Oxytocin increases itch sensitivity of mice in late pregnancy and its peripheral receptor mechanism of itch-promoting effect

Oxytocin (OXT) is secreted in a large amount during the middle and late pregnancy. Except for the regulation of functions related to childbirth, OXT is involved in the regulation of cognition, social behavior, addiction, pain and so on. Our aim is to confirm the increase of OXT content in mice in late pregnancy is the main cause of itch during pregnancy and observe whether exogenously administered OXT can induce or increase itch sensitivity. The research shows that itch sensitivity of mice increased significantly in late pregnancy and basically returned to normal one day after delivery. The number of OXT-positive neurons in paraventricular nucleus (PVN) and the content of OXT in serum of the late pregnant mice increased significantly, and decreased sharply after delivery. Intradermal injection of low concentration of OXT (0.2 nmol/L) could not induce scratching behavior in mice, but high concentration of OXT (5 nmol/L, 10 nmol/L) could do this in a dose-dependent manner. Low concentration of OXT significantly increased the itch sensitivity to histamine. Intradermal injection of oxytocin receptor (OXTR) or arginine vasopressin-1a receptor (AVPR1A) antagonist did not affect histamine-induced scratching behavior, but both reversed the increase of itch sensitivity in late pregnant mice or the facilitated itch sensitivity by OXT. Study suggests that both endogenous and exogenous increases in OXT can increase the body ’ s sensitivity to itch, and even induce itch directly. Pruritus during pregnancy is closely related to the increase of OXT content in vivo. In the periphery, the itch-promoting effect of OXT is mediated by OXTR and AVPR1A.


Introduction
Itch is an unpleasant sensation that can cause a desire or reflex to scratch, and often caused by some skin diseases, neurological diseases, systemic diseases, or mental illnesses (Yosipovitch et al., 2018). Itch can serve as a protective mechanism for the human body like pain, tactile sense, and thermesthesia, which indicating the body is experiencing external harmful stimuli or suffering from some kind of diseases, such as itch caused by mosquito bites or neurodermatitis. The persistent chronic itch can cause serious harm to both the physical and mental, and there is still a lack of efficient and low side effect therapeutic drugs in clinical treatment (Nowak and Yeung, 2017). Pruritus is also one of the common complaints among women in the middle and late stages of pregnancy, with approximately 15% of pregnant women experiencing itch (Bergman et al., 2013). The degree of itch during pregnancy varies from mild to severe. Severe itch not only affects quality of sleep and life, but may also potentially leading to or exacerbating depression, and even causing harm to the fetus (Bechtel, 2018). Pruritus during pregnancy is mainly caused by some pregnancy-specific conditions, such as polymorphic eruption of pregnancy (PEP), atopic eruption of pregnancy (AEP), pemphigoid gestationis (PG), and intrahepatic cholestasis in pregnancy (ICP), but it may also be related to other diseases or even normal physiological changes during pregnancy (Himeles and Pomeranz, 2022;Stefaniak et al., 2022). In addition, some systemic diseases (such as kidney disease, liver disease, endocrine disease, etc.), and drugs used in the treatment process, can also cause pruritus during pregnancy (Mian et al., 2020). More often, the cause of pruritus is unknown, and even there is no skin appearance in the early stage of the disease (Stefaniak et al., 2022). Some pruritus resolves spontaneously in the short term, while others continue until the termination of the pregnancy. Due to the limitation of medication during pregnancy, the clinical treatment of pruritus during pregnancy is very limited (Genovese et al., 2020). In addition, the transmission and regulation mechanism of itch sensation are not very clear, which also poses a great limitation to the development of clinical drugs and medication strategies.
It is widely accepted that pregnancy pruritus may be related to increased hormone secretion during pregnancy (Szczech et al., 2017). Oxytocin (OXT) is a hormone secreted in large amounts in the middle and late stages of pregnancy, which is mainly involved in the regulation of labor and lactation, as well as cognition, addiction, social behavior, and pain (McCormack et al., 2020). Previous studies on OXT analgesia have reported that intracerebroventricular injection of OXT can induce scratching behavior in mice (Schorscher-Petcu et al., 2010), and intrathecal OXT administration can also induced transient scratching behavior in mice (Meisenberg and Simmons, 1982;Van Wimersma Greidanus et al., 1990), which largely interferes with animal behavioral detection. Therefore, intraperitoneal administration of OXT has been recommended in studies of analgesic (Martinez-Lorenzana et al., 2008). Nevertheless, OXT has not been associated with pruritus, let alone gestational pruritus, and attempts have even been made to ignore the scratching behavior induced by OXT. Therefore, it is important to determine whether OXT can facilitate or induce itch and by what mechanisms, and whether OXT, which is secreted in increased amounts during the middle and late stages of pregnancy, is the main cause of pregnancy pruritus.
Oxytocin receptor (OXTR) is widely distributed in the mammalian nervous system and plays a physiological role by binding to OXT (Jurek and Neumann, 2018). Because the structure of OXT and arginine vasopressin (AVP) is extremely similar, arginine vasopressin-1a receptor (AVPR1A) can also bind to OXT (Swanson and Kuypers, 1980;Chini and Manning, 2007).
In view of the above reasons, we conducted the following studies using morphological, behavioral, and ELISA methods: a) observing the changes of itch sensitivity in mice during pregnancy and postpartum; b) detecting the changes of OXT levels in serum and PVN of mice at different stages of pregnancy and postpartum; c) determining the effects of different concentrations of exogenous OXT on itch sensitivity in mice; d) clarifying the peripheral receptor mechanism mediating the pruritic effect of OXT by the application of OXTR or AVPR1A antagonists. Through these studies, the relationship between OXT and pregnancy pruritus will be clarified, and the peripheral receptor mechanism of OXT in regulating pruritus will also be preliminarily clarified. This study will provide new theoretical basis and ideas for the treatment of pruritus, especially for clinical pregnancy pruritus.

