Trophoblast organoids with physiological polarity model placental structure and function

Three-dimensional organoid culture models from tissue-derived stem cells have emerged as important ex vivo systems to study a variety of aspects of the physiological and pathological states of their tissues of origin. Established organoid models often preserve key features of their source organs, including tissue organization and composition, expression signatures, immune responses and secretion profiles. Importantly, organoid cultures can be propagated long-term and can often be cryopreserved, and thus have the capacity to serve as powerful in vitro tools even in the absence of access to new donor tissue. Over the past several years, trophoblast organoids (TOs) derived from human placentas at different gestational stages have emerged as models for the study of trophoblast development and biology, congenital infections and innate immune defenses (Haider et al., 2018; Sheridan et al., 2020; Turco et al., 2018; Yang et al., 2022). We have shown that TOs can be derived and cultured from full-term human placental tissue and used to model trophoblast immunity and teratogenic viral infections (Yang et al., 2022).

In tissue-derived TOs models, trophoblast stem/progenitor cells are isolated from placental chorionic villi by serial dissociation with digest solution followed by mechanical disruption (in the case of full-term tissue), then are embedded within an extracellular matrix (ECM; such as Corning Matrigel) ‘domes’. The domes containing isolated trophoblast stem/progenitor cells are then submerged in growth factor cocktail-reconstituted growth medium to support stem/progenitor cell proliferation and differentiation, and promote their self-organization into mature organoid units. TOs differentiate to contain all trophoblast subtypes present in the human placenta, including proliferative cytotrophoblasts (CTBs), which differentiate into the multinucleated non-proliferative syncytiotrophoblast (STB), and invasive extravillous trophoblasts (EVTs). Human chorionic villi are covered by an outermost STB layer and an inner CTB layer that fuses to replenish the outer STB during pregnancy. However, TOs cultured as three-dimensional organoids embedded in ECM develop with the opposite polarity, and mature organoids contain an inward-facing STB (STBⁱⁿ) and an outward-facing CTB layer (Turco et al., 2018; Yang et al., 2022). This inverse polarity limits the utility of TOs for studies that require access to the STB layer. For example, STBⁱⁿ TOs might not recapitulate the vertical transmission route of teratogenic infections, the transport of nutrients and antibodies across the STB, or the directionality of hormones and other factors that are critical for communication to maternal tissues and cells.

To overcome the limitation of existing TO models, we developed a suspension culture method to change the polarity of TOs such that the STB layer is outward facing (STB^out). Similar approaches have been developed and applied to a variety of epithelial-derived organoid models (Co et al., 2019, 2021; Krüger et al., 2020; Li et al., 2020; Salahudeen et al., 2020). We show that this culture method also enhances the secretion of hormones and cytokines associated with the STB. Furthermore, we performed patch clamping of STBⁱⁿ and STB^out TOs to measure the size of cells comprising the outermost layer of these organoids and found that STB^out organoids are covered by large syncytia (>50 nuclei), whereas the STBⁱⁿ TOs contain smaller syncytia (<10 nuclei) and are largely composed of mononuclear cells. The STB^out TO culture model described here thus better reflects the physiological and pathological processes of the human placenta, which can facilitate studies to define the underlying mechanisms of normal and diseased placental conditions.

