Tacrolimus Induced Hypertension and Vascular Remodeling Includes Mechanisms of Cellular Senescence—The Protective Effect of Valsartan

The authors declare that all supporting data are available within the article and its Appendix. Upon reasonable request, additional information will be provided. S1

All animal procedures and experimental protocols were performed according to the guidelines from Directive 2010/63/EU of the European Parliament on the protection of animals used for scientific purposes or the NIH Guide for the Care and Use of Laboratory Animals and were approved by the Institutional Animal Care and Ethical Committee of Sun Yat‐sen University (IACUC‐2023‐0011).

Male C57Bl/6 mice (age: 8–10 weeks) were obtained from SLAC laboratory animal company (Shanghai, China). The mice were kept in climate‐controlled environments with a 12‐h light/dark cycle and fed standard laboratory food and tap water. Mice were randomly divided into 7 groups (Figure S1), including control, valsartan, tacrolimus, tacrolimus + valsartan, tacrolimus + Amlo, ABT‐263 and tacrolimus + ABT‐263. After 2 weeks of acclimatization, one group of mice received valsartan (30 mg/kg/day [31], V129241, Aladdin, China) by gavage while another group received Tac (5 mg/kg/day, HY‐13756, MCE, USA) intraperitonially. A third group received both valsartan and Tac. The fourth group received Amlo (10 mg/kg/day [32], HY‐B0317, MCE, USA) by gavage. Amlo is a common antihypertensive drug. The fifth group received ABT‐263, which has shown senolytic effects in several experimental models [33, 34]. The sixth group received a daily intraperitoneal injection of Tac as well as 4 cycles of treatment with ABT‐263 (37.5 mg/kg/day [35], HY‐10087, MCE, USA), where each cycle consisted of a daily intraperitoneal injection on 5 consecutive days followed by a 2‐day hiatus. The control group received the solvent. All groups were treated for 4 weeks. Mice were anesthetized using isoflurane inhalation, euthanized by cervical dislocation, and dissected to collect serum, kidneys and mesenteric arteries for further experiments.

Working solutions of all drugs were prepared such that they could be administered at a volume of 4 mL/kg. Tac working solution (1.25 mg/mL) was prepared by diluting the stock solution (Tac 100 mg/mL in DMSO) at a 1:80 (v/v) ratio with a mixture of saline and PEG‐300 (2:1). Working solutions of valsartan (7.5 mg/mL) and Amlo (2.5 mg/mL) were prepared by diluting the respective DMSO stock solutions (100X) in corn oil. The working solution of ABT‐263 (9.375 mg/mL) was obtained by diluting the stock solution (937.5 mg/mL in DMSO) at a 1:100 (v/v) ratio with 40% PEG‐300 in saline.

Mice were randomly allocated to different experimental groups using a computer‐generated random number sequence. The investigators involved in drug administration and outcome assessment (blood pressure measurement, tissue collection, and data analysis) were blinded to the group allocation throughout the experiment and analysis.

To evaluate vascular reactivity in renal afferent arterioles (Af‐Art), a microperfusion system was used. The isolation and microperfusion procedures have been previously described [18, 36]. In brief, mice were anesthetized with isoflurane. The kidneys were removed and sliced along the corticomedullary axis. The kidney slices were then kept in ice‐cold Dulbecco's modified Eagle's medium (DMEM). Next, the kidney slices were microdissected under a stereomicroscope. A single Af‐Art with attached glomerulus was isolated and then transferred to a chamber filled with DMEM (37°C) and placed on the stage of an inverted microscope. The Af‐Art was perfused using a micromanipulator system with concentric holding and perfusion pipettes. The vessel was equilibrated at 37°C for 15 min. Subsequently, 100 mM KCl was applied to evaluate the vascular reactivity and washed out. After 5 min, cumulative concentration response curves to Ang II (10⁻⁶ to 10⁻¹² mol/L) or PE (10⁻⁵ to 10⁻¹⁰ mol/L) were obtained. For vasodilatation, PE (10⁻⁵ mol/L) was applied to pre‐constrict Af‐Art, followed by the generation of cumulative concentration‐response curves to ACh (10⁻⁵ to 10⁻⁹ mol/L). All the experiments were performed within 2 h after the mice were killed. Only one or two Af‐Art were used from each mouse.

