Different phenotypes of non‐classical monocytes associated with systemic inflammation, endothelial alteration and hepatic compromise in patients with dengue

Abbreviations

ALT

alanine aminotransferase

APC

allophycocyanin

AST

aspartate aminotransferase

CBA

cytometric bead array

CRP

C‐reactive protein

DENV

dengue virus

DPBS

Dulbecco's phosphate‐buffered saline

ELISA

enzyme‐linked immunosorbent assay

FACS

fluorescence‐activated cell sorter

FITC

fluorescein isothiocyanate

IL

interleukin

MFI

Imean fluorescence intensity

MPs

microparticles

PBMCs

peripheral blood mononuclear cells

PCA

Principal component analysis

PE

phycoerythrin

PS

phosphatidylserine

SD

severe dengue

TLR

toll‐like receptor

TNF

tumor necrosis factor

WHO

World Health Organization

INTRODUCTION

Dengue is an acute viral infection¹ transmitted by the bite of the infected mosquitos belonging to the genus Aedes.² According to the World Health Organization (WHO), 50–100 million cases of infections by dengue virus (DENV) are registered every year across the world,², ³ of which approximately 60% are asymptomatic cases.² When the disease is manifested, 90–95% of the individuals develop similar symptoms, such as common flu‐type syndrome, which may be accompanied by warning signs. In most cases, dengue can be spontaneously resolved; the remaining 5–10% of cases can evolve into severe manifestations characterized by hemorrhages, organ damage, hypovolemic shock and death (severe dengue).², ³

The reasons for severe dengue in some patients are not completely understood. However, different components of the immune response seem to play a critical role in the pathogenesis of dengue. For instance, monocytes have been associated with both the control of viremia and dissemination of DENV infection. Monocytes produce high levels of cytokines, such as interleukin‐1β (IL‐1β), tumor necrosis factor‐α (TNF‐α) and IL‐6, and chemokines, such as CCL5, CCL4 and CCL3, in response to DENV infection,⁴, ⁵ thereby demonstrating their contribution to viral control and protective immune response against the infection.⁶ However, these cells act as the main target of DENV infection,⁷ which supports viral replication and damage to different tissues and organs, such as the alterations in vascular permeability, that can be observed during the infection.⁸, ⁹ Therefore, it has been considered that these cells play a central role in DENV immunopathology, particularly in severe disease manifestations. Monocytes are divided into three subsets based on the expression of the lipopolysaccharide membrane receptor (CD14) and low‐affinity Fcγ receptor (CD16) in the classical (CD14⁺⁺ CD16⁻), non‐classical (CD14⁺ CD16⁺⁺) and intermediate (CD14⁺⁺ CD16⁺) subsets; each one of these subsets has a differential function and response to the inflammatory processes. For example, it is important to highlight that non‐classical monocytes seem to be the main producers of inflammatory mediators in response to DENV infection,⁵, ¹⁰ as well as the major interacting monocytes with microvascular endothelium performing patrolling and surveillance functions in the basal conditions.¹¹, ¹²

Based on this information, we propose that the non‐classical monocytes possibly contribute considerably to the severe development of the disease. Hence, these cells must present the phenotypic changes in dengue patients related with different manifestations and clinical compromises observed during the infection. Given the above factors and with the purpose to enhance a greater understanding of the role that non‐classical monocytes play in DENV infection, monocyte subset counts were performed in dengue patients and in a healthy control group. Moreover, an extensive phenotypification of non‐classical monocytes was achieved based on the expression of markers associated with the activation, function and differentiation of mononuclear phagocytes. The results presented here indicate that dengue patients possess a smaller number of non‐classical monocytes; these cells produce TNF‐α and present a phenotype compatible with the cellular activation, migration and differentiation. These changes were associated with hepatic compromise, endothelial damage and high concentration of soluble mediators in the plasma of the patients. These cumulative observations suggest that non‐classical monocytes have a significant impact on the clinical outcomes in dengue patients.

Go to:

MATERIALS AND METHODS

Patients and controls

Patients diagnosed with DENV infection were recruited at the IPS Universitaria (Clínica León XIII, Medellín, Colombia) between October 2014 and February 2016. The sociodemographic data (gender and age), clinical and paraclinical characteristics, such as hematocrit level, platelet count, C‐reactive protein (CRP) level, transaminases aspartate aminotransferase (AST), alanine aminotransferase (ALT) and bilirubin information were collected from the patients’ medical histories or directly obtained from the patients.

A total of 147 dengue patient samples were clinically diagnosed according to the symptomatology and clinical presumptive diagnosis of the WHO.² However, only 63 patients, who were positive with the SD BIOLINE Dengue Duo Kit (Standard Diagnostics, Yongin, Corea del Sur), were finally included in the study. The positivity of the samples was confirmed by the Panbio Dengue IgM and IgG capture enzyme‐linked immunosorbent assay (ELISA) (Alere Waltham, MA).

The days of evolution of the disease corresponded to the day the symptoms were manifested, as reported by the patients, until the day sampling was performed. Patients were classified into two groups: dengue (D, n = 49) (without or with alarming signs) and severe dengue (SD, n = 14), (including severe bleeding, respiratory distress, severe organ involvement, shock and/or severe plasma leakage), according to the 2009 WHO criteria.² This classification was performed following the criteria of the physician‐in‐charge and an independent analysis of the information recorded in the medical histories that were performed and reviewed by the other two physicians (Berta N. Restrepo and Francisco J. Diaz).

We included 45 healthy individuals of the same gender and ages (±2 years) as the experimental groups, and from the same city as controls. Our study complies with the Declaration of Helsinki; the research protocol and informed consent form were approved by the Universidad de Antioquia's Medical Research Institute and IPS Universitaria Committees. All patients and controls provided a signed informed consent form.

Detection of viral NS1 antigen and IgG/IgM anti‐DENV antibodies

A rapid and immunochromatographic point‐of‐care testing to detect both DENV NS1 antigen and antibodies of DENV (IgG/IgM) was undertaken with the SD BIOLINE Dengue Duo Kit, according to the manufacturer's instructions.

The measurement of anti‐DENV IgG and IgM antibodies was performed with Panbio Dengue IgM Capture ELISA and Panbio Dengue IgG Capture ELISA Kits, according to the manufacturer's instructions. The results were quantified by an ‘index value’ defined as the ratio between the optical densities of the sample and the cut‐off value. The distinction between the primary and secondary infections was undertaken using the IgM/IgG ratio, obtained by dividing the index values of the two tests: a sample with an IgM/IgG ratio ≥ 1·4 was considered as the primary infection and the rest as the secondary infection. This ratio was not calculated when both the index values were < 1.¹³

Leukocyte counts and monocyte subset classification

The relative and absolute counts of total monocytes and the classification of monocytes into three subsets (classical: CD14⁺⁺ CD16⁻, non‐classical: CD14⁺ CD16⁺⁺ and intermediate: CD14⁺⁺ CD16⁺) were performed using flow cytometry as previously reported.¹⁴ A total peripheral blood anti‐coagulated with EDTA was stained with anti‐human HLA‐DR‐allophycocyanin (APC) ‐Cy7 (clone L243; BioLegend, San Diego, CA), CD14‐fluorescein isothiocyanate (FITC) MY4 (clone 322A‐1; Beckman Coulter, Hialeah, FL), CD16‐V450 (clone 3G8; Becton Dickinson Biosciences, San Jose, CA) and CD45‐phycoerythrin (PE) ‐Cy7 (clone HI30; BD Bioscience) antibodies for 20 min. Erythrocytes were lysed with 1‐ml BD FACS Lyzing Solution 1 × (BD Pharmingen, San Jose, CA) for 10 min. Two washes were performed using BD FACSFlow at 600 g for 5 min. Leukocyte (CD45⁺) counting was performed using 20 μl of Flow‐Count Fluorospheres (Beckman Coulter) with flow cytometry and manually with a hemocytometer and acetic acid.¹⁵ Cells were acquired immediately on the FACS Canto™ II flow cytometer with FACS DIVA software (BD Biosciences). The fluorescence minus one method was performed for each fluorochrome to determine the positive and negative events. Data were analyzed by performing the aggregates exclusion based on the FSC‐A and FSC‐H parameters using the flowjo 7.6.1 software (Tree Star, Ashland, OR). The relative and absolute counts of each subset were reported relative to the total monocytes.

