According to the 2023 World Health Organization (WHO) estimates, 1.28 billion adults between 30 and 79 years old suffer from hypertension and about 1 in 5 have it under control (1). Hypertension is currently the most prevalent risk factor for cardiovascular disease and one of the leading causes of premature death worldwide, with the risk being higher the younger the age of hypertension onset (2).

Blood pressure (BP) regulation is extremely complex, and several endocrine, nervous, vascular and immune factors are closely interconnected in hypertension pathogenesis (3). Water and sodium balance is undoubtedly one of them. As early as 1972, Guyton et al. demonstrated the influence of kidney in BP regulation, showing that increasing BP levels were associated with a significant increase in sodium and, consequently, water renal excretion aimed at reducing extracellular volume and restoring normal BP values (4). While sodium contribution to hypertension development has since been extensively and thoroughly studied, the role of water is still poorly defined.

Arginin-vasopressin (AVP), also known as antidiuretic hormone (ADH), is a 9-amino-acid peptide synthetized by the magnocellular neurons located in the supraoptic and paraventricular nuclei of hypothalamus and secreted in response to raised serum tonicity and/or reduced effective circulating arterial blood volume (5). Characteristically, AVP acts as an antidiuretic hormone by activating V2 receptors on the basolateral membrane of the principal cells in the collecting ducts. This is followed by Aquaporin 2 (AQP2) translocation to the luminal membrane and the formation of AQP2 channels, resulting in increased water permeability, water reabsorption and, finally, excretion of hyperosmotic urine (5). A possible role for AVP in hypertension development has been evoked since the 1980s (6–10), but, to date, is still controversial (11–13). Indeed, AVP was first identified as a pressor hormone (14), and, in vitro, its vasoconstrictor effect, mediated by V1a receptors on smooth muscle cells, has shown to be greater than that of angiotensin II (15, 16). In vivo, however, its importance is presumably limited by the activation of counter-regulatory mechanisms, including the baroreceptor reflex and the modulation of cardiac output by AVP itself (15, 16).

Interestingly, in 1958, Ginetzinsky demonstrated that hyaluronic acid (HA) metabolism in the renal interstitium is closely related to water handling and AVP action (17). Indeed, HA content in renal medullary significantly influences the process of urine concentration and dilution, mainly by altering the interstitial osmotic gradient that draws AVP-induced water reabsorption at the collecting duct level (18, 19). In humans, the degradation of high-molecular-weight HA into low-molecular-weight fragments is catalyzed by six different types of hyaluronidases, of which HYAL1 and HYAL2 are the most highly expressed and actively involved in HA metabolism (20, 21). Their activity is tightly regulated by several factors, including hormones involved in water balance, such as angiotensin II and AVP (22). Notably, AVP seems to accelerate HA catabolism in renal medullary by increasing both hyaluronidase synthesis and activity (23).

As hyaluronidase excretion in the urine has been shown to be inversely correlated to body water balance (19), the aim of our study was to assess urinary hyaluronidase activity and to study its association with BP and vasopressinergic system activation in response to an oral water load. Hypertensive patients were compared to normotensive men with and without a family history of hypertension. Since over 90% of circulating AVP is bound to platelets and its plasma half-life is only 5 to 20 min (24), the urinary AQP2/creatinine ratio, which reflects the shedding phenomenon of AQP2 from the apical luminal membrane of collecting duct principal cells, has been considered as valid surrogate for assessing renal sensitivity to AVP action, based on results obtained in a rat model of essential hypertension (25). To further confirm our results, we also assessed serum levels of copeptin, the 39-aminoacid C-terminal part of the AVP precursor (pre-proAVP), which is secreted into circulation in equimolar concentration with AVP, thus indicating AVP plasma levels (24).

