- Email: [email protected]
Fusogenic activity of reconstituted newcastle disease virus envelopes: a role for the hemagglutinin-neuraminidase protein in the fusion process
The International Journal of Biochemistry & Cell Biology 34 (2002) 403–413
Fusogenic activity of reconstituted newcastle disease virus envelopes: a role for the hemagglutinin-neuraminidase protein in the fusion process C. Cobaleda, I. Muñoz-Barroso, A. Sagrera, E. Villar∗ Departamento de Bioqu´ımica y Biolog´ıa Molecular, Universidad de Salamanca, Plaza Doctores de la Reina s/n, Edificio Departamental, Lab109, 37007 Salamanca, Spain Received 16 July 2001; received in revised form 16 September 2001; accepted 19 September 2001
Abstract Enveloped viruses, such as newcastle disease virus (NDV), make their entry into the host cell by membrane fusion. In the case of NDV, the fusion step requires both transmembrane hemagglutinin-neuraminidase (HN) and fusion (F) viral envelope glycoproteins. The HN protein should show fusion promotion activity. To date, the nature of HN–F interactions is a controversial issue. In this work, we aim to clarify the role of the HN glycoprotein in the membrane fusion step. Four types of reconstituted detergent-free NDV envelopes were used, on differing in their envelope protein contents. Fusion of the differerent virosomes and erythrocyte ghosts was monitored using the octadecyl rhodamine B chloride assay. Only the reconstituted envelopes having the F protein, even in the absence of HN protein, displayed residual fusion activity. Treatment of such virosomes with denaturing agents affecting the F protein abolished fusion, indicating that the fusion detected was viral protein-dependent. Interestingly, the rate of fusion in the reconstituted systems was similar to that of intact viruses in the presence of the inhibitor of HN sialidase activity 2,3-dehydro-2-deoxy-N-acetylneuraminic acid. The results show that the residual fusion activity detected in the reconstituted systems was exclusively due to F protein activity, with no contribution from the fusion promotion activity of HN protein. © 2002 Elsevier Science Ltd. All rights reserved. Keywords: Newcastle disease virus; Membrane fusion protein; HN–F interactions; R18 dequenching assay; Viral envelope reconstitution
1. Introduction The membrane of newcastle disease virus (NDV), an avian enveloped single-stranded RNA virus belongAbbreviations: F, fusion glycoprotein; HN, hemagglutinin-neuraminidase glycoprotein; HR, heptad repeats; M, matrix protein; NDV, newcastle disease virus; Neu5Ac2en, 2,3-dehydro-2-deoxyN-acetylneuraminic acid; DTT, dithiothreitol; FDQ, fluorescence dequenching; KNP, 120 mM KCl, 30 mM NaCl, 10 mM sodium phosphate pH 7.4; R18, octadecylrhodamine B chloride ∗ Corresponding author. Tel.: +34-923-294465; fax: +34-923-294579. E-mail address: [email protected] (E. Villar).
ing to the family of paramyxoviridae, contains two transmembrane glycoproteins: hemagglutinin-neuraminidase (HN) and fusion F protein [1]. In addition, a membrane-associated non-glycosylated M protein is located underneath the lipid bilayer, interacting with the HN protein and the nucleocapsid [2]. The F protein is directly responsible for the fusion between the viral envelope and the target membrane. For paramyxoviruses, the fusion mechanism has been proposed to occur at neutral pH. Nevertheless, we have recently shown that the fusion of NDV with cultured cells is enhanced at acidic pH [3]. The F protein is produced as a single inactive peptide, Fo, which, once
1357-2725/02/$ – see front matter © 2002 Elsevier Science Ltd. All rights reserved. PII: S 1 3 5 7 - 2 7 2 5 ( 0 1 ) 0 0 1 2 7 - 3
404
C. Cobaleda et al. / The International Journal of Biochemistry & Cell Biology 34 (2002) 403–413
cleaved by a cellular protease (reviewed in [4]), becomes the active F1–F2 form, with two peptides linked by a disulfide bond [5]. To date, four domains of the F1 polypeptide have been suggested to be involved in the fusion mechanism of NDV: the N-terminal fusion peptide [6] and three heptad repeat (HR) regions of the ectodomain, named HR1, HR2 and HR3 [7–9]. HN binds to sialic acid-containing receptors of the cell surface molecules (hemagglutinating activity) and it also displays receptor-destroying activity; namely, neuraminidase or sialidase activity. Both the sialidase and hemagglutinating activities depend on the lipid environment [10]. Moreover, it has been demonstrated that HN possesses a third activity; namely, fusion promotion activity [11,12]; this activity seems to be located in the stalk domain of the HN protein [13]. The complete mechanism of NDV-induced membrane fusion is still unknown. Like many other paramyxoviruses, NDV requires the presence of the two homotypic membrane glycoproteins in the same bilayer to induce fusion (revised in [14]). Although it seems clear that HN fusion promotion activity is required for F protein-mediated membrane fusion in most paramyxoviruses, the nature of this activity and the regions of both proteins involved in functional interactions remain open questions. In paramyxoviruses, type-specific HN–F interactions seem to be established before HN interacts with the target membrane receptor, in the rough endoplasmic reticulum (ER) of the host cell [15,16], before HN becomes able to interact with receptors. Recently, it has been proposed [9] that the HR3 domain of the F protein could be involved in the requirement of HN protein for fusion. In this sense, the authors reported the existence of an F mutant in the HR3 region that escapes the need for HN for fusion. In order to gain further insight into the requirements of HN–F interactions for NDV protein-mediated membrane fusion, we prepared several reconstituted viral envelopes from the paramyxovirus NDV “Clone 30”. These virosomes differed in the composition of their membrane-associated proteins: (i) virosomes containing the three viral membrane-bound proteins (HN–F–M envelopes); (ii) virosomes containing both transmembrane proteins (HN–F envelopes); (iii) virosomes containing the F protein only (F-envelopes); and (iv) virosomes containing viral lipid vesicles and no proteins. We examined the temperature- and pH-dependence of the fusion kinetics and the final
extent of fusion of the different reconstituted envelopes. The results were compared with those of intact virions, used as controls. For all the virosomes studied, the extent of fusion was 10 times lower than that of intact viruses, regardless of their protein composition. Fusion activity was seen to be proteindependent. We discuss these results in terms of the disappearance of native HN–F interactions after detergent disruption of the viral membrane and subsequent membrane reconstitution.
2. Materials and methods 2.1. Materials Octadecyl rhodamine B chloride (R18) was a product from Molecular Probes Inc. (Junction City OR; USA); DTT Tris, Triton X-100, BSA, SDS, and Sephadex G-75 were all from SIGMA (St. Louis, MO, USA). SM2 Bio-Beads (20–50 mesh) and reagents for SDS-PAGE were purchased from Bio-Rad (Richmond, CA) and all other reagents were of analytical grade. 2.2. Viruses NDV “Clone 30” was grown and purified mainly as described previously [3]. 2.3. Preparation of erythrocyte ghosts Pig (Sus scropha L var. domestica) blood was obtained from a local slaughterhouse and collected in vessels containing 0.15 M NaCl/1.5% (w/v) EDTA as anticoagulant. Ghosts were prepared by hypotonic lysis of the erythrocytes in 5 mM sodium phosphate, pH 8.0, at 4 ◦ C, as described previously [17]. 2.4. Reconstitution of NDV membrane components 2.4.1. Solubilization of NDV membranes This was performed according to the Triton X-100 solubilization procedure [18,19], with the modifications described previously [10]. Briefly, purified NDV particles were resuspended in 10 mM potassium phosphate, 1 M KCl buffer, pH 7.2, at a protein concentration of 5 mg/ml. Then, Triton X-100 was added
C. Cobaleda et al. / The International Journal of Biochemistry & Cell Biology 34 (2002) 403–413
405
to a final concentration of 2% (v/v). After 30 min incubation at room temperature with gentle shaking, the mixture was centrifuged at 200,000 × g for 90 min to remove the pelleted nucleocapsids. The supernatant or “solubilized envelopes” contained HN, F and M proteins, and viral lipids.
previously activated as described elsewhere [23]. Before removal of the detergent, R18 was added at the right proportion to ensure self-quenching (0.3 mM) and later linear dequenching upon fusion with the target membrane [24]. Then, Triton X-100 was removed using SM2 Bio-Beads as previously described [10].
2.4.2. Removal of M protein from solubilized envelopes This was achieved by precipitation of solubilized envelopes at low ionic strength [10]. Briefly, solubilized envelopes were extensively dialyzed against 10 mM Tris–HCl, 150 mM NaCl buffer, pH 6.0. After the dialysis procedure, M protein was pelleted by centrifugation at 12,000 × g for 30 min.
2.5. Labeling of NDV with R18
2.4.3. Removal of HN protein to obtain F-envelopes To prepare envelopes containing only the F protein, we used the modification of the method described by Tomasi and Loyter [20], as described in [21]. Briefly, pelleted viruses were resuspended in 20 mM Tris–HCl, 150 mM NaCl, 3 mM DTT buffer, pH 8.4, up to a protein concentration of 5 mg/ml, and incubated at 37 ◦ C for 2 h. During this time, DTT reduces disulfide bridges in viral glycoproteins. Then, the samples were dialyzed for 24 h against four changes of 10 mM Tris–HCl, 150 mM NaCl, 2 mM Ca2+ , 2 mM Mg2+ buffer, pH 7.4. This dialysis step removes DTT and allows the regeneration of disulfide bridges in F protein, but not in HN protein [20]. Viruses were pelleted as described above and resuspended up to a protein concentration of 10 mg/ml in 10 mM Tris–HCl, 150 mM NaCl, 850 mM KCl, 2 mM Ca2+ , 2 mM Mg2+ buffer, pH 7.4, containing 2% (v/v) Triton X-100. This sample was incubated at room temperature with gentle shaking for 60 min. Then, the mixture was centrifuged at 200,000 × g for 90 min to remove pelleted nucleocapsids and HN aggregates. The supernatant contained F and M proteins. Following this, M protein was removed by dialysis and centrifugation as described above. 2.4.4. Collection of reconstituted envelopes from solubilized envelopes The method used was based on the procedure previously described for Sendai virus [22] and uses the absorbent copolymer SM2 Bio-Beads for removing the detergent Triton X-100 (Bio-Rad, 20–50 mesh),
Labeling of intact NDV was accomplished as described previously [24], adapted from [25]. 2.6. R18 fusion assays The fusion assays were based on the relief of self-quenching of the probe which, when diluted in the target membrane as a result of fusion and/or probe transfer, gives rise to an increase in the fluorescence emission signal [25]. Continuous monitoring of the R18 fluorescence was carried out on a Hitachi F-4010 spectrofluorimeter (excitation wavelength, 560 nm; emission wavelength, 590 nm; slit widths, 5 and 10 nm for excitation and emission, respectively). All components in the cuvette were stirred continuously with a magnetic stirrer during the reaction time, and temperature was controlled by a thermostatically-controlled water bath. In a typical fusion experiment, 25 g of R18-labeled viruses or reconstituted envelopes (total proteins) in 5–15 l of KNP buffer, were added to KNP buffer at 37 ◦ C, up to a final volume of 2 ml. The fluorescence of the mixture was taken as zero. Then, 130 g of erythrocyte ghosts were added and the development of R18 fluorescence was monitored continuously for approximately 60 min. Fusion was stopped by the addition of Triton X-100 at 1% (v/v, final concentration), and the resulting fluorescence was taken as 100%. The percentage of fluorescence dequenching, % FDQ, at a given time was calculated according to the equation: Ft − F0 % FDQt = 100 (1) F100 − F0 where F0 and Ft are the fluorescence intensities at time zero and at the given time point, respectively; F100 is the fluorescence after the addition of Triton X-100, which was considered to result in infinite dilution of the probe [25].
