Genomic architecture and evolutionary relationship of BA.2.75: A Centaurus subvariant of Omicron SARS-CoV-2

In this study, we explored the genomic architecture and phylogenomic relationship of BA.2.75, a subvariant of Omicron SARS-CoV-2. A set of 1468 whole-genome sequences of BA.2.75, submitted by 28 countries worldwide were retrieved from GISAID and used for finding genomic mutations. Moreover, the phylogenetic analysis of BA.2.75 was performed by using 2948 whole-genome sequences of all sub-variants of Omicron along with the Delta variant of SAS-CoV-2. We detected 1885 mutations, which were further grouped into 1025 missense mutations, 740 silent mutations, 72 mutations in non-coding regions, 16 in-frame deletions, 02 in-frame insertions, 8 frameshift deletions, 8 frameshift insertions and 14 stop-gained variants. Additionally, we also found 11 characteristic mutations having a prevalence of 81–99% and were not observed in any of the previously reported variant of SARS-CoV-2. Out of these mutations K147E, W152R, F157L, E210V, V213G, G339H were found in the NTD, and G446S & N460K in the RBD region of the Spike protein, whereas S403L and T11A were present in the NSP3, and E protein respectively. The phylogenetic relationship of this variant revealed that BA.2.75 is descended from the Omicron sub-variant BA.5. This evolutionary relationship suggests that the surge of BA.5 infections can reduce the severity of the infections accredited to BA.2.75. These findings would also improve our knowledge and understanding that how genetic similarities in different variants of SARS-CoV-2 can prime the immune system to fight off the infection caused by one subvariant, after defeating the other.


Introduction
The Omicron variant of SARS-CoV-2 was first identified in Botswana, and South Africa in the early November 2021. On November 24, 2021, it was reported to the World Health Organization (WHO) and eventually declared as a variant of concern (VOC) on November 26, 2021 [1]. Though Omicron replicates about 70-times faster than the delta variant in the bronchi of infected individuals, but there are some evidences that it is less severe than previous strains/ variants, especially when compared to the delta variant [2]. However, it continued to spread a1111111111 a1111111111 a1111111111 a1111111111 a1111111111 the resulting file was left with a set of 1468, high coverage, genomic sequences GISAID considers genomes with length of >29,000 nucleotides as complete and assigns them the high coverage label when there are less than 1% of undefined bases. For the remaining 1468 good quality genome sequences, all the relevant information including their unique identifiers, collection & submission date, and submitting lab information are shown in supplementary materials (S1 Table). The aligned and filtered sequence file was used as an input to the Coronapp web application to obtain the nucleotide variations [14]. All the mutations found in the structural, nonstructural, and accessory proteins were mapped on the SARS-CoV-2 genome by using the Corona Antiviral Research Database (CoV-RDB) [15]. Additionally, the CoV-RDB [15] was also used for the comparison of BA.2.75 mutations with other sub-variants of Omicron SARS-CoV-2. The amino acid sequence of Spike protein was downloaded from the GenBank [16]. However, the three-dimensional (3D) structures of Spike protein used in this work was extracted from the Protein Data Bank (https://www.rcsb.org/), denoted as 6VYB, and its 3Dstructured graph highlighting mutations was developed by using PyMOL [17].
For the construction of the phylogenetic tree, 2948 whole-genome sequences of different variants of SARS-CoV-2 retrieved from the GISAID database were used. And the phylogenetic tree was constructed by using the NextStrain's Augur pipeline [18]. Sequences were again aligned to the SARS-CoV-2 reference genome (NC_045512.2) by using MAFFT [19] and a time-resolved phylogenetic tree was constructed with IQ-Tree [20] and TreeTime [21] under the generalized time reversible (GTR) substitution model [22] and was visualized with auspice [18].

Discussion
Here we present the mutations present in the genes/proteins and in the non-coding regions of BA.2.75; a subvariant of Omicron SARS-CoV2. And relate them with the in-vitro activity of authorized MAbs against the variants reported in this study based upon the published literature. In N-terminal domain of the Spike protein, 11 mutations including T19I, L24del, A27S, G142D, K147E, W152R, F157L, I210V, V213G, G257S, and G339H were observed. All of these mutations except T19I, L24del, and G142D and A27S have very high (>80%) prevalence due    (Table 3). However, the T19I (95%) mutations is also present in BA.2, BA.4 and BA.5 while G142D (56%) mutation is present in all other subvariants of Omicron SARS-CoV-2 (Table 3) [15]. Likewise, in the RBD region of the Spike protein, following 17 mutations S371S, S373P, S375F, D405N, K417N, N440K, S477N, T478K, E484A, Q498R, N501Y, Y505H and D574V were detected. Out of which following two missense mutations, G446S and N460K are only present in BA.2.75 subvariant with a frequency of 93% and 94% respectively, and, therefore, are most likely to be the characteristic mutations of this variant. However, determination of the extent to which the mutations in different variants of SARS-CoV-2 can reduce the monoclonal antibodies (MAbs) susceptibility is critical in the prevention and treatment of COVID-19 [25]. The G446S mutation in B.2.75 is reported to be associated in creating the high-level of resistance to imdevimab but not to cilgavimab [26]. In addition, D405N mutation (97% incidence) was also observed in the following Omicron variant BA.2, BA.4, and BA.5 (Table 3), and reported to reduce the susceptibility of these variants to etesevimab (16 to 26-fold) and casirivimab (11 to 14-fold) [3]. Four other missense mutations including N440K, E448A, S477N, and T478K which have 92% prevalence in BA.2.75 Omicron and are also known as the common RBD mutations. It is noteworthy that these mutations have a gradual increase in their prevalence since the start of the pandemic and are still present in all the Omicron SARS-CoV-2 variants (Table 3). Additionally, S371F, N501Y, and Y505H had 91% prevalence in BA.2.75, and are present in the RBD region of Spike protein. The S371F is an RBD core mutation which is also present in the Omicron BA.2, BA.4, and BA.5 variants. It had been reported to drastically reduced the susceptibility of these

