Patient consent and ethical considerations

We confirm that this research complies with all relevant ethical regulations. The study protocol was approved by the Norwegian Regional Committee for Medical and Health Research Ethics (REK, approval number REK no. 469419) and was conducted in accordance with the Declaration of Helsinki and institutional guidelines of Oslo University Hospital. The patient provided written informed consent for participation in the study and for publication of anonymized data. In addition to the approvals noted above, the patient was enroled in the IciStem cohort (IciS256) and the EU2Cure cohort (study no. 110723823).

Sample collection and preparation

Whole blood, serum and plasma samples were collected every 3–6 months from transplantation to 48 months follow-up, when the patient also underwent leukapheresis.

PBMC and CD4⁺ cell isolation

Whole blood was collected in a BD CPT Cell Preparation Tube (Becton Dickinson, cat. no. 362780) coated with sodium heparin, and peripheral blood mononuclear cells (PBMC) were isolated following the manufacturer’s instruction. In brief, CPT tubes were mixed by inverting the tube eight times and centrifuged (1,650g, 15 min, room temperature (RT)). The PBMC layer was transferred to a new tube (Corning 50-ml centrifuge tubes; Merck, cat. no. 430291) and washed twice with phosphate-buffered saline (PBS) (Euroclone, cat. no. ECB5004L), followed by centrifugation (300g, 15 min, RT). A portion of the PBMCs was cryopreserved in 10% dimethyl sulfoxide (DMSO) (Sigma-Aldrich, cat. no. D2650) and 90% foetal calf serum (FCS) (Biowest, cat. no. S181H) and stored at −150 °C for subsequent ELISpot and proliferation analyses. After isolation, PBMCs were resuspended in autoMACS rinsing buffer (Miltenyi Biotec, cat. no. 130-091-222) for positive selection through the magnetic bead isolation of CD4⁺ cells with human CD4⁺ MicroBeads (Miltenyi Biotec, cat. no. 120-000-440), following the manufacturer’s instructions using an autoMACS Pro Separator (Miltenyi Biotec, cat. no. 140-005-371.03). Pelleted CD4⁺ cells (300g, 15 min, RT and supernatant completely removed) were stored at −80 °C before lysis or DNA isolation.

Isolation of CD34⁺ cells from bone marrow aspirate

A bone marrow aspirate was drawn from the iliac crest using heparin as an anticoagulant. The aspirate was filtered through a 50-µM pore filter, and CD34⁺ cells were isolated using human CD34 MicroBeads (Miltenyi Biotec, cat. no. 130-046-702), following the manufacturer’s instructions using an autoMACS Pro Separator (Miltenyi Biotec). The cells were pelleted (300g, 15 min, RT), and the supernatant was removed. The CD34⁺ cells were stored at –80 °C until further analysis.

Colonoscopy and isolation of LPMCs

The patient underwent bowel lavage with 24 h of fasting and with administration of 10 mg sodium picosulphate (CitraFleet) 18 and 5 h before colonoscopy. The colonoscopy, which included the examination of the entire lower intestine and terminal ileum, was performed by an experienced gastroenterologist using a video colonoscope (Olympus). During the procedure, sets of pinch biopsies were collected with a single-use oval cup biopsy forceps with needle (EndoJaw 3.7 mm, Olympus) from sigmoid colon and terminal ileum, respectively. Biopsies for reservoir analysis with ddPCR were preserved in 10% DMSO (Merck, cat. no. 1.02931) and 90% FCS (Biowest, cat. no. S1860-500) and stored at −150 °C. In total, 20 biopsies from each site were pooled in a 50-ml Falcon tube with precooled 30 ml RPMI medium–L-glutamine (Lonza, cat. no. 12-702 F) with 10% FCS and 1% penicillin (100 U ml⁻¹)–streptomycin (100 μg ml⁻¹) (Gibco, cat. no. 10378016). The biopsies from each site were processed separately using an established enzymatic tissue digestion protocol. In brief, biopsies were washed twice in RPMI medium and transferred to digestion tubes containing preheated (37 °C) RPMI collagenase blend type H (1 mg ml⁻¹ final concentration; Sigma-Aldrich, cat. no. C8051-1G) and DNAse I (20 U ml⁻¹ final concentration; Invitrogen, cat. no. 18047-019). The tubes were incubated at 37 °C with gentle agitation for 40 min. Following enzymatic digestion, the samples were placed on ice, and the resulting tissue suspensions were repeatedly aspirated and expelled until a homogeneous single-cell suspension was obtained. The suspension was then filtered through a 70-μm cell strainer and washed once with cold medium. Lamina propria mononuclear cells (LPMCs) were cryopreserved in 10% DMSO (Sigma-Aldrich, cat. no. D2650) and 90% FCS (Biowest, cat. no. S181H) and stored at −150 °C until further analysis.

