Viral attachment to the host cell is critical for tissue and species specificity of virus infections. Recently, pattern of viral attachment (PVA) in human respiratory tract was determined for highly pathogenic avian influenza virus of subtype H5N1. However, PVA of human influenza viruses and other avian influenza viruses in either humans or experimental animals is unknown. Therefore, we compared PVA of two human influenza viruses (H1N1 and H3N2) and two low pathogenic avian influenza viruses (H5N9 and H6N1) with that of H5N1 virus in respiratory tract tissues of humans, mice, ferrets, cynomolgus macaques, cats, and pigs by virus histochemistry. We found that human influenza viruses attached more strongly to human trachea and bronchi than H5N1 virus and attached to different cell types than H5N1 virus. These differences correspond to primary diagnoses of tracheobronchitis for human influenza viruses and diffuse alveolar damage for H5N1 virus. The PVA of low pathogenic avian influenza viruses in human respiratory tract resembled that of H5N1 virus, demonstrating that other properties determine its pathogenicity for humans. The PVA in human respiratory tract most closely mirrored that in ferrets and pigs for human influenza viruses and that in ferrets, pigs, and cats for avian influenza viruses. Viral attachment to the host cell is critical for tissue and species specificity of virus infections. Recently, pattern of viral attachment (PVA) in human respiratory tract was determined for highly pathogenic avian influenza virus of subtype H5N1. However, PVA of human influenza viruses and other avian influenza viruses in either humans or experimental animals is unknown. Therefore, we compared PVA of two human influenza viruses (H1N1 and H3N2) and two low pathogenic avian influenza viruses (H5N9 and H6N1) with that of H5N1 virus in respiratory tract tissues of humans, mice, ferrets, cynomolgus macaques, cats, and pigs by virus histochemistry. We found that human influenza viruses attached more strongly to human trachea and bronchi than H5N1 virus and attached to different cell types than H5N1 virus. These differences correspond to primary diagnoses of tracheobronchitis for human influenza viruses and diffuse alveolar damage for H5N1 virus. The PVA of low pathogenic avian influenza viruses in human respiratory tract resembled that of H5N1 virus, demonstrating that other properties determine its pathogenicity for humans. The PVA in human respiratory tract most closely mirrored that in ferrets and pigs for human influenza viruses and that in ferrets, pigs, and cats for avian influenza viruses. Infections with human influenza A viruses of the subtypes H1N1 and H3N2 are important causes of respiratory tract disease. The most common lesion in immunocompetent individuals is tracheobronchitis.1Zambon MC Epidemiology and pathogenesis of influenza.J Antimicrob Chemother. 1999; 44: 3-9Crossref PubMed Scopus (191) Google Scholar Uncommonly, human influenza A virus infection causes severe pneumonia, which requires hospitalization and may be fatal. This pattern of disease contrasts with the ongoing outbreak of highly pathogenic avian influenza A virus infection of the subtype H5N1. In this outbreak, severe pneumonia is the most common lesion in the >300 patients with confirmed H5N1 virus infection, and the case fatality rate is over 50% (World Health Organization http://www.who.int/csr/disease/avian_influenza/country/cases_table_2007_06_04/en/index.html). Until now, there is no evidence that this avian virus has become efficiently transmissible among humans, which could result in a new pandemic.2Garcia-Sastre A Whitley RJ Lessons learned from reconstructing the 1918 influenza pandemic.J Infect Dis. 2006; 194: S127-S132Crossref PubMed Scopus (33) Google Scholar The increased interest in H5N1 virus infection has highlighted large gaps in our knowledge of the pathogenesis of influenza A virus infections in humans. An important factor in this pathogenesis is tissue tropism, which depends largely on the ability of the virus to attach to the host cell.3Baigent SJ McCauley JW Influenza type A in humans, mammals and birds: determinants of virus virulence, host-range and interspecies transmission.Bioessays. 