Anthropometric, clinical and biological features of study population are detailed in Table 1 . A total of 86 men aged 20 to 52 years were included. HT patients were older than NT subjects (p<0.0001 versus both NT- and NT+) and, as expected, had higher baseline SBP and DBP values (p<0.0001 for both). No difference in SBP and DBP was found between NT- and NT+ men. Likewise, baseline heart rate was similar among the groups. Subjects had normal renal function and no electrolyte disturbances. After a follow-up period of 8 to 11 years from the water load test, sustained arterial hypertension was diagnosed in 4 out of 29 (14%) NT- versus 11 out of 36 (30%) NT+ men.

As shown in Figure 1 , in response to oral water loading, SBP and DBP increased rapidly, with peak values at 20–40 min, and then reverted to basal or even lower levels. Although HT patients maintained higher BP values throughout the test (p<0.0001 versus normotensive subjects), this trend was similar in the three groups. By contrast, except for an early increase in NT+ men, heart rate was not significantly affected by the oral water load ( Figure 2 ). No differences in baseline plasma osmolarity, as well as serum copeptin levels, were observed among the groups. However, as expected, water load resulted in a significant reduction in both parameters (p<0.001 compared to baseline for all groups, Figures 3A, B ). Notably, at 120 minutes, while plasma osmolarity was similar among groups, NT+ subjects showed a blunted decrease in serum copeptin levels (p<0.05 versus NT- group).

Similarly, as illustrated in Figure 4 , in all groups, urinary hyaluronidase activity was significantly reduced by water load, reaching the lowest levels at 120 min (p<0.0001 versus baseline for all groups) and then gradually recovering to baseline values (p<0.01 versus baseline for HT patients, p<0.001 for NT- and NT+ subjects). In addition, while the trends over time of NT- and HT groups were overlapping, NT+ men showed significantly higher values at both 60 and 120 min after water loading (p<0.0001 versus NT-).

Assessing the AQP2/creatinine ratio in urine, we observed a similar time course ( Figure 5 ). Indeed, after water drinking, the AQP2/creatinine ratio rapidly decreased from 20 min onward to reach minimum values at 90–120 min (p<0.0001 for all groups at both times). Again, trends over time were comparable among groups, but NT+ and HT subjects had higher values than the NT- ones, especially in the first 2 hours of the test.

In light of these findings, the association between urinary hyaluronidase activity and the AQP2/creatinine ratio was studied. As depicted in Figure 6 , including all men, we found significant positive correlations at any time point, especially between 60 and 180 min after water load ( Figures 6B–D ). Interestingly, when we carried out regression analyses by groups, we observed globally consistent results, confirming the data obtained on the entire cohort ( Table 2 ).

When evaluating the associations between SBP and hyaluronidase activity, as well as AQP2/creatinine ratio, significant correlations were found only in the NT- group ( Tables 3 , 4 , respectively). In particular, while SBP was positively correlated with urinary hyaluronidase activity at baseline and at the end of the test (p<0.01 at 0 and p<0.05 at 240 min, Table 3 ), a significant positive correlation with the AQP2/creatinine ratio was detected at T0 and 40 min after water drinking (p<0.05 and p<0.01, respectively, Table 4 ), which corresponded to peak SBP levels ( Figure 1 ).

In our study, we first demonstrated that urinary hyaluronidase activity markedly decreases in response to an oral water load, reaching its lowest levels after 120 min. Notably, at this time, NT+ subjects showed significantly higher values than both NT- and HT men. Interestingly, the time course of serum copeptin levels and urine AQP2/creatinine ratio followed that of hyaluronidase activity with a close direct correlation. These data, along with the positive associations observed between SBP and both urinary hyaluronidase activity and AQP2/creatinine ratio, support the existence of a close relationship between renal hyaluronidases and vasopressinergic system activation, which could affect water metabolism under both normotensive and sustained hypertensive conditions.

