Evaluation of Glu11 and Gly8 of the H5N1 influenza hemagglutinin ...

Arch Virol (2008) 153: 247–257 DOI 10.1007/s00705-007-1088-9 Printed in The Netherlands

Evaluation of Glu11 and Gly8 of the H5N1 influenza hemagglutinin fusion peptide in membrane fusion using pseudotype virus and reverse genetics Y. Su1;2 , X. Zhu1 , Y. Wang1 , M. Wu1 , P. Tien1 1 2

Center for Molecular Virology, Chinese Academy of Sciences, Institute of Microbiology, Beijing, P.R. China Graduate School of the Chinese Academy of Sciences, Beijing, P.R. China

Received 17 May 2007; Accepted 26 September 2007; Published online 22 November 2007 # Springer-Verlag 2007

Summary

Influenza viruses gain entry into host cells by binding to cellular receptors and promoting the fusion of the viral envelope with the host cell membrane. The fusion peptide of influenza hemagglutinin (HA) is crucial for fusion. To examine the structural and functional roles of amino acids E11 and G8 of the H5 HA fusion peptide, a series of fusion mutants was generated. We determined the effect of each mutation on fusion activity and infection of rescued recombinant virus by polykaryon formation, cell–cell fusion assay, HA pseudovirus transduction and reverse genetics. Our findings indicate that E11V and E11A mutants dramatically inhibit fusion and that at position 11 a polar residue such as glutamic acid or serine may be desirable for preserving the fusion activity. More interestingly, one mutation (G8E) raised the threshold pH of polykaryon formation. Our results suggest that G8 as well as E11 play an important functional and structural role in membrane fusion and that the po-

Correspondence: Po Tien, Center for Molecular Virology, Chinese Academy of Sciences, Institute of Microbiology, Beijing 100080, P.R. China e-mail: [email protected]

larity of E11 is crucial for fusion activity. Finally, we developed an assay based on a reporter gene plus pseudotyped virus that could sensitively detect fusion activity.

Introduction Influenza virus enters mammalian cells by receptormediated endocytosis and subsequent fusion of the viral and endosomal membranes triggered in the pH 5 environment within the endosome [1, 2, 7, 12, 30]. Structural analysis has provided information about the ectodomains of both the native protein at the neutral pH [4, 30] and a portion of HA2 at a low pH [1, 3, 5]. The amino-terminal stretch of HA2 consisting of 20 amino acid residues, termed fusion peptide, is directly involved in the fusion reaction [29]. To elucidate the role of this highly conserved sequence in the fusion process, numerous mutations have been made at various positions within the fusion peptide region [8, 9, 11, 21, 23, 25, 28]. The unusually high content of glycine residues in the fusion peptide sequence have been the focus of intensive studies, which have included mutations at positions 1, 4, 8, and 13 of HA2 (Table 1). These

248

Y. Su et al.

Table 1. Amino acid substitution at selected positions of H5 HA HA

wt E11G E11V E11A E11S G8V G8E

329

R – – – – – –

Amino acid at position G

L

F

G

A

I

A

G

F

I

E

G

G

W

Q

G

M

V

D

G

– – – – – – –

– – – – – – –

– – – – – – –

– – – – – – –

– – – – – – –

– – – – – – –

– – – – – – –

– – – – – V E

– – – – – – –

– – – – – – –

– G V A S – –

– – – – – – –

– – – – – – –

– – – – – – –

– – – – – – –

– – – – – – –

– – – – – – –

– – – – – – –

– – – – – – –

– – – – – – –

glycines are thought to perform an important structural and=or functional role because of their conformational plasticity [3]. Among these glycine residues, G1 and G8 have been found to be more sensitive to mutations affecting fusogenicity than the other two [13, 25]. Substitutions of glycine at the N terminus with valine or glutamic acid have been found to inhibit fusion activity [11, 21, 28]. Meanwhile, substitution of G8 with alanine results in greatly impaired polykaryon formation [13, 25]. It has recently been determined that G1, G4 and G8 formed a smooth ‘‘glycine edge’’ at the top of the N-terminal arm of the fusion peptide [19]. The conserved glycine ridge containing the GxxxG motif is known to stabilize helix-helix interactions in both membrane and soluble proteins [22]. Furthermore, this glycine ridge appears to be required for the important interaction between the transmembrane (TM) domain and the fusion peptide. In the present study, we have replaced the G8 with valine or glutamic acid to examine the effects of these mutations

