A_2AR-mediated lymphangiogenesis via VEGFR2 signaling prevents salt-sensitive hypertension Free

Abstract

Aims

Excess dietary sodium intake and retention lead to hypertension. Impaired dermal lymphangiogenesis and lymphatic dysfunction–mediated sodium and fluid imbalance are pathological mechanisms. The adenosine A_2A receptor (A_2AR) is expressed in lymphatic endothelial cells (LECs), while the roles and mechanisms of LEC–A_2AR in skin lymphangiogenesis during salt-induced hypertension are not clear.

Methods and results

The expression of LEC–A_2AR correlated with lymphatic vessel density in both high-salt diet (HSD)–induced hypertensive mice and hypertensive patients. Lymphatic endothelial cell–specific A_2AR knockout mice fed HSD exhibited 17 ± 2% increase in blood pressure and 17 ± 3% increase in Na⁺ content associated with decreased lymphatic density (−19 ± 2%) compared with HSD-WT mice. A_2AR activation by agonist CGS21680 increased lymphatic capillary density and decreased blood pressure in HSD-WT mice. Furthermore, this A_2AR agonist activated MSK1 directly to promote VEGFR2 activation and endocytosis independently of VEGF as assessed by phosphoprotein profiling and immunoprecipitation assays in LECs. VEGFR2 kinase activity inhibitor fruquintinib or VEGFR2 knockout in LECs but not VEGF-neutralizing antibody bevacizumab suppressed A_2AR activation–mediated decrease in blood pressure. Immunostaining revealed phosphorylated VEGFR2 and MSK1 expression in the LECs were positively correlated with skin lymphatic vessel density and A_2AR level in hypertensive patients.

Conclusion

The study highlights a novel A_2AR-mediated VEGF-independent activation of VEGFR2 signaling in dermal lymphangiogenesis and sodium balance, which might be a potential therapeutic target in salt-sensitive hypertension.

Structured Graphical Abstract

Impaired dermal lymphangiogenesis and lymphatic dysfunction contribute to the pathogenesis of salt-sensitive hypertension. Modulation of lymphatic endothelial A_2AR-mediated VEGF-independent activation of VEGFR2 signaling, to regulate dermal lymphangiogenesis and sodium balance, might be a potential therapeutic target in salt-sensitive hypertension. A_2AR, adenosine A_2A receptor; KO, knock out; LEC, lymphatic endothelial cell; MSK1, mitogen- and stress-activated kinase 1; Na⁺, sodium; VEGF, vascular endothelial growth factor; VEGFR2, vascular endothelial growth factor receptor 2.

Open in new tabDownload slide

See the editorial comment for this article ‘Skin sodium, lymphatics, and blood pressure: a non-canonical mechanism of salt-sensitive hypertension', by R. Nosalski and T.J. Guzik, https://doi.org/10.1093/eurheartj/ehad290.

Introduction

Excess dietary sodium intake and retention play a central role in the pathogenesis of hypertension.¹ However, plasma sodium levels are maintained with a narrow range in both normotensive and hypertensive mammals,^2,3 and the mechanisms for salt-sensitive hypertension remain unclear. Recent studies implied that impaired lymphangiogenesis and lymphatic dysfunction–mediated sodium and fluid imbalance contribute to the development of salt-sensitive hypertension.^4,5 There is growing evidence that skin interstitium and the skin lymphatics are important regulators of both sodium (Na⁺) balance and blood pressure (BP).⁶ Sodium accumulates in the skin during a high-salt diet (HSD), which stimulates dermal macrophages to secrete vascular endothelial growth factor C (VEGFC), activating lymphangiogenesis locally.⁵ Inhibition of lymphangiogenesis exacerbates the hypertensive effect of HSD.^7,8 However, the mechanisms by which the lymphatic endothelial cells (LECs) participate in mitigating salt-sensitive hypertension are largely unstudied.

