Characterization of Mechanical Signature of Eutopic Endometrial Stromal Cells of Endometriosis Patients

Endometriosis affects 5–10% of women in reproductive age and causes pelvic pain and subfertility. Exact etiology of the disease is unknown. Here, we present a microfluidic platform for characterizing mechanical properties of eutopic endometrial stromal cells of endometriosis patients based on cellular deformability inside narrow microchannels. Primary human endometrial stromal cells were isolated from eutopic endometrium of endometriosis patients (4407 cells, from 7 endometriosis patients) and from disease-free women (4541 cells, from 6 control women) and were pumped through microchannels (formed of polydimethylsi- loxane (PDMS) by standard soft lithography, with dimensions of 8 × 20 × 150μm, as width × height × length) at a constant flow rate of 2 μL/min. High-speed imaging was used to capture videos of cells as they flow inside microchannels, and a computer vision code was used to track cells, measure their area, and calculate the time each cell takes to pass through the microchannel. Compared with their counterparts from control women, eutopic endometrial stromal cells from endometriosis patients showed significantly increased deformation index (1.65 ± 0.2 versus 1.43 ± 0.19, respectively, P value < 0.001), and higher velocity in travelling through narrow microchannels (96.530 ± 0.710 mm/s versus 57.518 ± 0.585 mm/s, respectively,P value < 0.001). The same difference in velocities between the two cell types was maintained after controlling for cell area. Eutopic endometrial stromal cells of endometriosis patients showed a mechanical phenotype characterized by high deformability and reduced stiff- ness. This mechanical signature can represent basis of a mechanical biomarker of endometriosis.

Mechanical stiffness . Endometrium . Stromal cells . Endometriosis . Microfluidics

Endometriosis is an estrogen-dependent disorder character- ized by the presence of endometrial glands and stroma outside the uterine cavity [ 1]. It affects around 5 –10% of women in their reproductive years, and up to 50% of women with chron- ic pelvic pain and/or infertility [2]. Treatment is either medical or surgical, and symptoms often recur after treatment discon- tinuation [ 3]. Severity of the symptoms, expenses and side effects of medications, and the need for multiple surgeries all reduce the quality of life of affected women [4]. Adding to the suffering of endometriosis patients are reports of increased risk of ovarian cancer in these patients [ 5]. The gold standard of endometriosis diagnosis is surgery through direct laparoscopic visualization of lesions [ 6]. Laparoscopy is invasive and requires anesthesia. It is not with- out risks including injury to a viscus, hemorrhage, or infec- tion. Moreover, laparoscopy is expensive and depends on the level of training of the surgeon [ 7]. Because of these Electronic supplementary material The online version of this article (https://doi.org/10.1007/s43032-019-00042-3) contains supplementary material, which is available to authorized users. * Essam Othman [email protected] * Mohamed Abdelgawad [email protected] 1 Center for Nanotechnology, Zewail City of Science and Technology, Giza, Egypt 2 Department of Obstetrics and Gynecology, Assiut University, Assiut, Egypt 3 Amsterdam University Medical Center, Location VUmc, Academic Endometriosis Center, Amsterdam, The Netherlands 4 Mechanical Engineering department, Assiut University, Assiut, Egypt 5 Department of Medical Microbiology and Immunology, Assiut University, Assiut, Egypt 6 Mechanical Engineering Department, American University of Sharjah, Sharjah, UAE Reproductive Sciences (2020) 27:364–374 https://doi.org/10.1007/s43032-019-00042-3 drawbacks, endometriosis patients suffer for 8–11 years from endometriosis-associated symptoms before a definitive diag- nosis is made [ 8]. A great deal of the puzzling nature of endometriosis stems from the fact that its etiology is not exactly known [ 9]. Described for the first time in 1927, the retrograde menstrua- tion theory of Sampson is a hypothesis that stood the test of time. It states that during menstruation, some viable endome- trial cells flow in a retrograde manner, against current, through the Fallopian tubes to reach the peritoneal cavity [10]. It is the most accepted theory by researchers to explain the develop- ment of endometriosis. However, because retrograde menstru- ation occurs in most cycling women with patent Fallopian tubes, other permissive factors must be playing in the back- ground to determine why a particular subgroup, not all, of women develops endometriosis [11]. Despite being a benign disorder, endometriosis has some common features with cancer [ 12, 13]. Endometriosis cells show high proliferation, enhanced survival, and diminished apoptosis [14], and the disease often recurs after surgical re- moval [ 15]. Moreover, endometriosis cells may disseminate via blood or lymphatic streams to initiate lesions in distant locations [ 16]. In addition, deep infiltrating endometriosis has been shown to harbor cancer-driver mutations [ 17]. Such similarities provoked investigators to look into some patho- physiologic mechanisms in cancer to see if they are operating in endometriosis as well. As cells change from benign to malignant nature, they un- dergo several changes at the level of their cytoskeleton [ 18]. This renders malignant cells several folds more elastic than their benign counterparts [19]. Being softer and more deform- able, metastatic cancer cells can penetrate through the colla- gen cross-linked fibers of damaged extracellular matrix at their primary site and squeeze themselves through narrow spaces between endothelial cells lining blood vessels to enter into the circulation. In addition, the high elasticity of malig- nant cells enables them to tolerate mechanical shears imposed on them by the circulation which is a main defense mechanism of the body against cancer metastasis [20]. Similarly, in endo- metriosis, endometrial cells leave their location in the uterus, flow against the current in the Fallopian tubes, and penetrate through damaged extracellular matrix of the peritoneal sur- faces where they implant. They can also squeeze themselves between endothelial cells lining blood or lymphatic capillaries to get implanted in faraway organs such as the lungs [ 16]. To this end, we hypothesize that eutopic endometrial stro- mal cells of endometriosis patients show lower mechanical stiffness and higher elasticity/deformability, contributing to their migratory and invasive nature, compared with their counterparts from disease-free women. Microfluidics emerged recently as a new high throughput tool for mechanical characterization of cells. Due to its matching size scale, microfluidics offers complete control over the ex vivo cellular environmentincluding the ability to induce mechanical stresses. Consequently, many microfluidics-based techniques were developed in the last decade to measure me- chanical stiffness of cells. These include wall-induced forces [21], shear stress–induced forces [22], hydrodynamic stretching [23], and optical stretchers [24]. Here, we propose a microfluidics-based platform for iden- tifying the mechanical properties of eutopic endometrial stro- mal cells of endometriosis patients. Our platform is based on wall-induced forces technique [21] in which cells are forced to flow inside a narrow microchannel with channel width smaller than cell size. Soft cells can deform easily and squeeze through this narrow microchannel, whereas stiff cells cannot easily deform and take longer time to flow through the same microchannel. Our findings have shown that cells from endo- metriosis patients could easily deform and flow inside the narrow microchannel at much higher velocities compared with cells from healthy women.

