Peptides Corresponding to the Heptad Repeat Sequence of Human Parainfluenza Virus Fusion Protein Are Potent Inhibitors of Virus Infection

VIROLOGY

223, 103–112 (1996) 0459

ARTICLE NO.

Peptides Corresponding to the Heptad Repeat Sequence of Human Parainfluenza Virus Fusion Protein Are Potent Inhibitors of Virus Infection QIZHI YAO and RICHARD W. COMPANS1 Department of Microbiology and Immunology, Emory University School of Medicine, Atlanta, Georgia 30322 Received April 8, 1996; accepted June 28, 1996 It has been suggested that a conserved heptad repeat region in paramyxovirus fusion (F) proteins is essential for viral fusion activity (Buckland et al., 1992; Sergel et al., 1994; Reitter et al., 1995). We have studied synthetic peptides containing the heptad repeat regions derived from the F proteins of human parainfluenza virus type 2 (PI2) and type 3 (PI3) for their function as potential inhibitors of virus-induced cell fusion as well as their effects on spread of viral infection. Two peptides containing sequences of heptad repeat B, adjacent to the transmembrane domain of the F protein, were synthesized for both PI2 and PI3 F proteins. We observed that the longer peptides [34 amino acids (a.a.) for PI2F or 35 a.a. for PI3F] which extend from heptad repeat B to the transmembrane domain showed complete inhibition of cell fusion induced by the respective virus as well as by the vaccinia-expressed F and HN proteins. The 50% effective concentration to inhibit virusinduced cell fusion was 2.1 mM for PI2 and 1.2 mM for PI3. Moreover, the inhibitory effects of each peptide on virusinduced cell fusion were found to be virus type-specific. These peptides were found to also inhibit viral entry and to prevent plaque formation when mixed with the virus inoculum. Furthermore, the peptides caused a reduction in virus yield when assayed 48 hr after low m.o.i. infection and in the size of viral plaques when added to the overlay. Shorter peptides (21 a.a. for PI2F or 24 a.a. for PI3F) which correspond to the partial sequence of heptad repeat B for PI2F and the entire heptad repeat B for PI3F showed partial inhibition of PI2- or PI3-induced cell fusion. These results indicate that peptides containing the heptad repeat B sequence have the potential to inhibit virus-induced cell fusion, virus entry, and spread of virus infection. q 1996 Academic Press, Inc.

INTRODUCTION

peptide sequence is highly conserved among paramyxovirus F proteins (Morrison and Portner, 1991) and is considered to be directly involved in mediating membrane fusion (Lamb, 1993). Two heptad repeat regions which have the potential to form a-helices are consistently present in the F proteins of paramyxoviruses. Heptad repeat A lies adjacent to the hydrophobic fusion peptide at the amino terminus of F1, while heptad repeat B lies immediately upstream of the transmembrane region (Buckland and Wild, 1989; Chambers et al., 1990). The function of the heptad repeat region is not known but may related to cell fusion activity (Buckland et al., 1992). The mechanism of virus-induced membrane fusion has not been well defined. Previous studies in our laboratory showed that cell fusion caused by human parainfluenza virus type 2 (PI2) and type 3 (PI3) glycoproteins requires coexpression of F and HN proteins from the same virus type, suggesting that a type-specific interaction between the homologous F and HN plays an important role in parainfluenza virus-induced cell fusion (Hu et al., 1992). Other reports have shown similar typespecific interactions among paramyxovirus glycoproteins (Horvath et al., 1992; Cattaneo and Rose, 1993; Malvoisin and Wild, 1993; Bousse et al., 1994; Heminway et al., 1994; Wild et al., 1994b). Many studies with influenza virus have demonstrated that a conformational change in the influenza HA protein activates its fusion activity

The human parainfluenza viruses are enveloped viruses which possess an outer limiting membrane containing two glycoproteins, the hemagglutinin-neuraminidase (HN) and fusion (F) proteins (Morrison and Portner, 1991). It was initially proposed that the HN protein provides an attachment activity which brings the F protein and the target membrane into close proximity, allowing F to reach the lipid bilayer to mediate fusion (Hsu et al., 1979). However, several lines of evidence indicate that the HN protein plays a more direct role in the induction of cell fusion (Miura et al., 1982; Heath et al., 1983; Merz and Wolinsky, 1983; Shibuta et al., 1983; Gittman and Loyter, 1984; Nussbaum et al., 1984; Tsurudome et al., 1986; Waxham and Wolinsky, 1986; Portner et al., 1987; Thompson and Portner, 1987), whereas the F protein is directly involved in virus-induced membrane fusion which is required for the penetration of the virus (Choppin and Compans, 1975; Choppin and Scheid, 1980). Activation of viral fusion activity by proteolytic cleavage of the F protein results in the formation of disulfide-linked F1– F2 subunits and exposure of a hydrophobic fusion peptide at the N-terminus of the F1 subunit (Homma and Ohuchi, 1973; Scheid and Choppin, 1974). The fusion 1

To whom reprint requests should be addressed.

0042-6822/96 $18.00

103

AID

VY 8092

/

6a1d$$$161

08-05-96 14:08:35

Copyright q 1996 by Academic Press, Inc. All rights of reproduction in any form reserved.

