A systematic review of observational studies on oxidative/nitrosative stress involvement in dengue pathogenesis.

OBJECTIVE
Our objective was to systematically review the published observational research related to the role of oxidative-nitrosative stress in pathogenesis of dengue.


METHODS
We searched electronic databases (PubMed, EMBASE, The COCHRANE library, ScienceDirect, Scopus, SciELO, LILACS via Virtual Health Library, Google Scholar) using the term: dengue, dengue virus, severe dengue, oxidative stress, nitrosative stress, antioxidants, oxidants, free radicals, oxidized lipid products, lipid peroxides, nitric oxide, and nitric oxide synthase. Articles were selected for review by title and abstract excluding letter, review, in vivo and in vitro studies, and duplicates studies. Selected articles were reviewed for study design, original purposes, sample size, main outcomes, methods, and oxidative-nitrosative stress markers values.


RESULTS
In total, 4,331 non-duplicates articles were identified from electronic databases searches, of which 16 were eligible for full text searching. Data from the observational studies originate from Asian countries (50%; 8/16), South American countries (31.2%; 5/16), and Central America and the Caribbean countries (18.8%; 3/16). Case-control study was the type of design most common in researches reviewed. The 1997 World Health Organization (WHO) dengue case classification criteria were used in all studies included in this review.


CONCLUSIONS
Based on published data found in peer-reviewed literature, oxidative and nitrosative stress are demonstrated by changes in plasma levels of nitric oxide, antioxidants, lipid peroxidation and protein oxidation markers in patients with dengue infection. Additionally, elevated serum protein carbonyls and malondialdehyde levels appear to be associated with dengue disease severity.


Introduction
Dengue is a systemic viral infection with a significant socioeconomic and disease burden in many tropical and subtropical regions all over the world. Using cartographic approaches, one recent estimate indicates 390 million dengue infections per year (95% credible interval 284-528 million), of which 96 million (95% credible interval 67-136 million) manifest clinically (with any severity of disease) 1 .
The etiologic agent of this tropical disease is dengue virus (DENV), member of the family Flaviviridae, with four different antigenic serotypes (DENV-1 to -4). The DENV genome of plus strand RNA encodes three structural proteins (capsid, prM and envelope) and seven non-structural proteins (NS1, NS2a, NS2b, NS3, NS4a, NS4b y NS5) 2,3 According to the PAHO, in 2013, for America reported 2,386,836 cases of dengue; of which 37,898 (1.6%) corresponded to severe dengue, with case fatality rate 0.06. Colombia, Andean country, reported 127,219 cases of dengue (55% of cases reported in Andean Sub-region); of which 3,377 (81.1% of cases reported in Andean Sub-region) corresponded to severe dengue 4 .
Clinical and epidemiological observations showed that dengue disease severity may vary according to age, ethnicity, genetic factors, immune status and underlying disease. It may also depend on the co-circulation of DENV serotypes and reinfection by different DENV serotypes [5][6][7][8][9] . In this respect, it has been proposed the involvement of DENV infection-derived oxidative stress on the severity of dengue. This is based on their ability to trigger the release of proinflammatory cytokines, including TNF-alpha, participating in collective action in the immunopathogenesis of dengue disease 10 .
By definition, oxidative stress is an imbalance between prooxidants and antioxidants in favour of the pro-oxidants 11,12 . Instead, nitrosative stress is defined as an indiscriminate nitrosilation of biological molecules 13 . In the absence of an appropriate compensatory response from endogenous antioxidant defense system, the activation of several stress-sensitive intracellular signaling pathways have been reported. This activation involves the production of gene products that can lead to cell death and/or pathophysiological conditions [13][14][15][16] .
Recently, Olagnier et al., 27 have reported that nuclear factorerythroid 2-related factor 2/Nrf-2 mediated oxidative stress response, iNOS signaling and production of NOS and ROS pathways were stimulated by DENV-2 infection of human monocyte-derived dendritic cells/Mo-DC. Also, a statistically significance decrease in SOD-2 mRNA levels was observed during treatment with ROS scavenger diphenyleneiodonium-DPI. In addition, these authors reported that DENV-2 infection was associated with NADPH oxidase-generated ROS accumulation.
These lines of evidence suggest that oxidative/nitrosative stress can be related to production of pathogenesis-related protein, increased susceptibility of mice to DENV infection, hemorrhage development in mice, proinflammatory cytokines and transcriptional factor expression, and DENV replication in various cell cultures.
Considering the aforementioned scenario, we performed a systematically review of observational studies evaluating the role of oxidative-nitrosative stress in pathogenesis of dengue. This review is important because understanding the involvement of oxidative and nitrosative stress in dengue pathogenesis could have potential implications for prognosis and treatment.

