Advancements in Microfluidic Systems for the Study of Female Reproductive Biology

Abstract

The female reproductive tract is a highly complex physiological system that consists of the ovaries, fallopian tubes, uterus, cervix, and vagina. An enhanced understanding of the molecular, cellular, and genetic mechanisms of the tract will allow for the development of more effective assisted reproductive technologies, therapeutics, and screening strategies for female specific disorders. Traditional 2-dimensional and 3-dimensional static culture systems may not always reflect the cellular and physical contexts or physicochemical microenvironment necessary to understand the dynamic exchange that is crucial for the functioning of the reproductive system. Microfluidic systems present a unique opportunity to study the female reproductive tract, as these systems recapitulate the multicellular architecture, contacts between different tissues, and microenvironmental cues that largely influence cell structure, function, behavior, and growth. This review discusses examples, challenges, and benefits of using microfluidic systems to model ovaries, fallopian tubes, endometrium, and placenta. Additionally, this review also briefly discusses the use of these systems in studying the effects of endocrine disrupting chemicals and diseases such as ovarian cancer, preeclampsia, and polycystic ovarian syndrome.

The female reproductive tract is a highly complex physiological system that enables pregnancy and sustains the fetus throughout the pregnancy. The reproductive tract consists of the ovaries, fallopian tubes, uterus, cervix, and vagina. These organs have a strong interdependence and communicate via paracrine and endocrine signals at a spatial–temporal level, which is necessary to achieve the reproductive system’s functions (1-3). Primary functions of the system include maturing the ova (or the egg cells), providing an environment to allow fertilization, and supporting the developing embryo. Additionally, the system also influences biological function in the endocrine, immune, cardiovascular, and skeletal systems through the production of steroid hormones (4-8).

The health and regulation of the female reproductive tract is vulnerable to genetic susceptibilities, environmental factors, and off-target effects from various drugs which can result in reproductive health concerns such as polycystic ovarian syndrome, ovarian cancer, and endometriosis (9-13). The Centers for Disease Control (CDC) reported that between 2015 and 2017, about 13.1% of women from 15 to 49 years of age had impaired fecundity, and 8.8% of married women in the same age group were infertile (14). During the same period, 12.7% of women opted for fertility treatments. Predictably, intense research has gone into assisted reproductive technologies (ART), and more than 5 million children have been conceived using such technologies (15). Traditionally, this research has largely focused on increasing the likelihood of continuing pregnancy, but recently has expanded into aspects that can influence the long-term health of the fetus, such as embryonic developmental competence, gene expression patterns, and epigenetic reprogramming (16-21). The key to development of more effective ART, therapeutics, and screening strategies for female specific disorders is better understanding of the molecular, cellular, and genetic mechanisms of the female reproductive tract. However, this has been challenging due to the lack of ideal model systems.

Studying human female reproductive biology can result in some ethical concerns, especially regarding pregnant subjects, designer embryos, and contraception. Additionally, the dynamic hormonal secretion patterns and anatomical inaccessibility of the system has made it difficult to systematically study causes of infertility and the early-stage events of disorders in the female reproductive tract. Thus, 2-dimensional static culture systems have traditionally been used to study different organs and cell populations of the tract where cells are grown on tissue culture treated plastic flasks (22, 23). These allow for convenient and effective cell-based assays that help elucidate mechanisms of diseases and drugs (24). Three-dimensional (3D) culture systems, such as organoids, have also been essential tools in studying reproductive biology (25-28). These systems use synthetic polymers such as hydrogels or natural extracellular matrix molecules and can better replicate certain characteristics of the in vivo tissue architecture and signaling (29, 30). However, these systems may not always reflect the cellular and physical contexts necessary to understand the dynamic hormonal exchange that is crucial for the functioning of the human female reproductive system (31-35).

