Skip to main content
NCBI home page
Emerging Microbes & Infections logoLink to Emerging Microbes & Infections
. 2026 Mar 31;15(1):2648891. doi: 10.1080/22221751.2026.2648891

Development of a progressive HIV-1-Like infection model in northern pig-tailed macaques using stHIV-1sv/G53D

Tian-Zhang Song a,#, Qing Li a,b,#, Li-Rong Zhang a,c, Yan Hu a,b, Ying Lu a, Yong-Tang Zheng a,CONTACT
PMCID: PMC13040574  PMID: 41914615

ABSTRACT

Simian immunodeficiency virus (SIV) and simian-human immunodeficiency virus (SHIV) infections in nonhuman primates closely approximate human immunodeficiency virus (HIV)/acquired immunodeficiency syndrome (AIDS), but intrinsic viral differences limit translational relevance. A physiologically accurate HIV-1 model remains elusive. Here, we established a progressive HIV-1-like infection model using northern pig-tailed macaques (NPMs) infected with stHIV-1sv/G53D and monitored for 200 days. Results showed persistent viremia, progressive CD4+ T-cell depletion, and influenza-like symptoms, mirroring clinical features of HIV-1 infection. Infected macaques recapitulated T-cell exhaustion, innate immune activation, chronic immunosuppression, multi-organ biomarker alterations, and microbial dysbiosis. Transient and attenuated antiviral immunity may underlie this disease trajectory. In summary, this model serves as a complementary platform for validating direct antiviral strategies targeting HIV-1.

KEYWORDS: HIV, AIDS animal model, nonhuman primate, northern pig-tailed macaque, stHIV-1sv/G53D

Introduction

Human immunodeficiency virus (HIV) exhibits narrow species tropism and causes acquired immunodeficiency syndrome (AIDS) through the progressive depletion of CD4+ T lymphocytes. Despite its zoonotic origin, conventional animal models – including rodents, lagomorphs, and rhesus macaques – fail to support robust HIV replication or faithfully recapitulate the full spectrum of human disease pathogenesis [1]. Contemporary HIV/AIDS research primarily relies on two experimental platforms: humanized mouse models challenged with HIV-1 and nonhuman primates (NHPs) infected with simian immunodeficiency virus (SIV) or simian-human immunodeficiency virus (SHIV) constructs.

Humanized murine models, established through xenotransplantation of human hematopoietic components into immunodeficient recipients, permit HIV-1 infection and have enabled in vivo studies of viral entry, antiretroviral efficacy, cell-type specificity, and latency dynamics [2,3]. However, these models are constrained by incomplete immune cell reconstitution, impaired lymphoid organogenesis, limited infection persistence, and aberrant tissue distribution of immune cells [4], thereby failing to capture systemic immunopathogenesis of AIDS, including widespread lymphoid destruction and multi-organ dysfunction. In contrast, NHPs – particularly rhesus macaques infected with pathogenic strains of SIV or SHIV – exhibit hallmark features of HIV-1 pathogenesis, including sustained viremia, CD4+ T-cell depletion, chronic immune activation, and progression to AIDS-related conditions. These models have been pivotal in elucidating critical aspects of mucosal transmission, early immune evasion, and viral reservoir formation at the tissue level [5]. Nonetheless, extensive genetic divergence between SIV/SHIV and HIV-1, which share only 40%–50% nucleotide identity, produces fundamental biological disparities. These include Vpx-mediated degradation of SAMHD1, altered macrophage tropism, and divergent replication dynamics, which fundamentally constrain the utility of SIV/SHIV models for studying authentic HIV-1 targets. As a result, critical intervention strategies – including broadly neutralizing antibodies, multi-epitope vaccine platforms, and antiretroviral combination – cannot be reliably assessed in SIV/SHIV-infected macaques [6–10].

Thus, an ideal model enabling sustained HIV-1 replication within an immunocompetent host remains critically lacking. Among Old World monkeys, only pig-tailed macaques (Macaca nemestrina group) exhibit natural susceptibility to HIV-1. This group includes three recognized subspecies: Sunda (M. nemestrina), northern (NPM, M. leonina), and Mentawai (M. pagensis) [11]. Our previous work identified a unique TRIM5-cyclophilin A fusion gene in NPMs that abrogates TRIM5α-mediated restriction, thereby permitting HIV-1 entry [12]. However, challenge with HIV-1NL4-3 in NPMs produced only transient viremia in the absence of disease, implicating residual post-entry restriction mechanisms [13]. To overcome these barriers, a synthetic HIV-1 construct, stHIV-1sv, was engineered by integrating SHIVKB9-derived env and SIVmac239 vif, allowing evasion of APOBEC3-mediated restriction and enabling replication in NPM-derived peripheral blood mononuclear cells (PBMCs) [14–16]. Although stHIV-1sv infection yielded higher acute-phase viral loads and fewer G-to-A hypermutations in vivo, persistent replication was constrained by interferon (IFN)-induced antiviral responses [15]. Sequence analysis of viruses persisting in chronically infected NPMs identified a G53D substitution in Vpu that enhanced Tetherin antagonism [17]. Incorporation of this mutation into stHIV-1sv (stHIV-1sv/G53D) improved virion release, conferred partial resistance to IFN, and attenuated immune activation in PBMCs [17]. Collectively, these iterative adaptations enabled sequential escape from TRIM5α, APOBEC3, and Tetherin, establishing stHIV-1sv/G53D as a candidate for modelling progressive HIV-1-like infection in NPMs. This study inoculated three NPMs with stHIV-1sv/G53D and conducted longitudinal analyses of virological, immunological, clinical, hematological, and microbial parameters over a 200-day period.

Materials and methods

Animals and ethics statement

Three healthy male NPMs, aged 7–8 years, along with fresh PBMCs from healthy NPMs, were obtained from the Kunming Primate Research Center, Kunming Institute of Zoology (KIZ), Chinese Academy of Sciences (CAS), China. All animals were maintained under standard conditions (24 °C; 12-h light/dark cycle), with ad libitum access to food and water. All procedures involving animals were approved by the Ethics Committee of KIZ, CAS (approval number: IACUC-PE-2024-04-001). Following written informed consent from all participants, PBMCs from healthy human donors and individuals with chronic HIV-1 infection (not receiving combination antiretroviral therapy) were provided by the Yunnan Provincial AIDS Care Centre in Kunming, China. Research involving human-derived samples was approved by the Ethics Committee of the Yunnan Infectious Diseases Hospital (approval number: 2022004).

In vitro PBMC infection

Viral stocks of HIV-1R3A, stHIV-1sv, and stHIV-1sv/G53D were prepared and titrated in our laboratory as previously described [15,17]. Fresh PBMCs from three healthy human donors and three NPMs (2 × 106 cells) were stimulated for 72 h with 25 μg/mL concanavalin A (ConA) and 0.05 IU/mL interleukin-2 (IL-2). Activated cells were infected with each viral strain at a multiplicity of infection (MOI) of 0.05 for 6 h. Subsequently, unbound viral particles were removed through three sequential washes with phosphate-buffered saline (PBS). Cells were then cultured for 6 days in fresh medium containing 0.05 IU/mL IL-2. Supernatants and cells were harvested at 0, 3, and 6 days post-infection (dpi) for subsequent analysis. HIV-1 p24 antigen levels in culture supernatants were quantified using a commercial enzyme-linked immunosorbent assay (ELISA) kit as previously reported [15,17]. All work with the engineered viral strains was conducted under biosafety conditions and protocols commensurate with those used for HIV-1 and SIV.

In vivo infection with stHIV-1sv/G53D and sample collection

All animal experiments were conducted under anesthesia using Zoletil 50 (Virbac, France). PBMCs from each of the three NPMs (2 × 106 cells) were infected in vitro with stHIV-1sv/G53D at an MOI of 0.05. After 3 days, autologous infected PBMCs were administered intravenously, together with 1 × 107 50% tissue culture infectious dose (TCID50) cell-free viral particles. Longitudinal monitoring of clinical parameters (body weight and rectal temperature), hematological indices (routine blood tests and plasma biochemistry profiles), and biological specimens (peripheral blood samples and rectal swabs) was conducted at 0, 10, 14, 18, 21, 26, 33, 40, 54, 68, 89, 142, 168, and 198 dpi. Blood was collected by venipuncture, and PBMCs were isolated using Ficoll density gradient centrifugation. Routine blood tests and plasma biochemistry analysis were performed using a veterinary blood cell counter (BC-5000Vet, Mindray, China) and an automatic biochemistry analyzer (E-100, Siemens, USA), respectively.

