First outbreaks and phylogenetic analyses of avian influenza H9N2 viruses isolated from poultry flocks in Morocco

11 poultry flocks from different regions of Morocco showing very high mortality, decrease in feed consumption and very severe respiratory signs including sneezing, coughing, rales and gasping were analysed by real time RT-PCR for the presence of nucleic acids of infectious bronchitis, Newcastle disease, virulent infectious bursal disease viruses and IAV [10, 19‐21]. Details on the flocks sampled in this study are presented in Table 1.

Five samples out of 11 with low Ct values in real time RT-PCR for IAV were selected for inoculation into the allantoic cavities of 9–day–old specific pathogen free chickens embryos, incubated for 48 h at 37 °C then chilled at 4 °C for 4 h before harvesting. The allantoic fluid was harvested, clarified, tested by real time RT-PCR for IAV genome, and then stored at −80 °C until use.

PCR amplification was performed in a thermocycler (GeneAmp PCR System 9700, Applied Biosystems) using segment specific primers described previously by Hoffman et al., (2001) [22] and the One step RT-PCR kit (Qiagen) following instructions of the manufacturer. Reactions were performed according to the following protocol: 50 °C for 30 min and 95 °C for 5 min, followed by 40 cycles of 95 °C for 30 s, 54 °C to 57 °C for 30 s, 72 °C for 1 min30 s, and a final elongation step of 6 min at 72 °C. PCR products of the expected length were purified with the Nucleospin gel and PCR cleanup kit (Macherey Nagel) according to the manufacturer’s instructions. Sanger sequencing method was carried out on all the segments using the same primers as used for the RT-PCR. Sequencing was performed on a 3130XL Applied Biosystems capillary sequencer at the Plateau de Génomique GeT-Purpan, UDEAR UMR 5165 CNRS/UPS, CHU PURPAN, Toulouse, France.

Complete sequence data of five H9N2 isolates were manually assembled using BioEdit software package version 5.0.9 [23]. The open source BLAST program (National Center for Biotechnology Information, Bethesda MD, http://​blast.​ncbi.​nlm.​nih.​gov/​Blast.​cgi) was used for sequence comparison. The best fitting nucleotide substitution model was estimated by means of hierarchical likelihood ratio approach using Mega 6.06 [24] for each sequences alignment. Phylogenetic analysis and tree construction were thus generated using the maximum likelihood method, General Time Reversible model, with 500 bootstrap replicates with Mega 6.06, and bootstrap values above 50 were labelled on major tree branches. Sequences were submitted to the EMBL/GenBank databases.

Up to the end of December 2015, no official report on the presence of any subtype of avian influenza in commercial poultry farms or backyard flocks had been recorded in Morocco. During the second week of January 2016, 11cases of chickens from different poultry production areas around the cities of Fez, Meknes, Kenitra and Rabat, were received for diagnosis purpose in our laboratory (Table 1). Most cases showed very high decrease in feed consumption and very severe respiratory signs (sneezing, coughing, rales and gasping). All cases showed a very rapid increase in mortality rates (2 % to 10 % per day between the 3rd and the 6th day after onset of disease). All affected flocks were over three weeks of age and some flocks were at the selling period. In a few cases, diarrhoea was reported. All flocks were vaccinated, as routinely performed, against Newcastle, infectious bronchitis and Gumboro diseases. By mid-February, all parts of the country were affected and the disease spread to layers and breeders farms, which showed very severe respiratory signs, mortality rates ranging from 2 to 15 %and drastic drop in egg production (up to 80 %) with no complete recovery after several weeks. Indeed, up to 10 weeks after the infection, egg production remained between 20 and 25 % lower than normal levels. Turkey flocks were similarly affected at different ages, with similar clinical signs as seen in chickens as well as sinusitis with facial swellings, and mortality rates around 10 %, but with a large inter-flock variability, likely due to other contributing factors.

Post mortem examinations revealed that all carcasses were congested, some showing oculonasal discharge and infraorbital sinusitis. The main consistent lesions were congested or haemorrhagic tracheas with fibrinous exudates in the lumen and presence of fibrinous casts in bronchi (Fig. 1), haemorrhagic lungs, pneumonia with fibrinous exudates and airsaculitis. In some cases congested and swollen kidneys, spleen and liver were observed.

Eleven samples from respiratory tissues collected from different cases and analysed by real time RT-PCR for the presence of Infectious bronchitis, Newcastle disease and virulent IBD viruses werenegative. However10 out of 11 samples were positive for IAV with cycle threshold (Ct) values ranging from 17.5 to 35.1 (Table 1).

The phylogenetic analyses showed that the Moroccan H9N2 viruses of broiler and breeder broiler origins were of the G1 lineage and closely related to each other (Fig. 2). The nucleotide sequence identities among the five Moroccan strains ranged from 98 to 100 %. Genes segments coding for polymerase basic 2 (PB2), polymerase basic 1 (PB1), acidic polymerase (PA), hemagglutinine (HA), nucleoprotein (NP), neuraminidase (NA), matrix (M) and non structural (NS) proteins all clustered with G1 lineage viruses, closely related to an A/pheasant/United Arab Emirates/D1521/2011-like H9N2 virus, as indicated by the trees topology (Figs. 2, 3) and by the high nucleotide sequence identity: 96.1 to 98.2 %.