Skip main navigation

Transcriptomic Response in the Heart and Kidney to Different Types of Antihypertensive Drug Administration

Originally publishedhttps://doi.org/10.1161/HYPERTENSIONAHA.121.18026Hypertension. 2022;79:413–423

Abstract

Certain classes of antihypertensive drug may exert specific, blood pressure (BP)-independent protective effects on end-organ damages such as left ventricular hypertrophy, although the overall evidence has not been definitive in clinical trials. To unravel antihypertensive drug-induced gene expression changes that are potentially related to the amelioration of end-organ damages, we performed in vivo phenotypic evaluation and transcriptomic analysis on the heart and the kidney, with administration of antihypertensive drugs to two inbred strains (ie, hypertensive and normotensive) of rats. We chose 6 antihypertensive classes: enalapril (angiotensin-converting enzyme inhibitor), candesartan (angiotensin receptor blocker), hydrochlorothiazide (diuretics), amlodipine (calcium-channel blocker), carvedilol (vasodilating β-blocker), and hydralazine. In the tested rat strains, 4 of 6 drugs, including 2 renin-angiotensin system inhibitors, were effective for BP lowering, whereas the remaining 2 drugs were not. Besides BP lowering, there appeared to be some interdrug heterogeneity in phenotypic changes, such as suppressed body weight gain and body weight-adjusted heart weight reduction. For the transcriptomic response, a considerable number of genes showed prominent mRNA expression changes either in a BP-dependent or BP-independent manner with substantial diversity between the target organs. Noticeable changes of mRNA expression were induced particularly by renin-angiotensin system blockade, for example, for genes in the natriuretic peptide system (Nppb and Corin) in the heart and for those in the renin-angiotensin system/kallikrein-kinin system (Ren and rat Klk1 paralogs) and those related to calcium ion binding (Calb1 and Slc8a1) in the kidney. The research resources constructed here will help corroborate occasionally inconclusive evidence in clinical settings.

High blood pressure (BP) or hypertension is a major risk factor for cardiovascular disease. Effective control of BP reduces the risk for end-organ damages such as stroke, left ventricular hypertrophy (LVH), heart failure, and chronic kidney disease (CKD). There are several classes of antihypertensive drug, with each acting through different targets and pathways. Studies have compared the antihypertensive effects on end-organ damages, on the assumption that certain drug classes exert specific, BP-independent organ-protective effects.1 The overall evidence has not been definitive, although some morbidity outcomes in clinical trials appear to differ between antihypertensive classes when used as first-line drugs.2 For example, moderate-certainty evidence shows that renin-angiotensin system (RAS) inhibitors (which include ACE [angiotensin-converting enzyme] inhibitors and ARBs [angiotensin receptor blockers]) and calcium-channel blockers do not differ for all-cause death, whereas RAS inhibitors decrease heart failure risk and contrarily increase stroke risk compared with calcium-channel blockers.

LVH, the most common complication of hypertension, is believed to be a maladaptive response to chronic pressure overload and constitutes a major risk factor for heart failure.3 This risk can be partially reversed by regression of LVH. However, it remains unclear whether there are appreciable differences in LVH regression among antihypertensive drug classes. A meta-analysis,4 involving randomized controlled trials conducted before 2009, demonstrates that RAS inhibitors are superior to other drug classes in the effect on LVH regression, although the findings are not consistently supported in all randomized controlled trials thereafter.3

Moreover, CKD and hypertension are closely interlinked. Chronic BP elevation can lead to impaired kidney function, which can conversely lead to worsening BP control. Hence, management of hypertension is important in patients with CKD. Certain antihypertensive classes are reported to provide reno-protective action independent of their BP-lowering effects. For example, RAS blockade appears to offer BP-independent reduction in proteinuria5 and is suggested to be first-line therapy for patients with CKD.6 Still, evidence from large randomized controlled trials remains insufficient to consolidate the renal protection of RAS blockade.

Accordingly, to unravel antihypertensive drug-induced gene expression changes that are potentially related to the amelioration of end-organ damages, we perform in vivo phenotypic evaluation and transcriptomic analysis on 2 target organs, heart and kidney, with administration of 6 antihypertensive drugs to 2 inbred rat strains: the stroke-prone spontaneously hypertensive rat (SHRSP) and its nonhypertensive control, Wistar Kyoto rats (WKY).7 We report experimental evidence for the presence of interdrug phenotypic heterogeneity and a global picture about genes showing prominent mRNA expression changes after treatment with different classes of antihypertensive drug.

Methods

The data that support the findings of this study are available from the corresponding author upon reasonable request.

Animal Procedures

All animal experiments conformed to the Guidelines for Animal Experiments of National Center for Global Health and Medicine and were approved by the Animal Research Committee of National Center for Global Health and Medicine.

This study used male rats of SHRSP and WKY, which were supplied by the Disease Model Cooperative Study Association (Kyoto, Japan). Rats were weaned at 4 weeks after birth and placed on normal rat chow (SP diet, Funabashi Farm, Japan). A selected number of rats were treated with antihypertensive drugs for 4 weeks from 12 to 16 weeks of age. All rats were laboratory animals and treated in compliance with institutional regulations.

Pharmacological Intervention

We chose 6 classes of antihypertensive drug: enalapril from ACE inhibitors, candesartan from ARBs, hydrochlorothiazide from diuretics, amlodipine from calcium-channel blockers, carvedilol from vasodilating β-blockers and hydralazine. We first conducted a pilot study, based on which we determined an appropriate dose for each drug.7 Systolic BP was measured every 3 to 4 days from 12 to 16 weeks of age by the tail-cuff method, in which 3 consistent BP readings were taken and averaged for each session. Thereafter, at 16 weeks of age, the rats were killed under pentobarbital anesthesia (200 mg/kg via intraperitoneal infusion), and the organs (kidney, heart, and retroperitoneal fat pad tissue) were excised for RNA analysis. The heart weight divided by body weight (Bw) (or adjusted heart weight) was used as a phenotypic variable reflecting LVH. The 24-hour proteinuria was also measured at 2 time points, before and after antihypertensive medication. Blood samples were drawn from the tail vein of the rats in an overnight (16-hour) fasting state.

