Efficacy of Antihypertensive Drugs and miR-632 Inhibition on Parietal Remodeling in a Model of Marfan Thoracic Aortic Aneurysm

1. Introduction

Marfan syndrome (MFS) is a genetic condition caused by mutations in the FBN1 gene coding for the extracellular glycoprotein fibrillin-1 [1]. MFS is associated with a high risk of early thoracic aortic aneurysm (TAA) development, in particular in the aortic root [1]. FBN1 mutation leads to Transforming Growth Factor β (TGF-β) signaling hyper-activation with early vascular wall weakness and TAA onset [2]. Despite clinical manifestations can be heterogeneous, the most life-threatening complications in MFS patients occur in the cardiovascular district with mitral valve prolapse, aortic insufficiency, aortic root dilatation and TAA dissection [3]. MFS TAA is characterized by early and severe degeneration of the aortic tunica media with disruption of extracellular matrix (ECM) homeostasis [4,5]. In particular, TGFβ hyperactivation promotes ECM remodeling through the induction of matrix metalloproteinases (MMPs), particularly MMP-9, and alterations in their endogenous inhibitors such as TIMP-1 with altered MMP-9/TIMP-1 balance with excessive degradation of structural ECM components, including elastin and collagen, thereby contributing to the progressive weakening and aneurysmatic dilation of the aortic wall [4,6].

The introduction of surgery and the use of antihypertensive drugs proved effective in the management of MFS patients [7]. In particular, recent advances in surgical and therapy improved life expectancy in MFS patients [7]. Initially, the standard-of-care for MFS patients was β-blockers (e.g., carvedilol, bisoprolol); the latter regulate systemic blood pressure and cardiac pulse wave reducing vascular tension and, consequently, the risk of root dilatation and TAA rupture [7]. Angiotensin converting enzyme inhibitors (ACEi) play an important role in the development of aortic stiffening, dilatation and rupture in MFS patients [8]. Recent advances suggest that the selective inhibition of angiotensin II type 1 receptors (AT1R) through Angiotensin Receptor Blockers (ARBs) is one of the most appropriate treatment [9]. Among ARBs, losartan is reported to prevent aortic dilation in MFS patients better than other antihypertensive drugs [10]. In some therapeutic protocols, losartan is administered in combination with β-blockers and, under medical control, also with ACE inhibitor [9,11]. Despite the recent improvement of the pharmacological strategies, MFS patients still undergo to early surgery. So, it is mandatory to find additional or more specific therapeutic approaches to prevent TAA progression in MFS patients. MFS TAA is characterized by an early and increased aortic remodelling compared with non-MFS ones [12]. We reported that a specific miR-632 upregulation in MFS TAA tissue and blood, induced by TGF-β signaling hyperactivation, contributes to severe and early degeneration of the aortic wall [12,13]. In particular, miR-632 deregulation induced vascular smooth muscle cells (VSMCs) dedifferentiation into myofibroblasts, which express extra domain A fibronectin (ED-A FN) and produce excessive collagen, leading to collagen accumulation and aortic fibrosis [13].

In the present study, we investigated the effect of antihypertensive drugs (β-blocker, ACE inhibitor and ARB) and a specific miR-632 inhibitor on the TGFinduced degenerative process of the aortic wall in a MFS TAA ex vivo model.

2. Materials and Methods

2.1. Patients and Aortic Sample Collection

For this study, we collected aortic tissue samples from non-MFS TAA patients undergoing elective surgical procedures in the Department of Cardiosurgery of Azienda Ospedaliera Universitaria Policlinico Tor Vergata of Rome. Eligibility criteria used were the follow: patients aged 30–70 years with TAAs (average of aortic diameter=53 mm), without coronary artery disease, endocarditis, aortic dissection, renal failure, liver disease, or tumor, and with a good ejection fraction. The diameter evaluation of ascending aorta was made both preoperatively and in the operating room by transthoracic echocardiography, and transesophageal echocardiography estimations were performed as follows: estimating the dimensions of the aortic annulus, the sinuses of Valsalva, and the proximal ascending aorta (above 2.5 cm of the sinotubular junction) in the parasternal long-axis view and evaluating the dimensions of the aortic arch from the suprasternal view (echocardiography-derived sizes were reported as internal diameter size [14]. Color Doppler was used to assess the presence and severity of aortic regurgitation and stenosis. The measurement of the aortic root and ascending aorta diameters was also carried out using helical computed tomography image analysis [14]. The surgical procedure was mostly the isolated ascending aorta replacement or the button Bentall operation, a modification of the original technique described by Kouchoukos et al. [14]. All operations were performed using crystalloid or hematic cardioplegia. The Territorial Ethics Committee approved the study (Protocol “PNRR-MR1-2022-12376699”) and all patients signed the informed consent form.

