In Silico Identification of Cardioprotective Targets for SGLT2 Inhibitors in Type 2 Diabetes-Associated Heart Failure and Coronary Artery Disease

The Diabetes mellitus (DM) is a metabolic illness marked by elevated blood sugar levels brought on by partial or total insulin insufficiency as well as/or insulin resistance ^[1]. Because of changes in dietary patterns, lifestyles of the populace, and economic progress, DM has been diagnosed in around 537 million individuals globally. Up to 11.3% in 2030 and 12.2% in 2045, the incidence rate is predicted to increase ^[2]. DM causes a variety of long-term issues that significantly impact the health and well-being of those who have it, in addition to elevated blood glucose levels. DM can cause cardiovascular disorders (CVD), which is one of the most concerning aspects of the condition. People with diabetes are disproportionately affected by cardiovascular disorders, which include heart failure, stroke, and coronary artery disease. Research has repeatedly demonstrated that those with diabetes have a two to four times higher risk of developing CVD than those without the disease ^[3]. Among the cardiovascular risks of DM, Heart failure (HF) and Coronary artery disease (CAD) are the frequently overlooked complications ^{[4, 5]}.

According to the latest studies, persons with DM have a 2- to 5-fold higher risk of HF and 2- to 4-fold higher risk of CAD than people without DM ^{[6, 7]}. Despite receiving therapy for this comorbid illness, the individuals continue to be at risk for CVD. So it is necessary to replace the therapy with some other medications. Sodium-glucose cotransporter 2 (or) sodium glucose-linked transporter inhibitors (SGLT2Is) are oral anti-diabetic drugs (OADs) that improve glycemic control in people with type 2 diabetes (T2DM). About 90% of the reabsorption of filtered glucose is carried out by the renal proximal convoluted tubule, and their mechanism of action includes the blockage of SGLT2 channels there ^[8]. Due to its hypoglycemic properties as well as its effect on protecting the heart and kidneys, SGLT2I has emerged as a potentially beneficial family of drugs. The United States Food and Drug Administration (USFDA) has approved several SGLT2Is, including Bexagliflozin (BEXA), Canagliflozin (CANA), Empagliflozin (EMPA), Dapagliflozin (DAPA), and Ertugliflozin (ERTU), and the Japan Pharmaceuticals and Medical Devices Agency (PMDA) has approved medications like Ipragliflozin (IPRA), Tofogliflozin (TOFO), Luseogliflozin (LUSEO), and Sotagliflozin (SOTA) for the treatment of DM ^[9-13] . Large-scale clinical studies have shown the cardiovascular advantages of SGLT2Is. There is the possibility that these profits have something to do with better glycemic management, which is precisely the processes behind these profits are being extensively studied. Treatment with one of the SGLT2Is, EMPA, decreased the risk of CV death and hospitalization for HF in diabetic patients. Hospitalizations for HF were assessed as a secondary outcome, with the composite of cardiovascular mortality, nonfatal MI, and nonfatal stroke serving as the primary outcome. Of the 7020 type-2 diabetes patients who were randomized to receive either empagliflozin or a placebo, 490 of 4687 patients (10.5%) experienced the primary endpoint, while approximately 2300 patients (12%) in the placebo group experienced 282 events hazard ratio (HR): 0.86, 95% confidence interval (CI): 0.74-0.99, p = 0.04). At a median of 3.1 years, there was a 35% reduction in HF hospitalisation (HR 0.65, 95% CI: 0.50-0.85).Both those at risk for HF and those with a confirmed diagnosis of HF experienced this impact, according to the first major SGLT2 inhibitor trial Empagliflozin Cardiovascular Outcome Event Trial in Type 2 Diabetes Mellitus Patients–Removing Excess Glucose (EMPA-REG OUTCOME) study^[14]. In the CANVAS (Canagliflozin Cardiovascular Assesment Study and Canagliflozin Cardivascular Assesment Study- Renal) trial, 10142 individuals with type 2 diabetes and a high cardiovascular risk were randomized to receive canagliflozin versus a placebo. The main result was a composite endpoint consisting of nonfatal myocardial infarction, nonfatal stroke, or death from cardiovascular causes^[15] . The biggest study population of type 2 diabetes mellitus with atherosclerotic cardiovascular disease or at risk for it (n = 17,160) was included in the DECLARE TIMI-58 (Dapagliflozin Effect on Cardiovascular Events-Thrombolysis in Myocardial Infarction 58) experiment. Compared to placebo, the key composite effectiveness objective of cardiovascular mortality or hospitalization for heart failure (HR 0.83; 95% CI: 0.73–0.95) was significantly lower in the study with dapagliflozin. This reduction was primarily due to a substantial decrease in HF hospitalizations (HR 0.73, 95% CI: 0.61–0.88)^[16]. In DAPA-HF (Dapagliflozin and Prevention of Adverse Outcomes in Heart Failure) study, dapagliflozin vs placebo was randomly given to 4,744 patients with HF New York Heart Association (NYHA) class II-IV with an ejection fraction of around 40%, regardless of diabetes, in addition to the best available medical treatment for HF. When compared to placebo, dapagliflozin decreased the individual endpoints of worsening heart failure event (HR:0.70; 95% CI: 0.59- 0.83, NNT= 27) and cardiovascular mortality (HR: 0.82; 95% CI: 0.69-0.98) as well as the primary endpoint of Worsening heart failure or cardiovascular mortality (HR: 0.74; 95% CI: 0.65-0.85). This impact was shown over the course of the 18.2 month trial period and was unaffected by the existence of diabetes^[17].  These four major trials suggest that SGLT2Is are a significant new treatment strategy for at-risk individuals with or without DM in treating HF and CAD. Although these studies have recommended medications (SGLT2Is) for the treatment of CVD, their primary targets and methods of action remain unclear, and understanding them is very important. Network pharmacology is a field of study grounded on multidirectional pharmacology and systems biology. It is used to perform target analysis and drug molecular design using particular nodes and biomolecular network analysis techniques. Network pharmacology is based on a complete methodology that is more effective for examining the target and route interactions between medications and disorders than standard experimental pharmacology methods^[18]. Thus, in this work, we employed network pharmacology analysis to systemically identify putative SGLT2I targets on the CV protection of T2DM. this will offer a strong foundation for more comprehensive research into the mechanism of action of SGLT2Is on T2DM with CAD and HF.