Animals
The female, male and pregnant mice used in this experiment were all adult C57BL/6 J mice that were purchased from the Laboratory Animal Center of Air Force Medical University (female and male mice were 8-10 weeks old, body weight was 20-25 g; Pregnant mice with variable body weight). In rearing, breeding, and related experiments, ambient temperature was maintained between 20 and 25 ℃, humidity between 55% and 65%, and a circadian rhythm of 6 am-6 pm with lights on and 6 pm-6 am with lights off was used to exclude the influence of environmental changes on the results of relevant experiments. Animals were fed and watered unlimited, and no more than six mice per cage were used to ensure adequate activity space. All the methods and procedures in this experiment were approved by the Animals care and Use Committees of Air Force Medical University (Xi'an China).

Behavioral Experiments
24 h before the establishment of the acute itch model, the hair on the back of the neck of the mice was removed by shaver. And the mice were placed and habituated in experimental site for 1 h for three consecutive days before the behavioral experiments. The pruritic agent histamine (His, H7250, Sigma) or chloroquine (CQ, C6628, Sigma) was injected intradermally into the back of the neck of the mice at a concentration of 10 mg/ml in a 15 μl volume using an insulin injection needle while the mice were awake. After that, the mice were placed in the designated area for video collection for 30 min, and the scratching times of the mice were statistically analyzed. On the basis of this acute itch model, we mixed other related reagents with His at a certain concentration to conduct behavioral experiments, and the data were obtained and analyzed statistically.

Preparation of pregnant mice at different stages of pregnancy
In order to ensure the exact time of pregnancy of the pregnant mice, the pregnancy time of the pregnant mice used in the experiment was accurately calculated. 24 (8-week-old) male (n = 6) and female (n = 18) mice were bought from Laboratory Animal Center and adapted to the environment for 3 days. At 6 PM on the third day, the mice were put in a cage with a male to female ratio of 1:3 and got pregnant. After 24 h, the male and female mice were separated, and the female mice were recorded as the first day of pregnancy. All female mice were subjected to subsequent experiments at the same time, and finally 8 female mice became pregnant and gave birth, served as the pregnant mice group. The acute itch model was established and the number of scratches was counted on the 5th (early pregnancy), 10th (middle pregnancy), 15th (late pregnancy) pregnancy days, and the 1st, 5th and 7th postpartum days. The day when the mice gave birth was recorded as the first postpartum day. At the same time, 8 non-pregnant female mice were selected as the control group to establish the acute itch model at the corresponding time points ( Fig. 1-A).