406
C. Cobaleda et al. / The International Journal of Biochemistry & Cell Biology 34 (2002) 403–413
When the effect of pH on fusion was analyzed, 10 mM sodium citrate, 120 mM KCl, 30 mM NaCl buffer, pH 4.0 was used for acidic pH; and 10 mM sodium carbonate, 120 mM KCl, 30 mM NaCl buffer, pH 10.0 for basic pH. In experiments in which viruses or envelopes were treated with different agents, incubations of the different envelopes were performed at 37 ◦ C for 30 min in a total volume of 200 l in 10 mM sodium phosphate, 120 mM KCl, 30 mM NaCl buffer, pH 7.4 containing 5% glutaraldehyde or 2 mM DTT as reduction agents, or in the presence of the neuraminidase inhibitor Neu5Ac2en (0.1 mM final concentration). 2.7. Preparation of liposomes Total lipids from NDV membrane were extracted by the method previously described [26]. The organic solvents were evaporated with N2 and then placed under a vacuum for 3 h to eliminate any residual solvent. Dried lipids were resuspended in PBS (pH 7.4) at a lipid concentration of 5 mg/ml. After 30 min at room temperature under an N2 atmosphere, samples were subjected to high-intensity ultrasonic irradiation for 5 min under a stream of N2 at 4–10 ◦ C using a Branson B-30 sonicator. Suspensions were then centrifuged at 100,000 × g for 60 min in order to pellet large liposomes. The supernatant was used as the source of lipids vesicles. 2.8. SDS–PAGE SDS polyacrylamide gel electrophoresis was performed in 5–15% polyacrylamide gradient gels. Gels were stained with silver nitrate. 2.9. Determination of Triton X-100 The amount of Triton X-100 in the samples was determined as previously described [27]. 2.10. Data analysis In order to quantify the dequenching due to fusion and due to non-specific probe transfer between membranes, data analysis was essentially as previously described by us [24]. Kinetic data were fitted by non-linear regression using the “SIMFIT” computer
package version 3.1, developed by W.G. Bardsley, University of Manchester, UK [28] using the sum of two exponential terms model: % FDQ = A1 [1 − exp(−k1 t)] + A2 [1 − exp(−k2 t)] (2) The goodness of each individual fit was evaluated using the X2 -test, the run and sign test of residuals, plots of residuals, the magnitude of relative residuals, the t-test for parameter redundancy and the R2 coefficient. Since we have previously shown [25] that the second exponential reflects non-specific probe transfer, only the values of the first exponential are shown, where the A1 parameter represents the asymptote of the specific part of the equation, while the k1 parameter represents the reaction rate.
3. Results and discussion A very useful way of studying the structural and functional properties of membrane proteins is to extract them from their native environment by solubilization with detergents, and then to reconstitute them in membrane systems with a defined lipid and/or protein composition [29]. These reconstitution procedures allow one to control the lipid and protein environment and to study the interactions that the protein of interest might establish with other membrane components by comparison with the same environments of native membranes. In the case of viral proteins, the membrane proteins in the reconstituted systems must be able to induce the fusion of virosomes with target membranes. These systems are important tools for the study of viral fusion mechanisms. Using this methodology, it has been possible to obtain functional reconstituted envelopes from different viruses such as influenza [30], Sendai [22,31], Semliki forest [32] or vesicular stomatitis viruses [33,34]. In the case of NDV, we have previously reconstituted the viral glycoproteins in different lipid envelopes [10] in order to characterize the dependence of the biological activities of HN on to the lipid composition of the bilayer. In the present work, our aim was to analyze the role of interactions among different viral envelope proteins in the fusogenic activity of NDV. To accomplish
C. Cobaleda et al. / The International Journal of Biochemistry & Cell Biology 34 (2002) 403–413
this goal, we prepared four different types of reconstituted envelopes: (i) envelopes containing the three viral membrane proteins, HN–F–M envelopes; (ii) envelopes containing the two transmembranal NDV proteins, HN–F envelopes; (iii) envelopes containing only the fusion protein or F-envelopes; (iv) lipid vesicles made from protein-free viral lipids (liposomes). As described in Section 2, reconstituted envelopes were prepared, by solubilization of viral particles in the presence of Triton X-100, followed by selective removal of undesired proteins (capsids, M or HN) and removal of the detergent by step-wise addition of SM2 Bio-Beads (Bio-Rad). This method allowed us to eliminate 99.9–100% of the Triton X-100, implying that the residual detergent was present in the reconstitued envelopes at a concentration between 0 and 0.069% (v/v) as determined for HN–F envelopes. Moreover, it has previously been shown [10] that this methodology produces homogeneous HN–F envelopes, as revealed by density gradient centrifugation analysis. The functionality of the HN protein in the reconstituted systems was assayed: in HN–F virosomes, HN showed sialidase and hemnagglutinating specific activity similar to that of intact NDV (not shown). In addition, quantification of sialidase activity in reconstituted envelopes before and after solubilization with Triton X-100 (2%, v/v final concentration) indicated that most of the neuraminidase activity (95%) was present on the external side of the reconstituted system
407
(not shown). This result suggests that the orientation of the viral envelope proteins in the bilayer of the virosomes would be similar to that of the intact viruses. The protein content of the reconstituted envelopes was analyzed by SDS-PAGE (Fig. 1). In the case of envelopes lacking M protein, we observed that it was not possible to acheive complete elimination of the M protein, 10% of residual M protein remaining. For F-envelopes, protein composition analysis revealed that the DTT reduction method used here (based on [20], see Section 2) afforded complete elimination of HN protein without affecting F protein gel mobility as detected by SDS-PAGE (Fig. 1). The fusogenic activity of intact NDV has been fully described earlier by us, using erythrocyte ghosts [24] or cultured cells [3] as target membranes. On using erythrocyte ghosts, as in the present work, NDV fusogenic activity was maximum at neutral pH and 37 ◦ C [24] but decreased dramatically at higher or lower pH values. Moreover, DTT and glutaraldehyde treatment of virions reduced the fusing activity to less than 20% [24]. The presence of the competitive inhibitor of HN neuraminidase activity, 2,3-dehydro-2deoxy-N-acetylneuraminic acid (Neu5Ac2en), elicited a strong decrease in the fusion parameters as well as in the initial rate of fusion (Fig. 2 and [24]), supporting a role for HN in the fusion process. Fig. 2 shows a graphic representation of the kinetic parameters A1 and k1 of the specific fusion reaction
Fig. 1. SDS-PAGE analysis of the reconstituted envelopes. Samples were electrophoresed in 5–15% polyacrylamide gradient gels and the gels were then stained with silver nitrate. Lanes 2 and 9, HN–F envelopes in the presence of the reducing agent DTT (40 mM); Lanes 3 and 8, F-envelopes in the absence of DTT; Lanes 4 and 7, F-envelopes in the presence of DTT; Lane 5, HN–F–M envelopes in the presence of DTT; Lane 6, molecular weight standards.
408
C. Cobaleda et al. / The International Journal of Biochemistry & Cell Biology 34 (2002) 403–413
Fig. 2. Kinetic constants for intact virions and reconstituted envelopes. Absolute values of the kinetic constants A1 (asymptotic value of specific fusion) and k1 (rate of specific fusion) (Eq. (2)) for the fusion process for NDV and for the different types of envelopes, using erythrocyte ghosts as membrane target. For each kind of sample, values in the absence (control) or in the presence of 0.1 mM of the sialidase inhibitor Neu5Ac2en (Inh) are shown. NDV, data for intact virus; NDV Inh, intact virus in the presence of the sialidase inhibitor; HN–F–M, virosomes composed by the three viral proteins; HN–F–M Inh, HN–F–M virosomes in the presence of the sialidase inhibitor; HN–F, virosomes composed by HN and F viral proteins; HN–F Inh, HN–F virosomes in the presence of the sialidase inhibitor; F, virosomes composed by the F protein; F Inh, F virosomes in the presence of the sialidase inhibitor; Lipos, liposomes lacking viral proteins; Lipos Inh, liposomes in the presence of the sialidase inhibitor. Data are from one representative experiment.
for the different types of virosomes as well as for intact viruses and liposomes. As mentioned above (see Section 2), taking into account that R18 non-specific transfer between membranes can occur (second exponential in Eq. (2)), only specific fusion kinetic parameters (first exponential in Eq. (2)) are shown, where A1 represents the extent of fusion, and k1 represents the reaction rate. Analysis of the kinetic constants A2 and k2 for non-specific probe transfer between reconstituted systems and erythrocyte membranes revealed no significant changes in them (not shown). For each sample, two different experiments are shown in Fig. 2; the control one under optimal conditions, i.e. pH 7.4, 37 ◦ C and 130 g of target membrane proteins, and a second experiment performed in the presence of
the neuraminidase inhibitor Neu5Ac2en (0.1 mM, final concentration), labeled with “Inh”. From the values of the kinetic constants shown in Fig. 2, several preliminary conclusions can be drawn: (i) for all types of envelopes (liposomes apart), the values of A1 and k1 are very similar; (ii) the k1 data for the three kinds of envelopes, representing the rate of the specific fusion reaction (see Eq. (2)), are one order of magnitude lower than that of the intact viruses under optimum conditions, and are very similar to the results obtained for intact NDV in the presence of the inhibitor Neu5Ac2en (NDV Inh); and (iii) for all three reconstituted envelopes, the difference between the kinetic constants in the presence or the absence of the sialidase inhibitor is not significant.