on the fusion activity of H5 and on the infection capability of rescued recombinant virus. Recent studies have shown that the fusogenic activity of the fusion peptide is closely correlated with the amino acid composition of its first 11 residues [26]. According to the structure of the fusion peptide by NMR and EPR spectroscopy, the 20 residues of the fusion peptide form two short helices at pH 5, which are separated by a 105 kink at N12 [14]. The residue E11 is involved in stabilizing the kink structure [19]. In addition, except for E11, most of the ionizable residues of the fusion peptide in the fusogenic state are neutral at pH 5.0 [6]. Thus, we hypothesized that the structural and biochemical features of E11 in the context of the fusion peptide might significantly influence HAmediated membrane fusion. In other studies on two synthetic mutants of influenza HA2 fusion peptide (residues 1–25), introduction of glutamic acid into the fusion peptide led to increased sensitivity of various biochemical properties to pH compared

Fig. 1. Amino acid sequence alignment of the fusion peptide genes from different subtypes of HA selected in this study. The figure was generated with BioEdit 7.0 software

Effect of mutations in influenza virus HA fusion peptide

to wild type [16]. These observations suggested that the nature of membrane fusion could be altered by substitution of this charged residue. A modified fusion peptide (E11V=E15V) of the influenza HA was reported to exhibit greatly increased fusion of lipid vesicles regardless of the pH value [17]. The susceptibility of E11 to some specific amino acid alteration may provide an interesting clue about the functional role of E11 in membrane fusion. A sequence comparison of the HA2 fusion peptides from various influenza viruses tested in this study indicated that G1, L2, G4, A5, I6, A7, G8, F9, I10, E11,G13, W14 and G16 of HA2 are highly conserved (Fig. 1). However, different from the influenza virus H3 HA, the H5 subtype protein contains glycine at position 12 instead of asparagine, indicative of stabilizing the structure of the kink in the HA fusion peptide [19]. We supposed that the glycine might contribute to a different fusion phenotype between the H5 HA and that of other subtypes. Most of the information available concerning HA fusion function has come from studies of the A=Aichi=2=68 (H3N2) strain of the virus. However, a greater understanding of the HA fusion properties of different viral strains is needed. The focus of our present work was to examine the fusion properties of H5 subtype protein from the highly pathogenic (HP) avian virus. Different from those of many influenza virus strains, the HA of HP H5 avian influenza virus (AIV) is expressed at the cell surface as a cleaved form [27] whose fusion-active state makes it convenient for studying membrane fusion. To investigate the structural and functional role of E11 and G8 of the H5 HA fusion peptide, we substituted E11 and G8 with some specific residues. Our study revealed that the E11 and G8 residues play an important functional and structural role in membrane fusion and that the polarity of E11 is crucial for fusion activity. Furthermore, the E11G and E11V mutants of H5 HA had some discrepancies in fusion activity compared with previous studies, and our study with the G8E mutant adds new knowledge in this area. Our results also demonstrate that virus-cell and cell–cell fusion have significant differences and that pseudotyped virus can

