Structure of the RNA Binding Domain of a DEAD-Box Helicase Bound to Its Ribosomal RNA Target Reveals a Novel Mode of Recognition by an RNA Recognition Motif

DEAD-box RNA helicases of the bacterial DbpA subfamily are localized to their biological substrate when a carboxy-terminal RNA recognition motif domain binds tightly and specifically to a segment of 23S ribosomal RNA (rRNA) that includes hairpin 92 of the peptidyl transferase center. A complex between a fragment of 23S rRNA and the RNA binding domain (RBD) of the Bacillus subtilis DbpA protein YxiN was crystallized and its structure was determined to 2.9 Å resolution, revealing an RNA recognition mode that differs from those observed with other RNA recognition motifs. The RBD is bound between two RNA strands at a three-way junction. Multiple phosphates of the RNA backbone interact with an electropositive band generated by lysines of the RBD. Nucleotides of the single-stranded loop of hairpin 92 interact with the RBD, including the guanosine base of G2553, which forms three hydrogen bonds with the peptide backbone. A G2553U mutation reduces the RNA binding affinity by 2 orders of magnitude, confirming that G2553 is a sequence specificity determinant in RNA binding. Binding of the RBD to 23S rRNA in the late stages of ribosome subunit maturation would position the ATP-binding duplex destabilization fragment of the protein for interaction with rRNA in the peptidyl transferase cleft of the subunit, allowing it to “melt out” unstable secondary structures and allow proper folding.     

Structure-Function Analysis of the Acyl Carrier Protein Synthase (AcpS) from Mycobacterium tuberculosis

We have solved the crystal structure of the acyl carrier protein synthase (AcpS) from Mycobacterium tuberculosis (Mtb) at 1.95 Å resolution. AcpS, a 4-phosphopantetheinyl transferase, activates two distinct acyl carrier proteins (ACPs) that are present in fatty acid synthase (FAS) systems FAS-I and FAS-II, the ACP-I domain and the mycobacterial ACP-II protein (ACPM), respectively. Mtb, the causal agent of tuberculosis (TB), and all other members of the Corynebacterineae family are unique in possessing both FAS systems to produce and to elongate fatty acids to mycolic acids, the hallmark of mycobacterial cell wall. Various steps in this process are prime targets for first-line anti-TB agents. A comparison of the Mtb AcpS structure determined here with those of other AcpS proteins revealed unique structural features in Mtb AcpS, namely, the presence of an elongated helix followed by a flexible loop and a moderately electronegative surface unlike the positive surface common to other AcpSs. A structure-based sequence comparison between AcpS and its ACP substrates from various species demonstrated that the proteins of the Corynebacterineae family display high sequence conservation, forming a segregated subgroup of AcpS and ACPs. Analysis of the putative interactions between AcpS and ACPM from Mtb, based on a comparison with the complex structure from Bacillus subtilis, showed that the Mtb AcpS and ACPM lack the electrostatic complementarity observed in B. subtilis. Taken together, the common characteristic of the Corynebacterineae family is likely reflected in the participation of different residues and interactions used for binding the Mtb AcpS to ACP-I and ACPM. The distinct features and essentiality of AcpS, as well as the mode of interaction with ACPM and ACP-I in Mtb, could be exploited for the design of AcpS inhibitors, which, similarly to other inhibitors of fatty acid synthesis, are expected to be effective anti-TB-specific drugs.     

Crystal structure of the RRM domain of poly(A)-specific ribonuclease reveals a novel m(7)G-cap-binding mode

Poly(A)-specific ribonuclease (PARN) is a processive 3′-exoribonuclease involved in the decay of eukaryotic mRNAs. Interestingly, PARN interacts not only with the 3′ end of the mRNA but also with its 5′ end as PARN contains an RRM domain that specifically binds both the poly(A) tail and the 7-methylguanosine (m⁷G) cap. The interaction of PARN with the 5′ cap of mRNAs stimulates the deadenylation activity and enhances the processivity of this reaction. We have determined the crystal structure of the PARN-RRM domain with a bound m⁷G triphosphate nucleotide, revealing a novel binding mode for the m⁷G cap. The structure of the m⁷G binding pocket is located outside of the canonical RNA-binding surface of the RRM domain and differs significantly from that of other m⁷G-cap-binding proteins. The crystal structure also shows a remarkable conformational flexibility of the RRM domain, leading to a perfect exchange of two α-helices with an adjacent protein molecule in the crystal lattice.     

Structure and Function of the FeoB G-Domain from Methanococcus jannaschii

FeoB in bacteria and archaea is involved in the uptake of ferrous iron (Fe²⁺), an important cofactor in biological electron transfer and catalysis. Unlike any other known prokaryotic membrane protein, FeoB contains a GTP-binding domain at its N-terminus. We determined high-resolution X-ray structures of the FeoB G-domain from Methanococcus jannaschii with and without bound GDP or Mg²⁺-GppNHp. The G-domain forms the same dimer in all three structures, with the nucleotide-binding pockets at the dimer interface, as in the ATP-binding domain of ABC transporters. The G-domain follows the typical fold of nucleotide-binding proteins, with a β-strand inserted in switch I that becomes partially disordered upon GTP binding. Switch II does not contact the nucleotide directly and does not change its conformation in response to the bound nucleotide. Release of the nucleotide causes a rearrangement of loop L6, which we identified as the G5 region of FeoB. Together with the C-terminal helix, this loop may transmit the information about the nucleotide-bound state from the G-domain to the transmembrane region of FeoB.     

