Rno-miR-199a-3p targets Nedd4 to promote sensitization of ST36 acupoints via mast cell activation in a rat model of knee osteoarthritis

Selecting routine points on related meridians is widely accepted as the foundational principle of acupuncture. When the body is suffering disease or injury, corresponding acupoints are thought to be activated and manifest in several sensitized forms. Sensitized acupoints hold high clinical value as a reection of disease activity on the body surface. Mast cells have been implicated in the process of acupoint sensitization but the underlying regulatory mechanisms remain unclear. In the present study, we evaluated ST36 as a sensitized acupoint in the monosodium iodoacetate-induced knee osteoarthritis rat model. We rst conrmed sensitization at the ST36 acupoint through decreases in the acupoint mechanical pain threshold and instructively found an accompanying increase in skin mast cell degranulation. Thereafter, we used highthroughput RNA sequencing to reveal potential molecular mechanisms of acupoint sensitization. We showed that rno-miR-199a-3p was highly expressed in the sensitized ST36 acupoint and its expression was associated with mast cells. Functional experiments revealed that overexpression of rno-miR-199a-3p increased mast cell histamine release whereas inhibition of rno-miR-199a-3p decreased histamine release. Mechanistically, we established rno-miR-199a-3p acted to inhibit neural precursor cell expressed developmentally down-regulated 4 (Nedd4) protein expression through miRNA-mediated targeting of the 3’-UTR of Nedd4 mRNA. Moreover, we found ectopic expression of Nedd4 antagonized histamine release in mast cells and blocked the actions of rno-miR-199a-3p overexpression. Thus, our study establishes that mast cells participate in the process of acupoint sensitization, and further reveals a novel miRNA-based mechanism which is crucial for further understanding of acupoint sensitization and acupuncture applications.


Introduction
Acupuncture was rst introduced to the Western world in the 20th century. Indeed, acupuncture is now recognized as an effective treatment and has become one of the most common auxiliary and alternative therapies worldwide [1]. The speci c effector sites for needle insertion in acupuncture are known as acupoints, and these are widely considered the core focus of acupuncture research [2]. Previous studies have shown that acupoints are not xed, rather their position and function can change dynamically according to physiological and pathological conditions [3]. In this process, the term acupoint sensitization refers to the dynamic transformation of acupoints from "silent state" (health) to "active state" (disease) [4,5]. When the body suffers disease or injury, the corresponding acupoints will be activated and appear in several sensitization forms, including the expansion of neuron receptive eld, pain sensitization and heat sensitization [6,7]. This phenomenon may gradually disappear with disease recovery. Acupoint sensitization signi cantly affects the receptive eld size and sensitivity of acupoints, thus enhancing their therapeutic functions, including receiving stimulation and regulating body functions [8]. In modern acupuncture and moxibustion clinical practice, the correct application of acupuncture treatments can guarantee the e cacy of disease treatment [9][10][11]. Therefore, selection of the speci c sites representing sensitized acupoints is of great signi cance for the prescription and clinical e cacy of acupuncture.

