miR-590-3p Alleviates diabetic peripheral neuropathic pain by targeting RAP1A and suppressing infiltration by the T cells

Background: MicroRNAs play a crucial role in diabetic peripheral neuropathic pain (DPNP). miR-590-3p is a novel miRNA and involved in multiple diseases. However, the pathological mechanism of miR-590-3p in DPNP needs to be elucidated. Materials and methods: The db/db mice and db/m mice were selected to mimic diabetes and control, respectively, for in vivo studies. The miR-590-3p agomir was injected into db/db mice and pain-related behavioral tests were performed. The interaction of miR-590-3p with target gene was confirmed by dual-luciferase reporter assay. The expression of target gene was determined by qRT-PCR and western blot assay. The levels of inflammatory cytokines were measured by enzyme-linked immunosorbent assay (ELISA). Results: miR-590-3p was down-regulated in diabetic peripheral neuropathy mice. More importantly, miR-590-3p agomir alleviated pain-related behavior, reduced TNF-α, IL-1β and IL-6 concentrations, and inhibited neural infiltration by immune cells in db/db mice. Interestingly, RAP1A was predicted to be the target of miR-590-3p by Targetscan, and was actually regulated by miR-590-3p. Finally, the rescue experiments proved that overexpression of RAP1A partially abrogated the suppressive impact of miR-590-3p on T cells proliferation and migration. Conclusion: miR-590-3p ameliorates DPNP via targeting RAP1A and inhibiting T cells infiltration, indicating that exogenous miR-590-3p may be a potential candidate for clinical treatment of DPNP.


INTRODUCTION
Diabetes has a high prevalence rate and has become a common chronic disease all over the world. It also induces a complication -diabetic neuropathy, which causes diabetic peripheral neuropathic pain (DPNP) (Davies et al., 2006). Diabetic patients with DPNP experience limb numbness, spontaneous pain, hyperalgesia, or allodynia (Calcutt 2004). Given that diabetic peripheral neuralgia (DPN) is a typical chronic neurogenic pain, it is difficult to treat, affects the physiological functions and quality of life of the patients, and brings a heavy burden to individuals and society (Boulton et al., 2004). The pathogenesis of DPNP is complex, including inflammation, repair process, and gene expression (Vanotti et al., 2007). Although the diverse treatment strategies have advanced, DPNP is still difficult to cure. Therefore, the pathogenesis of DPNP should be further clarified and new treatment strategies should be developed.
MicroRNA (miR), is a non-coding RNA, which generally has a length of 18-22-nucleotides. MiRs can bind to the 3'-UTR of the target gene to modulate the gene expression (Wu et al., 2013). Currently, the expression of miRs has been proved to be closely related to DPN. miR-146a influences the severity of DPN through the regulation of inflammation (Feng et al., 2018). Moreover, knockdown of miR-25 can significantly aggravate diabetic peripheral neuralgia via the production of the reactive oxygen species . Overexpression of miR-146a can suppress hyperglycemia-induced proinflammatory genes and alleviate diabetic peripheral neuralgia (Liu et al., 2017). Moreover, miR-590-3p is a novel miRNA involved in multiple diseases. The expression of miR-590-3p is reduced in Alzheimer's disease, suggesting that it may be related to neural degradation (Villa et al., 2011). More importantly, miR-590-3p is down-regulated in DPN model (Gong et al., 2015). However, the specific role and mechanism of miR-590-3p in DPN remain unknown.
RAP1A (Ras-associated protein 1A) is a member of the Ras-like GTPases family and plays a vital role in the cellmatrix and cadherin-mediated cell-cell contacts (Duchniewicz et al., 2006). In addition, RAP1A regulates T cells via augmenting lymphocyte responses and activating integrins (Sebzda et al., 2002). Moreover, inhibition of RAP1A alleviates neuropathic pain (Li et al., 2015;Fang et al., 2019). Also, accumulating evidence has proved that pro-inflammatory T cells migrate into the spinal cord in several pain models, and the immune cell infiltration is involved in diabetic neuropathy (Costigan et al., 2009;Grace et al., 2011;Agarwal et al., 2018). Given all these results, RAP1A might be related to the pathogenesis of DPNP.
The aim of our study was to elucidate the role of miR-590-3p in DPNP. We found that miR-590-3p was downregulated in db/db mice (type 2 diabetes model).
Overexpression of miR-590-3p alleviated DPNP in db/ db mice by targeting RAP1A, providing strong evidence for the role of altered miR-590-3p/RAP1A axis in the development of DPNP.

