Limits...
Intravital imaging reveals how BRAF inhibition generates drug-tolerant microenvironments with high integrin β1/FAK signaling.

Hirata E, Girotti MR, Viros A, Hooper S, Spencer-Dene B, Matsuda M, Larkin J, Marais R, Sahai E - Cancer Cell (2015)

Bottom Line: Intravital imaging of BRAF-mutant melanoma cells containing an ERK/MAPK biosensor reveals how the tumor microenvironment affects response to BRAF inhibition by PLX4720.Fibronectin-rich matrices with 3-12 kPa elastic modulus are sufficient to provide PLX4720 tolerance.We propose that paradoxically activated MAFs provide a "safe haven" for melanoma cells to tolerate BRAF inhibition.

View Article: PubMed Central - PubMed

Affiliation: Tumor Cell Biology Laboratory, Cancer Research UK London Research Institute, London WC2A 3LY, UK.

Show MeSH

Related in: MedlinePlus

Stromal Cells Provide Safe Haven to Tolerate BRAF Inhibition for Drug-Sensitive Human Melanoma Cells(A and B) Growth curves of the indicated melanoma cells in vitro (A) (mean ± SD) and in vivo (B) treated with DMSO (0.1% in vitro and 4% in vivo) or PLX4720 (1 μM in vitro and 25 mg/kg in vivo).(C) Intravital images of A375- and WM266.4-expressing EKAREV-NLS in nude mice. Images are from mice treated with DMSO (4%) or PLX4720 (25 mg/kg) for the indicated days. Arrows in high magnification images of DMSO-treated tumors indicate small cages created by thick collagen fibers (detected by second harmonic generation: SHG). In PLX4720-treated tumors, the regions in white squares are magnified and shown below. Distribution (scatterplots with mean ± SD) and histogram of ERK activities in A375 and WM266.4 tumors are shown at the bottom. Scale bars represent 500 μm (low magnification) and 100 μm (high magnification).(D) Frozen sections of WM266.4 tumors treated with DMSO or PLX4720 for 15 days were double-stained with anti-fibronectin (in green) and anti-tenascin-C (in red) antibodies. Scale bars represent 200 μm.(E) A375 cells stably expressing firefly luciferase grown subcutaneously in nude mice were treated with DMSO (4%), PLX4720 (25 mg/kg), PF562271 (50 mg/kg), or the combination. Bioluminescence from the tumors was analyzed at day 25 and shown (mean ± SD). (φ) Please note that DMSO group (n = 4) contains fewer mice than other groups because the higher growth rates of these tumors meant that some mice were killed before day 25.(F) A375 tumors treated with the indicated drug(s) for 29 days were stained with an anti-Ki-67 antibody. Scale bar represents 100 μm.(G) Ki-67 index was measured from three different positions in six different tumors treated with the indicated drugs for 18–29 days.Data are represented as mean ± SD. Please note that necrotic areas with myeloid cell infiltration in PLX4720 + PF562271 tumors were excluded from the analysis. See also Figure S6.
© Copyright Policy - CC BY-NC-ND
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC4402404&req=5

fig6: Stromal Cells Provide Safe Haven to Tolerate BRAF Inhibition for Drug-Sensitive Human Melanoma Cells(A and B) Growth curves of the indicated melanoma cells in vitro (A) (mean ± SD) and in vivo (B) treated with DMSO (0.1% in vitro and 4% in vivo) or PLX4720 (1 μM in vitro and 25 mg/kg in vivo).(C) Intravital images of A375- and WM266.4-expressing EKAREV-NLS in nude mice. Images are from mice treated with DMSO (4%) or PLX4720 (25 mg/kg) for the indicated days. Arrows in high magnification images of DMSO-treated tumors indicate small cages created by thick collagen fibers (detected by second harmonic generation: SHG). In PLX4720-treated tumors, the regions in white squares are magnified and shown below. Distribution (scatterplots with mean ± SD) and histogram of ERK activities in A375 and WM266.4 tumors are shown at the bottom. Scale bars represent 500 μm (low magnification) and 100 μm (high magnification).(D) Frozen sections of WM266.4 tumors treated with DMSO or PLX4720 for 15 days were double-stained with anti-fibronectin (in green) and anti-tenascin-C (in red) antibodies. Scale bars represent 200 μm.(E) A375 cells stably expressing firefly luciferase grown subcutaneously in nude mice were treated with DMSO (4%), PLX4720 (25 mg/kg), PF562271 (50 mg/kg), or the combination. Bioluminescence from the tumors was analyzed at day 25 and shown (mean ± SD). (φ) Please note that DMSO group (n = 4) contains fewer mice than other groups because the higher growth rates of these tumors meant that some mice were killed before day 25.(F) A375 tumors treated with the indicated drug(s) for 29 days were stained with an anti-Ki-67 antibody. Scale bar represents 100 μm.(G) Ki-67 index was measured from three different positions in six different tumors treated with the indicated drugs for 18–29 days.Data are represented as mean ± SD. Please note that necrotic areas with myeloid cell infiltration in PLX4720 + PF562271 tumors were excluded from the analysis. See also Figure S6.

