HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

J. Biol. Chem., Vol. 277, Issue 9, 7108-7117, March 1, 2002

This Article Abstract Full Text (PDF) All Versions of this Article: 277/9/7108    most recent M109045200v1 Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Lommer, B. S. Articles by Luo, M. Search for Related Content PubMed PubMed Citation Articles by Lommer, B. S. Articles by Luo, M. Social Bookmarking                      What's this?

Structural Plasticity in Influenza Virus Protein NS2 (NEP)^*

Barbara S. Lommer and Ming Luo

From the Department of Microbiology, University of Alabama at Birmingham, Birmingham, Alabama 35294-4400

Received for publication, September 19, 2001, and in revised form, December 18, 2001

	ABSTRACT

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

The cellular nuclear transport machinery relies on the assembly of specialized transport complexes between soluble transport receptors, transport substrates, and additional accessory proteins. This study focuses on the structural characteristics of influenza virus protein NS2 (NEP), which interacts with the nuclear export machinery during viral replication, and has been proposed to act as an adapter molecule between the nuclear export machinery and the viral ribonucleoprotein complex. For this purpose, we have purified recombinant NS2 under nondenaturing conditions, and have investigated its structure and aggregation state using optical spectroscopy, differential scanning calorimetry, as well as hydrodynamic techniques. Our results indicate that isolated NS2 exists as a monomer in solution, and adopts a compact, but very flexible conformation, which shows characteristics of the molten globule state under near physiological conditions. Proteolytic sensitivity suggests that, despite its overall plasticity, the structure of NS2 is heterogeneous. While the C terminus of the protein adopts a relatively rigid conformation, its N terminus, which is recognized by the nuclear export machinery, exists in a highly mobile and exposed state. It is proposed that the flexibility observed in the nuclear export domain of NS2 is an important element in the recognition of substrate proteins by the nuclear export machinery.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Active transport of macromolecules across the nuclear membrane is an important mechanism in the regulation of gene transcription and translation. Signal-mediated transport is typically initiated by soluble transport receptors that recognize cis-acting nuclear transport signals in their substrates and target their cargo to the nuclear pore complex. Translocation through the nuclear pore, release of substrate after transport, and directionality of the transport process are thought to be achieved mainly through action of the GTPase Ran and its regulatory factors (1-3).

Influenza virus, like many viruses that replicate in the nucleus, takes advantage of these cellular pathways to translocate its genome, mRNAs, and proteins across the nuclear membrane (4, 5). Three influenza virus proteins in particular have been found to interact with the nuclear transport machinery: nucleoprotein (NP),¹ matrix protein (M1), and the nonstructural protein 2 (NS2, also called NEP).

NP encapsidates the viral genomic RNA (6, 7), is able to shuttle between the nucleus and the cytoplasm (8), and is thought to catalyze the import of the viral ribonucleoprotein complexes (vRNPs) into the nucleus at early times after infection (9-11). In addition, NP interacts with the nuclear export receptor Crm1 in vitro and has been proposed to achieve the nuclear export of progeny vRNPs during virion assembly (12).

M1 is also transported into the nucleus (13) where it associates with the vRNPs (14). While it has been shown to be essential for vRNP export (14, 15), its role in this process is not clearly defined. M1 may escort the progeny vRNPs out of the nucleus (14), release the vRNPs from the nuclear matrix (16, 17), and/or prevent the reimport of the vRNPs after translocation (8).

The small protein NS2 is the latest influenza protein for which an interaction with the nuclear export machinery has been identified. NS2s of both type A and type B influenza viruses contain a HIV-1 Rev-like nuclear export signal (NES) at their N terminus that has been shown to interact with cellular nucleoporins in a yeast two-hybrid system, and that can mediate the nuclear export of heterologous proteins (18, 19). In addition, full-length NS2, as well as peptides corresponding to the NS2 NES, can functionally replace the HIV-1 Rev effector domain (18-20), which promotes the nuclear export of unspliced HIV-1 RNA (21). As in case of Rev, the integrity of the NS2 NES is essential for nuclear export of the protein, but not for interaction of NS2 with the soluble export receptor Crm1 (19, 22, 23). The functional importance of the recognition of NS2 by the nuclear export machinery is currently the subject of intense discussion. NS2 was found to be essential for export of progeny vRNPs in some studies (18, 23), and was also found indispensable for the generation of functional virus-like particles in a plasmid transfection system (19, 24, 25). In addition, the C terminus of NS2 has been shown to interact with M1 both in vivo and in vitro (26, 27). Based on these observations, it has been proposed that NS2 mediates the export of vRNPs by acting as an adapter protein between the nuclear export machinery and the M1·vRNP complex (18, 19, 23).

Rev-like NESs, as found in NS2, are thought to be crucial for stabilization of the ternary RanGTP·NES·Crm1 complex that is responsible for export of transport substrates by this Crm1-dependent pathway (22, 28). While the interactions involved in signal-mediated nuclear import have been the subject of intense study in recent years, the molecular processes involved in the assembly of this ternary export complex, as well as in the specific recognition of the export substrates by the nuclear transport machinery, remain ill defined. Structural studies of the nuclear export complex itself and of its principal constituents in solution will therefore provide valuable insights into the molecular mechanisms involved in the nuclear export process.

This study focuses on the structural characteristics of NS2 that allow it to interact with its ligands. Our results indicate that the nuclear export domain of isolated NS2 exists in an exposed and highly flexible conformation in solution. The C terminus of the protein, in contrary, adopts a relatively rigid structure, but lacks a tightly packed hydrophobic core under near physiological conditions. It is proposed that the high degree of flexibility observed in the nuclear export domain of NS2 is associated with the recognition of substrate proteins by the nuclear export machinery.

	EXPERIMENTAL PROCEDURES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Expression and Purification of Recombinant NS2-- The NS2 cDNA of influenza strain InfA/PR/8/34 was subcloned into a pET21b expression vector (Novagen). This construct was designed to allow expression of a full-length version of NS2 that carried no tags or additional residues. The correct NS2 sequence in the resulting pET21bNS2 plasmid was confirmed by DNA sequencing.

