Differential replication characteristic of reassortant avian influenza A viruses H5N8 clade 2.3.4.4b in Madin-Darby canine kidney cell

Abstract

Late in 2016, multiple reassortant highly pathogenic (HP) avian influenza virus (AIVs) H5N8 was detected. AIVs infect different isolated hosts with a specific viral tropism. In the current study, the whole genome of the Egyptian A/chicken/NZ/2022 was genetically characterized. The H5N8-A/Common-coot/Egypt/CA285/2016, A/duck/Egypt/SS19/2017 previously isolated in Egypt, and the recently circulating A/chicken/Egypt/NZ/2022 reassortant viruses' replication, pathogenicity, and viral load in comparison to the H5N1-Clade 2.2.1.2 were investigated on Madin-Darby canine kidney cell (MDCK), by using the cytopathic effect (CPE) percent and matrix-gene reverse transcription quantitative real-time polymerase chain reaction to compute the virus titer at various points in time. The A/chicken/Egypt/NZ/2022 virus was similar to the reassortant strain clade 2.3.4.4b discovered in farms in 2016. The 2 sub-groupings of hemagglutinin (HA) and neuraminidase (NA) genes were identified (I and II); the A/chicken/Egypt/NZ/2022 HA and NA genes were associated with subgroup II. The subgroup II of the HA gene was further divided into A and B owing to acquired specific mutations. The A/chicken/Egypt/NZ/2022 in our study was associated with subgroup B. The M, NS, PB1, and PB2 genes were shown to be clustered into clade 2.3.4.4b by full genome sequence analysis; however, the PA and NP genes were found to be associated with H6N2 viruses, which had particular mutations that improved viral virulence and mammalian transmission. The current results showed that the circulating H5N8 viruses were more variable than previous viruses analyzed in 2016 and 2017. Compared to other reassortant HPAI H5N8, and HPAI H5N1, the growth kinetics of A/chicken/Egypt/NZ/2022 had a high CPE without the addition of trypsin and the most viral copies with a significant difference (P < 0.01) in comparison to HPAI H5N8 and HPAI H5N1 reassortant viruses. Accordingly, the effective viral replication of A/chicken/Egypt/NZ/2022 in the MDCK than other viruses may play a factor in the spread and maintenance of specific reassortant H5N8 influenza virus in the field.

INTRODUCTION

The first highly pathogenic avian influenza viruses (HPAIVs), caused by the H5 subtype of type A influenza virus, was discovered in 1996 in domestic goose in Guangdong, China (Gs/GD lineage), and it has since been found in 80 countries, causing enormous economic damages. It was reported that the zoonotic prospect of this virus is likewise extremely strong (Lycett et al., 2019). Eight hundred sixty-two human infections with H5N1 were recorded worldwide between January 2003 and December 2020, with 455 fatalities; with 25 H5N6 cases were reported from China, 8 of which were lethal (Lycett et al., 2019).

Gs/GD lineage viruses were discovered in wild birds, in contrast to earlier HPAIVs, and this explains their rapid global spreading (Pantin-Jackwood et al., 2017). The Gs/GD viral lineage comprises 10 virus clades (0–9) and various subclades (Lee et al., 2016). Hemagglutinin (HA)-containing viruses belonging to Clade 2.3.4 acquired multiple neuraminidase (NA) genes, including N2, N5, and N8, by reassortment with other local AIV. These viruses were found in domestic birds in China in live poultry markets. Clade 2.3.4 has evolved into 2.3.4.4, which includes the H5N2, H5N6, and H5N8 subtypes that have caused pandemics with catastrophic losses to the poultry industry worldwide (European Food Safety Authority, 2020).

According to genetic studies, Korea was first exposed to 2 different genetic groupings (A and B) at the beginning of 2014 through migratory birds (Lee et al., 2018). In 2013 wild migrating birds from China were found to have the HPAI H5N8 clade 2.3.4.4 viruses (Lee et al., 2018). Long-distance migratory birds carrying group A carried it to North America in late 2014; eventually, among other viruses, it reassorted with local low pathogenic avian influenza viruses (LPAIV) to generate H5N2, which became the predominant virus circulating in North America in 2014–2015 and caused multiple outbreaks in chicken farms (Peyrot, 2020).

