De Novo Elastin Assembly Alleviates Development of Supravalvular Aortic Stenosis – Brief Report

Abstract

Background

A series of incurable cardiovascular disorders arise due to improper formation of elastin during development. Supravalvular aortic stenosis (SVAS), resulting from a haploinsufficiency of elastin (ELN), is caused by improper stress sensing by medial vascular smooth muscle cells (VSMCs), leading to progressive luminal occlusion and heart failure. SVAS remains incurable, as current therapies do not address the root issue of defective elastin.

Methods

We employ SVAS here as a model of vascular proliferative disease using both human induced pluripotent stem cell (hiPSC)-derived VSMCs and developmental Eln^+/− mouse models to establish de novo elastin assembly as a new therapeutic intervention.

Results

We demonstrate mitigation of vascular proliferative abnormalities following de novo extracellular elastin assembly through addition of the polyphenol epigallocatechin gallate (EGCG) to SVAS hiPSC-derived VSMCs and in utero to Eln^+/− mice.

Conclusions

We demonstrate de novo elastin deposition normalizes SVAS hiPSC-VSMC hyperproliferation and rescues hypertension and aortic mechanics in Eln^+/− mice, providing critical preclinical findings for the future application of EGCG treatment in humans.

Graphical Abstract

Introduction

Arterial obstruction in vascular proliferative diseases, including atherosclerosis, coronary restenosis, and supravalvular aortic stenosis (SVAS), is associated with hyperproliferation of vascular smooth muscle cells (VSMCs) and defective extracellular elastin, and represents a leading cause of morbidity and mortality in the developed world.^1-3 Elastin is fundamental to maintaining vascular biomechanics and physiology, while its disorganization leads to an array of vascular proliferative diseases,⁴ which remain difficult to treat due to the inability to adequately facilitate the assembly of functional elastin in disease states. These complications are further compounded by the lack of an effective human model to study elastin biomechanical signaling in vitro, thereby addressing long-standing questions in the field about the role of elastin in promoting cellular quiescence. Therefore, treatments of elastinopathy aimed at restoring functional elastic fibers and a human model for unraveling elastin biomechanical signaling remain in urgent need.

Here we utilize the condition SVAS, a primary elastinopathy, as a model of vascular proliferative disease resultant from elastin haploinsufficiency and elastic fiber disorganization. SVAS is characterized by improper stress sensing and severe, progressive hyperproliferation of VSMCs in the ascending aorta, occluding the aortic lumen and increasing vessel stiffness, ultimately leading to heart failure.⁴ Currently proposed therapies focus on downstream sequelae of elastin deficiency and do not address the root issue of defective extracellular elastin.^5,6 In order to treat SVAS it is essential to restore the extracellular assembly of functional elastic fibers, which has been difficult to replicate in smooth muscle.

Herein we describe rescue of defects caused by elastin haploinsufficiency through the addition of the polyphenol epigallocatechin gallate (EGCG) to facilitate the assembly of functional, extracellular elastic fibers (Figure 1A). Polyphenols have been implicated in participating in elastin biosynthesis.^7,8 However, previous studies relied on wild type VSMCs for elastin formation or non-human in vitro models, which do not offer effective insight into the role of elastin in human disease pathogenesis. Moreover, the functionality of cell-produced elastic fibers is not investigated, which would be critical for characterizing the physiological relevance of deposited elastin. For the first time, we show that the polyphenol EGCG stimulated de novo assembly of extracellular elastic fibers in vitro using our SVAS human induced pluripotent stem cell (hiPSC) model,⁹ alongside phenotypic rescue in an in utero mouse model of elastin insufficiency (Eln^+/−),^4,5 setting the stage for potential future application in humans (Figure 1A). Notably, we show that EGCG rescued SVAS hiPSC-VSMC hyperproliferation leading to normalization of downstream mechanosignaling marker expression. Further, in utero EGCG-treated Eln^+/− mice exhibited normalized blood pressures, aortic mechanical properties, and an improved organization of their elastic lamellae when compared to vehicle-treated Eln^+/− mice. Such studies provide essential preclinical findings for the potential future treatment of SVAS (Figure 1A).

