Schwarz DS and Blower MD (2014), The calcium-dependent ribonuclease XendoU promo...

XB-ART-49479

J Cell Biol 2014 Oct 13;2071:41-57. doi: 10.1083/jcb.201406037.

Show Gene links Show Anatomy links

The calcium-dependent ribonuclease XendoU promotes ER network formation through local RNA degradation.

Schwarz DS , Blower MD .

???displayArticle.abstract???
How cells shape and remodel organelles in response to cellular signals is a poorly understood process. Using Xenopus laevis egg extract, we found that increases in cytosolic calcium lead to the activation of an endogenous ribonuclease, XendoU. A fraction of XendoU localizes to the endoplasmic reticulum (ER) and is required for nuclear envelope assembly and ER network formation in a catalysis-dependent manner. Using a purified vesicle fusion assay, we show that XendoU functions on the surface of ER membranes to promote RNA cleavage and ribonucleoprotein (RNP) removal. Additionally, RNA removal from the surface of vesicles by RNase treatment leads to increased ER network formation. Using human tissue culture cells, we found that hEndoU localizes to the ER, where it promotes the formation of ER tubules in a catalysis-dependent manner. Together, these results demonstrate that calcium-activated removal of RNA from membranes by XendoU promotes and refines ER remodeling and the formation of tubular ER.

???displayArticle.pubmedLink??? 25287301
???displayArticle.pmcLink??? PMC4195833
???displayArticle.link??? J Cell Biol

Species referenced: Xenopus laevis
Genes referenced: dync1i1 dync1i2 endou endoul ktn1 rpl7a rps6 ssr1
???displayArticle.antibodies??? Dync1i1 Ab1 Endou Ab1 Ktn1 Ab1 Rpl7a Ab1 Rps6 Ab2 Ssr1 Ab1 Tuba4b Ab2

???attribute.lit??? ???displayArticles.show???

