We used 378 SARS-CoV-2–positive diagnostic respiratory samples tested at the reference laboratory in Benin during January 30–April 2, 2021, for genomic surveillance. All samples with cycle threshold <36 (Sarbeco E-gene assay; TIB Molbiol, https://www.tib-molbiol.de) were used for this study. To enable rapid prescreening of mutations known to affect the viral phenotype, we used 4 reverse transcription PCR (RT-PCR)–based single-nucleotide polymorphism (SNP) assays (VirSNiP; TIB Molbiol) targeting 9 hallmark mutations in 7 S codons of variants of concern (VOCs): B.1.1.7 (Alpha), B.1.351 (Beta), P.1 (Gamma), and B.1.617.2 (Delta) (Table 1). A total of 374 (98.9%) samples selected for the study tested positive for >1 mutation. Of those, ≈67.5% (255/378) showed the 69/70 deletion, 58.9% (223/378) the E484K mutation, 33.9% (128/378) the N501Y mutation, 30.4% (115/378) the P681H mutation, 14.8% (56/378) the L452R mutation, and 0.3% (1/378) the K417N or P681R mutation. The K417T or V1176F mutations associated with the Beta and Gamma VOCs were not detected. Approximately 22.2% (84/378) of samples were typeable to 1 of the lineages covered by the VirSNiP assays. According to SNP-based analyses, 14.8% (56/378) of the overall samples showed the mutation pattern of the Alpha variant, B.1.1.7, and 7.4% (28/378) of the B.1.525 variant. Frequent occurrence of the mutations under study suggests that earlier SARS-CoV-2 lineages not carrying those mutations have been replaced in Benin.

Figure 1

Figure 1. Genomic surveillance of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) lineages in Benin, 2021. A) Nonsynonymous mutations of Benin-derived SARS-CoV-2 sequences across the full genome. B) Spike mutations occurring in...

Definite lineage designation relies on the full genome sequence. We selected 68 (9 typeable and 59 nontypeable) samples according to unique mutational patterns covering the complete period of the study for a NimaGen/Illumina-based whole-genome sequencing workflow (Appendix). All near-full genomes generated within this study were deposited into GISAID (https://www.gisaid.org; accession nos. EPI_ISL_2932532–84 and EPI_ISL_2958658–72). Lineage assignment using the Pangolin COVID-19 Lineage Assigner version 3.0.2 (https://pangolin.cog-uk.io) confirmed SNP-based lineage prediction in all 9 typeable samples selected for whole-genome sequencing (Appendix). Despite robust lineage prediction based on unambiguous SNP-based results, our data demonstrate the limited use of VirSNiP assays for strain designation; however, these assays can detect relevant mutations of currently circulating variants. The 68 Benin-derived near-complete genomes were designated to 10 unique lineages, suggesting higher genetic diversity in Benin than ≈1 year before (3). During early 2021, lineages B.1.1.7 (22%), A.27 (19.1%), B.1.525 (17.6%), and B.1.1.318 (16.2%) were most prominent in Benin (Appendix). Despite presence of the mutation P681R (associated with the Delta VOC) in 1 sequence, that strain was typed as A.23.1, and no Delta variant was found. These data are consistent with recent online sequence reports from West Africa (A.E. Augustin, unpub. data, https://www.medrxiv.org/content/10.1101/2021.05.06.21256282v1; E.A. Ozer et al., unpub. data, https://www.medrxiv.org/content/10.1101/2021.04.09.21255206v3). A 100% consensus sequence of all 68 Benin-derived sequences showed 229 nonsynonymous nucleotide substitutions across the whole genome; 57 (24.9%) occurred in the S protein (Figure 1, panel A). Of note, variants with mutations in the S protein might alter the transmissibility and antigenicity of the virus (4). Internationally recognized VOCs to date share 16 S mutations in unique combinations (https://covariants.org/shared-mutations). The Benin-derived SARS-CoV-2 strains shared 10 unique S mutations reported in VOCs, although most of those strains were not defined as any VOC other than Alpha (Figure 1, panel B), suggesting convergent evolution of key mutations across different lineages (D.P. Martin et al., unpub. data, https://www.medrxiv.org/content/10.1101/2021.02.23.21252268v3; S. Cherian, unpub. data, https://www.biorxiv.org/content/10.1101/2021.04.22.440932v2). Putative higher fitness mediated by genomic change was consistent with more mutations in predominant lineages than in lineages found at lower frequencies (Figure 1, panel B).