The G protein–coupled adenosine receptors comprise A₁, A_2A, A_2B, and A₃ subtypes, which are activated upon binding with adenosine.⁹ The A_2A receptor (A_2AR) is abundantly expressed in various cell types, including endothelial cells, macrophages, and adipocytes, and regulates cardiovascular response.^10–14 Activation of A_2AR inhibits the release from immune cells of inflammatory factors, promotes vasodilation, regulates the secretion of adipokines, and thereby promotes cardiovascular homeostasis.^15–17 Endothelial A_2AR activation induces angiogenesis in wound healing and retinopathy.¹⁴ In normal rats, A_2AR antagonists reduce renal sodium and water excretion.¹⁸ Furthermore, deletion of A_2AR augments salt-induced BP elevation in Dahl salt-sensitive hypertensive animals.¹⁹

Herein, we report our discovery that A_2AR expression in LECs is positively correlated with lymphatic vessel density in the skin in HSD-induced hypertensive mice as well as in hypertensive patients. We observe that a reduction in A_2AR expression in LECs contributes to hypertension, because LEC-specific A_2AR knockout mice (A_2AR^LECKO) have increased BP associated with decreased lymphangiogenesis and reduced skin Na⁺ clearance. We further observe a VEGF-independent activation of VEGFR2 during A_2AR-mediated lymphangiogenesis. Consistent with this mechanism, we find that the VEGFR2 inhibitor fruquintinib or genetic VEGFR2 deletion in LEC, but not the VEGF antibody bevacizumab, reverses dermal lymphangiogenesis induced by the A_2AR agonist and increases BP in HSD mice. These observations are consistent with our hypothesis that A_2AR activation in LECs protects against high salt–induced hypertension by activating dermal lymphangiogenesis and restoring sodium balance in the skin.

Methods

Detailed materials and methods are presented in the supplementary material online.

Results

The lymphatic endothelial cell–A_2AR mediates skin lymphangiogenesis and Na⁺ clearance

We observed that A_2AR was the dominant adenosine receptor in mouse skin LECs (see Supplementary data online, Figure S1A). In cultured human LECs (HLECs), the adenosine receptor with the greatest increase in expression responsive to NaCl stimulation (40 mM) is A_2AR. In contrast, the expression of A₁R, A_2BR, and A₃R were unchanged in response to high NaCl (see Supplementary data online, Figure S1B). The expression of A_2AR was dose-dependently increased in HLECs by NaCl (see Supplementary data online, Figure S1C). Moreover, we found no apparent LEC death when NaCl concentration is below 40 mM. However, at high-concentration NaCl (80 mM), we observed an increase in surface markers of apoptosis and necrosis (see Supplementary data online, Figure S1D and E). Therefore, all in vitro experiments were performed under 40 mM. We next observed the time-dependent A_2AR expression in NaCl-treated HLECs and found that A_2AR expression increased during the first 6 h and then decreased (see Supplementary data online, Figure S1F).

Similarly, we observed an increase in A_2AR expression (66 ± 6%) in the cutaneous lymphatic endothelium, as well as serum adenosine in mice administered HSD, in parallel with increased dermal lymphatic capillary density (52 ± 11%) (Figure 1A–C). Intriguingly, in hypertensive patients, A_2AR expression (−33 ± 9%) was decreased in skin lymphatic vessels (−28 ± 9%, podoplanin⁺ cell), associated with decreased lymphatic capillary density as observed in 3D images (Figure 1D and Supplementary data online, Videos S1 and S2) and coimmunostaining (Figure 1E–G). These findings are consistent with the hypothesis that lymphatic endothelial A_2AR participates in HSD-driven lymphatic capillary density and sodium clearance in skin. We therefore hypothesized that the increased A_2AR expression in mice may enhance sodium clearance by stimulating lymphangiogenesis.

The lymphatic endothelial cell–A_2AR protects against salt-induced hypertension

We performed flow cytometric analyses of disaggregated skin from HSD mice and found an increased F4/80-LyVE1⁺ LECs, but not F4/80⁺LyVE1⁺ macrophages (see Supplementary data online, Figure S2A). Although it has been reported that macrophage deletion of A_2AR had no effect on hypertension,²⁰ we performed bone marrow (BM) transplantation assay from A_2AR global knockout mice (A_2AR^KO-BM)¹⁶ or control WT mice (A_2AR^WT-BM) to wild-type C57 mice to exclude the potential roles of macrophage A_2AR in high salt–induced hypertension. No statistical difference of BP, lymphatic vessel density, skin VEGF mRNA expression, and Na⁺ content was observed between A_2AR^WT-BM and A_2AR^KO-BM mice (see Supplementary data online, Figure S3A–E). These data indicate that Lyve1-Cre is appropriate for studying skin lymphatic system.

