Size-Exclusion Chromatography of Circulating Cardiac Troponin T

In a recent issue of Clinical Chemistry, Fahie-Wilson et al. (1) reported their use of gel-filtration chromatography and size-exclusion chromatography (SEC), combined with off-line measurement, to investigate circulating forms of cardiac troponin T (cTnT). We believe that the presented data did not justify the main conclusion of the paper, that cTnT circulates predominantly in the free, intact form in serum of patients with kidney failure and patients with acute coronary syndrome. Behavior of proteins in SEC is determined mainly by their hydrodynamic volume (expressed as Stokes radius, Rs). Nonglobular proteins are known to have a larger Rs than globular proteins with a comparable relative molecular mass (Mr), and cTnT is far from globular (2). Therefore, as is generally agreed, the separation of the nonglobular cTnT protein in an SEC column is better described with the Rs than the Mw(3). Hence, we used Rs to analyze the data from Fahie-Wilson et al.(1). We then calculated the Kav for the 2 cTnT reactive peaks from Fig. 2D of the Fahie-Wilson article (peak 1, Ve = 30.5 so Kav = 0.09; and for peak 2, Ve = 37.0 so Kav = 0.22). When these Kav were entered into equation 1, Stokes radii of 37 Å and 26 Å were obtained. In addition, we used equation 1 to calculate the highest Rs (Kav = 0) that could be separated with the column (Rs,max = 48Å). We believe these Rs do not correspond to the troponin complex and free cTnT. The Rs of the troponin complex has been determined to be 57Å (5). Hence, we may conclude that the first peak cannot correspond to the troponin complex, because this would have eluted at V0. Unfortunately, the Rs of cTnT has never been determined. It is plausible, however, that the first peak corresponds to intact free cTnT. Based on the Rs of bovine TnC (Mr, ∼18 400) and cTnI (Mr, ∼23 900) (24Å and 30Å, respectively (6), it is likely that the larger cTnT (Mr, ∼34 500) has an Rs of 37Å. The online SWISS-PROT database indicated 85%, 89%, and 99% homology between bovine and human cardiac troponin T, I, and C, respectively. The second peak would then correspond to an immunoreactive fragment. Interestingly, a small immunoreactive fraction just before 30 mL is visible in Fig. 2C and corresponds to the first peak from Fig. 2D. This peak could indicate a small amount of the intact protein present in acute coronary syndrome patients. Importantly, because no secondary confirmation was provided (e.g., sodium dodecyl sulfate-polyacrylamide gel electrophoresis or amino acid sequencing), the exact identity of the peaks remains unclear. In studies such as this, it is essential to use a fully characterized troponin TIC complex standard. Usually, it can be expected that the claims from suppliers are correct and thoroughly checked. However, when we contacted the supplier (SCIPAC), they were not able to provide data showing both intact free cTnT and TIC complex in their standard. In summary, the conclusions drawn from fundamentally incorrect interpretations of SEC data, without any secondary confirmation, and the use of a troponin complex standard that is not properly characterized may have been misleading to readers. It is more plausible that the correct conclusion from the study by Fahie-Wilson et al. (1) is that no intact free cTnT is present in patients with kidney failure, but that a smaller immunoreactive cTnT fragment is present.