceptive Neurotransmission in Mice

Zona incerta (ZI) is an integrative subthalamic region in nociceptive neurotransmission. Previous studies demonstrated that the rostral ZI (ZIR) is an important gamma–aminobutyric acid-ergic (GABAergic) source to the thalamic paraventricular nucleus (PVT), but whether the ZIR–PVT pathway participates in nociceptive modulation is still unclear. Therefore, our investigation utilized anatomical tracing, fiber photometry, chemogenetic, optogenetic and local pharmacological approaches to investigate the roles of the ZIRGABA+–PVT pathway in nociceptive neurotransmission in mice. We found that projections from the GABAergic neurons in ZIR to PVT were involved in nociceptive neurotransmission. Furthermore, chemogenetic and optogenetic activation of the ZIRGABA+–PVT pathway alleviates pain, whereas inhibiting the activities of the ZIRGABA+-PVT circuit induces mechanical hypersensitivity and partial heat hyperalgesia. Importantly, in vivo pharmacology combined with optogenetics revealed that the GABA-A receptor (GABAAR) is crucial for GABAergic inhibition from ZIR to PVT. Our data suggest that the ZIRGABA+–PVT pathway acts through GABAAR-expressing glutamatergic neurons in PVT mediates nociceptive neurotransmission.


Introduction
Pain is a complex sensory experience that has been well recognized as a major clinical, social, and economic problem. Although considerable progress has been made in elucidating the mechanisms of pain, studies about the basic pathway and circuit of the nervous system are still in progress [1,2].
As the hypothalamic region of the thalamus, zona incerta (ZI) contains mostly inhibitory neurons, including gamma-aminobutyric acid-ergic (GABAergic) neurons [3]. ZI is roughly divided into four components: the rostral ZI (ZIR), the dorsal ZI (ZID), the ventral ZI (ZIV) and the caudal ZI (ZIC). Distinct ZI components display diverse connectivity patterns and play multiple roles [4]. Studies have shown that the ZI is involved in a variety of functions such as defensive behaviors [5], feeding and drinking [6], sleep and circadian rhythms [7].
Morphological evidence reveals that the spinothalamic tract sends dense nociceptive inputs to ZI, suggesting a potential role of ZI in nociceptive neurotransmission [8,9]. In addition, the activities of GABAergic neurons within ZI were decreased in chronic pain rodents compared with the sham-treated group [10][11][12]. On the other hand, the ZI deep brain stimulation with 20 Hz decreased pain in humans [13]. These reports indicate that as a key nucleus, ZI might be involved in the pain neurotransmission and modulation. Previous literature has paid attention to the ZIV involvement in the regulation of pain information [14]. However, except for ZIV, different parts of ZI may respond differently to nociceptive stimuli, the roles of other parts of ZI in nociceptive neurotransmission are still needed to be elucidated.
The thalamic paraventricular nucleus (PVT), one of important thalamic midline nucleus, is mainly composed of glutamatergic neurons [15]. Previous research found that inhibition of PVT activities could alleviate visceral and neuropathic pain [16,17]. Since the PVT is lacks of the GABAergic inhibitory neurons, the modulation of PVT function relies on inhibitory afferents for excitation and inhibition balance in neurotransmission [18]. It is noted that ZIR is one of GABAergic source to PVT [19], recent studies demonstrated that the GABAergic ZIR neurons project to PVT neurons and contribute to conditioned feeding behavior [6,20]. However, the roles of the ZIR-PVT pathway in nociceptive neurotransmission have not been fully addressed.
Based on these considerations, we hypothesized that the GABAergic ZIR neurons project to PVT and participate in pain. In the present study, anatomical, chemogenetic, optogenetic and pharmacological approaches were applied to investigate the roles of the ZIR GABA+ -PVT pathway in nociceptive neurotransmission.

