Chronic SIV-infected rhesus macaques develop increased lesions following MPXV infection
To mimic mpox cases with and without HIV co-infection and compare their pathogenicity, we enrolled six rhesus macaques with chronic simian immunodeficiency virus (SIVmac239) infection into the SIV-MPXV (MS) group and six naïve macaques into the MPXV-alone (MP) group (Fig.1a and Supplementary Data S1). Prior to the MPXV challenge, the six macaques in the MS group were intravenously inoculated with 100 TCID50 of SIVmac239 and monitored for plasma viral loads over 115 days. During this period, all six macaques exhibited persistent SIV replication in peripheral blood (Supplementary Fig. S1a and Fig.1b) along with a decreased CD4+/CD8+ T cell ratio and CD4+ T cell counts (Supplementary Fig. S1b), confirming the successful establishment of a chronic and stable SIV infection model. Subsequently, all twelve macaques were intravenously challenged with 1 × 107 TCID50 of MPXV clade IIb (MPXV-B.1-China-C-Tan-CQ01), isolated from the first imported mpox case in China in 2022.
After the MPXV challenge, three macaques in the MP and MS group were monitored for 10 days and euthanized, while another three macaques in either group were continuously monitored for 35 days to measure viral loads in plasma and tissues, binding and neutralizing antibodies, cellular immune response, and cytokine levels. MPXV DNA copies in both the MP and MS groups exhibited a peak on days 7 and 10, reaching 4.18 × 105 and 5.89 × 105 copies/mL, respectively. These levels are comparable to those reported in previous studies of MPXV infection in non-human primates33,34, suggesting a typical viral replication pattern during the acute phase of the infection. Notably, despite the absence of significant differences in MPXV plasma loads between the two groups throughout the 35-day observation period (Fig.1b), the chronic SIV co-infection was associated with more severe skin lesions and other pathological manifestations. This indicates that factors other than plasma viral load may contribute to the enhanced pathogenicity in the co-infected animals. Deaths were not observed in either group (Supplementary Fig. S1c), and only mild weight loss was noted, with no significant differences between the two groups (Supplementary Fig. S1d). The body temperature of all twelve macaques peaked significantly at four days post-MPXV infection (dpi) and then returned to levels similar to those at one dpi, with no significant differences between the MP and MS groups (Supplementary Fig. S1e). These findings suggest that while the overall systemic manifestations, such as weight loss and fever, were not markedly affected by SIV co-infection, the local tissue damage, as evidenced by the skin lesions, was more pronounced.
Skin lesions are the most common and obvious symptom of MPXV infection, and close contact with skin lesions mainly leads to human-to-human transmission of MPXV. We continuously monitored lesion counts at 1, 4, 7, 10, 14, 21, 28, and 35 days post-MPXV infection. In both groups, skin lesions were first observed at 7 dpi and reached a peak at 10 dpi (Fig.1c). Notably, the macaques of MS group showed more skin lesion than those in the MP group on day 10 (group mean: 50.17 vs. 20.33) and 14 (group mean: 49.00 vs. 14.00) post-MPXV challenge (P < 0.0001 for all comparison). Disease duration was measured as the time from lesion onset to resolution and was charted for each individual animal (Fig. 1d). Lesions were distributed across all limbs, pygal region, soles, torso, and face of the macaques in both the MS and MP groups (Fig. 1e). Significant more lesions appeared on the limbs and pygal region of the macaques in the MS group than the MP group at 10 dpi. Although not statistically significant, other body parts also exhibited more lesions in the MS group (Fig. 1f). The increased number of skin lesions observed in the macaques with SIV-MPXV co-infection was similar to the higher incidence of skin lesions seen in HIV-positive patients infected with MPXV compared to those without HIV1. This may be due to the insufficient immune response in SIV-infected macaques, which fails to control MPXV replication. As a result, the virus spreads rapidly throughout the body, leading to more severe skin lesions. In addition, the higher MPXV DNA loads found in the skin lesions of the MS group than those in the MP group (P < 0.01, Fig. 1g) further confirm this.
To assess the severity of pathological changes in the skin lesions of rhesus monkeys, skin samples from the limbs were collected, fixed, sectioned, stained with hematoxylin and eosin (H&E), and analyzed histopathologically. Compared to normal skin, skin lesions in both the MP and MS groups had extensive increases in the spinous cell layers, moderate to severe epidermal thickening, and loosening of connective tissue on the skin surface. However, skin lesions in the MS group displayed additional signs of pathological damage, including spinous cell necrosis, fibroblast necrosis in the dermis, nuclear condensation and fragmentation, and increased inflammatory cell infiltration (primarily lymphocytes and granulocytes) compared to the MP group (Fig. 1h). In summary, although there were no significant differences in MPXV plasma viral loads, survival, or body temperature, chronic SIV co-infection resulted in more severe skin lesions than MPXV infection alone in rhesus macaques.
