Goldilocks and RNA: where Mg2+ concentration is just right.

Magnesium, the most abundant divalent cation in cells, catalyzes RNA cleavage but also promotes RNA folding. Because folding can protect RNA from cleavage, we predicted a 'Goldilocks landscape', with local maximum in RNA lifetime at Mg2+ concentrations required for folding. Here, we use simulation and experiment to discover an innate and sophisticated mechanism of control of RNA lifetime. By simulation we characterized RNA Goldilocks landscapes and their dependence on cleavage and folding parameters. Experiments with yeast tRNAPhe and the Tetrahymena ribozyme P4-P6 domain show that structured RNAs can inhabit Goldilocks peaks. The Goldilocks peaks are tunable by differences in folded and unfolded cleavage rate constants, Mg2+ binding cooperativity, and Mg2+ affinity. Different folding and cleavage parameters produce Goldilocks landscapes with a variety of features. Goldilocks behavior allows ultrafine control of RNA chemical lifetime, whereas non-folding RNAs do not display Goldilocks peaks of protection. In sum, the effects of Mg2+ on RNA persistence are expected to be pleomorphic, both protecting and degrading RNA. In evolutionary context, Goldilocks behavior may have been a selectable trait of RNA in an early Earth environment containing Mg2+ and other metals.

G-Quadruplexes in Human Ribosomal RNA.

rRNA is the single most abundant polymer in most cells. Mammalian rRNAs are nearly twice as large as those of prokaryotes. Differences in rRNA size are due to expansion segments, which contain extended tentacles in metazoans. Here we show that the terminus of an rRNA tentacle of Homo sapiens contains 10 tandem G-tracts that form highly stable G-quadruplexes in vitro. We characterized rRNA of the H. sapiens large ribosomal subunit by computation, circular dichroism, UV melting, fluorescent probes, nuclease accessibility, electrophoretic mobility shifts, and blotting. We investigated Expansion Segment 7 (ES7), oligomers derived from ES7, intact 28S rRNA, 80S ribosomes, and polysomes. We used mass spectrometry to identify proteins that bind to rRNA G-quadruplexes in cell lysates. These proteins include helicases (DDX3, CNBP, DDX21, DDX17) and heterogeneous nuclear ribonucleoproteins. Finally, by multiple sequence alignments, we observe that G-quadruplex-forming sequences are a general feature of LSU rRNA of Chordata but not, as far as we can tell, of other species. Chordata ribosomes present polymorphic tentacles with the potential to switch between inter- and intramolecular G-quadruplexes. To our knowledge, G-quadruplexes have not been reported previously in ribosomes.

Relationship between calorimetric profiles and differential melting curves for natural DNAs.

Many experiments demonstrate that regions with higher GC-content in natural DNAs unwind at higher temperatures adsorbing more heat than equivalently sized regions with lower GC-content. This simple observation implies that normalized calorimetric melting profiles (calorimetric cDMCs) will not be equivalent differential melting curves (DMCs). We propose simple expressions for long natural and random DNA sequences to reciprocally convert DMCs and corresponding calorimetric cDMCs. The expressions are confirmed by the Poland-Fixman-Freire method and an approach based upon mixtures of homopolymeric duplexes. Using these expressions and experimental calorimetric data, we demonstrate that the average relative deviation between DMC and cDMC is proportional to the temperature melting range of the helix-coil transition ΔT. Corresponding difference between melting temperatures is proportional to ΔT2. In general, sequence and ionic conditions influence the deviation through their effect on ΔT. On the basis of the developed approach, we propose a method to determine the thermodynamic melting temperature (ratio of calorimetric enthalpy and entropy of the helix-coil transition) for natural DNAs from optical DMCs without calorimetric experiments.

Yeast rRNA Expansion Segments: Folding and Function.