Immunofluorescence staining
Animals were administered in deep anesthesia by intraperitoneal injection of sodium pentobarbital (40 mg/kg body weight). The mice were then fixed by perfusion through the circulatory system with 0.1 M PB (phosphate buffer) (pH 7.2-7.4) containing 4% (w/v) paraformaldehyde (80096618, SCR). The perfused fixed mice were craniotomy, and the whole brain tissue was removed intact, placed in a paraformalin-fixed solution for 4 h, and then dehydrated in 0.1 M PB solution containing 30% sucrose (10021418, SCR). The removed brain tissues were trimmed and embedded with OCT (Leica, Germany), and then coronal sections were cryosectioned to produce 30 µm sections. After rinsing, sections containing PVN were picked out for subsequent immunofluorescence staining.

Enzyme linked immunosorbent assay (ELISA)
Animals were administered in deep anesthesia by intraperitoneal injection of sodium pentobarbital (40 mg/kg body weight). The eyeball of the mice was pulled out with ophthalmic forceps, and the ophthalmic artery blood was collected. The blood samples were allowed to stand and centrifuged, and the supernatant was taken for detection. Mice were killed by cervical dislocation after blood collection. Mouse oxytocin (OXT) ELISA kit (CSB-E09245m, CUSABIO) was used to detect the concentration of OXT in the serum of mice, and the professional statistical software Curve Expert was used to draw the standard curve and calculate the sample concentration.

Statistic analysis
All the experiments were conducted in a double-blind experiment, and the experimental results were analyzed by SPSS 20.0. Measurement data were expressed as mean ± standard error (Mean ± SEM), and statistical analysis was performed using One-Way ANOVA or Two-Way ANOVA. Dunnett's t test or SNK-q test was used for inter-group comparison if the overall result of One-Way ANOVA was statistically significant, and Bonferroni post test was used for inter-group comparison if the overall result of Two-Way ANOVA was statistically significant. P < 0.05 was considered statistically significant.

The itch sensitivity of mice increased in late pregnancy
Before the establishment of acute itch model, we observed the spontaneous scratching of normal female mice and late pregnant mice. The results showed that compared with normal female mice, late pregnant mice did not show obvious spontaneous scratching behavior ( Fig. 1-B t = 0.9792, P = 0.3506, Student t-test). However, for pruritus induced by intradermal injection of histamine (His), there were significant differences in scratching behavior among different pregnant mice group ( Fig. 1-C F (3,28) = 11.48, P < 0.005, One-Way ANOVA). Compared with the normal female mice, the scratching number of 5-day (P = 0.7283, Dunnett's t test) and 10-day (P > 0.9999, Dunnett's t test) pregnant mice was not significantly different, but the scratching number of 15-day pregnant mice was significantly increased (P < 0.0005, Dunnett's t test). The results suggested that the sensitivity of the mice to itch did not change in early pregnancy, but significantly increased in late pregnancy (15 days after pregnancy).
Both intradermal injection of His and chloroquine (CQ) significantly increased the scratching number of the late pregnant mice compared with normal male (Fig. 1-D F (2,21) = 49.29, P < 0.0001, One-Way ANOVA) and female mice ( Fig. 1-E F (2,21) = 10.82, P < 0.005, One-Way ANOVA). This suggests that the itch sensitivity to both His and CQ is increased in late pregnant mice. It was also observed that the scratching number induced by His injection in normal female mice was significantly higher than that in normal male mice (q=3.702, P < 0.05, SNK-q). It is suggested that the itch sensitivity of female mice is higher than that of male mice, and this phenomenon was not discussed further in this paper.

With the significant decrease of OXT content after delivery, the itch sensitivity of mice rapidly returned to normal level
In order to explore the changes of postpartum itch sensitivity in mice, acute itch models were established in late pregnancy and 1, 5 and 7 days postpartum. Another 10 pregnant mice were obtained by the same method as described before for this part of the experiment. The results showed that the scratching number of postpartum mice was significantly lower than that of the late pregnancy group (Fig. 2-A F (1.531,   13.78) = 29.92, P < 0.0001, One-Way ANOVA), and the scratching number of mice decreased significantly on the first day after delivery (q=8.601, P = 0.0009; Tukey's post hoc multiple comparison test), and the 5-day postpartum group and the 7-day postpartum group also showed a significant decrease (q=7.604, P = 0.0021; q=9.667, P = 0.0004, Tukey's post hoc multiple comparison test). However, the scratching number among different postpartum time groups had no significant difference. These results suggest that the increased itch sensitivity in late pregnant mice rapidly returns to normal level after delivery.
OXT is one of the major hormones involved in the physiological process of pregnancy. In order to verify that the increase of itch sensitivity in late pregnant mice is caused by the increase of OXT levels in vivo, we first measured the levels of OXT in the paraventricular nucleus (PVN) and serum of late pregnant mice, and compared them with the data of normal and postpartum female mice. The results showed that the number of OXT-ergic neurons in the PVN of the late pregnant mice ( Fig. 2-C) was significantly higher than that of the normal female mice (Fig. 2-B), and the number of OXT-ergic neurons in the PVN of the  and female mice. * * P < 0.01, * ** P < 0.001, Two-way ANOVA, n = 6. postpartum mice (Fig. 2-D) was significantly decreased (Fig. 2-E F (2,6) = 49.18, P = 0.0002, One-Way ANOVA). Compared with normal male, normal female, and 5-day postpartum female mice, the serum OXT concentration in late pregnant mice was significantly increased and decreased rapidly after delivery (Fig. 2-F F (3,14) = 37.96, P < 0.0001, One-Way ANOVA).