At the same time, group A moved westward across Europe, and by the end of 2014, there were numerous outbreaks (Napp et al., 2018). By the middle of 2016, dead wild birds bearing the reassortant HPAI H5N8 Clade 2.3.4.4 group B virus had been discovered in Qinghai Lake, China, and Uvs-Nuur Lake, south Russia. This virus had the PB2 (polymerase basic 2), PB1 (polymerase basic 1), PA (polymerase acidic), NP (Nucleoprotein), and M (Matrix) segments from Eurasian LPAI (Sobolev et al., 2021). These distinct viruses were subsequently combined with Eurasian LPAI viruses, which was dispersed throughout a wide geographic range in 2016 along waterfowl migration routes in Europe, Asia, Africa, and the Middle East (Meseko et al., 2018).

In late 2016 on Egypt's northern coast, wild birds with HPAI H5N8 Clade 2.3.4.4 group B were initially detected (Yehia et al., 2018). H5N8 has spread across Egypt since late 2016 and has supplanted H5N1 as the predominant virus in the country, despite no known human infection (European Food Safety Authority et al., 2021). Many HPAI H5N8 virus invasions have been discovered by genetic research (Tarek et al., 2021). H5N8 viruses of many reassortant viruses were found in late 2016 (Mady et al., 2019) and early 2017 (Salaheldin et al., 2018), but only one was the predominant circulating strain (Salaheldin et al., 2018; Yehia et al., 2020, 2022).

In viral isolation and propagation, Madin-Darby canine kidney (MDCK) is used instead of chicken embryonated eggs. Studying influenza virus development in cell lines facilitates comprehension of viral replication processes (Murakami et al., 2008). Due to its excellent sensitivity to numerous influenza viruses, the MDCK cell line is the most commonly utilized tool in influenza virus research (Liu et al., 2009). Due to a lack of knowledge of the virological and pathobiological traits of different reassortant H5N8 viruses.

In the current study, MDCK was used to study the replication, pathogenicity, and viral load of 2 reassortant H5N8 viruses found in wild birds and poultry in Egypt during their first introduction wave from 2016 to 2017 and the dominant circulating strain.

MATERIALS AND METHODS

Reference Viruses

HPAIV isolates (wild bird A/Common-coot/Egypt/CA285/2016 accession number: EPI1018198) were retrieved from wild birds in November 2016 (Selim et al., 2017), and A/duck/Egypt/SS19/2017-H5N8 (accession number: MH 893738) was retrieved in January 2017 from backyard ducks (Yehia et al., 2018).

A control strain representing HPAI H5N1 clade 2.2.1.2 from commercial chicken farms in Egypt, A/chicken/Egypt/15S75/2015 (accession number: EPI573317) was included. These strains were kindly provided by the Animal Health Research Institute's (AHRI) Reference Laboratory for Veterinary Control on Poultry Production, Cairo, Egypt.

Field Cases of the Virus

Based on the sudden onset of respiratory symptoms and death in 18 domestic bird populations and farms in 3 Egyptian governorates (Giza, Dakhlia, and Sharqia) between January and May 2022, pooled oropharyngeal swabs were retrieved from chickens and ducks. The samples' viral RNA was extracted according to the manufacturer's instructions using the QIAamp Viral RNA Mini Kit (Qiagene Digital Insights, Hilden, Germany).

All samples were subjected to the recognition of the M, HA, and NA genes of AIV using reverse transcription quantitative real-time polymerase chain reaction (RT-qPCR) (Fereidouni et al., 2012) by the AgPath Real-time Kit (Thermo Fisher Scientific Inc., Waltham, MA) and the RT-qPCR step one plus System (Applied Biosystems, Foster City, CA). In the subsequent stage, subtyping via RT-qPCR was carried out to determine the HA and NA genes (Hoffmann et al., 2016).

