High YKL-40 serum levels and its expression in the muscle tissues of patients with antisynthetase syndrome

The protein chitinase-3-like-1 (YKL-40) is rarely analyzed in patients with myositis. Therefore, we aimed to evaluate YKL-40 serum levels; correlate them with laboratory and clinical parameters, disease status, and treatment schemes; and analyze the YKL-40 expression in the muscle tissues of patients with antisynthetase syndrome (ASSD). This cross-sectional single-center study included 64 adult patients with ASSD who were age-, gender-, and ethnicity-matched to 64 healthy control individuals. Their YKL-40 serum levels were analyzed using the Enzyme-Linked Immunosorbent Assay (ELISA) kit method, while YKL-40 expression in muscle tissues was analyzed using an immunohistochemical technique. Disease status was assessed using the International Myositis Assessment and Clinical Studies Group (IMACS) set scores. The patients’ mean age was 44.8 ± 11.8 years, and median disease duration was 1.5 (0.0–4.0) years. These patients were predominantly female (82.8%) and Caucasian (73.4%). Most patients had stable disease. The median YKL-40 serum level was significantly higher in patients with ASSD when compared to the healthy individuals: 538.4 (363.4–853.1) pg/mL versus 270.0 (201.8–451.9) pg/mL, respectively; P < 0.001. However, YKL-40 serum levels did not correlate with any clinical, laboratory, disease status, or therapeutic parameters (P > 0.050), except tumor necrosis factor alpha (TNF-α) serum levels (Spearman’s correlation, rho = 0.382; P = 0.007). YKL-40 was highly expressed by inflammatory cells found in muscle biopsy specimens. High YKL-40 serum levels were observed in patients with ASSD and correlated positively with TNF-α serum levels. Moreover, YKL-40 was expressed by the inflammatory cells of the muscle tissue.

In rheumatoid arthritis, YKL-40 serum levels are not only increased but also correlate positively with the disease activity [11][12][13]. However, YKL-40 is not associated with radiographic progression or predictive values for disease remission [14]. In patients with systemic sclerosis, YKL-40 serum levels correlate mainly with joint involvement [16], lung involvement (fibrosis or reduction in the diffusing capacity for carbon monoxide [DLCO]), digital articular deformities through cutaneous retraction, and reduced patient survival [17]. High YKL-40 serum levels are also described in patients with dermatomyositis and polymyositis, correlating with the disease activity and severity and with pulmonary involvement [19][20][21].
However, no studies have assessed YKL-40 serum levels exclusively in patients with ASSD, which was the primary motivation for this study. We aimed to correlate YKL-40 serum levels with clinical and laboratory parameters, disease status, and therapeutic schemes, as well as to analyze the YKL-40 expression in muscle tissue specimens from patients with ASSD.

Patients and methods
This cross-sectional single-center study took place from 2017 to 2019 and included 66 adult patients with ASSD followed up at the outgoing clinic of our service. All patients fulfilled the modified ASSD criteria classification proposed by Connors et al. [1], but with modifications to increase its specificity, including the presence of at least two of the following three manifestations: muscle, joint, and lung involvement; and in addition to persistent fever; Raynaud's phenomenon; "mechanic's hands;" and the presence of serum anti-aminoacyl-tRNA synthetase autoantibody.
Muscle involvement was defined as the presence of progressive and proximal limb muscle weakness, increased serum levels of muscle enzymes, such as creatine phosphokinase (CPK), myopathic pattern on electroneuromyography, and muscle biopsy compatible with inflammatory myopathy. Joint involvement was defined as non-deforming and non-erosive arthritis or arthralgia. Additionally, as part of the internal protocol of our service, patients underwent computed tomography (CT) chest scans at disease onset and during follow-up. In this study, lung involvement was defined as the presence of incipient pneumopathy, pulmonary nodules, ground-glass opacities with or without bronchiectasis, and pulmonary fibrosis (e.g., honeycombing areas).