Figure 1. EGCG Facilitates Elastin Fiber Assembly In Vitro.

A, Overview of research approach. Top: SVAS hiPSC-VSMCs are normalized following EGCG treatment. Middle: In utero EGCG-treated Eln^+/− mice are born with physiologically normalized cardiovasculature. Bottom: Future application of EGCG in humans. B, Representative images of EdU incorporation and total nuclei number in control or SVAS hiPSC-VMSCs treated with vehicle or EGCG (n≥5). Scale bar: 200 μm. C, Quantification of percent EdU incorporation in B. D, Quantification of total cell number in B. E, Representative western blots for integrin β3 normalized to GAPDH in control and SVAS hiPSC-VSMCs treated with vehicle or EGCG (n=8). F, Quantification of normalized protein expression from E. G, Representative western blots for p-FAK and FAK normalized to GAPDH in control and SVAS hiPSC-VSMCs treated with vehicle or EGCG (n=4). H, Quantification of normalized protein expression from G. I, Immunofluorescence of decellularized ECM for elastin (ELN, green) and fibrillin-1 (red) derived from SVAS hiPSC-VSMCs cultured in medium containing vehicle or EGCG. Scale bar: 200 μm. J, Quantification of hot oxalic acid-digested insoluble elastin protein by the Fastin assay of SVAS hiPSC-VSMCs cultured with either vehicle or EGCG (n=12). Total α-elastin levels were normalized to the total amount of soluble protein derived from the cells which is expected to be directly proportional to total cell counts. K, qRT-PCR quantification of relative fold change in gene expression between SVAS hiPSC-VMSCs treated with vehicle or EGCG for fibrillin-1 (FBN1), fibulin-4 (FBLN4), fibulin-5 (FBLN5), matrix metalloproteinases 9 and 12 (MMP9 and MMP12), and ELN (n≥3). L, Schematic for naïve SVAS hiPSC-VSMC reseeding. Vehicle or EGCG-treated cells are decellularized for an elastin enriched (green) or unenriched (gray) ECM. Matrices are treated with or without elastase prior to reseeding. M, EdU incorporation in reseeded naïve SVAS hiPSC-VMSCs. Scale bar: 200 μm. N, Quantification of percent EdU incorporation (n=4). Data were analyzed via two-way ANOVA with Tukey’s multiple comparisons (C, D, F, H, and N), unpaired Mann-Whitney test (J) or two-tailed, unpaired Welch’s t-test (K). Mean values and S.E.M. indicated by the error bars are shown. n: biological replicates.

Our study additionally provides an ideal model for investigating elastin biomechanical signaling and identifying critical mediators via robust, human VSMC-produced extracellular elastic fibers which could have broad ramifications for future therapeutic development for vascular proliferative diseases.

Methods

The data that support the findings of this study are available from the corresponding author upon reasonable request.

Animal use for EGCG Testing in vivo

Eln^+/+ female 8-week-old mice C57BL/6J mice (Jackson Labs) were bred with Eln^+/− males to produce a mixture of wild-type and heterozygous offspring. Upon observable plug formation (embryonic day 0.5), pregnant dams were tracked until E16.5, when daily 10 mg/kg intraperitoneal injection of either DMSO vehicle or EGCG began and continued through pup weaning at postnatal day 21. Between 14 and 19 independent samples per group in each cohort were studied. All animal treatments and measurements were conducted under the approval of the Institutional Animal Care and Use Committee (IACUC) of the Yale School of Medicine.

