Research Article

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

Michael B. Steketee, Stavros N. Moysidis, Xiao-Lu Jin, Jessica E. Weinstein, Wolfgang Pita-Thomas, Hemalatha B. Raju, Siraj Iqbal, and Jeffrey L. Goldberg
PNAS November 22, 2011 108 (47) 19042-19047; https://doi.org/10.1073/pnas.1019624108

  1. Edited* by Robert Langer, Massachusetts Institute of Technology, Cambridge, MA, and approved October 17, 2011 (received for review January 1, 2011)

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.

Central nervous system (CNS) neurons fail to regenerate after injury or disease because of reduced intrinsic axon growth ability (1, 2), inhibitory molecules (3–6), and deficient neurotrophic factor signaling (7–9). Neurotrophins, like brain-derived neurotrophic factor (BDNF), activate tropomyosin-related kinase B (TrkB) receptors and are endocytosed by clathrin-dependent and -independent mechanisms into signaling endosomes (10, 11). These signaling endosomes signal persistently during retrograde (12) and anterograde (13, 14) transport in axons or dendrites (15, 16) directing neurite growth, survival, and cell migration (17, 18). Signaling endosomes are critical long-range communication links used by neurons in the central and peripheral nervous system during development and regeneration (17), whose dysfunction is linked to nervous system disorders (19–21). Therefore, studying signaling endosome localization and related functions in regulating neurite growth is vital. MNPs are emerging as flexible, multimodal nanoparticles that can be targeted to specific tissues or cells by molecular functionalization. To alter signaling endosome localization, we targeted functionalized MNPs to active TrkB signaling endosomes and demonstrate that magnetically manipulating their localization affects growth cone behavior and neurite growth.

Results

To load MNPs into TrkB signaling endosomes, 50-nm MNPs were functionalized with the anti-TrkB agonist antibody, 29D7, conjugated to Alexa 594. 29D7 activates TrkB and enhances retinal ganglion cell (RGC) survival and neurite growth in vitro and in vivo (22). Now, we show 29D7 facilitates rapid MNP endocytosis into TrkB signaling endosomes. In vivo, 29D7-MNPs (fMNPs) but not control MNPs (cMNP, functionalized with Alexa 594-conjugated IgG antibodies) injected intravitrealy were detected in RGC somas, axons, and dendrites in the retina and RGC axons in the optic nerve (Fig. S1). In primary RGCs cultured and allowed to extend neurites before a 15-min fMNP perfusion, fMNPs were detected as puncta in all (n = 45) growth cones, neurites, and somas (Fig. 1A). FMNP puncta were more prominent in growth cone central domains than peripheral domains (Fig. 1A and Fig. S2) or somas (Fig. S2), indicating selective endocytosis or transport. FMNPs increased phospho-TrkB (p-TrkB) that colocalized with 92% (821/847) of fMNP puncta (Fig. 1A and Fig. S2). In RGCs cultured with BDNF, fMNPs were not detected and failed to increase p-TrkB staining (Fig. 1B), suggesting fMNP endocytosis is TrkB-dependent. In contrast, cMNPs were virtually undetectable in RGC growth cones, neurites, and somas (Fig. 1C; n = 15 total puncta in 45 cells), and neither increased p-TrkB nor colocalized with p-TrkB puncta (4/15 puncta). Magnetic fMNP recovery from RGCs treated overnight in suspension cultures pulled down p-TrkB and fMNPs increased both phospho-ERK1/2 (p-ERK1/2) and phospho-Akt (p-Akt; Fig. 1E), consistent with 29D7 alone (22, 23), whereas cMNPs failed to pull down p-TrkB (Fig. 1D) or increase p-ERK1/2 or p-Akt (Fig. 1E).

Fig. 1.
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.)

To determine whether fMNP signaling endosomes induce neurite growth similar to BDNF, RGCs were incubated with either BDNF or fMNPs with or without the Trk receptor inhibitor K252a, which inhibits BDNF-induced neurite growth in RGCs (24). BDNF and fMNPs induced neurite growth similarly (Fig. S3A), indicating 29D7 bound to MNPs stimulates RGC neurite elongation like non-MNP bound 29D7 (22). Moreover, K252a inhibited fMNP-induced neurite growth significantly (Fig. S3B), indicating RGC neurite growth induced by fMNPs depends on Trk activation. Finally, we analyzed the ability of fMNPs to alter the transcription of genes known to promote neurite growth and to be up-regulated by BDNF (25, 26), including c-Fos, EGR2, KLF5, and GAP43, by quantitative RT-PCR. Both BDNF and fMNPs increased the expression of these genes at both 3- and 12-h time points, again in a K252a-sensitive manner (Fig. S4). Together, these results suggest fMNPs are endocytosed into active, long-lived TrkB signaling endosomes that stimulate neurite elongation, and signaling pathways and gene expression associated with neurite growth.

