Differential proteins from EVs identification based on tandem mass tags analysis and effect of Treg-derived EVs on T-lymphocytes in COPD patients

Abstract

Background

Chronic obstructive pulmonary disease (COPD) is a widespread respiratory disease. This study examines extracellular vesicles (EVs) and proteins contained in EVs in COPD.

Methods

Blood samples were collected from 40 COPD patients and 10 health controls. Cytokines including IFN-γ, TNF-α, IL-1β, IL-6, IL-8, and IL-17, were measured by ELISA. Small EVs samples were extracted from plasma and identified by transmission electron microscope (TEM), nanoparticle tracking analysis (NTA), and Western blot. Protein components contained in EVs were analyzed by Tandem Mass Tags (TMT) to identify differential proteins. Treg-derived EV was extracted and added to isolated CD8⁺, Treg, and Th17 subsets to assess its effect on T-lymphocytes.

Results

ELISA revealed higher levels of all cytokines and flow cytometry suggested a higher proportion of Treg and Th17 cells in COPD patients. After identification, TMT analysis identified 207 unique protein components, including five potential COPD biomarkers: BTRC, TRIM28, CD209, NCOA3, and SSR3. Flow cytometry revealed that Treg-derived EVs inhibited differentiation into CD8⁺, CD4⁺, and Th17 cells.

Conclusion

The study shows that cytokines, T-lymphocyte subsets differences in COPD and Treg-derived EVs influence T-lymphocyte differentiation. Identified biomarkers may assist in understanding COPD pathogenesis, prognosis, and therapy. The study contributes to COPD biomarker research.

Introduction

Chronic obstructive pulmonary disease (COPD) is a condition distinguished by the presence of respiratory symptoms and airflow limitation that lasts for an extended period of time. Toxic particles, principally cigarette smoke, have been identified as the most significant risk factor [1, 2]. The underlying mechanisms are complex and include an array of immune responses and secondary chronic inflammation. Specifically, T lymphocytes, including CD8⁺ T cells, CD4⁺ T cells, and Th17 cells, play a pivotal role in COPD progression and exacerbation by inducing oxidative stress and promoting the release of proinflammatory factors [3, 4].

As previously studied [5, 6], the Th1/Th2 imbalance commonly occurs in COPD. Th1-mediated responses play a role in the development of organ-specific autoimmune disorders, leading to lung tissue injury, airway remodeling, and emphysema [4, 7]. Th17 cells are another subset of effector T cells, secreting multiple cytokines, such as IL-17, IL-6, and CXCL 1, in the lung. Meanwhile, Di Stefano et al. [8] demonstrated that Th17 cells and associated cytokines may participate in COPD airway inflammation and tissue remodeling.

Extracellular vesicles (EVs) generally refer to particles known for a phospholipid bilayer that are released from cells without self-replication ability [9]. Due its compositional complexity, EVs widely serve important roles in most biological systems. Small EVs, consisting mostly of exosome, was described as < 200 nm in diameter. It has been reported that EVs enable to regulate immune response [10].

Treg, a T cell subpopulation, secrets inhibitory cytokines such as IL-10, IL-35 and TGF-β, to inhibit the biological function of effector T cells [11]. In addition, Tregs are reported to inhibit the activation of autoreactive T and B cells, thus involving autoimmune disease [12,13,14]. In addition, Tregs have also been found to regulate effector T cells via paracrine or endocrine exosomes, a subset of small EVs [15]. Specially, Treg-derived exosomes enhance interactions with other immune cells through their diverse membrane surface proteins, thus regulating the immune response. Exosomes also drive inflammatory responses and enhance immune activation in COPD pathogenesis [16]. Thus, small EVs were considered potential biomarkers for diagnosing and grading COPD severity.

In this study, we identified differences in peripheral blood cytokines and proteins in small EVs between COPD patients and healthy humans. In addition, we investigated the effect of Treg-derived EVs from COPD on CD4⁺ T cells, CD8⁺ T cells, and Th17 cells. Moreover, 207 differential proteins were identified by Tandem Mass Tags (TMT) and 5 proteins were selected and validated as potential biomarkers in COPD.