Page 3/20
The occurrence and development of acupoint sensitization is closely related to the microphysical environment and chemical changes of local tissues [4,12]. Increasing research efforts have broadly contributed to a deepening understanding of acupoints, with the realization that acupoints are not single static points, but change according to the physiological and pathological state of the body [5]. For example, it was found that when acute gastric mucosal injury leads to a neuronal in ammatory response, there were more sensitization points in a relatively concentrated area of the body surface, and these sensitization points have a greater correlation with acupoints [13,14]. Notably, mast cells are important immune cells and widely considered as potential effector cells in acupuncture treatment [15,16] and appear to be one of the key signal ampli cation factors in acupuncture effects [16]. In support of this concept, one pilot study found that the density of mast cells from the ST36 acupoint in rats was higher than a nearby sham point [17]. At the same time, mast cells numbers along with their degranulation rate are one of the important indicators of acupoint sensitization [4,18]. Therefore, it is necessary to clarify the structural changes and functional characteristics of mast cells at acupoints to improve acupuncture treatments and to better understand the connotation of acupoints.
The local microenvironment has also been identi ed as an important acupuncture target and is closely related to the effects of acupuncture [4]. Previous studies mainly focused on the effects of acupuncture on chemical mediators such as histamine, 5-HT and trypsin, and their regulatory mechanisms. However, other potential regulatory mechanisms during acupoint sensitization such as those involving microRNAs (miRNAs) have been ignored to some extent. At present, most related studies have focused on the regulatory role of miRNAs in acupuncture treatment [19,20], but whether miRNAs are involved in the regulation of acupoint sensitization is still unclear.
Knee osteoarthritis (KOA), also known as knee degenerative osteoarthritis, is a chronic and degenerative disease characterized by the degeneration, destruction, and hyperplasia of articular cartilage. In traditional Chinese medicine, KOA belongs to the category of bone arthralgia. Animal models of KOA have been commonly used in experimental acupuncture studies and have proved useful in the study of acupoint sensitization. In the context of this study, KOA model animals have been used to study the relationship between acupoint sensitization and the characteristics of mast cells in sensitized acupoints. For instance, one study demonstrated that elevated GlyT2 expression was a crucial mediator of ST35 acupoint sensitization in KOA rats [21], while early laser moxibustion signi cantly reversed monosodium iodoacetate-induced (MIA)-induced mechanical hyperalgesia [22].
Based on the limited information currently available, our study aimed to explore the contribution of miRNAs to the acupoint sensitization phenomenon in a rat model of KOA. We used a von Frey electronic pain meter and infrared thermal imager to detect the mechanical pain threshold and temperature of acupoints near the knee joint in KOA model rats. Differentially expressed miRNAs were then identi ed through high-throughput sequencing after extracting miRNAs from subcutaneous connective tissue of the sensitized acupoints. In parallel, we also determined the differential expression of miRNAs in mast cells extracted from the sensitized acupoints and assessed their function in regulating degranulation. Together this information provides a valuable reference for explaining the regulatory mechanisms governing acupoint sensitization. MIA-induced KOA model Rats (n = 68) were randomly assigned to two groups comprising normal saline (NS) and model groups (KOA). The NS group received a 50 µL injection of 0.9% saline into the left knee whereas rats in the KOA group were anesthetized with iso urane (RWD, Shenzhen, China) and subject to a single intra-articular injection of MIA (3 mg/50µL; Sigma,USA) dissolved in 0.9% saline in the left knee joint through the infrapatellar ligament, as previously described [23].

Model evaluation
We rst compared the diameter of left knee joint of all rat before and after treatment. After detecting the mechanical pain threshold, selected rats from the NS and KOA groups were used for pathological examination of knee cartilage. Brie y, the rats were sacri ced and the joints exposed before separating the upper and lower articular surfaces, stripping, and visually observing the muscle and ligament tissues. Thereafter, the bone and cartilage tissues 1 cm above the tibial plateau were xed in 4% formaldehyde solution and subjected to para n embedding and sectioning before conducting H&E staining. The articular cartilage pathology was then observed and imaged with a light microscope (Leica, Milan, Italy).

Left knee joint diameter measurements
The transverse diameters of the knee were measured at days 0 and 14 with a slide caliper (Mitutoyo, Japan) to measure the horizontal distance between the highest left and right points of the knee joints, respectively, when exed at 90°. Each knee was measured three times, and the mean values recorded.

Measurement of acupoint sensitization
Acupoint mechanical thresholds were determined by measuring the incidence of foot withdrawal in response to mechanical indentation of ST36. An electronic Von Frey algesimeter (IITC Life Science Inc, America) was used to detect the mechanical pain threshold of the ST36 acupoints in NS and KOA rats prior to treatment and after 14 days. The mechanical pain threshold was detected average for three times and the change rate of mechanical pain threshold was calculated as: (mean value after model-value before model)/(mean value before model).
Histologic examination of ST36 acupoint skin Skin tissues from the ST36 acupoint were xed with 10% formalin in PBS and then embedded in para n. Skin sections with a thickness of 5 mm was prepared and stained with toluidine blue before examination using light microscopy (Leica, Milan, Italy).

RNA extraction and miRNA sequencing
Tissue samples were ash frozen in liquid nitrogen and stored at − 80°C until nucleic acid extraction. In total, 200 mg of fresh frozen tissues were used to isolate total RNA with the mirVana™ miRNA Isolation Kit (Thermo Fisher Scienti c, Waltham, MA, USA). A NanoDrop spectrophotometer (Thermo) was used to estimate total RNA concentration while an Agilent 2100 Bioanalyzer (Agilent Technologies, Inc., Santa Clara, CA, USA) was used to measure the quantity and purity of small RNAs. Small RNA libraries were constructed using the TruSeq Small RNA Sample Preparation kit (Illumina, San Diego, CA) according to the manufacturer's protocols. Brie y, small RNA samples were ligated with 5' and 3' adapters, followed by reverse transcription-PCR (RT-PCR) for cDNA library construction and incorporation of index tags. The cDNA library fragments were puri ed separated on TBE PAGE gels and the fraction containing miRNA inserts were isolated. The cDNA library samples were pooled in equimolar amounts and used for cluster generation and sequence analysis in a single lane on an Illumina HiSeq2000 by Shanghai Biotechnology Corporation.

