ABSTRACT
Human alveolar epithelial cells (AECs) and alveolar macrophages (AMs) are the first lines of lung defense. Here, we report that AECs are the direct targets for H1N1 viruses that have circulated since the 2009 pandemic (H1N1pdm09). AMs are less susceptible to H1N1pdm09 virus, but they produce significantly more inflammatory cytokines than AECs from the same donor. AECs form an intact epithelial barrier that is destroyed by H1N1pdm09 infection. However, there is significant variation in the cellular permissiveness to H1N1pdm09 infection among different donors. AECs from obese donors appear to be more susceptible to H1N1pdm09 infection, whereas gender, smoking history, and age do not appear to affect AEC susceptibility. There is also a difference in response to different strains of H1N1pdm09 viruses. Compared to A/California04/09 (CA04), A/New York/1682/09 (NY1682) is more infectious and causes more epithelial barrier injury, although it stimulates less cytokine production. We further determined that a single amino acid residue substitution in NY1682 hemagglutinin is responsible for the difference in infectivity. In conclusion, this is the first study of host susceptibility of human lung primary cells and the integrity of the alveolar epithelial barrier to influenza. Further elucidation of the mechanism of increased susceptibility of AECs from obese subjects may facilitate the development of novel protection strategies against influenza virus infection.
IMPORTANCE Disease susceptibility of influenza is determined by host and viral factors. Human alveolar epithelial cells (AECs) form the key line of lung defenses against pathogens. Using primary AECs from different donors, we provided cellular level evidence that obesity might be a risk factor for increased susceptibility to influenza. We also compared the infections of two closely related 2009 pandemic H1N1 strains in AECs from the same donor and identified a key viral factor that affected host susceptibility, the dominance of which may be correlated with disease epidemiology. In addition, primary human AECs can serve as a convenient and powerful model to investigate the mechanism of influenza-induced lung injury and determine the effect of genetic and epigenetic factors on host susceptibility to pandemic influenza virus infection.
INTRODUCTION
Pandemic H1N1 (H1N1pdm09) virus spread into 30 countries during the first few weeks of 2009 and caused a worldwide pandemic (1). Different from seasonal influenza viruses, H1N1pdm09 virus commonly targets the lower respiratory tract, specifically the distal lung cells (2–4). Human alveolar epithelial cells (AECs) are a key part of lung defenses against respiratory pathogens and the primary targets for pandemic and severe influenza (3–7). Alveolar macrophages (AMs) are scavengers in the lung and known to be important in the limitation of influenza virus, including H1N1pdm09, in animal studies (8, 9). In recent studies of patients that died with H1N1pdm09 pneumonia, both AECs and AMs were found to be positive for viral antigens. However, it is not clear whether these two types of cells are equally susceptible to H1N1pdm09 infection.
Influenza virus is one of the leading causes of death in the United States, but there is a significant variation in disease outcomes among virus-infected individuals. Some patients do not develop symptoms, some recover after a few days of body aches and fever, and some individuals develop severe disease (e.g., respiratory distress pneumonia) and are admitted to intensive care units. It is important to investigate the susceptibility to influenza virus and identify high-risk populations for early prevention and treatment. However, it is difficult to study influenza virus susceptibility in patients because of the known confounding factors of prior immunization, unclear exposure history, and the existence of coinfections. Therefore, investigating influenza virus susceptibility in targeted cells carrying different genetic and epigenetic traits may provide an advantage to gain insights into the complicated mechanism of differential susceptibility to influenza virus infection and facilitate individualized prevention and treatment strategies.
In this study, we compared the infection of human AECs and AMs from the same donor and clarified that AECs, instead of AMs, are the primary targets for H1N1pdm09 viruses. However, AMs release higher levels of cytokines than do AECs after viral infection, despite the lower susceptibility to H1N1pdm09 viruses. We focused on human AECs and examined the cellular susceptibility to influenza virus infection. Interestingly, we noticed that AECs from obese individuals were more susceptible to H1N1pdm09 infection, whereas, gender, smoking, and age did not affect AEC susceptibility to influenza. Additionally, we found that two strains of H1N1pdm09 showed different infectivities. A/New York/1682/2009 (NY1682) caused a higher level of infection and induced more significant epithelial barrier injury in human AECs than did A/California/04/09 (CA04). A single Ala214Thr substitution in NY1682 hemagglutinin (HA) abolished the difference.
