key: cord-0969207-u7ka7rbc authors: He, Cheng‐Qiang; He, Mei; He, Hong‐Bin; Wang, Hong‐Mei; Ding, Nai‐Zheng title: The matrix segment of the “Spanish flu” virus originated from intragenic recombination between avian and human influenza A viruses date: 2019-07-15 journal: Transbound Emerg Dis DOI: 10.1111/tbed.13282 sha: f2b3d266cd19602866b54716c898729e39c30be7 doc_id: 969207 cord_uid: u7ka7rbc The 1918 Spanish flu virus has claimed more than 50 million lives. However, the mechanism of its high pathogenicity remains elusive; and the origin of the virus is controversial. The matrix (M) segment regulates the replication of influenza A virus, thereby affecting its virulence and pathogenicity. This study found that the M segment of the Spanish flu virus is a recombinant chimera originating from avian influenza virus and human influenza virus. The unique mosaic M segment might confer the virus high replication capacity, showing that the recombination might play an important role in inducing high pathogenicity of the virus. In addition, this study also suggested that the NA and NS segments of the virus were generated by reassortment between mammalian and avian viruses. Direct phylogenetic evidence was also provided for its avian origin. The 1918 "Spanish Flu" was the most devastating pandemic in modern history (Honigsbaum, 2018) . In 11 months between the spring of 1918 and the winter of 1919, it killed approximately 50 million people worldwide (Johnson & Mueller, 2002; Patterson & Pyle, 1991) . The morbidity pattern, together with the rapid disease progression to multiorgan failure and death, characterizes the influenza pandemic (Honigsbaum, 2018) . So far, its origin, its unusual epidemiologic features, and the basis of its pathogenicity remain elusive (Taubenberger & Morens, 2006) . Its pathogen, a variant of H1N1 influenza A virus (IVA) (Gibbs, Armstrong, & Gibbs, 2001; Kilbourne, 2006) , has the eight-segmented genome housed in an enveloped virion (Noda et al., 2006) . Knowing the origin of the virus might be a key for uncovering its epidemiologic features and pathogenicity basis . However, there are two controversies regarding its origin. The first one is whether the virus originated from mammalian IVA reassortment or directly from an avian virus (Antonovics, Hood, & Baker, 2006; Gibbs & Gibbs, 2006; Taubenberger, 2006; Vana & Westover, 2008) . And, the second one is whether homologous recombination shaped the formation of the virus (Gibbs et al., 2001; Worobey, Rambaut, Pybus, & Robertson, 2002) . Based on a comparison analysis of the amino acid (AA) sequences of the viral proteins, Taubenberger et al., (2005) . found that some proteins have the characteristics of avian viruses, and proposed that the virus was not a reassortant, but an avian virus in origin. However, this interpretation was questioned by several other groups because some genes of the virus were obviously clustered into swine or human branches in IVA phylogenetic trees (Antonovics et al., 2006; Gibbs & Gibbs, 2006; Vana & Westover, 2008) . Therefore, the virus is considered a reassortant from mammalian IVAs (Smith et al., 2009 ). Besides of genetic reassortment (Taubenberger, 2006) , some additional events might also contribute to the formation of the virus and the significant change of its phenotype triggering the pandemic (Basler et al., 2001; Reid, Fanning, Hultin, & Taubenberger, 1999; Webster, 1999) . Novel virulent variants of several other viruses have been shown to be generated through homologous recombination (Parrish et al., 2008; Sabir et al., 2016; Worobey, Rambaut, & Holmes, 1999) . It was thus proposed that homologous recombination also occurred in the HA gene and resulted in the increased virulence associated with the pandemic (Gibbs et al., 2001) . Unfortunately, the | 2189 HE Et al. recombination event was doubted because of the complete absence of phylogenetic evidence for recombination in HA (Worobey et al., 2002) . Of the eight segments in IVA, the segment 7 encodes the two matrix (M) proteins, M1 and M2 (Lamb, Lai, & Choppin, 1981; Winter & Fields, 1980) . The open reading frames (ORFs) of the M1 and M2 genes share the first nine codons at the N-terminus, and the C-terminus of