Human Endometrial Sample Collection We obtained endometrial tissue biopsies from 7 endometriosis patients and 6 control women. Endometriosis was diagnosed during laparoscopy performed to diagnose the cause of pelvic pain and/or infertility experienced by these women. The con- trol group consisted of women whose laparoscopic examina- tion revealed no abnormalities. Endometrial biopsies were ob- tained via curettage during the time of laparoscopic surgery in one session for the convenience of the patient and the research team. The same endometrial biopsy can be taken as an outpa- tient procedure using a pipelle device in office setting, without anesthesia. All laparoscopic surgeries were performed in the prolifer- ative phase of the menstrual cycle. Menstrual dates were assessed based on patient’s menstrual history and histological examination of endometrial biopsy. Women were excluded from the study if they had irregular menstrual cycles, had any pelvic pathology other th an endometriosis (such as adenomyosis, fibroids, non-endometriotic adhesions, etc.), re- ceived hormonal treatment, or were pregnant in the last 3 months before surgery. All participating women provided written informed consent. Institutional Review Board at Faculty of Medicine, Assiut University, approved the use of human endometrial tissue samples for this study. Endometrial Stromal Cell Culture Endometrial stromal cell cultures were established from eutopic endometrial biopsies of endometriosis patients and control women as described before [ 25]. Briefly, endometrial Reprod. Sci. (2020) 27:364–374 365 biopsies were sent immediately to the lab where they were washed several times with phosphate buffered saline (PBS) to remove blood clots. Endometrial tissues were minced into 1–2-mm pieces using small scalpel before they were enzymat- ically digested. Digestion was done in 10 ml of PBS contain- ing 0.1% collagenase type I and continued for 2 h in a shaking water bath at 37 °C. Endometrial stromal cells were separated through filtration using 40 and 20 μms i e v e s .T h ef i l t r a t ew a s centrifuged at 1200 rpm at room temperature for 10 min, and supernatant discarded. The cellular pellet was dissolved in 10 ml of DMEM-F12 supplemented with 10% fetal bovine serum (FBS) and cultured in cell culture incubator at 37 °C and 5% CO 2 until confluence. At the time of microfluidics experiment, cells were harvested with 0.05% trypsin-EDTA and washed with PBS, and cell suspension was loaded to the system. Characterization of Endometrial Stromal Cells Cells were stained with anti-human CD90 FITC –conjugated antibodies (Novus Biologicals; Centennial, CO, USA, cata- logue# NBP1 –96125), anti-human Vimentin Alexa Fluor® 488–conjugated monoclonal antibodies (R&D Systems; Minneapolis, MN, USA, cat alogue# IC2105G) or anti- cytokeratin (ThermoFisher; Waltham, MA, USA, catalogue#MA5 –18158) Alexa Fluor® 488 –conjugated monoclonal antibodies following fixation, permeabilization and intracellular staining against isotype-matched controls. Cells were acquired using the FACSCalibur flow cytometer and the Cell Quest Pro software. The acquired data were an- alyzed using FlowJo software. Microchannel Fabrication Standard soft lithography technique was used to fabricate the microfluidic devices [ 26]. A 20- μm-thick layer of negative photoresist (SU-8-2010) was deposited on a clean, dry silicon wafer by spin-coating (1000 rpm, 1 min) followed by soft bake (65 °C for 2 min followed by 5 min at 95 °C). Microchannel design was patterned using UV micro-pattern generator ( μPG101, Heidelberg Instruments, Heidelberg, Germany) at 60% of the power of the 70 mW UV diode. The SU-8 layer was baked again (95 °C, 12 min) then devel- oped for 6 min using diacetone alcohol (Sigma, Cairo, Egypt). Mechanical properties of the SU-8 layer were improved by hard baking at 200 °C for 30 min. Polydimethylsiloxane (PDMS) was prepared by mixing prepolymer base and curing agent at 10:1 ratio by weight. Negative replica of the master was created by casting in PDMS. The PDMS was cured at 100 °C for 45 min then it was peeled off; the master and holes were punched for inlets and outlets. PDMS slabs were bonded to clean microscope slides after treating it with a portable corona treater (Electro-Technic Products, Chicago, IL) for 2 min, as described elsewhere [ 27]. The final fabricated microchannel had constrictions of the following dimensions: 8 μm×2 0 μm × 150 μm (width × height × length). Experimental Setup The experimental setup used for characterizing the mechanical properties of endometrial ce lls is shown schematically in Fig. 1. The platform was placed on the stage of an inverted microscope (Olympus CKX53, Shinjuku, Tokyo, Japan), which was working in phase contrast mode with × 40 objec- tive. We typically started each experiment with 0.5 ml of cell suspension at a density of 7 × 10 5 cells/ml. A syringe pump (NE 4000, New Era Pump Systems, Farmingdale, NY , USA) was used to pump the cell suspension into the microchannel at af l o wr a t eo f2 μL/min. A high-speed camera (Basler ACA2000-340 km, Basler, Ahrensburg, Germany) was used to capture images of the flowing cells inside the constriction microchannel. Images were acquired using software built on LabVIEW (National Instruments, Austin, TX, USA). The frame rate of all