The Structure of the Neisserial Lipooligosaccharide Phosphoethanolamine Transferase A (LptA) Required for Resistance to Polymyxin

Gram-negative bacteria possess an outer membrane envelope consisting of an outer leaflet of lipopolysaccharides, also called endotoxins, which protect the pathogen from antimicrobial peptides and have multifaceted roles in virulence. Lipopolysaccharide consists of a glycan moiety attached to lipid A, embedded in the outer membrane. Modification of the lipid A headgroups by phosphoethanolamine (PEA) or 4-amino-arabinose residues increases resistance to the cationic cyclic polypeptide antibiotic, polymyxin. Lipid A PEA transferases are members of the YhjW/YjdB/YijP superfamily and usually consist of a transmembrane domain anchoring the enzyme to the periplasmic face of the cytoplasmic membrane attached to a soluble catalytic domain. The crystal structure of the soluble domain of the protein of the lipid A PEA transferase from Neisseria meningitidis has been determined crystallographically and refined to 1.4 Å resolution. The structure reveals a core hydrolase fold similar to that of alkaline phosphatase. Loop regions in the structure differ, presumably to enable interaction with the membrane-localized substrates and to provide substrate specificity. A phosphorylated form of the putative nucleophile, Thr280, is observed. Metal ions present in the active site are coordinated to Thr280 and to residues conserved among the family of transferases. The structure reveals the protein components needed for the transferase chemistry; however, substrate-binding regions are not evident and are likely to reside in the transmembrane domain of the protein.     

A novel two-domain architecture within the amino acid kinase enzyme family revealed by the crystal structure of Escherichia coli glutamate 5-kinase

Glutamate 5-kinase (G5K) makes the highly unstable product glutamyl 5-phosphate (G5P) in the initial, controlling step of proline/ornithine synthesis, being feedback-inhibited by proline or ornithine, and causing, when defective, clinical hyperammonaemia. We determined two crystal structures of G5K from Escherichia coli, at 2.9 A and 2.5 A resolution, complexed with glutamate and sulphate, or with G5P, sulphate and the proline analogue 5-oxoproline. E. coli G5K presents a novel tetrameric (dimer of dimers) architecture. Each subunit contains a 257 residue AAK domain, typical of acylphosphate-forming enzymes, with characteristic alpha(3)beta(8)alpha(4) sandwich topology. This domain is responsible for catalysis and proline inhibition, and has a crater on the beta sheet C-edge that hosts the active centre and bound 5-oxoproline. Each subunit contains a 93 residue C-terminal PUA domain, typical of RNA-modifying enzymes, which presents the characteristic beta(5)beta(4) sandwich fold and three alpha helices. The AAK and PUA domains of one subunit associate non-canonically in the dimer with the same domains of the other subunit, leaving a negatively charged hole between them that hosts two Mg ions in one crystal, in line with the G5K requirement for free Mg. The tetramer, formed by two dimers interacting exclusively through their AAK domains, is flat and elongated, and has in each face, pericentrically, two exposed active centres in alternate subunits. This would permit the close apposition of two active centres of bacterial glutamate-5-phosphate reductase (the next enzyme in the proline/ornithine-synthesising route), supporting the postulated channelling of G5P. The structures clarify substrate binding and catalysis, justify the high glutamate specificity, explain the effects of known point mutations, and support the binding of proline near glutamate. Proline binding may trigger the movement of a loop that encircles glutamate, and which participates in a hydrogen bond network connecting active centres, which is possibly involved in the cooperativity for glutamate.     

Structural and Functional Studies of the Yeast Class II Hda1 Histone Deacetylase Complex

Yeast class II Hda1 histone deacetylase (HDAC) complex is an H2B- and H3-specific HDAC in Saccharomyces cerevisiae consisting of three previously identified subunits, the catalytic subunit scHda1p and two non-catalytic structural subunits scHda2p and scHda3p. We co-expressed and co-purified recombinant yeast class II HDAC complex from bacteria as a functionally active and trichostatin-A-sensitive hetero-tetrameric complex. According to an extensive analysis of domain organization and interaction of all subunits (or domains), the N-terminal domain of scHda1p associates through the C-terminal coiled-coil domains (CCDs) of the scHda2p-scHda3p sub-complex, yielding a truncated scHda1pHDAC-scHda2pCCD2-scHda3pCCD3 complex with indistinguishable deacetylase activity compared to the full-length complex in vitro. We characterized the interaction of the HDAC complex with either single-stranded or double-stranded DNA and identified the N-terminal halves of scHda2p and scHda3p as binding modules. A high-resolution structure of the scHda3p DNA-binding domain by X-ray crystallography is presented. The crystal structure shows an unanticipated structural homology with the C-terminal helicase lobes of SWI2/SNF2 chromatin-remodeling domains of the Rad54 family enzymes. DNA binding is unspecific for nucleotide sequence and structure, similar to the Rad54 enzymes in vitro. Our structural and functional analyses of the budding yeast class II Hda1 HDAC complex provide insight into DNA recognition and deacetylation of histones in nucleosomes.     

