Trisomy 21 Alters Cell Proliferation and Migration of iPSC-Derived Cardiomyocytes on Type VI Collagen

Abstract

Purpose

Individuals with Down syndrome (DS) are 2000 times more likely to develop a congenital heart defect (CHD) than the typical population Freeman et al. in Am J Med Genet 80:213–217 (1998). The majority of CHDs in individuals with DS characteristically involve the atrioventricular (AV) canal, including the valves and the atrial or ventricular septum. Type VI collagen (COLVI) is the primary structural component in the developing septa and endocardial cushions, with two of the three genes encoding for COLVI located on human chromosome 21 and upregulated in Down syndrome (von Kaisenberg et al. in Obstet Gynecol 91:319–323, 1998; Gittenberger-De Groot et al. in Anatom Rec Part A 275:1109–1116, 2023).

Methods

To investigate the effect of COLVI dosage on cardiomyocytes with trisomy 21, induced pluripotent stem cells (iPSC) from individuals with DS and age- and sex-matched controls were differentiated into cardiomyocytes (iPSC-CM) and plated on varying concentrations of COLVI.

Results

Real time quantitative PCR showed decreased expression of cardiac-specific genes of DS iPSC-CM lines compared to control iPSC-CM. As expected, DS iPSC-CM had increased expression of genes on chromosome 21, including COL6A1, COL6A2, as well as genes not located on chromosome 21, namely COL6A3, HAS2 and HYAL2. We found that higher concentrations of COLVI result in decreased proliferation and migration of DS iPSC-CM, but not control iPSC-CM.

Conclusions

These results suggest that the increased expression of COLVI in DS may result in lower migration-driven elongation of endocardial cushions stemming from lower cell proliferation and migration, possibly contributing to the high incidence of CHD in the DS population.

Supplementary Information

The online version contains supplementary material available at 10.1007/s12195-023-00791-x.

Introduction

Cardiac septation of the atrioventricular (AV) canal occurs between embryonic day 28 and 50 (E28–E50) in humans and is necessary to ensure division of the upper atria and lower ventricles of the heart [4]. This process of AV septation begins with swelling of the endocardial cushions, termed the “cardiac jelly”, due to large deposits of extracellular matrix. Cells from the developing endocardium infiltrate and populate the cardiac jelly, undergo epithelial-to-mesenchymal transition (EMT), and proceed to remodel the tissue [5]. AV septa tissue remodeling is highly dependent on cell migration, proliferation, and differentiation to ensure endocardial fusion and complete septation [4]. Alternatively, abnormal development or malformation of the endocardial cushions of the AV septum can result in a septal defect. Trisomy 21, the cause of Down syndrome (DS), results from the full or partial non-disjunction of human chromosome 21 (Hsa21) during maternal meiosis and is a major risk factor for developing a congenital heart defect (CHD). Nearly 50% of infants with DS are born with a CHD. Of these, 45% are diagnosed with an atrioventricular septal defect (AVSD) and 35% are diagnosed with a ventricular septal defect (VSD) [1]. Previous studies have investigated the correlation of genetics and structural heart disease in mouse models, though this approach is limited in the predictive ability for structural heart defect development in humans due to the imperfect recapitulation of trisomy 21 in animal models [6].

Previous work investigating human cases of partial trisomy 21, where only a fraction of chromosome 21 was triplicated, revealed a narrow region within Hsa21 q22.2-22.3 predictive of CHD [7]. This DS-CHD region contains several key genes, including the genes encoding for the α1 and α2 subunits of type VI collagen (COL6A1, COL6A1; COLVI) [8]. COLVI is expressed in the endocardial cushion ECM concurrent with septation and endocardial cushion remodeling processes [3, 9]; is secreted by cardiac fibroblasts and cardiomyocytes[10]; provides tensile strength to the surrounding tissue [11] and plays a substantial role in multiple tissues to regulate biomechanical properties [11, 12]. Furthermore, it was demonstrated that human fetuses with DS show increased expression of COLVI in the endocardial cushions during heart development compared to typical fetuses [3]. Changes to ECM composition have been implicated in multiple cardiomyopathies and cell-ECM interactions are known to directly regulate cell behavior and gene expression during development.

