Tissue-specific expression of senescence biomarkers in spontaneously hypertensive rats: evidence of premature aging in hypertension

Animals

Four-week-old male spontaneously hypertensive rats (SHR/KyoMlac or SHR) (n = 40) and age/sex-matched normotensive wild-type rats (WMN/NrsMlac or WT) (n = 40) were acquired from the National Laboratory Animal Center of Thailand (NLAC), Mahidol University. Both groups of animals were housed in polycarbonate shoe box cages (two animals per cage) and the polycarbonate shelters were provided as enrichment devices. They were acclimatized in a controlled environment of 21 ± 1 °C, under 12-h light-dark cycle, 50–60% relative humidity. General health status was checked daily by veterinarians at the Central Animal Facility, Faculty of Science, Mahidol University (MUSC-CAF)—an AAALAC (Association for Assessment and Accreditation of Laboratory Animal Care)-accredited facility. They were fed standard laboratory food containing 1.0% calcium, 0.9% phosphorus and 4,000 IU/kg vitamin D (Perfect Companion Group Co., Ltd., Bangkok, Thailand) and reverse-osmosis (RO) water ad libitum. In an experiment that demonstrated the effect of antihypertensive agent on Timp1 expression, 24-week-old male SHR rats were randomly divided into two groups, i.e., (1) the vehicle group that was subcutaneously injected with distilled water, and (2) the treated group that was subcutaneously injected with 0.1 mg/kg/day propranolol everyday over a 12-week period (n = 8 per group). The experimental protocols have been approved by the Institutional Animal Care and Use Committee (IACUC), Faculty of Science, Mahidol University (No: MUSC61-050-452 and MUSC64-010-559). All studies related to animals were performed in accordance with relevant guidelines and regulations, including the ARRIVE guideline (https://arriveguidelines.org).

Blood pressure measurement and sample collection

Animals were housed at MUSC-CAF until reaching experimental endpoints at the age of 6, 12, 24, and 36 weeks, animals were randomly selected (Fig. 1B). Prior to euthanasia, assessment of systolic, diastolic and mean arterial blood pressures (SBP, DBP and MAP, respectively) was performed using non-invasive tail-cuff method (CODA; Kent Scientific Corp., Torrington, CT, USA) (n = 10/group). Briefly, animals were restrained in the restraining and acclimatized for at least 5 min, thereafter the blood pressure measurement was performed. At the end of the experiments, all animals were euthanized by intraperitoneal injection of 5 mg/kg xylazine (X-lazine; L.B.S. Laboratory Ltd., Part, Bangkok, Thailand) and 40 mg/kg tiletamine/zolazepam (Zoletil™ 100; Virbac Laboratories, Carros, France). After euthanasia, blood, heart, liver, kidneys, whole brain and long bones were collected. Serum samples were separated from whole blood for further analyses using enzyme-linked immunosorbent assay (ELISA) technique. Freshly isolated organs were kept frozen at –80 °C for RNA isolation and analyses of gene expression.

Tissue preparation and RNA isolation

Tissues from left ventricles and left lobe of liver were selected. Long bones were flushed with ice-cold phosphate-buffered saline to remove residual cells within bone marrow spaces. Bone specimens were chopped into small pieces. Whole brain tissues were placed with dorsal surfaces up on the cutting block. Razor blades were placed in fixed position to obtain coronal sections. In each section, hypothalamus was located and dissected as previously described (Heffner, Hartman & Seiden, 1980). Specimens from the kidneys were obtained after the surrounding connective tissues and renal capsules were carefully removed, they were then chopped into small pieces. Heart, liver, hypothalamus, and kidney were homogenized for RNA isolation following a protocol previously described (Tiyasatkulkovit et al., 2019). Bone tissues were homogenized using a Precellys Evolution Super Homogenizer with ceramic beads according to the manufacturer’s protocol (Bertin Instruments, Montigny-le-Bretonneux, France). Total RNAs were extracted using TRIZol reagent (Invitrogen, Carlsbad, CA, USA). Total RNA quality was assessed by measuring the absorbance at 260 and 280 nm using a NanoDrop-2000c spectrophotometer (Thermo Fisher Scientific, Waltham, MA, USA) and the 260/280 ratio ranged between 1.8 and 2.0.