Enalapril and hydrochlorothiazide were provided by Towa Pharmaceutical Co Ltd (Osaka, Japan), candesartan was provided by Takeda Pharmaceutical Co Ltd (Osaka, Japan), amlodipine was provided by Pfizer Inc (New York), carvedilol was provided by Daiichi-Sankyo Co Ltd (Tokyo, Japan), and hydralazine was provided by Sanwa Kagaku Kenkyusho Co Ltd (Nagoya, Japan).

Gene Expression Analysis

The RNA was extracted, the quality of RNA was assessed, and microarray analysis was performed as previously described,7,8 using the 35K CodeLink Bioarray (GE Healthcare United Kingdom) for the kidney and the Whole Rat Genome Microarray 4×44K (Agilent Technologies) for the heart (Figure 1).

Figure 1.

Figure 1. Transcriptomic analysis design.A, Transcriptomic response to 6 antihypertensive drugs was analyzed by microarray in the kidney (top) and the heart (bottom). Four drugs—amlodipine (AML), hydralazine (HYD), enalapril (ENA), and candesartan (CAN)—induced prominent BP reduction in the rat strains tested (stroke-prone spontaneously hypertensive rat [SHRSP] and Wistar Kyoto rats [WKY]), whereas the remaining 2 drugs—hydrochlorothiazide (HCTZ) and carvedilol (CAR)—did not. Specific solvent (Sol.) was prepared to dissolve CAN (Sol. A; 0.15% PEG300, 0.15% ethanol, 0.055% NaHCO3), HCTZ (Sol. B; 0.1 N NaOH), and CAR (Sol. C; 0.5% methylcellulose); each solvent was used as vehicle of the corresponding drug. B, Differential gene expression changes were investigated for 2 themes by intergroup comparison: (1) (4 effective BP-lowering drugs [AML, HYD, ENA, CAN]) group vs (control [or vehicle]) group in the heart of SHRSP (left side) and (2) (renin-angiotensin system [RAS] inhibitors [ENA+CAN]) group vs (control [or vehicle]) group in 2 organs of SHRSP and WKY (right side).

Quantitative PCR amplification was performed to validate mRNA expression of selected genes, as previously described.7 For quantitative PCR, PCR primer sequences for the target genes were designed originally or taken from literature (Table S1).

Microarray, Differential Expression, and Functional Network Analysis

In microarray analysis, differential expression was tested for each transcript by multiple regression with log2-transformed intensity and the type of experimental condition (such as strain category and administration of antihypertensive drug) used as the dependent and independent variables, respectively.

Differential expression was first examined between a drug treatment group and the corresponding nontreated (or control) group in each organ separately by strain or with both strains combined (Figure 1A). Transcriptomic response was then analyzed regarding 2 points: (1) effective BP-lowering in SHRSP and (2) RAS blockade versus other drug classes in both SHRSP and WKY (Figure 1B).

To discover biological pathways that were differentially expressed, the Reactome database (ver. 66, https://reactome.org) was sought with the WebGestalt website.9 Also, protein-protein interaction networks functional enrichment analysis was performed on the STRING database (ver. 11.5, https://string-db.org) by using the Cytoscape software platform (ver. 3.8.2, https://cytoscape.org). See Supplemental Material for more information.

Statistics

The results are expressed as means±SEM, and differences were analyzed using ANOVA or unpaired Student t test when comparing 2 groups means. P<0.05 was considered statistically significant after adjustment of multiple testing. Alternatively, when we examined data for transcriptomic response to a given drug, we regarded genes showing P<5×10–6 and P<5×10–5 as significant and suggestive significant, respectively, besides false discovery rate <0.05 for differential mRNA expression in this study.

Data Availability

Microarray data have been deposited at the EMBL-EBI ArrayExpress Archive of Functional Genomics Data under accession number (E-MTAB-9244, E-MTAB-10667, and E-MTAB-10672) and also at the NCBI Gene Expression Omnibus under accession number, GSE188350.

Results

Phenotypic Changes After Antihypertensive Drug Administration

Four antihypertensive drugs, including 2 RAS inhibitors—enalapril and candesartan—markedly lowered BP in SHRSP (Figure 2A), which are deemed effective drugs; 3 of them (other than amlodipine) also showed significant BP reduction in normotensive WKY. Hydrochlorothiazide and carvedilol did not decrease BP to an appreciable extent in both strains.

Figure 2.

Figure 2. Phenotype changes induced by antihypertensive drug treatment.A, Blood pressure changes from baseline (12 wk of age) to 14/28 days after treatment (ΔBP) are demonstrated for each drug and control (or vehicle). B, Body weight changes from baseline (12 wk of age) to 14/28 days after treatment (ΔBw) are demonstrated for each drug and control (or vehicle). Intergroup comparison (drug vs control group, n=5 or 6 per group) was performed at 2 timings, separately by strain. *P<0.05, **P<0.001 by unpaired t test. AML indicates amlodipine; CAN, candesartan; CAR, carvedilol; ENA, enalapril; HCTZ, hydrochlorothiazide; HYD, hydralazine; SHRSP, stroke-prone spontaneously hypertensive rat; and WKY, Wistar Kyoto rats.

Moreover, 4 of the tested drugs had significant impact on Bw gain. Compared with those in a control group, the rats undergoing enalapril treatment showed significantly smaller Bw gain from baseline to 14- and 28-days of treatment (eg, 0.3 versus 19.7 g for 14 days, 2.5 versus 26.5 g for 28 days; drug versus control group in SHRSP, P<0.01). Although less prominent in the degree of suppression, amlodipine, carvedilol, and hydrochlorothiazide showed significantly smaller Bw gain in either or both of the strains similar to enalapril (Figure 2B, Figure S1A, and Table S2). In connection with such impacts on Bw, 6 antihypertensive drugs all showed a significant decrease in adjusted adipose tissue weight, compared with the respective control group (Figure S1B). However, there was no apparent relation of suppressed Bw gain with either adjusted adipose tissue weight (Figure S1C) or plasma lipid levels (Figure S1D) in SHRSP.

In the heart of SHRSP, a significant decrease in adjusted heart weight was detectable for candesartan, enalapril, and amlodipine but not for hydralazine, among 4 effective drugs (Figure S1E). In the kidney of SHRSP, 4-week administration of antihypertensive drugs had no appreciable effects on proteinuria, except for hydrochlorothiazide, which showed borderline significant reduction of 24-hour proteinuria (−3.9±1.0 mg/d, adjusted P=0.054 by unpaired t test, versus a control group; Figure S1F).