In the context of Down syndrome, previous work has shown that cells with trisomy 21 exhibit altered behavior in the presence of COLVI compared to non-trisomic cells. For example, fibroblasts with trisomy 21 exhibit increased cell adhesion to COLVI, and decreased proliferation and migration when seeded on COLVI compared to non-trisomic fibroblasts [13, 14]. Furthermore, the proteome of fibroblasts with trisomy 21 is greatly altered in the presence of COLVI compared to other substrates such as fibronectin and type I collagen [13]. However, little is known about the impact of trisomy 21-modulated upregulation of COLVI on the microenvironment that regulates cardiomyocyte gene expression, proliferation, and migration. We therefore hypothesize that trisomy 21-caused genetic upregulation of COLVI impairs cell migration and proliferation of cardiomyocytes during septation of the AV canal. To test this hypothesis, we differentiated iPSC from individuals with DS along with age- and sex-matched controls into cardiomyocytes and analyzed gene expression. We then measured how trisomy 21 impacts iPSC-CM proliferation and migration when seeded on ECM components commonly found in the heart during development [3, 13, 14], particularly type I collagen, fibronectin, and varying concentrations of COLVI.

Materials and Methods

Down Syndrome and Control iPSC Lines

Three pairs of age- and sex-matched induced pluripotent stem cell (iPSC) lines from individuals with and without DS were obtained from the Crnic Institute Human Trisome Project Biobank (www.trisome.org, NCT02864108). iPSC lines were reprogrammed by the Stem Cell Biobank and Disease Modelling Service Core at the CU Anschutz Medical Campus using mod-RNA and mi-RNAs as previously described[37]. Information on these lines can be found in Table 1. iPSC lines were maintained using mTeSR Plus (StemCell Technologies) on Corning® Matrigel® matrix using 6-well tissue culture plates. Medium was replaced every other day and iPSCs were split every 3-5 days using a 1:10 ratio and 0.5 mM EDTA.

Table 1.

Induced pluripotent stem cell lines from Crnic Institute Human Trisome Project Biobank

Pair	Gates Lab ID	Cohort	Sex	Age (years)	Source	Heart condition
1	iLC62-6	Control	M	27.1	Urine-derived renal epithelial cells	None
1	iLD7-6	DS	M	30.3	Urine-derived renal epithelial cells	ASD
2	iLC42-3	Control	M	6.1	Urine-derived renal epithelial cells	None
2	iLD19-4	DS	M	6.2	Urine-derived renal epithelial cells	PFO/PDA
3	LC68BC-5	Control	M	39	Episomal reprogramming-PBMC reprogramming	None
3	LD49BC-3	DS	M	39.1	Episomal reprogramming-PBMC reprogramming	VSD

DS Down syndrome, ASD atrial septal defect, PFO patent foramen ovale, PDA patent ductus arteriosus, VSD ventricular septal defect, M male, F female.

Karyotypic Analysis

Cytogenetic analyses of iPSC lines from individuals with DS were performed by the University of Colorado Cancer Center Cytogenetics Core using standard GTL banding (G-banding) of metaphase chromosomes. Briefly, iPSCs were seeded in T75 flask on Matrigel and incubated for 2–3 days in mTeSR Plus (StemCell Technologies) until 70% confluent. Then, cells were incubated with 10 µg/mL Colcemid™ solution for 60 minutes at RT to arrest in metaphase. The iPSCs were dissociated by 1× Trypsin-EDTA and treated with a hypotonic solution containing 0.075 M potassium chloride for 20 min, and subsequently fixed with Carnoy’s solution containing three parts methanol and one-part acetic acid. Metaphase images were scanned with 60x/oil magnification with Cytovision (Leica Biosystems) GSL automated scanner connected to a Leica DM6000B motorized microscope. A total of 20 metaphases were examined in each cell line using Cytovision version 7.7 software for karyotyping, and the results were described in accordance with International System for Human Cytogenetic Nomenclature 2020 (ISCN).

