DC mobilization from the skin requires docking to immobilized CCL21 on lymphatic endothelium and intralymphatic crawling

Abstract

Dendritic cells (DCs) must travel through lymphatics to carry skin antigens into lymph nodes. The processes controlling their mobilization and migration have not been completely delineated. We studied how DCs in live mice respond to skin inflammation, transmigrate through lymphatic endothelium, and propagate in initial lymphatics. At steady state, dermal DCs remain sessile along blood vessels. Inflammation mobilizes them, accelerating their interstitial motility 2.5-fold. CCR7-deficient BMDCs crawl as fast as wild-type DCs but less persistently. We observed discrete depositions of CCL21 complexed with collagen-IV on the basement membrane of initial lymphatics. Activated DCs move directionally toward lymphatics, contact CCL21 puncta, and migrate through portals into the lumen. CCR7-deficient DCs arrive at lymphatics through random migration but fail to dock and transmigrate. Once inside vessels, wild-type DCs use lamellipodia to crawl along lymphatic endothelium and, sensing lymph flow, proceed downstream. DCs start drifting freely only in collecting lymphatics. These results demonstrate in vivo that the CCL21-CCR7 axis plays a dual role in DC mobilization: promoting both chemotaxis and arrest of DCs on lymphatic endothelium. Intralymphatic crawling, in which DCs combine active adhesion-based migration and directional cues from lymph flow, represents a new step in DC mobilization which may be amenable to regulation.

Figure 1.

Optical sectioning of the intact footpad skin from a live CD11c-EYFP mouse. (a–d) Images representing extended focus views of subsequent layers. (a) The upper layer of the epidermis, at 0–20 µm in depth, consists of cornified keratinocytes stained topically with SNARF. (b) The epidermis, at ∼20–40 µm, contains CD11c-EYFP⁺ star-shaped LCs. (c) The upper dermis, at ∼40–65 µm, contains collagen fibers, which generate a second harmonic signal, and blood vessels traced with QD655. (d) The deeper dermis, at ∼65–120 µm, contains CD11c-EYFP⁺ DDCs and lymphatics stained with anti-LYVE1 conjugated to Texas Red. (e–g) Three-dimensional reconstructions of all the above layers (e), epidermal LCs (f), and dermal lymphatics, DDCs and blood vessels shown from below (g). Images are representative of at least 15 paws of CD11c-EYFP⁺ mice examined.

Figure 2.

Inflammation of the footpad accelerates DDC movement in the dermis. (a and b) Extended focus snapshots of the dermis in the steady-state (a) or 24 h after injecting CFA s.c. (b). Tracks show the motion of selected DDCs through 45 min (arrowheads indicate the endpoints of tracked cells). (c and d) Charts representing the three-dimensional paths, normalized to their starting coordinates, taken by the cells tracked above. (e and f) Crawling velocities (e) and arrest coefficients (f). Shown is the percentage of time in which cells were immobile (slower than 2 µm/min). Data points represent individual cells (n of steady state = 535, n of CFA-treated = 305) and were pooled from six mice and 15 movies for each condition. Red bars denote the mean. Inflammation increased velocities and reduced arrest. *, P < 0.0001 for both.

Figure 3.

CCR7 plays a dual role in DCs mobilization, promoting chemotaxis and docking to lymphatics. After s.c. injection of LPS-activated CFSE-stained BMDCs to the footpad, WT DCs (a–c) moved linearly toward an initial lymphatic, compatible with chemotaxis. DCs that crossed the endothelium clustered in the proximal sections of initial lymphatics, either in blind ends (c, left rectangle) or in adjacent sections (c, right rectangle). (d) Within 24 h, WT cells have efficiently entered lymphatics. In contrast, CCR7^−/− cells (e) kept crawling in the interstitium bypassing the lymphatic vessels they encountered (four representative cells tracked), suggesting that CCR7 promotes DC adhesion to lymphatics. This pattern was also apparent (f) when both WT and CCR7^−/− DCs were co-injected. CCR7^−/− DCs crawled as fast as WT DCs in the dermis (P = 0.29; g) but were less persistent (P = 0.006; h), implying that CCR7 ligation is not essential for DC chemokinesis but participates in their chemotaxis. Persistence index was calculated by dividing the cell displacement by path length. Data points represent individual cells and were pooled from three mice for each condition. Red bars denote the mean. *, P = 0.006. (i) When BMDCs were co-injected with CCL21, their migration to the popliteal LN was not affected, indicating that misplaced chemokine did not misguide DCs and prevent random migration and docking on endothelial CCL21. Data indicate the percentage of injected DCs out of resident LN CD11c⁺ cells. Error bars denote SEM.

