PP2A attenuates thoracic aneurysm and dissection in mouse models of Marfan syndrome

Abstract

Background:

Recent studies show that hyperactivation of mTOR signaling plays a causal role in the development of thoracic aortic aneurysm (TAA) and dissection (AAD). Modulation of Protein phosphatase 2A (PP2A) activity has been shown to be of significant therapeutic value. In light of the effects that PP2A can exert on the mTOR pathway, we hypothesized that PP2A activation by small molecule activators of PP2A (SMAPs) could mitigate AA progression in Marfan Syndrome (MFS).

Methods:

Two distinct mouse models of MFS underwent daily oral administration of SMAP compound DT-061 to assess its therapeutic potential. Echocardiography was performed to monitor the growth of the aortic root and ascending aorta. Histologic evaluation was performed to assess alterations in the vascular wall. RNA sequencing, western blot and immunostaining were performed to decipher the underlying mechanisms by which DT-061 suppresses AA progression.

Results:

PP2A activity decreased while mTOR activity increased in both human and mouse aortas with MFS. Concordantly, oral administration of DT-061 increased PP2A activation, reducing aortic expansion in Marfan mice. DT-061 treatment also mitigated medial hypertrophy, elastin breakdown, and extracellular matrix deterioration in the ascending aorta, along with decreased metalloproteinase activities. Mechanistic studies suggest that DT-061 suppresses mTOR signaling and smooth muscle cell de-differentiation, contributing to its effects on thoracic AAD progression.

Conclusions:

These studies demonstrate a pathologic role of PP2A activity loss in the etiology of MFS and implicate that activation of PP2A may serve as a novel therapeutic strategy to limit MFS progression, including aortic aneurysm formation.

Graphical Abstract

Introduction

Marfan syndrome (MFS) is an autosomal dominant inherited disease that impacts various organs, including the cardiovascular system, eyes, skeleton, lungs, skin, and dura ¹. It arises from pathogenic mutations in the fibrillin-1-encoding gene FBN1 ². The mutant fibrillin-1 disrupts microfibril formation, leading to dysfunction in vascular smooth muscle cells (VSMCs) and medial degeneration. This results in the destabilization of the aortic wall, making it susceptible to hemodynamic injury. Aortic root dilatation, aneurysms, and dissection are the primary causes of mortality in MFS patients ^3,4. Due to incomplete knowledge of the underlying molecular mechanisms driving these observed phenotypes, effective targeted therapeutic strategies are lacking, with current treatments limited to blood pressure control drugs and prophylactic aortic root replacement ⁵. Therefore, deciphering novel molecular mechanisms and identifying new therapeutic targets is imperative for treating MFS patients.

Protein phosphatase 2A (PP2A), an important and ubiquitously expressed serine threonine phosphatase in mammalian cells, controls biological functions by dephosphorylating many critical cellular molecules, including Akt, p53, c-Myc and β-catenin ⁶. By modulating key signaling pathways, PP2A plays an essential role in an array of cellular processes, including cell proliferation, migration, and apoptosis. The holoenzyme of PP2A is a heterotrimer in mammalian cells composed of a scaffolding subunit A, a regulatory subunit B, and a catalytic subunit C, that together form the active trimeric structure capable of dephosphorylating substrate proteins. The scaffolding subunit A is a highly flexible protein that serves as the main platform surface for all components to bind and interact. The C subunit provides the catalytic enzymatic function to the phosphatase. The regulatory B subunits dictate substrate specificity, with each of the 16 different B subunits known to have unique substrate affinities ⁷. Targeting PP2A has been considered challenging given the structural diversity of the various heterotrimeric forms of the enzyme that exist at any given time in our cells ⁶. Recently, a novel class of small molecules exemplified by SMAP-061 (or DT-061, as referenced in Leonard et al. ⁸) has been reported by several groups to act as a molecular glue which fits in a binding pocket between scaffolding subunit A, the catalytic subunit C and the B56alpha regulatory subunit to effectively modulate PP2A heterotrimeric formation, driving downstream dephosphorylation of essential proteins ⁸. Thus, these small molecule activators of PP2A (SMAPs) can be used as important tools to study PP2A biology, including evaluating their therapeutic potential for the treatment of PP2A-driven diseases.

Diminished PP2A activity has been found in a variety of human pathologies, such as cancer, neurological disorders, cardiovascular diseases, and autoimmune diseases ^6,9-11. Reactivation of PP2A by SMAPs has demonstrated a therapeutic benefit in inhibiting the growth of multiple tumor types ^8,12-16. In cardiovascular disease, our recent studies have demonstrated that SMAP treatment markedly attenuated the progression of AngII-induced abdominal aortic aneurysm (AAA) formation in Apolipoprotein E -deficient mice (Apoe^−/−) ¹⁷. In a murine cardiac transplantation model, increased PP2A activity prevented cardiac rejection and prolonged allograft survival, highlighting its potential to improve alloengraftment ¹⁸. While aortic aneurysm can affect any part of the aorta, from its root to the abdominal segment, AAA and thoracic AA (TAA) are two distinct disease entities. The pathogenesis and the underlying molecular and cellular mechanisms driving AAA and TAA development and progression are believed to be fundamentally distinct. AAA is commonly associated with atherosclerosis, inflammation, and weakening of the vessel wall due to risk factors such as smoking, hypertension, and genetic predisposition. In contrast, TAA exhibits a stronger genetic predisposition, as seen in conditions like Marfan syndrome, due to defects of fibrillin-1 ¹⁹. Another significant distinction lies in the embryonic lineage of smooth muscle cells comprising the wall of the abdominal and thoracic aorta. VSMCs in the ascending aorta are derived from progenitor cells originating in the cardiac neural crest, a structure derived from the neural ectoderm during embryonic development ^20,21. Conversely, VSMCs in the abdominal aorta arise from progenitor cells originating from the paraxial mesoderm located in the somites ²². Considering the above understanding and the benefit conferred by PP2A activation on AAA and allograft acceptance in a murine cardiac transplantation model, it is tempting to speculate PP2A might also hold therapeutic potential for TAA. In the current study, we investigated the association of PP2A activity and TAA. Our results show that PP2A activity is strongly impaired in both MFS patients and mice. Importantly, PP2A reactivation strongly suppresses the progression of TAA in murine models of MFS, with repression of mTOR signaling and prevention of VSMCs dedifferentiation as the underlying mechanism.

