{"paper_id":"95a14512-d49d-4a3f-b6b7-d8bed3a9bafe","body_text":"1Scientific  RepoRtS  |         (2020) 10:9014  | https://doi.org/10.1038/s41598-020-65884-3\nwww.nature.com/scientificreports\nAcquired contractile ability in \nhuman endometrial stromal cells \nby passive loading of cyclic tensile \nstretch\nJeonghyun Kim1, takashi Ushida1,2, Kevin Montagne2, Yasushi Hirota3, osamu Yoshino4, \nt akehiro Hiraoka3, Yutaka osuga 3 & Katsuko S. furuakwa 1,2 ✉\nthe uterus plays an important and unique role during pregnancy and is a dynamic organ subjected \nto mechanical stimuli. it has been reported that infertility occurs when the peristalsis is prevented, \nalthough its mechanisms remain unknown. in this study, we found that mechanical strain mimicking \nthe peristaltic motion of the uterine smooth muscle layer enabled the endometrial stromal cells to \nacquire contractility. in order to mimic the peristalsis induced by uterine smooth muscle cells, cyclic \ntensile stretch was applied to human endometrial stromal cells. the results showed that the strained \ncells exerted greater contractility in three-dimensional collagen gels in the presence of oxytocin, due \nto up-regulated alpha-smooth muscle actin expression via the cAMp signaling pathway. these in vitro \nfindings underscore the plasticity of the endometrial stromal cell phenotype and suggest the possibility \nof acquired contractility by these cells in vivo and its potential contribution to uterine contractile \nactivity. this phenomenon may be a typical example of how a tissue passively acquires new contractile \nfunctions under mechanical stimulation from a neighboring tissue, enabling it to support the adjacent \ntissue’s functions.\nIt is now widely known that mechanical stimuli applied to various cell types can trigger intracellular signaling \nevents leading to physiological and pathological changes 1–5. The uterus allows implantation of the embryo and \nregulates its growth by supplying nutrients from the mother’s body6. It is also known as a dynamic organ that is \nmodulated by menstrual hormone changes during the menstrual cycle and pregnancy. The uterine wall consists \nof three layers, namely the endometrium, the myometrium, and the perimetrium\n7,8. While the inner layer of the \nendometrium is composed of epithelial cells and stromal cells, the thickest middle myometrial layer mainly con-\nsists of smooth muscle cells. The perimetrium is the thin outermost layer of connective tissue. The myometrium \nis known to show spontaneous contractile activity\n9, and undergoes remodeling by hyperplasia and hypertrophy \nduring pregnancy10.\nThe non-pregnant uterus also shows a distinct activity called “endometrium movement” throughout the \nmenstrual cycle, which is regulated by ovarian steroid hormones 11. Furthermore, the endometrium wave is \nknown to play a significant role during pregnancy in order to transport the fertilized egg/zygote through the \nutero-tubal cavities prior to implantation\n12. It has also been reported that infertility occurs when the mechanical \nstress induced by the endometrium wave is prevented13. Therefore, we believe that this mechanical stimulus from \nthe myometrium has a crucial role in physiological functions of the endometrium, such as menstruation and \npregnancy.\nIn this study, we hypothesized that the uterine peristalsis induced by uterine smooth muscle cells might affect \nthe contractile ability of endometrial stromal cells, an important function of the uterus for pregnancy. The endo-\nmetrial stromal cells have been thought to passively undergo strain stimulation under the contractile movement \nof uterine smooth muscle cells. However, we propose that stromal cells actually actively support the peristaltic \n1Department of Bioengineering, Graduate School of Engineering, the University of Tokyo, Tokyo, Japan. \n2Department of Mechanical Engineering, Graduate School of Engineering, the University of Tokyo, Tokyo, Japan. \n3Department of Obstetrics and Gynecology, School of Medicine, the University of Tokyo, Tokyo, Japan. 4Department \nof Obstetrics and Gynecology, School of Medicine, Kitasato University, Sagamihara, Japan. ✉e-mail: furukawa@\nmech.t.u-tokyo.ac.jp\nopen\n\n2Scientific  RepoRtS  |         (2020) 10:9014  | https://doi.org/10.1038/s41598-020-65884-3\nwww.nature.com/scientificreportswww.nature.com/scientificreports/\nmovement of the uterus. It is possible that the existence of these mechanisms helps to make the uterine peristaltic \nmovement, which plays an important role for implantation of fertilized eggs and pregnancy, more steady and \nreliable.\nResults\nReorientation of hESCs after applying 7 days of uniaxial cyclic strain. In this study, we loaded 15% \nof uniaxial cyclic strain at 0.1 Hz to hESCs for 7 days, as shown in Fig. 1(A). In order to quantify the reorienta-\ntion of hESCs after applying uniaxial cyclic strain for 7 days, we evaluated the changes in the cells’ angle from \nnormal microscope images of control and strained cells as shown in Fig. 1(B),(C). In Fig. 1(D),(E), mean angles \nof cells (or mean direction of elongation) in control and strained hESCs were 108°  (SD 69.1° ) and 91.3°  (SD \nFigure 1. (A) Schematic view of the experimental setup showing the side view of the Flexcell tension system \nand the top view of the Flexcell plate. Microscope images of (B) control and (C) strained cells after applying 7 \ndays of cyclic strain. The white bar indicates 300 μm. (D) Quantification of orientation changes in the strained \ncells. Graphs show the angle of cells (300 cells from 6 independent experiments). The bars represent the mean \n± standard error deviation (p-value was obtained from F-test; *p < 0.05, **p < 0.005). (E) Schematic plot of the \ncell distribution. The red and black arrows represent the strained and control group, respectively.\n\n3Scientific  RepoRtS  |         (2020) 10:9014  | https://doi.org/10.1038/s41598-020-65884-3\nwww.nature.com/scientificreportswww.nature.com/scientificreports/\n33.6°), respectively. The mean angle of strained hESCs compared to that of control cells was significantly different \n(p < 0.005). Moreover, the standard deviation of the angle in the strained cells was much smaller than that of the \ncontrol sample. This indicates the angle of the cells in the strained hESCs was much more uniform compared to \nthe more random orientation of the control hESCs. Hence, after 7 days, the strained cells reoriented perpendicu-\nlarly to the direction of strain and became elongated while the control cells were randomly oriented.