Cardiomyocyte Differentiation of iPSC

Down syndrome and control iPSC were differentiated using a previously established small molecule-based monolayer protocol with minor changes [38]. Briefly, iPSC were passaged at a 1:3 split ratio using Accutase (Millipore Sigma) into a 12-well plate coated with Matrigel. Cells were cultured in mTeSR Plus (StemCell Technologies) for 3-4 days until 100% confluent. Mesodermal induction was initiated on day 0 of differentiation by changing medium to RPMI 1640 with B27 supplement (RPMI/B27) minus insulin (Gibco) supplemented with 4 µM CHIR99021 (Selleck Chemicals). After 48 h, the medium was switched to RPMI/B27 minus insulin. On day 3, medium was replaced with RPMI/B27 minus insulin supplemented with 5 µM IWR-1 (Sigma Aldrich). After 48 h, the medium was replaced with RPMI/B27 minus insulin. On day 7, medium was changed to RPMI/B27 with insulin (Gibco) and refreshed every 2–3 days. Beating cardiomyocytes were observed between day 7 and day 10 of differentiation. From day 10 to day 14, metabolic purification of the cell population was performed by replacing medium with RPMI 1640 without D-glucose (Gibco) supplemented with B27 with insulin. Following purification, iPSC-derived cardiomyocytes were maintained in RMPI/B27 with insulin. When necessary, DS and control iPSC-CM were replated using Accutase in RPMI/B27 medium with insulin, supplemented with 15% FBS and 10 µM ROCK inhibitor (Y-27632, Sigma) to prevent cell death. Cell culture medium was replaced with RPM/B27 with insulin the next day.

Gene Expression Analysis Using RT-qPCR

Gene expression was measured using quantitative real-time PCR (RT-qPCR) analyses. Total RNA was harvested using TRIzol® Reagent (ThermoFisher) and column purified with RNeasy Mini Kit (Qiagen) following kit instructions. RNA quality and concentration was assessed using the NanoDrop ND-1000 Spectrophotometer. Reverse transcription of RNA was performed with the High-Capacity cDNA Conversion kit (Applied Biosystems) and QuantStudio3 thermocycler (ThermoFisher) according to manufacture instructions. TaqMan™ Fast Universal PCR MasterMix (2X; Applied Biosystems) was used alongside PCR assays (ThermoFisher, SI Table 1) to measure relative gene expression of cardiac genes: TNNT2, NKX2-5, GATA4, PLN, MYH6, MYH7; and DS-CHD related genes: COL6A1, COL6A2, COL6A3, HAS and HYAL2. Resulting Ct values were normalized to GAPDH and then to control iPSC-CM expression using the delta-delta Ct method. Data from three DS and three control lines were averaged and data is represented as mean with SEM.

Immunofluorescence Assay

Immunofluorescent staining was used to visualize protein expression of iPSC and iPSC-CM. Cells were washed with PBS, fixed with 4% paraformaldehyde and permeabilized with 0.5% Triton-X-100 (ThermoFisher). Non-specific antibody binding was blocked with 3% bovine serum albumin for 1 h at room temperature (RT). iPSCs were incubated with pluripotency-specific antibodies mouse anti-OCT3/4 (Santa Cruz Biotechnology, dilution factor 1:100); mouse anti-TRA-1-81 (BioLegend, dilution factor 1:250); and goat anti-SSEA-4 (BD Biosciences, dilution factor 1:250) for 1 h at RT. iPSC were then incubated with secondary antibodies (dilution factor 1:1000): Alexa Fluor 488 goat anti-Mouse IgG (H + L) (ThermoFisher) or Alexa Fluor 594 donkey anti-goat IgG (H + L) (ThermoFisher) for 1 h at RT. iPSC-CM were incubated with mouse anti-cardiac troponin T primary antibody (ThermoFisher, dilution factor 1:400) for 1 h at RT, then Alexa Fluor 546 goat anti-mouse secondary antibody (ThermoFisher, dilution factor 1:1000) for 1 h at RT. All iPSC and iPSC-CM samples were counterstained with DAPI (Vector Laboratories, dilution factor 1:1000) for 5 minutes at RT. Images were captured on Zeiss Axio Observer Z1 inverted phase contrast microscope using Zen Software (Zeiss).

Extracellular Matrix Components

To measure cell proliferation and migration on various ECM components, tissue culture treated and IncuCyte® ImageLock 96-well plates (Satorius) were coated with 12.5 µg/mL Matrigel diluted in DMEM/F12 (Gibco); 0.5 mg/mL rat tail type I collagen (Corning) diluted in 0.02 N acetic acid (Sigma Aldrich); 10 µg/mL bovine plasma fibronectin (Sigma) diluted in PBS; or 10-100 µg/mL human type VI collagen diluted in 0.02 N acetic acid. Solutions were incubated at 37 °C for 1 h. ECM components diluted in 0.02 N acetic acid were washed twice with PBS to remove excess acid before plating cells.