Flow cytometry

Absolute counts of T-cell subsets were determined directly from fresh ethylenediaminetetraacetic acid (EDTA)-K2-anticoagulated whole blood, rather than cryopreserved PBMCs. Briefly, 50 μL of undiluted whole blood was stained in TruCount absolute counting tubes (BD Biosciences, USA) using a panel of cross-reactive anti-human monoclonal antibodies that have been validated for immunophenotyping in NPMs [15,16,18–20], including anti-CD3 BV421 (clone SP34-2), anti-CD8 PE-Cy7 (clone PRA-T8), anti-CD95 BV711 (clone DX2), anti-HLA-DR BV510 (clone G46-6), and anti-Ki67 PE (clone B56) from BD Biosciences (USA), and anti-CD4 Percp/Cy5.5 (clone OKT4), anti-CD28 APC (clone CD28.2), anti-PD-1 PE (clone EH12.2H7), and anti-CTLA4 BV605 (clone BNI3) from BioLegend (USA). After incubation for 15 min at room temperature in the dark, red blood cells were lysed and samples were fixed by adding 500 μL of FACS lysing solution (BD Biosciences). Flow cytometric data were acquired on an LSRFortessaTM Cell Analyzer (BD Biosciences, USA) and analyzed using FlowJo v10.5.3 (FlowJo LLC, USA).

Plasma cytokine and microbial translocation marker quantification

Plasma concentrations of granulocyte-macrophage colony-stimulating factor (GM-CSF), IFN-α, interleukin-10 (IL-10), interleukin-6 (IL-6), IFN-gamma-induced protein 10 kDa (IP-10, also known as CXCL10), interleukin-8 (IL-8, also known as CXCL8), C-C motif chemokine ligand 5 (CCL5), and tumour necrosis factor-alpha (TNF-α) were measured using an NHP XL Cytokine Luminex® Performance Premixed Kit (FCSTM21, R&D, USA) according to the manufacturer’s instructions. Lipopolysaccharide (LPS) and fatty acid-binding protein 2 (FABP2), markers of microbial translocation, were quantified using a Monkey Lipopolysaccharide ELISA Kit (ml996522 V, MLBIO, China) and Monkey Fatty Acid-Binding Protein ELISA Kit (ml9956932 V, MLBIO, China), respectively. Optical density values were measured at 450 nm using an ELISA microplate reader (800TS, BioTek, USA).

stHIV-1sv/G53D real-time RT-PCR assay

Plasma viral RNA was extracted using a High Pure Viral RNA Kit (7798200, Roche, UK) in accordance with the manufacturer’s protocols. Total RNA from PBMCs was extracted using TRIzol Reagent (9109, Thermo, USA), and genomic DNA was purified using a TIANamp Genomic DNA Kit (DP304, Tiangen, China). Nucleic acid concentrations were quantified using a NanoDrop 2000 spectrophotometer (Thermo Fisher, USA). Quantification of stHIV-1sv/G53D was performed using primers and probes targeting the conserved gag region of HIVNL4-3, as described in our previous study [13,16]. Viral RNA was analyzed by one-step reverse transcription quantitative PCR (RT-qPCR) using an Evo M-MLV One-Step RT-qPCR Kit (AG11715, AG, China) on the QuantStudioTM 5 Dx Real-Time PCR System (Thermo, USA). DNA samples were amplified using Premix Ex TaqTM (RR390A, TAKARA, Japan) under identical thermocycling conditions. The lower limit of detection (LLoD) for plasma viral RNA was 50 copies/μL [16]. For cell-associated viral DNA and RNA, the LLoD was determined to be 5 copies per reaction [13].

16S rRNA sequencing

Total genomic DNA was extracted from rectal swabs. The hypervariable V4 region of the bacterial 16S rRNA gene was amplified by PCR using universal primers 515F (5′-GTGCCAGCMGCCGCGGTAA-3′) and 806R (5′-GGACTACHVGGGTWTCTAAT-3′). Purified amplicons were subjected to paired-end sequencing (2 × 250 bp) on the Illumina NovaSeq 6000 platform (Illumina, USA). Clean reads were obtained by removing barcodes, primer sequences, and chimeric fragments, followed by sequence splicing and quality filtering. High-quality reads were clustered into operational taxonomic units (OTUs) at 97% identity using the UPARSE algorithm. Taxonomic classification was performed using the SILVA132 SSU rRNA reference database (https://www.arb-silva.de) through the Mothur pipeline (v1.43.0), applying an 80% bootstrap confidence threshold. All laboratory procedures, including DNA extraction, library preparation, and sequencing, were performed at Novogene (China). Alpha diversity indices (Chao1 richness and Shannon diversity) were calculated using QIIME2 (v2020.6) to assess within-sample microbial diversity. Beta diversity was evaluated based on weighted UniFrac distances using QIIME2. Predictive functional profiling of microbial communities was performed using Tax4Fun2 (v0.3.1) with default parameters, which maps 16S rRNA gene sequences to Kyoto Encyclopedia of Genes and Genomes (KEGG) orthologs through a nearest-neighbor approach.

Bulk RNA sequencing (RNA-seq)

RNA-seq library construction and sequencing were performed at Novogene (China). In brief, at least 1 mg of high-quality total RNA (RNA integrity number (RIN) > 8) was used for RNA sample preparation, with libraries generated using a NEBNext® UltraTM RNA Library Prep Kit for Illumina® (NEB, USA), following the manufacturer’s recommendations. Sequencing was conducted on the Illumina sequencing platform (NovaSeq 6000, Illumina, USA). Clean data were generated by removing reads containing adapters, poly-N (>10%), and >50% low-quality (Q ≤ 5) bases. Reference genomes for Macaca mulatta (Mmul_10, release-104) and Homo sapiens (GRCh3, hg38) were downloaded from Ensembl. The Macaca mulatta genome was selected due to its more mature and functionally curated annotation, an approach consistent with prior transcriptomic studies in NPMs [21]. Differential gene expression analysis was performed using DESeq2, with differentially expressed genes (DEGs) defined by fold change > 2 and P < 0.05. Functional enrichment analyses were performed using the ClusterProfiler package in R (v4.2.3). Gene Ontology Biological Process (GO_BP) and KEGG enrichment analyses were performed using org.Mmu.eg.db and org.Hs.eg.db annotations for monkey and human data, respectively. Significantly enriched gene sets were defined by adjusted P-value (p.adj) < 0.05. Immune-related DEGs were identified using the ImmPort database. Protein-protein interaction (PPI) networks of DEGs were established using the web-based visualization software STRING.

Expression of DEGs

Total RNA extracted from PBMCs was reverse transcribed using a PrimeScriptTM Fast RT Reagent Kit with gDNA Eraser (RR092A, TAKARA, Japan). The cDNA of 28 genes was amplified using TB Green® Premix Ex TaqTM II (RR820A, TAKARA, Japan), with normalization to RPL13A expression determined in parallel reactions. Relative gene expression was expressed as ΔCt, representing the difference between the Ct of RPL13A and that of the target gene.

Plasma protein profiling

Plasma proteins were extracted using DB lysis buffer (8 M Urea, 100 mM TEAB, pH 8.5), followed by reduction with 1 M dithiothreitol (DTT) for 1 h and alkylation with sufficient iodoacetamide (IAM) for 1 h in the dark. Proteins were digested with trypsin overnight and purified using a C18 desalting column. Peptides were analyzed by liquid chromatography-tandem mass spectrometry (LC-MS/MS) in data-independent acquisition (DIA) mode. Chromatographic separation was conducted on a Thermo UHPLC system equipped with a C18 analytical column (ES906, 150 μm × 15 cm, 2 μm), and MS acquisition was performed using a Thermo orbitrap astral mass spectrometer in DIA mode. Raw data were processed with DIA-NN software using the Macaca_mulatta_uniprot_2024_07_26 protein database. Protein quantification results were statistically analyzed using t-tests, with differentially expressed proteins (DEPs) defined as those with significant quantitative differences between stHIV-1sv/G53D-infected and non-infected groups (up-regulated: P < 0.05, fold change > 1.2; down-regulated: P < 0.05, fold change < 0.83). The GO_BP and KEGG databases were used to analyze functional enrichment of protein families and pathways (p.adj < 0.05).

