Regulatory RNAs Study Document

R RNA 13.1 Overview of Regulatory RNAS RNA is ideally suited to regulate through base-pairing interactions Base-pairing interactions between a regulatory RNA and the RNA molecule it is targeting can directly obstruct function (by blocking interactions with other partners), or alter the target RNA structure (and thus potential interactions and functions), or recruit other factors that, in turn, mediate downstream events The base-pairing RNAs can be encoded on the strand of DNA that is antisense to the one encoding their target RNA; these antisense RNAs typically have extensive complementarity with their target Regulatory RNAs can also be encoded at a site that is entirely separate (in trans) from the genes encoding the target RNA or RNAs with which they base-pair, in which case they typically have more limited complementarity with their target or targets Flexibility in RNA structure allows for regulation via metabolite or protein binding Given the inherent flexibility of RNA molecules, which enables them to adopt diverse structures in the cell, it is not surprising that RNAs also can act as regulators by providing binding sites for metabolites or proteins RNA regulators often act in cis, meaning that they are directly connected to the transcript that they regulate Structured RNAs can also act in trans whereby they provide binding sites for proteins and influence their activities In addition, RNA can act as a regulator by effectively serving As a scaffold to bring two or more proteins into proximity with one another or with DNA Regulatory RNAs in bacteria and eukaryotes have similar biological effects but achieve them in different ways Regulatory RNAs in all organisms can impact the transcription, stability, or translation of the target RNA However, the way in which the regulatory RNAs are processed and the mechanisms and machinery through which they manifest their effects are different in organisms from the different kingdoms of life 13.2 Bacterial Base-pairing sRNAs Trans-encoded sRNAs can regulate ribosome binding or target mRNAs for degradation Trans-encoded base-pairing sRNAs in bacteria are commonly expressed in response to specific environmental conditions such as limiting iron or a shift from aerobic to anaerobic conditions The base-paired RNA duplex prevents the ribosome from binding, resulting in the decreased synthesis of the protein encoded by the mRNA. In selected circumstances, the sRNAs can have the opposite effect, actually facilitating translation of target mRNAs In these cases, the sRNA may base-pair with the 5' UTR of an mRNA and thereby prevent formation of RNA secondary structure that would otherwise block ribosome binding and inhibit translation Bacterial sRNAs can also stimulate the degradation of the target mRNAs by recruiting ribonucleases. In some cases, the sRNA itself is degraded together with the mRNA Expression of RyhB is induced when cells encounter low iron levels. Under these conditions, RyhB represses the synthesis of non-essential iron-containing enzymes, allowing the limited iron to be more effectively utilized by essential iron-requiring enzymes For many mRNA targets, ryhb base-pairs at or rear the translational start codon of the mRNA, thereby blocking ribosome binding On another target, ryhb base-pairs with part of the structured 5’ UTR of the mRNA to promote ribosome binding and increase synthesis of a gene product that is important for the production of siderophore molecules that facilitate iron uptake Trans-encoded sRNAs in many bacteria require the chaperone protein Hfq for base-pairing The base complementarity between the trans-encoded sRNAs and their mRNA targets is usually limited to a length of 10 - 20 base pairs The region of the sRNA involved in base-pairing with the target RNA is generally single-stranded and more conserved than the rest of the molecule Given the limited base-pairing interactions and the need for the SRNA to locate its specific target mRNAs among thousands of other transcripts, many bacterial sRNAs require the assistance of the RNA chaperone protein HFQ The sRNAs and target mRNAs appear to bind to different surfaces of the HFQ, and so it is thought that HFQ functions to bring two regions involved in base-pairing into proximity with each other The action of trans-encoded sRNAs can be modulated in multiple ways Sometimes, the levels of HFq or the sRNAs in the cell can be limiting relative to their mRNA targets. in these situations, competition for HFQ binding can have a direct impact on regulation In another twist on this theme, an mRNA can act as a "sponge" or "decoy" to sequester a given sRNA, in turn preventing this sRNA from acting on another target In the example shown, the