Sorting of internalized neurotrophins into an endocytic transcytosis pathway via the Golgi system: Ultrastructural analysis in retinal ganglion cells

Abstract

Subcellular pathways and accumulation of internalized radiolabeled neurotrophins NGF, BDNF, and NT-3 were examined in retinal ganglion cells (RGCs) of chick embryos by using quantitative electron microscopic autoradiography. All three neurotrophins accumulated in endosomes and multivesicular bodies. BDNF and NGF also concentrated at the plasma membrane, whereas NT-3 accumulated transiently in the Golgi system. The enhanced targeting of NT-3 to the Golgi system correlated with the anterograde axonal transport of this neurotrophin. Anterograde transport of NT-3, but not its internalization, was significantly attenuated by the tyrosine kinase (trk) inhibitor K252a. Abolishment of trk activity with K252a shifted NT-3 (and BDNF) away from the Golgi system and into a lysosomal pathway, indicating that trk activity regulated sorting of the ligand-receptor complex. Cross-linking of neurotrophins and immunoprecipitation with antibodies to the neurotrophin receptors p75, trkA, trkB, and trkC showed that the large majority of exogenous, receptor-bound NT-3 was bound to trkC in RGC somata, but during anterograde transport in the optic nerve most receptor-bound NT-3 was associated with p75, and after arrival and release in the optic tectum transferred to presumably postsynaptic trkC. These results reveal remarkable and unexpected differences in the intracellular pathways and fates of different neurotrophins within the same cell type. They provide first evidence for an endocytic pathway of internalized neurotrophic factors via the Golgi system before anterograde transport and transcytosis. The results challenge the belief that after internalization all neurotrophins are rapidly degraded in lysosomes.

Fig. 1.

A–C, Anterograde axonal transport of neurotrophic factors and cytochromec in chick RGCs. A, B, The relative amount of anterograde transport of radiolabeled NT-3 was plotted by dividing the amount measured by gamma counting in the tectum (counts per minute per specific activity in picograms) by the amount measured in the eye (counts per minute per specific activity in nanograms) at the time chick embryos were killed (20 hr after injection in the eye). A, The effects of coinjection of monensin (MON) (von Bartheld et al., 1996a), excess cold NT-3 (cold NT-3), excess cold NGF (cold NGF), or excess cold BDNF (cold BDNF), blocking p75 antibody (aP75), blocking trkC antibody (atrkC), and normal rabbit IgG (IgG). B, The effects of coinjection of monensin (MON), tyrosine kinase inhibitor K252a (K252a), vehicle (DMSO), wortmannin (WOR), LY294,002 (LY), and brefeldin A (BFA) (von Bartheld et al., 1996a) are indicated. C, The relative amount of anterograde transport of radiolabeled glial cell line-derived neurotrophic factor (GDNF), cardiotrophin-1 (CT1), and cytochrome c (CytC) was plotted by dividing the amount measured by gamma counting in the midbrain (picograms) by the amount measured in the eye (nanograms) at the time chick embryos were killed (20 hr after injection in the eye). The effects of coinjection of monensin (MON), excess cold GDNF, or CytC, and K252a are indicated. Error bars indicate SEM. The number of independent experiments is indicated. ***p ≤ 0.005; **p ≤ 0.01; *p ≤ 0.05.

Fig. 2.

Internalization of ¹²⁵I-NT-3 (10–20 ng/ml) in purified retinal ganglion cells from E18–21 chick embryos does not require tyrosine kinase activity. The amount of internalization in the presence of 1 μg/ml K252a (tyrosine kinase inhibitor) is plotted as the percentage relative to the values for vehicle (1 μg/ml DMSO), which were averaged to 100% internalization (∼40,000 cpm per plate). Nonspecific association of NT-3 (at 4°C) is indicated (dotted line). Error bars indicate SEM. The number of independent experiments (each in duplicate) is indicated. Thep level for confidence (t test) is indicated, showing no significant effect of K252a on NT-3 internalization.

Fig. 3.

A–D, In situ hybridization for neurotrophin receptors in the central region of the ganglion cell layer (GCL) of 16- to 18-d-old chick embryos. This area contains ∼85–90% retinal ganglion cells (Ehrlich, 1981). A, Cells labeled for p75 neurotrophin receptor mRNA. B, Cells labeled with probes specific for the tyrosine kinase domain of the chick trkB neurotrophin receptor. C, Cells labeled with probes specific for the tyrosine kinase domain of the chick trkC receptor. Scale bars, 5 μm.D, The intensity of labeling (Grains/Cell) is plotted against the relative frequency (percentage of cells with this labeling intensity). Note that trkB-labeled cells contain nearly twice the number of grains as p75- or trkC-labeled cells (after subtraction of background). The number of cells sampled is indicated.

Fig. 4.

