Molecular characterization of Brazilian FeLV strains in São Luis, Maranhão Brazil

Abstract

The feline leukemia virus (FeLV) belongs to the Retroviridae family and Gammaretrovirus genus, and causes a variety of neoplastic and non-neoplastic diseases in domestic cats (Felis catus), such as thymic and multicentric lymphomas, myelodysplastic syndromes, acute myeloid leukemia, aplastic anemia, and immunodeficiency. The aim of the present study was to carry out the molecular characterization of FeLV-positive samples and determine the circulating viral subtype in the city of São Luís, Maranhão, Brazil, as well as identify its phylogenetic relationship and genetic diversity. The FIV Ac/FeLV Ag Test Kit (Alere™) and the commercial immunoenzymatic assay kit (Alere™) were used to detect the positive samples, which were subsequently confirmed by ELISA (ELISA - SNAP® Combo FeLV/FIV). To confirm the presence of proviral DNA, a polymerase chain reaction (PCR) was performed to amplify the target fragments of 450, 235, and 166 bp of the FeLV gag gene. For the detection of FeLV subtypes, nested PCR was performed for FeLV-A, B, and C, with amplification of 2350-, 1072-, 866-, and 1755-bp fragments for the FeLV env gene. The results obtained by nested PCR showed that the four positive samples amplified the A and B subtypes. The C subtype was not amplified. There was an AB combination but no ABC combination. Phylogenetic analysis revealed similarities (78% bootstrap) between the subtype circulating in Brazil and FeLV-AB and with the subtypes of Eastern Asia (Japan) and Southeast Asia (Malaysia), demonstrating that this subtype possesses high genetic variability and a differentiated genotype.

Introduction

The feline leukemia virus (FeLV) belongs to the Retroviridae family and Gammaretrovirus genus. It has a lipoprotein envelope and two identical, non-complementary RNA strands as its genetic material. It is known to cause a variety of neoplastic and non-neoplastic diseases in domestic cats (Felis catus), such as thymic and multicentric lymphomas, myelodysplastic syndromes, acute myeloid leukemia, aplastic anemia, and immunodeficiency [1, 2], in addition to co-infections with the endemic feline coronavirus (FCoV), feline immunodeficiency virus (FIV), feline infectious peritonitis, and respiratory diseases [3].

The RNA is transcribed into double-stranded DNA (PIC) that migrates to the nucleus and integrates into the host genome. The provirus contains Long Terminal Repeat (LTR) sequences at the 5′ and 3′ ends, which exercise a regulatory and control function on the expression of viral genes. These are of great importance to the replication of retroviruses and as well as presenting promising and potentiating sequences in U3 are involved in virus-related oncogenesis [4]. LTRs flank retroviral genes. The gag gene encodes the internal structural proteins (nucleocapsid-p10, matrix-p15, capsid-p27), the pol gene encodes the proteins involved in viral replication (integrase-IN, protease-PR and RT-reverse transcriptase), and the env gene encodes the viral envelope proteins (gp70 and p15e) [3].

The gag gene is also abundantly expressed by an alternative route using the glycosylation pathway of the host cell. The post-translational modification of proteins by the addition of carbohydrates is an important event for their structural and functional diversification, as when modified they are less recognized by antibodies [5]. As approximately half of the p27 proteins synthesized by infected cells are released in the glycosylated form, this is likely to be a mechanism of immune response evasion [6].

One of the most important sites of variability in the FeLV genome is in the gene sequence that encodes the surface glycoprotein (SU). The accumulation of mutations during infection has been found to result in the alteration of viral biological properties, such as recognition and receptor affinity, replication kinetics, and/or the pathogenic potential of the variant. FeLV is therefore classified into four subgroups, FeLV-A, B, C, and T, which differ genetically through variations in the env gene sequence and functionally by interference tests (a virus of one subgroup interferes with the infection by another virus of the same subgroup, in the same cell) and viral neutralization, as well as the ability to replicate in non-feline tissues [7].