Microfluidic culture systems are a technology that has the potential to provide a more physiologically relevant model that overcomes limitations of traditional static cultures. Known by various names such as organ-on-a-chip or lab-on-a-chip, these devices can recapitulate the multicellular architecture and contacts between different tissues and thus, can enhance our understanding of in vivo mechanisms (36-38). Some of these devices offer precise control over fluid flow in a series of micrometer-sized channels, thus replicating the in vivo continuous or pulsatile perfusion (39-43). Cells and tissues in culture are exposed to mechanical cues, including fluid shear stress, tension, and compression, as well as chemical cues, such as spatiotemporally controlled gradients of secreted factors and steroid hormones (44-47). Microenvironmental cues largely influence cell structure, function, behavior, and growth (43, 48). Hence, the cellular response to externally introduced stimuli such as drugs and toxins can be more accurately represented in a microfluidic culture system (49-52).

Different organs such as the liver, kidney, lungs, and intestines have been modeled by adapting commercially available platforms such as the Emulate or Nortis chips (53-62). Instead, the reproductive field has evolved to develop unique systems that are tailored to support specific biological properties of different reproductive tissues such as elimination of polydimethylsiloxane (PDMS) to ensure availability of hydrophobic steroids, large tissue compartments for 3D follicle expansion, and recirculation of perfusion to ensure endocrine signaling (Fig. 1). Hence, this review is organized around female reproductive tract organs, including the ovaries, fallopian tubes, endometrium, and placenta, and discusses the setup of specific devices that were used to model the organ. The review also briefly discusses the use of these systems in studying effects of endocrine disrupting chemicals (EDCs) and diseases of the reproductive tract such as ovarian cancer and preeclampsia (Fig. 2). Application of microfluidic systems for the study of male reproductive system is beyond the scope of this review but has been extensively discussed in other review articles (63-68).

Figure 1.

Technical features of microfluidic systems that enable their application to the reproductive tract. Polydimethylsiloxane (PDMS) and polysulfone (PSF) are popular materials used in the fabrication of microfluidic devices. PDMS has a high tendency to absorb hydrophobic compounds, such as steroid hormones and drugs, which would impact reproductive function. Organs with a lumen benefit from an air-liquid interface to remain viable in culture but research on placental drug transport instead requires a closed fluid environment—both of these environments can be created in a microfluidic system. Unlike static culture systems, microfluidic systems can incorporate dynamic fluid flow and replicate physiochemical cues. Devices can be configured to recirculate media and reproduce endocrine loops or model circadian rhythm.

Figure 2.

Applications of microfluidics systems in studying the reproductive tract. Culturing tissues or cell populations from the female reproductive tract using microfluidic systems allows for the study of various biological processes such as ovulation, folliculogenesis, hormone biosynthesis, decidualization, postimplantation developmental events, and beating of cilia present on the fallopian tube epithelium (FTE). Microfluidic devices can assist in the study of enhanced assisted reproductive technologies (ART). Moreover, the systems can also help elucidate mechanisms, identify biomarkers, and develop therapeutic strategies for disorders of the female reproductive tract, such as high-grade serous ovarian cancer (HGSC). Microfluidic systems also provide a system for drug efficacy and toxicity, especially when used for co-culturing the tract’s organs with organs from other organ systems such as the liver and intestine.

Microfluidic Modeling of the Ovary

The ovary is the female gonad and is located near the lateral walls of the pelvic cavity. In addition to being the site of gametogenesis, the ovaries secrete steroid hormones, such as estrogen and progesterone, both of which modify the uterus and are crucial to normal reproductive development and fertility (69). The follicular phase, or the first half of the 28-day menstrual cycle, involves stimulation of ovarian follicles in response to follicle-stimulating hormone. Midway through the cycle, a spike in luteinizing hormone released by the pituitary gland triggers ovulation, an event where the preovulatory follicle ruptures and releases the oocyte. After ovulation, in the luteal phase, granulosa cells that surround the ruptured follicle undergo a process of luteinization which involves differentiation, cell enlargement, and increased progesterone production. A woman has a fixed number of oocytes at birth, unlike the sperm in males, and as a woman ages, the number of oocytes declines, resulting in menopause (70). Hence, microfluidic technology to rescue and mature the ovulated oocyte (known as the egg) might provide a route to preserving fertility, which would be particularly beneficial when treatments (such as cancer therapies) that are toxic to the arrested ovarian follicle reserve risk causing permanent infertility (17, 71, 72). Additionally, a microfluidic device that can model ovulation provides a system to study signaling pathways associated with infertility resulting from disease or compounds such as specific chemotherapy regimens (73-76).

