We agree that the discrepancy between our results on Xenopus cardiac MHCα expression reported in Afouda et al. (Afouda et al., 2008) and those of Samuel and Latinkic based on their Xenopus MHCα expression experiments (Samuel and Latinkic, 2010) might indeed be due to technical details. Our analysis is based on the full-length Xenopus MHCα sequence [accession no. AY913767 (Garriock et al., 2005)], using the PCR primers developed by Professor Asashima's research group (see Ariizumi et al., 2003): forward, 5′-GCCAACTCAAACCTCTCCAAGTTCCG-3′ and reverse, 5′-GGTCACGTTTTATTTGATGCTGATTAACAGG-3′, with an annealing temperature of 55°C, an expected product of 230 bp, and we routinely use 23 cycles. We do not know exactly which sequence Samuel and Latinkic refer to in their Correspondence (Samuel and Latinkic, 2010), but the partial sequence [accession no. S74057 (Logan and Mohun, 1993)] that this group has referred to in previous publications (e.g. Latinkic et al., 2003) might not be the same as the MHCα sequence we used. Garriock and colleagues (Garriock et al., 2005) have used genomic analysis to show that their sequence represents the Xenopus orthologue of mammalian MHCα and expression studies to argue that this Xenopus gene has a role in the myocardium similar to that of mammalian MHCα.
However, Samuel and Latinkic's issue is really with the early onset of Xenopus MHCα expression during development. Indeed, as they would expect, we find that other cardiac muscle differentiation markers (such as cTnI and MLC2) are only expressed after about stage 28. We do, however, regularly detect Xenopus cardiac MHCα expression long before this stage, as do others. Small and colleagues (Small et al., 2005) report expression of their Xenopus cardiac MHCα gene at stage 12.5 (their Fig. 4A), and Cox and Neff (Cox and Neff, 1995) reported expression of Xenopus cardiac MHC at stage 13. As Samuel and Latinkic point out, these findings do raise the issue of whether MHCα represents a specific marker for cardiomyocyte differentiation. For this reason, we used MHCα as an indirect target gene control in the figure that Samuel and Latinkic refer to [Fig. 3 in Afouda et al. (Afouda et al., 2008)], and not as a cardiomyocyte differentiation marker.
Samuel and Latinkic also argue that particular animal cap explants in which cardiomyogenesis is induced with an inducible GATA4 construct recapitulate cardiomyo genesis with the same developmental timing as normal cardiomyocyte differentiation in whole embryos. We have no firm evidence to suggest that this is not the case. However, some overexpression experiments in animal cap explants do show an accelerated cardiomyogenesis differentiation programme. For instance, myocardin overexpression results in precocious expression of cTnI at stage 12.5 (Small et al., 2005). In unpublished results with GATA4-induced cardiogenesis in animal cap explants, we have been able to detect TBX5 at control stage 12.5 and MLC2 at stage 24 by RT-PCR, which appears to us to be slightly earlier than would be expected of these marker genes. We would therefore not completely rule out the possibility that GATA4 or GATA6 overexpression-driven cardiomyogenesis can be slightly accelerated, or delayed, relative to whole-embryo controls, depending on the particular experimental conditions (expression levels, stage of induction, etc.). However, this would not detract from the proven validity of GATA4-induced animal cap explants as a powerful model for vertebrate cardiogenesis.