Using wire myograph, we assessed vascular reactivity in mesenteric arteries as previously described [18, 37]. First, mice were anesthetized with isoflurane and sacrificed. The abdominal cavity was opened, and the second branch mesenteric arteries were dissected under a stereomicroscope. The isolated mesenteric arteries were cut into rings (2 mm in length) and were kept in ice‐cold Krebs–Henseleit solution with the following composition (mmol/L): NaCl 112, KCl 5, NaHCO₃ 25, NaH₂PO₄ 1, MgCl₂ 0.5, CaCl₂ 2.5 and glucose 11.5. Then, the mesenteric arteries were mounted onto the wire myograph system. After 10 min of equilibration at 37°C, the resting tension was set according to the manufacturer's protocol. To confirm vessel reactivity, high‐potassium solution (KPSS, PSS with 80 mM KCl) was applied three times. Finally, cumulative concentration response curves were obtained to Ang II (10⁻⁶ to10⁻¹² mol/L), PE (10⁻⁵ to 10⁻¹⁰ mol/L) and ACh (10⁻⁵ to 10⁻⁹ mol/L).

Blood pressure in conscious mice was measured using non‐invasive tail‐cuff plethysmography (BP‐2000; Visitech Systems, NC, USA), as previously described [38]. Mice were acclimatized to the measurement process. After 3 days of training, the mice received valsartan, ABT‐263, and/or tacrolimus for 30 days, and blood pressure was measured every 2 days. Amlo combined with valsartan was administered for 28 days, and blood pressure was measured at the endpoint (Day 28).

Blood pressure in anesthetized mice was measured using a pressure transducer attached to a carotid artery cannula, as we previously described [18, 38]. First, mice were anesthetized with isoflurane (2%) and positioned on a temperature‐controlled table to maintain their body temperature at 37.0°C ± 1.0°C. Then, the left carotid artery was isolated and cannulated by a polyethylene catheter (PE‐10), which was connected to a Powerlab system (ADInstruments, CO, USA) to record blood pressure for 10 min.

Human umbilical vein endothelial cells (HUVEC, ATCC‐CRL‐4053, Manassas, USA) or mouse aortic vascular smooth muscle cells (MOVAS, ATCC‐CRL‐2797, Manassas, USA) were cultured in RPMI‐1640 medium (Gibco, NY, USA) supplemented with 10% fetal bovine serum (Tianhang Biotech, Zhejiang, China) and 1% penicillin–streptomycin (New Cell & Molecular Biotech, Suzhou, China) at 37°C, in a 5% CO₂ humidified environment.

Single Af‐Art with attached glomerulus were isolated as described above. The glomerulus was removed and Af‐Art (without glomerulus) were transferred into RLT buffer (RNeasy Mini Kit; Qiagen, The Netherlands). mRNA was extracted from Af‐Art, renal tissue, mesenteric arteries, HUVECs, and MOVAS as previously described. Total RNA was isolated using TRIzol Reagent (Invitrogen, Thermo Fisher Scientific; Cat. 15596026) and purified with the RNeasy Mini Kit (QIAGEN; Cat. 74104), including on‐column DNase digestion (RNase‐Free DNase Set; QIAGEN; Cat. 79254). RNA quantity/purity was measured by NanoDrop and Qubit using the Qubit RNA HS Assay Kit (Thermo Fisher Scientific; Cat. Q32852↗). Integrity was evaluated on a Qsep400 Bio‐Fragment Analyzer (BiOptic) with the RNA (R1) Cartridge, 4‐channel (Cat. C405110), following the user guide: heat‐denature at 70°C for 2 min, then place on ice for ≥ 5 min before loading; results were reported as RNA Quality Number (RQN) by Q‐Analyzer software. Samples with an RQN or RIN ≥ 9.0 were used for qPCR. The mRNA was then reverse transcribed into cDNA using a commercial kit (Bio‐rad, CA, USA). SYBR Green supermix (Biorad, CA, USA) was used to perform quantitative polymerase chain reaction (qPCR). All primers used are listed in Table S1. Data were analyzed and quantified with the comparative Ct method (2^−ΔΔCt) to calculate the relative mRNA expression level.

