Supplemental Information.

NGF trafficks together with TrkA to early endosomes in PC12 cells The signaling endosome hypothesis predicts that NGF/TrkA complexes are trafficked within endosomes (1–3). We reexamined the issue, focusing on a specific endosomal compartment, the early endosome; it is within early endosomes in the perinuclear region that NGF treatment results in persistent pMAPK1/2 in PC12 cells (4). We used BtNGF, which possesses full biological activity, in combination with a strepavidin-Alexa 488 conjugate that allowed for detection by immunofluoresence. Serum-starved PC12 cells were chilled and were incubated with 50 ng/ml BtNGF for 45 min on ice to allow surface binding but not internalization of BtNGF. This was confirmed by showing under these conditions that >95% of the BtNGF-streptavidin Alexa 488 signals were susceptible to subsequent acid-stripping in 0.2 N Acetic Acid, 0.4 M NaCl at 4°C for 5 min (data not shown). Binding of BtNGF to the PC12 cell surface was receptor-mediated as there was little signal for BtNGF-strepavidin Alexa 488 when excess unmodified NGF (1.0 μM) was included in the incubation (data not shown). We then prepared two sets of samples. One was rinsed and fixed immediately following the binding step (Fig.1S A). The second set was rinsed and shifted to 37°C for 30 min to allow for endocytosis of BtNGF (Fig.1S B). Both sets were then processed for immunostaining with MCTrk, a mouse antibody to Trk, and analyzed by confocal microscopy. The signal for Trk overlapped extensively with that for BtNGF with or without warming (Fig.1S A). However, BtNGF/Trk concentrated in the perinuclear region following warming (Fig.1S B). To ask whether or not NGF was present in early endosomes, we compared the pattern for BtNGF with that for an early endosomal marker, EEA1 (Fig. 1S C and D). Whereas surface-bound BtNGF showed little, if any, co-localization with EEA1 (Fig.1S C), following warming, BtNGF was markedly co-localized with EEA1 (Fig.1S D). Similarly, without warming, Trk showed little co-localization with Rab5B (Fig.1S E), another early endosomal marker protein. Following warming Trk was co-localized with Rab5B (Fig.1S F). As expected, pMAPK1/2 that was induced by NGF following warming (Fig.1S H vs G) was also present in the Rab5B-positive compartment (Fig.1S H). Our results thus place NGF and TrkA in early endosomes in the perinuclear region.

NGF treatment induces trafficking of Rap1 to and activates MAPK1/2 in early endosomes.

To investigate the trafficking events induced by NGF treatment, we examined the distribution of TrkA, Rap1 and pMAPK1/2. PC12 cells were grown on glass coverslips and were serum-starved overnight. The cells were treated with either the NGF vehicle (-NGF) or 50 ng/ml NGF for 30 min (+NGF). This time point was chosen to maximize trafficking of TrkA to early endosomes and activation of Rap1 (4, 5). Samples were then prepared for indirect immunofluorescence and analyzed using confocal microscopy. To outline the cell perimeter, in some cases (Fig.2S A, B, E, F) images were also captured using transmitted light and were superimposed onto the fluorescent image. In control cells (-NGF), Rap1 (Fig.2S A) and TrkA (Fig.2S E) were found dispersed throughout the cytoplasm (also see Fig. 4). NGF treatment resulted in marked concentration of both Rap1 (Fig.2S B vs A) and TrkA (Fig.2S F vs E) in the perinuclear region. The perinuclear localization of Rap1 in NGF-treated cells was also illustrated in studies in which the nucleus was stained using Hoechst 33258 (Fig.2S D vs C).

Next, we examined the subcellular localization of Rap1 with respect to TrkA, pMAPK1/2 and EEA1. Again, in cells treated with vehicle only, both Rap1 and TrkA were seen dispersed throughout the cytoplasm and showed little colocalization (Fig.2S G). As expected, no signal for pMAPK1/2 was seen in these samples (Fig.2S I). Treatment with NGF resulted in the concentration of Rap1 and TrkA in the perinuclear region (Fig.2S H). NGF treatment also resulted in the appearance of pMAPK1/2 in the perinuclear region, where pMAPK1/2 partially co-localized with Rap1 (Fig.2S J). Rap1 was partially co-localized with EEA1 before and after NGF treatment (Fig.2S K, L). We conclude that NGF treatment induces trafficking events that lead to the concentration of TrkA, Rap1 and pMAPK1/2 within early endosomes in the perinuclear region.

