From January 2019 to December 2020, 16 consecutive patients were enrolled in our Tertiary Center of Secondary Arterial Hypertension: 11 patients with APA and 5 patients with NFA. The diagnosis of PA was performed according to the Endocrine Society Guidelines (2). NFA was diagnosed in patients with adrenal incidentalomas and normal adrenal hormone secretion. Surgical resection was done if the NFA was larger than 6 cm or if the lesion enlarged by more than 20% (in addition to at least a 5-mm increase in maximum diameter) during a follow-up period of 12 months as suggested by the European Guidelines (15). Patients with confirmed biochemical and/or histopathological diagnoses of APA or NFA were included. Plasmatic cortisol, 24-h urinary cortisol, and 24-h urinary metanephrines were measured, and patients with cortisol and catecholamines overproduction were excluded from the study. Subjects with coronary artery disease (CAD), HF, valvular heart disease, pericardial disease, cerebrovascular disease, peripheral arterial disease, kidney or liver disease, or with a history of cerebrovascular or cardiovascular events were also excluded. Patients taking lipid-lowering drugs, hypoglycemic drugs, or corticosteroids were also excluded. All the study participants followed a regular dietary habit with a salt intake of 140 mEq/day and a potassium intake of 50–75 mEq/day. All patients had stable body weight for 6 months before the study. In order to determine hypertension-related organ damage, we recorded a transthoracic 2D-guided M-mode echocardiogram using commercially available equipment (Esaote MyLab Omega). Standard parasternal and apical views were obtained with patients lying in the left lateral decubitus position. Left ventricular mass (LVM) and LVM indexed for body surface area (LVMi) were calculated according to the American Society of Echocardiography (ASE) guidelines (16). A 24-h ambulatory blood pressure monitoring (ABPM) was performed using the oscillometric technique, which involves a portable lightweight, non-invasive monitor with a self-insufflating cuff (Spacelabs Medical, 90207, Issequah, WA, USA). ABPM readings were obtained at 15-min intervals from 6 A.M. to midnight and at 30-min intervals from midnight to 6 A.M. All patients underwent laparoscopic adrenalectomy by the same operator. The postoperative course was uneventful for all patients. During surgery, a small fragment (<5 mm) of AT was obtained from 2 different depots for each site: (a) surrounding adrenal neoplasia, (b) peri-renal, and (c) subcutaneous depots. Real-time quantitative polymerase chain reaction (PCR) for miR-132, miR-143, miR-221, miR-26a, and miR-26b was performed. The selection of miRNAs was based on data reported in previous studies (14, 17–24). After 1-year follow-up after adrenalectomy (median follow-up 12.1 ± 2.1 months), we collected data on anthropometric parameters and physical examination, pharmacological treatment, blood samples, and 24-h ABPM. All methods were performed in accordance with relevant guidelines and regulations. All subjects gave their informed consent for inclusion before they participated in the study. The study was conducted in accordance with the Declaration of Helsinki, and the protocol was approved by Sapienza University Research Committee (project identification code: RM118164220EB8B2).

An ABI PRISM 7500 Fast Real-Time PCR System (Life Technologies, Foster City, CA, USA) was used for the analysis process. Total RNA was isolated from paired surrounding adrenal neoplasia, peri-renal, and subcutaneous AT samples using TRIzol (Life Technologies, Grand Island, NY). cDNA was reversely transcribed from 5 ng total RNA samples using specific miRNA stem-loop primers from the QIAGEN^® RNeasy Lipid Tissue Mini Kit human panel. PCR products were synthesized from cDNA samples using sequence-specific primers from the QIAGEN^® miRCURY LNA RT Kit (QIAGEN GmbH, Hilden, Germany). We used hsa-miR-16 as endogenous controls to normalize the expression levels of miRNA target genes to correct the differences in the amount of cDNA loaded into PCRs (17). PCR was carried out by placing them 3 times on 96-well microtiter plates, and at the end of the reaction, the results were evaluated using ABI PRISM 7500 software. The cycle threshold value (Ct: cycle threshold) for each set of the three reactions was obtained by calculating the mean value.

