Abstract

1 Introduction

Endothelial dysfunction is characterized by an imbalance in the bioavailability of active substances of endothelial origin, being critical in the development of several autoimmune and inflammatory diseases such as rheumatoid arthritis (RA) (Sattar et al., 2003; González-Gay et al., 2008; Murdaca et al., 2012; Steyers and Miller, 2014; Yang et al., 2016). In this sense, key factors in the vasculopathy and inflammatory response include vascular cell adhesion molecule-1 (VCAM-1), one of the main responsibles of the firm adhesion of leukocytes to the endothelium (Dessein et al., 2005; Navarro-Hernández et al., 2009; Murdaca et al., 2012; Steyers and Miller, 2014; Wang et al., 2015; Kong et al., 2018), and the monocyte chemoattractant protein-1 (MCP-1), a chemokine that regulates the endothelial transmigration of monocyte (Hayashida et al., 2001; Rantapää-Dahlqvist et al., 2007; Deshmane et al., 2009; Dessein et al., 2013; Yang et al., 2016). Besides, asymmetric dimethylarginine (ADMA) is also involved in these disorders, regulating the vascular tone and inhibiting the nitric oxide synthesis (Dimitroulas et al., 2012; Murdaca et al., 2012; Steyers and Miller, 2014; Dimitroulas and Kitas, 2019; Zhao et al., 2019).

Interestingly, pulmonary endothelium dysfunction has also been described as a key stage for the development of lung damage and the subsequent onset and progression of interstitial lung disease (ILD). ILD represents one of the main causes of mortality in patients with RA, being the early detection of pulmonary involvement crucial to avoid an irreversible lung damage in these patients (Gómez-Carrera and Bonilla-Hernan, 2013; Bacha et al., 2017; Hyldgaard et al., 2017; Zamora-legoff et al., 2017; Bendstrup et al., 2019; Manfredi et al., 2021). However, the early diagnosis of RA-ILD⁺ constitutes a challenge for the clinicians because of the absence of symptoms in early or mild disease in some patients as well as the similarity of clinical, pathological, and epidemiological features with idiopathic pulmonary fibrosis (IPF), the most severe ILD (Paulin et al., 2015; Bendstrup et al., 2019; Atienza-Mateo et al., 2020; Manfredi et al., 2021).

In this context, we recently described that endothelial progenitor cells and angiogenic T cells, key cells in the vascular endothelium repair, are involved in the pathogenesis of ILD linked to autoimmune diseases (Pulito-Cueto et al., 2020; Pulito-Cueto et al., 2021; Pulito-Cueto et al., 2022). Hence, it is conceivable to think that other endothelial dysfunction-related markers could also contribute to the early identification of ILD in patients with RA. Nevertheless, this issue is far from being fully elucidated yet.

Accordingly, the aim of this study was to explore the functional role of VCAM-1, MCP-1 and ADMA, assessing both protein and mRNA, as potential biomarkers of pulmonary fibrosis in RA-ILD⁺.

2 Materials and methods

3 Results

3.1 Differences in the demographic and clinical features between the study groups

Demographic and clinical features including sex, age, smoking history, PFTs and HRCT patterns were collected from all the patients (Table 1).

TABLE 1

Demographic and clinical characteristics of all the patients included in the study.

	RA-ILD+ n = 21	RA-ILD− n = 25	IPF n = 21	p RA-ILD+ vs. RA-ILD−/RA-ILD+ vs. IPF
Sex (women), n (%)	9 (45.9)	15 (60.0)	7 (33.3)	0.84/0.53
Age at study (years), mean ± SD	66.5 ± 10.1	60.1 ± 11.8	69.2 ± 10.0	0.05/0.38
Smoking ever, n (%)	13 (65.0)	13 (52.0)	16 (76.2)	0.50/0.32
RA duration (years), mean ± SD	9.2 ± 10.2	4.1 ± 7.4	—	0.06/—
CRP (mg/dl), mean ± SD	1.1 ± 1.1	0.5 ± 0.5	—	0.04/—
ESR (mm/1st hour), mean ± SD	22.8 ± 27.2	14.4 ± 12.4	—	0.24/—
Disease activity			—
DAS28-CRP, mean ± SD	3.4 ± 1.4	2.2 ± 1.1	—	<0.01/—
DAS28-ESR, mean ± SD	3.4 ± 1.8	2.6 ± 1.6	—	0.12/—
Antibody status			—
RF+, n (%)	17 (81.0)	11 (44.0)	—	0.01/—
ACPA+, n (%)	19 (90.4)	15 (60.0)	—	0.02/—
Pulmonary function tests
FVC (% predicted), mean ± SD	95.2 ± 24.1	99.2 ± 16.0	84.9 ± 14.7	0.58/0.10
FEV1 (% predicted), mean ± SD	92.2 ± 21.0	94.9 ± 22.0	87.3 ± 19.6	0.67/0.43
FEV1/FVC (% predicted), mean ± SD	77.8 ± 9.1	93.6 ± 12.3	79.7 ± 7.8	<0.01/0.47
DLCO (% predicted), mean ± SD	43.3 ± 15.9	79.9 ± 20.0	43.6 ± 18.4	<0.01/0.97
HRCT
Pulmonary involvement	21 (100.0)	0 (0.0)	21 (100.0)	—
UIP pattern, n (%)	11 (52.4)	—	21 (100.0)	—
Probable UIP pattern, n (%)	2 (9.5)	—	0 (0.0)	—
NSIP pattern, n (%)	7 (33.3)	—	0 (0.0)	—
Non-NSIP pattern, n (%)	1 (4.8)	—	0 (0.0)	—