Divergence between prokaryotic and eukaryotic ribosomal RNA (rRNA) and among eukaryotic ribosomal RNAs is focused in expansion segments (ESs). Eukaryotic ribosomes are significantly larger than prokaryotic ribosomes partly because of their ESs. We hypothesize that larger rRNAs of complex organisms could confer increased functionality to the ribosome. Here, we characterize the binding partners of Saccharomyces cerevisiae expansion segment 7 (ES7), which is the largest and most variable ES of the eukaryotic large ribosomal subunit and is located at the surface of the ribosome. In vitro RNA-protein pull-down experiments using ES7 as a bait indicate that ES7 is a binding hub for a variety of non-ribosomal proteins essential to ribosomal function in eukaryotes. ES7-associated proteins observed here cluster into four groups based on biological process, (i) response to abiotic stimulus (e.g., response to external changes in temperature, pH, oxygen level, etc.), (ii) ribosomal large subunit biogenesis, (iii) protein transport and localization, and (iv) transcription elongation. Seven synthetases, Ala-, Arg-, Asp-, Asn-, Leu-, Lys- and TyrRS, appear to associate with ES7. Affinities of AspRS, TyrRS and LysRS for ES7 were confirmed by in vitro binding assays. The results suggest that ES7 in S. cerevisiae could play a role analogous to the multi-synthetase complex present in higher order organisms and could be important for the appropriate function of the ribosome. Thermal denaturation studies and footprinting experiments confirm that isolated ES7 is stable and maintains a near-native secondary and tertiary structure.

Ligation of RNA Oligomers by the Schistosoma mansoni Hammerhead Ribozyme in Frozen Solution.

The interstitial liquid phase within frozen aqueous solutions is an environment that minimizes RNA degradation and facilitates reactions that may have relevance to the RNA World hypothesis. Previous work has shown that frozen solutions support condensation of activated nucleotides into RNA oligomers, RNA ligation by the hairpin ribozyme, and RNA synthesis by a RNA polymerase ribozyme. In the current study, we examined the activity of a hammerhead ribozyme (HHR) in frozen solution. The Schistosoma mansoni hammerhead ribozyme, which predominantly cleaves RNA, can ligate its cleaved products (P1 and P2) with yields up to ~23 % in single turnover experiments at 25 °C in the presence of Mg(2+). Our studies show that this HHR ligates RNA oligomers in frozen solution in the absence of divalent cations. Citrate and other anions that exhibit strong ion-water affinity enhanced ligation. Yields up to 43 % were observed in one freeze-thaw cycle and a maximum of 60 % was obtained after several freeze-thaw cycles using wild-type P1 and P2. Truncated and mutated P1 substrates were ligated to P2 with yields of 14-24 % in one freeze-thaw cycle. A pool of P2 substrates with mixtures of all four bases at five positions were ligated with P1 in frozen solution. High-throughput sequencing indicated that 70 of the 1024 possible P2 sequences were represented in ligated products at 1000 or more read counts per million reads. The results indicate that the HHR can ligate a range of short RNA oligomers into an ensemble of diverse sequences in ice.

History of the ribosome and the origin of translation.

We present a molecular-level model for the origin and evolution of the translation system, using a 3D comparative method. In this model, the ribosome evolved by accretion, recursively adding expansion segments, iteratively growing, subsuming, and freezing the rRNA. Functions of expansion segments in the ancestral ribosome are assigned by correspondence with their functions in the extant ribosome. The model explains the evolution of the large ribosomal subunit, the small ribosomal subunit, tRNA, and mRNA. Prokaryotic ribosomes evolved in six phases, sequentially acquiring capabilities for RNA folding, catalysis, subunit association, correlated evolution, decoding, energy-driven translocation, and surface proteinization. Two additional phases exclusive to eukaryotes led to tentacle-like rRNA expansions. In this model, ribosomal proteinization was a driving force for the broad adoption of proteins in other biological processes. The exit tunnel was clearly a central theme of all phases of ribosomal evolution and was continuously extended and rigidified. In the primitive noncoding ribosome, proto-mRNA and the small ribosomal subunit acted as cofactors, positioning the activated ends of tRNAs within the peptidyl transferase center. This association linked the evolution of the large and small ribosomal subunits, proto-mRNA, and tRNA.

Evolution of the ribosome at atomic resolution.