Methods

The data that support the findings of this study are available from the corresponding author upon reasonable request. A detailed Methods section is available in the Data Supplement.

Study approval

Animal studies were approved by the Institutional Animal Care and Use Committee (IACUC) at Case Western Reserve University (IACUC protocol # 2013–0128) and Emory University (PROTO201800048). All mice used in this study received humane care in compliance with the Principles of Laboratory Animal Care formulated by the National Society for Medical Research and were conducted in accordance with the National Institute of Health (NIH) Guide for the Care and Use of Laboratory Animals. Human aorta samples from MFS patients were obtained during elective or emergency surgery for proximal aortic replacement from Emory University Hospital. Ascending aorta samples used as controls were obtained anonymously from multiorgan transplant donors. Informed consent was obtained from all human participants or their families. Patient clinical data were retrieved while maintaining anonymity. Discarded and decoded human aortas were reused according to the IRB protocol (IRB00109646) approved by the Human Investigation Review Committee at Emory University.

Drug treatment

In all experiments involving the administration of SMAP, a 5mg/kg dose of DT-061 was given orally twice a day to mice starting at the age of 5 weeks (Fbn1^mgR/mgR) or 8 weeks of age (Fbn1^C1039G/+) and continued until the end of the experiments. DT-061 was prepared in a N,N-dimethylacetamide/Kolliphor HS-15/diH20 solution. The control mice were orally given vehicle solution along the same schedule.

Statistics

Data are expressed as mean ± standard error of the mean (SEM) of at least 3 independent experiments. Statistical analysis was performed using PRISM software (version 9.00, GraphPad Software, San Diego, CA), and all data sets were checked for Gaussian normality distribution. A log-rank test was used to compare survival rates. For data from 2 groups that followed Gaussian normality distribution, an unpaired or paired Student’s t-test was performed. For data that did not follow Gaussian normality distribution, an unpaired Mann–Whitney U-test was performed. One-way ANOVA (followed by Tukey post hoc test) was performed to compare single variables in multiple groups. p-values < 0.05 were considered statistically significant. All statistical analyses were performed by two researchers in a blinded manner.

Availability of data and materials

All data generated or analyzed during this study are included in this published article and its supplementary information files. The datasets generated and/or analyzed during the current study have been deposited onto the GEO database (Accession # GSE278185).

Results

Inflammatory response and protein phosphorylation pathways were augmented in MFS.

To investigate the transcriptomic changes in MFS, we analyzed bulk RNA-seq datasets from both MFS patients and MFS mice (GSE145903.1 and GSE145903.2) ²³. We identified 1064 upregulated and 841 downregulated overlapping DEGs between mouse and human MFS aortas (Figure S1A). The volcano plots show the number of upregulated and down-regulated genes in MFS patients compared to the healthy human (Figure S1B) and in the ascending aorta of MFS mice relative to their WT controls (Figure S1C). The enrichment pathway analysis on DEGs in MFS patients compared to healthy human aortic samples shows the upregulation of pathways involved in the positive regulation of cell migration, positive regulation of phosphorylation, and the inflammatory response, among others. The downregulated pathways were related to cell morphogenesis, the Notch signaling pathway, extracellular matrix organization, PI5P, PP2a and IER3 regulation of PI3K/AKT signaling, regulation of smooth muscle contraction, etc. (Figure S1D). Similarly, we found the inflammatory response, positive regulation of cell migration and positive regulation of phosphorylation were also enriched in DEGs in MFS mice compared to the WT mouse samples (Figure S1E). Select associated genes are shown in the volcano plots (Figure S1B and C). Taken together, the above results indicate that both the inflammatory response and protein phosphorylation are activated in the progression of MFS.

PP2A activity decrease while mTOR activity increases in the aorta of both humans and mice with MFS.

Intrigued by the downregulation of PP2A in MFS as revealed by the pathway enrichment analysis of DEGs between normal and MFS human samples, we next assessed whether PP2A activity (indicated by methylated catalytic subunit C (PP2A-C)) is altered in MFS aortic tissues. First, we measured the expression of various PP2A subunits and carboxymethylation of the PP2A-C in human aortic samples from MFS patients. Western blot analysis displayed that the methylated PP2A-C (a key regulator of PP2A biogenesis and heterotrimer formation) (0.44-fold, p=0.03) and total level of PP2A-C (0.70-fold, p=0.003) were significantly lower in patients with MFS compared to matched controls (Figure 1A, Figure S2A). Immunofluorescence of human aortas staining with the VSMC marker myosin heavy chain (SM-MHC) shows that the decreased methylated PP2A-C expression was mainly found within the medial VSMCs layer (Figure 1B, Figure S3).

Figure 1. PP2A activity decreases while mTOR activity increases in the aorta of humans with MFS.

A. and C. Western blot analysis of ascending aorta samples of five normal and four MFS patients. Protein levels for different subunits of PP2A (A, B56a, C) and active PP2A (methyl-PP2A-C). HSP90 was used as a loading control (A). Levels of mTOR pathway proteins (C) were tested as indicated in the figure. Vinculin was used as loading control. n = 4. Quantifications are shown in Figure S2. B. and D. Immunofluorescence analysis of ascending aorta sections from normal and MFS humans. Samples were immunostained to detect methylated PP2A-C (red), SM-MHC (green), and DAPI (blue) (B), and pS6 (red), SM-MHC (green), and DAPI (blue) (D). Representative images from Intimal and Media aortic layers are shown. Scale bars = 100 μm.

