Active tuberculosis was diagnosed according to i) no previous history of TB; ii) positive acid-fast staining, positive Mtb culture, or positive Xpert Mtb/RIF test; and iii) no other lung diseases. The exclusion criteria of active TB were i) immunodeficiency, such as HIV, co-existing autoimmune diseases, or long-term steroid use; and ii) a history of anti-tuberculosis treatments, latent tuberculosis infection (LTBI), household contacts of newly diagnosed TB patients with positive interferon-gamma release assay (IGRA), and no evidence of clinical manifestations of active tuberculosis. Other diseases include pulmonary sarcoidosis, lung cancer, and other lung diseases.

A total of 50 patients with tuberculous pleurisy (15 women and 35 men; mean ± SEM age 44.78 ± 2.54 years, range 20 –82 years) were recruited from the Wuhan Pulmonary Hospital, Wuhan, China. A diagnosis of TPE was based on one of the following criteria: i) positive acid-fast staining, ii) pleural fluid or pleural biopsy specimens with growth of Mtb culture, iii) Mtb DNA testing positive, or iv) with clinical data highly suggestive of tuberculosis and pleural effusion improved after anti-tuberculosis treatment. Patients diagnosed with HIV or autoimmune diseases were excluded from the study. Malignant pleural effusion (MPE) refers to observed malignant cells in pleural effusion or pleural biopsy specimens. This study has been approved by the ethics committees of Wuhan Pulmonary Hospital, approval No. 2023 (43).

The pleural effusion samples were extracted using standard thoracic puncture techniques, and the peripheral blood samples were drawn simultaneously. Mononuclear cells were separated from the peripheral blood using Ficoll Paque (Tianjin HaoYang, Tianjin, China) gradient centrifugation within 24 hours after sampling, named pleural fluid mononuclear cells (PFMCs) and peripheral blood mononuclear cells (PBMCs), respectively. Then, the cells were re-suspended in the complete Roswell Park Memorial Institute (RPMI) 1640 medium.

A total of 1 × 10⁷ PFMCs were stimulated with lipopolysaccharide (LPS) at a concentration of 100 ng/mL, Bacillus Calmette-Guérin (BCG) at a ratio of 1:10 (PFMCs : BCG), and H37Ra at a ratio of 1:10 (PFMCs:H37Ra). After incubation at 37°C with 5% CO₂ for 24 hours, the total RNA was extracted using the TRIzol reagent (Invitrogen, Carlsbad, CA, USA). In addition, freshly separated PBMCs and PFMCs without any cultivation or stimulation were also extracted using the same method. The whole blood RNA was extracted using the QIAamp RNA Blood Mini Kit (QIAGEN, Valencia, CA, USA). Reverse transcription was performed using the kits manufactured by Applied Biosystems, Inc. (Foster City, CA, USA). Then, the PowerUp SYBR Green Master Mix kit was used for the quantitative real-time PCR. GAPDH was used as an internal control gene in the reaction system. The primer sequences are listed in Table 1 .

According to the protocol of the manufacturer, the concentrations of IL-32, TNF-α, IL-1β, IL-1Ra, IL-17, IL-6 (R&D, Minneapolis, MN, USA), and IFN-γ (BD, San Jose, CA, USA) in the cell culture supernatants, pleural effusion supernatants, or plasma of the participants were measured using the sandwich enzyme-linked immunosorbent assay (ELISA) method (R&D or BD Systems). In a 96-well plate, 200 μL of culture medium containing 5 × 10⁵ PFMCs was added alongside culture medium (control) or different stimuli: LPS (100 ng/mL), BCG (at a ratio of 1:10 with PFMCs), H37Ra (at a ratio of 1:10 with PFMCs), IL-32γ (50 ng/mL), TNF-α (20 ng/mL), IL-1Ra (20 ng/mL), or IFN-γ (20 ng/mL). After the cultivation at 37°C, the supernatant was removed in some experiments. Three freeze–thaw cycles were used to lyse the cells, then the intracellular cytokine concentrations were measured, and the results represented “cell-associated” IL-32 production; intracellular and secreted IL-32 concentrations were determined together, and the results represented “total” IL-32 production (9).

In order to clarify the cellular source of IL-32 production stimulated by Mtb, PFMCs were incubated at 37°C for 2 hours. The non-adherent cells that were transferred to another well were mainly lymphocytes, while the majority of adherent cells were monocytes and macrophages. The PFMCs, non-adherent cells, and adherent cells were stimulated with RPMI, LPS (100 ng/mL), BCG (at a ratio of 1:10 with PFMCs), and H37Ra (at a ratio of 1:10 with PFMCs). After 24 hours, the concentrations of “total IL-32” and IFN-γ were measured (9). In co-culture trans-well experiments, the non-adherent cells were added to the top well of the trans-well. Meanwhile, the adherent cells were plated in the lower compartment, stimulated with H37Ra, and co-cultured for 24 hours. Cells and medium were removed from the lower compartment and the top well, and the concentrations of “total IL-32” and IFN-γ were measured.