Like epithelial-derived organoids, TOs cultured within ECM domes have an inward facing apical surface (Co et al., 2019, 2021). However, the polarity of epithelial-derived organoids can be reversed by culturing of mature organoids under suspension culture conditions, which can occur within ∼24 h of initiating these cultures (Co et al., 2019, 2021; Krüger et al., 2020; Li et al., 2020). Given this, we developed a TO culturing approach that involved the culturing of organoids for 7 days in Matrigel domes to promote their differentiation and maturation, then the release of organoids from Matrigel. We used TOs isolated from full-term placental tissue that had been previously characterized (Yang et al., 2022). Once released, organoids were cultured for an additional period of 24–48 h in suspension with gentle agitation (schematic, Fig. 1A). Unlike epithelial organoids in which polarity reversal can be distinguished based on brightfield microscopy alone (Co et al., 2019, 2021), we were unable to clearly distinguish between TOs grown in suspension (STB^out) and those cultured in Matrigel domes (STBⁱⁿ) based on brightfield microscopy alone (Fig. 1B). To directly compare the architecture and viability of STBⁱⁿ versus STB^out TOs, we performed hematoxylin and eosin (H&E) staining of cryo-sections in parallel to immunostaining for a marker of CTBs (ITGA6) and the STB (the β-subunit of human chorionic gonadotropin, encoded by several genes, herein denoted CGB). Similar to what is seen with first-trimester TOs (Turco et al., 2018), we found that STBⁱⁿ organoids formed intracellular ‘mini-cavities’, with ITGA6-positive CTBs lining the outer surface of the organoids and the CGB-positive STB distributing across these intra-organoid cavities, which varied in size and quantity between organoids (Fig. 1C, top rows). Although we found that STB^out organoids also formed mini-cavities, the ITGA6-positive CTBs were largely localized to the intracellular compartment, with the CGB-positive STB on the outer surface (Fig. 1C, bottom rows). A schematic of the general architecture of these organoids is shown in Fig. 1C, left. In some cases, we did observe greater aggregation of and/or fusion between organoids in STB^out TOs grown in suspension (Fig. S1A). However, this aggregation could be avoided by limiting the number of organoids seeded into each well while in suspension culture (to <100 organoids) and dissociating aggregates by manual pipetting should aggregation occur during suspension culture.

To determine the kinetics of the formation of STB^out TOs, we cultured organoids for between ∼21 and 48 h and performed three-dimensional confocal microscopy for syndecan-1 (SDC-1), a cell surface proteoglycan that localizes to the apical surface of the STB, and a pan-trophoblast cytokeratin (cytokeratin-19; CYT-19, also known as AS3MT). We found that the STB began to appear at the outer surface of TOs cultured in suspicion by ∼20 h post-incubation but required 48 h in culture to reach the maximal exterior localization (Fig. 1D,E). We did not observe any increased toxicity during this time as assessed by light microscopy, H&E staining or lactate dehydrogenase levels in the medium (Fig. 1B,C; Fig. S1B).

To evaluate the efficiency of the STB^out model, and to determine whether contact between organoids was required for STB^out TO formation, we developed a U-bottomed ultra-low attachment 96-well plate-based culture format in which individual TO units were cultured in suspension in individual wells. To do this, we collected mature TO units derived from three unique donor tissues from Matrigel domes for a total of ∼40 organoids. Then, individual organoids were cleared from traces of Matrigel using a needle under a tissue culture microscope and placed into a well of a U-bottomed plate for 48 h suspension culturing (schematic, Fig. 2A,B, top). Individual organoids were then fixed and immunostained for ITGA6 and SDC-1 to determine the orientation of the CTB and STB. We retrieved 37 single suspension cultured organoid units after immunostaining and found that of these organoids, 34 exhibited >50% exterior SDC-1 immunolocalization compared to matched STBⁱⁿ organoids, in which only one of 37 organoids exhibited this orientation (Fig. 2C).

To further define the localization of key trophoblast cell populations between STBⁱⁿ and STB^out TOs, we performed whole-organoid immunostaining followed by three-dimensional confocal microscopy. SDC-1 localized to interior mini-cavities in STBⁱⁿ TOs (Fig. 3A; Movie 1). In contrast, SDC-1 almost exclusively localized to the outermost surfaces of STB^out TOs (Fig. 3A; Movie 2). Similarly, CGBs localized to the inner-most surfaces of STBⁱⁿ TOs and were surrounded by outer layers of ITGA6-positive CTBs (Fig. 3B; Movie 3). In contrast, CGB localization was on the exterior of STB^out TOs, with ITGA6-positive CTBs comprising the interior of organoids (Fig. 3B; Movie 4).

