In this study, 26 peripheral blood samples were collected (13 blood donors, 6 HTLV-1 asymptomatic carriers (HAC), and 7 HAM/TSP individuals). The distribution of the participants according to gender and age is shown in Supplementary Table 1. HAC and HAM/TSP individuals presented a mean age of 46 years (±17 years) and 56 years (±12 years) (Supplementary Table 1), respectively, with a mean proviral load (PVL) of 652.7 copies/10⁵ PBMCs and 6,727.5 copies/10⁵ PBMCs (Table 1), respectively. Figure 1 shows the distribution of PVL among infected groups. Patients with HAM/TSP presented higher PVL when compared with HAC (p = 0.0025, Mann–Whitney test). Those findings are also observed in other studies and reflect the state of infection in each patient (Olindo et al., 2005; Iwanaga et al., 2010; dos Furtado et al., 2012).

Previous data from our group showed that HTLV-1-infected cell lines release EVs with viral cargo and can induce the production of proinflammatory cytokines (Otaguiri et al., 2018). To verify if this is also valid for circulating EVs from HTLV-1 carriers, sEVs were isolated from the serum of HAC, HAM/STP individuals, and blood donors using the Exoquick exosome precipitation solution (System Biosciences). Serum-derived sEVs from all three groups presented a modal diameter up to 166.2 nm when evaluated by nanoparticle tracking analysis (Figure 2A). In addition, western blot analysis detected classic EV markers (Alix and CD9) and the absence of the viral protein p19 (Figure 2B). Since sEVs and HTLV-1 viral particles share the endosomal pathway during biogenesis, the exclusion of p19 in sEV extracts indicates none or low levels of contamination with HTLV-1 virions. The presence of EV markers in the protein extract of HTLV-1 particles portrays the shared pathway (Figure 2B).

Following sEVs characterization, we sought to determine the expression levels of tax and HBZ viral messenger RNA (mRNA) in serum-derived sEVs from HTLV-1 carriers. A quantitative real-time PCR (RT-qPCR) analysis was performed, and the amplification data were normalized to the expression of the β-actin gene within sEVs. From all 13 serum samples of HTLV-1 carriers, only three presented sEVs with detectable levels of tax and HBZ mRNA (Table 1). These samples corresponded to HAM/TSP individuals with PVL higher than 6,000 HTLV-1 copies/10⁵ PBMCs. Interestingly, HAM/TSP 1 showed PVL three times higher than the other patients with HAM/TSP at blood sampling. The notable PVL was accompanied by higher tax and HBZ expressions in sEVs when compared with other individuals—especially HBZ. In all three samples, the HBZ mRNA/tax mRNA ratio demonstrated 2.2 to 11.7 fold increase in HBZ expression over tax expression (Table 1). Therefore, the correlation between the sEVs viral transcripts and the HTLV-1 PVL was measured. As expected, our results indicated a positive correlation between tax and HBZ mRNA in sEVs and the patient's PVL (Spearman's r = 0.0381) (Figure 3).

Finally, we investigated whether or not serum-derived tax+ HBZ+ sEV triggers the transcription and secretion of inflammatory cytokines in PBMCs. Increasing doses of control sEV and tax+ HBZ+ sEV (500, 1,000, and 2,000 μg/ml) from HAM/TSP1 were co-cultivated with heterologous naive PBMCs and heterologous HTLV-1-infected PBMCs. Simultaneously, sEVs from CTRL1 were co-cultivated under the same conditions as a negative control for immunomodulation. The production of inflammatory cytokines (interferon (IFN)-γ, tumor necrosis factor (TNF)-α, IL-4, IL-6, IL-8, IL-10, and C–X–C motif chemokine 10 [CXCL10/IP-10]) was evaluated by quantitative PCR (qPCR) and the Luminex xMAP assay. The selected cytokines correspond to those commonly altered in symptomatic HTLV-1 carriers (Futsch et al., 2018).

Regarding naive PBMCs, it was observed that those treated with 2,000 μg/ml of control sEV presented higher IL-4 protein levels than untreated PBMCs (Kruskal–Wallis post-hoc Dunn's test, p = 0.0349) (Figure 4A). However, IL-4 transcripts levels did not alter significantly under the same treatment (Figure 4C). Tax+ HBZ+ sEV did not alter significantly any cytokine levels. Moreover, IFN-γ transcripts and protein levels increased in a dose-dependent manner after PBMCs treatment with tax+ HBZ+ sEV (Figures 4B,D).

As for the assays with infected cells, two PBMC samples were used: HAC 6, which corresponded to a patient with low PVL (11.3 HTLV-1 copies/10⁵ PBMCs), and HAC 7, a patient with high PVL (5,806.3 copies/10⁵ PBMCs). Those samples were used to evaluate the production of inflammatory cytokines by the adaptive immune response to HTLV-1.