LPMCs were thawed, washed and rested for 12 h before staining. Cells were then washed and stained for 30 min at RT with the following fluorochrome-conjugated antibodies: anti-CD45 PerCP-Cy5.5 (BD Biosciences, cat. no. 332784), anti-CD3 AmCyan (BD Biosciences, cat. no. 339186), anti-CD4 PE (BD Biosciences, cat. no. 566910) and anti-CD8 APC (BD Biosciences, cat. no. 555369). CD45⁺ cells were sorted on a BD FACS Melody cell sorter (BD Biosciences) by gating on lymphocytes (forward scatter (FSC)-height (H) versus side scatter (SSC)-H), followed by gating on single (FSC-H versus SSC-width) CD45⁺ cells. Sorted cells were pelleted and cryopreserved for downstream analyses.

HIV reservoir quantification in blood

Plasma HIV RNA quantification

Blood samples were collected in BD Vacutainer standard EDTA-containing tubes or BD Vacutainer plasma preparation tubes (PPTs, cat. no. 362788). Standard EDTA-containing tubes were centrifuged at 1,100g for 20 min within 6 h of sample collection. PPTs were centrifuged at 1,100g for 10 min within 2 h of collection before transportation. PPTs centrifuged before transportation were recentrifuged after transport to minimize the number of peripheral blood cells in the plasma. The plasma samples were transferred to a secondary tube and stored at −70 °C until the day of analysis. The quantification of HIV RNA in EDTA-plasma was performed by the Cobas AmpliPrep-Cobas TaqMan HIV test, v1.0 before 2010 (Roche, cat. no. 03542998) and v2.0 from 2010 (Roche, cat. no. 05212308). From 2018, Cobas HIV on the 6800 system (Roche) was used. All tests were performed according to manufacturer’s instructions, and the limit of quantification was 40 HIV RNA copies per millilitre until 2010 and 20 copies ml⁻¹ thereafter.

The ultrasensitive viral load was measured 48 months post HSCT. In brief, 9 ml of plasma were ultraconcentrated at 170,000g at 4 °C for 30 min. Viral RNA was extracted from the pellet using the m2000sp Abbott 244 RealTime HIV Assay device and the corresponding laboratory-defined application software. HIV RNA was quantified with a validated in-house calibration curve, set with a limit of detection (LOD) of 0.11 copies ml⁻¹.

HIV proviral DNA routine analysis

HIV qualitative proviral DNA analysis was performed regularly post HSCT. Total nucleic acids were isolated from 200 µl EDTA whole blood using the EZ1 Advanced XL extraction instrument with the DSP Virus Card and EZ1 DSP Virus Kit (Qiagen, cat. no. 62724). The elution volume was 150 µl. HIV DNA was amplified by an in-house multiplex nested PCR in six parallel reactions with the JA17 through JA20 primer sets, which target the pol gene³⁶, and the OG 462 and OG 502 primer set, which targets the vif region³⁷. See Supplementary Table 3 for primer sequences. The first PCR was performed in a total volume of 50 µl containing 10 µl DNA extract, 1× AccuStart II PCR ToughMix (Quanta Biosciences, cat. no. 95142-100), 0.25 µM each of the primers JA17_IF and JA20_IR, OG462_IF and OG502_IR and was run on a Perkin Elmer GeneAmp PCR System 2720 (Applied Biosystems, Thermo Fisher). The cycling conditions were 5 min at 95 °C, followed by 25 cycles with 26 s at 94 °C, 30 s at 50 °C and 59 s at 72 °C, and the final extension was at 72 °C for 5 min. The nested PCR was performed in a total volume of 50 µl containing 1 µl of the first PCR product and 0.25 µM each of primers JA18 IIF and JA19_IIR, OG465_IIF and OG483_IIR using the same PCR master mix and PCR conditions as described for the first PCR. See Supplementary Table 3 for primer sequences. The PCR products were visualized by agarose gel electrophoresis and ethidium bromide staining. A quantitative PCR assay using primers and probe targeting the human retinoblastoma (RB) gene were used as an internal extraction control³⁸. The PCR assay was performed in a total volume of 25 μl containing 5 µl DNA extract, 1× TaqMan Universal PCR Master Mix, no AmpErase UNG (Applied Biosystems, cat. no. 4324018), 0.4 μM of each forward and reverse primer and 0.2 μM probe. The forward primer RB1-2672F, the reverse primer RB1-2750R and the probe RB1-2727_FAM were synthesized by Tib-Molbiol. The thermal cycling was performed on AriaDx real-time PCR instrument (Agilent Technologies) with cycling conditions set at: initial denaturation at 95 °C for 10 min followed by 45 cycles of 95 °C for 15 s and 60 °C for 60 s.