249

serve as a sensitive and rapid assay for detecting fusion activity. Materials and methods Cells HeLa and 293T cells were maintained in Dulbecco’s modified Eagle’s medium (DMEM) supplemented with 10% (v=v) fetal calf serum (FCS) (Invitrogen, Karlsruhe, Germany). MDCK cells were kept in minimal essential medium (MEM) supplemented with 10% FCS and streptomycin (100 mg=ml) and penicillin (100 U=ml) at 37 C, 5% (v=v) CO2. Mutagenesis A cDNA copy of the entire wild-type(wt) AIV-HA and neuraminidase (NA) genes of A=chicken=Fujian=1042=2005 (H5N1) was cloned into the Xho I and Not I sites of the pCI-neo mammalian expression vector (Promega, Madison, WI) under the control of the cytomegalovirus (CMV) promoter. The point mutant plasmid clones were generated using a QuickChange site-directed mutagenesis kit (Stratagene) with wt HA as template (Table 1). Antibodies and immunofluorescence staining of HA-expressing cells Polyclonal chicken antiserum raised against AIV H5N1 was purchased from the Institute of Harbin Veterinary Research. HA-expressing 293T cell monolayers on coverslips were washed once with PBS and fixed with a chilled solution of 4% formaldehyde. The fixed cells were washed and incubated with a 1:200 dilution of anti-chicken polyclonal antibody at 37 C for 1 h, then washed and incubated with fluoresceinconjugated goat anti-chicken immunoglobulin G antibodies (Sigma, St. Louis, MO) at 1:1000 dilution. The slides were observed using a fluorescence microscopy (TE300; Nikon, Tokyo, Japan). Flow cytometric analysis of immunostained wt and mutant-HA-expressing cells Forty-eight hours after transfection with HA-expression plasmids, 293T and HeLa cells were detached and washed with ice-cold FACS buffer (1% bovine serum album diluted in phosphate buffer saline (PBS, pH 7.2), then resuspended in FACS buffer containing a 1:500 dilution of chicken antiHA polyclonal antibody. After a 1-h incubation on ice, the cells were washed twice and resuspended in 1 ml of FACS buffer containing a 1:200 dilution of fluorescein-isothiocyanate-conjugated goat anti-chicken IgG. Afterwards, the cells were incubated on ice for 1 h, washed, and analyzed on a FACSSCAN (FACScan, Becton Dickinson, Heidelberg, Germany).

250

Y. Su et al.

Cell–cell fusion assays

particles in serum-free DMEM was added to wells for 6 h at 37 C before addition of serum-containing medium. Two days post-transduction, the cells were lysed and the luciferase activity in 20 ml lysate was assayed in a luminometer using commercially available reagents (Promega).

HA-induced cell-to-cell fusion was quantitatively determined by measuring firefly luciferase gene expression. 293T effector cells in 6-well plates (5 105 cells per well) were transfected with expression plasmids encoding the HA envelope protein (1.5 mg), together with two reporter plasmids, pT7EMCVLuc (0.5 mg) and pRL-TK (0.05 mg) (Promega, Madison, WI) using lipofectAmine 2000 (Invitrogen, Carlsbad, CA). pT7EMCVLuc had a firefly luciferase gene under the control of the T7 promoter. pRL-TK had a renilla luciferase gene under the control of the TK promoter, and expression of the pRL-TK gene was used as an internal reference for monitoring firefly luciferase gene transfection efficiency. The target cells (293 T cells) in 6-well plates were transfected with 2 mg of plasmid encoding T7 RNA polymerase. Between 36 and 48 h after transfection, the effector cells were mixed with the target 293T cells and incubated for 5 h. The co-cultured cells were bathed in PBS at pH 5.0 for 3 min at 37 C and then incubated with DMEM containing 10% FCS at pH 7.2 for 7 h. Then, the cells were harvested in passive lysis buffer (Promega) and the lysates were assayed for luciferase activity using the dual-luciferase repoter assay system (Promega). Relative light units were measured with a luminometer (Promega). The cell fusion activity was determined by measuring firefly luciferase gene expression in the lysates of the cocultured cells after it was normalized to the basal luciferase gene expression.

Results

Syncytia formation

Cell surface expression of wt and mutant HA 5

HeLa cells (2 10 per well; six-well plates) were transfected with expression vectors for VSV-G or pCI-HA AIV HA envelope expression plasmid or its fusion mutants (2 mg) using lipofectamine 2000 (Invitrogen). Two days after transfection, the cells were incubated for 5 min at 37 C with fusion buffer (pH 5.0), and then with complete medium at 37 C for 4 h to allow the formation of syncytia. Then, the cells were fixed with 0.5% glutaraldehyde (Sigma) for 10 min and stained with crystal violet (Sigma). Pictures were taken at a 20 magnification. Pseudotype production and virus-to-cell fusion assay 293T cells (60–70% confluent) were transfected with AIV H5N1 HA and NA envelope expression plasmid plus HIV reporter plasmid (pNL-luc-E R ) by using the calcium phosphate precipitation method. Forty-eight hours after transfection, the culture supernatants were harvested and filtered through a 0.45-mm filter (Millipore, Watford, United Kingdom). Aliquots were stored at 70 C. The amount of HA-HIV-like particles released into the culture supernatants was quantitated by HIV-1 p24 antigen capture enzymelinked immunosorbent assay (ELISA) (Retro-Tek). Sub-confluent 293T cells (in 24-well plates) plated at a density of 2 104 per well were transduced with an equal amount pseudotyped virus (50 ng). The inoculum of pseudo-

