Elevated levels of advanced glycation end-products AGE10 in patients with reactive arthritis caused by Chlamydia trachomatis and Epstein-Barr virus infection

Abstract

Advanced glycation end products (AGEs), and particularly the unique AGE10 epitope, may be a potential biomarker of immunopathology in rheumatic diseases. They may be associated with inflammation, joint damage and ossification processes. AGE10 present in human and animal tissues could be detected with monoclonal antibody against melibiose-derived glycation product MAGE synthesized in anhydrous conditions. This MAGE product was different from the classic synthesis in water solution. The epitope was determined in serum with ELISA using these anti-MAGE monoclonal antibodies. This work aims to determine serum AGE10 levels in patients with reactive arthritis (ReA)-caused with Chlamydia trachomatis (group 2) and ReA with C. trachomatis during the reactivation of EBV infection (group 3). Additionally, ankylosing spondylitis (AS) patients (group 4) were involved in the study, due to the potential evolution of ReA toward AS. The control group maintained physiological AGE10 levels (316 µg/ml), while the combined infection group showed elevated AGE10 (850 µg/ml) compared to the chlamydial-only group (17 µg/ml). Fluorescent fAGE were at the highest level in AS patients. A striking finding was the complete absence of detectable AGE10 antigen in the AS group, coinciding with notably elevated immune complex AGE10–anti-AGE10 levels. A similar pattern was observed in patients with ReA caused by C. trachomatis alone (Group 2), albeit to a lesser extent. In contrast, both the control group and patients with ReA associated with EBV coinfection (group 3) displayed an inverse relationship, characterized by higher antigen levels and lower immune complex concentrations. Thus, diminished level of AGE10 could be caused, besides local accumulation, also by immune complexes formation, a pathogenic factor. Therefore, evaluating disease activity in ReA and AS is crucial to further our understanding of the pathophysiology of AGEs formation and predicting prognosis.

Introduction

Advanced glycation end-products (AGEs) – are a wide-ranging group of molecules formed as a result of a non-enzymatic reaction of reducing sugars with an amino group of proteins, lipids, and nucleic acids and contribute to the age-associated increase in inflammation or “inflammaging”, and are formed exogenously via the Maillard reaction and endogenously via other mechanisms [1]. In a homeostatic situation, the endogenous development of AGEs is prolonged, and the clearance rate is adequate to prevent their accumulation. Endogenous sources comprise glycation, polyol pathway and glyoxidation [2]. AGEs stimulate oxidative stress and inflammation by binding with cell surface receptors or crosslinking with body proteins, altering their structure and function, and binding to the type I transmembrane receptor of the immunoglobulin superfamily. The binding of circulating AGEs to full-length, membrane-bound receptors to AGEs (RAGE) initiates a signaling cascade: (1) leads to nuclear factor kappa-B (NF-κB) activation, promotes of the transcription of several pro-inflammatory cytokines and acute-phase proteins [3, 4]; (2) facilitates oxidative stress via the initiation of NADPH oxidase, that helps to the de novo production of AGEs, which can act as RAGE ligands [5]. These changes participate in multiple anomalies like the accumulation of AGEs in the development of type 1 and type 2 diabetes mellitus (DM), cardiovascular disease, rheumatoid arthritis (RA), nephropathy, and dysfunctions in bone refurbishing and neurological diseases, allergological diseases, cancer, metastasis, and other degenerative diseases [4–7].

Oxidative stress is produced by elevated reactive oxygen species (ROS) such as hydrogen peroxide, superoxide anions, hydroxyl radicals, nitric oxide, also NADPH oxidases (Nox) and nitric oxide synthases (NOS). RAGE activation increases ROS production by stimulating specific signaling cascades such as TGF-𝛽, NF-𝜅B, and Nox-1. The members of the AGE/RAGE signaling cascade (i.e. p38 MAPK and ERK1/2) were found to be phosphorylated upon RAGE activation [8]. Ligation of AGE to RAGE results in NF-κB migration to the nucleus and stimulates transcription of pro-inflammatory genes. It leads to up-regulation of endothelial adhesion molecules, such as secretory vascular cell adhesion molecule-1 (VCAM-1) [9]. Complex AGE/RAGE can influence both cellular and systemic immune responses. AGEs are produced and can accumulate during chronic inflammation and lead to dysregulation of cellular physiology and cellular signaling pathways [10]. Although RAGE is the primary receptor mediating the pro-inflammatory and pro-oxidative signaling of AGE, several other receptors are involved in AGE clearance, metabolism, and the induction of cellular responses. Particularly important are scavenger receptors, CD36, LOX-1, and galectin-3, which contribute to AGE removal and the regulation of immune responses [3].

Spondyloarthritis (SpA, also called spondyloarthropathy) is a group of joint and spine inflammatory diseases with various clinical manifestations. Five major subtypes of SpA are recognized based on the classification criteria proposed by the European Spondyloarthropathy Study Group (ESSG) [11]. These subtypes include reactive arthritis (ReA), ankylosing spondylitis (AS), psoriatic arthritis (PsA), inflammatory bowel disease-associated spondyloarthritis (IBD-SpA), and undifferentiated spondyloarthritis (uSpA).

Reactive arthritis (ReA) is usually triggered by a genitourinary or gastrointestinal infection, which contributes to the activation of T-lymphocytes. These immune cells release inflammatory mediators, leading to aseptic synovitis and an aberrant immune response in genetically predisposed individuals, particularly those carrying the HLA-B27 allele. The resulting sterile joint inflammation is typically characterized by asymmetric oligoarthritis, enthesitis, and may be accompanied by extra-articular manifestations such as conjunctivitis or uveitis.

Chlamydia trachomatis (C. trachomatis) is a Gram-negative obligate intracellular bacterial pathogen that causes urogenital and eye infection, which is the main reason for ReA, and exists in the human body and activates typical ReA [12]. Chlamydia is a unique parasite distinguished by its biphasic developmental cycle, alternating between the infectious Elementary Body (EB) and the metabolically active, replicative Reticulate Body (RB). Like a virus, Chlamydia penetrates the cell. And then, possessing its DNA like a bacterium, it suppresses the cell, robs it of energy, and forces it to produce its kind. When the cell dies, the cell membrane ruptures and new Chlamydia attack new cells [13]. Human leukocyte antigen HLA-B27 is detected in 60–80% of all patients with ReA [14, 15]. The hypothesis of molecular mimicry suggests that a cross-reactive peptide derived from an infecting bacterial pathogen stimulates T cells, which subsequently respond to an HLA-B27 associated “self-peptide” or to peptides derived from HLA-B27 directly [16].