Transcriptomic Response to Antihypertensive Medication Across Rat Strains and Organs

Volcano plots for microarray analysis are shown for 6 antihypertensive drugs by strain (SHRSP and WKY) and by organ (heart and kidney; Figure S2). For each drug, statistical significance was pronounced in SHRSP than WKY, while the shape of volcano plots appeared to be similar between the strains in general, likely to reflect drug-specific transcriptomic responses. Differential gene expression was prominent for enalapril in the heart and for amlodipine and hydralazine in the kidney among 6 tested drugs (Table S3). Besides organ-specific regulation, due to differences in the number of arrays per group (n=4 in the heart versus n=2 in the kidney) and the type of array platform (Agilent versus CodeLink), the distribution of statistical significance differed between the organs. Differentially expressed genes substantially overlapped between the drugs; the number of overlapping was largest between enalapril and amlodipine in the heart and between amlodipine and hydralazine in the kidney (Figure S3).

Transcriptomic Response to Effective BP-Lowering in SHRSP

For a subset of genes showing significant BP-dependent differential mRNA expression in the heart of SHRSP (n=496, P≤5×10−6 in meta-analysis of 4 effective drugs), a high level of correlation (r≥0.87) was detectable in the interdrug comparison between 4 effective drugs. On the contrary, for another subset of genes showing prominent BP-independent differential mRNA expression in the heart of SHRSP (n=45, P>0.001 in meta-analysis of 4 effective drugs and P≤5×10−6 in any of 6 drugs), interdrug correlation levels were not high, apart from a few pairs of antihypertensive drugs (Figure 3).

Figure 3.

Figure 3. Blood pressure (BP)-dependent and BP-independent gene expression changes in the heart of stroke-prone spontaneously hypertensive rat (SHRSP). Plots of interdrug correlation of gene expression changes (fold change in log2-scale) are depicted for genes (n=496) showing significant BP-dependent gene expression changes (upper right) and for genes (n=45) showing prominent BP-independent gene expression changes (lower left) in the heart of SHRSP. See main texts about the criteria for gene selection. AML indicates amlodipine; CAN, candesartan; CAR, carvedilol; ENA, enalapril; HCTZ, hydrochlorothiazide; and HYD, hydralazine.

A number of genes showed prominent differential mRNA expression by administration of effective drugs concordantly in 2 organs (Table 1) or in either of them (Tables S4 and S5). The concordant genes were found to be involved in the biological pathways/classifications, likely reflecting primary effects of BP-lowering on cardiac remodeling.

Table 1. Prominent Gene Expression Changes Induced by Effective Blood Pressure-Lowering Drugs Concordantly in the Heart and Kidney of SHRSP

Gene symbolChromosomeFour effective drugs combined*
HeartKidney
Position, rn6Array probeFC[−1SE–+1SE]P valueFDRPosition, rn6Array probeFC[−1SE – +1SE]P valueFDR
Postn2143 680 538A_44_P5252350.33[0.31–0.36]2.3×10−142.1×10−10143 687 430GE127170.42[0.36–0.48]3.6×10−50.026
Fcrl22186 596 106A_64_P1054770.60[0.56–0.64]3.3×10−91.1×10−6186 594 997GE11475320.60[0.56–0.65]2.3×10−50.024
Slc66a3 (Pqlc3)642 290 842A_42_P5226380.65[0.62–0.68]8.3×10−116.5×10−842 291 066GE166500.60[0.54–0.67]3.4×10−40.060
Loxl1863 073 023A_43_P199070.47[0.44–0.50]1.7×10−136.9×10−1063 072 174GE155320.71[0.66–0.76]2.4×10−40.054
Fn1978 900 248A_44_P6816600.54[0.50–0.57]5.5×10−115.0×10−878 900 390GE11560040.68[0.65–0.72]7.0×10−60.013
Serpine2985 561 300A_44_P4388630.57[0.54–0.60]6.5×10−131.5×10−985 561 287GE11722370.81[0.78–0.84]5.3×10−50.030
Col1a11082 761 378A_44_P3505210.55[0.51–0.59]4.5×10−102.2×10−782 762 677GE212220.50[0.43–0.59]6.4×10−40.073
Col8a11145 005 593A_44_P1406840.52[0.48–0.55]3.4×10−113.7×10−845 005 983GE11626440.60[0.54–0.67]2.7×10−40.055
Crlf11620 675 045A_64_P1247010.54[0.51–0.57]1.9×10−136.9×10−1020 675 099GE11286620.47[0.41–0.54]1.0×10−40.037
Mmp21915 542 956A_44_P9967290.78[0.75–0.81]5.7×10−74.9×10−515 543 320GE209940.76[0.72–0.79]3.4×10−50.026
Cdk12020 591 234A_42_P5808440.50[0.47–0.54]5.7×10−115.0×10−820 591 305GE221490.54[0.50–0.59]7.0×10−60.013

Four antihypertensive drugs (amlodipine, hydralazine, enalapril, and candesartan) effectively lowered BP in SHRSP and are designated as effective drugs, whereas the remaining 2 drugs (hydrochlorothiazide and carvedilol) are not effective in BP lowering. In the table, shown are a list of prominent gene expression changes induced by a category of effective drugs in the heart (≤5×10−6) and the kidney (P<1×10−3) of SHRSP. Gene expression was analyzed with Agilent and CodeLink arrays for the heart and the kidney, respectively. BP indicates blood pressure; FC, fold change; FDR, false discovery rate; and SHRSP, stroke-prone spontaneously hypertensive rat.

* Gene expression changes were evaluated with ANOVA by combining the impacts of 4 effective drugs in SHRSP (see Methods).

