All experimental protocols for animal research were approved by the Institutional Animal Care and Use Committees at the University of Massachusetts Medical School (UMMS) (PROTO202000131) following the National Research Council Publication Guide for the Care and Use of Laboratory Animals and NIH Guide for the Care and Use of Laboratory Animal. Mice were maintained under a standard 12-h light/dark cycle (lights on at 7:00 AM) with food and water ad libitum (room temperature 22 ± 2°C). CD-1 and C57BL/6J mice were purchased from the Jackson Laboratories and bred in the UMMS animal care facility. Red-shifted genetically encoded calcium indicators report lines using R-CaMP1.07 were generated by Michael Kotlikoff at Cornell University (Ohkura et al., 2012) (CHROMus line acta2-R-CaMP1.07; Jackson stock #028345). αSMA-hrGFP mice were provided by Alan Fine at Boston University (Paez-Cortez et al., 2013).

To induce adenomyosis in mice, we used a tamoxifen protocol as developed by Parrott et al. (2001). The stomach of neonatal mice when filled with milk is clearly white and readily recognizable. In our preliminary study, we found it was straight-forward to deliver 20 μl solution via injection to the stomach as confirmed with Chicago Blue dye. The stomachs of female neonatal CD-1 mice were injected on days 1 to 5 after birth (day of birth = day 0) with 1 mg/kg tamoxifen suspended in a peanut oil/lecithin/condensed milk mixture (2:0.2:3 v/v). Control mice received vehicle only. Mice on day 14 (i.e., postnatal day 14, PND14) or PND55 ± 3 in estrus were humanely euthanized using CO₂ and uteri were removed for histological examination or functional tests. These two stages were chosen for investigation because by PND14, the uterus’ basic adult configuration is established (Hu et al., 2004a, b) and by PND55, mice mature sexually. Moreover, investigating two developmental stages of mice after tamoxifen treatment allowed us to explore the natural histology and the root cause of adenomyosis pathogenesis.

Sexually mature female mice (7 weeks of age) were mated to fertile males, and copulation was confirmed by observing vaginal plugs the following morning. The morning when the plug was observed was designated as 1 dpc (day post copulation).

Implantation sites were visualized in 5 dpc pregnant mice by tail vein injection of 200 μL of 1% Chicago Blue dye; mice were humanely euthanized 20 min after injection. Dissected uteri were photographed in white light, and blue bands were counted as implantation sites.

Previous in vivo studies show that uterine peristalsis results from circular muscle contraction and relaxation (Eytan et al., 2001; Bulletti et al., 2004; Brosens et al., 2010). Therefore, we cut the uterus perpendicularly to its long axis such that uterine slices were able to contract and relax in the same direction as uterine peristalsis in vivo. Mice were humanely euthanized by carbon dioxide, followed by cervical dislocation. The uterus was removed and transferred to ice-cold Hanks’ balanced salt solution (HBSS, Cat#1387, Sigma Aldrich), supplemented with 20 mM HEPES buffer and adjusted to pH 7.4 (named as sHBSS). After removing adipose and connective tissue, the half of the uterus near the ovarian side was cut and embedded in 5% low melt agarose gel, as was previously done in other types of smooth muscle tissue (Bergner and Sanderson, 2002; Lu et al., 2021a). After the agarose was gelled at 4°C, the specimen was loaded into a compresstome vibratome (VF-300; Precisionary Instruments) and sectioned into 250 μm thick slices. The slices were collected in a 24-well plate with cold sHBSS buffer and examined under an optical microscope. Those slices with intact circular smooth muscle rings were selected for subsequent experiments.

Uterine slices were placed on cover glass (45 mm × 50 mm) and held in position with a nylon mesh with a hole cut in the middle. Another smaller cover glass was placed on top of the nylon mesh and sealed on both sides with silicone grease to create a perfusion chamber. The chamber was mounted on an inverted microscope (IX71, Olympus) with a 2× or 4× objective (Nikon, Tokyo, Japan). The perfusion flow rate was ∼700 μl/min driven by a gravity perfusion system via a VC-6 six channel valve controller (Warner Instruments Corp.). All experiments were conducted in a custom-made Plexiglas chamber complete with a custom-built objective heater and a thermal controller to regulate the temperature at ∼33°C.

