Serum proteomics links the cardiorespiratory biomarkers CTRC, OSM, and MMP-10 to exacerbation severity and number in patients with COPD Open Access

Chronic obstructive pulmonary disease (COPD) is a common lung disorder characterized by chronic airflow limitation, local and systemic inflammation, poor quality of life, and high mortality [1]. A key risk factor for COPD is long-term tobacco smoking, and the typical inflammatory profile associated with COPD encompasses local accumulation of cytotoxic T cells, alveolar macrophages, and neutrophils, as well as a systemic increase in circulating neutrophils [1]. Unfortunately, the precise mechanisms driving COPD are still poorly understood, and the available pharmacological therapies can only reduce symptoms and risk of progression or death in a modest manner. Given that exacerbations of COPD constitute a strong risk factor for poor prognosis, it seems feasible that further study of this complication may shed light on key pathogenic mechanisms [2].

Historically, research on COPD has centered on two common pulmonary comorbidities: emphysema and chronic bronchitis (CB). Nevertheless, a growing body of literature now indicates that cardiovascular disease (CVD) also plays an important role in the morbidity and mortality of patients with COPD. In fact, CVD is up to five times more likely to occur in patients with COPD compared with the general population [3], and CVD is a common cause of death among these patients [4]. Furthermore, the risk to experience and die of CVD may be enhanced up to four times during exacerbations in patients with COPD [5].

Given the large number of patients with COPD and the limited clinical resources available, numerous attempts have been made to identify easily accessed biomarkers that can help define and prioritize patients with a high risk for poor prognosis. Along these lines, C-reactive protein (CRP) in blood has been used to assess systemic inflammation in COPD [6], while several cytokines involved in neutrophil mobilization (e.g. IL-6, IL-8, and IL-17) have been used as local biomarkers of disease severity and risk for CVD in COPD patients [7-11]. Nevertheless, most biomarkers have been characterized in patients with stable COPD, who are easier to study, while the literature on biomarkers for exacerbations of COPD remains limited. To date, fibrinogen in plasma emerges as one of the best-documented protein biomarkers for the risk of exacerbations in COPD patients [12-14]. However, a single biomarker is unlikely to reflect all important aspects of COPD pathology, and the interpretation of a single protein in any body fluid can be confounded by common inflammatory conditions beyond COPD. Therefore, we think that there is a strong rationale for new studies aiming to identify novel biomarker signatures for critical aspects of COPD, including exacerbations and the risk of CVD.

In the present study, we aimed to identify systemic alterations in biomarkers of CVD during COPD exacerbations, using biomarkers of inflammation as a reference. For this purpose, we used a proteomic approach to assess 177 biomarkers in serum samples from a group of long-term smokers (LTS) with COPD and CB (COPD-CB). In this group, we performed longitudinal comparisons between paired samples collected during stable disease and an exacerbation, respectively. In addition, we performed cross-sectional comparisons between this COPD-CB group during stable disease and two control groups: clinically healthy non-smokers (HNS) and LTS without COPD and CB.

The patient material used in this study derives from a larger material that has previously been described in detail [9,15,16]. In brief, LTS with COPD-CB were recruited via advertisement and from the outpatient clinic of the Department of Respiratory Medicine of the Sahlgrenska Hospital, Gothenburg, Sweden. During the inclusion visit, we confirmed that all LTS with COPD-CB had an FEV1/FVC ratio of less than 0.7 (COPD) and had experienced a productive cough for at least three months per year for two consecutive years (CB). The exclusion criteria encompassed the following: (I) any additional lung diseases other than COPD-CB; (II) allergic sensitization to one or more common aeroallergens (Phadiatop^TM); (III) regular use of oral glucocorticoids; (IV) long-term oxygen therapy; (V) drug abuse; (VI) cancer; (VII) severe mental illness besides well-treated depression; (VIII) chronic inflammatory diseases such as rheumatoid arthritis and inflammatory bowel disease; and (IX) congestive heart failure or coronary artery disease. The aeroallergens screened for were plant pollens (birch, timothy grass, mugwort, olive, and pellitory), pet dander (dog, cat, and horse), house dust mite, and Cladosporium (mold). Serum samples were collected during an inclusion visit and every 15th week for 60 weeks. In addition, serum and sputum samples were collected whenever a patient with COPD and CB experienced an acute exacerbation, which was diagnosed according to established criteria [17,18]. In brief, patients experiencing dyspnoea and/or cough and sputum worsening less than 14 days were judged as having an exacerbation. This exacerbation sample was collected prior to emergency treatment. Moreover, a group of HNS and a group of LTS without COPD and CB were also recruited and used as control groups. All participants underwent a urine cotinine test to confirm smoking status, and long-term smoking was defined as more than 10 pack-years.

