Longitudinal changes of serum angiopoietin 1 (Ang-1) and angiopoietin 2 (Ang-2) associated with endothelial stability in dengue patients with different disease stages were studied. Serum Ang-1 and Ang-2 levels were measured in confirmed dengue fever (DF) patients on admission (DFA, n = 40) and discharge (DFD, n = 20); in dengue hemorrhagic fever (DHF) patients on admission (DHFA, n = 40), at critical stage (DHFC, n = 36), and on discharge (DHFD, n = 20); and in healthy controls (HC, n = 25).
KEYWORDS: dengue fever, dengue hemorrhagic fever, angiopoietin, endothelial dysfunction, biomarker
ABSTRACT
Longitudinal changes of serum angiopoietin 1 (Ang-1) and angiopoietin 2 (Ang-2) associated with endothelial stability in dengue patients with different disease stages were studied. Serum Ang-1 and Ang-2 levels were measured in confirmed dengue fever (DF) patients on admission (DFA, n = 40) and discharge (DFD, n = 20); in dengue hemorrhagic fever (DHF) patients on admission (DHFA, n = 40), at critical stage (DHFC, n = 36), and on discharge (DHFD, n = 20); and in healthy controls (HC, n = 25). DHFC had the highest serum Ang-2 and lowest Ang-1 levels compared to DFA, DHFA, and HC (P < 0.050). The ratio of serum Ang-2/Ang-1 in DHFC was the highest among all study categories tested (P < 0.001). Significant positive correlations were observed between serum Ang-1 and platelet count in DHFA (Pearson r = 0.653, P < 0.001) and between Ang-1 and pulse pressure in DHFC (r = 0.636, P = 0.001). Using a cutoff value of 1.01 for the Ang-2/Ang-1 ratio for DHFC, a sensitivity of 83.2% and a specificity of 81.2% discerning DF from DHF were obtained. Therefore, serum Ang-2/Ang-1 could be used as a biomarker for endothelial dysfunction in severe dengue at the critical stage.
INTRODUCTION
The hallmark of the pathogenesis of dengue hemorrhagic fever (DHF), a severe form of dengue infection, is high vascular permeability leading to plasma leakage, the most characteristic feature in DHF (1). Plasma leakage results in several clinical manifestations in DHF patients, including hemoconcentration, pleural and pericardial effusion, abdominal ascites, etc. (2). The duration of the leaking phase in DHF progresses quickly within 3 to 7 days of onset of illness and resettles within 1 to 2 days with fluid recovery following the leaking phase that persists for about 48 h (3). The immune response against the dengue virus (DENV) activates dengue virus-specific T lymphocytes (4), which produce high levels of different cytokines resulting in a storm of cytokines that acts on the endothelium leading to transient endothelial instability resulting in vascular permeability (5).
The Angiopoietin signaling system, comprising angiopietin-1 (Ang-1), angiopoietin-2 (Ang-2), and the Tie-2 receptor of vascular endothelial cells, is one of the mechanisms that influences vascular endothelial permeability (6). Ang-1 is produced by pericytes and smooth muscle cells surrounding the endothelial cell monolayer (7). Ang-2 is produced in endothelial cells, stored in Weibel-Palade bodies (WPBs) inside the endothelial cells, (8) and rapidly released upon exposure to inflammatory stimuli. Under regular conditions, Ang-1 binds to the Tie-2 receptor and initiates survival pathways resulting in endothelial cell quiescence. In contrast, due to inflammation triggered by infection, Ang-2 is released which binds to Tie-2 receptor and acts as an antagonistic ligand of Ang-1 promoting proinflammatory stimuli and destabilization of the cell-cell junction via the Rho kinase pathway (9). Loss of cell-cell contact plays an important role by facilitating cell migration and new vessel formation. This process causes capillary leakage and facilitates transmigration of leukocytes (10).
The role of serum Ang-2 has been shown as an important biomarker in breast cancer (11), non-small-cell lung cancer (12) and infectious diseases such as cerebral malaria (13). A previous study on dengue infection had reported elevated levels of Ang-2 levels and Ang-2/Ang-1 ratios in children (1 to 15 years old) with dengue shock syndrome (DSS) on admission to the hospital compared to dengue hemorrhagic fever patents (DHF) and healthy controls (14). The imbalance in plasma Ang-2 and Ang-1 levels was associated with thrombocytopenia and endothelial activation. Another study in children on admission had shown elevated Ang-2 levels in DSS patients compared to DHF and dengue fever (DF) patients. However, the differences in angiopoietin levels were not significant between DHF and DF, and no significant differences were observed after 48 h (15). Serum Ang-2 levels were significantly higher in dengue patients with plasma leakage than in those without plasma leakage (10).
In addition to the response to Ang-1 and Ang-2, endothelial cells can alter and regulate their permeability due to cross-reactivity of the dengue nonstructural protein 1 (NS1) with vascular endothelium (16, 17). The dengue NS1 proteins could trigger the release of more vasoactive cytokines from immune cells such as interleukin-6 (IL-6), tumor necrosis factor alpha (TNF-α), IL-1β, and IL-8. Moreover, recent studies have reported a higher cross-reactivity of anti-NS1 antibodies with several host molecules and cells, including thrombin, plasminogen, platelets and, more importantly, endothelial cells, via the molecular mimicry mechanism (18–23).