Mentions: The data described above demonstrate that MAFs provide a mechanism for mouse BRAF-mutant melanoma cell lines refractory to PLX4720 in vivo. However, the majority of human BRAF-mutant melanoma respond well to BRAF inhibition before the emergence of resistant disease over many months (Chapman et al., 2011). We were interested whether our findings regarding the role of the stroma in providing drug tolerance were relevant to the survival of melanoma cells in between the initial administration of BRAF inhibitors and the ultimate emergence of genetically resistant cells. Therefore, we examined ERK/MAPK activity and stromal changes in two models of human melanoma that exhibit a clear response to BRAF inhibition in vivo. Figures 6A and 6B show that A375 and WM266.4 human melanoma cells are sensitive to PLX4720 both in vitro and in vivo. However, tumors did not disappear and typically 2–4 mm of residual disease remained during PLX4720 treatment. We hypothesized that this residual disease may be supported by signals from the stroma. Intravital imaging of the EKAREV biosensor revealed that the residual disease after 11–14 days of PLX4720 treatment exhibited similar levels of ERK/MAPK signaling to the pre-treatment tumors (Figures 6C and S6A). Based on our analysis of 5555 and 4434 models, we predict the following: there should be changes in the stroma of PLX4720-treated A375 and WM266.4 tumors, the melanoma cells should not be intrinsically resistant to PLX4720 at this stage, but that they should use FAK- and Src-dependent signaling to sustain ERK/MAPK activity and survival. Both A375 and WM266.4 exhibited clear changes in collagen SHG when treated with PLX4720 (Figure 6C). This is consistent with increased matrix deposition and remodeling by fibroblasts. Immunohistochemical staining, and Gomori’s trichrome staining confirmed that the residual disease was rich in fibroblastic stroma and had higher levels of fibrillar collagen, FN and TNC (Figures 6D and S6A). Furthermore, conformation specific antibodies revealed increased active integrin β1 levels in PLX-treated tumors (Figure S6B).


Intravital imaging reveals how BRAF inhibition generates drug-tolerant microenvironments with high integrin β1/FAK signaling.

Hirata E, Girotti MR, Viros A, Hooper S, Spencer-Dene B, Matsuda M, Larkin J, Marais R, Sahai E - Cancer Cell (2015)

Stromal Cells Provide Safe Haven to Tolerate BRAF Inhibition for Drug-Sensitive Human Melanoma Cells(A and B) Growth curves of the indicated melanoma cells in vitro (A) (mean ± SD) and in vivo (B) treated with DMSO (0.1% in vitro and 4% in vivo) or PLX4720 (1 μM in vitro and 25 mg/kg in vivo).(C) Intravital images of A375- and WM266.4-expressing EKAREV-NLS in nude mice. Images are from mice treated with DMSO (4%) or PLX4720 (25 mg/kg) for the indicated days. Arrows in high magnification images of DMSO-treated tumors indicate small cages created by thick collagen fibers (detected by second harmonic generation: SHG). In PLX4720-treated tumors, the regions in white squares are magnified and shown below. Distribution (scatterplots with mean ± SD) and histogram of ERK activities in A375 and WM266.4 tumors are shown at the bottom. Scale bars represent 500 μm (low magnification) and 100 μm (high magnification).(D) Frozen sections of WM266.4 tumors treated with DMSO or PLX4720 for 15 days were double-stained with anti-fibronectin (in green) and anti-tenascin-C (in red) antibodies. Scale bars represent 200 μm.(E) A375 cells stably expressing firefly luciferase grown subcutaneously in nude mice were treated with DMSO (4%), PLX4720 (25 mg/kg), PF562271 (50 mg/kg), or the combination. Bioluminescence from the tumors was analyzed at day 25 and shown (mean ± SD). (φ) Please note that DMSO group (n = 4) contains fewer mice than other groups because the higher growth rates of these tumors meant that some mice were killed before day 25.(F) A375 tumors treated with the indicated drug(s) for 29 days were stained with an anti-Ki-67 antibody. Scale bar represents 100 μm.(G) Ki-67 index was measured from three different positions in six different tumors treated with the indicated drugs for 18–29 days.Data are represented as mean ± SD. Please note that necrotic areas with myeloid cell infiltration in PLX4720 + PF562271 tumors were excluded from the analysis. See also Figure S6.
© Copyright Policy - CC BY-NC-ND
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC4402404&req=5