Acute oral water intake is known to increase BP in several animal species (29) as well as in patients with orthostatic hypotension symptoms due to autonomic failure (30). In our cohort, we observed a significant and early increase in SBP and, to a lesser extent, DBP levels, which reached their highest values about 20 min after the oral water load. These findings, largely consistent with most of the literature (29, 31–34), have been reported to be associated with increased plasma norepinephrine levels, muscle sympathetic nerve activity and peripheral vascular resistance, which would be counterbalanced by vagal modulation and baroreflex activation (29, 32). In accordance with this, in response to the oral water load, we found no significant changes in heart rate, except for a modest increase in NT+ subjects at 20 min. Interestingly, SBP levels increased by 8.9 +/- 0.3 mmHg in HT group versus 4.6 +/- 0.3 mmHg and 2.7 +/- 0.3 mmHg in NT+ and NT- men, respectively, suggesting a greater sympathetic response to water drinking in HT patients. These data, confirming the results of Velasquez et al. (27), are in agreement with the finding of higher basal muscle sympathetic nerve activity in subjects with mild hypertension compared to the normotensive ones (32). In previous studies, no changes were detected in plasma volume, plasma renin activity and sodium balance (32). Furthermore, the pressor effect of drinking water did not appear to be related to water temperature or gastric distension (29) and was not obtained when pure water was replaced by salt water (32) or administered via intravenous infusion (29). Drinking water would rather activate postganglionic sympathetic neurons directly (33) or, alternatively, through a gastro-pressor reflex mediated by the osmoreceptors in the proximal gut and portal system (35, 36), triggering both sympathetic activation and reflex inhibition of AVP secretion (31). Indeed, whereas in response to oral water loading, Velasquez et al. did not show any difference in plasma AVP levels (27), Geelen et al. demonstrated a rapid decrease followed by a long-lasting inhibition of AVP secretion which was not accompanied by changes in the renin-angiotensin-aldosterone system (RAAS) (31).

Similar to what we observed in normotensive and hypertensive rats (data under submission), oral water load induced a significant decrease in urinary hyaluronidase activity with a nadir at 120 min, followed by an almost complete restoration to baseline physiological values within 4 hours. Meaningfully, NT+ men, despite having lower hyaluronidase activity prior to the test, maintained significantly higher levels than NT- subjects at both 60 and 120 minutes after water load. As far as we know, this is the first time that such a result has been described. Indeed, the effects of oral water load on hyaluronidase activity have at present been studied essentially in vitro (37–39) or in animal models (19, 23, 40, 41). In particular, in rats, Stridh et al. (41) demonstrated that, upon continuous hydration, medullary HA levels increased with a peak after about 2 hours and returned to control values after about 4 hours (41).

Although the inverse relationship between the renal excretion of hyaluronidase and body hydration has been known since 1958 (17, 37), the mechanism through which hyaluronidase activity influences water metabolism has not yet been fully elucidated. HA is a negatively charged, ubiquitously distributed linear glycosaminoglycan with a unique water-binding ability (42) that largely depends on its molecular weight. Indeed, when high-molecular-weight HA is above 0.2 mg/mL, inter- and intra-molecular interactions occur such that the volume occupied by HA increases, excluding other large molecules (19). This “steric exclusion” phenomenon might antagonize further water reabsorption and influence the osmotic activity in the intercellular matrix (43). Interestingly, in the kidney, the HA content is considerably higher in the inner medulla than in the cortex, where 80% of the filtered water is reabsorbed (19) and depends mainly on renal medullary interstitial cell (RMICs) activity, which is influenced by both medium osmolality and oxygen tension (38). In particular, high-molecular-weight HA, internalized in RMICs via the CD44-receptor, undergoes degradation by HYAL1 and HYAL2, which seems to be the main regulatory step determining HA content in the interstitial matrix (37, 38). A decrease in hyaluronidase activity thus results in an increase in high-molecular-weight HA content in the renal medullary, which would interfere with the urinary concentration process by modifying the kidneys’ ability to excrete or reabsorb adequate amounts of water. In addition to the “steric exclusion” effect, two other mechanisms have been proposed to explain this phenomenon. First, modification of the physicochemical characteristics of the interstitial matrix due to HA accumulation could compromise the medullary osmotic gradient. Alternatively, or complementarily, the functional oedema accompanying the increase in high-molecular-weight HA might widen the diffusion distances between tubules and blood vessels and hinder water reabsorption from the collecting ducts (18). In light of these considerations, the higher hyaluronidase activity found in NT+ subjects strongly suggests that men predisposed to develop hypertension have a lower diuretic response to water load, possibly resulting in blood volume expansion and ultimately, hypertension development.

