NTRAP interacts with Trk receptor tyrosine kinases. (A) The amino acid sequence of NTRAP. The first underlined region represents a RING finger domain, and the next five underlined regions are five tandem C₂H₂ zinc finger motifs. The two proline-rich regions are shown in bold. (B) Interaction of NTRAP with the intracellular domain of TrkA, TrkB, TrkC, or Met in yeast two-hybrid assays. Laminin was used as a negative control (CTL). The interaction strength was determined by measuring the activity of the reporter β-galactosidase. (C) Specificity of NTRAP antibody. Protein extracts from PC12 cells or E18.5 rat DRGs were blotted with the affinity-purified NTRAP antibody. (D) Enhancement of interaction between NTRAP and TrkC by NT3. Lysates of HEK293 cells were immunoprecipitated with anti-Myc antibodies, and immunoprecipitates were blotted with the anti-NTRAP antibody. Cell lysates (25 μg) were also blotted with antibodies to Myc, NTRAP, and α-tubulin for examination of expression. (E) Enhancement of interaction between NTRAP and TrkA by NGF. HEK293 cells were transfected with constructs expressing Flag-tagged TrkA and NTRAP. Cell lysates were immunoprecipitated with anti-Flag antibodies, and immunoprecipitates were blotted with anti-NTRAP antibodies. Cell lysates (25 μg) were also blotted with antibodies to Flag, NTRAP, and α-tubulin. (F) Interaction of endogenous NTRAP and TrkA. PC12 cell lysates were immunoprecipitated with the mouse IgG against TrkA, and immunoprecipitates were blotted with the rabbit IgG against NTRAP or TrkA. Cell lysates were blotted with the antibody to NTRAP. NGF treatment enhanced the association of the two proteins. (G) Tissue distribution of NTRAP mRNA in adult mice, as revealed by Northern hybridization. (H) High levels of NTRAP in the nervous system and liver of an E12.5 mouse embryo, as revealed by immunohistochemistry. (I) NTRAP expression in DRG ganglia of an E12.5 mouse embryo. Note that axons of DRG neurons also contain NTRAP (arrow). (J) Localization of NTRAP in the cytoplasm of DRG neurons. Scale bar, 10 μm.

Colocalization of NTRAP with Trk receptors in DRG neurons. Sections at 10 μm were obtained from dissected rat DRG at E18.5 and stained with the affinity-purified NTRAP antibody and an antibody that recognizes all three Trk receptors (A–C). The arrow indicates one neuron that expresses NTRAP but few Trk receptors. The boxed areas are shown in A1–C1 at higher magnification. Scale bars, 5 μm.

Identification of interaction domains in NTRAP and TrkC. (A) Schematic representation of NTRAP mutants. (B) Yeast cells expressing the intracellular domain of TrkC fused to LexA and VP16-NTRAP mutants were streaked onto a histidine-free plate containing 10 mM 3-AT and grown at 30°C for 8 d. The number of grown yeast cells is an indicator of the interaction strength between TrkC and an NTRAP mutant. (C) Interaction of TrkC with NTRAP mutants. The #3 C₂H₂ construct is comprised of the VP16 activation domain and the third C₂H₂ zinc finger motif. The NTRAP-ZFM construct is comprised of the VP16 activation domain and an NTRAP mutant in which the two cysteine residues in the third C₂H₂ zinc finger have been changed to two alanine residues. The β-galactosidase activity in yeast two-hybrid assays reflects the interaction strength between two proteins. Error bars, SEs. **p < 0.01 by Student's t test; n = 3. (D) Interaction of NTRAP with wild-type TrkC (WT) or TrkC mutants (ΔTK and K572N). Error bars, SEs. **p < 0.01 by Student's t test; n = 3.

