Skip to main content
NCBI home page

This is a preprint.

It has not yet been peer reviewed by a journal.

The National Library of Medicine is running a pilot to include preprints that result from research funded by NIH in PMC and PubMed.

bioRxiv logoLink to bioRxiv
[Preprint]. 2026 May 20:2026.01.26.700345. [Version 2] doi: 10.64898/2026.01.26.700345

Hyaluronan Underlies the Emergence of Form, Fate, and Function in Human Cardioids

Stefan M Jahnel 1,2, Anna Dimitriadi 1,3, Julia Kodnar 1, Vasileios Gerakopoulos 2, Yajushi Khurana 4, Maximilian Mayrhauser 1, Tobias Ilmer 1, Mohamed Amin Aguech 5, Adam R Hall 5,6, Keisuke Ishihara 4,*, Sasha Mendjan 1,2,*
PMCID: PMC13228570  PMID: 42239165

Abstract

The extracellular matrix (ECM) is crucial for organ development and disease. Yet, the interplay among cells, function, and the ECM during human cardiogenesis remains obscure. Using human cardioids, we discovered that cardiac mesoderm-synthesized hyaluronan (HA) underlies early cardiac functional development. HA drives cardioid cavity formation through hydrogel swelling and bioscaffolding, critical functions of the cardiac jelly in the early vertebrate heart. During an early developmental window, HA is essential for establishing cardiac cell identity, while at later stages, HA-generated forces promote beating function through mechanosensitive channels. Chamber-specific differences in mechanical sensitivity ensure robust contractions in multi-chambered tissues. Our findings reveal how a single endogenous ECM component orchestrates the co-emergence of form, fate, and function during human organogenesis, opening new avenues for bioengineering physiologically relevant organ models.

INTRODUCTION

Organogenesis unfolds through dynamic interactions between cells and their physicochemical environment (1). Progenitors produce the extracellular matrix (ECM), which, in turn, provides chemical and mechanical cues that direct cellular behavior (2–4). This cell-ECM feedback is a fundamental principle that underlies the emergence of functional organs. The first organ to form and function in the human embryo is the heart (5–8), which starts to beat around embryonic day 20 (9). The ECM is crucial for the functional development of the heart (10, 11), providing both biochemical signals that regulate the behavior of diverse cardiac cell types and the mechanical scaffolding that supports the hearťs rhythmic contraction. While numerous ECM components have been identified in cardiac development and disease (12, 13), the precise functional dissection of cell-ECM interactions orchestrating human cardiogenesis remains challenging.

The inaccessibility of the developing human embryo has long prevented direct mechanistic studies of human cardiogenesis. Recent advances in three-dimensional models of the human heart derived from human pluripotent stem cells (hPSCs) provide a powerful experimental platform, recapitulating self-organized chamber formation, regional identity specification, and contractile function in vitro (14–17). However, these systems have not been used to study the role of endogenous cardiac ECM, despite its central role in development and disease. Here, we leverage chamber-like human cardiac organoids (14, 15), or cardioids, to dissect how form, fate, and function emerge during early heart development. We discover that cell-synthesized hyaluronan (hyaluronic acid, HA) drives chamber morphogenesis, promotes cardiomyocyte differentiation, and establishes a mechanical microenvironment that supports the earliest heartbeats. Our findings reveal how HA, a key ECM component, coordinates the co-emergence of structure, identity, and function during human organ development.

RESULTS

Cardioids generate cavities that contain hyaluronan

Fluid-filled cavities are the hallmark of the chambered architecture of cardioids. While our previous studies showed cavities emerge by d2.5 (14, 15) (Fig. 1A), bright field microscopy (Fig. 1B) and cryo-sections provide limited understanding of 3D tissue structure. Therefore, we focused on left ventricle (LV) cardioids that generate the largest cavities of all cardioids (15) and performed whole-mount 3D imaging and volumetric analysis (Methods, Fig. 1C). We found that cardioids undergo a pancake-to-spherical shape transition from d1.5 and the subsequent 12 hours. From d2.5 onwards, cardioids adopt spherical morphology with multiple cavities. The total volume of cardioids expands rapidly between d2.0 and d4.5, subsequently plateauing at d5.5 (Fig. 1C’), with cavities comprising 30–40% of total volume between d2.5 and d5.5 (fig. S1A). The number of cavities increased over time (Fig. 1C’). To monitor the dynamics of cavity formation, we performed live volumetric imaging on CAG-GFP cardioids and observed that multiple cavities emerge, grow, and coalesce within cardioids (Fig. 1D, 1D’, fig. S1B, Movie S1, S2, S3). Scanning electron microscopy revealed a loosely connected cell arrangement (Fig. 1E), suggesting that the tissue is porous. To test the permeability of molecules in cardioids, we added fluorescently labelled Dextran to the culture media. Within 10 min, 150 kDa Dextran rapidly permeated the cavities (Fig. 1F). Thus, unlike mammalian blastocysts (18, 19) or many other organoids (20, 21), cardioid cavities rapidly expand in the absence of a tightly sealed tissue.

Figure 1. Cardioids generate cavities that contain hyaluronic acid.

Figure 1.

Open in a new tab

(A) Schematic of cardioid development and time course of cavity formation.

(B) Brightfield images of left ventricular cardioids at day 2.5 containing cavities. Examples are shown for cardioids generated from three cell lines.

(C) Top: Confocal images of cardioids labeled with AlexaFluor 647. Bottom: Surface renderings of the outer cardioid boundary (gray) and cavities (blue) of the same samples. Scale bar 400 micron. (C’) Top: Volume of cardioid and cavity at each time point. Bottom: Number of cavities per cardioid at each time point. Error bars indicate the standard deviation across n=4–12 samples per time point.

(D) Surface renderings of selected time points from the live microscopy data (Movie S2, S3), where 0 hr was defined as the beginning of the imaging. (D’) Volume quantification of cardioid and cavity of the live microscopy date. Error bars indicate the standard deviation across n=4 samples.

(E) Scanning electron microscope images show the porous nature of cardioids. Scale bar 100 micron.

(F) Confocal image of a cardioid (d4.5) incubated with 150 kDa FITC-Dextran overnight.

(G) Schematic of the human heart tube at day 22. Similar to the heart tube of other vertebrates, the cardiac jelly fills the space between the myocardium and endocardium.

(H) Cryo-section images of cardioids showing the localization of hyaluronic acid binding protein (HABP). Scale bar 100 microns.

How can such a porous tissue increase the volume in its acellular space? We considered the developmental mechanism employed by the vertebrate heart tube, where the large extracellular space between the endocardium and myocardium is occupied by a gelatinous substance termed the ‘cardiac jelly’ (22, 23), which largely consists of hyaluronic acid (HA) produced by cardiac mesoderm and cardiomyocyte progenitors (24–26) (Fig. 1G). The swelling of the HA hydrogel leads to the rapid volumetric expansion of the heart tube. To investigate whether a similar mechanism may operate in cardioids, we performed histological staining and found that HA is present throughout the cavities (Fig. 1H). To independently confirm HA identity and characterize its molecular weight, we performed single-molecule solid-state nanopore analysis on HA extracted from day 6.5 LV cardioids, revealing a mean molecular weight of 1,243 kDa (fig. S1C).

Hyaluronan synthesis by HAS2 drives cavity formation in cardioids

To test if HA plays a role in shaping cardioids, we enzymatically degraded HA using hyaluronidase (HAase). When d2.5 left ventricular cardioids were treated with HAase, they began to shrink (Fig. 2A), with diameters decreasing by 22 % after 10 hours (Fig. 2A’). Sections of cardioids revealed complete collapse of cardioid cavities following HAase treatment. (Fig. 2B, Movie S4). HAase treatment at d3.5, 4.5, and 5.5 showed a reduced yet consistent decrease in cardioid size (Fig. 2A’, Movie S5). These results suggest HA exerts an outward force that counteracts cell contractility to sustain cavity volume.

