International Journal of
Molecular Sciences
Article
Enhanced Cardiorenal Protective Effects of Combining SGLT2
Inhibition, Endothelin Receptor Antagonism and RAS
Blockade in Type 2 Diabetic Mice
Ander Vergara 1,2,† , Conxita Jacobs-Cacha 1,† , Carmen Llorens-Cebria 1 , Alberto Ortiz 3 ,
Irene Martinez-Diaz 1 , Nerea Martos 1 , Pamela Dominguez-Báez 1 , Mireia Molina Van den Bosch 1 ,
Sheila Bermejo 1,2 , Michael Paul Pieper 4 , Begoña Benito 5,6,7, * and Maria Jose Soler 1,2, *
1 Nephrology and Kidney Transplantation Research Group, Vall d’Hebron Institut de Recerca (VHIR),
Passeig Vall d’Hebron 119-129, 08035 Barcelona, Spain
2 Nephrology Department, Vall d’Hebron Hospital Universitari, Passeig Vall d’Hebron 119-129,
08035 Barcelona, Spain
3 IIS-Fundación Jiménez Diaz, Fundación Renal Iñigo Álvarez de Toledo-IRSIN, REDinREN,
Instituto de Investigación Carlos III, Universidad Autónoma de Madrid, Av. de los Reyes Católicos 2,
28040 Madrid, Spain
4 Cardio-Metabolic Diseases Research, Boehringer Ingelheim Pharma GmbH & Co. KG, Birkendorfer Str. 65,
88397 Biberach an der Riß, Germany
5 Cardiology Research Group, Vall d’Hebron Institut de Recerca (VHIR), Vall d’Hebron Hospital Universitari,
Vall d’Hebron Barcelona Hospital Campus, Passeig Vall d’Hebron 119-129, 08035 Barcelona, Spain
6 Cardiology Department, Vall d’Hebron Hospital Universitari, Passeig Vall d’Hebron 119-129,
Citation: Vergara, A.; Jacobs-Cacha,
C.; Llorens-Cebria, C.; Ortiz, A.;
08035 Barcelona, Spain
7 Deparment of Medicine, Universitat Autònoma de Barcelona, Av. de Can Domènech, 08193 Bellaterra, Spain
Martinez-Diaz, I.; Martos, N.;
* Correspondence: b.benito@vhebron.net (B.B.); m.soler@vhebron.net (M.J.S.); Tel.: +34-93-4894072 (M.J.S.)
Dominguez-Báez, P.; Van den Bosch,
† These authors contributed equally to the work.
M.M.; Bermejo, S.; Pieper, M.P.; et al.
Enhanced Cardiorenal Protective
Abstract: Treatments with sodium–glucose 2 cotransporter inhibitors (SGLT2i) or endothelin receptor
Effects of Combining SGLT2
antagonists (ERA) have shown cardiorenal protective effects. The present study aimed to evaluate
Inhibition, Endothelin Receptor
the cardiorenal beneficial effects of the combination of SGLT2i and ERA on top of renin–angiotensin
Antagonism and RAS Blockade in
system (RAS) blockade. Type 2 diabetic mice (db/db) were treated with different combinations of
Type 2 Diabetic Mice. Int. J. Mol. Sci.
2022, 23, 12823. https://doi.org/
an SGLT2i (empagliflozin), an ERA (atrasentan), and an angiotensin-converting enzyme inhibitor
10.3390/ijms232112823 (ramipril) for 8 weeks. Vehicle-treated diabetic mice and non-diabetic mice were included as controls.
Weight, blood glucose, blood pressure, and kidney and heart function were monitored during the
Academic Editors: Salvatore Piro,
study. Kidneys and heart were collected for histological examination and to study the intrarenal RAS.
Stefania Di Mauro and
Treatment with empagliflozin alone or combined significantly decreased blood glucose compared
Agnese Filippello
to vehicle-treated db/db. The dual and triple therapies achieved significantly greater reductions
Received: 9 September 2022 in diastolic blood pressure than ramipril alone. Compared to vehicle-treated db/db, empagliflozin
Accepted: 18 October 2022 combined with ramipril or in triple therapy significantly prevented GFR increase, but only the triple
Published: 24 October 2022
combination exerted greater protection against podocyte loss. In the heart, empagliflozin alone or
Publisher’s Note: MDPI stays neutral combined reduced cardiac isovolumetric relaxation time (IVRT) and left atrium (LA) diameter as com-
with regard to jurisdictional claims in pared to vehicle-treated db/db. However, only the triple therapy was able to reduce cardiomyocyte
published maps and institutional affil- area. Importantly, the add-on triple therapy further enhanced the intrarenal ACE2/Ang(1-7)/Mas
iations. protective arm of the RAS. These data suggest that triple therapy with empagliflozin, atrasentan and
ramipril show synergistic cardiorenal protective effects in a type 2 diabetic mouse model.
Keywords: diabetes; diabetic nephropathy; chronic kidney disease; sodium–glucose cotransporter 2
Copyright: © 2022 by the authors.
inhibitors; endothelin receptor antagonists
Licensee MDPI, Basel, Switzerland.