An in vitro study to examine the effect of dengue virus on primary human umbilical vein endothelial cells with respect to the regulation of different endothelial factors such as Ang-1, Ang-2, vascular endothelium-cadherin (VE-cadherin) and zona occluding-1 (ZO-1) suggested that the upregulation of Ang-2 and the downregulation of Ang-1 are strongly dependent on the endothelial damage in DV infected cells in a dose-dependent manner (24). Another study showed that knockout mice infected with DENV-3 treated with recombinant Ang-1 type I and type II interferon receptors had an extended survival rate (25). Moreover, human dermal microvascular endothelial cells (HMEC-1) treated with nonstructural protein 1 of DENV secreted high levels of Ang-2 (26).
Although several investigations have explored the associations between Ang levels and dengue severity, none have investigated the dynamics of Ang-1 and Ang-2 with the progression of dengue disease from admission to discharge through the critical stage among DHF patients. Therefore, the present study was designed to investigate the changing patterns of Ang-1 and Ang-2 levels for different disease stages of DF and DHF over the days of fever from admission and to progression to the critical stage in DHF. The potential use of these endothelial factors as biomarkers for endothelial dysfunction was analyzed. In addition, the association between clinical parameters and Ang-1 and Ang-2 levels in DF and DHF patients at different disease stages was studied.
MATERIALS AND METHODS
Recruitment of patients.
The maximum possible number of DHF patients was required to test the association between disease severity and Ang-1 and -2 levels. For this purpose, a total of 116 patients admitted to the Colombo North Teaching Hospital, Ragama, Sri Lanka (wards 9 and 12), were tested for dengue positivity. Patients with signs of plasma leakage, i.e., the presence of various degrees of abdominal ascites detected by ultrasonography and/or pleural effusion detected by chest X-ray, were categorized as DHF cases (n = 40). Age- and gender-matched DF patients (n = 40) were selected to match DHF patients. Blood samples were collected from June 2015 to June 2016. Ethical clearance was obtained from the Ethics Review Committee of the Faculty of Medicine, University of Colombo, Colombo, Sri Lanka (ERC-13-172). Informed written consent was obtained from each participant prior to sampling. Patients with other flaviviral infections and pregnant women were excluded from the study.
Disease confirmation, clinical characterization, and categorization of patients.
Patients with dengue were confirmed by detection of anti-dengue IgM antibody using enzyme-linked immunosorbent assay (ELISA; VirionSirion, Germany) if they were admitted to hospital on or after day 4 of fever and/or by using a rapid chromatographic detection of dengue nonstructural protein 1 (NS1) antigen (SD Bioline, South Korea) if they were admitted to hospital on or before day 4 of fever.
Clinical characterization of DF and DHF was done using comprehensive guidelines for the prevention and control of dengue and dengue hemorrhagic fever (27). Accordingly, patients with acute febrile fever with two or more of the following symptoms, including headache, retro-orbital pain, myalgia, athralgia, skin rash, leucopenia (white blood cell count < 5,000 cells/mm3), mild thrombocytopenia (100,000 to 150,000 cells/mm3), and mild (∼10%) hematocrit rise, were considered positive for dengue fever. World Health Organization criteria (27) were used to identify severe dengue, including acute onset of high fever lasting 2 to 7 days, thrombocytopenia (≤100,000/mm3) and one or more evidences of high vascular permeability (rising hematocrit ≥20% from baseline levels [baseline level for males is about 40% and for females about 36%], hypoproteinemia, and the presence of pleural effusion and abdominal ascites on chest X-ray or ultrasonography). For this study, patients who showed the most objective evidence of plasma leakage, i.e., abdominal ascites in various degrees ranging from pericholecystic fluid to massive ascites and pleural effusion, were included as confirmed DHF patients. Continuous hematological and serum biochemical measurements of patients were recorded during their hospital stay.
Recruitment of healthy controls.
Twenty-five healthy controls of both sexes in the age range of DHF patients were recruited for the study. Individuals who were negative for previous dengue or any flavivirus episodes and without fever during the week of sample collection were included in this study. Since almost all patients were residents of the Gampaha district of the Western province of Sri Lanka, healthy individuals in the community were recruited from the Gampaha district with prior approval from the Regional Director of Health Services, Gampaha district, Sri Lanka.
Sample collection and processing.
A diagnosis of dengue was made using rapid detection of NS1 antigen and/or IgM ELISA. The first blood sample of confirmed dengue patients was collected on admission to the hospital. The patients were monitored, and another sample was collected from patients who showed evidence of plasma leakage approximately within 24 h of the commencement of the leaking phase at the time point where the hematocrit increased by 20% from the baseline value and/or signs of plasma leakage such as pleural effusion and abdominal ascites were detected. This sample was considered the DHF sample at the critical stage (DHFC). At this stage, admission samples were categorized as DF on admission (DFA) and DHF on admission (DHFA). Another blood sample was collected from dengue patients on the day of discharge (DFD in DF patients and DHFD in DHF patients). DFA, DHFA, and DHFC patients were further categorized based on the day of fever the samples were collected (e.g., a DF patient admitted on day 3 of fever was denoted DFA-3).
Portions (5 ml) of blood were collected by venipuncture into a sterile plain tube. The serum was separated by centrifugation at 900 × g for 10 min, and clear sera were aliquoted and stored at –80°C.
Determination of serum Ang-1 and Ang-2 levels.