Generation of fusion peptide mutant virus by reverse genetics To obtain chimeric viruses containing HA or its fusion mutants and neuraminidase (NA) of H5N1 AIV in the background of A=WSN=33 (H1N1) virus, the H5N1 AIV HA or its fusion mutants and NA cDNA were introduced into the RNA expression plasmid pHH21. Influenza viruses were rescued from plasmid cDNA essentially as described by Neumann et al. [20]. Reassortant viruses contained the HA and NA genes from H5N1 virus in the genetic background of A=WSN=33(H1N1) were generated by transfecting 293T cells with 0.2 mg each of 12 plasmids for expression of the viral genomic RNAs and protein (kindly provided by Y. Kawaoka) using Lipofectamine 2000 transfection reagent (Invitrogen) following the manufacturer’s instructions. Supernatants from transfected cells at 3 or 4 days post-transfection were passaged once in MDCK cells and titrated on MDCK cells by plaque assay.

To determine the expression level of HA protein and its transportation to the cell surface, the 293T and HeLa cells transiently transfected with HA-expression plasmids were analyzed by FACS analysis to quantitate the HA expression level and by indirect immunofluorescence analysis to determine the localization of the proteins. As shown in Fig. 2A, the HA proteins were transported to the 293T cell surface, as indicated by immunofluorescence analysis. Results obtained from FACS assays revealed that a smaller percentage of cells expressed E11V, E11S and E11A mutants than the cells that expressed other mutants and wt (Table 1 and Fig. 2B, ranging from 42 to 142% of the wild-type level). Membrane fusion activity of mutant HAs measured by syncytia formation assay The pH at which influenza HA mediates membrane fusion can vary, and this property can have biological significance. To investigate the optimal fusion pH of these HA mutants, we performed the syncytia


251

Fig. 2. Cell-surface expression of wt HA and mutants. (A) Cell-surface expression of mutant proteins analysed by immunofluorescence. 293T cells were transfected with the expression plasmids encoding full-length wt and mutant HA proteins. At 40 h post-transfection, the cells were fixed with formalin, immunostained and photographed. (B) Cell-surface expression of mutant proteins analyzed by FACS. The values are expressed as percentages of that of the wt HA protein. The results are the averages of three independent experiments

formation assay at pH 4.8, 5.0, 5.3, 5.6 and 6.0. The capacity of wt and mutant HAs to mediate membrane fusion was assayed by observing the syncytia formation of HA-expressing cells. The mutant E11G was able to induce syncytia at pH 4.8, 5.0 and 5.3 like wt HA, but the polykaryon formation was consistently less extensive than that of wt HA (Fig. 3A and B) when the number of the syncytia from 5 fields of microscopy were counted. The number of syncytia decreased when the pH increased, while other mutants displayed no polykaryon formation at pH 4.8 and 5.0. Meanwhile, the VSV-G control could induce apparent polykaryon formation over a broad pH range. Different from a previous report [23] in which an E11G mutant of

A=Japan=305=57 (H2 subtype) HA showed no syncytial activity, our result suggests that the glycine at position 11 may be a substitutable one. This difference may be due to the viral subtype of HA, the type of cells or differences in HA expression levels. A particularly interesting observation was that the G8E mutant could induce distinct polykaryon formation only at pH 5.3, which is related to the glutamic acid substitution. Quantitative cell-to-cell fusion and pseudotyped virus infection assay for the HA and its mutants To examine the cell fusion activity of these H5 HA mutants, we established a highly sensitive and

252

Y. Su et al.

Fig. 3A and B. Low-pH-induced polykaryon formation of HeLa cells expressing wild-type or mutant HA proteins. HeLa cells were transfected with pCI-HA vectors containing the wild-type or mutant HA genes. The cells were tested at low pH values, 4.8, 5.0 and 5.3, for their ability to form polykaryons. The cells were then fixed and photographed