For protein expression, pET21bNS2 was transformed into Escherichia coli strain BL21(DE3) (Novagen), and the cells were grown to an A₆₀₀ of 0.8-1.0 at 37 °C in Luria-Bertani medium containing 200 µg/ml ampicillin. At this point, protein expression was induced by addition of 0.5 mM isopropyl--D-thiogalactopyranoside. After incubation for 20 h at 18 °C, the cells were harvested by centrifugation and resuspended in lysis buffer (50 mM NaCl, 20 mM Tris, 1 mM EDTA, pH 8.5). The cells were broken by sonication using a 550 Sonic Dismembrator (Fisher Scientific), and insoluble material was removed by centrifugation for 20 min at 10,000 rpm in a JA-20 rotor (Beckman).

NS2 was captured by gel filtration on a Sephacryl-S300 26/60 HR column (Amersham Biosciences, Inc.) from the soluble fraction of the cell lysate. The separation was performed using elution buffer A (150 mM NaCl, 20 mM Tris, 1 mM EDTA, pH 8.5), and NS2 was found to elute as a single peak at about 239 ml as confirmed by SDS-PAGE and Coomassie Brilliant Blue staining. The NS2 containing fractions were pooled and concentrated by ultrafiltration through a YM3 filter (Amicon). Further purification of NS2 was accomplished by gel filtration on a Sephadex-75 16/60 HR column (Amersham Biosciences, Inc.) using elution buffer B (150 mM NaCl, 20 mM HEPES, 1 mM EDTA, pH 7.5). Again, NS2 was found to form a single peak that eluted at about 78 ml, and the appropriate fractions were pooled and the buffer exchanged to 20 mM HEPES, 1 mM EDTA, pH 7.0, by ultrafiltration and dilution. Subsequently, NS2 was passed through a Hitrap SP column (Amersham Biosciences, Inc.), and bound proteins were eluted with 1 M NaCl, 20 mM HEPES, 1 mM EDTA, pH 7.0. While contaminant proteins were bound to the column, NS2 remained in the flow through.

Protein quantitation was performed by absorbance at 280 nm, using an extinction coefficient of 12,660 M¹ cm¹ calculated from the amino acid composition of the protein (29). Following purification, NS2 was dialyzed against protein storage buffer (150 mM NaCl, 20 mM HEPES, 1 mM EDTA, pH 7.5), concentrated to about 1 mg/ml by ultrafiltration, and transferred to 4 °C. Further buffer exchange and protein concentration was accomplished by dialysis and ultrafiltration as necessary.

Purity and Identity of Bacterially Expressed NS2-- Denaturing isoelectric focusing (IEF) of purified NS2 was performed using the PhastSystem^TM automated electrophoresis system (Amersham Biosciences, Inc.). IEF was carried out utilizing a dry IEF Phastgel^TM rehydrated with 6.25% Pharmalyte^TM 4-6.5 (Amersham Biosciences, Inc.) in 8 M urea, and protein detection was accomplished by staining with Coomassie Brilliant Blue. Before separation, NS2 samples were incubated for 2.5 h in protein storage buffer + 8 M urea. To assure completion of the focusing process, all samples were applied at both the cationic and the anionic end of the gel.

Matrix-assisted laser desorption ionization-mass spectrometry of purified NS2 was performed at the UAB Mass Spectrometry Core Facility on a Voyager Elite matrix-assisted laser desorption ionization-time of flight instrument (PerSeptive Biosystems) with delayed extraction. Internal calibration with reference proteins thioredoxin (11674.5 Da) and apomyoglobin (16952.5 Da) was used for molecular mass determination. For N-terminal sequencing, NS2 was electroblotted onto a polyvinylidiene difluoride membrane (Bio-Rad) and Automatic Edman degradation (6 cycles) of the NS2 band was carried out at the UAB Protein Sequencing Core Facility using a Model PI 2090E (Beckman) sequencer.

Near ultraviolet absorption spectroscopy of NS2 was performed on 0.5 mg/ml NS2 in protein storage buffer using a 1-cm cell. Experiments were carried out at 25 °C utilizing a Beckman DU 650 spectrophotometer, and the NS2 UV absorbance spectra were corrected by subtraction of the buffer baseline.

The ability of recombinant NS2 to associate with influenza virus matrix protein M1 of both viral and bacterial origin was assayed by far-Western blot similar as described in Ref. 26. Briefly, purified InfA/PR/8/34 virus was prepared as described in Ref. 30. Recombinant, hexahistidine-tagged InfA/PR/8/34 M1 protein was produced in E. coli using the pET21b expression vector. Protein expression was accomplished by growing BL21(DE3)pET21b M1 cells for 16 h at 37 °C in the absence of isopropyl--D-thiogalactopyranoside. Subsequently, the cells were harvested, resuspended in 0.5 M NaCl, 20 mM Tris, 5 mM imidazole, pH 7.9, and lysed by sonication as described above. Purification of M1 was achieved by incubation of the cell lysate supernatant with chelating Sepharose (Amersham Bioscience Inc.), and elution was performed according to the manufacturers instructions. Detection of M1 bound NS2 was accomplished utilizing monospecific, mouse anti-NS2 antibodies that had been raised against a hexahistidine-tagged, affinity-purified version of bacterially expressed NS2.²

Analytical Ultracentrifugation-- Sedimentation equilibrium ultracentrifugation of NS2 was performed at the UAB Analytical Ultracentrifuge Analysis Facility on a Beckman XL-A Analytical Ultracentrifuge using an AN-60 Ti rotor (Beckman). NS2 samples at a concentration of 0.3, 0.7, 1.0, and 1.3 mg/ml NS2 in protein storage buffer were stored for 1 week at 4 °C before analysis. Subsequently, double sector cells were loaded with 120 µl of protein sample or protein storage buffer, respectively, and sedimentation equilibrium data were obtained at 29,000, 34,000, and 48,000 rpm at 20 °C. Protein detection was carried out by absorbance at 280 nm after equilibrium was reached. The equilibrium data was edited by deletion of absorbance changes at the protein and reference meniscus interface. In addition, absorbance values greater than 1.5 at the steepest part of each absorbance versus radial position exponential were deleted to ensure that only data obeying Beer's law were used in the nonlinear regression analysis. The partial specific volume of NS2 was calculated as 0.7367 cm³/g from its amino acid sequence. For modeling sedimentation equilibrium data, the data were merged, and a global nonlinear least squares analysis of the edited data sets was performed using the software packages XL-A v4.0 (Beckman) and Origin v4.1 (Microcal). Fitting the experimental data to both a single ideal species model and to a monomer-dimer self-association model was attempted. In addition, whole cell apparent weight average molecular masses were determined from each individual data set as well.