Animals
Experiments were performed with adult male (6-12-week-old) wild-type (WT) C57BL/ 6J (purchased from the Laboratory Animal Center of the Fourth Military Medical University), GAD2-Cre (JAX#010802), Ai9-TdTomato (JAX#007909) mice (purchased from Jackson Laboratories) and GAD67-GFP. About 200 mice were used in the study, and the number of the mice per subgroups is described in the figure legends. All animals used were housed in a 12 h light/dark cycle at 22-25 • C with food and water given ad libitum. Animal care and used strictly followed institutional guidelines and governmental regulations. All protocols to reduce the suffer of the animal from surgical operation were performed in accordance with the Animal Care and Use Committees at The Fourth Military Medical University (NO: IACUC-20210356).

Stereotaxic Injections
Mice were anesthetized with intraperitoneal (i.p.) injection of pentobarbital sodium (40 mg/kg). Mice were then fixed on a stereotaxic apparatus (RWD Life Science, Shenzhen, China), and were stereotactically injected with fluorescent tracers or viruses through a glass micropipette (internal tip diameter 15-25 µm) attached to a 1 µL Hamilton microsyringe. Following injection, the needle was left in place for another 10 min before retraction.
The injection coordinates are listed as follows based on brain atlas: ZIR: anterior posterior (AP), −1; medial lateral (ML), 0.7; and dorsal ventral (DV), −4.67 mm. PVT: AP, −0.94; ML, 0; and DV, −3.25 mm. All tracers and viruses were injected by pressure through a microsyringe with a constant flow pump at a speed of 20 nl/min. All viruses were purchased from by BrainVTA Co., Ltd (Wuhan, China).
For microinjection experiments, mice were implanted with a cannula (RWD Life Science, China) above the PVT for infusion of GABA A R antagonists bicuculline (0.1 µg). The microinjector was 0.5 mm longer than the guide sleeve and was attached to a polyethylene tube that was attached to a 10 µL syringe. The drug was infused into the PVT over 2-3 min. The microinjector was left for a further 1 min to allow the drug to diffuse. The mice were placed in the experimental apparatus 10 min after administration connect optical fibers.
For the neural tracing, the tracers used in this study include the retrograde tracers CTB (200 nl, Sigma, Louis, MO, USA) and 594 retroBeads (200 nl, Lumafluor, New York, NY, USA) that were stereotaxically injected into PVT. After surgery, mice were allowed to survive for 7-10 day.

Histology
Animals were deeply anesthetized with excess pentobarbital sodium (100 mg/kg, i.p.) and transcardially perfused with 50 mL of 0.01 M phosphate-buffered saline (PBS, pH 7.4), followed by 150 mL of 4% paraformaldehyde in 0.1 M phosphate buffer (PB, pH 7.4) to fix the tissue. Brains were removed and soaked in 4% PFA for 2-4 h. Subsequently immersed into 30% sucrose dissolved in 0.1 M PB until they sunk to the bottom of the containers. The brains were cut into 30 µm transverse sections (CM1950; Leica, Heidelberg, Germany) and stored in PBS at 4 • C until immunostaining processed. All sections were blocked in 5% donkey serum for 30 min and then incubated with the primary antibodies at room temperature for 16-24 h and the secondary antibodies at room temperature for 4-6 h.

Whole-Cell Patch-Clamp Recordings
Mice were anesthetized with pentobarbital sodium (40 mg/kg, i.p.). Coronal slices (300 µm thick) containing the ZIR were cut on a vibrating microtome (Leica VT 1200s, Heidelberger, Nussloch, Germany) at 4 • C. Slices were then transferred to a submerged recovery chamber containing oxygenated artificial cerebrospinal fluid (ACSF) consisting of (in mM) 124 NaCl, 25 NaHCO 3 , 2.5 KCl, 1 NaH 2 PO 4 , 2 CaCl 2 , 1 MgSO 4 and 10 glucose saturated with 95% O 2 and 5% CO 2 . The recordings were performed in voltage-clamp or current-clamp mode using an Axon 700B amplifier (Molecular Devices). To test the effectiveness of the ChR2 viruses, optical stimulation (473 nm) was applied with a custom laser fiber. The stimulation pattern was similar as that used in vivo (5 Hz, 10 Hz, and 20 Hz pulse trains). To test the effectiveness of the ZIR GABA+ -PVT virus projection, clozapine evoked currents were recorded in voltage-clamp mode with membrane potential held at 70 mV, and cells were stimulated using 20 µM clozapine.

