IMAGES

Kinesin-5 is Essential for Growth-Cone Turning. (Current Biology 2008 Dec 23.)

The Microtubule-severing Proteins Spastin and Katanin Participate Differently in the Formation of Axonal Branches (Mol. Biol. Cell 2008 19:1485-98.)

Quantitative and Functional Analyses of Spastin in the Nervous System: Implications for Hereditary Spastic Paraplegia (J. Neuroscience 2008 28:2147-57.)

Kinesin-5 regulates the growth of the axon by acting as a brake on its microtubule array (J Cell Biology 2007 178:1081-91.)

Tau protects microtubules in the axon from severing by katanin. (J Neurosci 2006 26:3120-9.)

Antagonistic forces generated by cytoplasmic dynein and myosin-II during growth cone turning and axonal retraction. (Traffic 2006 7:1333-51.)

Effects of dynactin disruption and dynein depletion on axonal microtubules. (Traffic 2006 7:524-37.)

Regulation of Microtubule Severing by Katanin Subunits during Neuronal Development (J Neurosci 2005 25:5573-5583.)

Role of Cytoplasmic Dynein in the Axonal Transport of Microtubules and Neurofilaments (J Cell Biol 2005 168: 697-703.)

Monastrol, a Prototype Anti-Cancer Drug that Inhibits a Mitotic Kinesin, Induces Rapid Bursts of Axonal Outgrowth from Cultured Postmitotic Neurons (Cell Motil Cytoskeleton 2004 58: 10-16.)

Role of Actin Filaments in the Axonal Transport of Microtubules (J Neurosci 2004 24: 11291-11301.)

Distribution of the Microtubule-Related Protein Ninein in Developing Neurons (Neuropharm 2004 47: 677-683.)

Axonal Growth is Sensitive to the Levels of Katanin, a Protein that Severs Microtubules (J Neurosci 2004 24: 5778-5788.)

Expression of the Mitotic Kinesin Kif15 in Postmitotic Neurons: Implication for Neuronal Migration and Development (J Neurocytol 2003 32: 79-96.)

Microtubule Reconfiguration during Retraction induced by Nitric Oxide (J Neurosci 2002 22: 5982-5991.)

Microtubule Reconfiguration during Axogenesis (Journal of Neurocytology, 2002, in press)

Growing and Working with Peripheral Neurons (Methods in Cell Biology, in press)

Motor proteins regulate force interactions between microtubules and microfilaments in the axon (Nature Cell Biology 2000 2, 276 - 280.)

An Essential Role for Katanin in Severing Microtubules in the Neuron (J. Cell Biol. 1999 145: 305-315)

Cytoplasmic Dynein and Dynactin Are Required for the Transport of Microtubules into the Axon (J. Cell Biol. 1998 140: 391-401)

Depletion of a Microtubule-Associated Motor Protein Induces the Loss of Dendritic Identity (J. Neurosci. 2000 20: 5782-5791.)

Reorganization and Movement of Microtubules in Axonal Growth Cones and Developing Interstitial Branches (J. Neurosci. 1999 19: 8894-8908.)

Expression of the Mitotic Motor Protein Eg5 in Postmitotic Neurons: Implications for Neuronal Development (J. Neurosci. 1998 18: 7822-7835.)

Identification of a Microtubule-associated Motor Protein Essential for Dendritic Differentiation (J. Cell Biol. 1997 138: 833-843)

Inhibition of a Mitotic Motor Compromises the Formation of Dendrite-like Processes from Neuroblastoma Cells (J. Cell Biol. 1997 136: 659-668)

Expression of a Kinesin-Related Motor Protein Induces Sf9 Cells to Form Dendrite-Like Processes with Nonuniform Microtubule Polarity Orientation (J. Neurosci. 1996 16: 4370-4375.)

Kinesin-5 regulates the growth of the axon by acting as a brake on its microtubule array (J Cell Biology 2007 178:1081-91.)

Figure 1: Kinesin-5 depletion results in increased axonal length. (A), Western blot showing kinesin-5 protein levels in cultured sympathetic neurons treated with control siRNA (left) and kinesin-5 siRNA (right). Bottom panels show the GADPH loading control for both protein samples. (B), Quantification of mean axonal outgrowth reveal significantly longer axons of kinesin-5-depleted neurons at all time-points observed, with a 4.3 fold average increase in axonal length (mean ± SEM; 3 hr: control = 21.3 ± 6.45 µm, experimental = 82.8 ± 19.7 µm; 5 hr: control = 71.8 ± 20.0 µm, experimental = 335 ± 66.8 µm; 7 hr: control = 187 ± 37.1 µm, experimental = 774 ± 120 µm; *, p < 0.01) (n = 17 control siRNA, n = 21 kinesin-5 siRNA). (C), Fluorescence images of kinesin-5 immunostaining in control siRNA (top panels) and kinesin-5 siRNA treated neurons (bottom panels). The corresponding microtubule staining for these two cells is shown in the panels on the right. (D), DIC images of a control siRNA neuron taken at 3, 5, and 7 hours after re-plating. (E), DIC images of a kinesin-5-depleted neuron at 3, 5, and 7 hours after re-plating reveal the notable increases in axonal length quantified in (B). Scale bar = 20 µm.

Figure 2: Axonal branching is enhanced when kinesin-5 is depleted. (A-B), DIC images of typical control (A) and kinesin-5-depleted (B) neurons. Note the increase in branch number in axons depleted of kinesin-5. (C), Quantitative analysis of axonal branch number (normalized to total axonal length) revealed a statistically significant increase when compared to control siRNA treated neurons (mean ± SEM; control siRNA = 0.006 ± 0.002, kinesin-5 siRNA = 0.022 ± 0.002; p = 5.76 x 10-8). (D), Categorical analysis at various branching loci (n = 27 control siRNA; n = 37 kinesin-5 siRNA) revealed a statistically significant increase in branching frequency at all branch positions (mean ± SEM; Primary branch: control = 0.006 ± 0.002, kinesin-5 siRNA = 0.013 ± 0.002, *, p < 0.02; Secondary branch: control = 0.002 ± 0.001, kinesin-5 siRNA = 0.004 ± 0.001, *, p < 0.03; Tertiary branch: control = 0.001 ± 0.0004, kinesin-5 siRNA = 0.003 ± 0.001, *, p < 0.03; Quaternary branch: control = 0.00 ± 0.00, kinesin-5 siRNA = 0.001 ± 0.0005, *, p < 0.007). Scale bar = 20 µm.

Figure 3: Depletion of kinesin-5 reduces the distance of axonal retraction. (A-B), DIC images of control axons (A) and kinesin-5-depleted axons (B), before and 30 minutes after treatment with 0.3mM noc-7. (B) shows an example of kinesin-5-depleted axons that continued to elongate even in the presence of noc-7. Arrows demarcate the distal tips of growth cones before and after noc-7 treatment (C), Quantitative analysis of noc-7 induced axonal retraction revealed a statistically significant reduction in mean distance retracted in neurons depleted of kinesin-5 (mean ± SEM; control siRNA 29 ± 2.87 µm; kinesin-5 siRNA = 19.8 ± 2.45 µm; *, p = 0.011) (n = 54 control, n = 55 kinesin-5 siRNA). Scale bar = 20 µm.

Figure 4: Depletion of kinesin-5 enhances the transport frequency of short microtubules. (A), Time-lapse images of a neuron expressing GFP-tubulin reveal a short microtubule moving in the anterograde direction through a photobleached axon. White arrows mark the leading and trailing ends of the microtubule. (B), Quantification of short microtubule transport frequency using siRNA-based depletion of kinesin-5 resulted in the increased frequency of bi-directional short microtubule transport (anterograde: control siRNA = 0.752 events/minute; kinesin-5 siRNA = 1.35 events/minute; retrograde: control siRNA = 0.400 events/minute; kinesin-5 siRNA = 0.705 events/minute; *, p � 0.025). (C), Inhibition of kinesin-5 with monastrol produced a result similar to siRNA-based depletion of kinesin-5, with an increase in bi-directional transport frequency to levels approximately two-fold higher than observed in control neurons (anterograde: DMSO = 0.801 events/minute; monastrol = 1.438 events/minute; retrograde: DMSO = 0.458 events/minute; monastrol = 0.788 events/minute *, p < 0.01). Scale bar = 5 µm.

Figure 5: Depletion of kinesin-5 motor does not affect axonal transport of membranous elements. (A), Time-lapse images of rhodamine-dextran labeled vesicles being transported retrogradely within the axon. White arrows demarcate two mobile vesicles observed during the period of imaging. (B), Quantification of vesicular transport direction and frequency revealed no difference between control and kinesin-5-depleted neurons (mean ± SEM; anterograde: control = 0.983 ± 0.357, kinesin-5 = 1.33 ± 0.262; retrograde: control = 1.32 ± 0.201, kinesin-5 = 1.19 ± 0.206). (n = 14 control axons; n = 14 kinesin-5 depleted axons). (C), Analysis of mitochondrion transport revealed that the depletion of kinesin-5 does not have an effect on mitochondrion length (mean length ± SEM; control = 3.10 ± 0.232 µm, kinesin-5 siRNA = 2.55 ± 0.206 µm) or distance of transport (mean ± SEM; control = 3.91 ± 1.05 µm, kinesin-5 siRNA = 3.37 ± 0.742 µm). n = 38 control siRNA, n = 39 kinesin-5 siRNA. (D), Time-lapse images of a Mito-tracker FM-labeled mitochondrion being transported anterogradely within the axon. White arrowheads mark the moving mitochondrion. Scale bars = 5 µm.

Figure 6: Kinesin-5 depletion results in fewer bouts of retraction of axons and axonal branches. (A), Time-lapse, phase-contrast images of a control (left panels) and a kinesin-5-depleted neuron (right panels). Notice that while control axons form early branches, the branches typically undergo retraction events (left panel arrows). Kinesin-5-depleted branches display robust growth and are typically observed to elongate (right panel arrows). (B), Quantitative comparison of axonal growth profiles revealed a significant shift toward stepwise bouts of growth in axons depleted of kinesin-5 (*, p � 0.001; chi-square). (C), Analysis of branching profiles revealed a similar and significant shift toward stepwise bouts of growth when kinesin-5 was depleted (*, p � 0.001; chi-square). (D-E), Data distribution graphs of the stepwise growth/retraction of control and kinesin-5 depleted axons (D) and branches (E). Scale bar = 20 µm.

Figure 7: Axonal growth and microtubule transport frequency are affected by kinesin-5 overexpression. (A-B), DIC (top panels) and corresponding anti-GFP immunocytochemistry (bottom panels) in neurons transfected with a wild-type EGFP-kinesin-5 construct demonstrating that axon length is coordinated with the level of EGFP-kinesin-5 expression. (C), Quantitative analysis of mean axonal length revealed a statistically significant reduction in mean axonal length in neurons expressing EGFP-kinesin-5 when compared to GFP-control axons (mean ± SEM; EGFP-kinesin-5: 3hr = 25.6 ± 6.38 µm, 5hr = 48.1 ± 10.7 µm, 7hr = 98.1 ± 18.2 µm; *, p < 0.0005). (D), Analysis of short microtubule transport frequency in EGFP-kinesin-5 expressers revealed a significant reduction in anterograde transport, but no change in retrograde transport frequency (anterograde: GFP-control = 0.779 events/minute; EGFP-kinesin-5 = 0.408 events/minute; retrograde: GFP-control = 0.429 events/minute; EGFP-kinesin-5 = 0.449 events/minute *, p � 0.01). Scalebars in (A) and (B) = 30 µm.

Figure 8: Kinesin-5 elicits it effects in a force-dependent manner. Inhibition of kinesin-5 by monastrol treatment or depletion of kinesin-5 by siRNA produces mean axonal lengths that are approximately four times greater that neurons treated with DMSO (control) or control siRNA. When a wild-type kinesin-5 construct is introduced into control siRNA treated neurons, there is a corresponding diminution in mean axon length. Parallel experiments using a rigor mutant (T112N) kinesin-5 construct in control siRNA treated neurons (green bar) revealed that mean axonal lengths were indistinguishable from the control siRNA alone group (yellow bar). A similar effect was observed in kinesin-5-depleted neurons that were rescued by introduction of the wild-type EGFP-kinesin-5 (light blue bar), with mean axonal lengths that were statistically similar to control siRNA neurons expressing endogenous levels of the motor protein (mean ± SEM = 257 ± 46.1 µm, wild type rescue; 187 ± 37.1 µm, control siRNA alone; p > 0.05, 2-tailed t-test). Finally, in kinesin-5-depleted neurons which we attempted to rescue with the EGFP-kinesin-5 mutant (dark blue bar), the mean lengths of axons were found to be nearly 3 times longer than control siRNA treated group, and more than twice the length of kinesin-5 depleted neurons rescued with wild-type kinesin-5 (mean ± SEM = 546 ± 86.5 µm, rigor-mutant rescue; 257 ± 46.1 µm, wild-type rescue, p < 0.02; 187 ± 37.1 µm, control siRNA alone, p < 0.001, 2-tailed t-test). *, p < 0.02.

Tau protects microtubules in the axon from severing by katanin. (J Neurosci 2006 26:3120-9.)

Figure 1. Microtubules are not protected from P60-katanin-induced severing by Taxol. A and E show expression of GFP­P60-katanin (in green), whereas B, D, and F show immunostains for microtubules. A, B, Two cells, only one of which is expressing GFP­P60-katanin. The nonexpresser shows a normal dense splayed microtubule array, whereas the expresser shows far less microtubule mass and only a scattering of very short microtubules. C, D, A cell not expressing GFP­P60-katanin but treated with Taxol. A portion of the microtubules appears as thick bundles. E, F, A cell expressing GFP­P60-katanin in the presence of Taxol. Note that the microtubule mass is severely reduced, and both bundled and unbundled microtubules are very short. Scale bar: (in F ) A­F, 30 µm.

Figure 2. MAP2c protects microtubules from being severed by P60-katanin, but MAP1b does not. A, C, E, G, GFP­P60-katanin in green and immunostain for either plasmid-expressed MAP2 or MAP1b in red. B, D, F, H, Immunostains for microtubules. A, B, Cells that are not overexpressing P60-katanin. As shown in A and B, MAP1b expression does not cause abnormal bundling of microtubules. As shown in C and D, a cell overexpressing MAP1b and P60-katanin shows only a scattering of very short microtubules and severely reduced microtubule levels. (A neighboring cell not expressing MAP1b or P60-katanin displays a normal microtubule array). As shown in E and F, MAP2c expression causes the formation of dense bundles of microtubules. As shown in G and H, the microtubules in MAP2c-expressing cells show no indication of severing by overexpression of P60-katanin, and the microtubule mass is not reduced. Scale bar: A­H, 30 µm.

Figure 3. Results obtained with various tau constructs expressed together with P60-katanin. Cells in the top row were transfected with the tau constructs alone. Cells in the bottom row were transfected with the tau constructs together with P60-katanin. Four-repeat tau generates microtubule bundling (A) and no indication of microtubule severing (a). The microtubule-binding domain alone generates some bundling, but not as much as the four-repeat isoform (B), and provides some protection against severing (b), but not as much as the four-repeat or the three-repeat (not shown in figure) full-length versions of tau. Two versions of tau that lack the microtubule-binding domain, one naturally occurring (C, c) and one generated by experimental truncation (D, d), show no protection against severing. w/o, Without. Scale bar: (in d) 30 µm.

Figure 5. Quantitative analysis of MAP levels after treatment with siRNAs. Control siRNA and siRNAs to MAP2, MAP1b, and tau were transfected into hippocampal neurons. Cultures were fixed 1, 3, and 5d, respectively, after transfection and immunostained for each MAP to evaluate its level as a result of the siRNA. Immunofluorescence indicates depletion of MAP2, MAP1b, and tau in siRNA-treated neurons (b, d, f ), compared with control siRNA-treated neurons (B, D, F ), is shown. A, C, E, The quantification of total fluorescence intensity within the soma revealing the progressive loss of MAP2, MAP1b, and tau depletion by siRNA. By day 5, the loss of MAP2, MAP1b, and tau in siRNA treated neurons are 95, 99, and 99%, respectively. G, Western blot of whole-cell extracts probed with the MAP2, MAP1b, and tau Ab, confirming the protein-lowering effect. GAPDH was used as the internal control. Error bars represent SE.

Figure 7. Depletion of tau from cultured hippocampal neurons increases sensitivity to katanin-induced microtubule severing, whereas depletion of MAP1b or MAP2 does not. A­F, Microtubule immunostains of stage III hippocampal neurons. A, A neuron transfected with control siRNA and GFP. B, A neuron transfected with control siRNA and P60-katanin. C, A neuron transfected with MAP1b siRNA and GFP. D, A neuron transfected with MAP1b siRNA and P60-katanin. E, A neuron transfected with tau siRNA and GFP (axon is directed upwards in the panel). F, A neuron transfected with tau siRNA and P60-katanin. Note that the axons show no diminution in fluorescence intensity as a result of overexpression of P60-katanin in either control siRNA or MAP1b siRNA-treated neurons. However, a dramatic diminution was detected in tau-depleted neurons overexpressing P60-katanin. a­f, Glow-scale pseudocolored images of the axons in A­F, with white indicating the highest level, purple indicating the lowest level, and shades of red, orange, and yellow indicating intermediate levels. G, The quantification of microtubule mass in cell bodies, in minor processes, and in axons of stage III neurons in each group. Overexpression of P60-katanin in either control group or the three MAP siRNA-treated groups leads to a 33­40% diminution in microtubule mass from cell bodies (p < 0.01). Overexpression of P60-katanin in either control group or the three MAP siRNA-treated groups leads to a 60­71% diminution in microtubule mass from minor processes (p <0.0001). The microtubules in axons of tau depleted neurons were very sensitive to overexpression of P60-katanin, leading to a 63% diminution (p < 0.001). The microtubules in axons of the other three groups showed no diminution as a result of the overexpression of P60-katanin (p >0.05). Error bars represent SE. AFU, Arbitrary fluorescence units; Kat, katanin.

Antagonistic forces generated by cytoplasmic dynein and myosin-II during growth cone turning and axonal retraction. (Traffic 2006 7:1333-51.)

