Application of a novel branched-DNA assay to quantify killer immunoglobulin-like receptor (KIR) mRNA expression identifies tissue compartmentalization in naïve and SIV-infected rhesus macaques

Abstract

Natural killer and cytotoxic CD8+ T cells are essential effectors of innate systemic antiviral and antitumor immune responses. Their effector functions and target cell recognition are partly regulated by a polymorphic family of activating and inhibitory killer-cell immunoglobulin-like receptors (KIRs) that interact with major histocompatibility complex (MHC) class I molecules. In Rhesus macaques (Macaca mulatta) (RM), a well-established model for studying simian immunodeficiency virus (SIV) infection and cellular immunobiology, the KIR gene family is highly diverse but remains poorly characterized across tissue compartments. In this study, we used a branched DNA, amplification-free assay with high specificity and reproducibility to simultaneously measure mRNA expression of six KIR genes (Mamu-KIR1D, Mamu-KIR3DH5, Mamu-KIR3DL2, Mamu-KIR3DL4, Mamu-KIR3DH, and Mamu-KIR3DL0) in mononuclear cells from peripheral blood, lymph nodes, and spleen of both naïve and chronically lentivirus-infected RM. Our findings reveal tissue-specific expression patterns of KIR genes that are differentially affected by chronic lentivirus infection, emphasizing the importance of compartmentalized KIR regulation in viral pathogenesis. This proof-of-concept study presents a reliable and scalable framework for the detailed characterization of the KIR gene repertoire in non-human primate models, providing a valuable alternative to traditional qPCR for profiling gene expression in complex tissues.

1. Introduction

Natural killer (NK) cells and cytotoxic CD8⁺ T lymphocytes are crucial effectors of innate and adaptive immune responses, respectively, and have vital roles in antiviral and antitumor defense. NK cells provide rapid, usually antigen-independent, cytotoxic responses through various families of innate receptors, serve as effector cells for antibody-dependent cell-mediated cytotoxicity (ADCC), and can produce immunoregulatory cytokines, such as interferon (IFN)-γ, that help shape adaptive immunity. In contrast, CD8⁺ T cells recognize specific viral peptides presented by major histocompatibility complex class I (MHC-I) molecules, allowing targeted destruction of infected cells after antigen priming and clonal expansion. KIRs are mainly expressed on NK cells but are also found on subsets of CD8⁺ T cells, where they regulate activation through interactions with MHC-I molecules. This regulation allows KIRs to play a key role in balancing activation and inhibition signals that control the effector functions of NK cells and T cells^1,2. Together, these cytolytic populations work to limit viral replication and maintain immune surveillance.

KIR genes are located within the leukocyte receptor complex (LRC) on chromosome 19q13.4 and exhibit extensive polymorphism in both gene content and allelic variation³. This genetic diversity leads to significant differences in immune responses among individuals. Each KIR protein has either two (2D) or three (3D) immunoglobulin-like extracellular domains, and the length of its cytoplasmic tail determines its signaling function. Long-tailed KIRs (KIRxDL) typically deliver inhibitory signals through immunoreceptor tyrosine-based inhibitory motifs (ITIMs), while short-tailed KIRs (KIRxDS) interact with the adaptor protein DAP12 to activate signals via immunoreceptor tyrosine-based activation motifs (ITAMs)⁴.

Rhesus macaques (RM) are a vital model species for studying a wide range of immunological and infectious diseases, including vaccine testing, transplantation immunology, and chronic viral infections. Their close evolutionary relationship and similar immune system structure allow translational studies that bridge the gap between rodent models and human clinical research⁵. Additionally, RM serve as important non-human primate models for exploring KIR biology and host-pathogen interactions, as their KIR gene family displays genetic diversity and polymorphism that closely resembles that in humans. Infection with simian immunodeficiency virus (SIV) in RM mimics key aspects of human immunodeficiency virus (HIV) immunopathogenesis, such as the progressive loss of CD4⁺ T cells, systemic immune activation, and the formation of tissue-resident viral reservoirs^6,7. The ability to conduct controlled, longitudinal, and tissue-specific sampling enhances the value of this model, enabling detailed studies of mucosal immunity, viral persistence, and neuroinvasion that are challenging to perform in humans. These features make RM especially useful for studying immune changes caused by chronic viral infections. Indeed, chronic SIV infection leads to dynamic shifts in NK cell phenotypes and KIR expression, particularly in secondary lymphoid organs. An expansion of KIR3DL-expressing NK cell subsets has been observed in lymph nodes, likely driven by sustained antigenic stimulation and abnormal MHC-I expression. Moreover, the presentation of viral peptides by MHC-I molecules can influence KIR binding, affecting NK cell activation and the effectiveness of viral control^8,9.

