Binyamin D. Berkovits and Christine Mayr
Department of Cancer Biology and Genetics, Memorial Sloan Kettering Cancer Center, New York, NY
About half of human genes use alternative cleavage and polyadenylation (ApA) to generate mRNA transcripts that differ in the length of their 3' untranslated regions (3'UTRs) while producing the same protein. Here, we show that alternative 3'UTRs can differentially regulate the localization of membrane proteins. We demonstrate that the long 3'UTR of CD47 enables cell surface expression of CD47 protein, whereas the short 3'UTR promotes localization of CD47 protein to the endoplasmic reticulum. CD47 protein localization occurs post-translationally and independently of RNA localization. In this new pathway, the long 3'UTR of CD47 acts as a scaffold to recruit a protein complex consisting of the RNA-binding protein HuR, SET and ANP32A to the site of translation. This facilitates binding of SET and ANP32A to the newly translated cytoplasmic tail of CD47 and subsequent translocation of CD47 to the plasma membrane via activated RAC1. We also show that CD47 protein can have opposite functions regarding cell survival depending on whether it was generated by the short or long 3'UTR isoform. Thus, ApA contributes to the functional diversity of the proteome without changing the amino acid sequence. 3'UTR-dependent protein localization is not restricted to the regulation of CD47 localization. Also, the long 3'UTRs of CD44, ITGA1 and TNFRSF13C which are bound by HuR increase surface protein expression compared to their corresponding short 3'UTRs. It was shown that HuR binds to thousands of mRNAs 6-10 with about a third of them encoding membrane proteins. Therefore, we propose that 3'UTR-dependent protein localization is a widely used mechanism to control cell surface molecule expression and can be regulated by ApA.
Centro de Investigaciones Microbiolóas. Instituto de Ciencias. Beneméta Universidad Autóa de Puebla. Puebla, Méco.
The quality control of protein translation is essential for the appropriate cell physiology and survival. In eukaryotic cells, some mRNA molecules leave the nucleus after splicing but elicit decay in an EJC- or DSE-dependent manner for mammals and yeasts respectively, in order to arrest translation. The molecular mechanisms regulating NMD are still not fully understood, but recent data suggest that these processes are highly conserved in different organisms. Ustilago maydis is a basidiomycete that has been used as a model to study different molecular and cellular eukaryotic mechanisms. We have identified in Ustilago maydis the putative proteins that regulate NMD in this model. When comparing U. maydis and H. sapiens, identity for most putative NMD factors ranges between 40% and 60%. Interestingly, the most striking similarities at the protein level corresponded to the NMD factors eRF1 (71%), Upf1 (66%) and MAGOH (66%). NMD factors that have been described for human, but that are absent in the classic yeast model S. cerevisiae, were also found in U. maydis (MAGOH, SMG1, SMG5-7). Few NMD factors described in S. cerevisiae but absent in humans, are also found in U. maydis (Ski 7). Other factors are highly conserved as well. Moreover, almost 50% of genes in U. maydis are interrupted with introns, being intron retention the prevalent alternative splicing event, leading to a high probability of aberrant transcript generation, which could be directed to NMD. In order to provide evidence supporting that the homologs identified could perform equivalent biological activities, we performed molecular dynamics assays, including contact maps and dynamic simulations for UPF1 proteins from H. sapiens and U. maydis. Our results indicate that the putative factors identified in U. maydis show a very similar structure, mechanic stability, physicochemical properties and spatial organization in comparison to the human homolog. On the other hand, PCR experiments showed that the NMD factors identified are present in the genome of U. maydis and that these factors are constitutively expressed, as shown by RT-PCR assays. Moreover, NMD targets were also identified when the fungal culture is treated with emetin. Finally, a fungal mutant lacking Upf1 was obtained and resulted sub-lethal. Taken together, our results suggest an important degree of conservation between human and fungal control of NMD and led us propose that U. maydis could present both mechanisms observed in human and yeasts.