Data were expressed as mean and standard deviation (SD). Before statistical analysis, variables that showed a non-Gaussian distribution at the Kolmogorov–Smirnov test were transformed to achieve a normal distribution, and they were analyzed by non-parametric tests. Categorical variables were compared with the Fisher and chi-square tests. Continuous and categorical variables were compared at baseline and follow-up by the Mann–Whitney test. Relationships between continuous variables were assessed by calculating the Pearson correlation coefficient or the Spearman rank correlation coefficient when appropriate. Statistical analysis was performed using SPSS software (version 24 for Mac; IBM^®, SPSS^®Statistics, Italy). Significance was set at p < 0.05. The power of the study is 90.6% with a minimum sample size of 9 patients with beta 0.1 and alpha 0.05. The effect size is 1.61 (according to the Hedges' g formula).

Table 5 shows the quantitative expression of miRNAs (expressed in real-time PCR cycles) in the AT of patients with APA and NFA.

The correlation studies showed that plasma levels of C-reactive protein at baseline were positively correlated with real-time PCR cycles of miRNA-26b analyzed at surrounding adrenal neoplasia AT of the enrolled patients (r = 0.825; p = 0.043; Figure 1). Real-time PCR cycles for surrounding adrenal neoplasia AT miRNA-26b of enrolled patients also positively correlated with LVMi (r = 0.987; p = 0.002) and neck circumference (r = 0.818; p = 0.047).

The comparison between the quantitative expression of surrounding adrenal neoplasia AT miRNAs and follow-up data is shown in Figure 2.

Patients with reduced BMI values at follow-up had significantly higher real-time PCR values of miR26b than those without improvement (35.5 ± 3.7 Ct vs. 32.9 ± 0.99 Ct, p < 0.001).

Furthermore, patients with an improvement in neck circumference and LDL cholesterol values at follow-up were those with reduced expression of miR-143 at surrounding adrenal neoplasia AT at baseline (30.6 ± 8.2 Ct vs. 26.26 ± 2.4 Ct, p < 0.002 and 32.9 ± 7.2 Ct vs. 24.47 ± 2.32 Ct, p < 0.027, respectively). Finally, patients with reduced LVMi had significantly higher real-time PCR values of miR-143 than those with no improvement (30.67 ± 7.78 Ct vs. 22.6 ± 0.77 Ct, p < 0.01).

Previously, Heneghan et al. (10) showed a significant correlation of the tissue expression of miRNAs with respective circulating levels in several diseases, suggesting a possible role of such markers as non-invasive biomarkers for obesity and related diseases, and possible therapeutic targets. Among different miRNAs evaluated, these authors found a significant reduction of miRNA-132 expression in individuals with obesity compared to individuals without obesity, which is inversely related to BMI. Giardina et al. (51) evaluated that the variation of body composition is associated with different expression of some miRNAs, rather than with the use of different diet plans (low or high glycemic index and low-fat content), highlighting that in overweight and obese subjects, there was a significant relationship between variations of waist circumference and fat mass with the reduced tissue expression of miRNA-132.

Previously, Klöting et al. (17) evaluated the different tissue expression of some miRNAs between subcutaneous and omental fat deposits, highlighting a greater metabolic activity in the latter; these results were confirmed in our study, where we also observed, for some miRNAs, the different tissue expression in the visceral AT compared to the subcutaneous site, confirming a different metabolic role in these two districts.

The miRNA-132 is known to act in numerous ways in the regulation of complex metabolic pathways: through cAMP response element-binding protein, which regulates glucose homeostasis (52), and through the brain-derived neurotrophic factor, which is involved in the regulation of appetite and energy homeostasis (53). On the other hand, factors such as chronic hyperglycemia, altered cytokine profile, obesity, and systemic inflammation can alter the tissue expression of miRNAs, especially miRNA-132, playing a role in the link between AT dysfunctions and metabolic diseases such as type 2 diabetes mellitus (17).

Interestingly, miRNAs may have different effects depending on the tissue or organ evaluated; at the myocardial level, miRNA-132 appears to have different properties, especially in relation to the assessment of risk stratification of patients with HF. In recent controlled randomized studies conducted in chronic ischemic or hypertensive HF, the miR-132 inhibitor was assessed as a safe drug, determining a decrease of pro-BNP, significant narrowing of the QRS complex (54), reduced cardiomyocyte volume and interstitial fibrosis, improved capillary density, and increased left ventricular ejection fraction (55). At this level, miRNA-132 could act by inhibiting the expression of nuclear factor erythroid-2-related factor 2 (Nrf2), a pleiotropic protein, physiologically regulating the expression of several antioxidant genes and other cytoprotective enzymes, playing a key role in the maintenance of the functional integrity of cardiomyocytes and cardiac fibroblasts, and preventing maladaptive cardiac remodeling and HF (55).

