Transverse tubules (t-tubules) are largely absent in neonatal cells and develop rapidly after birth. In adult ventricular cells, a mobile Ca2+ buffer calmodulin exists predominantly near t-tubules in the cell. Ca2+ influx via the L-type Ca2+ channel (ICaL) triggers Ca2+ release from the sarcoplasmic reticulum (SR) via the ryanodine receptor (RyR), and Ca2+ flow in a restricted subspace located between the junctional SR and t-tubules. On the other hand, t-tubule network is largely absent, SR is relatively sparse, and ICaL expression is diminished in neonatal ventricular cells; as such, Ca2+ transient in neonates occurs predominantly at the periphery of the cell. In addition, the maximal binding capacity of Ca2+ buffer is much lower than that of Ca2+ buffer in adult ventricular myocytes. Here, we utilized human ventricular cell model (tenTusscher et al., 2006) and guinea pig ventricular cell model (Kuzumoto et al., 2008) to assess the differences between neonatal and adult ventricular cells in terms of intracellular Ca2+ dynamics and structure of t-tubules. In our previous study (Itoh et al., 2007), we showed that a developmental change in action potentials (APs) can be well reproduced with common sets of mathematical equations and by changing relative densities of ion currents on the Kyoto model, a guinea pig ventricular cell model. In this study, we first modified the human ventricular cell model by implementing the relative densities to represent developing ventricular cells at early and late embryonic stages. We then removed t-tubules and accompanying subspaces from the model to represent ventricular cells without t-tubules. As a result, removal of the subspace structure increased the amplitude of Ca2+ transient by four-fold and prolonged the basic cycle length by 800ms in early embryonic ventricular cell model. On the other hand, the amplitude of Ca2+ transient was decreased by 30 % in late embryonic ventricular cell model without subspace, which indicates that existence of the subspace structures affect the amplitude of Ca2+ transient in both early and late embryonic ventricular cell models. The Kyoto model was then expanded to represent developmental changes in the t-tubule structures by implementing the subspace structure to the model. We also considered predominant localization of the ICaL and INaCa (sodium-calcium exchange current) on t-tubules and assessed the effect of Ca2+ release via the INaCa to intracellular Ca2+ dynamics. We simulated APs and dynamic behaviors of currents with the Kyoto model considering the developmental changes of t-tubules structures in four different stages. As a result, the effects of the developmental changes in t-tubule structures and intracellular subspace at four different stages were simulated in terms of the excitation-contraction coupling.