Insights on Zika virus envelope gene conservation in American outbreaks

Abstract

Phylogenetic studies with Zika virus (ZIKV) have been conducted in Brazil. In this study, we sequenced 8 new sequences of the ZIKV envelope (E) gene from strains of cases from the Paraná and Mato Grosso do Sul states in 2016. A low phylogenetic signal was observed, with more than 40% of unresolved quartets, and the Maximum Likelihood Tree grouped all sequences in the Brazilian branches within the Asian genotype. In addition, a Shannon entropy analysis was conducted, showing a high stability in the E protein through the ZIKV polyprotein. Taken together, these results suggest a high degree of conservation in the ZIKV E gene from the recent American outbreaks.

Electronic supplementary material

The online version of this article (10.1007/s42770-020-00349-3) contains supplementary material, which is available to authorized users.

Between the years of 2015 and 2016, Brazil was surprised by an outbreak of an exanthematous disease with dengue-like symptoms [1–3]. Clinical investigations led to the finding that the causative agent was Zika virus (ZIKV), and over the course of the outbreak, the association with serious sequelae (Congenital Zika Syndrome (CZS) in children whose mothers were infected during pregnancy, and Guillain-Barre) was established [4]. Since then, the number of new ZIKV cases has decreased substantially, and by the end of 2019, the incidence of confirmed cases of ZIKV infection was only 5.1 cases per 100,000 inhabitants in Brazil [5].

Phylogenetic analyses have shown the occurrence of two main genotypes of ZIKV: African and Asian [6, 7]. Since 2015, ZIKV sequences obtained from outbreaks that occurred in different regions of Brazil were clustered within the Asian genotype [2, 3]. Here, using new ZIKV sequences from strains that occurred in the Paraná and Mato Grosso do Sul states, we presented a phylogenetic analysis based on the E gene.

Serum samples from patients were provided by Laboratório Central do Estado do Paraná (LACEN-PR). The diagnosis of ZIKV infection was defined by the qRT-PCR method, using a set of primers specific for E gene, as previously described [7]. A cut-off was established in a cycle threshold (CT) of 40. Total RNA was extracted from 140 μL of serum using the QIAmp® Viral RNA Mini kit (Qiagen, Courtaboeuf, France), following the manufacturer’s instructions, being immediately used for cDNA synthesis (Superscript III, Invitrogen, Carlsbad, CA) according to the manufacturer’s protocol. The use of human samples was approved by the Permanent Research Ethics Committee of the State University of Maringá (Protocol no. 1.751.311).

In order to sequence the entire ZIKV E gene, two pairs of primers were designed: ZVF1/ZVF1.2 and ZVF2/ZVR2.1 (Online resource – Table S1). Two PCR reactions were conducted: The first reaction was prepared to 20 μL, with 1X Buffer, 0.2 mM of each dNTP (10 mM dNTP mix), 1.5-mM MgCl₂, 1 U of Taq DNA polymerase (Invitrogen, Carlsbad, CA), primers ZVF1 and ZVR2.1 (0.5 μM each), and 4 μL of ZIKV cDNA. The conditions were initial denaturation at 94 °C for 5 min, 30 cycles of 94 °C for 1 min, 60 °C for 30 s, 72 °C for 2 min, followed by final extension at 72 °C for 10 min. For the second reaction, two semi-nested PCR were performed. Both were prepared to 20 μL, with 1X Buffer, 0.062 mM of each dNTP (10 mM dNTP mix), and 1.5-mM MgCl₂; primers ZVF1 and ZVR1.2 (0.5 μM each) and primers ZVF2 and ZVR2.1 (0.5 μM each) were used for the two fragments, respectively. Further, 1 U of Taq DNA polymerase (Invitrogen, Carlsbad, CA) and 2 μL of the product obtained in the first PCR were used. The PCR conditions were initial denaturation at 94 °C for 5 min, 35 cycles of 94 °C for 1 min, 58 °C for 30 s, 72 °C for 1 min and 30 s, followed by final extension at 72 °C for 10 min. The PCR products were sequenced by ACTGene Análises Moleculares Ltda. (Centro de Biotecnologia, UFRGS, Porto Alegre, RS, Brazil) using the ABI Prism® BigDye™ Terminator v3.1 (Applied Biosystems, Foster City, CA) and the AB 3500 Genetic Analyzer (Applied Biosystems, Foster City, CA). All ZIKV E gene sequences were assembled by the Phred/Phrap program and contigs were formed for each sequence [8, 9].

