Dipeptidyl peptidase-4 (DPP4) inhibition increases catecholamines without increasing blood pressure during sustained angiotensin-converting enzyme (ACE) inhibitor treatment

Abstract

Background:

Dipeptidyl peptidase-4 (DPP4) inhibitors comprise a class of oral diabetes medication that have the potential for off-target cardiovascular effects. We previously showed that DPP4 inhibition attenuates the hypotensive effect of acute angiotensin converting-enzyme (ACE) inhibition and increases norepinephrine. Here we investigated the effects of DPP4 during sustained ACE inhibition compared to during therapy with an angiotensin receptor blocker (ARB) or calcium channel blocker (neutral comparator) in a randomized, double-blinded crossover study.

Methods:

We enrolled 106 adults with type 2 diabetes (T2DM) and hypertension, and 100 received intervention. Subjects were randomized to one of three blood pressure arms: ramipril, valsartan, or amlodipine for a total of 15 weeks and received three one-week cross-over therapies in random order: placebo + placebo, sitagliptin + placebo, and sitagliptin + aprepitant separated by four-week washout.

Results:

We found that DPP4 inhibition increased norepinephrine during ramipril, but did not increase blood pressure. Aprepitant, a NK1 (substance P) receptor blocker, lowered standing heart rate during RAAS blockade with ramipril or valsartan.

Conclusions:

Increased catecholamines during concurrent ACE and DPP4 inhibition may contribute to cardiovascular complications in patients pre-disposed to heart failure.

Introduction

Dipeptidyl peptidase-4 (DPP4) inhibitors are widely used to treat type 2 diabetes mellitus (T2DM). DPP4 is a transmembrane protease that selectively cleaves the amino terminal dipeptide from peptides with a penultimate proline or alanine, such as the incretins glucagon-like peptide-1 (GLP-1) and glucose-dependent insulinotropic peptide (GIP).^1,2 In addition to preventing the degradation of incretins, DPP4 inhibitors prevent the degradation of several vasoactive peptides, including substance P and neuropeptide Y.^1,3 Substance P is released from sensory C fibers and activates sympathetic nerve terminals via the NK1 receptor.^4,5 This increased sympathetic activation could contribute to an increased risk of heart failure that has been observed in some clinical trials of DPP4 inhibitors.^6–9

Approximately two-thirds of patients with T2DM also have hypertension. Many of these patients take angiotensin-converting enzyme (ACE) inhibitors or angiotensin receptor blockers (ARBs) for renoprotection. Substance P is degraded by ACE as well as by DPP4, creating the potential for a drug-drug interaction between ACE inhibitors and DPP4 inhibitors. We have reported previously that DPP4 inhibition attenuates the hypotensive effect of acute ACE inhibition and increases norepinephrine (NE) concentrations.¹⁰ In the present study, we tested the hypothesis that DPP4 inhibition could affect the blood pressure or catecholamine response to sustained ACE inhibition in patients with T2DM and hypertension. We included two drug comparators. We studied the effect of DPP4 inhibition in patients treated with an ARB as a control to evaluate for an interaction of DPP4 inhibition with a class of drugs that interrupt the renin-angiotensin-aldosterone system (RAAS) without inhibiting ACE. We also studied the effect of DPP4 inhibition in patients treated with a calcium channel blocker to evaluate an interaction with an anti-hypertensive drug that does not interrupt the RAAS.

Methods

Data that support the findings of this study are included in this manuscript or available from the corresponding author upon reasonable request.

Study Subjects

Men and women, age 18 to 80 years, with T2DM and hypertension were studied. All participants provided written informed consent. The study was approved by the Vanderbilt Institutional Review Board and conducted according to the Declaration of Helsinki. T2DM was defined by an untreated hemoglobin A1c (HbA1c) ≥ 6.5% (48 mmol/mol), fasting plasma glucose ≥ 126 mg/dL (7.0 mmol/L), or 2-hour plasma glucose ≥ 200 mg/dL (11.1 mmol/L) following a 75 g oral glucose load, or treatment with anti-diabetic medication for at least six months. Hypertension was defined as a seated systolic blood pressure (SBP) ≥ 130 mm Hg or diastolic blood pressure (DBP) ≥ 80 mm Hg on at least three occasions or treatment with antihypertensive medications for at least six months. Pregnancy was excluded in women of childbearing potential by β-human chorionic gonadotropin testing. Participants with poorly controlled diabetes [HbA1c > 8.7% (72 mmol/mol)], long-term (>12 months) use of an anti-diabetic medication other than metformin, use of insulin, type 1 diabetes, or secondary hypertension were excluded.

