Increased circulating myeloid-derived suppressor cells in vivax malaria and severe falciparum malaria

Circulating myeloid-derived-suppressor-cells (MDSC) with immunosuppressive function are increased in human experimental Plasmodium falciparum infection, but have not been studied in clinical malaria. Using flow-cytometry, circulating polymorphonuclear-MDSC were evaluated in cryopreserved samples from patients with uncomplicated Plasmodium vivax (n = 8) and uncomplicated (n = 4) and severe (n = 16) falciparum malaria from Papua, Indonesia. The absolute number of circulating polymorphonuclear-MDSC were significantly elevated in severe falciparum malaria patients compared to controls (n = 10). Polymorphonuclear-MDSC levels in uncomplicated vivax malaria were also elevated to levels comparable to that seen in severe falciparum malaria. Control of expansion of immunosuppressive MDSC may be important for development of effective immune responses in falciparum and vivax malaria.


Background
Malaria remains a major life-threatening infectious disease in tropical and subtropical countries. The World Health Organization reported an increase in malaria cases and deaths in 2020, with an estimated 241 million malaria cases and 627,000 deaths worldwide [1]. Plasmodium falciparum is responsible for over 97% of global malaria cases and nearly all malaria deaths [1]. Outside of sub-Saharan Africa Plasmodium vivax also causes significant morbidity [1]. Failure to control the expansion of P. falciparum parasite biomass, including delayed treatment [2], can result in progression to severe malaria and death [3]. The way in which the immune system responds to Plasmodium infection remains incompletely understood, although failure to control Plasmodium infection and clinical disease are associated with an immunoregulatory immune response [4,5].
Myeloid-derived-suppressor-cells (MDSC) are a heterogenous group of polymorphonuclear-type (PMN-MDSC) or monocytic-type MDSC with strong immunosuppressive activity against T-cells, that can regulate the function of other immune cells. MDSC are emerging as key pathological mediators of disease including in cancer [6], sepsis [7,8] and SARS-CoV-2 [9,10], and also play a role in physiological maternalfetal tolerance [11,12]. PMN-MDSC are recognized as pathologically activated neutrophils and display distinct pro-inflammatory and immunosuppressive gene

Open Access
Malaria Journal † Gabriela Minigo and Steven Kho contributed equally to this work *Correspondence: steven.kho@menzies.edu.au signatures compared to classical neutrophils [13,14]. Recent work in controlled human malaria infections (CHMI) with P. falciparum sporozoites report expansion of PMN-MDSC in individuals developing blood-stage parasitaemia with suppression of CD4 + and CD8 + T cell proliferation [15]. MDSC expansion is also reported in murine models of malaria [16][17][18]. Despite their important immunoregulatory roles, MDSC have not been studied in patients with clinical malaria.
During peripheral blood separation by density gradient centrifugation, low-density PMN-MDSC co-purify with peripheral blood mononuclear cells (PBMC) [7], enabling their separation from high-density classical neutrophils sharing the same surface markers [6]. To determine whether PMN-MDSC are elevated in clinical malaria, we evaluated PMN-MDSC in cryopreserved PBMC samples from hospitalised patients with clinical P. falciparum and P. vivax malaria, including patients with severe disease, and compared their levels to uninfected individuals residing in the same endemic area.

Study site and samples
In the eastern Indonesian province of Papua the prevalence of P. falciparum and P. vivax are similar [19], with acute uncomplicated and severe malaria occurring in both children and adults [20]. In clinical studies conducted in Timika between 2011 and 2013 [21], peripheral venous blood samples were collected from malaria patients infected with P. falciparum or P. vivax attending Rumah Sakit Mitra Masyarakat Hospital, and had PBMC cryopreserved. Definitions for enrolment of uncomplicated and severe malaria cases (modified 2000 WHO research criteria; Table 2) are described previously [22]. PBMC from Timika household survey participants collected in the same period who were polymerase-chainreaction-negative for Plasmodium and had no history of fever in the preceding month were included as controls [23]. Participant demographic and clinical data were recorded on standardized forms. Differential full blood counts were collected on an automated analyser (Sysmex,

Data analysis
Flow cytometric data were analysed on FlowJo v10 (Tree-Star, Ashland, OR) and statistical analysis was conducted using Graphpad Prism v9 (La Jolla, CA

Study participants
PBMC preparations from a total of 38 individuals were examined, of whom 8 had uncomplicated P. vivax, 20 had P. falciparum infection (4 uncomplicated and 16 severe) and 10 were PCR-negative healthy controls (Table 1). All participants were adults (> 18 years) except for three  patients with P. falciparum infection who were aged 7, 9 and 13 years. There was an underrepresentation of Papuans in the control group; however, all were residents of Timika for at least 2 years prior to enrolment [23]. Neutrophil counts were higher in uncomplicated vivax and severe falciparum malaria relative to controls, and haemoglobin levels were reduced in severe falciparum malaria (Table 1). In severe malaria, 50% of adults (7/14) had neutrophilia (> 7400 neutrophils per microlitre of blood; Timika adult population mean plus two standard deviations, n = 794 household survey). The most common criteria for disease severity were cerebral malaria and hyperparasitaemia ( Table 2).