basepairing of an sRNA with an abundant mRNA, whose transcription is induced in the presence of high levels of chitosugars (oligosaccharides derived from the ubiquitous chitin polymer), can lead to the degradation of the sRNA, and thus its removal from the active pool of sRNAs so that it will no longer block translation of other mRNAs, including one encoding a porin for the uptake of chitosugars Antisense sRNAs repress the synthesis of potentially toxic proteins by modulating translation and mRNA stability Fewer simple antisense sRNAs have been studied in bacteria, but one theme that has emerged is that these RNAs tightly regulate potentially detrimental proteins such as transposases and toxins Although sense and antisense RNAs by definition share extensive complementarity, base-pairing between the RNAs typically is initiated through limited contact between two single-stranded regions in what has been termed a "kissing" complex The region of pairing can subsequently expand, ultimately blocking translation or serving as a substrate for ribonucleases such that double-stranded specific endonuclease RNase III 13.3 Eukaryotic sRNAs: miRNAs, siRNAs, and piRNAs The sRNAs in eukaryotes can be grouped into broad classes MicroRNAs (miRNAs) are specific short RNA sequences which are derived from a primary transcript that is generated in the nucleus from a specific gene encoded in the genome; these RNAs generally down-regulate cytoplasmic mRNAs through translational repression and mRNA decay and are integral to many regulatory networks Small interfering RNAs (siRNAs) are short RNA sequences derived from longer duplex RNAs found in the cytoplasm, including viral RNAs or various endogenous duplex RNAs There are multiple endogenous sources of duplex RNA, including RNA transcripts that have fold-back structures, two separate RNA molecules that are transcribed convergently, or antisense RNAs synthesized by RNA dependent RNA polymerases from RNA templates Sometimes denoted repeat-associated small interfering RNAs (rasiRNAs), these RNAs down-regulate transcription from diverse repetitive regions of the genome and fall into multiple subclasses 13.4 processing of eukaryotic sRNAs RNase III family enzymes generate miRNAs and siRNAs by endonucleolytic cleavage Broadly speaking, the RNase III family of endonucleases use a dimeric catalytic domain to promote the cleavage of double-stranded RNA at two sites In contrast, drosha and dicer, the two critical enzymes involved in miRNA and siRNA processing, both possess two distinct RNase III domains and so function as monomers A key difference between these proteins is that dicer contains a "helicase" domain while drosha does not; this domain is thought to be important in the protective "dicing" reaction that is typical of the dicer protein class Both RNase III enzymes appear to secure the 3' end of the RNA in a protein domain called the PAZ (or PAZ-like) domain and have the catalytic domain located a defined distance away Each of these proteins also relies on at least one accessory factor - DGCR8 for Drosha and TRBP for Dicer - which act in different ways; these RNA binding proteins appear to increase the accuracy and/or efficiency of processing of the precursor species miRNAs are encoded in the genome and are commonly processed sequentially by two distinct RNase III enzymes Like other RNA polymerase II products, the pri-miRNA transcripts are capped and polyadenylated and often contain introns, but do not necessarily encode A protein product. in general terms, the pri-miRNA transcript folds into a stem-loop structure (with an internal region of unpaired nucleotides) surrounded by 5' and 3' single-stranded extensions Drosha interacts with a specialized RNA-binding protein, DGCR8 (in humans), to form what is termed the microprocessor complex. This complex performs a cleavage reaction that removes the 5' and 31 extensions, liberating a 60 - to 70nucleotide transcript known as a pre-miRNA that maintains the form of a hairpin Importantly, like other RNase III enzymes, the RNAs generated by this processing carry monophosphates at their 5’ termini and dinucleotide overhangs with hydroxyl groups at their 3’ termini The hairpin structure generated by the microprocessor complex is subsequently recognized by a nuclear export factor (exportin 5) that delivers the pre-miRNA to the cytoplasm There, a second endonucleolytic cleavage reaction, generally referred to as "dicing", is catalyzed by dicer, another RNase III enzyme, in complex with a doublestranded RNA-binding protein The product of this single diver cleavage reaction (miR:miR*), again carries 5’ monophosphates and 3' overhangs of two nucleotides as characteristic of the cleavage products of RNase III enzymes