https://orcid.org/0000-0002-8699-2279
Figures
Abstract
Crimean-Congo haemorrhagic fever orthonairovirus (CCHFV) is a negative-sense, single-stranded RNA virus with a segmented genome and the causative agent of a severe Crimean-Congo haemorrhagic fever (CCHF) disease. The virus is transmitted mainly by tick species in Hyalomma genus but other ticks such as representatives of genera Dermacentor and Rhipicephalus may also be involved in virus life cycle. To improve our understanding of CCHFV adaptation to its tick species, we compared nucleotide composition and codon usage patterns among the all CCHFV strains i) which sequences and other metadata as locality of collection and date of isolation are available in GenBank and ii) which were isolated from in-field collected tick species. These criteria fulfilled 70 sequences (24 coding for S, 23 for M, and 23 for L segment) of virus isolates originating from different representatives of Hyalomma and Rhipicephalus genera. Phylogenetic analyses confirmed that Hyalomma- and Rhipicephalus-originating CCHFV isolates belong to phylogenetically distinct CCHFV clades. Analyses of nucleotide composition among the Hyalomma- and Rhipicephalus-originating CCHFV isolates also showed significant differences, mainly in nucleotides located at the 3rd codon positions indicating changes in codon usage among these lineages. Analyses of codon adaptation index (CAI), effective number of codons (ENC), and other codon usage statistics revealed significant differences between Hyalomma- and Rhipicephalus-isolated CCHFV strains. Despite both sets of strains displayed a higher adaptation to use codons that are preferred by Hyalomma ticks than Rhipicephalus ticks, there were distinct codon usage preferences observed between the two tick species. These findings suggest that over the course of its long co-evolution with tick vectors, CCHFV has optimized its codon usage to efficiently utilize translational resources of Hyalomma species.
Citation: S. Celina S, Černý J (2024) Genetic background of adaptation of Crimean-Congo haemorrhagic fever virus to the different tick hosts. PLoS ONE 19(4): e0302224. https://doi.org/10.1371/journal.pone.0302224
Editor: Engin Berber, Lerner Research Institute - Cleveland Clinic, UNITED STATES
Received: June 30, 2023; Accepted: March 28, 2024; Published: April 25, 2024
Copyright: © 2024 S. Celina, Černý. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: All relevant data are within the manuscript and its Supporting Information files.
Funding: The work of JC and SSC were funded by The Czech University of Life Sciences, grant number IGA 20223108. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
1. Introduction
The Crimean-Congo haemorrhagic fever orthonairovirus (CCHFV), recently renamed as Orthonairovirus haemorrhagiae by the International Committee on Taxonomy of Viruses (ICTV) Nairoviridae Study Group, is a member of the genus Orthonairovirus, family Nairoviridae and the etiologic agent of Crimean-Congo haemorrhagic fever (CCHF) disease [1]. It is a lipid-enveloped, single-stranded, negative-sense RNA virus with a segmented genome. Each of the three CCHFV genomic segments has a different function: S (small) segment encodes nucleocapsid protein (N), M (medium) segment the glycoproteins (Gn and Gc), and L (large) segment RNA-dependent RNA polymerases (RdRp) [2].
CCHF is the most widely distributed tick-borne viral disease in humans being endemic in many geographic regions across Africa, Asia, Eastern and Southern Europe, and the Middle East [2]. In Africa and Eurasia, CCHFV is among the deadliest human pathogens [3] and outbreaks of CCHF pose a significant threat due to its epidemic potential, high case fatality rates ranging from 5% to 80% [4], potential for nosocomial outbreaks, and challenges in treatment and prevention [5]. Therefore, due to its high potential for human-to-human transmission, the high risk of laboratory-acquired infections, and the lack of a specific vaccine, CCHFV is classified as a biosafety level 4 (BSL-4) agent [6].
The virus is transmitted through the bite of its main vector, ticks in the genus Hyalomma (Ixodidae). Hyalomma marginatum Koch, 1844 is the major vector of CCHFV in the Old World [2]. Additionally, various other species within the Hyalomma genus have also been reported to carry the CCHFV, such as H. excavatum Koch, 1844, H. lusitanicum Koch, 1844, H. rufipes Koch, 1844, and H. truncatum Koch, 1844 [7–10], playing significant roles as vectors in the Middle East, the Iberian Peninsula, and Africa, respectively [11]. However, the scope of potential vectors expands beyond Hyalomma species, with over 30 different tick species implicated in CCHFV transmission [11]. Ticks from other genera of Ixodidae, including Rhipicephalus and Dermacentor, are also capable of transmitting CCHFV. Rhipicephalus ticks, such as R. bursa Koch, 1844, and R. turanicus Pomerantsev, 1936 have been identified as carriers of CCHFV in regions spanning Albania, Turkey, Greece, and Russia [12–16]. Similarly, in genus Dermacentor, D. marginatus Sulzer, 1776 has tested positive for CCHFV in Turkey, Russia, and Spain [12,14,15,17,18], highlighting its potential as a vector for the virus. Despite the detection of CCHFV within D. marginatus eggs and its confirmed competency as a tick vector for the virus in laboratory studies, questions remain regarding their natural vectorial capacity within the enzootic cycle of the virus [11,19].
Ticks transmit CCHFV to various mammals by taking blood meals for their maturation and egg production. Nevertheless, humans can acquire CCHFV infection not only from tick bites but also from direct contact with the blood or tissues of infected animals or human patients. Other possible routes of transmission for CCHFV are through infected mother to offspring, sexual contact, and blood transfusion [20].
Due to its complex ecology, CCHFV is characterized by a great genetic diversity with complex evolutionary patterns [21]. CCHFV can be phylogenetically divided into eight distinguishable clades (Africa 1–3, Europe 1–3, and Asia 1 and 2). These clades differ not only by their geographic distribution and primary sequence but also by other features as pathogenic potentials. On the other hand, segment reassortment between the clades and long-range migration events of individual CCHFV lineages were observed, demonstrating the dynamic nature of CCHFV evolution [22,23].
In the genomes of each species, there is a distinct preference for specific synonymous codons, which encode the same amino acids, leading to unequal frequencies of codon usage within their respective genes [24,25]. This concept, referring to the differential usage of synonymous codons, is known as codon usage bias. Codon usage bias has been studied in a wide range of organisms, from prokaryotes to eukaryotes and viruses. However, how such biases arise is a much-debated area of molecular evolution. Different factors have been suggested to be related to codon usage bias, including gene expression level, guanine-cytosine content, guanine-cytosine skew, amino acid conservation, protein hydropathy, and transcriptional selection [26–29].