Protocol

Participants taking anti-hypertensive and/or anti-diabetic medications other than thiazide diuretics or metformin underwent washout from these other medicines for a minimum of three weeks (four weeks for spironolactone). Following washout, participants were randomized in a 1:1:1 ratio to the ACE inhibitor ramipril, the ARB valsartan, or the calcium channel blocker amlodipine at the following doses for 15 weeks: ramipril 5 mg/day(d) for three days, then 10 mg/d; valsartan 160 mg/d for three days, then 320 mg/d; or amlodipine 5 mg/d for three days, then 10 mg/d (Supplemental Figure S1). After the first four weeks of anti-hypertensive treatment, participants were randomized to the first of three one-week crossover therapies: placebo/d + placebo/d, sitagliptin 100 mg/d + placebo/d, and sitagliptin 100 mg/d + the NK1 receptor antagonist aprepitant (125 mg, then 80 mg/d). Each crossover treatment was separated by four weeks to avoid carryover effects (Supplemental Figure S1). The study was approved by the Vanderbilt Institutional Review Board and registered with ClinicalTrials.gov (NCT02130687). All participants underwent informed consent.

During the last day of each one-week crossover treatment period, participants presented to the Vanderbilt Clinical Research Center (CRC) after an overnight fast for a study day. Participants were studied in the supine position. They were given their anti-hypertensive medication and their last dose of crossover medications between seven and eight in the morning. Blood pressure and heart rate were measured via an automated oscillometric device every five minutes (Dinamap; Critikon, Carlsbad, CA) for four hours. After four hours, participants were asked to stand, and blood pressure and heart rate were measured every five minutes during this orthostatic challenge. Blood was drawn through the indwelling catheter for measurement of DPP4 activity, ACE activity, glucose, insulin, and catecholamines at baseline. Blood for measurement of catecholamines was collected hourly after drug administration for four hours and after one-half hour of standing.

A subset of 51 participants (15 in the ramipril group, 16 in the valsartan group, and 20 in the amlodipine group) participated in a pre-specified sub-study in which they ingested a mixed meal on each study day and the results of the mixed meal study have been published previously.¹¹

Laboratory Analyses

DPP4 activity was measured by cleavage of a colorimetric substrate (2 mM l-glycyl-l-prolyl p-nitroanilide hydrochloride; Sigma-Aldrich, St. Louis, MO), as previously reported.¹¹ ACE activity was measured by LabCorp (Burlington, NC). Plasma glucose was measured via glucose oxidase method using a YSI glucose analyzer (YSI Life Sciences, Yellow Springs, OH) immediately following collection. All remaining samples were stored at −80°C in aliquots until the time of assay. Plasma was collected in tubes containing aprotinin, and insulin was measured by radioimmunoassay (EMD Millipore, Billerica, MA). The assay cross-reacts with 38% intact proinsulin, but not with C-peptide (≤ 0.01%). Plasma catecholamines was measured via high-performance liquid chromatography (HPLC) using electrochemical detection, as previously described.^12,13

Statistical Analyses

Continuous data are presented as means ± standard deviation. We used generalized least squares linear models to evaluate for the possibility of carryover or period effect as detailed in Chapter 5 (pages 212–213) of Jones and Kenward.¹⁴ We evaluated for and found no evidence for carryover or period effects on mean arterial pressure (MAP), heart rate, glucose, or insulin. We used a Kruskal-Wallis test followed by Dunn’s test to compare blood pressure, heart rate, and catecholamines among anti-hypertensive treatment groups during concurrent placebo + placebo crossover therapy. A Wilcoxon signed-rank test was used to compare measurements among cross-over treatments within each anti-hypertensive treatment group. The portion of participants taking concomitant metformin or thiazide diuretic among the anti-hypertensive treatment groups was compared using a Pearson test. In addition, to exclude any possible confounding effect of concurrent thiazide diuretic use, we also fitted multivariable generalized least squares linear regression models for the outcomes with thiazide use included as a covariate. A two-sided p value <0.05 was considered significant. All analyses were completed using R3.6.3. (R Core Team 2020, https://www.R-project.org, Vienna, Austria).