Results

Robust viral replication and attenuated antiviral responses in stHIV-1sv/G53D-infected PBMCs from NPMs

Two macaque-adapted viral derivatives were previously constructed based on HIV-1NL4-3 (Figure 1(A)): stHIV-1sv, encoding an HIV-1 env gene from SHIVKB9 and a vif gene from SIVmac239, and stHIV-1sv/G53D, harbouring the Vpu-G53D mutation that enhances Tetherin antagonism. Both constructs retain dual tropism for CCR5 and CXCR4. In the present study, viral replication dynamics were evaluated by quantifying p24 antigen levels in the supernatants of HIV-1-challenged PBMCs derived from either humans (hPBMCs) or NPMs (mPBMCs). At 3 and 6 dpi, stHIV-1sv/G53D-infected mPBMCs exhibited markedly elevated p24 concentrations, reaching levels comparable to those observed in HIV-1R3A-infected hPBMCs (Figure 1(B)). In contrast, stHIV-1sv-infected mPBMCs showed substantially lower p24 expression, with stHIV-1sv/G53D achieving approximately 2.5- to 4-fold higher replication. These findings indicate that the G53D substitution in Vpu effectively counteracts restriction mechanisms in macaque cells, enabling efficient viral propagation in this otherwise nonpermissive host system.

Figure 1.

Six diagrams and charts comparing HIV-1NL4-3, stHIV-1sv, stHIV-1sv/G53D structures, p24 line graph, DEG bar charts, GO dot plots, and antiviral heatmap. The figure shows six related visuals summarizing viral constructs, replication, and gene expression in infected peripheral blood mononuclear cells. The first visual is a schematic diagram of three linear HIV genomes labeled HIV-1NL4-3, stHIV-1sv, and stHIV-1sv/G53D, each drawn from left to right with long terminal repeat, Gag, Pol, Vif, Vpr, Vpu, Env, Rev, Tat, and Nef boxes, and a legend indicating that some Env and Vif segments derive from SIVmac239, SHIVKB9, or the G53D variant. The second visual is a grouped dot and bar chart with the x axis listing stHIV-1sv, stHIV-1sv/G53D, and HIV-1R3A, and the y axis labeled p24 picograms per milliliter from 0 to 7000 at intervals of 1000. Dots representing day 3 clustered closely around the corresponding shaded bars for day 6, and stHIV-1sv/G53D and HIV-1R3A reached higher values than stHIV-1sv. All data are approximate. The third visual is two vertical bar charts labeled 3 days versus 0 days and 6 days versus 0 days, with the x axis listing stHIV-1sv, stHIV-1sv/G53D, and HIV-1R3A and the y axis labeled count of differentially expressed genes, ranging from negative to positive, where upper bars indicate up-regulated genes and lower bars indicate down-regulated genes. All data are approximate. The fourth visual consists of two dot plots showing Gene Ontology Biological Process terms for down-regulated genes at 3 days versus 0 days in stHIV-1sv/G53D and HIV-1R3A. The x axis is ratio from 0 to about 0.25, the y axis lists immune and defense process descriptions, dot size encodes gene count, and dot shading encodes negative log base 10 adjusted p value. All data are approximate. The fifth visual contains two similar dot plots for up-regulated Biological Process terms at 6 days versus 3 days, with immune and antiviral responses highlighted. All data are approximate. The sixth visual is a heatmap matrix with rows of antiviral genes, including PIM2, PRF1, MX1, OAS1, ISG15, IRF7, TRIM5, and others, and columns for stHIV-1sv/G53D and HIV-1R3A at 6 days, colored to show Z score expression changes from about negative 2 to positive 2 relative to 0 days, with some genes showing higher intensity in macaque adapted virus conditions. All data are approximate.

Open in a new tab

Viral replication dynamics and transcriptomic profiles in HIV-1-infected peripheral blood mononuclear cells (PBMCs). (A) Schematic comparison of genomic structures for stHIV-1sv, stHIV-1sv/G53D, and HIV-1NL4-3. (B) Concentrations of p24 antigen in culture supernatants of HIV-1-infected PBMCs. At a multiplicity of infection (MOI) of 0.05, Human PBMCs (hPBMCs) were infected with HIV-1R3A, and northern pig-tailed macaque PBMCs (mPBMCs) were infected with stHIV-1sv or stHIV-1sv/G53D. p24 levels were quantified by ELISA at 3 and 6 days post-infection (dpi). n = 3 per group per time point. (C) Number of differentially expressed genes (DEGs; fold change > 2, P < 0.05) identified by bulk RNA sequencing (RNA-seq) at 3 and 6 dpi compared to 0 dpi. (D) Top 10 down-regulated Gene Ontology Biological Process (GO_BP) terms when comparing 3 dpi to 0 dpi. Terms are ranked by descending adjusted P-value (p.adj). Blue terms highlight immune- and defense-related pathways. (E) Top 10 up-regulated GO_BP terms when comparing 6 dpi to 3 dpi. Terms are ranked by descending p.adj. Red terms highlight immune- and defense-related pathways. (F) Heatmap showing the expression (Z-score) of selected antiviral genes in HIV-1R3A-infected hPBMCs and stHIV-1sv/G53D-infected mPBMCs at 6 dpi relative to 0 dpi.

RNA-seq was performed to assess HIV-1-induced transcriptional changes in PBMCs. Relative to the pre-infection baseline (0 dpi), all infection systems at 3 dpi exhibited a predominance of down-regulated over up-regulated DEGs (Figure 1(C)). Notably, HIV-1R3A-infected hPBMCs generated 957 up-regulated DEGs at 6 dpi (vs. 0 dpi), representing a 3-fold increase relative to either stHIV-1sv- or stHIV-1sv/G53D-infected mPBMCs (Figure 1(C)). GO_BP enrichment analysis revealed that immune- and defense-related pathways were significantly down-regulated in both HIV-1R3A-infected hPBMCs and stHIV-1sv/G53D-infected mPBMCs when comparing 3 dpi to 0 dpi (Figure 1(D) and Supplementary Figure 1(A and B)), whereas no significant pathway changes were observed in stHIV-1sv-infected mPBMCs (Supplementary Figure 1(C and D)). Subsequently, PPI networks constructed from immune-related DEGs at 3 dpi (vs. 0 dpi) demonstrated coordinated suppression of innate immune regulators, including Toll-like receptors (TLR1, TLR2, TLR4, and TLR8) and cytokines (CXCL9 and TNFSF13B) in both HIV-1R3A-infected hPBMCs (Supplementary Figure 1(E)) and stHIV-1sv/G53D-infected mPBMCs (Supplementary Figure 1(F)), suggesting convergent immunological changes between the two systems. Notably, STAT1 and MX1 were down-regulated in the stHIV-1sv/G53D system, while IFNG was significantly up-regulated in HIV-1R3A-infected hPBMCs, suggesting divergent antiviral responses between the two systems.

To further delineate temporal dynamics, transcriptomic changes from 3 to 6 dpi were compared across all three infection models. stHIV-1sv-infected mPBMCs at 6 dpi (vs. 3 dpi) exhibited down-regulation of immune- and defense-related pathways, mirroring the profile observed in stHIV-1sv/G53D-infected mPBMCs at 3 dpi (vs. 0 dpi) and suggesting that the G53D variant accelerated immunomodulation in mPBMCs compared to the parental strain (Supplementary Figure 1(G and H)). Furthermore, both HIV-1R3A-infected hPBMCs and stHIV-1sv/G53D-infected mPBMCs showed up-regulation of immune- and defense-related pathways from 3 to 6 dpi (Figure 1(E) and Supplementary Figure 1(I and J)). However, their transcriptional profiles diverged significantly. In HIV-1R3A-infected hPBMCs, this up-regulation was characterized by strong activation of antiviral responses, with core antiviral genes (e.g. MX1, MX2, ISG15, DHX58, and OASL) forming central nodes in the up-regulated DEG network (Supplementary Figure 1(K)). In contrast, stHIV-1sv/G53D-infected mPBMCs exhibited primary enrichment of natural killer (NK) cell- and leukocyte-related pathways alongside sustained immune suppression (Supplementary Figure 1(J)), and failed to induce the robust antiviral gene programme seen in HIV-1R3A-infected hPBMCs (Figure 1(F) and Supplementary Figure 1(L)). Together, these findings indicate that the stHIV-1sv/G53D variant sustains high-level replication in mPBMCs by actively suppressing antiviral transcriptional programmes.