vira

AP: Virology

104

YAO AND COMPANS

upon exposure to low pH (Stegmann et al., 1987; Wharton, 1987; White and Wilson, 1987; Ruigrok et al., 1988; Doms et al., 1990). The HA2 subunit forms an extended coiled coil structure at low pH, in which the location of the fusion peptide has been rearranged to the tip of the HA trimer (Carr and Kim, 1993; Bullough et al., 1994)). In contrast, at neutral pH, the fusion peptides are sequestered near the base of the spike where they are unable to undergo interaction with the target membrane (Wilson et al., 1981). The F protein of paramyxoviruses, like influenza HA, appears to undergo a conformational change upon proteolytic cleavage (Hsu et al., 1981, 1982). By analogy with influenza HA, it has been proposed that an additional conformational change occurs in the F protein which exposes the fusion peptide at the right time and place (Sergel et al., 1993; Lamb, 1993). However, in paramyxoviruses this conformational change is not activated by low pH. Several possible mechanisms for triggering the conformational change of F protein in paramyxoviruses have been suggested (Lamb, 1993) including interaction of F with HN, its interaction with receptors, or interactions mediated by cellular proteins. Recent reports have indicated that apart from the fusion peptide, other domains within viral fusion proteins are also important in mediating the fusion event. The transmembrane and cytoplasmic domains were found to be necessary for complete fusion in influenza virus (Kemble et al., 1994). In a study of the cytoplasmic domains of PI2 and PI3 F proteins, we found that the sequence of PI3 F cytoplasmic domain can affect paramyxovirus-induced cell fusion (Yao and Compans, 1995). Moreover, several studies have suggested that a leucine zipper-like motif in the region near the transmembrane domain is necessary for syncytium formation in measles virus and Newcastle disease virus (NDV) (Buckland et al., 1992; Sergel et al., 1994; Reitter et al., 1995). Studies of human immunodeficiency virus type 1 (HIV1)-induced cell fusion suggested that a coiled-coil (leucine zipper-like) structure may be critical for a fusionrelated conformational change in gp41 (Wild et al., 1992; Dubay et al., 1992; Chen et al., 1993). Recent studies indicated that synthetic peptides corresponding to two leucine zipper-like motifs near the fusion peptide or adjacent to the transmembrane domain within the envelope protein gp41 (TM) display various extents of anti-viral activity (Wild et al., 1992, 1994a; Jiang et al., 1993a,b). Synthetic peptides corresponding to a predicted a-helical domain were reported to be potent inhibitors of HIV1 infection and fusion. In the present study, synthetic peptides derived from the PI2 and PI3 F protein were studied to further define functional domains involved in cell fusion. The peptide sequences were based on a highly conserved heptad repeat B region in the F protein which was predicted to form an extended amphipathic a-helix (O’Shea et al., 1989) structurally analogous to those in the fusion glycoproteins of several other fuso-

genic viruses, such as influenza HA and retroviruses. We have investigated the function of these F peptides as inhibitors of virus-induced cell fusion as well as their effects on the spread of virus infection in cell culture.

AID

vira

VY 8092

/

6a1d$$$161

08-05-96 14:08:35

MATERIALS AND METHODS Cells, viruses, and antibody CV-1, HeLa T4, LLC-MK2 , and Vero cells were maintained in Dulbecco’s minimal essential medium (DMEM) with 10% newborn calf serum (HyClone Laboratories, Inc., Logan, UT). PI2, PI3, and SV5 viruses were obtained from the Division of Research and Resources, National Institutes of Health (Bethesda, MD). PI2 virus was adapted to grow on LLC-MK2 cells while PI3 and SV5 viruses were grown on Vero cell, and titers of the viruses were determined on Vero cells. Recombinant vaccinia virus-T7 (vTF7.3) was obtained from Bernard Moss (NIH). vTF7.3 was grown on HeLa T4 cells, and titers of the virus were determined on CV-1 cells. Rabbit anti-PI2 antibody was described by Hu et al. (1992). Recombinant plasmids PI2F and PI2HN cDNA genes were cloned in pGEM-3 and pGEM-3Zf(0) plasmids (Promega Biotech, Madison, WI), respectively, as described previously (Hu et al., 1992). The PI3F and PI3HN cDNA clones were kindly provided by Dr. Mark Galinski. These two genes were subcloned into pGEM plasmid vectors. The SV5F and SV5HN cDNA clones were kindly provided by Dr. Robert A. Lamb and were subcloned in pGEM plasmids. Peptide synthesis and purification Peptides were synthesized on a Rainin Symphony Multiplex multiple peptide synthesizer (Rainin Instruments Company, MA). Using standard solid-phase synthesis techniques, all peptides were acetylated at the N terminus and amidated at the C terminus. Cleavage of peptides from the resin and removal of side-chain blocking groups were automatically performed on the instrument with trifluoroacetic acid and scavengers (5% thioanisol, 5% water, 2.5% ethanedithiol, 0.8 M phenol). The peptides were then washed with 12 ml cold ether four times, dissolved in water, and dried under vacuum overnight. Peptides were dissolved in dionized water. Peptides were then purified by reverse-phase HPLC on a MicrosorbMV C18 , 5-mm, 300-A column (Rainin Dynamax) using a water:70% acetonitrile plus 30% water gradient containing 0.1 % trifluoroacetic acid. All peptides were ú95% pure by analytical reverse-phase HPLC. The identity of the peptides was verified by mass spectrometry. Cell fusion assays Monolayers of HeLa T4 cells were infected with PI2, PI3, or SV5 at m.o.i. of 0.1. After the virus was adsorbed

