The blood samples were collected from 27 SPPT, 37 SNPT patients and 36 individuals without infection (Ctrl), and their demographic and clinical characteristics are shown in Table S1 . The tuberculosis patients were recruited from the Tuberculosis Prevention and Treatment Institute of Kashgar, the Second People’s Hospital of Aksu, and the Kuqa County Infectious Disease Hospital. The Ctrl group were recruited from the First Affiliated Hospital of Xinjiang Medical University. Here, the inclusion criteria of TB patients included: 1) the people were diagnosed with tuberculosis based on clinical symptoms and microbiological evidence according to Diagnosis for Pulmonary Tuberculosis (WS 288-2017); 2) the patients provided the signed permissions for the use of their clinical data for scientific purpose and informed consent for the anonymous publication of data. The exclusion criteria included: 1) the TB patients in treatment period; 2) the TB patients with other chronic or acute diseases such as pregnancy complications, cardiac dysfunction, renal disease, psychiatric disease, gastrointestinal disease, uncontrolled hypertension, and some severe stress states (including cardiovascular and cerebrovascular events, severe infection, traumatic surgery, and severe wasting diseases). The control group were included according to the following criteria: no prior TB exposure, no specific clinical manifestations of TB, negative results on chest imaging, sputum smear test, and T-SPOT tests (Hu et al., 2022).

EDTA blood samples were collected from each participant and then centrifuged at 1500 g for 10 minutes at room temperature. The plasma was aliquoted and stored at -80°C. For protein preparation, high-abundance proteins were removed using the Agilent Human 14 Multiple Affinity Removal System according to the manufacturer’s instructions. Sample lysis and protein extraction were performed using SDT buffer (4% SDS, 100mM Tris-HCl, 1mM DTT, pH 7.6). The proteins were then digested overnight at 37°C with trypsin (Pierce, Thermo Fisher Scientific). The obtained tryptic peptides were desalted using C18 cartridges (Empore™ SPE Cartridges C18 (standard density), bed I.D. 7 mm, volume 3 mL, Sigma) and concentrated by vacuum centrifugation. Protein quantification was performed using the BCA Protein Assay Kit (Bio-Rad, USA). Protein integrity was confirmed by SDS-PAGE and silver staining, and samples showing protein degradation were excluded from the proteomic analysis. For metabolite preparation, the plasma samples were thawed on ice, and 150 μL of each sample was transferred to a centrifuge tube. Then, 1 mL of a chloroform-methanol solution was added, followed by ultrasound treatment. The supernatant was collected, and 2 mL of a 1% sulfuric acid-methanol solution was added. The mixture was methylated at 80°C for 30 minutes. Subsequently, 1 mL of n-hexane was added for extraction, and 5 mL of water was used for washing. Before injection, 500 μL of the extracted supernatant was mixed with 25 μL of an internal standard. This study was approved by the Hainan Province Clinical Medical Center.

The proteomic analyses were conducted by Shanghai Applied protein technology Co., Ltd (Shanghai, China) as previously reported protocol (Wei et al., 2018). The TMT 10plex labeling procedures (Thermo Fisher Scientific, San Jose, CA, US) were performed using the manufacturer’s instructions. Within each TMT experiment, three Ctrl samples were labeled with channels 126, 127N, and 127C. Three SNPT samples were labeled with channels 128N, 128C, and 129N. Three SPPT samples were labeled with channels 129C, 130N, and 130C. Additionally, a standard sample was created by mixing equal amounts of proteins from the nine samples and labeled with channel 131. For the labeling, each individual sample (100 μg) was resuspended in 100 μL of 0.1 M TEAB buffer, followed by reduction, alkylation, and trypsin digestion at 37°C for 16 hours. The reagents were dissolved in LC-grade acetonitrile (41 μL per 0.8 mg of reagent). After labeling for 1 hour, the reaction was quenched with 8 μL of 5% hydroxylamine for 15 minutes at room temperature. All channels were then mixed, aliquoted, and dried using a speed-vac. Subsequently, the TMT-labeled peptides were fractionated via high pH RP chromatography using a Zorbax 300 Extend-C18 column (3.5 μm, 4.6 × 250 mm; Agilent). LC-MS/MS analysis was performed on a Q Exactive mass spectrometer (Thermo Scientific) that was coupled to Easy nLC (Proxeon Biosystems, now Thermo Fisher Scientific) for 60/90 min.

