EMILIN1/α9β1 Integrin Interaction Is Crucial in Lymphatic Valve Formation and Maintenance

Abnormal lymphatic valves in Emilin1^−/− mice. (A) PECAM-1 immunohistochemistry of mesenteric vessels and luminal valves of embryonic (E16.5 and E18.5) and postnatal (P0, P3, and P6) WT and Emilin1^−/− mice. Note the difference in shape of WT (arrowheads) and Emilin1^−/− (arrows) valves. (B) Luminal valves in chyle-filled mesenteric P6 WT and Emilin1^−/− mice. The arrows and arrowheads indicate abnormal (ring-shaped) and normal (V-shaped) valves, respectively. (C) Morphological evaluation of valve shape in P6 WT and Emilin1^−/− mesenteric lymphatic vessels. Black bar, normal V-shaped valves; white bar, abnormal ring-shaped valves (n = 3 animals per genotype; more than 100 valves were counted for each genotype). *, P < 0.05 (χ² test). (D and E) Quantification of the luminal valves in neonatal and postnatal WT and Emilin1^−/− mesenteric lymphatic vessels of CD1 (P0 and P6) (D) and C57BL/6 (P0 and P6) (E) mice (means and standard deviations [SD]; n = 3 animals per genotype with 10 vessels each; *, P < 0.05; **, P < 0.01; unpaired Student's t test). Scale bars, 50 μm (A) and 100 μm (B).

Defects of lymphatic collectors in adult animals. (A) Quantification of the luminal valves in adult (P21) CD1 WT and Emilin1^−/− mesenteric lymphatic vessels (means and SD; n = 3 animals per genotype with 10 vessels each; *, P = 0.01). (B) PECAM-1 staining of ear vessels in 6-month-old C57BL/6 mice. Note the well-formed V-shaped valves in WT animals compared to the narrow structures of collecting vessels in Emilin1^−/− mice. (C) Adult lymphatic mesenteric vessels (P21) stained for α-SMA (gray) and Prox1 (green) of WT and Emilin1^−/− mice. (D) Lumen sizes of the valves in adult (P21) WT and Emilin1^−/− mesenteric lymphatic vessels. Z-stack images were collected from whole-mount samples stained for α-SMA and Prox1; the resulting three-dimensional (3D) reconstruction by Volocity software combined with x and y projections was used to highlight and calculate the smallest (d_s) and largest (d_l) diameters of the valve lumen. The values obtained from calculation [(d_s + d_l)/2] are reported in the graph. Four mice per genotype (at least 5 valves/mesentery) were examined. *, P < 0.01. Scale bars, 25 μm (C) and 50 μm (B).

Abnormal lymphatic collectors and valves in Emilin1^−/− adult mice. Shown are serial semithin cross sections of P21 WT (A) and two representative Emilin1^−/− (B and C) mesenteric valves. Note the increased thickness of the lymphatic wall in Emilin1^−/− mice. The lumen of collecting vessels in the right-hand panel appears contracted, with folded valve leaflets. The sections marked “a” in panels A and C were subjected to ultrastructural analysis as shown in Fig. 5. Scale bars, 25 μm.

Ultrastructural analysis of mesenteric valves and lymphatic wall in adult mice. (A and B) TEM analysis of P21 WT (A) and Emilin1^−/− (B) mesenteric valves. The colored boxes in the semithin sections (upper left panels, corresponding to sections Aa and Ca, respectively, from Fig. 4) represent the areas that were selected for ultrathin sections. Emilin1^−/− mice show an increased number of mural cells/pericytes (P), which display a SMC/myofibroblast-like phenotype, and folded LECs. The arrow in panel B indicates a mitotic LEC. (C) TEM analysis of P21 WT and Emilin1^−/− mesenteric lymphatic vessel walls showing increased numbers of caveolae (arrows) and cytoplasmic filaments (F) and the presence of a basement membrane (arrowheads) at the surface of one representative Emilin1^−/− mural cell/pericyte. Scale bars, 25 μm (Aa and Ba), 1 μm (A and B), and 500 nm (C).

EMILIN1 deficiency leads to impaired lymphatic collector function. FITC-dextran (2,000 kDa) was subcutaneously injected into the tails of WT and Emilin1^−/− mice at P21 (A) and at 6 weeks (B) to visualize the draining lymphatic vasculature and lymph nodes. The mice were sacrificed after injection at the indicated times. Low-magnification images were taken to visualize lymph nodes, and high-magnification images (a, a′, a″, a‴, b, and b′) were taken to detail collectors corresponding to the boxed areas. Note the difference in fluorescence intensity between WT and Emilin1^−/− draining lymph nodes and collectors. The arrows and arrowheads indicate V- and ring-shaped valves, respectively. (C) Draining lymph nodes (inguinal, iliac, and mesenteric) were removed 30 min after injection and measured with a computer-interfaced GeniusPlus microplate reader (Tecan, Italy). Arbitrary fluorescence units (AU) in the graph represent the sums obtained from the fluorescence values of all draining lymph nodes/mouse. The graph reports the means ± SD of three independent experiments (n = 4 animals per genotype). Scale bar, 2 mm (0.5 mm in a, a′, a″, a‴, b, and b′).

