Molecular conservation and differential mutation on ORF3a gene in Indian SARS-CoV2 genomes.

A global emergency due to the COVID-19 pandemic demands various studies related to genes and genomes of the SARS-CoV2. Among other important proteins, the role of accessory proteins are of immense importance in replication, regulation of infections of the coronavirus in the hosts. The largest accessory protein in the SARS-CoV2 genome is ORF3a which modulates the host response to the virus infection and consequently it plays an important role in pathogenesis. In this study, an attempt is made to decipher the conservation of nucleotides, dimers, codons and amino acids in the ORF3a genes across thirty-two genomes of Indian patients. ORF3a gene possesses single and double point mutations in Indian SARS-CoV2 genomes suggesting the change of SARS-CoV2's virulence property in Indian patients. We find that the parental origin of the ORF3a gene over the genomes of SARS-CoV2 and Pangolin-CoV is same from the phylogenetic analysis based on conservation of nucleotides and so on. This study highlights the accumulation of mutation on ORF3a in Indian SARS-CoV2 genomes which may provide the designing therapeutic approach against SARS-CoV2.

Introduction

Since December 2019, the coronavirus disease (COVID-19) due to the severe acute respiratory syndrome (SARS) originating from Wuhan, China, has been causing a pandemic across the world [1,2]. The causative virus, SARS-CoV2 is a positive-stranded RNA virus with genome size approximately of 30,000 bases. The genome of SARS-CoV2 contains twenty-nine open reading frames (ORFs) [[1], [2], [3], [4], [5], [6], [7], [8], [10], [11],4]. Among the twenty-nine ORFs, there are sixteen nonstructural proteins (nsps), four structural proteins (E, M, N, S), and six or seven accessory proteins such as ORF3a, ORF6, ORF7a, ORF7b, ORF8 and ORF10 [[5], [6], [7]]. SARS-CoV2 has been thought to be evolved due to rapid mutation, and recombination with existing other coronavirus in the body. They can alter tissue tropism, cross the species barrier and adapt to different epidemiological situations [8]. Sequence similarity based phylogeny infers that the SARS-CoV2 forms a distinct lineage with Bat-SARS-like coronaviruses that belong to the genus Beta-coronavirus (β-CoVs) [9]. The SARS-CoV2 genomes have a significant sequential similarity with percentages 96.3%, 89%, and 82% with bat CoV, SARS-like CoV, and SARS-CoV, respectively, which confirms zoonotic origin of the SARS-CoV2 [10]. There are about 380 amino acid changes from the different proteins of SARS-CoV genomes to the proteins of present SARS-CoV2 genomes as reported so far [11]. The 348, 27 and 5 changes of amino acids occurred in different accessory proteins, S protein and N protein respectively [11]. The accessory proteins have a significant role in virus pathogenesis and these proteins regulate the interferon signalling pathways and the production of pro-inflammatory cytokines [12]. The ORF3a gene which encodes a protein of 274 amino acids, is the second largest sub-genomic RNA in the genome of SARS-CoV [13]. The ORF3a gene encodes a protein with TRAF, ion channel and caveolin binding domain [14]. Mutation in these region alters the NF-kB activation and NLRP3 inflammosome [13]. One of the important features of the ORF3a protein is the presence of a cysteine-rich domain as observed in the SARS-CoV genomes [15]. The ORF3a protein is expressed abundantly in infected and transfected cells, which localizes to intracellular and plasma membranes [16,17]. It induces apoptosis in transfected and infected cells [18]. In the SARS-CoV genomes, co-mutation between the ORF3a gene and the spike gene exists which suggests that the function of the ORF3a protein correlates with the spike protein [[19], [20], [21]]. Therefore, locating the mutation in ORF3a proteins might lead to understanding the functionality changes in the protein during viral spreading. On missense mutations of various proteins of SARS-CoV2 and related finding are presented in the articles [[22], [23], [24]]. Till today, no such study has been carried out to look for the existence of ORF3a variation in the Indian patients.

In this present study, we intend to transact the molecular arrangements of nucleotides, dimers, codons and amino acids of the ORF3a gene/protein sequences of SARS-CoV2 of the Indian patients and of CoVs of Bat and Pangolin in order to fetch the evolution connections (if there is any) and similarities and dissimilarities. This study would help to comprehend the effect of non-synonymous mutations in the accessory proteins of the SARS-CoV genomes collected from various geo-locations across the world. In addition, beyond sequence similarity based bioinformatics, this study opens us the hidden conservation of nucleotides, dimers, codons and amino acids over the accessory protein OR3a of three different hosts such as Bat, Pangolin and Human.

Findings on the dataset

Globally, as on May 14, 2020, among 2385 genomes, we see 118 different mutations in the ORF3a gene. Among these mutations, three changes the size of the gene ORF3a. Out of three changes, one is with deletion of one codon (MT358717-USA: WA), second with deletion of two codons (MT293186-USA: WA) and third with insertion of one codon (MT449656-USA: CA). The rest (115 in total), including accessions from India, contain ORF3a genes of SARS-CoV2 genomes with only point mutations. There are five major genomic groups with sizes (1068, 967, 100, 31, 30), the rest of the groups have sizes in one digit. We name the two largest groups as ORF3a-Type-1 and ORF3a-Type-2. Among them, there is just a difference of one point mutation (G to T) at the 117 position of the ORF3a gene across all the 967 SARS-CoV2 genomes. In all the groups, the number of point mutations is found to be at most 4, across the available genome data. The most divergent mutations are often found in the USA. Though 102 different position of ORF3a are globally found, but mutation in three positions which are exclusively in Indian SARS-CoV2 are considered for our study.

As on May 14, 2020, there are thirty-two complete genomes viz. MT451874, MT451876, MT451877, MT451878, MT451880, MT451881, MT451882, MT451883, MT451884, MT451885, MT451886, MT451887, MT451888, MT451889, MT451890, MT435079, MT435080, MT435081, MT435082, MT435083, MT435084, MT435085, MT435086, MT415320, MT415321, MT415322, MT415323, MT358637, MT012098 and MT050493 of SARS-CoV2 from Indian patients are available in the NCBI database and that are considered for this present study [25]. Note that, except the genomes MT012098, MT050493 all the other thirty genomes belong to the L-type as per classification made in the article [26]. A set of brief remarks on the accessory protein coding genes across the thirty-two genomes from the Indian patients is given in Table 1 . The proteins ORF7a, ORF6 and ORF10 are 100% conserved in the thirty-two SARS-CoV2 genomes of Indian origin. It is noteworthy that after some days (while the manuscript was under-review) on May 24, 2020, in the Indian genomes, some missense mutations over the protein ORF7a, ORF7b and ORF8 are found as reported in the article [27]. However, there are four different types of ORF3a genes that are found based on single-point mutations.