The STB is a primary producer of hormones required for pregnancy, including human chorionic gonadotropin (hCG) which is comprised of two subunits, CGB and CGA. We and others have shown that STBⁱⁿ TOs recapitulate this secretion (Turco et al., 2018; Yang et al., 2022). To determine whether there were differences in the secretion of hCG between STBⁱⁿ and STB^out TOs, we performed Luminex assays with conditioned medium from both STBⁱⁿ and STB^out TOs. To control for variability in organoid size and quantity, we also collected and quantified total protein, which was used to normalize these values. We found that there were significantly higher levels of hCG in medium collected from STB^out TOs compared to that from STBⁱⁿ TOs (∼10,000 pg/ml versus ∼7000 pg/ml, respectively) (Fig. 3C), consistent with the enhanced localization of the STB to the outer surfaces of TOs.

TOs can self-differentiate to contain small amounts of human leukocyte antigen G-positive (HLA-G⁺) EVTs (Sheridan et al., 2020; Yang et al., 2022). To determine whether this occurred or was altered by STB^out suspension culture conditions, we performed immunostaining for HLA-G in STBⁱⁿ and STB^out TOs. Consistent with our previous study (Yang et al., 2022), we found that STBⁱⁿ TOs differentiated to contain <10% HLA-G⁺ EVTs (Fig. 3D,E). In contrast, we were unable to detect any HLA-G⁺ cells in STB^out TOs, suggesting that suspension culture condition reduces spontaneous EVT differentiation. Although rates of EVT differentiation can be promoted by altering the composition of culture medium (Sheridan et al., 2020; Yang et al., 2022), this process requires an extended culture period of >3–4 weeks, which is beyond the time frame possible to culture STB^out TOs in suspension. Thus, it remains unclear whether these STB^out organoids can also be cultured to promote EVT differentiation.

In addition to hormones, the STB also secretes cytokines required to facilitate the establishment of tolerance and/or to defend the fetus from pathogen infection, such as the release of the antiviral type III interferons (IFNs) IFN-λs (Bayer et al., 2016). We have previously shown that TOs recapitulate this secretion and release a number of these cytokines, including IL-6 and IFN-λ2 (Yang et al., 2022). To determine whether STB^out TOs maintain this cytokine secretion or induce unique cytokines and chemokines compared to STBⁱⁿ TOs, we performed multiplex Luminex profiling of 73 cytokines and chemokines, a subset of which we previously showed were released from STBⁱⁿ TOs (Yang et al., 2022). We did not observe any secretion of cytokines and chemokines in STB^out TOs that were not also secreted from STBⁱⁿ TOs (Fig. 4A). However, we found that STB^out TOs secreted higher levels of two factors, IFN-λ2 (>7-fold increase) and IL-6 (>4-fold increase) (Fig. 4B–D). In contrast, other analytes such as MIF (also known as GRO-α) were secreted at similar levels between both STBⁱⁿ and STB^out TOs (Fig. 4A,E). One chemokine, CXCL1, which has been associated with decidual stromal cell responses to trophoblasts (Hess et al., 2007) and which uses SDC-1 as a co-receptor (Carey, 1997), was significantly reduced in STB^out TOs (>12-fold reduction) (Fig. 4B,F). Collectively, these results suggest that STB orientation in TOs has implications on immune secretion profiles.

