Stability of the HTLV-1 glycoprotein 46 (gp46) gene in an endemic region of the Brazilian Amazon and the presence of a significant mutation (N93D) in symptomatic patients

In 1980, the first human retrovirus was found in cells of the T lineage, which was denominated the Human T-cell Lymphotropic Virus (HTLV) [1]. Four variants of this virus are now known in humans, and are referred to as the Human T-cell Lymphotropic Virus, types 1–4 (HTLV-1, HTLV-2, HTLV-3, and HTLV-4). This virus belongs to the genus Deltaretrovirus, family Retroviridae [2]. Worldwide, only 2–5% of infected individuals present symptoms, while most are asymptomatic throughout their lives. The current data indicate that there may be 5–10 million carriers of HTLV-1 worldwide, with approximately 1.5 billion individuals living in endemic areas [3]. An estimated 2.5 million individuals are thought to be infected in Brazil [4], and in the Amazon region, infections by both HTLV-1 and HTLV-2 have been recorded in both urban and rural populations, with a certain degree of endemicity [4, 5]. While it may often be asymptomatic, HTLV-1 has been implicated in the development of diseases such as leukemia/Adult T-cell Lymphoma (ATL), HTLV-1 associated myelopathy/Tropical Spastic Paraparesis (HAM/TSP), and a number of other inflammatory diseases, including dermatitis, uveitis, arthritis, and strongyloidiasis [6].

The HTLV-1 genome has three regions that codify the precursors gag, pol and env. The glycoproteins of the env region mediate the binding of the virus to the surface receptors of the target cells. The gp46 surface Envelope glycoprotein is essential to the initial steps of the viral infection, and is the most immunogenic protein of all viral antigens. A large proportion of the neutralizing antibodies are directed towards the Envelope glycoprotein [7]. The reduced genetic diversity of the gp46 sequences is related to the fact that the HTLV-1 genome varies little, in general, given that it persists in an individual throughout the clonal expansion of infected cells. However, a certain amount of variation does exist among geographic regions and in some HTLV-1 subgroups, which may be associated with different levels of vulnerability to disease [8].

Understanding the immunogenic properties of the HTLV-1 surface Envelope glycoprotein will be crucial for the development of effective vaccines and immunological treatments to combat infections [8]. In Brazil, HTLV infection is not treated as a public health problem and is largely neglected, which in general means that the risk of transmission increases substantially [9]. The geographic region investigated in the present study is considered to be an area of endemism for this viral infection, as demonstrated in previous genomic studies [10‐12], although the genetic evolution of the virus (which is of paramount importance for the prevention of this pathology) has never been analyzed in this prominent region. The present study is based on the molecular characterization of the gp46 gene of HTLV-1, the analysis of its diversity and evolution, and the identification of the possible factors that determined the changes in its amino acids.

The Tropical Medicine Nucleus is renowned as a center of excellence for the treatment and monitoring of HTLV patients in northern Brazil, with more than 90% of its patients being resident in the metropolitan region of Belém. Between 2010 and 2015, 1929 samples of blood were analyzed, of which, 51 tested positive for HTLV-1. While these figures may not be representative of the prevalence of the virus in the Brazilian Amazon region, they may be consistent with the infection rates found in the metropolitan region of Belém.

The phylogeny of the p46 gene revealed that the aA HTLV-1 genotype is found in the study region, which is consistent with the data on blood donors from the state of Pará [20], and the predominance of this subtype in other regions of Brazil, in both patients with associated diseases [21] and asymptomatic individuals [22]. Gessain and Cassar [3] highlighted the role of migration in the dispersal of the aA subtype to many countries and populations. The slave trade between the fifteenth and nineteenth centuries may have played a determining role in the prevalence of the virus in Brazilian populations of African descent [23].

In the phylogenetic tree, the positive HTLV-1 samples were grouped in the aA clade, together with samples from Canada, China, France, and other regions of Brazil. Limitations of the data prevented comparisons with many samples, due to the lack of complete p46 sequences. Most of the samples analyzed are closely-related, reflecting the conservation of the virus, despite the fact that the individuals sampled were indicator cases, and not related, which reconfirms the known lack of genetic diversity in this virus [24, 25]. No other HTLV-1 subtypes were recorded in the present study, which is consistent with limited occurrence of other genotypes, such as subtypes d, e, and f, which are isolated in Africa, the Congo, and Gabon, respectively [26].

This region is likely to suffer greater selection pressure than others, due to its role in the expression of viral bonding proteins, reinforcing the conclusion that peptides derived from the the gp46 Envelope glycoprotein are strong candidates for the development of an effective vaccine [8].