\nUniaxial cyclic strain up-regulated α-SMA expression in heScs. As shown in Fig. 2(A), we analyzed \nby real-time PCR several stromal cell markers including CD10 and CD90 and uterine smooth muscle cell (SMC) \nmarkers such as ACTA2 and TAGLN. Particularly, ACTA2 plays a key role in the production of alpha-smooth \nmuscle actin (α-SMA), which belongs to the actin protein family and is involved in cell contraction. Cyclic strain \nslightly decreased endometrial stromal cell marker expression (0.90-fold change for CD10 and 0.86-fold change \nfor CD90), but the changes were not significant. With regards to ACTA2 and TAGLN, they were significantly \nup-regulated by cyclic strain (1.32-fold change for ACTA2, p < 0.005; and 1.59-fold change for TAGLN, p < 0.05). \nIn the same manner as ACTA2 and TAGLN, 7 days of cyclic strain significantly raised mRNA expression of \nFigure 2. Promoted expressions of smooth muscle cell markers in hESCs after applying cyclic strain for 7 \ndays, measured by real-time PCR and immunostaining. (A) mRNA expressions of endometrial stromal cell \nmarkers and smooth muscle cell markers in hESCs after applying cyclic strain for 7 days measured by real-\ntime PCR. (CD10, CD90, ACATA2, TAGLN, OXTR, DES, IL6, ANGPT1, and RAMP1). Graphs show the fold \nchange of mRNA expressions relative to RPL32 mRNA normalized to the control mean (n = 6). The bars \nrepresent the mean ± standard error (p-value was obtained from Student’s t-test; *p < 0.05, **p < 0.005). (B) \nImmunostaining of Vimentin (Vim) and smooth muscle actin (α-SMA) in hESCs after loading cyclic strain for \n7 days; Vim and α-SMA expression in control samples (left). Vim and α-SMA expression in strained samples \n(right). The arrows indicate the direction of cyclic strain. Scale bar = 100 μm.\n\n4Scientific  RepoRtS  |         (2020) 10:9014  | https://doi.org/10.1038/s41598-020-65884-3\nwww.nature.com/scientificreportswww.nature.com/scientificreports/\nthe oxytocin receptor (OXTR ), which is highly expressed in the myometrium to regulate uterine contraction \n(2.13-fold change for OXTR, p < 0.05).\nIn order to distinguish uterine SMCs and myofibroblasts, we also examined desmin (DES) and Interleukin 6 \n(IL6) expression, which are associated with myofibroblasts. Real-time PCR results also showed an up-regulation \nof DES mRNA expression (1.63-fold change, p  < 0.05) and a decrease in IL6 (0.68-fold change; p < 0.05) after \napplication of cyclic strain. On the other hand, mRNA expressions of Angiopoietin 1 (ANGPT1) (0.73-fold \nchange; p = 0.12) and Receptor activity modifying protein (RAMP1) (0.50-fold change, p < 0.005) were measured \nto distinguish uterine SMCs and vascular smooth muscle cells.\nAfter applying cyclic strain for 7 days, immunostaining for vimentin (Vim) and α -SMA was carried out to \ncheck for changes in stromal cell and smooth muscle cell marker expression as shown in Fig.  2(B). As a result, \nthere was an increase in the staining intensity of α -SMA in the strained cells while no significant change in the \nexpression of the stromal cell marker was observed after applying cyclic strain.\nc yclic strain increases cAMp production in heS c . In order to understand the effect of uniaxial cyclic \nstrain on hESCs, we measured the level of cAMP in hESCs after applying strain. Firstly, cAMP concentrations \nafter 15 mins of cyclic strain were measured. Figure 3(A) shows a transient and significant up-regulation in cAMP \nproduction (1.75-fold change; p < 0.005) in as little as 15 mins. In addition, Fig. 3(B) shows the levels of cAMP \nafter 7 days of cyclic strain. There was a non-significant increase in cAMP production immediately after 7 days \nof strain (1.62-fold change; p = 0.07). We then performed 7 days of strain, followed by a 2-hour break for cAMP \nlevels to stabilize, followed by an extra 15 mins of cyclic strain. After such a strain regimen, cAMP concentration \nwas significantly up-regulated (2.35-fold change; p < 0.05).\nSQ22536 and H-89 inhibit the up-regulation of α -SMA expression by cyclic strain in heScs. To \ndetermine whether the cAMP pathway is involved in the up-regulation of α -SMA expression under strain, we \nstretched hESCs in the presence or absence of the adenylyl cyclase inhibitor SQ22536 or the PKA inhibitor H-89. \nFigure 4(A)–(F) represent the fold changes in mRNA expression measured by real-time PCR in the presence or \nabsence of inhibitors. As in the previous experiment, the cyclic strain did not significantly affect CD10 or CD90 \nexpression in hESCs but significantly up-regulated ACTA2 (1.37-fold change) and TAGLN (1.68-fold change) \nexpression. By adding SQ22536, the up-regulation of ACTA2, TAGLN, and OXTR by cyclic strain was inhibited, \nwith a respectively 1.01-, 0.91-, and 1.00-fold change in ACTA2, TAGLN, and OXTR expression. Moreover, the \nuse of H-89 also showed an inhibiting effect on ACTA2 (0.47-fold change), TAGLN (0.55-fold change), and OXTR \n(0.73-fold change) expression. Moreover, the SQ22536 and H-89 non-significantly suppressed the increase in DES \n(0.81- and 1.12-fold change, respectively) expression.\nEffect of oxytocin on the hESC-mediated collagen I gel contraction. After applying uniaxial cyclic \nstrain to hESCs for 7 days, the strained cells were collected for a cell contraction assay to check the contractile \nability of the cells seeded in a three-dimensional collagen gel. Before the stress in the gel was released by detaching \nthe gels from the culture dish, the cells in the collagen I gel were treated with oxytocin (10 nM) to examine if in the \nFigure 3. Relative cAMP production levels measured by the cyclic AMP EIA kit. (A) Applying cyclic strain for \n15 mins significantly up-regulated cAMP production in hESCs. (B) cAMP levels were measured after 7 days of \nstrain, followed by a 2-hour break for cAMP levels to stabilize, followed by an extra 15 mins of cyclic strain. 