The origins and evolution of the ribosome, 3-4 billion years ago, remain imprinted in the biochemistry of extant life and in the structure of the ribosome. Processes of ribosomal RNA (rRNA) expansion can be "observed" by comparing 3D rRNA structures of bacteria (small), yeast (medium), and metazoans (large). rRNA size correlates well with species complexity. Differences in ribosomes across species reveal that rRNA expansion segments have been added to rRNAs without perturbing the preexisting core. Here we show that rRNA growth occurs by a limited number of processes that include inserting a branch helix onto a preexisting trunk helix and elongation of a helix. rRNA expansions can leave distinctive atomic resolution fingerprints, which we call "insertion fingerprints." Observation of insertion fingerprints in the ribosomal common core allows identification of probable ancestral expansion segments. Conceptually reversing these expansions allows extrapolation backward in time to generate models of primordial ribosomes. The approach presented here provides insight to the structure of pre-last universal common ancestor rRNAs and the subsequent expansions that shaped the peptidyl transferase center and the conserved core. We infer distinct phases of ribosomal evolution through which ribosomal particles evolve, acquiring coding and translocation, and extending and elaborating the exit tunnel.

Secondary structures of rRNAs from all three domains of life.

Accurate secondary structures are important for understanding ribosomes, which are extremely large and highly complex. Using 3D structures of ribosomes as input, we have revised and corrected traditional secondary (2°) structures of rRNAs. We identify helices by specific geometric and molecular interaction criteria, not by co-variation. The structural approach allows us to incorporate non-canonical base pairs on parity with Watson-Crick base pairs. The resulting rRNA 2° structures are up-to-date and consistent with three-dimensional structures, and are information-rich. These 2° structures are relatively simple to understand and are amenable to reproduction and modification by end-users. The 2° structures made available here broadly sample the phylogenetic tree and are mapped with a variety of data related to molecular interactions and geometry, phylogeny and evolution. We have generated 2° structures for both large subunit (LSU) 23S/28S and small subunit (SSU) 16S/18S rRNAs of Escherichia coli, Thermus thermophilus, Haloarcula marismortui (LSU rRNA only), Saccharomyces cerevisiae, Drosophila melanogaster, and Homo sapiens. We provide high-resolution editable versions of the 2° structures in several file formats. For the SSU rRNA, the 2° structures use an intuitive representation of the central pseudoknot where base triples are presented as pairs of base pairs. Both LSU and SSU secondary maps are available (http://apollo.chemistry.gatech.edu/RibosomeGallery). Mapping of data onto 2° structures was performed on the RiboVision server (http://apollo.chemistry.gatech.edu/RiboVision).

Secondary structure and domain architecture of the 23S and 5S rRNAs.

We present a de novo re-determination of the secondary (2°) structure and domain architecture of the 23S and 5S rRNAs, using 3D structures, determined by X-ray diffraction, as input. In the traditional 2° structure, the center of the 23S rRNA is an extended single strand, which in 3D is seen to be compact and double helical. Accurately assigning nucleotides to helices compels a revision of the 23S rRNA 2° structure. Unlike the traditional 2° structure, the revised 2° structure of the 23S rRNA shows architectural similarity with the 16S rRNA. The revised 2° structure also reveals a clear relationship with the 3D structure and is generalizable to rRNAs of other species from all three domains of life. The 2° structure revision required us to reconsider the domain architecture. We partitioned the 23S rRNA into domains through analysis of molecular interactions, calculations of 2D folding propensities and compactness. The best domain model for the 23S rRNA contains seven domains, not six as previously ascribed. Domain 0 forms the core of the 23S rRNA, to which the other six domains are rooted. Editable 2° structures mapped with various data are provided (http://apollo.chemistry.gatech.edu/RibosomeGallery).

RNA with iron(II) as a cofactor catalyses electron transfer.

Mg(2+) is essential for RNA folding and catalysis. However, for the first 1.5 billion years of life on Earth RNA inhabited an anoxic Earth with abundant and benign Fe(2+). We hypothesize that Fe(2+) was an RNA cofactor when iron was abundant, and was substantially replaced by Mg(2+) during a period known as the 'great oxidation', brought on by photosynthesis. Here, we demonstrate that reversing this putative metal substitution in an anoxic environment, by removing Mg(2+) and replacing it with Fe(2+), expands the catalytic repertoire of RNA. Fe(2+) can confer on some RNAs a previously uncharacterized ability to catalyse single-electron transfer. We propose that RNA function, in analogy with protein function, can be understood fully only in the context of association with a range of possible metals. The catalysis of electron transfer, requisite for metabolic activity, may have been attenuated in RNA by photosynthesis and the rise of O2.