Analytical Gel Filtration Chromatography-- Analytical gel filtration of purified NS2 was carried out on a Sephadex-75 16/60 HR gel filtration column (Amersham Biosciences, Inc.). NS2 at a concentration of 0.5 mg/ml in protein storage buffer was loaded onto the column, and the protein was eluted with the same buffer. If the gel permeation properties of NS2 were determined under denaturing conditions, NS2 was incubated for 4.5 h in protein storage buffer + 8 M urea prior to elution with this buffer. Protein elution was followed by absorption at 260 and 280 nm, and the elution profiles were corrected by subtraction of the appropriate buffer baselines. Purity and homogeneity of the observed NS2 peaks were confirmed by SDS-PAGE and by observation of the A₂₈₀/A₂₆₀ absorbance ratio over the NS2 peak as described in Ref. 31. For determination of the Stokes radius, the column was calibrated using the Amersham Biosciences Inc. gel filtration low molecular weight calibration kit. Blue dextran 2000 (Amersham Biosciences, Inc.) was used to determine the void volume of the gel filtration column both in the presence and absence of 8 M urea. The frictional ratio f/f₀ of NS2 was calculated according to Ref. 32 using the following equation,

(Eq. 1)

M is the molecular weight,  designates the Stokes radius,  is the partial specific volume, and N is Avogadro's number.

Circular Dichroism (CD) Spectroscopy-- CD spectroscopy of NS2 was carried out on an Aviv model 62DS CD spectrometer equipped with a thermostated cell holder. Far-UV CD spectra of NS2 were recorded at 25 °C on 0.8 mg/ml NS2 in 20 mM borate/boric acid, pH 7.5, using a 0.01-mm cell. The data were averaged over 2 repeat scans and corrected by subtraction of the buffer baseline. The secondary structure content of NS2 was estimated by analysis of the normalized NS2 spectrum (188-240 nm) as a combination of 3 reference spectra (-helix, -sheet, random coil) using the program PROSEC (Aviv Associates). To verify the results, a synthetic spectrum with the same structural components as determined for NS2 was created using the program BLEND (Aviv Associates) and compared with the original NS2 spectrum. Near-UV CD spectroscopy of NS2 was carried out at 25 °C using a 1-cm cell. Data was acquired on 1.0 mg/ml NS2 in protein storage buffer, or on 1.0 mg/ml NS2 in protein storage buffer containing 6 M guanidinium chloride (GdmCl). Each data set was averaged over 5 scans, and the spectra were corrected by subtraction of the appropriate buffer baselines.

Fourier Transform-Infrared (FT-IR) Spectroscopy-- Infrared absorption spectra of 10.0 mg/ml NS2 in protein storage buffer were acquired at room temperature (23.1 °C) using a PROTA FT-IR instrument (Bomen). The data were averaged over 400 scans, corrected by subtraction of the air, buffer, and cell baselines, and normalized. The secondary structure content of NS2 was determined by analysis of the NS2 amide I band as a linear combination of the spectra of 32 reference proteins of known x-ray crystallographic structure using the PROTA software (Bomen).

Fluorescence Spectroscopy-- Fluorescence spectroscopy of NS2 was performed on a SLM 8000C spectrofluorometer equipped with a thermostated cell holder. All experiments were carried out at 25 °C using a 1-cm cell, and samples were stirred during data acquisition to minimize artifacts due to photochemical degradation. After excitation at 290 nm, protein fluorescence emission spectra were collected on 0.026 mg/ml NS2 (A₂₈₀ = 0.05) in protein storage buffer, or in protein storage buffer containing 6 M GdmCl, respectively. Due to rapid photodegradation of NS2, fluorescence quenching experiments were carried out using individual NS2 samples containing 0-600 mM acrylamide. 4,4-Dianilino-1,1-binaphthyl-5,5-disulphonic acid (ANS) binding studies were performed at an excitation wavelength of 370 nm using a molar NS2/ANS ratio of 1:3. Spectra were corrected for buffer fluorescence and for variation of light source intensity and detector response versus wavelength (33). Analysis of NS2 quenching data was carried out according to Ref. 34 using the following modified form of the Stern-Volmer equation (35),

(Eq. 2)

where [Q] designates the concentration of the quenching agent and F₀ and F are the fluorescence intensities at wavelength  in the absence and presence of the quencher, respectively. While V characterizes the static component of the quenching process, K_SV describes its collisional component. The effective quenching constant K_SV(eff) was obtained from the initial slope of the Stern-Volmer plot and normalized for increase of the solution viscosity due to GdmCl as described in Ref. 34.

Conformational Stability of NS2-- Thermal denaturation of NS2 was monitored by CD spectroscopy at 220 nm. 0.026 mg/ml NS2 in 20 mM borate/boric acid, pH 7.5, in a 1-cm cell was heated under stirring from 10 to 90 °C using increments of 0.5 °C. The temperature inside the sample cell was monitored by directly inserting a thermistor into the cell. To assure reversibility of the denaturation process, the sample was allowed to cool to room temperature after the experiment was complete, and the far-UV CD spectrum of the renatured protein was compared with the spectrum of NS2 collected before the heating experiment.