In the current study, we analysed paired serum samples from LTS with COPD-CB collected during a stable COPD visit and an exacerbation visit, respectively. The majority of samples from stable COPD (20 out of 26) were collected during the inclusion visit. Because six included patients experienced an exacerbation during their inclusion visit, their matched samples from stable disease were harvested 15 weeks after their exacerbation pair. Finally, all LTS with COPD-CB included in this study experienced at least one exacerbation during the course of the study, and all exacerbation samples used in this study correspond to the first exacerbation of each patient. Exacerbations were classified as mild, moderate, or severe according to GOLD criteria [1]. In brief, mild exacerbations only required treatment with short-acting bronchodilators in the outpatient clinic, moderate exacerbations required treatment with oral glucocorticoids and/or antibiotics in the outpatient clinic, and severe exacerbations required hospitalization. Importantly, no patients included in this study experienced pneumonia during their inclusion visit, exacerbation visit, or in between these two visits.

Commercially available panels (Olink^® Target 96 Cardiovascular II and Inflammation panels) were used to measure protein biomarkers in serum samples by proximity extension assay [19] at the facilities of Olink^® proteomics (Uppsala, Sweden). In this method, two antibodies labeled with complementary DNA oligonucleotides bind a target protein in solution, and the close proximity allows the hybridization of the DNA oligonucleotides. The hybridized DNA labels are then extended by a DNA polymerase to generate a unique DNA barcode that can be amplified and detected via next-generation sequencing (NGS). In brief, this high-throughput method combines the specificity of antibody-based immunoassays with the powerful amplification potential of polymerase chain reaction (PCR) and next-generation sequencing (NGS). Each panel detected 92 proteins (Supplementary Tables 1 and 2) and there were seven proteins that overlapped between the two panels for a total of 177 unique proteins. Results are reported as normalized protein expression (NPX), an arbitrary unit in the log base 2 scale.

Samples were randomly assigned to wells in 96-well plates before delivery to Olink^® proteomics for analysis. Notably, each sample was assigned a codename before analysis to guarantee blind assessment.

Paired and unpaired comparisons between the two groups were assessed by paired and unpaired Student’s t-test, respectively. Multiple comparisons were performed via ANOVA followed by Tukey’s post hoc test to compare each group against the others. Correlations were determined by Spearman’s rank test. P values were corrected for multiple testing using the Benjamini–Hochberg method. Statistical analyses of proteomics data were done using Python 3·10·9, while all other statistical analyses were performed in Prism 9.3 (GraphPad).

The main study group consisted of 26 LTS with COPD and CB (COPD-CB group), while the two control groups consisted of 10 HNS and 10 LTS without COPD and CB, respectively. The main characteristics of all study participants during stable clinical conditions are summarized in Table 1 (see also “Methods” for more information). Although the median age was lower in the LTS control group when compared with the other groups (LTS vs COPD-CB: P=0.03 & LTS vs HNS: P=0.002 determined by Kruskal–Wallis test followed by Dunn’s multiple comparison test), the median number of pack-years was similar between the LTS group and the COPD-CB group (P=0.09 determined by Mann-Whitney test). In addition, the main clinical characteristics of LTS with COPD-CB during exacerbation are summarized in Table 2. In general, most exacerbations included in the present study (20 out of 26) were moderate, and only one exacerbation required hospitalization.

We found that the serum concentration of chymotrypsin C (CTRC) was decreased, while the corresponding concentrations of oncostatin M (OSM) and matrix metalloproteinase 10 (MMP-10) were increased, during exacerbation in the COPD-CB group (Figure 1A–C). We did not detect any statistically significant alteration in the serum concentrations of the other 174 proteins investigated during exacerbation. Notably, the serum concentrations of CTRC, OSM, and MMP-10 during stable disease in the COPD-CB group resembled those of the HNS and LTS control groups (Supplementary Figure S1). Furthermore, we did not detect any sex-related differences in the serum concentrations of CTRC, OSM, and MMP-10 (Supplementary Figure S2), and we found that having cardiovascular comorbidities at the time of inclusion did not have an impact on the serum concentrations of these three biomarkers (Supplementary Figure S3).

