Dual inhibition of sodium–glucose linked cotransporters 1 and 2 exacerbates cardiac dysfunction following experimental myocardial infarction

Ninety-nine 7-week old male rats Fischer F344 rats (Charles River Laboratories, Wilmington, MA, USA) were randomized to myocardial infarction or sham groups in a ratio of 3:1 using the left anterior descending (LAD) coronary artery of myocardial infarction that provides a robust model of adverse cardiac remodeling akin to its human counterpart [15]. Animals were then further randomized in a ratio of 1:1:1 to receive either vehicle, the dual SGLT1/2 inhibitor, T-1095 (150 mg/kg/day, Sun-Shine Chemical Technology Co., Ltd, Shanghai, China) dosed as previously described [16], or the selective SGLT2 inhibitor, dapagliflozin 0.5 mg/kg b.i.d (gift of Astra Zeneca, Gothenburg, Sweden) [17] by gavage. Doses of T-1095 (150 mg/kg/day or 0.1 wt/wt admixed in chow) and dapagliflozin (1 mg/kg/day) were based on previously published literature according to their ability to achieve an approximate 50% reduction in plasma glucose when administered to diabetic rats [18, 19].

Myocardial infarction was induced 1 week later by ligation of the LAD coronary artery with sham animals undergoing thoracotomy and incision of the pericardial sac, but not LAD ligation [20]. Two days later, rats underwent echocardiography, and were randomized on confirmation of MI and demonstration of similar MI size based upon fractional shortening and wall motion score index [21] and followed for a further 7 weeks before being terminated. All procedures were performed in the research vivarium under anesthesia using 2.5% isoflurane supplemented with 100% O₂ as previously undertaken and shown to effectively reduce consciousness without impairing cardiac function [22]. Urine was obtained for measurement of glucose concentration just prior to termination by cervical dislocation under isoflurane anesthesia. Because of the comparatively short duration of the study in comparison with the half-life of erythrocytes, fructosamine rather than hemoglobin A_1c was used provide an integrated index of glycemia. Serum fructosamine concentrations was, accordingly, determined by colorimetric assay using a commercial diagnostic assay (Roche Diagnostics, Canada) analyzed on the Olympus AU400 automated chemistry analyzer (Beckman Coulter, Canada).

Animals underwent echocardiography and cardiac catheterization as described below. Following these procedures, rats were terminated and their hearts and lungs removed for weighing and other assessments, as also described below. Tibial length was measured to provide a morphometric index for cardiac hypertrophy and lung weight [23].

Sample size estimates were based on similar studies previously reported by our group using this animal model [24]. Of the cohort of 99 rats, five did not show evidence of wall motion abnormalities indicative of infarction when echocardiography was performed 2 days after LAD ligation. A further 17 (8 from the vehicle group and 9 from the dapagliflozin group died during surgery or during the immediate postoperative recovery period following myocardial infarction. Two post-MI animals were withdrawn from the study when it became evident that there were insufficient quantities of T-1095 for them to complete. As such, the final number of animals that completed the study as per protocol were 22, 13 and 15 among the MI + vehicle, MI + dapagliflozin and MI + T-1095 groups, respectively and 12, 7 and 6 among the sham + vehicle, sham + dapagliflozin and sham + T-1095 groups, respectively.

All animals were housed 2/cage at the St. Michael’s Hospital Animal Research Vivarium in a temperature-controlled (22 °C) room with a 12-h light/dark cycle and ad libitum access to commercial standard rat chow. Enrichment, proper animal handling and anesthesia procedure were used to minimize the pain and distress of the animals during the study. All animal studies were approved by the St. Michael’s Hospital Animal Care Committee in accordance with the Guide for the Care and Use of Laboratory Animals (NIH Publication No. 85-23, revised 1996).

Transthoracic echocardiography was performed, as previously described [25], under light anaesthesia (1% isoflurane supplemented with 100% O₂), at 2 days and 8 weeks post MI, prior to sacrifice. Images were acquired using a high-frequency ultrasound system (Vevo 2100, MS-250 transducer, Visualsonics, Toronto, ON). Two dimensional long-axis images of the LV in parasternal long- and short-axis views with M-mode measurements at mid-papillary muscle level and linear dimensions were analyzed offline (Vevo 2100 software v. 1.8) using the standard leading edge-to-leading edge technique by a single investigator, masked to treatment.