Ankylosing spondylitis (AS), or spondyloarthritis (SpA), is an unknown etiology inflammatory disease characterized by chronic axial joint inflammation, mainly involving the sacroiliac joints, spinal processes, and paraspinal soft tissues [17]. An essential role in the pathogenesis of SpA is like in ReA, in their association with HLA-B27, which is the genetic feature of SpA, the phenotype found in about 90% of the cases with AS [12, 14, 18]. HLA-B27 may suggest a possible evolution of ReA toward AS [19].

Viral infections are a well-recognized cause of acute arthralgia and arthritis. Arthralgias are the most common joint manifestation of Epstein-Barr viral infection, with the rare case of arthritis causing occasional significant joint swelling. Arthritis symptoms are normally not long but can exist in joints for several months [20]. Epstein-Barr Virus (EBV) is involved in many immune evasion mechanisms, including penetration of host cell membranes. The life cycle of EBV, a large enveloped DNA virus, consists of three main phases: primary infection, latency, and lytic reactivation [21, 22].

Recently we have reported on the synthetic melibiose-derived glycation product MAGE, which mimics a unique epitope present in human and animal tissues [23]. This MAGE product was synthesized in anhydrous conditions, different from the classic synthesis in water solution. The physiological serum epitope called AGE10, was determined with ELISA using anti-MAGE monoclonal antibody [23, 24].

From the group of AGEs, it is also worth distinguishing pentosidine, a short one is one of the best characterized cross-linked glycation products [3]. Due to fluorescent properties (exc. ​​335 nm, em. 385 nm), it is possible to easily analyze this analyte in biological material, without separation from the sample and using highly specialized equipment such as mass spectrometry or high-performance liquid chromatography. Fluorescence measurement at wavelengths exc. 370 and em. 440 is also used to estimate the content of total fluorescent AGEs [6, 7]. Pentosidine is frequently measured in clinical and experimental studies as a marker of AGEs accumulation. Pentosidine levels reflect the extent of glycative stress in the body and are often used to assess the progression of chronic diseases associated with AGE-related tissue damage, such as diabetes, cardiovascular diseases, and inflammatory joint disorders [25]. Its stability and detectability make it a valuable biomarker for evaluating the impact of AGEs on disease pathogenesis and progression.

The current understanding suggests that both persistent C. trachomatis infection and EBV reactivation contribute to oxidative stress and immune dysregulation, which are key drivers of AGE formation [4, 10, 25]. However, the combined impact of these two distinct intracellular pathogens—a chronic bacterial stressor and an intermittently reactivating viral pathogen—on the systemic glycative burden (AGEs) in ReA patients has not been previously quantified. Given the known mechanism of AGE/RAGE signaling in amplifying inflammation and the strong association of both C. trachomatis and EBV with sustained immune activation, we hypothesize that the co-existence of C. trachomatis with EBV reactivation constitutes a synergistic pro-inflammatory and pro-oxidative “double hit”. It is therefore essential to differentiate these patient groups to determine if EBV reactivation acts as a critical amplifying factor, leading to significantly higher levels of AGEs, which may serve as biomarkers for a more severe or complex disease course.

Materials and methods

The study was conducted at the Lviv Regional Clinical Diagnostic Center and the Department of Clinical Immunology and Allergology of the Danylo Halytsky Lviv National Medical University in the period 2017–2019, following the principles of the seventh revision of the Helsinki Declaration of Human Rights (2013), the Council of European Convention on Human Rights and Biomedicine and the appropriate Laws of Ukraine. The research was approved by the Ethics Committee at the Danylo Halytsky Lviv National Medical University (protocol #6, 2017/03/08). All the participants signed informed consent. ReA was diagnosed with the 2015 European guideline on managing C. trachomatis [26]. The patients with AS were diagnosed by a combination of clinical criteria of inflammatory back pain and enthesitis – or arthritis with radiologic findings. The modified New York criteria and the Assessment of SpondyloArthritis International Society (ASAS) classification criteria were used to diagnose AS [27]. The ASAS classification criteria offer a way to classify patients with axial (ASAS axSpA) and peripheral (ASAS pSpA) spondyloarthritis. The axial criteria are met if a patient presents with sacroiliitis on imaging (X-ray or MRI) plus at least one SpA feature, or if they have the HLA-B27 allele plus at least two other SpA features, provided they have chronic back pain [27].

Inclusion criteria of the patients:

Both sexes adult patients were between 18 and 65 years old, who were first diagnosed with ReA due to Chlamydia infection, and AS, who had not yet received basic immunosuppressive therapy. Nine study group patients displayed chronic comorbid conditions: four with hypertension, four with hypothyroidism, and atopic dermatitis – in 1 patient, which could not affect the study results.

Exclusion criteria of the patients

were pregnancy, EBV, C. trachomatis as a manifestation of opportunistic infection in HIV infection, psychological diseases, and children. In addition, patients with diseases that could affect inflammatory markers, such as acute infections and hematological disorders, were excluded from this analysis.

Characteristics of the patient groups

Patients

One hundred twenty-three patients applied for visits with arthritis problems: in general, 88 patients with ReA, including 45 patients with ReA-caused C. trachomatis and 43 with ReA-combined C. trachomatis with EBV. For comparison purposes, two groups were taken: the first – 25 healthy control participants of appropriate age and sex and the second group, 35 patients with AS, detected for the first time. In addition, general laboratory, biochemical, and instrumental studies were performed on all patients. Table 1 shows detailed characteristics of the subjects of participants.

Table 1.

Number of participants included in the study

	Group 1: Control	Group 2: Reactive arthritis (ReA) C. trachomatis	Group 3: Reactive arthritis (ReA) C. trachomatis with EBV	Group 4: Ankylosing spondylitis (AS)
Number of participants	25	45	43	35
Men/women	15/10	37/8	24/19	30/15
Median age	25	30	27	26
Min/max of ages	19–35	18–45	18–38	24–47

We used the Development of a Disease Activity Index to assess reactive arthritis (DAREA) to evaluate the clinical manifestation of ReA [28], shown in Table 2.

