Rolling neutrophils form tethers and slings under physiologic conditions in vivo

Abstract

Human and mouse neutrophils are known to form tethers when rolling on selectins in vitro. Tethers are ∼0.2 μm thin, ∼5–10 μm‐long structures behind rolling cells that can swing around to form slings that serve as self‐adhesive substrates. Here, we developed a mouse intravital imaging method, where the neutrophil surface is labeled by injecting fluorescently labeled mAb to Ly‐6G. Venules in the cremaster muscle of live mice were imaged at a high frame rate using a confocal microscope equipped with a fast resonant scanner. We observed 270 tethers (median length 3.5 μm) and 31 slings (median length 6.9 µm) on 186 neutrophils of 15 mice. Out of 199 tether break events, 123 were followed by immediate acceleration of the rolling cell, which shows that tethers are load‐bearing structures in vivo. In venules with a high wall shear stress (WSS; > 12 dyn/cm²), median rolling velocity was higher (19 μm/s), and 43% of rolling neutrophils had visible tethers. In venules with WSS < 12 dyn/cm², only 26% of rolling neutrophils had visible tethers. We conclude that neutrophil tethers are commonly present and stabilize rolling in vivo.

Abbreviations

D_after

displacement after tether breakage

D_before

displacement before tether breakage

PSGL‐1

P‐selectin glycoprotein ligand 1

WSR

wall shear rate

WSS

wall shear stress

1. INTRODUCTION

In most tissues, neutrophil rolling is a necessary precursor to arrest, adhesion, and transmigration—a process collectively known as recruitment.1 Rolling is driven by the drag force and torque exerted by the flowing blood and opposed by selectin and integrin bonds breaking at the rear end of the rolling cell.2, 3, 4

Neutrophils form tethers to stabilize their rolling speed over a range of WSS.5 In a P‐selectin‐coated flow chamber, neutrophils roll as a result of the formation of anchoring bonds between the P‐selectin substrate and the PSGL‐1 expressed on the neutrophil microvilli. Studies conducted with contrast imaging microscopy showed that these anchoring bonds pull out tethers from the neutrophils starting at 2 dyn/cm² WSS.6, 7 Further studies conducted with fluorescent‐based microscopy of membrane‐labeled neutrophils revealed that tethers can swing around the rolling cell and form slings, providing a self‐adhesive substrate over which the cell can roll.8, 9 These experiments showed that as the anchoring bonds break, the tether disappears without leaving behind any membrane residue on the substrate. Only ∼15% of the tethers become slings,8, 9 presumably because most of the tethers are pulled back into the cell before they could swing in front of the cell. Tethers and slings stabilize neutrophil rolling via increasing the number of anchoring points between the substrate and the rolling cell.5 The load‐bearing property of these structures is supported by several observations: 1) with increasing WSS, the number of tethers increases;7 2) at a constant WSS, the cells showing more tethers roll at lower velocity;7 3) each tether break is followed by microjumps of the rolling neutrophil;7, 8, 9 and 4) detachment of slings from the substrate is followed by an increase of rolling speed.8

Imaging of tethers and slings in vivo is challenging, because tethers are thin (∼200 nm), below the resolution limit of light microscopy, transient, require high temporal resolution, and may go out of focus. Therefore, no good quality images or movies of neutrophil tethers or slings in vivo are available. One study reported tether‐ and sling‐like structures on rolling neutrophils in vivo, but no measurements of frequencies, numbers, or lengths of these structures were provided.8 Here, we developed an intravital imaging method that detects robustly and consistently tethers and slings on rolling neutrophils in vivo.

2. MATERIALS AND METHODS

Eight‐ to 16‐wk‐old male C57BL/6J mice were anesthetized with isoflurane inhalation, placed on a 37°C heating pad, and prepared for intravital microscopy via femoral artery cannulation and cremaster muscle exteriorization, as described in previous studies.10, 11 Cremaster venules were imaged with a Leica 25×/0.95 NA water‐immersion objective, and high temporal resolution was achieved by using a resonant scanner on a Leica SP8 confocal microscope, equipped with hybrid detectors (Leica Microsystems, Buffalo Grove, IL, USA).