AP: Virology

HEPTAD REPEAT SEQUENCE VIRAL INHIBITORS

at 377 for 1 hr, the inoculum was removed and DMEM with 2.5% of FBS was added to the well with or without addition of serial dilutions of each peptide. For vaccinia expression studies, confluent monolayers of cells were infected with vTF7.3 at a m.o.i. of 10 and incubated at 377 for 1 hr. After removal of the inoculum, cells were transfected with 2 mg of F and HN plasmid DNA with lipofectin (GIBCO-BRL, Gaithersburg, MD) in 0.5 ml of DMEM. After 24 hr (for virus-infected cells) or 18 hr (for transfected cells) of incubation at 377 in a 5% CO2 atmosphere, the formation of syncytia was compared with the infected or transfected cell control containing no additional peptide. Cell fusion was measured by polykaryon formation and expressed as (number of nuclei in polykaryons/number of total nuclei) 1 100. Cells were counted in five different fields under a light microscope. Dose– response curves were also generated using this method. The IC50 (50% inhibitory concentration) was determined using the Miller-Tainter method (Welkos and O’Brien, 1994). Plaque reduction assay A series of two-fold dilutions of F peptides together with appropriately diluted PI2 or PI3 virus were added to monolayers of Vero cells. After 1 hr incubation at 377, the inoculum was removed and overlaid with 2% white agar mixed at 1:1 ratio with 21 DMEM containing 2% FBS. Three days after overlaying, red agar mixed at a 1:1 ratio with 21 DMEM containing 2% FBS was overlaid to stain the cells. Plaque numbers were counted 1 day after neutral red staining and recorded. The Reed-Muench method (Welkos and O’Brien, 1994) was used to calculate the 50% inhibitory concentration for virus plaque formation. Surface immunofluorescence Indirect immunofluorescence was performed as described by Paterson et al. (1989). In general, HeLa T4 cells were grown on glass coverslips in 24-well plates and infected with wild-type PI2 virus as described above. At indicated times postinfection, cells were washed with ice-cold PBS three times before being fixed with 1% paraformaldehyde in PBS. Rabbit anti-PI2 antibody was added onto cell monolayers, and cells were then incubated at 47 for 30 min. Cells were washed with ice-cold PBS three times, and then goat anti-rabbit immunoglobulin G fluorescein-conjugated antibody (Southern Biotechnology Associates, Inc., Birmingham, AL) was added and was then incubated at 47 for 30 min. Cell surface fluorescence was examined by fluorescence microscopy with a Nikon Optiphot microscope.

105

essential for viral fusion activity (Buckland et al., 1992; Sergel et al., 1994; Reitter et al., 1995). In order to study the role of this region in parainfluenza virus-induced cell fusion, synthetic peptides corresponding to the heptad repeat B region near the transmembrane domain of the fusion protein were synthesized. Figure 1A shows a schematic diagram of two peptides (2F1B21 and 2F1B34) from the PI2F sequence. 2F1B21 [21 amino acids (a.a.)] extends from the beginning of the transmembrane domain to a predicted glycosylation site in the heptad repeat B region, while 2F1B34 (35 a.a.) extends from the beginning of the transmembrane domain to include the entire heptad repeat B region. For PI3 virus (Fig. 1B), peptide 3F1B24 (24 a.a.) corresponds to the heptad repeat B only, while 3F1B35 (35 a.a.) extends from the beginning of the transmembrane domain to include the entire heptad repeat B region. To determine whether these peptides have the potential to inhibit virus-induced cell fusion, the extent of cell fusion was compared at 24 hr post viral infection in infected cells incubated with or without the peptides. In HeLa T4 cells infected with PI2 virus at a m.o.i. of 0.1, cell fusion was reduced remarkably at 24 hr postinfection when incubated with 2F1B21 peptide at a concentration of 80 mM or more, while cell fusion was completely inhibited by the 2F1B34 peptide at a concentration of 4 mM. Similarly, in PI3 virus-infected HeLa T4 cells, the longer peptide 3F1B35 completely inhibited virus-induced cell fusion at a concentration as low as 2 mM, while the shorter peptide 3F1B24 only partially inhibited cell fusion at 80 mM or more. Figures 2A and 2D show that extensive syncytia are observed in HeLa T4 cells infected with either the PI2 or PI3 virus. Syncytium formation was dramatically decreased in HeLa T4 cells infected with the virus and then incubated with the shorter peptides 2F1B21 (Fig. 2B) or 3F1B24 (Fig. 2E), whereas no syncytia were seen in infected cells incubated with the longer peptides 2F1B34 (Fig. 2C) or 3F1B35 (Fig. 2F). These results demonstrate that F peptides corresponding to the heptad repeat B have the potential to inhibit virus-induced cell fusion. However, the heptad repeat region by itself does not appear to be sufficient for the full inhibitory activity and the extra sequences extending to the transmembrane domain are also important for maximum inhibitory activity. The length of the peptides may also play a role in their inhibitory activity. Inhibition of cell fusion is virus type-specific

Several reports have indicated that conserved heptad repeat regions in the paramyxoviruses fusion protein are

To determine the specificity of the inhibition of virusinduced cell fusion, HeLa T4 cells were infected with one virus type and incubated with F peptides derived from the F protein of another virus type. The inhibition of cell fusion was observed at 24 hr postinfection. As shown in Table 1, peptides corresponding to PI2F sequences were found to inhibit only cell fusion induced by PI2 but

AID

vira

RESULTS Peptides containing the F protein heptad repeat B inhibit virus-induced cell fusion

VY 8092

/

6a1d$$$161

08-05-96 14:08:35

AP: Virology

106

YAO AND COMPANS

FIG. 1. Schematic diagram of the F proteins of PI2 and PI3 viruses and amino acid sequences of the synthetic F peptides. (A) Full-length PI2F and amino acid sequences of peptides 2F1B21 and 2F1B34 are shown; (Ú É ) represents a predicted glycosylation site within the heptad repeat B region. Arrows represent cleavage sites. TM, transmembrane domain. (B) Full-length PI3F and amino acid sequences of peptides 3F1B24 and 3F1B35 are shown.

not PI3. Similarly, peptides from PI3F were found to inhibit only cell fusion induced by PI3 but not PI2 virus. In addition, none of the peptides from PI2 or PI3 were found to inhibit cell fusion induced by SV5. These results demonstrate that the presence of a heptad repeat region, which is predicted to form an extended a-helical structure, is not the only requirement for inhibitory activity, but that the specific amino acid sequences within heptad repeat B and adjacent sequences in the peptides for each virus type play an essential role in the inhibitory activity of the F peptides. Peptides derived from the F protein inhibit cell fusion by vaccinia-T7-expressed F and HN proteins In order to confirm that the F peptides inhibit cell fusion by an effect on viral glycoproteins and that they are not acting on other steps in the viral replication cycle, the vaccinia-T7 transient expression system was used to express homologous F and HN with or without addition of the inhibitory peptides in the culture media. In Table 2, HeLa T4 cells were infected with recombinant vTF7.3 and then transfected with homotypic F and HN genes. Six hours posttransfection, F peptides were added to the culture media. Inhibition of F- and HN-induced cell fusion was observed at 18 hr in the wells incubated with the F peptides. Similar to results obtained with virus infection, 2F1B34 (at a concentration of 16 mM) was found to completely inhibit cell fusion induced by coexpression of PI2

