On January 2, 2017, high death rates were reported in Indian peafowl (Pavo cristatus) in a backyard exotic poultry farm located in the town of Makilingaye (Tokombéré district of Mayo-Sava division), a village neighboring Nigeria in the far-north region of Cameroon. Samples were collected by the Cameroon Epidemio-Surveillance Network (Reseau d’Epidemio-Surveillance au Cameroun [RESCAM]) of the Ministry of Livestock, Fisheries and Animal Industries and were sent to the National Veterinary Laboratory (LANAVET) in Garoua, where the H5N8 subtype was diagnosed. Almost all the peafowl (103/107) died within ≈2 weeks. Death in chickens (Gallus gallus domesticus, 24/24) housed in the same compound was delayed and appeared 5 days later than in peafowl. Following the laboratory confirmation of the first outbreak, the Cameroon government, through the Ministry of Livestock, Fisheries and Animal Industries, implemented prompt and strong control measures to stop the spread of the virus and reduce the risk of human infections. Stamping out was deployed together with movement restrictions and virological surveillance; disinfection of premises and contact materials was intensified. The RESCAM team conducted a routine avian influenza survey in the Maroua, Yagoua, and Guidiguis central poultry markets in the far-north region. All the samples were analyzed at LANAVET Garoua and Annex Yaounde; H5N8 virus was detected in 5 birds (1 pigeon, 1 chicken, 2 guinea fowls, and 1 duck) out of 122 birds.

We sequenced the hemagglutinin (HA) and NA gene segments of the virus A/Indian peafowl/Cameroon/17RS1661-6/2017, identified from an Indian peafowl in Makilingaye, at the Istituto Zooprofilattico Sperimentale delle Venezie in Legnaro, Italy, along with the complete genomes of 2 positive samples collected from domestic birds (chicken and duck) and the partial genome (all segments except for polymerase acidic protein [PA] and polymerase basic protein 1 [PB1]) of a sample from a wild pigeon, all identified in the Maroua and Yagoua markets (Table). A detailed description of the methods used for sequencing and genetic analyses is provided in Technical Appendix 1 and details on the HA gene segments used for the analysis are given in Technical Appendix 2. We submitted consensus sequences to GenBank (accession nos. MG650618–41).

Figure

Figure. Maximum-likelihood phylogenetic tree of the HA gene of highly pathogenic avian influenza A(H5N8) virus from Cameroon (bold) and reference viruses. Arrows indicate location where Cameroon viruses were obtained. Bootstrap supports >600/1,000...

Topology of the phylogenetic tree based on the HA gene segment showed that the 4 H5N8 viruses detected in Cameroon in 2017 fell within genetic clade 2.3.4.4 group B (Figure; Technical Appendix 1 Figure 1). The HA gene segments of the 3 viruses from Makilingaye (A/Indian peafowl/Cameroon/17RS1661–6/2017) and Maroua market clustered together; similarity ranged from 99.5% (between the viruses collected from Makilingaye and Maroua markets) to 100% (between the 2 viruses from Maroua market). These segments grouped with H5N8 viruses collected during 2016–2017 in South Korea, Egypt, Italy, Siberia, and India (similarity between 99.3% and 99.6%). However, the virus A/duck/Cameroon/17RS1661-3/2017 identified in Yagoua market clustered separately and showed the highest similarity (99.3%) to a virus from India (A/duck/India/10CA01/2016) (Figure).

We confirmed the different clustering of the viruses collected from distinct markets in Cameroon by the analyses of the other gene segments, which suggests the occurrence of >2 distinct viral introductions in the country. Specifically, sequences of the 2 viruses from Maroua market were identical for all the available gene segments (Table) and clustered with viruses collected in Asia, Europe, and Egypt (Technical Appendix 1 Figures 2–8). On the other hand, the virus A/duck/Cameroon/17RS1661-3/2017 grouped with the A/duck/India/10CA01/2016 virus (similarity 94.3–99.9%), except for the nucleoprotein (NP) gene, where it showed the highest identity (99.3%) with a virus collected in 2016 in the Russian Federation (Uvs Nuur Lake). The lack of genetic information on other H5N8 viruses detected in Africa makes it impossible to pinpoint how these viruses entered Cameroon and spread, nor is it feasible to determine where the 2 introductions occurred. The time to the most recent common ancestor estimated for the HA gene (Technical Appendix 1) suggested that 2 introductions into Africa may have occurred almost simultaneously during March–December 2016 (Technical Appendix 1 Figure 9). However, analyses of the amino acid sequences show that A/duck/Cameroon/17RS1661-3/2017 possesses the N319K mutation in the NP protein, which has been reported to enhance polymerase activity and stimulate vRNA synthesis in mammalian cells (6). In addition, A/chicken/Cameroon/17RS1661-1/2017 contains the V100A mutation in the PA protein, which is an amino acid signature typical of human influenza viruses (7).

Nigeria was the first country in West Africa to report the presence of the HPAI H5N8 virus in November 2016 (8,9). Less than 2 months later, in January 2017, the virus was detected in the far-north region of Cameroon in Makilingaye, close to the Nigerian border. Considering the extensive and porous frontier between Cameroon and Nigeria, trade and movement of poultry might have played a key role in the spread of the virus. However, the involvement of wild birds cannot be excluded. The region contains several wetlands (Domayo River, Mayo Kani River, and Maga Lake) where different wild birds congregate, in particular during the dry season (December–April), when the virus was first detected. The almost simultaneous detection (early January 2017) of the H5N8 virus in poultry and wild birds in distant locations in Africa, such as Tunisia and Uganda, suggests that the role of wild birds in the introduction and/or dissemination of the virus in the region should not be overlooked. Of note, the first outbreak caused by the Gs/GD/96 H5 lineage in Cameroon, in 2006, also occurred in the far-north region of Cameroon at about the same time. This finding might suggest a common pathway for introduction of the virus into this area and highlights the need to improve surveillance in this region.

Although the epidemiologic and genetic data are insufficient to establish definite pathways and time of introduction of H5N8 virus into West Africa, this study demonstrates that >2 distinct H5N8 viruses entered Cameroon. This finding, together with the evidence that this event represents the third incursion of a Gs/GD/96-lineage H5 HPAI virus into Cameroon, again underlines the need to perform avian influenza surveillance on an ongoing basis for rapid identification and response to outbreaks in this area.