Peripheral blood mononuclear cells and immunophenotyping of non‐classical monocytes

The characterization of non‐classical monocytes was performed by immunostaining of the peripheral blood mononuclear cells (PBMCs), as per our previous report.¹⁴ These cells were obtained from the venous blood treated with anti‐coagulant EDTA and previously diluted with Dulbecco's phosphate‐buffered saline (DPBS; Gibco‐BRL, Grand Island, NY) at the ratio of 1 : 1, after centrifugation on Histopaque‐1077 (Sigma‐Aldrich, St. Louis, MO) at 900 g for 30 min. PBMCs were washed twice with DPBS and washing buffer (DPBS plus 1% bovine serum albumin and 0·01% NaN₃; Sigma‐Aldrich) at 300 g for 10 min at room temperature. The cells were counted manually with a hemocytometer and cell viability was determined by the exclusion of trypan blue (≥ 96%) (Sigma‐Aldrich). Thereafter, the cells were incubated with a blocking buffer (DPBS plus 1% bovine serum albumin, 0·01% NaN₃ and 10% inactivated fetal calf serum) for 15 min. Next, the cells were incubated for 25 min with anti‐human CD14, CD16 and HLA‐DR antibodies to define the non‐classical monocyte subset (CD14⁺ CD16⁺⁺) and with different combinations of the next anti‐human CCR2‐Alexa Fluor 657 (clone 48607), CD32‐PE (clone 3D3), CD64‐APC (clone 10.1), CD13‐PECy7 (clone L138), CD33‐PE (clone WM53), CD36‐APC (clone CB38), CD86‐PECy5 (clone IT2.2), CD11a‐PECy7 (clone HI111), CD11b‐PE (clone D12), CD54‐PECy5 (clone HA58) and CD69‐PE (clone FN50) antibodies obtained from BD Biosciences (San Jose, CA); anti‐human CX3CR1‐PE (clone 2A9‐1), CCR5‐PECy7 (clone J418F1), CD15‐PECy5 (clone W6D3), CD18‐APC (clone TS1/18) antibodies and their respective isotype controls obtained from BioLegend. The cells were washed with washing buffer at 600 g for 5 min. For intracellular staining with anti‐human CD68‐PE (clone Y1/82A; BD Biosciences) and TNF‐α‐APC (clone MAb11, BioLegend) antibodies, PBMCs were fixed with the Fix/permeabilization solution of the FoxP3/Transcription Factor Staining Buffer (eBioscience, San Diego, CA) for 60 min. Next, the cells were washed with the 1 × permeabilization solution from the same kit at 1700 g for 5 min. The cells were incubated with the respective antibodies in the permeabilization solution for 25 min. Finally, the cells were washed twice and 50 000 monocytes were immediately acquired with the FACS Canto™ II flow cytometer with facs diva software. Analysis of the data was performed with flowjo 7.6.1 software, as described for monocyte subset classification. All the results were expressed as the mean fluorescence intensity (MFI) and percentage of the positive cells according to the isotype control.

The cells were stained with Annexin‐V‐Cy5 (BD Pharmingen) after the panel staining, in the presence of the Annexin binding buffer (BD Pharmingen) for 20 min, to assess the phosphatidylserine (PS) exposure and to verify the activation phenotype of non‐classical monocytes in some patients. This measurement was performed simultaneously with the TNF‐α ⁺, CD69⁺ and extracellular CD68⁺ (without permeabilization) cells. In addition, LIVE/DEAD probe (Invitrogen, Thermo Fisher Scientific, Waltham, MA) was used to evaluate the cell death.

Detection of the viral E protein

DENV infection was evaluated in PBMCs by the intracellular detection of the viral E protein using the murine monoclonal anti‐flavivirus 4G2 antibody (clone D1‐4G2‐4‐15; Millipore, Darmstadt, Germany, IgG2a) for 25 min and the respective secondary anti‐mouse IgG2a‐PE (Jackson ImmunoResearch, West Grove, PA) antibody for an additional 25 min at 4° in the dark.¹⁶ These incubations were completed after the panel staining (see above), in which none of the antibodies were IgG2. For intracellular staining, PBMCs were fixed and permeabilized as discussed previously. Finally, the cells were washed twice and 50 000 monocytes were immediately acquired on the FACS Canto™ II flow cytometer with facs diva software. PBMCs stained with the panel plus the secondary antibody without the D1‐4G2‐4‐15 antibody were performed for all cases and considered as the controls to define the frequency of the positive cells to E protein using the Overtone subtraction algorithm of the flowjo 7.6.1 software.

Isolation and characterization of microparticles

Microparticles (MPs) were isolated from the venous blood treated with anti‐coagulant citrate and then characterized using flow cytometry, as per our previous report.¹⁴ Briefly, MPs were incubated with anti‐human HLA‐DR and CD105‐APC antibodies (clone 43A3; BioLegend) at the concentration of 350–650 MPs/μl for 25 min at 4° in the dark. The cells were washed with filtered BD FACS Flow at 16 900 g for 60 min at 4°. MPs were immediately acquired in the FACS Canto™ II flow cytometer with facs diva software. Analysis of the data was performed with flowjo 7.6.1 software, and the results were expressed as frequencies.

Cytokines measurement

Cytokines were evaluated using the Cytometric Bead Array (CBA; BD Pharmingen),¹⁷ according to the manufacturer's guidelines. Plasma samples from dengue patients were stored at −70° until evaluation. Then, these samples were thawed and mixed in a 1 : 1 : 1 proportion with CBA cytokine bead mix (IL‐1β, IL‐8, TNF‐α, IL‐6, IL‐10 and IL‐12p70) and PE detection reagent, incubated for 3 hr and washed at 200 g for 5 min. The samples were immediately acquired in the FACS Canto™ II flow cytometer with facs diva software and analyzed by the flowjo v10 software according to the standard curve provided by the kit.

Statistical analysis

Variables in this study were described as frequencies (relative count), absolute numbers (absolute count), measures of central tendency (means and medians) and measures of dispersion (minimal and maximal values, interquartile range and standard deviations) according to the distribution of the data and the nature of the variables. Monocyte subset comparisons between the total patients and the controls were performed by Mann–Whitney U‐test, and the differences among the groups were examined by Kruskal–Wallis test and Dunn's post hoc test. Correlations were determined using the Spearman's rank coefficient with a 95% confidence interval. Correlations with an r value 0·25–0·39 were considered as weak, 0·4–0·5 as low, 0·51–0·74 as medium and ≥ 0·75 as high. The principal component analysis (PCA) was used to generate an orthogonal transformation of the variables to identify the possible associations among linearly uncorrelated variables. Statistical significance was set at the critical values of P ≤ 0·05, *P ≤ 0·05, **P ≤ 0·01 and ***P ≤ 0·001. Analyses were performed using graphpad prism version 5.01 (GraphPad Software Inc., San Diego, CA) and statgraphics centurion XVI Version 16.1.18 (Statgraphics Corp., Rockville, MD) software.

Go to:

RESULTS

Clinical and sociodemographic data of dengue patients

Of the 63 patients included in this study, 49 were classified with dengue (six patients without warning signs and 43 with warning signs) and 14 with severe dengue. Owing to the small sample size, patients with dengue without warning signs could not be analyzed independently. Both groups of patients (with dengue and with severe dengue) had a median age of 31 years (range 18–77 years), and 57% of them were male (Table 1).

