Plasmodium vivax infection compromises reticulocyte stability

Abstract

The structural integrity of the host red blood cell (RBC) is crucial for propagation of Plasmodium spp. during the disease-causing blood stage of malaria infection. To assess the stability of Plasmodium vivax-infected reticulocytes, we developed a flow cytometry-based assay to measure osmotic stability within characteristically heterogeneous reticulocyte and P. vivax-infected samples. We find that erythroid osmotic stability decreases during erythropoiesis and reticulocyte maturation. Of enucleated RBCs, young reticulocytes which are preferentially infected by P. vivax, are the most osmotically stable. P. vivax infection however decreases reticulocyte stability to levels close to those of RBC disorders that cause hemolytic anemia, and to a significantly greater degree than P. falciparum destabilizes normocytes. Finally, we find that P. vivax new permeability pathways contribute to the decreased osmotic stability of infected-reticulocytes. These results reveal a vulnerability of P. vivax-infected reticulocytes that could be manipulated to allow in vitro culture and develop novel therapeutics.

Fig. 1. Development of a flow cytometry osmotic stability assay.

a Flow cytometry method for measuring RBC osmotic stability. Created with BioRender.com. b and c Representative flow cytometry forward scatter (FSC) side scatter by (SSC) plot (b) and FSC by FITC-phalloidin plot of a lysed (120 mOsm) RBC sample (c). Data representative of five independent experiments. d Representative immunofluorescent images of FITC–phalloidin binding RBC ghosts. Scale bar, 10 μm. Arrows indicate a RBC undergoing lysis. Data representative of two independent experiments. e Representative hemoglobin (Hgb) absorbance (black line) and flow cytometry (red line) lysis curves for normal RBCs. Error bars represent SD of n = 3 technical replicates. Data fit with least-squares regression fit curves of normalized data. f Normal (n = 6) and hereditary xerocytosis (HX) (n = 3) RBC lysis₅₀ values measured by flow cytometry and hemoglobin absorbance lysis assays. Unique symbols indicate biological replicates and lines match the lysis₅₀ values obtained from flow cytometry and hemoglobin absorbance assays. n.s. indicates no significant difference in lysis₅₀ (normal RBC p = 0.8, HX RBC p = 0.1) using paired two-sided Student’s t-test. g Normal RBC (n = 12) lysis₅₀ values measured by flow cytometry. Unique symbols indicate biological replicates. Horizontal lines and error bars represent mean ± SEM. h Normal RBC (n = 6) lysis₅₀ values over the course of 14 days of 4 °C storage. Unique symbols indicate biological replicates. Data fit with linear regression lines. Mean linear regression for all data depicted by solid line. Pearson correlation of lysis₅₀ and days of 4 °C storage_, r = 0.15. i 4 °C degree stored (n = 3) and cryopreserved (n = 3) RBCs lysis₅₀ values following transfer to P. vivax in vitro culture conditions. Unique symbols indicate biological replicates. Horizontal lines and error bars represent mean ± SEM. Asterisks denote significant differences in lysis₅₀ during culture (4 °C degree stored RBCs p = 0.02, Cryopreserved RBCs p = 0.04) using RM one-way ANOVA analysis.

Fig. 2. Osmotic stability dynamics during erythroid differentiation and reticulocyte maturation.

a Flow cytometry method for measuring osmotic stability of RBC subpopulations. Created with BioRender.com. b Representative lysis curves for normocytes (DNA− RNA− CD71−, purple line), old reticulocytes (DNA− RNA+ CD71, green dashed line), young reticulocytes (DNA− RNA+ CD71+, blue solid line), and RBC precursors (DNA+ CD71+, orange dashed line) from a bone marrow aspirate measured by flow cytometry. Data fit with least squares regression fit curves of normalized data. c Normocyte (purple data points), old reticulocytes (green data points), young reticulocytes (blue data points), and RBC precursors (orange data points) from bone marrow aspirates lysis₅₀ values (n = 6). Unique symbol indicate biological replicates. Horizontal lines and error bars represent mean ± SEM. Asterisks denote significant differences in lysis₅₀ values (p = 0.0001) using RM one-way ANOVA analysis. d RBC progenitor lysis₅₀ values during course of CD34+ differentiation in vitro (n = 2). Unique symbols indicate RBC progenitor lysis₅₀ values at days 9, 11, 14, 17, and 20 from two independent differentiations. n.s. indicates no significant difference in lysis₅₀ during course of differentiation (p = 0.2) using Friedman test. Inset are representative photos of RBC progenitors during in vitro differentiation. Scale bar, 10 μm. Colored bars below images represent proportion of RBC progenitor developmental stages (gray—proeythroblast, green—basophilic normoblast, blue—polychromatic normoblast, purple—orthochromatic normoblast, red—reticulocyte) present at day-9, -11, -14, -17, and -20 of differentiation.

