Ethics statement

The Indian rhesus macaques (Macaca mulatta) used in this study were obtained from the free-range colony on Morgan Island (South Carolina) or Covance Research Products (Princeton, NJ) (Supplementary Table 1). Animals were housed and maintained at the NCI Animal Facility at the NIH, Bethesda, MD. The NIH is accredited by AAALAC International and follows the Public Health Service Policy for the Care and Use of Laboratory Animals. Animal care was provided in accordance with the procedures outlined in the “Guide for Care and Use of Laboratory Animals” (National Research Council, 2011; National Academy Press, Washington, D.C.). Animals were handled in accordance with AAALAC standards in an AAALAC-accredited facility (OLAW, Animal Welfare Assurance A4149-01 for NIH). All animal care and procedures were carried out under protocols approved by the NCI and/or NIAID Animal Care and Use Committees before study initiation (ACUC; Protocol VB043). Animals were closely monitored daily for any signs of illness, and appropriate medical care was provided as needed. Animals were socially housed per the approved ACUC protocol and social compatibility. All clinical procedures, including biopsy collection, administration of anesthetics and analgesics, and euthanasia, were carried out under the direction of a laboratory animal veterinarian. Steps were taken to ensure the welfare of the animals and minimize discomfort of all animals used in this study. Animals were fed daily with a fresh diet of primate biscuits, fruit, peanuts, and other food items to maintain body weight or normal growth. Animals were monitored for mental health and provided with physical enrichment, including sanitized toys, destructible enrichment (cardboard and other paper products), and audio and visual stimulation.

Animal inoculation and treatments

Nineteen male and female rhesus macaques uninfected with SIV/SHIV, as demonstrated by several consecutive negative PCRs and seronegative for simian T-cell lymphotropic virus 1 at the initiation of the study, were randomized into three groups based on sex, age, weight, and their prior enrollment in other studies (Supplementary Table 1)³⁹. Five animals from the α-CD8/NK/Clodrosome®/HTLV-1A/C_oI-L group and ten animals from α-CD8/NK/Clodrosome®/HTLV-1A were treated for three consecutive days (Days −3, −2, −1) with an anti-CD8 monoclonal antibody the clone MT807R1 targeting the α/α chain of the CD8⁺ lymphocytes and NK cells, in addition to a single dose of Clodrosome® delivered in Liposome (Day −1) targeting the phagocytic cells. Four control animals in the IgG/Liposome/HTLV-1A/C_oI-L group were treated for three consecutive days (Days −3, −2, −1) with the isotype control antibody IgG OKT3, reactive against the human CD3 molecules, in addition to a single dose (Day −1) of Encapsome® corresponding to an empty Liposome. Both antibodies, anti-CD8 and IgG OKT3, were purchased from the NHP Reagent Resource Program (University of Massachusetts Medical School, Worcester, MA), while Clodrosome® (cat. #CLD-8909) and Encapsome® (cat. #CLD-8910) were purchased from Encapsula NanoSciences (Brentwood, TN). All treatments were administered intravenously at 5 mg/kg/dose/day prior to the intravenous inoculation of 1 × 10⁸ or 1.5 × 10⁸ lethally γ-irradiated 729.6 lymphoblastoid B-cell lines producing either HTLV-1A or HTLV-1A/C_oI-L, respectively. The inoculated cell number was normalized for p19 Gag antigen production and viral DNA level to reflect the amount used in our previous studies^33,35,39. Animals were monitored for over 21 weeks post viral inoculation and then euthanized to study viral dissemination and pathogenesis in tissues except for TMN, DG8Z, TiT, RA6, TRE, TZW, RH5, RKF, and DHF6 that were monitored for over 48 weeks. Except for randomization purposes, the sex of the animals was not considered as discriminating factor in our study. Since the aim of our study was to investigate the infectivity and the pathogenicity of the newly constructed chimeric molecular clone, both male and female mice were used in each group.

Biosafety statement

All experiments involving HTLV-1A or HTLV-1A/C_oI-L were conducted in a Biosafety Level 2 (BSL-2) laboratory with BSL-3 practices in accordance with the institutional biosafety guidelines. Standard microbiological practices were followed, including the use of appropriate Personal Protective Equipment (lab coats and gloves), performing aerosol-generating procedures in a certified Biological Safety Cabinet, and proper decontamination of waste materials.

Plasmid construction and verification

HTLV-1A/C_oI-L chimeric molecular clone

An artificially synthesized DNA fragment encompassing a portion of HTLV-1 subtype A envelope and the entire 3′end that spanned the orf-I, II, III, IV, and the 3′LTR of the HTLV-1C was introduced into a SalI/EcoRI cassette. The molecular clone HTLV-1A (pAB_D26)³⁴ was cleaved at the SalI/EcoRI sites and was used as backbone for the construction of the chimeric molecular clone HTLV-1A/C_oI-L. A molar ratio of 1:7 (Vector: Insert) was used for the ligation in the presence of Takara Ligation mix (cat. #6023 Takara Bio, San Jose, CA). After 10 min of incubation at room temperature (RT), 3 µl of the ligation mix was used to transform 50 µl of the One Shot Stbl3 Chemically Competent E. coli (cat. #C737303, Thermo Fisher Scientific, Waltham, MA). Briefly, the cells/ligation mix was incubated on ice for 30 min, and the cells were then heat-shocked for 30 s at 42 °C in a water bath. The vial was returned and incubated on ice for another 2 min, and 250 μl of prewarmed SOC medium was then added to the cells and incubated in a shaking incubator for 2 h at 27 °C. The mix was then streaked on a prewarmed kanamycin-resistant plate and incubated at RT for 2 to 3 overnight. The smallest colonies were separately collected, and mini cultures were started by adding 4 mL of Luria Bertani (LB) Broth (cat. #BLF-7030, KD Medical, Columbia, MD) supplemented with 50 μg/mL kanamycin. The cultures were incubated in a shaker at 27 °C for 24 to 48 h. Plasmids were isolated using the Wizard Plus SV Minipreps DNA purification System (cat. #A1460, Promega, Madison, WI) following the manufacturer’s instructions. A restriction digestion was then performed using SpeI and NheI/KpnI restriction enzymes. The molecular clones displaying the expected restriction digestion pattern were amplified by Maxiprep using Qiagen Plasmid Maxi Kits (cat. #12162, Qiagen, Germantown, MD) following manufacturer’s instructions. The viral genomic sequences were verified by sequencing. The sequence of the pAB_HTLV-1A/C_oI-L chimeric molecular clone was submitted to NCBI GenBank nucleotide database (accession number PP860917). The HTLV-1C viral nucleotide sequence used in the cloning was derived from virus isolated from PBMCs of an infected donor. The nucleotide sequence was obtained from the Laboratory of Dr. Damian JF Purcell¹⁸ (accession nos. PP596271, PP596272, PP596273, PP596274, PP596275, PP596276, PP596277, PP596278, PP596279, PP596280, PP596281, PP596282, PP596283, PP596284, PP596285, PP596286, PP596287, PP596288, PP596289, PP596290, PP596291, PP596292 for all patient proviruses).

HTLV-1A/C_E-L chimeric molecular clone

An artificially synthesized DNA fragment encompassing a portion of the polymerase, the entire envelope, and orf-I of HTLV-1 subtype C was introduced into a BbvCI/XbaI cassette. The chimeric molecular clone HTLV-1A/C_oI-L described above was cleaved at the BbvCI/XbaI sites and was used as backbone for the construction of the chimeric molecular clone HTLV-1A/C_E-L.