2003; 25: 657-671Crossref PubMed Scopus (187) Google Scholar Influenza A viruses attach to host cells by binding of the hemagglutinin (HA) protein to sialosaccharides on the host cell surface. The HAs of influenza A viruses from different host species differ in their specificity of binding. For example, HAs of human influenza A viruses preferentially recognize sialic acid (SA)-α-2,6-Gal-terminated saccharides (α-2,6-SA), whereas HAs of avian influenza viruses preferentially recognize SA-α-2,3-Gal-terminated saccharides (α-2,3-SA).4Rogers GN Paulson JC Receptor determinants of human and animal influenza virus isolates: differences in receptor specificity of the H3 hemagglutinin based on species of origin.Virology. 1983; 127: 361-373Crossref PubMed Scopus (698) Google Scholar, 5Suzuki Y Sialobiology of influenza: molecular mechanism of host range variation of influenza viruses.Biol Pharm Bull. 2005; 28: 399-408Crossref PubMed Scopus (362) Google Scholar, 6Connor RJ Kawaoka Y Webster RG Paulson JC Receptor specificity in human, avian, and equine H2 and H3 influenza virus isolates.Virology. 1994; 205: 17-23Crossref PubMed Scopus (665) Google Scholar These differences generally correspond with the variation in the type of SAs expressed at important sites for influenza A virus replication in the respective host species. For example, human tracheal epithelium expresses mainly α-2,6-SA,7Couceiro JN Paulson JC Baum LG Influenza virus strains selectively recognize sialyloligosaccharides on human respiratory epithelium: the role of the host cell in selection of hemagglutinin receptor specificity.Virus Res. 1993; 29: 155-165Crossref PubMed Scopus (367) Google Scholar whereas duck intestinal epithelium expresses mainly α-2,3-SA.8Ito T Couceiro JN Kelm S Baum LG Krauss S Castrucci MR Donatelli I Kida H Paulson JC Webster RG Kawaoka Y Molecular basis for the generation in pigs of influenza A viruses with pandemic potential.J Virol. 1998; 72: 7367-7373Crossref PubMed Google Scholar Therefore, the type and distribution of SA is considered to be an important factor in the susceptibility of different host species to influenza A viruses.9Suzuki Y Ito T Suzuki T Holland Jr, RE Chambers TM Kiso M Ishida H Kawaoka Y Sialic acid species as a determinant of the host range of influenza A viruses.J Virol. 2000; 74: 11825-11831Crossref PubMed Scopus (412) Google Scholar The SA recognized by influenza A virus is not only important in the host species range but also in its transmissibility. The latter was demonstrated in experimental infections of ferrets with the 1918 pandemic influenza virus. In this study, horizontal transmission was abolished by two amino acid mutations in the HA that caused a switch in binding preference from human α-2,6-SA to avian α-2,3-SA.10Tumpey TM Maines TR Van Hoeven N Glaser L Solorzano A Pappas C Cox NJ Swayne DE Palese P Katz JM Garcia-Sastre A A two-amino acid change in the hemagglutinin of the 1918 influenza virus abolishes transmission.Science. 2007; 315: 655-659Crossref PubMed Scopus (477) Google Scholar A traditional method for studying the tissue tropism of different influenza A viruses is to measure the distribution of SAs by use of lectin histochemistry. Lectin histochemistry makes use of the property of a wide variety of lectins to specifically bind to SAs.11Varki A Sialic acids as ligands in recognition phenomena.FASEB J. 1997; 11: 248-255Crossref PubMed Scopus (491) Google Scholar Studies on receptors for influenza virus have made use of the plant lectins Sambucus nigra agglutinin, which has a major specificity for α-2,6-SA, and Maackia amurensis agglutinin, which has a major specificity for α-2,3-SA.12Baum LG Paulson JC Sialyloligosaccharides of the respiratory epithelium in the selection of human influenza virus receptor specificity.Acta Histochem Suppl. 1990; 40: 35-38PubMed Google Scholar, 13Shinya K Ebina M Yamada S Ono M Kasai N Kawaoka Y Avian flu: influenza virus receptors in the human airway.Nature. 