Study cohort
To address whether lifelong exposure to intense P. falciparum transmission or concurrent asymptomatic P. falciparum-infections lead to altered frequency or function of DCs, we analyzed freshly isolated, primary myeloid DC activation by P. falciparum-iRBC lysate in vitro in a malaria-endemic setting in a cross-sectional study. In the rural village of Kambila, Mali, where malaria transmission is intense and seasonal [32], we enrolled 35 asymptomatic adults in the study. Peripheral venous blood was drawn at the peak of the malaria transmission season and transported to a laboratory at the Malaria Research and Training Centre in Bamako, Mali, where in vitro experiments were conducted immediately upon sample arrival. Plasma samples were cryopreserved and shipped to the NIH for cytokine and chemokine analysis. Asymptomatic infection was determined using PCR at the NIH, Rockville. Of the 35 enrolled Malian individuals, 8 were PCR-positive for peripheral P. falciparum blood-stage infection (Table 1). We observed a significant difference in age between the two groups: asymptomatically P. falciparum-infected adults were on average 9 years older than the uninfected group (uninfected: 42±8; infected: 51±5). Due to naturally acquired clinical immunity to the parasite, adults in this cohort rarely develop symptomatic infections [18]. To be able to characterize cytokine and chemokine plasma levels of symptomatic P. falciparum-infections along with asymptomatic infections, we added plasma samples from 19 children between the ages of 6 and 9 years into the study (Table 1). These children resided in a different village than the adult cohort. Malaria endemicity in this village was comparable to the study site of the adult cohort [33]. PBMC subset and in vitro mDC analyses were not performed with samples obtained from children due to the lower blood volumes obtained and lack of access to freshly isolated cells. The groups were mostly sex balanced (Table 1).
Table 1. Study cohort.
|
Subjects
|
Age
|
Sex
|
Symptomatic
|
|
n
|
median (IQR)
|
M/F
|
|
Adults uninfected
|
27
|
42 (37-49)
|
15/12
|
no
|
Adults infected
|
8
|
51 (46-55)
|
5/3
|
no
|
Children
|
19
|
8 (6-9)
|
11/8
|
|
Healthy baseline
|
|
|
|
no
|
Acute infected
|
|
|
|
yes
|
7 days post treatment
|
|
|
|
no
|
Subset distribution and apoptosis of DCs in malaria-exposed and asymptomatically-infected adults
First, enumeration of peripheral blood DC subsets was conducted to identify possible differences in DC numbers between malaria-exposed but uninfected and asymptomatically P. falciparum-infected adults. The human peripheral blood DC compartment can be identified by flow cytometry as lineage (CD3, CD14, CD19, CD20 and CD56) negative and HLA-DR positive cells [34]. This population can be further divided into myeloid (mDC) and plasmacytoid DC (pDC) subsets. pDCs can be classified by their expression of CD303. mDCs are composed of two classical subsets, CD1c+ and CD141+ DCs, and a third subset characterized by the expression of CD16 (Figure 1A) [35, 36]. We used this classification to assess the frequency of these DC subsets in malaria-exposed but uninfected and asymptomatically P. falciparum-infected Malian adults. FACS analysis percentages were used to normalize to subset cell number per µl of blood after determining the total number of PBMCs per sample/volume of blood (Figure 1). No significant difference in peripheral numbers of total DCs or mDC subsets was observed. pDC numbers in the circulation of asymptomatically P. falciparum-infected individuals, however, were found to be on average approximately half of the frequency of pDCs in uninfected individuals (Figure 1B).
A decrease in HLA-DR surface expression, which might lead to suboptimal T cell priming, has been described in acute malaria patients [28, 37-42]. Therefore, we asked whether altered HLA-DR expression was associated with asymptomatic P. falciparum-infection. We found that HLA-DR surface levels in uninfected and asymptomatically-infected Malians were comparable (Figure 1C). As expected, CD1c+ mDCs expressed the highest levels of HLA-DR on their surface, while CD16+ exhibited low surface HLA-DR (Figure 1C) [36, 43].
In patients with acute malaria, it has been reported that peripheral blood DCs undergo apoptosis [28]. In this study, no significant difference of total DC apoptosis was observed between uninfected and asymptomatically infected Malian adults (Figure 2). The infected group exhibited a rather large variability of DC apoptosis percentages, ranging from 2.75% to 69.5%.