Differential scanning calorimetry of NS2 was carried out on Model 6100 Nano II Differential Scanning calorimeter (Calorimetry Sciences Corp.) equipped with a capillary cell. Freshly purified NS2 at a concentration of 0.4, 0.7, and 1.0 mg/ml in 20 mM borate/boric acid, pH 7.5, and/or in 150 mM NaCl, 20 mM HEPES, 1 mM EDTA, pH 7.5, was heated from 5-100 or 5-80 °C, respectively, using heating rates between 0.5 and 2 °C/min. The denaturation data was corrected by subtraction of the buffer baseline and converted to partial molar heat capacity using the program CpCalc (Calorimetry Sciences Corp.). Reversibility of the denaturation process was confirmed by subsequent cooling of the sample to 5 °C, followed by additional rounds of heating. The partial heat capacities of the fully unfolded polypeptide were estimated according to Refs. 36 and 37.

Solvent-induced unfolding of NS2 was carried out by addition of GdmCl and was followed by fluorescence spectroscopy and far-UV CD spectroscopy. When unfolding was analyzed by fluorescence spectroscopy, 0.026 mg/ml NS2 was incubated in protein storage buffer containing 0-6 M GdmCl for 5 min at room temperature. Subsequently, fluorescence spectra were collected as described above using an excitation wavelength of 273 nm, as at this wavelength the folded and unfolded forms of NS2 showed no difference in UV absorbance. When NS2 unfolding was followed by far-UV CD spectroscopy, 0.26 mg/ml NS2 was incubated in 150 mM NaCl, 20 mM borate/boric acid, 0-6 M GdmCl, pH 7.5, for the same amount of time. CD spectra were collected at 25 °C using a 2-mm cell, and the data was corrected as described above. Completeness of the unfolding process was assured by examination of fluorescence and CD data of NS2 samples that had been exposed to GdmCl for different periods of time (5-90 min). Reversibility of the unfolding process was tested by examination of CD or fluorescence spectra of NS2 samples that had been unfolded by exposure to 6 M GdmCl and had subsequently been refolded by 1:10 (v/v) dilution in GdmCl-free buffer. The solvent-induced unfolding data of NS2 was analyzed assuming a two-state mechanism of unfolding. The equilibrium constant K and the Gibbs free energy change G of the unfolding reaction were calculated according to the following.

(Eq. 3)

(Eq. 4)

Here f_F and f_U are the fractions of folded and unfolded protein, respectively. y is the observed value at any point during the denaturation process. y_F and y_U are the values of y characteristic of the folded and unfolded state of NS2 and were obtained by extrapolation of the pre- and post-transition regions of the NS2 denaturation curves. R and T denote the gas constant and the absolute temperature, respectively. G(H₂O), the conformational stability of the protein in the absence of denaturant, was obtained by least squares analysis of the linear region of the G versus [GdmCl] plot as described in Ref. 38.

Protease Mapping of NS2-- Limited proteolysis experiments of NS2 were performed at 37 °C using immobilized trypsin (Sigma) and immobilized Staphylococcus aureus V8 protease (Pierce). Trypsin digestions were carried out for 0.5-16 h using 0.01-0.4 units of trypsin/mg of NS2 in 150 mM NaCl, 50 mM HEPES, pH 8.0. V8 protease digestions were performed for 1-24 h using 0.5-4 units of V8 protease/mg of NS2 in 150 mM NaCl, 20 mM HEPES, pH 7.8. Reactions were stopped by filtration through 0.2-µm Millipore Ultrafree centrifugal filtration units and analyzed by 13.5% SDS-PAGE and staining with Coomassie Brilliant Blue. N terminus sequencing of the digestion products was carried out as described for full-length NS2. A C-terminal hexahistidine-tagged version of NS2² that yielded the same digestion products as untagged NS2 was used to determine whether the identified NS2 fragments retained the same C terminus as the undigested protein. For this purpose, full-length, hexahistidine-tagged NS2 and its digestion fragments were blotted onto nitrocellulose, and retention of the C-terminal hexahistidine tag was analyzed by Western blot using monoclonal anti-His antibodies (Qiagen) and the monoclonal anti-NS2 antibodies described above.

	RESULTS

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Isolated, Recombinant NS2 is Pure and Active-- InfA/PR/8/34 NS2 protein was expressed in E. coli and purified from the soluble fraction of the bacterial lysate by a combination of size exclusion and cation exchange chromatography. The purity, identity, and homogeneity of isolated NS2 were analyzed using a number of different techniques. SDS-PAGE produced a single band that migrated with an apparent molecular weight of about 11,000, as observed for the viral protein (39). If the NS2 preparation was subjected to denaturing IEF, only one band was observed as well. Matrix-assisted laser desorption ionization-time of flight mass spectrometry revealed a single species of 14,382.8, which is consistent with the molecular weight of NS2 calculated from the amino acid sequence of the native protein (14,380.3) (40). N-terminal sequencing of the purified protein yielded the sequence MDPNT as expected for InfA/PR/8/34 NS2. The near-UV absorption spectrum of NS2 showed only marginal absorbance above 300 nm, confirming that the NS2 preparation was free of aggregates. The A_{260 nm}/A_{280 nm} ratio of the protein preparation was 0.69, indicating that it contained less than 1% nucleic acids (41, 42). Furthermore, recombinant NS2 was found to associate with influenza virus matrix protein M1 in a far Western blot assay that was carried out as described in Refs. 26 and 27 (data not shown). Based on these observations, isolated, recombinant NS2 was judged chemically pure and active with respect to binding to influenza virus matrix protein M1.

NS2 Exists as a Compact Monomer in Solution-- The aggregation state and gross conformation of isolated NS2 were probed by a combination of hydrodynamic techniques. Sedimentation equilibrium ultracentrifugation allows the determination of the molecular weight of intact proteins, independent of frictional effects due to molecule shape or hydration (43). NS2 was subjected to ultracentrifugation at protein concentrations between 0.3 and 1.4 mg/ml using a range of different centrifugation speeds (Fig. 1A). The average apparent molecular mass of NS2 was determined as 14,153 Da (with 95% confidence limits of 14,126 Da and 14,179 Da) by global analysis of all data sets assuming a single, non-interacting species. Upon separate examination of the individual data sets, no significant change of the apparent molecular mass of NS2 was observed with increased protein concentration, nor was significant non-ideality detected with increased rotor speed.