C. Cobaleda et al. / The International Journal of Biochemistry & Cell Biology 34 (2002) 403–413
409
Fig. 3. Kinetic fusion parameters for HN–F–M envelopes under different conditions. Relative values of the kinetic constants A1 (asymptotic value of specific fusion) and k1 (rate of specific fusion) (Eq. (2)) for fusion between HN–F–M envelopes and erythrocyte ghosts. Data were calculated as percentages of those of a control experiment in which HN–F–M envelopes were incubated at 37 ◦ C, pH 7.4, in the presence of 130 g of target membrane proteins (optimum conditions) taken as 100% (control). Neu5Ac2en, virosomes in the presence of 0.1 mM of the sialidase inhibitor. pH 4, pH 10; 15 and 25 ◦ C, HN–F–M envelopes were incubated in the presence of 130 g of target membrane proteins at the indicated pH or temperature values. Gh2x indicates a double amount of target membranes with respect to the control. Glut indicates virosomes treated with 5% of glutaraldehyde, as described in Section 2. Data are from one representative experiment.
In order to analyse the role of viral membrane proteins in the fusion activity of the reconstituted envelopes studied here, we performed several experiments in the presence of different agents and under different conditions. Figs. 3–5 represent the values of the A1 and k1 kinetic parameters of the fusion process performed under different experimental conditions for the HN–F–M, HN–F and F reconstituted envelopes, respectively. The data are expressed as percentages of those from a control experiment conducted under optimum conditions (pH 7.4, 37 ◦ C and 130 g of target membrane proteins) with the indicated type of envelopes. It should be noted that the data on the kinetic constants for reconstituted envelopes reveals greater variability than for the intact viruses, probably due to the fact that they were one order of magnitude lower (Fig. 2). From the data shown in Figs. 3–5, it can be seen that the fusogenic activity of the three envelopes studied is completely abolished after DTT or glutaraldehyde
treatment of virosomes. In addition, it is sensitive to changes in pH and temperature. This demonstrates that the fusing activity analyzed here, although severely diminished with respect to intact viruses, is indeed dependent on the virus envelope proteins. Thus, and as expected, fusion activity was totally lost in liposomes devoid of any protein (see Fig. 2). Surprisingly, fusion of HN–F virosomes with erythrocyte ghost was not abolished at pH 4. This result is difficult to interpret, although one explanation can be found in the activation of NDV fusion at acidic pH reported previously by us [3]. Why it should be exerted in HN–F virosomes but not in HN–F–M virosomes remains unclear. In sum, in reconstituted NDV envelopes the extent and rate of fusion, represented by the A1 and k1 kinetic parameters, respectively, were 10 times lower than those of intact viruses. The fusing activity of virosomes was very similar to that of intact NDV in the presence of the inhibitor Neu5Ac2en, which is known to act specifically on sialidase HN protein
410
C. Cobaleda et al. / The International Journal of Biochemistry & Cell Biology 34 (2002) 403–413
Fig. 4. Kinetic fusion parameters for HN–F envelopes under different conditions. Relative values of the kinetic constants A1 (asymptotic value of specific fusion) and k1 (rate of specific fusion) (Eq. (2)) for HN–F envelopes-erythrocyte ghosts fusion. Data were calculated as percentages of those of a control experiment in which HN–F envelopes were incubated at 37 ◦ C, pH 7.4, in the presence of 130 g of target membrane proteins (optimum conditions) taken as 100% (control). Neu5Ac2en, virosomes in the presence of 0.1 mM of the sialidase inhibitor. pH 4, pH 10; 4, 15, and 25 ◦ C, HN–F envelopes were incubated in the presence of 130 g of target membrane proteins at the indicated pH or temperature values. Gh2x indicates a double amount of target membranes with respect to the control. DTT and Glut indicates treatment of virosomes with 2 mM of DTT or 5% of glutaraldehyde, respectively, as described in Section 2. Data are from one representative experiment.
activity but not on hemagglutinating activity [35]. For all three types of reconstituted envelopes studied here, the residual fusing activity was the same, regardless of the presence of HN or M proteins. This suggests that membrane fusion is only due to the presence of the F protein without the participation of a coadjuvant active HN protein. Such a conclusion is also supported by the fact that the fusogenic activity of the envelopes was not affected by the presence of the sialidase inhibitor (Fig. 2), as in intact viruses. It is well documented that for many paramyxoviruses type specific HN–F interactions are required for fusion [14]. The fusion promotion activity of HN protein has been proposed as an effect of the viral attachment protein on the conformation of the fusion protein. After HN has interacted with the target membrane receptors, specific HN–F interactions would elicit a conformational change in F protein that
would allow the fusion peptide to be exposed and to interact with the target membrane, thereby triggering the fusion cascade [14,36]. Moreover, even a weakened interaction between HN and F is still sufficient to trigger fusion [37]. To date, the nature of these interactions remains to be elucidated, but from the data presented here it could be argued that the decrease in the fusogenic activity of NDV proteins after reconstitution would be due to the disappearance of functional interactions between HN and F proteins during the solubilization-reconstitution cycle. For most paramyxoviruses, including NDV and with the exception of simian virus 5 [12,38] and respiratory syncytial virus [39], it has been reported that the expression of F protein in the absence of HN protein does not induce cell–cell fusion. The HR1 and HR3 domains seem to be involved in the HN-independent fusing activity of the simian virus 5
C. Cobaleda et al. / The International Journal of Biochemistry & Cell Biology 34 (2002) 403–413
411
Fig. 5. Kinetic fusion parameters for F-envelopes under different conditions. Relative values of the kinetic constants A1 (asymptotic value of specific fusion) and k1 (rate of specific fusion) (Eq. (2)) for F-envelopes-erythrocyte ghosts fusion. Data were calculated as percentages of those of a control experiment in which F-envelopes were incubated at 37 ◦ C, pH 7.4, in the presence of 130 g of target membrane proteins (optimal conditions) taken as 100% (control). Neu5Ac2en, virosomes in the presence of 0.1 mM of the sialidase inhibitor. pH 4, pH 10, and 4 ◦ C, F-envelopes were incubated in the presence of 130 g of target membrane proteins at the indicated pH or temperature values, respectively. Glut indicates treatment of virosomes with glutaraldehyde 5%, as described in Section 2. Data are from one representative experiment.
F protein [40]. In NDV, a single amino acid change in the ectodomain of F protein (HR3 region) modifies the requirement for HN protein in syncytium formation [9]. In this sense, in the present report we detected fusion activity in HN-lacking virosomes. The residual fusion detected in F-envelopes is difficult to explain because the receptor-binding protein was not present. Several possibilities could be advanced in this regard: (i) the presence in the reconstituted systems of an undetected binding agent; (ii) the presence of residual amounts of HN protein, not detected by SDS-PAGE (Fig. 1); (iii) the F protein of NDV could have a residual F activity independent of HN protein; and (iv) in the reconstituted systems, F could escape the requirements of the attachment protein. After the reconstitution procedure used here, NDV F protein can acquire a given conformation that would allow it to develop attachment activity in absence of HN
protein [9]. In this context, for simian virus 5 it has been proposed that the F protein might interact with a specific receptor of the target membrane before the triggering of the fusion cascade, independently of the attachment function of HN protein [41]. Thus, the reconstituted envelopes studied here retained certain fusion activity, dependent only on the F protein, as demonstrated by the fact that this activity could not be inhibited by Neu5Ac2en (Fig. 2). The residual fusing activity present in the three reconstituted systems (including HN-lacking envelopes, i.e. F-envelopes) suggests that a small population of F proteins escapes the need for HN protein for fusion; this was also detected in intact viruses in the presence of the sialidase inhibitor Neu5Ac2en (Fig. 2 and [24]). The effect of the inhibitor on the fusion activity of intact NDV supports the idea that HN neuraminidase activity may affect F protein-mediated
412
C. Cobaleda et al. / The International Journal of Biochemistry & Cell Biology 34 (2002) 403–413
membrane fusion, although the possibility that this agent might alter HN fusion promotion itself cannot be ruled out. It has been reported [42] that, in NDV, HN fusion promotion does not correlate with sialidase activity since different HN mutants show changes in sialidase activity without fusion promotion activity being affected, and vice versa. Other studies [43] have suggested that the high receptor-destroying activity of HN might reduce the fusion promotion activity of the protein. Therefore, the role of neuraminidase activity in paramyxovirus F protein-mediated membrane fusion should be analyzed in more detail. Although we have failed to reconstitute fully functional NDV envelopes as compared with the fusion activity of intact viruses, the residual F activity detected here allows us to draw some conclusions about HN–F interactions in NDV-mediated fusion. NDV-derived virosomes proved to be 10 times less fusogenic than intact viruses. Our theory is that in HN–F–M and HN–F reconstituted envelopes, HN fails to activate F protein and to promote its fusion activity. In sum, the loss of fusion promotion detected here could be due to several reasons, in agreement with the proposals of Lamb [14] and Horvath et al. [12]: (i) after reconstitution, HN–F physical interactions are modified or destroyed, as discussed above; (ii) in the reconstituted systems, the ratio and/or distance between both glycoproteins in the membrane of the virosomes are different from those of intact virions which would not allow the HN fusion promotion signal. Further experiments must be performed in order to gain direct evidence in support of these hypotheses. In any case, the present work is a further contribution to the currently accepted view of the need for HN protein in paramyxovirus membrane fusion (revised by [14]), considering that the loss of the HN fusion promotion activity proposed here is due to the rupture of HN–F interactions that occur in the intact virus. Since the sialidase and hemagglutinating specific activities of HN in reconstituted systems (HN–F envelopes) were similar to those of intact virions, HN fusion promotion activity seems to be a more critical activity than the other two.