Transcriptomic Response to RAS Blockade

Differential mRNA expression could be detected between a group of rats undergoing RAS blocking treatment and their control group (Table 2, Tables S6 through S8). In the heart, RAS blockade induced prominent mRNA expression changes, assumably in a BP-independent manner, for a number of genes. The Gene Ontology terms noted for such genes are actin filament (Acta1, Actc1, and Micall2), diuretic hormone activity (Nppb), fatty acid biosynthetic/metabolic process (Abhd1, Acot2, and Hacd1), and extracellular matrix structural constituent (Cilp and Col4a1). In the kidney, prominent BP-independent mRNA expression changes were observed for a number of genes associated with several Gene Ontology terms, which include circulatory system development (Npy1r and Tnmd), calcium ion binding (Calb1 and Slc8a1), and endopeptidase activity (Ren and Klk1c9). In our data set, it is only the Cdk1 gene that showed prominent mRNA expression changes concordantly in 2 organs of 2 rat strains (Tables S6 and S7), although this appeared to be due principally to BP-lowering (Table 1 and Figure S4A). For part of the genes showing prominent RAS blockade–induced changes in microarray analysis, we validated their mRNA expression by quantitative PCR (Figure S4A through S4D and Supplemental Material).

Table 2. Prominent Gene Expression Changes Characteristically Induced by 2 Renin-Angiotensin System Blocking Drugs in the Heart and Kidney of Rats Combining SHRSP and WKY

Gene symbolChromosomePosition, rn6Array probeEnalapril and candesartan combined*EnalaprilCandesartanAmlodipineHydralazine
FC(−1SE–+1SE)P valueFDRFCP valueFCP valueFCP valueFCP value
Heart
Pdcd6131 591 270A_44_P10416160.85(0.83–0.86)4.3×10−134.4×10−100.84[4.2×10−9]0.86[1.9×10−7]0.96[0.089]0.94[0.027]
Actc13105 507 472A_64_P0788620.88(0.86–0.89)7.4×10−112.7×10−80.89[3.8×10−6]0.87[2.1×10−7]0.97[0.179]0.94[0.021]
Nppb5164 796 760A_42_P6384940.52(0.49–0.56)7.4×10−136.4×10−100.58[1.2×10−6]0.48[8.1×10−10]0.99[0.952]1.10[0.360]
Abhd1626 789 240A_43_P172131.36(1.31–1.41)2.0×10−119.0×10−91.39[6.6×10−8]1.33[2.1×10−6]1.13[0.029]1.15[0.015]
Acot26107 467 376A_42_P5976381.42(1.35–1.49)1.5×10−104.5×10−81.44[6.5×10−7]1.40[3.2×10−6]1.19[0.011]1.16[0.029]
Nudt4736 643 947A_43_P128651.47(1.40–1.54)1.6×10−121.1×10−91.49[3.3×10−8]1.45[1.6×10−7]1.13[0.061]1.14[0.045]
Cilp870 774 752A_44_P4368080.57(0.53–0.61)3.8×10−122.1×10−90.62[4.1×10−6]0.52[2.0×10−9]0.84[0.069]0.84[0.064]
Tax1bp31059 747 980A_44_P3545730.76(0.73–0.78)3.2×10−121.9×10−90.76[1.2×10−7]0.76[1.2×10−7]0.88[0.010]0.89[0.017]
Unc1191065 612 258A_64_P1340790.79(0.77–0.81)4.7×10−111.9×10−80.81[4.4×10−6]0.77[9.1×10−8]0.98[0.691]1.03[0.462]
Hap11088 258 250A_44_P1117231.37(1.32–1.42)6.9×10−136.2×10−101.44[5.2×10−10]1.31[1.5×10−6]1.01[0.876]1.06[0.233]
Micall21217 014 318A_44_P10184470.80(0.77–0.82)1.9×10−105.4×10−80.80[4.4×10−6]0.79[6.4×10−7]1.00[0.913]0.93[0.116]
Hopx1433 362 253A_44_P4045910.68(0.65–0.71)1.1×10−131.7×10−100.68[1.0×10−8]0.68[1.5×10−8]0.87[0.028]0.89[0.046]
Col4a11683 632 094A_64_P1495010.76(0.73–0.79)3.3×10−108.5×10−80.76[3.5×10−6]0.75[1.7×10−6]1.07[0.219]0.93[0.166]
Hacd11781 173 733A_64_P0845250.72(0.69–0.74)2.8×10−133.1×10−100.71[2.1×10−8]0.72[2.7×10−8]0.93[0.209]0.97[0.583]
Dpep21937 964 260A_42_P6337970.70(0.67–0.73)2.9×10−121.9×10−90.69[3.5×10−8]0.71[3.3×10−7]0.91[0.105]0.87[0.023]
Acta11956 675 553A_44_P2552360.54(0.52–0.57)2.1×10−214.8×10−170.54[2.7×10−14]0.54[2.6×10−14]0.98[0.753]0.86[0.023]
Kidney
Klk1c91100 089 588GE12648961.57(1.45–1.71)5.5×10−81.9×10−41.56[1.6×10−4]1.59[9.0×10−5]1.08[0.449]1.13[0.243]
Mmp93161 421 267GE208910.58(0.53–0.64)1.6×10−91.1×10−50.72[3.0×10−3]0.47[2.8×10−8]1.53[2.3×10−4]1.03[0.781]
Calb1529 562 373GE217920.63(0.59–0.67)3.8×10−134.3×10−90.68[4.3×10−6]0.58[1.4×10−8]1.44[1.2×10−5]1.06[0.393]
Slc8a164 256 192GE11097640.78(0.74–0.81)2.4×10−89.0×10−50.79[2.2×10−4]0.76[2.8×10−5]1.04[0.465]0.86[0.013]
Ren1350 502 963GE214298.00(6.55–9.77)2.3×10−257.8×10−216.34[2.8×10−12]10.09[1.1×10−14]2.57[2.4×10−6]1.36[0.066]
Npy1r1624 779 846GE145040.70(0.66–0.74)5.3×10−104.5×10−60.70[1.4×10−5]0.69[8.8×10−6]0.84[0.017]0.90[0.140]
TnmdX104 700 086GE12023540.56(0.51–0.63)2.2×10−76.7×10−40.58[4.8×10−4]0.55[1.2×10−4]0.85[0.234]0.90[0.466]

In the table, shown are a list of prominent, BP-independent (P>0.01 for both amlodipine and hydralazine) gene expression changes induced by 2 RAS inhibitors (enalapril and candesartan) in the rat heart and the kidney (P<5×10−10 and <5×10−6, respectively) for 2 drugs combined. Gene expression was analyzed with Agilent and CodeLink arrays for the heart and the kidney, respectively. BP indicates blood pressure; FC, fold change; FDR, false discovery rate; RAS, renin-angiotensin system; and SHRSP, stroke-prone spontaneously hypertensive rat.