HTLV-1C-LTR-Luc

An artificially synthesized DNA fragment of the 5′ long terminal repeat (LTR) region of HTLV-1C virus spanning the U3-R-U5 regions was introduced into an XhoI/HindIII cassette. The cassette was then swiped into the LTR-HTLV-1A-Luc at the XhoI/HindIII restriction sites⁵⁹. The bacterial transformation was performed as described above with few changes: i) a molar ratio of 1:3 (Vector:Insert), ii) incubation performed at 37 °C, and iii) ampicillin-resistant plate/media. The isolated plasmids were first verified by restriction digestion using XhoI/HindIII, and the plasmid displaying the expected restriction digestion pattern were sequenced and amplified by Maxiprep. The 5′LTR-HTLV-1C nucleotide sequence used in the cloning was derived from virus isolated from PBMCs of the same infected donor described above.

pRL-TK-Luc, HTLV-1A-LTR-Luc, and NF-κB-Luc⁶⁰

pSDM-12-HA, pSDM-p16A-HA, and pSDM-p16C-HA

Cassettes were digested with PmeI and BamH1 restriction enzymes and re-cloned into the PmeI and BamH1 digested pSDM101 vector⁶¹ (hereafter pSDM). All constructs were verified by sequencing.

Cell lines and primary cell culture

Adherent cell lines

HEK293T and HeLa cell lines were cultured in Dulbecco’s Modified Eagle’s Medium (DMEM) (Invitrogen, Carlsbad, CA) supplemented with 1% penicillin/streptomycin and 10% fetal bovine serum (FBS) (cat. #16140071, Life Technology, Carlsbad, CA). Suspension cell lines: Jurkat T-lymphoblastic and monocytic THP-1 cell lines were cultured in a Roswell Park Memorial Institute (RPMI 1640) medium (Invitrogen, Carlsbad, CA) supplemented with 1% penicillin/streptomycin and 10% FBS. Peripheral blood mononuclear cells: PBMCs were separated from blood of healthy human donors, naïve rhesus macaques, or heparinized human neonatal umbilical cord blood by density gradient centrifugation by Ficoll Plaque (GE Healthcare).

CD4⁺ cell isolation

CD4⁺ cells were isolated from cryopreserved human and non-human primate PBMCs using PE-conjugated antibodies and Anti-PE Microbeads (cat. #130-048-801, Miltenyi Biotec). Briefly, 30 × 10⁶ PBMCs were thawed, then incubated at 4 °C for 15 min with 155 µl of MACS buffer/tube and 45 µl of PE anti-CD4 (clone L200; cat. #550630, BD Biosciences) antibody. Following incubation, cells were washed with 3 ml buffer per tube. Cells in both tubes were then incubated at 4 °C for 15 min with 240 µl of MACS buffer and 60 µl of anti-PE microbeads for each isolation. Cells were then washed again with 3 ml buffer, resuspended with 500 µl of MACS buffer, and isolated using the AutoMACSpro (Miltenyi Biotec). CD4⁺ cell positive selection was performed following the Possel program. Mononuclear cells were cultured in RPMI supplemented with 1% penicillin/streptomycin, 20% FBS, and 100 U/ml of purified interleukin-2 (IL-2) (cat. #I2644, Millipore Sigma, Rockville, MD) in a 6-well plate at a density of 1 to 5 × 10⁶ cells per well. The medium was renewed twice a week, and fresh IL-2 was added to the cultures.

Transfection

HEK293T and HeLa cells were transfected using LipoD293 In vitro DNA transfection Reagent (cat. #SL100668, SignaGen Laboratories, Frederick, MD) following the manufacturer’s instructions. Briefly, cells were plated 18 h prior transfection, allowing the cells to reach a 70 to 80% confluency by the time of the transfection. The diluted LipoD293 (LipoD293 + media) was added to the diluted DNA (DNA + media) to a ratio LipoD293 (μl) – DNA (μg) of 3:1. The Mix was briefly vortexed and incubated at RT for 10–15 min. The mix was then added dropwise onto the media in each well, and the mixture was homogenized by gently swirling the plate. 12 h post-transfection, the LipoD293/DNA complex-containing media was replaced, and the transfection efficacy was checked 24–48 h post-transfection.

Establishment of stable Jurkat (T-cell) and monocytic THP-1 cell lines expressing viral proteins

Jurkat and THP-1 cell lines were transduced with GFP-expressing lentiviral empty vector control (pSDM) or pSDM expressing p12-HA, p16A-HA, or p16C-HA. Briefly, 80% confluent HEK293FT cells (6 × 10⁶ cells) cultured in 10 cm dish, were transfected using LipoD293 transfection reagent as described above with a DNA mix constituted of 12 µg of pSDM, pSDM-p12-HA, pSDM-p16A-HA, or pSDM-p16C-HA in addition to 8 µg of psPAX-2 (Packaging plasmid, Addgene) and 6 µg pMD2.G (Env, Addgene). 48 h post-transfection, supernatants of the different cell cultures were collected, centrifuged for 5 min at 500 × g, and filtered through a 0.22 µm filter. Lentiviruses were concentrated by centrifugation at 8000 × g for 3 h at 4 °C. Concentrated lentiviruses were resuspended in 100 µl of RPMI. Jurkat and THP-1 cells were centrifuged for 5 min at 300 × g, then resuspended in the virus solution and incubated for 5 min at 37 °C. The mix of cells-lentivirus is then centrifuged for 10 min at 500 × g, and the pellet is resuspended in 1 ml of complete medium and transferred in a 12-well plate for 72 h. Transduction efficiency was measured by flow cytometry (FACSCalibur, BD). Additionally, the expression of viral proteins (p12, p16A, and p16C) was verified by Western blot for HA.

Spin-infection

Heparinized human neonatal umbilical cord blood and CD4⁺ T-cells were spin-infected with γ-lethally irradiated 729.6 lymphoblastoid B-cell lines producing either HTLV-1A or HTLV-1A/C_oI-L viruses. The number of donor cells (729.6 B-cells) used to infect the acceptor cells (primary cells) was normalized for p19 Gag production and viral DNA level, and the ratio of acceptor:donor was 1:0.5 and 1:1 for HTLV-1A and HTLV-1A/C_oI-L, respectively. Briefly, after isolation, 1 to 5 × 10⁶ primary cells were washed twice with RPMI supplemented with 10%FBS and 1% penicillin /streptomycin and the pellets were resuspended with the γ-lethally irradiated 729.6 B-cell lines in a final volume of 4 ml of RPMI supplemented with 10%FBS and 1% penicillin /streptomycin in the presence of polybrene with a final concentration of 8 μg/ml (cat. #sc-134220, Santa Cruz Biotechnology, Dallas, Texas). The donor/acceptor-cell mix was then centrifuged at 2000 × g at RT for 1 h, the supernatant was discarded, and the pellet was resuspended in RPMI medium supplemented with 20% FBS, 1% penicillin/streptomycin, and 100 U of IL-2. The media was changed twice a week, and fresh IL-2 was added to the primary culture.

