Lymph/angiogenesis contributes to sex differences in lung cancer through oestrogen receptor alpha signalling

in Endocrine-Related Cancer
Correspondence should be addressed to C Pequeux: C.Pequeux@uliege.be
Restricted access

Oestrogen signalling pathways are emerging targets for lung cancer therapy. Unravelling the contribution of oestrogens in lung cancer development is a pre-requisite to support the development of sex-based treatments and identify patients who could potentially benefit from anti-oestrogen treatments. In this study, we highlight the contribution of lymphatic and blood endothelia in the sex-dependent modulation of lung cancer. The orthotopic graft of syngeneic lung cancer cells into immunocompetent mice showed that lung tumours grow faster in female mice than in males. Moreover, oestradiol (E2) promoted tumour development, increased lymph/angiogenesis and VEGFA and bFGF levels in lung tumours of females through an oestrogen receptor (ER) alpha-dependent pathway. Furthermore, while treatment with ERb antagonist was inefficient, ERa antagonist (MPP) and tamoxifen decreased lung tumour volumes, altered blood and lymphatic vasculature and reduced VEGFA and bFGF levels in females, but not in males. Finally, the quantification of lymphatic and blood vasculature of lung adenocarcinoma biopsies from patients aged between 35 and 55 years revealed more extensive lymphangiogenesis and angiogenesis in tumour samples issued from women than from men. In conclusion, our findings highlight an E2/ERa-dependent modulation of lymphatic and blood vascular components of lung tumour microenvironment. Our study has potential clinical implication in a personalised medicine perspective by pointing to the importance of oestrogen status or supplementation on lung cancer development that should be considered to adapt therapeutic strategies.

Downloadable materials

  • Suppl. Fig.1: Genotyping of Tie2-Cre+/ERalox/lox (Cre+) and Tie2-Cre-/ERalox/lox (Cre-) mice by PCR on DNA extracts from the tail. A. Expression of Cre-recombinase and of FGF2 used as internal positive control of PCR. B. Expression of ERalox/lox and of ERa-/- corresponding to esr1 without exon 2.
  • Suppl. Fig.2: ELISA quantification of (A) VEGFA and (B) bFGF in LLC-Luc lung tumour lysates from vehicle- and PHTPP-treated females.

 

      Society for Endocrinology

Article Information

Metrics

All Time Past Year Past 30 Days
Abstract Views 542 542 127
Full Text Views 66 66 11
PDF Downloads 20 20 5

Altmetrics

Related Articles

Figures

  • View in gallery

    Orthotopic graft of LLC-Luc cells into lung parenchyma grows faster in females than in males. (A) In vivo bioluminescent signals derived from LLC-Luc lung tumours imaged either in whole mice (upper panel) or in dissected pulmonary lobes (lower panel) from male and female mice. The right panel shows bioluminescent intensity quantification over time (n = 8), **P < 0.01, 2-way ANOVA. (B) Hematoxylin/eosin staining of LLC-Luc lung tumours from male (n = 24) and female (n = 23) mice (scale bar = 1mm) and quantification of lung tumour density (tumour area/total lung area) on the right panel, eight slides spaced with 50 µm were analysed per sample, *P < 0.05, Mann–Whitney. (C) In vivo bioluminescent signals derived from LLC-Luc lung tumours imaged in female, OVX female and OVX + E2 female mice. The right panel shows bioluminescent intensity quantification over time (n = 8), *P < 0.05, 2-way ANOVA. (D) Hematoxylin/eosin staining and quantification of lung tumour density (tumour area/total lung area) of LLC-Luc lung tumours from females and OVX female mice treated or not with E2 (scale bar = 1mm) (n = 8), 8 slides spaced with 50 µm were analysed per sample, *P < 0.05, 1-way ANOVA. (E) In vivo bioluminescent signals derived from LLC-Luc lung tumours imaged either in whole mice of male (n = 8), gonadectomised (Cx) male (n = 8) and Cx + E2 male (n = 7) animals. The right panel shows bioluminescent intensity quantification over time, 2-way ANOVA. (F) Hematoxylin/eosin staining and quantification of the lung tumour density (tumour area/total lung area) of LLC-Luc lung tumours from males (n = 8) and gonadectomised male mice treated (Cx + E2, n = 7) or not with E2 (Cx, n = 8) (scale bar = 1mm), 8 slides spaced with 50 µm were analysed per sample, 1-way ANOVA.