Human Induced Pluripotent Stem Cells

Human cell populations were derived using protocols approved by Yale University Human Investigation Committee. A previously established wild-type human induced pluripotent stem cell line and an elastin haploinsufficient stem cell line were employed in our research.⁹ Both hiPSC lines were originally produced through reprogramming vascular smooth muscle cells from arterial tissues, under Yale Institutional Review Board approval, using an hSTEMCCA polycistronic lentiviral vector encoding human OCT4, KLF4, SOX2, and c-MYC genes, followed by Cre recombinase-mediated removal of the ectopic reprogramming factors. To maintain pluripotency, the hiPSC lines were cultured on mouse embryonic fibroblast (MEF) feeder layers in DMEM/F12 medium containing 20% knock-out serum (GIBGO, Invitrogen) and 100 ng/mL FGF2, and passaged after reaching 70-80% confluency (~4-5 days) by ethylenediaminetetraacetic acid (EDTA; ThermoFisher) treatment.

Generation of VSMCs from hiPSCs

hiPSC-VSMCs were obtained via an embryoid body (EB)-based approach. Briefly, hiPSC colonies were expanded to 80% confluency on MEFs pre-seeded on tissue culture plates (Corning). Cells were then treated with 1 mg/mL Dispase II (Gibco; 17105041) for 7-9 minutes and mechanically dissociated and centrifuged. The iPSC pellet was resuspended in mTeSR1 supplemented with 1:100 (v/v) GFR-Matrigel and 5 μM ROCK inhibitor (Y-27632; Millipore) and transferred to a 6-well low attachment plate. Over the following 72 hour period, the mTeSR1 medium was gradually mixed with EB medium [(Dulbecco’s Modified Eagle Medium (DMEM; high glucose, ThermoFisher) supplemented with 10% fetal bovine serum (FBS, Gemini), 2 mM L-glutamine (ThermoFisher), 1% (v/v) non-essential amino acid (NEAA, ThermoFisher), 1% (v/v) penicillin/streptomycin (pen/strep, ThermoFisher), and 0.012 mM β-mercaptoethanol (Sigma-Aldrich)] in volume ratios of 2:1 (mTeSR1:EB medium) on day 1, 1:1 on day 2, and 1:2 on day 3. From days 4-6, EBs were cultured in 100% EB medium in suspension. Following this, EBs were collected and seeded onto gelatin-coated cell culture dishes for 6 additional days in EB medium to facilitate outgrowth of differentiated cells. To induce hiPSCs to the VSMC lineage, adherent EB-derived cells were dissociated with 0.05% trypsin-EDTA (ThermoFisher), re-seeded at 20,000 cells/cm² onto cell culture plates coated with GFR-Matrigel, and cultured in SmGM-2 expansion medium (Lonza). It typically took 7-10 days of culture in SmGM-2 medium to obtain proliferative hiPSC-derived VSMCs for additional studies.

Generation of Extracellular Elastin

To produce extracellular elastin in vitro, hiPSC-VSMCs were seeded at 20,000 cells/cm² in SmGM-2 medium. At 80% confluency, medium was exchanged for 3% FBS + 1 ng/mL TGF-β serum-reduced elastin-promoting culture medium supplemented with either DMSO vehicle or 7.5 μg/mL EGCG for 12 days. To maximize the effects of EGCG addition medium was replenished once every 72 hours.

Decellularization Assay

Following established protocols,¹⁰ cell monolayers were removed from the extracellular matrix (ECM) using a decellularization buffer [DPBS with Ca²⁺ and Mg²⁺ (ThermoFisher), 20 mM NH₄OH (JT Baker Chemical Company, JTB-9721-02), and 0.5% (v/v) Triton-X 100 (AmericanBIO, AB02025)]. Cells were washed once with DPBS supplemented with Ca²⁺ and Mg²⁺, and subsequently submerged in decellularization buffer and incubated for 2 minutes at 37°C. Buffer was discarded and decellularized matrices were gently washed with DPBS with Ca²⁺ and Mg²⁺ twice for 5 minutes each, with the first of these incubations occurring at 37°C and the second at room temperature. Matrices were then incubated overnight at 4°C in DPBS with Ca²⁺ and Mg²⁺ supplemented with 1:1000 DNase I (Roche, 04716728001) to remove any remaining debris. Decellularized matrices could then immediately proceed to blocking for immunostaining.

