Journal of General Virology (2014), 95, 2618–2626

DOI 10.1099/vir.0.067926-0

Polymorphisms in the haemagglutinin gene influenced the viral shedding of pandemic 2009 influenza virus in swine Alessio Lorusso,13 Janice R. Ciacci-Zanella,1,2 Eraldo L. Zanella,1,3 Lindomar Pena,4,54 Daniel R. Perez,4,5 Kelly M. Lager,1 Daniela S. Raja˜o,1 Crystal L. Loving,1 Pravina Kitikoon1 and Amy L. Vincent1 Correspondence

1

Amy L. Vincent

2

[email protected]

Virus and Prion Research Unit, National Animal Disease Center, USDA-ARS, Ames, IA, USA Laborato´rio de Virologia, Embrapa Suı´nos e Aves, Conco´rdia, Santa Catarina, Brazil

3

Universidade de Passo Fundo, Brazil

4

Department of Veterinary Medicine, University of Maryland, College Park, MD, USA

5

Virginia–Maryland Regional College of Veterinary Medicine, College Park, MD, USA

Received 13 May 2014 Accepted 13 August 2014

Interactions between the viral surface glycoprotein haemagglutinin (HA) and the corresponding receptors on host cells is one important aspect of influenza virus infection. Mutations in HA have been described to affect pathogenicity, antigenicity and the transmission of influenza viruses. Here, we detected polymorphisms present in HA genes of two pandemic 2009 H1N1 (H1N1pdm09) isolates, A/California/04/2009 (Ca/09) and A/Mexico/4108/2009 (Mx/09), that resulted in amino acid changes at positions 186 (S to P) and 194 (L to I) of the mature HA1 protein. Although not reported in the published H1N1pdm09 consensus sequence, the P186 genotype was more readily detected in primary infected and contact-naı¨ve pigs when inoculated with a heterogeneous mixed stock of Ca/09. Using reverse genetics, we engineered Ca/09 and Mx/09 genomes by introducing Ca/09 HA with two naturally occurring variants expressing S186/ I194 (HA-S/I) and P186/L194 (HA-P/L), respectively. The Ca/09 HA with the combination of P186/L194 with either the Ca/09 or Mx/09 backbone resulted in higher and prolonged viral shedding in naı¨ve pigs. This efficiency appeared to be more likely through an advantage in cell surface attachment rather than replication efficiency. Although these mutations occurred within the receptor-binding pocket and the Sb antigenic site, they did not affect serological crossreactivity. Relative increases of P186 in publicly available sequences from swine H1N1pdm09 viruses supported the experimental data, indicating this amino acid substitution conferred an advantage in swine.

INTRODUCTION The first influenza pandemic of the new millennium was caused by a novel H1N1 influenza virus initially identified in humans in April 2009 (Garten et al., 2009; Novel SwineOrigin Influenza A (H1N1) Virus Investigation Team, 2009). The pandemic 2009 H1N1 (H1N1pdm09) infection was characterized by high transmissibility in humans and, in general, by mild clinical symptoms. The end of the pandemic was declared in August 2010. Currently, the H1N1pdm09 virus circulates worldwide in humans (http:// 3Present address: Istituto Zooprofilattico Sperimentale dell’Abruzzo e del Molise ‘G. Caporale’, Teramo, Italy. 4Present address: Federal University of Paraı´ba, Joa˜o Pessoa, Brazil. Supplementary material is available with the online version of this paper.

2618

www.who.int/csr/don/2010_04_16/en/), and a representative H1N1pdm09 strain (A/California/7/2009) was chosen for the trivalent seasonal flu vaccine for 2010, 2011, 2012 and 2013 (CDC, 2010; http://www.cdc.gov/flu/about/season/ vaccine-selection.htm). H1N1pdm09 appeared with a gene segment combination that was not known to circulate in humans or animals prior to the time of emergence (Gibbs et al., 2009; Novel Swine-Origin Influenza A (H1N1) Virus Investigation Team, 2009). Six of the gene segments, including the segment encoding the haemagglutinin (HA) protein, were closest in sequence to those of the ‘triple reassortant internal gene influenza viruses that have been isolated from North American pigs since 1997–1998 (Lorusso et al., 2013). The remaining two segments (matrix and neuraminidase) bear the closest homology with those of the Eurasian avian-like swine viruses (Garten et al., 2009). 067926 G 2014 The Authors

Printed in Great Britain

HA mutations influence transmission of H1N1pdm09

The HA protein is a major determinant of virulence and host specificity as its binding site and binding pocket recognize sialic-acid-containing cell surface receptors on host epithelial cells (Ayora-Talavera et al., 2009; de Wit et al., 2010; Nicholls et al., 2008; Shinya et al., 2006). The HA protein of H1N1pdm09 belongs to the North American classical swine lineage – a descendent from the 1918 pandemic influenza virus (Garten et al., 2009; Lorusso et al., 2011, 2013). A previous report showed a clear distinction between the receptor-binding repertoire of A/ California/04/2009 (Ca/09) and a seasonal H1N1 virus (Childs et al., 2009). Furthermore, Ca/09 was shown to bind a considerable range of a2–3-linked sialyl probes in addition to the majority of a2–6-linked sialyl probes and binding was stronger than another H1N1pdm09 used in that study. The authors speculated that Ca/09 HA, with several amino acid residues that were not present in the pH1N1 consensus sequence, may have played a role in the difference in binding. By investigating the transmission properties of two early human isolates of H1N1pdm09 in pigs, including Ca/09, we detected a mixed population of viruses shed by Ca/09-infected pigs, with polymorphisms in the HA gene that were also present in the challenge inoculum. We demonstrated subsequently that these polymorphisms contributed to transmission in pigs.