Centro de Investigaciones Microbiolóas. Instituto de Ciencias. Beneméta Universidad Autóa de Puebla. Puebla, Méco
Ustilago maydis is a dimorphic basidiomycete fungus that causes the carbon disease in corn commonly known in Mexico as "cuitlacoche" since old Aztec times. This model has been useful in the study of cellular processes such as DNA repair and recombination, given the fact that they are highly conserved between mammals and higher basidiomycetes. Interestingly, splicing is a rather common mechanism occurring in Ustilago maydis, given that around 50% of its genes are interrupted and the prevailing mechanism for alternative splicing is intron retention. Regulatory cis-elements for most of the U. maydis introns are very similar to those described for human pre-mRNAs both in sequence composition and relative position. Data annotated for the genome of Ustilago maydis show a striking degree of similarity with human splicing factors. When comparing Ustilago and human protein sequences, identity for most putative splicing factors ranges between 40% and 60%. The most conserved factors were the core of snRNPs (Sm and Lsm proteins), as well as some components of snRNP U1 (U1-70K), snRNP U2 (SF3b complex) and snRNP U5. Moreover, protein modeling showed that secondary and tertiary structures are nearly identical between the homologs, suggesting that they could be functionally equivalent. Other conserved proteins involved in splicing regulation include CBP80, U2AF35 and U2AF65. Alternative splicing regulators were also identified, but while some SR proteins are highly similar, some hnRNP proteins are barely conserved. As for the human genes, alternative splicing regulation for the expression of different factors was also founded. For example, the hypothetical hnRNP A1 from Ustilago maydis corresponds to the locus UM02420. This putative protein shows an identity of 40% with the human hnRNP A1, with conserved RRM domains. Locus UM02420 is organized in 4 exons and 3 introns. RT-PCR experiments confirmed that alternative splicing actually occurs for this mRNA. Further experiments will be needed to analyze the regulation of this alternative splicing event and the functionality of the spliced products. Altogether, our results suggest that Ustilago maydis could be considered as a new fungal model to study the complex regulation of human splicing events.
1 Berlin Institute for Medical Systems Biology, Max Delbrünter for Molecular Medicine, Robert-Röe-Strasse 10, 13125 Berlin, DE;
2 Institute for Genome Sciences and Policy, Duke University Medical Center, Durham, NC 27708, USA;
3 Genetics and Development Biology Center, National Heart, Lung and Blood Institute, National Institutes of Health, Bethesda, MD 20892, USA
The splicing and maturation of primary RNA transcripts to mature RNA is a highly regulated and dynamic process coordinated by numerous ribonucleoprotein (RNP) complexes. Decades of research into pre-mRNA splicing have yielded mechanistic details regarding the biochemistry of spliceosome assembly, the recognition of splice sites, the definition of exons and/or introns, and the subsequent removal of introns to create mature mRNA. In addition to small nuclear RNPs (snRNPs) and RBP components of the major spliceosome, such as U2AF1 and SF1, there are numerous regulatory RBPs that can either promote or repress the usage of a splice site, thus determining the final mature transcript. The goal of this study is to investigate mechanisms governing splice site usage and efficiency via pulse-labeling of RNAs during a 60-minute time course. We conducted metabolic labeling of RNA with 4-thiouridine (4SU) for 7.5, 15, 30, and 60 minutes after 4SU incorporation followed by deep directional paired-end RNA-seq (~40 million read pairs per sample). We employ complementary approaches to examine splicing and stability of RNAs at the level of transcripts and individual splice-sites. Non-coding RNAs exhibit stronger correlations between splice-site usage, splice-site sequence strength, and intron length than protein-coding genes. Protein-coding transcripts are generally more completely spliced and abundant than non-coding transcripts. This is reinforced by primary transcript expression estimates indicating slower metabolism of non-coding genes relative to protein-coding genes. We also investigate the splicing kinetics of protein-coding-gene introns containing non-coding RNAs. Specifically, mirtron containing introns appear more spliced after 7.5 minutes than other protein-coding introns of similar length. We will integrate RBP binding data from numerous splicing factors to identify cases of positive or negative regulation of splice site usage not explained by splice site strength and/or intron length.
