Transcriptional Control of Axon Growth Ability

of a dissertation at the University of Miami. (May 2010) Dissertation supervised by Professor Jeffrey Goldberg. No. of pages in text. (167) Mammalian central nervous system (CNS) neurons lose their ability to regenerate their axons after injury during development. For example, optic nerve injury studies in hamsters have shown that optic nerve axons injured around the time of birth retain the ability to regenerate to their target, but this ability is lost during development (So et al., 1981). The development of an inhibitory CNS environment has been implicated in the inability of the adult CNS to regenerate, however there is also support for this loss being a result of changes in developmental programs intrinsic to the neurons themselves (Goldberg et al., 2002a; Goldberg, 2004). While some molecules have been identified as being involved in intrinsic mechanisms controlling axon growth, there is still much to be discovered. Using genes shown to be regulated in retinal ganglion cells (RGCs) during development (Wang et al., 2007), I performed an overexpression screen in embryonic primary neurons measuring changes in neurite growth. Of these genes, the most significant effect in neurite growth was seen with overexpression of Krüppel-like factor 4 (KLF4), resulting in a greater than 50% decrease in growth. KLF4 is a member of the KLF family of transcription factors which all possess a DNA binding domain containing 3 zinc finger motifs. Outside of the nervous system, KLF4 has been implicated in cancer (Black et al., 2001; Rowland and Peeper, 2006), mitotic growth arrest (Shields et al., 1996) and most recently in the induction of pluripotency (Yamanaka, 2008; Zhao and Daley, 2008). In the CNS, KLF4 has recently been implicated in increasing the sensitivity of cortical neurons to NMDA insult (Zhu et al, 2009), though no effect of KLF4 on neurite growth or regeneration has yet been described. I found that KLF4 overexpression in RGCs results in decreased neurite growth and neurite initiation. KLF4 overexpression also leads to decreases in polarity acquisition in hippocampal neurons, though even when they acquire polarity, they still display decreased neurite growth. Additionally, KLF4 knockout targeted to RGCs leads to an increased neurite growth ability and increased neurite initiation in vitro. In vivo, KLF4 knockout increases RGC axon regeneration after optic nerve injury. Interestingly, KLF4 is one of 17 members of the KLF family, known for their ability to act redundantly and competitively amongst family members for their binding sites. Therefore, we looked to see if other KLFs could affect neurite growth ability. 15 of 17 KLF family members are expressed in RGCs, and their overexpression results in differential effects on neurite growth in both cortical neurons and RGCs. Additionally, many of the family members are developmentally regulated in a manner that typically correlates with their ability to affect neurite growth. For example, KLF6 and -7, whose expression decreases during development, when overexpressed, increase neurite growth, whereas KLF9, whose expression increases developmentally, when overexpressed, decreases neurite growth. Surprisingly, there are multiple KLFs expressed in RGCs that are neurite growth-suppressors, and further study has revealed that the combination of KLF growth enhancers with KLF growth suppressors results in a suppressive or neutral phenotype (Moore et al., 2009), suggesting that to further enhance regeneration after injury in vivo, we will need to additionally remove the growth suppression from other KLF family members. Taken together, these data suggest that KLFs may play an important role in the intrinsic loss of axon growth and regeneration seen during development. Further characterization of downstream targets of KLF4 and other KLF family members may reveal specific neuronal gene targets that could mediate the phenotypic effects of these transcription factors. It is my hope that by determining the developmental programs that underlie the loss of intrinsic axon growth ability of CNS neurons, we may ultimately determine how to revert adult CNS neurons to their embryonic axon growth ability.