The expression of circulating hsa-miR-126-3p in dengue-infected Thai pediatric patients

ABSTRACT

Circulating hsa-miRNA-126 (CmiR-126) has been reported to involve in the pathogenesis of many infectious diseases including dengue virus infection. However, no prior study has been conducted to describe more details in dengue-infected pediatric patients. This study aimed to describe CmiR-126-3p in dengue-infected pediatric patients during the febrile and convalescent phases. Additionally, the correlations between CmiR-126-3p and other relevant clinical laboratory factors were investigated. Sixty paired-serum specimens collected during febrile and convalescent phases were retrieved from patients with dengue fever (DF) (n = 30) and dengue hemorrhagic fever (DHF) (n = 30). Thirty paired-serum specimens collected from non-dengue acute febrile illness patients (AFI) were included as the control group. CmiR-126-3p was determined using reverse transcription quantitative real-time polymerase-chain reaction (RT-qPCR). Relative miRNA expression was calculated as 2^−ΔCt using CmiR-16-5p for data normalization. CmiR-126-3p expression during febrile and convalescent phases in dengue-infected patients was significantly lower than AFI (p < 0.05). However, miRNA levels were not different (p > 0.05) compared between DF and DHF and between primary and secondary infection. CmiR-126-3p levels in DF in the convalescent were significantly higher than in the febrile phase (p = 0.025). No association between CmiR-126-3p and hematocrit, WBC level, platelet count, WBC differential count or dengue viral load was observed (p > 0.05). The data suggest that hsa-miR-126-3p involved in pathogenesis of dengue infection and may be a promising early and late biomarker for DENV infection. However, hsa-miR-126-3p alone cannot be used as a predictor for dengue severity.

Introduction

Dengue virus (DENV) infection remains a major health problem in many parts of the world. Approximately 2.5–3 billion people live in its endemic areas and nearly 400 million suffer from infection [1,2]. It is a positive-sense single-stranded RNA virus in the family Flaviviridae (genus Flavivirus), which is composed of four distinct serotypes (DENV1-DENV4) and is transmitted via Aedes mosquitoes [2,3]. The genome consists of one open reading frame (ORF) with approximately 11 kb containing 3 structural (pre-membrane (prM), envelope (E) and capsid (C)) and 7 non-structural (NS1, NS2A, NS2B, NS3, NS4A, NS4B, and NS5) genes [2]. Clinical symptoms are categorized as asymptomatic infection, undifferentiated fever (UF), dengue fever (DF) without hemorrhage, DF with unusual hemorrhage, dengue hemorrhagic fever (DHF), DHF with shock or dengue shock syndrome (DSS), and expanded dengue syndrome (EDS) or unusual manifestations with organ involvement [4]. Severe dengue infections commonly result from secondary infection with different serotypes from prior infection [5].

All age groups are susceptible to dengue infection but clinical pictures are quite different. In many countries, including Thailand, children aged 5–15 years old have the highest incidence of dengue infection [2, 6–8]. Moreover, they also have a higher risk for severe dengue infection than adult patients in the aspect of higher incidence of plasma leakage, shock, and lower platelet count [9–11]. The pathogenesis of plasma leakage is the interaction among DENV, host immune response, and endothelial cells, which affects the homeostasis, integrity, and functions leading to severe symptoms [12].

MicroRNAs (miRNAs, miR), the small non-coding RNAs in the length of 17–25 bp, regulate various cellular functions and involve in cellular physiological processes. They are released into blood and body fluids known as circulating miRNAs (CmiRNA or CmiR). Expression of miRNAs has a role in various diseases, such as cancers, cardiovascular disorders, and infectious diseases including viral infection [13]. In addition, miRNA expressions have been reported to involve various processes during dengue infection such as promoting or suppressing viral replication, helping the virus to evade host immune response, regulating inflammatory processes, and affecting vascular endothelial functions [14,15].

Homo sapiens (hsa)-miR-126-3p (known as miR-126) is located at intron 7 of epidermal growth factor-like domain 7 genes (EGFL7) on human chromosome 9 [16]. It is highly expressed in endothelial cells and involved in inflammation, angiogenesis, and vascular integrity [17–19]. Expressed miR-126 from endothelial cells decreases vascular cell adhesion molecule 1 (VCAM-1) expression and leukocyte adherence to those cells, which attenuates inflammation of endothelial cells [20]. CmiR-126 has been used as the biomarker for various diseases, such as sepsis, smoking-related diseases, type 2 diabetes mellitus, and coronary artery disease [21–24]. Because the pathogenesis of dengue infection involves the functions and integrity of endothelial cells, we hypothesize that the alteration of CmiR-126-3p may occur in patients during the period of infection.