Xiao et al (2017) designed the single and dual unit microfluidic systems, Solo-MFP and Duet-MFP, using an intermediate thermoplastic track-etched membrane (77). These devices were able to sustain murine ovarian explants, which contained follicles from all developmental stages, for the length of the human menstrual cycle. The device used pneumatic actuation technology to incorporate dynamic flow of media in their culture system and support murine follicle growth. By controlling exogenous follicle-stimulating hormone and luteinizing hormone, the group replicated maturation, ovulation, and granulosa cell luteinization. Moreover, they found that the steroid hormone profile produced by the cultured follicles mimicked that of an ovary in a human menstrual cycle. The inability of static culture to maintain proper nutrient and oxygen diffusion had previously made this challenging (78, 79).

Despite the huge potential for fertility preservation, it has been challenging to culture in vitro oocytes from immature ovarian follicles of large mammalian species because of their complex metabolic needs and long-term culture requirements (80-83). While it has previously been observed that both murine and human primary follicles behave similarly in static culture, it is essential to translate microfluidic development to larger mammalian species because of differences between mice and large mammals such as humans (84-92). For example, the menstrual cycle duration as well as follicle development period is longer in larger animals compared to mice in addition to the varied size of follicles between species (80, 93, 94). Hence, in vitro culturing of large species follicles requires a more complex and dynamic system than most static cultures can provide.

To create a dynamic system that is capable of improving survival and growth of follicles from large mammalian species in vitro, Nagashima et al (2017) built a microfluidic chip using polymethyl methacrylate (95). The group cultured both individual follicles and ovarian cortex tissues isolated from cats and dogs. The study evaluated the influence of flow on isolated ovarian follicles and effects of culture conditions on the follicular responsiveness to flow. This system enabled the study of mechanistic aspects of folliculogenesis, in particular primordial follicle activation and the interaction between microenvironment rigidity and follicle development. The understanding from large mammalian animal models is more relevant to human fertility preservation techniques and help in the development of enhanced ART.

Another approach that utilizes microfluidic technology for the purpose of fertility preservation has been described by He (2017), who used a nonplanar PDMS device to study microfluidic encapsulation of rodent ovarian follicles in core-shell hydrogel microcapsules (96). Their study showed that this technique could create biomimetic ovarian microtissue that recapitulated the mechanical heterogeneity in the extracellular matrix, which is essential to regulate follicle development. Thus, this technique could be used to facilitate follicle culture as a part of ART aiming to preserve the fertility of women.

Many current ARTs have operator-to-operator variation and lack standardization due to their reliance on skilled embryologists (97). For instance, ART clinics use enzymatic action of hyaluronidase to denude oocytes from the cumulus cell-corona cell mass that surround the oocyte (98). This involves the labor of mechanical pipetting and risks damaging the oocyte (99). A microfluidic device capable of denuding oocytes from the surrounding cumulus-corona cell mass in a continuous fluid flow in order to overcome these barriers was demonstrated by Weng et al (2018) (100). The group used standard PDMS soft lithography techniques to fabricate a microfluidic device that employed repeating constriction-expansion units of optimized geometries and surface features to completely denudate mouse oocytes. Additionally, their device did not use excessive mechanical stress, and hence it maintained developmental potential of the oocyte. Such devices have invaluable translational potential and could result in an automated and standardized processing of human samples in a clinical setting.

