Study subjects used in this study were enrolled from the Beijing PRIMO Clinical Cohort, a prospective study cohort of HIV-1-negative MSM to identify AHI at Beijing You’an Hospital, Beijing, China, since October 2006 (28). After enrollment, they were screened for plasma HIV-1 antibody, HIV-1 RNA level, clinical signs of acute infection, and syphilis status. Then they were followed up every 2 months, and HIV-1 antibody and syphilis were detected at each visit. AHI was defined as a negative or indeterminate HIV-1 antibody status but a positive HIV RNA (29). Once acute HIV-1-infected cases were identified, patients were followed up at 1, 2, 4, 8, and 12 weeks, and then every 3 months and syphilis status was monitored at each visit (Figure 1). Blood samples were collected, and peripheral blood mononuclear cells (PBMCs) and plasma were isolated and cryopreserved. Estimated date of infection was defined as the mid-point between the last HIV-1 antibody negative test and the first HIV-1 antibody positive test, or as 14 days prior to a positive RNA PCR assay on the same day as a negative HIV Enzyme Immunoassay. Patients whose infection time was longer than 180 days were defined as chronic HIV-1 infection (CHI). 26 AHI and 31 CHI patients were recruited randomly. The estimated infection date of AHI and CHI patients were 62 ± 33 and 565 ± 307 days, respectively. Early HIV-1 infection can be depicted as six discrete stages proposed by Fiebig et al. (30, 31). Stage I–II: HIV-1 RNA positive, ELISA negative with 3 patients; stage III–IV: HIV-1 RNA positive, ELISA positive, and Western blot negative or indeterminate with 4 patients; stage V–VI: HIV-1 RNA positive, ELISA positive, Western blot without/with p31 band with 19 patients. Syphilis was diagnosed on the basis of a compatible history, and the results of rapid plasma reagin (RPR) test (Shanghai Kehua Company, China) and T. pallidum particle agglutination assay (TPPA) (Fujirebio Diagnostics, Inc., Japan) (32). Based on the RPR and TPPA results, AHI and CHI patients were, divided into HIV⁺RPR⁺ group (n = 15, n = 18) and HIV⁺RPR⁻ group (n = 11, n = 13), respectively. All AHI and CHI patients were ART-naïve treated. 6 patients in acute HIV⁺RPR⁺ group and 4 patients in chronic HIV⁺RPR⁺ group received benzathine benzylpenicillin treatment for syphilis before their enrollment in this study. Patients with opportunistic infections or coinfections with tuberculosis, hepatitis B virus, or hepatitis C virus were excluded. Twenty age-matched male MSM HIV-1-negative controls (HC) were included as controls. The characteristics of all subjects are described in the Table 1.

This study and all relevant experiments have been approved by the Beijing You’an Hospital Research Ethics Committee and written informed consent was obtained from each participant according to the Declaration of Helsinki. All the participants provided written informed consent for their information, and clinical samples were stored and used for research. At enrollment, all subjects provided baseline demographic, clinical, and epidemiological information by completing a standardized questionnaire. The methods were carried out in accordance with approved guidelines and regulations.

Venous blood was collected into tubes containing EDTA as an anticoagulant. Whole blood cells analysis was performed on Uritest-3000 Fully Automated Hematology Analyzer (URIT medical electronic company, Shenzhen, China). Absolute counts of CD3⁺, CD4⁺, and CD8⁺ T lymphocytes were obtained as previously described (26).

Plasma HIV-1 viral load (copies/ml) were quantified using the COBAS AmpliPrep/COBAS TaqMan 48 Analyser (Roche Diagnostic, Branchburg, NJ, USA), or by real-time PCR (Abbott, Des Plaines, IL, USA), with a detection limit of 40 copies/ml of plasma.

HIV-1 infection status was determined by screening with a HIV-1/2 antigen/antibody combo enzyme immunoassay (Beijing Wantai Biological Medical Company, Beijing, China). Positive specimens were further conformed by using Western blot for HIV-1/2 (HIV Blot 2.0 MP Diagnostics, Singapore). The diagnosis of syphilis was performed with the RPR test (Shanghai Kehua Company, China) and TPPA (Fujirebio Diagnostics, Inc., Japan) (32). A seropositive result for TPPA was defined as the presence of a past or current syphilis infection, while a seropositive result for both TPPA and RPR was diagnosed as a current syphilis infection.