Fig. 4. Cell–cell fusion assay of 293T cells expressing the wt or mutant HA. Luciferase activities in lysates of cocultures of transfected 293T cells were measured to reflect the extents of the Env-mediated cell–cell fusion activities after normalization (firefly luciferase activity=sea pansy luciferase activity make sure which vector for which part of the figure is labeled), were measured. Each data point was determined for triplicate samples

quantitative reporter activation method. We used the luciferase reporter gene to enhance the sensitivity and used the internal standard to normalize the transfection efficiency of this reporter gene. The cell fusion activity determined using the luciferase assay showed that, despite an apparently reduced efficiency compared to the wt HA, the E11G mutant displayed lower cell fusion activity at pH 5.0 (Fig. 4). It is possible that the difference between the cell–cell fusion and polykaryon formation of E11G was due to the high sensitivity of luminescent readout, resulting in a remarkably increased positive value. However, other mutants displayed little or no cell–cell fusion activity. On the other hand, the G8E and G8V mutants exhibited an increased cell fusion activity at pH 5.3 while other mutants and wt showed a reduced tendency of cell fusion at the same pH. Although the cell–cell fusion assay is a useful method for studying virus fusion, differences between cell–cell fusion and virus-cell fusion exist, and hence there is a need for direct assays of viruscell fusion. Here, we utilized a simple system by generating a HIV-luc(AIV) pseudotyped virus to assess the virus-cell fusion ability of these HA mutants, which enabled us to study the fusion function of HA through a single-cycle transduction. The ability of the HA mutant proteins to promote viruscell fusion was assessed by titering the pseudotyped

253

Fig. 5. Infectivity of HA-pseudotyped HIV-luc viral particles on 293T cells. The titers for wt HA- pseudotyped virus were set as 100% to calculate the relative infectivity of the mutant viruses. A representative of two experiments is shown

virus on 293T cells (Fig. 5). Our results indicated that a substitution of serine at position 11 appears to be tolerated, which was different from the results observed using the cell–cell fusion assay (Fig. 6). It was noted that although valine is more hydrophobic than the glutamic acid of the wild-type peptide, the E11V mutant was significantly inactive in the virusto-cell fusion assay.

Fig. 6. Application of Harvey ball analysis to evaluate the results of the wt and six mutations in the different assays applied: cell-surface expression, polykaryon formation, reporter-gene-based cell–cell fusion, virus-cell fusion and reverse genetics. Levels of expression and fusion are relative to that of wt HA. 3 75–100%, 0–5%, 5–25%, 6 25– 50%, 50–75%

254

Y. Su et al.

Table 2. Virus titres of rescued virus in transfected 293T cell supernatants HA

Virus rescued

Virus titres (p.f.u.=ml)

wt E11G E11V E11A E11S G8V G8E

þ þ þ þ þ þ þ

2.3 105 2.7 104 2.0 l03 2.6 104 1.6 105 1.0 104 1.0 103

MDCK cell monolayers were infected with wild-type and its mutant HA recombinant virus. Titers of progeny virus released into the medium were determinded by plaque assay.

Infectivity of fusion peptide mutant viruses generated by reverse genetics To evaluate the effect of these mutations on virus replication and infectivity, the mutant HAs were incorporated into infectious viruses by reverse genetics. The reverse-genetics viruses bearing the mutant HAs and neuraminidase (NA) of AIV H5N1 in a background of genes of the WSN strain was rescued by co-transfecting 293T cells with 12 plasmids. As shown in Table 2, both the wt and mutant viruses could be rescued as infectious viruses. Notably, the E11V and E11A mutants, which displayed little or no fusion activity using polykaryon formation and cell–cell fusion assays, could be rescued. The yields of mutant viruses E11G, E11V, E11A, G8V and G8E were one to two orders of magnitude less than that of the rescued wt. In particular, the E11S mutant was generated at titers comparable to that of wt. These data demonstrate that the E11S mutation may offer an advantage for efficient virus replication, while the E11V and G8E mutation is likely to cause a defect in viral replication. Disscussion Structural information at the atomic and molecular level available for influenza HA indicates the existence of at least three fusion peptide structures, which impose distinct sequence constraints that are only now starting to be understood [8, 24]. Much of the previous work based on site-specific