These data suggest that pDC numbers are slightly lower in asymptomatically P. falciparum-infected compared to uninfected Malian adults while mDC numbers in circulation remain unchanged.
mDCs up-regulate costimulatory molecules and secrete IL-10, CXCL9 and CXCL10 in response to P. falciparum in malaria-exposed adults
Although it would be informative to analyze pDCs and mDCs from individuals living in an endemic area, we decided to focus only on mDCs for this study to enable a more detailed assessment and their ability to efficiently prime naïve T cells. A limitation of commonly conducted activation analyses using whole PBMCs are the various, possibly regulatory, cell types present in PBMCs. To avoid uncontrollable effects of other cell types, we enriched mDCs by negative selection depleting cells bound to magnetic beads with lineage-specific antibodies. Enriched mDCs were comprised of CD1c+ and CD141+ mDC subsets. Enrichment and in vitro incubation with P. falciparum blood-stage lysates were undertaken with freshly isolated cells to minimize the effects of handling, freezing, and thawing the cells. DC surface marker expression as well as cytokine and chemokine secretion commonly associated with mDC activation were analyzed after stimulation with P. falciparum-iRBC lysate for 24h. Uninfected RBC lysate was used as a negative control. To minimize batch effects we assessed mDC responses to the parasite for all 35 enrolled uninfected and asymptomatically P. falciparum-infected adults simultaneously. However, due to generally low numbers of DCs in circulation, some samples had too few DCs to successfully execute the assay. Out of the 27 uninfected adults, 21 had sufficient mDC numbers to conduct the experiments (Figure 3).
mDCs isolated from uninfected Malian adults were able to up-regulate the activation markers HLA-DR, CD80, CD86 and CD40 upon stimulation with parasite lysate (Figure 3A). Malian mDCs did not secrete significant amounts of the inflammatory cytokines IL-1β, IL-6 or TNF, but we measured a slight but significant increase of IL-10 in the culture supernatant when mDCs were stimulated with P. falciparum-iRBC lysate (Figure 3B). mDCs from malaria-exposed adults also secreted the Th1-associated chemokines CXCL9 (MIG) and CXCL10 (IP-10) as well as CCL2 but no CCL5 (Figure 3C).
Of the 8 asymptomatically infected individuals, only 4 had sufficient mDC numbers for analysis. Among these 4 individuals, no trends or significant differences were detected (Figure 4).
Plasma cytokine and chemokine levels in asymtomatically-infected adults and children with acute malaria
Next, plasma cytokine and chemokine levels in the described cohort of asymptomatically P. falciparum-infected and uninfected Malian adults were analyzed by bead-based immunoassays. In addition to the cytokines and chemokines tested in DC supernatants, cytokines commonly associated with T cell responses were tested.
We detected slightly elevated levels of IL-10 (Figure 5A) and CXCL9 (Figure 5B) in plasma of asymptomatically infected compared to uninfected adults. This observation corroborates our finding that DCs in these individuals respond to the parasite with IL-10 and CXCL9 in vitro (Figure 4). All other analyzed cytokines and chemokines, namely IL-6, TNF, IL-1β, IL-18, IFNγ, IFNα, CCL2, CCL5 and CXCL10, were comparable between the two groups (Figure 5), suggesting they were at baseline levels in both groups.
Finally, to compare these responses to symptomatic malaria, plasma cytokine and chemokine levels were measured longitudinally in Malian children before (healthy baseline, HB), during (malaria, Mal) and after (convalescence, Conv) an acute symptomatic malaria episode. Plasma obtained from US adults was used as a control. As expected, most inflammatory cytokine and chemokine levels tested increased significantly during acute malaria and decreased to baseline levels at the convalescent time point 7 days post treatment. Similar to asymptomatically infected adults (Figure 5A), plasma IL-10 levels exhibited the largest increase with a 150-fold increase during acute malaria compared to baseline (Figure 6A). Mean IL-6 and IFNγ levels increased roughly 30- and 3.5-fold, respectively. Surprisingly, the inflammatory cytokines IL-1β and TNF did not show significantly higher plasma levels during acute malaria. Overall IFNα levels in plasma were low, but we measured a significantly higher amount of this cytokine during acute malaria compared to convalescence (Figure 6A). The Th1-associated chemokines CXCL9 and CXCL10 exhibited a robust increase in plasma levels during acute malaria (Figure 6B).
Most plasma cytokine and chemokine levels returned to baseline by the convalescent timepoint, with the exception of IL-18, CCL5 and CXCL9 (Figure 6).
Taken together, these data indicate that IL-10 plays a prominent role in the cytokine response to P. falciparum in malaria-endemic settings.