Fig. 3. Osmotic stability of P. vivax-infected reticulocytes.

a Flow cytometry strategy for measuring osmotic stability Plasmodium-infected RBCs. Created with BioRender.com. b Representative lysis curves for P. vivax-infected CD71+ (DNA+ CD71+, blue line) and CD71− reticulocytes (DNA+ CD71−, red line) and uninfected CD71+ reticulocytes (DNA− CD71+, gray line) and normocytes (DNA− CD71−, purple line) from a cryopreserved Brazilian clinical P. vivax sample after 1-h in vitro culture. Data fit with least-squares regression fit curves of normalized data. c Developmental stage of in vitro cultured ex vivo Indian (n = 2) and cryopreserved Brazilian (n = 5) P. vivax clinical isolates. Horizontal lines and error bars represent mean ± SEM. Inset are representative images of P. vivax IDC development stages. Scale bar, 10 μm. Gray (stage I), green (stage II), blue (stage III), red (stage IV), yellow (stage V), purple (gametocyte) bars indicate IDC development stage. d P. vivax-infected reticulocytes from cryopreserved Brazilian (black symbols) and non-cryopreserved Indian (blue symbols) clinical P. vivax samples lysis₅₀ values after 1-, 8-, 16-, 24-, and 44-h of in vitro culture. Unique symbols indicate biological replicates. Horizontal lines and error bars represent mean ± SEM. Dashed line (154 mOsm) and gray shading indicate lysis₅₀ threshold for hemolytic anemias. Hashtag and n.s., significant and no significant difference in cryopreserved Brazilian and non-cryopreserved Indian P. vivax lysis₅₀ values, respectively (1-h p = 0.05, 8-h p = 0.6, 24-h p = 0.8, 44-h p = 1.0) using unpaired two-tailed Mann–Whitney test. Astericks and n.s., significant and no significant difference between Brazilian P. vivax-infected reticulocytes lysis₅₀ at 24- and 1-, 8-, 16-, 24- and 44-h of culture, respectively (1-h p = 0.3, 8-h p = 0.02, 16-h p = 0.04, and 44-h p = 0.9) using unpaired two-tailed Dunnett’s multiple comparisons test. e Cryopreserved Brazilian P. vivax-infected CD71+ and CD71− lysis₅₀ values normalized to lysis₅₀ values of uninfected CD71+ reticulocytes at 8- (n = 3), 16- (n = 5), and 24-h (n = 5) of culture. Unique symbols indicate biological replicates. Horizontal lines and error bars represent mean ± SEM. Astericks and n.s., significant and no significant difference between 24-h and 8- and 16-h of culture lysis₅₀ values, respectively (CD71+: 8-h p = 0.04, 16-h p = 0.006 and CD71-: 8-h p = 0.1, 16-h p = 0.05) using unpaired two-tailed Student’s t-test.

Fig. 4. Osmotic stability of P. falciparum-infected normocytes.

a Developmental stage of in vitro-cultured, cryopreserved P. falciparum (clone 3D7 P2G12) (n = 5). Bars represent the mean ± SEM. Inset are representative images of P. falciparum IDC development stages. Scale bar, 10 μm. Gray (stage I), green (stage II), blue (stage III), red (stage IV), yellow (stage V) bars indicate IDC development stage. b P. falciparum-infected RBC lysis₅₀ values normalized to lysis₅₀ values of corresponding uninfected RBCs at 24- and 44-h of in vitro culture of cryopreserved P. falciparum (clone 3D7 P2G12) (n = 5). Unique symbols indicate biological replicates. Horizontal lines and error bars represent mean ± SEM. n.s., no significant difference between P. falciparum-infected RBC lysis₅₀ values at 24 and 44-h of culture (p = 0.1) using paired two-tailed Student’s t-test. c Lysis curves of 24-h (gray dotted line) (n = 7) and 44-h (gray solid line) (n = 6) cryopreserved Brazilian P. vivax-infected reticulocyte cultures and 24-h (black dotted line) (n = 5) and 44-h (black solid line) (n = 5) cryopreserved P. falciparum-infected normocytes (clone 3D7 P2G12) cultures. Unique symbols indicate biological replicates. Data fit with least-squares regression fit curves of normalized data.

Fig. 5. P. vivax new permeability pathways increase the permeability of P. vivax-infected reticulocytes.

Sensitivity of cryopreserved Brazilian P. vivax-infected reticulocytes to a d-sorbitol and b l-alanine lysis in the presence and absence of NPP inhibitor furosemide in 8-, 16-, 24-, and 44-h cultures (n = 3). Unique symbols indicate biological replicates. Horizontal lines and error bars represent mean ± SEM. Hashtag and n.s., significant and no significant difference between furosemide treated and untreated P. vivax-infected reticulocyte lysis respectively (sorbitol: 8-h p = 0.6, 16-h p = 0.02, 24-h p = 0.04, 44-h p = 0.02 and alanine: 8-h p = 0.5, 16-h p = 0.01, 24-h p = 0.03, 44-h p = 0.04) using paired two-tailed Student’s t-test. Asterisks, significant difference between P. vivax-infected reticulocyte lysis at 8-h and 16-, 24- and 44-h of culture (sorbitol: 16-h p = 0.01, 24-h p = 0.02, 44-h p = 0.004 and alanine: 16-h p = 0.003, 24-h p = 0.003, 44-h p = 0.005) using unpaired two-tailed Dunnett’s multiple comparisons test. c Model of the osmotic stability dynamics within the reticulocyte compartment and the impact of P. vivax infection on reticulocyte osmotic stability. Created with BioRender.com.