Table 1

Accessory proteins coding genes with associated remarks based on the thirty two genomes from India.

Accessory Gene	Remarks based on the thirty two Indian genomes
ORF3a	Three single-point mutations (viz. G to T and C to T) are found in ORF3a gene across the thirty genomes.
ORF6	100% identical across all the thirty two genomes.
ORF7b	100% identical across all the thirty two genomes.
ORF7a	100% identical except in the genome MT435082.
	From 318th onwards 20 ambiguous base N are placed.
ORF10	100% identical across all the thirty genomes.
ORF8	100% identical except in the genomes MT435081 and MT435082.
	Note that MT435081 and MT35082 contain the truncated ORF8 gene.
	In the truncated genes there is a point mutation from C to T.
	Note that ORF8 and ORF7a are exactly of same length
	but it does not have any significant similarity.

In Indian patients, we found twenty-two ORF3a-Type-1 and seven ORF3a-Type-2 genomes among the thirty-two genomes of the Indian patients. The rest of the two types of mutations (we have seen 2 + 1 = 3 genomes) are Indian patients specific and have only one base difference with ORF3a-Type-2 and two bases differences from the 50 ORF3a-Type-1. We named these two Indian groups as ORF3a-Type-3 and ORF3a-Type-4 (refer to Table 2 ). The nucleotide frequencies, length and some associated remarks of the four types of ORF3a genes of SARS-CoV2 genomes of the Indian patients including the ORF3a genes of the pangolin and Bat CoV are presented in Table 2.

Table 2

ORF3a genes across different SARS-CoV2 and CoVs genomes of Pangolin and Bat.

ORF3a/Genome ID	Host	# of A	# of C	# of G	# of T	Length	Remarks
ORF3a-Type-1	Human	225	174	153	276	828	At 171th position, the base is G
ORF3a-Type-2	Human	225	174	152	277	828	W.r.t. ORF3a-Type-1 gene, at 171th position one mutation G to T occurred.
ORF3a-Type-3	Human	225	174	151	278	828	W.r.t. ORF3a-Type-2 gene, at 463rd position one mutation G to T occurred.
ORF3a-Type-4	Human	225	173	152	278	828	W.r.t. ORF3a-Type-2 gene, at 512th position one mutation C to T occurred.
MT040333	Pangolin	223	175	151	279	828	The query gene ORF3a
MT040334	Pangolin	224	173	152	279	828	826/828(99%)
MT040335	Pangolin	225	172	152	279	828	825/828(99%)
MT040336	Pangolin	224	173	152	279	828	826/828(99%)
KY417143	Bat	223	178	161	263	825	The query gene ORF3a
KY417144	Bat	234	179	152	260	825	749/827(91%)
KY417146	Bat	232	176	156	261	825	751/829(91%)
KY417147	Bat	227	179	158	261	825	807/825(98%)
KY417148	Bat	222	179	162	262	825	821/825(99%)
KY417149	Bat	225	181	158	261	825	795/825(96%)
KY417150	Bat	233	179	153	260	825	748/827(90%)
KY417151	Bat	236	180	151	258	825	745/827(90%)
KY417152	Bat	235	177	150	262	824	748/827(90%)

In the Table 2, it is found that the length of ORF3a gene of SARS-CoV2 genomes is 828 bases whereas the length of ORF3a gene of SARS-CoV was 825 bases. That is ORF3a gene in SARS-CoV and SARS-CoV2 encode amino acid sequence of length 274 and 275 respectively. Clearly, in the present SARS-CoV2 genomes, the one amino acid E, Glutamic acid is inserted after 240th aa of the ORF3 protein sequence into the ORF3a protein sequence which is shown in the Fig.1 .

Fig. 1

Amino acid Glutamic acid (E) insertion in ORF3a gene of SARS-CoV. Credit: NCBI.

The ORF3a protein of the SARS-CoV2 is also blasted (using NCBI-blastp suite) with other ORF3a proteins of Bat and Pangolin CoV. It resulted that the Glutamic acid at the 241 position matches with that of Pangolin-CoV which is shown in Fig.2 .

Fig. 2

Amino acid sequence alignment of ORF3a across Bat and Pangolin CoV with that of SARS-CoV2. Credit: NCBI.

So considering the mutations in ORF3a gene of the SARS-CoV2 genomes of Indian patients, there are four different ORF3a gene sequences of SARS-CoV2 are found, and they are referred as ORF3a-Type-1, 2, 3 and 4. These mutations over the gene ORF3a alter the amino acids viz. Q to H, D to Y and S to L), which is schematically presented in the Fig.3 .

Fig. 3

Mutations and associated alternation of amino acids in the four types of ORF3a genes.

The Fig.3 follows that the ORF3a-Type-3 is obtained by two single point mutation (G to T) from the ORF3a-Type-1. Likewise, the ORF3a-Type-4 is achieved by two single point mutations (G to T and C to T) from the ORF3a-Type-1. The genomes which contain the four different types of ORF3a genes of thirty-two SARS-CoV2 genomes of the Indian patients are mentioned in Table 3 . These data suggest that profiling of mutation on ORF3a genes in Indian patients is different from that of rest of world.

Table 3

SARS-CoV2 genomes of 32 Indian patients and their respective type based on the mutation in ORF3a genes.