For uterine contraction measurements, phase-contrast images were collected at a rate of 1 Hz with a CCD camera controlled by the μManager image acquisition software in the time-lapse mode. The contraction and relaxation of the uterus, after a 30 min incubation, were represented by the change of the whole uterine slice area, as described previously (Tan and Sanderson, 2014; Lu et al., 2021a). Briefly, the whole slice’s area was measured by adjusting the gray level threshold in the 1st frame of the image series (ImageJ > Image > Adjust > threshold) such that the entire area enclosed by the uterus was a single particle (i.e., object), and then counting the pixels in that particle (ImageJ > analyze > analyze particles) in ImageJ¹ (NIH). Changes in the uterine area were calculated as (Area_t − Area₀)/Area₀^∗100, where Area₀ is the slice area when the uterus is at relaxation and Area_t is the slice area at a time point. As uterine peristalsis and Ca²⁺ signals (see below) occurred as bursts (i.e., comprised of multiple peaks) or spikes (i.e., single peaks), the frequency of events was expressed as either the number of bursts plus the number of isolated spikes per second or the number peaks per second. Uterine peristalsis frequency were determined using the local maximum method of the Peak Analyzer in Origin 2020b. The values of local points and amplitude height% were finely tuned such that more than 95% of peaks were found. The missing peaks or falsely selected peaks were manually added or deleted by confirmation via visual inspection of the original images. When events were determined based on bursts or spikes, their full width at half maximum (FWHM) values were calculated. Please note that the contraction spikes in this study may not be correlated one to one with underlying electrical signals, i.e., action potentials, or electrical spikes.

For the uterine Ca²⁺ signal measurement, slices were loaded with sHBSS with 20 μM Cal-520 AM (AAT Bioquest, Inc., Sunnyvale, CA, United States) and 0.06% Pluronic F-127 in the dark at 33°C for 50 min, followed by sHBSS alone for an additional 30 min at room temperature. Uterine slices were then mounted in the perfusion chamber (described above), and fluorescence images were obtained with μManager in a custom-built wide-field digital imaging system (Navaroli et al., 2012). The camera was interfaced to an IX71, Olympus inverted microscope, and images were recorded with a 2× or 4× objective at a speed of 1 Hz. The 488 nm line of an argon-ion laser provided fluorescence excitation, with a shutter to control exposure duration; emission of the Ca²⁺ indicator was monitored at wavelengths >500 nm. The uterine slices’ fluorescence intensity was defined as the average intensity value of the circular smooth muscle region. This region was manually outlined at key frames when the uterine slice’s contraction or relaxation initiated, and outlines on intermediate frames were created by the ImageJ interpolate ROIs function which assumes a linear progression between the key frames. A linear bleach correction with a slope calculated using a period without any Ca²⁺ events was applied using the ImageJ Stacks-T function plugin. Changes in fluorescence intensity were expressed as (F_t − F₀)/F₀^∗100, i.e., ΔF/F₀ × 100, where F_t was the fluorescence intensity at a particular time, and F₀ was the mean fluorescence at rest. The frequency and FWHM of Ca²⁺ oscillations were calculated as described above in Origin 2020b.

Protein distribution was assessed by immunohistochemistry as described (Lu et al., 2021a). Uterine cryosections were 8 μm, and antibodies were mouse monoclonal antibody to MYH11 (ab683 clone 1G12, 1:200; Abcam) and Alexa Fluor 488-conjugated goat anti-mouse IgG (Cell Signaling Technology, dilution 1:500). Immunoreactivity was evaluated using a Leica TCS SP8 confocal laser scanning microscope system (Leica Microsystems Inc., Buffalo Grove, IL, United States).

Uterine histology was assessed with hematoxylin–eosin stain according to the manufacturer’s instructions (ab245880, Abcam). The stainings were evaluated using a Zeiss Axioplan 2 microscope equipped with a BFLY-U3-23S6C-C USB 3.0 Blackfly camera (Edmund Optics, Barrington, NJ, United States).

Uteri from mice were isolated and cleaned by removing connective tissues. Total cellular RNAs from specimens were isolated, cleaned and reverse-transcribed as described (Lu et al., 2021b).

qRT-PCRs were carried out to determine mRNA levels of genes with iTaq^TM Universal SYBR^® Green Supermix (Cat #172-5121; Bio-rad) as described (Lu et al., 2021b). Each gene’s expression level was calculated using the 2^–ΔCt method and normalized against the housekeeping gene β-actin. Those with C_t > 35 were considered low precision and counted as no expression as a 35-cycle PCR reaction should amplify even a single transcript to a measurable level. All the primers are listed in Table 1.