Materials and methods

Patients and controls

This study enrolled a total of 40 patients diagnosed with COPD from patient respiratory medicine clinics, and 10 healthy humans from physical examination population. Subjects recruited for this research were all aged between 40 and 80, meeting the diagnostic criteria of the Global Initiative for Chronic Obstructive Lung Disease (GOLD) as defined by a post-bronchodilator forced expiratory volume in one second (FEV1) to forced vital capacity (FVC) ratio less than 0.7 [17]. However, subjects were excluded if they had a diagnosis of cerebrovascular, immunodeficiency, active infection, or had used inflammatory/immune response-affecting drugs, or any acute or chronic conditions limiting participation.

The control group consisted of 10 healthy adult between the ages of 40 and 80 with no history of respiratory, cardiovascular, or other chronic diseases, infectious diseases, or inflammation, and were not taking any medication that could affect their inflammatory or oxidative status. The time period for inclusion was from January 2022 to December 2022. The study was approved by the ShaoXing University Hospital Affiliated Committee on Ethics (2021(yan)NO. 13). The demographic information of patients were listed in Table S1.

ELISA

After the blood samples were collected, the samples were centrifuged at 1800 g for 10 min to obtain serum. The content of all indicators was determined by corresponding kits as manufacturer’s instructions.

The information of kits used in ELISA was listed as follow: IFN-γ (MM-0033H1), TNF-α (MM-0122H1), IL-1β (MM-0181H1), IL-6 (MM-0049H1), IL-8 (MM-1558H1), and IL-17 (MM-0180H1). All kits were purchased from Jiangsu Enzyme Immune Industrial Co., Ltd.

Flow cytometry

10 of COPD blood samples were used to detect lymphocyte immune phenotype and the other blood samples were stored at -80 ℃ for Treg-derived EVs extraction and T cell sorting.

Blood samples were added to an EP tube containing anticoagulant within 2 h and then transferred to a new EP tube containing PBS (1:1, v/v). Subsequently, the dilute blood sample was added to equal lymphocyte separation medium (P8610, Solarbio). After centrifugation at 800 g for 20 min, the lymphocytes in the middle were absorbed by the dropper and then, washed with PBS and contrifuged at 600 g for 10 min, twice. Subsequently, the supernatant was discard and 100 µL PBS was added to resuspend cells. 2.5 µL of corresponding antibodies were added to the cell suspension and then incubated at 4 ℃ in the dark for 30 min. After incubation, the cells were washed twice with PBS and resuspended with 200 µL of PBS. Before boarding the machine, the cells passed through a 200 mesh screen. Lymphocyte immune phenotype was detected by multi-parameter flow cytometry (NovoCytewas, Agilent).

The antibodies used in flow cytometry were as follow: CD4 antibody (sc-19641 PerCP, 1:20), CD8 antibody APC (sc-1177 APC, 1:20), ISG20 antibody FITC CD25 (sc-514979 FITC, 1:200), IL-17 PE antibody (sc-374218 PE, 1:200). All the antibodies were purchased from SANTA CRU.

Extraction and identification of EVs from plasma

Blood samples were placed in a collection tube containing EDTA and then centrifuged in 2000 g at 4 ℃ for 15 min to obtain plasma samples. Subsequently, plasma samples were gradually centrifuged in 2000 g at 4 ℃ for 30 min and then in 10,000 g at 4 ℃ for 45 min. Supernatant was filtered 0.45 μm membrane and then centrifuged in 100,000 g for 70 min at 4 ℃. After resuspend in 500 µL pre-cooled 1× PBS, the isolated samples were identified by transmission electron microscope (TEM), nanoparticle tracking analysis (NTA), and Western blot.

TEM: Dropwise apply 20 µL of sample onto a copper mesh, followed by stained with 10 µL uranium dioxide acetate (GZ02625, Zhongjingkeyi Tech.). Absorb the liquid with filter paper. Dry for a little while and image sample at 100 kV electron microscopy (HT-7700, Hitachi).

NTA: Samples were unfrozen and then diluted with 1× PBS. After calibration and focusing, the sample was added to nanoparticle tracking analyzer (ZetaVIEW, PARTICLE METRIX) to analyze particle trajectories and calculate the particle size distribution and concentration.