Molecular paleontology: a biochemical model of the ancestral ribosome.

Ancient components of the ribosome, inferred from a consensus of previous work, were constructed in silico, in vitro and in vivo. The resulting model of the ancestral ribosome presented here incorporates ∼20% of the extant 23S rRNA and fragments of five ribosomal proteins. We test hypotheses that ancestral rRNA can: (i) assume canonical 23S rRNA-like secondary structure, (ii) assume canonical tertiary structure and (iii) form native complexes with ribosomal protein fragments. Footprinting experiments support formation of predicted secondary and tertiary structure. Gel shift, spectroscopic and yeast three-hybrid assays show specific interactions between ancestral rRNA and ribosomal protein fragments, independent of other, more recent, components of the ribosome. This robustness suggests that the catalytic core of the ribosome is an ancient construct that has survived billions of years of evolution without major changes in structure. Collectively, the data here support a model in which ancestors of the large and small subunits originated and evolved independently of each other, with autonomous functionalities.

RNA folding and catalysis mediated by iron (II).

Mg²âº shares a distinctive relationship with RNA, playing important and specific roles in the folding and function of essentially all large RNAs. Here we use theory and experiment to evaluate Fe²âº in the absence of free oxygen as a replacement for Mg²âº in RNA folding and catalysis. We describe both quantum mechanical calculations and experiments that suggest that the roles of Mg²âº in RNA folding and function can indeed be served by Fe²âº. The results of quantum mechanical calculations show that the geometry of coordination of Fe²âº by RNA phosphates is similar to that of Mg²âº. Chemical footprinting experiments suggest that the conformation of the Tetrahymena thermophila Group I intron P4-P6 domain RNA is conserved between complexes with Fe²âº or Mg²âº. The catalytic activities of both the L1 ribozyme ligase, obtained previously by in vitro selection in the presence of Mg²âº, and the hammerhead ribozyme are enhanced in the presence of Fe²âº compared to Mg²âº. All chemical footprinting and ribozyme assays in the presence of Fe²âº were performed under anaerobic conditions. The primary motivation of this work is to understand RNA in plausible early earth conditions. Life originated during the early Archean Eon, characterized by a non-oxidative atmosphere and abundant soluble Fe²âº. The combined biochemical and paleogeological data are consistent with a role for Fe²âº in an RNA World. RNA and Fe²âº could, in principle, support an array of RNA structures and catalytic functions more diverse than RNA with Mg²âº alone.

Domain III of the T. thermophilus 23S rRNA folds independently to a near-native state.

The three-dimensional structure of the ribosomal large subunit (LSU) reveals a single morphological element, although the 23S rRNA is contained in six secondary structure domains. Based upon maps of inter- and intra-domain interactions and proposed evolutionary pathways of development, we hypothesize that Domain III is a truly independent structural domain of the LSU. Domain III is primarily stabilized by intra-domain interactions, negligibly perturbed by inter-domain interactions, and is not penetrated by ribosomal proteins or other rRNA. We have probed the structure of Domain III rRNA alone and when contained within the intact 23S rRNA using SHAPE (selective 2'-hydroxyl acylation analyzed by primer extension), in the absence and presence of magnesium. The combined results support the hypothesis that Domain III alone folds to a near-native state with secondary structure, intra-domain tertiary interactions, and inter-domain interactions that are independent of whether or not it is embedded in the intact 23S rRNA or within the LSU. The data presented support previous suggestions that Domain III was added relatively late in ribosomal evolution.

The influence of Escherichia coli Hfq mutations on RNA binding and sRNA•mRNA duplex formation in rpoS riboregulation.