Open in a separate window

RA, rheumatoid arthritis; ILD, interstitial lung disease; IPF, idiopathic pulmonary fibrosis; SD, standard deviation; CRP, C-reactive protein; ESR, erythrocyte sedimentation rate; DAS, disease activity score; RF, rheumatoid factor; ACPA, anti-cyclic citrullinated peptide antibodies; FVC, forced vital capacity; FEV1, forced expiratory volume at first second; DLCO, diffusing capacity of the lung for carbon monoxide; HRCT, high resolution computed tomography; UIP, usual interstitial pneumonia; NSIP, non-specific interstitial pneumonia. Significant results are highlighted in bold.

Statistically significant differences were found between patients with RA-ILD⁺ and RA-ILD^- regarding the following demographic and clinical features: age, C-reactive protein (CRP) levels, disease activity score 28-CRP, rheumatoid factor and anti-cyclic citrullinated peptide antibodies status (Table 1). Likewise, we also observed differences in forced expiratory volume at first second/forced vital capacity and diffusing capacity of the lung for carbon monoxide (Table 1). However, no differences were disclosed between RA-ILD⁺ and IPF patients (Table 1).

3.2 VCAM-1, MCP-1 and ADMA associated with RA-ILD⁺

On the one hand, VCAM-1, MCP-1 and ADMA levels were increased in patients with RA-ILD⁺ in relation to RA-ILD^- patients (p < 0.01 in all cases, Figures 1A,C,E, Supplementary Table S1). Moreover, patients with RA-ILD⁺ exhibited a higher CCL2 mRNA expression than RA-ILD^- patients (p < 0.01, Figure 1D, Supplementary Table S1). Furthermore, a lower PRMT1 mRNA expression was found in patients with RA-ILD⁺ in comparison with their negative counterpants (p = 0.04, Figure 1F, Supplementary Table S1).

Open in a separate window

FIGURE 1

Differences in serum levels of VCAM-1 (A), MCP-1 (C) and ADMA (E) as well as mRNA expression of VCAM1 (B), CCL2 (D) and PRMT1 (F) between RA-ILD⁺ patients and the comparative groups. RA: rheumatoid arthritis; ILD, interstitial lung disease; IPF, idiopathic pulmonary fibrosis; VCAM-1, vascular cell adhesion molecule 1; MCP-1, monocyte chemoattractant protein-1; ADMA, asymmetric dimethylarginine. Significant results are highlighted in bold.

On the other hand, patients with RA-ILD⁺ showed significantly higher serum levels of VCAM-1, MCP-1 and ADMA compared to those with IPF (p < 0.01 in all cases, Figures 1A, C, E, Supplementary Table S1). Additionally, a lower expression of VCAM1, CCL2, and PRMT1 was observed in patients with RA-ILD⁺ when compared with those with IPF (p = 0.04, p = 0.05, p = 0.02, respectively, Figures 1B, D, F, Supplementary Table S1).

3.3 VCAM-1, MCP-1 and ADMA as diagnostic biomarkers for RA-ILD⁺

To evaluate the ability of serum VCAM-1, MCP-1 and ADMA levels for discriminating patients with RA-ILD⁺ from those with RA-ILD^-, ROC curves were drawn for each biomarker (Figure 2A). Interestingly, the AUC was 0.704, 0.765 and 0.695, respectively (Table 2). The optimal cutoff value for VCAM-1, MCP-1 and ADMA showing the best sensitivity and specificity was 1680.0 ng/ml, 468.2 pg/ml and 0.5795 μmol/L, respectively (Table 2).

