Disordered regions in C-domain structure of influenza virus M1 protein

Influenza virus matrix M1 protein is one of the main structural components of the virion performing also many different functions in infected cell. X-ray analysis data with 2.08 angstrom resolution were obtained only for the N-terminal part of M1 protein molecule (residues 2-158) but not for its C-terminal domain (159-252). In the present work M1 protein of A/Puerto Rico/8/34 (H1N1) virus strain in acidic solution was investigated with the help of tritium bombardment. Tritium label incorporation into M1 protein domains preferentially labeled the C-domain and inter-domain loops. Analytical centrifugation and dynamic light scattering experiments demonstrated increased hydrodynamic parameters (diameter) that may be explained by low degree of M1 structural organization. Computational analysis of M1 protein by intrinsic disorder predictions methods also demonstrated the presence of unfolded regions mostly in the C-domain and inter-domain loops. It is suggested, that influenza virus M1 polyfunctionality in infected cell is determined by its tertiary structure plasticity which in its turn results from the presence of unstructured regions.     

Spatial structure peculiarities of influenza A virus matrix M1 protein in an acidic solution that simulates the internal lysosomal medium

The structure of the C-terminal domain of the influenza virus A matrix M1 protein, for which X-ray diffraction data were still missing, was studied in acidic solution. Matrix M1 protein was bombarded with thermally-activated tritium atoms, and the resulting intramolecular distribution of the tritium label was analyzed to assess the steric accessibility of the amino acid residues in this protein. This technique revealed that interdomain loops and the C-terminal domain of the protein are the most accessible to labeling with tritium atoms. A model of the spatial arrangement of the C-terminal domain of matrix M1 protein was generated using rosetta software adjusted to the data obtained by tritium planigraphy experiments. This model suggests that the C-terminal domain is an almost flat layer with a three-α-helical structure. To explain the high level of tritium label incorporation into the C-terminal domain of the M1 protein in an acidic solution, we also used independent experimental approaches (CD spectroscopy, limited proteolysis and MALDI-TOF MS analysis of the proteolysis products, dynamic light scattering and analytical ultracentrifugation), as well as multiple computational algorithms, to analyse the intrinsic protein disorder. Taken together, the results obtained in the present study indicate that the C-terminal domain is weakly structured. We hypothesize that the specific 3D structural peculiarities of the M1 protein revealed in acidic pH solution allow the protein greater structural flexibility and enable it to interact effectively with the components of the host cell.     

Tritium planigraphy: Differences in the spatial structures of the influenza virus M1 protein in crystal,solution, and virion

The structure of the M1 protein of the influenza virus A/Puerto Rico/8/34 (PR8, subtype H1N1) in solution at acidic pH and in the composition of the virion has been studied by the tritium planigraphy method. A model of the spatial structure was constructed using a special algorithm simulating the experiment and a set of algorithms for predicting the secondary structure and disordered regions in proteins. The tertiary structure was refined using the Rosetta program. For a comparison of the structures in solution and inside the virion, the data of X-ray diffraction analysis for the NM domain were also used. The main difference in the structures of the protein in solution and the crystalline state is observed in the region of contact of N and M domains, which in the crystalline state is packed more densely. The regions of the maximum label incorporation almost completely coincide with unstructured regions in the protein that were predicted by the bioinformatics analysis. These regions are concentrated in the C domain and in loop regions between M, N, and C domains. The data were confirmed by analytical centrifugation and dynamic light scattering. Anomalous hydrodynamic dimensions and a low structuration of the M1 protein in solution were found. The polyfunctionality of the protein in the cell is probably related to its flexible tertiary structure, which, owing to unstructured regions, provides contact with various partner molecules.     

Differences in spatial structures of the influenza virus M1 protein in crystal, solution and virion

Spatial structure of the influenza virus A/Puerto Rico/8/34 (PR8, subtype H1N1) M1 protein in a solution and composition of the virion was studied by tritium planigraphy technique. The special algorithm for modeling of the spatial structure is used to simulate the experiment, as well as a set of algorithms predicting secondary structure and disordered regions in proteins. Tertiary structures were refined using the program Rosetta. To compare the structures in solution and in virion, also used the X-ray diffraction data for NM-domain. The main difference between protein structure in solution and crystal is observed in the contact region of N- and M-domains, which are more densely packed in the crystalline state. Locations include the maximum label is almost identical to the unstructured regions of proteins predicted by bioinformatics analysis. These areas are concentrated in the C-domain and in the loop regions between the M-, N-, and C-domains. Analytical centrifugation and dynamic laser light scattering confirm data of tritium planigraphy. Anomalous hydrodynamic size, and low structuring of the M1 protein in solution were found. The multifunctionality of protein in the cell appears to be associated with its plastic tertiary structure, which provides at the expense of unstructured regions of contact with various molecules-partners.     