Establishment of the HTLV-1A/C_oI-L producer cell line by electroporation

729.6 lymphoblastoid B-cell lines (hereafter 729.6 B-cells) (kindly provided by Patrick Green, Department of Veterinary Biosciences, Ohio State University), were transfected using the Amaxa Cell Line Nucleofector Kit V (cat. #VCA-1003, Lonza, Morristown, NJ) and following manufacturer’s instructions. Briefly, 5 × 10⁶ of low-passaged 729.6 B-cells were centrifuged at 600 × g for 10 min, and the pellet was resuspended in 100 μl of Nucleofector Solution. The cell/Nucleofector solution mix was then mixed with 5 μg of the pAB HTLV-1A/C_oI-L molecular clone and then transferred into a certified cuvette. Electroporation was performed using the pre-set optimized M-013 program on the Amaxa-Nucleofector II (Amaxa Biosystems). After the electroporation, 500 μl of RPMI media complemented with 20% FBS and 1% penicillin/streptomycin is immediately added into the cuvette, and the cells were incubated at RT for 10 min before transferring them into larger flasks containing 4.5 ml of media and incubated at 37 °C. 48 h post-electroporation the cells were collected by centrifugation and resuspended in RPMI media supplemented with 10%FBS, 1% penicillin/streptomycin, and 100 mg/ml of Geneticin (cat. #ant-gn-5, Invivogen, San Diego, CA), used as a selection agent. Following 4–5 weeks of selection period, viable cells were expanded in culture for further analysis.

Western blot

Cell lysates were separated by SDS-PAGE (cat. #NP0321, NuPAGE 4–12% Bis-Tris Protein Gels, Thermo Fisher Scientific, Waltham, MA) and transferred to a polyvinylidene difluoride membrane (Immobilon-P PVDF, Millipore Sigma, St. Louis, MO). The membranes were incubated overnight at 4 °C with primary antibodies to HTLV-1 p24 Gag (Mouse, Applied Biological Laboratories cat. #4310, 1:1000), HTLV-1 gp46 envelope (Mouse, Creative Biolabs, cat. #CBMAB-V208-1154-FY, 1:100), Tax (Tab172, 1:100 ref. ³⁴; Tax-LT-4, Mouse, Millipore Sigma cat. #MABF3063, 1:1000; Tax-1A3, Mouse, Abcam cat. #ab26997, 1:1000), GFP (Mouse, Thermo Fisher Scientific, cat. #MA5-15256), β-actin (Rabbit Cell Signaling Technology, D6A8, 1:1000), GAPDH (Rabbit Cell Signaling Technology, D3U4C, 1:1000), HA (Rabbit, Cell Signaling Technology, C29F4, 1:1000) in PBS containing 0.1% Tween 20 and 0.25% milk. Membranes were washed in PBS 0.1% Tween and exposed to a horseradish peroxidase-conjugated goat secondary anti-Mouse or anti-Rabbit antibody (Thermo Fisher Scientific NA931 and NA934, respectively; 1:10.000). Blots were developed with the use of either SuperSignal West Pico Plus or West Femto Maximum sensitivity substrate chemiluminescent detection system (cat. #34579 and 34095, respectively, Thermo Fisher Scientific). Proteins were visualized by chemiluminescence using a ChemiDoc Imaging System (Bio-Rad Laboratories, Hercules, CA). Full scan blots are in the Source data file.

Dual-Glo luciferase assay

HEK293T cells were transfected with HTLV-1A-LTR-Luc, HTLV-1C-LTR-Luc, or NF-κB-Luc reporter plasmids and either HTLV-1A or HTLV-1A/C_oI-L molecular clones with a reporter-molecular clone ratio of 2:1 and a LipoD293-DNA ratio of 3:1. The amount of DNA transfected was equalized by addition of a control vector (pcDNA 3.1). All the transfections were carried out in the presence of a pRL-TK-Luc vector in order to normalize the results for the transfection efficiencies. 48 h post-transfection, cells were extracted with Passive Lysis Buffer (Promega, Milwaukee, WI) and reporter activities were assayed using Dual-Glo reagent (cat. #E2940, Promega, Milwaukee, WI) following manufacturer’s instructions. Briefly, to each well containing 5 μl of cell lysate, 25 μl of Dual-Glo Reagent is added, and the mix is incubated on a shaker at RT for at 10 min then the firefly luminescence is measured in a luminometer. A volume of Dual-Glo Stop & Glo Reagent equal to the volume of Dual-Glo Reagent is then added to each well, and the mix is incubated at RT on a shaker for 10 min, and the Renilla luminescence is measured. Luciferase assays were performed with the VictorX4 2030 multilabel plate reader (PerkinElmer, Waltham, MA). Results were normalized for Renilla luciferase activity, and fold change induction was calculated by dividing each luciferase activity value by values of the cells transfected in the absence of the molecular clone plasmids (pcDNA 3.1).

RNA extraction and reverse transcription PCR

Total RNA was extracted from the stably infected cell lines, stably transduced Jurkat and THP-1 cell lines, as well as from macaque lung lobes using RNeasy plus mini kit (cat. #74034, Qiagen, Germantown, MD) following the manufacturer’s instructions. Briefly, the cells were harvested and lysed in the RLT Plus buffer. The homogenized lysates are then transferred to a gDNA eliminator spin column and centrifuged for 30 s at 8500 × g. 70% ethanol is then added to the flow-through, and the mix is transferred to the RNeasy spin column and centrifuged. The column is then washed 3 consecutive times using once the RW1 and twice the RPE buffers. Following the last wash, the column is air-dried by centrifugation at high speed for 1 min. The RNAs are then eluted in 50 μl of nuclease-free water by centrifugation. The eluted ARNs are quantified using a Nanodrop ND-2000 apparatus (Thermo Fisher Scientific). RNA samples were then subjected to Reverse transcription using QuantiTect Reverse transcription Kit according to the manufacturer’s instructions (cat. #205311; Qiagen, Germantown, MD). Briefly, 500 ng of total RNA was mixed with gDNA wipeout Buffer and RNase-free water. The mix is incubated for 3 min at 42 °C, then placed immediately on ice. A second mix of reverse transcriptase, reverse transcriptase buffer, and reverse transcriptase primers is then added to the first mix and incubated for 30 min at 42 °C, followed by an incubation at 95 °C for 3 min to inactivate the reverse transcriptase. The cDNA materials are then stored at −80 °C until use. Prior to RNA extraction from lung lobes, the tissues preserved in RNAlater™ were thawed on ice and centrifuged at 800 × g for 3 min at 4 °C. The supernatant was discarded and 600 μl of RPE buffer supplemented with β-Mercaptoethonal (10 μl in 1 ml) was added to each sample. The tissues were resuspended by pipetting until dissolution, and the lysates were then centrifuged through a QIAshredder (cat. #79654; Qiagen, Germantown, MD) homogenizer spin column at full speed for 2 min.

DNA extraction and viral DNA detection

Genomic DNA from the PBMC, bone marrow, and biopsies from the Lymphnodes the BAL, and the lung lobes were isolated from all animals during the time course of the study except the lungs lobes collected at the time of the sacrifice, using the DNeasy Blood and Tissue Kit (Qiagen) following the manufacturer’s instructions. 100 ng of DNA were used as templates for the first round of PCR amplification. Three microliters of the PCR reaction were then used as a template for nested PCR. The PCR conditions used were 94 °C for 2 min, followed by 35 cycles of 94 °C for 30 s, 55 °C for 30 s, 68 °C for 60 s, and a final extension at 68 °C for 7 min, and a hold at 4 °C. Platinum High Fidelity PCR SuperMix (Invitrogen, Carlsbad, CA) was used according to the manufacturer’s protocol. Correctly sized amplicons were identified by 1% agarose gel electrophoresis.

