Left ventricular hypertrophy is one of the most common forms of TOD occurring in hypertensive patients, associated with an increased cardiovascular disease-related morbidity and mortality (Ruilope and Schmieder, 2008; Katholi and Couri, 2011; Okwuosa et al., 2014; Shenasa and Shenasa, 2017). An increase in LV wall stress is the principal mechanical factor in the development of LVH. However, recent studies indicate that some other factors, including those connected with endothelial dysfunction and/or damage, may play a significant role in the development and maintenance of LVH (Yildiz et al., 2003; Sciacqua et al., 2006; Peng et al., 2014; Bartnicki et al., 2016). There is evidence that nitric oxide (NO) released from healthy endothelium prevents cardiac hypertrophy, whereas endothelin, a powerful microvascular constrictor and pro-inflammatory substance released from a dysfunctional endothelium, promotes cardiac hypertrophy (Shah, 2005; Bupha-Intr et al., 2012; Poulikakos et al., 2014; Archer et al., 2017). It may suggest that endothelial dysfunction precedes measurable LVH and that investigating therapies aimed at normalizing endothelial function may limit the development of LVH. The prevalence of LVH in hypertensive patients is very high, especially in those with RH, in which this complication occurs at a frequency ranging from 40% to 55% (Cuspidi et al., 2001, 2010). RH represents an extreme phenotype of hypertension, characterized by a high uncontrolled blood pressure (BP), despite sustained therapy with at least three different classes of antihypertensive medications, including a diuretic agent. Recent studies suggest that high susceptibility of RH patients to left ventricular remodeling may be associated with their worse endothelial status (Figueiredo et al., 2012; Chang-Min et al., 2014).

Due to the close relationship between endothelial abnormalities and LVH in hypertensive patients, a growing number of studies have been focused on the implication of parameters of endothelial dysfunction and/or damage in the diagnosis and management of cardiac hypertrophy (Palmieri et al., 2005; Salles et al., 2007; Yeboah et al., 2010). The presence of CECs has recently been recognized as a direct and specific marker for endothelial dysfunction and/or injury (Goon et al., 2005; Erdbruegger et al., 2006). The mechanisms of detachment of endothelial cells from vessel walls include their mechanical damage and/or apoptosis, proteolysis of subendothelial matrix proteins or a deficiency of anchoring proteins (Scheller et al., 2014). It was found that an elevated number of CECs is associated with an increased cardiovascular risk as well as worsening cardiovascular outcomes (Boos et al., 2006, 2007a, b). A related circulating cell population is CEPCs, which originate from bone marrow (Kawamoto and Asahara, 2007; Asahara et al., 2011; Scheller et al., 2014). CEPC numbers tend to increase in the peripheral blood following vascular injury. The mobilized CEPCs migrate to the sites of injured endothelium and differentiate into mature endothelial cells in situ (Dimmeler and Zeiher, 2004; Lee and Poh, 2014). Thus, a balance between the CEPCs and CECs seems to be critical for effective endothelial regeneration, which assures continuity of endothelial lining. Therefore, CEPCs/CECs ratio is treated as a reliable parameter of the body’s capacity for endothelial repair (Karthikeyan et al., 2011; Szpera-Goździewicz et al., 2017).

In our previous work, we demonstrated a higher number of CECs and a drastically lowered CEPCs/CECs ratio in patients with mild (MH) and RH (Budzyń et al., 2018). In the present study, for the first time, we tried to determine the potential of these cells in the prediction of LVH in the same group of hypertensive patients. Therefore, patients were divided into those with and those without LVH, and the level of CECs, CEPCs and their ratio were evaluated and compared to a normotensive control. Moreover, in each group of hypertensive patients, the correlation of CECs, CEPCs and their ratio with echocardiographic (ECG) incidences of LVH were also investigated.

The cardiac manifestations of systemic hypertension include LVH at early stages and dilatation with left ventricular dysfunction at late decompensated stages (Gatzoulis et al., 2000; Kahan and Bergfeldt, 2005; Cuspidi et al., 2012). Both have been associated with an increased incidence of ventricular arrhythmias and sudden cardiac death (Tin et al., 2002; Shenasa and Shenasa, 2017). LVH is an adjustment mechanism to chronic pressure overload in arterial hypertension, which allows the left ventricle to maintain volume output against the increased systolic pressure. The hypertrophy is a risk factor for morbidity and mortality in hypertension – the presence of LVH worsens the prognosis in hypertensive patients (Hawkins et al., 2007; Pavlíček et al., 2016). It means that an earlier recognition as well as an improved understanding of cardiac hypertrophy may lead to more effective therapeutic strategies for this cardiovascular risk factor.