The influenza C virus CM2 protein can alter intracellular pH, and its transmembrane domain can substitute for that of the influenza A virus M2 protein and support infectious virus production

The influenza C virus CM2 protein and a chimeric influenza A virus M2 protein (MCM) containing the CM2 transmembrane domain were assessed for their ability to functionally replace the M2 protein. While all three proteins could alter cytosolic pH to various degrees when expressed from cDNA, only M2 and MCM could at least partially restore infectious virus production to M2-deficient influenza A viruses. The data suggest that while the CM2 ion channel activity is similar to that of M2, sequences in the extracellular and/or cytoplasmic domains play important roles in infectious virus production.     

A mutation on influenza C virus M1 protein affects virion morphology by altering the membrane affinity of the protein

Reverse genetics has been documented for influenza A, B, and Thogoto viruses belonging to the family Orthomyxoviridae. We report here the reverse genetics of influenza C virus, another member of this family. The seven viral RNA (vRNA) segments of C/Ann Arbor/1/50 were expressed in 293T cells from cloned cDNAs, together with nine influenza C virus proteins. At 48 h posttransfection, the infectious titer of the culture supernatant was determined to be 2.51 x 10(3) 50% egg infectious doses/ml, which is lower than the number of influenza C virus-like particles (VLPs) (10(6)/ml) generated using the same system. By generating influenza C VLPs containing a given vRNA segment, we showed that each of the vRNA segments was similarly synthesized in the plasmid-transfected cells but that some segments were less efficiently incorporated into the VLPs. This finding leads us to speculate that the differences in incorporation efficiency into VLPs between segments might be a reason for the inefficient production of infectious viruses. Second, we generated a mutant recombinant virus, rMG96A, which possesses an Ala-->Thr mutation at residue 24 of the M1 protein, a substitution demonstrated to be involved in the morphology (filamentous or spherical) of the influenza C VLPs. As expected, rMG96A exhibited a spherical morphology, whereas recombinant wild-type of C/Ann Arbor/1/50, rWT, exhibited a mainly filamentous morphology. Membrane flotation analysis of the cells infected with rWT or rMG96A revealed a difference in the ratio of membrane-associated M1 proteins, suggesting that the affinity of M1 protein to the cell membrane is a determinant for virion morphology.     

Chemical ligation of the influenza M2 protein for solid‐state NMR characterization of the cytoplasmic domain

Solid‐state NMR‐based structure determination of membrane proteins and large protein complexes faces the challenge of limited spectral resolution when the proteins are uniformly ¹³C‐labeled. A strategy to meet this challenge is chemical ligation combined with site‐specific or segmental labeling. While chemical ligation has been adopted in NMR studies of water‐soluble proteins, it has not been demonstrated for membrane proteins. Here we show chemical ligation of the influenza M2 protein, which contains a transmembrane (TM) domain and two extra‐membrane domains. The cytoplasmic domain, which contains an amphipathic helix (AH) and a cytoplasmic tail, is important for regulating virus assembly, virus budding, and the proton channel activity. A recent study of uniformly ¹³C‐labeled full‐length M2 by spectral simulation suggested that the cytoplasmic tail is unstructured. To further test this hypothesis, we conducted native chemical ligation of the TM segment and part of the cytoplasmic domain. Solid‐phase peptide synthesis of the two segments allowed several residues to be labeled in each segment. The post‐AH cytoplasmic residues exhibit random‐coil chemical shifts, low bond order parameters, and a surface‐bound location, thus indicating that this domain is a dynamic random coil on the membrane surface. Interestingly, the protein spectra are similar between a model membrane and a virus‐mimetic membrane, indicating that the structure and dynamics of the post‐AH segment is insensitive to the lipid composition. This chemical ligation approach is generally applicable to medium‐sized membrane proteins to provide site‐specific structural constraints, which complement the information obtained from uniformly ¹³C, ¹⁵N‐labeled proteins.     