Persistent viremia and progressive T-cell depletion in stHIV-1sv/G53D-infected NPMs

In this study, a novel HIV-1-like infection model was established in three NPMs (#16223, #16229, and #17209) through infection with stHIV-1sv/G53D (Figure 2(A)), followed by longitudinal monitoring of virological and immunological dynamics over 200 days. Plasma viral loads peaked during the acute phase (2–3 weeks post-infection) at 0.4–2.6 × 106 copies/mL, with a gradual decline to 103–104 copies/mL over the next 8 weeks, ultimately stabilizing within this range for 100 days (Figure 2(B)). Quantification of viral gag DNA and RNA in mPBMCs confirmed sustained viral integration and replication, with both markers maintained above 104 copies/μg (DNA: 104–105; RNA: 104–106) for over 170 days (Figure 2(B)). It is noteworthy that in two animals (#16223 and #17209), both viral DNA and RNA in PBMCs demonstrated a decreasing trend from 177 to 198 dpi, indicating a potential decrease in the cell-associated viral reservoir during this late phase. Collectively, these data demonstrate that NPMs are susceptible to productive infection by stHIV-1sv/G53D, sustaining both persistent viremia and a detectable viral reservoir for at least 200 days.

Figure 2.

Five diagrams and seven line graphs tracking stHIV-1sv G53D infection in Macaca leonina NPMs, showing viral loads and T cell changes over 200 days. The figure shows an experimental diagram and multiple graphs summarizing stHIV-1sv G53D infection in Macaca leonina northern pig tailed macaques numbered 16223, 16229, and 17209. The first visual is a schematic diagram outlining blood collection from an animal, peripheral blood mononuclear cell (PBMC) isolation, ex vivo PBMC infection, followed by intravenous injection of the infected cells and stHIV-1sv/G53D into the animal. The second group contains three line graphs of viral measurements over 0 to 200 days post infection. The x axes are labeled days post infection. The y axes show plasma viral ribonucleic acid in logarithm 10 copies per milliliter, viral deoxyribonucleic acid in peripheral blood mononuclear cells in copies per microgram total ribonucleic acid, and viral ribonucleic acid in peripheral blood mononuclear cells in copies per microgram total ribonucleic acid. All three animals are plotted separately and show an early rise followed by partial decline and then persistent values at later time points. The third group has four line graphs showing total T cell counts, cluster of differentiation 4 positive T cell counts, cluster of differentiation 8 positive T cell counts, and the cluster of differentiation 4 to cluster of differentiation 8 ratio, all expressed as fold change to day 0 over 0 to 200 days. The traces fluctuate around a horizontal reference line at 1.0. The fourth visual set is a group of bar charts showing fold change to day 0 for markers cytotoxic T lymphocyte associated protein 4, programmed cell death protein 1, human leukocyte antigen DR, and Ki67 on cluster of differentiation 4 positive and cluster of differentiation 8 positive T cells at 0, 21, and 142 days post infection. The final group contains two bar charts showing frequencies of naive, central memory, and effector memory subsets within cluster of differentiation 4 positive and cluster of differentiation 8 positive T cells as fold change to day 0 at the same three time points. All data are approximate.

Open in a new tab

Viral persistence and T-cell immunopathology in stHIV-1sv/G53D-infected northern pig-tailed macaques (NPMs). (A) Schematic of stHIV-1sv/G53D infection protocol in NPMs. (B) Longitudinal virological monitoring. Plasma viral RNA (red), PBMC-associated viral DNA (blue), and PBMC-associated viral RNA (brown) were quantified by RT-PCR. (C) Dynamics of T-cell subsets and CD4/CD8 ratio (black) during stHIV-1sv/G53D infection. Absolute counts of total T cells (red), CD4+ T cells (blue), and CD8+ T cells (brown) were determined by flow cytometry. Data are presented as fold change relative to baseline (0 dpi). (D) Expression of immune markers in T-cell subsets. Levels of CTLA-4, PD-1, HLA-DR, and Ki67 in CD4+ and CD8+ T cells were quantified by flow cytometry and are shown as fold change relative to 0 dpi. (E) T-cell subset distribution. The frequencies of naïve (TN), central memory (TCM), and effector memory (TEM) subsets within CD4+ and CD8+ T cells were determined by flow cytometry and are presented as fold change relative to 0 dpi.

To assess immunopathological progression, peripheral T-cell subsets were analyzed longitudinally (Figure 2(C)). Total T-cell counts inconsistently fluctuated during the acute phase (temporary increase in #16223 and #17209), followed by a sustained decline to 60%–90% of baseline, plateauing at reduced levels for 150 days. CD4+ T cells exhibited a more pronounced and rapid depletion in all NPMs, declining to 40%–80% of pre-infection values within 4 weeks and remaining suppressed through 150 dpi. CD8+ T cells transiently expanded at 14 dpi before normalizing to baseline by 200 dpi, consistent with acute-phase immune activation observed in early HIV-1 infection. Notably, the CD4/CD8 ratio declined progressively between 40 and 200 dpi in all animals, reflecting a key clinical indicator of disease progression.

To delineate the dynamics of immune dysregulation, peripheral T cells from NPMs at 0, 21, and 142 dpi were profiled for exhaustion, activation, proliferation, and memory phenotypes. Analysis of T-cell exhaustion revealed distinct kinetics between subsets: CTLA-4 and PD-1 expression on CD4+ T cells was significantly elevated at 21 dpi and further increased at 142 dpi, whereas CD8+ T cells exhibited heterogeneous changes at 21 dpi that resolved into a uniformly high exhaustion marker expression at 142 dpi (Figure 2(D)). Acute T-cell activation, measured by HLA-DR expression, was markedly and concordantly upregulated on both CD4+ and CD8+ T cells at 21 dpi but declined substantially at 142 dpi (Figure 2(D)). In contrast, the proliferative marker Ki67 showed considerable inter-individual variation without a consistent directional trend across the three animals at either time point; therefore, the biological significance of the observed mean increase remains unclear (Figure 2(D)). Memory subset analysis demonstrated a consistent and progressive loss of effector memory T cells (TEM; CD28-CD95+) in both lineages at 142 dpi, while naïve (TN; CD28 + CD95-) and central memory (TCM; CD28 + CD95+) frequencies showed an overall increasing trend at 142 dpi (Figure 2(E)). Collectively, these immunophenotypic changes – progressive exhaustion, acute activation, and selective TEM attrition – recapitulate key features of HIV-1-induced T-cell immunopathology in the stHIV-1sv/G53D NPM model.

stHIV-1sv/G53D infection induces diverse pathophysiological alterations in NPMs

All infected macaques exhibited acute weight loss within 40 dpi and persistent low-grade fever beyond 150 dpi (Figure 3(A)). Systemic metabolic dysregulation emerged at 142 dpi, characterized by reductions in plasma albumin and glucose and elevations in globulin and total cholesterol, in contrast to the variable trends at 21 dpi (Figure 3(B)). Hepatic enzymes (aspartate aminotransferase [AST] and alanine aminotransferase [ALT]), renal markers (blood urea nitrogen [BUN] and creatinine [CRE]), and cardiac/muscle injury indicators (lactate dehydrogenase [LDH] and creatine kinase [CK]) showed relatively consistent elevations, despite notable inter-individual variability for some (e.g. AST, CK) (Figure 3(B)). For some markers (e.g. CK, BUN, CRE), absolute values remained within the normative range for macaques. Totally, this multi-system pattern of biomarker increases mirrors the early, often subclinical, organ perturbations observed in PLWH.

Figure 3.

