A peptide derived from a conserved domain of Sendai virus fusion protein inhibits virus-cell fusion. A plausible mode of action.

SV-201, a peptide derived from a conserved and potentially amphipathic region (amino acids 201-229) in the Sendai virus ectodomain, specifically inhibited virus-mediated hemolysis only when added to virions prior to their attachment to red blood cells. Sendai virus-mediated hemagglutinin assay in the presence of SV-201 demonstrated that the peptide does not disturb the binding of virions to the target red blood cells. A mutated peptide with 2 amino acids substitution, rendering the peptide neutral, was biologically inactive. A second mutant with 7 amino acids randomized at the N terminus keeping the hydrophobicity of the peptide unaltered was only slightly active. A hydrophobic peptide corresponding to the fusion peptide domain was also inactive. SV-201, the two mutants, and the fusion peptide bind similarly with high affinity to both negatively charged phosphatidylserine/phosphatidylcholine and zwitterionic phosphatidylcholine lipid vesicles, suggesting that the inhibitory effect is not due merely to membrane modulation. Fluorescence studies with rhodamine-labeled peptides and SV-201-induced inhibition assays, demonstrated that the SV-201 binding site is most probably located in the region corresponding to amino acids 201-229 of the Sendai virus fusion protein. The data presented here suggest that SV-201 disturbs a functional domain in the Sendai virus fusion protein, which is most probably associated with the assembly of the fusion protein and/or membrane apposition. The existence of homologous SV-201 regions in other viruses suggests that these regions may have a similar role, and their synthetic counterparts may act as inhibitors for the corresponding viruses.

Penetration of enveloped viruses into host cells involves the fusion of viral and cellular membranes and the subsequent transfer of genetic material into the target cell (1)(2)(3). Some of the enveloped viruses, including single-stranded RNA viruses such as influenza viruses, and alphaviruses, are internalized into the target cell by receptor-mediated endocytosis at acidic pH (4). However, the viral envelopes of e.g. paramyxoviruses and retroviruses including human immunodeficiency virus, fuse directly with the cellular plasma membrane at neutral pH (2,5).
Sendai virus, a member of the paramyxovirus family, con-tains two types of glycoproteins, hemagglutinin neuraminidase and fusion (F) glycoprotein (5,6). Although the hemagglutinin neuraminidase glycoprotein is responsible for the attachment of virions to the receptor of the target cell (7,8), it may also play some as yet undefined role in the fusion process (9). The fusion glycoprotein is believed to disrupt the target cell membrane and induce membrane fusion. The fusion protein monomer, synthesized as an inactive precursor (F0), is cleaved by a host proteolytic enzyme to form the biologically active protein consisting of two disulfide-linked subunits, F1 and F2 (7, 8, 10 -12). This processing exposes a hydrophobic region at the N terminus of the F1 subunit, which is highly conserved among paramyxovirus F proteins. This domain, designated "fusion peptide", is considered to be directly involved in promoting fusion with the target membrane (12,13). Evidence supporting this hypothesis includes: (1) site-directed mutagenesis in the fusion peptide regions of several enveloped viruses including influenza virus (1,14), HIV 1 (15,16), simian immunodeficiency virus (17), and SV5 (18), which severely affect the fusogenic activity of the corresponding viruses; and (2) studies with synthetic peptides that mimic the fusion peptide region of several viruses including influenza virus (19,20), HIV (21)(22)(23), simian immunodeficiency virus (24), and Sendai virus (25). In addition, the photoactive lipid probe preferentially labeled the fusion peptide domain of influenza hemagglutinin (26). The role of the F2 subunit in the paramyxovirus-mediated membrane fusion process is not yet clearly understood. Reduction of disulfide bonds on the F1-F2 fusion protein results in loss of fusion activity, which may mean that F2 is required for the function of the fusion protein (27). Although the role of the fusion peptide of the viral envelope glycoprotein has been substantiated, it is difficult to envision that the entire fusion process is determined solely through the small N-terminal region of the fusion protein. Indeed, recent reports point to the possible involvement of heptad repeats located adjacent to the fusion peptide region and the transmembrane anchor domain in the mediation of membrane fusion. Heptad repeats have been assumed to play a role in the assembly of the fusion protein of the influenza virus (28), HIV (29), and Sendai virus (30). Point mutations in the heptad repeats of the fusion proteins of the measles virus (31), the transmembrane gp41 of HIV (32,33), and Newcastle disease virus (34) severely affect viral activity. Furthermore, synthetic peptides corresponding to heptad repeats of HIV (35)(36)(37), Sendai virus (30), and several paramyxoviruses (38) can inhibit viral-induced cell fusion by their corresponding viruses. Hydropathy plot of Sendai virus fusion protein shows that in addition to the N-terminal fusion peptide and the transmembrane anchor domain, there are other significantly hydrophobic regions within the ectodomain of the fusion protein (39). This may indicate that additional regions may be involved with the membrane fusion process. To study this possibility, we selected SV-201 (amino acids 201-229), a domain with intermediate hydrophobicity, preceding the N-terminal heptad repeat and which potentially can form an amphipathic ␣-helix structure. This region is not only extremely conserved in the Sendai virus family but also has appreciable amino acid homology with the corresponding regions of other paramyxoviruses (Fig. 1). We have synthesized SV-201 and two mutant peptides, Mu1SV-201 and Mu2SV-201. In Mu1SV-201, 2 conserved amino acids were substituted by 2 negatively charged amino acids to retain the peptide with a zero net charge, and in Mu2SV-201, 7 amino acids at the N terminus were randomized keeping the hydrophobicity of the peptide unaltered. We found that SV-201 but not the two mutants and the hydrophobic 33-amino acid fusion peptide of the Sendai virus was a potent inhibitor of virusinduced hemolysis. SV-201 inhibited lysis only when it was incubated with the virions prior to, but not after, their attachment to red blood cells. However, both the wild type and mutant peptides could not inhibit virus-mediated agglutination of human red blood cells. These results, together with data obtained using fluorescently labeled wild type and mutant peptides, are discussed in terms of the possible role of the SV-201 domain of the Sendai virus fusion protein in the assembly of the fusion protein during the initial step of the fusion process and in membrane apposition.
Peptide Synthesis, Fluorescent Labeling, and Purification-The peptides were synthesized by the solid phase method on the corresponding resin (0.15 meq) as described previously (40,41). Double coupling was carried out with freshly prepared hydroxybenzotriazole-active esters of butyloxycarbonyl-amino acids. Labeling of the N terminus of a peptide was achieved as reported previously (42). Briefly, 15 mg of a resinbound peptide in its fully protected form was treated with trifluoroacetic acid (50% v/v in methylene chloride) to remove the butyloxycarbonylprotecting group from the N-terminal amino groups of the linked peptides. The resin-bound peptides were then treated with either (i) tetramethylrhodamine succinimidyl ester (3-4 eq) in dry dimethylformamide containing 5% v/v diisopropylethylamine or (ii) NBD-fluoride (2-3 eq) in dry dimethylformamide, which led to the formation of resin-bound N 1 -Rho or N 1 -NBD peptides, respectively. After 48 h, the resins were washed thoroughly with N,N-dimethylformamide and then with methylene chloride. The peptides were then cleaved from the resins by hydrogen fluoride and finally precipitated with ether. All the peptides were purified using reverse phase-high performance liquid chromatography on a C 4 reversed phase Vydac column (300 Å pore size). The column was eluted in 40 min using a linear gradient of 25-80% acetonitrile in water (containing 0.05% trifluoroacetic acid (v/v)), at a flow rate of 0.6 ml/min. The purified peptides were shown to be homogeneous (ϳ99%) by analytical high performance liquid chromatography. The peptides were subjected to amino acid analysis.
Preparation of Small Unilamellar Vesicles-Small unilamellar vesicles were prepared by sonication of PC or PS/PC (1:1 w/w) as described previously (43). Briefly, dry lipids were dissolved in CHCl 3 :MeOH (2:1 v/v). The solvents were evaporated under a stream of nitrogen, and the lipids were resuspended in PBS buffer (at a concentration of 7.2 mg/ml) by vortex mixing. The resulting lipid dispersions were sonicated (10 -30 min) in a bath-type sonicator (G1125SP1 Sonicator, Laboratory Supplies Company Inc., NY) until the turbidity had cleared. The lipid concentrations of the resulting preparations were determined by phosphorus analysis (44). Vesicles were visualized using a JEOL JEM 100B electron microscope (Japan Electron Optics Laboratory Co., Tokyo, Japan) by depositing a drop of vesicles on a carbon-coated grid and negatively staining with uranyl acetate. Examination of the grids demonstrated that the vesicles were unilamellar with an average diameter of 20 -50 nm.