Restricted domain mobility in the Candida albicans Ess1 prolyl isomerase

Ess1 is a peptidyl prolyl cis/trans isomerase that is required for virulence of the pathogenic fungi Candida albicans and Cryptococcus neoformans. The enzyme isomerizes the phospho-Ser-Pro linkages in the C-terminal domain of RNA polymerase II. Its human homolog, Pin1, has been implicated in a wide range of human diseases, including cancer and Alzheimer's disease. Crystallographic and NMR studies have demonstrated that the sequence linking the catalytic isomerase domain and the substrate binding WW domain of Pin1 is unstructured and that the two domains are only loosely associated in the absence of the substrate. In contrast, the crystal structure of C. albicans Ess1 revealed a highly ordered linker that contains a three turn α-helix and extensive association between the two tightly juxtaposed domains. In part to address the concern that the marked differences in the domain interactions for the human and fungal structures might reflect crystal lattice effects, NMR chemical shift analysis and ¹⁵N relaxation measurements have been employed to confirm that the linker of the fungal protein is highly ordered in solution. With the exception of two loops within the active site of the isomerase domain, the local backbone geometry observed in the crystal structure appears to be well preserved throughout the protein chain. The marked differences in interdomain interactions and linker flexibility between the human and fungal enzymes provide a structural basis for therapeutic targeting of the fungal enzymes.     

Multiple Interactions of the Intrinsically Disordered Region between the Helicase and Nuclease Domains of the Archaeal Hef Protein

Hef is an archaeal protein that probably functions mainly in stalled replication fork repair. The presence of an unstructured region was predicted between the two distinct domains of the Hef protein. We analyzed the interdomain region of Thermococcus kodakarensis Hef and demonstrated its disordered structure by CD, NMR, and high speed atomic force microscopy (AFM). To investigate the functions of this intrinsically disordered region (IDR), we screened for proteins interacting with the IDR of Hef by a yeast two-hybrid method, and 10 candidate proteins were obtained. We found that PCNA1 and a RecJ-like protein specifically bind to the IDR in vitro. These results suggested that the Hef protein interacts with several different proteins that work together in the pathways downstream from stalled replication fork repair by converting the IDR structure depending on the partner protein.     

Zinc- and pH-dependent conformational transition in a putative interdomain linker region of the influenza virus matrix protein M1

The matrix protein M1 of influenza A virus forms a shell beneath the viral envelope and sustains the virion architecture by interacting with other viral components. A structural change of M1 upon acidification of the virion interior in an early stage of virus infection is considered to be a key step to virus uncoating. We examined the structure of a 28-mer peptide (M1Lnk) representing a putative linker region between the N- and C-terminal domains of M1 by using circular dichroism, Raman, and absorption spectroscopy. M1Lnk assumes an alpha-helical structure in a mildly hydrophobic environment irrespective of pH, being consistent with the X-ray crystal structures of an N-terminal fragment of M1 at pH 7 and 4. In the presence of Zn(2+), on the other hand, M1Lnk takes a partially unfolded conformation at neutral pH with a tetrahedral coordination of two Cys residues and two His residues to a Zn(2+) ion in the central part of the peptide. Upon acidification, the peptide releases the Zn(2+) ion and refolds into the alpha-helix-rich structure with a midpoint of transition at pH 5.9. The pH-dependent conformational transition of M1Lnk strongly suggests that the interdomain linker region of M1 also undergoes a pH-dependent unfolding-refolding transition in the presence of Zn(2+). A small but significant portion of the M1 protein is bound to Zn(2+) in the virion, and the Zn(2+)-bound M1 molecule may play a special role in virus uncoating by changing the disposition of the N- and C-terminal domains upon acidification of the virion interior.     

Thermal stability of the three domains of streptokinase studied by circular dichroism and nuclear magnetic resonance.