2.2. Ex Vivo Transfection of TAA Tissues with miR-632 Inhibitor

After surgery, small aortic samples were immediately used for experiments. Before proceeding with experiments adventitia tunica was removed and endothelium was scraped, as reported [15]. Then, small fragments of fresh aortic tunica media from six non-MFS TAA patients, were adhered to the bottom of a 24- or 6-well multiwall [13]. Tissue samples were primed with TGF-β1 (10 ng/mL; Sigma Aldrich, St. Louis, MO, USA) for 24 hours [13]. After 24 hours, TGF-β1-enriched medium was removed and tissue fragments transfected with miRCURY locked nucleic acid (LNA)-anti-miR-632 (100nM) and scrambled LNA (negative controls) (Qiagen, Hilden, Germany). For RNA and protein extraction, samples were collected in four pools (two pools for scrambled LNA and two for LNA-anti-miR-632, each pool=3 patients) 24h and 72 h after transfection, respectively.

2.3. Ex Vivo Treatment of TAA Tissues with Antihypertensive Drugs

As reported above, small tunica media fragments of fresh aortic tissue samples from six different non-MFS TAA patients were adhered to the bottom of a 24- or 6-well multiwall, ex-vivo pre-treated TGFβ1 (with 10 ng/mL) for 24 hours and after 24 hours TGFβ1–enriched medium was removed, and fragments treated with a non-selective β-blocker (carvedilol), the ACE inhibitor (ramipril) and ARBs (losartan), alone or in combination using a concentration of 10µM, as reported [16,17,18]. For RNA and protein extraction, samples were collected in 2 pools (each pool=3 patients) for each treatment after 24 and 72 h, respectively. Experiments were performed in triplicate.

2.4. Gene Expression Analysis

RNA and miRNA extraction was performed using the TRI Reagent^® (Sigma Aldrich, St. Louis, MI, USA) according to manufacturer’s protocols, as reported [19]. A Micro-Volume spectrophotometer (EZDROP 1000 Micro-Volume spectrophotometer, Resnova, Roma, Italy) was used to measure the RNA concentration. Then, 700 ng of RNA/miRNA were reverse-transcribed using the microScript microRNA cDNA Synthesis Kit (Norgen Biotek, Thorold, Canada) or the SuperScript III (Invitrogen, Thermo Fisher Scientific, Waltham, MA, USA). Real-time PCR was carried out by using SYBR Green (BioRad, Hercules, CA, USA). We used MystiCq microRNA Primer HSA-miR-632 (Merck KGaA, Darmstadt, Germany) and Universal PCR Reverse Primer (Norgen Biotek) were used to analyze miR-632 expression. For gene expression analysis, cDNAs were amplified using specific primers for ED-A FN. HSA-miR-632 was normalized to the U6 expression, while other genes were normalized to the GAPDH expression (see Supplemental Table S1). Changes in target gene expression were calculated using the comparative ΔΔCT method [20]. Fold change was considered significant for values > 2.0 and < 0.5. Analyses were performed in triplicate.

2.5. Western Blot Analysis

Total protein extraction was performed on non-MFS TAA tunica media treated/transfected [13]. Briefly, tissues were homogenized and resuspended in RIPA buffer (Sigma Aldrich) with protease inhibitor cocktail (Sigma Aldrich) and phosphatase inhibitors (Sigma Aldrich). Protein concentration was determined by a spectrophotometric assay (BioRad). 70 μg of protein were separated using 8% polyacrylamide gel and then transferred from the gel to a filter. The latter was incubated with specific mouse monoclonal anti-ED-A FN antibody (1:200, Santa Cruz Biotechnology), Rabbit Polyclonal anti-MMP-9 (1:300, Proteintech Rosemont, Illinois USA), Rabbit Monoclonal anti-TIMP-1 (1:200, Abcam, Cambridge, UK). Normalization was performed using mouse polyclonal anti-α-tubulin (1:500,Sigma Aldrich). Detection and quantification were carried out using the ECL Reagent (BioRad) and LAS-4000 (Fujifilm, Tokyo, Japan). The intensity of the bands was analyzed by Image J software (National Institutes of Health, NIH, USA).