HIV DNA quantification in CD4⁺ T cells by ddPCR

Proviral total HIV-1 DNA was quantified from CD4⁺ T cells isolated from leukapheresis sampled 48 months post HSCT and HIV-negative control samples. Genomic DNA was isolated using the AllPrep DNA/RNA Mini Kit (Qiagen, cat. no. 80204) according to the manufacturer’s instructions. Total HIV-1 DNA and RPP30 was measured by ddPCR¹⁵. Reaction mixtures consisted of 1× Supermix no dUTP (Bio-Rad, cat. no. 1863025), 400 nM primers (Integrated DNA Technologies) and 125 nM probes (Thermo Fisher) targeting long terminal repeat (LTR)/gag, DNA template (4 µl and 2 µl for HIV and RPP30 reactions, respectively) and H₂O to a final volume of 22 µl. See Supplementary Table 3 for primer and probe sequences. Droplets were prepared using the Automated Droplet Generator (Bio-Rad) and cycled at 95 °C for 10 min, 45 cycles of 95 °C for 30 s, 59 °C annealing and extension for 60 s and, finally, 98 °C for 10 min. Droplet analysis was performed on a QX200 Droplet Reader (Bio-Rad) using the QX Manager software (Bio-Rad, version 2.1.0.25). HIV and human RPP30 reactions were performed separately in parallel, and the results were normalized to the amount of DNA input.

The cross-subtype intact proviral DNA assay (CS-IPDA) was used to assess intact proviral HIV DNA^39,40. CD4⁺ T cells were isolated from PBMCs (CD4⁺ T Cell Isolation Kit, Miltenyi Biotec, cat. no. 130-096-533), and genomic DNA was extracted using a method optimized to minimize DNA shearing⁴¹. First, cell pellets were resuspended in 3 M guanidine HCl (Life Technologies, cat. no. 24115) and 20 mM proteinase K (Qiagen, cat. no. 19133) and incubated at 56 °C for 1 h. Then, 6 M guanidine isothiocyanate (Merck, cat. no. 50983) was added and incubated at 42 °C for 10 min followed by precipitation of DNA using isopropanol (Merck, cat. no. I9516). Samples were resuspended in buffer EB (Qiagen, cat. no. 19086), and the extracted genomic DNA (gDNA) was then digested by Bg1I (New England Biolabs, cat. no. R0143L) overnight. Finally, DNA was precipitated using ethanol and resuspended in buffer EB at 150 ng µl⁻¹. The number of intact HIV copies per million CD4⁺ T cells was measured by multiplex ddPCR, targeting LTR/gag, 5′ pol and env, and the number of CD4⁺ T cells assayed was measured by the reference assay targeting RPP30 and deltaD, using the primers and probes in Supplementary Table 3. Reaction mixtures (final volume of 22 µl) were prepared as follows: HIV multiplex (LTR–gag/env/pol) reactions comprising 1× Supermix no dUTP (Bio-Rad, cat. no. 186-3024), 0.225 µM Env primer–probe mix, 0.228 µM 5′LTR/gag primer–probe mix, 0.288 µM 5′pol primer–probe mix, 3.7 µl nuclease-free water, 4 µl template. For the reference (RPP30/deltaD) reactions, 1× Supermix no dUTP (Bio-Rad, cat. no. 186-3024), 0.550 µM 5′RPP30 primer–probe mix, 0.550 µM 3′ RPP30 primer–probe mix, 0.550 µM deltaD primer–probe mix, 3 µl nuclease-free water and 2 µl template were mixed in a total volume of 22 µl. Droplets were prepared using the Automated Droplet Generator (Bio-Rad) and cycled at 95 °C for 10 min, 65 cycles of 94 °C for 30 s, 60.5 °C annealing/extension for 60 s and, finally, 98 °C for 10 min. Droplet analysis was performed on a QX200 Droplet Reader (Bio-Rad) using the QX Manager software (Bio-Rad, version 2.1.0.25). Gates for single-, dual- and triple-positive HIV-1 droplets were set on the basis of the positive control. HIV and human RPP30 reactions were performed separately in parallel, and the results were normalized to the amount of DNA input.

