Association between serum miR-221-3p and intravenous immunoglobulin resistance in children with Kawasaki disease

Intravenous immunoglobulin (IVIG) resistance was a major cause of coronary artery lesions in children with Kawasaki disease (KD). However, the cause of IVIG resistance in KD remains unknown. miR-221-3p has been confirmed involved in cardiovascular diseases and rheumatoid arthritis. The purpose of this study was to investigate the association between miR-221-3p and IVIG resistance in children with KD. Fifty-five KD patients and 29 healthy controls (HCs) were enrolled in this study. KD patients were divided into group of sensitive to IVIG (IVIG-response, n = 42) and group of resistant to IVIG (IVIG-resistance, n = 13), group of 10 KD patients with coronary artery lesions (CALs, KD-CALs) and group of 10 sex- and age-matched KD patients without CALs (KD-NCALs). Reverse transcription–quantitative polymerase chain reaction (RT-qPCR) was used to detect the levels of miR-221-3p. Compared with the HCs group, miR-221-3p were significantly increased in the KD group (p < 0.05), and the IVIG-resistance group had higher levels of miR-221-3p than those in the IVIG-response group (p < 0.05). CRP (C-reactive protein), PCT (procalcitonin), NLR (neutrophil–lymphocyte ratio) were positively correlated with miR-221-3p in KD patients. In addition, the group of IVIG resistance had a higher level of Kobayashi Score (p < 0.001). The receiver operating characteristic curve showed that miR-221-3p had a better value for diagnosis IVIG resistance in children with KD than Kobayashi Score with the AUC of 0.811 (95% CI, 0.672–0.951), 0.793 (95% CI, 0.618–0.968), respectively. Additionally, miR-221-3p was elevated (p < 0.05) and showed an AUC value of 0.83 (95% CI, 0.648–1.000, p < 0.05) for the prediction of the complication of coronary artery abnormalities in the group of KD with CALs. miR-221-3p might be involved in the pathogenesis of KD and IVIG resistance and miR-221-3p can be used as a new potential biomarker to predict IVIG resistance in children with KD.


Introduction
Kawasaki disease (KD) is an acute self-limiting febrile vasculitis that mainly affects children under five years old [1]. Since its first discovery in 1960s, KD has been identified worldwide and now has become the leading cause of acquired heart disease in children in developed countries [2]. The most severe complication of KD is coronary artery lesions, which could ultimately lead to coronary heart disease (CHD) in young adults, resulting in myocardial ischemia, infarction, and even sudden death [3][4][5]. Although a single dose of 2 g/kg intravenous immunoglobulin (IVIG) combined with aspirin regimen considered to be the standard treatment for KD, there is still approximately 20% of the cases are insensitive to IVIG, and around 10% of them develop into serious coronary artery abnormalities [6]. As a result, further research regarding the pathogenesis of IVIG resistance in children with KD is necessary.
miR-221, a member of antiangiogenic gene-regulating miRNAs family, is encoded by a co-transcribed cluster on the X chromosome and abundant distribution in endothelial and smooth muscle cells of human vessels [11]. Lots of studies have shown that microRNAs (miR-NAs) are involved in pathogenesis of some diseases, such as cardiovascular diseases, inflammatory diseases, and immunologic diseases [7][8][9][10]. Pandis et al. found that the expression levels of miR-221 were significantly higher in synovial fibroblast (SF) in patients with rheumatoid arthritis (RA) [12]. Yang et al. demonstrated that downregulation of miR-221 can significantly inhibit the levels of TNFα, IL-6, IL-1β and CXCL16 in fibroblast-like synoviocytes (FLS) cells in patients with RA [13]. Also, studies have found that miR-221 expression increases significantly in acute KD patients [14]. However, the relationship between miR-221 and IVIG resistance remains unclear. Thus, the purpose of this study is to investigate the role of miR-221-3p on IVIG resistance in children with KD.

