Abstract
Tricyclic antidepressants (TCAs), such as desipramine (DMI), are effective at managing neuropathic pain symptoms but often take several weeks to become effective and also lead to considerable side effects. Tianeptine (TIAN) is an atypical antidepressant that activates the mu-opioid receptor but does not produce analgesic tolerance or withdrawal in mice, nor euphoria in humans, at clinically-relevant doses. Here, we evaluate the efficacy of TIAN at persistently alleviating mechanical allodynia in the spared nerve injury (SNI) model of neuropathic pain, even well after drug clearance. After finding an accelerated onset of antiallodynic action compared to DMI, we used genetically modified mice to gain insight into RGS protein-associated pathways that modulate the efficacy of TIAN relative to DMI in models of neuropathic pain. Because we observed similar behavioral responses to both TIAN and DMI treatment in RGS4, RGSz1, and RGS9 knockout mice, we performed RNA sequencing on the NAc of TIAN- and DMI-treated mice after prolonged SNI to further clarify potential mechanisms underlying TIANs faster therapeutic actions. Our bioinformatic analysis revealed distinct transcriptomic signatures between the two drugs, with TIAN more directly reversing SNI-induced differentially expressed genes, and further predicted several upstream regulators that may be implicated in onset of action. This new understanding of the molecular pathways underlying TIAN action may enable the development of novel and more efficacious pharmacological approaches for the management of neuropathic pain.
Similar content being viewed by others
Introduction
Neuropathic pain is a debilitating condition that may affect up to 10% of the global population [1]. The induction and maintenance of neuropathic pain involve maladaptations in both peripheral and central nervous system sites, including thalamic, cortical and brainstem circuits as well as components of the brain reward pathway [2, 3]. The management of neuropathic pain conditions has been particularly challenging as currently-prescribed medications, such as anticonvulsants and antidepressants, can promote severe adverse effects, are not well tolerated, have a slow onset of action, and are not effective in all patients [4, 5]. On the other hand, the use of opioids is problematic as mu-opioid receptor (MOR) agonists promote analgesic tolerance, and their repeated administration risks physical dependence and transition to addiction [6]. Understanding the intracellular mechanisms triggered by both tricyclic antidepressants (TCAs) and opioid-targeting agents will provide critical information on signaling cascades that can be targeted for the development of more efficacious and safer treatments for neuropathic pain conditions.
The atypical antidepressant tianeptine (TIAN) has been shown effective for the treatment of mood disorders such as Major Depressive Disorder. Recent preclinical work has demonstrated that TIAN is a MOR agonist and that its antidepressant- and anxiolytic-like effects in mice are dependent on MORs, unlike standard monoamine-targeting antidepressants [7, 8]. TIAN has brief analgesic properties in murine models of acute pain, with acute analgesia peaking in 15–30 min, and is rapidly metabolized out of the system with blood and brain concentrations of TIAN and its metabolite returning to baseline levels within hours. Interestingly, TIAN does not produce tolerance or withdrawal in mice when used in a similar dosing strategy to morphine [9,10,11]. TIAN also induces CaMKII/PKA-mediated post-translational modification of AMPA receptors [12, 13], similar to other antidepressants. TIAN also shares structural similarities to TCAs, yet lacks affinity for the same monoaminergic receptors [14]. Despite these similarities, TIAN has an improved side effect profile when compared to TCAs. Due to the known antidepressant and acute analgesic properties of this compound, and the partial overlap in pain and affective circuits, we hypothesized that TIAN might also have antiallodynic effects that might lead to alternative options for effective neuropathic pain treatment.
Here, we use murine models to demonstrate that TIAN alleviates long-term mechanical hypersensitivity in the tibial spared nerve injury (SNI) model of neuropathic pain [15]. TIAN promotes these antiallodynic effects in the murine SNI model more rapidly than desipramine (DMI), a TCA with known antiallodynic properties [16]. Importantly, there is no tolerance to the antiallodynic effects of either antidepressant, unlike morphine.
Preclinical work by our group and others has previously identified intracellular modulatory pathways triggered by analgesics that modulate GPCR signaling [17,18,19,20]. Such signal transduction complexes contain Regulators of G protein Signaling (RGS proteins), which are dynamic modulators of opioid and antidepressant drug efficacy with distinct expression patterns in mood, nociception, and analgesia circuits [21,22,23,24,25]. RGS proteins control GPCR signaling amplitude and direction by multiple actions, including direct binding to activated Gα subunits and control of Gβγ availability to effectors [18, 26]. Using models of stress or neuropathic pain, we demonstrated that RGS4, a molecule expressed in several mood and nociception controlling circuits, including the striatum, is necessary for functional responses to monoamine-targeting antidepressants [27], while RGS9-2, a molecule primarily expressed in the striatum, delays the onset of antidepressant response and acts as a negative modulator of TCA and SNRI efficacy [23]. More recently, our group demonstrated that RGSz1, a small brain-localized RGS protein, acts as a potent negative modulator of opioid analgesia. However, the role of RGSz1 in functional responses to antidepressants has not been investigated [21].
Using constitutive knockout mouse lines lacking the genes encoding RGS4, RGS9 or RGSz1, we demonstrate that TIAN triggers antiallodynic, antinociceptive, and stimulatory behavioral signatures that are similar to those observed with monoamine-targeting antidepressants, such as DMI, and synthetic opioids, such as oxycodone. To determine potential differences in transcriptomic effects of TIAN and DMI, we performed RNA sequencing (RNA-seq) on the nucleus accumbens (NAc). We selected this brain region as our previous work with classical antidepressants has shown dynamic changes in G protein signaling and transcriptional adaptations in this region in models of neuropathic pain. The NAc plays a critical role in the modulation of chronic pain symptoms and treatment efficacy in humans as well as in preclinical models, and several members of the RGS family are expressed in this brain region [3, 28,29,30]. This analysis revealed distinct gene expression change profiles in the NAc, with TIAN more directly counteracting SNI-specific transcriptomic changes. Several predicted upstream regulators were identified that may explain the faster, persistent antiallodynic efficacy of TIAN.
Materials and methods
Animals and drug administration
Male mice (2–3 months old) were used for all behavioral assays, except for the experiments shown in Supplementary Fig. 1A, B that used both male and female mice. Three inbred, mixed background constitutive knockout (KO) lines and their wildtype (WT) counterparts were used: RGS4KO [22], RGSz1KO [21], and RGS9KO [31, 32]. Mice were housed on a 12-h light-dark cycle with food and water accessible ad libitum. All animal breeding, housing, and experimentation was performed in accordance with Mount Sinai Institutional Animal Care and Use Committee.
A 15 or 30 mg/kg solution of TIAN NaCl (Qingdao Sigma Chemical Co.; purity verified by NMR analysis) or 15 mg/kg of DMI NaCl (Sigma-Aldrich) was administered intraperitoneally (i.p.). For longitudinal experiments, TIAN and DMI were administered bidaily (BID) every 12 h.