Streptococcus equisimilis streptokinase (SK) is a single-chain protein of 414 residues that is used extensively in the clinical treatment of acute myocardial infarction due to its ability to activate human plasminogen (Plg). The mechanism by which this occurs is poorly understood due to the lack of structural details concerning both molecules and their complex. We reported recently (Parrado J et al., 1996, Protein Sci 5:693-704) that SK is composed of three structural domains (A, B, and C) with a C-terminal tail that is relatively unstructured. Here, we report thermal unfolding experiments, monitored by CD and NMR, using samples of intact SK, five isolated SK fragments, and two two-chain noncovalent complexes between complementary fragments of the protein. These experiments have allowed the unfolding processes of specific domains of the protein to be monitored and their relative stabilities and interdomain interactions to be characterized. Results demonstrate that SK can exist in a number of partially unfolded states, in which individual domains of the protein behave as single cooperative units. Domain B unfolds cooperatively in the first thermal transition at approximately 46 degrees C and its stability is largely independent of the presence of the other domains. The high-temperature transition in intact SK (at approximately 63 degrees C) corresponds to the unfolding of both domains A and C. Thermal stability of domain C is significantly increased by its isolation from the rest of the chain. By contrast, cleavage of the Phe 63-Ala 64 peptide bond within domain A causes thermal destabilization of this domain. The two resulting domain portions (A1 and A2) adopt unstructured conformations when separated. A1 binds with high affinity to all fragments that contain the A2 portion, with a concomitant restoration of the native-like fold of domain A. This result demonstrates that the mechanism whereby A1 stimulates the plasminogen activator activities of complementary SK fragments is the reconstitution of the native-like structure of domain A.     

A dihydropyridine receptor alpha1s loop region critical for skeletal muscle contraction is intrinsically unstructured and binds to a SPRY domain of the type 1 ryanodine receptor

The II-III loop of the dihydropyridine receptor (DHPR) alpha(1s) subunit is a modulator of the ryanodine receptor (RyR1) Ca(2+) release channel in vitro and is essential for skeletal muscle contraction in vivo. Despite its importance, the structure of this loop has not been reported. We have investigated its structure using a suite of NMR techniques which revealed that the DHPR II-III loop is an intrinsically unstructured protein (IUP) and as such belongs to a burgeoning structural class of functionally important proteins. The loop does not possess a stable tertiary fold: it is highly flexible, with a strong N-terminal helix followed by nascent helical/turn elements and unstructured segments. Its residual structure is loosely globular with the N and C termini in close proximity. The unstructured nature of the II-III loop may allow it to easily modify its interaction with RyR1 following a surface action potential and thus initiate rapid Ca(2+) release and contraction. The in vitro binding partner for the II-III was investigated. The II-III loop interacts with the second of three structurally distinct SPRY domains in RyR1, whose function is unknown. This interaction occurs through two preformed N-terminal alpha-helical regions and a C-terminal hydrophobic element. The A peptide corresponding to the helical N-terminal region is a common probe of RyR function and binds to the same SPRY domain as the full II-III loop. Thus the second SPRY domain is an in vitro binding site for the II-III loop. The possible in vivo role of this region is discussed.     

Involvement of an Arginine Triplet in M1 Matrix Protein Interaction with Membranes and in M1 Recruitment into Virus-Like Particles of the Influenza A(H1N1)pdm09 Virus

The influenza A(H1N1)pdm09 virus caused the first influenza pandemic of the 21st century. In this study, we wanted to decipher the role of conserved basic residues of the viral M1 matrix protein in virus assembly and release. M1 plays many roles in the influenza virus replication cycle. Specifically, it participates in viral particle assembly, can associate with the viral ribonucleoprotein complexes and can bind to the cell plasma membrane and/or the cytoplasmic tail of viral transmembrane proteins. M1 contains an N-terminal domain of 164 amino acids with two basic domains: the nuclear localization signal on helix 6 and an arginine triplet (R76/77/78) on helix 5. To investigate the role of these two M1 basic domains in influenza A(H1N1)pdm09 virus molecular assembly, we analyzed M1 attachment to membranes, virus-like particle (VLP) production and virus infectivity. In vitro, M1 binding to large unilamellar vesicles (LUVs), which contain negatively charged lipids, decreased significantly when the M1 R76/77/78 motif was mutated. In cells, M1 alone was mainly observed in the nucleus (47%) and in the cytosol (42%). Conversely, when co-expressed with the viral proteins NS1/NEP and M2, M1 was relocated to the cell membranes (55%), as shown by subcellular fractionation experiments. This minimal system allowed the production of M1 containing-VLPs. However, M1 with mutations in the arginine triplet accumulated in intracellular clusters and its incorporation in VLPs was strongly diminished. M2 over-expression was essential for M1 membrane localization and VLP production, whereas the viral trans-membrane proteins HA and NA seemed dispensable. These results suggest that the M1 arginine triplet participates in M1 interaction with membranes. This R76/77/78 motif is essential for M1 incorporation in virus particles and the importance of this motif was confirmed by reverse genetic demonstrating that its mutation is lethal for the virus. These results highlight the molecular mechanism of M1-membrane interaction during the formation of influenza A(H1N1)pdm09 virus particles which is essential for infectivity.     