Acknowledgements This work was supported by Grants fom the Spanish DGES (PM97-0160) and the Junta de Castilla y
León (SA 62/99) to E.V. The authors thank E. Diaz de Espada and José Antonio Rodr´ıguez Hernández, from Laboratories Intevret, Salamanca, Spain, for providing the lentogenic “Clone-30” NDV strain. C.Cobaleda and A.Sagrera were fellowship holders of the Plan de Formación del Profesorado Universitario y Personal Investigador, Ministerio de Educación y Cultura, Spain.
References [1] P.W. Choppin, R.W. Compans, in: H. Fraenkel-Conrat, R.R. Wagner (Eds.), Comprehensive Virology, Plenum Press, New York, 1975, pp. 95–178. [2] A. Garc´ıa-Sastre, J.A. Cabezas, E. Villar, Proteins of newcastle disease virus envelope: interaction between the outer hemagglutinin-neuraminidase glycoprotein, Biophys. Acta 999 (1989) 171–175. [3] K. San Román, E. Villar, I. Muñoz-Barroso, Acidic pH enhancement of the fusion of newcastle disease virus with cultured cells, Virology 260 (1999) 329–341. [4] T.G. Morrison, A. Portner, in: H. Fraenkel-Conrat, R.R. Wagner (Eds.), The Paramyxoviruses, Plenum Press, New York, 1991, pp. 347–382. [5] S. Iwata, A.C. Schmidt, K. Titani, M. Suzuki, H. Kido, B. Gotoh, M. Hamaguchi, Y. Nagai, Assignment of disulfide bridges in the fusion glycoprotein of Sendai virus, J. Virol. 68 (1994) 3200–3206. [6] R. Brasseur, M. Vandenbranden, B. Cornet, A. Burny, J.M. Ruysschaert, Orientation into the lipid bilayer of an asymmetric amphipathic helical peptide located at the N-terminus of viral fusion proteins, Biochim. Biophys. Acta 1029 (1990) 267–273. [7] P. Chambers, C.R. Pringle, A.J. Easton, Heptad repeat sequences are located adjacent to hydrophobic regions in several types of virus fusion glycoproteins, J. Gen. Virol. 71 (1990) 3075–3080. [8] R. Buckland, F. Wild, Leucine zipper motif extends, Nature 338 (1989) 547. [9] T.A. Sergel, L.W. McGinnes, T.G. Morrison, A single amino acid change in the newcastle disease virus fusion protein alters the requirement for HN protein in fusion, J. Virol. 74 (2000) 5101–5107. [10] I. Muñoz-Barroso, C. Cobaleda, G. Zhadan, V. Shnyrov, E. Villar, Dynamic properties of newcastle disease virus envelope and their relations with viral hemagglutinin-neuraminidase membrane glycoprotein, Biochim. Biophys. Acta 1327 (1997) 17–31. [11] T. Morrison, C. McQuain, L. McGinnes, Complementation between avirulent newcastle disease virus and a fusion protein gene expressed from a retrovirus vector: requirements for membrane fusion, J.Virol. 65 (1991) 813–822. [12] C.M. Horvath, R.G. Paterson, M.A. Shaughnessy, R. Wood, R.A. Lamb, Biological activity of paramyxovirus fusion
C. Cobaleda et al. / The International Journal of Biochemistry & Cell Biology 34 (2002) 403–413
[13]
[14] [15]
[16]
[17]
[18]
[19]
[20]
[21]
[22]
[23]
[24]
[25]
[26]
[27]
[28]
[29]
proteins: factors influencing formation of syncytia, J. Virol. 66 (1992) 4564–4569. J. Stone-Hulslander, T.G. Morrison, Mutational analysis of heptad repeats in the membrane-proximal region of newcastle disease virus HN protein, J. Virol. 73 (1999) 3630–3637. R.A. Lamb, Paramyxovirus fusion: a hypothesis for changes, Virology 197 (1993) 1–11. J. Stone-Hulslander, T.G. Morrison, Detection of an interaction between the HN and F proteins in newcastle disease virus-infected cells, J. Virol. 71 (1997) 6287–6295. S. Tong, R.W. Compans, Alternative mechanisms of interaction between homotypic and heterotypic parainfluenza virus HN and F proteins, J. Gen. Virol. 80 (1999) 107–115. T.L. Steck, J.A. Kant, Preparation of impermeable ghost and inside–out vesicles from human erythrocyte membranes, Meth. Enzymol. 31 (1974) 172–180. A. Scheid, L.A. Caliguiri, R.W. Compans, P.W. Choppin, Isolation of paramyxovirus glycoproteins. Association of both hemagglutinating and neuraminidase activities with the larger SV5 glycoprotein, Virology 50 (1972) 640–652. A. Scheid, P.W. Choppin, Isolation and purification of the envelope proteins of newcastle disease virus, J. Virol. 11 (1973) 263–271. M. Tomasi, A. Loyter, Selective extraction of biologically active F-glycoprotein from dithiothreitol-reduced Sendai virus particles, FEBS Lett. 131 (1981) 381–385. S. Bagai, A. Puri, R. Blumenthal, D.P. Sarkar, Hemagglutininneuraminidase enhances protein-mediated membrane fusion of reconstituted Sendai virus envelopes with cells, J. Virol. 67 (1993) 3312–3318. A. Vainstein, M. Hershkovitz, S. Israel, S. Rabin, A. Loyter, A new method for reconstitution of highly fusogenic Sendai virus envelopes, Biochim. Biophys. Acta 773 (1984) 181–188. P.W. Holloway, A simple procedure for removal of Triton X-100 from protein samples, Anal. Biochem. 53 (1973) 304– 308. C. Cobaleda, A. Garc´ıa-Sastre, E. Villar, Fusion between newcastle disease virus and erythrocyte ghosts using octadecyl rhodamine B fluorescence assay produces dequenching curves that fit the sum of two exponentials, Biochem. J. 300 (1994) 347–354. D. Hoekstra, T. de Boer, K. Klappe, J. Wilschut, Fluorescence method for measuring the kinetics of fusion between biological membranes, Biochemistry 23 (1984) 5675–5681. E.G. Bligh, W.J. Dyer, A rapid method of total lipid extraction and purification, Can. J. Biochem. Physiol. 37 (1959) 911– 917. D.L. Wright, D.T. Plummer, Multiple forms of acetylcholinesterase from human erythrocytes, Biochem. J. 133 (1973) 521–527. W.G. Bardsley, N.A.J. Bukhari, M.W.J. Ferguson, J.A. Cachaza, F.J. Burguillo, Evaluation of model discrimination, parameter estimation and goodness of fit in non-linear regression problems by test statistics distributions, Comput. Chem. 19 (1995) 75–84. A. Helenius, K. Simons, Solubilization of membranes by detergents, Biochim. Biophys. Acta 415 (1975) 29–79.
413
[30] K. Kawasaki, S.B. Sato, S.I. Ohnishi, Membrane fusion activity of reconstituted vesicles of influenza virus hemagglutinin glycoproteins, Biochim. Biophys. Acta 733 (1983) 286– 290. [31] D.J. Volsky, A. Loyter, An efficient method for reassembly of fusogenic Sendai virus envelopes after solubilization of intact virions with Triton X-100, FEBS Lett. 92 (1978) 190–194. [32] A. Helenius, M. Sarvas, K. Simons, Asymmetric and symmetric membrane reconstitution by detergent elimination, Eur. J. Biochem. 116 (1981) 27–35. [33] W.A. Petri, R.R. Wagner, Reconstitution into liposomes of the glycoprotein of vesicular stomatitis virus by detergent dialysis, J. Biol. Chem. 254 (1979) 4313–4316. [34] M. Paternostre, M. Viard, O. Meyer, M. Ghanam, M. Ollivon, R. Blumenthal, Solubilization and reconstitution of vesicular stomatitis virus envelope using octylglucoside, Biophys. J. 72 (1997) 1683–1694. [35] A. Garc´ıa-Sastre, C. Cobaleda, J.A. Cabezas, E. Villar, On the inhibition mechanism of the sialidase activity from newcastle disease virus, Biol. Chem. Hoppe-Seyler 372 (1991) 923–927. [36] K.A. Baker, R.E. Dutch, R.A. Lamb, T.S. Jardetzky, Structural basis for paramyxovirus-mediated membrane fusion, Mol. Cell 3 (1999) 309–319. [37] R. Deng, Z. Wang, P.J. Mahon, M. Marinello, A.MirzaM. Iorio, Mutations in the newcastle disease virus hemagglutinin-neuraminidase protein that interfere with its ability to interact with the homologous F protein in the promotion of Fusion, Virology 253 (1999) 43–54. [38] M. Ito, M. Nishio, M. Kawano, S. Kusagawa, H. Komada, Y. Ito, M.M. Tsurodome, Role of a single amino acid at the amino terminus of the simian virus 5 F2 subunit in syncytium formation, J. Virol. 71 (1997) 9855–9858. [39] R.A. Karron, D.A. Buonagurio, A.F. Georgiu, S.S. Whiteheat, J.E. Adamus, M.L. Clements-Mann, D.O. Harris, V.B. Randolph, S.A. Udem, S.A.B.R. Burphy, M.S. Sidhu, Respiratory syncitial virus (RSV) SH and G proteins are not essential for viral replication in vitro: clinical evaluation and molecular characterization of a cold-passaged, attenuated RSV subgroup B mutant, Proc. Natl. Acad. Sci. U.S.A. 94 (1997) 13961–13966. [40] M. Ito, M. Nishio, H. Komada, Y. Ito, M. Tsurodome, An amino acid in the heptad repeat 1 domain is important for the haemagglutinin-neuraminidase-independent fusing activity of simian virus 5 fusion protein, J. Gen. Virol. 81 (2000) 719– 727. [41] R.E. Dutch, S. Bagai-Joshi, R.A. Lamb, Membrane fusion promoted by increasing surface densities of the paramyxovirus F and HN proteins: comparison for fusion reactions mediated by simian virus 5 F, human parainfluenza type 3 F, and influenza virus HA, J. Virol. 72 (1998) 7745–7753. [42] T. Sergel, L.W. McGinnes, T.G. Morrison, The fusion promotion activity of the NDV HN protein does not correlate with neuraminidase activity, Virology 196 (1993) 831–834. [43] T. Bousse, T. Takimoto, W.L. Gorman, T. Takahashi, A. Portner, Regions on the hemagglutinin-neuraminidase proteins of human parainfluenza virus type 1 and Sendai virus important for membrane fusion, Virology 204 (1994) 506– 514.