Inhibition of NTRAP blocks NGF-induced differentiation of PC12 cells. (A) Stable expression of HA epitope-tagged C₂H₂ zinc finger region of NTRAP (ZF-HA) in PC12 cells. Two control lines (vector 1 and vector 2) and two expression lines (ZF-HA1 and ZF-HA2) of PC12 cells were established. (B) Representative images showing inhibition of NGF-induced differentiation of PC12 cells by expression of ZF-HA. The images were taken after 4 d of NGF treatment. (C) Quantification of PC12 cell differentiation. The experiment was performed in triplicate. Percentage of cells with neurites was significantly lower in ZF-HA1 and ZF-HA2 lines compared with either of the two control lines at 2, 4, or 6 d of NGF treatment (p < 0.05 by Student's t test). Error bars, SEs. (D) Effect of ZF-HA expression on viability of PC12 cells in the presence of NGF. MTT assays were repeated four times. Error bars, SEs. **p < 0.01 by Student's t test. (E) No effect of HA-ZA expression on survival of PC12 cells in the presence or absence of NGF, as revealed by TUNEL staining (n = 4). Error bars, SEs. (F) Representative images showing reduced NTRAP immunoreactivity in PC12 cells expressing NTRAP shRNA 2 d after transfection. The shRNA constructs were cotransfected into PC12 cells with a construct expressing EGFP. (G) Quantification of NTRAP immunoreactivity in PC12 expressing either control shRNA or NTRAP shRNA. Error bars, SEs. *p < 0.05 by Student's t test. (H) Western blot analysis of extracts prepared from HEK293 cells transfected with adeno-associated viral (AAV) constructs expressing both EGFP and shRNA (either control shRNA or NTRAP shRNA). Some cells were also cotransfected with a construct expressing shRNA-resistant NTRAP mRNA (NTRAPr). The numbers represent the ratio of NTRAP amount to α-tubulin amount. (I) Representative images showing NGF-induced differentiation of PC12 cells expressing shRNA. PC12 cells were transfected with an AAV construct expressing EGFP and shRNA or cotransfected with the AAV construct expressing EGFP and NTRAP shRNA and a construct expressing NTRAPr mRNA. Two days later, the cells were treated with 100 ng/ml NGF for 3 d. Scale bar, 10 μm. (J) Quantification of NGF-induced differentiation of PC12 cells expressing shRNA after 3 d of NGF treatment. Error bars, SEs. ***p < 0.001 by Student's t test (n = 6–7 coverslips for each condition). (K) Measurement of neurite length in PC12 cells expressing shRNA after 3 d of NGF treatment. Error bars, SEs. ***p < 0.01 by Student's t test (n = 20–22 cells for each condition).

NTRAP affects the Rap1-ERK1/2 pathway. (A) Representative Western blots indicating that the sustained activation of ERK1/2 by NGF is diminished in ZF-HA2 line of PC12 cells. After overnight serum starvation, PC12 cells were incubated with NGF (100 ng/ml) for 0, 5, 20, 60, or 120 min. Lysates were blotted with antibodies to p-ERK1/2. The same membrane was also blotted with antibodies to ERK1/2 for examination of expression. (B) Quantification of p-ERK1/2 levels. Intensities of p-ERK1/2 bands were measured using ImageJ software and normalized to levels of total ERK1/2 (n = 3). Error bars, SEM. *p < 0.05 by Student's t test when compared with line vector 1. (C) Representative Western blots showing that Rap1 activation is delayed in ZF-HA2 line of PC12 cells. After overnight starvation in serum-free medium, cells were treated with NGF (100 ng/ml) for 0, 30, or 60 min. (D) Representative images of Rap1 immunocytochemistry in vector 1 and ZF-HA2 lines of PC12 cells. The outline of PC12 cells and location of nucleus are revealed by phaloidin staining of actin filaments (green) and DAPI staining of DNA (blue), respectively. Scale bar, 5 μm. (E) Distribution of Rap1 in the cytoplasm of PC12 cells. Rap1 immunofluorescence was measured at 0.5-μm intervals from the edge of the nucleus to the cell periphery. Error bars, SEM. *p < 0.05 by Student's t test (n = 30 cells for each condition). (F) Normal Akt activation in PC12 cells expressing ZF-HA. After overnight serum starvation, PC12 cells were incubated with NGF (100 ng/ml) for 0, 5, 20, or 60 min. Lysates were blotted with antibodies to p-Akt. The same membrane was also blotted with antibodies to Akt for examination of expression. (G) Representative Western blots indicating that the sustained activation of ERK1/2 by NGF is decreased in DRG neurons expressing NTRAP shRNA. AAV-infected DRG neuronal cultures were incubated with NGF (100 ng/ml) for 0, 1, 2, 4, or 6 h. Lysates were blotted with antibodies to p-ERK1/2. The same membrane was then blotted with antibodies to ERK1/2 for examination of expression.