MATERIALS AND METHODS
Viruses.The 2009 pandemic H1N1 (H1N1pdm09) virus CA04 was kindly provided by NIAID BEI Bioresources. NY1682 was isolated from a patient in New York in April 2009. A/California/07/09 (CA07), isolated in California in July 2009, was provided by Ted Ross at the University of Georgia. A/Puerto Rico/8/1934 (PR8), a laboratory-adapted H1N1 virus, was originally provided by Kevan Hartshorn from Boston University. NY1682 viruses carrying different HA mutations were constructed using reverse genetics as described previously (10, 11). All viruses were propagated and titrated in MDCK cells by standard plaque assay.
Isolation and culture of human AECs and AMs.Human alveolar type II cells were isolated from de-identified healthy lung donors using a modified protocol as previously described (6, 12, 13). The Committee for the Protection of Human Subjects at National Jewish Health and the Committee for Oversight of Research and Clinical Training Involving Decedents (CORID) from University of Pittsburgh approved this study. Cells isolated from 40 donors with ages ranging from 2 to 80 years were used in this study. Briefly, the donated lung was perfused, subjected to lavage, and digested as described previously (13). The digested lung was then minced and the lung cell suspension was filtered through series of filters. After lysing of the red blood cells, AECs were purified through a discontinuous gradient and subsequent positive selection with EpCAM microbeads (Miltenyi Biotec Inc., San Diego, CA). The purity of isolated type II cells was around 90%, as evaluated by staining the cells with pro-SP-C and ATII-280, an ATII-specific marker (14), by flow cytometry (data not shown). The purity was further increased during the primary cell culture. For culture of human AECs, the isolated ATII cells were plated in Dulbecoo modified Eagle medium (DMEM) with 10% fetal bovine serum (FBS) on rat tail collagen-coated Transwell inserts (BD Biosciences, San Jose, CA), and after 2 days of adherence, media were switched to DMEM with 5% FBS. AECs were cultured for another 4 days before influenza virus infection. AMs from the same donor were isolated and cultured as described previously, and the purity of AMs was confirmed by CD68 staining (6).
Measurement of alveolar epithelial barrier function.Barrier function of cultured human AECs was evaluated by measuring transepithelial electrical resistance (TEER) and paracellular permeability. TEER was measured through the cultured monolayer using the EVOM2 (World Precision Instrument, Sarasota, FL) and represented as ohms per square centimeter. For paracellular permeability measurement, fluorescence-labeled dextran (Sigma-Aldrich, St. Louis, MO) was added to the apical compartment of the culture, and cell supernatant from the basal compartment was collected 3 h later for reading of the fluorescence intensity using a fluorescent plate reader (Bio-Tek, Vinooski, VT). Higher TEER and lower permeability indicate better barrier function.
Infection of primary human AECs and AMs.Before infection, human AECs and AMs were washed once and inoculated with H1N1 viruses at a multiplicity of infection (MOI) of 1 for 1 h at 37°C. After inoculation, the cells were washed and fresh media were added to the culture. At the desired time points postinoculation, samples of the culture supernatant from infected cells were collected for virus titration and cytokine analysis, and cells were fixed for detection of influenza virus antigen.
Detection of influenza virus infection by IFA and flow cytometry.For immunofluorescence assay (IFA), the cell monolayer was fixed with methanol and immunostained with mouse antinucleoprotein (anti-NP) (EMD Millipore) or rabbit antihemagglutinin (anti-HA) (kindly provided by NIAID BEI Resources) as previously described (6). Nuclei were counterstained with 4′,6-diamidino-2-phenylindole (DAPI; Vector Laboratories Inc., Burlingame, CA). Pictures from four randomly selected areas were taken for calculating the percentage of infection. For flow cytometry, single-cell suspensions were immunostained with fluorescein isothiocyanate (FITC)-labeled mouse anti-NP (Abcam, Cambridge, MA) and analyzed on a BD LSR II flow cytometer (BD Biosciences, San Jose, CA). The percentage of positive-stained cells was calculated using FACSDiva (BD Biosciences).
