Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2009 Aug;175(2):748-62.
doi: 10.2353/ajpath.2009.090047. Epub 2009 Jul 9.

Expression of RNA CCUG repeats dysregulates translation and degradation of proteins in myotonic dystrophy 2 patients

Elizabeth Salisbury  1 , Benedikt Schoser, Christiane Schneider-Gold, Guo-Li Wang, Claudia Huichalaf, Bingwen Jin, Mario Sirito, Partha Sarkar, Ralf Krahe, Nikolai A Timchenko, Lubov T Timchenko
Affiliations

Expression of RNA CCUG repeats dysregulates translation and degradation of proteins in myotonic dystrophy 2 patients

Elizabeth Salisbury et al. Am J Pathol. 2009 Aug.

Abstract

Myotonic dystrophy 2 (DM2) is a multisystem skeletal muscle disease caused by an expansion of tetranucleotide CCTG repeats, the transcription of which results in the accumulation of untranslated CCUG RNA. In this study, we report that CCUG repeats both bind to and misregulate the biological functions of cytoplasmic multiprotein complexes. Two CCUG-interacting complexes were subsequently purified and analyzed. A major component of one of the complexes was found to be the 20S catalytic core complex of the proteasome. The second complex was found to contain CUG triplet repeat RNA-binding protein 1 (CUGBP1) and the translation initiation factor eIF2. Consistent with the biological functions of the 20S proteasome and the CUGBP1-eIF2 complexes, the stability of short-lived proteins and the levels of the translational targets of CUGBP1 were shown to be elevated in DM2 myoblasts. We found that the overexpression of CCUG repeats in human myoblasts from unaffected patients, in C2C12 myoblasts, and in a DM2 mouse model alters protein translation and degradation, similar to the alterations observed in DM2 patients. Taken together, these findings show that RNA CCUG repeats misregulate protein turnover on both the levels of translation and proteasome-mediated protein degradation.