AID

VY 8092

/

6a1d$$$161

08-05-96 14:08:35

F plus PI2 HN and 3F1B35 (at a concentration of 8 mM) inhibited fusion by PI3 F plus PI3 HN, respectively, while 2F1B21 and 3F1B24 only partially inhibited cell fusion. These results indicate that peptides corresponding to heptad repeat B have the potential to inhibit virus-induced cell fusion by interfering with the activity of the viral glycoproteins in the fusion process. Inhibition of cell fusion is dose-dependent To determine the dose dependence of inhibition of cell fusion, increasing concentrations of peptides were added to the culture medium and the effect on cell fusion was determined. Figure 3 shows the dose response curves for the longer peptides 2F1B34 from PI2F and 3F1B35 from PI3F. At a concentration of 2 mM, peptide 2F1B34 was found to block about 50% of cell fusion, and fusion was completely blocked by increasing the peptide concentration to 4 mM. Peptide 3F1B35 was found to completely inhibit cell fusion at a concentration of as low as 2 mM. Using the Miller-Tainter method, the 50% inhibitory concentration for cell fusion (IC50) was determined to be 2.1 mM for 2F1B34 and 1.2 mM for 3F1B35. The shorter peptides 2F1B21 from PI2 and 3F1B24 from PI3 could only partially inhibit cell fusion even at peptide concentrations as high as 80 mM (data not shown). F peptides inhibit viral entry and spread To determine whether the F peptides interfere with the viral entry process, a plaque reduction assay was

vira

AP: Virology

HEPTAD REPEAT SEQUENCE VIRAL INHIBITORS

107

FIG. 2. Inhibition of cell fusion by F peptides. HeLa T4 cells were infected with PI2 or PI3 virus at a m.o.i. of 0.1. After incubation of 1 hr at 377, the inoculum was removed and DMEM with 2.5% FBS was added with or without addition of each peptide. Cells are shown at 24 hr postinfection. (A–C) Cells were infected with PI2 virus; (D–F) cells were infected with PI3 virus. (A and D) 2.5% DMEM only; (B) addition of 80 mM of 2F1B21; (C) 4 mM of 2F1B34; (E) 80 mM of 3F1B24; (F) 2 mM of 3F1B35.

performed by inoculating cells with virus and peptide together at 377 for 1 hr and then overlaying with agar without the addition of peptide. As shown in Table 3, a progressive reduction in the number of viral plaques was observed with increasing concentration of peptide in the virus inoculum. Using this method, the IC50 of F peptides for plaque formation was determined to be 5.1 mM for 2F1B34 and 2.2 mM for 3F1B35. These data indicate that the F peptides inhibit the viral entry process.

We also compared viral titers in the culture supernatant incubated postinfection at a m.o.i. of 0.1 with or without peptides in the culture media. The virus titers were not remarkably reduced in the supernatant of cells incubated with the peptide at 18 or 24 hr postinfection, indicating that the peptides do not affect the first round of virus replication when added after virus entry. However, a 15-fold decrease in the viral titer was observed at 48 hr post PI2 infection in the presence of 2F1B34 (Fig. 4).

TABLE 1 Inhibition of Virus-Induced Cell Fusion by the F Peptides Is Virus Type-specifica 2F1B21

3F1B24

3F1B35

/

0

/

0

/

0

/

0

/ /// ///

///b /// ///

0 /// ///

/// /// ///

/// / ///

/// /// ///

/// 0 ///

/// /// ///

Viruses PI2 PI3 SV5

2F1B34

a Hela-T4 cells were infected with PI2 or PI3 at a m.o.i of 0.1. After virus adsorbtion at 377 for 1 hr, the inoculum was removed and replenished with 2.5% DMEM with (/) or without (0) addition of peptides. Cell fusion extent was observed 24 hr postinfection. b (///) ú 60% of nuclei are present in syncytia; (/) õ20% of nuclei are present in syncytia; (0) no syncytia were seen.

AID

VY 8092

/

6a1d$$$161

08-05-96 14:08:35

vira

AP: Virology

108

YAO AND COMPANS TABLE 2 Inhibition of Cell Fusion Induced by Vaccinia-Expressed F and HN Proteinsa by the F Peptides 2F1B21

2F1B34

3F1B24

3F1B35

Plasmids

/

0

/

0

/

0

/

0

PI2F / PI2HN PI3F / PI3HN SV5F / SV5HN

/ /// ///

///b /// ///

0 /// ///

/// /// ///

/// / ///

/// /// ///

/// 0 ///

/// /// ///

a

Hela T4 cells were infected with vTF7.3 at a m.o.i. of 10 at 377 for 1 hr, and then cells were transfected with homologous F and HN genes. Six hours posttransfection, F peptides were added to the culture media. Inhibition of cell fusion was observed 18 hr posttransfection. b (///) ú60% of nuclei are present in syncytia; (/) õ20% of nuclei are present in syncytia; (0) no syncytia were seen.

