Temperature-sensitive migration dynamics in neutrophil-differentiated HL-60 cells

Abstract

Cell migration plays an essential role in wound healing and inflammatory processes inside the human body. Peripheral blood neutrophils, a type of polymorphonuclear leukocyte (PMN), are the first cells to be activated during inflammation and subsequently migrate toward an injured tissue or infection site. This response is dependent on both biochemical signaling and the extracellular environment, one aspect of which includes increased temperature in the tissues surrounding the inflammation site. In our study, we analyzed temperature-dependent neutrophil migration using differentiated HL-60 cells. The migration speed of differentiated HL-60 cells was found to correlate positively with temperature from 30 to 42 °C, with higher temperatures inducing a concomitant increase in cell detachment. The migration persistence time of differentiated HL-60 cells was higher at lower temperatures (30-33 °C), while the migration persistence length stayed constant throughout the temperature range. Coupled with the increased speed observed at high temperatures, this suggests that neutrophils are primed to migrate more effectively at the elevated temperatures characteristic of inflammation. Temperature gradients exist on both cell and tissue scales. Taking this into consideration, we also investigated the ability of differentiated HL-60 cells to sense and react to the presence of temperature gradients, a process known as thermotaxis. Using a two-dimensional temperature gradient chamber with a range of 27-43 °C, we observed a migration bias parallel to the gradient, resulting in both positive and negative thermotaxis. To better mimic the extracellular matrix (ECM) environment in vivo, a three-dimensional collagen temperature gradient chamber was constructed, allowing observation of biased neutrophil-like differentiated HL-60 migration toward the heat source.

Figure 1

Temperature-dependent attachment and migration dynamics of differentiated HL-60 cells. (a) Phase contrast images of control cell tracks at 30 °C (hypothermic), 37 °C (normal) and 42 °C (hyperthermic). A large number of cells detached under hyperthermic conditions. Scale bars: 100 µm. (b) Phase contrast images of cells at control conditions at 37 °C, heated at 42 °C and cooled back to 37 °C. Detached cells did not appear to be dead or apoptotic, as reduction of the temperature back to 37 °C resulted in cell reattachment to the substrate. Scale bars: 200 µm. (c) Cell attachment at different seeding conditions. In control samples under hyperthermic conditions, ~ 50% of the initially attached cells detached. 100 µg/ml bovine fibronectin (bFN) coating did not improve cell attachment compared to the control condition (no coating, 10% FBS), whereas a fivefold increase in FBS (50% FBS) in the media resulted in a significantly higher (~ 20%) number of attached cells throughout the entire (30–42 °C) temperature range. (d) The average cell speed ± 95% CIs of differentiated HL-60 cells at each experimental temperature. Cell speed increased linearly with increasing temperature for all conditions. Cells seeded in the presence of 50% FBS did not show any significant increase in average speed. 100 µg/ml bFN surface coating significantly reduced average cell speed at normal and hyperthermic temperatures. 39 and 78 cells were tracked for all conditions in hypothermic and normal/hyperthemic temperature ranges, respectively. At 39 and 40 °C on bFN coating, 117 cells were tracked. (e) The number of detached cells, as measured by automatically identified rounded cells, as a function of temperature. Data was captured every 10 s for 2.5 h. For (a–d) 2 independent experiments were performed. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001, N.S. not significant.

Figure 2

Temperature- and ECM-dependent direction correlation of differentiated HL-60 cell migration. (a–c) Direction autocorrelation analysis of differentiated HL-60 cells under 3 different conditions (control, 50% FBS and bFN coating) and 9 different temperatures between 30 and 42 °C. At hypothermic temperatures, directional autocorrelation peaks for all conditions. On bFN coating, cells maintain directionality for longer time periods. Error bars represent S.E.M. (d) Persistence length (µm) of differentiated HL-60 cells was calculated by multiplying the persistence time by average cell speed. The persistence time was evaluated by fitting Eq. (1) to the autocorrelation curves. At normal and hypothermic conditions, the persistence length was preserved. Error bars represent 95% CIs. (e) Average cell speed of differentiated HL-60 cells without coating (control), on 100 µg/ml bovine fibronectin coating (bFN coating), and on heat-treated bovine fibronectin coating (HT bFN coating). Heat treatment of the coating significantly reduced average cell speed. (f) Direction autocorrelation of differentiated HL-60 cells on control, bFN and HT bFN coatings at 37 °C. Heat treatment of bovine fibronectin did not influence cell persistence. Error bars are S.E.M. For all: 2 independent experiments were performed. **P < 0.01, ***P < 0.001, ****P < 0.0001.