Table 1

Sociodemographic, paraclinical and clinical data of patients with dengue

	Dengue	Severe dengue	Total patients
Frequency % (n)	78 (49)	22 (14)	100 (63)
Age in years, median (min value–max value)a	31 (18–70)	31 (18–77)	31 (18–77)
Gender % (n)a
Male	43 (21)	43 (6)	43 (27)
Female	57 (28)	57 (8)	57 (36)
Type of area % (n)a
Urban	71 (35)	86 (12)	75 (47)
Rural	29 (14)	14 (2)	25 (16)
Days of illness % (n)a
4–5 days	22·5 (11)	14 (2)	21 (13)
6–7 days	55 (27)	29 (4)	49 (31)
8–9 days	22·5 ± 5 (11)	57 (8)	30 (19)
Leukocyte count/μl (mean ± SD)a	3740 ± 1773	3244 ± 1420	3622 ± 1697
Platelet count/μl (mean ± SD)a	91 021 ± 50 557	41 443 ± 23 822***	79 826 ± 50 270
AST IU/l (mean ± SD)a	172·7 ± 126·1	494·5 ± 372·7*	238·4 ± 236·2
ALT IU/l (mean ± SD)a	141 ± 106·7	322·9 ± 286·3	176·6 ± 170·3
Bilirubin mg/dl (mean ± SD)a	0·626 ± 0·295	0·55 ± 0·383	0·611 ± 0·306
Hematocrit % (mean ± SD)a	42·56 ± 5·28	43·9 ± 3·92	42·75 ± 5·09
CRP mg/l (mean ± SD)a	2·71 ± 3·68	2·55 ± 3·49	2·69 ± 3·58
Type of infection % (n)
Primary	49 (24)	50 (7)	49 (31)
Secondary	51 (25)	50 (7)	51 (32)
Result of the rapid test % (n)
NS1+	10 (5)	0 (0)	8 (5)
IgM+	4 (2)	0 (0)	3 (2)
IgG+	8 (4)b	0 (0)	6 (4)
NS1+ IgM+	12 (6)	22 (3)	14 (9)
NS1+ IgG+	4 (2)	7 (1)	5 (3)
IgM+ IgG+	43 (21)	64 (9)	48 (30)
NS1+ IgM+ IgG+	19 (9)	7 (1)	16 (10)
Clinical signs of severity (n)
Severe bleeding	0	8	8
Severe organ involvement	0	4	4
Severe plasma leakage	0	2	2
Respiratory distress	0	4	4
Shock	0	4	4

Open in a separate window

^aData obtained from the medical record. CRP, C‐reactive protein; AST, aspartate aminotransferase; ALT, alanine aminotransferase; Min, minimum value; Max, maximum value; SD, standard deviation.

^bPatients positive to anti‐DENV IgG only in the rapid test were corroborated by ELISA, as explained in the Materials and methods section; in this case, the patients were positive to both IgG and IgM by ELISA.

*The Mann–Whitney U‐test was used for the comparison between patients with dengue and with severe dengue *P ≤ 0·05, **P ≤ 0·01 and ***P ≤ 0·001.

Regarding the paraclinical variables, platelet count was statistically lower in patients with severe dengue (41·443 ± 23·822) compared with patients with dengue (91·021 ± 50·557). In addition, the liver transaminase levels were increased in patients with severe dengue compared with dengue patients, being significant only for the AST levels, which were almost three times higher in the severe cases, 494·5 ± 372·7 versus 172·7 ± 126·1; although not significant, the ALT levels were also higher for severe cases, indicating specific hepatic compromise (Table 1). No additional differences were noted in these variables in patients according to their state of illness, neither for the type of infection (primary and secondary) nor for the time of evolution (data not shown).

The healthy controls (n = 45) showed a median of age of 30 years (range: 18–76 years) and 44% of them were male (data not shown).

Patients with dengue exhibited a low count of non‐classical monocytes

Total monocytes corresponding to CD45⁺ CD14⁺ HLA‐DR⁺ cells and monocyte subsets were determined according to the relative expression of CD14 and CD16 in classical (CD14⁺⁺ CD16⁻), intermediate (CD14⁺⁺ CD16⁺) and non‐classical (CD14⁺ CD16⁺⁺) monocytes and evaluated in dengue patients (see Supplementary material, Fig. S1a). An increase was noted in the relative count, but not in the absolute count, of total monocytes in all patients with dengue compared with that in the healthy controls (Fig. 1a). In addition, higher percentages of the intermediate monocytes were observed in patients compared with that in healthy controls (Fig. 1b,c). This difference was not observed in the number of cells belonging to this subset and no changes were detected in the count of classical monocytes (Fig. 1b,c). Finally, a significant decrease in the frequency and number of non‐classical monocytes in all dengue patients were noted in comparison with the controls (Fig. 1b,c); this decrease was not associated with the state of illness, time of evolution and type of infection (data not shown). In addition, a significant but low correlation was recorded between the percentage of CD14⁺ monocytes and the number of platelets/μl (r = −0·26; P = 0·04) and between the number of non‐classical monocytes and AST levels (r = −0·37; P = 0·036); suggesting that changes in the counts of non‐classic monocytes are influenced by hepatic compromise.

 

 

Open in a separate window

Figure 1

Decrease in the relative and absolute count of non‐classical monocytes in patients with dengue. (a) Relative (frequency, left panel) and absolute (right panel) counts of total monocytes (CD45⁺, CD14⁺ HLA‐DR⁺) in 63 patients with dengue and in 45 healthy controls. (b) Representative dot plots to define the monocyte subsets. The data from a dengue patient (right panel) and healthy control (left panel) are shown. (c) Frequency (top) and absolute counts (down) of the three monocyte subsets in 63 patients with dengue and in 45 healthy controls. The median is shown as a central tendency measure. Comparisons between groups were performed using the Mann–Whitney U‐test, *P ≤ 0·05, **P ≤ 0·01 and ***P ≤ 0·001.

The percentage of positive cells to the viral E protein in patients with dengue was similar among the monocyte subsets

To determine whether the changes recorded in the count of monocyte subsets were related to the differences in the infection of these cells, the expression of viral E protein was evaluated in these cells. A similar frequency of positive cells (Fig. 2) and mean fluorescence intensity (data not shown) was obtained for this protein among the classical, intermediate and non‐classical subsets in dengue patients. Although a high frequency of positive cells to viral E protein was observed in a few patients, we were unable to find any clinical or paraclinical characteristics related with this finding.

 

 

Open in a separate window

Figure 2

Similar percentage of positive cells to viral E protein in monocyte subsets of patients with dengue. (a) Representative histograms of the control cells (in red color, corresponding to staining without primary antibody) and the cells stained with the anti‐viral E protein primary antibody (in blue color) in the three monocyte subsets. The data from a healthy control and two dengue patients are shown (one with low frequency and one with high frequency of positive cells). (b) The frequency of positive cells to viral E protein in the three monocyte subsets of 58 patients with dengue. The median is shown as a central tendency measure. Comparisons among the groups were performed using the Kruskal–Wallis test and the Dunn's post hoc test.

High expression of CD64 but low CD16 expression in non‐classical monocytes was observed in dengue patients

As non‐classical cells were the most affected monocytes (in frequency and number) in dengue patients, a more detailed and extensive phenotype characterization of these cells was performed according to the expression of markers associated with the activation, function and differentiation of mononuclear phagocytes (see Supplementary material, Fig. S1b). First, the expression of FcγR CD16, CD32 and CD64, which mediates the antibody‐dependent enhancement, associated with the more severe manifestation of this disease, was evaluated in non‐classical monocytes.¹⁸, ¹⁹ A statistically significant decrease in the expression of CD16 in non‐classical monocytes of total patients was recorded in comparison with that recorded in controls (Fig. 3a), without differences among patients when divided according to the illness state and the infection type (Fig. 3a).

 

 

 

 

Open in a separate window

Figure 3

Low expression of CD16 and high expression of CD64, CCR2 and CCR5 in non‐classical monocytes of patients with dengue. The mean fluorescence intensity (MFI) of (a) CD16 and (b) CD64 FcR, (c) CCR2 and (d) CCR5 chemokine receptors in non‐classical monocytes of 63 patients with dengue and in 45 healthy controls. The results of patients divided according to the state of illness and the type of infection are shown in the middle and right panels, respectively. The median is shown as the central tendency measure. Comparisons between two groups were performed using the Mann–Whitney U‐test, *P ≤ 0·05, **P ≤ 0·01 and ***P ≤ 0·001.

An increase in the CD64 expression was detected in non‐classical monocytes of all the patients compared with that in the healthy controls. This result was mainly detected in patients with primary infection, but it did not differ according to the state of illness (Fig. 3b). These results could suggest that, in patients with IgG‐opsonized DENV, the virus can infect non‐classical monocytes mostly through CD64. Moreover, this molecule has been considered as a differentiation and activation marker of mononuclear phagocytes,²⁰, ²¹ which may explain the changes in the increased expression of CD64.