A data set with the ZIKV E gene sequences obtained in this study and 167 complete ZIKV sequences recovered from Genbank, representing different regions and genotypes, was created (Online resource – Table S2). The softwares MUSCLE [10] and jModelTest 2.1 [11] were used to align and determine the best-fit model of nucleotide substitution, respectively, and sequences were trimmed in order to keep the E gene sequence for the whole dataset. The phylogenetic signal of the ZIKV E gene was determined using the Tree-Puzzle 5.3 software [12, 13] and a phylogenetic tree by the Maximum Likelihood Method was constructed using the PhyML package as implemented in SEAVIEW software [14], with 1000 bootstrap pseudo-replicates [15]. All sequences were deposited in GenBank under the accession numbers MH379777 to MH379784 (Envelope gene).

Of the 33 samples provided by LACEN-PR, only 8 were sequenced, 4 from the state of Mato Grosso do Sul and 4 from the state of Paraná, Brazil. One of possible reasons for our poor recovery of samples that could be sequenced was the low viral load of ZIKV in the peripheral blood of patients (Online resource – Table S3). All sequences obtained in this study were grouped in the Brazilian branches within the Asian genotype (Fig. 1).

Fig. 1.

Maximum Likelihood Phylogenetic Tree using aligned ZIKV E gene sequences and obtained after 1000 pseudo bootstrap replicates. The GTR model was used with a variation of four Gamma distribution categories. The bootstrap value is above the clades that separate ZIKV genotypes. Clades highlighted with red color represent the East African genotype, clades highlighted in green represent the West African genotype, and clades highlighted in white the Asian/Americas genotype. The sequences obtained in this study, corresponding to the states of Paraná and Mato Grosso do Sul, are highlighted in red and yellow oblongs, respectively

The quartets analysis showed a low phylogenetic signal, with more than 40% of unresolved quartets (Online resource – Fig. S1). An informational entropy analysis of Shannon, based on the entire ZIKV polyprotein, was performed in BioEdit 7.2.0 software [16] using the same sequences of the phylogenetic analysis (Online resource - Table S2) and excluding sequences of E gene obtained in this study. Entropy analysis was used in order to evaluate the variability and complexity in an amino acid position, being calculated in bits using the Shannon formula: H(l) = − Σf(b,l)log(base 2)f(b,l), where H is the entropy at the position l and f(b,l) is the frequency with which residue b is found at the position l. As a result, entropy in E protein sequence was low, with few peaks in some amino acids positions, with higher entropy values in nonstructural regions (Fig. 2).

Fig. 2.

Informational entropy analysis of Shannon based on ZIKV polyprotein implemented on BioEdit 7.2.0. The entropy analysis was conducted based on the position of each amino acid in complete coding region of ZIKV polyprotein. The NS5 region shows the higher entropy values and E gene shows a low entropy

Different from our results, Faye et al. (2014), in a study with 43 ZIKV sequences sampled from 1947 to 2007 in Africa and Asia, observed a high phylogenetic signal in the ZIKV E gene [17]. However, it is important to emphasize that in our report, the sequences analyzed were from 2013 to 2016, which is a short period in comparison with the time of ZIKV occurrence in Africa and Asia. Therefore, the possibility of cumulating nucleotide substitutions in our strains is smaller than in the African strains analyzed by Faye et al. (2014) [17]. Besides, it is important to mention that ZIKV has reached the Americas recently and has affected an immunologically naïve population [18, 19].

In addition, it is known that several sites in the ZIKV E gene are under strong negative selection and possible nonsynonymous substitutions could not be fixed in the population [17, 20, 21]. In fact, the dual-host cycling of arboviruses may impact in the occurrence of nonsynonymous mutations in important genes, which means that the majority of those mutations are deleterious and removed by purifying selection [22]. Also, the short time of ZIKV occurrence in the Americas may reflect in a lack of positive selection in the E protein [17]. Finally, taking into account all discussion above and the establishment of a single lineage in Brazil, the potential role for founder effect should also be considered a reason for the low genetic diversity of ZIKV E gene observed in our study [23, 24]. Moreover, despite the short period of time analyzed in this study, it is possible that American ZIKV strains were not under a strong natural pressure, which led to a conservative E gene.

In conclusion, the sequences obtained in this study were grouped in the Brazilian branches, within the American clade and the Asian genotype. Altogether, our results showed a high degree of conservation on ZIKV E gene in the recent American outbreak strains, giving a good perspective to the development of new vaccines, treatments, and diagnostic methods.

Funding information

This work was supported by Fundação Araucária de Apoio ao Desenvolvimento Científico e Tecnológico do Estado do Paraná, Brazil (Grant. no. 040/2016).

Compliance with ethical standards

Conflict of interest

On behalf of all authors, the corresponding author states that there is no conflict of interest.

Ethical approval

The use of human samples was approved by the Permanent Research Ethics Committee of the State University of Maringá (Protocol no. 1.751.311).