Spared nerve injury
Tibial spared nerve injury (SNI) was performed on the left sciatic nerve of mice as previously described [29]. Briefly, a skin incision was made on the lateral surface of the hindlimb and underlying muscle was pulled apart with forceps to expose the sciatic nerve at the trifurcation of the tibial, peroneal, and sural branches. The peroneal and sural nerves were then tightly ligated with a silk 6-0 suture (Patterson Veterinary, CO) and distally severed. One to two millimeters were taken from the distal portion of the nerve to inhibit regeneration. A 6-0 silk suture was then used to close the muscle, and a 4-0 nylon suture was used to close the skin.
von Frey assay
Mechanical allodynia in SNI and sham mice was assessed with the von Frey (VF) assay (Stoelting Co, IL). For the comparison of TIAN to DMI in C57BL/6J animals, drug was administered four weeks after nerve injury to assess effects in prolonged nerve injury states. TIAN injections for the VF assay were initiated three days after nerve injury for RGS4KO and RGSz1KO mice, and nine days for RGS9KO due to a previously noted delayed onset of maximal mechanical allodynia in this mouse line [24]. After the three-day incubation period, mice were injected with TIAN BID for three days prior to VF assessment. VF was then performed 16 hours post-drug injection on staggered days. The ascending stimulus method was used to determine withdrawal thresholds [33]. An allodynic response was recorded as a hind paw withdrawal or licking, and the withdrawal threshold was recorded as the second filament force that elicited a positive response in three of five stimulations.
Locomotor activity assay
Mice were allowed to habituate to the locomotor activity chambers (Med Associates Inc, VT) for 60 minutes. Mice were then injected with saline (i.p.) and placed back in the locomotor chambers for another 50–60 minutes to assess locomotor response to an i.p. injection. Immediately after this session, mice were injected with TIAN (15 mg/kg i.p. for RGSz1WT/KO and RGS9WT/KO and 30 mg/kg for RGS4WT/KO and RGSz1WT/KO) and placed back in the chambers for 60 minutes. A lower dose was used for RGSz1 and RGS9 mice to detect an enhanced response to nearly sub-threshold doses based on prior studies with these mouse lines that demonstrated increased sensitivity to antidepressants or opioids [21, 23]. Data were recorded with Med Associates auto-quantifying software, and total combined X-axis and Y-axis beam breaks were calculated for a time course statistical analysis.
Hot plate assay
Thermal nociceptive responses were assessed in uninjured mice using the hot plate assay (IITC Life Sciences, CA) [21]. Briefly, the hot plate apparatus was pre-set to 52.1 °C, and mice were habituated in the room for 1 hour. Mice were injected i.p. with TIAN (15 or 30 mg/kg), and hind paw withdrawal times were recorded 15, 30, and/or 60 minutes after injection. Withdrawal time was measured as the time at which the first hind paw shake/lick or full jump was observed. We used a cut-off time of 40 seconds. Data were expressed as percent of the maximal possible effect (% MPE), which was calculated with the following equation: MPE = (TimeDrug − TimeBaseline)/(TimeCutoff − TimeBaseline).
RNA-seq and bioinformatic analysis
Four weeks after nerve injury surgery, male mice received TIAN, DMI, or saline for three weeks and were then sacrificed for tissue collection. NAc tissue was punched and flash frozen on dry ice. RNA was extracted from tissues using TRIzol-mediated phase separation (Invitrogen, MA). RNA samples were processed through whole-transcriptome library preparation using the Stranded mRNA Prep kit, (Illumina, CA) following the manufacturer’s instructions. Briefly, total RNA inputs were normalized to 100 ng/10 ul going into preparation. Total RNA was first enriched for mRNA prior to cDNA generation. 3’ ends of cDNA were then adenylated before ligation with adapters utilizing unique dual indices (96 UDIs) to barcode samples to allow for efficient pooling and high throughput sequencing. Libraries were enriched with PCR, with all samples undergoing 13 cycles of amplification prior to purification and pooling for sequencing. Completed libraries were quantified using Quant-iT reagents, and equimolar pools were generated and sequenced on the NextSeq 500/550 Hi Output Kit v2.5 (150 CYS) flow cell with 2x50bp paired-end reads, yielding a mean of 25 million paired-end reads per sample.
FASTQ files were aligned to the mouse genome (mm10) using the NGS-Data-Charmer pipeline (https://github.com/shenlab-sinai/NGS-Data-Charmer). In brief, the pipeline first trims the reads of adapters and poor-quality bases. The paired-end trimmed reads are then aligned to the genome using HISAT2 (version 2.2.1). Duplicate reads are then removed from the alignment files. To obtain a gene expression matrix, the unique reads are processed using FeatureCounts (flags: -p -O -t gene fraction) and the murine transcriptome (GENCODE vM25). In a QC step, samples were examined for the presence of any poor-quality samples. DESeq2 (version 1.32.0) was implemented to identify any differential expression. Differential transcripts were classified as having a p-nominal < 0.05, with a log2FC > |0.32| cutoff for comparison against SNI-Saline-specific DEGs. Visualizations of the DESeq2 results (RRHO2 plots, volcano plots, Venn/petal diagrams, and heatmaps) were generated in R (version 4.1.1). Pathway analysis was performed using Ingenuity Pathway Analysis (IPA, Qiagen, Germany). A p-nominal < 0.05 cutoff was used for pathway analysis.
Statistical analysis
Statistics were performed on Prism 10 (GraphPad, CA). Normality was tested using the D’Agostino & Pearson normality test for univariable analyses. The Mann–Whitney test was used for analyzes where a group was significantly non-normal. For the remaining univariable analyses, an F-test was performed and those comparisons with equal variance underwent an unpaired homoscedastic t-test. In contrast, those with significantly different variances underwent an unpaired t-test with a Welch’s correction. These tests were used for mechanical and dynamic allodynia measurements after dose escalation and withdrawal. Repeated measures two-way ANOVAs were used for longitudinal VF measurements, locomotor activity, and the hot plate assay. Sidak’s post hoc multiple comparisons test was used to compare several timepoints between animals of different genotypes within repeated measures assays across genotypes. Tukey’s post hoc multiple comparisons test was used to compare different drug treatments in wildtype animals at individual timepoints.
Results
TIAN promotes antiallodynic responses in the SNI model of neuropathic pain
We first investigated whether TIAN alleviates sensory hypersensitivity in models of neuropathic pain using the von Frey (VF) assay of punctate mechanical allodynia. We compared sustained antiallodynic efficacy in the SNI model (>16 hours post-injection) between TIAN (30 mg/kg i.p. twice daily (BID)) and DMI (15 mg/kg i.p. BID). As seen in Fig. 1A, we initiated drug treatment four weeks after nerve injury in male mice. TIAN or DMI was administered BID for three weeks. Neither TIAN nor DMI had any effect on mechanical hypersensitivity in sham animals (Fig. 1B). As shown in Fig. 1C, repeated TIAN or DMI treatment alleviated SNI-induced mechanical allodynia in the VF assay in a sustained manner, well after the expected clearance of TIAN and its active metabolite [34]. Importantly, TIAN treatment promoted significant alleviation of mechanical allodynia within 11 days post-drug (PD) administration (PD-D11), with a trending response effect by only PD-D7, whereas DMI did not produce a significant antiallodynic response until 15 days of treatment (PD-D15) (Fig. 1C, repeated measures two-way ANOVA Interaction F12,102 = 3.159, p = 0.0007; Tukey’s m.c. PD-D11 TIAN vs. Saline q = 4.742, df = 119, p = 0.0031; PD-D15 TIAN vs. Saline q = 6.853, df = 119, p < 0.0001; PD-D15 DMI vs. Saline q = 5.198, df = 119, p = 0.001).