Virus and Erythrocytes-Sendai virus (Z strain) was a generous gift of Prof. Michael Ovadia from Tel Aviv University, Israel. The virus was resuspended in buffer composed of 160 mM NaCl, 20 mM Tricine, pH 7.4, and stored at Ϫ70°C. Human blood was obtained from a blood bank and used fresh. Prior to use, erythrocytes were washed twice with PBS, pH 7.3, and diluted to the desired concentration with the same buffer.
Sendai Virus-induced Hemolysis of Human RBCs and Its Inhibition by SV-201-Virions, erythrocytes, and peptides were mixed in different orders of addition and in various amounts. Briefly, peptides were added at different concentrations to virions (35 hemagglutination units) and erythrocytes (2.2%) by three different methods: (i) Peptides were added to virions suspension followed by incubation (20 or 80 min, room temperature) to enable the binding of peptides to virions. Erythrocytes were then added and the suspension further incubated for 10 min. (ii) Virions were incubated with erythrocytes in an ice bath for 30 min followed by the addition of peptides, and further incubated for 20 min. (iii) Peptides were incubated with erythrocytes for 10 min at room temperature followed by the addition of virions, and further incubated for 20 min. At the end of each one of the three experiments, final incubation was always at 37°C for 40 min, followed by centrifugation at 5700 ϫ g for 8 min to remove intact cells. In all assays, duplicate samples were used, and two aliquots taken from the supernatant of each sample were placed in two wells of a 96-well plate. Each experiment was repeated 2-4 times. The amount of hemoglobin released was monitored by measuring the absorbance of the wells by using a plate reader at 540 nm.
Hemagglutinin Assay-Hemagglutinating activity of Sendai virus in the presence and absence of peptides was determined using fresh human RBCs in a 96-well microtiter plate by standard methods (27,45,46). In a control experiment, 100 l of 4% RBC were added to 100 l of Sendai virus suspension. In another control experiment the same amount of RBCs was added to solutions of SV-201, Mu1SV-201, and Mu2SV-201 in PBS (concentrations ranging from 0 to 15 M). In the actual experiment, Sendai virions were incubated with SV-201, Mu1SV-201, Mu2SV-201, and the fusion peptide (0 -15 M each) for 80 min to allow the binding of peptides to virions, followed by the addition of 4% RBCs (100 l) to each suspension. One hour after the addition of RBCs, the degree of agglutination was observed as described by others (45,46). The agglutinated RBCs settled in the bottom of the wells in aggregates to form a uniform carpet of cells, whereas unagglutinated RBCs formed a compact button of cells. All the experiments were done in duplicate using 2% RBCs and 35 hemagglutination units Sendai virions in a volume of 200 l.
NBD and Rhodamine Fluorescence Measurements-Fluorescence emission spectra of NBD-and Rho-labeled peptides either in PBS or in the presence of lipid vesicles were recorded at room temperature on a Perkin-Elmer LS-50B spectrofluorometer with the excitation monochromator set at 467 and 530 nm, respectively, with a 5-8-nm slit width. Measurements were performed in a 0.5-cm path length glass cuvette in a final reaction volume of 0.4 ml.

RESULTS
We report the identification, synthesis, biological function, and possible site of action of a 29-residue peptide, designated SV-201, derived from a conserved region in the ectodomain of Sendai virus fusion protein (amino acids 201-229). Fig. 1 depicts the sequence, designation, and location of SV-201 in the F1 domain of the fusion protein. A possible structure of this sequence could be an amphipathic ␣-helix, which could be involved in the assembly of proteins in solution as well as within membranes (47). A Shiffer and Edmundson wheel projection of the 18 most conserved amino acids of SV-201 reveals that the peptide has the potential to form an amphipathic ␣-helical structure (Fig. 2, inner circle), in which most of the hydrophobic amino acids are segregated on one surface with the hydrophilic amino acids on the opposite surface. Indeed, CD studies showed that SV-201 has ϳ70% ␣-helical structure in trifluoroethanol (48), a solvent which stabilizes the helical structure. Fig. 2 also shows the wheel structure of the three homologous regions of three other viruses. Interestingly, the most conserved regions face the hydrophobic surface. In addition, a net positive charge is conserved in the hydrophilic surface. The two mutant peptides, Mu1SV-201 and Mu2SV-201, served as controls (Fig. 1). In Mu1SV-201 a conserved glycine at the hydrophobic face (position 207) and a positively charged arginine in the hydrophilic face (position 205) were substituted with 2 negatively charged amino acids, rendering the peptide neutral. In Mu2SV-201, 7 amino acids at the N terminus were randomized, thus maintaining the same amino acid composition as in SV-201. The hydrophobic fusion peptide (25) served as a third control. CD experiments showed that Mu1SV-201 and Mu2SV-201 adopt 61 and 60% ␣-helical structure, respectively, in trifluoroethanol indicating that the mutations did not alter significantly the secondary structure. 2 The fusion peptide has been shown previously to adopt a predominantly ␣-helical structure (25). The peptides were then labeled with the fluorescent probes rhodamine and NBD to study their ability to assemble in solution and to bind phospholipid membranes.