Fiber Photometry
A fiber photometry system (Inper, Hangzhou, China) was used to record the fluorescence signals generated by 470 nm LED light and 410 nm LED light excitation. On the day of the experiment, mice were acclimated for 30 min in the behavioral test room, and basal fluorescence was recorded for 5 min after acclimation. Next, mice were stimulated with von Frey filament (0.07 g, 1.0 g) and shocked (0.5 s, 0.3 mA). For data analysis, fluorescence change (deltaF/F) was represented by (F − F0)/F0 and the F0 is the averaged fluorescence in baseline period. Mice without the correct viral transduction or the correct fiber-optic site were excluded from the analysis.

Chemogenetic Manipulation
All behavioral tests were performed 30 min after intraperitoneal injection of 0.1 mg/kg clozapine (Sigma) in chemogenetic manipulation. The different behavior experiments were separated by two days.

Von Frey Filament Test
Mice were habituated in in the plastic box (8 × 8 × 10 cm 3 ) for 3 successive days. Before the test, mice were placed in arena ahead of the test until they calmed down. Then, a standard protocol was used to vertically stimulate the left hind paws of mice with a series of von Frey hairs in logarithmic increments of stiffness (0.07-2.0 g) as our previous reports [21]. Each von Frey hair was placed vertically on the sole of the hind paw for 4-5 s. The minimal force that caused responsive behaviors (lifting, flicking, or licking responses) 3 times out of 5 times of tests was regarded as the paw withdraw threshold of this animal (cut-off value is set to 2.0 g).

Hargreaves Test
Mice were placed in a white plastic box on a glass floor. A radiant heat beam was focused on the hind paw. Each test was repeated 3 times with a minimum interval of 5 min. To avoid tissue damage, the cut-off latency for the test was 20 s.

Hotplate Test
Mice were exposed to hot plate at temperatures of 48, 52 and 56 • C. The movement of the mice was restrained by a 20 cm diameter plexiglass cylinder. The latency of the withdrawal response such as licking, flicking, or lifting the hind paw was recorded, and the cut-off time was set to 30 s. Each temperature is tested only once, 10 min apart.

Open Field Test (OFT)
The OFT was used to evaluate the locomotor activity and the anxiety-related behavior of mice. Mice were placed in the central of a polystyrene enclosure (50 × 50 × 40 cm) and allowed to move freely for 15 min and were videotaped separately. The following mice were then swabbed with 75% ethanol before being tested. At the end of the test, an automated analysis system was used to quantify the total distance traveled, velocity and the time spent in the central area.

Elevated plus Maze (EPM) Test
The EPM is made of a white Plexiglas apparatus consisting of two opposing open arms (30 × 5 cm), two opposing closed arms (30 × 5 cm), and a central area (5 × 5 cm). The platform was 50 cm above the floor. In general, the mice were placed alone in the central area of the maze and allowed to explore for 9 min. The time spent in the open arms was recorded by the automated analysis system.

Formalin Test
After a 30 min habituation period, mice were subcutaneously injected with 4% formalin (20 µL) in the left paw, and mice behaviors were videotaped for 60 min. The paw licking and biting times were divided into 5 min sections. Additionally, the acute (0-10 min) and inflammatory (10-45 min) phase pain responses were quantified.

Neuropathic Pain Model
We used SNI to establish a neuropathic pain model [17]. Mice received general anesthesia with pentobarbital sodium (40 mg/kg, i.p.). After disinfection, three branches of the sciatic nerve are clearly exposed through small incisions in the skin and muscle. The common peroneal nerve and the tibial nerve were ligated with 5-0 thread without damaging the sural nerve. The nerve was then severed between ligation and the distal nerve stump was removed 1-2 mm. Finally, the myofascial membrane and skin were closed.

Statistical Analysis
All data are presented as mean ± s.e.m. We used Prism 8 software (GraphPad, San Diego, CA, USA) for statistics. The data were statistically analyzed with two-tailed t-test for the raw data, one-way or two-way ANOVA with repeated measures for multiple comparisons. p < 0.05 was considered statistically significant.

