Abstract
While lizards and salamanders both exhibit the ability to regenerate amputated tails, the outcomes achieved by each are markedly different. Salamanders, such as Ambystoma mexicanum, regenerate nearly identical copies of original tails. Regenerated lizard tails, however, exhibit important morphological differences compared with originals. Some of these differences concern dorsoventral patterning of regenerated skeletal and spinal cord tissues; regenerated salamander tail tissues exhibit dorsoventral patterning, while regrown lizard tissues do not. Additionally, regenerated lizard tails lack characteristically roof plate-associated structures, such as dorsal root ganglia. We hypothesized that differences in neural stem cells (NSCs) found in the ependyma of regenerated spinal cords account for these divergent regenerative outcomes. Through a combination of immunofluorescent staining, RT-PCR, hedgehog regulation, and transcriptome analysis, we analyzed NSC-dependent tail regeneration. Both salamander and lizard Sox2(+) NSCs form neurospheres in culture. While salamander neurospheres exhibit default roof plate identity, lizard neurospheres exhibit default floor plate. Hedgehog signaling regulates dorsalization/ventralization of salamander, but not lizard, NSCs. Examination of NSC differentiation potential in vitro showed that salamander NSCs are capable of neural differentiation into multiple lineages, whereas lizard NSCs are not, which was confirmed by in vivo spinal cord transplantations. Finally, salamander NSCs xenogeneically transplanted into regenerating lizard tail spinal cords were influenced by native lizard NSC hedgehog signals, which favored salamander NSC floor plate differentiation. These findings suggest that NSCs in regenerated lizard and salamander spinal cords are distinct cell populations, and these differences contribute to the vastly different outcomes observed in tail regeneration.