In specific pathogen-free embryonated chicken egg, positive samples were isolated, and allantoic fluid was collected and tested by rapid hemagglutination test (HA) test (World organization for animal health (OIE), 2008). The partial HA sequence was performed on 5 positive isolates. One isolate (A/chicken/Egypt/NZ/2022) was selected for complete genome sequencing using the superccriptTM III first-strand synthesis technique (Thermo Fisher Scientific, MA) with specific primers (Naguib et al., 2015).

The PCR products were purified from the gel using QIAquick gel extraction kit (Qiagene). Cycle sequencing reactions were adopted on PCR outcomes via bigdye terminator v3.1 cycle sequencing kit (Applied Biosystems). The sequence outcomes were purified using solid  phase extraction (SPE) columns (Macherey-Nagel, Düren, Nordrhein-Westfalen, Germany), and then sequenced on an ABI prism 3130 genetic analyzer (Applied Biosystems).

The CLUSTAL-W tool and the MegAlign module of DNASTAR software (Lasergene version 7.2; DNASTAR, Madison, WI) were used to match the nucleotide sequences (DNAStar, 2001). Mutation and phylogenetic evaluation were performed using the MEGA version 6 program through a bootstrap of 1000 trials of the Clustal W alignment algorithm (Tamura et al., 2013) and the maximum likelihood tree technique (Sagulenko et al., 2018).

The phylogenetic trees were constructed among Egyptian viruses and other worldwide reference and vaccine strains. The DNASTAR program was used to determine the identity of the nucleotide sequence.

Cell Culture

MDCK (1×10⁶) given by the Egyptian holding company for Biological Products and Vaccines (VACSERA) (Giza, Egypt) were cultivated in Eagle's minimal essential medium (LONZA, Basel, Switzerland), which included 10% fetal calf serum (FCS) (Greiner Bio-One, Kremsmünster, Austria), 100 U ml⁻¹ streptomycin's (LONZA, Basel, Switzerland), 100 U mL⁻¹ penicillin (LONZA) and 2 mM L-glucose.

Cells were incubated in 5% CO₂ at 37°C. The 50% tissue culture infectious dose assay (TCID₅₀ mL⁻¹) was used to titrate each virus in MDCK cells, and the titers were determined using the Reed and Muench method (Reed and Muench, 1938). The multiplicity of infection estimate was calculated to be around 0.0001.

Assessment of the Percentage of Cytopathic Effect

MDCK (1×10⁶) were grown in 6-well microplates in Dulbecco's Modified Eagle Medium (DMEM) with 10% fetal bovine serum and penicillin/streptomycin antibiotic. Confluent cells were infected with 100 µL of each virus at 0.0001 multiplicity of infection and twice rinsed with PBS before being incubated for 1 h at 37°C with 5% CO₂. The supernatant fluid was then removed, and 1 mL of DMEM with 2% serum was added before the cells were incubated at 37°C with 5% CO₂ (Cottey et al., 2001).

Cells were then incubated for an additional 12, 24, 36, and 48 h. As a negative control, cells that had not been infected were produced. Cytopathic effect (CPE) was determined by measuring cell death using a hemocytometer and the 0.4% (w/v) trypan blue staining procedure at each time point.

Real-time growth Kinetics of HPAI-H5N8 Viruses in MDCK

To assess the kinetics of the growth of 3 different H5N8 AIV strains from Egypt. The supernatants of the infected cells were collected in triplicate at specific times after the viruses were introduced into MDCK. The growth kinetics of HPAI H5N8 viruses was calculated using RT-qPCR. The infected culture medium and cell were extracted separately at 12, 24, 36, 48, and 72 h post-infection (PI).

Following the manufacturer's instructions, the samples' total RNA was extracted using the QIAamp viral RNA mini kit (Qiagen). Using influenza type A matrix (M) gene, RT-qPCR was used to detect and quantify viruses (Spackman et al., 2002). The AIV matrix gene-specific RNA was quantified and expressed as log₁₀ genome copies/reaction in contrast to a standard curve based on a 10-fold dilution of an in vitro synthesized RNA template of the implanted viruses using the Stratagen MX3005P device (Agilent Technologies, Inc., Santa Clara, CA).