7 \ndays of strain induced a non-significant increase in cAMP production, but adding a 2-hour break followed by an \nextra 15 mins significantly up-regulated cAMP levels. Graphs show the fold change of cAMP production levels \nrelative to the amounts of DNA and normalized to the control mean. The bars represent the mean ± standard \nerror (n = 4) (p-value was obtained from Student’s t-test; *p < 0.05, **p < 0.005).\n\n5Scientific  RepoRtS  |         (2020) 10:9014  | https://doi.org/10.1038/s41598-020-65884-3\nwww.nature.com/scientificreportswww.nature.com/scientificreports/\nFigure 4. mRNA expressions measured by real-time PCR of (A) CD10, (B) CD90, (C) ACTA2, (D) TAGLN, (E) \nOXTR, and (F) DES in hESCs after applying cyclic strain in the presence or absence of the inhibitors SQ22536 \nand H-89. All the mRNA expressions were normalized to RPL32 expression and further normalized to control \nvalues. Applying cyclic strain for 7 days non-significantly down-regulated both endometrial stromal cell \nmarkers and up-regulated the smooth muscle cell markers. While the non-significant down-regulation of CD10 \nand CD90 were unchanged by addition of SQ22536 and H-89, both inhibitors significantly inhibited the up-\nregulation of ACTA2 and OXTR. The bars represent the mean fold change ± standard error between strained \nand control samples (n = 4) (p-values were obtained from ANOV A followed by Fisher’s LSD test; *p < 0.05). (G) \nSchematic diagram of the signaling pathway activated in hESCs in response to cyclic strain. Adenylyl cyclase \nlocated on the inner side of the plasma membrane converts ATP to intracellular cAMP . cAMP induced by cyclic \nstrain then promoted SMa marker α-SMA expression, via adenylyl cyclase and PKA.\nFigure 5. hESCs-mediated collagen I gel contraction after 7 days in response to oxytocin (10 nM). (A) control \nsample without oxytocin (B) control sample with oxytocin, (C) gel containing strained cells without oxytocin, \nand (D) gel containing strained cells with oxytocin. (E) The graph represents the gel area 7 days after release \nnormalized to the initial area. The bars represent the mean ± standard error of gel area (n = 4) (p-value was \nobtained from ANOV A followed by Fisher’s LSD; *p < 0.05, **p < 0.005; the statistical analysis was performed \nusing the original area measurement data).\n\n6Scientific  RepoRtS  |         (2020) 10:9014  | https://doi.org/10.1038/s41598-020-65884-3\nwww.nature.com/scientificreportswww.nature.com/scientificreports/\npresence of oxytocin, the strained cells exerted greater contractility, behaving like uterine SMC. Figure 5(A)–(D) \nshow the samples 7 days after releasing the gels from the culture dish in the presence or absence of oxytocin. The \nsurface areas of the samples were measured and quantified as shown in Fig. 5(E). The surface areas of the control \nsamples with and without oxytocin shrank to 71.6% (1.36 cm2) and 55.9% (1.06 cm2), respectively. On the other \nhand, the strained cells showed a significant elevation in their contractile ability both in the absence and in the \npresence of oxytocin, as gels shrank to 48.9% (0.93 cm\n2) and 36.7% (0.70 cm2), respectively, of their initial surface \narea. In particular, while the strained cells showed greater contractility than the control cells (p < 0.005), the addi-\ntion of oxytocin in the strained cells further enhanced their contractility (p < 0.05).\nDiscussion\nMechanical stimulation is crucial to the proper function of many different organs like cartilage, blood vessels or \nthe uterus\n14,15. For the uterus, which plays an important role during pregnancy and childbirth, the mechanical \nstress caused by the peristaltic movement of the fallopian tube due to the contraction of uterine smooth muscle \ncells promotes the migration of fertilized eggs\n16. In addition, infertility can also be caused by uterine fibroids that \nperturb the periodic uterine peristaltic movement induced by hormones 17. Therefore, although the number of \nstudies regarding mechanical stress responses in the uterus is still limited, the mechanical stress may play a role in \nhomeostasis and pathogenesis of the uterus.\nIn this study, we proposed a new hypothesis regarding the response to mechanical stress in the uterine tissues \nwith a hierarchical structure. The uterus has a three-layer structure consisting of, starting from the inside, epi-\nthelial cells, stromal cells, and smooth muscle cells. From our in vitro results, it is reasonable to imagine a similar \nmechanism in vivo, by which the peristaltic movement of the uterus caused by contraction of the outermost \nsmooth muscle layer enhances the contractility of the inner stromal cells. We believe that such a mechanism \nmight make the three-dimensional peristaltic movements of the uterus during pregnancy and childbirth more \nconsistent and reliable.\nIn two-dimensional culture, the tensile stress that mimics the peristaltic movement of the myometrium is \nreported to regulate the biochemical function of stromal cells to support a\n the differentiation process of endo-\nmetrium, decidualization18, but it is dubious whether the tensile stress is well loaded three-dimensionally on the \nstromal cells in the body. As shown in this study, acquisition of contractility by the stromal cells under strain may \nbe an effective mechanism to transmit the tensile stress from the smooth muscle layer in the body. In other words, \nit is suspected that contraction by uterine smooth muscle cells is transmitted to the inner stromal cells, whereby \neach individual stromal cell can contract steadily thanks to the increased expression of proteins such as α -SMA \nand oxytocin. Considering the fact that, in vivo, infertility can occur when the uterus is unable to contract prop-\nerly, these findings may represent a new mechanotransduction mechanism, by which contractility is transmitted \nfrom the outside to the adjacent inner cell layer.