qVOA

For qVOA determination, we set up a limiting dilution virus culture assay using 65 million CD4⁺T cells from leukapheresis. Cryopreserved PBMCs were subject to negative selection to isolate total CD4 T cells using the Miltenyi CD4⁺ T Cell Isolation Kit (cat. no. 130-096-533), following the manufacturers’ instructions. On the same day, CD4⁺ T cells were played in 96-well round bottom culture plates at 50,000 CD4⁺ T cells per well (total of 1,312 culture wells) and cultured in medium consisting of RPMI with L-glutamine, 1% streptomycin and penicillin, 10% FCS, recombinant human IL-2 (100 U ml⁻¹) (Gibco, cat. no. PHC0027) and conditioned media from a mix lymphocyte reaction culture. The cells were then stimulated with 1 µg ml⁻¹ phytohemagglutinin (Remel PHA Purified, Thermo Scientific, cat. no. R30852801) in the presence of irradiated allogeneic PBMCs from HIV-negative healthy donors (10,000 cells per well), with a total culture volume of 200 µl. Culture wells consisting of irradiated PBMCs from healthy donors (10,000 cells per well) in 200 µl media with PHA were included as negative controls, and wells consisting of irradiated PBMCs (10,000 cells per well) and 30,000 ACH2 cells in 200 µl media with PHA were included as positive controls. After 2 days, the PHA was washed away, and the cultures were supplemented with 10,000 MOLT4/CCR5 cells per well. On days 5, 7 and 9, the culture media was replenished. On day 9, the cultures were supplemented with a further 10,000 MOLT4/CCR5 cells per well. On day 12, 50 µl culture supernatants were transferred to cultures of TZMbl cells (10,000 cells per well; total volume 200 µl; cultured in DMEM + 10% FBS), and these were cultured for a further 3 days. The number of infectious HIV units per million CD4⁺ T cells was then measured in a firefly luciferase reporter assay in TZMbl cells using the Britelite plus Reporter Gene Assay System, 100 ml (Perkin Elmer, cat. no. 6066761) and calculated using limiting dilution analysis⁴². As the qVOA experiment yielded no positive culture wells, the limit of quantification was assigned as the infectious units per million cells (IUPM) value for one culture well of viral outgrowth in the same experimental conditions. ACH2 cells (cat. no. 349), MOLT4/CCR5 (cat. no. 4984) and TZM-bl cells (cat. no. 8129) were obtained from the NIH HIV Reagent Program.