2.6. Statistical Analysis

All results were reported as the mean ± standard error of mean (SEM). We analyzed normally distributed continuous and quantitative variables between different independent experimental groups. Differences were calculated by using the unpaired t-test and considered statistically significant for p value < 0.05. Statistical analyses were performed with SPSS software (IBM, SPSS Statistics, Chicago, IL, USA. version 23).

3. Results

3.1. Losartan Is the Most Effective in Reducing miR-632 Levels and Myofibroblastic Dedifferentiation in TGFβ1-Treated TAA Tunica Media

In order to analyze the effects of antihypertensive drugs on miR-632 expression and on VSMC phenotypic switch into myofibroblasts, non-MFS TA, we pre-treated tunica media tissue fragments with TGFβ1 that induces miR-632 up-regulation and represents a well-established model of MFS TAA [13]. TGFβ1−pretreated TAA fragments were successively treated with antihypertensive drugs: carvedilol, ramipril and losartan. Gene expression analysis revealed that losartan was the most effective treatment, significantly downregulating miR-632 levels (p< 0.01 vs TGFβ1; Figure 1A) and demonstrated a potent inhibitory effect on ED-A FN at both transcriptional and protein levels (p< 0.001 vs TGF1; Figure 1B-D). In comparison, ramipril showed a significant inhibitory effect only on miR-632 and ED-A FN mRNA expression but less than losartan (p< 0.01 vs TGFβ1; Figure 1 A,B) whereas its impact was lost at the protein level (Figure 1 C,D). Conversely, carvedilol failed to provide protection against TGFβ−induced miR-632 and ED-A FN overexpression, instead exacerbating the pro-fibrotic response (p< 0.05).

3.2. Combined Antihypertensive Treatment Reduces Losartan Inhibition of miR-632 and ED-A FN in TAA Tissues

Since losartan can be administered to MFS patients in combination with β-blocker and, in some cases, with ACEi (under medical supervision), we analyzed the effects of those combinations on miR-632 levels and ED-A FN expression in ex-vivo TGFβ1-stimulated TAA tunica media tissues. Our results highlighted that the potent inhibitory effect of losartan is notably attenuated when used in combination with other drugs. Specifically, the co-administration of carvedilol and losartan did not significantly reduced miR-632 and ED-A FN levels compared to losartan monotreatment (p<0.05 and p<0.01, respectively; Figure 2 A-D). Similarly, although the ramipril+losartan group showed lower expression levels than TGFβ1 control (p< 0.05 and p< 0.01), its efficacy was less than losartan alone (p< 0.05 and p< 0.01). Overall, those findings suggest that combining these pharmacological treatments does not provide a synergistic benefit; rather, it compromises the maximal efficacy of losartan on miR-632 and myofibroblastic dedifferentiation in this ex vivo TGFβ−treated ΤAA model.

3.3. Antihypertensive Drugs Differently Impact MMP-9 and TIMP-1 Expression in TGFβ−Treated ΤAA Tunica Media

Since MFS TAA is characterized by extensive tissue remodeling, we evaluated the impact of different antihypertensive treatments on key remodeling markers, specifically MMP-9 and TIMP-1.As shown in Figure 3 A,B in TGFβ−treated TAA tunica media fragments, carvedilol significantly upregulated MMP-9 and TIMP-1 expression (p< 0.05). Conversely, losartan induced a slight, though non-significant, reduction in MMP-9 levels. Despite these individual changes, the MMP-9/TIMP-1 ratio remained unaffected by any of the treatments. Furthermore, we investigated whether combined treatments of losartan with either carvedilol or ramipril) influenced those markers. As reported in Figure 4, the combined treatment of losartan with ramipril led to a significant increase in MMP-9 protein expression compared to TGFβ1 and losartan monotreatment (p< 0.05; Figure 4), which a consequent significant higher MMP-9/TIMP-1 ratio. In contrast, the addition of carvedilol did not induce significant changes in the expression of MMP-9 or TIMP-1, nor did it alter their relative ratio.