Similarly, the 3F1B35 peptide caused a 10-fold reduction in PI3 virus titers (not shown). These results indicate that the peptides strongly inhibit subsequent rounds of virus multiplication, presumably by preventing virus entry. As another indication of a reduction in spread of virus infection under multiple cycle conditions, the effect of the peptides on plaque size was determined. As shown in Fig. 5, when the longer PI2 F peptide was added to the agar overlay after the initial 1-hr virus adsorption period, the plaque size was reduced from about 3 mm in diameter in control cells to 1.5 mm in the presence of the 2F1B34 peptide. This result also indicates that the F peptides can inhibit multiple cycle replication of the virus, thus restricting viral spread. We also compared the distribution of infected cells by surface immunofluorescence at intervals post infection at m.o.i. of 0.1 in the presence or absence of F peptides. As shown in Fig. 6, at 18 hr postinfection (hpi), cultures infected with PI2 virus showed limited number of cells

with positive surface fluorescence for PI2 antigen, whereas the levels of cells expressing PI2 antigen on their surfaces increased by 24 and 48 hpi. In contrast, in cells incubated with peptide 2F1B34, there was no increase in numbers of cells showing antigen expression on the surface at 24 h or 48 hpi. These data indicate that in the presence of the F peptides, the infection process remains limited to the cells initially infected by the inoculum virus. DISCUSSION We have observed that peptides which contain a heptad repeat B sequence in the F1 subunit are specific inhibitors of PI2- or PI3-induced membrane fusion. The high specificity of their inhibitory effect was shown by the finding that each peptide was only able to inhibit fusion by the virus with the same F protein sequence, i.e., peptides derived from PI2F will inhibit fusion by PI2 virus but not PI3 virus and vice versa. The most active peptides extended from the second heptad repeat to the

TABLE 3 Inhibition of Plaque Formation by the F Peptidesa

FIG. 3. Dose-dependent inhibition of cell fusion by F peptides. HeLa T4 cells were infected with PI2 or PI3 virus at 377 for 1 hr and peptides were added after removal of the inoculum. Inhibition of cell fusion was observed at 24 hr postinfection. Cell fusion was measured by polykaryon formation and expressed as (number of nuclei in polykaryons/ number of total nuclei) 1 100. Cells were counted in five different fields under a light microscope.

AID

VY 8092

/

6a1d$$$161

08-05-96 14:08:35

2F1B34 (mM)

Plaque number b

3F1B35 (mM)

Plaque number b

0 0.5 1 2 4 8 16

50 46 37 30 21 17 8

0 0.5 1 2 4 8 16

45 38 25 18 12 10 6

a A series of two-fold dilutions of F peptides together with appropriately diluted PI2 or PI3 virus were added to monolayers of Vero cells. After 1 hr incubation at 377, the inoculum was removed and overlaid with 2% white agar mixed at a 1:1 ratio with 21 DMEM containing 2% FBS. Three days after overlaying, red agar mixed at a 1:1 ratio with 21 DMEM containing 2% FBS was overlaid to stain the cells. Plaque numbers were counted 1 day after neutral red staining and recorded. b The plaque numbers are the averages of three determinations at each peptide concentration.

vira

AP: Virology

HEPTAD REPEAT SEQUENCE VIRAL INHIBITORS

109

FIG. 4. Inhibition of virus production by F peptides. HeLa T4 cells were infected with PI2 virus at a m.o.i. of 0.1. After incubation at 377 for 1 hr, the inoculum was removed and 2.5% DMEM was added with or without addition of 4 mM of the synthetic peptides 2F1B21 and 2F1B34. Released virus titers were determined by plaque assay 48 hr postinfection.

beginning of the transmembrane domain (34 a.a. for PI2F and 35 a.a. for PI3F). Moreover, the F peptides were found to inhibit fusion by virus-infected cells as well as with vaccinia-expressed F and HN proteins and were also found to inhibit multiple cycle replication of the virus. Several reports have shown that synthetic peptides corresponding to a predicted coiled-coil region of gp41 of HIV-1 near the fusion peptide or the transmembrane domain have the potential to inhibit virus-induced cell fusion (Jiang et al., 1993a,b; Wild et al., 1992, 1994a). More recently studies of synthetic peptides corresponding to a conserved heptad repeat domain in paramyxovirus F proteins also showed a potent inhibitory effect on virus-induced cell fusion (Rapaport et al., 1995; Lambert et al., 1996). The mechanism by which these peptides inhibit HIV-1- or paramyxovirus-induced cell fusion is not fully understand. The peptides could alternatively interact

FIG. 6. Inhibition of the spread of viral infection. Monolayers of HeLa T4 cells were infected by PI2 virus at an m.o.i. of 0.1. At 18, 24, and 48 hr postinfection, cells were washed with PBS and incubated with rabbit anti-PI2 serum followed by FITC- conjugated goat anti-rabbit IgG. Cell surface fluorescence was examined with a Nikon Optiphot microscope. PI2 represents cells infected with PI2 virus only, while PI2 / 2F1B34 were incubated with 4 mM of peptide 2F1B34.

FIG. 5. Plaque size reduction by F peptides. Plaque assay was performed on Vero cells using PI2 virus. After 1 hr at 377, inoculum was removed and cells overlaid with a 1:1 mixture of 2% white agar with 21 DMEM containing 2% FBS (A) without addition of F peptide or (B) with addition of 4 mM of 2F1B34 in the overlay agar. Three days after overlaying, the agar was removed and the monolayer of cells was stained with crystal violet.

with the virus or with the cell membrane. Wild et al. (1995) proposed that peptide DP-107 corresponding to the leucine zipper-like motif near the fusion peptide region functions through interaction with the leucine zipper domain of gp41 and disruption of its functional role in the process of virus entry. Jiang et al. (1993a,b) proposed a different inhibition mechanism, in which the peptides interact with the N-terminal fusion domain of gp41, which disrupts insertion of the fusogenic sequence into the target cell membrane during virus entry. Rapaport et al. (1995) reported that peptides from the Sendai virus (SN) F protein located near the fusion peptide (similar to DP-107) did not have an anti-viral effect, while peptides derived from the leucine zipper-like region in the F protein adjacent to the transmembrane anchor inhibited the fusion of Sendai virus with human erythrocytes. Recent studies of peptides from the conserved heptad repeat region of other paramyxovirus fusion proteins, including respiratory syncytial virus (RSV), PI3, and measles virus (MV), also showed a potent inhibitory effect on viral fusion (Lambert et al., 1996). Our results are consistent with the results reported by Lambert et al. (1996). Furthermore, we found that the F peptides could also inhibit viral entry and multi-