Figure 4.

CCL21 depositions are concentrated in specific regions of LECs. Whole-mount confocal images of the skin of the footpad (a) or ear (b) stained for LYVE-1 and CCL21 showed a punctate expression of CCL21 on initial lymphatics (blue dots). (c) This pattern was even more obvious after isosurface rendering. (d and e) Triple staining for VE-cadherin, LYVE-1, and CCL21 shows alternate arrangement of VE-cadherin at button junctions and LYVE-1 at loose flaps. CCL21 puncta are located in the junction-free areas of the LECs. (f–h) Compared with the steady-state, inflammation, either through CHS (g) or CFA (h), did not affect the pattern of CCL21 puncta or increase the CCL21 signal after 24 h.

Figure 5.

CCL21 immobilization on the basement membrane of initial lymphatics may facilitate DC adhesion at CCL21-rich sites and passage through dedicated portals. (a) Collagen IV staining in whole-mount preparations of the ear skin shows perforations within the basement membrane of initial lymphatics (arrows in boxed region). (b) CCL21 staining reveals that several, but not all, of these perforations (arrows) are associated with CCL21 puncta. (c) Treatment of skin cross sections with type IV collagenase digested the basal membrane of initial lymphatics, markedly reducing collagen on lymphatics. Membrane-bound CCL21was dissociated, leaving behind small perinuclear depositions (arrows). (d) In the presence of the calcium chelator EDTA, collagenase IV treatment did not disrupt collagen (left) and CCL21 (right). (e–h) 24 h after contact sensitization, skin whole mounts were triple-stained for LYVE-1, CCL21, and MHC-II. MHC-II⁺ DCs accumulated outside initial lymphatics (e). (f) Enlargement of the boxed region in e shows in three dimensions a DC which extended protrusions toward the initial lymphatic vessel and contacted two CCL21-rich puncta. Two other examples of DCs contacting CCL21 are shown (g and h). (i–l) Similar results were obtained when the skin of CD11c-EYFP mice was analyzed. Collectively, these findings suggest that CCL21 is secreted from intracellular stores inside LECs and is immobilized on the basal membrane (often near preformed portals), to promote DC adhesion and site-specific transmigration.

Figure 6.

DCs trans-migrate through the endothelium in selected lymphatic sections. The footpad was injected with BMDCs s.c. with 24 h before imaging. (a) Sequential images (timed 0–40 min) demonstrate the typical dynamics of DC transmigration; within 20 min, a DC that adhered to abluminal surface of a lymphatic vessel trans-migrated into the lymphatic lumen and started crawling. (b) Two DCs cross through the same spot on the endothelium, presumably representing a preformed portal (Pflicke and Sixt, 2009).

Figure 7.

DCs actively crawl inside initial lymphatics using lymph flow as a directional cue. (a) Sequential images of BMDC propagation inside initial lymphatics. Circled is a DC crawling inside the lymphatic lumen. (b) At a higher magnification a DC is seen crawling using lamellipodia at the leading edge and a well defined uropod at the trailing edge. Sequences are representative of at least 13 independent experiments. (c) Under general anesthesia, which is known to reduce lymph flow, the percentage of immigrant GFP⁺ BMDCs from the footpad skin, as recorded in the popliteal LN 10 h after transfer, drops fivefold (P < 0.001). Data represent three independent experiments. Red bars denote the mean. *, P < 0.001. (d–f) In the presence of lymph flow, DCs move directionally in the lymphatics. Mice were injected s.c. with 5 µl of fluorescently stained saline (red). As fluid diffused through the interstitium and concentrated in lymphatics, DCs switched from random motility inside lymphatics (d) to directional crawling toward the LN (e). Tracks show several DCs following the same route down the lymphatic vessel (f). Under these conditions DCs crawled along the lymphatic vessel at a mean speed of 9.06 µm/min and a persistence index of 0.65.