FeLV-A is easily transmitted between cats and, once inside the cell, it can recombine with endogenous FeLV sequences (enFeLV) or mutate, resulting in other subtypes [8]. The replication of the B and C subtypes is only possible in the presence of FeLV-A, as many sequences which are important for replication are substituted in the recombinants. Only the A subgroup is transmitted between cats [6]. The host membrane protein used as a receptor is Feline high-affinity thiamine transporter 1 (feTHTR1) [9].

All isolates from naturally infected animals possess FeLV-A only or a combination with FeLV-B, FeLV-C, or both. Recombinant viruses are more pathogenic than FeLV-A due to different properties in the envelope proteins. FeLV-B is strongly associated with the onset of neoplasms and is more frequently isolated than FeLV-C, a rare subtype commonly seen in animals with non-regenerative anemia [10]. A fourth FeLV-T subtype, cytolytic to T lymphocytes, is associated with severe immunosuppression. This lymphotropic variant requires more complex means of penetrating the cell and does not replicate in feline fibroblasts. It appears that this subtype uses the FeLV-A receptor and an auxiliary mechanism with the enFeLV-encoded coreceptor [3, 11].

Polymerase chain reaction (PCR) is a sensitive test and when performed properly can be a good ally for confirming the diagnosis of feline viral leukemia, especially when there are discordant results between Enzyme-Linked Immunosorbent Assays and Indirect Immunofluorescence, or when the disease is suspected but no antigens have been detected. Depending on how it is performed, it can detect viral RNA or a provirus. PCR materials can be aspirated from bone marrow, tissues, and blood [12].

The present study aimed to perform the molecular characterization of FeLV-positive samples and determine the circulating viral subtype in cats from the city of São Luís, Maranhão, Brazil, as well as to evaluate phylogenetic relations and genetic diversity.

Materials and methods

Detection of FIV antibodies and FeLV, FeCoV, and FPLV antigens

Blood samples were collected from eighty cats from shelters in the city of São Luís, Maranhão, Brazil. Serum samples were used for the immunochromatographic tests using the SNAP Combo Plus (IDEXX® Laboratories, USA) and Ag Test Kit (Alere™) commercial kits to detect the FeLV p27 protein and antibodies to the FIV p24 protein. Stool specimens were analyzed for the Coronavirus (FeCoV) and Feline Panleukopenia (FPLV) using the Ag Test Kit (Alere™) and the results were interpreted according to the manufacturer’s recommendations.

Extraction of proviral DNA

Proviral DNA extraction was performed using the Wizard® Genomic DNA Purification kit (PROMEGA), following the manufacturer’s recommendations. In order to establish the same DNA concentration for all samples, they were quantified by spectrophotometry (NanoVue Plus Spectrophotometer) (GE Healthcare, Piscataway, NJ, USA) and subsequently stored at − 20 °C.

PCR for evaluation of the efficacy of extractions

The samples obtained were submitted to PCR for detection of the endogen that codifies the GAPDH enzyme (glyceraldehyde 3-phosphate dehydrogenase), according to Pinheiro et al. [13], to verify the quality of extraction and integrity of the DNA. The primers used were GAPDH F-5′ GGTGATGCTGGTGCTGAGTA-3′ and GAPDH R-3′ CCCTGTTGCTGTAGCCAAAT-5′. The PCR conditions were 5.0 μL of Green GoTaq® Flexi Buffer 5× (Promega, USA), 2.0 μL of each primer (5 pmol/μL—Invitrogen, USA), 0.5 μL of dNTP mix (10 mM—Promega, USA) 1.5 μL of MgCl₂ (10 mM—Promega, USA), 0.1 μL of Go Taq Flexi DNA Polymerase (500 U—Promega, USA), and 5.0 µL of DNA and ultrapure water DNase and RNase free (Invitrogen-Life Technologies®, USA), for a final volume of 25 μL. The PCR master mix plus 2 μL of water was used as negative control. The amplification conditions were 4 min at 95 °C, followed by 35 cycles at 95 °C for 30 s, 54 °C for 30 s, 72 °C for 50 s, and a final extension at 72 °C for 7 min.