* Gene expression changes are evaluated with Z score meta-analysis by combining the impacts of enalapril and candesartan in the heart and the kidney of rats combining SHRSP and WKY.

Functional Pathways/Networks Affected by Antihypertensive Drugs

To illuminate biological pathways or cellular processes, in which a set of genes acting in concert could be systematically affected by a particular drug treatment, we obtained gene-set enrichment analysis results with WebGestalt.9 We found that gene-sets in some pathways (eg, mitochondrial fatty acid beta-oxidation, extracellular matrix organization and immune system) were influenced commonly in 2 organs across effective drugs, whereas those in other pathways were influenced more prominently by 2 RAS inhibitors than amlodipine and hydralazine in either of the organs (eg, signaling by GPCR and hemostasis in the heart, and regulated necrosis in the kidney; Tables S9 and S10).

Besides gene-set enrichment analysis results, to gain more insight, we predicted protein-protein interaction networks constituting functional associations. In the microarray data sets, we could identify several clusters that form well-defined protein-protein interaction networks (involving ≥10 mutually interacting genes) in the heart; 3 clusters (clusters 1–3) were common between effective BP-lowering and RAS blockade, whereas one cluster (cluster 4, characteristic of glutathione metabolism, actin filament, etc) was specific to RAS blockade (Figure 4, Figures S5 and S6, Tables S11 and S12A through S12D, and Supplemental Material).

Figure 4.

Figure 4. Protein-protein interaction (PPI) networks of differentially expressed genes by renin-angiotensin system (RAS) blockade in the heart.A, Densely connected PPI networks are visualized with gene expression profiles (top), using the Cytoscape platform, for differentially expressed genes (P<5×10−6) that are induced by RAS blockade in the heart of stroke-prone spontaneously hypertensive rat (SHRSP) and Wistar Kyoto rats (WKY). Four clusters (from C1–C4, bottom) could be detected by the MCODE algorithm, involving ≥10 proteins within the subgraphs. Known biological functions predicted for the individual clusters are mitosis, etc for C1 (bottom, leftmost); extracellular matrix (ECM) organization, etc for C2 (bottom, middle left); fatty acid beta-oxidation, etc for C3 (bottom, middle right); and glutathione metabolism, etc for C4 (bottom, rightmost). Location of each cluster is indicated by dashed square in the top. In each of the clusters, top 3 hub-genes are circled in red. B, Top 10 enriched categories from gene-set enrichment analysis are depicted. See details about genes and GSEA categories in Tables S11B and S12B, respectively.

Moreover, as an approach to evaluating BP-dependent alterations, we compared gene-set enrichment analysis categories affected by BP-lowering drugs (false discovery rate <0.25) between hypertensive (SHRSP) and nonhypertensive (WKY) rat strains (Figure S7 and Table S12E). In the heart, the number of affected gene-sets was larger for SHRSP (n=292) than WKY (n=61), with 40 gene-sets overlapping. In the kidney, only a single gene-set was detectable as prominent alterations for WKY.

Discussion

We have examined phenotypic and transcriptomic responses to 6 antihypertensive drugs, which are commonly used in clinical settings to prevent hypertension-related cardiovascular complications. In the tested rat strains, 4 of 6 drugs, including RAS inhibitors, are effective for BP lowering, whereas the remaining 2 drugs are not. Besides BP lowering, there appears to be some interdrug heterogeneity in phenotypic changes, such as adjusted heart weight reduction and suppressed Bw gain (Figure 2 and Figure S1). For the transcriptomic response, a considerable number of genes show prominent mRNA expression changes either in a BP-dependent or BP-independent manner (Figure 3) with substantial diversity present between the heart and the kidney (Figure S2). Noticeable changes of mRNA expression can be induced particularly by RAS blockade, for example, for genes in the natriuretic peptide system in the heart and for those in the RAS/KKS and those related to calcium ion binding in the kidney.

Thus far, few studies have simultaneously examined 6 principal classes of antihypertensive drug in in vivo animal models as reported here. This helps us objectively recognize that hemodynamic impacts of antihypertensive drugs differ substantially and that some of them can induce significant phenotypic changes independent of BP-lowering. For BP changes, treatment with a β-blocker, carvedilol, induces less prominent BP-lowering than 4 effective drugs, even though it sufficiently suppresses the heart rate in SHRSP and WKY.7 These findings are mostly in accordance with previous studies by other investigators using SHR strains.10 Likewise, when given alone, hydrochlorothiazide induces relatively modest BP reduction in SHR.11

For cardiovascular complications, it has been reported that treatment with candesartan but not hydralazine induces regression of hypertensive LVH, even though both drugs significantly decrease BP in the rats.12 Our results not only support this but also demonstrate that treatment with enalapril and amlodipine similarly leads to significant suppression of adjusted heart weight (Figure S1E) in SHRSP. In addition, of note is the finding that 4 of the 6 drugs induce significant suppression of Bw gain, while all drugs induce reduction of adipose tissue weight in the rats (Figure S1B). A number of studies have reported the corresponding effects of enalapril treatment on body composition, that is, decrease in Bw and loss of fat mass, without alterations in serum lipid profile,13 setting up a hypothesis that increased hydration leads to loss of Bw and fat mass.14 However, since there is no apparent relationship between Bw and adjusted adipose tissue weight (Figure S1C), our data refute the hypothesis. Further investigation is required to clarify the underlying mechanisms.

Antihypertensive drug-induced BP lowering is pronounced in SHRSP than WKY, while a significant inter-strain BP difference is noticeable longitudinally on normal rat chow, that is, >80 mm Hg difference between 2 strains after 12 weeks of age.7 Accordingly, BP reduction can result from both strain diversity and antihypertensive treatment, either of which leads to prominent gene expression changes. On the contrary, mRNA expression of some genes is likely to be influenced by specific antihypertensive drugs via pharmacological activity. In this respect, when we look at gene expression changes in the heart of SHRSP, the number of genes showing significant BP-dependent expression changes is much larger than that of genes showing prominent BP-independent expression changes (Figure 3).