We excluded pregnant patients and those with other systemic autoimmune diseases (overlap syndrome), chronic or acute infection, neoplasia-associated myositis, and uncontrolled and serious comorbidities (e.g., liver disease, nephropathy, and heart disease). Observing the internal protocol of our service, a screening for neoplasms and other lung diseases was performed in all cases of myopathies.
The following pre-standard and pre-parameterized data were collected from the patients' interviews: demographic information (current age, gender, ethnicity), clinical information (disease duration; muscle, pulmonary, and joint involvement; Raynaud's phenomenon; presence of fever; and "mechanic's hands"), laboratory findings (serum levels of CPK, lactate dehydrogenase, aspartate aminotransferase, alanine aminotransferase), disease status, treatment (immunosuppressive, immunomodulatory, and glucocorticoid therapies), and comorbidities. Moreover, on the interview day, blood samples (20 mL) were taken for laboratory analysis after an eight-hour fasting.
Current clinical and laboratory disease status was assessed by applying the following questionnaires and International Myositis Assessment & Clinical Studies Group (IMACS) scores: Myositis Disease Activity Assessment Visual Analogue Scales (MYOACT) [23,24], Manual Muscle Testing (MMT-8) [25], overall disease assessment by a physician and by the patient using a visual analogue scale (VAS) [26], and the Health Assessment Quality (HAQ) [27].
Pulmonary assessment was performed using chest CT images. The following parameters were investigated: incipient pneumopathy, ground-glass opacities with or without bronchiectasis, pulmonary nodules, and pulmonary fibrosis (honeycombing areas) [1,28,29]. All chest CT images were analyzed independently by the two authors of the present study, who are experienced in connective tissue diseases (RCSC and GLBP). In case of disagreement between them, the images were discussed with radiology groups from our institute.
Chest CT images obtained during a period ranging from six months before to six months after the patients' participation were included in this study. When possible, the pulmonary assessment was supplemented by a pulmonary function test that considered the forced expiratory volume in one second (FEV1), forced vital capacity (FVC), FEV1/ FVC ratio, and DLCO.
The patients with ASSD were gender-, age-, and ethnicity-matched to healthy control individuals (patients' relatives or employees of our service) selected for convenience. All laboratory tests were performed on the control group, and they were asked to answer all interview questions, except those related to the disease data.
For the cytokine assessment, the material was centrifuged at 3000 rpm for 10 min immediately after blood collection (< 30 min), and the sera were aliquoted and stored at − 80°C for further cytokine analysis. The YKL-40 quantification was performed using a specific kit (Human CHI3L1 ELISA kit, RayBiotech, USA) and processed following the manufacturer's protocol. The interferon gamma (IFN-γ) and TNF-α tests were performed using the LUMINEX 100/200x xMAP technology (Millipore, USA), as described elsewhere [30].
Additionally, the YKL-40 expression and location were assessed in muscle tissues of three patients with ASSD, whose biopsies were taken from the vastus lateralis muscle upon diagnosis. Sequential 5-μm thick frozen sections were first stained by hematoxylin-eosin and, then, immunohistochemistry was performed. For immunohistochemical analysis, monoclonal antibodies (CD4, CD8, CD68, and YKL-40: Abcam, USA) were used. Frozen specimens were fixed for 10 min in acetone at 4°C. Endogenous peroxidase was blocked using 1% H 2 O 2 in absolute methanol three times for 10 min. After rinsing the specimens in phosphate buffered saline (PBS) (0.01 M, pH 7.4) for five minutes, these were incubated in fetal serum in a wet chamber for one hour at 37°C. Primary antibodies were diluted in PBS and albumin from 1% bovine serum was applied in wet chamber at 37°C overnight. Next, slides were washed in PBS and a secondary mouse biotinylated solution (StreptABComplex/HRP) was applied for 30 min at 37°C and rinsed in PBS. Then, the prepared StreptABComplex/HRP complex was applied and incubated for 30 min at 37°C. After rinsing in PBS and incubation, the reactions were visualized using a solution of chromogenic substrate (3,3′-diaminobenzidine tetrahydrochloride) for peroxidase. After a final rinse, hematoxylin counterstaining was performed. The slides were then mounted and covered with an aqueous-based mounting medium. All muscle specimens were prepared at the same time as a batch. Human amygdala was used as a positive control, while a muscle specimen from a patient with ASSD, but whose biopsy results were normal (without infiltrate inflammation), was used as a negative control. Inflammatory cell phenotyping (CD4, CD8, CD68, and YKL-40) was analyzed in the endomysium, perimysium, and pericapillary (endomysium and perimysium) areas in 10 fields (200x magnification).