Cell Proliferation

Three days after cell seeding, medium was replaced with RPMI/B27 with insulin supplemented with 10 µM EdU (Invitrogen). After 48 h, cells were fixed and immunostained for EdU and cardiac troponin T (ThermoFisher) according to the Click-iT™ EdU Cell Proliferation Kit for Imaging (Invitrogen) manufacturer instructions. Cell proliferation was quantified using ImageJ (NIH). At least two independent differentiations of each cell line were used and at least 300 cells per group were analyzed for each experiment.

Cell Migration

An in vitro scratch (wound) assay was performed using materials and software from Satorius. Three days after seeding iPSC-CM on ECM components in an IncuCyte® ImageLock (Satorius) 96-well plate, cells were treated with 10 µg/mL mitomycin C (ThermoFisher) to prevent proliferation. The WoundMaker® (Satorius) was used to create a scratch across the monolayer of cells, and cells were imaged every hour for 72 h using the IncuCyte® (Satorius) live-imaging system. Resulting videos were analyzed using the TrackMate plugin from ImageJ. Data is shown from three groups per sample and at least one independent differentiation of each cell line.

Statistics

Data are represented as the mean with standard error of the mean error bars. Statistical analysis was performed using GraphPad Prism 8 Software, using one-way ANOVA with Bonferroni correction for multiple comparisons. A value of p < 0.05 was considered significant in all tests.

Results

Characterization of DS and Control Induced Pluripotent Stem Cell Lines

Pluripotency of all iPSC lines and aneuploidy of DS iPSC lines (Table 1) were confirmed prior to differentiation. G-banding karyotypic analysis confirmed the presence of one extra copy of chromosome 21 in each cell from all iPSC lines from individuals with Down syndrome in this study (47,XX, + 21 or 47,XY, + 21; Fig. 1A). Immunofluorescent staining indicated positive expression of pluripotency markers OCT3/4, SSEA-4, and TRA-1-81 within all colonies from control and DS iPSC lines (Fig. 1B). These data confirm the aneuploid identity of DS iPSC and the pluripotency of all iPSC lines prior to differentiation.

Fig. 1.

Characterization of DS and control iPSC lines. A Representative karyogram of DS iPSC metaphase chromosomes showing (47,XX, + 21 karyotype. B Representative images of immunofluorescent staining of pluripotency markers OCT3/4 (green, nuclear), TRA-1-81 (green), SSEA-4 (red) and nuclear stain, DAPI (blue) using DS and control iPSC. Scale bar 50 µm.

Characterization of iPSC-Derived Cardiomyocytes

To confirm that Down syndrome and control cell lines can be successfully differentiated into a cardiac cell lineage, six iPSC lines (Table 1) were differentiated into cardiomyocytes using a small molecule, monolayer-based protocol (Fig. 2A). By day 7 to day 10 of differentiation, cells began spontaneously contracting in culture. By day 21 of differentiation, cells stained positive for the contractile-protein cardiac troponin T (cTnT) and sarcomeric structures were visible in all three iPSC-CM pairs (Fig. 2B). RT-qPCR analyzed the relative expression of the following cardiac markers from three DS and three control iPSC-CM lines: cardiac troponin T (TNNT2), cardiac transcription factor genes (NKX2-5 and GATA4), phospholamban (PLN), L-type calcium channel (CACNA1C), and myosin heavy chain 6 and 7 (MYH6 and MYH7). Analysis of cardiac gene expression using DS and control iPSC-CM indicated a statistically significant down regulation in the relative expression of TNNT2, NKX2-5, PLN, CACNA1C, MYH6, and MYH7 in DS iPSC-CM compared to control iPSC-CM (Fig. 2C). Furthermore, it is worth noting that DS iPSC-CM displayed higher expression of MYH6 relative to MYH7, indicating a more immature cardiomyocyte or atrial phenotype.

Fig. 2.

Cardiomyocyte differentiation of DS and control iPSC. A Schematic of cardiomyocyte differentiation of iPSC. B Immunofluorescent staining for cTnT (orange) and nuclear stain, DAPI (blue) of three pairs of control and DS iPSC-CM. Average relative gene expression of three DS and three control iPSC-CM lines for C cardiac markers and D genes implicated in DS-CHD. Bars represent the average gene expression of DS iPSC-CM relative to control iPSC-CM gene expression using the standard error of the mean, n = 3 cell lines per group, one independent differentiation, and three technical replicates per sample. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001, 2-way ANOVA using Bonferroni correction for multiple comparisons. Scale bar = 10 µm.

