Viral replication rate regulates clinical outcome and CD8 T cell responses during highly pathogenic H5N1 influenza virus infection in mice

Since the first recorded infection of humans with H5N1 viruses of avian origin in 1997, sporadic human infections continue to occur with a staggering mortality rate of >60%. Although sustained human-to-human transmission has not occurred yet, there is a growing concern that these H5N1 viruses might acquire this trait and raise the specter of a pandemic. Despite progress in deciphering viral determinants of pathogenicity, we still lack crucial information on virus/immune system interactions pertaining to severe disease and high mortality associated with human H5N1 influenza virus infections. Using two human isolates of H5N1 viruses that differ in their pathogenicity in mice, we have defined mechanistic links among the rate of viral replication, mortality, CD8 T cell responses, and immunopathology. The extreme pathogenicity of H5N1 viruses was directly linked to the ability of the virus to replicate rapidly, and swiftly attain high steady-state titers in the lungs within 48 hours after infection. The remarkably high replication rate of the highly pathogenic H5N1 virus did not prevent the induction of IFN-β or activation of CD8 T cells, but the CD8 T cell response was ineffective in controlling viral replication in the lungs and CD8 T cell deficiency did not affect viral titers or mortality. Additionally, BIM deficiency ameliorated lung pathology and inhibited T cell apoptosis without affecting survival of mice. Therefore, rapidly replicating, highly lethal H5N1 viruses could simply outpace and overwhelm the adaptive immune responses, and kill the host by direct cytopathic effects. However, therapeutic suppression of early viral replication and the associated enhancement of CD8 T cell responses improved the survival of mice following a lethal H5N1 infection. These findings suggest that suppression of early H5N1 virus replication is key to the programming of an effective host response, which has implications in treatment of this infection in humans.     

Curved EFC/F-BAR-domain dimers are joined end to end into a filament for membrane invagination in endocytosis

Pombe Cdc15 homology (PCH) proteins play an important role in a variety of actin-based processes, including clathrin-mediated endocytosis (CME). The defining feature of the PCH proteins is an evolutionarily conserved EFC/F-BAR domain for membrane association and tubulation. In the present study, we solved the crystal structures of the EFC domains of human FBP17 and CIP4. The structures revealed a gently curved helical-bundle dimer of approximately 220 A in length, which forms filaments through end-to-end interactions in the crystals. The curved EFC dimer fits a tubular membrane with an approximately 600 A diameter. We subsequently proposed a model in which the curved EFC filament drives tubulation. In fact, striation of tubular membranes was observed by phase-contrast cryo-transmission electron microscopy, and mutations that impaired filament formation also impaired membrane tubulation and cell membrane invagination. Furthermore, FBP17 is recruited to clathrin-coated pits in the late stage of CME, indicating its physiological role.     

Sequential aldol condensation catalyzed by hyperthermophilic 2-deoxy-d-ribose-5-phosphate aldolase

Genes encoding 2-deoxy-d-ribose-5-phosphate aldolase (DERA) homologues from two hyperthermophiles, the archaeon Pyrobaculum aerophilum and the bacterium Thermotoga maritima, were expressed individually in Escherichia coli, after which the structures and activities of the enzymes produced were characterized and compared with those of E. coli DERA. To our surprise, the two hyperthermophilic DERAs showed much greater catalysis of sequential aldol condensation using three acetaldehydes as substrates than the E. coli enzyme, even at a low temperature (25 degrees C), although both enzymes showed much less 2-deoxy-d-ribose-5-phosphate synthetic activity. Both the enzymes were highly resistant to high concentrations of acetaldehyde and retained about 50% of their initial activities after a 20-h exposure to 300 mM acetaldehyde at 25 degrees C, whereas the E. coli DERA was almost completely inactivated after a 2-h exposure under the same conditions. The structure of the P. aerophilum DERA was determined by X-ray crystallography to a resolution of 2.0 A. The main chain coordinate of the P. aerophilum enzyme monomer was quite similar to those of the T. maritima and E. coli enzymes, whose crystal structures have already been solved. However, the quaternary structure of the hyperthermophilic enzymes was totally different from that of the E. coli DERA. The areas of the subunit-subunit interface in the dimer of the hyperthermophilic enzymes are much larger than that of the E. coli enzyme. This promotes the formation of the unique dimeric structure and strengthens the hydrophobic intersubunit interactions. These structural features are considered responsible for the extremely high stability of the hyperthermophilic DERAs.     