Protein and phosphosite profilings of skin lesions exhibit differences between SIV-MPXV co-infection and MPXV single infection rhesus macaques
Disruptions in translation and proteostasis are frequently observed characteristics of pathogenic viruses. For a preliminary glance at SIV-MPXV co-infection induced molecular changes, limb skin lesions collected at 10 days post-MPXV infection from both the MP and the MS groups, and samples from eight other organs, were analyzed using liquid chromatography-mass spectrometry (LC-MC)-based proteomic and phosphoproteomic techniques. A total of 11,527 proteins (derived from 113,415 peptides) and 26,582 phosphosites (24,576 phosphopeptides) in 6,084 proteins were identified. To ensure high reliability, only proteins/sites present in > 50% of the samples in at least one group were included, resulting in 11,457 proteins and 26,086 phosphosites from 6056 proteins for further analysis.
For lesions from each macaque, paired normal adjacent skin was named peripheral lesion (MP_PL or MS_PL) and used as control samples (Fig. 2a). In the macaque skin (lesion and PL), a total of 9554 proteins (derived from 65,589 peptides) and 18,565 phosphosites (16,725 phosphopeptides) were identified, and 8028 proteins and 13,718 sites were considered highly reliable. Principal component analysis (PCA), a basic unsupervised clustering method, was used to evaluate differences among groups. Though there was overlap among skin lesion and PL groups, MP_lesion and MS_lesion groups were mostly separated at protein expression level (Supplementary Fig. S2a) and were totally separated at the phosphosite expression level (Supplementary Fig. S2b), revealing skin lesions had different protein and phosphosite expression profilings. Comparing to MP_PL, lesions in the MP group had 29 upregulated and 26 downregulated proteins (Supplementary Fig. S2c and Supplementary Data S2). Comparing to MS_PL, lesions in the MS group had 67 upregulated and 29 downregulated proteins (Supplementary Fig. S2d and Supplementary Data S2). Combining protein alterations into a scatter plot, it is obvious that proteins only upregulated in the MS group (66 proteins) were much more than proteins only upregulated in the MP group (28 proteins; Fig.2b and Supplementary Data S2). Using KEGG enrichment analysis, we identified that in the MS group, proteins associated with the Rap1 signaling pathway were especially upregulated. The severe immunosuppression and chronic immune activation induced by SIV infection35,36 may promote enhanced recruitment of inflammatory cells (e.g., neutrophils, monocytes/macrophages) to the skin lesions of the MS group. Activation of the Rap1 signaling pathway facilitates adhesion and migration of these cells37, potentially exacerbating local inflammatory infiltration and tissue destruction. Proteins only upregulated in the MP group were functionally enriched in the IL-17 signaling pathway, arachidonic acid metabolism, PPAR signaling pathway, et al. (Fig. 2c and Supplementary Data S2). Proteins only downregulated in the MP group were functionally enriched in carbohydrate digestion and absorption, adipocytokine signaling pathway, thyroid hormone signaling pathway, and apoptosis (Fig. 2c and Supplementary Data S2). These features in the skin lesions of the MP group reflected a more classic viral skin inflammatory response: robust IL-17-mediated neutrophil recruitment and barrier defense38, coupled with active arachidonic acid metabolism generating pro-inflammatory lipid mediators39. Concurrently, PPAR pathway activation serves to initiate anti-inflammatory responses and tissue repair40. Downregulation of metabolic pathways indicates energy redistribution toward glycolytic energy production and inflammatory/antiviral responses in infected and immune cells41. Based on the protein-protein interaction (PPI) annotations, PTGS2 (regulating arachidonic acid metabolism) showed regulatory potential in mpox infection because of its interaction with multiple proteins (Fig. 2d).