Fusogenic activity of reconstituted newcastle disease virus envelopes: a role for the hemagglutinin-neuraminidase protein in the fusion process C. Cobaleda, I. Muñoz-Barroso, A. Sagrera, E. Villar∗ Departamento de Bioqu´ımica y Biolog´ıa Molecular, Universidad de Salamanca, Plaza Doctores de la Reina s/n, Edificio Departamental, Lab109, 37007 Salamanca, Spain Received 16 July 2001; received in revised form 16 September 2001; accepted 19 September 2001
Abstract Enveloped viruses, such as newcastle disease virus (NDV), make their entry into the host cell by membrane fusion. In the case of NDV, the fusion step requires both transmembrane hemagglutinin-neuraminidase (HN) and fusion (F) viral envelope glycoproteins. The HN protein should show fusion promotion activity. To date, the nature of HN–F interactions is a controversial issue. In this work, we aim to clarify the role of the HN glycoprotein in the membrane fusion step. Four types of reconstituted detergent-free NDV envelopes were used, on differing in their envelope protein contents. Fusion of the differerent virosomes and erythrocyte ghosts was monitored using the octadecyl rhodamine B chloride assay. Only the reconstituted envelopes having the F protein, even in the absence of HN protein, displayed residual fusion activity. Treatment of such virosomes with denaturing agents affecting the F protein abolished fusion, indicating that the fusion detected was viral protein-dependent. Interestingly, the rate of fusion in the reconstituted systems was similar to that of intact viruses in the presence of the inhibitor of HN sialidase activity 2,3-dehydro-2-deoxy-N-acetylneuraminic acid. The results show that the residual fusion activity detected in the reconstituted systems was exclusively due to F protein activity, with no contribution from the fusion promotion activity of HN protein. © 2002 Elsevier Science Ltd. All rights reserved. Keywords: Newcastle disease virus; Membrane fusion protein; HN–F interactions; R18 dequenching assay; Viral envelope reconstitution
1. Introduction The membrane of newcastle disease virus (NDV), an avian enveloped single-stranded RNA virus belongAbbreviations: F, fusion glycoprotein; HN, hemagglutinin-neuraminidase glycoprotein; HR, heptad repeats; M, matrix protein; NDV, newcastle disease virus; Neu5Ac2en, 2,3-dehydro-2-deoxyN-acetylneuraminic acid; DTT, dithiothreitol; FDQ, fluorescence dequenching; KNP, 120 mM KCl, 30 mM NaCl, 10 mM sodium phosphate pH 7.4; R18, octadecylrhodamine B chloride ∗ Corresponding author. Tel.: +34-923-294465; fax: +34-923-294579. E-mail address: [email protected] (E. Villar).
ing to the family of paramyxoviridae, contains two transmembrane glycoproteins: hemagglutinin-neuraminidase (HN) and fusion F protein [1]. In addition, a membrane-associated non-glycosylated M protein is located underneath the lipid bilayer, interacting with the HN protein and the nucleocapsid [2]. The F protein is directly responsible for the fusion between the viral envelope and the target membrane. For paramyxoviruses, the fusion mechanism has been proposed to occur at neutral pH. Nevertheless, we have recently shown that the fusion of NDV with cultured cells is enhanced at acidic pH [3]. The F protein is produced as a single inactive peptide, Fo, which, once
1357-2725/02/$ – see front matter © 2002 Elsevier Science Ltd. All rights reserved. PII: S 1 3 5 7 - 2 7 2 5 ( 0 1 ) 0 0 1 2 7 - 3
404
C. Cobaleda et al. / The International Journal of Biochemistry & Cell Biology 34 (2002) 403–413
cleaved by a cellular protease (reviewed in [4]), becomes the active F1–F2 form, with two peptides linked by a disulfide bond [5]. To date, four domains of the F1 polypeptide have been suggested to be involved in the fusion mechanism of NDV: the N-terminal fusion peptide [6] and three heptad repeat (HR) regions of the ectodomain, named HR1, HR2 and HR3 [7–9]. HN binds to sialic acid-containing receptors of the cell surface molecules (hemagglutinating activity) and it also displays receptor-destroying activity; namely, neuraminidase or sialidase activity. Both the sialidase and hemagglutinating activities depend on the lipid environment [10]. Moreover, it has been demonstrated that HN possesses a third activity; namely, fusion promotion activity [11,12]; this activity seems to be located in the stalk domain of the HN protein [13]. The complete mechanism of NDV-induced membrane fusion is still unknown. Like many other paramyxoviruses, NDV requires the presence of the two homotypic membrane glycoproteins in the same bilayer to induce fusion (revised in [14]). Although it seems clear that HN fusion promotion activity is required for F protein-mediated membrane fusion in most paramyxoviruses, the nature of this activity and the regions of both proteins involved in functional interactions remain open questions. In paramyxoviruses, type-specific HN–F interactions seem to be established before HN interacts with the target membrane receptor, in the rough endoplasmic reticulum (ER) of the host cell [15,16], before HN becomes able to interact with receptors. Recently, it has been proposed [9] that the HR3 domain of the F protein could be involved in the requirement of HN protein for fusion. In this sense, the authors reported the existence of an F mutant in the HR3 region that escapes the need for HN for fusion. In order to gain further insight into the requirements of HN–F interactions for NDV protein-mediated membrane fusion, we prepared several reconstituted viral envelopes from the paramyxovirus NDV “Clone 30”. These virosomes differed in the composition of their membrane-associated proteins: (i) virosomes containing the three viral membrane-bound proteins (HN–F–M envelopes); (ii) virosomes containing both transmembrane proteins (HN–F envelopes); (iii) virosomes containing the F protein only (F-envelopes); and (iv) virosomes containing viral lipid vesicles and no proteins. We examined the temperature- and pH-dependence of the fusion kinetics and the final
extent of fusion of the different reconstituted envelopes. The results were compared with those of intact virions, used as controls. For all the virosomes studied, the extent of fusion was 10 times lower than that of intact viruses, regardless of their protein composition. Fusion activity was seen to be proteindependent. We discuss these results in terms of the disappearance of native HN–F interactions after detergent disruption of the viral membrane and subsequent membrane reconstitution.
2. Materials and methods 2.1. Materials Octadecyl rhodamine B chloride (R18) was a product from Molecular Probes Inc. (Junction City OR; USA); DTT Tris, Triton X-100, BSA, SDS, and Sephadex G-75 were all from SIGMA (St. Louis, MO, USA). SM2 Bio-Beads (20–50 mesh) and reagents for SDS-PAGE were purchased from Bio-Rad (Richmond, CA) and all other reagents were of analytical grade. 2.2. Viruses NDV “Clone 30” was grown and purified mainly as described previously [3]. 2.3. Preparation of erythrocyte ghosts Pig (Sus scropha L var. domestica) blood was obtained from a local slaughterhouse and collected in vessels containing 0.15 M NaCl/1.5% (w/v) EDTA as anticoagulant. Ghosts were prepared by hypotonic lysis of the erythrocytes in 5 mM sodium phosphate, pH 8.0, at 4 ◦ C, as described previously [17]. 2.4. Reconstitution of NDV membrane components 2.4.1. Solubilization of NDV membranes This was performed according to the Triton X-100 solubilization procedure [18,19], with the modifications described previously [10]. Briefly, purified NDV particles were resuspended in 10 mM potassium phosphate, 1 M KCl buffer, pH 7.2, at a protein concentration of 5 mg/ml. Then, Triton X-100 was added
C. Cobaleda et al. / The International Journal of Biochemistry & Cell Biology 34 (2002) 403–413
405
to a final concentration of 2% (v/v). After 30 min incubation at room temperature with gentle shaking, the mixture was centrifuged at 200,000 × g for 90 min to remove the pelleted nucleocapsids. The supernatant or “solubilized envelopes” contained HN, F and M proteins, and viral lipids.
previously activated as described elsewhere [23]. Before removal of the detergent, R18 was added at the right proportion to ensure self-quenching (0.3 mM) and later linear dequenching upon fusion with the target membrane [24]. Then, Triton X-100 was removed using SM2 Bio-Beads as previously described [10].
2.4.2. Removal of M protein from solubilized envelopes This was achieved by precipitation of solubilized envelopes at low ionic strength [10]. Briefly, solubilized envelopes were extensively dialyzed against 10 mM Tris–HCl, 150 mM NaCl buffer, pH 6.0. After the dialysis procedure, M protein was pelleted by centrifugation at 12,000 × g for 30 min.
2.5. Labeling of NDV with R18
2.4.3. Removal of HN protein to obtain F-envelopes To prepare envelopes containing only the F protein, we used the modification of the method described by Tomasi and Loyter [20], as described in [21]. Briefly, pelleted viruses were resuspended in 20 mM Tris–HCl, 150 mM NaCl, 3 mM DTT buffer, pH 8.4, up to a protein concentration of 5 mg/ml, and incubated at 37 ◦ C for 2 h. During this time, DTT reduces disulfide bridges in viral glycoproteins. Then, the samples were dialyzed for 24 h against four changes of 10 mM Tris–HCl, 150 mM NaCl, 2 mM Ca2+ , 2 mM Mg2+ buffer, pH 7.4. This dialysis step removes DTT and allows the regeneration of disulfide bridges in F protein, but not in HN protein [20]. Viruses were pelleted as described above and resuspended up to a protein concentration of 10 mg/ml in 10 mM Tris–HCl, 150 mM NaCl, 850 mM KCl, 2 mM Ca2+ , 2 mM Mg2+ buffer, pH 7.4, containing 2% (v/v) Triton X-100. This sample was incubated at room temperature with gentle shaking for 60 min. Then, the mixture was centrifuged at 200,000 × g for 90 min to remove pelleted nucleocapsids and HN aggregates. The supernatant contained F and M proteins. Following this, M protein was removed by dialysis and centrifugation as described above. 2.4.4. Collection of reconstituted envelopes from solubilized envelopes The method used was based on the procedure previously described for Sendai virus [22] and uses the absorbent copolymer SM2 Bio-Beads for removing the detergent Triton X-100 (Bio-Rad, 20–50 mesh),
Labeling of intact NDV was accomplished as described previously [24], adapted from [25]. 2.6. R18 fusion assays The fusion assays were based on the relief of self-quenching of the probe which, when diluted in the target membrane as a result of fusion and/or probe transfer, gives rise to an increase in the fluorescence emission signal [25]. Continuous monitoring of the R18 fluorescence was carried out on a Hitachi F-4010 spectrofluorimeter (excitation wavelength, 560 nm; emission wavelength, 590 nm; slit widths, 5 and 10 nm for excitation and emission, respectively). All components in the cuvette were stirred continuously with a magnetic stirrer during the reaction time, and temperature was controlled by a thermostatically-controlled water bath. In a typical fusion experiment, 25 g of R18-labeled viruses or reconstituted envelopes (total proteins) in 5–15 l of KNP buffer, were added to KNP buffer at 37 ◦ C, up to a final volume of 2 ml. The fluorescence of the mixture was taken as zero. Then, 130 g of erythrocyte ghosts were added and the development of R18 fluorescence was monitored continuously for approximately 60 min. Fusion was stopped by the addition of Triton X-100 at 1% (v/v, final concentration), and the resulting fluorescence was taken as 100%. The percentage of fluorescence dequenching, % FDQ, at a given time was calculated according to the equation: Ft − F0 % FDQt = 100 (1) F100 − F0 where F0 and Ft are the fluorescence intensities at time zero and at the given time point, respectively; F100 is the fluorescence after the addition of Triton X-100, which was considered to result in infinite dilution of the probe [25].