On the other hand, at the liver site, miRNA-132 showed a negative correlation with pathogenic factors inducing inflammation and insulin resistance-related molecular pathways, playing a key role in the development of non-alcoholic fatty liver disease (NAFLD) and non-alcoholic steatohepatitis (NASH). In patients with obesity, reduced levels of miRNA were correlated to higher behaviors of LDL and total cholesterol, high expression of inflammation molecules [IL-6 and tumor necrosis factor-α (TNF-α)], liver macrophage infiltration, high cellular expression of V-rel reticuloendotheliosis viral oncogene homolog A (RELA), and STEAP family member 4 (STEAP4/STAMP2) (23). Moreover, in an interesting study conducted on brown AT (BAT), physiologically involved in the control of energy homeostasis acting by dissipating energy to produce heat and inversely related to BMI, glucose level, and metabolic syndrome, Kariba et al. (56) found that BAT, induced by norepinephrine or cold exposure, beyond adipokines, can induce higher expression of miRNA-132, repressing hepatic Srebf1 expression, attenuating expression of lipogenic genes, and controlling the lipid metabolism in the liver.

Recently, Wang et al. (57) showed different expressions of some miRNAs in the AT in women with obesity (93% of miRNAs were downregulated, including miRNAs-143 family), with significant correlations with numerous biological functions and signaling pathways. Similar results were shown in subjects with insulin resistance (58), in obese subjects with evident metabolic alterations (59), in young subjects developing dyslipidemia in childhood obesity (60), and in murine models subjected to marked weight gain (61). In this regard, Takanabe et al. (62) demonstrated contrasting results, showing higher miRNAs values correlated with elevated body weight and mesenteric fat mass levels in mouse models subjected to a high-fat diet.

For the first time in the literature, we showed a significant reduction in the tissue expression of miRNA-143 in visceral AT in patients with PA, suggesting the pathophysiological role of this molecule in promoting metabolic disorders in patients with PA.

Physiologically, the miRNAs-143 family is highly abundant in AT (59, 63); recently, Xie et al. (64) showed a significant reduction of these miRNAs in response to an increased inflammatory level, in particular, secondary to elevated behaviors of TNF-α secreted by macrophages. Regarding oxidative stress, we previously found increased oxidative stress in patients with PA due to the direct effect of aldosterone on the production of NOX₂, the catalytic subunit of the main supplier of oxidative stress (NADPH oxidase) (6).

Altered expression of miRNA-143 results in impaired processes of differentiation, maturation, and adipogenesis of white adipocytes (65), inducing inhibition of new adipocyte proliferation, favoring triglycerides accumulation into mature adipocytes, characterized by larger volume adipocytes (66), and reducing the expression of specific adipocyte genes (proliferator-activated peroxisome receptor-γ2 and glucose transporter-4) (67). In this respect, in human adipocytes, Zhu et al. (68) showed inhibition of miRNA-143 by higher plasma levels of free fatty acids, leptin, and resistin.

In this regard, we showed that significant metabolic alterations (hyperglycemia, dyslipidemia, and high rate of metabolic syndrome) observed in patients with PA were associated with increased expression of leptin and resistin in the visceral AT and consensual reduced expression of adiponectin, suggesting a relevant role of these adipokines in the development of metabolic changes observed in patients with PA (3). In this study, we found significantly reduced tissue expression of miRNA-143 in patients with APA compared to subjects with NFA; and in addition, patients with lower tissue expression of miRNA-143 (expressed by greater number of real-time PCR cycles) showed more clinical improvement to the specific treatment of excess hormone, characterized by more evident reduction of neck circumference and LDL behaviors; moreover, lower tissue expression of miRNA-143 was associated with a significant reduction of myocardial mass of left ventricle during follow-up after definitive treatment.

Recent studies suggest the protective role of miRNAs in the cardiovascular system; regarding the crosstalk between AT and the cardiovascular system, it has been shown that, under shear stress, endothelial cells release increased the levels of miR-143, internalized by the smooth muscle cells (SMCs) of the middle tunic in order to avoid SMC de-differentiation, exerting a protective role against atherosclerotic plaque (69); lower levels of this miRNA can favor cardiovascular remodeling observed in patients with PA.

