Dapagliflozin Attenuates Sympathetic and Pressor Responses to Stress in Young Prehypertensive Spontaneously Hypertensive Rats

Abstract

Background:

Sodium-glucose cotransporter 2 inhibitor (SGLT2i), a class of anti-diabetic medications, is shown to reduce blood pressure (BP) in hypertensive patients with type 2 diabetes mellitus. Mechanisms underlying this action are unknown but SGLT2i-induced sympathoinhibition is thought to play a role. Whether SGLT2i reduces BP and sympathetic nerve activity (SNA) in a non-diabetic prehypertension model is unknown.

Methods:

Accordingly, we assessed changes in conscious BP using radiotelemetry and alterations in mean arterial pressure (MAP) and renal SNA (RSNA) during simulated exercise in non-diabetic spontaneously hypertensive rats (SHRs) during chronic administration of a diet containing dapagliflozin (DAPA, 0.5mg/kg/day) vs. a control diet.

Results:

We found that DAPA had no effect on fasting blood glucose, insulin, or hemoglobin A_1C levels. However, DAPA reduced BP in young (8-week old) SHRs as well as attenuated the age-related rise in BP in adult SHR up to 17-weeks of age. The rises in MAP and RSNA during simulated exercise (exercise pressor reflex activation by hindlimb muscle contraction) were significantly reduced after 4 weeks of DAPA (ΔMAP: 10±7 vs. 25±14 mmHg, ΔRSNA: 31±17 vs. 68±39 %, P<0.05). Similarly, rises in MAP and RSNA during mechanoreflex stimulation by passive hindlimb stretching were also attenuated by DAPA. Heart weight was significantly decreased in DAPA compared to the control group.

Conclusions:

These data demonstrate a novel role for SGLT2i in reducing resting BP as well as the activity of skeletal muscle reflexes, independent of glycemic control. Our study may have important clinical implications for preventing hypertension and hypertensive heart disease in young prehypertensive individuals.

Graphical Abstract

INTRODUCTION

Hypertension affects nearly half of the US population, including 103.3 million adults.¹ Although hypertension treatment reduces cardiovascular disease risk, treated hypertensive individuals are still more likely to develop target organ damage and suffer a cardiovascular disease event than those who have not been hypertensive ^{2, 3}. Thus, primordial prevention of hypertension is of paramount importance to optimally reduce the burden of cardiovascular disease attributable to hypertension. Hypertension is characterized not only by elevated resting blood pressure (BP) but also by an exaggerated rise in BP during exercise ^4–7. It is well known that the normal physiologic decline in systemic vascular resistance during dynamic exercise is greatly attenuated or absent in human hypertension ^{8, 9}. Previous studies from our laboratories and others have shown that the pressor response to exercise in prehypertension and established forms of hypertension is exaggerated, being largely mediated by a reflex originating from the contracting skeletal muscle known as the exercise pressor reflex (EPR) that causes peripheral vasoconstriction ^10–12. Nevertheless, mechanisms underlying this augmented exercise pressor response remain unknown and effective intervention has not been developed.

Sodium-glucose transport protein 2 inhibitor (SGLT2i) is a new class of anti-diabetic agents that lower blood glucose by promoting glycosuria. SGLT2i has been shown to induce a significant reduction in BP and cardiovascular disease risks in patients with type 2 diabetes mellitus. Mechanisms underlying this antihypertensive action are unknown but appear to be out of proportion to its diuretic action ^{13, 14}. SGLT2 inhibition has been shown to reduce arterial BP and selectively blunt the low frequency power of systolic BP, an indirect marker of sympathetic nerve activity (SNA), in a rat model of metabolic syndrome ¹⁵. This effect is more pronounced during nighttime ¹⁵, when the rodents are more active (i.e. increased physical activity). Such SGLT2i-mediated reduction in SNA is further attributed to improved baroreflex sensitivity in a diabetic rat model ¹⁶. Whether SGLT2i reduces BP and SNA at rest and during exercise as well as alters baroreflex control of SNA in a nondiabetic prehypertension model remains unknown.