Open in a separate window

FIGURE 2

ROC curves of VCAM-1, MCP-1 and ADMA for the screening of RA-ILD⁺ and RA-ILD⁻ (A) and RA-ILD⁺ and IPF (B). RA, rheumatoid arthritis; ILD, interstitial lung disease; IPF, idiopathic pulmonary fibrosis; VCAM-1, vascular cell adhesion molecule 1; MCP-1, monocyte chemoattractant protein-1; ADMA, asymmetric dimethylarginine.

TABLE 2

Receiver operating characteristic curves analysis for the discrimination of RA-ILD⁺ from RA-ILD⁻ and IPF.

	AUC (95% CI)	Optimal cutoff value	Sensitivity (%)	Specificity (%)
RA-ILD+ vs. RA-ILD−
VCAM-1	0.704 (0.520–0.888)	1680.0 ng/ml	68.4	87.5
MCP-1	0.765 (0.607–0.923)	468.2 pg/ml	75.0	78.3
ADMA	0.695 (0.542–0.849)	0.5795 μmol/L	33.3	100.0
RA-ILD+ vs. IPF
VCAM-1	0.813 (0.663–0.964)	1220.0 ng/ml	73.7	95.0
MCP-1	0.863 (0.746–0.979)	458.4 pg/ml	75.0	90.0
ADMA	0.783 (0.638–0.927)	0.4855 μmol/L	81.0	68.4

Open in a separate window

RA, rheumatoid arthritis; ILD: interstitial lung disease; IPF, idiopathic pulmonary fibrosis; AUC, area under the curve; CI, confidence interval; VCAM-1, vascular cell adhesion molecule 1; MCP-1, monocyte chemoattractant protein-1; ADMA, asymmetric dimethylarginine.

In addition, ROC curves were used for the differentiation of patients with RA-ILD⁺ from those with IPF (Figure 2B). Of note, the AUC was 0.813 for VCAM-1, 0.863 for MCP-1 and 0.783 for ADMA (Table 2). The optimal cutoff value was 1220.0 ng/ml, 458.4 pg/ml and 0.4855 μmol/L, respectively (Table 2).

3.5 Relationship of VCAM-1, MCP-1 and ADMA with clinical characteristics intrinsic of the rheumatic disease in RA-ILD⁺ and RA-ILD⁻ patients

Regarding RA-ILD⁺ patients, a positive partial correlation was found between MCP-1 levels and RA duration (r = 0.542, p = 0.037, Figure 3A, Supplementary Table S3). This was also the case when PRMT1 mRNA expression was assessed (r = 0.645, p = 0.013, Figure 3B, Supplementary Table S3). In addition, ADMA serum levels were positively partially correlated with CRP (r = 0.585, p = 0.022, Figure 3C, Supplementary Table S3). No relationship between VCAM-1, MCP-1 and ADMA protein levels and other clinical characteristics of patients with RA-ILD⁺ were found (Supplementary Table S3). Likewise, no other significant results were obtained regarding VCAM1, CCL2 and PRMT1 mRNA expression (Supplementary Table S3).

Open in a separate window

FIGURE 3

Relationship of MCP-1 (A), PRMT1 mRNA expression (B) and ADMA (C) with clinical characteristics of RA-ILD⁺ patients. RA, rheumatoid arthritis; MCP-1, monocyte chemoattractant protein-1; CRP, C-reactive protein; ADMA, asymmetric dimethylarginine.

With respect to patients with RA-ILD^-, no association was found between VCAM-1, MCP-1 and ADMA and clinical characteristics intrinsic of RA, both when assessing protein levels and mRNA expression (Supplementary Table S4).

4 Discussion

Endothelial dysfunction, an early stage in the vascular damage, is observed in patients with RA (González-Gay et al., 2008) and is also implicated in ILD (Gómez-Carrera and Bonilla-Hernan, 2013; Bacha et al., 2017; Hyldgaard et al., 2017; Zamora-legoff et al., 2017; Bendstrup et al., 2019; Manfredi et al., 2021). RA-ILD⁺ diagnosis is often delayed due to the heterogeneity and unpredictability of ILD, constituting a major concern for the clinicians (Paulin et al., 2015; Bendstrup et al., 2019; Atienza-Mateo et al., 2020; Manfredi et al., 2021).

Given that pulmonary vasculopathy is important for the development of ILD, endothelial dysfunction-related markers could help to the early identification of RA-ILD⁺. Interestingly, this study provides evidence, for the first time, of an association of VCAM-1, MCP-1 and ADMA with pulmonary fibrosis in RA-ILD⁺.