The diversity of the B regulatory subunits enables PP2A to modulate a majority of cellular signaling pathways including c-Myc ²⁴, MEK ²⁵, and mTOR ^26,27. Recent studies have demonstrated that mTOR activation induces a degradative VSMC phenotype ²⁸ . We next examined the mTOR activity in aortic samples of MFS patients. Compared to normal aortic samples, the phosphorylation of mTOR (2.33-fold, p=0.003), p70S6K (2.90-fold, p<0.001), ribosome protein S6 (12.02-fold, p=0.03), and 4EBP1 (4.42-fold, p<0.001) were significantly increased in the ascending aortas of MFS patients (Figure 1C, Figure S2B). Immunofluorescence analysis of phospho-S6 counter-stained with SM-MHC also revealed increased phosphorylation of S6 within the medial VSMCs (Figure 1D, Figure S4).

Consistent with the patients’ results, methylated PP2A-C (0.68-fold, p=0.0114) and the total level of PP2A-C (0.80-fold, p=0.0049) was decreased in the ascending aorta of Fbn1^mgR/mgR mice as compared to control Fbn1^WT/WT littermates by Western blot and immunofluorescence staining (Figure 2A and B, Figure S5B). Immunoblots also showed increased phosphorylation of mTOR (3.11-fold, p=0.00002), p70S6K (2.35-fold, p=0.007), ribosome protein S6 (32.50-fold, p<0.00001), and 4EBP1 (p=0.0002) as well as the total levels of ribosome protein S6 (p=0.00002), and 4E-BP1 (2.05-fold, p=0.0014), in the ascending aortas of Fbn1^mgR/mgR mice (Figure 2A and B). Phosphorylation of ribosome protein S6 and SM-MHC was also localized to intimal and medial VSMCs by immunofluorescence (Figure 2C), suggesting an increased mTOR activity in VSMCs.

Figure 2. PP2A activity decreases while mTOR activity increases in the aorta of Fbn1^mgR/mgR mice.

A-B. Western blot analysis (A) and Densitometric analysis (B) of protein levels in ascending aorta samples of WT and Fbn1^mgR/mgR (mgR) mice. Membranes were probed with the indicated antibodies to detect changes in the expression of different subunits of PP2A (A, B56a, C), active PP2A (methyl-PP2A-C), and the mTOR pathway. HSP90 or Vinculin were used as loading controls. n=6. C. Immunofluorescence staining of pS6 (red), SM-MHC (green), CD31 (purple), and DAPI (blue) in ascending aortic sections from normal and Fbn1^mgR/mgR mice. Scale bars = 50 μm. Med: media; Lu: lumen; Ad: Adventitia. Data are presented as mean ± SEM. Each point represents an individual biological replicate. Data was assessed for differences using a two-way ANOVA TEST. “ns” means no significant difference. *, p < 0.05 and **, p < 0.01. ***, p < 0.001 and ****, p < 0.0001.

These findings indicate that biased PP2A heterotrimer formation driven by loss of PP2A-C carboxymethylation and upregulation of the mTOR signaling pathway may play a role in the pathogenesis of MFS in both mice and humans.

Analysis of scRNA-seq data from Marfan mice reveals that PPP2CA may modulate VSMC phenotypic switching.

To corroborate the above results (Figures S1 and Figure 2-3). We re-analyzed the sc-RNAseq data (GSE227776) ²⁹ from mgR mice aortic tissues, with a focus on assessing Ppp2ca expression in vascular cell populations. Unsupervised clustering identified 7 different cell types which were subsequently identified and sorted based on DEG signatures (Figure S6A and B). The distribution of distinct cell types in WT and mgR mice is illustrated in Figure S6C. Based on the cell population results (Figure S6D), the reduced percentages of VSMCs (69.6% - 42.4%) and increased percentages of macrophages (MACs) (0.3% - 15.9%) indicate the loss of VSMCs and immune activation in TAA in mgR mice. We then conducted enrichment pathway analysis on DEGs between WT and mgR mice samples. The results suggest that the muscle structure development, actin cytoskeleton organization, vascular smooth muscle contraction, extracellular matrix organization and smooth muscle tissue development pathways are mostly enriched in WT aortic tissues, whereas the inflammatory response, cell chemotaxis, regulation of MAPK cascade, regulation of tumor necrosis factor production, regulation of peptidyl- tyrosine phosphorylation pathways are enriched in mgR mice aortic samples (Figure S7A and B). Furthermore, the hallmark gene set analysis of these DEGs from the GSEA database revealed an increased myogenesis pathways and decreased MTORC1 signaling in the WT group compared to the mgR mice group (Figure S7C and D).

Figure 3. PP2A activation attenuates aortic expansion in male Fbn1^mgR/mgR mice.

A. A workflow of DT-061 or vehicle-treated WT and Fbn1^mgR/mgR mice. B. Kaplan-Meier analysis of the survival time (log-rank test) in male Fbn1^mgR/mgR mice treated with vehicle (mgR) or DT-061 (mgR+DT-061), n=12-13. *, p < 0.05 C. Representative picture for the aorta morphology of WT and Fbn1^mgR/mgR mice after 4 weeks of vehicle or DT-061 treatment. Scale bars = 1 mm. D. Representative transthoracic echocardiography images of the ascending aorta of Fbn1^mgR/mgR male mice at baseline (Age of 5 weeks) and after treatment (Age of 9 weeks). E. Graph shows blood pressure in Fbn1^mgR/mgR male mice after 4 weeks of vehicle or DT-061 treatment. Systolic blood pressure (SBP) and diastolic blood pressure (DBP) are shown, n=6-12. Data was assessed for differences using an unpaired T test. “ns” means no significant difference. F. Quantification and statistical analysis of diameters of the aortic root (left) and proximal ascending aorta (right) in Fbn1^mgR/mgR male mice treated with DT-061 or vehicle for 4 weeks. n=13-20. Data are presented as means ± SEM, each point represents an individual biological replicate. Significant differences were evaluated with a two-way ANOVA test (F). **, p < 0.01. ***, p < 0.001 and ****, p < 0.0001.