Measurement of cytokine release.Culture supernatant was collected after 24 h of viral infection and used for detection of cytokine release by a multiplex assay, following the manufacturer's instructions (Bio-Rad, Hercules, CA). In addition, DuoSet enzyme-linked immunosorbent assay (ELISA) for human IFN-lambda 1 (interleukin 29 [IL-29]) was purchased from R&D Systems (Minneapolis, MN).
Statistical analysis.Statistical analysis was conducted in GraphPad Prism version 6.0 (GraphPad Software, San Diego, CA). Pairwise comparisons were tested for significance using the Wilcoxon matched-pairs test. Nonpaired comparisons were performed using the Mann-Whitney test. Multiple comparisons were performed by one-way analysis of variance followed by the appropriate multiple-comparison test. Differences were considered significant at a P value of <0.05.
RESULTS
Pandemic H1N1 virus primarily targets human AECs and causes epithelial barrier dysfunction.To clarify whether human AEC and AMs are equally susceptible to H1N1 virus infection, we infected primary AECs and AMs from the same donor with two H1N1pdm09 viruses, CA04 and NY1682, as well as a control H1N1 virus, PR8. At 24 h postinfection (hpi), we harvested the cells and examined viral infection by staining the viral antigen NP. As shown in a representative infection in Fig. 1A, human AECs were sufficiently infected with all three H1N1 viruses. Human AMs were permissive for PR8 infection, whereas they were poorly susceptible to the same amount of H1N1pdm09 viruses. In addition, the same concentration of H1N1pdm09 virus caused more infection of human AECs than AMs from the same donor (Fig. 1B). These results demonstrate that human AMs are relatively resistant to H1N1pdm09 viruses and that human AECs are the primary targets for H1N1pdm09 infections.
Infection of H1N1 viruses with primary human AECs and AMs from the same donor. (A and B) Infection of AECs and AMs. AECs and AMs from the same donor were cultured on glass coverslips and infected with H1N1 viruses at an MOI of 1. At 24 h postinoculation, cells were fixed with methanol and doubly stained for viral nucleoprotein (NP) (green) and cell markers (red): cytokeratin for AEC and CD68 for AM. Nuclei were counterstained with DAPI (blue). (A) Representative images of infected cells; (B) quantification of infections. Significant differences between AEC and AM groups are indicated as follows: *, P < 0.05, and **, P < 0.01; n = 3. (C) H1N1 virus causes barrier injury in AECs. AECs were cultured on the inserts and infected with H1N1pdm09 virus NY1682 at an MOI of 1. TEER was measured at different time points. Data represent results from one of four donors. Paracellular permeability (PP) was measured at 72 h postinfection; n = 4.
Because AECs are the physical barrier of the lung, we further examined the effect of viral infection on the barrier function of human AECs by comparing transepithelial electrical resistance (TEER) and paracellular permeability across the epithelial monolayer between infected and noninfected conditions. Healthy human AECs form an intact barrier with TEER over 1,000 Ω per cm2. Along with the viral replication, NY1682 virus infection time dependently reduced the TEER and significantly increased the paracellular permeability (Fig. 1C), indicating an impairment of the barrier function. Similar results were observed with PR8 and another H1N1pdm09 virus, CA04.