Microfluidic Modeling of the Fallopian Tube

The fallopian tubes in humans, called oviducts in other mammals, are tubes that originate near the ovaries and ultimately transition into the uterus. The epithelial lining of the fallopian tube consists of ciliated and secretory cells that make close contact with the gametes and early embryo (101-103). Additionally, the intricately folded tubular morphology, muscle contractions, and ciliary beating of these tubes largely influence fluid flux and movement of sperm and oocyte, enabling fertilization (104-107). In addition to providing a passageway for gametes and early-stage embryos to travel, the fallopian tubes also provide an optimal microenvironment for the embryo’s preimplantation development and epigenetic reprogramming (108-110).

Anatomical inaccessibility and difficulty in replicating in vivo gradients create challenges to studying the signaling pathways and processes of the fallopian tube that contribute to fertilization and developmental competence of the early embryo. Currently, in vitro oviductal or fallopian tubes models have used monolayer cultures of oviduct epithelial cells (111-113). However, fallopian tube cultures on plastic can undergo a rapid transformation of the differentiated, cuboidal–columnar oviduct epithelial cells into flattened cells, thus losing some of their cilia and secretory ability, compromising aspects of the in vivo tubular physiology (114, 115). Hence, a microfluidic model that maintains cell differentiation is beneficial in investigating the factors, such as hormones, growth factors, and shear stress that influence fertilization and preimplantation development.

Ferraz et al (2018) designed PDMS microfluidic device that maintained the in vivo–like morphology of bovine oviduct epithelial cells including a cuboidal to columnar pseudostratified epithelium with ciliated and secretory cells and function (114). In addition to supporting cell growth and differentiation similarly to that observed in vivo, their oviduct-on-a-chip model also produced bovine zygotes with a transcriptome and global methylation pattern similar to that of zygotes in vivo. This device provides a physiological microenvironment ideal for advancing our knowledge of gamete interaction, fertilization, embryogenesis, and zygote genetic reprogramming. Additionally, the device possesses a live imaging feature that can be useful in assessing the effects of drugs on gametes or embryos in real time.

In addition to its use in studying ovaries, the Solo-MFP designed by Xiao et al (2017) was also employed by Jackson-Bey et al (2020) to study the effects of high testosterone on function of human fallopian tube epithelium (116). Using the microfluidic device, the group cultured human fallopian tube epithelial tissue obtained from women undergoing salpingectomy. In addition to maintaining viability, the tissue retained cilia beating activity for 14 days. The study found that increased exposure to testosterone modifies expression of multiple genes that regulate cilia beating. These findings demonstrate the application of microfluidic systems in elucidating the mechanisms of subfertility in women living with hyperandrogenic disorders.

Microfluidic Modeling of the Uterus and Endometrium

The uterus is a muscular organ consisting of the corpus and cervix. The endometrium is 1 of 3 layers that lines the uterine cavity and its thickness varies throughout the menstrual cycle (2, 69). After ovulation, during the luteal phase, if the egg is fertilized by sperm, the resulting embryo implants into the endometrium. The uterus then provides mechanical protection and nutritional support to the embryo until delivery. However, if conception does not occur, the endometrial tissue of the uterine lining is discharged during menstruation. After menstruation, the endometrial tissue repairs, grows, and differentiates to prepare for the next cycle (69, 117).

Depending on the cyclic stimulation of ovarian sex steroids, endometrial stromal fibroblasts and vascular endothelial cells undergo morphological and biochemical changes to support embryo implantation and vascular function regulation (118). Various processes are involved in preparing the endometrial tissue for reproduction, but the details of these processes remain unclear (119). Decidualization of endometrial stoma is one such process that is critical for the successful establishment and maintenance of pregnancy to term (120, 121). This transformation involves differentiation of stromal fibroblasts in response to sex steroids and is marked by a transition to an epithelial-like cuboidal cell shape and secretion of pro-gestational proteins (122, 123). Despite the importance of decidualization in establishing embryo implantation, it has been difficult to investigate the triggering mechanism and the early stages of this process in a static culture due to the lack of vascular perfusion (124, 125). Hence, a dynamic microfluidic model of the endothelium may lead to a better understanding of the interaction between decidualization, pharmaceutical agents, or environmental toxicants, and fertility.