Functional replacement of the carboxy-terminal two-thirds of the influenza A virus NS1 protein with short heterologous dimerization domains

The NS1 protein of influenza A/WSN/33 virus is a 230-amino-acid-long protein which functions as an interferon alpha/beta (IFN-alpha/beta) antagonist by preventing the synthesis of IFN during viral infection. In tissue culture, the IFN inhibitory function of the NS1 protein has been mapped to the RNA binding domain, the first 73 amino acids. Nevertheless, influenza viruses expressing carboxy-terminally truncated NS1 proteins are attenuated in mice. Dimerization of the NS1 protein has previously been shown to be essential for its RNA binding activity. We have explored the ability of heterologous dimerization domains to functionally substitute in vivo for the carboxy-terminal domains of the NS1 protein. Recombinant influenza viruses were generated that expressed truncated NS1 proteins of 126 amino acids, fused to 28 or 24 amino acids derived from the dimerization domains of either the Saccharomyces cerevisiae PUT3 or the Drosophila melanogaster Ncd (DmNcd) proteins. These viruses regained virulence and lethality in mice. Moreover, a recombinant influenza virus expressing only the first 73 amino acids of the NS1 protein was able to replicate in mice lacking three IFN-regulated antiviral enzymes, PKR, RNaseL, and Mx, but not in wild-type (Mx-deficient) mice, suggesting that the attenuation was mainly due to an inability to inhibit the IFN system. Remarkably, a virus with an NS1 truncated at amino acid 73 but fused to the dimerization domain of DmNcd replicated and was also highly pathogenic in wild-type mice. These results suggest that the main biological function of the carboxy-terminal region of the NS1 protein of influenza A virus is the enhancement of its IFN antagonist properties by stabilizing the NS1 dimeric structure.     

Quantitation of the glycoprotein spike area on the surface of enveloped viruses

The density of glycoprotein (GP) distribution on the virion surface substantially influences the virus infectivity and pathogenicity. A method to quantitatively determine the area occupied by surface GP spikes was proposed for influenza virus (Flu) strain A/PR/8/34 on the basis of data of tritium bombardment and dynamic light scattering. The latter was used to measure the diameter of intact virions and subviral particles (Flu virions lacking GP spikes after bromelain digestion). Intact virions and subviral particles were bombarded with a hot tritium atom flux, and the specific radioactivity of the matrix M1 protein was analyzed. The tritium label was incorporated into the amino acid residues of a thin exposed protein layer and partly penetrated through the lipid bilayer of the viral envelope, labeling M1, located under the lipid bilayer. The tritium label distribution among different amino acid residues was the same in M1 isolated from subviral particles and M1 isolated from intact virions, demonstrating that the M1 spatial structure remained unchanged during proteolysis of GP spikes. The difference in specific radioactivity between the M1 proteins isolated from intact virions and subviral particles was used to calculate the GP-free portion of the viral surface. Approximating the Flu virion as a sphere, the GP-covered area was estimated at 1.4 × 10⁴ nm², about 40% of the total virion surface. This was consistent with the cryoelectron tomography data published for Flu strain A/X-31. The approach can be applied for other enveloped high pathogenic viruses, such as HIV and the Ebola virus.     