PCR for FeLV detection

For detection of FeLV, the amplification of a region of the gag gene was performed, using the primers 5′-ACTAACCAATCCCCACGC-3′ and 3′-ATGGCTGTCCCACTAGAG-5′ [14], resulting in a product of 450 bp and of the coding region LTR/U3 (5′-AAAATTTAGCCAGCTACTGCAG and 3′-GAAGGTCGAAACTCTTGGTCAACT)/5′-TTACTCAAGTATGTTCCCATG and 3′-CTGGGGAGCCTGGAGACTGCT, which amplified a product of 235 and 166 bp [15, 16]. A master mix solution with the previously described concentrations was used. The amplification conditions were initial denaturation of 5 min at 94 °C, followed by 35 cycles of 1 min at 94 °C (denaturation), 1 min at 54 °C (annealing) and 1 min at 72 °C (extension), and a final extension of 5 min at 72 °C, amplifying a fragment of 450 pb of the gag gene. For the LTR region, the amplification conditions were initial denaturation at 94 °C for 4 min, followed by 35 cycles of 1 min at 94 °C, 1 min at 50 °C (annealing), 2 min at 72 °C (extension), and a final step of 5 min at 72 °C, amplifying a fragment of 235 bp.

Nested PCR for FeLV subtypes (A, B and C)

A nested PCR was performed to detect the FeLV subtypes, following conventional PCR. The first step consisted of the use of the external oligonucleotides H20 and H18, according to Chen et al. [17], which are annealed in the pol gene sequences (toward the env gene) and the U3 region, according to Coelho et al. [18], amplifying the entire env gene and the U3 region of the 3′ LTR. The reaction occurred in the conventional GeneMate Series PCR thermocycler and consisted of an initial denaturation stage of 5 min at 94 °C, followed by 35 cycles of 1 min at 94 °C (denaturation), 1 min at 52 °C (annealing), 1 min and 30 s at 72 °C, and a final step of 5 min at 72 °C.

The internal oligonucleotides used amplify segments of the env gene, differentiating them from the FeLV subtypes. Two microliters of the first reaction product were used in the second step, with specific oligonucleotides for each subtype: RB59 and RB17 for FeLV-A, which generates a PCR product of 1072 bp [17]; RB53 and RB17, specific for FeLV-B, with a product of 866 bp [19]; and RB58 and RB47, specific for FeLV-C, which generates a 1755-bp product [18, 20]. Each subtype was tested in independent reactions; no tests were performed for subtype T.

The reaction products were evaluated on 1.5% agarose gel (UltraPure™—Invitrogen), in TAE1X buffer (20.3-mM KH₂PO4, 10.4-mM Tris acetate, 10-mM EDTA pH 8.0), stained with ethidium bromide (1.0 mg/μL). An aliquot of 10 μL of the amplicon was applied to the gel and, for visual comparison, a standard molecular marker of 100 bp was used (Invitrogen, batch number 1735579). The gel was subjected to electrophoresis for 45 min, at a constant voltage of 110 V and current of 2.0 A and then visualized through a translucent with exposure to ultraviolet light and photographed in the gel photodocumentation system (L-Pix Loccus Biotechnology).

Purification and DNA sequencing

Purification of the amplified product was performed using the Wizard® SV Gel and PCR Clean-Up System (Promega), following the protocol and the manufacturer’s recommendations. Amplicon sequencing was performed with the BigDyeTerminator v3.1 automated kit, using the ABI-Prism 3500 Genetic Analyzer (Applied Biosystems) sequencer.