High concentrations of OXT can directly induce pruritus, and low can facilitate pruritus
To investigate the effect of OXT on itch sensitivity in normal mice, different concentrations (0.2, 5, 10 nmol/L) of OXT (abs815864 Absin Shanghai) were intradermally injected into the back of neck of male mice. The results showed that intradermally injected OXT at a low concentration (0.2 nmol/L) did not induce significant scratching behavior in mice, while injection of OXT at higher concentrations (5 and 10 nmol/L) did (Fig. 3-A F (3,20) = 110.1, P < 0.0001, One-Way ANOVA), and the scratching behavior induced by high OXT concentration could last for more than 30 min (Fig. 3-B). These results indicate that high concentrations of OXT can directly induce pruritus in a concentrationdependent manner, but low concentrations of OXT cannot directly induce pruritus.
To test whether low concentration OXT can affect the sensitivity of mice to His, normal male and female mice were intradermally injected with His (His group) or a mixture of His and low concentration (0.2 nmol/L) OXT (His+OXT group), respectively, and video recording was performed and scratching behavior was statistically analyzed. The results showed that both male and female mice in the His+OXT group had a significantly increased scratching number compared with the His group ( Fig. 3-C t = 3.751, P = 0.0025; t = 4.402, P = 0.0006, Student ttest). This suggests that although low concentration OXT does not directly induce pruritus, it can facilitate the pruritus induced by His injection. At the same time, we again observed a higher sensitivity to His-induced pruritus in female mice than that of male mice (Fig. 3-C  F (1,20) = 9.466, P < 0.0001, Two-way ANOVA).