The miRNA-143 could act through different pathways; miR-143 was first identified as a positive regulator of human adipocyte differentiation in 2004 via effects on ERK5 signaling (70). Moreover, further analysis showed that miRNA-143 family overexpression or inhibition, respectively, decreased or increased the MAPK7, suggesting that MAPK7 is a target gene of miR-143-3p mediating cell proliferation and differentiation; inhibition of MAPK7 reduces adipocyte proliferation, playing an opposite role in pre-adipocyte proliferation and differentiation when compared with the miRNA-143 (71). Recent studies showed that HMGA2, a target gene of some miRNAs produced in adipocytes, which is a transcription factor that can regulate adipogenesis (60, 70, 72, 73), is upregulated in subjects with obesity and patients with diabetes (74) and correlated with increased body weight and body fat (75), whereas HMGA2 knockout mice have less fat, low-fat content, and are not susceptible to obesity-related diseases (76). Wang et al. showed that the overexpression of HMGA2 is related to the reduction of miRNA-143, accompanied by a greater inflammatory state in ATs (57), which is particularly associated with the activation of macrophages, with the production of pro-inflammatory cytokines, including TNF-α, IL-6, and IL-1β (77, 78).

We found a significant reduction of miRNA-26b expression on the AT surrounding the adrenal gland in patients with APA compared to the NFA group. In an interesting study, Wróblewski et al. (79) demonstrated that chronic or transient hyperglycemia induces changes in the expression of several adipokines, characterized by lower behaviors of adiponectin and higher production of IL-6 in human visceral preadipocytes, suggesting some mechanisms of development of high grade of inflammation in patients with diabetes. Regarding this, in addition to higher IL-6 expression, multiple miRNAs (in particular miRNA-26b family) were found to be downregulated in chronic hyperglycemia. In this respect, we found a significant negative relation between miRNA-26b expression and the circulating levels of C-reactive protein (expressed as a correlation between high real-time PCR cycles of miRNA-26b and elevated values of C-reactive protein: r = 0.825, p < 0.05). Strum et al. showed that TNF-α, leptin, and resistin induced downregulation of the hsa-miR-26b expression in adipocytes, the latter acting as an important mediator in regulating the obesity-related insulin sensitivity and inflammatory responses (21). Several studies reported a reduction of plasma miRNA-26 family in patients with diabetes (80, 81), and the overexpression of miRNA-26 in murine model protects against high-fat diet-induced obesity, whereas miR-26 deficiency results in AT hyperplasia (82).

Moreover, the expression level of miR-26b negatively correlates with increasing BMI and homeostasis model assessment for IR in human obese subjects. BMI is widely known as an independent risk factor for developing HF, CAD, stroke, and overall CVD death, and the reduction of BMI is related to the improvement of the metabolic and BP profile in these patients. We have demonstrated that patients with higher expression of miRNA-26b showed greater benefit by definitive treatment of aldosterone excess, evidenced by a greater reduction in BMI during follow-up after surgical treatment. miR-26b could promote insulin-dependent glucose uptake through glucose transporter type 4 translocation to the plasma membrane in human mature adipocytes. miR-26b modulates insulin-stimulated AKT activation via inhibition of its target gene, PTEN, and significantly increases insulin sensitivity via the PTEN/PI3K/AKT pathway (83). Recently, Ma et al. showed that miR-26b is highly abundance in AT, acting as an effective regulator of adipogenesis; overexpression of miR-26b increased expression of adipogenic marker genes and lipid accumulation in pre-adipocyte, suggesting its regulation on pre-adipocyte differentiation. miRNA-26b could act through inhibition of expression of fibroblast growth factor 21 (FGF21) mRNA, important growth factor in regulating glucose and lipid metabolism (84), or repressing expression of Fbxl19, encoding a component of SCF–type E3 ubiquitin ligase complexes (82).

The limitations of our study are the reduced number of cases studied; moreover, we did not evaluate the circulating levels of miRNAs, requiring to evaluate tissue expression of miRNAs, to verify different tissue expressions of miRNAs; in fact, in literature, there is a high variability of the results due to different methodologies used, and the evaluation of organs of different dimensions can induce insignificant changes in circulating miRNA levels, albeit a significantly modified tissue expression. Therefore, our study aimed to evaluate the tissue expression of miRNAs precisely to avoid possible methodological biases, and further studies will be necessary to evaluate the possible correlation between tissue expressivity and circulating levels of miRNA, validating the plasma dosage (85).