M1 overlaps with a region of M2. M1 consists of 252 AAs, while M2 consists of 97 AAs encoded by an alternatively spliced transcript. Lining the inner layer of the viral membrane and contacting the ribonucleoprotein (RNP) core, M1 is highly conserved and is the most abundant protein in viral particles (Reid, Fanning, Janczewski, McCall, & Taubenberger, 2002) . M1 regulates the nuclear export of viral RNPs (Bui, Wills, Helenius, & Whittaker, 2000; Martin & Helenius, 1991) , restricts viral replication (Liu & Ye, 2002) , inhibits viral transcription in the late stages of infection and the switch from replication to viral assembly (Perez & Donis, 1998; Ye, Baylor, & Wagner, 1989 ) and influences virus assembly and budding (Gomez-Puertas, Albo, Perez-Pastrana, Vivo, & Portela, 2000; Helenius, 1992; Latham & Galarza, 2001) . Anchored in the viral envelope, M2 serves as a transmembrane ion channel (Lamb, Zebedee, & Richardson, 1985; Sugrue & Hay, 1991) and plays key roles in both virion uncoating and viral budding (Grambas & Hay, 1992; Pinto, Holsinger, & Lamb, 1992) . To clarify the origin of the Spanish flu virus, we re-dissected the phylogenetic history of the virus and found that its M segment was a mosaic recombined from human and avian influenza viruses, showing that homologous recombination is truly linked with the most devastating pandemic. In addition, we also provided direct phylogenetic evidence for the avian origin of the virus. The segment sequences of the 1918 IVA isolate A/Brevig Mission/1/1918(H1N1) are from previous reports (Basler et al., 2001; Reid et al., , 2002 Reid, Fanning, Janczewski, Lourens, & Taubenberger, 2004; Reid, Fanning, Janczewski, & Taubenberger, 2000; Taubenberger et al., 2005) . Referring to previous studies (Smith et al., 2009; Vana & Westover, 2008) , eight segments of classical strains of different species (human, avian, swine, and horse) and A/Brevig Mission/1/1918(H1N1) were respectively concatenated to analyze the potential reassortment or recombination event between viruses of different species in A/Brevig Mission/1/1918(H1N1). Employing basic local alignment search tool (BLAST), reassortment between viruses of different species in these representative genomes was checked through comparing their sequence identity with IVAs deposited in GenBank (https ://blast.ncbi.nlm.nih.gov/Blast. cgi). Based on sequence identity to the query segment, the host species distribution of the top 100 sbjct viruses is shown in Table S1 . Except for PB1 of A/chicken/Rostock/45/1934 (H7N1) that may be associated with human IVA, no other reassortment events were found in these representative viruses. To dissect the phylogenetic history of the M segment of the Spanish flu virus, 23 viruses of different years (from 1902 to 2012) and different species were selected. Information about these viruses was included in virus name. The sequences of IVAs were aligned with the MUSCLE programme implemented in MEGA 6 (Tamura et al., 2011) . The phylogenetic histories of these viruses were inferred using the Maximum Likelihood (ML) or Neighbour-Joining (NJ) methods based on the best substitution model selected by the model test programme in MEGA 6 (Tamura et al., 2011) . The robustness of each lineage was tested using the bootstrap method (≥1,000 replicates). And the monophylogenetic lineage with a bootstrap value ≥70% was considered robust. Recombination analysis was carried out as per our previous studies . In brief, to distinguish the recombinant, similarity comparison of nucleotide sequences between the putative recombinants and their parents was performed using the sliding window method in the Simplot programme (Lole et al., 1999) . The sequence with contradiction identity was considered to be a putative recombinant. Integrating the Fisher's exact test method, we identified the putative breakpoints (p < 0.05) with the maximum chi-square value of information site. A set of statistically incongruent phylogenetic trees were recommended as the gold-standard approach for confirming the presence of recombination (Boni, Jong, Doorn, & Holmes, 2010) . Therefore, the recombinants were finally determined through the incongruent phylogenetic histories of different regions