Figure 2. Hyaluronic acid synthesis by HAS2 drives cavity formation in cardioids.

Figure 2.

Open in a new tab

(A) Brightfield image sequences show that hyaluronidase (HAase) treatment results in rapid shrinkage of day 2.5 cardioids and their cavities, while untreated cardioids continue growth. Scale bar 500 microns (A’) Size reduction with HAase treatment is quantified for cardioids at day 2.5 to 5.5. n=8 samples per condition. Shaded regions indicate standard deviation.

(B) Cryo-section images of untreated and HAase-treated (treatment d1.5–2.5) day 2.5 cardioids. Scale bar 250 microns.

(C) Heatmap visualization of the transcripts detected for HAS1, HAS2, and HAS3 genes during cardioid development. VST, variant stabilized transcript counts.

(D) Brightfield images of wild type (WT) and HAS2 KO cardioids at d5.5. Scale bar 500 microns. (D’) Diameter of cardioids derived from WT and HAS2 KO genotypes. Colors distinguish samples prepared from n=2 WT cell lines and n=3 HAS KO cell lines. n=15–16 samples per cell line per timepoint. Shaded regions indicate standard deviation.

(E) Representative cryo-section images of WT and HAS2 KO cardioids showing absence of cavities in HAS2 KO. Scale bar 500 microns.

(F) Brightfield images of WT and HAS2 KO; TetOn-GFP-HAS2 cardioids at day 5.5. Doxycycline was supplemented to the medium from day 1.5 to 5.5 for the indicated condition. Scale bar 500 microns. (F’) Diameter of cardioids derived from WT and HAS2 KO; TetOn-GFP-HAS2 genotypes. n=11–12 samples per cell line per timepoint for day 2.5 and 3.5. n=5 samples per cell line per timepoint for day 5.5. Shaded regions indicate standard deviation.

While the human genome encodes three members of the hyaluronic acid synthase family of enzymes, HAS2 is the only known isoform implicated in vertebrate heart development (25–27). Through transcriptomic profiling, we determined that HAS2 is the only synthase that is expressed during cardioid differentiation (Fig. 2C). To test if HAS2 is responsible for HA production, we established three independent HAS2 KO lines and generated cardioids (fig. S2A, B). HAS2 KO cardioids remained small in size relative to their WT counterparts throughout differentiation and did not exhibit readily observable cavities in brightfield images (Fig. 2D, Fig. 2D’, fig. S2C). Histological sections confirmed the absence of cavities in HAS2 KO cardioids (Fig. 2E). To test whether the morphology of HAS2 KO cardioids can be rescued genetically, we engineered a chemically inducible system for GFP-HAS2 expression in the HAS2 KO background (HAS2KO; TetOn-GFP-HAS2). Upon addition of doxycycline from d1.5, we observed GFP-HAS2 expression (fig. S2D) and an increase in cardioid size from d2.5 to d5.5 (Fig. 2F, Fig. 2F’). In summary, we identified a HA hydrogel, produced by HAS2, as the main driver of cavity formation in cardioids.

Hyaluronan is necessary for cardiac mesoderm specification

Animal studies have shown that HAS2 is essential for normal heart development and embryonic viability (25). Cardioid development is characterized by a transition from nascent mesoderm through cardiac mesoderm into functional cardiomyocytes (14) (Fig. 3A). To investigate the role of HA during LV cardioid development, we harvested HAS2-KO cardioids at different timepoints (d1.5, d2.5, d4.5, d7.5) and performed bulk RNA sequencing. Up to d1.5 HAS2-KO had no effect on the transcriptome. From d2.5 onward, the transcriptome of HAS2 KO samples diverged sharply from that of WT controls (Fig. 3B). To test if enzymatic degradation of HA has a similar effect to genetic deletion of HAS2, we started HAase treatment from d1.5 and followed with transcriptomic profiling (Fig. 3C). At the transcriptomic level, HAase-treated cardioids largely phenocopied stage-matched HAS2 KO cardioids at d2.5 (Fig. 3D). Both KO and HAase-treated cardioids downregulated pathways associated with cardiac muscle development and function attributed to a misregulation of typical cardiac regulatory (e.g. TBX5, NKX2–5, GATA4) and structural (e.g. TNNT2, TPM1, ACTC1, MYL7, MYL4, MYH9) genes (Fig. 3D, Fig. 3E, fig. S3A, fig. S3B, fig. S3C, Table S1, Table S2). On the other hand, KO and HAase-treated cardioids showed prolonged expression of genes linked to early mesoderm development (e.g. MESP1, MESP2, MIXL1, CER1) and a bias towards a paraxial/intermediate mesodermal (e.g. TCF15, DLL3, LHX1, FOXC1) (Fig. 3D, Fig. 3E) or non-mesodermal fate (fig. S3A). At d4.5 and d7.5, KO and HAase-treated cardioids continued to show the downregulation of cardiac/muscle-related pathways, while suggesting misspecification into non-cardiac cell fates (Fig. 3F, fig. S3D, fig. S3E, fig. S3F, fig. S3G, fig. S3H, fig. S3I). At these later stages, KO and HAase-treated cardioids downregulated genes that respond to muscle stretch (e.g. ANKRD1, NPPA, NPPB, MYH7, CSRP3) (Fig. 3G). At d7.5, both KO and HAase-treated samples showed aberrant regulation of genes involved in cell cycle and metabolic processes (fig. S3G, fig. S3H, fig. S3I). Overall our genetic and enzymatic perturbations in cardioids demonstrate that HA is required for a timely transition of nascent mesoderm into pre-cardiac mesoderm and cardiomyocytes.

Figure 3. HAS2 and hyaluronan are necessary for cardiac mesoderm specification.

Figure 3.

Open in a new tab

(A) Timeline of cardioid fate specification and expression of exemplary regulatory and structural genes.

(B) Number of differentially expressed genes of WT vs. HAS2 KO cardioids at different readout timepoints.

(C) Number of differentially expressed genes of untreated vs. HAase-treated cardioids (treatment either started at d1.5, or d2.5) at different readout timepoints.

(D) Enrichment scores of selected pathways differentially regulated between WT, HAS2 KO and HAase-treated cardioids at d2.5. See fig. S3A for complete list.

(E) Expression of selected genes differentially regulated between WT, HAS2 KO and HAase-treated cardioids at d2.5.

(F) Enrichment scores of selected pathways differentially regulated between WT, HAS2 KO and HAase-treated cardioids at d4.5. See fig. S3D for complete list.

(G) Expression of selected genes differentially regulated between WT, HAS2 KO and HAase-treated cardioids at d4.5.

(H) Images showing brightfield and TNNI-GFP signal of untreated vs. HAase-treated cardioids at d7.5. Scale bar 1000 microns.

(I) Immunostaining of cryosections of untreated vs. HAase-treated cardioids at d7.5 showing absence of cavities in treated cardioids. Note lower differentiation efficiency in samples treated from d1.5 (as compared to samples treated from d2.5). Scale bar 500 microns. Color bars in (D) and (F): GSVA enrichment score. Colors bars in (E) and (G): Scaled gene expression.