However, quantifying KIR gene expression in RM remains challenging due to high sequence similarity, polymorphism, and typically low transcript levels within the KIR gene family⁹. Conventional PCR-based or RNA-Seq methods often encounter amplification bias and cross-reactivity, which limit their ability to accurately measure individual transcripts. To address these issues, branched DNA (bDNA) assays offer high specificity and sensitivity for transcript quantification and have proven effective in measuring polymorphic immune receptor genes, including KIRs. In this study, we used a bDNA assay (QuantiGene Plex Assay, Thermo Fisher Scientific, Inc.) to measure the expression of six structurally and functionally distinct KIR genes—Mamu-KIR1D, -KIR2DL04, -KIR3DH, -KIR3DH5, -KIR3DL02, and -KIR3DL05—in mononuclear cells isolated from peripheral blood, spleen, and lymph nodes of naïve and chronically SIV/(simian-human immunodeficiency virus infection) SHIV-infected RM. These genes were chosen for their lineage diversity, evolutionary significance, and prior links to SIV immune responses¹⁰. By using the bDNA assay, we aimed to demonstrate its effectiveness as a sensitive and specific method for detecting subtle differences in KIR gene expression across various immune compartments and infection stages, thereby establishing a reliable approach for future studies on KIR biology in non-human primate models.

2. Methods

2.1. RM macaque samples

A total of sixteen adult Indian-origin RM aged between 3 and 7 years were used in this study. Animals were sourced from Biomere Inc. (Worcester, MA), Bioqual Inc. (Rockville, MD), the Washington National Primate Research Center (Seattle, MA), and the New England Primate Research Center (NEPRC; Southborough, MA). They were cared for in accordance with the recommendations of the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health, along with the suggestions of the Weatherall report, “The use of non-human primates in research.” All experimental protocols and procedures received approval from local Institutional Animal Care and Use Committees under protocols AUP19–018P (Bioqual), HU04637 (NEPRC), and VAC17–13 (Biomere). RM diets included standard monkey chow, supplemented daily with fresh fruit and vegetables, and water was available ad libitum. Social enrichment was provided and overseen by veterinary staff. Euthanasia and necropsy (when indicated) were conducted according to standard protocols. Both experimentally naïve and SIV- and SHIV-infected RM (Supplemental Table 2) were included in this study.

2.2. Mononuclear cell isolation

Peripheral blood for immunologic assays was collected in EDTA-treated tubes. Peripheral blood mononuclear cells (PBMCs) were isolated by density-gradient centrifugation (Ficoll or lymphocyte separation medium), washed in RPMI-1640 with 10% FBS, and RBCs removed via hypotonic ammonium chloride lysis. Cell viability and concentration were assessed via trypan blue exclusion.

At necropsy, the spleen and lymph nodes (LN) were harvested fresh and mechanically dissociated using scalpels, then pressed through a 70 μm nylon mesh. After washing, contaminating erythrocytes were lysed, and cells were resuspended in R10 medium. Cell viability and concentration were assessed via trypan blue exclusion. Final cell pellets from PBMCs, lymph nodes, and spleen were cryopreserved in 90% FBS / 10% DMSO and stored in liquid nitrogen vapor.

2.3. KIR selection and probe design

Thermo Fisher Scientific in Waltham, MA, developed custom probe sets for each target gene based on their NCBI accession numbers (Supplemental Table 3). Each magnetic bead set is covalently linked to capture probes specific to a single analyte. Following hybridization and signal amplification with the bDNA technology, the fluorescent signal intensity from each bead set was measured on a Luminex 2.0 instrument, allowing for target-specific detection of multiple transcripts.

