During pregnancy, intrahepatic cholestasis (ICP) causes excessive pruritus without a rash and elevations of total bile acids (TBAs) [1]. Incidences of ICP vary by geographic location and ethnicity, ranging from 0.2% to 2% [2]. After delivery, the mother’s symptoms usually disappear, whereas foetal exposure to ICP results in perinatal and long-term complications, including premature delivery, neonatal respiratory distress syndrome, sudden intrauterine foetal demise, stillbirth, and metabolic syndrome later in life [3, 4]. However, the aetiology of ICP remains unclear. Researchers recently found that the placenta’s core genes are largely responsible for regulating immune function; expression of these genes is upregulated in mild ICP, and in severe cases the expression is further increased, which indicates immune system might play a major role in ICP [5]. Recently, the T helper 17 (Th17) cell in pregnant women has received attention. ICP patients’ peripheral blood and placentas were reported to be significantly elevated in interleukin (IL)-17, a cytokine of Th17 cell [6, 7].

There are both positive and negative impacts on the immune system caused by Th17 cell, a relatively new subset of CD4${}^{+}$ T cells. It has been shown that excessive levels of Th17 cell and Th17-related cytokines are linked to disorders of pregnancy, including preeclampsia and repeated abortions, which could disrupt maternal and foetal immunity [8, 9, 10]. Many molecules play important roles in regulating Th17 cell differentiation and function. IL-17A is a proinflammatory cytokine and the major effector cytokine in Th17 cell. Additionally, as a Th17 signature transcriptional factor, retinoic acid-related orphan nuclear receptor gamma t (ROR$\gamma$t) is imperative for Th17 differentiation, which regulates the expression of key genes of Th17, like IL-17A and interleukin (IL)-23R [11]. Moreover, signal transducer and activator of transcription 3 (STAT3) is needed to differentiate Th17 under activation by interleukin (IL)-6 and transforming growth factor (TGF)-$\beta$ [12]. STAT3 contributes to the differentiation of Th17 cell via multiple pathways, including by directly binding to the ROR$\gamma$t gene (RORG) to promote its expression [13], inhibiting forkhead box protein (Foxp) 3 expression [12], directly activating IL-17A promoter [14]. Th17 cell differentiation involves three phases: the early phase, characterised by a transient induction of immune response pathways, including the expression of STAT3; the intermediate phase, marked by ROR$\gamma$t expression; and the late phase, characterised by the induction of signature cytokines of Th17 cell, including IL-17A [15]. Although several studies have examined IL-17’s role in ICP, few researchers have paid attention to the roles of STAT3 and ROR$\gamma$t in the modulation of Th17 differentiation and function in ICP, and whether a consistent and predictable relationship exists among STAT3, ROR$\gamma$, and Th17 cell in ICP and normal pregnancy. Therefore, the purpose of our study was to investigate the alterations of Th17 cell in placenta and peripheral blood of patients with ICP and to analyse their correlations with STAT3, ROR$\gamma$, and IL-17.

Women were recruited from August 2020 to October 2020 at West China Second Hospital, affiliated with Sichuan University. Our study was approved by the hospital’s ethics committee. Clinical information and samples were anonymized for statistical workup. Pregnancy healthcare profiles were created for all patients and they underwent routine pregnancy checks at our hospital. In the experimental and control groups were nine patients with singleton ICP and nine patients with normal-term singleton pregnancy,respectively. Delivery of all patients in both groups was by caesarean section. Based on Chinese Medical Association’s “Guidelines for diagnosis and treatment of intrahepatic cholestasis of pregnancy (2015)”, ICP was diagnosed.

Inclusion criteria for ICP group included the following: elevated TBA or aminotransferase; signs of prutitus or jaundice; no history of skin disease, gallstones, cholecystitis, liver cirrhosis, autoimmune diseases, acute and chronic infectious diseases, no history of infection with hepatitis virus, Epstein Barr virus, and cytomegalovirus; without any other obstetric complication. For the control group, the following criteria were used: normal liver or renal function; no history of gallstones, cholecystitis, liver cirrhosis, autoimmune diseases, acute and chronic infectious diseases, without history of infection with hepatitis virus, Epstein Barr virus, cytomegalovirus are not permitted; without any obstetrics complication.

Before the operation, 2 mL elbow venous blood was collected.Placental tissue was collected within 5 min of delivery. We incubated the paraffin-embedded placental sections with primary antibodies: anti-STAT3 (1:150), anti-IL-17A (1:200) (Proteintech Group, Chicago, IL, USA), and anti-ROR$\gamma$t (1:50) (Merck-Millipore, Bedford, MA, USA). Subsequently, these sections incubated with sheep anti-rabbit IgG (Biyuntian Biotechnology, Shanghai, China).

We homogenized placental tissues in ice-cold lysis buffer. Protein were subjected to sodium dodecyl sulphate–polyacrylamide gel electrophoresis and transferred to polyvinylidene difluoride membranes (STAT3: 100 V, 90 min; ROR$\gamma$t: 100 V, 60 min; IL-17A: 40 V, 90 min). We blocked the membranes in Tris-buffered saline and incubated the membranes with the primary and secondary antibody in turn.