Figure 3

Differentiated HL-60 cell thermotaxis in response to temperature gradients. (a) Schematic representation of the microfluidic temperature gradient chamber with cross sections in X and Y directions. Copper reservoirs were used as a heat source and a heat sink, with their temperatures held constant at 65 °C and 5 °C, respectively. (b) 3D Volume temperature gradient distribution computed by COMSOL Multiphysics. A 27–43 °C temperature gradient is generated along the 2 mm wide microchannel. (c) Temperature gradient effects on cell adhesion and migration. After application of the gradient, ~ 40% of cells remained attached. 72% of the attached cells were non-mobile. Of the mobile cells, approximately 50% migrated towards the heat sink (negative thermotaxis) and 40% migrated towards the heat source (positive thermotaxis). (d) Differentiated HL-60 cell speed in the presence of the temperature gradient. 25 cells in each direction were tracked. No significant differences in cell speed were found between those moving towards the heat source and those toward the heat sink, but both were approximately 50% slower than those in a constant 37 °C environment. (e) Migration tracks of 50 cells migrating both towards and away from the heat source. Cells that migrated at an angle of 60˚ or less with respect to temperature gradient direction (in this case the Y axis) are considered to have directed migration towards either the heat source or sink (red and blue triangles). 10% of cells showed non-directed migration. (f) Polar histogram of the relative angles from cell step to cell step calculated for 50 cells tracked under temperature gradient. Small steps less than 2 µm were disregarded. For all: 2 independent experiments were performed. Error bars represent 95% CIs of the data. ****P < 0.0001, N.S. not significant.

Figure 4

Fluid shear flow of a similar magnitude to that produced by temperature gradients does not induce directed migration. (a) Cell speed of differentiated HL-60 cells in the presence of a temperature gradient and under laminar flow of 30 and 40 µL/h. The average speed of 50 cells in total was evaluated for each condition. In the 2 temperature gradient chamber, 25 cells in each direction were tracked at a sampling interval of 15 s. No significant difference between migration speed of cells moving towards either heat source or sink was found by Welch’s t-test. When compared with cell speed in a constant 37 °C environment (sampling interval 10 s), cell speed was reduced by approximately 50%. Application of shear flow also resulted in significant cell speed reduction. Lower cell velocities could partially be an artifact of the lower cell tracking rate of 60 s, but the significant differences in cell speed under lower and higher flow rates indicate that cell migration speed has an inverse relationship to flow rate. (b) Migration tracks and polar histograms of θ for 50 cells migrating under laminar flow conditions. Application of 30 and 40 µL/hour flow rates did not result in directional cell migration.

Figure 5

Differentiated HL-60 cell thermotaxis in response to temperature gradients in 3D microchambers. (a) Schematic representation of the 3D microfluidic temperature gradient chamber with cross sections in X and Y directions. Copper reservoirs were used as a heat source and a heat sink, with their temperatures held constant at 65 °C and 5 °C, respectively. The microchamber was filled with a collagen scaffold containing cells. (b) Migration tracks and polar histograms of θ for 20 cells each, migrating in 3D collagen matrix with and without temperature gradients. Application of the temperature gradients in 3D revealed a higher percentage of cells migrating towards the heat source. (c) Cell speed of differentiated HL-60 cells in a 3D collagen matrix at a control temperature of 37 °C and in the presence of the temperature gradient. The average speed of 20 cells in each condition was evaluated. For all: Error bars represent 95% CIs of the data. 2 independent experiments were performed. N.S. not significant.