Table 2.

Clinical and general laboratory characteristics of ReA patients

		Control	Patients ReA with C. trachomatis	Patients ReA with C. trachomatis + EBV	P value
		N = 25	N = 45	N = 43
		1	2	3
DAREA index		-	9.71 ± 5.23	13.6 ± 9.45	P = 0.0184
Number of swollen joints		-	2.11 ± 1.71	3.15 ± 2.65	P = 0.0307
Number of painful joints		-	2.37 ± 2.43	3.74 ± 2.61	P = 0.0126
Assessment of pain by the patient		-	2.14 ± 1.35	2.47 ± 1.43	P = 0.2686
The patient’s general condition assessment		-	1.15 ± 1.32	1.65 ± 1.28	P = 0.0749
C-reactive protein (mg/dL)		0.2 ± 0.15	1.94 ± 1.32	2.59 ± 1.78	P1–2 < 0.0001 P1–3 < 0.0001 P2–3 = 0.0542
Chronic low-grade fever			45	43
Heel pain and difficulty walking			9	11
Enthesitis and dactylitis			36	34
Conjunctivitis			35	37
Uveitis			5	7
Urethritis or cystitis			45	43
Chronic fatigue syndrome			22	22
Differential Count
Leukocytes (109/L)		6.62 ± 4.27	9.45 ± 5.22	7.87 ± 4.43	P1–2 = 0.0238 P1–3 = 0.2598 P2–3 = 0.1303
Lymphocytes	%	29 ± 6.41	44.0 ± 8.42	48.0 ± 10.16	P1–2 < 0.0001 P1–3 < 0.0001 P2–3 = 0.0469
	109/L	1.92 ± 1.31	4.16 ± 2.26	3.78 ± 2.84	P1–2 < 0.0001 P1–3 = 0.0030 P2–3 = 0.4882
Neutrophils (%)	%	58.0 ± 28.45	37.0 ± 16.84	31 ± 14.03	P1–2 = 0.0002 P1–3 < 0.0001 P2–3 = 0.0736
	109/L	3.84 ± 3.14	3.50 ± 2.23	2.44 ± 1.73	P1–2 = 0.6001 P1–3 = 0.0204 P2–3 = 0.0150
Monocytes (%)	%	8.0 ± 2.48	14.0 ± 3.12	16.0 ± 4.87	P1–2 < 0.0001 P1–3 < 0.0001 P2–3 = 0.0402
	109/L	0.53 ± 0.22	1.32 ± 0.58	1.26 ± 0.55	P1–2 < 0.0001 P1–3 < 0.0001 P2–3 = 0.6201

Values in bold indicate significant differences between investigated groups.

Laboratory investigations

The levels of AGE10 were determined with ELISA according to the published procedure [7, 23, 24, 29–32], using mouse anti-MAGE monoclonal antibody obtained in the Hirszfeld Institute of Immunology and Experimental Therapy Polish Academy of Sciences, Wroclaw. The test sera were diluted 6 times, the resulting CV% were 3.3–31.3. AGEs were also measured fluorometrically, using natural fluorescence at 335/385 nm of excitation/emission, specific for pentosidine and at 370/440 nm of excitation/emission, specific for total fluorescent advanced glycation end products (fAGEs), as described in previous work [7, 33, 34]. The method of Turk et al. [35] was applied for recovering immune complexes (ICs) from serum, which were determined as previously. Flow Cytometric Immunophenotypic Test (FCIPT) was used to identify HLA-B27 in whole blood using The BD™ HLA-B27 on flow cytometer the BD FACSCalibur™. Detection of C. trachomatis was performed in peripheral blood, swabs, and scrapings using a GeneProof Chlamydia trachomatis PCR Kit (Triomed, Turkey). EBV in blood, saliva, and the mucosa of the pharynx was performed by PCR using GeneProof Epstein-Barr Virus (EBV) PCR Kit (Triomed, Turkey). These two tests were performed using “Rotor-Gene 6000” (Corbett Research, Australia) equipment. The serum evaluation of Human Anti-Chlamydia trachomatis IgA ELISA Kit (ab108718), IgM ELISA Kit (ab108722) and IgG ELISA Kit (ab108720), and for identification of EBV the human anti-Epstein-Barr virus IgG (ab108730) and IgM (ab108732) was performed using ELISA Kits (Abcam, UK) by the manufacturer’s instructions [7].

Statistics

The obtained data was processed by GraphPad Prism 8.0.1 and Statistica 13.3. The probability assessment was done using the unpaired t-test (t) for a parametric, normally distribute on and the non-parametric Mann-Whitney method in the data not normally distributed. The arithmetic mean value (M) and the standard deviation (SD) (M ± SD), two-tailed P value were calculated Where the data distribution was not normal, differences between groups were assessed using the Kruskal-Wallis test. The difference was considered probable at p < 0.05.

Results

Clinical results

Clinical data detailed analysis showed statistical differences between ReA patients caused C. trachomatis compared to patients with ReA-caused C. trachomatis + EBV (p = 0.0307) in number of swollen joints (2.11 ± 1.71 vs. 3.15 ± 2.65, respectively) and painful joints (p = 0.0126) (such as arthritis) (2.37 ± 2.43 vs. 3.74 ± 2.61, respectively). CRP was expressed in the control group (P < 0.0001) compared to patients with ReA-caused C. trachomatis and ReA-caused C. trachomatis + EBV (0.2 ± 0.15 mg/dl vs. 1.94 ± 1.32 mg/dl vs. 2.59 ± 1.78 mg/dl, respectively). However, between two groups of ReA patients, it was expressed without statistical differences (p = 0.0542). DAREA index (p = 0.0184) was calculated in patients with ReA with C. trachomatis (9.71 ± 5.23), compared to ReA with C. trachomatis + EBV (13.6 ± 9.45).