Figure 2

Figure 2. PRNT results of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) variants from Benin, 2021. Graphs compare results of neutralization tests for naturally infected persons (A) and persons who received the...

Because S mutations, individually or in combination, have been shown to afford viral escape to antibody-mediated immune responses, the high prevalence of variants with large numbers of these mutations circulating in Benin was cause for concern. To investigate whether and to what extent SARS-CoV-2 variants circulating in Benin and West Africa (5) evade neutralizing antibody responses, we isolated 4 lineages with unique mutational patterns (Table 2): an A.27 lineage isolate harboring the N501Y mutation; a B.1 isolate harboring the 69/70 deletion and the E484K and D614G mutations; a B.1.1.7 lineage isolate harboring the 69/70 deletion and the N501Y, D614G, and P681H mutations; and a B.1.214.2 lineage harboring the Q414K and D614G mutations (Figure 2). Additional isolation attempts of strains belonging to the frequently detected B.1.525 and B.1.318 lineages failed, likely because of degradation after repeated freeze-thaw cycles under tropical conditions. We tested neutralization potency of 6 serum samples from patients in Benin taken ≈8 days after RT-PCR–confirmed SARS-CoV-2 infection during early 2020 (6) and another 7 serum samples from persons in Europe 4 weeks after receiving the second dose of the Pfizer/BioNTech vaccine (BNT162b2; https://www.pfizer.com) (Appendix). Sampling was approved by the ethics committee of the Benin Ministry of Health (approval no. 030/MS/DC/SGM/DNSP/CJ/SA/027SGG2020) and of Charité-Universitätsmedizin Berlin (approval nos. EA1/068/20 and EA4/245/20). We compared neutralization titers with a SARS-CoV-2 strain (B.1.153) from January 2020 and the Beta strain (B.1.351), known to evade antibody-mediated neutralization (7). Despite the early sampling time after RT-PCR confirmation of SARS-CoV-2 infection, all 6 serum specimens from patients in Benin efficiently neutralized the early SARS-CoV-2 isolate carrying only the D614G mutation. In contrast, only 3 of those 6 serum specimens neutralized the B.1 isolate, the only isolate with the E484K mutation (Figure 2, panel A). Among the serum specimens from vaccinated persons, all neutralized the B.1 isolate, albeit at 1.5-fold lower titers than the early lineage B.1.153 isolate (by Friedman test and Dunn’s multiple comparisons test; p>0.99) (Figure 2, panel B). Those data were consistent with a recent report describing efficient neutralization of a B.1.525 strain from Nigeria by vaccinee-derived serum specimens (8). Of note, another strain classified as B.1.214.2 was neutralized more efficiently than all other tested lineages (Figure 2), highlighting that not every mutation in circulating lineages affords reduced antibody-mediated neutralization. Other hypothetically present fitness advantages of such strains will require detailed virologic investigation.

Our study is limited by patient-derived samples taken an average of 8 days after infection (7), which could imply incomplete maturation of antibodies. However, similar neutralization patterns between patient-derived and vaccinee-derived serum specimens suggest robustness of our data. Another limitation is that vaccinee-derived serum samples originated exclusively from Europe. Vaccine responses vary between populations, possibly influenced by genetic background and immune-modulating diseases (e.g., malaria or HIV) (9), highlighting the importance of testing serum samples from vaccinees in Africa for future studies. Of note, the efficacy trial of the Pfizer/BioNTech vaccine enrolled ≈40,000 participants, only ≈800 of whom were from Africa, and all of those from South Africa (10).

Our data highlight the importance of ongoing monitoring of population immunity to emerging SARS-CoV-2 variants in Africa and of using serum specimens from local settings for phenotypic characterizations. Vaccination programs in Africa should be accelerated urgently, emphasizing the importance of global access to vaccines.