Konopacki FA et al. (2016), ESCRT-II controls retinal axon growth by regula...

XB-ART-52527

Open Biol 2016 Apr 01;64:150218. doi: 10.1098/rsob.150218.

Show Gene links Show Anatomy links

ESCRT-II controls retinal axon growth by regulating DCC receptor levels and local protein synthesis.

Konopacki FA , Wong HH , Dwivedy A , Bellon A , Blower MD , Holt CE .

???displayArticle.abstract???
Endocytosis and local protein synthesis (LPS) act coordinately to mediate the chemotropic responses of axons, but the link between these two processes is poorly understood. The endosomal sorting complex required for transport (ESCRT) is a key regulator of cargo sorting in the endocytic pathway, and here we have investigated the role of ESCRT-II, a critical ESCRT component, in Xenopus retinal ganglion cell (RGC) axons. We show that ESCRT-II is present in RGC axonal growth cones (GCs) where it co-localizes with endocytic vesicle GTPases and, unexpectedly, with the Netrin-1 receptor, deleted in colorectal cancer (DCC). ESCRT-II knockdown (KD) decreases endocytosis and, strikingly, reduces DCC in GCs and leads to axon growth and guidance defects. ESCRT-II-depleted axons fail to turn in response to a Netrin-1 gradient in vitro and many axons fail to exit the eye in vivo These defects, similar to Netrin-1/DCC loss-of-function phenotypes, can be rescued in whole (in vitro) or in part (in vivo) by expressing DCC. In addition, ESCRT-II KD impairs LPS in GCs and live imaging reveals that ESCRT-II transports mRNAs in axons. Collectively, our results show that the ESCRT-II-mediated endocytic pathway regulates both DCC and LPS in the axonal compartment and suggest that ESCRT-II aids gradient sensing in GCs by coupling endocytosis to LPS.

???displayArticle.pubmedLink??? 27248654
???displayArticle.pmcLink??? PMC4852451
???displayArticle.link??? Open Biol
???displayArticle.grants??? [+]

Species referenced: Xenopus laevis
Genes referenced: dcc rab11a rab5a rab7a rpl5 snf8 tec ugcg vps25 vps36
???displayArticle.morpholinos??? vps25 MO1 vps25 MO2

???attribute.lit??? ???displayArticles.show???

