key: cord-0290725-sk5ov5xm authors: Nabakooza, Grace; Collins Owuor, David; de Laurent, Zaydah R.; Owor, Nicholas; Timothy Kayiwa, John; Jjingo, Daudi; Nyaigoti Agoti, Charles; James Nokes, David; Patrick Kateete, David; Mulindwa Kitayimbwa, John; David William Frost, Simon; Julian Lutwama, Julius title: Phylogenomic analysis of Uganda influenza type-A viruses to assess their relatedness to the vaccine strains and other Africa viruses: a molecular epidemiology study date: 2021-07-05 journal: bioRxiv DOI: 10.1101/2021.07.05.451078 sha: d1e9c95b061afc0aeeaa20fb58d2eb8736795248 doc_id: 290725 cord_uid: sk5ov5xm Background Genetic characterisation of circulating influenza viruses is essential for vaccine selection and mitigation of viral transmission. The current scantiness of viral genomic data and underutilisation of advanced molecular analysis methods on influenza viruses circulating in Africa has limited their extensive study and representation in the global influenza ecology. We aimed to sequence influenza type-A viruses (IAVs) that previously circulated in Uganda and characterised their genetic relatedness to the vaccine viruses and publicly available Africa IAVs. Methods This was an observational study nested to the Uganda national influenza surveillance programme. We used Next-generation sequencing to locally generate genomes from 116 A(H1N1)pdm09 and 118 A(H3N2) viruses collected between 2010 and 2018 from 7 districts across Uganda. A total of 206 hemagglutinin (HA), 207 neuraminidase (NA), and 213 matrix protein (MP) sequences were genetically compared to the WHO-recommended vaccines and other viruses isolated from Africa since 1994. Viral temporal and spatial divergence and circulating genetic clades were characterised using phylogenetic methods. Findings We successfully generated gene sequences for 91·9% (215/234) viruses. Uganda A(H1N1)pdm09 and A(H3N2) virus HA, NA, and MP proteins had 96·36-99·09%, 96·49-99·39%, and 97·48-99·95% amino acid similarity, respectively, to vaccines recommended from 2010 through 2020. The local viruses incorporated amino acid substitutions (AAS) in their antigenic, receptor binding, and glycosylation sites each year causing them to antigenically drift away from vaccines. For seasons when vaccine formulations differed, Uganda IAV antigenic sites had 1-2 extra AAS relative to the Southern than Northern hemisphere vaccine viruses. All Uganda IAVs carried the adamantine-resistance marker S31N but not the neuraminidase inhibitor (NAI) resistance markers H274Y and H275Y. However, some A(H1N1)pdm09 viruses had permissive substitutions V234I, N369K, and V241I typical of NAI-resistant viruses. The 2017-2018 A(H1N1)pdm09 viruses belonged to global genetic clade 6B.1, while the A(H3N2) viruses isolated in 2017 belonged to clades 3C.2a and 3C.3a. Uganda IAVs obtained before 2016 clustered distinctly from other Africa viruses while later viruses mixed with other Africa, especially Kenya and Congo, and global viruses. Several unique viral lineages (bootstrap >90) persisted in Uganda and other countries for 1-3 years. Interpretation The study reveals Uganda as part of the global influenza ecology with continuous importation, antigenic drift, and extensive local transmission of IAVs, presenting a potential risk of future outbreaks. For a country with limited health resources and where social distancing is not sustainable, viral prevention by vaccination should be prioritized. The notable viral diversity in Africa is a warning to countries to broaden and incorporate genome analysis in routine surveillance to monitor circulating and detect