To define the precise temporal requirement of HA in cardioid development, we compared HAase treatment initiated at d1.5 versus d2.5 (Fig. 3C). Direct comparison of these two conditions at d4.5 shows that the aforementioned effect is weakened in cardioids treated only from d2.5 on (Fig. 3F, Fig. 3G, fig. S3D, fig. S3E, fig. S3F, fig. S3G, fig. S3H, fig. S3I), suggesting that the presence of HA from d1.5 is critical for instructing cardioid development. To complement our transcriptomic experiments, we studied the expression of cardiac proteins in cardioids treated with HAase. HAase treatment from d1.5 largely reduced TNNI1-GFP expression at d7.5, while delaying the treatment to d2.5 reversed this effect in similarly size-reduced cardioids (Fig. 3H). While cells co-express NKX2–5 and cTNT by d7.5 in untreated cardioids, we found many NKX2–5-positive cells that failed to co-express cTNT in cardioids treated with HAase from d1.5 (Fig. 3I). Taken together, we identify d1.5 to d2.5 as the critical time window in cardioid development during which the presence of HA hydrogel is required for the proper specification of the cardiac mesoderm and subsequent formation of cardiomyocytes.

Loss of hyaluronan compromises the beating function of left ventricular cardioids

After finding that HA underlies morphogenesis and cell fate specification, we asked whether HA plays a role in cardioid physiology. The earliest contractions in LV cardioids begin around d5, with all samples showing consistent beatings by d5.5(14) (Fig. 4A, 4B). At d5.5, cardioids generated from HAS2 KO cell lines showed an overall reduced beating frequency compared to WT cardioids (KO3 showed no contractions) (Fig. 4B’). The difference in beating frequency for WT vs KO cell lines became statistically significant at day 6.5 and 7.5 (fig. S4A). Starting from day 5.5, contractile HAS2 KO cardioids (KO1, KO2) showed a significantly reduced contraction amplitude, which can be described as a ‘twitching’ behavior as opposed to the large amplitude beating of WT cardioids (Fig. 4B, Fig. 4B’’, fig. S4B, Movie S6). Overall, reductions in beating frequency and contraction amplitude functionally distinguishes HAS2 KO from WT cardioids.

Figure 4. Hyaluronan provides a physical environment that is necessary and sufficient for the beating function of LV cardioids.

Figure 4.

Open in a new tab

(A) Schematic of LV cardioid development highlighting the onset of contractile activity.

(B) Contraction traces of day 5.5 WT and HAS2 KO cardioids. (B’) Beating frequency and (B’’) Contraction Amplitude of day 5.5 WT and HAS2 KO cardioids. Colors distinguish samples prepared from n=2 WT cell lines and n=3 HAS KO cell lines. n = 16 samples per cell line per condition.

(C) Contraction traces of day 5.5 cardioids treated with HAase for different time windows. (C’) Beating frequency and (C’) Contraction Amplitude of day 5.5 cardioids treated with HAase for different time windows. Colors distinguish samples prepared from two cell lines. n=16 samples per cell line per condition.

(D) Schematic of experiment to test the effect of acute hyaluronic acid degradation and mechanical rescue strategies on the beating function of day 5.5 cardioids.

(E) Bright field image (top) and contraction traces (bottom) of a day 5.5 cardioid subjected to HAase treatment, followed by oil injection. Scale bar 500 microns. (E’) Top: Beating frequency of cardioids subjected to HAase treatment, followed by oil injection. Bottom: Beating frequency of cardioids subjected to HAase treatment, followed by mock injection. Colors distinguish samples prepared from three cell lines. n=5–8 samples per cell line per condition.

(F) Bright field image (top) and contraction traces (bottom) of a day 5.5 cardioid subjected to HAase treatment, followed by Yoda1 treatment. Scale bar 500 microns. (F’) Beating frequency of cardioids treated with HAase and Yoda1. (F’’) Classification of arrhythmia phenotypes. n=15 samples from two cell lines are combined. See also fig. S5D. Percentages indicate the fraction of beating samples in each condition. Asterisks indicate a statistically significant difference of p-value < 0.05.

To define the temporal HA requirement for contractile function, we treated cardioids with HAase for different time windows and assessed contraction at d5.5 and 6.5 (fig. S4C). When the treatment started at d2.5, beating frequency of the treated group was comparable to the control group at d5.5 (Fig. 4C’), but most cardioids stopped beating by d6.5 (fig. S4D). Contraction amplitude was significantly reduced relative to untreated samples (Fig. 4C’’, fig. S4E, Movie S7), similar to the ‘twitching’ phenotype observed in HAS2 KO cardioids. When HAase treatment started at d1.5, this completely abolished contractions in cardioids (Fig. 4C, Fig. 4C’, fig. S4D, Movie S7). In summary, our functional experiments revealed that HA is particularly important during the d1.5-d2.5 developmental window, when it contributes to the beating frequency and contraction amplitude to ensure proper LV cardioid function at later stages.

Hyaluronan provides a physical environment that is necessary and sufficient for the beating function of left ventricular cardioids

One plausible explanation of impaired beating of LV cardioids devoid of HA from d1.5 is that the cells fail to differentiate into functional cardiomyocytes (see Fig. 3). An additional possibility is that from d5.5, the physical properties of HA hydrogel contribute to the normal beating of LV cardioids. To directly test this scenario, we treated beating LV cardioids at d5.5 with HAase and immediately assessed beating (Fig. 4D). At 1 hour post HAase treatment, cardioids were reduced in size (Fig. 4E, fig. S4F) and showed a near complete loss of contractility (Fig. 4E, Fig. 4E’ at t1, Movie S8), indicating that acute HA degradation impairs cardioid beating. To ask if this beating defect can be physically rescued in the absence of HA, we manipulated tissue morphology and mechanics by injecting an inert oil droplet into the collapsed cavities of HAase-treated cardioids (Fig. 4D). Oil injections successfully increased cardioid size with diameters approaching values of untreated samples (Fig. 4E, fig. S4F at t2). Strikingly, oil-injected cardioids restored their beating, while mock-injected remained largely inactive (Fig. 4E, Fig. 4E’ at t2, Movie S8). Taken together, our results demonstrate that HA provides a physical environment that is essential for the normal beating of LV cardioids. However, these experiments did not address whether this physical environment is defined by tissue morphology or mechanics.

Mechanosensitive channels link hyaluronan-generated forces to left ventricular cardioid beating

Cardiomyocytes are mechanosensitive cells that detect and respond to physical forces through multiple molecular mechanisms (28). Key among these are mechanosensitive channels (MSCs) that translate tissue-scale forces to cellular electrophysiological responses (29). To test whether MSCs underlie cardioid contractility, we used the MSC inhibitor GsMTx-4, which broadly targets the Piezo and TRP families of mechanosensitive channels (30, 31). Treating d5.5 LV cardioids with 5.0 μM GsMTx-4 abolished contractions (fig. S4G, fig. S4H, Movie S9), but had no effect on tissue size (fig. S4G, fig. S4I). At a lower dose of 2.5 μM GsMTx-4, we observed that cardioids from one WT cell line continued to beat (fig. S4H, WT3), yet, these contractions were predominantly arrhythmic (fig. S4J, fig. S4K, fig. S4L, Movie S9). To ask if activation of a specific MSC results in cardioid contractions, we turned to the Piezo1 activator Yoda1 (32, 33) (Fig. 4D). Treating LV cardioids with Yoda1 increased the beating frequency (fig. S4M, Movie S10). Strikingly, Yoda1 treatment completely rescued the HAase-induced beating defects, including arrhythmia (Fig. 4F, Fig. 4F’, Fig. 4F’’, fig. S4N, fig. S4O, Movie S10), despite the Yoda1-treated cardioids remaining small in size (Fig. 4F, fig. S4P). Taken together our pharmacological manipulations suggest that it is not the HA-induced morphology per se but the mechanical forces generated by HA that act through MSCs and prime LV cardioids for rhythmical contractions.

