Splenic clearance of rigid erythrocytes as an inherited mechanism for splenomegaly and natural resistance to malaria

Abstract

Background: In malaria-endemic areas, subjects from specific groups like Fulani have a peculiar protection against malaria, with high levels of IgM but also frequent anaemia and splenomegaly. The mechanisms underlying this phenotype remain elusive.

Methods: In a cohort study set up in Benin, West Africa, after a careful evaluation of malaria-related phenotypes, we measured the deformability of circulating erythrocytes in genetically distinct groups (including Fulani) living in sympatry, using ektacytometry and microsphiltration, a mimic of how the spleen clears rigid erythrocytes. Heritability of erythrocytes deformability was calculated, followed by a genome-wide association study (GWAS) of the same phenotype.

Findings: Compared to non-Fulani, Fulani displayed a higher deformability of circulating erythrocytes, pointing to an enhanced clearance of rigid erythrocytes by the spleen. This phenotype was observed in individuals displaying markers of Plasmodium falciparum infection. The heritability of this new trait was high, with a strong multigenic component. Five of the top 10 genes selected by a population structure-adjusted GWAS, expressed in the spleen, are potentially involved in splenic clearance of erythrocytes (CHERP, MB, PALLD, SPARC, PDE10A), through control of vascular tone, collagen synthesis and macrophage activity.

Interpretation: In specific ethnic groups, genetically-controlled processes likely enhance the innate retention of infected and uninfected erythrocytes in the spleen, explaining splenomegaly, anaemia, cryptic intrasplenic parasite loads, hyper-IgM, and partial protection against malaria. Beyond malaria-related phenotypes, inherited splenic hyper-filtration of erythrocytes may impact the pathogenesis of other hematologic diseases.

Funding: ANR, National Geographic Society, IMEA, IRD, and Région Ile-de-France.

Keywords: Erythrocytes; Ethnic groups; Falciparum; Genome-wide association study; Heritability; Malaria; Spleen; Splenomegaly.

Figure 1

Overview of experimental procedures. Study location is indicated in the upper right corner (Birni district, Atacora department, Benin). SAGM: saline, adenine, glucose, mannitol. Hct: haematocrit. Ektacytometry was performed on 212 samples because of a shortage of polyvinypyrrolidone. PCR was performed on 424 samples because whole blood on filter paper could not be obtained in 97 subjects.

Figure 2

Splenomegaly is more prevalent in Fulani than in subjects from other ethnic groups. (a-d): Prevalence (%) of (a) splenomegaly, (b) anaemia according to the WHO definition, (c) positive rapid diagnostic test for malaria and (d) positive thick smears in Fulani (F) and non-Fulani (NF). (e1)Plasmodium falciparum density (median, IQR) in Fulani and non-Fulani, shown separately in children (e2) and adults (e3) on thick smears (only positive values are shown). (f) Concentration of total IgM (median, IQR) in plasma from Fulani and non-Fulani. *: 0·01<P value<0·05; **: 0·001<P value<0·01. No symbol is shown if the difference is not significant. Categorical variables were compared through Chi-squared of Fisher's exact test, as appropriate; parasite densities and total IgM values were compared through Mann-Whitney test. For each parameter, the number of individuals on which data were available was: splenomegaly, 521; anaemia, 508; positive rapid diagnostic test, 521; positive thick smear, 521; total IgM measurement, 419.

Figure 3

Circulating RBC are more deformable in Fulani than in subjects from other groups, and markers of recent P. falciparum infection are associated with enhanced RBC deformability in Fulani. (a) Elongation index (EI) of RBC measured by rotational ektacytometry at 30 Pa in Fulani (F) and non-Fulani (NF) (n = 248). Higher EI values indicate enhanced RBC deformability. (b) Retention or enrichment rates (RER%) of circulating RBC from Fulani (F) and non-Fulani (NF) determined by filtration through microsphere layers (microsphiltration) (n = 463). Positive RER (i.e., enrichment in test RBC downstream from filter) indicate enhanced RBC deformability (n = 463). (c) RER determined by microsphiltration in Fulani (n = 131) and non-Fulani (n = 282) subjects with both RDT and thick smear negative for malaria parasites (RDT- TS-), or with both tests positive (RDT+ TS+) or with only RDT positive (RDT+ TS-). (d) RER values in Fulani (n = 136) and non-Fulani (n = 271) when considering subjects with negative or positive PCR for P. falciparum. Box plots indicate median and IQR values; bars indicate 5th and 95th percentiles. NS: no significant difference; *: 0·01<P value<0·05; **: 0·001<P value<0·01; ***: P value<0·001. Ektacytometry data were analyzed through Mann-Whitney test; Microsphiltration data were analyzed through Student's t-test and ANOVA, as appropriate.

Figure 4

(a) Heritability of splenic RBC filtration. Using a variance component approach, the heritability of the deformability of circulating red blood cells (RBC) was determined either in relation with ethnicity in Fulani (F) or non-Fulani (NF) (left) or in the whole population depending on the status of P. falciparum infection (right). Additive heritability is estimated as the ratio of the additive genetic variance component to the total phenotypic variance. P-values were estimated using likelihood ratio tests. n: number of subjects; *: 0·01<P value<0·05; **: 0·001<P value<0·01; ***: P value<0·001; TS: thick smear; RDT: malaria rapid diagnostic test. Analysis was performed on 420 subjects. (b): Genome-wide association analysis for RBC deformability. (n = 323) (b1) Each point of the Manhattan plot shows a single-marker variant, with genomic positions across chromosomes on the X-axis, and the association level corrected by genomic control on the Y-axis. The red dashed line shows the genome-wide significant P-value of 5 × 10^–8. (b2): Gene-based test as computed by MAGMA. Each point of the Manhattan plot indicates a gene. Input SNPs were mapped to 19185 protein-coding genes. Genome wide significance (red dashed line in the plot) was defined at P = 0·05/19185 = 2·606 × 10^–6. (c1) Determinants of retention upon splenic filtration of circulating RBC. (c2) Malaria-related phenotypic features observed in Fulani (grey boxes), potential mechanisms underlying the increased retention of circulating RBC in Fulani (red boxes), and genes potentially involved in these mechanisms as per literature analysis, the specific mechanisms being indicated in details in Supplementary Table 4 (Bold letters).