	Figure 1. ESCRT-II co-localizes with early endosomal vesicles in RGC growth cones. (a–d) ESCRT-II immunoreactivity in GCs (a) co-labelled with F-actin (b; strongly labelling GC periphery including filopodia) and acetylated α-tubulin (c; predominantly staining axon shaft and GC central domain). Panel (d) shows a compound image of (a–c). Note ESCRT-II-positive granules in GC filopodia (insets a–d). (e) Cartoon depicts the involvement of individual Rabs with distinct elements of the endocytic pathway. (f–i) Co-localization of ESCRT-II with endosomal markers Rab5 (f), Rab11 (g), Rab4 (h) and Rab7 (i) in RGC GCs. Arrows point to spots where the signals visibly co-localize. GC outline indicated with yellow line. (j) Table shows average Manders' co-localization coefficients of Rabs and ESCRT-II in whole GC and filopodia. Scale bars, 10 µm.
	Figure 2. ESCRT-II knockdown reduces growth cone endocytosis and accelerates axon growth in vitro. (a) Verification of the MO-induced ESCRT-II knockdown. The morpholino targets the Vps25 subunit of ESCRT-II. Western blots were done in whole eye extracts. Graph shows the mean reduction of Vps25 band intensity in ESCRT-II MO-injected embryos normalized to control. (b–d) Balance of endo/exocytosis measured by the levels of FM4-64 dye loaded over 90 s into control MO (b) or ESCRT-II MO (c) GCs. GC outline indicated with yellow line. (d) Graph shows the average FM4-64 fluorescence in GCs normalized to control MO-injected axons. (e–i) In vitro RGC axon outgrowth assay. The growth of axons from embryos injected with control MO (e,f) or ESCRT-II MO (g,h) measured over 45 min; arrows in (f) and (h) indicate how far the axons extended. (i) Quantification of axon growth speed in vitro. p ≤ 0.05, **p ≤ 0.0001, Student's t-test. Scale bars, 5 µm (b,c), 20 µm (e–h).
	Figure 3. ESCRT-II knockdown impairs axon exit from the eye. (a,b) Images of in vivo ventral preparation of Xenopus embryos electroporated with GFP + control MO (a) or GFP + ESCRT-II MO (b). Note very few axons exiting the eye and coursing to optic tectum in (b). A vertical yellow line indicates the midline. (c,d) Sections of embryos' eyes electroporated with GFP + either a control MO (c; n = 6) or ESCRT-II MO (d; n = 7) stained with DAPI (blue) and GFP (green). (e) Quantification of (c) and (d). Graph shows the average number of labelled cells (vertical axis) and the corresponding number of axons in the optic path (horizontal axis) in both conditions. **p ≤ 0.001, Student's t-test. ONH, optic nerve head. Scale bars, 100 µm (a,b), 50 µm (c,d).
	Figure 4. Impaired Netrin-1 responsiveness in ESCRT-II-depleted growth cones. (a) In vitro RGC axon outgrowth assay. The experimental layout is shown on the top. The growth of axons from embryos injected with control MO, ESCRT-II MO and ESCRT-II MO + ESCRT-II mRNAs was measured from time −45 min to 0 min (without Netrin-1) and subsequently from 30 min to 75 min (with Netrin-1). Drawings show representative examples of quantified axons. (b) Quantification of (a); ***p ≤ 0.0001, paired (except for comparison of bars 2 and 4) Student's t-test.
	Figure 5. ESCRT-II co-localizes with DCC in growth cones. (a–c) Co-localization of ESCRT-II (a; red in c) and DCC (b; green in c) immunofluorescent signals in Xenopus RGC GCs. The signal overlap is especially visible in filopodia (indicated with arrows on insets below). (d) Table shows Manders' co-localization of DCC and ESCRT-II in whole GC and filopodia. (e–g) Proximity ligation assay confirming the close localization of ESCRT-II and DCC in RGC GCs (e). Yellow dots denote the sites where both probes interact. The known interaction of DCC with the large ribosomal subunit protein RPL5 [55] was used as a positive control (f). GCs are outlined with yellow lines. Graph (g) shows the quantification of the number of PLA puncta per unit area. *p ≤ 0.05, Student's t-test. Scale bars, 10 µm.
	Figure 6. ESCRT-II regulates the levels of DCC receptor in growth cones. (a–h) ESCRT-II knockdown leads to decreased DCC levels in GCs. (a–f) Representative examples of GCs from embryos injected with control MO (a,b), ESCRT-II MO (c,d) and ESCRT-II MO + ESCRT-II mRNAs (e,f), stained for ESCRT-II (a,c,e) and DCC (b,d,f). (g,h) Graphs showing the normalized signal intensities of ESCRT-II (g; black bars) and DCC (h; white bars). ***p ≤ 0.0001 compared with control, Students' t-test. (i) A representative western blot from eye extracts indicating that the decrease in DCC levels shown in (b,d) is global. GCs are outlined with yellow lines. Scale bars, 10 µm.
	Figure 7. ESCRT-II regulates surface levels of DCC in growth cones. (a–d) Immunostaining for total (a,c) and surface (b,d) DCC receptor in control (a,b) and ESCRT-II-depleted (c,d) GCs. For clarity, the signal intensities in (b,d) are increased by 30% compared with (a,c). GCs are outlined with yellow lines. (e–g) Graphs showing quantification of total (e) and surface (f) DCC levels and surface to total DCC ratios (g), normalized to the respective controls. *p ≤ 0.05 Mann–Whitney test. Scale bars, 10 µm.
	Figure 8. DCC rescues ESCRT-II knockdown phenotypes. (a–d) In vivo ventral view of the Xenopus optic path in stage 41 embryos whose right eye had been electroporated with control MO (a), ESCRT-II MO (b) and ESCRT-II MO + DCC mRNA (c). The numbers of axons exiting the eye and navigating in the optic pathway were counted and the quantification is shown in (d). OT, optic tract; OC, optic chiasm; tec, optic tectum. (e–o) In vitro turning assay. (e–j) Representative examples of RGC axons from embryos injected with control MO (e,h), ESCRT-II MO (f,i) and ESCRT-II MO + DCC mRNA (g,j) before (e–g) and after (h–j) being subjected to a Netrin-1 gradient ejected from a pipette (indicated with black arrowheads) set at 45° angle from the direction of growth. Growth measurement start point is indicated with horizontal black dotted line; dashed lines show the measured directions of growth at time 0 min and 45 min. (k–m) Traces of control (k), ESCRT-II MO (l) and ESCRT-II MO + DCC mRNA (m) axons growing for 1 h while exposed to Netrin-1 gradient (black arrowheads). (n) Cumulative distribution plot showing the turning angles of all measured axons. *p ≤ 0.05, ANOVA + uncorrected LSD Fisher's test. Scale bars, 20 µm.
	Figure 9. ESCRT-II KD decreases local protein synthesis and ESCRT-II-mRNA positive granules are trafficked along axons. (a–h) De novo protein synthesis in Xenopus RGC GCs in vitro. (a–c) AHA incorporation over 1 hour into non-stimulated RGC GCs. (a,b) Representative images of control (a) and ESCRT-II KD (b) GCs. (c) Graph showing the comparison of the two conditions. *p ≤ 0.01, Mann–Whitney test. (d–h) Puromycin labelling in GCs in response to bath application of Netrin-1 (300 ng ml−1) for 0, 25 and 40 min. (d–g) Representative images of control (d,e) and ESCRT-II KD (f,g) GCs fixed at time 0 (no stimulation: d,f) and 25 min of Netrin-1 stimulation (e,g). In each case, puromycin was added 10 min before fixation. (h) Graph represents measured fluorescence levels, indicative of puromycin labelling (see Material and methods). Blue trace, control MO; red trace, ESCRT-II knockdown. Data normalized to non-stimulated control MO. p ≤ 0.05, **p ≤ 0.01, Kruskal–Wallis test with Dunnis post hoc. GCs outlined in yellow. Scale bars, 5 µm. (i–j) ESCRT-II co-transport with β-actin mRNA in RGC axon. (i) Cartoon showing the experimental design. (j) Time-lapse (12 fpm) imaging of an RGC axon in vitro (outlined in yellow) expressing EGFP-ESCRT-II (green) and Cy3-β-actin mRNA (magenta). Co-localization is visible as white on merge images. Arrows point to an example of an ESCRT-II-positive granule moving together with Cy3-labelled β-actin mRNA. Scale bar, 5 µm. (k,l) A kymograph of the area marked with orange broken arrow in the merge images (k), showing the movement of one ESCRT-II and β-actin mRNA-positive granule (bottom to top of the three kymographs in l), while another one remains stationary (at the top of each kymograph in l).

References [+] :

Allison, An ESCRT-spastin interaction promotes fission of recycling tubules from the endosome. 2013, Pubmed