Non‐classical monocytes of patients with dengue show lower expression of fractalkine receptor (CX3CR1) and higher expression of CCR2, CCR5, CD11b and CD54 compared with healthy controls

As non‐classical monocytes have been involved in the patrolling and surveillance of the microvasculature endothelium,¹¹ the expression of chemokine receptors CX3CR1, CCR2 and CCR5, as well as the expression of the adhesion molecules CD11a, CD11b, CD18, and CD54 was evaluated on these cells of dengue patients. The expression of CCR2, CCR5 (Fig. 3c,d), CD11b and CD54 (Fig. 4a) increased and the expression of CX3CR1 decreased in non‐classical monocytes of patients with dengue. These changes do not seem to be associated with the state of illness, the type of infection (Fig. 3c,d) or the days of illness (Data not shown). However, patients with primary infection exhibited a higher expression of CCR2 and CCR5 in non‐classical monocytes as compared with that in patients with secondary infection (Fig. 3c,d). There were no differences for CD11a, and CD18 expression in non‐classical monocytes between the patients and controls (data not shown). These data indicate that non‐classical monocytes have less expression of CX3CR1 – a chemokine receptor that is important in endothelium patrolling.²² However, these cells have more expression of adhesion molecules and chemokine receptors associated with firm adhesion to the endothelium and cellular transmigration.

 

 

Figure 4

High expression of CX3CR1, CD11b and CD54 in non‐classical monocytes and the increase in the percentage of endothelium‐derived microparticles in patients with dengue. (a) The mean fluorescence intensity (MFI) of CX3CR1, CD11b and CD54 in non‐classical monocytes from 63 patients with dengue and in 45 healthy controls are shown. The results of the patients are divided according to the state of illness and the type of infection are shown in the middle and right panels, respectively. (b) Representative pseudocolor diagram of the HLA‐DR⁺ CD105⁺ microparticles analysis. The data from a dengue patient (middle panel) and healthy control (left panel) are shown. The frequency of the HLA‐DR⁺ CD105⁺ microparticles in 25 patients with dengue and in 20 healthy controls (right panel). The median is shown as a central tendency measure. Comparisons between groups were performed using the Mann–Whitney U‐test, *P ≤ 0·05, **P ≤ 0·01 and ***P ≤ 0·001.

Because previous results were contrasting with the expression of adhesion molecules and to evaluate the patrolling function of non‐classical monocytes over endothelium¹¹ in patients with dengue through an indirect approach, the frequency of vesicles derived from activated endothelium that were CD105⁺ and HLA‐DR⁺ were measured in these patients. An increase was noted in the percentage of endothelium‐derived MPs that were CD105⁺ HLA‐DR⁺ in patients with dengue compared with that in the healthy controls (Fig. 4b), which was not related to the state of illness (data not shown). These results suggest that even though non‐classical monocytes express some molecules associated with firm adhesion to the endothelium in patients with dengue, they do not achieve their surveillance function over this tissue in an efficient manner.

Non‐classical monocytes of patients with dengue exhibit higher expression of markers associated with differentiation and activation of mononuclear phagocytes

Continuing with the phenotype characterization of these cells, markers associated with differentiation and activation of mononuclear phagocytes were evaluated in non‐classical monocytes. No differences were noted in the expression of markers related to the differentiation of mononuclear cells CD33 and CD36 in non‐classical monocytes (data not shown). However, there was an increase in the expression of CD13 (data not shown), CD14, total CD68, CD86 and HLA‐DR in non‐classical monocytes of patients with dengue (Fig. 5a–d). Only for CD14, a lower expression was noted in non‐classical monocytes of patients with severe disease than in those with dengue (Fig. 5a). These data suggest that non‐classical monocytes of dengue patients exhibit changes in association with increased differentiation.

 

 

Open in a separate window

Figure 5

High expression of CD14, CD68, CD86 and HLA‐DR in non‐classical monocytes of patients with dengue. The mean fluorescence intensity (MFI) of (a) CD14, (b) total CD68 (intra and extracellular), (c) CD86 and (d) HLA‐DR in non‐classical monocyte subset from 63 patients with dengue and from 45 healthy controls. The results of the patients divided according to the state of illness and the type of infection are shown in the middle and right panels, respectively. The median is shown as the central tendency measure. Comparisons between groups were performed using the Mann–Whitney U‐test, *P ≤ 0·05, **P ≤ 0·01 and ***P ≤ 0·001.

However, no increase in the expression of CD36 – a traditional marker for monocyte differentiation²³ – and as CD64, CD54, CD86 and HLA‐DR expression may also be up‐regulated in mononuclear phagocytes during activation, the expression of other molecules induced by activation of these cells was evaluated. The expression of CD69 and extracellular CD68, as well as the percentage of TNF‐α ⁺ cells and those positive for the PS exposure increased in non‐classical monocytes of patients with dengue compared with those in the healthy controls (Fig. 6a). These data suggest that non‐classical monocytes are activated in patients with dengue and produce pro‐inflammatory factors, such as TNF‐α. In addition, as the exposure of PS has also been reported as a cell‐death marker, the positivity of some samples to LIVE/DEAD probe was determined. Non‐classical monocytes positive to Annexin‐V were also LIVE/DEAD⁺ (data not shown). In addition, significant and medium correlations were observed between the expression of CD69 and the number of platelets/μl (r = −0·64; P = 0·0008), as well as the frequency of TNF‐α ⁺ cells and bilirubin levels mg/dl (r = −0·55; P = 0·032) in patients with dengue (Fig. 6b). These results suggest that non‐classical monocytes not only exhibited a phenotype of differentiation and activation, but this was also positive to a death marker in patients with dengue. Some of the changes mentioned above correlate with the signs of dengue.

 

 

Figure 6

High expression of CD69 and CD68 (extracellular) and the frequency of tumor necrosis factor‐α‐positive (TNF‐α ⁺) and Annexin‐V⁺ cells in the non‐classical monocyte subset of patients with dengue. (a) The mean fluorescence intensity (MFI) of CD69, CD68 (extracellular) and the frequency of TNF‐α ⁺ and Annexin‐V⁺ cells in the non‐classical subset of 15–25 patients with dengue and 10–20 healthy controls. (b) Analysis of the correlation between MFI of CD69 on non‐classical monocytes and the number of platelets/μl, and the percentage of TNF‐α ⁺ cells in non‐classical monocytes and the blood levels of bilirubin in 23 patients with dengue. The median is shown as a central tendency measure. Comparisons between groups were performed using the Mann–Whitney U‐test *P ≤ 0·05, **P ≤ 0·01 and ***P ≤ 0·001. Correlation analyses were performed by the Spearman rank coefficients at 95% confidence intervals.

Changes observed in non‐classical monocytes associated with paraclinical and immunological variables of patients with dengue

The levels of the circulating cytokines TNF‐α, IL‐1β, IL‐6, IL‐8, IL‐10 and IL‐12p70 were measured with the purpose of examining the correlation of the systemic inflammatory response in dengue patients with changes observed in non‐classical monocytes through a further PCA. Significantly higher concentrations of IL‐6, IL‐8 and IL‐10 were noted in patients with dengue than in healthy controls, without any significant changes in the state of illness (Fig. 7a–c). The levels of IL‐10 were greater in patients with 4–5 days of evolution compared with patients with 8–9 days of evolution (Fig. 7c). No changes were detected in the circulating levels of TNF‐α, IL‐1β and IL‐12p70 in patients with dengue (data not shown).

 

 

Figure 7

High levels of circulating interleukin‐6 (IL‐6), IL‐8 and IL‐10 in patients with dengue. The plasmatic concentration of (a) IL‐6, (b) IL‐8 and (c) IL‐10 in 63 patients with dengue and in 45 healthy controls. The results of the patients divided according to the state and days of illness are shown in the middle and right panels, respectively. The median is shown as a central tendency measure. Comparisons between the two groups were performed using the Mann–Whitney U‐test. Other comparisons among the groups were performed using the Kruskal–Wallis test and the Dunn's post hoc test, *P ≤ 0·05, **P ≤ 0·01 and ***P ≤ 0·001.

