Research Progress on the Mechanisms of Central Post-Stroke Pain: A Review

Abstract

Central Post-Stroke Pain (CPSP) is a primary sequelae of stroke that can develop in the body part corresponding to the cerebrovascular lesion after stroke, most typically after ischemic stroke but also after hemorrhagic stroke. The pathogenesis of CPSP is currently unknown, and research into its mechanism is ongoing. To summarize current research on the CPSP mechanism and provide guidance for future studies. Use “central post-stroke pain,” “stroke AND thalamic pain,” “stroke AND neuropathic pain,” “post-stroke thalamic pain” as the search term. The search was conducted in the PubMed and China National Knowledge Infrastructure databases, summarizing and classifying the retrieved mechanism studies. The mechanistic studies on CPSP are extensive, and we categorized the included mechanistic studies and summarized them in terms of relevant pathway studies, relevant signals and receptors, relevant neural tissues, and described endoplasmic reticulum stress and other relevant studies, as well as summarized the mechanisms of acupuncture treatment. Studies have shown that the pathogenesis of CPSP involves the entire spinal-thalamo-cortical pathway and that multiple substances in the nervous system are involved in the formation and development of CPSP. Among them, the relevant receptors and signals are the hotspot of research, and the discovery and exploration of different receptors and signals have provided a wide range of therapeutic ideas for CPSP. As a very effective treatment, acupuncture is less studied regarding the analgesic mechanism of CPSP, and further experimental studies are still needed.

Graphical Abstract

Introduction

Stroke continues to be the third most common cause of disability and the second most common cause of death worldwide. The prevalence of stroke disease and the number of fatalities are both increasing globally (‘Global, regional, and national burden of stroke and its risk factors, 1990–2019: a systematic analysis for the Global Burden of Disease Study 2019’ 2021; Feigin et al. 2022). A prospective study found that 45.8% of post-stroke patients reported experiencing pain during a 6-month follow-up, indicating that pain symptoms are widespread among stroke survivors (Hansen et al. 2012). The influence of pain on a patient’s quality of life and subsequent rehabilitation is important (Naess et al. 2012; Westerlind et al. 2020). Central post-stroke pain (CPSP), complex regional pain syndrome (CRPS), shoulder pain, spasm-related pain, and headache are the most prevalent kinds of pain in people who have had strokes (Delpont et al. 2018). Because of its complicated symptoms, challenging diagnosis, and unclear mechanism, CPSP remains the focus of basic and clinical research.

Edinger proposed central post-stroke pain, now known as Dejerine–Roussy syndrome, in 1891, and it was later documented by French neurologists Jules Dejerine and Gustave Roussy in 1906: a thalamic lesion, slight hemiplegia, disturbance of superficial and deep sensibility, hemiataxia and hemiastereognosia, intolerable pain, and choreoathetoid movements (Andersen et al. 1995; Guédon et al. 2019). Following more than a century of research, it was shown that CPSP was not just related with thalamic lesions, but also with lesions occurring anywhere along the spinal-thalamic-cortical circuit (Kumar et al. 2009; Klit et al. 2009). Because CPSP is more difficult to distinguish from other types of neurological pain, R-D Treede et al. defined it as “pain arising as a direct consequence of a lesion or disease affecting the central somatosensory system” (Treede et al. 2008). Some authors define CPSP as a central neuropathic pain syndrome characterized by pain and sensory disturbances that can occur in body sites corresponding to cerebrovascular lesions after stroke, most commonly after ischemic stroke but also after hemorrhagic stroke, and excluding pain with an obvious cause, psychogenic pain, and peripheral neuropathic pain (Flaster et al. 2013; Vartiainen et al. 2016).

The clinical presentation of CPSP is not markedly different from that of other neuropathic pain disorders (Freeman et al. 2014; Jensen and Finnerup 2014). Pain sensations in CPSP sufferers are also not uniformly characterized and can manifest as pins and needles, squeezing, burning, electric shock, numbness, and so on (Oh and Seo 2015). Pain in CPSP patients can occur spontaneously or as a result of cold, heat, weariness, emotion, or other stimuli; however, spontaneous pain is more common, accounting for up to 85% of cases (Betancur et al. 2021). Aside from pain, the majority of CPSP patients experience sensory dullness, sensory hypersensitivity, and aberrant temperature perception (Oh and Seo 2015). The variety of CPSP symptoms is strongly tied to its molecular mechanisms, which are yet unknown, and research into its mechanism is ongoing. In this review, we synthesize existing studies in order to provide guidance for future research.

Literature Search and Organization

Use “central post-stroke pain,” “stroke AND thalamic pain,” “stroke AND neuropathic pain,” and “post-stroke thalamic pain,” as the search term. The search was conducted in the PubMed and China National Knowledge Infrastructure databases, and the search time limit was set to December 2022. A total of 1394 papers were found from the search, with 949 documents obtained after checking with EndNote X9 literature management software and 357 papers eventually obtained after reading the titles and abstracts to reject those that did not match. There were 84 review papers, 62 basic research papers, 155 clinical research papers, and 56 case reports in the final collected literature. The search process is described in Fig. 1.

Fig. 1.

Search process

Related Mechanisms Research

There is no unified consensus regarding the pathophysiology of CPSP, and the study of the relevant mechanisms spans a wide range of topics, such as neural pathways, nerve fibers, proteins, and cell receptors. The related hypotheses that have been put out include the inflammatory response, central sensitization theory, and de-inhibition theory. In order to offer theoretical backing for next study, this document compiles the most recent CPSP-related mechanism research.

Studies of the Relevant Pathways

TRN-VB Inhibitory Projections

Previous studies have found that most neurons in the spinal cord and thalamus are excited by pruritic and injurious stimuli (Davidson et al. 2012). The thalamic reticular nucleus (TRN) plays a crucial role in the dynamic regulation of sensory processing, the ventrobasal thalamus (VB) is the main projection target of the thalamus, and most VB neurons are excitatory, the main inhibitory input to the VB comes from the TRN (Dong et al. 2019).