Phycoerythrin (PE)-conjugated anti-human CD38 (HIT2) monoclonal antibody (mAb), allophycocyanin (APC)-conjugated anti-human HLA-DR (L243) mAb, phycoerythrin-cyanine 7 (PE-cy7)-conjugated anti-human CD3 (HIT3a) mAb, peridinin-chlorophyll protein complex cyanine 5.5 (PerCP-cy5.5)-conjugated anti-human CD27 (O323) mAb, and APC-conjugated anti-human CD45RA (HI100) mAb were purchased from Biolegend (San Diego, CA, USA). Fluorescein isothiocyanate (FITC)-conjugated anti-human pan TCRγδ mAb, PE-conjugated anti-human pan TCRγδ (IMMU510) mAb and FITC-conjugated anti-human pan Vδ₂TCR mAb (IMMU389) were purchased from Immunotech (Beckman Coulter, Fullerton, France). FITC-conjugated anti-human panVδ₁TCR (TS8.2) mAb was purchased from Pierce (Rockford, IL, USA). PE-conjugated IL-17A (SCPL1362) mAb and APC-conjugated interferon-γ (IFN-γ) (B27) mAb were purchased from BD Pharmingen (San Diego, CA, USA). The isotype control mAbs were purchased from the corresponding companies.

For surface staining, PBMCs were isolated from HC and HIV-1-infected patients with or without syphilis infection. Cells were washed with 1% bovine serum albumin in PBS and were labeled with LIVE/DEAD fixable viability stain 510 (BD Biosciences, San Jose, CA, USA), and dead cells were excluded. Then, cells labeled with specific surface antibodies: gating strategy on CD3⁺γδTCR⁺ cells, Vδ₁ and Vδ₂ T cells, or γδT_Naive (CD27⁺CD45RA⁺), γδT_CM (CD27⁺CD45RA⁻), γδT_EM (CD27⁻CD45RA⁻), and γδT_EMRA (CD27⁻CD45RA⁺) were displayed (Figure 2). Cytometer setup and tracking calibration particles were used to ensure that fluorescence intensity measurement was consistent in all experiments. Flow cytometry Comp-Beads kits (BD Biosciences, San Jose, CA, USA) were used for compensation. Forward angle and side scatter light gating were gated on lymphocytes and were used to exclude cell debris from the analysis. Forward height and forward area were used to exclude doublet cells. Cells were performed with a FACScan flow cytometer, as previously described (33, 34). The final analysis was performed with FlowJo software (version 7.6.2; Tree Star Inc., Ashland, OR, USA), which generated a graphical output. The strategies for the analysis of flow cytometry data are detailed in Figure 2.

Peripheral blood mononuclear cells were incubated with 50 ng/ml phorbol 12-myristate-13-acetate (PMA), 1 μg/ml ionomycin, and 10 μg/ml brefeldinA (Sigma, St Louis, MO, USA) at 37°C for 6 h. Cells were collected, washed, and labeled with specific surface antibodies. Then, cells were washed, fixed, permeabilized, and incubated with IL-17 and IFN-γ mAbs. Cells were performed with a FACScan flow cytometer. Cells were first gated on 510 negative cells (live cells), then gated on CD3⁺γδTCR⁺ cells, the expression of IL-17 or IFN-γ was analyzed (Figure 2). Data were analyzed by FlowJo software, as described above.

Data are expressed as mean ± SD. Statistical analysis was performed with GraphPad Prism software version 5.03 (GraphPad Software, San Diego, CA, USA). Statistical significance P values for differences between groups were assessed by Mann–Whitney tests and one-way ANOVA test. The statistical dependence between variables was assessed by performing Spearman’s rank correlation analysis. All tests were two-tailed, and values of P < 0.05 were considered significant.

26 AHI, 31 CHI patients, and 20 HIV-1-negative HC were enrolled in this study (Table 1). Both AHI and CHI patients were further divided into two groups (HIV⁺RPR⁺ or HIV⁺RPR⁻) based on the syphilis status. The information of all the subjects is presented in Table 1. The age, sex, and the nationality of the HIV-1-infected patients and HC are matched. The estimated date of infection and viral load of HC are unavailable. The absolute number of CD4⁺ T cells and CD4/CD8 ratio in AHI and CHI patients are much lower than that in HC (P < 0.001). The absolute number of CD3⁺ and CD8⁺ T cells between HIV-1-infected patients and HC are not significant different (P > 0.05, Table 1).