Multi-chamber architecture provides resilience to mechanical perturbation in vivo and in vitro

Finally, we asked whether the mechanical role of HA in early cardiac contractions is a general phenomenon beyond LV cardioids. To test this, we chose chick embryos as an accessible source for whole functional hearts. HAase injection into cultured chick embryos at HH8 increased the volume of the heart lumen at the expense of cardiac jelly as reported previously (34) (fig. S5A). In some cases, the volume of the heart tube expanded (26, 35) (fig. S5B), potentially caused by systemic blood pressure. Therefore, to study the beating function of whole hearts in mechanical isolation from the embryo, we explanted hearts from HH11–12 chick embryos and cultured them ex vivo. HAase treatment readily reduced the size of heart explants, while the size of control samples remained unchanged (Fig. 5A, Fig. 5A’, fig. S5C). However, the beating frequency was not significantly affected with or without HAase treatment (Fig. 5A’’, fig. S5D, Movie S11). This indicates that the chick embryonic hearts, like those of zebrafish(36), are functionally resilient to the acute degradation of HA.

Figure 5. Multi-chamber architecture provides resilience to mechanical perturbation in vivo and in vitro.

Figure 5.

Open in a new tab

(A) Experiment to test the effect of acute HA degradation on chick heart explants. Each row of the image panel shows the same heart explant before and after hyaluronidase treatment. t1 recording was done 4.3 hours after addition of HAase. Scale bar 500 microns. (A’) Area and (A’’) Beating frequency of heart explants before and after HAase. n=9 samples.

(B) Experiment to test the effect of acute HA degradation on day 7.5 atrial cardioids. Brightfield images and contraction traces of the same cardioid before and after HAase treatment. t1 recording was done 1 hour after addition of HAase. Scale bar 500 microns. (B’) Diameter and (B’’) Beating frequency of atrial cardioids before and after HAase treatment. Colors distinguish samples prepared from three cell lines. n=13–16 samples per cell line per condition.

(C) Schematic of experiment to assay the role of acute hyaluronic acid degradation on multi-chamber cardioids. LV-A multi-chamber cardioids are generated by fusing an atrial cardioid (labelled with a dye) and a left ventricular cardioid at day 3.5, allowed to develop, and assayed at day 7.5. t1 recording was done 1 hour after addition of HAase.

(D) Brightfield and fluorescent images of the same day 7.5 LV-A multi-chamber cardioid before and after HAase treatment. TNNI-GFP is expressed in both chambers, while only the atrial chamber is labelled with the dye DilC18(5)-DS. Contraction traces are shown for the corresponding timepoints. Scale bar 500 microns. (D’) Beating frequency of LV-A multichambered cardioids before and after HAase treatment. Colors distinguish samples prepared from three cell lines. n=14–16 samples per cell line per condition.

(E) Ca2+ signal propagation in GCaMP6f-expressing LV-A cardioids after HAase treatment. Scale bar 500 microns.

(F) Left: Heatmap indicates the time when Ca2+ signal first appeared. Right: DilC18(5)-DS signal indicates the region corresponding to the atrial chamber. Scale bar 500 microns. Percentages indicate the fraction of beating samples in each condition. Asterisks indicate a statistically significant difference of p-value < 0.05.

Unlike LV cardioids, the chick heart contains multiple heart regions, including the primitive atrium, which acts as the pacemaker region prior to the development of the sinoatrial node (37, 38). To dissect the region-specific functions of the human heart, we generated atrial cardioids (A) and treated them with HAase. We performed our experiments at d7.5, a timepoint when all A cardioids show robust beating. HAase treatment reduced cardioid size, suggesting that HA supports cavity morphogenesis of A cardioids (Fig. 5B, Fig. 5B’) similar to LV cardioids. However, the beating frequency of A cardioids decreased only slightly (Fig. 5B’’). Thus, in contrast to LV cardioids, A cardioids are functionally resilient to mechanical unloading.

Our findings in chick explants and A cardioids led us to ask if a multichamber architecture is sufficient to confer functional resilience to mechanical unloading. To this end we fused LV and A cardioids (15) at d3.5 and treated them with HAase at d7.5 (Fig. 5C). Untreated LV-A multichamber cardioids showed a beating frequency of 8.7±2.6 beats/30s, which is between the frequencies of isolated A (14.7±1.6 beats/30s) and LV (1.6±0.9 beats/30s) cardioids at this day (Fig. 5D, Fig. 5D’). HAase treatment reduced the area of LV-A multichamber cardioids by 27% (fig. S5E) and led to a decrease in beating frequency (Fig. 5D’ at t1). However, unlike LV cardioids, the majority of LV-A multichamber cardioids continued beating after HAase treatment. Brightfield movies confirmed contraction in both A and LV regions (Movie S12). To better understand the spatial and mechanistic basis of contractions in these conditions, we performed optical mapping of Ca2+-transients using cardioids expressing GCaMP6f (39). Upon HAase treatment, LV cardioids lost their Ca2+-transients, while they were maintained in A cardioids, consistent with their contractilities in the respective conditions (Movie S13). In multichamber cardioids, Ca2+-transients initiated in the atria and propagated to the ventricular portion in all cases for untreated and HAase-treated conditions, respectively (Fig. 5E, fig. S5F), indicating that atrial pacing can rescue the beating defect of LV cardioids. Altogether, we show that cardioid sub-compartments exhibit distinct mechanosensitivity and that the multichamber architecture extends functional resilience across distinct heart chambers.

DISCUSSION

Organ shape and function emerge from the dynamic interactions between differentiating stem cells and microenvironmental cues. In this study, we found that human cardioids endogenously produce a hyaluronic acid (HA) hydrogel that plays a fundamental role in the development of cardiac form, fate, and function. Within two days of cardioid development, progenitor cells begin to express the hyaluronic acid synthase, HAS2, which is responsible for producing the HA hydrogel that swells and drives cavity formation (Fig. 1, Fig. 2). HA-enriched cavities occupy up to 40% of the total volume of cardioids, contributing to rapid volumetric growth. In contrast to the widespread mode of lumen formation that involves active ion transport in polarized cellular sheets (18–21), cavity formation by hydrogel swelling creates an acellular space, accessible to large, freely diffusing molecules in a porous tissue. The HA hydrogel in cardioids is reminiscent of the cardiac jelly in the early vertebrate heart. The HA secreted by the cardiac mesoderm and cardiomyocyte progenitors via HAS2 is the major component of the cardiac jelly, driving the expansion of the myoendocardial space (23–26, 34). Thus, using cardioids, we uncover and dissect a general biophysical mechanism that rapidly generates acellular space in developing tissues (40, 41), providing the first mechanistic evidence for the role of HA in human cardiac morphogenesis.

Beyond its role in sculpting cardioid chambers, we found that HA is crucial for hierarchical specification of cardiac fate from nascent mesoderm (Fig. 3). Genetic and enzymatic degradation of HA led to aberrant expression of presomitic mesoderm genes and attenuated cardiac lineage commitment in cardioids. We established that HA mainly acts during a narrow developmental window (d1.5–2.5), and propose three mechanistic scenarios for the role of HA in cell fate regulation. First, an aberrant metabolism in HAS2 knockout cells (42, 43) and conditions with high hyaluronidase activity (44), resulting from disrupted UDP-sugar utilization or altered glycolytic flux, may instruct the cell cycle and progenitor lineage decisions (45, 46). Second, HA receptor-mediated signaling through receptors (e.g. CD44 (47)) may activate downstream transcriptional networks for cardiac differentiation (25). Third, mechanical forces generated by the HA hydrogel may directly regulate gene expression programs required for cardiac specification (48). Future studies using cardioids are well-positioned to dissect these scenarios to understand cell fate decisions at the onset of human cardiogenesis.