Accession	Geo_location	Collection_date	ORF3a type	Accession	Geo_location	Collection_date	ORF3a type
MT457403	Hyderabad	2020-03-25	Type-1	MT415321	India	2020-03-11	Type-1
MT451874	Surat	2020-04-24	Type-1	MT415322	India	2020-03-16	Type-1
MT451877	Surat	2020-04-26	Type-1	MT415323	India	2020-03-20	Type-1
MT451878	Surat	2020-04-27	Type-1	MT358637	Rajkot	2020-04-05	Type-1
MT451880	Surat	2020-04-26	Type-1	MT012098	Kerala State	2020-01-27	Type-1
MT451883	Ahmedabad	2020-04-26	Type-1	MT050493	Kerala State	2020-01-31	Type-1
MT451884	Ahmedabad	2020-04-26	Type-1	MT457402	Hyderabad	2020-03-24	Type-2
MT451886	Ahmedabad	2020-04-26	Type-1	MT451876	India: Surat	2020-04-26	Type-2
MT451887	Ahmedabad	2020-04-26	Type-1	MT451885	Ahmedabad	2020-04-26	Type-2
MT451889	Ahmedabad	2020-04-26	Type-1	MT451888	Ahmedabad	2020-04-26	Type-2
MT435079	Ahmedabad	2020-04-13	Type-1	MT435081	Ahmedabad	2020-04-13	Type-2
MT435080	Ahmedabad	2020-04-13	Type-1	MT435082	Ahmedabad	2020-04-13	Type-2
MT435083	Ahmedabad	2020-04-07	Type-1	MT435085	Gandhinagar	2020-04-22	Type-2
MT435084	Ahmedabad	2020-04-14	Type-1	MT451881	Ahmedabad	2020-04-26	Type-3
MT435086	Mansa	2020-04-21	Type-1	MT451882	Ahmedabad	2020-04-26	Type-3
MT415320	India	2020-03-01	Type-1	MT451890	Ahmedabad	2020-04-26	Type-4

In addition, as the references for establishing any evolutionary connections from the ORF3a gene perspective, ORF3a genes from the four CoV genomes of Pangolin viz. MT040333, MT040334, MT040335 and MT040336 and nine Bat CoV genomes viz. KY417143, KY417144, KY417146, KY417147, KY417148, KY417149, KY417150, KY417151 and KY417152 are considered for the present study. The corresponding phylogeny of the genomes based on sequential similarity of the ORF3a gene is given in the Fig.4 .

Fig. 4

Phylogeny (distance tree) of the thirty genomes based on sequential similarities of the ORF3a genes. Credit: NCBI.

The phylogeny shows that the ORF3a genes of CoVs across the three different hosts are mutually placed differently in the distance tree. The phylogeny reports that the ORF3a gene of four types of SARS-CoV2 genomes are sequentially very much closer to that of Pangolin-CoV, than Bat-CoV. The ORF3a-Type-3 and ORF3a-Type-4 genes are evolved from the ORF3a-Type-2 gene by single point mutations as reported in the phylogeny.

Among 1068 and 967 genomes having mutations of ORF3a-Type-1 and ORF3a-Type-2 respectively, one hundred each such examples of genomes with their respective geo-locations are given in the Table 4, Table 5 .

Table 4

List of accessions and respective geo-locations based on the NCBI blast of the query sequence ORF3a-Type-1 gene.

Accession	Geo_location	Accession	Geo_location	Accession	Geo_location	Accession	Geo_location
MT434758	India	MT418880	USA: VA	MT419855	USA: CA	MT412201	USA: Michigan
MT434759	India	MT418881	USA: VA	MT419856	USA: CA	MT412214	USA: Michigan
MT434760	India	MT418883	USA: VA	MT419857	USA: CA	MT412244	USA: WA
MT434786	USA: NY	MT418884	USA: VA	MT419858	USA: CA	MT412246	USA: WA
MT434796	USA: NY	MT418893	USA: VA	MT419859	USA: CA	MT412248	USA: WA
MT434800	USA: NY	MT418894	USA: VA	MT419860	USA: CA	MT412250	USA: WA
MT434813	USA: NY	MT419810	Puerto Rico	MT412134	China	MT412252	USA: WA
MT435079	India: Ahmedabad	MT419812	Puerto Rico	MT412136	USA: Michigan	MT412253	USA: WA
MT435080	India: Ahmedabad	MT419815	Puerto Rico	MT412137	USA: Michigan	MT412257	USA: WA
MT435083	India: Ahmedabad	MT419828	USA: CA	MT412138	USA: Michigan	MT412261	USA: WA
MT435084	India: Ahmedabad	MT419829	USA: CA	MT412139	USA: Michigan	MT412275	USA: WA
MT435086	India: Mansa	MT419830	USA: CA	MT412144	USA: Michigan	MT412281	USA
MT365028	Hong Kong	MT419831	USA: CA	MT412147	USA: Michigan	MT412290	USA: WA
MT365029	Hong Kong	MT419832	USA: CA	MT412157	USA: Michigan	MT412291	USA: WA
MT365030	Hong Kong	MT419833	USA: CA	MT412158	USA: Michigan	MT412295	USA: WA
MT365031	Hong Kong	MT419834	USA: CA	MT412159	USA: Michigan	MT412302	USA: CT
MT365032	Hong Kong	MT419835	USA: CA	MT412167	USA: Michigan	MT412303	USA: CT
MT428551	Kazakhstan	MT419837	USA: CA	MT412172	USA: Michigan	MT412312	USA: WA
MT428552	Kazakhstan	MT419839	USA: CA	MT412173	USA: Michigan	MT412316	USA: WA
MT428553	Kazakhstan	MT419841	USA: CA	MT412174	USA: Michigan	MT415320	India
MT429187	USA: Wisconsin	MT419842	USA: CA	MT412175	USA: Michigan	MT415321	India
MT429188	USA: Wisconsin	MT419845	USA: CA	MT412177	USA: Michigan	MT415322	India
MT318827		MT419846	USA: CA	MT412183	USA: Michigan	MT415323	India
MT270814	Hong Kong	MT419853	USA: CA	MT412193	USA: Michigan	MT415895	USA: VA
MT270815	Hong Kong	MT419854	USA: CA	MT412197	USA: Michigan	MT415896	USA: VA

Table 5

List of accessions and respective geo-locations based on the NCBI blast of the query sequence ORF3a-Type-2 gene.