PubMed Disclaimer

Figures

Figure 1
Figure 1
CCUG RNA repeats form RNA-protein complexes in the cytoplasm of myoblasts derived from DM2 patients. A: FISH hybridization detects CCUG RNA (green) in cytoplasm and in nuclei. The fluorescence in situ hybridization assay was performed with myoblasts from unaffected patients (control) and with myoblasts from DM2 patients. Bottom images show DAPI staining. Arrows show cytoplasmic foci generated by the expanded CCUG repeats. B: Northern blot analysis of RNA isolated from cytoplasm and from nuclei of normal (N) and DM2 myoblasts with CAGG16 probe. Bottom images show ethidium bromide staining of the gels. The 28S and 18S ribosomal RNAs show equal loading and integrity of the used RNAs. C: Detection of RNA CCUG repeats in nuclear and cytoplasmic RNAs of DM2 myoblasts by dot-blot hybridization. RNA, 5, 10, and 20 μg, were loaded on the filter and hybridized with CAGG16 probe. D: CUG and CCUG RNAs contain common elements, GC islands (shown in red). E: CCUG16 RNA forms three RNA-protein complexes with cytoplasmic extracts from C2C12 myotubes. EMSA was performed with cytoplasmic proteins isolated from C2C12 myotubes with CUG8, CCUG8, and CCUG16 probes, shown on the top of the gel. Arrows show three RNA-protein complexes formed with CCUG16 RNA. F: Complexes 1 and 2 contain CUGBP1. Increasing amounts of non-radioactive CCUG16 probe and Abs to CUGBP1 were incorporated in the binding reactions with cytoplasm and [32P]-CCUG16 probe. G: Homogenous MBP-CUGBP1 binds to CCUG RNA. UV cross-link was performed with CUG8, CCUG8, and CCUG16 probes. The filter was stained with Coomassie blue to verify equal loading of the MBP-CUGBP1. H: CUGBP1 binds to the long CCUG RNA. Increasing amounts of CUGBP1 were incubated with the long CCUG300 probe, treated with UV, and separated by denaturing gel. Upper image shows radioactive signals; the bottom image shows Coomassie stain of the same membrane. I: Amounts of high molecular weight complex 3 are increased in cytoplasm of DM2 patients. Cytoplasmic protein extracts from myoblasts derived from unaffected patients (control) and DM2 patients were examined by EMSA with CCUG16 probe. The bottom image shows results of EMSA analysis. Positions of high molecular weight complex 3 and free probe are shown by arrows. Upper panel shows a Coomassie stain of the fragment of independent gel with the same loadings of the cytoplasmic proteins as used for the EMSA.
Figure 2
Figure 2
Purification of complex 3. A: A diagram showing the strategy for the isolation of the complex. Twenty milligrams of cytoplasmic proteins from livers of wild-type aged mice were used for the isolation of the high molecular weight complexes. B: Detection of the complex 3 in each step of purification by EMSA/gel shift assay with CCUG16 RNA. The fractions containing complex 3 and complexes 3–1 and 3–2 are underlined. C: A representative picture showing the analysis of protein fractions after the second step of purification by UV cross-link with CCUG16 RNA and by Western blotting with antibodies to CUGBP1. A major portion of complex 3–1 is located in fractions marked with red line. The position of CUGBP1 in the fractions examined by UV cross-link is shown by open circles. D: Coomassie staining of proteins of complex 3–1 after size exclusion chromatography. The identity of these proteins was determined by mass spectroscopy. E: UV cross-link analysis of the purified complex. The purified complex was incubated with the CCUG16 probe and examined by UV cross-link assay. The arrow shows a position of CUGBP1. F: The composition of the purified complex 3–1 determined by Western blotting with antibodies shown on the left. Western blotting analysis of two preparations of the complex 3–1 is shown. G: The purified complex 3–1 (CUGBP1-eIF2) increases translation of C/EBPβ in a cell-free translation system. Two different preparations of the CUGBP1-eIF2 complex were added in the reticulocyte lysate programmed with C/EBPβ mRNA. Positions of LAP and LIP isoforms of C/EBPβ are shown.
Figure 3
Figure 3
DM2 muscle cells contain high amounts of the CUGBP1-eIF2 complex. A: The amount of CUGBP-eIF2 complex is increased in DM2 myoblasts. CUGBP1 was immunoprecipitated from the cytoplasmic extracts of normal control and DM2 myoblasts and examined by Western blotting with antibodies to eIF2α, eIF2β, and eR99. IgG; signals of the heavy chains of IgGs. B: The binding of complex 3 to CCUG16 and C/EBPβ mRNA is increased in DM2 myoblasts. EMSA was performed with C/EBPβ, p21, CCUG16, and c-fos probes with cytoplasmic proteins from myoblasts derived from unaffected patients (normal) and from patients with DM2. The position of the CUGBP1-eIF2 complex is shown by the arrow. C: Protein levels of CUGBP1 and its translational downstream targets, MEF2A and C/EBPβ, are increased in myoblasts from DM2 patients. Western blotting analyses of myoblasts from unaffected (normal) and DM2 patients with antibodies against CUGBP1, MEF2A, and C/EBPβ are shown. Membranes were re-probed with antibodies against β-actin. Low-mobility CUGBP1 isoforms migrating in the positions of hyperphosphorylated CUGBP1 are shown by red arrows. D: Protein levels of MEF2A and C/EBPβ are increased in DM2 muscle biopsies. Cytoplasmic (for CUGBP1) and nuclear extracts (for MEF2A and C/EBPβ) were examined by Western blotting with antibodies shown on the right. The membranes were re-probed with anti-β-actin.
Figure 4
Figure 4
CUGBP1 is associated with the expanded CCUG repeats in the cytoplasm of DM2 myoblasts and in cytoplasm of livers of CCTG TR mice. A: CUGBP1 is bound to the mutant CCUG RNA in cytoplasm of DM2 myoblasts. Cytoplasmic proteins from myoblasts derived from unaffected (normal) patients and patients with DM2 were fractionated by gel filtration. Western blot analysis of gel filtration fractions was performed with antibodies to CUGBP1 and β-actin. RNA from the same fractions was examined by blot hybridization with CAGG16 probe. B: Ectopically expressed CCUG300 RNA binds to the CUGBP1-eIF2 complex. Cytoplasmic proteins from C2C12 myoblasts transfected with an empty vector (control) and from cells transfected with a plasmid expressing CCUG300 were fractionated on size exclusion column SEC400. Optical density (OD280) profiles of gel filtrations are shown on the top. Gel filtration fractions were examined by UV cross-link with CCUG16 probe, by Western blotting with CUGBP1 and eIF2α antibodies, and by blot hybridization with CAGG16 probe (CCUG). The position of the CUGBP1-eIF2 complex associated with the CCUG repeats is shown below. C: CUGBP1 levels are increased in the livers of CCTG-transgenic mice. Western blot analyses of cytoplasmic liver extracts from three wild-type (WT) and three CCTG TR mice were performed with antibodies to CUGBP1 and β-actin. Bar graphs show the levels of CUGBP1 as ratios to β-actin. D: Complex 3 is abundant in the livers of CCTG-transgenic mice. Cytoplasmic proteins from wild-type and CCTG TR mice were fractionated on SEC400 column. Top image shows EMSA with CCUG16 probe, which located complex 3 within gel filtration fractions. The positions of complex 3 and free probe are shown by arrowheads. Western: Gel filtration fractions were analyzed by Western blotting with antibodies to CUGBP1, eIF2α and β-actin (control). Bottom image shows results of blot hybridization of the fractions with CAGG16 probe.