The nucleotides located between positions 2 and 8 of the miR product are often referred to as the "seed" region and are typically most critical in identifying the target mRNA The processing of some miRNAs involves variations such as methylation While the pathway pictured above is thought to be the most common for miRNA processing in animals, there are less typical pars through which miRNA processing can occur, including pathways in which ribonucleaseS other than drosha carry out the initial cleavages It is interesting to note that, in addition to the core cleavage steps described here, post-transcriptional modification of sRNAs can take place during the course of the RNA processing reaction The plant miRNAs are special among the miRNAs in that they typically target mRNAs with perfect complementarity, much like the siRNAs and piRNAs It is speculated that the function of such post-transcriptional modifications is to protect the 3' ends of the sRNAs from trimming (by exonucleases) and tailing (by polymerases), thus preserving these critical ends siRNAs are processed from double-stranded RNAs from a range of sources Instead, siRNAs are derived from double-stranded RNA that comes from several different sources; these sources may be endogenous or exogenous Endogenous duplex RNA can arise from the normal transcription of genomic loci that have extensive hairpin structures, or from the annealing of sense and antisense RNAs that have both been transcribed from a given locus Exogenous sources of double-stranded RNA include viral RNAs and a variety of duplex structures that have been synthetically introduced into cells by scientists for experimental purposes siRNA-mediated RNA interference is very clearly an important cellular process that has evolved to perform a variety of functions (including cellular defense against viral invasion), and this endogenous mechanism has been widely exploited to inactivate genes in genetic experiments Instead, the dicer enzyme iteratively "dices" the long duplex structures, sequentially cleaving along the duplex approximately every 20-25 base pairs. The cutting frame is set by interactions between the double-stranded RNA and the PAZ region of the dicer protein, and the position of the active sites of the RNase III enzyme Secondary siRNAs are generated by the action of rNA-dependent RNA polymerases In worms and plants, RNA-dependent RNA polymerases further contribute to the generation of siRNAs, typically by de novo synthesis of relatively short complementary copies of single-stranded precursor RNAs transcribed normally from DNA The WAGO proteins thus mediate the functions of secondary siRNAs, which include establishing centromere function, silencing some germline mRNAs, and serving as a defense against transposons and RNA viruses Some sRNAs are produced from longer precursor RNAs that are transcribed from repetitive regions of the genome The processed RNAs are typically somewhat longer than the miRNAs and siRNAs and broadly share the property of being transcribed from repetitive sequences in genomes, which include retrotransposons, DNA transposons, and microsatellite DNA sequences The most general view is that these sRNAs interfere with gene expression from the repetitive sequences, in some cases by targeting nascent mRNA transcripts for degradation by the Argonautes and in other cases by directly modulating chromatin structure, DNA methylation, and thus transcription at these loci sRNAs are derived from repeat sequences in fission yeast are processed in a manner similar to the siRNAs In this system, low level bidirectional transcription of pericentromeric repeat elements yields double-stranded RNA that is cleared by a dicer endonuclease to give duplexes similar to siRNA duplexes The transcription of the pericentromeric repeats during A particular time in the cell cycle is thought to provide the siRNAs that establish and maintain the centromeric heterochromatin Mammalian and fly piRNAs are generated from single-stranded precursors by a "ping-pong" mechanism Many of these regions are enriched in fragmented transposon remnants and are located in the most transposon-rich areas of the genome The biogenesis of piRNAs has been most extensively defined in drosophila where primary piRNAs are derived from singlestranded piRNA precursor RNAs which are antisense to transposon mRNAs Processing of these precursors is initiated by the endonucleolytic definition of the 5’ end of the piRNA by a nuclease known as zucchini. The cleaved RNA is then loaded onto piwi or aubergine, two piwi-clade argonaute proteins. Trimming of the 3' end happens through a process also involving the zucchini protein The primary piRNAs direct the cleavage of the transposon mRNAs and thereby