Background: Diabetic cardiomyopathy is recognised as an independent risk factor for heart failure progression. A key feature of diabetic cardiomyopathy is increased left ventricular stiffness and an inability for the heart to relax sufficiently. In addition to increased collagen deposition (fibrosis), advanced glycation end-product (AGE) crosslinking of the extracellular collagen network has been shown to be a key feature of diabetic myocardial stiffness. Considerably less is known about the intracellular origins of diabetic diastolic dysfunction, and AGE modification of regulatory and contractile proteins may play an important role. Troponin is an important dynamic protein required for Ca2+ regulation of myofilament function. Disturbances in myofilament Ca2+ sensitivity are a common observation in diabetic cardiomyocytes, and glycation of troponin may contribute to diastolic dysfunction in diabetes. The aim of this study was to determine whether the cardiac troponin complex (cTn), the primary regulator of cardiomyocyte contractile thin filament activation, is susceptible to AGE modification in vivo and in vitro.
Methods: Tandem mass spectrometric analysis (MS/MS) of tryptic peptides digested from purified human cTnC and cardiac troponin I (cTnI) (purchased commercially) and cTnI and cTnT from rodent ventricular heart homogenate (extracted from SDS-PAGE gel), was performed to identify amino acid sites of AGE (N ε-carboxymethyl-lysine (CML), N ε-carboxyethyl-lysine (CEL) and methylglyoxal-derived hydroimidazolone 1 (MG-H1) modification in vivo. Purified human cardiac troponin C (cTnC) was incubated in 2M glucose or 2M fructose to identify amino acid residues susceptible to hexose modification (the first step of AGE formation). MS/MS analysis of intact protein mass was performed to identify +162amu and +16amu cTnC mass shifts corresponding to hexose and oxidation modification respectively.
Results: MS/MS analysis of trypsin-digested human purified cTnC and cTnI revealed that CML modification was evident on cTnC at lys158 and on cTnI at lys121, 132, and 165. MG-H1 modification of cTnI at arg205 was also evident. Rodent cTnC, cTnI and cTnT was extracted from SDS-PAGE gel homogenate separation, trypsin-digested and analysis by MS/MS revealed that CML modification of cTnC was evident on lys21, cTnI on lys59 and 132, and cTnT on lys202. To compare the efficacy of glucose and fructose for hexose attachment, human purified cTnC was incubated in 2M glucose and fructose in vitro. cTnC exhibited 5.3±0.2 and 2.2±0.2 hexose attachments with 7 days of glucose and fructose incubation respectively. In 18 replicate samples, lys21 was most frequently observed to be modified by glucose, with 80% of samples showing attachment at this residue. All 5 replicate fructose-incubated samples exhibited fructose attachment on lys21 and 39. Other lysine residues also exhibited glucose and fructose attachment, but with lower frequency of detection. Oxidation was evident on all methionine residues of cTnC in >80% of fructose-incubated replicate samples. Glucose-induced oxidation of cTnC was less marked, with many methionine residues oxidized in only ∼10% of replicate samples.
Conclusions: This study provides the first evidence that the cTn complex is susceptible to AGE modification in vivo and in vitro. AGE modifications were located in functional domains of the cTn complex and may play a key role in altering regulation of contractile myofilament relaxation. In vitro cTnC studies identified lysine residues which are particularly susceptible to the early stages of AGE formation and demonstrate differential induction of hexose vs oxidation modification by glucose and fructose. These findings provide important information about glycation of the troponin complex and investigation of troponin glycation in diabetic cardiomyocytes is now warranted.