Anticonvulsants modulate neuronal activity by acting on calcium and sodium channels and are among the most prescribed pharmacological treatments for PDN. Gabapentinoids, including pregabalin and gabapentin, are widely used as first-line treatments for PDN. These drugs act on the α₂δ subunit of voltage-gated calcium channels to reduce neurotransmitter release, calming overactive nerves that contribute to NP [13]. Pregabalin is particularly effective in patients with multiple NP syndromes and improves associated sleep disturbances [14, 15]. Mirogabalin, a newer drug, demonstrates higher selectivity for the α₂δ-1 subunit, with improved efficacy and fewer side effects [16]. However, gabapentinoids are associated with adverse side effects such as drowsiness, dizziness, weight gain, and peripheral edema [17, 18]. In addition, there is a risk of abuse, especially with pregabalin [19].

Sodium channel blockers, such as carbamazepine and oxcarbazepine, are a family of epilepsy medications sometimes prescribed for PDN. They inhibit voltage-gated sodium channels to reduce nerve excitability and alleviate pain. Carbamazepine is effective but has black box warnings for severe dermatological reactions, agranulocytosis and aplastic anemia [20]. Oxcarbazepine exhibits its analgesic effect mostly through its active metabolite licarbazepine, and offers a safer profile with fewer side effects, which makes it a preferable choice in this class [21]. However, these sodium channel blockers also carry with many side effects such as drowsiness, dizziness, headaches, nausea, vomiting, and double vision [22–24].

Serotonin-norepinephrine reuptake inhibitors (SNRIs) are a class of antidepressant medications used to treat major depressive disorder, anxiety disorders, chronic NP, and other medical diseases [12]. SNRIs such as duloxetine and venlafaxine are very effective for PDN and are particularly valuable for patients with comorbid depression or anxiety. Duloxetine, an FDA-approved first-line treatment, inhibits serotonin and norepinephrine reuptake to modulate pain signaling in the CNS. It often provides rapid pain relief, sometimes within the first week of treatment [25], with fewer side effects compared to older medications like tricyclic antidepressants (TCAs) [26–28]. Venlafaxine has a similar mechanism of action but with lower binding affinity to serotonin and norepinephrine transporters [29]. SNRIs are generally well-tolerated, though side effects such as nausea, dry mouth, and dizziness are possible [30, 31]. However, the potential for developing serotonin syndrome, a potentially life-threatening condition caused by an excessive accumulation of serotonin, requires careful monitoring [32].

TCAs are a broad class of medications that were first discovered in the early 1950s. Common examples include amitriptyline, nortriptyline, and imipramine, which remain another option for treating PDN. They exert their analgesic effect by inhibiting serotonin and norepinephrine reuptake and targeting additional receptors, such as opioid receptors and ion channels [33, 34]. TCAs are highly effective for NP and can improve associated sleep disturbances. However, their complex pharmacology results in many side effects, including dry mouth, constipation, urinary retention, drowsiness, and memory impairment, making them less favorable compared to SNRIs [35–38].

Several opioid drugs are clinically used for managing moderate to severe PDN, including tapentadol, tramadol, morphine and oxycodone. Tapentadol is approved by the FDA as first-line use [39]. Morphine, a potent μ-opioid receptor (MOR) agonist, provides effective relief for acute pain [40], while oxycodone demonstrates similar potency with active metabolites that enhance its analgesic effects [41]. Tapentadol and tramadol uniquely combine MOR agonism with norepinephrine reuptake inhibition, offering comparable efficacy to oxycodone but with fewer gastrointestinal side effects, such as constipation [42, 43]. Although opioids are effective for some patients to manage PDN, they come with significant risks, such as tolerance, dependence, and respiratory depression, which need cautious use and close monitoring [44–48].

Capsaicin, a plant-derived alkaloid from chili peppers, provides localized pain relief by binding transient receptor potential vanilloid 1 (TRPV1) receptors on nociceptive fibers [49]. The FDA-approved 8% capsaicin patch is applied to affected areas, such as the feet, and offers significant pain reduction in certain patient subpopulations with minimal systemic side effects. However, burning sensations at the application site are a common adverse reaction [50]. Overall, capsaicin has very few reported side effects beyond this localized discomfort. Additionally, antioxidants such as alpha-lipoic acid and benfotiamine supplement are also used to treat PDN in some countries [12, 51–53]. But none of these nutritional supplements have been approved by the FDA for use in the United States.

In summary, while current medications for PDN offer some relief, they are often associated with a range of side effects that limit their therapeutic effectiveness and patient compliance. Alarmingly, fewer than one-third of PDN patients achieve satisfactory pain relief with existing therapies, underscoring the urgent need for innovative treatments and the identification of new therapeutic targets.

It is noteworthy that several TCAs used for PDN treatment also bind to S1R, but with much lower affinity than their primary pharmacological targets. As a result, S1R modulation may play a secondary or minimal role in the analgesic effects of these drugs. Recent research suggests that directly targeting S1R offers a novel mechanism for alleviating NP across various etiologies, including PDN. As explored in this review, S1R represents a promising therapeutic target that could lead to more effective and mechanistically distinct treatments for PDN.