In our data, the standard deviation of within-subject difference in SBP between placebo + placebo and sitagliptin + placebo was 9.4 mmHg and in DBP was 5.3 mmHg. With 30 subjects within an anti-hypertensive treatment group, we have 80% power to detect a significant difference of 5 mmHg in SBP or 2.8 mmHg in DBP for sitagliptin compared to placebo with a two-sided type I error rate of 0.05 based on a two-sample paired t-test.

Results

Supplemental Figure S2 shows the consort diagram for the study. Baseline characteristics of those who were randomized to and received ramipril, valsartan, and amlodipine appear in Table 1. Eighty-three percent of participants randomized to ramipril, 83% of those receiving valsartan, and 79% of those receiving amlodipine were prescribed metformin (p=0.88). Fifty percent of those randomized ramipril, 36% of those receiving valsartan, and 28% of those receiving amlodipine were taking a thiazide diuretic concurrently (p=0.20).

Table 1:

Baseline characteristics of patients randomized to and receiving an angiotensin-converting enzyme (ACE) inhibitor, an angiotensin receptor blocker (ARB), or a calcium channel blocker

Characteristic	Ramipril N=30	Valsartan N=30	Amlodipine N=33
Age, years	52.2±8.9*	58.4±10.3	58.3±11.2
Gender, N
Male: Female	16: 14	16: 14	16: 17
Race, N
Asian: Black: White: Other	1: 4: 24: 1	0: 7: 22: 1	1:10: 22: 0
Ethnicity, N
Hispanic, non-Hispanic, unknown	0: 29: 1	0: 30: 0	2: 31: 0
BMI, kg/M2	35.9±6.8	33.1±6.3	34.1±6.3
SBP, mmHg	129.6±12.7	135.3±9.7	134.7±10.1
DBP, mmHg	75.7±7.4*	80.8±7.5	80.2±8.3
HR, bpm	72.6±9.8	70.3±10.8	73.8±11.2
FBS, mg/dL	135.7±26.4	122.1±18.2	123.4±26.3
HbA1C, %	6.91±1.06	6.61±0.81	6.50±0.69

Abbreviations: Body mass index (BMI), systolic blood pressure (SBP), diastolic blood pressure (DBP), heart rate (HR), fasting blood sugar (FBS), hemoglobin A1C (HbA1C).

^*p<0.05 versus the other two anti-hypertensive treatment groups

We first evaluated the effect of treatment on ACE and DPP4 activity. ACE activity was significantly lower in the ramipril group compared to the valsartan or amlodipine groups (Table 2). DPP4 activity was significantly and similarly decreased during co-treatment with sitagliptin + placebo or sitagliptin + aprepitant, in all three anti-hypertensive treatment groups.

Table 2:

Effect of one-week treatment with placebo, DPP4 inhibitor or DPP4 inhibitor and NK1 receptor antagonist within each anti-hypertensive treatment group.