All patients were determined to be more likely to manifest chronic low-grade fever (37.2–38 °C). Heel pain and difficulty walking related to inflammation were expressed in patients with ReA with C. trachomatis 9 (20%) compared to patients with ReA with C. trachomatis + EBV in 11 (25.6%). The extra-articular manifestations were revealed as enthesitis and dactylitis in 80% of all groups of patients. Conjunctivitis was shown in patients with ReA with C. trachomatis in 35 (77.7%) compared to ReA with C. trachomatis + EBV 37 (86%). Uveitis was registered in 5 (11%) of ReA patients compared to ReA with C. trachomatis + EBV in 7 (16%). Diseases of the genitourinary system (symptoms of urethritis or cystitis) were observed occasionally in all patients with ReA with C. trachomatis and ReA with C. trachomatis + EBV. Chronic fatigue syndrome was elicited in patients with ReA with C. trachomatis in 22 (48.8%) and ReA with C. trachomatis + EBV in 22 (50%).

In the complete blood count (CBC), we observed an increased number of leucocytes in patients with ReA with C. trachomatis (9.45 ± 5.22%, p = 0.0238) and ReA with C. trachomatis + EBV (7.87 ± 4.43% p = 0.2598), compared to the control group (6.62 ± 4.27%). The differences between the two ReA groups were without statistical differences (p = 0.1303). The relative lymphocyte count was significantly higher (p < 0.0001) in patients with ReA infected with C. trachomatis (44 ± 8.42%) and further increased in those co-infected with C. trachomatis and EBV (48 ± 10.16%, p < 0.0001) compared to the control group (29 ± 6.41%). The weak statistical differences between the two ReA groups were observed (p = 0.0469). Absolute lymphocyte amount was more elicited (p < 0.0001) in patients with ReA with C. trachomatis (4.16 ± 2.26 × 10⁹/L) than in ReA patients with C. trachomatis + EBV (3.78 ± 2.84 × 10⁹/L, p = 0.0030) compared to the control group (1.92 ± 1.31 × 109/L). The differences between the two ReA groups were without statistical differences (p = 0.4882). The relative neutrophils amount was less elicited (p = 0.0002) in patients with ReA with C. trachomatis (37 ± 16.84%) compared to ReA patients with C. trachomatis + EBV (31 ± 14.03%, p < 0.0001) and the control group (58 ± 28.45%). The differences between the two ReA groups were without statistical differences (p = 0.0736). Absolute neutrophils amount was less elicited (p = 0. 0.0002) in patients with ReA with C. trachomatis (3.50 ± 2.23 × 10⁹/L) than in ReA patients with C. trachomatis + EBV (2.44 ± 1.73 × 10⁹/L, p = 0.0204) compared to the control group (3.84 ± 3.14 × 10⁹/L). The differences between the two ReA groups were statistically significant (p = 0.0150). The relative monocyte count was significantly higher (p < 0.0001) in patients ReA infected with C. trachomatis (14 ± 3.12%) and even higher in those co-infected with C. trachomatis and EBV (16 ± 4.87%, p < 0.0001) compared to the control group (8 ± 2.48%), with weak statistical differences observed between the two ReA groups (p = 0.0402). Absolute monocyte amount was less elicited (p < 0.0001) in patients with ReA with C. trachomatis (1.32 ± 0.58 × 10⁹/L) than in ReA patients with C. trachomatis + EBV (1.26 ± 0.55 × 10⁹/L, p < 0.0001) compared to the control group (0.53 ± 0.22 × 109/L). The differences between the two ReA groups were not statistical (p = 0.6201).

The next task was to verify chlamydial infection; all patients with ReA were determined to have classes of immunoglobulins (Ig) to C. trachomatis. Among the group of ReA patients with C. trachomatis, elevated IgA to C. trachomatis was found in 38 (84.5%) patients, in patients with combined infection (C. trachomatis + EBV) – in 34 (79.0%) patients. Antibody IgA to Chlamydia trachomatis was increased in both groups of patients: in patients with ReA with C. trachomatis in 21 (46.6%) and in patients with ReA with combined infection (C. trachomatis + EBV) – in 20 (46.5%) patients. The level of IgG to C. trachomatis was elevated in 40 (93.0%) in the group of patients with ReA with C. trachomatis and in patients with ReA with combined infection (C. trachomatis + EBV) – in 42 (96.7%) patients. The higher level of IgM to C. trachomatis in the group of patients with ReA with C. trachomatis was 16 (35.6%), and in patients with ReA with combined infection (C. trachomatis + EBV) – in 17 (39.5%) patients.

Patients underwent PCR studies of peripheral blood and mucous urogenital investigations. PCR in the blood of patients with ReA with C. trachomatis detected chlamydial DNA in 15 (33.3%) patients and with combined infection (C. trachomatis + EBV) in 12 (27.9%) patients. Chlamydia trachomatis DNA in the group of patients with ReA with C. trachomatis in urogenital scrapings was detected in 22 (48.9%) patients and patients with combined infection (C. trachomatis + EBV) – in 20 (46.5%).

HLA-B27 was revealed in 29 (64.4%) ReA patients compared to 24 (55.8%) patients with combined infection and 32 (91.4%) patients with AS (Group 4), while in control group of 25 it was present in only 8% of individuals.

Antibodies to EBV (IgG EBV) at levels six times or more above the reference range were detected in all patients with ReA with combined infection (C. trachomatis + EBV). At the same time, an elevated level of IgM of EBV was recorded in seven patients (16.3%) in the group of ReA patients with combined infection (C. trachomatis + EBV).

We analyzed the features of the detection of EBV DNA in patients with ReA. In patients with ReA with C. trachomatis, EBV DNA was not detected in any biological medium. As can be seen in Fig. 1, in patients with ReA with combined infection (C. trachomatis + EBV), DNA of EBV was detected in the blood, saliva, and mucosa of the pharynx simultaneously in 3 (6.93%) patients, in saliva and of the mucosa of the pharynx – in 24 (55.82%), in the mucosa of the pharynx – in 16 (32.6%) and only in saliva – in two patients (4.65%).

Fig. 1.