No dose of control sEV altered the transcription or secretion of the analyzed cytokines in both infected PBMCs. When compared with naive PBMCs, HAC 7 PBMCs showed much higher secretion of all inflammatory cytokines regardless of tax+ HBZ+ sEV treatment. However, the effects of tax+ HBZ+ sEV treatments on HAC 6 PBMCs followed a dose-dependent increase in IL-8 transcripts and protein levels when compared with the untreated PBMCs (Figures 5A,B). Interestingly, after treatment with 2,000 μg/ml of tax+ HBZ+ sEV, IL-8 levels were similar to those observed in PBMCs with high PVL (HAC 7) (Supplementary Figure 1). These data suggest an sEV-mediated increase in inflammatory responses.

Peripheral blood samples were collected from blood donors and HTLV-1-infected adult subjects of both sexes. Samples were fractionated into serum and PBMCs. Infected individuals were classified as HTLV-1 asymptomatic carriers (HAC) or HAM/TSP carriers according to the World Health Organization (WHO) diagnostic criteria.

In this work, cell lineages HeLa (ATCC-CCL-2), Jurkat (ATCC TIB-152), MT-2 (ECACC 08081401), and primary culture of human PBMCs were used. Cell culture media for HeLa expansion was D-MEM (Gibco) supplemented with 1% Penicillin-Streptomycin at 10,000 U/ml (Gibco) and 10% fetal bovine serum (Gibco). Jurkat, MT-2, and human PBMCs were cultured with RPMI 1640 supplemented with 1% Penicillin-Streptomycin at 10,000 U/ml (Gibco) and 10% fetal bovine serum (Gibco). All cells were cultivated at 37°C in a 95% humid atmosphere with 5% CO₂.

Human T-lymphotropic virus 1 PVL was quantified by multiplex qPCR as previously described (Rodrigues et al., 2022). Briefly, PBMCs were isolated by the density gradient method, using Ficoll Hypaque PLUS reagent (GE Healthcare) and genomic DNA extracted by the Gentra Puregene Blood kit (Qiagen). Using primers and probes to pol viral gene and the human β-globin endogenous gene, the number of integrated HTLV-1 provirus copies was determined using linear interpolation in a plasmidial standard curve. The PVL was calculated by the formula:

Serum samples were preprocessed through centrifugation at 3,000 × g for 15 min to remove larger vesicles (apoptotic bodies). Then, exoquick reagent (System Biosciences) was used to enrich EVs from 1.5 ml of human serum according to the manufacturer's recommendations. Briefly, serum samples were incubated with the one-fourth volume of the Exoquick polymer at 4°C for 1 h. The samples were then centrifuged at 1,500 × g for 30 min and the resulting pellet was resuspended in 400 μl of PBS for the following analyses.

The isolated EVs were characterized by size and protein markers. EVs concentration and size distribution were accessed by nanoparticle tracking analysis (NTA) (NS300 equipment, NanoSight). Images were acquired for 15 s in triplicate with the following parameters: scientific Complementary Metal-Oxide-Semiconductor (sCMOS) camera, green laser (532 nm), camera shutter 1,300, camera gain 512, and detection threshold 8. Protein markers were accessed by western blot. For protein extraction, 50 μl of EVs suspension were disrupted by 100 μl of radioimmunoprecipitation assay (RIPA) buffer (Sigma-Aldrich) plus 1% protease inhibitors (Sigma-Aldrich). Protein concentration was determined by the Bradford microplate assay (Bio-rad), and 30 μg of total proteins were applied in 4–20% Mini-PROTEAN® TGX™ Precast Protein Gels (Bio-rad). The migrated proteins were transferred to a polyvinylidene fluoride (PVDF) membrane (GE Healthcare) previously activated in methanol (Merck), followed by blocking with 10% milk (w/v) (Bio-rad) and 0.1% Tween 20 (v/v) (Sigma-Aldrich) in Tris Buffered Saline solution (TBS, 50 mM Tris, 150 mM NaCl). The primary antibodies used were 1:1,000 anti-Alix (Cell Signaling), 1:1,000 anti-GAPDH (Cell Signaling), 1:1,000 anti-CD9 (Abcam), and 1:1,000 anti-HTLV-1 p19 (Abcam).

