Lissencephaly-1 promotes the recruitment of dynein and dynactin to transported mRNAs - PubMed (original) (raw)

Lissencephaly-1 promotes the recruitment of dynein and dynactin to transported mRNAs

Carly I Dix et al. J Cell Biol. 2013.

Abstract

Microtubule-based transport mediates the sorting and dispersal of many cellular components and pathogens. However, the mechanisms by which motor complexes are recruited to and regulated on different cargos remain poorly understood. Here we describe a large-scale biochemical screen for novel factors associated with RNA localization signals mediating minus end-directed mRNA transport during Drosophila development. We identified the protein Lissencephaly-1 (Lis1) and found that minus-end travel distances of localizing transcripts are dramatically reduced in lis1 mutant embryos. Surprisingly, given its well-documented role in regulating dynein mechanochemistry, we uncovered an important requirement for Lis1 in promoting the recruitment of dynein and its accessory complex dynactin to RNA localization complexes. Furthermore, we provide evidence that Lis1 levels regulate the overall association of dynein with dynactin. Our data therefore reveal a critical role for Lis1 within the mRNA localization machinery and suggest a model in which Lis1 facilitates motor complex association with cargos by promoting the interaction of dynein with dynactin.

PubMed Disclaimer

Figures

Figure 1.

A biochemical screen for novel components of mRNA transport complexes. (A) Cartoon depicting methodology used to identify novel components of mRNA transport complexes. For simplicity only one of the two streptavidin-binding aptamers in the in vitro transcribed RNA is shown. SA, streptavidin; red star, mutation in localization signal; green circle, protein enriched on WT signal. (B) Top panels, immunoblots showing that Egl, BicD, and Dhc are strongly enriched on the ILS, HLE, and TLS signals compared with mutated nonlocalizing equivalents (ILSas, HLEr12, TLSU6C, and TLSΔCA). Bottom panel, ethidium bromide–stained TBE-urea gel showing equivalent amounts of WT and mutant RNAs after coupling to beads, washing, and elution under the same conditions as the pull-downs from extract. Note the two bands of aptamer-HLE RNA and its mutant equivalent, which have sizes consistent with monomeric and dimeric forms (see Materials and methods).

Figure 2.

Identification of known and novel components of mRNA transport complexes by mass spectrometry. (A) Venn diagram showing the number of proteins classed as enriched on the ILS, HLE, and TLS localization signals by MS/MS using our selection criteria (>4 normalized spectral counts [nSCs] present on WT signal and >80% of the total normalized spectral counts for a WT and paired mutant signal present on the WT signal; nSCs are spectral counts normalized to correct for differences in the total number of spectra between individual MS runs; see Materials and methods and

Table S3

legend for details). Proteins were only classed as enriched on the ILS if they fulfilled the selection criteria in the two independent experiments. Proteins classed as enriched on the TLS in A are from the comparison of the WT element to the mutant ΔCA, which has a stronger inhibitory effect than the U6C mutant on recruitment of known components of the localization machinery (B). (B) nSCs observed for the 11 proteins enriched on all three localization signals tested. The nSCs for the two biological replicates of ILS vs. ILSas experiments are shown separately for comparison, as are the data for the TLSU6C mutant. Note that differences in nSCs between different proteins are not a good reflection of abundance, as these values are heavily influenced by the protein’s molecular mass and how well individual peptides are detected by MS. (C) Immunoblot showing the enrichment of Dic, Dlic, p50Dmn, CLIP-190, and Lis1 on active localization signals. The Lis1-interacting protein NudE is not detectable on any signal.

Figure 3.

CLIP-190 is not required for apical mRNA transport in embryos. (A) Immunoblot confirming loss of CLIP-190 protein (arrows) in clip-190 mutant ovary extracts. The clip190KO allele was generated by homologous recombination; Df is the genomic deficiency Df(2L)BSC294, which removes the clip-190 locus. (B–D) Motile properties of particles of the localizing h RNA after injection into embryos of the indicated maternal genotypes. See

Table S2

for full details of motile properties and number of particle tracks analyzed. **, P < 0.01 (ANOVA test). Error bars represent SEM.