Western blot: 5× RIPA lysate (P0013B, Beyotime) was added to samples, mixed on ice for 30 min, then measured protein concentration with a BCA kit (P0012, Beyotime). Then, SDS-PAGE and PVDF transfer performed. PVDF member (10600023, GE Healthcare Life) sealed with 5% skim milk TBST. Primary antibody added, incubated 4 ℃ overnight and room temperature 1 h. Then visualized with ECL reagents using gel imaging system (ChemiScope 3000mini, CLINX).

The antibodies used in Western blot were as follow: GITR antibody (Ab264419, Abcam); TSG101 antibody (ab125011, Abcam); CD9 antibody (A1703, Abclonal); CD63 antibody (A5271, Abclonal); CD25 antibody (13517, CST); CD39 antibody (Ab300065, Abcam); CD73 antibody (Ab133528, Abcam); CTLA-4 antibody (Ab237712, Abcam); Goat anti-Rabbit IgG Peroxidase Conjugated (AP132P, Merck Millipore).

TMT quantitative proteomics analysis

The differential proteins components in small EVs were identified by TMT as previous studies [18,19,20]. Concentrations of protein components were detected by BCA kits (P0012, Beyotime). Subsequently, the protein samples were transferred to a centrifuge tube to unify the volume. Add 5 µL of 1 M dithiothreitol (43819, Sigma), and keep at 37 ℃ for 1 h; add 20 µL 1 M iodoacetamide (I1149, Sigma), and react at room temperature for 1 h, avoiding light; transfer all samples to the ultrafiltration tube and centrifugate, then discard supernatant. Add 100 µL UA (8 M urea, 100 mM Tris-HCl, pH 8.0) to the ultrafiltration tube, and discard the collection solution after centrifugation, repeating twice. Add 0.5 M TEAB (90360, Sigma) 100 µL, and discard the supernatant after centrifugation, and repeat three times; Update the collection tube, add trypsin (V5113, Promega; protein: enzyme = 50:1) to tube, and then enzymolysis at 37 ℃ for 12–16 h; use C18 Cartridge (WAT023590, Waters) to desalinate the peptide segment, and add 40 µL of 0.1% formic acid (FA; 5330020050, Sigma) solution after lyophilization. Peptides were appropriately taken from each sample peptide group and marked with kit from Thermo. Then, samples were analyzed by HPLC (Easy nLC/ Ultimate 3000, Thermo Scientific) and MS (Q Exactive HF-x, Thermo Scientific).

T test was used to determine significance with a P value. The BH method was applied to correct P value to obtain q value, set at 1.5 difference multiples (FC) and q < 0.05. Significance was determined as follows: FC ≥ 1.5 and q < 0.05 for up-regulation, FC ≤ 0.667 and q < 0.05 for down-regulation, and 0.667 < FC < 1.5 or q > 0.05 for no significant change.

The q_value was calculated as follow:

$$\:{\text{q}}_{\text{i}}=\frac{{p}_{i}\cdot m}{i}$$

Note: p_i, i^th hypothesis test’s original p_value; m, total number of hypothesis tests, i, sorted position of the i^th hypothesis test.

Identified differential proteins are compared with the EuKaryotic Orthologous Groups (KOG) database to predict the possible functions of these proteins and perform functional classification statistics [21].

All differential proteins were mapped to terms in the Gene Ontology database (http://www.geneontology.org/), and the number of proteins associated with each term was calculated [22]. Differential proteins were significantly enriched to GO terms.

KEGG pathway annotation and counting of identified differential proteins [23]. Both GO and KEGG analysis were dentified by hypergeometric tests, with P-value ≤ 0.05 as the threshold for significant enrichment.

The p_value was calculated as follow:

$$\:\text{p}=1-\sum_{\text{i}=0}^{\text{m}-1}\frac{\left(\frac{\text{M}}{i}\right)\left(\frac{\text{N}-\text{M}}{\text{n}-\text{i}}\right)}{\left(\frac{N}{\text{n}}\right)}$$

Note: N, total proteins number with annotation; n, differential proteins number; M, proteins number annotated to a certain entry; m, differential proteins number annotated to a certain entry.