Echocardiography is now frequently used in the evaluation of LVH in the hypertensive patients (Zheng et al., 2010; Pavlíček et al., 2016). It has proven to be a reliable and reproducible technique in the assessment of cardiac anatomy and function. Assessing left ventricular mass (LVM) and/or LVM indexed by body surface area (LVMI) is the most common role of echocardiography in hypertensive patients. Recent studies indicate that endothelial dysfunction is linked with LVH parameters in the general population independently of other cardiovascular risk factors (Lazdam et al., 2012), and there is evidence that a declining endothelial function may identify an adverse cardiac phenotype in healthy young adults (Yeboah et al., 2010).

Nitric oxide plays the central role in endothelial function. Recent experimental studies have firmly established that NO released from endothelial cells exerts several specific effects on myocardial function, analogous to the endothelial regulation of vascular wall function (MacCarthy and Shah, 2000). In particular, these include the selective enhancement of myocardial relaxation, diastolic LV function and reduction in myocardial O₂ consumption (Rastaldo et al., 2007; Ruiz-Hurtado and Delgado, 2010). Moreover, it must considered that endothelial cells secrete other vasodilation and vasoconstrictor factors which modulate cardiac performance, and their imbalance contributes to pathological cardiac remodeling (Segers et al., 2018). Accordingly, there is little doubt that endothelial damage and/or dysfunction has a larger impact on cardiac function and heart failure progression than currently anticipated (Segers et al., 2018). In this context, parameters of endothelial dysfunction can be recognized as potential diagnostic and prognostic biomarkers of the cardiac remodeling process in hypertensive individuals, even though there is not quite enough evidence from large clinical trials.

Only a few studies demonstrate a relationship between endothelial dysfunction and LVH in hypertensive patients. An increased concentration of biochemical markers of endothelial dysfunction, such as inter-cellular adhesion molecule 1 (ICAM-1), vascular cell adhesion protein 1 (VCAM-1) and E-selectin, were observed in hypertensive patients with LVH (Kuroda et al., 2001; Malmqvist et al., 2002). Moreover, endothelial damage reflected by abnormal albuminuria was independently associated with LVH occurrence in hypertensive patients, including those with RH (Plavnik et al., 2002; Salles et al., 2007; Nabbaale et al., 2015). However, all of the previously analyzed parameters of endothelial dysfunction are not highly specific and their level is subject to intra-patient variability and also strongly depends on renal function. Therefore, the main aim of our study is to examine the association of specific circulating markers of endothelial damage: CECs and CEPCs with ECG parameters of LVH in MH and RH, the latter being a well-known predisposing factor for cardiac hypertrophy development. We observed that a higher number of CECs reflecting endothelial damage already occurs in patients without LVH and keep rising in patients with LVH, in both the MH and RH group. Ercan et al. (2003) obtained results similar to ours, i.e., a worse endothelial function assessed by FMD in hypertensive patients with and without LVH in comparison to the control group. These observations suggest that endothelial injury may occur prior to LVH development and contribute significantly to cardiac remodeling. Moreover, our studies indicate that the participation of endothelial dysfunction in cardiac hypertrophy may be a mechanism that commonly occurs in various clinical types of hypertension. The important question is whether endothelial perturbation is restricted to coronary circulation or rather spreads out in the peripheral circulatory bed. There are findings indicating that patients with non-ischemic heart failure including LVH tend to have coronary rather than systemic endothelial dysfunction (Canetti et al., 2003; Dini et al., 2009; Shantsila et al., 2012). It has been proven that coronary endothelial impairment is strictly connected with reduced NO synthase expression and activity in cardiac tissue, consequently leading to the deterioration of cardiac contractility, cardiac remodeling and fibrosis (Lapu-Bula and Ofili, 2007). However, in our study an elevated number of CECs was detected in peripheral blood samples suggesting the systemic nature of endothelial dysfunction in hypertensive patients including those with LVH. Our results are in accordance with previous observations made by other authors who detected peripheral endothelial dysfunction in ischemic as well as non-ischemic heart failure (Kubo et al., 1991; Bitar et al., 2006). Interesting results have been obtained by Heitzer et al. (2005) who demonstrated that a lower vasodilator response to acetylcholine and to sodium nitroprusside, that reflected peripheral endothelial dysfunction, is an early manifestation in the progression of left ventricular dysfunction and may even precede the symptoms of heart failure. It seems that the exact localization of the site of endothelial dysfunction contributing to LVH may be difficult mainly due to the complexity of interactions between different segments of the vascular tree. The factors liberated by dysfunctional endothelium in some specific areas of circulation may affect endothelium localized elsewhere. Therefore, in further studies, the concept of endothelium as one organ should be established as the most appropriate.