The ZIR GABA+ -PVT Pathway Involvement in Nociception
First, to determine the connections between ZI and PVT, we injected the retrograde tracer 594 retroBeads into PVT of C57BL/6J mice ( Figure 1A). We found that the retrograde labeled neurons mainly occupied the ZIR and were rarely observed in the other parts of ZI ( Figure 1C,D), which is consistent with previous study [19].
Next, to clarify the types of neurons that ZIR projects to PVT, we applied the retrograde tracer cholera toxin B subunit (CTB) into PVT of glutamic acid decarboxylase-green fluorescence protein (GAD67-GFP) knock-in mice ( Figure 1E). We found that there were many GFP-labeled positive neurons in ZI of GAD67-GFP mice, consistent with the data for glutamic acid decarboxylase 2 (GAD2) mRNA distribution shown in the Allen Brain Atlas (http://www.brain-map.org) (accessed on 1 January 2023). The retrograde neurons doublelabeled with CTB/GFP were observed in ZIR ( Figure 1H1-I3). The results showed that approximately 72.2% of the CTB-labeled were co-expressed with GFP in ZIR ( Figure 1G), suggesting that most neurons in ZIR project to PVT are GABAergic.
grade tracer cholera toxin B subunit (CTB) into PVT of glutamic acid decarboxylase fluorescence protein (GAD67-GFP) knock-in mice ( Figure 1E). We found that ther many GFP-labeled positive neurons in ZI of GAD67-GFP mice, consistent with th for glutamic acid decarboxylase 2 (GAD2) mRNA distribution shown in the Allen Atlas (http://www.brain-map.org) (accessed on 1 January 2023). The retrograde ne double-labeled with CTB/GFP were observed in ZIR ( Figure 1H1-I3). The results sh that approximately 72.2% of the CTB-labeled were co-expressed with GFP in ZIR ( 1G), suggesting that most neurons in ZIR project to PVT are GABAergic.  We bred GAD2-cre mice with Ai9 reporter mice, in which the GAD2 + neurons were labeled with tdTomato. To investigate whether the ZIR-PVT pathway is involved in nociceptive transmission, we injected rAAV-hSyn-eGFP into the right ZIR of GAD2-Ai9 mice ( Figure 2A1-A4). Three weeks later, the mice were applied with 4% formalin (20 µL) in the left hind paw [14], and perfused after 2 h. Fos protein (FOS) is widely used as a biomarker of early neuronal activation in relation to the pathophysiology of pain [22,23]. Immunohistochemical staining showed that in response to nociceptive stimuli, many FOSpositive neurons were observed in PVT ( Figure 2B3). The anterogradely labeled axonal terminals (green) from the ZIR showed GAD2-positive (red), and made close connections with the FOS positive neurons (blue) in PVT ( Figure 2C1-C4). These data indicate that the activated PVT neurons receives GABAergic projections from the ZIR in nociception.
To further confirm the distinct relationships between the ZIR-PVT pathway and the pain-related responses, we applied in vivo fiber photometry recording to monitor their Ca 2+ levels in free moving mice. We stereotaxically injected anterograde transmonovirus rAAV-hSyn-Cre into ZIR combined with the application of ultrasensitive protein calcium sensors (cre-dependent jGCaMP7s) into PVT, and implanted an optical fiber above PVT for long-term recordings of jGCaMP7s fluorescence for in vivo calcium imaging via fiber photometry ( Figure 3D). We found an increase in the activities of ZIR neurons projecting to PVT following 1.0 g von Frey filament stimuli ( Figure 2F,G) and shock ( Figure 2H,I), whereas no significance in Ca 2+ signals were detected by the non-nociception stimuli (0.07 g von Frey filament, Figure 2J,K), indicating that the activities of PVT neurons receiving afferents from ZIR were specifically evoked by nociceptive stimuli.
Taken together, our results indicate that in response to nociceptive stimuli, these activated PVT neurons receives GABAergic afferents from ZIR. Importantly, an in vivo fiber photometry recording indicated that the excitability of the ZIR GABA+ -PVT pathway was increased in nociception.

