Conformational changes of Newcastle disease virus envelope glycoproteins triggered by gangliosides Laura Ferreira, Enrique Villar and Isabel Mun ˜ oz-Barroso Departamento de Bioquı ´ mica y Biologı ´ a Molecular, Universidad de Salamanca, Spain We have investigated the conformational changes of New- castle disease virus (NDV) glycoproteins in response to receptor binding, using 1,1-bis(4-anilino)naphthalene-5,5- disulfonic acid (bis-ANS) as a hydrophobicity-sensitive probe. Temperature- and pH-dependent conformational changes were detected in the presence of free bovine gan- gliosides. The fluorescence of bis-ANS was maximal at pH 5. The binding of bis-ANS to NDV was not affected by chemicals that denature the fusion glycoprotein, such as reducing agents, nor by the presence of neuraminidase inhibitors such as N-acetyl neuramicic acid. Gangliosides partially inhibited fusion and hemadsorption, but not neuraminidase hemagglutinin-neuraminidase glycoprotein (HN) activity. A conformational intermediate of HN, trig- gered by the presence of gangliosides acting as receptor mimics, was detected. Our results indicate that, upon binding to free gangliosides, HN undergoes a certain conformational change that does not affect the fusion glycoprotein. Keywords: NDV; bis-ANS; conformational intermediates; paramyxovirus receptors; gangliosides. Newcastle disease virus (NDV) is an avian enveloped virus belonging to the family of Paramyxoviridae, genus Avulavirus. The membrane contains two transmembrane glycoproteins, hemagglutin-neuraminidase (HN) and the fusion (F) protein [1]. HN binds to sialic acid-containing receptors at the cell surface through its hemagglutinating activity (receptor-binding activity) and it also displays neuraminidase or sialidase activity (receptor-destroying activity), which probably prevents the aggregation of the viral progeny. In addition, a third activity (the so-called fusion promotion activity) has been proposed for the HN protein [2–4]. The F protein is directly responsible for the fusion between the viral envelope and the target mem- brane. For paramyxoviruses, the fusion mechanism has been proposed to occur at neutral pH; nevertheless, we have previously shown that the fusion of NDV with cultured cells is enhanced at acidic pH [5]. The F protein is produced as a single inactive polypeptide, F o ,which, once cleaved by a cellular protease (reviewed in [6]), becomes the active F 1 -F 2 form, with two peptides linked by a disulfide bond [7]. To date, three domains of the F 1 polypeptide have been suggested to be involved in the fusion mechanism of NDV. These are the N-terminal fusion peptide [8] and two heptad repeat (HR) regions of the ectodomain, one (HR1) located adjacent to the fusion peptide, and the other (HR2) at the C-terminal adjacent to the transmembrane domain [9–11]. Once activated, the F protein is thought to undergo a series of conformational changes that result in exposure of the fusion peptide and interaction of the HR1 and HR2 domains. A six-helix bundle has emerged as the fusion core structure of many viral fusion proteins, the N-terminal HR forming the inner core, surrounded by antiparallel C-terminal helices along the grooves located between the helices of the central HR coiled-coil. The formation of this structure is believed to pull the viral and cell membranes into close proximity for merging. The complete mechanism of NDV-induced membrane fusion remains unknown. As with many other paramyxo- viruses, NDV needs type-specific HN–F interactions that must be present in the same bilayer to induce fusion (reviewed in [12]). It has been proposed that the interaction of HN with the cellular receptor induces conformational changes in the HN protein that activates the F protein [12], although the nature of such changes is obscure. In the present study we analyzed the possible conform- ational changes ocurring in NDV envelope glycoproteins when interacting with free gangliosides as receptor mimics. These changes were revealed through use of the fluorescent probe 1,1-bis(4-anilino)naphthalene-5,5-disulfonic acid (bis-ANS), which is nonfluorescent in aqueous solution but increases its quantum yield when bound to hydrophobic groups [13,14]. We observed that bis-ANS fluorescence was maximal at 37 °C and at acidic pH. As reduction of the disulfide bond of the F protein did not affect bis-ANS Correspondence to I. Mun ˜ oz-Barroso and E. Villar, Departamento de Bioquı ´ mica y Biologı ´ a Molecular, Universidad de Salamanca, Edificio Departamental Laboratory 108, Plaza Doctores de la Reina s/n, 37007 Salamanca, Spain. Fax: + 34 923 294579, Tel.: + 34 923 294465, E-mail: imunbar@usal.es and evillar@usal.es Abbreviations: bis-ANS, 1,1-bis(4-anilino)naphthalene-5,5-disulfonic acid; DMEM, Dulbecco’s modified Eagle’s medium; F protein, fusion glycoprotein; FDQ, fluorescence dequenching; HA, influenza hemagglutinin; Had, hemadsorption; HN, hemagglutinin- neuraminidase glycoprotein; HR, heptad repeat; KNP, 120 m M KCl, 30 m M NaCl, 10 m M sodium phosphate pH 7.4; NDV, Newcastle disease virus; NeuAc, N-acetylneuraminic acid; p.f.u., plaque formation units; R 18 , octadecylrhodamine B chloride. (Received 7 November 2003, accepted 9 December 2003) Eur. J. Biochem. 