The Escherichia coli RNA binding protein Hfq plays an important role in regulating mRNA translation through its interactions with small non-coding RNAs (sRNAs) and specific mRNAs sites. The rpoS mRNA, which codes for a transcription factor, is regulated by several sRNAs. DsrA and RprA enhance translation by pairing to a site on this mRNA, while OxyS represses rpoS mRNA translation. To better understand how Hfq interacts with these sRNAs and rpoS mRNA, the binding of wt Hfq and eleven mutant Hfqs to DsrA, RprA, OxyS and rpoS mRNA was examined. Nine of the mutant Hfq had single-residue mutations located on the proximal, distal, and outer-edge surfaces of the Hfq hexamer, while two Hfq had truncated C-terminal ends. Hfq with outer-edge mutations and truncated C-terminal ends behaved similar to wt Hfq with regard to binding the sRNAs, rpoS mRNA segments, and stimulating DsrA•rpoS mRNA formation. Proximal surface mutations decreased Hfq binding to the three sRNAs and the rpoS mRNA segment containing the translation initiation region. Distal surface mutations lowered Hfq's affinity for the rpoS mRNA segment containing the (ARN)(4) sequence. Strong Hfq binding to both rpoS mRNA segments appears to be needed for maximum enhancement of DsrA•rpoS mRNA annealing. OxyS bound tightly to Hfq but exhibited weak affinity for rpoS mRNA containing the leader region and 75 nt of coding sequence in the absence or presence of Hfq. This together with other results suggest OxyS represses rpoS mRNA translation by sequestering Hfq rather than binding to rpoS mRNA.

The stoichiometry of the Escherichia coli Hfq protein bound to RNA.

The Escherichia coli RNA binding protein Hfq is involved in many aspects of post-transcriptional gene expression. Tight binding of Hfq to polyadenylate sequences at the 3' end of mRNAs influences exonucleolytic degradation, while Hfq binding to small noncoding RNAs (sRNA) and their targeted mRNAs facilitate their hybridization which in turn effects translation. Hfq binding to an A-rich tract in the 5' leader region of the rpoS mRNA and to the sRNA DsrA have been shown to be important for DsrA enhanced translation initiation of this mRNA. The complexes of Hfq-A(18) and Hfq-DsrA provide models for understanding how Hfq interacts with these two RNA sequence/structure motifs. Different methods have reported different values for the stoichiometry of Hfq-A(18) and Hfq-DsrA. In this work, mass spectrometry and analytical ultracentrifugation provide direct evidence that the strong binding mode of the Hfq hexamer (Hfq(6)) for A(18) and domain II of DsrA (DsrA(DII)) involve 1:1 complexes. This stoichiometry was also supported by fluorescence anisotropy and a competition gel mobility shift experiment using wild-type and truncated Hfq. More limited studies of Hfq binding to DsrA as well as to the sRNAs RprA, OxyS, and an 18-nt segment of OxyS were also consistent with 1:1 stoichiometry. Mass spectrometry of cross-linked samples of Hfq(6), A(18), and DsrA(DII) exhibit intensity corresponding to a ternary 1:1:1 complex; however, the small intensity of this peak and fluorescence anisotropy experiments did not provide evidence that this ternary complex is stable in solution.

E. coli DNA associated with isolated Hfq interacts with Hfq's distal surface and C-terminal domain.

The RNA-binding protein Hfq has been studied extensively for its function as a modulator of gene expression at the post-transcriptional level. While most Hfq studies have focused on the protein's interaction with sRNAs and mRNAs, Hfq binding to DNA has been observed but is less explored. During the isolation of Hfq from Escherichiacoli, we found genomic DNA fragments associated with the protein after multiple steps of purification. Sequences of 41 amplified segments from the DNA fragments associated with Hfq were determined. A large fraction of the DNA segments were predicted to have significant helical axis curvature and were from genes associated with membrane proteins, characteristics unexpected for non-specific binding. Analysis by analytical ultracentrifugation indicated that rA(18) binding to Hfq disrupts Hfq-DNA interactions. The latter observation suggests Hfq binding to DNA involves its distal surface. This was supported by a gel mobility shift assay that showed single amino acid mutations on the distal surface of Hfq inhibited Hfq binding to duplex DNA, while six of seven mutations on the proximal surface and outer circumference of the hexamer did not prevent Hfq binding. Two mutated Hfq which have portions of their C-terminal domain removed also failed to bind to DNA. The apparent K(d) for binding wild type Hfq to several duplex DNA was estimated from a gel mobility shift assay to be ~400nM.

Mechanism of RNA double helix-propagation at atomic resolution.

