Introduction

Malaria is a major parasitic disease with high morbidity and mortality worldwide, with 228 million cases occurring only in the year 2018 [1]. In humans this disease is caused by five different species of Plasmodium [2] of which P. vivax is the species with greatest epidemiological interest as it is the most widely distributed species in regions outside of the African continent. P. vivax is responsible for high morbidity caused by the infection, and brings social and economic losses to the countries where this disease is present [3, 4]. Cases of severe malaria due to infection by this species have been described in the literature more frequently [5–8].

In 2018 more than 194,000 cases of the disease occurred in Brazil, of which more than 99% are from the Amazon region [9]. Another important health issue in the same area is the high circulation of enteroparasites [10–12] which, during a co-infection with Plasmodium vivax, may alter the immunopathological course. A previous study conducted by our group [13] analyzed the impact of co-infection Plasmodium-enteroparasite in the municipality of Oiapoque, located in a border area between Brazil and French Guiana. The results showed an increase in the levels of the cytokines TNF-α, and IL-10 in individuals co-infected with P. vivax and enteroparasitoses. However, this data raises new questions about the impact of other inflammatory immunological markers since it was not possible to verify the influence of co-infection on the cytokines profile that characterizes the Th1 and Th2 pattern.

Receptors present on the surface of innate immune cells can recognize components of microorganisms and initiate the inflammatory response. Among these, the Triggering Receptor Expressed on Myeloid Cells 1 (TREM-1) is a receptor expressed in macrophages, monocytes, and neutrophils that is known to increase inflammatory responses of myeloid cells in innate and adaptive responses that were initiated by the Toll-like receptors [14, 15]. The increase in serum levels of the soluble form of TREM-1 (sTREM-1) has been associated with more severe profiles in other parasitic diseases, such as visceral leishmaniasis [16], and has also been reported in malaria by P. falciparum [17–19].

In addition to sTREM-1, several cytokines have been described that play a role in controlling infection and symptom development [20]. Between them, TNF-α, interleukin 10 (IL-10) and IL-6 are molecules that can be a powerful inflammatory marker of severity during malaria. IL-6 has often been associated with increased severity during malaria caused by P. vivax in Brazil’s extra-amazon [21] and Amazon regions [22] however this cytokine has not yet been evaluated in this scenario of malaria coinfection with enteric pathogens.

This study therefore aims to evaluate the levels of inflammatory mediators in malarial patients and to compare with individuals co-infected with enteroparasites residents of the municipality of Oiapoque, Amapá state, North Brazil.

Material and methods

Results

Serum levels of sTREM-1 and IL-6 were measured in patients with malaria and controls. Considering the high prevalence of enteroparasitosis circulating in the region and that this infection can alter the immune course of malaria this was also measured the levels in two other comparison groups: co-infected and enteroparasites. Results showed a higher level of sTREM-1 and IL-6 in individuals with malaria compared with individuals without the infection (P < 0.05). The level of sTREM-1 was similar between the groups, although a decreasing trend was seen for the co-infected group (Fig 1A). Interestingly for IL-6, in the malaria group the level was 6.9 higher than control (P < 0.0001) and the results showed higher levels of this cytokine compared to all other groups (P < 0.005). In the co-infected group, the IL-6 levels were higher than in the control group (P < 0.002) (Fig 1B).

Download:

• PNG
  larger image
• TIFF
  original image

Fig 1. Differential levels of inflammatory mediators during malaria with sTREM-1 and IL-6 serum levels were associated with malaria infection and IL-6 serum levels were also associated during enteroparasites co-infection.

(A) sTREM-1 measured in serum samples from malaria-infected individuals (n = 34), co-infected patients (n = 116), enteroparasite patients (n = 120) and healthy controls (n = 95). Bars represent the mean ± standard error (*P < 0.05, by Mann Whitney test). The asterisk represents a statistically significant difference between the malaria and the control and group. (B) IL-6 measure in serum samples from malaria-infected individuals (black; n = 34), co-infected patients (blue; n = 116), enteroparasite patients (purple, n = 120) and healthy controls (pink, n = 95). Bars represent the mean ± standard error (****P < 0.0001, by Mann Whitney test). The asterisk represents a statistically significant difference between the malaria and the control and group. (C) The ratio between the levels of IL-10 and IL-6 in serum samples from malaria-infected individuals (n = 34) and co-infected patients (n = 116). Bars represent the mean ± standard error (***P < 0.001, by Mann Whitney test). The asterisk represents a statistically significant difference between the malaria and the control and co-infected group. (D) Ratio between the levels of IL-10 and TNF-α in serum samples from malaria infected (n = 34) and co-infected individuals (n = 116). Bars represent the mean ± standard error (*P < 0.05, by Mann Whitney test). The asterisk represents a statistically significant difference between the malaria and the control and co-infected group.