The codon usage bias in RNA viruses is mainly affected by compositional constraints under mutational pressure and natural selection [30,31]. Many studies on codon usage bias in various viruses have shown that the main driver shaping codon usage patterns is mutational pressure than natural selection [32–34]. However, for many DNA and RNA viruses, mutational pressure is not the only factor on establishing codon usage patterns [35,36]. Compared with prokaryotic and eukaryotic genomes, the viral genome has certain features, such as dependence on its hosts for replication, protein synthesis, and transmission of proteins. The interaction between virus and host is considered to affect survival, adaptation and evolution of virus, as well as its ability to evade the host’s immune system. In many viruses, including CCHFV, an optimal codon usage pattern has been shown to be an important factor in better adaptation of these viruses to their hosts [37,38]. Moreover, major ecological shifts in the evolution of viruses have been linked to variations in their codon usage [39]. On the other hand, codon usage pattern deoptimization leads to decrease of fitness in many viruses [40–42].
Previously, an analysis describing the adaptation of CCHFV codon usage pattern to its vertebrate hosts was performed [36], but the tick species (except for Hyalomma ticks) were not included in the study, despite the fact that arthropod vectors have at least the same effect on arbovirus evolution as their vertebrate hosts [43].
Therefore, we performed a comprehensive analysis of codon usage patterns of three genomic segments (S, M, and L) of CCHFV isolates from Hyalomma and Rhipicephalus (no sequences of CCHFV strains isolated from other tick species are available) using available sequence data. While Hyalomma species are the primary virus reservoirs and vectors for CCHFV, there is experimental evidence suggesting the potential involvement of ticks from the Rhipicephalus genus in its transmission; however, concrete evidence supporting their significant role in viral maintenance and natural transmission remains inconclusive [19]. Recent studies have detected CCHFV antigen within a Rhipicephalus species, along with viral RNA in different Rhipicephalus species across Albania, Kosovo, Greece, Turkey, and Iran [12–16,44–46]. Notably, this highlights the exclusive presence of CCHFV strains belonging to Europe 2 (clade VI) within R. bursa species, indicating their potential role as vectors of the virus [13]. Conversely, R. sanguineus sensu lato ticks are commonly linked with the Europe 1 clade. Further laboratory investigations are essential to establish the vector competence of Rhipicephalus species [13]. Moreover, a recent comprehensive review categorizes R. bursa among confirmed vectors for CCHFV [11]. This classification is based on documented infection rates, records, and observations spanning over 30 distinct tick species [11]. Given this context, our study aimed to investigate the codon usage patterns of CCHFV in relation to both Hyalomma and Rhipicephalus tick species, providing a deeper understanding of its adaptation to various vector species.
2. Materials and methods
2.1 Data collection
In this study, complete or nearly complete genome sequences of CCHFV strains isolated from tick species were analyzed, while partial sequences, that could lead to biased results in terms of codon usage, were excluded. We compared nucleotide composition and codon usage patterns among the all CCHFV strains i) which sequences and other metadata as locality of collection and date of isolation are available in GenBank and ii) which were isolated from in-field collected tick species. These criteria fulfilled 70 sequences (24 coding for S, 23 for M, and 23 for L segment) of CCHFV isolates originating from different representatives of Hyalomma and Rhipicephalus genera, and were retrieved from the National Center for Biotechnology Information (NCBI) GenBank database (https://ncbi.nlm.nih.gov/nuccore/) on December 2019, together with information about isolation date, collection locality, and tick species. Sequence details were compiled in S1 Table.
2.2 Phylogenetic analysis
Sequences were divided into groups according to CCHFV segments and aligned using MAFFT (v7.427). Maximum clade credibility (MCC) trees were constructed using Bayesian evolutionary analysis by sampling trees (BEAST, version 1.10.4) [47,48] using HKY as the nucleotide substitution model with gamma distributed rate heterogeneity and a relaxed molecular clock. The Markov chain Monte Carlo (MCMC) algorithm was executed for 100-million generations, with the initial 10% of the chain discarded as burn-in using the TreeAnnotator package integrated within the BEAST software [47]. The resulting MCC tree files were subsequently imported into the Figtree tool (http://tree.bio.ed.ac.uk/software/figtree/) for tree visualization and to estimate the time to the most recent common ancestor (tMRCA). Clades with Bayesian posterior probabilities exceeding 0.5 were displayed in the trees. A total of 70 sequences were used in this analysis.
2.3 Nucleotide composition analysis
Nucleotide compositional properties of the CCHFV coding sequences were calculated using CAIcal server (http://genomes.urv.es/CAIcal/) [49]. The overall frequency of occurrence of nucleotides (A%, C%, U% and G%), frequency of each nucleotide at the third site of synonymous codons (A3%, C3%, U3% and G3%), frequencies of occurrence of nucleotides GC at the first (GC1), second (GC2) and third synonymous codon positions (GC3), the mean frequencies of nucleotide GC at the first and the second position (GC12), overall GC and AU contents, and AU and GC contents at the third codon positions (AU3, GC3) were calculated. AUG and UGG that are only the codons for Met and Trp (no synonymous codons) along with the termination codons (UAG, UAA, and UGA) which do not encode any amino acids were excluded from the analyses since they were not expected to show any codon usage bias.
2.4 Analysis of the effective number of codons (ENC)
An effective number of codons (ENC) analysis was used to assess codon usage bias in CCHFV segments isolated from Hyalomma and Rhipicephalus ticks, calculated with CodonW software (v1.4.4) (http://sourceforge.net/projects/codonw). ENC values typically range from 20 to 61, with lower values signifying extreme codon usage bias and higher values indicating the opposite.
To explore whether the codon usage of given strains is solely due to mutational pressure or selection pressure, an ENC-plot was produced. To determine the correlation between the expected ENC and the GC3s values, the expected ENC values were calculated for different GC3s using the method of Singh et al. (2016) [50]:
where “s” indicates GC contents at the 3rd synonymous codon positions (GC3s%). When data points align with or near the standard curve, it suggests predominantly mutational pressure, while points falling below indicate codon usage subject to natural selection.
2.5 Relative Synonymous Codon Usage (RSCU) analysis
RSCU analysis was performed to compare the codon usage preferences of three different segments of CCHFV being isolated from Hyalomma or Rhipicephalus. RSCU was calculated using CAIcal server. RSCU values greater than 1 show positive codon usage bias and are described as “abundant” codons. The values less than 1 show negative codon usage bias and are described as “less-abundant” codons. A RSCU value of 1 indicates no bias in codon usage.
Hyalomma- and Rhipicephalus-isolated CCHFV coding sequences were compared with codon usage values of their natural vectors obtained from Codon and Codon Pair Usage Tables (CoCoPUTs) (https://hive.biochemistry.gwu.edu/review/codon2) [51], accessed in June 2020, and analyzed using CAIcal server.