Isolated mesenteric arteries and kidney tissue were fixed in 4% formaldehyde, embedded in paraffin and sliced at 5‐μm thickness for histological and immunohistochemical staining. Hematoxylin & eosin (H&E), Masson, Elastin van Gieson (EvG) staining methods were used for histopathological analyses of mesenteric arteries, while H&E and Masson staining were applied to kidney tissue. Measurements of luminal radius, media thickness, intensity of collagen, and scoring for hyalinosis were conducted as described in detail by Gan et al. [5] and Chiasson et al. [39], respectively. Briefly, using ImageJ software, measurements were taken from four anatomical positions (12, 3, 6, and 9 o'clock) per slide and averaged to determine luminal radius, media thickness, collagen intensity, and media‐to‐lumen ratio. Hyalinosis was scored in 10 randomly selected fields per section for both the left and right kidneys of each mouse using a 4‐point scale: 1 = hyalinosis evident in less than 10% of the field, 2 = 10%–25%, 3 = 25%–50%, and 4 = more than 50%. Immunohistochemistry was performed following standard protocols. Briefly, slices were deparaffinized, washed twice with xylene, and later rehydrated. Subsequently, the sections were blocked and incubated overnight at 4°C with one of the following primary antibodies: anti‐renin (1:2000, Abcam, AB212197↗, UK), anti‐AGTR1 (1:100, Bio Excellence, 25 343‐1‐AP, China), anti‐P16 (1:200, Santa Cruz Biotechnology, sc‐1661, USA), anti‐P21 (1:200, Servicebio, GB12153, China) and anti‐γH2AX (1:200, Servicebio, GB111841↗ ‐100, China) at 4°C overnight. The next day, primary antibodies were removed, and the slides were washed 5 times with PBS before incubation with the secondary antibodies for 1 h. Finally, the sections were washed three more times with PBS and imaged.

Blood samples were collected from the retro‐orbital venous sinus from anesthetized mice and allowed to clot at room temperature for 30 min. After centrifugation at 3000 rpm, 4°C for 15 min, the supernatant plasma was collected and stored at −80°C until analysis. Plasma Ang II and renin levels were detected using commercial ELISA kits: Mouse Ang II ELISA Kit (E‐EL‐M2612, Elabscience Biotechnology, China) and Mouse REN (Renin) ELISA Kit (E‐EL‐M0061, Elabscience Biotechnology, China), respectively, according to the manufacturer's instructions.

HUVEC or MOVAS cells were cultured as described above. Cells were seeded onto 96‐well plates. After 24 h, cells were treated with Ang II (10⁻⁶ mol/L HY‐13948, MCE, USA), valsartan (10 μmol/L V129241, Aladdin, China), ABT‐263 (1 μmol/L HY‐10087, MCE, USA) and/or Tac (5 μmol/L HY‐13756, MCE, USA) for 24 h. Levels of SA‐β‐galactosidase (SA‐β‐gal) in the cells and in mesenteric arteries was measured using a Senescence β‐Galactosidase Staining Kit (Beyotime, Shanghai, China) according to the manufacturer's instructions.

To ensure unbiased evaluation, the individuals who administered the treatments and conducted the daily observations were different from the individuals who performed the data analysis. The analyst was kept blind to the group allocation until the final statistical tests were completed.

Statistical analyses were performed using Prism 10. One‐way ANOVA or two‐way ANOVA followed by Dunnett's post hoc test or Tukey's post hoc test for multiple comparisons, respectively. 2‐tailed Student's t‐test was performed when only two groups were compared. Data are presented as mean ± SEM, and a p < 0.05 is considered statistically significant.

mRNA levels of angiotensinogen (AGT), renin, Ang II receptor type 1a (AT₁Ra), and angiotensin converting enzyme (ACE) were elevated in renal tissue of Tac‐treated mice (Figure 1A). Additionally, expression levels of AT₁Ra and renin were significantly increased in Af‐Art (Figure 1B) from Tac‐treated mice. Plasma levels of Ang II (43.71 pg/mL versus 89.65 pg/mL, p = 0.007, Figure 1C) and renin (25.82 ng/mL versus 37.22 ng/mL, p = 0.0041, Figure 1D) were also significantly elevated compared to untreated animals. Immunohistochemical analyses confirmed these findings: in Tac‐treated mice, the area of renin positive staining was more pronounced than in controls (Figure 1E). The AT₁Ra positive staining in renal small arteries (Figure 1F) and mesenteric arteries (Figure 1G) did not differ between Tac‐treated mice and controls.

