Research ArticleCANCER

Unexpected contribution of lymphatic vessels to promotion of distant metastatic tumor spread

See allHide authors and affiliations

Science Advances  08 Aug 2018:
Vol. 4, no. 8, eaat4758
DOI: 10.1126/sciadv.aat4758
  • Fig. 1 Lymphangiogenesis in metastasis-bearing lungs.

    4T1-luc2 cells (1 × 105) were orthotopically inoculated, and after 21 days, the primary tumors were removed. Twenty-one days after the operation, mice were sacrificed for analysis. (A) Representative images of a naïve lung and a metastasis-bearing lung. White dashed line indicates the edge of a metastatic nodule. Scale bars, 500 μm. (B) Quantification of Thy1+ area in the lungs of control mice and non-nodule covered lung area of 4T1-injected mice (n = 12). (C) Quantification of the lung-draining (tracheobronchial) lymph node (LN) weight (n = 12). (D) Representative images of 4T1 cells metastasized to a tracheobronchial lymph node. Scale bar, 50 μm. (E and F) Prox1-GFP mice were inoculated with 5 × 105 B16F10-luc2-tdTomato cells intravenously and were sacrificed on days 7, 14, and 21 after injection. (E) Representative images of lungs and tracheobronchial lymph nodes. Lymph nodes were imaged in situ using a fluorescence stereomicroscope. Scale bars, 500 μm. (F) Correlation between lung metastasis (quantified by bioluminescence signal from B16F10-luc2-tdTomato cells) and lymph node weight (n = 11). Nonparametric correlation (Spearman) with nonlinear regression analysis was performed.

  • Fig. 2 High lymphatic density and lymphatic invasion in metastasis-bearing lungs correlate with poor prognosis of melanoma patients.

    (A and B) Representative images of lymphatic hotspots (stained for podoplanin) around (A) and within (B) metastases in the lung. The metastasis area is labeled with “M.” Scale bars, 200 μm. (C and D) Quantification of the percentage of lymphatic area around (C) and within (D) metastases in the lungs of patients without (white bars; n = 11) and with (black bars; n = 15) lung-draining lymph node metastasis. (E and F) Kaplan-Meier survival analysis of patients with low lymphatic area (below median; n = 13) and high lymphatic area (above median; n = 13) around (E) and within (F) metastases in the lungs. (G) Representative image of lymphatic invasion in a melanoma lung metastasis. Scale bar, 200 μm. Tissues shown in (B) and (G) were located within a larger metastatic region containing cancer cells (see fig. S3C). (H) Quantification of incidence of lymphatic invasion in patients without (n = 11) and with (n = 15) lung-draining lymph node metastasis. Fisher’s exact test was performed. (I) Kaplan-Meier survival analysis of patients with (+; n = 14) and without (−; n = 12) lymphatic invasion. The time axes indicate the time from detection of lung metastases until death. Log-rank (Mantel-Cox) test was performed.

  • Fig. 3 Characterization of CCSP-rtTA × tet-O–VEGF-C mice after doxycycline treatment for 2 weeks.

    (A) VEGF-C mRNA expression in the lung (normalized to RPLP0). The average expression level in WT × tet-O–VEGF-C mice was set to 1 (n = 6). (B) VEGF-C protein levels in the lung detected by enzyme-linked immunosorbent assay (ELISA) (n = 6). (C) Quantification of the VEGFR3+ LV area (n = 4). (D) Representative images of VEGFR3 staining in the lung. Scale bars, 1 mm. (E) Representative high-magnification images of different regions in the lung stained for VEGFR3. Scale bars, 200 μm. PA, pulmonary artery; PV, pulmonary vein. (F) Representative images of MECA-32 staining in the lung. Scale bars, 50 μm. (G) Quantification of MECA-32+ blood vessel area in the lung (n = 5).

  • Fig. 4 Increased metastasis in CCSP-rtTA × tet-O–VEGF-C mice after tail vein injection of B16F10 melanoma cells.