Gnecco et al (2019) designed a model of the endometrial perivascular stroma by integrating a porous membrane in a dual chamber microfluidic device in PDMS (126). The porous membrane enabled diffusion of small molecules as well as high-resolution bright field and fluorescent imaging. This model sustained an in vitro co-culture of human endometrial stromal and endothelial cells for 4 weeks. While stimulating the hormonal changes associated with the human menstrual cycle, the group observed successful differentiation of stromal cells into functional decidual cells. Additionally, the group controlled the microfluidic properties of the device to subject the cells to experimental laminar perfusion that mimics the hemodynamic forces derived from endometrial blood flow. They found that hemodynamic forces induced secretion of specific endothelial cell-derived prostanoids that enhanced endometrial perivascular decidualization. This device has further potential to identify bidirectional crosstalk between cells and hemodynamics that may cause endometriosis, pregnancy failure, and infertility.

In addition to the uterine vascularization, the quality of the embryo is also influenced by postimplantation processes (127). The initial postimplantation morphological development in human embryo is marked by the apical–basal polarization and lumenogenesis of the epiblast that results in the pro-amniotic cavity (128, 129). However, there are barriers to studying these processes. Maintaining human embryos in vitro beyond the blastocyst stage remains challenging due to suboptimal protocols (130, 131). Moreover, bioethical guidelines restrict in vitro culture of human embryos to either 14 days postfertilization or to the onset of primitive streak development (132). Microfluidic systems can help in study of postimplantation developmental processes as shown by a human pluripotent stem cells (hPSCs)-based device developed by Zheng at al (2019) (133). Geltrex was preloaded in the device and the material contracted during gelation, forming concave gel pockets that acted as a 3D environment for cells loaded in the device. The model successfully recapitulated early postimplantation developmental events, such as lumenogenesis of the epiblast, differentiation of primordial germ cells, and formation of pro-amniotic cavity and embryonic sac. The control and scalability offered by their microfluidic model makes it an apt tool for investigating the influence of drugs and toxins on pregnancy.

When studying drug safety and efficacy, it is important to recognize that the absorption, distribution, metabolism, excretion, and toxicity of a drug is dictated by multiple cell types that are linked by intricate mechanisms (134). Edington et al 2018 fabricated a microfluidic platform with 10 organs: liver, lung, gut, endometrium, brain, heart, pancreas, kidney, skin, and skeletal muscle (135). The platform comprises 3 layers that involve a pneumatic bottom layer, a fluidic top layer, and an intermediate membrane. The top layer was made from a polysulfone plastic and the intermediate membrane was made of polyurethane. The platform sustained the phenotypic functionality of all 10 modules for 4 weeks and the circulating media flow allowed them to interact and exchange endogenously produced molecules. Because one of the organs included was the endometrium, the platform can be employed for studying pharmacokinetics during drug discovery as well as for disease modeling.

Microfluidic Modeling of Placenta

The placenta is a highly specialized organ in the human body and is composed of trophoblasts, connective tissue, basal lamina, and the fetal endothelium (136). The placenta is critical to a successful pregnancy as it acts as a barrier between the fetal circulation and maternal blood, and thus, it tightly regulates the exchange of endogenous and exogenous substances between the mother and the fetus. The lack of understanding about this key organ’s inner workings is a barrier to elucidating the effects of maternally administered drugs on the developing fetus. Additionally, a range of complications of pregnancy, such as preeclampsia and intrauterine growth restriction, have been linked to abnormalities in placenta structure and function (137, 138).

Traditionally, in vitro and ex vivo models that mimic physiological drug transport across the maternal-fetal interface in the human placenta have been inadequate (139). In vitro models used have been based on static culture of trophoblast monolayers in Transwell inserts and ex vivo models used extracorporeal perfusion of cannulated whole human placenta (140-143). While some toxicology studies have used animal models of pregnancy, there are considerable interspecies differences in placenta, and the 1960s tragedy of thalidomide raised questions about the predictive value of animal data (144). To better assess the safety of medications in pregnancy, one option would be to screen drug compounds for their ability to cross the placenta using a microfluidic model that captures the multilayered architecture, intercellular junctions, and hemodynamic environment of the placental barrier.