The peptide mixture was loaded onto a reverse phase trap column (Thermo Scientific Acclaim PepMap100, 100 μm*2 cm, nanoViper C18) connected to the C18-reversed phase analytical column (Thermo scientific EASY column, 10 cm, ID75 μm, 3 μm, C18-A2) in buffer A (0.1% formic acid), and then separated with a linear gradient of buffer B (84% acetonitrile and 0.1% Formic acid) at a flow rate of 300 nL/min by Intelli Flow technology. LC–MS/MS analysis was operated in positive ion mode. Full MS scans were acquired in the mass range of 300-1800 m/z. Automatic gain control (AGC) target was set to 1e6, and maximum inject time to 50 ms. The dynamic exclusion duration was set for 60.0 s. Survey scans were acquired at a resolution of 70,000 at m/z 200 and resolution for HCD spectra was set to 35,000 at m/z 200, and isolation width was 2 m/z. Normalized collision energy was 30 eV and the underfill ratio was defined as 0.1%. Proteome Discoverer (v.1.4) was used to search all the Q Exactive raw data. False discovery rate (FDR) was set as ≤0.01. The protein ratios were calculated as the median of only unique peptides of the protein. All peptide ratios were normalized by the median protein ratio, and the median protein ratio should be “1” after the normalization.

LC-PRM/MS analysis were performed on 32 DEPs obtained from the TMT quantitative proteomic analysis (Shanghai Applied Protein Technology Co., Ltd, China) using the procedure described previously (Cao et al., 2022). Briefly, peptides were first prepared, with an internal standard reference of a Peptide Retention Time Calibration Mixture (Thermo Scientific). Tryptic peptides were desalted using C18 stage tips and subjected to reversed-phase chromatography on an Easy nLC-1200 system (Thermo Scientific) with 1-hour gradients. PRM analysis was performed on a Q Exactive HF mass spectrometer (Thermo Scientific) using optimized parameters for collision energy, charge state, and retention times. The mass spectrometer was operated in positive ion mode with specific parameters: full MS1 scan resolution of 70,000, AGC target values of 3.0×10^-6, and a maximum ion injection time of 200 ms. This was followed by 20 PRM scans at 35,000 resolution (at m/z 200) with AGC set to 3.0×10^-6 and a maximum injection time of 200 ms. Targeted peptides were isolated with a 1.5 Th window, and ion activation/dissociation was performed at a normalized collision energy of 27 using higher energy dissociation (HCD) in the collision cell. Data analysis was conducted using Skyline (MacCoss Lab, University of Washington), quantifying signal intensities for significantly altered proteins relative to each sample with normalization to a standard reference.

The metabolic properties of fatty acids obtained from proteomic analyses were then measured by targeted lipidomics using GC-MS analysis as previously described (Wen et al., 2020). In brief, the sample of metabolites were separated on an Agilent DB-WAX capillary column (30 m × 0.25 mm ID × 0.25 μm) gas chromatography system. The temperature programming was as follows: 50°C for 3 min, 220°C for 20 min, and 250°C for 10 min. An Agilent 6890N/5975B gas chromatography-mass spectrometer was used for analysis. The electron bombardment ionization (EI) source, SIM scanning mode, and electron energy were 70 eV.

We analyzed the differentially expressed proteins (|FC| ≥ 1.2, P-value < 0.05) performing pathway and functional annotation analysis using the Ingenuity Pathway Analysis (IPA, QIAGEN, USA) platform. We conducted core and comparison analyses to obtain canonical pathway and disease and biofunctions, which were evaluated by P-value (P-value < 0.05) and Z-score (Z-score ≠ 0). P-value < 0.05 indicated that the item was significantly enriched; Z-score > 0 indicated that the item was activated and Z-score < 0 indicated that the item was inhibited. Statistical analyses were performed using R software v4.1.0, and P-value < 0.05 indicated statistical significance.