2.6 Codon adaptation index (CAI) analysis
CAI was performed to gain insight into the codon preferences of CCHFV in relation to its tick species. The CAI values vary from 0 to 1, and higher values indicate higher levels of codon usage bias towards the codons used in highly expressed genes [40]. The most frequent codons signify the highest relative adaptation to the host, and sequences having higher CAI are known to be favoured over sequences having lower CAI. The CAI analysis was performed using CAIcal server for Hyalomma (H. anatolicum, H. asiaticum, H. dromedarii, H. excavatum, H. lusitanicum, H. marginatum, H. rufipes and H. truncatum) and Rhipicephalus (R. bursa and R. sanguineus) species. The reference data sets showing codon frequencies for these tick species were obtained from the CoCoPUTs database.
In addition, the expected CAI values (eCAI) at the 95% confidence interval were calculated in order to ascertain whether statistically significant differences in CAI values result from codon preferences, and to provide statistical support to CAI analyses, a Kolmogorov-Smirnov test for the eCAI was also applied [52].
2.7 Correspondence analysis (COA)
COA analysis was performed to detect the variation of codon usage data [58]. The COA was performed based on RSCU values for CCHFV strains (S, M, and L segments) isolated from Hyalomma and Rhipicephalus, and the distribution of the strains in the plane of the first two principal axes of COA was determined. CodonW (v1.4.4) software was used in order to examine the codon usage indices.
2.8 Selection analysis
The dN/dS ratio (ω) is used to compare the non-synonymous substitution rate per non-synonymous site (dN) with the synonymous substitution rate per synonymous site (dS). This ratio estimates the selective pressures acting on a coding sequence by identifying the fraction of codons that are evolving under purifying/negative selection (ω < 1), nearly neutral evolution (ω = 1), or positive/diversifying selection (ω > 1). To estimate the selection pressure on the S, M, and L segments of CCHFV isolated from Hyalomma and Rhipicephalus ticks, the Hypothesis Testing Using Phylogenies software v2.2.4 (HyPhy) was used [53]. Pervasive site-specific selection pressure analysis was performed in this study, utilizing Fast Unconstrained Bayesian AppRoximation (FUBAR) method accessible at the Datamonkey webserver [54] (https://www.datamonkey.org/). The dN/dS estimates for selection inference were analyzed using the posterior probability of ≥0.9.
2.9 Statistical analysis
A statistically significant difference between all nucleotide compositional properties of Hyalomma- and Rhipicephalus-isolated CCHFV strains (A, C, U, G, A3, U3, G3, C3, AU, GC, GC1, GC2, AU3, GC3, GC12), and among ENC and CAI values was addressed by applying the t-test and Wilcoxon & Mann-Whitney test with Bonferroni correction (p < 0.05) in GraphPad Prism 9.
3. Results
3.1 Hyalomma and Rhipicephalus originating CCHFV isolates are phylogenetically isolated
The phylogenetic trees of the S and L segments showed the genomic sequences were clearly assembled into clusters in relation to their tick hosts (Fig 1). The results revealed that CCHFV strains isolated from Rhipicephalus were genetically distant from Hyalomma-isolated strains. This highlighted the evidence of strong selection pressure on host adaptation.
The tree node shows the tMRCA in years. Hyalomma- and Rhipicephalus-isolated strains are distinguished in branches of different shapes (circle for Hyalomma-isolated strains and triangle for Rhipicephalus-isolated strains). Phylogenetic clades of strains are represented in the legend with different colors.
In contrast, the phylogenetic tree for the M segment did not exhibit a clear separation of Hyalomma- and Rhipicephalus-isolated CCHFV strains. Notably, CCHFV strains isolated from R. bursa in Greece (Europe 2) formed a distinct cluster from Hyalomma strains, while strains from R. bursa in Turkey (Europe 1) clustered together with Hyalomma-isolated strains. Previous studies showed that codon usage patterns in viral genomes can be influenced by geographic origins [36,50,55–58]. Therefore, the clustering of M segment strains may reflect both host adaptation and geographic distribution.
3.2. Nucleotide composition of CCHFV isolates from Hyalomma and Rhipicephalus ticks is different
Nucleotide composition strongly affects codon usage patterns. Important are especially nucleotides at the third position of codons (A3, U3, G3, C3) [37,39,66]. There were statistically significant differences between Hyalomma- and Rhipicephalus-isolated CCHFV strains in the frequency of A3 and G3 in all segments (Fig 2; S2 Table).
Nucleotide content distribution and composition in Hyalomma- and Rhipicephalus-isolated S (A), M (B), and L (C) segments, respectively. Standard deviation is marked in the plot by the error bars. Asterisk (*) shows a significant difference between variables (p < 0.05).
Further, C3 was significantly different between Hyalomma- and Rhipicephalus-isolated CCHFV strains in S and L (but not M) segments (Fig 2; S2 Table). No statistically significant difference was observed for U3 in any genomic segment of CCHFV strains isolated from Hyalomma and Rhipicephalus.
Moreover, GC content at all positions of codons (GC1, GC2, and GC3) and GC content at the first plus the second positions of codons (GC12) also play an important role in influencing overall codon usage preferences [32,59]. In S and L segments, GC content significantly differed between Hyalomma- and Rhipicephalus-isolated CCHFV strains on almost all codon positions (Fig 2; S2 Table). In M segment, GC content significantly differs between Hyalomma- and Rhipicephalus-isolated CCHFV strains only at GC3 and GC12 positions. Nevertheless, results of these analyses showed substantial differences in frequencies of occurrence of nucleotides between Hyalomma- and Rhipicephalus-isolated CCHFV variants (p < 0.05).
3.3 Hyalomma- and Rhipicephalus-isolated CCHFV strains preferentially use different codons
At S segments, G/C-ended codons were preferred over A/U-ended codons in CCHFV strains isolated from both Hyalomma and Rhipicephalus tick species. But, Rhipicephalus-isolated CCHFV strains had a strong preference to the A-ended codons whereas Hyalomma-isloated ones had a higher frequency in U-ended codons (Fig 3, Table 1).
At M segments, A/U-ended codons were highly preferred in CCHFV strains isolated from both Hyalomma and Rhipicephalus tick species. M segments of Hyalomma-isolated CCHFV strains showed a higher preference to the A-ended codons, while M segments of CCHFV strains isolated from Rhipicephalus had a preference to highly use U-ended codons. Regarding the G/C-ended codons, M segments of CCHFV strains isolated from both tick species used different codons, but both preferred to use rather C-ended than G-ended codons (Fig 3, Table 1).
At L segments, CCHFV strains isolated from both tick species also had a high preference for A/U-ended codons. Hyalomma-isolated strains prefer A-ending codons, while Rhipicephalus-isolated strains prefer U-ending codons. However, L segments of Rhipicephalus-isolated CCHFV strains had higher frequencies in C-ended codons over G-ended codons (Fig 3, Table 1).