Although previous studies revealed that CmiRNAs could be used as biomarkers for dengue infection and disease severity, CmiR-126-3p had not been well investigated [25, 26]. Recently, downregulation of plasma hsa-miR-126-5p was observed in dengue-infected adult patients with warning signs (WS) and severe dengue (SD) compared to the patients with uncomplicated dengue (UD) [27].

There has been limited information about serum hsa-miR-126-3p during the course of dengue infection and no prior study investigated CmiR-126-3p in pediatric patients with dengue infection. Moreover, the dynamics of hsa-miR-126-3p expression and the correlation with other relevant parameters have not been elucidated. The purpose of this study was to explore serum hsa-miR-126-3p expression in dengue-infected patients with different clinical categories. We also presented the changes of CmiR-126-3p expression in the febrile and convalescent period and the correlation of CmiR-126-3p expression with blood and viral parameters.

Materials and methods

Patient and specimen recruitment

Serum specimens were randomly selected from the archived serum samples from the previous studies ‘Dengue infection in children in Ratchaburi, Thailand: A cohort study. I. Epidemiology of symptomatic acute dengue infection in children, 2006–2009 and Dengue infection in children in Ratchaburi, Thailand: A cohort study. II. Clinical manifestation’ [7, 8]. Briefly, Sixty-paired serum specimens collected during the febrile (day 0–7 of illness) and convalescent phase (7–14 days later from first specimen collection) from confirmed dengue-infected pediatric patients (30 DF and 30 DHF) were used for this study. Thirty-paired serum samples collected from pediatric patients diagnosed as acute febrile illness without dengue-infection (AFI) were used as the control group. All enrolled participants were aged between 4 and 14 years old. Demographic data such as patients’ characteristics, clinical diagnosis, and laboratory data were analyzed. The patients whose serum specimen was collected only 1 time, whose clinical and laboratory data were absent were excluded from this study. This study was approved by The Ethics Committee of Faculty of Tropical Medicine, Mahidol University, Thailand (MUTM 2018–022-01). Procedures conducted in this study were in accordance with the standard criteria of the responsible committee on human experimentation (The Human Ethical Research Committee of Faculty of Tropical Medicine, Mahidol University, Thailand) and with the Declaration of Helsinki 1975 (revised in 2000).

Selection of candidate miRNA

In this study, hsa-miR-126-3p was chosen based on the previous reviews mentioning roles of miR-126 and vascular diseases including previous findings related to dengue infection [17, 19, 28, 29]. Endogenous serum hsa-miR-16-5p was used as internal control and reference miRNA for data normalization [22, 30–32].

MiRNA extraction and cDNA synthesis

Total miRNAs were extracted from 200 µL of each serum sample using miRNeasy mini kit (Qiagen, Hilden, Germany) following the instruction manual. Then, the concentration and quality of extracted miRNA were verified using NanoDrop^TM 2000 Spectrophotometer (ThermoFisher Scientific, Massachusetts, USA). Each extracted miRNA was aliquoted in a small volume and kept at −80°C until use.