Statistical Analysis

SPSS version 26 was used to code and enter the data (IBM Corp., Armonk, NY) (Pallant, 2020). Analysis of variance (ANOVA) and a hoc test were performed when comparing more than 2 groups (Chan, 2003). The P-value cutoff for statistical significance was 0.05 or lower.

RESULTS

Genetic Characterization of the Field Virus

Ten out of 18 field cases from 3 governorates were positive for the H5N8 subtype. The HAPI (H5N1) and LPAI (H9N2) subtypes presently prevalent in Egypt were not found in any of the examined samples. The partial HA sequence was performed on 5 positive isolates and aligned with other Egyptian strains in the gene bank (the result was not shown). The A/chicken/Egypt/NZ/2022 virus was chosen according to the dominant HA sequence for complete genome sequences and submitted to the National Center for Biotechnology Information (NCBI, Bethesda, MD) under the accession numbers as shown in Table 1.

Table 1.

Accession number of field strain A/chicken/Egypt/NZ/2022.

Field sample genes names	Gene Bank Accession number	Sub group
A/chicken/Egypt/PB1/NZ/2022	ON847353	Russian like H5N8 Reassortant
A/chicken/Egypt/PB2/NZ/2022	ON847352	Russian like H5N8 Reassortant
A/chicken/Egypt/PA/NZ/2022	ON908611	European like H5N8 Reassortant
A/chicken/Egypt/HA/NZ/2022	ON847347	Sub-group II-B
A/chicken/Egypt/NP/NZ/2022	ON847351	European like H5N8 Reassortant
A/chicken/Egypt/NA/NZ/2022	ON847348	Sub-group II
A/chicken/Egypt/M/NZ/2022	OP003527	Russian like H5N8 Reassortant
A/chicken/Egypt/NS/NZ/2022	ON847350	Russian like H5N8 Reassortant

The Egyptian strains HA and NA genes belonged to clade 2.3.4.4b, and were divided into 2 subgroups I, and II (Figures 1A and 1 B). The subgroup II in the HA gene gained unique mutations that led it to be further divided into A and B (Figure 1A). The Egyptian strains (A/chicken/Egypt/NZ/2022) HA gene belongs to subgroup II B, and NA gene belong to subgroup II (Figures 1A and 1 B).  However, the M, NS, PB1, and PB2 genes belong to clade 2.3.4.4b (Figures 2A, 2B, 2E, 2F), but the PA and NP genes were clustered with H6N2 viruses (Figures 2 C and 2D).

Figure 1.

(A) Phylogenetic tree of the HA gene. All Egyptian strains cluster in clade 2.3.4.4b, with two minor subgroups (I, II) and subgroup II subdivided into A and B. The Egyptian H5N8 viruses used in this study were indicated by a black dot, and the selected field strain sequenced in this study with a red dot. (B) Phylogenetic tree of the NA gene of HPAI H5N8. The Egyptian strains cluster in clade 2.3.4.4b with two minor subgroups (I and II). The Egyptian H5N8 viruses used in this study was indicated by a black dot, and the selected strain sequenced in this study with red dot.

Figure 2.

Phylogenetic tree of internal genes. The pb1 and pb2 genes of all Egyptian strains cluster in clade 2.3.4.4b (Russian-like reassortant strain), but the NP and PA genes are closely related to those of H6N2 virus. (A) phylogenetic tree of the pb2 gene. (B) phylogenetic tree of the pb1 gene. (C) phylogenetic tree of the PA gene. (D) phylogenetic tree of the NP gene. (E) phylogenetic tree of the MP gene. (F) phylogenetic tree of the NS gene. A black dot indicates Egyptian H5N8 viruses used in this study, and the selected strain sequenced in this study with a red dot.

Specific mutations in the A/chicken/Egypt/NZ/2022 virus may affect the virus's virulence and mammalian dissemination. The HA gene showed a distinct mutation (175L and 236D, 522A, 140A) previously identified in all Egyptian strains isolated in 2020-2021, as well as 106I, 201I, 213I, 245S, 265A, 295M, reported in the NA gene. In addition, all isolates had the I312V mutation, a molecular indicator of oseltamivir resistance in the NA gene.