Elastase Assay

Porcine Pancreatic Elastase (Elastase, Lyophilized; Worthington-Biochem, LS002292) reconstituted to 30 μg/mL in basal DMEM was added to decellularized extracellular matrices for 15 minutes at 37°C to selectively degrade elastin in the ECM while leaving other proteins intact. Naïve, previously untreated hiPSC-VSMCs cultured in SmGM-2 growth medium were dissociated and subsequently reseeded at 20,000 cells/cm² on top of these matrices for further analysis.

Blood Pressure Measurements

Anesthesia was induced and maintained by inhaled isoflurane (3% induction, 1% maintenance) in every mouse for hemodynamic testing for a duration of 15 minutes. The right carotid artery was dissected and a 1F French transducer-tipped catheter (ADInstruments, SPR-1000) was inserted to the ascending aorta for blood pressure measurement. Data was collected using a PowerLab digital data acquisition system (ADInstruments, PL2604) and analyzed on a computer using LabChart software.

Derivation and Mechanical Assessment of Murine Aortic Segments

Following exsanguination via DPBS perfusion, aortic segments of approximately 1 mm in length were harvested from mice and kept in DPBS at 4°C until measurement. Precise measurements were taken for each sample using a digital caliper and mounted into a Instron 5960 microtester to evaluate the stress versus strain relationship. The samples were mounted onto two appropriately sized stainless-steel pins, one anchored to a 10N load cell and the other to the stationary arm of the Instron 5960 microtester. For the explanted mouse vessel segments, a rectangular approximation was used to calculate the cross-sectional area for murine aortic vessels. For a rectangular geometry $A=w\ast t$. Width ($w$) was determined by the size of the vessel segment’s long axis (~1 mm), which was directly measured for each segment using an electronic caliper. Thickness ($t$) was approximated by the diameter of one of the stainless-steel pins used for mounting the vessels to the Instron device. The Instron 5960 microtester would make real-time measurements and calculations of engineering strain ($\epsilon$) and engineering stress ($\sigma$) using these inputs. The strain was determined with the standard formula $\epsilon =\frac{\Delta L}{L_0}$. Note that the initial length ($L_0$) was approximated by the combined diameter of the top and bottom stainless-steel pins, while the change in length ($\Delta L$) was derived by subtracting $L_0$ from the length of a given time $(L_t):\Delta L=L_t-L_0$. Stress was measured by dividing the applied force (F) by the current cross-sectional area: $\sigma =\frac{F}{A}$. Each aortic segment was cyclically pre-stretched for three cycles at 10% strain, then progressively stretched until failure.

Histology

Following exsanguination via DPBS perfusion, aortic segments were harvested from mice and directly fixed in 4% PFA at 4°C overnight. PFA was then exchanged for 70% ethanol for transport to Yale Pathology Tissue Services. For Elastic Verhoeff-Van Gieson staining (EVG staining), a specific stain for elastic fibers, graft sections were processed by Yale Pathology Tissue Services based on standard protocols.

Quantification and Statistical Analysis

All graphic illustrations and statistical analyses were completed using GraphPad Prism 8. One-way ANOVA with Tukey’s multiple comparisons test was applied for comparison among more than two groups. Two-way ANOVA followed by Tukey’s multiple comparisons test was utilized to investigate the effect of two independent variables on a dependent variable among different experimental groups. Two-tailed, unpaired Welch's t-test or Mann-Whitney test was used to determine the significant differences between the control and the experimental groups. A p-value lower than 0.05 was considered significant. Numerical data were reported in format of mean ± S.E.M from three or more independent experiments. Sample size (n) for each analysis stands for the number of biological replicates and can be found in the figure legends. The statistical details of each experiment can also be found in figure legends and related results.

