Enhanced tryptophan-kynurenine metabolism via indoleamine 2,3-dioxygenase 1 induction in dermatomyositis

Objectives Extrahepatic tryptophan (Trp)-kynurenine (Kyn) metabolism via indoleamine 2,3-dioxygenase 1 (IDO1) induction was found to be associated with intrinsic immune regulation. However, the Trp-Kyn metabolism–associated immune regulation in dermatomyositis (DM) remains unknown. Therefore, we aimed to investigate the clinical relevance of the Trp-Kyn metabolism via IDO1 induction in DM. Methods Liquid chromatography-mass spectrometry (HPLC–MS) was used to examine the serum Kyn and Trp concentrations in DM. In addition, we used X-tile software to determine the optimal cutoff value of the Kyn/Trp ratio, a surrogate marker for Trp-Kyn metabolism. Spearman analysis was performed to evaluate the association of Trp-Kyn metabolism with muscle enzymes and inflammatory markers. Results DM patients had significantly higher serum Kyn/Trp ratio (× 10−3) when compared with the healthy controls. The serum Kyn/Trp ratio was positively correlated with the levels of muscle enzymes and inflammatory markers. In addition, the serum Kyn/Trp ratio significantly decreased (36.89 (26.00–54.00) vs. 25.00 (18.00–37.00), P = 0.0006) after treatment. DM patients with high serum Kyn/Trp ratio had a significantly higher percentage of muscle weakness symptoms (62.5% vs. 20.0%, P = 0.019) and higher levels of LDH (316.0 (236.0–467.0) vs. 198.0 (144.0–256.0), P = 0.004) and AST (56.5 (35.0–92.2) vs. 23.0 (20.0–36.0), P = 0.002)) than those with low serum Kyn/Trp ratio. Multiple Cox regression analyses identified ln(Kyn/Trp) (HR 4.874, 95% CI 1.105–21.499, P = 0.036) as an independent prognostic predictor of mortality in DM. Conclusions DM patients with enhanced Trp-Kyn metabolism at disease onset are characterized by more severe disease status and poor prognosis. Intrinsic immune regulation function via enhanced Trp-Kyn metabolism by IDO1 induction may be a potential therapeutic target in DM. Key Points • HPLC–MS identified increased serum Kyn/Trp ratio in DM patients, which positively correlated with levels of muscle enzymes and inflammatory markers and was downregulated upon treatment. • Cox regression analyses identified ln(Kyn/Trp) as an independent prognostic predictor of mortality in DM. • Monitoring intrinsic immune regulation function should be considered a potential therapeutic target in DM patients. Supplementary Information The online version contains supplementary material available at 10.1007/s10067-022-06263-3.


Introduction
Dermatomyositis (DM) is an autoimmune inflammatory disorder characterized by the presence of pathognomonic skin lesions (heliotrope rash, Gottron's papules/sign, poikiloderma, etc.) with or without muscular and extra-muscular manifestations (primarily lung involvement, followed by heart, esophagus, gastrointestinal tract, etc.) [1]. Disease activity is reflected by levels of muscle enzymes or inflammatory markers. Moreover, interstitial lung disease (ILD) and malignancy are frequently correlated with disease severity and prognosis [2][3][4].
Autoimmune-mediated pathological inflammation plays a crucial role in the pathogenesis of DM [5]. Corresponding anti-inflammatory therapeutic strategies, including glucocorticoids and immunosuppressants, were widely used in clinical practice [6,7]. However, the immune system itself can limit excessive inflammatory responses in normal physiology, termed "immune regulation." Immune homeostasis is a highly dialectical situation comprising immune regulation and inflammation. They interlace with each other and once the balance is broken, disease occurs. The function of immune regulation would have also altered under the disease conditions, and we propose that the intrinsic immune regulation function should provoke new ideas for disease monitoring and treatment.
The extrahepatic tryptophan (Trp)-kynurenine (Kyn) metabolism is a crucial metabolic foundation underlying the intrinsic immune regulation function [8]. In the later phase of inflammation, IFN-γ induces indoleamine 2,3-dioxygenase 1 (IDO1), the rate-limiting enzyme for Trp-Kyn metabolism, in extensive cells (monocytes, macrophages, plasmacytoid dendritic cells, etc.) converting Trp to Kyn to drive the intrinsic immune regulation function. Impairment in IDO1 induction exacerbates the underlying inflammatory pathology, and reversing this defect would cause improvement [9]. The serum Kyn/Trp ratio equals to IDO1 activity and allows an estimation for level of Trp-Kyn metabolism, essentially serving as an indicator of intrinsic immune functional status. Trp-Kyn metabolism increased in many diseases with chronic immune activation, such as chronic inflammatory, autoimmune, and allergic diseases, as well as cancer and atherosclerosis [10]. Enhanced extrahepatic Trp-Kyn metabolism was reported to be associated with poor prognosis in cancer or disease activity in chronic inflammatory disorders [11,12]. Collectively, we hypothesized that the activity of Trp-Kyn metabolism reflects the disease severity and may affect the prognosis in DM.
Interestingly, it has been reported that Trp metabolism is significantly altered in DM [13]. However, further clinical information about its downstream Trp-Kyn metabolism is limited. Therefore, we measured the serum levels of Trp and Kyn and calculated the serum Kyn/Trp ratio to determine the association of Trp-Kyn metabolism with clinical characteristics, treatment, and clinical outcomes in DM patients.