Data are presented as means ± SE of N mice unless otherwise stated. Statistical analyses of characteristic differences of normal and adenomyotic mice were carried with unpaired Student’s t-test in the GraphPad Prism 8. The dose responses to oxytocin were analyzed using a two-way ANOVA, and implantation sites were tested with a Chi-Square statistic. The significance levels using mouse number were showed as follows: N.S., no statistical significance (P > 0.05), ^∗P < 0.05, ^∗∗P < 0.01, ^∗∗∗P < 0.001, ^****P < 0.0001.

We used a vibratome to cut the uterine transversely into 250 μm thickness slices to study uterine peristalsis. We first confirmed myometrial structure utilizing a line of the transgenic mouse in which hrGFP is expressed under the control of the α-smooth muscle actin (αSMA) gene promoter (Paez-Cortez et al., 2013). Figure 1A shows that the precision-cut uterine slice has two layers of GFP positive SMCs: the inner circular one as a ring and the outer longitudinal one as bundles. These structural arrangements in smooth muscle cells indicate uterine slices keep their in situ cell architecture.

We then assessed the contractile properties of the uterine slices. After a 30 min perfusion with Ca²⁺ buffer at 33°C, uterine slices developed stable spontaneous rhythmic phasic contractions and relaxations (Figure 1B) that persist for hours (data not shown). These activities changed the uterine slices’ size, i.e., the uterine slices underwent radial and isotonic contraction just as they would in vivo. As revealed in the Kymographs in Supplementary Figure 1, each rhythmic contraction and relaxation induced the same spatial change in the uterine slice. To quantify the contractions and relaxations, we measured the entire uterine slice’s size over time using ImageJ software’s particle detection function. Figure 1B(ii) shows the time course of changes in the size of a uterine slice. Given the mouse uterus is a tube structure, and coordinated uterine circular smooth muscle contractions traveling through the tube constitute peristalsis (van Gestel et al., 2003; Kissler et al., 2004; Lesny and Killick, 2004; Fanchin and Ayoubi, 2009; Moliner et al., 2021), it is arguable that rhythmic contractions and relaxations in uterine slices reflect uterine peristalsis in vivo. Significantly, the peristalsis frequency of uterine slices (see below) is close to that of the whole uterus in vivo (Crane and Martin, 1991; Meirzon et al., 2011; Zhang et al., 2019). Hence, we use the term “uterine peristalsis” to describe rhythmic contractions and relaxations in uterine slices from now on in this study.

To gain insight into the cellular signals responsible for uterine peristalsis in uterine slices, we measured Ca²⁺ signals with Cal-520 as Ca²⁺ is the primary signal for smooth muscle contraction. As shown in Figure 1B(i) and Supplementary Video 1, Ca²⁺ signals in the slices were not constant but oscillated. Moreover, Ca²⁺ oscillations were temporally associated with uterine peristalsis in a one-to-one manner [Figures 1B(i,ii),C and Supplementary Video 1]. Amplitudes of Ca²⁺ oscillations and changes in the uterine slice size showed a biphasic relationship (i.e., the two different fitting lines in two segments shown in Figures 1D,E). At the lower Ca²⁺ levels, there was a positive correlation indicated by the red fitting line between Ca²⁺ increase and the decrease in uterine area (r = 0.9659, p < 0.0001) when using the single event shown in Figure 1C (Figure 1D) and also when using all the events shown in Figure 1B (red fitting line, r = 0.8334, p < 0.0001) from the same slice (Figure 1E). But, at higher Ca²⁺ levels (i.e., those greater than 5% ΔF/F₀ in this set of data), there were no correlations (two blue fitting lines; r = 0.2051, p = 0.1191, Figure 1D; r = 0.1117, p < 0.0075, Figure 1E). Visual inspection showed that this occurred when both Ca²⁺ oscillation frequency and contraction magnitudes reached and stayed at a plateau. The results in Figure 1 indicate that Ca²⁺ oscillations underlie uterine peristalsis.