Human AMs are less permissive for H1N1 pdm09 viruses but produce more cytokines and chemokines than AECs from the same donors.Figure 1A indicated that human AMs from the same donor were resistant to H1N1 viruses compared to AECs. We were curious whether these cells were still able to stimulate a cytokine response. We compared the cytokine responses stimulated by different strains of H1N1pdm09 viruses, including CA04, NY1682, and CA07, which has been the human vaccine strain since 2009, between AECs and AMs from additional three donors. Data shown in Fig. 2 indicate that both AECs and AMs release cytokines in response to viral infections, although under certain conditions, the difference did not reach statistical significance because of the limited number of donors tested. Surprisingly, human AMs released more of the proinflammatory cytokines interleukin 1β (IL-1β), tumor necrosis factor alpha (TNF-α), IL-6, and gamma interferon (IFN-γ) than did AECs despite the lower level of infection with H1N1pdm09 viruses. Consistent to the previous finding with PR8 infection (6), AMs were not able to secrete significant amount of IL-29 (IFN-λ1) in response to H1N1pdm09 infection. There was no significant difference in stimulation of cytokine responses in human AMs by different strains of H1N1pdm09 viruses (Fig. 2). CA04 induced greater cytokine and chemokine responses than NY1682 and CA07 in AECs, especially in IL-29 induction; no differences were detected between NY1682 and CA07 in initiating a cytokine response in either cell type. There was no increase of monocyte chemoattractant protein 1 (MCP-1) or IL-8 in H1N1pdm09 virus-infected AECs, but these cells released a significant amount of IP-10 and RANTES, indicating that they may play a role in chemoattraction of lymphocytes and monocytes to the infected site.
Cytokine and chemokine responses in human AECs and AMs infected with H1N1 viruses. Human AECs and AMs from the same donors were infected with same amount of H1N1 viruses PR8, CA04, NY1682, and CA07 (MOI = 1), and culture supernatant was collected for detection of the cytokine and chemokine responses by multiplex assay and ELISA by following the manufacturer's instructions. Significant differences from the mock control are indicated as follows: *, P < 0.05; **, P < 0.01; and ***, P < 0.001; n = 3.
Human AECs from different donors vary in their susceptibilities to H1Npdm09 viruses.To test whether the differences in susceptibility to influenza virus observed in clinical practice and animal studies could be shown at the cellular level, we inoculated primary AECs isolated from different donors with H1Npdm09 viruses at a multiplicity of infection (MOI) of 1 and evaluated cellular susceptibility to the infections using immunofluorescence assay to detect and quantify the infected cells. The same concentration of virus caused significantly different infection rates in cells from different donors (Fig. 3A). It was noted that AECs from one donor, a 74-year-old female nonsmoker with a body mass index (BMI) of 27.9, was highly permissive to both CA04 and NY1682 for unknown reasons, and almost 100% of cells from this donor were infected by each virus. Cells from this donor were also highly susceptible to other H1N1 viruses, including CA07 and PR8 (data not shown). To further investigate what factors contribute to different susceptibilities to H1N1pdm09 virus, we excluded this specific donor and other donors with missing information for age, gender, smoking history, and BMI and then compared the percentages of CA04 infection among different groups. It should also be noted that child donors were only included within the age comparison; their ages were 2, 2, and 7 years. The rest of the donors were adults. Compared to nonobese donors (BMI range between 18 and 30), AECs from obese donors (BMI ≥ 30) displayed a much higher infection rate with CA04, with a level of infection of 40.0% versus 26.9%, respectively (P = 0.042) (Fig. 3B). These results suggest that AECs from obese subjects are more susceptible to H1N1pdm09 infection than AECs from nonobese subjects. There was no difference in AEC susceptibility between males and females or smokers and nonsmokers. Also, no statistically significant difference was detected in susceptibility among different age groups, although AECs from children seemed to have a higher percentage of infection than adult (18 to 65 years) or elderly (≥65 years) groups (42.6% in children versus 29.7% in adult and 29.1% in elderly donors) (Fig. 3B). Similar trends with NY1682 infections were also observed (data not shown).
Variation in human AEC susceptibility to H1N1pdm09 viruses. AECs from the same donor were infected with CA04 or NY1682 at an MOI of 1. At 24 hpi, cells were fixed with methanol and immunostained with anti-NP. Data represent percentages of positive infected cells for each virus. Each spot indicates one donor. (A) Differential susceptibilities to H1N1pdm09 viruses; (B) effects of age, gender, smoking, and obesity on AEC susceptibility to CA04. BMI, body mass index; NS, nonsmoker; Ex-S, ex-smoker; S, smoker.
