Fig 1.
E. histolytica adhesion forces are stronger on FN than on glass.
A. Schematic representation of a laminar flow chamber. B. Representative images of adherent amoeba on non-coated glass and on FN-coated glass in the absence of flow (T0) and at a flow rate of 350 μL/s (T 62.5s). scale bar: 40 μm. C. D. Amoebae adhering to non-coated glass (blue curve) or FN-coated glass (red curve). C. The percentage of bound cells as a function of the flow rate. The number of cells was counted in every 10th image of the video, and the mean rupture force was determined from five independent runs on non-coated glass (n cells = 181) and two runs on FN-coated glass (n cells = 37). D. The amoebae’s adhesion forces were calculated from the friction force when the interaction with the glass surface was broken. Amoeba were considered as half-spheres with a mean diameter of 30 μm on FN-coated glass and 15 μm on non-coated glass. Differences were analyzed using an unpaired t-test (p = 0.0008, ***). Videos of two representative runs on each substrate are provided as supplementary information (S1 Video).
Fig 2.
Morphological diversity of F-actin-containing adhesive structures formed by E. histolytica on an FN substrate.
A. Actin structures found within amoeba adhering to FN-coated or non-coated glass. Fixed amoebae were labelled for F-actin (red) and Arp3 (green) in the first and second panels and for F-actin (red) and paxillin (green) in the third and fourth panels. The first and third rows correspond to two-dimensional confocal microscopy images of the merging of brightfield and the projection of z-stack fluorescent images. The Pearson’s coefficients for the correlation between F-actin and Arp3 in each representative structure through the Z-stack was 0.67 for the adhesion plate (AP), 0.90 for the podosome-like (PL) structure in a ring, and 0.76 for PL structure in a cluster; for F-actin and paxillin was 0.79 for AP, 0.73 for PL structure in a ring and 0.87 for PL structure in a cluster. The second and fourth rows show the three-dimensional reconstructions of a z-stack of images of the above structures. The xy axis (in red) indicates the ventral side of the cell adhering to the substrate. B. Quantification of the F-actin structures present in E. histolytica seeded on non-coated glass or FN-coated glass. The proportions of amoebae showing adhesive plates (APs), rings of dots (RDs), and clusters of dots (CDs) are shown. The left panel indicates the proportion of amoeba containing APs and actin dots (ACs) on non-coated glass (G) and the proportion of amoeba containing APs and PLs on FN-coated glass (F). The percentages of amoeba showing an actin structure in one image were expressed in a box plot: n = 113 cells on non-coated glass and n = 145 cells on FN-coated glass; n = 503 dots on non-coated glass and n = 1145 dots on FN-coated glass. The results of Mann-Whitney tests are indicated. ns: not significant. ****P<0.0001; ***P<0.001.
Fig 3.
Differential effects of inhibitors (Y27632, CK666 and wortmannin) on the F-actin adhesive structures formed by E. histolytica.
Amoebae migrating on non-coated glass or FN-coated glass were treated with the following compounds: (A) 5 μM Y27632 for 4 h, (B) 20 μM CK666 for 2 h, and (C) 3 μM wortmannin for 2 h. The percentage of amoeba per image containing F-actin labelled structures (such as an adhesive plate (AP), an actin dot (AD) and a podosome-like structure (PL)) was plotted for each condition. On FN: with or without Y27632, n = 17 cells for each condition; with CK666, n = 14 cells; without CK666, n = 15 cells; with Wtmn, n = 11 cells; without Wtmn, n = 42 cells. On non-coated glass: with Y27632, n = 21 cells; without Y27632, n = 8; with or without CK666, n = 21 cells; without Wtmn, n = 20 cells; with Wtmn, n = 21 cells. N = 4 experiments. (D) Volume and sphericity of dots present in E. histolytica as a function of the substrate composition and the drug treatment. For cells seeded on non-coated glass: control, n = 503 dots; Y27632, n = 389 dots; CK666, n = 273 dots; Wtmn, n = 341 dots. For cells seeded on FN-coated glass: control, n = 1265 dots; Y27632 n = 504 dots; CK666, n = 432 dots; Wtmn, n = 522 dots. Mann-Whitney test: ns: not significant. ****P<0.0001; ***P<0.001, **P<0.01, *P<0.1.
Fig 4.
E. histolytica adopts a fan-shape morphology on a FN substrate.
Confocal micrographs of amoebae loaded on FN-coated glass or non-coated glass for 1 h and labelled for F-actin. (A) Amoebae on non-coated glass (left) and on FN-coated glass (right). White arrowheads and white arrows indicate examples of areas of high F-actin density. A quantification of cell actin fluorescence in the image did not reveal a difference between the two conditions (S2 Fig). Scale bar: 10 μm (B) Examples of cells with elongated shape, showing a higher F-actin concentration on the smallest side of the cell (first row) or a reinforcement of cortical F-actin (second row). Examples of cells with a fan-like shape, showing F-actin fibres as a spine on the broader side that could radiate into the cell like the spokes of a bicycle wheel, at the leading edge (third row) and in multiple dots (fourth row). Scale bar: 5 μm.
Fig 5.
On FN, the amoeba lengthens in the presence of CK666.