406
C. Cobaleda et al. / The International Journal of Biochemistry & Cell Biology 34 (2002) 403–413
When the effect of pH on fusion was analyzed, 10 mM sodium citrate, 120 mM KCl, 30 mM NaCl buffer, pH 4.0 was used for acidic pH; and 10 mM sodium carbonate, 120 mM KCl, 30 mM NaCl buffer, pH 10.0 for basic pH. In experiments in which viruses or envelopes were treated with different agents, incubations of the different envelopes were performed at 37 ◦ C for 30 min in a total volume of 200 l in 10 mM sodium phosphate, 120 mM KCl, 30 mM NaCl buffer, pH 7.4 containing 5% glutaraldehyde or 2 mM DTT as reduction agents, or in the presence of the neuraminidase inhibitor Neu5Ac2en (0.1 mM final concentration). 2.7. Preparation of liposomes Total lipids from NDV membrane were extracted by the method previously described [26]. The organic solvents were evaporated with N2 and then placed under a vacuum for 3 h to eliminate any residual solvent. Dried lipids were resuspended in PBS (pH 7.4) at a lipid concentration of 5 mg/ml. After 30 min at room temperature under an N2 atmosphere, samples were subjected to high-intensity ultrasonic irradiation for 5 min under a stream of N2 at 4–10 ◦ C using a Branson B-30 sonicator. Suspensions were then centrifuged at 100,000 × g for 60 min in order to pellet large liposomes. The supernatant was used as the source of lipids vesicles. 2.8. SDS–PAGE SDS polyacrylamide gel electrophoresis was performed in 5–15% polyacrylamide gradient gels. Gels were stained with silver nitrate. 2.9. Determination of Triton X-100 The amount of Triton X-100 in the samples was determined as previously described [27]. 2.10. Data analysis In order to quantify the dequenching due to fusion and due to non-specific probe transfer between membranes, data analysis was essentially as previously described by us [24]. Kinetic data were fitted by non-linear regression using the “SIMFIT” computer
package version 3.1, developed by W.G. Bardsley, University of Manchester, UK [28] using the sum of two exponential terms model: % FDQ = A1 [1 − exp(−k1 t)] + A2 [1 − exp(−k2 t)] (2) The goodness of each individual fit was evaluated using the X2 -test, the run and sign test of residuals, plots of residuals, the magnitude of relative residuals, the t-test for parameter redundancy and the R2 coefficient. Since we have previously shown [25] that the second exponential reflects non-specific probe transfer, only the values of the first exponential are shown, where the A1 parameter represents the asymptote of the specific part of the equation, while the k1 parameter represents the reaction rate.
3. Results and discussion A very useful way of studying the structural and functional properties of membrane proteins is to extract them from their native environment by solubilization with detergents, and then to reconstitute them in membrane systems with a defined lipid and/or protein composition [29]. These reconstitution procedures allow one to control the lipid and protein environment and to study the interactions that the protein of interest might establish with other membrane components by comparison with the same environments of native membranes. In the case of viral proteins, the membrane proteins in the reconstituted systems must be able to induce the fusion of virosomes with target membranes. These systems are important tools for the study of viral fusion mechanisms. Using this methodology, it has been possible to obtain functional reconstituted envelopes from different viruses such as influenza [30], Sendai [22,31], Semliki forest [32] or vesicular stomatitis viruses [33,34]. In the case of NDV, we have previously reconstituted the viral glycoproteins in different lipid envelopes [10] in order to characterize the dependence of the biological activities of HN on to the lipid composition of the bilayer. In the present work, our aim was to analyze the role of interactions among different viral envelope proteins in the fusogenic activity of NDV. To accomplish
C. Cobaleda et al. / The International Journal of Biochemistry & Cell Biology 34 (2002) 403–413
this goal, we prepared four different types of reconstituted envelopes: (i) envelopes containing the three viral membrane proteins, HN–F–M envelopes; (ii) envelopes containing the two transmembranal NDV proteins, HN–F envelopes; (iii) envelopes containing only the fusion protein or F-envelopes; (iv) lipid vesicles made from protein-free viral lipids (liposomes). As described in Section 2, reconstituted envelopes were prepared, by solubilization of viral particles in the presence of Triton X-100, followed by selective removal of undesired proteins (capsids, M or HN) and removal of the detergent by step-wise addition of SM2 Bio-Beads (Bio-Rad). This method allowed us to eliminate 99.9–100% of the Triton X-100, implying that the residual detergent was present in the reconstitued envelopes at a concentration between 0 and 0.069% (v/v) as determined for HN–F envelopes. Moreover, it has previously been shown [10] that this methodology produces homogeneous HN–F envelopes, as revealed by density gradient centrifugation analysis. The functionality of the HN protein in the reconstituted systems was assayed: in HN–F virosomes, HN showed sialidase and hemnagglutinating specific activity similar to that of intact NDV (not shown). In addition, quantification of sialidase activity in reconstituted envelopes before and after solubilization with Triton X-100 (2%, v/v final concentration) indicated that most of the neuraminidase activity (95%) was present on the external side of the reconstituted system
407
(not shown). This result suggests that the orientation of the viral envelope proteins in the bilayer of the virosomes would be similar to that of the intact viruses. The protein content of the reconstituted envelopes was analyzed by SDS-PAGE (Fig. 1). In the case of envelopes lacking M protein, we observed that it was not possible to acheive complete elimination of the M protein, 10% of residual M protein remaining. For F-envelopes, protein composition analysis revealed that the DTT reduction method used here (based on [20], see Section 2) afforded complete elimination of HN protein without affecting F protein gel mobility as detected by SDS-PAGE (Fig. 1). The fusogenic activity of intact NDV has been fully described earlier by us, using erythrocyte ghosts [24] or cultured cells [3] as target membranes. On using erythrocyte ghosts, as in the present work, NDV fusogenic activity was maximum at neutral pH and 37 ◦ C [24] but decreased dramatically at higher or lower pH values. Moreover, DTT and glutaraldehyde treatment of virions reduced the fusing activity to less than 20% [24]. The presence of the competitive inhibitor of HN neuraminidase activity, 2,3-dehydro-2deoxy-N-acetylneuraminic acid (Neu5Ac2en), elicited a strong decrease in the fusion parameters as well as in the initial rate of fusion (Fig. 2 and [24]), supporting a role for HN in the fusion process. Fig. 2 shows a graphic representation of the kinetic parameters A1 and k1 of the specific fusion reaction
Fig. 1. SDS-PAGE analysis of the reconstituted envelopes. Samples were electrophoresed in 5–15% polyacrylamide gradient gels and the gels were then stained with silver nitrate. Lanes 2 and 9, HN–F envelopes in the presence of the reducing agent DTT (40 mM); Lanes 3 and 8, F-envelopes in the absence of DTT; Lanes 4 and 7, F-envelopes in the presence of DTT; Lane 5, HN–F–M envelopes in the presence of DTT; Lane 6, molecular weight standards.
408
C. Cobaleda et al. / The International Journal of Biochemistry & Cell Biology 34 (2002) 403–413
Fig. 2. Kinetic constants for intact virions and reconstituted envelopes. Absolute values of the kinetic constants A1 (asymptotic value of specific fusion) and k1 (rate of specific fusion) (Eq. (2)) for the fusion process for NDV and for the different types of envelopes, using erythrocyte ghosts as membrane target. For each kind of sample, values in the absence (control) or in the presence of 0.1 mM of the sialidase inhibitor Neu5Ac2en (Inh) are shown. NDV, data for intact virus; NDV Inh, intact virus in the presence of the sialidase inhibitor; HN–F–M, virosomes composed by the three viral proteins; HN–F–M Inh, HN–F–M virosomes in the presence of the sialidase inhibitor; HN–F, virosomes composed by HN and F viral proteins; HN–F Inh, HN–F virosomes in the presence of the sialidase inhibitor; F, virosomes composed by the F protein; F Inh, F virosomes in the presence of the sialidase inhibitor; Lipos, liposomes lacking viral proteins; Lipos Inh, liposomes in the presence of the sialidase inhibitor. Data are from one representative experiment.
for the different types of virosomes as well as for intact viruses and liposomes. As mentioned above (see Section 2), taking into account that R18 non-specific transfer between membranes can occur (second exponential in Eq. (2)), only specific fusion kinetic parameters (first exponential in Eq. (2)) are shown, where A1 represents the extent of fusion, and k1 represents the reaction rate. Analysis of the kinetic constants A2 and k2 for non-specific probe transfer between reconstituted systems and erythrocyte membranes revealed no significant changes in them (not shown). For each sample, two different experiments are shown in Fig. 2; the control one under optimal conditions, i.e. pH 7.4, 37 ◦ C and 130 g of target membrane proteins, and a second experiment performed in the presence of
the neuraminidase inhibitor Neu5Ac2en (0.1 mM, final concentration), labeled with “Inh”. From the values of the kinetic constants shown in Fig. 2, several preliminary conclusions can be drawn: (i) for all types of envelopes (liposomes apart), the values of A1 and k1 are very similar; (ii) the k1 data for the three kinds of envelopes, representing the rate of the specific fusion reaction (see Eq. (2)), are one order of magnitude lower than that of the intact viruses under optimum conditions, and are very similar to the results obtained for intact NDV in the presence of the inhibitor Neu5Ac2en (NDV Inh); and (iii) for all three reconstituted envelopes, the difference between the kinetic constants in the presence or the absence of the sialidase inhibitor is not significant.