To determine whether fMNP signaling endosomes were trafficked into nascent neurites, purified RGCs, lacking neurites, were incubated overnight with either fMNPs or cMNPs in BDNF(-) suspension cultures and then plated on laminin-coated coverslips where they could extend new neurites (Fig. 2 A and B). FMNPs and cMNPs were detected in 95% and 7% of RGC somas, and 80% and 0% of RGC neurites and growth cones, respectively (Fig. 2 C and D and Fig. S2; n = 30 each). FMNP signaling endosomes moved bidirectionally in neurites between soma and growth cones independently (Fig. 3 and Movie S1). Fast-moving puncta appeared oblong because of motion blurring during acquisition, whereas slowly moving or stationary puncta were round and exhibited oscillatory movements, differing from puncta bound to the substrate, which failed to move. Transport rates varied from 0 to 12 μm/s, comparable with fast axonal transport (27). Mean transport rates were similar in anterograde (0.22 ± 0.03 μm/s; n = 50) and retrograde (0.23 ± 0.03 μm/s; n = 50) directions (Fig. 3B). FMNP signaling endosomes frequently entered and exited growth cones via the neurite (Movie S1), suggesting regulated fMNP trafficking into and out of growth cones. Thus, purified RGCs endocytose both fMNPs and cMNPs, but only anti-TrkB fMNP signaling endosomes are trafficked and transported in nascent neurites.

Fig. 2.
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.
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).

To alter fMNP signaling endosome transport, defined magnetic forces (Fig. S5) were applied to fMNP-loaded RGCs via an electromagnetic needle. Force was controlled by varying the distance between the needle tip and the RGC. Interestingly, at approximately 15 pN, net fMNP signaling endosome transport halted anterogradely, and fMNP signaling endosomes moved retrogradely out of distal neurites into proximal neurites and somas (Fig. S6 and Movie S2). Over a 20-min exposure, the mean fluorescent light units decreased 90.1 ± 7% in neurites and increased 41.5 ± 4.8% in somas (n = 3). This net retrograde transport was not accompanied by process retraction or evacuation. Other vesicles, mitochondria, and vacuoles continued to move bidirectionally (Movie S3 and Movie S6). Thus, a focal 15-pN magnetic force biased fMNP signaling endosome transport away from growth cones. At forces >15 pN, fMNPs changed from punctate to diffuse in neurites and growth cones, indicating we were unable to pull MNP signaling endosomes toward the magnet without disrupting their integrity (n = 5; Movie S4). Thus, we were unable to pull signaling endosomes toward the magnet without disrupting endosome integrity similar to results in nonneuronal cells (28). We did not further examine either the cellular or molecular effects induced by disrupting signaling endosome integrity.

To determine whether altering fMNP signaling endosome transport affected growth cone motility or neurite growth, a 15-pN magnetic force was applied to unloaded, cMNP-, or fMNP-loaded RGCs. A continuous magnetic force failed to alter growth cone motility or neurite growth in unloaded or cMNP-loaded RGCs (n = 20; Fig. 4A and Movie S5). Because cMNPs were detected in a small percentage of cells (e.g., Fig. 2), cMNPs were a good control for endocytosed, non-TrkB–signaling MNPs. Neither prolonged magnetic forces over 1 h nor increasing the force to 100 pN detectably altered lamellar and filopodial activity or neurite growth rate in either control group (e.g., Fig. 5 B and C). Moreover, in the absence of an external magnetic field, loading fMNPs into RGC signaling endosomes failed to alter either lamellar and filopodial activities or neurite growth rate (Fig. 5 B and C). Thus, neither pN magnetic fields applied to unloaded or cMNP-loaded RGCs, nor fMNP loading into signaling endosomes alone, alters growth cone motility or neurite growth.

Fig. 4.
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.
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).