HIV reservoir quantification in tissues

Ileum and sigmoid colon biopsies (8 and 12 pieces, respectively) obtained 48 months post HSCT were cryopreserved as described above. After thawing, the biopsies were processed to isolate lamina propria leucocytes (LPLs)⁴³ by incubating them in Hank’s Balanced Salt Solution (HBSS; without Ca²⁺/Mg²⁺, Gibco, cat. no. 14170-070) containing 1 mM dithiothreitol (DTT; Sigma-Aldrich, cat. no. D5545-1G) and 1 mM EDTA (Gibco, cat. no. 15575-020) for 25 min, at RT with constant shaking, to remove the epithelial layer. Then, biopsies were transferred to fresh complete media, containing antifungal and antibiotics, and one to two biopsies per well were cultured overnight in six-well low-binding plates (Costar, cat. no. CLS-3471-24EA). Afterwards, culture supernatants were collected to recover the released cells, and the remaining tissue was disrupted by gentle pipetting. Finally, cell suspensions were cleared from tissue debris using a 40-μm nylon mesh (BD Falcon, cat. no. 734-0002). LPMCs were resuspended in PBS (Gibco) supplemented with 2 mM EDTA and 2% FBS and stained for 30 min at 4 °C with the following fluorochrome-conjugated antibodies: anti-CD45 FITC (Becton Dickinson, cat. no. 555482), anti-CD3 APC (Becton Dickinson, cat. no. 555335) and anti-CD8 PerCP (Becton Dickinson, cat. no. 345884). CD45⁺ cells were sorted using a BD FACS Aria II cell sorter (BD Biosciences) by first gating on singlets (FSC-H versus FSC-area (A)) and then selecting CD45⁺ cells from the lymphocyte gate (SSC-A versus FSC-A). Sorted CD45⁺ cells, defined as LPLs, were pelleted and resuspended in lysis buffer at a concentration of 5 × 10⁴ cells μl⁻¹ (minimum volume of 10 μl). The lysis buffer consisted of 0.1% Triton X-100 (Sigma-Aldrich, cat. no. T8787) and 400 μg ml⁻¹ proteinase K (Ambion, cat. no. AM2546) in 10 mM Tris–HCl (pH 9.0, Sigma-Aldrich, cat. no. T2819). Cell lysates were incubated at 55 °C for 15–20 h, followed by proteinase K inactivation at 95 °C for 5 min and then stored at −20 °C. Cells were lysed, and HIV DNA was quantified by LPL-viral-DNA ddPCR assay targeting LRT and gag⁴³. Total proviral HIV DNA was measured with probes targeting 5′ LTR and gag (FAM/HEX-ZEN-Iowa BlackFQ double-quenched probes, Integrated DNA Technologies). See Supplementary Table 3 for primer and probe sequences. Potential primer–template mismatches from viral sequence variation were avoided by using two primer–probe sets targeting conserved regions of the HIV-1 genome (5′ LTR and gag). For each reaction, 2 µl lysate was mixed with ddPCR Supermix for Residual DNA quantification (Bio-Rad, cat. no. 186–4037), 0.9 µM primers and 0.25 µM probe and H₂O to a final volume of 20 µl. The annealing and extension step was set at 57 °C to quantify vDNA using the C1000 Touch Thermal Cycler (Bio-Rad) and subsequently analysed using a QX100 droplet reader (Bio-Rad) and the QuantaSoft v.1.6 software (Bio-Rad). PBMCs from HIV-negative donors were used as negative controls and assayed in each plate to set the positive and negative threshold for ddPCR analysis, and the number of those negative control wells was the same as replicas for each sample. Undetectable values are represented as the value of the LOD. LOD was defined as the lowest HIV DNA concentration that would produce one positive target in the set of wells analysed. Total cell counts were calculated from RPP30 measurements across all wells, and the LOD is reported as HIV copies per 10⁶ cells. Undetectable samples are reported as the value of the LOD. Intact proviral HIV DNA was analysed by IPDA as described above.

Formalin-fixed biopsy samples collected at days 45 and 155 post HSCT were processed for DNA extraction using the AllPrep DNA/RNA FFPE kit (Qiagen, cat. no. 80234) on a QIAcube automated workstation (Qiagen), following the manufacturer’s instructions and eluted in 100 µl. Extracted DNA was quantified on a Qubit fluorometer and assessed for purity using Nanodrop spectrophotometer before downstream analysis. Total proviral HIV DNA was quantified by ddPCR using the QX200 ddPCR System (Bio-Rad Laboratories), targeting the 5′ LTR and gag regions, as described above in this section, using 4.45 µl eluate as input for LTR/gag and 1.1 µl input for RPP30⁴³.

CCR5 genotyping and viral coreceptor tropism analysis

Testing for the CCR5Δ32 allele was performed in a buccal swab sample from the donor at DKMS Life Science Lab in Dresden, Germany and in a blood sample from the patient at Histogenetics Lab in NY, USA.