Cell fusion dramatically increases the surface area of the fused cell. As cell surface area is proportional to its membrane capacitance (Cm) (Hodgkin and Huxley, 1952), patch clamp, a quantitative electrophysiological technique (Gillis, 1995; Neher and Sakmann, 1976), can be used to evaluate cell size. We therefore utilized patch clamping to calculate the size of the cells and syncytia comprising the exterior cellular surface of STBⁱⁿ versus STB^out TOs (schematic, Fig. 5A). When a small voltage step (10 mV) was applied to TO lines derived from three unique donor placentas, the capacitive current from STB^out TOs showed much slower decay than the capacitive current from STBⁱⁿ TOs (Fig. 5B). The average Cm values in STBⁱⁿ were 0.238 nF, 0.076 nF and 0.104 nF for code 1, code 2 and code 3, respectively, whereas the average Cm in STB^out TOs were 2.812 nF, 1.603 nF and 1.899 nF, respectively for the three codes (Fig. 5C). Interestingly, the Cm of the surface trophoblasts in STB^out TOs exhibited a Gaussian distribution in all lines tested (Fig. 5D–F). In stark contrast to the broader distribution of the Cm from STBⁱⁿ TOs centered at 0.113 nF, 0.051 nF and 0.087 nF for code 1, code 2, and code 3, respectively, the Cm from the STB^out TOs was largely centered at 3.350 nF, 1.230 nF and 1.622 nF, which is ∼20–30-fold larger than in STBⁱⁿ TOs. It is worth noting that extremely large syncytia are readily observed on the surfaces of STB^out TOs (Fig. S2A). We recorded these cells from five independent STB^out TOs and found that they exhibited unmeasurable cell capacitance (Fig. S2B). This is likely due to the space clamp issue for syncytia with extremely large surface areas (Spruston et al., 1993). The smallest Cm measured in our study, which represents the Cm of a single mononuclear cell, was ∼30 pF. Syncytialization involves fusion of plasma membranes, and Cm is directly proportional to the cell surface area. Therefore, the extent of syncytialization in STBⁱⁿ and STB^out TOs can be estimated by dividing the Cm of the syncytia by the single cell capacitance of 30 pF. By performing these calculations, we found that the surface trophoblasts observed in the STBⁱⁿ samples were predominantly composed of individual cytotrophoblasts (CTBs) and syncytia with limited fusion (less than 10 nuclei). On the other hand, the surface trophoblasts in the STB^out samples are primarily comprised of syncytia with a greater number of nuclei, exceeding 50 nuclei.

In this study, we develop a method to culture trophoblast organoids under conditions that reflect their physiological cellular orientation in vivo. This model facilitates access to the STB layer while also maintaining key features associated with STBⁱⁿ TOs, including their three-dimensional morphology, the presence of distinct trophoblast subpopulations, and the secretion of pregnancy related hormones and immune factors. STB^out TOs have several advantages over STBⁱⁿ TOs. For example, STB^out TOs naturally self-reorganize with an STB outward-facing surface and do not require extensive manipulation to develop this outer layer. In addition, as STB^out TOs are cultured in suspension, the lack of ECM allows for applications in which this scaffold presents a barrier to diffusion, such as studies of microbial infections or antibody uptake.

For epithelial organoids grown in ECM domes with basal-out polarity, microinjection can serve as an option to directly access the enclosed apical surface (Bartfeld et al., 2015; Bartfeld and Clevers, 2015). However, in contrast to epithelial-derived organoids, which often form clear cystic structures, TOs have heterogeneous mini-cavities, which makes microinjection of these organoids difficult. Additional methods have been applied to epithelial-derived organoids, such as seeding dissociated organoid fragments onto Transwell inserts (Good et al., 2019; VanDussen et al., 2015). However, this approach compromises the three-dimensional nature of organoids, which might impact their function. The method we describe here avoids several of these challenges, as STB^out TOs maintain their three-dimensional structure and do not require their disruption to generate. It is unclear whether STBⁱⁿ TOs undergo similar mechanisms of polarity reversal as do epithelial-derived organoids, which undergo relocalization of junction-associated proteins to mediate this process, or whether culturing in suspension instead promotes CTB fusion on the organoid surface. Given that the surface of STB^out TOs is covered by very large syncytia, it is possible that suspension culturing promotes the fusion of CTBs on the organoid surface rather than inducing a relocalization of the STB from the inner to outer organoid surface. This fusion could be promoted by factors including low levels of shear stress during suspension culturing, which has been proposed to enhance rates of CTB fusion (Brugger et al., 2020). Fluid shear is known to impact myriad aspects of epithelial cell function, including formation of microvilli (Miura et al., 2015), suggesting that this shear likely also impacts rates of CTB fusion.

