Molecular Epidemiology and Evolutionary Dynamics of Human Influenza Type-A Viruses in Africa: A Systematic Review

Genomic characterization of circulating influenza type-A viruses (IAVs) directs the selection of appropriate vaccine formulations and early detection of potentially pandemic virus strains. However, longitudinal data on the genomic evolution and transmission of IAVs in Africa are scarce, limiting Africa’s benefits from potential influenza control strategies. We searched seven databases: African Journals Online, Embase, Global Health, Google Scholar, PubMed, Scopus, and Web of Science according to the PRISMA guidelines for studies that sequenced and/or genomically characterized Africa IAVs. Our review highlights the emergence and diversification of IAVs in Africa since 1993. Circulating strains continuously acquired new amino acid substitutions at the major antigenic and potential N-linked glycosylation sites in their hemagglutinin proteins, which dramatically affected vaccine protectiveness. Africa IAVs phylogenetically mixed with global strains forming strong temporal and geographical evolution structures. Phylogeographic analyses confirmed that viral migration into Africa from abroad, especially South Asia, Europe, and North America, and extensive local viral mixing sustained the genomic diversity, antigenic drift, and persistence of IAVs in Africa. However, the role of reassortment and zoonosis remains unknown. Interestingly, we observed substitutions and clades and persistent viral lineages unique to Africa. Therefore, Africa’s contribution to the global influenza ecology may be understated. Our results were geographically biased, with data from 63% (34/54) of African countries. Thus, there is a need to expand influenza surveillance across Africa and prioritize routine whole-genome sequencing and genomic analysis to detect new strains early for effective viral control.


Definitions and Outcomes
We defined a subtype as a group of viruses with a specific combination of HA and NA genes, for example, H1N1 and H3N2 [7]. A viral lineage is a group of genetically related viruses of the same subtype sharing a common ancestor. A strain is a virus exhibiting minor mutations different from other viruses of the same lineage or subtype.
A phylogeny is a tree-like structure reconstructed using viral gene sequences showing evolutionary relationships between the sampled viruses from a common ancestor. The tree has internal nodes, branches, and tips representing ancestors, genetic distances from the nearest ancestors, and sampled viruses, respectively.
Phylogenetic clusters and clades are often interchanged [31]. Here, we defined a cluster as a subtree of viruses sharing a common ancestor and may share epidemiological linkage [32]. In contrast, a clade is a cluster of viruses with characteristic amino acid substitutions in their HA proteins subunits (HA1 or HA2) [27]. Phylodynamics is the study of evolutionary processes that shape the phylogenies, such as changes in viral effective population sizes [Ne(t)], evolutionary rate, basic reproductive number (R 0 ), selection pressures, and reassortment [25,33]. Additional definitions are provided in Table S1.

Screening & Eligibility First Identification
Articles after duplicates removed (n= 947)

Articles screened (n = 947)
Excluded articles irrelevant to review's subject and objectives (n = 871) Articles assessed for eligibility (n = 76) Articles excluded (n = 16): 12 conference abstracts, 1 case report, 2 preprints, and 1 study done in American soldiers in Djibouti Articles included in the final analysis (n = 71)

Included
Articles assessed for eligibility (n = 71)

Deduplication
Duplicated articles excluded (n = 390 ) A total of 87% (62/71) of the studies generated and analyzed new viral sequences, and 13% (9/71) analyzed online sequences. Of the studies that generated new sequences, 53% (33/62) and 37% (23/62) sequenced IAVs locally and abroad, especially in the United Kingdom (U.K.) and the United States (U.S.), respectively. Two studies sequenced IAVs both locally and abroad, and four studies did not report where sequencing was performed.
A total of 56% (35/62) of the studies used Sanger sequencing technology and were published between 1996 and 2020. Only 8% (5/62) of the studies used next-generation sequencing (NGS) and were published between 2011 and 2021 [75,76,81,97,98]. Two studies used both Sanger and NGS [26,73]. Another study did not clearly report whether Sanger or NGS was used [94]. Three studies used pyrosequencing or Sanger and pyrosequencing or Sanger and targeted HA analysis. Sixteen studies did not describe the technology used.

