Limits...
Integrin alphav-mediated inactivation of p53 controls a MEK1-dependent melanoma cell survival pathway in three-dimensional collagen.

Bao W, Strömblad S - J. Cell Biol. (2004)

Bottom Line: We found that integrin alphav inactivated p53 and that suppression of p53 activity by dominant negative p53 or p53-small interfering RNA obviated the need for integrin alphav for melanoma cell survival in 3D-collagen and for tumor growth in vivo.Furthermore, we found that melanoma cell integrin alphav was required for MAPK kinase (MEK) 1 and extracellular signal-regulated kinase (ERK)1/2 activity in 3D-collagen, whereas inhibition of MEK1 activity induced apoptosis.Surprisingly, MEK1 and ERK1/2 activities were restored in integrin alphav-negative melanoma cells by suppression of p53, whereas concomitant block of MEK1 induced apoptosis.

View Article: PubMed Central - PubMed

Affiliation: Department of Laboratory Medicine, Karolinska Institutet, Stockholm, 141 86, Sweden.

ABSTRACT
Integrin alphav is required for melanoma cell survival and tumor growth in various models. To elucidate integrin alphav-mediated melanoma cell survival mechanisms, we used a three-dimensional (3D) collagen gel model mimicking the pathophysiological microenvironment of malignant melanoma in the dermis. We found that integrin alphav inactivated p53 and that suppression of p53 activity by dominant negative p53 or p53-small interfering RNA obviated the need for integrin alphav for melanoma cell survival in 3D-collagen and for tumor growth in vivo. This indicates that integrin alphav-mediated inactivation of p53 functionally controls melanoma cell survival. Furthermore, we found that melanoma cell integrin alphav was required for MAPK kinase (MEK) 1 and extracellular signal-regulated kinase (ERK)1/2 activity in 3D-collagen, whereas inhibition of MEK1 activity induced apoptosis. Surprisingly, MEK1 and ERK1/2 activities were restored in integrin alphav-negative melanoma cells by suppression of p53, whereas concomitant block of MEK1 induced apoptosis. This suggests that integrin alphav controls melanoma cell survival in 3D-collagen through a pathway involving p53 regulation of MEK1 signaling.

Show MeSH

Related in: MedlinePlus

Integrin αv inhibits p53 activity in 3D-collagen. (A) Nuclear extracts from integrin αv-negative M21L cells cultured in 3D-collagen for 5 d were used for detecting p53 DNA-binding activity by EMSA. Specific p53 supershift was obtained by anti-p53 mAb Pab421. Specific or nonspecific competition was generated by incubating the nuclear extracts with the 32P-labeled probe in the presence or absence of 50-fold excess of unlabeled specific or nonspecific oligonucleotides. (B and D) p53-DNA binding activity was detected by EMSA in M21 (αv+) and M21L (αv−) cells (B) and in M0 (αv−) and M0-αv (αv+) cells (D) under 2D conditions (d 0) or within 3D-collagen for the indicated times. Large bar graphs show quantifications of p53 DNA-binding activity EMSA supershift bands displayed immediately below. Boxed bar graphs show the mean ± SD of the ratio of p53 DNA-binding activity bands between M21L (αv−) and M21 (αv+) (B) or between M0 (αv−) and M0-αv (αv+) (D) at each time point based on three to four independent experiments as described in Materials and methods. (* −P < 0.05; ** −P < 0.01, as compared with the corresponding p53 DNA-binding activity ratios (αv−/αv+) at d 0 using unpaired two-tailed t test). (C and E) p53 protein levels were determined by Western blotting in M21 (αv+) and M21L (αv−) cells (C) and in M0 (αv−) and M0-αv (αv+) cells (E). Bar graphs show mean ± SD of quantifications of p53 protein levels relative to actin loading control from three independent experiments. (F and G) Phosphorylated p53 at Ser 15 and Ser 20 were detected by Western blotting in M21 (αv+) and M21L (αv−) cells (F) and in M0 (αv−) and M0-αv (αv+) cells (G). The protein levels of p53 and actin served as controls. (H–O) The levels of acetylated p53-K382, p300, PUMA, Apaf1, Bax, and Bcl-2 proteins were detected by Western blotting in M21 (αv+) and M21L (αv−) cells (H, J, L, and N) and in M0 (αv−) and M0-αv (αv+) cells (I, K, M, and O) cultured under 2D conditions (d 0) or in 3D-collagen for the indicated times. Bar graphs (H and I) show quantifications of p53-K382-Ac bands relative to total p53 levels of the blots shown below, whereas bar graphs in J–M show quantifications of the respective targeted protein bands relative to the corresponding actin levels in blots displayed below the respective graph. Statistical analyses were performed using t-test by analyzing the ratios between the M21L and M21 or M0 and M0-αv values at each time point compared with the corresponding ratios at d 0 as described in Materials and methods based on three independent experiments. p53-K382-Ac (H and I): P < 0.001 for M21L/M21 at d 7 and M0/M0-αv at d 3 and d 5; P < 0.05 for M21L/M21 at d 3. p300 (J and K): P < 0.001 for M0/M0-αv at d5; P < 0.05 for M21L/M21 at d 5 and d 7 and for M0/M0-αv at d 3. PUMA (L and M): P < 0.001 for M21L/M21 at d 5; P < 0.05 for M21L/M21 at d 7 and M0/M0-αv at d 5. Displayed blots and gel shifts are representative among at least three independent experiments.
© Copyright Policy
Related In: Results  -  Collection


