Oncogene (2008) 27, 3201–3210
& 2008 Nature Publishing Group All rights reserved 0950-9232/08 $30.00 www.nature.com/onc
ORIGINAL ARTICLE
Epidermal growth factor receptor activation in prostate cancer by three novel missense mutations CQ Cai1,7, Y Peng1,7, MT Buckley2, J Wei1, F Chen1, L Liebes2, WL Gerald3, MR Pincus4,5, I Osman2,6 and P Lee1,4 1 Department of Pathology, New York University School of Medicine, New York, NY, USA; 2Department of Medicine, New York University School of Medicine, New York, NY, USA; 3Department of Pathology, Memorial Sloan-Kettering Cancer Center, New York, NY, USA; 4New York Harbor Healthcare System, New York, NY, USA; 5Department of Pathology, State University of New York, New York, NY, USA and 6Department of Urology, New York University School of Medicine, New York, NY, USA
While epidermal growth factor receptor (EGFR) dysregulation is known to play a critical role in prostate carcinogenesis, there has been no direct evidence indicating EGFR mutations induce tumorigenesis in prostate cancer. We previously identified four novel EGFR somatic mutations in the EGFR tyrosine kinase domain of prostate cancer patients: G735S, G796S, E804G and R841K. In this study, we investigated the oncogenic potential of these somatic mutations by establishing stable clonal NIH3T3 cells expressing these four mutations and WT EGFR to determine their ability to increase cell proliferation and invasion. In the absence of the EGF ligand, cell proliferation was readily increased in G735S, G796S and E804G mutants compared to WT EGFR. The addition of EGF ligand greatly increased cell growth and transforming ability of these same EGFR mutants. Matrigel invasion assays showed enhanced invasion with G735S, G796S and E804G mutants. Western blot analysis showed that these EGFR mutations enhanced cell growth and invasion via constitutive and hyperactive tyrosine phosphorylation and led to the activation of mitogen-activated protein kinase (MAPK), signal transducer and activator of transcription 3 (STAT3) and Akt pathways. Our findings demonstrate the oncogenic activation of three novel EGFR somatic missense mutations in prostate cancer. Molecules that regulate the mechanisms of their oncogenic activation represent novel targets for limiting tumor cell progression, and further elucidation of these mutations will have utility in prostate cancer treatment. Oncogene (2008) 27, 3201–3210; doi:10.1038/sj.onc.1210983; published online 14 January 2008 Keywords: EGFR; mutation; prostate cancer
Correspondence: Dr P Lee, Department of Pathology, New York University School of Medicine, 423 E. 23rd Street, Room 6140N, New York, NY 10010, USA. E-mail:
[email protected] and I Osman, Department of Urology and Medicine, New York University School of Medicine, 522 First Avenue, SML 405, New York, NY 10016, USA. E-mail:
[email protected] 7 These authors have contributed equally to this work. Received 19 April 2007; revised 8 October 2007; accepted 10 October 2007; published online 14 January 2008
Introduction Epidermal growth factor receptors (EGFRs) are a family of transmembrane protein tyrosine-kinase receptors that are targeted by a variety of growth signals or ligands and form homo- or heterodimers in a variety of cell types (Ushiro and Cohen, 1980; Schreiber et al., 1983). Upon ligand binding, phosphorylation of tyrosine kinase creates docking sites for signaling molecules containing PTB or SH2 domains, which then leads to a phosphorylation cascade of downstream proteins, including members of the Akt, mitogen-activated protein kinase (MAPK) and signal transducer and activator of transcription 3 (STAT3) pathways (Yarden and Sliwkowski, 2001; Citri and Yarden, 2006). Aberrant signaling or abnormal activation through ligand overproduction, receptor overexpression or gain-of-function mutations can drive tumorigenesis in a broad spectrum of human cancers (Yarden and Sliwkowski, 2001; Hubbard, 2005). Dysregulation of EGFR expression and function is a common feature of a number of tumors. Recently, either point or deletion mutations of the EGFR kinase domain have been associated with a number of human cancers, including lung, head/neck and brain tumors (Paez et al., 2004; Pao et al., 2004; Bell et al., 2005; Lee et al., 2005, 2006; Nagahara et al., 2005). The L858R EGFR mutation, also alternatively identified as L834R for the human EGFR sequence occuring in lung and head neck cancer, has been extensively studied and serves as a prototype for EGFR kinase domain mutational studies in many other cancers. L858R is located in the activation loop encoded by exon 21 (Pao et al., 2004; Arteaga, 2006; Zhang et al., 2006; Choi et al., 2007). Several groups have demonstrated the role of increased EGFR signaling in prostate carcinogenesis and progression (Sherwood et al., 1998; Ratan et al., 2003; Torring et al., 2003). We have recently reported the discovery of four somatic missense mutations in the tyrosine kinase domain of EGFR in prostate cancer patients (Douglas et al., 2006). The discovery of these mutations in the EGFR is important because it may lead to the identification of progressive yet treatable disease in cancer patients. In this study, we evaluated the
EGFR activation in prostate cancer CQ Cai et al
3202
biological function of these four EGFR mutants in terms of increased cellular proliferation and invasion. Furthermore, we assessed the ability of these EGFR tyrosine kinase domain mutations to differentially activate downstream signaling pathways, including Akt, MAPK and STAT3.