The whole blood was diluted using RPMI 1640 medium and incubated with stimulant and blocker in turn. The blood was then incubated for 5 h at 37 °C with 5% CO${}_2$. Subsequently, the blood was incubated with APC-conjugated anti-human CD4 antibody (Biolegend, CA, USA), fixed with Fix Medium A (Fix & Perm kit, Invitrogen, CA, USA). After discarding the supernatant, the cells were permeabilised with Permeabilization Medium B (Fix & Perm kit, Invitrogen), incubated with FITC-conjugated anti-human IL-17A antibody (Biolegend).

Analysis of the data were conducted using IBM SPSS 23 software and R version 4.0.4 (IBM corp., Armonk, NY, USA). For the quantitative variables, the results were presented as the mean value $\pm$ standard deviation and the results were compared using t-tests for normally distributed variables or equivalent non-parametric tests otherwise. Chi-square tests was used to compare the data for categorical variables. For the correlation matrix, Pearson’s parametric correlation test was utilized. Using Pearson correlation scatterplots, we examined the correlations among the variables within groups. Two-sided p-values 0.05 were considered statistically significant in all tests.

Our study highlights the alters of STAT3, ROR$\gamma$t and Th17 cell in ICP along with the correlations between Th17 cell and ICP. We found that the ICP group had substantially elevated placental STAT3, ROR$\gamma$t, and IL-17A protein expression levels, as well as a greater peripheral blood Th17 cell count, relative to those in the control group. Our finding of an increased count of Th17 cell is consistent with the finding of previous studies [7, 16, 17]. Additionally, placental IL-17A protein showed significantly positive relationships with placental STAT3 and ROR$\gamma$t expression in the control group; in contrast, the ICP group did not show such a correlation. This indicated that the differentiation and function of Th17 cell in ICP differed from those in normal pregnancy. Our results can thus provide novel insights into the correlations between Th17 cell and ICP.

As a key player that defends against external pathogens and mediates immune rejection in the host, Th17 cell are vital in mediating maternal-foetal immunity. Recent studies have shown that an increased Th17 cell count is associated with ICP as it disrupts the delicate immunity and tolerance balance at the maternal-foetal interface [6, 7]; however, the mechanism underlying differentiation and functional regulation of Th17 cell in ICP remains unclear. We found that STAT3/ROR$\gamma$t/IL-17A proteins were mainly distributed in the trophoblast cells of the placenta, which is an important component of maternal-foetal immunity. Our study suggested that Th17 cell might be closely related to the maternal-foetal immune response. This result is consistent with those reported in previous studies [6, 7].

STAT3 is indispensable for the differentiation of Th17 cell. Our study revealed that STAT3 protein was distributed in the nucleus and cytoplasm of trophoblasts. Shochet et al. [18] found that STAT3 promoted the invasion of trophoblast cell and facilitated embryo implantation in primary pregnancy. Additionally, Ye et al. [19] demonstrated that STAT3 is mostly located in trophoblasts and ICP participants have higher STAT3 levels compared to controls.Our study demonstrated increased expression levels of STAT3 protein in the placenta of the ICP group, which is consistent with earlier studies. STAT3 can regulate Th17’s differentiation via directly binding to the ROR$\gamma$t encoding gene and promoting its expression [13]. Our study confirmed positive correlations between placental STAT3 and ROR$\gamma$t were presented in ICP and control groups.

ROR$\gamma$t is the lineage-specific transcription factor expressed in Th17 cell, which is required to initiate the Th17 differentiation and maintain Th17 cell increase [20]. ROR$\gamma$t${}^{-/-}$ mice lack tissue-infiltrating Th17 cell, and autoimmune disease is alleviated in this model [21]. Here, we demonstrated there was higher level of placental ROR$\gamma$t protein in the ICP group compared to control group. Furthermore, placental IL-17A protein showed remarkably positive relationships with placental STAT3 and ROR$\gamma$t protein in control group, but not in ICP group. IL-17 is a specific effector of Th17. IL-17A, IL-17B, IL-17C, IL-17D, IL-17E, and IL-17F are six fellows of IL-17 family [22]. Among them, the family prototype is IL-17A. IL-17F and IL-17A share high homology, and their coding genes are located in the same chromosome region; however, IL-17F has poor stability. A complex network of cytokines and transcription factors was required for Th17 cell differentiation and function. Du et al. [23] reported there were substantially increased levels of IL-6 but decreased levels of TGF-$\beta$ in ICP patients serum. IL-6 and TGF-$\beta$ can activate the STAT3 signalling pathway to promote Th17 cell differentiation, at the same time Th17 cell secrete IL-6 [24]. The increased proportion of Th17 cells via autoimmune regulation would inhibit the inflammatory responses [15]; hence, the effects of STAT3 and ROR$\gamma$t on the differentiation of Th17 cell are likely inhibited in ICP. However, since small number of patients in our study, these results may change if a larger patient cohort is analysed. Further study is still needed to validate this.