Codon over- and underrepresentation analysis emphasized that RSCU values of the majority codons ranged from 0.6 to 1.6. Interestingly, we found that the nucleotide frequencies at the end of most over-represented codons (RSCU > 1.6) differ between Hyalomma- and Rhipicephalus-isolated CCHFV strains (Fig 4). Pro, Ala, and Arg are over-represented in all Hyalomma- and Rhipicephalus isolated CCHFV genomic segments. However, Hyalomma-isolated strains have a strong preference on GCA for Ala, while Rhipicephalus ones prefer to use GCC. Moreover, Hyalomma- and Rhipicephalus isolated L segments differ on codon usage preferences for Arg, where AGA is over-represented in Hyalomma-isolated strains, and Rhipicephalus ones show a higher tendency to use AGG codon (Fig 4).
Analysis of ENC values alone showed significant differences between the Hyalomma- and Rhipicephalus- isolated CCHFV strains for M and L but not for S segments (p < 0.05) (Fig 5A). While the S segment exhibited relatively similar ENC values for both Hyalomma (53.33 ± 1.33) and Rhipicephalus (52.85 ± 0.41), the M segment had higher ENC values for Hyalomma-isolated strains (50.89 ± 0.45) compared to Rhipicephalus-isolated strains (49.50 ± 1.91), suggesting a difference in codon usage bias among CCHFV isolates from two tick hosts. Further, the L segment showed the greatest difference in ENC values, with Hyalomma-isolated strains (51.92 ± 0.32) having markedly higher values than Rhipicephalus-isolated strains (47.64 ± 2.74), indicating a significant variation in codon usage bias of the L segment of CCHFV strains isolated from both hosts. When ENC was plotted as a function of GC3s (GC content at the third synonymous codon position), we could see a weak but significant codon usage bias for both Hyalomma- and Rhipicephalus-isolated CCHFV strains (Fig 5B). Further, it was apparent that both Hyalomma- and Rhipicephalus- isolated CCHFV strains form separated clusters. The results of COA revealed that CCHFV strains isolated from Hyalomma and Rhipicephalus ticks are separated into distinct clusters based on codons used by these strains (Fig 5C). This separation cannot be explained based on affiliation to the CCHFV phylogenetic clades. It clearly shows that Hyalomma- and Rhipicephalus-isolated CCHFV strains have considerable variation in codon usage patterns.
A) Comparison of the values for ENC between Hyalomma- and Rhipicephalus-isolated CCHFV strains for S, M, and L segments. Standard deviation is marked in the plot by the error bars. B) ENC-GC3 plots of Hyalomma- and Rhipicephalus-isolated CCHFV genomes. ENC values (Y-axis) was plotted against the GC content at the third synonymous codon positions (GC3s values, X-axis). The curve (red line) indicates the expected codon usage if GC compositional constraints alone account for the codon usage bias. Hyalomma- and Rhipicephalus-isolated CCHFV strains are marked by different symbols (circle for Hyalomma-isolated strains and triangle for Rhipicephalus-isolated strains). Different colors mark for different CCHFV phylogenetic clades. C) COA values are based on the RSCU values of Hyalomma- and Rhipicephalus-isolated S, M, and L segments, respectively. D) nCAI values of Hyalomma-isolated strains to Hyalomma and Rhipicephalus codon usage, and Rhipicephalus-isolated strains to Hyalomma and Rhipicephalus codon usage in relation to S, M, and L segments, respectively.
3.4 Both Hyalomma- and Rhipicephalus-isolated CCHFV strains show higher codon adaptation index values for Hyalomma tick species
All genomic segments of both Hyalomma- and Rhipicephalus-isolated CCHFV strains show significantly higher codon adaptation index values for Hyalomma than for Rhipicephalus tick species (p < 0.05). Interesting is the fact that M and L segments of Rhipicephalus-isolated CCHFV strains show significantly higher CAI values to Hyalomma ticks than the M and L segments of Hyalomma-isolated CCHFV strains (p = 0.0001) (Fig 5D).
3.5 The pattern of selection in Hyalomma- and Rhipicephalus-isolated CCHFV strains varies
We utilized the FUBAR method, a site-specific pervasive selection approach, to assess the selective pressure on 70 CCHFV sequences isolated from both Hyalomma and Rhipicephalus ticks. The estimated dN/dS ratios were as follows: 0.04 and 0.02 for S segment, 0.05 and 0.17 for M segment, and 0.04 and 0.07 for L segment isolated from Hyalomma and Rhipicephalus, respectively. The results further revealed that the Hyalomma- and Rhipicephalus-isolated S segments had 380 and 103 sites identified under negative/purifying selection, respectively. For the Hyalomma- and Rhipicephalus-isolated M segments, 1696 and 397 sites were identified under negative/purifying selection, respectively. However, no sites were identified under positive/diversifying selection in both segments isolated from two ticks. Lastly, on the Hyalomma- and Rhipicephalus-isolated L segments, 2858 and 334 sites were identified under negative/purifying selection, respectively. Additionally, 1 site was identified under positive/diversifying selection for Rhipicephalus-isolated L segments.
Overall, these findings suggest that there are variations in the selective pressures acting on CCHFV isolates of different tick hosts, which may be influenced by the segment-specific codon usage biases and the tick species from which they were isolated.
4. Discussion
The differences in genome composition and codon usage patterns between Hyalomma- and Rhipicephalus-isolated CCHFV variants can influence viral fitness, evolution, and the ability to replicate within different tick species as well as mammalian hosts.
Phylogenies of S and L genomic segments of Hyalomma- and Rhipicephalus-isolated CCHFV variants revealed a clear phylogenetic separation of Rhipicephalus-isolated CCHFV strains from Hyalomma ones. These findings diverge from previous phylogenetic analyses, which primarily emphasized the influence of geographical spots on codon usage patterns [36]. Instead, our study highlights the evidence of strong selection pressure on host adaptation, which is in agreement with CAI analysis. Our findings suggest that, beyond geographical factors, vector host species may significantly impact the codon usage patterns of the virus. Interestingly, the results of the phylogenetic analysis showed a phylogenetic separation among Europe 2 and Europe 1 clades in terms of the ticks in which the strains are vectored. Europe 1 circulates in various geographic regions, including southern Russia, Turkey, the Balkan Peninsula, and Iran. The strains belong to Europe 1 (clade V) are known to be highly pathogenic in humans and Hyalomma species were observed to harbor particularly this clade. Likewise, the infection caused by strains belonging to Europe 2 (clade VI) has mild or non-pathogenic effects on humans, and this clade is exclusively found in ticks of the Rhipicephalus genus, in particular R. bursa [13,60]. In contrast to the clear separation observed in the phylogenetic trees of the S and L segments based on hosts, the M segment tree did not reveal a distinct clustering of Hyalomma- and Rhipicephalus-isolated CCHFV strains. Previous studies showed that geographic origins might influence the codon usage patterns of the viral genome [36,50,55–58], and our findings suggest that a combination of both host adaptation and geographical origin may have contributed to the observed patterns in codon usage bias of Hyalomma- and Rhipicephalus-isolated CCHFV M segments. Additionally, potential variations in two different ticks hosting the same genetic structure of the virus should be explored, utilizing artificial feeding experiments to facilitate such investigations, thereby offering valuable suggestions for further studies in this field.
