High-Throughput 3D-Printed Model of the Feto-Maternal Interface for the Discovery and Development of Preterm Birth Therapies

Abstract

Spontaneous preterm birth (PTB) affects around 11% of births, posing significant risks to neonatal health due to the inflammation at the fetal-maternal interface (FMi). This inflammation disrupts immune tolerance during pregnancy, often leading to PTB. While organ-on-a-chip (OOC) devices effectively mimic the physiology, pathophysiology, and responses of FMi, their relatively low throughput limits their utility in high-throughput testing applications. To overcome this, we developed a three-dimensional (3D)-printed model that fits in a well of a 96-well plate and can be mass-produced while also accurately replicating FMi, enabling efficient screening of drugs targeting FMi inflammation. Our model features two cell culture chambers (maternal and fetal cells) interlinked via an array of microfluidic channels. It was thoroughly validated, ensuring cell viability, metabolic activity, and cell-specific markers. The maternal chamber was exposed to lipopolysaccharides (LPS) to induce an inflammatory state, and proinflammatory cytokines in the culture supernatant were quantified. Furthermore, the efficacy of anti-inflammatory inhibitors in mitigating LPS-induced inflammation was investigated. Results demonstrated that our model supports robust cell growth, maintains viability, and accurately mimics PTB-associated inflammation. This high-throughput 3D-printed model offers a versatile platform for drug screening, promising advancements in drug discovery and PTB prevention.

Keywords: 3D-printed high-throughput model; drug screening platform; feto-maternal inflammation; feto-maternal interface organ-on-chip; infection; pregnancy; preterm birth.

Figure. 1.

Visual representation of a pregnant woman with a growing fetus enveloped by fetal membranes and placenta, forming the feto-maternal interface. Maternal and fetal cells are co-cultured in a 3D-printed scaffold, employing 3D printing technology and a cell type-specific culture environment, to mimic the FMi for efficient high-throughput drug screening applications. (Created with BioRender.com)

Figure. 2.

(A) The device design aimed to mimic the feto-maternal interface, featuring two separated cell culture chambers interconnected by an array of microchannels. The microchannels had a diameter of 430 μm, while the chambers had a height of approximately 10 mm and a diameter of 6.35 mm. Top and side views of the 3D-printed device filled with different colored dyes in each chamber for easy visualization (scale bar = 1.5 mm). (B) The 3D-printed device is designed for seamless integration into wells of a 96-well plate, facilitating high-throughput studies and enabling effortless multi-pipette operations. (C) Cytokine absorption was assessed by introducing known concentrations of recombinant cytokines IL-6 and IL-8 into the devices. Absorption was evaluated after 72 h using ELISA. Devices coated with Parylene C demonstrated decreased cytokine absorption compared to uncoated devices. Quantitative values are presented as mean ± SEM and statistical analysis was performed using the Mann-Whitney U and Student’s t-test, with significance indicated by *p<0.05 (N=3–6). (D) Drug compound absorption was evaluated by introducing known concentrations of drug compounds Ruxolitinib (JAK-inhibitor) and U0126 (MAPK inhibitor) into the devices and assessing absorption after 24 h via mass spectroscopy. Devices were coated with Parylene C for this experiment. Quantitative values are presented as mean ± SEM, and statistical analysis was performed using the Mann-Whitney U and Student’s t-test, with significance indicated by *p<0.05(N=3–4).

Figure. 3.

(A) Propagation of LPS-FITC under different device conditions through the microchannel array was examined at multiple time points. At the 72 h time, LPS propagation and absorption were not observed in the absence of microchannels. However, with the presence of microchannels, propagation reached approximately 50%. When GelMA or GelMA with cells was introduced, the propagation stabilized at 25%. Values are depicted as mean ± SEM. Statistical analysis was performed using the Mann-Whitney U and Student’s t-test, with significance indicated by *p<0.05 (N=3). (B&C) Drug compound propagation between the two chambers through the microchannels was evaluated at various time intervals. At the 6 h time point, a minor propagation of the JAK inhibitor was detected, whereas the MAPK inhibitor exhibited slightly higher propagation as early as the 1 h time point. Values are depicted as mean ± SEM.

Figure. 4:

(A) Decidual cells and amnion epithelial cells were subjected to 3D culture within GelMA matrices, and their viability was evaluated at 24-, 48-, and 72-hours post-culture using live (green – calcein AM) and dead (red – ethidium homodimer) fluorescent stains (scale bar = 50 μm). (B&C) Cell cytotoxicity was assessed via the lactate dehydrogenase (LDH) assay, while cellular metabolism, indicative of cell viability, was evaluated using the AlamarBlue assay at the 72-hour time point following the 3D culture. Data presented are mean ± SEM. Statistical analysis was performed using the Mann-Whitney U and Student’s t-test, with significance indicated by *p<0.05 (N=4 for LDH assay and N=5 for Alamar Blue assay). (D) Cell morphology was assessed by comparing cells cultured in 2D vs. 3D environments, focusing on the expression of specific markers: cytokeratin-18 (CK-18) and vimentin for epithelial cells, and vimentin alone for decidual cells (scale bar = 20 μm).

Figure. 5:

(A) Cellular viability assessed via the lactate dehydrogenase (LDH) assay under various disease conditions: D1 (LPS on the maternal side), D2 (LPS and Poly IC on the maternal side), D3 (LPS on both maternal and fetal chambers), at 6 and 24-hour intervals. (B) Cellular metabolism, indicative of cell viability, was evaluated using the AlamarBlue assay under the aforementioned disease conditions and time points. (C) Interleukin-6 (IL-6) cytokine concentrations were quantified within maternal and fetal chambers across varying disease states at 6-hour and 24-hour time points. (D) Interleukin-8 (IL-8) cytokine concentrations were measured within the maternal and fetal chambers across various disease conditions at 6-hour and 24-hour hours. Blue bars represent healthy states, whereas different disease conditions are represented by light to dark red bars. Values are presented as mean ± SEM, and statistical analysis was conducted using the Mann-Whitney U test and Student’s t-test, with statistical significance indicated as *p<0.05 & **p<0.01 (N=6 for all cases).

Figure. 6:

(A) Cellular viability assessed via the lactate dehydrogenase (LDH) assay under different conditions: Healthy, disease (LPS on the maternal side), drug-1 (JAK inhibitor treatment), drug-2 (MAPK inhibitor treatment), at 6 and 24-hour intervals. (B) Cellular metabolism, indicative of cell viability, was evaluated using the AlamarBlue assay under the aforementioned conditions and time points. (C) Interleukin-6 (IL-6) cytokine concentrations were assessed within the maternal and fetal chambers across healthy, diseased (LPS-induced), and drug-treated models at 6-hour and 24-hour intervals. (D) Interleukin-8 (IL-8) cytokine concentrations were measured within the maternal and fetal chambers across various disease conditions at 6-hour and 24-hour hours. Blue bars depict the healthy state, while red bars represent the diseased state. The two drug-treated states are visualized in green (drug1: JAK inhibitor treatment) and dark green (drug2: MAPK inhibitor treatment). Values are presented as mean ± SEM, and statistical analysis was conducted using the Mann-Whitney U test and Student’s t-test, with statistical significance indicated as *p<0.05, **p<0.01, ***p<0.001, ****p<0.0001 (N=9 for all cases).