Figure 5
Figure 5
Complex 3–2 contains the 20S catalytic core of the proteasome. A: Coomassie staining of the complex 3–2 after CHT1 column. Identity of proteins within the complex was determined by mass spectroscopy. Positions of protein markers are shown on left. B: The purified CCUG-binding complex 3–2 contains 20S proteasome. The purified complex was tested by Western blot assay with antibodies against subunits α2, α4 and α7 of the 20S proteasome. Cytoplasmic protein extract from C2C12 myotubes (cyto) and anti-HuR were used as controls. C: Hsp70 protein is associated with the proteasome complex. Proteasome complex was immunoprecipitated from the cytoplasm of old mouse livers by Abs to the α4 subunit of the 20S proteasome and Hsp70 was examined in this IP by Western blotting. Third lane shows Hsp70 in the α4-IP from cytoplasmic liver extract treated with ATP. IgG; heavy chains of IgGs used for IP. The filter was re-probed with Abs to α2 and α4 subunits of the 20S proteasome. D: Complex 3–2 directly binds to RNA CCUG repeats. Left image: UV cross-link assay of the 20S and 26S proteasomes (BioMol International, LP) and the complex 3–2 with RNA CCUG16. Right image shows the UV cross-link of these complexes with C/EBPβ RNA probe (negative control). E and F: Complex 3–2 contains proteins of 20S proteasome. Proteins were fractionated by ion exchange chromatography and fractions were examined by gel shift assay with CCUG16 probe (upper) and by Western blot with antibodies against α2, α4, α6, and α7 subunits of the 20S proteasome (middle). The fragment of gel containing the complex 3–2 was cut out, loaded onto a 4 to 20% polyacrylamide gel and examined by Western blotting with antibodies against α2, α4 and α7 subunits of the 20S proteasome. G: RNA CCUG repeats bind with the 20S core complex in DM2. FISH hybridization with CAGG16 probe (green) was performed using myoblasts from DM2 patients and normal control myoblasts following immunofluorescent analyses (top) with antibodies against the α4 subunit of the 20S proteasome. The positions of the proteasome associated with CCUG repeats in the cytoplasm are shown by arrows.
Figure 6
Figure 6
Proteasome activity is inhibited in DM2 myoblasts leading to the increased stability of short-lived proteins in DM2. A: Hsp70–20S proteasome complex is increased in DM2 myoblasts. Proteasome was immunoprecipitated from myoblasts derived from unaffected (N) and DM2 patients and probed with Abs to Hsp70 (upper panel). Two middle panels show examination of α2 and α7 subunits of the 20S proteasome within α4 IPs. Bottom panel shows Coomassie staining of heavy chains of immunoglobulins (IgGs) used for IP. First lane in all panels is a control with preimmune serum. Fourth lane in all panels shows the results of IP-Western with cytoplasmic extract from DM2 myoblasts containing ATP. B: Western blot analyses of control (N) and DM2 myoblasts with antibodies to p21, c-myc and CUGBP1. Protein loading was verified by Western blotting with antibodies against β-actin. C: The activity of proteasome is inhibited in DM2 myoblasts. Control and DM2 myoblasts were treated with MG132 and protein levels of p21, c-myc and CUGBP1 were examined. β-actin or cross-reactive molecule (CRM) signals show equal protein loading. D: Amounts of ubiquitin-protein conjugates are increased in DM2 myoblasts. Total protein extracts from untreated control and DM2 myoblasts and myoblasts treated with MG132 were examined by Western blotting with antibodies to ubiquitin. Bottom image shows a Coomassie staining of the filter. E: Amounts of p21-Ub conjugates are increased in DM2 myoblasts. Upper image shows levels of p21 in extracts used for detection of ubiquitin-p21 conjugates. Middle image: p21 was immunoprecipitated from control myoblasts derived from unaffected patients and from DM2 myoblasts and p21 IPs were probed with antibodies to ubiquitin. Bottom image shows Coomassie stain of heavy chains of immunoglobulins (IgG) after analysis of p21-IPs. F: Stability of p21, c-myc and CUGBP1 proteins is increased in DM2 myoblasts. Control and DM2 myoblasts were treated with cycloheximide to inhibit protein translation. Proteins were analyzed by Western blotting assay with antibodies against p21, c-myc and CUGBP1. 20 μg of proteins from DM2 myoblasts and 50 μg of proteins from control myoblasts were used to compensate for the increase of p21, c-myc and CUGBP1 in DM2. Membranes were re-probed with anti-β-actin to verify protein loading. G: Protein levels of p21, c-myc, and CUGBP1 were calculated as ratios to β-actin in control (red) and DM2 (blue) myoblasts. The picture shows percent (y axis) of each protein at 1, 2, and 4 hours (x axis) after addition of cycloheximide relative to 0 time point.
Figure 7
Figure 7
Expression of CCUG RNA repeats in control myoblasts from unaffected patients and from C2C12 myoblasts increases the levels of C/EBPβ isoforms and stabilizes c-myc, p21 and CUGBP1 proteins. A: Transfected CCUG RNA is located in nuclei and in cytoplasm. Upper. C2C12 myoblasts were transfected with a plasmid expressing CCUG36 repeats. FISH hybridization with CAGG16 probe (green) was performed. Cytoplasmic CCUG RNA is shown by arrows. Bottom images show a merge of nuclei stained with DAPI (blue) and CCUG RNA. B: Protein levels of C/EBPβ, p21, c-myc and CUGBP1 are increased in human primary myoblasts derived from unaffected patients transfected with RNA CCUG repeats. Western blotting was performed with total protein extracts isolated from control human myoblasts transfected with RNA CCUG16 and with control AU42 RNA. Un; protein extracts from un-transfected cells. C: CCUG repeats increase the half-life of p21, c-myc and CUGBP1. Control human myoblasts were transfected with control AU-rich RNA (AU42) and CCUG16 RNA, protein synthesis was blocked by cycloheximide and protein levels of p21, c-myc, CUGBP1 and C/EBPβ were determined. The bottom part shows a similar experiment in mouse C2C12 myoblasts containing higher levels of C/EBPβ isoforms LAP and LIP. D: Bar graphs show the half-life of the proteins calculated as a percentage of the 0 time point. E: A hypothetic model suggesting the role of RNA CCUG repeats in the disruption of protein synthesis and protein degradation in DM2 muscle.
See this image and copyright information in PMC