To explore the role of the lymphatic endothelial A_2AR in hypertension, we generated the LEC-specific Adora2a receptor gene deficient (Lyve1-Cre⁺ A_2AR^fl/fl; abbreviated as A_2AR^LECKO) and control littermate (Lyve1-Cre⁻ A_2AR^fl/fl; abbreviated as WT) mice. The A_2AR expression was decreased (−77 ± 5%) in LECs, but not significantly altered in skin macrophages and vascular ECs from A_2AR^LECKO mice (see Supplementary data online, Figure S2B). A_2AR deletion in LECs had no effect on dermal lymphatic capillary density, BP, skin Na⁺, and Cl⁻ content and concentration with low-salt diet (LSD) (Figures 2A and B and S2C).

Intriguingly, with HSD, we observed an increase in dermal lymphatic capillary density (59 ± 7%) and an increase in BP (29 ± 4%) in WT mice compared with LSD-fed WT mice. In comparison, in the A_2AR^LECKO mice, there was a blunted increase in lymphatic capillary density (37 ± 5%) and an exaggerated increase in BP (45 ± 2%) with HSD (Figure 2A and B). Thus, A_2AR deficiency in LECs inhibits cutaneous lymphangiogenesis and increases BP in response to HSD. We also visualized lymphatic vessels by Evans blue dye lymphangiography and found fine distinct lymphatic vessels in HSD-treated WT mice. In contrast, HSD-treated A_2AR^LECKO mice showed dilated lymphatic vessels and Evans blue dye extravasated from these dilated vessels (Figure 2C and Supplementary data online, Videos S3 and S4). Na⁺ content and concentration were both increased, but Cl⁻ content and concentration were unchanged in the skin of HSD-A_2AR^LECKO mice compared with HSD-WT mice (Figure 2D). No statistical differences in plasma Na⁺ and Cl⁻ concentration were observed between HSD-A_2AR^LECKO and HSD-WT littermate mice (see Supplementary data online, Figure S2D). The decreased cutaneous lymphatic capillary density in A_2AR-deficient mice was further confirmed by quantitative immunofluorescence microscopy analysis using the lymphatic marker LyVE1 (Figure 2E).

We also utilized a tamoxifen-inducible Prox1Cre^ERT2 mouse line to specifically delete LEC–A_2AR in adult and generated Prox1Cre^ERT2+ A_2AR^fl/fl (abbreviated as A_2AR^PLECKO) and control littermate A_2AR^fl/fl mice. Tamoxifen administration extinguished A_2AR expression in LECs of A_2AR^PLECKO mice (see Supplementary data online, Figure S4A). In response to HSD, A_2AR^PLECKO mice exhibited an attenuated dermal lymphatic density and exaggerated increase in systolic BP (SBP) and cutaneous Na⁺ accumulation compared in comparison to control A_2AR^fl/fl mice (see Supplementary data online, Figure S4B–D).

We next used a mild mouse model with 4% salt diet plus 0.9% saline water to drink and detected change in BP of A_2AR^LECKO and littermate WT mice by radiotelemetry BP monitoring. Both SBP and diastolic BP were elevated in WT mice (Figure 2F). Blood pressure was elevated to a greater degree in the A_2AR^LECKO mice, accompanied with reduced dermal lymphatic capillary density (−30 ± 3%), and increased skin Na⁺ content (from 0.10 ± 0.003 to 0.11 ± 0.003 mmol/g) and Na⁺ concentration (from 0.083 ± 0.003 to 0.094 ± 0.002 mmol/L, Figure 2F–H).

To determine if other causes of hypertension were exacerbated in the A_2AR^LECKO mouse, we induced hypertension using angiotensin II infusion. We observed no differences in dermal lymphatic capillary density or BP between A_2AR^LECKO and WT littermates treated with angiotensin II (see Supplementary data online, Figure S5A–C).

To determine if there is a direct link between lymphangiogenesis and BP, we treated mice with recombinant VEGFC. As expected, VEGFC infusion significantly increased skin lymphatic capillary density in HSD-WT mice from 56 ± 1 to 74 ± 2 AU and normalized the decreased skin lymphangiogenesis in HSD-A_2AR^LECKO mice from 44 ± 2 to 60 ± 1 AU (see Supplementary data online, Figure S6A). Similarly, VEGFC administration normalized skin Na⁺ clearance and BP in HSD-A_2AR^LECKO mice (see Supplementary data online, Figure S6B and C). The effect of LEC–A_2AR on lymphangiogenesis was confirmed in A_2AR homozygous knockout HLECs generated by CRISPR/Cas9 methodology [documented by quantitative polymerase chain reaction (qPCR) and western blot; Supplementary data online, Figure S7A]. Consistent with the in vivo data, NaCl inhibited HLEC proliferation, migration, and tube formation with reduction of branch points and total tube length (see Supplementary data online, Figure S7B–D). This effect of NaCl was exacerbated by A_2AR knockout (see Supplementary data online, Figure S7B–D).