Primers

All primers used in the study (Supplementary Table 5) were designed using Primer3 software (v. 0.4.0) and produced by Eurofins Genomics LLC (Louisville, Kentucky).

Transcriptional characterization of HTLV-1A/C_oI-L

The primers (Supplementary Table 5) used to detect the different mRNA viral species in the 729.6 B-cells stably infected cell lines were designed using Primers3 (v.0.4.0); either on the exon sequence for gag-pro-pol, tax-orf-II, usHBZ, and β-actin or boundary spanning primers overlapping the splice sites of the singly (env, orf-I, orf-II, p21^rex, and sHBZ) and doubly spliced (tax/rex and rex-orf-I) mRNAs. The PCR was performed in the presence (+) or absence (−) of the Reverse Transcriptase (see section above). The PCR conditions used for amplification were 94 °C for 2 min, followed by 35 cycles of 94 °C for 30 s, 57 °C for 30 s, 68 °C for 60 s, and a final extension at 68 °C for 7 min, and a hold at 4 °C. Platinum High Fidelity PCR SuperMix (cat. #12532016, Invitrogen, Carlsbad, CA) was used according to the manufacturer’s protocol. The PCR products were separated on a 1% agarose gel and visualized by ethidium bromide staining. Correctly sized amplicons were identified by 1% agarose gel electrophoresis. Sanger sequencing was carried at Center for Cancer Research Genomics Core at the National Cancer Institute, NIH; on all transcripts to check for any depletion or mutation that would affect the amino acid sequence of the encoded protein. The failure to detect a signal in the absence of reverse transcriptase confirmed that the mRNA is free of plasmid or genomic DNA contamination. Full scan gels are in the Source data file.

HTLV p19 antigen ELISA

p19 Gag detection was measured by ELISA assay in the culture supernatants of 1 × 10⁶ cells washed and seeded for 24 h in a 24-well plate in 1 mL of complete RPMI following the manufacturer’s instructions (ZeptoMetrix, Buffalo, NY). The p19 Gag detection in infected primary cell cultures was measured as accumulation of p19 Gag antigen without normalization of the cell number.

HTLV-1 p24 antibody titer

HTLV-1 p24 antibodies in plasma samples from macaques were detected and quantified against purified HTLV-1 p24 protein using an ELISA assay. Each well of the ELISA plate was coated with HTLV-1 p24 core antigen (Prospec, cat. #hiv-108-c) at 500 ng per well and incubated overnight at 4 °C. Wells were then emptied and blocked with Superblock Blocking Buffer in PBS (cat. #37515, Thermo Fisher Scientific, Waltham, MA) for 1 h at room temperature. Serially diluted samples were added to the wells and incubated overnight at 4 °C. The plate was then washed with PBS Tween 20 and incubated for 1 h at 37 °C with the diluted goat anti-human IgG HRP (Kirkegaard & Perry Lab Inc., Gaithersburg, MD). The plate was washed, and the 1-Step Ultra TMB-ELISA (cat. #34029, Thermo Fisher Scientific, Waltham, MA) was added to the wells and incubated for 30 min at room temperature. The reaction was stopped by adding _KPL TMB Stop Solution (cat. #5150-0019), and the plate was read at 450 nm using VictorX4 2030 multilabel plate reader (PerkinElmer, Waltham, MA).

HTLV serology

Reactivity to specific HTLV-1 viral antigens in the plasma of the animals was detected with the use of a commercial HTLV-1 western immunoblot assay (GeneLabs Diagnostics, Redwood City, CA) following the manufacturer’s instructions. Briefly, pre-hydrated strips are incubated for 1 h at RT on a rocking platform with 20 μL of animal sera in the presence of a blotting buffer. The strips are washed three times, then incubated for 1 h with the working conjugated solution. After the washes the Substrate solution is added to the strips and incubated at RT for 15 min. The strips are then washed and dried.

Proximity extension assay (PEA)

Plasma and BAL samples

Protein quantification was executed employing the Olink® Target 48 Cytokine panel* (Olink Proteomics AB, Uppsala, Sweden) in accordance with the manufacturer’s protocols. This method leverages the Proximity Extension Assay (PEA) technology, as extensively detailed by Assarsson et al.⁶². This specific PEA methodology enables the concurrent assessment of 45 distinct analytes. Briefly, we used pairs of oligonucleotide-labeled antibody probes, each tailored to selectively bind to their designated protein targets. Probe pairs mix was incubated with 1 μl of plasma or BAL fluid. Probes that encountered their cognate proteins are then in close spatial proximity, and their respective oligonucleotides engage in pair-wise hybridization. A DNA polymerase was used to amplify the polymerized DNA and to create distinct PCR target sequences. Subsequently, we detected and quantified these newly formed DNA sequences through utilization of a microfluidic real-time PCR platform, specifically the Biomark HD system by Fluidigm (Olink Signature Q100 instrument). Data validation to uphold data integrity was conducted with the Olink NPX Signature software, specifically designed for the Olink® analysis: the application was used to import data from the Olink Signature Q100 instrument and process the data. Data normalization procedures were executed employing an internal extension control and calibrators, thereby effectively mitigating any inherent intra-run variability. The ultimate assay output was reported in pg/ml, predicated upon a robust 4-parameter logistic (4-Pl) fit model, thereby ensuring precise absolute quantification. Comprehensive insights into the assay’s validation parameters, encompassing limits of detection, intra- and inter-assay precision data, and related metrics, are available at www.olink.com.

Output from the Olink software was further processed to extrapolate values for samples that were below the lower limit of quantification (LLOQ) and above the upper limit of quantification (annotated as >ULOQ). The Olink software interpolates values below the LLOQ through fitting the 4-Pl model to a distinct minimum limit of detection for each plate of samples run, and values below this interpolation range are set to NaN. Since these values are below the limit of detection but not truly missing, for each assay, we determined a universal below detection value by taking the mode LLOQ across 22 plates, divided by 10,000 (which was below all interpolated values in this extensive historical dataset), and set all NaN to this assay-specific below detection value. Samples that were above the ULOQ were set to the ULOQ value for the indicated target assay from the plate on which the sample was run. Finally, true missing values annotated as No Data were converted to NA to be systematically treated as missing.

THP-1 stably transduced cells

THP-1 p12-HA-GFP or THP-1 p16C-HA-GFP stable cell lines were plated in 24-well plate at a density of 1 × 10⁶ cells/mL. 24 h post-treatment with PMA, the culture supernatant was removed, and a fresh medium was added. 24 h later, the culture supernatant was collected and stored at −80 °C.