AID

vira

VY 8092

/

6a1d$$$161

08-05-96 14:08:35

AP: Virology

110

YAO AND COMPANS

ple cycle replication. One possible mechanism for this inhibitory effect could involve an interaction of the peptides with the viral glycoproteins, affecting a conformational change in the F protein which may be required for fusion. Alternatively, the peptides may block the specific interaction of F and HN which is required for the fusion process. However, it has not yet been established that the peptides interact directly with the viral glycoproteins. We found that the functional concentrations of the purified peptides were comparable with those reported by others for HIV and paramyxoviruses (Wild et al., 1994a; Rapaport et al., 1995; Lambert et al., 1996). Peptide 3F1B35 corresponds to the PI3 peptide designated T210 by Lambert and co-workers. We found that 3F1B35 inhibited 50% of virus-induced cell fusion at a concentration of 1.2 mM (4.8 mg/ml), while in their assay T-210 was reported to have an EC50 of 2.4 mg/ml. Furthermore, they compared a series of overlapping 35-amino-acid peptides and reported that peptides containing additional amino acids at the N-terminal side of heptad repeat B domain showed little or no activity, while peptides extending to the C-terminal side showed more potent antiviral activity. Similarly, we observed that peptides which extend from the heptad repeat B to the beginning of the transmembrane domain were more potent inhibitors of virus-induced cell fusion than a peptide corresponding to the heptad repeat region only. Moreover, the F peptides derived from one virus could only inhibit cell fusion induced by this specific virus but not others. Lambert and co-workers (1996) also observed a high level of virus specificity of the inhibitory peptides, although some PI3 peptides were cross-reactive with respiratory syncytial virus and measles virus. Although the length of the peptides may be crucial to the inhibitory effect in both PI2 and PI3 viral-induced cell fusion, the specificity of peptides for fusion by a specific paramyxovirus is presumably related to differences in the F or HN sequences of these viruses. Sequence comparison of these regions showed no sequence homology except for a common leucine zipper-like motif which has the potential to form a coiled-coil secondary structure. Moreover, a chimeric HN containing an 82-amino acid region derived from SNHN just outside the membrane-spanning segment and the remainder derived from P13 HN is sufficient for almost full fusion activity when coexpressed with the SN F protein (Tanabayashi and Compans, 1996). Other recent studies with other paramyxoviruses (NDV and PI2) have also indicated that a similar region of HN is involved in fusion promotion (Deng et al., 1995; Tsurudome et al., 1995). It is likely that these HN sequences could interact with the sequences in the F protein which are just external to the transmembrane region. The heptad repeat B region and the sequences between the heptad repeat B and transmembrane domain may therefore be involved in F–HN interaction. We observed that the F peptides showed an inhibitory

AID

VY 8092

/

6a1d$$$161

08-05-96 14:08:35

effect on viral entry as well as the spread of infection after low m.o.i inoculation. HIV or Sendai virus-derived peptides (Wild et al., 1994a; Rapaport et al., 1995) did not interfere with the attachment of the virus to the target cells, and we have made similar observations for PI2 and PI3 peptides (data not shown). The F peptides therefore appear to inhibit viral entry at a stage after viral attachment, which is likely to be the fusion of the viral envelope with the cell membranes. In addition, we observed that the F peptides interfere with fusion induced by the F and HN glycoproteins expressed using the vaccinia-T7 expression system. Although cell fusion could be completely inhibited by the F peptides at similar concentrations, the F peptides did not completely block virus entry, as seen in a plaque reduction assay. In addition, the concentration of peptides required for 50% inhibition of fusion was somewhat lower than that required for 50% inhibition of plaque formation in both PI2 and PI3 virus. The peptides therefore appear to be more effective inhibitors of cell–cell fusion than fusion of the virus to the cell. Similarly, the HIV-1 peptide DP-178 which corresponds to the leucine zipper-like motif near the N-terminus of the transmembrane domain was a more potent inhibitor of cell–cell fusion than of infection by cell-free virus (Wild et al., 1994a). Although the reason for the difference in potency of the peptides in these two processes is unclear, it may reflect differences in the processes of cellfree virus transmission vs cell to cell fusion. Many peptides have short half lives when administered systemically. Because the F peptides are likely to be interacting with the external domains of the viral envelope proteins, it is also likely that they could be acting at the surfaces of infected cells and thus could prevent virus spread during multiple cycles of infection. It is therefore possible that the F peptides would inhibit virus infection when administered to the respiratory tract by aerosol. This may provide an approach for inhibiting virus infection in vivo, which may lead to control of the serious respiratory diseases in infants which result from parainfluenza virus infections. ACKNOWLEDGMENTS We thank Dr. Kiyoshi Tanabayashi for valuable discussions and suggestions. We also thank Lawrence R. Melsen for photography and Tanya Cassingham for assistance in preparing the manuscript. This study was supported by Research Grant AI 12680 from the National Institutes of Allergy and Infectious Diseases, NIH.

REFERENCES Bousse, T., Takimoto, T., Gorman, W. L., and Portner, A. (1994). Regions on the hemagglutinin-neuraminidase proteins of human parainfluenza virus type 1 and Sendai virus important for membrane fusion. Virology 204, 506–514. Buckland, R., and Wild, F. (1989). Leucine zipper motif extends. Nature 338, 547. Buckland, R., Malvoisin, E., Beauverger, P., and Wild, F. (1992). A leucine zipper structure present in the measles virus fusion protein is not