To make tethers and slings visible, the highly expressed, neutrophil‐specific cell‐surface molecule Ly‐6G was labeled via injecting via the femoral artery catheter 5 µl (2.5 µg) Alexa Fluor 647 or 10 µl (2 µg) PE‐labeled anti‐Ly6G mAb (Clone 1A8; BioLegend, San Diego, CA, USA).

A vessel with rolling and tether‐forming neutrophils was recorded for several minutes. In the same vessel, WSR was measured via measuring maximal blood‐flow velocity with particle‐tracking velocimetry.12 For this, the acquisition settings were changed to a high frame‐rate recording (see Fig. 2C), and 0.5 µm Nile Red fluorescent particles (Spherotech, Lake Forest, IL, USA), diluted in saline, were injected through the femoral artery catheter close to the branch entering the cremaster.11 Each time a small volume (∼30 μl) was injected, a shower of fluorescent beads was observed in the vessel.

Figure 2.

Load‐bearing properties of tethers in vivo. (A) Frame‐to‐frame displacement tracking of a rolling neutrophil; delay between frames is 210 ms. The blue columns indicate the period where the tether breaks. Red dots indicate the frames of the example sequences shown above the chart. The left and middle sequences show large jumps; the right example shows no jump of the neutrophil after tether break. (B) The ratio between D_before and D_after tether break for 199 tether break events is shown. (C) Velocity and WSR measurement. After acquiring a vessel where rolling neutrophils with tethers were visible, acquisition settings were changed to rapid imaging to trace the frame‐to‐frame displacement of fluorescent particles. The white arrows indicate a circulating fluorescent particle on 3 consecutive frames; the magenta, dashed frame indicates the location where the neutrophil shown on the left side was detected. Venule diameter was measured on the transillumination images. (D, upper, left and middle) Each dot is a vessel. (Right) The relationship between rolling velocity and WSS is shown on a cumulative frequency chart; each dot is a cell with tether(s), and these cells are grouped based on the WSS of the vessel in which they roll. (Lower) The data are grouped based on the WSS ranges (above or below 12 dyn/cm²), introduced in the upper right chart. (Left) The ratio between cells, with and without tether; each dot is a recorded sequence collected from 7 (red group) or 5 (blue group) vessels. (Middle) The total tether count per cell; each dot is a cell. (Right) Maximum tether length; each dot is a tether. Original scale bars, 8 μm, except C, bottom, where it is 20 μm. Significances include the following: Student’s t test, *P < 0.0001; **P < 0.05

WSR was calculated as 0.625 × v_max × 8 × 2.2/d,13 where v_max , the maximum flow velocity was measured via tracking the fastest bead in the vessel, and d, the vessel diameter, was measured on the transillumination images. WSS is linearly related to WSR by plasma viscosity, as the area near the endothelium is free of blood cells.12 Mouse plasma viscosity is ∼1 cP,14 which is equal to 0.01 s × dyn/cm²; thus, a WSR of 100 s⁻¹ corresponds to a WSS of 1 dyn/cm². The recordings were analyzed with Leica LAS X (Leica Microsystems), Imaris 8.3 (Bitplane, Concord, MA, USA), and FIJI software (http://fiji.sc/). Data were analyzed, and charts were made with Excel (Microsoft, Redmond, WA, USA) and Prism 7 (GraphPad Software, La Jolla, CA, USA). Statistical analyses included 2‐tailed Student’s t test.

3. RESULTS AND DISCUSSION

Tethers were observed in over a 100 neutrophils rolling in mouse cremaster venules under mild inflammation induced by surgical trauma. Under these conditions, rolling is known to be P‐selectin and PSGL‐1 dependent.11, 15, 16 At least one‐third of the analyzed neutrophils showed tethers. This is a lower bound, because tethers can easily be missed if they are in a plane that is not in focus. Most observations were made using anti‐Ly6G‐AF647 mAb, but similar results were obtained with anti‐Ly6G‐PE mAb (Fig. 1A).

Figure 1.