In Situ Chemogenetic Activation of the ZIR GABAergic Neurons Attenuates but Inhibition of ZIR Promotes Nociceptive Behaviors
We observed that the ZIR GABA+ -PVT pathway activated in nociception. Next, we characterized the functional roles of ZIR in nociceptive behaviors with the chemogenetic methods of designer receptors exclusively activated by designer drugs (DREADDs). The rAAV encoding the cre-dependent excitatory human M3 muscarinic receptor (hM3Dq) or inhibitory (hM4Di) designer receptor fused with enhanced yellow fluorescence protein (EYFP) were injected into bilateral ZIR of GAD2-Cre mice ( Figure 3A). The control mice were injected with the rAAV-EF1a-DIO-EYFP (EYFP group). Three weeks later, at least 30 min after intraperitoneal injection of clozapine (0.1 mg/kg), behavioral data showed that compared with the EYFP group, the mechanical pain threshold was distinctly increased in the hM3Dq group (hM3Dq-Clozapine vs. EYFP-Clozapine: p = 0.0087). On the other hand, the mechanical pain threshold was decreased when chemogenetic inhibition of ZIR GABAergic neurons (hM4Di-Clozapine vs. EYFP-Clozapine: p = 0.0352, Figure 3C).
Collectively, these data suggest that chemogenetic activation of ZIR GABAergic neurons is sufficient to induce antinociceptive effect, whereas inhibition of ZIR GA-BAergic neurons results in mechanical and partial heat hypersensitivity. Effects of chemogenetic manipulation of ZIR GABAergic neurons on heat thresholds in hotplate test, n = 9 for EYFP group, n = 7 for the hM3Dq group, n = 7 for the hM4Di group. * p < 0.05, ** p < 0.01. One-way ANOVA with Dunnett's multiple comparisons test for (C-G).

Optogenetic Activation of the ZIR GABA+ -PVT Pathway Alleviates Nociception but Inhibition Induces Hyperalgesia
Our previous results suggested an essential role of in situ ZIR GABAergic neurons in nociception, next we aimed to observe the effects of the pathway from the ZIR GA-BAergic neurons to PVT on nociceptive neurotransmission by optogenetic manipula- (E-G) Effects of chemogenetic manipulation of ZIR GABAergic neurons on heat thresholds in hotplate test, n = 9 for EYFP group, n = 7 for the hM3Dq group, n = 7 for the hM4Di group. * p < 0.05, ** p < 0.01. One-way ANOVA with Dunnett's multiple comparisons test for (C-G).