Accession	Geo_location	Accession	Geo_location	Accession	Geo_location	Accession	Geo_location
MT434782	USA: NY	MT434817	USA: NY	MT419822	Puerto Rico	MT412216	USA: Michigan
MT434788	USA: NY	MT435081	India: Ahmedabad	MT419851	USA: CA	MT412217	USA: Michigan
MT434789	USA: NY	MT435082	India: Ahmedabad	MT412187	USA: Michigan	MT412218	USA: Michigan
MT434790	USA: NY	MT435085	India: Gandhinagar	MT412188	USA: Michigan	MT412219	USA: Michigan
MT434791	USA: NY	MT429183	USA: Wisconsin	MT412189	USA: Michigan	MT412220	USA: Michigan
MT434792	USA: NY	MT429184	USA: Wisconsin	MT412190	USA: Michigan	MT412221	USA: Michigan
MT434793	USA: NY	MT429185	USA: Wisconsin	MT412191	USA: Michigan	MT412222	USA: Michigan
MT434794	USA: NY	MT429186	USA: Wisconsin	MT412192	USA: Michigan	MT412223	USA: Michigan
MT434795	USA: NY	MT429189	USA: Wisconsin	MT412194	USA: Michigan	MT412224	USA: Michigan
MT434797	USA: NY	MT429190	USA: Wisconsin	MT412195	USA: Michigan	MT415894	USA: VA
MT434798	USA: NY	MT429191	USA: Wisconsin	MT412196	USA: Michigan	MT415897	USA: VA
MT434799	USA: NY	MT432195	USA: Louisiana	MT412198	USA: Michigan	MT415898	USA: VA
MT434801	USA: NY	MT422806	USA: FL	MT412199	USA: Michigan	MT415899	USA: VA
MT434802	USA: NY	MT422807	USA: FL	MT412200	USA: Michigan	MT415900	USA: VA
MT434803	USA: NY	MT418889	USA: VA	MT412202	USA: Michigan	MT415901	USA: VA
MT434804	USA: NY	MT418890	USA: VA	MT412203	USA: Michigan	MT415902	USA: VA
MT434805	USA: NY	MT418891	USA: VA	MT412204	USA: Michigan	MT415903	USA: VA
MT434806	USA: NY	MT418892	USA: VA	MT412205	USA: Michigan	MT415904	USA: VA
MT434808	USA: NY	MT419811	Puerto Rico	MT412206	USA: Michigan	MT415905	USA: VA
MT434809	USA: NY	MT419814	Puerto Rico	MT412207	USA: Michigan	MT415906	USA: VA
MT434810	USA: NY	MT419817	Puerto Rico	MT412209	USA: Michigan	MT415907	USA: VA
MT434811	USA: NY	MT419818	Puerto Rico	MT412211	USA: Michigan	MT415908	USA: VA
MT434812	USA: NY	MT419819	Puerto Rico	MT412212	USA: Michigan	MT415909	USA: VA
MT434815	USA: NY	MT419820	Puerto Rico	MT412213	USA: Michigan	MT415910	USA: VA
MT434816	USA: NY	MT419821	Puerto Rico	MT412215	USA: Illinois	MT415912	USA: VA

So these two types of ORF3a gene having one base difference belong to a large class of SARS-CoV2 genomes across different geo-locations as shown in Table 4, Table 5. It is noted that the NCBI blast results no genome from China having 100% similarity with the ORF3a-Type-2 gene. That is the one point mutation (G to T) in the ORF3a-Type-2 gene that has happened outside the patients of China. It is worth mentioning that the OF3a-Type-3 and ORF3a-Type-4 genes were blasted in the NCBI database and do not find any 100% similar sequence with 100% query coverage. Hence, these two type of mutations in the gene ORF3a are unique in Indian patients.

Methods

In order to determine the molecular level conservation and descriptions of the ORF3a genes across different hosts as mentioned, some methods are discussed [[28], [29], [30], [31], [32], [33]], which would be subsequently used.

Nucleotide conservation Shannon entropy

Shannon entropy is a measure of the amount of information (measure of uncertainty). Conservation of each of the four nucleotides has been determined using Shannon entropy [34,35]. Note that it is assumed log (0) = 0 for smooth calculation of the SE. For a given sequence of length l, the conservation SE (Conv_SE) is calculated as follows:where ; f represents the occurrence frequency of a nucleotide N in the given sequence.

Dimer conservation Shannon entropy

The conservation of usages of all possible sixteen dimers (words of length two consisting letters from the set {A,  T,  C,  G}) has been determined using Shannon entropy as follows. For a given sequence of length l, the conservation of dimers (Dim_SE) is calculated as follows:where ; g represents the number of occurrences of a dimer m in the given sequence.

Codon conservation Shannon entropy

The conservation of usages of all possible sixty four codons has been determined using Shannon entropy as follows [36]. For a given sequence of length l, the conservation of codons (Codon_SE) is calculated as follows:where ; h represents the number of occurrences of a codon r in the given sequence.

Amino acid conservation Shannon entropy

The conservation of twenty amino acids usages across the primary protein sequence encoded by the gene ORF3a has been determined using Shannon entropy as follows. For a given amino acid sequence corresponding to a RNA sequence (ORF3a gene) of length l, the conservation of codons (AA_SE) is calculated as follows:where ; k represents the number of occurrences of an amino acid s in the given sequence.

In addition to the different conservation SEs, some basic derivative features such as nucleotide frequency and density, frequency of all possible sixteen dimers, frequency of codon usages, frequency of amino acids in the protein sequence encoded by the ORF3a gene, GC content, pyrimidine density are obtained for a given ORF3a gene sequence [28,30]. It is worth noting that the first positive frame has been considered to determine codons and double nucleotides over a given gene.

Results

For each of the seventeen different ORF3a genes (including the genomes of SARS-CoV2, Pangolin and Bat CoV) a feature vector is defined which comprises the nucleotides, dimers, codons and amino acids frequencies and associated conservation in the ORF3a genes. Based on these feature vectors corresponding to each of the seventeen sequences, the nearest neighbourhood joining phylogeny is built up for each of the molecular conservation of nucleotides, dimers, codon and amino acids.

Frequency and conservation of nucleotides over ORF3a gene

The counts of the nucleotide bases, length, GC content and pyrimidine density (py density) and the conservation Shannon entropy (ConV_SE) of the seventeen ORF3a genes across three different hosts are tabulated in Table 6 .

Table 6

Molecular descriptions of the gene ORF3a across different hosts.