271, 581–588 (2004) Ó FEBS 2004 doi:10.1111/j.1432-1033.2003.03960.x fluorescence, we suggest that the binding of NDV to free gangliosides results in the formation of a conformational intermediate of HN. Materials and methods Materials Bis-ANS and octadecyl rhodamine B chloride (R 18 )were from Molecular Probes Inc. (Junction City, OR, USA). Bovine brain gangliosides, disialoganglioside GD1a, lacto- cerebrosides, dithiothreitol, 2-mercaptoethanol, N-acetyl- neuraminic acid and Triton X-100 were all from SIGMA (St. Louis, MO, USA). Cell culture media were from BIO Whittaker (Walkersvile, Maryland, USA). Fresh blood from healthy donors (with their consent) was obtained from the Blood Bank of the University Hospital in Salamanca (Spain). Cells and viruses NDV ÔClone 30Õ was grown and purified essentially as described elsewhere [15]. COS-7, HeLa and Vero cells were obtained from the American Type Culture Collection and were maintained in Dulbecco’s Modified Eagle Medium (DMEM) supplemented with L -glutamine (580 mgÆL )1 ), penicillin/streptomycin (100 UÆmL )1 /100 lgÆmL )1 ), and heat-inactivated fetal bovine serum at 10% (v/v) for COS-7 and HeLa cells and at 5% (v/v) for Vero cells. For the fusion experiments, COS-7 cells grown in monolayers were detached with trypsin/EDTA. Trypsin was inactivated by the addition of DMEM. The cells were washed twice with 15 m M Hepes buffer (130 m M NaCl, 5m M KCl, 2 m M CaCl 2 ,1m M MgCl 2 ,10m M glucose), pH 7.4, and resuspended at 2.5 · 10 6 cells in 200 lLof Hepes buffer, pH 7.4. Ganglioside-induced conformational changes Bis-ANSwasaddedto20lgofNDVat3l M in Hepes buffer (final volume, 2 mL) under constant stirring in the fluorimeter cuvette. Then, different concentrations of gan- gliosides were added, and the fluorescence progress curve was recorded for 2 min on a Hitachi F-4010 spectrofluo- rimeter (excitation, 395 nm; emission, 500 nm; slit widths 5 and 10 nm for excitation and emission, respectively). The background fluorescence resulting from gangliosides was calculated. To analyse the bis-ANS data, the relative fluorescence (I rel ) was calculated according to [16]: I relðtÞ ¼ðI ðtÞ À I NDVþbis-ANS Þ=ðI maxBG ÀI bis-ANS Þð1Þ where I (t) is the fluorescence intensity at any given time, I NDV + bis-ANS is the fluorescence intensity of bis-ANS in the presence of the virus, I maxBG is the final fluorescence intensity of bis-ANS in the presence of gangliosides, and I bis-ANS is the fluorescence intensity of bis-ANS in aqueous solution. R 18 fusion assays Dequenching fusion assays were accomplished as described previously [5]. Briefly, purified NDV was labeled with the fluorescent probe octadecylrhodamine (R 18 )andafter mixing with the target cells, the progress curve of R 18 fluorescence was spectrofluorimetrically monitored. Neuraminidase assays Neuraminidase activity was determined by a fluorimetric procedure using 2¢-(4-methylumbelliferyl)a- D -N-acetyl- neuraminic acid as substrate [17]. Hemadsorption assays The hemadsorption (Had) activity of HN protein was determined according to [18] with modifications. HeLa cells were plated in 24-well plate 12 h before infection. Then, the cell monolayers were infected with NDV at 1 multiplicity of infection. At 24 h postinfection, the cells were washed twice with NaCl/P i (10 m M KH 2 PO 4 ,150m M NaCl, pH 7.2) and incubated for 30 min at 4 °C with 2% of human erythro- cytes. After washing twice with NaCl/P i , adsorbed erythro- cytes were lysed in 50 m M NH 4 Cl for 5 min at 4 °C. The lysates were clarified by centrifugation and Had activity was quantified by measuring the absorbance at 540 nm and subtracting the background absorbance obtained with uninfected cells. To analyse the effect of gangliosides on Had activity, the cells were incubated in the presence of different concentrations of gangliosides for 10 min at 37 °C before the addition of red blood cells. Results and discussion It has been shown previously that the exposure of hydro- phobic regions of viral proteins as a consequence of conformational changes can be analyzed by means of the hydrophobic-sensitive probe bis-ANS [14,16,19]. This water-soluble fluorophore undergoes a strong increase in its quantum yield upon binding to hydrophobic sites [16] and has been used to study protein structural changes [14,16,19,20]. The exposure of hydrophobic segments of NDV envelope proteins triggered by gangliosides was tested by measuring binding to bis-ANS. Initially, the effect of increasing concentrations of NDV or gangliosides on bis- ANS fluorescence was studied. For these experiments, at zerotimetheviruswasaddedto37°C-prewarmed Hepes buffer at pH 7.4 in the fluorimeter cuvette, followed by the addition of 3 l M of bis-ANS. The fluorescence emitted, taken as I NDV+bisANS , became stablilized