We investigated whether RAS inhibition with valsartan can mitigate the adverse effects of Tac on vascular function and structure. Contractile responses to Ang II and PE in mesenteric arteries were significantly enhanced in Tac‐treated mice, while ACh‐induced vasodilation was impaired (Figure 2A–C). Valsartan treatment improved both contractile and relaxation responses. Histological analyses using H&E, Masson's trichrome, and EVG staining revealed that Tac treatment increased medial thickness, media‐to‐lumen ratio, and collagen deposition in mesenteric arteries—all of which were ameliorated by valsartan (Figure 2D). No changes were observed in elastin abundance under Tac treatment.

We further assessed the effect of chronic administration of Tac on vascular structure and function in Af‐Art. Chronic Tac administration increased the constriction of Af‐Art to Ang II (Figure 3A,B), while PE responses were unaffected (Figure 3C). However, ACh‐induced dilation was significantly suppressed (Figure 3D). Co‐administration of valsartan effectively reversed these functional impairments (Figure 3B,D).

Histological examination using H&E and Masson staining revealed structural damage in preglomerular vessels (Figure 3E–I) and small renal arteries (Figure S2) from Tac‐treated mice, including increased media thickness, media‐lumen ratio, collagen deposition, and arteriolar hyalinosis. Valsartan largely reversed these structural alterations (Figures 3E–I and S2).

We next examined whether valsartan or Amlo could mitigate Tac‐induced hypertension. Tac treatment increased both systolic and mean arterial blood pressure in conscious mice (Figure 4A,C), while diastolic pressure and heart rate remained unchanged (Figure 4B,D). Similar elevations in systolic and mean blood pressure were observed in anesthetized mice (Figure 4E–G). Valsartan and Amlo effectively rescued the Tac‐induced hypertension (Figures 4A–G and 5B). Tac, valsartan, Amlo, or their combination did not significantly alter body weight (Figure 5A).

To explore the role of cellular senescence in the protective effects of valsartan, we treated HUVECs (Figure 6A,B) and MOVAS (Figure 6D,E) cells with Tac, valsartan, and/or Ang II, followed by SA‐β‐gal staining and qPCR for p16 and p21 (Figures 6C,F, and S3A–D). Mesenteric arteries were isolated from the 5 groups of mice and stained for SA‐β‐gal (Figure 7A). The number of SA‐β‐gal positive cells was significantly increased following Tac or Ang II treatment and was markedly reduced by co‐treatment with valsartan or Amlo, but the anti‐senescent effect of Amlo was less than valsartan. Immunohistochemical and qPCR analysis further confirmed increased expression of senescence markers (p16, p21, and γH2AX) in mesenteric arteries, renal small arteries, Af‐Art, and renal tissue from Tac‐treated mice (Figures 7B–E and S4–S6). Valsartan normalized the expression of these markers. Amlo also had a predominantly beneficial effect on the expression of these markers (Figure 7B,C). RNA integrity was tested in renal tissue and mesenteric artery (Figure S7).

To further evaluate the role of cellular senescence in Tac‐induced microvascular injury and hypertension, the senolytic agent ABT‐263 was utilized. SA‐β‐gal staining showed that ABT‐263 suppressed Tac‐ and Ang II‐induced senescence in both HUVECs and MOVAS cells (Figure 8A–D). ABT‐263 also alleviated vascular damage in mesenteric, Af‐Art, and renal arteries (Figures 9A–D and S8). Wire myography and microperfusion studies demonstrated that ABT‐263 reversed Tac‐induced changes in vascular reactivity to both constrictors and dilators in mesenteric arteries and Af‐Art (Figure 9E–H). Furthermore, ABT‐263 significantly reduced Tac‐induced hypertension (Figure 9I–L).

Long term Tac treatment activated the classical RAS pathway, as indicated by increased expression of its key components (Figure 1). These findings are consistent with previous studies, demonstrating elevated renin activity and Ang II concentrations in both plasma and kidney tissue following CNI treatment [15, 28, 29]. Thus, our model may reflect Tac‐induced RAS activation. The exact mechanisms underlying RAS stimulation remain debated and can involve both direct induction of renin production and indirect effects such as intrarenal vasoconstriction and reduced renal perfusion [40]. Tac administration led to structural remodeling in both systemic and renal arterioles, accompanied by substantial vascular dysfunction (Figure 1). Renal arteriolar remodeling is a hallmark of CNI nephropathy in both clinical and experimental settings [39, 41, 42]. We observed that Tac‐induced remodeling of Af‐Art is associated with increased contractility and reduced dilator responses (Figure 3). A similar functional impairment was noted in small mesenteric resistance vessels (Figure 2). These alterations are likely to contribute to the development of hypertension by increasing total peripheral resistance and reducing renal perfusion [43]. Although not all studies agree on vascular functional outcomes, there is considerable evidence supporting heightened vasoconstrictor sensitivity and impaired vasodilator responses in Tac treated animals [17, 18, 19, 41, 44, 45, 46, 47, 48, 49, 50].