First, we found that VCAM-1 serum levels were increased in patients with RA-ILD⁺ in relation to those with RA-ILD^- and IPF. In this context, it has been described that VCAM-1 is induced by both inflammatory cytokines and hypoxia (Peter et al., 1999; Dessein et al., 2005; Navarro-Hernández et al., 2009; Cook-Mills et al., 2011; Murdaca et al., 2012; Wang et al., 2015; Kong et al., 2018; McGroder et al., 2019), which may explain the increase of this protein observed in our RA-ILD⁺ patients. Likewise, its ability to stimulate the fibroblast proliferation may be related to the fibrotic properties of this disease (Agassandian et al., 2015; McGroder et al., 2019). Moreover, our findings revealed a decrease of VCAM1 expression in RA-ILD⁺ patients with respect to the comparative groups, suggesting the involvement of post-transcriptional regulatory mechanisms (Greenbaum et al., 2003; Stevens and Brown, 2013; Li et al., 2020).

In the second place, we also disclosed a relationship of MCP-1 with lung involvement in RA since patients with RA-ILD⁺ presented higher levels of MCP-1 than those with RA-ILD⁻ and IPF. Of note, we noticed that information on the role of MCP-1 in RA-ILD⁺ was limited to a single study in Asian population that also showed the same differences between RA-ILD⁺ and RA-ILD⁻ patients (Ling et al., 2010). In this regard, a profibrotic role of MCP-1 has been previously described in other autoimmune diseases such as systemic sclerosis (Hasegawa et al., 1999; Schmidt et al., 2009; Yalçinkaya et al., 2016). The higher levels of MCP-1 in RA-ILD⁺ may also be caused by the chronic systemic exposure to proinflammatory cytokines characteristic of RA (Hayashida et al., 2001; Rantapää-Dahlqvist et al., 2007; Deshmane et al., 2009; Lin et al., 2014). It is noteworthy that an increase of CCL2 expression was observed in our patients with RA-ILD⁺ compared to those with RA-ILD⁻, whereas IPF patients showed the highest CCL2 expression. These results could indicate that CCL2 expression increases in response to ILD severity, supporting its known profibrotic role (Hasegawa et al., 1999; Schmidt et al., 2009; Ling et al., 2010). Moreover, we found a positive correlation between MCP-1 serum levels and RA duration in RA-ILD⁺ patients, which is in line with its key endothelial function reported in RA (Hayashida et al., 2001; Rantapää-Dahlqvist et al., 2007; Deshmane et al., 2009; Lin et al., 2014).

Finally, RA-ILD⁺ patients showed an increase of ADMA serum levels compared to both RA-ILD⁻ and IPF patients, pointing out ADMA as a biomarker of RA-ILD⁺. The highest ADMA levels in RA-ILD⁺ patients make sense considering that ADMA has been previously described as a contributor of lung fibrosis and that its degradation may be inhibited by hypoxia and inflammatory cytokines, common processes of this disease (Ito et al., 1999; Millatt et al., 2003; Sattar et al., 2003; Sandoo et al., 2012; Janssen et al., 2013). Indeed, a positive correlation of ADMA and CRP was found in our cohort of RA-ILD⁺ patients. It should also be mentioned that RA-ILD⁺ patients presented the lowest PRMT1 expression, probably as a result of post-transcriptional regulatory mechanisms (Greenbaum et al., 2003; Stevens and Brown, 2013; Li et al., 2020). Furthermore, PRMT1 expression was positively correlated with RA duration in RA-ILD⁺ patients, confirming its previously described relationship with RA (Sandoo et al., 2012; Dimitroulas and Kitas, 2019; Zhao et al., 2019).

Considering all these findings, we can hypothesize that an additive effect combining the vascular damage caused by chronic inflammation and other features of RA as well as that associated with the fibrotic processes typical of ILD is related to increased serum concentrations of VCAM-1, MCP-1 and ADMA in RA-ILD⁺. In addition, serum levels of VCAM-1, MCP-1 and ADMA higher than 1680.0 ng/ml, 468.2 pg/ml and 0.5795 μmol/L may differentiate RA-ILD⁺ from RA-ILD^-. Likewise, serum levels higher than 1220.0 ng/ml, 458.4 pg/ml and 0.4855 μmol/L, respectively, may discriminate RA-ILD⁺ from IPF. These results could be relevant in the clinical practice for the diagnosis of RA-ILD⁺.

In conclusion, our study suggests that VCAM-1, MCP-1 and ADMA can be useful biomarkers to identify the presence of ILD in patients with RA, as well as to discriminate between RA-ILD⁺ and IPF, contributing to the early diagnosis of RA-ILD⁺.

Acknowledgments

References