Differential protein expression analysis

First, we identified potential targets for “COPD” or “Treg” from the Genecard database (http://www.genecards.org) [24] and OMIM database (http://omim.org/) [25], respectively. Related “COPD” or “Treg” targets were defined as the intersection of potential targets between Genecard and OMIM databases. Subsequently, the differential proteins identified by TMT analysis were converted into gene names byUniPort (https://www.uniprot.org/) [26]. Differential genes intersect with COPD and Treg potential target genes, respectively, to obtain potential key targets (Table 1). We selected five proteins, including three up-regulation and two down-regulation targets, that have been widely reported to be related to COPD or own multiple functions. Then, differential targets corresponding to these five proteins were selected as potential biomarkers for COPD. Subsequently, their expressions were measured by Western blot to validate the TMT analysis.

Table 1 Network pharmacology analysis

The antibodies used in this experiment were listed as follow: BTRC Antibody (DF6534, Affinity; 1:1000); NCOA3 Antibody (AF4055, Affinity; 1:1000); TRIM28 Antibody (Bioss, bs-3581R; 1:1000); CD209 Antibody (DF2295 Affinity; 1:1000); SSR3 Antibody (DF2778, Affinity; 1:1000); Anti-rabbit IgG, HRP-linked Antibody (7074, CST; 1:6000); Vinculin Antibody (AF5122, Affinity; 1:1000); GAPDH Antibody (10494-1-AP, Proteintech; 1:10000).

T cell sorted by magnetic bead and identification for purity

Magnetic bead method: The cells isolated from peripheral blood of COPD patients were divided into three samples and adjusted to 1 × 10⁷ per mL after resuspension in 80 µL buffer. Then 20 µL of CD4, CD8, and CD25 antibodies were added to each sample, respectively, followed by incubation at 4 ℃ for 15 min. Subsequently, each sample added 1–2 mL buffer and centrifuged at 300 g for 10 min, followed by adding 500 µL buffer and sufficient mixing.

Place the LS column in a MACS separator (130-042-302, Miltenyi Biotec) and rinse with 3 mL running buffer (130-091-211, Miltenyi Biotec); add the incubated cell suspension to the sorting column to collect unlabeled cells; wash the column with 3 mL buffer and repeat 3 times to collect unlabeled cells. Then, remove the column from the separator and place it on a suitable collection tube, flush the labeled cells with 5 mL buffer.

Microbeads used were as follow: CD4 MicroBeads, human (130-045-101); CD8 MicroBeads, human (130-045-201); CD25 MicroBeads II, human (130-092-983). All MicroBeads were purchased from Miltenyi Biotec.

Subsequently, T cells separated by magnetic beads were sorted by flow cytometer in mode purity as mentioned in Section “Flow cytometry” and then cultured with RPMI-1640 for further experiments. Before further experiments, cultured cells were verified for purity by flow cytometry.

Effect Treg-derived EVs on lymphocyte differentiation in vitro

Treg-derived EVs extraction: Isolated Tregs were collected to extract Treg-derived EV. After differential centrifugation, filtration, and ultracentrifugation, Treg-derived EV were isolated, followed by identification via TEM, NTA, and Western blot, as described in Section “Extraction and identification of EVs from plasma”.

T lymphocyte treated with EVs: After sorting and purity identification, isolated CD4⁺ T cells, CD8⁺ T cells, and Th17 cells were seeded into 6-wells plates, respectively. When the cell concentration reached 1 × 10⁵ per mL, cells were incubated with 1 µg/mL small EVs extracted from Tregs for 24 h, with a control group receiving no EVs. The frequency of CD4⁺ T cells, CD8⁺ T cells and Th17 cells were then detected by flow cytometry as mentioned in Section“Flow cytometry”.

Statistical analysis

Data was displayed as mean ± standard error. Normal distribution data was analyzed through ANOVA, followed by Turkey test for pairwise comparisons. For uneven variance, Dunnett’s T3 was applied. For non-normal data, Kruskal Wallis H test was instead. Significance was determined by p < 0.05.