ORF3a/genome ID	Den A	Den C	Den G	Den T	GC content	Py density	Conv_SE
ORF3a-Type-1	0.2717	0.2101	0.1848	0.3333	39.4928	54.3478	0.9811
ORF3a-Type-2	0.2717	0.2101	0.1836	0.3345	39.3720	54.4686	0.9806
ORF3a-Type-3	0.2717	0.2101	0.1824	0.3357	39.2512	54.5894	0.9801
ORF3a-Type-4	0.2717	0.2089	0.1836	0.3357	39.2512	54.4686	0.9802
MT040333	0.2693	0.2114	0.1824	0.3370	39.3720	54.8309	0.9801
MT040334	0.2705	0.2089	0.1836	0.3370	39.2512	54.5894	0.9800
MT040335	0.2717	0.2077	0.1836	0.3370	39.1304	54.4686	0.9798
MT040336	0.2705	0.2089	0.1836	0.3370	39.2512	54.5894	0.9800
KY417143	0.2703	0.2158	0.1952	0.3188	41.0909	53.4545	0.9867
KY417144	0.2836	0.2170	0.1842	0.3152	40.1212	53.2121	0.9843
KY417146	0.2812	0.2133	0.1891	0.3164	40.2424	52.9697	0.9849
KY417147	0.2752	0.2170	0.1915	0.3164	40.8485	53.3333	0.9862
KY417148	0.2691	0.2170	0.1964	0.3176	41.3333	53.4545	0.9873
KY417149	0.2727	0.2194	0.1915	0.3164	41.0909	53.5758	0.9866
KY417150	0.2824	0.2170	0.1855	0.3152	40.2424	53.2121	0.9846
KY417151	0.2861	0.2182	0.1830	0.3127	40.1212	53.0909	0.9843
KY417152	0.2852	0.2148	0.1820	0.3180	39.6845	53.2767	0.9829

The density of each nucleotide bases across the seventeen ORF3a genes are plotted in the Fig.5 .

Fig. 5

Nucleotide density of four bases across the seventeen ORF3a genes. The numbers 1, 2, 3, denote the ORF3a gene/Genome ID from the top to bottom of the first column of Table 6, respectively.

In each ORF3a gene the density of T is maximum and G is minimum. Also, it is noted the density of C dominates that of G over all the ORF3a genes of three different hosts. The ORF3a genes are pyrimidine-rich with percentage approximately 53% across different genomes as mentioned in the Table 6. Also, the ORF3a possesses the highest GC content across the Bat CoV genomes and which is ranging from 39.68% to 41.34%. After a single mutation, the GC content of ORF3a-Type-2 is slightly reduced to 39.5% from that of the ORF3a-Type-1 gene. The GC content of Pangolin CoVs is turned out to be minimum and that is 39.13%. The GC content of ORF3a-Type-2 gene and ORF3a of MT040333 is identical though the density of G and C are slightly different in the respective sequences. The ORF3a genes across fifteen different genomes of CoV of the three hosts are highly conserved with equally likely presence of the four nucleotide bases as the Conv_SE for all the genes are turned out to be approximately 1.

Based on features of the ORF3a gene across the seventeen CoV genomes, as mentioned in the Table 6, a phylogeny has been developed as shown in Fig.6 .

Fig. 6

Phylogenetic relationships among the seventeen CoV genomes based on the densities of nucleotides of ORF3a gene.

The phylogeny depicts that the ORF3a-Type-1 and ORF3a-Type-2 gene of the SARS-CoV2 genomes of the Indian patients are very close to each other (belong to the 4th level of the tree). At the 6th level of the phylogenetic tree, the ORF3a-Type-3 and that of the genomes MT040334 and MT040336 of CoV of Pangolin belong and naturally they are co-evolved from the previous evolutionary levels of the tree. The ORF3a-Type-4 gene and ORF3a of the Bat CoV genome MT040335 belong to the binary branch of 4th level of the phylogenetic tree. It is also inferred from the Fig.3 that the ORF3a genes of four types of SARS-CoV2 and CoV-Pangolin are evolved from the ORF3a gene of the Pangolin CoV genome MT040333. On the other side, ORF3a gene of Bat CoV are distantly placed in the tree. Among the nine genomes of Bat CoV, the pair of genomes {KY417143,  KY417149} and {KY417144,  KY417150} are the nearest as they belong to the sixth level of the tree.

Frequency and conservation of dimers over ORF3a gene

All possible words consisting letters from the set {A,  T,  C,  G} of length two are commonly known as dimers. The frequency of dimers and conservation Shannon entropy of dimers (Dim_SE) over the seventeen ORF3a genes across various genomes of CoV are presented in the Table 7 . Also, a graphical representation of the frequency of the dimers of four types (dimers start with A, T, C and G) are given in Fig.7 .

Table 7

Frequency of dimers of the gene ORF3a and associated dimer conservation Shannon entropy.

ORF3a/genome ID	AA	AC	AG	AT	CA	CC	CG	CT	GA	GC	GG	GT	TA	TC	TG	TT	Dim_SE
ORF3a-Type-1	70	62	35	57	55	31	17	71	45	35	26	47	54	46	75	101	0.9705
ORF3a-Type-2	70	62	34	58	55	31	17	71	44	35	26	47	55	46	75	101	0.9702
ORF3a-Type-3	70	62	34	58	55	31	16	72	43	35	26	47	56	46	75	101	0.9694
ORF3a-Type-4	70	62	34	58	54	31	17	71	44	35	26	47	56	45	75	102	0.9698
MT040333	60	57	36	69	59	28	15	73	41	42	29	39	62	48	71	98	0.9706
MT040334	60	57	37	69	58	27	15	73	42	42	29	39	63	47	71	98	0.981
MT040335	61	56	37	70	58	27	15	72	42	42	29	39	63	47	71	98	0.9705
MT040336	61	55	37	70	58	28	15	72	42	42	29	39	62	48	71	98	0.9709
KY417143	59	62	38	63	68	22	20	68	42	51	25	43	53	43	78	89	0.9726
KY417144	64	62	38	69	70	29	20	60	40	48	26	38	59	40	68	93	0.9738
KY417146	62	60	38	71	69	26	21	60	40	47	29	40	60	43	68	90	0.9756
KY417147	63	64	38	61	71	23	20	65	39	49	26	44	53	43	74	91	0.9729
KY417148	59	62	38	62	68	23	21	67	42	52	25	43	52	42	78	90	0.9733
KY417149	61	64	40	59	69	23	20	69	39	52	24	43	55	42	74	90	0.9727
KY417150	64	62	38	68	69	29	21	60	40	48	26	39	59	40	68	93	0.9746
KY417151	65	62	38	70	72	29	20	59	39	49	26	37	59	40	67	92	0.9735
KY417152	64	61	37	72	70	29	19	59	39	47	26	38	61	40	68	93	0.9727

Fig. 7

Bar-plot of the frequencies of dimers of ORF3a genes.

