DYNLRB1 is essential for dynein mediated transport and neuronal survival

Abstract

The cytoplasmic dynein motor complex transports essential signals and organelles from the cell periphery to the perinuclear region, hence is critical for the survival and function of highly polarized cells such as neurons. Dynein Light Chain Roadblock-Type 1 (DYNLRB1) is thought to be an accessory subunit required for specific cargos, but here we show that it is essential for general dynein-mediated transport and sensory neuron survival. Homozygous Dynlrb1 null mice are not viable and die during early embryonic development. Furthermore, heterozygous or adult knockdown animals display reduced neuronal growth, and selective depletion of Dynlrb1 in proprioceptive neurons compromises their survival. Conditional depletion of Dynlrb1 in sensory neurons causes deficits in several signaling pathways, including β-catenin subcellular localization, and a severe impairment in the axonal transport of both lysosomes and retrograde signaling endosomes. Hence, DYNLRB1 is an essential component of the dynein complex, and given dynein's critical functions in neuronal physiology, DYNLRB1 could have a prominent role in the etiology of human neurodegenerative diseases.

Keywords: Axonal transport; Dynein complex; Neurodegeneration; Neuronal survival; dynlrb1.

Fig. 1

Depletion of Dynlrb1 reduces axon outgrowth in DRG neurons. A: siRNA screen for effects of dynein complex subunits on neuronal growth (n > 100 cells for each). Positive hits (Dync1i2, Dync1li2, Dynlrb1 and Dctn6) were validated in at least three independent experiments (n > 500 cells each). ** p < .01, and *** p < .001 (Student's t-test and one-way ANOVA). B: Fluorescence images of cultured DRG neurons from adult Thy1-YFP mice transfected with either siControl or siDynlrb1. Neurons were re-plated 24 h after siRNA transfection, and images were acquired 48 h after re-plating. YFP signal in green. Scale bar, 200 μm. C: Fluorescent images of cultured DRG neurons from wild type Thy1-YFP mice and Thy1-YFP- Dynlrb^tm1a/+ mice (YFP signal in green). Cells were imaged every hour for a period of 48 h. Scale bar, 100 μm. D: Quantification of the time-lapse imaging experiment described in C. Mean ± SEM, *** p < .001, n = 3, two-way ANOVA.

Fig. 2

Conditional Dynlrb1 depletion affects survival of proprioceptive neurons in vivo. A: Abnormal clenched hind-limb phenotype in RNX-Dynlrb1^−/− mice suspended from the tail, compared to the typical limb extension of wild type animals. B: Rotarod test (acceleration set at 20 rpm in 240 s) showing that RNX-Dynlrb1^−/− mice are unable to balance themselves on the rod. Mean ± SEM, *** p < .001, n > 6 mice per genotype, two-way ANOVA. C: Representative catwalk gait traces for wild type and RNX-Dynlrb1^−/− mice. The position of the paws are shown in different colors as indicated, RF - right front, RH - right hind, LF - left front, LH - left hind. D: Step sequence regularity index and Paw stands from the catwalk gait analyses of C shows impairment in RNX-Dynlrb1^−/− mice compared to wild type. Mean ± SEM, * p < .05 *** p < .001, n > 6 mice per genotype, one-way ANOVA followed by Tukey's HSD post hoc correction for multiple comparisons. E: Representative DRG ganglia sections from wild type and RNX-Dynlrb1^−/− mice, stained with either NFH (green) for proprioceptive neurons or CGRP (red) for nociceptors. Scale bar, 50 μm. F: Quantification of NFH and CGRP positive neurons from the experiment described in G. Mean ± SEM, * p < .05, n = 3 per genotype, unpaired t-test.

Fig. 3

Characterization of Dynlrb1 knockdown in the DRG in vivo. A: DRG ganglia sections from mice transduced with either AAV9-shControl or AAV9-shDynlrb1. AAV9 infected cells express the Venus reporter (green). Sensory neurons are labeled with β3 tubulin (red). Scale bar, 50 μm. B: Percentage of GFP-positive neurons normalized to β3 tubulin-positive neurons in DRG sections of the experiment described in A. C: Quantitative PCRs on RNA extracted from DRG neuron cultures transduced with AAV9-shControl or AAV9-shDynlrb1 for 7 days. AAV9-shDynlrb1 transduced neurons showed 40% reduction of Dynlrb1 mRNA compared to shControl. Mean ± SEM, *** p < .001, n = 6, unpaired t-test. D: Tail suspension reveals a clenched hind limb phenotype in mice transduced with AAV9-shDynlrb1 compared to AAV9-shControl-treated animals. E: Rotarod tests (acceleration set at 20 rpm in 240 s) on mice transduced as shown, 32 days after viral injection. AAV9-shDynlrb1-transduced mice show significantly reduced performance in the test. Mean ± SEM, * p < .05, ** p < .01, *** p < .001, n > 11 mice per group, two-way ANOVA. F: Representative traces from catwalk gait analyses on mice transduced as described in (e). The position of the paws are shown in different colors as indicated, RF – right front, RH - right hind, LF - left front, LH - left hind. G: Catwalk analysis of experiment described in C. The step sequence regularity index as well as the duration of the paw stands are impaired for both front paws in mice transduced with AAV9-shDynlrb1. Mean ± SEM, * p < .05, *** p < .001, n > 11 mice per group, unpaired t-test.

Fig. 4

RNA-seq of cultured Dynlrb^tm1a/+ DRG neurons highlights signaling deficits. A: Bar charts representing the numbers of differentially expressed genes (contrast analysis in EdgeR, FDR < 0.1) in wild type and Dynlrb^tm1a/+ DRG neurons at the indicated time points in culture. B: Heat map representing the differential expressed genes in wild type and Dynlrb^tm1a/+ mice at 48 h in culture (Log10(FPKM)). C: Ingenuity pathway analyses on the 48 h differentially expressed genes dataset, ranked by –log(p values). The Wnt – β-catenin pathway (highlighted in red) was selected for further validation. D: Representative cell bodies of wild type and Dynlrb^tm1a/+ DRG neurons cultured for 48 h before immunostaining as shown. Scale bar, 20 μm. E: Quantification of nuclear versus cytoplasmic β-catenin in the experiment described in C. Mean ± SEM, * p < .05 ** p < .01, n > 3 mice per genotype, one-way ANOVA followed by Tukey's HSD post hoc correction for multiple comparisons.

Fig. 5

Axonal retrograde transport deficits after Dynlrb1 depletion. A: Representative kymographs of Lysotracker Red tracking on wild type and Dynlrb^tm1c/tm1c DRG neurons transduced for 96 h in culture as indicated (Ad5-Cre, Adenovirus5 expressing Cre-GFP) for 96 h. Diagonal lines indicate retrogradely moving carriers. B: Velocity distributions from the experiment shown in A, Mean ± SEM, n > 32 axons per group over 3 independent biological repeats. C: Fraction of moving versus stationary carriers in the experiment described in A. Mean ± SEM, *** p < .001, n > 32 axons per group over 3 independent biological repeats, two-way ANOVA. D: Representative kymographs of TeNT H_CT tracking on wild type and Dynlrb^tm1c/tm1c DRG neurons transduced for 96 h in culture as indicated. E: Velocity distributions from the experiment shown in (d). Mean ± SEM, n > 36 axons per group over 3 independent biological repeats. F: Fraction of moving versus stationary carriers in the experiment described in D. Mean ± SEM, ** p < .01, n > 36 axons per group over 3 independent biological repeats, two-way ANOVA.