Introduction

Circulating microparticles (MPs) are a heterogeneous population of small cellular vesicles ranging in size from 0.1 µm to 1 µm.¹ Recent studies have shown that MPs can transport cellular proteins, microribonucleic acids, messenger ribonucleic acids, DNA, and membrane receptors from one cell to another, contributing to intercellular communication.^1-3 Under physiological conditions, MPs are present in small quantities, but their levels tend to increase in various diseases, suggesting a role in the pathogenesis of inflammation and procoagulant activity, which can lead to vascular disorders and other complications.^2,4

Their prothrombotic effects stem from the presence of surface phosphatidylserine and tissue factor, enhancing thrombin and fibrin generation.^5,6 Elevated levels of MPs are associated with thrombosis and disseminated intravascular coagulation, but they also support vascular homeostasis—some MPs stimulate angiogenesis, endothelial cell proliferation, and the reconstruction of the vascular barrier.^5,6

MP levels reflect a balance between their production (eg, during apoptosis or cell activation) and clearance (via enzymatic degradation, phagocytosis, or hepatic or pulmonary uptake).¹ Their short half‑life (approximately 5.8 h), as compared with markers, such as C‑reactive protein (CRP; approximately 19 h), suggests a potential for early detection of disease activity.⁷

Although the role of MPs in various diseases has been extensively studied,^6,8 data on their involvement in inflammatory bowel disease (IBD) remain limited.⁷ Few studies have examined the correlations between circulating MPs and clinical disease activity scores, such as the Crohn Disease Activity Index (CDAI).^9,10 A deeper understanding of MP subpopulations and their relationship with disease activity may offer insights into their potential as biomarkers and therapeutic targets in CD. While circulating MPs have been previously studied in the context of inflammatory and thrombotic diseases, there is still a significant gap in our understanding of their specific role in the pathophysiology of CD. Understanding the entire pathomechanism of exacerbation development is also important due to the observed increasing number of patients with CD in different regions of the world.^11,12 Current research is limited, and most of it does not closely analyze the subpopulations of MPs or their relationship with the clinical activity of the disease as assessed by the CDAI. Furthermore, there is a lack of data linking the presence of specific subpopulations of MPs with markers of platelet (PLT) activation and endothelial damage, as well as with PLT‑derived growth factor (PDGF) levels—a factor involved in both angiogenesis and fibrosis. This study aimed to fill this gap by providing a comprehensive quantitative and qualitative analysis of MPs in the context of clinical activity of CD, considering their potential role as biomarkers and mediators of vascular dysfunction.

MPs mainly originate from platelets, endothelium, and leukocytes, retaining markers that allow for identification of their origin and function using flow cytometry.⁵ In this study, we quantified circulating MPs and PDGF in CD patients, examined their cellular origin, and explored associations with the disease activity (as measured by the CDAI).

Patients and methods

The inclusion criteria for the study were: confirmed diagnosis of CD based on current clinical, endoscopic, histopathologic, or imaging criteria, and age of at least 18 years. The control group consisted of healthy adults without chronic illnesses and no symptoms related to the digestive system, not on immunosuppressive, anti‑inflammatory, or antibiotic therapy. Exclusion criteria for CD and control groups comprised the presence of ulcerative colitis, autoimmune diseases, cancer, pregnancy, or lack of consent. All participants were recruited between 2013 and 2015 at the Clinic of Gastroenterology and Hepatology of the University Hospital in Kraków, Poland. The study was approved by the Bioethics Committee of the Jagiellonian University (KBET/135/B/09). Venous blood samples were collected from all participants in a fasting state into tubes containing sodium citrate.

Basic laboratory tests were performed, including complete blood count (white blood cell [WBC] and PLT count), coagulation parameters (prothrombin time [PT], activated partial thromboplastin time [APTT], and international normalized ratio [INR]), fibrinogen concentration (measured using the Clauss method), and CRP concentration.

MPs were isolated from PLT‑poor plasma using centrifugation. The MP analysis was conducted with flow cytometry. The following monoclonal antibodies (BioLegend, San Diego, California, United States) targeting specific MP subpopulations were used for identification: CD41 (a PLT marker), CD62P (a PLT activation marker), and CD31 (an endothelial cell marker). The following reagents were used for MP determination: (all from BioLegend) Annexin V–FITC (fluorescein isothiocyanate); mouse (immunoglobulin G1 [IgG]) IgG1 antibody against human integrin α chain 2b (CD41) labelled with allophycocyanin (APC); mouse IgG1 antibody against human P‑selectin (CD62P) labelled with phycoerythrin (PE); and recombinant monoclonal and polyclonal antibodies developed in the mouse, rabbit, rat, goat, and Armenian hamster against platelet endothelial cell adhesion molecule (PECAM‑1, also known as CD31), conjugated with Pacific Blue (PB). Appropriate isotype controls were also employed: mouse IgG1‑PE, mouse IgG1‑APC, and mouse IgM‑Pacific Blue. Additionally, Annexin V Binding Buffer (10 × concentrate; BD Biosciences, San Jose, California, United States) and PBS without ions (1 × concentrate; Corning Inc., Corning, New York, United States) were used. Both the total number of MPs and the following selected subpopulations were analyzed: CD41+, CD41+CD62P+, CD41+CD31+, and CD41–CD31+.

