Variable impact of the hemagglutinin polybasic cleavage site on virulence and pathogenesis of avian influenza H7N7 virus in chickens, turkeys and ducks

Abstract

Avian influenza viruses (AIV) are classified into 16 hemagglutinin (HA; H1-H16) and 9 neuraminidase (NA; N1-N9) subtypes. All AIV are low pathogenic (LP) in birds, but subtypes H5 and H7 AIV can evolve into highly pathogenic (HP) forms. In the last two decades evolution of HPAIV H7 from LPAIV has been frequently reported. However, little is known about the pathogenesis and evolution of HP H7 from LP ancestors particularly, in non-chicken hosts. In 2015, both LP and HP H7N7 AIV were isolated from chickens in two neighbouring farms in Germany. Here, the virulence of these isogenic H7N7 LP, HP and LP virus carrying a polybasic HA cleavage site (HACS) from HP (designated LP-Poly) was studied in chickens, turkeys and different duck breeds. The LP precursor was avirulent in all birds. In contrast, all inoculated and contact chickens and turkeys died after infection with HP. HP infected Pekin and Mallard ducks remained clinically healthy, while Muscovy ducks exhibited moderate depression and excreted viruses at significantly higher amounts. The polybasic HACS increased virulence in a species-specific manner with intravenous pathogenicity indices of 3.0, 1.9 and 0.2 in chickens, turkeys and Muscovy ducks, respectively. Infection of endothelial cells was only observed in chickens. In summary, Pekin and Mallard were more resistant to HPAIV H7N7 than chickens, turkeys and Muscovy ducks. The polybasic HACS was the main determinant for virulence and endotheliotropism of HPAIV H7N7 in chickens, whereas other viral and/or host factors play an essential role in virulence and pathogenesis in turkeys and ducks.

Figure 1

In vitro characterisation of recombinant H7N7 viruses in this study. The replication efficiency in MDCKII (a) and CEK (b) cells at an MOI of 0.001 for 1, 8, 24, 48 and 72 hours post infection is illustrated. Results are shown as the mean and standard deviations of all values of triplicates of two independent runs. Cell-to-cell spread was determined by measuring the diameter of plaques induced by the indicated viruses in MDCKII cells. The size of plaques induced by HP was set as 100%. The results are expressed as mean and standard deviation. (c) The cleavability of HA0 into HA1 and HA2 subunits was studied using Western Blot after the infection of MDCKII cells at an MOI of 1 PFU per cell of indicated viruses in the presence (+) or absence (−) of trypsin. The viral proteins were detected by polyclonal chicken serum against H7N1 at a ratio 1:500 after separation in a 10% polyacrylamide gel. NC refers to the mock control (non-infected cells). (d) The original Western Blot figure is available as a Supplementary Fig. S1.

Figure 2

Clinical scoring and virus excretion in experimentally infected chickens, turkeys and Muscovy ducks. Clinical scoring after oculonasal (a) or intravenous (b) infection was calculated by dividing the sum of the arithmetic mean values of daily scores by 10 (the number of observation days). The PI and IVPI values for each virus ranged from 0 (avirulent) to 3 (highly virulent) and the results are expressed as mean and standard deviation. Virus excretion at 4 days post inoculation in oropharyngeal (c,d) and cloacal (e,f) swabs of inoculated (c and e) and contact (d and f) birds was determined by RT-qPCR against standard curves using ten-fold dilutions of HPAIV H7N7 and the mean and standard deviations were expressed as equivalent plaque forming unit pro ml (eq. PFU/mL). Asterisks indicate significant differences at p ≤ 0.05 (*) or 0.01 (**) or 0.0001 (****).

Figure 3

Distribution of influenza A antigen in different organs of inoculated chickens. Distribution of NP antigen in the parenchyma (a) and endothelium (b) of different organs of two inoculated chickens at 4 days post inoculation was detected by immunohistochemistry. The intensity and distribution of NP signals are scaled 0 (no signal), 1 (focal to oligofocal), 2 (multifocal) or 3 (confluent to diffuse). The severity of lymphatic necrosis, apoptosis and/or depletion (c) and necrotizing inflammation (d) was assessed by histopathological examination on a scale 0 (negative), 1 (low), 2 (moderate) or 3 (high). Values are shown as median and standard deviation scores of two chickens.