In our study, the changes in hyaluronidase activity were accompanied by broadly comparable variations over time in serum copeptin levels and urinary AQP2/creatinine ratio, which we have previously shown to reliably reflect body water balance and vasopressinergic system activation (24, 25). Moreover, a positive correlation between urinary hyaluronidase activity and AQP2/creatinine ratio was found at any time point of the test, when analyzing both the entire cohort and each subject group separately, indicating the existence of a close relationship between hyaluronidase activity and AVP. The presence of vasopressinergic V1 (44) and V2 receptors (45) has been demonstrated in RMICs and AVP has been found to enhance both the expression and synthesis of hyaluronidases. In addition, it increases sodium and urea concentration to levels optimal for hyaluronidases activity (23) by inducing the expression of the epithelial sodium channel ENaC and the urea transporter UT-A1, respectively (5). Interestingly, with regard to animal models, the Brattleboro rats, which genetically lack AVP, show higher HA content in the renal interstitium associated with exaggerated diuresis (23), while gerbils, having 4 times higher AVP levels than rats, have 37% lower papillary HA content which could facilitate water reabsorption to ensure water conservation (40).

Significantly, prior to oral water load and 4 hours after, that is under physiological conditions in the context of a sodium- and potassium-controlled diet, urinary hyaluronidase activity was positively correlated with SBP levels, which, in turn, were positively associated with the urinary AQP2/creatinine ratio at 0 and 40 min, corresponding to BP rise in response to the oral water load. As mentioned above, the role of AVP in hypertension development has long been debated. Although patients suffering from syndrome of inappropriate ADH secretion (SIADH) have normal BP levels (11) and some authors have found no differences in plasma AVP between normotensive and hypertensive subjects (12), others have shown that individuals with hypertension have significantly higher levels of urinary and circulating AVP, which positively correlate with DBP (7, 8, 14, 27). In addition, in a variety of rat models of hypertension, including spontaneously hypertensive and DOCA-salt sensitive rats, higher levels of AVP in urine and plasma, as well as increased pressor responsiveness to AVP, have been described (6). Notably, in these animals, the antidiuretic effect of AVP appears to be necessary to make possible the expansion of circulating volume upon which these hypertensive forms depend (6). In particular, according to Share and Crofton, for AVP to contribute to hypertension, the pressor responsiveness to it must be sufficiently high, which implies both increased vascular smooth muscle cell reactivity and impaired baroreceptor activity (6).

Thus, both the higher hyaluronidase activity levels and the increased AQP2 urinary shedding indicate that in men with a family history of hypertension, the kidneys’ ability to excrete water in response to an oral load is reduced and/or impaired. In the long-term, this altered vasopressinergic response could lead to the blood volume expansion that contributes to hypertension development in its early stages, before the sodium and RAAS contribution becomes predominant. The lack of a positive linear correlation between SBP and hyaluronidase activity, or AQP2/creatinine ratio, observed in NT+ and HT men, could be consistent with the loss of the appropriate vasopressinergic system regulation. The resulting alterations in water balance might influence BP through two distinct pathways. Indeed, as previously proposed (18, 39), AVP could act in the renal medullary either by regulating the permeability of the apical membrane of the collecting duct principal cells through AQP2 translocation via V2 receptors, or by modifying the HA content in the interstitial matrix via V1 receptors. In particular, while the role of hyaluronidase activity would be primarily to provide conditions facilitating water flow, AQP2 channels could be recruited when higher water volumes need to be moved to rapidly restore water balance ( Figure 7 ).