Serum Ang-1 and Ang-2 levels were measured using commercially available kits (Ray Biotech) according to the manufacturer’s instructions. The kits were selected after reviewing available commercial kits for their high sensitivity, where the lowest detection limit for Ang-1 is 30 pg/ml and that for Ang-2 is 10 pg/ml, and high specificity since they do not have cross-reactivity with many human cytokines and other mediators. Samples were tested in duplicate, and the concentrations of serum Ang-1 and Ang-2 were interpolated using standard curves of the respective recombinant human proteins. Threefold dilution series ranging from 18,000 to 24.9 pg/ml for Ang-1 and from 3,000 to 4.12 pg/ml for Ang-2 were prepared using 50 ng/ml of recombinant human Ang-1 stock solution and 130 ng/ml of recombinant human Ang-2 stock solution, respectively. Assay diluent solutions provided in the kit were used for the preparation of standards. Before starting the assay, the serum samples and all required reagents were brought to room temperature. Sera were then diluted using assay diluents provided in each test kit (Ang-1 [1:20] and Ang-2 [1:5]). The dilution factors were optimized for both assays.
Data analysis.
Statistical analysis was performed using the Statistical Package for the Social Sciences (SPSS) v16.0. Kolmogorov-Smirnov tests were done to determine the distributions of data. Data were summarized using descriptive statistics and are presented as means ± the standard deviations (SD). A chi-square test was used to compare proportions. Continuous data with two categories were compared using the independent-sample t test. Repeat observations were tested using the paired t test. One-way analysis of variance (ANOVA) with the Bonferroni post hoc correction was used to compare continuous data with more than two categories. Spearman correlation coefficient was used to test the association between two continuous variables. A receiver operating characteristic (ROC) curve was used to obtain the optimum cutoff value of a variable that best discriminates two categories of subjects. A P value of <0.05 was considered significant.
Data availability.
All relevant data are available in the article.
RESULTS
Forty acute dengue-positive patients each from DF and DHF categories were enrolled in the study. In the DF patients, 40 samples were obtained on admission (DFA), and 20 samples were obtained at discharge (DFD). In DHF patients, 40 DHFA, 36 DHFC, and 20 DHFD samples were obtained. Twenty-five healthy controls were recruited for the study. The demographic profile of participants is given in Table 1.
TABLE 1.
Demographic profile of study participants
| Demographic profile | Study participants |
||
|---|---|---|---|
| DF (n = 40) | DHF (n = 40) | HC (n = 25) | |
| Mean age ± the SD (yr) | 28.5 ± 12.6 | 29.2 ± 10.5 | 28.5 ±11.8 |
| Gender (male:female) | 1.29:1.0 | 1.38:1.0 | 1.09:1.0 |
| Mean BMIa ± the SD (wt/ht2) | 19.3 ± 4.2 | 21.8 ± 2.6 | 22.6 ± 3.9 |
BMI, body mass index.
DHF patients were confirmed using evidence of plasma leakage based on clinical, X-ray, and ultrasonography evidence. However, we did not quantify the extent of plasma leakage. Ultrasonography results showed that the majority of patients (79%) had abdominal ascites to various degrees, ranging from pericholesyctic fluid to massive ascites, and 45% of DHF patients had evidence of a pleural effusion in the right side on chest X-ray.
The means of clinical and biochemical parameters were calculated in different stages of the disease in DF and DHF patients, including on admission, at the critical stage (for DHF), and on discharge (Table 2). DHF patients had a significantly lower pulse pressure on admission than DF patients on admission (P = 0.015). DHF patients had higher body temperatures (P = 0.002), platelet counts (P = 0.017), and neutrophil counts (P = 0.006) and lower levels of alanine aminotransferase (ALT) (P = 0.040) and lymphocytes (P = 0.020) on admission than those at the critical stage.
TABLE 2.
Changes of clinical and biochemical parameters in samples of patients at different stages of disease
| Parameter | Mean ± SDa
|
||||
|---|---|---|---|---|---|
| DF patients on admission (n = 40) | DF patients on discharge (n = 20) | DHF patients on admission (n = 40) | DHF patients at critical stage (n = 36) | DHF patients on discharge (n = 20) | |
| Body temp (°C) | 38.3 ± 0.7AC | 36.5 ± 0.1 | 38.3 ± 0.9EF | 37.1 ± 0.6 | 37.1 ± 0.5 |
| Pulse rate (beats/min) | 81.4 ± 7.8 | 78.2 ± 3.1D | 85.6 ± 9.1 | 81.3 ± 7.1 | 83.0 ± 3.3 |