Neutrophils form tethers and slings in vivo while rolling in venules. (A) Ly6G‐AF647 (magenta)‐ or Ly6G‐PE (red)‐labeled rolling neutrophils, observed from a side‐ or bottom‐view perspective, show tethers. Consecutive frames, separated by dashed lines, are shown in top‐to‐bottom order. Arrows indicate the last visible tether in each example. The vessel lumen is indicated by the diffuse background signal given by the unbound antibody. The chart shows the size distribution of the observed tethers. (B, top) A neutrophil (bottom view) that forms 3 tethers while it passes through the field of view. The image was made via overlaying 4 frames; numbers indicate the relative order of frames. (Left) Side view of a neutrophil, which forms 2 tethers at the same time. While the rear tether grows (anchoring point marked with white arrows), the second tether is visible in a single frame only. The graph shows the total count of tethers per neutrophil observed as the neutrophil passed through the field of view. (C) Selected frames from a record of a tether‐to‐sling transition are shown in top‐to‐bottom order. The delay of a frame compared with the previous frame is indicated. The tether disappeared, and sling appeared during the 210 ms delay between the second and third frame. The sling landed on the endothelium, and the neutrophil wrapped it up. White arrows indicate the last image of the tether and the first image of the sling. (D) Examples of neutrophils with sling (indicated with white arrows). The slings were rolled up by the neutrophils. Chart shows the size distribution of observed slings. (E) Number of tethers, slings, and cells with tether or sling detected during imaging the indicated amount of vessels of the indicated amount of mice. Rolling direction is always from left to right. Frame order is top to bottom or left to right. t/s, tether/sling. All scale bars, 8 μm

Tethers ranged in length between 1 and 42 μm (median 3.5 µm; Fig. 1A). Several cells showed multiple tethers while traveling through the field of view (Supplemental Movie 1), but very few cells showed multiple tethers at the same time, and an example is shown (Fig. 1B). Although several slings were observed, only in some cases was it possible to observe the entire tether‐to‐sling transition in side view (Fig. 1C and Supplemental Movie 2). Sling lengths ranged between 2 and 36 μm (median 6.9 µm; Fig. 1D). Occasionally, a residual tether material is left behind on the endothelium after the tether disappeared (an example is visible on Fig. 1C and on Supplemental Movie 2). The frequency and quantitative importance of this phenomenon is unknown. In total, 270 tethers and 31 slings (11% of the number of tethers) were observed on 186 neutrophils rolling in 31 venules of 15 mice (Fig. 1E).

Next, we addressed the effect of tethers on neutrophil rolling. In agreement with the in vitro reports, we observed in vivo that after each tether break, the cell velocity increases immediately. An example of this is shown in Fig. 2A, displaying frame‐to‐frame the displacement and tether breakage of the same cell rolling over 52 frames (10.9 s). We analyzed the velocity after the tether break in 148 neutrophils and expressed the data as the ratio of D_after and D_before (Fig. 2B). In 138 of 199 observed tether breaks (69%), this ratio was above 1, which was highly significant (P < 0.0001). The median value of this ratio was 1.3 (Fig. 2B).

We measured the WSS and neutrophil rolling velocity in 16 venules (diameter range 14–59 µm), which showed tether‐forming neutrophils (Fig. 2D, upper). WSS and rolling velocity increased with decreasing venule diameter. To assess the relationship between WSS and rolling velocity or tether properties, we stratified the venules into low WSS (< 12 dyn/cm²) and high WSS (> 12 dyn/cm²) groups. Rolling velocity and the ratio of neutrophils with/without tethers were significantly higher in the high WSS group (median: 3.76 vs. 18.81 µm/s, P < 0.0001; and 26 vs. 43%, P < 0.05; Fig. 2D, upper right and lower left). Only venules with at least 10 rolling cells were analyzed. The count of tethers per cell did not change significantly (1.69 vs. 1.8; Fig. 2D, lower middle). Tether length was significantly less in the high WSS group (median: 3.68 vs. 3.34 µm, P < 0.05; Fig. 2D, lower right).

We conclude that tethers and slings on P‐selectin dependently rolling neutrophils are commonly present in vivo. Most tethers are load bearing, based on the observed cell acceleration after the tether break—the correlation between WSS and ratio of cells with tether(s).

Supporting information

Supporting Information

Supporting Information

Marki A, Buscher K, Mikulski Z, Pries A, Ley K. Rolling neutrophils form tethers and slings under physiologic conditions in vivo. J Leukoc Biol. 2018;103:67–70. 10.1189/jlb.1AB0617-230R