Then, by screening Ppp2ca-expressing cells in WT and mgR samples, we found that Ppp2ca gene is expressed in all cell clusters (Figure S8A), supporting the potential importance of Ppp2ca in regulating these various cellular functions. However, when comparing the Ppp2ca-expressing cell proportions in WT and mgR groups (Figure S8B), Ppp2ca-expressing cells are significantly reduced in mgR mice (72.5% - 43.3%), suggesting that Ppp2ca positive VSMCs may play an important role in maintaining the physiological phenotype and function. We then conducted enrichment pathway analysis on DEGs between Ppp2ca⁺ cells and Ppp2ca⁻ cells in all cell populations (Figure S8C) and in SMC subpopulations (Figure S8D), separately. The results showed the same enrichment pathways included in actin cytoskeleton organization, vascular smooth muscle contraction, extracellular matrix organization, focal adhesion, elastic fiber formation, blood vessel development, Extracellular matrix (ECM)-receptor interaction, integrin cell surface interactions and post-translational protein phosphorylation are upregulated DEGs in Ppp2ca⁺ cells compared to Ppp2ca⁻ cells. Moreover, the most enriched pathways in down-regulated DEGs are immunity-modulated pathways. Furthermore, the transcription factor targets enrichment analysis (Figure S8E) of DEGs between Ppp2ca⁺ versus Ppp2ca⁻ SMC cells show that the top affected transcription factor targets include SRF, AP1, SMAD, BACH1, STAT4, PPAR, GATA1, HNF4, NFE2, AP3, TATA, etc., all of which have been shown to be important for the regulation of SMC differentiation. Collectively, these results point to the possibility that Ppp2ca may regulate the SMC phenotype switching seen in MFS.

PP2A activation attenuates aortic expansion and skeletal abnormalities in MFS mice.

The above findings prompted us to postulate that diminished PP2A activity and concomitant hyperactivation of mTOR may drive the dedifferentiation of smooth muscle cells and that PP2A activation might mitigate MFS progression. To test this hypothesis, we leveraged the use of a specific PP2A modulator, DT-061 in a mouse model of MFS to evaluate its therapeutic potential. We compared TAA progression in MFS mice treated with either DT-061 or vehicle. DT-061 was administered through oral gavage starting at 5 weeks of age. The progression of TAA was tracked using echocardiography, and the survival of MFS mice, including the incidence rates of aortic dissection and rupture, was observed up to 4 weeks and 40 days after drug administration, respectively (Figure 3A). The lethality of Fbn1^mgR/mgR male mice treated with vehicle was 31.25%. All deaths of the Fbn1^mgR/mgR mice were due to ruptured ascending aneurysms as revealed by necropsy. Very excitingly, all Fbn1^mgR/mgR male mice treated with DT-061 survived (Figure 3B). Similar results for female mice are shown in Figure S9A, where to the end of the experiment, no Fbn1^mgR/mgR mice treated with DT-061 died, and the survival rate of Fbn1^mgR/mgR mice treated with vehicle was only 80% (Figure S9A). Due to the limited number of female mice, the differences between DT-061 and the vehicle group were not statistically significant.

We next evaluated the aortic diameters at 4 weeks post-administration of DT-061 or vehicle in Fbn1^mgR/mgR mice using in vivo transthoracic echocardiography. All mice were randomly distributed into either the DT-061 or vehicle treatment groups. No differences in the ascending aorta and aortic root dimensions were observed when the treatment was initiated (Figure 3C-F, Figure S9C-D). After 4 weeks of treatment with vehicle, significant dilatation (Figure 3C and D) of the ascending aorta was noted in the Fbn1^mgR/mgR male mice (aortic root 2.20 ± 0.09 mm, ascending aorta 2.18 ± 0.08 mm) compared to the baseline (aortic root 1.66 ± 0.05 mm, ascending aorta 1.40 ± 0.05 mm) (Figure 3F). Similar findings were obtained in female Fbn1^mgR/mgR mice (Aortic root 2.25± 0.08 mm, ascending aorta 2.04± 0.12 mm) compared with baseline female (Aortic root 1.65± 0.04 mm, ascending aorta 1.35± 0.04 mm) (Figure S9C and D). Notably, in line with our hypothesis, DT-061 inhibited aortic root and ascending aorta enlargement in both males (Figure 3F) and females (Figures S9C and D). BP measurements indicated no difference between Fbn1^mgR/mgR mice treated with DT-061 or vehicle in both male and female mice (Figure 3E and Figure S9B).

To confirm this phenotype, Fbn1^C1039G/+ mice were used as a second independent MFS mouse model in this study. Both female and male mice were randomly distributed into either DT-061 or vehicle treatment groups. Treatment was administered to 8-week-old mice and in vivo transthoracic echocardiography was performed to measure the ascending aorta enlargement at 2, 4, and 6 months after treatment (Figure S10A). Similar to the results in Fbn1^mgR/mgR mice, we observed considerable growth in the ascending aorta and aortic root as mice aged. However, when male mice were treated with DT-061, this expansion was significantly reduced. In female mice, DT-061 helped alleviate aortic root dilation but did not affect the ascending aorta (Figure S10B-E).

In addition, a computed tomographic (CT) scan confirmed that both Fbn1^mgR/mgR and Fbn1^C1039G/+ mice generally presented severe skeletal deformities similar to those seen in human MFS while DT-061 treatment attenuated the kyphoscoliosis seen in both murine MFS models (Figure S11 and S12).

Taken together, these data suggest that DT-061 treatment significantly attenuates aortic dilation and kyphoscoliosis progression in Fbn1^mgR/mgR mice as well as Fbn1^C1039G/+ mice and prolongs the survival of Fbn1^mgR/mgR mice.

