First serological evidence of H9 low pathogenic avian influenza virus exposure among rural population in Sidi Kacem province in Morocco: a cross-sectional study

¹Research Center (BIOBIO), Laboratory of Biodiversity, Ecology and Genome, Faculty of Sciences, University Mohammed V in Rabat, Rabat, Morocco, ²Avian Pathology Unit, Animal Diseases and Public Health department, Institut Agronomique et Vétérinaire Hassan II, Rabat, Morocco, ³Laboratory of bacteriology, Pasteur Institute of Morocco, MoH, 1, Place Charles Nicolle, 20360, Casablanca, Morocco, ⁴Laboratory of Biology and Health URAC34-Metabolic and Immunologic Pathology Research Team, Faculty of Science, Ben M´sik, Hassan II University, Casablanca, Morocco

^&Corresponding author
Hind Ezzine, Research Center (BIOBIO), Laboratory of Biodiversity, Ecology and Genome, Faculty of Sciences, University Mohammed V in Rabat, Rabat, Morocco

Introduction    

Avian influenza (AI) viruses are highly contagious as a zoonotic pathogen [1], highly variable and widespread in birds [2]. Low Pathogenic Avian Influenza (LPAI) causes a mild infection in birds and generally does not pose a significant threat to human health. However, strains of Highly Pathogenic Avian Influenza (HPAI) are often lethal to birds and easily transmitted between susceptible species [3]. In deed, HPAI viruses that naturally cause acute clinical signs in broilers, turkeys, and other farmed birds have been associated only with H5 and H7 subtypes [4]. Rarely, AI viruses can adapt its characteristics to circulate in mammals. Over the past century, these viruses have contributed to pandemics in humans and influenza virus diversity in pigs and were the source of the currently circulating canine influenza in dogs [5].

In addition, the emergence of pandemic influenza viruses envisages a gene flow from the aquatic avian reservoir to humans via reassortment in pigs. The study of the recent outbreaks of H5N1 in Hong Kong since 1997 [6] and the isolation of the H9N2 avian virus in humans raise hypotheses about the risk of the emergence of a new pandemic virus. These H9N2 viruses have amino acids in their hemagglutinin, indicating their potential to infect humans directly [7]. In a study comparing the HA sequence of 16 H9N2 viruses to the first strain isolated in Morocco, all the Moroccan H9N2 viruses carried 158N, H183 amino acid substitutions (H3 numbering) in their receptor binding site related to human virus like receptor specificity [8]. In addition, some of H9N2 viruses contain gene segments closely related to those of A/Hong Kong/156/97 (H5N1/97, H5N1) or A/Quail/Hong Kong/G1/97 (G1-like, H9N2) [9]. Thus, avian influenza H9 is a severe health hazard [7].

A small number of influenza virus subtypes circulate in humans, such as H1N1 and H3N2, resulting in seasonal infections or occasional pandemics [10]. Other AI subtypes, such as H5N1, H5N6, H6N1, H7N2, H7N3, H7N4, H7N7, H7N9, H9N2, H10N7 and H10N8, may occasionally cause sporadic infections and/or deaths [11-17] without sustainable human-to-human transmission [18]. The adaptability of influenza viruses to circulate between different species raises the possibility of a highly pathogenic variant that could lead to a devastating pandemic and devastating economic losses. Therefore, the control of this disease at the human-animal interface is an important part of the control and preparedness strategies against Influenza pandemics [19,20].

In Morocco, the first outbreaks of H9N2 influenza virus were recorded in early 2016, affecting different types of poultry production. The virus has spread rapidly throughout the country and is now endemic in poultry [8,21]. Multiple mutations associated with the mammalian host have been detected which would promote transmission of the virus from avian hosts to mammals, including humans, other virulence-related mutations have also been identified [21]. Therefore, vigilance is required in the human health sector to detect any possible human contamination by this subtype, especially since human cases of H9N2 have been reported by the WHO, including 17 cases in China [18,22]. This study aims to identify the potential circulation of H9 avian influenza in farmworkers and habitants with poultry contact in Sidi Kacem province in Morocco.

Methods     

Study design: we conducted a cross-sectional study.

Study period and setting: this study was conducted on sera collected between November 1, 2017 and May 30, 2018, in 36 farms of 11 municipalities in the province of Sidi Kacem in the northwest of Morocco. This province has a total area of 4.060 km² (of which, 3024.25 km² is rural) and an estimated population of 514 000 (67% rural). It is composed of 29 municipalities (5 urban and 24 rural) whose economy is based primarily on agriculture. Livestock is one of the concerns of many farmers in the province of Sidi Kacem. This province encompasses 22 679 breeding farms for all types of livestock. There is traditional breeding of a limited amount of poultry on these farms. The province is also included in one of the paths of migratory birds.

Participants: we used a convenience sample. Our study was carried out on 185 samples taken, as part of a seroprevalence study of brucellosis in Morocco [23], from a population of breeders and their entourage living in farms in 11 communes of Sidi Kacem (Figure 1). The farm residents raise various animals around their houses, including poultry. The sera were shared with an anonymous database.