Seven charts showing stHIV-1sv/G53D-infected NPM body measures, blood markers, cytokines, GSEA plots, microbiota diversity, and pathways. The figure shows seven graphics summarizing clinical, blood, and microbiome measurements in stHIV-1sv/G53D-infected northern pigtailed macaques over time. A large two-part line graph tracks body weight (left y-axis) and rectal temperature (right y-axis) versus days post infection, with values plotted for animals 16223, 16229, and 17209. Both y-axes show fold change or difference relative to day 0. Body weight initially decreased then increased, whereas temperature increased over the course of 150 days. A grouped dot and bar chart displays fold change in plasma albumin, globulin, albumin to globulin ratio, glucose, cholesterol, triglyceride, aspartate aminotransferase, alanine aminotransferase, blood urea nitrogen, creatinine, lactate dehydrogenase, and creatine kinase at 0, 21, and 142 days, with several markers increased at 142 days; all data are approximate. A two-part line graph shows fold change in red blood cells and monocytes in peripheral blood versus days post infection. Red blood cells showed minimal variation, whereas monocytes initially increased then decreased. A grouped dot and bar chart shows plasma cytokines granulocyte macrophage colony stimulating factor, interferon gamma, interleukin 4, interleukin 10, interleukin 12, interleukin 17, interleukin 8, C C motif chemokine ligand 5, tumor necrosis factor alpha, and microbial markers fatty acid binding protein 2 and lipopolysaccharide at 0, 21, and 142 days, with several cytokines and lipopolysaccharide elevated at 21 or 142 days; all data are approximate. Four gene set enrichment analysis line plots compare day 21 or day 142 with day 0 for complement and coagulation cascades, cholesterol metabolism, biological process immune response, and biological process antigen processing and presentation, each including a running enrichment score curve, ranked genes, and a normalized enrichment score value; all data are approximate. A two-part line graph shows rectal microbiota alpha diversity by Chao 1 and Shannon indices versus time, with peaks around 21 days then partial decline; all data are approximate. A heatmap chart displays predicted microbial functional pathways on the y axis and days post infection on the x axis at 0, 21, and 142 days, with a color scale representing Z score from enrichment to depletion; all data are approximate.

Open in a new tab

Clinical symptoms, hematological changes, and microbial signatures in stHIV-1sv/G53D-infected NPMs. (A) Longitudinal changes in body weight and rectal temperature. Body weight (red) is presented as fold change relative to 0 dpi; rectal temperature is presented as the difference from 0 dpi. (B) Plasma biochemical parameters indicative of metabolic (left) and organ function (right). Measurements were performed using an automatic biochemistry analyzer. Data are presented as fold change relative to 0 dpi. (C) Hematological profiles from complete blood count analysis. Measurements were performed using a veterinary blood cell counter. Data are presented as fold change relative to 0 dpi. (D) Plasma concentrations of cytokines and microbial translocation markers. Cytokines (left) and microbial translocation markers (right) were measured by Luminex assay and ELISA, respectively. Data are presented as fold change relative to 0 dpi. (E) Gene Set Enrichment Analysis (GSEA) of plasma proteomic data at 21 and 142 dpi compared to 0 dpi. (F) Alpha diversity of the rectal microbiota, assessed by Chao1 (richness) and Shannon (diversity) indices. (G) Heatmap of predicted microbial functional pathways. The colour scale represents Z-scores of relative abundance, where red to blue indicates enrichment to depletion. Only pathways with a significant change (P < 0.05) at any time point are displayed.

Peripheral immunological alterations extended beyond CD4+ T-cell depletion (Figure 3(C and D)). Red blood cell (RBC) counts remained stable, whereas acute infection induced monocytosis accompanied by a proinflammatory cytokine profile. Marked and consistent increases were observed at 21 dpi for IFN-α, IL-10, IL-8, CCL5, and TNF-α. GM-CSF, IL-6, and IP-10 showed a trend of elevation at this time point, though with considerable inter-individual variability, collectively indicating early immune hyperactivation. By the chronic stage (142 dpi), the immune profile transitioned to a more immunosuppressive state, characterized by declining monocyte counts and diminished cytokine expression. Levels of IFN-α, IL-10, IL-6, IP-10, IL-8, CCL5, and TNF-α were consistently reduced. While mean concentrations of GM-CSF showed a downward trend, these changes exhibited variability across animals. Collectively, these dynamic changes delineate a biphasic immunopathological trajectory: acute hyperinflammation followed by chronic immunosuppression. Additionally, plasma levels of LPS and FABP2 were elevated at both 21 and 142 dpi, reflecting compromised intestinal barrier integrity and sustained microbial translocation, key drivers of HIV-1-associated immune dysregulation (Figure 3(D)).

To obtain a multi-omics view of systemic host response and local microbial ecology, we performed integrated profiling of plasma proteomes and rectal microbiomes at 0, 21, and 142 dpi. Although relatively few DEPs were detected between the infected and uninfected groups (Supplementary Figure 2(A)), Gene Set Enrichment Analysis (GSEA) revealed dynamic changes in host biological pathways over time. At 21 dpi, significant enrichment or suppression was detected in seven GO_BP terms and 13 KEGG pathways compared to 0 dpi. By 142 dpi, two GO_BP terms and nine KEGG pathways remained differentially regulated (Supplementary Figure 2(A and B)). Early infection was associated with down-regulation of complement-related functions and cholesterol metabolism (Figure 3(E)). By 142 dpi, functional down-regulation extended to antigen presentation and broader immune response pathways, consistent with declining cytokine levels and progressive lymphocyte dysfunction.

Microbiome analysis revealed elevated alpha diversity (Chao1 and Shannon indices) in infected individuals (Figure 3(F)), and clear beta diversity separation based on weighted UniFrac distance (Supplementary Figure 2(C)), indicating significant gut microbial restructuring post-infection. Taxonomic profiling showed a substantial post-infection shift in the relative abundance of dominant phyla, with Firmicutes levels decreasing and Bacteroidota levels increasing (Supplementary Figure 2(D)). At the genus level, notable alterations included Ligilactobacillus depletion and Prevotella enrichment. Functional predictions using Tax4Fun2 revealed microbial enrichment in pathways associated with aging and immune regulation, alongside suppression of core metabolic functions including lipid, amino acid, and carbohydrate metabolism. In contrast, nucleotide, energy, vitamin, and cofactor metabolism pathways were up-regulated (Figure 3(G)). These findings indicate that stHIV-1sv/G53D infection induces profound remodelling of the gut microbiome, potentially contributing to sustained immune disruption and intestinal metabolic imbalance.

Attenuated antiviral immunity in stHIV-1sv/G53D-infected NPMs

To compare host responses between PLWH and the stHIV-1sv/G53D macaque model, transcriptomic profiling was performed using PBMCs from untreated PLWH and infected NPMs. GO_BP enrichment analysis of DEGs revealed that chronic HIV-1 infection in humans was characterized by marked up-regulation of antiviral pathways alongside suppression of immune-related pathways (Figure 4(A)). A similar but less pronounced pattern emerged during acute stHIV-1sv/G53D infection in NPMs, which exhibited modest activation of antiviral programmes and concurrent downregulation of immune-related pathways (Figure 4(B)), partially reflecting the transcriptional landscape observed in PLWH. However, in the chronic phase of stHIV-1sv/G53D infection, immunosuppressive signalling predominated without significant enrichment of antiviral pathways relative to pre-infection levels (Figure 4(C)).

Figure 4.

Eight charts showing GO biological process enrichment, immune gene counts, protein interaction networks, and qPCR results in HIV studies. The figure shows eight visuals that compare immune related transcriptional changes in untreated people living with HIV and stHIV-1sv/G53D infected northern pig tailed macaques. Three dot charts display Gene Ontology biological process enrichment. Each chart has the x axis labeled Ratio from 0 to about 0.25 with ticks at regular intervals and the y axis listing biological process descriptions. Dot size represents gene counts from about 4 to about 60 and a vertical color scale shows negative log 10 adjusted p values from about 1.0 to about 15. The first chart contrasts human up regulated pathways with human down regulated pathways. The second chart compares macaque up regulated pathways at 21 days post infection with down regulated pathways at the same time. The third chart compares macaque up regulated pathways at 142 days post infection with down regulated pathways at 142 days. All data are approximate. A vertical bar chart summarizes counts of immune related differentially expressed genes for humans and macaques at 21 and 142 days post infection, with separate bars for up and down regulation on the y axis. Three circular protein protein interaction network diagrams show clusters of immune genes for human up, human down, and macaque up signatures, with nodes labeled by gene symbols and edges connecting interacting genes. A fourth circular network diagram displays macaque down regulated genes at 21 and 142 days. Finally, grouped bar charts present quantitative polymerase chain reaction delta Ct values for selected hub genes in humans and macaques at 0, 21, and 142 days, using the y axis labeled delta Ct. All data are approximate.