For statistical analysis, the Kolmogorov-Smirnov test was used to assess each parameter distribution. The clinical and demographic characteristics were expressed as mean ± standard deviation (SD), for continuous variables, or as frequencies and percentages (%), for categorical variables. The median (interquartile 25th to 75th) was calculated for continuous variables that did not have a normal distribution. Comparisons between patient and control parameters were made using the Mann-Whitney test and Student t-test for continuous variables, while the Fisher's exact test or chi-square test were used to assess categorical variables. Spearman's correlation was used for analyzing the correlation (rho) between continuous variables and YKL-40 values. Values between 0 and ± 0.333 were considered weak; values between ± 0.333 and ± 0.666 were considered moderate; and those between ±0.666 and ± 1.000 were considered strong." The relation between YKL-40 values and other variables (identified as the presence or absence of a certain characteristics) was performed using the Student t-test. To analyze the factors that together influence YKL-40, a multiple regression was performed considering all variables with P value < 0.10 as explanatory and the value of YKL-40 as dependent. A value of P < 0.05 was adopted to indicate statistical significance. All analyses were performed using the statistical software SPSS version 22.0 (Chicago, IL, USA).

Results
Sixty-six patients with ASSD were initially included in the present study. Two patients were excluded, one due to overlap and the other due to neoplasia-associated myositis. No patients had acute or chronic infection, were pregnant, or had severe and decompensated comorbidities. Therefore, 64 patients with ASSD were evaluated and age-, gender-, and ethnicity-matched to 64 healthy controls (Fig. 1).
The initial and cumulative clinical manifestations, autoantibody panels, and comorbidities of patients with ASSD are shown in Table 1.
The current disease status of patients with ASSD at the time of inclusion in this study, lung parameters, and treatments are shown in Table 2.
Concerning the immunohistochemical analysis, YKL-40 was expressed by inflammatory cells. Figure 2 shows a muscle biopsy specimen from a patient with ASSD.

Discussion
To the best of our knowledge, this study is the first to show the relevance of YKL-40 in patients with ASSD. High YKL-40 serum levels were observed in patients with ASSD and correlated positively with TNF-α serum levels. Besides, YKL-40 was expressed by inflammatory cells in muscle tissues.
Although ASSD is a rare disease, this study employed strict inclusion criteria to increase the specificity in the selection of the involved patients. In this context, a significant sample was included and paired by age, gender, and ethnicity with healthy individuals, thus allowing for greater accuracy in comparative outcomes. Finally, information on patients was based on parameterized and pre-standardized data to ensure reliable study data.