RESULTS Nasal shedding and transmission profiles differ between Ca/09wt and Mx/09wt in pigs Pigs were infected with WT Ca/09 (Ca/09wt) and A/Mexico/ 4108/2009 (Mx/09wt), and a reverse-genetics-engineered Mx/09 (rgMx/09). In Studies 1–3, Ca/09wt demonstrated nasal shedding at higher titres in infected and contact pigs, as well as increased transmission to contact pigs compared with Mx/09wt (Fig. 1). In both studies, primary pigs infected with Ca/09wt had titres significantly higher than Mx/09wt at 3 and 5 days post infection (p.i.) (P,0.05), with a peak titre

4 3 2 1

Pigs infected with Mx/09wt reached peak titres at 5 days p.i. and became negative by 9 days p.i. (Fig. 1a). Mx/09wt direct-contact pigs shed at lower titres and were statistically different from those of Ca/09 in each respective study at all time points, except at 9 and 11 days post-contact (P,0.05). Direct-contact pigs became negative by 11 days post-contact (Fig. 1b). Respiratory-contact pigs remained negative for viral shedding for the entire experiment in Study 1 (Fig. 1c). However, exposure to virus occurred, as revealed by ELISA that detected specific nucleoprotein (NP) antibodies in all serum samples collected at 19 days post-contact (Table S1, available in the online Supplementary Material), and shedding occurred in the respiratory contacts of Study 3, although this was consistently lower than that detected in Ca/09wt groups (Fig. 1c). In order to compare the reverse-genetics-derived viruses with WT virus, we inoculated pigs by the same route of infection with rgMx/09 (Fig. 1). The pattern of shedding for rgMx/09 in primary infected pigs was similar to that observed for Mx/09wt (Fig. 1a). Primary pigs infected with rgMx/09 yielded very low titres at 3 days p.i. that were significantly lower than Ca/09wt (P,0.05). Primary rgMx/ 09-infected pigs reached peak nasal swab titres at 5 days p.i. with no significant difference compared with Ca/09 and Mx/09wt, but were negative at 7 days p.i. However, viral shedding remained decreased as compared with Ca/09wt. Direct-contact pigs had detectable virus in nasal swabs at 3 days post-contact, reaching the peak titre at 5 days postcontact with no significant differences to Ca/09wt directcontact pigs at either time points (Fig. 1b). However, the direct-contact pigs infected with rgMx/09 were negative by 7 days post-contact (P,0.05). Nasal swabs from

(c)

5

5 Virus shed (log10 TCID50 ml–1)

(b)

5

Virus shed (log10 TCID50 ml–1)

Virus shed (log10 TCID50 ml–1)

(a)

at 5 days p.i. and becoming negative by 9 days p.i. (Fig. 1a). Ca/09wt direct-contact pigs (Fig. 1b) started shedding at 3 days post-contact and peaked at 5 days post-contact, whereas respiratory-contact pigs typically began shedding at 5 days post-contact and peaked at 5 or 7 days post-contact (Fig. 1c). Direct-contact pigs in the Ca/09 group were negative for shedding at 11 days post-contact.

4 3 2 1 0

0 3

7

5

Time (day p.i.)

9

11

4 3 2 1 0

3

5

7 9 Time (day p.i.)

11

3

5

7 9 Time (day p.i.)

11

Fig. 1. Virus shedding (mean±SD) in nasal swab filtrates over time of primary infected (a), direct-contact (b) and respiratorycontact (c) pigs infected with Ca/09wt (Study 1, $; Study 3, .), Mx/09wt (Study 1, h; Study 3, e) and rgMx/09 (Study 2, D). Transmission was monitored by titrating the amount of virus in the nasal swabs of pigs. http://vir.sgmjournals.org

2619

A. Lorusso and others

respiratory-contact pigs infected with rgMx/09 yielded very low titres of virus at 3 days post-contact with the peak at 7 days post-contact and were negative by 9 days postcontact. Titres at 3 days post-contact were significantly lower (P,0.05) than those observed in the Ca/09wt respiratory-contact group, whereas no statistically significant differences were found at the remaining time points. Genotypes of Ca/09 HA impacted nasal shedding and transmission profiles As binding and affinity of HA with its receptor is a critical factor for transmission, we investigated if Ca/09wt HA played a role in transmission by introducing Ca/09 HA into Mx/09 by reverse genetics in comparison with parental Ca/ 09 and Mx/09 strains. During the process of cloning the Ca/ 09 HA gene, a double population of H1N1pdm09 viruses involving two residues in the HA gene at positions 186 and 194 were detected. The combinations isolated in our study were S186/I194 and P186/L194. Therefore, we introduced Ca HA-S/I and Ca HA-P/L into the Ca/09 and Mx/09 backbones, obtaining Ca HA-P/L, Ca HA-S/I, Mx/09 : Ca HA-P/L and Mx/09 : Ca HA-S/I, and then infected the pigs. When we introduced the Ca/09 HA variants onto the Mx/09 backbone, the Mx/09 : Ca HA-P/L tended to shed at higher titres than Mx/09 : Ca HA-S/I in infected pigs, although not significantly different from each other (Fig. 2a). The dynamics of shedding for these two mutants were similar in the primary infected pigs (Fig. 2a), whereas direct-contact pig titres had higher titres for a longer duration for Mx/ 09 : Ca HA-P/L (Fig. 2b). For Mx/09 : Ca HA-P/L, nasal swabs from direct-contact pigs were positive up to 9 days post-contact. Respiratory-contact pigs infected with Mx/ 09 : Ca HA-P/L shed at higher viral titres at 5, 7 and 9 days post-contact (P,0.05) (Fig. 2c). The same trends were observed for Ca/09 : Ca HA-P/L in the respiratory-contact group, with the highest titres at 7 days post-contact, and statistically significant differences in titres compared with the Ca/09 : Ca HA-S/I group at 9 and 11 days post-contact (P,0.05) (Fig. 2c).