Itch-promoting effect of OXT is mediated by OXTR and AVPR1A
OXT has only one membrane receptor, oxytocin receptor (OXTR), but due to the structural similarity between OXT and arginine vasopressin (AVP), OXT can also bind to arginine vasopressin receptor 1A (AVPR1A), that is, OXT can bind to both two membrane receptors in Fig. 4. The pro-pruritic effect of OXT is mediated by OXTR or AVPR1A. (A) The scratching behavior of normal male mice (Male) did not change after intradermal injection of both His and OXTR antagonist (or AVPR1A antagonist). (B) The scratching behavior of normal female mice (Female) did not change after intradermal injection of both His and OXTR antagonist (or AVPR1A antagonist). (C) OXTR and AVPR1A antagonists can inhibit the increased itch sensitivity in late pregnant mice (Preg). * ** * P < 0.0001, One-Way ANOVA, n = 9. (D) The facilitation of His-induced itch by OXT was inhibited by intradermal injection of either OXTR or AVPR1A antagonists in normal male mice. * ** *P < 0.0001, One-way ANOVA, n = 10. (E) The facilitation of His-induced itch by OXT was inhibited by intradermal injection of either OXTR or AVPR1A antagonists in normal female mice. * P < 0.05, * * P < 0.005, One-way ANOVA, n = 10. (F) The sensitivity of itch was slightly decreased in normal female (Female) and pregnant (pregnant) mice after intraperitoneal injection of non-specific estrogen antagonist. * P < 0.05, * * P < 0.01, * ** *P < 0.0001, Two-way ANOVA, n = 6. vivo.
To explore the receptor that mediates OXT's pruritic effect, antagonists of OXTR (L-368899, 3377, Tocris) or AVPR1A (anti-AVPR, d(CH2) 5[Tyr(Me)2]AVP, 2641, Tocris) were added to His injected during the establishment of acute itch model. 27 pregnant mice were prepared again in the same way as before and randomly divided into 3 groups (Preg+His, Preg+His+L368 and Preg+His+anti-AVPR) for experiment. At the same time, 10 male and 10 female mice were selected for three times of experiments with different groups of pruritic agents (+His, +His+L368 and +His+anti-AVPR), and the interval between each experiment was 1 week. The results showed that antagonists of either OXTR or AVPR1A had no effect on itch sensitivity in normal male and female mice by His injected (Fig. 4-A F (2,27) = 0.3262, P = 0.7245, Oneway ANOVA; 4-B F (2,27) = 0.3262, P = 0.9241, One-way ANOVA), but both antagonists could inhibit the increase of itch sensitivity in late pregnancy mice (Fig. 4-C F (2,24) = 22.49, P < 0.0001, One-way ANOVA). This suggests that OXT mediates the increase of itch sensitivity in late pregnancy mice through OXTR and AVPR1A.
As the level of estrogen is also significantly increased during pregnancy, especially in late pregnancy, in order to test whether the increase of itch sensitivity is also related to the increase of estrogen, we applied a non-selective antagonist of estrogen (ICI 182780, 1047 t, Ocri). The results showed that ICI 182780 could reduce the pruritic sensitivity of His in normal female mice as well as in late pregnant mice (Fig. 4-F  F (1,20) = 150.6, P < 0.0001, Two-way ANOVA). However, after ICI 182780 treatment, the pruritic sensitivity of pregnant mice was still relatively high, and the scratching number of histamine injection was still much higher than that of normal female mice (Fig. 4-F His+Vehicle: t = 8.543, P < 0.0001; His+ICI 182780: t = 8.815, P < 0.0001, Bonferroni post tests). Therefore, estrogen is not the main factor of increased itch sensitivity in late pregnancy mice. At the same time, it also suggests that estrogen may be the key factor responsible for the higher itch sensitivity of female mice than that of male mice, that is, estrogen has pro-pruritic effect.