2006; 440: 435-436Crossref PubMed Scopus (1094) Google Scholar Although useful for determining the distribution of SAs in tissues, these lectin histochemistry techniques are only an indirect measure of influenza A virus attachment to host tissues. They do not account for other variables that influence the binding specificity. For HA, these include glycosylation and sialylation close to the receptor binding site5Suzuki Y Sialobiology of influenza: molecular mechanism of host range variation of influenza viruses.Biol Pharm Bull. 2005; 28: 399-408Crossref PubMed Scopus (362) Google Scholar; for the receptor, these include type of SA, alternative linkages,14Wu W Air GM Binding of influenza viruses to sialic acids: reassortant viruses with A/NWS/33 hemagglutinin bind to alpha2,8-linked sialic acid.Virology. 2004; 325: 340-350Crossref PubMed Scopus (33) Google Scholar and sulfation and fucosylation of the saccharide residues.15Gambaryan A Tuzikov A Pazynina G Bovin N Balish A Klimov A Evolution of the receptor binding phenotype of influenza A (H5) viruses.Virology. 2006; 344: 432-438Crossref PubMed Scopus (165) Google Scholar To circumvent these problems, we made use of virus histochemistry to study the pattern of virus attachment (PVA) in respiratory tissues. This method, modified from Couceiro et al,7Couceiro JN Paulson JC Baum LG Influenza virus strains selectively recognize sialyloligosaccharides on human respiratory epithelium: the role of the host cell in selection of hemagglutinin receptor specificity.Virus Res. 1993; 29: 155-165Crossref PubMed Scopus (367) Google Scholar directly displays the attachment of influenza virus to tissues. By use of this method, we recently determined that H5N1 virus attachment in the human respiratory tract is progressively more abundant toward the alveoli, where the virus attaches predominantly to type II pneumocytes and alveolar macrophages.16van Riel D Munster VJ de Wit E Rimmelzwaan GF Fouchier RA Osterhaus AD Kuiken T H5N1 virus attachment to lower respiratory tract.Science. 2006; 312: 399Crossref PubMed Scopus (555) Google Scholar This attachment pattern fits with the limited pathology data on H5N1 virus infection in humans, which show diffuse alveolar damage as the primary lesion.17Uiprasertkul M Puthavathana P Sangsiriwut K Pooruk P Srisook K Peiris M Nicholls JM Chokephaibulkit K Vanprapar N Auewarakul P Influenza A H5N1 replication sites in humans.Emerg Infect Dis. 2005; 11: 1036-1041Crossref PubMed Scopus (231) Google Scholar, 18To KF Chan PK Chan KF Lee WK Lam WY Wong KF Tang NL Tsang DN Sung RY Buckley TA Tam JS Cheng AF Pathology of fatal human infection associated with avian influenza A H5N1 virus.J Med Virol. 2001; 63: 242-246Crossref PubMed Scopus (386) Google Scholar, 19Ng WF To KF Lam WW Ng TK Lee KC The comparative pathology of severe acute respiratory syndrome and avian influenza A subtype H5N1: a review.Hum Pathol. 2006; 37: 381-390Abstract Full Text Full Text PDF PubMed Scopus (121) Google Scholar Our results on H5N1 virus were supported by recent experimental studies in human ex vivo lung cultures, which demonstrated H5N1 virus replication in the lower respiratory tract (LRT).13Shinya K Ebina M Yamada S Ono M Kasai N Kawaoka Y Avian flu: influenza virus receptors in the human airway.Nature. 2006; 440: 435-436Crossref PubMed Scopus (1094) Google Scholar, 20Nicholls JM Chan MC Chan WY Wong HK Cheung CY Kwong DL Wong MP Chui WH Poon LL Tsao SW Guan Y Peiris JS Tropism of avian influenza A (H5N1) in the upper and lower respiratory tract.Nat Med. 2007; 13: 147-149Crossref PubMed Scopus (283) Google Scholar Although limited studies have been done on the human trachea,7Couceiro JN Paulson JC Baum LG Influenza virus strains selectively recognize sialyloligosaccharides on human respiratory epithelium: the role of the host cell in selection of hemagglutinin receptor specificity.Virus Res. 1993; 29: 155-165Crossref PubMed Scopus (367) Google Scholar the PVA for human influenza A viruses in the LRT is not known. This information is important to understand better the pathogenesis of influenza pneumonia, which is centered on the LRT. In addition, it is not clear whether the PVA of H5N1 virus that we observed is unique among avian influenza viruses and therefore may in part explain its ability to cause respiratory disease in humans. Finally, the PVA of human influenza A virus in experimental animals is not known. This