Figure 1: Dynein depletion enhances nitric oxide-induced axonal retraction. Phase-contrast images show neuronal morphologies immediately before (A and D) and 30 min after (B and E) exposure to a nitric oxide donor. A control-siRNA-treated neuron (A and B) retracted its axons over significant distances in response to nitric oxide, showing typical ‘retraction bulbs’ at axon tips (described in Results). However, a DHC-depleted neuron (D and E) retracted its axons over much greater distances than the control neuron and showed more dramatic retraction bulbs at axon tips. Panels (C) and (F) are the microtubule staining of the control neuron and the DHC-depleted neuron, respectively. In both cases, microtubules are abundant in retracted axons and appear as coiled arrays with no apparent depolymerization during the retraction (see inset in F). G) Quantification of the percentage of axons retracted in response to nitric oxide among control and DHC-depleted neurons. H) The average distance (in microns) retracted per axon is quantified and compared among different groups of neurons. Distances retracted by DHC-depleted axons are approximately two times greater than those in control neurons (*p < 0.05, two-tailed t-test). Bar, 10 µm.

Figure 2: Growth cone turning is diminished in dynein-depleted growth cones. Growth cone behaviors of control neurons (A­H). Neurons cultured after transfection with control or DHC siRNA for 2 days were replated onto patterned laminin substrates and allowed to extend axons for ~15 h. The cultures were then fixed and stained for laminin (red) and microtubules (green), and growth cones at borders between laminin-containing and laminin-free zones were imaged and then scored for turning as described in the text. The figure shows examples of control growth cones scored as turning (A­D) or not turning (E­H). Categorization and quantification of growth cone turning revealed a 58% reduction in the frequency of turning in DHC-depleted growth cones compared to control-siRNA-treated growth cones (I). Bar, 8.3 µm.

Figure 3: Representative growth cones at borders of laminin-containing and laminin-free zones stained to reveal microtubules (purple), actin filaments (green), and laminin (laminin-free zones are dark pink). In control growth cones (A and B), microtubules extend from the central domain of the growth cone into the peripheral domain and align with actin bundles of filopodia (arrowheads in A and B). In contrast, in dynein-depleted growth cones (C and D), microtubules are confined to the central domain, and examples of microtubules extending into the peripheral domain and aligning with actin bundles of filopodia are much less common than in controls. (A) and (C) are examples of non-spread growth cones, whereas (B) and (D) are examples of spread growth cones. Bar, 8.3 ?m.ones (I). Bar, 8.3 µm.

Figure 5: Microtubule advance and invasion of growth cone filopodia are markedly reduced in dynein-depleted growth cones. (A) and (B) Still-frame images extracted from live cell movies of a control siRNA (A) and DHC siRNA growth cone (B). In control growth cones, EGFP-EB3 comets frequently enter filopodia and commonly extend toward filopodia tips, while in dynein-depleted growth cones, EGFP-EB3 comets fail to enter filopodia (arrows in A and B). (C) Quantification of filopodial invasion by EGFP-EB3 comets reveals a significant decrease in the frequency of microtubule entry into dynein-depleted growth cones over a 5-min imaging period (*, p < 0.001, chi-square analysis). (D) Measurements of EB3 excursion depth reveal that, on the average, comet excursions proceed significantly farther toward the distal tip of growth cone filopodia in control-siRNA-treated neurons than do comet excursion in dynein-depleted neurons [*, p < 0.05, two-tailed Student’s t-test; mean distance ± SEM = 2.500 ± 0.313 ?m (control), and 1.433 ± 0.303 ?m (DHC)]. (E) Analysis of excursion velocities show that EB3 comets move significantly slower in dynein-depleted growth cones compared to control-siRNA-treated growth cones [*, p < 0.00001, two-tailed Student’s t-test; mean velocity ± SEM = 0.095 ± 0.005 ?m (control); 0.065 ± 0.004 ?m (DHC)]. (F) Kymographs of EGFP-EB3 excursion in neurons treated with control siRNA and DHC siRNA. In control growth cones, EB3 comets display progressive and consistent forward movements, while in dynein-depleted growth cones, EB3 comets remain near their starting positions, displaying bouts of assembly that are unable to advance the position of the polymerizing microtubule. Bar, 5 µm.

Figure 6: Depletion of DHC does not affect the rates of retrograde actin flow. (A) Still-frame images extracted from live cell movies of a control-siRNA-treated growth cone. The three time-points depicted show the progressive movement of polyethylenamine-coated beads undergoing retrograde transport from the distal growth cone toward the central region of the growth cone. Arrows demarcate three beads that underwent substantial movement during the observed time period. (B) Quantification of retrograde flow data revealed that there is no statistically significant variation in the velocities of retrograde movement in neurons treated with control or DHC siRNA (p > 0.1, two-tailed t-test; n = 47 control, n = 51 DHC). Arrows demarcating actin rich filopodia in the top panels were transcribed onto tubulin immunostaining images (bottom panels) to indicate microtubule/actin colocalization. Bar, 5 µm.

Figure 7: Inhibition of myosin-II results in alterations in microtubule organization in dynein-depleted growth cones. Neurons were depleted of DHC using siRNA and then treated with 50 ?M blebbistatin for 30 min. Cultures were then double stained to reveal actin filaments (top panels) and microtubules (bottom panels). With vehicle (DMSO) alone, dynein-depleted growth cones displayed actin that was normally distributed (A) and microtubules that were generally twisted and confined to the central zone of the cone (a). After blebbistatin treatment, the microtubules were no longer twisted but extended forward into peripheral zone to invade virtually every actin rich extension or filopodium (B/b/C/c). Arrows demarcating actin rich filopodia in the top panels were transcribed onto tubulin immunostaining images (bottom panels) to indicate microtubule/actin colocalization. Bars, 10 µm.

Figure 8: Inhibition of myosin-II relieves inhibition of microtubule advance into dynein-depleted growth cones and promotes microtubule invasion into filopodia. (A) and (B) Still-frame images extracted from live cell movies of a dynein-depleted (A) and dynein depleted, blebbistatin-treated growth cone (B). In neurons depleted of dynein alone, mRFP-EB3 comets appear disoriented and fail to enter growth cone filopodia, while in dynein-depleted growth cones treated with blebbistatin, mRFP-EB3 comets regain the ability to enter growth cone filopodia (arrows in A and B). (C) Analysis of mRFP-EB3 comet entry into filopodia revealed that the frequency of filopodial invasion in blebbistatin-treated, dynein-depleted growth cones is enhanced compared to growth cones treated with blebbistatin and control siRNA [frequency = 4.50 events/filopodium (DHC siRNA + blebbistatin); frequency = 2.58 events/filopodium (control siRNA + blebbistatin)]. These invasion frequencies are similar to those measured in neurons treated with control siRNA alone. (D) The combination of dynein depletion and blebbistatin treatment resulted in excursion depths statistically similar to blebbistatin-treated control neurons [p > 0.3; 5.44 ?m (DHC siRNA + blebbistatin); 3.71 ?m (control siRNA + blebbistatin)]. (E) mRFP-EB3 velocity analysis show that myosin-II inhibition increases comet velocities in both control and dynein-depleted growth cones when compared to siRNA treatment alone, while no significant difference is observed between controls and dynein-depleted growth cones that have been treated with blebbistatin [p > 0.05; mean ± SEM = 0.201 ± 0.013 ?m (control + blebbistatin), and 0.150 ± 0.008 ?m (DHC + blebbistatin)]. (F) Kymographs of mRFP-EB3 comets reveal that blebbistatin treatment relieves the inhibition of microtubule advance observed when DHC is depleted; note that the kymograph of a comet after DHC depletion and blebbistatin treatment is similar to the kymographs after treatment with control siRNA alone (Figure 5F). Bar, 5 µm.

Figure 9: Depletion of DHC decreases distal growth cone microtubule mass when microtubule assembly is inhibited. (A/a) Images of control-siRNA-treated growth cones in the presence of vinblastine (Vin) alone to inhibit microtubule (MT) assembly. (B/b) Control-siRNA treated growth cones with a combined treatment of vinblastine and blebbistatin (Bleb) to inhibit microtubule assembly and myosin-II activity. In control growth cones treated with siRNA/vinblastine, the microtubules extend into the distal regions of the cone but do not appear to extend into filopodia. The combined vinblastine/blebbistatin treatment enhances distal microtubule levels, and microtubules can be clearly seen within growth cone filopodia. When DHC is depleted in the presence of vinblastine, the levels of distal microtubules are greatly reduced and appear primarily constrained to the central zone of the growth cone (C/c). Treatment with vinblastine/blebbistatin enhances levels of distal microtubules and facilitates their extension toward and into growth cone filopodia (D/d). (E) Graph showing that microtubule dynamics are suppressed significantly and proportionally with increasing concentrations of vinblastine, as measured by the frequency of EGFP-EB3 comets appearing within the growth cone. Statistical comparison of vinblastine titrations with 0.1% DMSO revealed that each of the drug treatments resulted in a significant reduction in the number of EGFP-EB3 comets with respect to time (*, p < 0.00003, two-tailed t-test), with the 50 nM vinblastine treatment resulting in the most substantial suppression of microtubule dynamics (p < 0.000004; 0.1% DMSO versus 50 nM vinblastine; two-tailed t-test). (F) Graph representing the relative levels of growth cone microtubule mass in the presence of vinblastine. Measurements of microtubule mass were calculated per unit area for the distal 25% of the growth cone and revealed that under conditions of DHC depletion alone, there was a 43.7% reduction in microtubule mass compared to neurons treated with control siRNA (p < 0.00001, two-tailed t-test; control, n = 17; DHC, n = 16). Addition of blebbistatin restored distal growth cone levels of microtubules to similar levels in cultures treated with both control (n = 17) and DHC siRNA (n = 15). These increases in distal microtubule mass were significantly different from the DHC siRNA group in the absence of blebbistatin (*, p < 0.0001, two-tailed t-test) but was not statistically different from the control siRNA group (p > 1, two-tailed t-test). A, B, C and D depict microtubules (violet) and phalloidin (green) staining. a, b, c and d represent the original anti-B-tubulin and Cy3-immunofluorescence images. Bar, 10 µm.

Effects of dynactin disruption and dynein depletion on axonal microtubules. (Traffic 2006 7:524-37.)

Figure 1. Axonal morphology and Golgi dispersion in cultured rat sympathetic neurons as a result of dynactin disruption. Panels A and D show a control neuron and P50-dynamitin-myc-expressing neuron (high expresser), respectively, stained with an antibody to myc. Shown here is a control neuron that was not transfected for myc, but for quantitative analyses, we routinely used control neurons transfected to express the myc-tag. Immunofluorescence labeling for tubulin reveals microtubule organization within the control (B) and experimental (E) neuron. Note that the distal regions of the axons and the branching regions of the P50-dynamitin-myc-expressing neurons are sometimes thickened, with microtubules somewhat more splayed apart compared to controls (see arrows). Golgi apparatus morphologies labeled by Golgi-58K protein immunostaining revealed compact, center-located Golgi apparatus in a control neuron (C), and fragmented and dispersed Golgi in a P50-dynamitin-myc-expressing neuron (F). Scale bars, 20 µm.

Figure 2: P50-dynamitinoverexpression results in marked redistribution of axonal neurofilaments. Panels A and B show neurofilament staining in a control neuron (A) and a neuron overexpressing P50-dynamitin (B). The scale bar represents 19 µm. Panel C shows quantitative analyses of neurofilament staining intensity along the length of the axons indicated by arrowheads in panels A and B. Note the dramatic accumulation of neurofilaments in the distal axon of the P50-dynamitin overexpresser compared to the control and the relatively lower neurofilament staining in the proximal axon of the overexpresser compared to the control. Similar analyses on the population of axons studied revealed that these differences are consistent and statistically significant.

Figure 3: P50-dynamitin overexpression reduces anterograde microtubule transport frequency. (A) Time-lapse images reveal a microtubule moving in the anterograde direction through the photobleached region. Red arrows mark the leading and trailing ends of the microtubule. Subsequent staining for myc confirmed that this axon was from a neuron overexpressing P50-dynamitin. (B) The frequency (events/min) of anterograde microtubule transport was significantly reduced in the axons of P50-dynamitin-myc-expressing neurons (chi-square test, *, p < 0.001) as compared to control axons, with no significant difference observed in retrograde transport frequency (chi-square test, p > 0.05). (C) Histogram depicting the mean velocities of microtubule movements in the axons of control [velocity ± SEM = 1.70 ± 0.105 µm/second (anterograde); 1.70 ± 0.116 µm/second (retrograde)] and P50-dynamitin-myc-expressing neurons [velocity ± SEM = 1.90 ± 0.098 µm/second (anterograde); 1.77 ± 0.103 mm/second (retrograde)]. Comparison of these two groups showed no statistically significant variation in microtubule transport velocity regardless of movement directionality (two-tailed t-test, anterograde p > 0.05, retrograde p > 0.05).

Figure 4: P50-dynamitin overexpression inhibits the fastest instantanous velocities of EGFP-EB3 excursions in short-term neuronal cultures. Shown are still images extracted from time-lapse movies of EGFP-EB3 comets within axons and graphs of data extracted from those movies. In control axons, individual EGFP-EB3 comets were seen to move at instantaneous comet velocities faster than 0.6 µm/s (row A). In axons overexpressing P50-dynamitin, EGFP-EB3 excursions faster than 0.6 mm/s were absent (row B). Light and dark arrows in first 3 panels of rows A/B denote initial positions and successive movements of EGFP-EB3 comets, respectively. Arrows in graphs of row A denote instantaneous comet velocities over 0.6 µm/second. Scale bar is 1 µm. (C) Kymographs of comets from the axons of P50-dynamitin-overexpressing neurons (right panel) show a steeper slope than kymographs of comets from control axons (left panel). Kymographs of the P50-dynamitin-overexpressers were more irregular than controls, corresponding to a lack of detectable motion of the comet between individual frames of the movie.

Figure 5: P50-dynamitin overexpression decreases the velocities of EGFP-EB3 excursions in short-term neuronal cultures by selective depletion of fastest instantaneous velocities. In all regions of the axonal shaft (A­C) and within neuronal cell bodies (D), P50-dynamitin overexpression greatly reduced the frequency of EGFP-EB3 comet instantaneous velocities faster than 0.6 µm/second. Blue lines and bars are control data and red lines and bars are experimental (P50-dynamitin-overexpressing cells) data in all graphs. (E) displays axon and cell body data combined, showing the overall differences in instantaneous velocities between controls and P50-dynamitin overexpressing neurons. (F) is a histogram of all EGFP-EB3 comet instantaneous velocities, clearly displaying a reduction in the frequency of comets moving at velocities above 0.6 µm/second within neurons overexpressing P50-dynamitin (red bars) as compared to control neurons (blue bars). In control neurons, 15.7% of the EGFP-EB3 excursions were faster than 0.6 µm/second, whereas in P50-dynamitinoverexpressing neurons only 0.5% of the EGFP-EB3 excursions were faster than 0.6 µm/second, reflecting a 97% reduction (statistically significant difference; p < 0.001, chi square test) in the frequency of fastest comets (G).

Figure 6: EB3 excursion velocities are not diminished in longer term neuronal cultures by either P50-dynamitin overexpression or DHC siRNA. (A), time-lapse images showing the progression of EB3 comets traveling anterogradely down the axons of either a control-siRNA-treated neuron (left column) or a DHC-siRNA-treated neuron (right column). Positional markers for both time points indicate the progression of EB3 comet excursions. (B), quantification of mean EB3 excursion velocities revealed that, while P50-dynamitin- myc-expressing neurons were indistinguishable from controls (velocity ± SEM = 0.149 ± 0.0049 µm/second), DHC siRNA treated cells displayed a slight but significant increase in excursion velocity compared to controls [velocity ± SEM = 0.182 ± 0.0059 µm/ second (DHC) and 0.157 ± 0.0034 µm/second (control); two-tailed student t-test, *, p < 0.001].

Figure 7: DHC depletion results in stunted axonal outgrowth growth and misalignment of microtubules when neurons are depleted of DHC for 3­4 days and then re-plated. (A), quantification of total axonal length per neuron. There is no significant difference between control and P50-dynamitin-myc-expressing neurons. However, the total axonal length of DHC-depleted neurons is much shorter than that of control neurons, with a 70% decrease in length (*p < 0.05, t-test). (B and C), low magnification images of microtubule staining of a control neuron and a DHC depleted neuron, respectively. (D­I), high magnification images of growth cones of a control neuron (D and E) and two DHC-depleted neurons (F, G, H and I). In the control growth cone, microtubules form a tight bundle in the axonal shaft, and splay apart in the growth cone. Overlay of microtubule fluorescence image (red) and F-actin fluorescence image (green) (D) reveals microtubules that invade filopodia (arrows in D and E). In DHC-depleted axons, microtubules (G and I) start splaying apart in the distal segment of the axon (arrows) and display marked misalignment and curvature of the microtubules back on themselves. Overlay of microtubules and F-actin reveals that, in contrast to the situation in the control growth cone, virtually no microtubules invade the filopodia of DHC-depleted growth cones (F and H). Scale bars, 20 µm.

Figure 9: Misalignment of axonal microtubules when microtubule assembly is inhibited and dynactin is also disrupted. (A­C) axonal morphology and microtubule array of dynactin-intact neurons when grown in the presence of 16 nM vinblastine, revealed by b-tubulin immunolabeling. (A´­C´) enlarged images of the enclosed areas in (A­C), revealed smooth, tight microtubule bundles in axons. (D­F) axonal morphology and microtubule array of dynactin-disrupted neurons, by P50-dynamitin overexpression, in the presence of 16 nM vinblastine. (D´­F´) enlarged images of enclosed areas in (D­F), revealed failure to form tight microtubule bundles in the axon, especially in the distal region. Microtubules splayed apart in the axonal shaft and became disoriented relative to each other. Scale bar: (A­C) and (D­F), 20 microns; (A´­C´) and (D´­F´), 10 microns.