A Flow diagram of procedures for testing TIAN and DMI prolonged antiallodynic effects and performing tissue collection. B No differences in mechanical threshold were observed between saline, TIAN, and DMI in sham animals. C TIAN treatment (30 mg/kg i.p. BID) resulted in significant antiallodynic effects relative to saline by day 11 post-treatment, whereas DMI resulted in significant antiallodynic effects by day 15 (n = 6–8/group). *p < 0.05 two-way r.m. ANOVA Tukey’s m.c. PD-D# post-drug administration day #, with measurements taken 16 hours after the previous drug injection.
We also confirmed that TIAN similarly reduces mechanical allodynia in both male and female mice after prolonged nerve injury. As seen in Supplementary Fig. 1A, B, i.p. administration of 30 mg/kg of TIAN twice-daily for three weeks increased normalized paw thresholds in both male and female mice at week 10 post-SNI, without affecting thresholds in sham animals (Males: two-way ANOVA Interaction F1,38 = 8.355, p = 0.0063; Tukey’s m.c. SNI-TIAN vs. SNI-Saline q = 3.802, df = 38, p = 0.0497; Females: two-way ANOVA Interaction F1,35 = 8.945, p = 0.0051; Tukey’s m.c. SNI-TIAN vs. SNI-Saline q = 3.984, df = 35, p = 0.0378).
Based on earlier observations that RGS proteins play dynamic and distinct modulatory roles in the antiallodynic efficacy and onset of actions of typical antidepressants such as DMI, we implemented the SNI model in three RGS knockout lines to determine if the antiallodynic actions of TIAN are modulated by similar mechanisms as DMI. We behaviorally profiled the acute antinociceptive, acute locomotor, and prolonged antiallodynic efficacy of TIAN in each of these RGS mouse lines. We then compared the antiallodynic effects of TIAN to those of DMI in each line as well.
RGSz1 negatively regulates the acute antinociceptive, but not the lasting antiallodynic, effects of TIAN
Previous findings demonstrated a powerful role of RGSz1 in negative modulation of acute analgesic responses to typical MOR agonists [21]. Here we used mice lacking the RGSz1 gene to determine if this RGS protein also impacts the antinociceptive and antiallodynic efficacy of TIAN, which acts as an atypical MOR agonist [34, 35]. As shown in Fig. 2A, RGSz1KO mice display a dose-dependent and prolonged increase in antinociceptive behavior in the hot plate paradigm, compared to their RGSz1WT counterparts (repeated measures two-way ANOVA Genotype Factor 15 mg/kg F1,19 = 6.648, p = 0.0184; 30 mg/kg F1,19 = 6.467, p = 0.0198; Sidak’s m.c. 15 mg/kg at 30 min RGSz1KO vs. WT t = 3.268, df = 38, p = 0.0046). Knockout of RGSz1 prevents TIAN-induced locomotor stimulation 10–20 minutes after drug injection (Fig. 2B; repeated measures two-way ANOVA Genotype Factor 15 mg/kg F1,18 = 4.177, p = 0.0559; Sidak’s m.c. t = 3.728, df = 378, p = 0.0047), in line with prior evidence that this mutant line shows reduced locomotor responses to opioids [21].
A RGSz1KO mice show higher thermal thresholds than RGSz1WT mice after TIAN administration in a dose-dependent fashion, with a genotype effect in both the 15 mg/kg cohort and the 30 mg/kg cohort (*p < 0.05 two-way r.m. ANOVA genotype factor; n = 10/11 animals/group). Significantly elevated thermal thresholds were observed in RGSz1KO mice 30 min after 15 mg/kg i.p. TIAN (**p < 0.01 two-way r.m. ANOVA Sidak’s m.c.) as highlighted by the blue shading. B RGSz1KO mice displayed significantly lower locomotor activation than RGSz1WT mice at 10–20 min after i.p. Tianeptine injection (**p < 0.01 two-way r.m. ANOVA Sidak’s m.c.; n = 10 animals/group). C No differences in recovery from mechanical allodynia were observed after prolonged DMI treatment (10 mg/kg i.p. BID) (n = 6–7 animals/group). D No differences in antiallodynic effects were seen between RGSz1WT and RGSz1KO in the 15 mg/kg TIAN cohort. E No differences in antiallodynic effects were seen between RGSz1WT and RGSz1KO in the 30 mg/kg TIAN cohort.
For mechanical allodynia testing, we first confirmed previously published findings that demonstrated a lack of baseline sensitivity differences between RGSz1WT and KO animals at early time points after nerve injury (WT: 1.188 g ± 0.0856 (n = 16); KO: 1.125 ± 0.0793 (n = 16); unpaired t-test t = 1.393, df = 30, p = 0.174). We next tested the effects of RGSz1KO on recovery from SNI-induced mechanical allodynia at 5 mg/kg and 10 mg/kg of DMI. We did not observe any differences in VF responses between RGSz1KO and RGSz1WT animals (Fig. 2C). We did not observe any differences between RGSz1KO and RGSz1WT animals 16 h post-TIAN injection at either 15 or 30 mg/kg BID (Fig. 2D, E) . A summary of preclinical studies to date on the effects of RGSz1 genetic manipulations and opioid/antidepressant effects on sensory and locomotor behavioral phenotypes is shown in Supplementary Table 1.
RGS4 positively regulates the acute antinociceptive and lasting antiallodynic effects of TIAN
RGS4 has been implicated in the regulation of both opioid and antidepressant drug actions. Here, we show that genetic inactivation of RGS4 reduced the antinociceptive efficacy of TIAN in the hot plate assay (Fig. 3A; repeated measures two-way ANOVA Genotype Factor F1,29 = 7.208, p = 0.0119; Sidak’s m.c. 15 min t = 3.974, df = 87, p = 0.0004), consistent with the previously observed overall reduction in opioid antinociceptive efficacy in RGS4KO mice compared to their WT counterparts. Knockout of RGS4 led to drastically reduced locomotor activation relative to WT animals 10–40 minutes after TIAN injection (Fig. 3B; repeated measures two-way ANOVA Genotype Factor F1,16 = 1.811, p = 0.197; Sidak’s m.c. 10–20 min t = 4.859, df = 320, p < 0.0001, 20–30 min t = 3.968, df = 320, p < 0.0018, 30–40 min t = 3.686, df = 320, p = 0.0053). These findings suggest that RGS4 is a positive modulator of both TIAN’s acute antinociceptive and locomotor effects.
