Mesenchymal Stem Cells Transplantation for Neuropathic Pain Induced By Peripheral Nerve Injury in Animal Models: A Systematic Review

Abstract

Neuropathic pain is defined as a lesion or disease of the somatosensory system, currently remaining a challenging condition to treat. Mesenchymal stem cells (MSCs) transplantation is emerging as a promising strategy to alleviate the neuropathic pain conditions induced by peripheral nerve injury. The aim of this systematic review was to assess the efficacy and safety of MSCs transplantation in neuropathic pain induced by peripheral nerve injury in controlled animal studies, and thus to yield evidence-based decision making. Following the PRISMA guidelines, PubMed, Cochrane Central Library, Embase, and CINAHL were searched for preclinical controlled animal studies from the inception to April 16, 2020. Seventeen studies are included in this review. Substantial heterogeneity is observed regarding the animal's species, models of neuropathic pain, regimen of MSCs transplantation, and outcome of measures across the included studies. Both mechanical allodynia and thermal hyperalgesia could be significantly attenuated by transplanted MSCs. The MSCs-elicited analgesic effect is independent of the type of MSCs, time of administration, and route of delivery, and is efficiently enhanced by genetic transfection with fibroblast growth factor, proenkephalin, and glial cell line-derived neurotrophic factor. The migration of MSCs after intrathecal or intravenous injection has been shown to be directed toward the surface of dorsal spinal cord or dorsal root ganglions on the ipsilateral side of injury. No adverse effects have been reported. The accumulating evidence demonstrates the therapeutic effect of MSCs-based cell therapy on prevention and alleviation of the neuropathic pain induced by peripheral nerve injury in rat or mouse models. The robust preclinical studies are deserved to optimize the regimen of MSCs transplantation and to promote the translation of the MSCs-based therapy into clinical studies.

Introduction

Neuropathic pain is defined as a direct consequence of a lesion or disease affecting the somatosensory system either in the periphery or centrally [1,2]. The prevalence of neuropathic pain is estimated to range from 6.9% to 10% of population, and an estimated 3.75 million cases of neuropathic pain have been reported in the United States [3]. The positive symptoms associated with neuropathic pain are presented with stimulus-dependent hyperalgesia and allodynia or stimulus-independent pain such as shooting, stabbing, or burning [4]. An interdisciplinary strategy, centered on pharmacological medicine including anticonvulsants, antidepressants, antiarrhythmics, cannabidiol, and opioids, is currently used for the treatment of neuropathic pain conditions [5 –7]. However, the analgesic effect is limited due to dose-induced side effect [5]. As such, novel therapeutic strategies with long-lasting effect are warranted to be investigated.

More attention has been focused on the therapeutic potential of mesenchymal stem cells (MSCs)-based cellular therapy in a broad range of disorders, including spinal cord injury, osteoarthritis, Crohn's disease, graft-versus-host disease, and neuropathic pain conditions. MSCs are easily harvested from bone marrow, adipose tissue, and umbilical cord, and induced to differentiate into the cell lineages of endoderm, mesoderm, and ectoderm. In addition to the differentiation capacity, the trophic, anti-inflammatory, and immunomodulatory properties possessed by MSCs through secreting paracrine factors enable MSCs to treat the neuroinflammation disease [8]. In vitro and in vivo studies have demonstrated the positive role of MSCs in inhibiting the neuroinflammation and demyelination of the nervous system [9 –11].

Currently, large bodies of preclinical randomized controlled animal experiments are available, providing the evidence that MSCs transplantation could play important roles in the alleviation of neuropathic pain symptoms, such as allodynia and hyperalgesia. It is worth looking into the therapeutic potential of MSCs-based cellular therapy in establishing a regenerative, anti-inflammatory, and immunomodulatory environment, contributing to the arrest or possible reversal of the intractable condition of neuropathic pain. However, the evidence derived from these animal studies is less robust owing to small sample size, heterogeneous designs, and conflicting outcomes. Moreover, some issues regarding the MSCs transplantation for the treatment of neuropathic pain, including optimal does, route of delivery, survive, migration, mechanisms of action, are still unclear, necessitating the further characterization before being introduced into clinical trials.

The purpose of the systematic review is to assess the efficacy and safety of MSCs transplantation in neuropathic pain induced by peripheral nerve injury in controlled animal studies, to provide a comprehensive overview of literature, to appraise the experimental designs, to elucidate the underlying mechanisms, and to facilitate the MSCs-based cellular therapies from bench to bedside.

Methods

The study protocol was finalized a priori by all authors, in which the objectives, electronic search strategy, inclusion/exclusion criteria for study selection, data collection, outcomes of interest, and analytical approaches were defined. The reporting of this systematic review was conducted in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) statement. The objective outcomes regarding the treatment of MSCs transplantation in neuropathic pain include the following: behavior test; migration, survival, and differentiation of MSCs; adverse effects; and mechanisms of action.

Search strategy

The comprehensive bibliographic search was performed in Medline (PubMed), Cochrane Central Library, Embase, and CINAHL from the inception to April 16, 2020. The following medical subject heading terms (MeSH) and free words were used in combination with Boolean operators (AND, OR, NOT): mesenchymal stem cells, mesenchymal stromal cells, mesenchymal progenitor cells, mesenchymal precursor cells, marrow stromal cells, Wharton's jelly cells, pain, neuropathic pain, neuralgia, allodynia, hyperalgesia, and hypersensitivity. Two independent reviewers (Q.W. and H.C.H.) screened the titles and abstracts of articles retrieved by the initial search. The full texts of potentially eligible studies were reviewed. Discrepancies between the reviewers were resolved through discussion, and final decision was reached by consensus with a third reviewer (C.Q.H.). The reference lists of included studies were searched manually for potentially relevant articles.

Inclusion and exclusion criteria

All the controlled animal experiments evaluating the efficacy and safety of MSCs transplantation in neuropathic pain models induced by peripheral nerve injury were included in this systematic review. The language was restricted to English. Exclusion criteria included the following: (1) not controlled animal studies; (2) the neuropathic pain secondary to other disease models, central nerve system diseases, or diabetic neuropathy; (3) cell-free therapy, for example, condition medium, MSCs-derived extracellular vesicles or exosomes, MSC secretome; (4) irrelevant to the subject, clinical trials, reviews and meta-analyses, editorials, perspectives, and letters to the editors; (5) without the outcomes of interest.