While the effects of ECM components on cardiac differentiation and cell fate had been studied (10, 11), cardioids uniquely enabled the investigation of early functional consequences, revealing a purely mechanical role of HA for cardiac contraction in a human model (Fig. 4). Based on our acute perturbation and rescue experiments, we propose that the swelling forces generated by HA hydrogel establish a permissive mechanical condition. We speculate that HA stretches the cardiomyocytes, mimicking the state of the heart under preload, supporting robust contractions in left ventricular cardioids. Our findings extend the well-established mechanosensitive nature of cardiomyocytes (49, 50) by demonstrating that these cells actively synthesize ECM components to engineer their own mechanical microenvironment for optimal contraction. We identified the mechanosensitive ion channel Piezo1 as a crucial mediator of ECM-derived swelling force and cellular response. Future studies will leverage the experimental accessibility of cardioids to address how ECM-derived mechanical signals are transmitted through changes in membrane potential and Ca2+ signaling to ultimately regulate the earliest contractions of the heart (51).

We generated cardioids corresponding to different heart regions and discovered chamber-specific mechanosensitivity, achieving a level of functional dissection previously unattainable (Fig. 5). Strikingly, in contrast to left ventricular cardioids, the contraction of atrial cardioids showed resilience to HA removal. When integrated into multi-chambered cardioids, the atrial portion drives the contractions in the left ventricular cardioid even after hyaluronidase treatment. The resilience of multi-chamber cardioids mirrors the mechanical robustness of whole heart explants. In the human embryo, the first heartbeat occurs around 20 days post-fertilization and is thought to originate from the region corresponding to the future left ventricle. As the atrial region develops, it assumes the role of pacemaker with increasingly consistent tempo. Left ventricular cardioids develop contractility two days earlier than atrial cardioids (15), but whether regional differences in ECM production (52) contribute to this accelerated developmental timing and pacemaking remains unexplored. We speculate that the complementary chamber-specific characteristics in ECM profile, mechanosensitivity, and developmental timing ensure a robust and timely emergence of organ-scale physiology in multi-chambered hearts.

Unlike exogenous matrices that dominate current organoid culture and tissue engineering approaches (53, 54), endogenous cell-synthesized ECM provides a dynamic, physiologically relevant environment for coordinating form, cell fate, and tissue function. Our findings indicate that cardiac tissue generation from human PSCs should incorporate mechanical cues as early as day 1.5 of differentiation. This contrasts with most current methods for generating human cardiomyocytes (55), which employ 2D differentiation conditions for multiple weeks before replating, scaffolding, or bioprinting for functional analysis. Our left ventricular cardioid protocol achieves high cardiomyocyte differentiation efficiency (14) and robust beating by day 5.5, representing one of the fastest timelines from hPSCs to contractile tissue. We attribute this efficiency to early 3D self-organization of cardiac progenitor cells that capture endogenous cell-secreted HA for efficient cardiac lineage commitment. Furthermore, the accumulated HA establishes a favorable physical environment for robust contraction. Consistent with this principle, other studies differentiating human PSCs at high density similarly report chamber formation and accelerated contractility (16, 17, 56). These findings collectively suggest that recapitulating the mechanical microenvironment during early organogenesis may be a generalizable strategy to accelerate the generation of more functional and physiologically relevant tissues. More broadly, by understanding and rewiring cellular programs for dynamic ECM synthesis, we can better control form, fate, and function to engineer advanced organoids.

Limitations and Future Directions

While cardioids offer a powerful modular platform for studying cardiac biology, their simplicity and our current understanding impose important limitations. As cardioids do not incorporate mechanical input from the foregut endoderm, this model may mimic heart formation in cardia bifida embryos, where foregut constriction is abolished (Fig. S5G, Fig. S5H). Moreover, we cannot currently direct HA generation between myocardial and endocardial tissue layers to reproduce the tube-within-tube architecture of the developing heart. Thus, we still lack directional flow or associated mechanical cues like wall shear forces as well as direct measurements of hydrogel rheology and tissue-scale forces. New technologies for in situ mechanical actuation and measurements may overcome this gap. Finally, the physicochemical profile of the ECM in cardioids remains unknown. Characterizing the molecular weight of HA and the presence of other ECM components is essential for understanding how this ECM is dynamically remodeled and sensed by cells. Such insights may enable dissection of later developmental events like trabeculation and valvulogenesis in this tractable model of the human heart.

Supplementary Material

Supplement 1
Download video file (1.7MB, mp4)
Supplement 2
Download video file (2MB, mp4)
Supplement 3
Download video file (1.1MB, mp4)
Supplement 4
Download video file (642.2KB, mp4)
Supplement 5
Download video file (3.8MB, mp4)
Supplement 6
Download video file (710.3KB, mp4)
Supplement 7
Download video file (370.1KB, mp4)
Supplement 8
Download video file (556.7KB, mp4)
Supplement 9
Download video file (1.2MB, mp4)
Supplement 10
Download video file (544.1KB, mp4)
Supplement 11
Download video file (7.2MB, mp4)
Supplement 12
Download video file (245.5KB, mp4)
Supplement 13
Download video file (3.4MB, mp4)
Supplement 14
media-14.xlsx (49.6MB, xlsx)
Supplement 15
media-15.xlsx (54.1KB, xlsx)
Supplement 16

LIST OF SUPPLEMENTARY MATERIALS:

Supplementary Materials

Materials and Methods

Figs. S1 to S5

Tables S1 and S2

Movies S1-S13

ACKNOWLEDGEMENTS:

We thank all members of the Mendjan and Ishihara labs for their help and discussions. We thank Katarzyna Warczok and Megan Shi for lab management. We are grateful for IMBA/IMP/Vienna Bio Center core facilities (Bioinformatics (Maria Novatchkova), BioOptics, Molecular Biology Service, Electron Microscopy, Histology, Next Generation Sequencing, and Stem Cell Core facilities) for their services. We thank the Allen Institute and HeartBeat.bio for cell lines.

Funding:

This work was supported by the Austrian Academy of Sciences (OEAW), an ERC Advanced Grant (CardioGROWTH, 101201790) to S.M.; an Austrian Science Fund (FWF) grant (ID: PAT4435223) to S.M.; an American Heart Association award (25CDA1432232) to K.I.; and a NIH grant (R01GM134226) to A.R.H.

Footnotes

Competing interests: S.M. is a co-founder and scientific advisory board member of Heartbeat Bio, a IMBA spin-off; IMBA filed patent applications with S.M (EP4540370A1) and S.M.J. and S.M. as co-inventors (EP4483888A3).

Data, code and materials availability:

RNA-seq data has been deposited to the NCBI Gene Expression Omnibus and is accessible through the GEO accession number GSE317879. 3D imaging data will be deposited in the BioImage Archive. Custom analysis code is available from the corresponding authors upon reasonable request. Cell lines and plasmids are available upon reasonable request from the corresponding authors and may require a material transfer agreement.

