Toll-like receptor ligands differentially affect cardiomyocyte function

Elizabeth Belcher, Sian E. Harding, Trupti A. Patel, Timothy W. Evans and Jane A. Mitchell. Myocardial Systems Biology, Cardiothoracic Pharmacology and Critical Care, NHLI, Imperial College, Dovehouse Street, London, SW3 6LY, UK.

Infection is now a recognised risk factor for cardiovascular deaths. Pathogen associated molecular patterns (PAMPs) are sensed in the immune system by pathogen response receptors (PRRs), including Toll-like receptors (TLRs). TLR4 mediates responses to most Gram-negative bacterial LPSs whilst TLR2 mediates responses to Gram-positive lipoproteins in macrophages and blood vessels (Akira et al., 2001; Jimenez, et al., 2005). TLR2 forms a heterophilic dimer with TLR1 or TLR6 to discriminate between diacyl or triacyl lipopeptides. TLR3 recognises viral dsRNA, whilst TLR9 recognises bacterial CpG DNA; both receptors being intracellular (Takeda et al., 2005). Gram-positive bacteria directly reduce the number of contracting cardiomyocytes (Patel et al., this meeting), but the specific role of individual TLRs is unknown. The aim of this study was to determine the role of TLR signalling in this response.

Ventricular myocytes were isolated from adult male Sprague Dawley rats (250-350 g) and cultured for 24 h or 48 h with S. aureus (108 cfu/mL), FSL-1 (TLR2/6; 300 μg/mL), PAM 3CSK4 (TLR2/1; 300 μg/mL), LPS (TLR4; 1 μg/mL), PolyI:C (TLR3; 25 μg/mL), CpG control (2 μg/mL), CpG (TLR9; 2 μg/mL) or co-stimulated with FSL-1 (300 μg/mL) and PAM3CSK4 (300 μg/mL). Cardiomyocytes contracted upon electrical stimulation (0.5 Hz, 1 mM Ca2+) and proportions of contracting, non-contracting and arrhythmic cells were counted. A contraction was detectable when shortening with each beat was ≥ 0.25%. Analysis of data was by one-way ANOVA, followed by Dunnett’s post-test.

S. aureus , FSL-1 or PAM 3CSK4 had no significant effect on the proportion of contracting myocytes at 24 h. (Control 68.8 ± 3.6%; S. aureus 58.4 ± 10.3%; FSL-1 59.4 ± 12.1%; PAM3CSK 62.6 ± 6.7%; n=5-6). Similarly, there was no effect of LPS (Control 88.9 ± 1.5%; LPS 81.3 ± 4.1%; n=6), CpG ( non-CpG oligodinucleotide 73.0 ± 4.6%; CpG 62.4 ± 9.7%; n=6) or PolyI:C (Control 75.6 ± 3.7%; PolyI:C 76.9 ± 3.2%; n=5). Co-stimulation with FSL-1 and PAM3CSK caused a significant reduction in the number of contracting myocytes at 24 h (Control 75.6 ± 3.7%; FSL-1+ PAM 3CSK 33.9 ± 11.2%; n=7, p<0.01). At 48 h S. aureus, FSL-1, but not PAM3CSK, reduced the proportion of myocytes that had contractile activity compared to control (Control 70.9 ± 4.7%; S. aureus 20.8 ± 8.6%, p<0.01; FSL-1 30.8 ± 10.9%; PAM3CSK 54.5 ± 8.8%; n=5-6, p<0.01). At 48 h co-stimulation with FSL-1 and PAM3CSK caused a reduction in the number of contracting myocytes (Control 52.1 ± 4.3%; FSL-1+ PAM3CSK 15.1 ± 3.5%; n=5, p<0.01). LPS, CpG or PolyI:C had no effect at 48 h.

These findings demonstrate that PAMPs differentially modulate cardiomyocyte function and that the effects of Gram-positive bacteria on these cells are mediated by the TLR2/6 heterodimer. This effect is enhanced by the TLR2/1 interaction. These data have implications for our understanding of the role of infection in mediating myocardial dysfunction.

Akira et al., (2001) Nat Immunol 2:675;
Jimenez et al., (2005) Proc Natl Acad Sci 102(12):4637-42;
Takeda et al. (2005) Int Immunol 17(1):1-14. Funded by MRC/BHF.