Given that respiratory infections are a major cause of COPD exacerbations, we determined whether having pathogenic bacteria in sputum cultures obtained during exacerbations influenced the serum concentrations of CTRC, OSM, and MMP-10. Notably, patients with pathogenic bacteria in sputum cultures had a lower serum concentration of CTRC during exacerbations than patients without these bacteria (Figure 2A). In contrast, the serum concentrations of OSM and MMP-10 during exacerbation were similar between patients with and without pathogenic bacteria in sputum cultures (Figure 2B and C).

We analysed arterial blood oxygen in all patients of the COPD-CB group who had a transcutaneous oxygen saturation below 94% during exacerbation. Interestingly, CTRC and OSM displayed strong positive and negative correlations, respectively, with the relative arterial oxygen saturation (Figure 3A and B) and with the arterial partial pressure of oxygen (Figure 3C and D). However, MMP-10 did not correlate with either of those parameters (Supplementary Figure S4).

To further characterize the clinical relevance of CTRC, OSM, and MMP-10 during exacerbation, we assessed the relationship between these proteins and the blood concentrations of CRP and leukocytes. On one hand, a decrease in the serum concentration of CTRC during exacerbation is associated with higher blood concentrations of CRP, monocytes, and neutrophil elastase (Figure 4A–C). On the other hand, an increase in the serum concentration of OSM during exacerbation is associated with higher blood concentrations of CRP, neutrophils, and myeloperoxidase (Figure 4D–F). Notably, MMP-10 did not correlate with any of the biochemical parameters summarized in Table 2 (not shown).

To further characterize the clinical relevance of CTRC, OSM, and MMP-10 during exacerbation, we assessed the relationship between these proteins and the blood concentrations of CRP and leukocytes. On one hand, a decrease in the serum concentration of CTRC during exacerbation is associated with higher blood concentrations of CRP, monocytes, and neutrophil elastase (Figure 4A–C). On the other hand, an increase in the serum concentration of OSM during exacerbation is associated with higher blood concentrations of CRP, neutrophils, and myeloperoxidase (Figure 4D–F). Notably, MMP-10 did not correlate with any of the biochemical parameters summarized in Table 2 (not shown). Serum concentrations of CTRC and OSM display strong correlations with those of other cardiovascular biomarkers during exacerbations.

We investigated the association between the serum concentrations of CTRC, OSM, and MMP-10 during exacerbations and those of the other cardiovascular biomarkers listed in Supplementary Table S1. While the serum concentration of CTRC only displayed a strong correlation with brother of CDO (BOC; Figure 5A), the serum concentration of OSM displayed strong and positive correlations with six other cardiovascular biomarkers (Figure 5B–G): interleukin-1 receptor antagonist (IL-1ra), lectin-type oxidized low-density lipoprotein receptor 1 (LOX-1), IL-6, pentraxin 3 (PTX3), chemokine ligand 1 (CXCL1), and carcinoembryonic antigen-related cell adhesion molecule 8 (CEACAM8). In contrast, the serum concentration of MMP-10 during exacerbation failed to display any statistically significant correlation with those of other cardiovascular markers tested. (not shown).

We also determined whether the serum concentrations of CTRC, OSM, and MMP-10 measured during stable disease in the COPD-CB group were associated with the number of exacerbations that occurred during the study. Notably, the concentration of MMP-10 was higher during stable disease among patients in the COPD-CB group who had two or more exacerbations compared with those who had one exacerbation only (Figure 6A). Moreover, in this group, the concentration of MMP-10 during stable disease displayed a moderate positive correlation with the number of exacerbations (Figure 6B). In addition, patients with a blood eosinophil count ≥ 0·3 × 10⁹/L, who are known to have a higher risk for COPD exacerbations, also had a higher concentration of MMP-10 during stable disease (Figure 6C). Notably, neither CTRC nor OSM displayed any statistically significant associations or differences regarding the number of exacerbations or blood eosinophil count in the COPD-CB group during stable disease (Supplementary Figure S5).

Finally, we assessed the relationship between the serum concentrations of CTRC, OSM, and MMP-10 and the clinical characteristics listed in Table 1 during stable disease. Notably, we found that the concentrations of OSM displayed a positive correlation with those of blood neutrophils and monocytes (Figure 7A and B), while the concentrations of MMP-10 displayed a positive correlation with smoking history (Figure 7C). Nonetheless, we did not detect any statistically significant correlation between the serum concentrations of CTRC, OSM, or MMP-10 and key lung function outcomes (FEV1, FEV1/FVC, and DLCO), or the blood concentrations of CRP, lymphocytes, or basophils (not shown).