This article is an open access article
distributed under the terms and
conditions of the Creative Commons
Attribution (CC BY) license (https://
1. Introduction
creativecommons.org/licenses/by/ Diabetes is the leading acquired risk factor for an accelerated progression of CKD to
4.0/). kidney failure [1]. Around 20 per cent of incident patients on kidney replacement therapy
Int. J. Mol. Sci. 2022, 23, 12823. https://doi.org/10.3390/ijms232112823 https://www.mdpi.com/journal/ijms
Int. J. Mol. Sci. 2022, 23, 12823 2 of 23
reach this stage due to diabetic kidney disease (DKD) [2]. Since the EMPA-REG OUTCOME
clinical trial in 2016 [3], sodium–glucose cotransporter 2 inhibitors (SGLT2i) have repeatedly
been shown to delay DKD progression in patients with preserved [4,5] and decreased
glomerular filtration rates (GFR) [4,6] when added on top of renin–angiotensin system
(RAS) blockade. SGLT2i represents the first drug class that has delayed DKD progression
since the beneficial effects of RAS blockers were demonstrated [7,8]. More recently, selective
endothelin receptor subtype A antagonists (ERA) [9] reduced albuminuria and delayed
DKD progression [10]. Atrasentan, which has a selectivity for type A versus B receptor of
1200:1, is one of the best studied ERAs [11]. The SONAR trial demonstrated that atrasentan
improved the composite renal outcome of doubling of serum creatinine or kidney failure
after a median follow-up of 2.2 years when compared to patients treated with placebo [11].
However, as previously observed for avosentan [12], fluid retention, and anemia were
more frequent in the atrasentan arm [11]. This favors heart failure and limits the use of
ERAs in clinical practice. In fact, ERAs are not available for routine treatment of DKD.
In contrast, SGLT2i displays diuretic effects [13], increases hemoglobin levels [14], and
decreases the risk of heart failure [15,16]. For this reason, the combination of an SGLT2i with
ERA may improve the safety of the latter group [17]. In addition, considering the different
mechanisms of action, combining an SGLT2i with ERA may provide synergistic cardiorenal
protective effects. Recently, a post hoc analysis of the SONAR trial in patients with type
2 diabetes and chronic kidney disease observed that the combination of atrasentan and
SGLT2i was beneficial in terms of less fluid retention and a larger decrease in albuminuria
compared to ERA alone [17].
In the present study, we analyzed the renal and heart protective effects of combining
ERA and SGLT2i on top of RAS blockade in the db/db mouse, which is a widely used
rodent model for type 2 diabetes research. Our main outcome was to study the effect of
these drugs on early diabetic hyperfiltration, and our secondary outcomes were to identify
kidney and heart protective effects at the tissue level. Moreover, we studied the impact of
ERAs and SGLT2i on the intrarenal RAS, a pathway clearly involved in the development of
diabetic nephropathy [18].
2. Results
2.1. Empagliflozin, Atrasentan, Ramipril, or Their Combinations Did Not Modify Body Weight or
Body Fat
After 8 weeks of treatment, vehicle-treated db/db mice weighed 13.0 g (95% CI: 9.6–16.3)
more than vehicle-treated db/m mice (Table 1). No differences were found between vehicle-
treated db/db mice and mice treated with empagliflozin, ramipril, atrasentan, or their
combinations, except for the ATR + RAM group. The latter group showed a decrease in
body weight not accompanied by a reduction in body fat volume (Table 1), suggesting that
the effect is unrelated to the study treatment. In line with the higher body weight, db/db
mice also showed a higher body fat volume (18.4 cm3 , IQR: 16.1–23.5) than vehicle-treated
db/m mice (3.3 cm3 , IQR: 1.9–4.9) (Table 1). Thus, body fat was 39.5% (95% CI: 32.3–46.7)
higher in db/db mice than in db/m mice when values were adjusted to body volume.
Further, only the group treated with the triple therapy showed increased subcutaneous and
intra-abdominal fat volume as compared to the vehicle-treated db/db despite the body
weight not being significantly higher in this group (Table 1).
Table 1. Body weight, body fat volume, and fat distribution at the end of the experiment of v
treated db/m mice, vehicle-treated db/db mice, and db/db mice treated with empagliflo
rasentan, ramipril, or their combinations.
db/db
db/db db/db db/db db/db db/db( $ 42.9(3 40.8(3 44.4(3 40.9(3 40.9(3
weight, body6.8–46.0)
fat volume, and fat distribution at the end of6.9–45.0) *(31.3
the experiment of vehicle-
26.7–29.9) 6.0–46.7) 9.8–49.4) .6–43
/>(g) 6.3–46.6) –40.6)
treated db/m mice, vehicle-treated db/db mice, and db/db mice treated with empagliflozin, atrasen-
Body 18.4
tan, ramipril, or their combinations.
3.3(1. 22.0(1 19.1(1 25.4(1 19.4(1 21.4(1 19.3vo $(1
9–4.9) 6.5–30.3) 6.5–25.4) 9.8–30.0) 5.8–25.7) 4.9–22.9) .4–25
lume (cm3) 6.1–23.5)
db/db db/db
db/db
db/db db/db db/db
db/m db/db EMP + RAM +
Subcutane- EMP
12.4 EMP + RAM 15.9ATR ATR + RAM RAM
ATR
2.0(1.
