Cardioprotective and Anti-Inflammatory Effects of Interleukin Converting Enzyme Inhibition in Experimental Diabetic Cardiomyopathy

Thirty-six male Sprague-Dawley (SD) rats (Charles River, Sulzfeld, Germany) weighing 300–330 g were maintained on a 12:12 h light:dark cycle and fed with standard chow. Diabetes was induced by a single intraperitoneal injection of streptozotocin (STZ) (70 mg/kg i.p. diluted in 0.1 mol/l sodium citrate buffer, pH 4.5; Sigma, Mannheim, Germany) in 24 animals (7). Twelve remaining SD animals were treated with vehicle (0.1 mol/l i.p. sodium citrate buffer, pH 4.5) and are referred to as the control group. Diabetic animals were randomly divided into two groups and treated either with the ICE inhibitor (ICEI) pralnacasan (50 mg/kg twice a day, n = 12; STZICEI; Hoechst-Aventis, Frankfurt, Germany) or with placebo (STZ, n = 12) for 6 weeks. Hyperglycemia (glucose >22 mmol/l) was confirmed 7 days later using a reflectance meter (Acutrend; Boehringer, Mannheim, Germany) and at the end of the study (glucose >30 mmol/l). This investigation conforms to the Guide for the Care and Use of Laboratory Animals published by the U.S. National Institutes of Health (NIH publication no. 85-23, revised 1985).

Six weeks after induction of diabetes, rats were anesthetized with intraperitoneal administration of pentobarbital (60 mg/kg i.p.) and were held in the half left-lateral position. Rats were allowed to breathe spontaneously during the echocardiographic studies. A commercially available echocardiographic machine equipped with a 15-MHz transducer (SONOS 5500, Philips, Germany) was used to record the long and short axis of the left ventricle as previously described (15). Measurement of the long axis was performed to evaluate end-diastolic volume. The M-mode recordings derived from the short axis were traced to obtain the fractional shortening as an index of left ventricular systolic function.

Two days after echocardiography, rats were intubated, artificially ventilated, and then subjected to open chest left ventricular catheterization under ketamin/xylazin anesthesia using a 2F pressure catheter (Millar-Instruments, Houston, TX) as reported earlier (3,4). The left ventricular pressure (LVP) that developed and the peak rate of rise in LVP (dP/dt_max) characterizing systolic performance were analyzed. Furthermore, the maximum rate of LVP decay (dP/dt_min), a hemodynamic parameter for left ventricular diastolic relaxation, and the left ventricular end diastolic pressure (LVEDP) were determined.

Transverse cryosections were embedded in optimal cutting temperature compound (Tissue Tek; Sakura Finetek), frozen in liquid nitrogen, and stored at −80°C. Serial 5-μm–thick cryosections were placed on 10% poly-l-lysine precoated slides and fixed in cold acetone. After blocking endogenous peroxidase activity, sections were incubated with an avidin/biotin blocking kit (Vector Laboratories, Burlingame, CA). Staining was performed with the following primary antibodies at the dilutions given (45 min, room temperature): for mouse anti-rat: ICAM-1 (clone 1A29, 1:100; Serotec, Munich, Germany), VCAM-1 (clone 5F10, 1:50; BabCO, Richmond, VA), CD18 (clone WT3, 1:50; Serotec), CD11a (clone WT1, 1:100; Serotec), CD11b (clone MRC OX-42, 1:100; Biozol, Eching, Germany), COL 1 (1:200; Abcam Limited, Cambridge, U.K.), Col3 (clone FH-7A, 1:40.000; Abcam Limited), and CD68 (clone ED1, 1:2,000; Serotec); for goat anti-rat: TNF-α (1:25; R&D Systems, Wiesbaden, Germany) and IL1-β (1:25; R&D Systems); and for rabbit anti-rat: IL-18 (1:25; R&D Systems). Sections used as negative control were incubated in PBS instead of a primary antibody. The mouse anti-rat antibodies were detected using a biotinylated goat anti-mouse IgG (1:400; Dianova, Hamburg, Germany), the goat anti-rat by biotinylated rabbit anti-goat IgG (1:333; Dianova), and the rabbit anti-rat by biotinylated goat anti-rabbit IgG (1:333; Dianova). Sections were incubated with peroxidase-conjugated streptavidin (Vectastatin ABC Elite; Vector Laboratories) (30 min, room temperature). Antigen antibody complexes were visualized by 3-amino-9-ethylcarbazole (Merck, Darmstadt, Germany). The slides were counterstained (Mayer's hematoxylin) and mounted in Kaiser's gelatin (both from Merck). Total collagen content of the sections stained with Sirius Red (Polyscience, Warrington, PA) was measured under circularly polarized light as recently reported (16). Immunohistochemical stainings were quantified by digital image analysis, as described in detail elsewhere (17,18). All available fields (>30 fields) of the left ventricular were measured.