after a few seconds. Then, gangliosides were added and the progress curve of the fluorescence emission at 500 nm was recorded for 2 min. To process the data, I rel was calculated according to Eqn (1). I NDV+bis-ANS was subtracted from the intensity of fluorescence observed at any given time and this was relatedtothetermI maxBG –I bis-ANS , i.e. the maximal fluorescence of gangliosides in the absence of virus (I maxBG ) after subtracting the fluorescence emission of the probe in buffer (I bis-ANS ). Figure 1 shows the dose–response curves of bis-ANS fluorescence in the presence of different concentrations of NDV. As can be seen, after 20 lgof NDV the fluorescence reached a plateau, suggesting saturation. Similarly, different concentrations of bovine brain gangliosides or of the disialoganglioside GD1a 582 L. Ferreira et al.(Eur. J. Biochem. 271) Ó FEBS 2004 (5–30 lg) were added to 20 lgofNDVinthepresenceof bis-ANS (Fig. 2), and it was found that 10 lg of bovine brain gangliosides and 5 lg of GD1a were sufficient to observe saturation under the conditions of the experiment. The observed saturation of the extent of bis-ANS fluores- cence (Figs 1 and 2) may indicate that the conformational change undergone by NDV glycoproteins in the presence of gangliosides is limited (see below). To assess the specific effect of gangliosides, 20 lg of NDV was preincubated in the presence of 10 lgofbovine gangliosides at 37 °C for 10 min. Then, an additional 10 lg was added and no increase in bis-ANS fluorescence was detected above background (Table 1). We interpret these results as pointing to the irreversibility of the conformational change triggered by gangliosides. In addi- tion, neutral glycolipids such as lactocerebrosides did not lead to an increase in the fluorescence of bis-ANS (Table 1). The temperature-dependence of bis-ANS fluorescence at neutral pH in the presence of NDV and gangliosides was analyzed (Fig. 3). The fluorescence of bis-ANS in the Fig. 1. Effect of NDV concentration on bis-ANS fluorescence. At zero time, 3 l M bis-ANSwasaddedto37°C prewarmed buffer containing different concentrations of NDV, after which 25 lg of bovine gangliosides was added. Fluorescence was recorded continuously over 2 min at excitation and emission wavelengths of 395 and 500 nm, respectively. The relative fluorescence, I rel , is shown (see Materials and methods). (A) Kinetics of bis-ANS fluorescence at different NDV concentrations from a representative experiment. (B) Relative fluorescence of bis-ANS, I rel ,at 90 s of reaction at the desired NDV concentration. Data taken from different experiments similar to that shown in (A). Data are means ± SE of at least three independent experiments. Fig. 2. Effect of ganglioside concentrations on bis-ANS fluorescence. At zero time, 3 l M bis-ANSwasaddedto37°C prewarmed buffer containing 20 lg of NDV, after which different concentrations of bovine brain gangliosides or GD1a were added. Fluorescence was recorded continuously for 2 min at excitation and emission wavelengths of 395 and 500 nm, respectively. The relative fluorescence, I rel , is shown (see Materials and methods). (A) Kinetics of bis-ANS fluorescence at different ganglioside concentrations from a representative experiment. (B) Relative fluorescence of bis- ANS, I rel , at 90 s of reaction at the desired ganglioside concentration; (d), bovine brain gangliosides; (m), GD1a. Data taken from different experiments similar to that shown in (A). Data are means ± SE of at least two independent experiments. Table 1. Effect of preincubation of NDV with different agents on the fluorescence of bis-ANS. NDV (20 lg) was incubated in the presence of 10 lg of bovine brain gangliosides, 10 m M NeuAc or 50 m M NeuAc for 10 min at 37 °C. Then, 3 l M bis-ANSwasaddedtoprewarmed buffer at 37 °C containing 20 lg of treated virus, after which 10 lgof bovine gangliosides or 10 m M NeuAcwereaddedfortriggeringthe conformational change. Fluorescence was recorded continuously over 2 min at excitation and emission wavelengths of 395 and 500 nm, respectively. I rel at90minofreactionareshown(seeMaterialsand methods). Preincubation Trigger of the conformational change I rel – Bovine gangliosides (10 lg) 6.98 – NeuAc (10 m M ) 15.45 – Lactocerebrosides (10 lg) 0.85 Bovine gangliosides (10 lg) Bovine gangliosides (10 lg) 0.6 Bovine gangliosides (10 lg) NeuAc (10 m M ) 12.71 NeuAc (10 m M ) Bovine gangliosides (10 lg) 8.31 NeuAc (50 m M ) NeuAc (10 m M ) 1.97 Ó FEBS 2004 Conformational changes in NDV glycoproteins (Eur. J. Biochem. 