Lymphatic endothelial cell–A_2AR deletion decreases Na⁺ transport caused by lymphatic dysfunction

To provide direct evidence that A_2AR regulates lymphatic function, especially Na⁺ removal, we measured the transport of Evans blue from the peripheral tissue to the popliteal lymph node (LN) and then lumbar LN. Transport of Evans blue, quantified by VISQUE fluorescence imaging system, from foot to popliteal and lumbar LN was unchanged in LSD-A_2AR^LECKO compared with LSD-WT mice (Figure 3A and C). We observed decreased transport of Evans blue (−30 ± 3%) in HSD-A_2AR^LECKO mice compared with HSD-WT mice (Figure 3A and C–E). Furthermore, we evaluated Na⁺ transport by detecting Na⁺ concentration in the LN at specified times after subcutaneous injection of NaCl. Na⁺ concentrations in both popliteal (0.016 ± 0.0003 mmol/L vs. 0.014 ± 0.0005 mmol/L) and lumbar (0.014 ± 0.0004 vs. 0.013 ± 0.0004 mmol/L) LNs of HSD-A_2AR^LECKO mice were significantly reduced in comparison to HSD-WT mice, although Na⁺ concentrations between LSD-A_2AR^LECKO and LSD-WT mice were not changed (Figure 3A and B).

Because A_2AR did not affect angiotensin II–induced hypertension, we also assessed renin–angiotensin–aldosterone system and renal function. The plasma renin activity, angiotensin II, and aldosterone were not changed in HSD-treated A_2AR^LECKO and WT mice (Figure 3F). Functional renal analyses by metabolic cage experiments revealed no significant difference of feed and water consumption, Na⁺ consumption of feed and drink, and urine Na⁺ and Cl⁻ between A_2AR^LECKO and WT mice with HSD (Figure 3G–I). These data indicate that the LEC–A_2AR deficiency exacerbates salt-induced hypertension via aggravating lymphatic dysfunction to decrease dermal lymphangiogenesis and Na⁺ transport.

Pharmacological activation of A_2AR induces lymphangiogenesis and skin Na⁺ clearance and decreases blood pressure

To determine whether A_2AR activation could protect against HSD-induced hypertension, we treated HSD-fed A_2AR^LECKO and WT mice with the A_2AR-specific agonist CGS21680. This A_2AR agonist increased lymphatic capillary density (40 ± 9%) and decreased BP (−23 ± 12%) in HSD-WT mice (Figure 4A and B). Decreased Na⁺ (−19 ± 11%) and unchanged Cl⁻ content and concentration were observed in the skin of HSD-WT mice treated with CGS21680 (Figure 4C). More importantly, these beneficial effects of A_2AR activation including lymphangiogenesis, skin Na⁺ clearance, and BP reduction were absent in HSD-treated A_2AR^LECKO mice (Figure 4A–C). Consistently, A_2AR agonist CGS21680 could increase HLEC proliferation, migration, and tube formation in vitro, which were also attenuated after A_2AR knockout (Figure 4D–F). These observations suggest that pharmacological activation of A_2AR protects against HSD-induced hypertension via inducing dermal lymphangiogenesis and promoting skin Na⁺ clearance.