2.4. Lysate preparation for QuantiGene analysis

Mononuclear cells isolated from PBMC, spleen, and LN were processed following the manufacturer’s instructions using the appropriate QuantiGene Sample Processing Kits (Thermo Fisher Scientific cat. No. QS0100) and tissue homogenates kit (Thermo Fisher Scientific cat. No. QS0106). Briefly, the supplied lysis mixture was pre-warmed to 37 °C for 30 minutes and supplemented with Proteinase K (20 μL per mL of lysis mixture) to create the working lysis mixture, which was then diluted 1:2 with RNase-free water. Cell pellets of mononuclear cells from each tissue source were resuspended in this working solution at approximately 1,000 cells per microliter. Samples were mixed by pipetting, briefly vortexed, and incubated at 50–55 °C for 30 minutes. After incubation, lysates were vortexed again and checked for viscosity; if necessary, additional pipetting or incubation was performed to achieve a uniform consistency. About 350 μL of lysate per sample was prepared, aliquoted, and stored at −20 °C until analysis.

2.4. bDNA assay

Following cell lysis, samples were hybridized with custom-designed QuantiGene 2.0 Plex probe sets (Thermo Fisher Scientific, Waltham, MA) that specifically targeted gene transcripts of interest. This hybridization process was performed overnight at 54°C in a 96-well plate containing magnetic capture beads pre-conjugated with the respective probe sets for each gene target. The hybridization mixture included blocking reagents and was agitated to maintain a consistent suspension of beads, thereby optimizing binding efficiency. Probe designs were based on NCBI accession numbers for each transcript (Supplemental Table 3). After hybridization, unbound materials were removed through serial washes using a magnetic plate washer. Signal amplification was achieved by sequential hybridization with a pre-amplifier, amplifier, and labeled probe oligonucleotides, according to the manufacturer’s protocol¹¹. The final detection step involved streptavidin-conjugated phycoerythrin (SAPE), and fluorescence intensity was measured using a Luminex 200 system. Raw data were collected as median fluorescence intensity (MFI) for each bead set, and expression values were normalized against the reference gene HPRT1. All samples were analyzed in duplicate, and results were accepted only when the intra-assay variation (coefficient of variation) remained within acceptable limits.

2.5. Statistical analysis

Raw data were collected as median fluorescence intensity (MFI) values for each bead set, and gene expression was normalized to the housekeeping gene HPRT1. All samples were tested in duplicate, and only results with acceptable intra-assay variability (i.e., low coefficient of variation) were used in further analyses.

The data were imported into QuantiGene Plex Analysis Software (Thermo Fisher Scientific, Waltham, MA), where plate layouts were configured and background subtraction was performed using wells containing media alone. Quality control (QC) metrics were reviewed, and the software flagged no critical errors. Normalization to HPRT1 was carried out using default software settings, and normalized median fluorescence intensity (normMFI) values were exported as CSV files. These normalized values were then imported into R v4.4.2, where they were scaled by a factor of 10³ to account for the initial sample input concentration. The processed data were subsequently imported into GraphPad Prism for visualization and statistical analysis. Differences in gene expression across tissues were assessed using the nonparametric Kruskal–Wallis test, followed by uncorrected Dunn’s post hoc comparisons. Statistical significance was set at p < 0.05.

3. Results

3.1. Assay performance and reproducibility across plates

QuantiGene Plex analysis software performs multiple quality control (QC) checks, such as bead count, coefficient of variation, and linearity assessments, each capable of flagging aberrant runs for review. Our samples were processed in several batches during the study. Comparisons of complete datasets showed no significant differences between runs, and QC metrics flagged no plates. Additionally, normalization using internal housekeeping genes (below) controls for technical variability across runs. Importantly, our findings are consistent with previous reports demonstrating the reproducibility of QuantiGene Plex (QGP) branched-DNA technology across different plates and experiments. Independent tests have shown low inter-plate coefficients of variation (CV, typically less than 5–10%) when data are normalized using stable reference genes, confirming that technical variability is minor compared to biological differences^12,13. For example, telomere length measurement with the QGP assay achieved inter-plate CVs of 3–5%, emphasizing its reliability in long-term studies¹⁴. High-throughput applications have also verified that plate-to-plate variability remains within acceptable limits, with housekeeping normalization ensuring accurate comparisons across plates. Overall, our results and the existing literature demonstrate that QuantiGene assays are highly dependable over time, supporting their use in comparative and multiplexed gene expression research¹⁵.