After its initial use in Study 1 and subsequent use in Study 3, the Ca/09wt stock was shown to be a mixed population with substitutions in the HA gene (Table 1). Unfortunately, it was not possible to analyse the original challenge stocks of Mx/09wt and Ca/09wt used in Study 1 as the heterogeneous nature of the virus stock was realized only after the initial study and further subsequent passage of WT virus stocks. However, as a double population was recovered from the primary infected pigs in Study 1, we assumed that a mixed HA population was likely present in the inoculum stock. HA sequences demonstrated a relative increase of subpopulations of WT virus after in vivo replication in pigs (Table 1). In Study 1, Ca/09-infected pigs shed virus from nasal secretions; three of eight clones contained P186, five of eight clones contained S186 and eight of eight clones contained L194. However, in nasal swabs from the directcontact pigs, seven of eight clones contained P186 and one of eight clones contained S186; one of eight clones contained I194 and seven of eight clones contained L194. For the nasal secreted viruses from respiratory-contact pigs, eight of eight clones contained P186 and eight of eight clones contained L194. For the virus shed from pigs inoculated with Mx/09wt in Study 1, one of eight clones contained P186, seven of eight clones contained S186 and eight of eight clones contained L194. In the direct-contact pigs, one of eight clones contained P186 and seven of eight clones contained S186, whereas six of eight clones contained L194 and two of eight clones contained I194. No RNA was isolated from respiratory-contact pigs infected with Mx/ 09wt as all were negative, as shown by virus isolation. The virus passages used for the inocula of Ca/09wt and Mx/ 09wt for Study 3 were found to contain the mixed virus genotypes (Table 1). The Ca/09wt stock contained three of six plaques with P186 and three of six plaques with S186. I194 was found in one of six plaques in combination with S186, whereas the remaining clones contained L194. P186 was found in combination only with L194. The S186 and

4 3 2 1 0

(c) 5 Virus shed (log10 TCID50 ml–1)

(b) 5 Virus shed (log10 TCID50 ml–1)

Virus shed (log10 TCID50 ml–1)

(a) 5

P186 was more frequently detected after in vivo pig infection

4 3 2 1 0

3

5

7 Time (day p.i.)

9

11

4 3 2 1 0

3

5

7 Time (day p.i.)

9

11

3

5

7

9

11

Time (day p.i.)

Fig. 2. Virus shedding (mean±SD) in nasal swab filtrates over time of primary infected (a), direct-contact (b) and respiratorycontact (c) pigs infected with Ca/09 : Ca HA-P/L ($), Ca/09 : Ca HA-S/I (#), Mx/09 : Ca HA-P/L (&) and Mx/09 : Ca HA-S/I (h). Transmission was monitored by titrating the amount of virus in the nasal swabs of pigs. 2620

Journal of General Virology 95

HA mutations influence transmission of H1N1pdm09

Table 1. HA sequence analysis from viral RNA extracted from nasal swab filtrates of infected or contact pigs of Study 1 and from virus stocks and nasal swab filtrates of infected or contact pigs of Study 3 and 4 Virus Study 1 Ca/09wt Infected Direct Respiratory Mx/09wt Infected Direct Respiratory Study 3 and 4 Ca/09wt* inoculum Infected Direct Respiratory Mx/09wt* inoculum Infected Direct Respiratory rgMx/09 HA-S/I inoculum Infected Direct Respiratory rgMx/09 HA-P/L inoculum Infected Direct Respiratory rgCa/09 HA-S/I inoculum Infected Direct Respiratory rgCa/09 HA P/L inoculum Infected Direct Respiratory

HA 186

HA 194

P: 3/8; S: 5/8 P: 7/8; S: 1/8 P: 8/8

L: 8/8 L: 7/8; I: 1/8 L: 8/8

P: 1/8; S: 7/8 P: 1/8; S: 7/8

L: 8/8 L: 6/8; I: 2/8

ND

ND

P: 3/6; S: 3/6 P: 8/8 P: 8/8 P: 8/8 P: 1/6; S: 5/6 P: 3/8; S: 5/8 P: 2/8; S: 6/8 P: 3/8; S: 5/8 S: 6/6

L: 5/6; I: 1/6 L: 8/8 L: 8/8 L: 8/8 L: 6/6 L: 8/8 L: 8/8 L: 8/8 I: 6/6

S: 8/8 S: 8/8 S: 8/8 P: 6/6

I: 8/8 I: 8/8 I: 8/8 L: 6/6

P: 8/8 P: 8/8 P: 8/8 S: 6/6

L: 8/8 L: 8/8 L: 8/8 I: 6/6

S: 8/8 S: 8/8 S: 8/8 P: 6/6

I: 8/8 I: 8/8 I: 8/8 L: 6/6

P: 7/7 P: 8/8 P: 8/8

L: 7/7 L: 8/8 L: 8/8

ND,

None detected *RNA was extracted from isolated plaques of Ca/09wt and Mx/09wt.

I194 residues were found naturally occurring and rescued together in the clone containing Ca HA-S/I, whereas P186 and L194 were rescued in the clone containing Ca HA-P/L. The S186 residues were more commonly found with L194. S186 and L194, reflecting the H1N1pdm09 consensus sequence from sequences available in GenBank. All clones from the Ca/09wt primary infected pigs contained P186 (eight of eight) and L194 (eight of eight). The identical amino acid pairing was found in the direct-contact and respiratory-contact pigs, suggesting that P186 and L194 http://vir.sgmjournals.org