CD4⁺ T cells and CD8⁺ T cells were isolated by magnetic-activated cell sorting (MACS) using the CD4⁺ T-cell isolation kit and CD8⁺ T-cell isolation kit (Miltenyi Biotec, Bergisch Gladbach, Germany). Human IL-32 APC-conjugated antibody (No. IC30402A) was obtained from R&D Systems (Minneapolis, MN, USA). The CD4⁺ T cells and CD8⁺ T cells from patients with or without TPE were fixed, permeabilized, and stained for intracellular cytokine IL-32 before measurements were taken. Flow cytometry staining was performed using the BD FACS Canto II flow cytometry, and the data were analyzed using FlowJo software.

All analyses were conducted using GraphPad Prism version 9 or SPSS version 23. The normal distribution variables were expressed as mean ± standard error of the mean (SEM), while the non-normal distribution variables were expressed with the interquartile range (IQR). Pearson’s correlation coefficients were computed to assess the relationship between IL-32 and other variables. In addition, Student’s t-test or Mann–Whitney U tests were conducted to examine the differences between every two groups. When the p-value was less than 0.05, the difference was considered statistically significant.

In total, 50 patients aged 20 to 82 with TBP were enrolled in this study. Among the patients, the male-to-female ratio of cases was 7:3, with an average age of 44.8. The medians of the erythrocyte sedimentation rate (ESR) and C-reactive protein (CRP) in the peripheral blood of the TBP patients were 66 mm/h (IQR 38.5–75.5) and 69.00 (IQR 23.93–82.95) mg/L, respectively. The medians of lactate dehydrogenase (LDH), adenosine deaminase (ADA), and IL-32 in TBP were 403 (IQR 286.5–681) U/L, 41.9 (IQR 30.5–53.35) U/L, and 354.3 (IQR 198.42–529.2) pg/mL, respectively ( Table 2 ). The analysis revealed that the level of IL-32 in the TPE was positively correlated with the LDH (Pearson’s r = 0.286, p < 0.05) and ADA (Pearson’s r = 0.391, p < 0.01) in TPE. Additionally, the IL-32 level in TPE showed a positive correlation with CRP in the peripheral blood (Pearson’s r = 0.293, p < 0.05) ( Table 3 ).

The study analyzed the levels of IL-32 mRNA in active pulmonary tuberculosis, latent pulmonary tuberculosis, healthy individuals, and other diseases. Results showed that IL-32 mRNA expression was higher in latent tuberculosis patients compared to that in active pulmonary tuberculosis and healthy individuals. The IL-32 mRNA in active tuberculosis patients was lower than that in healthy individuals (all p < 0.05) ( Figure 1 ). In addition, compared with the other groups, in the group of other diseases (including pneumonia, malignant tumors, and various other lung infections, which need to make a differential diagnosis with TB), the expression of IL-32 mRNA was the lowest (p < 0.05).

Furthermore, the IL-32 expressions in the TPE, MPE, and transudative pleural effusion were checked. The average concentrations of IL-32 in TPE were significantly higher than in MPE and transudative pleural effusion (such as the history of heart disease) (all p < 0.05). There were no differences between MPE and transudative pleural effusion (p > 0.05) ( Figure 2 ).

The concentrations of IL-32 in the TPE and paired peripheral blood were measured, and the results indicated that the concentration of IL-32 in the TPE was significantly higher than that in the paired peripheral blood (p < 0.0001, Figure 3A ), and IL-32 mRNA in the PFMCs was significantly higher than that in the PBMCs (p < 0.01, Figure 3B ). Furthermore, IL-32α, IL-32β, and IL-32γ mRNA levels in the PFMCs were higher than those in the paired PBMCs (all p < 0.05), while no difference in IL-32δ was observed between the two groups (p > 0.05) ( Figure 3C ). The ratios of IL-32α/IL-32γ ( Figure 3E ) and IL-32β/IL-32γ ( Figure 3F ) in the PFMCs were higher than those in the PBMCs (p < 0.05). However, IL-32α/IL-32β ( Figure 3D ) and IL-32δ/IL-32γ ( Figure 3G ) did not differ between the PFMCs and PBMCs, indicating that the alternative splicing of IL-32γ into IL-32α and IL-32β was significantly enhanced in the PFMCs compared with the paired PBMCs. These data suggest that different IL-32 subtypes may have different roles and regulated patterns in the local TPE and peripheral blood immune inflammatory response in patients with TBP.