Extracellular vesicle RNA extraction was performed by lysis and phase separation followed by purification using the QIAamp Viral RNA Mini Kit (QIAGEN). Precipitated EV was added with 750 μl of TRIzol LS reagent (Thermofisher Scientific) and 150 μl of chloroform. The samples were vigorously vortexed and incubated at room temperature for 10 min. Phase separation was observed after centrifugation at 12,000 × g for 15 min at 4°C. For each 200 μl of aqueous phase obtained, 1 μl of linear polyacrylamide was added to enrich the RNA yield (GenElute LPA, Sigma-Aldrich) and 200 μl of isopropanol (Merck). After overnight incubation, samples were loaded onto the column provided by the QIAamp Viral RNA Mini kit, and extraction proceeded according to the manufacturer's instructions.

Complementary DNA (cDNA) synthesis was performed from 200 ng of RNA using the High-Capacity cDNA Reverse Transcription kit (Thermofisher Scientific) and specific antisense primers for tax (5′-GAGAAACTTACCCATGGTGTTGG-3′ and nucleotides 5,169–5,191), HBZ (5′-GAGCCGATAACGCGTCCAT-3′ and nucleotides 1,086–1,104), and the endogenous β-actin genes (5′-CCAGGGCAGTGATCTCCTTCT-3′ and nucleotides 1,025–1,045).

The relative quantification of viral genes was performed by singleplex qPCR, using specific primers and probes for tax (5′-ATCCCGTGGAGACTCCTCAA-3′, sense, nucleotides 5,098–5,117, 5′-GAGAAACTTACCCATGGTGTTGG-3′, antisense, nucleotides 5,169–5,191, and 5′-FAMAGCTGCATGCCCAAGAMGB-3′, probe, nucleotides 5,115–5,130), HBZ (5′-TACATCGTCACGCCCTACTGG-3′, sense, nucleotides 1,026–1,046, 5′-GAGCCGATAACGCGTCCAT-3′, antisense, nucleotides 1,086–1,104, and 5′-FAMATCAGATCACCTGGGACCMGB-3′, probe, nucleotides 1,062–1,079), and endogenous β-actin genes. The amplification of the endogenous gene used the Human ACTB Endogenous Control (Applied Biosystems). The reactions were performed in duplicate using TaqMan Universal PCR Master Mix (Applied Biosystems) and the ABI 7500 sequence detection platform (Applied Biosystems). The results of gene expression of tax and HBZ were expressed in Relative Expression Units (REUs), according to the calculation described (Albesiano et al., 2003):

Peripheral blood mononuclear cells were isolated by the density gradient method, using Ficoll Hypaque PLUS reagent (GE Healthcare). A total of 10⁵ cells were plated in 24-well plates (Greiner) and co-cultured with increasing doses of sEVs (500, 1,000, and 2,000 μg/ml) isolated from a patient with HAM/TSP (HAM/TSP 1) and a blood donor (CTRL 1), both paired by sex and age. In addition, we co-cultured PBMCs with the Exoquick polymer used to precipitate sEV (vehicle) as a negative control. For the assay, RPMI 1640 medium (Gibco) was used supplemented with 1% Penicillin-Streptomycin at 10,000 U/ml (Gibco) and 10% exosome-depleted fetal bovine serum (Gibco). After 48 h at 37°C, cells and culture supernatant were collected and stored at −80°C.

Seven inflammatory cytokines and chemokines were evaluated in culture supernatant by the Luminex xMAP assay: INF-γ, TNF-α, IL-4, IL-6, IL-8, IL-10, and CXCL10/IP-10. Data acquisition was performed using the Luminex MAGPIX System-EMD Millipore equipment, and the acquired data were analyzed using the Milliplex Analyst v3.5 software (Millipore). The detection procedure was performed according to the manufacturer's instructions.

The altered analytes were confirmed by gene expression. PBMCs were submitted to RNA extraction by the RNeasy Mini Kit (Qiagen) according to the manufacturer's recommendations. cDNA was synthesized by reverse transcriptase reaction from 50 ng of RNA, using the High-Capacity cDNA Reverse Transcription kit (Thermofisher Scientific) according to the manufacturer's instructions. Gene expression analysis was performed using the TaqMan system, with specific probes for the INF-γ (Hs00989291_m1), IL-4 (Hs00174122_m1), and IL-8 (Hs00174103_m1) genes. The normalization of the reaction was performed by the endogenous β-actin gene (ACTB, Hs99999903_m1), and cytokine expression levels were calculated by the delta-delta Ct method.

For the confection of graphs and inferential analysis, we used the GraphPad Prism software, version 8.0.2. The difference between the PVL presented by symptomatic and asymptomatic participants was investigated by the Mann–Whitney test. The correlation between PVL and levels of viral transcripts in EV was evaluated by Spearman's correlation test. Finally, the response to EV treatment observed in PBMCs was assessed by the Kruskal–Wallis test. For all analyses, data were expressed as mean ± standard deviation (SD) and a bilateral value of p ≤ 0.05 was considered statistically significant.