vira

AP: Virology

HEPTAD REPEAT SEQUENCE VIRAL INHIBITORS

111

required for its tetramerization but is essential for fusion. J. Gen. Virol. 73, 1703–1707. Bullough, P. A., Hughson, F. M., Skehel, J. J., and Wiley, D. C. (1994). Structure of influenza haemagglutinin at the pH of membrane fusion. Nature 371, 37–43. Carr, C. M., and Kim, P. S. (1993). A spring-loaded mechanism for the conformational change of influenza hemagglutinin. Cell 73, 823–832. Cattaneo, R., and Rose, J. K. (1993). Cell fusion by the envelope glycoproteins of persistent measles viruses which caused lethal human brain disease. J. Virol. 67, 1493–1502. Chambers, P., Pringle, C. R., and Easton, J. J. (1990). Heptad repeat sequences are located adjacent to hydrophobic regions in several types of virus fusion glycoproteins. J. Gen. Virol. 71, 3075–3080. Chen, S.-L., Lee, C.-N., Lee, W.-R., McIntosh, K., and Lee, T.-H. (1993). Mutational analysis of the leucine zipper-like motif of the human immunodeficiency virus type 1 envelope transmembrane glycoprotein. J. Virol. 67, 3615–3619. Choppin, P. W., and Compans, R. W. (1975). Reproduction of paramyxoviruses. In ‘‘Comprehensive Virology’’ (H. Fraenkel-Conrat and R. R. Wagner, Eds.), Vol. 4, pp. 95–178. Plenum, New York. Choppin, P. W., and Scheid, A. (1980). The role of viral glycoproteins in adsorption, penetration and pathogenicity of viruses. Rev. Infect. Dis. 2, 40–61. Deng, R., Wang, Z., Mirza, A. M., and Iorio, R. M. (1995). Localization of a domain on the paramyxovirus attachment protein required for the promotion of cellular fusion by its homologous fusion protein spike. Virology 209, 457–469. Doms, R. W., White, J. M., Boulay, F., and Helenius, A. (1990). Influenza virus hemagglutinin and membrane fusion. In ‘‘Cellular Membrane Fusion: Fundamental Mechanisms and Applications of Membrane Fusion Techniques’’ (J. W. D. Hoekstra, Ed.). Dekker, New York. Dubay, J. W., Roberts, S. J., Brody, B., and Hunter, E. (1992). Mutations in the leucine zipper of the human immunodeficiency virus type 1 transmembrane glycoprotein affect fusion and infectivity. J. Virol. 66, 4748–4756. Gittman, A. G., and Loyter, A. (1984). Construction of fusogenic vesicles bearing specific antibodies: targeting of reconstituted Sendai virus envelopes towards neuraminidase-treated human erythrocytes. J. Biol. Chem. 259, 9813–9820. Heath, T. D., Martin, F. J., and Macher, B. A. (1983). Association of ganglioside-protein conjugates into cell and Sendai virus: requirement for the HN subunit in viral fusion. Exp. Cell Res. 149, 163–175. Heminway, B. H., Yu, Y., and Galinski, M. S. (1994). Paramyxovirus mediated cell fusion requires co-expression of both the fusion and hemagglutinin-neuraminidase glycoproteins. Virus Res. 31, 1–16. Homma, M., and Ohuchi, M. (1973). Trypsin action on the growth of Sendai virus in tissue culture cells. III. Structural differences of Sendai viruses grown in eggs and tissue culture cells. J. Virol. 12, 1457– 1463. Horvath, C. M., Paterson, R. G., Shaughnessy, M. A., Wood, R., and Lamb, R. A. (1992). Biological activity of paramyxovirus fusion proteins: factors influencing formation of syncytia. J. Virol. 66, 4564– 4569. Hsu, M.-C., Scheid, A., and Compans, R. W. (1979). Reconstitution of membranes with individual paramyxovirus glycoproteins and phospholipid in cholate solution. Virology 95, 476–491. Hsu, M.-C., Scheid, A., and Choppin, P. W. (1981). Activation of the Sendai virus fusion protein (F) involved a conformational change with exposure of a new hydrophobic region. J. Biol. Chem. 256, 3557– 3563. Hsu, M.-C., Scheid, A., and Choppin, P. W. (1982). Enhancement of membrane-fusion activity of Sendai virus by exposure of the virus to basic pH is correlated with a conformational change in the fusion protein. Proc. Natl. Acad. Sci. USA 79, 5862–5866. Hu, X., Ray, R., and Compans, R. W. (1992). Functional interactions between the fusion protein and hemagglutinin-neuraminidase of human parainfluenza viruses. J. Virol. 66, 1528–1534.