information is important to help select the most appropriate animal model for influenza pneumonia. Of particular interest is the domestic pig, which is permissive for both human and avian influenza A virus infections and may thus act as a “mixing vessel” for the generation of reassortant viruses.21Webster RG Bean WJ Gorman OT Chambers TM Kawaoka Y Evolution and ecology of influenza A viruses.Microbiol Rev. 1992; 56: 152-179Crossref PubMed Google Scholar Therefore, we here describe the PVA of two currently circulating subtypes of human influenza A virus (H3N2 and H1N1) and low pathogenic avian influenza viruses (H5N9 and H6N1) to compare these with the PVA of highly pathogenic avian influenza A virus H5N1 in human respiratory tract. Furthermore, we determined the PVA of these human and avian influenza A viruses in respiratory tract of known experimental animals. To determine the PVA of human and avian influenza A viruses in the trachea and LRT of humans, we used two low pathogenic avian influenza viruses (H5N9 and H6N1), a highly pathogenic avian influenza A virus H5N1 isolate, and two recently circulating human influenza viruses (H3N2 and H1N1). We determined whether attachment occurred to epithelial cells in the trachea or LRT (including bronchi, bronchioles, and alveoli) and to alveolar macrophages. The PVA of all of the above viruses was also determined in mammalian species, which are used for experimental influenza A virus infections. Animals included were cynomolgus macaque (Macaca fascicularis), European shorthair cat, ferret, Yorkshire-Landrace pig, and C57/BL6 mouse. Influenza virus A/Netherlands/35/05 (H1N1) and A/Netherlands/213/03 (H3N2) are recent human isolates grown on Madin-Darby canine kidney cells and were kindly provided by the National Influenza Center (Rotterdam, The Netherlands). Influenza virus A/Mallard/Sweden/79/02 (H5N9) and A/Mallard/Sweden/81/02 (H6N1) were obtained from cloacal swabs of migratory mallard ducks (Anas platyrhynchos) during ongoing influenza virus surveillance of wild birds and were subsequently passaged twice in embryonated hens' eggs.22Munster VJ Wallensten A Baas C Rimmelzwaan GF Schutten M Olsen B Osterhaus AD Fouchier RA Mallards and highly pathogenic avian influenza ancestral viruses, northern Europe.Emerg Infect Dis. 2005; 11: 1545-1551Crossref PubMed Scopus (195) Google Scholar Influenza virus A/Vietnam/1194/04 (H5N1) was isolated from a fatal human case. The virus was kindly provided by Dr. W. Lim (Queen Mary Hospital, Hong Kong, People's Republic of China), and propagated once in Madin-Darby canine kidney cells. The H1N1, H3N2, and H5N1 viruses, isolated from humans, were grown in Madin-Darby canine kidney cells. The supernatant was harvested and cleared by low-speed centrifugation. The H5N9 and H6N1 viruses, isolated from mallards, were grown in the allantoic cavity of 11-day-old embryonated hens' eggs. The allantoic fluid was harvested after 2 days and cleared by low-speed centrifugation. Cleared supernatants and allantoic fluid samples were subsequently centrifuged 2 hours at 85,000 × g in a SW28 rotor at 4°C. The virus pellet was resuspended in 2 ml of phosphate-buffered saline (PBS), loaded on a 20 to 60% sucrose (w/w) gradient, and centrifuged overnight at 300,000 × g in a SW41 rotor at 4°C. To deplete the sucrose, the viruses were additionally centrifuged 2 hours at 85,000 × g in a SW28 rotor at 4°C, and the virus was resuspended in PBS. H5N1 virus was inactivated by dialyzing against 0.1% formalin for 3 days. All other viruses were inactivated by incubation with 1:1 (v/v) 10% formalin for 1 hour at room temperature. After inactivation, virus suspensions were dialyzed against PBS. Inactivation was confirmed by failure to passage on Madin-Darby canine kidney cells. Viruses were labeled by mixing concentrated, inactivated viruses, suspended in PBS, with an equal volume of 0.1 mg/ml fluorescein isothiocyanate (FITC) (Sigma, St. Louis, MO) in 0.5 mol/L bicarbonate buffer (pH 9.5) for 1 hour with constant stirring. To lose all unbound FITC, labeled viruses were dialyzed against PBS. To check for the continued capacity for hemagglutination by the inactivated viruses, the hemagglutination titer of the viruses was determined after formalin inactivation and FITC labeling.23Rimmelzwaan GF Baars M Claas EC Osterhaus AD Comparison of RNA hybridization, hemagglutination assay, titration of infectious virus and immunofluorescence as methods for monitoring influenza virus replication in vitro.J Virol Methods. 