The conversion of a nucleic acid from single strands to double strands is thought to involve slow nucleation followed by fast double-strand propagation. Here, for RNA double-strand propagation, we propose an atomic resolution reaction mechanism. This mechanism, called the stack-ratchet, is based on data-mining of three-dimensional structures and on available thermodynamic information. The stack-ratchet mechanism extends and adds detail to the classic zipper model proposed by Porschke (Porschke, D. Biophysical Chemistry 1974, 2, pp. 97-101). Porschke's zipper model describes the addition of a base pair to a nucleated helix in terms of a single type of elementary reaction; a concerted process in which the two bases, one from each strand, participate in the transition state. In the stack-ratchet mechanism proposed here a net base-pairing step consists of two elementary reactions. Motions of only one strand are required to achieve a given transition state. One elementary reaction preorganizes and stacks the 3' single-strand, driven by base--base stacking interactions. A second elementary reaction stacks the 5' strand and pairs it with the preorganized 3' strand. In the stack-ratchet mechanism, a variable length 3' stack leads the single-strand/double-strand junction. The stack-ratchet mechanism is not a two-state process. A base can be (i) unstacked and unpaired, (ii) stacked and paired, or (ii) stacked and unpaired (only on the 3' strand). The data suggests that helices of DNA and of RNA do not propagate by similar mechanisms.

DNA denaturation under freezing in alkaline medium.

It is generally accepted that DNA conserves its secondary structure after a freeze-thaw cycle. A negligible amount of degradation occurs after this procedure. Degradation becomes appreciable only after multiple cycles of freezing and thawing. In this study, we have found that a single freeze-thaw cycle in alkaline medium (pH>or=10.8) gives rise to denaturation of calf thymus DNA, although the melting temperature of intact DNA in the solution used for the freeze-thaw experiments is higher than 60 degrees C. The degree of denaturation is almost independent of the regime of freezing. The melting curve obtained after DNA is frozen at -2 degrees C and then thawed is almost the same as after a freezing carried out in liquid nitrogen (-196 degrees C). However, incubation in the same solution at 0 degrees C for 24 hours without freezing does not give rise to any denaturation. The degree of denaturation caused by freezing increases with pH (if pH>or=10.8) and decreases with Na2CO3 concentration at fixed pH and [Na+], although Na2CO3 decreases the melting temperature of intact DNA. A preliminary treatment of DNA with cisplatin or transplatin (0.01 Pt atoms per nucleotide) gives rise to a full recovery of the DNA secondary structure after freezing and thawing similar to what occurs after heating DNA to 100 degrees C and quick cooling. Possible mechanisms that may cause DNA denaturation during a freeze-thaw cycle in alkaline medium are discussed.

Effect of Hfq on RprA-rpoS mRNA pairing: Hfq-RNA binding and the influence of the 5' rpoS mRNA leader region.

The rpoS mRNA encodes a stress response transcription factor in Escherichia coli. It is one of a growing number of mRNAs found to be regulated by small RNAs (sRNA). Translation initiation of rpoS mRNA is enhanced by two sRNAs, DsrA and RprA, that pair to the same site near the rpoS start codon in the presence of the Hfq protein. In this work, we examine the interaction of E. coli Hfq with RprA and two portions of the rpoS mRNA leader region. One rpoS RNA, rpoS-L, contained the entire 565-nucleotide leader region, while the other, rpoS-S, contained the 199-nucleotide sequence surrounding the start codon. An RNase H assay indicated both rpoS RNAs have similar secondary structures in the translation initiation region. Hfq formed two complexes with RprA in a gel mobility assay with binding parameters similar to values previously determined for DsrA. Unlike DsrA, Hfq binding to RprA was inhibited by poly(A) and influenced by Hfq mutations on both the distal and proximal surfaces. Hfq increased the level of RprA binding to both rpoS RNAs but showed a much larger enhancement when rpoS-L, the entire leader region, was examined. The lower affinity of RprA for rpoS-L versus rpoS-S in the absence of Hfq suggests that Hfq overcomes an inhibitory structure within rpoS-L in stimulating RprA binding. Similar results were obtained with DsrA. The results indicate that the full upstream leader sequence of rpoS mRNA influences Hfq-facilitated annealing of RprA and DsrA and is likely to be involved in its regulation.