View larger version (11K): [in this window] [in a new window]   Fig. 1.   The aggregation state and shape of isolated NS2. A, analysis of NS2 by sedimentation equilibrium ultracentrifugation. NS2 at a concentration of 0.4-1.3 mg/ml in protein storage buffer (150 mM NaCl, 20 mM HEPES, 1 mM EDTA, pH 7.5) was stored at 4 °C for 1 week, and then analyzed at multiple centrifugation speeds. All data was fitted globally to a single, non-interacting species model. The sedimentation equilibrium profile shown was obtained using a protein concentration of 0.7 mg/ml and a centrifugation speed of 29,000 rpm, and is representative of all data sets. The solid line represents the global fit. B, analysis of NS2 by analytical gel filtration. Estimation of the Stokes radius of NS2 was carried out by gel filtration using a protein concentration of 0.5 mg/ml in protein storage buffer or protein storage buffer + 8 M urea. The protein standards used were: 1) ribonuclease A (13.7 kDa); 2) chymotrypsinogen A (25.0 kDa); 3) ovalbumin (43.0 kDa); and 4) albumin (67.0 kDa). The Kav value represents the normalized elution behavior of the solutes.

In contrast to sedimentation equilibrium ultracentrifugation, analytical gel filtration is sensitive to the Stokes radius of a particle, rather than to its molecular weight (32). This parameter can be used to obtain the frictional ratio of a protein, which is a measure of its departure from a smooth, unhydrated sphere of equal molecular weight. When NS2 was analyzed by analytical gel filtration under near physiological conditions, it was found to elute in a single, homogeneous peak. Comparison of the elution volume of NS2 with the elution volumes of several globular reference proteins estimated the Stokes radius of NS2 as 16.6 Å (Fig. 1B). This value was found to correspond to a molecular weight of about 15,000, consistent with the molecular weight of NS2 determined by analytical ultracentrifugation. The frictional ratio of NS2 was calculated as 1.03, a ratio typical of compact proteins (32). In the presence of 8 M urea, a dramatic shift in the elution volume of NS2 was observed, which eluted with a Stokes radius of 35.6 Å and a frictional ratio of 2.21. In contrast, the void volume of the gel filtration column did not change significantly (44.55 versus 43.47 ml in the absence of urea), indicating that the permeation properties of the gel filtration matrix did not change under these conditions. These results indicate that isolated NS2 exists as a globular, non-interacting monomer under near physiological conditions, and adopts an extended conformation only after solvent-induced denaturation.

Nonetheless, in NS2 preparations that had been stored for extended periods of time (several weeks), a small amount (10% of total protein) of NS2 dimer could be found by SDS-PAGE and gel filtration similar as reported by Ward et al. (27) for a glutathione S-transferase fusion form of NS2. This interaction was not found in fresh samples (1-2 weeks old) and usually coincides with the appearance of small amounts of irreversibly precipitated protein.

NS2 Adopts a Folded, but Highly Flexible Conformation-- The structure and dynamics of NS2 in solution were studied using different methods of optical spectroscopy. The far-UV CD spectrum of a protein is dominated by the contribution of its peptide bonds, and can be described as a superposition of spectra distinctive of the different types of regular secondary structure present in the polypeptide (44). Purified NS2 was probed by far-UV CD spectroscopy under a variety of buffer conditions to obtain an estimate of its secondary structure composition. At near physiological conditions, the CD spectrum observed for NS2 displays two minima centered at 208 and 222 nm and is typical of a folded protein containing a considerable amount of -helical structure (45) (Fig. 2A). No significant changes of the spectrum were observed if the pH of the solution was varied between 6.0 and 11.0, whereas at pH values close to the calculated isoelectric point of NS2 (pI 5.1) reversible precipitation of the protein occurred. Likewise, variation of the ionic strength of the solution between 0 and 2 M NaCl did not lead to significant changes in the CD spectrum. Quantitative determination of the secondary structure content of NS2 was carried out as described in Ref. 45 using CD data between 188 and 240 nm recorded at 25 °C. The secondary structure of NS2 was found to consist of 57.8% -helix, 15.9% -sheet, and 26.3% reverse turn or random coil. Furthermore, the secondary structure of NS2 was probed by FT-IR spectroscopy, a technique which relies on the amide bands observed in the infrared absorption spectrum of polypeptides. These bands are formed by overlapping absorption bands corresponding to the different types of secondary structure and can be analyzed as a combination of the spectra of reference proteins of known crystallographic structure (46, 47). The infrared absorption spectrum of NS2 displays a strong amide I band centered at 1653 cm¹, as expected for a predominantly -helical conformation (46) (Fig. 2B). From this data, the secondary structure elements of NS2 were estimated as 42.6% -helix, 17.2% -sheet, 17.0% reverse turns, and 24.7% random coil using the PROTA procedure (Bomen). These results indicate that isolated NS2 contains a significant amount of secondary structure in solution. The agreement between the estimates of the secondary structure composition of NS2 obtained by CD and FT-IR spectroscopy can be considered satisfactory in light of the differences between the basic spectra libraries utilized by both methods.

View larger version (14K): [in this window] [in a new window]   Fig. 2.   Secondary structure content of NS2. A, far-UV circular dichroism spectra of NS2. Solid line, NS2 in 150 mM NaCl, 20 mM borate/boric acid, pH 7.5, 25 °C. +++, NS2 in 150 mM NaCl, 20 mM borate/boric acid, pH 7.5, 80 °C. , NS2 in 150 mM NaCl, 20 mM borate/boric acid, 6 M guanidinium chloride, pH 7.5, 25 °C. []MRW, mean molar ellipticity. B, infrared absorbance spectrum of NS2 in 150 mM NaCl, 20 mM HEPES, 1 mM EDTA, pH 7.5, recorded at room temperature (23.1 °C).

