A Neural Circuit From Paraventricular Nucleus of the Thalamus to the Nucleus Accumbens Mediates Inflammatory Pain in Mice

Male C57BL/6J mice (procured from Hangzhou Hangsi Biotechnology Co. Ltd), aged 8–12 weeks, were used in all experiments of this study. All mice were housed in cages equipped with a dedicated ventilation system under standard laboratory conditions (12 h light/12 h dark cycle, lights on 08:00 a.m. to 08:00 p.m., temperature of 23°± 2°C, and humidity of 50% ± 10%) with ad libitum access to standard lab mouse pellet food and water. All animal care, use, and procedures in this study were approved by the Ethical Review Committee of Laboratory Animal Welfare of Huzhou Central Hospital and conformed to the ethical guidelines for animal experimentation, and all experimental protocols were conducted according to the National Institute of Health Guide for Care and Use of Laboratory Animals (IACUC Protocol: 150005A2). The experimental protocols and the functional tests employed in this study were meticulously designed to minimize the utilization of animals and mitigate any potential discomfort they may experience. We eliminated the mice that did not get the virus correctly and ended up with a total of 53 mice in our study.

A volume of 25 µL of CFA was subcutaneously injected into the right hind paw with a 20‐gauge micro‐injector to establish a mouse model of inflammatory pain (Nagakura et al. 2003).

We assessed the pain severity in mice by measuring their thermal and mechanical pain thresholds. Investigators responsible for the behavioral test were blinded to which animals represented treatments or controls.

The Hargreaves method was employed to determine the thermal pain threshold (Hargreaves et al. 1988). The mice were individually habituated in transparent acrylic enclosures on an elevated glass table in a temperature‐controlled and noise‐free room, allowing for 1 h of habituation before the test. A mobile heat‐producing radiant heat source was focused on their right hind paw to stimulate the plantar surface. The paw withdrawal latencies (PWLs) were defined as the time from the light‐on to a paw withdrawal or paw licking being recorded. The basic PWLs of mice predominantly remained 15–20 s by adjusting the radiant light intensity, and this intensity remained constant throughout the duration of the experiment. The stimulus instrument was programmed with an automatic cut‐off time of 20 s to prevent thermal radiation‐induced tissue damage. The PWLs were measured for 5 repeats/time points/animal, removing the maximum and minimum values and reserving the last three for analysis.

The mechanical pain threshold of the mice was evaluated using the von Frey test as previously described (Chaplan et al. 1994). The mice were acclimated in transparent acrylic enclosures on a wire mesh platform in a temperature‐controlled, quiet room for 1 h before the commencement of the experiment. Similarly, during a period of relative quiescence and reduced engagement in exploratory and other activities, each von Frey filament (ranging from 0.008, 0.02, 0.04, 0.07, 0.16, 0.4, 0.6, 1.0, and 1.4 g with logarithmically incremental stiffness) was applied perpendicularly to the plantar surface of the right hind paw with sufficient force to bend the filament. The von Frey filaments were kept for a duration of 4–6 s or until eliciting a paw withdrawal response to assess the paw withdrawal thresholds (PWTs). Lifting, shaking, or licking the paw indicated a positive response and prompted the next weaker filament, and the absence of a paw withdrawal response prompted the next stronger filament. Three positive behaviors were recorded as the mice's mechanical pain threshold. The von Frey filament, which made mice three positive behaviors, was recorded as the mice's PWTs.