MATERIALS AND METHODS:

SGLT2 Inhibitor-Related Targets:

The Canonical SMILES of SGLT2Is namely BEXA, DAPA, EMPA, CANA, ERTU, IPRA, TOFO, LUSEO, and SOTA were gathered using Pubchem (https://pubchem.ncbi.nlm.nih.gov/), which is a favoured chemical knowledge source with many different uses. PubChem has had several updates during the last two years. Over 120 data sources have contributed data to PubChem. PubChem comprises a diverse range of chemical entities, including lipids, polysaccharides, siRNA, miRNA, and chemically modified biopolymers, however small molecules make up the majority of its contents ^[19]. Then the SMILES datas were entered into the SwissTargetPrediction database (http://www.swisstargetprediction.ch/), by which Nine SGLT2 inhibitors' possible targets were determined by entering their Canonical SMILES into a target prediction algorithm. The Swiss target prediction technique estimates almost all of a small molecule's macromolecular targets while providing details on chemical constituents and biological activities ^[20]. In addition to SwissTargetPrediction database, we used another database known as DrugBank (https://go.drugbank.com/) to gather SGLT2Is related Genes. DrugBank is a distinct bioinformatics and chemical informatics database that includes information on 11,628 pharmaceuticals together with chemical properties, pharmacological targets, protein data, and other relevant information ^[21]. According to the data collected from these 2 databases, Duplicate genes present in each drug data were removed by using Excel and the resultant data of all the drugs were used.

Drug-Disease-Targets Prediction:

The terms T2DM, HF, CAD were retrieved from GeneCards (https://www.genecards.org/) and Disgenet (https://www.disgenet.org/) in order to identify potential target genes. Over 7000 human genes are included in the GeneCards database, and moreover, every single gene has a recognized gene symbol^[22] and DisGeNET is a comprehensive platform for discovery created to answer a range of queries related to the genetic basis of human disorders. DisGeNET is one of the largest collections of its sort currently accessible, with over 380,000 correlations between over 16,000 genes and 13,000 disorders^[23]. According to the data collected from these 2 databases, Duplicate genes present in each disease data were removed by using Excel, and the resultant data of all the diseases were used. According to the Targets of SGLT2Is and the disorders, we obtained a Venn diagram on Bioinformatics & Evolutionary Genomics (https://bioinformatics.psb.ugent.be/webtools/Venn/) and the intersected Genes were obtained known as HubGenes.

Construction of the Network Model:

Two network models, which include 1) a network model between nine SGLT2 inhibitors and their corresponding targets, and 2) a network model between the points where medications and diseases intersect, were constructed using a software tool called Cytoscape v3.10.1. Cytoscape can effectively represent the interactions between proteins, DNA, and genes in order to show network linkages^[24].

Core-Target screening for SGLT2 inhibitors in the management of diabetes mellitus with HF and CAD:

Analysis of protein-protein interactions inside the PPI network can identify core regulatory genes. One useful resource for finding PPI information is the STRING database (https://stringdb.org/), which has a plethora of data regarding known and predicted protein-protein interactions of different species^[25]. HubGenes were loaded into the STRING database, "Multiproteins" was chosen, and "Homo Sapiens" was chosen as the organism. The results were filtered with an interaction score of >0.9, disconnected nodes in the network were removed, and the remaining parameters were left unchanged. The TSV file was exported into the Cytoscape 3.10.1 application for additional analysis. The degrees of targets were determined using the Cytohubba plugin, which is based on Cytoscape, in order to identify the core targets in the common targets between drugs and diseases.

Protein-Protein Interaction (PPI) Network Construction of Core-Targets:

The STRING internet platform was used to build the PPI network of critical targets. The Gene Name list of target proteins was uploaded to the network once the multiple proteins module was chosen, confining the species to Homo sapiens and setting the confidence score to > 0.9.  The PPI diagram has solid circles for each gene, with the protein structure shown in the center of each circle and lines of various colors connecting the circles. Each line depicts a biological event that occurs between two proteins, such as signal transmission, gene expression regulation, cell migration, and so forth ^[24].

Functional Enrichment Analysis:

Genes are categorized by gene ontology (GO) into three groups: molecular function (MF), cellular component (CC), and biological process (BP). The signaling pathways implicated in genes were examined using the Kyoto Encyclopedia of Genes and Genomes (KEGG) ^[26]. For the GO and KEGG pathway studies in this work, the Database for Annotation, Visualization, and Integrated Discovery (David) (https://david.ncifcrf.gov/summary.jsp) was utilized. The SRplot was used to visualize the Bar and Bubble chart of GO and KEGG Functional enrichment analysis.

Molecular docking analysis:

SGLT2 inhibitors were molecularly docked to the top two targets in the PPI network and the first 3 key targets in the "pathway-Target" network. The Protein Data Bank (PDB, https://www.rcsb.org/) provided the 3D structure of the main targets (Homo sapiens mode and X-Ray diffraction value less than 3) in PDB format. Additionally, the Pubchem Database (https://pubchem.ncbi.nlm.nih.gov/) provided the 3D structure of the drugs. According to the discussion by Rizvi et al in 2013, the Autodock tool 1.5.7 platform was used to carry out the molecular docking study ^{[27, 28]}.

RESULTS:

Exploring SGLT2 Inhibitors' Potential Related Targets

The Canonical SMILES of SGLT2Is namely BEXA, DAPA, EMPA, CANA, ERTU, IPRA, TOFO, LUSEO, and SOTA were gathered using PubChem online platform (Table 1). By importing their Canonical SMILES into SwissTargetPrediction database, we obtained 100 Targets for Bexagliflozin, 100 Targets for Dapagliflozin, 100 Targets for Empagliflozin, 100 Targets for Canagliflozin, 100 Targets for Ertugliflozin, 100 Targets for Ipragliflozin, 100 Targets for Tofogliflozin, 100 Targets for Luseogliflozin, and 100 targets for Sotagliflozin. 7 Bexagliflozin Targets, 11 Dapagliflozin Targets, 10 Empagliflozin Targets, 8 Canagliflozin Targets, 0 Ertugliflozin Target, 0 Ipragliflozin Target, 0 Tofogliflozin Target, 0 Luseogliflozin Target, and 0 Sotagliflozin Target were obtained using DrugBank database. After deduplicating each drug targets retrieved from the two databases and combining, a total of 932 SGLT2Is targets were found (Table 2). Then entering them into Cytoscape v3.10.1 to construct a network model between nine SGLT2 inhibitors and their corresponding targets. There are 277 nodes (268 predicted targets and 9 drugs) and 932 edges. Where the SGLT2 inhibitors are shown as Blue-colored V nodes, drug targets (predicted targets) are represented by Orange nodes, and interactions between the medications and targets are represented by the edges (Figure 1). This illustrates a potential way that SGLT2 inhibitors and various targets interact.