lead to the production of additional piRNAs that correspond to the sense strand of the transposon mRNAs Both classes of piRNA (those derived from the sense and those derived from the antisense strand)are loaded onto several piwi family proteins by mechanisms that are poorly understood The iterative process of generating piRNAs is termed a "ping-pong cycle", where existing piRNAs bound to a piwi family protein define the cleavage position within a target RNA to form a new piRNA (which will subsequently bind to another piwi family protein, and so on) the piRNAs that are produced via this ping-pong reaction typically have A uracil at the 5' end (as do many miRNAs) and an adenine bias at position 10 in the secondary piRNA 13.5 Loading of Argonaute family proteins with eukaryotic sRNAs Regardless of how the eukaryotic sRNAs are generated, all carry out their functions in association with Argonaute family proteins. These proteins bind and orient sRNAs to facilitate their binding to fully or partially complementary target RNAs Argonaute proteins typically possess four core domains that interact with different parts of the regulatory RNA: an n-terminal domain; a Paz domain that binds the 3' end of bound RNA; a mid (for middle) domain that interacts with the 5' end of bound RNA; and an RNase H-related PIWI domain that interacts along the length of the sRNA The Paz, mid, and Piwi domains together make extensive sequence-independent contacts with the backbone of the "seed" region, thus pre-orienting this portion of the guide RNA in a helical conformation that facilitates the scanning of cellular RNAs for complementary sequences miRNA and siRNA sorting on Argonaute proteins involves identification of the guide strand and elimination of the passenger strand The two strands of this duplex RNA are called the guide and passenger strands. Once the duplex species has been generated, it is loaded onto the appropriate argonaute protein. One of the RHA strands (the miR or siR guide strand) is retained, while the other strand (the miR* or siR* passenger strand) is selectively removed This "sorting" problem is complex. It appears that a number of factors help to determine which of the two strands is the guide strand and is thus loaded onto the argonaute protein, and which is the passenger strand that is released and degraded Sorting appears to be affected by a number of factors, including: (1) the dicer protein and its interacting protein partners that are involved in the final RNA cleavage step; (2) the thermodynamic properties of the duplex RNA; and (3) the destination argonaute protein By contrast, there are four argonaute proteins in humans; all are thought to be involved in miRNA-mediated silencing mechanisms under normal circumstances, and there appear to be for specialized sequence preferences for dictating miRNA binding to specific argonaute proteins For imperfect duplexes (miR:miR*, the duplex composed of the intended miRNA (miR) and its nearly complementary sequence (miR*)), it is not certain how the passenger strand is selectively removed, though the action of a helicase may well be helpful Single-stranded piRNAs are loaded onto piwi proteins as they are generated Independent of the route taken, the output of the loading process for an classes of eukaryotic sRNAs is a singlestranded guide RNA bound to an argonaute-type protein - this ribonucleoprotein complex is commonly referred to as RISC, RNA-induced silencing complex 13.6 Gene Silencing by small eukaryotic RNAs Partially and fully complementary pairings to target RNAs induce the argonaute protein to adopt distinct conformations The most important parring region for the miRNA is referred to as the "seed" sequence, which generally encompasses nucleotide positions 2-8 of the miRNA species and defines its principal cellular targets In these cases, extensive base-pairing with downstream portions of the miRNA does not typically occur; the absence of extended duplex formation allows the 3' end of the miRNA to remain securely bound to the Paz domain By contrast, in the presence of fully complementary target RNA, the argonaute-bound guide siRNA base-pairs with its target an along its length to form extended duplex nuclei acid When such extended duplex forms, the 3' end of the guide RNA is released from the Paz domain of the argonaute protein, leading to conformational changes that are thought to activate the cleavage activity of the RNase H PIWI domain of the argonaute protein miRISC typically targets 3' UTRs to promote translational repression and mRNA decay These complexes [miRISC] are generally found in the cytoplasm of the cell where mRNAs are targeted for silencing. Functional miRNA binding sites are predominantly found in the 