A Knockout of RGS4 reduced the acute antinociceptive efficacy of TIAN in the hot plate assay (*p < 0.05 two-way r.m. ANOVA genotype factor), specifically at 15 min post-injection (***p < 0.001 two-way r.m. Sidak’s m.c.; n = 14–17 animals/group). B RGS4KO mice displayed significantly lower locomotor activity than RGS4WT mice 10–40 min after 30 mg/kg TIAN injection (**p < 0.01, ****p < 0.0001 two-way r.m. ANOVA Sidak’s m.c:.n = 9 animals/group). C Knockout of the RGS4 gene reduced the antiallodynic properties of DMI (10 mg/kg i.p. BID) relative to RGS4WT counterparts (*p < 0.05 two-way r.m. ANOVA genotype factor), on day 13 and 15 post-injection (*p < 0.05 two-way r.m. ANOVA Sidak’s m.c.; n = 7 animals/group). D There was a sustained reduction in the antiallodynic efficacy of TIAN in RGS4KO mice compared to their wild type counterparts (**p < 0.01 two-way r.m. ANOVA genotype factor), with a significant reduction in antiallodynic efficacy on day 21 post-SNI (*p < 0.05 two-way r.m. ANOVA Sidak’s m.c.; n = 11-12 animals/group).
As with prior studies [36], we did not observe any baseline differences in mechanical thresholds in the VF assay between RGS4KO and RGS4WT animals at early time points after nerve injury (WT: 1.145 g ± 0.0813 (n = 11); KO: 1.067 ± 0.0829 (n = 12); unpaired t-test t = 0.6767, df = 21, p = 0.506). We observed a significant reduction in the antiallodynic efficacy of DMI in RGS4KO animals on and after day 13 post-drug administration (Fig. 3C; repeated measures two-way ANOVA Genotype Factor F1,12 = 5.779, p = 0.0333; Sidak’s m.c. day 13 t = 3.074, df = 60, p = 0.0158, day 15 t = 2.92, df = 60, p = 0.0244). We also observed a lasting negative modulation of the antiallodynic effects of TIAN in RGS4KO mice, with a significant reduction in efficacy 18 days after initiation of treatment (repeated measures two-way ANOVA Genotype Factor F1,21 = 13.62, p = 0.0014; Sidak’s m.c. day 21 t = 3.205, df = 147, p = 0.0116; Fig. 3D). A summary of preclinical studies to date on the effects of RGS4 genetic manipulations and opioid/antidepressant effects on sensory and locomotor behavioral phenotypes is shown in Supplementary Table 2.
RGS9-2 positively regulates the acute antinociceptive and lasting antiallodynic effects of TIAN
RGS9-2 has been previously shown to be a positive regulator of methadone and fentanyl but a negative regulator of morphine in the hot plate antinociceptive assay [37]. Here, we performed the hot plate assay in the RGS9KO line to determine whether RGS9 also modulates responses to TIAN. The acute antinociceptive efficacy of TIAN (15 mg/kg) was drastically reduced in RGS9KO animals (Fig. 4A; repeated measures two-way ANOVA Genotype Factor F1,13 = 10.42, p = 0.0066; Sidak’s m.c. 15 min t = 2.661, df = 39, p = 0.0334, 30 min t = 2.733, df = 39, p = 0.0279). This suggests that TIAN engages molecular pathways similarly to fentanyl and methadone. RGS9KO did not affect TIAN-induced locomotor activation (Fig. 4B).
A Knockout of the RGS9 gene leads to a reduction of acute antinociceptive efficacy TIAN (15 m/kg i.p.) in the hot plate assay (**p < 0.01 two-way r.m. ANOVA genotype factor) 15 and 30 min post-injection (*p < 0.05 two-way r.m. ANOVA Sidak’s m.c.; n = 7–8 animals/group). B No differences were observed in locomotor stimulation by a 15 mg/kg TIAN injection between RGS9WT and RGS9KO animals (n = 5–7/group). C RGS9KO animals demonstrate an antiallodynic response to a subtherapeutic dose of DMI (5 mg/kg i.p. BID) (**p < 0.01 two-way r.m. ANOVA genotype factor), on days 4, 10, 12, and 17 post-injection (*p < 0.05, **p < 0.01 two-way r.m. ANOVA Sidak’s m.c.; n = 4–6 animals/group). D RGS9KO displayed an accelerated onset of antiallodynic effects to low-dose TIAN (15 mg/kg i.p. BID) compared to their RGS9WT counterparts (**p < 0.01 two-way r.m. ANOVA genotype factor), specifically on days 6 and 9 post-injection (*p < 0.05 two-way r.m. ANOVA Sidak’s m.c.; n = 7-9 animals/group). E No mechanical threshold differences were observed between RGS9WT and RGS9KO SNI animals after dose escalation nor during spontaneous withdrawal. F No differences in mechanical thresholds were observed between RGS9WT and RGS9KO sham animals treated with 15 mg/kg i.p. TIAN BID. G Elevating the dose to 30 mg/kg BID and spontaneous withdrawal did not affect mechanical thresholds in the sham cohort.
We did not observe baseline sensitivity differences between RGS9-2WT and KO animals at early time points after nerve injury (WT: 1.185 g ± 0.131 (n = 13); KO: 1.2 ± 0.0633 (n = 15); unpaired Welch’s t-test t = 0.1058, df = 17.49, p = 0.917), as previously described [23]. RGS9-2 has also been shown to act as a negative regulator of antidepressant action on allodynic behaviors [23]. Indeed, RGS9KO resulted in enhanced alleviation of mechanical hypersensitivity on day 10 of DMI administration at a sub-threshold dose (5 mg/kg) compared to RGSWT animals (Fig. 4C; repeated measures two-way ANOVA Genotype Factor F1,8 = 15.38, p = 0.0044; Sidak’s m.c. day 10 t = 3.239, df = 40, p = 0.012, day 12 t = 3.077, df = 40, p = 0.0187, day 17 t = 3.887, df = 40, p = 0.0019). Similarly, 6–9 days after the initiation of TIAN administration at 15 mg/kg, half the dose used in wild type animals to achieve a response, SNI groups of RGS9KO mice showed significantly higher mechanical thresholds than their wildtype counterparts, with no differences between sham groups (Fig. 4D, F; repeated measures two-way ANOVA Genotype Factor F1,14 = 9.906, p = 0.0071; Sidak’s m.c. day 15 t = 3.211, df = 98, p = 0.0125, day 18 t = 2.857, df = 98, p = 0.036). At 30 mg/kg TIAN, where the dose is closer to saturation, we did not observe a difference in the antiallodynic efficacy of the drug in RGS9KO animals compared to RGS9WT groups, nor 5 days after cessation of TIAN administration (Fig. 4E, G). A summary of preclinical studies to date on the effects of RGS9 genetic manipulations and opioid/antidepressant effects on sensory and locomotor behavioral phenotypes is shown in Supplementary Table 3.
TIAN induces nucleus accumbens transcriptomic changes different than DMI
In order to identify differences between the effects of TIAN and DMI that might underlie differences in therapeutic dynamics, we carried out RNA-seq studies. Briefly, we administered TIAN, DMI, or saline for three weeks, beginning one month after SNI or sham surgery. We then extracted and sequenced the NAc, a region of interest based on its role in modulating antidepressant drug efficacy in models of peripheral nerve injury [38]. Our analysis revealed 770 differentially expressed genes (DEGs) (391 up and 379 down; p-nom < 0.05; Fig. S1A) in NAc from TIAN-treated animals under SNI conditions and 1947 DEGs (1063 up and 884 down; p-nom < 0.05; Fig. S1C) under sham. TIAN effects on NAc transcriptomics appeared to be injury state-dependent, as SNI-TIAN neuro-related top canonical pathways included “FGF Signaling” (p = 0.000465, Z = +2.121), “Reelin Signaling in Neurons” (p = 0.0028, Z = +2.121), and “Ephrin B Signaling” (p = 0.00412, Z = +2.000) (Fig. S1B), while Sham-TIAN neuro-related pathways included downregulated “CREB Signaling in Neurons” and “Neuroinflammation Signaling” (p = 0.000201, Z = −1.905) (Fig. S1D). Very little overlap in the top 20 pathways was observed between SNI and sham conditions after TIAN treatment.