Phosphorylation is one of the most pivotal biological mechanisms for regulating cellular processes. Given that nearly all cell signaling pathways rely on phosphotransfer reactions and that dysregulated kinase activity is implicated in numerous diseases, kinases represent an appealing target for therapeutic intervention42,43. Therefore, we also conducted the phosphorylation analysis on multi-organ samples of rhesus macaques with SIV-MPXV co-infection and MPXV infection alone. Comparing the MP_lesion to the MP_PL samples, 137 upregulated phosphosites located in proteins regulating phagosome, protein export, neutrophil extracellular trap formation, platelet activation, C-type lectin receptor signaling pathway, IL-17 signaling pathway, and tight junction (Supplementary Fig. S2e and Supplementary Data S3). The detected protein phosphorylation suggests significant inflammatory responses and the reprogrammed host’s baseline gene expression pattern, which may enhance the efficiency of viral genome transcription and translation, thus promoting virus propagation within infected cells44,45. The proteins of MP_lesion containing 75 downregulated phosphosites were associated nucleocytoplasmic transport and thermogenesis (Supplementary Fig. S2e and Supplementary Data S3). Compared to the MS_PL group, the MS_lesion group had 204 upregulated and 81 downregulated phosphosites. In the MS_lesion groups, proteins containing upregulated phosphosites were functionally enriched in chemokine signaling pathway, Fc gamma receptor-mediated phagocytosis, B cell receptor signaling, endocytosis, neutrophil extracellular trap formation, platelet activation, natural killer cell- mediated cytotoxity, NF-kappa B signaling pathway, and C-type lectin receptor signaling pathway. Proteins containing downregulated phosphosites were functionally associated with toll and immune deficiency signaling pathway (Supplementary Fig. S2f and Supplementary Data S3). Although the pronounced upregulation of phosphosites within these signaling pathways may significantly bolster the host’s antiviral immune responses, the overall effectiveness of these mechanisms might be attenuated due to the profound immune dysregulation caused by HIV infection, especially the marked impairment of CD4+ T-cell functionality. Referring to reported kinase-phosphosite relationships, kinase predication was performed based on phosphosite expression intensity. CDK1, FYN, TTK, LYN, SRC, YES1, CSK, GRB2, and ZAP70 were predicted to be activated, while CAMK2D, ILK, GSK3B, CAMK2D, CAMK2G, PRKG1, PRKACA, and PRKACG were inhibited in the MP skin lesions (Fig. 2e and Supplementary Data S4). CDK1, CAMK2A, AURKB, ILK, AURKA, PLK1, TTK, and PLK4 were predicated to be activated while PIK3CD, SGK1, SGK3, RIMS1, and PLK3 were inhibited in the MS skin lesions (Fig.2f and Supplementary Data S4). CDK1 and TTK were predicted to be activated in both the MP_lesion and MS_lesion groups. However, CAMK2A and ILK were predicted to have opposite activity in the MP_lesion and MS_lesion groups. Eight motifs were summarized among reliable phosphosites, showing proline, arginine, lysine, and glutamate residues most commonly appeared near phosphosites (Fig. 2g). Referring to drug-protein relationships recorded in DrugBank, several Food and Drug Administration (FDA)-approved drugs targeting identified kinases in MPXV-induced skin lesions were listed (Supplementary Fig. S2g and Supplementary Data S4). These aforementioned therapeutic agents hold potential to provide novel avenues for the treatment of mpox.
In summary, MPXV induced distinct protein expression profiles in skin lesions depending on the presence or absence of SIV co-infection.
Chronic SIV infection enhances MPXV replication and caused systematic dysfunction in MPXV-infected macaques
To understand whether SIV only exacerbates local skin lesions or widely induces systemic dysfunction in MPXV-infected macaques, samples from lymph nodes, immune tissues, the gastrointestinal tract, and other organs were collected for viral load measurements. Compared to the MP group, the MPXV DNA loads in various organs were significantly higher in the MS group, which was different from the phenomena that both the MP and MS groups had similar levels of viremia (Fig. 3a). For further molecular investigation, proteomic and phosphoproteomic data of eight organs, including inguinal lymph node (ILN), spleen, cerebral cortex (CC), heart, lung, liver, kidney, and rectum were processed for analyses (Fig. 3b). To establish baseline data for the study, we selected another three healthy rhesus monkeys (group Control) in this section of the experiment and conducted proteomic and phosphogenomic analysis on the multiple organs. Compared to MPXV-infected alone, MPXV loads in SIV-MPXV co-infection macaques were significantly increased in the