The X-ray crystallographic structures of the human S1R in complex with several known agonists and antagonists were successfully resolved at high resolution, offering significant insights into its molecular architecture and ligand interactions. The first breakthrough came in 2016 with the disclosure of human S1R crystal structures in complex with an antagonist (PD144418; Figure 2 and Figure 3) and an ambiguous ligand (4-IBP) [63]. Later in 2018, additional cocrystal structures of S1R with one agonist ((+)-pentazocine), two antagonists (haloperidol and NE-100; Figure 2) were published [61]. As illustrated in Figure 3, these crystal structures reveal S1R adopts a homotrimeric overall architecture, with each protomer containing a single transmembrane domain (Figure 3A). The ligand binding pocket in the S1R structure is located in the cytosolic domain within a β-barrel region, which is predominantly hydrophobic and occluded by the solvent. Key molecular interactions between the receptor and its ligands have been identified based on these crystal structures. Specifically, the anionic sidechain of Glu172 forms a salt-bridge with the cationic amine of the ligands, while the protonated Asp126 forms a hydrogen bond with Glu172 to retain the interaction network. Tyr103 interacts with ligands through π−π stacking and forms a hydrogen bond with Glu172 to stabilize its orientation. Additionally, several hydrophobic residues (Trp89, Leu95, Tyr103, Phe107, and Trp164) are involved in extensive nonpolar interactions with the hydrophobic or aromatic regions of the ligands (Figure 3B).

Given the highly occluded nature of the binding pocket revealed by the original crystal structures, the mechanisms of ligand enter and exit for S1R remain an interesting topic of study. In 2022, the crystal structures of Xenopus laevis S1R in complex with PRE084 (an agonist) and E-52862 (an antagonist) suggest that ligands likely access the binding site via conformational changes in the carboxy-terminal two-helix bundle, rather than the cupin-fold domain [60]. More recently in 2024, several cocrystal structures of Xenopus laevis with neurosteroid ligands, progesterone and DHEA sulfate, were resolved [59]. These findings highlight distinct binding modes of endogenous steroids and underscore the versatility of S1R’s ligand interactions.

Beyond static crystal structures, biochemical and cellular studies have demonstrated that S1R exhibits dynamic ligand-dependent oligomerization: antagonists stabilizing high-molecular-mass oligomers and agonists promoting their dissociation into monomers or dimers [89, 90]. However, X-ray crystallography studies have consistently revealed a homotrimeric architecture for S1R [59–61, 63]. This discrepancy may arise due to the stabilizing conditions used in crystallization, which can favor specific oligomeric conformations. The trimeric arrangement observed in X-ray structures might represent a stable intermediate or a functionally relevant oligomeric form captured in the absence of membrane dynamics and cellular signaling. These findings highlight the importance of studying S1R in diverse experimental conditions to fully understand its conformational flexibility and functional implications. Understanding the conformational changes induced by different ligand classes can aid in the rational design of novel compounds with optimized binding properties and functional outcomes. These advances in S1R structural biology have not only significantly enhanced our understanding of its ligand recognition mechanisms but also provides a solid foundation for the structure-based design of novel S1R-targeting compounds.

S1R interacts with several ion channels involved in pain modulation, including voltage-gated ion channels (e.g., Na_v1.5, K_v1.2, and Ca_v2.2), calcium-activated potassium channels (e.g., SK3), and ligand-gated ion channels [e.g., N-methyl-d-aspartate (NMDA) receptors] [78, 81, 82]. Notably, S1R has been shown to bind directly to the GluN1 subunit of the NMDA receptors [79, 82]. The GPCR nociceptors binding to S1R include cannabinoid CB1 and MOR [80, 83]. A direct interaction between MOR and S1R has also be described in transfected human embryonic kidney (HEK) cells [83]. Importantly, S1R antagonists have been shown to potentiate opioids’ (e.g., morphine, oxycodone) analgesia without exacerbating their side effects, such as dependence, tolerance, and constipation [91–93]. Studies using S1R KO mice and selective antagonists have revealed S1R’s critical roles in the modulation of pain of different etiologies, including chemical-induced hyperalgesia (formalin and capsaicin tests) [94, 95], central and peripheral neuropathy [64, 96, 97], osteoarthritis [98], and cancer pain [99]. In this review, we focus on NP, with a particular emphasis on PDN.

As summarized in Table 1, S1R has been extensively studied in preclinical NP models of multiple modalities. For instance, in a contusion spinal cord injury (SCI) model of central NP, S1R KO didn’t develop mechanical allodynia and thermal hyperalgesia in mice, while E-52862 (aka S1RA or MR309; Figure 2) alleviated mechanical allodynia and thermal hyperalgesia after both acute and prolonged treatments [100]. Similarly, S1R KO mice showed resistance to the development of mechanical allodynia and hyperalgesia in peripheral nerve injury models, such as paclitaxel-induced NP and partial sciatic nerve ligation (PSNL) [101, 102]. Selective S1R antagonists (e.g., BD-1047, S1RA, CM304, and PW507; Figure 2) have demonstrated analgesic efficacy in attenuating mechanical allodynia and hyperalgesia in a range of models, including chemotherapy (paclitaxel, cisplatin, and oxaliplatin) induced neuropathy [97, 101, 103], PSNL [102, 104, 105], and chronic constriction injury (CCI) [97, 106–108]. Furthermore, in clinical trials, E-52862 demonstrated significant efficacy to mitigate chronic oxaliplatin-induced peripheral neuropathy [109, 110]. These findings underscore S1R’s essential roles in amplifying pain responses across diverse pain conditions.