A benefit of TOs is their ability to recapitulate the hormone and cytokine secretion observed in primary trophoblasts and chorionic villous tissue explants (Yang et al., 2022), which is not recapitulated in standard trophoblast cell lines (Bayer et al., 2016). However, given that STBⁱⁿ TOs are embedded in Matrigel, many of these STB-associated factors would be secreted into the center of the organoid structure or perhaps into the surrounding ECM. We found that STB^out TOs not only recapitulate the release of these factors, but that some factors were secreted at significantly higher levels than those observed in STBⁱⁿ TOs. The mechanistic basis for this is likely two-fold and could include the increase in syncytia size on the STB^out TO surface as well as the direct release of these factors into the culture medium. However, we found that STB^out organoids had substantially lower levels of HLA-G⁺ EVTs, suggesting that there is reduced EVT differentiation, which could impact EVT-specific secretion profiles. It is not clear whether methods to promote EVT differentiation previously applied to TOs derived from full-term tissue (Yang et al., 2022) could also be applied to the STB^out TO system. However, the extended time to perform this procedure (>3 weeks) makes it is unlikely that STB^out TOs would be amenable to this process.

We leveraged the power of electrophysiology to define the size of cells and syncytia covering the surface of STBⁱⁿ and STB^out TOs. These studies verified the high efficiency of the STB^out TO system and provided quantitative measurements of the number of nuclei comprising syncytia. Based on these findings, we estimate that syncytia covering STB^out TOs were comprised of at least 60 nuclei as well as some syncytia that were too large to be measured by patch clamping. These studies not only confirmed the presence of syncytia on the outer surface of STB^out TOs but provide a strong proof of concept for the application of this approach to quantitatively measure syncytial size on the surfaces of TOs, which could be applied to a variety of biological questions.

There are several applications that could benefit from the STB^out TO model we describe here, which results from the ability to access the STB directly and the absence of reconstituted basement membrane matrix such as Matrigel. This is particularly advantageous for studies involving infection modeling as the STB, but not CTB, is highly resistant to infection by viruses, bacteria and parasites (Megli and Coyne, 2022). Thus, proper orientation of the STB is critical to fully model the barrier to infection that the STB forms in vivo. In addition, growth of STB^out TOs also eliminates the physical barrier that matrix like Matrigel poses to diffusion of large microbes, such as bacteria and parasites, which are often unable to penetrate these substances. Similarly, given that the STB is primarily responsible for the development of passive immunity during human pregnancy due to its capacity to transport antibodies (Firan et al., 2001), modeling transplacental transport of antibodies across the STB is critical to define the specificity of these processes. Although placental explants have been used previously to placental transport (Firan et al., 2001), the limited viability of explant cultures in vitro and their genetic intractability limits mechanistic studies. Additional applications could include studies on the transplacental transport of environmental agents, channel activity on the STB surface, and identification of factors that directly influence CTB fusion and/or STB formation. Thus, the model we describe here significantly expands the experimental applications of TOs.

However, the STB^out model we describe does have limitations. We have applied the protocol described above to multiple lines of TOs derived from unique full-term placental tissue. Although we would anticipate that this protocol can be adapted to TOs derived from early gestation tissue as well as from iPSC-derived organoids, it is possible that some of the steps described would require further optimization for such models. The most significant limitation of this approach is that STB^out organoids cannot be passaged to maintain this physiological polarity phenotype long-term, and TOs with this orientation must be generated for each experiment. Thus, generation of STB^out TOs is considered a terminal culture approach, like existing EVT differentiation methods (Sheridan et al., 2020). However, given that STB^out polarity is maintained post-culturing for several days, these organoids can be utilized for studies post-generation.

The model described here provides an organoid system that recapitulates the cellular orientation of the human placenta in vivo and provides evidence that this system can be used to model key aspects of STB structure and function. In addition, given that we have developed a system for single organoid culturing in a multi-well format, this approach could be used for mid-to-high throughput screening approaches. Collectively, our method described here can be used model key aspects of placental physiology and development.