For viral coreceptor tropism analysis, HIV RNA was isolated from 400 µl EDTA-plasma collected in 2006, using the protocol Total_NA_Plasma_100_400_V3_2 on the MagNA Pure Compact extraction instrument (Roche) and Magna Pure Compact Nucleic Acid Isolation Kit I (Roche, cat. no. 03730964001). The elution volume was 50 µl. HIV RNA was amplified by nested reverse transcriptase PCR (RT–PCR) using the primers targeting the V3-loop of the env gene⁴⁴. The RT–PCR was performed in a total volume of 50 µl using One-Step RT–PCR kit (Qiagen, cat. no. 210212) according to the manufacturer’s instructions, 10 µl RNA extract, 0.2 µM each of primers V3-forward and V3-reverse, 0.4 mM deoxynucleoside triphosphates each and 20 U RNasin (Promega, cat. no. N2111). See Supplementary Table 3 for primer sequences. RT–PCR was run on a Perkin Elmer GeneAmp PCR System 2720 (Biosystems) under the following cycling conditions: 30 min at 50 °C, 15 min at 95 °C, followed by 35 cycles with 30 s at 95 °C, 30 s at 57 °C and 90 s at 72 °C, and the final extension was at 72 °C for 10 min. The nested PCR was performed in a total volume of 50 µl containing 1 µl RT–PCR product, 1× AccuStart II PCR ToughMix (Quanta Biosciences, cat. no. 95142-04 K) and 0.2 µM each of primers V3-nested-for and V3-nested-rev and run on a Perkin Elmer GeneAmp PCR System 2720 (Applied Biosystems) under the following cycling conditions: 3 min at 94 °C, followed by 33 cycles with 30 s at 94 °C, 30 s at 57 °C and 60 s at 72 °C, and the final extension was at 72 °C for 5 min.

Before cycle sequencing, the nested PCR product was diluted 1:50 in PCR-grade water. The cycle sequencing was performed by forward and reverse priming using the PCR primers V3-nested-for and V3-nested-rev, BigDye Terminator v1.1 Cycle Sequencing Kit (Applied Biosystems, cat. no. 4337450) and 2 µl diluted PCR product. The amplification was performed using a Perkin Elmer GeneAmp PCR System 2720 (Applied Biosystems) with cycling conditions set at: 25 cycles at 96 °C for 10 s, 50 °C for 5 s and 60 °C for 150 s and kept at 4 °C until ready for purification with BigDye XTerminator (Applied Biosystems, cat. no. 4376486). Nucleotide sequences were determined by Sanger’s dideoxy chain termination method using the automated capillary DNA sequencer, Applied Biosystems 3130xl Genetic Analyser (Thermo Fisher). Coreceptor tropism was determined by analysing the V3 sequence using the online bioinformatics tool geno2pheno[coreceptor] (https://coreceptor.geno2pheno.org/). The submitted sequence returned a score termed the false positive rate, which is defined as the probability of incorrectly classifying an R5 strain as X4 with a cut-off more than 10% (ref. ¹²).

HIV antibody detection

Plasma samples from multiple time points were analysed using a qualitative western blot, with the MP Diagnostics HIV Blot 2.2 kit (MP Biomedicals, cat. no. 11030-018), following the manufacturer’s instructions.

The sample from 48 months post HSCT was analysed by the standard version of the LIASON XL HIV Ag/Ab assay (DiaSorin, cat. no. 310260). The same platform was used to measure avidity⁴⁵. Samples were diluted in duplicate in 200-μl aliquots at 1:10 in PBS or 1 M guanidine, using 20 μl of plasma to 180 μl of PBS or guanidine and incubating at RT for 10 min. Both guanidine- and PBS-treated duplicates were tested using the manufacturer’s recommended procedure for the LIASON XL HIV Ag/Ab assay. The avidity results were reported as an avidity index, which was calculated as a ratio of the S/C of the sample incubated in guanidine to the S/C of the sample incubated in PBS. Using a selection of samples from the IciStem cohort, a standard curve was created to compare values from the LIASON XL HIV Ag/Ab to the avidity versions of the VITROS Anti-HIV assay. These results were then compared with earlier cohort data⁴⁵.

HIV-specific T cell response assay

The IFN-γ ELISpot and lymphocyte proliferation assays were performed on cryopreserved PBMCs. PBMCs used for ELISpot were left for 3 h at 37 °C, whereas for lymphocyte proliferation assay, they were processed immediately. Positive control samples from PWH (n = 2) were obtained from the ROADMAP study⁴⁶. A confirmed HIV-negative buffy coat donor served as the negative control (n = 1).