IL-17 is an important inflammatory mediator of immunity and several studies have focused on its role in ICP. Ren et al. [6, 25] found that IL-17 levels in the decidual tissue and peripheral blood of ICP patients were significantly elevated compared with healthy pregnancy; they positively correlated with serum TBA. However, Kirbas et al. [17] found serum IL-17 level was significantly increased in ICP patients than normal pregnant patients, but no significant relationship between IL-17 and TBA was abserved. Based on these results, no definite conclusion could be reached on the correlation between IL-17 and TBA ICP patients. We found that placental IL-17A protein levels and peripheral blood Th17 cell levels were significantly higher in ICP group than control group. Our results are similar to previous studies [6, 25]. Additionally, our study also revealed positive correlations among maternal serum TBA, DBIL, AST, and ALT and placental STAT3 and ROR$\gamma$t proteins in both the ICP and normal groups. Moreover, maternal serum TBA, DBIL, and AST levels positively related with placental IL-17A protein and the proportion of Th17 cell. TBA measurement is regarded as a gold-standard method for diagnosing and assessing ICP [26, 27]. In the setting of ICP, liver injuries typically have a hepatocellular pattern, which is not surprising and thus AST and ALT are usually elevated in patients with ICP. However, there was no relationship between Th17 cell and maternal serum ALT. Another previous study [28] used maternal routine liver tests to reliably exclude ICP and showed that combined laboratory scores including AST, GGT, ALK, and TB, but not ALT, contribute to this prediction ability. This needs to be evaluated in further studies. In some cases, patients with ICP develop jaundice. ICP patients with jaundice are considered to have a more severe form of the condition and are more likely to have adverse pregnancy outcomes [29]; moreover, maternal DBIL levels may be another important predictor of adverse pregnancy outcomes. Our data indicated that Th17 cells might be a potential marker for ICP prediction and assessment.

Bile acid can induce the production of inflammatory mediators like IL-23 [30], and IL-17 can promote the these inflammatory mediators production, which in turn induces Th17 cell expansion through a positive feedback loop [31]. IL-17 could act on bile duct epithelial cells and promot e the Th17 cell differentiation and aggregation. Th17 cells around the bile duct, which may interfere with bile acid metabolism and induce ICP [30]. Moreover, the expansion in Th17 cell might further aggravate the inflammatory response at the maternal-foetal interface [32]. We found significant positive correlations between placental IL-17A protein and the proportion of Th17 in both the ICP and control groups. These results indicated the proportion of peripheral blood Th17 could reflect immune activity at the maternal-foetal interface.

Our study revealed ICP patients were associated with earlier gestational days, lower neonatal weights and higher ratio of placental weight and neonatal weight comparing with normal pregnancy. The difference in gestational days may affected the comparisons’ accuracy of Th17 cell between normal and ICP groups. However, previous and recent studies both found Th17 cell count keep stable during all pregnancy [33, 34]. Additionlly, the mean gestational days in both groups was over 37 weeks. So the gestational days difference may has little effect on Th17 cells in our study. The difference of ratio of placental weight and neonatal weight in two group may reflect impaied placenta function,which need further study.

Our study presents some limitations. We studied the expression levels of STAT3/ROR$\gamma$t /IL-17A in placenta only at the protein level and not at the mRNA level. The histological features of the placenta were not recorded when the placenta tissue were collected. In addition, all ICP patients in our study were treated with ursodeoxycholic acid. As reported by Lee et al. [35], ursodeoxycholic acid colud inhibit the function of Th17 cells.There was no high quliaty study that examined ursodeoxycholic acid effect on Th17 cell in ICP. In our study, we did not examine the alters in Th17 cell before and after treatment with ursodeoxycholic acid. In future studies, we plan to include a larger number of samples and observe the effect of ursodeoxycholic acid for Th17 cell in patients with ICP, simultaneously. Moreover, in vitro experiments will be carried out to verify the results.

In summary, our data indicated that with an increase of Th17 cells, the roles of STAT3 and ROR$\gamma$t in the differentiation and function of Th17 in ICP are likely inhibited and attenuated. Th17 might play a negative role in ICP by aggravating inflammatory responses and disrupting the immune balance at the maternal-foetal interface. Excessive accumulation of Th17 cells may be associated with adverse outcomes of ICP.

YXW is responsible for research performing and manuscript writing; YYH is responsible for the design of the research study; QC is responsible for the data analysis; QHC and DS is responsible for the help and advice. All authors contributed to editorial changes in the manuscript. All authors read and approved the final manuscript.

The study was approved by the Ethics Committee of West China Second Hospital, Sichuan University, Chengdu, China and the Ethics board number is 2020-67. All patients signed an informed consent form.

We are very grateful to the patients and staff participated in this study.

This work was supported by Applied Research in China Sichuan Province [grant number 2022YFS0043]; the Applied Research in China Sichuan Province [grant number 2020CDZG-23]; Clinical research fund of West China Second University Hospital [grant number KL107].

The authors declare no conflict of interest.