Accordingly, we conducted studies to determine the cardiovascular and sympathetic responses to dapagliflozin, a SGLT2i, in young prehypertensive spontaneously hypertensive rats (SHRs) at rest and in response to muscle contraction (i.e. simulated physical activity via EPR activation) as well as during baroreceptor stimulation.

METHODS

The authors declare that all supporting data are available within the article and its online supplementary files.

Animal Models

All experiments were performed in 4- to 6-week-old male SHRs (Charles River Laboratories). The animals were housed in standard rodent cages on 12 h light dark cycles and were randomly divided into either a group fed a control powered chow diet (T2016, 16.4% protein 4% fat and 48.5% carbohydrate, Envigo Teklad Diets, Madison, WI) or the same powdered food containing dapagliflozin (DAPA; 0.5 mg/kg/day, Advanced ChemBlocks Inc., Hayward, CA) for 4 weeks. All studies were conducted in accordance with the US Department of Health and Human Services NIH Guide for the Care and Use of Laboratory Animals. The outlined procedures were approved by the Institutional Animal Care and Use Committee of the University of Texas Southwestern Medical Center.

Blood Collection, ELISA, and Metabolic Cage Study

Fasting blood glucose and 3-hydroxybutyrate (3-OHB) levels were assessed by glucometer and ketometer, respectively. Additionally, fasting blood hemoglobin A_1c (HbA_1c) levels were measured by a rapid test kit. Fasting insulin concentration was determined in blood plasma obtained from the tail using ELISA (Crystal Chem, Chicago, IL). Food intake, water intake, urine volume, and urinary excretion of creatinine and sodium were measured during metabolic cage studies ¹⁷.

Experimental Protocols

Chronic experiment: conscious direct measurement of BP.

SHRs were implanted with BP transmitters (PA-C10, Data Sciences International, St. Paul, MN) as described previously ¹⁸. Conscious 24-hour BP, HR, and locomotor activity were recorded using the DSI system.

Acute experiment 1: skeletal muscle reflex testing.

Mean arterial pressure (MAP), HR, and renal SNA (RSNA) were continuously measured at rest and during distinctly separate stimulation of the EPR, mechanoreflex (a mechanically-sensitive component of the EPR), and metaboreflex (a metabolically-sensitive component of the EPR) following pre-collicular decerebration as described previously ^19–21. The pre- and post-decerebration hemodynamics were assessed via an arterial line placed in the carotid artery.

Acute experiment 2: baroreflex testing.

Resting baroreflex control of HR and RSNA, changes in HR and RSNA were obtained in response to an intravenous bolus infusion of phenylephrine (100 μg/ml) and nitroprusside (100 μg/ml) via the femoral vein while arterial pressure was recorded from the femoral artery ^{22, 23}.

Statistical Analyses

Data were analyzed using two-way ANOVA, post hoc Fisher’s least significant difference and Sidak test, Student’s unpaired t-test and Mann-Whitney U nonparametric test. Data normality was assessed with the Shapiro-Wilk test. The significance level was set at P < 0.05. Results are presented as mean ± SD. All analyses were performed using GraphPad Prism 9 (San Diego, CA).