To determine whether Ca²⁺ oscillations occurring only in USMCs cause uterine peristalsis, we studied uterine slices from acta2-R-CaMP1.07 mice. In these mice, acta2-R-CaMP1.07, a Ca²⁺ reporter, is expressed only in smooth muscle cells (Ohkura et al., 2012; Lu et al., 2021a). Supplementary Figure 2 and Supplementary Video 2 show that Ca²⁺ oscillations revealed by R-CaMP1.07 signals were temporally associated with uterine peristalsis, arguing that Ca²⁺ signals indeed originate from USMCs and cause the peristalsis.

To investigate the potential changes in uterine peristalsis and Ca²⁺ signals in adenomyosis, we adapted a tamoxifen protocol developed by Parrott et al. (2001) to induce adenomyosis in CD-1 mice (Green et al., 2005; Mehasseb et al., 2009, 2010a). In control mice at PND14, the circular muscle layer was well developed to form an intact ring structure and did not contain endometrial tissues (Figure 2A, left). After treatment with tamoxifen, the circular muscle layer degenerated, with some muscle cells replaced with endometrial tissue and glands. However, the outer longitudinal layer appeared intact without invasion of endometrial tissue (Figure 2A, right). To examine the muscle layers’ changes, we immunostained uterine tissues with the antibody for Myh11, a smooth muscle-specific marker. As shown in Figure 2B (left), Myh11 positive cells formed a complete and intact circular ring and bundles in the longitudinal layer in control mice. However, in tamoxifen-treated mice, Myh11 positive cells in the circular muscle layer were interrupted by endometrial glands and stromal tissues, while we detected no such interruption in the longitudinal layer (Figure 2B, right).

In control mice at PND55, glands and muscle layers were more developed (Figure 2C) than PND14. Both glands and the circular muscle layer had clear boundaries, and no glands were detected in the circular muscle layer (Figure 2C, left). In tamoxifen-treated mice, the circular muscle layer degenerated further and was occupied and disrupted by variable sizes of endometrial glands (Figure 2C, right). These features were also evident when Myh11 was stained in both uterine slices (Figure 2D).

To further characterize the tamoxifen-induced adenomyosis model, we assessed the expression of genes that are potentially essential for uterine smooth muscle contraction (Wray, 1993; Bulletti et al., 2004; Sanborn, 2007; Young, 2007; Aguilar and Mitchell, 2010; Guerra and Hurt, 2019). The transcript for α-SMA, a primary structural protein in smooth muscle, was reduced by 41.38 ± 3.35% and 66.56 ± 0.88% at PND55 and PND14, respectively, in the tamoxifen-treated group compared to the control (Figure 3A). Among estrogen receptors, Esr1 expression was increased at both ages after treatment with tamoxifen, while Esr2 and Gper1 stayed low at both ages (Figure 3B). Oxytocin receptor was increased by 3.34 ± 0.89 fold at PND55 and 14.75 ± 4.88 fold at PND14 in the tamoxifen-treated group compared to the control group (Figure 3C).

We also measured the expression of the genes that may play a role in the pathogenesis of adenomyosis (Oh et al., 2013; Artymuk et al., 2019; Inoue et al., 2019; Kim et al., 2020; Adamyan et al., 2021). Kras and Tgb1 stayed at the same level in control and tamoxifen-treated mice at both ages (Figure 3C). Tgbr1 and β-catenin were reduced at PND55 while increased at PND14 in tamoxifen-treated mice compared to control mice (Figure 3C). E-cadherin expression was not different at PND55, but its level was increased by 2.39 ± 0.14 fold at PND14 (Figure 3B). Collagens 1 and 3 were increased in the tamoxifen-treated group at both ages compared to the control group (Figure 3A).

To evaluate the impact of adenomyosis on reproduction, we measured the success rate of embryo implantation and the number of implanted embryos at 5 dpc (a vaginal plug as day 1) (Figure 4A). We found that the rate of embryo implantation was markedly decreased in tamoxifen-treated mice (Figure 4B). For those mice which had successful implantations, the number of implanted embryos was significantly smaller in tamoxifen-treated mice than in control mice (Figure 4C). Hence tamoxifen-induced adenomyosis causes mice to become less fertile.