Data extraction and tabulation

The following data are extracted and recorded in Table 1: the author, year of publication, recipient animal (species, strain, sex), neuropathic pain models, number of samples allocated to MSCs or control groups, type of graft (autologous, syngeneic, allogeneic, or xenogeneic), intervention regimen of MSCs group (source, total dose of transplanted cells, vehicle, route of delivery) and control group (vehicle and route of delivery), intervention time points (time from inducing the neuropathic pain to cell transplantation), observation time slot, outcome of measures and mechanisms of action. The data of outcomes of interest were extracted for the final time point when a serial of time points of assessment was performed.

Table 1.

Characteristics of the In Vivo Studies Included

Study	Animal species	Sex	Model	Type of graft	Sample (M/C)	Allocation		Intervention time	Observation time	Outcomes
Study	Animal species	Sex	Model	Type of graft	Sample (M/C)	MSCs (dose/volume/vehicle/route)	Control (volume/vehicle/route)	Intervention time	Observation time	Outcomes
Teng et al. [12]	Wistar rats	Male	CCD	Allogeneic	5/5	Rat BMSCs 1 × 10⁶/15 μL/PBS/i.t.	15 μL/PBS/i.t	On days 4 and 5 postsurgery	On day 28 postsurgery	① ②④
Forouzanfar et al. [13]	Wistar rats	Male	CCI	Allogeneic	8/8	Rat AD-MSCs Rat FGF1-AMSCs 1 × 10⁶/200 μL/PBS/i.v	Rat fibroblast 1 × 10⁶/200 μL/PBS/i.v	On days 0 and 1 postsurgery	On day 14 postsurgery	① ③④
Sun et al. [14]	SD rats	Male	CCI	Xenogeneic	13/13	Human BMSCs Human hPPE-BMSCs 6 × 10⁶/10 μL/PBS/i.t.c	10 μL/PBS/i.t	On day 7 postsurgery	On day 17 postsurgery	①
Mert et al. [15]	SD rats	Male	CCI	Allogeneic	7/7	Rat AD-MSCs 2 × 10⁶/2 mL/Medium/i.p. or 1 × 10⁶/1 mL/Medium/locally	2 mL/Medium/i.p. or 1 mL/Medium/locally	On day 1 or 7 postsurgery	On day 35 postsurgery	①
Liu et al. [16]	SD rats	Male	CCI	Allogeneic	5/3	Rat AD-MSCs/BMSCs 2 × 10⁵/100 μL/PBS/i.t or i.v 5 × 10⁵/100 μL/PBS/i.t or i.v	100 μL/PBS/i.t or i.v	On day 7 or 9 postsurgery	On day 42 postsurgery	① ③
Li et al. [17]	SD rats	Male	SNL	Allogeneic	8/8	Rat BMSCs Rat IL-1β pretreated BMSCs 2.5 × 10⁶/25 μL/PBS/i.t.c	25 μL/PBS/i.t.c	Before surgery or on day 7 postsurgery	On day 14 postsurgery	① ④
Fischer et al.[18]	SD rats	Male	TNI	Allogeneic	10/8	Rat BMSCs 2.5 × 10⁵/10 μL/PBS/i.t	10 μL PBS/i.t	At the same time with surgery	On day 35 postsurgery	① ③
Chiang et al.[19]	SD rats	Male	CCI	Xenogeneic	18/18	Human AF-MSCs 0.5 × 10⁶/200 μL/PBS/i.v	200 μL/PBS/i.v	—	On day 28 postsurgery	①
Chen et al. [20]	SD rats	Male	SNL	Xenogeneic	10/10	Human UC-MSCs 1 × 10⁶/20 μL/PBS/i.t.c	20 μL/PBS/i.t.c	On day 3 postsurgery	On day 14 postsurgery	① ④
Yu et al. [21]	SD rats	Male	SNL	Allogeneic	10/10	Rat EGFP/GDNF-BMSCs Rat EGFP-BMSCs 2 × 10⁴/2 μL/PBS/i.g	2 × 10⁴/2 μL/PBS/i.g	At the same time with surgery	On day 14 postsurgery	① ③
Chen et al. [22]	CD1 mice	Male	CCI	Allogeneic	6/6	Mice BMSCs 1 × 10⁵/10 μL/PBS/i.t; 2.5 × 10⁵/10 μL/PBS/i.t	10 μL/PBS/i.t	On day 4 or 14 postsurgery	On day 42 postsurgery	① ③④
Zhang et al. [23]	SD rats	Male	SNL	Allogeneic	10/10	Rat BMSCs 1 × 10⁵/10 μL/PBS/i.t	10 μL/PBS/i.t	On day 7 postsurgery	On day 24 postsurgery	①
Schafer et al. [24]	SD rats	Female	PSNL	Allogeneic	10/10	Rat BMSCs 1 × 10⁶/15 μL/PBS/i.t	15 μL/PBS/i.t	On days 2,3,4 postsurgery	On day 21 postsurgery	① ④
Sacerdote et al. [25]	C57BL/6 mice	Male	CCI	Xenogeneic	3/3	Human AD-MSCs 5 × 10⁵/200 μL/PBS/i.v 1 × 10⁶/200 μL/PBS/i.v	200 μL/PBS/i.v	On day 7 postsurgery	On day 21 postsurgery	① ④
Siniscalco et al. [26]	CD1 mice	Male	SNI	Xenogeneic	18/18	Human BMSCs 2 × 10⁶/100 μL/PBS/i.v	100 μL/PBS/i.v	On day 4 postsurgery	On day 90 postsurgery	① ④
Siniscalco et al. [27]	C57BL/6 mice	Male	SNI	Xenogeneic	5/5	Human BMSCs 5 × 10⁴/5 μL/PBS/l.c.v	5 μL/PBS/l.c.v	On day 4 postsurgery	On day 21 postsurgery	① ④
Musolino et al. [28]	SD rats	Male	SLNC	Allogeneic	10/10	Rat BMSCs 2 × 10⁵/4 μL/PBS/i.g	4 μL/PBS/i.g	At the same time with surgery	On day 56 postsurgery	①

Neuropathic pain animal models: CCD, chronic compression of dorsal root ganglion; CCI, chronic constriction injury; PSNL, partial sciatic nerve ligation; SLNC, single ligature nerve constriction; SNI, sciatic nerve injury; SNL, spinal nerve ligation; TNI, tibial nerve injury.