Compared with the other 2 reassortant H5N8 and H5N1 viruses used in the present study, all 3 forms of the Egyptian reassortant HPAI H5N8 viruses had the PLREKRRKRGL at the HA cleavage site and PQGEGRRKKK/GLF in H5N1 virus, according to the deduced amino acid sequences for those viruses conferring high pathogenicity (Table 2). The mutations V127, L550 and L672 in PA, V504 in PB2, and S64 S42 in NS1 and P69 in M2 confer pathogenicity on mammals (Table S1).

Table 2.

The cleavage and antigenic sites of the 3 separate reassortant H5N8 and H5N1 viruses.

	Antigenic site A			Antigenic site B			Antigenic site E			Cleavage site
	133	140	141	154	156	184	71	83	86
Item	A	T	P	N	A/V	A	I	A	A	P	L	R	E	R	R	R	K	K	R	R	G	L	F
ON847347-A/chicken/Egypt/NZ/2022	-	A	P	-	-	-	-	-	-	-	-	-	-	K	-	-	-	*	*	-	-	-	-
A/duck/Egypt/SS19/2017	-	-	P	-	-	-	-	-	-	-	-	-	-	K	-	-	-	*	*	-	-	-	-
A/common-coot/Egypt/CA285/2016	-	-	P	-	-	-	-	-	-	-	-	-	-	K	-	-	-	*	*	-	-	-	-
A/chicken/Egypt/15S75/2015 (H5N1)	S	R	-	-	-	-	L	T	-	-	Q	G	-	G	-	-	-	-	K	*	-	-	-

^⁎= not found

- = the same mutation, A= Alanine, R= Arginine, N= Asparagine, D= Aspartic acid, C= Cysteine, Q= Glutamine, E= Glutamic acid, G= Glycine, H= Histidine, I= Isoleucine, K= Lysine, M= Methionine, F= Phenylalanine, P= Proline, S= Serine, T= Threonine, W= Tryptophan, Y= Tyrosine, and V= Valine.

Additionally, it was discovered that the viral proteins PB2, PB1, PA, NP, M1, M2, NS1, and NS2 included the mammalian preference alterations L13P, 398Q, and G70 in PB1, NP, and NS2, respectively, in all viruses (Table S2). The 57Q and 214K mutations in the PA and NP viral proteins favor mammalian preference, and the 473R mutation in PB1 that suppresses polymerase activity was present in the A/Common-coot/Egypt/CA285/2016 virus.

In PA and pb2, the A/chicken/Egypt/NZ/2022 had 615R and 558T, which favor mammalian preference. The 714S mutation in PB2 of the A/duck/Egypt/SS19/2017 lowers mammalian infectivity. Multiple uncharacteristic mutations were observed in Table S3.

Investigation of CPE in MDCK

The occurrence of CPE on MDCK was seen within 12 h PI and subsequently reaching a peak within 72 h PI in all viruses. Whilst, no CPE was observed in the uninfected cells.

Compared to other viruses, the A/chicken/Egypt/NZ/2022 had a significant difference (P < 0.05) with the highest CPE. It can be recognized at 12, 24, 36, and 48 h PI with 31.6 ± 2.9%, 50 ± 5%, 68.3 ± 2.9%, and 75%, respectively (Figure 3). This was followed by A/chicken/Egypt/15S75/2015–H5N1 which showed higher CPE than H5N8-A/Common-coot/Egypt/CA285/2016 and A/duck/Egypt/SS19/2017 with significant difference (P < 0.05). It can be recognized at 24, 36, and 48 h PI with 41.6 ± 2.8%, 68.3 ± 2.8% and 73.3 ± 2.9%, respectively. There was the same CPE with no significant differences (P > 0.05) between H5N8-A/Common- coot/Egypt/CA285/2016 and A/duck/Egypt/SS19/ 2017 at all time points. The CPE was observed as cell rounding, cell detachment, and/or adhering cell clustering. In addition, cell mortality and detachment were observed in the absence of trypsin, indicating that all tested viruses were highly pathogenic (Figure 4).

