Translation of branched-chain aminotransferase-1 transcripts is impaired in cells haploinsufficient for ribosomal protein genes
2014; Elsevier BV; Volume: 42; Issue: 5 Linguagem: Inglês
10.1016/j.exphem.2013.12.010
ISSN1873-2399
AutoresTamara C. Pereboom, Albert Bondt, Paschalina Pallaki, Tim D. Klasson, Yvonne J. Goos, Paul Essers, Marian J.A. Groot Koerkamp, Hanna T. Gazda, Frank C. P. Holstege, Lydie Da Costa, Alyson W. MacInnes,
Tópico(s)Cancer-related gene regulation
ResumoDiamond-Blackfan anemia (DBA) is a bone marrow failure syndrome linked to mutations in ribosomal protein (RP) genes that result in the impaired proliferation of hematopoietic progenitor cells. The etiology of DBA is not completely understood; however, the ribosomal nature of the genes involved has led to speculation that these mutations may alter the landscape of messenger RNA (mRNA) translation. Here, we performed comparative microarray analysis of polysomal mRNA transcripts isolated from lymphoblastoid cell lines derived from DBA patients carrying various haploinsufficient mutations in either RPS19 or RPL11. Different spectrums of changes were observed depending on the mutant gene, with large differences found in RPS19 cells and very few in RPL11 cells. However, we find that the small number of altered transcripts in RPL11 overlap for the most part with those altered in RPS19 cells. We show specifically that levels of branched-chain aminotransferase-1 (BCAT1) transcripts are significantly decreased on the polysomes of both RPS19 and RPL11 cells and that translation of BCAT1 protein is especially impaired in cells with small RP gene mutations, and we provide evidence that this effect may be due in part to the unusually long 5'UTR of the BCAT1 transcript. The BCAT1 enzyme carries out the final step in the biosynthesis and the first step of degradation of the branched-chain amino acids leucine, isoleucine, and valine. Interestingly, several animal models of DBA have reported that leucine ameliorates the anemia phenotypes generated by RPS19 loss. Our study suggests that RP mutations affect the synthesis of specific proteins involved in regulating amino acid levels that are important for maintaining the normal proliferative capacity of hematopoietic cells. Diamond-Blackfan anemia (DBA) is a bone marrow failure syndrome linked to mutations in ribosomal protein (RP) genes that result in the impaired proliferation of hematopoietic progenitor cells. The etiology of DBA is not completely understood; however, the ribosomal nature of the genes involved has led to speculation that these mutations may alter the landscape of messenger RNA (mRNA) translation. Here, we performed comparative microarray analysis of polysomal mRNA transcripts isolated from lymphoblastoid cell lines derived from DBA patients carrying various haploinsufficient mutations in either RPS19 or RPL11. Different spectrums of changes were observed depending on the mutant gene, with large differences found in RPS19 cells and very few in RPL11 cells. However, we find that the small number of altered transcripts in RPL11 overlap for the most part with those altered in RPS19 cells. We show specifically that levels of branched-chain aminotransferase-1 (BCAT1) transcripts are significantly decreased on the polysomes of both RPS19 and RPL11 cells and that translation of BCAT1 protein is especially impaired in cells with small RP gene mutations, and we provide evidence that this effect may be due in part to the unusually long 5'UTR of the BCAT1 transcript. The BCAT1 enzyme carries out the final step in the biosynthesis and the first step of degradation of the branched-chain amino acids leucine, isoleucine, and valine. Interestingly, several animal models of DBA have reported that leucine ameliorates the anemia phenotypes generated by RPS19 loss. Our study suggests that RP mutations affect the synthesis of specific proteins involved in regulating amino acid levels that are important for maintaining the normal proliferative capacity of hematopoietic cells. Diamond-Blackfan anemia (DBA) is a congenital disorder that typically presents in the first year of life and is characterized by a lack of erythrocyte progenitor cells, red cell aplasia, bone marrow failure, and, in some cases, physical abnormalities [1Vlachos A. Muir E. How I treat Diamond-Blackfan anemia.Blood. 