TO lines used in this study were derived from human full-term placentas as described and characterized previously (Yang et al., 2022). Human tissue used in this study was obtained through the UPMC Magee-Womens Hospital Obstetric Maternal & Infant Database and Biobank or from Duke University after approval was received from the University of Pittsburgh or Duke University Institutional Review Board (IRB) and in accordance with the guidelines of the University of Pittsburgh and Duke University human tissue procurement. Informed consent was obtained for all tissue donors and all clinical investigation have been conducted according to the principles expressed in the Declaration of Helsinki. For passaging and culturing, TOs were plated in Matrigel (Corning 356231) domes, then submerged with prewarmed complete growth medium [Advanced DMEM/F12 (Life Technologies, 12634-010) supplemented with 1× B27 (Life Technologies, 17504-044), 1× N2 (Life Technologies, 17502-048), 10% FBS (vol/vol, Cytiva HyClone, SH30070.03), 2 mM GlutaMAXTM supplement (Life Technologies, 35050-061), 100 µg/ml Primocin (InvivoGen, ant-pm-1), 1.25 mM N-acetyl-L-cysteine (Sigma, A9165), 500 nM A83-01 (Tocris, 2939), 1.5 µM CHIR99021 (Tocris, 4423), 50 ng/ml recombinant human EGF (Gibco, PHG0314), 80 ng/ml recombinant human R-spondin 1 (R&D systems, 4645-RS-100), 100 ng/ml recombinant human FGF2 (Peprotech, 100-18C), 50 ng/ml recombinant human HGF (Peprotech, 100-39), 10 mM nicotinamide (Sigma, N0636-100G), 5 µM Y-27632 (Sigma, Y0503-1MG) and 2.5 µM prostaglandin E2 (PGE2, R&D systems, 22-961-0)] as described previously (Yang et al., 2022). Cultures were maintained in a 37°C humidified incubator with 5% CO₂. Medium was renewed every 2–3 days. At ∼5–7 days after seeding, TOs were collected from Matrigel domes, digested in prewarmed TrypLE Express (Gibco, 12605-028) at 37°C for 8 min, then mechanically dissociated into small fragments using an electronic automatic pipettor and further manually pipetting, if necessary, followed by seeding into fresh Matrigel domes in 24-well tissue culture plates (Corning 3526). Propagation was performed at 1:3–1:6 splitting ratio once every 5–7 days. For the first 4 days after re-seeding, the complete growth medium was supplemented with an additional 5 µM Y-27632 (Sigma, Y0503).

To generate STB^out TOs, mature STBⁱⁿ organoids cultured as described above were first released from Matrigel domes using cell recovery solution (Corning, 354253) on ice with constant rotation at high speed (>120 rpm) for 30–60 min, pelleted (200 g for 2 min), washed one time with basal medium [Advanced DMEM/F12 (Life Technologies, 12634-010) supplemented with 2 mM GlutaMAX supplement, 10 mM HEPES (Gibco, 15630-106) and 1× penicillin-streptomycin (Lonza, 17-602E)], and then resuspended in complete growth medium supplemented with 5 µM Y-27632. Organoids were then carefully transferred using a FBS pre-coated wide orifice p200 pipette tips (Fisher Scientific, 02-707-134) into an ultra-low attachment 24-well plate (Corning, 3473). One dome containing ∼500 organoids units can be dispensed into up to five wells of a 24-well plate with <100 organoids units per well. TOs were evenly distributed in the wells prior to culturing in a 5% CO₂ 37°C incubator for suspension culture of 1–2 days. Constant orbital rotating was introduced into suspension culture to improve polarity reversal efficiency (Thermo Fisher Scientific, 88881103). Medium was renewed daily, and any aggregates dissociated using a FBS pre-coated wide orifice p200 pipette tip.