Blundell et al (2018) created a placenta-on-a-chip model by setting up a co-culture of human trophoblasts and human placental villous endothelial cells that grew in apposition on a fibronectin-coated semipermeable membrane in a compartmentalized microfluidic device (145). By precisely manipulating various key control parameters, the group was able to account for biochemical and physical factors that influence the dynamics of placental transfer such as utero-placental and umbilical blood flow, barrier thickness, and concentration gradients (146). The continuous perfusion and dynamic culture conditions of their model maintained cell viability and reproduced native morphological characteristics of the placenta such as generation of microvilli. To portray the applicability of their device to investigate placental drug transport, the group provided proof-of-principle for the role of breast cancer resistance protein (BCRP; an efflux transporter highly expressed on trophoblast cells) in mediating transport of glyburide, an oral drug used to treat diabetes in pregnancy (147). As the device is able to replicate pathological features of different placental disorders, it has the potential to serve as an in vitro disease model that can be used to develop new therapeutic strategies.

In addition to drugs, a microfluidic placenta model can also be used to analyze the interaction and transport of environmental contaminants and food additives through the placenta. A good example is environmental nanomaterials, which are prevalent due to the emerging nanotechnology industry (148, 149). Pregnant women are likely to encounter these particles through air or water. Yin et al (2019) investigated placental responses to environmental nanomaterials by co-culturing a human choriocarcinoma cell line and human umbilical vein endothelial cells on a PDMS chip (150). Additionally, through the addition of extracellular matrix using Matrigel, the group replicated in vivo microenvironment on the chip. Once the morphology and functionality of the model was confirmed, they exposed the model to titanium dioxide nanoparticles and found increased permeability due to ruptures in the placental barrier and impaired maternal immune cell behavior.

The placenta barrier facilitates gas exchange to the fetus, ensuring that the fetus is supplied adequate levels of oxygen and that carbon dioxide is removed (151). Hence, microfluidic models of placenta have clinical applicability in neonatal respiratory care of infants born prematurely that often have structurally and functionally immature lungs. Currently, these infants are given mechanical ventilation or extracorporeal membrane oxygenation. However, these can cause long-term complications or even death (152, 153). The premature transition from liquid to gas ventilation puts these infants at risk of breathing disorders such as bronchopulmonary dysplasia (154). Hence, an approach where extracorporeal gas exchange is maintained by the placenta is preferred.

Partridge et al (2017) created an extracorporeal device that was able to support an extremely premature fetal lamb for 4 weeks without apparent physiologic derangement or organ failure (155). They overcame the obstacle of circulatory failure due to preload or afterload imbalance imposed on the fetal heart by oxygenator resistance and pump-supported circuits by using a pumpless arteriovenous circuit along with a closed fluid environment that continuously exchanged fluid. This fluid exchange ensured nutritional and oxygen supply to the fetus, mimicking amniotic fluid. As such, this device may eventually not only be used to support premature infants but also to directly administer therapeutic agents to the isolated fetus.

Microfluidic Modeling of Diseases and Toxic Exposures

In addition to understanding reproductive processes and enhancing ART, microfluidic models of the female reproductive tract also have an application in studying exposures to EDCs as well as diseases of the female reproductive tract such as ovarian cancer and preeclampsia.

Ovarian Cancer

With a 10-year survival rate of <30%, ovarian cancer is one of the 5 leading causes of cancer death in females, likely due to insufficient early detection methods (156-158). The most common histotype of ovarian cancer is high-grade serous carcinomas (HGSC) and it is derived from the epithelial cells of the fallopian tube (156, 159, 160). One potential diagnostic strategy that incorporates the use of microfluidics is to utilize extracellular vesicles derived from biofluids for early and noninvasive detection of the disease.