Results

Compared to control hiPSC-VSMCs, SVAS patient-derived cells displayed significantly higher proliferation rates and mechanosignaling protein expression of integrin β3 (ITGB3) and p-FAK (Figure 1B-H), reflecting clinical sequelae.³ Notably, we demonstrated addition of EGCG specifically rescued SVAS hyperproliferation and mechanosignaling protein expression to levels comparable to control cells (Figure 1B-H), without any notable toxicity at the efficacious dose of 7.5 μg/mL, determined through dose-response TUNEL cytotoxicity and EdU proliferation assays (Figure S1), suggesting this compound may have therapeutic utility and normalize mechanosignaling in the context of elastin defective human VSMCs.

We next examined whether EGCG could affect elastin assembly, implicated in VSMC mechanosensing and growth.¹¹ SVAS hiPSC-VSMCs cultured in an optimized elastin promoting medium of 3% serum supplemented with 1 ng/mL TGF-β in the presence of 7.5 μg/mL EGCG or DMSO vehicle were decellularized¹⁰ to observe protein abundance in the ECM, where elastin can engage in biomechanical signaling. Compared to vehicle-treated SVAS hiPSC-VSMCs, EGCG-treated matrices showed abundant deposition of elastic fibers in the ECM (Figure 1I). Importantly, immunofluorescence imaging revealed that these elastic fibers overlapped and co-stained with the glycoprotein fibrillin-1 (FBN1), which acts as the microfibrillar scaffold upon which tropoelastin is deposited and crosslinked into functional elastic fibers,⁴ indicating a potential physiological role for this newly deposited elastin. To quantify and confirm the identity of the protein, we performed a Fastin insoluble elastin quantification assay. Fastin quantification showed that EGCG addition to cultured SVAS hiPSC-VSMCs significantly increased the amount of insoluble elastin produced by these cells in either the cell membrane or extracellular space compared to vehicle treatment (Figure 1J). These results further suggest the ability of EGCG to facilitate the assembly of elastic fibers de novo in hiPSC-VSMCs. EGCG also affected the expression of other important proteins involved in elastogenesis, upregulating FBN1, fibulin-4 (FBLN4) and fibulin-5 (FBLN5) that have been shown to support the assembly and maintenance of elastin (Figure 1K).¹² Additionally, EGCG treatment decreased the expression of the elastin-degrading matrix metalloproteinases MMP9 and MMP12, promoting the stability of nascent fibers (Figure 1K), while treatment increased the expression of tissue inhibitor of metalloproteinase 1 (TIMP1) and 3 (TIMP3), which likely further contributes to the inhibition of MMPs and decreased elastic fiber degradation (Figure S2). Interestingly, the expression level of elastin mRNA itself did not significantly increase (Figure 1K), suggesting that EGCG may act to facilitate assembly of already produced tropoelastin molecules, rather than modify the synthesis of tropoelastin itself. Taken together, these results suggest that EGCG helps facilitate and maintain the assembly of extracellular elastic fibers in SVAS hiPSC-VSMCs, which may lead to the subsequent rescue of hyperproliferation and biomechanical signaling.

To further investigate a causal link between newly deposited ECM elastin and SVAS hiPSC-VSMC hyperproliferation, we decellularized vehicle or EGCG-treated SVAS hiPSC-VSMCs, leaving a native or EGCG-treated ECM. Naïve SVAS hiPSC-VSMCs with no previous EGCG interaction were subsequently reseeded onto these decellularized matrices (Figure 1L). Naïve SVAS hiPSC-VSMCs seeded onto decellularized EGCG-treated matrices led to a significantly reduced proliferation rate compared to those seeded onto vehicle-treated matrices, despite these cells never having been treated with EGCG themselves (Figure 1M, left two panels, 1N). To further validate that it was the newly synthesized elastin leading to these effects, we pretreated decellularized matrices with transient pancreatic elastase for 15 minutes at 30 μg/mL, as this dose led to the selective degradation of elastic fibers over fibrillin-1 and collagen by immunofluorescence (Figure S3).¹³ Pretreatment of decellularized matrices with transient elastase prior to cell reseeding abrogated the proliferative rescue effect (Figure 1M, bottom two panels, 1N), suggesting EGCG inhibits cellular hyperproliferation via promoting the deposition of functional elastic fibers.