delimited by the putative recombination breakpoints. Shimodaira-Hasegawa test was implemented to prove whether phylogenetic trees estimated from different regions were significantly different employing the Tree test programme (http://aix1.uotta wa.ca/~saris bro/). In order to dissect the phylogenetic history of the M segment of the Spanish flu virus, 23 IVA representatives including H1N1 human (n = 11), classical H1N1 swine (n = 2), mixture avian (n = 8) and H7N7 equine (n = 2) were used to infer the origin of the M segment of the virus. According to the phylogenetic history constructed from the complete M segments of these viruses, the Spanish flu virus belongs to the same monophylogentic lineage as the human H1N1 viruses Notably, this difference was only attributed to the Spanish flu virus. In order to distinguish whether the cluster of the Spanish flu virus into the avian branch was due to homologous recombination or convergent evolution adapting to avian host, we reconstructed its phylogenetic history using the third nucleotides of codons in the M1 and M2 ORFs between the two breakpoints (Figure 3b') . The topologies of the two trees were identical (Figure 3b and b') . Therefore, it is more likely that homologous recombination caused the Spanish flu virus to jump into the avian branch rather than convergent evolution adapting to avian host. (Figure 4a and b) . In addition, according to the sequence similarity and bootscan analysis of the concatenated IVA genome and phylogenetic reconstruction of NA and NS, the NA and NS segments of the Spanish flu virus might be re-assorted from avian IVAs ( Figure S1 ). Within the HA segments, there is also a crossover site in the sequence similarity plot ( Figure S1A ). However, there is no phylogenetic evidence that the similarity crossover of the HA gene is due to recombination ( Figure S1C and D) , which is consistent with the previous report (Worobey et al., 2002) . (Antonovics et al., 2006; Gibbs & Gibbs, 2006; Smith et al., 2009; Vana & Westover, 2008) . However, the Spanish flu virus has multiple proteins carrying the unique AAs of the avian branch, suggesting its progenitors had infected avian before its outbreak; and thus, the mutations adaptive to birds might have been fixed in virus genome during circulation in avian host (Basler et al., 2001; Reid et al., , 2004 Reid et al., , 2000 Taubenberger et al., 2005) . Interestingly, in this study, by introducing a bird IVA lineage isolated in the early 20th century into the dataset, we found that the NA and NS segments of the Spanish flu virus are more likely from avian IVA. Moreover, approximately 300 nt fragment of its M segment is also inherited from avian IVA. These results provided direct evidence for the avian origin of the virus. The two major advantages of recombination over mutation are that recombination accelerates the rate at which advantageous genetic combinations are produced and allows more efficient removal of deleterious mutations (Simon-Loriere & Holmes, 2011) . This could also result in the change of host tropism and virulent phenotypes, as occurred with the human immunodeficiency virus (HIV) (Lemey, Rambaut, & Pybus, 2006) and Middle East respiratory syndrome coronavirus (Sabir et al., 2016) . Even for negative RNA viruses, recombination has also been responsible for outbreaks of bovine ephemeral fever virus in cattle (He et al., 2016) , bat rabies virus in skunk and raccoon (Ding, Xu, Sun, He, & He, 2017) and severe fever with thrombocytopenia syndrome bunyavirus in human . For the Spanish flu virus, Gibbs et al., (2001) . proposed its HA was recombined from human and swine IVAs, and the recombination played the key role in the outbreak of the virus. However, this recombination evidence was thought to be invalid because the phylogenetic histories of HA were not different in the regions inside and outside the proposed recombination breakpoints (Worobey et al., 2002) . Here, we provided robust phylogenetic evidence that the M segment was a chimera inherited from a human virus and a bird virus, showing that homologous recombination does be the important genetic mechanism driving the formation of the Spanish flu virus, although recombination might infrequently