To further investigate the inductive effect of different Mtb stimuli on IL-32 production, freshly obtained PFMCs were stimulated with LPS, BCG, and H37Ra, and then the gene expressions and concentrations of IL-32 were determined. As shown in Figure 4A , the IL-32 mRNA expression increased after the addition of LPS, BCG, and H37Ra (all p < 0.05). Simultaneously, the production of IL-32 induced by the various stimuli was detected. Higher levels of total IL-32 production were observed in response to stimulation with H37Ra (227.4 ± 23.9 pg/mL, p < 0.001) and BCG (151.8 ± 22.6 pg/mL, p < 0.01), while LPS (71.6 ± 8.7 pg/mL, p < 0.001) induced a more moderate increase in IL-32 production ( Figure 4B ). Additionally, no IL-32 was detected in the culture supernatant of the PFMC or RPMI group, suggesting that the IL-32 levels in the culture supernatant of the PFMC and RPMI groups were even lower than the lowest value limit of the IL-32 cytokine detection kit.

As shown in Figure 5 , in the TPE, the levels of TNF-α ( Figure 5A ), IFN-γ ( Figure 5B ), and IL-1Ra ( Figure 5F ) showed a significant correlation with IL-32 levels (r = 0.3207, p < 0.05; r = 0.3172, p < 0.05; r = 0.3627, p < 0.01), while no significant correlation was observed for other cytokines such as IL-6 ( Figure 5C ), IL-1β ( Figure 5D ), and IL-17 (all p > 0.05) ( Figure 5E ).

To further evaluate the correlation between IL-32 and the above cytokines, PFMCs were stimulated with human recombinant protein IFN-γ, TNF-α, and IL-1Ra. As shown in Figure 6A , PFMCs stimulated with IFN-γ protein induced a 3.3-fold change in IL-32 mRNA expression (p < 0.05). However, TNF-α and IL-1Ra failed to stimulate the expression of IL-32 mRNA in PFMCs. In addition, IFN-γ enhanced the production of IL-32 significantly (p < 0.001, Figure 6B ), while IL-32 was not detected in the culture medium control group, TNF-α group, and IL-1Ra group.

Considering IL-32γ was the most active isoform (12, 13), PFMCs were treated with human reorganizing IL-32γ protein, and the results showed that the IL-32γ treatment positively induced the TNF-α mRNA expression ( Figure 7B ) and secretion ( Figure 7E ), and compared with the RPMI control group, the difference was statistically significant (p < 0.05). However, there was no difference in the expression and production of other cytokines between IL-32γ-treated PFMCs and untreated cells ( Figures 7A, C, D, F ).

PFMCs were separated into non-adherent cells and adherent cells. PFMCs, adherent cells, and non-adherent cells were stimulated with RPMI, H37Ra, BCG, and LPS for 24 hours. Compared to that in non-adherent lymphocytes, the concentration of “total IL-32” in stimulated adherent cells was significantly higher (p < 0.05) ( Figure 8A ). Compared with that in PFMCs, the total production of IL-32 by separated adherent cells and non-adherent cells was reduced (all p < 0.05) ( Figure 8A ). The concentration of IFN-γ in the stimulated PFMCs was significantly higher compared to either stimulated adherent or non-adherent lymphocytes (p < 0.01) ( Figure 8B ). In addition, no significant difference was detected as to the production of IFN-γ between stimulated adherent cells and non-adherent cells (p > 0.05) ( Figure 8B ).

To further verify whether the increase of IL-32 in adherent cells is related to direct cell–cell contact with non-adherent cells, trans-well co-culture assays were conducted. PFMCs, adherent cells, non-adherent cells, and co-cultured cells were separately stimulated with H37Ra for 24 hours. The results revealed higher IL-32 production in PFMCs compared to adherent cells and co-cultured cells (p < 0.05) ( Supplementary Figure 1A ). No significant difference was found as to IL-32 expression between adherent cells and co-cultured cells (p > 0.05) ( Supplementary Figure 1A ). Moreover, IFN-γ production by PFMCs was higher than that of adherent cells and co-cultured cells (p < 0.01), while no difference was observed between adherent cells and co-cultured cells in terms of IFN-γ levels (p > 0.05) ( Supplementary Figure 1B ).

Research has shown an increase in IL-32 expression in macrophages in TPE compared to non-TPE (14). Therefore, the expression of IL-32 was further compared within T cells between TPE and non-TPE. CD4⁺ and CD8⁺ T cells were purified from TPE and non-TPE, respectively. The flow cytometry results revealed that the expression of IL-32 was enhanced in the CD4⁺ T cells purified from TPE compared with the CD4⁺ T cells purified from non-TPE (p < 0.001) ( Figure 9A ). Similarly, IL-32 was also enhanced in the CD8⁺ T cells purified from TPE compared with that from non-TPE (p < 0.01) ( Figure 9B ).