new viruses. This knowledge can inform virus selection for vaccine production and assist in developing cost-effective virus control strategies. Funding This work was supported by the Makerere University-Uganda Virus Research Institute Centre of Excellence for Infection and Immunity Research and Training (MUII). MUII is supported through the Developing Excellence in Leadership, Training and Science (DELTAS) Africa Initiative (Grant no. 107743). The DELTAS Africa Initiative is an independent funding scheme of the African Academy of Sciences (AAS), Alliance for Accelerating Excellence in Science in Africa (AESA), and supported by the New Partnership for Africa’s Development Planning and Coordinating Agency (NEPAD Agency) with funding from the Wellcome Trust (Grant no. 107743) and the UK Government. The work was also funded in part by a Wellcome Trust grant (102975). Novel influenza type-A viruses (IAVs) cause human respiratory infections that lead to social 75 lockdowns, economic losses, and millions of deaths 1 . Genetic characterisation of IAVs is important 76 to differentiate them from other viruses causing similar clinical symptoms for effective viral control 77 and prevention. Seasonal influenza-related illnesses kill 290000-650000 people globally per year, 78 mostly in sub-Saharan Africa 2 . Influenza accounts for 21·7% and 10·1% of the influenza-like 79 illnesses (ILI) and severe acute respiratory illnesses (SARI) in Africa, respectively, and circulates all- year-round with discernible influenza peaks in North and South Africa 3 . Uganda's annual epidemics 81 have two major peaks between May and November and usually constitute multiple IAV types and 82 subtypes responsible for 13% and 6% of the ILI and SARI cases, respectively 4 . Vaccination and antiviral treatment are the best ways to prevent and control viral transmission 5 . However, the multi-segmented IAVs continuously mutate, especially in the antigenic surface genes, Well-sampled and developed collect sufficient data on IAV evolution patterns, drug sensitivity 6 , and 90 circulating and emerging genetic clades 7 for vaccine selection 5 . However, the high cost of whole- We aimed to explore the feasibility of using Next-Generation Sequencing (NGS) in a low-income 97 setting to generate WGs of Uganda IAVs sampled in 2010-2018 and compare the generated 98 sequences with vaccine viruses and publicly available Africa IAV sequences collected in 1994-2019. We analysed HA carrying antigenic sites which trigger host immune responses and NA and matrix 100 protein (MP) genes targeted by antiviral drugs. This work birthed an East African network of 101 influenza molecular epidemiologists which we hope to expand across Africa. Samples with a broad fragment size spectrum (>250 bp) were normalized manually to 2nM. 5µL per 135 sample library were pooled, denatured using Sodium Hydroxide (NaOH), and diluted to 12.5 pmol. Diluted libraries were spiked with 5% Phi-X control (Illumina, San Diego, CA, USA) and sequenced 137 using the Illumina MiSeq (Illumina Inc., San Diego, California, USA) generating 2x250bp paired 138 reads per sample. We assessed sequencing efficiency based on the read (coverage depth) and gene segment 140 (genome length) count generated per swab sample. The funder of the study played no role in study design, data collection, data analysis, data 189 interpretation, or writing of the report. Sociodemographic characteristics of sampled patients The UVRI-NIC laboratory tested 18353 patients between 22 nd October 2010 and 9 th May 2018. Three A(H1N1)pdm09 and one A(H3N2)] sampled patients' swabs lacked