Organs-on-a-chip to understand reproduction

History of cell culture

Cell culture, the process of growing cells in a controlled artificial environment, has a rich history marked by significant scientific advancements. The cell theory, which posits that living organisms are composed of cells, was introduced by Matthias Schleiden and Theodor Schwann in the early 19th century, laying the groundwork for cell biology (Vasil 2008). At the beginning of the 20th century, Ross Harrison and Leo Loeb kept tissue or pieces of organs alive in test tubes in liquid medium (Taylor 2014). By the mid-20th century, the use of antibiotics in cell culture greatly reduced contamination issues. The first continuous human cell line, HeLa cells, which were created in 1951 produced from Henrietta Lacks’ cervical cancer cells, revolutionized medical science. Advances in culture media compositions, like Eagle’s Minimal Essential Medium made by Harry Eagle, standardized culture media, thus enhancing the consistency in experiments (Eagle 1959). The next decades brought innovations in developmental biology, drug discovery, and regenerative medicine, with the advent of three-dimensional (3D) cell culture and organoids offering more physiologically relevant models. Cell culture is of paramount importance in science and medicine and continues to drive forward scientific and therapeutic advancements.

In vitro models – the endometrium

In vitro models have become increasingly important for advancing our understanding of tissue function and mechanisms. This is particularly true for the endometrial lining of the uterus which is the first point of contact for a developing embryo to implant and settle in for the duration of the pregnancy. In humans, implantation is successful only if the endometrium has remodeled to a ‘receptive’ state which is driven by ovarian sex hormones (Lessey and Young 2019). This complex process has prompted extensive research, and as a result, the endometrium has been studied outside the body in various ways. To retain key attributes of native endometrial tissue, including tissue architecture, composition, and direct intercellular contact, endometrial explants were used for short-term studies (Schafer et al. 2011). However, substantial tissue degradation occurred rapidly and does not allow for long-term studies. As a result, organotypic endometrial models were created, which usually included the major cell types, stromal and epithelial cells embedded in a matrix to mimic the scaffolding and structure of uterine tissue. The endometrial response to hormones is well replicated in these models, demonstrating the importance of paracrine and endocrine signaling amongst various cell types (Bläuer et al. 2005; Murphy et al. 2022).

A widely adopted 3D culture method of endometrial epithelial cells was developed by two independent groups, Turco et al. (2017) and Boretto et al. (2017). Both luminal and glandular-like endometrial epithelial cells were present in organoids, cultured in Matrigel, and maintained significant endometrial characteristics, including glycogen production and response to hormones (Boretto et al. 2017). The organoids could be maintained for more than 4–6 months and were able to be passaged. Other organoid models have been developed to further mimic the endometrium by including an additional cell type: stromal cells. A scaffold-free multicellular organoid included both primary endometrial epithelial and stromal cell type (Wiwatpanit et al. 2020). Primary epithelial and stromal cells, plated on low-adhesion agarose micromolds, self-organized to reflect the native tissue with one cell layer of polarized epithelial cells surrounding and inner cluster of stromal cells. The organoids expressed functional markers, including estrogen receptor, progesterone receptor, and androgen receptor, and responded to menstrual cycle hormones. Another two-cell organoid was constructed and termed an assembloid, bringing together epithelial organoids in Matrigel, surrounded by a bed of stromal cells on the exterior, which responded to hormones and kinase inhibitors (Rawlings et al. 2021). Organoid technology has changed how researchers study the endometrium and its diseases, with a number of reviews written in recent years (Campo et al. 2020; Alzamil et al. 2021; Cadena et al. 2021; Fitzgerald et al. 2021; Heremans et al. 2021; Song and Fazleabas 2021; Murphy et al. 2022).

The female reproductive tract

As part of the Tissue Chip Program, the EVATAR system, the first multi-organ microphysiological model of the female reproductive tract, was developed through a collaboration between reproductive biologists and engineers (Xiao et al. 2017) (Fig. 1a). To integrate five tissues into a single system, a practical and scalable microfluidic device was engineered. This device consisted of modules, similar to a well in a 24-well plate, with each module large enough to house 3D cultures and explants. Each module had flow ports to facilitate flow of media through all the modules. The flow of media was powered with electromagnetic actuated micropumps and built for precise management of flow rates.

Fig. 1.