First-strand (cDNA) synthesis was performed using TaqMan™ Advanced miRNA cDNA Synthesis Kit (Applied Biosystems: ThermoFisher Scientific, Massachusetts, USA) by following the instruction manual and all processes were run with PCR thermocycler (GeneAmp™ PCR System 9700, Applied Biosystems: ThermoFisher Scientific, Massachusetts, USA). The reaction is composed of poly-A tail addition, adaptor ligation reaction, reverse transcription, and pre-amplification. Briefly, poly-A tail synthesis was performed by adding 2 μL of mastermix composing of 10X poly (A) buffer, ATP, poly (A) enzyme into 3 µL of extracted miRNA. Polyadenylation reaction was performed at 37°C for 45 minutes, then the reaction was terminated at 65°C for 10 minutes. Adapter ligation was continued by adding 5X DNA ligase buffer, 50% PEG 8000, 25X ligation adapter, and RNA ligase. The ligation reaction was set at 16°C for 60 minutes. The reverse transcription step was done after finishing the ligation step by using 5X reverse transcription (RT) buffer, 25 mM dNTP mix, 20X universal RT primer, and 10X reverse transcription enzyme mix. The reaction was incubated at 42°C for 15 minutes and enzymatic activity was terminated at 85°C for 5 minutes. In order to increase the sensitivity of miRNA detection, miR-Amp reaction was performed by adding 5 μL of cDNA with 45 μL of PCR reagent containing 2X miR-Amp mastermix and 20X miR-Amp primer mix. Fourteen cycles of PCR reaction were done in the condition of 95°C for 5 minutes (enzyme activation), 95°C for 3 seconds (denaturation), 60°C for 30 seconds (annealing/extension), and 99°C for 10 minutes for terminating the reaction. The PCR product was kept at −20°C up to 2 months prior to performing quantitative PCR.

MiRNA expression using real-time RT-PCR

Extracted miRNA solutions from dengue-infected patients and control group were explored using TaqMan quantitative real-time polymerase-chain reaction (TaqMan qPCR) with TaqMan®Fast Advanced Mastermix (Applied Biosystems: ThermoFisher Scientific, Massachusetts, USA) by following the instruction manual. Primers and TaqMan probes of hsa-miR-126-3p were commercially available primer-probe assays (Assay ID: 477887_mir) (ThermoFisher Scientific, Massachusetts, USA). Additionally, hsa-miR-16-5p (Assay ID: 477860_mir) was selected as internal control and reference miRNA for data normalization. The Ct > 35 was stated as having no miRNA expression [22].

Briefly, 1 μL of cDNA was pipetted into each 0.1 mL PCR tube containing 9 μL of mastermix solution (10 μL of 2X TaqMan®Fast Advanced mastermix, 1 μL of 20X TaqMan® Advanced miRNA assay, and 8 μL of RNase-free water). The qPCR reaction was performed with StepOnePlus Real-Time PCR System machine (Applied Biosystems: ThermoFisher Scientific, Massachusetts, USA) in the condition of 95°C, 20 seconds for enzyme activation, followed by 40 cycles of denaturation step (95°C, 1 second) and annealing/extension step (60°C, 20 minutes). Expression of hsa-miR-126-3-p was analyzed as relative quantification by normalizing its cycle threshold (Ct) value with the Ct of hsa-miR-16-5p. Relative quantification in different specimens and time points was calculated as 2^−ΔCt (ΔCt = The average Ct of each hsa-miRNA – The average Ct of hsa-miR-16-5p) [22, 33].

Statistical analysis

All data were presented as tables and figures. Statistical analysis was performed with the IBM SPSS version 23.0 (SPSS Inc., Chicago, Illinois, USA) and GraphPad Prism version 9.3.0.463 (GraphPad Software Inc., San Diego, California, USA). Quantitative data such as age, body temperature, duration of fever (DOF), hemoglobin concentration, hematocrit, platelet count, white blood cell (WBC) count, differential WBC, and the level of serum miRNAs were presented as median and interquartile range (IQR) values. The comparison between or among groups was analyzed using Mann-Whitney U Test, Wilcoxon test, or Kruskal–Wallis test where appropriate. Qualitative data such as demographic and laboratory data were presented as percentages. A Chi-squared test was used for comparing results between or among groups. The correlation between each miRNA expression and viral load, blood parameters, or other factors was explored using Spearman’s rank correlation statistics. The p-value <0.05 was considered statistically significant.

Results

Patient characteristics

The demographic data of 90 enrolled patients (30 AFI, 30 DF, and 30 DHF) were presented in Table 1. Dengue-infected patients (DF and DHF) were older than AFI (p = 0.010). Dengue-infected patients had a significantly longer duration of fever than AFI (p < 0.001). Secondary dengue infection was more frequent in DHF. DHF had significantly higher hematocrit (Hct) values than DF and AFI (40.10% vs 38.10% vs 35.95%, respectively; p = 0.011). Platelet count values in DHF were significantly lower than DF and AFI (97.00 × 10³ vs 150.00 × 10³ vs 260.00 × 10³, respectively; p < 0.001). Other hematologic parameters including WBC count and differential WBCs among the three groups were not significantly different (p > 0.05). Dengue viral load was not statistically significant compared between DF and DHF (4.626 vs 4.525, p = 0.520).