NY1682 has a higher infectivity and causes greater epithelial injury than CA04.Data in Fig. 1 and 3 indicate a difference in infectivity between the H1N1pdm09 viruses NY1682 and CA04. NY1682 had a higher infectivity than CA04. To explore whether NY1682 enters cells faster than CA04, we incubated NY1682 or CA04 with MDCK cells, the cells used for propagation of influenza virus, at 4°C for 2 h before moving cells to the incubator at 37°C and then examined the viral infection at 1, 4, and 6 hpi by immunostaining for viral NP with infected culture. As early as 4 hpi, there were more NP-stained cells in the NY1682-infected cultures (Fig. 4A). By 6 hpi, there were many more positive-stained cells in both infections, but significantly more cells were infected by NY1682 than by CA04. We carried out similar experiments with AECs but stained the cells with viral hemagglutinin (HA) instead of NP and observed the same result at 9 hpi (Fig. 4A). These results suggest that NY1682 enters cells faster than CA04. We then compared the abilities of the two H1N1pdm09 viruses to replicate and damage the epithelial barrier created by cultured AECs. Using the same amount of infectious virus, NY1682 replicated to almost a 10-fold-higher value than CA04 in human AECs (Fig. 4B) and induced a greater reduction in TEER (Fig. 4C) at 24 h postinfection. Interestingly, NY1682 stimulated a much smaller amount of cytokines, such as IP-10, RANTES, and IL-29 (data not shown), which was consistent with the results shown for AECs in Fig. 2. These results demonstrate that NY1682 has a higher infectivity than CA04, which results in more robust replication of the NY1682 in host cells.
Comparison of viral entry, replication, and effect on AEC barrier between NY1682 and CA04 infections. (A) Viral entry in MDCK cells and human AECs. MDCK cells and human AECs were inoculated with the same amount of CA04 or NY1682 at 4°C for 2 h and washed with DMEM before incubation at 37°C. Cells were harvested at designated time points for immunostaining with anti-influenza virus NP (green) for MDCK cells or antihemagglutinin (anti-HA) (green) for AECs; nuclei were counterstained with DAPI (blue). The experiments were repeated twice. (B and C) Viral replication and effect on human AEC barrier. AECs were infected with and without CA04 or NY1682 (MOI = 1) or treated with recombinant human IL-1β (10 ng/ml) as a positive control. (B) Culture supernatant was collected for evaluation of infectious virus release by plaque assay as described previously (6). **, P < 0.01 between two groups. (C) TEER was measured at 24 h posttreatment. Significant differences from the mock control are indicated as follows: *, P < 0.05; **, P < 0.01; ***, P < 0.001; n = 4. # indicates a significant difference from the CA04 group.
Thr214Ala substitution in HA enhances the H1N1 virus infectivity.Influenza virus HA protein is the main determinant of virus entry and receptor binding. To determine whether the difference in infectivities is caused by the difference in HA sequences, we sequenced the HA genes from these two virus stocks. Alignment of HA proteins showed that the CA04 has 5 amino acid residue substitutions compared with the sequence from NY1682: S100P, A214T, A232V, D293N, and V338I. To test whether these differences contribute to the differential infectivies, we first created a new NY1682 strain carrying HA-S100P/D293N/V338I using reverse genetics. This virus has the same sequence as the CA07 strain; therefore, we designated this virus NY1682-CA07HA. We then infected human AECs with CA04, NY1682, and NY1682-CA07HA viruses and quantified the percentages of infected cells by flow cytometry and also measured IL-29 secretion from infected cells. NY1682-CA07HA behaved similarly to wild-type (wt) NY1682 virus in its abilities to infect cells (Fig. 5A) and stimulated IL-29 secretion (Fig. 5B). These results suggest that the amino acid differences at positions 100, 293, and 338 are not important for the differential infectivities between NY1682 and CA04; instead positions 214 and 238 require more investigation.