We next wanted to assess the potential for EGCG treatment to rescue deficits associated with elastin insufficient mice as a preclinical model for SVAS patients. As functional elastin is deposited predominantly during the perinatal stages of development, ⁴ we elected to pursue an in utero drug treatment time course for elastin insufficient pups as a practical regimen for functional elastin restoration (Figure 1A). Pregnant dams were injected intraperitoneally daily with 10 mg/kg EGCG or vehicle from embryonic day 16.5 through pup weaning at postnatal day 21. This dose was chosen as it is well within the range of daily injection tolerance in toxicology studies,¹⁴ and preliminary dosing at the maximum indicated daily dose of 21 mg/kg/day led to observed toxicity in neonates. Consistent with previous findings, Eln^+/− pups analyzed at one month of age displayed significant hypertension compared to control mice (Figure 2A-B),⁴ while harvested aortic tissue rings showed a significantly decreased failure strain (Figure 2C-D). Importantly, both hypertensive and failure strain phenotypes were rescued in Eln^+/− pups treated with EGCG, with pressure traces, mean arterial pressures, and aortic failure strain comparable to control mice (Figure 2A-D). There were also no significant differences in Young’s modulus or maximum tensile strain across phenotypes or treatment conditions (Figure S4A-B). Elastic Verhoeff-Van Gieson (EVG) staining of ascending aorta sections revealed a similar number of elastin layers in EGCG- or vehicle-treated Eln^+/− mice, though the darker EVG staining and smooth appearance of these fibers in Eln^+/+ or drug treated Eln^+/− mice potentially suggests an improved quality (Figure S4C). Additional proteins important in elastic fiber assembly and maintenance were also augmented following EGCG treatment, as immunofluorescence staining of aortic sections demonstrated an increased abundance of LTBP-4 and MFAP-4 localized to lamellar units in EGCG-treated elastin heterozygous mice compared to vehicle (Figures 2E-H), further suggesting an enhanced organizational structure of the elastic lamellae.^15,16 Notably, EGCG treatment in Eln^+/+ mice did not appear to provide additional effects in modulating blood pressure, failure strain, elastic fiber structure, nor the expression of ancillary elastogenic proteins (Figure 2), likely due to the sufficient and competent elastin biogenesis machinery present in Eln^+/+ animals. Overall, EGCG in utero treatment appears to significantly mitigate cardiovascular abnormalities in elastin insufficient mouse models, providing essential preclinical results for the potential future use of this drug in the treatment of SVAS in humans.

Figure 2. In Utero Treatment of EGCG Rescues Cardiovascular Dysfunction in Elastin Insufficient Mice.

A, Representative blood pressure traces from the ascending aorta of 1-month-old mice treated in utero with 10 mg/kg/day vehicle or EGCG. Average peak systolic and diastolic pressures (mmHg) are shown. B, Quantification of ascending aorta mean arterial pressures (n≥4). C, Representative tensile stress versus strain plots for mouse ascending aorta segments (~1 mm length). D, Failure strain quantification for ascending aorta segments (n=3). E, Representative immunofluorescence images of the elastic fiber-associated marker latent TGF-β binding protein 4 (LTBP4, far-red). Elastin autofluorescence is shown in green. Nuclei were counterstained by DAPI. Dashed lines indicate outer boundary for medial VSMCs; *: Indicates aortic lumen. Scale bar: 100 μm. F, Quantification of the mean fluorescence intensity normalized by area of LTBP4 (n=4). G, Immunofluorescence imaging of the elastic fiber-associated marker microfibril-associated glycoprotein 4 (MFAP4, red). Elastin autofluorescence is shown in green. Nuclei were counterstained by DAPI. Dashed lines indicate outer boundary for medial VSMCs; *: Indicates aortic lumen. Scale bar: 100 μm. H, Quantification of the mean fluorescence intensity normalized by area of MFAP4 (n=5). Data were analyzed via two-way ANOVA with Tukey’s multiple comparisons (B, D, F, and H). Mean values and S.E.M. indicated by the error bars are shown. n: biological replicates.