A_2AR regulates VEGFR2 activity via promoting MSK1 phosphorylation

To gain insight into of the mechanisms of LEC–A_2AR-mediated lymphangiogenesis, we performed phosphoprotein profiling assay in HLECs following pharmacological A_2AR activation. The protein microarray identified phosphorylated VEGFR2 (Tyr1175) expression was increased 1.8-fold in HLECs, whereas VEGFR2 protein expression was not changed, after activation of A_2AR (see Supplementary data online, Table S4 and Figure 5A and B), which were confirmed by western blot in both skin LECs and NaCl-treated HLECs with A_2AR agonist stimulation (Figure 5C). In contrast, the expression of phosphorylated VEGFR2 (Tyr1175) and mitogen- and stress-activated kinase 1 (MSK1) (Ser376) were not changed after treatment of A_2AR^LECKO mice or A_2AR knockout HLECs with angiotensin II (see Supplementary data online, Figure S5D–F). It is known that VEGFR2 undergoes endocytosis and that phosphorylated VEGFR2 could translocate to the nucleus.^21,22 We found that A_2AR activation induces VEGFR2 endocytosis and increases phosphorylated VEGFR2 in the nucleus (see Supplementary data online, Figure S8C and D). The expression of VEGF family, including VEGFA, VEGFB, VEGFC, and VEGFD, in NaCl-treated HLECs were not changed following A_2AR knockout (Figure 5D) or A_2AR activation (Figure 5E). These studies indicate that VEGFR2 phosphorylation and internalization induced by A_2AR might be independent of VEGF.

We next determined whether A_2AR directly interacted with VEGFR2. Immunoprecipitation in 293T cells transfected with V5-A_2AR and HA-VEGFR2 revealed that A_2AR did not directly interact with VEGFR2 (see Supplementary data online, Figure S8A). Therefore, we employed protein microarray to identify potential A_2AR downstream signaling that may activate VEGFR2 phosphorylation. We found that phosphorylated MSK1 (Ser376) was increased following A_2AR activation (Figure 5A and B), which was confirmed by western blot in both skin LECs and HLECs treated with A_2AR agonist and NaCl (Figure 5C and D). Immunostaining data showed that in CGS-treated mice, there was increased phosphorylation of MSK1 (Ser376) and VEGFR2 (Tyr1175) in LECs (see Supplementary data online, Figure S8B and C). Immunoprecipitation assay in 293T cells transfected with V5-tagged A_2AR and Flag-tagged MSK1 vector showed that A_2AR directly interacted with MSK1 (Figure 5F). To further confirm the result in vivo, we performed immunoprecipitation in HSD-treated mice and found that A_2AR directly bound to MSK1 (Figure 5F).

To investigate the effects of phosphorylated site Ser376 of MSK1 on VEGFR2 activity and endocytosis, we mutated Ser376 of MSK1 to Ala376 [MSK1(376A)] and performed western blots in MSK1 or mutated MSK1(376A)-transfected HLECs under NaCl stimulation. An increase in phosphorylated MSK1 (Ser376) and phosphorylated VEGFR2 (Tyr1175) was observed in the nucleus of MSK1-transfected HLECs compared with Empty-transfected HLECs (Figure 5G). In comparison, phosphorylated MSK1 (Ser376) and phosphorylated VEGFR2 (Tyr1175) in the nucleus of MSK1(376A)-transfected HLECs were significantly decreased compared with MSK1-transfected HLECs. Similar results were observed in the nucleus of HLECs by utilizing confocal immunostaining (Figure 5H). We also performed in vitro EdU and wound healing assays in MSK1 or mutated MSK1(376A)-transfected HLECs and found that MSK1 overexpression promoted cell proliferation and migration, while mutated MSK1(376A) had no effect on HLEC proliferation and migration (Figure 5I and J). These studies suggest that MSK1(376A) plays an important role in VEGFR2 activation and internalization and HLEC function.

We also utilized the MSK1 antagonist (CKI7) and found that MSK1 antagonist inhibited phosphorylation of MSK1 and VEGFR2 (see Supplementary data online, Figure S9A) and decreased HLEC proliferation, migration, and tube formation (see Supplementary data online, Figure S9B–D). These studies suggest that A_2AR activation regulates LEC functions via MSK1-dependent, VEGF-independent VEGFR2 activation and internalization.

Pharmacological inhibition or lymphatic endothelial-specific genetic deletion of VEGFR2 blocks the beneficial effects of A_2AR activation in high-salt diet mice

To determine whether VEGF-independent VEGFR2 signaling is involved in A_2AR activation–mediated phenotype in HSD mice, we utilized the antineoplastic agent VEGFR2 tyrosine kinase inhibitor fruquintinib or VEGF-neutralizing antibody bevacizumab. Bevacizumab scavenges VEGF to prevent its activation of VEGFR2, whereas fruquintinib inhibits processes downstream of VEGFR2 activation. Mice fed HSD were pretreated with the antagonists and then received the A_2AR agonist CGS21680. CGS21680 treatment increased lymphatic capillary density (from 19.0 ± 0.3% to 27.0 ± 0.4%) and skin Na⁺ clearance (Na⁺ content from 0.1 ± 0.001 to 0.0794 ± 0.004 mmol/g) and decreased BP (from 120 ± 3 to 97 ± 5) in HSD-WT mice. Fruquintinib pretreatment decreased lymphatic capillary density (by −36 ± 1%) and skin Na⁺ clearance (by −19 ± 3%) and increased BP (by 42 ± 8%). In contrast, bevacizumab did not influence these parameters in CGS21680-treated HSD mice (Figures 6A–C and S10A).

