Limits...
Time since onset of disease and individual clinical markers associate with transcriptional changes in uncomplicated dengue.

van de Weg CA, van den Ham HJ, Bijl MA, Anfasa F, Zaaraoui-Boutahar F, Dewi BE, Nainggolan L, van IJcken WF, Osterhaus AD, Martina BE, van Gorp EC, Andeweg AC - PLoS Negl Trop Dis (2015)

Bottom Line: Sequential whole blood samples from DENV infected patients in Jakarta were profiled using affymetrix microarrays, which were analysed using principal component analysis, limma, gene set analysis, and weighted gene co-expression network analysis.Clinical diagnosis (according to the WHO classification) does not associate with differential gene expression.Overall, we see a shift in the transcriptome from immunity and inflammation to repair and recovery during the course of a DENV infection.

View Article: PubMed Central - PubMed

Affiliation: Department of Viroscience, Erasmus Medical Center, Rotterdam, The Netherlands.

ABSTRACT

Background: Dengue virus (DENV) infection causes viral haemorrhagic fever that is characterized by extensive activation of the immune system. The aim of this study is to investigate the kinetics of the transcriptome signature changes during the course of disease and the association of genes in these signatures with clinical parameters.

Methodology/principle findings: Sequential whole blood samples from DENV infected patients in Jakarta were profiled using affymetrix microarrays, which were analysed using principal component analysis, limma, gene set analysis, and weighted gene co-expression network analysis. We show that time since onset of disease, but not diagnosis, has a large impact on the blood transcriptome of patients with non-severe dengue. Clinical diagnosis (according to the WHO classification) does not associate with differential gene expression. Network analysis however, indicated that the clinical markers platelet count, fibrinogen, albumin, IV fluid distributed per day and liver enzymes SGOT and SGPT strongly correlate with gene modules that are enriched for genes involved in the immune response. Overall, we see a shift in the transcriptome from immunity and inflammation to repair and recovery during the course of a DENV infection.

Conclusions/significance: Time since onset of disease associates with the shift in transcriptome signatures from immunity and inflammation to cell cycle and repair mechanisms in patients with non-severe dengue. The strong association of time with blood transcriptome changes hampers both the discovery as well as the potential application of biomarkers in dengue. However, we identified gene expression modules that associate with key clinical parameters of dengue that reflect the systemic activity of disease during the course of infection. The expression level of these gene modules may support earlier detection of disease progression as well as clinical management of dengue.

Show MeSH

Related in: MedlinePlus

Overlap of differential gene expression with signatures identified in other studies.(a) indicates the signatures generated in this study, (d) those of the other dengue studies. (b-c) and (e-f) indicate the overlap with the signatures of day 0 and day 4, respectively.
© Copyright Policy
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC4358925&req=5

pntd.0003522.g003: Overlap of differential gene expression with signatures identified in other studies.(a) indicates the signatures generated in this study, (d) those of the other dengue studies. (b-c) and (e-f) indicate the overlap with the signatures of day 0 and day 4, respectively.

Mentions: Over the past few years, several studies have examined the transcriptional profile of dengue infections. Three cross-sectional studies (Tolfvenstam et al., Long et al. and Loke et al.) [6,7,18] were similar in the type of sample used (whole blood) and the included data on time since onset of symptoms, allowing these studies to be compared to results from our cohort (Fig. 3A, Table 2). Tolfvenstam et al. and Long et al. have a fairly large overlap in differentially expressed genes (Fig. 3D), presumably because both studies included patients early (<72 hours) after onset of disease. The signatures published by Loke et al. have little overlap with those of the other studies (1 and 5 genes only, Loke et al. DF and DHF signatures combined), most likely due to the fact that patients were included at a later time point after the onset of disease (3–6 days after onset). To compare our results to these studies, we compared the early and late general dengue signatures to those of the other studies. 48% of differentially expressed genes in the signature from Tolfvenstam et al. and 63% of Long et al. are also part of our day 0 dengue gene signature (collected <48 hours after onset of disease) (Fig. 3B). On the contrary, only 1% of the genes in the DF and 7% of the genes in the DHF signature in Loke et al. were similar to our day 0 signature (Fig. 3C). In contrast, our day 4 signature showed the greatest similarity with the signatures in Loke et al. (68% DF, 68% DHF; Fig. 3F), but much less so with those from Tolfvenstam et al. and Long et al. (18% and 35%, respectively; Fig. 3E). Our results therefore concur with all three studies and confirm that these signatures can occur within one cohort, but at different time points after onset of symptoms. In conclusion, time since the onset of symptoms accounts for most of the transcriptome differences between mRNA profiling studies in dengue patients.