Ultrastructure of a retinal ganglion cell (RGC) and an amacrine cell (AC) in the ganglion cell layer of an 18-d-old chick embryo. The plasma membrane of the RGC is marked by black dots; the plasma membrane of the AC is marked with asterisks. Note the abundance of rough endoplasmic reticulum (ER) in the RGC, and the paucity of such organelles in the AC. G, Golgi system;M, mitochondrion. This retina was injected with NT-3 and processed after 10 hr. Scale bar, 1 μm.

Fig. 5.

A–F, Examples of accumulation of radiolabeled internalized neurotrophins over organelles in E18 retinal ganglion cells (A–E) and tectal cells (F). A, Accumulation at the plasma membrane (PM); BDNF after 10 hr. Three silver grains touch the PM. B, Accumulation in multivesicular bodies (MVB); NT-3 after 20 hr.C, Accumulation in dense endosomes (ES); NT-3 after 10 hr. D, Accumulation in the endoplasmic reticulum (ER); NT-3 after 20 hr. E, Accumulation in the Golgi apparatus (G) and Golgi-associated vesicles; NT-3 after 10 hr. F, Accumulation in a lysosome (LS); NT-3 after 20 hr and anterograde transport in RGC axons and axo-dendritic transfer to a tectal cell in a 16-d-old chick embryo. Scale bars (shown inA for A–D), 500 nm; (shown in F for E,F), 500 nm.

Fig. 6.

A–D, Effect of K252a and trkC activity on the distribution of internalized neurotrophins in retinal ganglion cells (RGC) (A–C) and SDS-PAGE analysis of internalized neurotrophins in purified RGCs (D).A, Labeling densities of internalized BDNF and NT-3 in the Golgi system of RGCs with and without K252a. The analysis was done in triplicate; significance was determined by unpaired ttest. B, Labeling densities of internalized BDNF and NT-3 in lysosomes and endosomes of RGCs with and without K252a. The analysis was done in triplicate; significance was determined by unpaired t test. C, Quantification of anterograde transport of 50–80 ng radiolabeled NGF when coinjected in the eye with 50–60 ng cold NT-3 or BDNF. The number of experiments is indicated. Significance was determined by unpaired ttest. Error bars indicate SEM. D, SDS-PAGE (15%) of internalized neurotrophins NGF, BDNF, and NT-3 recovered from purified immunopanned RGCs after 10 hr. Each sample was run with an adjacent sample of native same factor (Na). The molecular weight is indicated. Arrow indicates the dye front. Note that much of the BDNF recovered from RGCs is cleaved, whereas virtually all the NGF and NT-3 is intact protein by this analysis.

Fig. 7.

A–C, Cross-linking and immunoprecipitation of radiolabeled neurotrophins internalized and anterogradely transported by retinal ganglion cells (RGCs). A, Binding of NT-3 to p75, trkB, and trkC receptors cross-linked with EDC or DSS and immunoprecipitated with chicken-specific neurotrophin receptor antibodies. The relative amount of total specific precipitation (after subtraction of nonspecific precipitation) is shown separately for each tissue (purified RGCs, optic chiasm, and optic tectum) and for each receptor. Totals for p75, trkB, and trkC (from top tobottom) add up to 100% for each tissue and cross-linker. B, Binding of NT-3 to receptors when radiolabeled NT-3 was added to the same tissues and then cross-linked and immunoprecipitated. Note the much more extensive precipitation with trkB antibodies (binding to trkB) when NT-3 was added to normal lysates of the same tissues rather than introduction of NT-3 in vivo. C, Immunoprecipitation of internalized radiolabeled BDNF and NT-3 binding to receptors in RGCs purified by immunopanning. Note that BDNF binds to p75 as well as trkB, whereas NT-3 binds almost exclusively to trkC receptors at 12 hr; binding of NT-3 to trkC in RGCs is transient, because it is much reduced 44 hr after injection (C). Error bars indicate SEM. The number of experiments is indicated.

Fig. 8.

Diagram summarizing proposed subcellular pathways of internalized NT-3 (black dots) in a retinal ganglion cell. At least two pathways of internalized NT-3 can be distinguished. A lysosomal pathway of NT-3 may be common to all neurotrophins and may involve binding to the p75 receptor (U) or binding of NT-3 to trkC receptor (Y) in the presence of the tyrosine kinase inhibitor K252a. The neurotrophin is degraded in lysosomes (LYS). Alternatively, a novel pathway of NT-3 internalized in endosomes fuses with membranes of the Golgi apparatus:Golgi Pathway. This sorting requires tyrosine kinase activity (presumably trkC, Y), and this pathway may join that of newly synthesized neurotrophins as well as p75 receptor (U) from the endoplasmic reticulum (ER) via the Golgi into an anterograde axonal path (von Bartheld et al., 2001). After passage through the Golgi system, internalized NT-3 is packaged in presumptive large dense-core vesicles (LDCV) (Wang et al., 2001) for anterograde axonal transport. In this pathway, internalized NT-3 binds preferentially to p75. Anterogradely transported NT-3 is released from RGC axon terminals in the tectum, and after release it binds predominantly to trkC in tectal cells where it accumulates in multivesicular bodies (MVB) (von Bartheld et al., 1996a).