We next observed the effect of fruquintinib or bevacizumab on BP of colorectal cancer patients without history of hypertension before hospitalization. The BP of patients was increased after fruquintinib or bevacizumab treatment (see Supplementary data online, Table S2). We then grouped the patients into bland diet group and HSD group according to their dietary preferences. We found a significant increase in BP in high-salt group compared with bland diet patients with fruquintinib, while no difference of BP was observed between bland diet and high-salt patients with bevacizumab (Figure 6D). These data demonstrate that fruquintinib increases BP mainly in patients with HSD preference.

Moreover, we obtained conditional VEGFR2^fl/fl mice and crossed with Prox1Cre^ERT2 mice to generate Prox1Cre^ERT2+VEGFR2^fl/fl (VEGFR2^PLECKO) mice. In these animals, tamoxifen treatment reduced VEGFR2 expression in LECs (by −70 ± 3%), which was confirmed by qPCR and immunostaining (Figure 6E and F). VEGFR2 deletion in LECs specifically reversed the effects of A_2AR activation to increase lymphangiogenesis and skin Na⁺ clearance and to reduce BP in HSD-treated mice (Figure 6G and I). Furthermore, the increased transport of Evans blue from peripheral tissue to popliteal LN in HSD mice pretreated with A_2AR agonist CGS21680 was blocked when VEGFR2 was deleted in LECs (Figure 6J). These data confirmed that the effects of A_2AR activation on high salt–induced hypertension were through downstream VEGFR2 signaling.

In accordance with the in vivo studies, fruquintinib reversed A_2AR agonist–mediated increase of LEC proliferation and migration, whereas bevacizumab has no effect on LEC functions (see Supplementary data online, Figure S10B and C). Knockdown of VEGFR2 in vitro cultured HLEC also attenuated CGS-induced proliferation, migration, and tube formation (see Supplementary data online, Figure S11A–D). These studies further indicate that A_2AR activates VEGF-independent VEGFR2 signaling, which increases dermal lymphangiogenesis to protect against HSD-induced hypertension.

Phosphorylated VEGFR2 and MSK1 levels positively correlate with lymphatic markers in the skin of hypertensive patients

We finally evaluated whether phosphorylated VEGFR2 and MSK1 expression correlated with markers of lymphatic vessels (podoplanin) and Na⁺ content and BP in the skin tissue of hypertensive patients. We observed that phosphorylated MSK1 (Figure 7A) and VEGFR2 (Figure 7B) were decreased in hypertensive patients, which values also correlated with increased skin Na⁺ retention and decreased podoplanin expression (Figure 7C–F). These data provide evidence of morphological and functional changes in cutaneous lymphatics of hypertensive patients resulting in reduced sodium clearance. Thus, restoring the structure and function of cutaneous lymphatics may represent a novel therapeutic target for hypertension.

Discussion

We provide evidence for a role of the LEC–A_2AR in mediating skin lymphangiogenesis, sodium clearance, and BP. Lymphatic endothelial cell–A_2AR expression was increased in a positive relationship with skin lymphatics in HSD-induced hypertensive mice. Furthermore, A_2AR^LECKO have reduced lymphatic vessel density, increased skin sodium, and an exaggerated increase in BP with HSD. The A_2AR-induced lymphangiogenesis in high-salt conditions is regulated by MSK1-mediated VEGF-independent VEGFR2 internalization and activation (Figure 7G). Finally, we show that in hypertensive patients, there is a decrease in lymphatic vessels and an increase in skin sodium concentration. Furthermore, the expression of A_2AR, as well as phosphorylated VEGFR2 and MSK1, is reduced in skin tissue of hypertensive patients and is inversely correlated with Na⁺ content and concentration. Therefore, LEC–A_2AR-mediated VEGF-independent activation of VEGFR2 signaling mediates dermal lymphangiogenesis, sodium balance, and BP.