Patients
This was a single-center, retrospective, observational study. From January 2017 to September 2020, we enrolled 57 newly diagnosed and treatment-naïve DM patients admitted to the Department of Dermatology at Shanghai Rui Jin Hospital. We also included 10 age-and sex-matched healthy controls (HCs). Classic DM (CDM) or clinically amyopathic DM (CADM) was diagnosed based on Bohan and Peter's criteria or modified Sontheimer's definitions [14,15]. Treatment-naïve DM was defined as having no history of corticosteroid or immunosuppressant use before admission. All patients completed an initial screening before enrollment to evaluate organ involvement and tumor. ILD was diagnosed based on characteristic abnormalities on high-resolution computed tomography (HRCT) and clinical presentations, with the exclusion of pulmonary infections determined by two experienced pulmonologists. All DM patients were divided into two groups according to the optimal cutoff value of the serum Kyn/Trp ratio (× 10 −3 ) determined by X-tile software [16], the high serum Kyn/Trp ratio group (> 38.50, n = 32) and the low serum Kyn/Trp ratio group (≤ 38.50, n = 25). The Ruijin Hospital Ethics Committee approved this study according to Confidential Agreement No. 2016 (105), and informed consent was obtained from all participants.

Data collection
Patient general characteristics and clinical information were retrieved from hospital electronic medical records by a dedicated person from the Department of Information Technology. All patients were followed up until the endpoint of this study (June 30, 2021), and data were collected without any missing information. Survival time was defined as the time to death or censoring date from the hospital admission date. Electronic medical records were used to collect and collate the corresponding clinical characteristics and laboratory data. Cutaneous Dermatomyositis Disease Area and Severity Index (CDASI) was performed by two dermatologists (H.C. and D.W.), as reported previously [17]. The serum Kyn/Trp ratio was calculated to represent a surrogate marker for Trp-Kyn metabolism level, and X-tile software was used to determine the optimal cutoff value.

Measurement of serum tryptophan and kynurenine concentrations
Sera were separated by centrifugation at 3000 rpm for 5 min at room temperature and stored frozen at − 80 °C away from light until use. The serum concentrations of Kyn and Trp were quantified using an Agilent 1100 high-performance liquid chromatography-mass spectrometer (HPLC-MS) system. Serum samples were pre-treated using protein precipitation. A 10 µL aliquot of the supernatant was injected into the HPLC-MS system equipped with an Agilent ZORBAX SB-C18 column (2.1 mm × 150 mm, 5 µm) to acquire the mass spectra peak area under the selected ion monitor (SIM) mode. The serum concentrations of Trp and Kyn were evaluated using an internal standard calibration curve. The calibration curve was established as follows. Taking the concentrations of Trp or Kyn calibration samples as the X-axis, the ratio of the mass spectra peak area between Trp or Kyn to the internal standard (metronidazole) as the Y-axis, the regression equation of concentration and mass spectra peak area was obtained using the least squares method.

Statistical analysis
Continuous variables with normal distribution are reported as the mean ± standard deviation (SD). Continuous variables with a skewed distribution are reported as medians with interquartile ranges (IQRs). Categorical variables are presented as frequencies (percentages). We used the Student's t-test for continuous variable comparisons or the Mann-Whitney U test for non-parametric continuous variable comparisons. Categorical variables were compared using the chi-squared test or Fisher's exact test. The results were corrected for multiple comparisons using Benjamin-Hochberg false discovery rate (FDR). Spearman's rank correlation analysis was used for non-parametric correlation analysis. Survival analysis was performed using the Kaplan-Meier method and log-rank test. Cox proportionalhazards regression analysis was used for the assessment of prediction of mortality. The Wilcoxon signed rank-sum test was used to compare the data before and after medication. Statistical significance was set at P < 0.05. Statistical analysis and graphing were performed using GraphPad Prism 8 software and R software (R version 4.0.4, http:// www.R-proje ct. org).
Patient key characteristics are listed in Table 1 (complete  characteristics are reported in Supplementary Table 3). No statistical differences in age at diagnosis, sex, and body mass index (BMI) were found between the two groups.
DM patients with high serum Kyn/Trp ratio had a significantly higher frequency of CDM (62.5% vs. 16%, P = 0.008) and a lower frequency of CADM (37.5% vs. 84.0%, P = 0.008) than those with low serum Kyn/Trp ratio, in concert with a higher percentage of muscle weakness symptoms (62.5% vs. 20.0%, P = 0.019) and higher levels of LDH