1H, 15N, and 13C resonance assignments for a monomeric mutant of the HIV-1 capsid protein

The mature fullerene cone-shaped capsid of the human immunodeficiency virus 1 is composed of about 1,500 copies of the capsid protein (CA). The CA is 231 residues long, and consists of two distinct structural domains, the N-terminal domain and the C-terminal domain (CTD), joined by a flexible linker. The wild type CA exhibits monomer-dimer equilibrium in solution through the CTD-CTD dimerization. This CTD-CTD interaction, together with other intermolecular interdomain interactions, plays significant roles during the assembly of the mature capsid. In addition, CA-CA interactions also play a role in the assembly of the immature virion. The CA also interacts with some host cell proteins within the viral replication cycle. Thus, the capsid protein has been of significant interest as a target for designing inhibitors of assembly of immature virions and mature capsids and inhibitors of its interactions with host cell proteins. However, the equilibrium exhibited by the wild-type CA protein between the monomeric and dimeric states, along with the inherent flexibility from the interdomain linker, have hindered attempts at structural determination by solution NMR and X-ray crystallography methods. In this study, we have utilized a CA protein with W184A and M185A mutations that abolish the dimerization of CA protein as well as its infectivity, but preserve most of the remaining properties of the wild type CA. We have determined the detailed solution structure of the monomeric W184A/M185A-CA protein using 3D-NMR spectroscopy. Here, we present the detailed sequence-specific NMR assignments for this protein.     

Characterization of the 1918 "Spanish" influenza virus matrix gene segment

The coding region of influenza A virus RNA segment 7 from the 1918 pandemic virus, consisting of the open reading frames of the two matrix genes M1 and M2, has been sequenced. While this segment is highly conserved among influenza virus strains, the 1918 sequence does not match any previously sequenced influenza virus strains. The 1918 sequence matches the consensus over the M1 RNA-binding domains and nuclear localization signal and the highly conserved transmembrane domain of M2. Amino acid changes that correlate with high yield and pathogenicity in animal models were not found in the 1918 strain. Phylogenetic analyses suggest that both genes were mammalian adapted and that the 1918 sequence is very similar to the common ancestor of all subsequent human and classical swine matrix segments. The 1918 sequence matches other mammalian strains at 4 amino acids in the extracellular domain of M2 that differ consistently between avian and mammalian strains, suggesting that the matrix segment may have been circulating in human strains for at least several years before 1918.     

Biophysical characterization of the unstructured cytoplasmic domain of the human neuronal adhesion protein neuroligin 3

Cholinesterase-like adhesion molecules (CLAMs) are a family of neuronal cell adhesion molecules with important roles in synaptogenesis, and in maintaining structural and functional integrity of the nervous system. Our earlier study on the cytoplasmic domain of one of these CLAMs, the Drosophila protein, gliotactin, showed that it is intrinsically unstructured in vitro. Bioinformatic analysis suggested that the cytoplasmic domains of other CLAMs are also intrinsically unstructured, even though they bear no sequence homology to each other or to any known protein. In this study, we overexpress and purify the cytoplasmic domain of human neuroligin 3, notwithstanding its high sensitivity to the Escherichia coli endogenous proteases that cause its rapid degradation. Using bioinformatic analysis, sensitivity to proteases, size exclusion chromatography, fluorescence correlation spectroscopy, analytical ultracentrifugation, small angle x-ray scattering, circular dichroism, electron spin resonance, and nuclear magnetic resonance, we show that the cytoplasmic domain of human neuroligin 3 is intrinsically unstructured. However, several of these techniques indicate that it is not fully extended, but becomes significantly more extended under denaturing conditions.     

Hydrogen exchange and proteolytic degradation of ribonuclease A. The local splitting of the native structure and the conformation of loop segmentes

The limited proteolytic sites or nicksites are present only in one of the five loops of the RNase A molecule. The splitted loop 15-23 connects two structural domains in the hinge region of the interdomain contacts of the V-shaped molecule. The other four loops are inside two domains, 64-71 and 112-115 in the domain I (1-19, 47-81, 102-106) and 36-42 and 88-95 in the domain II (20-46, 82-101). Because of enhanced chain flexibility of the splitted loop in the pH-dependent conformational isomerization, deformation of its structure is slighter under the influence of the intermolecular contacts in the crystal lattice and more significant changes occur in loop conformation at the formation of the 3D swapped dimer of the RNase A molecule. The proteolytic splitting of the 15-23 loop proceeds due to the local fluctuation of the native protein structure.