Niger (5)
Nigeria ( The geographical distribution of studies that did genomic characterization of influenza type-A viruses (H1N1, H1N1pdm09, and H3N2) sampled in Africa. Each country is highlighted based on the absolute number of studies that analyzed sequences of influenza viruses collected from that country. For each country, the study count also includes any study that included at least one sequence from that country in their virus clade classification using the European Center for Disease Control (ECDC) guidelines [27]. Abbreviations: CAR = Central African Republic. Countries not shown: Cape Verde (n = 1), Reunion (n = 3), Seychelles (n = 4), Mauritania (n = 1), Mauritius (n = 10), and Mayotte (n = 1).

Phylogenetic Clustering and Circulating Clades among H1N1pdm09 Viruses
Phylogenies showed that Africa H1N1pdm09 virus strains were variants of the A/California/07/2009(H1N1) strain that diverged into multiple clades that co-circulated every season. Analysis of concatenated whole viral genomes (concat-8) formed similar but deeper phylogenies (with a higher number of nodes between root and furthest tip) than those of individual genes [54,55]. Pandemic H1N1pdm09 viruses sampled from different African countries mixed. While sH1N1pdm09 strains clustered distinctly according to time and location (household or province or country) of sampling [47,53,55,60,62,102,103].   Table S9.
Despite the multiple introductions and early diversification of H1N1pdm09 strains, two highly supported viral clusters (I and II, bootstrap > 80%) persisted for >1.8 years in Western Africa between 2010 and 2012 [60]. Cluster I contained strains with H1 substitutions S145T and R276K that circulated in Senegal, Ghana, and Ivory Coast between March 2011 and January 2013, similar to a France-origin strain sampled earlier in 2010. The largest cluster II contained both Western Africa (n = 26, including the A/Ghana/763/2011(H1N1)-clade 8 strain) and non-Western Africa (Ethiopia, France, Italy, Norway, Sweden, and Minnesota, n = 11) strains with H1 substitutions A15T, N490D, T491K, and V537, sampled during the November 2010-October 2012 and November 2012-January 2013 seasons, respectively. All cluster II non-Western Africa strains clustered monophyletically (bootstrap = 100%) [60]. Phylogeographic analysis by Owuor using the Bayesian Ancestral state reconstruction (BSSVS) confirmed the introduction of H1N1pdm09 strains into Africa between 2009 and 2018 from outside Africa, with the highest viral migration rate of 1.48 (Bayes factor, BF ≥ 3) from East and Southeast Asia [98]. Introduction events were inferred as clusters with bootstrap ≥ 80%, size of n ≥ 2 strains, and >80% of strains sampled from Africa. The BSSVS model estimated strongly supported H1N1pdm09 viral migration from Northern to Southern Africa (BF ≥ 1000), between countries (100 ≤ BF < 1000), and within the country (Kenya) (rate = 0.81-0.93, BF ≥ 1000) [98]. Overall, H1N1pdm09 strains sampled in Africa belonged to similar globally recognized clades. However, most clades first circulated elsewhere before their introduction in Africa, except clades 8 and 9 were unique to Africa ( Figure S10).

Population Dynamics, Evolutionary Rates, Selection, and Reassortment among H1N1pdm09 Viruses
Upon viral introduction in Kenya, the population sizes (genomic diversity) of H1N1pdm09 strains fluctuated continuously, with the highest peaks observed during the 2009-2010 and 2018 seasons [98].
Using a relaxed molecular clock, concatenated whole genomes of Kenya pH1N1pdm09 strains were estimated to evolve at a mean rate of 4.9 × 10 −3 (2.6-7. respectively, exhibited lower rates of nonsynonymous (dN) than synonymous substitutions (dS) with dN/dS ratio of 0.6-0.8 in their HA1 genes, confirming purifying (negative) selection [61,66]. Single likelihood ancestor counting (SLAC) and fixed-effects likelihood (FEL) predicted no codon site under selection at a p-value <0.05 [61].
South Africa and Cameroon H3N2 strains sampled between 2009 and 2016 also continuously evolved from vaccine strains in their N2 and MP genes [78,104].