Figure 10: Disruption of dynactin inhibits invasion of microtubules into filopodia when microtubule assembly is inhibited. (A­B) immunostaining of microtubules. (A´­B´) overlay of microtubules (green) and phalloidin-staining of actin (red). In control neurons, microtubules entered the peripheral region of the growth cone, some of which invaded filopodia along actin bundles (see Figure 7). P50-dynamitin overexpressers, microtubule distribution in the growth cone was generally similar to controls (data not shown). In a neuron exposed to 16 nM vinblastine (A and A´), which arrests the assembly of microtubules, fewer microtubules invaded the peripheral region of the growth cone and aligned with the actin bundles. In a P50-dynamitin overexpressing neuron treated with 16 nM vinblastine (B and B´), distal microtubules failed to invade the periphery of the growth cone. (C), analysis of the number of microtubules that entered each filopodium along the actin bundle. Under normal culture conditions, control growth cones and dynactin-disrupted growth cones revealed similar numbers. However, in the presence of 16 nM vinblastine, the number of microtubules that entered each filopodium was much diminished in dynactin-disrupted growth cones relative to that in controls (with vinblastine treatment alone) (*p < 0.05, t-test, two-tailed). Scale bar, 20 µm.

Regulation of Microtubule Severing by Katanin Subunits during Neuronal Development (J Neurosci 2005 25:5573-5583.)

Figure 1. Biochemical determination of the ratio of P60 to P80 katanin. A , Calibration of anti-P60 and anti-P80 antibodies. Standard curves were obtained using 20-160 fmol of P60 or 20-80 fmol of P80 katanin. B-D , Concentration of P80 and P60 katanin in developing rat tissues. Tissues were collected and combined from 12 E15 and 12 E18 rats, six to eight P0 or P6 rats, and four to six P12 and adult (Ad) rats. Protein extracts from 1000 µg of combined tissues from different developmental stages were prepared in sample buffer. One-half of the extract was used for Western blot analysis with anti-P80 antibody, and the other half was used simultaneously for Western blot analysis with anti-P60 antibody. The optical densities were measured for bands corresponding to full-length P80 and P60 katanin, and protein concentrations were read from standard curves. Each bar in B-D represents the average amount of katanin found in tissue from 4-12 animals. P60/P80 ratios were calculated by dividing P80 concentration (in fmol) by P60 concentration (in fmol). B , Concentration of P80 and P60 katanin in developing rat heart. C , Concentration of P80 and P60 katanin in developing rat cerebral cortex. D , Concentration of P80 and P60 katanin in developing rat hippocampus. Even when total protein concentration in the adult cerebral cortex sample (Adc) was two times higher than in the adult hippocampus sample (Ad), the P60 concentration was still clearly higher in the hippocampus.

Figure 2. Expression of P60 and P80 katanin in neuronal and non-neuronal tissues. A , C , E , G , E13 mouse sections immunolabeled for P60 katanin. B , D , F , H , E13 mouse sections immunolabeled for P80 katanin. A , High levels of P60 katanin are present along the lengths of axons exiting the fifth cranial nerve (arrows). B , Expression of P80 katanin in axons exiting the fifth cranial nerve is considerably lower than that of P60 katanin (arrows). C , P60 katanin expression is abundant in peripheral processes (arrows) of DRG sensory neurons. D , The same processes of DRG sensory neurons express P80 katanin (arrows) but at a significantly lower level. E , At E13 in the developing cerebral cortex, high levels of P60 katanin are present only in cells and processes near the lateral ventricle (arrows). Labeled cells are radial glia or recently born neurons. F , P80 katanin is also present mostly in cells near the ventricle in E13 cortex (arrows). G , H , In E13 heart, both P60 and P80 katanin are more uniformly distributed in developing cardiac muscle cells. Scale bar: A-D , G-H , 230 µm; E , F , 115 µm. 5TH CRANIAL N., Fifth cranial nerve; VL, lateral ventricle; SC, spinal cord.

Figure 3. Expression of P60 and P80 katanin in the cerebral cortex and hippocampus. A , C , E , E16 mouse cerebral cortex sections ( A , C ) and an E16 mouse hippocampus section ( E ) immunolabeled for P60 katanin. B , D , F , E16 mouse cerebral cortex sections ( B , D ) and an E16 hippocampus section ( F ) immunolabeled for P80 katanin. G , P0 hippocampus section in situ hybridized with a riboprobe specific for P60 katanin. H , P0 hippocampus section in situ hybridized with a riboprobe specific for P80 katanin. A , By E16, P60 is detectably expressed in cells adjacent to the ventricle (arrows) but also in cells far from ventricular zones, which are postmigratory neurons of the early cortical plate (arrowheads). B , P80 expression is very weak in cells adjacent to ventricle (arrows) but is clearly detectible in cells that have migrated toward the pia (arrowheads). C , D , The P60 and P80 expression, respectively, is elevated in these sections within superficial layers of the E16 cerebral cortex, where the postmigratory neurons are localized. E , F , P60 katanin expression and P80 expression in E18 hippocampus is also high in early postmigratory neurons (arrowheads). G , H , At postnatal day 1 (P0), both P60 and P80 are highly expressed in layers containing hippocampal neurons, including the dentate gyrus (arrowheads) and Ammon's horn (arrows). Scale bar: A-D , G , H , 230 µm; E , F , 115 µm. VL, Lateral ventricle.

Figure 4. Comparison of P60 and P80 katanin expression patterns in developing cultured hippocampal neurons. Microtubule staining (-tubulin) is shown in red, and katanin staining is shown in green. The top row shows P60 katanin ( A-D ), and the bottom row shows P80 katanin ( E-H ). P60 and P80 are both present throughout all compartments of the neuron; processes that appear red-orange have lower levels of katanin, whereas processes that appear green have higher levels of katanin. A , E , Neurons in stage 2/3 before dendritogenesis. B , C , F , G , Neurons in early stage 4 (arrows). D , H , Neurons in late stage 4. Note that P80 is more concentrated in cell bodies compared with P60. Note that P60 levels are dramatically higher during early stage 4 compared with earlier and later developmental stages. No such spike in expression is observed with P80. Scale bar, 80 µm.

Figure 5. Quantification of endogenous P60 and P80 katanin levels in cell bodies, processes, and tips of processes of cultured hippocampal neurons at different developmental stages. A-D , Neurons in early stage 4. Microtubule staining (-tubulin) is shown in red, and katanin staining is shown in green. A , B , P60 katanin. C , D , P80 katanin. Arrowheads point to growth cones, which are notably enriched for P60 but not P80. A and C are lower magnification and B and D are higher magnification of the tip regions (growth cone) of axons. The arrows point to branch points, which were also somewhat enriched for P60 but not P80. Quantification of data obtained from various types of processes and various regions of the neuron are shown in E-J . Data ( y -axis) are expressed as the ratio of P60 or P80 obtained with each antibody to -tubulin calculated using AFUs. Absolute levels of P60 and P80 cannot be compared with this approach. Note that the two subunits of katanin do not show identical changes in the various compartments (see Fig. 4 ). Stages of development [as defined by Dotti et al. (1988)] are shown as roman numerals in these graphs. There is a gradual increase in P80 levels within the cell body as the neuron transitions into early and late stage 4 (1.6-2.1 times compared with stage 1; p < 0.001). Compared with stage 3 axons, the levels of P80 in early and late stage 4 axons only increased by 1.36-1.5 times ( p < 0.05). In late dendrites, the levels are approximately triple those in the early immature processes ( p < 0.001). With regard to P60, there is a sharp increase in early stage 4 cell bodies, axons (2.5 times; p < 0.001), and dendrites (2.25 times; p < 0.001) compared with stage 1. The levels of P60 came down in the late stage 4 cell bodies (1.8 times; p < 0.001), axons (1.25 times; p < 0.05), and dendrites ( p > 0.05) compared with stage 1. At all developmental stages, there is more P80 at the tips of the processes compared with the processes themselves (1.6-2.5 times; p < 0.01). With regard to P60, there is no significant difference between the tips and the shafts in the immature processes and the stage 3 axons ( p > 0.05), but there is a very dramatic enrichment at the tips of both axons and dendrites in early stage 4 (3-6 times; p < 0.0001). Scale bar: A , C , 30 µm; B , D , 50 µm.

Figure 6. Effects of katanin constructs on microtubule levels in fibroblasts. A-D , Immunostaining of microtubules in RFL6 cells overexpressing EGFP, katanin P80, con80, or P60. E-G , Quantification of the data on microtubule levels and katanin (P80 and P60) levels. P80 and con80 resulted in partial but not dramatic loss of microtubule mass (decreased by 18 and 16%, respectively; p < 0.01) ( B , C , E ), even when expressed at high levels (5 times more than control; p < 0.001) ( F ). P60, in contrast, caused much greater severing and loss of microtubule mass (reduced by 63%; p < 0.001), even when expressed at much lower levels (28% increase; p < 0.01). Notably, P80 and con80, although not dramatically reducing microtubule levels, resulted in a loss of the tight concentration of microtubules in the centrosomal region ( B , C ). All of the microtubules in the cells overexpressing P60 are relatively short (a few micrometers in length), whereas the cells overexpressing P80 or con80 still display many long microtubules. Data ( y -axis) are expressed in AFUs. Scale bar, 100 µm.

Figure 7. Quantitative analyses of process number and length on cultured hippocampal neurons at different developmental stages overexpressing katanin P80, con80, and P60. A-D , Immunostaining of EGFP for the cells that were transfected with the various constructs using EGFP as the tag. These cells were plated for 1 d, transfected, and allowed to express for 1 d. E-H , Quantification of data on the levels of katanin P80, con80, and P60 in the cell body, on process length, and on process number in hippocampal cultures at various days after plating, induced to express EGFP, P80-katanin, P60-katanin, or con80. The constructs were expressed for only 1 d after various days in culture. Day 1 refers to cells that were transfected before plating. Note that day 2 represents a particularly sensitive developmental stage with regard to process number. The P60 construct diminishes process number, whereas the P80 constructs increase process number ( A-D , H ) ( p < 0.05 between control and P80/con80; p < 0.0001 for control and P60). With regard to the length of the processes, P80 and P60 overexpression decreased total process length by 30 and 50%, respectively ( p < 0.01). Day 4 is sensitive in terms of process number and length, with all constructs diminishing both ( p < 0.01). The number of days in culture is shown on the x -axis. The y -axis shows the total length of all of the processes extended by the neuron in micrometers ( n > 20). At all time points, overexpression of P80 and con80 increases the protein level within the cell body by 2.3 times, whereas P60 increases total levels by 33% ( E , F ) ( y -axis is in AFUs). Scale bar, 30 µm.

Figure 8. Quantitative analyses of microtubule levels in different regions of cultured hippocampal neurons overexpressing P80, con80, and P60. Aa and Bb are microtubule immunostainings of late stage 3 hippocampal neurons. a and b are the glow-scale pseudocolored images, with white indicating the highest level and purple indicating the lowest level (see color-coded bar). Aa shows a control neuron expressing EGFP, and Bb shows a neuron overexpressing P60 katanin. Arrowheads point to the axon, and small arrows indicate twisted microtubules that are sometimes observed in neurons overexpressing P80, con80, or P60 (note also the bright structure in the top left of B and D ; (this is a small bundle of these contorted microtubules). Note that the axon shows no diminution in fluorescence intensity as a result of P60 overexpression, whereas the cell body and immature ("minor") processes are dimmer. Cc shows the quantification of microtubule mass in stage 3 neurons. Overexpression of the various katanin constructs in stage 3 neurons resulted in a 30-58% decrease in microtubule levels in cell bodies and minor processes ( p < 0.01). In axons, overexpression of P80 or con80 did not cause any microtubule diminution (no significant difference; p > 0.05), but overexpression of P60 caused a 20% loss of microtubules ( p < 0.01). Dd shows the quantification of microtubule mass in stage 4 neurons. Overexpression of the various katanin constructs caused microtubule levels in the cell bodies to decrease by 44-50% ( p < 0.01), which is similar to the results obtained from the cell bodies of stage 3 neurons. In the stage 4 axons, like stage 3 axons, P80 or con80 did not cause any microtubule diminution ( p > 0.05). However, at stage 4, there was also no diminution in microtubule levels in axons resulting from expression of the P60 construct ( p > 0.05). In dendrites, there were 20 and 23% diminution in microtubule levels as a result of P80 and con80, respectively, and a 44% diminution in microtubule levels as a result of the P60 construct ( p < 0.05). Scale bar, 30 µm.

Role of Cytoplasmic Dynein in the Axonal Transport of Microtubules and Neurofilaments (J Cell Biol 2005 168: 697-703.)

Figure 1:

Figure 2:

Figure 3:

Figure 4:

Monastrol, a Prototype Anti-Cancer Drug that Inhibits a Mitotic Kinesin, Induces Rapid Bursts of Axonal Outgrowth from Cultured Postmitotic Neurons (Cell Motil Cytoskeleton 2004 58: 10-16.)

Fig. 2. Effects of monastrol and taxol on dissociated cultures of rat sensory neurons and a murine carcinoma cell line. Cultures are immunostained for III-tubulin. A–C: Neurons after 24 h of exposure to vehicle alone (A), monastrol (B), or the lower dose of taxol (C). Axons are about half as long in the presence of monastrol compared to controls, but neuronal morphology is otherwise similar to controls. Arrow in A indicates a non-neuronal cell (slightly fluorescent due to nonspecific labeling with the secondary antibody), of which there are fewer after monastrol treatment. Axonal growth is severely stunted in the presence of taxol, and neuronal morphology is dramatically abnormal. Microtubules are also abnormal in taxol-treated cultures (see text). D,E: Summary of data from neuronal cultures and the carcinoma cell line. Note that monastrol and the higher dose of taxol both produce a noticeable inhibition of cell proliferation while the lower dose of taxol does not. Bar in A (A–C) 10 m.

Fig. 3. Effects of 4-h monastrol treatment on already-grown axons from explant cultures. A: Portion of an explant culture from superior cervical ganglia showing axon tips ( arrows show examples) that were monitored for growth. B: Summary of data. Axonal growth is increased relative to controls in explant cultures treated with monastrol in the case of both sensory (DRG) and sympathetic (SCG) neurons. Bar in A 20 m.

Role of Actin Filaments in the Axonal Transport of Microtubules (J Neurosci 2004 24: 11291-11301.)

 

Figure 1: Microtubules align with actin filament bundles during early axogenesis. Shown are images of neurons fixed and stained for fluorescence visualization of actin filaments (panel A) and microtubules (panel B) during early axonal outgrowth. Large lamellae developed in which microtubules and actin bundles formed along the axes of presumed future axons. Microtubules and actin bundles were seen to colocalize in some cases (see arrows in A and B). Bar, 10 microns.

 

 

 

Figure 2: Actin depletion alters peripheral microtubule organization. Shown are images of neurons fixed and stained for fluorescence visualization of actin filaments (left column) and microtubules (right column) before axonal outgrowth (panels A-D) and after significant axonal outgrowth had occurred (panels E-H). Before axogenesis, neurons developed broad lamellae in which microtubules extended to the periphery (panel B), but did not invade the actin-rich cortical region (panel A). Treatment for 30 minutes with latrunculin depleted virtually all of the filamentous actin (panel C), which resulted in the extension of microtubules to the edge of the lamella (panel D). Note that microtubules also fail to curve back on themselves after actin depletion. Control neurons showed dense bundles of actin filaments occupying the peripheral regions of growth cones and filopodia (panel E). Individual microtubules (arrows in F) were seen to align with some of these actin bundles (arrows in E). After actin depletion (panel G and H), the growth cones typically collapsed into rounded bulbs, with microtubules swirling in a disorganized fashion within the bulb (panel H). Arrows point to actin bundles in A and E, and to microtubules in B, D and F. Bar, 10 microns.

Figure 3: Effects of actin depletion on microtubule transport in the axon. (A), time-lapse images reveal a microtubule moving in the anterograde direction through the photobleached region. Brightness and contrast were adjusted to best reveal the moving microtubule. Red arrows mark the leading and trailing ends of the microtubule. (B), analysis of microtubule transport events demonstrated that the frequency (events/min) of anterograde transport was significantly reduced in actin-depleted axons (*p<0.001, chi-square), while the frequency of retrograde transport was not significantly affected. (C), histogram of microtubule lengths displays no significant difference between control and latrunculin treated neurons. (D), histogram depicting the mean velocity distributions of both anterograde and retrograde microtubule transport in control and latrunculin treated neurons. A significant increase in mean velocity was detected in both anterograde and retrograde microtubule movements (p<0.005 and p<0.001, respectively; 2-tailed t-test). (E), histogram showing the distributions of the instantaneous velocities of microtubules chosen randomly from the total observed microtubule population shown in (D). No significant difference was detected between the instantaneous velocities of moving microtubules between control and actin-depleted axons. (F), the combined data from (D) and (E) depicting microtubule velocity change as a function of transport directionality shows a significant increase in the mean microtubule transport velocities that is bidirectional (*p<0.01; 2-tailed t-test). A small but insignificant increase is seen in comparing control and latrunculin treated instantaneous velocities. Bar, 5 microns.

Figure 4: Outward transport of microtubule transport requires actin filaments when microtubule density is low. Panel A is a schematic of the drug treatment regime used in this experiment. Microtubules are drawn in red. Panel B shows a control neuron (actin filaments present) after a 15 minute exposure to vinblastine (following treatment with and removal of nocodozole) in which short microtubules (white) were seen to invade distal regions where actin filaments (green) were also localized. Arrows in panel B denote several examples of microtubules that have moved outward to the cell periphery and into processes. Panels C and D show fluorescently labeled actin filaments and microtubules, respectively, in a neuron that has been partially depleted of actin filaments using a lower drug concentration. Microtubules invade peripheral regions of the neuron where actin filaments remain. Arrows in panels C and D denote regions where microtubules and actin filaments co-localized. Panels E and F show phalloidin-staining and microtubules, respectively, in a neuron that has been completely depleted of actin filaments. Short microtubules were seen in the cytoplasm (panel F), but they were not transported to the cell periphery and many appeared curved and not outwardly oriented as in control cells (see panel B). Arrows in F denote examples of mal-oriented and/or curved microtubules. Bar, 10 microns.