$
14.1(1 12.3(1 12.7(9. 12.8(9. 12.6(1
42.9 40.8 44.4 *(13.3
40.9 36.1 * 40.9
(g) (26.7–29.9) 3–3.4)
(36.3–46.6) (36.8–46.0) 1.3–18.9)
(36.0–46.7) 0.5–14.8)
(39.8–49.4) (36.9–45.0) 9–15.7)
(31.3–40.6) 5–14.4)
(37.6–43.0) .6–16
ume (cm3) 0.6–14.4) –18.0)
Body fat 3.3 $ 22.0 19.1 25.4 19.4 21.4 19.3
18.4
volume (cm3 ) Intra-ab-
(1.9–4.9) (16.1–23.5)
6.8
(16.5–30.3) (16.5–25.4) 7.0(5.9
(19.8–30.0) 9.0
(15.8–25.7) 6.8(5.9
(14.9–22.9) 7.8(5.4
(16.4–25.7)
1.4(0. 8.5(5.3 6.7(5 *(6.6–
Subcutaneous fat 2.0 12.4 $
6–1.7) 14.1 12.3
–10.7) 15.9 *
–9.6) 12.7 12.8
–10.0) 12.6
–8.9) –9.7
volume (cm3 ) volume (cm3)
(1.3–3.4) (10.6–14.4) .4–9.0)
(11.3–18.9) (10.5–14.8) (13.3–18.0) 11.1)
(9.9–15.7) (9.5–14.4) (11.6–16.7)
Intra-abdominal 1.4 6.8 $ db/m:
8.5 non-diabetic
7.0 mice treated 9.0with
* vehicle. db/db:
6.8 diabetic 7.8mice treated with6.7 vehicle. db/db
fat volume (cm3 ) (0.6–1.7) (5.4–9.0) (5.3–10.7)
diabetic mice(5.9–9.6) (6.6–11.1)
treated with empagliflozin. (5.9–10.0)
db/db EMP + RAM: (5.4–8.9) (5.1–9.7)
diabetic mice treated with em
db/m: non-diabeticflozinmice and ramipril.
treated db/db db/db:
with vehicle. EMP +diabetic
RAM +mice ATR: diabetic
treated mice
with treated
vehicle. with
db/db empagliflozin,
EMP: diabetic ramip
mice treated with atrasentan.
empagliflozin. db/db
db/db ATR:EMP + RAM:
diabetic diabetic
mice treatedmice treated
with with empagliflozin
atrasentan. db/db ATR and ramipril.
+ RAM: diabetic mice
db/db EMP + RAM + ATR: diabetic mice treated with empagliflozin, ramipril, and atrasentan. db/db ATR:
with atrasentan and ramipril. db/db RAM: diabetic mice treated with ramipril. Values are shown
diabetic mice treated with atrasentan. db/db ATR + RAM:$diabetic mice treated with atrasentan and ramipril.
db/db RAM: diabetic dian andtreated
mice interquartile
with ramipril. Values are p
range (Q1–Q3). < 0.05as
shown vehicle-treated db/db micerange
median and interquartile vs. vehicle-treated
(Q1–Q3). db/m
* p < 0.05
$ p < 0.05 vehicle-treated anymice
db/db treated db/db mice compared
vs. vehicle-treated db/m mice. with
* p vehicle-treated
< 0.05 any treateddb/db
db/dbmice.
mice compared
with vehicle-treated db/db mice.
2.2. Empagliflozin Alone or in Combination Reduced Blood Glucose
2.2. Empagliflozin Alone or in Combination Reduced Blood Glucose
After 8 weeks of treatment, vehicle-treated db/db developed hyperglycemi
After 8 weeks of treatment,
mg/dL, 95% CI: vehicle-treated
359–429) (Figuredb/db developed
1A,B). hyperglycemiaalone
Only empagliflozin, (394 mg/dL,
or in combinati
95% CI: 359–429)duced fasting blood glucose (mean reduction of 213 mg/dL in the 3reduced
(Figure 1A,B). Only empagliflozin, alone or in combination, groups) in db/d
fasting blood glucose (mean
(Figure 1B). reduction
Thus, of 213 mg/dL
empagliflozin preservesin its
theblood-glucose-lowering
3 groups) in db/db mice effect unde
(Figure 1B). Thus, empagliflozin preserves its blood-glucose-lowering effect under dual
or triple therapy combinations. In addition, empagliflozin, in monotherapy or com
or triple therapy combinations.
increased In addition,
glucosuria (p < 0.001empagliflozin, in monotherapy
for empagliflozin’s main effect)or(Figure
combined,
1C). Howeve
increased glucosuria (p < 0.001 for empagliflozin’s main effect) (Figure 1C). However, this
effect was milder in the EMP + RAM + ATR group.
effect was milder in the EMP + RAM + ATR group.