1998; 74: 57-66Crossref PubMed Scopus (176) Google Scholar Archival paraffin-embedded human tissue sections were obtained from the Department of Pathology, Erasmus MC. Archival paraffin-embedded animal tissue sections were obtained from the Department of Virology, Erasmus MC, or from the Department of Pathobiology, Faculty of Veterinary Medicine, University of Utrecht (Utrecht, The Netherlands). All tissues selected were from individuals without histological lesions or evidence of respiratory tract infection at the time of death. Three individuals per species were analyzed. Formalin-fixed paraffin-embedded tissues were deparaffinized with xylene and hydrated using graded alcohols. FITC-labeled influenza viruses were incubated overnight at 4°C at a titer of 50 to 100 hemagglutinating units/50 μl. For the visualization by light miscroscopy, FITC label was detected with a peroxidase-labeled rabbit-anti-FITC (Dako, Glostrup, Denmark). The signal was amplified with a tyramide signal amplification system (Perkin Elmer, Boston, MA) according to the instructions of the manufacturer. Peroxidase was revealed with 3-amino-9-ethyl-carbazole (Sigma), resulting in a bright red precipitate. Tissues were counterstained with hematoxylin and embedded in glycerol-gelatin (Merck, Darmstadt, Germany). Attachment of influenza virus to tissues was visible as granular to diffuse red staining on the apical surface of epithelial cells. Cytoplasmic staining in epithelial cells and staining of other cell types was seen occasionally. For each tissue tested, in each run, an omission control was included to check for nonspecific amplification. To validate the method, we incubated labeled H5N1 virus and H3N2 virus with human trachea and mallard duck intestine. The pattern of attachment of both viruses to these tissues was as expected. H5N1 virus, which has retained a preference for α-2,3-SA,15Gambaryan A Tuzikov A Pazynina G Bovin N Balish A Klimov A Evolution of the receptor binding phenotype of influenza A (H5) viruses.Virology. 2006; 344: 432-438Crossref PubMed Scopus (165) Google Scholar bound abundantly to duck intestinal epithelium, which expresses mainly α-2,3-SA,8Ito T Couceiro JN Kelm S Baum LG Krauss S Castrucci MR Donatelli I Kida H Paulson JC Webster RG Kawaoka Y Molecular basis for the generation in pigs of influenza A viruses with pandemic potential.J Virol. 1998; 72: 7367-7373Crossref PubMed Google Scholar and bound rarely to human tracheal epithelium, which expresses mainly α-2,6-SA.7Couceiro JN Paulson JC Baum LG Influenza virus strains selectively recognize sialyloligosaccharides on human respiratory epithelium: the role of the host cell in selection of hemagglutinin receptor specificity.Virus Res. 1993; 29: 155-165Crossref PubMed Scopus (367) Google Scholar In contrast, H3N2 virus, which has a preference for α-2,6-SA, bound abundantly to human tracheal epithelium and bound poorly to duck intestinal epithelium. H5N1 virus attachment in human lung was detected as described above, but peroxidase was not yet revealed. Type II pneumocytes were detected by incubation with a monoclonal mouse anti-human surfactant apoprotein A (PSP-A) antibody (Dako) for 1 hour at room temperature, followed by incubation with an alkaline phosphatase-labeled goat anti-mouse IgG2b (Southern Biotechnology Associates, Inc., Birmingham, AL) for 1 hour at room temperature. Peroxidase was revealed as described above, and alkaline phosphatase was revealed with 5-bromo-4-chloro-3-indolyl phosphate/nitro blue tetrazolium substrate system (Dako), resulting in a dark blue precipitate. Sections were not counterstained. Omission of H5N1 virus or an IgG2b isotype control (instead of mouse anti-human PSP-A) was included as a negative control in each run. The human influenza A viruses H3N2 and H1N1 had a similar PVA to the human respiratory tract (Table 1; Figure 1). Virus attachment in the trachea