2010; 116: 3715-3723Crossref PubMed Scopus (125) Google Scholar]. There is also a significantly increased risk for cancer in patients with DBA, predominantly acute myeloid leukemia, although osteogenic sarcomas, colon carcinomas, and female genital cancers also have been reported [2Albers J. Rajski M. Schönenberger D. et al.Combined mutation of Vhl and Trp53 causes renal cysts and tumours in mice.EMBO Mol Med. 2013; 5: 949-964Crossref PubMed Scopus (44) Google Scholar, 3Vlachos A. Rosenberg P.S. Atsidaftos E. Alter B.P. Lipton J.M. Incidence of neoplasia in Diamond-Blackfan anemia: a report from the Diamond-Blackfan Anemia Registry.Blood. 2012; 119: 3815-3819Crossref PubMed Scopus (206) Google Scholar]. Approximately 70% of DBA patients carry mutations in ribosomal protein (RP) genes. Mutations are found in RPS19 in about 25% of all DBA cases, and mutations in RPS7, RPS10, RPS17, RPS24, RPS26, RPS27, RPS29, RPL5, RPL11, RPL27, RPL31, and RPL35a also have been reported at lower frequencies (as have two cases of patients with GATA1 mutations) [4Boria I. Garelli E. Gazda H.T. et al.The ribosomal basis of Diamond-Blackfan anemia: mutation and database update.Human Mutat. 2010; 31: 1269-1279Crossref PubMed Scopus (169) Google Scholar, 5Doherty L. Sheen M.R. Vlachos A. et al.Ribosomal protein genes RPS10 and RPS26 are commonly mutated in Diamond-Blackfan anemia.Am J Hum Genet. 2010; 86: 222-228Abstract Full Text Full Text PDF PubMed Scopus (179) Google Scholar, 6Farrar J.E. Nater M. Caywood E. et al.Abnormalities of the large ribosomal subunit protein, Rpl35a, in Diamond-Blackfan anemia.Blood. 2008; 112: 1582-1592Crossref PubMed Scopus (186) Google Scholar, 7Sankaran V.G. Ghazvinian R. Do R. et al.Exome sequencing identifies GATA1 mutations resulting in Diamond-Blackfan anemia.J Clin Invest. 2012; 122: 2439-2443Crossref PubMed Scopus (246) Google Scholar, 8Ito E. Yoshida Y. Yusuke O. et al.Identification of two new DBA genes, RPS27 and RPL27, by whole-exome sequencing in Diamond-Blackfan anemia patients.Blood (ASH Annual Meeting Abstracts). 2012; 120: 984Google Scholar, 9Fisher R. Henson A. Quarello P. et al.Insights into diagnosis and etiology of Diamond Blackfan anemia by analysis of pre-rRNA processing.Blood (ASH Annual Meeting Abstracts). 2012; 120: 3476Google Scholar, 10Mirabello L. Ballew B.J. Giri N. et al.RPS29 is mutated in a multi-case Diamond Blackfan anemia family.Blood (ASH Annual Meeting Abstracts). 2012; 120: 511Google Scholar]. These RP gene mutations are invariably haploinsufficient; however, they have been found in a variety of types including single-base changes, small insertions, and deletions as well as large genomic deletions [4Boria I. Garelli E. Gazda H.T. et al.The ribosomal basis of Diamond-Blackfan anemia: mutation and database update.Human Mutat. 2010; 31: 1269-1279Crossref PubMed Scopus (169) Google Scholar, 11Kuramitsu M. Sato-Otsubo A. Morio T. et al.Extensive gene deletions in Japanese patients with Diamond-Blackfan anemia.Blood. 2012; 119: 2376-2384Crossref PubMed Scopus (44) Google Scholar, 12Quarello P. Garelli E. Brusco A. et al.High frequency of ribosomal protein gene deletions in Italian Diamond Blackfan anemia patients detected by Multiplex Ligation-dependent Probe Amplification (MLPA) assay.Haematologica. 