Quantitative real-time reverse-transcription polymerase chain reaction (qRT-PCR)

Synthesis of cDNA was performed by reverse transcription of 1 μg total RNA using iScript cDNA synthesis kit (Bio-Rad, Hercules, CA, USA). Two housekeeping genes were included, i.e., 18S rRNA for heart, liver, and kidney samples and β-actin for long bones and hypothalamus samples. Consistency of reverse-transcription reactions was evaluated with variation coefficient less than (5%; n = 10 for each group). qRT-PCR and melting-curve analyses were performed using SsoFast Eva Green Supermix (Bio-Rad, Hercules, CA, USA) and QuantStudio 3 RT-PCR System (Applied Biosystems, Foster City, CA, USA). The PCR reactions were set for 40 cycles. Each cycle was fixed at 95 °C for 60 s, 54–58 °C for 30 s, and 72 °C for 30 s. Annealing temperatures for all primer pairs are listed in Table 1. These primers have been validated for specificity and efficiency using conventional RT-PCR. Relative gene expression was calculated based on 2^–ΔΔCt method (Livak & Schmittgen, 2001).

Enzyme-linked immunosorbent assay (ELISA)

Levels of TIMP metallopeptidase inhibitor 1 (Timp1) in the serum were determined by ELISA using Quantikine enzyme-linked immunosorbent assays (catalog no. RTM100; R&D systems, Minneapolis, MN, USA). We performed quality assessment of this technique. The intra- and inter-assay coefficients of variation were 3.1% and 6.7%, respectively. In addition, the detection limit was 0.0035 ng/mL. In brief, serum samples were diluted by 200-fold with the Calibrator Diluents (Calibrator Diluent RD 5–17). Rat Timp1 standards at various concentrations were used to obtain the standard curve with r² at 0.7921. Monoclonal antibody specific for rat Timp1 was used to pre-coat a microplate and serum samples were added for 2 h. Next, polyclonal antibody specific for rat Timp1 was added for 2 h at room temperature. Substrate solution was then added to each well and incubated for 30 min. The color reagents were used to stop reaction. The optical density was determined at 450 nm by microplate reader (model M965; Metertech, Taipei City, Taiwan). Serum Timp1 concentrations were expressed in ng/mL.

Determination of plasma levels of O-methylated metabolites

Plasma levels of O-methylated metabolites of the catecholamines, i.e., metanephrine, normetanephrine and 3-methoxytyramine, were determined by liquid chromatography tandem mass spectrometry (LC-MS/MS) at Ramathibodi Hospital, Mahidol University, Bangkok, Thailand.

Immunofluorescent staining for p16^Ink4a

Liver was fixed overnight in 4% paraformaldehyde in 0.1 M phosphate-buffered saline (PBS). After being embedded in paraffin, the specimens were cut cross-sectionally into 4-µm sections. All sections were incubated in sodium citrate solution (pH 6.0) at 95 °C for 40 min and then blocked with 5% fetal bovine serum in PBS containing 0.1% Tween-20 at room temperature for 2 h. Sections were then incubated with anti-CDKN2A/p16^Ink4a antibody (catalog no. 54210; Abcam, Waltham, MA, USA) with 1:250 dilution at 4 °C for overnight in moist chamber. They were then washed and incubated with goat anti-mouse IgG (catalog no. ab150116; Abcam, Waltham, MA, USA) with 1:500 dilution at room temperature for 1 h. Vectashield Plus antifade mounting medium (catalog no. H-1900; Vector Laboratories, Newark, CA, USA) containing DAPI at 1:5,000 dilution was applied, and fluorescent signals were visualized under a fluorescent microscope (Ni-U model eclipse, Nikon, Japan).