\nIt is possible that such a phenomenon occurs not only in the uterus, but also in blood vessels as reported in the \nliterature. When the tensile stress is applied to vascular endothelial cells such as human umbilical cord endothelial \ncells (HUVEC), smooth muscle actin expression is dramatically increased\n19–21. It might imply that the vascular \nendothelial cells loaded with tensile stimulation may not differentiate into vascular smooth muscle cells, and \nindividual vascular endothelial cells might have instead acquired the contractility under the tensile stimulation \ninduced by mechanical activation of intracellular signaling pathways, thus potentially enhancing the efficiency of \nblood vessel contraction. Our experiments in hESCs have shown that the strain slightly increases the gene expres-\nsion levels of SMC markers, ACTA2\n22 and TAGLN23. On the other hand, the immuno-staining data still showed \nstrong staining levels of vimentin, a marker for stromal cells24, even with or without strain. Although Additionally, \nthe α -SMA staining levels in ESC were extremely low compared to those in the SMC in vivo24. However, the strain \ndid not apparently increase those of α-SMA in ESC even after stretching. Taking into consideration those results, \nit is reasonable to conclude that strain did not differentiate ESC into SMC, but made ESC acquire the ability to \ncontract. This phenomenon might be similar to the response of HUVEC under strain.\nIn the field of cell and tissue engineering, mechanical stimuli have been identified as a significant factor to \ninduce physiological changes by activating intracellular signaling pathways. While many researchers have \nreported on the effects and roles of mechanical stimuli in various cell models, studies of the uterus in response to \nmechanical stimuli are limited. There have been only a few studies reporting the effect of mechanical stretch on \nhESCs, particularly focusing on the expression of specific genes such as IGFBP1 or interleukin-8 (IL-8), which are \nassociated with decidualization or inflammatio\n18,25. In this study, we aimed to understand the effects of mechani-\ncal stimuli on the hESCs, particularly on their contractility.\nWe applied 15% of uniaxial cyclic strain to hESCs at 0.1 Hz for up to 7 days. First, application of uniaxial cyclic \nstrain to hESCs induced rearrangement in the direction perpendicular to the strain axis while control cells were \nrandomly distributed. Moreover, both real-time PCR and immunostaining showed that the cyclic strain induced \nan increase in the expression of α-SMA. As well as the up-regulation of α-SMA, real-time PCR results showed a \nsignificant up-regulation in the mRNA expression of oxytocin receptor (OXTR) after loading cyclic strain for 7 \ndays, a gene which is highly expressed in the myometrium\n26–28. The oxytocin receptor, however, is also expressed \nin the endometrium and its expression varies during the non-pregnant cycle, depending on ovarian steroid hor-\nmones, such as progesterone and estrogen29,30. Kunz’s group showed that oxytocin increases the frequency of the \nendometrial wave31. Moreover, it is known that up-regulation of the oxytocin receptor before the onset of labor \nduring pregnancy induces the production of prostaglandin F2α , which results in an increase in endometrial con-\ntraction. This study is the first to suggest that mechanical stimuli play a role in endometrial stromal cells in the \nacquisition of contractility, by up-regulating α-SMA and oxytocin receptor expression.\nIn addition, we showed the strained cells to be different from myofibroblasts. While there is a lack of desmin \naccumulated in myofibroblasts, the expression of DES is known to be relatively abundant in the myometrium\n32,33. \nMoreover, IL6 production level is known to be elevated in myofibroblasts 34– 36. Application of cyclic strain \n\n7Scientific  RepoRtS  |         (2020) 10:9014  | https://doi.org/10.1038/s41598-020-65884-3\nwww.nature.com/scientificreportswww.nature.com/scientificreports/\nfor 7 days significantly induced the expression of DES  and also down-regulated the expression of IL6, which \nindicates the strained hESCs were distinct from myofibroblasts. On the other hand, an essential marker abun-\ndantly expressed in vascular smooth muscle cells, ANGPT1, was non-significantly but slightly reduced by cyclic \nstrain37–40. RAMP1 mRNA expression, which is specific to vascular smooth muscle cells in uterine arteries, was \nsignificantly down-regulated by cyclic strain, indicating that the strained cells were distinct from vascular smooth \nmuscle cells\n41–43.\nIn this study, a cAMP production assay was performed after loading strain on hESCs. cAMP is a secondary \nmessenger produced from adenosine triphosphate (ATP) and is known to regulate endometrial stromal cells for \ndecidualization during the menstrual cycle\n44. There is a report that addition of estradiol in uterine cells evoked an \nincrease in cAMP levels, and the cAMP pathway via adenylyl cyclase is involved in this mechanism45. Moreover, \nthe phenotype induced in bone marrow-derived MSCs by cAMP treatment suggests those cells could serve as a \nsource of endometrial stem/progenitor cells\n46. Thus, cAMP has a significant regulatory role in the uterus just like \nhormones such as estrogen and progesterone. In this study, we therefore focused on the involvement of cAMP in \nresponse to cyclic strain.\nHere we also report that cyclic strain up-regulated cAMP production in hESCs, implying that the cAMP sig-\nnaling pathway may be involved in the up-regulation of α-SMA expression under stretch. Applying cyclic strain \nfor as little as 15 mins induced the up-regulation of cAMP production in hESCs. After up-regulating α -SMA \nexpression in hESCs under cyclic strain for 7 days, we also examined whether cAMP production was responsive \nto strain. Since cAMP production is usually transiently induced, cells strained for 7 days were subjected to a \n2-hour break (static condition) to stabilize the level of cAMP , followed by 15 mins of cyclic strain. This resulted in \na significant up-regulation of cAMP production, showing that stretch was able to induce cAMP production both \nbefore and after applying cyclic strain for 7 days.