Differentiation of HL-60 cells into primed neutrophils for the evaluation of antiapoptotic effects of poorly soluble nanoparticles

Abstract

Background: Neutrophil apoptosis is an important determinant of intensity and duration of neutrophilic inflammation. The interaction of poorly soluble nanoparticles with primed neutrophils can reduce their natural apoptosis rates. This reaction may contribute to pathogenic consequences of increased neutrophilic inflammation. Toxicological studies aiming to identify hazards of such materials with primary neutrophils are however challenging due to the short life span of these cells and a high donor to donor variability. Our purpose was the establishment of a culturable neutrophil-like cell line as a suitable model for studies of antiapoptotic effects of poorly soluble combustion-derived environmental nanoparticles. Therefore, differentiation protocols for the myeloid HL-60 cell line based on commonly used differentiation inducers all-trans retinoic acid (ATRA) and dimethyl sulfoxide (DMSO) were established and compared.

Results: The data demonstrate that only a combined cell treatment with ATRA and DMSO for a period of 5 days leads to the complete HL-60 differentiation with the typical neutrophil morphology and characteristic features of neutrophil maturation like cell cycle arrest, increase in differentiation marker CD11b, loss of proliferation marker CD71, and increased phagocytic capacity. Exposure of cells differentiated with ATRA + DMSO to carbon nanoparticles or proinflammatory cytokine granulocyte-macrophage colony-stimulating factor (GM-CSF) revealed a delay of apoptosis causally linked to intracellular reactive oxygen species (ROS). These data verified our earlier findings with human peripheral primed neutrophils from donors with slightly elevated proinflammatory blood plasma factors. Moreover, completely differentiated HL-60 cells possessed similar levels of L-selectin CD62L as neutrophils with primed immunophenotype, thus representing the biology of primed inflammatory neutrophils.

Conclusion: Neutrophil-like HL-60 cells differentiated according to our protocol could be an appropriate substitute cell line model for studies on the effects of inhalable nanomaterials on primed inflammatory neutrophils like lung neutrophils. For such toxicological studies our cell model is preferable to peripheral neutrophils, as blood neutrophils not always occur in a primed state and primed lung neutrophils from donors are not available for this purpose.

Fig 1. Effect of ATRA and DMSO on proliferation and cell cycle/apoptosis of HL-60 cells.

a The time courses of HL-60 cell density after starting the differentiation with 1 µM ATRA or/and 1% DMSO, monitored for a period of 5 days. Data are presented as mean ± SEM, n ≥ 19, * p ≤ 0.05. b Distribution of the cell cycle phases (G0/G1, S, and G2) and apoptosis (sub-G1) after induction of differentiation with ATRA or/and DMSO before and after 5 days of treatment. Data are presented as mean ± SEM, n ≥ 9. c, d The time courses of S phase (%S) and apoptosis (%sub-G1) after starting the differentiation with ATRA or/and DMSO, monitored for a period of 5 days. Data are presented as mean ± SEM, n ≥ 15. e Representative gating strategy and histograms from typical flow cytometry measurement of the cell cycle. f Representative histogram for determination of DNA content and different cell cycle phases in undifferentiated and differently treated HL-60 cells.

Fig 2. Effect of ATRA and DMSO on morphology and phagocytic capacity of HL-60 cells.

a HL-60 cells were treated with 1 µM ATRA and 1% DMSO alone and in combination for 5 days and morphology was assessed after May-Gruenwald-Giemsa staining. Representative images are shown. b Phagocytic capacity of undifferentiated, differently treated HL-60 cells at day 5 of differentiation protocol, and freshly isolated human peripheral blood neutrophils (PMN). Data are presented as mean ± SEM, n ≥ 3, * p ≤ 0.05. c Representative dot plots from typical flow cytometry measurement for determination of pHrodo positive, phagocytic cells.

Fig 3. Changes in cell surface markers upon differentiation of HL-60 cells.

a, b The time courses of CD11b and CD71 positive cells after starting the differentiation with 1 µM ATRA or/and 1% DMSO, monitored for a period of 5 days. c Representative gating strategy and histograms from typical flow cytometry measurements for determination of CD11b/CD71 positive HL-60 cells. d, e Representative histograms from typical flow cytometry measurements for determination of CD16 (FcγRIII) and CD66b expression in HL-60 cells and freshly isolated human peripheral blood neutrophils (PMN). f, g Changes in CD64 (FcγRI) and CD32 (FcγRII) positive cells after induction of differentiation with ATRA or/and DMSO before and after 5 days of treatment. Data are presented as mean ± SEM, n ≥ 3, * p ≤ 0.05.

Fig 4. Changes in CD62L antigen upon differentiation of HL-60 cells.

a The time courses of CD62L positive cells after starting the differentiation with 1 µM ATRA or/and 1% DMSO, monitored for a period of 5 days. b CD62L staining on freshly isolated human peripheral blood neutrophils (PMN) and fully differentiated neutrophil-like HL-60 cells (combined treatment ATRA + DMSO day 5). The data from human neutrophils (see Fig 4b) are originated from the earlier publication [21] in order to demonstrate the similarity of L-selectin levels in human primed responder neutrophils and differentiated cultured HL-60 cells. Data are presented as mean ± SEM. a n ≥ 3, b primed PMN n = 42, non-primed PMN n = 12, diff. HL-60 cells n = 11. * p ≤ 0.05.

Fig 5. Effect of granulocyte-macrophage colony-stimulating factor (GM-CSF) and carbon nanoparticles (CNP) on apoptosis and ROS production in HL-60 cells.

Undifferentiated and differently treated HL-60 cells (day 5) or freshly isolated human peripheral blood neutrophils (PMN) at the density of 2*10⁶ cells/ml were exposed to 20 ng/ml GM-CSF or 33 µg/ml CNP. Controls were cells before treatment (ctrl. 0 h) and incubated untreated cells (ctrl. 1 h/18 h). a, d DCF fluorescence determined at 1 h after exposure. b, e Apoptosis (%hypodiploidy or %sub-G1) determined at 18 h after exposure. c Apoptosis (% total Annexin binding) determined at 18 h after exposure. The data from human neutrophils (see Fig 5d and 5e) are originated from the earlier publication [21] in order to demonstrate the similarity of ROS levels and apoptosis behaviour in response to proinflammatory cytokine GM-CSF and carbon nanoparticles in human primed responder neutrophils and differentiated cultured HL-60 cells. Data are presented as mean ± SEM. a n ≥ 4, b n = 4, c n = 6, d PMN n = 77, diff. HL-60 cells n = 44, e PMN n = 123, diff. HL-60 cells n = 42. * p ≤ 0.05.