Bioinformatics analysis
Raw FASTQ sequences were generated and demultiplexed using the Illumina CASAVA v1.8 pipeline. Per base sequence quality was then assessed using the FastQC toolkit (http://www.bioinformatics.babraham.ac.uk/projects/fastqc). Brie y, the 3' adapter sequences were trimmed, the read size ltered (16-35nt), unique reads counted and low abundance reads (< 10 reads) discarded. Unique sequence reads were then aligned to the rat genome and miRBase_v16 using the miRanalyzer web server tool (http://bioinfo2.ugr.es/miRanalyzer/miRanalyzer.php). MiRanda database was applied to predict the targeted binding of the differential miRNA and the 3'UTR of mRNA obtained in the previous step. The address of the miRanda database is http://www.microma.org/. The main functions of the predicted target genes regulated by differentially expressed miRNAs were determined using KEGG functional classi cations by database for annotation, visualization, and integrated discovery (DAVID 6.8). The address of KEGG database is http://www.genome.jp/kegg/.

Real-time quantitative PCR (RT-qPCR)
RNA was extracted using TRIzol® reagent (Invitrogen; Thermo Fisher Scienti c, USA) and subsequently treated with DNase I (Thermo Fisher Scienti c, USA ). Primers were designed using Primer Express 3.0.1 and synthesized by Sangon Biotech (Shanghai) Co. Ltd. First-strand cDNA synthesis was carried out by using a Reverse Transcription System Kit according to the manufacturer's instructions (#11801-025, OriGene Technologies, USA). The indicated miRNAs were ampli ed from cDNA with the following speci c primers: To detect Nedd4 mRNA expression, the primers used were as follows: Nedd4 forward CGGAGGACGAGGTATGGGAG Nedd4 reverse AAGGACTCCACTCATCGGGT Actin forward CACCCGCGAGTACAACCTTC Actin reverse CCCATACCCACCATCACACC U6 was applied as an internal control for miR-199a-3p while actin was used for assays measuring Nedd4. RT-qPCR reactions were performed and analyzed with a QuantStudio Dx Real-Time Instrument (Thermo Fisher Scienti c, USA) and QuantStudio™ Design & Analysis software using the following reaction conditions: 50℃,2 min; 95℃ 10 min (95℃, 15 s; 60℃, 1 min) x 40 cycles. Expression levels were calculated using the 2 −ΔΔ CT data analysis method.
Isolation and culture of ST36 skin mast cells Skin mast cells were isolated according to a previous report using aseptic techniques [24]. Brie y, the ST36 acupoint skin was cut into comb-like pieces and transferred to medium containing Dulbecco's Phosphate-Buffered Saline (DPBS) with Ca 2+ and Mg 2+ (Beyotime) containing 2.4 U/ml Dispase Type II (Solarbio) and incubated overnight at 4℃. The next day, the epidermis was removed from the dermis with ne forceps. The homogenized dermis was then transferred for secondary digestion for 60 min at 37℃ with DPBS containing collagenase, hyaluronidase, and DNase (all from Solarbio). The resulting suspension was ltered through a mesh sieve and centrifuged at 400 g for 10 min at 4 ℃. Cell pellets were then resuspended in cold DPBS and after further centrifugation, resuspended in MACS buffer (Thermo Fisher Scienti c, USA) and ltered sequentially through cell strainers. After centrifugation, the cells were suspended in 1 ml of cytokine free MC culture medium (Basal Iscove's medium (Gibco) with 10% fetal bovine serum (HyClone), penicillin-streptomycin (HyClone), nonessential amino acids (Procell, Wuhan, China), and α-monothioglycerol (RHAMN, Shanghai, China) and cultured at 37℃ in a cell culture incubator (Thermo) for 24 h. Mast cells were identi ed by toluidine blue staining MiRNA / plasmid transfection Rat skin mast cells and HEK-293 cells were used in transfection experiments as indicated. Control rno-miRNA (rno-miRNA-NC), rno-mir-199a-3p mimic and rno-mir-199a-3p inhibitors were purchased from RiboBio (Guangzhou, China). Nedd4 related plasmids were synthesis by TsingKe Biological Technology (Chengdu, China). For miRNA / plasmid transfection, cells were seeded in six-well dishes at a density of 2.0×10 5 cells per well 24 h prior to transfection. Thereafter, transfections were performed with Lipofectamine 2000 (Invitrogen, Carlsbad, Calif) and the cells harvested 48 h later for analysis.