Figure 1 Interaction network to indicate SGLT2Is and its targets

SGLT2i	Pubchem (cid)	Molecular Formula	CANONICAL SMILES
Bexagliflozin	25195624	C24H29ClO7	C1CC1OCCOC2=CC=C(C=C2)CC3=C(C=CC(=C3)C4C(C(C(C(O4)CO)O)O)O)Cl
Canagliflozin	24812758	C24H25FO5S	CC1=C(C=C(C=C1)C2C(C(C(C(O2)CO)O)O)O)CC3=CC=C(S3)C4=CC=C(C=C4)F
Empagliflozin	11949646	C23H27ClO7	C1COCC1OC2=CC=C(C=C2)CC3=C(C=CC(=C3)C4C(C(C(C(O4)CO)O)O)O)Cl
Dapagliflozin	9887712	C21H25ClO6	CCOC1=CC=C(C=C1)CC2=C(C=CC(=C2)C3C(C(C(C(O3)CO)O)O)O)Cl
Ertugliflozin	44814423	C22H25ClO7	CCOC1=CC=C(C=C1)CC2=C(C=CC(=C2)C34C(C(C(C(O3)(CO4)CO)O)O)O)Cl
Ipragliflozin	10453870	C21H21FO5S	C1=CC=C2C(=C1)C=C(S2)CC3=C(C=CC(=C3)C4C(C(C(C(O4)CO)O)O)O)F
Tofogliflozin	46908929	C22H26O6	CCC1=CC=C(C=C1)CC2=CC3=C(COC34C(C(C(C(O4)CO)O)O)O)C=C2
Luseogliflozin	11988953	C23H30O6S	CCOC1=CC=C(C=C1)CC2=CC(=C(C=C2C)OC)C3C(C(C(C(S3)CO)O)O)O
Sotagliflozin	24831714	C21H25ClO5S	CCOC1=CC=C(C=C1)CC2=C(C=CC(=C2)C3C(C(C(C(O3)SC)O)O)O)Cl

Table 1 – Information on nine SGLT2 inhibitors from pubchem

Drugs	Swiss target prediction (stp)	Drug bank (db)	Targets before de-duplicating	Duplicates	Removed	Targets after de-duplication
Bexagliflozin	100	7	107	1	1	106
Canagliflozin	100	8	108	1	1	107
Empagliflozin	100	10	110	1	1	109
Dapagliflozin	100	11	111	1	1	110
Ertugliflozin	100	0	100	0	0	100
Ipragliflozin	100	0	100	0	0	100
Tofogliflozin	100	0	100	0	0	100
Luseogliflozin	100	0	100	0	0	100
Sotagliflozin	100	0	100	0	0	100

Table 2 Total drug targets

Figure 2a Venn Diagram and 2b Protein-protein Interaction

Figure 3 SGLT2I-Target-Disease network construction

Figure 4 Gene Ontology (GO) Enrichment analysis of the targets of SGLT2Is in treating T2DM with CVD.

Figure 5 KEGG Pathway Enrichment analysis of the targets of SGLT2Is in treating T2DM with CVD.

Figure 6 Construction of Pathway Target Network.

Figure 7 Molecular Docking study A) Ipragliflozin-MAPK1; B) Sotagliflozin-MAPK1; C) Canagliflozin-PIK3CA; D) Ipragliflozin-PIK3CD; E) Tofogliflozin-MAPK1; F) Canagliflozin-MAPK1; G) Ertugliflozin-MAPK1; H) Tofogliflozin-PIK3CD; I) Empagliflozin-MAPK1; J) Dapagliflozin-MAPK1.

SGLT2Is	Pubchem ID	Targets	PDB ID	Binding energy (kcal/mol)
Bexagliflozin	25195624	EGFR	1M17	-2.78
		MAPK1	1PME	-4.22
		PIK3CA	3HHM	-3.65
		PIK3CD	5DXU	-2.65
Canagliflozin	24812758	EGFR	1M17	-3.86
		MAPK1	1PME	-5.35
		PIK3CA	3HHM	-5.53
		PIK3CD	5DXU	-4.34
Dapagliflozin	9887712	EGFR	1M17	-2.67
		MAPK1	1PME	-4.52
		PIK3CA	3HHM	-3.33
		PIK3CD	5DXU	-4.50
Empagliflozin	11949646	EGFR	1M17	-3.17
		MAPK1	1PME	-4.84
		PIK3CA	3HHM	-3.96
		PIK3CD	5DXU	-3.59
Ertugliflozin	44814423	EGFR	1M17	-2.58
		MAPK1	1PME	-4.97
		PIK3CA	3HHM	-3.80
		PIK3CD	5DXU	-3.88
Ipragliflozin	10453870	EGFR	1M17	-3.58
		MAPK1	1PME	-5.88
		PIK3CA	3HHM	-4.33
		PIK3CD	5DXU	-5.14
Luseogliflozin	11988953	EGFR	1M17	-2.65
		MAPK1	1PME	-4.20
		PIK3CA	3HHM	-3.08
		PIK3CD	5DXU	-2.95
Sotagliflozin	24831714	EGFR	1M17	-3.63
		MAPK1	1PME	-5.62
		PIK3CA	3HHM	-3.65
		PIK3CD	5DXU	-3.60
Tofogliflozin	46908929	EGFR	1M17	-4.52
		MAPK1	1PME	-5.26
		PIK3CA	3HHM	-3.91
		PIK3CD	5DXU	-4.88

Table 4 – Molecular docking findings of SGLT2Is

DISCUSSION:

Diabetes mellitus is a chronic metabolic disorder characterized by elevated blood glucose levels due to improper insulin synthesis or utilization by the body. Among the most concerning aspects of diabetes is its potential to result in cardiovascular disease (CVD). For those with diabetes, cardiovascular diseases such heart failure, stroke, and coronary artery disease are leading causes of death and disability. People with diabetes have a two to four times higher risk of developing CVD than people without the disease, according to numerous studies ^[3]. A lot of interest and attention was generated when it was discovered in clinical trials that SGLT2 inhibitors were useful in treating CVD in individuals with DM ^{[14, 15]}. However, the exact mechanism of SGLT2 inhibitors in the treatment of Diabetes induced CVD is still unknown. In this work, we investigated the SGLT2 inhibitor’s therapeutic mechanism in the treatment of T2DM complicated by CVD using the network pharmacology approach and molecular docking study. According to the combination of PPI network and Pathway-Target network, the top 4 core targets of SGLT2 inhibitors Mitogen-activated protein kinase 1 (MAPK1), Epidermal growth factor receptor (EGFR), Phosphatidylinositol-4,5-Bisphosphate 3-Kinase Catalytic Subunit Alpha (PIK3CA) and Phosphatidylinositol-4,5-Bisphosphate 3-Kinase Catalytic Subunit Delta (PIK3CD) acting on DM with HF and CAD were obtained. KEGG analysis was used to filter the key pathways, which mostly involved metabolic processes, lipid and atherosclerosis, chemical carcinogenesis - reactive oxygen species, AGE-RAGE signaling pathway in diabetic complications, and EGFR tyrosine kinase inhibitor resistance.