Aromatic groups that are immobilized in an asymmetric environment in well structured proteins produce highly characteristic bands in the near-UV region of the CD spectrum that result from the interaction of the aromatic side chains with neighboring groups (48). The near-UV CD spectrum of NS2, which contains two tryptophans and one tyrosine, but no intrinsically active disulfide bonds, was recorded to probe the stability of its tertiary structure. Interestingly, the near-UV CD spectrum obtained at near physiological buffer conditions was virtually zero, and resembled the spectrum of the protein in the presence of 6 M GdmCl (Fig. 3). These observations suggest that under native buffer conditions, the aromatic groups of NS2 are able to rotate freely and undergo no strong interactions with other groups within NS2 (48). To further analyze the tightness and stability of the tertiary structure of NS2, the protein was probed by tryptophan fluorescence spectroscopy. The wavelength at which the maximum fluorescence emission of tryptophan is observed depends on the polarity of its immediate environment, and, in native proteins, is usually significantly blue-shifted compared with the emission maximum of free tryptophan in solution (44). Furthermore, the efficiency and pattern with which the tryptophan fluorescence of a protein is quenched by the collisional quencher acrylamide provides a measure of the dynamics of the protein, and of the exposure its tryptophan residues to the solvent (34, 50). When the tryptophan fluorescence emission spectrum of NS2 was recorded under near physiological buffer conditions, its emission maximum was found to exhibit only a small blue-shift compared with its emission maximum under denaturing conditions. Addition of acrylamide led to rapid quenching of the tryptophan fluorescence, comparable to tryptophan quenching under denaturing conditions (Table I). The Stern-Volmer plot of the quenching reaction showed upward curvature, indicating that both tryptophan residues of NS2 were nearly equally assessable to the solvent (Fig. 4) (34). In conjunction with the observations made by near-UV CD spectroscopy, these results suggest a dynamic, loosely packed tertiary structure in the vicinity of the aromatic residues of NS2.

View larger version (16K): [in this window] [in a new window]   Fig. 3.   Near-UV circular dichroism spectra of NS2. Spectra were recorded at 25 °C using a protein concentration of 1.0 mg/ml NS2 in a 1-cm cell. , NS2 in 150 mM NaCl, 20 mM HEPES, 1 mM EDTA, pH 7.5. , NS2 in 150 mM NaCl, 20 mM HEPES, 1 mM EDTA, 6 M guanidinium chloride, pH 7.5.

View larger version (15K): [in this window] [in a new window]   Fig. 4.   Acrylamide quenching of the intrinsic fluorescence of NS2. Quenching experiments were carried out at 25 °C using an excitation wavelength of 290 nm. , NS2 in 150 mM NaCl, 20 mM HEPES, 1 mM EDTA, pH 7.5. , NS2 in 150 mM NaCl, 20 mM HEPES, 1 mM EDTA, 6 M guanidinium chloride, pH 7.5.

To determine whether the flexibility observed by near-UV CD and tryptophan fluorescence spectroscopy is due to local disorder in the proximity of the aromatic groups, or if it reflects a more general property of the NS2 structure, binding experiments with the hydrophobic dye ANS were carried out. This probe strongly associates with hydrophobic clusters in compact proteins that possess pronounced secondary structure but lack a tightly packed tertiary structure. In contrast, coil-like proteins, -helical and -sheet structural homopolypeptides as well as proteins containing a rigid hydrophobic core show only very little affinity for ANS (51). In the presence of NS2, the fluorescence intensity of ANS was found to increase 7-fold, and its emission maximum was significantly blue-shifted compared with the emission maximum of ANS in the absence of the protein, pointing to a strong interaction of the probe with NS2 (Table I) (51). In contrast, if NS2 was added to ANS in the presence of 6 M GdmCl, no such interaction was observed. These results suggest that NS2 possesses a loosely packed hydrophobic core that is accessible to the solvent, as expected for a folded protein with a dynamic tertiary structure.

Denaturation by heat or chemical denaturants can also be used to probe the structure and conformation of proteins in solution. The temperature-induced denaturation of small globular proteins is commonly a two-state transition, even so the temperature-denatured state of these proteins is often compact, and typically retains a substantial amount of secondary structure (52). When thermal unfolding of NS2 was followed by far-UV CD spectroscopy, a broad, nonlinear transition was observed that was highly reversible. The end point of this transition was reached at ~80 °C, at which point NS2 was found to retain about 50% of its secondary structure (Figs. 2A and 5A). To further explore the character of this transition, the enthalpy change associated with temperature-induced denaturation of NS2 was measured directly using differential scanning calorimetry. Temperature-induced denaturation of small, native proteins is typically accompanied by a single, distinct heat absorption peak of several J × K¹ × g¹ (53). Surprisingly, the partial specific heat capacity function of NS2 was found to increase monotonously with increasing temperature, and was devoid of any excess heat absorption (Fig. 5B). At 25 °C, the partial specific heat capacity of NS2 of 1.65 J × K¹ × g¹ was significantly lower than the partial specific heat capacity expected for the fully unfolded protein (2.06 J × K¹ × g¹), and was in a range typically observed for native proteins of similar size (37). At 75 °C, the partial specific heat capacity of NS2 (2.06 J × K¹ × g¹) approached the value expected for the fully unfolded polypeptide chain (2.23 J × K¹ ×g¹), but did not quite reach it. These results indicate that NS2 assumes a compact conformation at room temperature, and partially unfolds upon heating to 80 °C. As the increase in heat capacity upon protein denaturation is believed to be due to the exposure of nonpolar groups to the solvent (54), the lack of excess heat absorption upon denaturation of NS2 indicates that isolated NS2 possesses an only loosely packed hydrophobic core.

View larger version (12K): [in this window] [in a new window]   Fig. 5.   Thermal denaturation of NS2. A, thermal denaturation of NS2 as monitored by far-UV circular dichroism spectroscopy at 220 nm. 0.026 mg/ml NS2 in 20 mM borate/boric acid, pH 7.5, was heated under stirring from 10 to 90 °C using increments of 0.5 °C/min. []MRW, mean molar ellipticity. B, thermal denaturation of NS2 as monitored by differential scanning calorimetry. 1.0 mg/ml NS2 in 150 mM NaCl, 20 mM HEPES, 1 mM EDTA, pH 7.5, was heated from 5 to 80 °C using a scan rate of 1 °C/min.