Materials and Methods
Search strategy This review was guided by the standard PRISMA protocol (Preferred Reporting Items for Systematic Reviews and Metaanalysis) 28 and was registered on PROSPERO, an international database of prospectively registered systematic reviews in health and social care managed by Center for Review and Dissemination, University of York, on 12 November 2014; http://www.crd.york. ac.uk/PROSPERO (CRD42014014878). PubMed, EMBASE, The COCHRANE library, ScienceDirect, Scopus, SciELO, LILACS via Virtual Health Library, Google Scholar databases were searched for articles using a combination of descriptors to select the studies of interest.

Study selection
After finding previously published studies in the databases with the descriptors "dengue" OR "dengue virus" combined with "oxidative stress" OR "nitrosative stress" OR "antioxidants" OR "oxidants" OR "free radicals" OR "oxidized lipid products" OR "lipid peroxides" OR "superoxide dismutase" OR "thioredoxin reductase" OR "nitric oxide" OR "nitric oxide synthase", we performed an analysis on the inclusion/exclusion criteria. This electronic search strategy was supplemented by scanning the reference lists of all articles to identify additional studies that may have been missed during the initial search.
As inclusion criteria, we used observational studies that evaluated oxidative-nitrosative stress in dengue pathogenesis. Exclusion criteria were impossible extraction of data, no control group, dates from mosquito cells cultures, the editorial, comments, case reports, letter to the editor, conference abstract, review articles, proteomics, in vitro and in vivo studies. When multiple publications from the same study population were available, we included the most recent publication.
Additionally, two authors reviewed the studies independently in case of disagreement a third author was consulted. In this systematic review, there is no restriction regarding to language, publication period, patient age (children or adult), or study design.
Data extraction strategy Articles were selected for review by title and abstract. After reading full-text articles, we extracted data relating to study design, original purposes, sample size, main outcomes, methods, and oxidative stress markers values.
The systematic computerized literature search of published observational studies was carried out in June 2014.

Results
Identification of studies In total, 4,331 non-duplicates articles were identified from electronic databases searches, of which 16 were eligible for full text searching (Fig.1). Table 1 presents the citation, definitions and characteristics of each included studies, respectively.
Study design. Case-control study was the type of design most common in studies reviewed.
Follow-up time. Approximately 63% of observational studies reported a 7-day follow-up period. About a quarter of the included studies did not report follow-up of participants. Dengue case classification system. The 1997 World Health Organization (WHO) dengue case classification criteria were used in all studies included in this review.

Discussion
To our knowledge, no previous reviews on the involvement of oxidative-nitrosative stress in dengue pathogenesis have been performed. In the present systematic review, 16 articles concerning this subject were included.
The role of nitric oxide as strong immunomodulator and its association with viral infections, both in vivo and in vitro, have been clearly demonstrated 29,30 but there are incongruous results in some studies in DENV-infected patients. For example, although elevated serum levels of nitric oxide have been observed in sera of patients with dengue fever (DF) in comparison to age-matched healthy controls 25,31 or to patients with dengue hemorrhagic fever (DHF) 25 , Trairatvorakul et al. 32 , reported that serum nitric oxide levels were significantly higher in healthy children than in DF or DHF patients. These studies were characterized by the low case/ control ratio (≤1:0.5). Now, the two first reports are consistent with high total nitric oxide synthase activity and high L-arginine transport via system y + L reported in platelet obtained from DF and DHF patients compared with healthy adult controls 33,34 .
A study by Thai investigators reported that nitric oxide levels of primary and secondary DF, and primary DHF were not significantly different on the fever, defervescent or convalescent days in plasma of 37 DENV-infected children, while the level of serum nitric oxide was significantly lower on the fever day than on the defervescent day among children with secondary DHF. In addition, serum nitric oxide levels had an inverse correlation with the level of DENV viraemia in patients with secondary DHF 22 , but have not been linked to viral serotype infection 31 .
These observations are consistent with the findings of Yen et al. 17 , who reported a temporal coincidence between iNOS upregulation and free radical production with hemorrhage development in DENV-infected mice.
Blood nitrite and nitrate levels are frequently assessed as an index of systemic nitric oxide production. The spectrophotometric assay based on the Griess reagent is the most commonly used method to determine nitrite/nitrate concentration in biological matrixes 35 . In the (Table 2), the contrasting results may be explained by differences in nitrate reduction and deproteinization methods used or by differences in characteristics of populations studied.
As regard lipid peroxidation products as biomarkers for oxidative stress, several studies have found high serum levels of MDA and 4-hydroxyalkenals (4-HAE) in dengue patients in comparison to healthy adult controls 10,[36][37][38] . Furthermore, others reported high serum MDA concentrations in DHF and dengue shock syndrome (DSS) patients than in DF patients. They also reported a positive correlation between serum MDA and TNF-a levels in all DENV-infected patients 10 .    Table 1. Observational evidences of oxidative and nitrosative stress involvement in dengue pathogenesis Interestingly, a recent study reported that DENV-2-infected mice showed alterations in oxidative stress by increasing the level of malondialdehyde (MDA) 18 . These data together provide direct evidence for use end products of lipid peroxidation as prognostic biomarker in dengue disease.