MI size was estimated 2 days after LAD ligation by measuring the percentage of endocardial circumferential extent of LV akinesis at end-diastole using 2D short-axis images of the LV at mid-papillary muscle level. Corrected LV mass was calculated using the Devereux and Reichek “cube” formula that includes a 20% correction for overestimation of LV mass based on data derived from previously reported validation studies [26, 27]. Fractional shortening (FS%) was calculated according to the formula: FS% = (LVIDd − LVIDs)/LVIDd × 100, where LVIDd and LVIDs are end-diastolic diameter and end-systolic diameter respectively, as previously described [28]. Three consecutive cardiac cycles were averaged for all analyses.

Cardiac catheterization was performed as previously published [25]. Briefly, rats were anaesthetized with 2.5% isoflurane, intubated using a 14 gauge catheter and ventilated using a pressure controlled ventilator (TOPO ventilator, Kent Scientific, Torrington, CT). Adequacy of anaesthesia was assessed by lack of response to surgical manipulation and loss of muscular tone. Rats were placed in the supine position on a water circulating heating pad and a 1.4 F pressure–volume (PV) catheter (SPR-838; Millar Instruments, Inc., Houston, TX) was inserted into the right carotid and advanced into the left ventricle and PV loops were generated. All pressure–volume (PV) loops were obtained with the ventilator turned off for 5–10 s and the animal apnoeic.

Data were acquired and recorded with a MPVS ultra^® data acquisition system (Millar Instruments) and LabChart Pro software (CHART 8.1 ADInstruments Inc., Colorado Springs, CO) under steady-state and following inferior vena cava occlusion (preload reduction). Conductance signals acquired with the Millar catheter were calibrated with estimated LV volumes derived echocardiographically using a two-point calibration method, matching LV maximal and minimal conductance signals and end-diastolic and end-systolic volumes (EDV and ESV) measured in long-axis view for each individual rat, respectively. Using the pressure conductance data, a range of functional parameters were then calculated.

To determine the extent of extracellular matrix deposition, sections were stained with antibodies to types I and III collagen (Southern Biotech, Birmingham, AL). The abundance of matrix within the non-infarct zone was then quantified as previously described [29]. To isolate the non-infarct zone from the infarct and the peri-infarct zone, the infarct and a 2 mm zone on either side of it were excluded from analysis. The proportional area staining brown within the remaining myocardium composed the non-infarct zone was then assessed using computer-assisted image analysis (Aperio ImageScope, Leica Biosystems Inc., Concord, Ontario, Canada), as previously described [30]. The whole non-infarct zone was used for quantification of ECM in order to prevent possible bias from using selected fields. For sham animals, the ECM content of the entire LV was quantitated by the same method, as described above.

The extent of cardiac myocyte hypertrophy was determined on haematoxylin and eosin stained sections, as previously reported [31]. In brief, stained sections were scanned with the Aperio Ultra-Resolution Digital Scanning System (Aperio Technologies Inc., Vista, CA), and the images were analyzed with The NDP.view2 viewer software (Hamamatsu Photonics, Hamamatsu City, Shizuoka Pref., Japan). Cardiac myocytes with elliptical nuclei in transverse section were selected. Diameter measurements were taken membrane to membrane across the narrowest point that crosses the nucleus. The average diameter of 30–50 myocytes per animal was measured based on the method previously reported [32].

The abundance of SGLT 1 and 2 were assessed by immunoblot, as previously described [33] using specific isoform antibodies (Santa Cruz Biotechnology, Dallas, TX). In brief, small pieces of heart tissue were rapidly collected from the non-infarct zone, snap frozen in liquid nitrogen and stored at − 80 °C until use. Lysates were prepared by homogenizing frozen heart tissue in ice-cold lysis buffer (50 mM Tris, pH 7.4, 150 mM NaCl, 1 mM EDTA, 1 mM EGTA, 0.5% Triton X-100 containing 1 mM Na₃VO₄, 50 mM NaF, 25 mM β-glycerophosphate, 10 mM Na pyrophosphate, 10 µg/ml aprotinin, 10 µg/ml leupeptin, 1 mM PMSF, 1 mM DTT). After 30 min incubation on ice, lysates were centrifuged to remove cell debris. Protein concentration was determined using Bradford reagent (Bio-rad, Hercules, CA) and standardized with known amounts of BSA. Western blot analysis was performed by resolving 50 µg of total protein on a 10% SDS-polyacrylamide gel by denaturing discontinuous gel electrophoresis according to the Laemmli method, and transferring to PVDF membranes (Roche, Mannheim, Germany). After incubating in blocking solution (5% nonfat dry milk, Tris buffered saline (pH 7.5), 0.1% Tween 20), membranes were immunoblotted with primary antibodies at 1:1000 dilution overnight at 4 °C. Membranes were then washed three times (5 min each) with TBST and blots were incubated with HRP-conjugated secondary antibodies (Dako) for 1 h at room temperature. The membranes were washed three times, and proteins were detected by the ECL system (Roche). Rat kidney served as positive control for SGLT2. Signals were digitized then analyzed with ImageJ software (NIH, Bethesda, MD). Results were calculated as the mean of at least three experiments relative to ß-actin immunolabelling. Kidney homogenates served as positive controls for both SGLT1 and − 2.

