Malaria systems immunology: Plasmodium vivax induces tolerance during primary infection through dysregulation of neutrophils and dendritic cells

Abstract

Objectives: To dissect the transcriptional networks underpinning immune cells responses during primary Plasmodium vivax infection of healthy human adults.

Methods: We conducted network co-expression analysis of next-generation RNA sequencing data from whole blood from P. vivax and P. falciparum controlled human malaria infection (CHMI) of healthy naïve and malaria-exposed volunteers. Single cell transcription signatures were used to deconvolute the bulk RNA-Seq data into cell-specific signals.

Results: Initial exposure to P. vivax induced activation of innate immunity, including efficient antigen presentation and complement activation. However, this effect was accompanied by strong immunosuppression mediated by dendritic cells via the induction of Indoleamine 2,3-Dioxygenase 1(IDO1) and Lymphocyte Activation Gene 3 (LAG3). Additionally, P. vivax induced depletion of neutrophil populations associated with down regulation of 3G-protein coupled receptors, CRXCR1, CXCR2 and CSF3R. Accordingly, in malaria-exposed volunteers the inflammatory response was attenuated, with a decreased class II antigen presentation in dendritic cells. While the immunosuppressive signalling was maintained between plasmodium species, response to P. falciparum was significantly more immunogenic.

Conclusions: In silico analyses suggest that primary infection with P. vivax induces potent immunosuppression mediated by dendritic cells, conditioning subsequent anti-malarial immune responses. Targeting immune evasion mechanisms could be an effective alternative for improving vaccine efficacy.

Keywords: Controlled human malaria infection; Dendritic cells; Immunoregulation; Malaria; Transcriptomics; Vaccines.

Fig. 1

Gene co-expression analysis (CEA) indicates development of immunosuppression during P. vivax infection in naïve individuals. (A) Visual representation of whole transcriptome analysis of whole blood from human naïve volunteers during the controlled human malaria infection (CHMI). The analysis identified eight main clusters. (BioLayoutExpress3D, r = 0.92, MCL = 1.7). Lines (edges) represent the similarity between transcript expressions; circles (nodes) represent transcripts. Clusters of co-expressed genes are coded by colour. Enrichment of gene ontology terms in clusters was done using ToppGene²⁶) Mean (± SEM) expression profiles for clusters 1–8, pre-challenge (pre, blue bars) and on diagnosis day (Dx, grey bar). (B) Mean (±SEM) expression profiles for clusters 1–8, pre challenge (blue bars) and at the diagnosis (grey bars). (For interpretation of the references to colour in this figure, the reader is referred to the web version of this article.)

Fig. 2

Changes in gene expression during malaria infection in malaria naive and malaria-exposed volunteers. (A) Comparison of differentially expressed genes induced during P. vivax infection in naïve vs malaria-experienced volunteers, red: up-regulated genes, blue: down-regulated genes. (DEG identification: EdgeR, FDR p value <0.05) (B) Overexpressed biological pathways (REACTOME database) identified using weighted gene-set enrichment analysis (WGSEA) comparing MN and ME individuals using a Kolmogorov–Smirnov non-parametric rank statistic with Benjamini and Yekutieli FDR multiple testing adjustment method (significance level was set at 0.05). Gene lists were ranked based on fold change with 1 × 10⁶ permutations. (For interpretation of the references to colour in this figure, the reader is referred to the web version of this article.)

Fig. 3

Bulk RNA sample deconvolution into specific cell proportions. Estimated proportions of B cells, monocytes, mDC, neutrophils, NK and T cells based on single-cell-specific gene expression (n = 5, unpaired t-test; *P < 0.05) in naïve (MN) and malaria-exposed (ME) pre (grey bars) and post (orange bars) exposure to P vivax. Mean of n = 6  ±  SEM shown. (For interpretation of the references to colour in this figure, the reader is referred to the web version of this article.)

Fig. 4

P. vivax induced immunosuppression is mediated by IDO1. Paired analysis of key regulatory genes during a P. vivax CHMI (n = 5, paired t-test; *P < 0.05). Normalised gene counts (scaled TPM) shown in naïve (MN) and malaria-exposed (ME) Pre (grey dots) and post (orange dots) exposure to P vivax. (For interpretation of the references to colour in this figure, the reader is referred to the web version of this article.)

Fig. 5

Distinct gene expression profiles in induced by different Plasmodium species Comparative analysis of selected immunomodulatory genes DEG during P. vivax and P. falciparum CHMI. Mean/median fold change in gene expression level between post and pre-exposure to P. falciparum and P. vivax shown for each gene. Genes down-regulated on exposure: blue, genes up-regulated on exposure: red. Black outline: significant FDR corrected p value. Green outline: neutrophil-associated genes, not detected in PBMC. Genes are ranked by fold change in P. falciparum CHMI. (For interpretation of the references to colour in this figure, the reader is referred to the web version of this article.)