Phylogenetic analysis and subtypes genetic identification

The following reference sequences of FeLV subtypes A and B, available from the GenBank database (National Center for Biotechnology Information—NCBI), were used: Japan: (AB847295); (AB847267); (AB847261); (AB847250); (AB847227); (AB847175); (AB847233); (AB847302); (AB847252); (AB847231); (AB847206); (AB847205); (AB847244); (AB847165); (AB847300); (AB847299); (AB847168); (AB847210); (AB847236); (AB847218) [21]. Malaysia (HQ727892); (HQ197369); (HQ197374); (JF815543); and (JF815538) [22]. Taiwan: (GQ979612); (GQ465833); (GQ327961). BRAZIL: (DQ821499MGSubA); (DQ821502MGSubA); (EU090948MGSubA); (AY745878MGSubA); (DQ821501MGSubB); (DQ821500MGSubB); (DQ871351MGSubB); (EU048345SubB); (EU090943SubB); (EU048360MGSubB); (EU436639MGSubB); (EU048353SubB); (EU048356SubB); (EU436640SubA); (EU048352MGA); (EU43664MGA); (EU048357MGA); and (DQ871353MGSubA) [18]. Reference sequences were included for the phylogenetic analyses, with the aim of identifying the possible subtypes of the FeLV cloned samples from São Luís.

The sequences were edited and aligned using the Clustal W 1.4 [23] program of the Bioedit 7.0 package [24]. Phylogenetic analysis was generated in the Mega 6.0 program [25], unrooted in order to better understand the identification of the subtypes. The selected model for tree construction was the Kimura 2-parameter [26] and the parameters were based on the Analysis of Neighbor Clusters/NJ. Cluster significance was estimated by the analysis of 1000 bootstrap replicates [27].

Cloning and insert amplification in the plasmid

The 450- pb fragments amplified in PCR were extracted from the gel and purified using the Wizard® SV Gel and PCR Clean-Up System (Promega) kit. An appropriate amount of insert was calculated for each sample and introduced into the plasmid vector pGEM®-T and pGEM®-T Easy Vector Systems (Promega, USA) (1:3 plasmid-insert ratio), according to the manufacturer’s instructions. The extraction of the plasmid was performed using the Wizard® Plus SV Minipreps DNA Purification Systems kit (Promega, USA), according to the instructions. To confirm the presence of the insert in the plasmid, the PCR described above was performed and the fragments were excised from the gel and purified. The purified DNA was sequenced, confirming the cloning of the samples.

Ethics approval

All experiments with animals were conducted in accordance with the guidelines for experimental procedures of the Conselho Nacional de Controle de Experimentação Animal (CONCEA) (the National Council for the Control of Animal Experimentation) and approved by the Ethical Committee on Animal Experimentation of the Universidade Estadual do Maranhão (Protocol number 041/2014).

Results

Frequency of feline samples positive for FeLV and co-infections

Of a total of 80 samples tested, 5.0% (4/80) were positive for FeLV, 15% (12/80) of the animals were infected with FIV, and 2.5% (2/80) were positive for coronavirus and 0% (0/80) for panleukopenia. Co-infection (FIV/FeLV) was observed in 1.25% (1/80) of the animals. FeLV infection was observed in 3.75% (3/80) of females and 1.25% (1/80) of males, while 6.25% (5/80) of females and 8.75% (7/80) of males exhibited anti-FIV antibodies.

Clinical signs

The clinical examination found halitosis, severe salivation, gingivostomatitis, pharyngitis, oral bleeding, ulcerations on the dorsal and lateral portions of the tongue, palatoglossal arch lesions, cachexia, lethargy, diarrhea, and respiratory signs in animals exhibiting co-infections, caused by immunosuppression.

PCR GAPDH

There was successful amplification of the GAPDH enzyme gene in the tested samples, resulting in a 709-bp product. The use of primers or probes directed at a known sequence of DNA in samples from mammals has been used as a standard in several studies [28].