\nIn order to determine the importance of cAMP in the stretch-induced up-regulation of α -SMA expression, \nwe carried out inhibitor tests using the adenylyl cyclase inhibitor SQ22536 and the PKA inhibitor H-89, since \nadenylyl cyclase regulates cAMP production while PKA is a well-known cAMP-dependent protein kinase\n47,48. \nFigure 4(G) schematically illustrates the signaling pathway activated in the hESCs under cyclic strain, as dis-\ncussed in this study. Adenylyl cyclase is an enzyme located on the inner side of the plasma membrane and usually \nactivated by G proteins. Activation of adenylyl cyclase under cyclic strain converts adenosine triphosphate (ATP) \nto cAMP , an intracellular second messenger. We showed that cyclic strain increased the cAMP production level in \nhESCs. This transient up-regulation of cAMP in hESCs by cyclic strain was consistent with previous studies using \nother cell models under mechanical stimuli such as cyclic strain or static compressive strain\n49,50. As suggested by \nthe inhibitor tests, the increase in intracellular cAMP levels is essential for the up-regulation of α-SMA expression \nin hESCs by cyclic strain. Interestingly, the addition of H-89 led to a decrease in the mRNA expression of SMC \nmarkers and OXTR under cyclic strain. The result may imply that H-89 did not only specifically inhibit the cAMP \nsignaling pathway, but also activated other signaling pathways which inhibit SMC markers and OXTR expression \nunder strain. Further studies will be required to address this point.\nThe cell contraction assay using collagen gels was then carried out to measure the cells’ contractile ability. As \na result, the samples strained for 7 days showed an increased contractile ability compared to control samples. \nMoreover, the contractility in the strained samples was significantly enhanced in the presence of oxytocin, indi-\ncating that the strained hESCs behaved like uterine smooth muscle cells. It has been reported that the enhanced \ncontractile ability of ESC may help to minimize defects in an endometrial wound model and promote endome-\ntrial tissue repair in vivo\n51. In ruminants, the level of oxytocin receptor is known to increase during the diestrus \nphase, reach its maximum value during the proestrus phase, and then decline during the estrus phase 52,53. \nPeristaltic patterns such as intrauterine pressure and strain have also been reported to change during the estrous \ncycle\n54. Up-regulation of OXTR mRNA expression induced by mechanical stimulation might contribute to stop-\nping the bleeding during the menstrual cycle by strengthening the contractility of the endometrium in response \nto oxytocin. Although the endometrium is exposed to a dynamic environment induced by the myometrium, the \neffect of mechanical stimuli on the ESCs with regard to their contractility remained unknown. In this study, we \nsuggested that the enhanced contractility in the strained cells was due to the up-regulation of α-SMA expression \nand the oxytocin receptor. While several studies have shown that biochemical stimulation using cytokines or \nplatelet-derived growth factor (PDGF) increased the contractility in hESCs in vitro\n51,55, we are the first to report \nthat mechanical stimuli also allowed endometrial stromal cells to acquire greater contractility while keeping their \noriginal cell phenotype. In other words, mechanical stimulation might help to control the dynamic and active \nfunctions of endometrial stromal cells. It has been reported that OXTR  is not expressed in stromal cells in vivo \nby immuno-staining\n27 or in situ hybridization56. However, in this paper, we reported that strain up-regulated \nOXTR mRNA expression in stromal cells by real-time PCR, which enables more sensitive signal detection than \nimmune-staining and in situ hybridization. However, our current studies have only been performed in vitro, and \nthe relevance of our findings will be examined during further studies. Although there are limitations to direct \nextrapolation of in vitro result to the in vivo context, passive strain stimulation of stromal cells caused by uterus \nSMCs might trigger active stromal cell contraction.\nIn summary, we report that applying uniaxial cyclic strain significantly up-regulates the expression of α-SMA \nas well as cAMP production. Together, the results show that strained hESCs acquire greater contractility, thus \nbehaving more like uterine smooth muscle cells. Furthermore, these findings may imply that contractile move-\nments by the myometrium have a significant role in inducing endometrial stromal cells to acquire the ability to \ncontract in vivo for physiological functions of the endometrium. This newly reported phenomenon might be a \ntypical example of how a tissue passively acquires new contractile functions under mechanical stimulation from \na neighboring tissue, enabling it to support the adjacent tissue’s functions.\n\n8Scientific  RepoRtS  |         (2020) 10:9014  | https://doi.org/10.1038/s41598-020-65884-3\nwww.nature.com/scientificreportswww.nature.com/scientificreports/\nMethods\nisolation and culture of h eScs. Endometrial biopsies were obtained from 38~48-year-old female \npatients who had regular menstrual cycles. Fresh human endometrial stromal cells were isolated and cultured \nas previously reported 57–59. The purity of the cell source was greater than 98% 59. This study was approved by \nthe Institutional Review Board of the University of Tokyo in accordance with the Declaration of Helsinki, and \neach patient gave informed consent for sample collection. We cultured hESCs in DMEM/Ham’s F12 (Sigma) \nsupplemented with 2.5% charcoal-stripped FBS (Funakoshi) and 1% of Antibiotic-Antimycotic (GIBCO) in a \nhumidified incubator at 37 °C with 5% CO\n2. The culture medium was changed every 3 or 4 days. For inhibitor \ntests, we used the adenylyl cyclase inhibitor SQ22536 and the protein kinase A (PKA) inhibitor H-89 (both from \nCayman Chemical). SQ22536 (100 μM) and H-89 (10 μM) were added to the cells’ culture medium just before \nloading cyclic strain.\nUniaxial cyclic strain loading. In this study, we loaded cyclic strain using the Flexcell tension system (FX-\n4000; Flexcell International Corporation) placed in a humidified incubator at 37 °C with 5% CO2. The Flexcell is \ncomputer-operated and applies its strain by vacuum. In this study, 15% uniaxial strain was applied at 0.1 Hz as \nillustrated in Fig. 1. During application of the uniaxial cyclic strain, the culture medium was changed every 2 or 3 \ndays. For control samples, hESCs were cultured in identical Flexcell plates but without strain.