  • View in gallery

    E2 increases LLC-Luc cell proliferation in vivo, but not in vitro. (A) Western blot of oestrogen receptors (ER) on LLC-Luc cell lysates: ERa (66 kDa), ERb (56 kDa), GPER (42 kDa). HSC70 (70 kDa) was used as loading control. Proteins isolated from mouse ovary and testis were used as positive controls. (B) In vitro proliferation of LLC-Luc cells treated with vehicle, E2 (10−9 M) or FBS 10% (positive control) for 24, 48 and 72 h, ***P < 0.001 vs vehicle, n = 6, 2-way ANOVA. (C) In vitro proliferation of LLC-Luc cells treated with vehicle, E2 (10−10 M to 10−7 M) or FBS 10% (positive control) for 48 h, ***P < 0.001 vs vehicle, n = 6, 1-way ANOVA. (D) In vitro viability of LLC-Luc cells treated with vehicle, E2 (10−10 M to 10−7 M) or cisplatin (10−4 M, positive control) for 48 h, ***P < 0.001 vs vehicle, n = 6, 1-way ANOVA. (E) In vitro proliferation of LLC-Luc cells treated with ERa, ERb or GPER selective antagonists (MPP 10−8 M, PHTPP 10−8 M, G15 10−7 M) in combination or not with E2 (10−9 M), n = 6, 1-way ANOVA. (F) In vivo LLC-Luc cell proliferation analysed by EdU immunofluorescent staining (scale bar = 500 µm in upper panel, scale bar = 100 µm in lower panel) in OVX female mice treated or not with E2. (G) Quantification of EdU density (EdU area/lung tumour area) on LLC-Luc tumours from OVX females (n = 26) and OVX + E2 females (n = 23), *P < 0.05, t-test. (H) Double-immunofluorescent staining and their zoom of LYVE1 (red) and ERa (green), colocalisation of stainings (yellow) and DAPI (blue) in LLC-Luc lung tumours, scale bar = 50 µm. Uterus tissue was used as positive control, scale bar = 50 µm. (I) Double-immunofluorescent staining and their zoom of CD31 (red) and ERa (green), colocalisation of stainings (yellow) and DAPI (blue) in LLC-Luc lung tumours, scale bar = 50 µm.

  • View in gallery

    E2 increases lymph/angiogenesis through ERa. (A) Immunofluorescent staining (scale bar = 50 µm) of LYVE1 (green) or CD31 (red) in LLC-Luc lung tumours from female (n = 11), OVX (n = 15) or OVX + E2-treated (n = 18) female mice, and from male (n = 12), Cx (n = 5) or Cx + E2-treated (n = 5) male mice. (B) Quantification of LYVE1 density (LYVE1 area/lung tumour area) and CD31 density (CD31 area/lung tumour area) in lung tumours of these mice, *P < 0.05; **P < 0.01, Mann–Whitney or 1-way ANOVA. (C) VEGFC, (D) VEGFD, (E) VEGFA and (F) bFGF levels measured by Milliplex and reported to the total amount of protein (mg) in lung tumour lysates issued from male, female, OVX female and OVX + E2 female mice, *P < 0.05, 1-way ANOVA. (G) CD31/LYVE1 staining (for each condition: left panel, scale bar = 1mm; right panel, scale bar = 500 µm) and quantification (CD31 or LYVE1 area/total cornea area) in cornea of female mice (n = 10), OVX (n = 10) or OVX + E2 (n = 10) mice, *P < 0.05; ***P < 0.001, 1-way ANOVA, or (H) in cornea of male (n = 18) or castrated male mice (Cx, n = 8), Mann–Whitney.

  • View in gallery

    Prolymphangiogenic and proangiogenic effects of E2 are mediated by ERa. (A) Hematoxylin/eosin (H/E) staining of LLC-Luc lung tumours issued from Tie2-Cre/ERalox/lox male (n = 7) or female (n = 7) mice and from Tie2-Cre+/ERalox/lox male (n = 8) or female (n = 5) mice (scale bar = 1mm) and quantification of the tumour density (tumour area/total lung area), 8 slides spaced by 50 µm were analysed per sample, *P < 0.05, Mann–Whitney. (B) Hematoxylin/eosin (H/E) staining of LLC-Luc lung tumours from Tie2-Cre/ERalox/lox OVX (n = 6), OVX + E2 (n = 6) female mice and from Tie2-Cre+/ERalox/lox OVX (n = 7), OVX + E2 (n = 5) female mice (scale bar = 1mm) and quantification of the tumour density (tumour area/total lung area), 8 slides spaced by 50 µm were analysed per sample, *P < 0.05, Mann–Whitney. (C) LYVE1 immunofluorescent staining (scale bar = 50 µm) and quantification (LYVE1 area/lung tumour area) in LLC-Luc lung tumours in Tie2-Cre+/ERalox/lox male (n = 8), female (n = 5), female OVX (n = 7) and female OVX + E2 (n = 5) mice, Kruskal–Wallis. (D) CD31 immunofluorescent staining (scale bar = 50 µm) and quantification (CD31 area/lung tumour area) in LLC-Luc lung tumours in Tie2-Cre+/ERalox/lox male (n = 8), female (n = 5), female OVX (n = 7) and female OVX + E2 (n = 5) Kruskal–Wallis. (E) LYVE1/CD31 staining (scale bar = 500 µm) and quantification (LYVE1 or CD31 area/total cornea area) of blood and lymphatic vessels in cornea of Tie2-Cre+/ERalox/lox male (n = 8), female (n = 5), OVX female (n = 7) and OVX + E2 female (n = 5) mice, Kruskal–Wallis. (F) LYVE1/CD31 staining (scale bar = 500 µm) and quantification (LYVE1 or CD31 area/total cornea area) of blood and lymphatic vessels in cornea of wild-type female mice treated with vehicle (n = 11), ERa antagonist (MPP 10−8 M, n = 13), ERb antagonist (PHTPP 10−8 M, n = 12) or GPER antagonist (G15 10−7 M, n = 9), *P < 0.05, Kruskal–Wallis.