Synchrotron small-and wide-angle X-ray scattering (SAXS/WAXS)

Synchrotron—a cyclic electron accelerator (1.2 GeV) operated by Synchrotron Light Research Institute (SLRI), Nakhon Ratchasima, Thailand—was utilized to generate high-quality intense X-ray beams. To determine the collagen alignment of the bone, SAXS and WAXS measurements were performed at BL1.3W: SAXS/WAXS, SLRI. Fresh frozen femoral cortical bone specimens from SHR and WT at 24 weeks of age were collected. Each bone sample was held to the aluminum frame with Kapton tape during simultaneous SAXS/WAXS measurements. An exposure time of 180 s and X-ray wavelength of 1.38 Å were used in both SAXS and WAXS techniques. The distances to detector for SAXS and WAXS were 4,969 and 149 mm, providing q range of 0.072–0.725 nm^–1 and 2θ_Cu range of 20–55°, respectively, and these protocols were calibrated using SEBS block co-polymer for SAXS and 4-bromobenzoic acid for WAXS. The data were background subtracted by using Kapton tape measurement. The SAXS and WAXS scattering patterns were recorded with a MarCCD165 and LX170-HS detectors, respectively. The raw data were processed by the SAXSIT program, which was developed by SLRI staff to obtain scattering curves. 1D-SAXS analysis was used to characterize a d-spacing value, calculated through Bragg’s Law using the following equation:

$d=2n\pi /q$where d is d-spacing, n is the number of periods and q is the SAXS scattering curves of samples that were averaged to each of their corresponding types, i.e., WT and SHR, before calculating d-spacing.

The periodic structure of the arrangement of the intrafibrillar collagen molecules was determined by 2D-SAXS analysis. The 1D-SAXS curve was characterized by a d-spacing value. The mineral crystals of carbonated apatite, which contributed to diffraction peaks, were evaluated from WAXS curves. The crystal lengths along the c-axis were indicated from the full-width half maximum (FWHM) of the (002) reflection at 2θ_Cu about 29.7° and the crystal widths in the ab-plane were determined from the (310) reflection which is perpendicular to (002) at 2θ_Cu about 39.5° (Lange et al., 2011). Pseudo Voigt function was used to fit in each peak (Fig fitting) using FityK software (Wojdyr, 2010). Subsequently, the FWHM of the fitted peak was used to calculate the crystal dimensions using Scherrer’s equation (Patterson, 1939):

$D=K\lambda /(\beta cos\theta )$where D is the average crystal size corresponding to the reflection; K is the Scherrer constant depending on the crystal shape (typically about 0.9); λ is the X-ray wavelength; $\beta$ is the FWHM of the peak in radians $2\theta$ located at scattering angle $\theta$.

The instrumental broadening contribution of the hydroxy apatite crystallite was corrected by the previously reported method using the Lab-6 standard (Lange et al., 2011).

Histology of the aorta

After euthanasia, the aorta was collected and washed with ice-cold sterile normal saline solution, fixed in 4% paraformaldehyde for 12 h at 4 °C. Tissues were dehydrated and cleared by graded ethanol and xylene, respectively, and embedded in paraffin. Paraffin blocks were cut into 5-µm thickness and stained with hematoxylin and eosin (H & E) for histological examination. Slides were visualized under a light microscope (Ni-U model eclipse; Nikon, Tokyo, Japan) with cellSens image acquisition software (Olympus, Tokyo, Japan).

Transmission electron microscopy (TEM)

The freshly collected left ventricular cardiac tissue was analyzed for ultrastructural morphology of vascular endothelium. Cardiac tissue was fixed with 2.5% glutaraldehyde in 0.1 M phosphate buffered saline (PBS) at 4 °C for 24 h. The sample was washed three times in pre-chilled PBS at 4 °C for 5 min each. The post-fixation of samples was performed under darkness using 1% osmium tetroxide (OsO₄) in 0.1 M PBS for 2 h and then washed three times with 0.1 M PBS for 5 min each. Cardiac tissue was dehydrated with graded ethanol series (70%, 80%, 90%, 95% twice and 100% three times for 30 min each). Cardiac tissue was dehydrated using graded ethanol series ranging from 70% to absolute ethanol with each step lasting 30 min. Subsequently, the samples were permeated twice with pure propylene oxide (PO) and embedded in Araldite 502 resin. The resin-embedded tissue blocks were then sliced into 60–70 nm in thickness. These ultra-thin sections were mounted on copper grids and subjected to staining with uranyl acetate for 45 min followed by 0.1% lead citrate for 45 min. Finally, the prepared samples were observed under a TEM (model Tecnai 20G2; Philips, Cambridge, MA, USA) at the Microscopic Center, Faculty of Science, Burapha University, Chonburi, Thailand.