recorded videos was around 3000~4000 frame per second. These images were transferred to a comput- er to be analyzed offline using computer vision. Image Analysis We used an in-house built computer vision code based on LabVIEW software [ 28], to analyze the recorded videos of cells flowing inside the microchannel. The computer vision code detects and tracks the cell once it enters the constriction and calculates its projected area which was used as an indica- tion of cell size. The real-time velocity of cells inside the constriction was calculated by measuring the distance be- tween the center of mass of the cell in two successive frames as in Eq. 1. Velocity ¼ traveled distance in two successive frames frame time ð1Þ The average cell velocity inside the microchannel was cal- culated by averaging the real-time velocities all over the con- striction length. Visual inspection was used to confirm that each cell flowed nonstop inside the microchannel and kept continuous contact with microchannel walls. Any cell that did not satisfy these two conditions, due to channel blockage or being small in size, was excluded from data analysis. Choice of the proper width of the microchannel was important to enable testing the maximum number of cells in light of the above two criteria. We tested different channel widths (7, 8, 10, and 12 μm) before we decided to use the reported width of 8 μm. Channels with smaller width got blocked frequently, which affected the throughput of the experiments. Whereas for 10 and 12 μm channels, some cells passed the channel 366 Reprod. Sci. (2020) 27:364–374 without touching channel walls or without deforming (diam- eter of endometrial stromal cells ranges between 8 and 30 μm). Also, data of multiple cells flowing through the chan- nel together were excluded and not used in data analysis. Statistical Analysis Statistical analysis was done using Statistical Package of Social Scientists (SPSS), version 20 (Chicago, IL, USA). Data analysis was done using parametric statistics. Groups were compared using the independent sample Student’s t test. V elocity of eutopic endometrial stromal cells from endometri- osis patients and control women was expressed as mean ± standard error of the mean (SEM). Statistical significance was reached if P <0 . 0 5 .

Clinical Characteristics of Study Subjects These are summarized in Table 1. Characterization of Endometrial Stromal Cells Our endometrial stromal cells were more than 95% positive for anti-human Vimentin and anti-human CD90 and were negative for anti-cytokeratin (Fig. 2), verifying their identity as endometrial stromal cells. Phases of Endometrial Stromal Cellular Passage Through Microchannels While flowing inside the microchannel, eutopic endometrial stromal cells from endometriosis patients showed a different behavior compared with their counterparts from healthy wom- en. Cell passage through the microchannel can be divided into two phases: transient and equilibrium. In the transient phase, the cell deforms gradually from spherical shape to a plug-like shape to fit inside the microchannel. In the equilibrium phase, the cell stops deforming and flows inside the microchannel with almost a constant velocity (i.e., linear profile of distance versus time). Eutopic endometrial stromal cells from endome- triosis patients took less time to deform in the transient phase and also passed through the microchannel in a shorter total time (as shown Fig. 3), indicating that they may be less stiff than cells from healthy women. Deformability of Eutopic Endometrial Stromal Cells as They Travel Through Microchannels Deformation index of cells, which is defined as the ratio between cell length inside the microchannel divided by its original diameter, was found to be higher in cells from endometriosis patients than that from control women (Fig. 4). Analysis of 234 cells from patients and controls showed that endometrial cells f rom endometriosis patients had a significantly higher deformation index of 1.65 ± 0.2 (mean ± SD) compared with a deformation index of 1.43 ± 0.19 (mean ± SD) for cells from healthy women ( P value < 0.001, Table 2). Fig. 1 Schematic of the time of flight principle for differentiating between cells based on its deformability. A high-speed camera (HSC) was used to capture images of cells as they flow through the region of interest (ROI) and the velocity of each cell inside the microchannel was calculated Reprod. Sci. (2020) 27:364–374 367 Velocity of Eutopic Endometrial Stromal Cells Derived from Endometriosis Patients and Control Women as a Surrogate of Their Cellular Stiffness We assessed the velocity of 4407 individual eutopic endome- trial stromal cells derived from seven endometriosis patients and 4541 cells derived from six control women inside our microchannel system. As shown in Fig. 5, the mean velocity of eutopic endometrial stromal cells derived from all endome- triosis patients (96.530 ± 0.710 mm/s) is significantly higher than that of their counterparts derived from control women (57.518 ± 0.585 mm/s); P value is < 0.001. Supplementary table 1 shows velocities of eutopic endometrial stromal cells derived from individual endometriosis cases and control women. Supplementary video S1 shows flow of eutopic en- dometrial stromal cells from a control woman inside the microchannel