Optogenetic Activation of the ZIR GABA+ -PVT Pathway Alleviates Nociception but Inhibition Induces Hyperalgesia
Our previous results suggested an essential role of in situ ZIR GABAergic neurons in nociception, next we aimed to observe the effects of the pathway from the ZIR GABAergic neurons to PVT on nociceptive neurotransmission by optogenetic manipulations. We injected rAAV-EF1a-DIO-ChR2-EYFP and rAAV-EF1a-DIO-eNpHR3.0-EYFP into ZIR of GAD2-Cre mice ( Figure 4A). The rAAV-EF1α-DIO-EYFP were injected as the control. Three weeks later, the opto fibers were implanted above PVT.
Above all, the effectiveness of the ChR2 virus by the whole-cell current-clamp recordings was examined. The brain slices were performed using 5 Hz, 10 Hz, and 20 Hz pulse trains stimulation, results showed that the blue laser pulses induced time-locked action potential firing in the neurons expressing ChR2-EYFP in ZIR ( Figure 4C). Moreover, the blue laser pulse significantly induced inhibitory postsynaptic currents (IPSC) in PVT ( Figure 4D). These results proved that the PVT neurons could be strongly inhibited by ZIR stimulation.
Behavioral data showed that optogenetic activation of the ZIR GABA+ -PVT pathway with blue laser light (473 nm, 20 Hz, 5-10 mW) induced a significantly increase in the threshold of mechanical pain (Light on vs. Light off in the ChR2 group: p = 0.0050; EYFP vs. ChR2 in the Light-on state: p = 0.0492). The mice received the virus without ChR2 expression did not show any difference when the light was on (Light on vs. Light off in the EYFP group: p > 0.9999, Figure 4E). There was no significant effect in Hargreaves test (Light on vs. Light off in the ChR2 group: p = 0.9989; EYFP vs. ChR2 in the Light-on state: p = 0.5166, Figure 4F Inversely, optogenetic inhibition of the ZIR GABA+ -PVT pathway with yellow light (589 nm, 5 mW) induced mechanical hypersensitivity (Light on vs. Light off in the eN-pHR3.0 group: p = 0.0016; EYFP vs. eNpHR3.0 in the Light-on state: p = 0.0033, Figure 4J). Inhibition of the ZIR GABA+ -PVT pathway also significantly decreased the latency withdrawal thresholds in Hargreaves test (Light on vs. Light off in the eNpHR3.0 group: p = 0.0063; EYFP vs. eNpHR3.0 in the Light-on state: p = 0.0356, Figure 4K). However, the results from hotplate test did not show any significant differences (48 Therefore, our findings demonstrate that the specific activation of the ZIR GABA+ -PVT pathway alleviates mechanical hyperalgesia. However, inhibition of the ZIR GABA+ -PVT pathway is sufficient to evoke mechanical hyperalgesia and but partial heat hyperalgesia.
In formalin-induced inflammatory pain model, compared with the control group, activation of the ZIR GABA+ -PVT pathway significantly shortened the time spent on licking paws in the inflammatory phase (second phase, 10-45 min) but not the acute phase (first phase, 0-5 min, acute: hM3Dq-Clozapine vs. mcherry-Clozapine: p = 0.8867; inflammatory: hM3Dq-Clozapine vs. mcherry-Clozapine: p < 0.0001, Figure 5F-H). The duration of paw licking decreased from 377.3 ± 41.7 s to 118.1 ± 14.7 s. The results demonstrate that the activation of the ZIR GABA+ -PVT pathway might alleviate formalin-induced inflammatory pain in mice.
Taken together, chemogenetic activation of the ZIR GABA+ -PVT pathway might exert analgesic effects in inflammatory pain and neuropathic pain in mice.

The ZIR GABA+ -PVT Pathway Modulates Nociception through GABA-A Receptor (GABA A R)-Expressing Neurons in PVT