Previously, it has been shown that codon usage bias, or the preference for one type of codon over another, can be greatly influenced by the overall nucleotide composition in the genome [32,59]. The nucleotide composition analysis revealed substantial differences on frequencies of occurrence of nucleotides between Hyalomma- and Rhipicephalus-isolated CCHFV variants. Despite the differences, it is apparent that the M and L segments of Hyalomma- and Rhipicephalus-isolated CCHFV strains are AU- rich; and A/U-ended codons appear to be preferred, indicating that the usage of optional codons might be influenced by compositional constraints resulting in the presence of mutational pressure. This is consistent with previous report indicating a substantial portion of mammalian-host isolated CCHFV strains are enriched with AU [36]. Further, this result is similar to the other RNA viruses such as West Nile virus [61], dengue virus [62], Marburg virus [59], Ebola virus [63] and bluetongue virus [57] where A/U-ended codons appear to be preferred. However, S segments isolated from both tick species are more GC rich and preferentially use G/C-ended codons. The biological importance of this condition is uncertain; therefore, it is important to investigate the factors influencing different nucleotide frequencies of CCHFV segments [64].
Previous studies on codon usage bias have also suggested that the composition of amino acids is a key factor in determining the nucleotide contents at the first and second codon positions of viral genomes, while the variation in proteins was forced by functional selection. However, at the third codon positions of a viral gene, a large proportion of the possible alterations (69%) result in synonymous or silent mutations that are not constrained by the functional selection of amino acids [64]. Based on RSCU values, we explored the different codon usage preferences of Hyalomma- and Rhipicephalus-isolated CCHFV sequences. In the previous study, it was observed that Hyalomma-isolated CCHFV strains exhibited a preference for C-ended codons [36]. However, our study revealed that only CCHFV S segments isolated from Hyalomma and Rhipicephalus ticks demonstrated a preference for C-ended codons, while M and L segments show contrary preferences toward A- and U-ended codons. Remarkably, Hyalomma-isolated M segments have a higher tendency to use A-ended codons, while strains isolated from Rhipicephalus have a strong preference to use U-ended codons. For L segment, although the two hosts have similar preferences for having more A-ended codons over U, they exhibit different codon preferences for the same amino acids such as Arg, Cys, and Thr. Furthermore, the previous study noted that CCHFV strains isolated from Hyalomma ticks exhibited a preference for CGC codons for Arg, while AGA and AGG were less favored [36]. Conversely, our analysis revealed an over-representation of AGA and AGG codons across S, M, and L segments of CCHFV isolates from Hyalomma and Rhipicephalus ticks. Regarding the codon preference for Ala, the previous study reported a strong preference for GCC in Hyalomma-isolated strains [36]. However, our findings showed that GCA was the preferred codon for Ala in Hyalomma-isolated S, M, and L segments. Additionally, our study revealed a strong preference for CCA codons for Pro in both Hyalomma- and Rhipicephalus-isolated CCHFV strains. This contrasts with the findings of the previous study [36], which indicated a preference for CCC codons in Hyalomma-isolated CCHFV strains.
All genomic segments of Hyalomma- and Rhipicephalus-isolated CCHFV strains have remarkably different codon usage patterns. The codon usage bias of CCHFV isolated from two tick genera was found to be low, with ENC values higher than 35. Similar low codon usage bias has also been reported among several other RNA viruses for instance, Zika virus (ENC: 53.32) [58], Ebola virus (ENC: 57.23) [65], chikungunya virus (ENC: 55.56) [35], classical swine fever virus (ENC: 51.7) [66], foot-and-mouth virus (ENC: 51.53) [67], hepatitis C virus (ENC: 52.62) [68], Venezuelan equine encephalitis virus (ENC: 56.51) [56], and West Nile virus (ENC: 53.81) [61]. It has been indicated that the low codon usage bias of the virus is beneficial for the efficient replication in its host cells and the reduced competition between virus and its hosts for protein synthesis. Similarly, all segments of tick-isolated CCHFV strains show rather high ENC values which indicate low codon usage bias. It suggests that the evolution of low codon bias within CCHFV coding sequences has allowed it to successfully maintain its survival cycle within its tick vectors as well as mammalian hosts.
The influence of a tick vector on CCHFV codon usage pattern was also visible from CAI analysis. Comparative analysis of CAI values revealed that both Hyalomma- and Rhipicephalus-isolated CCHFV strains display higher adaptation to use the codons that are preferred by Hyalomma tick species. These results suggest that over the course of its long co-evolution with tick vectors, CCHFV has optimized its codon usage patterns to utilize the translational resources of Hyalomma species more efficiently than that of Rhipicephalus ticks which are vectors used only by specific CCHFV. Higher genetic adaptation of CCHFV strains isolated from the tick species of two genera, favoring the use of codons preferred by Hyalomma ticks, can be attributed to the role of Hyalomma species as the primary vectors of CCHFV. Contrarily, a lower adaptation of Hyalomma- and Rhipicephalus-isolated CCHFV segments to Rhipicephalus codon usage highlights that Rhipicephalus ticks are rather occasional vectors or evolutionary new vectors that are used by CCHFV in areas where Hyalomma ticks are either absent or at least rare.
The selection analysis of Hyalomma- and Rhipicephalus- isolated S, M, and L segments suggest that CCHFV isolates from these hosts are subject to strong purifying (negative) selection, as has been previously observed for other RNA viruses such as Zika virus, the West Nile virus, the dengue virus, the yellow fever virus, and the tick-borne encephalitis virus, which exhibit low dN/dS ratios ranging from 0.019 to 0.066 [69].