| Pulse pressure (mm Hg) | 35.3 ± 5.0C | 35.1 ± 7.1 | 35.2 ± 7.5F | 31.2 ± 5.2G | 38.5 ± 6.3 |
| Respiratory rate (breaths/min) | 23.1 ± 3.0A | 21.3 ± 1.4 | 23.3 ± 2.6 | 23.6 ± 2.0 | 23.1 ± 1.2 |
| Platelet count (103/mm3) | 81.0 ± 33.5C | 66.6 ± 45.8 | 61.2 ± 62.3EF | 30.4 ± 26.1 | 38.5 ± 14.5 |
| Hematocrit (%) | 40.9 ± 4.4 | 41.5 ± 2.2 | 42.0 ± 4.5 | 42.1 ± 3.8 | 39.2 ± 3.1 |
| Lowest leukocyte count (×1,000) | 3.2 ± 1.7C | 4.8 ± 2.3 | 4.4 ± 3.2 | 6.4 ± 3.3E | 7.1 ± 3.2 |
| AST (U/liter) | 152.6 ± 201.8A | 285.0 ± 182.0 | 225 ± 264.0 | 270.2 ± 247.5 | 263.1 ± 245.2 |
| ALT (U/liter) | 105.2 ± 142.1A | 286.5 ± 196.6 | 123.1 ± 148.0E | 185.9 ± 156.4 | 171.3 ± 109.5 |
| Hemoglobin (g/dl) | 13.5 ± 1.9 | 12.4 ± 3.0 | 14.5 ± 2.1 | 13.9 ± 1.3 | 13.5 ± 1.3 |
| Red blood cells (×106/μl) | 4.7 ± 0.6 | 4.8 ± 0.4 | 5.2 ± 2.1 | 4.9 ± 0.4 | 4.7 ± 0.5 |
| Neutrophils (%) | 55.1 ± 17.2AC | 39.1 ± 17.8 | 54.7 ± 19.6EF | 39.5 ± 14.3 | 40.8 ± 13.5 |
| Lymphocytes (%) | 34.8 ± 16.4 | 45.1 ± 12.4 | 32.1 ± 14.9EF | 45.2 ± 12.4 | 45.4 ± 10.2 |
| Monocytes (%) | 7.7 ± 3.7C | 11.8 ± 7.6 | 8.1 ± 5.4 | 11.1 ± 5.8 | 11.6 ± 5.7 |
| Eosinophils (%) | 1.7 ± 2.1 | 2.9 ± 1.5 | 1.4 ± 1.6 | 1.6 ± 1.0 | 2.3 ± 1.2 |
| Basophils (%) | 1.8 ± 4.2 | 1.5 ± 1.2 | 1.2 ± 1.3 | 2.8 ± 5.2 | 0.7 ± 0.3 |
Values with significant differences are indicated in boldface. Significant differences are indicated by superscript capital letters as follows: A, (between) DFA and DFD (paired t test); B, DFA and DHFA (independent-sample t test); C, DFA and DHFC (paired t test); D, DFD and DHFD (independent-sample t test); E, DHFA and DHFC (independent-sample t test); F, DHFA and DHFD (paired t test); G, DHFC and DHFD (paired t test; P < 0.050). AST, aspartate aminotransferase; ALT, alanine aminotransferase.
Changes in serum ANG-1, ANG-2, and ANG-1/2 levels in different patient samples and healthy controls.
The tests were performed in duplicate, and the intra-assay coefficient of variances (CV) were 7.4% and 12.0% for Ang-1 and Ang-2, respectively. The reproducibility of the tests was assessed considering the standard series of serum Ang-1 and Ang-2, which was measured in two panels, and the interassay CVs were 13.7% for Ang-1 and 14.5% for Ang-2.
The serum Ang-2/Ang-1 ratio was significantly higher in DHFC (3.88 ± 3.25) samples than in all other sample categories (DFA, 0.49 ± 0.55; DFD, 0.60 ± 0.52; DHFA, 1.19 ± 2.39; DHFD, 1.01 ± 2.46; HC, 0.18 ± 0.14, P < 0.001) (Fig. 1A).
FIG 1.
Comparison of serum Ang-2/Ang-1 (A), Ang-2 (B), and Ang-1 (C) levels in different sample categories. Categories include dengue fever (DF, circle), and dengue hemorrhagic fever (DHF, black diamond) in different disease stages and healthy controls (HC, black square). ***, Significant differences between DHFC samples and all other categories of samples (P < 0.001). Significance was determined using one-way ANOVA with the Bonferroni post hoc correction for independent samples. A paired sample t test was used for repeated measures.
Serum Ang-2 levels were the opposite of Ang-1, where the DHFC samples had the highest Ang-2 levels (10.68 ± 4.49 ng/ml), which were significantly higher than those of DFA samples (5.63 ± 3.96 ng/ml; P < 0.001), DHFA samples (6.56 ± 3.27 ng/ml; P < 0.001), DFD samples (5.82 ± 2.87 ng/ml; P = 0.002), and HC samples (4.23 ± 1.93 ng/ml; P < 0.008) (Fig. 1B).
DHFC samples had the lowest serum Ang-1 levels (9.41 ± 6.66 ng/ml) compared to DHFA samples (20.57 ± 14.30 ng/ml; P = 0.022), DFA samples (19.87 ± 11.83 ng/ml; P = 0.010), DHFD samples (22.94 ± 15.28 ng/ml; P = 0.038), and HC samples (32.82 ± 14.30 ng/ml; P < 0.001). The highest level of Ang-1 was recorded in HC samples and was significantly higher than in all patient categories (P < 0.050) except DHFD (P = 0.218) (Fig. 1C).
ROC curve analyses were done for the Ang-1, Ang-2, and Ang-2/Ang-1 levels. The curves were plotted considering the levels in DHFC samples against all other samples, including those of healthy controls. Serum Ang-2/Ang-1 had the highest area under the curve (91.2%) compared to serum Ang-1 and Ang-2 analyzed separately. The Ang-2/Ang-1 ratio had a sensitivity of 83.2% and a specificity of 81.2% using a cutoff value of 1.01 (Fig. 2). This suggests that serum Ang-2/Ang-1 has the potential to be used as a biomarker for the critical stage of DHF using a cutoff value of 1.01.
FIG 2.
Comparison of ROC curves of serum Ang-2 and Ang-1 levels in DHFC samples compared to all other samples.
Changes of serum Ang-1, Ang-2, and Ang-2/Ang-1 levels by patient category and day of fever.