2012; 97: 1813-1817Crossref PubMed Scopus (44) Google Scholar]. The hematopoietic phenotypes of DBA vary dramatically. Anemia in some patients is relatively mild and treatable with corticosteroids. Other patients depend on chronic blood transfusions, and the most severe cases require hematopoietic stem cell transplantation [13Lipton J.M. Ellis S.R. Diamond-Blackfan anemia: diagnosis, treatment, and molecular pathogenesis.Hematol Oncol Clin North Am. 2009; 23: 261-282Abstract Full Text Full Text PDF PubMed Scopus (137) Google Scholar]. The phenotypes of DBA also depend to some extent on which RP gene is mutated; for example, physical defects, such as cleft palate and thumb abnormalities, tend to be observed only in patients with mutations in RPL5 or RPL11 [14Gazda H.T. Sheen M.R. Vlachos A. et al.Ribosomal protein L5 and L11 mutations are associated with cleft palate and abnormal thumbs in Diamond-Blackfan anemia patients.Am J Hum Genet. 2008; 83: 769-780Abstract Full Text Full Text PDF PubMed Scopus (307) Google Scholar]. In zebrafish models, it also has been established that the haploinsufficient loss of some but not all RP genes can lead to tumor development [15Amsterdam A. Sadler K.C. Lai K. et al.Many ribosomal protein genes are cancer genes in zebrafish.PLoS Biol. 2004; 2: E139Crossref PubMed Scopus (344) Google Scholar]. Until the very recent discovery of GATA1 mutations linked to DBA, RP gene mutations have been the sole focus of the genetics underlying the disease. The general belief that DBA is a disease of ribosome failure and translation impairment has led to its classification as a "ribosomopathy" along with other diseases linked to genes involved in ribosome biogenesis including Shwachman-Diamond syndrome, dyskeratosis congenita, and 5q-myelodysplastic syndrome [16Narla A. Ebert B.L. Ribosomopathies: human disorders of ribosome dysfunction.Blood. 2010; 115: 3196-3205Crossref PubMed Scopus (542) Google Scholar, 17Cmejlova J. Dolezalova L. Pospisilova D. Petrtylova K. Petrak J. Cmejla R. Translational efficiency in patients with Diamond-Blackfan anemia.Haematologica. 2006; 91: 1456-1464PubMed Google Scholar]. Mutations in RP genes found in DBA patients have been shown to impair the processing of rRNA and to effect the production of the large or small ribosomal subunit, depending on the RP mutation involved [18Moore 4th, J.B. Farrar J.E. Arceci R.J. Liu J.M. Ellis S.R. Distinct ribosome maturation defects in yeast models of Diamond-Blackfan anemia and Shwachman-Diamond syndrome.Haematologica. 2010; 95: 57-64Crossref PubMed Scopus (27) Google Scholar, 19Robledo S. Idol R.A. Crimmins D.L. Ladenson J.H. Mason P.J. Bessler M. The role of human ribosomal proteins in the maturation of rRNA and ribosome production.RNA. 2008; 14: 1918-1929Crossref PubMed Scopus (169) Google Scholar]. These results have suggested that not all RP gene mutations are equal, because some mutations have more severe effects on rRNA processing or the polysome profiles than others. Another example of this is the physical malformations mentioned previously that so far seem to manifest solely in patients with RPL5 and RPL11 mutations. The fact that DBA mutations in RP genes are so closely linked to the critical cellular process of translation has led to the hypothesis that one of the etiologies of DBA is due to an alteration of the mRNA transcripts (or transcriptome) loaded onto the translation machinery, namely ribosomes and polysomes. Although the translational capacity of cells with DBA-linked mutations overall is diminished compared with normal control cells [17Cmejlova J. Dolezalova L. Pospisilova D. Petrtylova K. Petrak J. Cmejla R. Translational efficiency in patients with Diamond-Blackfan anemia.Haematologica. 