  • View in gallery

    Lung tumour treatment with ERa/ERb antagonists or with tamoxifen. (A) In vivo bioluminescent signals and quantification of LLC-Luc lung tumours in female mice treated with vehicle (control group), ERa antagonist (MPP, 1 mg/kg), ERb antagonist (PHTPP, 1 mg/kg), n = 8; *P < 0.05, 2-way ANOVA. (B) Hematoxylin/eosin staining of LLC-Luc lung tumours (scale bar = 1mm) and quantification of tumour density (lung tumour area/total lung area) in females treated with vehicle; MPP; PHTPP and Tamoxifen (Tmx), *P < 0.05, Mann–Whitney. (C) Immunofluorescent staining (scale bar = 50 μm) and quantification (LYVE1 area/lung tumour area) of LYVE1 positive vessels in LLC-Luc lung tumours of female mice treated with vehicle (n = 7), MPP (n = 8); PHTPP (n = 7) or Tmx (n = 8), *P < 0.05, **P < 0.01, 1-way ANOVA. (D) Immunofluorescent staining of CD31 positive blood vessels in LLC-Luc lung tumours (scale bar = 50 μm) and quantification of the CD31 density (CD31 area/lung tumour area) in tumours of female mice treated with vehicle (n = 16), MPP (n = 8); PHTPP (n = 8) or Tmx (n = 8), **P < 0.01, ***P < 0.001, Kruskal -Wallis. Milliplex analysis of (E) VEGFC, (F) VEGFD, (G) VEGFA, (H) bFGF concentrations in LLC-Luc lung tumour lysates from vehicle-, MPP- and Tmx-treated females; *P < 0.05, **P < 0.01, 1-way ANOVA. (I) In vivo bioluminescent signals and quantification of LLC-Luc-derived bioluminescence in lungs of male mice treated with vehicle (control group), ERa antagonist (MPP, 1 mg/kg), ERb antagonist (PHTPP, 1mg/kg) (n = 8), 2-way ANOVA. (J) Hematoxylin/eosin staining of LLC-Luc lung tumours (scale bar = 1mm) and quantification of tumour density (lung tumour area/total lung area) from males treated with vehicle, MPP, PHTPP and Tamoxifen (Tmx), Kruskal–Wallis. (K) Immunofluorescent staining (scale bar = 50 μm) and quantification (LYVE1 area/lung tumour area) of LYVE1 positive vessels in LLC-Luc lung tumours of male mice, treated with vehicle (n = 8), MPP (n = 6); PHTPP (n = 7) or Tmx (n = 7), 1-way ANOVA. (L) Immunofluorescent staining of CD31 positive blood vessels in LLC-Luc lung tumours (scale bar = 50 μm) and quantification of the CD31 density (CD31 area/lung tumour area) in lung tumours of male mice (n = 17), treated with vehicle (n = 8), MPP (n = 7); PHTPP (n = 7) or Tmx (n = 7), Kruskal–Wallis.

  • View in gallery

    Lymph/angiogenic vasculature and ERa status in human lung adenocarcinoma biopsies. (A) Characteristics of human lung tumours (histology, age, smoker status). (B) Representative immunohistochemical staining of ERa (blue) in human lung tumour biopsies according to sex (men: n = 23, women: n = 51) and separated as ERa-positive (ERa+) and ERa-negative (ER−) tumours, scale bar = 50 µm. (C) Immunofluorescent staining of PDPN and quantification of PDPN density (PDPN stained tumour area/lung tumour area) in human lung tumours according to sex (men n = 23, women n = 49) and ERa expression (ERa+: me n = 14, women n = 30; Era−: men n = 9, women n = 19), *P < 0.05; **P < 0.01; ***P < 0.001, Mann–Whitney, scale bar = 250 µm on upper panel and 50 µm on lower panel. (D) Immunohistochemical staining of CD31 and quantification (CD31 stained tumour area/lung tumour area) in human lung tumours according to sex (men n = 23, women n = 51) and ERa expression (ERa+: me n = 14, women n = 30; ERa−: men n = 9, women n = 19), *P < 0.05, t-test, scale bar = 250 µm on upper panel and 50 µm on lower panel. (E) Representative ERa (green), PDPN (red) and CD31 (red) mRNA detection by RNAscope on human lung tumour sections (scale bar = 10 µm). Endometrium is used as positive control for ERa expression. Cell nuclei are stained with DAPI.

PubMed

Google Scholar