In contrast to temperature-induced denaturation, solvent-induced unfolding of native proteins by high concentrations of denaturants is accompanied not only by the loss of the tertiary structure of the protein, but leads to almost complete unfolding and loss of the compact protein conformation (52). When the unfolding of NS2 by GdmCl was followed by far-UV CD and tryptophan fluorescence spectroscopy, two nearly identical, highly cooperative unfolding curves were obtained (Fig. 6). As expected, these results indicate that the unfolding of NS2 is dominated by the unfolding of its secondary structure. Assuming a two-state mechanism of unfolding, the conformational stability G(H₂0) of NS2 was estimated as about 6 kcal/mol as described in Ref. 38 (Table II). This value is at the lower end of the range of stabilities commonly observed for solvent-induced unfolding of globular proteins (G(H₂0)  5-10 kcal/mol) (55). The m value, which represents the dependence of the free energy of unfolding on the denaturant concentration, was found to be roughly proportional to the molecular weight of NS2, indicating that the solvent-induced unfolding transition of this protein indeed follows a two-state mechanism (56, 57).

View larger version (14K): [in this window] [in a new window]   Fig. 6.   Solvent-induced unfolding of NS2. NS2 was incubated for 5 min in 150 mM NaCl, 20 mM HEPES, 1 mM EDTA, 0-6 M guanidinium chloride, pH 7.5, or in 150 mM NaCl, 20 mM borate/boric acid, 0-6 M guanidinium chloride, pH 7.5, respectively. , NS2 unfolding monitored by circular dichroism spectroscopy at 220 nm. , NS2 unfolding monitored by fluorescence emission at 342 nm.

Protease Mapping Reveals a Bipartite Structure of NS2-- To gain further insight into the structure of NS2, protein digestion experiments using trypsin and V8 protease were carried out. Limited proteolysis of a folded protein is thought to be dominated by the stereochemistry and the dynamics of the protein conformation, and has been used successfully to probe the structure of native as well as partly folded proteins (58). Both trypsin and V8 protease are highly specific and possess a large number of theoretical cut sites throughout the primary sequence of NS2, making them suitable probes to study the structure of this protein. When NS2 was digested using a range of trypsin:NS2 ratios, one major digestion fragment of about 6 kDa was observed by SDS-PAGE. In addition, a minor band of slightly smaller size was obtained. Similarly, if NS2 was digested using V8 protease, only one digestion fragment was observed that migrated with an apparent molecular weight of 6,000-7,000 on SDS-PAGE (Fig. 7). The major protease cut sites of NS2 were mapped to residues Glu-47 and Arg-51 by N terminus sequencing of the V8 digestion fragment and the major trypsin digestion fragment, respectively. In addition, N terminus sequencing of the minor trypsin digestion fragment revealed a second, less assessable trypsin cleavage site at position Arg-61. A hexahistidine-tagged version of NS2² showed the same cut sites as conventionally purified NS2, and was probed for retention of the C-terminal hexahistidine tag by Western blot. All fragments were found to contain hexahistidine tags indicating that they retained the C terminus of the full-length protein (data not shown). The most N-terminal digestion sites that were fully protected in the NS2 fragments were Glu-63 and Lys-64, while Glu-112 and Arg-114 were the most C-terminal digestion sites that were not assessable to their proteases. These results indicate that residues 63-115 of NS2 form the minimal core of the protease-resistant region of NS2. No fragments corresponding to the N terminus of NS2 were observed, even if very low quantities of proteases were employed and very short digestion times were used. As limited digestion of globular proteins typically yields nicked species that retain their overall fold under nondenaturing conditions (58), these results suggests a highly exposed and flexible structure for the N terminus of NS2. The C terminus of the protein, in contrary, assumes a comparatively protease-resistant conformation.

View larger version (36K): [in this window] [in a new window]   Fig. 7.   Protease mapping of NS2. Trypsin digestions of NS2 were carried out at 37 °C using 0.4 units of immobilized trypsin/mg NS2 in 150 mM NaCl, 50 mM HEPES, pH 8.0. V8 protease digestions of NS2 were performed using 4.0 units of immobilized V8 protease/mg of NS2 in 150 mM NaCl, 20 mM HEPES, pH 7.8. Reactions were stopped by filtration through 0.2-µm filtration units, and the samples were analyzed by 13.5% SDS-PAGE and staining with Coomassie Brilliant Blue.

	DISCUSSION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

In this study we have characterized the structure of isolated influenza virus protein NS2 in solution using a combination of biochemical and biophysical techniques. For this purpose, NS2 was expressed in E. coli and purified to homogeneity from the soluble fraction of the cell lysate using conventional techniques. The recombinant protein was capable of interacting with influenza virus protein M1, as found for both the viral protein, and for NS2 produced in infected cells (26, 27). In addition, recombinant NS2 was quite soluble and stable in a variety of buffers for extended periods of time, and did not exhibit the extensive aggregation and precipitation usually observed for misfolded proteins. While one study reported partial phosphorylation of NS2 produced in infected cells (59), another group found no phosphorylation of NS2 in their system (20). The structure of recombinant NS2 was therefore considered representative of at least one form of NS2 present in vivo.

If the structure and aggregation state of NS2 was probed by analytical gel filtration and analytical ultracentrifugation, isolated NS2 was found to exist as a globular, non-interacting monomer. While far-UV CD and FT-IR spectroscopy detected the presence of a significant amount of secondary structure in NS2, near-UV CD and fluorescence spectroscopy revealed that the aromatic residues of the protein were highly assessable to the solvent, and did not undergo strong interactions with neighboring groups. Furthermore, isolated NS2 strongly interacted with the hydrophobic dye ANS suggesting that hydrophobic clusters within the structure of the protein were assessable to the solvent as well (51). While GdmCl-induced unfolding of NS2 was characterized by a highly cooperative transition, no distinct heat absorption peak was associated with temperature-induced denaturation of the protein as determined by differential scanning calorimetry. Taken together, these results suggest the presence of a fluid-like hydrophobic core in NS2 that lacks rigid packing interactions. Nonetheless, the nonlinear transition observed if thermal unfolding of the protein was monitored by far-UV CD spectroscopy may point to the presence of some areas of more tightly packed structure within NS2, whose enthalpy change may be too small to be detected.