3' UTRs of the targeted mRNAs miRNAs typically pair imperfectly with their targets and silence them through translational repression and mRNA decay As a consequence of this particular mismatch, the PIWI domain is generally not able to cleave mRNAs targeted by miRNAs. Typically the pairing of the miRNA through the seed region to the short complementary sequence is sufficient for repressing translation of the target Degradation of the target mRNA is thought to occur subsequent to translational repression via normal RNA decay pathways, which involve deadenylation, decapping, and exonucleolytic degradation steps Typically, full complementarity is observed in plants and, as such, cleavage by the argonaute protein is activated and occurs precisely at the position in the target mRNA between nucleotides 10 and 11 of the guide RHA In these organisms, endonucleolytic cleavage, followed by exonucleolytic decay, is the principal mechanism of miRNA-mediated silencing An siRISC targets cytoplasmic RNAs for endonucleolytic cleavage and degradation An siRISC, composed of an argonaute protein and an siRNA, is often found in the cytoplasm where it silences gene expression of complementary targets, although there is increasing evidence that these argonaute complexes have nuclear roles In contrast to miRISC, the RNAs targeted by siRISC tend to be fully complementary to the siRNA, which potently activates the complex to cleave ("slice") the target mRNA, exactly between the target positions complementary to nucleotides 10 and 11 of the siRNA Double-stranded viral RNAs are processed to generate siRNAs, which subsequently bind to Argonaute and then eventually target their own RNAs for silencing 13.7 viral defense by the CRISPR systems in bacteria and archaea CRISPR sequences are associated with genes encoding enzymes capable of cleaving DNA and RNA Soon after the repetitive CRISPR sequences were identified, it became clear that these sequences are found adjacent to conserved sets of CRISPR-associated (or cas) genes, which encode proteins containing domains likely to be useful in different nuclei acid-based transactions Generally speaking, the CRISPR systems carry out two main processes in which sequences from foreign DNA are extracted to be integrated into the CRISPR array as new spacers; and interference, in which the CRISPR locus is expressed, spacers direct the cleavage of foreign nucleic acids, and immunity is conferred For the interference step, ribonucleases are needed to process the RNAs that serve as guides; structural proteins are needed to cradle the RNA guides and assist them in finding target DNA or RNA species, and, again, nucleases are needed to catalyze the degradation of DNA and RNA targets Broadly, there are several main types of CRISPR systems with different sets of associated protein factors, and among them, mechanistic similarities and distinctions in how the key processes are carried out Acquisition of CRISPR spacer sequences requires DNA integration In general, new invader-derived fragments are incorporated one at a time, in a polarized manner relative to the leader-adjacent end of the CRISPR array. Thus the order of spacers reflects the history of invasion of the organism Cas 1 and Cas 2, which possess endonuclease domains and bind to DNA, are central to spacer acquisition, but they often function in concert with cellular enzymes, such as the DNA recombination enzymes ReeBCD, that provide substrates for the integration, and other proteins such as the DNA bending protein IHF CRISPR-based interference involves an RNA intermediate This transcript [CRISPR RNA] is then processed by cleavage within repeat-derived sequences into shorter fragments termed crRNAs, each containing a spacer sequence that directs the targeting of foreign nucleic acids In the type I system, adjacent palindromic repeat sequences base-pair, thus providing a double-stranded substrate for the Cas6e endonuclease, which cleaves within each repeat sequence to generate individual crRNA species In the type II system, which has been extensively studied due to its relative simplicity, a precursor RNA transcript pairs with a trans-acting "tracer RNA" through the constant region in the repeat, and an endogenous bacterial RNase III enzyme (assisted by Cas9) is responsible for processing the RNA into crRNA fragments Independent of how they are generated, the resulting short spacer-containing RNA fragments bind to a set of Cas proteins - we broadly refer to these complexes as the crRNA:Cas complexes Strikingly, this structure reveals how, in type I systems, every sixth nucleotide is effectively sequestered into the cascade complex to allow the crRNA "seed" sequence to assume an extended conformation that can effectively