RESULTS

Morphometric characteristics, fasting blood parameters, and 24-hour metabolic parameters and locomotor activity for control and DAPA treated SHRs are presented in Table 1. The DAPA group showed significant reductions in body weight as well as epididymal fat pad weight (1.07±0.35 vs. 1.98±0.61 g, P<0.01) and epididymal fat pad weight-to-body weight ratios (5.25±1.21 vs. 7.94±1.99 mg g⁻¹, P<0.01) as compared with the control group. Heart weight and heart weight-to-tibial length ratios were also significantly decreased in young DAPA-treated rats compared to control rats. There were no significant differences in heart weight-to-body weight, lung weight, lung weight-to-body weight, and lung weight-to-tibial length ratios. Kidney weights were not different between young DAPA and control SHRs while the DAPA groups showed greater kidney weight-to-tibial length ratios (table S1). Triceps surae (gastrocnemius and soleus) weight was slightly reduced by DAPA in prehypertensive SHRs (1.05±0.19 vs. 1.28±0.13 g, P<0.01) while triceps surae weight-to-body weight ratios were not different between the 2 groups (5.0±0.37 vs. 5.2±0.53 mg g⁻¹, P>0.1). Heart weight (0.92±0.08 vs. 1.08±0.08 g, P<0.05, table S1) and heart weight-to-tibial length ratios (25.8±1.7 vs. 29.9±1.2 mg ml⁻¹, P<0.05, table S1) were also significantly decreased in adult SHRs fed DAPA vs. control chow for 4 weeks. However, we did not find a significant difference in soleus, gastrocnemius, or plantaris muscle weight and soleus, gastrocnemius, or plantaris muscle weight-to-body weight ratios between adult DAPA SHRs and control SHRs (table S1). Fasting blood glucose and HbA_1c levels were not significantly different between DAPA and control groups. However, fasting levels of 3-OHB were significantly higher in DAPA than the control group while fasting plasma insulin was not different between the 2 groups. Metabolic cage studies showed that food intake was similar between the 2 groups. However, water intake was significantly higher in DAPA SHRs than control SHRs. Similarly, urine volume as well as 24-hour urinary glucose and sodium excretion were significantly higher, and plasma sodium concentration was significantly lower in DAPA rats than control rats. There were no significant differences in plasma creatinine levels, estimated glomerular filtration rate, or 24-hour urinary potassium and creatinine excretion between DAPA and control groups. Locomotor activity for 24-hours was significantly lower in DAPA treated rats compared to control rats.

Table 1.

Morphometric characteristics, blood parameters, and 24-hour metabolic parameters and locomotor activity.

Variable	Control	DAPA
Morphometric characteristics
Body weight (g)	227 ± 32	186 ± 30**
Heart weight (g)	0.80 ± 0.06	0.65 ± 0.07**
Heart weight/body weight (mg g−1)	3.27 ± 0.17	3.28 ± 0.27
Heart weight/tibial length (mg mm−1)	25.1 ± 1.4	21.4 ± 1.7**
Lung weight (g)	1.54 ± 0.28	1.37 ± 0.13
Lung weight/body weight (mg g−1)	6.26 ± 1.05	6.87 ± 0.71
Lung weight/tibial length (mg mm−1)	48.3 ± 8.4	44.9 ± 6.2
Triceps surae weight (g)	1.28 ± 0.13	1.05 ± 0.19**
Triceps surae weight/body weight (mg g−1)	5.2 ± 0.53	5.0 ± 0.37
Epididymal fat pad weight (g)	1.98 ± 0.61	1.07 ± 0.35**
Epididymal fat pad weight /body weight (mg g−1)	7.94 ± 1.99	5.25 ± 1.21**
Fasting blood parameters
Glucose (mg dl−1)	78 ± 20	65 ± 27
HbA1c (%)	4.9± 0.4	4.8± 1.1
Plasma Insulin (ng ml−1)	0.91± 0.45	1.18± 0.38
3-OHB (mmol l−1)	1.9± 0.7	2.7 ± 0.9*
Non-fasting plasma concentrations
Creatinine (mg dl−1)	0.24 ± 0.02	0.22 ± 0.02
Sodium (mmol l−1)	143 ± 1.7	139 ± 1.6*
24-hour metabolic parameters and locomotor activity
Food intake (g day−1)	18 ± 4	22 ± 4
Water intake (ml day−1)	24 ± 2	47 ± 8**
Urine volume (ml day−1)	5 ± 2	30 ± 3**
Urinary creatinine (mg day−1)	3.9 ± 1.5	4.2 ± 1.2
Urinary sodium (mmol day−1)	0.88 ± 0.35	1.65 ± 0.04*
Urinary potassium (mmol day−1)	1.6 ± 0.5	2.1 ± 0.2
Urinary glucose (mg day−1)	5.7 ± 4.6	4,033 ± 11.7**
Estimated glomerular filtration rate (ml min−1)	1.17 ± 0.52	1.32 ± 0.28
Locomotor activity (counts min−1)	5 ± 1	3 ± 1**

Values are means ± SD. DAPA, dapagliflozin-treated rats; HbA_1c, hemoglobin A_1c; 3-OHB, 3-hydroxybutyrate. Data were analyzed using Student’s unpaired t-test or Mann-Whitney U nonparametric test.