Figure 1. Blood glucose, urine glucose, blood pressure (BP), and heart rate of vehicle-treated
db/m mice, vehicle-treated db/db mice, and db/db mice treated with empagliflozin, atrasentan,
ramipril, or their combinations. (A) Mean blood glucose over time in mg/dL. Error bars display
standard deviation (SD). (B) Blood glucose (mg/dL) after 8 weeks of treatment. (C) Urine glucose to
plasma glucose ratio at the end of the treatment. (D) Systolic blood pressure in mmHg. (E) Diastolic
blood pressure in mmHg. (F) Heart rate in bpm. db/m: non-diabetic mice treated with vehicle.
db/db: diabetic mice treated with vehicle. db/db EMP: diabetic mice treated with empagliflozin.
db/db EMP + RAM: diabetic mice treated with empagliflozin and ramipril. db/db EMP + RAM +
Int. J. Mol. Sci. 2022, 23, 12823 4 of 23
ATR: diabetic mice treated with empagliflozin, ramipril, and atrasentan. db/db ATR: diabetic mice
treated with atrasentan. db/db ATR + RAM: diabetic mice treated with atrasentan and ramipril.
db/db RAM: diabetic mice treated with ramipril. Factorial ANOVA main effect results are displayed
next to the graph. PDiabetes : diabetes’ main effect. PEMP : empagliflozin’s main effect. PRAM :
ramipril’s main effect. PATR : atrasentan’s main effect. $ p < 0.05 vehicle-treated db/db mice vs.
vehicle-treated db/m mice. * p < 0.05 any treated db/db mice compared with vehicle-treated
db/db mice.
2.3. Dual or Triple Combination Therapy Further Reduced Blood Pressure
Although no differences were observed in BP before starting the treatments (Figure S1),
at the end of the experiment, systolic and diastolic BP were slightly lower in vehicle-treated
db/db than in non-diabetic controls (Figure 1D,E). Ramipril in monotherapy significantly
reduced systolic BP by 13.2 mmHg (95% CI: 4.8–21.6) compared with vehicle-treated
db/db mice, while empagliflozin or atrasentan alone had no impact on systolic or diastolic
BP. Ramipril in combination with empagliflozin or atrasentan reduced both systolic and
diastolic BP as compared with vehicle-treated db/db mice, which was not observed with
ramipril alone (Figure 1D,E). Thus, combination therapy offers a larger decrease in BP
than ramipril alone, but triple therapy does not further reduce BP as compared with dual
therapy. No evident differences were observed in heart rate (Figure 1F).
2.4. Empagliflozin Combined with Ramipril or Triple Therapy Prevent Diabetic
Glomerular Hyperfiltration
Albuminuria and GFR were measured as surrogate markers of diabetic nephropathy.
Consistent with the development of hyperfiltration, GFR increased in db/db mice after 8
weeks on vehicle Figures 2A and S2) compared to non-diabetic db/m mice. Monotherapy
approaches with either ramipril, empagliflozin, or atrasentan did not prevent hyperfil-
tration (Figure 2A), although ramipril’s main effect was significant (p = 0.005), suggesting
that the ACEi component has an impact in GFR modulation. Interestingly, the combina-
tion of empagliflozin and ramipril as well as the triple therapy prevented hyperfiltration
(92 ± 311 µL/100 g/min increase from baseline in vehicle-treated db/db mice vs. 192 ±
335 µL/100 g/min decrease in db/db mice treated with the combination of empagliflozin
and ramipril, p = 0.037, and 189 ± 407µL/100 g/min decrease in the mice treated with the
triple therapy, p = 0.025) (Figure 2A). Additionally, water intake was maintained in all the
groups at the end of the treatment (Figure S3). Thus, dehydration is an unlikely cause of
the reduction in GFR observed in these two groups.
Albuminuria increased in vehicle-treated diabetic mice (1575 µg/mg, 95% CI: 922–3268,
vs. db/m mice) (Figure 2B). However, there was a large variability in albuminuria levels
and despite a trend towards a reduced albuminuria for all active treatment arms, it only
reached statistical significance for empagliflozin monotherapy (UACR 897 µg/mg, 95% CI:
216–2131) (Figure 2B).
tion of empagliflozin and ramipril as well as the triple therapy prevented hyperfilt
(92 ± 311 µL/100 g/min increase from baseline in vehicle-treated db/db mice vs. 192
µL/100 g/min decrease in db/db mice treated with the combination of empaglifloz
ramipril, p = 0.037, and 189 ± 407µL/100 g/min decrease in the mice treated with the
therapy, p = 0.025) (Figure 2A). Additionally, water intake was maintained in a
Int. J. Mol. Sci. 2022, 23, 12823 5 ofunlikely
groups at the end of the treatment (Figure S3). Thus, dehydration is an 23 ca
the reduction in GFR observed in these two groups.
Figure 2. Glomerular filtration rate difference (∆GFR) from baseline and urinary albumin-to-
creatinine ratio (UACR) at the end of the experiment in vehicle-treated db/m mice, vehicle-treated
db/db mice, and db/db mice treated with empagliflozin, atrasentan, ramipril, or their combina-
tions. (A) ∆GFR measured in µL/100 g body weight/min. Difference was calculated subtracting pre-
treatment GFR to post-treatment GFR in each mouse. (B) UACR at the end of the experiment. db/m:
non-diabetic mice treated with vehicle. db/db: diabetic mice treated with vehicle. db/db EMP: dia-
betic mice treated with empagliflozin. db/db EMP + RAM: diabetic mice treated with empagliflozin
and ramipril. db/db EMP + RAM + ATR: diabetic mice treated with empagliflozin, ramipril, and
atrasentan. db/db ATR: diabetic mice treated with atrasentan. db/db ATR + RAM: diabetic mice
treated with atrasentan and ramipril. db/db RAM: diabetic mice treated with ramipril. Factorial
ANOVA main effect results are displayed next to the graph. PDiabetes : diabetes’ main effect. PEMP :
empagliflozin’s main effect. PRAM : ramipril’s main effect. PATR : atrasentan’s main effect. $ p < 0.05
vehicle-treated db/db mice vs. vehicle-treated db/m mice. * p < 0.05 any treated db/db mice
compared with vehicle-treated db/db mice.