EBV DNA percentage of ReA patients with combined infection (C. trachomatis + EBV) in different biological mediums: 1: blood + saliva + mucosa of the pharynx; 2: saliva + mucosa of the pharynx; 3: mucosa of the pharynx; 4: saliva

Laboratory study results

To quantify the overall impact of group status on serum AGE-related markers, we first estimated non-parametric effect sizes from the omnibus Kruskal–Wallis tests. AGE10 showed a large between-group effect (ε² = 0.565; 95% CI 0.435–0.748), indicating that a substantial proportion of variability in AGE10 is attributable to group differences. fAGE total showed a moderate effect (ε² = 0.208; 95% CI 0.037–0.521), consistent with meaningful but less pronounced group separation. By contrast, Pentosidine (ε² = 0.026; 95% CI − 0.046–0.311) and IC AGE10-anti-AGE10 (ε² = 0.105; 95% CI 0.032–0.316) exhibited small effects with wide confidence intervals, reflecting limited precision due to modest sample sizes.

Given the pronounced group effect for AGE10 and the moderate effect for fAGE total, we next examined detailed between-group differences for these markers using Kruskal-Wallis tests followed by Dunn’s post-hoc comparisons with Benjamini–Hochberg FDR correction (Figs. 2, 3 and 4).

Fig. 2.

Total fluorescent AGEs (fAGE) concentration in patients’ serum in all investigated groups as measured with 370/440 nm of excitation/emission, respectively. Box plots show the median (horizontal line), 25th and 75th percentiles (box boundaries, interquartile range, IQR), and whiskers extending to the most extreme data points within 1.5 × IQR from the box. Individual data points beyond the whiskers are shown as outliers. Color coding: Control (blue), ReA with C. trachomatis (purple), ReA with C. trachomatis + EBV (orange), AS (red). Median (IQR) fAGE concentrations were: Control 1249.0 (1080.0–1414.2) a.u., ReA with C. trachomatis 936.8 (613.6–1062.7) a.u., ReA with C. trachomatis + EBV 1154.3 (1084.5–1287.8) a.u., and AS 1521.2 (1280.3–2022.5) a.u. Group differences were assessed using the Kruskal–Wallis test (Kruskal–Wallis H = 10.56, p = 0.014), followed by Dunn’s post-hoc comparisons with Benjamini–Hochberg false discovery rate (FDR) correction; for fAGE, significant differences were observed for Control vs. ReA with C. trachomatis (q = 0.0446) and ReA with C. trachomatis vs. AS (q = 0.0054). Exact sample sizes were: Control n = 20, ReA n = 13, ReA with C. trachomatis + EBV n = 6, AS n = 8

Fig. 3.

Pentosidine concentration in patients’ serum in studied groups as measured with fluorescence at 335/385 nm of excitation/emission, respectively. Color coding: Control (blue), ReA with C. trachomatis (purple), ReA with C. trachomatis + EBV (orange), AS (red)

Fig. 4.

AGE10 antigen and AGE10 immune complexes across patient groups. (A) Box plots showing AGE10 antigen concentrations (µg/ml) in different patient groups. (B) Box plots displaying immune complex (IC) AGE10–anti-AGE10 concentrations (µg/ml) across patient groups. (C) Combined index log(IC + 1) − log(AGE10 + 1) across the four study groups. The global p-value above the plot is from the Kruskal–Wallis test; p-values shown on the connecting lines represent Dunn post-hoc pairwise comparisons with Benjamini–Hochberg correction. (D) Line graph connecting group means ± SD, illustrating the inverse relationship between AGE10 antigen levels (blue line, increasing trend) and IC AGE10–anti-AGE10 levels (red line, decreasing trend) across patient groups. Correlation analyses are exploratory and should be interpreted cautiously in view of modest sample sizes and potential residual confounding by disease group and other clinical factors

Analysis of fluorescent AGEs (fAGEs) in patient serum revealed significant differences between groups (Kruskal–Wallis H = 10.56, p = 0.014; Fig. 2). Post-hoc Dunn’s tests with Benjamini–Hochberg FDR correction showed higher fAGE concentrations in AS compared with ReA caused by C. trachomatis (median 1521.2 [IQR 1280.3–2022.5] vs. 936.8 [613.6–1062.7] a.u.; q = 0.0054) and higher fAGE concentrations in Controls compared with ReA caused by C. trachomatis (median 1249.0 [1080.0–1414.2] vs. 936.8 [613.6–1062.7] a.u.; q = 0.0446). No other pairwise comparisons reached significance after FDR correction.

Pentosidine levels, when compared across all groups using the Kruskal-Wallis test, showed no statistically significant differences (Fig. 3).

By contrast, AGE10 concentrations demonstrated marked variation among groups (Kruskal-Wallis H = 23.42, p = 5.1 × 10⁻⁶; Fig. 4A). Median AGE10 levels (IQR) were 157.4 (11.1–231.8) µg/ml in Controls, 0.0 (0.0–0.0) µg/ml in ReA caused by C. trachomatis, 906.3 (893.7–1042.0) µg/ml in ReA with C. trachomatis + EBV, and 0.0 (0.0–0.0) µg/ml in AS. Dunn’s post-hoc tests with Benjamini-Hochberg FDR correction revealed significant differences between several group pairs: Control vs. ReA caused by C. trachomatis (q = 0.0063), Control vs. ReA with C. trachomatis + EBV (q = 0.0282), Control vs. AS (q = 0.0038), ReA caused by C. trachomatis vs. ReA with C. trachomatis + EBV (q = 8.0 × 10⁻⁵), and ReA with C. trachomatis + EBV vs. AS (q = 7.8 × 10⁻⁵). We also compared serum IC AGE10–anti-AGE10 concentrations among the four groups (Control, ReA with C. trachomatis, ReA with C. trachomatis plus EBV, and AS, Fig. 4B). Kruskal-Wallis analysis showed a trend towards higher IC levels in patient groups compared with Controls, but this did not reach statistical significance (H = 6.79, p = 0.07). Post-hoc Dunn tests with Benjamini-Hochberg correction did not reveal any significant pairwise differences between individual groups (all q ≥ 0.08).

Free AGE10 and IC AGE10–anti-AGE10 are biologically related, we constructed a simple combined index to reflect their relative balance in each subject, defined as log(IC + 1) − log(AGE10 + 1). Positive values of this index indicate relatively higher IC compared to free AGE10. This index showed significant differences across the four study groups (Control, ReA with C. trachomatis, ReA with C. trachomatis plus EBV, and AS; Kruskal–Wallis H = 22.33, p = 5.6 × 10⁻⁵). Dunn’s post hoc tests with Benjamini-Hochberg correction showed that the index was significantly higher in AS than in Controls (q = 0.0017) and ReA with C. trachomatis plus EBV (q = 8.0 × 10^(−5)), and higher in ReA with C. trachomatis than in ReA with C. trachomatis plus EBV (q = 0.0045), while other pairwise differences were not significant after correction (all q > 0.05, Fig. 4C and D).