Center for Thrombosis and Hemostasis (CTH) University Medical Center of the Johannes Gutenberg University, Mainz, Gemany
Current high throughput approaches for RNA isoform profiling (i.e. RNA-seq and microarray) provide opportunities to gain comprehensive and high quality data, yet pose serious challenges in terms of bioinformatic analysis. In order to further characterize alternative isoforms of representative transcripts, independent non-high throughput methods are used (e.g. qRT-PCR, RACE, northern blot, etc.). Each of these methods has both advantages and limitations. For example, multiple sets of primers are needed to characterize alternative transcript isoforms by qRT-PCR. Moreover, specificity becomes an issue for this method when working with families of genes with high (up to 95%) homology. Additionally, RACE (Rapid amplification of cDNA ends) approaches cannot detect alternative events within the cores of transcripts. However the transcriptome diversity has emerged to be much wider than previously expected and many transcript isoforms are yet to be discovered. Considering this, direct visualization of transcript variants and their dynamic changes in a fast, sensitive and specific manner without employing bioinformatic analysis would be highly desirable.
Northern blotting has previously been modified to omit radioactivity, but further advancement have become secondary due to the advent of novel high throughput techniques. Here we present a custom protocol for non-radioactive, highly sensitive northern blotting, which can detect endogenous transcripts in as low as 1 µg of total RNA. A modified probe generation process excludes the competing antisense DNA strand thus further increasing specificity and sensitivity. The protocol can easily be scaled up for screening larger numbers of transcripts in a short timeframe avoiding bioinformatic burden for data analysis. In a proof of concept experiment we have used this protocol to detect and quantify the dynamics of previously unannotated endogenous transcript isoforms in a wide screen of targets. Taken together, the current protocol enables the performance of fast, highly sensitive, non-radioactive, scalable and cost effective northern blotting. limitations. For example, multiple sets of primers are needed to characterize alternative transcript isoforms by qRT-PCR. Moreover, specificity becomes an issue for this method when working with families of genes with high (up to 95%) homology. Additionally, RACE (Rapid amplification of cDNA ends) approaches cannot detect alternative events within the cores of transcripts. However the transcriptome diversity has emerged to be much wider than previously expected and many transcript isoforms are yet to be discovered. Considering this, direct visualization of transcript variants and their dynamic changes in a fast, sensitive and specific manner without employing bioinformatic analysis would be highly desirable.
Northern blotting has previously been modified to omit radioactivity, but further advancement have become secondary due to the advent of novel high throughput techniques. Here we present a custom protocol for non-radioactive, highly sensitive northern blotting, which can detect endogenous transcripts in as low as 1 µg of total RNA. A modified probe generation process excludes the competing antisense DNA strand thus further increasing specificity and sensitivity. The protocol can easily be scaled up for screening larger numbers of transcripts in a short timeframe avoiding bioinformatic burden for data analysis. In a proof of concept experiment we have used this protocol to detect and quantify the dynamics of previously unannotated endogenous transcript isoforms in a wide screen of targets. Taken together, the current protocol enables the performance of fast, highly sensitive, non-radioactive, scalable and cost effective northern blotting.
Cold Spring Harbor Laboratory Library and Archives, Cold Spring Harbor, NY
mRNA is at the crossroads of several different research traditions, including molecular biology, cell biology, structural biology, biotechnology, and biomedicine. As an institution that has been at the forefront of molecular biology and genetics research and education since 1890, CSHL has made valuable contributions to the development of mRNA. Notable meetings and courses that played a key role in discussions about mRNA, as well as presentation of research results, are the Cold Spring Harbor Symposia on Quantitative Biology and the famed Phage Course, which was taught each summer at the Lab from 1945 - 1970.
The CSHL Archives houses a rich repository of rare books, manuscripts, photographs, and scientific reprints documenting research, meetings, courses, and life at the Lab since 1890. The Archives makes its collection available to scholars, graduate students and writers interested in the history of molecular biology and genetics. In this poster, we present a collection of photos, figures, manuscripts, reprints, and other materials related to the history of mRNA at CSHL. These materials highlight some of the key discussions and discoveries that have advanced our knowledge of the structure, function, and importance of mRNA in basic research and human health and disease.
The poster is divided into three sections:Howard Hughes Medical Institute, Department of Biochemistry and Biophysics, University of Pennsylvania School of Medicine, Philadelphia, PA 19104.