fig6: Stromal Cells Provide Safe Haven to Tolerate BRAF Inhibition for Drug-Sensitive Human Melanoma Cells(A and B) Growth curves of the indicated melanoma cells in vitro (A) (mean ± SD) and in vivo (B) treated with DMSO (0.1% in vitro and 4% in vivo) or PLX4720 (1 μM in vitro and 25 mg/kg in vivo).(C) Intravital images of A375- and WM266.4-expressing EKAREV-NLS in nude mice. Images are from mice treated with DMSO (4%) or PLX4720 (25 mg/kg) for the indicated days. Arrows in high magnification images of DMSO-treated tumors indicate small cages created by thick collagen fibers (detected by second harmonic generation: SHG). In PLX4720-treated tumors, the regions in white squares are magnified and shown below. Distribution (scatterplots with mean ± SD) and histogram of ERK activities in A375 and WM266.4 tumors are shown at the bottom. Scale bars represent 500 μm (low magnification) and 100 μm (high magnification).(D) Frozen sections of WM266.4 tumors treated with DMSO or PLX4720 for 15 days were double-stained with anti-fibronectin (in green) and anti-tenascin-C (in red) antibodies. Scale bars represent 200 μm.(E) A375 cells stably expressing firefly luciferase grown subcutaneously in nude mice were treated with DMSO (4%), PLX4720 (25 mg/kg), PF562271 (50 mg/kg), or the combination. Bioluminescence from the tumors was analyzed at day 25 and shown (mean ± SD). (φ) Please note that DMSO group (n = 4) contains fewer mice than other groups because the higher growth rates of these tumors meant that some mice were killed before day 25.(F) A375 tumors treated with the indicated drug(s) for 29 days were stained with an anti-Ki-67 antibody. Scale bar represents 100 μm.(G) Ki-67 index was measured from three different positions in six different tumors treated with the indicated drugs for 18–29 days.Data are represented as mean ± SD. Please note that necrotic areas with myeloid cell infiltration in PLX4720 + PF562271 tumors were excluded from the analysis. See also Figure S6.
Mentions: The data described above demonstrate that MAFs provide a mechanism for mouse BRAF-mutant melanoma cell lines refractory to PLX4720 in vivo. However, the majority of human BRAF-mutant melanoma respond well to BRAF inhibition before the emergence of resistant disease over many months (Chapman et al., 2011). We were interested whether our findings regarding the role of the stroma in providing drug tolerance were relevant to the survival of melanoma cells in between the initial administration of BRAF inhibitors and the ultimate emergence of genetically resistant cells. Therefore, we examined ERK/MAPK activity and stromal changes in two models of human melanoma that exhibit a clear response to BRAF inhibition in vivo. Figures 6A and 6B show that A375 and WM266.4 human melanoma cells are sensitive to PLX4720 both in vitro and in vivo. However, tumors did not disappear and typically 2–4 mm of residual disease remained during PLX4720 treatment. We hypothesized that this residual disease may be supported by signals from the stroma. Intravital imaging of the EKAREV biosensor revealed that the residual disease after 11–14 days of PLX4720 treatment exhibited similar levels of ERK/MAPK signaling to the pre-treatment tumors (Figures 6C and S6A). Based on our analysis of 5555 and 4434 models, we predict the following: there should be changes in the stroma of PLX4720-treated A375 and WM266.4 tumors, the melanoma cells should not be intrinsically resistant to PLX4720 at this stage, but that they should use FAK- and Src-dependent signaling to sustain ERK/MAPK activity and survival. Both A375 and WM266.4 exhibited clear changes in collagen SHG when treated with PLX4720 (Figure 6C). This is consistent with increased matrix deposition and remodeling by fibroblasts. Immunohistochemical staining, and Gomori’s trichrome staining confirmed that the residual disease was rich in fibroblastic stroma and had higher levels of fibrillar collagen, FN and TNC (Figures 6D and S6A). Furthermore, conformation specific antibodies revealed increased active integrin β1 levels in PLX-treated tumors (Figure S6B).

Bottom Line: Intravital imaging of BRAF-mutant melanoma cells containing an ERK/MAPK biosensor reveals how the tumor microenvironment affects response to BRAF inhibition by PLX4720.Fibronectin-rich matrices with 3-12 kPa elastic modulus are sufficient to provide PLX4720 tolerance.We propose that paradoxically activated MAFs provide a "safe haven" for melanoma cells to tolerate BRAF inhibition.

View Article: PubMed Central - PubMed

Affiliation: Tumor Cell Biology Laboratory, Cancer Research UK London Research Institute, London WC2A 3LY, UK.

Show MeSH
Related in: MedlinePlus