The abundance of atrial natriuretic peptide (ANP) was assessed by measuring its mRNA by quantitative real-time PCR in left ventricular tissue stored at − 80 °C as previously described [34]. In brief, SYBR green-based measurement of gene expression were performed on QuantStudio 7 Flex Real-Time PCR System (Applied Biosystems, Foster City, CA) according to the manufacturer’s instructions using the predesigned sequence-specific primers for ANP from Integrated DNA Technologies (Coralville, IA). Data were analyzed using Applied Biosystems Comparative CT method.

Data are expressed as mean ± SEM unless otherwise specified. Between group differences were analyzed by one way ANOVA with Fisher’s Protected Least Significant Difference test post hoc. All statistics were performed using GraphPad Prism 6 for Mac OS X (GraphPad Software Inc., San Diego, CA). A p value of < 0.05 was regarded as statistically significant.

No differences in baseline in body weight among the six groups were evident prior to randomization. When compared with rats that had been treated with dapagliflozin, those randomized to receive T-1095 had lower body weights in the post-MI setting (Table 1). Left ventricular weight, indexed to tibial length, increased following myocardial infarction with lower values noted in rats that had received dapagliflozin but lower still in those randomized to T-1095. Lung weight indexed to tibial length was increased in all animals that had undergone coronary artery ligation but to a greater extent in rats that had received T-1095 compared with dapagliflozin-treated animals.

Fructosamine concentrations were lower in both dapagliflozin and T-1095 when compared with vehicle-treated control animals. In rats that had undergone LAD ligation, however, fructosamine concentrations were lower than in vehicle-treated control animals though less so among rats that had received T-1095 (Table 1). No glucosuria was detected in untreated animals but was readily apparent in animals that received either T-1095 or dapagliflozin (Table 1). SGLT1 was easily detected by immunoblot of cardiac tissue but was unaffected by either myocardial infarction or treatment group assignment (Fig. 1). SGLT2 on the other hand, while abundantly expressed in the kidney could not be detected in cardiac tissue (data not shown).

The expression of ANP mRNA was assessed as a marker of extracellular fluid volume and cardiac wall stress. Its abundance was increased in the hearts of all animals that had undergone experimental myocardial infarction but less so in those that had received dapagliflozin when compared with either vehicle or T-1095 treated rats (Table 1).

Ejection fraction and fractional shortening were reduced in all animals that undergone experimental myocardial infarction when compared with sham surgery animals (Fig. 2, Table 2). Ejection fraction was, however, reduced to an even greater extent in those rats that had received T-1095 than those treated with either vehicle or dapagliflozin. Similar changes were also noted in left ventricular mass that while increased in all groups that had undergone LAD ligation, this was less so in those that had received T-1095 when compared with either vehicle- or dapagliflozin-treated animals.

Global left ventricular contractility, as measured by the rate of left ventricle pressure rise in early systole (dP/dt_max), was reduced in the post-MI setting but to a greater extent in those that had received T-1095 when compared with either vehicle or dapagliflozin-treated animals (Fig. 2, Table 3). The rapid, early decrease in LV pressure during the period of isovolumic relaxation (dP/dt_min) that reflects diastolic function was also reduced in rats that had undergone LAD ligation (Fig. 3). The extent of the reduction differed according to treatment assignment with those animals randomized to receive T-1095 experiencing the greatest reduction. This same pattern was also evident in other indices of diastolic function including both Tau, that assesses the early energy-dependent phase of relaxation as well as the end-diastolic pressure volume relationship (EDPVR) that provides an index of passive LV compliance (Fig. 3, Table 3). Systolic blood pressure was similar in all groups with the exception of post-MI rats receiving T-1095 where it was slightly lower.

Myocyte dimensions differed according to treatment assignment. Animals that received T-1095 did not show evidence of hypertrophy in the post-MI setting, contrasting those in rats randomized to vehicle or dapagliflozin (Fig. 4). The extent of fibrosis in the left ventricle in the area remote from the site of infarction also differed according to group assignment. Increased type III collagen was evident in the interstitium of those rats that had received either vehicle or T-1095 but not in those that had been treated with dapagliflozin (Figs. 5, 6). A different pattern was observed for type I collagen where increased deposition was only found in animals that had received T-1095 following experimental myocardial infarction (Figs. 5, 6).