From the Fig.7, it is noticed that frequency of the dimers starting with the letter T is the highest over the gene ORF3a across the seventeen distinct genomes. The frequency decreases over the dimers with the first letter A, C and G respectively. The dimers TT and CG attain maximum and minimum frequency over the ORF3a gene across the fifteen genomes. In all the four types of ORF3a genes the frequencies of the dimers AG, AT, CA, CG, CT, GA, TA, TC and TT are varying as observed in the Table 7. The frequency of usages of most of the dimers in the ORF3a genes of four types dominates that of the Pangolin and Bat CoVs. The Dim_SE follows that the ORF3a genes across all the genomes are conserved with all sixteen dimers with nearly equal probability of occurrences. The Dim_SE of the ORF3a-Type-1 and ORF3a of the genome MT040335 of Pangolin-CoV are identical although the frequency of respective dimers are different. It is noted that all the dimers are equally likely to appear and conserved in the ORF3a-Type-3 and ORF3a-Type-4 genes.

Based on the frequency of dimers across the ORF3a genes over the genomes the following phylogeny is made in Fig.8 .

Fig. 8

Phylogenetic relationships among the seventeen CoV genomes based on the frequency of dimers of ORF3a genes.

The phylogeny based on the frequency distribution of the dimers over the ORF3a genes across various genomes of different hosts follows that ORF3a genes of SARS-CoV2 genomes of Indian patients and genomes of Pangolin-COV are co-evolved by belonging into the same level of the tree. In the other branch of the phylogenetic tree ORF3a genes of the Bat CoV are placed and among them the genomes KY417144 and KY417150 are the nearest based on the dimers usages over the gene ORF3a as found in the Fig.8.

Codon conservations and associated descriptions of ORF3a gene

The frequency usages of all the codons over the ORF3a genes across the SARS-CoV2 genomes of Indian patients including genomes of Pangolin and Bat CoVs are given in Table 8 . All the twenty amino acids are present in the protein sequence of ORF3a although the codons CCC, CGA, GGG, TAG and TGA are thoroughly absent from the ORF3a genes across all the genomes. The ORF3a genes of SARS-CoV2 genomes of the Indian patients as well as of Pangolin CoV contain one CGC while that of the Bat CoV do not contain the codon CGC. This presence of the codon CGC (codes Arginine) deviates the ORF3a gene of SARS-CoV2 and Pangolin CoV from that of the Bat-CoV. In contrast, ORF3a genes of the genomes of Bat-CoV contain the codon GCG (encode Alanine) while the ORF3a genes of four types of SARS-CoV2 genomes do not contain it. It is found that the frequency of GAG, GTG in ORF3a genes of Bat-CoV dominates that of the other two genomes. The most preferred stop codon across all the ORF3a genes of various CoV genomes is TAA. The most frequently used codon ATT and GTT which encode Isoleucine and Valine respectively in ORF3a across all the observed genomes. The ORF3a genes possess clearly codon biases in encoding the various amino acids as evident from the codon frequency usages.

Table 8

Frequency of codon usages over the gene ORF3a across the seventeen CoV genomes.

