ASK1 promotes uterine inflammation leading to pathological preterm birth

It is widely accepted that enhanced uterine inflammation associated with microbial infection is a main causative factor for preterm birth. However, little is known about the molecular basis by which inflammation is associated with preterm birth. Here, we demonstrate that apoptosis signal-regulating kinase 1 (ASK1), a member of the mitogen-activated protein 3-kinase family, facilitates inflammation-induced preterm birth and that inhibition of ASK1 activity is sufficient to suppress preterm birth. ASK1-deficient pregnant mice exhibited reduced incidence of lipopolysaccharide (LPS)-induced preterm birth. ASK1 was required for the induction of LPS-induced inflammatory responses related to preterm birth, including pro-inflammatory cytokine production in the uterus and peritoneal cavities. In addition, selective suppression of uterine ASK1 activity through a chemical genetic approach reduced the incidence of LPS-induced preterm birth. Moreover, translational studies with human choriodecidua demonstrated that ASK1 was required for LPS-induced activation of JNK and p38 and pro-inflammatory cytokine production. Our findings suggest that ASK1 activation is responsible for the induction of inflammation that leads to preterm birth and that the blockade of ASK1 signaling might be a promising therapeutic target for preventing preterm birth.

known to trigger pathologic inflammatory responses associated with enhanced production of pro-inflammatory cytokines in the development of various diseases, including cancers, autoimmune diseases, and microbial infections [11][12][13] . With respect to the pathogenesis of preterm birth, several recent studies using LPS-induced mouse models and human choriodecidua have suggested that the activation of the JNK and p38 pathways underlies the pathogenesis of inflammation-induced preterm birth 14,15 . However, findings implicating the involvement of stress-activated signaling systems in preterm birth are limited, and a mechanism detailing how JNK and p38 receive upstream signals and induce inflammatory responses that trigger preterm birth remains unknown.
The activities of JNK and p38 are tightly controlled by theirs upstream MAP kinase kinases and MAP kinase kinase kinases (MAP3Ks). Among MAP3Ks, apoptosis signal-regulating kinase 1 (ASK1) is accepted as a key player in the regulation of the activities of the JNK and p38 pathways, and plays pivotal roles in the pathogenesis of various diseases, including cancers, infections, and neurodegenerative diseases [16][17][18] . Intriguingly, ASK1 activation in macrophages and other inflammatory cells is crucial for Toll-like receptor 4 (TLR4)-mediated innate immunity responses triggered by LPS, which results in the production of pro-inflammatory cytokines; in fact, ASK1-deficient mice have demonstrated resistance to LPS-induced sepsis 19 . Moreover, because ASK1 is activated only under pathological conditions, it has drawn a great deal of attention as a therapeutic target 20 . However, very little has been determined regarding the involvement of ASK1 in pregnancy disorders, including preterm birth. In the present study, we demonstrate that ASK1 is crucial for promoting infection-induced uterine inflammation leading to preterm birth by regulating the JNK and p38 pathways, based on findings from an LPS-induced preterm birth model utilizing two independent types of ASK1-genetically-engineered mice, as well as in vitro studies using human choriodecidua, thus, implicating ASK1 as a potential therapeutic target for preterm birth.

ASK1 deficiency suppresses LPS-induced preterm birth.
To examine the involvement of ASK1 in preterm birth, we initially assessed the expression of ASK1 in the uterus. ASK1 is reportedly expressed ubiquitously in mice, however, protein expression in the organs related to the female reproductive system remained unknown. Utilizing samples from ASK1-deficient (ASK1 −/− ) pregnant mice as negative controls, we confirmed that ASK1 protein is substantially expressed in the uterus, cervix, and myometrium (Fig. 1A). Then, to assess the roles of ASK1 in preterm birth, we used a preterm-birth mouse model induced by transvaginal injection of LPS into the cervix 21 , which mimics the pathological condition of chorioamnionitis resulting from bacterial infection ascending from the vagina up to the uterus, in wild-type mice and ASK1 −/− pregnant mice.