Table 1.

Demographic, clinical, and laboratory data of all groups in this study.

Parameter	DF(n = 30)	DHF(n = 30)	AFI* (n = 30)	p-value**
Age (years)	9.00 (7.00–10.00)	10.50 (8.75–12.00)	8.50 (7.00–10.00)	0.010
Sex:*** Male Female	19 (63.33) 11 (36.67)	17 (56.67) 13 (43.33)	17 (56.67) 13 (43.33)	0.832
DOF (days)	6.00 (5.00–7.00)	6.00 (5.00–7.00)	3.00 (2.00–5.00)	<0.001
Type of infection:*** Primary Secondary	9.00 (30.00) 21.00 (70.00)	2.00 (6.67) 28.00 (93.33)	NA NA	0.04****
Hematocrit	38.10 (35.15–40.20)	40.10 (36.83–43.00)	35.95 (32.03–37.88)	0.011
Hemoglobin (g/dL)	12.30 (11.40–13.23)	13.05 (12.23–14.05)	12.25 (11.45–13.23)	0.090
WBCcount (cells/μL)	2,900.00(2,458.00–4,492.00)	4,100.00(2,475.00–5,625.00)	6,025.00(3,665.00–9,338.00)	0.074
Platelet count(x103/μL)	150.00(123.30–186.30)	97.00(55.50–165.00)	260.00(213.50–282.30)	<0.001
% Neutrophils	56.50 (35.75–76.25)	51.00 (37.00–68.00)	60.00 (46.50–69.75)	0.640
% Lymphocytes	37.50 (16.75–51.50)	41.00 (24.00–51.00)	33.00 (20.50–46.50)	0.825
% Atypicallymphocytes	0.50 (0.00–8.00)	2.00 (0.00–4.50)	0.00 (0.00–0.50)	0.138
% Monocytes	5.00 (2.00–8.00)	4.00 (2.50–7.00)	3.50 (2.50–8.50)	0.933
% Eosinophils	0.00 (0.00–1.00)	0.00 (0.00–1.00)	1.00 (0.00–4.00)	0.479
% Basophils	0.00 (0.00–0.00)	0.00 (0.00–0.00)	0.50 (0.00–1.25)	0.141
dengue viral load (Log PFU equivalent/mL)	4.626(1.000–8.017)	4.525(1.000–7.554)	NA	0.520

Results were presented as median and interquartile range (IQR) values.

DOF = Duration of fever

*AFI referred to patients with acute febrile illness and diagnosed as non-dengue infection.

** p-value was based on Kruskal–Wallis test.

***Results were presented as frequency and (percentage).

**** p-value was based on Fisher exact test compared between DF and DHF.

NA = Not available; DF = dengue fever; DHF = dengue hemorrhagic fever

The levels of CmiR-126-3p expression

The levels of hsa-miR-126-3p expression in febrile and convalescent serum in all three groups of patients were reported as relative quantification using hsa-miR-16-5p for data normalization. The results were presented in Table 2 as median and IQR values.

Table 2.

The expression of serum hsa-miR-126-3p among DF, DHF, and AFI during febrile and convalescent phases [median (IQR)].

	Febrile phase	Convalescent phase	p-value*
DF (n = 30)	0.349 (0.117–0.590)	0.433 (0.275–0.660)	0.025
DHF (n = 30)	0.391 (0.246–0.576)	0.468 (0.302–0.614)	0.115
DF+DHF (n = 60)	0.368 (0.200–0.583)	0.443 (0.299–0.615)	0.006
AFI (n = 30)	1.250 (0.602–2.077)	1.165 (0.569–1.960)	0.730
p-value**(DF vs DHF vs AFI)	<0.001	<0.001
p-value***(DF vs AFI)	<0.001	<0.001
p-value***(DHF vs AFI)	<0.001	<0.001
p-value***(DF vs DHF)	0.295	0.535
p-value***(DF+DHF vs AFI)	<0.001	<0.001

Results were presented as median and interquartile range (IQR) values.

AFI referred to patients with acute febrile illness and diagnosed as non-dengue infection.

DF = dengue fever; DHF = dengue hemorrhagic fever

* p-value was based on Wilcoxon matched-paired signed rank test.

