Greater Endothelial Activation, Weibel-Palade Body Release and Host Inflammatory Response to Plasmodium vivax, compared with Plasmodium falciparum: A Prospective Study in Papua, Indonesia

Abstract

Pathogenic mechanisms underlying vivax malaria are poorly understood, with few studies comparing endothelial and inflammatory responses with falciparum malaria. In adults with uncomplicated vivax or falciparum malaria, we compared plasma measurements of endothelial Weibel-Palade body release (angiopoietin-2) and activation (ICAM-1, E-selectin), as well as selected cytokines. Despite a lower median parasite count, angiopoietin-2 concentrations were higher in patients with vivax malaria, compared with falciparum malaria. Per peripheral parasite, median plasma angiopoietin-2, ICAM-1, E-selectin, interleukin-6, and interleukin-10 concentrations were higher in patients with malaria due to Plasmodium vivax. P. vivax induces greater endothelial Weibel-Palade body release and activation and greater host inflammatory responses, compared with Plasmodium falciparum.

Plasmodium vivax causes 70–390 million clinical cases of malaria annually [1]. Previously considered to be benign, recent studies from Indonesia, Papua New Guinea, and India have challenged this view, demonstrating that P. vivax is associated with severe malaria and death [1–3]. Despite the increasing recognition of its clinical and public health importance, little is known about the pathogenesis of vivax malaria [4]. Few studies have compared host responses in disease caused by P. vivax and Plasmodium falciparum, the 2 major species that cause malaria [4].

There are fundamental differences in the biology of P. vivax and P. falciparum [4]. P. falciparum invades all erythrocytes and can cause high parasitemias, whereas P. vivax invades reticulocytes, limiting the total number of parasitized erythrocytes to <2% of the red cell mass [4]. Although central to the pathogenesis of P. falciparum [5], endothelial cell cytoadherence of parasitized erythrocytes and microvascular sequestration have not been definitively reported with P. vivax [4]. These differences are thought to explain the increased risk of progressing to severe disease with falciparum malaria, compared with vivax malaria. In contrast, plasma tumor necrosis factor (TNF) concentrations, associated with severe disease in falciparum malaria, can be higher prior to the paroxysms of vivax malaria. This has been linked to a stronger host response per parasitized erythrocyte to P. vivax [6, 7].

Weibel-Palade body (WPB) release and endothelial activation are major pathogenic processes in falciparum malaria, associated with increased cytoadherence, sequestration, microvascular dysfunction, impaired organ perfusion, severe disease, and death [8]. The role of endothelial activation and WPB release in the pathophysiology of vivax malaria is not known and neither are the comparative mechanisms underlying endothelial activation by sequestering and nonsequestering parasites [4]. In nonimmune returned travellers, plasma concentrations of endothelial adhesion molecules are increased in vivax malaria [9, 10], comparable to levels found in uncomplicated falciparum malaria [9, 10]. In endemic areas, plasma levels of von Willebrand Factor, found in WPBs and platelets, and the non– endothelial-specific adhesion molecule intercellular adhesion molecule-1 (ICAM-1), have been reported to be higher in falciparum malaria, compared with vivax malaria [11, 12]. There are no data on specific endothelial activation markers from endemic regions or regions where complicated vivax malaria is seen. Similarly, there are no data comparing endothelium-derived autocrine mediators of endothelial activation, such as angiopoietin-2, that have been associated with impaired microvascular perfusion and fatal outcome in falciparum malaria [8].

We hypothesized that endothelium-specific activation and inflammatory responses in uncomplicated P. vivax infections would be greater than those in patients infected with P. falciparum malaria. In a prospective observational study of adults with uncomplicated vivax and falciparum malaria in Indonesia, we compared the plasma levels of angiopoietin-2 (Ang-2), a specific measure of endothelial WPB release and autocrine activation, endothelial activation markers (E-selectin and ICAM-1), and selected pro- and anti-inflammatory cytokines.

Methods

This prospective observational study was conducted at the Mitra Masyarakat Hospital (Papua, Indonesia). Ethical approval was obtained from the Ethics Committees of the National Institute of Health Research and Development in Indonesia and the Menzies School of Health Research in Australia. Written informed consent was obtained from all patients.

Participants were aged ≥18 years and had (1) uncomplicated P. falciparum or P. vivax monoinfection by microscopy, (2) hemoglobin level >60 g/L, (3) history of fever in the preceding 48 h, (4) no warning signs or criteria for severe malaria according to World Health Organization guidelines, and (5) for P. falciparum, >1000 asexual parasites/μL. The clinical and laboratory characteristics of the patients with P. falciparum have been reported elsewhere [8, 13]. The decision on the need for inpatient treatment was made by local physicians, independent of the research team, on the basis of clinical indications. Clinical data were collected using a standardized data form. Venous blood was collected in lithium heparin, was plasma separated, and was stored at −70°C within 30 min.