2006; 91: 1456-1464PubMed Google Scholar], there appears to be selective specificity in terms of which mRNAs are translated. For example, in murine erythroblasts transduced with RPS19 or RPL11 shRNA constructs, it has been previously shown that these cells preferentially translate specific sets of mRNAs, examples being the reduced translation of Bag1 and Csde1 [20Horos R. Ijspeert H. Pospisilova D. et al.Ribosomal deficiencies in Diamond-Blackfan anemia impair translation of transcripts essential for differentiation of murine and human erythroblasts.Blood. 2012; 119: 262-272Crossref PubMed Scopus (120) Google Scholar]. In the RP gene haploinsufficient zebrafish models of cancer, it also has been shown the tumor cells do not translate the tumor suppressor p53 transcript, despite no decrease in p53 mRNA levels, p53 gene mutations, or global translation rates [21MacInnes A.W. Amsterdam A. Whittaker C.A. Hopkins N. Lees J.A. Loss of p53 synthesis in zebrafish tumors with ribosomal protein gene mutations.Proc Natl Acad Sci U S A. 2008; 105: 10408-10413Crossref PubMed Scopus (110) Google Scholar]. It also has been shown in mice that mutation of RpL38 results in the downregulation of a specific set of Homeobox mRNAs with no global impairments of protein translation [22Kondrashov N. Pusic A. Stumpf C.R. et al.Ribosome-mediated specificity in Hox mRNA translation and vertebrate tissue patterning.Cell. 2011; 145: 383-397Abstract Full Text Full Text PDF PubMed Scopus (369) Google Scholar]. However, a systematic large-scale investigation of preferentially translated transcripts in normal human versus DBA patient cells with RP gene mutations has not been done to date. Normal and DBA patient lymphoblastoid cell lines (LCLs) were obtained after informed consent in accordance with the Declaration of Helsinki and with institutional review board approval at the Boston Children's Hospital. All cell lines were established in the laboratory of Dr. Hanna Gazda, using a standard Epstein-Barr virus (EBV) immortalization of isolated mononuclear cells [23Neitzel H. A routine method for the establishment of permanent growing lymphoblastoid cell lines.Hum Genet. 1986; 73: 320-326Crossref PubMed Scopus (556) Google Scholar]. Genotyping of seven unrelated DBA patients revealed the following mutations: RPS19 (107-2): acceptor splice site mutation IVS2-1g>a; Intron2. RPS19 (KB): acceptor splice site mutation IVS2-1g>c; Intron2. RPS19 (120-3) 13insA; Frameshift at codon 5, stop at 75. RPL11 (231-1) 160insA; Frameshift at codon 54, stop at 60. RPL11 (70-3) c.314_315delTT; Frameshift at codon 105, stop at 121. RPL11 247-3 Donor splice site IVS1 +2t>c; Intron1. RPS17 200delGA; Frameshift at codon 67, stop at 86. All cells were maintained in RPMI + 10% fetal calf serum (FCS) under standard conditions. Unless otherwise noted, most experiments were performed with RPS19 (KB) and RPL11 (120-3). All steps of this protocol were performed at 4°C or on ice. Gradients of 17% to 50% sucrose (11 mL) in gradient buffer (110 mM KAc, 20 mM MgAc, and 10 mM HEPES pH 7.6) were poured the evening before use. LCLs were lysed in 500 μl polysome lysis buffer (gradient buffer containing 100 mM KCl, 10 mM MgCl, 0.1% NP-40, and freshly added 2 mM DTT and 40 U/ml RNAsin [Promega, Leiden, The Netherlands]), using a Dounce tissue grinder (Wheaton, Millville, NJ, USA). The samples were centrifuged at 1200g for 10 min to remove debris. Protein concentrations of each sample were calculated by Bradford analysis (Sigma, Zwijndrecht, The Netherlands), equalized using lysis buffer, and loaded onto the sucrose gradients. The gradients were ultracentrifuged for 2 hours at 120,565g in an SW41 Ti rotor (Beckman-Coulter, Woerden, The Netherlands). The gradients were displaced into a UA6 absorbance reader (Teledyne