PP2A activation mitigates the impairment of vascular wall integrity and inhibits MMP activity in Fbn1^mgR/mgR mice

Disruption of the orderly elastic lamellae, progressive dilatation, and rupture of the aorta represents the natural progression of MFS in patients without surgical intervention. The Fbn1^mgR/mgR mice demonstrate the same progression of events. By examining the histological changes, we found that aortic sections from Fbn1^mgR/mgR mice had a significantly thicker tunica media area (WT 0.93±0.07 vs. mgR 2.13±0.07. p<0.0001), greater elastin fragmentation (WT 0.35±0.02 vs. mgR 0.12±0.01. p<0.0001), and collagen (WT 0.33±0.03 vs. mgR 0.16±0.02. p=0.0009) compared with their WT counterparts (Figure 4A-C). Interestingly, Fbn1^mgR/mgR mice treated with DT-061 showed less medial hypertrophy (mgR 2.13±0.07 vs. mgR+DT-061 1.74±0.14. p=0.0367), fragmentation of elastic fibers (mgR 0.12±0.01 vs. mgR+DT-061 0.22±0.01. p<0.0001) and collagen (mgR 0.16±0.02 vs. mgR+DT-061 0.25±0.03. p=0.0418) compared with vehicle group (Figure 4A-C). Additionally, MMPs in situ activity (Figure 4D) was increased in Fbn1^mgR/mgR mouse aortas (WT 0.01±0.003 vs. mgR 0.50±0.08. p=0.0001) and suppressed after DT-061 treatment (mgR 0.50±0.08 vs. mgR+DT-061 0.06±0.04. p=0.0001).

Figure 4. PP2A activation mitigates loss of vascular wall integrity and inhibits MMPs activity in Fbn1^mgR/mgR mice.

A-C. Representative images and quantification of aorta cross-sections of male Fbn1^mgR/mgR (mgR) mice treated with vehicle (mgR+vehicle) or DT-061 (mgR+DT-061) for 4 weeks and stained with Hematoxylin and Eosin (HE), n=5-6 (A), or Verhoeff–van Gieson stain (EVG) to assess elastic fiber architecture, n=9-11 (B), and Masson’s trichrome to assess collagen deposition, n=5-6 (C). D. In situ zymography to assess MMPs activity, n=4-6 (D). Scale bars = 50 μm. Data are presented as means ± SEM. Each point represents an individual biological replicate. Significant differences were evaluated using a one-way ANOVA test. **, p < 0.01. ***, p < 0.001, and ****, p < 0.0001.

RNA-seq demonstrates that PP2A activation promotes the enrichment of genes associated with smooth muscle contractility in Fbn1^mgR/mgR mice

To decipher the mechanism underlying the therapeutic effects of PP2A modulation, RNA-Seq was performed to identify genes differentially expressed in the ascending aortas of Fbn1^mgR/mgR mice treated with either DT-061 or Vehicle for 4 weeks. Ascending aorta of 9-week-old mice were collected and subjected to RNA-seq. The data revealed drastically different gene expression profiles in which 598 genes were upregulated, and 1633 genes were downregulated in ascending aortic tissues of DT-061 treated Fbn1^mgR/mgR mice compared with those of control mice (Figure S13A and B). We then performed KEGG enrichment pathway analysis on DEGs to identify biological processes important for ascending aortic aneurysm development in Fbn1^mgR/mgR mice with DT-061 treatment. Data suggest that the top affected processes include cytokine-cytokine receptor interaction, chemokine signaling pathway, NF-KB signaling pathway, Th1, Th2 and Th17 cell differentiation, vascular smooth muscle contraction, among others (Figure 5A). DEGs that affect the vascular smooth muscle contraction pathway were mapped with Pathview (Figure S13C). Moreover, the comparative analysis of gene expression in Fbn1^mgR/mgR mice aorta treated with DT-061 versus the vehicle group from Molecular Signatures Database using GSEA Hallmark gene set indicated 14 gene sets were significantly enriched in the vehicle group at FDR < 25%: Interferon gamma and alpha response, allograft rejection, inflammatory response, TNFα signaling via NFκB, IL2-STAT5 signaling, coagulation, p53 pathway, mitotic spindle, IL6-JAK-STAT3 signaling, mTORC1 signaling, PI3K-AKT-mTOR signaling, apoptosis, and glycolysis pathways, while 2 gene sets, myogenesis and fatty acid metabolism, were enriched in DT-061-treated Fbn1^mgR/mgR mice (Figure 5B). As shown in Figure 5C, myogenesis was most enriched in the DT-061-treated group, whereas MTORC1 signaling and PI3K-AKT-MTOR signaling were enriched in the vehicle-treated group. These findings provided additional guidance for the subsequent exploration of the molecular mechanisms and PP2A substrates driving the markedly attenuated TAA formation in DT-061-treated MFS mice.

Figure 5. Bulk RNA sequencing demonstrates that PP2A activation promotes the enrichment of genes associated with smooth muscle contractility in Fbn1^mgR/mgR mice.

A. Top enriched Kyoto Encyclopedia of Genes and Genomes (KEGG) pathways analysis of differentially expressed genes (DEGs) from RNAseq data obtained from ascending aortic sample between Fbn1^mgR/mgR mice treated with DT-061 versus Vehicle. B. Hallmark gene signature (MsigDB; as obtained by GSEA) shows the significantly enriched gene sets in aortic tissues sorted as in A. C. Representative result of the Gene Set Enrichment Analysis (GSEA) showing enrichment of the gene signature (obtained in B) for the Myogenesis, MTORC1, and PI3K-AKT-mTOR pathways.