In the IFN-γ ELISpot assay, 1 × 10⁵ cells per well were seeded in Multiscreen 96-well plates (Millipore, MSIPS4W10) precoated with unconjugated anti-human IFN-γ monoclonal antibody (clone 1-D1K; Mabtech, cat. no. 3420-3-1000). PBMCs were stimulated and incubated overnight (~18 h). For each sample replicate, eight conditions were examined: three negative controls (DMSO), three HIV-1 peptide pools each at a concentration of 1 μg ml⁻¹ (Gag (JPT, PM-HIV-Gag), Nef (JPT, PM-HIV-Nef), Pol (JPT, PM-HIV-Pol)), two positive controls (CEF peptide pool (JPT, PM-CEF-S-3, 1 μg ml⁻¹) and staphylococcal enteroxin B (SEB, 1 μg ml⁻¹)). Following incubation, plates were washed and incubated for 1 h with biotinylated anti-human IFN-γ monoclonal antibody (clone 7-B6-1, Mabtech, cat. no. 3420-6-250). After washing, plates were incubated for 1 h with streptavidin–alkaline phosphatase conjugate (Mabtech, cat. no. 3310-10-1000). Spots were developed by the AP conjugate substrate kit (Bio-Rad, cat. no. 1706432) and counted using the CTL Immunospot reader (S6 Entry M2, Cellular Technology). Responses were quantified as spot-forming cells (SFC) per 10⁶ input cells. A response was considered positive if the background-subtracted spot count exceeded the highest of the following three criteria: ≥5 spots per well, mean number of spots in negative-control wells plus three standard deviations or three times the mean number of spots in negative-control wells.

For the lymphocyte proliferation assay, PBMCs (1 million per millilitre) were stained with 5 µM CellTrace Violet (Thermo Fisher, C34557) for 20 min at RT. Subsequently, the reaction was quenched by adding RPMI glutamine supplemented with penicillin–streptomycin and 10% FCS at five times the original staining volume and incubated for 5 min at RT. Stained cells were seeded in 96-well U-bottom plates at 2 × 10⁵ cells per well and stimulated with three different HIV peptide pools: Gag (JPT, PM-HIV-Gag), Nef (JPT, PM-HIV-Nef) and Pol (JPT, PM-HIV-Pol) at a final concentration of 1 µg ml⁻¹ for 6 days. DMSO served as an assay negative control, whereas anti-human CD3 and anti-human CD28.2 (Nordic Biosite) were used as positive controls.

On day six, cells were stained for viability (LIVE/DEAD Fixable NEAR-IR, Invitrogen) by incubating for 20 min at RT. Cells were then treated with Human TruStain FcX (BioLegend, cat. no. 422302) for 10 min at RT, followed by 30 min of surface staining (PE/Dazzle 594 anti-human CD3 (Biolegend, cat. no. 317345), PE/Fire 700 anti-human CD4 (Biolegend, cat. no. 344665) and APC anti-human CD8 (Biolegend, cat. no. 344721)) in Brilliant Stain Buffer Plus (BD Horizon, cat. no. 566385) at RT. Samples were acquired on the MACSQuant16 (Miltenyi Biotec) and analysed using FlowJo v10.8.1 (BD Biosciences). The percentage of proliferating cells was determined by gating cells dimmer than the undivided peak with the DMSO control subtracted. The gating strategy is illustrated in Extended Data Fig. 2.

Chimerism analysis

The assessment of donor chimerism by short tandem repeat PCR analysis was performed by the Department of Forensic Medicine, Oslo University Hospital, as part of routine practice following HSCT, on the basis of donor and patient profiles obtained before HSCT. The method enables quantitative detection of recipient DNA with an analytical sensitivity of approximately 1%. DNA was isolated from peripheral blood (total nuclear cells) and CD2⁺ lymphocytes selected by magnetic beads (Dynabeads CD2, Invitrogen, cat. no. 111590), gut leucocytes (CD45⁺ cells sorted by flow cytometry as described above) and bone marrow stem cells (CD34⁺ cells) as described above. DNA was extracted using the EZ1&2 DNA blood kit (Qiagen, cat. no. 951054) and eluted in 350 µl. PCR amplification and short tandem repeat analysis were performed using the Promega PowerPlex Fusion 6C System (Promega).

Histology

Endoscopic biopsies were formalin-fixed and paraffin-embedded, cut at 4 µm and stained with haematoxylin and eosin. The slides were scanned, and images were obtained with Slideviewer (3DHISTECH).

Reporting summary

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