The HTLV-1 subtypes compared in the evolutionary tree also revealed relatively low levels of divergence. In the case of the nucleotide diversity, Wolfe et al. (2005) [27] recorded a rate of 1% for the complete HTLV-1 genome, while a more specific analysis of the env region [28] revealed rates of 7.8–8.0%. While the nucleotide diversity found in the present study was intermediate between these extremes, the analyses were limited by a lack of data on complete p46 sequences.

Approximately two-thirds of the patients analyzed in the present study were symptomatic, a rate similar to that recorded in the endemic region of Salvador, Brazil, where 84.3% of patients reported some type of symptom, in particular pain, during their treatment [29]. This relatively high rate may be related to the nature of the subset of patients treated in the outpatients clinic of the TMN/UFPA, which includes family cases, screened blood donors, and other individuals under investigation. The presence of persistent symptoms may also be related to the mean age of the patients (45.2 years), given the potential role of this factor in the occurrence of symptoms related to infection by HTLV-1, in particular pain [30].

These factors may also have contributed to the relatively high incidence of HAM/TSP (17.5%), in comparison with populations from the Caribbean, where the incidence was 1–5% [31], and other populations, with rates of around 5% [32, 33]. Two of the six patients with HAM/TSP presented amino acid mutations, one with S72G and the other with L70I and S103P. Mota-Miranda et al. (2013) [8] recorded the S72G, N42H and F14S mutations in patients with HAM/TSP. In all cases, there was some alteration of the amino acid sequence of the p46 gene.

In the present study, 21 of the samples presented amino acid mutations, of which, S35 L, F14S, and S72G have been described previously, and most are associated with specific functional domains [8]. The most common mutations were S72G, N93D and S192P, and, of these, N93D was the most relevant, given that all the samples with this mutation were associated with similar symptoms of pain and sensitivity. While nothing is known of this mutation, it may be linked to motor symptoms and possibly even HAM/TSP, and may be restricted to the region of the present study. This mutation involves a large segment of the sequence of the p46 gene (25-190aa/53-75aa/75-101aa/86-107aa/90-94aa), which may cause alterations to the Receptor Binding Domain (RBD), the predominant region of the linear epitopes, the functional domain, and the amino acids associated with the interactions with neuropilin 1.

The S192P mutation, while relatively frequent, could not be linked systematically to symptoms, given that two patients were asymptomatic. While it was also among the most frequent mutations, S72G was also found in an individual with more than one amino acid mutation in the sequence of the gp46 gene. This mutation has been associated with patients with HAM/TSP and has also been found in samples from Gabon, Martinique, and Guadaloupe [8]. In this case, it seems reasonable to suggest an association with the development of motor symptoms and/or HAM/TSP. In some protein domains, distinct epitopes have been identified in asymptomatic individuals and patients with HAM/TSP. The latter tend to have a larger set of env epitopes in comparison with the asymptomatic individuals, indicating that this diversity affects the cytotoxicity of the CD8-positive T cells, and may be related to the hyper-immune response in individuals with HAM/TSP [34].

Subject 2454 presented the most amino acid mutations in its sequence, with five (S192P, S194 T, L200H, L210H, and L213P), but nevertheless remained asymptomatic throughout the study period, which suggests that these changes did not influence the symptomatology of the patient. None of these mutations has been described previously, and while they affect some domains with well-defined functions, it seems likely that, in this case, other factors that determine the infection are more influential.

Considering that the gp46 glycoprotein is involved directly in the mechanism of adsorption of the virus by the receptor cell, and thus in the transmission of the virus [35], many of the domains presents in the sequence of the gene have a direct influence on the function of this mechanism, which implies that mutations in this sequence may favor or hinder the transmission of the virus. In 63% of the cases in which family transmission of HTLV-1 was observed, some amino acid mutation was also observed, although there was no significant difference in comparison with the individuals in which no family transmission was recorded. This emphasizes the need for a larger sample size for the more systematic evaluation of the relationship between these mutations and the transmission of the virus. Once the most relevant protein domains are located, it may be possible to define clinical markers for the diagnosis of the disease.

While the HTLV-1 genome varies little, the amino acid mutations in the gp46 gene may modify the structure or antigenicity of the principal neutralizing epitopes. These modifications may have a direct effect on the efficacy of the neutralizing antibodies, and may be related to the clinical manifestations, dissemination, and pro-viral charge [36]. The identification of the env epitopes responsible for the activation of the immunological system may also be useful for the development of a vaccine [8].

A single HTLV-1 subtype (aA) was recorded in the metropolitan region of Belém. The codifying region of the gp46 gene was highly stable, with a low rate of evolution. There was a predominance of certain symptoms, such as pain and sensitivity, dysautonomia and motor disorders, and HAM/TSP was common in infected patients. Amino acid mutations were related to some symptoms, but not to any greater probability of family transmission of the infection. The rare mutation N93D was found invariably in patients with oligosymptoms associated with HAM/TSP.