Stable kinesin and dynein assemblies drive the axonal transport of mammalian prion protein vesicles

Abstract

Kinesin and dynein are opposite-polarity microtubule motors that drive the tightly regulated transport of a variety of cargoes. Both motors can bind to cargo, but their overall composition on axonal vesicles and whether this composition directly modulates transport activity are unknown. Here we characterize the intracellular transport and steady-state motor subunit composition of mammalian prion protein (PrP(C)) vesicles. We identify Kinesin-1 and cytoplasmic dynein as major PrP(C) vesicle motor complexes and show that their activities are tightly coupled. Regulation of normal retrograde transport by Kinesin-1 is independent of dynein-vesicle attachment and requires the vesicle association of a complete Kinesin-1 heavy and light chain holoenzyme. Furthermore, motor subunits remain stably associated with stationary as well as with moving vesicles. Our data suggest a coordination model wherein PrP(C) vesicles maintain a stable population of associated motors whose activity is modulated by regulatory factors instead of by structural changes to motor-cargo associations.

Figure 1. PrP^C Vesicles are Transported Bidirectionally in Wild-type Hippocampal Axons

(A) Top panels: sequential images of YFP-PrP^C vesicle movement in a hippocampal axon. Vesicles moving bidirectionally (*), in a retrograde direction (•), and a stationary one (◊) are followed for a period of 14 seconds. Middle panel: kymograph generated from movie in (A). Bottom panel: same kymograph depicting individual particle traces generated by particle tracking software. (B) Population breakdown of YFP-PrP^C vesicles. (C) Mean segmental velocity; (D) Segmental velocity histograms. Red lines show mean. (E) run length, (F) pause duration, and (G) pause frequency of YFP-PrP^C vesicles. Nv = # vesicles; Np = # pauses; Nt = # tracks; Ns = # segments. All values are shown as mean ± SEM. See also Figure S1.

Figure 2. PrP^C Vesicles Associate with Kinesin-1 and Dynein

(A) Schematic diagram of a membrane flotation experiment showing the 8/35 fraction used as starting material for the vesicle immunoisolation in (B). Wild-type post-nuclear supernatant (PNS) obtained from wild-type mouse brain homogenate was bottom loaded. Buffers used did not contain detergent to prevent breaking of membranes. (B) An antibody against KLC1 was used to pull down associated membrane components from 8/35 fractions, including PrP^C-containing vesicles. KHC antibody recognizes mostly Kinesin-1C. UNB = unbound fraction; imm = immunoisolation. Anti-GFP was used as a control. (C) Deconvolved images of vesicles stained with antibodies against PrP^C and KLC1, Kinesin-1C, Kinesin-1A, or DHC1. Arrows point to some colocalization events. (D) Run length and (E) pause frequency in DHC1 shRNA axons. All values are shown as mean ± SEM. **p<0.01, *p<0.05, permutation t-test. (F) Segmental velocity histograms (shown as percent of segments) of YFP-PrP^C transport in wild-type and DHC1 shRNA axons. Red and light blue curves represent the overall and predicted Gaussian modes, respectively. Ns = # segments; Nt = # tracks. See also Figure S2.

Figure 3. PrP^C Vesicular Transport is Inhibited in Kinesin Light Chain Mutant Axons

(A) Representative kymograph of YFP-PrP^C vesicle movement in wild-type (top panel), KLC1 -/- (middle panel), and KLC2 shRNA (bottom panel) hippocampal axons. (B-E) Transport parameters in KLC1 -/- and KLC2 shRNA axons. (B) Population breakdown of YFP-PrP^C vesicles (Nv = # vesicles), (C) run length, (D) estimated run length, and (E) pause frequency. Numbers inside bars are segments (run length in C), and tracks (pause frequency in E). All values are shown as mean ± SEM. ***p<0.001, **p<0.01, *p<0.05, permutation t-test. (F) Segmental velocity histograms (shown in percent of segments) in wild-type, KLC1 -/-, and KLC2 shRNA axons. Red and light blue curves represent the overall and predicted Gaussian modes, respectively. (G) Anterograde and retrograde wild-type segmental velocity histograms (shown as percent of segments) were reconstituted from adding together histograms of KLC1 -/- and KLC2 shRNA axons (in F). See also Figure S3.