Quantitative real-time RT-PCR (ABI PRISM 7900 HT Sequence Detection System software, version 2.2.2.; Perkin Elmer) of 100 ng cDNA was used to quantify rat left ventricular ICAM-I, VCAM-I, interferon (IFN)-γ, Bax, and L32 cDNA levels. The cDNA expression levels of these genes were normalized to L32 cDNA. Conventional PCR products of rat ICAM-I, VCAM-I, IFN-γ, and L32 were obtained with the primers designed for real-time RT-PCR and were cloned into pGEM-T Easy vector (Promega) to be used as DNA standards. The sequences of the primer sets used in this study were as follows: for ICAM-1 5′-GTCTCATGCCCGTGAAATTATG-3′ and 5′-TTTTCTCCCAGGCATTCTCT-3′; for VCAM-1 5′-GGAGGTCTACTCATTCCCTGAAGA-3′ and 5′-ACCGTGCAG TTGACAGTGACA-3′; for IFN-γ 5′-GAATCGCACCTGATCACTAACTTCT-3′ and 5′-CCTCGAACTTGGCGATGCT-3′; and for L32 5′-AACCGAAAA GCCATCGTAGAAA-3′ and 5′-CCTGGC GTTGGGATTGG-3′.

Left ventricular samples were homogenized in lysis buffer containing proteinase and phosphatase inhibitors. An equal amount of protein (10–30 μg) was loaded into 10% SDS-PAGE. Total Akt, phosporylated Akt-Ser 473, Bax, Bcl-2, and tubulin (Cell Signaling Technology, Danvers, MA) were detected with each specific antibody. The blots were visualized with an enhanced chemiluminescene system (Amersham Bioscience, Buckinghamshire, U.K.). Quantitative analysis of the intensity the bands was performed with National Institutes of Health Image 1.63 software.

JMP statistical software 4.0.1 (SAS Institute, Cary, NC) was used to analyze significant differences between the groups. Data were compared conducting the Wilcoxon-Kruskal-Wallis analysis and the honestly significant difference post hoc analysis using Tukey-Kramer test (α < 0.05). Measurements were expressed as means ± SE. Bivariate correlations were analyzed using Spearman's correlation coefficient. A P value <0.05 was deemed statistically significant in all tests performed.

Body weight decreased in both STZ groups to a similar degree and in a significant manner when compared with the nondiabetic controls (STZICEI 223 ± 8 and STZ 225 ± 10 vs. control 383 ± 12 g, P < 0.05). Blood glucose levels were markedly increased in STZICEI and STZ (32 ± 2 vs. 31.3 ± 2 mmol/l, respectively, P = NS) yet found to be normal in controls (4.2 ± 0.3 mmol/l). The body weight–to–heart weight ratio remained unchanged after 6 weeks of diabetes in both STZ groups compared with controls (STZICEI 2.42 ± 0.08 and STZ 2.52 ± 0.1 vs. control 2.77 ± 0.08 g/mg, P < 0.05).

Most parameters for left ventricular function, derived from invasive pressure measurements as well as from echocardiography, were significantly impaired in the STZ group compared with those of controls: fractional shortening (−20%, P < 0.05), LVP (−49%, P < 0.05), dP/dt_max (−63%, P < 0.05), LVEDP (+100%, P < 0.05), and dP/dt_min (+60%, P < 0.05). ICEI treatment could improve systolic (LVP +40%, P < 0.05) and diastolic (LVEDP −33%; dP/dt_min +62%, both P < 0.05) dysfunction in the STZICEI group, and this compared significantly with those functions in the STZ group. dP/dt_max was even normalized when STZICEI was compared with controls. The left ventricular function data presented are not physiological data, since open chest measurement and the anesthesia protocol used are both known to be cardiodepressive in comparison with those of conscious animals (19). Echocardiography did not reveal any significant left ventricular dilatation, as shown by unchanged LV end diastolic diameter values found for all groups (Table 1).