3.2. Baseline KIR Expression and Tissue Distribution in Naïve Animals

In naïve RM, KIR transcript levels demonstrated clear tissue-specific patterns across PBMC, LN, and spleen. LN exhibited the highest expression of KIR3DH5 and KIR3DL2, both approaching 3.0 log₁₀-transformed median fluorescence intensity (MFI), while splenic tissue was enriched for KIR3DL0 and KIR3DH, with expression values of ≥3.0 log₁₀ MFI. In contrast, PBMCs consistently showed the lowest expression across all six KIR genes (approximately 1.5–2.3 log₁₀ MFI) and exhibited minimal variability between animals (Figure 1). These findings underscore that KIR expression is not evenly distributed across compartments but rather reflects specialized tissue environments.

Figure 1. Within-animal variation in KIR expression across tissues of naïve RMs.

Paired line plots show normalized median fluorescence intensity (MFI) values for six KIR genes measured in PBMC, lymph node (LN), and spleen from naïve animals. Each light blue line represents one animal, with dark dots marking group medians. Black lines connect medians across tissues. P-values were calculated using ratio-paired t-tests comparing LN vs PBMC and spleen vs PBMC.

Pairwise comparisons within individual naïve animals confirmed significant differences between compartments for specific transcripts. KIR1D expression was consistently higher in LN than in PBMC (p = 0.0017), with a similar upward trend seen in spleen versus PBMC (p = 0.0537). KIR2DL4 also exhibited significant compartmental bias, with LN expression exceeding that in PBMC (p = 0.0042). For the other genes—KIR3DL0, KIR3DL2, KIR3DH, and KIR3DH5—expression varied across tissues; however, no pairwise differences were statistically significant (p-values ≥ 0.09) (Figure 1).

Overall, these data demonstrate that even without infection, KIR expression remains strongly compartmentalized, with LN and spleen serving as preferred sites for specific transcripts like KIR1D and KIR2DL4. This baseline understanding provides essential context for interpreting infection-related changes discussed below.

3.3. KIR Expression Across Tissues in Acute and Chronic Infection States

Bar graph analyses (Figure 2A) revealed overall tissue-level patterns: in naïve animals, LN expression was enriched for KIR3DL2 and KIR3DH5, while splenic tissue showed higher levels of KIR3DL0 and KIR3DH. PBMCs consistently had the lowest expression of all genes. Infected animals showed several decreases, especially in the spleen, where KIR3DL0 and KIR3DH levels decreased by nearly one log. KIR1D and KIR2DL4 also trended lower, while LN expression of KIR3DL2 decreased, and PBMC levels increased slightly. The heatmaps (Figures 2B and 2C) provided infection-specific details. In the LN (Figure 2B), acute SIV animals exhibited modest changes compared to naïve controls, with slight decreases in KIR1D and KIR2DL4. Chronic SIV animals showed broader reductions, especially in KIR3DL0, KIR3DL2, and KIR3DH. Chronic SHIV animals displayed more varied responses, with some resembling chronic SIV and others maintaining expression levels closer to naïve animals. In earlier analyses, all infected animals were grouped under a single SIV/SHIV+ category to compare the overall effects of infection with ]transcript expression by dividing animals into three groups: acute SIV, chronic SIV, and chronic SHIV. In the spleen (Figure 2C), acute SIV showed little change, whereas chronic SIV animals had the most notable reductions, notably in KIR3DL0, KIR3DH, and KIR1D. Chronic SHIV animals showed variable outcomes, with reductions in some transcripts but partial preservation in others. Across tissues and infection groups, no transcripts demonstrated consistent upregulation compared to naïve controls.

Figure 2. KIR gene expression in tissues of naïve and SIV/SHIV-infected RM macaques.

Normalized median fluorescence intensity (normMFI) values for six KIR genes (KIR1D, KIR2DL4, KIR3DH5, KIR3DL0, KIR3DL2, and KIR3DH) were quantified by bDNA assay in peripheral blood mononuclear cells (PBMC), lymph nodes (LN), and spleen. (A) Green bars represent naïve animals and orange bars represent SIV/SHIV-infected animals. Dots indicate individual animals; bars show mean ± SEM. P-values are provided in Supplemental Table 1. Y-axes are plotted on a log₁₀ scale. (B–C) Heatmaps display the log₂ fold-change in normMFI values for infected groups compared to naïve controls. (B) LN comparisons for SIV acute, SIV chronic, and SHIV chronic. (C) Spleen comparisons for the same groups.