Results Construction of clonal NIH3T3 cells stably expressing wild-type and mutated EGFR The four kinase domain mutants, that is, G735S, G796S, E804G and R841K were created in retroviral mammalian expression pBabe vectors by site-directed mutagenesis using WT EGFR as a template (Figures 1a–d). To express either WT or mutated EGFR in clonal stable cell lines, we transduced pBabe-G735S, pBabe-G796S, pBabe-E804G, pBabe-R841K, WT EGFR and pBabe (vector alone) into NIH3T3 cells with
DNA
Amino Acid
735
AGT
Serine
WT
GGT
Glycine
puromycine as the selection marker. As shown in Figures 2a and b, expression of WT EGFR and mutant EGFR (G735S, G796S, E804G and R841K) in the absence and presence of EGF was analysed by western blot. Net band intensities using a densitometer of G735S, G796S, E804G and R841K are 22 223±781 (mean±s.e.m.) in the absence of EGF ligand. In the presence of EGF ligand, EGFR was also expressed at comparable levels with 31 835±1247 (mean±s.e.m.), indicating that the system was valid. b-Actin was used as a loading control for the western blot analysis. Three clonal stable cell lines for each mutant, as well as WT EGFR and vector alone (pBabe) were used in the experiments detailed below. Increased cell proliferation by mutated EGFR To determine the growth regulatory effects on cell proliferation of the four EGFR kinase domain mutations (Figures 1a–d) occurring in prostate cancer, we performed cell counting experiments using NIH3T3
DNA
Amino Acid
796
AG C
Serine
WT
GGC
Glycine
DNA
Amino Acid
DNA
Amino Acid
804
GGA
Glycine
841
AAG
Lysine
WT
GAA
Glutamic Acid
WT
AGG
Arginine
Figure 1 Confirmation of G735S, G796S, E804G and R841K EGFR mutations in pBabe-G735S, pBabe-G796S, pBabe-E804G and pBabe-R841K constructs by sequencing. Oncogene
EGFR activation in prostate cancer CQ Cai et al
3203
stable cell lines transduced with each mutant, as well as WT EGFR and vector alone. The experiments were performed in both the absence and presence of EGF (50 ng ml1). Cell counts were performed in triplicate with three independent clones. Addition of 50 ng ml1 EGF to the cells expressing WT EGFR receptor increased cell growth, indicating that the system was appropriate for this study. Two of the four mutants (G796S and E804G) showed that their expression directly increased cell growth compared to WT EGFR, both in the absence of EGF (Figure 2c), and more dramatically, in the presence of EGF (Figure 2d). In the absence of EGF, cells with mutant E804G showed the fastest growth with a cell number of 264.6±4.3 (mean±s.e.m.) ( 104) on day 5 compared to WT (156.7±4.8), followed by G796S with 236.7±7.5. This suggested that the mutated receptors were capable of inducing autonomous cell growth. The growth stimulation was more evident with the addition of EGF ligand for all mutants E804G, G796S, G735S and R841K with cell numbers of 331.3±5.9, 276.6±6.4, 219.3±5.8 and 205.3±3.5, respectively (Figure 2d), compared to WT EGFR (185.3±2.9) on day 5. The increased cell proliferation effect over WT was most
significant for mutants E804G and G796S in both the absence and presence of EGF. Notably, in the absence of EGF, cells expressing mutant R841K did not reveal any growth advantage compared to cells with pBabe vector alone or WT EGFR. With the addition of EGF on day 5, this mutant showed a slight increase in growth over the WT EGFR. Mutations G735S, G796S and E804G increased cell transforming ability of mutated EGFR To assess the transforming potential of these EGFR kinase mutations, anchorage-independent growth assays were performed (Di Fiore et al., 1987; Danielsen and Maihle, 2002). The G735S, G796S and E804G mutants were able to transform NIH3T3 cells to anchorageindependent growth in the absence of exogenous EGF (Figures 3a and b). Cells stably expressing mutants G735S, E804G and G796S, respectively, showed 12.0-, 11.0- and 5.0-fold increases in their ability to form colonies in suspension media versus cells stably expressing WT EGFR. As previously described (Di Fiore et al., 1987; Greulich et al., 2005), WT EGFR transformed cells after the EGF addition. Upon the addition of EGF to the soft agar medium, there was a