Discussion

Our study represents the first demonstration of enhanced elastic fiber assembly to alleviate phenotypic SVAS disease. We show that the polyphenol EGCG normalized SVAS hiPSC-VSMC hyperproliferation and mechanosignaling protein expression abnormalities by enriching ECM elastin content. We further demonstrate that this elastin-enriched ECM is itself necessary and sufficient to prevent further SVAS hyperproliferation, providing direct evidence of elastin as a quiescent matrix protein and a decellularized extracellular elastin model as a novel platform to unravel critical mediators in the elastin mechanosignaling cascade. We then leveraged these discoveries to characterize a preclinical proof-of-principle model of enhanced elastic fiber assembly via EGCG as a potential treatment for cardiovascular dysfunction in vivo associated with elastin insufficiency.

It is important to note that earlier studies of SVAS disease often relied on an Eln^−/− mouse model, which does not reflect human disease, led to severe growth retardation or premature death in treated pups, and focused primarily on downstream factors of elastin signaling.^5,6 As such, these treatments do not address the fundamental insult of elastin insufficiency and cannot ultimately offer a curative solution to SVAS. To provide effective, lasting treatment, functional ECM elastin must be provided to restore proper biomechanical stress sensing, thereby naturally correcting downstream aberrant hyperproliferation resulting from ELN haploinsufficiency. Our combined human SVAS hiPSC-VSMC and Eln^+/− mouse models represent a robust approach for addressing the root issue of rescuing dysfunctional ECM elastic fibers. The in vitro and in vivo findings presented here suggest that EGCG is a promising candidate for SVAS treatment as it acts to facilitate the assembly and maintenance of functional elastic fibers, and that this restoration is sufficient to normalize phenotypic disease hallmarks. Our in utero treatment regimen further suggests early introduction of EGCG during the perinatal time course of elastin assembly can rescue elastin insufficient mice from apparent hypertension and improve the quality of elastic fibers produced. Due to the hereditary nature of SVAS, our findings establish a foundation for EGCG as a novel in utero therapy for patients with currently incurable elastin disorders.

It is also important to investigate the mechanism of action of EGCG-facilitated elastic fiber assembly and SVAS rescue in greater detail in follow-up studies. We hypothesize that the negatively charged hydroxyl side chains of the polyphenol exhibit beneficial polar interactions with positively charged lysine residues on tropoelastin monomers, stabilizing and coordinating coacervation of monomers and the productive assembly of polymeric elastin. Stabilized monomers, oligomers, or polymers of elastin might then feedback to the VSMCs, promoting cellular quiescence, augmenting transcription of elastic fiber (FBLN4 and 5) and elastin-stabilizing genes (TIMP1 and 3), and downregulating the expression of elastin-degrading genes (MMP9 and 12) for effective fiber assembly. This hypothesis is supported by our findings in Figure 1I, which demonstrate a cooperative deposition pattern between fibrillin-1 and elastin in the presence of EGCG treatment, whereas without EGCG there appears to be a compensatory overabundance of fibrillin-1 in the ECM without bound tropoelastin. Likewise, the insignificant change in expression of the tropoelastin gene (Figure 1K) versus the significant increase in insoluble elastin production (Figure 1J) further supports the role of EGCG as a coordinator of elastic fiber assembly from already secreted tropoelastin molecules rather than as a stimulator of tropoelastin gene expression. Alternatively, it is also possible that EGCG may be acting to increase the translational efficiency of tropoelastin mRNA, potentially via controlling mRNA sequestering and/or release or through interactions with mediators such as microRNAs which could increase or decrease polysome density.