PMN-MDSC expansion in acute uncomplicated and severe malaria
Circulating PMN-MDSC were phenotyped by flow cytometry based on side-scatter (SCC) properties and cell surface markers. After gating on SCC hi , PMN-MDSC were identified as CD15 + CD66b + CD11b + CD14 − cells (Fig. 1A) Fig. 1B). PMN-MDSC levels in uncomplicated vivax malaria were comparably elevated to levels seen in severe falciparum malaria. The expansion of circulating PMN-MDSC seen in the 4 cases of uncomplicated falciparum malaria did not reach statistical significance (median 559 [range: 103-8020] per µL blood, p = 0.47). PMN-MDSC levels remained significantly greater in severe falciparum malaria after excluding children (p = 0.005). Among the different severity criteria in patients with severe disease, those with jaundice displayed a trend towards higher circulating PMN-MDSC levels compared to those without jaundice ( Fig. 2A) and those with acute kidney injury (AKI) had reduced PMN-MDSC levels (Fig. 2B); the analyses were likely underpowered with neither comparisons reaching statistical significance. Patients with cerebral malaria or hyperparasitaemia displayed no obvious differences in PMN-MDSC levels compared to patients with other identified severity criteria (Fig. 2C, D).

PMN-MDSC relationships with neutrophils, CD4+ T cells, parasitaemia and age
Whether circulating PMN-MDSC levels were associated with automated neutrophil counts was assessed. Despite sharing surface markers and ancestry [6], there was no significant correlations between MDSC and neutrophils in uncomplicated vivax malaria (r s = − 0.50, p = 0.22; Fig. 2E) or severe falciparum malaria (r s = 0.12, p = 0.68; Fig. 2F). Relationships with CD4 + T cells and lymphocyte counts were examined to gain insight into MDSC immunosuppressive function. In accordance with previous reports from the same study area [23], loss of circulating CD4+ T cells in uncomplicated vivax and severe falciparum malaria relative to controls was observed (Table 1)

Discussion
This is the first study evaluating MDSC in clinical malaria. Patients with uncomplicated P. vivax, P. falciparum and severe falciparum malaria display increased PMN-MDSC in circulating blood. Circulating PMN-MDSC were at comparably high levels in uncomplicated P. vivax and severe falciparum malaria. PMN-MDSC may contribute to the greater immunoregulatory response previously observed in P. vivax infections [4]. Severe falciparum malaria was not associated with higher levels of PMN-MDSC compared to uncomplicated disease, though the power to test this was limited. The expansion of circulating PMN-MDSC seen in clinical P. falciparum infections is consistent with elevated PMN-MDSC reported in CHMI developing P. falciparum parasitaemia [15]. With their known immunosuppressive function, our findings suggest that PMN-MDSC are important immune regulators in clinical disease and may contribute to malaria-induced immunosuppression. PMN-MDSC mediate T cell dysfunction and have been shown to suppress ex-vivo T cell proliferation in CHMI volunteers [15]. In the present cohort, loss of circulating CD4 + T cells was not associated with PMN-MDSC levels in peripheral blood, nor were there any correlations between PMN-MDSC and lymphocyte counts. Larger studies directly assessing the immunosuppressive activity of circulating PMN-MDSC in clinical malaria are warranted. Analysis of severe falciparum malaria as a single group indicated similar PMN-MDSC levels to uncomplicated malaria, though the sample size was underpowered to confirm this observation. The greater expansion of PMN-MDSC in severely ill patients with jaundice compared to those without was not statistically significant but is plausible, and is consistent with a previous study reporting increased PMN-MDSC in cancer patients with related jaundice [24]. Conversely, the trend towards reduced circulating PMN-MDSC in AKI compared to other severity criteria is in-line with cells of this phenotype being recruited to the site of injury in mice models [25], potentially minimising the numbers remaining in peripheral blood. Migration of circulating MDSC to the tissues is also reported in experimental murine cerebral malaria [26]. However, in the present cohort, the similar circulating PMN-MDSC levels seen in cerebral and non-cerebral severe malaria did not reflect this.
PMN-MDSC are pathologically activated neutrophils distinguished from classical neutrophils by density gradient separation [6]. In the present study, the absence of correlations between circulating PMN-MDSC and automated neutrophil counts in uncomplicated vivax and severe falciparum malaria further confirmed that our identification of PMN-MDSC from co-purification with low-density PBMC were unlikely to be contaminated with high-density classical neutrophils, and that PMN-MDSC expansion was not related to the neutrophilia associated with severe falciparum malaria seen here and elsewhere [27]. Inflammatory factors and parasite-derived molecules have been shown to trigger MDSC development in parasitic infections [16]. Furthermore, murine malaria studies suggest that MDSC can contribute directly to parasite clearance and prevent pathology [18]. In patients with malaria, no relationship between circulating PMN-MDSC and parasitaemia was observed, suggesting lack of MDSC anti-parasitic activity or direct parasite-driven effects on MDSC expansion. It was not feasible to evaluate soluble measures of inflammation or more accurate measures of parasite biomass to assess these relationships further. Patient age was associated inversely with the number of circulating PMN-MDSC in uncomplicated vivax malaria, in contrast to the agerelated increase in PMN-MDSC frequency that is known to occur physiologically in humans [28]. It is speculated that chronic exposure to P. vivax may be affecting the agerelated alterations in MDSC homeostasis, though the data supporting this should be interpreted with caution given the small sample size in our cohort.

Conclusion
Circulating PMN-MDSC expand in clinical vivax and severe falciparum malaria, and may contribute to the immunosuppression previously observed in clinical disease from both species. In uncomplicated P. vivax infections, PMN-MDSC concentrations were influenced by age and were at least as high as levels in severe P. falciparum infections. In severe disease, the level of PMN-MDSC expansion may contribute to different severe manifestations. Larger studies on the kinetics of PMN-MDSC (as well as monocytic-MDSC) in clinical disease and their role in asymptomatic infections are warranted to evaluate MDSC as a potential control target to improve immune responsiveness for both species.