were selected in the swine host when challenged with a heterogeneous virus stock. A similar finding was detected for the Mx/09 stock, but with a lower ratio of P186 to S186. P186 was present in only one of six plaques, whereas the remaining five plaques contained S186. No plaques were shown to possess I194 and all contained L194 in the Mx/ 09wt stock used in Study 3. Virus shed from primary infected pigs in the Mx/09wt group demonstrated P186 in three of eight clones and S186 in the remaining clones, whereas virus from the direct-contact pigs showed two of eight clones with P186 and six of eight clones with S186. Virus from respiratory-contact pigs infected with Mx/09wt showed three of eight clones with P186 and five of eight clones with S186. Virus from the infected, direct-contact and respiratory-contact pigs invariably contained L194. Analysis of reverse-genetics-engineered stock viruses used for challenge in Studies 3 and 4 showed that they invariably contained S186 and I194 for Mx/09 : Ca HA-S/I and Ca/ 09 : Ca HA-S/I, and P186 and L194 for Mx/09 : Ca HA-P/L and Ca/09 : Ca HA-P/L (Table 1). The same residues were found in all clones from viruses isolated from nasal secretions from the infected, direct-contact and respiratory-contact pigs of each virus group, thus suggesting that S186P was not an adaptive change in the swine host in this study, but it appeared that selection in favour of P186 occurred when pigs were challenged with the heterogeneous WT mixed stock. P186/L194 conferred an advantage in cell surface attachment In order to establish whether the two amino acid combinations play a role in cell attachment in vitro, Madin–Darby canine kidney (MDCK) cells were infected with virus clones Ca/09 : Ca HA-S/I or Ca/09 : Ca HA-P/L and virus at the cell surface (attachment) versus intracellular virus (infectivity) was evaluated. After 1 h of incubation, the Ca HA-P/L virus produced a significantly higher percentage of positively stained MDCK cells. Staining was primarily due to cellsurface-attached virus versus intracellular virus (Fig. 3a) as permeabilization did not significantly increase the percentage of positive cells. Likewise, the Ca HA-P/L-infected MDCK cells demonstrated significantly higher percentages of positively stained cells at 24 and 48 h p.i. (Fig. 3b) when compared with MDCK cells infected with Ca HA-S/I. Although there was a trend for the Ca HA-P/L virus to show higher virus titres in the supernatant, no time points demonstrated a statistically significant difference (Fig. 3c), indicating again that the increased percentages of positively stained cells were likely due to an attachment advantage rather than an advantage in viral replication, which was consistent with the growth curves (Fig. S1). Distribution of P186 The prevalence of P186 in deduced HA protein sequences was low in human isolates of H1N1pdm09 over all of the years analysed. Only 0.4 % of human H1N1pdm09 isolated 2621

A. Lorusso and others

(b) *

20

10

20

10

HA-S/I HA-P/L

105 104 103 102 101 100

0 8

pe

16

24 36 Time (h p.i.)

48

8

16

24 36 Time (h p.i.)

48

A-

H

A-

P/ L

no P/ L

106

30

rm

rm pe

pe S/ I

AH

HA-S/I HA-P/L

(c)

*

H

H

A-

S/ I

no

pe

rm

rm

0

* 40

Virus titre (log10 TCID50)

30

Influenza-positive MDCK cells (%)

Influenza-positive MDCK cells (%)

(a)

Fig. 3. Attachment and infectivity assays as measured by FACS on a flow cytometer. (a) MDCK cells were infected at m.o.i. 0.1 and collected at 1 h p.i. to assess the percentage of cells stained for influenza. Cells were either stained without permeabilization (cell surface virus; ‘no perm’) or after permeabilization (cell surface and intracellular virus; ‘perm’). (b) MDCK cells were then infected at m.o.i. 0.001, and collected at 8, 16, 24, 36 and 48 h p.i. for evaluation of percentage of cells staining positive for influenza. (c) Supernatants were collected at each time point and titrated for quantity of virus.

from 2009 contained P186, whereas nearly all of the remainder contained S186. The number of human H1N1pdm09 isolates found with P186 appeared to show a slight increase, with 6 % in 2010 and 2011, but 1 % in 2012–14 (Table 2). However, in pigs, P186 had a more apparent increase, beginning with 1 % of H1N1pdm09 in 2009, but increasing to 6 % in 2010, 16 % in 2011 and 57 % in 2012. The frequency of P186 in swine was only 4 % in 2013–14, but this was based on only 27 available sequences. The lower number of sequences available for analysis may have biased the frequencies detected in 2012– 2014 compared with 2009–2011, especially for swine viruses.

The S186P mutation found in H1N1pdm09 was found in other H1 lineages in addition to H1N1pdm09 (data not shown). From 1930 to 1998, 100 % of H1 from the North American swine lineage had P186. From 1933 to 1997, only 6 % of the human H1 isolates sequenced contained P186 until gaining prevalence from 1998 to 2009, with all the human 2009 seasonal H1 having P186 (Ye et al., 2010). I194 was extremely rare in the context of H1 viruses of swine and human origin, and the position was more consistently occupied by an L residue in the sequences included in the analysis.

DISCUSSION Table 2. Frequency of P/S at position 186 and I/L at position 194 in publicly available H1N1pdm09 sequences in viruses isolated from human and swine in 2009–2014 Amino acid numbering is based on H3 as the reference. Year

2009 2010 2011 2012 2013–2014

2622

Human

Swine

HA 186 (%)

HA 194 (%)

HA 186 (%)

HA 194 (%)

P: 0.4 S: 99.5 P: 6 S: 93 P: 6 S: 94 P: 1 S: 96 P: 1 S: 98

I: 0.8 L: 99 I: 0 L: 100 I: 0.5 L: 99.5 I: 0.4 L: 99.4 I: 0 L: 100

P: 1 S: 99 P: 6 S: 92 P: 16 S: 84 P: 57 S: 43 P: 6 S: 94

I: 0.5 L: 99.5 I: 0.6 L: 99.4 I: 1 L: 99 I: 3 L: 97 I: 0 L: 100

Our results confirm that the prototypic WT H1N1pdm09 human isolates Ca/09wt and Mx/09wt had similar pathogenicity in pigs, were transmissible among pigs and exhibited serological cross-reactivity to each other (Table S2). However, Ca/09wt demonstrated nasal shedding at higher titres and increased transmission to contact pigs compared with Mx/09wt. This difference was dramatically evident in our first study where, unexpectedly, nasal swabs from the respiratory-contact pigs in the Mx/09wt group remained negative for the entire experiment, but all of the pigs sero-converted (Table S1). To confirm our initial findings, the in vivo study with the two WT viruses (Ca/ 09wt and Mx/09wt) was repeated and compared with a homogeneous culture of reverse-genetics-engineered rescued Mx/09 (rgMx/09). When given to pigs, rgMx/09 showed a viral shedding profile that increased compared with Mx/09wt, but was still consistently less than Ca/09wt. Ca/09wt and Mx/09wt differed in either three or four residues of the HA1 protein (Table 3). During the process Journal of General Virology 95

HA mutations influence transmission of H1N1pdm09

Table 3. Amino acid differences between published Ca/09 and Mx/09 genomes Gene position PA 578 HA 186 HA 194 HA 200 HA 324 HA 430 NP 101