ORF3a/Genome ID	AAA	AAC	AAG	AAT	ACA	ACC	ACG	ACT	AGA	AGC	AGG	AGT	ATA	ATC	ATG	ATT	Codon_SE
ORF3a-Type-1	7	4	4	4	6	2	3	13	3	2	1	5	7	5	4	9	0.9245
ORF3a-Type-2	7	4	4	4	6	2	3	13	3	2	1	5	7	5	4	9	0.9243
ORF3a-Type-3	7	4	4	4	6	2	3	13	3	2	1	5	7	5	4	9	0.9238
ORF3a-Type-4	7	4	4	4	6	2	3	13	3	2	1	5	7	5	4	9	0.9249
MT040333	6	1	4	6	10	1	4	10	4	3	2	4	5	2	4	14	0.9151
MT040334	6	1	4	6	11	1	4	10	4	3	2	4	5	2	4	14	0.915
MT040335	6	1	4	7	11	1	4	9	4	3	2	4	5	2	4	14	0.9153
MT040336	6	1	4	7	10	1	4	9	4	3	2	4	5	2	4	14	0.9159
KY417143	8	5	4	5	10	1	4	7	3	2	0	4	3	4	6	13	0.925
KY417144	8	6	4	7	9	3	3	8	4	2	0	2	5	5	5	13	0.9283
KY417146	8	6	4	6	9	3	3	8	4	2	0	2	5	6	5	12	0.9277
KY417147	8	5	4	5	10	1	4	7	2	3	1	4	3	5	6	13	0.9285
KY417148	8	5	4	5	10	1	4	7	3	2	0	4	3	4	6	13	0.9275
KY417149	7	5	5	6	11	1	3	7	3	2	1	5	3	4	5	14	0.9259
KY417150	8	6	4	7	9	3	3	8	4	2	0	2	5	5	5	13	0.9289
KY417151	8	6	4	7	9	3	3	8	4	2	0	2	5	5	5	13	0.925
KY417152	8	6	4	7	9	3	3	8	4	2	0	2	5	5	5	13	0.9111
ORF3a/Genome ID	CAA	CAC	CAG	CAT	CCA	CCC	CCG	CCT	CGA	CGC	CGG	CGT	CTA	CTC	CTG	CTT
ORF3a-Type-1	5	4	4	4	3	0	2	7	0	1	0	1	1	5	2	10
ORF3a-Type-2	5	4	3	5	3	0	2	7	0	1	0	1	1	5	2	10
ORF3a-Type-3	5	4	3	5	3	0	2	7	0	1	0	1	1	5	2	10
ORF3a-Type-4	5	4	3	5	3	0	2	7	0	1	0	1	1	5	2	10
MT040333	7	3	1	6	6	0	2	5	0	1	0	0	2	4	3	13
MT040334	7	2	2	6	5	0	2	5	0	1	0	0	2	4	3	13
MT040335	7	2	2	6	5	0	2	5	0	1	0	0	2	4	3	13
MT040336	7	2	2	6	6	0	2	5	0	1	0	0	2	4	3	13
KY417143	7	3	4	5	6	0	4	3	0	0	0	1	2	6	4	10
KY417144	6	3	5	4	7	0	3	3	0	0	0	0	4	4	2	10
KY417146	5	2	7	5	6	0	2	3	0	0	0	0	4	5	2	9
KY417147	8	3	3	5	7	0	3	3	0	0	0	1	1	5	3	11
KY417148	7	3	4	5	6	0	4	3	0	0	0	1	2	6	3	10
KY417149	8	3	3	5	6	0	4	3	0	0	0	1	3	4	4	12
KY417150	6	3	5	4	6	0	4	3	0	0	0	0	4	4	2	10
KY417151	6	3	5	4	8	0	2	3	0	0	0	0	4	4	1	10
KY417152	6	2	4	6	8	0	2	3	0	0	0	0	4	4	1	10
ORF3a/Genome ID	GAA	GAC	GAG	GAT	GCA	GCC	GCG	GCT	GGA	GGC	GGG	GGT	GTA	GTC	GTG	GTT
ORF3a-Type-1	10	6	1	7	3	3	0	7	4	3	0	7	7	3	1	14
ORF3a-Type-2	10	6	1	7	3	3	0	7	4	3	0	7	7	3	1	14
ORF3a-Type-3	10	5	1	7	3	3	0	7	4	3	0	7	7	3	1	14
ORF3a-Type-4	10	6	1	7	3	3	0	7	4	3	0	7	7	3	1	14
MT040333	8	6	1	9	3	3	0	10	4	4	0	7	2	2	2	13
MT040334	8	6	1	9	3	3	0	10	4	4	0	7	2	2	2	13
MT040335	8	6	1	9	3	3	0	10	4	4	0	7	2	2	2	13
MT040336	8	6	1	9	3	3	0	10	4	4	0	7	2	2	2	13
KY417143	4	6	4	7	6	1	1	9	3	6	0	5	2	4	5	14
KY417144	3	8	4	5	8	4	1	6	4	5	0	6	2	4	2	12
KY417146	3	7	4	7	7	4	3	5	4	5	0	6	2	4	2	13
KY417147	4	6	4	7	6	2	1	10	3	6	0	5	3	3	4	13
KY417148	4	6	4	7	6	2	1	9	3	6	0	5	2	3	5	14
KY417149	4	7	4	5	6	1	1	11	1	7	0	5	3	4	4	12
KY417150	3	8	4	5	8	4	1	6	4	5	0	6	2	4	2	12
KY417151	3	8	4	5	9	4	2	6	4	5	0	6	1	5	2	11
KY417152	3	8	4	5	8	4	2	6	4	5	0	6	2	4	2	12
ORF3a/Genome ID	TAA	TAC	TAG	TAT	TCA	TCC	TCG	TCT	TGA	TGC	TGG	TGT	TTA	TTC	TTG	TTT
ORF3a-Type-1	1	9	0	8	8	4	0	3	0	4	6	3	3	6	9	8
ORF3a-Type-2	1	9	0	8	8	4	0	3	0	4	6	3	3	6	9	8
ORF3a-Type-3	1	10	0	8	8	4	0	3	0	4	6	3	3	6	9	8
ORF3a-Type-4	1	9	0	8	7	4	0	3	0	4	6	3	4	6	9	8
MT040333	1	8	0	8	7	3	0	6	0	7	6	2	3	4	4	10
MT040334	1	8	0	8	7	3	0	6	0	7	6	2	3	4	4	10
MT040335	1	8	0	8	7	3	0	6	0	7	6	2	3	4	4	10
MT040336	1	8	0	8	7	3	0	6	0	7	6	2	3	4	4	10
KY417143	1	7	0	10	6	2	0	4	0	7	6	1	3	5	4	8
KY417144	1	6	0	11	6	1	1	4	0	4	6	4	2	6	5	9
KY417146	1	6	0	11	7	1	0	4	0	4	6	4	2	6	5	9
KY417147	1	9	0	8	6	2	0	4	0	4	6	3	4	6	4	7
KY417148	1	7	0	9	6	2	0	4	0	7	6	2	3	6	5	7
KY417149	1	8	0	9	5	2	1	4	0	6	6	1	2	5	4	8
KY417150	1	6	0	11	6	1	1	4	0	4	6	4	2	6	5	9
KY417151	1	6	0	11	6	1	0	4	0	4	6	4	3	6	5	9
KY417152	1	6	0	11	6	1	0	4	0	4	6	4	3	6	5	9

Over the seventeen different genomes of SARS-CoV2, Pangolin and Bat, the codons are not as conserved as the nucleotides and dimers were in the ORF3a gene due to the codon biases. The Codon_SE of ORF3a genes across the genomes are ranging from 0.9111 to 0.9289 and this emerges to a certain degree of uncertainty of codon conservation over the gene.

The following phylogeny of the seven genomes is made by using the frequency of codon usages over the gene ORF3a, as shown in Fig.9 .

Fig. 9

Phylogenetic relationships among the seventeen CoV genomes based on the frequency of codon usages in ORF3a gene across fifteen genomes.

Based on frequency of codon usages and conservation of codon in the ORF3a genes, the four types of SARS-CoV2 genomes of the Indian patients are distantly placed from the Pangolin and Bat CoVs as chalked out in the phylogenetic tree. The closest distribution of codons in the gene ORF3a over the pair of genomes KY417143 and KY417148 of Bat-CoV is noted. This phylogeny in the Fig. 9 depicts that the ORF3a gene of genomes of the Indian patients and that of Bat CoV are co-evolved from the same origin.

