Differential replication of avian influenza H9N2 viruses in human alveolar epithelial A549 cells

Abstract

Avian influenza virus H9N2 isolates cause a mild influenza-like illness in humans. However, the pathogenesis of the H9N2 subtypes in human remains to be investigated. Using a human alveolar epithelial cell line A549 as host, we found that A/Quail/Hong Kong/G1/97 (H9N2/G1), which shares 6 viral "internal genes" with the lethal A/Hong Kong/156/97 (H5N1/97) virus, replicates efficiently whereas other H9N2 viruses, A/Duck/Hong Kong/Y280/97 (H9N2/Y280) and A/Chicken/Hong Kong/G9/97 (H9N2/G9), replicate poorly. Interestingly, we found that there is a difference in the translation of viral protein but not in the infectivity or transcription of viral genes of these H9N2 viruses in the infected cells. This difference may possibly be explained by H9N2/G1 being more efficient on viral protein production in specific cell types. These findings suggest that the H9N2/G1 virus like its counterpart H5N1/97 may be better adapted to the human host and replicates efficiently in human alveolar epithelial cells.

Figure 1

Replication of H9N2 subtypes in different cell types and chicken embryonated eggs. (A) A549, (B) LMH and (D) MDCK cells were infected by H1N1, H9N2/G1, H9N2/Y280 or H9N2/G9 at an m.o.i. of 0.01, and (C) chicken eggs were infected with the viruses at TCID₅₀ of 2 × 10⁶. Viral titres of the infected cells were measured at 6 h, 24 h and 48 h post-infection (pi). Each point represents the mean of viral titers for three independent experiments and the titres were statistically analyzed by the two tailed, paired t-test; *:p < 0.05; Mock, uninfected cells.

Figure 2

Expression of viral nucleoprotein and matrix protein in A549 cells. A549 cells were infected with H1N1, H9N2/G1, H9N2/Y280 or H9N2/G9 at an m.o.i. of 2 and cultured with DMEM without adding N-Tosyl-L-phenylalaninechloromethyl ketone-treated trypsin. (A) The infected cells were stained with FITC-conjugated monoclonal antibody specific for influenza nucleoproteins at 24 h post-infection (pi) and were visualized by fluorescence microscopy. Mock-treated cells (at the lower right corner) were counter-stained with DAPI. (B) A549 cells were infected with H9N2/G1, H9N2/Y280 or H9N2/G9 at an m.o.i. of 2 or mock treated. Total proteins were harvested at 8 h or 24 h post-infection (pi) and M1 proteins of influenza A were examined by Western analysis. (C) A549 cells were infected with influenza H1N1 or H9N2/G1 viruses at an m.o.i. of 2 or mock treated. Total proteins were harvested at 8 h or 24 h pi and M1 proteins of influenza A were examined by Western analysis. Equal loading of protein samples was determined with anti-actin antibodies. The densities of the protein bands were determined by using Bio-Rad Quantity One imaging software. The values in the parenthesis are relative M1 protein intensities compared with those of the corresponding actin. Mock, uninfected cells; M1, matrix protein; UT, untreated; DAPI, 4'-6-Diamidino-2-phenylindole.

Figure 3

Quantitative analysis of the RNA levels of M1 and PA in A549 cells. (A) A549 cells were infected with H9N2/G1, H9N2/Y280 or H9N2/G9 at an m.o.i. of 2 or mock treated. Total RNA samples of the virus-infected cells were collected at 3 h, 6 h and 24 h post-infection (pi) and assayed by quantitative RT-PCR. The mRNA levels of (A) M1 and (B) PA, normalized to β-actin gene, were compared to the uninfected samples. Mock, uninfected cells; M1, matrix protein; PA, acid polymerase protein.