Despite the novel and innovative findings, our study has some limitations. First, we do not currently have measurements of plasma AVP levels, urine osmolarity, or urine sodium and potassium concentrations during the water load test, which could have been important for a more complete explanation of our results. Second, hyaluronidase activity was not assessed between 0 and 60 min after the oral water load, when BP increased and the AQP2/creatinine ratio began to drop. Third, women were excluded from our study, limiting the generalization of our findings. It is, however, well-known that the incidence and severity of hypertension are lower in females than in age-matched males due to the protective effects of estrogens, which beside acting on cardiovascular and renal systems, regulate the expression and secretion of AVP by the hypothalamic paraventricular nuclei (46). Recent data obtained in rats also demonstrate that estradiol prevents salt-dependent hypertension by suppressing salt-induced GABA-ergic excitation in AVP-secreting neurons, resulting in AVP secretion that plays a crucial role in the development and maintenance of hypertension in these animals (47). To avoid any potential misleading interpretation of our results, we thus decided to exclude women, including those of postmenopausal age. Indeed, as study subjects were young men aged 20 to 52 years, the inclusion of postmenopausal women would have introduced an additional potentially confounding factor related to age. Further, our sample size is quite limited, but to the best of our knowledge, our study is the first to assess hyaluronidase activity under hyperhydration conditions in men and to evaluate its relationship with BP regulation not only in hypertensive patients, but also in normotensive subjects with a family history of hypertension. In this regard, it is noteworthy that after a follow-up period of 8–11 years from the water load test, 30% of NT+ men were diagnosed as hypertensive compared to 14% of NT- men. This does not mean that all NT+ subjects will necessarily develop this disease during their lifetime. However, since the men’s age at the follow-up ranged from 28 to 47 years, having both parents suffering from sustained hypertension is confirmed as a strong predisposing factor for the early development of arterial hypertension. Finally, since this is a noninterventional study, the associations we have pointed out allow us only to speculate about the potential role of AVP in BP regulation, but not to establish a causality link between the AVP-induced changes in HA metabolism and the predisposition to hypertension. Further confirmation of our hypotheses may come from data obtained under water deprivation conditions, which are currently being processed.

In conclusion, in our study we demonstrated for the first time the existence in men of a close association between urinary hyaluronidase activity and vasopressinergic system activation, expressed by serum copeptin levels and urinary AQP2/creatinine ratio. These findings confirm that HA handling is highly involved in renal water metabolism under the influence of AVP, which could play a key role in BP regulation under both physiological and water-load conditions by alternatively targeting hyaluronidase activity and the AQP2 channel system. With the reservation of being confirmed in larger samples, our results contribute to shed light on the complex interactions involved in the endocrine regulation of BP and highlight some pathophysiological aspects that may be deeply implicated in hypertension susceptibility, paving the way for novel potential target to act upon in the early stages of hypertension development.

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

The studies involving humans were approved by Ethics Committee of University Hospital of Parma. The studies were conducted in accordance with the local legislation and institutional requirements. The participants provided their written informed consent to participate in this study.

AnC: Data curation, Investigation, Software, Writing – original draft, Writing – review & editing. AB: Conceptualization, Data curation, Writing – original draft, Writing – review & editing, Methodology, Software. IV: Conceptualization, Data curation, Investigation, Writing – review & editing. GF: Data curation, Investigation, Writing – original draft. VV: Data curation, Investigation, Writing – review & editing. PC: Data curation, Investigation, Writing – review & editing. AM: Conceptualization, Writing – review & editing, Methodology, Project, Resources, Supervision. AdC: Conceptualization, Data curation, Funding acquisition, Investigation, Writing – review & editing, Methodology, Project, Supervision.