Route of delivery: i.g, intraganglionic injection; i.p, intraperitoneal; i.t, intrathecal; i.t.c, intrathecal catheter implantation; i.v, intravenous; l.c.v, lateral cerebral ventricle injection.

MSCs: BMSCs, bone marrow mesenchymal stem/stromal cells; AD-MSCs, adipose tissue mesenchymal stem/stromal cells; UC-MSCs, umbilical cord mesenchymal stem/stromal cells; AF-MSCs, amniotic fluid mesenchymal stromal cells.

Others: FGF, fibroblast growth factor; IL-1β, interleukin-1β; hPPE, human proenkephalin; EGFP, enhanced green fluorescent protein; GDNF, glial cell line-derived neurotrophic factor.

Outcomes: ①behavior test; ②migration, survival and differentiation of MSCs; ③adverse effects; ④mechanisms of action.

Quality assessment

The methodological quality of the included studies was evaluated using the Systematic Review Center for Laboratory Animal Experimentation's (SYRCLE) risk of bias tool for animal studies. This tool is composed of 10 items reflecting the 6 aspects of the risk of bias: (1) selection bias (sequence generation, baseline characteristics, and allocation concealment); (2) performance bias (random housing and blinding); (3) detection bias (random outcome assessment and blinding); (4) attrition bias (incomplete outcome data); (5) reporting bias (selective outcome reporting); and (6) other sources of bias. Indicating “yes” means a low risk of bias, “no” means a high risk of bias, and “unclear” means no sufficient details to measure the risk of bias.

Statistical analysis

Given the overall study heterogeneity in neuropathic pain models, regimen of MSCs transplantation, observation time, reporting of outcomes among the eligible studies, a formal statistical analysis or meta-analysis was attempted but not performed. Ultimately, a purely descriptive presentation of available data was adopted.

Results

Study characteristics

The flow chart of identification and selection process of literature is shown in Fig. 1. The initial literature search yielded 1,454 potentially relevant records, of which 287 duplications were excluded. After screening titles and abstracts, 68 records were retrieved for full-text evaluation. Ultimately, 17 studies were included and assessed in this systematic review.

FIG. 1.

Flow diagram of literature search and included studies.

Table 1 summarizes the basic characteristics of the 17 eligible studies [12 –28], all of which had been published between 2007 and 2019. Animals included Sprague Dawley rats in 11 studies (64.7%) [14 –21,23,24,28], Wistar rats in 2 studies (11.8%) [12,13], CD1 mice in 2 studies (11.8%) [22,26], and C57/BL6 mice in 2 studies (11.8%) [25,27]. Totally, 312 rats and 73 mice were involved in the eligible studies. Of which, 175 rats (51.1%) and 41 mice (56.2%) were allocated in the MSCs transplantation group, while 137 rats (48.9%) and 32 mice (43.8%) in the control group. The male rats or mice were used in most studies (94.1%) [12 –23,25–28], while the female mice were tested in only one study (5.9%) [24]. Among the animal models of neuropathic pain induced by peripheral nerve injury, chronic constriction injury (CCI) and spinal nerve ligation models were applied in seven studies (41.2%) [13 –16,19,22,25] and four studies (23.5%), respectively.

The regimens of MSCs transplantation for treatment of neuropathic pain in animal models are summarized as follows. The allogeneic transplantation of MSCs was applied in 11 studies (64.7%) with the donor cells obtained from rats [12,13,15 –18,21,23,24,28] or mice [22], while the xenogeneic transplantation applied in 6 studies (35.3%) with the donor cells from human [14,19,20,25 –27]. The sources of transplanted MSCs included bone marrow (BMSCs) in 12 studies (70.6%) [12,14,16 –18,21–24,26 –28], adipose tissue (AD-MSCs) in 4 studies (23.5%) [13,15,16,25], umbilical cord (UC-MSCs) in 1 study (5.9%) [20], and amniotic fluid (AF-MSCs) in 1 study (5.9%) [19]. The majority (88.2%) of studies used PBS as the control intervention [12,14,16 –28], while the remaining two studies (11.8%) adopted condition medium [15] and fibroblast cells [13] as controls, respectively.

To improve the therapeutic efficacy of transplantation, MSCs were genetically transfected with fibroblast growth factor-1 (FGF-1) [13], proenkephalin (PPE) [14], and glial cell line-derived neurotrophic factor (GDNF) [21], as well as pretreated with interleukin-1β (IL-1β) [17]. The MSCs were delivered into the rats or mice mainly through intrathecal injection in six studies (28.6%) [12,16,18,22 –24], intravenous injection in five studies (23.8%) [13,16,19,25,26], intrathecal catheter implantation in three studies (14.3%) [14,17,20], and intraganglionic injection in two studies (9.5%) [21,28]. The doses of transplanted MSCs were 2 × 10⁵–6 × 10⁶ and 1 × 10⁵–2 × 10⁶ for rats and mice, respectively. The intervention time when MSCs were transplanted was reported to be within 1 week after surgery of neuropathic pain model in 12 studies (70.6%) [12,14 –17,20,22 –27], immediately after surgery in 4 studies (23.5%) [13,18,21,28], and before surgery in only 1 study (5.9%) [17]. The MSCs-elicited analgesic effect on neuropathic pain was observed within 2–6 weeks after MSCs transplantation.

Risk of bias

The results of risk of bias for each study are summarized in Table 2. According to SYRCLE risk of bias assessment, none of the studies were judged as having a low risk of bias across all domains. The method of random sequence generation was not clearly described, thereby the risk of bias in the domain of sequence generation was judged as “unclear” in all studies. The majority of studies (88.2%) [12,14,16 –28] did not adequately describe the method used to conceal allocation, which was evaluated as high risk. Blinding for caregivers or researchers was incompletely described in all the studies, whereas blinding for outcome assessment was described in 11 studies (64.7%) [12,13,15,16,18,19,21,22,25 –27]. Moreover, the domains of random housing and random outcome assessment were assessed as high risk in nearly all the studies (88.2%) [12,14,16 –28]. All studies were judged as having unclear risk of attrition and reporting bias.