Among all 177 serum proteins targeted in the current study, only three were substantially altered during exacerbations in the COPD-CB group: CTRC, OSM, and MMP-10. Notably, these three proteins have previously been linked to CVD [20-25], though they have never been investigated in the context of COPD exacerbations. Moreover, we found that the serum concentrations of CTRC and OSM during exacerbations correlated with two aspects of blood oxygen saturation, CRP levels, and blood leukocytes, all of which are used in the clinic to assess the severity of an exacerbation. Additionally, the serum concentrations of CTRC and OSM during exacerbations correlated with multiple other cardiovascular biomarkers. In contrast, we found that the serum concentrations of MMP-10 during stable disease correlated with the total number of exacerbations in the COPD-CB group. Finally, we also found that the serum concentrations of OSM and MMP-10 during stable disease correlated with blood leukocytes and smoking history, while high MMP-10 concentrations were associated with high blood eosinophil counts, thus linking these serum proteins to known risk factors in COPD. Nevertheless, the serum concentrations of CTRC, OSM, and MMP-10 during stable disease in the COPD-CB group were similar to those of the control groups. Of note, it is important to emphasize that all exacerbation samples were taken upon arrival at the clinic and prior to emergency treatment. Therefore, it is possible that samples collected after treatment or at later time points might have yielded different results.

Chymotrypsin C (CTRC), also known as caldecrin or elastase 4, is a protease produced by pancreatic acinar cells that can degrade trypsinogen and trypsin [26] and has potent calcium-decreasing (hypocalcemic) properties in blood [27]. Importantly, reduced production of CTRC is associated with a number of cardiovascular, respiratory, and endocrine diseases. For instance, extracellular vesicles isolated from blood of patients who experienced myocardial infarction contain less CTRC than those isolated from blood of healthy controls [20]. In addition, loss-of-function mutations in the CTRC gene result in chronic pancreatitis due to unrestrained trypsin activity [28]. Importantly, studies on animal models indicate that unrestrained trypsin activity can also lead to emphysema [29], and thus a decrease in CTRC expression might be relevant for the development of emphysema in patients with COPD. In fact, a previous study demonstrated that expression of the CTRC gene is downregulated in patients with severe or very severe COPD (GOLD stages 3–4) when compared to those with milder disease (GOLD stage 2) [30]. In the present study, we show that the serum concentration of CTRC is decreased during exacerbations of COPD, and that this decrease is associated with a sharper decline in blood oxygen levels and a marked increase in the blood levels of CRP, monocytes, and neutrophil elastase—a key enzyme involved in emphysema development. Moreover, we show that a decrease in the serum concentration of CTRC during exacerbations of COPD is associated with a similar decrease in the cardiovascular biomarker BOC, a decrease that has previously been linked to hypoglycemia—a known risk factor for CVD [31]. Given the previously published evidence linking myocardial infarction with reduced CTRC content in blood extracellular vesicles, our current findings on reduced systemic levels of CTRC during exacerbations of COPD are fully compatible with an increased cardiorespiratory risk in this clinical setting.

Oncostatin M (OSM) is a multifunctional cytokine that originally derived its name from its ability to inhibit the proliferation of numerous cancer cell lines [32]. Nevertheless, OSM can be released by leukocytes under different circumstances [33], and an increase in OSM has been associated with cardiorespiratory diseases and severe infections. First, the blood concentration of OSM is increased in patients with heart failure [21]. In this context, OSM can have multifaceted effects: while OSM has a protective effect on the heart immediately after myocardial infarction [22,23], continuous stimulation with OSM can in turn lead to heart failure [23,24]. Second, the concentration of OSM is increased in the airways of patients with stable COPD [34] or severe asthma [35], as well as in the blood of patients with bacterial septicemia [36]. Moreover, a recent study implicates these increased levels of OSM in respiratory exacerbations, given that deletion of the OSM gene or treatment with an anti-OSM blocking antibody prevented bacterial-driven exacerbations in a mouse model of allergic asthma [37]. Notably, we now show that the blood concentration of OSM increases during exacerbations in patients with COPD and that this increase is associated with lower blood oxygen levels and higher blood levels of CRP, neutrophils, and myeloperoxidase—an enzyme implicated in both COPD progression and CVD risk [38,39]. In addition, we now show that an increase in the blood concentration of OSM during exacerbations of COPD is linked to an increase in the blood concentration of six other cardiovascular biomarkers: IL-1ra, LOX-1, IL-6, PTX3, CXCL1, and CEACAM8, all of which have been linked to increased risk for CVD [40-45]. Thus, our results further strengthen the association between increased OSM levels and a higher cardiorespiratory risk.