heart, lung, kidney, and rectum and had upward trends in the brain and liver (Supplementary Fig. S3a). Notably, MPXV burden in the inguinal lymph node and spleen of the MS group macaques were both 1.2 times those in the MP group macaques (Supplementary Fig. S3b). For other organs, compared to macaques infected with MPXV alone, macaques co-infected with SIV-MPXV appear to have a higher burden of MPXV (Supplementary Fig. S3c). In the eight organs for proteomic analysis, 11,481 proteins (112,559 peptides) and 26,326 phosphosites (24,352 phosphopeptides) were identified, and 11,357 proteins and 25,688 sites were considered highly reliable. MPXV single infection or SIV-MPXV co-infection had no significant effect on the count of identified proteins and phosphosites (Fig.3c,d). However, SIV-MPXV organs commonly had more downregulated proteins and phosphosites than MPXV-alone ones (Fig.3c and Supplementary Data S5). Unsupervised clustering revealed that the tested eight organs were clearly separated based on LC-MS-identified proteins (Supplementary Fig. S3d) and phosphosites (Supplementary Fig. S3e), revealing that each organ had its unique characteristics. Compared to the control macaques, each organ had its unique upregulated and downregulated proteins in the MP (Supplementary Fig. S4a and Supplementary Data S6) and MS (Supplementary Fig. S4b and Supplementary Data S6) groups. To focus on the effect of chronic SIV infection on MPXV-infected macaques, proteins of each organ in the MS group were directly compared to the MP group, and proteins that were unique upregulated or downregulated in each organ were extracted for KEGG enrichment analyses (Fig. 3e). Compared to the MP group, we identified that the cerebral cortex had upregulated neurotrophin signaling pathway, apoptosis, and neutrophil extracellular trap formation, the heart had downregulated oxytocin signaling pathway, the kidney had downregulated ECM-receptor interaction, the liver had upregulated protein digestion and absorption and downregulated RNA polymerase, the lung had downregulated ErbB signaling pathway, base excision repair, longevity regulating pathway, Th1 and Th2 cell differentiation, and notch signaling pathway, and the rectum had upregulated N-glycan biosynthesis in the MS group (Fig. 3e and Supplementary Data S6). These disturbances indicated that chronic SIV infection exacerbates the burden on multiple tissues of rhesus monkeys infected with MPXV.
Using the phosphosite expression levels, kinases were predicted in each organ in the MP and MS groups (Fig. 3f, Supplementary Fig. S4c, d). comparing the MS to MP groups, the observed differences, particularly in cell cycle control (PLK1, PLK4, AURKA, AURKB, CDK1, BUB1, BUB1B, TTK, NEK2), immune signaling (SYK, ZAP70, LYN, FYN, SRC, YES1, GRAP2), PI3K/MTOR pathways, and cytoskeletal organization (CAMK2A, CAMK2G, ILK, SRC), strongly suggested alterations in cellular proliferation, immune responses, metabolic homeostasis, and structural integrity. In summary, although exacerbated skin lesions were nearly the only apparent physical signs, chronic SIV infection caused systemic dysfunction, promoted inflammatory responses, and affected cell mitosis and cytoskeleton regulation.
Chronic SIV infection impairs the immune responses and functional heterogeneity in the spleen and lymph node
In the absence of antiviral treatment (ART), HIV infection is characterized by the gradual loss of CD4+ T cells and progressive immune deficiency that leads to opportunistic infections, and ultimately death46. Early CD8-cell depletion studies in experimentally SIV-infected rhesus macaques demonstrated that CD8+ T cells are critical for controlling virus replication in vivo47,48. Cellular immune responses against cross-reactive orthopoxviruses of rhesus monkeys were evaluated at 28 dpi to understand the impact of chronic SIV infection. The vaccinia virus (VACV), a poxvirus with 90% sequence homology to MPXV49, was used to assess specific cellular immune responses, as its induced T-cell response is largely cross-reactive with MPXV epitopes. Flow cytometry was performed to analyze VACV-specific cellular immune responses in the MP and MS groups. The results showed that stimulation of peripheral blood mononuclear cells (PBMCs) with VACV resulted in an obvious decrease in the proportion of VACV-specific CD8+ T cells producing interferon-γ (IFN-γ), interleukin-2 (IL-2), and tumor necrosis factor-α (TNF-α) in the MS group compared to the MP group (Fig. 4a and Supplementary Fig. S5a). In contrast, there were no significant differences in the proportion of IFN-γ, IL-2, and TNF-α secreting CD4+ T cells between the MP and MS groups. These findings indicate that chronic SIV infection impairs specific CD8+ T cell responses against MPXV in rhesus monkeys, while having no significant effect on CD4+ T cell immunity.