There is a causal relationship between dietary salt intake and BP. Reduction in salt intake from the current world average of ∼10 g/day to the World Health Organization recommended level of <5 g/day lowers BP and reduces the risk of cardiovascular disease and all-cause mortality.^23–25 Previous studies have confirmed high salt intake elevated BP in both rodents^7,26 and humans.^3,27 The molecular mechanisms of high salt intake and BP elevation are still unclear. There is growing evidence that skin is a large reservoir of Na⁺ ions which are stored in an osmotically inactive form on glycosaminoglycans.⁶ Local hypertonicity due to extensive accumulation of Na⁺ within the skin was reported to induce macrophage accumulation and subsequently to secrete VEGFC via upregulating transcription factor TonEBP, resulting in lymphangiogenesis within the skin.⁵ High levels of salt exposure also elevated BP via upregulating the NLRP3 inflammasome, pyroptotic and apoptotic caspases, and interleukin-1β transcription in monocytes.²⁶

The increased lymphangiogenesis in the skin seems to be adaptive in maintaining BP in high salt-load states, though one recent study revealed that dermal lymphatic vessels are not involved in salt storage or BP regulation.²⁸ However, more studies showed that administration of exogenous VEGFC, or VEGFC overexpression, attenuates HSD-induced BP elevation and related target organ damage.^7,29,30 Furthermore, HSD-induced extracellular sodium accumulation causes lymph flow increase, effect which is blocked by depletion of macrophages/monocytes.⁷ These studies suggest that lymphangiogenesis has an important role in BP control in high-salt conditions. Herein, we utilized two models of HSD: an angiotensin II and 8% HSD model, and a less intense mouse model with 4% salt diet. Blood pressures in HSD-treated mice were assessed by the tail–cuff method, as well as radiotelemetry, as the latter method was more sensitive to the smaller BP changes in the 4% salt diet–treated mice. Both models confirmed the association of lymphangiogenesis with BP.

A related question is whether the function of the skin lymphatics is affected by high dietary salt, a question that had not been addressed previously. We herein provide direct evidence that high-salt treatment inhibits the transport of Na⁺ from the skin to the LNs, associated with an increase in skin sodium and BP. Lymphatic endothelial cell–A_2AR activation mediates skin lymphangiogenesis and skin sodium clearance and downregulates BP. Our hypothesis was supported by observations of the effects of HSD or A_2AR activation on lymphatic transport of Na⁺, as assessed by changes in Na⁺ concentration over time in regional LNs after subcutaneous injection of NaCl. Methods with greater temporal and spatial resolution are needed to confirm and refine our observations.

A_2AR is abundantly expressed in immune cells and throughout the cardiovascular system. Activation of A_2AR is known to induce a Th2 response and the production of antiinflammatory factors, attenuating cardiovascular inflammation and related organ damage.¹⁷ Furthermore, A_2AR induces vasodilation by activating the release of endothelium-derived nitric oxide and by opening ATP-sensitive potassium channels to hyperpolarize vascular smooth muscle.³¹ We herein show that HLEC–A_2AR is increased and then decreased with time-dependent high-salt conditions in vitro. Though the number of skin samples is small, and skin Na⁺ concentrations and lymphangiogenesis might be affected by antihypertensive drugs or concomitant diseases, we observed a reduced LEC–A_2AR expression in the skin of hypertensive patients.

The molecular mechanism of high salt–mediated A_2AR expression needs further investigation. The core promoter region of the A_2AR gene has binding sites for the transcription factor cAMP response element–binding protein (CREB).³² Meanwhile, CREB is the target of salt-inducible kinases, which could be phosphorylated and activated responsive to high-salt conditions.³³ The detailed mechanism of CREB in regulation of A_2AR expression needs to be further explored in the future.