EMILIN1 colocalizes with α9 integrin in valve leaflets. Immunofluorescence staining for the localization of EMILIN1 (green) and α9 integrin (A′, red or gray) in embryonic (E18.5) (A and A′), neonatal (P0) (B, C, and D), and mature (P6) (E) mesenteric lymphatic vessels. The arrows indicate lymphatic valves, which were positive for both EMILIN1 and α9 integrin staining. (C) Magnification of the boxed area in panel B. The dotted lines in panel A outline the lymphatic vessels. (B and C) Mesenteric vessels are indicated as follow: a, artery; v, vein; l, lymphatic. (D) Four confocal longitudinal cross sections of a mesenteric valve showing EMILIN1 expression at both abluminal (z1) and luminal (z2, z3, and z4) sides of the valve and colocalization with α9 integrin in the valve leaflets. (E) Representative P6 mesenteric lymphatic valve stained for EMILIN1 (green) and α9 integrin (red). Scale bars, 150 μm (B), 50 μm (A, C, and D), and 25 μm (E).

EMILIN1 is an α9 integrin-specific ligand in the lymphatic vasculature system. (A) Percentages of HMVEC-LLy and HMVEC-dLyNeo cells adhering to EMILIN1 (E1), gC1q, E933A, and bovine serum albumin (BSA). The data are expressed as means and SD of three independent experiments with 6 replicates. (B) Adhesion of HMVEC-LLy and Jurkat cells (positive control for α4 integrin expression). The cells were preincubated with anti-α4 integrin subunit monoclonal antibody (MAb) (P1H4) or anti-α9 integrin subunit MAb (Y9A2). The data are expressed as means and SD of three independent experiments with 6 replicates. (C) Proliferation of HMVEC-LLy cells cultured on tissue culture plates (plastics) and incubated for 48 h in the presence of fibronectin, EIIIA, E1, or gC1q. The data are expressed as mean percentages and SD of the number of Ki67-positive cells of three independent experiments. Six fields were examined for each condition. (D) Representative images of Ki67 (green) and α-SMA (red) staining of WT and Emilin1^−/− P6 mesenteric vessels. The dotted lines outline the lymphatic vessels. The insets are magnifications of the boxed areas showing the presence of a higher number of Ki67-positive cells in Emilin1^−/− lymphatic valves. (E) Quantification of the Ki67-positive cells of luminal valve-associated areas in WT and Emilin1^−/− P6 mesenteric lymphatic vessels. Z-stack images were collected from whole-mount samples; the resulting 3D reconstruction by Volocity software was used to highlight and calculate the valve area and count the associated Ki67-positive cells. Four mice per genotype (at least 10 valves/mesentery) were examined. (F) Haptotactic movement of HMVEC-dLyNeo cells toward gC1q or EIIIA in the presence or absence of the function-blocking monoclonal anti-α9 integrin subunit. (G) Phalloidin staining for actin cytoskeleton visualization of HMVEC-dLyNeo after 1 h of adhesion on FN, EMILIN1, EIIIA, or gC1q. (H) Wound-healing scratch performed on HMVEC-dLyNeo in the presence of the function-blocking monoclonal anti-α9 integrin subunit or of the monoclonal antibody anti-gC1q. Scratch healing was determined by measuring the shortest distance between scratch edges at 0 and 20 h in each field of view. At least 3 different fields were measured per scratch. (I and J) Wound-healing assays of WT and Emilin1^−/− (E1-/-) LAECs. The graphs report the speed of closure (μm/h) of WT and Emilin1^−/− LAECs in the presence of culture medium without (I) or enriched with (J) supplements and growth factors. (K) Immunodetection for EMILIN1 staining of LAEC transduced with scrambled or short hairpin RNA EMILIN1 (shEMILIN1) lentiviral particles. (L and M) Velocity of scrambled (scr) and EMILIN1 silenced (shE1) LAECs in a scratch assay in the presence of culture medium without (L) or enriched with (M) supplements and growth factors. The data shown in the graphs are the means and SD of three independent experiments each. *, P < 0.05; **, P ≤ 0.02; ***, P = 0.0001. Scale bars, 50 μm (D, G, and K) and 25 μm (insets).