2.5. Combination Therapy with Empagliflozin, Ramipril, and Atrasentan Increased Protective
Effects against Diabetic Kidney Injury
Diabetic nephropathy was evaluated in kidney histology by measuring glomerular
mesangial matrix area using PAS staining and by evaluating podocyte density. Mesangial
matrix expansion was observed in vehicle-treated db/db mice (7.9%, 95% CI: 5.5–10.0,
higher mesangial matrix area than non-diabetic mice), consistent with histological diabetic
nephropathy (Figure 3A,B). Empagliflozin, ramipril, and atrasentan, alone or in combina-
tion, significantly reduced mesangial matrix expansion in db/db mice (Figure 3B). Thus,
dual or triple therapy did not further add to the effect of monotherapy in this regard.
In addition, vehicle-treated db/db mice showed a reduced podocyte density when com-
pared to non-diabetic controls. All treatments except atrasentan monotherapy were able
to recover podocyte density. However, the triple therapy showed the greatest increase in
podocyte density (754 podocytes per mm2 (IQR: 607–814) in the vehicle-treated db/db
vs. 914 podocytes per mm2 (IQR: 855–1048) in db/db EMP + ATR + RAM) (Figure 3C,D).
Additionally, empagliflozin slightly improved renal tubulointerstitial collagen deposition
(p = 0.017 for empagliflozin’s main effect), although this trend was not identified in all the
groups receiving empagliflozin (Figure S4).
Int. J. Mol. Sci. 2022, 23, x FOR PEER REVIEW 6o
Int. J. Mol. Sci. 2022, 23, 12823 6 of 23
Figure 3. Glomerular mesangial matrix
Figure 3. Glomerular mesangialarea and
matrixpodocyte
area and density in vehicle-treated
podocyte db/m mice, db/m m
density in vehicle-treated
vehicle-treated db/db mice, and db/db mice treated with empagliflozin, atrasentan,
vehicle-treated db/db mice, and db/db mice treated with empagliflozin, atrasentan, ramipril, ramipril
their combinations.
or their combinations. (A) Representative
(A) Representative PAS-stained
PAS-stained glomerular glomerular microphotographs
microphotographs depicting depicting g
merular mesangial
glomerular mesangial matrix expansion
matrix expansion (×400 magnification).
(×400 magnification). (B) Percentage
(B) Percentage of glomerular
of glomerular mesan- mesang
matrix
gial matrix with with respect
respect to totalto total glomerular
glomerular tuft (C)
tuft area. area.Representative
(C) Representative
WilmsWilms
tumortumor 1 stained glom
1 stained
ular microphotographs showing podocyte density in each experimental
glomerular microphotographs showing podocyte density in each experimental group (×400 mag- group (×400 magnificatio
(D) Podocyte number per mm22 of glomerular area. db/m: non-diabetic mice treated with vehi
nification). (D) Podocyte number per mm of glomerular area. db/m: non-diabetic mice treated
db/db: diabetic mice treated with vehicle. db/db EMP: diabetic mice treated with empaglifloz
with vehicle. db/db: diabetic mice treated with vehicle. db/db EMP: diabetic mice treated with
db/db EMP + RAM: diabetic mice treated with empagliflozin and ramipril. db/db EMP + RAM
empagliflozin. db/db
ATR: EMPmice
diabetic + RAM: treateddiabetic mice treated with
with empagliflozin, empagliflozin
ramipril, and ramipril.
and atrasentan. db/dbdiabetic m
db/db ATR:
EMP + RAMtreated diabetic
+ ATR:with mice treated with empagliflozin, ramipril, and atrasentan.
atrasentan. db/db ATR + RAM: diabetic mice treated with atrasentan db/db ATR: and ramip
diabetic micedb/db
treated with diabetic
RAM: atrasentan. mice db/db ATR
treated + RAM:
with diabetic
ramipril. miceANOVA
Factorial treated with
mainatrasentan andare display
effect results
nextRAM:
ramipril. db/db to the diabetic
graph. PDiabetes : diabetes’
mice treated main
with effect. Factorial
ramipril. PEMP: empagliflozin’s
ANOVA main main effect.
effect PRAM: ramipr
results
main effect. P : atrasentan’s main effect. $ p < 0.05 vehicle-treated db/db mice vs. vehicle-trea
are displayed next to the graph. PDiabetes : diabetes’ main effect. PEMP : empagliflozin’s main effect.
ATR
db/mmain
PRAM : ramipril’s mice.effect.