R841K
E804G
G796S
G735S
WT
pBabe
E804G
R841K
EGF
G796S
G735S
WT
pBabe
No EGF
EGFR
300
400
pBabe
pBabe + EGF
WT
WT + EGF
G735S
G735S + EGF
300
G796S
200
Cell Number (x104)
Cell Number (x104)
250
E804G R841K
150
100
G796S + EGF E804G + EGF
200
R841K + EGF
100
50 0
0 0
2
3
Day
4
5
0
2
3
4
5
Day
Figure 2 Enhanced cell proliferation by epidermal growth factor receptor (EGFR) mutation in the absence and presence of EGF. (a, b) Establishing clonal NIH3T3 cell lines stably expressing WT and mutants G735S, G796S, E804G and R841K EGFR. G735S, G796S, E804G and R841K and WT EGFR were expressed at comparable levels in the absence of EGF ligand (a, upper panel). In the presence of EGF ligand (50 ng ml1; b, upper panel), EGFR was also expressed at comparable levels. The same blots of the EGFR panels were reprobed with anti-b-actin antibody as a loading control (a, b, lower panels). (c, d) For growth curves, three independent single-clonal cell lines were chosen based on their similar expression level of WT or mutant EGFRs and performed at least in triplicate. One representative experiment of three is shown. (c) Increased growth of NIH3T3 cells stably expressing EGFR mutants G796S and E804G in the absence of EGF. (d) Dramatically increased growth of NIH3T3 cells stably expressing EGFR mutants G735S, G796S and E804G in the presence of EGF (50 ng ml1). NIH3T3 cells expressing R841K mutation showed comparable growth rate as WT EGFR in the absence of EGF (c) and moderately increased growth in the presence of EGF (d). Oncogene
EGFR activation in prostate cancer CQ Cai et al
3204 WT
pBabe
G735S
G796S
E804G
R841K
No EGF
EGF
R841K
E804G
G735S
WT
pBabe
R841K
EGF
E804G
WT
pBabe
Colonies
40
G796S
No EGF
EGF 60
G796S
No EGF
G735S
80
EGFR
20
0 pBabe
WT
G735S
G796S
E804G
R841K
Figure 3 Enhanced cell transforming ability by epidermal growth factor receptor (EGFR) mutations. (a) Cells expressing the prostate cancer-derived EGFR mutants had markedly increased transformation potential. Microphotographs were taken at 4 objective. In the absence of EGF (upper panel), colonies formed by cells stably expressing the three oncogenic mutants G735S, G796S and E804G were generally less in number and smaller in size, compared to those in the presence of 50 ng ml1 EGF (lower panel). The cells stably expressing vector construct alone and mutant R841K did not form colonies, and WT formed colonies only in the presence of 50 ng ml1 EGF but substantially less than those formed by cells expressing G735S, G796S and E804G. (b) Graphical representation of colonies counted microscopically in (a). Error bars represent s.d. (c) EGFR was expressed at a similar level (upper panels) for all samples when anchorage-independent growth was performed. EGFR immunoblots were re-probed with b-actin antibody as a loading control (lower panels) in both the absence (left panels) and presence (right panels) of EGF.
significant increase in the number of colonies (Figures 3a and b) of cells expressing WT EGFR and all three mutants G735S, G796S and E804G. There were 3.7-, 2.6- and 2.3-fold increases in the number of colonies for cells expressing mutants G735S, E804G and G796S, respectively, over cells expressing WT EGFR. The sizes of colonies in media with EGF were larger than that without EGF, especially for cells expressing mutant G735S (Figure 3a). Cells expressing EGFR mutant R841K failed to induce colony formation in either the presence or absence of EGF. The above cell growth and anchorage-independent experiments strongly indicate the oncogenic effects of EGFR mutations G735S, G796S and E804G in prostate cancer.