Ca/09

Mx/09

M P I* T I V D

L S L A V T G

*Found only in the deposited Ca/09 HA sequence (GenBank accession number GQ280797); H3 numbering.

of cloning of the Ca/09 HA gene, we detected a double population of H1N1pdm09 viruses involving two residues in the HA gene at positions 186 and 194. The combinations isolated in our study were S186/I194 and P186/L194. The combination found in the current H1N1pdm09 consensus, S186/L194, was not identified at the time; however, the S186/I194 combination (Ca HA-S/I) was reported in the earliest publicly available Ca/09 sequence (GenBank accession number GQ280797). The alternative form, P186/L194 (Ca HA-P/L), differed from the H1N1pdm09 consensus sequence in position 186, as the vast majority of H1N1pdm09 isolates contained S186. However, Mx/09 HA matched the H1N1pdm09 consensus sequence in having S186 and L194. Likewise, our rgMx/09 clone contained S186 and L194. In the Ca/09wt and Mx/09wt heterogeneous stocks, subpopulations of virions carrying amino acid variations at position 186 and 194 were present; however, P186 and L194 were more frequently detected in pigs rather than S186 and I194. When compared with viruses carrying S186/I194, the P186/L194 combination in the context of the Ca/09 HA gene was shown to confer higher rates of replication in the upper respiratory tracts of infected pigs. Respiratory-contact pigs significantly highlight this difference, shedding virus at higher titres and for a prolonged duration. In pigs challenged directly with reverse-genetics-engineered viruses with either Ca HA-S/I or HA-P/L, similar trends in nasal shedding with both H1N1pdm09 backbones were detected, but with slightly higher titres reached by the viruses with Ca HAP/L. The most notable differences were observed in the shedding of virus from pigs in the respiratory-contact groups. For the reverse-genetics-engineered viruses carrying the HA-P/L gene, the shedding was of a higher magnitude as well as prolonged in duration, thus suggesting that the mutant HA carrying P186/L194 was more efficient in nasal shedding compared with S186/I194. This trend was consistent on both the Ca/09 and the Mx/09 backbones, although the magnitude and the duration over which statistical differences were detected between the HA-S/I and HA-P/L clones varied between the two backbones. The mechanism that allowed Ca/09wt and the clones carrying Ca HA-P186 and/or L194 to replicate to higher http://vir.sgmjournals.org

titres than Mx/09wt and Ca HA-S186 and/or I194-carrying clones in the upper respiratory tract of pigs in vivo (Figs 1 and 2) and in MDCK cells in vitro (Fig. S1) was suggested to be due to a greater efficiency of attachment during the initial infection and after waves of replication (Fig. 3). As the shedding profile of rgMx/09 (with S186/L194) closely resembled that which we observed with Mx/09 : Ca HA-S/I, we speculate that the conservative amino acid change L194I did not affect the replication and transmissibility of Mx/ 09 : Ca HA-S/I. The role of P186 in increased shedding of H1N1pdm09 in pigs was supported additionally by the phenotype of Ca/09 : Ca HA-S/I with a pattern more similar to rgMx/09 than the phenotype of Ca/09 : Ca HA-P/ L with a pattern more similar to Ca/09wt. This suggests that the single amino acid change S186P in Ca/09 HA was responsible for the increased nasal detection and transmission. Although Mx/09wt in Study 3 showed increased transmission when compared with Mx/09wt in Study 1 and rgMx/ 09, the magnitude of shedding was consistently reduced in comparison with Ca/09 in the same study cohorts. Subsequently, we confirmed that a mixed population of HA variants was present in the Mx/09wt challenge stock, although seemingly with a lower proportion of P186 as compared with the Ca/09wt challenge stock. The Ca/09wt stock used for the challenge in Study 3 contained an equal distribution of plaques with S186 and P186. However, when the viral stock was given to pigs, viruses containing HA with P186 appeared to be selected rapidly, and P186 was invariably detected in the nasal swabs from directcontact and indirect respiratory-contact pigs. A similar scenario can be assumed for the viral stocks used in Study 1 as both S186 and P186 were present in the primary challenged pigs, but again only P186 was identified as being shed from the nasal secretions from the contact pigs. This suggests that P186 has a selective advantage in terms of infection in pigs in the context of Ca/09 HA. The number of plaques with P186 was found to be lower with respect to S186 in the challenge stocks for Mx/09 for Study 3, but P186 was nonetheless present upon retrospective analysis. The number of P186 plaque clones remained lower than S186 in direct-contact and respiratory-contact pigs in Studies 1 and 3. This likely contributed to the variation in the shedding profile we observed in the Mx/09wt groups in Studies 1 and 3, and rgMx/09. We speculate that a differential representation of the population of viruses carrying P186 in the two replicates with Mx/09wt challenge stocks affected the nasal shedding dynamics. However it cannot be ruled out that the variation observed during Mx/09 infection was also the result of natural physiological differences that can be observed during in vivo studies with outbred pigs. The selective advantage of P186 in direct-contact and respiratory-contact pigs infected with the mixed populations found naturally in both Ca/09wt and Mx/09wt virus stocks strongly supports the important role of this amino acid moiety in the context of not only Ca/09 HA, but also Mx/ 2623