getmorefigures.php?uid=PMC2172581&req=5

fig3: Integrin αv inhibits p53 activity in 3D-collagen. (A) Nuclear extracts from integrin αv-negative M21L cells cultured in 3D-collagen for 5 d were used for detecting p53 DNA-binding activity by EMSA. Specific p53 supershift was obtained by anti-p53 mAb Pab421. Specific or nonspecific competition was generated by incubating the nuclear extracts with the 32P-labeled probe in the presence or absence of 50-fold excess of unlabeled specific or nonspecific oligonucleotides. (B and D) p53-DNA binding activity was detected by EMSA in M21 (αv+) and M21L (αv−) cells (B) and in M0 (αv−) and M0-αv (αv+) cells (D) under 2D conditions (d 0) or within 3D-collagen for the indicated times. Large bar graphs show quantifications of p53 DNA-binding activity EMSA supershift bands displayed immediately below. Boxed bar graphs show the mean ± SD of the ratio of p53 DNA-binding activity bands between M21L (αv−) and M21 (αv+) (B) or between M0 (αv−) and M0-αv (αv+) (D) at each time point based on three to four independent experiments as described in Materials and methods. (* −P < 0.05; ** −P < 0.01, as compared with the corresponding p53 DNA-binding activity ratios (αv−/αv+) at d 0 using unpaired two-tailed t test). (C and E) p53 protein levels were determined by Western blotting in M21 (αv+) and M21L (αv−) cells (C) and in M0 (αv−) and M0-αv (αv+) cells (E). Bar graphs show mean ± SD of quantifications of p53 protein levels relative to actin loading control from three independent experiments. (F and G) Phosphorylated p53 at Ser 15 and Ser 20 were detected by Western blotting in M21 (αv+) and M21L (αv−) cells (F) and in M0 (αv−) and M0-αv (αv+) cells (G). The protein levels of p53 and actin served as controls. (H–O) The levels of acetylated p53-K382, p300, PUMA, Apaf1, Bax, and Bcl-2 proteins were detected by Western blotting in M21 (αv+) and M21L (αv−) cells (H, J, L, and N) and in M0 (αv−) and M0-αv (αv+) cells (I, K, M, and O) cultured under 2D conditions (d 0) or in 3D-collagen for the indicated times. Bar graphs (H and I) show quantifications of p53-K382-Ac bands relative to total p53 levels of the blots shown below, whereas bar graphs in J–M show quantifications of the respective targeted protein bands relative to the corresponding actin levels in blots displayed below the respective graph. Statistical analyses were performed using t-test by analyzing the ratios between the M21L and M21 or M0 and M0-αv values at each time point compared with the corresponding ratios at d 0 as described in Materials and methods based on three independent experiments. p53-K382-Ac (H and I): P < 0.001 for M21L/M21 at d 7 and M0/M0-αv at d 3 and d 5; P < 0.05 for M21L/M21 at d 3. p300 (J and K): P < 0.001 for M0/M0-αv at d5; P < 0.05 for M21L/M21 at d 5 and d 7 and for M0/M0-αv at d 3. PUMA (L and M): P < 0.001 for M21L/M21 at d 5; P < 0.05 for M21L/M21 at d 7 and M0/M0-αv at d 5. Displayed blots and gel shifts are representative among at least three independent experiments.