BP-dependent gene expression changes appear to be generally common between the heart and the kidney (Table 1). Such genes are involved in a wide range of mechanosensitive pathways, for example, regulation of tissue fibrosis and extracellular matrix (Figure 4 and Figures S5 and S6), as vigorously investigated in studies of pressure-overloaded hearts and heart failure.15 Fibrosis is also a major outcome of CKD, and inflammation can precede or progress along with fibrosis in many cases of hypertension-induced end-organ damages. Among the key mediators of tissue fibrosis, periostin (Postn), which is secreted as a nonstructural component of extracellular matrix and mediates signals in cell-matrix crosstalk, has drawn attention as a biomarker and a target of therapy in the heart and the kidney.16,17 Our results support the biological relevance of Postn to hypertension treatment.

Prominent mRNA expression changes by BP lowering are detectable for cyclin-dependent kinase 1 (Cdk1; Table 1). Cdk1 has been identified primarily as a key regulator of cell cycle in mitosis and meiosis, and is essential for cell proliferation and survival with relation to cancer cells and fibroblasts.18Cdk1 has recently been shown to activate cardiac fibroblasts into myofibroblasts, thereby assisting atrial fibrosis.19 Moreover, it is known that a profibrotic cytokine, TGF-β1, activates p53 pathways, which can suppress the expression of CDK1, in the progression and development of fibrotic disorders.20 Thus, whether it is the cause or the consequence of hypertension-related fibrosis, we find Cdk1 as a target molecule that warrants further investigation.

BP-dependent mRNA expression changes of apelin (Apln) are worthy of note. Apln is an endogenous peptide, belongs to the adipokine family with extensive tissue distribution and has many actions in the cardiovascular system.21 The mechanisms underlying complex effects of Apln remain incompletely understood, which is likely to cause its tissue-specific pattern of mRNA expression changes (Figure S4B).

RAS blockade is estimated to provide a 20% relative risk reduction for cardio-renal disease compared with other non-RAS-blocking therapies.22 According to meta-analysis of clinical trials, there has been no evidence for a difference in total mortality or cardiovascular outcomes between ACE inhibitors and ARBs.23 In accordance with these previous reports, we find a number of genes to show prominent mRNA expression changes by 2 RAS inhibitors equivalently; the list of genes is mostly nonoverlapping between the heart and the kidney (Table 2).

In the heart, of particular note for RAS blockade are Nppb and Acta1; both genes are known to be hypertrophic markers. In SHRSP and WKY, Nppb but not Nppa is downregulated specifically by RAS inhibitors, which is consistent with the previous study showing that mRNA expression of BNP is distinctly regulated from that of ANP.24 Also, skeletal α-actin (encoded by the Acta1 gene) is a marker of pathological and pressure overload-induced hypertrophy.25 mRNA expression of Acta1, as well as Actc1 (Table S13), is markedly decreased in both rat strains by RAS inhibitors (Figure S4A) to a similar extent.

In the kidney, RAS blockade and thiazide diuretics significantly increase mRNA expression of Ren and rat Klk1 paralogs (Klk1c4 and Klk1c9), the latter of which we found to be plausible target genes for a BP quantitative trait locus in the rats.7 Several previously unnoticed genes also show prominent mRNA expression changes by RAS inhibitors in the kidney (Table 2 and Table S6B). For instance, carbindin (Calb1) is an intracellular protein presumably involved in the reabsorption of calcium and magnesium in the kidney; hypertension is reported to affect Calb1 expression, although the exact regulatory mechanisms remain unknown.26 Tenomodulin (Tnmd), appreciably expressed in the adipose tissue, is indicated to act as a protective factor attenuating insulin resistance in obesity.27Tnmd is significantly downregulated by RAS inhibitors both in the kidney and adipose tissue (Figure S4D), supporting the pharmacological role of RAS inhibitors in modulation of insulin resistance.28,29

Furthermore, our protein-protein interaction networks functional enrichment analysis validates a number of biological pathways to be influenced by antihypertensive treatment in the rat either in a BP-dependent or BP-independent manner, in particular, discovering a single cluster, which is specific to RAS blockade (Figure 4).

There are several limitations in the present study. First, the homogenized whole organ was used for mRNA expression analysis, leading to the inability to evaluate heterogeneity in cellular component. Second, the current analysis with microarray platforms may not be sufficient in genome-wide coverage and statistical power to identify differential gene expression. Third, antihypertensive effects on gene expression changes were investigated in the stage before full development of hypertension and manifestation of apparent, pathological end-organ damages in genetically hypertensive rats, which further necessitates the more pathophysiological relevant study in a more advanced stage. Fourth, the current findings need to be pursued regarding the point that some of the impaired pathways, for example, defective cell cycle regulation and apoptosis,30 may be present from the neonatal period, impacting on susceptibility to end-organ damages in SHR strains.

Perspective

Our data reveal that 4 classes of antihypertensive drug, including 2 RAS inhibitors, effectively lower BP in the rat strains tested, with the presence of interdrug phenotypic heterogeneity. For transcriptomic response in the heart and the kidney, many genes show prominent mRNA expression changes in a BP-dependent manner, while the expression changes in some genes are relatively specific to RAS inhibitors. Such RAS inhibitor-specific genes consist of known ones (eg, Nppb and Acta1 in the heart) and previously unnoticed ones (eg, Calb1 and Tnmd in the kidney). The research resources constructed in this study will be the foundation to pursue the hemodynamic and pharmacological modification by antihypertensive medication and its consequences on cardiovascular complications, and will help corroborate evidence, which is occasionally inconclusive, in clinical settings.

Article Information

Acknowledgments

We are grateful to the research staff at Research Institute, National Center for Global Health and Medicine for their technical assistance with DNA analysis.

Nonstandard Abbreviations and Acronyms

ACE

angiotensin-converting enzyme

ARB

angiotensin receptor blocker

BP

blood pressure

Bw

body weight

CKD

chronic kidney disease

LVH

left ventricular hypertrophy

RAS

renin-angiotensin system

SHRSP

stroke-prone spontaneously hypertensive rat

WKY

Wistar Kyoto rat

Footnotes

Supplemental Material is available at https://www.ahajournals.org/doi/suppl/10.1161/HYPERTENSIONAHA.121.18026.