271)583 presence of NDV but in the absence of gangliosides [I NDV+bisANS from Eqn (1)], was independent of tempera- ture (data not shown). No increase in fluorescence was observed at 4 °C, whereas it increased gradually at 15 and 25 °C, showing a sharper increase after 30 °C. These data are comparable to those of the temperature-dependence of NDV fusion with cultured cells reported by us previously [5]. In both cases, we failed to detect an increase in fluorescence at 4 °C. Nevertheless, it has been established that the HN protein of paramixoviruses can bind to the sialoglycosides of the cell surface at 4 °C [21], suggesting that NDV may interact with gangliosides at this tempera- ture. However, this binding seems to be insufficient to trigger any conformational change in NDV glycoproteins detectable with the bis-ANS technique. The pH-dependence of the fluorescence of bis-ANS in the presence of NDV and gangliosides was also analyzed (Fig. 4).Atzerotime,NDVwasaddedto37 °C-prewarmed buffer at the desired pH, followed by the addition of 3 l M bis-ANS and then gangliosides; next I rel was calculated as described above. Figure 4A shows the kinetics of bis-ANS fluorescence in the presence of NDV and gangliosides at different pH values; Fig. 4B depicts the final extent of bis- ANS fluorescence after 90 s of virus–ganglioside contact. The extent of bis-ANS fluorescence at the different pH values assayed occurred in the following order: pH5>pH5.5>pH6.5>pH7.Ithasbeenreported that the increase in fluorescence at low pH can be partly explained in terms of the protonation of negatively charged groups, which facilitates the binding of bis-ANS [14]. In this sense, we detected a slight increase in the fluorescence of bis-ANS in the presence of NDV upon lowering the pH [I NDV+bis-ANS in Eqn (1)], but these figures were subtracted from the fluorescence intensity emitted in the presence of gangliosides (Eqn 1). In Fig. 4B, a sharp increase in fluorescence intensity at pH < 6.5 can be seen. The fluorescence intensity observed at pH 5 was about twice the value seen at pH 7.4. This difference is smaller than that reported for viruses that show a pH-dependent entry mechanism, since for influenza virus Korte and Herrman (1994) [14] have reported that bis-ANS fluorescence is five times higher at acidic than at neutral pH. Our data on the Fig. 4. pH-dependence of the bis-ANS fluorescence. At zero time, 3 l M bis-ANSwasaddedto37 °C prewarmed buffer at the desired pH containing 20 lg of NDV, after which 15 lg of bovine gangliosides was added. Fluorescence was recorded continuously over 2 min at excitation and emission wavelengths of 395 and 500 nm, respectively. The relative fluorescence, I rel , is shown (see Materials and methods). (A) Kinetics of bis-ANS fluorescence at different pHs from a representative experiment (B) Relative fluorescence of bis-ANS, I rel , at 90 s of reaction at different pHs. Data are means ± SE of two independent experiments. Fig. 3. Temperature-dependence of bis-ANS fluorescence. At zero time, 3 l M bis-ANS was added to buffer prewarmed to the desired temperature containing 25 lg of NDV, after which 10 lg of bovine gangliosides was added. Fluorescence was recorded continuously over 2 min at excitation and emission wavelengths of 395 and 500 nm, respectively. The relative fluorescence, I rel , is shown (see Materials and methods). (A) Kinetics of bis- ANS fluorescence at different temperatures from a representative experiment (B) Relative fluorescence of bis-ANS, I rel , at 90 s of reaction at the desired temperature. Data are means ± SE of two independent experiments. 584 L. Ferreira et al.(Eur. J. Biochem. 271) Ó FEBS 2004 pH-dependence of bis-ANS fluorescence indicate that the conformation of NDV proteins triggered at acidic pH exposes a higher number of hydrophobic fluorophore- binding sites. It is interesting to note the similarities between the pH-dependence of NDV fusion activity reported previously by us [5] and that of bis-ANS fluorescence, pointing to the maximal extent of both fusion and bis-ANS fluorescence at pH 5.0. We have previously hypothesized [5] that NDV might use the endocytic pathway as a secondary mechanism of entry. If the conformational change under- gone by HN protein after receptor binding (see below) is activated at acidic pH, as well as NDV fusion activity, the present data confirm our hypothesis concerning the acidic pH enhancement of NDV entry. Moreover, the pH- dependence of viral entry seems debatable. In this sense, Mothes et al. [22] have reported that the entry of the avian leukosis virus, a retrovirus, into the host cell depends on a low pH step that acts after receptor binding. For these authors, partial conformational changes in env protein in the presence of soluble receptors may be due to receptor priming rather than complete activation. Additionally, it has been recently reported [23] that the SER paramyxovirus shows a low-pH-dependent fusion activity. We performed a series of experiments to elucidate whether the binding of bis-ANS to hydrophobic sites of NDV glycoproteins was located in F and/or HN protein. First, NDV was incubated in the presence of 10 m M 2-mercaptoethanol or 2 m M dithiothreitol (agents that reduce the disulfide bonds of the F protein) for 30 min at 37 °C before the addition of bis-ANS and gangliosides. As deduced by PAGE analysis (data not shown), treatment of viruses with 2-mercaptoethanol led to the loss of F 0 protein. In another series of experiments, viruses