Extracellular recordings were made in free-ranging mice by Peng-Fei Liu et al. to look at how neurons react to itchy and painful stimuli (Liu et al. 2022). They discovered that the majority of the neurons in the VB reacted to mechanical stimuli that caused pain and scratching as a result of pruritus. Inhibiting VB neurons decreased scratching behavior and relieved pathological pain in mice, but activating VB excitatory neurons increased pruritogen-induced scratching behavior. The experiments also revealed that the ventral TRN predominantly innervates the VB and there are monosynaptic GABAergic connections between TRN neurons and VB neurons. Moreover, TRN-VB inhibitory projections are involved in the processing of itch and pain signals, and optogenetic activation of TRN-VB inhibitory projections inhibits scratching behavior, whereas removal of TRN enhances scratching behavior. Activation of TRN-VB inhibitory projections relieves pathological pain without affecting basal nociceptive thresholds. These results suggest that VB neuronal activity correlates with both scratching and pain-related behaviors. Manipulation of VB or TRN-VB inhibitory projections can modulate both pruritus and pain.

MED1/BDNF/TrkB Pathway

Neuropathic pain, including CPSP, is frequently connected with emotional problems, such as depression. Neuropathic pain-induced depression behavior is associated with the dysregulation of Brain-Derived Neurotrophic Factor (BDNF) and its tropomyosin receptor kinase B (TrkB) signaling in brain regions, including the hippocampus (Boccella et al. 2019). Mediator Complex Subunit 1 (MED1), can regulate the expression ability of BDNF (Nagpal et al. 2018). Rosmara Infantino et al. were the first to characterize the role of the BDNF/TrkB pathway in stroke-induced pathogenic alterations, as well as its possible role in neuroplastic changes (Infantino et al. 2022). Compared to the sham-operated group, rats in the thalamic hemorrhage (TH) model group showed a significant increase in the expression of MED1 and TrkB in the hippocampal region and a significant decrease in the expression of BDNF. Also, Rosmara Infantino et al. found significant overexpression of MED1 in human post-mortem specimens after stroke.

Glutamatergic Neuronal Circuits

Previous experiments have demonstrated that many brain regions, such as the thalamus, amygdala, prefrontal cortex (PFC), and anterior cingulate cortex (ACC), are involved in the central regulation of chronic pain (Saab 2012). The thalamus serves as a gateway to the cerebral cortex, and thalamocortical components are required for pain expression. However, it is unknown which independent thalamocortical circuits determine pain symptoms in individual tissue injury or depressive states. Xia Zhu et al. found in mouse experiments that excitation of primary somatosensory cortex glutamatergic neurons (S1^Glu) neurons were increased by posterior thalamic nucleus (PO^Glu) under tissue damage conditions, whereas the inhibition of the anterior cingulate cortex GABA-containing neurons to glutamatergic neurons (ACC^GABA→Glu) by the parafascicular thalamic nucleus (PF^Glu) are reduced under depression-like conditions and initiate pain behavior (Zhu et al. 2021). This suggests that PO^Glu projections to S1^Glu mediate nociceptive hypersensitivity due to tissue damage. Meanwhile, the pathway from the PF^Glu to the ACC^GABA→Glu mediates nociceptive abnormalities associated with depression. This experiment demonstrates that different subregions of the thalamic nuclei are applied to alleviate nociceptive abnormalities caused by different pathological conditions.

According to the studies mentioned above, there are multiple neurological pathways that can cause central post-stroke pain. It could affect the thalamus area, hippocampus region, frontal cortex, and anterior cingulate cortex, among other brain regions. Excitatory and inhibitory effects within particular brain regions, as well as the transmission of associated neurotransmitters, can enhance or attenuate pathological pain; additionally, neural connections between brain regions can mediate various nociceptive abnormalities and nociceptive hyperalgesia. The corresponding summary is shown in Table 1.

Table 1.

Summary of CPSP-related mechanisms

Related mechanisms research		Relevant mechanisms or key findings
Relevant pathways	TRN-VB inhibitory projections	Manipulation of VB or TRN-VB inhibitory projections can modulate both pruritus and pain
	MED1/BDNF/TrkB pathway	The expression of MED1 and TrkB was significantly increased, while the expression of BDNF was significantly decreased
	Glutamatergic neuronal circuits	POGlu → S1Glu mediate nociceptive hypersensitivity due to tissue damage; PFGlu → ACCGABA→Glu mediates nociceptive abnormalities associated with depression
Studies of the signals and receptors	SDF1-CXCR4 signaling	The increase of SDF1-CXCR4 signaling contributes to the neuroinflammatory response and the maintenance of CPSP
	P2X7 Receptor	The activation of the P2X7 receptor enhances the production of inflammatory cytokines and the associated neuronal damage
	P2X4 Receptors	Promotes inflammatory responses, mediates downstream pathways, and participates in the sensitization mechanism of central post-stroke pain
	BDNF	BDNF upregulation following thalamic damage is a major factor in CPSP
	PUFA/GPR40 Signaling	Mechanical nociceptive sensitization may be regulated by astrocyte activation and stimulation of GPR40 signaling
	GABA	Reduced GABAergic transmission in VB promotes thermal nociceptive hypersensitivity in chronic inflammatory pain
	NLRP3 inflammasome	a. Activation of NLRP3 leads to a decrease in GABAergic release and ultimately to CPSP b. Enhanced inflammatory response of microglia, contributing to CPSP
	Nitric oxide synthase (NOS)	Early mechanical allodynia in CPSP can be induced by enhanced NOS activity in the spinal cord due to DDAH1 overexpression
	LPA	Pain signaling leads to LPA production, while promoting microglia activation and IL-1β production and further promotes LPA production through LPA1 and LPA3 signaling
Studies of the relevant neural tissue	Microglia	Promotes neuroinflammation and maladaptive neural reorganization
	Primary afferent neurons A and C fibers	The sensitivity of Aδ and Aβ fibers was significantly increased in the presence of middle cerebral artery obstruction. The sensitivity of C and Aβ fibers was increased in bilateral carotid artery occlusion
	Neurosteroids	Upregulation of neurosteroids has a protective effect on pain

