Contribution of novel biomarkers to adverse cardiac remodelling and myocardial dysfunction in a rodent model of pressure overload

Abstract

Background: Hypertension drives the development of concentric left ventricular hypertrophy (LVH) and adverse cardiac remodelling. Concentric LVH is characterised by an increase in wall thickness relative to chamber size. La Place’s law posits that concentric LVH initially serves as an adaptive response to maintain optimal wall tension. Over time, the sustained pressure overload causes maladaptive remodelling, leading to a transition towards an eccentrically-remodelled phenotype marked by chamber dilation and thinning of the ventricular walls. This maladaptive remodelling involves complex molecular mechanisms, including novel biomarkers such as pentraxin-3 (PTX-3), microRNAs (miRNAs), and transcription factors, which may play important roles in these pathological changes. However, their functional roles and contribution to hypertensive cardiac remodelling and dysfunction remains elusive. Aim: In the present thesis, we elucidated the functional role of three novel biomarkers; namely, PTX-3, specific miRNAs, and the runt related transcription factor 1 (RUNX-1), in the development of hypertensive cardiac pathology. Specifically, we assessed the contribution of PTX-3 in the hypertrophic myocardium, and evaluated whether PTX-3 is amenable to modulation using two anti-hypertensive therapies with differential effects on the development of concentric LVH. Additionally, we sought to determine the patterns of expression and functional roles of three miRNAs of interest in the development of LVH, fibrosis, and overall myocardial function. Finally, we examined the role of RUNX-1 during the transition towards an eccentrically-remodelled phenotype, and explored its putative involvement in fibrotic and apoptotic mechanisms. Methods: Spontaneously hypertensive rats (SHRs) and normotensive Wistar Kyoto rats (WKY), were utilised to investigate hypertensive cardiac pathology at different stages. To determine the role of PTX-3 in the development of LVH, 3-month-old SHRs received either 20 mg/kg/day ix hydralazine (SHR+H, n=6), 40 mg/kg/day captopril (SHR+C, n=6), or plain gelatine cubes (untreated SHR, n=7) for four months. Measurements were conducted at 7-months-old when LVH was established. To investigate the role of miR-146a-5p, miR-155-5p, and miR-29b-5p in the development of LVH and cardiac dysfunction, seven-month-old SHR (n=7 male, n=10 female), and normotensive WKY (n=7 male, n=9 female) were used. Lastly, to investigate the role of expression of RUNX-1 and its putative involvement in fibrotic and apoptotic mechanisms, 3- month-old rats (n=9 SHR, and n=9 WKY) and 12-month-old rats (n=9 SHR, and n=9 WKY) were used. Different sets of rats were used for each sub-study. Blood pressure (BP) was measured via tail-cuff plethysmography. Cardiac geometry and function were assessed through M-mode and Doppler transthoracic echocardiography. Circulating levels of inflammatory markers, mRNA, and miRNA expressions were quantified using enzyme-linked immunosorbent assays (ELISA) and RT-PCR, respectively. Left ventricular (LV) collagen deposition, cardiomyocyte size, macrophage infiltration, and cardiomyocyte nuclei density were determined histologically. Results: At 7 months, circulating PTX-3 levels were elevated in untreated SHR compared to WKY; mean difference [95% CI], −0.51 [−0.87 to −0.15, P = 0.004], correlating positively with increased systolic blood pressure (SBP) (r = 0.722; P= 0.0001) but showing no independent associations with indices of concentric LVH: LVPWTd (r = 0.372; P = 0.097), normalized heart mass (r = 0.216; P = 0.565) and normalized LV mass (r = 0.343; P = 0.109). Treatment with captopril and hydralazine reduced circulating levels of PTX-3; mean difference [95% CI], 0.58 [0.20 to 0.96, P = 0.002] and 0.68 [0.30 to 1.06, P = 0.0004], with captopril exhibiting a more pronounced effect in reversing LVH (P = 0.011 for RWT). Despite these changes in plasma PTX- 3, its LV cardiomyocyte mRNA expression remained similar across groups (all P > 0.05). Concurrently, miR-146a-5p expression was