Figure 3.

Cytopathic effect (CPE) percent of tested viruses in Madin-Darby canine kidney cell (MDCK) to explain the difference in growth between viruses by measuring the CPE percent at different time points. *There are no significant differences between groups at P value ≤ 0.05.

Figure 4.

Cytopathic effect (CPE) of tested viruses in Madin-Darby canine kidney cell (MDCK) to explain the effect of viruses in the cell culture at different time points post-infection showing the difference in CPE between the reassortant H5N8 and H5N1 viruses starting from cell lysis to large plaque and control that showed no CPE.

Real-Time Growth Kinetics of HPAI H5N8 Viruses in MDCK

When compared to other viruses, the A/chicken/Egypt/NZ/2022 virus had the greatest degree of multiplicity with a significant difference (P < 0.05) in growth kinetics. It can be recognized after 12 h PI with 2.1± 0.6 log₁₀ EID₅₀. After 72 h PI, the concentration increased to 5.7±0.08 log₁₀. The A/chicken/Egypt/15S75/2015–H5N1 showed the second ranking of viral load just after A/chicken/Egypt/NZ/2022 with a significant difference (P < 0.05) when compared with other viruses except A/duck/Egypt/SS19/2017.

It can be recognized after 12 h PI with 0.8±0.9 log₁₀ EID₅₀. After 72 h PI, the concentration increased to 5.0±0.1 log₁₀ EID₅₀. The A/duck/Egypt/SS19/2017 had a higher viral load than the A/Common-coot/Egypt/CA285/2016 with concentrations of 2 ± 0.5, 1.4± 0.1 log₁₀ EID₅₀ respectively, at 12 h PI and achieving a maximum concentration of 4.5±0.07, 3.4±0.5 log₁₀ EID₅₀ after 72 h PI with a significant difference (P < 0.05) (Figure 5).

Figure 5.

Growth curve based on M-genome copies, and comparison of the 3 reassortant studied H5N8 viruses and H5N1 virus grown in Madin-Darby canine kidney cell (MDCK) based on measuring M genome copies at different time points.

DISCUSSION

HPAIV subtype H5N1 clade (2.2.1.2) formed an epizootic in Egypt until the end of 2014 (European Food Safety Authority et al., 2017) due to the continued evolution of viruses. Since May or June of 2016 (Pantin-Jackwood et al., 2016), pandemic waves caused by HPAI H5N8 clade 2.3.4.4b have spread to Asia, Europe, the Middle East, and Africa. Additionally, clade 2.3.4.4 H5N8 viruses were found in Egypt's migratory bird population at the end of 2016 (Selim et al., 2017). Its gradual spread throughout Egypt has caused substantial economic losses and the segregation of a number of diverse poultry-producing enterprises (Yehia et al., 2020).

Some months after the virus appeared in Egypt, a genotyping investigation of HPAI H5N8 viruses revealed multiple introduction of reassortant strains (Salaheldin et al., 2018). The A/Common-coot/Egypt/CA285/2016 was found in wild birds in 2016 (Selim et al., 2017), whereas the A/duck/Egypt/SS19/2017 was found in backyard ducks and commercial duck farms in 2017 (A/duck/Egypt/F446/2017), which are still widely spread in bird flocks today (Yehia et al., 2018, 2020). All 3 of these genotypes showed high nucleotide similarity at the level of HA, NA, M, and NS gene segments close to 2.3.4.4b Russia–Mongolia 2016 reassortant isolates (e.g., A/great crested grebe/Uvs-Nuur Lake/341/2016 (H5N8), with PB2, PB1, PA and NP segments originated from different influenza viruses circulated in Asia and Europe (Yehia et al., 2020).

The current study's objective was to assess the variations in viral growth and replication among 3 representative HAPI H5N8 reassortant viruses in cell culture, as well as the capacity of cells to deal with different reassortant viruses and the potential for transmission to mammalian hosts. Two reassortant HPAI H5N8 viruses were selected from the 2.3.4.4b clade based on phylogenetic studies and the other reassortant strain that became dominant circulated in Egypt. We studied their replication in the MDCK cell culture system by measuring viral load by RT-qPCR and CPE percent at specific intervals.