* p < 0.05
PATRany treated db/db
: atrasentan’s mainmice compared
effect. with
$ p < 0.05 vehicle-treated
vehicle-treated db/db
db/db mice. φ p < 0
mice
combinations
vs. vehicle-treated of empagliflozin
db/m mice. * p < 0.05 any or atrasentan
treated db/dbwith mice
ramipril db/db vs.
compared ramipril
with db/db.
vehicle-treated
ϕ
db/db mice. p < 0.05 combinations of empagliflozin or atrasentan with ramipril db/db vs. ramipril
2.6. Atrasentan Alone Did Not Improve Diastolic Dysfunction, but Did Not Interfere with Pro
db/db.
tection Exerted by Ramipril or Empagliflozin
2.6. Atrasentan Alone Did Not Improve
Vehicle-treated Diastolic
diabetic mice Dysfunction, but Did
showed incipient Notof
signs Interfere with
diastolic dysfunction afte
Protection Exerted by Ramipril or Empagliflozin
weeks (Tables 2 and S1). This was observed by higher LA diameter (2.29 mm, IQR: 2.1
Vehicle-treated
2.49) and diabetic mice showedrelaxation
longer isovolumetric incipient signs
timesof diastolic
(IVRT, 19.9dysfunction after 8 than no
ms, IQR: 18.2–20.8)
weeks (Table 2 and Table S1). This was observed by higher LA diameter
diabetic littermates (1.83 mm, IQR: 1.75–2.06, and 14.3 ms, IQR: 12.5–17.0, (2.29 mm, IQR:
respectivel
2.15–2.49) and longer isovolumetric relaxation times (IVRT, 19.9 ms, IQR: 18.2–20.8)
Ramipril in monotherapy or combined with atrasentan improved IVRT in db/db mice than
non-diabetic= littermates (1.83 mm, main
0.039 for ramipril’s IQR: 1.75–2.06,
effect) butand
did14.3
notms, IQR: 12.5–17.0,
modify respectively).
the LA diameter (Table 2). On
Ramipril inempagliflozin
monotherapymonotherapy
or combined with atrasentan improved IVRT in db/db
or combined with ramipril as dual or triple therapy dmice
(p = 0.039 for ramipril’s
creased both main
IVRT effect)
and LAbut did notinmodify
diameter the LA
db/db mice. diameter
Thus, (Table 2).
both ramipril Only
and empagliflo
empagliflozin monotherapy or combined with ramipril as dual or triple therapy decreased
provide cardiac protection, although this was more comprehensive for empagliflozin. T
both IVRT and
ple LA diameter
therapy did in db/db
not mice.
further Thus,
add both structure
cardiac ramipril andandempagliflozin provide but its
function protection,
cardiac protection, although this was more comprehensive
rasentan component did not detract from it either. for empagliflozin. Triple therapy
did not further add cardiac structure and function protection, but its atrasentan component
did not detract
Table from it either.
2. Echocardiographic measurements after 8 weeks of treatment in vehicle-treated db/m m
vehicle-treated db/db mice, and db/db mice treated with empagliflozin, atrasentan, ramipril, or th
combinations.
Int. J. Mol. Sci. 2022, 23, 12823 7 of 23
Table 2. Echocardiographic measurements after 8 weeks of treatment in vehicle-treated db/m mice,
vehicle-treated db/db mice, and db/db mice treated with empagliflozin, atrasentan, ramipril, or
their combinations.
db/db db/db db/db db/db db/db db/db
db/m db/db
EMP EMP+RAM EMP+RAM+ATR ATR ATR+RAM RAM
LA diameter 1.83 2.29 $ 2.07 * 2.09 * 2.11 * 2.29 2.16 2.49
(mm) (1.75–2.06) (2.15–2.49) (2.00–2.15) (2.06–2.15) (1.99–2.28) (2.08–2.54) (1.85–2.29) (1.92–2.50)
LV end-diastolic 3.63 3.61 3.61 3.65 3.58 3.77 3.53 3.49
diameter (mm) (3.50–3.77) (3.35–3.74) (3.48–3.87) (3.40–3.76) (3.38–3.68) (3.54–3.86) (3.40–3.79) (3.35–3.63)
LV end-systolic 2.28 2.18 2.31 2.14 2.19 2.40 * 2.12 2.14
diameter (mm) (2.17–2.34) (2.05–2.44) (1.88–2.42) (2.04–2.46) (2.03–2.27) (2.31–2.52) (1.93–2.47) (1.91–2.29)
IVS thickness 0.89 0.90 0.85 0.89 0.89 0.90 0.87 0.91
(mm) (0.85–0.93) (0.87–0.98) (0.79–0.98) (0.81–0.92) (0.82–0.92) (0.85–0.95) (0.80–0.89) (0.89–0.91)
Ejection 72.1 73.1 75.2 74.7 75.4 71.2 76.7 74.2
fraction (%) (70.4–75.6) (70.7–74.2) (69.9–78.3) (70.9–80.0) (70.0–79.6) (69.7–73.0) (71.9–78.1) (72.7–76.2)
14.3 19.9 $ 15.1 * 14.5 * 13.0 * 17.5 14.7 * 17.7
IVRT (ms)
(12.5–17.8) (17.5–21.4) (13.3–16.8) (12.0–15.6) (11.5–16.4) (14.5–24.8) (10.8–18.9) (10.8–19.9)
LA: left atrium. LV: left ventricle. IVS: interventricular septum. IVRT: isovolumetric relaxation time. db/m:
non-diabetic mice treated with vehicle. db/db: diabetic mice treated with vehicle. db/db EMP: diabetic mice
treated with empagliflozin. db/db EMP + RAM: diabetic mice treated with empagliflozin and ramipril. db/db
EMP + RAM + ATR: diabetic mice treated with empagliflozin, ramipril, and atrasentan. db/db ATR: diabetic
mice treated with atrasentan. db/db ATR + RAM: diabetic mice treated with atrasentan and ramipril. db/db
RAM: diabetic mice treated with ramipril. Values are shown as median and interquartile range (Q1–Q3). $ p < 0.05
vehicle-treated db/db mice vs. vehicle-treated db/m mice. * p < 0.05 any treated db/db mice compared with
vehicle-treated db/db mice.