Flow cytometry analysis

For the whole blood cell phenotyping, 100 μl of fresh EDTA whole blood was stained with Fluorochrome-conjugated mAbs. 1μl of the following antibodies were used: FITC anti-CD8 (clone DK25; cat. #FCMAB176F; EMB Millipore Corp.), BB700 anti-CD14 (clone M5E2; cat. #745790; BD Biosciences), PE-Cy5 anti-CD95 (clone DX2; cat. #305610; BioLegend), PE-Cy7 anti CD159 (NKG2a) (clone Z199; cat. #B10246; Beckman Coulter), APC anti-CD66abce (clone TET2; cat. #130-118-539; Miltenyi Biotec), Alexa 700 anti-CD3 (clone SP34-2; cat#.557917; BD Biosciences), APC-Cy7 anti-CD11b (clone ICRF44; cat. #557754; BD Biosciences), BV421 anti-CD16 (clone 3G8; cat. #562874; BD Biosciences), BV570 anti-CD20 (clone 2H7; cat. #302332; BioLegend), BV750 anti-CD4 (clone L200; cat. #747202; BD Biosciences), BV786 anti-CD45 (clone D058-1283; cat. #563861; BD Biosciences), BUV496 anti-CD28 (clone CD28.2; cat. #741168; BD Biosciences), BUV563 anti-CD49d (clone 9F10; cat. #749455; BD Biosciences), BUV661 anti-HLA-DR (clone G-46-6; cat. #612980; BD Biosciences), BV711 anti-CD11c (clone B-ly6; cat. #741139; BD Biosciences), BV650 anti-CD123 (clone 7G3; cat. #572392; BD Biosciences), BUV805 anti-CD8 (clone SK1; cat. #612889; BD Biosciences). Blue LIVE/Dead viability dye (cat. #L23105; Thermo Fisher Scientific, Waltham, MA) was used to exclude dead cells. T-cell lineages were identified following the gating strategy; i) Singlets/Live/CD45⁺/CD20⁻CD14⁻/CD3⁺CD4⁺ for CD4⁺ T-cells, ii) CD4⁺CD8⁻/CD95⁺ for CD4⁺ memory helper T-cells, iii) CD4⁺CD95⁻ for CD4⁺ naïve helper T-cells, iv) CD4⁺CD95⁻CD28⁻ for CD4⁺ effector memory helper T-cells, v) CD4⁺CD95⁻CD28⁺ for CD4⁺ central memory helper T-cells, vi) CD3⁺CD4⁻CD8⁺ for CD8⁺ cytotoxic T-cells (CTLs), vii) CD3⁺CD8⁺CD95⁺ for memory CTLs, viii) CD8⁺CD95⁻ for naïve CTLs, ix) CD8⁺CD95⁺CD28⁻ for effector memory CTLs, x) CD8⁺CD95⁻CD28⁺ for central memory CTLs. NK and neutrophils were identified following the gating strategies Singlets/Live/CD45⁺/CD3⁻CD20⁻/CD14⁻/CD8⁺NKG2a⁺, and Singlets/Live/CD45⁺/CD20⁻CD3⁻/CD8⁻/CD123⁻CD11c⁻/CD14⁻CD16⁻/CD66abce⁺, respectively. Monocyte populations were identified as Singlets/Live/CD45⁺/HLA⁻DR⁺CD20⁻/CD3⁻CD8⁻ and differentiated by the expression of CD14 and CD16⁶³. Classical monocytes were identified as CD14⁺CD16⁻, intermediate monocytes as CD14⁺CD16⁺, and non-classical monocytes as CD14⁻CD16⁺. Briefly, following the 30 min staining at RT, the red blood cells were lysed by incubating the samples with the BD FACS Lysing solution (cat. #349202 BD Biosciences, San Jose, CA) for 10 min at RT. Samples were washed with PBS and resuspended in 1% ultrapure formaldehyde (cat. #1008B-10 Tousimis, Rockville, MD). Flow cytometry acquisitions were performed on a FACSymphony A5 and examined using FACSDiva software (BD Biosciences) by acquiring all stained cells. Data was further analyzed using FlowJo v10.1 (TreeStar, Inc., Ashland, OR).

To measure neutrophils, monocytes, myeloid dendritic cells (mDC), and plasmacytoid dendritic cells (pDC) in bronchoalveolar lavage (BAL) and whole blood of the animals, 200 μL of whole EDTA blood and 1 × 10⁶ cells freshly isolated from BAL were stained with Blue LIVE/DEAD viability dye (cat. #L34962, Thermo Fisher Scientific) to exclude dead cells. 5μl The following antibodies were used for cell surface staining: FITC anti-CD66abce (clone TET2; cat. #130-116-522; Miltenyi Biotec), BB700 anti-CD162 (clone KPL-1; cat. #745768; BD Biosciences), Alexa 700 anti-CD3 (clone SP34-2; cat. #557917; BD Biosciences), Alexa 700 anti-CD20 (clone 2H7; cat. #560631; BD Biosciences), APC-Cy7 anti-CD11b (clone ICRF44; cat. #47-0118-42; Invitrogen™), BV480 anti-CD11c (clone 3.9; cat. #748269; BD Biosciences), BV650 anti-CD8 (clone RPA-T8; cat. #563821; BD Biosciences), BV750 anti-CD206 (clone 19.2; cat. #746891; BD Biosciences), BV786 anti-CD45 (clone D058-1283; cat. #563861; BD Biosciences), BUV395 anti-123 (clone 7G3; cat. #564195; BD Biosciences), BUV496 anti-CD16 (clone 3G8; cat. #612944; BD Biosciences), BUV563 anti-CD163 (clone GH1/61; cat. #741402; BD Biosciences), BUV661 anti-HLA-DR (clone G-46-6; cat. #612980; BD Biosciences), BUV737 anti-CD64 (clone 10.1; cat. #564426; BD Biosciences), BUV805 anti-CD14 (clone M5E2; cat. #612902; BD Biosciences). Subsequently, cells were permeabilized with Foxp3/Transcription Factor Staining Buffer Set (Invitrogen, cat. #00-5523-00) according to manufacturer’s recommendation. The following antibodies were used for intracellular staining: PE anti-MPO (clone MPO455-8E6; cat. #12-1299-42; Invitrogen™), BV421 anti-IL-8 (clone G265-8; cat. #563310; BD Biosciences), BV605 anti-TNF-α (clone mAB11; cat. #502936; BioLegend), and BV711 anti-IL-10 (clone JES3-9D7; cat. #564050; BD Biosciences). Samples were washed with PBS and resuspended in 1% ultrapure formaldehyde (cat. #1008B-10 Tousimis, Rockville, MD). Flow cytometry acquisitions were performed on a FACSymphony A5 and examined using FACSDiva software (BD Biosciences) by acquiring all stained cells. Data was further analyzed using FlowJo v10.1 (TreeStar, Inc., Ashland, OR). In the blood, (i) neutrophils were identified following the gating strategies Singlets/Live/CD45⁺/CD3⁻CD20⁻CD8⁻/CD123⁻CD11c⁻/CD14⁻CD16⁻/CD66abce⁺ cells⁶⁴, (ii) monocytes were identified as Singlets/Live/CD45⁺/CD3⁻CD20⁻CD8⁻/HLA⁻DR⁺/FSC-A^lowSSC-A^low and differentiated by the expression of CD14 and CD16 as previously published⁶⁵, classical monocytes (CD14⁺CD16⁻), Intermediate monocytes (CD14⁺CD16⁺), and non-classical as (CD14⁻CD16⁺); (iii) pDC and mDc were identified following the gating strategies Singlets/Live/CD45⁺/CD3⁻CD20⁻CD8⁻/HLA-DR⁺/CD14⁻/CD123⁺CD11c⁻ cells and Singlets/Live/CD45⁺/CD3⁻CD20⁻CD8⁻/HLA-DR⁺/CD14⁻/CD123⁻CD11c⁺ cells, respectively⁶⁶. In the BAL, (i) neutrophils were identified following the gating strategies Singlets/Live/CD45⁺/CD206⁻/CD163⁻/CD3⁻CD20⁻CD8⁻/CD123⁻CD11c⁻/CD14⁻CD16⁻/CD66abce⁺ cells, (ii) infiltrated monocytes were identified as Singlets/Live/CD45⁺/CD206⁻/CD163⁻/CD3⁻CD20⁻CD8⁻/HLA⁻DR⁺/FSC-A^lowSSC-A^low and differentiated by the expression of CD14 and CD16 as previously published, classical monocytes (CD14⁺CD16⁻), Intermediate monocytes (CD14⁺CD16⁺), and non-classical as (CD14⁻CD16⁺); (iii) pDC and mDc were identified following the gating strategies Singlets/Live/CD45⁺/CD206⁻/CD163⁻/CD3⁻CD20⁻CD8⁻/HLA-DR⁺/CD14⁻/CD123⁺CD11c⁻ cells and Singlets/Live/CD45⁺/CD3⁻CD20⁻CD8⁻/HLA-DR⁺/CD14⁻/CD123⁻CD11c⁺ cells, respectively.