In our analysis, a selected strain's HA and NA genes (A/chicken/Egypt/NZ/2022) clustered in clade 2.3.4.4b, was determined as previously reported (Yehia et al., 2018, 2022). Phylogenetic analysis indicated that the Egyptian  HPAI H5N8 virus divided HA and NA into 2 subgroups. According to Yehia et al., (2020), the HA subgroups split into 2 groups, and nearly all subgroup II isolates were gathered in 2018. Subgroup II expanded into subgroups A and B, collected in 2019 and 2020–2022, respectively. As previously mentioned, our study's HA and NA genes belong to subgroups II B and II, respectively (Salaheldin et al., 2018).

Analysis of the genes encoding the internal proteins of the selected strain demonstrated grouping with clade 2.3.4.4b, except NP and PA, which were closely connected to H6N2 that were discovered in Germany in late 2016, India in 2017, and Egypt in 2017 (Selim et al., 2017). The genotype of the selected strain matched the prevalent HPAI H5N8 genotype in Egypt in 2022.

A recent study found that the pervasive use of multiple H5N1 and H5N2 vaccinations forces the virus to acquire vaccine resistance, thereby favoring escape mutations at antigenic sites (Abdelwhab et al., 2016). This analysis determined mutations at antigenic site A position 140, as previously mentioned by Yehia et al. (2020). It is necessary to conduct additional research on how this mutation affects the virus' antigenicity. The NA gene sequencing study revealed that the I312V, A245S, and T265A mutations, which are molecular markers of oseltamivir and zanamivir resistance found in the first report of H5N8 virus infection in humans (Ding et al., 2022). Additionally, the V106I alters the virus's capacity to tolerate acidic conditions, increasing the stability of the virus (Elderfield et al., 2014).

It is interesting to note that the H5N8 viruses discovered in Egypt had undergone a number of mammalian preference modifications that likely originated from other influenza virus subtypes that increase the likelihood of transmission from birds to mammals such as V127, L672, and L550 in PA, S64, and P69 in M2 in all viruses; L13P, 398Q, G70, V504, P69, and S42 in PB1, NP, NS2, PB2, M2, and NS1 respectively similar to the previously mentioned findings by Moatasim et al. (2019).

Understanding how numerous elements, such as the virus's molecular genetics, the host cell's receptor binding capabilities, and certain other factors related to the host, affect cell culture systems for viral isolation is essential for successful and efficient viral propagation (Abdelwhab et al., 2016). Human and AIVs can be efficiently reproduced on MDCK cells due to the exposure of both α2,6 and α2,3 linked sialic acid receptors (Matrosovich et al., 2003). The kind of receptors found on the cell surface was correlated with the efficiency of AIVs replication in its host. Using MDCK in the current study was recommended by the World Health Organization (WHO) for the growth of several influenza viruses. The MDCK cell line also encourages the development of the most HPAIVs (Romanova, 2017).

Currently, the pathogenicity of AIVs is linked to numerous essential amino acids at their cleavage sites (Byrd-Leotis et al., 2017). The polygenic process of influenza virus replication in a host cell depends on the cell's endocytic pathways for viral genome entry and transfer and signaling activation (Eisenreich et al., 2019). Our findings demonstrated that all the tested strains were HPAI, which showed CPE in MDCK without adding trypsin as previously mentioned by Shankar et al. (2009). This aligns with the HA gene sequencing findings and validates several basic amino acid patterns, including PLREKRRKR/GLF, at the cleavage site (Yehia et al., 2018).

The ability of the HPAI H5N8 and HPAI H5N1 viruses to replicate in MDCK tissue culture was assessed separately using the CPE percent and real-time growth kinetics of HPAI H5N8 viruses (Li et al., 2009). The proportion of CPE reflects the viral load present in the cells when comparing the virus load produced by each virus using RT-qPCR estimation. As previously mentioned, all HPAI H5N8 and H5N1 viruses in the present study could effectively replicate in MDCK when the viral titer was high as suggested by Moatasim et al. (2019).