Statistical analysis

The results are expressed as means ± standard errors of mean (SEM). The two-group data were analyzed by unpaired Student’s t-test. Unless otherwise specified, statistical significance was considered when P < 0.05. All data were analyzed by GraphPad Prism 9 (GraphPad, San Diego, CA, USA).

SHR exhibited an elevation of blood pressure without systemic change in O-metabolites of catecholamines and Timp1 levels in the blood

First, we evaluated the model of systemic hypertension used in this study. Longitudinal assessment of blood pressure parameters in SHR revealed persistent elevation of all three parameters of blood pressure, i.e., systolic, diastolic and mean arterial pressure. Quantitative analysis showed that the elevations reached statistical significance as early as 6 weeks and persisted up to 36 weeks of age (Figs. 2A–2C). Our findings on the degrees of change in these parameters in SHR compared to those normotensive WT controls that characterized the model were consistent with previous report (Okamoto & Aoki, 1963). In addition, SHR manifested thickening of the smooth muscle layer and enlargement of the aorta as compared to WT (Figs. S1A and S1B). Nevertheless, the endothelium and underlying intimal layers remained intact without apparent histological abnormality. TEM analysis of the cardiac vessels also showed that the endothelial cells of SHR were apparently normal (Fig. S1C).

Chronic elevation in blood pressure found in SHR led us to ask whether there was a progressive increase in sympathetic activation and SASP production in these hypertensive rats. Here, we determined the levels of O-methylated metabolites of the catecholamines, i.e., metanephrine, normetanephrine and 3-methoxytyramine, as proxy indicators for sympathetic activation. Plasma levels of O-methylated metabolites in SHR did not differ from those of WT (Fig. 3A). Likewise, the levels of Timp1—a circulating SASP that indicated ECM remodeling (Arpino, Brock & Gill, 2015). in SHR and WT rats were not different in all age groups (Fig. 3B). These data suggested that elevated blood pressure in our SHR model did not associate with widespread activation of sympathetic nervous system or systemic production of Timp1.

Upregulation of senescence-associated CDKI gene expression was found in heart and liver of SHR

The absence of systemic changes in Timp1 levels suggested that the onsets of hypertension-induced senescence were not synchronized among organs. Alternatively, susceptibilities to hypertension-induced senescence in those tissues were variable. Thus, we used the organ-based approach to dissect changes in the expression of various senescence-associated markers. We first focused on the heart since cardiac tissue remodeling and hypertrophy caused by pressure loads have been reported in hypertensive patients and animal models of essential hypertension (Lorell & Carabello, 2000; Pagan et al., 2022). Indeed, the expression of transcripts encoding two senescence-associated CDKIs, Cdkn2a and Cdkn1a, was strongly induced in the cardiac tissue of the 6-week-old SHR (Figs. 4A and 4E). This overt upregulation of Cdkn2a and Cdkn1a remained evident at 12, 24, and 36-week-old SHR (Figs. 4B–4D, 4F–4H). These increases in the expression of CDKI genes were consistent with the temporal profile of elevated blood pressure found in SHR.

Identification of senescence markers in cardiac tissue but not in the blood of SHR implied that elevated blood pressure selectively induced premature aging process in certain tissues. To characterize these changes, assessment of Cdkn2a and Cdkn1a expression in the major vascular beds that were likely to be impacted by systemic hypertension, i.e., those of the liver, bone, kidney and hypothalamus. Indeed, we observed different changes in the pattern of Cdkn2a and Cdkn1a expression among these selected tissues (Figs. 4A–4H). Similar to the cardiac tissues, the expression of Cdkn2a and Cdkn1a mRNAs in the liver of SHR was significantly increased throughout the study. Consistently, an immunostaining analysis in hepatocytes showed an increase in p16 protein expression in SHR compared to WT (Fig. S2). On the other hand, these CDKI genes were downregulated in the hypothalamus of SHR. Likewise, the expression of Cdkn1a mRNA was significantly suppressed in the renal tissues. There was no change in Cdkn1a mRNA expression in bone tissues. Taken together, it was likely that chronic hypertension found in SHR initiated organ-specific senescence (i.e., heart and liver) via induction of CDKI expression.