The correlation between AGE10 [µg/ml] and immune complexes AGE10–anti-AGE10 [µg/ml] was analysed across four patient groups. In the AS group (Ankylosing Spondylitis, n = 8), correlation could not be calculated because AGE10 antigen levels exhibited zero variance, with all values approximating 0 µg/ml. In the Control group, Spearman’s ρ = − 0.502 (p = 0.034) indicated a significant moderate negative correlation, whereas in ReA, ρ = 0.561 (p = 0.073) suggested a non-significant positive trend. In ReA + EBV, Pearson’s r = 0.961 (p = 0.009; R² = 0.923) reflected a significant strong positive correlation.

Correlation methods were selected according to data distribution and the form of the relationship between variables. In the Control and ReA groups, AGE10 values showed marked deviations from normality (skewness, zero-inflation) on Shapiro–Wilk testing and visual inspection of histograms and Q–Q plots; therefore, we used Spearman rank correlation, which does not assume normally distributed data and is robust to outliers. In contrast, in the ReA + EBV group the scatter plot suggested an approximately linear relationship with less pronounced skew, so Pearson’s correlation was applied to quantify this strong linear association between AGE10 antigen and IC AGE10–anti-AGE10. A particularly striking finding was the complete absence of detectable AGE10 antigen in the AS group, coinciding with high immune complex AGE10–anti-AGE10 levels, while patients with ReA caused by C. trachomatis alone (Group 2) showed the same pattern, but to a lesser degree, and the Control group and patients with ReA associated with EBV coinfection (Group 3) showed the opposite pattern, with high levels of antigen and low levels of immune complex. The AS group demonstrated a clear trend toward higher concentrations of immune complexes compared to other groups (Fig. 4A, B). For better visualization the relationships of AGE10 and immune complexes the box plots are shown on Fig. 4C and the connectivity between groups is on Fig. 4D. One may conclude that, in addition to local accumulation, immune complex formation may be another pathogenic factor leading to reduced AGE10 levels.

Potential confounders considered a priori included age, sex and underlying disease group. Because of the modest sample sizes and collinearity between diagnostic group and AGE10/IC AGE10–anti-AGE10 profiles, we did not perform multivariable adjustment. Correlation results are therefore interpreted as exploratory and hypothesis-generating, and residual confounding by clinical characteristics cannot be excluded.

Discussion

In our investigations, we observed significant differences in the levels of fluorescent AGEs (fAGE) across patient groups. The AS group exhibited the highest fAGE levels, which were significantly elevated compared to both the control group and patients with ReA caused solely by C. trachomatis (see Fig. 2). Patients with combined infection (C. trachomatis + EBV) displayed intermediate fAGE levels, approximately 1.4 times higher than those with chlamydial infection alone, suggesting an additive effect of EBV in elevating fAGE.

Regarding AGE10 levels, our analysis revealed distinct group differences. The control group maintained physiological AGE10 levels, while the combined infection group showed elevated AGE10 compared to the chlamydial-only group. Notably, patients in the AS group exhibited virtually undetectable levels of AGE10. Consequently, the ratio of immune complexes to AGE10 was markedly high in both the chlamydial-only and AS groups (see Fig. 4). In contrast, pentosidine levels did not differ significantly among the groups (Fig. 3). It is interesting to note that the AS group presented a much higher total fAGE level compared to pentosidine, indicating that an additional portion of fluorescent compounds – or alternative AGE species – contributes to the overall fAGE measurement in AS patients.

These observations, together with the broader context of AGE metabolism and immune complex formation, suggest distinct pathogenic mechanisms between AS and ReA. In AS patients, the elevated fAGE levels alongside undetectable AGE10, combined with high immune complex ratios, may reflect a scenario of enhanced RAGE receptor activation and inflammatory signaling, potentially driven by mechanisms such as accelerated AGE10 degradation or increased sRAGE expression acts as a “decoy” receptor, capturing circulating AGEs and reducing their availability [36] Fig. 5. Low level of AGE10 has been also found in serum of patients with allergic rhinitis and chronic Epstein-Barr virus infection at different stages of virus persistence [7]. Immunochemical experiments revealed distinct lower level of circulating serum AGE10 in patients with Alzheimer’s disease, in relation to healthy controls [30]. In both of these diseases specific immune complexes AGE10–anti-AGE10 were found, which could contribute to pathogenicity [7, 30]. Conversely, in ReA, particularly in cases triggered by combined C. trachomatis and EBV infection, the data suggests that the additive effect of EBV may amplify oxidative stress and glycation processes, leading to higher fAGE and AGE10 levels compared to chlamydial infection alone. The distinct patterns observed here may indicate that, in AS, a unique or additional pathogenic mechanism is at work, one that fosters chronic inflammation and tissue damage even in the absence of detectable AGE10. These findings underscore the importance of considering both the qualitative and quantitative aspects of AGE formation in disease states. They also highlight the potential of targeting specific glycation pathways and immune complex generation as a therapeutic strategy, particularly in conditions where mixed infections exacerbate autoimmunization and disease severity.

Fig. 5.