Figure 4. PrP^C Vesicular Transport is Inhibited in Kinesin-1C Mutant Axons

Transport parameters in Kinesin-1A -/-, Kinesin-1C -/-, and Kinesin-1B-cre axons. (A,E) Population breakdown of YFP-PrP^C vesicles (Nv = # vesicles), (B, F) run length, (C, G) pause frequency, (D, H) segmental velocity. ***p<0.001, **p<0.01, *p<0.05, permutation t-test (black asterisks), Wilcoxon-Mann-Whitney test (red asterisks). Numbers inside bars are segments (run length in B and F, segmental velocity in D and H), and tracks (pause frequency in C and G). All values are shown as mean ± SEM. (I) Segmental velocity histograms (shown as percent of segments) in wild-type, Kinesin-1A -/-, and Kinesin-1C -/- axons. Red and light blue curves represent the overall and predicted Gaussian modes, respectively. See also Figure S4.

Figure 5. Composition of Motor Subunits on PrP^C Vesicles

(A) Representative immunofluorescence image of a hippocampal axon stained with antibodies against PrP^C, KLC1, and DHC1. Insets show enlargement with arrows pointing to three and two point sources in KLC1 and DHC1 channels, respectively, that associate with PrP^C vesicles. Dots represent the location of fitted Gaussian functions. (B) Percentage of PrP^C vesicles that have only KLC1, only DHC1, both, or no motor subunits associated with them. Inside bars are the numbers of vesicles for each category. All values are shown as mean ± SEM. (C-D) Gaussian intensity amplitude distributions comparing the frequency of PrP^C vesicles associated with (C) DHC1 and (D) KLC1 intensities, in wild-type and Kinesin-1C -/- axons. (E-F) Histograms of the same Gaussian intensity amplitude distributions shown in (C, D), depicting percent of PrP^C vesicles associated with (E) DHC1 and (F) KLC1. Red and light blue curves represent the overall and predicted Gaussian modes, respectively. Red open circles point to intersections between modes (G-H) Distribution of PrP^C vesicles with (G) one or (H) both associated motor subunits. Numbers in boxes are percentages of PrP^C vesicles in each category. Color gradient represents higher to lower percentage of PrP^C vesicles in each category. Nv = # vesicles. See also Figure S5.

Figure 6. Association of Motor Subunits with Stationary and Moving YFP-PrP^C Vesicles

(A) Left panel: schematic diagram of a microfluidic chamber device. Transfected neurons are shown in green. Right panel: inset of two axons transfected with YFP-PrP^C growing through a single microchannel (outlined with dotted lines) (B) Kymograph of YFP-PrP^C movement in the wild-type hippocampal axon shown in (A). Time of paraformaldehyde application is indicated by green dotted line. The panel below the kymograph is of the deconvolved image of the same fixed axon showing the YFP-PrP^C channel. Red squares correspond to the same anterograde-moving vesicles in the kymograph that have been mapped to those in the fixed deconvolved image. Deconvolved images (inset) were taken of all three fixed/stained channels. Point sources were fitted with Gaussian functions (colored dots). (C, D) Scatter plots of KLC1 versus DHC1 Gaussian intensity amplitudes of (C) all moving and stationary mapped vesicles from n = 7 axons (with and without associated motor subunits), and of (D) stationary vesicles with detected motor subunits.

Figure 7. A Stable Motor Association and Coordination Model of PrP^C Vesicle Transport

(A) A stable motor subunit composition on anterograde, retrograde, or stationary vesicles is depicted, but only a subset of those are active to drive transport in either direction. Our data suggests a coordination model whereby Kinesin-1 and dynein act as alternating activators of opposite polarity transport. Number of motors depicted is arbitrary. See Discussion for details. (B) Activation of retrograde motility requires the vesicle association of a complete Kinesin-1 holoenzyme comprised of both KLC1 and Kinesin-1C. Removal of either subunit downregulates activation of DHC1, but does not dissociate DHC1 from PrP^C vesicles. Thus, Kinesin-1C can activate DHC1 via interaction with KLC1.