3.4. miR-632 Specific Inhibition Is the Most Effective in Reducing Aortic Wall Degeneration Markers in TGFβ−Treated ΤAA Tunica Media

Following our comparative analysis of antihypertensive agents, we further evaluated the efficacy of losartan, the clinical standard for MFS patients, in comparison with a specific miR-632 inhibitor. In TGFβ−treated TAA tunica media fragments, losartan treatment successfully reduced the expression of miR-632 and ED-A FN at both the transcriptional and protein levels (p< 0.01 vs TGFβ1; Figure 5 A-D). However, the specific inhibition of miR-632 resulted in a more pronounced downregulation of those markers (p< 0.001 vs TGFβ1).Regarding the proteolytic balance, while losartan maintained it unaltered (Figure 6 A,B), miR-632 inhibitor exerted a stronger impact by reducing MMP-9 and upregulating TIMP-1 protein expression (p< 0.01; Figure 6 A,B), thereby strongly decreasing the MMP-9/TIMP-1 ratio (p< 0.001; Figure 6 A,B).

4. Discussion

The pharmacological management of MFS TAA primarily focuses on mitigating the hemodynamic and molecular triggers of aortic wall remodeling. Our findings provide a detailed comparison of how conventional antihypertensive agents—carvedilol, ramipril, and losartan—and novel epigenetic targets, such as miR-632, modulate ECM degradation that characterize MFS TAA. Previous studies from our group demonstrated that miR-632 expression is significantly upregulated in both TAA tissue and blood of MFS patients and plays a key role in promoting aortic wall degeneration [13,19]. In particular, in MFS aorta, TGFβ hyperactivation induced miR-632 upregulation leading to aortic fibrosis, ECM disorganization and progressive weakening of the aortic wall [13]. In this context, our findings demonstrated that among conventional antihypertensive drugs, losartan, used as a monotreatment, is the most effective in reducing the fibrotic markers (i.e., miR-632 expression and ED-A FN). Those findings support the hypothesis of a complex molecular crosstalk between the Renin-Angiotensin System and the TGFβ pathway. In fact, previous reports demonstrated that angiotensin II type 1 receptor blockade by losartan attenuates TGFβ signaling, a central pathway in MFS aortopathy pathogenesis [21,22,23]. Indeed, losartan was shown to inhibit both canonical and non-canonical TGF-β pathways, thereby limiting downstream profibrotic responses [24,25]. Regarding the effects of losartan on the proteolytic profile, our results suggest also role in maintaining ECM homeostasis rather than inducing a robust inhibition of remodeling markers. In our experimental model, losartan treatment showed a slight trend toward reducing MMP-9 levels, while leaving the overall MMP-9/TIMP-1 ratio substantially unchanged. Those findings suggest that while losartan, through its blockade of the AT1 receptor and partial modulation of the TGFβ pathway, contributes to the stabilization of the tunica media, it does not exert a predominant effect on the proteolytic balance.

In contrast, ramipril, acting upstream as an ACE inhibitor, may fail to completely prevent local Angiotensin II synthesis likely due to alternative pathways like chymases [26]), explaining its more modest effect on miR-632 and ED-A FN expression failing to translate this reduction to the protein level. Regarding the proteolytic profile, ramipril monotreatment did not significantly modulate MMP-9 or TIMP-1 expression. In our experimental conditions, the protein levels remained largely stationary, with no detectable shift in the MMP-9/TIMP-1 ratio. This lack of effect suggests that, in this specific MFS TAA model, ACE inhibition alone is insufficient to alter the proteolytic profile of the tunica media. This observation is consistent with the presence of alternative local pathways for Angiotensin II synthesis, such as chymases as already reported[26], which may render the proteolytic markers unresponsive to ramipril treatment, thereby maintaining the remodeling markers at their baseline levels. As concerning the effects of carvedilol we did not documented a significant modulation of miR-632 levels compared to the TGFβ1-treated control. However, despite this lack of epigenetic regulation, we observed that carvedilol promoted a further increase in ED-A FN expression. This discrepancy suggests that carvedilol may negative influence aortic remodeling through pathways independent of miR-632. The upregulation of ED-A FN by a non-selective beta-blocker could be linked to its complex interaction with the TGFβ-signaling axis. While carvedilol is clinically established for its role in reducing hemodynamic stress, several studies provide conflicting evidence regarding the anti-fibrotic potential of β-blockers. Some studies suggest that β-adrenergic ligands may paradoxically influence extracellular matrix deposition under specific pathological conditions [27,28]. Our observations align with these discrepancies: while some research highlights the protective effects of β-blockade [29], others report minimal or no impact depending on the experimental model utilized [30]. Notably, to the best of our knowledge, no previous studies have specifically addressed the influence of carvedilol on VSMC dedifferentiation in human aorta, making our findings a significant contribution to understanding its molecular limitations in MFS TAA. As concerning the proteolytic profile, carvedilol induced a significant upregulation of both MMP-9 and TIMP-1 protein expression, suggesting that it could trigger a broad activation of the ECM regulatory system, which nevertheless maintained a stable MMP-9/TIMP-1 ratio. This finding aligns with conflicting evidence in the literature already reported above. Given its non-selective nature, further studies are required to evaluate whether more selective β-blockers might exert different effects on those molecular pathways.