Laboratory analysis: the samples were analyzed in the laboratory of the Avian Pathology Unit at the Hassan II Agronomic and Veterinary Institute in Rabat. H9 avian influenza virus antibodies in the sera samples were detected by using the haemagglutination inhibition (HI) test. The HI is the standard method for serological detection of influenza virus infection in humans. The sera obtained were treated with RDE (Receptor Destroying Enzyme) by diluting one part of the serum with three parts of enzyme and incubated overnight in a water bath at 37°C. Then, the enzyme was inactivated by incubating for 30 minutes at 56°C, followed by adding six parts of 0.85% saline to obtain a final dilution of 1/10. Hemagglutination inhibition (HI) assays were performed in 96-well U-bottom microtiter plates with 0.5% poultry erythrocytes [24]. To assess the levels of influenza H9 antibody titers, we used the HI assay using H9 antigens. Twice serial dilutions of the collected serum (25 μL) were prepared, and then 4 HA units of influenza H9 virus were added to each dilution. Then, after 30 minutes of incubation, 25 μl of 0.1% washed chicken red blood cells were added to each well, and the mixture was incubated for 30 minutes at room temperature. The endpoint of HI (the highest dilution of serum causing complete inhibition) was noted and recorded as common log2 values of the highest dilution of HI.

Statistical analysis: a statistical analysis was performed using Epi-Info7 and Excel software. It concerned demographic, epidemiological characteristics such as age, gender and profession; then the H9 status and the antibody titer.

Ethical aspects: the sera samples used in this study were collected for another study granted, under number 34/16 dated 07/01/2016, by the Ethics Committee for Biomedical Research of Mohammed V University of Rabat (Faculty of Medicine and Pharmacy of Rabat).

Results     

A total of 185 sera were received and analyzed at the Avian Pathology Unit of IAV, of which 37.8% (70) were negative for H9, and 62.2% (115) were positive. H9 antibody titers ranged from (3 log₂ to 6 log₂) with 25.9% having a titer equal to 4 log₂, cases with a titer less than or equal to 2 log 2 (37.8%) were considered negative (Table 1). Anti-H9 were found in all municipalities, with positivity rates ranging from 38.9% in the Bir Taleb municipality to 100% in the Sidi Azouz municipality (Figure 2). Among the 115 positive cases, 65 (56.5%) were female. The study series ranged in age from 18 to 83 years. The mean age was 45.0 + 14.9 and 46.0 + 16.7 years old for the positive and negative cases, respectively, without any statistically significant difference (P=0.68). Indeed, 65.5% of the seropositive cases were in the 25-64 age group. The highest positivity rate (68%) was recorded in the 25-40 age group (Figure 3).

The majority of those sampled were housewives (42.4%), while farmers, livestock keepers and students accounted for 33.2%, 21.7% and 1.6%, respectively, and 1.1% were in other occupations, namely, a driver transporting livestock and an Imam. The latter two were H9 positive, while the positivity rate ranged from 66.7% among farmers to 52.5% among livestock keepers (Figure 4). Regarding educational level, 48% of the sera tested belonged to persons who had never attended school, 10% had preschool level, 26% and 14% had attended primary and secondary school respectively, and only 2% had a higher education level. The positivity rate ranged from 46.9% among those with primary school education to 77.8% among those who had attended only preschool level. This rate was 60.0% among those with secondary school level, 67.4% among those who had never attended school and 75.0% among those with high education level.

Discussion     

Avian Influenza (AI) viruses circulating in birds are a serious issue to public health, in fact, they could be transmitted from avian species to mammals and humans, thus causing public health threats [25]. Avian Influenza viruses are divided on the Low Pathogenic and High Pathogenic AI, according to their potential to kill chickens and/or induce severe symptoms in commercial flocks [26]. H9N2 is considered one of the most critical subtypes for avian sector, responsible for low pathogenic avian influenza. H9N2 is considered as an endemic virus within the poultry population of Eurasia and Africa. In fact, in the Eurasian region, the H9N2 subtype has caused enormous economic losses to the poultry industry due to decreased egg production and high mortality combined with other infections [27]. The same symptoms and economic issues were reported in Morocco, since the first detection of H9N2 in January 2016 [8,28]. To control and to prevent infection caused by H9N2 AIV, inactivated vaccines have been used in chickens, these inactivated vaccines have protected against clinical signs [29], unfortunately, all vaccination strategies implemented in Morocco since 2016 were not been able to eradicate completely the virus in chicken [30].

Therefore, H9N2 AIVs still circulated in farmed chickens during the past decade [31], with the influenza virus subtype evolving rapidly and may acquire mutations that would facilitate its multiplication in mammals, including humans [32]. In fact, our study revealed for the first time the presence of H9 antibodies on samples collected from a breeder and their entourages within the rural population of 11 municipalities in Sidi Kacem. These results might be explained by the contact and interaction with animals, including poultry and wild birds, and confirm the genomic finding and the adaptability of H9 to human receptors, therefore inducing an immune response measured by HA [8]. The positivity rate in this study was 62.2%, close to that recorded at the farm level in Pakistan (45.5%) [7], lower than that found by Hadipour (87%) among chicken farmers [33] and significantly higher than the general population in central China (22.72%) [34]. In other studies, serological investigations of AIV antibodies in poultry were conducted, the seroprevalence was variable, indeed Bangladesh have reported (9.82%) [35], Bosnia-Herzegovina (2.87%) [36], Nigeria (4.4%) [37] and Pakistan (14.3%) [38]. The difference between those results might be explained by the size and the characteristics of the sampled population, the used tests, but confirm the circulation of H9 virus among poultry.

