STUDIES ON THE MECHANISMS OF LIGAND-INDUCED NUCLEAR LOCALIZATION OF CELLULAR RETINOIC ACID BINDING PROTEIN TYPE II

Abstract

Retinoic acid (RA), a derivative of vitamin A, is well know for its ability to act as a transcriptional modulator, and it is critically involved in a number of cellular processes. The cellular retinoic acid binding proteins (CRABP-I and II) are small cytosolic proteins that bind to RA with a high affinity, and they are thought to play a role in the biological activities of RA. The expression of CRABP-II has been shown to selectively enhance the activity of a group of ligand-inducible transcription factors known as the retinoic acid receptors (RARs), and CRABP-II has also been shown to accumulate in the nucleus in response to RA treatment. CRABP-II is thought to enhance RAR activity by enhancing the movement of RA into the nucleus, where RAR functions as a transcription factor. The question addressed in this work was that of the mechanism by which CRABP-II accumulates in the nucleus in response to RA. Abolishing nuclear localization of CRABP-II, and the resulting effects on RAR activity were addressed. CRABP-II was tagged to a nuclear export signal (NES), which actively excludes a protein from the nucleus. Subsequently, the NES functionality was verified by fluorescence microscopy. Functional assays for RAR activity were performed, and it was determined that accumulation of CRABP-II in the nucleus was necessary for it to enhance the activity of RAR. To address possible mechanisms for the nuclear localization of CRABP-II, analysis of its primary and tertiary structures were performed. No nuclear localization signals could be identified in the primary sequence of CRABP-II. Comparisons of the electrostatic surface potentials of apo- and holo- crystal structures of CRABP-II identified a region of the protein that appeared to become more basic in response to RA binding. Three basic residues in the region were identified as potentially important for the change in the surface potential. Further analysis identified the residues as being very similar in orientation to a classical primary nuclear localization signal. The three basic residues of CRABP-II were converted to alanines by mutagenesis, and the recombinant mutant was purified from E. Coli. Using the purified WT and mutant proteins, it was determined that mutation of the residues did not significantly change the proteins? sensitivity to urea-induced denaturation, and that the mutations did not significantly disturb the ability of CRABP-II to bind RA. The three basic residues were then mutated in mammalian expression vectors, and the effects of mutating them were evaluated by microscopy. It was determined that the three basic residues are required for RA induced nuclear localization of CRABP-II. Further, it was determined that Leptomycin-B (LMB), an inhibitor of nuclear export, had no effect on the RA-mediated localization of the WT or the mutant protein, suggesting that CRABP-II nuclear localization is mediated by an import machinery. Functional assays for RAR activity were used to address the functional significance of the three basic residues. It was determined that the mutant protein was unable to enhance the activity or RAR to the same degree as WT, and further that the enhancement by the mutant was similar to that obtained using the NES-tagged WT protein. Further research was done to identify the proteins involved with the RA induced nuclear import of CRABP-II. Recombinant CRABP-II-GST was shown to interact with importin?, a known nuclear import-related protein, in an RA dependent fashion. The NLS CRABP-II mutant did not interact with importin? even in the presence of RA, further supporting the hypothesis that RA induced nuclear localization requires the three NLS residues. Overall, this work suggests a unique mechanism for nuclear import, in which a non-classical NLS emerges in CRABP-II in response to RA binding.