	Ramipril			Valsartan			Amlodipine
	Placebo/placebo	Sitagliptin/placebo	Sitagliptin/aprepitant	Placebo/placebo	Sitagliptin/placebo	Sitagliptin/aprepitant	Placebo/placebo	Sitagliptin/placebo	Sitagliptin/aprepitant
ACE activity, U	15.4±9.2*	14.7±9.4*	13.5±5.4*	37.2±14.8	36.7±13.1	36.9±13.2	37.0±14.6	40.2±14.9	35.8±13.7
DPP4 activity, nmol/mL/min	20.6±6.2	8.8±6.5†	7.7±3.4†	19.4±7.1	7.8±4.1†	6.7±3.5†	20.3±6.8	7.3±3.0†	7.0±4.3†
FBS, mg/dL	118.0±21.8	107.6±18.4†	107.7±17.5†	112.0±21.6	103.7±18.1†	99.75±15.6†	123.8±34.3	112.5±30.0†	109.1±21.4†
Insulin, μU/mL	26.2±13.8	22.6±13.0	26.0±18.5	21.4±16.2	19.9±14.5	16.8±10.6	20.7±10.4	20.7±12.0	20.2±16.3
SBP, mmHg
Supine	124.8±18.7	123.2±13.6	120.5±15.5	131.3±13.2	129.8±15.1	130.6±14.8	130.2±11.8	128.7±13.8	126.2±13.1†
Standing	127.0±19.4	124.7±14.4	121.6±19.3	131.4±11.6	133.1±14.9	131.1±14.9	129.8±13.7	130.8±15.6	130.2±14.9
DBP, mmHg
Supine	72.9±10.0	73.2±8.7	70.2±8.3‡	76.2±9.7	75.7±9.8	75.7±10.0	76.6±7.1	75.8±7.6	74.2±6.7†
Standing	78.0±10.4	76.8±9.7	73.5±9.1†	80.1±9.2	80.1±10.1	79.0±9.4	79.6±8.9	78.4±8.0	77.9±8.3
MAP, mmHg
Supine	90.2±12.5	89.9±9.7	87.0±10.2	94.5±10.0	93.7±10.8	94.0±10.8	94.4±7.5	93.4±8.8	91.5±7.8†
Standing	94.3±12.5	92.8±10.2	89.5±11.8†	97.2±8.2	97.8±10.8	96.3±9.9	96.3±9.6	95.9±9.5	95.3±9.4
HR, bpm
Supine	66.6±7.7	66.3±8.2	66.2±8.1	66.2±9.2	65.9±8.5	65.1±8.5	69.6±9.8	70.4±9.8	69.4±9.8
Standing	81.3±12.6	84.3±13.5	81.0±9.9‡	81.0±13.3	81.0±12.2	78.7±11.8‡	85.7±14.8	85.2±11.5	85.6±12.7

Supine hemodynamic data reflect the average over four hours. Standing hemodynamic data reflect the average value over one-half hour of standing.

^*p<0.05 versus valsartan or amlodipine treatment groups,

^†p<0.05 versus placebo + placebo treatment,

^‡p<0.05 versus sitagliptin + placebo.

DBP indicated diastolic blood pressure, FBS fasting blood sugar, HR heart rate, MAP mean arterial pressure, SBP systolic blood pressure

Glucose and insulin concentrations were similar in the three anti-hypertensive groups (Table 2). Glucose was significantly and similarly diminished during treatment with sitagliptin + placebo and sitagliptin + aprepitant compared to during placebo + placebo in all three anti-hypertensive treatment arms. There was no effect of any treatment on insulin.

We next evaluated hemodynamic parameters. Supine SBP, DBP, MAP and HR were similar among the ramipril, valsartan, and amlodipine treatment groups during placebo + placebo (Table 2). There was no effect of concurrent sitagliptin treatment (sitagliptin + placebo) on supine SBP, DBP, or MAP in any of the anti-hypertensive treatment groups (Figure 1), even when thiazide diuretic use was included as a covariate. In the amlodipine treatment group, supine SBP, DBP and MAP were significantly lower during concurrent sitagliptin + aprepitant compared to placebo + placebo (Table 2 and Figure 1). In the ramipril treatment group, standing DBP and MAP were significantly lower during sitagliptin + aprepitant compared to placebo + placebo and supine DBP was significantly lower during concurrent sitagliptin + aprepitant compared to sitagliptin alone. Also, during ramipril and valsartan therapy, standing HR was significantly lower during concurrent sitagliptin + aprepitant compared to sitagliptin alone.