were incubated in the presence of 50 m M of N-acetylneuraminic sialic acid (NeuAc)for30minat37°C. This compound is both a product and an inhibitor of the neuraminidase activity of the HN protein through binding to its active site [17]. Neither treatment affected the emission of bis-ANS fluor- escence with respect to the control (NDV without treat- ment) when gangliosides were added to treated viruses in the bis-ANS assay (Table 1 and data not shown). To test the possibility that bis-ANS might bind nonspecifically to 2-mercaptoethanol-treated-virus, we performed the follow- ing experiment. Twenty micrograms of virus, both treated and nontreated with the reducing agent, were preincubated in the presence of 10 lg of bovine gangliosides for 10 min at 37 °C. Then, a further 10 lg of gangliosides was added in the bis-ANS assay. In both cases, no increase in bis-ANS fluorescence was detected above the background level, unlike the findings on treated virus not preincubated in the presence of gangliosides (data not shown). We therefore assume that the fluorescence of bis-ANS of reduced virus in the presence of gangliosides would not be due to the nonspecific binding of the probe to 2-mercaptoethanol- treated-NDV. Because viruses treated with these reducing agents are fusion-deficient (data not shown), this seems to indicate that the newly exposed hydrophobic binding sites are not located within the F protein. On the other hand, the increase in bis-ANS fluorescence did not vary after pre- incubation with NeuAc, suggesting that the binding of gangliosides, the putative agents of the conformational change, did not compete with the neuraminidase inhibitor NeuAc. To test this hypothesis, we performed a direct binding assay between the sialic acid NeuAc and NDV using the bis-ANS technique. Our data indicate that, similarly to gangliosides, NeuAc leads to an increase in the fluorescence of bis-ANS in the presence of NDV (Table 1). We performed a series of experiments to analyze the relationship between ganglioside and NeuAc binding sites. NDV was incubated in the presence of 10 lg of bovine brain gangliosides or 10 m M NeuAc for 10 min at 37 °C. Then, 3 l M bis-ANS was added to prewarmed buffer at 37 °C containing 20 lg of treated virus, after which an additional 10 lg of bovine gangliosides or 10 m M NeuAc were added to trigger the conformational change. Our results revealed that preincubation of NDV in the presence of NeuAc did not abolish the increase in fluorescence when gangliosides were added, but it did abolish it when additional NeuAc was added. By contrast, preincubation of NDV in the presence of gangliosides did not abolish the increase in fluorescence when NeuAc was added, but it did so when additional gangliosides were added (Table 1). Our conclusion is that the binding sites for NeuAc do not compete with the binding sites for gangliosides. As mentioned above, we detected the exposure of hydro- phobic binding sites of NDV proteins as measured by the increase in bis-ANS emission intensity (Figs 1–4), triggered by gangliosides. We assume that the new hydrophobic binding sites must belong to the envelope glycoproteins of NDV as the fluorophore shows a pronounced affinity for the hydrophobic sites of proteins in comparison with its affinity for lipids ([14] and references therein). The next step was to investigate whether the presence of gangliosides might exert some effect on NDV envelope glycoprotein activities. First, the fusion of NDV with COS-7 cells was analyzed by assaying the dequenching of the R 18 incorporated into the viral membrane (see Materials and methods). For this, 20 lg of R 18 -labeled NDV was incubated in the presence of 25 lg of bovine gangliosides for 10 min at 37 °C. Then, 2.5 · 10 6 COS-7 cells were added and the dequenching of R 18 fluorescence was recorded for 30 min. Data from a typical experiment are depicted in Fig. 5. As can be seen, fusion was not abolished although it was partially inhibited, showing an inhibition of the extent of fusion of about 27% as compared with controls at 30 min of virus–cell contact. This reduction was slightly lower if gangliosides were added to the virus–cell mixture (at time zero) without preincubation (20% as compared with control). As we observed that the denatur- ation of F protein by the cleavage of disulfide bonds did not exert any effect on bis-ANS fluorescence (data not shown), we assume that the partial inhibition of fusion exerted by gangliosides could be an indirect effect on fusion due to a certain inhibition of the virus binding to cells in the presence of gangliosides. As discussed below, gangliosides would bind to HN, lowering its interaction with COS-7 cell receptors and subsequently fusion of the virus with the cells. In addition, the ability of the gangliosides to inhibit HN hemadsorption activity was analyzed. NDV-infected HeLa cells were incubated in the presence of different concentrations of gangliosides for 1 h at 37 °C before the addition of red blood cells. As shown in Fig. 6, the data indicated that the inhibition of the Had activity of NDV HN protein exerted by bovine brain gangliosides was dose-dependent. Taken together, these results strongly suggest a specific interaction Ó FEBS 2004 Conformational changes in NDV glycoproteins (Eur. J. Biochem. 