https://doi.org/10.1371/journal.pone.0270007.g001

Given the importance of the pro-and anti-inflammatory balance during malaria, a comparison was performed between the ratio of the main pro-inflammatory (IL-6 and TNF-α) and anti-inflammatory (IL-10) cytokines identified in the malaria and coinfected groups. In this analysis, the coinfected group had a higher level of IL-10 than the inflammatory cytokines, compared to the profile seen in the malaria group (Fig 1C and 1D).

Considering previous findings that these samples showed an increase in anemia in the coinfected group [13], an evaluation of the three main inflammatory mediators identified in the coinfected group with and without anemia was carried out to identify if there is a difference in the levels of these cytokines in these groups. It was observed from this analysis that the group of individuals with anemia had higher levels of IL-6, TNF- α, and IL-10 cytokines than the individuals without anemia (P < 0.05) (Fig 2A–2C).

Download:

• PNG
  larger image
• TIFF
  original image

Fig 2. Higher levels of inflammatory mediators were correlated with anemia in patients coinfected with P. vivax and enteroparasites.

(A) IL-6 levels in serum samples from co-infected patients with anemia (n = 15) and without anemia (n = 48). Bars represent the mean ± standard error (*P < 0.05, by Mann Whitney test). The asterisk represents a statistically significant difference between the malaria and the control and co-infected group. (B) TNF-α levels in serum samples from co-infected patients with anemia (n = 15) and without anemia (n = 48). Bars represent the mean ± standard error (*P < 0.05, by Mann Whitney test). The asterisk represents a statistically significant difference between the malaria and the control and co-infected group. (C) IL-10 levels in serum samples from co-infected patients with anemia (n = 15) and without anemia (n = 48). Bars represent the mean ± standard error (*P < 0.05, by Mann Whitney test). The asterisk represents a statistically significant difference between the malaria and the control and co-infected group.

https://doi.org/10.1371/journal.pone.0270007.g002

Discussion

The inflammatory response generated during malaria is essential for parasite clearance however enhanced inflammation can trigger immunopathological damage to the host [24, 25]. This is the first study that describes the levels of sTREM-1 in individuals infected with P. vivax and has identified significantly increased levels in infected individuals (P < 0.05).

The sTREM-1, the soluble portion of TREM-1, has been extensively investigated as a potential biomarker of severity [16, 20, 26–28]. In parasitic infections, high levels of sTREM-1 were observed in patients with Visceral Leishmaniasis who died, and the levels of sTREM-1 were negatively correlated with leukocyte count and hemoglobin concentration and were associated with increased parasitemia [16]. Similarly, a study conducted in Uganda with patients infected with malaria by P. falciparum identified sTREM-1 as a new biomarker of severe and fatal malaria in children with this malaria profile [26].

Our study is thus the first to include only non-severe patients infected by P. vivax which is the most widely distributed species of Plasmodium worldwide [1, 3] and observe a difference in sTREM-1 levels when compared to healthy controls. However, when comparing with the coinfected and enteroparasite groups no significant difference was seen, although a decreasing trend in the coinfected group was observed. This may be due to the fact that all the individuals included here had mild symptoms and the main findings already described with sTREM-1 and parasitic diseases correlate this protein with a severity prognosis [16, 20, 26].

Furthermore, experimental studies of sTREM-1 induction with LPS in animal models demonstrated that when low-dose LPS (2 ng/kg) was used, the increase in sTREM-1 was transient and decreased progressively within the first 24 hours [29]. If the same concept applies for malaria, the low parasitemia should influence the data found here considering that many samples have not detected sTREM-1 levels. It must be considered that the serum used in this analysis was collected and refrigerated between 2014 and 2015 years and the sTREM-1 in some samples may have degraded over time. Therefore, the results described here could differ if they are made with fresh samples, we acknowledge.

Our results also demonstrated that IL-6 serum levels increased 6.9 times in malaria patients compared to healthy individuals. In addition, IL-6 was associated positively with IL-10, which is compatible with the mild forms in the patients included in this study. Athough, higher IL-6 production during P. falciparum infection was observed to be associated with increased clinical episodes [30], the IL-6 levels do not change the memory response development in P. vivax, since no variations were observed in the production of anti-PvMSP-1₁₉ IgG antibodies in our previous study with these same samples [13].