Figure 4.

Net minus end–directed transport of apically localizing transcripts requires Lis1. (A) Fluorescent immunoblot showing equivalent levels of Egl, BicD, and Dhc between WT and lis1 mutant ovaries. Comparison of the fluorescence intensity in Lis1 blots reveals that Lis1 levels in the heterozygous and trans-heterozygous genotype are ∼65% and ∼25% of WT, respectively (for quantification see

Fig. S2 A

). (B) Immunostaining of MTs (α-tubulin, red) and centrosomes (α-centrosomin, green) in cycle 14 blastoderm embryos from WT and lis1E415/lis1k11702 mutant mothers, revealing that centrosome position and MT organization and length are not perturbed by reduced Lis1 levels. Nuclei, blue (DAPI). Ap, apical; Ba, basal. (C) Still images from time-lapse movies of injected Alexa Fluor 488–labeled h RNA, revealing a strong defect in apical transport in embryos from lis1E415/lis1k11702 mothers compared with WT (see corresponding

Video 4

). Red and yellow arrows mark injection site and apically localized RNA, respectively. Time after injection is shown. (D–I) Motile properties of h (D–F) and asK10 (G–I) RNA particles after injection into the stated maternal genotypes. In D–F, +TG indicates the presence of two copies of a weakly expressed lis1 transgene. See

Table S2

for full details of motile properties and number of particle tracks analyzed. *, P < 0.05; **, P < 0.01; ***, P < 0.001 (ANOVA test). Error bars represent SEM. Bars: (B and C) 10 µm.

Figure 5.

Size is not the sole factor dictating the sensitivity of RNP transport to reduced Lis1 levels. (A) Cross-sectional fluorescent area of RNPs containing different RNAs. Values shown are means of the mean area of RNPs per WT embryo. In A and C, n is number of embryos analyzed. Note that the values obtained are unlikely to represent the true size of RNPs due to light scattering in the embryo, but allow the relative size of RNPs formed by different species to be evaluated (no overt differences in the size of fluorescent RNPs were observed between WT and lis1 mutant genotypes, e.g.,

Video 4

). (B) Fluorescent, nonenzymatic in situ hybridization demonstrating a reduction in the apical accumulation of endogenous h transcripts (green) in lis1E415/lis1k11702 cycle 14 blastoderm embryos compared with WT. Nuclei, red (DAPI). Note that h is transcribed in stripes. Bar, 10 µm. (C) Box-and-whisker plot revealing a statistically significant decrease in the apical/basal fluorescence of endogenous h signal in lis1 mutant embryos compared with WT. Whiskers mark the first and 95th percentile, box marks 50th percentile, crosses mark the maximum and minimum values, and the hollow square and horizontal line represents the mean and median, respectively. ***, P < 0.001 (unpaired t test).

Figure 6.

Lis1 promotes the recruitment of dynein and dynactin to RNAs. (A) Immunoblots showing that the recruitment of Dhc, but not Egl and BicD, to the ILS is dramatically reduced when complexes are assembled from lis1 mutant ovary extract using aptamer-based RNA pull-downs. Recruitment of proteins to the ILSas mutant from WT extract is shown as a specificity control. Note that reduced Lis1 levels in the starting extract are not apparent for the heterozygous lis1E415/lis1+ combination due to nonlinear enzyme-coupled chemiluminescent detection (reduction of Lis1 levels in heterozygotes is revealed by nonenzymatic detection with fluorescent antibodies [Figs. 4 A, 7 C, and

Fig. S4 D

]). (B) Immunoblots showing a reduced association of p150Glued and p50Dmn with the aptamer-associated ILS when complexes are assembled from lis1 mutant ovary extract compared with WT. Blots for Egl and Dhc recruitment to signals are shown as controls. (C) Immunoblots assessing protein recruitment from WT and lis1 mutant ovary extracts to a localizing 800-nt piece of the K10 3′UTR (K103′UTR) and an antisense nonlocalizing mutant version (asK103′UTR). Reduced Lis1 levels have no effect on Egl binding, but clearly decrease the association of Dhc with both RNA species. We confirmed that the association of p150Glued with asK103′UTR was also reduced using the lis1 mutant versus WT extract (