^*P < 0.05 compared with control rats

^**P < 0.01 compared with control rats.

Body weight, control: n=18, DAPA: n=17; heart and lung weight and tibial length, control: n=11, DAPA: n=10; triceps surae weight, control: n=11, DAPA: n=8; epididymal fat pad weight, control: n=11, DAPA: n=9; glucose, control: n=12, DAPA: n=9; HbA_1c and plasma insulin, control: n=8, DAPA: n=7; 3-OHB, control: n=14, DAPA: n=11; plasma creatinine and sodium, food and water intake, urine volume, urinary creatinine, sodium, potassium, and glucose as well as estimated glomerular filtration rate, control: n=3, DAPA: n=3; locomotor activity, control: n=7, DAPA: n=7.

Effects of SGLT2i on BP and Progression of Hypertension in SHRs

Dapagliflozin administration beginning at 4 weeks of age markedly attenuated daytime/nighttime MAP in prehypertensive SHRs by 8-weeks of age compared with SHRs fed a control diet likewise beginning at 4-weeks old (daytime: 119±6 vs. 130±6 mmHg, nighttime: 122±6 vs. 133±7 mmHg, P<0.01, Figure 1A). A significant reduction in HR in DAPA rats was also observed compared to control rats (daytime: 302±13 vs. 336±18 bpm, nighttime: 322±10 vs. 358±9 bpm, P<0.01, Figure 1A). To further determine if SGLT2i attenuates the progression of hypertension, additional telemetry BP and HR measurements were performed in a separate group of 12-week old SHRs randomized to DAPA chow vs. control and followed prospectively for a total of 5 weeks. At baseline (week 0), daytime/nighttime MAP and HR were similar between DAPA and control groups (Figure 1B). Following 5 weeks of the DAPA vs. control chow, however, nighttime MAP was significantly attenuated in the DAPA group compared to the control group (144±5 vs. 165±19 mmHg, P<0.05, Figure 1B). There was a tendency for daytime MAP to be lower in the DAPA than the control group, although the decrease did not reach statistical significance (P=0.08 for ANOVA time × drug interaction, Figure 1B). In addition, daytime/nighttime HRs were substantially lower in the DAPA group than those in the control group (daytime: 271±5 vs. 299±6 bpm, nighttime: 279±3 vs. 327±5 bpm, P<0.01, Fig. 1B) which was evident within 1 week of treatment. In addition, the bradycardic effect of DAPA during nighttime was persistent after 5 weeks of treatment compared to control SHRs (P<0.01, Figure 1B). In contrast, the heart rate lowering effects of DAPA during the daytime was attenuated after 5 weeks of administration to similar levels observed in the control SHRs (Figure 1B).

Figure 1. Conscious mean arterial pressure (MAP) and heart rate (HR) measured continuously by radiotelemetry over 12-hr light-dark cycles for 2 days in control and dapagliflozin-treated (DAPA) spontaneously hypertensive rats.

A, MAP and HR during a 24-hr period as well as daytime/nighttime averaged MAP and HR when treatment was initiated at 4 weeks of age (n=7/group). B, Time-course of daytime/nighttime BP and HR changes over 5 weeks when treatment was initiated at the age of 12 weeks old (n=3/group). Values are means ± SD. Data were analyzed by two-way ANOVA with repeated measures followed by post hoc comparisons using Fisher’s least significant difference or Sidak test. * P < 0.05 compared with control. ** P < 0.01 compared with control.

Effects of SGLT2i on Skeletal Muscle Reflex Function

Pre- and post-decerebration baseline hemodynamics and RSNA are shown in table S1. Baseline MAP was significantly lower in DAPA than in the control group under 1% isoflurane, similar to the findings with conscious BP measurements. After decerebration, the difference in baseline BP persisted between groups. However, baseline HR and RSNA to noise ratio pre- and post-decerebration were not significantly different between the 2 groups.