Nested PCR for FeLV-A/B/C env gene

2350- and 1072-bp products were obtained for the FeLV env gene, for the determination of the FeLV-A subtype, as described by Chen et al. [17] and 866 bp for the determination of the FeLV-B subtype, according to Pan et al. [19]. There was no amplification of the 1755-bp product, characteristic of the FeLV-C subtype. Subtype results by nested PCR demonstrated that the four FeLV-positive samples were amplified for subtypes A and B, with an AB combination and no amplification for subtype C and/or ABC. Nested PCR also allowed the identification of FeLV subtypes, revealing 39.6% (21/53) for subtype A, 1.9% (1/53) for subtype B, and 54.7% (29/53) for FeLV-AB. Subtype C or the ABC combination were not found.

FeLV phylogenetic analysis

The amplifications of the cloned samples of cats from the city of São Luís, based on the gag gene, were compared with 46 gag/FeLV sequences obtained from GenBank, giving a total of 50 analyzed samples. Phylogenetic analysis demonstrated the existence of two clades (Clades I, II, and III) and that the São Luís samples formed a subclade (Clade I) that was isolated from all the sequences used as reference, including those obtained from Minas Gerais, Brazil. The strong support of this clade (99% bootstrap) shows that São Luís samples form a variant or genotype which is isolated and varies from the others, but the similarity, despite its low statistical support (78% bootstrap), shows that this variable genotype is closer to the subtypes that occur in East Asia (Japan) and Southeast Asia (Malaysia) than those that occur in Minas Gerais. The FeLV subtypes A and B from Minas Gerais curiously relied on a differentiated clade from samples from Europe (the USA and the UK), with high statistical support (98% bootstrap). The genetic identity of the gag gene, despite similarity with subtypes A and B, reveals that the circulating viral subtype in São Luís is highly genetically variable, characterizing a circulating subtype with a different genotype from those already obtained and available. Clade III grouped only FeLV sequences from the gag gene occurring in Taiwan (Fig. 1).

Fig. 1.

Unrooted phylogenetic tree based on Neighbor-Joining and Kimura 2 model parameters, obtained from 1000 bootstrap replies. Colors represent the respective localities where the FeLV gag gene was obtained. [] São Luís sequences. [] Japanese genotype I, II and III sequences. [] Malaysia sequences. [] Minas Gerais sequences. [] USA sequences. [] Taiwan sequences. Scale bar represents 10% divergence between sequences

The samples from São Luís aligned with those from Japan and Malaysia but not from those from Minas Gerais, most likely because they share the same transmission route, a similar phylogenetic relationship to the FeLV genotypes of Japan and/or even a subgroup, but with a different genotype, considering the possibility of genetic recombination of the samples with enFeLV or other endogenous receptors.

According to the proposal of Kawamura et al. [21], the FeLV circulating virus from São Luís, based on the analysis of the gag gene, are variants of genotype I (GI/1, GI/2, GI/3, GI/4, GI/5, GI/6, and GI/7) (Clade I), which represent virus strains circulating in Japan. Clade I clearly shows that the genetic variants of this gene are not structured by geographic distribution, as can be seen in FeLV env analysis. The sequences obtained from São Luís which could have been grouped with samples of genotype III (GIII/1 and GIII/ 2) for FeLV samples from Europe and America showed no similarity to the same (Clade II). The present study demonstrates the need for genetic recombination studies to provide better understanding of the pathogenesis, diversity, and genetic variability of the virus in Brazil and also its defense mechanisms against the immune response of the infected hosts.

The molecular identification of the circulating viral subtype in the city of São Luís, Maranhão, Brazil, exhibited similarity to FeLV-AB, but proved to be a highly differentiated genotype and so may be another variable of the gene. Further work with genetic recombination analyses is needed, however, to provide a better understanding of the molecular mechanisms of FeLV transmission, in order to increase knowledge of Gammaretrovirus evolutionary patterns.