Among the top four key targets, the first, MAPK1, belongs to the MAPK protein family and is often referred to as ERK1(extracellular signal-regulated kinase 1) ^[29]. MAPKs are a class of serine/threonine protein kinases that, upon activation, phosphorylate a broad variety of downstream targets across several compartments to carry out highly specialized physiological responses ^[30]. Multiple biochemical signals can be integrated by MAPK, which is involved in many distinct cellular processes such as development, transcription regulation, differentiation, and proliferation. As a member of the MAPK family, cardiac hypertrophy and dysfunction have also been linked to ERK1 activation ^[31]. ERK 1 has been shown to have an impact on diabetes, as evidenced by the connection between dysregulated ERK 1 levels and dysregulation in a type 1 diabetic mice model ^[32]. Unlike empagliflozin or dapagliflozin, Canagliflozin inhibits ERK1/2 phosphorylation in lipopolysaccharide-stimulated human coronary artery endothelial cells, which lessens Hexokinase II expression and lowers inflammation ^[33]. This suggests that one of the possible targets in T2DM with HF and CAD is MAPK1. The next one is EGFR, the epidermal growth factor receptor is a protein on the cell surface that binds to the epidermal growth factor to control cell growth, proliferation, and survival. It also plays a role in atherogenesis, neointimal hyperplasia, blood pressure regulation, and reactive oxygen. Angiotensin II receptor type 1 (AT1) and EGFR receptors were hypothesized to be involved in the enhanced activity that led to this outcome.
According to Xu et al., the loss of plasmalemmal caveolae in mesenteric vascular smooth muscle cells may be connected to the increased myogenic constriction of the mesenteric artery in heart failure ^[34]. Resistance arterial dysfunction in type 2 diabetes was discovered to be influenced by increased EGFR phosphorylation by Belmadani et al ^[35]. This suggests that EGFR is another important target in DM with HF and CAD. Phosphoinositide 3-kinases (PI3Ks) are lipid kinases that create 3-phosphoinositides to regulate important intracellular signaling and vesicle trafficking functions. The final two genes in the top five core targets are PIK3CA and PIK3CD. Class I PI3Ks are mostly activated to catalyze PtdIns(4,5)P2 phosphorylation, which yields PtdIns(3,4,5)P3, in response to stimulation of cell surface receptors^[36]. A catalytic component known as p110 is joined with a regulatory subunit to form these heterodimeric enzymes^[37]. The isoform nomenclature for PI3Kα, β, δ, and γ is determined by the catalytic p110 subunit found in a Class I PI3K heterodimer; that is, p110α (PIK3CA gene), p110β (PIK3CB), p110δ (PIK3CD), or p110γ (PIK3CG) are present in each of these isoforms, respectively [36]. Thrombosis can result in acute myocardial infarction and sudden cardiac mortality in cases of coronary heart disease, which is frequently linked to atherosclerosis ^[38]. Though its involvement in platelet function is relatively minor, PI3Kα has been implicated in IGF1-mediated increase of platelet function, according to early investigations utilizing the PI3Kα selective inhibitor PIK75. Moreover, research on inhibitors revealed that PI3Kα may possibly be involved in Platelet glycoprotein VI GPVI-mediated platelet function ^[39-41]. Following that, two groups created a mouse model in which the megakaryocytic lineage's p110α had been selectively eliminated. researchers discovered that PI3Kα is the isoform that preserves platelet function, even though PI3Kβ and α both contribute to Insulin-like growth factor 1 IGF1-mediated Protein kinase B (AKT) phosphorylation. This was achieved by combining genetic and pharmacological techniques, such as cautious titration of PI3Kα and β inhibitors ^[42]. The processes of integrin αIIbβ3 activation and α-granule release mediated by PAR4, thrombin, CRP, and fucoidan, and thrombus formation on a collagen-coated surface under flow, were not affected ^[42]. On the other hand, p110α deletion, except PI3Kα inhibition, led to a synergistic augmentation of TPO mediated priming of platelet activity through increased phosphorylation of ERG1/2 and creation of TxA2. This suggests that PI3Kα has a unique negative regulatory role in platelet function that is independent of kinase. Additionally, research indicates that PI3Kα plays many functions in platelet function; inhibitors and PI3Kα deficiency reduce ADP release at low GPVI activation levels and impair platelet adhesion to vWF under shear. In addition to PI3Kβ, PI3Kα is involved in the platelet priming effect of anti-phospholipid antibodies ^[43,44]. More significantly, inhibition and deletion of PI3Kα reduced arterial thrombosis without affecting bleeding time, indicating that it may be a target for the treatment of CVD ^[44]. In both human and mouse platelets, the expression of the PI3Kδ isoform is small and has minimal effect on platelet function ^[45-47]. It is yet unknown how important PI3K inhibition is for myocardial infarction (MI). The significance of PI3K inhibition for myocardial infarction (MI) is still unclear. In animal models, the Class I PI3Kδ/γ inhibitor TG100-115 has been shown to provide cardio protection by reducing the formation of infarcts and maintaining myocardial function, even when administered up to three hours after cardiac infusion. This led to its entry into phase I and II clinical trials for acute MI ^[48]. According to published research, PI3K suppression may have cardioprotective and anti-thrombotic properties. Nevertheless, direct binding cannot be confirmed by our docking results for PIK3CA and PIK3CD “(-3.08 to -5.53 kcal/mol)”. Validation by experimentation is necessary.