and bronchi was more abundant than in the bronchioles. In trachea, bronchi, and bronchioles, virus predominantly attached to the surface of ciliated epithelial cells, occasionally attached to goblet cells, and rarely attached to bronchiolar nonciliated cuboidal cells. In alveoli, virus attached more to type I than to type II pneumocytes and very rarely to alveolar macrophages.Table 1Virus Attachment in Respiratory Tract of Humans and Five Animal SpeciesTracheaBronchusBronchioleAlveolusVirusSpeciesScorePredominant cell typeScorePredominant cell typeScorePredominant cell typeScorePredominant cell typeH3N2Human++†Goblet cells occasionally positive.cil++†Goblet cells occasionally positive.cil+cil+Type IMouse−−−±Ferret++*Submucosal glands positive.cil±*Submucosal glands positive.±+Type IMacaque−−−−Pig++†Goblet cells occasionally positive.cil++†Goblet cells occasionally positive.cil++cil+Type ICat−−−−H1N1Human++†Goblet cells occasionally positive.cil++†Goblet cells occasionally positive.cil+cil+Type IMouse−−−±Ferret+*Submucosal glands positive.cil+*Submucosal glands positive.cil±+Type IMacaque−−−−Pig++†Goblet cells occasionally positive.cil++†Goblet cells occasionally positive.cil++cil+Type ICat−−−−H5N1Human−*Submucosal glands positive.+*Submucosal glands positive.cil+non-cil+‡Alveolar macrophages positive.Type IIMouse++Both+non-cil+non-cil+Type IIFerret−−±+Type IIMacaque−±±+Type IPig−−−+Type IICat−−+non-cil+‡Alveolar macrophages positive.Type IIH5N9Human−*Submucosal glands positive.+*Submucosal glands positive.cil+non-cil+‡Alveolar macrophages positive.Type IIMouse++Both++non-cil++non-cil+Type IIFerret−−−±Macaque−−±±Pig−−±±Cat−−±+‡Alveolar macrophages positive.Type IIH6N1Human±*Submucosal glands positive.±*Submucosal glands positive.+non-cil+‡Alveolar macrophages positive.Type IIMouse++Both+non-cil+non-cil+Ferret−−−±Type IIMacaque−±±+Type IPig−−±+Type IICat−−±non-cil+‡Alveolar macrophages positive.Type IIThe mean abundance of cells to which virus attached was scored as follows: −, no attachment; ±, attachment to rare or few cells; +, attachment to a moderate number of cells; ++, attachment to many cells. Where possible, the predominant cell type to which virus attached is indicated: ciliated cells (cil), nonciliated cuboidal cell (non-cil), type I pneumocytes (type I), or type II pneumocytes (type II).* Submucosal glands positive.† Goblet cells occasionally positive.‡ Alveolar macrophages positive. Open table in a new tab The mean abundance of cells to which virus attached was scored as follows: −, no attachment; ±, attachment to rare or few cells; +, attachment to a moderate number of cells; ++, attachment to many cells. Where possible, the predominant cell type to which virus attached is indicated: ciliated cells (cil), nonciliated cuboidal cell (non-cil), type I pneumocytes (type I), or type II pneumocytes (type II). The avian influenza viruses H5N9 and H6N1 showed a similar PVA to tissues of the human respiratory tract that also resembled the PVA of H5N1 virus (Table 1; Figure 1). Attachment to the apical cell membrane was usually granular for H6N1 virus and more diffuse for H5N9 virus. In contrast to the human influenza A viruses, attachment of the avian influenza viruses was rare in the trachea and increased progressively toward the bronchioles. The avian influenza viruses also preferentially attached to different cell types than the human influenza A viruses: to acinar cells of the tracheal and bronchial submucosal glands (Figure 2) and to mucus at these sites, to nonciliated cuboidal cells in the bronchioles, and to type II pneumocytes and alveolar macrophages in the alveoli. To confirm that avian influenza viruses attached to type II pneumocytes, human lung tissue was double stained with H5N1 virus and PSP-A, which is a surfactant produced specifically by type II pneumocytes (Figure 3).Figure 3Confirmation of H5N1 virus attachment to type II pneumocytes in human alveoli by staining for human surfactant apoprotein A (PSP-A). H5N1 virus attachment is visible as red staining on the apical cell surface, whereas PSP-A expression, characteristic for type II pneumocytes, is visible as diffuse dark blue staining in the cytoplasm.View Large Image Figure