Study subjects
Fifty-five KD patients (KD group) consisting of 37 males and 18 females (average age 2.648±1.832 years) and twenty-nine healthy controls (HCs) children consisting of 19 males and 10 females (average age 3.191±1.075 years) were enrolled in this study from April to August of 2019 consecutively. All subjects met the criteria of American Heart Association (AHA) Scientific Statement. An IVIG resistance was defined by recrudescent or persistent fever at least 36 hours after the initial infusion of IVIG [2] and Kobayashi system [15] was used to predict IVIG resistance. The CALs were examined by the echocardiography with an internal diameter of an artery ≥ 2.5 mm (0 to 3 years); ≥ 3.0 mm (3 to 9 years) and ≥ 3.5 mm (9 to 14 years), or if the internal diameter of a segment was at least 1.5 times that of an adjacent coronary artery; an internal diameter of an artery > 4 mm was defined as CAA [16]. This study was approved by the institutional ethics board of Children's Hospital of Chongqing Medical University, and written informed consents were obtained from all subjects or guardians.

Sample collection
The blood samples were collected from the KD patients on the day of admission before given IVIG and aspirin and immediately centrifuged at 4°C for 10 min at 3,000 revolutions/min to isolate the serum. The same procedures were conducted to the HCs, and all serum samples were stored at -80 °C.

RNA extraction and real-time qPCR assay
RNA was extracted using a miRNeasy serum/plasma kit (Qiagen, Hilden, Germany, Cat. No: 217184) according to the manufacturer's instructions. The primer of miR-221-3p was quantitated by reverse transcription and Real-Time qPCR analysis using Qiagen-based RT-qPCR assay (Qiagen, Hilden, Germany, Cat. No: 339320). The cycling conditions were 95°C for 2 min followed by 40 cycles at 95 °C for 10 s and 56 °C for 1 min. All reactions were performed in triplicate. The expression was quantified using the 2 −ΔΔ Cq values and hsa-miR-103a-3p served as a control for calibrating the expression levels of miR-221-3p.

Statistical analysis
Demographic and clinical characteristics of subjects were expressed as mean values and as total numbers according to the applied statistical methods. The distribution of these variables was evaluated with Shapiro-Wilk normality test. Normally distributed variables were compared using t test. The comparison of rate examined with Chi-square test. The differences in levels of miR-221-3p in serum between KD patients and HCs were analyzed by 2 −ΔΔ Cq method. The correlation of miR-221-3p with other clinical characteristics was assessed with Pearson correlation. The diagnostic accuracy of miR-221-3p was measured by the receiver-operating characteristic (ROC) curves and the area under the ROC curve (AUC). Reported p values are two-sided and a p value less 0.05 was considered the difference significantly. All analyses were performed by Statistical Package for Social Sciences for Windows 24.0 (SPSS Inc., Chicago, IL, USA).

The expression of miR-221-3p in KD patients and healthy controls
Demographic features between KD group and HCs group were examined and there was no significant difference in age and gender between these two groups (p > 0.05). The qRT-PCR method was used to detect the expression of miR-221-3p in 55 KD patients and 19 healthy controls. The results showed that the level of serum miR-221-3p was significantly higher in KD patients than that of in healthy controls (p = 0.002, Fig. 1).
Interestingly, serum miR-221-3p was significantly association with total dosage of IVIG and administration times of IVIG in KD group (r = 0.501, p < 0.001; r = 0.558, p < 0.001, respectively), while no relationship with the onset age, gender, and weight (all p > 0.05, Table 1).

Clinical characteristics in IVIG resistance group and IVIG response group
A total of 42 IVIG responders and 13 IVIG non-responders were enrolled in this study. IVIG non-responders were (n = 13) more likely to have a longer duration of fever, a higher total dosage of IVIG (p < 0.001), higher levels of mean platelet volume (MPV), platelet distribution width (PDW), neutrophils, total bilirubin (TLB), alanine transaminase (ALT), aspartate aminotransferase (AST), CRP, PCT, NLR, prothrombin and lower levels of lymphocyte, platelet count (PLT), and albumin (ALB) than that of IVIG responders group, while age, gender, weight and time point of IVIG were not different ( Table 2).