Figure 1:

Supine mean arterial pressure (MAP) and heart rate measured on the 7^th day of crossover treatment with placebo + placebo, sitagliptin 100mg/d + placebo, or sitagliptin 100 mg/d + aprepitant 80mg/d in individuals with hypertension and type 2 diabetes (T2DM) who were taking ramipril (10 mg/d), valsartan (320 mg/d), or amlodipine (10 mg/d). Crossover treatments were administered on the 5th, 10th, and 15th week of anti-hypertensive treatment in random order. *p<0.05 versus placebo + placebo, †p<0.05 versus sitagliptin + placebo

During placebo + placebo treatment, supine norepinephrine (NE) and dihydroxyphenylglycol (DHPG) concentrations were lower in the ramipril group compared to in the amlodipine group (Figure 2). Supine DHPG was also lower in the ramipril group compared to the valsartan group. In the ramipril treatment group, but not the valsartan or amlodipine group, supine NE concentrations were increased during treatment with sitagliptin + placebo compared to placebo + placebo (Figure 2), even after controlling for thiazide diuretic use. Co-treatment with aprepitant did not alter this effect except at the four-hour time point. In the ramipril treatment group, supine DHPG concentrations were increased during treatment with sitagliptin + placebo compared to placebo + placebo and co-administration of aprepitant prevented this effect. The supine DHPG:NE ratio was also decreased compared to placebo + placebo (5.54±4.56) during sitagliptin + placebo (3.86±2.26, p=0.002) but not during sitagliptin + aprepitant (5.09±5.66), even after controlling for thiazide diuretic use. There was no difference in NE or DHPG among crossover treatments in the valsartan or amlodipine treatment groups. There was no effect of any treatment on epinephrine, dopa, dopamine, or dopac concentrations (data not shown).

Figure 2:

Supine norepinephrine (NE) and dihyroxyphenylglycol (DHPG) concentrations measured on the 7th day of crossover treatment with placebo + placebo, sitagliptin 100mg/d + placebo, or sitagliptin 100 mg/d + aprepitant 80mg/d in individuals with hypertension and type 2 diabetes (T2DM) who were taking ramipril (10 mg/d), valsartan (320 mg/d), or amlodipine (10 mg/d). Crossover treatments were administered on the 5th, 10th, and 15th week of anti-hypertensive treatment in random order. *p<0.05 versus placebo + placebo, †p<0.05 versus sitagliptin + placebo

We also assessed the effect of treatment on standing BP, HR, and catecholamines. During placebo + placebo standing SBP, DBP, MAP, and HR were similar among anti-hypertensive treatment arms (Table 2). During standing, circulating NE was significantly lower in the ramipril treatment group compared to the valsartan and amlodipine treatment groups (Figure 3). The change in NE with standing was also smaller in the ramipril group compared to the amlodipine group (385.0±338.7 pg/mL versus 609.7±420.0 pg/mL, p=0.048). In addition, the DHPG:NE ratio was significantly greater during standing in the ramipril group compared to the amlodipine group (Figure 3). There was no effect of sitagliptin + placebo or sitagliptin + aprepitant on standing NE concentrations in any anti-hypertensive treatment group. In the ramipril treatment group, standing DHPG concentrations were decreased during sitagliptin + aprepitant compared to during sitagliptin + placebo.

Figure 3:

Standing norepinephrine (NE), dihydroxyphenolglycol (DHPG), and DHPG/NE on the 7th day of crossover treatment with placebo + placebo, sitagliptin 100mg/d + placebo, or sitagliptin 100 mg/d + aprepitant 80mg/d in individuals with hypertension and type 2 diabetes mellitus (T2DM) who were taking ramipril (10 mg/d), valsartan (320 mg/d), or amlodipine (10 mg/d). Crossover treatments were administered on the 5^th, 10^th, and 15^th week of anti-hypertensive treatment in random order. *p<0.05 versus both valsartan and amlodipine during placebo+placebo, †p<0.05 versus sitagliptin+placebo in ACE inhibitor-treated participants, ‡p<0.05 versus amlodipine during placebo+placebo

Discussion

We have reported previously that concurrent DPP4 inhibition attenuates hypotensive response to acute ACE inhibition.¹⁰ In the present study, we examined the effect of DPP4 inhibition on blood pressure in individuals with hypertension and diabetes during sustained treatment with an ACE inhibitor, an ARB, or a calcium channel blocker. The DPP4 inhibitor sitagliptin did not affect blood pressure during treatment with any of the classes of anti-hypertensive medication. In contrast, concurrent DPP4 inhibition significantly increased circulating NE concentrations in those individuals treated with the ACE inhibitor ramipril. Addition of the NK1 receptor antagonist aprepitant did not negate the effect of sitagliptin on NE concentrations during ramipril treatment, suggesting that during sustained ACE inhibition concurrent DPP4 inhibition increased fasting norepinephrine through an NK1 receptor-independent mechanism.

