Fig 2: CCL21/CCR7 axis has lymphangiogenic potential in vitro. (A) Real-time PCR of CCR7 and CCL21 mRNA expression in HMVEC-dLy. (B) Western blot of CCR7 and CCL21 protein expression in HMVEC-dLy. GAPDH was used as an internal control. (C) CCL21 protein secretion by HMVEC-dLy as measured by ELISA. Data are represented as mean ± SD (n = 3). (*) indicates significant difference (p < 0.05). (D) HMVEC-dLy proliferation in response to CCL21/6Ckine (0, 100, 200, 350 ng/ml) performed by the measurement of BrdU. Data are presented as mean ± SD (n = 4, p < 0.001). (E) HMVEC-dLy proliferation in response to CCR7 neutralizing antibody (0, 5, 10, 20 μg/ml). Data are presented as mean ± SD (n = 4, p < 0.0003). (F) Representative images of HMVEC-dLy migration (40× magnification). (G) Quantification of HMVEC-dLy cellular migration in response to CCL21/6Ckine (0, 100, 200, and 350 ng/ml). (H) Quantification of HMVEC-dLy migration in response to CCR7 neutralizing antibody (0, 5, 10, 20 μg/ml). Bars in (G, H) represent mean number of migrated cells ± SD (n = 4, p < 0.005). (I) Representative micrographs of HMVEC-dLy tubular network formation in response to CCL21/6Ckine (0, 100, 200, 350 ng/ml). Bar equals 100 μm. (J) Quantification of total length of tubular structures formed by HMVEC-dLy corresponding to (I) as determined by ImageJ. (K) Representative images of inhibition of HMVEC-dLy tubular network formation. Bar equals 100 μm. (L) Quantified HMVEC-dLy tube formation corresponding to images shown in (K). In (J and L) data are presented as mean ± SD (n = 4, p < 0.0001). (M) Quantified HMVEC-dly proliferation, migration, and tube formation in response to treatment with VEGFR-3 neutralizing antibody (0, 1, 2.5, 5 μg/ml). Data are presented as mean ± SD. Different superscripts represent a statistical significant difference.

Fig 3: CCL21/CCR7 pair correlates with lymphangiogenic markers in a panel of breast cancer tissues. (A) Quantitative real-time PCR analysis of CCL21, CCR7 and VEGF-C mRNA expression in control (adjacent non-tumor) and tumoral tissues. Data are represented as a mean ± SD. (*) indicates significant differences (p < 0.05). (B), (C), (D) mRNA expression level of CCR7 is positively correlated with the expression of lymphatic vascular markers (LYVE1, Podoplanin, VEGF-C) in primary breast cancer samples; Pearson’s coefficient indicates strong correlations. (E) mRNA expression level of CCR7 and blood vessel marker CD31 in primary breast cancer samples. Pearson’s coefficient suggests little to no correlation between the two variables. Data represent mean values from three independent experiments.

Fig 4: CCR7 activation by CCL21 regulates VEGF-C secretion via the PI3K/AKT signaling pathway. (A) Western blot time course analysis of AKT and ERK1/2 activation in MDA-MB-231 cells following treatment with CCL21 for a total of 60 minutes. Phosphorylation of AKT at Ser 473 and ERK1/2 is observed over the entire duration of stimulation compared to control, untreated cells. (B) Densitometric analysis revealed that phosphorylation of both pathways is significantly induced after 5 min of CCL21 treatment. (C) Western blot and (D) Densitometric analysis indicating that CCL21-induced phosphorylation is dependent on CCR7 activation. Western blot demonstrating that PI3K/AKT inhibitors LY 294002 (E) Akt-1/2 (F), and ERK1/2 inhibitor U0126 (G) can block CCL21/CCR7-mediated phosphorylation in an inhibitor dose-dependent manner. For all experiments, total AKT and total ERK1/2 confirmed the equivalent loading of lanes. (H-J) PI3K/AKT but not ERK1/2 specific inhibitors significantly block CCL21 mediated VEGF-C secretion measured with ELISA. Data are represented as means ± SD (n = 3). Different superscripts indicate statistically significant differences (p < 0.05).

Fig 5: CCL21/CCR7 axis has lymphangiogenic potential in vivo . (A) Macroscopic digital images of angioreactors collected with surrounding tissues. (B) Fluorescence assay analysis of LYVE1, Prox1, and CD31 markers as expressed by mouse lymphatic endothelial cells and blood vascular endothelial cells recruited into the angioreactors. Data are presented as mean relative fluorescent units ± SEM. (*) indicates significant difference (p < 0.005). (C) Quantitative real-time PCR analysis of LYVE1 and CD31 mRNA expression in the cellular contents of angioreactors. Expression levels are normalized to actin (ACTB). (*) indicates significant difference (p < 0.05) (n = 4). (D to G). Representative images of immunofluorescence localization of CD31 (red), LYVE1 (green), Podoplanin (green), and Prox1 (green) in serial sections of angioreactors containing MCF-7 mock vs. CCL21 KI MCF-7 cells (D), (E) and MDA-MB-231 mock vs. CCR7 KD MDA-MB-231 cells (F), (G). Nuclei are stained with DAPI (blue). Scale bar equals 50 μm. (H) Representative images of immunofluorescence localization of CD31 (red), LYVE1 (green), Podoplanin (green), and Prox1 (green) in serial sections of angioreactors containing CCR7 KD MDA-MB-231 cells and recombinant human VEGF-C (30 ng/μl). Nuclei are stained with DAPI (blue). Scale bar equals 50 μm. (I) Quantification of MVD and LVD. “Hot spot” scores for CD31-LYVE1, CD31-Podoplanin, and CD31-Prox1 were calculated by means of Image J (40× magnification). Data are presented as mean of “hot spot” ± SEM.