A microfluidic-based device from Dorayappan et al (2019) isolated exosomes from culture media and cancer patient samples in a cost-effective and highly specific manner (161). Mass spectroscopy of these samples allowed them to identity exosome cargo proteins that are differentially expressed in HGSC cell lines compared with normal cells. These protein biomarkers may serve as part of an early detection strategy for cancer using biofluids. In fact, Zhang et al (2019) demonstrated using their microfluidic system termed MINDS (multiscale integration by designed self-assembly) that elevated folate receptor alpha (FRα) protein in ovarian cancer extracellular vesicles can be used as a biomarker for liquid biopsy-based cancer diagnosis (162). Their system integrated micro-patterning and 3D nanostructured herringbone mixer for molecular recognition and flow manipulation to detect ovarian cancer using as little as 2 microliters of plasma.

Ferraz et al (2020) created an oviduct-on-a-chip model using dog oviduct tissue discarded during routine ovary-hysterectomy (163). The model was responsive to hormonal stimulation patterns associated with the menstrual cycle. Moreover, using CRISPR-Cas9 to knock out the oncogene p53, the group was able to recapitulate serous tubal intraepithelial carcinoma, a premetastatic form of HGSC, as well as events associated with human serous tubal intraepithelial carcinoma, such as decreased ciliation, loss of cell polarization, and increased cell proliferation. In vitro models like these could help study early-stage events and identify novel biomarkers of ovarian cancer.

Preeclampsia

Preeclampsia is a hypertensive pregnancy complication that affects approximately 3.4% of pregnancies in the United States (164, 165). The disease is characterized by high maternal blood pressure that decreases blood supply, reducing fetal access to oxygen and nutrients. This condition increases the risk for organ failure, stroke, preterm birth, and stillbirth. Women who opt for ART such as egg donation, donor insemination, or in vitro fertilization are at a higher risk for preeclampsia (166-168). The disease has been associated with higher levels of circulating and placental soluble fms-like tyrosine kinase-1 (sFlt-1), an anti-angiogenic protein that disrupts angiogenic balance by antagonizing the proangiogenic factor, vascular endothelial growth factor (VEGF) (169). Use of microfluidic systems in preeclampsia research can open new avenues for treatment as shown by Trapiella-Alfonso et al (2019) (170). The group suggested restoring the angiogenic balance in patients by using VEGF-coated magnetic beads to capture sFlt-1 and release phosphatidylinositol-glycan biosynthesis class F protein (PIGF), a proangiogenic factor. The group utilized a microfluidic device to mimic a small-scale extracorporeal circulation system to assess the validity of their concept and they found that their approach resulted in a 40% reduction of sFlt-1 and up to 2-fold increase of free PIGF. Thus, microfluidic systems can be used to validate proof-of-concepts of new therapeutic approaches for preeclampsia.

Endocrine Disrupting Chemicals

EDCs are a class of chemicals that can mimic, block, or interfere with hormones in the body’s endocrine system due to similarities in their chemical structures (171). These are present in the environment as plastics (bisphenols), plasticizers (phthalates), fungicides (vinclozolin), and pharmaceutical agents (diethylstilbestrol) (172). Recently, it has been found that these chemicals, even in small amounts, can exert toxic effects on the female reproductive system and adversely affect a woman’s reproductive outcomes (173). Specifically, bisphenols have been suggested to hinder with uterine development and reproductive function by interacting with estrogen receptors and altering expression of estrogen-regulated genes (174-178). EDCs have also been reported in higher levels in patients with polycystic ovarian syndrome (179, 180).

The tunability of microfluidic systems offers the ability to characterize the effects of EDC exposure. The devices described in this review can be manipulated to examine the effects of EDC on specific organs such as ovaries, fallopian tubes, endometrium, and placenta as well as on physiological processes of ovulation, fertilization, implantation, embryogenesis, and placental transport. Given the complexity of effects resulting from EDC exposure, coupled with the high level of interaction between the organs of the reproductive system, a microfluidic device that cultures an interconnected network of multiple organs would be more informative about the collective response to EDCs.