Brain stereotaxic surgeries and injections were conducted following the previously established protocols (D.‐B. Li et al. 2019; X. Li et al. 2021; H. Liu et al. 2021). The mice were immobilized in a stereotaxic apparatus (RWD Life Technology Co. Ltd., Shenzhen, China) under general anesthesia with 1% pentobarbital sodium (40 mg/kg, dissolved in saline, ip). All the viruses were purchased from Brain VTA Technology Co. Ltd. (Wuhan, China). The corresponding viruses were injected into the PVT (unilateral coordinates, Bregma: AP: −1.4 mm; LM: +0.1 mm; DV: −3.0 mm, 4°) and the NAc (bilateral coordinates, Bregma: AP: +1.4 mm; LM: ±0.8 mm; DV: −4.6 mm, 0°) using glass pipettes with a tip diameter of approximately 20 µm controlled by a programmable nanoliter injector. The stereotaxic coordinates mentioned in this study are based on the stereotaxic atlas of The Mouse Brain in Stereotaxic Coordinates (the second edition, George Paxinos and Keith B.J. Frank edited). The speed of virus injection was 100 nL/min and the pipette tip was left in place at the injection site for 10 min post‐injection before being slowly withdrawn. To illuminate the PVT^Glut→NAc pathway, the optical fiber implantations were performed using a stereotaxic apparatus. The optical fibers (ceramic ferrule: diameter 2.50 mm; optical fiber: 200 µm core diameter, 0.39 NA, FT200EMT, Thorlabs) were precisely implanted in the PVT (unilateral coordinates, Bregma: AP: −1.4 mm, LM: −0.1 mm, DV: −3.0 mm, 4°). The optical fibers were securely affixed to the skull using dental cement. All mice were given a recovery period of 3–4 weeks after the surgery, during which all viruses were adequately expressed.

The technique of optogenetics enables the targeted investigation of neuronal circuitry and precise manipulation of behavioral responses (Xie, Wang, and Bonin 2018). In the optogenetic experiments, retrograde recombination adeno‐associated cyclization recombination enzyme (Cre) virus‐encoded the broad‐spectrum promoter CaMKIIα (rAAV‐CaMKIIα‐Cre) into the NAc. This allowed rAAV encoding Cre recombinase‐inducible channelrhodopsin (rAAV‐EF1α‐DIO‐ChR2‐EYFP) or rAAV encoding Cre recombinase‐inducible enhanced halorhodopsin 3.0 (rAAV‐EF1α‐DIO‐NpHR3.0‐EYFP) or rAAV encoding Cre‐dependent control EYFP virus (rAAV‐EF1α‐DIO‐EYFP) injected into the PVT to express, targeting PVT^Glut→NAc projecting neurons and their axon terminals. When the viruses were adequately expressed, a fiber core implanted in the mouse brain was connected to a combined laser generator and the stimulator (Newdoon, Hangzhou, China), used to generate a 473 nm wavelength of the blue laser or 589 nm of the yellow laser. In the behavioral experiments, the parameters for blue light stimulation were as follows: wavelength of 473 nm, frequency of 20 Hz, wave width of 10 ms, stimulation energy of 10 mW, and a single stimulation lasting for 1 h for activating PVT neurons in normal mice; the PVT neurons in CFA inflammatory pain mice were inhibited using yellow light and the stimulation parameters were as follows: wavelength of 589 nm, frequency of 0.1 Hz, 8‐s‐on/2‐s‐off, stimulation energy of 10 mW, and a single stimulation lasting for 2 h.

The mice were deeply anesthetized using pentobarbital sodium (40 mg/kg) and then perfused transcardially with phosphate‐buffered saline (PBS) until a yellowing of the liver was observed and sequentially perfused with 4% paraformaldehyde. The brains were removed, collected, and postfixed in 4% paraformaldehyde for 6–8 h, followed by dehydrated in 20% sucrose at 4°C overnight. Subsequently, they were transferred to a solution of 30% sucrose at 4°C overnight again. The brains of mice were preserved by freezing them in optimal cutting temperature compound Tissue‐Tek OCT. Subsequently, the frozen brains were sliced coronally into sections at a 30 µm thickness by using a cryotome (VT1000S, Leica Microsystems). These sections were then collected in PBS for staining and observation purposes.