Plasma PDGF concentration was measured using the enzyme‑linked immunosorbent assay, following the manufacturer’s instructions (Sunred Biological Technology Co., Shanghai, China). The results are expressed in pg/ml.

Results

The study included 78 participants divided into 3 groups: 36 patients with active CD, 20 patients in remission, and 22 healthy controls. Mean (SD) age was 39.3 (6.3) years in the active CD group, 35.6 (13.3) years in the remission group, and 34.9 (10.3) years in the control group. Men predominated in both study CD groups (active CD group, 29 men and 7 women; remission group, 14 men and 6 women). Median (IQR) disease duration was 4.5 (2–8) years in the active CD group and 4 (1–8) years in the remission group, with 2.5% of the patients being newly diagnosed (up to 6 months prior). Treatment regimens included immunosuppressive therapy (19 patients in the active CD group and 12 in the remission group), 5‑ASA (34 individuals in the active CD group and 19 in the remission group), steroids (16 participants in the active CD group and 5 in the remission group), antibiotics (11 patients in the active CD group and 2 in the remission group), and biologics (3 individuals in the active CD group and none in the remission group).

Table 1 outlines the assessed parameters. The CDAI score was higher in the active CD group than the remission group (P <⁠0.001). The patients with active CD also showed elevated CRP levels and PLT count, as compared with the controls (P <⁠0.05). Coagulation parameters (INR, PT, and APTT) and WBC count did not differ between the groups. The number of MPs, including PLT‑derived MPs (CD41+), was higher in the active CD group than in the control and remission groups (P <⁠0.001), as were the activated MPs (CD41+CD62P+ and CD41+CD31+). PDGF concentration was elevated in the patients with active CD in comparison with the controls (P = 0.001), indicating increased PLT activation and inflammation.

Multiple linear regression analyses were performed on log‑transformed nonparametric data to identify factors associated with MPs, CD41+ MPs, CD31+ MPs, CD41–31+ MPs, CD41+CD62P+ MPs, CD41+CD31+ MPs, CD41+CD62P+CD31+ MPs, and PDGF. All 78 participants were included in these analyses. The model fit was reported with R², adjusted R², and bias‑corrected R². The associations of independent variables were reported using β and P values. The PDGF model showed the strongest fit (R² = 0.93; adjusted R² = 0.73; P = 0.05), with positive effects of WBC (β = 4.32; P = 0.007), APTT (β = 2.31; P = 0.009), fibrinogen (β = 1.61; P = 0.02), 5‑ASA (β = 1.54; P = 0.02), and antibiotics (β = 3.77; P = 0.004); and negative effects of PLT (β = –1.06; P = 0.02) and CRP (β = –3.11; P = 0.01). The MP model was significant (R² = 0.74; adjusted R² = 0.49; P = 0.03), positively associated with CDAI (β = 0.69; P = 0.01) and negatively with PLT (β = –0.4; P = 0.04), and immunosuppression (β = –0.53; P = 0.01). The CD41–31+ MPs model was also significant (R² = 0.88; adjusted R² = 0.65; P = 0.04), positively associated with CDAI (β = 0.9; P = 0.04) and negatively with immunosuppression (β = –0.55; P = 0.01). Other models were insignificant. These findings indicate that disease activity and treatment, particularly immunosuppression, are important determinants of circulating MP and PDGF levels.

Discussion

The MPs analyzed in this study were small membrane vesicles secreted by various types of cells. Previous studies suggest that circulating MPs may play a role in the pathomechanism of CD.^8-10 Our findings show that the concentration of both MPs and their selected subpopulations (Table 1) were considerably elevated in the patients with active disease, as compared with those in remission and the healthy controls. This suggests that MPs may be involved in the inflammatory process characteristic of CD exacerbations, which has also been highlighted in other studies.^9,10 The most significant differences were observed in the total number of MPs and the following subpopulations: CD41+ (PLT‑derived), CD41+CD31+ (with PECAM‑1 expression), CD41–CD31+ (endothelial, lacking features of PLT origin), and activated MPs, specifically CD41+CD62P+. This suggests that MPs are involved not only in systemic inflammation and coagulation but also in endothelial injury and vascular remodeling—key components of CD exacerbations.