We next asked whether applying a magnetic force to fMNP signaling endosomes alters growth cone motility or neurite elongation. In contrast to control RGCs, a 15-pN force applied to fMNP-loaded RGCs for 1–5 min was sufficient to halt neurite growth without retraction. (Figs. 4B and 5, Movie S6, and Movie S7). The growth cone's central domain stopped advancing, and both lamellar and filopodial protrusions immobilized in the peripheral domain. Despite peripheral immobilization, protrusion continued in the central domain (Fig. 5A and Movie S6); organelles, including mitochondria and vacuoles, transported in the peripheral and central domains and neurite (Movie S3 and Movie S6); and retrograde actin flow (29) was visible in immobilized filopodia and lamella (Movie S6). Approximately 15–30 min after removing the magnet, protrusion and central domain advance resumed, typical of neurite growth (Movie S8). This delayed recovery indicates the force did not simply hold the growth cone but likely altered signaling presumably because of altered fMNP signaling endosome localization. Interestingly, even as the central domain advanced after recovery, the previous central domain failed to consolidate into neurite and previously immobilized lamella and filopodia failed to remobilize (Fig. 4B and Movie S8). Thus, in fMNP-loaded RGCs, a focal magnetic force immobilizes peripheral domain protrusions concomitant with reversibly inhibiting peripheral protrusive activity and neurite growth.

Finally, to determine whether these changes are a general response to altered fMNP signaling endosome localization, dorsal root ganglion (DRG) cells were loaded with fMNPs. Like RGCs, fMNPs were detected as puncta in the growth cones, neurites, and somas and were transported bidirectionally in DRG neurites (Fig. S7A). Similar to RGCs, a 15-pN magnetic force reversibly inhibited peripheral protrusions and neurite elongation (5/5 cells; Fig. S7B). Thus, regulated signaling endosome localization is likely a general mechanism regulating growth cone motility and neurite growth.

Discussion

This study shows TrkB signaling endosome localization is critical to growth cone motility and neurite elongation by documenting several findings. First, MNPs functionalized with a TrkB receptor agonist that stimulates neurite growth by activating TrkB-dependent signaling pathways are endocytosed into RGC and DRG signaling endosomes and transported in nascent neurites between the soma and growth cones. These fMNP signaling endosomes are similar in size, transported at rates similar to Trk signaling endosomes in PC12 cells (30) and cortical neurons (31), and their transport did not affect growth cone motility or neurite growth rate on their own. Because signaling endosome dysfunction is linked to numerous nervous system diseases (19–21), fMNP signaling endosome studies may provide insight into targeting nanotherapeutics by identifying cargo- or domain-specific adaptor and motor proteins (32), and by identifying the effects nanoconjugation have on the endocytosis and trafficking of antireceptor agonist antibodies (33) that may impact nanotherapeutic efficacy.

Second, a focal magnetic force alters TrkB/fMNP signaling endosome transport without altering transport as a whole, allowing the specific manipulation and study of signaling endosome localization. These studies revealed a difference in anterograde and retrograde transport, because the former was selectively disrupted by a 15-pN magnetic force. Interestingly, the stall forces for kinesin and dynein are in the 7- to 8-pN range (34). Kinesin and dynein transport endosomal cargos along microtubules by unique mechanisms and their stall forces differ depending on cargo load (35, 36). However, the precise mechanisms controlling motor proteins/cargo specification are unknown. Targeting MNPs to endosomal cargoes and novel micromagnetic tools (37) may facilitate finer force application to specific subcellular regions or even individual organelles to better investigate the roles different motor proteins or motor protein isoforms (38) play in transporting specific cargoes.

Third, altering TrkB/fMNP signaling endosome localization in the distal neurite alters lamellar and filopodial motility and turnover in the growth cone's peripheral domain and halts neurite growth, suggesting a linked mechanism. Previous studies showed ERK1/2 and Akt activities are critical to neurite growth promoted by neurotrophins (24, 39). We now extend these studies to show that TrkB signaling endosome localization regulates motility on a finer scale. Altering TrkB signaling endosome transport inhibited normal filopodial and lamellar turnover and protrusion. Because protrusions remained immobilized despite retrograde flow, filamentous actin (F-actin) polymerization was not likely inhibited in the peripheral domain (40–42), and such F-actin polymerization is insufficient to support protrusive initiations (43, 44). These data suggest signaling endosome localization in the distal neurite regulates F-actin–driven protrusive initiations but not F-actin retrograde flow. These data also suggest that altering signaling endosome localization may provide a unique method to study retrograde flow, myosin activity, and protrusive activity in the peripheral domain of the growth cone. Moreover, altered signaling endosome localization may provide a unique method for evaluating current hypotheses on the role that TrkB signaling endosome localization plays in other known signaling endosome-dependent processes such as cell migration (18), synaptic plasticity (45, 46), calcium signaling (47, 48), and learning and memory (49).