As a number of studies have previously reported, ASK1 −/− mice were viable and fertile, and displayed no apparent abnormal phenotypes in the female reproductive or other organs. In addition, no spontaneous preterm or post-term birth was induced in ASK1 −/− pregnant mice, and ASK1 −/− pups from ASK1 −/− pregnant mice were born and grew without any apparent abnormalities. First, the status of the uterine ASK1 signal transduction pathway, converging on stress-activated MAPKs, JNK, and p38, was examined by immunoblotting in the LPS-induced preterm birth model. Using a phospho-ASK1 antibody that specifically recognizes a critical threonine residue in ASK1 22 , which is the activating phosphorylation site, we found that cervical application of LPS enhanced the phosphorylation status of ASK1 both in the cervix and the myometrium of WT pregnant mice (Fig. 1B,C). In addition, we found that LPS-induced phosphorylation of JNK and p38 were suppressed in the cervix and myometrium of ASK1 −/− pregnant mice, compared with that in WT mice. These results suggested that uterine ASK1 is activated by LPS and that ASK1 is required for the activation of the JNK and p38 pathways in response to LPS in the uterus. Next, we examined the frequency of preterm birth and found that the incidence of delivery within 48 hours following LPS injection was significantly lower in ASK1 −/− pregnant mice than in WT mice (Fig. 1D). This result indicated that ASK1 functions to facilitate preterm birth induced by LPS.

ASK1 facilitates LPS-induced inflammatory responses leading to preterm birth. Preterm birth
is characterized by locally enhanced inflammation in the uterus caused by microbial infection, including the production of pro-inflammatory cytokines and the infiltration of inflammatory cells into the uterus, which induces labor and cervical ripening, leading to the preterm birth 7 . LPS-induced activation of the JNK and p38 pathways is well known to facilitate the inflammatory response by also inducing cytokine expression as well 11 . Next, to assess the contribution of ASK1 to LPS-induced inflammation leading to preterm birth, we used ELISA to compare the levels of pro-inflammatory cytokines in the peritoneal fluid between WT and ASK1 −/− pregnant mice after LPS injection. The levels of IL-1β, IL-6, TNF-α, and the murine IL-8 homologue, CXCL2, in the peritoneal fluid were markedly increased following LPS treatment, and the elevated levels of these cytokines were significantly suppressed in ASK1 −/− pregnant mice compared with those in WT mice ( Fig. 2A-D). Moreover, LPS-induced elevated levels of TNF-α and CXCL2 in the myometrium were also significantly reduced in ASK1 −/− pregnant mice compared with WT mice (Fig. 2E,F). Among inflammatory cells amplifying the inflammation related to the pathogenesis of preterm birth, macrophages are the predominant subtype residing in the uterus 23 . Macrophages infiltrating the cervix are known to play critical roles in driving the inflammatory process that facilitates the cervical ripening mediated by the production of matrix metalloproteinases (MMPs) 24 . Therefore, we examined the state of macrophage infiltration in the cervix after LPS using immunohistochemical staining for F4/80, a marker for macrophages. LPS-induced cervical infiltration of macrophages with immunoreactivity for F4/80 was markedly visible in WT pregnant mice but was significantly less frequent in ASK1 −/− mice (Fig. 2G,H). Furthermore, we found that LPS-induced elevated levels of MMP8, a key player in the development of cervical ripening in preterm birth 25 , are significantly suppressed in the cervix of ASK1 −/− pregnant mice compared with WT mice (Fig. 2I). These results suggested that ASK1 plays crucial roles in the promotion of the inflammatory response that is closely associated with the development of LPS-induced preterm birth.  