Since the neurons from ZIR project to PVT were mainly GABAergic neurons, the underlying mechanism of the ZIR GABA+ -PVT pathway in nociceptive modulation is needed to be specified in detail. Therefore, we injected cre-dependent anterograde virus (rAAV-DIO-mcherry) into the right ZIR of GAD2-Cre mice ( Figure 6A). Calcium-calmodulin dependent protein kinase II (CaMKII) is generally considered to be a marker for excitatory neurons in thalamus [24]. Herein, we performed theCaMKII/GABA A R double-labeled immunostaining in GAD2-Cre mice receiving rAAV-DIO-mcherry injection into ZIR. Results showed that the CaMKII/GABA A R double-labeled neurons densely distributed in PVT ( Figure 6B), indicating these excitatory neurons expressing GABA A R. Furthermore, the anterogradely labeled axonal terminals (red) from the ZIR made close connections with GABA A R/CaMKII double-labeled neurons in PVT ( Figure 6B). The immunostainings indicate that the inhibitory effects of GABAergic terminals from ZIR GABA+ to glutamatergic PVT neurons might be mediated by GABA A R located at PVT. We next detected the effects of GABA A R on PVT in nociceptive neurotransmission by locally microinjected GABA A R antagonists bicuculline (0.1 µg) in PVT with saline infused as control. The cannula implantation regions were confirmed at the end of all behavioral tests ( Figure 6C). We found that pharmacological inhibition of GABA A R in PVT decreased the mechanical pain threshold and thermal hyperplasia (von Frey test: Bicuculline vs. Saline: p < 0.0001; Hargreaves test: Bicuculline vs. Saline: p = 0.0104, Figure 6D,E), revealing that the GABA inhibition of PVT via GABA A R potentially contributes to pain hypersensitivity.
To further confirm the roles of GABA A R in the ZIR GABA+ -PVT pathway, we injected rAAV-EF1a-DIO-ChR2-EYFP into ZIR ( Figure 6F). After microinjection of bicuculline for 20 min, blue laser stimulation was performed. Results demonstrated that bicuculline decreased the mechanical pain threshold, which could be mostly rescued by activating the ZIR GABA+ -PVT pathway (Bicuculline vs. Saline: p < 0.0001; Bicuculline + ChR2 vs. Bicuculline: p = 0.0021; Bicuculline + ChR2 vs. Saline: p = 0.0042, Figure 6G). This effect was also found in the Hargreaves test (Bicuculline vs. Saline: p = 0.0423; Bicuculline + ChR2 vs. Bicuculline: p = 0.0057, Figure 6H). The significant increase in pain threshold from 0.40 ± 0.02 g (Bicuculline group) to 0.69 ± 0.06 g (Bicuculline + ChR2 group) in the von Frey test and 9.48 ± 0.96 s (Bicuculline group) to 13.83 ± 0.97 s (Bicuculline + ChR2 group) in the Hargreaves test. These findings demonstrate that GABA A R is crucial for GABAergic inhibition from ZIR to PVT pathway in nociceptive neurotransmission. Therefore, our results suggest that the ZIR GABA+ -PVT pathway plays a critical role in regulating nociceptive neurotransmission through GABA A R-expressing glutamatergic neurons in PVT. culline: p = 0.0021; Bicuculline + ChR2 vs. Saline: p = 0.0042, Figure 6G). This effect was also found in the Hargreaves test (Bicuculline vs. Saline: p = 0.0423; Bicuculline + ChR2 vs. Bicuculline: p = 0.0057, Figure 6H). The significant increase in pain threshold from 0.40 ± 0.02 g (Bicuculline group) to 0.69 ± 0.06 g (Bicuculline + ChR2 group) in the von Frey test and 9.48 ± 0.96 s (Bicuculline group) to 13.83 ± 0.97 s (Bicuculline + ChR2 group) in the Hargreaves test. These findings demonstrate that GABAAR is crucial for GABAergic inhibition from ZIR to PVT pathway in nociceptive neurotransmission.
Therefore, our results suggest that the ZIR GABA+ -PVT pathway plays a critical role in regulating nociceptive neurotransmission through GABAAR-expressing glutamatergic neurons in PVT.