mutagenesis and synthetic fusion peptide analogues have been focused on the first three glycine residues, and in particular the glycine at the N-terminus [11, 13, 16, 21, 25, 28]. To further understand the significant roles of the glycine ridge and G8 residue, we introduced the bulkier hydrophobic residue valine or charged polar residue glutamic acid at position 8. Our results revealed that the G8E mutant also decreased fusion efficiency and increased the pH threshold for fusion activity. It has been found that the HA2 G4E substitution decreased fusion efficiency and elevated the pH threshold for activation of the process [9, 10]. The G4E=G8E double mutant has been observed to increase the fusion activity of HA as well as the pH sensitivity of various biochemical properties [16]. In combination, these results suggest that incorporation of negatively charged residues into HA impart a great sensitivity to pH changes, and the protonation of glutamic acid residues appears to lower the energy barrier. This is necessary for the low-pH-induced conformational transition of HA2 to the fusion-active state, allowing it to undergo the fusion-inducing conformational change at a higher pH. In the present study, similar to the fusion phenotype of the mutant G1V [21], substitution of G8 within HA2 by bulky non-polar valine resulted in greatly impaired fusogenicity of HA. Although the hydrophobic effect of valine could serve as a driving force for fusion, the disruption of the glycine edge might remarkably affect its interaction with other helices in membrane or some residues of the TM domain and lead to impaired fusion activity. According to the cell fusion and infection data of G8E and G8V mutants, the fusogenicity of HA was affected more significantly by glutamic acid than by valine, suggesting that the electrostatic repulsion generated by glutamic acid played a more important role in changing the fusion activity of HA than the space occupied by valine. It was recently proposed that the residues from E11 to G20 of peptides form a stable amphipathic segment. Four polar residues (E11, N12, E15 and D19) are arranged as a hydrophilic face, whereas three non-polar residues (W14, M17, and I18) form a hydrophobic face [15]. In this study, replacement of the polar and hydrophilic E11 residues of H5 HA


with a slightly hydrophilic residue, glycine, could preserve the fusion activity, albeit at lower levels than that of the wild type. However, the E11G mutant has been reported to induce fusion of erythrocytes with mammalian cells at the same efficiency and pH as the wild-type protein [11]. The difference for the E11G mutant between our result and previous work was likely the result of the different techniques or the fusion peptide of different subtype used. In the present study, the hydrophilicity of alanine was lower than that of glycine or serine. The E11A mutants showed a lower level of cell– cell fusion and recombinant virus infection, consistent with a previous observation that the E11A mutation causes hemifusion [18]. Thus, it is reasonable to suggest that this specific arrangement of hydrophobic and hydrophilic residues is crucial for the formation of a specific conformation that may be essential for membrane fusion. In agreement with this, when E11 was substituted by a non-polar and hydrophobic residue, valine, the fusion activity was particular low, suggesting that the valine substitution had a dramatic inhibition effect on fusion function. The dramatically hydrophobic and non-polar effect of valine may be disadvantageous for maintaining the fusion-active conformation, which is related to the model of the fusion peptide. The polar face of the amphipathic a-helix is pointed inward to reduce the unfavorable free energy caused by immersing these residues of the polar face in the hydrophobic milieu [3]. However, the E11V=E15V double mutant has been reported to have higher membrane fusion activity than the wild-type peptide [17]. The difference between E11V and E11V=E15V might be due to the double mutation or the fusion peptide of different subtype used. Taken together, for position 11 of the fusion peptide, the glycine was more suitable for this position than alanine or valine for maintaining fusion activity, and the ideal arrangement would include polar and hydrophilic residues like glutamic acid and serine. Our studies using reverse genetics showed that the degree and efficiency for incorporation of HA into infectious virus correlates with its virus-cell fusion activity. Previously, using a variety of cell– cell fusion assays including polykaryon formation