Studies of the Signals and Receptors

SDF1-CXCR4 Signaling

Previous studies suggest that SDF1-CXCR4 signaling in the hemorrhagic area is involved in the maintenance of nociceptive hyperalgesia in thalamic hemorrhagic stroke (THS) by regulating thalamic neuroinflammation (Yang et al. 2017a, b). Ting Liang et al. studied the degeneration and loss of spinal dorsal horn neurons after intra-thalamic collagenase (ITC) injection and discovered that spinal thalamic tract (STT) neurons sustained secondary apoptosis and death via retrograde axonal degeneration after primary THS (Liang et al. 2022). This degenerative deterioration is followed by secondary neuroinflammation, as evidenced by activation of microglia and astrocytes in the dorsal horn of the spinal cord and elevation of SDF1-CXCR4 signaling. This study demonstrates that subsequent spinal cord damage and neuroinflammation, in addition to primary thalamic injury, play a key role in maintaining pain hypersensitivity following THS. Through axonal degenerative processes, primary thalamic nerve injury can cause subsequent neuronal death and neuroinflammation in the dorsal horn of the spinal cord. Secondary spinal neuritis and pain hypersensitivity following THS are regulated by SDF1-CXCR4 signaling. Ting Liang et al. showed that the dorsal horn of the spinal cord may be an important site for clinical treatment of CPSP.

Fei Yang et al. found through experiments that both intra-thalamic collagenase (ITC) and SDF1 injections can induce CPSP and then the process can be blocked and reversed by intra-thalamic injections of AMD-3100 (a selective CXCR4 antagonist) and inhibitors of microglia or astrocyte activation(Yang et al. 2017a, b). Moreover, the long-term increased expression of SDF1 and CXCR4 that accompanies microglia and astrocyte activation after ITC can be blocked by AMD-3100 and YC-1 (selective inhibitors of HIF-1α). SDF1 and its receptor CXCR4 appear to play an important role in the establishment and maintenance of thalamic hemorrhagic CPSP via hypoxia-inducible factor 1α (HIF-1α)-mediated microglia–astrocyte neuronal connections. HIF-1α is a transcription factor that is linked to the expression of SDF1 and CXCR4 in hypoxic conditions. ITC-induced thalamic hemorrhage may maintain CPSP by upregulating SDF1-CXCR4 signaling via HIF-1α, followed by positive feedback regulating the neuroinflammatory microenvironment induced by microglia–astrocyte–neuron interactions.

P₂X₇ Receptor

Extracellular nucleotide levels in the central nervous system (CNS) may rise, activating numerous cell surface purinoceptors. Many of these activities are controlled by activated P₂X₇ purinoceptor subtypes. P₂X₇ receptors are widely thought to play a role in pro-inflammatory actions in the CNS (Monif et al. 2016). The inhibition of P₂X₇ receptors has been shown to diminish microglia activation and inflammation. P₂X₇ receptor activation is assumed to be mediated by the production of interleukin-1β (IL-1β) and adenosine triphosphate (ATP) from microglia or macrophages and is linked to the harmful transmission and chronic neuropathic pain (Englezou et al. 2015).

Yung-Hui Kuan et al. created a rat model of thalamic hemorrhage and discovered a significant increase in P₂X₇ expression in reactive microglia in the tissue surrounding the thalamic lesion 5 weeks after the hemorrhage (Kuan et al. 2015). Thalamic P₂X₇ receptors are directly involved in pain transmission and hypersensitivity responses. Furthermore, in CPSP rats, early treatment with the P₂X₇ receptor antagonist prevented increased neuronal excitability, reduced microglia and macrophage aggregation, and hindered abnormal pain. The findings support the hypothesis that persistent CPSP is caused by P₂X₇ receptor activation following brain tissue injury and subsequent elevation of inflammatory cytokines. Traumatized cells at the site of lateral thalamic lesions produce substantial amounts of intracellular ATP and increase IL-1β secretion by reactive microglia into adjacent tissues, including the synaptic gap, according to Yung-Hui Kuan et al. ATP and IL-1β then increase glutamate release, increasing the frequency of neuronal bursts in the thalamus-glutamate pathway. P₂X₇ receptor antagonists inhibit the activation of P₂X₇ receptors in glutamatergic nerve endings and microglia and then inhibit the abnormal overproduction of IL-1β, which effectively inhibits the overexcitation of medial thalamus (MT) and ACC neurons in response to injurious stimuli and regulates CPSP to a normal state.

Another study, conducted by Yung-Hui Kuan et al., discovered that P2X7 receptor expression was elevated at sites near CPSP brain injury and was linked to CNS immunoreactive cells (i.e., reactive microglia) (Kuan et al. 2018). The study suggests that inhibiting or blocking neuronal hyperexcitability and reversing abnormal oscillations may have anti-injury effects by targeting P2X7 receptors. Furthermore, early P2X7 receptor antagonist treatment of stroke patients may prevent activation of microglial P2X7 receptors in the peri-lesional tissue, reducing the release of local inflammatory cytokines and associated neuronal damage.

P₂X₄ Receptor

P₂X₄ receptor is mainly expressed in microglia and belongs to the purinergic P₂ receptor family, which has been widely studied in neuropathic pain (Inoue 2019). P₂X₄ receptor are ligand-gated Ca²⁺ channels, Ca²⁺ can be transmitted to various of intracellular effectors, mediate the downstream signaling pathways, and lead to the expression of pro-inflammatory mediators in the inflammatory response (Layhadi and Fountain 2017).

Tumor necrosis factor-α (TNF-α) is one of the cytokines recognized to be involved in the pathophysiology of neuropathic pain. TNF-α acts on neuronal TNFR1 receptors to increase excitatory synaptic strength and, in turn, induces endocytosis of gamma-aminobutyric acid a receptor (GABAaR) to decrease inhibitory synaptic strength, disrupting the balance of the central nervous system’s facilitatory and inhibitory systems (Pribiag and Stellwagen 2013). Jiajie Lu et al. found that elevated TNF-α levels and GABAaR endocytosis in CPSP could be inhibited by blocking the P₂X₄ receptor (Lu et al. 2021). In addition, antagonizing TNF-α increases cell surface GABAaR expression and mechanical pain threshold, and down-regulation of TNFR1 reverses cell surface GABAaR endocytosis and attenuates mechanical hypersensitivity. Thus, neuropathic pain is mediated in part through post-stroke-induced P₂X₄R/TNF-α/TNFR1/GABAaR signaling. This signaling pathway may provide a new therapeutic strategy for CPSP.