RNA-seq of NAc from DMI-treated animals revealed 4527 DEGs (2560 up and 1967 down; p-nom < 0.05; Fig. S1E) in an SNI state and 2054 DEGs (1261 up and 793 down; p-nom < 0.05; Fig. S1G) under sham. NAc from DMI-treated SNI animals affected several neuroplasticity ontologies, including “Synaptogenesis Signaling Pathway” (p = 7.45E−07, Z = +2.325), “CREB Signaling in Neurons” (p = 0.0000103, Z = +5.883), and “Reelin Signaling in Neurons” (p = 0.000383, Z = +2.646) (Fig. S1D). DMI had a strong effect on neuronal pathways in the sham condition as well, including “Axonal Guidance Signaling” (p = 1.2e−08, Z = NaN), “Opioid Signaling” (p = 1.75e−06, Z = +2.535), “Synaptogenesis Signaling” (p = 2.42e−06, Z = +4.323), and “SNARE Signaling” (p = 1.7e−05, Z = +2.558). The “EIF2 Signaling” pathway, which has been implicated in stress-induced depression models, was substantially downregulated (p = 6.23E−13, Z = −4.529) [39, 40]. Of note, pathways in TIAN signaling were driven by far fewer DEGs than DMI on average (9.22 ± 0.86 vs. 66.44 ± 6.24), emphasizing a lower transcriptomic effect on NAc cells after TIAN treatment.
When assessing commonly-regulated p-nom < 0.05 DEGs between TIANvSaline_SNI, DMIvSaline_SNI, TIANvSaline_Sham, and DMIvSaline_Sham, we observed very little overlap between conditions (Fig. 5A). This suggests that both drugs have highly condition-specific effects on the NAc transcriptomic pathways. However, when utilizing rank-rank hypergeometric overlap, a threshold-free strategy, we determined that broader gene expression changes overall are concordant between TIAN and DMI under SNI conditions (Fig. 5B).
A Venn diagram comparing TIANvSal_SNI, TIANvSal_Sham, DMIvSal_SNI, and DMIvSal_Sham conditions (n = 4–6 animals/condition). B RRHO plot depicting threshold-free concordance between TIAN and DMI effects on gene expression in SNI animals, despite vast differences in DEGs. C Volcano plot depicting DEGs from the direct TIANvDMI_SNI comparison (blue = p-adj < 0.1, yellow = p-nom < 0.05). D Top canonical pathways predicted by Qiagen IPA from the TIANvDMI_SNI comparison (p-nominal < 0.05). E, F Bar graph depicting upstream regulators predicted to be activated (top) and inhibited (bottom) in SNI animals treated with TIAN compared to DMI (p value of overlap < 0.05). G Venn diagram comparing SNIvSham_Saline DEGs to TIANvSal_SNI and DMIvSal_SNI DEGs. H Union heatmap comparing TIANvSal_SNI and DMIvSal_SNI p-nom < 0.05 gene expression changes to SNIvSham_Sal DEGs, a reference control for nerve injury-induced DEGs, on a log2FC > |0.32| cutoff. I–K GO: Molecular Function (GO: MF; Enrichr) mechanisms associated with shared, TIAN-unique, and DMI-unique counter-regulated, SNI-specific DEGs (p-nom < 0.05), respectively.
Despite this broad concordance, differences in therapeutic efficacy can arise from differential expression of key genes under specific injury conditions. Therefore, we directly compared TIAN vs. DMI DEGs under SNI conditions, which resulted in 5630 DEGs (2495 up and 3135 down; p-nom < 0.05; Fig. 5C). IPA pathway analysis of the TIANvDMI_SNI comparison highlighted that, relative to DMI, TIAN alters GPCR and CREB signaling pathway transcriptomics much less (Fig. 5D). Interestingly, using IPA upstream regulator analysis, TIAN effects may involve an upregulation of Levodopa- or alpha-synuclein-associated cascades (Fig. 5E), and a downregulation of broader HDAC- and NKX2-1-associated signaling (Fig. 5F). Previous literature has suggested that administration of L-DOPA to the NAc can alleviate SNI-induced synaptic long-term depression [41]. Furthermore, our group and others have shown a potent role of HDAC inhibition (HDAC1/5/6) in direct antiallodynic and antidepressant efficacy [23, 42,43,44,45, 30]. These differences in transcriptional regulation may contribute to the quicker onset of TIAN antiallodynic action during SNI.
In order to further assess gene changes induced by TIAN and DMI within the realm of DEGs induced by SNI alone, we compared TIANvSal_SNI, DMIvSal_SNI, and SNIvSham_Sal conditions. We observed few overlapping genes across all three conditions and few DEGs co-regulated between DMIvSal_SNI & SNIvSham_Sal and TIANvSal_SNI & SNIvSham_Sal (Fig. 5G). Following a more conservative approach, we added a log2FC ≥ |0.32| expression threshold and plotted the three conditions in a union heatmap (Fig. 5H). Interestingly, while TIAN treatment resulted in a much lower number of DEGs, at face value, gene changes associated with TIAN appeared to more directly counteract SNI-specific changes in the NAc compared to DMI.
Of the 1465 DEGs from the SNIvSham_Sal comparison (log2FC ≥ |0.32|, p-nom < 0.05), TIAN treatment resulted in the significant (p-nom < 0.05) counter-regulation of 58 genes, while DMI treatment counter-regulated 104 genes. Only 22 genes were commonly counter-regulated by both TIAN and DMI treatment. In agreement with the union heatmap above, 100% (58/58) of significant DEGs from TIAN treatment under SNI contra-regulated DEGs from the SNIvSham_Sal comparison. In contrast, only 43.3% (104/240) DEGs from DMI treatment under SNI contra-regulated SNI-induced DEGs.
To better understand the molecular pathways associated with these shared and TIAN- or DMI-specific injury-related DEGs, we used Enrichr’s GO: Molecular Function gateway. As seen in Fig. 5I, genes commonly regulated by TIAN and DMI treatment were associated with metabolic activity, such as “Flavin Adenine Dinucleotide Binding” and “Oxidoreductase Activity”. TIAN-specific counter-regulation of the SNI transcriptomic signature was primarily associated with ion-mediated signaling, such as “Sodium Channel Regulatory Activity”, “Gap Junction Channel Activity”, and “High Voltage-gated Calcium Channel Activity”, and growth factor signaling, including “Type 1 Fibroblast Growth Factor Receptor Binding” and “Growth Factor Activity”. DMI-specific counter-regulation, on the other hand, was primarily associated with the inflammatory cascade, such as “Vascular Endothelial Growth Factor-activated Receptor Activity” and “CCR5 Chemokine Receptor Binding”, as well as GPCR and Transmembrane Receptor Kinase activity. This pathway analysis further reinforced that TIAN and DMI engage unique transcriptional pathways in nerve injury states.