In prior studies, DPP4 inhibition increased NE concentrations and blood pressure during acute ACE inhibition and increased post-prandial norepinephrine concentrations in individuals with hypertension under a variety of conditions.^11,15 In the present study, sitagliptin did not increase blood pressure during chronic ramipril therapy even though norepinephrine and DHPG concentrations were increased. This finding suggests that with longer concurrent administration of ACE and DPP4 inhibitors, there may be adrenergic receptor down-regulation leading to attenuation of the hypertensive effect of increased catecholamines. In support of this, heart rate was not increased when sitagliptin was given together with chronic ramipril. Increased fasting norepinephrine during combined ACE inhibition and DPP4 inhibition may nevertheless have implications for patients with heart failure, where sympathetic activation is associated with poor outcomes.^16–18

Prior literature provides conflicting data on the anti-adrenergic effects of ACE inhibition and ARB therapy. Preclinical studies in spontaneously hypertensive rats, for example, have shown a decrease in NE concentrations during ACE inhibition compared to controls and concentrations were similar to rats without hypertension.¹⁹ RAAS inhibition with an ACE inhibitor or ARB have also been reported to decrease sympathetic tone and ARB therapy may reduce release of NE from sympathetic nerves.²⁰ In humans, however, ACE inhibition and ARB therapy did not reduce norepinephrine release and concentrations of epinephrine and norepinephrine were not significantly different compared to placebo.^21–23 We found that during placebo, supine NE and DHPG were lower during ACE inhibition compared to during calcium channel blockade therapy and DHPG was also lower during ACE inhibition compared to ARB therapy.

Co-treatment with aprepitant decreased DHPG during sitagliptin back to concentrations measured during placebo in ACE inhibitor-treated patients. DHPG is derived from intraneuronal metabolism of NE after reuptake via the norepinephrine transporter (NET).²⁴ DHPG concentrations reflect uptake of prevailing NE concentrations, and the ratio of DHPG to NE is considered to be an index of NET activity.²⁵ NET is localized in the pre-synaptic neuron in lipid rafts as is the NK1 receptor.^26,27 NK1 receptor activation mediates NET downregulation via protein kinase C in vitro. Interestingly, the effects of sitagliptin and aprepitant in ACE inhibitor-treated participants differed in the supine and standing positions. Sitagliptin decreased the ratio of DHPG:NE in supine ACE inhibitor-treated patients, and aprepitant reversed this effect. This suggests that while sitagliptin increased supine NE through an NK1-independent mechanism, sitagliptin prevented the reuptake of NE via an NK1-dependent mechanism. In contrast, while not significant sitagliptin tended to decrease the standing DHPG:NE ratio regardless of whether aprepitant was given concurrently. Studies in vitro suggest that high concentrations of NE prevent the downregulation of NET by NK1 receptor activation.²⁸ This may account for the lack of effect of aprepitant on the DHPG:NE ratio in ramipril-treated participants during standing or in valsartan- and amlodipine-treated participants.

The findings of the present study contrast our early observation regarding the effect of sitagliptin on post-prandial norepinephrine concentrations. Sitagliptin increased post-prandial concentrations of NE in both ACE inhibitor treatment and ARB treatment groups in a prespecified sub-study.¹¹ Moreover, this effect was mediated by substance P as aprepitant (NK1/ substance P receptor blocker) prevented the effect of sitagliptin to increase catecholamines in this setting.¹¹ It is possible that substance P release is increased after a meal compared to in the fasting state in individuals with diabetes, as previously reported in animals.²⁹ Postprandial substance P secretion has not been well studied in humans; however, recent data show that cultured sensory nerves secrete substance P in response to GLP-1 and GIP.³⁰ Since intact (active) GLP-1 and GIP secretion are increased after a meal during sitagliptin,³¹ increased substance P could contribute to post-prandial catecholamine release.

