Circulating lncRNA ITSN1‐2 is upregulated, and its high expression correlates with increased disease severity, elevated inflammation, and poor survival in sepsis patients

Background This study aimed to assess the correlation of long noncoding (lnc) RNA intersectin (ITSN) 1‐2 expression with disease risk, severity, inflammation, and survival in sepsis patients. Methods Three hundred and nine intensive care unit (ICU)‐treated sepsis patients and 300 healthy controls were consecutively recruited in this study. Blood samples were collected from all sepsis patients within 24 hours after admitted to ICU and from healthy controls at the time of health screening, and the expression of lncRNA ITSN1‐2 in plasma was detected by quantitative polymerase chain reaction. Disease severity was assessed by physicians using acute physiology and chronic health evaluation (APACHE) II score on day 1 after ICU admission. Additionally, the plasma inflammatory cytokines (including tumor necrosis factor α (TNF‐α), interleukin 1β (IL‐1β), IL‐6, IL‐8, IL‐10, and IL‐17) were measured by enzyme‐linked immunosorbent assay (ELISA) kits. Results lncRNA ITSN1‐2 was highly expressed in sepsis patients compared to healthy controls and could differentiate sepsis patients from healthy controls with area under the curve (AUC) 0.777 (95% CI: 0.740‐0.813). lncRNA ITSN1‐2 expression was positively correlated with APACHE II score, C‐reactive protein (CRP), TNF‐α, IL‐6, and IL‐8 levels, but negatively correlated with IL‐10 level. In addition, lncRNA ITSN1‐2 was highly expressed in non‐survivors compared to survivors and could distinguish survivors from non‐survivors in sepsis patients with AUC 0.654 (95% CI: 0.581‐0.726). Conclusion Circulating lncRNA ITSN1‐2 is upregulated, and its high expression associates with increased disease severity and inflammation as well as poor prognosis in sepsis patients.

diagnostic markers have been discovered and analytical techniques have improved, the delay of diagnosis for sepsis is still very common due to acute, critical illness, contributing to poor treatment outcomes, increased hospitalization duration and cost, and even death. Therefore, exploring reliable sepsis biomarker for early diagnosis and prognostic prediction is an urgent need for helping establish clinical treatment strategies.
Long noncoding (lnc) RNA is class of RNA longer than 200 nucleotides and does not encode proteins. 7 Accumulating studies have revealed that multiple lncRNAs are implicated in acute inflammatory response against microbial infection through transcriptionally regulating inflammatory gene expression. 7 lncRNA intersectin (ITSN) 1-2 is one of lncRNA family members located in chromosome 21 with its NONCODE gene ID NONHSAG032726.2, and its length was 451 bp with NONCODE transcript ID NONHSAT081856.2, starting from 33976355 to end site 33976982. 8 It is reported that lncRNA ITSN1-2 high expression is correlated with increased disease risk and activity as well as inflammation degree of rheumatoid arthritis (RA). 8 Considering that RA and sepsis are both characterized as systematic inflammatory response and the latter presents with even more severe inflammation, and our preliminary research in a small population observed that lncRNA ITSN1-2 was extremely upregulated in sepsis patients, we speculated that lncRNA ITSN1-2 may be involved in the development and progression of sepsis. Thus, the objective of this study was to assess whether lncRNA ITSN1-2 could distinguish sepsis from healthy controls, and its correlation with disease severity, inflammation, and survival in sepsis patients.

| Samples and data collection
Blood samples were collected from all sepsis patients within 24 hours after admitted to ICU and from healthy controls at the time of health screening. Also, the characteristics of sepsis patients were recorded including age, gender, and BMI, and the levels of serum creatinine (Scr), albumin, white blood cell (WBC), and C-reactive protein (CRP) were assayed routinely. In addition, disease severity of sepsis was assessed by physicians using acute physiology and chronic health evaluation (APACHE) II score on day 1 after ICU admission. All patients were treated in accordance with current guidelines for treatment of sepsis and followed up for 30 days (30-day survival was recorded as well).