Table 2.

Assessments of Risk of Bias of the Included Studies by SYRCLE's Tool

Study	Selection bias			Performance bias		Detection bias		Attrition bias	Reporting bias	Other
Study	Sequence generation	Baseline characteristics	Allocation concealment	Random housing	Blinding	Random outcome assessment	Blinding	Incomplete outcome data	Selective outcome reporting	Other sources of bias
Teng et al. [12]	U	Y	N	N	N	N	Y	U	U	U
Forouzanfar et al. [13]	Y	Y	Y	Y	Y	Y	Y	U	U	U
Sun et al. [14]	U	Y	N	N	N	N	N	U	U	U
Mert et al. [15]	Y	Y	Y	Y	Y	N	Y	U	U	U
Liu et al. [16]	U	Y	N	N	N	N	Y	U	U	U
Li et al. [17]	U	Y	N	N	N	N	N	U	U	U
Fischer et al. [18]	U	Y	N	N	N	N	Y	U	U	U
Chiang et al. [19]	U	Y	N	N	N	N	Y	U	U	U
Chen et al. [20]	U	Y	N	N	N	N	N	U	U	U
Yu et al. [21]	U	Y	N	N	N	N	Y	U	U	U
Chen et al. [22]	U	Y	N	N	N	N	Y	U	U	U
Zhang et al. [23]	U	Y	N	N	N	N	N	U	U	U
Schafer et al. [24]	U	Y	N	N	N	N	N	U	U	U
Sacerdote et al. [25]	U	Y	N	N	N	N	Y	U	U	U
Siniscalco et al. [26]	U	Y	N	N	N	N	Y	U	U	U
Siniscalco et al. [27]	U	Y	N	N	N	N	Y	U	U	U
Musolino et al. [28]	U	Y	N	N	N	N	N	U	U	U

SYRCLE, Systematic Review Center for Laboratory Animal Experimentation; Y, “yes” means a low risk of bias; N, “no” means a high risk of bias; U, “unclear” means no sufficient details to measure the risk of bias.

Outcomes of interest

Analgesic effect of MSCs transplantation on neuropathic pain

The neuropathic pain model is characterized by mechanical allodynia or thermal hyperalgesia, which were assessed by paw mechanical withdrawal threshold and paw thermal withdrawal latency, respectively [29]. The results from the eligible studies show that both mechanical allodynia and thermal hyperalgesia were significantly attenuated by transplantation of BMSCs [14,17,18,21,22,26,27], AD-MSCs [13,15,16,25], UC-MSCs [20], and AF-MSCs [19]. In Sacerdote et al.'s study, the repeated administrations were able to prolong the MSCs-elicited analgesic effect on neuropathic pain [25]. Furthermore, the genetic transfection of MSCs with FGF-1 [13], PPE [14], and GDNF [21] as well as pretreatment of MSCs with interleukin-1β (IL-1β) [17] has been demonstrated to boost the MSCs-elicited analgesic effect on neuropathic pain.

In some studies, the analgesic effect of BMSCs transplantation on neuropathic pain was found to be more appreciable on reduction of mechanical allodynia but not thermal hyperalgesia [12,23,28]. Teng et al. [12] showed that intrathecal injection of BMSCs resulted in a profound and transient reduction of mechanical allodynia in rats. It is notable that the transplanted BMSCs after intrathecal injection had no beneficial effect on pain sensations such as allodynia and hyperalgesia in female mice [24].

Migration, survival, and differentiation of MSCs

The migration of transplanted MSCs in animal models of neuropathic pain was investigated in 11 studies [12,13,16 –18,20 –22,24,26,27]. Of these, 7 studies [16 –18,20,22,24,26] showed that the migration of transplanted MSCs after intrathecal injection was directed toward the surface of dorsal spinal cord or dorsal root ganglions (DRGs) on the ipsilateral side of peripheral nerve injury. In addition, the migration of MSCs after intravenous injection was investigated in three studies [13,16,26]. One study by Siniscalco et al. [26] indicated that the engrafted BMSCs after intravenous injection were detected around the surface of spinal cord or DRGs, while the other two studies by Forouzanfar et al. [13] and Liu et al. [16] did not locate the MSCs after intravenous injection.

The survival and differentiation of transplanted MSCs were observed in two studies [21,22]. Yu et al.'s study indicated that the BMSCs after intraganglion injection could survive for up to 3 weeks on the surface of DRGs, without disrupting the normal pattern of satellite glial cell apposition [21]. Similarly, Chen et al. found that the survival of transplanted BMSCs could last 84 days after intrathecal injection [22]. It is intriguing that the BMSCs engrafted into the subarachnoid space or DRGs eventually did not differentiate into neurons, glial cells, or monocytes in the microenvironment of injured peripheral nerve [21,22].

Adverse effects of MSCs

No adverse effect of transplanted MSCs in animal models of neuropathic pain has been reported in the recent studies [13,16,18,21 –23,25]. After MSCs transplantation, all the mice/rats were monitored for any abnormal changes in weight, locomotion, food and fluid intake, and survival. At last, no macroscopic toxicity, tumors, behavioral changes, or death were observed, confirming the safety of intrathecally or intravenously transplanted MSCs. Analysis of the cerebrospinal fluid also confirmed that no tumor was detected after intrathecal injection of BMSCs [23].

Mechanisms responsible for analgesic effect of MSCs

The underlying mechanisms how the transplanted MSCs migrate to the injured sites and elicit an analgesic effect on neuropathic pain were investigated in 10 studies [12 –15,17,20,22,24 –27]. The lesion of peripheral nerve leads to the release of stromal cell-derived factor-1 (SDF-1), monocyte chemoattractant protein-1 a (MCP-1a), and C-X-C motif chemokine ligand 13 (CXCL13), which could induce the engrafted MSCs expressing the corresponding chemokine receptors, for example, C-X-C chemokine receptor type 4 (CXCR4), C-C chemokine receptor type 2 (CCR2), and CXCR5 [13,17,22]. It has been shown that the CXCL12/CXCR4 axis mediated the migration of intrathecally transplanted BMSCs to lumbar DRGs after peripheral nerve injury in mice [22], while BMSCs could directionally migrate to the surface of the ipsilateral spinal cord through CXCL13/CXCR5 axis [17].