Hai-Feng Lu et al. established a novel model of hemorrhagic stroke in rats; this model is based on a hemorrhagic stroke lesion with intra-thalamic autologous blood (ITAB) injection in the ventral posterolateral nucleus of the thalamus (Lu et al. 2018). In this model, there was a significant increase in P₂X₄ receptor expression in the microglia surrounding the post-hemorrhagic thalamic lesion. P₂X₄ receptor blockade alleviated mechanical abnormal pain in CPSP rats. Mechanical abnormal pain was significantly reduced after antidepressants and antiepileptics were administered. P₂X₄ receptor expression was also found to be significantly reduced after treatment with these drugs. These findings also imply that P₂X₄ receptor are involved in the mechanism of pain sensitization following a central stroke.

BDNF

Previous research has demonstrated that brain-derived neurotrophic factor (BDNF) is an important neurotrophic factor released by astrocytes or microglia and that BDNF overexpression in proliferating astrocytes or microglia following damage is a crucial factor linked with neuropathic pain (Boakye et al. 2021).

Yung-Hui Kuan et al. discovered that CPSP rats had higher levels of BDNF at peri-lesional locations. BDNF receptor blockers might erase aberrant pain and nociceptive hyperalgesia in CPSP rats, inhibit heightened noxious responses in the MT-ACC pathway, and inhibit enhanced spontaneous MT activity (Kuan et al. 2018). It was also discovered that BDNF mediated the thalamic injurious response to CPSP, BDNF mRNA was elevated in MT following CPSP, and that rapid injection of BDNF receptor blockers decreased the MT injurious response.

Hsi-Chien Shih et al. established a CPSP rat model and discovered that mechanical and thermal aberrant pain was caused following thalamic ventral basal complex injury in rats (Shih et al. 2017). After 4 weeks, the number of neurons in the brain damage area reduced, astrocytes, microglia, and P₂X₄ receptor increased, and BDNF gene expression rose. These findings imply that BDNF upregulation following thalamic damage is a major factor in CPSP.

PUFA/GPR40 Signaling

Polyunsaturated fatty acid (PUFA) such as docosahexaenoic acid (DHA) and eicosapentaenoic acid (EPA) are abundant in brain tissue and have neuroprotective qualities during ischemia and brain injury. Furthermore, astrocytes play a key role in PUFA production and release PUFA to neuronal populations. PUFA can activate GPR40 which is a free fatty acid receptor 1 (FFA1). GPR40 has anti-injurious effects and pain reduction at CPSP via the release of β-endorphins (Figueroa et al. 2013; Eady et al. 2014).

Shinichi Harada et al. investigated the role of GPR40 in CPSP by developing a bilateral carotid artery occlusion (BCAO) mouse model, as well as the role of astrocytes (Harada et al. 2014). The mechanical nociceptive sensitization threshold in the hind paw of mice was found to be significantly lower after the BCAO test compared to before the BCAO test. However, DHA or GW9508 (a GPR40 agonist) injections via the lateral ventricle significantly reduced BCAO-induced mechanical nociceptive hypersensitivity; these effects were also reversed by the GW1100 (a GPR40 antagonist). The above studies suggest that BCAO-induced mechanical nociceptive sensitization may be regulated by astrocyte activation and stimulation of GPR40 signaling. Activation of GPR40 could be a potential therapeutic target for the control of pain symptoms in CPSP.

GABA

The VB includes the ventral posterolateral (VPL) and the ventral posteromedial (VPM), which transmit nociceptive information to the cerebral cortex. Both the VPM and VPL are regulated by GABAergic regulation from the thalamic reticular nucleus (TRN). GABA released from presynaptic vesicles binds to synaptic GABAaRs as a neurotransmitter and mediates transient or phasic inhibition; whereas, ambient GABA binds to extrasynaptic GABAaRs to mediate persistent or tonic inhibition (Zheng et al. 2014).

Chan Zhang et al. reported in their study the reduction of GABA levels in VB of rats with chronic inflammatory pain induced by complete Freund’s adjuvant (CFA) (Zhang et al. 2017). Microinjection of GABAaR agonist and optogenetic activation of TRN-VB pathway alleviated thermal nociceptive hypersensitivity in chronic inflammatory pain. In contrast, microinjection of extrasynaptic GABAaR agonists or selective knockdown of synaptic GABAaR subunits exacerbates thermal nociceptive sensitization in chronic inflammatory pain. These results confirm the protective effect of synaptic GABAergic transmission on thermal nociceptive hypersensitivity in chronic inflammatory pain. Reduced GABAergic transmission in VB promotes thermal nociceptive hypersensitivity in chronic inflammatory pain.

NLRP3 Inflammasome

The inflammasome is a complex composed of multiple proteins, which regulate the body’s innate immune system and sense microorganisms, metabolites, and stress responses. NLR pyrin domain-containing 3 (NLRP3) is a member of the NLR inflammasome family, and large amounts of endogenous and exogenous substances can lead to oligomerization of NLRP3 inflammasome, thereby stimulating the secretion and maturation of pro-inflammatory cytokines (IL-18 and IL-1β), which are involved in regulating the inflammatory response of the body (Sutterwala et al. 2014). MicroRNAs (miRNAs/miRs) are endogenous small noncoding RNAs with a size of ~ 22 nucleotides. miRNAs serve important roles in essential biological activities, including cell proliferation, differentiation, and apoptosis (Shukla et al. 2011).

Previous trials have demonstrated that miR-223 can suppress inflammation and prevent collateral damage. miR-223 has been shown to negatively regulate NLRP3 during the development of inflammatory responses (Haneklaus et al. 2012). Huang, T. et al. found that mir-223 expression levels were significantly reduced in the CPSP model (Huang et al. 2022). NLRP3, IL-18, and IL-1β expression levels were significantly increased. miR-223 agonist significantly reduced thalamic pain and significantly decreased NLRP3, IL-18, and IL-1β levels. In addition, the introduction of miR-223 antagonist into the VPL nucleus of mice could mimic thalamic pain and significantly increase the levels of NLRP3, IL-18, and IL-1β. These results suggest that miR-223 can inhibit NLRP3 inflammasome activity and ameliorate CPSP in mice due to thalamic hemorrhage by downregulating NLRP3 expression and significantly attenuate pain sensitization responses, such as cold hypersensitivity, mechanical hypersensitivity, and thermal hyperalgesia.