NGF-induced localization of NTRAP in late endosomes. (A) Colocalization of NTRAP with the early endosome marker, EEA1. The graph shows percentage of NTRAP immunoreactivity that was colocalized with EEA1. Scale bar, 10 μm. (B) Colocalization of NTRAP with the late endosome marker, Rab7. PC12 cells were starved for serum overnight and then incubated with/without NGF (100 ng/ml) for 20 min. NGF treatment increased colocalization of NTRAP and Rab7. Scale bar, 10 μm.

NTRAP controls TrkA trafficking and ERK activation in endosomes. (A) Fractionation of membranous organelles. Membranous components isolated from PC12 cells were separated in iodixanol density gradients. The numbers inside the diagram indicate the concentration of iodixanol. Immunoblots show the distribution of EEA1, Rab7, and NTRAP isolated from vector 1 line of cells. (B) Distribution of the TrkA receptor among the four interfaces before and after NGF treatment. The numbers show the distribution of TrkA among the four interfaces. (C) Distribution of ERK1/2 among the four interfaces before and after NGF treatment. The numbers show the distribution of ERK1/2 among the four interfaces.

Colocalization of NTRAP with signaling endosome markers in DRG neurons. Dissected dorsal root ganglia from E18.5 rat embryo were sectioned at 10 μm, stained with antibodies to NTRAP and EEA1 (A–C) or to NTRAP and NGF (D–F), and examined with a confocal microscope. Scale bars, 5 μm.

Down-regulation of NTRAP reduces neuronal survival and CREB activation promoted by retrograde neurotrophic signaling. (A) Reduced NTRAP immunoreactivity in cultured DRG neurons infected with NTRAP shRNA AAV compared with those infected by control shRNA AAV. (B) Reduced levels of NTRAP in cultured DRG cells infected with NTRAP shRNA AAV, as revealed by Western blots. (C) Quantification of NTRAP levels. NTRAP signals on Western blots were normalized to those of α-tubulin. Error bars, SEs. p < 0.05 by Student's t test, n = 3. (D) DAPI staining of cultured DRG neurons. Arrows indicate apoptotic cells with fragmented or condensed nuclei. Scale bar, 5 μm. (E) Quantification of apoptotic neurons in compartmentalized neuronal cultures. Neurotrophins (NTs) were applied to distal axons (DA) or cell bodies (CB) as indicated. EGFP-positive cells with fragmented or condensed nuclei were counted as apoptotic cells. Error bars, SEs. **p < 0.01 by Student's t test; n = 3. (F) Activation of CREB by neurotrophins applied to distal axons, revealed by immunocytochemistry with an antibody specific to p-CREB. AAV-infected neurons were marked by EGFP. P-CREB immunoreactivity was lower in neurons infected by NTRAP shRNA AAV than in neurons infected by control shRNA AAV. Scale bar, 10 μm. (G) Reduced CREB activation by neurotrophins applied to distal axons in DRG neurons expressing NTRAP shRNA. Neurons positive for EGFP and neurons positive for both EGFP and p-CREB were counted and the numbers were used to calculate percentage of p-CREB–positive cells. Error bars, SEs. **p < 0.01 by Student's t test; n = 3.