A multivariate PCA was performed to evaluate the presence of an association between the alterations observed in the non‐classical monocytes (blue lines) and select variables collected from patients (Fig. 8). This analysis revealed that two components can explain the 70·15% value of the variability in the system, component 1 with 45·08% and component 2 with 25·07%. These components allowed a clear observation of the three main groups of variables based on their eigenvalues. The first included variables IL‐6, IL‐10, IL‐8, MFI of CD69, CD11b, CCR2, CCR5 and percentage of Annexin‐V⁺ and TNF‐α ⁺ cells, suggesting that non‐classical monocytes that were activated and underwent a change in phenotype to a migration profile could be contributing to the elevated concentration of cytokines in these patients and dying in response to the infection. The second group was formed by the number of non‐classical monocytes, the percentage of endothelium‐derived MPs, the CRP values and CX3CR1 MFI. In this group, low number of non‐classical monocytes together with the decreased expression of fractalkine receptor could contribute to a decreased removal of MPs (thereby favouring the endothelial compromise) and with part of the systemic inflammatory response observed in these patients. Finally, the third group included the variables AST, ALT, MFI of CD68, HLA‐DR, CD14, CD86 and CD54 molecules, which suggested that most changes related with the activation and differentiation of non‐classical monocytes could be associated with the hepatic involvement observed in this disease.

 

 

Figure 8

Association between paraclinical and non‐classical monocyte variables in patients with dengue. Principal component analysis showing the correlations among paraclinical variables aspartate aminotransferase (AST), alanine aminotransferase (ALT), C‐reactive protein (CRP) in red lines; interleukin‐6 (IL‐6), IL‐8, IL‐10 in orange; the percentage of microparticles in green line; the number of total monocytes, the number of non‐classical monocytes, the percentage of tumor necrosis factor‐α‐positive (TNF‐α ⁺) and annexin‐V⁺ in non‐classical monocytes and the mean fluorescence intensity (MFI) of CCR5, CCR2, CD11b, CD69, CD68, HLA‐DR, CD86, CD14, CD54 and CX3CR1 in non‐classical monocytes of dengue patient in blue lines.

Go to:

DISCUSSION

The findings of this study demonstrate that dengue patients show a marked decrease in the frequency and number of non‐classical monocytes, accompanied by an increase in the frequency of the intermediate subset; these changes do not seem to depend on the state of the disease, the type of infection, or the days of the disease. Our findings partially agreed with other reports,⁴, ²⁴ including Kwissa et al.,⁴ who found a decrease in both the frequency and absolute count of the classical and non‐classical monocytes in Thai patients infected with DENV, as well as an increase in frequency and an absolute number of intermediate monocytes. However, Azeredo et al.²⁴ showed a decrease in the frequency of the classical subset along with an increase in the intermediate subset, without apparent changes in non‐classical monocytes in mild and severe dengue patients of the Rio de Janeiro city in comparison with the healthy controls. Recently, flow cytometry analysis of PBMCs carried out by Fialho et al.²⁵ demonstrated that Brazilian patients, with dengue fever, experience an increase in the frequency of CD14⁺ CD16⁺  and a decrease in the frequency of CD14⁺ CD16^– monocytes compared with that in the healthy controls. These contrasting results can be explained by the difference in the methodological approaches used, as these studies only discriminate the total monocytes according to the CD14 and CD16 expression. The importance of including the CD45 and HLA‐DR markers in the characterization of monocyte subsets to eliminate some spurious cells from the analysis, such as natural killer cells, granulocytes and erythrocytes, are already established.²⁶ In addition, an increase in the frequency of total monocytes in patients with dengue could be influenced by the significant increase in the intermediate subset; however, the low number of non‐classical monocytes did not significantly affect the total number of monocytes, possibly because of their low frequency in the peripheral blood.

The decrease in non‐classical monocytes has also been reported in other inflammatory diseases, such as systemic lupus erythematosus.¹⁵ Despite different hypotheses proposed to explain these alterations, such as the exposition to treatment or certain cytokines, the causes of this finding remain largely unknown.²⁷, ²⁸ The low count of non‐classical monocytes in dengue patients could be explained by the preferential infection of this subset by DENV; however, no differences were observed in the percentage of positive cells to viral E protein among the monocyte subsets in this study; this measurement was considered to be a reflection of in vitro infection.⁵ It should be noted that the absence of differences in these results might be due to the time elapsed since the viral infection, because all patients had ≥ 4 days from the beginning of the symptoms until monocyte evaluation. Hence, our data do not discard the existence of differences in the frequency of positive monocytes to viral E protein in the earlier days of infection. The frequency of the infected cells was not previously evaluated ex vivo in monocyte subsets, but the NS3 and prM viral antigens were measured in total monocytes.⁸ To the best of our knowledge, this is the first research to evaluate ex vivo the infection of monocyte subsets in patients with dengue. On the other hand, the virus seems to replicate in a similar proportion, in both CD16⁺ and CD16^– previously enriched monocytes (CD14⁺) infected in vitro with DENV‐2.⁵ As the approach used in our study is not precise to determine the infection, the evaluation of monocyte subsets isolated from infected patients with molecular techniques such as real‐time PCR would be ideal owing to their greater sensibility.²⁹

In general, our data suggest that the decrease in the number of non‐classical monocytes is not associated with differences in the percentage of infected cells; hence, it is possible that this decrease would be alternatively explained by a functional response of these monocytes to DENV infection. To partially evaluate this hypothesis, changes in the phenotype of these phagocytes were studied, especially, the expression of receptors associated with phagocytosis and adhesion, chemokine receptors, and markers of cell differentiation and activation. Because the clinical manifestations of patients with dengue were highly heterogeneous and some of the correlation analyses performed between non‐classical monocytes and paraclinical variables achieved significant but low values, a multivariate PCA analysis was performed. It was found that variables evaluated in non‐classical monocytes were divided into three groups together with some of the warning and severity signs of the disease.

1) The systemic inflammatory response (high plasmatic levels of IL‐6, IL‐10, IL‐8 and the percentage of TNF‐α‐positive cells) was associated with an increase in the MFI of CD69, CD11b, CCR2, CCR5 and positive cells for Annexin‐V in non‐classical monocytes. This finding suggests that the activation of non‐classical monocytes and their potential migration and death could be related to the increase in the amount of soluble pro‐inflammatory mediators in dengue. Previously, a high number of positive cells for the ligand of CCR5 (CCL5 or RANTES) was reported in the hepatic tissue of dengue patients with lethal outcomes.³⁰ Similarly, it has been reported that hepatocytes can be infected with DENV and produce ligands for these receptors.³¹ In comparison with the wild‐type mice, CCR2^−/− and CCR5^−/− mice infected with DENV‐2 exhibited a decrease in lethality, lower levels of transaminases and lower frequency of infected macrophages in the liver.³², ³³ Hence, the increase in the expression of these molecules in non‐classical monocytes of dengue patients could indicate a potential migration of these cells toward the hepatic tissue; this phenomenon has also been demonstrated in other diseases, such as atherosclerosis³⁴ and lupus nephritis.³⁵ In addition, it is established that in vitro DENV infection of CD16⁺ monocytes induces a higher production of TNF‐α, IL‐1β, IL‐6, CCL2, CCL3 and CCL4 than the CD16^– monocytes.⁵ The high production of pro‐inflammatory cytokines by non‐classical monocytes has been described in other inflammatory contexts, such as sepsis and lupus.³⁶ Hence, these monocytes seem to be the main subset producing pro‐inflammatory mediators like TNF‐α in response to dengue infection, thereby contributing to the cytokine storm that characterizes patients with severe dengue.³⁷, ³⁸ It is tempting to propose that non‐classical monocytes that have acquired this migratory phenotype contribute to the high inflammatory response observed in the severe dengue cases.

To discriminate whether the frequency of Annexin‐V⁺ monocytes was more related to activation than to cell death, the LIVE/DEAD positivity was evaluated in non‐classical monocytes. Despite the death‐cell type not being defined in the present work, Annexin‐V⁺ non‐classical monocytes showed alterations in the permeability of the plasma membrane. Therefore, the induction of death in these cells could be an alternative explanation for the decrease in both the percentage and the absolute number of non‐classical monocytes in dengue patients. An increase in the death of total human monocytes infected with DENV‐2 in vitro has been previously reported;⁵, ³⁹, ⁴⁰ two types of cell death have been described, namely, apoptosis⁴¹ and pyroptosis.⁴⁰ In contrast, an increase in the viability of both CD16⁻ and CD16⁺ monocytes has been observed after DENV infection in vitro.⁵ These discrepancies may be attributed to the differences in the type of samples evaluated and the time of virus exposition, among other reasons. Hence, additional studies are required to determine whether the non‐classical monocytes die in response to DENV infection.