VEGF family members and their receptors are crucial for development and maintenance of both the blood and lymphatic vascular networks. Among these, VEGFC and VEGFD are the primary growth factors regulating the lymphatic vasculature via the VEGFR3 tyrosine kinase pathway.^34,35 VEGFC also activates VEGFR2 in LECs by promoting heterodimerization between VEGFR2 and VEGFR3, and regulates LEC survival, proliferation, migration, and lymphangiogenesis.^36–38 It is well established that VEGF-independent VEGFR upregulation in endothelial cells contributes to angiogenesis.³⁹ We herein provide direct evidence that VEGF-independent VEGFR2 phosphorylation and activation in LECs promote lymphangiogenesis in high-salt conditions, which is mediated by the A_2AR–MSK1 signaling pathway. MSK1 activates ERK1/2 and p38 MAPK pathways and endothelial cell function.^40,41 MSK1 contains an N-terminal kinase domain (NTKD) in the AGC kinase family and a C-terminal kinase domain (CTKD) from the calmodulin kinase family. MSK1 activation causes CTKD autophosphorylation of NTKD which can phosphorylate substrates such as histone H3 and transcription factors CREB and ATF1 in the nucleus.⁴⁰ Previous studies demonstrated that VEGFR2 internalization is required for downstream signaling to regulate endothelial cell migration, proliferation, and survival.^42,43 Here, we show that A_2AR mediates a VEGF-independent VEGFR2 endocytosis and phosphorylation and promotes LEC proliferation, migration, and tube formation in high-salt conditions.

Anti-VEGF–neutralizing antibody bevacizumab and VEGFR2 tyrosine kinase inhibitor fruquintinib are both used in the treatment of malignancies.^44,45 They may each cause hypertension as a side effect, because they each antagonize VEGF-mediated vasodilation.⁴⁴ However, they act differently on the VEGF pathway. Bevacizumab scavenges VEGF, whereas fruquintinib blocks phosphorylation of VEGFR.⁴⁶ Consistent with a VEGF-independent action of A_2AR as described above, we showed that during high-salt conditions, fruquintinib, but not bevacizumab, blocks A_2AR activation–induced dermal lymphangiogenesis and increases BP. These results were supported by our observations in cancer patients with HSD, though the number of subjects was small. Greater understanding of the vascular effects of the VEGF/VEGFR pathway could lead to novel therapeutic approaches to reduce the hypertensive effects of antiangiogenic therapies in cancer patients. However, the animal sample size in our study was low, which limited adequate power to ensure clinically important meaning or difference.

Our results suggest that LEC–A_2AR might be a potential therapeutic target in salt-sensitive hypertension because it mediates VEGF-independent activation of VEGFR2 signaling to increase dermal lymphangiogenesis and reduce skin sodium retention. However, A_2AR agonists are reported to have detrimental effects, which may require the development of more selective A_2AR ligands that preferentially act on skin lymphatics.⁴⁷ Moreover, previous studies demonstrated that the effects of adenosine receptors differed according to sex.^48,49 The animals used in the present study were male, and we will investigate the sex differences of LEC–A_2AR receptor in dermal lymphangiogenesis and sodium balance in the future. Finally, it is important to note that, whereas we have documented a VEGF-independent effect of LEC–A_2AR, our results do not exclude a role for VEGF in salt-sensitive hypertension as it enhances dermal lymphangiogenesis.

Our research reveals the importance of lymphatic endothelial A_2AR in the progression of high salt–induced hypertension. Lymphatic endothelial cell–A_2AR activation increases dermal lymphangiogenesis and skin sodium clearance by promoting phosphorylation of MSK1 and VEGFR2 endocytosis and activation. Notwithstanding the much more advanced complexity of human salt-sensitive hypertension in comparison with the mouse model, our novel findings suggest an attractive possibility that activating A_2AR-mediated lymphangiogenesis may be a potential therapeutic target in the treatment of salt-sensitive hypertensive patients.

Supplementary data

Supplementary data are available at European Heart Journal online.

Declarations

Disclosure of Interest

All authors declare no conflict of interest for this contribution.

Data Availability

All data associated with are present in the paper or the supplementary material online. All resources generated in this study are available upon request from the corresponding author upon providing a completed material transfer agreement.

Funding

This study was supported by the National Key R&D Program of China (2022YFA1104200), National Natural Science Foundation of China (82070456, 81922004, 81870180, and 81900434), Shanghai Science and Technology Commission (21XD1420500 and 19JC1414600), Shanghai Rising-Star Program (21QA1401400), Young Health Talents of Shanghai Municipal Health Commission (2022YQ069), Innovative research team of high-level local universities in Shanghai and a key laboratory program of the Education Commission of Shanghai Municipality (ZDSYS14005), Huangpu Municipal Health Commission (HLQ202102), National Institutes of Health (NIH) R01s HL133254 and HL148338, Cancer Prevention and Research Institute of Texas CPRIT RP150611, and Medical Engineering Joint Fund of Fudan University (YG2022-26).

Ethical Approval

Ethical Approval was not required.

Pre-registered Clinical Trial Number

None supplied.

References

Author notes