Jiang, S., Lin, K., Strick, N., and Neurath, A. R. (1993a). HIV-1 inhibition by a peptide. Nature 365, 113. [Letter] Jiang, S., Lin, K., Strick, N., and Neurath, A. R. (1993b). Inhibition of HIV-1 infection by a fusion domain binding peptide from the HIV-1 envelope glycoprotein GP41. Biochem. Biophys. Res. Commun. 195, 533–538. Kemble, G. W., Danieli, T., and White, J. M. (1994). Lipid-anchored influenza hemagglutinin promotes hemifusion, not complete fusion. Cell 76, 383–391. Lamb, R. A. (1993). Paramyxovirus fusion: A hypothesis for changes. Virology 197, 1–11. Lambert, D. M., Barney, S., Lambert, A. L., Guthrie, K., Medinas, R., Davis, D. E., Bucy, T., Erickson, J., Merutka, G., and Perreway, S. R., Jr. (1996). Peptides from conserved regions of paramyxovirus fusion (F) proteins are potent inhibitors of viral fusion. Proc. Natl. Acad. Sci. USA 93, 2186–2191. Malvoisin, E., and Wild, T. F. (1993). Measles virus glycoproteins: studies on the structure and interaction of the hemagglutinin and fusion proteins. J. Gen. Virol. 74, 2365–2372. Merz, D. C., and Wolinsky, J. S. (1983). Conversion of nonfusing mumps virus infections by selective proteolysis of the HN glycoprotein. Virology 131, 328–340. Miura, N., Uchida, T., and Okada, Y. (1982). HVJ (Sendai virus)-induced envelope fusion and cell fusion are blocked by monoclonal anti-HN protein antibody that does not inhibit hemagglutinin activity of HVJ. Exp. Cell Res. 141, 409–420. Morrison, T. G., and Portner. (1991). Structure, function, and intracellular processing of the glycoprotein of Paramyxoviridae. In ‘‘The Paramyxoviruses’’ (D. W. Kingsbury, Ed.), pp. 347–382. Plenum, New York. Nussbaum, O., Sakai, N., and Loyter, A. (1984). Membrane bound antiviral antibodies as receptors for Sendai virions in receptor depleted erythrocytes. Virology 138, 185–197. O’Shea, E., Rutkowski, R., and Kim, P. (1989). Evidence that the leucine zipper is a coiled-coil. Science 243, 538–543. Paterson, R. G., Shaughnessy, M. S., and Lamb, R. A. (1989). Analysis of the relationship between cleavability of paramyxovirus fusion protein and length of the connecting peptide. J. Virol. 63, 1293–1301. Portner, A., Scroggs, R. A., and Metzger, D. W. (1987). Distinct functions of antigenic sites of the HN glycoprotein of Sendai virus. Virology 156, 61–68. Rapaport, D., Ovadia, M., and Shai, Y. (1995). A synthetic peptide corresponding to a conserved heptad repeat domain is a potent inhibitor of Sendai Virus-cell fusion: an emerging similarity with functional domains of other viruses. EMBO J. 14, 5524–5531. Reitter, J. N., Sergel, T., and Morrison, T. G. (1995). Mutational analysis of the leucine zipper motif in the Newcastle disease virus fusion protein. J. Virol. 69, 5995–6004. Ruigrok, R. W., Aitken, A., Calder, L. J., Martin, S. R., Skehel, J. J., Wharton, S. A., Weis, W., and Wiley, D. C. (1988). Studies on the structure of the influenza virus HA at the pH of membrane fusion. J. Gen. Virol. 69, 2785–2795. Scheid, A., and Choppin, P. W. (1974). Identification and biological activities of paramyxovirus glycoproteins. Activation of cell fusion, hemolysis and infectivity by proteolytic cleavage of an inactive precursor protein of Sendai virus. Virology 57, 475–490. Sergel, T., McGinnes, L. W., Peeples, M. E., and 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. Sergel, T. G., McQuain, C., and Morrison, T. G. (1994). Mutations in the fusion peptide and heptad repeat regions of the Newcastle disease virus fusion protein block fusion. J. Virol. 68, 7654–7658. Shibuta, H., Nozawa, A., Shioda, T., and Kanda, T. (1983). Neuraminidase activity and syncytial formation in variants of parainfluenza 3 virus. Infect. Immun. 41, 780–788. Stegmann, T., Booy, F. P., and Wilschut, J. (1987). Effects of low pH on

AID

vira

VY 8092

/

6a1d$$$161

08-05-96 14:08:35

AP: Virology

112

YAO AND COMPANS

influenza virus. Activation and inactivation of the membrane fusion capacity of the hemagglutinin. J. Biol. Chem. 262, 17744–17749. Tanabayashi, K., and Compans, R. W. (1996). Functional interaction of paramyxovirus glycoproteins: Identification of a domain in Sendai virus HN which promotes cell fusion, J. Virol., in press. Thompson, S. D., and Porter, A. (1987). Localization of functional sites on the hemagglutinin-neuraminidase glycoprotein of Sendai virus by sequence analysis of antigenic and temperature sensitive mutants. Virology 160, 1–8. Tsurudome, M., Hamada, A., Hishiyama, M., and Ito, Y. (1986). Monoclonal antibodies against the glycoproteins of mumps virus: Fusion inhibition by anti-HN monoclonal antibody. J. Gen. Virol. 67, 2259– 2265. Tsurudome, M., Kawano, M., Yuasa, T., Tabata, N., Nishino, M., Komada, H., and Ito, Y. (1995). Identification of regions on the hemagglutininneuraminidase protein of human parainfluenza virus type 2 important for promoting cell fusion. Virology 213, 190–203. Waxham, M. N., and Wolinsky, J. S. (1986). A fusing mumps virus variant selected from a nonfusing parent with the neuraminidase inhibitor 2deoxy-2,3-dehydro-N-acetylneuraminic acid. Virology 151, 286–295. Welkos, S., and O’Brien, A. (1994). Determination of median lethal and infectious doses in animal model systems. Methods Enzymol. 235, 29–39. Wharton, S. A. (1987). The role of influenza virus haemagglutinin in membrane fusion. Microbiol. Sci. 4, 119–124.

White, J. M., and Wilson, I. A. (1987). Anti-peptide antibodies detect steps in a protein conformational change: Low pH activation of the influenza virus HA. J. Cell Biol. 105, 2887–2896. Wild, C. T., Oas, T., McDanal, C. B., Bolognesi, D., and Matthews, T. J. (1992). A synthetic peptide inhibitor of human immunodeficiency virus replication: Correlation between solution structure and viral inhibition. Proc. Natl. Acad. Sci. USA 89, 10537–10541. Wild, C. T., Shugars, D. C., Greenwell, T. K., McDanal, C. B., and Matthews, T. J. (1994a). Peptides corresponding to a predictive a-helical domain of human immunodeficiency virus type 1 gp41 are potent inhibitors of virus infection. Proc. Natl. Acad. Sci. USA 91, 9770– 9774. Wild, C. T., Greenwell, T. K., Shugars, D., Rimsky-Clarke, L., and Matthews, T. J. (1995). The inhibitory activity of an HIV type 1 peptide correlates with its ability to interact with a leucine zipper structure. AIDS Res. Hum. Retroviruses 11, 323–325. Wild, T. F., Fayolle, J., Beauverger, P., and Buckland, R. (1994b). Measles virus fusion: Role of the cysteine-rich region of the fusion glycoprotein. J. Virol. 68, 7546–7548. Wilson, I. A., Skehel, J. J., and Wiley, D. C. (1981). Structure of the hemagglutinin membrane glycoprotein of influenza virus at 3 A resolution. Nature (London) 289, 366–373. Yao, Q., and Compans, R. W. (1995). Differences in the role of the cytoplasmic domain of human parainfluenza virus fusion proteins. J. Virol. 69, 7045–7053.

AID

vira

VY 8092

/

6a1d$$$161

08-05-96 14:08:35

AP: Virology