Multiwell microfluidic platforms to study interactions of female reproductive tract tissues. (a) EVATAR, the first microphysiological system of the female reproductive tract was developed to promote interactions between multiple tissue types through microchannels connecting the wells. (b) LATTICE is a plug-and-play microfluidic platform that can accommodate up to eight large tissues or organ models. It builds on the capabilities of EVATAR but with simpler engineering and lower cost. Images of EVATAR and LATTICE were reproduced from Xiao et al. (2017) and Campo et al. (2023), respectively.

Within this platform, tissue mimics of the ovary, fallopian tube, endometrium, cervix, and non-reproductive tissue (liver) were connected with continuous media flow through each compartment. Mouse ovarian explants were treated with gonadotropic hormones (FSH and LH) to mimic a human menstrual cycle over 28 days. In response, estradiol and progesterone was secreted in a pattern resembling the human 28-day cycle, peaking during the follicular and luteal phases. Human fallopian tube explants, endometrial cells in decellularized scaffolds, human ectocervix explants, and human liver microtissues cultured together in the microfluidic device maintained their integrity and function throughout the 28-day period. These tissues exhibited robust responses to fluctuating levels of ovarian hormones, showing improvements compared to single cultures. Moreover, the use of a shared universal media highlighted the exchange of paracrine factors among the integrated tissues.

The success of this proof-of-concept device propelled the next phase of the study, which was to build a device that was more accessible in engineering and cost as well as to model a disease using the multi-organ format. A second-generation prototype was built named LATTICE (Campo et al. 2023), based on the design of the first platform (Fig. 1b). The device consisted of a power unit and a compatible culture plate with microfluidic channels that facilitated dynamic media mixing and flow between tissue wells. The device was made from biocompatible materials that did not absorb lipophilic compounds like hormones and drugs and the plate dimensions resembled a 24-well culture plate format.

Polycystic ovarian syndrome (PCOS) was the multi-organ reproductive disease chosen to be represented in LATTICE. A hyperandrogenic ovary was used as a source of high androgens, commonly found in women with PCOS, by increasing the FSH/LH ratio added to the mouse ovaries (Campo et al. 2023). Within the microfluidic device, the hyperandrogenic ovary promoted a significant downregulation of cilia beating in the fallopian tube explants. Ultimately, with the inclusion of other tissue types relevant to PCOS like adipose tissue, pancreatic islets, and endometrium, drugs can be screened and their impact on each tissue type can be assessed. As the multi-well platform is able to replenish nutrients and eliminate waste, thereby promoting prolonged physiologic tissue function, interactions between tissues types relevant to reproductive diseases like endometriosis, endometrial cancer, ovarian aging, and ovarian cancer, can be modeled using OoC, obtaining insights into new mechanisms of development and growth in a controlled setting.

Human cervix chips

To study the impact of infection on the lower female reproductive tract, one group developed a six-chamber organ-on-a-chip model (VCD-OOC) that mimicked the female reproductive tract during pregnancy (Tantengco et al. 2022). These chambers housed vaginal epithelial cells, cervical epithelial and stromal cells, and decidual cells. Using this model, this group demonstrated a similar response in the cells to a Ureaplasma parvum infection in the VCD-OOC as found in the animal model of infection.

Another group built a cervix chip to replicate the human cervical epithelial–stromal interface (Izadifar et al. 2024). Similar to the native cervix tissue, the cells in the cervix chip displayed a functional epithelial barrier, producing mucus and responding to hormones. Using this model, this study demonstrated responses to healthy versus dysbiotic microbial infection (specifically Lactobacillus crispatus and Gardnerella vaginalis) including tissue innate immune responses, barrier integrity, cell viability, proteomic profiles, and mucus composition, which were significantly different between the two microbial communities. This human cervix chip was shown to be a physiological representation of the cervix tissue and a useful model to study the cervix and associated diseases.

Senescence in ovarian aging

Cellular senescence has attracted significant attention due to its impact on overall health, with cooperative groups like the Cellular Senescence Network (SenNet) working to create an atlas of senescence across multiple tissues in human health (SenNet Consortium 2022; Suryadevara et al. 2024). Senescent cells are in a state of permanent proliferative arrest, accumulate with biological age, and create a microenvironment that can negatively affect tissue function, also called the bystander effect (Waaijer et al. 2012; Childs et al. 2015). This is due to the secreted factors of senescent cells, termed Senescence-Associated Secretory Phenotype (SASP), which include pro-inflammatory cytokines, chemokines, growth factors, proteases, and lipids. There is a growing body of research associating cellular senescence with ovarian aging (Secomandi et al. 2022; Wu et al. 2024); however, much remains to be learned to fully understand the precise role of senescence and the bystander effect on ovarian function and aging.