Figure 4

Distribution of influenza A NP antigen in different organs in chickens. Distribution of influenza A NP antigen in different organs in chickens at 4 days post inoculation with LP (a–c), LP-Poly (d–f) or HP (g–i) displaying variable level of organ tropism ranging from a minimum of none to a maximum of coalescing nucleoprotein within brain (a,d,g), lung (b,e,h) and heart (c,f,i). Arrows point to parenchymal cells with influenza A virus nucleoprotein-positive nuclei (neuroglia cells, pneumocytes or cardiomyocytes). Arrowheads point to endothelial cells with influenza A virus nucleoprotein-positive nuclei. Immunohistochemistry; avidin-biotin-peroxidase complex method; polyclonal rabbit anti-influenza A FPV/Rostock/34-virus-nucleoprotein antiserum; 3-amino-9-ethyl-carbazol (red-brown); hematoxylin counterstain (blue); Nomarski contrast; Bars = 50 µm.

Figure 5

Distribution of influenza A antigen in different organs of inoculated turkeys. Distribution of NP antigen in the parenchyma (a) of different organs of three inoculated turkeys at 4 days post inoculation was detected by immunohistochemistry. NP was not detected in the endothelium of any organ of any of the three examined turkeys. The intensity and distribution of NP signals are scaled 0 (no signal), 1 (focal to oligofocal), 2 (multifocal) or 3 (confluent to diffuse). The severity of necrotizing inflammation (b) and lymphatic necrosis, apoptosis and/or depletion (c) was assessed by histopathological examination on a scale 0 (negative), 1 (low), 2 (moderate) or 3 (high). Values are shown as median and standard deviation scores of three turkeys.

Figure 6

Distribution of influenza A NP antigen in different organs in turkeys. Distribution of influenza A NP antigen in different organs in turkeys at 4 days post inoculation with LP (a–c), LP-Poly (d–f) or HP (g–i) displaying variable level of organ tropism ranging from a minimum of none to a maximum of coalescing nucleoprotein within brain (a,d,g), lung (b,e,h) and heart (c,f,i). Arrows point to parenchymal cells with influenza A virus nucleoprotein-positive nuclei (neuroglia cells, macrophages/pneumocytes or cardiomyocytes). Immunohistochemistry; avidin-biotin-peroxidase complex method; polyclonal rabbit anti-influenza A FPV/Rostock/34-virus-nucleoprotein antiserum 61; 3-amino-9-ethyl-carbazol (red-brown); hematoxylin counterstain (blue); Nomarski contrast; Bars = 50 µm.

Figure 7

Virus excretion and seroconversion in inoculated and/or contact Pekin, Mallard and Muscovy ducks. Shown are the results of virus excretion at 4 days post inoculation in oropharyngeal (OP) and cloacal (CL) swabs using RT-qPCR expressed by equivalent Log10 PFU/mL in oculonasal (ON) inoculated Pekin, Mallard and Muscovy ducks with HPAIV H7N7. (a) Excretion of LP, LP-Poly or HP in inoculated (b) and contact (c) Muscovy ducks were also determined by RT-qPCR. All values are shown as mean and standard deviation of positive samples. Antibody titres at the end of the experiment in ducks inoculated ON or intravenously (IV) were measured by an NP antibody inhibition ELISA at the end of the experiments. Results are shown as 100- optical density (OD) reading. Lower and upper dashed lines indicate the 45–55% cut-off ratio where samples lower than 45% are negative, from 45 to 55% are questionable and over 55% are positive. (d) Statistical significance shown in asterisk indicate P values ≤ 0.05 (*), ≤ 0.01 (**), ≤ 0.001 (***) or ≤ 0.0001 (****).