Parasite counts were determined by microscopy of Giemsa-stained thick and thin fields by experienced research micros-copists. Hemoglobin level, biochemistry, acid-base parameters, and lactate levels were measured with a bedside analyzer (i-STAT). White blood cell counts were measured by Coulter counter. Plasma concentrations of ICAM-1, E-selectin, and Ang-2 were measured by enzyme-linked immunosorbent assay (R&D), and TNF, interleukin (IL)-6, IL-4, and IL-10 levels were measured by cytometric bead assay, as described elsewhere [8, 13]. For statistical purposes, concentrations below the lower limit of detection were assigned a value of one-half the lower limit. Concentrations were expressed as median values and, for cytokines with median values above the lower detection limit, as a ratio of peripheral parasitemia, giving plasma concentrations per peripheral parasite. Reference values for plasma Ang-2, representative of this population, were those from a control group reported elsewhere [8, 13].

Continuous variables were compared between P. vivax and P. falciparum by T test or Mann-Whitney U test, as appropriate. Pearson’s or Spearman’s methods were used to estimate correlations, as appropriate. All analyses were performed with Stata (version 9.2; StataCorp). A 2-sided value of P < .05 was considered to be significant.

Results

From February 2005 through February 2006, 47 patients with uncomplicated vivax malaria, 78 with uncomplicated falciparum malaria, and 18 healthy controls were enrolled. There was no difference in baseline characteristics between the groups, except P. vivax infection was associated with a lower parasitemia (Table 1) and a lower frequency of hospitalization (24 patients [51%]), compared with P. falciparum (78 patients [100%]). No patient developed severe disease or died.

Table 1.

Baseline Characteristics and Concentrations of Endothelial Activation Markers and Cytokines

Characteristic	Healthy control subjects (n = 18)	Patients with falciparum malaria (n = 78)	Patients with vivax malaria (n = 47)
Age, mean years (range)	25 (18–44)	28 (18–56)	23 (18–35)
Male sex	13 (74)	32 (67)	33 (71)
Papuan highlander ethnicity	15 (87)	59 (77)	42 (90)
Duration of fever before presentation, median days (IQR)	NA	2 (1–5)	3 (2–4)
Systolic blood pressure, mean mmHg (range)	128 (96–136)	110 (80–134)	113 (94–140)
Respiratory rate, mean breaths/min (range)	20 (18–26)	23 (16–32)	25 (18–44)
Temperature, mean °C (range)	35.6 (35–36.7)	36.5 (34.8–40.2)	36.3 (35.1–40.2)
Required hospitalizationa	0 (0)	78 (100)	24 (51)
White blood cell count, mean cells/μL × 103 (95% CI)	ND	5.9 (5.4–6.9)	6.5 (5.8–7.2)
Hemoglobin level, mean g/L (range)	ND	121 (70–170)	129 (60–163)
Parasite density,a geometric mean parasites/μL (range)	0 (0–0)	14,900 (850–127,000)	3918 (572–18,700)
Soluble ICAM-1 level, mean pg/mL (95% CI)	ND	569 (516–623)	580 (520–639)
Soluble ICAM-1 per parasite,a median pg/mL/PRBC × 10−4 (IQR)	NA	0.1 (0.05–0.9)	1.2 (0.6–2.6)
Soluble E-selectin level, mean pg/mL (95% CI)	ND	106 (95–118)	102 (85–119)
Soluble E-selectin per parasite,a mean pg/mL/PRBC × 10−5 (95% CI)	NA	0.3 (0.09–1.4)	2 (1.2–5.4)
Plasma angiopoietin-2 level,a median pg/mL (IQR)	2300 (1900–3600)	5300 (3900–8200)	7000 (4200–10,600)
Plasma angiopoietin-2 per parasite,a median pg/mL/PRBC × 10−3 (IQR)	NA	0.1 (0.04–0.6)	1.8 (0.7–3)
Plasma TNF level,b median pg/mL (IQR)	ND	1.25 (1.25–1.25)	1.25 (1.25–1.25)
Plasma IL-6 level, median pg/mL (IQR)	ND	27 (10–123)	15 (6–78)
Plasma IL-6 per parasite,a median pg/mL/PRBC × 10−6 (IQR)	NA	1.41 (0.37–3.2)	3.7 (1.1–15.9)
Plasma IL-4 level,b median pg/mL (IQR)	ND	1.25 (1.25–1.25)	1.25 (1.25–1.25)
Plasma IL-10 level, median pg/mL (IQR)	ND	130 (51–571)	74 (26–290)
Plasma IL-10 per parasite,a median pg/mL/PRBC × 10−5 (IQR)	NA	0.5 (0.1–2.7)	1.8 (0.6–7.1)

NOTE. Data are no (%) of persons, unless otherwise indicated. CI, confidence interval; ICAM-1, intercellular adhesion molecule-1; IL, interleukin; IQR, interquartile range; NA, not applicable; ND, not done; PRBC, parasitized red blood cell; TNF, tumor necrosis factor.

^aP < .05 by Kruskal-Wallis or Mann-Whitney test for Plasmodium vivax vs Plasmodium falciparum.