Result

COPD alters peripheral blood cytokines and lymphocyte immunophenotype

As shown in Fig. 1, the ELISA results revealed that cytokines levels, including IFN-γ, TNF-α, IL-1β, IL-6, IL-8, and IL-17, in COPD patients peripheral blood were higher those in healthy humans (p < 0.05 or p < 0.01).

Fig. 1

Pro-inflammation cytokines increased in COPD patients. (A-F) The content of (A) IFN-γ, (B) TNF-α, (C) IL-1β, (D) IL-6, (E) IL-8, and (F) IL-17 in peripheral blood of COPD ($\:\text{n}=40$) and healthy control ($\:\text{n}=10$) groups. $^{\blacktriangle}$p<0.05, $^{\blacktriangle\blacktriangle}$p<0.01 vs. healthy control

Subsequently, we detected lymphocyte immunophenotypes, including CD8⁺ T cells, CD4⁺CD25^- T cells, CD4⁺CD25⁺ Tregs, and CD4⁺IL-17⁺ Th17 cells, to assess immune response alteration in COPD. Interestingly, there were no significant differences in CD8⁺ T cell counts between COPD and control groups (Fig. 2A). However, as shown in Fig. 2C-E, the proportion of other subsets of CD4⁺ T cells increased notably in COPD patients, with substantial increases in both CD4⁺CD25^- T cells and CD4⁺CD25⁺ Tregs (p < 0.01). Similar changes in Th17 cells were also observed.

Fig. 2

Lymphocyte immunophenotype alteration. (A) Flow quadrant diagram of CD8. (B) Proportion of CD8⁺ T cells (quadrant 1 + 2 in A). (C) Flow quadrant diagram of CD4 and CD25. (D-E) Proportion of (D) CD4⁺CD25⁺ T cells (quadrant 2 in C) and (E) CD4⁺CD25⁻ T cells (quadrant 3 in C). (F) Flow quadrant diagram of CD4 and IL-17. (G) Proportion of CD4⁺ IL-17⁺ T cells (quadrant 2 in F). $^{\blacktriangle}$p < 0.05, $^{\blacktriangle\blacktriangle}$p < 0.01 vs. healthy control

Identification of Treg-derived EV

Subsequently, we tried to extract EVs from peripheral blood. As shown in Fig. 3, both the Treg-derived EVs isolated from health control and COPD groups were identified by TEM, NTA and Western blot.

Fig. 3

Identification of small EVs extracted from peripheral blood. (A-B) The representative images of EVs in (A) healthy control and (B) COPD group under TEM. Magnification ×10k, scale bar 1 μm; Magnification ×20k, scale bar 500 nm; Magnification ×30k, scale bar 200 nm; Magnification ×60k, scale bar 100 nm. (C) The representative particle size distribution diagram of healthy control group. (D) Average particle size. (E) The representative particle size distribution diagram of COPD group. (F) Average particle concentration. (G) Protein bands for EV biomarkers ($\:\text{n}=5$). M, marker

Samples were observed as tea tray-like vesicles with double concave bilayer membrane structure (Fig. 3A and B). In addition, the NTA showed that the size distribution was within 200 nm, with an average diameter of 136.9 and 141.6 nm, respectively (Fig. 3C and D). Moreover, no significant difference in concentration was observed between the control and COPD groups (Fig. 3E and F).

Western blot revealed the expression of TSG101, a solute protein present in EVs, as well as CD63 and CD9, transmembrane proteins critical to EVs content sorting, confirmed as EVs. In addition, GM130 and GRP 94, representative impurity proteins derived from cells, were almost undetected in both control and COPD EVs (Fig. 3G). Interestingly, CD39, CD25, CD73, GITR, and CTLA-4, which are specific proteins associated with Tregs, were observed to be present in EVs. Notably, the CD73 blot showed a higher gray release in control samples compared to the COPD sample (Fig. 3G). However, this Western blot assay was a qualitative experiment with no internal reference. More precise assays are needed in the future to demonstrate differences in CD73 expression.