constriction. Supplementary video S2 is the same as video S1 after being processed using the machine vision program which shows velocity of each cell while flowing inside the microchannel. Supplementary videos S3 and S4 are similar videos but for eutopic endometrial stromal cells from an endometriosis patient. Table 1 Clinical characteristics of study subject Case/control Age Indication of surgery Operative findings Endometriosis case# 1 30 years Secondary infertility Peritoneal endometriotic spots Endometriosis case# 2 31 years Secondary infertility Ovarian endometriotic cyst Endometriosis case# 3 29 years Primary infertility Ovarian endometriotic cyst Endometriosis case# 4 26 years Primary infertility Ovarian endometriotic cyst Endometriosis case# 5 25 years Chronic pelvic pain Ovarian endometriotic cyst Endometriosis case# 6 30 years Primary infertility Bilateral ovarian endometriotic cysts Endometriosis case# 7 32 years Chronic pelvic pain Peritoneal endometriotic spots Control# 1 30 years Primary infertility Normal laparoscopic findings Control# 2 36 years Secondary infertility Normal laparoscopic findings Control# 3 34 years Primary infertility Normal laparoscopic findings Control# 4 32 years Secondary infertility Normal laparoscopic findings Control# 5 32 years Primary infertility Normal laparoscopic findings Control# 6 25 years Secondary infertility Normal laparoscopic findings Fig. 2 Characterization of eutopic endometrial stromal cells. Flow cytometry analysis after staining with a Vimentin Alexa Fluor® 488 – conjugated monoclonal antibodies, b CD90 FITC –conjugated antibodies, and c anti-cytokeratin Alexa Fluor® 488 –conjugated monoclonal antibodies . Shaded histograms were stained with representative isotype-matched control antibodies 368 Reprod. Sci. (2020) 27:364–374 Distribution of Velocities of Eutopic Endometrial Stromal Cells Inside Microchannels When a histogram is plotted for the distribution of velocities of eutopic endometrial stromal cells from endometriosis pa- tients and control women inside the microchannels, cells from endometriosis patients show more or less a normal distribution curve (skewness is 0.35), whereas cells from control women Fig. 3 I Series of pictures showing an endometrial cell while deforming to squeeze through the narrow microchannel. II Series of pictures of a cell passing through a microchannel (8 μmW×2 0 μmH× 150 μm L). The machine vision code we developed captured the cell once it enters the microchannel and calculated its area and the time it took to pass through the microchannel. III Displacement vs. time curve for two cells imposed on the same figure, one cell from a healthy woman (cell area = 254μm 2), and one cell from an endometriosis patient (cell area = 242 μm2). The cell from the endometriosis patient passed through the channel in a much shorter time (i.e., with higher velocity) Fig. 4 S c a t t e rp l o to fd e f o r m a t i o nindex (cell length inside the microchannel divided by its original diameter) of cells from healthy women and endometriosis patients. Figure is based on analysis of 125 cells from endometriosis patients and 109 cells from healthy women. Endometrial cells from endometriosis patients had a significantly higher deformation index of 1.65 ± 0.2 (mean ± SD) compared with a deformation index of 1.43 ± 0.19 (mean ± SD) for cells from control women Table 2 Average deformation index of eutopic endometrial stromal cells of endometriosis patients and control cells. Data expressed as mean ± standard deviation Endometriosis cells (n =1 2 4 ) Control women cells (n = 110) P value Deformation index (DI) 1.65 ± 0.2 1.43 ± 0.19 P value <0 . 0 0 1 Reprod. Sci. (2020) 27:364–374 369 show tendency of skewness to the right (skewness of 1.06), as seen in Fig. 6. Cellular Velocity Versus Cell Size Inside Microchannels Cell velocity inside the microchannel depends on cell size in addition to cellular stiffness. A stiffer cell will impose larger forces on channel walls when it deforms inside it resulting in higher friction and longer passage time. Same applies for larger cells that may be less stiff but can still apply high forces on channel walls because of the higher deformation it experiences to pass through the narrow microchannel. Consequently, cell size had to be included as another parameter when comparing cell velocity from patients and control women. Therefore, we presented data in the form of a heat map where cell size is plotted on the x-axis and cell velocity is plotted on the y-axis (Fig. 7). Each point in the heat map represents data from one cell. When many points coincide on top of each other, they are assigned a different color. As can be seen from Fig.7, larger cells from both patients and controls do take longer time to pass through the channel. Moreover, it is also clear that eutopic endometrial stro- mal cells from endometriosis patients exhibit higher velocities when passing through the microchannel compared with cells of the same size from control women which is a reflection of the lower stiffness of cells from endometriosis patients. The data s h o w ni nF i g .7 is the result of analyzing 8948 cells from 7 endometriosis patients and 6 control cases. The heat map of cells from each individual endometriosis patient and control woman is included insupplementary information.