NY1682-HA-S100P/D293N/V338I (rNY1682-CA07HA) behaves similarly to NY1682 in the ability to infect human AECs. Human AECs were infected with recombinant NY1682 carrying HA-S100P/D293N/V338I (CA07HA), wild-type NY1682, and CA04. Percentages of infection were compared by flow cytometry, and induction of IL-29 release was measured by ELISA. (A) Infectivity; (B) IL-29 release. Data represent results from one of three donors.
Therefore, we reconstructed two more NY1682 viruses by individually replacing Ala with Thr at position 214 (NY1682-A214T) and Ala with Val at position 238 (NY1682-A238V) separately. We also introduced another single Ser-to-Pro substitution at position 100 (P100S) into NY1682 as a control virus (NY1682-S100P). We then infected AECs with these viruses and compared the percentages of infection by flow cytometry. As shown in Fig. 6, there was no difference in the percentage of infection between wt NY1682 and NY1682 carrying S100P/D293N/V338I (CA07HA), A238V, or S100P. In contrast, the A214T substitution reduced the infectivity of NY1682 significantly, to a level similar to that of CA04 (P < 0.01). These data demonstrate that the T214A substitution in HA confers the difference in H1N1pdm09 infectivity in human AECs.
Replacement of Ala with Thr at position 214 of NY1682-HA reduces the infectivity of NY1682. AECs infected with wild-type NY1682, CA04, or recombinant NY1682 viruses carrying HA-S100P/D293N/V338I (CA07HA), -A214T, -Al238V, or -S100P were fixed with 4% paraformaldehyde and immunostained with FITC-labeled anti-influenza virus NP. Positive infected cells were detected by flow cytometry, and percentages of infection were compared among different viruses; n = 5. **, P < 0.01; ***, P < 0.001.
DISCUSSION
We previously developed a primary culture system to study human AECs (13) and reported their innate immune response to influenza virus infections (6, 12, 15–17). In the present study, we found that cultured human AECs form an intact barrier with a resistance over 1,000 Ω per cm2 and that influenza virus infection results in a time-dependent drop in TEER and significantly increased paracellular permeability across the AEC monolayer, indicating an injury to the epithelial barrier. This culture system provides an ideal model to study how influenza viruses disrupt the lung epithelial barrier and cause pulmonary edema, which is often seen in severe flu cases, including avian and pandemic flu (3, 4, 18, 19).
H1N1pdm09 virus targets the lower respiratory tract (20, 21). Although viral antigens have been detected in both AECs and AMs from the lungs, there are controversial publications about whether human AMs are directly targeted by H1N1pdm09. Gill and colleagues reported that influenza virus antigen was mainly detected in AMs (22). This finding is in contrast to results from another H1N1pdm09 study reported by Shieh et al., in which viral antigen was predominantly found in AECs (3). The study by Gill et al. used a New York strain virus, while the study by Shieh et al. used a virus originating from Atlanta, GA. Although there is no detailed description about the two strains used in these two studies, the discrepancy observed in the fatal H1N1pdm09 cases (3, 22) suggests that different strains of H1N1pdm09 virus may differ in cell tropism. Actually, several studies have shown that different strains targeted different organs and that variation in the hemagglutinins (HAs) of H1Npdm09 viruses altered the receptor binding and virulence (23–28). A recent study also showed that different strains differ in their levels of replication in the lungs and cells in bronchoalveolar lavage fluid in mice (29). However, there is no study investigating the different strains on AECs and AMs from the same donor in vitro. In the current study, we have shown that the same infective dose of H1N1pdm09 viruses CA04 and NY1682 led to significantly less infection in AMs than in AECs from the same donor (Fig. 1); this is also true with another H1N1pdm09 virus, CA07, the current human vaccine strain (data not shown). These results confirm that AMs are more resistant to H1N1pdm09 viruses (3, 21, 30) and might not be the primary target for H1N1pdm09 infections. However, human AMs produce predominantly proinflammatory cytokines IL-1β, TNF-α, IFN-γ, and IL-6 and significant amounts of the chemokines IP-10, RANTES, IL-8, and MCP-1 (Fig. 2), despite the lower susceptibility to H1N1pdm09. It is well accepted that an excessive cytokine response, or cytokine storm, is associated with influenza-associated fatal diseases (31–33). Although infected human AECs also release a lot of chemokines and cytokines, the degree is lower than for AMs. Therefore, human AMs are likely the major driving force for the inflammatory cell infiltration in vivo and contribute significantly to H1N1pdm09-associated pathology. In addition, AMs are known to be able to take up virus particles and infected apoptotic epithelial cells and therefore clear the virus in the lungs (8, 9). We speculate that there might have been a dysfunction of AMs in certain fatal 2009 H1N1 infections, which resulted in impaired viral clearance and therefore accumulation of viral antigens in AMs.