Western blotting
Whole cell lysates were extracted using NP-40 Lysis buffer (Beyotime, Shanghai, China) and protein concentrations quanti ed using a BCA protein assay kit (Beyotime Biotechnology, Shanghai, China). Western blotting was performed according to standard procedures using rabbit polyclonal anti-Nedd4 antibody, mouse monoclonal anti-ag antibody and rabbit monoclonal anti-actin antibody along with appropriate horseradish peroxidase (HRP)conjugated secondary antibodies (Proteintech, Chicago, USA). Actin protein levels served as a loading control.
Fluorescence in situ hybridization RNA uorescence in situ hybridization (FISH) was performed as described previously on sections of para n-embedded tissues [25]. Brie y, sections were dewaxed at 62℃ for 2 hours and then rehydrated in DEPC dilution (Amresco,USA). The sections were then boiled for 10-15 minutes, treated with proteinase K (20 ug/ml Servicebio) and then washed 3 times in PBS. Thereafter, the sections were incubated for 1 h at 37℃ in pre-hybridization solution (Servicebio) before hybridization overnight with miRNA probes in hybridization solution. The sections were then washed with different concentrations of SSC solution (Servicebio). Cell nuclei were counterstained with DAPI (Servicebio) before mounting and observation using an epi uorescence microscope (Nikon Eclipse CI, Nikon, Japan). A Cy3-labelled rno-miR-199a-3p probe purchased from Shanghai Sinomics Corporation was used to detect miR-199a-3p (CACAAATTCGGTTCTACAGGGTA) [18]. Cyanine 3 was detected at excitation wavelengths of 510-560 nm and emission wavelength of 590 nm while DAPI was detected by excitation wavelengths of 330-380 nm and emission wavelength of 420 nm.

Histamine release assay
Rat mast cells were treated with 50 µm substance P (SP) (Selleck,Shanghai, China) for 30 min before collected supernatants to measure the amounts of secreted histamine using a histamine ELISA kit according to the manufacturer's instructions (Elabscience Biotechnology Co. Ltd, Wuhan, China). Values were measured using a spectrophotometer (Epoch, BioTECH, USA) and data expressed as nanograms per milliliter of histamine.

Statistical analysis
All statistical analyses were performed using SPSS 19.0 software (Chicago, IL, USA). Statistical tests ANOVA and a t-test were used in results as described. When ANOVA was used, a Tukey post hoc test was performed for means comparison. Two-sided P < 0.05 were considered signi cant.

Construction and evaluation of the KOA rat model
We constructed a KOA model by injecting monosodium iodoacetate into the left knee joint cavity of rats and compared this to a control group injected with the same volume of normal saline (NS). After 14 days we observed a series of pathological changes in rat synovial tissues in the KOA rats including roughness in the surface of articular cartilage, local defects and increases or decreases in joint uid along with obvious osteophytes found around the cartilage. In comparison, no signi cant pathological changes were evident in the NS group (Fig. 1A). In parallel, we assessed the pathological changes by examining H&E stained synovial sections. As anticipated, there were no obvious changes in the NS group tissues whereas a number of pathological changes were evident in the KOA group including articular cartilage defects, brous tissue proliferation and chondrocyte necrosis (Fig. 1B). Additionally, from the functional viewpoint, the treated leg in the KOA group appeared limp and the rats were reluctant to exercise the affected limb. Moreover, KOA group rats showed higher scores in pain stimulation response, gait, joint activity, and joint swelling than the NS group. Vernier caliper measurements also revealed that the diameter of left knee joints before and after treatment was increased in both NS and KOA groups (Fig. 1C). However, the joint diameter increases in the KOA group were signi cantly more than the NS group (P < 0.05), whereas the NS group changes were not statistically signi cant (P < 0.05). Together these data indicate that injection of MIA induced rat KOA-related cartilage damage, thus establishing the suitability of the model to study KOA.