In the enrichment of KEGG pathway, EGFR tyrosine kinase inhibitor resistance, AGE-RAGE signaling pathway in diabetic complications, Lipid and atherosclerosis and Metabolic pathways were found to be the significant pathways.   The first KEGG pathway is EGFR tyrosine kinase inhibitor resistance. in 2005, the first in vivo publications that demonstrated a direct link between increased EGFR/ErbB1 signalling and vascular dysfunction brought on by diabetes.These reports indicated that pharmacological inhibition of this RTK corrected the pathology in the renal and carotid arteries, as well as the mesenteric vascular bed ^[49-51]. The first studies on vascular EGFR were carried out in rats with streptozotocin-induced type 1 diabetes, a preclinical model of the disease. The results of these studies were then repeated in the vasculature of animals with type 2 diabetes ^[35,52,53], indicating that dysregulated EGFR/ErbB1/HER1 signalling was a prevalent mediator of vascular issues in both T1 and T2DM.In rats that were spontaneously hypertensive, EGFR inhibition decreased the generation of reactive oxygen species in the left ventricle and decreased the size of the hypertensive myocarditis ^[54]. Consequently, there may be a shared effect of EGFR tyrosine kinase inhibitor resistance to quicken the progression of DM with CVD. The next one, Lipid and atherosclerosis (AS), the primary cause of cardiovascular disease (CVD) overall is atherosclerosis, an inflammatory illness. According to C-reactive protein (CRP), atherosclerosis is classified as an inflammatory disease with a high concentration of activated monocytes/macrophages and T lymphocytes in lesions ^[55-57]. The cells within the lesions generate significant quantities of cytokines, primarily proinflammatory ones ^[58,59]. Even though there are other factors contributing to the development of AS, DM is a well-known cause of AS ^[60]. The results of a comprehensive review and meta-analysis indicate that drugs that lower blood sugar, like SGLT2 inhibitors, can significantly minimize the risk of AS events ^[61]. Three distinct cell types are involved in the production of atherosclerotic plaques: monocytes/macrophages, vascular smooth muscle cells (VSMCs), and endothelial cells (ECs). Excessive lipids and lipoproteins build up in the subendothelial matrix during the first stage as a result of endothelial dysfunction. An accumulation of oxidized low-density lipoprotein (Ox-LDL) initiates this process by inducing the release of adhesion molecules such as monocyte chemoattractant protein 1 (MCP-1), macrophage colony-stimulating factor (M-CSF), intercellular cell adhesion molecule-1 (ICAM-1), and vascular cell adhesion protein 1 (VCAM-1). Subsequently, monocytes in circulation go towards the intima, where they multiply, undergo macrophage differentiation, and merge with lipoproteins to create foam cells. Tumour necrosis factor alpha (TNFα) and IL-1 are pro-inflammatory cytokines that are expressed by monocytes and macrophages. Pro-fibrosis chemicals can also be secreted by VSMCs that migrate out of the medial layer. Necrotic cores are formed when foam cells and SMCs die as a result of the ongoing migration of monocytes into the intima and the buildup of lipids and lipoproteins. Atherosclerotic plaques will eventually form when platelets clone to create thrombi. Acute cardiovascular and cerebrovascular events can arise from arterial blockage or plaque rupture ^[62-64]. Numerous metabolic traits, such as hypertension, dyslipidemia, obesity, and hyperglycemia, are independent risk factors for atherosclerosis ^[65]. These metabolic diseases contribute to the pathophysiology of atherosclerosis by promoting the production of plaques and oxidative stress, inflammation, endothelial dysfunction, and impaired autophagy ^[66]. The next one is the AGE-RAGE signaling pathway in diabetic complications; AGE-RAGE signaling is a well-researched cascade in diabetes mellitus. It has been discovered that vascular calcification in diabetes can be directly mediated by the AGE-RAGE signaling pathway. Furthermore, because the route causes oxidative stress, elevated inflammation, and heightened extracellular matrix buildup that results in diastolic and systolic dysfunction, it may potentially have an influence on diabetes consequences. When diabetes problems arise, activation of the AGE-RAGE signaling pathway may result in excessive generation of advanced glycation end products that harm cardiomyocytes and induce heart failure (HF), which, like outcomes of a different study's pathway analysis, suggests that the AGE-RAGE signaling pathway in diabetes complications may affect the onset of HF ^[67-69]. The next one is metabolic pathways, the heart always needs energy, and it uses a range of sources, including glucose and free fatty acids. The heart lacks sufficient energy sources due to poor glucose utilization in stress-related disorders such as T2DM and HF. Fatty acids and ketones offer an alternate energy source in this dire situation. The metabolism of these secondary energy stores gives the heart an increased supply of ATP ^[70-72]. Insulin-mediated glucose utilization is diminished in persons with type 2 diabetes. It is more dependable for free fatty acids (FFAs) to provide enough energy. Nevertheless, metabolizing FFAs requires more oxygen than metabolizing glucose ^[73]. Reduced heart metabolic efficiency and insufficient ATP generation are the results of this metabolic condition. Numerous investigations have demonstrated that SGLT-2 inhibitor therapy protects diabetic individuals by enhancing cardiac energy metabolism. β-hydroxybutyrate (βOHB) is a ketone that is promoted by empagliflozin. According to some research, βOHB might serve as a more affordable energy source and "super fuel" for those suffering from diabetic cardiovascular disease ^[74]. As previously discussed, when a person has diabetes and heart failure, glucose is no longer their main energy source. through encouraging the ketone body formation from energy sources, empagliflozin reverses cardiac remodelling and improves systolic function and myocardial energy metabolism. This improves poor cardiac remodelling in heart failure patients ^[75]. The administration of empagliflozin decreased ventricular preload and elevated plasma ketones in a diabetic db/db mouse model fed on a fasting diet. There is a positive correlation between myocardial ATP generation and elevated ketone levels ^[76]. Furthermore, ketone bodies are acknowledged as more than just a different kind of energy. Additionally, they can function as an inflammatory suppressant by blocking nucleotide-binding oligomerization domain-like receptor P3 (NLRP3) pathways. Contrasting results have been reported by other investigations, though. Empagliflozin alleviated heart failure symptoms in mice, however ketone bodies did not mediate this benefit, according to Verma et al. (2018). Rather, they saw more oxidation of traditional energy sources like glucose and fatty acids. This implies that SGLT2 inhibitors may have multifactorial and context-dependent metabolic effects ^[77,78]. In summary, our study methodically clarified the underlying molecular pathways by which SGLT2 inhibitors interfered with T2DM and CVD at the same time using network pharmacology and molecular docking. Thus, additional experimental results should be examined to attain the dependability of anticipated outcomes through both in vitro and in vivo methods.