It has been demonstrated that the engrafted MSCs were able to release the trophic factors, such as FGF-1, protecting sensory neurons from the insult caused by inflammatory cytokines derived from peripheral nerve injury [13]. Furthermore, the transplanted MSCs have been shown to play a key role in suppressing the phenotypic activation of microglia and astrocyte, leading to the inhibition of neuroinflammation. Administration of BMSCs [17,22], UC-MSCs [20], or AD-MSCs [15] by intrathecal injection significantly decreased the levels of activated microglia and proinflammatory cytokines (TNF-α, IL-1β, IL-6, and IL-17), and increased the anti-inflammatory cytokines (TGF-β1 and IL-10) in the animal models of neuropathic pain. Similar results were reported in the studies with the AD-MSCs [13,25] and BMSCs [26] transplanted by intravenous injection in the neuropathic pain models. In addition, lateral cerebral ventricle injection of BMSCs has been shown to attenuate the mRNA levels of proinflammatory IL-1β, as well as the neural β-galactosidase in the prefrontal cortex, with a parallel reduction of activated microglia and astrocyte in the brain [27]. Besides, the purine receptor P2X4R, a protein expressed exclusively in microglia, was found to be involved in intrathecally transplanted MSCs-induced suppression of activated microglia and neuropathic pain [12].

Apart from the above-mentioned mechanisms, the release of PPE opioid peptides [14], reduction of reactive oxygen species [23], and restoration of normal inducible nitric oxide synthase expression in spinal cord [25] have been shown to be implicated in MSCs-elicited analgesic effect on neuropathic pain.

Discussion

The systematic review was conducted to evaluate the efficacy and safety of MSCs transplantation on neuropathic pain induced by peripheral nerve injury in animal models. Our findings were based on a thorough and critical analysis of all the included studies. The therapeutic efficacy of transplanted MSCs in neuropathic pain was supported by attenuation of mechanical allodynia and thermal hyperalgesia, well integration to the surrounding tissue of spinal cord or DRGs without spontaneous differentiation, suppression of activated microglia and astrocyte, and inhibition of neuroinflammation. No adverse effects of transplanted MSCs were reported during the treatment of neuropathic pain.

Despite the substantial heterogeneity observed, the analgesic effect of MSCs transplantation on neuropathic pain induced by peripheral nerve injury in mice or rats has been demonstrated in almost all the included studies. The opposite result was reported in only one study by Schafer et al. [24], who showed that the mechanical allodynia and thermal hyperalgesia were not significantly altered in the female rats after intrathecally injected MSCs. An increasing body of evidence supports a sexual dimorphism in the induction and maintenance of neuropathic pain [30]. As demonstrated in Sorge et al.'s study [31], the involvement of microglial cells in mediating mechanical pain hypersensitivity was confined to male mice, whereas the female mice achieved similar levels of pain hypersensitivity using adaptive immune cells, likely T-lymphocytes. Hence, further studies are needed to clarify the sex difference of MSCs-elicited analgesic effect on neuropathic pain.

So far, there has been no consensus regarding the optimal regimen of MSCs transplantation to prevent and alleviate the neuropathic pain in animal models. The MSCs-elicited analgesic effect on neuropathic pain has been shown to be independent of the type of MSCs [16], dose of administration [15,16,22,25], route of delivery [15,16], and time of intervention [15,17,22]. Liu et al. have found that the comparisons of efficacy of transplanted MSCs between intravenous versus intrathecal injection, bone marrow versus adipose tissue, and small dose (2 × 10⁵/100 μL) versus high dose (5 × 10⁵/100 μL) were not significantly different in the reversal of hyperalgesia induced by CCI of the sciatic nerve in rats [16]. Based on the literature review, the regimen of MSCs transplantation could be summarized as follows: MSCs were mainly harvested from bone marrow and transplanted through intrathecal injection at the dose of 1 × 10⁵–6 × 10⁶/10–100 μL or through intravenous infusion at the dose of 5 × 10⁵–1 × 10⁶/200 μL, either before the operation or 1 week postoperation. The robust evidence from the rigorous studies in the multiple laboratories is warranted to verify the above-mentioned regimen of MSCs transplantation for the treatment of neuropathic pain.

The enhancement of MSCs-elicited analgesic effect has been found through the genetic transfection of MSCs with FGF-1 [13], PPE [14], and GDNF [21]. This incorporation of genetic engineering into MSCs-based cellular therapy has been proved to be more efficient than what the transplantation of MSCs alone did [13,14,21]. The basal level of anti-inflammatory, immunomodulatory, and regenerative factors secreted by MSCs could be enhanced by genetic manipulation designed to target a particular mechanism of disease [32]. Based on the underlying mechanism of neuropathic pain, the analgesic peptides such as inhibitory neurotransmitters (eg, β-endorphin [33]), anti-inflammatory peptides (eg, IL-10 [34], fractalkine [35]), neurotrophins (NT-3 [33], VEGF [36]), and soluble receptors (eg, soluble tumor necrosis factor receptor [37]) have the potential to be employed for genetic modification of MSCs to treat neuropathic pain conditions. In addition, the inhibitory effect of MSCs on the release of proinflammatory cytokines and neuropathic pain could be augmented by pretreatment of MSCs with IL-1β [17].