**p-value was based on Kruskal–Wallis test.

***p-value was based on Mann Whitney U test.

The expression of serum hsa-miR-126-3p during febrile phase among the three groups was significantly different (p < 0.001). AFI had significantly higher level of CmiR-126-3p compared to DF and DHF and combined DF+DHF, respectively (1.250 vs 0.349, p < 0.001; 1.250 vs 0.391, p < 0.001, 1.250 vs 0.368, p < 0.001, respectively). However, no significant difference in CmiR-126-3p expression was observed between DF and DHF (0.349 vs 0.391, p = 0.295).

Similarly, the serum hsa-miR-126-3p expression during convalescent phase in AFI was higher than DF, DHF, and DF+DHF (1.165 vs 0.433, 0.468, and 0.443, respectively; p < 0.001 for all comparison). No significant difference in hsa-miR-126-3p expression between DF and DHF was observed (0.433 vs 0.468, p = 0.535).

The levels of CmiR-126-3p in dengue-infected patients diagnosed as primary or secondary infection

Serum hsa-miR-126-3p expressions in dengue-infected patients were compared between primary and secondary infection. Results were presented in Table 3.

Table 3.

The expression of serum hsa-miR-126-3p during febrile and convalescent phases in dengue-infected patients (DF+DHF) diagnosed as primary or secondary infection [median (IQR)].

	Primary infection	Secondary infection	p-value*
Febrile phase	0.335 (0.119–0.591)	0.374 (0.204–0.576)	0.747
Convalescent phase	0.433 (0.282–0.939)	0.445 (0.301–0.594)	0.781

Results were presented as median and interquartile range (IQR) values.

*p-value was based on Mann Whitney U test.

During the febrile phase, CmiR-126-3p expression in patients with primary infection and secondary infection was not significantly different (0.335 vs 0.374, p = 0.747). The expression during the convalescent phase also revealed no significant difference (0.433 vs 0.445, p = 0.781).

The dynamic of CmiR-126-3p levels

CmiR-126-3p expression during febrile and convalescent phases was explored and presented in Table 2 and Figure 1. CmiRNA-126-3p expression during convalescent phase was significantly higher than febrile phase in DF and combine DF+DHF (0.443 vs 0.349, p = 0.025 and 0.443 vs 0.368, p = 0.006, respectively). Although CmiR-126-3p expression in DHF had a trend to be higher during the convalescent phase, it was not statistically significant (0.468 vs 0.391, p = 0.115). There was no difference in CmiR-126-3p expression in febrile and convalescent sera in AFI.

Figure 1.

The dynamic of median hsa-miR-126-3p levels in serum of DF, DHF, and AFI during febrile and convalescent phases. The expression levels were calculated with the formula 2^−ΔCt and hsa-miR-16-5p was selected as internal and reference control for data normalization. (*Statistically significant difference after analyzed by Wilcoxon matched-pairs signed rank test; p< 0.05).

The correlation between serum CmiR-126-3p expression and laboratory parameters

The correlations between hsa-miR-126-3p expression in dengue-infected patients and laboratory results (hematocrit, WBC count, platelet count, differential WBC, and viral load) were investigated. Results were presented in Figure 2. No association between hsa-miR-126-3p expression and all investigated laboratory parameters was observed (p > 0.05).

Figure 2.

The correlations of serum hsa-miR-126-3p expression with hematocrit (a), WBC count (b), platelet count (c), differential WBC (d-f), and viral load (g).

Discussion

Hsa-miR-126-3p is one of the endothelial-cell-related miRNAs and is highly expressed in endothelial cells [17,19]. The change of expression level may reflect the dysregulation of endothelial cells and pathological disturbance of blood vessels, which correlate with the pathogenesis of vascular-related diseases including dengue infection. Vascular leakage and endothelial dysfunctions remain the hallmarks of severe dengue infection [12, 34]. The purpose of this study was to evaluate the expression of serum hsa-miR-126-3p in dengue-infected patients. We also explore the dynamics of miRNA expression and the correlation between CmiR-126-3p and blood and virological parameters. To our knowledge, this is the first study demonstrating the putative expression of miR-126 in serum of patients with dengue infection and the correlation with other relevant factors, particularly in pediatric patients.

