In the research article, “PERSISTENCE LENGTH OF HUMAN CARDIAC α-TROPOMYOSIN MEASURED BY SINGLE MOLECULE DIRECT PROBE MICROSCOPY,” Loong et al. (PLoS One 2012;7(6):e39676) report that the apparent persistence length of wild-type tropomyosin is about 45 nm, which is an unusually low value for the persistence length of a coiled-coil.
Loong et al. refer to our work (Li et al., J. Mol. Biol., 2010, 395:327-339), where more typical values were found, stating: “A more recent EM study measured the average Lp from 16 molecules of bovine cardiac αTm to be ~102 nm, presumably at room temperature,” implying later in their paper that our work might be inaccurate by additionally stating, “We also note a general overestimation of Lp when fewer than ~100 molecules are included in data analysis, which suggests a large data set is necessary for reliable estimates of Lp using similar techniques.”
While Loong et al. mention another paper of ours (Sousa et al., Biophys. J. 2010 99:862-868, their reference [45]), they fail to note that in the Sousa study a large sample size of >500 tropomyosin molecules was used, which resulted in the same ~100 nm apparent persistence length value as in our prior sampling. In fact in a newly published paper of ours (Li et al., Biochem. Biophys. Res. Commun., 2012, in press at the time of this posting), the apparent persistence length on yet another sample of >200 wild-type tropomyosin molecules was measured with the same outcome of ~100 nm for the apparent persistence length. Moreover, Loong et al. do not adequately consider our extensive Molecular Dynamics on atomic models of tropomyosin (Li et al., J. Mol. Biol., 2010, 395:327-339), where again a ~100 nm apparent persistence length was found for the tropomyosin molecule. Here, sample size could not be an issue, since about 30,000 snapshots of tropomyosin were recorded and analyzed.
Therefore, inadequate sampling cannot explain the large difference in apparent persistence length values reported by us and by Loong et al. We suggest a number of reasons that may account for the low apparent persistence length computed by Loong et al. for coiled-coil tropomyosin:
1. Loong et al. used AFM to image tropomyosin molecules which had been adsorbed onto poly-lysine treated mica, which thus presented a large density of positive charges on the substrate surface. Hence, strong attractive forces will draw the negatively-charged tropomyosin (approximately -54e at neutral pH) to the positively-charged mica substrate. The influence of this interaction on tropomyosin tertiary structure is unknown. However, there is good reason to expect that tropomyosin structure will be perturbed during the adsorption process and consequently corresponding persistence length results will be affected, since there is precedence for macromolecular distortion of proteins occurring on mica substrates: For example, adsorption of smooth muscle myosin onto mica for AFM analysis was found to prevent normally occurring myosin head-head interactions, as the heads appear to become more attracted to the mica than to each other (see “Cryo-atomic force microscopy of unphosphorylated and thiophosphorylated single smooth muscle myosin molecules.”, Sheng et al., J. Biol. Chem. 2003, 278:39892-39896). Studies by Muecke et al. (J. Mol. Biol. 2004, 335:1241-1250) showed that substrate-protein interactions can indeed affect persistence length. Using AFM they found that the apparent persistence length of intermediate filaments is lower on a mica substrate than it is on glass or on carbon films used for EM. Muecke et al. explain that “for an interaction energy in the range of the thermal energy, the filaments will freely equilibrate on the support before they become adsorbed. In this case, the elastic properties are conserved during the adsorption process, so that the measured persistence length is equal to that of a filament equilibrated in a dilute solution.” (In contrast) “For an interaction energy bigger than the thermal energy, the filaments will be “caught” by the support before having equilibrated so that the filaments are “fixed” into a contour resembling a normal (i.e. perpendicular) projection of the actual three-dimensional contour onto the support. Such a “capture” mechanism yields more condensed filaments on the support and hence a smaller apparent persistence length is revealed." In the case of experiments by Loong et al., it is thus likely that the tropomyosin was captured by the poly-lysine coated mica, and hence prevented from relaxing freely in the local 2D environment above the support. Consequently, the apparent persistence value reported for tropomyosin by Loong et al. may not accurately reflect the persistence length value of interest, namely that of a freely fluctuating molecule.
2. The AFM images of tropomyosin in the Loong et al. article (Figs. 1A and 1B) reveal a mixture of some relatively straight and some slightly curved molecules. For tropomyosin to display a persistence length of ~45 nm on a two-dimensional surface as claimed, the end-to-end bending of the 42 nm long tropomyosin should be ~50 degrees on average according to the relationship cos theta = exp (-s/2Lp), where s = 42 nm. Given the presence of many nearly straight molecules depicted, other molecules should be bent much more than 50 degrees, in order to yield an average of 50 degrees. Such large end-to-end angles cannot be seen in these figures. This raises the question how such short apparent Lp could have been computed by Loong et al.? A possible explanation is that over-sampling during the skeletonizing along the long axis of the images may have introduced local bends that are not justified given the low resolution of the images. For example, the kinks seen at the ends of the skeletons in Fig. 2H are likely to have been exaggerated, given the poor resolution of the original image in Fig. 2B. The low resolution level of the images can easily be recognized by the exaggerated thickness (in relation to the length) of the tropomyosin images (the tropomyosin molecule is ~20 times narrower than it is long, which is not what is seen in the case of these images). The artificial introduction of local bends in the skeletons of the images artificially reduces the resulting apparent persistence length.
We therefore conclude that the present AFM measurement of the apparent persistent length of tropomyosin molecules led to an under-estimation of the biologically relevant value for the isolated molecule. Moreover, were tropomyosin to have a persistence length as short as approximately one times its own length, as claimed by Loong et al., then the isolated molecule would oscillate end-to-end by about 70 degrees on average (see Li et al. 2010, J. Struct. Biol., 107:313-318) and thus would be of insufficient stiffness to cooperatively regulate thin filament activation of myosin ATPase over distances of about one to two times its own length.
William Lehman
Xiaochuan (Edward) Li
Stefan Fischer