Finally, engineered MNPs may provide a flexible delivery and release system controlled by noncontact forces (50, 51), verifiable by MRI (52, 53). Increasingly, MNPs are used in wide-ranging biological and clinical applications including gene (54) and drug delivery (55), tumor targeting (56, 57), intraaxonal labeling (58, 59), stem cell tracking (60), and intraaxonal drug delivery (61). The MNPs studied here are nontoxic to ocular structures when studied in mice (62). Magnetic forces can target MNPs to specific cellular populations (63) and release MNP encapsulated therapeutics at target sites (64). Our result showing that fMNPs that stimulate neurite growth can be endocytosed into RGC cells bodies and then transported anterogradely in neurites both in vitro and in vivo support methods for targeting nanotherapeutics to inaccessible regions of the CNS. In summary, fMNP strategies will expand our knowledge of endosomal trafficking and axon growth and raise the potential for delivering nanotherapeutics to the injured nervous system.

Methods

Animals.

Experiments conformed to the Association for Research in Vision and Ophthalmology Statement for the Use of Animals in Ophthalmic and Vision Research and were approved by the Institutional Animal Care and Use Committee and the Institutional Biosafety Committee at the University of Miami.

Cell Culture.

RGCs were purified from embryonic day 20 to postnatal day 8 Sprague–Dawley rats (Harlan Laboratories) to >99% purity by immunopanning (65), cultured on poly-d-lysine (70 kDa, 10 μg/mL) and 2 μg/mL laminin-coated glass bottom dishes (P35G-1.0-2-C, MatTeK) in H-SATO containing Hibernate-E (BrainBits) with 5 mg/mL insulin, 1 mM sodium pyruvate, 1 mM l-glutamine, 40 ng/mL triiodo-thyronine, 5 mg/mL N-acetyl cysteine, B27, with or without 50 ng/mL BDNF, 10 μg/mL ciliary neurotrophic factor, and 5 μM forskolin. DRGs were dissected from P4 Sprague–Dawley rats and cultured as above with the addition of 25 ng/mL NGF. Unless noted, reagents were purchased from Sigma.

MNP Functionalization and Loading.

Rat anti-mouse IgG1 superparamagnetic MACS MicroBeads (12.5 μL, 50 nm; Miltenyl Biotec) were incubated with Alexa 594-conjugated mouse anti-TrkB agonist antibody (29D7, 10 μg/mL; Wyeth), for 10 min in PBS (200 μL, 0.02% BSA). FMNPs were centrifuged, washed twice with PBS, and resuspended in 1.0 mL of H-SATO. Control MNPs were incubated with goat anti-rat IgG Alexa 594 (Invitrogen) and prepared as above. RGCs were preloaded by incubating ≈5 × 105 cells/mL with MNPs for 12–16 h in suspension cultures under constant agitation at 37 °C. After incubation, fMNP-loaded RGCs were resuspended in fresh H-SATO and cultured on PDL/laminin as above.

Electromagnet.

An electromagnet was constructed as described (66). Briefly, a machined, aluminum cylinder was wrapped 7600 times with 30-AWG magnet wire (RadioShack). The resulting coil was 88 mm in length and 13 mm in diameter. A 3.2-mm diameter, 155-mm-long, 1018 cold rolled steel rod was fitted into the cylinder and sharpened to a point. Constant current was supplied by a 6–12 V DC, 2500 mA AC/DC power adapter (Radio Shack) controlled by a clarostat potentiometer (Honeywell) and measured by a digital multimeter (RadioShack). Force was calibrated with 100 nm MNPs (Chemicell; catalog no. 4202-5) in glycerol (4080 CP; Sigma) because this diameter most closely matched the fMNP endosomes measured (e.g., Fig. 3). To calculate the force, a constant 164 mA current was delivered to the electromagnet and fluorescent time-lapse recordings of MNPs (n = 5) moving to the electromagnet tip were recorded (AxioVision; Zeiss). Stoke's law (F = 6πμRv) was used to calculate the force and generate a power law curve (Fig. S3).

Neurite Growth Assay.

Purified RGCs were cultured with either BDNF or fMNPs with or without K252a (1 μM, Sigma) for 24 h, fixed with 4% paraformaldehyde in PBS for 15 min, washed in PBS, extracted with 0.2% Triton X-100 for 5 min, washed in PBS, labeled with anti-β-tubulin (Tuj1, 1:250; Abcam) overnight at 4 °C, and then incubated with goat anti-mouse Alexa 594 (1:1,000, Invitrogen) for 4 h at room temperature. Six random fields were imaged at 10×, and total neurite length/RGC was measured in Axiovision (Zeiss).