Continued
Similar concentrations of MDA were reported using Satoh method 10,38 . Both patients and healthy controls, Klassen et al., 37 reported lower serum levels of MDA using Jentzch correction method. Instead, Gil et al. 36 , reported high serum MDA concentrations in dengue patients. This sensible difference could be explained by detection of MDA in combination with 4-HAE and/or sensitivity and reproducibility of the method 39 .
Although THs levels are considered a measure of overall oxidative damage 40 , there are no consistent pattern of THs in dengue disease. Gil et al. 36 , reported that serum THs concentrations were significantly decreased in dengue patients than in healthy adults. By contrast, the same group researchers recently reported elevated serum THs levels in dengue patients in comparison to adult controls 41 . In dengue cases, unlike controls, both studies reported similar serum THs concentrations.
During febrile stage of dengue infection, high concentrations of urinary F 2 -IsoP and plasma HETEs levels have been detected in comparison with convalescent stage and also with healthy adults 42   With respect to PCOs content, it has been used as an important biomarker of protein oxidative damage in oxidative stress-related diseases in humans 44 . Elevated serum PCOs levels have been reported in dengue patients in comparison to adult controls. Moreover, elevated serum PCOs concentrations were associated with dengue disease severity 38,45 . In both studies reported similar serum PCOs levels using method of Levine.
At all ages, elevated serum SOD levels and reduced GPx levels have been reported in dengue patients in comparison with controls 36,41,47 . By contrast, no differences were found in plasma antioxidants nutrients (except of retinol) concentrations between dengue patients and adult controls, but in this study only 9 patients and 12 healthy controls were evaluated 37 .
Results of one study have shown that serum GPx levels were associated with spontaneous bleeding events and with serum triglycerides levels in dengue patients. This finding suggests that the intensity of oxidative stress can influence the clinical presentation of dengue 48 . In this regard, a recent study reported that increased triglycerides levels were observed mainly in severe dengue patients 49 . Based on these data, it is likely that serum lipid profile in DENV-infected patients plays an important role in dengue severity.
Free radical production and alterations in antioxidants status occur during dengue disease 37 . The major antioxidant defense system appears to function by scavenging radical free radical. Nonenzymatic scavenger like GSH has been shown to be an effective protector against oxidative damage 46 . There are differences in the reports on the comparison of GSH concentrations in dengue patients and healthy controls. Gil et al. 36 , and Klassen et al. 37 , reported that serum GSH levels were not significantly different between patients and controls, whereas Gil et al. 41 , reported that serum GSH concentrations were significantly lower in dengue patients than in healthy adults controls. It is important to note two aspects: (i) these three researches were conducted in Central America and the Caribbean countries and (ii) the number of patients studied was small (n ≤22).
Despite the limitations of systematic reviews of observational studies, given the evidences presented here, oxidative-nitrosative stress is demonstrated by changes in plasma levels of nitric oxide, antioxidants, lipid peroxidation and protein oxidation markers in patients with dengue infection.
Additionally, elevated serum PCOs and MDA levels appear to be associated with dengue disease severity, expressed in terms of the 1997-WHO dengue case classification system. Accordingly, more data are needed to establish an association between oxidativenitrosative stress and dengue severity in the context of 2009-WHO dengue classification scheme.

Conclusions
Based on published data found in peer-reviewed literature, oxidative and nitrosative stress are demonstrated by changes in plasma levels of nitric oxide, antioxidants, lipid peroxidation and protein oxidation markers in patients with dengue infection. Additionally, elevated serum protein carbonyls and malondialdehyde levels appear to be associated with dengue disease severity.
There have been many studies that target the severity of dengue infections. However, our understanding is not complete yet. Better understanding of relation between oxidative-nitrosative stress and dengue pathogenesis will lead us to developing better prognostic strategies of this pathology.