Acupoint Sensitization Detection And Determination Of Sensitization State
As a frequently used acupoint for acupuncture treatments involving KOA, we selected ST36 as the acupoint for our sensitization studies. A von Frey electronic pain meter was used to detect the mechanical pain threshold of the relevant acupoints of each rat before and after the model treatments.
The calculated changes in pain threshold rates were used as the criterion for determining sensitization acupoints. As shown in Fig. 2A, there was no signi cant change in the ST36 pain threshold in the NS group, but notably the pain threshold in the KOA group was signi cantly decreased. This proposes that the ST36 acupoints in the KOA group were sensitized.
As described in the introduction, mast cell accumulation and degranulation are important signs of acupoint sensitization [4]. Interestingly, we observed a non-signi cant trend that the total number of mast cells in the ST36 acupoint was increased in the KOA group compared to the NS group (P > 0.05; Fig. 2B).
Since it is known that mast cells are concentrated in the ST36 acupoint [17] this fact may have affected the level of signi cance. Nonetheless, the levels of mast cell degranulation in the sensitized ST36 acupoints were signi cantly higher in KOA compared to the NS treatments, along with their degranulation rate (P < 0.05; Fig. 2C and 2D). Phenotypic examination using toluidine blue staining clearly revealed more degranulated mast cells in the skin at the ST36 acupoint in the KOA group compared to the NS group ( Fig. 2E and 2F). Together these data indicate that ST36 acupoint sensitization in KOA was associated with increased mast cell degranulation.

RNA-seq reveals differentially expressed miRNAs in the sensitized ST36 acupoint
We hypothesized that miRNAs would be likely mediators of acupoint sensitization. To test this, we employed transcriptome sequencing to identify differentially expressed miRNAs in skin tissues between sensitized ST36 acupoints in the NS and KOA groups. This approach identi ed a total of 8 miRNAs whose expression was signi cantly different between the KOA and NS groups. Of these, 1 miRNA was down-regulated, and 7 miRNAs were up-regulated in the KOA group. To verify the transcriptomic data, we randomly selected a subset of samples (n = 3/each group) and used RT-qPCR (n = 6/each group) to measure 4 of the differentially expressed miRNAs identi ed in the RNA-seq analysis, speci cally rno-miR-199a-3p,rno-miR-199a-5p rno-miR-205 rno-miR-214-3p, and rno-miR-543-3p. Using U6 snRNA as a reference gene we found that there was a similar tendency between the RNA-seq and RT-qPCR data for all 4 miRNAs, however, only rno-miR-205 and rno-miR-199a-3p were signi cantly changed (P < 0.05; Fig. 3A).
Next to investigate potential pathways involved in KOA modelling, we utilized the Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway analysis from DAVID. The predicted actions of the differentially expressed miRNAs included effects on genes involved in mast cell activation pathways such as the Fc epsilon RI signaling [26,27] (Fig. 3B). This prediction together with our ndings led us to focus on the potential role of rno-miR-199a-3p in acupoint sensitization.

High expression of rno-miR-199a-3p in the ST36 acupoint contributes to mast cell activation
Our preceding results established that rno-miR-199a-3p was highly expressed in the ST36 acupoint skin of the KOA group rats. In order to determine whether the increased expression was associated with mast cells, we utilized uorescence in situ hybridization (FISH) to reveal the cellular location of rno-miR-199a-3p. We observed that rno-miR-199a-3p uorescence signals were noticeably higher in mast cells in the KOA group skin compared to the NS group (Fig. 4A), suggesting increased rno-miR-199a-3p levels in mast cells in the KOA group. To con rm this notion, we extracted mast cells from ST36 acupoint skin from the KOA and NS groups and measured the relative levels of rno-miR-199a-3p using RT-qPCR. Indeed, these results con rmed that the expression level of rno-miR-199a-3p in KOA rats ST36 acupoint mast cells was signi cantly increased (Fig. 4B).
To further validate the role of rno-miR-199a-3p in mast cells, we transfected rno-miR-199a-3p inhibitors or NC controls into cultured mast cells. The results showed rno-miR-199a-3p expression was signi cantly reduced in inhibitor-treated mast cells compared to mock or NC-treated cells (Fig. 4C). Instructively, inhibition of rno-miR-199a-3p levels resulted in signi cantly decreased histamine release from mast cells after treatment with substance P (SP) (Fig. 4E), which is a commonly used technique to stimulate mast cells in the eld of acupuncture research [29,30]. Alternatively, rno-miR-199a-3p mimics were also transfected into rat skin mast cells to overexpress rno-miR-199a-3p. As anticipated, after transfection the measured levels of rno-miR-199a were signi cantly increased (Fig. 4D) and such overexpression promoted increased HA release from mast cells after treatment with SP (Fig. 4E). Taken together, these data indicate that upregulation of rno-miR-199a-3p signi cantly increases histamine release in mast cells, proposing a functional link between rno-miR-199a-3p and the activation of mast cells in the ST36 acupoint in KOA.