In response to endothelial damage, the amount of CEPCs should rise adequately. Although in our study the number of CEPCs demonstrated a tendency to rise in hypertensive patients with LVH, it did not reach statistical significance. Similar results were obtained by other authors, who did not observe any statistical difference in the total number of CEPCs in patients with essential hypertension (Delva et al., 2007; Marketou et al., 2014). However, they assessed CEPC levels in the general population of hypertensive patients, not considering their classification based on the absence/presence of LVH. Other authors, who divided hypertensive patients using LVH criteria observed even lowered levels of CEPCs in those with ECG evidence of LVH in comparison to those who did not have LVH (Lee et al., 2011). To find an explanation for this phenomenon it must be taken into consideration that the amount of CEPCs is regulated by two different mechanisms: release of CEPCs from bone morrow and the survival and life span of CEPCs. The various factors common to hypertensive milieu, such as angiotensin II, aldosterone, catecholamines and cytokines liberated by injury tissues are suspected to activate bone morrow stem cells, which undergo differentiation and mobilization (Montaniel and Harrison, 2016). That mechanism explains an elevated number of bone morrow derived immune cells in hypertensive patients which infiltrate peripheral organs causing their dysfunction (Wenzel et al., 2011; Youn et al., 2013; Santisteban et al., 2015). According to a previously described concept, CEPCs which originate from bone morrow probably undergo an analogous process: enhanced activation and mobilization which should result in their rise in the circulation. However, in the circulation the amount of CEPCs is regulated by other mechanisms connected with their lifespan and death, such as apoptosis. Matsumoto et al. who observed an increased activity of caspase-3 in CEPCs isolated from patients with aortic valves stenosis, has indicated that enhanced apoptosis is a significant cause of the reduced CEPCs level (Matsumoto et al., 2009). It means that the enhanced mobilization of CEPCs from bone morrow may be masked by their enhanced elimination from circulation due to apoptosis. Therefore, as a consequence, the overall number of CEPCs observed in circulation do not rise apparently or may even fall despite their continuous liberation from bone morrow. These speculations could be supported by the evaluation of the apoptotic status of CEPCs in hypertensive individuals, which will be our aim in further studies.

Described mechanisms may also explain the positive relationship between CEPCs and ECG parameters of LVH: LVM as well as LVMI was found in RH patients and confirmed by other authors (Oliveras et al., 2008; Kalyva et al., 2016). This correlation may indicate mobilization of CEPCs from bone morrow rising adequately to organ damage, despite the fact that CEPCs level in the circulation do not elevate proportionally, due to failure mechanisms described above. Interestingly, CEPCs remained an independent predictor of LVH in the general population of hypertensive individuals even after the adjustment for other confounders, such as the type of hypertension, SBP and DBP, age and sex, BMI, inflammation, and renal function parameters. It means that LVH progression may be directly associated with an increased liberation of CEPCs from bone morrow. However, considering other described mechanisms controlling CEPCs turnover, the final number of CEPCs seems to be inadequate for endothelial regeneration and organ repair. It is reflected by a lowered CEPCs/CECs ratio which already occurs in hypertensive individuals without LVH and falls further in those with LVH.