Analysis of differential protein components in EVs by TMT

TMT was used to identify differential protein components in EVs. A total of 207 differential proteins were identified. Of these, 118 were up-regulated, while 89 showed a decrease. All these protein alterations were graphically represented in a volcano plot and also organized into clusters via a heatmap (Fig. 4A and B). GO and KEGG enrichment analyses were then performed to enrich the differential expression genes in GO terms and KEGG pathways.

Fig. 4

Differential proteins analysis by TMT. Red represents up-regulated differential proteins, while green dots represent down-regulated differential proteins, with black representing no significant difference. (A) Volcano plot. (B) Heatmap. (C) Top 20 enriched GO term. (D) Top 20 enriched KEGG pathway. (E) KOG function classification

As shown in Fig. 4C, we noted that differential proteins enriched in extracellular matrix, collagen trimer and collagen-containing extracellular matrix, suggesting that differential proteins may involve biological function related to collagen and matrix. Notably, a high abundance of rich factors was observed in complement activation, lectin pathway, cell surface pattern recognition receptor signaling pathway, and positive regulation of opsonization, indicating that differential proteins may primarily influence immune response.

Interestingly, these differential proteins were found to be associated with immune-related diseases, such as Type I diabetes mellitus, autoimmune thyroid disease, allograft rejection, and systemic lupus erythematosus. In addition, immune pathways, such as phagosome, intestinal immune network for IgA production, and spliceosome, were also identified (Fig. 4D).

Furthermore, we classified the functions of differential proteins by cluster of Orthologous Groups of proteins (COG). As shown in Fig. 4E, the functions were mainly force in signal transduction mechanisms, extracellular structures, cytoskeleton, amino acid transport and metabolism and posttranslational modification, protein turnover, and chaperones, except for poorly characterized proteins. Interestingly, the functional classifications were almost consistent with the findings of the previously mentioned GO term or KEGG pathway.

Differential protein expression in COPD group and healthy control

Table 1 illustrated the specific targets and biomarkers that we sought to identify through network pharmacology. To support our results, we quantified the levels of 4 proteins related to both COPD and Tregs, including BTRC, NCOA, TRIM28, and CD209, as well as SSR3, a predicted down-regulation protein, in the peripheral blood of both COPD patients and healthy controls.

Western blot was used to assess the expression levels of these proteins (Fig. 5). Consistent with our TMT analysis, there were elevated levels of BTRC, TRIM28, and CD209, and reductions in NCOA3 and SSR3.

Fig. 5

Relative expressions of potential biomarkers. (A-E) The expression levels of (A) BTRC, (B) NCOA3, (C) TRIM28 ,(D) CD209, and (E) SSR3. (F) Protein bands. $^{\blacktriangle}$p < 0.05, $^{\blacktriangle\blacktriangle}$p < 0.01 vs. healthy control

Treg-derived EV promoted the differentiation of lymphocyte from COPD patients

As shown in Fig. 6, distinct T-lymphocyte subsets were identified by flow cytometry. The purity of CD8⁺ T cells, Th17 cells, and Tregs was 96.9%, 96.1%, and 95.8%, respectively, indicating successful isolation. Subsequently, isolated Tregs were used to extract EVs.

Fig. 6

Isolated T-lymphocytes identification. (A-C) The identification and purity of (A) CD8⁺, (B) Th17, and (C) Tregs

To study the effects of EVs on T cell differentiation, we added EVs as described above to the previously identified CD4⁺ T cells, CD8⁺ T cells, and CD4⁺ IL-17⁺ Th17 cells. As shown in Fig. 7, CD4⁺ T cells, CD8⁺ T cells, and CD4⁺ IL-17⁺ Th17 cells decreased post-treatment with small EVs.

Fig. 7

Differentiation in CD4⁺ T cells, CD8⁺ T cells, and Th17 cells. (A) Flow quadrant diagram of CD4 and CD8 in control and EV groups. (B-C) Proportion of (B) CD4⁺ T cells and (C) CD8⁺ T cells. (D) Flow quadrant diagram of CD4 and IL-17 in control and EV groups. (E) Proportion of Th17 cells. $^{\blacktriangle}$p < 0.05, $^{\blacktriangle\blacktriangle}$p < 0.01 vs. healthy control

Discussion

Our study revealed that peripheral blood cytokine levels significantly increased in COPD patients, as well as alterations in T-lymphocyte subsets. Further investigation determined that EVs secreted by Tregs were capable of influencing differentiation of T-lymphocyte subset. Through TMT analysis of protein components in EVs, 207 differential proteins were detected, with 5 proteins ultimately identified as potential biomarkers for COPD. These results were subsequently validated in COPD patients and healthy controls.