ISCO, Wierde, Belgium) using a syringe pump (Brandel, Gaithersburg, MD, USA), containing 60% sucrose. Absorbance was recorded at an OD of 254 nm. Cells were plated on day 0 at 20,000 cells/well of a 12-well plate in 1mL RPMI + 10% FCS, each cell line plated in 12 wells. Each of three wells were counted in triplicate each day for 4 days using a CASY Cell Counter (Roche, Basel, Switzerland). RNA was isolated from the polysomal peaks of the displaced sucrose gradients using the Trizol LS reagent (Invitrogen, Bleiswijk, The Netherlands). Microarrays used included Human 70-mer oligos (Operon, Human V2 AROS) spotted onto Codelink Activated slides (Surmodics, Eden Prairie, MN, USA). RNA amplifications, labeling, and hybridizations were performed as previously described [24Roepman P. Wessels L.F. Kettelarij N. et al.An expression profile for diagnosis of lymph node metastases from primary head and neck squamous cell carcinomas.Nature Genet. 2005; 37: 182-186Crossref PubMed Scopus (332) Google Scholar]. cRNA (700 ng) coupled to Cy3 and Cy5 fluorophores (Amersham, Diegem, Belgium) were hybridized on a Tecan HS4800PRO and scanned on an Agilent scanner (G2565BA) at 100% laser power, 100% photomultiplier tubes. After data extraction using Imagene 8.0 (BioDiscovery, Hawthorne, CA, USA), print-tip Loess normalization was performed [25Yang Y.H. Dudoit S. Luu P. et al.Normalization for cDNA microarray data: a robust composite method addressing single and multiple slide systematic variation.Nucleid Acids Res. 2002; 30: e15Crossref PubMed Scopus (2812) Google Scholar] on mean spot intensities without background subtraction. Dye bias was corrected based on a within-set estimate as previously described [26Margaritis T Lijnzaad P. van Leenen D. et al.Adaptable gene-specific dye bias correction for two-channel DNA microarrays.Mol Syst Biol. 2009; 5: 266https://doi.org/10.1038/msb.2009.21Crossref PubMed Scopus (36) Google Scholar]. Data were analyzed using analysis of variance (R version 2.2.1/MAANOVA version 0.98-7) (http://www.r-project.org/). In a fixed-effect analysis, sample, array, and dye effects were modeled. p values were determined by a permutation F2-test in which residuals were shuffled 5,000 times globally. Genes with p < 0.05 after familywise error rate (FWER; or false discovery rate [FDR]) multiple-testing correction were considered significantly changed. Data files were uploaded to ArrayExpress under the accession number: E-MTAB-1427. GO terms were analyzed using Database for Annotation, Visualization and Integration Discovery (DAVID) [27Huang da W. Sherman B.T. Lempicki R.A. Systematic and integrative analysis of large gene lists using DAVID bioinformatics resources.Nat Protoc. 2009; 4: 44-57Crossref PubMed Scopus (24210) Google Scholar, 28Huang da W. Sherman B.T. Lempicki R.A. Bioinformatics enrichment tools: paths toward the comprehensive functional analysis of large gene lists.Nucleic Acids Res. 2009; 37: 1-13Crossref PubMed Scopus (9790) Google Scholar]. Antibodies against human BCAT1 protein were custom designed, made, purified, and enzyme-linked immunosorbent assay tested by Eurogentec (Belgium). Peptides containing the sequence DIQYGREESDWTIVLS (antigen code: EP112427-KLH-GLUTA) were used to immunize rabbits under the 3-month program. Lymphocyte-specific kinase (LCK) antibodies (433) and actin antibodies (1616) were purchased from Santa Cruz Biotech. The p53 antibody was purchased from Cell Signaling (2527S). RNA was isolated from the polysome peaks using Trizol LS (Invitrogen). The RNA pellet was resuspended in 20 μL H2O. All RNA was used to create cDNA using Cloned-AMV First Strand cDNA Synthesis Kit (Invitrogen) according to manufacturer's protocol using Oligo(dT)20 primer. Gene expression was analyzed using a MyiQ Single Color Real-Time polymerase chain reaction (PCR) Detection System on a Bio-Rad iCycler. 