Sorting of beta-actin mRNA and protein to neurites and growth cones in culture

Abstract

The transport of mRNAs into developing dendrites and axons may be a basic mechanism to localize cytoskeletal proteins to growth cones and influence microfilament organization. Using isoform-specific antibodies and probes for in situ hybridization, we observed distinct localization patterns for beta- and gamma-actin within cultured cerebrocortical neurons. beta-Actin protein was highly enriched within growth cones and filopodia, in contrast to gamma-actin protein, which was distributed uniformly throughout the cell. beta-Actin protein also was shown to be peripherally localized after transfection of beta-actin cDNA bearing an epitope tag. beta-Actin mRNAs were localized more frequently to neuronal processes and growth cones, unlike gamma-actin mRNAs, which were restricted to the cell body. The rapid localization of beta-actin mRNA, but not gamma-actin mRNA, into processes and growth cones could be induced by dibutyryl cAMP treatment. Using high-resolution in situ hybridization and image-processing methods, we showed that the distribution of beta-actin mRNA within growth cones was statistically nonrandom and demonstrated an association with microtubules. beta-Actin mRNAs were detected within minor neurites, axonal processes, and growth cones in the form of spatially distinct granules that colocalized with translational components. Ultrastructural analysis revealed polyribosomes within growth cones that colocalized with cytoskeletal filaments. The transport of beta-actin mRNA into developing neurites may be a sequence-specific mechanism to synthesize cytoskeletal proteins directly within processes and growth cones and would provide an additional means to deliver cytoskeletal proteins over long distances.

Fig. 1.

Localization of actin isoforms within cultured neurons. Cortical neurons cultured for 4 d were double-labeled with an isoform-specific antibody to β-actin or γ-actin (left column) and phalloidin–TRITC (right column). A, B, γ-Actin was distributed throughout the cell body (arrow) and neurites and resembled phalloidin staining. C,D, Localization of β-actin at tips of minor neurites (arrowhead). Low levels of β-actin were observed in the cell body (arrow). Phalloidin labeled actin filaments throughout the cell body (arrow) and minor neurites (arrowhead). E,F, γ-Actin was distributed throughout the axon and growth cone, as was phalloidin staining. G,H, β-Actin was enriched within axonal growth cones (arrows). Only weak labeling was observed in the axon shaft. Not all filopodia were labeled (arrowheads) despite the presence of F-actin (phalloidin). Scale bar, 8.5 μm.

Fig. 2.

Localization of β-actin protein visualized in optical sections via image processing. Cortical neurons cultured for 4 d were double-labeled with phalloidin (rhodamine) and an isoform-specific antibody to β-actin (fluorescein). Images were superimposed after restoration and then registered (see Materials and Methods). Shown here is a single optical section (250 nm) from thez-series. A, Phalloidin labeled actin filaments throughout the cell body and neurites, whereas the β-actin isoform was concentrated within the distal tips of minor processes. The overlap is indicated by the presence of white pixels. B, In distal axons, phalloidin labeling is distributed throughout the neurite and growth cone, whereas the β-actin isoform is localized to the peripheral margin; note the apparent fibrillar distribution within filopodia. Scale bar, 10 μm.

Fig. 3.

Localization of HA–actin protein. Cortical neurons were transfected with an RSV vector containing β-actin bearing an HA epitope tag. A, Detection of HA–actin within the growth cone of a minor process (arrow).B, Differential interference contrast (DIC) optics.C, Detection of HA–actin within an axonal growth cone (arrow) near another neuron. D, DIC optics. E, Schematic drawing showing the location of the HA sequences between the coding region and the 3′-UTR. Scale bar, 10 μm.

Fig. 4.

Intraneuronal distribution of β-actin and γ-actin mRNA. Cortical neurons were cultured for 4 d, at which time most neurons have distinguishable axonal and dendritic processes.A, Hybridization of biotinated probes to γ-actin mRNA within the cell body (arrow). B, Hybridization of digoxigenin-labeled probes to β-actin mRNA within the cell body (arrow). C, Differential interference contrast (DIC) microscopy of the cell body (arrow), minor neurites, and a single axon. The axon is considerably longer than the minor neurites and cannot be photographed in entirety at this magnification. Shown here is the initial segment.D, Absence of γ-actin mRNAs from the axonal growth cone (arrow). E, Localization of β-actin mRNA within the axonal growth cone (arrow).F, DIC image of axonal growth cone (arrow) from this axon. The axon in this cell is ∼150 μm in length. Scale bar, 10 μm.

Fig. 5.