| Statistical analysis
Statistical analysis was performed using SPSS 22.0 software (SPSS Inc, USA) and GraphPad Prism 6.01 (GraphPad Software, USA).
Normal distributed continuous variable was presented as mean value ± standard deviation, skewed distributed continuous variable was expressed as median (25th-75th quantiles), and categorized variable was presented as count (percentage). Differences of lncRNA ITSN1-2 relative expression were determined by Wilcoxon rank-sum test. Correlation analyses were performed using Spearman's test.
Receiver operating characteristic (ROC) curves and the area under the ROC curve were used to assess the diagnostic value of lncRNA ITSN1-2 levels for sepsis and the ability to discriminate between survivors and non-survivors. All tests were all two-sided, and Pvalue <0.05 was considered significant.

| Baseline characteristics
There were 309 patients in sepsis group, including 204 (66.0%) males as well as 105 (34.0%) females, and their mean age was 57.3 ± 9.7 years ( Table 1). The median APACHE II score was 16.0

| lncRNA ITSN1-2 expression in sepsis patients and healthy controls
The lncRNA ITSN1-2 expression was higher in sepsis patients compared to healthy controls (P < 0.001) ( Figure 1A). The ROC curve disclosed that lncRNA ITSN1-2 expression could distinguish sepsis patients from healthy controls with area under the curve (AUC) 0.777 (95% confidence interval (CI): 0.740-0.813) ( Figure 1B). The sensitivity and specificity were 59.5% and 86.3% at the best cutoff point that was defined as the point where the sum of sensitivity and specificity was maximum, and lncRNA ITSN1-2 expression was 1.820. These data suggested that lncRNA ITSN1-2 might be a biomarker for sepsis risk.

| Comparison of lncRNA ITSN1-2 expression between non-survivors and survivors
The lncRNA ITSN1-2 expression was decreased in survivors compared to non-survivors (P < 0.001) ( Figure 3A). The ROC curve revealed that the AUC of lncRNA ITSN1 and APACHE II score was 0.654 (95% CI: 0.581-0.726) and 0.549 (95% CI: 0.481-0.616), respectively, and lncRNA ITSN1-2 expression present a good value on distinguishing survivors and non-survivors in sepsis patients, whose sensitivity and specificity at the best cutoff point were 92.1% and 40.4%, respectively, and lncRNA ITSN1-2 expression was 4.059, which implied that lncRNA ITSN1-2 might be a prognostic marker in sepsis patients ( Figure 3B)