Figure 5: Live-cell visualization with GFP-EB3 of microtubule behaviors in neurons depleted of actin filaments prior to axon formation. Panels A-D are still images extracted from a time-lapse movie of a single GFP-EB3 expressing neuronal cell body with a large lamellar region. See supplemental movie 1. Arrows in A, C and D denote GFP-EB3 comets. Prior to depletion of filamentous actin, microtubules were mostly constrained to the central region of the cell body, presumably due to the actin-rich lamellae found in the periphery (panel A). At the time of drug addition to the culture, the GFP-EB3 appeared to temporarily dissociate from the plus-ends of the microtubules (panel B). After actin filaments were depleted, plus-ends of microtubules were seen to invade more peripheral regions of the neuron (panels C and D). Prior to drug treatment, roughly 60% of the excursions occurred from the cell center outward to the cell periphery, whereas most of the others occurred across the cell (panels E and I). After drug treatment, there was an increase in these outward excursions as well as a small increase in inward excursions and a corresponding diminution in excursions going across the cell (panels F and I). Arrowheads in E and F indicate the direction of GFP-EB3 excursions and red color indicates inward directionality. There was also a small increase in the velocity and a greater increase in the duration of GFP-EB3 excursions after actin depletion (panels G and H). Blue bars indicate control data and red bars indicate experimental (actin-depleted) data in G-I. Bar, 5 microns.

Figure 6: Live-cell visualization with GFP-EB3 of microtubule behaviors in axons and growth cones depleted of actin filaments. Neurons that had been transfected to express GFP-EB3 were permitted to grow axons for several hours before treatment with latrunculin. Panels A and B show still images extracted from time-lapse movies of control and actin depleted axons, respectively. White arrows in A and B point out GFP-EB3 comets and grey arrows denote the length and duration of examples of GFP-EB3 excursions. The mean velocity of GFP-EB3 excursions in actin-depleted axons was slightly decreased (panel C, darker bar; lighter bars are controls). The mean duration of the excursions was dramatically increased (panel D, darker bar). Directionality of the excursions was unchanged after depletion of actin filaments, with greater than 95 percent of microtubules showing forward excursions (panel E, lighter bars are controls, darker bars are actin-depleted data). Panels F and G are images of control growth cones extracted from a time lapse movie where GFP-EB3 comets (see arrows) were seen to move outward and invade filopodia. The first 3 panels of row K show still images extracted from a time-lapse movie of a growth cone after actin depletion showing chaotic movements of GFP-EB3 comets which indicate highly disorganized microtubules (arrows point out several comets). Also, GFP-EB3 excursion velocity was significantly increased and excursion duration was notably decreased in growth cones depleted of actin filaments (panels I and J, darker bars). The schematics in panels H and the last panel of row K show the paths and directions of representative individual plus end excursions in control and drug treated growth cones, respectively. Bar, 5 microns in A, B, F and G; 1 micron in row K.

Figure 7: Microtubules track along other microtubules after actin depletion. Shown in this figure are still images from a time lapse movie of a region of a lamella of a neuron depleted of actin after being transfected to express GFP-EB3 as in figure 5. Panels on the right are inverse images that show the lengths of the microtubules (with weak fluorescence) better than the conventional images. At the cell edge, microtubule trajectories were clearly disorganized and gave the impression of tumbling chaotically (region of arrowheads in top 2 panels). See supplemental movie 2. In the central region, microtubules were seen to follow identical trajectories as other microtubules, as shown by individual frames that capture multiple GFP-EB3 comets co-linear with one another (sets of arrows). The inverse images demonstrated that the common trajectories follow along underlying pre-existing microtubules which acted as tracks. Similar co-linear excursions were not as commonly observed in control neurons. Bar, 5 microns.

Figure 8: Directionality of GFP-EB3 excursions in axons grown in the absence of actin filaments indicates uniform or nearly uniform microtubule polarity orientation. Neurons were cultured overnight in the presence of latrunculin to inhibit the formation of a filamentous actin network and then induced to grow axons the following day in the presence of the drug (see panel A for treatment regime). Axons extended in the absence of actin filaments were about half as long as control axons and were thicker in diameter. More GFP-EB3 “comets” were seen across the axon width (arrows in panels B-D). See supplemental movie 3. The growth cone was essentially absent in the drug treated axon, and instead appeared as a club-like region, dense in microtubule ends, at the distal end of the growing axon (arrowhead region denoted in panel B). The mean velocity was unchanged compared to control axons, while the duration of GFP-EB3 excursions increased (panels F-G; darker bars are actin-depleted data, lighter bars are control data). The percentage of forward excursions was only slightly lower than in control axons, indicating that microtubules in the axon achieved uniform (or nearly uniform) plus-end-distal polarity orientation in the absence of actin filaments (panel H). Bar, 5 microns.

Distribution of the Microtubule-Related Protein Ninein in Developing Neurons (Neuropharm 2004 47: 677-683.)

Figure 2. Ninein in a migrating neuron. (A) DIC image of a cerebellar granule neuron with a cell body that had translocated in prior timelapse images. Note leading process extending to upper right from cell body at lower left; cell body appears to be translocating on a process of another cell. (B) GFP–ninein fluorescence, same cell as in A. Note bright fluorescence from region of centrosome and surrounding areas; leading process has faint GFP–ninein fluorescence, better seen with increased contrast (inset). (C) Time-lapse images at higher magnification of cell body in B. Note that the distribution of pericentrosomal ninein changes from image to image. Scale bar = 16 m m (A,B); 10 m m (C).

Figure 3. Ninein in neurons with processes. (A) DIC image of a neuron (upper right) that recently extended a process onto another process from a second neuron (lower left). (B) GFP–ninein fluorescence, same cell as A. Note that regions of higher GFP–ninein are scattered around the nucleus, distinct from the apparent centrosome (arrow). (C) DIC image of a neuron with several processes. Two processes have bulbous endings (arrowheads), while two are more tapered (arrows). (D) GFP–ninein fluorescence, same cell as C. Note that GFP–ninein is only detected in processes, not in the cell body. Also note how GFP–ninein conforms to the shape of the processes; bulbous in bulbous process endings (arrowheads) and more extended in tapered processes (arrows). (E) DIC image of a neuron with a process. (F) GFP–ninein fluorescence, same cell as E. Apparent centrosome contains GFP–ninein (arrowhead) along with a filamentous structure (arrow) directed from the centrosome toward the process. Diffuse GFP–ninein is present throughout the cell body and at lower levels within the process (asterisk). (G) Cerebellar cell immunolabeled for b-tubulin. (H) Same cell as in G, immunolabeled for ninein. Note abundant ninein throughout cell body. Ninein particles (arrowheads) are also present in processes, but in lesser amounts. Scale bar = 16 m m (A–D); 10 m m (E, F); 3.5 m m (G, H).

Figure 4. Ninein associated with retraction. (A–C) GFP–ninein fluorescence in a non-neuronal cell, images from a time-lapse series. GFP–ninein is present at high levels at the apparent centrosome (arrow). Particles of GFP–ninein are present away from the centrosome in one region of the cell. In panels B and C, the region containing high levels of ninein retracted. (D and E) GFP–ninein in a spontaneously retracting neuronal process. Process partially retracts between panels D and E (arrowheads). GFP–ninein is present within the more proximal process (arrow) from which the retracting processes branch (arrowheads). The cell body also contains GFP–ninein. (F) Cultured cerebellar cell with morphology characteristic of a migrating cell, immunostained for ninein. Ninein-containing particles are scattered throughout the cell, but are concentrated within the apparent trailing process (arrowhead). (G) Same cell as G immunostained for b-tubulin, which is also abundant in the trailing process. Scale bar = 30 m m (A–E); 8 m m (F, G).

Axonal Growth is Sensitive to the Levels of Katanin, a Protein that Severs Microtubules (J Neurosci 2004 24: 5778-5788.)

Figure 4. P60-katanin levels decrease with in 24 hr after axons encounter target cells. Basilar pontine neurons were grown in culture and double-labeled for -tubulin and P60-katanin. In some cultures, the neurons were presented with cerebellar cells, mostly granule neurons, which are a physiological target for the basilar pontine axons; axonal growth ceases as the axons encounter the granule cells. A , Axons and growth cones immunolabeled for -tubulin after 24 hr of extension from explants from basilar pontine nuclei explants growing on a laminin substrate in the absence of cerebellar cells. B , Same field as A , labeled for P60-katanin. Regions of the growth cone (shown here is a growth cone that has just recently bifurcated and given rise to two early axonal branches) that are particularly rich in P60-katanin have lower levels of tubulin immunofluorescence, probably reflecting microtubule-severing activity ( A, B , arrows). C , Basilar pontine axons immunolabeled for -tubulin (arrowhead) after 24 hr of coculture with dissociated cerebellar neurons. Cell bodies seen are cerebellar cells, mostly granule cell targets of basilar pontine axons. D , Same field as C , immunolabeled for P60-katanin. Although cerebellar cells often stain brightly for P60-katanin (asterisk, for example), basilar pontine axons have little or no immunoreactivity (arrowhead). E , MeanP60-katanin immunofluorescence in 10 µm axon segments at or near distal tips or in 10 µm segments located 35-100 µm more proximally within the same groups of axons. Distal and proximal segments were measured in nine axons contacting cerebellar cells (Cereb. Cells) and in nine axons not contacting cerebellar cells (Laminin). Mean katanin levels are significantly reduced in distal regions of basilar pontine axons contacting cerebellar cells, by 60% at or near growth cones or other types of axonal endings compared with basilar pontine axons not in contact with cerebellar cells in the same cultures ( p < 0.03). Mean katanin levels in proximal axon segments were significantly reduced by 40% in basilar pontine axons contacting cerebellar cells compared with axons not contacting target cells in the same culture ( p < 0.02). Mean katanin levels were in general significantly lower when comparing proximal and distal regions of the same axons regardless of whether they were in contact with cerebellar cells (30% reduction, axons not contacting cerebellar cells, p < 0.03; 46% reduction, axons contacting cerebellar cells, p < 0.005). F , Example of high katanin levels in a growth cone (arrowhead) and adjacent distal axon segment in a basilar pontine axon growing on laminin. Katanin levels are reduced in more proximal regions of this axon. G , Example of low katanin levels in a growth cone (arrowhead) in contact with target granule cells (asterisks) in the same culture as the axon in F . Note that katanin levels are also low in more proximal regions of the same axons, whereas katanin levels in the granule neurons are high. Scale bar: A-D , 25 µm; F, G , 14 µm.

Figure 5. Expression of P60-katanin wild-type and dominant-negative constructs suppresses axonal outgrowth, but not in proportion to the levels of expression. GFP constructs of the wild-type or dominant-negative P60-katanin were transfected into rat sympathetic neurons. A shows immunostaining for P60-katanin in a neuron that was transfected with GFP alone, whereas B shows two neurons that were transfected with the wild-type construct. Axonal outgrowth is less robust in the neurons expressing the wild-type construct. C shows quantitative data on P60-katanin levels in control (CON) neurons and neurons expressing the wild-type (WT) constructs. The construct raises the total level of immunoreactivity by a mean of 16%, with fairly low variability from neuron to neuron. The dominant-negative construct expressed equally well (data not shown). D shows a marked diminution in axonal length relative to control neurons of neurons expressing the wild-type construct and an even greater diminution in neurons expressing the dominant-negative (DN) construct. E shows axonal length of individual neurons plotted against quantitative data on levels of GFP fluorescence after expression of the wild-type construct. There is no correlation between the amount of expression and the degree to which axonal outgrowth was stunted. A similar lack of correlation was observed with regard to the amount of expression of the dominant-negative and whether we quantified GFP fluorescence or P60-katanin fluorescence. Scale bar, 40 µm.

Figure 6. Morphological features differ in neurons expressing either the P60-katanin wild-type or a dominant-negative construct. Phase-contrast and fluorescence images of the transfected neurons after 8 hr of axonal out growth are shown. The neurons were still alive at the time the images were acquired. Fluorescence images are of GFP and are not augmented with antibody staining to enhance the intensity. In A , GFP alone is expressed in control neurons; expression does not alter the growth properties of the axon compared with nonexpressing neurons. Neurons expressing the dominant-negative construct often show a single short axon with a notable curve ( B ) or a single short axon with an abnormally thickened distal region ( C ). D-F show three other examples of neurons expressing the dominant-negative construct showing other morphologies. Neurons with no outgrowth at all were rare (an example is in shown in D ), except with the replating experimental regimen (see Results). Neurons expressing the wild-type P60-katanin construct also rarely showed no axonal outgrowth at all (except in the replating regimen) but showed a notable reduction in size on the rare occasion when there was no outgrowth ( G ). Neurons expressing the wild-type P60-katanin construct typically showed short axons, often multiple in number ( H ). I-K show three other examples of neurons expressing the wild-type construct with varying degrees of axonal outgrowth. Neurons shown in both phase-contrast and fluorescence are labeled with capital and small letters, respectively, whereas neurons shown only in fluorescence are labeled with capital letters. Scale bar, 25 µm.

Figure 7. Replating of neurons expressing P60-katanin constructs results in two populations of wild-type-expressing neurons with regard to axonal outgrowth. Rat sympathetic neurons were allowed to express the constructs for 2 d and were then triturated, replated, and allowed to grow axons. Control neurons (expressing GFP alone) grew robust axonal arbors in 8 hr. Neurons expressing the dominant-negative construct were stunted in their growth, with approximately half showing no growth at all and the other half clearly stunted compared with controls. Neurons expressing the wild-type construct could be divided into two populations; half of the neurons grew no axons at all, whereas the other half grew axons at highly variable levels. The second population of wild-type expressers, analyzed apart from the first population, was not statistically different from the control neurons but showed more variability. In fact, some of these wild-type expressers grew longer axons than any of the control neurons, an example of which is shown in B . Also shown in B is a wild-type expresser that shows no axonal growth at all. The neurons in A and B are stained for microtubules to reveal their morphology. GFP fluorescence is shown in green. C shows the data graphically. CONT, Control (GFP alone); WT1, wild-type (including process-bearing neurons and neurons with no processes); DN1, dominant-negative (including process-bearing neurons and neurons with no processes); WT2, wild-type (only process-bearing neurons); DN2, dominant-negative (only process-bearing neurons). See Results for more details on data analysis. Scale bar: A, B , 50 µm.

Figure 8. Expression of the P60-katanin dominant-negative construct alters the microtubule array of cultured neurons. Both non-neuronal fibroblastic cells ( A-C ) and neurons ( D-F ) that exist within the cultures derived from rat sympathetic ganglia are shown. Optical sections were taken at different planes. Control non-neuronal cells show a typical radial array of microtubules emanating from a centrosomal region ( A ). Non-neuronal cells overexpressing the dominant-negative construct show an abnormal accumulation of microtubules, reflecting the expected partial inhibition of release ( B , on a plane that accentuates the centrosome) and an apparent increased microtubule length compared with control cells ( C , on a plane below the centrosome). A control neuron is shown in D , and an abnormal accumulation of microtubules, reflecting the expected partial inhibition of release, is also observed in neurons overexpressing the dominant-negative construct ( E , on a plane that accentuates the centrosome). Neurons overexpressing the dominant-negative construct also showed an apparent increased microtubule length compared with control cells ( F , on a plane below the centrosome). Note that the increases in microtubule length are only representative of alterations in microtubule length, because individual microtubules may traverse more than one optical section. Scale bar: A-F , 10 µm.

Figure 9. Expression of the P60-katanin wild-type construct alters the microtubule array of cultured neurons. Both non-neuronal fibroblastic cells ( A, B ) and neurons ( C-E ) that exist within the cultures derived from rat sympathetic ganglia are shown. Optical sections were taken at various planes. Non-neuronal cells expressing the construct show an absence of microtubules from the central region of the cell body, an overall diminution in total microtubule levels, and a notable abundance of very short microtubules. Microtubules several micrometers in length were also present ( A ), suggesting that some regions of the microtubules may be less susceptible to severing by katanin, perhaps because of microtubule-associated proteins that block access to the lattice. Neurons expressing the construct show comparable results ( C-E ). E is at a plane below the nucleus against the culture dish, revealing that not all microtubules scatter to the cell periphery after enhanced katanin-induced severing. Scale bar: A-E , 10 µm.

Expression of the Mitotic Kinesin Kif15 in Postmitotic Neurons: Implication for Neuronal Migration and Development (J Neurocytol 2003 32: 79-96.)

Figure 3. Kif15 co-localizes with actin fibers in interphase RFL-6 rat fibroblasts. A-C. RFL-6 cells probed with anti-tubulin (first column), and anti-Kif15 (second column) antibodies. Merged images are shown in the third column (tubulin in red, Kif15 in green, overlay in yellow). The arrowheads in B, C (last panel) indicate an apparent boundary to the limit of Kif15 immunostaining that does not extend to the cell periphery. The arrow in C indicates presumptive centrosome staining by Kif15 antibody (two centrioles can be resolved). D. RFL-6 cell immunostained with actin (first column), Kif15 (second column), and the merged image shown in the last column (actin in green, Kif15 in red, overlay in yellow). Kif15 shows no apparent co-localization with microtubules during interphase, but shows a tight co-localization with actin.

Figure 4. Kif15 and tubulin co-localize in cultured rat sympathetic neurons. Neurons were plated overnight on polylysine and not further treated with laminin or matrigel. A. This neuron, which had extended only lamellae, was probed with anti-tubulin (left panel) and anti-Kif15 (middle panel). Arrows indicate an example of a microtubule (or small microtubule bundle) that is immunopositive for Kif15. Kif15 can be detected on some, but not all, microtubules. Right panel is the merged image (tubulin in red, Kif15 in green, overlay in yellow). B,C. These neurons had extended short, slow-growing axons. Commonly, neurons grown in this fashion have growth cones that intermittently stall, during which time they display tight knots of microtubules in the distal regions of the axons. B. Neurons probed with anti-tubulin (left), and anti-Kif15 (middle); merge (right) of tubulin (red) and Kif15 (green). C. Neurons probed with anti-actin (left), anti-Kif15 (middle); merge (right) of actin (red) and Kif15 (green). Staining for Kif15 and microtubules show a great deal of overlap, particularly in the regions where microtubules appear as bundles. No apparent co-localization with actin was detected. Arrows indicate examples of the distal enrichment of Kif15 frequently present in axons. The distal enrichment of Kif15 was observed in axons (and axonal branches) with stalled growth cones, but was not observed in distal regions of rapidly growing axons (and axonal branches) in which microtubules were more splayed apart (see middle growth cone of uppermost neuron in A). Bar, 25µm (A); 20µm (B,C).