Neuropeptide Y (NPY) concentrations may also contribute to NE release, and this may be context dependent.³² DPP4 degrades NPY 1–36 to NPY 3–36, and NPY 3–36 activates prejunctional Y2 receptors to decrease release of norepinephrine from the sympathetic nerve terminal.^33,34 We have previously reported increased concentrations of NPY 1–36 and decreased levels of NPY 3–36 during sitagliptin.¹¹ In the setting of sitagliptin, decreased NPY 3–36 may contribute to increased norepinephrine release compared to during placebo.

This study has a few limitations. Although anti-hypertensive drugs were administered for a total 15 weeks, each crossover treatment was administered for one week. It is possible that with longer exposure to sitagliptin there may be further alterations such as the downregulation of receptors for substrates of DPP4. The study was not powered to allow for stratified analysis of effects of sitagliptin in specific racial or gender groups. We did not measure substance P concentrations as currently available immunoassays are not specific for substance P and detect its degradation products. During the study, participants could continue metformin or thiazide diuretics if necessary for safety, and as metformin is first line for therapy of type 2 diabetes. Metformin is not known to increase neurohormone concentrations such as NE and epinephrine.³⁵ There were no differences in the proportion of individuals taking either metformin or thiazide diuretics among anti-hypertensive treatment groups. In addition, the effects of sitagliptin, placebo, and aprepitant were measured within participants in a crossover design. Lastly, we excluded any confounding effect of concurrent thiazide diuretic use by conducting an analysis in which thiazide use was included as a covariate.

Perspectives

DPP4 inhibitors and ACE inhibitors are often used together in patients with diabetes and hypertension. DPP4 inhibition increases norepinephrine (NE) and attenuates the hypotensive effect of acute angiotensin converting-enzyme (ACE) inhibition. DPP4 inhibition significantly increased supine fasting norepinephrine concentrations in individuals with hypertension and T2DM taking an ACE inhibitor chronically, but not those taking an ARB or CCB. The increase in fasting norepinephrine concentrations were not mediated via a substance P receptor mechanism, whereas the substance P inhibitor decreased supine DHPG concentrations. Despite increased NE concentrations during concurrent sustained ACE inhibition and DPP4 inhibition, blood pressure and heart rate were not increased, consistent with adaptation to increased catecholamines. Nevertheless, increases in catecholamines may have detrimental cardiovascular effects particularly in patients pre-disposed to heart failure.

Pathophysiologic Novelty and Relevance.

“What is new”:

• Concurrent DPP4 inhibition increases fasting supine norepinephrine and DHPG concentrations in patients with T2DM and hypertension taking an ACE inhibitor chronically but does not increase blood pressure.

• Concurrent DPP4 inhibition does not increase norepinephrine or DHPG concentrations taking an ARB or a calcium channel blocker.

• Substance P (NK1) receptor antagonism reduces fasting concentrations of DHPG but not norepinephrine during combined DPP4 and ACE inhibition.

“What is relevant”:

• DPP4 inhibitors and ACE inhibitors are commonly used together in patients with diabetes and hypertension

• Understanding the impact of combining these classes of drugs versus combining DPP4 inhibitors with alternative antihypertensive medications that block the renin-angiotensin-aldosterone system through a non-enzymatic mechanism or that do not interact directly with the renin-angiotensin-aldosterone system will inform therapy.

“What are the Pathophysiological Implications”:

During sustained ACE inhibition but not during angiotensin receptor blockade or calcium channel blockade treatment with a DPP4 inhibitor increased supine circulating NE and DHPG concentrations. There was no effect on blood pressure or supine heart rate. Co-treatment with a substance P (NK1) receptor inhibitor decreased DHPG but not NE. Increased catecholamines may contribute to cardiovascular complications in patients pre-disposed to heart failure.