It was well known that immune system homeostasis in the respiratory tract was achieved by regulating pro-inflammatory and anti-inflammatory cytokines by subsets of CD4⁺ T cells [4]. For example, Th1/Th2 imbalance has also been observed in patients with COPD, as seen in previous studies [27,28,29,30]. Our research found that COPD patients have higher levels of IFN-γ and TNF-α, secreted by Th1 cells, indicating an increase in Th1. Similar to previous findings [27, 31], IFN-γ is known to inhibit the activity and function of Th2 cells. This further supports the existence of a Th1/Th2 imbalance in COPD. In addition, flow cytometry results revealed a substantial increase in Th17 cells and their related cytokines, including IL-17 and IL-6, in peripheral blood of COPD patients, indicating a high degree of transformation towards Th17 cells. This is consistent with flow cytometry results. Moreover, IL-8, more commonly known as CXCL8, serves as a key chemoattractant neutrophil infiltration [32]. Hitomi Fujie reported that IL-8 was up-regulated by pro-inflammatory cytokines, including IL-1β, TNF-α, and Th17 cytokine [33], which were also found to increase in our study. Moreover, IL-8 is also a potent inducers to stimulate bronchial epithelial cells and fibroblasts to produce IL-6. Both IL-8 and IL-6 enhanced mucus production, leading to tissue injury and COPD symptoms [34, 35].

Moreover, Tregs, a critical subset of T cells involved in immune response suppression through various mechanisms [13], have attracted substantial attention in many chronic inflammatory and autoimmune disorders [36]. However, the differentiation of Tregs exhibited notable variations depending on lung compartment assessed and COPD stages [37]. In general, although Tregs play a multifaceted role in COPD progression, they benefit COPD patients by maintaining the balance between Th17 and Tregs [4]. In our study, Tregs were higher in peripheral blood of COPD patients. We believed that this enhancement was due to the immune response, which was verified by the increased level of pro-inflammatory cytokines as mentioned above.

As in previous studies, EVs produced by immune cells play a crucial role in regulating immune responses [38]. For example, Treg-derived EVs have been shown to promote the differentiation of other T cells towards the Treg phenotype [39]. In our study, treatment with small EVs significantly inhibited differentiation to CD8⁺, CD4⁺ T cells and Th17 cells, suggesting that Treg-derived EVs in COPD patients can suppress immune response. However, this finding appears to contradict the fact that both CD4⁺, CD8⁺ and Th17 were higher in COPD patients. We hypothesized that the observed relationship might be linked to chronic inflammation in COPD patients, leading to an abnormal upregulation of the immune cells such as CD4⁺, CD8⁺ and Th17 [40]. Moreover, the difference in the effect on Treg-derived EVs between healthy donors and COPD patients may be one possible explanation for the contradictory results. For example, protein components, like CD9 and CD73, showed significant individual differences in Western blot assay. As Smyth reported, immunomodulatory molecules, including TLA-4, CD73, and GITR, contained in EVs can be secreted by activated Tregs, which were also identified in our study [41]. These molecules were also identified in our study. In previous studies, CTLA-4 was found to promote immune regulation by binding to B7, or by secretion of IL-10 to activate CD8⁺ Treg [42, 43]. Additionally, CD73 has been shown to enhance immune tolerance through the cAMP signaling pathway [44]. We therefore propose that these protein components in EVs play a critical role in immune regulation.