To explore the molecular mechanisms of SPPT and SNPT patients, we performed TMT quantitative proteomic analyses on plasma samples from 27 SPPT, 37 SNPT patients, and 36 Ctrl individuals ( Table S1 ). The results displayed distinct protein expression profiles among the three groups, and 161 differentially expressed proteins (DEPs) were identified (|FC| > 1.2, P-value < 0.05) ( Figures 1A, B ; Table S2: Additional Table 1 ). Compared with Ctrl, 44 up-regulated and 50 down-regulated DEPs were obtained in SNPT, and 38 up-regulated and 82 down-regulated DEPs were detected in SPPT ( Figure 1C ); 30 and 55 DEPs were specially expressed in the SNPT/Ctrl and SPPT/Ctrl groups, respectively ( Figure 1D ). Some of them were further verified by PRM targeted proteomic mass spectrometry ( Figure S1 ), which displayed a consistent pattern obtained from TMT analysis.

We then analyzed the top 20 up- and down-regulated DEPs in the SNPT/Ctrl and SPPT/Ctrl groups ( Figure 2 ). Of the top 20 up- and down-regulated DEPs, eight and 11 were shared by the two groups ( Figures 2A, B ).

For the top 20 up-regulated DEPs ( Figure 2A ; Table S3 ), seven of eight shared DEPs were associated with inflammation/immune response (CRP, APMAP, C3, B2RA39, PF4, LBP and IGFBP2), indicating increased inflammatory response in TB patients regardless of SNPT or SPPT. Further analysis about 12 specially up-regulated DEPs showed the stronger immune response in SPPT than in SNPT: seven of the 12 specifically upregulated DEPs in the SPPT/Ctrl group were associated with innate immune response (ORM2, ORM1, P0DOX2, SIRPB1, LILRA2, GOLM1, and MBL2), while five of the specifically upregulated DEPs in the SNPT/Ctrl group were related to anti-inflammation (FGL1, CHIT1, LRG1, SERPINA3, and B4DLL8).

For the top-20 down-regulated DEPs ( Figure 2B ; Table S3 ), nine of 11 shared DEPs in the two groups were involved in metabolism and antioxidant stress (Q53H26, B4DF70, CA1, APOA1, S100A4, LTF, AK1, B2RAN2 and S100A6), demonstrating decreased metabolism and antioxidant capacity in TB patients. On the other hand, in the SNPT/Ctrl group, seven out of nine DEPs that were specifically down-regulated were also related to metabolism and antioxidant stress (HEL-S-282, BLVRB, B4DM82, CNDP1, APOA4, NME1, and TPI1). This observation suggested a decrease in metabolism and antioxidant capacity in SNPT patients. Among the nine specifically down-regulated DEPs in the SPPT/Ctrl group, four were related to lipid metabolism (KRT17, B2R8I2, ATP8B3, and PI16), and the remaining five were associated with injury/remodeling processes (F6KPG5, CALM2, ITIH1, ZBTB18, and B2R582). These findings indicated an inhibition of lipid metabolism and damage repair processes in SPPT patients.

Additionally, we detected six up-regulated and eight down-regulated DEPs in the SPPT/SNPT group. For the up-regulated DEPs, five were associated with inflammation/immune responses; for the down-regulated DEPs, five were related to metabolism, two of which were associated with lipid metabolism (PGLYRP2 and KNG1) ( Table S4 ). These reflected disordered immune responses and metabolism compared to SNPT.

We then classified all DEPs into six expression patterns based on the protein expression trends among the three groups ( Figure 3A ; Table S5, S6 ), and further focused on the top10 proteins in the LMH and HML patterns with continuously up- or down-regulated expression trends from Ctrl to SNPT and then to SPPT ( Figure 3B ). Of the top10 proteins in the LMH pattern, six were related to innate immune regulation (ORM2, ORM1, CRP, FGL1, P0DOX2, and PF4), indicating continuously enhanced body inflammatory response as increasing bacterial loads. Conversely, among the top 10 proteins of HML pattern, four were related to lipid metabolism, including APOA1, KRT17, B2R8I2, and ATP8B3, demonstrating continuously increased lipid metabolism disorder with the increase of bacterial loads.

For the LHM and MHL patterns (with higher expressed proteins in SNPT compared to SPPT) ( Figure S2 ; Table S6 ), six out of 13 proteins (~46%), were related to anti-inflammatory responses, indicating their crucial role in protecting the body from excessive inflammatory responses in SNPT. For the 15 DEPs in the HLM and MLH patterns (with lower expressed proteins in SNPT compared to SPPT), four were associated with immune response (B2R6V9, DDT, B2RBZ5, and S6B294), and 10 were related to metabolism and antioxidant stress ( Figure S2 ; Table S6 ).

