Activation of Drosophila hemocyte motility by the ecdysone hormone

Abstract

Drosophila hemocytes compose the cellular arm of the fly's innate immune system. Plasmatocytes, putative homologues to mammalian macrophages, represent ∼95% of the migratory hemocyte population in circulation and are responsible for the phagocytosis of bacteria and apoptotic tissues that arise during metamorphosis. It is not known as to how hemocytes become activated from a sessile state in response to such infectious and developmental cues, although the hormone ecdysone has been suggested as the signal that shifts hemocyte behaviour from quiescent to migratory at metamorphosis. Here, we corroborate this hypothesis by showing the activation of hemocyte motility by ecdysone. We induce motile behaviour in larval hemocytes by culturing them with 20-hydroxyecdysone ex vivo. Moreover, we also determine that motile cell behaviour requires the ecdysone receptor complex and leads to asymmetrical redistribution of both actin and tubulin cytoskeleton.

Keywords: Cell culture; Cell migration; Cytoskeleton dynamics; Drosophila; Ecdysone; Hemocytes.

Fig. 1.. Ecdysone causes a shift in hemocyte morphology and distribution.

(A) Images of the presumptive thorax/abdominal region with hemocytes labelled by PxnGAL4-UASGFP; CrqGAL4-UASGFP. At the Late L3 stage, hemocytes are attached to the larval integument and are localized to segmental patches and have rounded morphology (yellow dashed squares in the low-magnification top rows indicate the region magnified in the middle panels (2× digital zoom); in these medium-magnification middle row panels, the red dashed squares indicate the region magnified (2× digital zoom from middle panels) in the high-magnification panels in the bottom row). At 1 h APF, the hemocytes have started to migrate out of these patches and begin to assume polarized morphology (middle panel, yellow arrow head). The zoomed image presented in the yellow box, under 1 h APF, was taken from the last frame of a live time-lapse and at the most ventral position within the Z-stack. This image was chosen as the best representative of hemocyte patch dispersal. At 2 h APF, hemocytes have migrated laterally from the dorsal patches as well as detached from the integument, as they are no longer in view at the dorsal plane of focus. Most of the population has adopted a spindle/polarized shape (bottom panel). At 24 h APF the hemocytes have colonized most of the thorax and abdomen and display vacuolation associated with phagocytosis of larval tissues and secretion of ECM. These in vivo morphologies are akin to those found in ex vivo culture (see B). (B) A histogram showing the percentages of cellular morphologies and behaviours between LIII (−ecdysone) (blue bars), WPP (red bars), and LIII +ecdysone (yellow bars) hemocyte populations ex vivo. Black bar with a star (*) represents significant difference. Sample images of each morphological class are shown at the bottom.

Fig. 2.. Ex vivo ecdysone hormone induces motile membrane behaviour in early LIII hemocytes.

(A) A series of fixed point track plots from LIII, WPP, and LIII +ecdysone hemocyte populations in ex vivo cultures. Both X and Y axis represent distance in µm. (B) A collage of sample WPP and LIII +ecdysone hemocytes ex vivo. The box to the right contains an outline of each still overlaid with directional arrows; circles, that are colour coded corresponding to each time point in the collage, indicates the approximate centroid region of the cell. (C) Histogram showing mean migration velocity and standard deviation exhibited across the hemocytes analysed.

Fig. 3.. Cytoskeletal analysis showing actin and microtubule distribution in LIII, WPP, LIII +ecdysone and LIII EcRB2_(F645A) +ecdysone hemocytes.

Fixed hemocyte images, taken at 63× magnification, showing DAPI, F-actin, and β-tubulin staining. DAPI (blue) labels the nucleus of the cell, phalloidin (red) labels polymerised actin (F-actin), and β-tubulin (green) labels the microtubules of the cell; merged images on the right. Representative hemocytes were selected from Or-R LIII, Or-R WPP, Or-R LIII + ecdysone, and He_GAL4 UAS-EcRB2_(F645A) LIII + ecdysone. Arrows point out the presence of asymmetrical actin and microtubule cytoskeletal sub-units in polarised hemocytes. Scale bar is to 10 µm.

Fig. 4.. The ecdysone receptor is required for hemocyte motility.

(A) Top left and right panels show a pooling of UAS-EcR-dominant-negative control tracks, whilst bottom left and right show the respective experimental results when these constructs, UAS-EcRB2_(F645A) and UAS-EcRA_(W650A), are expressed by the hemocyte specific driver HmL_GAL4 (see supplementary material Fig. S2 for the He_GAL4 driven UAS-EcRB2_(F645A) track plot). (B) Percentages of hemocytes that exhibited either a symmetrical or an asymmetrical cytoskeletal phenotype amongst the four groups tested previously shown in Fig. 3 – Or-R LIII (blue bars), Or-R WPP (Red bars), Or-R LIII +ecdysone (yellow bars) and He_GAL4 driven UAS-EcRB2_(F645A) LIII +ecdysone (Purple bars). ‘N’ indicates the number of hemocytes tested in each sample population, ‘S’ indicates symmetry, ‘A–S’ asymmetry and the (*) indicates significant differences between the bracket bars. (C) UAS-EcRB2_(F645A) expressed in vivo in hemocytes precludes the changes in morphology and distribution observed in wild-type hemocytes at WPP (compare with second and third rows of Fig. 1A).