Activation of PP2A suppresses mTOR signaling and promotes the upregulation of VSMC contractile markers both in cell culture and in vivo

To confirm the involvement of PP2A-mediated downregulation of mTOR signaling and the enrichment of VSMC contractile markers, the ascending aorta from Fbn1^mgR/mgR mice treated with either DT-061 or vehicle were collected and immunoblot analysis was performed. As shown in Figure 6A, increased levels of mTOR phosphorylation (mgR 2.54±0.21 vs. mgR+DT-061 1.93±0.14, p=0.0039), p-p70-S6K (mgR 1.68±0.17 vs. mgR+DT-061 0.57±0.06, p<0.001), p-S6 (mgR 11.00±1.09 vs. mgR+DT-061 1.61±0.31, p<0.001), and total S6K (mgR 2.01±0.19 vs. mgR+DT-061 1.12±0.21, p=0.006) in Fbn1^mgR/mgR mice were repressed by DT-061 treatment. Conversely, the expression of VSMC contractile markers including SM-MHC (mgR 0.64±0.06 vs. mgR+DT-061 0.80±0.04, p=0.0369), calponin1 (mgR 0.49±0.13 vs. mgR+DT-061 1.07±0.04, p=0.0003), and sm22α (mgR 0.68±0.09 vs. mgR+DT-061 1.06±0.04, p=0.0009) in Fbn1^mgR/mgR mice were augmented by PP2A modulation (Figure 6A, Figure S14). Increased PP2A activity as measured by PP2A C subunit methylation was also noted with DT-061 treatment (Figure 6A, Figure S14).

Figure 6. Activation of PP2A suppresses mTOR signaling and promotes the upregulation of VSMC contractile markers both in vivo and in vitro.

A. Western blot analysis of protein levels in ascending aorta samples of WT and Fbn1^mgR/mgR (mgR) mice treated with either vehicle or DT-061. n=3-6. B-C. Protein levels of HASMCs treated with DT-061 (5 μM, pretreatment 30-min) and PDGF (30 ng/ml, 30-min), n=4 (B) or PDGF, n=3 (30 ng/ml, 24-hour) (C). Membranes were probed with the indicated antibodies to detect changes in the expression of different subunits of PP2A (A, B56a, C), active PP2A (methyl-PP2A-C), VSMC contractile markers, and the mTOR pathway. HSP90, Vinculin, and α-actin were used as loading controls. Quantifications are shown in Figure S14-15.

HASMCs were then used to validate affected molecular pathways in a cell-based assay. Incubation of HASMCs with PDGF-BB enhanced mTORC1 signaling as demonstrated by the hyperphosphorylation of mTOR (2.906-fold, p=0.006) and its target signaling molecules p70S6K (1.917-fold, p=0.05), S6 (1.684-fold, p=0.001), and 4E-BP1 (1.793-fold, p<0.001). However, treatment of HASMCs with DT-061 strongly reduced phosphorylation of mTOR (0.534-fold, p=0.02), p70S6K (0.430-fold, p=0.03), S6K (0.580-fold, p=0.001), and 4E-BP1 (0.859-fold, p=0.05) as compared to control DMSO treated cells (Figure 6B, Figure S15A). Additionally, DT-061 treatment promoted the expression VSMC contractile markers and alleviated the loss of SM-MHC (PDGF-BB 0.39±0.06 vs. PDGF-BB+DT-061 1.36±0.05, p<0.001), Calponin1 (PDGF-BB 0.10±0.03 vs. PDGF-BB+DT-061 0.44±0.04, p=0.003), SMA (PDGF-BB 0.07±0.01 vs. PDGF-BB+DT-061 0.55±0.09, p<0.001) and SM22α (PDGF-BB 0.26±0.02 vs. PDGF-BB+DT-061 0.70±0.02, p=0.0082) induced by PDGF-BB (Figure 6C, Figure S15B). Altogether, DT-061 treatment correlates with PP2A activation, induces downregulation of the mTORC signaling pathway and increases the expression of VSMC markers.

Discussion

VSMCs exhibit remarkable plasticity in various vascular diseases. Current paradigms indicate that alterations in VSMC function as well as transdifferentiation of VSMC populations play important causal roles in aneurysm formation and aortic dissection. Despite our growing understanding of VSMC biology and the mechanistic basis for thoracic aortic aneurysm and dissection (TAAD) etiology, identifying and developing novel agents to prevent (inhibit) pathological VSMC transformation or revert the modulated VSMC to a more physiological contractile phenotype remains an unmet need. Towards this end, our current studies have identified that small molecule activators of PP2A can mitigate VSMC dedifferentiation in TAAD. The central finding of our study is that PP2A activation limits TAAD in two mouse models of MFS. In both MFS patients and mice, there is a marked reduction of PP2A activity. Most notably, in both widely utilized murine MFS models (Fbn1^mgR/mgR and Fbn1^1039G/+), we show that PP2A activation by DT-061 strongly suppresses thoracic aortic aneurysm progression. Furthermore, in Fbn1^mgR/mgR mice, PP2A modulation also prevents thoracic aortic dissection and reduces mortality. At the mechanistic level, PP2A maintains VSMCs in their differentiated state, a critically important action that involves suppressing mTOR signaling, a key pathway driving VSMC phenotypic transformation in TAAD in both MFS and non-MFS.

Protein phosphorylation is a critical form of translational modification that plays an essential regulatory role in regulating a wide range of cellular functions in health and disease. Dysregulated protein phosphorylation has been linked to the development of numerous diseases. PP2A, a critical serine/threonine phosphatase, has emerged as a vital regulator of cardiovascular biology ^11,17,30. Prior genetic and pharmacological studies have implicated a central role for dysregulated PP2A activity as a driver of cardiac and vascular diseases ^17,31-33. In this study, we aimed to establish a correlation between PP2A function and Marfan TAA by analyzing PP2A expression and its carboxymethylation status, a key regulator of the PP2A-B regulatory subunit in aortic tissues from both human patients and mouse MFS models. Our results reveal a strong inverse relationship between TAA and PP2A methylation, suggesting that compromised PP2A activity — specifically, the loss of binding of the growth-suppressive PP2A-B regulatory subunits — may drive the development and progression of TAA. To corroborate these observations, we analyzed the MFS scRNA-seq dataset to gain insight into ppp2ca expression (the main catalytic subunit isoform of the PP2A holoenzyme) in VSMCs at the single cell level. A dramatic reduction of ppp2ca-expressing VSMCs was observed in MFS. Furthermore, enrichment pathway analysis of DEGs pointed to compromised VSMC contractility, actin cytoskeleton integrity, ECM organization, and elastin fiber formation in ppp2ca-expressing cells. These observations, in conjunction with the significantly reduced activity of several transcription factors critical for VSMC differentiation, strongly support a potential nodal role for PP2A in maintaining VSMC contractile function.