Changes in Ang-1, Ang-2, and Ang-2/Ang-1 levels in different disease stages of DF and DHF patients by day of fever are shown in Fig. 3. In both DFA and DHFA, the mean serum Ang-2 levels were significantly increased on day 5 of fever compared to day 3 of fever (P < 0.050). There was a significant increase in serum Ang-2/Ang-1 levels in DHFA on day 5 of fever (DHFA-5) compared to DHFA-3 (P = 0.001). The highest Ang-2 levels were observed in DHFC-5 and were significantly higher than for DHFA-5 (P = 0.037) and DHFC-7 (P = 0.007). The same trend was observed with serum Ang-2/Ang-1 levels. Serum Ang-1 levels in DFA, DHFA, and DHFC samples by day of fever were similar. A declining trend of Ang-1 levels was observed in both DF (DFA) and DHF (DHFA) samples on admission with progression in terms of days of fever. The mean values of serum Ang-2/Ang-1, Ang-2, and Ang-1 are given in Table 3.
FIG 3.
Changes in serum Ang-1 (A), Ang-2 (B), and Ang-2/Ang-1 (C) by day of fever in different dengue samples. Dengue fever patients on admission (DFA, gray circle), dengue hemorrhagic fever patients on admission (DHFA, black diamond), and dengue hemorrhagic fever patients at critical stage (DHFC, black square). Significance is expressed as P values determined using an independent-sample t test. A paired sample t test was used for repeated measures.
TABLE 3.
Changes in serum Ang-2/Ang-1, Ang-2, and Ang-1 levels in samples of dengue patients in different stages of the disease by day of fever and healthy controls
| Patient categorya | No. of patients | Mean ± SD |
||
|---|---|---|---|---|
| Ang-2/Ang-1 | Ang-2 (ng/ml) | Ang-1 (ng/ml) | ||
| DFA | 40 | 0.49 ± 0.55 | 5.63 ± 3.96 | 19.87 ± 11.83 |
| DFA-3 | 17 | 0.45 ± 0.45 | 4.02 ± 2.86 | 21.50 ± 12.27 |
| DFA-4 | 11 | 0.72 ± 0.82 | 6.20 ± 5.32 | 16.98 ± 9.54 |
| DFA-5 | 12 | 1.05 ± 0.98 | 7.89 ± 4.71 | 17.25 ± 9.74 |
| DFD | 20 | 0.60 ± 0.52 | 5.82 ± 2.87 | 20.75 ± 19.47 |
| DHFA | 40 | 1.19 ± 2.39 | 6.56 ± 3.27 | 20.57 ± 14.30 |
| DHFA-3 | 20 | 0.69 ± 1.26 | 5.32 ± 3.44 | 26.23 ± 15.23 |
| DHFA-4 | 11 | 0.52 ± 0.57 | 7.06 ± 3.02 | 18.35 ± 10.25 |
| DHFA-5 | 9 | 2.72 ± 3.87 | 8.20 ± 2.87 | 15.36 ± 16.25 |
| DHFC | 36 | 3.88 ± 3.25 | 10.68 ± 4.49 | 9.41 ± 6.66 |
| DHFC-5 | 20 | 2.63 ± 3.96 | 11.99 ± 3.96 | 10.18 ± 8.98 |
| DHFC-6 | 9 | 2.19 ± 1.99 | 10.75 ± 5.46 | 8.86 ± 7.60 |
| DHFC-7 | 7 | 0.73 ± 0.39 | 6.92 ± 3.00 | 11.45 ± 6.04 |
| DHFD | 20 | 1.01 ± 1.46 | 7.33 ± 3.02 | 22.94 ± 15.28 |
| HC | 25 | 0.18 ± 0.14 | 4.22 ± 1.93 | 32.82 ± 14.30 |
DFA, dengue fever on admission; DFA-3, dengue fever patients admitted on day 3 of fever; DFD, dengue fever patients on discharge; DHFA, dengue hemorrhagic fever patients on admission; DHFC, dengue hemorrhagic fever at critical stage; DHFD, dengue hemorrhagic fever on discharge; HC, healthy controls.
Association between serum Ang-1 and Ang-2 levels and clinical parameters.
The associations between clinical parameters and serum Ang-1 and Ang-2 levels in the different samples were assessed. Significant associations were observed between Ang-1 and Ang-2 levels and clinical parameters in DHF patients on admission and at the critical stage (Table 4). Serum Ang-1 levels were positively correlated with platelet count in DHFA samples (r = 0.653, P < 0.001). Serum Ang-1 levels were positively correlated with pulse pressure in DHF patients at the critical stage (r = 0.636, P = 0.001).
TABLE 4.
Significant associations between Ang-1 and Ang-2 levels and clinical parameters in DHF patients on admission and at the critical stage
| Category | Test parameter | Clinical parameter | Spearman rank correlation coefficient | Significance (P) |
|---|---|---|---|---|
| DHFA | Ang-1 | Platelet count (103/mm3) | 0.653 | <0.001 |
| Eosinophil (%) | 0.454 | 0.023 | ||
| Basophil (%) | 0.410 | 0.046 | ||
| Ang-2 | Leukocyte (%) | 0.442 | 0.028 | |
| Basophil (%) | 0.445 | 0.030 | ||
| DHFC | Ang-1 | Pulse rate (beats/min) | 0.560 | 0.030 |
| Pulse pressure (mm Hg) | 0.636 | 0.001 | ||
| AST (U/liter) | 0.542 | 0.030 | ||
| Ang-2 | Age (yr) | 0.477 | 0.046 | |
| AST (U/liter) | 0.499 | 0.049 |
DISCUSSION
In this study, we have demonstrated significantly higher levels of serum Ang-2 and decreased levels of serum Ang-1 in DHF samples at the critical stage of the disease. Since plasma leakage is the main characteristic feature of DHF at the critical stage, our results show that serum Ang-1 levels are low and that serum Ang-2 levels are high during this phase of vascular permeability. However, no significant differences were observed in serum Ang-1 and Ang-2 levels between DHFA and DFA samples, suggesting that vascular involvement of severe dengue infection is less likely during the early stage of the disease.