Animals.
All animal experiments were performed in accordance with the Guide for the Care and Use of Vol. 67, No 4/2020 587-593 https://doi.org/10.18388/abp.2020_5451 Laboratory Animals and were approved by the Ethics Committee of the First Affiliated Hospital of Soochow University. The 20-week-old male db/db mice and db/m mice use in this study were purchased from Nanjingjunke (China). miR-590-3p Agomir and NC agomir (Genepharma, China) were successfully injected into 20-week-old mice according to the manufacturer's protocol. We delivered miR-590-3p agomir or NC agomir to db/db mice via tail vein once a week for 4 weeks under anesthesia with isoflurane. Before sacrificing the mice, pain-related behavioral tests were performed. Before the end of the animal experiments, the blood glucose in one drop of tail blood and body weight were determined with a blood glucose monitoring system (Bayer, Germany) and electronic balance (Meitele, Switzerland), respectively. Finally, the dorsal root ganglion (DRG) tissues were extracted, frozen and stored for further experiments.
Isolation of T cells and transfection. pleens from db/m mice were collected and disaggregated, and then the erythrocytes were lysed with Lysing Buffer (BD Biosciences, USA). The T cells were isolated using CD4 + T cell Isolation Kit (Miltenyi Biotec, Germany). Next, the T cells were cultured and stimulated with 0.5 μg/ml anti-CD3e (BD Biosciences, USA) in RPMI-1640 medium (Gibco, USA) containing 10% fetal bovine serum (FBS; Gibco, USA) at 37°C.
Pain-related behavioral tests. The pain-related behavioral tests were performed as follows: the mice were placed on a hot plate (approx. 55°C), and then the reaction time (latency of the first recoil or struggle of the rear paw) was recorded. To avoid skin damage, we applied a cutoff time of 10 seconds. The shortened withdrawal latency indicated hyperalgesia (Chopra et al., 2010). In addition, each mouse's tail-swinging delay was determined by dipping the distal end of the tail 2-3 cm into a vessel filled with cold water (10±1°C). The data on the duration of the struggle, tail-swinging reaction, or tail-swinging were recorded. The threshold value was set at 15 seconds. A shorter soaking time indicated hyperalgesia which is also attributed to a central mechanism (Kamboj et al., 2010).
Western blot. The cells or DRG tissue were lysed with RIPA lysis buffer (Beyotime, China) to isolate the proteins, and the protein concentration was determined using a BCA kit (Beyotime, China). The equal amount of protein was separated by SDS-PAGE and transferred onto a PVDF membrane. Then the membranes were washed with TBST (Tris Buffered Saline with Tween) buffer three times and incubated with anti-RAP1A antibody (1:800; Abcam, UK) or anti-β-actin antibody (1:800; Abcam, UK) overnight at 4°C. After washed with TBST buffer for three times, the membranes were then incubated with Goat Anti-Rabbit IgG H&L (HRP) (1:8 000; Abcam, UK) or Goat Anti-Mouse IgG H&L (HRP) (1:8 000; Abcam, UK) for 2 h. Finally, an electrochemiluminescence kit (Beyotime, China) was used to measure the corresponding protein expression levels.
ELISA assay. The levels of IL-1β, TNF-α and IL-6 in DRG tissues were evaluated by ELISA. After the preparation of the samples, the concentration of IL-6, IL-1β and TNF-α in the sample supernatant was determined with IL-6, IL-1β and TNF-α ELISA kit (Elabscience, China), respectively. Each measurement was repeated three times.
Immunohistochemistry (IHC). The collected DRG tissues were subjected to IHC staining. After fixation with 10% paraformaldehyde, the samples were sliced horizontally, blocked with goat serum (Jackson Immu-noResearch, USA), and probed with anti-CD4 antibody (1:200; Abcam, UK). DAB and hematoxylin were used to stain the tissue slices. RAP1A staining was visualized using a light microscope (Olympus, Japan).
Transwell assay. After transfection, T cells were seeded onto the upper chamber of a Transwell plate (filters diameter: 6.5 mm, pore size: 5 μm; Corning, USA). The lower chamber was filled with medium containing sphingosine-1-phosphate (S1P; Sigma, USA). After 4 hours of incubation, the cells in the lower chamber were harvested and counted under a light microscope (Olympus, Japan).
Data statistics. The data in this study were analyzed with SPSS 20.0 (SPSS, USA) and presented as mean and Standard Deviation (S.D.). The Student's t-test or one-way analysis of variance was performed for comparison between two groups or more than two groups, respectively. P < 0.05 was considered to indicate a significant difference.