Ongoing studies of senescence in the ovary use the LATTICE platform which provides new features well-suited to understanding ovarian aging. Current in vitro methods involve the treatment of single cell type with specific inducers of senescence in static cultures (Huang et al. 2022; Veronesi et al. 2023) with waste accumulation potentially causing artificial responses. LATTICE continuously replenishes media, removing any stress responses that could confound data. The larger wells in the LATTICE plate also allow for the culture of multiple ovarian follicles at time. A customized agarose micromold was made to hold multiple multilayer secondary follicles in a single focal plane, allowing for constant time-lapse monitoring and automated analysis (Zaniker et al. 2024). Finally, the multi-well platform was designed to recapitulate organ–organ interactions in vitro with control of directionality of these interactions. The systemic effects of the aging ovary on other organ systems is an active area of research.

Embryo and fertility research

The OoC technology can enhance current embryo and fertility research by offering more precise, controlled, and scalable models. Studies in embryo implantation face significant challenges due to the complex and dynamic interactions between maternal and fetal tissues, as well as the ethical and practical limitations of in vivo research. One group used a microfluidic device to gain understanding on preeclampsia which is considered to be characterized by inadequate invasion of extravillous placental trophoblast cells into the uterine arteries (Abbas et al. 2017). This device was comprised of three channels with the fetal extravillous trophoblasts (EVTs) in the middle channel and flow of media creating cytokine gradients in the outer channels. This study demonstrated that granulocyte-macrophage colony-stimulating factor, a cytokine that is produced by activated decidual natural killer (NK) cells, stimulated the migration of EVTs with specific directionality.

Another study developed an intricate 3D maternal–fetal interface using OoC technology termed implantation-on-a-chip (Park et al. 2022). In this microengineered model, primary human EVTs isolated from clinical specimens, endometrial stromal fibroblasts from endometrial tissue, endothelial cells to mimic maternal vessels, and uterine NK cells isolated from endometrial tissues, were included in a device consisting of three parallel lanes, each for a specific cell type. The model demonstrated in vivo-like directional migration of EVTs towards microengineered maternal vessels, and showed that decidual cells could impact the invasion process. They were also successful in demonstrating the important role of uterine NK cells in influencing EVT invasion. Novel and important interactions and secreted factors were identified in this study, offering new insights into early human pregnancy dynamics.

Embryo development and screening

Improvements in preimplantation embryo culture, a critical component of assisted reproductive technology (ART), have lagged behind advancements in embryo selection (Vajta et al. 2021). The challenge lies in replicating complex in vivo conditions for in vitro embryo culture. Although human ART data is less conclusive, animal models, particularly bovines, are notably affected by these in vitro conditions. Differences in developmental competence, cryotolerance, and pregnancy outcomes are significant between embryos developed in vitro and those derived in vivo (Abe et al. 2002; Rizos et al. 2002a, 2002b). It was reported that the pregnancy and live birth rates of embryos produced in vitro remained lower compared to in vivo-produced embryos with pregnancy rates at 40.1% and 64.1%, respectively (Ealy et al. 2019). Despite these statistics, over 2 million bovine embryos were collected or produced in 2021, with in vitro-produced embryos surpassing those produced in vivo (Viana 2022).

To improve embryo quality in vitro, an in-depth understanding of the in vivo condition in the local environment of the developing embryo is needed, including the paracrine actions that occur between the embryo and the maternal tissues like the oviduct and the endometrium. Furthermore, biophysical forces during embryo transport in vivo can also be considered. Studies have mimicked the fluid mechanical stimulation of the embryos that occur in the oviduct using dynamic culture systems (Kim et al. 2009; Matsuura et al. 2010; Esteves et al. 2013; Gualtieri et al. 2024). Advanced culture systems such as these hold significant potential for enhancing embryo culture and ART success.

Improving ART in animals has profound impacts across various domains. It enables the selection and propagation of genetically superior animals, to enhance productivity and disease resistance. ART techniques also play a crucial role in conserving endangered species and preserving valuable genetic diversity. Moreover, ART can increase the efficiency of breeding programs, which in turn boosts agricultural productivity and economic returns. Beyond economic benefits, advancements in ART contribute to broader research in reproductive biology and medicine, benefiting animal welfare. By offering alternatives to traditional breeding methods, ART can also improve ethical considerations in animal husbandry, promoting sustainable practices and reducing environmental impact.