In the present study, we developed a high throughput microfluidics platform to characterize the mechanical signa- ture of eutopic endometrial stromal cells of endometriosis pa- tients based on cellular deformability. Our results have shown that eutopic endometrial stromal cells of endometriosis pa- tients are less stiff, more deformable, and exhibit higher ve- locities in traversing narrow microchannels than their counter- parts from endometriosis-free women. To the best of our knowledge, this is the first study to investigate the mechanical stiffness of eutopic endometrial stromal cells of endometriosis patients. An important advan- tage of the current study is the high throughput nature of our microfluidics platform in which we tested large number of individual endometrial stromal cells from endometriosis Fig. 5 V elocity of eutopic endometrial stromal cells of endometriosis patients and control women inside microchannels. Error bars represent standard error of the mean (SEM). *P value < 0.001 Fig. 6 A histogram showing distribution of eutopic endometrial stromal cells from endometriosis patients and control women according to their velocity inside the microchannel system 370 Reprod. Sci. (2020) 27:364–374 patients and control women. To confirm the accuracy of our results, and as a check on the validity of the image processing software for measuring the velocity of cells inside microchannels [28], we measured the velocity of a group of cells manually and compared it with the velocity automatical- ly calculated by the software. Both values were identical. Prior research has confirmed the invasiveness of eutopic endometrial stromal cells [29]. When co-cultured with perito- neal explants, endometrial stromal cells breached the intact mesothelial cell layer in 24 h of co-culture [ 30]. Similarly, when plated with dispersed peritoneal mesothelial cells, endo- metrial stromal cells extended pseudopodia under the meso- thelial cell layer [ 31]. In addition, endometrial stromal cells were found to invade peritoneal mesothelial cells plated on matrigel-coated chambers. This effect was enhanced by activin A and associated with production of MMP-2 and MMP-9 [ 32]. Moreover, ectopic endometrial stromal cells were more invasive than their eutopic counterparts, which were more invasive than endometrial stromal cells of disease-free women. The difference in invasiveness is related to increased expression of ezrin, a member of ezrin/radixin/ meosin family of proteins which act as linkers between actin filaments and plasma membrane proteins [33]. Because cellu- lar motility and invasiveness are inversely proportional to their mechanical stiffness [34], it is not surprising that the motile and invasive eutopic endometrial stromal cells of endometri- osis patients are less mechanically stiff, as our results have shown. The vesico-elastic properties of cells are determined by characteristics of their cytoskeleton [ 35]. F-actin, vimentin intermediate filaments, and microtubules represent the main constituents of cellular cytoskeleton structures. F-actin fila- ments provide high degree of resistance to deformation. They polymerize to form tertiary structures known as actin stress fibers with the help of different actin-binding proteins to provide the cell with high mechanical integrity. The inter- mediate vimentin filaments allow moderate degree deforma- tion and tolerate high degrees of mechanical stress, even at levels at which F-actin are unable to keep their mechanical integrity. Microtubules do not have enough tensile stiffness to provide mechanical support to the cytoskeleton, but they work together with other elements to stabilize the cellular cytoskel- eton. The degree of mechanical stiffness and cellular deformability exhibited by a particular cell type depends on the concentration and molecular composition/regulation of the cytoskeleton proteins contained in these cells [36]. Key regulators of the cytoskeleton protein dynamics are members of the small Rho- GTPase family of proteins (Rho, Rac, and Cds42) [37]. ROCKII, which is a downstream effector of Rho, activates myosin light chain and regulates ezrin/radixin/ moesin family of actin remodeling proteins in the cell [ 38]. Eutopic endometrial stromal cells of endometriosis patients showed more enhanced migratory phenotype than their coun- terparts from control women as a result of higher activation of the Raf-1/Rho/ROCKII pathway [39]. In addition, Ectopic en- dometrial stromal cells containedsignificantly higher levels of Fig. 7 V elocity of eutopic endometrial stromal cells versus cell size (projected cell area). a Eutopic endometrial stromal cells from control women. Cellular velocity = 57.518 ± 0.585 mm/s (mean ± SEM). Cells count = 4541. b Eutopic endometrial stromal cells from endometriosis patients. Cellular velocity = 96.530 ± 0.710, mm/s (mean ± SEM). Cells count = 4407 Reprod. Sci. (2020) 27:364–374 371 phosphorylated ezrin/radixin/moesin cytoskeletal protiens than cells of eutopic endometrium, or control women [40]. Vinculin is another actin remodeling proteins that was shown to be dys- regulated in endometriosis [ 41]. In addition, ovarian steroid