Relative gene expression of the three alpha subunits of type VI collagen (COL6A1, COL6A2, COL6A3), hyaluronan synthase−2 (HAS2), and hyaluronidase−2 (HYAL2) was measured using RT-qPCR. These genes for the synthesis (HAS2) and enzymatic degradation (HYAL2) of hyaluronic acid (also known as hyaluronan) were included in the analysis because hyaluronic acid is also expressed in the endocardial cushions during development [15] and the synthesis of hyaluronic acid is directly upregulated by overexpression of COL6A2 [16]. Results indicated the relative expression of COL6A1, COL6A2, COL6A3, HAS2, and HYAL2 was significantly upregulated in the DS iPSC-CM lines compared to control iPSC-CM (Fig. 2D). Ultimately, results from analyzing relative gene expression confirmed that trisomy 21 and control cells could be successfully differentiated into CM. Overall, DS iPSC-CM exhibited decreased expression of CM genes compared to control iPSC-CM, but higher expression of genes located on chromosome 21 and markers implicated in DS-CHD.

Trisomy 21 Alters Proliferation on Type VI Collagen

To test how iPSC-CM proliferation is impacted by Down syndrome, control and DS iPSC-CM were seeded on different ECM components commonly found in the endocardial cushions during development: Matrigel (MGEL, a control), type I collagen (COL1), fibronectin (FN), and increasing concentrations of type VI collagen (10, 25, 50, 100 µg/mL COLVI), where “10COLVI” indicates a type VI collagen concentration of 10 µg/mL (see Fig. 3). Following culture on ECM components, cells were stained using EdU to measure proliferation and cTnT to visualize sarcomeric proteins (Fig. 3A). Proliferation of control and DS iPSC-CM was not statistically different when cells were seeded on Matrigel, fibronectin, type 1 collagen, or lower concentrations of type VI collagen (10COLVI and 25COLVI) (Fig. 3B). However, as the concentration of type VI collagen increased to 50 and 100 µg/mL (50COLVI and 100COLVI, respectively), proliferation of DS iPSC-CM drastically decreased. Meanwhile, proliferation of control iPSC-CM gradually increased on increasing type VI collagen concentrations (Fig. 3B). These results indicate that cell proliferation on higher concentrations of COLVI is altered by trisomy 21.

Fig. 3.

Proliferation of DS and Control iPSC-CM on ECM components. A Representative images of proliferation over 48 h using immunofluorescent staining of EdU (green), cTnT (orange), and DAPI (blue) of control and DS iPSC-CM on ECM components. B Quantitative analysis of proliferation using the three pairs of DS and control iPSC-CM on ECM components: MGEL: Matrigel; COL1: type I collagen; FN: fibronectin; COLVI: type VI collagen (10, 25, 50, 100 µg/mL). n > 2 differentiations for each cell line; n = 3 biological replicates per experiment. *p < 0.05, 1-way ANOVA using Bonferroni correction for multiple comparisons. Scale bar = 50 µm

Cell Migration on ECM Components

To measure differences in DS and control iPSC-CM migration on different ECM components, a scratch (wound) assay was performed using the three iPSC-CM pairs and ECM components: Matrigel (MGEL), type I collagen (COL1), fibronectin (FN), and type VI collagen (10, 25, 50, 100 µg/mL COLVI). Following scratch, iPSC-CM were imaged for 72 h using the IncuCyte® (Satorius) live-imaging system (Fig. 4A). Resulting videos were analyzed for average cell displacement (µm) and average cell velocity (µm/hr). Results showed a significant decrease in average cell displacement of DS iPSC-CM seeded on 25 and 100 µg/mL COLVI (25COLVI, 100COLVI) compared to control iPSC-CM after 72 h (Fig. 4B) with no significant differences in average cell velocity (Fig. 4C). Moreover, when seeded on even higher concentrations of COLVI of 250 and 500 µg/mL, DS iPSC-CM displacement and motility into the ‘scratched area’ was further decreased (SI Fig. 1), as measured by Relative Wound Density (RWD) percent. RWD is the percentage of spatial cell density in the scratched area relative to the spatial cell density outside the scratched area at each time point. Under typical conditions, RWD will increase as cells migrate over time. The results from these experiments concluded that trisomy 21 alters cardiomyocyte migration on higher concentrations of COLVI compared to controls.