271)585 of the viral proteins with gangliosides, which would act as receptor mimics. In this sense, free gangliosides might compete with the actual receptors of the cell surface, inhibiting viral glycoproteins activities (Figs 5 and 6). Other simple molecules have previously been used to trigger conformational changes on viral receptors as soluble CD4 that induces certain conformational changes upon binding to envelope glycoproteins of HIV and SIV [20,24]. Viral HN glycoprotein has three different biological activities, sialidase or neuraminidase, hemagglutinating or receptor-binding, and fusion promotion. Although there is a considerable body of evidence both in favour of and against the topological separation of the neuraminidase and recep- tor-binding site ([18] and references therein), the crystal structure of NDV HN protein [25] supports the notion of a single site. Recently, on the basis of their crystallographic data on the HN protein of NDV, Crennell et al. [25] have proposed the existence of a single sialic acid recognition site switchable between both activities: the binding site or catalytic site. As summarized above, here we assayed (a) the effect of neuraminidase inhibitors on bis-ANS fluorescence and (b) the effect of gangliosides on neuraminidase and hemagglutinating activities. The extent of bis-ANS fluores- cence triggered by gangliosides was not affected by the neuraminidase inhibitor NeuAc (Table 1). Moreover, the presence of gangliosides did not exert any effect on the neuraminidase activity of HN protein (data not shown), although they did inhibit Had in a dose-dependent manner (Fig. 6). In addition, gangliosides and NeuAc did not compete for their binding sites in the bis-ANS assay (Table 1). Taken together, these data suggest that ganglio- sides bind to the receptor-binding site of HN protein and that this binding is not altered by the presence of neuraminidase inhibitors. Therefore, the data presented here together with those from our previous work [19,26] fail to account for the topological coincidence of both sites, although they do not allow us to propose their separation. In current models of membrane fusion induced by viral proteins, exposure of the fusion peptide that triggers membrane merging is a consequence of the conformational change of the F protein that involves the two heptad repeat regions (revised in [12]). The nature of these interactions and changes is not completely understood, although it has been established that for viruses that fuse with the target membrane through a pH-independent mechanism, such as most paramixoviruses and retroviruses, the conformational change of the F protein must be triggered after receptor binding. Upon comparing the 3D structure of HN, both alone and in a complex with the neuraminidase substrate 2-deoxy-2,3-dehydro-N-acetylneuraminic acid, Takimoto et al. [27] suggested that receptor binding induces a structural change in the hydrophobic surface of the HN protein that disrupts physical HN–F interactions, triggering the activation of F protein to initiate membrane fusion. Despite this, our results indicate that the binding alone of simple molecules is insufficient to induce strong HN conformational changes that would in turn affect the F protein. In other words, the conformational changes induced by free receptor mimic molecules are only partial. As we have shown here, NDV glycoproteins undergo conformational changes in the presence of gangliosides, as indicated by the exposure of new hydrophobic binding sites for the bis-ANS probe. Our data strongly support the idea that binding to these receptor mimics induces a conform- ational change in HN protein. Our observation that inactivation of the F protein did not affect the extent of bis-ANS fluorescence suggests that the fusion protein does not undergo any conformational change in the presence of gangliosides. Therefore, functional HN–F interactions in vivo, i.e. interactions that drive fusion, may need a more Fig. 5. Effect of bovine gangliosides on NDV fusion with COS-7 cells. R 18 -labeled NDV (20 lg) was incubated in the presence of 25 lgof bovine gangliosides for 10 min at 37 °C. Then, 2.5 · 10 5 COS-7 cells were added and the sample was incubated at 37 °C for 30 min under continuous stirring. Fusion was monitored continuously, as described in Materials and methods, by measuring the dequenching of R 18 .(d) Control; (m) virus and gangliosides with preincubation; (j)virusand gangliosides without preincubation. Fig. 6. Dose–response effect of ganglioside inhibition of hemadsorption. HeLa cells were infected with NDV at 1 multiplicity of infection. At 24 h postinfection, the cells were incubated in the presence of different concentrations of gangliosides for 10 min at 37 °C and then incubated for 30 min at 4 °C with 2% of human erythrocytes. The rate of hemadsorption in comparison with controls was calculated by meas- uring the absorbance at 540 nm of the erythrocytes bound to NDV- infected cells after lysing in 50 m M NH 4 Cl. Data are means ± SE of two independent experiments. 