There are two putative theories for the participation of NLRP3 inflammasome in CPSP: one is that NLRP3 activation causes cortical damage while decreasing thalamic downstream projection fibers. This condition may cause a decrease in GABAergic release, increasing the excitability of VB neurons. This finally results in the emergence of CPSP. Another explanation is that NLRP3 inflammasome boost microglia’s inflammatory response while injuring the thalamus and prolonged inflammation induces GABAergic changes in TRN, reducing VB interneuron function and leading to CPSP (Li et al. 2018). Studies have shown that inhibition of NLRP3 inflammatory vesicles using MCC950 (a NLRP3 inflammasome inhibitor) has therapeutic potential in ischemic stroke models. Therefore, targeted inhibition of NLRP3 activation may be a promising therapeutic approach for CPSP (Ismael et al. 2018).

Nitric Oxide Synthase (NOS)

Previous experiments have found that cerebral infarction can lead to secondary degeneration in non-ischemic regions, such as evidence that nitric oxide synthase (NOS) appears upregulated in the spinal cord after stroke in mice (Matsuura et al. 2018). NOS activity is influenced by a variety of factors, such as ischemic stress or injurious stimuli. Dimethylarginine dimethylaminohydralase 1 (DDAH1) is a methylated arginine-degrading enzyme that is widely present in a variety of central and peripheral tissues and is involved in nitric oxide synthase (NOS) activity regulation.

Wataru Matsuura et al. discovered a significant increase in mechanically triggered withdrawal responses in mice on days 1 and 3 following bilateral carotid artery occlusion (BCAO) surgery (Matsuura et al. 2019). DDAH1 protein expression was increased on day 1 but not on day 3 following BCAO. Injection of PD404182 (the DDAH1 inhibitor) significantly decreased the mechanical hypersensitivity reaction on day 1 following BCAO, but not on day 3. Furthermore, on days 1 and 3 after BCAO, intravenous treatment of L-NAME (the NOS inhibitor) dramatically reduced mechanical allodynia and inhibition of DDAH1 reduced mechanical nociceptive hyperalgesia and NOS activity. The findings show that early mechanical allodynia in CPSP can be induced by enhanced NOS activity in the spinal cord due to DDAH1 overexpression.

LPA

Previous experiments have demonstrated that lysophosphatidic acid receptors (LPA) play a key role in the molecular mechanism of neuropathic pain development and that strong pain signals lead to the production of LPA in the dorsal horn of the spinal cord, which in turn promotes microglia activation and IL-1β production and further promotes LPA production through LPA1 and LPA3 signaling (Ueda and Neyama 2017).

Hiroshi Ueda et al. used photochemically induced thrombosis (PIT) to create a new model of central neuropathic CPSP (Ueda et al. 2019). In addition, they discovered that electrical stimulation hypersensitivity, as well as thermal and mechanical nociceptive hypersensitivity, disappeared in mice lacking LPA1 and LPA3. Both thermal and mechanical pain sensitization were significantly reversed when LPA1 and LPA3 antagonists were administered sequentially. Several LPA molecules were significantly increased in somatosensory S-I and medial dorsal thalamus (MD), but not in striatum and ventroposterior thalamus. These results suggest that LPA1 and LPA3 signaling play a key role in the occurrence and maintenance of CPSP.

By aggregating research on related signaling and receptors, we discovered that the majority of them are active or overexpressed during the inflammatory response. (e.g., SDF1-CXCR4 signaling, P2X4 Receptor, P2X7 Receptor, GABA, NLRP3 inflammasome). This disrupts the nervous system’s natural equilibrium, which adds to aberrant modifications in downstream pathways and, ultimately, generates numerous nociceptive disorders. We also discovered that microglia and astrocytes express or mediate practically all signaling and receptors. However, whatever signals or receptor has a major role in the development of central post-stroke pain is an issue that needs to be investigated more in future. The corresponding summary is shown in Table 1.

A diagram of the above-related molecular mechanisms is shown in Fig. 2.

Fig. 2.

Schematic diagram of the molecular mechanism

Studies of the Relevant Neural Tissue

Microglia

Previous experiments have shown that microglia activation is involved in the pathophysiological processes of CPSP. The role of microglia in the altered thalamocortical plasticity associated with TH has been described (Hiraga et al. 2020). Microglia, as immune cells of the CNS, participate in early defensive processes following injury or disease. Microglia are hypothesized to contribute to long-term sensory impairment in neuropathic pain disorders by promoting neuroinflammation and maladaptive neuronal remodeling (Guida et al. 2020).

N-palmitoylethanolamide and luteolin (PEALut) is a neuroprotective substance that shows immunomodulatory and anti-inflammatory effects by reducing microglia activation. Rosmara Infantino et al. observed in a rat model of TH that repeated administration of PEALut significantly reduced the mechanical hypersensitivity response after TH due to its reduction of early microglia activation at the peri-lesion site (Infantino et al. 2022). This demonstrates that microglia can promote the development of CPSP. Shin-ichiro Hiraga et al. discovered microglia activation in the damaged primary somatosensory cortex and damaged thalamus using a mouse model of focal TH (Hiraga et al. 2020). The effect of pharmacological ablation of microglia on TH was also studied, and it was discovered that the treatment totally prevented the development of TH-induced nociceptive abnormalities.

Primary Afferent Neurons A and C Fibers

Kazunori Takami et al. focused on changes in the sensitivity of primary sensory neurons (including A and C fibers) and changes in pain thresholds in response to mechanical stimuli after stroke (Takami et al. 2011). By constructing a CPSP model of left middle cerebral artery obstruction (MCAO) mice, it was found that on day 3 after MCAO, current stimulation thresholds measured at 5 Hz (i.e., C-fiber stimulation) in the left hind paw and right hind paw of the MCAO group were similar to those before MCAO, whereas current stimulation thresholds at 250 Hz (i.e., Aδ-fiber stimulation) and 2000 Hz (i.e., Aβ-fiber stimulation) were significantly lower compared to pre-MCAO. It was suggested that the sensitivity of Aδ and Aβ fibers was significantly increased, while the sensitivity of C-fiber was not changed. The left hind paw retraction threshold (PWT) was significantly lower at day 1 and day 3 after MCAO compared to day 0, whereas the difference in the right hind PTW to day 0 was not statistically significant. It is inferred that myelin A fiber-specific hypersensitivity reaction after stroke may be a cause of the above symptoms. It is also known that mechanical ectopic pain occurs ipsilateral to the middle cerebral artery obstruction.