In contrast to what was found for sTREM, IL-6 levels were significanty lower in the coinfected group compared to the malaria group. The malaria group showed higher values of this cytokine compared with all other groups, demonstrating that IL-6 levels are highly stimulated during P. vivax infection. Interestingly, the difference observed between malaria and the coinfected group seems to indicate a modulation of inflammation in individuals coinfected with enteroparasites. As expected, the IL-10/IL-6 ratio was lower in malaria group when compared to the coinfected, showing that in the coinfected the response was leaning more towards a suppressive profile. This may indicate a potential protective effect of co-infection on malaria by P. vivax. In malaria by P. falciparum, this protective effect has already been seen in acute renal failure and jaundice [31] and co-infection with Ascaris lumbricoides was protective against cerebral malaria [32]. A study conducted in Africa with people coinfected with P. falciparum and soil borne helminth, coinfection led to increased IL-10 and IL-6 levels and a down-regulation of TGF-β [33]. In our study with P. vivax, the coinfected had increased IL-10/IL-6 and IL-10/TNF-alpha ratios and decreased IL-6 levels when compared to patients who were infected with malaria infection only.

However, another possible scenario must be considered. In the acute phase of infection, a predominantly pro-inflammatory response is important for parasitemia control and parasite clearance [34]. Given that all the individuals were included in the study at the time of diagnosis it is likely that most of them were in the acute phase and therefore this immunoregulation may be favorable to the parasite, ensuring maintenance of the infection.

Additionally, increased anemia was seen in the coinfected compared to the malaria group [13] which may support the hypothesis that the suppressive profile may not be beneficial for the patients. Anemia during malaria is a multifactorial etiology. The incidence and severity increase in specific groups such as children under 5 years old, pregnant women, and can also be caused due to resistance during anti-malarial treatment, hemolysis, malnutrition, and other associated infections such as HIV, parvovirus B19, Intestinal helminths [35]. However, among these characteristics, only enteroparasite coinfection was present in the individuals included in our study.

It is worth of noting that despite the decrease in IL-6, this coinfected group showed higher values of TNF-α than malaria [13]. TNF-α is an important pro-inflammatory cytokine and during malaria has been correlated with the development of anemia [36]. The anemia in these patients may thus be developed as a result of the inflammatory response of these cytokines and the generation of a damage-associated molecular pattern (DAMPs) [37, 38] and/or due to a strong TH2 response that favors the persistence of the parasite and is generated by the natural cycle of Plasmodium replication that leads to red blood cells lysis.

As expected, in malaria patients, parasitemia was positively correlated with the anti and pro-inflammatory cytokines IL-10 and TNF-α. In a study conducted in a municipality in Amazonas, Brazil increased IL-10 and IL-6 were associated with parasite load [22]. In our study, although there was a tendency to increase, no significant association was found. For sTREM-1 no association with the cytokine panel or with the clinical variables parasitemia, gametocytes, and hemoglobin levels has been identified.

We also demonstrated a potential modulation of IL-6 during coinfection with intestinal parasites. It is important to mention that most malaria-endemic regions have a high prevalence of enteroparasites [31, 32, 39]. Nevertheless, few studies have focused on assessing the impact of co-infection on malaria prognosis. One limitation that we acknowledge is the inability to separate enteroparasites into the helminth, protozoan, and both groups in our analysis due to the low number of individuals in the groups.

Therefore, studies are needed to determine the impact of these immune molecules on each of these groups, especially sTREM-1 which has been scarcely studied in these populations. Moreover, studies have demonstrated the impact of polymorphisms in the TREM-1 gene on the susceptibility, inflammatory role and prognosis of malaria and other diseases [17, 40–44]. It is also necessary therefore to assess the genetic variability of TREM-1 in populations of endemic areas and their impact on the course of malaria by P. vivax. Finally, is the coinfection with enteroparasites beneficial during P. vivax infection in non-severe cases, and is the anemia developed during coinfection caused by immune damage or due to parasite persistence? are two key questions that need further clarification.

Supporting information

Acknowledgments

We acknowledge and thank all those who accepted to participate in the present study, as well as to the professionals of the Hospital of the municipality of Oiapoque, in which the activities were developed in particular to the employees: Flaviano Feitosa and Leudilene Marques for assistance in sample collection and technical support.