1 μL cDNA was used in combination with FastStart High Fidelity PCR reagents (Roche) and 1 μL 3,75x SYBR Green (Sigma) in a volume of 25 μL total. Human β-actin was amplified using primer pairs 5'-TTTTGAATGATGAGCCTTCG-3' and 5'-AGCCTTCATACATCTCAAGTTG-3', human BCAT1 using 5'-TTCAACTCGTGATACACCAA-3' and 5'-ATTCCTGTGCTAGAGAGCAT-3'. The data were analyzed as previously described [29Pereboom T.C. van Weele L.J. Bondt A. MacInnes A.W. A zebrafish model of dyskeratosis congenita reveals hematopoietic stem cell formation failure resulting from ribosomal protein-mediated p53 stabilization.Blood. 2011; 118: 5458-5465Crossref PubMed Scopus (56) Google Scholar]. Biological triplicates were used, and the ratio between BCAT1 and actin calculated, setting the ratio of normal controls to 1. For p53 Westerns, cells were subject to 25 grays γ-irradiation and then lysed 6 hours later, or subjected to 6 hours of 20 μM MG132 (Sigma) treatment, or both. Cells were lysed in buffer containing protease inhibitors cocktail tablets (Roche) and normalized to 20 μg per sample before being run on 8% Sodium Dodecyl Sulfate/Polyacrylamide gel electrophoresis (SDS/PAGE) gels and blotted with Lck (2787 Cell Signaling, Leiden, The Netherlands) or p53 (2527 Cell Signaling) at a dilution of 1:1000. Horseradish peroxidase-conjugated secondary antibodies (GE Healthcare, Diegem, Belgium) were diluted 1:5000. Detection was performed with enhanced chemiluminescence reagents (Promega). Total RNA was isolated from cells with Trizol (Invitrogen) and 10 μg run on a 1% agarose/formaldehyde gel and transferred to nitrocellulose membranes as previously described [21MacInnes A.W. Amsterdam A. Whittaker C.A. Hopkins N. Lees J.A. Loss of p53 synthesis in zebrafish tumors with ribosomal protein gene mutations.Proc Natl Acad Sci U S A. 2008; 105: 10408-10413Crossref PubMed Scopus (110) Google Scholar]. The I.M.A.G.E. clone IRATp970B1053D of human BCAT1 (Source Bioscience) was used as a template to amplify a PCR product with primers 5'-ACGGGCATCCCCACACTCGG-3' and 5'-TGCGGCTTGCCAGCTTAGGA-3'. cDNA generated from normal LCLs (Invitrogen) was used as a template to amplify a PCR product for actin with primers 5'-GGCATCCTCACCCTGAAGTA-3' and 5'-TGTTGGCGTACAGGTCTTTG-3'. These PCR products were labeled with 32P-dCTP (Perkin-Elmer, Groningen, The Netherlands), using the Random Primer Labeling Kit (Invitrogen). Blots were probed, washed, and exposed overnight to a Phospho-Screen (Molecular Dynamics, Sunnyvale, CA, USA) that was scanned by a Typhoon Trio+ Variable Mode Imager (GE Healthcare). Analysis of the blots was performed with Quantity One software (BioRad, Veenendaal, The Netherlands). LCLs were incubated in media without methionine or cysteine (Sigma); 45 min later, 200 μCi of [35S]-methionine/cysteine (Perkin-Elmer) was added to each flask for 30 min. Cells were lysed and 1 mg of protein was incubated overnight with an equal volume of BCAT antibody (or 3 μL actin antibody) at 4°C, then precipitated with Protein A/G Plus agarose beads (Calbiochem, San Diego, CA, USA) for 2 hours at 4°C. Samples were run on an 8% SDS/PAGE gel that was fixed in 10% acetic acid/45% ethanol/45% H20 for 1 hour, followed by 30 min in water and 1 hour in Amplify (GE Healthcare) before being dried (BioRad gel dryer) and exposed to autoradiography film (Kodak). Analysis was performed with a GS-800 densitometer and Quantity One software (BioRad). The following primers were used to PCR the 5'UTR of the BCAT1 gene (NM_001178094) from I.M.A.G.E. clone IRATp970B1053D (Source Bioscience, Berlin, Germany): Forward 5'-CGGAGCTCGGTGGATGCTGCGGCATCGG-3' and reverse 5'-CGAAGCTTACCGTGCGCTCCTCTCCAGG-3'. The PCR product was subcloned into the pGL4.10 luciferase vector (Promega) with