Mentions: Active p53 can induce the mitochondrial apoptotic pathway (Vousden and Lu, 2002) and activation of p53 has also been associated with block of integrin αvβ3 during angiogenesis (Strömblad et al., 1996, 2002). Therefore, we investigated whether integrin αv may also regulate p53 activity in melanoma cells cultured in 3D-collagen. Using electrophoretic mobility shift analysis (EMSA), we examined p53 DNA-binding activity by detecting specific p53 supershift bands with an anti-p53 mAb. First, we verified the specificity by specific and nonspecific competition (Fig. 3 A). Then, p53 activities were examined in M21 (αv+) and M21L (αv−) cells after exposure in 3D-collagen. We found no difference in p53 activity between M21 (αv+) cells and M21L (αv−) cells before exposure in 3D-collagen (d 0), indicating that p53 activity is not regulated by melanoma cell integrin αv under 2D culture conditions. However, the p53 activity was clearly higher in M21L (αv−) cells as compared with M21 (αv+) cells after 5–7 d in 3D-collagen (Fig. 3 B), whereas p53 protein levels displayed no difference (Fig. 3 C). Consistently, we found that the p53 activity was higher in M0 (αv−) cells than in M0-αv (αv+) cells after exposure in 3D-collagen (Fig. 3 D), but without any change in p53 protein levels (Fig. 3 E). Together, our results demonstrate that melanoma cell integrin αv inhibits p53 DNA-binding activity within a 3D dermal collagen environment.


Integrin alphav-mediated inactivation of p53 controls a MEK1-dependent melanoma cell survival pathway in three-dimensional collagen.

Bao W, Strömblad S - J. Cell Biol. (2004)

Integrin αv inhibits p53 activity in 3D-collagen. (A) Nuclear extracts from integrin αv-negative M21L cells cultured in 3D-collagen for 5 d were used for detecting p53 DNA-binding activity by EMSA. Specific p53 supershift was obtained by anti-p53 mAb Pab421. Specific or nonspecific competition was generated by incubating the nuclear extracts with the 32P-labeled probe in the presence or absence of 50-fold excess of unlabeled specific or nonspecific oligonucleotides. (B and D) p53-DNA binding activity was detected by EMSA in M21 (αv+) and M21L (αv−) cells (B) and in M0 (αv−) and M0-αv (αv+) cells (D) under 2D conditions (d 0) or within 3D-collagen for the indicated times. Large bar graphs show quantifications of p53 DNA-binding activity EMSA supershift bands displayed immediately below. Boxed bar graphs show the mean ± SD of the ratio of p53 DNA-binding activity bands between M21L (αv−) and M21 (αv+) (B) or between M0 (αv−) and M0-αv (αv+) (D) at each time point based on three to four independent experiments as described in Materials and methods. (* −P < 0.05; ** −P < 0.01, as compared with the corresponding p53 DNA-binding activity ratios (αv−/αv+) at d 0 using unpaired two-tailed t test). (C and E) p53 protein levels were determined by Western blotting in M21 (αv+) and M21L (αv−) cells (C) and in M0 (αv−) and M0-αv (αv+) cells (E). Bar graphs show mean ± SD of quantifications of p53 protein levels relative to actin loading control from three independent experiments. (F and G) Phosphorylated p53 at Ser 15 and Ser 20 were detected by Western blotting in M21 (αv+) and M21L (αv−) cells (F) and in M0 (αv−) and M0-αv (αv+) cells (G). The protein levels of p53 and actin served as controls. (H–O) The levels of acetylated p53-K382, p300, PUMA, Apaf1, Bax, and Bcl-2 proteins were detected by Western blotting in M21 (αv+) and M21L (αv−) cells (H, J, L, and N) and in M0 (αv−) and M0-αv (αv+) cells (I, K, M, and O) cultured under 2D conditions (d 0) or in 3D-collagen for the indicated times. Bar graphs (H and I) show quantifications of p53-K382-Ac bands relative to total p53 levels of the blots shown below, whereas bar graphs in J–M show quantifications of the respective targeted protein bands relative to the corresponding actin levels in blots displayed below the respective graph. Statistical analyses were performed using t-test by analyzing the ratios between the M21L and M21 or M0 and M0-αv values at each time point compared with the corresponding ratios at d 0 as described in Materials and methods based on three independent experiments. p53-K382-Ac (H and I): P < 0.001 for M21L/M21 at d 7 and M0/M0-αv at d 3 and d 5; P < 0.05 for M21L/M21 at d 3. p300 (J and K): P < 0.001 for M0/M0-αv at d5; P < 0.05 for M21L/M21 at d 5 and d 7 and for M0/M0-αv at d 3. PUMA (L and M): P < 0.001 for M21L/M21 at d 5; P < 0.05 for M21L/M21 at d 7 and M0/M0-αv at d 5. Displayed blots and gel shifts are representative among at least three independent experiments.
© Copyright Policy
Related In: Results  -  Collection