\nMeasurement of cell reorientation under cyclic strain.  In order to measure the cells’ orientation \nangles, we used normal microscope images (×5) taken from the center of Flexwell plate to cover a wide range of \nthe sample, that is further to avoid sample vialing. By using ImageJ, a line was drawn from the bottom of the cell to \nthe top of the cell along the cell’s major axis. After drawing the line, the angle between this line and the horizontal \naxis was measured using ImageJ. We selected 50 cells per each sample from 6 different experiments so as to select \nentirely in terms of area distribution.\nReal time-pcR. To measure mRNA expression in the different samples from the Flexcell system, we carried \nout real-time PCR. After finishing loading cyclic strain, the cells were rinsed with PBS and immediately lysed \nwith Trizol reagent (Invitrogen) before RNA extraction and cDNA synthesis using the ReverTra Ace qPCR RT \nMaster Mix with gDNA Remover (Toyobo). CD10 and CD90 were used as endometrial stromal cell markers while \nACTA2 and transgelin (TAGLN) are highly expressed in SMC. Oxytocin receptor (OXTR) was also examined as a \nuterine smooth muscle cell marker while desmin (DES) and interleukin 6 (IL6) were utilized to distinguish uter-\nine SMCs from myofibroblasts. Similarly, the expression of the vascular smooth muscle cell markers angiopoietin \n1 (ANGPT1) and receptor activity modifying protein 1 (RAMP1) were also measured. All genes were normalized \nto RPL32 expression and further normalized to the control samples. Primer sequences and amplicon sizes are \nlisted in Table 1.\nimmunostaining. After 7 days of cyclic strain, hESCs were immediately fixed with 4% paraformaldehyde. \nThe fixed samples were permeabilized with 0.2% Triton-X 100 in PBS for 3 min and washed with PBS 3 times. \nThen non-specific binding was blocked with PBS containing 1% BSA before covering the cells with anti-α-SMA \nantibody (Abcam) at a 1/500 dilution for 1 hour at room temperature or overnight at 4° C. After revealing the \nantibody using the DAB peroxidase substrate kit (Vector Laboratories), the samples were mounted on glass slides \nfor visualization and storage.\nc yclic adenosine monophosphate (cAMp ) measurement. The samples were collected and lysed \nwith 0.1 M HCl. After centrifugation at 1,000 g the supernatant was decanted and stored at − 80 °C until assay. \nCyclic AMP concentrations were measured using the cyclic AMP EIA kit (Cayman Chemical) according to the \nmanufacturer’s instructions. The absorbance at 412 nm was measured with an EnSpire Multimode Plate Reader \n(PerkinElmer). The cAMP concentrations were normalized to the DNA amounts quantified using the Quant-iT \nPicoGreen dsDNA Reagent and Kit (Invitrogen).\nc ell contraction assay. After the samples were subjected to cyclic strain for 7 days, they were trypsinized \nand resuspended in medium at a density of 2.0 ×  106 cells/ml. By using the collagen-based Cell Contraction \nGene Forward primer Reverse primer\nAmplicon \nsize (bp)\nRPL32 GCCCAAGATCGTCAAAAAGA GTCAATGCCTCTGGGTTT 98\nCD10 TCCACTGGAGATCAGCCTTT TATCGGGAACTGGTCTCAGG 237\nCD90 CTAGTGGACCAGAGCCTTCG TGGAGTGCACACGTGTAGGT 235\nTAGLN AGGTCTGGCTGAAGAATGGC TTCAAAGAGGTCAACAGTCTGG 199\nACTA2 CTGAGCGTGGCTATTCCTTC TTCTCAAGGGAGGATGAGGA 133\nOXTR TTCTTCGTGCAGATGTGGAG ACGAGTTCGTGGAAGAGGTG 149\nDES CTGAGCAAAGGGGTTCTGAG TGGCAGAGGGTCTCTGTCTT 135\nIL6 CACACAGACAGCCACTCACC TTTTCTGCCAGTGCCTCTTT 139\nANGPT1 GAAGGGAACCGAGCCTATTC GCTCTGTTTTCCTGCTGTCC 108\nRAMP1 CCTCACCCAGTTCCAGGTAG GAACCTGTCCACCTCTGCAT 157\nTable 1. Primer List.\n\n9Scientific  RepoRtS  |         (2020) 10:9014  | https://doi.org/10.1038/s41598-020-65884-3\nwww.nature.com/scientificreportswww.nature.com/scientificreports/\nAssay kit (Cell Biolabs, Inc.), we prepared a collagen lattice with bovine type I collagen at a concentration of \n3.0 mg/ml. As indicated in the protocol of the kit, the collagen gel was polymerized in the presence of cells and \nincubated for two days to allow stresses to develop within the gel. Before the stress was released by detaching \nthe gel from the culture dish, oxytocin (10 nM) was added to promote contraction as in SMC. Seven days after \nreleasing the gels from the culture dish, pictures were taken and the surface area of the sample was measured \nusing ImageJ.\nStatistical analysis. The statistical significance was assessed using Student’s t-test, F-test, or ANOV A fol-\nlowed by Fisher’s least significant difference (LSD). P-values below 0.05 were regarded as significant.\nReceived: 20 December 2019; Accepted: 21 April 2020;\nPublished: xx xx xxxx\nReferences\n 1. Kim, J., Montagne, K., Ushida, T. & Furukawa, K. Enhanced chondrogenesis with upregulation of PKR using a novel hydrostatic \npressure bioreactor. Bioscience, Biotechnology, and Biochemistry 79, 239–241 (2014).\n 2. Kim, J., Montagne, K., Nemoto, H., Ushida, T. & Furukawa, K. S. Hypergravity down-regulates c-fos gene expression via ROCK/\nRho-GTP and the PI3K signaling pathway in murine ATDC5 chondroprogenitor cells. PLoS ONE 12, e0185394 (2017).\n 3. Ting, S. Y . W ., Montagne, K., Nishimura, Y ., Ushida, T. & Furukawa, K. S. Modulation of the Effect of Transforming Growth \nFactor-β3 by Low-Intensity Pulsed Ultrasound on Scaffold-Free Dedifferentiated Articular Bovine Chondrocyte Tissues. Tissue Eng \nPart C Methods 21, 1005–1014 (2015).\n 4. Furukawa, K. S., Imura, K., Tateishi, T. & Ushida, T. Scaffold-free cartilage by rotational culture for tissue engineering. J. Biotechnol. \n133, 134–145 (2008).\n 5. Kim, J. & Adachi, T. Cell Condensation Triggers the Differentiation of Osteoblast Precursor Cells to Osteocyte-Like Cells. Front. \nBioeng. Biotechnol. 7, (2019).\n 6. Levasseur, M. C. Utero-ovarian relationships in placental mammals: role of uterus and embryo in the regulation of progesterone \nsecretion by the corpus luteum. A review. Reprod Nutr Dev 23, 793–816 (1983).\n 7. Padykula, H. A. Regeneration in the primate uterus: the role of stem cells. Ann. N. Y. Acad. Sci. 622, 47–56 (1991).\n 8. Chan, R. W . S., Schwab, K. E. & Gargett, C. E. Clonogenicity of human endometrial epithelial and stromal cells. Biol. Reprod. 