The mTOR (mechanistic target of rapamycin) signaling pathway is intricately involved in modulating the VSMC phenotype, with growing evidence implicating its aberrant activation in the pathogenesis of TAA and TAAD ^34-37. mTORC1 activation in VSMCs promotes phenotypic changes characterized by increased proliferation, migration, and production of extracellular matrix proteins (such as collagen and elastin), while simultaneously downregulating the VSMC contractile machinery ^38,39. These alterations in the VSMC phenotype disrupt the delicate balance between vessel wall remodeling and stability, leading to weakening of the aortic wall and predisposing it to aneurysm formation and dissection. This is exemplified by elegant studies in Tuberous sclerosis complex (Tsc) deficient mice (Tsc1^−/−), which present with TAAD, a phenotype reversed by mTORC1 inhibition. Mitigation of hyperactivation of mTORC1 in Tsc1^−/− rescues the abnormal VSMC phenotype, findings consistent with what is observed in Tsc2^+/− mice ⁴⁰, which showed neointima formation in a carotid artery injury model. Additionally, pharmacological mTORC1 inhibition by rapamycin has been shown to prevent aortic rupture and pseudoaneurysm formation in AngII-infused Apoe-null mice ⁴¹. Collectively, the existing literature supports a role for targeting mTORC1 to modulate the VSMC phenotype as a therapeutic strategy for preventing or attenuating the progression of aortic aneurysm and dissection. However, effective pharmacological agents that modulate mTORC1 signaling have remained inadequately explored.

Toward that end, the antiparallel changes in PP2A and mTOR signaling in MFS TAA are very intriguing (Figure 1&2), both from a scientific and therapeutic perspective. The inverse functional relationship between PP2A and mTOR has been implicated in a variety of pathophysiological processes across different disease states and cell types, with strong evidence for dysregulation of this axis in cancer biology ^42,43. The strong functional control of mTOR by PP2A lead us to postulate that a similar mechanism might be operative in TAA. Considering the well-documented role of mTOR in driving VSMC de-differentiation, we posited that activation of PP2A may downregulate mTOR signaling, thereby promoting a differentiated VSMC phenotype. Our hypothesis was substantiated by robust in vitro and in vivo data presented in this manuscript. First, as shown in Figure 5, RNA-seq of ascending aortic tissues revealed that, in contrast to the strong enrichment of myogenesis seen in DT-061-treated TAA tissues, mTORC1 signaling and PI3K-Akt-mTOR were highly enriched in the control group. Second, in MFS ascending aortic tissues, the reduction of SMC contractile proteins (SMMHC, CNN1, α–SMA), was reversed by DT-061. PP2A modulation was associated with potent inhibition of mTOR activation (p-mTOR), as well as its downstream effector p-p70 S6K and p-S6. Third, inhibition of mTOR signaling by treatment with DT-061 blocked PDGF-induced dedifferentiation of HASMCs. Together, our results strongly implicate that dampening of mTOR activity accounts for the pro-contractile action of PP2A in VSMCs.

An important aspect of these current studies was to investigate the biological effects of modulation of PP2A function using pharmaceutically tractable SMAPs in both male and female TAA models. It is well appreciated that sexual dimorphism affects the incidence and complications of AAD in TAA ⁴⁴. While TAA statistically affects more men than women, with AADs (both type A and B) seen approximately twice as frequently in men, studies have shown that TAA growth is accelerated in women and poses a higher risk for acute aortic syndrome, including earlier or more extensive dissections ⁴⁵. Unfortunately, in preclinical animal studies, most previous work primarily focused on male models of the disease ^46,47. In light of these observations, we sought to rigorously address the mechanisms underlying sex as a biological variable in AAD and to explore the therapeutic potential of SMAPs in treating MFS TAA. To this end, we conducted our studies in both male and female animals. The results of our studies are illuminating. In the two mouse models of MFS, oral administration of SMAP strongly slowed aneurysmal growth. Additionally, SMAPs markedly reduced aortic dissection in Fbn1^mgR/mgR model in both male and female mice. These benefits on aneurysm growth and progression were corroborated by the improvement of mouse kyphosis – a symptom associated with aortic manifestation of the disease in MFS patients and mice. Although the exact influence of sex hormones on the maintenance of vascular wall homeostasis and the underlying mechanisms involved in these observed phenotypes remain largely unknown, our results offer proof of principle evidence that SMAPs possess a unique ability to retard aneurysm progression in MFS, potentially via a unifying mechanism of action irrespective of sex hormone function.