Several preclinical models of diabetes have been used to investigate S1R’s roles in the development of PDN, including chemical-induced, genetically modified, and diet-induced neuropathy models.

Streptozotocin (STZ), a toxin causing selective pancreatic β-cell destruction, induces physiological and behavioral changes resembling type 1 diabetes in humans [117–119]. In STZ-induced diabetic models, S1R has been found to be overexpressed in dorsal root ganglion, implicating its role in the development and maintenance of PDN [112]. BD-1047 inhibited STZ-induced PDN, such as mechanical allodynia and thermal hyperalgesia, after treatment for 7 days [intrathecal (i.t.)] [112]. Similarly, E-52862 significantly reduced mechanical allodynia both after acute [intraperitoneal (i.p.)] and 7-day [i.p., twice daily (b.i.d.)] treatments [97]. Furthermore, another potent and selective S1R antagonist, PW507, demonstrated robust efficacy in alleviating both mechanical allodynia and thermal hyperalgesia after acute (i.p.) and 14-day treatments (i.p., b.i.d.) [111].

In the nicotinamide (NA)-STZ model, which resembles type 2 diabetes by combining β-cell damage and insulin resistance, haloperidol and its analog LMH-2 (Figure 2) exhibited efficacy in alleviating PDN symptoms [subcutaneous (s.c.)], such as mechanical allodynia and thermal hyperalgesia [113].

In the Zucker diabetic fatty (ZDF) rat model, a genetic model of type 2 diabetes characterized by obesity and insulin resistance, E-52862 treatment produced significant reductions in mechanical allodynia and thermal hyperalgesia. These effects were observed after both acute (i.p.) administration and a sustained 14-day twice-daily (b.i.d.) regimen, emphasizing the compound’s robust therapeutic efficacy in this context [114].

Meanwhile, the high-fat diet (HFD)-induced neuropathy model, which simulates the metabolic dysfunction associated with diet-induced obesity, provided additional insights into S1R’s role in PDN. In this model, S1R KO mice did not develop mechanical allodynia and thermal hyperalgesia. Complementing these findings, BD-1047 (i.t.) effectively reduced both mechanical allodynia and thermal hyperalgesia in the same model [115].

Together, these findings across diverse preclinical studies reinforce the therapeutic promise of S1R antagonists in mitigating PDN, offering hope for improved management of this debilitating condition in diabetic patients. However, the precise mechanisms by which S1R antagonists alleviate PDN remain unclear. S1R has been reported to interact with the NR1 subunit of the NMDA receptor, a key player in the development of NP [82, 120–122]. In diabetic neuropathy models induced by STZ, ZDF, or HFD, nerve injuries activate the NMDA receptor, contributing to central sensitization and heightened pain sensitivity. Antagonizing S1R disrupts its interaction with NR1, thereby reducing NMDA receptor activity and dampening pain signaling and amplification [79]. However, other potential mechanisms should not be overlooked. For instance, S1R antagonists have been shown to mitigate STZ-induced peripheral neuropathy by downregulating high-mobility group box 1 (HMGB1), a DNA-binding protein and a proinflammatory modulator for chronic pain [112, 123]. Further research is necessary to elucidate the full scope of pharmacological mechanisms underlying S1R antagonist efficacy.

Initially, S1R was mistakenly categorized as a subtype of opioid receptors, leading to the belief that its binding was off-target and undesirable. For example, haloperidol (Figure 2), a butyrophenone-based antipsychotics, demonstrates potent binding affinity for S1R (K_i = 2.3 nM) and has served as a reference S1R antagonist for over four decades. However, haloperidol preferentially binds to dopamine receptors with sub-nano molar binding affinity [103, 124]. Early selective S1R antagonists, such as NE-100 and BD-1047, also exhibit potent S1R binding affinity (K_i < 10 nM) and have been widely used as research probes [125, 126]. Following the reclassification of S1R as a distinct receptor in the 1990s and the advancements in S1R structural biology in 2010s, increasing interests have emerged in developing potent and selective S1R ligands for multiple diseases. Over the years, a variety of S1R antagonists with various chemical structures have been developed, including pyrrolidine-ones, substituted piperazines, benzothiazole-ones, spirocyclics, di-substituted pyrazoles and triazoles (Figure 2) [104, 127–132]. Several of these S1R antagonists (BD-1047, E-52862, PW507, CM304 and its radio-tracer [¹⁸F]FTC-146) have been extensively studied in preclinical and clinical pain research.