2) The endothelial damage (increase in MPs) was associated with a low count of non‐classical monocytes, high levels of CRP, as well as a decrease in the MFI of fractalkine receptor (CX3CR1). This observation suggests that non‐classical monocytes do not accomplish their function of endothelial surveillance, possibly by a decrease in the fractalkine receptor, which could contribute to some of the systemic inflammatory responses observed in dengue patients. Non‐classical monocytes also have a patrolling function over the endothelium of microvasculature in the basal state, removing cellular detritus and MPs in a way that is dependent on CX3CR1 expression.⁴² However, when these monocytes are activated through Toll‐like receptor 7 (TLR‐7) and TLR‐8,¹² their expression of adhesion molecules like CD54 increases, they are retained in the endothelial surface and so can mediate endothelial alterations.¹¹ These observations could explain the lower percentage and number of circulating non‐classical monocytes that we recorded in our patients, considering that an increased expression of CD54 and CD11b on non‐classical monocytes was observed in dengue patients. The increase in the expression of CD54 in total monocytes of dengue patients has also been previously described.²⁴ Taking this information into consideration, along with an increase in the frequency of endothelium‐derived MPs, it is possible to hypothesize that the non‐classical monocytes could mediate part of the endothelial dysfunction characteristic of this disease. Moreover, it is worth mentioning that MPs have been proposed previously as biomarkers of predisposition to develop severe manifestations in dengue infection.⁴³

3) The hepatic compromise (increase in AST and ALT) was associated with markers related to the activation and differentiation²⁰, ²¹, ⁴⁴ of mononuclear phagocytes, such as CD86, HLA‐DR, CD68 and CD14, suggesting a relationship between the effector response of these monocytes and the hepatic damage present in these patients. The increase in the expression of CD14 and intracellular CD68 detected in non‐classical monocytes of dengue patients could suggest that these cells are in the process of differentiation in response to infection. This hypothesis agrees with previous reports that human monocytes infected with DENV‐2 in vitro show changes in differentiation, which are like those induced by colony‐stimulating factor.⁴ In addition, the increase in the expression of CD64, CD86, HLA‐DR, CD68 (extracellular), CD69 and in the percentage of cells TNF‐α ⁺ and Annexin‐V⁺ were consistent with the activation phenotype of this subset of monocytes. It was important to highlight that these changes were observed in non‐classical monocytes, but not in the other monocyte subsets (see Supplementary material, Fig. S2 and data not shown), which suggested that despite all subsets being infected, only the non‐classical monocytes responded strongly to the virus. The activation of total monocytes by DENV has been reported previously in dengue patients assessing molecules, such as CD86, CD209, CD11c, CD54, TLR‐2 and TLR‐4.⁸, ²⁴, ⁴⁵ In addition, the activation of these monocytes has been shown in vitro through an increase of inducible nitric oxide synthase⁴⁶ and the production of pro‐inflammatory cytokines.⁵, ³⁹, ⁴⁵, ⁴⁷ However, this is the first study to report specifically the activation of the non‐classical subset in dengue patients. Therefore, our results suggest that an increase in activation markers reported previously in total monocytes of patients with dengue could be explained, albeit partially, by the changes observed in the non‐classical monocytes.

The association between the activation profile of non‐classical monocytes with the observed hepatic damage had not been previously reported in dengue infection; however, this association had been demonstrated in other diseases, such as chronic hepatitis B⁴⁸ and liver cirrhosis.⁴⁹ Moreover, the tissue compromise induced by the non‐classical monocytes has also been observed in autoimmune diseases.⁵⁰ Therefore, it is tempting to propose that, in dengue patients, non‐classical monocytes that acquire differentiation and activation phenotypes seem to contribute to the hepatic injury.

On the other hand, two important results need to be taken into consideration that require future studies to understand their potential implication in the immunopathology of dengue: (i) Some molecules (CD64, CCR2 and CCR5) were expressed differentially in non‐classical monocytes between the primary and secondary infections, which suggested that the response of these cells seemed to be different in the secondary exposure to the virus. This result may be due to the antibody‐dependent enhancement effect¹⁹ or may correspond to an adaptation of these cells related to memory induction of the innate immune system.⁵¹ (ii) Although, there was an increase in the CD14 expression in non‐classical monocytes of dengue patients, this expression decreased in severe dengue patients. It is possible that this alteration corresponds to a release of the CD14 receptor from the membrane into a soluble form.⁵², ⁵³ In addition, a previous study reported a significant correlation between the increase in soluble circulating CD14 and the decrease in platelet count⁵⁴ in patients at different stages of dengue. Moreover, the manner in which soluble CD14 activates the surrounding monocytes is not yet fully understood.⁵⁵ Therefore, it is possible to propose that the activation of non‐classical monocytes by DENV can induce the release of CD14 in patients, thereby contributing to the development of severe dengue.⁵⁶

Finally, this study has two main limitations. First, a support approach or technique was missing in the results, such as a gene expression profile, because of the limited number of non‐classical monocytes detected in patients. Nevertheless, the transcriptomics of monocyte subsets in the healthy individuals has been reported previously,⁵⁷ which has shown the differences in the expression of a variety of genes and markers, similar to those observed in cells of the controls, such as the higher expression of CD36 in classical monocytes and CX3CR1 in non‐classical monocytes.⁵⁷ Therefore, it is possible to speculate that some of the changes observed in non‐classical monocytes of our patients were present at the transcriptional level. New techniques, such as RNA single‐cell sequencing, need to be implemented in the future study to clarify this issue. Second, it is important to note that most non‐severe dengue cases in this study involved patients with the alarm signs at the moment of recruitment; hence, it is possible that these patients were evolving toward severity, which could explain the lack of significant differences between dengue and severe dengue patients in most of the evaluated variables.

Our results suggest that changes in the non‐classical monocytes of patients with dengue are associated with different manifestations of this disease, which may have contributed to the immunopathology of dengue. Therefore, we propose that the profile of molecules described in this study can be evaluated in the future studies to detect patients with a possible predisposition to develop a hepatic compromise, pro‐inflammatory response, or endothelial damage.

Go to:

DISCLOSURES

The authors declare the absence of any commercial or financial relationships that could be interpreted as a potential conflict of interest.

Go to:

AUTHOR CONTRIBUTIONS

JNG performed the experiments, analyzed the data and wrote the manuscript. JAC contributed to sample preparation, carried out some data acquisition and contributed to the final version of the manuscript. MR, BNR and FJD provided a conceptual framework, helped in data analysis and interpretation of the data, and contributed to the final version of the manuscript. PAV and DC designed the study and experiments, supervised the research and wrote the manuscript. All authors approved the final version of the manuscript and agreed to be accountable for all aspects of the work.

Go to:

SUPPORTING INFORMATION

Figure S1. Representative pseudocolor dot plots showing the gating strategy for the analysis of total monocytes (CD45⁺ CD14⁺ HLA‐DR⁺) and subsets: classical (CD14⁺⁺ CD16⁻), intermediate (CD14⁺⁺ CD16⁺) and non‐classical (CD14⁺ CD16⁺⁺) in (a) lysed blood and (b) peripheral blood mononuclear cells (PBMCs). Representative histograms of the analysis of some markers and respective isotype controls in each monocyte subset are shown in PBMCs. SSC‐A: Side Scatter.

Click here for additional data file.^{(41M, tif)}

Figure S2. Expression of CD86 and the frequency of tumor necrosis factor‐α‐positive (TNF‐α ⁺) and Annexin‐V⁺ cells in the monocyte subsets of patients with dengue

Click here for additional data file.^{(1.0M, tif)}

 

Click here for additional data file.^{(15K, docx)}

Go to:

ACKNOWLEDGEMENTS

This research was funded by COLCIENCIAS (grant number 111556933247) and Universidad de Antioquia (Sostenibilidad, Sistema Universitario de Investigaciones, CODI). We thank Dr. Patricia Sierra, coordinator of the infection control program IPS Universitaria. We thank the technical research committee of the IPS Universitaria for its support for carrying out the research (IN07‐2014). The authors wish to thank Anne‐Lise Haenni from the Université Paris‐Diderot for her positive revision and English proofreading.

Go to:

CONTRIBUTOR INFORMATION

Paula A. Velilla, Email: oc.ude.aedu@allilev.aluap.