miR-590-3p was down-regulated in diabetic peripheral neuropathy mice
Compared to db/m mice, the blood glucose in db/ db mice was markedly elevated, suggesting the suc-miR-590-3p Ameliorates diabetic peripheral neuropathic pain via targeting RAP1A cessful establishment of diabetes models (Fig. 1A).
Besides, compared to db/m mice, the body weight was higher in db/db mice. For pain-related behavior test, the hind paw reaction time and tail withdrawal latency were lower in db/db mice compared to those in db/m mice, suggesting that db/db mice were in hyperalgesia and suffered from diabetic neuropathy pain (Fig. 1B). Interestingly, compared to db/m mice, the db/db mice exhibited a reduction of miR-590-3p levels in DRG tissue (Fig. 1C). These results confirmed that diabetic peripheral neuropathy mice models were successfully established and miR-590-3p was down-regulated in diabetic peripheral neuropathy mice.

miR-590-3p inhibited the production of proinflammatory mediators and neural infiltration by the immune cells
An excessive amount of blood glucose stimulates the production of pro-inflammatory cytokines (including IL-6, IL-1β and TNF-α) and directs the cells toward inflammation (Rains & Jain 2011). When compared to the control, the levels of IL-6, IL-1β, and TNF-α in db/db mice were increased, however, they were significantly reduced by miR-590-3p agomir (Fig. 3A). Accumulating evidence has shown a significant neural infiltration by the immune cells in DPNP patients (Younger et al., 1996;Alexandraki et al., 2006). The infiltration by the immune cells was determined by IHC staining of CD4. The expression of CD4 in DRG tissue of db/db mice was elevated (Fig. 3B). After the injection of miR-590-3p agomir into db/db mice, the expression of CD4 was reduced (Fig. 3B). These findings proved that exogenous miR-590-3p inhibited the production of pro-inflammatory mediators and neural infiltration by immune cells in db/db mice.

Figure 1. miR-590-3p was down-regulated in diabetic peripheral neuropathy mice
The 20-week-old db/db mice (diabetes) and db/m mice (control) were studied. (A) Blood glucose and weight were measured. (B) The tail immersion and hot plate tests were performed. (C) The expression of miR-590-3p was determined by qRT-PCR in DRG tissue. Ten mice per group. ***p<0.001.

miR-590-3p inhibited proliferation and migration of T cells via targeting RAP1A
To clarify whether miR-590-3p affected the production of pro-inflammatory mediators and neural immune cells infiltration in a RAP1A-dependent manner, the isolated T cells were co-transfected with RAP1A3-overexpression vector or the corresponding empty vector and NC mimic or miR-590-3p mimic. Western blot verified that miR-590-3p mimic partially reduced the effect of RAP1A overexpression (Fig.  5A). More importantly, CCK-8 assay indicated that miR-590-3p mimic markedly reduced cell viability, whereas this effect was reversed by overexpression of RAP1A (Fig. 5B). Furthermore, Transwell assay results showed that miR-590-3p mimic reduced cell migration ability, and the effect was reversed by overexpression of RAP1A (Fig. 5C). With these results, we concluded that overexpression of RAP1A partially abrogated the suppressive impact of miR-590-3p on T cells proliferation and migration.