Overall, the results of this study show the strong difference in codon usage patterns between the CCHFV strains isolated from different tick species, which may mirror the differences in evolutionary processes that forms these virus strains. However, our study has identified a few limitations that should be considered when interpreting our findings. Firstly, the sample size of 70 sequences (24 for S segment, 23 for M segment, and 23 for L segment) of CCHFV isolates from different representatives of the Hyalomma and Rhipicephalus genera may not be sufficient to provide robust conclusions. Additionally, there are limitations in the amount of full CCHFV genome sequences isolated from Rhipicephalus and the lack of CCHFV sequences from Dermacentor species, despite experimental evidence suggesting their potential involvement in virus transmission and the frequent detection of viral RNA in these species [11–16,18,19,44–46]. Thus, generating full genome sequences of CCHFV strains isolated from Dermacentor ticks could provide valuable insights into the virus’s adaptation to additional tick species. These limitations highlight the importance of generating full genome sequences of CCHFV strains from their tick hosts to achieve a more thorough understanding of the virus-vector coevolution. Despite a few limitations, this research not only provided knowledge about the variation in CCHFV codon usage patterns in relation to their two vectors but also contributed to analyzing the factors that influence the adaptation of the virus to its vector species. In silico studies are highly important in the case of CCHFV as it is regarded to be a BSL-4 pathogen, and therefore, studies on the virus are very limited.
5. Conclusion
Our study suggests that analysis of codon usage bias of Hyalomma- and Rhipicephalus-isolated CCHFV strains can provide an alternative strategy to understand the evolution and genetic background of adaptation of CCHFV to its vector species. Our findings indicate that CCHFV strains isolated from Hyalomma and Rhipicephalus have significant differences in codon usage variations and patterns. Furthermore, our study highlighted that Hyalomma- and Rhipicephalus-isolated CCHFV strains have a higher tendency to use the codons that are preferred by species of Hyalomma genus. The results of this study indicate the strong effect of evolutionary processes on codon usage patterns and highlight the evidence of strong selection pressure on host adaptation while codon usage bias patterns in Hyalomma- and Rhipicephalus-isolated CCHFV M segments may result from a combination of host adaptation and geographical origin. The data analyzed in this study contribute to our understanding of the virus’s evolution and genetic adaptation to its vector species.
Supporting information
S1 Table. The strain name, accession number, origin, isolation host and collection date of polyprotein-coding region of each CCHFV isolates used in this study.
https://doi.org/10.1371/journal.pone.0302224.s001
(XLSX)
S2 Table. Nucleotide composition analysis of S, M, and L segments of CCHFV isolated from Hyalomma and Rhipicephalus species (%).
Significant values on the 95% confidence limit between strains isolated from two tick hosts are marked in bold (p<0.05).
https://doi.org/10.1371/journal.pone.0302224.s002
(XLSX)
References
- 1. Thomas Postler, Kuhn H Jens. Rename all species in the family to comply with the ICTV-mandated binomial format (Bunyavirales: Nairoviridae). ICTV [International Committee on Taxonomy of Viurses] TaxoProp 2021.017M.N.v1.Nairoviridae_sprenamed. 2021. 2021 [cited 30 Jun 2023]. Available: https://talk.ictvonline.org/files/proposals/animal_dsrna_and_ssrna-_viruses/m/animal_rna_minus_under_consideration/12837.
- 2. Bente DA, Forrester NL, Watts DM, McAuley AJ, Whitehouse CA, Bray M. Crimean-Congo hemorrhagic fever: History, epidemiology, pathogenesis, clinical syndrome and genetic diversity. Antiviral Res. 2013;100: 159–189. pmid:23906741
- 3. Shayan S, Bokaean M, Shahrivar MR, Chinikar S. Crimean-Congo Hemorrhagic Fever. Lab Med. 2015;46: 180–189. pmid:26199256
- 4. Mertens M, Schmidt K, Ozkul A, Groschup MH. The impact of Crimean-Congo hemorrhagic fever virus on public health. Antiviral Res. 2013;98: 248–260. pmid:23458713
- 5. WHO. Crimean-Congo haemorrhagic fever. 31 Jan 2013 [cited 26 Sep 2022]. Available: https://www.who.int/news-room/fact-sheets/detail/crimean-congo-haemorrhagic-fever.
- 6. Ergönül Ö. Crimean-Congo haemorrhagic fever. Lancet Infect Dis. 2006;6: 203–214. pmid:16554245
- 7. Akuffo R, Brandful JAM, Zayed A, Adjei A, Watany N, Fahmy NT, et al. Crimean-Congo hemorrhagic fever virus in livestock ticks and animal handler seroprevalence at an abattoir in Ghana. BMC Infect Dis. 2016;16: 324. pmid:27392037
- 8. Gonzalez JP, Cornet JP, Wilson ML, Camicas JL. Crimean-Congo haemorrhagic fever virus replication in adult Hyalomma truncatum and Amblyomma variegatum ticks. Res Virol. 1991;142: 483–488. pmid:1803413
- 9. Madbouly MH, Kemp GE, Causey OR, David-West TamS. Congo Virus from Domestic Livestock, African Hedgehog, and Arthropods in Nigeria. Am J Trop Med Hyg. 1970;19: 846–850. pmid:5453910
- 10. Estrada-Peña A, Palomar AM, Santibáñez P, Sánchez N, Habela MA, Portillo A, et al. Crimean-Congo hemorrhagic fever virus in ticks, Southwestern Europe, 2010. Emerg Infect Dis. 2012;18: 179–80. pmid:22261502
- 11. Nasirian H. Ticks infected with Crimean-Congo hemorrhagic fever virus (CCHFV): A decision approach systematic review and meta-analysis regarding their role as vectors. Travel Med Infect Dis. 2022;47: 102309. pmid:35318129
- 12. Orkun Ö, Karaer Z, Çakmak A, Nalbantoğlu S. Crimean-Congo hemorrhagic fever virus in ticks in Turkey: A broad range tick surveillance study. Infection, Genetics and Evolution. 2017;52: 59–66. pmid:28433738