SacI and HindIII restriction enzymes (Promega). LCLs were grown in three different flasks per line. One million cells per sample were collected and lysed in 100 μL buffer V (Amaxa Cell Line Nucleofector Kit V cat. No. VCA-1003), containing a mix of either pGL4.10 5'UTR BCAT1 (400 ng) and pRL-CMV (Promega) (100 ng) or pGL4.10 Tata (empty vector) (400 ng) and pRL-CMV (100 ng). Cells were transfected using Nucleofector (Lonza, Basel, Switzerland) program V-001 and plated in a six-well plate for 24 hours with RPMI media +10% FCS and 1% pen/strep. Dual-Luciferase Reporter Assay (Promega) was performed per the manufacturer's instructions. Polysomal microarray analysis was performed on EBV immortalized LCLs derived from either patients with DBA or normal controls. We selected three lines carrying RPS19 mutations (two with splice site mutations, one with a small insertion leading to an early stop codon) or RPL11 mutations (two with a small insertion or deletion leading to an early stop codon, one with a splice site mutation). To ensure that the EBV immortalization of these cells did not affect the ribosome biogenesis phenotypes of the various mutations, we performed sucrose density fractionations with the lysates from these cells along with three different normal control lines, and generated polysome profiles, followed by isolation of the mRNAs in the polysomal fractions. The experimental set-up is illustrated in Figure 1A. Polysome profiles from normal controls demonstrate that the 40S and 60S peaks are the same size as each other, indicating a balanced stoichiometry between small and large ribosomal subunits (Fig. 1B). In LCLs derived from DBA patients carrying RPS19 gene mutations, we found a decrease of the 40S peak compared with the 60S peak (Fig. 1C). In contrast, the profiles of patient LCLs carrying RPL11 gene mutations show a decrease in the 60S subunit peak, corresponding to the large ribosomal subunit (Fig. 1D). Additionally, in the profiles from the RPL11 LCLs, we observed the presence of halfmers, representing transcripts that are bound to single 40S subunits that have not joined with the 60S subunit (as would be expected given the temporal sequence of ribosome subunit binding to mRNAs and an imbalance of too few large ribosomal subunits). These profiles affirmed for us the use of their polysomal fractions for the subsequent mRNA isolation and microarray analysis. To characterize the lymphoblastoid cell lines further, we performed standard cell proliferation assays and found that both mutations in RPS19 and RPL11 result in slower growing cells compared with cells derived from a healthy control (Fig. 2A). To assess levels of apoptosis in the LCLs, we performed Western blot analysis using antibodies against the p53 tumor suppressor protein. The results revealed that the basal levels of p53 stabilization are higher in cells only with RPS19 mutations, suggesting that the slower growth of RPL11 LCLs is not a result of increased apoptosis (Fig. 2B). The application of DNA damage by γ-irradiation and/or the inhibition of the proteasome in cells with a mutation in RPS19 resulted in a substantial increase in p53 stabilization in both normal and RPS19 mutant cells, suggesting that the p53 pathway is still intact in the mutant cells and that the basal level of p53 stabilization in the mutant cells is still well below the maximum response (Supplementary Figure E1; online only, available at www.exphem.org). mRNA transcripts were isolated from the polysomal fractions of the six DBA patient lines and three independent controls. Each sample was analyzed in biological duplicate, resulting in 18 microarrays. FWER identified a set of 77 transcripts that were differentially loaded (FDR < 0.05) onto polysomes of RPS19 and/or RPL11 cells compared with normal