For Sources of Funding and Disclosures, see page 423.

Correspondence to: Norihiro Kato, Research Institute, National Center for Global Health and Medicine, Tokyo, 162-8655, Japan. Email

References

  • 1. Michel MC, Brunner HR, Foster C, Huo Y. Angiotensin II type 1 receptor antagonists in animal models of vascular, cardiac, metabolic and renal disease.Pharmacol Ther. 2016; 164:1–81. doi: 10.1016/j.pharmthera.2016.03.019CrossrefMedlineGoogle Scholar
  • 2. Chen YJ, Li LJ, Tang WL, Song JY, Qiu R, Li Q, Xue H, Wright JM. First-line drugs inhibiting the renin angiotensin system versus other first-line antihypertensive drug classes for hypertension.Cochrane Database Syst Rev. 2018; 11:CD008170. doi: 10.1002/14651858.CD008170.pub3MedlineGoogle Scholar
  • 3. Soliman EZ, Prineas RJ. Antihypertensive therapies and left ventricular hypertrophy.Curr Hypertens Rep. 2017; 19:79. doi: 10.1007/s11906-017-0777-3CrossrefMedlineGoogle Scholar
  • 4. Fagard RH, Celis H, Thijs L, Wouters S. Regression of left ventricular mass by antihypertensive treatment: a meta-analysis of randomized comparative studies.Hypertension. 2009; 54:1084–1091. doi: 10.1161/HYPERTENSIONAHA.109.136655LinkGoogle Scholar
  • 5. Jafar TH, Stark PC, Schmid CH, Landa M, Maschio G, de Jong PE, de Zeeuw D, Shahinfar S, Toto R, Levey AS; AIPRD Study Group. Progression of chronic kidney disease: the role of blood pressure control, proteinuria, and angiotensin-converting enzyme inhibition: a patient-level meta-analysis.Ann Intern Med. 2003; 139:244–252. doi: 10.7326/0003-4819-139-4-200308190-00006CrossrefMedlineGoogle Scholar
  • 6. Xie X, Liu Y, Perkovic V, Li X, Ninomiya T, Hou W, Zhao N, Liu L, Lv J, Zhang H, Wang H. Renin-Angiotensin system inhibitors and kidney and cardiovascular outcomes in patients with CKD: a Bayesian network meta-analysis of randomized clinical trials.Am J Kidney Dis. 2016; 67:728–741. doi: 10.1053/j.ajkd.2015.10.011CrossrefMedlineGoogle Scholar
  • 7. Takeuchi F, Liang YQ, Isono M, Tajima M, Cui ZH, Iizuka Y, Gotoda T, Nabika T, Kato N. Integrative genomic analysis of blood pressure and related phenotypes in rats.Dis Model Mech. 2021; 14:dmm048090. doi: 10.1242/dmm.048090CrossrefMedlineGoogle Scholar
  • 8. Kato N, Liang YQ, Ochiai Y, Birukawa N, Serizawa M, Jesmin S. Candesartan-induced gene expression in five organs of stroke-prone spontaneously hypertensive rats.Hypertens Res. 2008; 31:1963–1975. doi: 10.1291/hypres.31.1963CrossrefMedlineGoogle Scholar
  • 9. Liao Y, Wang J, Jaehnig EJ, Shi Z, Zhang B. WebGestalt 2019: gene set analysis toolkit with revamped UIs and APIs.Nucleic Acids Res. 2019; 47:W199–W205. doi: 10.1093/nar/gkz401CrossrefMedlineGoogle Scholar
  • 10. Del Mauro JS, Prince PD, Allo MA, Santander Plantamura Y, Morettón MA, González GE, Bertera FM, Carranza A, Gorzalczany SB, Chiappetta DA, et al. Effects of third-generation β-blockers, atenolol or amlodipine on blood pressure variability and target organ damage in spontaneously hypertensive rats.J Hypertens. 2020; 38:536–545. doi: 10.1097/HJH.0000000000002284CrossrefMedlineGoogle Scholar
  • 11. Wienen W, Schierok HJ. Effects of telmisartan, hydrochlorothiazide and their combination on blood pressure and renal excretory parameters in spontaneously hypertensive rats.J Renin Angiotensin Aldosterone Syst. 2001; 2:123–128. doi: 10.3317/jraas.2001.013CrossrefMedlineGoogle Scholar
  • 12. Kojima M, Shiojima I, Yamazaki T, Komuro I, Zou Z, Wang Y, Mizuno T, Ueki K, Tobe K, Kadowaki T. Angiotensin II receptor antagonist TCV-116 induces regression of hypertensive left ventricular hypertrophy in vivo and inhibits the intracellular signaling pathway of stretch-mediated cardiomyocyte hypertrophy in vitro.Circulation. 1994; 89:2204–2211. doi: 10.1161/01.cir.89.5.2204LinkGoogle Scholar
  • 13. Santos EL, de Picoli Souza K, Guimarães PB, Reis FC, Silva SM, Costa-Neto CM, Luz J, Pesquero JB. Effect of angiotensin converting enzyme inhibitor enalapril on body weight and composition in young rats.Int Immunopharmacol. 2008; 8:247–253. doi: 10.1016/j.intimp.2007.07.021CrossrefMedlineGoogle Scholar
  • 14. Thornton SN. Increased hydration can be associated with weight loss.Front Nutr. 2016; 3:18. doi: 10.3389/fnut.2016.00018CrossrefMedlineGoogle Scholar
  • 15. Saucerman JJ, Tan PM, Buchholz KS, McCulloch AD, Omens JH. Mechanical regulation of gene expression in cardiac myocytes and fibroblasts.Nat Rev Cardiol. 2019; 16:361–378. doi: 10.1038/s41569-019-0155-8CrossrefMedlineGoogle Scholar
  • 16. Landry NM, Cohen S, Dixon IMC. Periostin in cardiovascular disease and development: a tale of two distinct roles.Basic Res Cardiol. 2018; 113:1. doi: 10.1007/s00395-017-0659-5CrossrefMedlineGoogle Scholar
  • 17. Prakoura N, Hadchouel J, Chatziantoniou C. Novel targets for therapy of renal fibrosis.J Histochem Cytochem. 2019; 67:701–715. doi: 10.1369/0022155419849386CrossrefMedlineGoogle Scholar
  • 18. Kalous J, Jansová D, Šušor A. Role of cyclin-dependent kinase 1 in translational regulation in the M-Phase.Cells. 2020; 9:E1568. doi: 10.3390/cells9071568CrossrefMedlineGoogle Scholar
  • 19. Sai C, Yunhan J, Zhao J, Yu Z, Yun Z, Zhezhe C, Fuqin T, Yingbin X, Ruiyan M. Cyclin Dependent Kinase 1 (CDK1) activates cardiac fibroblasts via directly phosphorylating paxillin at Ser244.Int Heart J. 2019; 60:374–383. doi: 10.1536/ihj.18-073CrossrefMedlineGoogle Scholar
  • 20. Higgins SP, Tang Y, Higgins CE, Mian B, Zhang W, Czekay RP, Samarakoon R, Conti DJ, Higgins PJ. TGF-β1/p53 signaling in renal fibrogenesis.Cell Signal. 2018; 43:1–10. doi: 10.1016/j.cellsig.2017.11.005CrossrefMedlineGoogle Scholar
  • 21. Parikh VN, Liu J, Shang C, Woods C, Chang AC, Zhao M, Charo DN, Grunwald Z, Huang Y, Seo K, et al. Apelin and APJ orchestrate complex tissue-specific control of cardiomyocyte hypertrophy and contractility in the hypertrophy-heart failure transition.Am J Physiol Heart Circ Physiol. 2018; 315:H348–H356. doi: 10.1152/ajpheart.00693.2017CrossrefMedlineGoogle Scholar
  • 22. Romero CA, Orias M, Weir MR. Novel RAAS agonists and antagonists: clinical applications and controversies.Nat Rev Endocrinol. 2015; 11:242–252. doi: 10.1038/nrendo.2015.6CrossrefMedlineGoogle Scholar
  • 23. Li EC, Heran BS, Wright JM. Angiotensin converting enzyme (ACE) inhibitors versus angiotensin receptor blockers for primary hypertension.Cochrane Database Syst Rev. 2014; 2014:CD009096. doi: 10.1002/14651858.CD009096.pub2MedlineGoogle Scholar
  • 24. Nakagawa O, Ogawa Y, Itoh H, Suga S, Komatsu Y, Kishimoto I, Nishino K, Yoshimasa T, Nakao K. Rapid transcriptional activation and early mRNA turnover of brain natriuretic peptide in cardiocyte hypertrophy. Evidence for brain natriuretic peptide as an “emergency” cardiac hormone against ventricular overload.J Clin Invest. 1995; 96:1280–1287. doi: 10.1172/JCI118162CrossrefMedlineGoogle Scholar
  • 25. Djalinac N, Ljubojevic-Holzer S, Matzer I, Kolesnik E, Jandl K, Lohberger B, Rainer P, Heinemann A, Sedej S, von Lewinski D, et al. The role of stretch, tachycardia and sodium-calcium exchanger in induction of early cardiac remodelling.J Cell Mol Med. 2020; 24:8732–8743. doi: 10.1111/jcmm.15504CrossrefMedlineGoogle Scholar
  • 26. Hemmingsen C. Regulation of renal calbindin-D28K.Pharmacol Toxicol. 2000; 87(suppl 3):5–30.MedlineGoogle Scholar
  • 27. Senol-Cosar O, Flach RJ, DiStefano M, Chawla A, Nicoloro S, Straubhaar J, Hardy OT, Noh HL, Kim JK, Wabitsch M, et al. Tenomodulin promotes human adipocyte differentiation and beneficial visceral adipose tissue expansion.Nat Commun. 2016; 7:10686. doi: 10.1038/ncomms10686CrossrefMedlineGoogle Scholar
  • 28. Shiuchi T, Cui TX, Wu L, Nakagami H, Takeda-Matsubara Y, Iwai M, Horiuchi M. ACE inhibitor improves insulin resistance in diabetic mouse via bradykinin and NO.Hypertension. 2002; 40:329–334. doi: 10.1161/01.hyp.0000028979.98877.0cLinkGoogle Scholar
  • 29. Huang Y, Li Y, Liu Q, Zhang J, Zhang Z, Wu T, Tang Q, Huang C, Li R, Zhou J, et al. Telmisartan attenuates obesity-induced insulin resistance via suppression of AMPK mediated ER stress.Biochem Biophys Res Commun. 2020; 523:787–794. doi: 10.1016/j.bbrc.2019.12.111CrossrefMedlineGoogle Scholar
  • 30. Deblois D, Tea BS, Beaudry D, Hamet P. Regulation of therapeutic apoptosis: a potential target in controlling hypertensive organ damage.Can J Physiol Pharmacol. 2005; 83:29–41. doi: 10.1139/y05-001CrossrefMedlineGoogle Scholar