Limitations: Instead of identifying causative pathways, network pharmacology finds connections. Binding energies below the usual threshold for high-confidence interactions (≤ -7.0 kcal/mol) were shown by the molecular docking data. As a result, these results should only be regarded as hypothesis-generating. Before any claim of direct target engagement can be made, experimental confirmation utilizing methods such cellular temperature shift assays, isothermal titration calorimetry, or surface plasmon resonance is necessary.

CONCLUSION

A network pharmacology analysis identified 27 possible SGLT2 inhibitor targets linked to coronary artery disease and heart failure associated with type 2 diabetes. With the exception of metabolic enzymes, the most promising candidate targets for additional validation are EGFR, MAPK1, PIK3CA, and PIK3CD. KEGG pathway analysis showed that lipid metabolism, atherosclerosis pathways, and AGE-RAGE signaling were all involved. Weak binding energies (-2.58 to -5.88 kcal/mol) were suggested by molecular docking, and direct target engagement must be confirmed through experimental validation. This study provides a hypothesis generating framework for future mechanistic investigations.

REFERENCES

1. Organization WH. WHO; 2016. Global Report on Diabetes. Publications of the World Health Organization Switzerland. 2016:1-88.
2. Tian Z, Fan Y, Sun X, Wang D, Guan Y, Zhang Y, et al. Predictive value of TCM clinical index for diabetic peripheral neuropathy among the type 2 diabetes mellitus population: A new observation and insight. Heliyon. 2023.
3. Forbes JM, Cooper ME. Mechanisms of diabetic complications. Physiological reviews. 2013;93(1):137-88.
4. Control CfD, Prevention. National diabetes fact sheet: national estimates and general information on diabetes and prediabetes in the United States, 2011. Atlanta, GA: US department of health and human services, centers for disease control and prevention. 2011;201(1):2568-9.
5. Kenny HC, Abel ED. Heart failure in type 2 diabetes mellitus: impact of glucose-lowering agents, heart failure therapies, and novel therapeutic strategies. Circulation research. 2019;124(1):121-41.
6. Collaboration ERF. Diabetes mellitus, fasting blood glucose concentration, and risk of vascular disease: a collaborative meta-analysis of 102 prospective studies. The lancet. 2010;375(9733):2215-22.
7. Avogaro A, Azzolina D, Fadini G, Baldi I. Incidence of heart failure in patients with type 1 diabetes: a systematic review of observational studies. Journal of Endocrinological Investigation. 2021;44:745-53.
8. Paolisso P, Bergamaschi L, Gragnano F, Gallinoro E, Cesaro A, Sardu C, et al. Outcomes in diabetic patients treated with SGLT2-Inhibitors with acute myocardial infarction undergoing PCI: The SGLT2-I AMI PROTECT Registry. Pharmacological research. 2023;187:106597.
9. Poole RM, Prossler JE. Tofogliflozin: first global approval. Drugs. 2014;74:939-44.
10. Vaduganathan M, Januzzi Jr JL. Preventing and treating heart failure with sodium-glucose co-transporter 2 inhibitors. The American Journal of Medicine. 2019;132(10):S21-S9.
11. Poole RM, Dungo RT. Ipragliflozin: first global approval. Drugs. 2014;74:611-7.
12. Takahashi T, Yamamoto K. Pharmacological and clinical profile of a new SGLT2 inhibitor, luseogliflozin (Lusefi®). Nihon yakurigaku zasshi Folia pharmacologica Japonica. 2015;146(3):150-8.
13. Markham A, Keam SJ. Sotagliflozin: first global approval. Drugs. 2019;79(9):1023-9.
14. Zinman B, Wanner C, Lachin JM, Fitchett D, Bluhmki E, Hantel S, et al. Empagliflozin, cardiovascular outcomes, and mortality in type 2 diabetes. New England Journal of Medicine. 2015;373(22):2117-28.
15. Neal B, Perkovic V, Mahaffey KW, De Zeeuw D, Fulcher G, Erondu N, et al. Canagliflozin and cardiovascular and renal events in type 2 diabetes. New England Journal of Medicine. 2017;377(7):644-57.
16. Wiviott SD, Raz I, Bonaca MP, Mosenzon O, Kato ET, Cahn A, et al. Dapagliflozin and cardiovascular outcomes in type 2 diabetes. New England Journal of Medicine. 2019;380(4):347-57.
17. McMurray JJ, Solomon SD, Inzucchi SE, Køber L, Kosiborod MN, Martinez FA, et al. Dapagliflozin in patients with heart failure and reduced ejection fraction. New England Journal of Medicine. 2019;381(21):1995-2008.
18. Gosak M, Markovič R, Dolenšek J, Rupnik MS, Marhl M, Stožer A, et al. Network science of biological systems at different scales: A review. Physics of life reviews. 2018;24:118-35.
19. Kim S, Chen J, Cheng T, Gindulyte A, He J, He S, et al. PubChem 2023 update. Nucleic acids research. 2023;51(D1):D1373-D80.
20. Daina A, Michielin O, Zoete V. SwissTargetPrediction: updated data and new features for efficient prediction of protein targets of small molecules. Nucleic acids research. 2019;47(W1):W357-W64.
21. Wishart DS, Feunang YD, Marcu A, Guo AC, Liang K, Vázquez-Fresno R, et al. HMDB 4.0: the human metabolome database for 2018. Nucleic acids research. 2018;46(D1):D608-D17.
22. Sharma A, Demissei BG, Tromp J, Hillege HL, Cleland JG, O'Connor CM, et al. A network analysis to compare biomarker profiles in patients with and without diabetes mellitus in acute heart failure. European Journal of Heart Failure. 2017;19(10):1310-20.
23. Piñero J, Queralt-Rosinach N, Bravo A, Deu-Pons J, Bauer-Mehren A, Baron M, et al. DisGeNET: a discovery platform for the dynamical exploration of human diseases and their genes. Database. 2015;2015:bav028.
24. Mai Z, Li H, Chen G, Chen E, Liu L, Lun Z, et al. A bioinformatics investigation into the pharmacological mechanisms of sodium-glucose co-transporter 2 inhibitors in diabetes mellitus and heart failure based on network pharmacology. Cardiovascular Drugs and Therapy. 2022;36(4):713-26.
