Genetic characterization of influenza A(H3N2) viruses circulating in coastal Kenya, 2009-2017.

BACKGROUND
Influenza viruses evolve rapidly and undergo immune driven selection, especially in the hemagglutinin (HA) protein. We report amino acid changes affecting antigenic epitopes and receptor-binding sites of A(H3N2) viruses circulating in Kilifi, Kenya, from 2009 to 2017.


METHODS
Next-generation sequencing (NGS) was used to generate A(H3N2) virus genomic data from influenza-positive specimens collected from hospital admissions and health facility outpatients presenting with acute respiratory illness to health facilities within the Kilifi Health and Demographic Surveillance System. Full-length HA sequences were utilized to characterize A(H3N2) virus genetic and antigenic changes.


RESULTS
From 186 (90 inpatient and 96 outpatient) influenza A virus-positive specimens processed, 101 A(H3N2) virus whole genomes were obtained. Among viruses identified in inpatient specimens from 2009 to 2015, divergence of circulating A(H3N2) viruses from the vaccine strains A/Perth/16/2009, A/Texas/50/2012, and A/Switzerland/9715293/2013 formed 6 genetic clades (A/Victoria/208/2009-like, 3B, 3C, 3C.2a, 4, and 7). Among viruses identified in outpatient specimens from 2015 to 2017, divergence of circulating A(H3N2) viruses from vaccine strain A/Hong Kong/4801/2014 formed clade 3C.2a, subclades 3C.2a2 and 3C.2a3, and subgroup 3C.2a1b. Several amino acid substitutions were associated with the continued genetic evolution of A(H3N2) strains in circulation.


CONCLUSIONS
Our results suggest continuing evolution of currently circulating A(H3N2) viruses in Kilifi, coastal Kenya and suggest the need for continuous genetic and antigenic viral surveillance of circulating seasonal influenza viruses with broad geographic representation to facilitate prompt and efficient selection of influenza strains for inclusion in future influenza vaccines.


| INTRODUC TI ON
Seasonal influenza viruses infect 5%-15% of the global population annually, resulting in 290 000-650 000 deaths each year. 1,2 The disease burden is highest in developing countries especially in sub-Saharan Africa, [3][4][5] where influenza viruses may circulate year-round without clear seasonality; this is in contrast to the clear seasonality observed in temperate climatic regions. 6 Safe influenza vaccines exist, 1 but effectiveness depends on host immune responses and how well the vaccine strains match the strains in circulation. 7 Vaccine effectiveness can be low when there is a mismatch between vaccine selected strains and circulating viruses. 8 Influenza A viruses (IAV) cause the majority of influenza-associated disease burden and are further classified into subtypes based on the combination of their hemagglutinin (HA) and neuraminidase (NA) surface glycoproteins. 1 IAV, especially A(H3N2) virus, evolve rapidly and undergo immune driven selection. 9 This occurs through changes in viral antigenic epitopes that result in evasion of immune recognition and mainly involves mutations in the HA and NA gene segments. 10,11 The HA glycoprotein is the primary target of host neutralizing antibodies, which inhibit the binding of HA to sialic acid receptors present on epithelial cell membranes of the upper respiratory tract. 12 Influenza A(H3N2) virus HA possesses defined antigenic epitopes (five sites designated A through E) and receptor-binding sites. 13 Accumulation of mutations at these antigenic sites results in viral escape from the host immune response. 14,15 These sequence drifts on the HA from accumulated mutations are observed more frequently in A(H3N2) virus than A(H1N1) virus. 8 which led to a low or null vaccine effectiveness for that season. [17][18][19][20] As vaccine effectiveness may not be fully explained by antigenic analysis using the hemagglutinin inhibition (HI) assay, the availability of high-throughput platforms to characterize HA genetic groups, for example, next-generation sequencing (NGS) techniques, can provide more timely information to evaluate protection afforded by vaccination.
Currently, the government of Kenya is considering recommending annual influenza vaccine for young children. 21 As an influenza vaccination program is implemented, there will be a need to establish genetic and antigenic viral surveillance which could be used to assess how well the vaccine performs and inform public health decisions on vaccination strategies. 7 We characterized the genetic changes in A(H3N2) viruses circulating in coastal Kenya using full-length HA sequences generated through next-generation sequencing (NGS) from respiratory specimens collected from inpatient and outpatient sentinel surveillance sites in coastal Kenya from 2009 to 2017.

| Sample sources and molecular screening
The samples used in this study were collected from health facilities within the Kilifi Health and Demographic Surveillance System (KHDSS) on the coast of Kenya. 22 [22][23][24][25] Samples were stored in viral transport medium (VTM) at −80°C prior to molecular screening and subsequent processing. 24,26,27 Samples were screened for a range of respiratory viruses, in-

| Bioinformatic analysis
Contiguous nucleotide sequence (contigs) assembly was carried out using the FLU module of the Iterative Refinement Meta-Assembler (IRMA), 29 which performs iterative segment-level read sorting based on Lineage Assignment By Extended Learning (LABEL), 30 and iteratively refines the references to optimize the final assembly based on the Striped Smith-Waterman (SSW) algorithm. 31

| Prediction of potential glycosylation sites
Potential N-linked glycosylation sites on the HA were determined using the NetNGlyc 1.0 server (http://www.cbs.dtu.dk/servi ces/ NetNG lyc/), in which a threshold of >0.5 suggests glycosylation.

| Ethics
Ethical clearance for the study was granted by the KEMRI-Scientific and Ethical Review Unit (SERU# 3103) and the University of Warwick Biomedical and Scientific Research Ethics Committee (BSREC# REGO-2015-6102). Informed consent was sought and received from the study participants for the study.

| Direct sequencing of IAV from clinical specimens using NGS
The specimen processing flow for IAV NGS is shown in Figure 1

| Analysis of A(H3N2) virus HA gene sequences
We The receptor-binding site of A(H3N2) virus is highly conserved at amino acids positions 98, 136,153,183,190,194,195, and 228 on HA1. However, we did not observe any substitution in these conserved positions in the Kilifi strains.

| D ISCUSS I ON
In this study, we characterized the circulating A(H3N2) viruses in

CO M PE TI N G I NTER E S TS
No competing interests were disclosed.