REFERENCES AND NOTES:

  • 1.Barresi M. J. F., Gilbert S. F., Developmental Biology (Oxford university press, New York, International 13th ed., 2023). [Google Scholar]
  • 2.Walma D. A. C., Yamada K. M., The extracellular matrix in development. Development 147, dev175596 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 3.Mao Y., Wickström S. A., Mechanical state transitions in the regulation of tissue form and function. Nat. Rev. Mol. Cell Biol. 25, 654–670 (2024). [DOI] [PubMed] [Google Scholar]
  • 4.Andrews T. G. R., Priya R., The Mechanics of Building Functional Organs. Cold Spring Harb. Perspect. Biol. 17, a041520 (2025). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5.Kirby M. L., Cardiac Development (Oxford University Press, 2007). [Google Scholar]
  • 6.Sizarov A., Ya J., de Boer B. A., Lamers W. H., Christoffels V. M., Moorman A. F. M., Formation of the Building Plan of the Human Heart. Circulation 123, 1125–1135 (2011). [DOI] [PubMed] [Google Scholar]
  • 7.Meilhac S. M., Buckingham M. E., The deployment of cell lineages that form the mammalian heart. Nat. Rev. Cardiol. 15, 705–724 (2018). [DOI] [PubMed] [Google Scholar]
  • 8.Mendjan S., Deyett A., Yelon D., Coordination of cardiogenesis in vivo and in vitro. Nat. Rev. Mol. Cell Biol. 27, 19–34 (2026). [DOI] [PubMed] [Google Scholar]
  • 9.Männer J., When Does the Human Embryonic Heart Start Beating? A Review of Contemporary and Historical Sources of Knowledge about the Onset of Blood Circulation in Man. J. Cardiovasc. Dev. Dis. 9, 187 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10.Derrick C. J., Noël E. S., The ECM as a driver of heart development and repair. Development 148, dev191320 (2021). [DOI] [PubMed] [Google Scholar]
  • 11.Lockhart M., Wirrig E., Phelps A., Wessels A., Extracellular matrix and heart development. Birt. Defects Res. A. Clin. Mol. Teratol. 91, 535–550 (2011). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12.Rienks M., Papageorgiou A.-P., Frangogiannis N. G., Heymans S., Myocardial Extracellular Matrix. Circ. Res. 114, 872–888 (2014). [DOI] [PubMed] [Google Scholar]
  • 13.Silva A. C., Pereira C., Fonseca A. C. R. G., Pinto-do-Ó P., Nascimento D. S., Bearing My Heart: The Role of Extracellular Matrix on Cardiac Development, Homeostasis, and Injury Response. Front. Cell Dev. Biol. 8 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14.Hofbauer P., Jahnel S. M., Papai N., Giesshammer M., Deyett A., Schmidt C., Penc M., Tavernini K., Grdseloff N., Meledeth C., Ginistrelli L. C., Ctortecka C., Šalic Š., Novatchkova M., Mendjan S., Cardioids reveal self-organizing principles of human cardiogenesis. Cell 184, 3299–3317.e22 (2021). [DOI] [PubMed] [Google Scholar]
  • 15.Schmidt C., Deyett A., Ilmer T., Haendeler S., Caballero A. T., Novatchkova M., Netzer M. A., Ginistrelli L. C., Juncosa E. M., Bhattacharya T., Mujadzic A., Pimpale L., Jahnel S. M., Cirigliano M., Reumann D., Tavernini K., Papai N., Hering S., Hofbauer P., Mendjan S., Multi-chamber cardioids unravel human heart development and cardiac defects. Cell 186, 5587–5605 (2023). [DOI] [PubMed] [Google Scholar]
  • 16.Lewis-Israeli Y. R., Wasserman A. H., Gabalski M. A., Volmert B. D., Ming Y., Ball K. A., Yang W., Zou J., Ni G., Pajares N., Chatzistavrou X., Li W., Zhou C., Aguirre A., Self-assembling human heart organoids for the modeling of cardiac development and congenital heart disease. Nat. Commun. 12, 5142 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17.Volmert B., Kiselev A., Juhong A., Wang F., Riggs A., Kostina A., O’Hern C., Muniyandi P., Wasserman A., Huang A., Lewis-Israeli Y., Panda V., Bhattacharya S., Lauver A., Park S., Qiu Z., Zhou C., Aguirre A., A patterned human primitive heart organoid model generated by pluripotent stem cell self-organization. Nat. Commun. 14, 8245 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18.Dumortier J. G., Le Verge-Serandour M., Tortorelli A. F., Mielke A., de Plater L., Turlier H., Maître J.-L., Hydraulic fracturing and active coarsening position the lumen of the mouse blastocyst. Science 365, 465–468 (2019). [DOI] [PubMed] [Google Scholar]
  • 19.Chan C. J., Chan C. J., Costanzo M., Costanzo M., Ruiz-Herrero T., Ruiz-Herrero T., xö nke G. M., Mönke G., Petrie R. J., Petrie R. J., Bergert M., Bergert M., Diz-Muñoz A., xñ oz A. D.-M., Mahadevan L, Hiiragi T., Hydraulic control of mammalian embryo size and cell fate. 571, 112–116 (2019). [DOI] [PubMed] [Google Scholar]
  • 20.Lee B. H., Fuji K., Petzold H., Seymour P., Yennek S., Schewin C., Lewis A., Riveline D., Hiraiwa T., Sano M., Grapin-Botton A., Permeability-driven pressure and cell proliferation control lumen morphogenesis in pancreatic organoids. Nat. Cell Biol. 28, 113–124 (2026). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21.Yang Q., Xue S.-L., Chan C. J., Rempfler M., Vischi D., Maurer-Gutierrez F., Hiiragi T., Hannezo E., Liberali P., Cell fate coordinates mechano-osmotic forces in intestinal crypt formation. Nat. Cell Biol., 1–32 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 22.Davis C. L., The cardiac jelly of the chick embryo. Anat Rec 27, 201–202 (1924). [Google Scholar]
  • 23.Barry A., The functional significance of the cardiac jelly in the tubular heart of the chick embryo. Anat. Rec. 102, 289–298 (1948). [DOI] [PubMed] [Google Scholar]
  • 24.Nakamura A., Manasek F. J., Experimental studies of the shape and structure of isolated cardiac jelly. J. Embryol. Exp. Morphol. 43, 167–183 (1978). [PubMed] [Google Scholar]
  • 25.Camenisch T. D., Spicer A. P., Brehm-Gibson T., Biesterfeldt J., Augustine M. L., Calabro A., Kubalak S., Klewer S. E., McDonald J. A., Disruption of hyaluronan synthase-2 abrogates normal cardiac morphogenesis and hyaluronan-mediated transformation of epithelium to mesenchyme. J. Clin. Invest. 106, 349–360 (2000). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26.Camenisch T. D., Biesterfeldt J., Brehm-Gibson T., Bradley J., McDonald J. A., Regulation of cardiac cushion development by hyaluronan. Exp. Clin. Cardiol. 6, 4–10 (2001). [PMC free article] [PubMed] [Google Scholar]
  • 27.Smith K. A., Chocron S., von der Hardt S., de Pater E., Soufan A., Bussmann J., Schulte-Merker S., Hammerschmidt M., Bakkers J., Rotation and Asymmetric Development of the Zebrafish Heart Requires Directed Migration of Cardiac Progenitor Cells. Dev. Cell 14, 287–297 (2008). [DOI] [PubMed] [Google Scholar]