C. Cobaleda et al. / The International Journal of Biochemistry & Cell Biology 34 (2002) 403–413
409
Fig. 3. Kinetic fusion parameters for HN–F–M envelopes under different conditions. Relative values of the kinetic constants A1 (asymptotic value of specific fusion) and k1 (rate of specific fusion) (Eq. (2)) for fusion between HN–F–M envelopes and erythrocyte ghosts. Data were calculated as percentages of those of a control experiment in which HN–F–M envelopes were incubated at 37 ◦ C, pH 7.4, in the presence of 130 g of target membrane proteins (optimum conditions) taken as 100% (control). Neu5Ac2en, virosomes in the presence of 0.1 mM of the sialidase inhibitor. pH 4, pH 10; 15 and 25 ◦ C, HN–F–M envelopes were incubated in the presence of 130 g of target membrane proteins at the indicated pH or temperature values. Gh2x indicates a double amount of target membranes with respect to the control. Glut indicates virosomes treated with 5% of glutaraldehyde, as described in Section 2. Data are from one representative experiment.
In order to analyse the role of viral membrane proteins in the fusion activity of the reconstituted envelopes studied here, we performed several experiments in the presence of different agents and under different conditions. Figs. 3–5 represent the values of the A1 and k1 kinetic parameters of the fusion process performed under different experimental conditions for the HN–F–M, HN–F and F reconstituted envelopes, respectively. The data are expressed as percentages of those from a control experiment conducted under optimum conditions (pH 7.4, 37 ◦ C and 130 g of target membrane proteins) with the indicated type of envelopes. It should be noted that the data on the kinetic constants for reconstituted envelopes reveals greater variability than for the intact viruses, probably due to the fact that they were one order of magnitude lower (Fig. 2). From the data shown in Figs. 3–5, it can be seen that the fusogenic activity of the three envelopes studied is completely abolished after DTT or glutaraldehyde
treatment of virosomes. In addition, it is sensitive to changes in pH and temperature. This demonstrates that the fusing activity analyzed here, although severely diminished with respect to intact viruses, is indeed dependent on the virus envelope proteins. Thus, and as expected, fusion activity was totally lost in liposomes devoid of any protein (see Fig. 2). Surprisingly, fusion of HN–F virosomes with erythrocyte ghost was not abolished at pH 4. This result is difficult to interpret, although one explanation can be found in the activation of NDV fusion at acidic pH reported previously by us [3]. Why it should be exerted in HN–F virosomes but not in HN–F–M virosomes remains unclear. In sum, in reconstituted NDV envelopes the extent and rate of fusion, represented by the A1 and k1 kinetic parameters, respectively, were 10 times lower than those of intact viruses. The fusing activity of virosomes was very similar to that of intact NDV in the presence of the inhibitor Neu5Ac2en, which is known to act specifically on sialidase HN protein
410
C. Cobaleda et al. / The International Journal of Biochemistry & Cell Biology 34 (2002) 403–413
Fig. 4. Kinetic fusion parameters for HN–F envelopes under different conditions. Relative values of the kinetic constants A1 (asymptotic value of specific fusion) and k1 (rate of specific fusion) (Eq. (2)) for HN–F envelopes-erythrocyte ghosts fusion. Data were calculated as percentages of those of a control experiment in which HN–F envelopes were incubated at 37 ◦ C, pH 7.4, in the presence of 130 g of target membrane proteins (optimum conditions) taken as 100% (control). Neu5Ac2en, virosomes in the presence of 0.1 mM of the sialidase inhibitor. pH 4, pH 10; 4, 15, and 25 ◦ C, HN–F envelopes were incubated in the presence of 130 g of target membrane proteins at the indicated pH or temperature values. Gh2x indicates a double amount of target membranes with respect to the control. DTT and Glut indicates treatment of virosomes with 2 mM of DTT or 5% of glutaraldehyde, respectively, as described in Section 2. Data are from one representative experiment.
activity but not on hemagglutinating activity [35]. For all three types of reconstituted envelopes studied here, the residual fusing activity was the same, regardless of the presence of HN or M proteins. This suggests that membrane fusion is only due to the presence of the F protein without the participation of a coadjuvant active HN protein. Such a conclusion is also supported by the fact that the fusogenic activity of the envelopes was not affected by the presence of the sialidase inhibitor (Fig. 2), as in intact viruses. It is well documented that for many paramyxoviruses type specific HN–F interactions are required for fusion [14]. The fusion promotion activity of HN protein has been proposed as an effect of the viral attachment protein on the conformation of the fusion protein. After HN has interacted with the target membrane receptors, specific HN–F interactions would elicit a conformational change in F protein that
would allow the fusion peptide to be exposed and to interact with the target membrane, thereby triggering the fusion cascade [14,36]. Moreover, even a weakened interaction between HN and F is still sufficient to trigger fusion [37]. To date, the nature of these interactions remains to be elucidated, but from the data presented here it could be argued that the decrease in the fusogenic activity of NDV proteins after reconstitution would be due to the disappearance of functional interactions between HN and F proteins during the solubilization-reconstitution cycle. For most paramyxoviruses, including NDV and with the exception of simian virus 5 [12,38] and respiratory syncytial virus [39], it has been reported that the expression of F protein in the absence of HN protein does not induce cell–cell fusion. The HR1 and HR3 domains seem to be involved in the HN-independent fusing activity of the simian virus 5
C. Cobaleda et al. / The International Journal of Biochemistry & Cell Biology 34 (2002) 403–413
411
Fig. 5. Kinetic fusion parameters for F-envelopes under different conditions. Relative values of the kinetic constants A1 (asymptotic value of specific fusion) and k1 (rate of specific fusion) (Eq. (2)) for F-envelopes-erythrocyte ghosts fusion. Data were calculated as percentages of those of a control experiment in which F-envelopes were incubated at 37 ◦ C, pH 7.4, in the presence of 130 g of target membrane proteins (optimal conditions) taken as 100% (control). Neu5Ac2en, virosomes in the presence of 0.1 mM of the sialidase inhibitor. pH 4, pH 10, and 4 ◦ C, F-envelopes were incubated in the presence of 130 g of target membrane proteins at the indicated pH or temperature values, respectively. Glut indicates treatment of virosomes with glutaraldehyde 5%, as described in Section 2. Data are from one representative experiment.
F protein [40]. In NDV, a single amino acid change in the ectodomain of F protein (HR3 region) modifies the requirement for HN protein in syncytium formation [9]. In this sense, in the present report we detected fusion activity in HN-lacking virosomes. The residual fusion detected in F-envelopes is difficult to explain because the receptor-binding protein was not present. Several possibilities could be advanced in this regard: (i) the presence in the reconstituted systems of an undetected binding agent; (ii) the presence of residual amounts of HN protein, not detected by SDS-PAGE (Fig. 1); (iii) the F protein of NDV could have a residual F activity independent of HN protein; and (iv) in the reconstituted systems, F could escape the requirements of the attachment protein. After the reconstitution procedure used here, NDV F protein can acquire a given conformation that would allow it to develop attachment activity in absence of HN
protein [9]. In this context, for simian virus 5 it has been proposed that the F protein might interact with a specific receptor of the target membrane before the triggering of the fusion cascade, independently of the attachment function of HN protein [41]. Thus, the reconstituted envelopes studied here retained certain fusion activity, dependent only on the F protein, as demonstrated by the fact that this activity could not be inhibited by Neu5Ac2en (Fig. 2). The residual fusing activity present in the three reconstituted systems (including HN-lacking envelopes, i.e. F-envelopes) suggests that a small population of F proteins escapes the need for HN protein for fusion; this was also detected in intact viruses in the presence of the sialidase inhibitor Neu5Ac2en (Fig. 2 and [24]). The effect of the inhibitor on the fusion activity of intact NDV supports the idea that HN neuraminidase activity may affect F protein-mediated
412
C. Cobaleda et al. / The International Journal of Biochemistry & Cell Biology 34 (2002) 403–413
membrane fusion, although the possibility that this agent might alter HN fusion promotion itself cannot be ruled out. It has been reported [42] that, in NDV, HN fusion promotion does not correlate with sialidase activity since different HN mutants show changes in sialidase activity without fusion promotion activity being affected, and vice versa. Other studies [43] have suggested that the high receptor-destroying activity of HN might reduce the fusion promotion activity of the protein. Therefore, the role of neuraminidase activity in paramyxovirus F protein-mediated membrane fusion should be analyzed in more detail. Although we have failed to reconstitute fully functional NDV envelopes as compared with the fusion activity of intact viruses, the residual F activity detected here allows us to draw some conclusions about HN–F interactions in NDV-mediated fusion. NDV-derived virosomes proved to be 10 times less fusogenic than intact viruses. Our theory is that in HN–F–M and HN–F reconstituted envelopes, HN fails to activate F protein and to promote its fusion activity. In sum, the loss of fusion promotion detected here could be due to several reasons, in agreement with the proposals of Lamb [14] and Horvath et al. [12]: (i) after reconstitution, HN–F physical interactions are modified or destroyed, as discussed above; (ii) in the reconstituted systems, the ratio and/or distance between both glycoproteins in the membrane of the virosomes are different from those of intact virions which would not allow the HN fusion promotion signal. Further experiments must be performed in order to gain direct evidence in support of these hypotheses. In any case, the present work is a further contribution to the currently accepted view of the need for HN protein in paramyxovirus membrane fusion (revised by [14]), considering that the loss of the HN fusion promotion activity proposed here is due to the rupture of HN–F interactions that occur in the intact virus. Since the sialidase and hemagglutinating specific activities of HN in reconstituted systems (HN–F envelopes) were similar to those of intact virions, HN fusion promotion activity seems to be a more critical activity than the other two.