2.7. Only Triple Therapy Prevents Cardiomyocyte Hypertrophy While Different Interventions
Protect against LV Fibrosis
Cardiomyocyte hypertrophy was evaluated in H&E-stained LV sections. An increased
cardiomyocyte area was observed in vehicle-treated db/db mice (142 µm2 , IQR: 117–156,
in the db/db group vs. 106 µm2 , IQR: 98–127, in non-diabetic controls), consistent with
cardiomyocyte hypertrophy (Figure 4A,B). Only triple therapy significantly decreased the
size of cardiomyocytes (124 µm2 , IQR: 103–132, p = 0.031 vs. vehicle-treated db/db mice)
(Figure 4B).
Histological results were consistent with those of LV myosin heavy-chain gene expres-
sion. Diabetes was associated with unchanged α-myosin heavy-chain (α-MHC) (Figure 4E)
but increased β-myosin heavy-chain (β-MHC) gene expression (2.8-fold, IQR: 1.2–6.9,
increase vs. db/m mice, p = 0.042) (Figure 4F). All interventions except atrasentan de-
creased β-MHC. Additionally, dual or triple therapy increased the α-MHC/β-MHC ratio,
preventing the isoform shift to a more economical myosin heavy isoform (Figure 4G).
Further, myocardial fibrosis measured by Picrosirius red staining was increased in
vehicle-treated diabetic mice (Figure 4C). All active treatments protected against collagen
accumulation (Figure 4C,D). In this regard, there was an increase (1.5-fold, IQR: 0.9–1.7)
in LV collagen I gene expression in vehicle-treated db/db mice with an almost significant
diabetes’ main effect (p = 0.065) (Figure 4H). All interventions either in monotherapy or
in combination lead to significantly lower LV collagen I gene expression, except for the
combination of empagliflozin and ramipril (Figure 4H).
2.8. Ramipril Activated the Intrarenal ACE2/Ang(1-7)/Mas Pathway and This Is Maximized by
Triple Therapy
To evaluate the impact of interventions on the intrarenal RAS, first, we analyzed the
kidney gene expression of angiotensinogen (AGT), which was 2.3-fold (IQR: 1.6–3.6) higher
in vehicle-treated db/db mice than in db/m (Figure 5A). This increase in renal AGT gene
expression correlated with increased urine AGT excretion in the vehicle-treated db/db as
compared to their non-diabetic littermates (Figure 5B). These findings depict an increased
renal expression of the main precursor of RAS in a diabetic milieu. The triple therapy signif-
icantly reduced AGT mRNA expression but increased AGT urine excretion (Figure 5A,B).
In this sense, only diabetes’ main effect was significant (p < 0.001 in both cases) suggesting
Int. J. Mol. Sci. 2022, 23, 12823 8 of 23
Int. J. Mol. Sci. 2022, 23, x FOR PEER REVIEW 8 of 24
that these variations are not related to the triple therapy effect. No relevant differences
were observed in serum AGT levels (Figure 5C).
Figure 4. Left Figure
ventricle cardiomyocyte
4. Left hypertrophy,
ventricle cardiomyocyte fibrosis,
hypertrophy, and gene
fibrosis, expression
and gene expression ofofα-MHC,
α-MHC, β- β-
MHC and collagenMHC and collagen
I alpha-1 I alpha-1
chain chain in vehicle-treated
in vehicle-treated db/m mice,db/m mice, vehicle-treated
vehicle-treated db/dbmice,
db/db mice, and
and
db/db mice treated with empagliflozin, atrasentan, ramipril, or their combinations. (A) Repre-
db/db mice treated with empagliflozin, atrasentan, ramipril, or their combinations. (A) Repre-
sentative hematoxylin-eosin microphotographs of transversally sectioned cardiomyocytes (×400
sentative hematoxylin-eosin
magnification). (B)microphotographs of transversally
Cardiomyocyte transversal sectioned
area in µm2 measured in cardiomyocytes
H&E-stained sections. (×400
(C)
magnification).Representative Picrosirius-red-stained
(B) Cardiomyocyte transversalleftarea
ventricle
in µm 2 measured
sections (×400 magnification).
in H&E-stained(D) Left sections.
ventricle
Picrosirius-red-stained area in µm2. (E) mRNA expression of α-myosin heavy-chain (α-MHC). (F)
(C) Representative Picrosirius-red-stained left ventricle sections (×400 magnification). (D) Left ven-
mRNA expression of β-myosin heavy-chain (β-MHC). (G) α-MHC/β-MHC ratio. (H) mRNA ex-
tricle Picrosirius-red-stained
pression of collagenareaI in µm2 .chain.
alpha-1 (E) mRNA expression
An almost of increase
significant α-myosin heavy-chain
in collagen (α-MHC).