Amino acids conservations and associated descriptions of ORF3a gene

The frequency of amino acids over the gene ORF3a across the genome of Indian patients, Pangolin and Bat are presented in the Table 9 . All the twenty amino acids are present over the gene ORF3a across all the genomes and it is turned out that the ORF3a protein is Luicine-rich with percentage approximately 10%. It is worth mentioning that the ORF3a gene of SARs-CoV genomes was cystine rich. The frequency of the amino acids Methionine and Arginine are the lowest among all over the ORF3a genes across the genomes. In the ORF3a gene of Type-1 and Type-2 the frequency of Glutamine and Histidine are altered from 9 to 8 and 8 to 9 respectively. The frequencies of Aspertic acid (D), Leucine (L) are 12 and 30 respectively in the ORF3a-Type-3 gene while those of D and L are 13 and 31 in the ORF3a-Type-4 gene of SARS-CoV2 genomes of the Indian patients. The frequencies of Serine and Tyrosine are increased by 1 in ORF3a while it switches from the Type-3 to Type-4 of SARS-CoV2 genomes of Indian patients.

Table 9

Amino acids frequencies over the ORF3a protein sequence across the seventeen genomes.

ORF3a/genome ID	A	R	N	D	C	Q	E	G	H	I	L	K	M	F	P	S	T	W	Y	V	AA_SE
ORF3a-Type-1	13	6	8	13	7	9	11	14	8	21	30	11	4	14	12	22	24	6	17	25	0.9553
ORF3a-Type-2	13	6	8	13	7	8	11	14	9	21	30	11	4	14	12	22	24	6	17	25	0.9553
ORF3a-Type-3	13	6	8	12	7	8	11	14	9	21	30	11	4	14	12	22	24	6	18	25	0.9549
ORF3a-Type-4	13	6	8	13	7	8	11	14	9	21	31	11	4	14	12	21	24	6	17	25	0.9549
MT040333	16	7	7	15	9	8	9	15	9	21	29	10	4	14	13	23	25	6	16	19	0.9587
MT040334	16	7	7	15	9	9	9	15	8	21	29	10	4	14	12	23	26	6	16	19	0.9579
MT040335	16	7	8	15	9	9	9	15	8	21	29	10	4	14	12	23	25	6	16	19	0.9594
MT040336	16	7	8	15	9	9	9	15	8	21	29	10	4	14	13	23	24	6	16	19	0.9602
KY417143	17	4	10	13	8	11	8	14	8	20	29	12	6	13	13	18	22	6	17	25	0.9607
KY417144	19	4	13	13	8	11	7	15	7	23	27	12	5	15	13	16	23	6	17	20	0.961
KY417146	19	4	12	14	8	12	7	15	7	23	27	12	5	15	11	16	23	6	17	21	0.9603
KY417147	19	4	10	13	7	11	8	14	8	21	28	12	6	13	13	19	22	6	17	23	0.9607
KY417148	18	4	10	13	9	11	8	14	8	20	29	12	6	13	13	18	22	6	16	24	0.9619
KY417149	19	5	11	12	7	11	8	13	8	21	29	12	5	13	13	19	22	6	17	23	0.9602
KY417150	19	4	13	13	8	11	7	15	7	23	27	12	5	15	13	16	23	6	17	20	0.961
KY417151	21	4	13	13	8	11	7	15	7	23	27	12	5	15	13	15	23	6	17	19	0.9607
KY417152	20	4	13	13	8	10	7	15	8	23	27	12	5	15	13	15	23	6	17	20	0.9619

A typical frequency distribution of amino acids in ORF3a genes across the seventeen genomes are presented in Fig.7. The frequencies of amino acids Isoleucine, Methionine, Phenylalanine and Tryptophan are invariant in ORF3a gene across the SARS-CoV2 and Pangolin-CoV genomes among three hosts (Fig.10).

Fig. 10

Frequency distribution of amino acids over the ORF3a genes of SARS-CoV2 genomes of the Indian patients, Pangolin-CoV and Bat CoV from left to right.

The AA_SE follows that the conservation of amino acids of ORF3a over the genome of Indian patients is invariant under mutation. It is noted that the ORF3a genes over the CoV genomes of Pangolin and Bat possess higher conservation of amino acids than that of SARS-CoV2 genomes of the Indian patients. ORF3a gene over the genomes KY417148 and KY417152 attain the highest amount of amino acid conservation as found in the Table 9.

Based on the frequency distribution of amino acids the following phylogeny (Fig.11 ) of the seventeen genomes are established. At the fifth level of the phylogenetic tree the pairs of genomes {ORF3a − Type − 1,  ORF3a − Type − 2}, {MT040335,  MT040336} and {KY417144,  KY417150} belong as leaf nodes and this imply the co-evolution of the ORF3a gene from the same parental origin.

Fig. 11

Phylogenetic relationships among the seventeen CoV genomes based on the frequency of amino acids in ORF3a proteins.

Conclusions

Among all the accessory proteins of SARS-CoV2, ORF3a is found to be very much important in playing virus pathogenesis as it possesses various mutations which are linked with that of the spike proteins. As mentioned, there are different mutations happened at various locations of the ORF3a gene of the SARS-CoV2 genomes of Indian patients and those mutations lead to alternation of amino acids. Among the mutations, the ORF3a-Type-3 and ORF3a-Type-4 mutations are restricted to only the Indian patients based in Ahmedabad so far it is identified. These mutations (Q to H, D to Y, S to L) are located near TRAF, ion channel, and caveolin binding domains respectively, suggesting that Type-3 and Type-4 might have effect on NLRP3 inflammasome activation. This unique non-synonymous mutations might affect the virulence of the virus and this needs a special attention from pathogenesis perspective by the medical scientists. A set of ORF3a genes of the Pangolin and Bat-CoVs were taken into consideration to investigate the evolutionary relationship from the phylogenies based on the nucleotides, dimers, codons and amino acids over the gene ORF3a across various genomes of CoVs. Based on conservations of nucleotide bases over the ORF3a genes, it is turned out that the ORF3a genes of four types of SARS-CoV2 and CoV-Pangolin are evolved from the ORF3a gene of the Pangolin CoV genome MT040333. It is worth noting that the ORF3a genes of Pangolin and Bat-CoV genomes are much closer than that of SARS-CoV2, from the phylogenetic analysis of codon and amino acids conservations. From the molecular conservation analysis, it is emerged that the ORF3a genes across the seventeen genomes of SARS-CoV2 along with that of Pangolin and Bat-CoVs are co-evolved from the same origin.

Author contributions

SH conceived the problem. SH, PPC, PB and SSJ analysed the data and result. SH wrote the initial draft which was checked and edited by all other authors to generate the final version.

Declaration of Competing Interest

The authors do not have any conflicts of interest to declare.