To perform single-organoid unit suspension culture, each individual mature STBⁱⁿ TO unit was picked out from domes with a sterilized needle (BD, 305125), including removal of any trace Matrigel matrix without compromising organoid integrity using a light microscope. Isolated organoids were then placed into a U-bottomed well of an ultra-low attachment 96-well spheroid microplate (Corning, 4515) for 2 days of suspension culturing as described above.

Conditioned medium (CM) was collected from original STBⁱⁿ in domes as described previously (Yang et al., 2022). To harvest CM from STB^out TOs in suspension culture, the suspension culture 24-well plate was tilted for ∼2 min to sediment organoids to one side of the well, then carefully aspirate the supernatant medium without disturbing the bottom organoids. Parallel STBⁱⁿ and STB^out TOs wells used for CM collection contained approximately same initial number of organoids for following analysis.

STBⁱⁿ TOs in Matrigel domes were released and collected as described previously (Yang et al., 2022), then total protein was extracted using RIPA buffer containing proteinase inhibitor and sonication at 10 amplitudes (QSONICA sonicators, model # Q55), after 10 min incubation on ice, lysis solution was centrifuged at high speed (>12,000 g) for 10 min, and finally the supernatant was collected as the organoids lysate. For the STB^out TOs total protein extraction, the same protocol was used except skipping the organoids releasing step. Total protein quantifications were performed using a BCA Protein assay kit (Pierce, 23227) according to the manufacturer's instructions.

STBⁱⁿ TOs were immunostained as described previously (Yang et al., 2022). For staining of STB^out TOs in suspension, the same protocol described was used, but the releasing of organoids from Matrigel was omitted. The following antibodies or reagents were used: SDC-1 (1:500; Abcam, ab128936), hCG-β (1:200; Abcam, ab243581), ITGA6 (1:500; Invitrogen, MA5-16884), HLA-G (1:200; Abcam, ab52454 and ab283260), cytokeratin-19 (1:500; Abcam, ab9221), Alexa Fluor 488 goat anti-mouse IgG secondary antibody (2 drops/ml; Invitrogen, R37120). Alexa Fluor 488 goat anti-rabbit IgG secondary antibody (2 drops/ml; Invitrogen, R37116), Alexa Fluor 488 goat anti-rat IgG secondary antibody (2 drops/ml; Invitrogen, A11006), Alexa Fluor 594 goat anti-mouse IgG secondary antibody (2 drops/ml; Invitrogen, R37121), Alexa Fluor 594 goat anti-rabbit IgG secondary antibody (2 drops/ml; Invitrogen, R37117), Alexa Fluor 633 goat anti-mouse IgG secondary antibody (1:1000; Invitrogen, A21052) and Alexa Fluor 647 goat anti-rat IgG secondary antibody (1:1000; Invitrogen, A21247), Images were captured using a Olympus Fluoview FV3000 inverted confocal microscope or a Zeiss 880 Airyscan Fast Inverted confocal microscope and contrast-adjusted in Photoshop (Adobe) or Fiji (National Institutes of Health). In some cases, colors were changed for optimal visualization using Imaris (Oxford Instruments) or Fluoview software (Olympus). Image analysis and generation of three-dimensional movies was performed using Imaris (version 9.2.1).

Organoid preparation and embedding, cryosectioning, H&E staining, and immunostaining were performed as previously described (Lancaster and Knoblich, 2014), using an H&E stain kit (Abcam, ab245880) according to the manufacturer's instructions or standard procedure for immunostaining. Briefly, the collected STBⁱⁿ and STB^out TOs were fixed (4% PFA in 1× PBS) and permeabilized (0.5% Triton X-100 in 1× PBS), then submerged in 20% sucrose solution overnight, then finally embedded into the 7.5% gelatin and 10% sucrose embedding solution and stored at −80°C. Cryosectioning of frozen blocks of organoids was performed using a cryotome (Leica, CM1950) at 10 µm thickness. For immunostaining of cryosections, frozen sections were warmed to room temperature, then immunostaining performed with the antibodies as described above. Images were captured on a Keyence BZ-X810 all-in-one fluorescence microscope and contrast-adjusted in Photoshop.