To further explore the potential of pharmacological synergy, we investigated in our model the effects of combined antihypertensive treatments. While ramipril and carvedilol monotherapies showed limited efficacy, their combination with losartan significantly improved the outcome by effectively reducing ED-A FN levels. Those results strongly support that the addition of losartan is essential to achieve a modulation of this marker. In relation to the proteolytic profile, our results demonstrate that losartan modulates remodeling markers differently depending on the specific drug combination. Specifically, when losartan was combined with carvedilol, we observed a notable reduction in both MMP-9 and TIMP-1 levels compared to carvedilol monotreatment. This stabilization likely results from the complementary action of these drugs: while carvedilol reduces hemodynamic wall stress, losartan antagonizes the AT1R/TGFβ signaling axis, which is the primary driver of protease overexpression in MFS aortopathy [31]. This synergy suggests that losartan can ‘correct’ the broad matrix activation previously induced by beta-blockers alone, restoring a more favorable proteolytic environment.

In contrast, the combination of losartan and ramipril paradoxically led to a significant upregulation of MMP-9, increasing the MMP-9/TIMP-1 ratio. This pro-proteolytic shift following a total blockade of the Renin-Angiotensin System (RAS) has been documented in other vascular models and may be attributed to the ‘Angiotensin II escape’ or the redirection of signaling toward the AT2 receptor. Indeed, studies have shown that dual RAS inhibition can increase local bradykinin levels and AT2R activation, which in certain pathological contexts can stimulate, rather than inhibit, MMP activity [32]. Furthermore, clinical evidence from the ONTARGET trial has highlighted that combining ACE inhibitors and ARBs does not necessarily provide superior tissue protection and may even lead to adverse remodeling responses compared to monotreatment [33]. In our experimental model, the specific inhibition of miR-632 emerged as a new potent therapeutic strategy, consistently outperforming conventional pharmacological drugs. While losartan acts upstream by antagonizing the AT1 receptor and indirectly modulating TGFβ signaling, miR-632 inhibition directly targets downstream molecular effectors directly involved in VSMC phenotypic switching and ECM remodeling.

Notably, miR-632 inhibition demonstrated a greater efficacy compared to losartan, as it not only suppressed its own expression and ED-A FN levels more effectively but also significantly reduced MMP-9 and TIMP-1 protein expression, thereby restoring a favorable proteolytic balance. Those results suggest that miR-632 plays a central regulatory role in the MFS aortopathy, acting as a master downstream effector of the TGFβ pathway. Consequently, targeting miR-632 may represent a more specific and possible therapeutic approach than standard pharmacological therapies. It is important to note that our experiments were conducted on aneurysmal aortic tissues surgical excised. While our findings underscore the potency of miR-632 inhibition in advanced aneurysmatic disease, they do not preclude the possibility that losartan might exert more significant protective effects if administered during the earlier phases of aneurismatic progression of MFS aortopathy.

5. Conclusions

In conclusion, our study suggests losartan as the most effective conventional drug for reducing fibrotic markers in this ex vivo MFS model. Notably, our data also suggest that caution is required regarding combined pharmacological treatments since certain drug combinations may paradoxically attenuate losartan efficacy or negatively affect the aortic proteolytic balance. Furthermore, while the miR-632 inhibition showed a great ability to counteract aortic remodeling in an ex vivo model of MFS TAA, although further experimental validation is needed to evaluate its safety and actual therapeutic potential in vivo. Overall, those findings emphasize the importance of choosing targeted of therapeutic approach to optimize the management of aortic wall degeneration and delay the progression of thoracic aortic aneurysms in MFS patients.