www.nature.com/scientificreports www.nature.com/scientificreports/ Selective inhibition of ASK1 kinase activity elicits therapeutic effects on LPS-induced preterm birth. Our results that ASK1 deficiency suppressed LPS-induced preterm birth, as demonstrated using ASK1 −/− pregnant mice, raised the possibility that strategies to inhibit ASK1 kinase activity might possess therapeutic potential with respect to preterm birth. In order to investigate this possibility, we took a chemical genetic approach, termed analogue sensitive kinase allele (ASKA), in which ASK1 kinase activity can be pharmacologically controlled by the use of an ATP analogue 26 (Fig. 3A). In this technology, an ASKA harbors mutations in the ATP binding pocket of its kinase domain, and while it can accommodate native ATP and function normally, its kinase activity is competitively blocked by an ATP analogue. We utilized knock-in mice harboring an ASKA of ASK1 (ASK1 ASKA ), in which selective pharmacological inhibition of ASK1 kinase activity can be achieved by the administration of an ATP analogue, 1Na-PP1. As previously reported 26 , the function of ASK1 in ASK1 ASKA mice in response to oxidative stress in the absence of 1Na-PP1 is comparable to that in WT mice. However, the oxidative stress-induced activation of ASK1 is robustly blocked by treatment with 1Na-PP1 in ASK1 ASKA mice, but not in WT mice. Then, utilizing ASK1 ASKA pregnant mice mated with ASK1 ASKA males, we performed transvaginal injection of LPS into the cervix and confirmed that the LPS-induced phosphorylation status of ASK1 in the cervix as well as the myometrium is substantially blocked by the application of 1Na-PP1 into the cervix in ASK1 ASKA pregnant mice (Fig. 3B). Intriguingly, similar to the preterm birth phenotype, transvaginal treatment with 1Na-PP1 into the cervix significantly reduced the incidence of preterm birth in ASK1 ASKA pregnant mice, compared with that in ASK1 ASKA pregnant mice without 1Na-PP1 treatment (Fig. 3C). As shown in Fig. 1D, in the absence of 1Na-PP1, the incidence of LPS-induced preterm birth in ASK1 ASKA pregnant mice was comparable to that in WT mice, suggesting that ASK1 functions normally in response to LPS in ASK1 ASKA pregnant mice. As there was no difference in preterm birth rates between WT mice with or without 1Na-PP1, the compound had no effect by itself on preterm birth induced by LPS. These results suggested that uterine ASK1 activity is responsible for LPS-induced preterm birth, and that a pharmacological approach to inhibit ASK1 kinase activity can elicit therapeutic effects.

Suppression of ASK1 activity by the ASK1-specific inhibitor K811 reduced LPS-induced inflammation in human choriodecidua.
To extend these mouse findings to the human situation, we conducted in vitro studies using explant cultures of choriodecidua isolated from human term placentas from normal pregnancies. Choriodecidua, which infectious pathogens colonize in the initial stages of chorioamnionitis, plays a central role in triggering detrimental excessive inflammatory responses ascending to the intra-amniotic cavity by producing a number of pro-inflammatory cytokines 27 . Therefore, we explored the involvement of the ASK1-JNK and p38 pathways in LPS-induced responses in human choriodecidua by using a recently-developed ASK1-specific inhibitor, K811 28 . Immunoblotting analysis revealed that treatment with K811 blocked the LPS-induced phosphorylation status of ASK1 in human choriodecidua (Fig. 4A). Moreover, the phosphorylation of JNK and p38 induced by LPS in human choriodecidua was almost completely suppressed by K811 treatment. These results suggested that ASK1 is responsible for LPS-induced activation of the JNK and p38 pathways in human choriodecidua. Next, we conducted expression profiling of pro-inflammatory cytokines induced by LPS. mRNA expression levels of the pro-inflammatory cytokines, IL-1β, IL-6, IL-8, and TNF-α, were markedly increased in choriodecidua by LPS treatment, and the increased expression of these cytokines was significantly suppressed by the administration of K811 (Fig. 4B-E). We also found that LPS induced increased mRNA expression of MMP8, a key molecule whose overproduction is closely associated with the development of chorioamnionitis and preterm premature rupture of membranes 29 , in choriodecidua, and that K811 treatment significantly reduced the production of MMP8 (Fig. 4F). These results indicated that ASK1 contributes to the induction of excessive inflammatory responses related to preterm birth in response to LPS in humans as well.