Subsequently, we revealed the potential role of biomarker proteins released from Tregs in COPD progression and prognosis through TMT technology. Through GO enrichment analysis, we noted that the biologic function of differential proteins is primarily mediated in immune response progression, such as complement activation and cell surface pattern recognition receptor, and collagen-containing extracellular matrix (ECM). Consistent with this, when analyzing the KEGG enrichment of this data, it is found that pathways enriched in autoimmune diseases are abundant. Furthermore, Tregs, which are known to secrete anti-inflammatory cytokines, have also been shown to induce differentiation of other T cells through small EVs, thus participating in immune regulation. Notably, based on the KOG function classification of these proteins, it was found that many were involved in the cytoskeleton and extracellular structure. This observation is consistent with GO analysis results, which indicated enrichment for collagen trim and extracellular matrix. When exposed to toxic particles, the major risk factor for COPD, fibroblasts are directly activated to produce collagen I, collagen III, and fibronectin. These findings are related to previous literature, specifically [45]. As a result, inflammatory cells, factors, and mediators work together to stimulate mucous cell formation, mucus secretion, fibroblast growth, and ECM protein synthesis. This process ultimately leads to constriction and remodeling of the airway, resulting in airflow limitation [46, 47].

Based on TMT identification results, five potential biomarkers were screened and validated in COPD patients. BTRC protein, also referred to as βTrCP, is an E3 ubiquitin protein ligase encoded by the BTRC gene. As Yan reported, BTRC is an important miRNA target that is involved in regulating many epithelial-mesenchymal transition molecules, thereby promoting fibrosis progression [48]. Furthermore, EVs derived from endothelial progenitor cells have been observed to express miRNAs that directly target BTRC mRNA, leading to lung tissue damage and immune suppression [49]. This may be why differentiation to CD8⁺, CD4⁺ T cells and Th17 cells was inhibited by EVs treatment in our study. TRIM28, a member of the tripartite motif-containing superfamily, functions as a transcriptional corepressor in cancer. Peng and colleagues reported that TRIM28 can activate autophagy in glioma, promoting cell proliferation [50]. Consistent with our study, the KEGG analysis suggests that the phagosome pathway, which regulates the interaction related to endosomes and lysosomes, is particularly enriched [23]. Furthermore, TRIM28 facilitates the recruitment of immunosuppressive myeloid-derived suppressor cells in non-small cell lung cancer, which subsequently suppresses CD8⁺ T cell activity and promotes resistance to anti-PD-1 therapy [51]. In addition, CD209 has also been shown to be a frequent and common marker for EVs in both sarcoidosis and hypersensitivity pneumonitis [52]. Moreover, researchers have suggested that NCOA3 has a significant role in regulating multiple pro-fibrotic transcriptional programs, and thus consider pharmaceutical inhibition of NCOA3 as a promising therapeutic approach for managing fibrotic diseases [53]. SSR3, a protein complex associated with translocations, functions as a glycosylated endoplasmic reticulum membrane receptor. Hu et al. conducted a comprehensive analysis and reported that SSR3 may be a potential oncogene in humans. They suggested that SSR3 may play a role in tumorigenesis and cancer immunity, and high levels of SSR3 expression have been linked to immunosuppression [54]. Notably, COPD patients had higher subsets of T lymphocytes and lower levels of SSR3 expression in our study. Based on these findings, we propose that these proteins may be potential biomarkers in COPD.

However, there were still some limitations in our studies. For example, our experimental design was not convincing enough to demonstrate the effect of Treg-derived EV on lymphocyte differentiation. For future research, we plan to conduct a native T lymphocyte model under specific conditions, which stimulate CD8 + or Th17 differentiation, to evaluate the suppression of Treg-derived EVs in COPD patients on T lymphocyte differentiation. Moreover, this study lacks a comparison between Treg-derived EVs from healthy donors and COPD patients. Future research will include an additional group treated with Treg-derived EV from healthy donors for a more comprehensive assessment.

In summary, we found a higher cytokine content and expression of subsets of T lymphocytes in COPD patients. In addition, we found specific surface proteins, such as CD39, CD25, CD73, GITR, and CTLA-4, in EVs. We determined that small EVs play a key role in promoting lymphocyte differentiation. In addition, 207 differential proteins were identified through TMT analysis and five potential COPD biomarkers were selected, including BTRC, TRIM28, CD209, NCOA3, and SSR3.