G735S and E804G invaded through the reconstituted basement membrane system (Figure 4a) and showed 4- and 11-fold increases over WT invasion, respectively. In the presence of EGF, cells expressing mutants G735S, G796S and E804G revealed greater ability to invade through Matrigel membrane and showed 6-, 13and 25-fold increases in invasion over WT invasion, respectively (Figures 4a and b). These changes are statistically significant (Po0.001). Cells expressing EGFR mutant R841K were not able to invade either in the presence or absence of EGF. EGFR was expressed at a similar level for all samples when invasion assays were performed, and anti-b-actin immunoblots served as loading control (Figure 4c).
Enhanced invasion by EGFR mutants We further examined the in vitro invasion ability of NIH3T3 cells stably expressing EGFR mutants compared with those expressing WT EGFR by BD Matrigel invasion assays (Chu et al., 1993; Sato et al., 1994). NIH3T3 is a noninvasive cell line (Sato et al., 1994). Expression of WT EGFR promoted cell invasion only in the presence of EGF (50 ng ml1) (Figure 4a). In contrast to WT, in the absence of EGF, cells expressing mutants
Constitutive and hyperactivation of EGFR kinase domain by oncogenic mutation EGF activates EGFR through phosphorylation of specific tyrosine residues in its intracellular kinase domain. There are five tyrosine residues in the tail region of the intracellular kinase domain capable of being phosphorylated upon activation, that is, pTyr845, pTyr992, pTyr1045, pTyr1068 and pTyr1173, leading to subsequent activation of various signal transduction
Oncogene
EGFR activation in prostate cancer CQ Cai et al
3205
a
pBabe
WT
G735S
G796S
E804G
R841K
No EGF
EGF
b
1500 No EGF
R841K
G796S
E804G
WT
pBabe
R841K
EGF
E804G
G796S
WT
500
G735S
No EGF
G735S
c
1000
pBabe
Cell Number
EGF
EGFR 0 pBabe
WT
G735S
G796S
E804G
R841K
Figure 4 Stimulation of cell invasion by epidermal growth factor receptor (EGFR) mutations. (a) Elevated invasive potential conferred by mutant EGFR from prostate cancer patients. Microphotographs were taken at 40 objective both in the presence and in the absence (upper panels) and presence (lower panels) of EGF (50 ng ml1). (b) Graphical representation of invasive colonies counted microscopically in (a). Error bars represent s.d. (c) EGFR immunoblots show similar expression levels of EGFR receptors in all conditions used in (a). EGFR immunoblots were re-probed with b-actin antibody as a loading control (lower panels).