Mast cell activation through rno-miR-199a-3p results from inhibition of Nedd4
MicroRNAs function as post-transcriptional gene regulators by targeting speci c mRNAs to cause downstream effects on the levels of their corresponding proteins. We next used the TargetScan database to identify the likely targets of rno-miR-199a-3p in mast cells. Of the candidate genes identi ed, we selected Nedd4 for further investigation because of its known functions related to mast cell activation [31]. As shown in Fig. 5A, the potential miRNA binding site was located at bases 366 to 373 in the Nedd4 3' UTR. On this basis, we constructed pmiR-GLO luciferase reporter vectors containing the 3'-UTR of the Nedd4 mRNA with the predicted rno-miR-199-3p binding sequence intact (wild-type) or mutated as illustrated. Then to test whether Nedd4 was directly regulated by rno-miR-199a-3p, we transfected the wild-type or mutant vectors in combination with either rno-miR-199a-3p mimics or the control NC mimics into HEK-293 cells. We found that the introduction of the rno-miR-199a-3p mimics inhibited the luciferase reporter activity of the wild-type but not the mutant vector (Fig. 5B), indicating that rno-miR-199a-3p targets Nedd4 mRNA though the identi ed binding site. Substantiating this result, transfecting cultured mast cells with rno-miR-199a-3p mimics resulted in signi cant decreases in the levels of Nedd4 mRNA (Fig. 5C) and protein (Fig. D). These results indicate that rno-miR-199a-3p can directly inhibit Nedd4 expression by targeting Nedd4 mRNA.
6. Rno-mir-199a-3p Promotes Mast Cell Activation Via Nedd4 Lastly, we sought to verify whether the functional effects of rno-miR-199a-3p on mast cell activation were mediated by targeting of Nedd4. Towards this we performed rescue assays where the reduced levels of Nedd4 resulting from rno-miR-199a-3p inhibition were reversed by ectopic expression of Flag-tagged Nedd4. Notably, the overexpression of Flag-Nedd4 alone in cultured mast cells resulted in signi cant reductions in HA release. In contrast, simultaneous transfection of Flag-Nedd4 with rno-miR-199a-3p mimics predominantly blocked increases in HA release (Fig. 6A). Taken together, these data indicate that rno-miR-199a-3p promotes mast cell activation by regulating Nedd4 expression.