The peak BP and RSNA responses to activation of the EPR during static muscle contraction were significantly attenuated in DAPA SHRs compared to control SHRs (ΔMAP: 10±5 vs. 28±15 mmHg, ΔRSNA: 43±17 vs. 96±48 %, P<0.05, respectively, Figure 2A and 2B). The HR change in response to EPR stimulation, however, was not different between the 2 groups. Importantly, muscle tension developed during ventral root stimulation was similar between control and DAPA rats (0.2±0.1 vs. 0.2±0.1 kg, P>0.1, Figure 2A and 2B).

Figure 2. Cardiovascular and sympathetic responses to activation of the exercise pressor reflex (EPR) by electrically induced muscle contraction in control and dapagliflozin-treated (DAPA) spontaneously hypertensive rats (SHRs).

A, Representative tracings of arterial blood pressure (ABP) as well as raw and normalized renal sympathetic nerve activity (RSNA, %) during a 30 sec static muscle contraction in a control and DAPA animal. B, Summary data showing peak changes in mean arterial pressure (MAP) [control, n=10; DAPA, n=10], heart rate (HR) [control, n=10; DAPA, n=10] RSNA [control, n=9; DAPA, n=7], and developed tension [control, n=10; DAPA, n=10] in response to EPR activation. Values are means ± SD. Data were analyzed by Student’s unpaired t-test. * P < 0.05 compared with control. ** P < 0.01 compared with control.

Similar to the responses during EPR stimulation, the pressor and sympathetic responses evoked by mechanoreflex activation during passive muscle stretch were also markedly lower in DAPA than control rats (ΔMAP: 10±7 vs. 25±14 mmHg, ΔRSNA: 31±17 vs. 68±39 %, P<0.05, Figure 3A and 3 B). The change in HR in response to mechanoreflex stimulation, however, was not significantly different between the two groups. The developed muscle tension during passive stretch was similar between control and DAPA SHRs (0.4±0.2 vs. 0.3±0.1 kg, P>0.1, Figure 3A and 3B).

Figure 3. Cardiovascular and sympathetic responses to activation of the mechanoreflex via passive muscle stretch in control and dapagliflozin-treated (DAPA) spontaneously hypertensive rats (SHRs).

A, Representative tracings of arterial blood pressure (ABP) as well as raw and normalized renal sympathetic nerve activity (RSNA, %) during a 30 sec passive stretch in a control and DAPA animal. B, Summary data showing peak changes in mean arterial pressure (MAP) [control, n=9; DAPA, n=10], heart rate (HR) [control, n=9; DAPA, n=10], RSNA [control, n=8; DAPA, n=8], and developed tension [control, n=9; DAPA, n=10] in response to mechanoreflex activation. Values are means ± SD. Data were analyzed by Student’s unpaired t-test. * P < 0.05 compared with control. ** P < 0.01 compared with control.

There was a tendency for MAP and RSNA responses to metaboreflex activation induced by capsaicin administration to be lower in the DAPA group compared to the control group, but the difference did not reach statistical significance (ΔMAP: 38±18 vs. 58±24 mmHg, P=0.07, ΔRSNA: 58±29 vs. 95±45 %, P=0.08, Figure 4A and 4B). The magnitude of the capsaicin-induced increase in HR was not different between control and DAPA animals (Figure 4B).

Figure 4. Cardiovascular and sympathetic responses to activation of the metaboreflex via intra-iliac arterial capsaicin administration in control and dapagliflozin-treated (DAPA) spontaneously hypertensive rats (SHRs).

A, Representative tracings of arterial blood pressure (ABP) as well as raw and normalized renal sympathetic nerve activity (RSNA, %) during capsaicin injection in a control and DAPA animal. B, Summary data showing peak changes in mean arterial pressure (MAP) [control, n=9; DAPA, n=9], heart rate (HR) [control, n=9; DAPA, n=9], and RSNA [control, n=9; DAPA, n=7] in response to metaboreflex activation. Values are means ± SD. Data were analyzed by Student’s unpaired t-test.