Data normalization is one of the critical points to be considered during performing miRNA expression. In this study, we had selected endogenous hsa-miR-16-5p as the reference for data normalization [22, 30–32]. We ensure that endogenous hsa-miR-16-5p is suitable for our experiment because its median Ct among three groups was not different in both febrile and convalescent specimens (data not shown). These imply the consistent expression of hsa-miR-16-5p in each serum specimen. In addition, AFIs were included in this study instead of using healthy volunteers to reduce the bias and ensure our results were related to dengue-infection, not fever.

The significant decrease in CmiR-126-3p expression in dengue-infected patients compared to AFI during both febrile and convalescent periods indicates that miR-126-3p may have a role involving in the pathogenesis of dengue infection and may be the early and late candidate biomarker for dengue infection. However, the decreased expression of CmiR-126-3p is not specific as it was also found in many other diseases/conditions, such as psoriasis [35], influenza virus A (H1N1) infection [36], type 2 diabetes mellitus [37], intracerebral hemorrhage [38], sepsis [21] and smokers [22]. The anti-inflammatory effects of hsa-miR-126-3p are mediated by reduction of vascular inflammation, angiogenesis, cell proliferation, and immune-cell trafficking to inflammatory sites. Moreover, the expression of hsa-miR-126-3p from endothelial cells decreases vascular cell adhesion molecule 1 (VCAM-1) expression and leukocyte adherence to vascular cells resulting in the attenuation of inflammation [20]. The downregulation of CmiR-126-3p in dengue-infected patients indicates that anti-inflammatory processes in endothelial cells are suppressed in dengue-infected patients and the endothelial or vascular inflammation may be one important pathophysiology in dengue infection. Endothelial dysfunctions during dengue infection may be the effects of cytokines and chemokines released from dengue-infected dendritic cells, monocytes and macrophages, dengue-infected endothelial cells, and dengue NS-1 protein [39]. However, the specific mechanisms among hsa-miR-126-3p, endothelial cells and DENV remain unknown.

There are currently no clinical or laboratory tests that can accurately predict development of severe complication in dengue patients during the early stages of infection. Therefore, finding a candidate test is important. One of our exploratory objective is to describe the difference in hsa-miR-126-3p expression compared between DF and DHF. There was no statistical difference in CmiR-126-3p expression compared between DF and DHF. Our results are similar to a previous study demonstrating the similar levels of miR-126 expression in febrile PBMCs of DF and DHF patients [29]. On the other hand, a recent study revealed that plasma hsa-miR-126-5p was more downregulated in dengue-infected adult patients with WS and SD compared with UD patients [27]. However, this previous study was performed in adult patients and used different criteria for classifying clinical symptoms in dengue-infected patients. The similar levels of CmiR-126-3p expression between DF and DHF patients may imply that hsa-miR-126-3p is not correlated with dengue severity, particularly vascular leakage in pediatric patients. The finding that the CmiR-126-3p level was persistently low throughout the febrile and convalescent phases also indicates that CmiR-126-3p is not correlated with the short-lived (48–72 hours) vascular leakage occurring in severe dengue (DHF or DSS). In addition, the finding that CmiR-126-3p levels in the febrile phase were similar between primary and secondary dengue infection further supports that hsa-miR-126-3p alone is not correlated dengue severity. We, therefore, conclude that CmiR-126-3p level is not correlated with dengue severity or disease progression.

Although the diagnostic specificity of CmiR-126-3p may be low compared to other dengue diagnostic tests, such as RT-PCR, dengue NS1, and serological tests, it may be useful in the cases that have ambiguous other dengue diagnostic results or in the cases that have inconsistency between clinical diagnosis and other dengue diagnostic tests.

The correlation between serum hsa-miR126-3p and laboratory parameters was investigated. All of the laboratory parameters were obtained during febrile phase. There was no correlation between serum hsa-miR-126-3p expression and either Hct, WBC count, platelet count, %neutrophils, %monocytes, % lymphocytes, or dengue viral load. These findings are consistent with the finding that the CmiR-126-3p level is not correlated with dengue severity.