In Eln^+/− mice EGCG treatment led to a rescue of LTBP4 and MFAP4 levels, which may contribute to elastic fiber assembly and maintenance through interactions with fibulin-5 and enhancement of tropoelastin self-assembly.^15,16 We will investigate the potential effect of EGCG on TGF-β bioavailability in the ECM through the retention of the growth factors by LTBP4 and MFAP4 in follow-up studies. Additionally, future efforts are warranted to identify the cellular targets of EGCG via affinity-based pull-down methods with tag-conjugated EGCG and SVAS hiPSC-VSMC lysates. The respective contributions of SVAS hiPSC-VSMC hyperproliferative rescue due to EGCG-mediated elastic fibers versus EGCG cell cycle inhibition will be investigated via siRNA-mediated knockdown of each target, EdU incorporation, and extracellular ELN immunofluorescence.

It could be interesting in the future to test whether extending the in utero treatment period could even further improve the observed rescue effects in elastin insufficient mice. As elastin expression in the mouse begins as early as embryonic day 12,¹⁷ beginning treatment at this stage could lead to even more pronounced effects. Similarly, direct treatment of newborn pups, rather than through lactation, could lead to improved elastic fiber formation during early cardiovascular development and growth. Physiological rescue effects described here could also be assessed for maintenance in aged mice.

In this study, effective elastic fibers were derived by culturing SVAS hiPSC-VSMCs in 3% serum supplemented with 1 ng/mL TGF-β in the presence of 7.5 μg/mL EGCG, revealing the efficacy of EGCG in promoting elastic fiber formation even with only one functional copy of the elastin gene. In the in vitro cell culture model, it is actually quite difficult to produce extracellular elastin in the absence of EGCG, even in control cells with healthy elastin genes.⁷ Based on robust, EGCG-assisted extracellular ELN formation by SVAS hiPSC-VSMCs, we have already demonstrated here that elastin indeed acts as a quiescent matrix, as SVAS hiPSC-VSMCs reseeded onto elastin-rich matrices exhibited significantly reduced proliferation rates, while pretreatment of the matrix with transient elastase blunted this effect. Thus, our decellularized extracellular elastin model may allow direct investigation into putative elastin complex receptors³ to identify therapeutic targets for elastinopathies. In follow up studies, reseeded SVAS cells could be pretreated in suspension with siRNAs to knock down putative elastin complex receptors, followed by characterization of hyperproliferative rescue maintenance or abrogation. With this strategy, we could effectively identify the essential receptors responsible for elastin signal transduction, which would be invaluable not only for developing treatments for SVAS but for a wide range of elastinopathies. Future endeavors will also be made to unravel the fundamental mechanisms by which EGCG modulates the above elastin mechanosensing pathways and alleviates vascular pathology caused by elastin haploinsufficiency in mice with larger sample sizes.

Further, due to the concomitantly observed rescue of essential elastic fiber-associated factors in vitro, including FBLN4 and FBLN5, and in vivo of LTBP4 and MFAP4, our EGCG-facilitated elastic fiber assembly approach may have the potential to be broadly applicable in the treatment of many vascular disorders,^{12,13,15,18,19} after carefully confirming its efficacy through human clinical trials in the future. Additionally, due to the anti-inflammatory and elasto-protective effects of EGCG, supplements could be taken or selectively delivered to critical sites throughout life as a non-invasive strategy for mitigation of SVAS disease progression, elastin degradation, and other vascular proliferative diseases such as atherosclerosis and coronary restenosis beyond the perinatal stages.

Highlights.

• Robust assembly of functional, extracellular elastic fibers in vitro via the polyphenol epigallocatechin gallate is sufficient to rescue disease hallmarks in human vascular smooth muscle cells.

• Demonstration of elastic fiber quiescent function through novel decellularized elastin platform which could be leveraged to investigate elastin mechanosignaling.

• In vivo rescue of elastin insufficient disease phenotypes in mice treated in utero with epigallocatechin gallate.

Non-standard Abbreviations and Acronyms:

SVAS

Supravalvular aortic stenosis

hiPSC

human induced pluripotent stem cell

VSMC

vascular smooth muscle cell

EGCG

epigallocatechin gallate

ECM

extracellular matrix

EVG

Elastic Verhoeff-Van Gieson