Among DHF patients, serum Ang-1 levels were high in DHFA samples (on admission) and subsequently decreased at the critical stage of the disease (DHFC samples); there was then an increase on discharge. Conversely, there was a clear rise of serum Ang-2 from admission to the critical stage; the highest serum Ang-2 levels were observed in DHFC samples at the critical stage, and then the levels declined on discharge. The overall trend of declining serum Ang-1 and rising Ang-2 levels from admission to the critical stage suggests that the levels of these endothelial stabilization markers may be associated with disease progression (admission to critical stage) in DHF patients. The restoration of Ang-1 and Ang-2 levels to baseline levels at discharge similar to healthy controls provides further evidence of their role in the plasma leakage phase.
High serum Ang-2 levels in DHFC samples may be due to release of more Ang-2 from WPBs. Studies have shown that elevated proinflammatory cytokines such as TNF-α (28), IL-1, and IL-6, etc. (29), in severe dengue patients could act on WPB releasing more Ang-2 from endothelial cells (9).
Even though serum Ang-2 was high in DHFC category samples, the levels changed depending on the day of fever in the progression to the critical stage. The highest serum Ang-2 levels were observed in DHFC-5 with a decline in DHFC-7 levels. This analysis shows the dynamics of the levels of angiopoietins even in the same category of patients with the day of fever further highlighting the importance of the day of fever for in vitro studies. Serum Ang-2 levels of DFA and DHFA samples increased with day of fever, even though endothelial dysfunction was not evident in DF patients.
It has been reported that serum Ang-1 is also released by platelets and that there is a positive association between thrombocytopenia and decreased levels of serum Ang-1 in severe dengue infection (14, 30), suggesting that low platelet levels may also contribute to the reduction of serum Ang-1 enhancing more Ang-2 binding to the Tie-2 receptor due to the competition of binding between Ang-1 and Ang-2 to the Tie-2 receptor.
A strong association was observed between serum Ang-1 levels and pulse pressure in DHF patients at the critical stage. Pulse pressure is defined as the difference between systolic and diastolic blood pressures. Narrowing pulse pressure is evident in DHF patients at the leaking stage and occurs due to the reduction of the pressure build up in the blood vessels due to the decrease in blood volume, i.e., plasma due to high vascular permeability (31). Serum Ang-1 plays an important role in maintaining the integrity of blood vessels by keeping the cell-cell junction closer. This association between DHFC samples and pulse pressure is probably due to plasma leakage as a result of the reduction of Ang-1 levels, probably also due to a low platelet count, a finding reported for the first time in severe dengue infection.
All the associations between serum Ang-1 and Ang-2 and clinical parameters were observed in the different disease stages of DHF patients. In this study, serum Ang-2 levels in DHF patients at the critical stage were positively associated with age (r = 0.477, P = 0.046). It is also important to measure biochemical parameters in HC and to determine the association between these endothelial markers and biochemical parameters in HC which would provide a baseline for changes in these parameters.
This preliminary study was done to investigate the role of serum Ang-1 and Ang-2 levels in severe dengue infection. We found that the serum Ang-2/Ang-1 ratio may be a potential biomarker for identifying and confirming endothelial dysfunction at the critical stage of the disease; the ratio has a sensitivity of 82.6% and a specificity of 80.4% using a cutoff value of 1.02. Even though we performed these assays in duplicate and in different panels, it is necessary to conduct these assays in different laboratory settings and with samples from different locations of the world to assess reproducibility and also to validate the results with unknown samples before implementing these tests routinely in the hospital setting.
In order to establish the Ang-2/Ang-1 ratio as a marker to predict severe dengue, it is important to assay the Ang-2/Ang-1 ratio before patients enter the critical stage or at an early point in the critical stage (e.g., within 6 h of the onset of critical stage). Therefore, DHFC samples could be further subcategorized based on different time frames of the critical window (e.g., 12, 24, 36, and 48 h after the critical stage of DHF), and assessments made just before the critical stage would provide a more precise idea on using these markers for the early identification of DHF.
Our results suggest that high Ang-2 levels and low Ang-1 levels may trigger endothelial dysfunction, resulting in severe dengue infection progressing from the DF to DHF critical stage through the binding of Ang-1 and Ang-2 to the Tie-2 receptor cell that signals endothelial stabilization pathways. The potential use of anti-Ang-2 blocking antibodies to arrest the destabilization of the endothelium would protect endothelial stability. This study suggests that DHF patients who enter the critical stage on day 5 of fever are the most appropriate study group to perform in vitro experiments to determine the role of Ang-2 in endothelial dysfunction. Further studies should be done to determine the kinetics of these angiopoietins in DHF patients on each day of fever during the hospital stay. Studying the genetic background and expression of Ang-1, Ang-2, and the Tie-2 receptor should be addressed in future studies with a view to finding therapeutic options for the management of severe dengue infections.
Conclusion.
A significantly higher serum Ang-2/Ang-1 ratio was observed in DHFC samples than in all other sample categories. The Ang-2/Ang-1 ratio may be used as a potential biomarker for severe dengue infection; a cutoff value of 1.01 has a sensitivity of 83.2% and a specificity of 81.2%. Ang-1 and Ang-2 appear to regulate endothelial stabilization pathways.