The results from the current study demonstrate a significant variation in cellular susceptibility to H1N1pdm09 infections (Fig. 3). To our knowledge, this is the first study to investigate the susceptibility of AECs from different donors to influenza virus infection. We observed an increased risk of H1N1pdm09 infections in AECs from obese donors, as the percentage of infection increased from 26.9% to 40.0% (P = 0.042) (Fig. 3). These results suggest that cells from obese individuals might be more vulnerable to H1N1pdm09 infection. Reports from epidemiological studies have shown that obesity is associated with increased mortality in patients infected with H1N1pdm09 (22, 34–37). A recent mouse study also suggested an impaired viral clearance in obese mice during influenza virus infection (38). However, the exact mechanism remains unknown. We speculate that the increased risk for influenza-associated mortality might be due to increased percentages of infected cells in the lungs, leading to a greater viral burden in obese individuals. Since over one-third of the population of the United States is obese (http://www.cdc.gov/obesity/data/adult.html), there is an urgent need to investigate why and how obesity affects host susceptibility to influenza virus infection.
We did not observe any effect of gender or smoking history on the cellular permissiveness to H1N1pdm09 infection. Age is a well-known determinant of clinical outcomes with the worse outcome for seasonal influenza in infancy and in the elderly. Children are more susceptible to both seasonal and pandemic influenza viruses. We observed that young AECs seem to be more permissive to H1N1pdm09 infection, but we had only three young donors in our study. More donors will be needed to determine whether AECs from young donors are more susceptible to H1N1pdm09 infections in the future. In addition, aging is known to increase susceptibility to many diseases, including influenza, because of the declined immune function in the elderly. Surprisingly, we did not observe a significant difference between AECs from adults aged 18 to 65 and those aged 65 to 80 (Fig. 3B); this might be due to the limited number of elderly donors.
Although there is a trend of increased susceptibility to H1N1pdm09 virus in AECs from obese donors (Fig. 3B), this difference seems unable to fully explain the significant variation observed in tested donors (Fig. 3A). Specifically, we found that almost 100% of AECs from one nonobese donor were positive for all tested H1N1 viruses. These results suggest that other host factors, such as genetic background, might affect cellular susceptibility to influenza. In fact, influenza virus infections with inbred mouse strains have demonstrated the importance of genetic factors to variation of susceptibility to influenza (39, 40). Studies with family clusters of avian influenza virus H5N1 and case studies of H1N1pdm09 patients also imply that genetic factors affect susceptibility to pandemic influenza (41–45). Considering the distinct genetic and epigenetic background of each donor and difference between human and mouse studies, investigating the effect of genetic or epigenetic factors on influenza susceptibility in well-controlled human AECs, the cells most relevant to influenza, may provide unique advantage because the response in the same type of cells will not be affected by prior influenza history, vaccination, or other environmental exposure.