Show All Figures
getmorefigures.php?uid=PMC2172581&req=5

fig3: Integrin αv inhibits p53 activity in 3D-collagen. (A) Nuclear extracts from integrin αv-negative M21L cells cultured in 3D-collagen for 5 d were used for detecting p53 DNA-binding activity by EMSA. Specific p53 supershift was obtained by anti-p53 mAb Pab421. Specific or nonspecific competition was generated by incubating the nuclear extracts with the 32P-labeled probe in the presence or absence of 50-fold excess of unlabeled specific or nonspecific oligonucleotides. (B and D) p53-DNA binding activity was detected by EMSA in M21 (αv+) and M21L (αv−) cells (B) and in M0 (αv−) and M0-αv (αv+) cells (D) under 2D conditions (d 0) or within 3D-collagen for the indicated times. Large bar graphs show quantifications of p53 DNA-binding activity EMSA supershift bands displayed immediately below. Boxed bar graphs show the mean ± SD of the ratio of p53 DNA-binding activity bands between M21L (αv−) and M21 (αv+) (B) or between M0 (αv−) and M0-αv (αv+) (D) at each time point based on three to four independent experiments as described in Materials and methods. (* −P < 0.05; ** −P < 0.01, as compared with the corresponding p53 DNA-binding activity ratios (αv−/αv+) at d 0 using unpaired two-tailed t test). (C and E) p53 protein levels were determined by Western blotting in M21 (αv+) and M21L (αv−) cells (C) and in M0 (αv−) and M0-αv (αv+) cells (E). Bar graphs show mean ± SD of quantifications of p53 protein levels relative to actin loading control from three independent experiments. (F and G) Phosphorylated p53 at Ser 15 and Ser 20 were detected by Western blotting in M21 (αv+) and M21L (αv−) cells (F) and in M0 (αv−) and M0-αv (αv+) cells (G). The protein levels of p53 and actin served as controls. (H–O) The levels of acetylated p53-K382, p300, PUMA, Apaf1, Bax, and Bcl-2 proteins were detected by Western blotting in M21 (αv+) and M21L (αv−) cells (H, J, L, and N) and in M0 (αv−) and M0-αv (αv+) cells (I, K, M, and O) cultured under 2D conditions (d 0) or in 3D-collagen for the indicated times. Bar graphs (H and I) show quantifications of p53-K382-Ac bands relative to total p53 levels of the blots shown below, whereas bar graphs in J–M show quantifications of the respective targeted protein bands relative to the corresponding actin levels in blots displayed below the respective graph. Statistical analyses were performed using t-test by analyzing the ratios between the M21L and M21 or M0 and M0-αv values at each time point compared with the corresponding ratios at d 0 as described in Materials and methods based on three independent experiments. p53-K382-Ac (H and I): P < 0.001 for M21L/M21 at d 7 and M0/M0-αv at d 3 and d 5; P < 0.05 for M21L/M21 at d 3. p300 (J and K): P < 0.001 for M0/M0-αv at d5; P < 0.05 for M21L/M21 at d 5 and d 7 and for M0/M0-αv at d 3. PUMA (L and M): P < 0.001 for M21L/M21 at d 5; P < 0.05 for M21L/M21 at d 7 and M0/M0-αv at d 5. Displayed blots and gel shifts are representative among at least three independent experiments.
Mentions: Active p53 can induce the mitochondrial apoptotic pathway (Vousden and Lu, 2002) and activation of p53 has also been associated with block of integrin αvβ3 during angiogenesis (Strömblad et al., 1996, 2002). Therefore, we investigated whether integrin αv may also regulate p53 activity in melanoma cells cultured in 3D-collagen. Using electrophoretic mobility shift analysis (EMSA), we examined p53 DNA-binding activity by detecting specific p53 supershift bands with an anti-p53 mAb. First, we verified the specificity by specific and nonspecific competition (Fig. 3 A). Then, p53 activities were examined in M21 (αv+) and M21L (αv−) cells after exposure in 3D-collagen. We found no difference in p53 activity between M21 (αv+) cells and M21L (αv−) cells before exposure in 3D-collagen (d 0), indicating that p53 activity is not regulated by melanoma cell integrin αv under 2D culture conditions. However, the p53 activity was clearly higher in M21L (αv−) cells as compared with M21 (αv+) cells after 5–7 d in 3D-collagen (Fig. 3 B), whereas p53 protein levels displayed no difference (Fig. 3 C). Consistently, we found that the p53 activity was higher in M0 (αv−) cells than in M0-αv (αv+) cells after exposure in 3D-collagen (Fig. 3 D), but without any change in p53 protein levels (Fig. 3 E). Together, our results demonstrate that melanoma cell integrin αv inhibits p53 DNA-binding activity within a 3D dermal collagen environment.