Diana Castaño, Email: oc.ude.aedu@onatsac.anaid.

Go to:

REFERENCES

1. Halstead SB. Dengue. Lancet 2007; 370:1644–52. [PubMed] [Google Scholar]
2. WHO . Dengue: Guidelines for Diagnosis, Treatment, Prevention and Control. New Edition. Geneva: WHO; Guidelines Approved by the Guidelines Review Committee, 2009. [Google Scholar]
3. Guzman MG, Harris E. Dengue. Lancet 2015; 385:453–65. [PubMed] [Google Scholar]
4. Kwissa M, Nakaya HI, Onlamoon N, Wrammert J, Villinger F, Perng GC et alDengue virus infection induces expansion of a CD14⁺CD16⁺ monocyte population that stimulates plasmablast differentiation. Cell Host Microbe 2014; 16:115–27. [PMC free article] [PubMed] [Google Scholar]
5. Wong KL, Chen W, Balakrishnan T, Toh YX, Fink K, Wong SC. Susceptibility and response of human blood monocyte subsets to primary dengue virus infection. PLoS ONE 2012; 7:e36435. [PMC free article] [PubMed] [Google Scholar]
6. Kurane I, Innis BL, Nimmannitya S, Nisalak A, Rothman AL, Livingston PG et alHuman immune responses to dengue viruses. Southeast Asian J Trop Med Public Health 1990; 21:658–62. [PubMed] [Google Scholar]
7. Kou Z, Quinn M, Chen H, Rodrigo WW, Rose RC, Schlesinger JJ et alMonocytes, but not T or B cells, are the principal target cells for dengue virus (DV) infection among human peripheral blood mononuclear cells. J Med Virol 2008; 80:134–46. [PubMed] [Google Scholar]
8. Durbin AP, Vargas MJ, Wanionek K, Hammond SN, Gordon A, Rocha C et alPhenotyping of peripheral blood mononuclear cells during acute dengue illness demonstrates infection and increased activation of monocytes in severe cases compared to classic dengue fever. Virology 2008; 376:429–35. [PMC free article] [PubMed] [Google Scholar]
9. Carr JM, Hocking H, Bunting K, Wright PJ, Davidson A, Gamble J et alSupernatants from dengue virus type‐2 infected macrophages induce permeability changes in endothelial cell monolayers. J Med Virol 2003; 69:521–8. [PubMed] [Google Scholar]
10. Yona S, Jung S. Monocytes: subsets, origins, fates and functions. Curr Opin Hematol 2010; 17:53–9. [PubMed] [Google Scholar]
11. Carlin LM, Stamatiades EG, Auffray C, Hanna RN, Glover L, Vizcay‐Barrena G et alNr4a1‐dependent Ly6C^low monocytes monitor endothelial cells and orchestrate their disposal. Cell 2013; 153:362–75. [PMC free article] [PubMed] [Google Scholar]
12. Cros J, Cagnard N, Woollard K, Patey N, Zhang SY, Senechal B et alHuman CD14^dim monocytes patrol and sense nucleic acids and viruses via TLR7 and TLR8 receptors. Immunity 2010; 33:375–86. [PMC free article] [PubMed] [Google Scholar]
13. Falconar AK, de Plata E, Romero‐Vivas CM. Altered enzyme‐linked immunosorbent assay immunoglobulin M (IgM)/IgG optical density ratios can correctly classify all primary or secondary dengue virus infections 1 day after the onset of symptoms, when all of the viruses can be isolated. Clin Vaccine Immunol 2006; 13:1044–51. [PMC free article] [PubMed] [Google Scholar]
14. Burbano C, Villar Vesga J, Orejuela J, Muñoz C, Vanegas A, Vásquez G et alPotential involvement of platelet‐derived microparticles and microparticles forming immune complexes during monocyte activation in patients with systemic lupus erythematosus. Front Immunol 2018; 9:322. [PMC free article] [PubMed] [Google Scholar]
15. Burbano C, Vasquez G, Rojas M. Modulatory effects of CD14⁺CD16⁺ + monocytes on CD14⁺⁺CD16^– monocytes: a possible explanation of monocyte alterations in systemic lupus erythematosus. Arthritis Rheumatol 2014; 66:3371–81. [PubMed] [Google Scholar]
16. Arboleda Alzate JF, Rodenhuis‐Zybert IA, Hernandez JC, Smit JM, Urcuqui‐Inchima S. Human macrophages differentiated in the presence of vitamin D3 restrict dengue virus infection and innate responses by downregulating mannose receptor expression. PLoS Negl Trop Dis 2017; 11:e0005904. [PMC free article] [PubMed] [Google Scholar]
17. McDonald SM, Matheson LA, McBane JE, Kuraitis D, Suuronen E, Santerre JP et alUse of monocyte/endothelial cell co‐cultures (in vitro) and a subcutaneous implant mouse model (in vivo) to evaluate a degradable polar hydrophobic ionic polyurethane. J Cell Biochem 2011; 112:3762–72. [PubMed] [Google Scholar]
18. Balsitis SJ, Williams KL, Lachica R, Flores D, Kyle JL, Mehlhop E et alLethal antibody enhancement of dengue disease in mice is prevented by Fc modification. PLoS Pathog 2010; 6:e1000790. [PMC free article] [PubMed] [Google Scholar]
19. Guzman MG, Alvarez M, Halstead SB. Secondary infection as a risk factor for dengue hemorrhagic fever/dengue shock syndrome: an historical perspective and role of antibody‐dependent enhancement of infection. Adv Virol 2013; 158:1445–59. [PubMed] [Google Scholar]
20. Pilling D, Fan T, Huang D, Kaul B, Gomer RH. Identification of markers that distinguish monocyte‐derived fibrocytes from monocytes, macrophages, and fibroblasts. PLoS ONE 2009; 4:e7475. [PMC free article] [PubMed] [Google Scholar]
21. Li Y, Lee PY, Kellner ES, Paulus M, Switanek J, Xu Y et alMonocyte surface expression of Fcγ receptor RI (CD64), a biomarker reflecting type‐I interferon levels in systemic lupus erythematosus. Arthritis Res Ther 2010; 12:R90. [PMC free article] [PubMed] [Google Scholar]
22. Ancuta P, Liu KY, Misra V, Wacleche VS, Gosselin A, Zhou X et alTranscriptional profiling reveals developmental relationship and distinct biological functions of CD16⁺  and CD16^– monocyte subsets. BMC Genom 2009; 10:403. [PMC free article] [PubMed] [Google Scholar]
23. Castano D, Garcia LF, Rojas M. Increased frequency and cell death of CD16⁺monocytes with Mycobacterium tuberculosis infection. Tuberculosis 2011; 91:348–60. [PubMed] [Google Scholar]
24. Azeredo EL, Neves‐Souza PC, Alvarenga AR, Reis SR, Torrentes‐Carvalho A, Zagne SM et alDifferential regulation of toll‐like receptor‐2, toll‐like receptor‐4, CD16 and human leucocyte antigen‐DR on peripheral blood monocytes during mild and severe dengue fever. Immunology 2010; 130:202–16. [PMC free article] [PubMed] [Google Scholar]
25. Fialho LG, Torrentes‐Carvalho A, Cunha RV, Faria N, Gandini M, Cipitelli M et alInduced nitric oxide synthase (iNOS) and indoleamine 2,3‐dioxygenase (IDO) detection in circulating monocyte subsets from Brazilian patients with Dengue‐4 virus. Virol Rep 2017; 7:9–19. [Google Scholar]
26. Poli A, Michel T, Theresine M, Andres E, Hentges F, Zimmer J. CD56^brightnatural killer (NK) cells: an important NK cell subset. Immunology 2009; 126:458–65. [PMC free article] [PubMed] [Google Scholar]
27. Frankenberger M, Ekici AB, Angstwurm MW, Hoffmann H, Hofer TP, Heimbeck I et alA defect of CD16‐positive monocytes can occur without disease. Immunobiology 2013; 218:169–74. [PubMed] [Google Scholar]
28. Ziegler‐Heitbrock L. Blood monocytes and their subsets: established features and open questions. Front Immunol 2015; 6:423. [PMC free article] [PubMed] [Google Scholar]