Schematic model summarizing the proposed mechanisms of AGE10 and AGE10–immune-complex involvement in ReA and AS. Panel A - Sources of AGE formation. Multiple upstream triggers—including Chlamydia trachomatis, Epstein–Barr virus, HLA-B27 misfolding stress, intestinal dysbiosis, chronic inflammation, and metabolic dysregulation/hyperglycemia- converge on oxidative/glycative stress (ROS + reactive carbonyls). This promotes the formation of AGE10 and fluorescent AGEs (fAGE) through non-enzymatic glycation of proteins. Panel B - Divergent AGE10, immune-complex, and fAGE profiles across diagnostic groups. ReA associated with C. trachomatis + EBV shows high AGE10 levels, very few immune complexes, and moderate fAGE. ReA associated with C. trachomatis alone exhibits low AGE10, moderate immune-complex formation, and low–moderate fAGE. Ankylosing spondylitis (AS) is characterized by chronic autoinflammation, absence of AGE10, the highest AGE10–antibody immune-complex formation, and the highest fAGE levels. Panel C - Downstream immune and inflammatory pathways. AGE10 and AGE10–antibody immune complexes activate RAGE, triggering oxidative stress via NADPH oxidase and downstream PKCζ and IKKβ. RAGE signaling activates NF-κB, MAPK/AP-1, PI3K/Akt, JAK/STAT, and JNK pathways. These transcriptional programs induce pro-inflammatory cytokines (IL-6, IL-1β, TNFα), IL-8, IL-2, MMPs, and proliferation/survival-related genes. Immune-complex deposition enhances complement activation (C3→C5) and macrophage-driven inflammation. Persistent signaling promotes epigenetic dysregulation through alterations in DNA and histone methylation. Created in https://BioRender.com

ReA is an autoimmune disease that develops in response to an infection. It is unknown the main mechanism by which the interaction of the pathogen with the host occurs in ReA. Multiple mechanisms are likely involved in patients with ReA and infection of Chlamydia trachomatis: (1) the antigens of C. trachomatis decussate reaction with their proteins, provoking and immortalizing an autoimmune response through mediation type 2 T helper (Th2) cells and joint impairment have been coupled with a Th2 cytokine profile; (2) synovitis in ReA is mediated by pro-inflammatory cytokines – IL-6 and IL-17, which is one of the major cytokines increased in the synovial fluid; insufficiency of the regulatory mechanism can result in the elevation of pro-inflammatory cytokine production and a poor prognosis of the course of the disease; (3) abnormalities in antigen presentation due to downregulation of TLR-4 costimulatory receptors. Subsequent studies implicated Toll-like receptors (TLRs) type 2 (TLR-2) polymorphism associated with acute ReA, suggesting that persistent infection can play a role in developing associated ReA [37].

Research suggests that the Epstein-Barr virus (EBV) infection could potentially contribute to the development of various systemic autoimmune diseases (SADs) – (rheumatoidal arthritis (RA), systemic lupus erythematosus (SLE), Sjögren’s syndrome (SS), systemic scleroderma (SSc), and organ-specific autoimmune diseases – autoimmune thyroiditis, multiple sclerosis (MS), etc [21, 38–40]. In our study, we observed that coinfection with C. trachomatis and EBV was associated with a more severe clinical course of ReA individuals. Patients with combined infection exhibited a higher number of swollen and tender joints, indicating more extensive joint involvement. Additionally, these patients had significantly higher disease activity scores (DAREA) and elevated CRP levels, reflecting greater systemic inflammation. Depending on the host immune system and environmental factors, some patients can eventually suffer from EBV-related diseases, as a result of EBV infection of other cell types (T cells, NK cells, NKT cells, monocytes/macrophages, etc.). The EBV infection in human lymphocytes under in vitro conditions could cause the expansion of non-specific B lymphocytes and T CD8 + cells, leading to the production of polyclonal antibodies and the activation of cytotoxic T lymphocytes [41]. It remains unclear whether the role of EBV is primarily in initiating SADs (e.g., by molecular mimicry) or is simply due to the chronic relapsing-remitting nature of EBV infections [21]. However, the extent of this relationship was not known. The link between EBV and autoimmunity has been suspected but remains controversial [39–43]. Since EBV can relocate between epithelial and B cells, SADs often occur as overlapping syndromes with symptoms and characteristic autoantibodies (e.g., antinuclear antibodies and rheumatoid factors (RF)) suggestive of epithelial and/or B cell infection [21]. In B cells, EBV can also downregulate the MHC-II expression and Toll-like receptors and alter the interaction between MHC-II and the T cell receptors. The final “victory” of the immune system must rely on the cellular immune control of EBV infection and the interaction of T cells, NK cells, and NKT cells [21].

In several autoimmune diseases, an elevated range of cytokines, e.g., IFN-ɣ or TNF, might suppress the production and secretion of antibodies, impairing the response to the AGEs. Moreover, it is discovered that the concentration of AGEs in serum is increasing with patient age. They might cumulate in an organism throughout aging because AGEs are long-lasting molecules [44]. It was also discovered that IL-6 can promote B lymphocyte differentiation and the production of antibody-activated T lymphocytes. At the same time, IL-1 and IL-6 may promote the proliferation of smooth muscle cells, increase endothelial permeability, and cause vascular injury. IL-1β, IL-6, and TNF-α levels in the serum of patients with positive RAGE gene promoter methylation were significantly lower than those of patients with negative RAGE gene promoter methylation. Methylation is the main mechanism of epigenetics; it consists of changing the DNA molecule without changing its nucleotide sequence and affecting the transcription level. It is a part of the mechanism of gene expression regulation [45]. The nuclear histones are subjected to methylation and thus contribute to regulating chromatin status and transcriptional activity. One of the concepts of autoimmunization is related to this [46]. DNA methylation has been offered to participate in gene expression regulation and T-cell differentiation. The dysregulated T cells participate in different disease states, including autoimmune diseases. Th1 cells revealed an IFN-γ-demethylated promoter and fight with bacteria by production IFN-γ. A developing change in methylation states also differentiates CD4⁺ T cells from Th17 cells. Cooperation between DNA methylation and conserved intergenic elements controls transcription at the IL-17 locus [47]. Therefore, an increase in the level of AGEs may cause the activation of DNA-associated processes that stimulate the development of autoimmune complications.

Patients with ReA-caused C. trachomatis and EBV infection showed common properties of activation immune mechanisms, namely, activation of T cells, NK cells, NKT cells, monocytes/macrophages, etc., production of pro-inflammatory cytokines; breakage of regulatory mechanisms through influence on NF-κB; and downregulation of TLR-4 costimulatory receptors.