Shigeyuki Tamiya et al. investigated the development of mechanical and thermal nociceptive hypersensitivity responses, as well as alterations in current stimulation thresholds, in primary afferent neurons from mice with BCAO (Tamiya et al. 2013). Compared to pre-BCAO, the sensitivity of C and Aβ fibers (at 5-Hz and 2000-Hz stimulation, respectively) was increased on day 3 after BCAO, while the sensitivity of Aδ fibers (at 250-Hz stimulation) was unchanged. This concludes that the mechanical and thermal hypersensitivity responses of the two hind paws in the BCAO model may involve alterations in C and Aβ fibers. The BCAO model, in contrast to the focal cerebral ischemia model, showed distinct changes in temperature hypersensitivity and primary afferent neurons. These findings imply that the mechanism of pain varies depending on the location and extent of cerebral ischemia.

Neurosteroids

Researchers discovered steroids in the nervous system in the 1980s and detected a series of enzymes for steroid production in the nervous system (Lloyd-Evans and Waller-Evans 2020). Fluctuations in neurosteroid concentrations modulate various physiological responses, including anxiety and stress. Neurosteroids can act through steroid hormone nuclear receptors and GABAaRs, among others, while key enzymes in steroidogenesis, as well as synaptic and extrasynaptic GABAaRs, are widely expressed in the thalamus (Zorumski et al. 2013; Waldvogel et al. 2017). Translocator protein (TSPO) mediates steroid synthesis and is widely present in steroid-synthesizing tissues, including the brain.

Meng Zhang et al. investigated the levels of neurosteroids and the steroidogenic enzyme TSPO in the lateral thalamus of normal and neuropathic pain rats, as well as their role in pain regulation (Zhang et al. 2016). It was discovered that during the chronic phase of neuropathic pain, neurosteroid levels in the lateral thalamic area increased, as did TSPO expression in thalamic neurons. TSPO activation increased neurosteroid hormone levels and decreased mechanical nociceptive hyperalgesia, whereas TSPO inhibition had the opposite effect. GABAaRs are also involved in neurosteroids’ analgesic effects. According to the findings, increased TSPO expression in the chronic phase of neuropathic pain leads to the overexpression of a series of neurosteroids in the lateral thalamic nucleus, which may enhance the inhibitory activity of GABAaRs and have a pain-reducing impact.

From the above findings, it is clear that in addition to the activation of microglia, which play an immune role, the hypersensitivity of nerve fibers and the upregulation of neurosteroid levels are closely related to the production of nociceptive abnormalities in the development and progression of central post-stroke pain. The corresponding summary is shown in Table 1.

The mechanisms of the above-mentioned relevant neural tissues are shown in Fig. 3.

Fig. 3.

Schematic diagram of the relevant nerve tissue

Endoplasmic Reticulum Stress

The pathological state of stroke can alter the protein folding process, leading to the accumulation of unfolded proteins in the endoplasmic reticulum (ER) and triggering a condition called endoplasmic reticulum stress. Intense or sustained endoplasmic reticulum stress can lead to chronic inflammation, apoptosis, and autophagy. Enhanced excitatory neurotransmission and neuronal de-inhibition induced by endoplasmic reticulum stress can lead to central sensitization (Maier et al. 2014; Chanaday et al. 2021). In this case, the unfolded protein response (UPR) is activated to re-establish endoplasmic reticulum homeostasis, and persistent UPR is also an inflammatory lesion that can elicit a defensive innate immune response. Previous experiments demonstrated that cells under severe endoplasmic reticulum stress can induce UPR-dependent inflammatory pathways, which further exacerbates the innate inflammation caused by primitive damage (Sprenkle et al. 2017).

Epoxyeicosatrienoic acids (EETs) are widely distributed in the brain and are important regulators of cerebral blood flow regulation, axonal growth, and neuronal survival. Endogenous EET is rapidly degraded by soluble epoxide hydrolase (sEH) within a few seconds. This short in vivo half-life limits the use of EETs in clinical applications. However, the development of sEH inhibitors can stabilize the levels of EETs and thus extend their half-lives to improve the biological activity of EETs (Pillarisetti and Khanna 2015). Epoxyeicosatrienoic acids (EETs)/soluble epoxide hydrolase inhibitors (sEHi) are considered to be emerging targets that play an important role in the regulation of pain and neuroinflammation.

Tongtong Liu et al. experimentally found endoplasmic reticulum stress at the peri-thalamic injury site in CPSP rats, with swollen endoplasmic reticulum luminae in most neurons at the thalamic injury site (Liu et al. 2021). The expression of major endoplasmic reticulum stress receptors and their downstream targets were upregulated from day 7 after CPSP induction and persisted for at least 1 month, while the level of sEH was elevated. Subsequent inhibition of sHE attenuated endoplasmic reticulum stress and peri-thalamic neuroinflammation in CPSP rats while attenuating CPSP-induced mechanical nociceptive hyperalgesia. The present study provides evidence that endoplasmic reticulum stress around the stroke site may activate glial cells and lead to further inflammatory responses, which constitute a vicious cycle that disrupts the self-repair system of the central nervous system and ultimately produces central sensitization and persistent pain. In contrast, drugs targeting EET signaling have great potential to treat CPSP by inhibiting excessive endoplasmic reticulum stress and neuroinflammatory responses, as well as preserving the normal thalamic inhibitory state.

Another Study

Previous experiments confirmed that LPA1 receptor signaling initiates peripheral nerve injury-induced neuropathic pain by altering the expression of pain-related genes/proteins. Halder, S. K. et al. established an animal model of mild cerebral ischemia for pain-related experiments (Halder et al. 2013). The LPA1 receptor signaling pathway was found to play a crucial role in the development of neuropathic pain induced by cerebral ischemia through stimulation of primary afferent Aδ and Aβ fibers.