Discussion
Our findings reveal that the atypical antidepressant TIAN alleviates mechanical and dynamic allodynia in the SNI model of neuropathic pain. This is an important observation for the management of neuropathic pain conditions, as TIAN acts faster than typical antidepressants, may produce fewer adverse effects, and can be administered for long periods of time without analgesic tolerance or physical dependence [34].
As the development of refined analogs of atypical antidepressants holds promise for safer and more effective chronic pain treatment strategies, the need for an in vivo screening method that allows comparisons with various clinically used opioids and antidepressants is substantial. In vivo screening assays using genetically modified mice may also provide a better understanding of intracellular pathways associated with the therapeutic effects of novel and currently prescribed medications. Such approaches will highlight G protein signaling components mediating functional responses to medications used to manage acute and chronic pain symptoms. Our group has previously demonstrated unique behavioral signatures of constitutive and conditional RGS knockout mouse lines in response to clinically used opioids and antidepressants [18]. While the RGS family members we tested show overlapping yet distinct patterns of expression, each protein plays unique roles in addiction, physical dependence, analgesia, and tolerance, as well as in the modulation of the actions of mood-stabilizing drugs. Here, we tested the antinociceptive, antiallodynic, and locomotor stimulating properties of TIAN in RGSz1, RGS4, and RGS9-2 knockout mouse lines. This approach allowed us to compare behavioral signatures to those observed with opioid analgesics and typical antidepressants, gain insight into G protein modulatory pathways that control the onset of action and efficacy of TIAN, and to better define mechanisms underlying antiallodynic and antinociceptive responses to this atypical antidepressant.
RGS4 is expressed in several stress and nociception-modulating regions, including the prefrontal cortex, NAc, and thalamus. Our earlier work showed that RGS4 positively modulates the antinociceptive efficacy of fentanyl and methadone while it has no effect on morphine-induced analgesia or analgesic tolerance [22]. Here, we observe a similar positive modulatory role of RGS4 in the acute antinociceptive efficacy of TIAN to that we observed with synthetic opioids. Our group has previously found that RGS4 promotes the actions of TCAs, SNRIs, and SSRIs [27], as knockout of RGS4 blunted responses to standard drug doses. Similar to its action on typical antidepressants, RGS4 promotes the antiallodynic efficacy of TIAN. This is a critical piece of information, as it suggests that the sustained effects of TIAN require RGS4-regulated pathways, which appear to be essential for therapeutic responses to most clinically available antidepressants.
RGS9-2 is enriched in the striatum and expressed in low amounts in several other sites modulating analgesia and emotional responses. RGS9-2 has been shown to modulate functional responses to opioids in an agonist-dependent manner. Specifically, it serves as a negative modulator of morphine-induced analgesia, but like RGS4, it positively regulates fentanyl, methadone, and oxycodone-induced analgesia. RGS9-2 also plays a dynamic role in drug reward sensitivity and in the development of physical dependence [31]. Here, we show that RGS9-2 promotes the antinociceptive actions of TIAN in the hot plate assay in a similar manner to methadone and fentanyl but not morphine. These data further support earlier work demonstrating that the actions of TIAN are mediated by distinct signaling pathways to those of morphine [46, 47]. In terms of antidepressant drug responses, RGS9-2 actions negatively regulate the antiallodynic efficacy of DMI and Duloxetine while also delaying their onset of action. Consistent with the observations on monoamine-targeting antidepressants, RGS9-2 delays the onset of TIAN action. These data suggest that the antiallodynic actions of TIAN and DMI are commonly regulated by RGS9-2.
Unlike RGS9-2 and RGS4, which show agonist-dependent modulatory roles, the brain-expressed RGSz1 protein acts as a negative modulator of the analgesic actions of morphine and synthetic opioids in models of acute and chronic pain [21]. Our findings suggest a similar negative modulatory role of RGSz1 in acute antinociceptive responses to TIAN. Thus, RGSz1 plays a unique role as a negative modulator of TIAN antinociceptive actions, while RGS4 and RGS9-2 act as positive modulators. Future work will further elucidate if RGSz1 complexes that modulate TIAN actions involve Gαz and/or additional Gαi subunits. We speculate that the phenotypes we observe result from RGS actions in various analgesic circuits, and additional RGS family members may also modulate the antinociceptive actions of TIAN.
Interestingly, we show that RGSz1 does not affect the efficacy or onset of antiallodynic effects of monoamine-targeting antidepressants or TIAN. While RGSz1 is expressed throughout the brain, it has primarily been shown to modulate opioidergic responses by actions in the periaqueductal gray [21]. Thus, it is unsurprising that sustained antiallodynic responses to TCAs and TIAN are not affected by RGSz1.
Earlier studies have reported brief acute locomotor activating effects of TIAN [34]. Here, we show that RGS4 and RGSz1 positively regulate the locomotor activating effects of TIAN, similar to observations with synthetic opioids. Importantly, the locomotor activating effects of TIAN are not affected by RGS9-2, which is a key component of addiction-related plasticity in the striatum.
Given the similar regulation of TIAN and DMI by RGS proteins and the role of the NAc in the regulation of chronic pathological pain states and pain perception [28, 48, 49], we decided to use RNA-seq in the NAc of TIAN, DMI, or saline-treated SNI or sham animals to begin to identify potential transcriptomic mechanisms that might underlie the observed difference in onset of persistent antiallodynic effects between these two drugs. We found that TIAN causes nerve injury-specific transcriptomic changes in the NAc that differ substantially from DMI and more-specifically counteract SNI-induced gene expression changes. Although TIAN induced fewer gene expression changes in the NAc compared to DMI, the changes that did occur appeared to more directly counteract the injury-associated transcriptional pattern observed after prolonged SNI. Our prediction analyses implicated RGS-associated pathways after TIAN treatment of SNI animals, such as JAK-STAT signaling, which is essential for the regulation of neuronal plasticity [50] and μ-/δ-opioid receptor signaling [51]. Considering other groups have identified synaptic long-term depression as a key mechanism of NAc involvement in pain processing [41], these TIAN-induced molecular pathways hold special relevance for treating neuropathic pain. Our study also highlighted potentially new links between TIAN and epigenetic mechanisms underlying chronic pain states, such as HDAC inhibition.
Analysis of TIAN-mediated counter-regulation of SNI-induced DEGs further revealed the drug’s role in the regulation of ion channels and growth factor signaling. Prior studies have demonstrated sodium and calcium-permeable current/channel dysregulation in the NAc under withdrawal, depression-, and pain-like states [52, 53]. However, a comprehensive view of the effects of prolonged peripheral nerve injury NAc sodium/calcium channel regulation does not yet exist. A recent study also demonstrated that prolonged peripheral nerve injury and localized inflammatory perturbation can result in enhanced neurogenesis [54], a process often resulting from changes in neurotrophic/growth factor signaling, which is implicated in mood disorders [55]. However, the explicit effects of peripheral nerve injury on NAc growth factor signaling are understudied, as most efforts have focused on the hippocampus due to its neurogenerative capacity.