^bMedian concentrations per parasite not determined for median values below the lower limit of detection.

Plasma Ang-2 concentrations were significantly increased in patients with vivax malaria, compared with patients with falciparum malaria and healthy individuals (P = .01; Figure 1). Plasma ICAM-1 and E-selectin concentrations were comparable between species (Table 1). There was no significant difference in absolute TNF, IL-6, IL-4, and IL-10 concentrations between patients with vivax and falciparum malaria (Table 1). However, when expressed as concentrations per peripheral parasite (for mediators with median values greater than the lower limit of detection), median plasma concentrations of endothelial activation markers, IL-6, and IL-10 were all higher in patients with malaria due to P. vivax (Table 1). There were no significant correlations found between ICAM-1, E-selectin, or Ang-2 concentrations with TNF, IL-6, IL-4 or, IL-10 levels in patients with vivax or falciparum malaria.

Figure 1.

Plasma concentrations of the Weibel-Palade body product, angiopoietin-2, in patients with uncomplicated Plasmodium falciparum and Plasmodium vivax malaria (P = .01). Horizontal lines indicate median values.

Discussion

Despite a lower peripheral parasitemia, plasma Ang-2 concentrations were higher in acute uncomplicated vivax malaria than in uncomplicated falciparum malaria. This suggests that P. vivax is intrinsically capable of inducing greater release of Ang-2 from endothelial WPBs than P. falciparum. Although absolute plasma ICAM-1, E-selectin, IL-6, and IL-10 concentrations were comparable between species, per parasite concentrations were higher in P. vivax than in P. falciparum. These results extend previous indications of a greater proinflammatory response to P. vivax, compared with P. falciparum [6, 7], and suggest that, for a given parasite biomass, P. vivax induces greater endothelial activation and greater production of both pro- and anti-inflammatory cytokines.

The per-parasite calculations of plasma concentrations assume that parasite-induced cytokine or adhesion molecule expression is not saturated at the levels of parasitemia found in uncomplicated malaria. This is likely, given the much greater concentrations of plasma cytokines and endothelial adhesion molecules reported in severe falciparum malaria with a higher parasite biomass [8, 13]. Furthermore, because of significant sequestration of parasites in falciparum but not vivax malaria, peripheral parasitemia underestimates total P. falciparum biomass, making the per parasite differences in plasma concentrations between species conservative estimates.

Our findings provide additional evidence that the effects of malaria parasites on endothelial adhesion molecule expression and WPB exocytosis may not be primarily dependent on direct contact of parasitized red cells with endothelial cells [14]. Although increased endothelial cell ICAM-1 expression is clearly associated with and mediates cytoadherence of parasitized red cells to endothelial cells in fatal falciparum malaria [5], in vitro evidence suggests that P. falciparum induces ICAM-1 expression independent of binding affinity, with a major role for soluble parasite factors [14]. Cytoadherence will increase the approximation of P. falciparum with endothelium under flow conditions in vivo. Nevertheless, WPB release and endothelial activation are greater in malaria due to P. vivax, a parasite with, at most, a paucity of endothelial cytoadherence and sequestration, suggesting that soluble parasite factors may be more important in inducing endothelial activation with both Plasmodium species.

Factors causing WPB exocytosis and Ang-2 release include thrombin, histamine, and activation of Toll-like receptors 2 and 4 [8]. Parasite factors proposed to mediate Ang-2 release in P. falciparum [8] include the Toll-like receptors 2 and 4 ligand glycophosphotidylinositol [4] and a soluble histamine-like molecule [15]. The homologues in P. vivax are unknown. Lipids derived from and specific to P. vivax trigger proinflammatory responses but have not been characterized in detail [6].

The current and previous studies indicate greater proinflammatory cytokine responses to P. vivax infection, compared with P. falciparum infection. However, the lack of correlation between endothelial adhesion molecules and any of the plasma cytokine concentrations in malaria from either species in the current study and other studies in P. falciparum [14] supports the view that cytokines are not the primary cause of endothelial activation or WPB release in either P. falciparum [8, 13] or P. vivax infection.

Ang-2 concentrations are significantly increased in P. vivax infection but are reported to be even higher in severe falciparum malaria [8], a likely reflection of the much higher parasite biomass found in severe disease due to P. falciparum. WPB exocytosis, as evidenced by elevated Ang-2 concentrations [8], is associated with mortality in both severe falciparum malaria [8] and bacterial sepsis, but the role of Ang-2 and endothelial activation in the pathophysiology of uncomplicated and severe vivax malaria is not yet known. We were unable to exclude concurrent bacteremia, which may also increase Ang-2 concentrations; however, this is rare in nonsevere malaria, and all inpatients with vivax malaria recovered with use of antimalarial therapy alone.

This study demonstrates another important difference in the pathobiology of P. vivax and P. falciparum, with uncomplicated vivax malaria associated with significantly greater WPB exocytosis and endothelial activation than uncomplicated falciparum malaria, despite a lower parasitemia. The mechanisms underlying the greater inflammatory response to P. vivax require further characterization.