ViewerDownload Hi-res image Download (PPT) The PVA of human influenza A viruses to respiratory tract tissues of some experimental animal species resembled that of humans more than others (Table 1; Figure 2, Figure 4, Figure 5). The PVA of ferrets and pigs resembled that of humans most closely, because they were the only two species in which the viruses attached to the surface of ciliated epithelial cells in the airways and to type I pneumocytes in alveoli. In ferrets, there also was virus attachment to submucosal glands and mucus (Figure 2). In mice, there was no attachment to trachea, bronchi, or bronchioles and only occasional attachment to alveolar epithelial cells of indeterminate type. In macaques and cats, virus attachment was not observed or was rarely observed at any level of the respiratory tract.Figure 5Attachment of H3N2 virus and H5N1 virus to pig trachea and alveoli.View Large Image Figure ViewerDownload Hi-res image Download (PPT) The PVA of H5N9 and H6N1 viruses to respiratory tract tissues resembled that of H5N1 virus in each of the five animal species (Table 1). As for human influenza A viruses, the PVA of avian influenza viruses to respiratory tract tissues of some animal species resembled that of humans more than others. The PVA of cats, ferrets, and pigs (Figure 5) resembled that of humans most closely, with rare virus attachment in the trachea and bronchi, rare to occasional attachment to nonciliated cuboidal cells in the bronchioles, and predominant attachment to type II pneumocytes in the alveoli. In cats, there also was virus attachment to alveolar macrophages, as in humans. The PVA of macaques also resembled that of humans, except that virus attachment was predominantly to type I instead of type II pneumocytes, although attachment to type II pneumocytes was also observed. In mice, virus attachment was most abundant in the trachea and became progressively weaker toward the alveoli, which was opposite to the PVA in humans. However, virus attachment to submucosal glands in mice mirrored that in humans (Figure 2). This study shows for the first time the PVA of human influenza A viruses in the human respiratory tract, from the trachea down to the alveoli. This information is important to understand better the pathogenesis of influenza pneumonia. The PVA of these viruses differs markedly from that of an H5N1 strain.16van Riel D Munster VJ de Wit E Rimmelzwaan GF Fouchier RA Osterhaus AD Kuiken T H5N1 virus attachment to lower respiratory tract.Science. 2006; 312: 399Crossref PubMed Scopus (555) Google Scholar These differences may explain, at least in part, the contrasts in localization and severity of respiratory disease between these virus infections in humans. The common presentation of human influenza A virus infection is tracheobronchitis, which fits with the abundant attachment of these viruses to tracheal and bronchial epithelium (Table 1; Figure 1). Although rarely, human influenza A viruses can cause severe pneumonia. This fits with the ability of human influenza A viruses to attach to the human LRT. The difference in disease outcome between human influenza A viruses and H5N1 virus infection, where the primary lesion is severe pneumonia,24Beigel JH Farrar J Han AM Hayden FG Hyer R de Jong MD Lochindarat S Nguyen TK Nguyen TH Hien TT Nicoll A Touch S Yuen KY Avian influenza A (H5N1) infection in humans.N Engl J Med. 2005; 353: 1374-1385Crossref PubMed Scopus (1168) Google Scholar fits with differences in virus attachment in the alveoli. Human influenza A viruses attached primarily to type I pneumocytes (Table 1; Figure 1), which are less numerous than type II pneumocytes (40 versus 60% of alveolar epithelial cells)25Castranova V Rabovsky J Tucker JH Miles PR The alveolar type II epithelial cell: a multifunctional pneumocyte.Toxicol Appl Pharmacol. 1988; 93: 472-483Crossref PubMed Scopus (126) Google Scholar and have low metabolic activity.26Stevens A Lowe J Histology. MO, Mosby, St. Louis1993: 124-142Google Scholar We therefore speculate that if attachment leads to infection, virus production by type I pneumocytes is relatively low and can more easily be controlled by the host innate immune response. In addition, a large