Luminex assays were performed using the following kits according to the manufacturer's instructions: hCG Human ProcartaPlex Simplex Kit (Invitrogen, EPX010-12388-901), Bio-Plex Pro Human Inflammation Panel 1 IL-28A/IFN-λ2 (Bio-Rad, 171BL022M), Bio-Plex Pro Human Inflammation Panel 1, 37-Plex (Bio-Rad, 171AL001M), and Bio-Plex Pro Human Chemokine Panel, 40-Plex (Bio-Rad, 171AK99MR2). Plates were washed using the Bio-Plex wash station (Bio-Rad, 30034376) and read on a Bio-Plex 200 system (Bio-Rad, 171000205). All samples from both polarity conditions (STBⁱⁿ and STB^out) were tested in duplicate, and each condition was performed with at least three biological replicates from unique placental tissue. All measurements were normalized to total protein of either STBⁱⁿ or STB^out TOs wells, quantified as described above.

Cytotoxicity assays were performed using the CytoTox96^® Non-Radioactive Cytotoxicity Assay kit (Promega, G1780) according to the manufacturer's instructions. All samples were tested in duplicate, and each condition (STBⁱⁿ or STB^out) was performed with at least three biological replicates.

To seed collected original STBⁱⁿ TOs onto the round glass coverslips (VWR, 76305-514) pre-coated with thin layer of Matrigel (Corning, 356231), each round coverslip was evenly distributed with ∼40 μl of Matrigel and carefully transferred into each well of regular 24-well plate to polymerize in a 37°C incubator for ∼20 min. Then, organoids were harvested as described above and evenly dispensed onto the Matrigel pre-coated surface of coverslips to settle down in a 5% CO2 37°C incubator for 3–4 h to ensure that the majority of organoids attached onto the matrix coating of the coverslip. For the STB^out TOs coat-seeding, the same protocol described above was used except omitting the release of organoids from Matrigel domes.

All results were recorded in whole-cell configurations using an Axopatch 200B amplifier (Molecular Devices) and the pClamp 10 software package (Molecular Devices). The glass pipettes were pulled from borosilicate capillaries (Sutter Instruments) and fire-polished using a microforge (Narishge) to reach a resistance of 2–3 MΩ. The pipette solution (internal) contained (in mM): 140 CsCl, 1 MgCl₂, 10 HEPES, 0.2 EGTA. pH was adjusted to 7.2 by CsOH. The bath solution contained (in mM): 140CsCl, 10HEPES, 1 MgCl₂. pH was adjusted to 7.4 by CsOH. All experiments were at room temperature (22–25°C). All the chemicals for solution preparation were obtained from Sigma-Aldrich. Once the whole-cell configuration was established, a 10-mV voltage command was delivered to the cell from a holding potential of 0 mV. The corresponding capacitive current was recorded. Membrane capacitance of the cell was calculated using Clampfit software (Molecular Devices) based on the following equation, ⁠, where C_m is the membrane capacitance, Q is the stored charge across the cell membrane, V is membrane voltage, I is current, and t is time. For the histogram plot, the bins (x-axis) were set as (pF): 0–20, 20–100, 100–200, 200–500, 500–1000, 1000–2000, 2000–5000 and 5000–10,000. The bars on the histogram were set in the middle of each bin.

All experiments reported in this study have been reproduced using a minimum of three independent organoids lines derived from unique placental tissues. All statistical analyses were performed using Clampfit (Molecular Devices), Excel (Microsoft) or Prism software (GraphPad). Data are presented as mean±s.d., unless otherwise stated. Statistical significance was determined as described in the figure legends. Parametric tests were applied when data were distributed normally based on D'Agostino–Pearson analyses; otherwise, nonparametric tests were applied. For all statistical tests, a P<0.05 was considered statistically significant, with specific P-values noted in the figure legends.

We gratefully acknowledge the Duke Light Microscopy Core Facility for their technical support and assistance for this work.