A. Examples of micrographs from six randomly selected motile trophozoites (left panel). After the segmentation of 17 cells, the trophozoites were contoured manually (middle panel) and the roundness index (Rnd) was determined (right panel). Rnd ranged from 28% to 85%. B. Distribution of Rnd values for cells migrating on the two substrates in the presence or absence of the inhibitors. On FN: control, n = 223 cells; with Y27632, n = 149 cells; with CK666, n = 36 cells; with Wtmn, n = 36 cells. On non-coated glass (G): control, n = 206 cells; with Y27632, n = 105 cells; with CK666, n = 73 cells; with Wtmn, n = 39 cells.
Fig 6.
E. histolytica moves faster on non-coated glass than on FN, and longitudinal movements are faster than lateral movements.
The mean velocity of trophozoites fluorescently labelled and loaded on non-coated glass (A, C) or FN-coated glass (A, B) in the presence (B, C) or absence (A) of the inhibitors Y27632, CK666 and Wtmn. Images were acquired in real time. Using Icy software, the cells were segmented, and the velocity of each cell was averaged over the whole recording. On FN: control, n = 223 cells; with Y27632, n = 249 cells; with CK666, 357 cells; with Wtmn, n = 36 cells. On non-coated glass: control, n = 206 cells; with Y27632, n = 105 cells; with CK666, n = 73 cells; with Wtmn, n = 39 cells. T-test: ns: not significant. ****P<0.0001; ***P<0.001, **P<0.01, *P<0.1.
Fig 7.
Fan-shaped E. histolytica exhibits a lateral migration mode.
A. Trajectories of cell displacements. Trophozoites fluorescently labelled with a red cell tracker and migrating on FN-coated glass were imaged at 300 ms intervals for 3 min and segmented with the Active Contours plugin. The displacement trajectories were processed with the Track Manager in Icy software. Representative video: Sup. Data 2) B. Individual segmented cells. An elongated cell moving in a winding trajectory in the same direction as the cell axis (upper row). A fan-shaped cell moving in a straight path perpendicular to the cell axis (bottom row). C. A schematic representation of the correlation between the cell migration angle and the type of movement. D. Examples of E. histolytica’s migration modes. A trophozoite with the displacement angle mostly below 40° at different times during longitudinal movement on non-coated glass. A trophozoite with the displacement angle mostly above 40° during characteristic lateral movement on FN-coated glass. Representative video: S3 Video.
Fig 8.
The lateral movement of fan-shaped amoeboid cells is not pressure-driven.
Measurements of pressure, velocity, and force for trophozoites on (A) non-coated glass, and (B) FN-coated glass. The snapshots displayed were rendered using Bioflow Display. The values are laid on top of the raw images at different time points. The force and pressure are measured relative to the cell’s viscosity constant. The colour scale differs for A and B: blue low to red high. Pressure [s-1]: A (0–4.1), B (0–2.1); velocity [μm s-1]: A (0–1.9), B (0–1.1); force [m-1s-1]: A (0–1.7⋅10−8), B (0–19.3⋅10−8). Longitudinal movement: T1: 220 ms, T2: 330 ms, T3: 440 ms. Lateral movement: T1: 210ms, T2: 315ms, T3: 420 ms. C. Four different quantitative variables that differ when comparing the two types of movement (see the Materials and Methods). “Pressure gradient” refers to the difference in pressure values within a cell, “flow speed” is the magnitude of the flow velocity, “variations” refers to the variance in time and is meant to quantify how much the velocity changes, and “force” refers to the magnitude of the force vector. Longitudinal movements appeared to be faster and pressure-driven, whereas lateral movements rely more on force. Unequal-variances t-test. *: 0.05; **: 0.01; ***: 0.001; ****: 0.0001 (n = 5,6 per condition).
Fig 9.
Models of signalling features during the movement of Entamoeba histolytica.
Three scenarios are summarized by considering possible pathways for the intracellular dynamics of trophozoites subjected to various stimuli. A. The diagram represents data from our in vitro experiments on trophozoites loaded onto non-coated glass or FN-coated glass. The activity of Pi3 kinase (Pi3K) is the main regulator, independent of the nature of the substrate. Pi3K can be inhibited by wortmannin (Wtmn), leading to a loss of morphology and adhesive structures and less cell migration. Pi3K is known to act selectively on small GTPases (see S1 Fig). Next, phosphoinositides activate ROCK in a substrate-dependent manner. ROCK is responsible for the formation of adhesion plaques (APs) and actin dots (ADs) on non-coated glass, and podosome-like structures (PLs) on FN-coated glass. In the presence of Y27632, the balance of these structures is modified (shown inside the square). Through various effectors, Pi3K signalling also activates Arp2/3 complex assembly, which is responsible for a rapid change towards a fan-shaped cell morphology on FN-coated glass. Inhibition of Arp2/3 by CK666 is associated with a lower proportion of fan-shaped cells. B. A working hypothesis for E. histolytica’s movements in areas of tissue lacking FN. We hypothesize that elongated trophozoites move primarily in an amoeboid manner; they are able to penetrate a “soft” environment (such as mucus and fibrillar collagen I in the ECM), as previously observed in a human intestinal explant model. The actomyosin cytoskeleton is essential for this activity because it generates the intracellular pressure gradient that pushes the cell body forward. C. A working hypothesis for E. histolytica’s movements in areas of tissue rich in FN. Trophozoites can specifically bind to a FN receptor, which triggers a signalling pathway responsible for morphological changes to a fan-shaped cell and the formation of podosomes, which attach the amoeba strongly to FN-coated glass and reduce the migration speed. Podosomes have an impact on E. histolytica’s invasion in the intestine because they are FN degradation sites thought to facilitate the destruction of the colonic architecture and thus allow the parasite’s progression into the tissues.