Discussion
In this study, we present that the ZIR GABA+ -PVT pathway is involved in nociceptive neurotransmission. Our study found that these PVT neurons which received ZIR projections were activated by nociceptive stimulation. Furthermore, we confirmed that activation of ZIR GABAergic neurons contribute to antinociception, whereas inhibition of ZIR induce mechanical hypersensitivity and partial heat hyperalgesia. Moreover, we also found that the activated the ZIR GABA+ -PVT pathway alleviated nociception, whereas when inhibited, this pathway induced hyperalgesia, and these effects were acted through GABA A R-expressing neurons in PVT.
GABA is an important inhibitory neurotransmitter in the central nervous system (CNS) and is essential for the balance between neuronal excitation and inhibition [25,26]. Excitation and inhibition imbalance in the activity of the thalamus is thought to be one of the important causes of pain [27]. Importantly, ZI has unidirectional feedforward inhibition control on other nuclei which are enriched excitatory neurons in thalamus, leading to the rapid and strong effects on neurotransmission of thalamus [28,29]. For instance, the pathological increased activity of mediodorsal thalamus (MD) and the posterior complex of the thalamus (PO) activity after spinal cord injury are associated with ZI inhibition [30,31]. This evidence suggests that ZI plays a key role in the inhibitory modulation of the excitation and inhibition balance in thalamus. However, there are still few studies on the involvement of ZI inhibition in thalamus in nociceptive neurotransmission.
As an important component of the midline thalamic nuclei, the PVT consists primarily of excitatory glutamatergic neurons that are thought to relay pain information across various cortical and subcortical regions [32]. Previous studies showed an increased c-Fos protein expression and phosphorylation of extracellular signal-regulating kinases (pERK) within PVT in hyperalgesic mice [33]. The hyperexcitation of electrophysiological characteristics of PVT neurons was observed in spinal nerve ligation (SNL) mice [34], indicating that the excitation and inhibition balance of PVT neurons plays an important role in pain, and inhibition of abnormal excitation of PVT might be essential for the occurrence and development of pain. In the present study, our results suggest that the ZIR GABA+ -PVT pathway participates in nociceptive neurotransmission and the enhanced inhibitory inputs from ZIR to PVT reduced pain.
GABAergic signaling pathway is crucial for the neural circuit in nociceptive neurotransmission by acting on GABA A R [35,36]. For instance, microinjection of the GABA A R agonist muscimol into the ventrobasal thalamus (VB) relieves thermal hyperalgesia in chronic inflammatory pain [37], and antagonists of the GABA A R bicuculline applied to the dorsal medial prefrontal cortex (dmPFC) could produce analgesic effects [38]. In addition, a recent study revealed that the injection of muscimol into PVT effectively relieve mechanical pain in SNL mice [34]. We provide morphological and behavioral evidence that the distribution of GABA A R at the ZIR GABA+ -PVT pathway is crucial in nociceptive neurotransmission. These findings show that activation of GABA A R in PVT might be a useful therapeutic strategy for antinociception.
The connections between different sectors or neuron types contribute to multiple roles of ZI in physiological functions [39][40][41]. For example, The ZIV GABAergic neurons have the opposite effects to ZIV parvalbumin (PV)-positive neurons on pain modulation [14]. The activated neural circuit from the central amygdala neurons expressing protein kinase Cdelta (PKCδ) to ZIV-GABAergic cells produces pain hypersensitivity in mice [42]. The A13 dopamine neurons in the middle of ZI are rapidly activated by acute nociceptive stimuli [43]. Our behavioral data provide evidence that ZIR is another important region that inhibition of GABAergic ZIR neurons induced behavioral hyperalgesia whereas activation of GABAergic ZIR neurons induced antinociception. However, in our study, we also admit that the virus injections might have additional impacts on other ZI regions except for ZIR, but the center of virus injection was in ZIR. Intriguingly, the somatostatin (SOM), calretinin (CR), and vesicular glutamate transporter-2 (VgluT2) expressing cells in ZIR encoded and modulated different components of anxiety [44]. It suggests that the different subpopulations of ZIR project to PVT may play more complex roles in pain, such as pain-related affective aspects, which are worth for our further exploration.
The effects of the ZIR GABA+ -PVT pathway on different forms of pain stimulation increase the complexity of the neural circuit in processing pain sensation. As shown in the present study, the modulations by chemogenetic and optogenetic methods in ZIR GABA+ or the ZIR GABA+ -PVT pathway mainly influence on mechanical hyperalgesia and thermal nociception induced by Hargreaves test. However, in hotplate test, another behavioral test for evaluating thermal nociception, results showed that there was no significance between groups by manipulations of ZIR GABA+ or ZIR GABA+ -PVT. This may be associated with fact that the mice were free to move when the hotplate test was conducted, thus influenced by other components, such as defensive and freezing behaviors. Previous research has shown that ZIR is engaged in modulating defense behaviors and freezing behaviors [45]. These findings indicate ZIR is an integrative node that modulates multiple behaviors. Taken together, our results reveal the function of the ZIR GABA+ -PVT pathway in modulating nociceptive neurotransmission.
In summary, we employed morphological, fiber optic recording, behavioral and other methods to clarify the roles of ZIR GABA+ -PVT pathway in the regulation of nociceptive neurotransmission.

Conclusions
It is universally acknowledged that pain remains a severe global health problem. The nature of neural circuits underlying the diverse components of the complex, multidimensional experience of pain is not well understood. Our study aims to investigate the circuit connectivity from the inhibitory ZIR GABA+ to excitatory PVT in pain using multiple techniques and methods, suggesting that the endogenous GABAergic inhibition is essential for excitation and inhibition balance in neurotransmission, which might be involved in nociception. Nowadays, the knowledge of neural connectivity and their dynamic regulation via functional structural plasticity has been confirmed that provide the beneficial effects for forward translation back to human therapies, for instance, neurostimulation and neuromodulation [1]. Based on the advanced knowledge for neural circuits, deep brain stimulation implantation surgery is an established treatment modality for certain CNS disorders, such as chronic pain [46]. Therefore, our findings can potentially provide a strategy for using of clinical approaches targeting specific neural circuits for pain management in future.