The current study was based on the search for anti-influenza H9 antibodies in asymptomatic people, the detection of viral RNA in those showing signs in favor of influenza would be a better way to confirm their infection by this type of virus. In addition, we could not estimate the seroprevalence because we conducted this study on 185 sera collected as part of another brucellosis seroprevalence study which has a different prevalence from influenza A H9, thus, our sample is not representative. Moreover, as this virus includes a human-like receptor specificity [39], it has the potential to transmit to humans. Furthermore, it has recently been recognized that AI subtype H9 has shared gene segments with highly zoonotic viruses such as H7N9 that would contribute to the emergence of the next influenza pandemic [40,41]. In fact, Bennani et al. have reported that H9N2 isolates from broiler flocks in Morocco, carried 158N, H183 amino acid substitutions (H3 numbering) in their receptor binding site related to human virus like receptor specificity [8]. Actually, all H9 duck isolates, except for two H9N1 and one wild-duck H9N2 (WDk/ST/4808) virus lineages, had amino acid residue leucine (L) at position 226 (H3 numbering) and glycine (G) at position 228 at the receptor-binding site.

It is important to highlight that since 1999, several cases of human infection with H9N2 virus have been reported in Asia and are generally associated with mild disease [42], and the transmission of H9N2 virus appears to be exclusively from birds to humans, including highly pathogenic strains. Recently, many H9N2 cases have been found in China, most likely due to the ongoing screening for zoonotic H7N9, and in Egypt and Bangladesh due to ongoing screening for zoonotic H5N1 infections; however, no human-to-human transmission by H9N2 virus has been reported [43,44], although reassortment of H9N2 with other highly pathogenic viruses of type H5 and H7 is possible [44], because H9 could donate its gene segments to other emerging avian influenza viruses, including H5N2, H6N1, H7N7, H7N9 and H10N8 [25,45-50].

In 2013, an H5N1 virus detected in a patient in Canada after a stay in China contained 2 polymerase genes from an H9N2 virus [28]. The H5N2 viruses expressing the HA gene have been isolated with internal genes from both H5N1 viruses and H9N2 viruses [51]. Actually, as part of the ONE HEALTH approach, the detection of antibodies anti-H9 in human population, and the notification of H9 human cases, in many countries, encourages us to strengthen the epidemiological surveillance taking into account the risk of a new pandemic for which this subtype may be responsible, especially since avian influenza A(H9N2) virus infections in humans are usually mild or even subclinical [33]. This study had certain limitations. It is based on the search for anti-influenza H9 antibodies in asymptomatic people, the detection of viral RNA in those showing signs in favor of influenza would be a better way to confirm their infection by this type of virus. In addition, we could not estimate the seroprevalence because we conducted this study was not representative, a study with a representative random sample would have made it possible to estimate the seroprevalence of this infection in the province of Sidi Kacem.

Conclusion     

This study demonstrate that subtype H9 has been circulating in human population in the province of Sidi Kacem in Morocco, after its detection in poultry flocks among the country. It is therefore necessary to monitor this subtype through the human influenza sentinel surveillance system. Finally, our finding reinforces the worth value of the ONE HEALTH approach in public health and multisectoral collaboration.

Competing interests     

The authors declare no competing interests.

Authors' contributions     

Hind Ezzine and Mariam Laatifi designed the study and performed data analysis and interpretation of results. Hind Ezzine, Mohamed Elazhari and KF helped with data collection. Hind Ezzine, Mohamed Elazhari, Kaoutar Faddane, Sara Rakani, Saadia Nassik and Mariam Naciri assisted with laboratory analysis and interpretation of laboratory results. Hind Ezzine and Mariam Laatifi wrote the paper. Saadia Nassik and Mariam Naciri provided pedagogic support, corrected the manuscript, and approved its final version to be published. All the authors have read and agreed to the final manuscript.

Acknowledgments     

We acknowledge critical support from all active members of the Avian Pathology Unit, Animal Diseases and Public Health department, Agronomic and Veterinary Institute Hassan II, Rabat, Morocco.

Tables and figures     

Table 1: anti-H9 antibody titers in Sidi Kacem farms, Morocco, November 2017-May 2018

Figure 1: Sidi Kacem municipalities where the sera were collected, Morocco, November 2017-May 2018

Figure 2: distribution of cases (positive and negative) and positivity rate by municipality, Sidi Kacem, Morocco, November 2017-May 2018

Figure 3: distribution of anti-H9 positivity rate by age group at the studied farms, Sidi Kacem province, Morocco, November 2017-May 2018

Figure 4: positivity rate by occupation at studied farms, Sidi Kacem, Morocco November 2017-May 2018

References