hormone treatment of eutopic endometrial stromal cells from endometriosis patients induced a promigratory phenotype char- acterized by cytoskeleton alteration including loss of stress fi- bers, progressive localization of actin toward the edge of the cell membrane, and simultaneous presence of numerous stress fiber arcs [42]. Focal adhesion kinase is an estrogen-regulated molecule residing at points of contact with extracellular matrix forming a signaling complex to mediate important cellular functions including cytoskeleton remodeling. In endometriosis, focal adhesion kinase was shown to be dysregulated in the eutopic endometrium, and its levels correlate with the disease stage and pain symptoms during the secretory phase of the menstrual cycle [43]. The research evidence for dysregulation of cytoskeleton elements and their regulatory pathways in eutopic endometrial cells of endometriosis patients can provide some explanation of our findings of altered mechanical proper- ties and reduced stiffness of these cells. Mechanical properties of living cells have emerged as a pos- sible biomarker for predicting health state of cells [ 44]. For example, many types of cancers were reported to have de- creased cellular stiffness. These include breast [45], lung [46], pancreatic [47], ovarian [ 48], and bladder cancers [ 49]. This change in mechanical properties between healthy and diseased cells leads to the emergence of mechanical properties as a bio- marker for diagnostics, eliminating the need for conventional biomarkers, thus reducing examination time and cost [35]. The difference in cellular deformability/mechanical stiffness be- tween eutopic endometrial stromal cells of endometriosis pa- tients and those of disease-free women expressed as the velocity of cells as they travel through microchannels, under the condi- tions described in our system, can establish the basis for a non- surgical test to differentiatebetween women with and without endometriosis. Compared with the current standard diagnostic modality of endometriosis, i.e., laparoscopy, our proposed mechanical biomarker of the disease is much less invasive, as it only requires an endometrial biopsy that can be taken as an office procedure. Considering costs, our technique requires one dis- posable microfluidic chip carrying 10 microchannels and cost- ing only $3 which is nothing compared with the expenses of laparoscopy which could amount to 4289 ± $3313 [ 50]. Although dissociation of the endometrial biopsy followed by cell culture for few days is required to obtain sufficient num- ber of cells to perform our test, still this is cheaper, more accessible, and more convenient than scheduling a laparosco- py. It has to be noted here that cost of capital equipment for both techniques (laparoscopic equipment for laparoscopy and an inverted microscope, high-speed camera, and a syringe pump for our microfluidic technique) is similar. Our study is not without limitations. As we were trying to prove the concept of a difference in the mechanical properties in endometrial cells between women with and without endo- metriosis, we used a complex platform involving cell culture equipment, inverted microscope, a microchip, high-speed camera, and computer vision software. In order to be able to confirm our results in a larger cohort of patients, our setup should be simplified. The envisioned platform will comprise a chip containing the required narrow microchannel integrated with on-chip pumping [51]. Eutopic endometrial stromal cell can be loaded on the inlet reservoir and pumped through the narrow microchannel. Built-in optics, CMOS sensor [52], and a field-programmable gate array (FPGA) module can be in- corporated and programmed to perform computer vision steps to find average cell velocity inside the microchannel. To re- place the cell culture step and increase the level of automation, the microfluidic platform can be further developed to perform tissue dissociation and debris filtration on chip [ 53]. To pre- vent clogging of microchannels, a cross-flow filter [54]c a nb e built at the microchannel entrance to hold large cells that may clog the microchannel. Cell sorting based on size prior to passage into the microchannel constriction is also possible. In conclusion, we characterized the mechanical signa- ture of eutopic endometrial st romal cells of endometriosis patients using a high throughput microfluidics platform implying velocity of cells inside microchannel constriction as surrogate for cellular stiffness. We found that eutopic endometrial stromal cells of endometriosis patients have increased deformation index and exhibit higher velocity inside our microchannel system (i.e., lower stiffness and higher deformability) compared with their counterparts of control women. These particular biomechanical features of eutopic endometrial stromal cells of endometriosis can lay the foundation for identifying a mechanical biomarker of the disease. Acknowledgments The authors would like to acknowledge Professor Felice Petraglia, University of Florence, Italy, and Dr. Felice Arcuri, Siena University, Italy, for providing the protocol for endometrial stromal cell isolation and culture. Funding Information The study was funded by a grant from Science and Technology Development Fund of Egypt (STDF) to E.O. (grant ID # 5525). Microchannels used in this study were fabricated at the clean room of the Faculty of Engineering which was established through a grant from the Science and Technology Development Fund of Egypt (STDF) to M.A. (grant ID # 4918). Compliance with Ethical Standards All participating women provided written informed consent. Institutional Review Board at Faculty of Medicine, Assiut University, approved the use of human endometrial tissue samples for this study. Conflict of Interest The authors declare that they have no conflict of interest. 372 Reprod. Sci. (2020) 27:364–374