Hyperpolarization-activated cyclic nucleotide-gated (HCN) channels are widely expressed in peripheral sensory neurons and in the central neuronal system. HCN channel activity plays an important role in behavioral and physiological processes. Misregulation of HCN channel activity has been demonstrated to cause neurological and psychological disorders, including pain, epilepsy, and anxiety (Du et al. 2013). Previous experiments have demonstrated that HCN activity in the upstream injurious pathway may be an important factor in chronic pain conditions. Weihua Ding et al. observed the effects of injecting ZD7288 (the HCN channel blocker) into the VPL on injurious behavior in rats with neuropathic pain or simple arthritis (Ding et al. 2016). ZD7288 was found to reduce mechanical nociceptive hyperalgesia and thermal nociceptive hyperalgesia in rats with chronic pain. In the thalamus, immunoreactivity of HCN1 and HCN2 subunits was enhanced in both models of rats. These results suggest that increased HCN channel activity in the superior injurious pathway of the thalamus is associated with chronic neuropathic pain and inflammatory pain.

Melatonin is a neurohormone secreted by the pineal gland and extra-pineal tissues that controls various physiological phenomena, such as circadian rhythms and emotional behavior. Melatonin also plays a key role in pain regulation. Melatonin therapy has been shown to be effective in fibromyalgia, migraine, and irritable bowel syndrome (Danilov and Kurganova 2016). A study by Tavleen Kaur et al. found a dramatic decrease in endogenous melatonin levels after thalamic injury in rats (Kaur et al. 2022). Since the thalamus has a strong non-photoregulatory influence on circadian rhythmicity, pineal melatonin production, and secretion (Mishima 2012), researchers believe that the phenomenon is due to damage to neurons in the thalamus, which prevents the production of endogenous melatonin in the pineal gland, leading to sudden changes in rest, sleep, and activity behavior in animals. Serum melatonin levels in CPSP rats were decreased but significantly restored after 3 weeks of exogenous melatonin therapy. Exogenous melatonin may substitute for endogenous melatonin production, and exogenous melatonin may assist improve not just activity behavior but also pain associated with CPSP and sleep cycle disturbance. Melatonin could thus be used as an alternative medication to address pain and sleep disturbances in stroke patients.

Acupuncture Treatment Mechanism

There are various pharmacological treatments for CPSP, which can be designed according to the neural pathways, signals, and receptors involved in the study of CPSP-related mechanisms. However, the cost of drug treatment is relatively high, and there is a great deal of uncertainty about the effectiveness of treatment. Therefore, non-pharmacological therapies are a very important option for CPSP patients, especially for refractory CPSP. In a recent systematic review, the safety and efficacy of different non-pharmacological interventions in the treatment of CPSP were evaluated, including 11 studies (Xu et al. 2020). The results showed that non-pharmacological therapies such as acupuncture, electroacupuncture, transcranial direct current stimulation, and acupoint injections all improved pain symptoms in patients with CPSP.

As a traditional Chinese therapy, acupuncture has been clinically proven to be effective in treating the pain symptoms of CPSP patients (Li et al. 2017; Sokal et al. 2015). A comparison of the efficacy and safety of acupuncture and western medicine in the treatment of post-stroke thalamic pain revealed that acupuncture was more effective than western medicine (Yang et al. 2022). Studies have shown that acupuncture analgesia mainly affects the neuroexcitability of the pain source, which in turn affects the release of the corresponding neurotransmitters, thus achieving the effect of treating central nervous system pain (Zhang et al. 2014). Furthermore, as a non-pharmacological therapy, acupuncture has low side effects, low cost, quick effect, and simple operation, making it well worth promoting. However, there have not been many research on the mechanism of acupuncture for the treatment of CPSP, therefore we compiled a list of the ones that have been done.

The cause of neuronal death in neuropathic pain is not only apoptosis but also autophagy. Autophagy is intrinsically active in the CNS and can prevent neuronal damage and neurodegeneration after stroke (Tan et al. 2014). When brain tissue is damaged, autophagy can degrade damaged organelles, eliminate damaged components and toxic metabolites, or provide a source of nutrients needed to maintain metabolism, ATP levels, dynamic cellular homeostasis, and survival (Nah et al. 2015). One of the markers of autophagy is the shift from the soluble form of LC3 (LC_3-I) to the autophagic vesicle-associated form (LC_3-II), and another marker of autophagy, p62, is increased after autophagy occurs.

Studies have shown that Electroacupuncture (EA) can provide analgesia by downregulating COX-2 expression in the spinal cord of rats with neuropathic pain (Ji et al. 2017). It was also found that EA could improve tolerance to cerebral ischemia/reperfusion (I/R) in rats by inhibiting autophagy, and it was demonstrated that EA treatment reduced the number of autophagosomes while decreasing the ratio of LC3B-II/LC3B-I in the peri-infarct cortex (Liu et al. 2016).

Gui-Hua Tian et al. (Tian et al. 2016) assessed the expression of pain-related behavioral responses, neuronal apoptosis, glial cell activation, and pain signaling-related factors (β-catenin, COX-2, and NK-1R) by constructing a CPSP mouse model. The electroacupuncture stimulation of ‟Baihui” and ‟ZuSanli” revealed that low-frequency EA (2 Hz) significantly reduced the size of brain tissue damage and hematoma and inhibited neuronal apoptosis, thus providing analgesic effects. Meanwhile, high-frequency EA (15 Hz) treatment had a strong inhibitory effect on the activation of abnormal astrocytes, and the expression of pain signaling-related factors was downregulated, thus reducing inflammation and producing a strong analgesic effect. The relative apoptotic rates in both the neocortex and hippocampus of CPSP rats were significantly reduced after electroacupuncture stimulation, while the best effect was obtained with low-frequency EA. This study confirmed that EA treatment was effective in relieving CPSP and that the neocortex and hippocampus of the brain were significantly associated with pain production. Low-frequency EA and high-frequency EA treatments exert analgesic effects by inhibiting neuronal apoptosis and abnormal astrocyte activation in the brain.

