Nrf2-driven CD36 and HO-1 gene expression in circulating monocytes correlates with favourable clinical outcome in pregnancy-associated malaria

Abstract

Background: Pregnancy-associated malaria (PAM) constitutes one of the most severe forms of malaria infection leading to fetal growth restriction and high risk of infant death. The severity of the pathology is largely attributed to the recruitment of monocytes and macrophages in the placenta which is evidenced by dysregulated inflammation found in placental blood. Importantly, CD36(+) monocytes/macrophages are also thought to participate in the tight control of the pro- and anti-inflammatory responses following Plasmodium detection through elimination of apoptotic cells and malaria-infected erythrocytes, internalization and recycling of oxidized forms of low-density lipoprotein and collaboration with TLR2 in pro-inflammatory response. Interestingly, previous work demonstrated that CD36 expression was upregulated on inflammatory macrophages following stimulation of the Nrf2 transcription factor, whilst the PPARγ pathway was inhibited and non-functional in the same inflammatory conditions. This current study examined the possible role of Nrf2-driven gene expression, CD36 and Haem-Oxygenase-1 (HO-1), in PAM clinical outcomes.

Methods: Clinical data and biological samples including peripheral blood mononuclear cells were collected from 27 women presenting PAM. Polychromatic flow cytometry was used to characterize innate immune cell subpopulations and quantify CD36 protein expression level on monocytes. mRNA levels of CD36, PPARγ, Nrf2 and HO-1 were determined by qPCR and related to clinical outcomes. Finally, the capacity of monocytes to modulate CD36 expression upon rosiglitazone or sulforaphane treatment, two respective PPARγ or Nrf2 activators, was also investigated.

Results: The CD36 receptor, mostly expressed by CD14(+) circulating monocytes, statistically correlated with increased infant birth weights. Interestingly, mRNA levels of the transcription factor Nrf2 and the enzyme HO-1 also correlated with lower parasitaemia and increased infant birth weight, while PPARγ mRNA levels did not. Finally, monocytes isolated from low infant birth weight pregnant women were capable of up-regulating CD36 via the Nrf2 pathway ex vivo.

Conclusions: Altogether these results suggest that Nrf2-driven CD36 and HO-1 expression on innate immune cells could contribute to a protective and detoxifying mechanism during PAM. More powered and mechanistical studies are however needed to strengthen the conclusions of this study.

Fig. 1

Relations between PAM, monocyte sub-populations and CD36 expression. a Gating strategy for the identification of monocyte sub-populations. Forward and side scattering parameters (FSC and SSC) were first used to gate monocytes among other leucocytes. Monocytes were sub-divided into CD14⁺ CD16⁻/CD14⁺ CD16⁺/CD14^low CD16⁺, according to CD14 and CD16 staining characteristics. Plots shown are representative examples. b Percentages of each monocyte sub-set found in PAM monocytes. c CD36 protein expression according to monocyte sub-populations in PAM. Geometric mean fluorescence intensity (MFI) was measured by flow cytometry for each monocyte sub-set. Values were compared by the Mann–Whitney U-test

Fig. 2

Relation between CD36 protein and mRNA expression and birth weight in PAM. a, c CD36 MFI and mRNA expression levels were compared according to birth weight threshold of 2.5 kg, by the two-factor covariance analysis (ANOVA). b, d CD36 MFI and mRNA expression levels were positively related to birth weight values by the Spearman correlation test

Fig. 3

Correlations between CD36, PPARγ, Nrf2 and HO-1 mRNA levels expressed by monocytes sampled in Beninese women presenting malaria at delivery. a Correlations between CD36/PPARγ, CD36/Nrf2 and PPARγ/Nrf2. b Correlations between HO-1/CD36, HO-1/PPARγ and HO-1/Nrf2. Statistical significance was tested by the Spearman correlation test. ***P < 0.0005, **P < 0.005, *P < 0.5

Fig. 4

Influence of Nrf2, PPARγ and HO-1 mRNA levels on clinical parameters in PAM. Correlations between Nrf2 or PPARγmRNA levels were related to peripheral (a) and placental parasitaemia (b). c HO-1 mRNA level correlation to birth weight. Values were compared by the Mann–Whitney U-test

Fig. 5

Capacity of monocytes to modulate CD36 expression in PAM. Monocytes were treated with sulforaphane (10 µM) and rosiglitazone (5 µM), two respective agonists of Nrf2 and PPARγ for 4 h before the detection of CD36 mRNA level by qRT-PCR. Left panel Monocytes sampled in women with low infant birth weights (≤2.5 kg) produced higher levels of CD36 mRNA after sulforaphane activation than monocytes related to high infant birth weight (>2.5 kg). Right panel Activation of these two groups of monocytes with rosiglitazone did not lead to any difference. Monocytes activated by sulforaphane were also compared to monocytes activated by rosiglitazone. No difference of capacity to produce CD36 mRNA was found, whatever the group of monocytes. Values were compared by the Mann–Whitney U-test

Fig. 6

Correlations between CD36, PPARγ, Nrf2, HO-1 mRNA levels and IL-10 and CD163 mRNA levels expressed by monocytes sampled in Beninese women presenting malaria at delivery. a Correlations to IL-10. b Correlations to CD163. Statistical significance was tested by the Spearman correlation test. ***P < 0.0005, **P < 0.005, *P < 0.5