Survival analysis according to serum Kyn/Trp ratio
In our cohort, the cumulative survival rate was 86.19%. Survival analysis showed that DM patients with high Trp-Kyn metabolism had worse overall survival than those with low Trp-Kyn metabolism (log-rank P = 0.0033) (Fig. 3). Patients with high Trp-Kyn metabolism had a higher frequency of comorbidities, including interstitial lung disease or malignancy, and the main causes of death were respiratory failure due to interstitial lung disease (3/9, 33.3%) or the exacerbation of malignancy (6/9, 66.7%). Cox proportional-hazards regression model showed that ln(Kyn/Trp) was an independent predictor of mortality in the cohort under investigation. After adjusting for BMI, WBC, malignancy, hypertension, and diabetes mellitus, ln(Kyn/Trp) (HR 4.874, 95% CI 1.105-21.499, P = 0.036) was still an independent predictor of mortality ( Table 2).

Discussion
The essential role of immunity is "self-nonself" discrimination [18]. Inflammation can be concluded as the activation of immune cells during the defense and clearance of "nonself." However, they were always limited and restrained by intrinsic immune regulation function in a feedback loop to restore immune homeostasis. Trp-Kyn metabolism level is an important indicator for such intrinsic immune functional status. In this study, we explored the clinical association of the Trp-Kyn metabolism via IDO1 induction in DM and demonstrated that (1) an increased serum Kyn/Trp ratio in DM correlated with disease activity; (2) Cox regression analyses identified ln(Kyn/Trp) as an independent prognostic predictor of mortality; and (3) serum Kyn/Trp ratio decreased upon treatment.    In previous basic research, it has been identified that higher Trp oxidation in the muscle of patients with polymyositis contributes to the mitochondrial proteins' alteration and mitochondrial dysfunction, which is a common pathway shared in muscle disease [19]. Later, a bio-informatic metabolomic study identified the aminoacyl-tRNA biosynthesis, Trp biosynthesis, and metabolism were perturbed pathways in DM and showed that the decreased Trp levels had diagnostic values for DM [13]. In this study, the clinical aspect of Trp metabolism in DM was considered through the disease activity, treatment, and prognosis.
Trp-Kyn metabolism via IDO1 induction has also been reported with clinical relevance in other autoimmune diseases. For example, in SLE patients, enhanced Trp-Kyn metabolism correlated with disease activity and severe fatigue and could predict disease activation [20][21][22][23]. Elevated IDO1 activity and increased regulatory T cells have been discovered and correlated with disease severity and interferon-positive signatures in patients with Sjogren's syndrome [24,25]. Moreover, serum IDO1 activity was confirmed to be increased in psoriasis, where immune cells from patients with psoriasis are defective in inducing IDO1 upregulation in response to inflammatory stimuli [26].
In this study, we observed that patients with high Kyn/ Trp ratio had more muscle involvement (CADM vs. CDM, muscle enzymes), which is in congruence with previous basic research, indicating that an enhanced extrahepatic Trp-Kyn metabolic pathway and implying that the subsequent intrinsic immune regulation function could play a role in muscle tissue damage and repairment. It is reasonable to speculate that Trp-Kyn metabolism may also contribute to the robustness of the tissue reconstruction system by maintaining immune homeostasis [27]. Skeletal muscle is the largest reservoir of amino acids in the body; it has been reported that aminoacyl-tRNA, phenylalanine, and tyrosine biosynthesis pathways were also altered in DM. Muscle metabolism-related changes in immune function may be a new therapeutic target for DM.  What's more, the serum Kyn/Trp ratio correlated with disease activity indicators. However, there was no correlation between the serum Kyn/Trp ratio and CDASI (activity or damage). We hypothesized that the CDASI indicates the severity of skin lesions while the Kyn/Trp ratio represents the whole level of immune status. Given the alterations of Trp-Kyn metabolism before and after treatment, the level of ln(Kyn/Trp) showed potential prognostic value for survival. This would implicate the underlying impairment in the intrinsic immune regulation function in patients with DM.