Ling Zheng et al. (Zheng et al. 2020) showed that the LC3B-II/I ratio was significantly higher and P62 was significantly lower when brain tissue was damaged. Therefore, it can be hypothesized that CPSP can induce damage and cause autophagy. Moreover, the addition of COX-2 and β-catenin inhibitors in the inflammation model can inhibit the high expression of autophagy-related proteins in hippocampal primary brain cells. This suggests a reciprocal regulatory relationship between β-catenin, COX-2, and LC3. It was also found that EA downregulated the expression of COX-2 and β-catenin in the damaged hippocampus while downregulating the expression of autophagy marker protein (LC3B-II/I, p62). This experiment demonstrates that EA may suppress autophagy in the hippocampus by lowering β-catenin/COX-2 protein production, hence decreasing CPSP (see Fig. 4).

Fig. 4.

Acupuncture treatment mechanism diagram

Discussion

In this review, we summarize the research on the mechanisms associated with central pain after stroke. According to the above, the site of injury in CPSP is not limited to the thalamus; injury at any site in the spinal cord-thalamus-cortex neural pathway can cause sensory abnormalities and nociceptive hyperalgesia. In the study of the mechanism of CPSP, the construction of animal models is very important. The most common way to construct a CPSP model is to induce thalamic hemorrhage by intra-thalamic injection of collagenase (ITC) and experiments have shown that this model can well trigger mechanical nociceptive hypersensitivity and warm nociceptive hypersensitivity in animals. The bilateral carotid artery occlusion (BCAO) mouse model and the middle cerebral artery obstruction (MCAO) mouse model are frequently used in ischemic animal model experiments.

This article briefly classifies current mechanistic studies, including relevant neural pathways, relevant signals and receptors, and relevant neural tissues, and also describes endoplasmic reticulum stress and other mechanistic studies. Based on the summary above, it is clear that current research on signals and receptors in the nervous system is extensive. A large number of studies have found that the activation or inhibition of different receptors or signals plays a great role in the formation and development of CPSP, so setting up corresponding targeted stimulants or inhibitors according to different signals and receptors is the current hotspot for drug treatment of CPSP. We also summarized the mechanism of acupuncture treatment for CPSP, and it is known that acupuncture relieves CPSP pain symptoms by inhibiting the autophagy of neurons in the hippocampus, which is the focus of current research. However, when compared to other pharmacological and non-pharmacological therapies, there are a scarcity of research on the mechanisms of acupuncture therapy.

Central post-stroke pain has a significant impact on patients’ lives, and researching the disease’s mechanism remains a hot and demanding research topic. How should future research directions be decided based on existing research? For many scholars, this is a key question. In order to facilitate future pertinent studies and give researchers ideas, this review paper sorts, summarizes, and classifies the most recent relevant mechanism studies. It also provides explanations in the form of figures and tables. The writers of this report also concentrated on the mechanism of acupuncture treatment of this condition. Although there are a paucity of research in this field, acupuncture treatment has been extensively regarded and recognized internationally in recent years as a traditional Chinese therapy, and the efficacy of acupuncture treatment for CPSP has been proven to be superior in the authors’ clinical work. As a result, the authors feel that acupuncture could be an important breakthrough in future study on the pathogenesis and therapeutic treatment of CPSP.

Because of the limited number of articles reviewed, this review primarily summarizes studies on relevant molecular mechanisms based on animal experiments and does not discuss clinical studies on CPSP. Meanwhile, due to limited access to the original text of some studies, the summary of molecular mechanisms in this review has some deficiencies and limitations, which we hope will be supplemented and corrected by future studies.

Conclusion

A large number of mechanistic studies on CPSP have been conducted. According to research, the pathogenesis of CPSP involves the entire spinal-thalamo-cortical pathway, and multiple substances in the nervous system are involved in the formation and development of CPSP. The discovery and exploration of various receptors and signals have resulted in a plethora of therapeutic ideas for CPSP. Acupuncture, despite being a highly effective treatment, has received little attention in terms of the analgesic mechanism of CPSP, and more experimental research is required.

Abbreviations

ACC

Anterior cingulate cortex

ACC^GABA→Glu

Anterior cingulate cortex GABA-containing neurons to glutamatergic neurons

ATP

Adenosine triphosphate

BCAO

Bilateral carotid artery occlusion

BDNF

Brain-derived neurotrophic factor

CFA

Complete Freund’s adjuvant

CPSP

Central post-stroke pain

CXCR4

C-X-C motif chemokine receptor 4

DHA

Docosahexaenoic acid

EA

Electroacupuncture

EETs

Epoxyeicosatrienoic acids

EPA

Eicosapentaenoic acid

ER

Endoplasmic reticulum

GABA

Gamma-aminobutyric acid

GABAaR

Gamma-aminobutyric acid a receptor

GRP40

Free fatty acid receptor 1 (FFA1)

HCN

Hyperpolarization-activated cyclic nucleotide-gated

HIF-1α

Hypoxia-inducible factor 1α

I/R

Ischemia/reperfusion

IL-1β

Interleukin-1β

ITAB

Intra-thalamic autologous blood

ITC

Intra-thalamic collagenase

LPA

Lysophosphatidic acid receptors

MD

Medial dorsal thalamus

MED1

Mediator complex subunit 1

MT

Medial thalamus

NLRP3

NLR pyrin domain-containing 3

NOS

Nitric oxide synthase

PEALut

N-palmitoylethanolamide and luteolin

PFC

Prefrontal cortex

PF^Glu

Parafascicular thalamic nucleus

PIT

Photochemically induced thrombosis

PO^Glu

Posterior thalamic nucleus

PUFA

Polyunsaturated fatty acid

PWT

Paw retraction threshold

S1^Glu

Primary somatosensory cortex glutamatergic neurons

SDF1

Stromal cell-derived factor 1

sEH

Soluble epoxide hydrolase

STT

Spinal thalamic tract

TH

Thalamic hemorrhage

THS

Thalamic hemorrhagic stroke

TNF-α

Tumor necrosis factor-α

TrkB

Tropomyosin receptor kinase B

TRN

Thalamic reticular nucleus

TSPO

Translocator protein

UPR

Unfolded protein response

VB

Ventrobasal thalamus

VPL

Ventral posterolateral

VPM

Ventral posteromedial

Author Contributions

YC: collected and summarized the literatures and wrote the first draft of this review. JH and YC: were involved in the collection and arrangement of the literatures. BW: reviewed and revised the paper. All authors contributed to the final version of the manuscript.

Funding

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

Data Availability

Data availability is not applicable to this article as no new data were created or analyzed in this study

Declarations

Conflict of interest

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.