ACKNOWLEDGMENTS
We thank the consultants, nurses, and other staff of wards 9 and 12 of the Colombo North Teaching Hospital, Ragama, Sri Lanka, for the recruitment of patients and sampling. We thank the staff of the Institute of Biochemistry, Molecular Biology and Biotechnology, University of Colombo, Colombo, Sri Lanka, and the staff of the Faculty of Science, University of Colombo, Colombo, Sri Lanka, for logistic assistance. We especially thank the individuals who donated their blood voluntarily.
This study was financially supported by the National Science Foundation of Sri Lanka (grant RG/2014/HS/04).
REFERENCES
- 1.Cohen SN, Halstead SB. 1966. Shock associated with dengue infection: I. Clinical and physiologic manifestations of dengue hemorrhagic fever in Thailand, 1964. J Pediatr 68:448–456. doi: 10.1016/s0022-3476(66)80249-4. [DOI] [PubMed] [Google Scholar]
- 2.Halstead SB. 2002. Dengue. Curr Opin Infect Dis 15:471–476. doi: 10.1097/00001432-200210000-00003. [DOI] [PubMed] [Google Scholar]
- 3.Nimmannitya S. 1987. Clinical spectrum and management of dengue haemorrhagic fever. Southeast Asian J Trop Med Public Health 18:392–397. [PubMed] [Google Scholar]
- 4.Chaturvedi UC, Agarwal R, Elbishbishi EA, Mustafa AS. 2000. Cytokine cascade in dengue hemorrhagic fever: implications for pathogenesis. FEMS Immunol Med Microbiol 28:183–188. doi: 10.1111/j.1574-695X.2000.tb01474.x. [DOI] [PubMed] [Google Scholar]
- 5.Rothman AL, Ennis FA. 1999. Immunopathogenesis of dengue hemorrhagic fever. Virology 257:1–6. doi: 10.1006/viro.1999.9656. [DOI] [PubMed] [Google Scholar]
- 6.van Meurs M, Kümpers P, Ligtenberg JJ, Meertens JH, Molema G, Zijlstra JG. 2009. Bench-to-bedside review: angiopoietin signaling in critical illness: a future target? Crit Care 13:207. doi: 10.1186/cc7153. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Hirschberg R, Wang S, Mitu GM. 2008. Functional symbiosis between endothelium and epithelial cells in glomeruli. Cell Tissue Res 331:485–493. doi: 10.1007/s00441-007-0526-z. [DOI] [PubMed] [Google Scholar]
- 8.Fiedler U, Scharpfenecker M, Koidl S, Hegen A, Grunow V, Schmidt JM, Kriz W, Thurston G, Augustin HG. 2004. The Tie-2 ligand angiopoietin-2 is stored in and rapidly released upon stimulation from endothelial cell Weibel-Palade bodies. Blood 103:4150–4156. doi: 10.1182/blood-2003-10-3685. [DOI] [PubMed] [Google Scholar]
- 9.Scharpfenecker M, Fiedler U, Reiss Y, Augustin H. 2005. The Tie-2 ligand angiopoietin-2 destabilizes quiescent endothelium through an internal autocrine loop mechanism. J Cell Sci 118:771–780. doi: 10.1242/jcs.01653. [DOI] [PubMed] [Google Scholar]
- 10.van de Weg CAM, Pannuti CS, van den Ham H-J, de Araújo ESA, Boas LSV, Felix AC, Carvalho KI, Levi JE, Romano CM, Centrone CC, Rodrigues CLDL, Luna E, van Gorp ECM, Osterhaus ADME, Kallas EG, Martina BEE. 2014. Serum angiopoietin-2 and soluble VEGF receptor 2 are surrogate markers for plasma leakage in patients with acute dengue virus infection. J Clin Virol 60:328–335. doi: 10.1016/j.jcv.2014.05.001. [DOI] [PubMed] [Google Scholar]
- 11.Li P, He Q, Luo C, Qian L. 2015. Diagnostic and prognostic potential of serum angiopoietin-2 expression in human breast cancer. Int J Clin Exp Pathol 8:660–664. [PMC free article] [PubMed] [Google Scholar]
- 12.Fawzy A, Gaafar R, Kasem F, Ali SS, Elshafei M, Eldeib M. 2012. Importance of serum levels of angiopoietin-2 and survivin biomarkers in non-small cell lung cancer. J Egypt Natl Cancer Inst 24:41–45. doi: 10.1016/j.jnci.2011.12.006. [DOI] [PubMed] [Google Scholar]
- 13.Lovegrove FE, Tangpukdee N, Opoka RO, Lafferty EI, Rajwans N, Hawkes M, Krudsood S, Looareesuwan S, John CC, Liles WC, Kain KC. 2009. Serum angiopoietin-1 and -2 levels discriminate cerebral malaria from uncomplicated malaria and predict clinical outcome in African children. PLoS One 4:e4912. doi: 10.1371/journal.pone.0004912. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Michels M, van der Ven AJAM, Djamiatun K, Fijnheer R, de Groot PG, Griffioen AW, Sebastian S, Faradz SMH, de Mast Q. 2012. Imbalance of angiopoietin-1 and angiopoetin-2 in severe dengue and relationship with thrombocytopenia, endothelial activation, and vascular stability. Am J Trop Med Hyg 87:943–946. doi: 10.4269/ajtmh.2012.12-0020. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Rampengan NH, Daud D, Warouw S, Ganda IJ. 2015. Serum angiopoietin-2 as marker of plasma leakage in dengue viral infection. Am J Clin Exp Med 3:39–43. doi: 10.11648/j.ajcem.20150301.15. [DOI] [Google Scholar]