Time-Lapse Microscopy.

During time-lapse recordings, H-SATO volume was increased to 2.7 mL and overlaid with 2.0 mL of mineral oil (Sigma) to minimize evaporation. Cultures were maintained at 37 °C with a heated stage (Zeiss) and recorded with a Zeiss camera at 12 frames/min with differential interference contrast (DIC) optics (Plan Apo 63×/1.40 DM objective; Zeiss).

Western Blots.

Western blots were done as described (22). Approximately 5 × 105 RGCs were incubated in H-SATO with fMNPs, cMNPs, or 29D7 antibody (10 μg/mL), at 37 °C, 10% CO2. Cells were collected by centrifugation, frozen, and stored at −80 °C. Posttransferred polyvinylidene fluoride (PVDF) membranes (Millipore) were incubated for 1 h in block (3% BSA, 0.1% Tween-20 at pH 7.6) and then at 4 °C for 3 d in block with rabbit anti-phospho–Trk (1:500; Cell Signaling Technology; catalog no. 9141), or overnight with anti-Akt (1:1,000, Cell Signaling Technology; catalog no. 9272), anti-phospho-Akt (1:1,000, Invitrogen; catalog no. 700392), anti-phospho–ERK1/2 (1:1,000, Sigma; catalog no. E7028), or anti-ERK1/2 (1:1,000, Sigma; catalog no. 7028). Subsequently, the membranes were incubated with horseradish peroxidase (HRP)-conjugated anti-rabbit IgG (1:2,500; Santa Cruz) for 2 h and developed by HRP chemiluminescence (Pierce). To determine total Trk or β-actin, PVDF membranes were stripped (Bio-Rad) for 5 min and then incubated with either rabbit anti-TrkA (1:500; BD; catalog no. 610101) or mouse anti-β-actin overnight and then redeveloped as above with appropriate secondary antibodies.

Quantitative RT-PCR.

RNA was extracted from 5 × 105 purified RGCs treated with BDNF or fMNPs with or without 1 μM K252a by using RNeasy (Qiagen) and digested with 2.7 U/μL DNase. RNA (200 ng) was reverse transcribed by using iScript cDNA synthesis kit (Bio-Rad) according to the manufacture's instructions. For gene expression analysis, 1 μL of cDNA (equivalent to 5 ng of RNA) was amplified by RT-PCR with primer annealing at 60 °C. Fluorescence was read after the extension step at 72 °C on an iQ5 multicolor RT-PCR detection system (Bio-Rad) by using the qPCR master mix for SYBR Green (iQ SYBR Green; Bio-Rad). Relative expression levels were normalized to 18S. The primer sets used were as follows: Egr2: forward 5′-TGAGATGAAG CTCCAGCTGACACA-3′, reverse 5′-AAGAACACAGAAGGGCGGTAGTGT-3′; c-Fos: forward 5′-ACGGAGAATCCGAAGGGAAAGGAA-3′, reverse 5′-TCTGCAA CGCAGACTTCTCGTCTT-3′; Gap43: forward 5′-TAAGAAACGGCTTTCCACGTTGCC-3′, reverse 5′-TAAGCCACACTGTTGGACTTGGGA-3′; KLF5: forward 5′-ACCTCCGTCCTATGCTGCTACAAT-3′, reverse 5′-TCGGACAGGTTGGATATTTGGCGA-3′; 18S: forward 5′-GAACTGAGGCCATGATTAAGAG-3′, reverse 5′-CATTCTTGGCAAATGCTTTC-3′.

Intravitreal Injections.

Adult female Sprague–Dawley rats were anesthesized by i.p. injection of 60 mg/kg ketamine and 7.5 mg/kg xylazine. Proparacaine hydrochloride 0.5% (Falcon Pharmaceuticals) was administered to eyes before fMNP or cMNP intravitreal injection. Buprenorphine (0.05 mg/kg; Bedford Laboratories) was administered as a postoperative analgesic. Rats were perfused after 24 h with 4% paraformaldehyde (PFA). Eyes were removed, postfixed, and cryoprotected by immersion in 4% PFA, 30% sucrose overnight. Eyes were embedded in O.C.T. (Sakura Finetek) and frozen in isopentane (Spectrum Chemicals) on dry ice. Ten-microgram cryosections were thaw-mounted on Superfrost Plus slides. To detect MNPs, sections were incubated in 1% hydrogen peroxide, 0.2% Trition X-100 in PBS for 30 min, washed in PBS, incubated overnight with goat anti-mouse-HRP or donkey anti-goat-HRP (1:1,000; Santa Cruz Biotechnology), and then amplified by tyramide signal amplification (Invitrogen; T20922).