Time since onset of disease and individual clinical markers associate with transcriptional changes in uncomplicated dengue.

van de Weg CA, van den Ham HJ, Bijl MA, Anfasa F, Zaaraoui-Boutahar F, Dewi BE, Nainggolan L, van IJcken WF, Osterhaus AD, Martina BE, van Gorp EC, Andeweg AC - PLoS Negl Trop Dis (2015)

Overlap of differential gene expression with signatures identified in other studies.(a) indicates the signatures generated in this study, (d) those of the other dengue studies. (b-c) and (e-f) indicate the overlap with the signatures of day 0 and day 4, respectively.
© Copyright Policy
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC4358925&req=5

pntd.0003522.g003: Overlap of differential gene expression with signatures identified in other studies.(a) indicates the signatures generated in this study, (d) those of the other dengue studies. (b-c) and (e-f) indicate the overlap with the signatures of day 0 and day 4, respectively.
Mentions: Over the past few years, several studies have examined the transcriptional profile of dengue infections. Three cross-sectional studies (Tolfvenstam et al., Long et al. and Loke et al.) [6,7,18] were similar in the type of sample used (whole blood) and the included data on time since onset of symptoms, allowing these studies to be compared to results from our cohort (Fig. 3A, Table 2). Tolfvenstam et al. and Long et al. have a fairly large overlap in differentially expressed genes (Fig. 3D), presumably because both studies included patients early (<72 hours) after onset of disease. The signatures published by Loke et al. have little overlap with those of the other studies (1 and 5 genes only, Loke et al. DF and DHF signatures combined), most likely due to the fact that patients were included at a later time point after the onset of disease (3–6 days after onset). To compare our results to these studies, we compared the early and late general dengue signatures to those of the other studies. 48% of differentially expressed genes in the signature from Tolfvenstam et al. and 63% of Long et al. are also part of our day 0 dengue gene signature (collected <48 hours after onset of disease) (Fig. 3B). On the contrary, only 1% of the genes in the DF and 7% of the genes in the DHF signature in Loke et al. were similar to our day 0 signature (Fig. 3C). In contrast, our day 4 signature showed the greatest similarity with the signatures in Loke et al. (68% DF, 68% DHF; Fig. 3F), but much less so with those from Tolfvenstam et al. and Long et al. (18% and 35%, respectively; Fig. 3E). Our results therefore concur with all three studies and confirm that these signatures can occur within one cohort, but at different time points after onset of symptoms. In conclusion, time since the onset of symptoms accounts for most of the transcriptome differences between mRNA profiling studies in dengue patients.

Bottom Line: Sequential whole blood samples from DENV infected patients in Jakarta were profiled using affymetrix microarrays, which were analysed using principal component analysis, limma, gene set analysis, and weighted gene co-expression network analysis.Clinical diagnosis (according to the WHO classification) does not associate with differential gene expression.Overall, we see a shift in the transcriptome from immunity and inflammation to repair and recovery during the course of a DENV infection.

View Article: PubMed Central - PubMed

Affiliation: Department of Viroscience, Erasmus Medical Center, Rotterdam, The Netherlands.

ABSTRACT

Background: Dengue virus (DENV) infection causes viral haemorrhagic fever that is characterized by extensive activation of the immune system. The aim of this study is to investigate the kinetics of the transcriptome signature changes during the course of disease and the association of genes in these signatures with clinical parameters.

Methodology/principle findings: Sequential whole blood samples from DENV infected patients in Jakarta were profiled using affymetrix microarrays, which were analysed using principal component analysis, limma, gene set analysis, and weighted gene co-expression network analysis. We show that time since onset of disease, but not diagnosis, has a large impact on the blood transcriptome of patients with non-severe dengue. Clinical diagnosis (according to the WHO classification) does not associate with differential gene expression. Network analysis however, indicated that the clinical markers platelet count, fibrinogen, albumin, IV fluid distributed per day and liver enzymes SGOT and SGPT strongly correlate with gene modules that are enriched for genes involved in the immune response. Overall, we see a shift in the transcriptome from immunity and inflammation to repair and recovery during the course of a DENV infection.

Conclusions/significance: Time since onset of disease associates with the shift in transcriptome signatures from immunity and inflammation to cell cycle and repair mechanisms in patients with non-severe dengue. The strong association of time with blood transcriptome changes hampers both the discovery as well as the potential application of biomarkers in dengue. However, we identified gene expression modules that associate with key clinical parameters of dengue that reflect the systemic activity of disease during the course of infection. The expression level of these gene modules may support earlier detection of disease progression as well as clinical management of dengue.

Show MeSH
Related in: MedlinePlus