Supplemental Figure 1: BtNGF was internalized into EEA1-positive endosomes in PC12 cells. PC12 cells were cultured for 24–48 hrs on cover glasses coated with matri-gel and were serum-starved overnight. Cells were chilled and incubated with BtNGF (50 ng/ml) for 45 min on ice. Cells were rinsed and were immediately processed for immunostaining (A, C, E, G). A parallel sample was shifted to 37°C for 30 min to allow endocytosis of BtNGF prior to immunstaining (B, D, F, H). BtNGF was detected with a streptavidin-Alexa 488 conjugate (green, A-D). The following specific primary antibodies were used: a mouse antibody to Trk, B3 (1/200) for TrkA (A, B, E, F); a mouse antibody to EEA1 (1/50) for EEA1 (1/50) (C, D); a rabbit antibody to Rab5B (1/200) for Rab5B (E, F, G, H); a mouse antibody to pMAPK1/2 (1/100) for pMAPK1/2 (G, H). Secondary antibody conjugated to either Alexa 568 or Alexa 488 was used to visualize the primary antibody. Co-localization of proteins is denoted by the white signal. The size of the collection box is 30 μm × 30 μm.

Supplemental Figure 2: NGF induced trafficking of Rap1, TrkA and pMAPK1/2 to the perinuclear region. PC12 cells were cultured as in Figure 1. Following pretreatment with either the NGF vehicle alone (-NGF) or with NGF (50 ng/ml) (+NGF), indirect immunofluorescence was carried out using confocal microscopy as described in Materials and Methods. Rap1 (A-D, G-L) was immunostained using a specific rabbit antibody (1/250) and detected with an Alexa 568 goat anti-rabbit IgG conjugate. TrkA (E-F, G-H) was detected using B3 (1/200) and visualized with either an Alexa-568 goat anti-mouse IgG conjugate (E-F) or an Alexa-488 goat anti-mouse IgG conjugate (G-H). The signals for pMAPK1/2 and EEA1 were each detected using a mouse monoclonal antibody specific for these proteins as in Fig. 1. The dilutions of primary antibodies were as follows: 1/100 for pMAPK1/2 and 1/50 for EEA1. An Alexa 488 goat anti-mouse IgG conjugate was used to visualize these antibodies. Co-localization of antigens is denoted by the white signal. In A, B, E and F, a corresponding image from the transmitted light detector was simultaneously captured and superimposed onto the fluorescence image of Rap1 or TrkA to illustrate the outline of the cell. In C and D, Rap1 (magenta) was co-stained with Hoechst 33258 to reveal the nucleus (blue). The resulting samples were captured and analyzed using a Nikon E800 epifluorescence microscope. 25 cells were examined in each condition (A-L) and a typical representative image was shown. The size of the collection box is 30μm × 30 μm for A-B, E-L, 60 μm × 60 μm for C, D.


Howe CL, Mobley WC. Signaling endosome hypothesis: A cellular mechanism for long distance communication. J Neurobiol 2004;58:207–16.

Ye H, Kuruvilla R, Zweifel LS, Ginty DD. Evidence in support of signaling endosome-based retrograde survival of sympathetic neurons. Neuron 2003;39:57–68.

MacCormick M, Moderscheim T, van der Salm LW, Moore A, Pryor SC, McCaffrey G, Grimes ML. Distinct signalling particles containing ERK/MEK and B-Raf in PC12 cells. Biochem J 2005;387:155–64.

Wu C, Lai CF, Mobley WC. Nerve growth factor activates persistent Rap1 signaling in endosomes. J Neurosci 2001;21:5406–16.

Mochizuki N, Yamashita S, Kurokawa K, Ohba Y, Nagai T, Miyawaki A, Matsuda M. Spatio-temporal images of growth-factor-induced activation of Ras and Rap1. Nature 2001;411:1065–8.

tra_718_sm_8_11gfig1.jpg238KSupporting info item
tra_718_sm_8_11gfig2.jpg260KSupporting info item

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.