255

assays and a fluorescent dye transfer assay [11, 13, 16, 18, 21, 25] based on the expressed HA, the fusogenic properties of a number of mutants have been studied for changes in their fusion ability. Employing two cell–cell fusion assays, polykaryon formation and the luciferase reporter assay, we observed differences between cell–cell fusion and virus-cell fusion (Fig. 6), which are likely related to a strict requirement for a high level of HA expression for the cell–cell fusion assay. According to our observations, although the E11V was rescued at significantly reduced efficiency compared to the wild-type virus, this mutant was able to mediate biologically functional fusion activity. This indicated that the mutant had greatly impaired fusion activity but was not totally fusion inactive, in contrast to the result from the cell–cell fusion assay. The E11S mutation was rescued at a level comparable to the wt, but no polykaryon formation and little cell fusion activity were detected. The difference between reverse genetics and the cell– cell fusion assay might be due to differences in the E11S mutant expression level caused by the serine substitution. In our studies, the mutant HA proteins had been inserted into pseudotyped virus to assess the ability of the HA proteins to direct fusion with susceptible cells. Although the pseudotyped viruses undergo a single cycle of replication, they are capable of efficiently transducing cells in an HA-mediated manner, which enables us to quantitatively measure virus-cell fusion. Comparing the results with standard cell–cell fusion assays and reverse genetics, the use of pseudotyped HIV=AIV virus to analyze the function of the HA fusion mutants is positively correlated to the rescued recombinant virus assay (Fig. 6). However, the cell–cell fusion assay held some limitations in assessing the fusion properties of some low-expression-level mutants that may affect the extent of fusion. This observation also supports the notion that virus-cell and cell–cell fusion have significant differences. In combination, pseudotyped HA-HIV-luc titer assays were able to accurately reflect the normal virus infection process. This assay is more convenient, safer and more sensitive than the standard cell– cell fusion assays or traditional virus-cell fusion

256

assays with live virus, particularly for studying infectious influenza virus – the highly pathogenic H5 influenza virus. In addition, our observations on the E11G, E11S, E11V and G8E mutants of H5 HA provide new insights into the mechanism of fusion activity.

Y. Su et al.

9.

10.

Acknowledgements 11. We thank Dr. Changmei Liu and Huaiyi Yang for many helpful discussions and Fulian Liao and Xiaoxuan Qi for their expert technical assistance. This work was supported by Ministry of Science and Technology of the People’s Republic of China (Grant Nos. 2005CB523001 and 2004BA519A12). Dr. Min Wu, Department of Biochemistry and Molecular Biology, University of North Dakota, a visiting scholar of IM, CAS supported by the K. C. Wong Education Foundation, Hong Kong.

12.

13.

References 1. Bullough PA, Hughson FM, Skehel JJ, Wiley DC (1994) Structure of influenza haemagglutinin at the pH of membrane fusion. Nature 371: 37–43 2. Carr CM, Kim PS (1993) A spring-loaded mechanism for the conformational change of influenza hemagglutinin. Cell 73: 823–832 3. Chang DK, Cheng SF, Deo Trivedi V, Yang SH (2000) The amino-terminal region of the fusion peptide of influenza virus hemagglutinin HA2 inserts into sodium dodecyl sulfate micelle with residues 16–18 at the aqueous boundary at acidic pH. Oligomerization and the conformational flexibility. J Biol Chem 275: 19150–19158 4. Chen J, Lee KH, Steinhauer DA, Stevens DJ, Skehel JJ, Wiley DC (1998) Structure of the hemagglutinin precursor cleavage site, a determinant of influenza pathogenicity and the origin of the labile conformation. Cell 95: 409–417 5. Chen J, Skehel JJ, Wiley DC (1999) N- and C-terminal residues combine in the fusion-pH influenza hemagglutinin HA(2) subunit to form an N cap that terminates the triple-stranded coiled coil. Proc Natl Acad Sci U S A 96: 8967–8972 6. Choi HS, Huh J, Jo WH (2006) Electrostatic energy calculation on the pH-induced conformational change of influenza virus hemagglutinin. Biophys J 91: 55–60 7. Cross KJ, Burleigh LM, Steinhauer DA (2001) Mechanisms of cell entry by influenza virus. Expert Rev Mol Med 2001: 1–18 8. Cross KJ, Wharton SA, Skehel JJ, Wiley DC, Steinhauer DA (2001) Studies on influenza haemagglutinin fusion

14.

15.

16.

17.

18.

19.

20.