SV-201 Is a Potent Inhibitor of Sendai Virus-mediated Hemolysis Only If Added Prior to the Attachment of the Virus to
Erythrocytes-The hemolytic activity of Sendai virus is associated with fusion of the virus with target RBCs (49). The assay was performed as follows. Virions were initially incubated with RBCs at room temperature to allow their attachment to the cells. Subsequent incubation at 37°C resulted in RBCs lysis. The extent of lysis was measured by absorbance at 540 nm (characteristic of hemoglobin).  inhibition was observed (column F). Thus, SV-201 could inhibit Sendai virus-induced hemolytic activity only if interacted first with virions. Column G shows that SV-201 (4 M) also has no inhibitory effect when incubated with RBCs before virions were added. Since ϳ70% of the peptide remains free in the solution in the presence of RBCs (data not shown), the attachment of virions to RBCs is probably kinetically faster than binding of SV-201 to virions. It should be noted that the wild type, mutant, and fusion peptides had no hemolytic activity up to the maximum concentration tested (100 M) (data not shown), thus eliminating the possibility that they contribute to the hemolytic activity during the fusion process. Overall, these results suggest that SV-201 inhibits Sendai virus-mediated hemolytic activity by interfering with some of the steps that take place during the attachment of virions to the target cells.
A Possible Mechanism of Inhibition-The results of Fig. 3 suggest that SV-201 can interfere with a functional domain of the Sendai virus fusion protein only when it is allowed to interact with the virions before the addition of RBCs. To search for a possible mechanism for the inhibitory effect of SV-201, hemagglutinin activity of Sendai virions was examined in the presence of the peptides. Moreover, the wild type SV-201 and the two mutants were tested for their ability to self associate, to coassemble in solution, and to bind phospholipid membranes.
SV-201 Does Not Inhibit the Binding of Sendai Virions to Target Red Blood Cells-Sendai virus-induced hemagglutinin assay was performed in the presence and absence of the peptides to examine whether they disturb the binding of virions to target RBCs. Binding of Sendai virions to human red blood cells causes their agglutination (27,50,51). In a control experiment agglutination of human red blood cells was readily observed upon their incubation with virions. In another control experiment the same amount of RBCs was added to solutions of SV-201, Mu1SV-201, Mu2SV-201, and the fusion peptide in PBS (in the concentration range used for the inhibition assay) to examine whether the peptides themselves agglutinate RBCs. No hemagglutination was observed, which ruled out hemagglu-tinin activity of the peptides (Table I) For these measurements, the peptides were labeled at their N termini by either rhodamine or NBD. Identical results were obtained with both fluorescent probes, and therefore, only the data with the rhodamine-labeled peptides are presented. Since the fluorescence of rhodamine is quenched when several molecules are in close proximity, an increase in fluorescence is expected when an aggregated rhodamine-labeled peptide dissociates, a process that occurs when the peptide is cleaved by a proteolytic enzyme. When equal concentrations of Rho-SV-201, Rho-Mu1SV-201, and Mu2SV-201 (0.10 M each) were dissolved in methanol, a solvent which does not promote peptides' aggregation, the three peptides exhibited the same fluorescence as expected (data not shown). However, in PBS the fluorescence intensity of Rho-SV-201 was much lower than that of Rho-Mu1SV-201 and Rho-Mu2SV-201 (Fig. 4, time point 2), suggesting that the former is in a higher oligomeric state than the other two peptides. Fig. 4 shows the time response profile of the changes of rhodamine fluorescence upon the addition of proteinase K. The data revealed that upon addition of proteinase K (time point 2), the fluorescence of Rho-SV-201 increased drastically (5 times), whereas that of Rho-Mu1SV-201 and Rho-Mu2SV-201 changed only slightly. It should be noticed that both Rho-labeled mutants dissociate when they are diluted in PBS, as evidenced from the spontaneous dequenching of rhodamine fluorescence prior to the addition of proteinase K. Eventually, the wild type and mutant Rho-labeled peptides (0.10 M each) exhibited very similar fluorescence after degradation with proteinase K. This indicates that the lower fluorescence of Rho-SV-201 in PBS is due to its oligomerization in aqueous solution. The finding that Mu2SV-201 is not selfassociated in solution although it has the same amino acid composition and similar structure as that of the wild type peptide indicates that the self-association of SV-201 is sequence specific and not due to its hydrophobic nature.