25. Szklarczyk D, Morris JH, Cook H, Kuhn M, Wyder S, Simonovic M, et al. The STRING database in 2017: quality-controlled protein–protein association networks, made broadly accessible. Nucleic acids research. 2016:gkw937.
26. Wang A, Li Z, Zhuo S, Gao F, Zhang H, Zhang Z, et al. Mechanisms of Cardiorenal Protection With SGLT2 Inhibitors in Patients With T2DM Based on Network Pharmacology. Frontiers in cardiovascular medicine. 2022;9:857952.
27. Rizvi SMD, Shakil S, Haneef M. A simple click by click protocol to perform docking: AutoDock 4.2 made easy for non-bioinformaticians. EXCLI journal. 2013;12:831.
28. Morris GM, Huey R, Lindstrom W, Sanner MF, Belew RK, Goodsell DS, et al. AutoDock4 and AutoDockTools4: Automated docking with selective receptor flexibility. Journal of computational chemistry. 2009;30(16):2785-91.
29. McKay M, Morrison D. Integrating signals from RTKs to ERK/MAPK. Oncogene. 2007;26(22):3113-21.
30. Qi M, Elion EA. MAP kinase pathways. Journal of cell science. 2005;118(16):3569-72.
31. Wang S, Ding L, Ji H, Xu Z, Liu Q, Zheng Y. The role of p38 MAPK in the development of diabetic cardiomyopathy. International journal of molecular sciences. 2016;17(7):1037.
32. Tessaro FHG, Ayala TS, Bella LM, Martins JO. Macrophages from a type 1 diabetes mouse model present dysregulated Pl3K/AKT, ERK 1/2 and SAPK/JNK levels. Immunobiology. 2020;225(2):151879.
33. Uthman L, Kuschma M, Römer G, Boomsma M, Kessler J, Hermanides J, et al. Novel anti-inflammatory effects of canagliflozin involving hexokinase II in lipopolysaccharide-stimulated human coronary artery endothelial cells. Cardiovascular Drugs and Therapy. 2021;35:1083-94.
34. Xu Y, Henning RH, Sandovici M, van der Want JJ, van Gilst WH, Buikema H. Enhanced myogenic constriction of mesenteric artery in heart failure relates to decreased smooth muscle cell caveolae numbers and altered AT1‐and epidermal growth factor‐receptor function. European journal of heart failure. 2009;11(3):246-55.
35. Belmadani S, Palen DI, Gonzalez-Villalobos RA, Boulares HA, Matrougui K. Elevated epidermal growth factor receptor phosphorylation induces resistance artery dysfunction in diabetic db/db mice. Diabetes. 2008;57(6):1629-37.
36. Vanhaesebroeck B, Stephens L, Hawkins P. PI3K signalling: the path to discovery and understanding. Nature reviews Molecular cell biology. 2012;13(3):195-203.
37. Tsolakos N, Durrant T, Chessa T, Suire S, Oxley D, Kulkarni S, et al. Quantitation of class IA PI3Ks in mice reveals p110-free-p85s and isoform-selective subunit associations and recruitment to receptors. Proceedings of the National Academy of Sciences. 2018;115(48):12176-81.
38. Otsuka F, Yasuda S, Noguchi T, Ishibashi-Ueda H. Pathology of coronary atherosclerosis and thrombosis. Cardiovascular diagnosis and therapy. 2016;6(4):396.
39. Hers I. Insulin-like growth factor-1 potentiates platelet activation via the IRS/PI3Kα pathway. Blood, The Journal of the American Society of Hematology. 2007;110(13):4243-52.
40. Kim S, Garcia A, Jackson SP, Kunapuli SP. Insulin-like growth factor-1 regulates platelet activation through PI3-Kα isoform. Blood, The Journal of the American Society of Hematology. 2007;110(13):4206-13.
41. Gilio K, Munnix IC, Mangin P, Cosemans JM, Feijge MA, Van der Meijden PE, et al. Non-redundant roles of phosphoinositide 3-kinase isoforms α and β in glycoprotein VI-induced platelet signaling and thrombus formation. Journal of Biological Chemistry. 2009;284(49):33750-62.
42. Blair TA, Moore SF, Williams CM, Poole AW, Vanhaesebroeck B, Hers I. Phosphoinositide 3-kinases p110α and p110β have differential roles in insulin-like growth factor-1–mediated Akt phosphorylation and platelet priming. Arteriosclerosis, thrombosis, and vascular biology. 2014;34(8):1681-8.
43. Terrisse A-D, Laurent P-A, Garcia C, Gratacap M-P, Vanhaesebroeck B, Sie P, et al. The class I phosphoinositide 3-kinases α and β control antiphospholipid antibodies-induced platelet activation. Thrombosis and Haemostasis. 2016.
44. Laurent P-A, Hechler B, Solinhac R, Ragab A, Cabou C, Anquetil T, et al. Impact of PI3Kα (phosphoinositide 3-kinase alpha) inhibition on hemostasis and thrombosis. Arteriosclerosis, Thrombosis, and Vascular Biology. 2018;38(9):2041-53.
45. Kim M-S, Pinto SM, Getnet D, Nirujogi RS, Manda SS, Chaerkady R, et al. A draft map of the human proteome. Nature. 2014;509(7502):575-81.
46. Zhang J, Vanhaesebroeck B, Rittenhouse SE. Human platelets contain p110δ phosphoinositide 3-kinase. Biochemical and biophysical research communications. 2002;296(1):178-81.
47. Senis YA, Atkinson BT, Pearce AC, Wonerow P, Auger JM, Okkenhaug K, et al. Role of the p110δ PI 3-kinase in integrin and ITAM receptor signalling in platelets. Platelets. 2005;16(3-4):191-202.
48. Doukas J, Wrasidlo W, Noronha G, Dneprovskaia E, Fine R, Weis S, et al. Phosphoinositide 3-kinase γ/δ inhibition limits infarct size after myocardial ischemia/reperfusion injury. Proceedings of the National Academy of Sciences. 2006;103(52):19866-71.
49. Benter IF, Yousif MH, Griffiths SM, Benboubetra M, Akhtar S. Epidermal growth factor receptor tyrosine kinase‐mediated signalling contributes to diabetes‐induced vascular dysfunction in the mesenteric bed. British journal of pharmacology. 2005;145(6):829-36.
50. Benter IF, Yousif MH, Hollins AJ, Griffiths SM, Akhtar S. Diabetes-induced renal vascular dysfunction is normalized by inhibition of epidermal growth factor receptor tyrosine kinase. Journal of vascular research. 2005;42(4):284-91.