Figure 5. Kif15 co-localizes with microtubules in rapidly growing neurons. Rat sympathetic neurons were exposed for 2hrs to matrigel (A) or laminin (B,C) before fixation. Cultures were immunostained for tubulin (first column), and Kif15 (second column); merges are shown in the last column (tubulin in red, Kif15 in green, overlay in yellow). Staining for Kif15 and microtubules show a great deal of overlap, particularly in the regions were the microtubules appear as bundles. Growth cones stall less frequently when exposed to these growth factors, and more often display splayed microtubules in their distal regions; such splayed microtubules did not exhibit the particularly high enrichment for Kif15 that was observed within stalled growth cones.

Figure 6. Immunofluorescence of RFL-6 fibroblasts treated with latrunculin or nocodazole. A, D. Treatment of RFL-6 cells with a low concentration of latrunculin spares microtubules (D, left panel) and denser actin bundles (A, left panel). Kif15 immunofluorescence co-localizes with actin bundles (A, middle and right panels) but not with microtubules (D, middle and right panels). Even the denser bundles of microtubules, which appear to be induced by the latrunculin treatment, do not co-localize with Kif15. Merged images (right panels): A, actin in red, Kif15 in green; D, microtubules in red, Kif15 in green; overlays in yellow. B, E. Treatment with a high concentration of latrunculin causes a drastic change in cell morphology but preserves microtubules (E, left panel) while almost eliminating Kif15 and actin immunolocalization (middle panels and B, left panel). Arrows in B point to patches of remaining F-actin that continue to co-localize with Kif15, but microtubules do not co-localize with Kif15 as observed with neurons. Merged images (right panels): B, actin in red, Kif15 in green; E, microtubules in red, Kif15 in green; overlays in yellow. C, F. Treatment of RFL-6 cells with nocodazole disassembles most microtubules (F, left panel) but does not detectably affect actin or Kif15 immunolocalization (C and F, middle panel). Bar, 50µm.

Figure 7. Immunofluorescence of rat sympathetic neurons treated with latrunculin or nocodazole. A. A low concentration of latrunculin abolishes the F-actin staining of the neurons (left panel) without loss of Kif15 immunofluorescence or localization (middle panel). Arrow indicates the distal enrichment of Kif15 frequently observed in axons with stalled growth cones. Merged image in right panel shows actin in red, Kif15 in green. B. Microtubules remain after latrunculin treatment (left panel), and Kif15 immunostaining (middle panel) still closely corresponds to some microtubules. C. Nocodazole disassembles much of the microtubule mass, particularly the more labile microtubules in the distal region of the axon and the lamellae extending from the cell body (left panel). Kif15 continues to co-localize with the microtubules remaining in the shaft of the axon, but no longer shows its typical pattern of staining in the regions of the neuron where microtubules are significantly depleted (middle panel). Arrow indicates the absence of Kif15 distal enrichment where microtubules have been depleted. B,C. Right panels: actin in red, Kif15 in green, overlay in yellow. Bar, 20µm.

Figure 8. Ratio images of KIF15 to tubulin in older cultures of sympathetic neurons. Rat sympathetic neurons were cultured for extended times to allow the development of dendrites. Images of the double immunolabeled neurons were digitally processed to produce ratio images which indicate the relative fluorescence intensities of KIF15 to tubulin as variable gray levels (black is high KIF15 to tubulin, white is low KIF15 to tubulin). Kif15 is dramatically enriched within dendrites as they appear (note black tone; arrows point to dendrites in the last panel). Cell bodies consistently have a high ratio at all stages of development. Bar, 30µm.

Figure 9. Kif15 is enriched in mitotic neuronal precursors and migrating neurons in neonatal mouse brain. A-D, Kif15 immunohistochemistry and in situ hybridization in neonatal mouse cerebellar cortex. A. P1 cerebellum. Kif15-immunoreactivity is highest in the developing external germinal layer (“EGL”) that contains granule cell precursors. Immunoreactivity is also present in a deeper layer, both in cell bodies (arrowheads) and their superficially directed processes (arrows), probably belonging to Bergmann glia. Scattered immunoreactive cells are present in the presumptive internal granule layer (asterisks for examples). B. P8 cerebellum, higher magnification. Cells in the superficial EGL have high levels of immunoreactivity, as do cell bodies (white asterisks) and processes (arrowheads) in the developing molecular and internal granule layers, probably corresponding to Bergmann glia or granule cells. Premigratory granule cells of the deeper EGL exhibit lower levels of immunoreactivity (black asterisks for examples), as do cells in the developing molecular and internal granule layers (arrows), which probably correspond to granule cells at various stages of their migration from EGL to IGL. Inset: High magnification DIC image of an immunoreactive cell in the molecular layer that has the form of a migratory granule cell with processes and is associated with a second process. C. P5 cerebellum, Kif15 in situ hybridization. Many cells in the EGL possess Kif15 transcript, as do cells in a deeper layer (arrowheads), probably corresponding to Bermann glia. Scattered Kif15-positive cells are located the EGL and in and around the deeper layer of expressing cells (arrows). D. Control in situ hybridization, Kif15 sense probe. Little or no hybridization is detected using sense probe. E-G. Kif15 immunohistochemistry, sagittal sections, P8 brain near lateral ventricle. E. Immunoreactive cells are found adjacent to the ventricle (“Vent.”), and at a distance (arrowheads). F. Boxed region in E at higher magnification. Immunoreactive cells are located adjacent to ventricle (asterisks), and at a distance (arrows), including cells in subventricular zones. Process-bearing cells further from ventricle are also immunoreactive (arrowheads). G. Cells associated with the rostral migratory stream are Kif15-immunoreactive (between arrows). The olfactory bulb (right) is the destination of this stream. The ventricle is to the left (arrowhead) but outside the field of view. Bar, 40 µm (A, F); 20 µm (B); 13 µm (B, inset); 160 µm (C,D,E,G).

Figure 10. Confocal images of double-label immunofluorescence for ßIII-tubulin (A,D,G), Kif15 (B,E,H) and the merged Kif15/ßIII-tubulin images (C,F,I) in cerebellar cortex (A-C), a subventricular zone (D-F) and septum (G-H) from P8 mouse. A. Cerebellum, ßIII-tubulin. Cells in the Purkinjie cell layer (PCL) and the IGL are immunoreactive, while the superficial EGL has little immunoreactivity. Arrowheads indicate that cells near the PCL that have high Kif15 immunoreactivity (see B,C) are also ßIII-tubulin immunoreactive. B. Same cerebellar section as (A), Kif15. Many cells in the EGL (between arrows) are highly immunoreactive, as are cells (arrowheads) near the PCL, probably Bergmann glia and some granule cells. C. Same section as (A,B), Kif15 and ßIII-tubulin merge. Kif15 and ßIII-tubulin are co-localized in cells near the PCL (arrowheads). Cells in the superficial EGL express high levels of Kif15 (green), but little ßIII-tubulin (red), while cells of the IGL express less Kif15, and high levels of ßIII-tubulin. D. Subventricular zone and surrounding regions, ßIII-tubulin. Although ßIII-tubulin is low in some regions (asterisk), many cells near the ventricle expressing high levels of Kif15 also express ßIII-tubulin (arrows). The axons of developing corpus colosum (“CC” and surrounding regions) are rich in ßIII-tubulin. E. Same section as (D), Kif15. Many cells near the lateral ventricule (arrowheads) and more distant (arrows) are highly immunoreactive. F. Same section as (D,E), Kif15 and ßIII-tubulin merge. Many cells and processes express either Kif15 (green) or ßIII-tubulin (red). Cells that express both Kif15 and ßIII-tubulin appear yellow or orange and are common in some regions (arrowheads for examples). G. Septum, ßIII-tubulin. Little or no ßIII-tubulin is detected in choroid plexus, and in the cells that line the septum. H. Same section as (G), Kif15. Cells of the choroid plexus and cells that line the surface of the septum (arrow) express high levels of Kif15. Processes and cell bodies at varying distances from the surface also show high levels of Kif15 (asterisk, arrowheads for examples). I. Kif15 and ßIII-tubulin are co-localized (orange) in punctate structures probably representing processes both near the surface of the septum (asterisk) and in deeper regions (arrows). Kif15 and ßIII-tubulin are also co-localized in process-bearing cells, clearly visible in deeper regions of the septum (arrowheads). Comparisons with (H) indicate that Kif15 is present in the cytoplasm and processes of these cells, but is obscured by high levels of ßIII-tubulin labeling when they are both labeled. In (I) only the nuclei, which lack ßIII-tubulin, can clearly be seen to contain Kif15. Bar, 50µm (A-C); 25µm (D-G).

Microtubule Reconfiguration during Retraction induced by Nitric Oxide (J Neurosci 2002 22: 5982-5991.)

Figure 2.   Reconfiguration of cytoskeletal components in axons exposed to NOC-7. Double fluorescence labeling of microtubules and microfilaments reveals reconfiguration of both cytoskeletal components in axons exposed to NOC-7. For each axon, the first fluorescence image shows microtubules, the second image shows microfilaments, and the third image shows a color overlay in which microtubule staining is red and microfilament staining is green . A , In a control axon, microtubules appear as a tight straight bundle, and microfilaments are concentrated in the growth cone. B , This axon was fixed immediately after the growth cone had collapsed in response to NOC-7, before any retraction of the axonal shaft occurred. Bending of the microtubules and redistribution of the microfilaments are already observed at this stage. C , This axon was fixed after retraction was well underway. Microtubules are reconfigured into coiled and sinusoidal bundles, and microfilaments are concentrated in the most distal region of the retracted axon. B, C , Phase-contrast images before and after retraction are shown at a lower magnification. Scale bars, 20 µm.

Figure 3.   Microtubule reconfiguration in axons induced to retract by NOC-7. Single-label studies for microtubules in retracting axons are shown here. Phase-contrast images are shown (at a lower magnification) above corresponding -tubulin images. Numerals represent the minutes after NOC-7 addition to the culture. A , Filamentous microtubule bundles are present in the retraction bulb ( arrows ). B , Microtubules have accumulated in a distal-proximal manner in the retracted region (the arrow shows the brightest region at the tip of the retraction bulb). The coiling of microtubule bundles in the shaft corresponds to the contour of the axon ( arrowhead ). C , Sinusoidal bending of microtubule bundles accommodates the shortening of the axon ( arrows ). Scale bars, 20 µm.

Figure 4.   Retraction of axons in response to nocodazole. Axons treated for 30 min with nocodazole retracted with a morphology distinct from that of axons retracting in response to NOC-7. Treatment with 10 µg/ml nocodazole induced axons to retract with a morphology characterized by bead formation along the axon ( arrow ), outgrowth of fine lateral extensions ( arrowhead ), and the withering of the axonal shaft. Immunofluorescence staining of -tubulin revealed a dramatic depletion of microtubules in axons treated with nocodazole. Scale bar, 20 µm.

Figure 5.   Quantitative data on microtubule levels in axons induced to retract by treatment with either NOC-7 or nocodazole. The method used to quantify microtubule levels is shown in A . Phase-contrast images of a representative neuron show two axons before and after 30 min of treatment with 1 m M NOC-7. Both axons retracted significantly. For each axon, the original length ( L O ) and length after retraction ( L R ) were measured. After simultaneous fixation/extraction to preserve microtubules while releasing free tubulin, neurons were immunostained for -tubulin. Total fluorescence intensity within the axon ( It ) and background fluorescence intensity in a region alongside the axon (Ib) were measured in AFU. The difference between It and Ib represents the fluorescence intensity from the axon, adjusted for any nonspecific background fluorescence. To enable us to compare the microtubule levels in axons of different lengths, we calculated for each axon the fluorescence per micrometer of axonal length (AFU per micrometer) by dividing the (It  Ib) by either L O and L R for treated axons (see the equations) and by L O only for control axons. B , Quantification of NOC-7-treated axons compared with controls. C , Quantification of nocodazole-treated axons compared with controls. B, C , The first bar in each graph shows control microtubule levels in parallel cultures prepared identically at the same time as the experimental cultures for the immunofluorescence analyses. The second bar in each graph shows the microtubule levels (per micrometer) in drug-treated axons, calculated for L O . The third bar in each graph also shows microtubule levels in treated axons, but calculated for L R . Separate control axons were analyzed for each experiment, and their mean AFU per micrometer values were normalized to 100 so that data from the different experiments could be compared. The two-tailed Student's t test (* p  < 0.05) shows a significant difference between control axons and NOC-7-treated axons when the calculations are performed using L R . However, there is no significant difference between the control and the NOC-7-treated axons when L O is used. In the case of the nocodazole-treated axons, there is a significant difference between controls and the treated axons calculated in both ways. Thus, there is no detectable microtubule loss in axons induced to retract by NOC-7; in contrast, there is dramatic microtubule loss in axons induced to retract by nocodazole. Scale bar, 20 µm.

Figure 6.   Analyses on the ratio of tyrosinated/total tubulin in microtubules during axonal retraction. A , Control axons. B , Axons retracting in response to treatment for 30 min with NOC-7. The first panel of each set shows staining for tyrosinated tubulin, whereas the second panel shows staining for total -tubulin. Tyrosinated tubulin staining is brighter distally, but so too is total tubulin. The third panel of each set shows a ratio image of tyrosinated/total tubulin (see Materials and Methods and Results). The ratio image is displayed in a pseudocolor scale in which white is the most intense and black is the least intense. (There is no white in these axons; hence, yellow is the brightest.) Note that the ratio is clearly highest distally in the control axons, indicating a concentration of microtubules particularly rich in tyrosinated tubulin. This same distal concentration is observed in retracting axons, indicating no loss in the distal enrichment of these tyrosinated tubulin-rich microtubules. Scale bar, 20 µm.

Figure 7.   Axonal retraction induced by NOC-7 in the presence of taxol. A , Phase-contrast images of an axon before any treatment, after 30 min in taxol alone, and after 30 min of treatment with NOC-7 in the continued presence of taxol. Note that the taxol treatment did not prohibit axonal growth when applied without the NOC-7, nor did it prohibit axonal retraction after the NOC-7 was added. B , Microtubule immunofluorescence studies of three different axons: (1) a control axon, (2) an axon treated for 30 min with taxol alone, and (3) an axon treated with taxol for 30 min and then treated with both taxol and NOC-7 for an additional 30 min. Axons are shown in a quantitative pseudocolor scale in which white is the most intense and black is the least intense. Note that the taxol-treated axon is brighter than the control axon. The taxol/NOC-7-treated axon is even brighter yet (per micrometer). C , Quantitative studies on fluorescence intensity, in arbitrary fluorescence units. Taxol treatment alone increased microtubule levels by ~30% above control levels. Axons retracting in response to NOC-7 (in the continued presence of taxol) displayed a much higher yet fluorescence intensity per micrometer when calculated for length after retraction ( L R ). However, when calculated for original length ( L O ), the intensity per micrometer was indistinguishable from that of the axons treated with taxol alone. Thus, taxol treatment elevated total microtubule levels above controls, but this stimulation of microtubule assembly did not prohibit axonal retraction in response to NOC-7. The two-tailed Student's t test (* p  < 0.05) shows a significant difference between control axons and axons treated with taxol and a significant difference between control axons and axons treated with taxol followed by NOC-7. There was also a significant difference between axons exposed only to taxol and axons exposed to taxol followed by NOC-7, if calculations were performed using L R . However, there was no significant difference between axons exposed only to taxol and axons exposed to taxol followed by NOC-7 if calculations were performed using L O . Scale bars, 20 µm.

Microtubule Reconfiguration during Axogenesis (Journal of Neurocytology, 2002, in press)

Phase-contrast images of rat sympathetic neurons two hours after plating, and after addition of matrigel. By two hours on polylysine, the cell bodies have extended modest lamellae (panel A). By thirty minutes after adding the matrigel, the lamellae begin to expand (panel B), often encompassing 3-4 times more surface area than prior to adding the matrigel (panel C). Still within the first hour of adding the matrigel, thin cylindrical axons tipped by flat growth cones extend from the periphery of the lamellae. The lamellae associated with the cell body gradually collapse as axons develop (panels D-H). Bar, 30 m.

Higher magnification phase-contrast images of lamellae in the early phases of axogeneseis.

In panel A, a darkened thin filopodium can be seen (marked by an arrow) forming at the periphery of a lamella. Panel B shows crimping and dividing of a lamella into multiple nascent axons. Panel C shows a dark swirl around the edge of the lamella (marked by an arrow). A similar dark swirl gives rise to a nascent axon in panel D (marked by arrow). Panel E shows more crimping of a lamella. Panel F shows how such crimping gives rise to multiple axons. Panel G shows a less common phenomenon with a very dark band (marked by arrow) extending through the lamella and leading into a nascent axon. Panel F shows an early axon with a large growth cone extending from a lamella while the lamella itself is also crimping in the region directly continuous with the cell body. Bar, 17 m.

Live-cell images obtained over a period of 90 minutes of a neuron cultured on polylysine with no addition of matrigel.

Microtubules in the lamella appear as a dense mass of mostly non-aligned polymers, many of which swirl around parallel to but not close to the edge of the lamella. At 12 minutes, a bud of the lamella expands outward (marked by an arrow), but the microtubules do not move into the bud, which subsequently collapses back, as can be seen in the 57 minute time point. The dense microtubule array appears to splay and open-up in one area at 57 minutes, but even by 90 minutes, it is apparent that this does not result in any major reconfiguration of microtubules, which still do not align or invade to the edge of the lamella. Sometimes aster-like formations of microtubules could be observed (see lower right area in the lower two panels). In the 90 minute panel, the larger arrow points to the swirl of microtubules that appear to be swept back from the leading edge of the lamella, while the smaller arrow points to an aster-like formation of microtubules. Bar, 10 m.