To investigate the molecular mechanism of SPPT and SNPT, we further performed pathway and functional analyses using IPA. First, 40 significantly enriched pathways (P-value < 0.05) were identified in the SNPT/Ctrl and SPPT/Ctrl groups ( Table S7, Table S2: Additional Table 2 and Additional Table 4 ). Among the top 10 shared significantly enriched pathways in the two groups ( Figure S3 ), five were related to immune response (“Acute Phase Response Signaling”, “Production of Nitric Oxide and Reactive Oxygen Species in Macrophages”, “IL-12 Signaling and Production in Macrophages”, “Clathrin-mediated Endocytosis Signaling”, and “Granulocyte Adhesion and Diapedesis”), and three were associated with lipid metabolism (“LXR/RXR Activation”, “FXR/RXR Activation”, and “Atherosclerosis Signaling”). These indicated dysfunctional immune response and lipid metabolism in the SNPT and SPPT patients.

500 significantly enriched biofunctions (P-value < 0.05) were then identified in the SNPT/Ctrl group and SPPT/Ctrl groups ( Table S8 , Table S2 : Additional Table 3 and Additional Table 5 ), among which 40 were significantly activated/inhibited in at least one group (|Z-score| ≥ 2, Figure 4A ). Herein, 21 of the 40 functions (more than half) belonged to immune related functions ( Figure 4A ), all of which were significantly activated (Z-score ≥ 2) in at least one group; fatty acid metabolism was significantly inhibited in the SPPT/Ctrl group. IPA biofunction analysis also showed dysfunctional immune and fatty acid metabolism in SNPT and SPPT.

We further explored fatty acid metabolism in SPPT and SNPT patients, and found gradually inhibition of fatty acid metabolism from Ctrl to SNPT then to SPPT ( Figure 4A ). Among eight enriched lipid metabolism related functions in the two groups (P-value < 0.05), it was worth noting that only one lipid metabolism related function (“Fatty acid metabolism”) was significantly inhibited (Z = -2.177) in the SPPT/Ctrl group, while it was inhibited (Z = -1.052) in the SNPT/Ctrl group ( Figure 4B ). This indicated more serious inhibition of fatty acid metabolism in SPPT than in SNPT, as previously reported in our previous metabolic findings about SPPT and SNPT (Hu et al., 2022).

To verify the above findings, we further performed targeted lipidomics using GC-MS analysis (Wen et al., 2020). The results indeed showed more severe inhibition of lipid metabolism in SPPT than in SNPT, especially for unsaturated fatty acids ( Figure S4 , Table S2 : Additional Table 6 ). In the SPPT/Ctrl group, eight out of 17 fatty acid molecules showed downregulation, among which six displayed significant downregulation with P-value < 0.05, including two polyunsaturated fatty acids (PUFA); in the SNPT/Ctrl group, most fatty acid molecules (14/17 = 82.4%) showed up-regulation, among which nine displayed significant upregulation with P-value < 0.05, including six unsaturated fatty acid molecules.

To explore the immune mechanism of SPPT and SNPT, we focused on 21 immune-related biofunctions, containing 19 innate- and two acquired-immune related ones ( Figure 4A ). In the SNPT/Ctrl group, all 11 immune-related biofunctions (including two shared and nine specifically enriched ones) were associated with migration/movement and binding/adhesion of immune cells, among which two specifically enriched ones belonged to acquired immune-related biofunctions and were related to antigen-presenting cell movement. In the SPPT/Ctrl group, five of 12 immune-related functions (including two shared ones) were associated with migration/movement and binding/adhesion of immune cells. The other ten specifically enriched items were related to movement, chemotaxis, infiltration and even death of immune cells, suggesting stronger innate immune response in SPPT patients.