A genetic divergence matrix based on haplotype analysis was obtained. A total of 33 haplotypes were obtained and were analyzed in MEGA 6.0. The genetic distance matrix generated with the Kimura-2-Parameter distance model and neighbor-joining algorithm for the gag gene is shown in Fig. 2. The genetic divergence showed low rates of divergence between the São Luís and Japan haplotypes, with values varying from 0 to 0.4%. Among the São Luís and Malaysia haplotypes, rates of divergence greater than 0–0.6% were observed. The highest rates of divergence were observed between São Luís and Taiwan (0.3–1%). These results again show that the gag gene samples obtained in Minas Gerais differ from those from São Luís, with values ranging from 0.4 to 0.6% (moderate to high variation), showing low similarity between the sequences and confirming the presence of the genetic variations of the FeLV gag gene in Brazil.

Fig. 2.

Matrix of genetic divergence estimated from the Kimura algorithm 2-parameters (1993) based on gag gene sequence

Discussion

In a study in Minas Gerais, Brazil, Teixeira et al. [29] observed 32.5% (13/40) positivity for FeLV in animals housed in shelters. When analyzed by sex, a frequency of 22.5% was found in females and 10% in males, values higher than those observed in São Luís (Maranhão). The percentage of positivity observed in São Luís (Maranhão) was higher than the results observed by Marçola et al. [30], who found 1.5% (3/200) of animals positive for FeLV and 2% (4/200) for FIV in the Federal District. However, the values observed for FIV in São Luís (Maranhão) were lower than those reported by Lara et al. [31], who identified an occurrence of 14.8% (67/454) of positive cats in 13 cities in the state of São Paulo and by Souza et al. [32], in the city of Rio de Janeiro, who verified positivity of 20.2%. They were also lower than the data of Caldas et al. [33], who found that 37.5% (15/40) of animals in the state of Rio Grande do Sul were reactive. While this confirms a high occurrence of positive animals, those with the clinical suspicion of FIV were selected only in the last case, which may have contributed to a higher frequency.

Martins et al. [34], in a study in São Luís, Maranhão found that cats were positive for FIV in three independent tests: 10.83% (13/120) by PCR, 9.17% (11/120) in SNAP combo plus, and 10.83% (13/120) using indirect ELISA and that only one animal was positive for FeLV in SNAP combo plus (0.83%—1/120). This FeLV-positive animal was not positive for FIV in any of the three tests. Compared to animals from shelters, there was an increase in the number of animals positive for FeLV, as well as for FIV when using the serological test, with one positive animal for both infections.

Most of the clinical signs in cats with FIV and/or FeLV are caused by secondary infections that arise as a result of the immunosuppression caused by these retroviruses. The clinical signs associated with these diseases vary greatly, depending on the age of the animal at the time of exposure to the virus, the efficacy of the immune system response, exposure and possible co-infection, virus pathogenicity, and virus subtype, mainly in FeLV [10, 11, 35, 36]. In general, the cats presented pale mucous membranes, dyspnea, lethargy, anorexia, progressive weight loss, fever, gingivitis, stomatitis, uveitis, diarrhea, and abscesses that did not heal [37].

Since blood cells may have inhibitory components that lead to false-negative results in PCR [38], the enzyme-encoding gene glyceraldehyde 3-phosphate dehydrogenase (GAPDH) was used to verify the presence of amplifiable DNA. All the samples amplified the gene, confirming that there was no DNA degeneration or presence of PCR inhibitors, thus reducing the possibility of false-negative samples.

The PCR technique used in the study identified proviral DNA (virus sequences that were integrated into the host genome). Retrovirus replications require the reverse transcription of viral RNA into double-stranded DNA, which integrates into the chromosomal DNA of the host. Therefore, when the host cell undergoes transcription, it also transcribes and translates the viral mRNA [39].

Hypothetically, some animals may have had latent infections in the bone marrow and were not diagnosed by serology and total blood PCR (leukocyte layer). The challenge for the clinician is to determine which test result is most likely to reflect the true infection phase of the disease. It is possible that infected cats are antigenic, but have negative blood cultures for the virus if the infection is sequestered in another tissue compartment, such as the bone marrow or lymphoid system [40].