Alterations in the microtubule array after the addition of matrigel. Most of the microtubules are aligned with their distal ends directed toward the edge of the lamella.

At 0 minutes, two particularly dense bundles of microtubules are apparent at the outer edges of the lamella, which has already begun to crimp into axons. These bundles are not surrounded on all sides by cell membrane. By 20 minutes, one of these microtubule bundles has given rise to an early axon, with the lamella crimping around it to form a cylinder. The other bundle shows slower development over 20 and 40 minutes, but has produced a second distinct axon by 60 minutes. The region of the lamella between the two early axons has still not collapsed into a discrete axon by 60 minutes; the microtubules are generally aligned but not yet tightly bundled. In the growth cones at the tips of the newly-forming axons, the microtubules once again splay apart from the bundle (as seen of the second axon to form in the 60 minute time point). Bar, 15 m.

Microtubule bundling.

In this sequence, a bud of a lamella that has started to crimp to form an axon. There are several aligned microtubules at 0 minutes, but they have not yet formed a tight bundle. At 7, 18, and 30 minutes, the microtubules can be seen gradually moving together into a bundle as this bud of a lamella is transformed into a cylindrical axon tipped by a growth cone. Bar, 12 m.

Splaying of microtubules from a bundle in a newly-formed growth cone.

A bundle of microtubules extends into a large growth cone and bends around as seen at 0 minutes. At 18 minutes, the bundle loosens, and individual microtubules begin to splay off the bundle into the growth cone. At 29 minutes, more splaying is observed, with a portion of the bundle breaking free from the main bundle. At 44 and 46 minutes, additional microtubule movements during the splaying can be observed. The fact that movements can be very fast is evidenced by marked differences observed between these latter time points, which are only 2 minutes apart. Bar, 12 m.

Fluorescence labeling of cultures after fixation.

Microtubules are shown in green and actin filaments are shown in red. Panel A shows a newly-formed bundle of microtubules similar to that shown in figure 5. The contrast of the actin-staining (in red) has been digitally increased so that the perimeter of the lamella is as clear as possible. Notice that the actin-rich cortex has not yet compressed tightly around the bundle in the more proximal region of the axon. Panel B shows an aster formation of microtubules, marked by an arrow. Bar (both panels), 5 m.

Electron microscopic analyses on neurons not treated with matrigel.

Panel A shows that most of the actin bundles swirl around the periphery of the lamella, parallel to its edge. In buds of the lamella that extend outward, the actin bundles are oriented parallel to the direction of outgrowth of the bud (panel B). Microtubules are generally sparse in the actin-rich regions, but are not entirely excluded. Some microtubules align parallel to the actin bundles (an example is marked by the larger arrow in panel B), while other microtubules are more perpendicular to the actin bundles (an example is marked by the two smaller arrows in panel B). In deeper regions of the lamellae farther from the periphery, actin filaments are not detectable, and the microtubules are generally short and scrambled in appearance (panels C and D.) Bar, 0.75 m.

Electron microscopic analyses of cells exposed to matrigel for 30 minutes.

A substantial diminution of actin levels and a dramatic reorganization of the filaments that persist are apparent. Discrete rib-like bundles of actin filaments extend into the filopodia (panels A-C). These bundles sometimes end blindly deeper in the lamella, but often curve around the periphery of the lamella from one filopodium to the next (panel C; see arrow). Microtubules often align with the path of the actin bundles (as in panels D and E; see arrows). As the filopodia expand into nascent axons, microtubules begin to occupy the space. In addition, the microtubules take on the general alignment of the actin bundles that previously dominated these regions (as in panel F, see arrow). Bar, 0.75 m.

Electron microscopic analyses showing further changes in microtubule configuration one hour after addition of matrigel.

The microtubule profiles are many times longer than in cells not treated with matrigel (see also figure 11 and compare with figure 8). Many of the microtubules show a cortical distribution that mirrors the earlier distribution of actin filaments. Panel A shows microtubules swirling in a tight bundle around the edge of the lamella, in a very similar distribution to that of the actin filaments after 30 minutes shown in figure 9C. In panels A and B, it is apparent that those microtubules that do not tightly hug the edge of the lamella are scattered and less organized relative to one another. In panel B, the transition from organized to less-organized microtubules is less distinct than in panel A, probably illustrating the recruitment of scattered microtubules into the bundle. Panel C shows the swirl of microtubules around the edge of the lamella continuing directly into a newly-formed axon. Microtubules are also recruited into the axon from deeper regions of the lamella (see lower part of panel C). Bar, 1.0 m.

Electron microscopic analyses showing additional features of microtubule configuration after one hour in matrigel.

Panels A and B show deeper regions of the lamella, not near its edge or near regions of axonal formation. The microtubules are longer and more organized relative to one another than in cells not treated with matrigel. Small groups of microtubules are aligned with one another, and the microtubules can be seen “drawing” together in small bundles in a zipper-like fashion. A particularly good example is marked by arrowheads in panel B. Panel C shows a microtubule bundle around the edge of a lamella that expands into a primitive growth cone (at the right side of the panel, see arrowheads). The microtubules can be seen splaying apart from the bundle. Bar, 0.75 m.

Growing and Working with Peripheral Neurons (Methods in Cell Biology, in press)

Dissection of DRGs from a day 10 chicken embryo.

After decapitatation, the embryo was pinned to the dissection plate with its ventral side up. The organs in the chest and abdomen were removed. Working under a dissecting microscope, the connective tissues surounding the spinal column and DRGs were carefully removed. Shown in the middle of the view in this figure is the lumbosacral region of the spinal column (the top of the photo is rostral). On the left-side of the figure, the lumbosacral DRGs and their roots are highlighted in pink and the sympathetic chain is highlighted in blue. The sympathetic chain has been removed on the right side of the figure to expose the DRGs beneath. Arrows point to three DRGs.

Dissection of superior cervical ganglion from neonatal rat.

(A): A neonatal rat pup was fixed on a dissecting plate with 3 syringe needles after anesthesia. (B): The two large salivary glands (one of which is outlined by arrowheads) along the midline were exposed after the neck skin was cut open. (C): After the salivary glands were removed, the sternocleidomastoid muscle (bracketed by arrows) was revealed on both sides. (D): Transection of the sternocleidomastoid muscle and strap muscles beneath has exposed the carotid artery and its branch (shown at higher magnification). The nodose ganglion (indicated by the arrowhead) is located directly lateral to the branch point of the carotid artery and the sympathetic ganglion (indicated by the arrow) is located directly beneath the branch point.

Dissection of superior cervical ganglion from neonatal rat.

(A): A neonatal rat pup was fixed on a dissecting plate with 3 syringe needles after anesthesia. (B): The two large salivary glands (one of which is outlined by arrowheads) along the midline were exposed after the neck skin was cut open. (C): After the salivary glands were removed, the sternocleidomastoid muscle (bracketed by arrows) was revealed on both sides. (D): Transection of the sternocleidomastoid muscle and strap muscles beneath has exposed the carotid artery and its branch (shown at higher magnification). The nodose ganglion (indicated by the arrowhead) is located directly lateral to the branch point of the carotid artery and the sympathetic ganglion (indicated by the arrow) is located directly beneath the branch point.

Dissociated embryonic chick DRG culture.

Embryonic day 10 chick DRG neurons were plated on plain glass coverslips in L-15-based medium thickened with methylcellulose. By 16 to 20 hours after being plated, DRG neurons had generated axons tipped with growth cones. Some neurons generate several simple axons without branches (A), while others generate axons which bifurcate (B).

Rat sympathetic neuronal culture.

(A) When plated in L-15-based medium on a poly-d-lysine coated coverslip, neurons attach firmly to the substrate and generate small lamella and numerous filapodia. Most neurons do not grow processes under these conditions. (B) When plated in N2-serum free medium, sympathetic neurons show somewhat more robust outgrowth, with broader lamellae, and sort stumpy processes with large growth cones.

Laminin-induced axogenesis from sympathetic neurons.

Sympathetic neurons were treated with laminin in N2-serum free medium. (A) and (B), 30 minutes after laminin treatment: broad lamellae have extended from the base of the cell bodies. Early processes tipped with large growth cones have begun to protrude from the periphery of the lamellae (arrowhead) and some short axons are already seen (arrow). (C) By 3 hours after laminin treatment, sympathetic neurons had generated long, thin axons with complex branches.

Long-term rat sympathetic culture.

Neonatal rat sympathetic neurons were plated and maintained as described in the text. A typical neuron is shown in this figure. After roughly two weeks in culture, the neuron has developed several short, thick and tapering dendrites, one of which is marked by an arrow.

Motor proteins regulate force interactions between microtubules and microfilaments in the axon (Nature Cell Biology 2000 2, 276 - 280.)

Effect of nocodazole, nocodazole and latrunculin, or nocodazole and NEM-modified-S1 on axonal retraction.

Time-lapse images of axons subjected to the indicated treatments. a-c, Axon treated with nocodazole. Images were obtained before drug application (a), and 15 min (b) and 30 min (c) after. d, e, Axon treated with nocodazole and latrunculin. Images were obtained before drug application (d) and 30 min after (e). f, g, Axon treated with nocodazole and injected with NEM-modified S1. Images were obtained before experimental treatment (f) and 30 min after (g). Axons frequently, but not always, showed substantial beading (as in g) along their lengths in response to nocodazole treatment. Scale bar represents 15 µm.

Effect of recombinant dynamitin on axonal retraction.

Time-lapse images of axons injected with dynamitin. a-c, Images were obtained shortly after injection (a), and 15 min (b) and 30 min (c) after. No axonal beading was observed, but axons showed marked sinusoidal bending during retraction. d-f, An axon attached to another cell part way along its length (thick arrows). Images were obtained shortly after injection (d), and 15 min (e) and 30 min (f) after. Axons retracted disto-proximally. Sinusoidal bending mainly occurred distal to the point of attachment to the other cell over the first 30 min of retraction. Thin arrows show 'strands' trailing behind the distal tip of the retracting axon. Scale bar represents 15 µm.

Quantitative immunofluorescence analysis of microtubule concentration and distribution in retracting axons.

After experimental manipulations, cells were fixed and prepared using a general antibody against -tubulin. Red represents the lowest fluorescence intensity and white represents the highest. a, Control axon. b, Axon treated with nocodazole for 30 min. c, d, Axons 30 min after injection of neurons with dynamitin. All axons except the control retracted by at least a third of their original lengths. Note that the nocodazole-treated axon shows a clear diminution in polymer concentrations, whereas axons of dynamitin-injected neurons show significantly higher concentrations of polymer. Total fluorescence intensities in a, c and d are roughly equal, showing that little or no microtubule depolymerization occurs during retraction. The high levels of fluorescence in distal regions of retracting axons indicate that microtubules may retreat disto-proximally from the axon tip during retraction. d shows that the trailing strands often left during dynamitin-induced retraction contain microtubules. Scale bar represents 18 µm.

Effects on axons of dynamitin in combination with latrunculin or with NEM-modified S1.

Time-lapse images of axons subjected to the indicated treatments. a, b, Axon pretreated with latrunculin for 15 min and injected with recombinant dynamitin {AU: OK?}. Images were obtained before experimental manipulation (a) and 30 min after injection (b). c, d, Axon co-injected with dynamitin and NEM-modified S1. Images were obtained before injection (c) and 30 min after (d). No retraction was observed with either treatment. Scale bar represents 15 µm.

An Essential Role for Katanin in Severing Microtubules in the Neuron (J. Cell Biol. 1999 145: 305-315)

Western blotting and immunofluorescence analyses of katanin in rat sympathetic neurons.

Western blot analyses were performed using affinity-purified polyclonal antibodies that recognize either the 80- (A, left) or 60-kD (A, right) katanin subunit. The blots reveal the presence of both subunits within developing rat sympathetic ganglia. Shown here are samples of superior cervical ganglia from 4-d animals (G) alongside samples from HeLa cells used as a positive control (H). Similar results were obtained with ganglia obtained from E18 and newborn rat pups. Molecular weight standards are indicated in the margin. Immunofluorescence analyses on cultured rat sympathetic neurons revealed the presence of both katanin subunits throughout the neuron at all stages of development. B and C show freshly plated neurons, while D shows a 3-d culture. B is a cell that had flattened against the substrate, but had not yet begun to extend processes. C is a cell that had begun to extend short processes. D is a cell that had extended a complex axonal arbor (and had begun to show signs of dendritic differentiation). Optical sections were obtained with a confocal microscope, and images were depicted in glow-scale pseudocolor to assist in determining whether there was a focal enrichment of katanin in the cell body that might correspond to the centrosome. No such enrichment was observed, indicating that katanin is not specifically concentrated at the centrosome. Shown here are analyses with the antibody to the 60-kD katanin subunit. Similar results were obtained with the antibody to the 80-kD subunit (not shown). Bar, 15 µm.

Immunoelectron microscopic analyses on katanin distribution in rat sympathetic neurons.

Shown are immunoelectron micrographs from neurons stained with the antibody against the 60-kD katanin subunit and a second antibody conjugated to 5-nm colloidal gold particles. In all neurons examined, katanin immunoreactivity was detected within or around the pericentriolar material (A-C, thicker arrows point to the centrosome). In addition, katanin immunoreactivity was typically associated with the clusters of amorphous cytoplasmic material that persist extraction (A). Occasionally, immunoreactivity was found at a discrete site along the length of a microtubule (A, thin arrow). There was no correlation between the number of microtubules attached to any given centrosome and the number of gold particles associated with it. C is a rare centrosome with several attached microtubules, but the number of gold particles is no higher than with other centrosomes. Immunoreactivity for katanin was also detected in axons (D) and dendrites (not shown). D is shown at slightly higher magnification than A-C. Bar (A-C), 0.40 µm. Bar (D), 0.30 µm.

Effects of microinjecting neurons with the function-blocking katanin antibody.

A shows tracings of typical control neurons roughly 6 h after plating, while B shows tracings of typical antibody-injected neurons (injected roughly 45 min after plating and photographed 5-6 h later). The levels of axon outgrowth were markedly reduced in the antibody-injected cells compared with the uninjected cells. C shows an immunofluorescence micrograph of the microtubule array within a typical control (uninjected) neuron. There is a widespread and relatively even distribution of microtubules throughout the cell body. D shows a neuron that grew no processes after injection with the viable katanin antibody. Microtubules appear throughout the cell body, but, as with uninjected cells, it is impossible to discern whether or not the microtubules are attached to the centrosome. E shows a neuron that extended short processes after injection of the viable katanin antibody. The microtubules have reorganized such that it is clear that several microtubules are attached to a "point source" in the cell body (arrow). In addition, these microtubules are unusually long, extending to the periphery of the cell body. F shows a neuron that grew somewhat longer processes after injection of the viable katanin antibody. In this cell, it is not possible to discern the attachment of microtubules to a point source, but it is clear that many of the microtubules are clustered together near the center of the cell body. There is a particularly bright region of staining within the cluster of microtubules that may correspond to the centrosome. In addition, some of the individual microtubules that can be resolved appear to be unusually long (note the long microtubules extending from the bright cluster toward the right). Bar (A and B), 66 µm. Bar (C-F), 10 µm.

Electron micrographs of centrosomes from control and katanin antibody-injected neurons.

Neurons treated as described in Fig. 3 were examined by electron microscopy. Shown in this figure are the electron micrographs (A and C) and corresponding tracings to make clearer the microtubules and centrosome, shown in black and gray, respectively (B and D). Control neurons show few or no microtubules attached to the centrosome (A and B). By contrast, antibody-injected neurons show many attached microtubules (C and D). Bar, 0.5 µm.

Quantification of microtubule mass and microtubule ends in control and katanin antibody-injected neuronal cell bodies.

To quantify total microtubule mass in the cell body of the neuron, cultures were prepared for immunofluorescence visualization of microtubules 4-5 h after a portion of the neurons had been injected with the function-blocking katanin antibody. There was an increase of roughly 48% in the total microtubule mass in the cell bodies of the antibody-injected neurons compared with the cell bodies of the control neurons. Examples of a cell body of a control and injected neuron displayed in a standard pseudocolor scale are shown in A and B, respectively. The pseudocolor scale is shown in B, with red indicating the most and purple the least intense levels of staining. To quantify microtubule ends, rhodamine-labeled tubulin was injected into the cell bodies of uninjected neurons and neurons that had been injected with the katanin antibody 4-5 h before the injection of the tubulin. 1 min was permitted for the incorporation of fluorescent tubulin onto the free ends of the microtubules, after which the cells were extracted in a microtubule-stabilizing buffer and rapidly imaged. In control cells (C and E), there were four to five times fewer microtubule ends as in antibody-injected cells (D and F). Bar, 5 µm.

Studies using a pharmacologic regime to test the role of katanin in releasing microtubules from the neuronal centrosome.

A schematically illustrates the pharmacologic regime for revealing the transport of microtubules from the centrosome to the periphery of the neuronal cell body. Under these conditions, a synchronized burst of microtubules is assembled at the centrosome, and then rapidly released and conveyed to the cell periphery. Here, the katanin antibody was injected shortly after the addition of nocodazole. Shown in B-E are cells prepared for immunofluorescence visualization of microtubules. B shows a cell that had not been injected, with microtubules concentrated beneath the periphery of the cell body after completion of the pharmacologic regime. C-E show three examples of cells that had been injected with the antibody. In all cases, the microtubules remained clustered within a discrete region of the cell body, and did not distribute beneath its periphery. In some cases, a clear point of attachment could not be clearly discerned (C), but in most cases the attachment of the microtubules to the centrosome could be discerned (D and E, arrows). In many cases, the entire centrosome/microtubule complex tended to relocate from cell center to cell periphery. Bar, 8 µm.