Fig. 4.

Migration of DS and Control iPSC-CM on ECM Components. A Representative brightfield images of DS and control iPSC-CM over 72hrs following scratch at 0 h. B Average cell displacement (µm) and C average cell velocity (µm/h) over 72 h of control and DS iPSC-CM seeded on different ECM components. Results from three cell lines of DS and control iPSC-CM migration were averaged, with each point representing the average of each cell line. MGEL: Matrigel; COL1: type I collagen; FN: fibronectin; COLVI: type VI collagen (10, 25, 50, 100 µg/mL). *p < 0.05 and **p < 0.01 using 1-way ANOVA and Bonferroni correction for multiple comparisons. Scale bar = 600 µm.

Discussion

In this study, we demonstrate the use of iPSCs from individuals with Down syndrome to identify differences in cardiomyocyte behavior compared to euploid controls. The use of iPSCs has been revolutionary in the field of bioengineering and disease modeling. Several groups have applied iPSC-CM to model a variety of cardiovascular diseases [17–20]; however, the use of iPSC-CM from patients with DS is relatively novel, with few groups having differentiated DS iPSC into cardiomyocytes [17]. Much more common is a neuronal differentiation of iPSCs from patients with DS to study neurodevelopment [21]. However, there is a significant need to understand the cardiac developmental processes that lead to the prevalence of DS-CHD pathologies. Towards this goal, human iPSC technologies present a uniquely advantageous approach to investigate the effects of trisomy 21 on cell behavior. Here, we tested our hypothesis that trisomy 21 perturbs cell-ECM interacts that drive developmental processes such as cell proliferation, migration and gene expression in Down syndrome, especially when seeded on COLVI.

In our cardiomyocyte differentiations, we noticed no qualitative differences between DS and control cells as both groups spontaneously contracted by day 10 of differentiation and stained positive for cTnT. Quantitatively, DS iPSC-CM did exhibit decreased expression of cardiomyocyte-specific genes relative to control iPSC-CM. We do not believe this is due to a differentiation deficiency and instead may suggest that DS iPSC-CM are either more fetal-like compared to control iPSC-CM or are delayed in development. Moreover, the ratio between myosin heavy chains 6 and 7 genes (MYH6/MYH7) can indicate functional maturity of cardiomyocytes [22, 23]. Data from RT-qPCR indicated that DS iPSC-CM have a much higher MYH6/MYH7 ratio than control iPSC-CM. This fetal-like nature, or immaturity of cells with trisomy 21 has been reported in other cell types [24], though this study does not distinguish whether DS iPSC-CM are unable to mature or rather have a delay in maturation.

Previous studies have shown the mechanical properties of the developing endocardial cushions result from the composition and proportion of glycosaminoglycans (e.g. hyaluronic acid), collagen, and cells, with each contributing to roughly 30% of total cushion volume [25]. In DS, overexpression of type VI collagen a2-subunit (COL6A2) is known to directly upregulate hyaluronan synthase−2 (HAS2) and increase intra/extracellular production of hyaluronic acid—a glycosaminoglycan also located in the endocardial cushions during development [16]. Meanwhile, expression of the main degrading enzyme of HA, hyaluronisase−2 (HYAL2), remains unchanged, therefore increasing the net production of intra/extracellular hyaluronic acid [16]. Our data from RT-qPCR confirms upregulation of all three alpha subunits of COLVI, in addition to upregulation of HAS2 and HYAL2. Previous work showcasing the effects of tissue ECM composition on tissue biomechanics [26–28] leads us to postulate that significant increases of COLVI and HA stiffens the extracellular matrix of developing endocardial cushions in Down syndrome, that can lead to aberrant mechanotransduction. In fact, previous research has found that increasing concentrations of hyaluronic acid increases isotropic stiffness, or bulk modulus, in a collagen-HA hydrogel system [28].Though it remains unknown if cells with trisomy 21 differentially sense or respond to changes in tissue mechanics.