Having characterized the pathohistology of the mouse model of adenomyosis induced by tamoxifen, we next studied uterine peristalsis and Ca²⁺ signals in these mice using precision-cut uterine slices. Our initial preliminary experiments noted that the Ca²⁺ indicator, Cal-520, mildly altered the uterine peristalsis pattern. Therefore, to better determine uterine peristalsis characteristics in normal and adenomyotic mice, we examined uterine peristalsis without Cal-520. Figures 5A,B show typical recordings of uterine peristalsis in control CD-1 mice and adenomyotic CD-1 mice at PND55. Of note, in control mice, uterine peristalsis occurred as bursts that consisted of multiple peaks (Figure 5A). However, in adenomyotic mice, uterine peristalsis appeared as spikes with a single peak or bursts with a few peaks (Figure 5B). Given the unique contractile patterns of uterine peristalsis, we analyzed the uterine peristalsis frequency in two ways. One is to count a burst or a spike as one peristalsis event, and the other an individual peak within bursts or in spikes as one event. When counting based on the burst and spike, uterine peristalsis frequency was significantly increased in adenomyotic mice compared with control mice (Figure 5C). However, when counting based on the peak, uterine peristalsis frequency was not significantly different between the two groups (Figure 5D). On average, peristaltic burst in control mice oscillated 7.57 ± 0.51 times, while in adenomyotic mice, this value was 4.13 ± 0.73 with over 50% of events presenting as spikes (i.e., with just one peak) (Figure 5E). As expected, full width at half maximum (FWHM) of the uterine peristalsis was significantly longer in control mice than in adenomyotic mice when bursts or spikes were counted as events (Figure 5F).

Force generated by uterine peristalsis should be proportional to the change in the uterine area. We found that change in uterine area per uterine peristalsis (with the burst or spike as the event) was smaller in adenomyotic uterine slices than in control slices in terms of an absolute value (Figure 5G) or a relative value (Figure 5H). To further quantify the change in force, we calculated the product of frequency (Figure 5C) and area under the curve (AUC), the area between the curve and the x-axis (Figure 5I), designated as contraction intensity (Figure 5J). Figure 5J reveals that contraction intensity was weaker in adenomyotic uterine slices than in control slices.

The above analyses indicate that uterine peristalsis activates as bursts in normal adult mice, while it turns into spikes in adenomyotic mature mice. Overall, uterine peristalsis in adult adenomyotic mice produces weaker contractions than in normal adult mice.

We loaded uterine slices with Cal-520 to get insight into the characteristics of Ca²⁺ oscillations underlying normal and dysfunctional uterine peristalsis. As shown in Figures 6A,B where typical Ca²⁺ oscillations from mature control mice and adenomyotic mice are displayed, Ca²⁺ oscillations exhibited as bursts in control slices but as spikes in adenomyotic slices. When counting a burst or a spike as an event, the frequency of Ca²⁺ oscillations was higher in adenomyotic slices than in control slices (Figure 6C). However, when counting based on the peaks, the frequency of Ca²⁺ oscillations was not statistically different in both control slices and adenomyotic slices (Figure 6D). The number of oscillations per event (Figure 6E) and FWHM (Figure 6F) in the slices from adult adenomyotic mice were smaller than those from control mice.

We further determined the amplitude of Ca²⁺ oscillations in both types of uterine slices. We found that amplitude with ΔF/F₀ as a proxy of intracellular Ca²⁺ concentration ([Ca²⁺]_i) was smaller in adenomyotic slices than in control slices (Figure 6G). Finally, we calculated the product of frequency and AUC, designated as calcium intensity. As shown in Figure 6H, the calcium intensity was significantly smaller in adenomyotic slices than control slices. These results indicate that Ca²⁺ oscillations change from bursts in adult control uteri into spikes with weaker intensity in adult adenomyotic uteri, consistent with uterine peristalsis features in Figure 5.

Uterine histoarchitecture is well established at PND 14 in mice (Figure 2A). However, to the best of our knowledge, no study on uterine peristalsis at this stage in normal mice and adenomyotic mice exists. Figure 7A shows typical uterine peristalsis at low and high temporal resolutions from a neonatal control mouse. Uterine peristalsis rose quickly and was maintained at a plateau with small oscillations. Visual inspection of the uterine slice indicates that the sustained plateau resulted from sequential contractions. The frequency of uterine peristalsis at this stage was 0.0067 ± 0.00062 Hz when each burst or spike was counted as one event (Figure 7C). This value was 0.075 ± 0.0051 Hz when each peak was counted as one event (Figure 7D). On average, a burst was the sum of 12.16 ± 0.55 individual contractions (Figure 7E), lasting for 60.57 ± 3.34 seconds at FWHM (Figure 7F).