Cytoplasmic Dynein and Dynactin Are Required for the Transport of Microtubules into the Axon (J. Cell Biol. 1998 140: 391-401)

Evidence that the pharmacological strategy reveals microtubule transport, not microtubule assembly.

a and b show a neuron treated with nocodazole, rinsed free of the drug, immediately injected with fluorescent tubulin, and then prepared for immunofluorescence visualization of microtubules 30 min later. The rhodamine image (depicting fluorescent tubulin that had incorporated into microtubule polymer [a]) and the fluorescein image (immunofluorescence image of total microtubule mass [b]) are similar in appearance, indicating complete or near-complete incorporation of fluorescent tubulin into polymer. c and d show a neuron treated with nocodazole, rinsed free of the drug, simultaneously exposed to 50 nM vinblastine, injected with fluorescent tubulin (see Materials and Methods), and then prepared for immunofluorescence visualization of microtubules 30 min later. The immunofluorescence image shows the expected peripheral concentration of microtubules (d). The rhodamine image shows only faint background fluorescence and a fluorescent "scar" made by the injection (c). No fluorescent microtubule polymer was detected. Bar, 5 µm.

SDS-PAGE gel showing the purity of the recombinant dynamitin preparations (a) and micrographs showing Golgi distribution in a control neuron (b) and a neuron 6 h after microinjection with recombinant dynamitin (c).

a shows Coomassie blue staining of native (lanes 1 and 3) and boiled (lanes 2 and 4) dynamitin preparations purified either by nickel-affinity (lanes 1 and 2) or by two-step chromatography (lanes 3 and 4). The analyses confirm the presence of a single major band of the expected size, 50 kD, common to both preparations. In b, the control neuron shows a discrete Golgi apparatus located near the nucleus. In c, the Golgi elements are dispersed throughout the injected cell. Bar, 5 µm.

Effects on axon outgrowth of recombinant dynamitin.

(a) A neuron 7 h after plating. Hundreds of microns of axons have grown. (b) An immunofluorescence image showing the dense microtubule array of the same cell after extraction and fixation. (c) A typical neuron 9 h after microinjection with the recombinant dynamitin. The cell has extended a broad lamellipodium but no axons. (d) An immunofluorescence image of the microtubule array within this cell. The microtubules splayed into the lamellipodia but did not form dense bundles as they did within the axons of control neurons. (e) One of a small number of neurons injected with recombinant dynamitin that extended processes (shown 9 h after microinjection). In these cases, the processes were significantly shorter than the axons of control neuron and also had a broad flat appearance that was rather different from the thin cylindrical shape of the control axons. (f) An immunofluorescence image of the microtubule array within this cell. The microtubules within these processes are more spayed apart and less paraxial than those within the control axons. Bar, 10 µm.

Neurons at various stages of the pharmacological regime designed to reveal the outward progression of microtubules from the centrosome.

The cells shown here were injected with the boiled recombinant dynamitin before rinsing out the nocodazole (see Results for details). The results were indistinguishable from those obtained on uninjected cells. (a) A neuron that had been treated with nocodazole, microinjected with the boiled protein, rinsed free of the drug for 3 min, and then prepared for immunofluorescence visualization of microtubules. A dense radial array of short microtubules is apparent at the centrosome, as are a few unattached microtubules of roughly the same length. (b) A neuron treated in similar fashion and then exposed to 50 nM vinblastine for 15 min. Microtubules are essentially absent from the central region of the cell body and are loosely packed beneath its periphery. (c) A neuron exposed to vinblastine for 30 min. The microtubules are more tightly packed at cell periphery. The microtubules vary somewhat in length from cell to cell, with some of the microtubules appearing longer in the cell shown in b than those in the cell shown in a. However, the differences are no greater than those observed among different cells before or after vinblastine treatment. Bar, 3 µm.

Examples of neurons treated with nocodazole, injected with recombinant dynamitin, rinsed free of the drug for 3 min, and then immunostained to reveal microtubules.

(a) A neuron with a discrete site of nucleation similar to control neurons. The microtubules are fewer and somewhat longer than the microtubules in similarly treated control neurons. (b) A neuron in which the site of nucleation is somewhat more diffuse than in a similarly treated control (uninjected or injected with boiled dynamitin protein) neuron, but the level of microtubule reassembly is similar. Small white arrows indicate the perimeter of the cell body where it is not apparent from the immunofluorescence image. Bar, 3 µm.

Examples of neurons treated with nocodazole, injected with recombinant dynamitin, rinsed free of drug for 7.5 min, exposed to vinblastine for 15 min, and then immunostained to reveal microtubules.

In no such case did the microtubules vacate the central region of the cell body and concentrate beneath the cell periphery, as was the case with similarly treated uninjected neurons or neurons injected with boiled protein. (a) A neuron in which most of the microtubules remained attached to their apparent nucleation site, which had become rather diffuse in appearance. (b) A neuron in which the nucleation site had split into two, and most of the microtubules remained attached to these sites. A few unattached microtubules were present at cell periphery, but the vast majority were not. (c) A neuron in which the microtubules were dispersed throughout the cytoplasm and not attached to any apparent nucleation sites. Small white arrows indicate the perimeter of the cell body where it is not apparent from the immunofluorescence image. Bar, 4 µm.

Depletion of a Microtubule-Associated Motor Protein Induces the Loss of Dendritic Identity (J. Neurosci. 2000 20: 5782-5791.)

Morphology of cultured rat sympathetic neurons revealed by neurofilament immunostaining.

Shown here are examples of the morphology of rat sympathetic neurons that were cultured in the presence of OP-1 to promote robust dendritic differentiation. After 2 weeks in culture, the vast majority of the neurons showed thick, tapering dendrites. Cultures were immunostained with an antibody to a poorly phosphorylated neurofilament epitope that is highly enriched in dendrites compared to axons. Most of the neurons showed three or more robust thick, tapering curvaceous dendrites that were at least 30-50 µm in length. Some neurons showed a somewhat less robust dendritic arbor, but clearly showed unmistakable dendrites. Fewer than 10% of the neurons showed no robust dendrites (d). Scale bar, 30 µm.

Morphology of cultured rat sympathetic neurons exposed to CHO1/MKLP1-antisense for 1 d revealed by neurofilament immunostaining.

Shown here are neurons treated with sense or antisense oligonucleotides specific for CHO1/MKLP1. The neurons shown here were immunostained with the same neurofilament antibody as in Figure 1. a shows a sense control with a robust dendritic arbor similar to that observed in typical control neurons. The remaining panels show antisense-treated neurons with clearly altered morphologies. The neuron shown in b displays dendrites that are longer than typical control dendrites, less tapered but still curvaceous. The vast majority of the experimental dendrites lost their curvaceous appearance and appeared straight and taut. c shows a neuron with dendrites that are somewhat thinner and less tapered than control dendrites, and less curvaceous. At least one of the dendrites is longer than typical controls. The remaining panels (d-h) show neurons in more advanced stages of their loss of dendritic morphology. The tapered regions of the dendrites appear to wither in a distoproximal fashion, but the actual length of the processes is difficult to assess because of bundling with neighboring axons. Scale bar, 40 µm.

Morphology of individual neurons revealed by Lucifer yellow dye injections.

Immunostains suggest that during antisense treatment, the tapering regions of the dendrites wither, but it is unclear whether the entire length of the process increases or decreases because of the bundling of the dendrites with axons from neighboring cells. To investigate this issue, we injected Lucifer yellow into individual neurons to reveal their neuritic arbor. a shows an untreated neuron, b shows a sense-treated neuron, and c and d show antisense-treated neurons. Treatments were for 1 d. In the untreated and sense-treated neurons, the tips of the dendrites are clearly distinguishable. Typical dendrites were 30-50 µm in length. In the antisense-treated neurons, the dendrites are substantially longer than controls. In many cases, the dendrites had elongated to well over 100 µm over the first day in antisense. Scale bar, 30 µm.

Immunofluorescence analyses on the levels of CHO1/MKLP1 in control and antisense-treated neurons.

After 1 d in antisense, cultures were fixed in cold methanol, and CHO1/MKLP1 was visualized using immunofluorescence microscopy. DIC images of the fixed cells are shown in the left-hand column, whereas corresponding immunofluorescence images are shown in the right-hand column. a and a' show an untreated neuron with robust dendrites and strong CHO1/MKLP1 immunoreactivity in the cell body and dendrites but not the axonal network. b and b' show another untreated neuron with somewhat less robust dendrites. The immunoreactivity is correspondingly somewhat less intense. c and c' show an antisense-treated neuron wherein the withered tapered regions of dendrites are still apparent. Immunoreactivity is significantly lower than in control neurons. d and d' show an antisense-treated neuron wherein the dendrites have almost completely reverted to an axonal morphology. The immunoreactivity is even lower yet. Scale bar, 35 µm.

Electron microscopic analyses on dendrites of control and CHO1/MKLP1-antisense-treated neurons.

Top panels show phase-contrast micrographs of embedded samples. a is an untreated neuron, whereas b and c are antisense-treated. The remaining panels show electron micrographs from the indicated regions. MT, Microtubule; NF, neurofilament bundles typical of dendrites; R, ribosomes. Control neurons show abundant ribosomes near the cell body and fewer but still abundant levels farther down the length of the dendrite. Microtubules are scattered near the cell body, but more paraxial (but still not as paraxial as in axons) farther down the length of the dendrite. Neurofilament bundles are plentiful throughout. Ribosomes are still plentiful near the cell body of antisense-treated dendrites, but are dramatically diminished with distance, and virtually absent from the thinner distal regions of the original dendrite and the newly grown regions. Neurofilament bundles remain plentiful throughout most of the original length of the dendrite, but only sparse unbundled neurofilaments are present in the newly grown regions. Microtubules appear at relatively lower levels within the more proximal regions of the original dendrite, but at higher levels more distally. The microtubules appear to be more paraxial than in controls especially in the more distal regions of the antisense-treated dendrites. The levels of internal membranous elements (presumably Golgi) also appear to diminish with antisense treatment. Thus, the cytoplasm of the antisense-treated dendrites becomes more axonal in character. The change in microtubule distribution is consistent with the idea that the minus-end-distal microtubules are moving out of the dendrite back toward the cell body, and the plus-end-distal microtubules are translocating anterogradely during antisense treatment. Scale bar, 0.5 µm.

Microtubule polarity analyses on control and CHO1/MKLP1-antisense treated cultures.

The standard "hooking" procedure was used to assess microtubule polarity orientation. Clockwise hooks indicate plus-end-distal microtubules, whereas counterclockwise hooks indicate minus-end-distal microtubules. a shows the data from two control dendrites, whereas b shows the data from six antisense-treated dendrites. The data are expressed as the percentages of clockwise hooks in different sampled regions. In control dendrites, slightly more than half of the hooks were clockwise in proximal and middle regions, with progressively higher percentages in the more distal regions. In dendrites that had completely thinned along their lengths, the percentage of clockwise hooks was >95%, indicating uniformly plus-end-distal microtubules (data not shown). The dendrites shown in b are "transitional," in the sense that they still showed withering tapered regions. In all cases, the thinner distal regions showed predominantly or entirely clockwise hooks. In two of the six cases, the proportion of plus-end-distal microtubules was significantly higher than controls at corresponding sites throughout the length of the dendrite. In four of the six cases, this increase in the proportion of plus-end-distal microtubules was observed throughout most of the length of the dendrite, except in the most proximal region near the cell body, which actually showed a reduction in the proportion of clockwise hooks. Examples of the electron micrographs are shown in c-e, and the corresponding regions of the dendrite from which the electron micrographs were taken are shown by lettered-marked arrows in b. Scale bar: a, b, 40 µm; c-e, 0.45 µm.

Tau-1 and MAP-2 immunofluorescence analyses on control and CHO1/MKLP1 antisense-treated neurons. a and b show an untreated and an antisense-treated neuron, both stained with the tau-1 antibody.

The dendrites of untreated cultures stained lightly and were clearly surrounded by more intensely-staining axons (a). The withering regions of the antisense-treated dendrites showed no apparent increase in staining for tau-1 and were still surrounded by more intensely stained axons (b). a shows an untreated neuron, whereas d and e show antisense-treated neurons, all stained for full-length MAP-2. The dendrites of untreated cultures stained intensely for full-length MAP-2, whereas the axons showed little or no staining (c). As the dendrites thinned and elongated in response to antisense-treatment, there was no immediate diminution in MAP-2 staining (d), although the intense MAP-2 staining was gradually diminished as the thick tapering regions of the dendrites became progressively more withered (e). Scale bar, 12 µm.

Reorganization and Movement of Microtubules in Axonal Growth Cones and Developing Interstitial Branches (J. Neurosci. 1999 19: 8894-8908.)

Formation of a MT loop in the central region of a large paused growth cone.

A low-power phase image of a cortical neuron (A), taken 1 hr and 16 min before injection with TMR-tubulin, is shown. The neuron has extended an axon with a prominent terminal growth cone. B, An enlarged image of the growth cone in A shows that MTs (phase dark band in growth cone) are beginning to curve during the initial formation of the MT loop. C, A series of fluorescent images of the same growth cone shown in A and B after injection of TMR-tubulin into the neuron. Note the prominent MT loop at 0:00 hr and the short MTs (arrows) throughout the peripheral lamellipodium. The loop uncurls and reforms several times over a period of 20 hr, without elongation of the axon. In the matching DIC images (D), vesicles and mitochondria remain within the loop. Scale bar: A, 30 µm; B-D, 10 µm.

Invasion of MTs into multiple branches extending from a large paused growth cone.

At 0:00 hr a large paused growth cone is beginning to extend a small process filled with MTs (1). Several hours later (2:16hr) MTs begin to invade two other small processes (2, 3). At 4:47 hr these small processes (2, 3) have extended and are heavily invested with MTs. In another small process (4, 4:47hr) MTs invade distal regions. Almost 6 hr later (10:25hr), both processes 1 and 4 have extended to form prominent branches. Note that the concentration of MTs in the branches increases, whereas the concentration of MTs in the paused growth cone decreases. Scale bar, 10 µm.

Movement and fragmentation of individual MTs in a paused growth cone.

A, A large paused growth cone shows a prominent MT loop in the central region. MT movements shown in image sequences B and C occur in regions of the growth cone indicated by boxes. In sequence B a MT elongates while moving rapidly into the peripheral lamellipodium (0-30 sec) and then shortens while moving laterally (40-60 sec). In sequence C a MT elongates (0-30 sec) and then fragments into two shorter MTs (40 sec). The shorter MT segment remains stationary without elongating or shortening (50-70 sec), whereas the longer MT segment grows slightly (50 sec) and then fragments a second time (60-70 sec). Sequences B' and C' highlight in yellow the MT shown in B and fragmented MT shown in C (arrowheads in last frame point to three MT fragments). Scale bar, 5 µm.

Individual MTs moving independently in all directions within a growth cone lamellipodium.

Low-magnification image A shows a growth cone that emerged from the cell body of a newly plated neuron that was just beginning to extend processes. Arrow indicates the lamellipodium shown at higher power in B and B'. During the sequence (B), MTs move out of the central region of the growth cone and explore its periphery. In B' two of these MTs have been highlighted yellow or blue for emphasis. In C the trajectories of both MTs are summarized. The numerals in C refer to frame numerals in B and B'. The lamellipodium is outlined in white to show changes in shape during the sequence. In D composite images resulting from subtraction of the time-lapse images in B from one another in sequence are shown. The region that contained an individual MT in the first frame appears black, and the region to which the MT moved appears white. The yellow and blue arrows point to the positions of the MTs highlighted yellow and blue in B'. All regions in which no MT movement occurred appear uniformly gray (see Material and Methods). Frame numerals correspond to 15 sec intervals. Scale bars: A, 5 µm; B-D, 2 µm.

Organization of MTs in large paused growth cones revealed by immunocytochemistry.

Phase images in A-D were obtained from living growth cones. Fluorescent images in A'-D' were obtained from the same growth cones after extraction of free tubulin, fixation, and immunostaining for -tubulin. The fluorescent images in A'-C' show that MTs form prominent loops in the central growth cone domain and that individual MTs extend into peripheral regions of the lamellipodia and into filopodia. D shows a large, flat, branching region that resembles a growth cone on the axon shaft. The axon is also tipped by a growth cone. In both regions MTs penetrate into the lamellipodia and filopodia (D'). Scale bar, 20 µm.

Splaying apart of the MT array within the axon shaft before interstitial branching.

The sequence of black and white with matching pseudocolor images shows changes in the MT array before (A, A') and during (B, B') development of an interstitial branch. Pseudocolor images are shown to indicate fluorescent intensity from low to high (scale in A'). Arrows in the first image (A) point to two regions where MTs are splayed apart in comparison to the bundled array in the nonbranching region of axon shaft (arrowhead). The axon is labeled to indicate proximal (P) and distal (D) segments. Six minutes later the MTs in the upper region (arrow) have formed a bundle, whereas those in the lower region (arrow) remain splayed. At 28 min MTs have invaded a filopodial process (arrow). Five hours later (B) MTs invade an interstitial branch elongating in the position of the filopodium (arrow in A, 0:28). Arrows in both frames in B indicate distal ends of the MTs. Time is shown in hours and minutes. Scale bar, 5 µm.

Retrograde movement of a MT in an axon branch.

A low-magnification image is shown of a microinjected neuron after extraction and fixation (A) to show the amount of fluorescent tubulin incorporated into MTs. The position of the interstitial axon branch imaged in B is indicated by the arrow. Between 0 and 10 sec (B) a MT elongates and moves anterogradely in the branch. Between 20 and 40 sec (B) the MT shortens and moves retrogradely toward the brightly labeled axon shaft. The sequence of images in B' shows the MT moving in B. These images have been prepared by maximizing the contrast and minimizing the brightness of the images in B to bring out the MT of interest. The branch has been outlined in white. Arrows in B' indicate a common reference point. MTs in a living fibroblast (C) have morphologies and dimensions similar to those imaged in the axon branch (B). Arrows in C point to single MTs, and the arrowhead points to a bundle of MTs. In D, MTs in the branch were traced in white at time 0 (B) and the same MT shown in B' is traced in gray. In E the presence of MTs in this axon branch after extraction and fixation confirms that the fluorescent structures imaged in B and traced in D are actual MTs. Time is shown in seconds. Scale bars: A, 20 µm; B-E, 5 µm.

Withdrawal of a long MT from an axon branch.