YKL-40 is seldom described in studies regarding systemic autoimmune myopathies [19][20][21]. In this context, Hozumi et al. [19] showed high YKL-40 serum levels in Table 2 Demographic, current disease status, pulmonary involvement, treatment, and the cytokines serum levels of the patients with antisynthetase syndrome. Laboratory profiles of the control group patients with dermatomyositis and polymyositis with associated interstitial lung disease, and YKL-40 serum levels were associated with a worse prognosis. Furthermore, the authors correlated YKL-40 levels negatively with DLCO and arterial oxygen pressure (PaO 2 ), finding a high expression of YKL-40 in alveolar macrophages and bronchial epithelial cells. Notably, the authors found that 42% of their patients had positive anti-aminoacyl-tRNA synthetase autoantibodies. Therefore, they might have included patients with ASSD and not necessarily pure dermatomyositis and polymyositis cases with associated interstitial lung disease. Additionally, patients with neoplasm associated myositis were included. Gao et al. [20] also observed high YKL-40 serum levels in patients with dermatomyositis and polymyositis, which correlated positively with interstitial lung disease and with the disease activity and severity evaluated via MYOACT, C-reactive protein, ferritin, and erythrocyte sedimentation rates. However, the authors did not analyze anti-aminoacyl-tRNA synthetase autoantibodies in their patients.
Jiang et al. [21] observed a positive correlation between YKL-40 serum levels and patients with clinically amyopathic dermatomyositis and positive anti-MDA-5, especially in those with severe interstitial lung disease.
In our study, high YKL-40 serum levels were also observed in patients with ASSD. However, YKL-40 did not correlate with any clinical and laboratory parameters, disease status, or therapeutic schemes. There was also no correlation with any pulmonary parameters (e.g., patient's lung symptoms, chest CT images, and pulmonary function tests). Since this was a cross-sectional analysis, a prospective analysis of these parameters would be interesting.
Unlike limited studies regarding YKL-40 in myositis, a limitation of our study was the fact that we did not specifically analyze the YKL-40 expression in lung biopsy specimens. Nonetheless, the YKL-40 expression analyzed in muscle specimens reinforces the hypothesis that YKL-40 expression is systemic, not being limited to lung involvement.
YKL-40 can be expressed and secreted by various types of cells, including macrophages, neutrophils, and chondrocytes. Its expression is regulated by several cytokines, which mediate the proliferation, tissue remodeling, and production of other inflammatory mediators. In ASSD and systemic autoimmune inflammatory disease, the elevated YKL-40 serum levels might derive come from systemic inflammatory cells or local cells (e.g., blood, muscle, skin, lungs). This was the first study to demonstrate the increased expression of YKL-40 in the inflammatory cells of muscle tissue obtained from a patient with ASSD.
As previously mentioned, YKL-40 is regulated by various pro-inflammatory cytokines [8,9]; therefore, we assessed the TNF-α and IFN-γ serum levels. High TNF-α serum levels were observed in patients with ASSD, similar to what occurs in other systemic autoimmune myopathies [31,32]. TNF-α can induce YKL-40 production [8]. These data corroborated our study, in which a positive correlation was found between YKL-40 and TNF-α serum levels.
Although IFN-γ is involved in the ASSD pathophysiology, we did not find high cytokine serum levels in the  [31,33]. Therefore, it is difficult to detect. In the pathophysiology of systemic autoimmune myopathies, the participation of Th1 and Th2 cells is described as follows: IFN-γ is produced by initially present Th1 cells and, then, Th1 cells regulate Th2 cells using other inflammatory cytokines, including YKL-40 [34,35].
The present study had some limitations. First, spirometry was not performed in all patients because of various internal circumstances, such as patients with difficulty in performing spirometry (dyspnea) and lack of available spirometry equipment at patient selection, even with these tests being carried out only in a later phase of the ASSD diagnosis and treatment (> six months). Second, the degree of dyspnea, six-minute walking distance, and  oxygen saturation were not analyzed. Third, the analysis of pneumopathy in CT was only descriptive, since there is no criterion in sytemic autoimmune myopathies that quantifies the degree of lung involvement.

Conclusions
This study showed high YKL-40 serum levels in patients with ASSD, and these correlated positively with TNF-α serum levels. Although other studies have shown the expression of YKL-40 in lungs of patients with various systemic autoimmune myopathies, we have shown high YKL-40 levels in inflammatory muscle tissue cells.