| D ISCUSS I ON
In the present study, we observed that: (1) lncRNA ITSN1-2 was highly expressed in sepsis patients compared to healthy controls, and it presented a good diagnostic value for sepsis; (2) lncRNA ITSN1-2 expression was positively correlated with APACHE II score and inflammatory markers; and (3) lncRNA ITSN1-2 was highly expressed in non-survivors compared to survivors, and F I G U R E 1 Comparison of lncRNA ITSN1-2 expression between sepsis patients and healthy controls. A, LncRNA ITSN1-2 expression was higher in sepsis patients compared to healthy controls (median value 2.271 vs 0.903). B, ROC curve displayed that lncRNA ITSN1-2 present good diagnostic value of sepsis, and its expression was 1.820 at best cutoff point that was defined as the point where the sum of sensitivity and specificity was maximum. Wilcoxon rank-sum test was performed to compare lncRNA ITSN1-2 expression between sepsis patients and healthy controls. LncRNA ITSN1-2, long noncoding RNA intersectin 1-2; ROC, receiver operating characteristic called lncRNA IL-eRNA promotes IL-1β production by activating TLR signaling pathway in lipopolysaccharide (LPS)-induced human monocytes. 12 As to sepsis, several lncRNAs have been reported to act as positive or negative regulators. 13,14 For example, lncRNA HOX transcript antisense RNA (HOTAIR) facilitates TNF-α production in cardiomyocytes of LPS-induced sepsis mice by activating nuclear factor (NF)-κB pathway. 13 In addition, lncRNA interleukin-7 receptor α-subunit (IL-7R) negatively regulates E-selectin, vascular cell adhesion molecule-1 (VCAM-1), IL-6, and IL-8 production via interacting with the human IL-7R gene, thereby attenuated the LPS-induced proinflammatory response. 14 Briefly, several lncRNAs could regulate inflammatory response in inflammatory diseases, particularly in sepsis.
In clinical trials, dysregulated lncRNAs are found in sepsis patients and these lncRNAs display good diagnostic values for sepsis. [15][16][17] For instance, plasma lncRNA nuclear-enriched abundant transcript (NEAT1) is highly expressed in sepsis patients compared to healthy controls, which could distinguish sepsis patients from healthy controls with AUC 0.730. 15 Also, lncRNA ENST00000452391.1 is highly expressed in sepsis compared to healthy controls and ROC curve displays its good diagnostic value with AUC 0.866. 17 However, for lncRNA ITSN1-2, there is limited information about its role in inflammatory disease, just one study discloses that high expression of ln-cRNA ITSN1-2 is observed in RA patients, and it presents with good diagnostic value (AUC 0.898). Given that sepsis arises from dysregulated inflammatory response to infection, even more severe than RA, we hypothesized that lncRNA ITSN1-2 might have diagnostic value in sepsis. In the present study, we uncovered that the expression of cir- be promising indicators for worse prognosis in sepsis patients. 21,22 Currently, the prognostic value of lncRNAs in sepsis has rarely been investigated and only a few lncRNAs are reported in this respect. 15,17 For example, lncRNA NEAT1 expression is higher in non-survivors compared to survivors and present good predictive value for sepsis-related survival with AUC 0.641. 15 Moreover, lncRNA ENST00000452391.1 high expression is correlated with reduced survival in sepsis patients. 17 Although above studies display the prognostic value of several ln-cRNAs in sepsis, there is no information about the role of lncRNA ITSN1-2 in prognosis of sepsis patients. Thus, in this study, we discovered that lncRNA ITSN1-2 was highly expressed in sepsis non-survivors compared to survivors and ROC curve disclosed lncRNA ITSN1-2 F I G U R E 3 Comparison of lncRNA ITSN1-2 expression between sepsis non-survivors and survivors. A, LncRNA ITSN1-2 expression was higher in sepsis non-survivors compared to survivors (median value 3.146 vs 2.018). B, ROC curve displayed that lncRNA ITSN1-2 present good diagnostic value of survival, and its expression was 4.059 at best cutoff point that was defined as the point where the sum of sensitivity and specificity was maximum. Wilcoxon rank-sum test was performed to compare lncRNA ITSN1-2 expression between sepsis non-survivors and survivors. LncRNA ITSN1-2, long noncoding RNA intersectin 1-2; ROC, receiver operating characteristic could distinguish survivors and non-survivors with AUC 0.654 in sepsis patients, which could be explained by that: (1) lncRNA ITSN1-2 contributed to elevated disease severity and systematic inflammation, which was validated in our aforementioned data, thereby caused poor prognosis in sepsis patients; 1,3 (2) lncRNA ITSN1-2 might attenuate cell response to sepsis treatment, such as anti-infection and anti-inflammation drugs, thereby worsened prognosis in sepsis patients. However, this hypothesis needed further investigation to confirm.
Limitations existed in the present study. Firstly, all enrolled patients were from single center and ICU; thus, multicentric and population-based sample was needed to validate the diagnostic and prognostic values of lncRNA ITSN1-2 in sepsis. Secondly, the detailed mechanisms of how lncRNA ITSN1-2 associated with inflammation and organ dysfunction were not explored. Thirdly, black race has been reported to be an important risk factor for the development of sepsis and a predictor of poor outcomes in sepsis compared to white race, while our study only investigated the role of lncRNA ITSN1-2 in sepsis in the yellow race and whether the results were applicable to other races was unknown. 23 Finally, we have performed the correlation of lncRNA ITSN1-2 with inflammatory cytokines levels, and the poor correlations were observed (eg, with IL-10 and IL-17), which might be resulted from the extreme value that might be correlated with the regulation of lncRNA to influence inflammatory factors levels.
Summarily, circulating lncRNA ITSN1-2 is upregulated and its high expression associates with increased disease severity and inflammation as well as poor prognosis in sepsis patients.

CO N FLI C T S O F I NTE R E S T
The authors declare that they have no conflicts of interest.