70, \n1738–1750 (2004).\n 9. Bulletti, C. et al. Abnormal uterine contractility in nonpregnant women. Ann. N. Y. Acad. Sci. 828, 223–229 (1997).\n 10. Shynlova, O., Mitchell, J. A., Tsampalieros, A., Langille, B. L. & Lye, S. J. Progesterone and gravidity differentially regulate expression \nof extracellular matrix components in the pregnant rat myometrium. Biol. Reprod. 70, 986–992 (2004).\n 11. van Gestel, I., IJland, M. M., Hoogland, H. J. & Evers, J. L. H. Endometrial wave-like activity in the non-pregnant uterus. Hum. \nReprod. Update 9, 131–138 (2003).\n 12. Bulletti, C. & de Ziegler, D. Uterine contractility and embryo implantation. Curr. Opin. Obstet. Gynecol. 18, 473–484 (2006).\n 13. IJland, M. M., Evers, J. L., Dunselman, G. A., Volovics, L. & Hoogland, H. J. Relation between endometrial wavelike activity and \nfecundability in spontaneous cycles. Fertil. Steril. 67, 492–496 (1997).\n 14. Montagne, K. et al. High hydrostatic pressure induces pro-osteoarthritic changes in cartilage precursor cells: A transcriptome \nanalysis. PLoS ONE 12, e0183226 (2017).\n 15. Slager, C. J. et al. The role of shear stress in the generation of rupture-prone vulnerable plaques. Nat Rev Cardiol 2, 401–407 (2005).\n 16. Matsuura, K. et al. Improved development of mouse and human embryos using a tilting embryo culture system. Reproductive \nBioMedicine Online 20, 358–364 (2010).\n 17. Y oshino, O. et al. Decreased pregnancy rate is linked to abnormal uterine peristalsis caused by intramural fibroids. Hum Reprod 10, \n2475–2479 (2010).\n 18. Harada, M. et al. Mechanical stretch upregulates IGFBP-1 secretion from decidualized endometrial stromal cells. American Journal \nof Physiology - Endocrinology and Metabolism 290, E268–E272 (2006).\n 19. Ishisaki, A., Hayashi, H., Li, A.-J. & Imamura, T. Human Umbilical Vein Endothelium-derived Cells Retain Potential to Differentiate \ninto Smooth Muscle-like Cells. J. Biol. Chem. 278, 1303–1309 (2003).\n 20. Cevallos, M. et al. Cyclic strain induces expression of specific smooth muscle cell markers in human endothelial cells. Differentiation \n74, 552–561 (2006).\n 21. Shoajei, S., Shahdpour, M. T., Shokrgozar, M. A. & Haghighipour, N. Alteration of human umbilical vein endothelial cell gene \nexpression in different biomechanical environments. Cell Biology International 38, 577–581 (2014).\n 22. Konrad, L. et al. Composition of the stroma in the human endometrium and endometriosis. Reproductive Science 25(7), 1106–1115 \n(2018).\n 23. Tawfik, O. et al. Transgelin, a novel marker of smooth muscle cell differentiation, effectively distinguishes endometrial stromal \ntumors from uterine smooth muscle tumors. Int. J. Gynecol. Obstet. Reprod. Med. Res. 1(1), 26–31 (2014).\n 24. Santoso, E. G. et al. Application of detergents or high hydrostatic pressure as decellularization processes in uterine tissues and their \nsubsequent effects on in vivo uterine regeneration in murine models. PLoS One 24;9(7), e103201 (2014).\n 25. Harada, M. et al. Mechanical Stretch Stimulates Interleukin-8 Production in Endometrial Stromal Cells: Possible Implications in \nEndometrium-Related Events. The Journal of Clinical Endocrinology & Metabolism 90, 1144–1148 (2005).\n 26. Havelock, J. C. et al. Human myometrial gene expression before and during parturition. Biol. Reprod. 72, 707–719 (2005).\n 27. Loddenkemper, C. et al . Use of oxytocin receptor expression in distinguishing between uterine smooth muscle tumors and \nendometrial stromal sarcoma. Am. J. Surg. Pathol. 27, 1458–1462 (2003).\n 28. Blanks, A. M., Shmygol, A. & Thornton, S. Preterm labour. Myometrial function in prematurity. Best Pract Res Clin Obstet Gynaecol \n21, 807–819 (2007).\n 29. Kimura, T., Tanizawa, O., Mori, K., Brownstein, M. J. & Okayama, H. Structure and expression of a human oxytocin receptor. Nature \n1992 356:6369 356, 526–529 (1992).\n 30. Robinson, R. S., Mann, G. E., Lamming, G. E. & Wathes, D. C. The effect of pregnancy on the expression of uterine oxytocin, \noestrogen and progesterone receptors during early pregnancy in the cow. J Endocrinol 160, 21–33 (1999).\n 31. Kunz, G., Noe, M., Herbertz, M. & Leyendecker, G. Uterine peristalsis during the follicular phase of the menstrual cycle: effects of \noestrogen, antioestrogen and oxytocin. Hum. Reprod. Update 4, 647–654 (1998).\n 32. Schürch, W ., Seemayer, T. A., Lagacé, R. & Gabbiani, G. The intermediate filament cytoskeleton of myofibroblasts: an \nimmunofluorescence and ultrastructural study. Virchows Arch A Pathol Anat Histopathol 403, 323–336 (1984).\n 33. Eyden, B. P ., Hale, R. J., Richmond, I. & Buckley, C. H. Cytoskeletal filaments in the smooth muscle cells of uterine leiomyomata and \nmyometrium: an ultrastructural and immunohistochemical analysis. Virchows Arch A Pathol Anat Histopathol 420, 51–58 (1992).\n\n\n10Scientific  RepoRtS  |         (2020) 10:9014  | https://doi.org/10.1038/s41598-020-65884-3\nwww.nature.com/scientificreportswww.nature.com/scientificreports/\n 34. Yu, J. et al. IL-6 downregulates transcription of NTPDase2 via specific promoter elements. Am. J. Physiol. Gastrointest. Liver Physiol. \n294, G748–56 (2008).\n 35. Shimada, M. et al. IL-6 secretion by human pancreatic periacinar myofibroblasts in response to inflammatory mediators. J. Immunol. \n168, 861–868 (2002).\n 36. Catarzi, S. et al. Oxidative state and IL-6 production in intestinal myofibroblasts of Crohn’s disease patients. Inflamm. Bowel Dis. 17, \n1674–1684 (2011).\n 37. Iivanainen, E. et al. Angiopoietin-regulated recruitment of vascular smooth muscle cells by endothelial-derived heparin binding \nEGF-like growth factor. FASEB J 17, 1609–1621 (2003).\n 38. Jeansson, M. et al. Angiopoietin-1 is essential in mouse vasculature during development and in response to injury. The Journal of \nClinical Investigation 121, 2278–2289 (2011).\n 39. Hellström, M., Kalén, M., Lindahl, P ., Abramsson, A. & Betsholtz, C. Role of PDGF-B and PDGFR-beta in recruitment of vascular \nsmooth muscle cells and pericytes during embryonic blood vessel formation in the mouse. Development 126, 3047–3055 (1999).\n 40. Rensen, S. S. M., Doevendans, P . A. F . M. & van Eys, G. J. J. M. Regulation and characteristics of vascular smooth muscle cell \nphenotypic diversity. Neth Heart J 15, 100–108 (2007).