Discussion
Acupuncture has been used as an effective alternative therapy for knee arthritis worldwide. A variety of evidence shows that acupuncture can reduce the pain associated with knee osteoarthritis, improve knee joint function, and often cooperates with other drugs [32,33]. According to traditional Chinese medicine, acupoints used in clinical practice usually represent the re ex points of speci c diseases (i.e., sensitized acupoints) [3]. Extensive clinical research on acupuncture has made clear the regularity of acupoint selection in KOA treatment, and provided important insights for acupoint selection. On this basis, knee osteoarthritis represents an ideal condition to study the process of acupoint sensitization.
Acupuncture treatments for knee osteoarthritis are typically prescribed through several acupoints around the knee such as GB34 and ST36 [34]. For our study, we selected ST36 as the target acupoint for experimental analysis and found that the pain threshold of ST36 changed after KOA modeling in rats. Many studies on acupoint sensitization have shown this process is closely related to mast cells [4,14,21].
For example, in a gastrointestinal mucosa chemical injury model the Evans blue (EB) exudation points in corresponding acupoint skin tissues increased, along with mast cell aggregation and degranulation rates.
Our study provides additional support for this concept where we found de nitive evidence showing increased mast cell degranulation rates in ST36 sensitized acupoints after KOA modeling.
Mast cells are important immune cells [35,36] which are mainly derived from bone marrow precursor cells but circulate in the blood to reach target tissues where they differentiate and mature [37,38]. For example, mature mast cells are observed in the skin, nasal cavity and intestinal mucosa [37,39]. Mast cells respond to a variety of physical and chemical stimuli, such as viral infections, bacterial invasion, mechanical and heat stimulation, causing them to release cytokines and chemokines [40,41]. Additionally, mast cell activation also results in the release active substances such as histamine (HA), tryptase and 5-hydroxytryptamine (5-HT) [42,43]. Through exocytosis, these substances may promote local skin allergic reactions, from which pain sensitization occurs. The sensitization of acupoints is related to the aggregation and degranulation rate of mast cells with the release of trypsin, 5-HT and HA likely playing a role [44]. Moreover, others have provided clear evidence that neuropeptide release from mast cells in the local skin microenvironment is also important for acupoint sensitization. Substance P is a prokinetic brain gut peptide widely distributed in nervous system and gastrointestinal nervous system [45][46][47] and represents a signaling substance recognized by the nervous, endocrine, and immune systems. In mast cells, SP occurs in the context of a positive feedback loop where SP promotes mast cell degranulation, which in turn, release substances including SP to promote further mast cell degranulation in a time and dose-dependent manner. SP is also is of the substances that cause the meridian effect which appears to require the cooperation of mast cells [30]. Consistently, we found in our study that mast cells within the sensitized ST36 acupoints showed increased activation and degranulation. Further mechanistic studies also revealed that expression level of rno-miR-199a-3p in the ST36 skin tissue was signi cantly increased and speci cally associated with mast cells. We also modelled mast cell activation in vitro using by treating cells with SP and assessing degranulation through the release of HA. Using rno-miR-199a-3p inhibitors and mimics, the key nding was made that the expression levels of rno-miR-199a-3p positively in uenced mast cell degranulation rates. Furthermore, we showed this effect on mast cell degranulation and histamine release was attributed to rno-miR-199a-3p targeting of Nedd4 mRNA which inhibited Nedd4 protein expression. Therefore, the rno-miR-199a-3p-Nedd4 axis may be one of the key regulatory processes involved in acupoint sensitization.
In summary, our study proposes for the rst time that a miRNA-mediated regulatory mechanism is responsible for regulating acupoint sensitization. However, there are some limitations of the study that must be acknowledged. We judged the occurrence and degree of sensitization by the decrease of acupoint mechanical pain threshold and the change rate. However, whether the difference of miRNA expression can be used as the standard to judge acupoint sensitization remains to be answered. Nor did we evaluate all other miRNAs identi ed by this analysis. Thus in order to determine the diagnostic criteria of sensitization in animal experiments, it will be necessary to further carry out larger sampling and adopt additional research methods. Furthermore, while our research suggests that miR-199a-3p was associated with mast cells and promotes mast cell degranulation, its origin is still unknown. It has been reported that miR-199a-3p also exists in exosomes[48-50] and therefore miR-199a-3p could conceivably be transported to the ST36 acupoint by exosomes. For example, an attractive hypothesis is that the increased rno-miR-199a-3p originates from secretions in the injured knee joint and is transported via exosomes to mast cells in the ST36 skin acupoint. However, this hypothesis needs further research.

Collusion
In this study,We found that ST36 acupoints in the KOA group were sensitized. Rno-miR-199a-3p was highly expressed in the sensitized ST36 acupoint and its expression was associated with mast cells.
Overexpression of rno-miR-199a-3p increased mast cell histamine. Mechanism, rno-miR-199a-3p may promotes mast cell activation by regulating Nedd4 expression. Together, our study establishes that mast cells participate in the process of acupoint sensitization, and further reveals the rno-miR-199a-3p-Nedd4 axis may be one of the key regulatory processes involved in acupoint sensitization. Availability of data and materials The datasets used and/or analysed during the current study are available from the corresponding author on reasonable request.

Con icts of Interest
The authors declare no con icts of interest.   Rno-miR-199a-3p inhibits Nedd4 expression by targeting the 3' UTR region of Nedd4 mRNA. A. TargetScan database analysis identi ed a potential binding site for rno-miR-199a-3p in the Nedd4 3'-UTR.
As illustrated, wild-type and mutant reporter constructs were prepared in the pmiR-GLO reporter vector. B. Luciferase reporter assay conducted in HEK-293 cells after transfection with the wild-type or mutant pmiR-GLO reporter constructs in combination with either rno-miR-199a-3p mimics or control NC mimics.