A. Lorusso and others

09 HA. However, we cannot rule out that the L194I mutation may have played a role in shedding with the P186S mutation by limiting the shedding of viruses with Ca HA-S/I specificity. Interestingly, it was demonstrated that a rgCa/09 carrying P186/L194 grew at a higher magnitude in MDCK than rgCa/09 viruses with S186/L194 or S186/I194 (Suphaphiphat et al., 2010), supporting our hypothesis that L194 played a minor role in the increased shedding pattern of the clones with Ca HA-P/L. P186 alone was found to confer higher replication rates to mice-adapted H1N1pdm09 in cell culture systems, and to be responsible for the high pathogenicity of mice-adapted H1N1pdm09 in mice and ferrets (Ilyushina et al., 2010; Ye et al., 2010). Although we did not specifically investigate the pathogenicity of each cloned recombinant virus, we had no evidence that P186 impacted virulence in pig; pulmonary lesions of primary infected pigs inoculated with similar titres of challenge stocks of Ca/09wt with a higher proportion of virions with P186 compared with Mx/09wt and Ca/09 : Ca HA-S/I were not different with respect to each other in terms of percentage of lung affected with pneumonia or the microscopic character of the lung lesions (data not shown). Residues at position 186 and 194 occur within the receptorbinding pocket (Yang et al., 2010) and the Sb antigenic site, and thus are likely to play a role in the interaction of the HA with the receptor. However, antigenicity in the haemagglutination inhibition assay was not affected by mutations at these positions in our studies (Table S2). S186 may be a human adaptation, whereas P186 is a swine adaptation in the context of the classical North American swine lineage H1, as has been suggested previously (Matrosovich et al., 1997; Meroz et al., 2011), and is supported by the engineered viruses at position 186 in our swine studies. This predilection for P186 in swine was also supported by the deduced amino acid sequence analysis from the publicly available 2009–2014 H1N1pdm09 sequences from swine and humans, and our report directs further interest in the S186P substitution. Although the number of H1N1pdm09 reported globally in swine declined drastically after 2011 and also in relative proportion to other lineages of H1 swine viruses in the United States (Anderson et al., 2013), the virus became established in most swine-producing regions and likely still circulates. The available sequences from swine in public databases are highly biased by the contributions from the USA, which may give a false indication that the H1N1pdm09 is declining in swine worldwide. However, the increased relative frequency of S186 in swine in 2013–14 may be an indication of new human-to-swine spillover events rather than continued circulation of endemic swine-adapted viruses. Further analyses are required to confirm this speculation. As H1N1pdm09 becomes established as a human seasonal strain, continued sharing between human and swine is highly likely. Our swine study highlights the importance of global networks and databases for sharing data from influenza surveillance and research to understand virus dynamics in both host species. To this end, it is critical to maintain the increased level of monitoring in swine seen after 2009. 2624

METHODS Viruses and cells. Ca/09wt and Mx/09wt were kindly provided by the Centers for Disease Control and Prevention (CDC), Atlanta, Georgia. These viruses were provided as passage 1 viruses in MDCK cells. Low-passage stocks were generated and maintained at 280 uC. To facilitate analysis of the HA gene in combination with the Ca/09 and Mx/09 backbones, stocks were prepared using viruses rescued by reverse genetics as indicated below. Viruses were titrated in MDCK cells to determine the TCID50 as described below. MDCK cells were maintained in modified Eagle’s medium (MEM) (Sigma-Aldrich) containing 5 % FBS (Sigma-Aldrich). 293-T human embryonic kidney cells were cultured in Opti-MEM I (Gibco) containing 5 % FBS. Amino acid differences between Ca/09 and Mx/09 viral proteins, as predicted by RNA sequence, are summarized in Table 3. In vivo Study 1. A total of 30 healthy pigs (4 weeks old) were inoculated intratracheally with 2 ml Ca/09 (n515 pigs) or Mx/09 (n515) at 105 TCID50 ml21. All pigs came from a herd free of swine influenza virus, porcine reproductive and respiratory syndrome virus, and Mycoplasma hyopneumoniae. The pigs were treated with ceftiofur crystalline free acid (Pfizer) to reduce bacterial contaminants preceding the start of the study. The two groups were housed in individual isolation rooms (Biosafety Level 3 Agriculture), and cared for in compliance with the Institutional Animal Care and Use Committee of the National Animal Disease Center. Five pigs remained unchallenged as negative controls. The pigs were anaesthetized by intramuscular injection of a cocktail of ketamine (8 mg kg21), xylazine (4 mg kg21) and Telazol (6 mg kg21; Fort Dodge Animal Health) followed by virus inoculation. Pigs were euthanized humanely with a lethal dose of pentobarbital (Sleepaway; Fort Dodge Animal Health) at the appropriate time during the course of the study. At 2 days p.i., five pigs were introduced in direct contact with the infected pigs. An additional five pigs were introduced as a respiratory-contact group, in an elevated deck ~2 m from the primary infected and direct-contact pigs, and separated by steel gating. Nasal swabs were taken and placed in 2 ml MEM on 0, 3, 5, 7, 9 and 11 days p.i. and post-contact to evaluate nasal virus shedding, and stored at 280 uC until the end of the study. Five inoculated pigs per group were euthanized on 3, 5 and 7 days p.i., and five control pigs were euthanized on 7 days p.i. Direct-contact and indirect-contact pigs were euthanized at 19 days post-contact. After euthanasia, each lung was lavaged with 50 ml MEM to obtain bronchoalveolar lavage fluid (BALF). Each nasal swab sample was subsequently thawed and vortexed for 15 s, centrifuged for 10 min at 640 g, and the supernatant passed through a 0.45 mm filter. Virus isolation and titration from nasal swabs and BALF were performed as described previously (Vincent et al., 2009). Generation of recombinant viruses. The eight gene segments of

Mx/09 (passage 2 in MDCK cells) were amplified by reverse transcription-PCR procedures with SuperScript III One-Step RT-PCR Platinum Taq HiFi (Invitrogen) and cloned in the bidirectional reverse genetics plasmid (Hoffmann et al., 2000; Ye et al., 2010). Similarly, the HA gene of Ca/09 (passage 2 MDCK) was amplified and cloned in two naturally occurring variants, HA-S/I and HA-P/L. HA-S/I contained S186 and I194 (H3 numbering) identical to the deposited sequence of Ca/09 (GenBank accession number GQ280797), whereas HA-P/L contained P186 and L194. The combination S186/L194 that corresponded to the H1N1pdm09 consensus sequence was not detected in our Ca/09 HA sequence at the time we started the experiments. Five recombinant viruses were rescued: rgMx/09, Mx/09 : Ca HA-S/I, Mx/ 09 : Ca HA-P/L, Ca/09 : Ca HA-S/I and Ca/09 : Ca HA-P/L. The reversegenetics-engineered viruses were recovered as described previously (Hoffmann et al., 2000). Rescued viruses were partially sequenced in each segment to confirm the desired gene constellation. Full-length HA gene segments were sequenced in order to avoid extraneous mutations and titrated in MDCK cells. Journal of General Virology 95

HA mutations influence transmission of H1N1pdm09

experimental design as described for Study 1, a total of 13 pigs were intratracheally inoculated with 2 ml rgMx/09 at 105 TCID50 ml21. At 2 days p.i., eight pigs were introduced in direct contact with the infected pigs and eight pigs were introduced in respiratory contact per each virus group. Nasal swabs were collected at the same time points as in Study 1 and processed as described for Study 1. Five inoculated pigs were euthanized at 5 days p.i. to evaluate lung pathology and BALF was processed as described above.