\n 41. Frayon, S., Cueille, C., Gnidéhou, S., de Vernejoul, M. C. & Garel, J. M. Dexamethasone increases RAMP1 and CRLR mRNA \nexpressions in human vascular smooth muscle cells. Biochemical and Biophysical Research Communications 270, 1063–1067 (2000).\n 42. Zhang, Z., Dickerson, I. M. & Russo, A. F . Calcitonin gene-related peptide receptor activation by receptor activity-modifying \nprotein-1 gene transfer to vascular smooth muscle cells. Endocrinology 147, 1932–1940 (2006).\n 43. Nikitenko, L. L. et al. Differential and cell-specific expression of calcitonin receptor-like receptor and receptor activity modifying \nproteins in the human uterus. Mol. Hum. Reprod. 7, 655–664 (2001).\n 44. Logan, P . C., Ponnampalam, A. P ., Steiner, M. & Mitchell, M. D. Effect of cyclic AMP and estrogen/progesterone on the transcription \nof DNA methyltransferases during the decidualization of human endometrial stromal cells. Mol. Hum. Reprod. 19, 302–312 (2013).\n 45. Aronica, S. M., Kraus, W . L. & Katzenellenbogen, B. S. Estrogen action via the cAMP signaling pathway: stimulation of adenylate \ncyclase and cAMP-regulated gene transcription. Proc. Natl. Acad. Sci. USA 91, 8517–8521 (1994).\n 46. Aghajanova, L., Horcajadas, J. A., Esteban, F . J. & Giudice, L. C. The bone marrow-derived human mesenchymal stem cell: potential \nprogenitor of the endometrial stromal fibroblast. Biol. Reprod. 82, 1076–1087 (2010).\n 47. Walsh, D. A. & Van Patten, S. M. Multiple pathway signal transduction by the cAMP-dependent protein kinase. FASEB J  8, \n1227–1236 (1994).\n 48. Goel, M., Zuo, C.-D. & Schilling, W . P . Role of cAMP/PKA signaling cascade in vasopressin-induced trafficking of TRPC3 channels \nin principal cells of the collecting duct. Am. J. Physiol. Renal Physiol. 298, F988–96 (2010).\n 49. Ngan, P . et al. The interactive effects of mechanical stress and interleukin-1 beta on prostaglandin E and cyclic AMP production in \nhuman periodontal ligament fibroblasts in vitro: comparison with cloned osteoblastic cells of mouse (MC3T3-E1). Arch. Oral Biol. \n35, 717–725 (1990).\n 50. Fitzgerald, J. B. et al. Mechanical compression of cartilage explants induces multiple time-dependent gene expression patterns and \ninvolves intracellular calcium and cyclic AMP . J. Biol. Chem. 279, 19502–19511 (2004).\n 51. Matsumoto, H. et al. Regulation of proliferation, motility, and contractility of human endometrial stromal cells by platelet-derived \ngrowth factor. The Journal of Clinical Endocrinology & Metabolism 90, 3560–3567 (2005).\n 52. Murata, T., Narita, K., Honda, K. & Higuchi, T. Changes of receptor mRNAs for oxytocin and estrogen during the estrous cycle in \nrat uterus. J. Vet. Med. Sci. 65, 707–712 (2003).\n 53. Fuchs, A. R. et al. Oxytocin and vasopressin receptors in bovine endometrium and myometrium during the estrous cycle and early \npregnancy. Endocrinology 127, 629–636 (1990).\n 54. Bulletti, C. et al. Uterine contractility during the menstrual cycle. Human Reproduction 15, 81–89 (2000).\n 55. Kimatrai, M., Oliver, C., Abadía-Molina, A. C., García-Pacheco, J. M. & Olivares, E. G. Contractile activity of human decidual \nstromal cells. The Journal of Clinical Endocrinology & Metabolism 88, 844–849 (2003).\n 56. Takemura, M. et al. Expression and localization of oxytocin receptor gene in human uterine endometrium in relation to the \nmenstrual cycle. Endocrinology 132, 1830–1835 (1993).\n 57. Koga, K. et al. Demonstration of angiogenin in human endometrium and its enhanced expression in endometrial tissues in the \nsecretory phase and the decidua. The Journal of Clinical Endocrinology & Metabolism 86, 5609–5614 (2001).\n 58. Taguchi, A. et al. Resveratrol Enhances Apoptosis in Endometriotic Stromal Cells. Am. J. Reprod. Immunol. 75, 486–492 (2016).\n 59. Hirota, Y . et al. Possible implication of midkine in the development of endometriosis. Hum. Reprod. 20, 1084–1089 (2005).\nAcknowledgements\nThe authors thank Dr. Katie Ryan for helpful advice on cell reorientation under cyclic strain. This work was \nsupported by Research Fellowships from the Japan Society for the Promotion of Science for Y oung Scientists \n(Grant Numbers JP17J07735), Grants-in-aid for Scientific Research from the Japanese Ministry of Education, \nCulture, Sports, Science and Technology, and grants from the Translational Systems Biology and Medicine \nInitiative from the Ministry of Education, Culture, Sports, Science and Technology of Japan, and from the Japan \nAgency for Medical Research and Development (AMED; Development of Manufacturing System for Regenerative \nMedicinal Cells from Human Mesenchymal Stem Cells).\nAuthor contributions\nF .S.K., U.T., H.T., H.Y ., Y .O., and O.Y . conceived and supervised the study; K.J., F .S.K. and U.T. designed \nexperiments; K.J., and H.T. performed experiments; K.J. and M.K. analyzed data; K.J. wrote the manuscript; M.K., \nF .S.K. and U.T. made manuscript revisions.\ncompeting interests\nThe authors declare no competing interests.\nAdditional information\nCorrespondence and requests for materials should be addressed to K.S.F .\nReprints and permissions information is available at www.nature.com/reprints.\nPublisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and \ninstitutional affiliations.\n\n11Scientific  RepoRtS  |         (2020) 10:9014  | https://doi.org/10.1038/s41598-020-65884-3\nwww.nature.com/scientificreportswww.nature.com/scientificreports/\nOpen Access This article is licensed under a Creative Commons Attribution 4.0 International \nLicense, which permits use, sharing, adaptation, distribution and reproduction in any medium or \nformat, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Cre-\native Commons license, and indicate if changes were made. The images or other third party material in this \narticle are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the \nmaterial. If material is not included in the article’s Creative Commons license and your intended use is not per-\nmitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the \ncopyright holder. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/.\n \n© The Author(s) 2020","source_license":"CC0","license_restricted":false}