To investigate the reversibility of the oligomerization process, unlabeled SV-201 was added to Rho-SV-201. We observed a dequenching of rhodamine fluorescence, which indicates the reversible nature of SV-201 association in solution (Fig. 5).  Fig. 6A demonstrates that the inhibitory effect of SV-201 increases with longer incubation time with virions, whereas the lower inhibitory effect of Mu2SV-201 is not increased with longer incubation time (Fig. 6C). Both Mu1SV-201 (Fig. 6B) and the fusion peptide (data not shown) were inactive. These data   (Fig. 7), pointing to the location of the NBD-labeled segments in hydrophilic environments (55)(56)(57). However, in the presence of PC vesicles in PBS, the fluorescence emission maxima of the three NBD-labeled peptides shifted to 531 Ϯ 1 nm, concomitant with large increases in fluorescence intensities (Fig. 7), indicating their binding to membranes. The three NBD-labeled peptides exhibited similar blue shifts and fluorescence enhancement also in the presence of negatively charged PS/PC vesicles (data not shown). The NBD-labeled fusion peptide of Sendai virus has a similar magnitude of blue shift and fluorescence enhancement in the presence of vesicles (42). Similar magnitudes of blue shifts are observed when surface-active NBD-labeled peptides interact with lipid membranes (54 -56) and are consistent with location of the NBD on the surface of the membrane (57). It might be speculated from the ability of SV-201 to bind lipid vesicles that the corresponding 201-229 amino acid domain of the Sendai virus fusion protein binds to the target cell membrane and assists in membrane apposition. NBD-labeled Mu1SV-201, Mu2-SV-201, as well as the fusion peptide exhibited the same emission maximum and almost identical fluorescence enhancement compared with wild type peptide in the presence of PC and PS/PC vesicles. Therefore, the inhibitory activity of SV-201 is most likely not the result of a nonspecific effect resulting from membrane modulation as has been seen with derivatives of hydrophobic peptides (58 -60). DISCUSSION We have described here the biological function and a possible site of interaction of a synthetic segment SV-201, correspond- ing to amino acids 201-229 in the Sendai virus fusion protein adjacent to the heptad repeat SV-163, which is located near the fusion peptide (30). A summary of the results of the functional studies is shown in Table I. Recent studies have demonstrated the inhibitory effects of synthetic peptides modeled after heptad repeat regions in several viruses including HIV, Sendai, and other paramyxoviruses (30,35,36,38,61). However, the peptides induced similar inhibitory effects whether they were incubated with the virions before or after their attachment to the target cell. Unlike the reported heptad repeats, SV-201 specifically inhibited the hemolytic activity of Sendai virus if interaction with the virions took place prior to their attachment to RBCs (Fig. 3). When virions were added to a mixture of RBCs and SV-201, Sendai virus-induced hemolysis was not inhibited (Fig. 3). This may be due to two reasons: (i) a major portion of the peptide is bound to RBCs, leaving only a small amount of the peptide to interact with the virions; and (ii) viral attachment to RBCs is faster than SV-201 binding to virions. The first possibility was ruled out following our observation that more than 70% of SV-201 remains free in solution in the presence of RBCs (data not shown). Therefore, it is probable that the faster kinetics involved in attachment of virions to RBCs protects the virions from the interference of SV-201. The data presented here suggest a possible binding site and a stage of the fusion process that might be affected by SV-201.
A Possible Binding Site for SV-201-Fluorescence dequenching experiments with rhodamine-labeled and unlabeled peptides demonstrated that SV-201 self-associates in aqueous solution, which is not the case for both of its mutants (Fig. 5). The fact that Mu2SV-201 cannot self-associate in aqueous solution despite having the same amino acid composition (same hydrophobicity) as the wild type SV-201, suggests that the oligomerization of SV-201 is a sequence-specific process and not due merely to nonspecific hydrophobic interactions. Moreover Direct interference with step 1 can be ruled out based on our experiments with the Sendai virus-mediated hemagglutinin assay in the presence of the peptides. The adhesion of the viral and erythrocyte membranes is mediated by the hemagglutinin neuraminidase protein, which binds to a receptor on the cell surface. It has been reported that binding of Sendai virions to human red blood cells causes their agglutination (27,50,51). We observed that SV-201, Mu1SV-201, and Mu2SV-201 could not inhibit Sendai virus-induced agglutination of human RBCs, indicating that the peptides probably do not affect the binding of the hemagglutinin neuraminidase protein of Sendai virus to the sialic acid receptors on the RBCs.