Reference data show a link between oxidative stress, AGEs formation, and autoimmune diseases. Since RA and AS are characterized by chronic inflammation, it is essential to determine the impact of non-spontaneous glycation damage on differential proteins that have been poorly recognized and linked with AS pathogenesis. Enhanced oxidative stress leads to the accumulation of AGEs and inflammatory disease, which has been observed in patients with RA [1]. AGEs also have been linked to the onset of disease as a neo-autoantigen in RA [4]. The absence of association with autoantibodies, including RF (hence the term “seronegative”), increased the frequency of AS, and the association with HLA-B27 confirms that ankylosing spondylitis belongs to autoimmune diseases. In our investigation, HLA-B27 was revealed in 29 (64.4%) ReA patients compared to 24 (55.8%) patients with combined infection and 32 (91.4%) patients with AS. The presence of HLA-B27 is believed to potentiate reactive arthritis by presenting bacterial antigens (e.g., C. trachomatis) to T cells, altering self-tolerance of the host immune system, increasing TNF-α production, promoting the invasion of microbes in the gut, and delaying clearance of causative organisms [48]. Because of the clinical interrelationships between ReA and AS, it has been proposed that AS may represent a chronic form of ReA, in which it may be difficult to identify the triggering pathogen if that event occurred in the remote past [14]. Thus, our results are consistent with these authors hypothesis, as we observe the highest ratio of immune complexes to free AGE10 thus in AS group. Therefore, evaluating disease activity in AS is crucial to further our understanding of the pathophysiology of AGEs formation and predicting prognosis.

Thus, most authors proved that glycation promotes the formation of AGEs and functions as a ligand for RAGE; both are important in activating the inflammatory signaling cascade in disease conditions. The non-enzymatic process results in glycation, nitrosylation, oxidation/reduction, etc., of the targeted protein. To expand pathogenic knowledge, it is also important to identify the differential occurrence of glycated proteins [4]. Many extracellular and/or intracellular endogenous stress/risk factors can maintain AGEs that affect proteins’ physiological role, leading to altered oxidative and inflammatory signaling cascades. The glycation process has been shown to promote not just amyloid formation but also fibril development and cytotoxicity. High concentrations of AGEs molecules (free, peptide‑bound, and protein‑bound) are found in the blood plasma, particularly in diabetic patients. The researchers found that the AGEs concentration positively correlated with different pathological conditions of patients and contributed to the formation of AGEs in cells and complications [49]. There is also a positive correlation between AGEs accumulation and the duration of the disease. The increased presence of AGEs in patients’ blood is still discussed. However, the authors suggested that AGEs mostly accumulate in tissues rather than plasma proteins [50].

Two basal mechanisms that explain the role of AGEs in diseases and disorders are (i) the covalent crosslinking of serum and extracellular matrix (ECM) proteins, lipids, and DNA, resulting in their biochemical impairment and cell disruption and (ii) the interaction of AGEs with their receptors, especially on the AGE-RAGE axis, launching a series of cascade reactions and signaling pathways, proliferation, autophagy, and apoptosis [51–53]. Chen and Guo also distinguish four more modes: (i) ROS generation (responsible for oxidative stress); (ii) mitochondrial function impairment; (iii) AGEs as antigens inducing immune responses; and (iv) allergies caused by AGEs [3].

Finally, AGEs are markers of glycation process in organisms; they can affect tissues structurally (e.g., through crosslinking with proteins, they damage extracellular matrix) and functionally (e.g., due to interactions with receptors). Our study found high AGEs levels as a byproduct of oxidative stress in patients with combined infection and ReA. It is another mechanism (via complex AGEs/RAGEs, pro-inflammatory cytokines, and TLRs) of the development of autoimmunity. We propose that heightened AGE accumulation acts as a gateway to various immunopathological disorders, culminating most significantly in the manifestation of autoimmune complications.

The level of AGE10 is considered as biochemical marker of tissue damage, especially in case if increased level, whereas lowered level may indicate the formation of its immune complexes. In patients with reactive arthritis with C. trachomatis, a reduced level of AGE10 products is observed, allowing to speculate that glycation products immune complexes causing clinical symptoms that occur in these patients. The AGE10 could be also found in immune complexes of sera from patients with Alzheimer disease [30] and patients with allergic rhinitis and chronic Epstein-Barr virus infection [7]. Thus diminished level of AGE10 could be caused, besides local accumulation, also by immune complexes formation, a pathogenic factor. In cases of reactive arthritis with combined C. trachomatis + EBV infection the amount of AGE10 was much higher than in control group, what indicates that combined infections result in high content of AGE10.

We anticipate our approach will be a foundational in cell biology due to AGE10 compound is most likely related to an advanced lipoxidation end-product. Thus, further studies relevant to the discovery of MAGE may contribute to clarifying disease mechanisms and to the development of novel therapeutic options for several diseases.

Our findings highlight distinct patterns of AGE metabolism and immune complex formation between ReA and AS patients, suggesting different pathogenic mechanisms in these conditions. The marked differences in AGE profiles – particularly the contrasting patterns of AGE10 levels and immune complex formation – not only enhance our understanding of disease mechanisms but also emphasize the importance of differential diagnosis in clinical practice (Fig. 5). Moreover, these observations suggest that therapeutic approaches may need to be tailored according to the specific triggering factors in ReA, especially in cases of mixed infections, and might require different strategies in AS where AGE metabolism appears to follow a unique pattern. These insights open new avenues for both diagnostic refinement and therapeutic intervention, potentially leading to more personalized treatment strategies in inflammatory joint diseases.

Author contributions

I.K., S.Z., A.H., M.L., and V.C. carried out patient selection, laboratory work, and preparation of the preliminary manuscript draft. A.K. performed laboratory work, data analysis, and contributed to writing the initial version of the manuscript. E.J. conducted statistical analysis, prepared figures, and contributed to manuscript revisions. A.G. conceived the study, prepared the initial manuscript draft, and supervised the project. All authors reviewed and approved the final version of the manuscript.

Funding

The authors declare that no funds, grants, or other support were received during the preparation of this manuscript.

Data availability

The data that support the findings of this study are not openly available due to reasons of sensitivity and are available from the first author upon reasonable request. Data are located in controlled access data storage at Danylo Halytsky Lviv National Medical University (Lviv, Ukraine).

Declarations

Ethics approval

This study was performed in line with the principles of the Declaration of Helsinki. The research was approved by the Ethics Committee at the Danylo Halytsky Lviv National Medical University (protocol #6, 2017/03/08).

Consent to participate

Informed consent was obtained from all individual participants included in the study.

Clinical trial number

not applicable.

Competing interests

The authors declare no competing interests.