Some studies have suggested that CEPC numbers may be affected by age and gender (Fadini et al., 2008; Pelliccia et al., 2009; Rousseau et al., 2010). This hypothesis is derived from observations indicating that the incidence of cardiovascular diseases is lower in cycling women, but after the age of between 45 and 54 y it drastically increases, reaching that of age-matched men within 10 y after menopause (Rousseau et al., 2010). This phenomenon is probably associated with the protective effect of endogenous estrogens on the vasculature. The impact of estrogens on endothelium, including CEPCs level and function, has emerged as a possible mechanism for this phenomenon (Masuda et al., 2007; Fadini et al., 2008; Li et al., 2013). Although estrogens appear to increase bone marrow−derived CEPCs production, reports on sex differences in CEPC levels have been studied in small populations with less consistent findings. Stauffer et al. (2008) found no differences in CEPCs in 29 menopausal women compared with men. Similarly, no sex-related differences in healthy young men and women were reported by Ruszkowska-Ciastek et al. (2015). In contrast, some authors indicate that premenopausal women had a higher CEPCs number compared with menopausal women and age-matched men (Fadini et al., 2008; Lemieux et al., 2009). Most women who qualified for our study were aged above 51 y which corresponds to the average age of menopause occurrence. To minimize the influence of the women participating in our research, we additionally decided to examine all parameters only in the group of men with the exclusion of women. Similarly, as in the general group of hypertensive patients, the CECs were higher in men with and without LVH compared with men in the control group, regardless of the type of hypertension. However, among men with RH, the greatest elevation of CECs, indicating the most severe endothelial damage, occurred in those with LVH complication. Interestingly, in the same group of RH men, a significant elevation of CEPCs was observed, which may be a response to the endothelial injury. These observations, obtained in hypertensive men, differ from those obtained in the general group of patients, including men and women, and probably confirm some gender differences in the endothelial function reported previously (Routledge et al., 2012; Stanhewicz et al., 2018). A growing number of studies suggest that the risk factors and mechanisms of endothelial injury may be strongly sex-dependent and should be analyzed in a population of men and women separately (Routledge et al., 2012; Cattaneo et al., 2017). Some authors speculate that the identification of sex specific mechanisms of endothelial dysfunction may improve patient outcomes across a wide variety of cardiovascular diseases (Bacon et al., 2011). In the present work, the number of patients was too small for the examination of gender variation in the analyzed endothelial parameters, to produce reliable results. However, future clinical studies, concerning the role of endothelium in hypertension, should be considered from a gender perspective.

Admittedly, the assumption about the important role of endothelial injury in LVH development is not innovative and was reported previously by other authors, however, our observations, that this process is accompanied by the disruption in endothelial regeneration before LVH occurrence, are crucial and may have some important clinical implications. They offer insight into the possible biological mechanism underlying LVH pathology and, in this way, enhance the chances of developing an effective targeted therapy for its treatment. A recovery of a sufficient number of functional CEPCs able to regenerate damaged endothelium may have beneficial effects; it may prevent hypertension related complications, including LVH, but may also lead to the regression of hypertension itself.

We are aware of the limitations of our study. It can be agreed that the number of participants was small. Another limitation was the lack of assessment of functional capacity of CEPCs. It is very likely that apart from their level in circulation, this subpopulation of endothelial cells may have an altered functional characteristic, which has a potential impact on cardiac hypertrophy development in hypertensive individuals. Despite these limitations, an important finding derived from our study is the fact that CEPCs may potentially serve as a screening biomarker in hypertensive patients to stratify the risk of LVH and initiate preventive interventions. However, larger clinical studies are necessary to determine the optimal diagnostic threshold values of CEPCs in differentiating hypertensive patients with a high risk of LVH development. It must also be considered that due to the deficiencies in our understanding of CEPCs pathophysiology, these endothelial cells are far from being a serious contender as a clinically useful biomarker. Thus, further experimental research is required to clarify the mechanism underlying the involvement of CEPCs in the pathogenesis of LVH to verify their predictive values.

In conclusion, our results suggest that endothelial injury, observed as an increased CECs number, and its impaired regeneration, reflected by a lowered CEPCs/CECs ratio precede LVH occurrence and may have a significant role in LVH development, regardless of the clinical severity of hypertension. Moreover, our data shows that CEPCs number is independently associated with LVH in hypertensive individuals, even after the adjustment of some potential confounders. Thus, CEPCs could potentially serve as a screening biomarker in hypertensive patients to stratify risk of LVH and initiate preventive interventions. However, more research is needed to delineate the pathophysiological effects of CEPCs on LVH and to verify their predictive values.

The studies involving human participants were reviewed and approved by the Local Bioethical Committee of Poznań University of Medical Sciences (No. 163/17). The patients and participants provided their written informed consent to participate in this study.

MagB contributed to the design of the study and drafted the manuscript. MagB, BG, MacB, MarK, BB-K, AO, AB, MI, and MagK contributed to the conception of the study. MagB, BG, BB-K, AO, and MI contributed to the data acquisition. MagB, MacB, MarK, BB-K, AO, AB, and MagK contributed to the analysis of the study. All authors critically revised the manuscript, gave final approval, and agreed to be accountable for all aspects of work ensuring integrity and accuracy.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.