Step 3 of the fusion event involves destabilization of the lipid bilayers. SV-201 is significantly hydrophobic, adopts an amphipathic ␣-helical structure, and binds strongly to membranes irrespective of their charge (Fig. 7). These properties raise the possibility that the inhibitory effect of SV-201 is a result of its ability to modulate the membrane, a property characteristic of other inhibitors of Sendai virus, which are composed of derivatives of enantiomers of short hydrophobic peptides, some of which are derived from the hydrophobic fusion peptide of the virus (58,59,62). However, this possibility can also be ruled out considering the following findings. First, both mutants, Mu1SV-201 and Mu2SV-201, and the fusion peptide have the same binding ability to phospholipid membranes, yet only SV-201 is a potent inhibitor. Whereas the hydrophobicity of SV-201 was slightly altered in Mu1SV-201, Mu2SV-201 retains the same amino composition of SV-201 with only 7 amino acids in the N terminus randomized. Second, SV-201 cannot inhibit fusion when incubated first with erythrocytes, which would be expected if it directly affects the properties of the membrane. Indeed, the above mentioned small hydrophobic peptides are inhibitors whether they are incubated first with virions or with erythrocytes (63). Third, SV-201 inhibits at the M range, which is about 2 orders of magnitude less than that found with peptides that modulate the structure of the membrane (6). Fourth, the inhibitory effect of SV-201 is time dependent, whereas membrane binding and modulation take place within seconds to minutes (25,43,64). We can exclude also the possibility that SV-201 interferes with step 4, which is the actual fusion event, since we have shown that SV-201 cannot inhibit fusion after the binding of virions to target cells.
Thus SV-201 probably disturbs step 2, i.e. close approach of the two lipid bilayers of the viral and cellular membranes. The finding that SV-201 has high affinity to phospholipid membranes and specifically self-associates in solution suggests, therefore, two plausible roles of the 201-229 amino acid domain in step 2 of Sendai virus-mediated membrane fusion. (i) It is possible that this domain is involved in membrane apposition since it is not perturbed by the corresponding SV-201 after the attachment of virions to target cells. This is further supported by the fact that SV-201 binds to both negatively charged (PS/ PC) and zwitterionic (PC) phospholipid vesicles (Fig. 6), suggesting that the 201-229 amino acid domain can bind to the target cell membrane and thus assist in bringing the viral and cellular membranes closer to facilitate fusion. A similar role has been proposed for the leucine zipper motifs present in influenza hemagglutinin and HIV gp41 (65,66). (ii) A conformational change or a change in the assembly of Sendai virus fusion protein may take place before the apposition of viral and target membranes, followed by the attachment of virions to target cell. One may speculate that the amino acid 201-229 domain is involved in these changes. The ability of SV-201 to assemble in an aqueous environment (Figs. 4 and 5) points to the involvement of this domain in the assembly of the fusion protein. By interacting with this domain before virions attachment to target RBCs, the synthetic peptide SV-201 could interfere with the necessary conformation change and/or assembly and thus inhibit the hemolytic activity of Sendai virus. This is in line with the hypothesis suggested recently based on the inhibitory effect of heptad repeat regions in other paramyxoviruses (38). At this point we cannot rule out any of the possibilities. The finding that SV-201 is a potent inhibitor only at a certain step before attachment of the virions to target cells, suggests that its binding site (most probably the domain of amino acids 201-229) is no longer accessible once the virions are already attached to cells. Alternatively, irreversible conformational changes are induced following viral attachment to RBCs. Therefore, even if SV-201 binds to its counterpart in the fusion protein, it cannot induce further disorganization of the fusogenic complex.
The finding that synthetic heptad repeats located at similar positions in the fusion proteins of several viruses have inhibitory effects, together with the finding that homologous SV-201 regions exist in other viruses (Fig. 1), suggests that these regions may play a role similar to that of the amino acids 201-229 domain in the Sendai virus and may act as inhibitors for their corresponding viruses.