Another highlight of the current work is that the inhibitory effect of PP2A activation is manifested in two preclinical MFS models (Fbn1^mgR/mgR and Fbn1^1039G/+). The hypomorphic Fbn1^mgR/mgR model recapitulates the severe phenotype of MFS (profound aortic dilation) resulting in a reduced lifespan due to spontaneous aortic rupture (within three months after birth) similar to what is seen in human MFS ^48,49. Thus, this model more faithfully replicates the natural progression of TAA initiation to dissection observed in human MFS. With this in mind, the bulk of our studies were performed using this model. Most importantly, our results demonstrated that this model is well suited for testing whether PP2A activation mitigates TAAD. We were able to substantiate our hypothesis that administration of a PP2A activator prolongs animal survival and reduces mortality due to aortic rupture. Another genetic model (Fbn1^1039G/+), which harbours a p.Cys1041Gly mutation in the Fbn1 gene, has also been utilized in MFS aneurysm studies. This model causes classic manifestations of Marfan syndrome in humans (Cys1039Tyr) and presents proximal aortic aneurysm, however, it rarely develops aortic dissection, hence limiting its utility to study TAAD ⁴⁹. Nevertheless, to ascertain if pharmacological modulation of PP2A is effective in limiting aneurysmal progression, we conducted similar studies in this model and monitored the impact of SMAP treatment on aortic root growth. Collectively, our data from two distinct MFS models unequivocally support a role for PP2A modulation in protecting against aortic aneurysm and dissection, and favours follow-up investigation to test their effects in animals with established TAAD.

Our current studies open exciting opportunities for investigating PP2A’s role and its underlying mechanisms in TAAD and for developing new therapies. Our study clearly demonstrated that SMAP (DT-061) prevents TAAD, prompting future therapy studies in animals with established TAAD. Regarding VSMC phenotype alterations in TAAD, increasing evidence from lineage tracing and single cell RNAseq studies indicates a diverse array of VSMC fates ^29,50-53. Future research is needed to explore DT-061's effects on VSMC subpopulations. We understand that oral administration of DT-061 targets multiple cell types in vivo. Endothelial cells (ECs) and VSMCs in the vasculature work together as an interactive unit to sustain normal vascular function via paracrine mediators. Importantly, growing evidence points to the importance of endothelial dysfunction in the pathological remodelling of the aortic media ^54-57. Our human VSMC culture data suggest that DT-061's benefits are cell-autonomous, highlighting VSMC as a major target. However, DT-061's influence on the endothelium may also preserve the differentiated VSMC phenotype. Future inquiries will address PP2A's regulation of the EC-VSMC unit and how PP2A activation affects their interaction. Finally, we acknowledge that the limited MFS patient samples in this study is an inherent limitation due to the paucity of accessible MFS samples for detailed molecular analysis. To clearly correlate changes in PP2A activity with MFS TAA, a well-powered assessment of MFS specimens from both sexes is necessary.

In summary, our studies demonstrate the role of PP2A dysregulation in the etiology of MFS and implicate PP2A activation as a novel therapeutic strategy to limit the progression of human MFS, including the development and rupture of aortic aneurysms.

Perspectives

Thoracic aortic aneurysms and dissections (TAAD), while often asymptomatic, are one of the most high-risk cardiovascular diseases, causing premature mortality due to sudden and life-threatening aortic rupture. Unfortunately, despite our increased understanding of this disease, there is a sobering reality that effective pharmacological approaches for the prevention and treatment of TAAD remain a critically unmet need. Our current findings illuminate the efficacy of SMAPs in slowing the progression of TAAD in preclinical mouse models of MFS. Notably, our studies revealed critical mechanistic insights into the action of PP2A in maintaining VSMC homeostasis and normal vessel function. Although MFS is the only genetically predisposed TAAD model used in this study, given the central importance of VSMCs in aortic health, as their dysfunction is intimately causal to genetic and non-genetic forms (sporadic) of TAAD, the fact that pharmacological activation of PP2A suppresses pathological VSMC phenotypic alterations implicate that targeting PP2A has broad implications not only for future research into PP2A biology but also for the design of new therapies for disease states in which vascular homeostasis is compromised.

Novelty and Relevance.

What is new?

• Protein Phosphatase 2A (PP2A) activity is strongly reduced in thoracic aortic aneurysms (TAA) in Marfan syndrome (MFS).

• Small-molecule activators of PP2A (SMAPs) strongly suppresses the progression of thoracic aortic aneurysm and prevents aortic dissection in two MFS mouse models.

• Inhibition of mTOR signaling and preservation of vascular smooth muscle cells (VSMC) in the differentiated state account for the beneficial effects of PP2A activation on thoracic AAD progression.

What is Relevant?

• Optimal PP2A activity is required for normal vascular homeostasis, while compromised PP2A activity renders aortas susceptible to TAAD.

• Loss of VSMC contractility and the attendant aortic dysfunction are hallmark features in TAAD pathology. PP2A activation prevents aberrant VSMC phenotypic alterations which contributes to its therapeutic potential in slowing down the advancement of TAAD.

• Remarkable improvement of skeletal deformations by PP2A activation highlights the broad benefit of drugs targeting PP2A besides the vascular system.

Clinical/Pathophysiological Implications?

Deficiency of PP2A activity represents a novel pathophysiological mechanism of TAAD in MFS. Activation of PP2A holds promise as a novel therapeutic approach to prevent and alleviate aortic malfunction in TAAD and potentially other pathological manifestations seen in MFS.

Non-standard Abbreviations and Acronyms

TAA

Thoracic aortic aneurysm

PP2A

Protein phosphatase 2A

SMAPs

Small molecule activators of PP2A

MFS

Marfan Syndrome

VSMC

Vascular smooth muscle cell

AAA

Abdominal aortic aneurysm

Apoe

Apolipoprotein E

DEGs

Differentially expressed genes

GEO

Gene expression omnibus

WT

Wild-type

scRNA-seq

Single cell RNA-seq

UMAP

Uniform manifold approximation and projection

GSEA

Gene set enrichment analysis

FDR

False discovery rate

IACUC

Institutional animal care and use committee

NIH

National institute of health

BP

Blood pressure

EVG

Verhoeff van gieson

MMPs

Metalloproteinases

KR

Kyphosis ratio

HASMCs

Human aortic VSMCs

RNA-seq

Transcriptome sequencing

GO

Gene ontology

KEGG

Kyoto encyclopedia of genes and genomes

SEM

Standard error of the mean

TAAD

Thoracic aortic aneurysm and dissection

ECM

Extracellular matrix

TSC

Tuberous sclerosis complex

EC

endothelial cells