Ultimately, this study emphasizes the importance of future investigation of the impact of TIAN on G protein-RGS protein-protein interactions, as well as its effects on signal transduction events in reward circuitry and other components of the nociceptive pathway. Furthermore, given known robust sex differences in pain pathophysiology and antidepressant drug efficacy, it will be crucial to determine if the observed antiallodynic effect of TIAN in female mice occurs through similar mechanisms to those found in males in this study. Information from this work can also be used to optimize the actions of TIAN and guide the development of compounds that preferentially activate intracellular pathways that promote strong antiallodynic effects for the management of neuropathic pain.
References
Dahlhamer J, Lucas J, Zelaya C, Nahin R, Mackey S, DeBar L, et al. Prevalence of chronic pain and high-impact chronic pain among adults—United States, 2016. Morb Mortal Wkly Rep. 2018;67:1001–6.
Latremoliere A, Woolf CJ. Central sensitization: a generator of pain hypersensitivity by central neural plasticity. J Pain. 2009;10:895–926.
Serafini RA, Pryce KD, Zachariou V. The mesolimbic dopamine system in chronic pain and associated affective comorbidities. Biol Psychiatry. 2020;87:64–73.
Kelly K, Posternak M, Alpert JE. Toward achieving optimal response: understanding and managing antidepressant side effects. Dialogues Clin Neurosci. 2008;10:409–18.
Swann AC. Major system toxicities and side effects of anticonvulsants. J Clin Psychiatry. 2001;62:16–21.
Benyamin R, Trescot AM, Datta S, Buenaventura R, Adlaka R, Sehgal N, et al. Opioid complications and side effects. Pain Physician. 2008;11:S105–20.
Kasper S, Olié JP. A meta-analysis of randomized controlled trials of tianeptine versus SSRI in the short-term treatment of depression. Eur Psychiatry. 2002;17:331–40.
Nickel T, Sonntag A, Schill J, Zobel AW, Ackl N, Brunnauer A, et al. Clinical and neurobiological effects of tianeptine and paroxetine in major depression. J Clin Psychopharmacol. 2003;23:155–68.
Bilge SS, Bozkurt A, Ilkaya F, Çiftcioǧlu E, Kesim Y, Uzbay TI. The antinociceptive effects of intravenous tianeptine in colorectal distension-induced visceral pain in rats: the role of 5-HT 3 receptors. Eur J Pharm. 2012;681:44–9.
Heo BH, Shin JY, Park KS, Lee HG, Choi JI, Yoon MH, et al. Effects of tianeptine on the development and maintenance of mechanical allodynia in a rat model of neuropathic pain. Neurosci Lett. 2016;633:82–6.
Kim WM, Lee SH, Jeong HJ, Lee HG, Choi JI, Yoon MH. The analgesic activity of intrathecal tianeptine, an atypical antidepressant, in a rat model of inflammatory pain. Anesth Analg. 2012;114:683–9.
Kole MHP, Swan L, Fuchs E. The antidepressant tianeptine persistently modulates glutamate receptor currents of the hippocampal CA3 commissural associational synapse in chronically stressed rats. Eur J Neurosci. 2002;16:807–16.
Svenningsson P, Bateup H, Qi H, Takamiya K, Huganir RL, Spedding M, et al. Involvement of AMPA receptor phosphorylation in antidepressant actions with special reference to tianeptine. Eur J Neurosci. 2007;26:509–17.
McEwen BS, Chattarji S, Diamond DM, Jay TM, Reagan LP, Svenningsson P, et al. The neurobiological properties of tianeptine (Stablon): from monoamine hypothesis to glutamatergic modulation. Mol Psychiatry. 2010;15:237–49.
Shields SD, Eckert WA, Basbaum AI. Spared nerve injury model of neuropathic pain in the mouse: a behavioral and anatomic analysis. J Pain. 2003;4:465–70.
Backonja MM, Serra J. Pharmacologic management part 1: better-studied neuropathic pain diseases. Pain Med. 2004;5:S28–47.
Luttrell LM, Maudsley S, Bohn LM. Fulfilling the promise of ‘biased’ G protein-coupled receptor agonism. Mol Pharm. 2015;88:579–88.
Sakloth F, Polizu C, Bertherat F, Zachariou V. Regulators of G protein signaling in analgesia and addiction. Mol Pharm. 2020;119:119206.
Senese NB, Kandasamy R, Kochan KE, Traynor JR. Regulator of G-protein signaling (RGS) protein modulation of opioid receptor signaling as a potential target for pain management. Front Mol Neurosci. 2020;13:5.
von Zastrow M. Proteomic approaches to investigate regulated trafficking and signaling of G protein–coupled receptors. Mol Pharmacol 2021;99:392–8.
Gaspari S, Purushothaman I, Cogliani V, Sakloth F, Neve RL. Suppression of RGSz1 function optimizes the actions of opioid analgesics by mechanisms that involve the Wnt/β-catenin pathway. Proc Natl Acad Sci USA. 2018; 115:E2085–94.
Han MH, Renthal W, Ring RH, Rahman Z, Psifogeorgou K, Howland D, et al. Brain region specific actions of regulator of G protein signaling 4 oppose morphine reward and dependence but promote analgesia. Biol Psychiatry. 2010. https://doi.org/10.1016/j.biopsych.2009.08.041.
Mitsi V, Terzi D, Purushothaman I, Manouras L, Gaspari S, Neve RL. RGS9-2-controlled adaptations in the striatum determine the onset of action and efficacy of antidepressants in neuropathic pain states. Proc Natl Acad Sci USA. 2015;112:E5088–97.
Terzi D, Gaspari S, Manouras L, Descalzi G, Mitsi V, Zachariou V. RGS9-2 modulates sensory and mood related symptoms of neuropathic pain. Neurobiol Learn Mem. 2014;115:43–8.
Sutton LP, Khalatyan N, Savas JN, Martemyanov KA. Striatal rgs7 regulates depression-related behaviors and stress-induced reinstatement of cocaine conditioned place preference. ENeuro. 2021;8:ENEURO.0365-20.2020.
Hollinger S, Hepler JR. Cellular regulation of RGS proteins: modulators and integrators of G protein signaling. Pharm Rev. 2021;54:8–59.
Stratinaki M, Varidaki A, Mitsi V, Ghose S, Magida J, Dias C, et al. Regulator of G protein signaling 4 is a crucial modulator of antidepressant drug action in depression and neuropathic pain models. Proc Natl Acad Sci USA. 2013;110:8254–9.
Baliki M, Petre B, Torbey S, Herrmann K, Huang L, Schnitzer T, et al. Corticostriatal functional connectivity predicts transition to chronic pain. Nat Neurosci. 2013;15:1117–9.
Descalzi G, Mitsi V, Purushothaman I, Gaspari S, Avrampou K, Loh YHE, et al. Neuropathic pain promotes adaptive changes in gene expression in brain networks involved in stress and depression. Sci Signal. 2017;10:1–18.
Pryce, KD. et al. Oxycodone withdrawal induces HDAC1/HDAC2-dependent transcriptional maladaptations in the reward pathway in a mouse model of peripheral nerve injury. Nat. Neurosci. 2023:1–16.
Gaspari S, Cogliani V, Manouras L, Anderson EM, Mitsi V, Avrampou K, et al. RGS9-2 modulates responses to oxycodone in pain-free and chronic pain states. Neuropsychopharmacology. 2017;42:1548–56.