29. Gurukumar KR, Priyadarshini D, Patil JA, Bhagat A, Singh A, Shah PS et alDevelopment of real time PCR for detection and quantitation of dengue viruses. Virol J 2009; 6:10. [PMC free article] [PubMed] [Google Scholar]
30. de‐Oliveira‐Pinto LM, Marinho CF, Povoa TF, de Azeredo EL, de Souza LA, Barbosa LD et alRegulation of inflammatory chemokine receptors on blood T cells associated to the circulating versus liver chemokines in dengue fever. PLoS ONE 2012; 7:e38527. [PMC free article] [PubMed] [Google Scholar]
31. Suksanpaisan L, Cabrera‐Hernandez A, Smith DR. Infection of human primary hepatocytes with dengue virus serotype 2. J Med Virol 2007; 79:300–7. [PubMed] [Google Scholar]
32. Guabiraba R, Marques RE, Besnard AG, Fagundes CT, Souza DG, Ryffel B et alRole of the chemokine receptors CCR1, CCR2 and CCR4 in the pathogenesis of experimental dengue infection in mice. PLoS ONE 2010; 5:e15680. [PMC free article] [PubMed] [Google Scholar]
33. Marques RE, Guabiraba R, Del Sarto JL, Rocha RF, Queiroz AL, Cisalpino D et alDengue virus requires the CC‐chemokine receptor CCR5 for replication and infection development. Immunology 2015; 145:583–96. [PMC free article] [PubMed] [Google Scholar]
34. Tacke F, Alvarez D, Kaplan TJ, Jakubzick C, Spanbroek R, Llodra J et alMonocyte subsets differentially employ CCR2, CCR5, and CX3CR1 to accumulate within atherosclerotic plaques. J Clin Investig 2007; 117:185–94. [PMC free article] [PubMed] [Google Scholar]
35. Yoshimoto S, Nakatani K, Iwano M, Asai O, Samejima K, Sakan H et alElevated levels of fractalkine expression and accumulation of CD16⁺ monocytes in glomeruli of active lupus nephritis. Am J Kidney Dis 2007; 50:47–58. [PubMed] [Google Scholar]
36. Mukherjee R, Kanti Barman P, Kumar Thatoi P, Tripathy R, Kumar Das B, Ravindran B. Non‐classical monocytes display inflammatory features: validation in sepsis and systemic lupus erythematous. Sci Rep 2015; 5:13886. [PMC free article] [PubMed] [Google Scholar]
37. Rothman AL. Immunity to dengue virus: a tale of original antigenic sin and tropical cytokine storms. Nat Rev Immunol 2011; 11:532–43. [PubMed] [Google Scholar]
38. Pang T, Cardosa MJ, Guzman MG. Of cascades and perfect storms: the immunopathogenesis of dengue haemorrhagic fever‐dengue shock syndrome (DHF/DSS). Immunol Cell Biol 2007; 85:43–5. [PubMed] [Google Scholar]
39. Espina LM, Valero NJ, Hernandez JM, Mosquera JA. Increased apoptosis and expression of tumor necrosis factor‐αcaused by infection of cultured human monocytes with dengue virus. Am J Trop Med Hyg 2003; 68:48–53. [PubMed] [Google Scholar]
40. Tan TY, Chu JJ. Dengue virus‐infected human monocytes trigger late activation of caspase‐1, which mediates pro‐inflammatory IL‐1beta secretion and pyroptosis. J Gen Virol 2013; 94:2215–20. [PubMed] [Google Scholar]
41. Torrentes‐Carvalho A, Azeredo EL, Reis SR, Miranda AS, Gandini M, Barbosa LS et alDengue‐2 infection and the induction of apoptosis in human primary monocytes. Mem Inst Oswaldo Cruz 2009; 104:1091–9. [PubMed] [Google Scholar]
42. Auffray C, Fogg D, Garfa M, Elain G, Join‐Lambert O, Kayal S et alMonitoring of blood vessels and tissues by a population of monocytes with patrolling behavior. Science 2007; 317:666–70. [PubMed] [Google Scholar]
43. Punyadee N, Mairiang D, Thiemmeca S, Komoltri C, Pan‐Ngum W, Chomanee N et alMicroparticles provide a novel biomarker to predict severe clinical outcomes of dengue virus infection. J Virol 2015; 89:1587–607. [PMC free article] [PubMed] [Google Scholar]
44. Li Y, Lee PY, Sobel ES, Narain S, Satoh M, Segal MS et alIncreased expression of FcγRI/CD64 on circulating monocytes parallels ongoing inflammation and nephritis in lupus. Arthritis Res Ther 2009; 11:R6. [PMC free article] [PubMed] [Google Scholar]
45. Chunhakan S, Butthep P, Yoksan S, Tangnararatchakit K, Chuansumrit A. Vascular leakage in dengue hemorrhagic Fever is associated with dengue infected monocytes, monocyte activation/exhaustion, and cytokines production. Int J Vasc Med 2015; 2015:917143. [PMC free article] [PubMed] [Google Scholar]
46. Neves‐Souza PC, Azeredo EL, Zagne SM, Valls‐de‐Souza R, Reis SR, Cerqueira DI et alInducible nitric oxide synthase (iNOS) expression in monocytes during acute Dengue Fever in patients and during in vitro infection. BMC Infect Dis 2005; 5:64. [PMC free article] [PubMed] [Google Scholar]
47. Bosch I, Xhaja K, Estevez L, Raines G, Melichar H, Warke RV et alIncreased production of interleukin‐8 in primary human monocytes and in human epithelial and endothelial cell lines after dengue virus challenge. J Virol 2002; 76:5588–97. [PMC free article] [PubMed] [Google Scholar]
48. Zhang JY, Zou ZS, Huang A, Zhang Z, Fu JL, Xu XS et alHyper‐activated pro‐inflammatory CD16 monocytes correlate with the severity of liver injury and fibrosis in patients with chronic hepatitis B. PLoS ONE 2011; 6:e17484. [PMC free article] [PubMed] [Google Scholar]
49. Panasiuk A, Zak J, Kasprzycka E, Janicka K, Prokopowicz D. Blood platelet and monocyte activations and relation to stages of liver cirrhosis. World J Gastroenterol 2005; 11:2754–8. [PMC free article] [PubMed] [Google Scholar]
50. Wahl SM, Allen JB, Welch GR, Wong HL. Transforming growth factor‐βin synovial fluids modulates Fc gamma RII (CD16) expression on mononuclear phagocytes. J Immunol 1992; 148:485–90. [PubMed] [Google Scholar]
51. Yoshida K, Ishii S. Innate immune memory via ATF7‐dependent epigenetic changes. Cell Cycle 2016; 15:3–4. [PMC free article] [PubMed] [Google Scholar]
52. Rosch M, Helm M, Strecker W, Bock KH, Kinzl L, Bruckner UB et alLiberation of soluble CD14 (sCD 14) in plasma of trauma patients. Unfallchirurg 1997; 100:805–10. [PubMed] [Google Scholar]
53. Rokita E, Menzel EJ. Characteristics of CD14 shedding from human monocytes. Evidence for the competition of soluble CD14 (sCD14) with CD14 receptors for lipopolysaccharide (LPS) binding. APMIS 1997; 105:510–8. [PubMed] [Google Scholar]
54. Yong YK, Tan HY, Jen SH, Shankar EM, Natkunam SK, Sathar J et alAberrant monocyte responses predict and characterize dengue virus infection in individuals with severe disease. J Transl Med 2017; 15:121. [PMC free article] [PubMed] [Google Scholar]
55. Landmann R, Link S, Sansano S, Rajacic Z, Zimmerli W. Soluble CD14 activates monocytic cells independently of lipopolysaccharide. Infect Immun 1998; 66:2264–71. [PMC free article] [PubMed] [Google Scholar]
56. Curtis AM, Edelberg J, Jonas R, Rogers WT, Moore JS, Syed W et alEndothelial microparticles: sophisticated vesicles modulating vascular function. Vasc Med 2013; 18:204–14. [PMC free article] [PubMed] [Google Scholar]
57. Wong KL, Tai JJ, Wong WC, Han H, Sem X, Yeap WH et alGene expression profiling reveals the defining features of the classical, intermediate, and nonclassical human monocyte subsets. Blood 2011; 118:e16–31. [PubMed] [Google Scholar]

Articles from Immunology are provided here courtesy of British Society for Immunology