    (A) Schematic of the B16F10 tumor study. Dox, doxycycline; i.v., intravenous. (B) Representative images of lung sections with metastatic nodules 14 days after tumor cell injection (indicated by staining for RFP from B16F10-luc2-tdTomato cells). Scale bars, 1 mm. (C) Quantification of tdTomato+ area in the lung 14 days after tumor cell injection (n = 10). (D) Quantification of tdTomato+ cells per lung 24 hours after intravenous injection of 1 × 106 B16F10-luc2-tdTomato cells. (E) Representative images of lymphatic invasion (white arrows) with orthogonal projection. Scale bars, 100 μm. Fisher’s exact test was performed. (F) Quantification of the lung-draining lymph node weight (n = 10). (G) Number of mice with different numbers of distant organs with metastasis (zero, one, two, or three organs affected) and example images of metastases (yellow arrows) in the liver and intestine. Data were pooled from two rounds of studies.

  • Fig. 5 Increased metastasis in CCSP-rtTA × tet-O–VEGF-C mice in the 4T1 spontaneous breast cancer metastasis model.

    (A) Schematic of the 4T1 tumor study. PR, postremoval; s.c., subcutaneous. (B) Representative in vivo bioluminescence images of metastases on day 21 after primary tumor removal. (C) Quantification of bioluminescence signal over the whole body on days PR7, PR14, and PR21 by IVIS imaging. For data from days PR7 and PR14, WT × VEGFC, n = 8, and CCSP × VEGFC, n = 11. Four mice reached the euthanization criteria before the end time point; thus, for data from day PR 21, WT × VEGFC, n = 7, and CCSP × VEGFC, n = 8. ns, not significant. (D) Number of mice with different numbers of distant organs with metastasis (zero, one, two, or three organs affected). Data were pooled from three rounds of studies. Only mice with detectable metastases (by ex vivo IVIS imaging) in the lungs were quantified.

Supplementary Materials

  • Supplementary material for this article is available at http://advances.sciencemag.org/cgi/content/full/4/8/eaat4758/DC1

    Fig. S1. High-magnification images of 4T1 metastases in lung sections stained for Thy1 and cytokeratin.

    Fig. S2. Clinical characteristics of study cohorts.

    Fig. S3. Lymphatic density and its correlation with prognosis.

    Fig. S4. Correlation of LV density and immune cell infiltration in lung metastases of melanoma patients.

    Fig. S5. Quantification of the number and size distribution of lung metastatic nodules in the B16F10 model.

    Fig. S6. Quantification of VEGF-C levels and analysis of LVs in different organs.

    Fig. S7. Characterization of CCSP-rtTA × tet-O–VEGF-C mice on BALB/c background.

    Fig. S8. Comparison of 4T1 primary tumor growth and lung metastasis without VEGF-C induction.

    Fig. S9. Quantification of nodule numbers and size distribution of the lung metastases on day PR21 in the 4T1 model.

    Fig. S10. 4T1 metastases in distant organs.

  • Supplementary Materials

    This PDF file includes:

    • Fig. S1. High-magnification images of 4T1 metastases in lung sections stained for Thy1 and cytokeratin.
    • Fig. S2. Clinical characteristics of study cohorts.
    • Fig. S3. Lymphatic density and its correlation with prognosis.
    • Fig. S4. Correlation of LV density and immune cell infiltration in lung metastases of melanoma patients.
    • Fig. S5. Quantification of the number and size distribution of lung metastatic nodules in the B16F10 model.
    • Fig. S6. Quantification of VEGF-C levels and analysis of LVs in different organs.
    • Fig. S7. Characterization of CCSP-rtTA × tet-O–VEGF-C mice on BALB/c background.
    • Fig. S8. Comparison of 4T1 primary tumor growth and lung metastasis without VEGF-C induction.
    • Fig. S9. Quantification of nodule numbers and size distribution of the lung metastases on day PR21 in the 4T1 model.
    • Fig. S10. 4T1 metastases in distant organs.

    Download PDF

    Files in this Data Supplement:

Navigate This Article