Expression of miR-221-3p and Score of Kobayashi System in IVIG resistance group and IVIG response group
Statistical tests revealed that the level of serum miR-221-3p was significantly higher in group of resistance to IVIG than that of group of IVIG responses (1.782 ± 0.7 vs 1 ± 0.552, p = 0.002) (Fig. 2). There were also higher Kobayashi scores in KD patients with IVIG resistance (p < 0.001) ( Table 3).

The ROC Curves Analysis of serum miR-221-3p in predicting IVIG resistance in children with KD
Receiver operating characteristic (ROC) curve analysis was used to explore the usefulness of serum miR-221-3p and Kobayashi Score to predict unresponsiveness to IVIG in children with KD. The results showed that serum miR-221-3p reflected obvious separation between group of IVIG resistance and group of IVIG response; the AUC was  (Fig. 3).
In addition, miR-221-3p was elevated in KD patients with CALs (p = 0.008) and showed an AUC value of 0.83 (95% CI, 0.648-1.000, p = 0.013) for the prediction of the complication of coronary artery abnormalities in 10 patients with CALs compared to 10 sex-and age-matched patients without CALs (Supplementary Table 1).

Discussion
The characterization of KD is an inflammatory in all the medium-sized arteries, especially coronary artery lesions (CALs). Accumulating studies reported that 10 to 20% of KD patients' unresponsiveness to initial treatment with IVIG, whom had increased risk to develop of CALs [17,18]. Early identification of IVIG resistance may be important to reduce CALs in children with KD. However, the pathogenesis of resistance of IVIG in children with KD remains unknown.
Total number of 55 KD patients were enrolled in this study, out of which 13 of them (23.6%) were diagnosed with IVIG resistance according to the 2017 American Heart Association guidelines. Our observation showed that patients with IVIG resistance have a longer duration of fever, a higher total dosage of IVIG and significantly higher levels of CRP, N%, PCT, NLR, ALT and AST than that of group of IVIG responders, and coincident with other reports [19][20][21]. The IVIG resistance group also showed lower levels of lymphocyte, PLT, albumin and sodium concentrations than that of IVIG responding group. Up to now, these laboratory data have been used as predictors of the IVIG non-responders [15,[22][23][24][25]. These results may collectively reflect the presence of severe vascular inflammation in children with KD and result in IVIG resistance.
Evidences have shown that the inflammatory cytokines and chemokines, such as TNF-α, IL-1, IL-6, and ICAM-1, are excessively increased during the acute febrile phase, indicating immune activation may contribute to the pathogenesis of KD [26,27]. Recently, numerous studies have implicated that Toll-like receptors (TLRs) signaling pathways in inflammatory vascular pathological conditions may   play an important role in development of KD [28][29][30][31]. Our previous study clarified that TLR4/NF-κB signaling pathway was involved in the process of coronary artery lesions. These results may suggest that regulating expression of TLRs will protect children with KD from coronary artery lesions [32]. miR-221-3p, one of the miRNAs that is conserved among vertebrates, was detected in immunologic and inflammatory disease as well as cardiovascular disease. Evidence has shown that miR-221-3p could be a regulator of TLRs [13]. Regulation of miR-221-3p can affect the level of proinflammatory cytokines, such as TNF-α, IL-1β and IL-6, in FLS cells when stimulated by LPS [33]. Upregulation of miR-221-3p in TLR4 can enhance the secretion of IL-6 and IL-8, and downregulation the level of anti-inflammatory factors of CXCL13 and IL-10 [34]. Consistent results were also observed that there was a relationship between miR-221-3p and expression of IL-6, TNF-α and activation of NF-kB and MAPKs in inflammatory lung disorder [35].
In our study, we found that serum miR-221-3p was elevated in children with acute phase of KD, which is consistent with previous study [14]. For the first time, we reported that serum miR-221-3p was extremely higher in group of IVIG non-responders than that of group of IVIG responders. The serum miR-221-3p levels have significantly correlated with many key inflammatory indexes, such as N%, CRP, PCT, NLR and even total dosage of IVIG and given times of IVIG in KD patients. These results elicited that miR-221-3p may exhibit an important role in initiating inflammatory reaction involved in pathogenesis of KD and IVIG resistance.
Interleukin-6 (IL-6) is one of the pro-inflammatory with pleiotropic activity. It is released by not only T cells and macrophages, but also vascular endothelial cells, fibroblasts, and many other cells in response to various stimulation including TLR ligands and proinflammatory cytokines, such as TNF-α or IL-1 [36]. It also has correlation with the acute phase of KD and IVIG non-responsiveness [20]. SOCS3 is a member of the IL-6 signaling family, and it directly interacts with Janus kinase (JAK) signal transducer and inhibits of transcription 3 (STAT3) pathway and regulates immune reaction, cell growth and differentiation. According to a recent study via bioinformatics analysis and dual-luciferase reporter gene assays, SOCS3 was found to be the target gene of miR-221-3p in hepatocellular carcinoma (HCC), which demonstrated that regulating the miR-221-3p/SOCS3 axis may affect the JAK/STAT signaling pathway and ultimately alter HCC tumor growth [37,38]. Furthermore, Chatterjee PK et al. [39] had reported that IL-6/STAT3 pathway may be an underlying mechanism of inflammation in vascular endothelial. Whether miR-221-3p/SOCS3 axis is involved in the development of KD and IVIG resistance needs further study.
Kobayashi score (KS) has been used to predict CALs and IVIG resistance. The parameters composed of the KS scores include age in months, times of illness at initial treatment, N%, PLT, AST, sodium, and CRP to generate a scoring model, and had a specificity of 68% and a sensitivity of 86%; however, the risk-scoring system is limited by regions and is not suitable for global promotion. Lin et al. [40] found the sensitivity of KS system to the KD patients in Taiwan was only 61%. When applied to the UK patients, the sensitivity and specificity of KS are even lower to 58% and 35% [41]. Here, the results have shown that serum miR-221-3p had a better value than the Kobayashi system to predict IVIG resistance with the AUC of 0.811, and miR-221-3p may also be a good biomarker for coronary artery abnormalities in KD patients with the AUC of 0.83 by ROC analysis.
In patients with KD, numerous studies have suggested miRNAs as strong circulating biomarkers with high diagnostic as well as prognostic function. Yun et al. [42] found that levels of microRNA-200c and miR-375-5p were elevated in KD patients. Chu et al. [43] reported bone marrow-derived miR-223 worked as an endocrine genetic signal in vascular endothelial cells and participated in vascular injury from Kawasaki Disease via clinical and experimental research. Serum miR-92a-3p was reported to relate to KD and may be as a new diagnosis and prognostic biomarker of coronary artery abnormalities in KD and become a new drug target in the future [14]. As a result, with the development of the technologies for quantifying miRNAs in clinics, we believe that the evaluation of miR-221-3p as quickly as the other parameters in clinical practice seems feasible.