The biologic effects of MSCs are mediated through their interaction with host tissues, requiring the MSCs to be in proximity to the targeted area. The accumulating evidence indicated that the migration of engrafted MSCs after intrathecal or intravenous injection was directed toward the surface of dorsal spinal cord or DRGs tissue on the ipsilateral side of peripheral nerve injury [13,16 –18,20 –22,24,26]. The tracking of engrafted MSCs in the enrolled studies was preformed through the fluorescent or bioluminescent labeling (eg, chloromrthylbenzamido [22], ⁹⁹mTc-HMPAO [13], bromodeoxyuridine [24]), and the genetic modification of MSCs with green fluorescent protein [17,21]. Indeed, the number of transplanted MSCs with specific fluorescence that were detected in the spinal cord or DRGs was quite few in the enrolled studies. In this regard, recent advances in imaging techniques are of significance to facilitate the tracking of transplanted MSCs, and further to enhance the understanding of MSCs-elicited analgesic effect on neuropathic pain.

The survival of engrafted MSCs is pivotal for successful long-term effectiveness of cell-based therapy for chronic pain. There have been only 2 in vivo studies available with relevance to the survival of allogeneic BMSCs in the treatment of neuropathic pain [21,22]. Chen et al.'s study demonstrated the long-term survival of allogeneic BMSCs in the DRGs tissue after intrathecal injection. However, the number of labeled BMSCs was dramatically reduced after 3 months of transplantation [22]. Recent studies have indicated that the allogeneic MSCs seem not be “immune privileged,” and the alloreactive antibodies against donor MSCs could be generated after transplantation [38]. Other factors that negatively affect the survival of transplanted MSCs include peripheral nerve injury-induced inflammatory response, reactive oxidative stress, and proapoptotic cytokines [39,40], as well as initial cell culture condition, passage number, and hypoxic environment [41]. The understanding of the survival and functional profile of transplanted MSCs after intrathecal or intravenous injection is of significance to better utilize the MSCs-based cellular therapy in the treatment of neuropathic pain.

The possible mechanisms involved in MSCs-elicited analgesic effect on neuropathic pain included the chemokine-induced directional migration toward nerve injury site [17,22], suppression of activated microglia-induced neuroinflammation [13,15,17,20,22], and modulation of proinflammatory and anti-inflammatory cytokines [13 –15,17,20,22,25 –27]. Of these, the MSCs-mediated suppression of nerve-injury-induced microglia activation in neuropathic pain was extensively studied. Microglia in the dorsal horn of spinal cord are strongly activated after peripheral nerve injury, mediating the release of neuroinflammatory cytokines IL-1β and TNF-α, and thereby responsible for the pathogenesis of neuropathic pain [42]. The MSCs transplanted by intrathecal, intravenous, or lateral cerebral injections all have been demonstrated to play a key role in the attenuation of activated microglia-mediated proinflammatory cytokines [13 –15,17,20,22,25 –27]. However, exactly how transplanted MSCs affect the activated microglia has yet to be fully elucidated in these studies. For example, which molecules and receptors are mainly involved in MSCs-elicited allogenic effect? and which microglia receptors are targeted specifically by MSCs? and whether the sexual dimorphism of MSCs-induced inactivation of microglia exists?

It has been demonstrated that the TGF-β1 and IL-10 released by MSCs could be implicated in the MSCs-elicited suppression of activated microglia [17,22]. The alleviation of neuropathic pain by intrathecally transplanted MSCs may be associated with the inhibition of P2X4 receptor in microglia [12]. In addition, in vitro studies provided compelling evidence for MSCs-elicited microglia inactivation. It has been shown that MSCs were capable of reprogramming microglia into a phenotype characterized by increased phagocytic activity and upregulated expression of anti-inflammatory mediators [43]. The MSCs secretome could inhibit lipopolysaccharide-induced microglia activation and inflammatory response through sphingosine kinase/S1P signaling [44]. Collectively, it is critical to elucidate the mechanisms underlying the MSCs-induced inactivation of microglia, which will enhance the MSCs-elicited analgesic effect on neuropathic pain.

In addition, the MSCs-based analgesic effect on neuropathic pain could be attributed to paracrine secretion of extracellular vesicles or exosomes, which play a fundamental role in cell-to-cell communication. The intrathecal injection of MSCs-derived exosome has been shown to reverse nerve ligation-induced mechanical and thermal hypersensitivities of rats [45]. It is suggested that the analgesic effects of MSCs-derived exosomes may involve their actions on neuron and immune cells. As a novel therapeutic approach, the manipulation of sensory neuron-macrophage/microglia communication, such as the regulation of microRNA encapsuled in MSCs-derived exosome, could be used to halt the development of neuropathic pain [46]. Further studies are needed to advance the knowledge of MSCs-derived exosome in the treatment of neuropathic pain through regulation of neuron–microglia communication.

This systematic review showed several limitations. First, substantial heterogeneity was observed among the included studies. The heterogeneity mainly existed in species and strains of animals, models of neuropathic pain, the regimen of MSCs transplantation (source of harvest, total dose, vehicle, route of delivery, intervention and observation time), and outcome of measures in the enrolled studies, which precluded more rigorous and robust analysis, as well as pooling of the results through meta-analysis. Second, the assessment of quality through SYRCLE's risk of bias tool indicated that almost all the included studies had unclear risk for most of the domains analyzed due to the lack of detailed documentation of the methodology in their publications. As thus, further effort should be made to standardize the outcomes analyzed and reported in each study, so as to improve the credibility and reliability of future publications. Third, the evidence was insufficient to verify which type of MSCs or route of transplantation could be superior than others. Further research is warranted to verify the optimal regimen of MSCs transplantation. Finally, the power of this systematic review was limited by the fact that the studies investigating the therapeutic efficacy of MSCs were currently limited to small animal models. Continuous studies need to go forward to relatively large animal models, and ultimately to clinical trials.

Conclusion

Based on the results of this systematic review, MSCs-based cellular therapy may be a promising strategy to alleviate the neuropathic pain induced by peripheral nerve injury in rat or mice models. The improvement of research design and standardization of reporting are deserved to facilitate the generation of robust and rigorous evidence for MSCs-elicited analgesic effect on neuropathic pain. Continuous research is still required to identify the appropriate transcriptional factors involved in MSCs-elicited analgesic effect, to modify the genome of MSCs for directional migration and sustained survival, and to elucidate the mechanisms underlying the MSCs-mediated inactivation of microglia in the treatment of neuropathic pain. Currently, there has not been registered clinical trial of MSCs treatment for neuropathic pain in ClinicalTrials.gov. Further studies in human clinical trials are warranted to advance our knowledge of the benefit.