Open in a new tab

Transcriptomic profiling of PBMCs from untreated PLWH and stHIV-1sv/G53D-infected NPMs. (A–C) GO_BP enrichment analysis. Shown are the top 10 up- (red) and down-regulated (blue) terms in (A) PLWH vs. healthy donors, (B) NPMs at 21 dpi vs. 0 dpi, and (C) NPMs at 142 dpi vs. 0 dpi. Terms coloured in red/blue are immune- and defense-related pathways and in gray are not significant (p.adj > 0.05). Ranking is by descending p.adj. (D) Numbers of up- and down-regulated immune-related DEGs (filtered via the ImmPort database). (E–G) PPI networks of key immune-related DEGs from (E) PLWH vs. healthy, (F) NPMs 21 dpi vs. 0 dpi, and (G) NPMs 142 dpi vs. 0 dpi. Nodes represent genes, with colour indicating expression direction (red: up-regulated; blue: down-regulated) and colour intensity reflecting node degree (darker = higher connectivity/importance). (H) qPCR validation of selected hub genes. Expression levels were normalized to RPL13A and are presented as ΔCt values. n = 3 per group per time point.

Immune-associated DEGs were further filtered using the ImmPort database. During acute infection, NPMs displayed a 50% reduction in up-regulated immune-related DEGs compared to PLWH, with no up-regulated immune-related DEGs detected during the chronic stage (Figure 4(D)). Subsequent PPI network analysis revealed that key antiviral effectors, such as STAT1, DDX58, MX1, and ISG15, occupied central regulatory positions driving immune activation in PLWH (Figure 4(E)). In contrast, NPMs infected with stHIV-1sv/G53D exhibited only a limited subset of these antiviral gene signatures (ISG15, IFIT1, and IFIT5) during acute infection (Figure 4(F)). Additionally, expression of several critical immune mediators, including proinflammatory cytokines (IL1B, TNF, and IFNG) and chemokines (CCL3, CXCL8, and CXCL10), was markedly down-regulated in NPMs, implying a more profound immunosuppressive phenotype compared to PLWH (Figure 4(G)). Finally, qPCR analysis of selected DEGs validated these transcriptomic patterns (Figure 4(H), Supplementary Table 1). Collectively, these findings demonstrate that stHIV-1sv/G53D infection elicits a transient and insufficient antiviral response in NPMs, likely contributing to persistent viral replication and progressive immune system impairment in vivo.

Discussion

This study demonstrated that infection with stHIV-1sv/G53D induced sustained plasma viremia and progressive systemic depletion of CD4+ T cells in NPMs. In addition, the model partly recapitulated key clinical and immunopathological features of human HIV-1 infection, including febrile responses, weight loss, T-cell exhaustion, transient innate immune hyperactivation, altered biochemical indices across multiple organ systems, and perturbations in microbial composition.

Effective modelling of AIDS pathogenesis requires both persistent viral replication and gradual CD4+ T-cell attrition – core features of human HIV-1 infection. Comparative analysis of successive viral constructs in NPMs revealed stepwise enhancement of pathogenicity. Initial infection with HIV-1NL4-3 generated transient viremia peaking at 104–105 copies/mL, which resolved by week 6 without CD4+ T-cell decline [13]. Introduction of the stHIV-1sv strain, engineered to evade APOBEC3 restriction, yielded increased acute viremia (105–106 copies/mL) and detectable viral RNA through week 8 (102–103 copies/mL), yet still failed to elicit CD4+ T-cell depletion [15]. In contrast, infection with the stHIV-1sv/G53D variant, harbouring an additional Vpu-G53D mutation that confers antagonism of Tetherin, resulted in markedly elevated acute-phase viremia (106–107 copies/mL), a stable chronic viral setpoint (103–104 copies/mL), progressive loss of CD4+ T cells, and overt clinical signs. This trajectory closely approximates that observed in untreated PLWH, encompassing triphasic viral dynamics (acute spike, rapid decline, and stabilization), biphasic CD4+ T-cell depletion (early drop followed by partial stabilization), and AIDS-related clinical symptoms including weight loss and fever [18,22–24].

Compared with our previously established pathogenic SIVmac239 infection model in NPMs, stHIV-1sv/G53D infection elicited comparable peripheral immune signatures during early and chronic stages. Both models exhibited acute-phase upregulation of HLA-DR expression on T cells, indicative of early systemic immune activation, and sustained elevation of PD-1 expression on CD4+ and CD8+ T cells across acute and chronic stages, reflecting progressive T-cell exhaustion [18]. These shared immunophenotypic trajectories confirm the capacity of the stHIV-1sv/G53D model to replicate essential features of lentivirus-induced immune dysfunction. However, divergent immunological profiles emerged during late-stage infection. By 142 dpi, stHIV-1sv/G53D-exposed macaques showed reduced HLA-DR expression across both CD4+ and CD8+ T-cell subsets and broad attenuation of immune activation markers, in contrast to the persistent inflammatory profile typically observed in SIVmac239-infected macaques and PLWH. This immunological dampening may reflect a distinct viral adaptation strategy conferred by the engineered Vpu-G53D mutation, which has been implicated in immunosuppressive signalling. These findings raise the possibility that stHIV-1sv/G53D promotes immune quiescence to facilitate long-term persistence while evading host immune surveillance.

The capacity of stHIV-1sv/G53D to attenuate chronic immune activation appears to arise from functional adaptation of the Vpu protein. Infection with HIV-1NL4-3 in NPMs triggers a robust IFN response that potently suppresses viral replication [25]. IFN-mediated antiviral activity primarily operates through host restriction factors, including TRIM5α, APOBEC3G, and Tetherin. To counteract these defenses, SIV and HIV-1 encode accessory proteins, such as Vif, Vpu, and Nef, to antagonize restriction factors in a species-specific manner, a property that also constrains cross-species transmission [26]. However, adaptive mutations in these proteins can expand antagonistic capacity to new hosts and thereby enable zoonotic spread [27]. Vpu represents a paradigmatic example. HIV-1 Vpu efficiently induces degradation of human Tetherin through proteasomal or lysosomal pathways [28–30], but generally fails to antagonize Tetherin orthologs from most NHPs, explaining the inability of wild-type HIV-1 to establish productive infection in monkeys, aside from rare exceptions such as strain DH12 [31]. In contrast, Vpu encoded by SIVcpz effectively antagonizes human Tetherin and played a critical role in the cross-species transmission events that generated the pandemic HIV-1 group M lineage [32]. Beyond facilitating virion release, Vpu also inhibits host NF-κB signalling, thereby limiting pro-inflammatory cytokine production and promoting viral replication [33]. Prior in vitro analyses demonstrated that the G53D mutation in Vpu markedly reduces production of key pro-inflammatory mediators, including type I and type II IFN, TNF-α, IL-1β, and IP-10 [17]. These cytokines, particularly IFN, serve as major upstream activators of the JAK-STAT signalling pathway. Consistent with this mechanism, stHIV-1sv/G53D infection failed to induce canonical JAK-STAT pathway genes, such as STAT1, MX1, and ISG15, in infected mPBMCs in vitro. This suppressive phenotype persisted in vivo. Acute infection was associated with a transient increase in plasma cytokines, including type I IFN, accompanied by only modest induction of a limited subset of antiviral genes, such as ISG15 and IFIT5 in mPBMCs. During the chronic phase, plasma cytokine concentrations declined in parallel with the absence of significant enrichment of antiviral transcriptional programmes. Together, these findings indicate that the engineered Vpu-G53D substitution confers a dual mechanistic advantage that underpins the pathogenic phenotype of stHIV-1sv/G53D. The mutation enhanced proteasome-mediated degradation of macaque Tetherin, thereby facilitating efficient virion release [17], while simultaneously suppressing pro-inflammatory cytokine production, including IFN and TNF-α. Attenuation of these upstream signals inhibited activation of downstream antiviral pathways, such as JAK-STAT signalling, establishing an immunologically permissive state conducive to persistent viral replication.