First, we investigated the effect of syphilis on the frequencies of total γδ, Vδ₁, and Vδ₂ T-cell in HIV-1-infected patients. The frequencies of total γδ, Vδ₁, and Vδ₂ T cells among three groups: HC (n = 20), HIV⁺RPR⁻ (n = 24), and HIV⁺RPR⁺ (n = 33) patients were compared. As shown in Figure 3A, the frequencies of total γδ T cells were not significantly different among the three groups. The frequencies of Vδ₁ T cells was significantly higher in the HIV⁺RPR⁻ group than in HC, but no significant difference was found compared with HIV⁺RPR⁺ group. The frequencies of Vδ₂ T cells was significantly lower in both HIV⁺RPR⁻ and HIV⁺RPR⁺ groups compared with HC. However, we found no significant difference in Vδ₂ T-cell frequency between HIV⁺RPR⁻ and HIV⁺RPR⁺ groups (Figures 3B,C). Based on the above results, we assumed that HIV-1 infection status might affect the frequencies of total γδ, Vδ₁, and Vδ₂ T cells. Therefore, we split the HIV⁺RPR⁻ and HIV⁺RPR⁺ groups into subgroups of patients with 26 AHI and 31 CHI patients. The frequencies of total γδ, Vδ₁, and Vδ₂ T cells were compared between the HIV⁺RPR⁻ and HIV⁺RPR⁺ subgroups with AHI and CHI. We found no significant difference of the frequencies of total γδ T cells between HIV⁺RPR⁻ and HIV⁺RPR⁺ groups for either acute or CHI (Figure 3D). The frequencies of Vδ₁ T cells were significantly lower in the HIV⁺RPR⁺ group than in the HIV⁺RPR⁻ group for AHI patients. By contrast, in CHI, no significant difference of the frequencies of Vδ₁ T cells was found between the HIV⁺RPR⁺ and HIV⁺RPR⁻ groups. Moreover, in the HIV⁺RPR⁺ group, the frequencies of Vδ₁ T cells were significantly higher in CHI patients than in AHI patients (Figure 3E). Conversely, in AHI patients, the frequencies of Vδ₂ T cells were significantly higher in the HIV⁺RPR⁺ group than in the HIV⁺RPR⁻ group. However, no significant difference of the frequencies of Vδ₂ T cells between HIV⁺RPR⁻ and HIV⁺RPR⁺ groups was observed in CHI patients. Moreover, for HIV⁺RPR⁺ group, the frequencies of Vδ₂ T cells in AHI patients were significantly higher than that in CHI patients (Figure 3F). Taken together, these results suggest that syphilis has differential effects on the frequencies of Vδ₁ and Vδ₂ T cells in patients with HIV-1 infection, depending on that this infection is acute or chronic.

Memory CD4⁺ T-cell depletion is a major cause of HIV-1 pathogenesis and disease progression (35, 36). Like CD4⁺ T cells, γδ T cells can be split into naïve and memory T-cell subsets on the basis of their surface expression of CD45RA and CD27 (37). Herein, we found that the frequencies of naïve (CD27⁺CD45RA⁺, T_Naïve) γδ T cells in HIV⁺RPR⁺ group in AHI patients were significantly increased compared with HC, and HIV⁺RPR⁻ patients with both acute and CHI. The frequencies of T_Naïve γδ T cells in HIV⁺RPR⁺ group in CHI patients were significantly higher compared with HC, but no significant differences were shown compared with the other 3 groups (Figure 4A). In AHI patients, the frequencies of central memory (CD27⁺CD45RA⁻, T_CM) γδ T cells in the HIV⁺RPR⁻ group were significantly lower than that in HC and HIV⁺RPR⁺ group. In CHI patients, the frequencies of T_CM γδ T cells were significantly higher than that in HIV⁺RPR⁻ group in AHI patients (Figure 4B). Moreover, the frequencies of effector memory (CD27⁻CD45RA⁻, T_EM) γδ T cells in the HIV⁺RPR⁻ and HIV⁺RPR⁺ groups in both AHI and CHI patients were significantly lower than that in HC. However, there were no significant differences of the frequencies of T_EM γδ T cells between the HIV⁺RPR⁻ and HIV⁺RPR⁺ groups in both AHI and CHI patients (Figure 4C). Finally, we found the frequencies of terminally differentiated (CD27⁻CD45RA⁺, T_EMRA) γδ T cells in HIV⁺RPR⁻ group in AHI patients were significantly higher than that in HC and HIV⁺RPR⁺ groups in both AHI and CHI patients. There was no significant difference between the HIV⁺RPR⁻ and HIV⁺RPR⁺ groups in CHI patients (Figure 4D).

Different γδ T-cell subsets exhibit different effector functions. CD27⁻ Vγ₉Vδ₂ T cells are considered as the major producer of IL-17 (38, 39) and neutrophils can regulate IL-17 production by γδ T cells (40–42). Therefore, we analyzed the relationships between the frequencies of T_Naïve, T_CM, T_EM, and T_EMRA γδ T cells and the frequencies of neutrophils (Figure S1 in Supplementary Material). We found that the frequencies of T_EM γδ T cells were positively correlated with the frequencies of neutrophils. However, no relationships were observed between the frequencies of T_Naïve, T_CM, and T_EMRA γδ T cells and the frequencies of neutrophils (Figure S1 in Supplementary Material).