Footnotes

Author Disclosure Statement

No competing financial interests exist.

Funding Information

This work was supported by grants from the National Natural Science Foundation of China, no. 8170226 (Qian Wang).

References

Scholz

and Woolf

. (2007). The neuropathic pain triad: neurons, immune cells and glia. Nat Neurosci, 10:1361–1368.

Baron

, Binder

and Wasner

. (2010). Neuropathic pain: diagnosis, pathophysiological mechanisms, and treatment. Lancet Neurol, 9:807–819.

Van Hecke

, Austin

, Khan

, Smth

, and Torrance

(2014). Neuropathic pain in the general population: a systematic review of epidemiological studies. PAIN, 155:654–662.

Jensen

and Finnerup

. (2014). Allodynia and hyperalgesia in neuropathic pain: clinical manifestations and mechanisms. Lancet Neurol, 13:924–935.

Gilron

, Baron

and Jensen

. (2015). Neuropathic pain: principles of diagnosis and treatment. Mayo Clin Proc, 90:532–545.

De Gregorio

, McLaughlin

, Posa

, Ochoa-Sanchez

, Enns

, Lopez-Canul

, Aboud

, Maione

, Comai

, and Gobbi

. (2019). Cannabidiol modulates serotonergic transmission and reverses both allodynia and anxiety-like behavior in a model of neuropathic pain. Pain, 160:136–150.

Belardo

, Iannotta

, Boccella

, Rubino

, Ricciardi

, Infantino

, Pieretti

, Stella

, Paino

, et al. (2019). Oral cannabidiol prevents allodynia and neurological dysfunctions in a mouse model of mild traumatic brain injury. Front Pharmacol, 10:352.

Kimbrel

, Kouris

, Yavanian

, Chu

, Qin

, Chan

, Singh

, McCurdy

, Gordon

, et al. (2014). Mesenchymal stem cell population derived from human pluripotent stem cells displays potent immunomodulatory and therapeutic properties. Stem Cells Dev, 23:1611–1624.

Jaimes

, Naaldijk

, Wenk

, Leovsky

, and Emmrich

. (2017). Mesenchymal stem cell-derived microvesicles modulate lipopolysaccharides-induced inflammatory responses to microglia cells. Stem Cells, 35:812–823.

10.

Thomi

, Surbek

, Haesler

, Joerger-Messerli

, and Schoeberlein

. (2019). Exosomes derived from umbilical cord mesenchymal stem cells reduce microglia-mediated neuroinflammation in perinatal brain injury. Stem Cell Res Ther, 10:105.

11.

Zhang

, Guo

, Yin

, Peng

, Pang

, Ma

, Zhang

, Xie

, Chen

, et al. (2020). RNA-Seq reveals underlying transcriptomic mechanisms of bone marrow-derived mesenchymal stem cells in the regulation of microglia-mediated neuroinflammation after subarachnoid hemorrhage. Stem Cells Dev, 29:562–573.

12.

Teng

, Zhang

, Yue

, Chen

, Qu

, Wei

, and Jia

. (2019). Intrathecal injection of bone marrow stromal cells attenuates neuropathic pain via inhibition of P2X(4)R in spinal cord microglia. J Neuroinflammation, 16:271–271.

13.

Forouzanfar

, Amin

, Ghorbani

, Ghazavi

, Ghasemi

, Sadri

, Mehri

, Sadeghnia

, and Hosseinzadeh

. (2018). New approach for the treatment of neuropathic pain: fibroblast growth factor 1 gene-transfected adipose-derived mesenchymal stem cells. Eur J Pain, 22:295–310.

14.

Sun

, Zhang

, Li

, Long

, and Sun

. (2017). Intrathecal administration of human bone marrow mesenchymal stem cells genetically modified with human proenkephalin gene decrease nociceptive pain in neuropathic rats. Mol Pain, 13:1–10.

15.

Mert

, Kurt

, Altun

, Celik

, Baran

, and Gunay

. (2017). Pulsed magnetic field enhances therapeutic efficiency of mesenchymal stem cells in chronic neuropathic pain model. Bioelectromagnetics, 38:255–264.

16.

Liu

, Hua

, Shen

, Yin

, Yang

, Cheng

, Aijun Liu, Wang

, and Cheng

. (2017). Comparative efficacy of multiple variables of mesenchymal stem cell transplantation for the treatment of neuropathic pain in rats. Mil Med, 182:175–184.

17.

, Deng

, Wang

, Yang

, Li

, Zhang

, and Yuan

. (2017). Interleukin-1beta pre-treated bone marrow stromal cells alleviate neuropathic pain through CCL7-mediated inhibition of microglial activation in the spinal cord. Sci Rep, 7:42260.

18.

Fischer

, Wang

, Xiang

, Bai

, Yu

, and Hogan

. (2017). Inhibition of neuropathic hyperalgesia by intrathecal bone marrow stromal cells is associated with alteration of multiple soluble factors in cerebrospinal fluid. Exp Brain Res, 235:2627–2638.

19.

Chiang

, Liu

, Sheu

, Chen

F-C

, Chen

C-J

, Su

H-L

, and Pan

H-C

. (2016). Feasibility of human amniotic fluid derived stem cells in alleviation of neuropathic pain in chronic constrictive injury nerve model. PLoS One, 11:e0159482.

20.

Chen

, Chen

, Yao

, Shu

, Feng

, Hu

, Hai

, Yao

, and Chen

. (2016). Intrathecal injection of human umbilical cord-derived mesenchymal stem cells ameliorates neuropathic pain in rats. Neurochem Res, 41:3250–3260.

21.

, Fischer

, Ebert

, Wu

H-E

, Bai

, and Hogan

. (2015). Analgesia for neuropathic pain by dorsal root ganglion transplantation of genetically engineered mesenchymal stem cells: initial results. Mol Pain, 11:1–13.

22.

Chen

, Park

, Xie

, and Ji

. (2015). Intrathecal bone marrow stromal cells inhibit neuropathic pain via TGF-beta secretion. J Clin Invest, 125:3226–3240.

23.