I expression was
(F) mRNA expression of β-myosin heavy-chain (β-MHC). (G) α-MHC/β-MHC ratio. (H) mRNA
observed in the vehicle-treated db/db group vs. the db/m: non-diabetic mice treated with vehicle.
db/db: diabetic
expression of collagen mice chain.
I alpha-1 treated An
withalmost
vehicle.significant
db/db EMP:increase
diabetic mice treated with
in collagen empagliflozin.
I expression was
db/db EMP + RAM: diabetic mice treated with empagliflozin and ramipril. db/db EMP + RAM +
observed in the vehicle-treated db/db group vs. the db/m: non-diabetic mice treated with
ATR: diabetic mice treated with empagliflozin, ramipril, and atrasentan. db/db ATR: diabetic mice vehicle.
db/db: diabetic micewith
treated treated with vehicle.
atrasentan. db/db ATR db/db
+ RAM:EMP: diabetic
diabetic mice mice
treatedtreated with empagliflozin.
with atrasentan and ramipril.
db/db EMP + db/db
RAM:RAM: diabetic
diabetic micemice treatedwith
treated with ramipril. Factorialand
empagliflozin ANOVA main effect
ramipril. db/dbresults
EMP are+displayed
RAM +
next to the graph. PDiabetes: diabetes’ main effect. PEMP: empagliflozin’s main effect. PRAM: ramipril’s
ATR: diabetic mice treated with empagliflozin, ramipril, and atrasentan. db/db ATR: diabetic mice
treated with atrasentan. db/db ATR + RAM: diabetic mice treated with atrasentan and ramipril.
db/db RAM: diabetic mice treated with ramipril. Factorial ANOVA main effect results are displayed
next to the graph. PDiabetes : diabetes’ main effect. PEMP : empagliflozin’s main effect. PRAM : ramipril’s
main effect. PATR : atrasentan’s main effect. $ p < 0.05 vehicle-treated db/db mice vs. vehicle-treated
db/m mice. * p < 0.05 any treated db/db mice compared with vehicle-treated db/db mice.
Int. J. Mol. Sci. 2022, 23, x FOR PEER REVIEW 10 of 24
Int. J. Mol. Sci. 2022, 23, 12823 9 of 23
Figure 5.5. Kidney,
Figure Kidney,urine,
urine,andandserum
serumangiotensinogen
angiotensinogenand andrenal
renalrenin
reninlevels
levelsininvehicle-treated
vehicle-treated
db/m
db/mmice,
mice,vehicle-treated
vehicle-treateddb/dbdb/dbmice,
mice,and
anddb/db
db/dbmice micetreated
treatedwith
withempagliflozin,
empagliflozin,atrasentan,
atrasentan,
ramipril,
ramipril, or their combinations. (A) Angiotensinogen (AGT) mRNA expression. (B) UrineAGT-to-
or their combinations. (A) Angiotensinogen (AGT) mRNA expression. (B) Urine AGT-to-
creatinine
creatinineratio.
ratio.(C)
(C)Serum
SerumAGT AGTvalues.
values. (D)
(D)Renin
ReninmRNAmRNA expression.
expression. (E)
(E)Representative
Representativekidneykidney
sections
sections showing
showing specific renin staining
specific renin staining((×400 magnification). Increased
×400 magnification). Increasedjuxtaglomerular
juxtaglomerularcell cellstaining
stain-
ing (grey arrows) is observed in treatments with ramipril and or ramipril combination with em-
(grey arrows) is observed in treatments with ramipril and or 2ramipril combination with empagliflozin
pagliflozin and/or atrasentan. (F) Renin stained area in µm . db/m: non-diabetic mice treated with
and/or atrasentan. (F) Renin stained area in µm2 . db/m: non-diabetic mice treated with vehicle.
vehicle. db/db: diabetic mice treated with vehicle. db/db EMP: diabetic mice treated with empagli-
diabetic
db/db:db/db
flozin. EMP mice treateddiabetic
+ RAM: with vehicle. db/dbwith
mice treated diabetic miceand
EMP:empagliflozin treated with db/db
ramipril. empagliflozin.
EMP +
db/db+ EMP
RAM RAM: diabetic
ATR:+diabetic mice with
mice treated treated with empagliflozin
empagliflozin, ramipril, and ramipril.
and db/db
atrasentan. EMPATR:
db/db + RAM dia-+
betic diabetic
ATR:mice micewith
treated treated with empagliflozin,
atrasentan. db/db ATR +ramipril, and atrasentan.
RAM: diabetic db/db
mice treated ATR:
with diabetic mice
atrasentan and
ramipril. db/db
treated with RAM: diabetic
atrasentan. db/db mice
ATR treated
+ RAM:withdiabetic
ramipril. Factorial
mice treatedANOVA main effect
with atrasentan results
and are
ramipril.
displayed
db/db RAM: nextdiabetic
to the graph.
mice treated : diabetes’
PDiabeteswith main
ramipril. effect. ANOVA
Factorial PEMP: empagliflozin’s main are
main effect results effect. PRAM:
displayed
ramipril’s
next to themain effect.
graph. PATR:: atrasentan’s
PDiabetes diabetes’ main main effect.
effect. PEMPp :