Identification of Cys385 in the isolated kinase insertion domain of heme-regulated eIF2 alpha kinase (HRI) as the heme axial ligand by site-directed mutagenesis and spectral characterization

Heme-regulated eIF2alpha kinase (HRI) is an important enzyme that modulates protein synthesis during cellular emergency/stress conditions, such as heme deficiency in red cells. It is essential to identify the heme axial ligand(s) and/or binding sites to establish the heme regulation mechanism of HRI. Previous reports suggest that a His residue in the N-terminal region and a Cys residue in the C-terminal region trans to the His are axial ligands of the heme. Moreover, mutational analyses indicate that a residue located in the kinase insertion (KI) domain between Kinase I and Kinase II domains in the C-terminal region is an axial ligand. In the present study, we isolate the KI domain of mouse HRI and employ site-directed mutagenesis to identify the heme axial ligand. The optical absorption spectrum of the Fe(III) hemin-bound wild-type KI displays a broad Soret band at around 373nm, while that of the Fe(II) heme-bound protein contains a band at 422nm. Spectral titration studies conducted for both the Fe(III) hemin and Fe(II) heme complexes with KI support a 1:1 stoichiometry of heme iron to protein. Resonance Raman spectra of Fe(III) hemin-bound KI suggest that thiol is the axial ligand in a 5-coordinate high-spin heme complex as a major form. Electron spin resonance (ESR) spectra of Fe(III) hemin-bound KI indicate that the axial ligands are OH(-) and Cys. Since Cys385 is the only cysteine in KI, the residue was mutated to Ser, and its spectral characteristics were analyzed. The Soret band position, heme spectral titration behavior and ESR parameters of the Cys385Ser mutant were markedly different from those of wild-type KI. Based on these spectroscopic findings, we conclude that Cys385 is an axial ligand of isolated KI.     

Low-intensity pulsed ultrasound (LIPUS) induces RANKL, MCP-1, and MIP-1beta expression in osteoblasts through the angiotensin II type 1 receptor

Constant mechanical stress is essential for the maintenance of bone mass and strength, which is achieved through the cooperative functions of osteoblasts and osteoclasts. However, it has not been fully elucidated how these cell types mediate mechanical signals. Low-intensity pulsed ultrasound (LIPUS) therapy is a recently developed method for application of mechanical stress, and is used clinically to promote bone fracture healing. In the present study, we applied LIPUS to osteoblasts at different stages of maturation and analyzed their chemokine and cytokine expression. In comparison with their immature counterparts, mature osteoblasts expressed significantly higher levels of mRNAs for the receptor activator of nuclear factor kappa B ligand (RANKL), monocyte chemoattractant protein (MCP)-1, and macrophage-inflammatory protein (MIP)-1beta after a few hours of LIPUS treatment. Intriguingly, protein and mRNA expression of angiotensin II type 1 receptor (AT1), a known mechanoreceptor in cardiomyocytes, was detected in osteoblasts, and the level of expression increased significantly during cell maturation. Furthermore, LIPUS-induced extracellular signal-regulated kinase (ERK) phosphorylation and RANKL/chemokine expression was abrogated by a specific AT1 inhibitor. Thus, AT1 may play one of the essential roles in bone metabolism as a mechanoreceptor of osteoblasts.     

Crystal structure of Ufc1, the Ufm1-conjugating enzyme

Ubiquitin and ubiquitin-like protein-conjugating enzymes play central roles in posttranslational modification processes. The ubiquitin-fold modifier 1 (Ufm1), one of a variety of ubiquitin-like modifiers, is covalently attached to target proteins via Uba5 and Ufm1-conjugating enzyme 1 (Ufc1), which are analogous to the E1 and E2 ubiquitylation enzymes. As Ufm1-related proteins are conserved in metazoa and plants, the Ufm1 system likely plays important roles in various multicellular organisms. Herein, we report the X-ray structure of human Ufc1 determined at 1.6 A resolution. The Ufc1 structure comprises a canonical E2 domain and an additional N-terminal domain. The Uba5 binding site on Ufc1 was assigned by structural comparison of Ufc1 and Ubc12 and related mutational analyses. In addition, we show that the N-terminal unique domain of Ufc1 contributes to thermal stability.     