In our study, valsartan attenuated both the structural and functional vascular impairments induced by Tac (Figures 2 and 3), suggesting that RAS mediates a significant portion of these effects. This conclusion is further supported by previous animal studies and clinical data demonstrating the beneficial effects of RAS inhibitors in kidney transplant recipients [16, 51, 52]. A cardinal feature of CNI nephropathy is renal arteriolar hyalinosis, which correlates inversely with GFR and renal blood flow, and directly with blood pressure [53, 54]. We show that valsartan ameliorated Tac‐induced renal arteriolar hyalinosis, implicating RAS in its pathogenesis and highlighting the protective effect of RAS inhibition (Figure 3). RAS activation appears to be responsible for the observed hypertension in the mice, which was alleviated by valsartan (Figure 4).

Sodium retention and hypertension are well‐known side effects of CNI therapy (see review [55]). Tac enhanced vasoconstriction and reduced vasodilation in Af‐Art, promoted arteriolar hyalinosis, and increased medial thickening, all of which contribute to elevated preglomerular resistance. Notably, increased preglomerular vascular resistance has been shown to promote sodium and water retention in RAS‐activated states [56, 57].

Our study highlights the role of RAS in Tac‐induced vascular remodeling and dysfunction. Ang II, the primary effector of RAS, exerts its effects via the AT₁ receptor, triggering multiple intracellular signaling cascades—including PKC, ERK, and MAPK pathways—that promote inflammation, fibrosis, and dysfunction in endothelial and vascular smooth muscle cells [58, 59]. In vitro, Tac and Ang II increased the levels of the senescence marker SA‐β‐gal in cultured HUVECs and MOVAS cells and in mesenteric arteries (Figure 6). These effects were prevented by valsartan, suggesting that Tac promotes senescence in vascular cells via RAS activation. Notably, the combined administration of Tac and Ang II did not exacerbate senescence beyond the individual treatments, but valsartan effectively mitigated the effect in both conditions, indicating that local RAS activation is a key mediator. Tac‐treated mice exhibited increased expression of senescence markers (p16, p21, and γH2AX) on both mRNA and protein levels in mesenteric arteries and Af‐Art, which was normalized by valsartan (Figure 7). The senolytic agent ABT‐263 also attenuated Tac‐induced senescence, vascular dysfunction, and hypertension, further underscoring the pathophysiological role of senescence in Tac‐related vascular injury (Figures 8 and 9). Ang II promotes premature senescence via AT₁ receptor mediated mechanisms, primarily by inducing oxidative stress and inflammation, two well‐established triggers of cellular senescence [60, 61, 62]. There is a strong link between vascular senescence and hypertension; numerous studies in both hypertensive patients and animal models demonstrate that vascular smooth muscle senescence contributes to vascular remodeling and elevated blood pressure [63, 64, 65, 66, 67]. Our data show that senescence in small mesenteric arteries and renal Af‐Art is a significant contributor to Tac‐induced vascular pathology.

Tac induces hypertension through a combination of mechanisms, among them increased vessel contractility and decreased dilation as well as water and sodium retention [17, 18, 19, 20, 21, 22, 23, 24, 25]. The present study shows activation of the RAS, which may mediate a large part of Tac‐induced effects, including induction of cellular senescence. Application of the antihypertensive drug Amlo reduced senescence marker expression; however, to a lesser extent compared to valsartan (Figure 7). This observation supports the idea that the anti‐senescence effect of valsartan may include a blood pressure‐independent component. Taken together, Tac treatment activates RAS, causing microvascular injury and hypertension in mice, together with increased expression of cellular senescence markers. Treatment with the senolytic agent ABT‐263 significantly reduces these side effects. In addition, valsartan attenuates these effects through its antihypertensive and well‐known cellular effects, which may include anti‐senescence.