Several limitations inherent to this proof-of-concept study delineate the interpretive boundaries of the current findings and define priorities for subsequent work. First, the small cohort size (n = 3), although typical for exploratory NHP studies, limits statistical power and precludes rigorous quantification of inter-individual variability in key disease parameters such as viral set points, CD4+ T-cell depletion kinetics, and systemic injury severity. Accordingly, while the reproducibility of core disease features supports model feasibility, the breadth of biological heterogeneity remains unresolved. Second, the observation window of approximately 200 days, although sufficient to establish chronic infection, did not encompass advanced stages of HIV-1 disease marked by severe immunodeficiency and opportunistic infections. Third, evidence for multi-organ pathology is currently indirect, relying on circulating biochemical markers in the absence of histopathological validation, thus precluding confirmation of the precise nature and extent of tissue injury. Fourth, infection was established exclusively through intravenous challenge, a strategy that ensures reliable systemic infection but fails to recapitulate the dominant mucosal transmission routes responsible for most HIV-1 acquisition events in humans, thereby limiting insight into early mucosal immune responses and portal-of-entry dynamics. Fifth, therapeutic responsiveness was not evaluated, as combination antiretroviral therapy was not administered, leaving critical parameters such as viral suppression, reservoir stability, immune reconstitution, and post-treatment rebound uncharacterized. Sixth, transcriptomic analyses relied on the closely related Macaca mulatta reference genome, which may introduce mapping biases and obscure host-specific genetic determinants unique to NPMs. These limitations delineate key priorities for advancing model development and translational applicability, including expanding cohort size to quantify biological variation and ensure reproducibility; introducing rectal and vaginal mucosal challenge to establish physiologically relevant transmission and enable investigation of mucosal immunity; extending longitudinal observation to capture late-stage disease progression; implementing antiretroviral intervention to evaluate therapeutic responsiveness, reservoir dynamics, and immune reconstitution; and conducting terminal analyses with comprehensive tissue sampling to map viral reservoirs, validate biomarkers, and confirm histopathological correlates. Systematic resolution of these priorities will enable transition from proof-of-concept to a rigorously validated, high-fidelity platform for mechanistic and therapeutic HIV-1 research.

Despite current limitations, the stHIV-1sv/G53D-infected NPM model establishes a valuable translational platform for HIV/AIDS research, with distinct strengths in two key domains. First, the viral construct is predominantly HIV-1-derived and is evaluated in a NHP species with pharmacokinetic, immunological, and physiological characteristics that more closely approximate human biology than rodent systems. This model thus enables rigorous in vivo assessment of therapeutic agents targeting HIV-1-specific elements, including small molecules and biologics that act on viral proteins such as Vpu or on conserved epitopes of the native HIV-1 envelope – targets that are not faithfully modelled in SIV-based systems [34]. Second, presentation of authentic HIV-1 antigens within an intact and fully competent immune system provides a strong framework for vaccine and antibody research. Notably, the model supports evaluation of broadly neutralizing antibody (bNAb) efficacy, investigation of vaccine-elicited mucosal and systemic immune responses, and generation of predictive in vivo data for bNAb-based prevention and treatment strategies, areas in which humanized mouse models and conventional SHIV challenges show intrinsic limitations [35,36]. However, the absence of direct histopathological assessment of organ injury and tissue viral reservoirs, together with immunological differences relative to human HIV-1 infection, constrains application to studies of HIV-associated organ pathology, reservoir quantification, latency-reversing interventions, AIDS-associated malignancies, opportunistic infections, or immune checkpoint inhibitor-based therapies [37–39]. Within this context, the stHIV-1sv/G53D NPM model is best positioned as a complementary experimental platform for validation and optimization of direct antiviral strategies targeting HIV-1.

Author contributions

Conceptualization: Yong-Tang Zheng, Tian-Zhang Song; Data curation: Tian-Zhang Song, Qing Li, Li-Rong Zhang, Yan Hu; Data interpretation: Tian-Zhang Song, Qing Li, Li-Rong Zhang, Yan Hu, Ying Lu; Formal analysis: Yong-Tang Zheng, Tian-Zhang Song; Investigation: Tian-Zhang Song, Qing Li, Li-Rong Zhang, Yan Hu; Project administration: Yong-Tang Zheng, Tian-Zhang Song; Software: Tian-Zhang Song, Qing Li, Li-Rong Zhang, Yan Hu; Supervision: Yong-Tang Zheng, Tian-Zhang Song; Writing: Yong-Tang Zheng, Tian-Zhang Song.

Supplementary Material

Revised Supplementary Figure 2.tif
TEMI_A_2648891_SM4245.tif (2.8MB, tif)
Revised Supplementary Table 1.xlsx
TEMI_A_2648891_SM4244.xlsx (12.2KB, xlsx)
Revised Supplementary Figure 1.tif
TEMI_A_2648891_SM4243.tif (4.6MB, tif)

Funding Statement

This work was supported by the Yunnan Province Science and Technology Department under Grant 202401AS070076, 202402AA310011 and 202403AC100011; the National Natural Science Foundation of China under Grant U23A20473, 82202508 and 82350710801, and the National Key R & D Program of China under Grant 2023YFC2306700.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Data availability statement

All data from this study are available from the corresponding authors upon reasonable request. The Sequencing data from mPBMC and 16s have been deposited in the Genome Sequence Archive (GSA) under accession codes CRA024824 (Shared link: https://ngdc.cncb.ac.cn/gsa/s/qul78yiK) and CRA024732 (Shared link: https://ngdc.cncb.ac.cn/gsa/s/zmR14xCJ). Sequencing data from hPBMC have been deposited in the GSA for Human under accession codes HRA011256 (Shared link: https://ngdc.cncb.ac.cn/gsa-human/s/K7I7sCQB). The proteome data have been deposited in the OMlX under accession codes OMIX009907 (Shared link: https://ngdc.cncb.ac.cn/omix/preview/598zHGj8).

Supplemental Material

Supplemental data for this article can be accessed online at https://doi.org/10.1080/22221751.2026.2648891.