Changes in SASP-related genes were consistent with the expression profile of CDKI gene expression in SHR

Next, we investigated the progression of senescence in these selected set of tissues by assessing the levels of SASPs. To further verify that induction of senescence in SHR was organ-specific, we evaluated the expression of Timp1 gene in those five organs. Indeed, mRNA levels of Timp1 gene were significantly upregulated in the heart and liver of SHR from the age of 6–36 months compared to age-matched WT rats (Figs. 5A–5D). Interestingly, the upregulation of Timp1 gene expression in SHR was markedly diminished by propranolol treatment (Fig. S3). Expression of Timp1 gene in bone revealed biphasic pattern, whereby the levels were initially decreased at 6 weeks and increased at 24 and 36 weeks of age. Furthermore, we found that the levels of Timp1 mRNA were significantly decreased in kidneys and hypothalamus in all agegroups. These data were similar to the pattern of CDKI mRNA expression. Figures 5E–5H show the levels of another matrix-associated senescence marker, matrix metallopeptidase 12 (Mmp12). The profile of Mmp12 gene expression in liver and kidneys (Figs. 5E–5H) were similar to those of Timp1 gene expression. Notably, the expression of Mmp12 gene was significantly increased in hepatic tissues of SHR. Levels of Mmp12 mRNA in renal tissue were decreased at 6 and 12 weeks old. These findings further support the association between elevated blood pressure and senescence phenotypes in heart and liver. In contrast, appearance of senescence phenotypes was delayed in tissues from kidneys and hypothalamus of SHR.

Inflammatory cytokines, such as IL-6 and CXCL1, were also included as SASPs. We found an early increase in Il6 mRNA levels in the heart of 6–12-week-old SHR (Figs. 6A and 6B). Upregulation of Il6 gene was further identified in the liver of 24–36-week-old SHR (Figs. 6C and 6D). Expression of Il6 mRNA was significantly decreased in kidney of SHR (at 6, 12 and 36 weeks of age) and hypothalamus (at 36 weeks of age). Consistently, expression of Cxcl1 mRNA was markedly upregulated in the liver of SHR at 12, 24, and 36 weeks of age (Figs. 6E–6H). Moreover, we found that the levels of Cxcl1 transcripts were increased in bone of SHR at 36 weeks of age (Fig. 6H).

SAXS/WAXS indicated misalignment of bone collagen fiber in SHR

Bone is a mineralized biological material that consists of cells embedded in the ECM. The mature mineral matrix is organized in a complex hierarchical network made from two major nanophases including collagen fibrils and hydroxyapatite, elongated platelet-like carbonated calcium phosphate particles (Hong et al., 2022). Because of the upregulation of Timp1 expression in the bone of SHR at the age of 24 and 36 weeks, it might also alter the ECM remodeling in the bone. Therefore, the collagen fiber alignment of the bone was determined by synchrotron SAXS/WAXS. As shown in Fig. 7A, SAXS analysis demonstrated a smaller peak at 1^st through 6^th orders in SHR, which suggested the poorly order and broad distribution of d-spacing in collagen fiber at 65.5 nm in SHR as compared to WT. In addition, the 2D-SAXS analysis showed the collagen oriented predominantly diagonal in WT, whereas SHR showed random orientation of collagen fiber (Fig. 7B). A ring that indicated collagen alignment in WT was more preferred orientation than that in SHR. Moreover, WAXS analysis of crystallite dimensions revealed a decrease in the crystal length and width in the bone of SHR (Figs. 7C–7D), which was calculated from FWHM of the (002) and (310) reflections (i.e., they indicated more crystallized collagen structure in WT as compared to SHR). Taken together, SHR was associated with the presence of disoriented collagen fibers and lower crystal dimensions in bone, which could be a consequence of hypertension and/or senescence.