According to Torres et al. [41], animals exposed to FeLV in which no antigenemia was proven were able to eliminate the virus before its integration into the host genome (abortive infection) or succeeded in retaining viral replication, in which non-productive infected cells were maintained in the circulation and the tissues (regressive infection). Animals with reduced ability to contain the progression of infection presented antigenemia only in the first few weeks and at moderate levels (distinct from other groups) of proviral DNA (latent infection). Infected cells are latent, but potentially productive. Reactivation is more expected in this case.

Animals that cannot contain the progression of infection have elevated levels of p27 and proviral DNA for up to 3 years after exposure (progressive infection). The latent infection is the absence of viremia and persistence of culturable virus in the bone marrow, with a high reactivation potential [42]. To evaluate the stage of the infection it is therefore necessary to perform a quantitative PCR, in addition to concomitant clinical evaluation and bone marrow PCR, for the detection of latent infection.

All the samples submitted to conventional PCR for the detection of part of the gag gene were tested for the LTR/U3 region, since the provirus contains repeated sequences at the 5′ and 3′ ends and has a regulatory and control function for the expression of the viral genes [3].

After penetrating the cell, the RNA genome, contained in a nucleus of non-glycosylated proteins associated with the viral reverse transcriptase, is transcribed into the cytoplasm in the form of a double strand of DNA. Such transcription involves two jumps of reverse transcriptase, from the 5′ to the 3′ region, which result in the duplication of the sequences located at the ends of the virion RNA, forming redundant sequences termed long terminal repeats (LTR) [43].

Gene expression in retroviruses is driven by the LTRs, which are generated at the end of the proviral genome by reverse transcription in the U3, R, and U5 RNA regions. The U3 region is particularly important for the regulation of gene expression as it contains the transcriptional promoter and accentuated sequences [44]. The synthesized double-stranded DNA is transported to the nucleus prior to its integration into the chromosomal DNA of the host by the integrase enzyme; the transcription of cellular DNA by RNA polymerase which initiates at 5′LTR and terminates at 3′LTR generates new virions [1].

The env gene encodes the proteins present in the viral envelope, such as glycoprotein 70 (gp70) and transmembrane protein 15E (p15E). The gp70 is extremely relevant for determining the degree of pathogenicity of the virus and its tropism. P15E is responsible for the binding of gp70 to the host cell membrane, causing immunosuppression and non-regenerative anemia, observed in cats with persistent infection. The pol gene is responsible for reverse transcriptase and integrase [10, 45].

Hardy et al. [45] found a higher percentage of FeLV-A and suggested that the presence of the subtype undergoes an environmental influence. In closed groups, where the animals do not have external contact, the subtype tends to be the same in the entire group of animals. In populations where the virus was introduced by external vertical transmission, a balanced distribution of FeLV-A and FeLV-AB, as occurs in the natural environment, is expected. In populations in which animals have external access, cats with different subtypes are expected. Samples analyzed in the city of São Luís were obtained from a closed environment (shelter), with the frequent introduction of new individuals, which lived in the same place and shared drinking and feeding troughs. For this reason, it is believed that there is a significant possibility of finding more than one subtype.

FeLV-B is generated from recombination between FeLV-A and the endogenous retroviral sequences [3, 46]. It has less similar identities between sequences and is usually associated with lymphomas [8]. Subgroup C is the most pathogenic as it causes aplastic anemia. It arises from mutations in the env gene of FeLV-A. FeLV-T was recognized and isolated in felines that developed FeLV-associated immunodeficiency, with 96% similarity to FeLV-A. It arises from mutations in the env gene sequence of this same subtype [45].