Although this study shows that hsa-miR-126-3p alone is not correlated with dengue severity, it is still possible that hsa-miR-126-3p may have a role in the complex pathogenesis of dengue severity, particularly by acting in concert with other miRNA or other pathogenic factors. Hsa-miR-126-3p may directly affect endothelial cells and regulate inflammation during dengue infection in many possible mechanisms. Firstly, hsa-miR-126-3p regulates endothelial cell permeability by inhibiting phosphatidylinositol 3‑kinase (PI3K) and mitogen-activated protein kinases (MAPK) signaling pathways. It also indirectly regulated phosphorylation of cofilin and other proteins affecting endothelial cell junctions by inhibiting sprouty related EVH1 domain containing 1 (SPRED1) gene. Secondly, due to the expression of tumor necrosis factor alpha (TNF-α) activating nuclear factor kappa B (NFκB) and interferon regulatory factor 1 (IRF1), resulting in VCAM-1 activation and induction of leukocyte adherence and inflammation, expression of hsa-miR-126-3p may act as anti-inflammatory functions by preventing leukocyte migration and inflammation by inhibiting VCAM-1. Moreover, hsa-miR-126-3p involves in the interaction cascade with other miRNAs, such as miR-155 and miR-221/222, with the intermediate transcription factor called ETS proto-oncogene 1 (Ets-1). Ets-1 induces miR-126 expression, which results in the inhibition of VCAM-1 and attenuation of inflammation and T-cell recruitment to vascular tissues during pathological changes [17]. Therefore, further studies are needed to verify the role of hsa-miR-126-3p in severe dengue infection.

In this study, the suggested guidelines such as using the isolated areas for PCR preparation, aliquoting reagents prior to work, and avoiding the splashing of reagents [40] were followed to prevent cross contamination and to avoid false-positive PCR.

There are some limitations of this study. Firstly, there was no data from the previous study for comparison with our results, especially CmiR-126-3p expression during the convalescent phase. Secondly, this study did not include dengue-infected patients with other organ complications. Thirdly, the sample size in each group of the study is quite small. Moreover, we cannot compare CmiR-126-3p expression among patients with different serotypes due to the small number of patients in each serotype. Further studies need to be conducted in different cohorts with larger sample sizes and with various clinical symptoms to confirm our pilot findings. In addition, although this study revealed an interesting fact that CmiR-126-3p was decreased in dengue infection, there was no evidence that this CmiR was associated with dengue hemorrhagic fever implying that it was not associated with vascular leakage. Unfortunately, because we used stored serum samples, we cannot perform the association of this CmiR and the pro- and anti-inflammatory cytokines, which had been reported to be related to hsa-miR-126-3p and plasma leakage. Therefore, additional studies are needed to explain the findings in this study. Moreover, other miRNAs related to hsa-miR-126-3p or pathogenesis of severe dengue infection should be further investigated to find out the suitable biomarkers for severe dengue disease.

Conclusion

In summary, we demonstrate the low expression of serum hsa-miR-126-3p in dengue-infected patients in both febrile and convalescent phases compared with AFI. The different hsa-miR-126-3p expressions in dengue-infected patients do not depend on primary or secondary infection as well as the disease severity. CmiR-126-3p may be the alternatively promising biomarker for dengue infection during the early or late phase of infection. However, we cannot demonstrate the potential role of CmiR-126-3p as a candidate miRNA for predicting dengue severity as well as the role for miRNA-based therapies to prevent vascular permeability in dengue-infected patients. Further studies with other miRNAs related to hsa-miR-126-3p or pathogenesis of severe dengue episode need to be investigated.

Funding Statement

This work was supported by The Thailand Research Fund (TRF) and Office of the Higher Education Commission, Thailand under Grant [MRG6180126].

Disclosure statement

No potential conflict of interest was reported by the author(s).

Authors’ contributions

All authors contributed to this project including manuscript preparation. M.S. designed and performed experiments, reviewed and analyzed data, and wrote the first draft of the manuscript. J.R and P.S. performed the experiments and collected data. S.C. and K.L. designed the experiments and reviewed data including manuscript. C.S. designed the experiments, reviewed data and involved in manuscript review, editing, and correction. All authors have reviewed and approved the final version of this manuscript.

Ethical declaration

This project was reviewed and approved by The Ethics Committee of Faculty of Tropical Medicine, Mahidol University, Thailand (MUTM 2018-022-01). Procedures conducted in this study were in accordance with the standard criteria of the responsible committee on human experimentation (The Human Ethical Research Committee of Faculty of Tropical Medicine, Mahidol University, Thailand) and with the Declaration of Helsinki 1975 (revised in 2000).