Conclusions
In conclusion, our study showed that KD patients with IVIG resistance had a higher level of miR-221-3p and a better value than Kobayashi score to predict IVIG resistance, miR-221-3p may be used as a proinflammatory factor and a potential biomarker of IVIG resistance in KD and provide a novel mechanism and a new therapeutic target for KD.

Study limitations
There are limitations in this study. Firstly, the sample size was relatively small and it was finished in a single center. Secondly, we did not collect serum samples from KD patients after being treated with IVIG in subacute and convalescent.
Authors' contributions Fengchuan Jing has a substantial contribution to the conception and design of this study, acquisition of data, analysis of data and draft of the article. Haobo Weng contributes to acquisition of data, analysis of data and interpretation of data. Qiongfei Pei contributes to interpretation of data. Jing Zhang contributes to interpretation of data. Ruixi Liu * has a substantial to design of the study and revise the article. Qijian Yi * has a substantial contribution to revise the article critically for important intellectual content and final approval of the version to be published. All authors approved the final manuscript as submitted.
Funding This work was supported by the Outstanding Young Talent Training Program Class B of Children's Hospital Affiliated to Chongqing Medical University; Award Number: None; Recipient: DR. Ruixi Liu.
Data availability All datasets used or analyzed during the current study are available from the corresponding author on reasonable request.