Novelty and Significance

What Is New?

  • Transcriptomic response to 6 principal classes of antihypertensive drug was simultaneously examined in in vivo animal models.

  • A considerable number of genes showed prominent mRNA expression changes induced by antihypertensive drugs either in a blood pressure-dependent or blood pressure-independent manner with substantial diversity between the heart and the kidney.

What Is Relevant?

  • The research resources constructed here will help corroborate occasionally inconclusive evidence in clinical settings.

  • Noticeable changes of gene expression were induced particularly by renin-angiotensin system blockade, in accordance with its assumable organ-protective effects.

Summary

This study reports experimental evidence for the presence of interdrug phenotypic heterogeneity and a global picture about genes showing prominent mRNA expression changes after treatment with different classes of antihypertensive drug.

eLetters(0)

eLetters should relate to an article recently published in the journal and are not a forum for providing unpublished data. Comments are reviewed for appropriate use of tone and language. Comments are not peer-reviewed. Acceptable comments are posted to the journal website only. Comments are not published in an issue and are not indexed in PubMed. Comments should be no longer than 500 words and will only be posted online. References are limited to 10. Authors of the article cited in the comment will be invited to reply, as appropriate.

Comments and feedback on AHA/ASA Scientific Statements and Guidelines should be directed to the AHA/ASA Manuscript Oversight Committee via its Correspondence page.