Apo and Calcium-Bound Crystal Structures of Cytoskeletal Protein Alpha-14 Giardin (Annexin E1) from the Intestinal Protozoan Parasite Giardia lamblia

Alpha-14 giardin (annexin E1), a member of the alpha giardin family of annexins, has been shown to localize to the flagella of the intestinal protozoan parasite Giardia lamblia. Alpha giardins show a common ancestry with the annexins, a family of proteins most of which bind to phospholipids and cellular membranes in a Ca²⁺-dependent manner and are implicated in numerous membrane-related processes including cytoskeletal rearrangements and membrane organization. It has been proposed that alpha-14 giardin may play a significant role during the cytoskeletal rearrangement during differentiation of Giardia. To gain a better understanding of alpha-14 giardin's mode of action and its biological role, we have determined the three-dimensional structure of alpha-14 giardin and its phospholipid-binding properties. Here, we report the apo crystal structure of alpha-14 giardin determined in two different crystal forms as well as the Ca²⁺-bound crystal structure of alpha-14 giardin, refined to 1.9, 1.6 and 1.65 Å, respectively. Although the overall fold of alpha-14 giardin is similar to that of alpha-11 giardin, multiwavelength anomalous dispersion phasing was required to solve the alpha-14 giardin structure, indicating significant structural differences between these two members of the alpha giardin family. Unlike most annexin structures, which typically possess N-terminal domains, alpha-14 giardin is composed of only a core domain, followed by a C-terminal extension that may serve as a ligand for binding to cytoskeletal protein partners in Giardia. In the Ca²⁺-bound structure we detected five bound calcium ions, one of which is a novel, highly coordinated calcium-binding site not previously observed in annexin structures. This novel high-affinity calcium-binding site is composed of seven protein donor groups, a feature rarely observed in crystal structures. In addition, phospholipid-binding assays suggest that alpha-14 giardin exhibits calcium-dependent binding to phospholipids that coordinate cytoskeletal disassembly/assembly during differentiation of the parasite.     

The Crystal Structure and Activity of a Putative Trypanosomal Nucleoside Phosphorylase Reveal It to be a Homodimeric Uridine Phosphorylase

Purine nucleoside phosphorylases (PNPs) and uridine phosphorylases (UPs) are closely related enzymes involved in purine and pyrimidine salvage, respectively, which catalyze the removal of the ribosyl moiety from nucleosides so that the nucleotide base may be recycled. Parasitic protozoa generally are incapable of de novo purine biosynthesis; hence, the purine salvage pathway is of potential therapeutic interest. Information about pyrimidine biosynthesis in these organisms is much more limited. Though all seem to carry at least a subset of enzymes from each pathway, the dependency on de novo pyrimidine synthesis versus salvage varies from organism to organism and even from one growth stage to another. We have structurally and biochemically characterized a putative nucleoside phosphorylase (NP) from the pathogenic protozoan Trypanosoma brucei and find that it is a homodimeric UP. This is the first characterization of a UP from a trypanosomal source despite this activity being observed decades ago. Although this gene was broadly annotated as a putative NP, it was widely inferred to be a purine nucleoside phosphorylase. Our characterization of this trypanosomal enzyme shows that it is possible to distinguish between PNP and UP activity at the sequence level based on the absence or presence of a characteristic UP-specificity insert. We suggest that this recognizable feature may aid in proper annotation of the substrate specificity of enzymes in the NP family.     

A C-Terminal Phosphatase Module Conserved in Vertebrate CMP-Sialic Acid Synthetases Provides a Tetramerization Interface for the Physiologically Active Enzyme

The biosynthesis of sialic acid-containing glycoconjugates is crucial for the development of vertebrate life. Cytidine monophosphate-sialic acid synthetase (CSS) catalyzes the metabolic activation of sialic acids. In vertebrates, the enzyme is chimeric, with the N-terminal domain harboring the synthetase activity. The function of the highly conserved C-terminal domain (CSS-CT) is unknown. To shed light on its biological function, we solved the X-ray structure of murine CSS-CT to 1.9 Å resolution. CSS-CT is a stable shamrock-like tetramer that superimposes well with phosphatases of the haloacid dehalogenase superfamily. However, a region found exclusively in vertebrate CSS-CT appears to block the active-site entrance. Accordingly, no phosphatase activity was observed in vitro, which points toward a nonenzymatic function of CSS-CT. A computational three-dimensional model of full-length CSS, in combination with in vitro oligomerization studies, provides evidence that CSS-CT serves as a platform for the quaternary organization governing the kinetic properties of the physiologically active enzyme as demonstrated in kinetic studies.