Chronic SIV/SHIV infection had the most pronounced impact on the spleen, where several KIR transcripts showed a downward trend compared to naïve animals (Figure 2A; Supplemental Table 1). KIR1D expression slightly decreased in the spleen (Δ = 22120. 30 log₁₀ MFI; fold change 0.50×). Although this difference did not reach statistical significance (p = 0. 1810), the consistent downward trend across animals and its confinement to the splenic compartment suggest a localized effect rather than random variation. Levels in PBMC and LN remained stable (p > 0.80), highlighting the spleen’s unique response to infection.

KIR2DL4 showed a similar pattern, with a mild decrease in splenic expression (Δ = −0. 30; 0.62×; p = 0. 3277), while PBMC and LN levels showed no change (p > 0.99). Among all transcripts, KIR3DL0 exhibited the most notable decline—a complete log reduction in splenic expression (Δ = −1.00; fold change 0.57×). Although this trend was consistent, it did not reach statistical significance (p = 0.2721). KIR3DH followed a similar pattern, decreasing from approximately 3.0 to 2. 5 log₁₀ MFI (Δ = −0. 50; 0.57×; p = 0.2721), with minimal variation in other compartments (Figure 2A). These parallel decreases suggest a coordinated biological effect within splenic NK cell populations, despite a limited sample size reducing statistical power.

KIR3DL2 displayed a different pattern. In naïve animals, this transcript was enriched in LN, but this pattern decreased in SIV/SHIV+ RM (Δ = −0. 30; fold change 0.44×; p = 0. 8665). PBMC levels increased nearly fivefold (p = 0. 1797). Although not statistically confirmed, these contrasting shifts imply a possible redistribution of KIR3DL2⁺ subsets from lymphoid tissues to circulation. In contrast, KIR3DH5 remained remarkably stable across all compartments (Δ = 0.00; p ≥ 0.58), demonstrating resistance to infection-driven changes.

No transcripts were upregulated in the spleen during infection. Outside the spleen, PBMC and LN profiles generally remained consistent, except for the apparent shift in KIR3DL2 toward the bloodstream (Figure 2B–C). Even though formal statistical significance was not achieved (p = 0.1447–0.3277), the consistency, magnitude, and tissue-specific nature of these changes - particularly the 1.0 log₁₀ reduction in KIR3DL0 - support the idea that chronic SIV/SHIV infection selectively disrupts NK cell repertoires within lymphoid tissues rather than causing overall systemic effects.

4. Discussion:

In this study, we analyzed the expression of six killer immunoglobulin-like receptor (KIR) genes—KIR1D, KIR3DH5, KIR3DL2, KIR3DL4, KIR3DH, and KIR3DL0—in PBMCs, spleen, and LN mononuclear cells from both SIV/SHIV-negative and SIV/SHIV-positive RM. To conduct these analyses, we used multiplexed, bead-based branched-DNA (bDNA) technology, a non-enzymatic hybridization method that avoids amplification bias and enables direct RNA quantification from tissue lysates. This platform is particularly valuable for quantifying complex gene families, such as KIRs, where high sequence similarity and frequent copy number variation make quantitative PCR (qPCR) challenging. Its multiplexing ability allows for the simultaneous detection of up to 80 preselected transcripts within a single well, thereby saving sample input and enabling consistent comparisons across tissues. Although moderately sensitive (approximately 1–2 × 10^3 transcripts per well) and limited in dynamic range (~3 log units), bDNA assays are precise and reproducible. Importantly, because they are bead-based and probe-driven, they cannot discover novel transcripts; targets must be known beforehand and included in the assay design. Within these limits, bDNA is well-suited for quantifying pre-characterized, structurally complex families, such as KIRs. Our results revealed consistent enrichment of KIR1D, KIR2DL4, and KIR3DL0 transcripts in the spleens of naive animals compared to PBMCs and LN, highlighting tissue-specific compartmentalization (Figure 1). Although these differences did not reach conventional thresholds for statistical significance, their reproducibility across animals suggests biological relevance. In SIV/SHIV⁺ animals, these splenic enrichments were lost, indicating disruption of KIR compartmentalization during infection (Figure 2A).