Matrix metalloproteinase 10 (MMP-10), also known as stromelysin-2 or transin-2, is released by macrophages and, to a lesser extent, epithelial cells in most tissues upon injury or infection. Importantly, a study on mice suggests that production of MMP-10 increases in the heart following myocardial infarction [25]. In addition, several lines of evidence implicate MMP-10 in the development of the pulmonary comorbidity emphysema. First, smokers and patients with emphysema have increased concentrations of MMP-10 in blood and bronchoalveolar lavage samples [46,47]. Second, mice lacking the MMP10 gene fail to develop emphysema even after 6 months of chronic exposure to cigarette smoke [48]. Third, MMP-10 has been identified as a putative driver of alveolar wall remodeling in a large gene-wide association study in patients with COPD [48]. Furthermore, we now complement these findings by showing that the serum concentration of MMP-10 is increased in exacerbations of COPD when compared to stable disease in paired study samples. In contrast, a previous study failed to detect a statistically significant difference in the plasma concentration of MMP-10 during exacerbations of COPD and stable disease in unpaired study samples [49]. However, this discrepancy might be easily explained by a key factor: the patient material used in that study was very small (n = 6 patients per group). In addition, it is unclear whether that study considered having allergy as a reason for exclusion, and all their patients with COPD exacerbations had community-acquired pneumonia, two factors that may have confounded their results. Finally, we also show that the serum concentration of MMP-10 during stable disease correlates with a higher number of exacerbations and pack-years, as well as blood eosinophil counts above the established threshold of 0·3 × 10⁹ /L, all of which are risk factors for disease progression [1]. We therefore think that our results are compatible with a positive association between a systemic increase in MMP-10 during exacerbations of COPD and an elevated cardiorespiratory risk.

In conclusion, while previous studies have implicated CTRC, OSM, and MMP-10 in CVD, our study forwards original evidence that the serum concentrations of these proteins bear potential as cardiorespiratory biomarkers in patients with COPD and CB. Collectively, these biomarkers display substantial alterations during exacerbations and associated with the number and severity of exacerbations plus additional risk factors such as elevated blood eosinophils. This novel evidence provides a rationale for future assessment of CTRC, OSM, and MMP-10 in larger prospective studies on cardiorespiratory risk in patients with COPD and CB that include patients with varying exacerbation severity.

Data are available upon reasonable request to corresponding author. European Data Regulations preclude open deposition of sensitive personal data (including proteomics) into public repositories.

This study was co-sponsored by AstraZeneca and one of the co-authors (Georgios Stratelis) was partially employed by AstraZeneca at the time of the investigation, as stated in the Funding section, and authors’ affiliations, respectively. None of the study sponsors exerted any specific influence on the design of the study or on the interpretation of the results. All decisions regarding these matters were made on a scientific basis by the team of authors. All authors confirm that they have no further potential conflicts of interest to disclose.

This work was supported by project grants from the Swedish Heart-Lung Foundation (20210286), Swedish Research Council (2021-01527), and an unrestricted research grant from AstraZeneca.

A.L. conceived, provided funding for, and supervised the research project, after initial discussions with E.I.C., G.S., and T.J. After this, E.I.C., K.A., N.P. and A.L. outlined the research protocols and coordinated the investigations. E.I.C., K.A., and N.P. performed the research analyses. K.A., A.E.J., and A.L. collected human samples. E.I.C. and A.L. reviewed the data. N.O. supervised the statistical analysis. T.J. and G.S. supervised the data interpretation regarding cardiovascular disease. E.I.C. and A.L. wrote the manuscript. E.I.C., G.S., N.P., A.E.J. and A.L. critically reviewed the contents of the manuscript. All authors read and approved the final version of the manuscript.

All participants provided written informed consent before inclusion, and all procedures and handling of patient information were conducted in accordance with the ethical permit approved by the Swedish Ethical Review Authority in Gothenburg (Diary No. S 233 – 03, T 286 – 04, and T 521 – 06).