Characterization of the denatured structure of pyrrolidone carboxyl peptidase from a hyperthermophile under nondenaturing conditions: role of the C-terminal alpha-helix of the protein in folding and stability

The cysteine-free pyrrolidone carboxyl peptidase (PCP-0SH) from a hyperthermophile, Pyrococcus furiosus, can be trapped in the denatured state under nondenaturing conditions, corresponding to the denatured structure that exists in equilibrium with the native state under physiological conditions. The denatured state is the initial state (D1 state) in the refolding process but differs from the completely denatured state (D2 state) in the concentrated denaturant. Also, it has been found that the D1 state corresponds to the heat-denatured state. To elucidate the structural basis of the D1 state, H/D exchange experiments with PCP-0SH were performed at pD 3.4 and 4 degrees C. The results indicated that amide protons in the C-terminal alpha6-helix region hardly exchanged in the D1 state with deuterium even after 7 days, suggesting that the alpha6-helix (from Ser188 to Glu205) of PCP-0SH was stably formed in the D1 state. In order to examine the role of the alpha6-helix in folding and stability, H/D exchange experiments with a mutant, A199P, at position 199 in the alpha6-helix region were performed. The alpha6-helix region of A199P in the D1 state was partially unprotected, while some hydrophobic residues were protected against the H/D exchange, although these hydrophobic residues were unprotected in the wild-type protein. These results suggest that the structure of A199P in the D1 state formed a temporary stable denatured structure with a non-native hydrophobic cluster and the unstructured alpha6-helix. Both the stability and the refolding rate decreased by the substitution of Pro for Ala199. We can conclude that the native-like helix (alpha6-helix) of PCP-0SH is already constructed in the D1 state and is necessary for efficient refolding into the native structure and stabilization of PCP-0SH.     

Mutagenesis of lysine 62, asparagine 64, and conserved region 1 reduces the activity of human ecto-ATPase (NTPDase 2)

The human ecto-ATPase (NTPDase 2) contains conserved motifs including five apyrase conserved regions (ACRs) and four conserved regions (CRs) as well as conserved lysine and arginine residues that are also present in other cell surface E-NTPDases. Some of the positively charged amino acids may be involved in ATP binding. The protein also contains six potential N-linked glycosylation sites. Results obtained with seven lysine and six arginine mutants indicate the importance of K62 that is located in CR1, K182, which is downstream of ACR3, and R155, which immediately follows CR3. Mutation of asparagine at the six potential N-linked glycosylation sites individually to glutamine established the importance of N64 in CR1 and N443 in ACR5 in protein function and expression. Mutation of N64, which is conserved in all cell surface NTPDases, results in the expression of an unstable protein, the activity of which is only manifested in the presence of concanavalin A. Both K62 and N64 reside in CR1 that is conserved in all cell surface NTPDases. In the sequence of the CR1 of human ecto-ATPase, 58WPADKENDTGIV69, 65DTG67 is similar to the phosphate-binding motif (DXG) in ACR1 and 4. The D65A and G67A mutants have reduced protein expression and activity. Mutations of other residues in CR1 to alanine led to partial to complete loss of protein expression and activity except for P59. The alanine mutants of the three acidic amino acid residues, D61, E63, and D65, all have decreased affinity for divalent ions. D61 can be substituted by glutamate, but E63 appears to be invariable. Taken together, these results indicate that CR1, which follows ACR1 in the cell surface NTPDases, is an essential structural element in these enzymes.     

Membrane topology of aspartate:alanine antiporter AspT from Comamonas testosteroni

We cloned the aspT gene encoding the L-aspartate:L-alanine antiporter AspTCt in Comamonas testosteroni genomic DNA. Analysis of the nucleotide sequence revealed that C. testosteroni has an asp operon containing aspT upstream of the l-aspartate 4-decarboxylase gene, and that the gene order of the asp operon of C. testosteroni is the inverse of that of Tetragenococcus halophilus. We used proteoliposomes to confirm the transport processes of AspTCt. To elucidate the two-dimensional structure of AspTCt, we analysed its membrane topology by means of alkaline phosphatase (PhoA) and beta-lactamase (BlaM) fusion methods. The fusion analyses revealed that AspTCt has seven transmembrane segments (TMs), a large cytoplasmic loop containing approximately 200 amino acid residues between TM4 and TM5, a cytoplasmic N-terminus, and a periplasmic C-terminus. These results suggest that the orientation of the N-terminus of AspTCt differs from that of tetragenococcal AspT, even though these two AspT orthologues catalyse the same transport reactions.