These data highlight the importance of distinguishing between infection states when interpreting KIR expression. When all infected animals were grouped together (SIV/SHIV⁺), the primary outcome was a general decrease in splenic transcripts. However, stratification revealed that acute SIV animals maintained relatively preserved expression, whereas chronic SIV animals displayed the most pronounced reductions across both lymph nodes and the spleen. This pattern indicates that chronicity, rather than infection itself, drives the erosion of tissue-resident KIR repertoires. Responses in chronic SHIV animals were more variable, suggesting differences between SIV and SHIV in how they affect NK compartments - likely influenced by viral lineage, immune activation, or host–virus interactions¹⁶. Although not statistically significant, the consistent downward trend in changes in chronic SIV, in contrast to the variability seen in chronic SHIV, supports the biological relevance of these findings. Overall, these results suggest that chronic lentiviral infection modifies KIR expression in a tissue-specific and infection-dependent manner, with the spleen and LN being the most affected tissues. (Figure 2B,C).

The high expression of KIR1D in the spleens of naïve animals is notable, given its frequent classification as a truncated, possibly non-functional pseudogene¹⁷. Consistent with previous transcriptomic studies, our data confirm that KIR1D is actively transcribed in vivo, suggesting that it may identify specific NK or KIR+ T-cell subsets. Notably, the observed reduction of KIR1D in the spleens of SIV/SHIV⁺ animals aligns with our broader finding—and previous reports—that chronic SIV infection changes KIR repertoires in a tissue-dependent manner^18,19. In SIV/SHIV⁺ animals, KIR2DL4 expression was significantly decreased (Figure 2A), likely reflecting a combination of NK dysfunction, chronic activation/exhaustion, and dysregulated pro-inflammatory cytokines. Furthermore, the splenic microenvironment, including stromal support and cytokine availability, plays a key role in maintaining KIR2DL4⁺ populations²⁰. KIR3DL2, an inhibitory receptor crucial for NK education, trended higher in uninfected spleens and has been associated with better control of SIV replication and slower disease progression in RM. In contrast, activating receptors such as KIR3DH and KIR3DH5 showed broadly consistent expression across tissues and infection states, reflecting their wider distribution among systemic NK cells.

While these findings highlight the strengths of multiplexed, bead-based bDNA technology, some important considerations should be kept in mind when interpreting the results. Because the assay is probe-driven and limited to predefined transcripts, it cannot detect uncharacterized or novel gene products. However, this design ensures high specificity and is well-suited for structurally complex, well-annotated families such as KIRs. Additionally, the moderate sensitivity and dynamic range mean that very low-abundance transcripts might be below detection levels, but the advantage is consistent quantification of moderate-to-abundant targets across multiple tissues. Lastly, our analyses were performed on bulk tissue lysates, which do not identify the exact cellular sources of KIR expression. Although single-cell or allele-specific approaches could improve resolution in future studies, this research demonstrates that a carefully chosen, robust, and targeted assay can capture biologically meaningful differences across tissues and infection states.

Together, these findings highlight both methodological and biological insights from this study. Using a reproducible, targeted bDNA platform, we accurately measured KIR expression across tissues and confirmed compartmentalization in naïve macaques. We demonstrated that chronic SIV and SHIV infections weaken this specialization in a tissue- and infection-dependent manner. The combination of assay robustness and biological reproducibility emphasizes the reliability of these results. Importantly, the data suggest that chronic infection disrupts KIR⁺ NK and T-cell populations in secondary lymphoid tissues, which has implications for innate immune dysfunction in lentiviral disease.

Concluding remarks

This study highlights the effectiveness of bDNA technology in analyzing complex immunogenetic patterns in nonhuman primates. By allowing the sensitive, specific, and reproducible measurement of multiple KIR transcripts without enzymatic amplification, the assay overcomes the challenges associated with gene families that exhibit extensive sequence similarity. Its high-throughput capability enables the analysis of gene expression from limited tissue samples, allowing detection of subtle yet biologically significant changes across immune compartments. These methodological benefits establish bDNA as a valuable platform for immune gene profiling in translational infection models, where limited tissue access and genetic complexity often limit experimental design. The bDNA assay not only enhances the ability to measure gene expression accurately but also reveals the dynamics of immune dysfunction during persistent viral infection. These insights are particularly relevant to HIV and other chronic viral diseases, where monitoring KIR repertoires and maintaining cytotoxic activity in lymphoid tissues may be crucial for preserving an effective immune defense.

Data availability

Raw, pre-processed, and meta data have been deposited to FigShare (https://10.6084/m9.figshare.30005296). Additional study details are outlined within this manuscript.