Transport of β-actin mRNA into processes after treatment with db-cAMP. Cortical neurons cultured for 4 d in N2 supplements were transferred to MEM for 3 hr. Cells were fixed and hybridized with digoxigenin-labeled probes specific to β-actin mRNA. Probes were detected by using fluorochrome-conjugated antibodies, and images were acquired with a cooled CCD camera (see Materials and Methods). A, β-Actin mRNA was detected in the cell body, but the signal no longer was observed in growth cones (arrow). C, After 15 min in db-cAMP, β-actin mRNA granules were observed in processes (arrow) but were not yet detectable within growth cones (arrowhead). Shown here is a signal within an axonal process. E, After 1 hr, β-actin mRNA granules were observed within growth cones. Shown here is a hybridization signal within the distal axon (arrowhead) and growth cone (arrow). G, γ-Actin mRNA was confined to the cell body (arrow). No signal was observed within the axonal growth cone (arrowhead). Shown here is a cell after 1 hr of treatment with db-cAMP. B, D, F, H, DIC optics. Scale bar, 10 μm.

Fig. 6.

Visualization of β-actin mRNA by three-dimensional digital imaging microscopy. To visualize actin mRNA in cortical neurons (4 d in culture) with higher resolution than conventional epifluorescence microscopy, we took a series of optical sections (100 or 250 nm) from each cell and further processed them by using deconvolution algorithms and applying a point spread function (Fay et al., 1989). A, Localization of β-actin mRNA within a single optical section (250 nm) of a cell body and minor processes (unprocessed image). The fluorescence intensity within the neurite shaft was low. A concentration of β-actin mRNA was observed within the growth cone. B, The same image after restoration. A punctate distribution was observed throughout most of the processes. Note the concentration of β-actin mRNA within one of the growth cones (arrow). C,D) Localization of β-actin mRNA within an axon and its growth cone. The cell body is at the top of the image, and the axon extends downward, terminating in an elaborate growth cone. Note the concentration of β-actin mRNA granules within the central domain (arrow) and few granules within peripheral regions (curved arrow). Scale bar, 10 μm.

Fig. 7.

Colocalization of β-actin mRNA with microtubules. β-Actin mRNA was detected with rhodamine, and tubulin protein was detected with fluorescein; then the two processed images were superimposed. Pixels that contained both fluorochromes appearedwhite in optical sections, whereas red pixels denote probe that is not within the same pixel as anti-tubulin (green pixels). The majority of β-actin mRNA granules colocalized with microtubules (white pixels). Scale bar, 5 μm.

Fig. 8.

Distance between β-actin mRNA granules and microtubules, as compared with randomized signals. The distance of β-actin mRNA (brightest voxels) to the nearest tubulin voxel was compared with a randomized distribution. This analysis was performed on a three-dimensional data set from 100 nm optical sections. The mean and SD of the random distribution are shown. The observed distribution of β-actin mRNA is significantly closer to the microtubules than a random distribution.

Fig. 9.

Ultrastructural visualization of polyribosomes within growth cones. Cortical neurons were grown on coverslips and processed for thin sectioning parallel to the monolayer, as described in Materials and Methods. A, An axonal growth cone viewed at low magnification. Polyribosomes are observed in the distal segment of the axon and central region of the growth cone. An area of interest (arrow) adjacent to filopodia is photographed at higher magnification in B to visualize the proximity of a polyribosome to a microtubule (arrow).C, Central region of a growth cone from another neuron that contains polyribosomes in clusters (arrowhead) and between microtubules (arrow). D, A large axonal growth cone from a third cell with polyribosomes near microtubules (arrowhead) and in clusters between microtubules (arrow). Note other membranous organelles and mitochondria. Scale bar, 0.5 μm.

Fig. 10.

Colocalization of β-actin mRNA and translational components. Shown are double labeling for β-actin mRNA (rhodamine) and EF1α or 60S ribosomes (fluorescein).A, EF1α and (B) β-actin mRNA in the distal field of an axonal process and its terminal branches. Punctate fluorescence for EF1α colocalized with β-actin mRNA granules (arrows). Image processing indicated that β-actin mRNA and EF1α occupied the same pixel coordinates. Note the identical spacing between the punctate distribution patterns in both rhodamine (β-actin mRNA) and fluorescein (EF1α). A granule containing EF1α does not colocalize with β-actin mRNA (arrowheads). C, D, Ribosomal subunit (60S) and actin mRNA in an axonal growth cone.C, Cluster of four granules that contain 60S protein (arrow) and (D) actin mRNA (arrow). A granule containing the 60S subunit does not contain actin mRNA (arrowheads). Scale bar, 5 μm.