Influenza has been a persistent public health problem mainly due to its ever-evolving genetic sequence. In this study, we have shown that different H1N1pdm09 strains cause differential infectivities in human AECs. NY1682 infects more AECs than CA04 and results in greater barrier injury (Fig. 1C, 3A, and 4C). In addition, different strains also differ in their abilities to stimulate cytokine responses (Fig. 2), as reported by other studies (15, 19, 29). CA07 seems to behave similarly to NY1682, and CA04 virus stimulates more proinflammatory cytokines in human AECs than NY1682 and CA07, especially IL-29 (Fig. 2 and 5), although CA04 enters cells more slowly and replicates to a much lower degree than NY1682 (Fig. 4A and B). IL-29 is the major antiviral interferon in human AECs (6, 12); the lower level of IFN-λ induced by NY1682 may give the virus a replication advantage over CA04 in human AECs (46).
It is well understood that factors from the virus, including HA, significantly contribute to the pathology of influenza virus infection (26, 46). We and others previously reported that an Asp222Gly substitution in HA from the H1N1pdm09 virus has been associated with severe or fatal disease outcome (27, 28, 47), likely because it binds to α2-3-linked sialic acid predominantly expressed by lower respiratory cells, especially human alveolar type II cells (7, 15). In addition, D225G mutation enhances pathology of H1N1pdm09 in mice (48). In the current study, we have shown that NY1682 virus enters cells much faster than CA04 and causes a higher infectivity. By reverse genetics, we were able to identify that a single Ala214Thr substitution in HA is responsible for the different infectivity observed between these two H1N1pdm09 viruses (Fig. 6). Interestingly, the ratio of HA-214A to HA-214T in H1N1pdm09 viruses increased gradually from 2009 to 2012 but then decreased dramatically in 2013 and almost to extinction in 2014 (Fig. 7), which could be correlated with the dramatic rise in deaths of young adult and children in the early 2014 flu season (http://apps.washingtonpost.com/g/page/national/influenza-hospitalizations/822/).
Proportion of H1N1pdm09 viruses that contain an alanine or threonine at HA residue 214 from 2009 to 2014. All pH1N1 HA sequences in GenBank were downloaded and grouped by the years of sample collection. For each year, the HA protein sequences were aligned and all amino acids at residue 214 were illustrated based on their relative frequencies using the WebLogo application (49).
Conclusion.We investigated host susceptibility to H1N1pdm09 infections in human AECs and AMs and showed that AECs are the primary targets for H1N1pdm09 virus and AMs are the major driving force in releasing proinflammatory cytokines. Obesity appears to be a risk factor for increased H1N1pdm09 infection in AECs. Further study is needed to determine why and how AECs from obese subjects are more susceptible to influenza virus. In addition, primary cultured human AECs will continue to provide a unique and powerful model to investigate the mechanism of influenza-induced lung injury and determine the genetic and epigenetic influence on host susceptibility to influenza viruses and variants within a lineage.
ACKNOWLEDGMENTS
We give special thanks to the International Institute for the Advancement of Medicine (IIAM) for the assistance in receiving donated lungs for this study. We thank Karen Edeen (National Jewish Health) and Daniel Mellon at (University of Pittsburgh) for assistance in lung isolation. We appreciate Kevin J. McHugh for assistance with the multiplex assay.
This work was supported by National Institutes of Health grants R03AI101953 (to J.W.), R01HL113655 (to J.W.), and U01AI082982 (to R.M.) and startup funding from the University of Pittsburgh (to J.W.).
FOOTNOTES
- Received 17 July 2015.
- Accepted 9 September 2015.
- Accepted manuscript posted online 16 September 2015.
- Address correspondence to Jieru Wang, Jieru.wang{at}chp.edu.
↵* Present address: Emily Travanty, Public Health Microbiology and Serology, Laboratory Services Division, State of Colorado; Bin Zhou, New York University, New York, New York, USA; David E. Wentworth, National Center for Immunization and Respiratory Diseases, Centers for Disease Control and Prevention, Atlanta, Georgia, USA.
Citation Travanty E, Zhou B, Zhang H, Di YP, Alcorn JF, Wentworth DE, Mason R, Wang J. 2015. Differential susceptibilities of human lung primary cells to H1N1 influenza viruses. J Virol 89:11935–11944. doi:10.1128/JVI.01792-15.
REFERENCES
- Copyright © 2015, American Society for Microbiology. All Rights Reserved.