Discussion
Utilizing two types of mice genetically engineered at ASK1 (ASK1-deficient mice and knock-in mice harboring an ASKA of ASK1), we revealed that ASK1 plays a critical role in the development of inflammation-associated preterm birth. Moreover, in vitro studies in human choriodecidua using an ASK1-specific inhibitor to suppress the activity of ASK1 demonstrated that ASK1 is responsible for LPS-induced inflammatory cytokine production related to preterm birth, by controlling the JNK and p38 pathways. ASK1, a stress-activated MAP3K that regulates the activities of JNK and p38, plays key roles in stress responses under pathologic conditions 17 . A number of studies have so far reported that ASK1 is intimately associated with the progression of various diseases in which aberrant inflammatory conditions are involved 20 . However, the contribution of ASK1 to disorders of the female reproductive system, including obstetric complications, remained completely unknown. We initially found that ASK1 is robustly expressed in the uterus, pointing to the possibility that ASK1 functions in response to certain pathologic conditions during pregnancy. Then, considering the close involvement of ASK1 signaling pathways in innate immune response-induced inflammation 19 , the current study aimed to assess the roles of ASK1 in preterm birth. As pointed to by the abundant expression of ASK1 in the uterus, we found that uterine ASK1 is activated in response to cervical LPS injection and contributes to the activation of JNK and p38. Moreover, ASK1-deficient pregnant mice exhibited lower incidence of preterm birth induced by LPS, suggesting that ASK1 functions to facilitate inflammation-associated preterm birth. Thus, we report for the first time the involvement of ASK1 in the development of pregnancy complications, especially in inflammation-induced pathological preterm birth.
Excessive inflammation caused by microbial infection ascending into the cervix and uterus from the vagina is a key driving force in the development of preterm birth 6 . Increased inflammatory responses, characterized by enhanced production of pro-inflammatory cytokines and infiltration of leukocytes such as macrophages, are involved in every aspect of the preterm birth process, including induction of labor, cervical ripening, and rupture of the membrane. Several studies have reported the involvement of JNK or p38 in the pathogenesis of this (2020) 10:1887 | https://doi.org/10.1038/s41598-020-58653-9 www.nature.com/scientificreports www.nature.com/scientificreports/  www.nature.com/scientificreports www.nature.com/scientificreports/ infection-induced inflammatory process leading to preterm birth 14,15 . A recent study by Pirianov et al. reported that JNK activation contributes to the progression of LPS-induced preterm birth by facilitating uterine inflammation 14 . Intriguingly, this report also showed that the inhibition of JNK showed therapeutic effects, not only on preterm birth but also on excess inflammation-induced fetal brain damage. However, it has not been fully elucidated how the JNK/p38 pathways are involved in the enhanced inflammation leading to preterm birth. In the current study, we took advantage of a preterm birth model induced by transvaginal injection of LPS into the cervix, which mimics the pathological condition of chorioamnionitis, and found that LPS-induced production of IL-1β, IL-6, TNF-α, and the murine IL-8 homologue, CXCL2, was demonstrably reduced in ASK1 −/− mice compared to WT mice, as was the number of infiltrating uterine macrophages. All of these pro-inflammatory cytokines are reported to be locally elevated around the uterus after LPS injection in mice and to be involved in the progression to preterm birth 21 . In humans, IL-1β, IL-6, and TNF-α are increased in the myometrium, amniotic fluid, and decidua of pregnant women presenting with preterm birth, and are markers for the presence of infection 3 . Moreover, our findings using human choriodecidua revealed that ASK1 is required for LPS-induced MMP8 expression and production of IL-1β, IL-6, TNF-α, and IL-8. It has been reported that compared to the amnion, the choriodecidua is the major source of proinflammatory cytokines, including IL-1β, IL-6, IL-8, and TNF-α, which are closely associated with the development of preterm labor 27,30 . Shoji et al. reported that p38 activation promotes the production of these pro-inflammatory cytokines in human choriodecidua 15 , pointing to the significance of p38 in preterm birth and supporting our results. Thus, our results from ASK1-deficient mice combined with human choriodecidua strongly suggest that ASK1 plays crucial roles in the development of LPS-induced inflammatory responses leading to preterm birth.