control cells. Cluster analysis of the set illustrates that the vast majority of differentially loaded transcripts were found in the RPS19 cells, whereas the polysomal transcripts of the RPL11 cells predominantly clustered within the normal controls (Fig. 3A). This indicates that in terms of loading mRNA transcripts onto polysomes, the mutations in the RPS19 gene exert a stronger deregulation than the mutations in the RPL11 gene, which do not appear to affect transcript loading in a significant manner. We performed FWER and FDR statistical analysis on the microarray analysis to determine which transcripts in DBA patient polysomal fractions were significantly altered compared with their levels in normal control cells. The number of transcripts with altered polysomal expression in RPL11 cells compared with controls was surprisingly low; less than 10 transcripts were significantly changed using either statistical method. This is in contrast to the number of changes found in the polysomal fractions of RPS19 cells compared with controls, which showed a difference of either more than 70 altered transcripts using FWER (Supplementary Table E1; online only, available at www.exphem.org) or more than 350 when using FDR (Supplementary Table E2; online only, available at www.exphem.org). This dramatic difference between RPS19 and RPL11 polysomal fractions is not unexpected given the microarray clustering analysis. However, we found that despite the considerably low number of altered transcripts in the RPL11 polysomal fractions compared with normal controls, most of them overlapped with transcripts found altered in RPS19 polysomal fractions. This is depicted in the Venn diagrams shown in Figures 3B and 3C, using analysis from the more rigorous FWER method. We then analyzed the gene ontology (GO) terms of the altered transcripts using FWER (Fig. 3D). No significant terms were found in the RPL11 lists subject to analysis (most likely due to the very small number of altered transcripts). In contrast, the GO analysis of the RPS19 cells reveals significant changes in several categories of gene functions (Fig. 3D). One increased set of genes being translated on polysomes has a GO function related to the induction of apoptosis, which is consistent with the well-established role of p53 tumor suppressor protein stabilization in response to RP gene mutations [30Dutt S. Narla A. Lin K. et al.Haploinsufficiency for ribosomal protein genes causes selective activation of p53 in human erythroid progenitor cells.Blood. 2011; 117: 2567-2576Crossref PubMed Scopus (289) Google Scholar]. Figure 3D also reveals that a set of downregulated transcripts in RPS19 cells has a GO function related to amino acid biosynthesis, which we explored further. Finally, the GO terms reveal a large change in transcripts that are likely unique to lymphoblastoid cells, including lymphocyte activation, antigen presentation, and immunoglobulin-mediated immune response, which we did not pursue. Table 1 lists the overlapping transcripts, the levels of which are found significantly altered in both RPS19 and RPL11 polysomal fractions compared with controls. The only transcript found increased in both RPS19 and RPL11 polysomal fractions is the p56 LCK. We used this data to verify with Western blot analysis that protein levels of LCK were in fact increased in mutant DBA LCLs, including several mutant lines not used for the microarray analysis (Fig. 4A). Given the lymphocytic nature of the cell lines used for immortalization, we suspected that this result may be cell line-specific, and although the observation provided a good biological validation of the microarray results, p56 LCK was not considered for further analysis.Table 1Significantly altered transcripts in mutant RPS19 and RPL11 polys
Referência(s)