References

  • 1.Victor Garcia J. Humanized mice for HIV and AIDS research. Curr Opin Virol. 2016;19:56–64. doi: 10.1016/j.coviro.2016.06.010 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 2.Hasler MF, Speck RF, Kadzioch NP.. Humanized mice for studying HIV latency and potentially its eradication. Curr Opin HIV AIDS. 2024;19(3):157–167. doi: 10.1097/COH.0000000000000855 [DOI] [PubMed] [Google Scholar]
  • 3.Han R, Su L, Cheng L.. Advancing human vaccine development using humanized mouse models. Vaccines (Basel). 2024;12(9):1012. doi: 10.3390/vaccines12091012 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 4.Fraker S, Atkinson B, Heredia A.. Humanized mouse models for preclinical evaluation of HIV cure strategies. AIDS Rev. 2022;24(3):139–151. doi: 10.24875/AIDSRev.22000013 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5.Evans DT, Silvestri G.. Nonhuman primate models in AIDS research. Curr Opin HIV AIDS. 2013;8(4):255–261. doi: 10.1097/COH.0b013e328361cee8 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6.Thippeshappa R, Kimata JT, Kaushal D.. Toward a macaque model of HIV-1 infection: roadblocks, progress, and future strategies. Front Microbiol. 2020;11:882. doi: 10.3389/fmicb.2020.00882 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7.Garcia-Tellez T, Huot N, Ploquin MJ, et al. Non-human primates in HIV research: achievements, limits and alternatives. Infect Genet Evol. 2016;46:324–332. doi: 10.1016/j.meegid.2016.07.012 [DOI] [PubMed] [Google Scholar]
  • 8.Hirsch VM, Dapolito G, Goeken R, et al. Phylogeny and natural history of the primate lentiviruses, SIV and HIV. Curr Opin Genet Dev. 1995;5(6):798–806. doi: 10.1016/0959-437x(95)80014-v [DOI] [PubMed] [Google Scholar]
  • 9.Hu SL. Non-human primate models for AIDS vaccine research. Curr Drug Targets Infect Disord. 2005;5(2):193–201. doi: 10.2174/1568005054201508 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10.Sattentau QJ, Stevenson M.. Macrophages and HIV-1: An unhealthy constellation. Cell Host Microbe. 2016;19(3):304–310. doi: 10.1016/j.chom.2016.02.013 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11.Agy MB, Frumkin LR, Corey L, et al. Infection of Macaca nemestrina by human immunodeficiency virus type-1. Science. 1992;257(5066):103–106. doi: 10.1126/science.1621083 [DOI] [PubMed] [Google Scholar]
  • 12.Liao CH, Kuang YQ, Liu HL, et al. A novel fusion gene, TRIM5-Cyclophilin A in the pig-tailed macaque determines its susceptibility to HIV-1 infection. AIDS. 2007;21(Suppl 8):S19–S26. doi: 10.1097/01.aids.0000304692.09143.1b [DOI] [PubMed] [Google Scholar]
  • 13.Pang W, Zhang GH, Jiang J, et al. HIV-1 can infect northern pig-tailed macaques (Macaca leonina) and form viral reservoirs in vivo. Sci Bull. 2017;62(19):1315–1324. doi: 10.1016/j.scib.2017.09.020 [DOI] [PubMed] [Google Scholar]
  • 14.Hatziioannou T, Ambrose Z, Chung NP, et al. A macaque model of HIV-1 infection. Proc Natl Acad Sci. 2009;106(11):4425–4429. doi: 10.1073/pnas.0812587106 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15.Pang W, Song JH, Lu Y, et al. Host restriction factors APOBEC3G/3F and other interferon-related gene expressions affect early HIV-1 infection in northern pig-tailed macaque (Macaca leonina). Front Immunol. 2018;9:1965. doi: 10.3389/fimmu.2018.01965 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16.Pang W, Shao Y, Zhuang XL, et al. Genomic evidence for the nonpathogenic state in HIV-1-infected northern pig-tailed macaques. Mol Biol Evol. 2023;40(5):msad101. doi: 10.1093/molbev/msad101 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17.Lu Y, Pang W, Zhang MD, et al. A novel Vpu adaptive mutation of HIV-1 degrades Tetherin in northern pig-tailed macaques (Macaca leonina) mainly via the ubiquitin-proteasome pathway and increases viral release. J Virol. 2023;97(4):e0020023. doi: 10.1128/jvi.00200-23 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18.Zhang MX, Zheng HY, Jiang J, et al. Predict disease progression from T-cell phenotypes in northern pig-tailed macaques (Macaca leonina) during SIVmac239 infection. Immunology. 2017;152(4):574–579. doi: 10.1111/imm.12794 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19.Song TZ, Zhang MX, Xia YJ, et al. Parasites may exit immunocompromised northern pig-tailed macaques (Macaca leonina) infected with SIVmac239. Zool Res. 2018;39(1):42–51. doi: 10.24272/j.issn.2095-8137.2018.015 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20.Song TZ, Zhang MX, Zhang HD, et al. Glucose metabolism disorder induces spermatogenic dysfunction in northern pig-tailed macaques (Macaca leonina) with long-term SIVmac239 infection. Front Endocrinol. 2021;12:745984. doi: 10.3389/fendo.2021.745984 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21.Yang W, Liu LB, Liu FL, et al. Single-cell RNA sequencing reveals the fragility of male spermatogenic cells to Zika virus-induced complement activation. Nat Commun. 2023;14(1):2476. doi: 10.1038/s41467-023-38223-z [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 22.Kazer SW, Walker BD, Shalek AK.. Evolution and diversity of immune responses during acute HIV infection. Immunity. 2020;53(5):908–924. doi: 10.1016/j.immuni.2020.10.015 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23.Coffin J, Swanstrom R.. HIV pathogenesis: dynamics and genetics of viral populations and infected cells. Cold Spring Harb Perspect Med. 2013;3(1):a012526. doi: 10.1101/cshperspect.a012526 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24.Zhang MX, Zheng HY, Jiang J, et al. Northern pig-tailed macaques (Macaca leonina) maintain superior CD4+ T-cell homeostasis during SIVmac239 infection. Eur J Immunol. 2018;48(2):384–385. doi: 10.1002/eji.201747284 [DOI] [PubMed] [Google Scholar]
  • 25.He WQ, He XY, Lu Y, et al. HIV-1 but not SIVmac239 induces higher interferon-α antiviral state in chronic infected northern pig-tailed macaques (Macaca leonina). Microbes Infect. 2022;24(5):104970. doi: 10.1016/j.micinf.2022.104970 [DOI] [PubMed] [Google Scholar]
  • 26.Luo MT, Mu D, Yang X, et al. Tree shrew cells transduced with human CD4 and CCR5 support early steps of HIV-1 replication, but viral infectivity is restricted by APOBEC3. J Virol. 2021;95(16):e0002021. doi: 10.1128/JVI.00020-21 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 27.Nakano Y, Yamamoto K, Ueda MT, et al. A role for gorilla APOBEC3G in shaping lentivirus evolution including transmission to humans. PLoS Pathog. 2020;16(9):e1008812. doi: 10.1371/journal.ppat.1008812 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28.Mangeat B, Gers-Huber G, Lehmann M, et al. HIV-1 Vpu neutralizes the antiviral factor Tetherin/BST-2 by binding it and directing its beta-TrCP2-dependent degradation. PLoS Pathog. 2009;5(9):e1000574. doi: 10.1371/journal.ppat.1000574 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 29.Mitchell RS, Katsura C, Skasko MA, et al. Vpu antagonizes BST-2-mediated restriction of HIV-1 release via beta-TrCP and endo-lysosomal trafficking. PLoS Pathog. 2009;5(5):e1000450. doi: 10.1371/journal.ppat.1000450 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 30.Waheed AA, Kuruppu ND, Felton KL, et al. In COS cells Vpu can both stabilize tetherin expression and counteract its antiviral activity. PLoS One. 2014;9(10):e111628. doi: 10.1371/journal.pone.0111628 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 31.Shingai M, Yoshida T, Martin MA, et al. Some human immunodeficiency virus type 1 Vpu proteins are able to antagonize macaque BST-2 in vitro and in vivo: Vpu-negative simian-human immunodeficiency viruses are attenuated in vivo. J Virol. 2011;85(19):9708–9715. doi: 10.1128/JVI.00626-11 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 32.Sauter D, Schindler M, Specht A, et al. Tetherin-driven adaptation of Vpu and Nef function and the evolution of pandemic and nonpandemic HIV-1 strains. Cell Host Microbe. 2009;6(5):409–421. doi: 10.1016/j.chom.2009.10.004 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 33.Heusinger E, Kirchhoff F.. Primate lentiviruses modulate NF-κB activity by multiple mechanisms to fine-tune viral and cellular gene expression. Front Microbiol. 2017;8:198. doi: 10.3389/fmicb.2017.00198 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 34.Policicchio BB, Pandrea I, Apetrei C.. Animal Models for HIV Cure Research. Front Immunol. 2016;7:12. doi: 10.3389/fimmu.2016.00012 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 35.Nishimura Y, Martin MA.. Of mice, macaques, and men: broadly neutralizing antibody immunotherapy for HIV-1. Cell Host Microbe. 2017;22(2):207–216. doi: 10.1016/j.chom.2017.07.010 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 36.Sok D, Burton DR.. Recent progress in broadly neutralizing antibodies to HIV. Nat Immunol. 2018;19(11):1179–1188. doi: 10.1038/s41590-018-0235-7 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 37.Banga R, Perreau M.. The multifaceted nature of HIV tissue reservoirs. Curr Opin HIV AIDS. 2024;19(3):116–123. doi: 10.1097/COH.0000000000000851 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 38.Chen H, Meng J, Lin L.. The mechanism of RNA virus-mediated tumorigenesis. iCell. 2024;1:xewe3812. doi: 10.71373/xewe3812 [DOI] [Google Scholar]
  • 39.Lee J, Whitney JB.. Immune checkpoint inhibition as a therapeutic strategy for HIV eradication: current insights and future directions. Curr Opin HIV AIDS. 2024;19(4):179–186. doi: 10.1097/COH.0000000000000863 [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Revised Supplementary Figure 2.tif
TEMI_A_2648891_SM4245.tif (2.8MB, tif)
Revised Supplementary Table 1.xlsx
TEMI_A_2648891_SM4244.xlsx (12.2KB, xlsx)
Revised Supplementary Figure 1.tif
TEMI_A_2648891_SM4243.tif (4.6MB, tif)

Data Availability Statement

All data from this study are available from the corresponding authors upon reasonable request. The Sequencing data from mPBMC and 16s have been deposited in the Genome Sequence Archive (GSA) under accession codes CRA024824 (Shared link: https://ngdc.cncb.ac.cn/gsa/s/qul78yiK) and CRA024732 (Shared link: https://ngdc.cncb.ac.cn/gsa/s/zmR14xCJ). Sequencing data from hPBMC have been deposited in the GSA for Human under accession codes HRA011256 (Shared link: https://ngdc.cncb.ac.cn/gsa-human/s/K7I7sCQB). The proteome data have been deposited in the OMlX under accession codes OMIX009907 (Shared link: https://ngdc.cncb.ac.cn/omix/preview/598zHGj8).

Articles from Emerging Microbes & Infections are provided here courtesy of Taylor & Francis

RESOURCES