the cytoplasm, while WT and S1027D were less well-merged. Furthermore, the number of puncta seen in the cytoplasm was also more than that of WT or S1027D. These results also support the idea that dephosphorylation of theSer1027 residue in ULK2 enhances its autophagic activity by increased association with LC3-II. Confocal microscopy of ULK2 WT, S1027A, and S1027D with endogenous Kap2 in HEK293 cells also supports our notion that ULK2 S1027D binds more with Kap2 and more is transported to the nucleus, due to PubMed ID:http://www.ncbi.nlm.nih.gov/pubmed/19666584 the fact that this 
protein can expose more of its PY-NLS motif to Kap2 because of its weak association with ATG13 or FIP200. In contrast to this, ULK2 WT or S1027A binds to Kap2 to a lesser extent because of the masking of the PY-NLS motif through the binding of Atg13 or FIP2000 and less is transported to the nucleus. Without PKA phosphorylation, PY-NLS of ULK2 seems to be masked through the steric inhibition through its strong association with ATG13 or FIP200. The Ser462 residue of ULK2 also matches well with the consensus sequence information, and it is possible for it to be phosphorylated by PKA . We constructed S462A and S462D mutants, transfected HEK293 cells, the performed PubMed ID:http://www.ncbi.nlm.nih.gov/pubmed/19667315 the same experiment as with S1027A and S1027D. However, we did not observe any change in the subcellular localization of the S462A or S462D mutant or in their protein-protein interaction with Atg13 or FIP200 in the western blot pattern with a PKA phosphorylation specific antibody, when compared with the S1027A or S1027D mutant. The apoptotic ability of the ULK2 S462A and S462D mutant was not changed either, compared with that of the ULK2 WT. Thus, these results suggest that the phosphorylation of ULK2 Ser1027 by PKA is the specific regulatory site for its interaction with ATG13 or FIP200 and for its subcellular localization. In order to obtain more evidence that the subcellular localization of ULK2 is controlled by PKA phosphorylation, the endogenous ULK2 localization in HEK293 cells was observed using confocal microscopy with the treatment of a PKA activator or inhibitor . Compared with the untreated cells, the cytoplasmic localization of 15 / 22 PY-NLS Motif and Ser1027 Residue Phosphorylation of ULK2 ULK2 with H89 treatment was not INK1117 dramatically changed. However, an increased nuclear localization of ULK2 with the treatment of FSK was observed, suggesting that the subcellular localization of ULK2 is changed by PKA phosphorylation. In addition, ULK2 was associated more with Kap2 in FSK-treated cells than in normal or H89-treated cells, consistent with Fig 6K6M. PCC between ULK2 and Kap2 was measured as shown in Fig 6N, Fig 6O, and Fig 6P using quantitative confocal microscopy. These results also support the notion that PKA phosphorylation enhances the association of ULK2 and Kap2. However, it is not clear at present the reason why the PKA inhibitor is less effective than FSK in the induction of a change to a nuclear localization of ULK2. The nuclear or cytoplasmic fluorescence intensity profile also confirm that the PKA-phosphorylated ULK2 shows an increased localization to the nucleus than to the cytoplasm. Using quantitative confocal microscopy, the Fn/t ratio of normal DMEM-grown HEK293 cells, H89-treated cells, or FSK-treated cells was also determined. Consistent with the PCC and ULK2 S1027A/D mutant results, the activation of PKA phosphorylation by FSK promotes ULK2 nuclear transport by an increase in ULK2 and Kap2 interaction. However, the inhibition o