1. Fassbender A, V odolazkaia A, Saunders P , Lebovic D, Waelkens E, De Moor B, et al. Biomarkers of endometriosis. Fertil Steril. 2013;99(4):1135–45. 2. Eskenazi B, Warner ML. Epidemiology of endometriosis. Obstet Gynecol Clin N Am. 1997;24(2):235–58. 3. Falcone T, Flyckt R. Clinical management of endometriosis. Obstet Gynecol. 2018;131(3):557–71. 4. Nnoaham KE, Hummelshoj L, Webster P , d ’Hooghe T, de Cicco NF, de Cicco NC, et al. Impact of endometriosis on quality of life and work productivity: a multicenter study across ten countries. Fertil Steril. 2011;96(2):366–373.e8. 5. Pearce CL, Templeman C, Rossing MA, Lee A, Near AM, Webb PM, et al. Association between endometriosis and risk of histolog- ical subtypes of ovarian cancer: a pooled analysis of case-control studies. Lancet Oncol. 2012;13(4):385–94. 6. Dunselman GA, V ermeulen N, Becker C, Calhaz-Jorge C, D’Hooghe T, De Bie B, et al. ESHRE guideline: management of women with endometriosis. Hum Reprod. 2014;29(3):400–12. 7. HarkkiSiren P , Kurki T. A nationwide analysis of laparoscopic com- plications. Obstet Gynecol. 1997;89(1):108–12. 8. Ballard K, Lowton K, Wright J. What ’s the delay? A qualitative study of women’s experiences of reaching a diagnosis of endome- triosis. Fertil Steril. 2006;86(5):1296–301. 9. Patel BG, Lenk EE, Lebovic DI, Shu Y , Y u J, Taylor RN. Pathogenesis of endometriosis: interaction between endocrine and inflammatory pathways. Best Pract Res Clin Obstet Gynaecol. 2018;50:50–60. 10. Sampson JA. Peritoneal endometriosis due to the menstrual dissem- ination of endometrial tissue into the peritoneal cavity. Am J Obstet Gynecol. 1927;14:422–69. 11. Giudice LC. Endometriosis. N Engl J Med. 2010;362(25):2389 –98. 12. Giudice LC, Kao LC. Enometriosis. Lancet. 2004;364(9447): 1789–99. 13. Matsuzaki S, Canis M, Pouly JL, Darcha C. Soft matrices inhibit cell proliferation and inactivate the fibrotic phenotype of deep en- dometriosis stromal cells in vitro. Hum Reprod. 2016;31(3):541 – 53. 14. Burney RO, Giudice LC. Pathogenesis and pathophysiology of endometriosis. Fertil Steril. 2012;98(3):511–9. 15. Guo SW . Recurrence of endometriosis and its control. Hum Reprod Update. 2009;15(4):441–61. 16. Davis AC, Goldberg JM. Extrapelvic endometriosis. Semin Reprod Med. 2017;35(1):98–101. 17. Anglesio MS, Papadopoulos N, A yhan A, Nazeran TM, Noë M, Horlings HM, et al. Caner- associated mutations in endometriosis without cancer. N Engl J Med. 2017;376(19):1835–48. 18. Lekka M, Gil D, Pogoda K, Duli ńska-Litewka J, Jach R, Gostek J, et al. Cancer cell detection in cancer tissue sections using AFM. Arch Biochem Biophys. 2012;518(2):151–619. 19. Suresh S. Nanomedicine: elastic clues in cancer detection. Nat Nanotechnol. 2007;2(12):748–9. 20. Ciasca G, Papi M, Minelli E, Palmieri V , De Spirito M. Changes in cellular mechanical properties during onset or progression of colo- rectal cancer. World J Gastroenterol. 2016;22(32):7203–14. 21. Hou HW, Li QS, Lee GYH, Kumar AP , Ong CN, Lim CT. Deformability study of breast cancer cells using microfluidics. Biomed Microdevices. 2009;11(3):557–64. 22. Herbig M, Krater M, Plak K, Muller P , Guck J, Otto O. Real-time deformability cytometry: label-free functional characterization of cells. In: Hawley TS, Hawley RG, editors. Flow cytometry proto- cols, vol. 2018. 4th ed; 1678. p. 347 –69. 23. Gossett DR, Tse HT, Lee SA, Ying Y , Lindgren AG, Y ang OO, et al. Hydrodynamic stretching of single cells for large population mechanical phenotyping. Proc Natl Acad Sci U S A. 2012;109(20): 7630–5. 24. Guck J, Ananthakrishnan R, Mahmood H, Moon TJ, Cunningham CC, Käs J. The optical stretcher: a novel laser tool to micromanipulate cells. Biophys J. 2001;81(2):767–84. 25. Carrarelli P , Funghi L, Bruni S, Luisi S, Arcuri F, Petraglia F. Naproxen sodium decreases prostaglandins secretion from cultured human endometrial stromal cells modulating metabolizing enzymes mRNA expression. Gynecol Endocrinol. 2016;32(4):319–22. 26. Duffy DC, McDonald JC, Schueller OJA, Whitesides GM. Rapid prototyping of microfluidic systems in poly(dimethylsiloxane). Anal Chem. 1998;70(23):4974–84. 27. Haubert K, Drier T, Beebe D. PDMS bonding by means of a por- table, low-cost corona system. Lab Chip. 2006;6(12):1548–9. 28. Esmaeel AM, ElMelegy T, Abdelgawad M. Multi-purpose machine vision platform for different microfluidics applications. Biomed Microdevices. 2019;21:1–13. https://doi.org/10.1007/s10544-019- 0401-1. 