Findings in ASK1 ASKA mice demonstrated that pharmacological inhibition of uterine ASK1 activity reduced the incidence of preterm birth induced by LPS. The ASKA technology is generally considered a powerful tool that allows the validation of specific responses to the inhibition of any protein kinase in vivo in knock-in animals 31 . The pharmacological application of an ATP analogue achieves selective competitive inhibition of the target kinase. Moreover, ASK1 ASKA enables the more precise investigation of ASK1 function without affecting protein expression of ASK1, since ASK1 deficiency induces the degradation of ASK2, a MAP3K molecule that is stabilized and functions in response to reactive oxygen species (ROS) only by forming a complex with ASK1 32 . Therefore, the findings from ASK1 ASKA mice showing that cervical treatment with 1Na-PP1 to suppress ASK1 activity reduced the incidence of preterm birth provided more solid in vivo evidence that uterine ASK1 activity is responsible for LPS-induced preterm birth, and that pharmacological inhibition of ASK1 kinase activity could be a novel therapeutic approach. As concerns human evidence, we demonstrated that a specific ASK1 inhibitor, K811, suppressed the LPS-induced activation of JNK and p38 and inflammatory responses related to preterm birth in human choriodecidua. Taken together, these findings strongly suggested that local uterine ASK1 inhibition could elicit anti-inflammatory therapeutic effects on inflammation-driven preterm birth. Several studies have reported that blockade of the JNK or p38 pathways have therapeutic potential for inflammation-induced preterm labor 14,15 . However, negative results from JNK and p38 inhibitors in clinical trials for various diseases, partly due to toxicity issues, indicate that the blockade of JNK or p38 is not desirable 13,33 . In contrast to the embryonic lethal phenotypes of mice deficient in p38 or JNK1/2, a number of studies have revealed that ASK1-deficient mice develop normally and display no abnormal phonotypes under normal physiological conditions 20 , indicating that ASK1 is not required for the maintenance of normal homeostatic mechanisms. ASK1 is a stress-responsive MAP3K, and it is activated and plays crucial roles only when cells are under stress. Thus, inhibition of ASK1 activity in preterm birth may aid in preventing ASK1-dependent activation of JNK and p38 in response to excessive ROS, which induces pathologic conditions, without interfering with the maintenance of homeostasis, including normal immune and anti-infection mechanisms. Theoretically, therapeutic strategies to inhibit ASK1 may carry low risks for side effects. Although further studies are needed to validate the safety of ASK1 inhibitors during pregnancy in animal models, our results suggest that ASK1 is an attractive target with respect to preterm birth, as it may enable the suppression of exaggerated inflammation cascades, which may lead to the improvement of perinatal outcomes.
In conclusion, our current study has provided valuable evidence from mouse and human models not only pointing to the significance of ASK1 signaling pathways in inflammation-associated preterm birth, but also establishing an innovative approach to inhibit preterm birth involving the anti-inflammatory effects of targeting ASK1.

Animals. All animal experiments were conducted according to the protocol approved by the Animal Research
Committees of the Graduate School of Medicine, and of the Graduate School of Pharmaceutical Sciences of the University of Tokyo. All mice utilized in this study were of the C57BL/6J genetic background, and female mice at 8-12 weeks of age were used for all experiments. ASK1-deficient (Map3k5/ASK1 −/− ) mice and knock-in mice harboring an analogue-sensitive kinase allele (ASKA) of ASK1 (ASK1 ASKA ) mice were generated as described previously 26,34 . All mice were maintained at room temperature in a humidity-controlled room with a 12 hour-light/12 hour-dark cycle and had free access to food and water during the entire experimental period.