occur between IVAs. The avian HA protein binds preferentially to 2,3-linked sialic acids, whereas human HA protein binds preferentially to 2,6-linked sialic acids. This determines the host difference (Matrosovich & Klenk, 2003) . However, outbreaks of avian IVAs, such as H5N1 and H7N9, have been found in human being (Nga et al., 2019) , suggesting that avian IVA can potential infect human. Several studies have also shown that HA of H7N9 can bind both human-and avian-type receptors (Belser et al., 2013; Watanabe et al., 2013; Zhang et al., 2013) . Therefore, it is not strange that the Spanish flu virus is a mosaic from avian and human IVAs. Based on this study and previous reports (Smith et al., 2009; Taubenberger et al., 2005; Vana & Westover, 2008) can also increase the virulence of the virus (Malaspinas, Malaspinas, Evans, & Slatkin, 2012; Qi et al., 2018) . Here, we found that the recombination has shaped a new M segment that is significantly different from any avian and mammalian IVAs, which may give the virus a unique biological phenotype. It has been known that the disease course and pathological damage caused by the virus are associated with the capability of replicating to titers higher than those of other strains (Wolbach, 1919) . The M protein is the regulator of IVA growth (Yasuda, Bucher, & Ishihama, 1994; Yasuda, Toyoda, Nakayama, & Ishihama, 1993) . Small changes in M can have large effects on replication phenotype (Reid et al., 2002) . M can also affect the virulence of the virus (Brown, Liu, Kit, Baird, & Nesrallah, 2001; Smeenk, Wright, Burns, Thaker, & Brown, 1996) . Through the recombination, the Spanish flu virus acquired the seven unique AAs from the avian IVA in M1 and M2. Previous studies have shown that the mosaic proteins might have given the Spanish flu virus a high replication capacity. The recombinant M segments of its offspring strains PR34 (A/Puerto Rico/8/34 (H1N1)) and WSN33 (A/WSN/33 (H1N1)) ( Figure S2 ) have been shown to confer high-growth characteristics in single-gene reassortant strains (Yasuda et al., 1994) , suggesting that the recombination may give the Spanish flu virus a high replication power, and thus, influence its pathogenicity. The 1957 Asian flu and 1968 Hong Kong flu pandemics also involved IVAs of avian and human origins (Kawaoka, Krauss, & Webster, 1989) . To see whether their M segments are the direct Figure S2 ). In conclusion, this study showed that the M segment of the Spanish flu virus was a recombinant originating from human and avian IVAs, while the NA and NS segments were generated by reassortment between mammalian and avian IVAs, providing direct evidence of its avian origin. Moreover, the recombination might be associated with the high replication capacity of the virus and thus play an important role in its high pathogenicity. https://orcid.org/0000-0001-9795-0616 Molecular virology: was the 1918 flu avian in origin? Sequence of the 1918 pandemic influenza virus nonstructural gene (NS) segment and characterization of recombinant viruses bearing the 1918 NS genes Pathogenesis and transmission of avian influenza A (H7N9) virus in ferrets and mice Guidelines for identifying homologous recombination events in influenza A virus Pattern of mutation in the genome of influenza A virus on adaptation to increased virulence in the mouse lung: Identification of functional themes Role of the influenza virus M1 protein in nuclear export of viral ribonucleoproteins A permanent host shift of rabies virus from Chiroptera to Carnivora associated with recombination Antigenic and genetic characteristics of swineorigin 2009 A(H1N1) influenza viruses circulating in humans Recombination in the hemagglutinin gene of the 1918 "Spanish flu Molecular virology: was the 1918 pandemic caused by a bird flu? Influenza virus matrix protein is the major driving force in virus budding Maturation of influenza A virus hemagglutinin-estimates of the pH encountered during transport and its regulation by the M2 protein Discovery of severe fever with thrombocytopenia syndrome bunyavirus strains originating from intragenic recombination Identification of three H1N1 influenza virus groups with natural recombinant genes circulating from New