Previous studies have shown significant changes in cell attachment on COLVI coating concentrations between 0.5 and 500 μl/ml [29]. At a concentration of 20 μl/ml, researchers saw induction of cardiac myofibroblast differentiation [30]. Others have observed effects of COLVI coatings on migration of human pluripotent stem cell-derived cardiomyocytes starting at 2.5 μl/ml [31]. As COLVI is the major structural molecule in developing cardiac septa and is expressed concurrent with enhanced cell migration [32], we might expect to see differences in migration and proliferation pertaining to septation at the higher range of COLVI concentrations.

Septation occurs when mesenchymal cells of epicardial origin undergo an endothelial to mesenchymal transition (EMT) and proliferate and migrate in the developing cushions [33]. This invasion and migration involves crosstalk with muscular myocardium and will not occur if myocardium is not present [34]. Future studies plan to investigate this endocardial-mesenchymal-myocardial signaling in septation and differences in cells with trisomy 21.

Moreover, to test how trisomy 21 affects cell-ECM interactions with COLVI, we measured cell proliferation and migration of the iPSC-CM when plated on various ECM components and increasing concentrations of COLVI. Data from our studies quantifying cell proliferation over 48 h indicated that DS iPSC-CM proliferated much less on higher concentrations of COLVI (50 to 100 µg/mL). Meanwhile, control iPSC-CM proliferation increased as COLVI concentration increased. It is important to note, though, that even in iPSC maintenance cultures on Matrigel, we observed qualitatively that the DS iPSC lines proliferated and slower than the matched controls—a phenomenon commonly observed in the field of Down syndrome research [34, 35]. Additionally, we found that DS iPSC-CM average displacement, but not average velocity, was decreased significantly in the presence of higher concentrations of COLVI. Type VI collagen is highly expressed in the developing AV endocardial cushions in humans, concurrent to tissue remodeling processes of cell migration and proliferation [3]. This increased secretion of COLVI by DS cardiac cells may be leading to differences in cell proliferation and migration that can affect development of the endocardial cushions during development. Future work will need to investigate how altered DS cell behavior on COLVI may impact processes during septation, such as myogenesis later in development during septa muscularization.

While CHD is more common in females with Down syndrome [36], the three donors with both Down syndrome and congenital heart defects that were available to us happened to all be from male donors. We recognize this as a limitation of this study and was not intentional. Having recognized this, we will make a greater attempt to balance the sex of cell donors in future studies.

The promising approach of using human induced pluripotent stem cell (iPSC) technology presents an attractive alternative platform for studying DS-CHD and provides many advantages over in vivo models: iPSC-derived cardiomyocytes are phenotypically immature, more accurately recapitulate fetal cell behavior during development, and can be reprogrammed from non-invasively sourced samples. This project utilized iPSC-derived cardiomyocytes as a disease model to investigate the role of trisomy 21 cell proliferation and migration in response to varying concentrations of COLVI and other ECM components commonly found in the heart during development, such as type I collagen and fibronectin. Previous studies have used primary dermal fibroblasts from skin to study DS-CHD by modeling behavioral differences amongst DS and control cells on COLVI, though these cells do not originate from cardiac cell lineage. Due to the observed differences in cell migration and proliferation of iPSC-CM from DS and control lines, we argue that cardiomyocytes more accurately model mechanisms behind septal defect formation. This is because fetal-like iPSC-derived cardiomyocytes retain properties of myofibroblasts, including significant migration and proliferation required for septation.

In conclusion, our study found iPSC-derived cardiomyocytes from individuals with DS have decreased relative expression of cardiac-specific genes, but increased relative expression of genes encoding type VI collagen, HAS2 and HYAL2. We also observed decreased proliferation and migration of DS iPSC-CM when seeded on high concentrations of type VI collagen compared to control iPSC-CM. This data suggests that high concentrations of COLVI in developing endocardial cushions, along with differences in response of DS iPSC-CM to COLVI, could contribute to higher rate of CHD in the Down syndrome population.

Supplementary Information

Below is the link to the electronic supplementary material.

Author Contributions

Conceptualization: RSR; methodology: RSR; formal analysis and investigation: RSR, AKS, KMP, JRR and MCVD; writing—original draft preparation: RSR; writing—review and editing: RSR, JRR, KMP, MCVD, AKS, JGJ; funding acquisition: JRR and JGJ; resources JRR and JGJ; supervision: JGJ.

Data Availability

The data that support the findings of this study are openly available in Mendalay at 10.17632/hpzbk2rcxs.1.

Declarations

Competing interests

The authors have no competing interests to declare that are relevant to the content of this article.