Preterm birth model. The day of vaginal plug detection after the first day of mating was defined as day 0 of gestation. Lipopolysaccharide (LPS) derived from Escherichia coli (O111: B4; Sigma-Aldrich Japan, Tokyo, Japan) was injected transvaginally into the cervix on gestational day 15, as previously reported 21  www.nature.com/scientificreports www.nature.com/scientificreports/ sacrificed at one and eight hours after LPS injection. The abdominal cavity was washed with 2 ml Milli Q water to obtain peritoneal fluid samples. To block ASK1 kinase activity in ASK1 ASKA pregnant mice, 4-Amino-1-tert-butyl-3-(19-naphthyl) pyrazolo [3,4-d] pyrimidine (1Na-PP1 952.2 µg/kg, Calbiochem, U.S.A.) was injected into the cervices of ASK1 ASKA pregnant mice mated with ASK1 ASKA male mice through the intracervical route. LPS (1.0 µg) was injected into the cervices of WT and ASK1 ASKA pregnant mice simultaneously with 1Na-PP1 on gestational day 15. Mice were sacrificed at two hours after LPS and 1Na-PP1 injection for sample collection.
Measurement of inflammatory cytokine levels by ELISA. Levels of mouse IL-1β, IL-6, TNF-α, and CXCL2 in peritoneal fluid were measured by enzyme linked immunosorbent assay kit (Quantikine; R&D Systems, U.S.A.) according to the manufacturer's protocol as described previously 36 . Immunohistochemistry. The tissues were fixed with 4% paraformaldehyde. The paraffin-embedded blocks and sections were obtained from Genostaff. (Genostaff Co., Ltd., Japan). Tissue sections were de-paraffinized with xylene and rehydrated through an ethanol series and PBS. For immunohistochemistry of F4/80, antigen retrieval was performed by enzyme treatment with Proteinase K. Endogenous peroxidase was blocked with 0.3% H 2 O 2 in methanol for 30 min, followed by incubation with Protein Block (Genostaff) and the avidin/biotin blocking kit (Vector, U.S.A.). The sections were incubated with anti-F4/80 rat monoclonal antibody (BIO-RAD, Japan) at 4°C overnight. They were then incubated with biotin-conjugated rabbit anti-rat IgG (Vector) for 30 min at room temperature, followed by the addition of peroxidase conjugated streptavidin (Nichirei, Japan) for 5 min. Peroxidase activity was visualized by diaminobenzidine. The sections were counterstained with Mayer's Hematoxylin (MUTO, Japan), dehydrated, and then mounted with Malinol (MUTO). For the quantification of F4/80 immunohistochemistry, the number of positively stained cells that have overlapping positive signals in the cell nucleus confirmed with the staining with hematoxylin was counted in 4 fields of randomized and blinded slides (200x magnification). Prior to immunostaining with F4/80, the specificity of the antibody was confirmed by using the spleen as a positive control and incubating biotin-conjugated rabbit anti-rat IgG as a secondary antibody without the primary antibody as a negative control.
Explant culture of human choriodecidua. Experiments were conducted under the approval of the institutional review board of our institution (No. 11538-(3)), and informed consent was obtained from all pregnant women. Experiments were performed in accordance with the relevant guidelines and regulations of medical and health research involving human subjects. Human term membranes were collected from the placentas of cesarean sections performed at 37-38 weeks of gestation. Cases without maternal or fetal complications were selected for sample collection. Briefly, membranes isolated from the placenta were washed with sterile PBS and soaked in hyaluronidase (Nakalai tesque, Japan) for 30 minutes. A layer of choriodecidua was carefully scraped from amnion membranes. The tissues were cut into small pieces. Tissue explants (0.1 g) were placed in each well of a 24-well plate and cultured in DMEM/F12 (ThermoFisher SCIENTIFIC, Japan) containing 10% fetal bovine serum (FBS, ThermoFisher SCIENTIFIC) as described previously 15 . The ASK1 inhibitor K811 was developed and confirmed