Nanoparticle-mediated signaling endosome localization regulates growth cone motility and neurite growth

Abstract

Understanding neurite growth regulation remains a seminal problem in neurobiology. During development and regeneration, neurite growth is modulated by neurotrophin-activated signaling endosomes that transmit regulatory signals between soma and growth cones. After injury, delivering neurotrophic therapeutics to injured neurons is limited by our understanding of how signaling endosome localization in the growth cone affects neurite growth. Nanobiotechnology is providing new tools to answer previously inaccessible questions. Here, we show superparamagnetic nanoparticles (MNPs) functionalized with TrkB agonist antibodies are endocytosed into signaling endosomes by primary neurons that activate TrkB-dependent signaling, gene expression and promote neurite growth. These MNP signaling endosomes are trafficked into nascent and existing neurites and transported between somas and growth cones in vitro and in vivo. Manipulating MNP-signaling endosomes by a focal magnetic field alters growth cone motility and halts neurite growth in both peripheral and central nervous system neurons, demonstrating signaling endosome localization in the growth cone regulates motility and neurite growth. These data suggest functionalized MNPs may be used as a platform to study subcellular organelle localization and to deliver nanotherapeutics to treat injury or disease in the central nervous system.

Fig. 1.

FMNPs are endocytosed and colocalize with activated TrkB receptors in primary neurons. (A) Anti-TrkB antibody fMNPs but not cMNPs were detectable in RGC neurites and growth cones as puncta of varying sizes that colocalized with antibodies against phospho-TrkB (α-p-TrkB), in the absence of BDNF. (B) In RGCs cultured with BDNF, neither binding nor endocytosis of fMNPs were detected. (C) CMNPs were not detected in either RGC growth cones or neurites. (D) By α-p-TrkB Western blot, fMNPs but not cMNPs increased p-TrkB (Top), total Trk (Middle), and β actin (Bottom). After fMNP recovery, p-TrkB was not detected in the supernatants (supnt.). (E) FMNPs but not cMNPs increased phospho-ERK1/2 and phospho-Akt. (Scale bar: 10 μm.)

Fig. 2.

FMNP signaling endosomes are transported into nascent RGC neurites. Both fMNPs and cMNPs are detected as discrete puncta in newly plated RGC somas after overnight loading in BDNF(-) suspension cultures. (A) FMNP puncta were robust in RGC somas (arrowheads) and colocalized with TrkB in most but not all fMNP puncta (arrows). (B) CMNPs were also detected as puncta (arrowheads) in some RGC somas, but these puncta were less numerous and usually failed to colocalize with TrkB. (C) DIC and fluorescent images of a fMNP-loaded RGC demonstrate anterograde transport into nascent neurites (arrows) and growth cones (arrowhead). (D) DIC and fluorescent images of a cMNP-loaded RGC lacking transport into either the neurite or growth cone. (Scale bars: 10 μm.)

Fig. 3.

FMNP signaling endosomes are trafficked bidirectionally in nascent RGC neurites. (A) Discrete fMNP puncta were detected in RGC neurites and growth cones (GC). Within neurites, fMNP signaling endosomes were transported both anterogradely (filled arrowhead) and retrogradely (open arrowhead) between the soma (right) and growth cone. Fluorescent images were inverted to maximize contrast. Time in seconds (s) is indicated. (Scale bar: 10 μm.) (B) Average rate of fMNP-loaded signaling endosome transport in anterograde (A) and retrograde (R) directions was similar (mean ± SEM; n = at least 50 vesicles from 10 neurons).

Fig. 4.

In fMNP-loaded RGCs, focal magnetic force alters growth cone motility and halts neurite growth. (A) In control RGCs, a constant 15-pN force failed to alter either growth cone motility or neurite growth rate. This growth cone extended new lamella (l) and filopodia (f), and the neurite continued to grow at ≈50 μm/h throughout the recorded time period. (B) In fMNP-loaded RGCs, a 15-pN force applied for 3 min was sufficient to immobilize both lamellar and filopodial protrusions in the peripheral domain and halt neurite growth. Both neurite growth and central domain (c) advance was immediately stalled. All active lamella (l) and filopodia (f) immobilized for 20 min after removing the magnet (compare 0′ and 23′). By 35 min, both lamellar (l1 and I2) and filopodial protrusions reinitiated at the leading edge in concert with resumed central domain advance (compare 23′ and 55′). Previously immobilized lamella and filopodia (f1–f3) remained immobilized but could still support new lamellar (e.g., l2) protrusions. Time in minutes (′) is indicated. Electromagnet tip indicated by black arrows. (Scale bars: 5 μm.)

Fig. 5.

In fMNP-loaded RGCs, protrusive activity persists in the central domain, despite immobilization in the peripheral domain elicited by a focal magnetic force. (A) A representative lamellar growth cone loaded with fMNP signaling endosomes was immobilized during and after a 3-min exposure to a 15 pN force (start at time 0). Within 5 min, neurite elongation and central domain advance halted, and lamellar motility in the peripheral (p) domain immobilized. The central domain (c) and distal neurite (n) widened (compare 0′ to 5′). However, filopodial (f) protrusions with small lamella (l) protruded from the central domain and then extended above the immobilized lamellar domain before cycling retrogradely to the base of the growth cone where they were absorbed. During peripheral domain immobilization, fMNP puncta were detectable and moved dynamically in the central domain (compare 8′45′′ and 8′50′′). Approximately 15 min after removing the magnet, protrusive activity at the leading edge resumed in concert with central domain advance. Time in minutes (′) and seconds (′′) is indicated. (Scale bar: 5 μm.) (B) A focal 15-pN force halted neurite elongation in neurites loaded with fMNPs but not unloaded or cMNP-loaded RGCs. The neurite growth rate was unaltered in fMNP-loaded RGCs in the absence of a focal magnetic force. (C) In fMNP-loaded RGCs, lamellar and filopodial initiations, the number of moving lamella and filopodia, and the number of filopodia were all reduced by a focal magnetic force compared with control RGCs with a magnetic force or fMNP-loaded RGCs without a magnetic force. (Values normalized to activity during the first 5 min of recording. In B and C, n = at least 3 per condition; *P < 0.0001).