We have shown that γδ T cells were over-activated in HIV-1-infected patients, and associated with disease progression (26, 43). Here, we investigated the effects of syphilis on γδ T-cell activation by comparing γδ T-cell activation among HC, the HIV⁺RPR⁺ and HIV⁺RPR⁻ groups in both AHI and CHI patients. Consistent with our previous results, we found that the frequencies of CD38⁺γδ T cells were significantly higher in both the HIV⁺RPR⁺ and HIV⁺RPR⁻ groups in AHI and CHI patients, compared with HC. In AHI patients, the frequencies of CD38⁺γδ T cells in HIV⁺RPR⁺ group were significantly lower than that in HIV⁺RPR⁻ group (Figure 5A). In addition, we found that, in AHI patients, the frequencies of HLA-DR⁺ γδ T cells and CD38⁺HLA-DR⁺γδ T cells were significantly lower in the HIV⁺RPR⁺ group than in the HIV⁺RPR⁻ group (Figures 5B,C). By contrast, in CHI patients, the frequencies of HLA-DR⁺ γδ T cells and CD38⁺HLA-DR⁺γδ T cells were significantly higher in the HIV⁺RPR⁺ group than that in the HIV⁺RPR⁻ group (Figures 5B,C). Moreover, the frequencies of CD38⁺, HLA-DR⁺, and CD38⁺HLA-DR⁺γδ T cells in CHI patients, both HIV⁺RPR⁺ and HIV⁺RPR⁻ groups, were significantly higher than that in HIV⁺RPR⁺ group in AHI patients (Figures 5A–C). In a previous study, it was reported that neutrophils inhibited γδ T-cell activation (44), we therefore analyzed the relationship between γδ T-cell activation and the percentage of neutrophils. We found that there were no correlations between the frequencies of CD38⁺, HLA-DR⁺, and CD38⁺HLA-DR⁺ γδ T cells and the percentages of neutrophils (Figure S2 in Supplementary Material). Thus, HIV-1 infection appears to lead to γδ T-cell over-activation. However, syphilis can differentially affect γδ T-cell activation status, depending on the acute or chronic nature of the HIV-1 infection.

Previous studies have shown that IL-17 levels in the peripheral blood or central nervous system (CFS) increase in patients with asymptomatic neurosyphilis and secondary syphilis, possibly as part of the immune response to T. palladium infection (19, 20, 27, 45). γδ T cells are a major source of IL-17, which is involved in inflammation and immune response (46, 47). Therefore, we assessed and compared the frequencies of IL-17-producing and IFN-γ-producing γδ T cells in HC, and in the HIV⁺RPR⁺ and HIV⁺RPR⁻ patients with acute and CHI. We found that the frequencies of IL-17-producing γδ T cells were significantly higher in CHI patients than in HC, particularly for the HIV⁺RPR⁺ group (Figure 6A). Surprisingly, we found a significant difference in the frequencies of IL-17-producing γδ T cells between the HIV⁺RPR⁺ and HIV⁺RPR⁻ groups was observed in CHI patients, but not in AHI patients. Moreover, the frequencies of IL-17-producing γδ T cells in HIV⁺RPR⁺ group in CHI patients were significantly higher than that in both HIV⁺RPR⁺ and HIV⁺RPR⁻ groups in AHI patients (Figure 6A). In addition, the frequency of IFN-γ-producing γδ T cells was significantly lower in the HIV⁺RPR⁺ group in CHI patients than that in HC. The frequencies of IFN-γ-producing γδ T cells in HIV⁺RPR⁺ group in AHI patients were significantly higher than that in HIV⁺RPR⁺ group in CHI patients. However, there was no significant difference in the frequency of IFN-γ-producing γδ T cells between the HIV⁺RPR⁺ and HIV⁺RPR⁻ groups in both AHI and CHI patients (Figure 6B). Thus, both IL-17 and IFN-γ may be involved in γδ T-cell-mediated immune response to syphilis, which seems to depend on HIV-1 disease stage. Recent studies have shown that neutrophils can suppress γδ T-cell activation, proliferation, and IFN-γ production (44, 48). However, Coffelt et al. reported that neutrophils promoted IL-17 production by γδ T cells, leading to tumor metastasis (49). We therefore hypothesized that the differences in γδ T-cell functions in AHI and CHI patients might be associated with neutrophils. We analyzed the relationships between IL-17- or IFN-γ-producing γδ T cells and the percentage of neutrophils. We found that the frequencies of IL-17-producing γδ T cells were positively correlated with the percentage of neutrophils (Figure 6C). However, no correlation was found between the frequencies of IFN-γ-producing γδ T cells and the percentage of neutrophils (Figure 6D). Taken together, these results suggest that syphilis may lead to the recruitment of neutrophils to local sites, where they promote the production of IL-17 by γδ T cells, leading to inflammation, immune activation, and an acceleration of HIV-1 disease progression.