K562 erythroleukemia line as a possible reticulocyte source to culture Plasmodium vivax and its surrogates

Abstract

Establishing an in vitro "red blood cell matrix" that would allow uninterrupted access to a stable, homogeneous reticulocyte population would facilitate the establishment of continuous, long-term in vitro Plasmodium vivax blood stage cultures. In this study, we have explored the suitability of the erythroleukemia K562 cell line as a continuous source of such reticulocytes and have investigated regulatory factors behind the terminal differentiation (and enucleation, in particular) of this cell line that can be used to drive the reticulocyte production process. The Duffy blood group antigen receptor (Fy), essential for P. vivax invasion, was stably introduced into K562 cells by lentiviral gene transfer. miRNA-26a-5p and miRNA-30a-5p were downregulated to promote erythroid differentiation and enucleation, resulting in a tenfold increase in the production of reticulocytes after stimulation with an induction cocktail compared with controls. Our results suggest an interplay in the mechanisms of action of miRNA-26a-5p and miRNA-30a-5p, which makes it necessary to downregulate both miRNAs to achieve a stable enucleation rate and Fy receptor expression. In the context of establishing P. vivax-permissive, stable, and reproducible reticulocytes, a higher enucleation rate may be desirable, which may be achieved by the targeting of further regulatory mechanisms in Fy-K562 cells; promoting the shift in hemoglobin production from fetal to adult may also be necessary. Despite the fact that K562 erythroleukemia cell lines are of neoplastic origin, this cell line offers a versatile model system to research the regulatory mechanisms underlying erythropoiesis.

Figure 1

Generation of Fy-transduced K562 variants. (A) Schematic of the SEW-Fy vectors used to introduce the different Fy-variants into the K562 erythroleukemia cell line. (B) Confirmation by flow cytometry using an allophycocyanin (APC)-labeled anti-Fy antibody that the Fy receptor is expressed on the surface of all Fy-transduced K562 variants (red), not transduced control K562 cells (blue).

Figure 2

Selection of Fy-transduced K562 variants. (A) Results of the binding experiments with the different Fy-transduced K562 variants using fixed K562 monolayers and free P. cynomolgi merozoites (one representative experiment of six is shown), each Fy variant was plated in quadruplicates (vertical bars indicate standard deviations). (B) Invasion of P. cynomolgi merozoites into the Fy^b-long-K562 variant; the red arrows indicate the different moments of parasite binding (at 20 and 30 min) and invasion/growth inside the cell (at 8 and 29 h). (C) The response to the induction of differentiation of Fy^b-long-transduced K562 cells using SAHA is clone (Nos. 1 and 2) dependent. Neg=non-transduced control K562 cells.

Figure 3

Effect of miRNA downregulation on the proliferation rate and production of hemoglobin of Fy-K562. Fy^b-longK562 cells were transduced with the empty locker vector, the locker-miR-26a-5p and the locker miR-30a-5p, the locker-miR-26a-5p or the locker miR-30a-5p. (A) The proliferation rate of these different Fy^b-longK562 cell lines are compared under four different conditions: untreated (black solid line), treated with macrophages (gray solid line), treated with 6 nmol/L MIT (red solid line), and treated with macrophages and 6 nmol/L MIT (blue solid line). Data are presented as means ± SEM. Tests were performed two-sided. *p < 0.05 (Mann–Whitney U test). N = 4 independent experiments. (B) K562 cell pellets at different time points of differentiation under four different conditions: untreated (1), treated with 6 nmol/L MIT (2), treated with macrophages (3), and treated with macrophages and 6 nmol/L MIT (4). One of four representative experiments is shown.

Figure 4

Effect of miRNA downregulation on the production of hemoglobin by benzidine staining. (A,B) Fy^b-longK562 cells either double transduced with the locker miR-26a-5p and the locker miR-30a-5p vectors (A) or transduced with the empty locker control vector (B) were cultured under four different conditions for 35 days. Every 7 days, 3 × 10⁵ cells were stained with benzidine, and cytospins were performed. Bar = 20 µm. One of four representative experiments is shown.

Figure 5

Effect of miRNA down regulation on the mRNA expression levels of hemoglobin chains. (A–D) Real-time PCR expression analysis of α- and γ-globin chain mRNA in Fy^b-longK562 cells transduced with the empty locker vector (A), locker-miR-26a-5p and locker miR-30a-5p (C26) (B), locker-miR-26a-5p (C), or locker miR-30a-5p (D) vector. Cells were cultured under four conditions. Relative fold changes in expression (normalized to GAPDH) were calculated by the ΔΔCT method and values are expressed as 2^–ΔΔCT (n = 4 independent experiments). Data are presented as means ± SEM. Tests were performed two-sided. *p < 0.05 (Mann–Whitney U test).

Figure 6

Effect of miRNA downregulation on the expression of different surface markers. (A–D) Fy^b-longK562 cells were transduced with the empty locker vector, C26, locker miR-30a-5p, or locker-miR-26a-5p, and cultured under four conditions for 35 days. Every 7 days, flow cytometric analysis of CD45 (A), CD71 (B), CD235a (C), and DARC receptor (D) were performed. Data are presented as means ± SEM. Tests were performed two-sided. *p < 0.05 (Mann–Whitney U test). N = 4 independent experiments.

Figure 7

Effect of miRNA downregulation on the morphology of K562 cells during erythroid differentiation. (A,B) Fy^b-longK562 cells either double transduced with locker-miR-26a-5p and locker-miR-30a-5p vectors (clone 26) (A) or transduced with the empty locker control vector (B) were cultured under four conditions for 35 days. Every 7 days, cytospins of 3 × 10⁵ cells were performed and stained with Giemsa solution. Bar = 20 µm. One of four representative experiments is illustrated.