Zhang

, Song

, Ko

, and Lee

. (2014). Intrathecal administration of mesenchymal stem cells reduces the reactive oxygen species and pain behavior in neuropathic rats. Korean J Pain, 27:239–245.

24.

Schafer

, Berger

, Deumens

, Goursaud

, Hanisch

, and Hermans

. (2014). Influence of intrathecal delivery of bone marrow-derived mesenchymal stem cells on spinal inflammation and pain hypersensitivity in a rat model of peripheral nerve injury. J Neuroinflammation, 11:157.

25.

Sacerdote

, Niada

, Franchi

, Arrigoni

, Rossi

, Yenagi

, de Girolamo

, Panerai

, and Brini

. (2013). Systemic administration of human adipose-derived stem cells reverts nociceptive hypersensitivity in an experimental model of neuropathy. Stem Cells Dev, 22:1252–1263.

26.

Siniscalco

, Giordano

, Galderisi

, Luongo

, de Novellis

, Rossi

, and Maione

. (2011). Long-lasting effects of human mesenchymal stem cell systemic administration on pain-like behaviors, cellular, and biomolecular modifications in neuropathic mice. Front Integr Neurosci, 5:79.

27.

Siniscalco

, Giordano

, Galderisi

, Luongo

, Alessio

, Di Bernardo

, de Novellis

, Rossi

, and Maione

. (2010). Intra-brain microinjection of human mesenchymal stem cells decreases allodynia in neuropathic mice. Cell Mol Life Sci, 67:655–669.

28.

Musolino

, Coronel

, Hokfelt

, and Villar

. (2007). Bone marrow stromal cells induce changes in pain behavior after sciatic nerve constriction. Neurosci Lett, 418:97–101.

29.

Jaggi

, Jain

and Singh

. (2011). Animal models of neuropathic pain. Fundam Clin Pharmacol, 25:1–28.

30.

Coraggio

, Guida

, Boccella

, Scafuro

, Paino

, Romano

, Maione

, and Luongo

(2018). Neuroimmune-driven neuropathic pain establishment: a focus on gender differences. Int J Mol Sci, 19:281.

31.

Sorge

, Mapplebeck

, Rosen

, Beggs

, Taves

, Alexander

, Martin

, Austin

J-S

, Sotocinal

, et al. (2015). Different immune cells mediate mechanical pain hypersensitivity in male and female mice. Nat Neurosci, 18:1081–1083.

32.

Wyse

, Dunbar

and Rossignol

. (2014). Use of genetically modified mesenchymal stem cells to treat neurodegenerative diseases. Int J Mol Sci, 15:1719–1745.

33.

Machelska

, Schroff

, Oswald

, Binder

, Sitte

, Mousa

, Rittner

, Brack

, Labuz

, et al. (2009). Peripheral non-viral MIDGE vector-driven delivery of beta-endorphin in inflammatory pain. Mol Pain, 5:72.

34.

Min

, Kim

, Park

, Cho

, and Oh

I-H

. (2007). IL-10-transduced bone marrow mesenchymal stem cells can attenuate the severity of acute graft-versus-host disease after experimental allogeneic stem cell transplantation. Bone Marrow Transplant, 39:637–645.

35.

Holmes

, Arnott

, Vanderplank

, Kerr

, Longbrake

, Popovich

, Imai

, Combadiere

, Murphy

, et al. (2008). Intra-neural administration of fractalkine attenuates neuropathic pain-related behaviour. J Neurochem, 106:640–649.

36.

Chattopadhyay

, Krisky

, Wolfe

, Glorioso

, Mata

, and Fink

. (2005). HSV-mediated gene transfer of vascular endothelial growth factor to dorsal root ganglia prevents diabetic neuropathy. Gene Ther, 12:1377–1384.

37.

Hao

, Mata

, Glorioso

, and Fink

. (2007). Gene transfer to interfere with TNFalpha signaling in neuropathic pain. Gene Ther, 14:1010–1016.

38.

Ankrum

, Ong

and Karp

. (2014). Mesenchymal stem cells: immune evasive, not immune privileged. Nat Biotechnol, 32:252–260.

39.

van de Vyver

. (2017). Intrinsic mesenchymal stem cell dysfunction in diabetes mellitus: implications for autologous cell therapy. Stem Cells Dev, 26:1042–1053.

40.

Szabó

, Fajka-Boja

, É Kriston-Pál, Hornung

, Makra

, Kudlik

, Uher

, Katona

, Monostori

, et al. (2015). Licensing by inflammatory cytokines abolishes heterogeneity of immunosuppressive function of mesenchymal stem cell population. Stem Cells Dev, 24:2171–2180.

41.

Bahsoun

, Coopman

, Forsyth

, and Akam

. (2018). The role of dissolved oxygen levels on human mesenchymal stem cell culture success, regulatory compliance, and therapeutic potential. Stem Cells Dev, 27:1303–1321.

42.

Inoue

and Tsuda.

Microglia in neuropathic pain: cellular and molecular mechanisms and therapeutic potential. Nat Rev Neurosci 2018; 19:138–152.

43.

Hegyi

, Kornyei

, Ferenczi

, Fekete

, Kudlik

, Kovacs

, Madarasz

, and Uher

. (2014). Regulation of mouse microglia activation and effector functions by bone marrow-derived mesenchymal stem cells. Stem Cells Dev, 23:2600–2612.

44.

Marfia

, Navone

, Hadi

, Paroni

, Berno

, Beretta

, Gualtierotti

, Ingegnoli

, Levi

, et al. (2016). The adipose mesenchymal stem cell secretome inhibits inflammatory responses of microglia: evidence for an involvement of sphingosine-1-phosphate signalling. Stem Cells Dev, 25:1095–1107.

45.

Shiue

, Rau

, Shiue

, Hung

Y-W

, Li

Z-X

, Yang

, and Cheng

J-K

. (2019). Mesenchymal stem cell exosomes as a cell-free therapy for nerve injury-induced pain in rats. Pain, 160:210–223.

46.

Simeoli

, Montague

, Jones

, Castaldi

, Chambers

, Kelleher

, Vacca

, Pitcher

, et al. (2017). Exosomal cargo including microRNA regulates sensory neuron to macrophage communication after nerve trauma. Nat Commun, 8:1778.