Effects of SGLT2i on Baroreflex Function

DAPA administration had no significant effects on the maximum gain, slope coefficient, midpoint pressure, or minimum value of output for both baroreflex control of RSNA and HR (table S2 and Figure 5A–5C). The operating MAP for the baroreflex control of HR tended to shift downward in the DAPA group (Figure 5C) but the difference did not reach statistical significance (P=0.08, table S2).

Figure 5. Baroreflex control of renal sympathetic nerve activity (RSNA) and heart rate (HR) determined during infusion of phenylephrine and nitroprusside in control and dapagliflozin-treated (DAPA) spontaneously hypertensive rats.

Graphs showing baroreflex curves relating A, mean arterial pressure (MAP) to RSNA (baseline=100%) [control, n=3; DAPA, n=2], B, MAP to RSNA (maximum=100%) [control, n=3; DAPA, n=2], and C, MAP to HR [control, n=4; DAPA, n=3]. Values are means ± SD. Data were analyzed by two-way ANOVA with repeated measures.

DISCUSSION

The major findings from this investigation are fourfold. First, chronic administration of SGLT2i induces a sustained reduction in resting BP and HR in young prehypertensive SHRs and prevents the progression of hypertension leading to reduction in heart weight. Second, SGLT2i attenuates cardiovascular and sympathetic responses to muscle contraction that simulates exercise in this prehypertensive rat model. Third, this beneficial effect of SGLT2i is evident without any detectable changes in blood glucose or plasma insulin. Fourth, SGLT2i does not have significant effects on baroreflex function in SHRs. Collectively, our present study provides the first direct evidence for the sympathoinhibitory actions of SGLT2i that may have an implication for the prevention of hypertension.

In a previous report, SGLT2i was shown to reduce renal and cardiac norepinephrine (NE) content of mice fed a high-fat diet ²⁴. In contrast, studies in alloxan-induced diabetic rabbits showed that empagliflozin attenuated the maximum RSNA response to hypotension but had no impact on resting RSNA, MAP or HR ²⁵. Previous studies in humans have suggested an inhibitory influence of SGLT2is on muscle SNA measured via intraneural recordings in patients with type 2 diabetes mellitus with or without heart failure ^{26, 27}. However, previous studies have not determined the effects of SGLT2i on SNA in non-diabetic hypertension or prehypertension models. To address this gap in knowledge, SNA was directly measured during simulated exercise in both SGLT2i treated rats as well as their controls. Results determined that SGLT2i significantly reduced the MAP and SNA responses to electrically-induced muscle contraction. In normal physiological states, a reflex originating from the working skeletal muscle (i.e. the EPR) is important in regulating BP and cardiovascular function during physical activity. The EPR is mediated by both activation of mechanically-sensitive muscle nerve afferents known as the mechanoreflex ²⁸ and by stimulating the chemically-sensitive component of the EPR known as the metaboreflex ²⁹. When experimentally isolated in our study, dapagliflozin treatment significantly attenuated muscle mechanoreflex activity (i.e. during muscle stretch) and had a similar impact on muscle metaboreflex function (i.e. during intra-arterial capsaicin administration) though the latter was not statistically significant. The mechanisms underlying the attenuation of EPR function and its mechanoreflex and metaboreflex components remain unknown. The reduction in body weight after SGLT2i treatment is one potential mechanism. However, higher body weight alone is not likely to contribute to augmented EPR responses in control SHRs compared to SGLT2i treated animals since untreated SHRs have been shown to have greater BP and SNA responses to muscle contraction despite similar body weight to Wistar-Kyoto (WKY) normotensive controls ^{30, 31}. As another possibility, reduced visceral fat may contribute to lowering EPR-mediated responses after dapagliflozin administration. However, previous studies have shown similar levels of epididymal fat and intra-renal fat in SHRs when compared to WKY when both consume a normal fat diet. This suggests that the augmented BP and SNA responses to EPR activation in SHR compared to WKY are independent of visceral fat ³². Another potential mechanisms are improvements in the glycemic milieu after SGLT2i treatment. However, this possibly is unlikely to account for the differences noted between groups in the current study since fasting blood glucose, insulin, and HbA_1c were not altered by dapagliflozin treatment.