sociodemographic data. Of the 230 swabs with data, 65·22% (150/230) and 34·78% (80/230) were from ILI and SARI cases, The newly-generated sequences were deposited in GISAID under the accessions EPIISL498819- Uganda IAV HA1 proteins continuously drifted away from 2010-2020 season vaccines (Table 1) . In 2012, 2013, 2019, and 2020 when formulations differed, Uganda viruses carried 1-2 extra AAS 240 relative to SH than NH vaccines viruses. Since Uganda's largest part lies north of the equator, the We observed 16 unique amino acid substitutions (UAAS) across the five antigenic sites amongst the 244 107 A(H1N1)pdm09 viruses (Table 1A) . Ranking from the most variable site, Sa, Ca2, Sb, Ca1, and 245 Cb had 4, 4, 3, 3, and 2 AAS, respectively. Substitution S164T, S185T, S203T, H138R, and S74R 246 was most frequent at site Sa, Sb, Ca1, Ca2, and Cb, respectively. Seventy-percent (54/77) and 100% of 2010-2016 viruses carried S185T and S203T, respectively. Ninety-percent (27/30) of the 248 2017-2018 viruses carried S164T and S74R. There were 93 UAAS across the five antigenic sites amongst the 99 A(H3N2) viruses (Table 1B) . We observed mutated receptor binding sites (RBS) (A186T, S190V, and D222E) and glycosylation 256 sites (GS) (A186T and N125D) of A(H1N1)pdm09 viruses previously described 18, 22 . A(H3N2) viruses 257 had more mutated RBS in the 130-loop (135, 138, and 140), 190-helix (186, 192, 194, and 196) , N1 proteins lacked the neuraminidase inhibitors (NAIs) resistance substitution H275Y. However, 266 7·55% (8/106) N1 proteins had T362I (1), I117M (2), Y155H (2), and V234I (3) N2 proteins lacked the NAI-resistance H274Y (N2 numbering) but 17·8% (18/101) carried Y155F (2), E/D221D/K/E (15), and H374N (1) We define a group as a cluster with at least three Uganda sequences in the African gene tree. Uganda A(H1N1)pdm09 viruses collected before 2016 clustered uniquely towards the root, while the 309 2017-2018 viruses mixed with others from Eastern, Central, Western, and Southern Africa (sFigure 16 7-9, Appendix pp 19-21). We observed 9 groups in each A(H1N1)pmd09 gene tree. Forty-four 311 percent (4/9) of the H1 and N1 (2,4, 5, and 6), and MP (2, 4, 5, and 7) were unique to Uganda 312 (sTable 6, Appendix pp 25). Uganda A(H3N2) viruses collected in 2008-2016 and some before April 2017 clustered uniquely 314 closest to the root (sFigure 10-12, Appendix pp 22-24). There were 10, 11, and 15 groups in the H3, 315 N2, and MP gene trees, respectively. Two H3 (1 and 2), one N2 (6), and eight MP (1, 3, 5, 6, 8, 9, Accessions for sequences used in this study are provided in sTable 7 (Appendix pp 26-27). We have demonstrated the feasibility of NGS whole-genome sequencing of IAVs in low- This is the first and largest study to sequence Uganda A(H1N1)pdm09 and A(H3N2) virus 376 WG locally, and internationally to describe a detailed temporal and spatial genetic diversity 19 and evolution patterns of Africa IAVs. Our sequences add significantly to publicly available 378 Africa data. However, we used pre-collected data with a biased geographical sampling. Therefore, we did not consider geographical location in the swab randomisation. We (3), S124N a (1), T128A a (11), T135K a (2), A138S a (6), I140K a (9), R142G a (12), R142K a (5), S144N a (13), S144K a (2), S145N a (2), S146G a (1), (2), T128A a (5), T135K a (2), A138S a (5), I140K a (5), G142K a (1), G142R a (7), S144N a (7), S145N a (2), S146G a (1), (2), T128A a (5), K131T a (13), T135K a (2), A138S a (5), I140K a (5), K142G a (5), K142R a (7), S144N a (7), S145N a (2), S146G a (1), (1), S124N a (1), T128A a (5), I140K a (1), R142G a (6), N145S a (6), N91S e (13), A128T a (8), S138A a (8), G142K a (1), G142R a (7), K144N a (7), K144S a (6), S159F b (2), S159Y b (6), S193F b (13) (2), T128A a (5), K131T a (13), T135K a (2), A138S a (5), I140K a (5), G142R a (7), G142K a (1), S144N a (7), S145N a (2), S146G a (1), (1), S91N e (1), S124R a (1), S124N a (1), A128T a (6), S138A a (13), R140K a (6), R140I a (8), G142K a (2), G142R a (4), N144K a (1), N144S a (6), L157S b (1), S159F b (7), S159Y b (6), (14) A128T a (8), K135T a (11), F137S a (13), S138A a (8), T160K b (7), S193F b (13), D225N d (2), SH= Southern hemisphere vaccine. NH= Northern hemisphere vaccine. SNH= vaccine shared for both Southern and Northern hemispheres. A(H1N1)pdm09 HA1 antigenic sites are Sa= a, Sb =b, Ca1=c, Ca2 = d, Cb= e as described by Caton et al., 1982 . A(H3N2) HA1 antigenic sites are A=a, B=b, C=c, D=d, and E=e as described by Wiley and Skehel, 1987 . The frequency of an amino acid substitution observed at a given antigenic site in Uganda HA1 proteins is reported as (n). The substitutions in BOLD are shared against the SH and NH vaccines of that given year. # = amino acid insertion. Mutations against the NH vaccine strains are colored yellow for visualization purposes only. Recommended vaccines for each influenza season are adopted from WHO (https://www.who.int/influenza/vaccines/virus/recommendations/en/) H156Q b (1), L157S b (1), Y159F b (7), Y159S b (6), K160T b (14), N171K d (7), P194L b (27), A196T b (2), Q197R b (5), S198P b (7), S198A b (2), X203T d# (27), K207Q d (1), I214T d (3), N225D d (7), S262N e(1)H156Q b (1), Y159F b (2), Y159S b (5), K160T b (7), K171N d (8), A196T b (2), Q197R b (1), S198P b (5), S198A b (2), I214T d (2), D225N d (2), G275S c (1), K276R c (1), K278N c (2),H311QH156Q b (1), Y159F b (2), Y159S b (5), K160T b (6), N171K d (5), P194L b (13), A196T b (2), Q197R b (1), S198P b (5), S198A b (2), I214T d (2), D225N d (2), Q261R e (13), G275S c (1), K276R c (1), K278N c (2),H311QQ156H b (6), L157S b (1), V186G b (6), Q197H b (1),I214TH156Q b (1), Y159F b (2), Y159S b (5), K160T b (6), K171N d (8), I186G b (13), A196T b (2), Q197R b (1), S198A b (2), S198P b (5), I214T d (2), F219S d (13), D225N d (2), G275S c (1), K276R c (1), K278N c (2), Q311H c (3),K160T b (6), M168V a (2), N171H d (1), V186G b (14), Q197R b (2), I214T d (1), X219S d# CLUSTER '0' '119' '15' '25' '39' '92' '98 EBB5305_A_HA-H3_Uganda_UVRI_Entebbe_022_2015-08-19_cluster35 MBA0351_A_HA-H3_Uganda_UVRI_Mbarara_001_2012-11-19_cluster35 ARU1409_A_HA-H3_Uganda_UVRI_Arua_001_2014-11-07_cluster116 EBB5301_A_HA-H3_Uganda_UVRI_Entebbe_021_2015-08-18_cluster35 FTL1327_A_HA-H3_Uganda_UVRI_Fortportal_004_2015-11-09_cluster116 KSW3436_A_HA-H3_Uganda_UVRI_Kawaala_020_2013-01-22_cluster35 EBB2907_A_HA-H3_Uganda_UVRI_Entebbe_010_2011-11-3_cluster13 KSW5016_A_HA-H3_Uganda_UVRI_Kawaala_010_2016-07-29_cluster35 KSW3397_A_HA-H3_Uganda_UVRI_Kawaala_019_2012-11-26_cluster35 KSW4413_A_HA-H3_Uganda_UVRI_Kawaala_006_2014-11-17_cluster35 EBB2897_A_HA-H3_Uganda_UVRI_Entebbe_009_2011-10-31_cluster13 FTL0878_A_HA-H3_Uganda_UVRI_Fortportal_003_2014-04-16_cluster35 KSY0882_A_HA-H3_Uganda_UVRI_Kisenyi_002_2011-10-10_cluster13 FTL1378_A_HA-H3_Uganda_UVRI_Fortportal_005_2015-11-30_cluster97 KIS1073_A_HA-H3_Uganda_UVRI_Kitebi_004_2012-12-17_cluster35 EBB2778_A_HA-H3_Uganda_UVRI_Entebbe_006_2011-09-28_cluster13 EBB3737_A_HA-H3_Uganda_UVRI_Entebbe_013_2013-03-11_cluster35 KSW4850_A_HA-H3_Uganda_UVRI_Kawaala_008_2016-04-27_cluster35 EBB4376_A_HA-H3_Uganda_UVRI_Entebbe_014_2013-12-03_cluster35 TOR1664_A_HA-H3_Uganda_UVRI_Tororo_003_2015-11-10_cluster35