Although this is an observational study, this research has important implications. Once inflammation was initiated, the intrinsic immune regulation function also occurred (Fig. 5A). It may be driven by extrahepatic Trp-Kyn metabolism via IDO1 induction to limit the excessive inflammatory reactions (Fig. 5B) [28]. Such findings can be explained as "Trp depletion" and "Kyn production" [29]. The former leads to the inhibition of effector immune cells via the general control nonderepressible two kinase (GCN2K) signaling pathway. The latter activates the aromatic hydrocarbon receptor (AhR) signaling pathway, which induces antiinflammatory signaling [30]. A Trp-Kyn metabolism via the IDO1 induction comic is shown in Supplementary Fig. 1. So we proposed that other than "anti-inflammation" therapeutic strategies, intrinsic immune regulation function would be a promising pipeline for the treatment and surveillance of DM.
The current first-line treatment of DM is still mainly empirical glucocorticoids with immunosuppressants [31]; due to its side-effects, new approaches are still in need. Trp-Kyn metabolism could be an essential target for restoration of intrinsic immune regulation function with potentially translational implications and may be a source of future work. First, regular aerobic exercise of a certain intensity can promote Fig. 3 Survival curves according to the optimal cutoff for serum Kyn/Trp ratio in all DM patients. Survival analysis (n = 57) was performed by the Kaplan Meier method and logrank test. Abbreviations: IDO1, indoleamine 2,3-dioxygenase 1; DM, dermatomyositis  [32]. Intensive aerobic exercise and resistance training can reduce disease activity and promote muscular improvement in DM [33,34]. Shifting the Kyn pathway after exercise could be one reason that non-pharmacological treatment exercises effectively treat DM. Second, changes in food composition and lifestyle, including physical activity and weight loss, have been shown to influence Trp availability [35,36]. The regulation of Trp content showed clinical potential in the prevention and treatment of immune-related disorders, which may also have therapeutic potential in DM [37]. Recently, IDO1 inhibitors have been developed and entered the clinical assessment phase and show promise for cancer therapy by strengthening the antitumor immune response [38]. Regulating IDO1 expression and activity may be a potential therapeutic strategy for DM.
Overall, we added additional evidence to the literature pool on recognizing another potential marker for clinical outcome prediction among patients with DM. A better understanding of such a signaling pathway could help physicians to differentiate between the DM subtypes. More importantly, our study showed that intrinsic immune regulation function is also of paramount importance for immune homeostasis and could also be a target for therapy in DM.
Importantly, our investigation has some limitations that require discussion. First, the presented data demonstrate associations rather than causal relationships or mechanistic insights. Second, without directly assessing IDO1 expression in serum, we can only hypothesize that increased Trp-Kyn metabolism in DM may associate with altered intrinsic immune regulation function. Interventional clinical  Immune homeostasis and local immune regulation via the tryptophan-kynurenine metabolic pathway via IDO1 induction. A Inflammation and immune regulation co-occur to maintain dynamic immune homeostasis. B IDO1 induction upon inflammatory stimuli drives the extrahepatic tryptophan-kynurenine metabolic pathway to play a role in local immune regulation. ① Pro-inflammatory stimuli; ② IDO1 induction; ③ enhanced extrahepatic tryptophan-kynurenine pathway; ④ altered phenotype of immune cells; ⑤ immune tolerance and immune suppression; ⑥ immune homeostasis. Abbreviations: Trp, tryptophan; Kyn, kynurenine; IFN-γ, interferon-gamma; IDO1, indoleamine 2,3-dioxygenase 1; M1, M1 macrophages; Th1, CD4 + T helper 1; Treg, regulatory T cell studies are needed to determine whether modification of intestinal tryptophan pathways affects the severity of DM. Third, the sample size of our investigation is rather limited. Further studies with larger sample sizes and prospective cohorts are required to confirm our findings. Fourth, other disease-specific outcome measures such as physician global disease activity scores, manual muscle testing, or patient reported outcomes had not been done to comprehensively evaluate the severity of the disease.
In the present study, we determined an immunometabolic feature of enhanced Trp-Kyn metabolism and its clinical relevance in treatment-naïve adult DM patients. The role of immune regulation via the Trp-Kyn pathway triggered by IDO1 induction in DM needs to be further investigated for its clinical potential in the future. Monitoring and intervention in the functional state of immune regulation in the human body need to be considered as a potential therapeutic target in DM.