Notably, in the SNPT/Ctrl group, two acquired immune-related biofunctions, “Cell movement of antigen presenting cells” (P-value < 0.05, Z-score = 2.084) and “Migration of antigen presenting cells” (P-value < 0.05, Z-score = 2.179) were significantly activated ( Figure 4A ), while they were not significantly activated in the SPPT/Ctrl group (Z-score = 0.576 and Z-score = 0.478). In addition, we also detected three down-regulated antigen presenting cells related functions in the SPPT/Ctrl group [(“Activation of antigen presenting cells” (Z-score = -1.028); “Phagocytosis of antigen presenting cells”, Z-score = -1.577; “Immune response of antigen presenting cells”, Z-score = -1.482)] ( Figure 4C ). A significantly down-regulated inducible T cell co-stimulator ligand/key molecule of activating T cells (ICOSLG/LOC102723996) (P-value < 0.05, FC = 0.82) were also detected in SPPT patients. The above findings suggested that antigen-presenting cell activation might be a critical determinant of different bacterial loads/symptoms between SNPT and SPPT.

To further investigate antigen-presenting and acquired immunity in SNPT and SPPT, we analyzed all enriched acquired immune-related functions and corresponding DEPs ( Figure 5 ; Table S9 ). The results showed stronger activation of antigen presenting and acquired immune related functions/proteins in SNPT than in SPPT. Herein, a total of 17 functions related to antigen-presenting and acquired immunity were identified in the two groups, including six shared ones, four specifically enriched ones in SNPT, and seven specifically enriched ones in SPPT.

First, among the six shared functions, all of them showed stronger activation or weaker inhibition of antigen presenting and acquired immune in SNPT than in SPPT. For antigen-presenting, in addition to the two significantly activated ones in SNPT (described in the previous section), “Activation of antigen-presenting cells” function was activated in SNPT but inhibited in SPPT, and the inhibition degree of “Phagocytosis of antigen-presenting cells” function was lower in SNPT than in SPPT. Secondly, among the four significantly enriched functions in SNPT, half of them were related to antigen-presenting and activated. Thirdly, among the seven significantly enriched functions in SPPT, most of them were inhibited, including “Immune response of antigen-presenting cells”. Overall, compared to SNPT, the inhibition of antigen-presenting cells (APCs) activation in SPPT might cause impaired acquired immunity, subsequently resulting in a higher bacterial load in SPPT than in SNPT.

Further analysis revealed that some proteins might be associated with differential antigen-presenting and acquired immune responses between the two groups. First, a total of 45 DEPs were enriched in the aforementioned APC/acquired immune related functions. Among them, 15 were shared in the two groups and exhibited consistent expression trends in both SNPT and SPPT patients, which were therefore not subjected to further analysis. The remaining 30 proteins were specifically enriched in the acquired immunity-related functions of either SNPT or SPPT, suggesting that they might play crucial roles in differential genotypes and phenotypes between SPPT and SNPT.

For seven specifically enriched proteins in SNPT, four of five up-regulated proteins (SPP1, CHI3L1, APOE, and NRP2) have been reported to be involved in promoting T cell activation and facilitating pathogen clearance (van den Elzen et al., 2005; Dela Cruz et al., 2012; Shan et al., 2014; Roy et al., 2017); one of two down-regulated proteins (MPO) has been reported to be associated with inhibition of antigen presenting cell activation (Odobasic et al., 2013).

For 23 specifically enriched proteins in SPPT, six of seven up-regulated proteins (SOD2, LILRA2, IGFBP2, FCGR3A/FCGR3B, TIMP1, and ICAM1) were associated with suppression of antigen presentation and T-cell activation, and upregulation of Treg(Marchesi et al., 2013; O'Mahony et al., 2016; Seo et al., 2019; Keller et al., 2021; Robilliard et al., 2022; Zhang et al., 2022). Here, Treg cells have been reported to play roles in suppressing a variety of immune cell activation, including B cells, CD4+, and CD8+ T cells (Seo et al., 2019). On the other hand, nine of 16 down-regulated proteins (ICOSLG/LOC102723996, ATRN, CDH13, ECM1, ARHGDIB, LECT2, IGF2, FN1, and PON1) have been reported to play roles in antigen presenting and T cell activation(Tang et al., 2000; Chitnis and Khoury, 2003; Zhang et al., 2006; Jodaa Holm et al., 2016; Bai et al., 2018; Bijen et al., 2018; He et al., 2018; Luo et al., 2019; Belfiore et al., 2023).