De novo emergence of the mutation E484K in a SARS-CoV-2 B.1.1.7 lineage variant

Since its first description in December 2020¹, the SARS-CoV-2 variant of concern VOC-202112/01 (also known as lineage B.1.1.7, 20I/501Y.V1 or, recently, according to the WHO, simply alpha²) has been spreading all over the world. In our geographical area, it became predominant from the beginning of March 2021, accounting for more than 90% of new infections in June 2021³. This is mainly due to the mutations that this variant accumulates in the gene that encodes the spike (S gene), especially in the receptor-binding domain (RBD). These mutations, especially the N501Y mutation that it shares with the beta (B.1.351 or 20H/501Y.V2) and gamma (P.1 20J/501Y.V3) variants of concern, among others, are related to an increased binding affinity of the spike with angiotensin-converting enzyme II and an increase in transmissibility⁴. These last two variants also share the E484K mutation, also in the RBD, which could be related to a certain degree of escape from the action of the vaccines⁵. Therefore, when the first sequences of the B.1.1.7 lineage with the E484K mutation emerged, the British authorities declared them variants of concern VOC-202102/02⁶. However, the cluster in which these variants were framed does not seem to have been as successful and represents only 0.225% of the sequences of the B.1.1.7 lineage included in the GISAID (Global Initiative on Sharing All Influenza Data).

In accordance with our centre’s variant screening protocol, all SARS-CoV-2 positive samples with Ct <32 are tested using the Allplex™ SARS-CoV-2 Variants I Assay kit (Seegene, Korea), which simultaneously detects H69/V70, E484K and N501Y mutations. At the end of April, we identified a sample which was positive for all three targets studied. In addition to the sample from the patient who had all three mutations, samples from the other three positive cases from the family cluster of which the patient had been a close contact were sequenced, presenting a profile compatible with the B.1.1.7 lineage but without the E484K mutation. For the sequencing of the viral genome, the Ion AmpliSeq SARS-CoV-2 Research Panel (Thermo Fisher Scientific, USA) was used⁷. Libraries were prepared following the manufacturer’s instructions and loaded onto a 540 chip and the Ion GeneStudio™ S5 (Thermo Fisher Scientific, USA) platform. The genome was assembled using the IRMA plugin⁸ and its consistency was checked using the Integrative Genomics Viewer (IGV) program⁹. The Nextstrain webApp¹⁰ was also used for both clade assignment and visualisation of mutations. The sequences obtained by next-generation sequencing confirmed the results of the PCR variant screening techniques. The test sample presented the G23012A mutation that conditions the amino acid change in the E484K S gene. None of the other three samples had that mutation in the assembly, and the A readings at position 23012 accounted for less than 1% of the readings in each of the samples. Except for the G23012A mutation, the four strains of the family cluster were identical and had both the characteristic mutations of the B.1.1.7 lineage and some of their own (see Table 1).

Table 1.

Coverage of the mutations present in the family cluster.

	Test sample	Relative 1	Relative 2	Relative 3
C241T	2935	7356	4892	3110
C913T	5499	10533	7634	5110
C3037T	6899	10817	8752	8597
C3267T	1870	6997	4489	3645
C4464T	7318	13962	8950	5797
A5041G	3407	6980	4543	2567
C5388A	4309	8448	5309	3474
G5763A	7673	12951	9983	10079
C5986T	2074	3458	1213	449
T6954C	2822	7202	3136	933
11288-11297	6911	11107	8345	7157
C11668T	5339	7078	3973	2795
C12439T	6903	11349	7622	6920
C14407T	4004	6209	3489	2521
C14408T	4006	6230	3496	2525
C14676T	5486	8911	6115	4265
T15096C	3185	5267	3462	3199
C15279T	1811	6655	4096	2838
T16176C	6050	11721	8117	9837
C18647T	2354	6501	4085	3055
21766-21772	5587	6261	5807	5072
21994-21997	7153	7985	6569	5075
G23012A	8771
A23063T	2137	5422	4300	1691
C23271A	4375	7485	6071	5999
A23403G	13281	16004	13387	13398
C23604A	9736	12985	11754	14791
C23709T	7735	11037	9594	7600
T24506G	14234	17157	15581	13242
G24914C	9398	14695	11772	7804
C27972T	25717	28812	39283	27683
G28048T	25829	28835	39267	27589
A28095T	12218	16000	19763	6728
A28111G	12192	15990	19727	6687
28274	9517	11135	18433	8315
G28280C	10624	12112	19790	8916
A28281T	10686	12175	19865	8933
T28282A	10689	12181	19875	8936
C28320T	10853	12294	20071	9036
G28881A	6831	10521	12661	5721
8882A	6840	10531	12673	5728
G28883C	6841	10531	12673	5728
C28977T	6715	10284	12336	5378

Despite belonging to the B.1.1.7 lineage and also exhibiting the E484K mutation, the sequence of our sample did not share the rest of the characteristic mutations of VOC-202102/02. The appearance of this mutation was therefore probably independent to those found in February 2021 in the United Kingdom, similar to other synchronous appearances of this mutation. In this case, the epidemiological chain of transmission is quite clear, since the infection of the patient with the E484K mutation appeared later: one week after the rest, and when the patient had been in isolation for six days for being a close contact of a confirmed case, so this mutation probably emerged de novo, either in the patient himself or in any of the other three members of the cluster after taking their respective samples. Fortunately, the patient had no subsequent close contacts and no new variants belonging to the B.1.1.7 lineage with the E484K mutation have been observed in the daily screenings carried out in our health organisation.