586 L. Ferreira et al.(Eur. J. Biochem. 271) Ó FEBS 2004 complex environment than the presence alone of a putative receptor, in this case gangliosides. In addition, the con- formational change undergone by the HN protein after binding to gangliosides can be completed in the presence of the correct target, i.e. the cell membrane. The binding of viruses to the host cell surface is a more complex phenom- enon than a mere bimolecular interaction between a viral protein and a cellular receptor. In this sense, viral binding may occur through multiple interactions among several viral and cellular molecules, accompanied by conforma- tional changes in viral proteins. Therefore, a major task would be to study the conformational changes of NDV proteins in the presence of cells. Nevertheless, the high extent of bis-ANS binding to hydrophobic sites of the cell surface did not allow us to use this assay with intact cells as targets (data not shown). The existence of conformational intermediates for viral proteins such as influenza HA [16], vesicular stomatitis virus fusion protein [19,28] or HIV envelope glycoproteins [20] has been reported. Additionally, the binding of bis- ANS to different viral glycoproteins [14,16,19,20] has been correlated with the fusion activation of the proteins. Nevertheless, here we observed a conformational inter- mediate of the HN protein prior to membrane merging, confirming that changes leading to fusion might be slow in the virus upon binding to the target membrane [16]. The newly exposed hydrophobic sequence of the HN protein triggered by gangliosides is not clear. We suggest several possibilities: (a) the HR stalk region, which has been proposed to be responsible for HN–F interactions [29]; (b) the interfaces of HN dimers, which presumably dissociate after ganglioside binding [27]; or (c) sequential conformational changes in HN protein, as proposed for other viral proteins [16]. In summary, here we have demonstrated that ganglio- sides bind to NDV, inducing the exposure of hydrophobic binding sites for bis-ANS. We propose that the binding site for gangliosides would be the receptor-binding site of HN protein, triggering the conformational change detected here. Our results indicate that the bis-ANS assay would also be useful for studying conformational changes in viral proteins that do not require an acidic pH to start fusion and that simple molecules such as gangliosides can be used as receptor mimics for triggering these changes. Acknowledgements This work was partially supported by the Spanish Fondo de Investigaciones Sanitarias, FIS (PI021848) and Junta de Castilla y Leo ´ n (SA 064/02) grants to E. V.; L. F. is a predoctoral fellowship supported by the Ministerio de Ciencia y Tecnologı ´ a, Spain (Grant DGES PM97-0160). We thank Drs E. Dı ´ ez Espada and J. A. Rodrı ´ guez from Intervet Laboratories (Salamanca, Spain) for providing the lentogenic ÔClone 30Õ strain of NDV. Thanks are also due to N. Skinner for language corrections and proofreading the manuscript. References 1. Choppin, P.W. & Compans, R.W. (1975) Reproduction of para- myxoviruses. In Comprehensive Virology (Fraenkel-Conrat,H.& Wagner, R.R., eds), pp. 95–178. Plenum Press, New York. 2. Morrison, T., McQuain, C. & McGinnes, L. (1991) Com- plementation between avirulent Newcastle disease virus and a fusion protein gene expressed from a retrovirus vector: require- ments for membrane fusion. J. Virol. 65, 813–822. 3. Horvath, C.M., Paterson, R.G., Shaughnessy, M.A., Wood, R. & Lamb, R.A. (1992) Biological activity of paramyxovirus fusion proteins: factors influencing formation of syncytia. J. Virol. 66, 4564–4569. 4. Sergel, T., McGinnes, L.W., Peeples, M.E. & Morrison, T.G. (1993) The attachment function of the newcastle disease virus hemagglutinin-neuraminidase protein can be separated from fusion promotion by mutation. Virology 193, 717–726. 5. San Roma ´ n, K., Villar, E. & Mun ˜ oz-Barroso, I. (1999) Acidic pH enhancement of the fusion of Newcastle disease virus with cul- tured cells. Virology 260, 329–341. 6. Morrison, T.G. & Portner, A. (1991) Structure, function and intracellular processing of the glycoproteins of Paramixoviridae.In The Paramyxovirus (Fraenkel-Conrat, H. & Wagner, R.R., eds), pp. 347–382. Plenum Press, New York. 7. Iwata, S., Schmidt, A.C., Titani, K., Suzuki, M., Kido, H., Gotoh, B., Hamaguchi, M. & Nagai, Y. (1994) Assignment of disulfide bridges in the fusion glycoprotein of Sendai virus. J. Virol. 68, 3200–3206. 8. Brasseur, R., Vandenbranden, M., Cornet, B., Burny, A. & Ruysschaert, J.M. (1990) Orientation into the lipid bilayer of an asymmetric amphipathic helical peptide located at the N-terminus of viral fusion proteins. Biochim. Biophys. Acta 1029, 267–273. 