The use of c‐Fos as a neuronal activation marker is widespread, and its strengths lie in low expression levels under basal conditions and rapid induction within minutes after acute challenges or stimuli, reaching maximal levels (Kovács 1998). The immunofluorescence staining was performed as follows. First, the floating brain sections were washed thrice with PBS and twice with 0.4% Tris‐buffered saline (TBS) followed by blocking for 1 h in 1% donkey serum diluted in 0.4% TBS at room temperature. Subsequently, the sections were incubated overnight at 4°C with primary rabbit anti‐c‐Fos polyclonal antibody (c‐Fos (9F6) Rabbit mAb, 1:1000, Cell Signaling Technology, USA). The following day, the sections were washed three times in TBS after being rewarmed for 2 h at room temperature. Subsequently, they were incubated with secondary biotinylated goat anti‐rabbit antibody (IgGH&L Alexa Fluor594, 1:1000; Abcam, UK) for 2 h at room temperature and then overnight at 4°C. After being washed three times with PBS, the brain section slices were mounted on the slide glass. The slides were air‐dried and sealed using anti‐fluorescence quenching DAPI sealing tablets before being stored in a dark environment at −20°C. The visualization and quantification of c‐Fos positive neurons were conducted utilizing fluorescence microscopy.

The estimated sample sizes were based on our past experience performing similar experiments. GraphPad Prism version 8.0 software was utilized for the statistical analysis of experimental data and the generation of statistical charts. The results were presented as means ± standard errors of the means (SEM). For the assessment of acute inflammatory pain mice and normal mice, a two‐way ANOVA was employed to compare the PWLs and PWTs among different groups. Unpaired two‐tailed t‐test was used for comparing two groups with normally distributed data. Statistical significance was considered at p values below 5% probability.

We first observed the alterations in nociceptive behaviors of mice with inflammatory pain induced by intra‐plantar injection of CFA. The experimental group was administered CFA to induce inflammatory pain, and the control group was injected with saline, followed by separate assessments of thermal and mechanical pain sensitivity in normal mice and those with inflammatory CFA injection at 4 h and 3 days post‐injection. Statistical analysis revealed that compared to the control group, the PWLs and PWTs of the right hind paw in mice injected with CFA were significantly reduced (n = 6, 6) (Figure 1A,B).

To investigate the activation of neurons in the PVT under conditions of inflammatory pain, CFA was administered to the experimental group via injection into the right hind paw of the mice, while saline was injected into the right hind paw of the mice in the control group, and the brain tissue samples were collected from the mice after they were sacrificed after 4 h and 3 days post‐injection, separately. The expression of c‐Fos in PVT was detected by immunoluminescence. The results revealed there was an increase in c‐Fos protein expression within this area among mice experiencing persistent inflammatory pain induced by CFA (Figure 1C,D), which means that some of the neurons in this area of the brain region have been activated.

To prove the anatomic projection relationship between the PVT and the NAc, we specifically labeled the PVT glutamatergic neurons projected to the NAc by injecting the retrograde tracer fluorescently labeled virus rAAV‐CaMKIIa‐EYFP (Figure 2A). After 28 days of virus injection, brain tissues were collected from the sacrificed mice. Images showed that the reverse transport fluorescently labeled virus injected in the NAc was expressed in PVT glutamatergic cell bodies. The virus was detected in the cell bodies of glutamatergic neurons in the PVT and their axon terminals in the NAc, providing further structural evidence for a substantial population of glutamatergic neurons in the PVT projecting to the NAc (AP: +1.4 mm; LM: ±0.8 mm; DV: −4.6 mm, 0°) (Figure 2B). Subsequently, we specifically labeled the PVT glutamatergic NAc by injecting the retrograde tracer fluorescently labeled virus rAAV‐CaMKIIa‐EYFP using the same methodology (Figure 2C). After 25 days post‐virus injection, CFA was administered into the right hind paw of the mice. Three additional days later, brain tissues were collected from the sacrificed mice for subsequent analysis of c‐Fos expression in the PVT using immunoluminescence. The c‐Fos immunoreactive cells were visualized in red, while the glutamatergic neurons projecting to the NAc within the PVT were labeled in green. Finally, we discovered that approximately 91.19% of the neurons were co‐labeled, indicating a potential association between the PVT^Glut→NAc circuit and inflammatory pain (Figure 2D).