A study by Coelho et al. [18] used nucleotide sequences corresponding to the LTR/U3 region plus the FeLV provirus gag gene and env gene for the phylogenetic analysis of samples from Minas Gerais. Of the samples analyzed, 19 were classified as FeLV-B by nested PCR and were obtained from animals with clinical signs of infection. However, 15 samples classified as FeLV-A were obtained from cats with no clinical signs. This analysis compared the gag gene with others obtained from GenBank. The env gene of FeLV samples from Minas Gerais State were compared to several exogenous isolates of FeLV and the endogenous provirus (enFeLV) from the same region. The results revealed the circulation of FeLV-B in a large part of the domestic cat population and found that FeLV-A sample sequences were closely identified with natural isolates.

The sequences of the samples obtained from São Luís had a closer genetic identity to the subgroups A and B of Japan and Malaysia, but also exhibited high genetic variation. The animals presented several clinical signs which were both associated and not associated with the disease, in contrast to the findings of Coelho et al. [18]. According to Stewart et al. [8], FeLV-A is easily transmitted between cats and, once inside cells, can recombine with the enFeLV (endogenous FeLV) sequences or undergo mutations, giving rise to the recombinant subtypes B and C, respectively. The most isolated variant cited was FeLV-B, formed by the recombination between FeLV-A and enFeLV.

Although these endogenous sequences are not capable of generating active infections, the envelope portions recombine with FeLV-A, resulting in FeLV-B, which is capable of colonizing cells previously infected with FeLV-A, as both use different receptors [47]. According to Hardy et al. [45], FeLV-A was detected in 100% of the infected animals; co-infection with FeLV-B was found in 49%; and co-infection with FeLV-C was found in 1%.

Another analysis described in literature was performed by Finoketti et al. [14] and showed that the sequenced FeLV sample had a closer identity to samples of subtype B, which also aligned with those described by Coelho et al. [18]. The authors indicate the need for a larger number of samples to undergo nucleotide sequencing and alignment in the state of Rio Grande do Sul.

FeLV is a good model for investigating natural genetic diversity, since recombination with the exogenous and endogenous DNA sequences can be observed among the virus subgroups. However, it is yet to be determined if the recombinant sequences affect the molecular mechanisms of FeLV transmission [21]. The mechanism by which recombination occurs among retroviruses is known as template switching during reverse transcription [48]. The generation of recombinant gag genes may be due to the expression of enFeLV in target cells. It is not known in which loci enFeLV are active and associated with the generation of recombinant viruses [21].

Further analysis of enFeLV expression would provide a greater understanding of the generation of recombinant viruses. Recombinant retroviruses can play an important role in virulence. Phylogenetic analysis demonstrated that the FeLV gag gene can be classified in genotypes I, II, and III. Genotype I is a large genetic group and may still be classified in clades 1–7 in Japan. FeLV genotypes based on env genes in Japan are strongly associated with geographic distribution. As the genotype data for FeLV gag and env genes were consistent, the association between the FeLV genotype and the geographic distribution in Japan of these two genes may also be consistent. The present study found that the gag gene of FeLV differs genetically and structurally. FeLV subtypes generally contain endogenous retroviral sequences in their genomes and are therefore viral particles [21].

Author contributions

The author’s NSM, APSR JMB, JJA, LLR, LLA, RMO, LSS, AVCN, RAO, RCC, FAM, JKPR, and ALAS performed experiments, conceived the study, draft, and revised the manuscript. ALAS provided financial support. All authors read and approved the final manuscript.

Funding

This work was funded by support from the Fundação de Coordenação de Aperfeiçoamento de Pessoal de Nível Superior—CAPES and FAPEMA. Dr. Ana Lucia.

Data availability

The raw data used in this study are available from the corresponding author upon reasonable request.

Declarations

Conflict of interest

The authors declare that they have no conflict of interest.

Ethical approval

This study was approved by the Ethical Committee on Animal Experimentation of the Universidade Estadual do Maranhão (Protocol Number 041/2014).

Consent to participate

Not applicable.

Consent for publication

All authors consent for publication.