Bottom Line: We found that integrin alphav inactivated p53 and that suppression of p53 activity by dominant negative p53 or p53-small interfering RNA obviated the need for integrin alphav for melanoma cell survival in 3D-collagen and for tumor growth in vivo.Furthermore, we found that melanoma cell integrin alphav was required for MAPK kinase (MEK) 1 and extracellular signal-regulated kinase (ERK)1/2 activity in 3D-collagen, whereas inhibition of MEK1 activity induced apoptosis.Surprisingly, MEK1 and ERK1/2 activities were restored in integrin alphav-negative melanoma cells by suppression of p53, whereas concomitant block of MEK1 induced apoptosis.

View Article: PubMed Central - PubMed

Affiliation: Department of Laboratory Medicine, Karolinska Institutet, Stockholm, 141 86, Sweden.

ABSTRACT
Integrin alphav is required for melanoma cell survival and tumor growth in various models. To elucidate integrin alphav-mediated melanoma cell survival mechanisms, we used a three-dimensional (3D) collagen gel model mimicking the pathophysiological microenvironment of malignant melanoma in the dermis. We found that integrin alphav inactivated p53 and that suppression of p53 activity by dominant negative p53 or p53-small interfering RNA obviated the need for integrin alphav for melanoma cell survival in 3D-collagen and for tumor growth in vivo. This indicates that integrin alphav-mediated inactivation of p53 functionally controls melanoma cell survival. Furthermore, we found that melanoma cell integrin alphav was required for MAPK kinase (MEK) 1 and extracellular signal-regulated kinase (ERK)1/2 activity in 3D-collagen, whereas inhibition of MEK1 activity induced apoptosis. Surprisingly, MEK1 and ERK1/2 activities were restored in integrin alphav-negative melanoma cells by suppression of p53, whereas concomitant block of MEK1 induced apoptosis. This suggests that integrin alphav controls melanoma cell survival in 3D-collagen through a pathway involving p53 regulation of MEK1 signaling.

Show MeSH
Related in: MedlinePlus