The device termed “EVATAR” is one such platform that is capable of supporting 5 interconnected organs and uses embedded electromagnetic actuation (77). Xiao et al (2017) employed this device with co-cultured human liver spheroids, mouse ovarian explants, human fallopian tube epithelium, human endometrium, and human cervix tissues. The group was successfully able to maintain tissue viability and hormonal responsiveness for 28 days. By introducing liver on the platform, the potential toxicity of EDC metabolism products can also be investigated.

Future Directions and Challenges

While microfluidic systems have made many advancements that were previously challenging with static cultures, there are still hurdles to overcome. For instance, further development is needed to improve the material used, fabrication processes, and other practical aspects of the technology.

The fabrication of microfluidic devices often requires access to specialized cleanroom facilities. The EVATAR system assembly needs special equipment, as the process involves producing an intermediate thermoplastic track-etched membrane from materials such as polycarbonate or polyethylene terephthalate (47). An alternative popular material is PDMS which has a user-friendly protocol, optical clarity, and requires commonly available lab equipment (181). However, PDMS has the tendency to absorb hydrophobic molecules, specifically hormones, making its applicability in reproductive research questionable (182, 183). The lack of alternatives and the ability to perform shorter experiments with certain coatings of the PDMS has made it acceptable for some studies (184-186). However, when studying the altered response of biological processes due to exposure to drugs, xenobiotics, pathogens, and environmental toxicants, it is important to quantify and account for potential material–compound absorption or interaction before concluding the findings.

Another popular material for constructing microfluidic devices is polysulfone (PSF) (135). This rigid machinable thermoplastic offers several advantages over PDMS and is often used in medical devices since it has US Food and Drug Administration (FDA) food grade approval (21CFR177.1655) as well as United States Pharmacopeia (USP) Class VI biocompatibility (187). The material is easy to sterilize as it can be autoclaved and is impenetrable to chemical solvents (188). Compared with most elastomers such as PDMS, polysulfone has significantly lower adsorption and absorption of hydrophobic compounds, making it suitable for studies involving drugs and hormones (189). While the material lacks the optical clarity and oxygen permeability of PDMS, neither of these factors are significant challenges as most microfluidic systems are open and have an air-liquid interface where gas exchange can occur.

Each microfluidic device produces a different fluid flow and shear stress profile. There is a large amount of optimization required to determine the ideal settings and configurations for replicating an organ’s in vivo structure and microenvironment. Moreover, the technical details become more challenging when culturing multiple cell lines or tissues on a single platform, as this creates a need to find a flow rate, sample volume, and scaling for the size of tissues. Research is needed into developing a media composition that is compatible with multiple organ model systems or cell types, as media needs to be circulated throughout the device to ensure interaction through secreted factors and hormones.

Conclusion

Studying the physiology of a system as intricate as the female reproductive tract is challenging. Microfluidic systems integrate engineering and biology to revolutionize cell culturing and experimenting techniques. The ability to integrate cellular and tissue level cytokine and endocrine loops into signaling pathways and to engineer precise temporal manipulations of microliter volume of fluids and hormones allows the reproducible study of the female reproductive tract in a physiologically relevant setting. While progress is required to optimize the use of microfluidic systems, this technology presents a unique opportunity to study human embryology, reproduction, and disorders of the female reproductive tract.

Glossary

Abbreviations

3D

3-dimensional

ART

assisted reproductive technologies

EDC

endocrine disrupting chemical

HGSC

high-grade serous carcinoma

PDMS

polydimethylsiloxane

PIGF

phosphatidylinositol-glycan biosynthesis class F protein

sFlt-1

soluble fms-like tyrosine kinase-1

Additional Information

Disclosures: The authors have no conflicts of interest to disclose.

Data Availability

Data sharing is not applicable to this article because no data sets were generated or analyzed during the present study.