Efferocytosis assay

The effect of the expression of p12 and p16C in apoptotic cells and how it affects the frequency of cells conducting efferocytosis was assessed by Efferocytosis Assay kit (cat. #601770, Cayman Chemical Company, Ann Arbor, MI) and CellTrace™ Far Red (cat. #C34564, Thermo Fisher Scientific, Waltham, MA), following manufacturers’ instructions. The efferocytosis assay was conducted using either monocytic THP-1 cell line or primary CD14⁺ cells isolated from human PBMCs as effector efferocytes, and stably transduced Jurkat cell lines expressing either GFP or p16C-HA-GFP or p12-HA-GFP as target cells.

Effector cells

CD14⁺ monocytes were isolated from cryopreserved healthy human donors’ PBMCs (n = 3) using human CD14 MicroBeads (cat. #130-050-201, Miltenyi Biotec) and following manufacturer instructions. Briefly, 30 × 10⁶ PBMCs were thawed and incubated with 60 µl microbeads and 240 µl buffer at 4 °C for 15 min. Cells were then washed with 3 ml buffer and resuspended in 500 µl of buffer. Positive selection was performed using the AutoMACSpro (Miltenyi Biotec) following the Possel program. At the end of the separation, CD14⁺ cells were resuspended in R10 (RPMI media, supplemented with 10% FBS and 1% antibiotic/antimycotic), counted, and washed once with PBS.

Monocytic THP-1 cells grown in R10 were counted and washed once with PBS. Washed THP-1 and CD14⁺ cells were stained with CytoTell™ Blue provided in the Efferocytosis Assay kit by following manufacturer’s instructions. Briefly, cells were resuspended in buffer (1 × 10⁷ cells/ml), an equal volume of buffer containing 2X CytoTell™ Blue (stock diluted 1:200 in kit buffer) was added, incubated at 37 °C for 30 min, washed three times with R10, resuspended in R10, and used for the efferocytosis assay.

Target cells

Stably transduced Jurkat-GFP and Jurkat-p16C-HA, and Jurkat-p12-HA express GFP; therefore, the dye CFSE contained in the Efferocytosis Assay kit could not be used. Target cells were stained with CellTrace™ Far Red following manufacturer instructions (cat. #C34564, Thermo Fisher Scientific, Waltham, MA).

Briefly, 40 × 10⁶ cells per cell line were washed twice with PBS, stained with Far Red staining solution (diluted 1:1000 with PBS) for 20 min at 37 °C. Cells were then washed twice with R10 and treated with apoptosis inducer. The apoptosis of the target cell lines was induced by treatment with Staurosporine Apoptosis inducer provided in the Efferocytosis Assay kit. Briefly, cells were resuspended in R10 media containing Staurosporine (stock diluted 1:1000) and incubated at 37 °C for 3 h. At the end of incubation, cells were washed twice with R10, resuspended 1 × 10⁶ cells/ml in R10, and used for the efferocytosis assay.

Coculture of target and effector cells

THP-1 or CD14⁺ effector cells and Jurkat apoptotic target cells were cultured alone (as controls) or cocultured at a ratio of one effector cell to three target apoptotic cells. For each THP-1 and human donor the experiment was done coculturing the cells in triplicate, with each tube containing 200,000 effector cells and 600,000 target cells. Cells were incubated at 37 °C for 2, 4, 8, 12, 18, and 24 h. At the end of the coculture, cells were washed with PBS, fixed with 1% paraformaldehyde in PBS, and acquired with a flow cytometer. Flow cytometry acquisitions were performed on a FACSymphony A5 and examined using FACSDiva software (BD Biosciences) by acquiring all stained cells. Data were further analyzed using FlowJo v10.1 (TreeStar, Inc.). The frequency of CD14⁺ efferocytes was determined as the frequency of GFP⁺ cells (target cells) in the CytoTell™ Blue⁺ cells (effector cells). The CellTrace™ Far Red was not used in the gating strategy to focus on the GFP⁺ cells that express the HTLV protein. Therefore, representing the frequency of effector cells that engulfed the apoptotic target cells. Gating strategy: FSC/SSC/Single cells/CytoTell™ Blue⁺/GFP⁺.

Immunofluorescence

HeLa cells transfected with PSDM p12 or p16C were cultivated and stained in µ-Slide 8 Well (ibidi) plate. Cells were fixed with freshly prepared 4.0% paraformaldehyde in PBS at RT for 5 min and rinsed thrice with PBS. Cells were permeabilized with PBS containing 0.5% Triton X-100 for 5 min at room temperature, rinsed thrice with PBS, and blocked with 4% BSA blocking buffer in PBS for 1 h at RT. Cells were then rinsed with PBS and incubated with an anti-HA antibody (Cell Signaling) for 1 h at RT (dilution 1:100). Following rinsing with PBS, cells were washed with 0.1% Triton X-100 in PBS. Cells were then incubated for a further 1 h in the presence of secondary antibodies, Alexa Fluor 568 (Thermo Fisher Scientific), dilution 1:1000. Again, cells were rinsed once with 0.1% Triton X-100 in PBS and washed three times with PBS. Cells were then incubated with 1 μg/ml DAPI (Thermo Fisher Scientific) in PBS for 30 min at RT. Cells were rinsed three times and kept at 4 °C in PBS until ready to image using the fluorescence microscope. Measurement of HA intensity in the peripheral region around the nucleus and the nuclei was used to calculate a ratio between the periphery/nucleus. Paired Student t-test was used for statistical evaluation.

Histopathology

Representative samples from each of the seven lung lobes were collected from each animal in accordance with the approved protocols by the NCI and/or NIAID Animal Care and Use Committees (ACUC; Protocol number: VB043) and preserved in 10% neutral buffered formalin for a minimum of 21 days. Subsequently, for all animals, one section of lung tissue from each of the seven lobes underwent processing using an automatic processor and was embedded into paraffin blocks. These blocks were then sectioned using a manual microtome and placed onto charged slides. Afterward, the slides were dried in an 80 °C oven for 1 h prior to Hematoxylin and Eosin (H&E) staining. The H&E staining process was carried out using the Sakura® Tissue-TekR Prisma™ automated Stainer, involving the application of commercial hematoxylin, clarifier, bluing reagent, and eosin-Y. A regressive staining method was employed, intentionally overstaining the tissues, and then utilizing a differentiation step (clarifier/bluing reagents) to remove excess stain. Finally, the slides were cover-slipped using the Sakura® Tissue-Tek™Glass® automatic cover slipper and allowed to dry before being evaluated in a blinded fashion for both control tissue and infected tissue.