Discussion
Clinically, it can be observed that pruritus during pregnancy can be relieved immediately once the pregnancy is over in most gestational pruritus patients. Therefore, we speculate that gestational pruritus may be related to changes in the content of a specific substance during pregnancy. The duration of pregnancy in mice is generally 19-20 days. In our study, we observed that the sensitivity of pregnant mice to histamine-induced pruritus was significantly increased after 15 days of pregnancy and returned to normal level at postpartum 1 day, which was similar to the characteristics of human pruritus during pregnancy. In addition, compared with normal male and female mice, the sensitivity to histamine and chloroquine in late pregnancy mice is increased, which determines the types of pruritus during pregnancy that are both histamine-dependent and histamine-independent.
During pregnancy, various hormones change significantly in the body, among which OXT comes into our sight as a hormone that increases more in the middle and late pregnancy. So far, there are more than 190 reports in the literature that OXT has a significant analgesic effect both centrally and peripherally Goodin et al., 2015;Tracy et al., 2015;Xin et al., 2017). Eliava found that small OXT-ergic neurons in the PVN can exert analgesic effects through peripheral and central modes of action (Eliava et al. (2016). Although the anatomy of afferents to small OXT-ergic neurons is unknown, ascending sensory projections from the spinal cord as well as ascending projections from sympathetic neurons appear to activate such neurons (Strack et al., 1989;Huang and Weiss, 1999;Gerendai et al., 2001;Affleck et al., 2012;Goodin et al., 2015). In addition, acute pain stimulation can activate OXT neurons and promote the release of OXT into peripheral blood, resulting in peripheral analgesia. Ablation OXTR-ergic neurons by intrathecal injection of OXT-saponins complex (OXT-Sap) produced hyperalgesia (Matsuura et al., 2016). In addition to the peripheral mechanism, OXT is involved in the regulation of pain in the forebrain, midbrain and brain stem. Although the involvement of PVN OXT-ergic neurons in the regulation of itch has hardly been mentioned in previous literatures, it has indeed been reported that intrathecal injection of OXT can induce transient scratching behavior in mice (Yaksh et al., 2014;Xin et al., 2017). It's reported recently that intrathecal OXT induced itch-scratching behaviors by activating spinal OXT receptors which expressed on the GRP neurons, and activates GRP/GRPR pathway to trigger itch-scratching behaviors in mice (Guo et al., 2020). Given that OXT can be involved in the regulation of pain and potentially the regulation of itch, we conducted a series of studies on OXT as a possible factor inducing itch during pregnancy.
To investigate the changes of OXT in pregnant mice, we observed the changes of OXT-ergic neurons in PVN by immunohistochemical staining. The results showed that the number of PVN OXT-ergic neurons increased significantly during the late pregnancy, and then decreased rapidly on the first postnatal day. Since OXT cannot cross the blood-brain barrier, the effects of OXT in the periphery are mainly exerted by the release of PVN-synthesized OXT into the blood by the neurohypophysis. Next, the serum OXT concentration in different groups of normal female mice, late pregnant mice, and mice at different postpartum periods was detected by ELISA. Similar results with immunostaining were obtained: the serum OXT concentration in late pregnant mice was significantly increased, and then rapidly decreased to normal level after delivery. Combined with the analysis of the itch behavior of mice, it is suggested that the changes of OXT levels in mice during different pregnancy and postpartum were positively correlated with the changes of itch sensitivity.
In order to confirm that the increase of itch sensitivity during the late pregnancy is indeed caused by the increase of OXT concentration in vivo, the effect of exogenous increase of OXT content on itch sensitivity in normal mice was observed. The results showed that intradermal injection of high concentration OXT could directly induce pruritus in normal mice in a concentration-dependent manner. Intradermal injection of OXT at low concentrations did not induce pruritus directly, but it increased itch sensitivity in both male and female mice. These results suggest that OXT can indeed facilitate pruritus in the periphery, and even induce pruritus directly.
Previous studies have observed the expression of OXTR and AVPR1A in skin tissues by immunofluorescence double labeling, and the coexistence of the two receptors in nerve terminals has been observed (Manzano-Garcia et al., 2018). This provided morphological evidence for pruritus induced by peripheral injection of high concentrations of OXT and for that low concentrations of OXT facilitated pruritus induced by pruritic agent. Since OXTR and AVPR1A are both distributed in the periphery and co-expressed to some extent, which receptor mediates the pro-pruritic effect of OXT? Or are both receptors involved? Using a behavioral pharmacology approach, we injected OXTR or AVPR1A specific antagonists intradermally to observe the effect on itch behavior of mice. The results showed that the two antagonists had no effect on the pruritus induced by intradermal injection of histamine in normal male and female mice, but they could reverse the increased itch sensitivity in pregnant mice. Moreover, after OXT was given with low concentration to increase the itch sensitivity of mice, intradermal injection of these two receptor antagonists could reverse the pro-pruritic effect of OXT. These results suggest that the pro-pruritic effect of OXT may be mediated by OXTR and AVPR1A receptors at the peripheral level. Ru-Long Li has reported that OXTR did not mediate the pro-pruritic effect of OXT (Rulong et al., 2019). But in his study, the concentration of histamine is 10 times lower than that used in our experiments, and the number of scratches of mice is also lower a lot. This may account for the different results.
During the experiment, we also found that when the same pruritic agent was given, the number of scratches of female mice was significantly higher than that of male mice, and when estrogen receptor antagonist was given, the number of scratches decreased, which proved that estrogen had a regulatory effect on pruritus, and the specific mechanism needs to be further studied.
Our study demonstrated that OXT is a major factor in the pathogenesis of pruritus in late pregnancy, and the increased OXT in late pregnancy is the main reason of pruritus, and this pro-pruritic effect is mediated by OXTR and AVPR1A receptors. Our findings provide new insights into the clinical management of pruritus during pregnancy and other pruritus disorders associated with OXT.

Author Contributes
Gang Yang, Zhen-Yu Wu, Er-Ping Xi and Hui Li were involved in the design of the study. Gang Yang, Xiao-Dong Wang, Lu-Ying Chen and Zhen-Yu Wu were involved in the behavior tests. Gang Yang and Xiao-Dong Wang were involved in the ELISA test. Gang Yang and and Lu-Ying Chen were involved in the Immunofluorescence staining. The draft manuscript was written, reviewed, and edited by Gang Yang, Zhen-Yu Wu, Er-Ping Xi and Hui Li. All authors contributed to and approved the final manuscript.

Funding
This work was supported by National Natural Science Foundation of China [Grants number 32171137].

Declaration of Competing Interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.