The cardiac cytokine content correlated negatively with the cardiac contractility parameter dP/dt_max (TNF-α vs. dP/dt_max, r = −0.69; IL1-β vs. dP/dt_max, r = −0.48; IL-18 vs. dP/dt_max, r = −0.71, all P < 0.05).

Both the histochemically marked proteins ICAM-1 (200%, P < 0.05) and VCAM-1 (800%, P < 0.05) as well as their mRNA abundance (100 and 220%, respectively, both P < 0.05) were found to increase significantly in STZ compared with the controls. These changes were normalized in STZICEI, in contrast to those of the controls (Fig. 1).

Immunohistochemical staining of proinflammatory cytokines revealed a significant increase of IL1-β (240%, P < 0.05), IL-18 (740%, P < 0.05), and TNF-α (950%, P < 0.05) abundance in the cardiac tissue in STZ compared with those of the controls. This increase could be distinctly reduced by ICEI treatment for IL1-β (−40%, P < 0.05), IL-18 (−70%, P < 0.05), and TNF-α (−80%, P < 0.05) in the STZICEI rats compared with STZ. Nevertheless, none of the cytokines could be normalized after ICEI treatment. Immunohistological staining of cytokines demonstrated a typical focal pattern, which is not restricted to the cytoplasm of cells having decreased immunoreactivity toward the periphery. This is consistent with the local secretion of these factors by cytokine-expressing cells. IFN-γ mRNA was not specifically regulated under diabetes conditions or after ICEI treatment (Fig. 2).

STZ hearts demonstrated significantly increased macrophage (CD68⁺ 1,400%, P < 0.05) and leukocyte (CD18⁺ 900%, P < 0.05) infiltrates in comparison with the controls. This could be reduced upon ICEI treatment by 50% (CD68) and 60% (CD18) (both P < 0.05) when STZ and STZICEI were compared. Nevertheless, leukocyte and macrophage (+500 and +300%, both P < 0.05) infiltrates were still significantly increased when STZICEI was compared with the controls (Fig. 3).

Collagen type I and III were increased (both +300%, P < 0.05) in STZ, and both were statistically different compared with controls. These changes led to a significant increase (+200%, P < 0.05) in total collagen content. In contrast, collagen type I, type III, and total collagen content were normalized due to ICEI treatment compared with controls (Fig. 4).

Western blot measurements showed that STZ-induced diabetes downregulated the phosphorylated-to-total ratio (P < 0.05), indicating a decrease in the activation state. This downregulation was normalized due to ICEI treatment compared with controls (Fig. 5).

Diabetes was found to increase cardiac apoptosis, as indicated by the significantly decreased Bcl-2/tubulin protein levels (−29%, P < 0.05) and increased Bax/tubulin levels (+26%). After treatment with the ICEI, Bcl-2 and Bax levels were seen to be normalized, resulting in a significantly decreased BCL-2/Bax ratio (−39%, P < 0.05) (Fig. 5).

The salient finding of the present study is that pharmacological inhibition of ICE improves cardiac dysfunction in STZ-induced diabetic cardiomyopathy. This was accompanied by decreased cardiac inflammation, fibrosis, and apoptosis despite severe hyperglycemia in an animal model of type 1 diabetes.

Diabetic cardiomyopathy is associated with left ventricular dysfunction (3,4,7,20). We recently showed that increased cytokine levels in the cardiac tissue correlate with the degree of cardiac failure in experimental diabetes (6), a finding that may explain their multiple cardiodepressive effects (21). Additionally, cardiac fibrosis impairs cardiac function due to increased stiffness in the cardiac muscle and replacement of contractile muscle (5,22). This is in agreement with our findings concerning impaired systolic and diastolic function, as was shown in the current study by invasive LVP and echocardiographic data. We demonstrate an inverse correlation between cardiac contractility and the cardiac cytokines, suggesting a direct and causative contribution of cytokines in diabetic cardiomyopathy. Therefore, it is intriguing to speculate that a reduction of cardiac cytokines by anticytokine-based therapy might indeed be beneficial in treating STZ-induced diabetic cardiomyopathy.