genetic mechanism, origin and population dynamic of bovine ephemeral fever virus Homologous recombination as an evolutionary force in the avian influenza A virus Unpacking the incoming influenza virus Spanish influenza redux: Revisiting the mother of all pandemics Updating the accounts: Global mortality of the 1918-1920 "Spanish" influenza pandemic Avian-to-human transmission of the PB1 gene of influenza A viruses in the 1957 and 1968 pandemics Influenza pandemics of the 20th century Sequences of mRNAs derived from genome RNA segment 7 of influenza virus: Colinear and interrupted mRNAs code for overlapping proteins Influenza virus M2 protein is an integral membrane protein expressed on the infectedcell surface Formation of wild-type and chimeric influenza virus-like particles following simultaneous expression of only four structural proteins HIV evolutionary dynamics within and among hosts Restriction of viral replication by mutation of the influenza virus matrix protein Full-length human immunodeficiency virus type 1 genomes from subtype C-infected seroconverters in India, with evidence of intersubtype recombination Estimating allele age and selection coefficient from time-serial data Nuclear transport of influenza virus ribonucleoproteins: The viral matrix protein (M1) promotes export and inhibits import Natural and synthetic sialic acidcontaining inhibitors of influenza virus receptor binding Zoonotic diseases from birds to humans in Vietnam: Possible diseases and their associated risk factors Architecture of ribonucleoprotein complexes in influenza A virus particles Cross-species virus transmission and the emergence of new epidemic diseases The geography and mortality of the 1918 influenza pandemic The matrix 1 protein of influenza A virus inhibits the transcriptase activity of a model influenza reporter genome in vivo Influenza virus M2 protein has ion channel activity Emergence and adaptation of a novel highly pathogenic H7N9 influenza virus in birds and humans from a 2013 human-infecting Low-Pathogenic Ancestor Origin and evolution of the 1918 "Spanish" influenza virus hemagglutinin gene Novel origin of the 1918 pandemic influenza virus nucleoprotein gene Characterization of the 1918 "Spanish" influenza virus matrix gene segment Characterization of the 1918 "Spanish" influenza virus neuraminidase gene The 1918 flu and other influenza pandemics: "over there" and back again Co-circulation of three camel coronavirus species and recombination of MERS-CoVs in Saudi Arabia Why do RNA viruses recombine? Mutations in the hemagglutinin and matrix genes of a virulent influenza virus variant, A/FM/1/47-MA, control different stages in pathogenesis Dating the emergence of pandemic influenza viruses Structural characteristics of the M2 protein of influenza A viruses: Evidence that it forms a tetrameric channel MEGA5: Molecular evolutionary genetics analysis using maximum likelihood, evolutionary distance, and maximum parsimony methods The origin and virulence of the 1918 "Spanish" influenza virus Influenza: The mother of all pandemics Characterization of the 1918 influenza virus polymerase genes Origin of the 1918 Spanish influenza virus: A comparative genomic analysis Characterization of H7N9 influenza A viruses isolated from humans 1918 Spanish influenza: The secrets remain elusive Cloning of influenza cDNA ino M13: The sequence of the RNA segment encoding the A/PR/8/34 matrix protein Comments on the pathology and bacteriology of fatal influenza cases, as observed at Camp Devens Massachusetts Widespread intraserotype recombination in natural populations of dengue virus Questioning the evidence for genetic recombination in the 1918 Growth control of influenza A virus by M1 protein: Analysis of transfectant viruses carrying the chimeric M gene Regulatory effects of matrix protein variations on influenza virus growth Transcription-inhibition and RNA-binding domains of influenza A virus matrix protein mapped with anti-idiotypic antibodies and synthetic peptides H7N9 influenza viruses are transmissible in ferrets by respiratory droplet Additional supporting information may be found online in the Supporting Information section at the end of the article. How to cite this article