Zachariou V, Georgescu D, Sanchez N, Rahman Z, DiLeone R, Berton O, et al. Essential role for RGS9 in opiate action. Proc Natl Acad Sci USA. 2003. https://doi.org/10.1073/pnas.2232594100.
Avrampou K, Pryce KD, Ramakrishnan A, Sakloth F, Gaspari S, Serafini RA, et al. RGS4 maintains chronic pain symptoms in rodent models. J Neurosci. 2019;39:8291–304.
Samuels BA, Nautiyal KM, Kruegel AC, Levinstein MR, Magalong VM, Gassaway MM, et al. The behavioral effects of the antidepressant tianeptine require the Mu-opioid receptor. Neuropsychopharmacology. 2017;42:2052–63.
Gassaway MM, Rives ML, Kruegel AC, Javitch JA, Sames D. The atypical antidepressant and neurorestorative agent tianeptine is a μ-opioid receptor agonist. Transl Psychiatry. 2014;4:e411–5.
Stratinaki M, Varidaki A, Mitsi V, Ghose S, Magida J, Dias C, et al. Regulator of G protein signaling is a crucial modulator of antidepressant drug action in depression and neuropathic pain models. Proc Natl Acad Sci USA. 2013;110:8254–9.
Psifogeorgou K, Papakosta P, Russo SJ, Neve RL, Kardassis D, Gold SJ, et al. RGS9-2 is a negative modulator of μ-opioid receptor function. J Neurochem. 2007;103:617–25.
Mitsi V, Terzi D, Purushothaman I, Manouras L, Gaspari S, Neve RL, et al. RGS9-2–controlled adaptations in the striatum determine the onset of action and efficacy of antidepressants in neuropathic pain states. Proc Natl Acad Sci USA. 2015;112:E5088–97.
Lin LC, Sibille E. Somatostatin, neuronal vulnerability and behavioral emotionality. Mol Psychiatry. 2015;20:377–87.
Girgenti MJ, Wohleb ES, Mehta S, Ghosal S, Fogaca MV, Duman RS. Prefrontal cortex interneurons display dynamic sex-specific stress-induced transcriptomes. Transl Psychiatry. 2019;9:292.
Ren W, Centeno MV, Berger S, Wu Y, Na X, Liu X, et al. The indirect pathway of the nucleus accumbens shell amplifies neuropathic pain. Nat Neurosci. 2016;19:220–4.
Sakloth F, Manouras L, Avrampou K, Mitsi V, Serafini RA, Pryce KD, et al. HDAC6-selective inhibitors decrease nerve-injury and inflammation-associated mechanical hypersensitivity in mice. Psychopharmacology. 2020;237:2139–49.
Zhang J, Ma J, Trinh RT, Heijnen CJ, Kavelaars A. An HDAC6 inhibitor reverses chemotherapy-induced mechanical hypersensitivity via an IL-10 and macrophage dependent pathway. Brain Behav Immun. 2022;100:287–96.
Sanna MD, Galeotti N. The HDAC1/c-JUN complex is essential in the promotion of nerve injury-induced neuropathic pain through JNK signaling. Eur J Pharm. 2018;825:99–106.
Sanna MD, Guandalini L, Romanelli MN, Galeotti N. The new HDAC1 inhibitor LG325 ameliorates neuropathic pain in a mouse model. Pharm Biochem Behav. 2017;160:70–5.
Gaspari S, Papachatzaki MM, Koo JW, Carr FB, Tsimpanouli ME, Stergiou E, et al. Nucleus accumbens-specific interventions in RGS9-2 activity modulate responses to morphine. Neuropsychopharmacology. 2014;39:1968–77.
Psigfogeorgou K, Terzi D, Papachatzaki MM, Varidaki A, Ferguson D, Gold SJ, et al. A unique role of RGS9-2 in the striatum as a positive or negative regulator of opiate analgesia. J Neurosci. 2011;31:5617–24.
Chang PC, Pollema-Mays SL, Centeno MV, Procissi D, Contini M, Baria AT, et al. Role of nucleus accumbens in neuropathic pain: linked multi-scale evidence in the rat transitioning to neuropathic pain. Pain. 2014;155:1128–39.
Baliki MN, Chialvo DR, Geha PY, Levy RM, Harden RN, Parrish TB, et al. Chronic pain and the emotional brain: specific brain activity associated with spontaneous fluctuations of intensity of chronic back pain. J Neurosci. 2006;26:12165–73.
Pallaki P, Georganta EM, Serafimidis I, Papakonstantinou MP, Papanikolaou V, Koutloglou S, et al. A novel regulatory role of RGS4 in STAT5B activation, neurite outgrowth and neuronal differentiation. Neuropharmacology. 2017;117:408–21.
Georgoussi Z, Leontiadis L, Mazarakou G, Merkouris M, Hyde K, Hamm H. Selective interactions between G protein subunits and RGS4 with the C-terminal domains of the mu- and delta-opioid receptors regulate opioid receptor signaling. Cell Signal. 2006;18:771–82.
Zhang XF, Hu XT, White FJ. Whole-cell plasticity in cocaine withdrawal: reduced sodium currents in nucleus accumbens neurons. J Neurosci. 1998;18:488–98.
Goffer Y, Xu D, Eberle SE, D’amour J, Lee M, Tukey D, et al. Calcium-permeable AMPA receptors in the nucleus accumbens regulate depression-like behaviors in the chronic neuropathic pain state. J Neurosci. 2013;33:19034–44.
García-González D, Dumitru I, Zuccotti A, Yen TY, Herranz-Pérez V, Tan LL, et al. Neurogenesis of medium spiny neurons in the nucleus accumbens continues into adulthood and is enhanced by pathological pain. Mol Psychiatry. 2021;26:7851.
Schmidt HD, Duman RS. The role of neurotrophic factors in adult hippocampal neurogenesis, antidepressant treatments and animal models of depressive-like behavior. Behav Pharm. 2007;18:391–418.
Funding
This work was supported by NINDS NS086444 and NS111351 (VZ), the Hope for Depression Research Foundation (JAJ), and NIH T32 GM007280 (RAS).
Author information
Authors and Affiliations
Contributions
RAS and VZ contributed to the conception and design of this work. RAS, ME, EAP, and FS acquired, analyzed, and interpreted data. RAS, EAP, LS, JAJ, and VZ contributed to drafting and revising the work. VZ, JAJ, and LS provided a final approval of the submitted version to be published. RAS and VZ agree to be accountable for all aspects of the work.
Corresponding author
Ethics declarations
Competing interests
VZ declares that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest. JAJ is an inventor on tianeptine-related patents owned by Columbia University. RAS, ME, EAP, FS, LS, and VZ declare no competing interests.
Additional information
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Supplementary information
Rights and permissions
Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
About this article
Cite this article
Serafini, R.A., Estill, M., Pekarskaya, E.A. et al. Tianeptine promotes lasting antiallodynic effects in a mouse model of neuropathic pain. Neuropsychopharmacol. 48, 1680–1689 (2023). https://doi.org/10.1038/s41386-023-01645-w
Received:
Revised:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1038/s41386-023-01645-w