Influenza Research Database (Squires et al., 2012). The total numbers of sequences of H1N1pdm09 HA per year obtained from viruses isolated from humans were 2006 from 2009, 1636 from 2010, 1407 from 2011, 490 from 2012 and 918 from 2013–14, and these were subjected to the analysis. In addition, 188, 172, 108, 30 and 27 HA protein sequences of H1N1pdm09 HA isolated from swine from 2009, 2010 2011, 2012 and 2013–14, respectively, were analysed. Sequences from 2013 through May 2014 were combined due to fewer numbers of sequences being available, especially for swine.

In vivo Study 3. To evaluate the nasal shedding profile of the HA-S/I

Statistical analysis. ANOVA was utilized to assess virus titre means

and HA-P/L clones on the Mx/09 backbone, a total of 20 pigs were intratracheally inoculated with 2 ml of Ca/09wt (n55 pigs), Mx/09wt (n55), Mx/09 : Ca HA-S/I (n55) or Mx/09 : Ca HA-P/L (n55) at 26105 TCID50 ml21. At 2 days p.i., five pigs were introduced in direct contact with the infected pigs and five pigs were introduced in respiratory contact per each virus group. Nasal swabs were collected and processed as described above.

by treatment group and to generate subsequent graphs (GraphPad). Response variables shown to have a significant effect (P,0.05) by treatment group were subjected to pair-wise comparison using the Tukey–Kramer test.

In vivo Study 2. Using identical containment rooms and a similar

Supplementary Methods. Methods used for ELISA, haemaggluti-

nation inhibition, and growth curve assays are available in the online Supplementary Material.

In vivo Study 4. To further evaluate the nasal shedding profile of the

HA-S/I and HA-P/L clones on the Ca/09 backbone, 10 pigs were intratracheally inoculated with 2 ml Ca/09 : Ca HA-S/I (n55) or Ca/ 09 : Ca HA-P/L (n55) at 26105 TCID50 ml21 in a similar study as above. Five pigs were introduced in direct contact with the infected pigs and five pigs were introduced in respiratory contact per each virus group at 2 days p.i. RNA isolation and sequence analysis. Challenge inoculum from

each group of Studies 3 and 4 as well as nasal swab filtrates in MEM from three pigs per virus group equally distributed from the infected, direct-contact and respiratory-contact groups from Studies 1, 3 and 4 were selected and individually processed. RNA isolation was carried out using the MagMAX-96 Total RNA isolation kit (Ambion) in accordance with the manufacturer’s protocol. A 500 bp fragment of the HA gene flanking the receptor-binding site coding sequence was amplified. Attachment and infectivity assays. For the assay, confluent MDCK cells in 12-well plates were infected at m.o.i. 0.1 in duplicate wells with virus clones Ca/09 : Ca HA-S/I or Ca/09 : Ca HA-P/L. After 1 h of incubation at 37 uC, supernatant was harvested and frozen at 280 uC for titration, and cells were harvested by gentle scraping, pipetting and pelleting by centrifugation for staining and flow cytometric analysis. To evaluate virus at the cell surface (attachment) versus intracellular virus (infectivity), the cells were fixed with 4 % paraformaldehyde in PBS and resuspended in PBS with 5 % BSA and 0.1 % sodium azide (FACS buffer) or permeabilized with FACS buffer with 0.1 % saponin (PERM buffer). A mAb against NP (IgG2a, clone HB-65; ATCC) was diluted in PERM or FACS buffer and added to the cells for 15 min. Cells were washed and the secondary antibody (goat anti-mouse IgG2a-PE; Southern Biotech) diluted in PERM or FACS buffer was added and incubated for 15 min. Cells were washed, and data were acquired using a FACScan flow cytometer (Becton Dickinson) and analysed using CellQuest software (Becton Dickinson).

For the attachment and infectivity time-course assay, confluent MDCK cells in 12-well plates were infected at m.o.i. 0.001 in duplicate wells with virus clones Ca/09 : Ca HA-S/I or Ca/09 : Ca HA-P/L. After 8, 16, 24, 36 and 48 h incubation at 37 uC, supernatant was harvested and frozen at 280 uC for titration, and cells were harvested by gentle scraping, pipetting and pelleting by centrifugation for staining and flow cytometric analysis. Cells were stained and processed in PERM buffer and analysed by flow cytometry as described above. Distribution of S/P186 from publicly available sequences.

H1N1pdm09 HA protein sequences were retrieved from publicly available databases, and the amino acid frequency at positions 186 and 194 analysed using the Analyse Sequence Variation tool at the http://vir.sgmjournals.org

ACKNOWLEDGEMENTS The authors thank Michelle Harland and Hillary Horst for technical assistance; David Alt and Lea Ann Hobbs for sequencing; and Dr Becky Jepsen, Brian Pottebaum, Jason Huegel and Jason Crabtree for assistance with animal studies. Isolates of H1N1pdm09 were generously provided by the late Dr Alexander Klimov (Centers for Disease Control and Prevention). Mention of trade names or commercial products in this paper is solely for the purpose of providing specific information and does not imply recommendation or endorsement by the US Department of Agriculture (USDA). Funding was provided by USDA-ARS and DHHS-CDC. USDA is an equal opportunity provider and employer.