The flexibility observed in NS2 was intriguing, as with a length of 121 residues the protein is by far large enough to fold independently. The unusual set of characteristics found in the structure of NS2 (compactness, presence of a pronounced secondary structure, but absence of a tightly packed tertiary structure) are attributes typically connected with the molten globule state of proteins (52, 60). While this state had first been identified as a kinetic intermediate in protein folding, since then it has also been found under equilibrium conditions (usually at acidic pH or in the presence of alcohol) in a number of proteins. While the "classical" molten globule model does not contain strong tertiary interactions, it is thought to retain a native-like tertiary fold and a loosely packed hydrophobic core (52, 60). Furthermore, a relatively tightly packed subdomain has since been found in the molten globule state of a number of proteins that has been proposed to be responsible for the two-state transition observed during solvent-induced unfolding of the molten globule state (61-65). Protease mapping of NS2 suggests that the flexibility observed in this protein is not distributed equally over the whole molecule as well, but that its N terminus assumes a significantly more flexible and solvent exposed conformation than its C terminus, which adopts a comparatively protease-resistant structure. Interestingly, both tryptophan residues of NS2 are located within this C-terminal domain and both Trp-65 and Trp-78 are surrounded by protease cut sites that appear inaccessible to their the proteases (Glu-63, Lys-64, Arg-66, Glu-67, Glu-74, Glu-75, Arg-77, Glu-81, and Glu-82). Nonetheless, the tertiary structure of NS2 in the vicinity of these aromatic groups was found to be only loosely packed by near-UV CD and tryptophan fluorescence spectroscopy, indicating that the C terminus of NS2, while more structured than its N terminus, does not represent a rigidly packed domain as well. In conclusion, the structure of NS2 appears to assume all the characteristics typically described for the molten globule state. But in contrast to most proteins that are known to be capable of assuming this conformational state, NS2 adopts its molten globule-like structure under near physiological conditions and retains at least partial activity in this conformation. These observations suggest that the plasticity observed in NS2 may be connected to its function in vivo.

In recent years, molten globule-like conformations as found in NS2, as well as partly folded and even unfolded states of proteins have been found to exert important functions in vivo (66, 67). The flexibility and adaptability of these structures are thought to mediate the transport of proteins through membranes (68, 69), to facilitate the specific recognition of RNA by proteins (70, 71), and to play a role in ribosome assembly (72). Protein-protein interactions in particular often seem to be mediated by disordered states of proteins. During macromolecular assembly, plastic states are known to allow the formation of extremely intimate interactions, which could not be formed by the assembly of rigid structures (73). Furthermore, binding-induced structure formation of highly flexible proteins has also been shown to direct the assembly of multiprotein complexes (67, 74), and to allow precise regulation of binding events by phosphorylation or interaction with auxiliary factors (75-77). In addition, protein plasticity is thought to a play key role in the ability of proteins such as p21^{Waf1/Cip1/Sdi1} to interact with multiple ligands (78).

The HIV-1 Rev-like NES of NS2 is located in its highly flexible N-terminal region (18, 19, 23). In contrast to most proteins investigated, the interaction of both NS2 and HIV-1 Rev with the nuclear export receptor Crm1 is independent of the integrity of the NES (19, 22, 23, 79, 80), indicating that regions surrounding the NES itself may be involved in the optimal recognition of transport substrates by Crm1. While the nuclear export domains of many proteins assume a stable conformation in the absence of the nuclear export machinery, others, like the effector domain of NS2 do not (21, 81-87). These observations indicate that protein plasticity may play a role in the assembly of the nuclear export complex and in the recognition of cargo proteins by the nuclear export machinery.

Interestingly, if influenza virus-infected cells were exposed to the cytotoxin leptomycin B, export of the proposed cargo of NS2, the vRNPs, was efficiently inhibited. But even so the binding of Rev-like NESs to Crm1 is thought to be inhibited by leptomycin B, the cellular localization of NS2 was not or only slightly affected (12, 49, 88). Similar results have also been reported if influenza virus-infected cells were treated with U0126, an inhibitor of Raf signaling. Under these conditions, the nuclear export of vRNPs was strongly inhibited, but only moderate nuclear accumulation of NS2 was observed, even so RNA export by a NS2-RevRBD fusion protein was abrogated (20). To explain these contradictory results, Watanabe et al. (88) have proposed that NS2 assumes two different forms in infected cells, one which is associated transiently with the vRNPs and whose export is leptomycin B-sensitive, and one which is not associated with the vRNPs, and which is exported by a different, leptomycin B-insensitive export pathway. Our results show that the structure of NS2 is characterized by a high degree of plasticity, indicating that NS2 may indeed exist in multiple conformations in vivo.

	ACKNOWLEDGEMENTS

We thank Dr. M. Hughes for providing the NS2 cDNA, and Drs. C. Brouillette, D. D. Muccio, and H. C. Cheung for use of their equipment. We are grateful to T. Bogart, B. Pybus, and Drs. I. Protassevitch, P. Jackson, and J. Xing for technical assistance and helpful discussions. We thank Drs. P. Prevelige and J. B. Finley for a critical reading of this manuscript.

	FOOTNOTES

^* The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

To whom correspondence should be addressed. Tel.: 205-934-4259; Fax: 205-934-0480; E-mail: ming@cmc.uab.edu.

Published, JBC Papers in Press, December 20, 2001, DOI 10.1074/jbc.M109045200

² B. S. Lommer, J. F. Kearney, and M. Luo, unpublished results.

	ABBREVIATIONS

The abbreviations used are: NP, nucleoprotein; NS2, nonstructural protein 2 (also referred to as NEP, nuclear export protein); Ran, Ras-related nuclear protein; M1, matrix protein; vRNP, viral ribonucleoprotein complex; Crm1, chromosome maintenance protein 1; Rev, regulator of expression of virion proteins; NES, nuclear export signal; HIV, human immunodeficiency virus; GdmCl, guanidinium chloride; FT-IR spectroscopy, Fourier Transform-infrared spectroscopy; ANS, 4,4-dianilino-1,1-binaphthyl-5,5-disulphonic acid; RevRBD, RNA binding domain of Rev.

	REFERENCES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES