KIS colocalizes and associates with β-actin mRNA in cortical neurons. (A) Cortical neurons expressing HA-KIS were analyzed by FISH with a β-actin mRNA probe and subsequently processed for immunofluorescence with an anti-HA antibody. Bar, 5 μm. Profiles of HA-KIS (green) and β-actin mRNA (red) fluorescence values are shown for the indicated tracing. (B) Input extracts and αFLAG IP from cortical neurons infected with a FLAG-KIS lentiviral construct or the empty vector were analyzed by Western blotting. (C) IP shown in panel B were used to determine β-actin/GAPDH mRNA ratios by real-time RT-PCR. The mean values of six measurements from three independent experiments and the respective confidence limits (α = 0.05) were plotted.

KIS overexpression stimulates myr-dGFP^Actb-3′^UTR expression levels in neurites. (A) In vivo fluorescence image of a representative cell cotransfected with myr-dGFP^Actb-3′^UTR and DsRed. Bar, 5 μm. (B, C, and D) Cortical neurons were cotransfected with myr-dGFP^Actb-3′^UTR and either DsRed-, DsRed-KIS-, or DsRed-KIS^KD-expressing constructs and incubated for 14 to 16 h. The levels of myr-dGFP^Actb-3′^UTR mRNA were determined by real-time RT-PCR analysis from equivalent amounts of total RNA, and the mean values and confidence limits (α = 0.05) from three measurements of two independent experiments are plotted (B). The total fluorescence in myr-dGFP^Actb-3′^UTR particles (n > 650) was quantified as described in Materials and Methods. Median values with confidence limits (α = 0.05) (C) and logarithmic distributions (D) were plotted. ctrl, control.

KIS plays a relevant role in neuritic outgrowth. (A) HEK293T cells were cotransfected with FLAG-KIS and either KIS siRNA or a scrambled control siRNA. Samples were analyzed by Western blotting to detect FLAG-KIS. The corresponding region of the blot stained with Coomassie blue (CB) is shown as a loading control. (B) Cortical neurons were cotransfected with a GFP reporter vector and either control or KIS siRNA and analyzed 16 h after transfection. Representative cells are shown for each condition. Bar, 20 μm. (C and D) Mean values and confidence limits (α = 0.01) for the total neurite length per cell (C) and the number of neurites longer than 3 μm in each cell (D) were plotted.

KIS colocalizes with RNA and KIF3A. (A) Cortical neurons infected with a lentiviral vector expressing HA-KIS were labeled with Syto14 to detect RNA and analyzed by immunofluorescence with an anti-HA antibody. Bar, 5 μm. The inset shows a reduced bright-field image. Profiles of HA-KIS (green) and Syto14 (red) fluorescence values are shown for the indicated tracings. (B) Western blot analysis of endogenous KIS and lentiviral HA-KIS proteins expressed by cortical neurons. In this experiment, the infection efficiency was ca. 27%, as deduced from immunofluorescence analysis. Lentiviral HA-KIS and endogenous KIS protein levels attained in HEK293T cells are also shown. (C) Subcellular fractionation of cortical cells expressing FLAG-KIS. The final supernatant (S3) and pellets (P3), incubated with (+) or without (−) RNase prior to the last centrifugation, were analyzed by Western blotting with antibodies against KIF3A, eEF1A, NonO, PSF, and hnRNP A/B proteins. FLAG-KIS and NonO proteins were expressed from lentiviral vectors and detected with anti-FLAG and anti-NonO antibodies. (D) Cortical neurons were analyzed by immunofluorescence to detect lentiviral HA-KIS and endogenous KIF3A proteins. Bar, 5 μm. The inset shows a reduced bright-field image. Profiles of HA-KIS (green) and KIF3A (red) fluorescence values are shown for the indicated tracings.

KIS overexpression increases neuritic GFP^Actb-3′^UTR levels in a kinase-dependent manner. (A) In vivo fluorescence images of two representative neurons cotransfected with GFP^Actb-3′^UTR and DsRed fusions to KIS or the KIS^KD mutant, which are shown in the corresponding insets as short-exposure images to observe only the red signal in soma. Bars, 10 μm. (B) Fourteen to 16 h after transfection with GFP^Actb-3′^UTR or GFP-expressing vectors and either DsRed-, DsRed-KIS-, or DsRed-KIS^KD-expressing constructs, green fluorescence in neurites and red fluorescence in soma were quantified and normalized to those of DsRed cells. The mean values (n ≥ 50) and confidence limits for the mean (α = 0.01) are plotted. ctrl, control. (C) Correlation analysis between normalized values of neuritic GFP^Actb-3′^UTR and normalized values of somatic DsRed, DsRed-KIS, and DsRed-KIS^KD levels in the transfected cells analyzed in panel B. (D) Correlation analysis for neuritic GFP signal as in panel C.

KIS interacts with KIF3A and RNP-transported mRNAs in postnatal brain extracts. (A) Whole brains from newborn mice were used in immunoprecipitation assays with anti-KIS, anti-KIF3A, or preimmune antisera. IP and a sample corresponding to 1/200 of the input were analyzed by Western blotting to detect endogenous KIS and KIF3A proteins. (B) IP with anti-KIS and the preimmune (preimm) antisera shown in panel A were used to quantify the relative levels of the indicated mRNAs by real-time RT-PCR. Gene symbols and descriptions are as follows: ACTB, β-actin, cytoplasmic; GRIA1, glutamate receptor, ionotropic, AMPA1 (alpha 1); GRIA2, glutamate receptor, ionotropic, AMPA2 (alpha 2); ARC, activity-regulated cytoskeletal-associated protein; MAPT, microtubule-associated protein Tau; MTAP2, microtubule-associated protein 2; NTRK2, neurotrophic tyrosine kinase, receptor TrkB, type 2; TFRC, transferrin receptor; CAMK2a, calcium/calmodulin-dependent protein kinase II alpha; SYAP1, synapse-associated protein 1. Ratios to GAPDH mRNA were obtained, and the mean values from three independent immunoprecipitation experiments and respective confidence limits (α = 0.05) were plotted. In addition, two sets of independent data were obtained for β-actin mRNA (ACTB). PGAM1 and a second measurement for GAPDH (2) were included as controls.

Interactions of KIS with NonO, KIF3A, and eEF1A in the yeast two-hybrid assay and HEK293T cells. (A) Domain structure of KIS. The numbers refer to amino acids. (B) Two-hybrid interactions of full-length KIS and versions that lack the kinase domain (KISΔKM) or the UHM domain (KISΔUHM) with NonO, KIF3A, and eEF1A. Residues present in bait constructs, as well as positive clones isolated from prey libraries, are shown in parentheses. Full (++), partial (+), and no (−) complementation for growth are indicated. (C) Cells were plated in synthetic complete medium without histidine in the presence of 25 mM aminotriazole and incubated at 30°C for 3 days. (D) HEK293T cells expressing FLAG-tagged versions of full-length KIS (wt), a kinase-dead mutant (K54A^KD), kinase domain (ΔUHM), UHM domain (ΔKM), or empty vector (−) as a negative control were used for immunoprecipitation with anti-FLAG antibody. The input and anti-FLAG IP were analyzed by Western blotting to detect endogenous NonO and eEF1A or cotransfected HA-KIF3A and FLAG-KIS proteins. The amounts of NonO, HA-KIF3A, and eEF1A proteins in IP were quantitated, and data relative to the respective immunoprecipitated FLAG-KIS proteins are indicated. NonO, HA-KIF3A, and eEF1A were barely detectable in ΔKM IP even after the exposure time was increased 2.5-fold. An immunoglobulin G cross-reactive band is indicated by an asterisk. (E) HEK293T cells expressing FLAG-KIS or transfected with empty vector (−) as a negative control were used for immunoprecipitation with anti-FLAG antibody as described above. Extracts were incubated on ice for 2 h with 10 μg/ml RNase A (+) or 100 U/ml RNase inhibitor (−) prior to immunoprecipitation. The input and anti-FLAG IP were analyzed by Western blotting to detect endogenous NonO and eEF1A or cotransfected HA-KIF3A and FLAG-KIS proteins.