The recent observation that most RNA polymerase II nascent transcripts in animal cells fail to produce full gene-length transcripts when U1 snRNP (U1) is inactivated revealed a new dimension in gene expression regulation. This essential U1 protective activity, which allows transcription to go farther (telescripting), suppresses premature cleavage and polyadenylation (PCPA) from cryptic signals (PASs) in introns. It is separate from U1's splicing function; both depend on U1's base pairing through its 5' sequence, but 5' modified U1 variants that cannot function in splicing are still effective in telescripting. Our studies have further shown that partial U1 inactivation with an antisense oligonucleotide to its 5'-end dose dependently shortens mRNAs. Widespread mRNA shortening of 3' untranslated regions (3'UTRs) occurs in cancer, proliferating cells and activated immune cells and neurons. 3'UTRs have many mRNA regulatory elements, including miRNAs targets that generally repress translation, and their removal contributes to oncogenicity. We have investigated the potential role of U1 telescripting in these phenomena. We found that moderate U1 decrease shortened 3'UTRs of hundreds of genes in HeLa cancer cells while its over-expression lengthened 3'UTRs of >2,000 genes that are already shortened in these cells. Concomitantly, U1 increase dose-dependently attenuated protein synthesis, cell migration and invasiveness, while its decrease drastically enhanced these cancer cell characteristics. U1-dependent 3'UTR length changes recapitulated cancer-causing miRNA target deregulation in numerous oncogenes. The strong correlation between U1's bi-directional effects on nascent transcripts' length and cell phenotype suggests that mRNA shortening in cancer can be explained by a deficit in U1 telescripting. Our findings establish telescripting's major role in gene regulation. Notably, the ability of U1 modulation alone to profoundly modify cancer aggressiveness suggests telescripting as a potential target for moderating tumorigenesis.
1Department of Biochemistry, Purdue University, West Lafayette, IN
2Purdue University Center for Cancer Research, Purdue University, Hansen Life Sciences Research Building, Rm 141, 201 S. University St., West Lafayette, Indiana 47907
3Next Generation Sequencing Core Facility, Feinberg School of Medicine, 750 Lake Shore Drive, Rubloff 11-144, Northwestern University, Chicago, IL
Dbp2 is a member of the DEAD-box protein family in S. cerevisiae with characterized ATPase and helicase activity in vitro. DEAD-box RNA helicases are a class of enzymes that utilize ATP hydrolysis to remodel RNA and/or RNA-protein (RNP) composition. Dbp2 has been proposed to utilize its helicase activity in vivo to promote RNA-protein complex assembly of both messenger (m)RNAs and long non-coding (lnc)RNAs. Previous work from our laboratory demonstrated that loss of DBP2 enhances the lncRNA-dependent transcriptional induction of the GAL genes by abolishing glucose-dependent repression. Herein, we report that either a rapid carbon source switch or glucose deprivation results in rapid export of Dbp2 to the cytoplasm. Genome-wide RNA sequencing identified a new class of antisense hexose transporter transcripts that are specifically upregulated upon loss of DBP2. Further investigation revealed that both sense and antisense HXT transcripts are aberrantly expressed in DBP2-deficient cells and that this expression pathway can be partially mimicked in wild type cells by glucose depletion. We also find that Dbp2 promotes ribosome biogenesis and represses alternative ATP-producing pathways, as loss of DBP2 alters expression of ribosome biosynthesis (snoRNAs and associated proteins) and respiration genes. This suggests that Dbp2 is a key integrator of nutritional status and transcriptional programs required for energy homeostasis. Because Dbp2 is a characterized RNA helicase, we propose that this metabolic control occurs through modulation of RNA or RNP structure of cell pathway-specific transcripts.