29. Weimar CHE, Macklon NS, Uiterweer EDP , Brosens JJ, Gellersen B. The motile and invasive capacity of human endometrial stromal cells: implications for normal and impaired reproductive function. Hum Reprod Update. 2013;19(5):542–57. 30. Witz CA, Dechaud H, Montoya-Rodriguez IA, Thomas MR, Nair AS, Centonze VE, et al. An in vitro model to study the pathogenesis of the early endometriosis lesion. Ann N Y Acad Sci. 2002;955: 296–307 Discussion 340–292, 396–406. 31. Witz CA, Cho S, Centonze VE, Montoya-Rodriguez IA, Schenken RS. Time series analysis of transmesothelial invasion by endome- trial stromal and epithelial cells using three-dimensional confocal microscopy. Fertil Steril. 2003;79(Suppl. 1):770–8. 32. Ferreira MC, Witz CA, Hammes LS, Kirma N, Petraglia F, Schenken RS, et al. Activin A increases invasiveness of endome- trial cells in an in vitro model of human peritoneum. Mol Hum Reprod. 2008;14:301–7. 33. Ornek T, Fadiel A, Tan O, Naftolin F, Arici A. Regulation and activation of ezrin protein in endometriosis. Hum Reprod. 2008;23(9):2104–12. 34. Wenwei X, Roman M, Byungkyu K, Lijuan W, John M, Todd S. Cell stiffness is a biomarker of the metastatic potential of ovarian cancer cells. PLoS One. 2012;7:e46609. 35. Quan FS, Kim KS. Medical applications of the intrinsic mechanical properties of single cells. Acta Biochim Biophys Sin (Shanghai). 2016;48(10):865. 36. Suresh S. Biomechanics, and biophysics of cancer cells. Acta Biomater. 2007;3(4):413–38. 37. Lawson CD, Ridley AJ. Rho GTPase signaling complexes in cell migration and invasion. J Cell Biol. 2018;217(2):447–57. 38. Matsui T, Maeda M, Doi Y , et al. Rho-kinase phosphorylates COOH-terminal threonines of ezrin/radixin/moesin (ERM) pro- teins and regulates their head-to-tail association. JCell Biol. 1998;140(3):647–57. 39. Y otova IY , Quan P , Leditznig N, Beer U, Wenzl R, Tschugguel W. Abnormal activation of Ras/Raf/MAPK and RhoA/ROCKII signal- ling pathways in eutopic endometrial stromal cells of patients with endometriosis. Hum Reprod. 2011;26(4):885–97. 40. Y otova I, Quan P , Gaba A, Leditznig N, Pateisky P , Kurz C, et al. Raf-1 levels determine the migration rate of primary endometrial stromal cells of patients with endometriosis. J Cell Mol Med. 2012;16(9):2127–39. 4 1 . W uZ Y ,Y a n gX M ,C h e n gM J ,Z h a n gR ,Y eJ ,Y iH ,e ta l . Dysregulated cell mechanical properties of endometrial stromal cells from endometriosis patients. Int J Clin Exp Pathol. 2014;7(2):648–55. 42. Gentilini D, Vigano P , Somigliana E, Vicentini LM, Vignali M, Busacca M, et al. Endometrial stromal cells from women with Reprod. Sci. (2020) 27:364–374 373 endometriosis reveal peculiar migratory behavior in response to ovarian steroids. Fertil Steril. 2010;93(3):706–15. 43. Mu L, Zheng W , Wang L, Chen X-J, Zhang X, Y ang JH. Alteration of focal adhesion kinase expression in eutopic endometrium of women with endometriosis. Fertil Steril. 2008;89(3):529–37. 44. Cross SE, Jin YS, Rao J, Gimzewski JK. Nanomechanical analysis of cells from cancer patients. Nat Nanotechnol. 2007;2(12):780–3. 45. Li QS, Lee GY , Ong CN, Lim CT. AFM indentation study of breast cancer cells. Biochem Biophys Res Commun. 2008;374:609–13. 46. Panzetta V , Musella I, Rapa I, V olante M, Netti PA, Fusco S. Mechanical phenotyping of cells and extracellular matrix as grade and stage markers of lung tumor tissues. Acta Biomater. 2017;57: 334. 47. Nguyen A V , Nyberg KD, Scott MB, Welsh AM, Nguyen AH, Wu N, et al. Stiffness of pancreatic cancer cells is associated with in- creased invasive potential. Integr Biol (Camb). 2016;8(12):1232 – 45. 48. Xu W, Mezencev R, Kim B, Wang L, McDonald J, Sulchek T. Cell stiffness is a biomarker of the metastatic potential of ovarian cancer cells. PLoS One. 2012;7:e46609. 49. Lekka M, Laidler P , Gil D, Lekki J, Stachura Z, Hrynkiewicz AZ. Elasticity of normal and cancerous human bladder cells studied by scanning force microscopy. Eur Biophys J. 1999;28(4):312–6. 50. Fuldeore M, Chwalisz K, Marx S, Wu N, Boulanger L, Ma L, et al. Surgical procedures and their cost estimates among women with newly diagnosedendometriosis: a US database study. J Med Econ. 2011;14(1):115–23. 51. Y okokawa R, Saika T, Nakayama T, Fujita H, Konishi S. On-chip syringe pumps for picoliter-scale liquid manipulation. Lab Chip. 2006;6(8):1062–6. 52. Wang KI, Salcic Z, Y eh J, Akagi J, Zhu F, Hall CJ, et al. Biosensors and bioelectronics toward embedded laboratory automation for smart lab-on-a-chip embryo arrays. Biosens Bioelectron. 2013;48: 188–96. 5 3 . A l - M o f t yS ,E l b a d r iN ,A l t a y y e bA ,O m a rO ,E l s a y e dM , Shamseldine A, et al. One stop lab-on-chip platform for tissue processing and cell sample preparation. 20 th International Conference on Miniaturized Systems for Chemistry and Life Sciences, MicroTAS 2016, Dublin, Ireland. 54. Ji HM, Samper V , Chen Y , Heng CK, Lim TM, Y obas L. Silicon- based microfilters f or whole blood cell separation. Biomed Microdevices. 2008;10(2):251–7. Publisher’sN o t eSpringer Nature remains neutral with regard to jurisdic- tional claims in published maps and institutional affiliations. 374 Reprod. Sci. (2020) 27:364–374