9. Chambers, P., Pringle, C.R. & Easton, A.J. (1990) Heptad repeat sequences are located adjacent to hydrophobic regions in several types of virus fusion glycoproteins. J. General Virol. 71, 3075– 3080. 10. Buckland, R. & Wild, F. (1989) Leucine zipper motif extends. Nature 338,547. 11. Sergel, T.A., McGinnes, L.W. & Morrison, T.G. (2000) A single amino acid change in the Newcastle Disease Virus fusion protein alters the requirement for HN protein in fusion. J. Virol. 74, 5101–5107. 12. Lamb, R.A. & Kolakofsky, D. (2001) Paramyxoviridae: the viruses and their replication. In Fundamental Virology (Fields, B.N., Knippe, D.M. & Kato, A., eds), pp. 1305–1340. Lippincot- Raven, NY. 13. Rosen, C.G. & Weber, G. (1969) Dimer formation from 1-Anili- no-8-naphthalenesulphonate catalyzed by bovine serum albumin. A new fluorescent molecule with exceptional binding properties. Biochemistry 8, 3915–3920. 14. Korte, T. & Herrmann, A. (1994) pH-dependent binding of the fluorophore bis-ANS to influenza virus reflects the conformational change of hemagglutinin. Eur Biophys. J. 23, 105–113. 15. Garcı ´ a-Sastre, A., Cabezas, J.A. & Villar, E. (1989) Proteins of Newcastle disease virus envelope: Interaction between the outer hemagglutinin-neuraminidase glycoprotein and the inner non-glycosylated matrix protein. Biochim. Biophys. Acta 999, 171– 175. 16. Korte, T., Ludwig, K., Booy, F.P., Blumenthal, R. & Herrmann, A. (1999) Conformational intermediates and fusion activity of influenza virus hemagglutinin. J. Virol. 73, 4567–4574. 17. Garcı ´ a-Sastre, A., Cobaleda, C., Cabezas, J.A. & Villar, E. (1991) On the inhibition mechanism of the sialidase activity from New- castle Disease Virus. Biol. Chem. Hoppe-Seyler 372, 923–927. 18. Iorio, R.M., Field, G.M., Sauvron, J.M., Mirza, A.M., Deng, R., Mahon, P.J. & Langedijk, J.P. (2001) Structural and functional relationship between the receptor recognition and neuraminidase activities of the Newcastle disease virus hemagglutinin-neur- aminidase protein: receptor recognition is dependent on neur- aminidase activity. J. Virol. 75, 1918–1927. Ó FEBS 2004 Conformational changes in NDV glycoproteins (Eur. J. Biochem. 271)587 19. Carneiro, F.A., Ferradosa, A.S. & Da Poian, A.T. (2001) Low pH-induced conformational changes in vesicular stomatitis virus glycoprotein involve dramatic structure reorganization. J. Biol. Chem. 276, 62–67. 20. Jones, P.L., Korte, T. & Blumenthal, R. (1998) Conformational changes in cell surface HIV-1 envelope glycoproteins are triggered by cooperation between cell surface CD4 and co-receptors. J. Biol. Chem. 273, 404–409. 21. Haywood, A.M. (1994) Virus receptors: binding, adhesion strengthening, and changes in viral structure. J. Virol. 68,1–5. 22. Mothes, W., Boerger, A.L., Narayan, S., Cunningham, J.M. & Young, J.A. (2000) Retroviral entry mediated by receptor priming and low pH triggering of an envelope glycoprotein. Cell 103, 679– 689. 23. Seth, S., Vincent, A. & Compans, R.W. (2003) Activation of fusion by the SER virus F protein: a low-pH-dependent paramyxovirus entry process. J. Virol. 77, 6520–6527. 24. Sattentau, Q.J., Moore, J.P., Vignaux, F., Traincard, F. & Poignard, P. (1993) Conformational changes induced in the envelope glycoproteins of the human and simian immunodeficiency viruses by soluble receptor binding. J. Virol. 67, 7383–7393. 25.Crennell,S.,Takimoto,T.,Portner,A.&Taylor,G.(2000) Crystal structure of the multifunctional paramyxoviruses hemagglutinin-neuraminidase. Nat. Struct. Biol. 7, 1068–1074. 26. Mun ˜ oz-Barroso, I., Cobaleda, C., Zhadan, G., Shnyrov, V. & Villar, E. (1997) Dynamic properties of Newcastle Disease Virus envelope and their relations with viral hemagglutinin-neur- aminidase membrane glycoprotein. Biochim. Biophys. Acta 1327, 17–31. 27. Takimoto, T., Taylor, G.L., Connaris, H.C., Crennell, S.J. & Portner, A. (2002) Role of the hemagglutinin protein in the mechanism of paramyxovirus-cell membrane fusion. J. Virol. 76, 13028–13033. 28. Puri, A., Winick, J., Lowy, R.J., Covell, D., Eidelman, O., Walter, A. & Blumenthal, R. (1988) Activation of vesicular stomatitis virus fusion with cells by pretreatment at low pH. J. Biol. Chem. 263, 4749–4753. 29. Stone-Hulslander, J. & Morrison, T.G. (1999) Mutational analysis of heptad repeats in the membrane proximal region of Newcastle disease virus HN protein. J. Virol. 73, 3630–3637. 588 L. Ferreira et al.(Eur. J. Biochem. 271) Ó FEBS 2004 . Conformational changes of Newcastle disease virus envelope glycoproteins triggered by gangliosides Laura Ferreira, Enrique Villar and Isabel. structural changes [14,16,19,20]. The exposure of hydrophobic segments of NDV envelope proteins triggered by gangliosides was tested by measuring binding to bis-ANS. Initially, the effect of increasing. indicate that the conformational change undergone by NDV glycoproteins in the presence of gangliosides is limited (see below). To assess the specific effect of gangliosides, 20 lg of NDV was preincubated