HTLV-1-Gag RNAscope

The presence of HTLV-1 virus was detected by staining 5 μm FFPE Rhesus macaque lung sections with the RNAScope® 2.5 LS probe V-HTLV-1-GAG (cat. #495058, ACD) with the RNAscope LS Multiplex Fluorescent Assay (cat. #322800, ACD) using the Bond RX auto-stainer (Leica Biosystems) with a tissue pretreatment 15 min at 95 °C with Bond Epitope Retrieval Solution 2 (Leica Biosystems), 15 min of Protease III (ACD) at 40 °C, and 1:750 dilution of OPAL™ 570 reagent (AKOYA, Biosciences®). The RNAscope 2.5 LS Negative Control Probe (Bacillus subtilis dihydrodipicolinate reductase (dapB) gene, cat. #312038) was used as a negative control. The RNAscope® LS 2.5 Positive Control Probe Mmu-PPIB (peptidylprolyl isomerase B) (cat#457718) encoding cyclophilin B was used as a technical control to ensure the RNA quality of tissue sections was suitable for staining. Slides were digitally imaged using a PhenoImager® HT 2.0 (AKOYA, Biosciences®). It is important to note that RNAscope, immunohistochemistry, as well as histopathology assays were performed on consecutive slices from the same tissue fragment.

Immunohistochemistry

Immunohistochemistry staining was performed on LeicaBiosystems BondRX autostainer with the following conditions: Epitope Retrieval 1 (Citrate 20′), CD3 (cat. #MCA1477 Bio-Rad, 1:100 60′) with secondary antibody Rabbit anti-Rat IgG (Vector Laboratories), CD20 (cat. #M0755, DAKO/Agilent, 1:200 30′), Microglia/Iba1 (cat. #CP290, Biocare, 1:500 30′), Smooth Muscle Actin (cat. #ab5694, Abcam, 1:500 30′), and the Bond Polymer Refine Detection Kit (cat. #DS9800, LeicaBiosystems). Isotype control reagents were used in place of primary antibodies for the negative controls. Slides were removed from the Bond autostainer, dehydrated through ethanols, cleared with xylenes, and cover-slipped.

Viral score calculation

To assess viral infectivity of HTLV-1A and HTLV-1A/C_oI-L, four serological and virological assays (virus variables) measured at different time points were used to calculate timepoint-derived viral scores, variable-derived viral scores, and combination of those. All scores were then compared between the experimental treatment groups. Virus variables included viral DNA detection of gag (gag_PCR), viral DNA detection of orf-I/II (orf-I/II_PCR), number of bands in the HTLV serology assay, and HTLV-1 p24 antibody titer (p24_titer). Virus variables were normalized to 0–1 scale: specifically, gag_PCR and orf-I/II_PCR values were already binary (0/1), number_of_bands was divided by the maximum possible number of bands (=9), and log₁₀(p24_titer) was divided by the maximum log₁₀(p24_titer) value (= 4).

First, we calculated Timepoint-derived scores that combine the different virus variables for each animal at each time point. To give emphasis to the number of assays that showed a positivity for each time point and animal, the fraction of variables that have any signal was calculated. Finally, these 2 scores were combined by calculating the combined timepoint-derived score to merge the effects of the 2 previous scores.

1. (1).

   Timepoint-derived score for each timepoint and animal was calculated as the sum of all normalized virus variables divided by the total number of virus variables (=4), such that each tp_score_{WeekXX,AnimalY} has a range of 0–1.

   $$\begin{array}{c}{{{\mathrm{tp}}}\_{{\mathrm{score}}}}_{{{\mathrm{Week}}}\; {{\mathrm{XX}}},{{\mathrm{AnimalY}}}} \\=\frac{{{{\mathrm{gag}}}\_{{\mathrm{PCR}}}}_{{\mathrm{normWeekXX}},{\mathrm{AnimalY}}}+{{{\mathrm{pX}}}\_{{\mathrm{PCR}}}}_{{\mathrm{normWeekXX}},{{\mathrm{AnimalY}}}}+{{{\mathrm{number}}}\_{{\mathrm{of}}}\_{{\mathrm{bands}}}}_{{{\mathrm{norm}}}\; {{\mathrm{Week}}}\; {{\mathrm{XX}}},{{\mathrm{AnimalY}}}}+{{\log }_{10}\left(p24{\_titer}\right)}_{{norm\; Week\; XX},{AnimalY}}}{{Total}{Number}{of}{Virus}{Variables}}\end{array}$$

2. (2).

   The fraction of variables that have any signal was calculated as the ratio of the number of virus variables not equal to zero divided by the total number of virus variables (=4), such that each fracVars_notZero_{WeekXX,AnimalY} has a range of 0–1.

   $$ {{\mathrm{fracVars}}}\_{{\rm{notZeroWeekXX}}},{{\mathrm{AnimalY}}} \\ =\frac{{{\mathrm{Number}}}\,{{\rm{of}}}\,{{\rm{Variables}}}\_{{{\rm{Not}}}\,{{\rm{Zero}}}}_{{\rm{WeekXX}}},{{\rm{AnimalY}}}}{{{\rm{Total}}}{{\rm{Number}}}{{\rm{of}}}{{\rm{virus}}}{{\rm{Variables}}}}$$

3. (3).

   The combined timepoint-derived score for each timepoint/animal was calculated as the average of the tp_score_{WeekXX,AnimalY} from step 1, and fracVars_notZero_{WeekXX,AnimalY} from step 2. As above, each tp_comb_score_{WeekXX,AnimalY} has a range of 0–1.

   $$ {{{\rm{tp}}}\_{{\rm{comb}}}\_{{\rm{score}}}}_{{{\rm{WeekXX}}},{{\rm{AnimalY}}}} \\ =\frac{{{{\rm{tp}}}\_{{{\rm{score}}}}_{{{\rm{WeekXX}}}},{{\rm{AnimalY}}}}+{{{\rm{fracVars}}}\_{{\rm{notZero}}}}_{{{\rm{WeekXX}}},{{\rm{AnimalY}}}}}{2}$$

   Then, for each animal, timepoint-derived scores for each timepoint were combined to get a set of composite timepoint-derived scores across all timepoints.

4. (4).

   tp_comb_scores (from step 3) were averaged across timepoints, such that tp_comb_scores__Mean has a range of 0–1.

   $${{\mathrm{tp}}}\_{{\rm{comb}}}\_{{\rm{scores}}}\_{{\rm{Mean}}}=\frac{{\sum }_{i={Baseline}}^{n={Week}21}{{\rm{tp}}}{{\rm{\_}}}{{\rm{comb}}}{{\rm{\_}}}{{\rm{score}}}_{{\rm{i}}}}{{Total}{Number}{of}{Time}{points}}$$

5. (5).

   The fraction of tp_comb_scores (from step 3) with any signal was calculated as the ratio of the number of timepoints with tp_comb_score not equal to zero divided by the total number of timepoints (=8), such that fracTPscores_notZero has a range of 0–1.

   $${{\mathrm{frac}}}{{\rm{TPscores}}}\_{{\rm{notZero}}}=\frac{{\sum }_{i={Baseline}}^{n={Week}21}{{\mathrm{tp}}}\_{{\rm{comb}}}\_{{\rm{score}}}_{i}\,\ne\, 0}{{Total}{Number}{of}{Time}{points}}$$

6. (6).