Image Analysis.

To characterize fMNP endosome responses to a focal magnetic force, we analyzed at least 10 RGC neurites from both fMNP-loaded and non-fMNP-loaded RGCs. The number of arbitrary fluorescence light units (FLU) was measured in Axiovision at 0 and 20 min. To monitor effects on the neurites, cells were recorded with DIC optics in between fluorescent exposures at 12–15 frames per min. To quantify the fMNP response to a focal magnetic force, three fMNP-loaded neurites exposed to a constant 15-pN force and three fMNP-loaded neurites exposed to fluorescent light only were analyzed. FLU units were measured in Axiovision. Exposure times were matched in each experiment. For presenting p-Trk colocalization data in Figs. 1 and 2 and Fig. S2, all images were processed identically in Photoshop (Adobe) from experiments conducted in parallel. All other images were processed in Photoshop to enhance contrast for easier visualization during publication without altering data integrity.

To quantify responses to a focal magnetic force, we selected the first three cells from each experimental group, unloaded, cMNP-loaded, and fMNP-loaded, that met the following criteria: (i) The recorded period lasted at least 15 min, and (ii) the recording was sufficiently clear to visualize all motile elements over the recorded period. To compare data from growth cones differing in both size and activity levels, data from individual growth cones were normalized by expressing the values as a percentage of the first 5 min of exposure to the magnetic field. Growth cone protrusions were analyzed as described (67). Briefly, the following lamellipodial and filopodial activities were measured over 2.5-min intervals: (i) moving lamella extending or retracting; (ii) lamellar and (iii) filopodial initiations that extended at least 1 μm; (iv) filopodial number; and (v) moving filopodia exhibiting shortening, elongation, or lateral movements.

Acknowledgments

We benefited from generous gifts of anti-TrkB antibody, clone 29D7, from Wyeth Pharmaceuticals and anti-phospho–TrkB antibody from Moses Chao, and we thank Alan Halpern for intellectual discussions, Eleut Hernandez for excellence in animal husbandry, and Gabriel Gaidosh for expert microscopy. We thank the National Institutes of Health for Grants EY017971 and NS061348 (to J.L.G.), P30-EY014801 (to the University of Miami), and National Research Service Award T32NS007044 (to M.B.S.), a grant from Orthopedic Development, Inc. (to J.L.G.), and an unrestricted grant from Research to Prevent Blindness (to the University of Miami).

Footnotes

  • ↵1To whom correspondence should be addressed. E-mail: jgoldberg{at}med.miami.edu.

  • Author contributions: M.B.S. and J.L.G. designed research; M.B.S., S.N.M., X.-L.J., J.E.W., W.P.-T., and H.B.R. performed research; M.B.S., S.N.M., X.-L.J., and S.I. analyzed data; and M.B.S. and J.L.G. wrote the paper.

  • The authors declare no conflict of interest.

  • ↵*This Direct Submission article had a prearranged editor.

  • This article contains supporting information online at www.pnas.org/lookup/suppl/doi:10.1073/pnas.1019624108/-/DCSupplemental.

Back to top
Article Alerts
Email Article
Citation Tools
Request Permissions
Nanoparticle-mediated signaling endosome localization regulates growth cone motility and neurite growth
Michael B. Steketee, Stavros N. Moysidis, Xiao-Lu Jin, Jessica E. Weinstein, Wolfgang Pita-Thomas, Hemalatha B. Raju, Siraj Iqbal, Jeffrey L. Goldberg
Proceedings of the National Academy of Sciences Nov 2011, 108 (47) 19042-19047; DOI: 10.1073/pnas.1019624108
Digg logo Reddit logo Twitter logo Facebook logo Google logo Mendeley logo
Proceedings of the National Academy of Sciences: 108 (47)
Table of Contents

Submit

Article Classifications

Jump to section

You May Also be Interested in

Multi-color molecular model
Enzymatic breakdown of PET plastic
A study demonstrates how two enzymes—MHETase and PETase—work synergistically to depolymerize the plastic pollutant PET.
Image credit: Aaron McGeehan (artist).

Similar Articles