- 13. Anna Papa, Anastasia Kontana, Katerina Tsioka, Ilias Chaligiannis, Smaragda Sotiraki. Molecular detection of Crimean-Congo hemorrhagic fever virus in ticks, Greece, 2012–2014. Parasitol Res. 2017;116: 3057–3063. pmid:28920177
- 14. Tsapko N V., Volynkina AS, Evchenko AY, Lisitskaya Y V, Shaposhnikova LI. Detection of Crimean-Congo hemorrhagic fever virus in ticks collected from South Russia. Ticks Tick Borne Dis. 2022;13: 101890. pmid:34953335
- 15. Yesilbag K, Aydin L, Dincer E, Alpay G, Girisgin AO, Tuncer P, et al. Tick survey and detection of Crimean-Congo hemorrhagic fever virus in tick species from a non-endemic area, South Marmara region, Turkey. Exp Appl Acarol. 2013;60: 253–261. pmid:23229492
- 16. Papa A, Velo E, Kadiaj P, Tsioka K, Kontana A, Kota M, et al. Crimean-Congo hemorrhagic fever virus in ticks collected from livestock in Albania. Infection, Genetics and Evolution. 2017;54: 496–500. pmid:28827176
- 17. Gargili A, Midilli K, Ergonul O, Ergin S, Alp HG, Vatansever Z, et al. Crimean-Congo Hemorrhagic Fever in European Part of Turkey: Genetic Analysis of the Virus Strains from Ticks and a Seroepidemiological Study in Humans. Vector-Borne and Zoonotic Diseases. 2011;11: 747–752. pmid:21028961
- 18. Moraga‐Fernández A, Ruiz‐Fons F, Habela MA, Royo‐Hernández L, Calero‐Bernal R, Gortazar C, et al. Detection of new Crimean–Congo haemorrhagic fever virus genotypes in ticks feeding on deer and wild boar, Spain. Transbound Emerg Dis. 2021;68: 993–1000. pmid:32738065
- 19. Gargili A, Estrada-Peña A, Spengler JR, Lukashev A, Nuttall PA, Bente DA. The role of ticks in the maintenance and transmission of Crimean-Congo hemorrhagic fever virus: A review of published field and laboratory studies. Antiviral Res. 2017;144: 93–119. pmid:28579441
- 20. Pshenichnaya NY, Leblebicioglu H, Bozkurt I, Sannikova IV, Abuova GN, Zhuravlev AS, et al. Crimean-Congo hemorrhagic fever in pregnancy: A systematic review and case series from Russia, Kazakhstan and Turkey. International Journal of Infectious Diseases. 2017;58: 58–64. pmid:28249811
- 21. Anagnostou V, Papa A. Evolution of Crimean-Congo Hemorrhagic Fever virus. Infection, Genetics and Evolution. 2009;9: 948–954. pmid:19560561
- 22. Chinikar S, Ghiasi SM, Moradi M, Goya MM, Shirzadi MR, Zeinali M, et al. Geographical Distribution and Surveillance of Crimean-Congo Hemorrhagic Fever in Iran. Vector-Borne and Zoonotic Diseases. 2010;10: 705–708. pmid:20854025
- 23. Hewson R, Gmyl A, Gmyl L, Smirnova SE, Karganova G, Jamil B, et al. Evidence of segment reassortment in Crimean-Congo haemorrhagic fever virus. Journal of General Virology. 2004;85: 3059–3070. pmid:15448369
- 24. Plotkin JB, Kudla G. Synonymous but not the same: the causes and consequences of codon bias. Nat Rev Genet. 2011;12: 32–42. pmid:21102527
- 25. Plotkin JB, Dushoff J, Desai MM, Fraser HB. Codon Usage and Selection on Proteins. J Mol Evol. 2006;63: 635–653. pmid:17043750
- 26. Chen Y, Chen Y-F. Analysis of synonymous codon usage patterns in duck hepatitis A virus: a comparison on the roles of mutual pressure and natural selection. Virusdisease. 2014;25: 285–293. pmid:25674595
- 27. Ermolaeva MD. Synonymous codon usage in bacteria. Curr Issues Mol Biol. 2001;3: 91–7. pmid:11719972
- 28. Knight RD, Freeland SJ, Landweber LF. A simple model based on mutation and selection explains trends in codon and amino-acid usage and GC composition within and across genomes. Genome Biol. 2001;2: research0010.1. pmid:11305938
- 29. Quax TEF, Claassens NJ, Söll D, van der Oost J. Codon Bias as a Means to Fine-Tune Gene Expression. Mol Cell. 2015;59: 149–161. pmid:26186290
- 30. Shields D, Sharp P, Higgins D, Wright F. “Silent” sites in Drosophila genes are not neutral: evidence of selection among synonymous codons. Mol Biol Evol. 1988. pmid:3146682
- 31. Stenico M, Lloyd AT, Sharp PM. Codon usage in Caenorhabditis elegans: delineation of translational selection and mutational biases. Nucleic Acids Res. 1994;22: 2437–2446. pmid:8041603
- 32. Jenkins GM, Holmes EC. The extent of codon usage bias in human RNA viruses and its evolutionary origin. Virus Res. 2003;92: 1–7. pmid:12606071
- 33. Sharp PM, Emery LR, Zeng K. Forces that influence the evolution of codon bias. Philosophical Transactions of the Royal Society B: Biological Sciences. 2010;365: 1203–1212. pmid:20308095
- 34. Wong EH, Smith DK, Rabadan R, Peiris M, Poon LL. Codon usage bias and the evolution of influenza A viruses. Codon Usage Biases of Influenza Virus. BMC Evol Biol. 2010;10: 253. pmid:20723216
- 35. Butt AM, Nasrullah I, Tong Y. Genome-Wide Analysis of Codon Usage and Influencing Factors in Chikungunya Viruses. PLoS One. 2014;9: e90905. pmid:24595095
- 36. Rahman SU, Yao X, Li X, Chen D, Tao S. Analysis of codon usage bias of Crimean-Congo hemorrhagic fever virus and its adaptation to hosts. Infection, Genetics and Evolution. 2018;58: 1–16. pmid:29198972
- 37. dos Passos Cunha M, Ortiz-Baez AS, de Melo Freire CC, de Andrade Zanotto PM. Codon adaptation biases among sylvatic and urban genotypes of Dengue virus type 2. Infection, Genetics and Evolution. 2018;64: 207–211. pmid:29792991
- 38. Li G, Shi L, Zhang L, Xu B. Componential usage patterns in dengue 4 viruses reveal their better evolutionary adaptation to humans. Front Microbiol. 2022;13. pmid:36204606
- 39. Di Paola N, Freire CC de M, Zanotto PM de A. Does adaptation to vertebrate codon usage relate to flavivirus emergence potential? PLoS One. 2018;13: e0191652. pmid:29385205
- 40. de Fabritus L, Nougairède A, Aubry F, Gould EA, de Lamballerie X. Attenuation of Tick-Borne Encephalitis Virus Using Large-Scale Random Codon Re-encoding. PLoS Pathog. 2015;11: e1004738. pmid:25734338