Guosheng Qu, Xiaolong Dong, Carol Lyn Piazza, Venkata R. Chalamcharla, Sheila Lutz, M. Joan Curcio, and Marlene Belfort
Group II introns are commonly believed to be the progenitors of spliceosomal introns, but they are notably absent from nuclear genomes. Barriers to group II intron function in nuclear genomes therefore beg examination. A previous study showed that nuclear expression of a group II intron in yeast results in nonsense-mediated decay and translational repression of mRNA, and that these roadblocks to expression are group II intron-specific. To determine the molecular basis for repression of gene expression, we investigated cellular dynamics of processed group II intron RNAs, from transcription to cellular localization. We showed that pre-mRNA mislocalize to the cytoplasm, where the RNAs are targeted to foci. Furthermore, tenacious mRNA-pre-mRNA interactions, based on intron-exon binding sequences, result in reduced abundance of spliced mRNAs. Nuclear retention of pre-mRNA pre- vents this interaction and relieves these expression blocks. In addition to providing a mechanistic rationale for group II intron- specific repression, our data support the hypothesis that RNA silencing of the host gene contributed to expulsion of group II introns from nuclear genomes and drove the evolution of spliceosomal introns.
Department of Biological Sciences, State University of New York at Albany
Neurons use post-transcriptional control of gene expression to coordinate the supply of critical cytoskeletal proteins, such as the medium neurofilament (Nefm), with axonal growth dynamics in response to extracellular cues encountered by the growing axons. Thus, to have a thorough understanding of how such post-transcriptional control is regulated through the interactions of RNA binding proteins with their targeted cis-regulatory elements within the nefm pre- and mature mRNA, one needs to study these questions within the context of the intact, developing nervous system. We have established a model system for rapidly evaluating the expression of reporter genes bearing mutated cis-regulatory domains in vivo through injection of modified plasmid DNA into early stage Xenopus embryos. In using this method to study cis-regulatory elements of the nefm gene's 3' untranslated region (3'UTR), we discovered that splicing of a specific intron was required for robust transgene protein expression, regardless of promoter or cell type. Transgenes utilizing the nefm 3'UTR but substituting other introns expressed little or no protein, indicating that the required cis-elements were specific to this intron. Surprisingly, all constructs bearing different introns yielded comparable levels of fully spliced, virtually identical mature message. This finding demonstrated that although splicing was required, it alone was insufficient for robust reporter protein expression, and further supported the conclusion that poor protein expression resulted from defects in mRNA translation as opposed to transcription or splicing per se. Furthermore, results from co-immunoprecipitation experiments with heterogeneous nuclear ribonucleoprotein K, which is required for nefm translation, indicated that the nefm intron promoted this RNA binding protein's association with its mature target mRNA. In summary, we have utilized a simple in vivo system to demonstrate in an intact, developing vertebrate nervous system that splicing of a specific intron, rather than splicing in general, was required for translational regulation by an RNA binding protein associated with the final spliced message
1Department of Biochemistry and 2Molecular and Cellular Biology Program, University of Iowa, Iowa City, IA
3Institute for Genome Sciences and Policy, Duke University, Durham, NC 27708
4The Genome Institute, Washington University in St. Louis, St. Louis, MO 63108
The Myc-Max heterodimer is a transcription factor that regulates expression of a large number of genes. Genome occupancy of Myc-Max is thought to be driven by Enhancer box (E-box) DNA elements (CACGTG or variants) to which the heterodimer binds in vitro in the context of accessible chromatin. By analyzing ChIP-Seq datasets, we demonstrated that the positions occupied by Myc-Max across the human genome correlate with the RNA polymerase II (Pol II) transcription machinery better than with E-boxes. Metagene analyses showed that in promoter regions, Myc was uniformly positioned about 100 bp upstream of essentially all promoter proximal paused polymerases with Max about 15 bp upstream of Myc. We re-evaluated the DNA binding properties of full length Myc-Max proteins using electrophoretic mobility shift assays (EMSA) and universal protein-binding microarrays (PBM). EMSA results demonstrated Myc-Max heterodimers display significant sequence preference, but have high affinity for any DNA. Quantification of the relative affinities of Myc-Max for all possible 8-mers using PBM assays showed that sequences surrounding core 6-mers significantly affect binding. Compared to the surrounding core 6-mers significantly affect binding. Compared to the in vitro sequence preferences, Myc-Max genomic occupancy measured by though not completely, independent of sequence specificity. Our results indicate that the genomic occupancy of Myc cannot be explained solely by its intrinsic DNA specificity and suggest that the transcription machinery and associated promoter accessibility play a predominant role in Myc recruitment.