Fig. S3 D

). (D) Cartoon summarizing the results from A–C and

Fig. S3 D

. Localizing RNAs recruit dynein–dynactin complexes to sites on localization signals bound by Egl–BicD complexes, and to additional sites on the RNA bound via an unknown protein(s) (factor X), also present on nonlocalizing RNAs (see also Bullock et al., 2006). Reduced Lis1 levels decrease the affinity of dynein and dynactin complexes for both signals and nonsignal sites.

Figure 7.

Lis1 regulates the association of dynein and dynactin components. (A) Immunoblots showing that reduced Lis1 levels do not perturb the composition of dynein complexes immunoprecipitated from ovary extracts using anti-Dic antibodies. Anti-GFP antibodies were used as a control. Note that Lis1 is not detectably associated with dynein complexes immunoprecipitated from lis1E415/lis1k11702 extract, suggesting that the Lis1 concentration in these extracts is limiting for efficient binding to dynein. Although Dlic levels appeared different in WT and lis1 mutant extracts in the experiment shown, this difference did not affect Dlic incorporation into dynein complexes. (B) Immunoblots showing reduced levels of p150Glued precipitated with GFP-Dlic (using GBP pull-downs) in lis1E415/lis1k11702 versus WT ovary extract (see

Fig. S4 B

for quantification of the reduced signal in multiple trials). (C) Immunoblots showing reduced levels of Dhc and Dic precipitated with GFP-p50Dmn (using GBP pull-downs) in lis1E415/lis1+ versus WT embryo extract (see

Fig. S4 C

for quantification of the reduced Dhc signal in multiple trials). Fluorescent detection was used for the Lis1 blot in this panel. The lower molecular weight species of GFP-p50Dmn precipitated in this experiment is presumably a degradation product. (D) Approximately fourfold overexpression (OE) of Lis1 (quantification not depicted) increases the amount of Dhc, Dic, and BicD coprecipitated with GFP-p50Dmn from ovary extract. Lis1 was overexpressed by driving UAS-lis1 in the germ line with MTD-GAL4. Note that, unlike in embryo extract (C), association of Lis1 with GFP-p50Dmn is not readily detectable in WT ovary extract (D). This may be due to tissue-specific differences in the abundance, affinities, or stoichiometries of the proteins participating in the interaction.

References

1. Akhmanova A., Steinmetz M.O. 2008. Tracking the ends: a dynamic protein network controls the fate of microtubule tips. Nat. Rev. Mol. Cell Biol. 9:309–322 10.1038/nrm2369 - DOI - PubMed
1. Amrute-Nayak M., Bullock S.L. 2012. Single-molecule assays reveal that RNA localization signals regulate dynein–dynactin copy number on individual transcript cargoes. Nat. Cell Biol. 14:416–423 10.1038/ncb2446 - DOI - PMC - PubMed
1. Bullock S.L., Ish-Horowicz D. 2001. Conserved signals and machinery for RNA transport in Drosophila oogenesis and embryogenesis. Nature. 414:611–616 10.1038/414611a - DOI - PubMed
1. Bullock S.L., Zicha D., Ish-Horowicz D. 2003. The Drosophila hairy RNA localization signal modulates the kinetics of cytoplasmic mRNA transport. EMBO J. 22:2484–2494 10.1093/emboj/cdg230 - DOI - PMC - PubMed
1. Bullock S.L., Nicol A., Gross S.P., Zicha D. 2006. Guidance of bidirectional motor complexes by mRNA cargoes through control of dynein number and activity. Curr. Biol. 16:1447–1452 10.1016/j.cub.2006.05.055 - DOI - PubMed

Lissencephaly-1 promotes the recruitment of dynein and dynactin to transported mRNAs - PubMed (original) (raw)