- 41. Nougairede A, De Fabritus L, Aubry F, Gould EA, Holmes EC, de Lamballerie X. Random Codon Re-encoding Induces Stable Reduction of Replicative Fitness of Chikungunya Virus in Primate and Mosquito Cells. PLoS Pathog. 2013;9: e1003172. pmid:23436995
- 42. Manokaran G, Sujatmoko , McPherson KG, Simmons CP. Attenuation of a dengue virus replicon by codon deoptimization of nonstructural genes. Vaccine. 2019;37: 2857–2863. pmid:31000413
- 43. Sexton NR, Ebel GD. Effects of Arbovirus Multi-Host Life Cycles on Dinucleotide and Codon Usage Patterns. Viruses. 2019;11. pmid:31336898
- 44. Sherifi K, Rexhepi A, Berxholi K, Mehmedi B, Gecaj RM, Hoxha Z, et al. Crimean-Congo Hemorrhagic Fever Virus and Borrelia burgdorferi sensu lato in Ticks from Kosovo and Albania. Front Vet Sci. 2018;5: 38. pmid:29560357
- 45. Sherifi K, Cadar D, Muji S, Robaj A, Ahmeti S, Jakupi X, et al. Crimean-Congo Hemorrhagic Fever Virus Clades V and VI (Europe 1 and 2) in Ticks in Kosovo, 2012. PLoS Negl Trop Dis. 2014;8: e3168. pmid:25255381
- 46. Fakoorziba MR, Naddaf-Sani AA, Moemenbellah-Fard MD, Azizi K, Ahmadnia S, Chinikar S. First phylogenetic analysis of a Crimean-Congo hemorrhagic fever virus genome in naturally infected Rhipicephalus appendiculatus ticks (Acari: Ixodidae). Arch Virol. 2015;160: 1197–1209. pmid:25742932
- 47. Drummond AJ, Rambaut A. BEAST: Bayesian evolutionary analysis by sampling trees. BMC Evol Biol. 2007;7: 214. pmid:17996036
- 48. Suchard MA, Lemey P, Baele G, Ayres DL, Drummond AJ, Rambaut A. Bayesian phylogenetic and phylodynamic data integration using BEAST 1.10. Virus Evol. 2018;4. pmid:29942656
- 49. Puigbò P, Bravo IG, Garcia-Vallve S. CAIcal: A combined set of tools to assess codon usage adaptation. Biol Direct. 2008;3: 38. pmid:18796141
- 50. Singh NK, Tyagi A, Kaur R, Verma R, Gupta PK. Characterization of codon usage pattern and influencing factors in Japanese encephalitis virus. Virus Res. 2016;221: 58–65. pmid:27189042
- 51. Alexaki A, Kames J, Holcomb DD, Athey J, Santana-Quintero L V., Lam PVN, et al. Codon and Codon-Pair Usage Tables (CoCoPUTs): Facilitating Genetic Variation Analyses and Recombinant Gene Design. J Mol Biol. 2019;431: 2434–2441. pmid:31029701
- 52. Puigbò P, Bravo IG, Garcia-Vallvé S. E-CAI: a novel server to estimate an expected value of Codon Adaptation Index (eCAI). BMC Bioinformatics. 2008;9: 65. pmid:18230160
- 53. Pond SLK, Frost SDW, Muse S V. HyPhy: hypothesis testing using phylogenies. Bioinformatics. 2005;21: 676–679. pmid:15509596
- 54. Murrell B, Moola S, Mabona A, Weighill T, Sheward D, Kosakovsky Pond SL, et al. FUBAR: A Fast, Unconstrained Bayesian AppRoximation for Inferring Selection. Mol Biol Evol. 2013;30: 1196–1205. pmid:23420840
- 55. Chen Y, Shi Y, Deng H, Gu T, Xu J, Ou J, et al. Characterization of the porcine epidemic diarrhea virus codon usage bias. Infection, Genetics and Evolution. 2014;28: 95–100. pmid:25239728
- 56. Wang H, Liu S, Lv Y, Wei W. Codon usage bias of Venezuelan equine encephalitis virus and its host adaption. Virus Res. 2023;328: 199081. pmid:36854361
- 57. Yao X, Fan Q, Yao B, Lu P, Rahman SU, Chen D, et al. Codon Usage Bias Analysis of Bluetongue Virus Causing Livestock Infection. Front Microbiol. 2020;11. pmid:32508755
- 58. Wang H, Liu S, Zhang B, Wei W. Analysis of Synonymous Codon Usage Bias of Zika Virus and Its Adaption to the Hosts. PLoS One. 2016;11: e0166260. pmid:27893824
- 59. Nasrullah I, Butt AM, Tahir S, Idrees M, Tong Y. Genomic analysis of codon usage shows influence of mutation pressure, natural selection, and host features on Marburg virus evolution. BMC Evol Biol. 2015;15: 174. pmid:26306510
- 60. Volynkina A, Lisitskaya Y, Kolosov A, Shaposhnikova L, Pisarenko S, Dedkov V, et al. Molecular epidemiology of Crimean-Congo hemorrhagic fever virus in Russia. PLoS One. 2022;17: e0266177. pmid:35551538
- 61. Moratorio G, Iriarte A, Moreno P, Musto H, Cristina J. A detailed comparative analysis on the overall codon usage patterns in West Nile virus. Infection, Genetics and Evolution. 2013;14: 396–400. pmid:23333335
- 62. Lara-Ramírez EE, Salazar MI, López-López M de J, Salas-Benito JS, Sánchez-Varela A, Guo X. Large-Scale Genomic Analysis of Codon Usage in Dengue Virus and Evaluation of Its Phylogenetic Dependence. Biomed Res Int. 2014;2014: 1–9. pmid:25136631
- 63. Luo W, Roy A, Guo F, Irwin DM, Shen X, Pan J, et al. Host Adaptation and Evolutionary Analysis of Zaire ebolavirus: Insights From Codon Usage Based Investigations. Front Microbiol. 2020;11. pmid:33224111
- 64. van Hemert F, van der Kuyl AC, Berkhout B. Impact of the biased nucleotide composition of viral RNA genomes on RNA structure and codon usage. Journal of General Virology. 2016;97: 2608–2619. pmid:27519195
- 65. Cristina J, Moreno P, Moratorio G, Musto H. Genome-wide analysis of codon usage bias in Ebolavirus. Virus Res. 2015;196: 87–93. pmid:25445348
- 66. Tao P, Dai L, Luo M, Tang F, Tien P, Pan Z. Analysis of synonymous codon usage in classical swine fever virus. Virus Genes. 2009;38: 104–112. pmid:18958611
- 67. Zhong J, Li Y, Zhao S, Liu S, Zhang Z. Mutation pressure shapes codon usage in the GC-Rich genome of foot-and-mouth disease virus. Virus Genes. 2007;35: 767–776. pmid:17768673
- 68. Hu J, Wang Q, Zhang J, Chen H, Xu Z, Zhu L, et al. The characteristic of codon usage pattern and its evolution of hepatitis C virus. Infection, Genetics and Evolution. 2011;11: 2098–2102. pmid:21907310
- 69. Lin J-J, Bhattacharjee MJ, Yu C-P, Tseng YY, Li W-H. Many human RNA viruses show extraordinarily stringent selective constraints on protein evolution. Proceedings of the National Academy of Sciences. 2019;116: 19009–19018. pmid:31484772