DISCUSSION
In the current study, the diabetic peripheral neuropathy mice models were successfully established and we found that miR-590-3p was down-regulated in diabetic peripheral neuropathy mice. Moreover, miR-590-3p agomir reduced the production of pro-inflammatory mediators and neural infiltration by the immune cells in db/ db mice. Interestingly, RAP1A was predicted to be the direct target of miR-590-3p by Targetscan, and miR-590-3p regulated the expression of RAP1A. Finally, the res-cue experiments proved that overexpression of RAP1A partially rescued the suppressive effects of miR-590-3p on T cells proliferation and migration. Together, these findings suggested that exogenous miR-590-3p may be beneficial to the clinical treatment of DPNP.
The hyperglycemic state may induce the development of DPN, which induces oxidant and inflammatory mediators, thereby leading to deleterious effects in tissues (Anitha et al., 2006;Martyn et al., 2008). Although DPN has been studied, the molecular mechanism underlying inflammatory mediators has not been fully explored (Feldman et al., 2017). Accumulating evidence has confirmed that miRNAs are involved in DPN (Zhu & Leung 2015). For instance, miR-146a and miR-155 regulate inflammation in diabetic peripheral neuropathy (Feng et al., 2018;Chen et al., 2019). Meanwhile, hsa-miR-590-3p was identified as an LDHAsuppressing microRNA and can significantly improve glucose metabolism in T2D mice model . Moreover, a recent study demonstrated that miR-590-3p is down-regulated in DR cells . Similarly, the expression of miR-590-3p was found to be decreased in db/ db mice, which provided strong evidence for its vital role in diabetes in vivo. Furthermore, miR-590-3p agomir alleviated DPNP in db/db mice. Nevertheless, the mechanism of how miR-590-3p alleviates DPNP remains unknown.

Figure 5. miR-590-3p inhibited T cells proliferation and migration via targeting RAP1A
The RAP1A3-overexpression vector or the corresponding empty vector and NC mimic or miR-590-3p mimic were co-transfected into T cells. (A) The expression of RAP1A was detected by western blot. (B) The cell proliferation was determined using the CCK-8 kit. (C) Four hours after transfection, the cell migration ability was evaluated by Transwell assay. Every experiment was repeated three times. *p<0.05, **p<0.01, ***p<0.001.
we demonstrated that miR-590-3p inhibited neural infiltration by T cells. Besides, under hyperglycemia, TNF-α production aggravates in neural tissues, thereby inducing nerve damage, and finally resulting in the development of diabetic polyneuropathy (Satoh et al., 2003). Many central nervous system paradigms proved that IL-6 induces axonal regeneration, which may be related to immune cells proliferation and migration (Leibinger et al., 2013;Carmel et al., 2015). Thus, we suspected that miR-590-3p affects the T cells migration and proliferation, which needed more experiment in the next step.
RAP1A is confirmed to be a target of a series of miRNAs and is involved in cell migration and proliferation in non-small cell lung cancer, breast cancer, cervical cancer and prostate cancer (Xiang et al., 2015;Cao 2019;Lu et al., 2020). In this study, RA-P1A was proved to be a direct target of miR-590-3p. As we suspected, miR-590-3p inhibited T cells proliferation and migration in DPNP, which was reversed by over-expression of RAP1A. These findings suggested that miR-590-3p inhibited T cells proliferation and migration via targeting RAP1A in DPNP, which was consistent with the previous research. A recent study pointed out that lncMIR205HG acts as a natural decoy for miR-590-3p and leads to the process of head and neck squamous cell carcinoma (Di Agostino et al., 2018). Besides, RAP1A contributes to cell migration via regulating the MAPK/ERK pathway (Fujita et al., 2005;. Given all these reports, we hypothesized that lncMIR205HG may target miR-590-3p and activate the MAPK/ERK pathway in DPNP. However, this hypothesis needs more experiments to confirm it in the future.
In summary, our data demonstrated for the first time that miR-590-3p inhibited the production of proinflammatory mediators and neural infiltration by T cells in DPNP. Moreover, additional in vitro studies indicated that miR-590-3p inhibited T cells proliferation and migration by targeting RAP1A. Our findings provide a novel insight into the molecular mechanism and molecular basis of treating DPN with exogenous miR-590-3p. Although the treatment strategy of exogenous miR-590-3p can be implemented, the interaction network of the miR-590-3p in diabetes remains to be elucidated.