Acknowledgements This work was supported by Grants fom the Spanish DGES (PM97-0160) and the Junta de Castilla y
León (SA 62/99) to E.V. The authors thank E. Diaz de Espada and José Antonio Rodr´ıguez Hernández, from Laboratories Intevret, Salamanca, Spain, for providing the lentogenic “Clone-30” NDV strain. C.Cobaleda and A.Sagrera were fellowship holders of the Plan de Formación del Profesorado Universitario y Personal Investigador, Ministerio de Educación y Cultura, Spain.
References [1] P.W. Choppin, R.W. Compans, in: H. Fraenkel-Conrat, R.R. Wagner (Eds.), Comprehensive Virology, Plenum Press, New York, 1975, pp. 95–178. [2] A. Garc´ıa-Sastre, J.A. Cabezas, E. Villar, Proteins of newcastle disease virus envelope: interaction between the outer hemagglutinin-neuraminidase glycoprotein, Biophys. Acta 999 (1989) 171–175. [3] K. San Román, E. Villar, I. Muñoz-Barroso, Acidic pH enhancement of the fusion of newcastle disease virus with cultured cells, Virology 260 (1999) 329–341. [4] T.G. Morrison, A. Portner, in: H. Fraenkel-Conrat, R.R. Wagner (Eds.), The Paramyxoviruses, Plenum Press, New York, 1991, pp. 347–382. [5] S. Iwata, A.C. Schmidt, K. Titani, M. Suzuki, H. Kido, B. Gotoh, M. Hamaguchi, Y. Nagai, Assignment of disulfide bridges in the fusion glycoprotein of Sendai virus, J. Virol. 68 (1994) 3200–3206. [6] R. Brasseur, M. Vandenbranden, B. Cornet, A. Burny, J.M. Ruysschaert, Orientation into the lipid bilayer of an asymmetric amphipathic helical peptide located at the N-terminus of viral fusion proteins, Biochim. Biophys. Acta 1029 (1990) 267–273. [7] P. Chambers, C.R. Pringle, A.J. Easton, Heptad repeat sequences are located adjacent to hydrophobic regions in several types of virus fusion glycoproteins, J. Gen. Virol. 71 (1990) 3075–3080. [8] R. Buckland, F. Wild, Leucine zipper motif extends, Nature 338 (1989) 547. [9] T.A. Sergel, L.W. McGinnes, T.G. Morrison, A single amino acid change in the newcastle disease virus fusion protein alters the requirement for HN protein in fusion, J. Virol. 74 (2000) 5101–5107. [10] I. Muñoz-Barroso, C. Cobaleda, G. Zhadan, V. Shnyrov, E. Villar, Dynamic properties of newcastle disease virus envelope and their relations with viral hemagglutinin-neuraminidase membrane glycoprotein, Biochim. Biophys. Acta 1327 (1997) 17–31. [11] T. Morrison, C. McQuain, L. McGinnes, Complementation between avirulent newcastle disease virus and a fusion protein gene expressed from a retrovirus vector: requirements for membrane fusion, J.Virol. 65 (1991) 813–822. [12] C.M. Horvath, R.G. Paterson, M.A. Shaughnessy, R. Wood, R.A. Lamb, Biological activity of paramyxovirus fusion
C. Cobaleda et al. / The International Journal of Biochemistry & Cell Biology 34 (2002) 403–413
[13]
[14] [15]
[16]
[17]
[18]
[19]
[20]
[21]
[22]
[23]
[24]
[25]
[26]
[27]
[28]
[29]
proteins: factors influencing formation of syncytia, J. Virol. 66 (1992) 4564–4569. J. Stone-Hulslander, T.G. Morrison, Mutational analysis of heptad repeats in the membrane-proximal region of newcastle disease virus HN protein, J. Virol. 73 (1999) 3630–3637. R.A. Lamb, Paramyxovirus fusion: a hypothesis for changes, Virology 197 (1993) 1–11. J. Stone-Hulslander, T.G. Morrison, Detection of an interaction between the HN and F proteins in newcastle disease virus-infected cells, J. Virol. 71 (1997) 6287–6295. S. Tong, R.W. Compans, Alternative mechanisms of interaction between homotypic and heterotypic parainfluenza virus HN and F proteins, J. Gen. Virol. 80 (1999) 107–115. T.L. Steck, J.A. Kant, Preparation of impermeable ghost and inside–out vesicles from human erythrocyte membranes, Meth. Enzymol. 31 (1974) 172–180. A. Scheid, L.A. Caliguiri, R.W. Compans, P.W. Choppin, Isolation of paramyxovirus glycoproteins. Association of both hemagglutinating and neuraminidase activities with the larger SV5 glycoprotein, Virology 50 (1972) 640–652. A. Scheid, P.W. Choppin, Isolation and purification of the envelope proteins of newcastle disease virus, J. Virol. 11 (1973) 263–271. M. Tomasi, A. Loyter, Selective extraction of biologically active F-glycoprotein from dithiothreitol-reduced Sendai virus particles, FEBS Lett. 131 (1981) 381–385. S. Bagai, A. Puri, R. Blumenthal, D.P. Sarkar, Hemagglutininneuraminidase enhances protein-mediated membrane fusion of reconstituted Sendai virus envelopes with cells, J. Virol. 67 (1993) 3312–3318. A. Vainstein, M. Hershkovitz, S. Israel, S. Rabin, A. Loyter, A new method for reconstitution of highly fusogenic Sendai virus envelopes, Biochim. Biophys. Acta 773 (1984) 181–188. P.W. Holloway, A simple procedure for removal of Triton X-100 from protein samples, Anal. Biochem. 53 (1973) 304– 308. C. Cobaleda, A. Garc´ıa-Sastre, E. Villar, Fusion between newcastle disease virus and erythrocyte ghosts using octadecyl rhodamine B fluorescence assay produces dequenching curves that fit the sum of two exponentials, Biochem. J. 300 (1994) 347–354. D. Hoekstra, T. de Boer, K. Klappe, J. Wilschut, Fluorescence method for measuring the kinetics of fusion between biological membranes, Biochemistry 23 (1984) 5675–5681. E.G. Bligh, W.J. Dyer, A rapid method of total lipid extraction and purification, Can. J. Biochem. Physiol. 37 (1959) 911– 917. D.L. Wright, D.T. Plummer, Multiple forms of acetylcholinesterase from human erythrocytes, Biochem. J. 133 (1973) 521–527. W.G. Bardsley, N.A.J. Bukhari, M.W.J. Ferguson, J.A. Cachaza, F.J. Burguillo, Evaluation of model discrimination, parameter estimation and goodness of fit in non-linear regression problems by test statistics distributions, Comput. Chem. 19 (1995) 75–84. A. Helenius, K. Simons, Solubilization of membranes by detergents, Biochim. Biophys. Acta 415 (1975) 29–79.
413
[30] K. Kawasaki, S.B. Sato, S.I. Ohnishi, Membrane fusion activity of reconstituted vesicles of influenza virus hemagglutinin glycoproteins, Biochim. Biophys. Acta 733 (1983) 286– 290. [31] D.J. Volsky, A. Loyter, An efficient method for reassembly of fusogenic Sendai virus envelopes after solubilization of intact virions with Triton X-100, FEBS Lett. 92 (1978) 190–194. [32] A. Helenius, M. Sarvas, K. Simons, Asymmetric and symmetric membrane reconstitution by detergent elimination, Eur. J. Biochem. 116 (1981) 27–35. [33] W.A. Petri, R.R. Wagner, Reconstitution into liposomes of the glycoprotein of vesicular stomatitis virus by detergent dialysis, J. Biol. Chem. 254 (1979) 4313–4316. [34] M. Paternostre, M. Viard, O. Meyer, M. Ghanam, M. Ollivon, R. Blumenthal, Solubilization and reconstitution of vesicular stomatitis virus envelope using octylglucoside, Biophys. J. 72 (1997) 1683–1694. [35] A. Garc´ıa-Sastre, C. Cobaleda, J.A. Cabezas, E. Villar, On the inhibition mechanism of the sialidase activity from newcastle disease virus, Biol. Chem. Hoppe-Seyler 372 (1991) 923–927. [36] K.A. Baker, R.E. Dutch, R.A. Lamb, T.S. Jardetzky, Structural basis for paramyxovirus-mediated membrane fusion, Mol. Cell 3 (1999) 309–319. [37] R. Deng, Z. Wang, P.J. Mahon, M. Marinello, A.MirzaM. Iorio, Mutations in the newcastle disease virus hemagglutinin-neuraminidase protein that interfere with its ability to interact with the homologous F protein in the promotion of Fusion, Virology 253 (1999) 43–54. [38] M. Ito, M. Nishio, M. Kawano, S. Kusagawa, H. Komada, Y. Ito, M.M. Tsurodome, Role of a single amino acid at the amino terminus of the simian virus 5 F2 subunit in syncytium formation, J. Virol. 71 (1997) 9855–9858. [39] R.A. Karron, D.A. Buonagurio, A.F. Georgiu, S.S. Whiteheat, J.E. Adamus, M.L. Clements-Mann, D.O. Harris, V.B. Randolph, S.A. Udem, S.A.B.R. Burphy, M.S. Sidhu, Respiratory syncitial virus (RSV) SH and G proteins are not essential for viral replication in vitro: clinical evaluation and molecular characterization of a cold-passaged, attenuated RSV subgroup B mutant, Proc. Natl. Acad. Sci. U.S.A. 94 (1997) 13961–13966. [40] M. Ito, M. Nishio, H. Komada, Y. Ito, M. Tsurodome, An amino acid in the heptad repeat 1 domain is important for the haemagglutinin-neuraminidase-independent fusing activity of simian virus 5 fusion protein, J. Gen. Virol. 81 (2000) 719– 727. [41] R.E. Dutch, S. Bagai-Joshi, R.A. Lamb, Membrane fusion promoted by increasing surface densities of the paramyxovirus F and HN proteins: comparison for fusion reactions mediated by simian virus 5 F, human parainfluenza type 3 F, and influenza virus HA, J. Virol. 72 (1998) 7745–7753. [42] T. Sergel, L.W. McGinnes, T.G. Morrison, The fusion promotion activity of the NDV HN protein does not correlate with neuraminidase activity, Virology 196 (1993) 831–834. [43] T. Bousse, T. Takimoto, W.L. Gorman, T. Takahashi, A. Portner, Regions on the hemagglutinin-neuraminidase proteins of human parainfluenza virus type 1 and Sendai virus important for membrane fusion, Virology 204 (1994) 506– 514.