In contrast to glucose and insulin, we found a significant increase in circulating ketone levels after dapagliflozin treatment, which has also been demonstrated in previous studies ³³. 3-OHB is one of the ketone bodies shown to decrease firing of peripheral sympathetic neurons via inhibition of G-protein coupled receptor 41 ^{34, 35} via the Pertussis Toxin-sensitive Gβγ–Phospholipase C-ERK1/2 pathway ³⁶. In addition, chronic exposure to ketones has been shown to reduce the activity of protein kinase C ³⁷, a serine/threonine kinase involved in transmission of nociceptive signals from peripheral nerves to the spinal cord ³⁸. Additional studies are needed to determine if elevations in ketone levels mediate the beneficial effects of SGLT2i on sympathetic outflow and BP at rest and during exercise.

Our radiotelemetry data demonstrated the potential benefits of SGLT2i treatment on young SHRs prior to the development of sustained hypertension. Furthermore, we exhibited the beneficial effects of SGLT2i in attenuating the progression of hypertension, when initiating treatment in early to mid-adulthood from the age of 12 to 17 weeks old, the time period in which hypertension normally reaches stable levels in SHRs ^{18, 39}. Previous studies have also shown that a selective SGLT2 inhibitor, luseogliflozin, reduced BP during the daytime and nighttime in adult SHR/NDcp ¹⁵, a rat model of metabolic syndrome that displays higher levels of visceral fat than traditional SHRs ⁴⁰. Our study showed similar effects with the use of dapagliflozin, another selective SGLT2i ⁴¹, on prehypertensive SHRs that had a markedly lower level of baseline BP than adult SHR/NDcp. Unlike studies in SHR/NDcp, however, we found a persistent reduction in HR during the nighttime after dapagliflozin treatment. The discrepancy in study results may be related to the difference in experimental protocols such as the animal strain utilized for investigation.

The mechanisms underlying reductions in HR induced by dapagliflozin remain unknown. Previous studies have demonstrated enhanced cardiac sympathetic neurotransmission in prehypertensive SHRs as evidenced by greater tachycardia during right stellate stimulation while the HR response to vagus stimulation and exogenous NE administration were similar when compared to WKY controls ⁴². Treatment with dapagliflozin may have reduced such elevations in cardiac sympathetic neurotransmission. We did not measure cardiac SNA directly and we cannot exclude the possibility that DAPA may alter cardiac sympathetic neurotransmission in our study. However, it is our contention that cardiac neurotransmission alone cannot explain the reduction in BP because of the difference in the time course of the decrease in HR vs. BP during telemetry studies. We found that DAPA induced a significant reduction in daytime and nighttime HR within 1 week of drug administration while daytime and nighttime MAP remained the same until later in the time course. Impaired baroreflex control of HR has also been noted in SHRs beginning from 9 weeks of age ^{43, 44}. In contrast, baroreflex control of SNA remains preserved in SHRs ⁴⁴. However, we did not find evidence that the gain of baroreflex control of HR or RSNA was modified by dapagliflozin in our study. Only the MAP operating point for the baroreflex control of HR range tended to be shifted to a lower BP range in our investigation. Thus, it is unclear if the change in operating points explains the reduction in HR induced by dapagliflozin in our study. Recent studies have demonstrated a direct inhibitory effect of SGLT2i on the late sodium channel in the ventricular myocyte of a murine model of heart failure ⁴⁵, which may offer protective effects against ventricular arrhythmias and sudden cardiac death. Empagliflozin, dapagliflozin, and canagliflozin have been shown to directly inhibit cardiac Na⁺/H⁺ exchange flux ⁴⁶, which are recognized as a major mechanism underlying atrial fibrillation ⁴⁷. Whether cardiac neurotransmission or cardiac sodium channel function are altered by SGLT2i therapy remains to be determined.