  • 28.Garoffolo G., Pesce M., Mechanotransduction in the Cardiovascular System: From Developmental Origins to Homeostasis and Pathology. Cells 8, 1607 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 29.Peyronnet R., Nerbonne J. M., Kohl P., Cardiac Mechano-Gated Ion Channels and Arrhythmias. Circ. Res. 118, 311–329 (2016). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 30.Suchyna T. M., Johnson J. H., Hamer K., Leykam J. F., Gage D. A., Clemo H. F., Baumgarten C. M., Sachs F., Identification of a peptide toxin from Grammostola spatulata spider venom that blocks cation-selective stretch-activated channels. J. Gen. Physiol. 115, 583–598 (2000). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 31.Bae C., Sachs F., Gottlieb P. A., The mechanosensitive ion channel Piezo1 is inhibited by the peptide GsMTx4. Biochemistry 50, 6295–6300 (2011). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 32.Syeda R., Xu J., Dubin A. E., Coste B., Mathur J., Huynh T., Matzen J., Lao J., Tully D. C., Engels I. H., Petrassi H. M., Schumacher A. M., Montal M., Bandell M., Patapoutian A., Chemical activation of the mechanotransduction channel Piezo1. eLife 4 (2015). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 33.Jiang F., Yin K., Wu K., Zhang M., Wang S., Cheng H., Zhou Z., Xiao B., The mechanosensitive Piezo1 channel mediates heart mechano-chemo transduction. Nat. Commun. 12, 869 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 34.Nakamura A., Manasek F. J., An experimental study of the relation of cardiac jelly to the shape of the early chick embryonic heart. J. Embryol. Exp. Morphol. 65, 235–256 (1981). [PubMed] [Google Scholar]
  • 35.Baldwin H. S., Lloyd T. R., Solursh M., Hyaluronate degradation affects ventricular function of the early postlooped embryonic rat heart in situ. Circ. Res. 74, 244–252 (1994). [DOI] [PubMed] [Google Scholar]
  • 36.Vignes H., Vagena-Pantoula C., Prakash M., Fukui H., Norden C., Mochizuki N., Jug F., Vermot J., Extracellular mechanical forces drive endocardial cell volume decrease during zebrafish cardiac valve morphogenesis. Dev. Cell 57, 598–609.e5 (2022). [DOI] [PubMed] [Google Scholar]
  • 37.MOORMAN A. F. M., CHRISTOFFELS V. M., Cardiac Chamber Formation: Development, Genes, and Evolution. Physiol. Rev., doi: 10.1152/physrev.00006.2003 (2003). [DOI] [PubMed] [Google Scholar]
  • 38.Kamino K., Hirota A., Fujii S., Localization of pacemaking activity in early embryonic heart monitored using voltage-sensitive dye. Nature 290, 595–597 (1981). [DOI] [PubMed] [Google Scholar]
  • 39.Chen T.-W., Wardill T. J., Sun Y., Pulver S. R., Renninger S. L., Baohan A., Schreiter E. R., Kerr R. A., Orger M. B., Jayaraman V., Looger L. L., Svoboda K., Kim D. S., Ultrasensitive fluorescent proteins for imaging neuronal activity. Nature 499, 295–300 (2013). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 40.Munjal A., Hannezo E., Tsai T. Y.-C., Mitchison T. J., Megason S. G., Extracellular hyaluronate pressure shaped by cellular tethers drives tissue morphogenesis. Cell 184, 6313–6325.e18 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 41.Michaut A., Mongera A., Gupta A., Tarazona O. A., Serra M., Kefala G.-M., Rigoni P., Lee J. G., Rivas F., Hall A. R., Mahadevan L., Guevorkian K., Pourquié O., Extracellular volume expansion drives vertebrate axis elongation. Curr. Biol. 35, 843–853.e6 (2025). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 42.Chanmee T., Ontong P., Izumikawa T., Higashide M., Mochizuki N., Chokchaitaweesuk C., Khansai M., Nakajima K., Kakizaki I., Kongtawelert P., Taniguchi N., Itano N., Hyaluronan Production Regulates Metabolic and Cancer Stem-like Properties of Breast Cancer Cells via Hexosamine Biosynthetic Pathway-coupled HIF-1 Signaling. J. Biol. Chem. 291, 24105–24120 (2016). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 43.Kobayashi T., Chanmee T., Itano N., Hyaluronan: Metabolism and Function. Biomolecules 10, 1525 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 44.Kim P. K., Halbrook C. J., Kerk S. A., Radyk M., Wisner S., Kremer D. M., Sajjakulnukit P., Andren A., Hou S. W., Trivedi A., Thurston G., Anand A., Yan L., Salamanca-Cardona L., Welling S. D., Zhang L., Pratt M. R., Keshari K. R., Ying H., Lyssiotis C. A., Hyaluronic acid fuels pancreatic cancer cell growth. eLife 10, e62645 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 45.Yiangou L., Grandy R. A., Osnato A., Ortmann D., Sinha S., Vallier L., Cell cycle regulators control mesoderm specification in human pluripotent stem cells. J. Biol. Chem. 294, 17903–17914 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 46.Leonardi M., Paychère Y., Naef F., Cell-cycle inhibition preserves robust development but rebalances lineages in mouse gastruloids. bioRxiv [Preprint] (2026). 10.64898/2026.01.08.698406. [DOI] [Google Scholar]
  • 47.Ponta H., Sherman L., Herrlich P. A., CD44: From adhesion molecules to signalling regulators. Nat. Rev. Mol. Cell Biol. 4, 33–45 (2003). [DOI] [PubMed] [Google Scholar]
  • 48.Happe C. L., Engler A. J., Mechanical Forces Reshape Differentiation Cues That Guide Cardiomyogenesis. Circ. Res. 118, 296–310 (2016). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 49.Tang X., Bajaj P., Bashir R., Saif T. A., How far cardiac cells can see each other mechanically. Soft Matter 7, 6151–6158 (2011). [Google Scholar]
  • 50.Nitsan I., Drori S., Lewis Y. E., Cohen S., Tzlil S., Mechanical communication in cardiac cell synchronized beating. Nat. Phys. 12, 472–477 (2016). [Google Scholar]
  • 51.Tyser R. C. V., Srinivas S., The First Heartbeat—Origin of Cardiac Contractile Activity. Cold Spring Harb. Perspect. Biol. 12, a037135 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 52.Henley T., Goudy J., Easterling M., Donley C., Wirka R., Bressan M., Local tissue mechanics control cardiac pacemaker cell embryonic patterning. Life Sci. Alliance 6, e202201799 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 53.Heo J. H., Kang D., Seo S. J., Jin Y., Engineering the Extracellular Matrix for Organoid Culture. Int. J. Stem Cells 15, 60–69 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 54.Hofer M., Lutolf M. P., Engineering organoids. Nat. Rev. Mater. 6, 402–420 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 55.Lyra-Leite D. M., Gutiérrez-Gutiérrez Ó., Wang M., Zhou Y., Cyganek L., Burridge P. W., A review of protocols for human iPSC culture, cardiac differentiation, subtypespecification, maturation, and direct reprogramming. STAR Protoc. 3, 101560 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 56.Ma Z., Wang J., Loskill P., Huebsch N., Koo S., Svedlund F. L., Marks N. C., Hua E. W., Grigoropoulos C. P., Conklin B. R., Healy K. E., Self-organizing human cardiac microchambers mediated by geometric confinement. Nat. Commun. 6, 7413 (2015). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 57.Chen G., Gulbranson D. R., Hou Z., Bolin J. M., Ruotti V., Probasco M. D., Smuga-Otto K., Howden S. E., Diol N. R., Propson N. E., Wagner R., Lee G. O., Antosiewicz-Bourget J., Teng J. M. C., Thomson J. A., Chemically defined conditions for human iPSC derivation and culture. Nat. Methods 8, 424–429 (2011). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 58.Stuart H. T., Stirparo G. G., Lohoff T., Bates L. E., Kinoshita M., Lim C. Y., Sousa E. J., Maskalenka K., Radzisheuskaya A., Malcolm A. A., Alves M. R. P., Lloyd R. L., Nestorowa S., Humphreys P., Mansfield W., Reik W., Bertone P., Nichols J., Göttgens B., Silva J. C. R., Distinct Molecular Trajectories Converge to Induce Naive Pluripotency. Cell Stem Cell 25, 388–406.e8 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 59.Rivas F., Zahid O. K., Reesink H. L., Peal B. T., Nixon A. J., DeAngelis P. L., Skardal A., Rahbar E., Hall A. R., Label-free analysis of physiological hyaluronan size distribution with a solid-state nanopore sensor. Nat. Commun. 