Pralnacasan, an available oral pharmacological ICEI, is used in the field of “classical” inflammatory diseases, such as rheumatoid arthritis, osteoarthritis, and inflammatory bowel disease in humans as well as in animals models (11,23–26). ICE exists as an inactive precursor in the plasma membrane. It requires two cleavages before becoming enzymatically active. This activation can be induced by a variety of stimuli, including general cell activation after endothelial adhesion, ischemia, increased radical oxygen species, high glucose levels, and by proinflammatory cytokines like TNF-α (11,25,27–29), all known to be present in diabetic cardiomyopathy. The ICEI pralnacasan (HMR 3480) used in the current study prevents an ICE-mediated cleavage from the cytokines IL1-β and IL-18 to their biologically active proinflammatory forms (14).

IL1-β is known to be an important mediator in inflammatory diseases by stimulating lymphocytes as well as bone marrow cells, inducing the acute phase reaction, and acting as an endogenous pyrogen (30). Distinctly different from IL1-β, which may also be cleaved by other proteases (31), active IL-18 is cleaved solely by ICE (32). Pomenrantz et al. (25) showed a direct cardiodepressive effect of IL-18 after ischemia and reperfusion injury in human atrial myocardium. They conclude that not only the proinflammatory cytokines TNF-α and IL1-β, but also IL-18 are responsible for cardiac dysfunction and may therefore become a new target for therapeutic intervention in cardiac diseases. This is supported by findings in heart failure patients, where it has been shown that IL-18 content was higher in myocardial tissue than in age-matched healthy control subjects (33). Coherent with these findings, our data suggest a role of IL-18 in diabetic cardiomyopathy. Many of its effects are known to be IFN-γ dependent, and, therefore, it was suggested that IL-18 is a mediator mainly of TH1-response inflammatory diseases (32). Further studies will have to focus on the role of IL-18 in order to clarify its direct effects in cardiac failure.

In the present study, IL1-β and IL-18 expression were reduced after treatment with ICEI. Furthermore, ICE inhibition decreased abundance of immunocompetent leukocytes, macrophages, the cytokine TNF-α, as well as the adhesion molecules ICAM-1 and VCAM-1. Although it is known that, for example, cardiomyocytes are also involved in the production of cytokines (34), it is intriguing to speculate that reduced levels of cardiac IL1-β and IL-18 (due to ICE inhibition) may have reduced cardiac inflammation by decreasing endothelial CAMs abundance. This is accompanied by a decreased number of immunocompetent cells invading into the cardiac tissue, a finding also supported by Wyman et al. (35), who showed that IL-18 acts as a priming agent for immunocompetent cells in human sepsis.

The anti-inflammatory properties of ICE treatment influenced intracellular pathways as well. In agreement with others (36), we found a significant reduction of the activated phosphorylation state of Akt in diabetic animals. This was normalized due to ICEI treatment. Akt is known to regulate many survival pathways of the cardiac cell (9). Alongside other pathways, this regulates cardiac contractility and also cardiac apoptosis (10). In line with these findings and in agreement with Syed et al. (37), who showed increased apoptosis after myocardial ischemia in mice overexpressing cardiac ICE, we demonstrated that reduced cardiac apoptosis could be indicated by normalized Bcl-2/Bax levels due to ICEI treatment. Furthermore, ICE inhibition (38) and/or ICE gene deficiency (39) itself reduced apoptosis in other animal models, suggesting a direct effect of ICE on cardiac apoptosis as well.

Diabetes is characterized by increased extracellular matrix deposition of collagen in the interstitium of the diabetic hearts (40,41). One of the many modulating parameters for fibrosis is increased cardiac proinflammatory cytokine content (42). Coherently, we showed that increased collagen type I, type III, and total collagen levels in the cardiac tissue of the STZ group were all reduced successive to ICEI treatment. This was associated with an improvement of cardiac compliance, as indicated by a reduction of LVEDP.

In conclusion, ICE inhibition reduces the development of STZ-induced diabetic cardiomyopathy. Further studies will have to investigate whether this form of anticytokine-based pharmacological intervention improves the common treatment concept and whether it can be passed on into the human pathophysiology of diabetes.

This study was supported by the Deutsche Forschungsgesellschaft (SFB-TR-19; Z3).

Pralnacasan was a kind gift of Aventis-Sanofi (Frankfurt, Germany).

We thank M. Kastner for proofreading the manuscript.