Similar articles

See all similar articles

Cited by

See all "Cited by" articles

References

    1. Harper PS. Myotonic dystrophy. London: W.B. Saunders,; 2001
    1. Ranum LP, Day JW. Myotonic dystrophy: RNA pathogenesis comes into focus. Am J Hum Genet. 2004;74:793–804. - PMC - PubMed
    1. Aslanidis C, Jansen G, Amemiya C, Shutler G, Mahadevan M, Tsilfidis C, Clen C, Alleman J, Wormskamp NG, Vooijs M, Buxton J, Johnson K, Sweets HLM, Lennon GG, Carrano AV, Korneluk RG, Wieringa B, deJong PJ. Cloning of essential myotonic dystrophy region and mapping of the putative defect. Nature. 1992;355:548–551. - PubMed
    1. Liquori CL, Ricker K, Moseley ML, Jacobsen JF, Kress W, Naylor SL, Day JW, Ranum L. Myotonic dystrophy type 2 caused by a CCTG expansion in intron of ZNF9. Science. 2001;293:864–867. - PubMed
    1. Day JW, Ricker K, Jacobsen JF, Rasmussen LJ, Dick KA, Kress W, Schneider C, Koch MC, Beilman GJ, Harrison , Dalton JC, Ranum LP. Myotonic dystrophy type 2: molecular, diagnostic and clinical spectrum. Neurol. 2003;60:657–664. - PubMed

Publication types