Reassortment between H1N1pdm09 and H3N2 Viruses and Zoonotic Exchanges
Time-resolved phylogenies showed the cocirculation of multiple IAV lineages and clades each year in Africa as observed globally [130,131]. Africa IAVs phylogenetically mixed with earlier global strains, suggestive of viral introductions into Africa from the rest of the world. Circulating strains in a given year replaced strains in the previous year. However, some H1N1pdm09 lineages persisted through consecutive seasons in the United Kingdom [132], China [113], and Western Africa [60]. Phylogeographical and tree trunk analysis confirmed significant viral migration from South Asia and Europe to Africa (Bayes Factor > 1000) but low migration rates from Africa to elsewhere [21,98,112,133]. Another study by Chan et al. fitted a probabilistic model on the global H1N1 and H3N2 virus HA1 sequences and confirmed Europe and North America as the main importers of IAVs to Africa [86]. Together, the studies confirm Africa as a sink for new IAVs. However, this pattern could be an artifact of inconsistent and insufficient viral sampling in Africa [134].
Various automated methods based on phylogenetic incongruence, TMRCAs, genetic distances, and coalescent modeling have detected intra-and inter-subtype reassortment among global IAVs [26,113,138,139,[149][150][151][152][153]. Global H3N2 strains reassorted more at a rate of 0.3-0.55 events per lineage per year than the H1N1pdm09 (0.1-0.45) [151]. However, reassortment among Africa IAVs has largely been inferred manually based on viral sequences occupying different positions between phylogenies of two or more gene segments [58,60,73,78]. Owuor used the automated reassortment detection tool (GiRaF) and identified no reassortant H3N2 strains in Kilifi (coastal Kenya) between December 2015 and December 2016 [98]. Contrastingly, our ongoing GIRAF analysis of~1500 Africa virus whole genomes sampled between 1994 and 2020 identified~573 reassortants, some of which circulated in Kilifi [154]. Such differences could be due to different sampling periods, given that the number of reassortants observed depends on the number of genomes sampled (genomic diversity) [150,154]. Inter-subtype reassortment events observed in Africa between 2009 and 2013 involved human-avian and human-human viral gene exchanges [72,100,105]. However, resulting reassortants could have been less fit to circulate [155]. Africa and global H1N1pdm09 strains sampled in humans and swine mixed phylogenetically indicative of a bidirectional zoonosis and continuous evolution of H1N1pdm09 strains among swine worldwide [42].

Conclusions
Phylogenetic and phylogeographic analyses showed that multiple introductions of new strains from outside Africa and extensive local viral transmission sustain the high genomic diversity, continuous antigenic drift, and persistence of influenza viruses among African human populations. Circulating strains had several new amino acid substitutions affecting major antigenic and N-linked glycosylation sites in their hemagglutinin (HA) proteins, which could dramatically affect antibody recognition among vaccinated and exposed unvaccinated populations. The role of reassortment and zoonosis in the evolution and diversification of IAVs in Africa needs to be determined. We observed substitutions and clades and persistent viral lineages unique to Africa. Therefore, Africa's contribution to the global influenza ecology should not be understated. Thus, there is a need to expand influenza surveillance across Africa and prioritize routine whole-genome sequencing using modern NGS technology and genomic analysis to monitor circulating strains and early detection of emerging ones. Such knowledge could inform public health policies and appropriate vaccine development and selection.
Supplementary Materials: The following supporting information can be downloaded at: https: //www.mdpi.com/article/10.3390/microorganisms10050900/s1, Table S1: Terms and definitions; Table S2: Keywords and their corresponding mesh terms included in the search strategy; Table S3: Study article inclusion and exclusion criteria; Table S4: Study quality and risk-of-bias assessment scheme; Table S5: List of all African countries with or without viral sequence data analysed in the included studies; Table S6: Study quality and risk of bias assessment;  Figure S10: Viral diversification and distribution of genetic clades among H1N1pdm09 viruses that circulated in Africa versus elsewhere during the 2009-2020 seasons; Table S11