- 16.Glasner DR, Puerta-Guardo H, Beatty PR, Harris E. 2018. The good, the bad, and the shocking: the multiple roles of dengue virus nonstructural protein 1 in protection and pathogenesis. Annu Rev Virol 5:227–253. doi: 10.1146/annurev-virology-101416-041848. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Puerta-Guardo H, Glasner DR, Espinosa DA, Biering SB, Patana M, Ratnasiri K, Wang C, Beatty PR, Harris E. 2019. Flavivirus NS1 triggers tissue-specific vascular endothelial dysfunction reflecting disease tropism. Cell Rep 26:1598–1613. doi: 10.1016/j.celrep.2019.01.036. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Chuang Y-C, Lei H-Y, Lin Y-S, Liu H-S, Wu H-L, Yeh T-M. 2011. Dengue virus-induced autoantibodies bind to plasminogen and enhance its activation. J Immunol 187:6483–6490. doi: 10.4049/jimmunol.1102218. [DOI] [PubMed] [Google Scholar]
- 19.Lin C-F, Lei H-Y, Shiau A-L, Liu H-S, Yeh T-M, Chen S-H, Liu C-C, Chiu S-C, Lin Y-S. 2002. Endothelial cell apoptosis induced by antibodies against dengue virus nonstructural protein 1 via production of nitric oxide. J Immunol 169:657–664. doi: 10.4049/jimmunol.169.2.657. [DOI] [PubMed] [Google Scholar]
- 20.Chuang YC, Lin YS, Liu HS, Wang JR, Yeh TM. 2013. Antibodies against thrombin in dengue patients contain both anti-thrombotic and pro-fibrinolytic activities. Thromb Haemost 110:358–365. doi: 10.1160/TH13-02-0149. [DOI] [PubMed] [Google Scholar]
- 21.Beatty PR, Puerta-Guardo H, Killingbeck SS, Glasner DR, Hopkins K, Harris E. 2015. Dengue virus NS1 triggers endothelial permeability and vascular leak that is prevented by NS1 vaccination. Sci Transl Med 7:304ra141. doi: 10.1126/scitranslmed.aaa3787. [DOI] [PubMed] [Google Scholar]
- 22.Jayathilaka D, Gomes L, Jeewandara C, Jayarathna GSB, Herath D, Perera PA, Fernando S, Wijewickrama A, Hardman CS, Ogg GS, Malavige GN. 2018. Role of NS1 antibodies in the pathogenesis of acute secondary dengue infection. Nat Commun 9:5242. doi: 10.1038/s41467-018-07667-z. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Modhiran N, Watterson D, Muller DA, Panetta AK, Sester DP, Liu L, Hume DA, Stacey KJ, Young PR. 2015. Dengue virus NS1 protein activates cells via Toll-like receptor 4 and disrupts endothelial cell monolayer integrity. Sci Transl Med 7:304ra142. doi: 10.1126/scitranslmed.aaa3863. [DOI] [PubMed] [Google Scholar]
- 24.Ong SP, Ng ML, Chu J. 2013. Differential regulation of angiopoietin 1 and angiopoietin 2 during dengue virus infection of human umbilical vein endothelial cells: implications for endothelial hyperpermeability. Med Microbiol Immunol 202:437–452. doi: 10.1007/s00430-013-0310-5. [DOI] [PubMed] [Google Scholar]
- 25.Phanthanawiboon S, Limkittikul K, Sakai Y, Takakura N, Saijo M, Kurosu T. 2016. Acute systemic infection with dengue virus leads to vascular leakage and death through tumor necrosis factor-α and Tie2/angiopoietin signaling in mice lacking type I and II interferon receptors. PLoS One 11:e0148564. doi: 10.1371/journal.pone.0148564. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Singh S, Anupriya MG, Modak A, Sreekumar E. 2018. Dengue virus or NS1 protein induces trans-endothelial cell permeability associated with VE-cadherin and RhoA phosphorylation in HMEC-1 cells preventable by angiopoietin-1. J Gen Virol 99:1658–1670. doi: 10.1099/jgv.0.001163. [DOI] [PubMed] [Google Scholar]
- 27.World Health Organization. 2011. Comprehensive guidelines for prevention and control of dengue and dengue haemorrhagic fever. WHO Regional Publication SEARO. World Health Organization, Geneva, Switzerland. [Google Scholar]
- 28.Yen Y-T, Chen H-C, Lin Y-D, Shieh C-C, Wu-Hsieh BA, Wu-Hsieh BA. 2008. Enhancement by tumor necrosis factor alpha of dengue virus-induced endothelial cell production of reactive nitrogen and oxygen species is key to hemorrhage development. J Virol 82:12312–12324. doi: 10.1128/JVI.00968-08. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Hober D, Poli L, Roblin B, Gestas P, Chungue E, Granic G, Imbert P, Pecarere JL, Vergez-Pascal R, Wattre P. 1993. Serum levels of tumor necrosis factor-alpha (TNF-α), interleukin-6 (IL-6), and interleukin-1β (IL-1β) in dengue-infected patients. Am J Trop Med Hyg 48:324–331. doi: 10.4269/ajtmh.1993.48.324. [DOI] [PubMed] [Google Scholar]
- 30.Li JJ, Huang YQ, Basch R, Karpatkin S. 2001. Thrombin induces the release of angiopoietin-1 from platelets. Thromb Haemost 85:204–206. doi: 10.1055/s-0037-1615677. [DOI] [PubMed] [Google Scholar]
- 31.Halstead SB. 1980. Dengue haemorrhagic fever: a public health problem and a field for research. Bull World Health Organ 58:1–21. [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Data Availability Statement
All relevant data are available in the article.