At 12, 24, 36, and 48 h after infection, A/chicken/Egypt/NZ/2022 showed a more significant degree of cytopathogenesis compared to cells infected with other H5N8 and H5N1 viruses, followed by A/chicken/Egypt/15S75/2015-H5N1 at 12, 24, 36, and 48 h after infection with a significant difference (P < 0.05). It then reached its maximal replication 72 h after infection. Because of the virus' invasion and reproduction in the cells, the cytopathic alterations varied from cell rounding to full cell death (Schrom and Bablanian, 1981). The CPE percentage between the A/Common-coot/Egypt/CA285/2016 and the A/duck/Egypt/SS19/2017 at different time points was the same.

In addition, the virus load in A/chicken/Egypt/NZ/2022 virus outscored other viruses in cell culture with significant difference (P < 0.05), followed by A/chicken/Egypt/15S75/2015-H5N1 when compared with other viruses with significant difference (P < 0.05) in all viruses except A/duck/Egypt/SS19/17 in which there was no significant difference (P > 0.05).

A/duck/Egypt/SS19/2017 outscored A/Common-coot/Egypt/CA285/2016 by 0.6 and 1.0 log₁₀ with significant difference (P < 0.05) from 12 h to 72 h PI. A previous study that evaluated the growth kinetics of H5N1 and H5N8 viruses using PCR discovered that H5N8 had higher replication than other viruses (Moatasim et al., 2019). Based on these current investigations, it might be suggested that the A/chicken/Egypt/NZ/2022 has a higher incidence of infection and better replication in MDCK compared to the other 2 reassortant HPAI H5N8 viruses. This may aid in the emergence and persistence of a specific strain of the H5N8 influenza virus in the field.

The presence of 615R and 558T in PA and PB2, respectively, could explain the high propagation of A/chicken/Egypt/NZ/2022 in MDCK as a mammalian cell (Chen et al., 2006), whereas A/Common-coot/Egypt/CA285/2016 and A/duck/Egypt/SS19/2017 had 473R and 714S in PB1 and PB2, respectively, which decreased polymerase activity and replication affinity in mammalian cell (Feng et al., 2012). Early work has shown that PB2 affects the virulence and efficiency of AIV replication in rats (Bi et al., 2015). The impact of various polymerase reassortments and molecular markers on the A/chicken/Egypt/NZ/2022 virus growth in MDCK and mice will be the subject of future study.

According to earlier research, to explain why HPAI H5N8 has spread more widely than HPAI H5N1, this may be explained by the display maintained fitness between its HA and NA activities (Yu et al., 2017). H5N8 has increased NA activity compared to the other H5Nx reassortant (Diederich et al., 2015). The H5 subtype clade 2.3.4.4 HA, in combination with N2, N6, or N8, increased viral fitness. Furthermore, the H5 subtype of HA combined with N8 viruses expanded more successfully than the N2 virus, demonstrating that the match of HA/NA boosted virus replication prior findings that the balance between HA and NA had a significant influence on viral growth (Yen et al., 2011).

Thus, our observation confirmed that the dominant genotype 2.3.4.4 in Egyptian chickens was more pathogenic and demonstrated extra effective viral multiplication and transmissibility. It also displayed a higher viral titer compared to HPAI H5N1, indicating the sustainable development of some H5N8 genotypes and a decline in the occurrence of H5N1 clade 2.2.1.2 (Yu et al., 2017; Yehia et al., 2022). This was supported by the fact that the H5N8 clade 2.3.4.4 virus was much more invasive in MDCK cells than the H5N1 clade 2.2.1.2 virus.

CONCLUSION

Comparative replication studies of the 3 HPAI H5N8 viruses' capacities have shown that the A/chicken/Egypt/NZ/2022 strain was significantly more pathogenic, more virulent and capable of rapid multiplication and invasion than other H5N8 and H5N1 viruses in MDCK cells with high viral load.