peptide mutants generated by reverse genetics. Embo J 20: 4432–4442 Daniels RS, Downie JC, Hay AJ, Knossow M, Skehel JJ, Wang ML, Wiley DC (1985) Fusion mutants of the influenza virus hemagglutinin glycoprotein. Cell 40: 431–439 Doms RW, Gething MJ, Henneberry J, White J, Helenius A (1986) Variant influenza virus hemagglutinin that induces fusion at elevated pH. J Virol 57: 603–613 Gething MJ, Doms RW, York D, White J (1986) Studies on the mechanism of membrane fusion: site-specific mutagenesis of the hemagglutinin of influenza virus. J Cell Biol 102: 11–23 Gruenke JA, Armstrong RT, Newcomb WW, Brown JC, White JM (2002) New insights into the spring-loaded conformational change of influenza virus hemagglutinin. J Virol 76: 4456–4466 Han X, Steinhauer DA, Wharton SA, Tamm LK (1999) Interaction of mutant influenza virus hemagglutinin fusion peptides with lipid bilayers: probing the role of hydrophobic residue size in the central region of the fusion peptide. Biochemistry 38: 15052–15059 Han X, Bushweller JH, Cafiso DS, Tamm LK (2001) Membrane structure and fusion-triggering conformational change of the fusion domain from influenza hemagglutinin. Nat Struct Biol 8: 715–720 Huang Q, Chen CL, Herrmann A (2004) Bilayer conformation of fusion peptide of influenza virus hemagglutinin: a molecular dynamics simulation study. Biophys J 87: 14–22 Kantchev EA, Cheng SF, Wu CW, Huang HJ, Chang DK (2004) Secondary structure, phospholipid membrane interactions, and fusion activity of two glutamate-rich analogs of influenza hemagglutinin fusion peptide. Arch Biochem Biophys 425: 173–183 Korte T, Epand RF, Epand RM, Blumenthal R (2001) Role of the Glu residues of the influenza hemagglutinin fusion peptide in the pH dependence of fusion activity. Virology 289: 353–361 Lai AL, Park H, White JM, Tamm LK (2006) Fusion peptide of influenza hemagglutinin requires a fixed angle boomerang structure for activity. J Biol Chem 281: 5760–5770 Li Y, Han X, Lai AL, Bushweller JH, Cafiso DS, Tamm LK (2005) Membrane structures of the hemifusioninducing fusion peptide mutant G1S and the fusionblocking mutant G1V of influenza virus hemagglutinin suggest a mechanism for pore opening in membrane fusion. J Virol 79: 12065–12076 Neumann G, Watanabe T, Ito H, Watanabe S, Goto H, Gao P, Hughes M, Perez DR, Donis R, Hoffmann E, Hobom G, Kawaoka Y (1999) Generation of influenza A viruses entirely from cloned cDNAs. Proc Natl Acad Sci U S A 96: 9345–9350

Effect of mutations in influenza virus HA fusion peptide 21. Qiao H, Armstrong RT, Melikyan GB, Cohen FS, White JM (1999) A specific point mutant at position 1 of the influenza hemagglutinin fusion peptide displays a hemifusion phenotype. Mol Biol Cell 10: 2759–2769 22. Russ WP, Engelman DM (2000) The GxxxG motif: a framework for transmembrane helix–helix association. J Mol Biol 296: 911–919 23. Schoch C, Blumenthal R (1993) Role of the fusion peptide sequence in initial stages of influenza hemagglutinin-induced cell fusion. J Biol Chem 268: 9267–9274 24. Skehel JJ, Cross K, Steinhauer D, Wiley DC (2001) Influenza fusion peptides. Biochem Soc Trans 29: 623–626 25. Steinhauer DA, Wharton SA, Skehel JJ, Wiley DC (1995) Studies of the membrane fusion activities of fusion peptide mutants of influenza virus hemagglutinin. J Virol 69: 6643–6651 26. Vaccaro L, Cross KJ, Kleinjung J, Straus SK, Thomas DJ, Wharton SA, Skehel JJ, Fraternali F (2005) Plas-

257

27.

28.

29.

30.

ticity of influenza haemagglutinin fusion peptides and their interaction with lipid bilayers. Biophys J 88: 25–36 Walker JA, Kawaoka Y (1993) Importance of conserved amino acids at the cleavage site of the haemagglutinin of a virulent avian influenza A virus. J Gen Virol 74 (Pt 2): 311–314 Wharton SA, Martin SR, Ruigrok RW, Skehel JJ, Wiley DC (1988) Membrane fusion by peptide analogues of influenza virus haemagglutinin. J Gen Virol 69 (Pt 8): 1847–1857 Wharton SA, Calder LJ, Ruigrok RW, Skehel JJ, Steinhauer DA, Wiley DC (1995) Electron microscopy of antibody complexes of influenza virus haemagglutinin in the fusion pH conformation. EMBO J 14: 240–246 Wilson IA, Skehel JJ, Wiley DC (1981) Structure of the haemagglutinin membrane glycoprotein of influenza virus at 3 A resolution. Nature 289: 366–373