Our study is limited by several factors including a lack of molecular mechanisms linking SGLT2i therapy to reductions in resting BP and HR as well as the attenuation of EPR dysfunction in prehypertensive SHRs. Since dapagliflozin also induced diuresis and natriuresis, we cannot determine the relative contribution of renal vs. neural mechanisms in SGLT2i-induced BP reduction. However, enhanced natriuresis and diuresis alone should have resulted in reflex activation from reduced intravascular volume resulting in increases in RSNA. Thus, our study suggested that the reduction in RSNA observed is a direct sympathoinhibitory action of dapagliflozin, which may in turn contribute to BP reduction and increased natriuresis. In addition, baroreflex function was examined under closed-loop conditions after infusion of phenylephrine and nitroprusside, which assesses both neural and peripheral arcs. In other studies, carotid sinus baroreflex function is assessed alone under open-loop conditions which may hinder direct comparisons between investigations ⁴⁸. Despite these limitations, we were able to demonstrate a sustained reduction in BP and HR under resting conditions after dapagliflozin via radiotelemetry, which allows assessment of hemodynamic changes in conscious states. The reduction in BP and SNA was also observed during stressful stimuli (i.e. simulated exercise), which was accompanied by a reduction in the heart weight. However, since the absolute hindlimb skeletal muscle weight was also slightly reduced in prehypertensive DAPA treated SHRs, we cannot exclude the possibility that the heart weight reduction is due to reduced overall body size. Nevertheless, our study findings may have important clinical implications introducing SGLT2i as a novel strategy in preventing hypertension and hypertensive heart disease in nondiabetic adult populations.

PERSPECTIVES

Hypertensive individuals are still more likely to suffer from cardiovascular events than the general population, despite considerable advances in treatments for the disease ^{2, 3}. As such, it is of high importance to develop new, effective, preventive treatment aimed at reducing the burden of cardiovascular disease in affected individuals. Currently, pharmacologic treatment is not available or approved for the prevention of hypertension but rather used once the disease manifests. Although previous studies have implicated the potential role of several antihypertensive drugs such as angiotensin receptor blockers, in preventing progression to hypertension, efficacy of treatment is modest and the generalizability of study findings to younger adults is limited as the trials have been conducted in middle-aged to older adults ^{49, 50}. Accordingly, only lifestyle intervention is recommended according to current guidelines, which has modest efficacy and often requires complex behavioral interventions ^{51, 52}. The present study is the first to demonstrate a novel role for SGLT2i in preventing hypertension and reducing adiposity in a prehypertensive nondiabetic model similar to the effects of caloric restriction ⁵³, which is difficult to sustain lifelong with nonpharmacologic intervention alone. An additional benefit of SGLT2i therapy in attenuating an augmented pressor response to exercise was also demonstrated in our study. Although previous studies have shown increased risk of genital infection associated with SGLT2i, which may limit its long-term use, the discontinuation rates due to adverse events were not different from placebo groups in clinical trials of patients with or without type 2 diabetes mellitus to date ^{54, 55}. Randomized clinical trial is needed to determine efficacy and safety of SGLT2i on preventing hypertension, particularly in younger adults.

Novelty and Relevance.

What Is New?

• Sodium-glucose cotransporter 2 inhibitor (SGLT2i) administration reduces sympathetic and blood pressure (BP) responses to muscle contraction in young prehypertensive spontaneously hypertensive rats (SHRs).

• SGLT2i reduces conscious resting BP and heart rate in prehypertensive SHRs and mitigates the progression of hypertension.

• These beneficial effects of SGLT2i are not explained by augmented baroreflex function or changes in blood glucose.

What is Relevant?

• SGLT2i attenuated resting BP as well as the rise in BP and sympathetic activity during simulated exercise in a prehypertensive experimental model. These findings provide the first direct evidence that SGLT2i exerts antihypertensive effects by inhibiting sympathetic nerve activity.

Clinical/Pathophysiological Implications?

• SGLT2i may prevent hypertension and hypertensive-related cardiovascular damage in young prehypertensive individuals.

Nonstandard Abbreviations and Acronyms:

DAPA

dapagliflozin

EPR

exercise pressor reflex

HR

heart rate

NE

norepinephrine

RSNA

renal sympathetic nerve activity

SGLT2i

sodium-glucose transport protein 2 inhibitor

SHRs

spontaneously hypertensive rats

SNA

sympathetic nerve activity

WKY

Wistar-Kyoto

3-OHB

3-hydroxybutyrate