	Figure 1. A calcium-dependent nuclease activity in X. laevis egg extract. (A) CaCl2, MgCl2, or MnCl2 were added to extract at increasing concentrations for 60 min. RNA was isolated from extract, 5â€² end-labeled, and run on a denaturing acrylamide gel. (B) Total RNA was isolated from reactions supplemented with no metal, CaCl2, EGTA alone, or EGTA and CaCl2 in the same reaction. RNAs were run on denaturing acrylamide gels and stained with SYBR Green II. (C) Precleared egg extract was run over a Heparin HiTrap affinity column, and flow-through was collected. CaCl2 was added to the flow-through and run over a Heparin HiTrap column preequilibrated with CaCl2. Proteins were eluted with a salt gradient from 0.1 M to 1 M KCl. Peak proteinâ€“containing fractions were pooled and run on a 10% SDS-PAGE gel and Coomassie stained. A band running just under 37 kD in fractions 7 + 8 and 9 + 10 (arrow) was submitted for mass spec analysis and identified XendoU as the putative nuclease. (D) Peak protein fractions from C were run on 10% SDS-PAGE gels containing 32P-labeled RNA in the resolving portion of the gel. Gels were soaked in buffer to remove SDS and renature proteins, then soaked in buffer containing 1 mM of CaCl2, MgCl2, MnCl2, or no metal overnight at 4Â°C and imaged by PhosphorImager.
	Figure 2. XendoU is required for calcium-mediated RNA degradation in vitro. (A) Recombinant wild-type XendoU was incubated with increasing concentrations of CaCl2, MgCl2, MnCl2, and total RNA from egg extract. RNA was visualized on a denaturing acrylamide gel stained with SYBR Green II. (B) Recombinant H162A and K224A were incubated with metal and RNA and visualized as in A. A very faint band (arrow) appears in the presence of calcium and K224A. (C) XendoU or mock-depleted (IgG) cytosol was incubated with 1 mM CaCl2. Recombinant wild-type XendoU (rXU) was added to XendoU-depleted, mock-depleted (IgG), or undepleted cytosol and analyzed as in A. (D) Western blot of XendoU demonstrating the antibody-mediated depletion in CSF extract as compared with undepleted (CSF) or mock-depleted (IgG) extracts. (E) Cytosol or cytosol mixed with light membranes were incubated with increasing concentrations of CaCl2, and RNA was visualized as in A.
	Figure 3. XendoU plays a role in nuclear envelope formation and proper ER morphology. (A) Demembranated X. laevis sperm nuclei were incubated in mock-depleted (IgG) or XendoU-depleted CSF in the presence of GFP-Histone H1 and Vybrant DiI membrane dye. 10â€“15 random fields were taken from live images every 10 min and the percent nuclei with closed nuclear envelopes was counted. n = 3. (B) IgG mock-depleted (top) or XendoU-depleted (bottom) CSF was incubated at room temperature in the presence of 1 mM CaCl2 and 1Ã— PBS + 15% glycerol (PBSg, protein storage buffer) for 60 min followed by staining of membranes by octadecyl rhodamine and imaged live. (C) Mock-depleted (top) or XendoU-depleted (bottom) extracts were incubated with recombinant proteins (wild-type [left], H162A [middle], or K224A [right]) on ice for 30 min followed by addition of 1 mM CaCl2 and incubated for 60 min at room temperature. Membranes were stained and imaged as above. (D) 10â€“12 randomly selected fields were taken for each condition in B and C in three separate egg extracts. Three-way junctions between ER tubules were counted for each field to assess network formation. Statistical comparison of IgG + PBSg to experimental samples was performed using a one-sample t test. Comparison of XendoU-depleted to rescued extract was performed using an unpaired Studentâ€™s t test. Error bars indicate SD. Bars, 10 Âµm.
	Figure 4. A subpopulation of XendoU localizes to ER membranes. (A) Cytosol (cyt) and light membrane (lm) preps (âˆ¼20Ã—) were isolated as described in the Materials and methods. Western blots were performed on each fraction for Î±-tubulin, TRAPÎ±, and XendoU. (B) ER networks were formed in flow cells and fixed with paraformaldehyde and glutaraldehyde. Immunofluorescence for XendoU was performed with Î±XendoU antibody and Î±-rabbit Cy3 secondary antibody (red). Alexa fluor 488 Concanavalin A (conA) was added in with secondary antibodies to detect glycoproteins and the ER (green). The boxed region is magnified 5Ã— in the lower panels. Bars: (top) 10 Âµm; (bottom) 2 Âµm. (C) Light membranes (lm) were washed in buffers containing 200 mM (wlm 200) or 500 mM (wlm 500) KCl. Membranes were washed twice and resuspended in the same original volume. RNA was isolated from each membrane prep (prep A and B), run on denaturing acrylamide gels, and stained with SYBR Green II. (D) Western blots for XendoU, dynein, ribosomal protein S6 (Rib S6), ribosomal protein L7a (Rib L7a), and TRAPÎ± were performed on light membranes (lm) and membranes washed in 200 mM KCl (wlm 200) or 500 mM KCl (wlm 500). TRAPÎ± serves as a loading and wash control.
	Figure 5. XendoU plays a role in vesicle fusion and controls local RNA degradation on membranes. (A) Light membranes were washed one time in buffer containing 200 mM KCl (wlm 200) and incubated in the absence of ATP and GTP (no fusion) or in the presence of ATP and GTP (vesicle fusion). Alternatively, wlms were incubated for 30 min at RT with 5 ÂµM control (IgG) or XendoU antibody followed by the addition of ATP and GTP. Aliquots were mixed with octadecyl rhodamine at 15 min, incubated for an additional 15 min at RT, and imaged live. Representative images of each condition are shown. (B) RNA was isolated from membrane pellets from the end points of reactions containing IgG or XendoU antibodies in A, 5â² end-labeled with [32P-Î³]ATP as described, and run on a denaturing gel. (C) Membranes were pelleted from standard vesicle fusion reactions (+ATP +GTP) or control reactions (âATP âGTP), and RNA was isolated from the supernatant, run on a denaturing gel, and imaged with SYBR Green II. (D) Western blots of XendoU, ribosomal protein S6 (RibS6), ribosomal protein L7a (RibL7a), dynein, and TRAPÎ± on membranes after vesicle fusion (+ATP +GTP) or in the absence of fusion (âATP âGTP) in the presence of IgG or XendoU antibody. (E) Wlms were mock-treated or RNaseA treated (0.01 ng/Âµl, 0.1 ng/Âµl, or 1 ng/Âµl), then washed once more, and RNA was isolated and imaged as in C. (F) RNaseA-treated vesicles from E were incubated and imaged as in A at 20 min and 50 min after addition of ATP and GTP. Bars, 10 Âµm.
	Figure 6. Human EndoU localizes to the ER and controls ER morphology. (A) Schematic of the structures of XendoU, human EndoU2, and EndoU-short. The locations of conserved catalytic residues and putative signal sequence/transmembrane domains are indicated. Residue 112 in XendoU is where the homology between the X. laevis and human proteins begins; the N termini are relatively unconserved. (B) Quantitative RT-PCR showing effective knockdown of EndoU with esiRNAs directed against the coding region and 3â² UTR. (C) HeLa cells were cotransfected with EndoU2-GFP and mCherry-Sec61Î², and analyzed by fluorescence microscopy. (D) HeLa cells were cotransfected with EndoU-short-GFP and mCherry-Sec61Î² and analyzed as in C. (E) Cells treated with control esiRNAs (left) or esiRNAs against the coding region (middle) or 3â² UTR (right) of EndoU were transfected with mCherry-Sec61Î² and analyzed by fluorescent microscopy. (F) Quantification of the percentage of cells exhibiting expanded ER sheets in control, EndoU coding, and EndoU 3â² UTR RNAi. Error bars indicate standard deviation. At least 200 cells were scored and the experiment was performed in triplicate. (G) HeLa cells were treated with control or EndoU 3â² UTR esiRNAs, then transfected with wild-type or catalytically dead EndoU2-GFP or EndoU-short-GFP (resistant to RNAi, lacking the 3â² UTR) and mCherry Sec61Î². The percentage of cells exhibiting expanded ER sheets is quantified. Error bars indicate standard deviation. At least 200 cells from three replicates of each condition were scored. Inset panels in C and D are magnified 3Ã. Bars: (main panels) 10 Âµm; (insets) 2 Âµm.
	Figure 7. Model of XendoU nuclease activity on membranes. (A) The ER network exists as a mixture of tubules and sheets. A decrease in XendoU results in the expansion of sheets. (B) Ribosomes, ribonucleoproteins (RNPs), XendoU, Atlastin, and (closed) calcium channels are localized to membrane vesicles containing Ca2+. Dimerization of Atlastin leads to eventual membrane fusion and calcium release through calcium channels on the membrane. XendoU binds calcium and mediates local RNA degradation (mRNAs, rRNAs, other RNAs), resulting in the release of ribosomes, RNPs, and RNA from the surface of membranes.

References [+] :

Allan, Protein phosphatase 1 regulates the cytoplasmic dynein-driven formation of endoplasmic reticulum networks in vitro. 1995, Pubmed, Xenbase