elevated in the hypertrophic myocardium of SHR (P x < 0.0001), correlating with SBP (r = 0.7; P = 0.0004), normalised LV mass (r = 0.5; P = 0.011), RWT (r = 0.5; P = 0.029), collagen fraction area (r = 0.4; P = 0.027), and diastolic dysfunction (r = -0.5; P = 0.027 for e’). Myocardial macrophage infiltration was greater in SHR compared with WKY (P < 0.0001) but similar between male and female rats (P = 0.159). Perivascular collagen deposition was also greater in SHR compared with WKY (P <0.05) but similar between male and female rats (P = 0.280). When disaggregated by sex to delineate sexually dimorphic features, sample clustering was observed in the principal component analysis biplot and a permutational multivariate analysis of variance (PERMANOVA) test revealed highly significant sex-specific differences in PCA scores (Pillai’s Trace = 0.72, P < 0.001). In moderation analyses illustrating the relationship between circulating concentrations of VCAM-1 and myocardial macrophage infiltration while considering the moderating effects of miR-146a-5p and SBP, the two-way interaction term between VCAM-1 and miR-146a-5p was highly significant (β = -0.0038, SE = 0.0002, P < 0.001), indicating a strong negative association with myocardial macrophage infiltration. In contrast, the two-way interaction term between VCAM-1 and SBP did not reach statistical significance (β = 0.00001769, SE = 0.0000204, P = 0.386), indicating that the combined effect of VCAM-1 and SBP does not significantly influence macrophage infiltration. The Wald chi- squared test result with two degrees of freedom (χ2 = 299.8, P < 0.001) showed that the slopes of the interaction effects differ significantly from each other. At 3 months, SHRs exhibited greater normalised heart mass (P = 0.049), increased LV mass (P = 0.003), and elevated cardiomyocyte nuclei number (P = 0.001). By 12 months, SHRs showed reduced PWTd (P = 0.005), lower RWT (P = 0.012), and enlarged LVIDd (P = 0.007). Diastolic function parameters at 12 months included elevated E/e′ ratio (P = 0.020), reduced a’ velocity (P = 0.027), and decreased e′/a′ ratio (P = 0.006). Molecular analyses revealed increased MMP-9 expression (P = 0.033), reduced Bax expression (P xi = 0.019), reduced cleaved PARP-1 (P = 0.002), a trend toward a lower Bax/Bcl-2 ratio (P = 0.062), and elevated cleaved caspase-3 (P = 0.065). RUNX-1 expression declined with age (P = 0.0002) but increased with hypertension (P = 0.001). RUNX-1 expression correlated positively with normalised LV mass (r = 0.54, P < 0.0001), RWT (r = 0.54, P = 0.002), and Bax/Bcl-2 ratio (r = 0.44, P = 0.010), and negatively with LVIDd (r = -0.57, P = 0.001) and MMP9 expression (r = - 0.48, P = 0.013). Conclusions: Our study provides novel mechanistic insights into the molecular underpinnings of hypertensive cardiac remodelling. While PTX-3 likely reflects the localised inflammatory milieu associated with hypertension, it may not be directly involved in the development of LVH in SHR. Transcriptionally, miR-146a-5p appears to play a significant role in hypertension-induced LV remodelling. Specifically, miR-146a-5p emerged as a key counter-mechanism, modulating VCAM-1-mediated transendothelial macrophage infiltration into the hypertrophic myocardium, inflammation, and oxidative stress by enhancing IκBa and SOD2 expression. RUNX-1 expression was positively associated with markers of concentric cardiac remodeling. Over time, RUNX-1 expression progressively declined, coinciding with a transition from concentric to eccentric remodeling in SHR. This temporal decrease in RUNX-1 expression appears to be linked to the shift in remodeling phenotype observed in SHR. Consequently, modulating RUNX-1 activity could represent a therapeutic strategy to prevent the transition from compensated hypertrophy to heart failure. While our findings provide valuable insights into the expression patterns and potential roles of PTX-3, RUNX-1, and specific miRNAs in hypertensive cardiac remodeling, they remain associative in nature. Future studies employing targeted molecular interventions are necessary to establish definitive mechanistic and functional roles for these potential biomarkers.