   The combination timepoint-derived score across all timepoints was then calculated as the average of the tp_comb_scores__Mean (from step 4) and the fracTPscores_notZero (from step 5)

   $${{{\rm{TP}}}\_{{\rm{COMBO}}}}=\frac{{{{\rm{tp}}}\_{{\rm{comb}}}\_{{\rm{scores}}}\_{{\rm{Mean}}}}+{{{\rm{fracTPscores}}}\_{{\rm{notZero}}}}}{2}$$

   Next, we employed a different approach and calculated variable-derived viral scores individually for each virus variable. Similarly, in order to give emphasis to the total magnitude of the results of each assay, the variable-derived score for each animal was calculated by combining the different timepoints for each virus variable separately. To give emphasis to the number of assays that showed a positivity, the fraction of timepoints that have any signal was calculated for each assay and animal. Finally, these 2 scores were combined by calculating the combined variable-derived score to merge the effects of the 2 previous scores.

7. (7).

   Variable-derived score for each variable and animal was calculated as the sum of all normalized virus variables divided by the total number of timepoints (=8), such that each var_score_{variableZ,AnimalY} has a range of 0–1.

   $$\frac{{{{\mathrm{var}}}{\_ {\mathrm{score}}}}_{{\mathrm{variablrZ}},{{\mathrm{AnimalY}}}}=}{\frac{{{\mathrm{variableZ}}}_{{\mathrm{normBaseline}},{{\mathrm{AnimalY}}}}+{{\mathrm{variableZ}}}_{{\mathrm{normWeek}}03,{{\mathrm{AnimalY}}}}+\cdot \cdot \cdot+{{\mathrm{variableZ}}}_{{\mathrm{normWeek}}21,{\mathrm{AnimalY}}}}{{Total}{Numbe}r{of}{Time}{points}}}$$

8. (8).

   The fraction of timepoints that have any signal was calculated as the ratio of the number of timepoints not equal to zero divided by the total number of timepoints (=8), such that each fracTPs_notZero_{variableZ,AnimalY} has a range of 0–1.

   $$ {{{\mathrm{fracTPs}}}\_{{\mathrm{notZero}}}}_{{{\mathrm{variableZ}}},{{\mathrm{AnimalY}}}} \\ =\frac{{{\mathrm{Number}}}\,{{\mathrm{of}}}\,{{\mathrm{Time}}}\,{{{{\mathrm{points}}}\_{{\mathrm{NotZero}}}}}_{{variableZ},{AnimalY}}}{{Total}{Number}{of}{Time}{points}}$$

9. (9).

   The combined variable-derived score for each variable/animal was calculated as the average of the var_score_{variableZ,AnimalY} from step 7. and fracTPs_notZero__variableZAnimalY from step 8. As above, each var_comb_score_{variableZ,AnimalY} has a range of 0–1.

   $$ {{\mathrm{var}}}{\_{{\rm{comb}}}\_{{\rm{score}}}}_{{{\rm{variableZ}}},{{\rm{AnimalY}}}} \\ =\frac{{{\mathrm{var}}{\_ {\rm{score}}}}_{{{\rm{variableZ}}},{{\rm{AnimalY}}}}+{{{\rm{fracTPs}}}\_{{\rm{notZero}}}}_{{{\rm{variableZ}}},{{\rm{AnimalY}}}}}{2}$$

   Next, for each animal, variable-derived scores for each variable were combined to get a set of composite variable-derived scores across all variables.

10. (10).

    var_comb_scores (from step 9) were averaged across variables, such that var_comb_scores__Mean has a range of 0–1.

    $${{\mathrm{tp}}}\_{{\rm{comb}}}\_{{\rm{scores}}}\_{{\rm{Mean}}}=\frac{{\sum }_{i={Baseline}}^{n={Week}21}{{\mathrm{tp}}}\_{{\rm{comb}}}\_{{\rm{score}}}_{{\rm{i}}}}{{Total}{Number}{of}{Time}{points}}$$

11. (11).

    The fraction of var_comb_scores (from step 9) with any signal was calculated as the ratio of the number of variables with var_comb_score not equal to zero divided by the total number of variables (=4), such that fracVARscores_notZero has a range of 0–1.

    $${{\mathrm{fracVARscores}}}\_{{\rm{notZero}}}=\frac{{\sum }_{i={{\mathrm{gag}}}\_{{\mathrm{PCR}}}}^{n={{\rm{p}}}24\_{{\mathrm{titer}}}}{{\mathrm{var}}}{\_{\mathrm{comb}}}\_{{\rm{score}}}\_{{\rm{NotZero}}}_{{\rm{i}}}}{{Total}{Number}{of}{Time}{points}}$$

12. (12).

    The combination variable-derived score across all variables was then calculated as the average of the var_comb_scores__Mean (from step 10) and the fracVARscores_notZero (from step 11)

    $${{{\rm{VAR}}}\_{{\rm{COMBO}}}}=\frac{{{{\rm{Var}}}\_{{\rm{comb}}}\_{{\rm{scores}}}\_{{\rm{Mean}}}+{{\rm{fracVARscores}}}\_{{\rm{notZero}}}}}{2}$$

    Finally, for each animal, composite scores combining timepoint-derived and variable-derived scores were calculated.

13. (13).

    Mean timepoint-derived combined score (from step 4) and mean variable-derived combined score (from step 10) were averaged.

    $$ {{{\rm{viral}}}\_{{\rm{score}}}\_{{\rm{Mean}}}} \\ =\frac{{{{\rm{tp}}}\_{{\rm{comb}}}\_{{\rm{scores}}}\_{{\rm{Mean}}}}+{{\mathrm{var}}}{\_{{\rm{comb}}}\_{{\rm{scores}}}\_{{\rm{Mean}}}}}{2}$$

14. (14).

    Fraction not zero timepoint-derived combined score (from step 5) and fraction not zero variable-derived combined score (from step 11) were averaged.

    $$ {{{\rm{viral}}}\_{{\rm{score}}}\_{{\rm{fracScores}}}\_{{\rm{notZero}}}} \\ =\frac{{{{\rm{fracTPscores}}}\_{{\rm{notZero}}}}+{{{\rm{fracVARscores}}}\_{{\rm{notZero}}}}}{2}$$

15. (15).

    Combination timepoint-derived score (from step 6) and combination variable-derived score (from step 12) were averaged.

\({{{\rm{viral}}}\_{{\rm{score}}}\_{{\rm{COMBO}}}}=\frac{{{{\rm{TP}}}\_{{\rm{COMBO}}}}+{{{\rm{VAR}}}\_{{\rm{COMBO}}}}}{2}\)

Other analytical methods

Significant changes in cell population frequency relative to baseline were modeled by fitting generalized estimating equations, as implemented by the geeglm function from the geepack R package. Mann–Whitney/Wilcoxon tests were used to determine differences between cell population levels or fold-changes between experimental groups. Heatmaps were generated using the pheatmap R package, and alluvia were manually drawn to guide the eye through the data. Full code can be found at https://github.com/NCI-VB/franchini_HTLVchimera_sarkis. Since our research design was hypothesis-generating, exploratory research, all p values are reported as nominal values without adjusting for multiple comparisons.

Reporting summary

Further information on research design is available in the Nature Portfolio Reporting Summary linked to this article.