What is more, the CD41–CD31+ subpopulation showed a marked and independent association with clinical disease activity (CDAI, β = 0.9; P = 0.04), and negatively correlated with immunosuppressive treatment (β = –0.55; P = 0.01). These findings suggest that endothelial MPs may reflect ongoing vascular injury, and that immunosuppressive therapy could help preserve endothelial integrity. Supporting this concept, Leonetti et al¹⁰ observed in the mouse model that MPs derived from patients with CD, both in the active phase of the disease and in remission, impaired endothelial‑dependent vasodilation. More surprisingly, MPs from the patients in remission caused stronger effects, indicating their involvement in endothelial and vascular dysfunction in CD. Gaetani et al¹³ also demonstrated higher concentrations of CD41+ MPs and angiogenic mediators, such as vascular endothelial growth factor, angiopoietin‑1, fibroblast growth factor, and PDGFα in patients with active CD, which indicates the promotion of angiogenesis in vitro. Our data confirm these findings and suggest that MPs may act as active participants in pathological vessel remodeling and endothelial disruption in vivo. However, no changes in the levels of MPs were observed in the remission group and the control group, which does not confirm the impact of the endothelial subpopulation on the endothelial function of blood vessels. Moreover, Cromer et al¹⁴ described the key role of endothelial dysfunction in the pathogenesis of IBD. This reinforces the hypothesis that MPs—especially those of endothelial and PLT origin—could serve as molecular links between immune activation and vascular injury in CD. In CD, MPs from the endothelial cells and PLTs can combine immune activation with endothelial damage, exacerbating vascular dysfunction.¹⁴

The increase in the concentration of CD41+CD31+ and CD41+CD62P+ MPs during the active phase of CD (Table 1) may indicate their involvement in PLT activation and endothelial damage through the inflammatory process. However, although CD41+CD62P+ MPs were significantly elevated in the patients with active disease, they did not show an independent predictive value in regression analyses, which may limit their specificity as biomarkers. This subpopulation appears to reflect generalized PLT activation rather than specific inflammatory signaling. In contrast, CD41–CD31+ MPs exhibited a stronger and independent correlation with disease activity, making them more specific biomarkers for endothelial damage. Additionally, CD41+ MPs were markedly higher in the active phase of the disease than in remission, supporting the hypothesis that PLT activation plays a key role in the pathophysiology of CD.

Similar results were obtained by Agnholt et al,¹⁵ who demonstrated an increased number of PLT‑derived and CD62P+ MPs in patients with active CD, as compared with those in remission, and observed that the count of those MPs correlated with clinical disease activity. The researcher suggested that they may be indicators of an inflammatory process, but did not explain the mechanism involved. However, Gaetani et al¹³ linked high levels of CD41+ MPs in patients with active CD to pathological vessel growth and subsequent endothelial damage. In our findings, the total number of MPs and CD41+ MPs was considerably higher in the patients with active CD than the individuals in remission. This suggests that the increased number of MPs may be a consequence of both cell activation and damage.

Our study also assessed the concentration of PDGF, which was higher in the active phase of the disease. In the multivariate linear regression, PDGF concentrations were independently predicted by CRP (β = –3.11; P = 0.01), WBC (β = 4.32; P = 0.007), PLT (β = –1.06; P = 0.02), APTT (β = –2.31; P = 0.009), fibrinogen (β = 1.61; P = 0.02), and treatment, demonstrating complex interactions between inflammation, leukocyte activity, coagulation, type of drugs used, and vascular remodeling. These results are confirmed by Kumagai et al,¹⁶ who demonstrated that PDGF and its receptors are present in the areas of inflammation and fibrosis in IBD, and their expression by leukocytes suggests that PDG may act as a chemoattractant. In fibrotic areas, activated fibroblasts and myofibroblasts expressed PDGF, indicating its involvement in fibrogenesis via stimulation of fibroblast proliferation and extracellular matrix production.¹⁶

It is also worth emphasizing that no notable differences were found between the patients in remission and healthy individuals in most parameters, suggesting normalization of the MP profile during inflammation absence. This finding supports the idea that MPs are not merely chronic background markers, but dynamically reflect the disease activity and potentially therapeutic response.

To sum up, substantial elevations in specific MPs and PDGF in the active phase of CD suggest their potential clinical application as noninvasive biomarkers for monitoring disease activity and vascular complications. Although the total number of MPs and the CD41+ subpopulation show some associations with the disease activity, it is the CD41–CD31+ subpopulation that seems to best reflect advanced inflammatory changes and endothelial damage. The increased concentration of PDGF in the active disease phase is also worth noting—its relationship with both angiogenesis and fibrotic processes suggests involvement in vascular remodeling in a chronic inflammatory state. In the future, MPs and PDGF may also represent therapeutic targets aimed at reducing vascular injury and tissue remodeling.

Nevertheless, certain limitations must be acknowledged. The relatively small control group and the cross‑sectional nature of the study (lack of assessment of MP levels over time) are notable. Another limitation of our study is that although an a priori power calculation was performed, the remission group did not reach the minimum estimated sample size, which may have affected the statistical power of some comparisons. Still, the results provide valuable preliminary insights that should be confirmed in larger, adequately powered studies. The conclusions appear interesting, and future studies, in addition to increasing the number of patients, should take into account other coagulation markers (eg, D‑dimer) and endothelial activation indicators (eg, intercellular adhesion molecule or vascular cell adhesion molecule‑1) in order to further analyze the role of MPs in vascular pathology in CD.