During this sequence a long MT within an axon branch moves retrogradely toward the axon shaft and folds back on itself in the expanded region of the branch point. In A, bright fluorescent speckles (indicated by arrowheads) at proximal and distal locations on the MT were used as fiduciary marks to follow retrograde movement. In A', the MT was traced in yellow, and the fluorescent speckles were marked with red, green, and blue dots. The tracings in A' show the distribution of the MTs in the proximal region of the branch. The distal tip of the growth cone (indicated by the arrowhead in A') exhibits motility and does not retract, showing that the MT is withdrawing from a stable branch. P and D refer to proximal and distal segment of the axon, respectively. Time is shown in seconds. Scale bar, 5 µm.

Withdrawal of MTs from an interstitial branch and proximal regions of a large terminal growth cone and invasion of MTs into two newly formed branches.

At time 0:00 hr MTs form a loop in the paused growth cone and invade a newly formed interstitial branch (arrow). This interstitial branch elongates over the next 4 hr (arrows at times 2:16hr and 4:42hr) as MTs continue to invest the branch. At 10:27 hr MTs withdraw from this interstitial branch (arrow), which has thinned and retracted slightly. Concomitantly, MTs invade two recently formed branches (arrowheads). Scale bar, 10 µm.

Redistribution of MTs in axon branches by invasion of some branches and withdrawal from others.

A large paused growth cone (0:00hr) has extended a long filopodium (arrowhead) and a new axon tipped by a growth cone (arrow). By 2:23 hr the newly formed axon has continued to elongate (arrow), but the filopodium does not grow (arrowhead). Two hours later (4:41hr) MTs in the filopodium have increased (arrowhead), whereas the main axon has paused (arrow). Six hours later (10:31hr) the filopodium has elongated into a prominent branch containing many MTs (arrowhead), whereas the axon and its MTs have retracted (arrow). Scale bar, 10 µm

Expression of the Mitotic Motor Protein Eg5 in Postmitotic Neurons: Implications for Neuronal Development (J. Neurosci. 1998 18: 7822-7835.)

Expression of MmEg5 mRNAs in mouse tissues and cultured cells determined by in situ hybridization.

A and B show cultured mouse neuroblastoma cells hybridized with the MmEg5 antisense and sense riboprobes, respectively. Autoradiographs C and D are representative of the hybridization pattern obtained with MmEg5 antisense and sense riboprobes, respectively, at E15.5. Hybridization signal was detected within the submandibular salivary gland (sg), cartilage primordium of the body of the hyoid bone (cb), thymus gland (tg), liver (l), gut (g), heart (h), tongue (t), kidney (k), lung (lu), and epithelial cells of the hindlimbs (hl). Shown in E and F, respectively, are a film autoradiograph and corresponding dark-field illumination of a transverse section of the cerebral cortex (Cx) at E15.5 obtained with the MmEg5 antisense riboprobe. Et indicates epithalamus, and SVZ indicates subventricular zone. LV indicates lateral ventricle. Autoradiographs G-I are representative of the hybridization patterns obtained with MmEg5 antisense riboprobe at P7, P21, and adult mouse brain (Ad), respectively. Cb, cerebellum; Hip, hippocampus; Ob, olfactory bulb. Autoradiograph J is representative of the adult mouse brain hybridization pattern obtained with MmEg5 sense riboprobe. K, Dark-field illumination of the cerebellum hybridized for MmEg5 at P7. At P7, the external granular cell layer (egl) and the internal granular cell layer (igl) are labeled. Scale bar: A, B, 6 µm; C, D, 0.3 cm; E, 0.1 cm; F-K, 0.2 cm.

Expression of Eg5 mRNAs in cultured rat sympathetic neurons determined by in situ hybridization.

Cultured sympathetic neurons were hybridized with either the radioactive (large panels) or with the digoxygenin-labeled (small panels) antisense (A-D) or sense (E) riboprobe for MmEg5. Sympathetic neurons were obtained from superior cervical ganglia of newborn rat pups and were grown for 1, 3, 7, and 14 d. In situ hybridization analyses show that mRNAs encoding Eg5 were expressed at 1, 3, and 7 d. At 14 d, the hybridization signal was similar to that detected at 3 d with the sense riboprobe control. Note also that Eg5 mRNAs are downregulated during in vitro development. Scale bar, 10 µm.

The polyclonal HsEg5 antibody recognizes MmEg5 protein in neuroblastoma cells.

The polyclonal antibody generated from the N-terminal motor region of the human Eg5 molecule reveals the same distribution of Eg5 protein during different phases of mitosis in mouse neuroblastoma cells as observed in HeLa cells with an antibody directed against the tail region of HsEg5 (Blangy et al., 1995). Arrows indicate interphase cells in various panels. Each panel also shows one or more cell in a particular stage of mitosis. A, Prophase; B, Metaphase; C, Anaphase; D, Telophase. Scale bar, 10 µm.

Immunofluorescence analyses on the distribution of Eg5 in cultured rat hippocampal neurons at early stages of development.

At stage 2 (A), the protein is localized within cell bodies and within most minor processes. Most typically, it is present at the tips of the minor processes, but sometimes along their lengths. At stage 3, the protein is still present within the cell body and minor processes. In some axons, the protein was no longer observed at the distal tip of the process (B, arrow). In most cases, the protein was observed at the distal tips of the early axon (C, D) and within branches of the axons (D). At stage 4 (E), protein levels were significantly diminished throughout the neuron. Very low levels of protein were sometimes detected at dendrite tips (arrow). Scale bar, 10 µm.

Immunofluorescence analyses on the distribution of Eg5 in cultured rat hippocampal neurons at stage 5 of development.

Shown are stage 5 hippocampal cultures double-immunostained for -tubulin in A-C to reveal cellular morphology and in A'-C' to show Eg5 distribution. Eg5 protein levels are significantly higher than at stage 4. The protein is localized within the tips of dendrites (A'), as well as within newly forming sprouts of dendrites (B', C'). Scale bar, 5.5 µm.

Immunofluorescence analyses on the distribution of Eg5 in developing cultured rat sympathetic neurons.

A, B, and B' show neurons cultured for 6 hr. A shows that Eg5 is present within cell bodies, lamellipodia, short processes resulting from coalescence of lamellipodia, and within the distal tips of early axons. B is a -tubulin double-stain to reveal morphology. B' shows that Eg5 is present within the cell bodies and distal tips of somewhat longer axons. The remaining panels show older cultures (3 and 14 d) double-immunostained for a dendrite-enriched neurofilament protein in C-E to reveal morphology and for Eg5 in C'-E'. At 3 d, Eg5 is concentrated at the dendrite tip (C', D') as well as in dendritic branches (D'). At 14 d, Eg5 is still detected in the cell body, but is no longer concentrated at the dendrite tip(E'). Scale bar: A, 15 µm; B-E', 10 µm.

Immunofluorescence analyses on the distribution of Eg5 in cultured hamster cortical neurons.

Shown are hamster cortical neurons double-immunostained for -tubulin in A and B and for Eg5 in A' and B'. Eg5 immunostain colocalizes with a subpopulation of microtubule polymer within the growth cone. Scale bar, 13 µm.

Identification of a Microtubule-associated Motor Protein Essential for Dendritic Differentiation (J. Cell Biol. 1997 138: 833-843)

Analysis of CHO1 mRNA expression in cultured sympathetic neurons and CHO cells by in situ hybridization.

Hybridization of the digoxigenin-labeled probe with native mRNA is indicated by the presence of a reddish reaction product within the cell. (A and C) Hybridization with antisense probe on sympathetic neurons and CHO cells, respectively. (B and D) Hybridization with sense probe on sympathetic neurons and CHO cells, respectively. CHO1 mRNA hybridization is prominent within the cell bodies of the sympathetic neurons (A, large arrow). Labeling is also apparent within the cytoplasm of the nonneuronal cells (A, small arrow) within the cultures. No labeling was detected with the sense CHO1 probe in the neurons or the nonneuronal cells within these cultures (B). As in the neuron cultures, labeling was observed in the CHO cells after hybridization with the antisense probe (C) but not with the sense probe (D). In both cases, the labeling was more intense in the nucleus than in the cytoplasm. Bar, 30 µm.

Localization of CHO1/MKLP1 within developing hippocampal neurons in culture.

Immunofluorescence images are shown both overlaid on their corresponding DIC images (A-D) and on their own (A-D). (A and B) In freshly plated neurons that had not yet grown processes, diffuse staining is present within the cell body but not within the lamellipodia. (A and B) After the formation of immature processes, the staining remains diffuse and restricted to the cell body. (C and D) After one of the early processes had differentiated into the axon, the staining is somewhat more patchy but still almost completely confined to the cell body. (C and D) As the remaining minor processes differentiated into dendrites, the staining takes on the appearance of discrete elongated patches. Staining is present both within the cell body and the developing dendrites, but it remains absent from the axon. Bar, 20 µm.

The localization of CHO1/MKLP1 within flatter hippocampal neurons in culture shown at higher magnification.

Staining is particularly patchy and sometimes filamentous. Bar, 10 µm.

Localization of CHO1/ MKLP1 within cultured sympathetic neurons and its association with the microtubule array.

(A and B) DIC and CHO1 immunofluorescence images, respectively, of a field containing two neurons that had developed robust dendrites. The staining within the dendrites is patchy in nature, but the individual patches are somewhat shorter than those in the hippocampal dendrites. (C and D) DIC and CHO1 immunofluorescence images of a neuron treated for 30 min with 10 µg/ml nocodazole to depolymerize a substantial portion of the microtubule polymer. The staining becomes diffuse. Bar, 15 µm.

Phase micrographs and microtubule polarity analyses on sympathetic neurons cultured with no oligonucleotides (A) or 1 µM AS2 (B and C).

(A) Neurons respond to the addition of OP1 to the culture medium by differentiating robust dendrites as indicated by their thick tapering morphology. (B and C) Most neurons grown in antisense form only processes with the thin cylindrical morphology of axons (B), but a small number form the processes with an ambiguous morphology that is slightly thicker than typical axons (C). In microtubule polarity analyses, hooked appendages on the microtubules indicate the polarity orientation of the microtubule. Plus end-distal microtubules show clockwise hooks and minus end-distal microtubules show counterclockwise hooks. (D) In dendrites from oligonucleotide-free cultures, ~60% of the microtubules show clockwise hooks and 40% show counterclockwise hooks. (E) In the thin cylindrical processes that form in AS2, ~99% of the microtubules show clockwise hooks. (F) In the processes with an ambiguous morphology, ~94% of the microtubules show clockwise hooks. Bars: (A-C) 25 µm; (D-F) 60 nm.

Inhibition of a Mitotic Motor Compromises the Formation of Dendrite-like Processes from Neuroblastoma Cells (J. Cell Biol. 1997 136: 659-668)

Phase-contrast micrographs of cultured mouse neuroblastoma cells.

A and B show normal interphase cells extending axon-like processes in A and dendrite-like processes in B. C shows an axon-like process extended by a cell in a culture treated with db cAMP. D shows dendrite-like processes extended by a cell in a culture treated with retinoic acid. Bar, 15 µm.

Microtubule polarity analyses performed on processes extended by neuroblastoma cells.

In the axon-like processes extended by normal interphase cells, the hooked appendages on the microtubules are predominantly clockwise (A). In the dendritelike processes extended by these cells, roughly equal proportions of the hooks are clockwise and counterclockwise (B). In the dendrite-like processes extended in the presence of retinoic acid, roughly equal proportions of the hooks are clockwise and counterclockwise (C). In the axon-like processes extended in the presence of db cAMP, the hooks are predominantly clockwise (D). Bar, 0.08 µm.

Northern blot and immunofluorescence visualization of CHO1 message and CHO1/MKLP1 protein respectively in neuroblastoma cultures.

A shows a Northern blot analysis designed to detect the presence of the 3.6-kb CHO1 message in neuroblastoma cultures. Messenger RNA from cultures treated with either retinoic acid (RA), db cAMP, or neither reagent were analyzed. Detectable levels of CHO1 message were found under all three conditions. The RA condition is shown here. B and C show immunofluorescence images of mitotic neuroblastoma cells immunostained for tubulin (red) and CHO1/MKLP1 (green). B shows a cell in anaphase, with CHO1/MKLP1 staining concentrated in the midzone. C shows a cell in telophase, with CHO1/ MKLP1 staining concentrated in the midbody. Diffuse staining for CHO1/MKLP1 is also present throughout the cells. Bar, 5 µm.

Immunofluorescence analyses of CHO1/MKLP1 distribution in neuroblastoma cells.

Staining is shown in green, while cellular morphology is revealed by overlays of the DIC (differential-interference contrast) images of the cells. A shows dendrite-like processes of normal interphase cells. B shows dendrite-like processes of retinoic acid-treated cultures. One axonlike process is also apparent in B, and is marked with an arrow. C shows axon-like processes extended by a normal interphase cell. D shows axon-like processes extended by cells treated with db cAMP. In all cases, the cytoplasm within the cell body stains for CHO1/MKLP1. In some but not all cases, nuclear staining is also apparent. In all cases, the dendrite-like processes stain for CHO1/MKLP1, with high levels of staining proximally and gradually decreasing levels with distance from the cell body. The axonlike processes show virtually no staining for CHO1/MKLP1. Bar, 10 µm.

Effects of CHO1/MKLP1 sense and antisense oligonucleotides on morphology, CHO1/MKLP1 immunofluorescence, and microtubule polarity orientation.

A-C are cells treated with sense oligonucleotides, while D-F are cells treated with antisense oligonucleotides. The sense-treated cultures were entirely similar to controls in that they showed many dendrite-like processes (A) that stained for CHO1/MKLP1 (B) and contained microtubules of both orientations (C). The antisense-treated cultures contained almost exclusively axon-like processes (D). The staining for CHO1/MKLP1 was virtually obliterated from these cultures (E), and the microtubules within the processes were of uniform polarity orientation (F). Bars: (A, B, D, and E) 20 µm; (C and F) 0.1 µm.

Expression of a Kinesin-Related Motor Protein Induces Sf9 Cells to Form Dendrite-Like Processes with Nonuniform Microtubule Polarity Orientation (J. Neurosci. 1996 16: 4370-4375.)

Phase and tubulin-immunofluorescence images of Sf9 cell induced to express a fragment of CHO1/MKLP1.

a, Phase-contrast micrograph of a moth ovarian Sf9 cell 3 d after infection with a baculovirus construct encoding the N-terminal half (containing the motor domain) of CHO1/MKLP1. b, Same cell immunostained for tubulin. Within 3 d of infection, up to 40% of infected cells had extended at least one process, and a portion had extended several. In the process-bearing cells, tubulin staining was concentrated within the processes. Scale bar, 8 µm.

Microtubule polarity determination in a thick tapering dendrite-like process of an Sf9 cell induced to express a fragment of CHO1/MKLP1.

a, Tracing of a cell that extended a thick tapering process after being induced to express the N-terminal half of CHO1/MKLP1. Arrows indicate points along the length of the process at which MT polarity orientation was determined. b-d, Electron micrographs of process cross-sections at points indicated in a. Clockwise hooks indicate MTs oriented with their plus ends distal to the cell body, whereas counterclockwise hooks indicated MTs with their minus ends distal to the cell body. e, Schematic showing the MT polarity data obtained from six processes analyzed at multiple points along their lengths. All cells chosen for analysis had extended a single process 35-50 µm in length. Four additional processes were analyzed at their midpoints only and produced results consistent with those shown here. Numerals above arrows indicate the fraction of counterclockwise to clockwise hooks counted per region and the percentage of counterclockwise hooks. The proportion of hooked MTs was high, nearly 90%, and of these, only ~15% were ambiguous. Scale bars: a, 10 µm; b-d, 0.15 µm; e, 10 µm.

Microtubule polarity determination in the cell body and in an early process of an Sf9 cell induced to express a fragment of CHO1/MKLP1.

a, Tracing of a cell that extended a short nascent process after being induced to express the N-terminal half of CHO1/MKLP1. Arrows indicate points at which MT polarity was assessed. b, Electron micrograph of the MT array at the point indicated in a. c, Electron micrograph of an MT bundle observed just under the plasma membrane in the cell body. As noted in the legend to Figure 2, clockwise hooks indicate the plus ends of MTs, whereas counterclockwise hooks indicate the minus ends of MTs. In b, observations were made from the vantage point of the distal end of the process, therefore clockwise hooks indicate MTs with their plus ends distal to the cell body, whereas counterclockwise indicate MTs with their minus ends distal to the cell body. In c, MT bundles were clustered just beneath the plasma membrane and, hence, there was no point of reference from which to judge MT polarity orientation. Therefore, hooks indicate the degree of uniformity of the polarity of bundled MTs. In total, five nascent processes between 3 and 10 µm in length were analyzed at one point near the cell body, and the results were as follows: (1) 68 clockwise/12 counterclockwise, 85% plus end distal; (2) 61 clockwise/14 counterclockwise, 81% plus end distal; (3) 37 clockwise/6 counterclockwise, 86% plus end distal; (4) 42 clockwise/10 counterclockwise, 81% plus end distal; (5) 106 clockwise/8 counterclockwise, 93% plus end distal. Five MT bundles were analyzed from three different cell bodies: (1) 11 clockwise/0 counterclockwise, 100% uniformity; (2) 0 clockwise/14 counterclockwise, 100% uniformity; (3) 2 clockwise/17 counterclockwise, 89% uniformity; (4) 11 clockwise/1 counterclockwise, 92% uniformity; (5) 6 clockwise/1 counterclockwise, 86% uniformity. Scale bars: a, 4 µm; b, c, 0.125 µm.

Microtubule polarity determination in a thick tapering dendrite-like process of an Sf9 cell induced to express a fragment of CHO1/MKLP1 while exposed to vinblastine to suppress microtubule assembly.

a, Uninfected Sf9 cells immunostained for MTs. b, Immunostain for MTs in Sf9 cells 3 d after infection with baculovirus construct encoding N-terminal half of CHO1/MKLP1. c, Immunostain for MTs in an infected Sf9 cell that developed in the presence of 10 nM vinblastine. Arrow shows the cell body, which is nearly devoid of MTs. Inset, The MT array at the midpoint of a process that developed in the presence of vinblastine. Clockwise hooks indicate MTs with their plus ends distal to the cell body, whereas counterclockwise hooks indicate MTs of the opposite orientation. Approximately half of the MTs are of each orientation. Scale bar: a-c, 9 µm; inset, 0.3 µm.

Movies