9, 1037 (2018). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 60.Erxleben D. A., Rivas F., Smith I., Poddar S., DeAngelis P. L., Rahbar E., Hall A. R., High-fidelity and iterative affinity extraction of hyaluronan. Proteoglycan Res. 2, e70008 (2024). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 61.Erxleben D. A., Poddar S., Elaguech M. A., Hall A. R., Analytical Assessment of Hyaluronan and Its Heavy Chain Modification with Solid-State Nanopore Sensors. Annu. Rev. Anal. Chem. 19, 227–243 (2026). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 62.Sala L., van Meer B. J., Tertoolen L. G. J., Bakkers J., Bellin M., Davis R. P., Denning C., Dieben M. A. E., Eschenhagen T., Giacomelli E., Grandela C., Hansen A., Holman E. R., Jongbloed M. R. M., Kamel S. M., Koopman C. D., Lachaud Q., Mannhardt I., Mol M. P. H., Mosqueira D., Orlova V. V., Passier R., Ribeiro M. C., Saleem U., Smith G. L., Burton F. L., Mummery C. L., MUSCLEMOTION: A Versatile Open Software Tool to Quantify Cardiomyocyte and Cardiac Muscle Contraction In Vitro and In Vivo. Circ. Res. 122, e5–e16 (2018). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 63.Love M. I., Huber W., Anders S., Moderated estimation of fold change and dispersion for RNA-seq data with DESeq2. Genome Biol. 15, 550 (2014). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 64.Yu G., Wang L.-G., Han Y., He Q.-Y., clusterProfiler: an R package for comparing biological themes among gene clusters. Omics J. Integr. Biol. 16, 284–287 (2012). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 65.Hänzelmann S., Castelo R., Guinney J., GSVA: gene set variation analysis for microarray and RNA-seq data. BMC Bioinformatics 14, 7 (2013). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 66.Susaki E. A., Tainaka K., Perrin D., Kishino F., Tawara T., Watanabe T. M., Yokoyama C., Onoe H., Eguchi M., Yamaguchi S., Abe T., Kiyonari H., Shimizu Y., Miyawaki A., Yokota H., Ueda H. R., Whole-brain imaging with single-cell resolution using chemical cocktails and computational analysis. Cell 157, 726–739 (2014). [DOI] [PubMed] [Google Scholar]
  • 67.Muntifering M., Castranova D., Gibson G. A., Meyer E., Kofron M., Watson A. M., Clearing for Deep Tissue Imaging. Curr. Protoc. Cytom. 86, e38 (2018). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 68.Ishihara K., Mukherjee A., Gromberg E., Brugués J., Tanaka E. M., Jülicher F., Topological morphogenesis of neuroepithelial organoids. Nat. Phys. 19, 177–183 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 69.Haase R., Royer L. A., Steinbach P., Schmidt D., Dibrov A., Schmidt U., Weigert M., Maghelli N., Tomancak P., Jug F., Myers E. W., CLIJ: GPU-accelerated image processing for everyone. Nat. Methods 17, 5–6 (2020). [DOI] [PubMed] [Google Scholar]
  • 70.Schindelin J., Arganda-Carreras I., Frise E., Kaynig V., Longair M., Pietzsch T., Preibisch S., Rueden C., Saalfeld S., Schmid B., Tinevez J.-Y., White D. J., Hartenstein V., Eliceiri K., Tomancak P., Cardona A., Fiji: an open-source platform for biological-image analysis. Nat. Methods 9, 676–682 (2012). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 71.Berg S., Kutra D., Kroeger T., Straehle C. N., Kausler B. X., Haubold C., Schiegg M., Ales J., Beier T., Rudy M., Eren K., Cervantes J. I., Xu B., Beuttenmueller F., Wolny A., Zhang C., Koethe U., Hamprecht F. A., Kreshuk A., ilastik: interactive machine learning for (bio)image analysis. Nat. Methods 16, 1226–1232 (2019). [DOI] [PubMed] [Google Scholar]
  • 72.van der Walt S., Schönberger J. L., Nunez-Iglesias J., Boulogne F., Warner J. D., Yager N., Gouillart E., Yu T., scikit-image: image processing in Python. PeerJ 2, e453 (2014). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 73.Sullivan C. B., Kaszynski A., PyVista: 3D plotting and mesh analysis through a streamlined interface for the Visualization Toolkit (VTK). J. Open Source Softw. 4, 1450 (2019). [Google Scholar]
  • 74.Stirling D. R., Swain-Bowden M. J., Lucas A. M., Carpenter A. E., Cimini B. A., Goodman A., CellProfiler 4: improvements in speed, utility and usability. BMC Bioinformatics 22, 433 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 75.Mi X., Chen A. B.-Y., Duarte D., Carey E., Taylor C. R., Braaker P. N., Bright M., Almeida R. G., Lim J.-X., Ruetten V. M. S., Wang Y., Wang M., Zhang W., Zheng W., Reitman M. E., Huang Y., Wang X., Li L., Deng H., Shi S.-H., Poskanzer K. E., Lyons D. A., Nimmerjahn A., Ahrens M. B., Yu G., Fast, accurate, and versatile data analysis platform for the quantification of molecular spatiotemporal signals. Cell 188, 2794–2809.e21 (2025). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 76.Kim Y., Wang K., Lock R. I., Nash T. R., Fleischer S., Wang B. Z., Fine B. M., Vunjak-Novakovic G., BeatProfiler: Multimodal In Vitro Analysis of Cardiac Function Enables Machine Learning Classification of Diseases and Drugs. IEEE Open J. Eng. Med. Biol. 5, 238–249 (2024). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 77.Archit A., Freckmann L., Nair S., Khalid N., Hilt P., Rajashekar V., Freitag M., Teuber C., Spitzner M., Tapia Contreras C., Buckley G., von Haaren S., Gupta S., Grade M., Wirth M., Schneider G., Dengel A., Ahmed S., Pape C., Segment Anything for Microscopy. Nat. Methods 22, 579–591 (2025). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 78.Chapman S. C., Collignon J., Schoenwolf G. C., Lumsden A., Improved method for chick whole-embryo culture using a filter paper carrier. Dev. Dyn. 220, 284–289 (2001). [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplement 1
Download video file (1.7MB, mp4)
Supplement 2
Download video file (2MB, mp4)
Supplement 3
Download video file (1.1MB, mp4)
Supplement 4
Download video file (642.2KB, mp4)
Supplement 5
Download video file (3.8MB, mp4)
Supplement 6
Download video file (710.3KB, mp4)
Supplement 7
Download video file (370.1KB, mp4)
Supplement 8
Download video file (556.7KB, mp4)
Supplement 9
Download video file (1.2MB, mp4)
Supplement 10
Download video file (544.1KB, mp4)
Supplement 11
Download video file (7.2MB, mp4)
Supplement 12
Download video file (245.5KB, mp4)
Supplement 13
Download video file (3.4MB, mp4)
Supplement 14
media-14.xlsx (49.6MB, xlsx)
Supplement 15
media-15.xlsx (54.1KB, xlsx)
Supplement 16
NIHPP2026.01.26.700345v2-supplement-16.pdf (3.3MB, pdf)

Data Availability Statement

RNA-seq data has been deposited to the NCBI Gene Expression Omnibus and is accessible through the GEO accession number GSE317879. 3D imaging data will be deposited in the BioImage Archive. Custom analysis code is available from the corresponding authors upon reasonable request. Cell lines and plasmids are available upon reasonable request from the corresponding authors and may require a material transfer agreement.

Articles from bioRxiv are provided here courtesy of Cold Spring Harbor Laboratory Preprints

RESOURCES