PLOS ONE
variants to etesevimab (143 to 630-fold), casirivimab (14 to 28-fold), imdevimab (11 to 126-fold), sotrovimab (5.5 to 21-fold), and tixagevimab (6.3 to 31-fold) [3]. However, the N501Y is present in alpha, beta, gamma, and all variant of Omicron SARS-CoV-2 and reported to mediate the viral entry by enhancing the binding ability of spike protein with ACE-2 receptors [27]. Likewise, S373P, T376A, K417N, and Q498R are present in the RBD region and had 88-89% prevalence in BA.2.75 and are present in all variant of omicron SARS-CoV-2 except K417N which is in RBM mutation and bind with ACE-2 and is also present in Beta, Gamma, and some other Omicron variants. In addition to RBD mutation, D614G, H655Y and N679K in SD2 region had 100% prevalence while P681H mutation which was located proximal to the S1/S2 furin cleavage site had 99% incidence. The H655Y is involved in increasing the Spike protein cleavage and replication and it is present in Gamma and Omicron variants and many other lineages of SARS-CoV-2. And the D614G, and P681H are the among the preliminary mutations which are present in Alpha, Beta, and each of the Omicron variant [24,28]. The increase in positive charge associated with this mutation is appeared to influence virus tropism by increasing S1/S2 cleavage in human airway epithelial cells [29]. Moreover, N764K, and D796Y in ACE2 region and Q954H, & N969K in HR1 region have 95-96% and 96-97% prevalence respectively and are known as the most frequent mutations in all the variants of Omicron SARS-CoV-2 [30,31].
Among the non-structural proteins of BA.2.75 Omicron SARS-CoV-2, T24I, S403L, and G489S, in NSP3; T492I, L264F, L438F in NSP4; P132H in NSP5; P314L, G662S in RdRp protein; R392C in NSP13; I42V in NSP14; and T112I in NSP15 had 99-100% prevalence. However, we suggest G662S in RdRp and S403L in the NSP3 as the characteristic mutations of BA.2.75 Omicron SARS-CoV-2. It is also evident from the fact that their incidence in the global samples of BA.2.75 is 99% and secondly these mutations had not been observed in any other variant of SARS-CoV-2. Moreover, P314L of RdRp had 100% incidence in BA.2.75 and is located very close to the hydrophobic cleft of RdRp which is the target of some antiviral drugs like remdesivir and favipiravir [24,32]. Hence, the occurrence of highly prevalent mutations in this region of RdRp suggest that this variant is likely to have resistance to various antiviral therapeutic agents. In addition to changing its sensitivity to antiviral drugs, this mutation might be involved in affecting the replication speed of the virus which is the basic function of the RdRp protein [33]. Further to the antiviral treatments, it has been reported that the vaccinated individuals who had suffered with some breakthrough infection from B.1, or B.5 had got  T  19  I  --I  --I  I   L  24  del  -------A  27  S  ---- protective level of (hybrid) immunity against BA.2.75 compared with those who were only vaccinated (3-doses), which shows high level of cross-immunity between B.1, B.5, and B.2.75 variants [34].

PLOS ONE
severity of cases accredited to BA.2.75. This may lead to enhance our knowledge and understanding of the fact that these genetic similarities can prime the immune system to successfully fight-off with one of these subvariant, after defeating an infection from the other.

Conclusion
We discovered a total of 11 characteristics mutations for BA.2.75, including 9 mutations in the Spike protein, 2 in non-structural proteins, and 1 in the Envelop protein of BA.2.75. Out of these characteristic mutations K147E, W152R, F157L, I210V, V213G, G339H are present in the NTD; G446S and N460K in RBD; S403L in NSP3; G662S in RdRp and T11A in E protein. However, the phylogenetic analysis revealed that BA.2.75 is descended from BA.5, an Omicron sub-variant. This evolutionary link between BA.2.75 and BA.5 is critical in determining whether the surge of BA.5 infection will decrease the severity of cases caused by BA.2.75, suggesting that that the development of immunity and cross-protection between these two variants is also possible. Hence, it is also very likely that a single vaccine may be used to develop the protective level of immunity against both, BA.5 and BA.2.75.