In adenomyotic mice at PND 14, uterine peristalsis comprised a single contraction or occasional multiple contractions, as shown in both low and high temporal resolution recordings (Figure 7B). Overall, uterine peristalsis at PND14 in adenomyotic mice comprised many fewer contractions (Figure 7E) and lasted a much shorter time, as assessed by FWHM (Figure 7F), when compared to control mice. Uterine peristalsis frequency was higher in adenomyotic mice than in control mice (Figure 7C) when a burst or a spike was counted as an event. However, there was no difference when the frequency calculation was based peak numbers (Figure 7D).

To assess the force generated by uterine peristalsis, we compared changes in the uterine area and contraction intensity between control mice and adenomyotic mice at PND14. Both uterine area changes in either absolute terms (Figure 7G) or relative terms (Figure 7H) were smaller in adenomyotic mice than in control mice. The AUC (Figure 7I) and contraction intensity (Figure 7J) caused by uterine peristalsis were smaller and weaker, respectively, in adenomyotic mice than in control mice.

Loading the slices with Cal-520 enabled us to measure Ca²⁺ signals corresponding to uterine peristalsis in control and adenomyotic mice at PND14. Ca²⁺ signals exhibited as bursts in control mice (Figure 8A). In adenomyotic mice, Ca²⁺ signals were largely spikes with a quick rise and a slower decay (Figure 8B). When a burst or a spike was counted as an event, Ca²⁺ oscillations had a higher frequency (Figure 8C) and shorter FWHM (Figure 8F) in adenomyotic mice than in control mice. When counting events based on the peak, Ca²⁺ oscillation frequency was not significantly different between control and adenomyotic groups (Figure 8D); Each burst comprised 11.97 ± 1.43 oscillations in control mice and 1.66 ± 0.15 in adenomyotic mice (Figure 8E). ΔF/F₀, i.e., [Ca²⁺]_i, and calcium intensity were smaller in adenomyotic mice than in control mice at PND14 (Figures 8G,H).

Oxytocin contributes to adenomyosis as proposed by the ‘tissue injury and repair’ hypothesis (Nie et al., 2010; Leyendecker and Wildt, 2011), hence we examined whether oxytocin affects Ca²⁺ signals and uterine peristalsis. At PND55, oxytocin increased the frequency of uterine peristalsis and Ca²⁺ oscillations in a dose-dependent manner in control mice and adenomyosis mice when bursts or spikes were used for event counting (Figure 9). For uterine peristalsis, oxytocin increased its frequency throughout the dose range tested (Figure 9B), but for Ca²⁺ responses, oxytocin at 0.1 μM yielded the maximal increase (Figure 9E). However, there were no significant differences in both responses between two groups (Figures 9B,E).

Besides frequency, oxytocin also increased amplitudes of uterine peristalsis and Ca²⁺ oscillations. Because oxytocin changed the signals’ baseline, we calculated AUCs to reflect changes in amplitude of both responses by oxytocin. We found that oxytocin dose-dependently increased the AUCs of uterine peristalsis (Figure 9C) and Ca²⁺ signals (Figure 9F). However, we found no difference in both responses between control mice and adenomyotic mice at PND55 (Figures 9C,F).

As there is no study on oxytocin’s effect on uterus’s contractile responses in neonatal mice, we studied them. Figure 10A shows original traces of uterine peristalsis in response to oxytocin in control mice and adenomyosis mice at PND14. In contrast to adult mice at PND55, there was a striking difference in response to oxytocin between control mice and adenomyotic mice at PND14. In control mice, oxytocin at concentrations between 10^–9 to 3 × 10^–5 M only modestly increased uterine peristalsis frequency and AUC (Figures 10B,C). However, in adenomyotic mice, oxytocin at 10^–9 M caused a marked increase in uterine peristalsis both in terms of frequency and AUC. This effect was dose-dependent within the range of 10^–9 to 3 × 10^–6 M (Figures 10B,C).

We further examined Ca²⁺ signals in response to oxytocin in control and adenomyotic mice at PND14 (Figure 10D). Like uterine peristalsis, oxytocin increased the frequency of Ca²⁺ oscillations in a dose-dependent manner in both control and adenomyosis mice (Figure 10E). However, it caused these oscillations at lower concentrations in adenomyotic mice. It also increased the oscillations more in adenomyotic mice than in control mice (Figures 10E,F). Both Ca²⁺ signals and uterine peristalsis analyses show that adenomyotic uteri at PND14 became hypersensitive to oxytocin.