51. Yousif M, Benter I, Akhtar S. The role of tyrosine kinase‐mediated pathways in diabetes‐induced alterations in responsiveness of rat carotid artery. Autonomic and Autacoid Pharmacology. 2005;25(2):69-78.
52. Choi S-K, Galán M, Partyka M, Trebak M, Belmadani S, Matrougui K. Chronic inhibition of epidermal growth factor receptor tyrosine kinase and extracellular signal-regulated kinases 1 and 2 (ERK1/2) augments vascular response to limb ischemia in type 2 diabetic mice. The American journal of pathology. 2012;180(1):410-8.
53. Palen DI, Matrougui K. Role of elevated EGFR phosphorylation in the induction of structural remodelling and altered mechanical properties of resistance artery from type 2 diabetic mice. Diabetes/metabolism research and reviews. 2008;24(8):651-6.
54. Zeng S-y, Yan Q-j, Yang L, Mei Q-h, Lu H-q. Inhibition of the ROS-EGFR Pathway Mediates the Protective Action of Nox1/4 Inhibitor GKT137831 against Hypertensive Cardiac Hypertrophy via Suppressing Cardiac Inflammation and Activation of Akt and ERK1/2. Mediators of inflammation. 2020;2020.
55. Ross R. Atherosclerosis—an inflammatory disease. New England journal of medicine. 1999;340(2):115-26.
56. Hansson GrK, Libby P, Schönbeck U, Yan Z-Q. Innate and adaptive immunity in the pathogenesis of atherosclerosis. Circulation research. 2002;91(4):281-91.
57. Binder CJ, Chang M-K, Shaw PX, Miller YI, Hartvigsen K, Dewan A, et al. Innate and acquired immunity in atherogenesis. Nature medicine. 2002;8(11):1218-26.
58. Hansson GK, Holm J, Jonasson L. Detection of activated T lymphocytes in the human atherosclerotic plaque. The American journal of pathology. 1989;135(1):169.
59. Frostegård J, Ulfgren A-K, Nyberg P, Hedin U, Swedenborg J, Andersson U, et al. Cytokine expression in advanced human atherosclerotic plaques: dominance of pro-inflammatory (Th1) and macrophage-stimulating cytokines. Atherosclerosis. 1999;145(1):33-43.
60. Stamler J, Vaccaro O, Neaton JD, Wentworth D, Group MRFITR. Diabetes, other risk factors, and 12-yr cardiovascular mortality for men screened in the Multiple Risk Factor Intervention Trial. Diabetes care. 1993;16(2):434-44.
61. Ghosh-Swaby OR, Goodman SG, Leiter LA, Cheng A, Connelly KA, Fitchett D, et al. Glucose-lowering drugs or strategies, atherosclerotic cardiovascular events, and heart failure in people with or at risk of type 2 diabetes: an updated systematic review and meta-analysis of randomised cardiovascular outcome trials. The lancet Diabetes & endocrinology. 2020;8(5):418-35.
62. Lusis AJ, Fogelman AM, Fonarow GC. Genetic basis of atherosclerosis: part I: new genes and pathways. Circulation. 2004;110(13):1868-73.
63. Weber C, Noels H. Atherosclerosis: current pathogenesis and therapeutic options. Nature medicine. 2011;17(11):1410-22.
64. Xu J, Hirai T, Koya D, Kitada M. Effects of SGLT2 inhibitors on atherosclerosis: lessons from cardiovascular clinical outcomes in type 2 diabetic patients and basic researches. Journal of Clinical Medicine. 2021;11(1):137.
65. Aboonabi A, Meyer RR, Singh I. The association between metabolic syndrome components and the development of atherosclerosis. Journal of human hypertension. 2019;33(12):844-55.
66. Xu J, Kitada M, Ogura Y, Koya D. Relationship between autophagy and metabolic syndrome characteristics in the pathogenesis of atherosclerosis. Frontiers in cell and developmental biology. 2021;9:641852.
67. Bodiga VL, Eda SR, Bodiga S. Advanced glycation end products: role in pathology of diabetic cardiomyopathy. Heart failure reviews. 2014;19:49-63.
68. Fukami K, Yamagishi S-i, Okuda S. Role of AGEs-RAGE system in cardiovascular disease. Current pharmaceutical design. 2014;20(14):2395-402.
69. Tao Y-g, Huang X-F, Wang J-y, Kang M-r, Wang L-j, Xian S-x. Exploring molecular mechanism of huangqi in treating heart failure using network pharmacology. Evidence-Based Complementary and Alternative Medicine. 2020;2020.
70. Sowton AP, Griffin JL, Murray AJ. Metabolic profiling of the diabetic heart: toward a richer picture. Frontiers in Physiology. 2019;10:639.
71. Braunwald E. Mechanisms of Cardiorenal Effects of Sodium-Glucose Cotransporter 2 Inhibitors: JACC State-of-the-Art Review. 2020.
72. Tran DH, Wang ZV. Glucose metabolism in cardiac hypertrophy and heart failure. Journal of the American Heart Association. 2019;8(12):e012673.
73. Correale M, Lamacchia O, Ciccarelli M, Dattilo G, Tricarico L, Brunetti ND. Vascular and metabolic effects of SGLT2i and GLP-1 in heart failure patients. Heart Failure Reviews. 2023;28(3):733-44.
74. Ferrannini E, Baldi S, Frascerra S, Astiarraga B, Heise T, Bizzotto R, et al. Shift to fatty substrate utilization in response to sodium–glucose cotransporter 2 inhibition in subjects without diabetes and patients with type 2 diabetes. Diabetes. 2016;65(5):1190-5.
75. Gaborit B, Ancel P, Abdullah A, Maurice F, Abdesselam I, Calen A, et al. Effect of empagliflozin on ectopic fat stores and myocardial energetics in type 2 diabetes: the EMPACEF study. Cardiovascular diabetology. 2021;20:1-14.
76. Abdurrachim D, Manders E, Nicolay K, Mayoux E, Prompers JJ. Single dose of empagliflozin increases in vivo cardiac energy status in diabetic db/db mice. Cardiovascular Research. 2018;114(14):1843-4.
77. Youm Y-H, Nguyen KY, Grant RW, Goldberg EL, Bodogai M, Kim D, et al. The ketone metabolite β-hydroxybutyrate blocks NLRP3 inflammasome–mediated inflammatory disease. Nature medicine. 2015;21(3):263-9.
78. Verma S, Rawat S, Ho KL, Wagg CS, Zhang L, Teoh H, et al. Empagliflozin increases cardiac energy production in diabetes: novel translational insights into the heart failure benefits of SGLT2 inhibitors. JACC: Basic to Translational Science. 2018;3(5):575-87.