To further investigate whether optogenetic activation of the PVT^Glut→NAc projection was sufficient to induce nociceptive behaviors in naive mice, we administered retrograde virus rAAV‐CaMKIIα‐Cre into the NAc, which allowed rAAV‐EF1α‐DIO‐ChR2‐EYFP (n = 8) for the channelrhodopsin‐2 (ChR2) group or rAAV‐EF1α‐DIO‐EYFP (n = 6) for the EYFP control group injected into the PVT to express, targeting PVT^Glut→NAc projecting neurons and their axon terminals (Figure 3A,B). Optical fibers were implanted above the PVT for subsequent optogenetic stimulation. At 28 days post‐injection of the virus, the optogenetic stimulation was performed by employing an optic fiber connected to a laser light source that emitted a wavelength of 473 nm, facilitating ChR2 photo‐activation (Figure 3C). The experimental results show that through a single 1‐hour‐long stimulation, the activation of the PVT^Glut→NAc circuit resulted in decreased PWLs and PWTs, with effects returning to the baseline level within 2 h after cessation of laser irradiation (Figure 3D,E). Then the PVT^Glut→NAc circuit in mice was stimulated with continuous periodic blue laser light for 1 h per day over a period of 5 days. We did the behavioral assessment on first day, third day, and fifth day when the light stimulation time was full for an hour and similar results were observed in the somatic optogenetic experiment. Finally we did the behavioral assessment on the sixth day without light stimulation; painful behaviors were not found in the absence of light stimulation (Figure 3F,G). These findings suggest that activation of the PVT^Glut→NAc circuit can induce nociceptive behaviors in naive mice.

Next, we attempted to ascertain the analgesic effects of inhibiting the PVT^Glut→NAc circuit in the mice with inflammatory pain. Similarly, we administered rAAV‐CaMKIIα‐Cre into the NAc and rAAV‐EF1α‐DIO‐NpHR3.0‐EYFP (n = 8) or rAAV‐EF1α‐DIO‐EYFP (n = 6) into the PVT to specifically express in PVT^Glut→NAc projecting neurons and their axon terminals (Figure 4A,B). The optical fibers were implanted above the PVT for the subsequent optogenetic stimulation. After 25 days of virus injection, CFA was injected into the right hind paw of the mice. After an additional 3 days, optogenetic stimulation was performed using an optic fiber connected to a laser light source emitting 589 nm for NpHR photo‐activation (Figure 4C). Surprisingly, despite a single 2‐h stimulation, compared with the EYFP control group, no changes were observed in PWLs and PWTs upon optogenetic inhibition of the PVT^Glut→NAc neural circuit (Figure 4D,E). The findings suggest that single light stimulation does not exert inhibitory effects on pain behavior in mice. Considering the activated state of neurons, it is plausible that brief light stimulation fails to reduce their excitability; therefore, we conducted periodic light stimulation. To examine the regulatory function of the PVT^Glut→NAc circuit in mice on inflammatory pain, we applied continuous periodic yellow laser stimulation for 2 h daily over a span of 5 days. We conducted the behavioral assessment on the first, third, and fifth days and observed that optogenetic inhibition of the PVT^Glut→NAc circuit significantly enhanced PWLs on Days 3 and 5, as well as PWTs on Day 5. Subsequently, we performed a behavioral assessment on the sixth day without light stimulation, revealing painful behaviors in the absence of light stimulation (Figure 4F,G). These findings suggest that inhibiting the PVT^Glut→NAc circuit can alleviate pain in mice with inflammatory pain.

The peer review history for this article is available at https://publons.com/publon/10.1002/brb3.70218↗.