Scan nucleic acids for complementary sequences For the type II system, the crRNA binds to Cas9, which is composed of two distinct lobes, one of which is principally involved in binding to the crRNA, while the other is critical for DNA recognition and for subsequent cleavage of the target Once formed, all crRNA:Cas complexes (with the exception of type III and type VI systems which find complementary sequences in RNA) share the similar task of needing to identify complementary DNA sequences, initially through a critical interaction with the 7- to 8-nucleotide long "seed" sequence and to then extend the pairing interaction The structure that forms when one strand of DNA is displaced by complementary pairing between the crRNA and the other strand of the DNA is referred to as an r loop; it is thought to trigger conformational changes in the crRNA:Cas complexes that activate (or recruit) the nucleases for cleavage 13.8 RNA - mediated regulation in cis Cis-encoded RNA elements regulate transcription, translation, or splicing of mRNAs Sequences at the 5’ end of bacterial mRNAs can impact transcription elongation by a phenomenon known as attenuation, in which short self-complementary regions in the RNA base-pair with one another and thereby impact transcription Translation of a small open reading frame rich in codons for a specific amino acid in the 5' UTR of a gene encoding enzymes required for the synthesis of the amino acid affected the downstream structure of the mRNA, and thus transcription termination In perhaps the simplest case, a segment of RNA functions like a "thermometer", sensing temperature and reporting the results in an interpretable fashion In the pathogenic bacteria in which they are commonly found, RNA thermometers make it possible to turn on the synthesis of virulence regulators and chaperone proteins when the bacteria are exposed to elevated temperatures in the host during infection In another permutation, uncharged tRNA molecules bind specifically to the 5' untranslated sequence of an mRNA sequence that encodes their cognate aminoacyl tRNA synthetase In addition to these specific examples - in which RNA regulates gene expression on its own - there is an increasing appreciation for the wide range of regulatory proteins that bird to mRNAs, both upstream and downstream of the ORF, to modulate all steps in the lifetime of an mRNA Riboswitch RNA elements respond to a wide range of small molecules One particularly well-characterized class of cis -encoded RNA regulatory element folds into exquisite secondary structures that bind to small molecules and can regulate downstream events, including transcription,translation, and even RNA splicing In general, the small molecule (that is, metabolite) binds to the aptamer domain and induces changes in the structure of the effector domain, which brings about changes in gene expression The effector domain can encompass a transcription terminator sequence whose activity can either be enhanced or diminished when a ligand binds to the aptamer domain Alternatively, the effector domain can effectively sequester or release the ribosome-binding site, thus either blocking or facilitating translation initiation In a typical scenario, the association of a particular metabolite with a riboswitch regulates (either negatively or positively) the synthesis of a gene product encoded by the downstream ORF that is involved in synthesizing,metabolizing, or transporting the very same metabolite In general terms, riboswitch elements illustrate the amazing versatility of RNA to adopt complex, functionally diverse three-dimensional structures that rival those formed by proteins 13.9 Protein-binding regulatory RNAs Protein-binding regulatory RNAs can Titrate a protein away from its target Given the hundreds of RNA-binding proteins and the ability of RNA to told into a wide range of structures, it is not surprising that some regulatory RNAs act by sequestering proteins away from their usual target sites When CsrB is not induced, CsrA binds to the 5' untranslated regions of target mRNAs and modulates their stability or translation However, when CsrB levels increase during specific stages of growth, the CsrA protein is titrated away from these target by the many binding sites present in CsrB RNA Not surprisingly, the 6S RNA effectively sequesters one form of RNA polymerase that is expressed during stationary phase growth, thus limiting overall transcription by this RNA polymerase at this stage Intriguingly, this mimic can even be subsequently used AS a substrate for RNA polymerase; during exponential growth - when 6S-mediated inhibition is no longer advantageous - the regulatory RNA transcript is extended by several nucleotides by the bound RNA polymerase