To further assess the impact of chronic SIV infection on humoral immune responses against MPXV infection, the plasma neutralizing antibody levels against MPXV were initially measured using a 50% plaque reduction neutralization titer (PRNT50). Neutralizing antibodies against MPXV peaked at 28 dpi (Fig. 4b). The antibody titer in plasma from the MS group macaques was significantly lower than that in the MP group (1:59 vs. 1:100, P < 0.05), indicating an impaired humoral immune response against MPXV due to chronic SIV infection. By 180 dpi, the PRNT50 neutralizing antibody titers were 1:46 and 1:22 in the plasma of the MP and MS groups, respectively, reflecting the persistence of humoral immunity after MPXV infection. In addition, total IgG antibody levels in plasma collected at 28 dpi were tested against six MPXV antigens by ELISA, including A35, B6R, H3L, M1R, A29, and E8L. The results showed no significant difference in specific binding antibodies to these six key MPXV antigens between the two groups (Supplementary Fig. S5b).
Plasma cytokine and chemokine levels in plasma at 10 dpi of the naïve, SIV-alone, MPXV-alone, and SIV-MPXV macaques were also detected. The MS group monkeys had significantly higher levels of B cell activating factor (BAFF), C-X-C motif chemokine 10 (CXCL10), and C-C motif chemokine ligand 5 (CCL5) compared to both the MP and SIV groups (P < 0.05 for all comparisons, Fig. 4c). The plasma level of suppression of tumorigenicity 2 (ST2) in the MS group was higher than that in the MP group but lower than in the SIV group, indicating that MPXV infection inhibits ST2 production (P < 0.05 for all comparisons). In addition, the growth/differentiation factor-15 (GDF-15) showed no significant difference between MPXV-infected monkeys with or without SIV infection (P > 0.05). We found high levels of IL-6 and IFN-γ in the plasma of macaques infected with SIV-MPXV, and the concentrations of C-reactive protein (CRP) and serum amyloid A1 (SAA1) in the MS group were 1.7- and 1.2-fold those in the MP group, respectively, which showing intense inflammatory responses in the co-infected macaques. The significantly low levels of cellular and humoral immunity against MPXV in the macaques highlight the substantial damage caused by SIV and MPXV to the immune system.
Further combining the proteomic analyses of immune organs, differentially expressed proteins were calculated among control, MP, and MS macaques for ILN and spleen, respectively. Then, differentially expressed proteins were separated into eight expression clusters in the ILN (Fig. 4d) and seven clusters in the spleen (Fig. 4e), respectively. In the lymph nodes, control, MP, and MS macaques had gradually decreased T cell receptor signaling pathway and Th1, Th2, and Th17 cell differentiation (cluster 4). Adhesion pathways were downregulated in the MS ILN (cluster 5) but were not decreased in the MP ILN (cluster 1; Fig. 4d and Supplementary Data S7). In the spleen, the control, MP, and MS groups had gradually higher complement and coagulation cascades, NF-kappa B signaling pathway, RIG-I-like receptor signaling, pyrimidine metabolism, and natural killer cell-mediated cytotoxicity (cluster 1) and gradually downregulated ECM-receptor interaction and focal adhesion (cluster 6). DNA replication seemed enhanced in both the MP and MS spleens (cluster 7; Fig. 4e and Supplementary Data S7). Researchers have also confirmed the downregulation of cell-cell adhesion molecules in cells infected with MPXV, which may lead to the entry of MPXV into cells during the infection process32.
Comparing the differentially expressed phosphosites between MP/control (Ctr) and MS/Ctr, both lymph node and spleen showed unique up- and down-regulated phosphosites in the MP and MS groups (Supplementary Fig. S5c, d and Supplementary Data S8). In the lymph node, MPXV infection alone especially upregulated phosphosites associated with mTOR signaling pathway and AMPK pathway, while SIV-MPXV co-infection especially upregulated phosphosites associated with DNA replication, ribosome, and protein processing (Supplementary Fig. S5e and Supplementary Data S8). In the spleen, the upregulated protein in the MP group were associated with ribosome biogenesis in eukaryotes, homologous recombination, and antigen processing and presentation. While the MS group showed more upregulated proteins related to the regulation of actin cytoskeleton, proteasome, and arginine and proline metabolism (Supplementary Fig. S5fand Supplementary Data S8). Using kinase prediction, the kinases (PLK1, PLK2, CDK3, CDK4, CDK6, CDK16, and CDK18) associated with DNA replication were identified to be activated in lymph node and spleen in both MP and MS groups (Fig. 4f, g and Supplementary Data S9). The MAPK3 was especially activated in the spleen in the SIV-MPXV co-infection group (Fig. 4g and Supplementary Data S9). Based on DrugBank, drugs targeting altered kinases in the MP or MS immune organs were also listed (Fig. 4h and Supplementary Data S9).
