Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2014 May 1;123(18):e100-9.
doi: 10.1182/blood-2013-12-541698. Epub 2014 Mar 20.

Glycophorin C (CD236R) Mediates Vivax Malaria Parasite Rosetting to Normocytes

Affiliations
Free PMC article

Glycophorin C (CD236R) Mediates Vivax Malaria Parasite Rosetting to Normocytes

Wenn-Chyau Lee et al. Blood. .
Free PMC article

Erratum in

  • Blood. 2015 Dec 17;126(25):2765

Abstract

Rosetting phenomenon has been linked to malaria pathogenesis. Although rosetting occurs in all causes of human malaria, most data on this subject has been derived from Plasmodium falciparum. Here, we investigate the function and factors affecting rosette formation in Plasmodium vivax. To achieve this, we used a range of novel ex vivo protocols to study fresh and cryopreserved P vivax (n = 135) and P falciparum (n = 77) isolates from Thailand. Rosetting is more common in vivax than falciparum malaria, both in terms of incidence in patient samples and percentage of infected erythrocytes forming rosettes. Rosetting to P vivax asexual and sexual stages was evident 20 hours postreticulocyte invasion, reaching a plateau after 30 hours. Host ABO blood group, reticulocyte count, and parasitemia were not correlated with P vivax rosetting. Importantly, mature erythrocytes (normocytes), rather than reticulocytes, preferentially form rosetting complexes, indicating that this process is unlikely to directly facilitate merozoite invasion. Although antibodies against host erythrocyte receptors CD235a and CD35 had no effect, Ag-binding fragment against the BRIC 4 region of CD236R significantly inhibited rosette formation. Rosetting assays using CD236R knockdown normocytes derived from hematopoietic stem cells further supports the role of glycophorin C as a receptor in P vivax rosette formation.

Figures

Figure 1
Figure 1
Experimental overview. Flowchart showing the summary of methodology applied in this study, and the respective results (figures) are shown in boxes with dotted lines. Note that *P < .05 and **P < .001 indicate these isolates are the same, and used for more than 1 experiment.
Figure 2
Figure 2
Rosetting kinetics. (A) Plot showing the kinetics of rosetting development in 47 P vivax isolates matured ex vivo. (B) Representative images of rosettes formed by different stages of P vivax and P falciparum are shown after Giemsa subvital staining process.
Figure 3
Figure 3
Factors affecting rosetting formation in P vivax and P falciparum. (A) Comparison of rosetting rate between P vivax and P falciparum isolates from the Thailand-Myanmar border. The median percentage of P vivax IRBCs is significantly higher than P falciparum; P (2-tailed) < .001. (B) Rosetting rate of P vivax and P. falciparum isolates found in different human ABO blood groups. No significant association was found between ABO phenotype of the malaria and occurrence of P vivax (P = .28) or P falciparum (P = .20). (C) Plot of rosetting rate of P vivax and P falciparum isolates against the original parasitemia of malaria patients presenting for treatment. No significant correlation was observed between patient parasitemia and rosetting for P vivax (Spearman r = −0.10; 95% CI [−0.39 to 0.20]; P = .50) or P falciparum (Spearman r = −0.06; 95% CI [−0.38 to 0.27]; P = .71). (D) Rosetting rate of P vivax and P falciparum isolates against peripheral reticulocyte counts in malaria patients. No significant correlation was observed between rosetting and patient reticulocyte counts for either P vivax (Spearman r = 0.33; 95% CI [−0.04 to 0.62]; P [2-tailed] = .07) or P falciparum (Spearman r = −0.13; 95% CI [−0.49 to 0.28]; P = .52).
Figure 4
Figure 4
P vivax rosetting preference. (A) Percentage of rosettes associated with at least 1 reticulocyte in P vivax and P falciparum isolates in patient isolates already containing late-stage parasites (no ex vivo maturation). (B) Differences in rosetting rate of P vivax and P falciparum isolates in environment with <1% reticulocytes and approximately 50% reticulocytes (achieved by concentrating host reticulocytes on a 75% Percoll gradient). Lines connect paired observations. Altering the reticulocyte concentration had little effect on P vivax rosette formation; however, the number of normocytes attached to the P vivax IRBC was notably reduced in the treatment of enriched reticulocytes. Interestingly, an increase in the available reticulocytes reduced the ability of P falciparum to rosette (P < .01). (C) Reticulocytes were rarely associated with rosettes irrespective of the group or species observed. Here, a single normocyte is rosetting on a P vivax schizont (red arrow). The yellow arrows point to Heilmeyer stage IV reticulocytes subvitally stained with Giemsa. The tip of a glass micropipette with an internal diameter of 6 um is shown for scale.
Figure 5
Figure 5
Antibody blocking rosette formation. (A) Schematic diagram of the target sites of anti-CD236R antibody clone BRIC 10 and anti-CD236R antibody clone BRIC 4 on human CD236R structure. The target site of trypsin on this sialoglycoprotein is shown by the black arrow. (B-C) Rosetting inhibition in P vivax and P falciparum caused by Fab fragments specifically targeting the BRIC 10 and BRIC 4 locations on CD236R. BRIC 4 showed a significant reduction in rosetting of P falciparum isolates (P < .0001) and P vivax isolates (P < .0001) studied. A paired Student t test was conducted. (D) Rosetting inhibition by mouse anti-human CD35 antibody. Unlike in P vivax, this antibody significantly reduced rosetting rate of P falciparum isolates tested (P < .0001). (E) Comparison of rosetting rates between the control and cells incubated with Fab fragments of mouse anti-human glycophorin A antibody from P falciparum and P vivax isolates recruited. There was no significant difference between the control group and the “anti-glycophorin A” group in P vivax and P falciparum isolates studied.
Figure 6
Figure 6
Transgenic approach to investigate the role of CD236R in P vivax rosetting. (A) Experimental design of schizont P vivax rosetting assay with cultured RBCs (cRBCs) generated from CD34+ hematopoietic stem cells. The stable knockdown of CD236R (glycophorin C) is obtained using Sigma lentivector with GFP and shRNA against CD236R cassette expression. The 2 subsets of cells: CD236R knockdown and CD236R+ cells were separated by flow cytometry using GFP expression, 1 day before performing the rosetting assay with P vivax schizonts isolated by magnetic sorting. (B) Flow cytometry histograms showing CD236R expression in GFP-positive cells (CD236R knockdown cells [green line]), GFP-negative (CD236R+ cells [black line]), and unstained cells (gray line). (C) Plots showing schizont P vivax rosetting with cRBC with CD236R knockdown (CD236R KD) and without knockdown (CD236+) with bright field (BF), Hoechst, and GFP detection. (D) Frequency of schizonts with or without rosetting in the presence of CD236R knockdown cRBCs or of CD236+ cRBCs showing a significant difference in proportion of schizonts able to form rosettes between the 2 different types of cRBCs (CD236R KD and CD236+).

Similar articles

See all similar articles

Cited by 17 articles

See all "Cited by" articles

Publication types

MeSH terms

Feedback