REFERENCES Anderson, T. K., Nelson, M. I., Kitikoon, P., Swenson, S. L., Korslund, J. A. & Vincent, A. L. (2013). Population dynamics of cocirculating

swine influenza A viruses in the United States from 2009 to 2012. Influenza Other Respir Viruses 7 (Suppl. 4), 42–51. Ayora-Talavera, G., Shelton, H., Scull, M. A., Ren, J., Jones, I. M., Pickles, R. J. & Barclay, W. S. (2009). Mutations in H5N1 influenza

virus hemagglutinin that confer binding to human tracheal airway epithelium. PLoS ONE 4, e7836. CDC (2010). Update: influenza activity - United States, August 30, 2009-March 27, 2010, and composition of the 2010-11 influenza vaccine. MMWR Morb Mortal Wkly Rep 59, 423–430. Childs, R. A., Palma, A. S., Wharton, S., Matrosovich, T., Liu, Y., Chai, W., Campanero-Rhodes, M. A., Zhang, Y., Eickmann, M. & other authors (2009). Receptor-binding specificity of pandemic influenza A

(H1N1) 2009 virus determined by carbohydrate microarray. Nat Biotechnol 27, 797–799. de Wit, E., Munster, V. J., van Riel, D., Beyer, W. E., Rimmelzwaan, G. F., Kuiken, T., Osterhaus, A. D. & Fouchier, R. A. (2010). Molecular

determinants of adaptation of highly pathogenic avian influenza H7N7 viruses to efficient replication in the human host. J Virol 84, 1597–1606. Garten, R. J., Davis, C. T., Russell, C. A., Shu, B., Lindstrom, S., Balish, A., Sessions, W. M., Xu, X., Skepner, E. & other authors (2009).

Antigenic and genetic characteristics of swine-origin 2009 A(H1N1) influenza viruses circulating in humans. Science 325, 197–201. Gibbs, A. J., Armstrong, J. S. & Downie, J. C. (2009). From where did the

2009 ‘swine-origin’ influenza A virus (H1N1) emerge? Virol J 6, 207. 2625

A. Lorusso and others

Hoffmann, E., Neumann, G., Kawaoka, Y., Hobom, G. & Webster, R. G. (2000). A DNA transfection system for generation of influenza A

Novel Swine-Origin Influenza A (H1N1) Virus Investigation Team (2009). Emergence of a novel swine-origin influenza A (H1N1) virus

virus from eight plasmids. Proc Natl Acad Sci U S A 97, 6108–6113.

in humans. N Engl J Med 360, 2605–2615.

Ilyushina, N. A., Khalenkov, A. M., Seiler, J. P., Forrest, H. L., Bovin, N. V., Marjuki, H., Barman, S., Webster, R. G. & Webby, R. J. (2010).

Shinya, K., Ebina, M., Yamada, S., Ono, M., Kasai, N. & Kawaoka, Y. (2006). Avian flu: influenza virus receptors in the human airway.

Adaptation of pandemic H1N1 influenza viruses in mice. J Virol 84, 8607–8616.

Nature 440, 435–436.

Lorusso, A., Vincent, A. L., Harland, M. L., Alt, D., Bayles, D. O., Swenson, S. L., Gramer, M. R., Russell, C. A., Smith, D. J. & other authors (2011). Genetic and antigenic characterization of H1

influenza viruses from United States swine from 2008. J Gen Virol 92, 919–930.

Squires, R. B., Noronha, J., Hunt, V., Garcı´a-Sastre, A., Macken, C., Baumgarth, N., Suarez, D., Pickett, B. E., Zhang, Y. & other authors (2012). Influenza research database: an integrated bioinformatics

resource for influenza research and surveillance. Influenza Other Respir Viruses 6, 404–416.

Lorusso, A., Vincent, A. L., Gramer, M. E., Lager, K. M. & CiacciZanella, J. R. (2013). Contemporary epidemiology of North American

Suphaphiphat, P., Franti, M., Hekele, A., Lilja, A., Spencer, T., Settembre, E., Palmer, G., Crotta, S., Tuccino, A. B. & other authors (2010). Mutations at positions 186 and 194 in the HA gene of the

lineage triple reassortant influenza A viruses in pigs. Curr Top Microbiol Immunol 370, 113–132.

2009 H1N1 pandemic influenza virus improve replication in cell culture and eggs. Virol J 7, 157.

Matrosovich, M. N., Gambaryan, A. S., Teneberg, S., Piskarev, V. E., Yamnikova, S. S., Lvov, D. K., Robertson, J. S. & Karlsson, K. A. (1997). Avian influenza A viruses differ from human viruses by

Vincent, A. L., Swenson, S. L., Lager, K. M., Gauger, P. C., Loiacono, C. & Zhang, Y. (2009). Characterization of an influenza A virus

recognition of sialyloligosaccharides and gangliosides and by a higher conservation of the HA receptor-binding site. Virology 233, 224–234. Meroz, D., Yoon, S. W., Ducatez, M. F., Fabrizio, T. P., Webby, R. J., Hertz, T. & Ben-Tal, N. (2011). Putative amino acid determinants of

the emergence of the 2009 influenza A (H1N1) virus in the human population. Proc Natl Acad Sci U S A 108, 13522–13527. Nicholls, J. M., Chan, R. W., Russell, R. J., Air, G. M. & Peiris, J. S. (2008). Evolving complexities of influenza virus and its receptors.

Trends Microbiol 16, 149–157.

2626

isolated from pigs during an outbreak of respiratory disease in swine and people during a county fair in the United States. Vet Microbiol 137, 51–59. Yang, H., Carney, P. & Stevens, J. (2010). Structure and receptor

binding properties of a pandemic H1N1 virus hemagglutinin. PLoS Curr 2, RRN1152. Ye, J., Sorrell, E. M., Cai, Y., Shao, H., Xu, K., Pena, L., Hickman, D., Song, H., Angel, M. & other authors (2010). Variations in the

hemagglutinin of the 2009 H1N1 pandemic virus: potential for strains with altered virulence phenotype? PLoS Pathog 6, e1001145.

Journal of General Virology 95

Polymorphisms in the haemagglutinin gene influenced the viral shedding of pandemic 2009 influenza virus in swine.

Interactions between the viral surface glycoprotein haemagglutinin (HA) and the corresponding receptors on host cells is one important aspect of influ...
213KB Sizes 0 Downloads 3 Views