pathways. We examined the phosphorylation of these five tyrosine residues, in both the absence and presence of EGF ligand (50 ng ml1), by western blot analysis in NIH3T3 cells stably expressing the WT and four mutants using site-specific phosphotyrosine antibodies (Figure 5a). Each western blot analysis was performed in triplicate and scanned to obtain average levels of total and phospho proteins at each site of tyrosine phosphorylation for each mutant compared with WT EGFR. The levels of expression were normalized to b-actin in Figures 5b and c. In the absence of EGF, cells expressing the three mutants (E804G, G735S, G796S) revealed increased phosphorylation in the order of magnitude E804G>G735S>G796S at tyrosine sites with the following order of activity: Y992>Y1173> Y1068>Y845 (Figure 5a, left panel and 5b). The most prominent activation (phosphorylation) mutation was in cells expressing mutant E804G at site Y1068 (average band intensity as 72.8±22.6), followed by G735S (23.5±6.4) and then G796S (21.7±5.6) (Figure 5a, left panel) compared to zero signal of WT EGFR. Cells harboring mutant R841K showed phosphorylation at only Y992, but at a similar level to WT (Figure 5a, left panel). Addition of EGF ligand induced phosphorylation at levels above WT in the following order of magnitude in cells expressing mutants E804G>G796S>G735S, and
in contrast to the absence of ligand, the presence of EGF ligand resulted in the increase in phosphorylation at all five tyrosine residues in the following order of activity Y1068>Y1173XY992>Y845>Y1045, compared to WT (Figure 5a, right panel and 5c). The most prominent activation mutation was in cells expressing mutant E804G at site Y1068 (3.3-fold increase over WT), followed by slight increase in activation for G796S (2.3-fold increase over WT) and G735S (1.5-fold increase over WT) mutations. Cells expressing EGFR mutant R841K showed phosphorylation at only Y992, but at a lower level of expression comparable to that in WT EGFR in the presence of EGF. EGF-independent and -dependent activation of Akt, STAT3 and MAPK pathways with EGFR oncogenic mutants EGFR exerts its function through the activation of multiple signal transduction pathways (Figure 6a), including Akt/PKB, STAT3 and MAPK. We first employed monoclonal total and phospho-Akt antibodies to determine if the EGFR mutations could lead to either constitutively active or hyperactive Akt activity in the presence or absence of EGF ligand stimulation. Total Akt was expressed in comparable levels either in the absence (Figure 6b, left panel and 6c) or presence Oncogene
EGFR activation in prostate cancer CQ Cai et al
3206 R841K
E804G
G796S
G735S
WT
pBabe
R841K
EGF E804G
G796S
G735S
WT
pBabe
No EGF
Y845 Y992 Y1045 Y1068 Y1173
EGFR
pBabe
250
WT
G735S
G796S
E804G
R841K
Level of Phosphorylation
200
150
100
50
0
Tyr 845
Tyr 992 pBabe
WT
Tyr 1045 G735S + EGF
Tyr 1068 G796S +EGF
E804G +EGF
Tyr1173 R841K +EGF
250
Level of Phosphorylation
200
150
100
50
0
Tyr 845
Tyr 992
Tyr 1045
Tyr 1068
Tyr1173
Figure 5 Activation of tyrosine kinase domain by epidermal growth factor receptor (EGFR) mutation. (a) Phosphorylation of four kinase domain mutations occurring in prostate cancer by western blot analysis using site-specific phospho-tyrosine antibodies in the absence (left panels) and presence (right panels) of EGF. The expression level of WT and mutant EGFRs was equal in all blots and b-actin was used as a loading control. (b) Graphical representation of EGFR phosphorylation sites for cells expressing mutants G735S, G796S, E804G and R841K in the absence of EGF ligand expressed as fold change over WT EGFR. (c) Graphical representation of EGFR phosphorylation for cells expressing mutants G735S, G796S, E804G and R841K in the presence of EGF ligand (50 ng ml1) expressed as fold change over WT EGFR.
Oncogene
EGFR activation in prostate cancer CQ Cai et al
3207 EGF Receptor 845
STAT5
1045
degradation
992
Akt/PKB
1068
MAPK/ERK
STAT3
1173
R841K
E804G
G796S
G735S
WT
pBabe
R841K
EGF E804G
G796S
WT
G735S
pBabe
No EGF
P-Akt Akt P-Erk Erk P-STAT3 STAT3 EGFR
pBabe
300
WT
G735S
G796S
E804G
R841K
200
100
0 P-Akt
Akt
P-Erk pBabe
300
Erk
P-STAT3 G735S + EGF
WT
G796S +EGF
STAT3 E804G +EGF
R841K +EGF
200
100
6
0 P-Akt
Akt
P-Erk
Erk
P-STAT3
STAT3
Figure 6 Activation of signal transduction pathways by epidermal growth factor receptor (EGFR) mutations. (a) Graphical representation of the activation of molecular signaling pathways. (b) The levels of total and phospho proteins for Akt, mitogenactivated protein kinase (MAPK) and signal transducer and activator of transcription 3 (STAT3) in cells stably expressing mutant EGFRs G735S, G796S, E804G and R841K in the absence (left panels) and presence (right panels) of EGF (50 ng ml1). The expression level of WT and mutant EGFRs was equal in all blots and b-actin was used as a loading control. (c) Graphical representation of Akt, MAPK and STAT phosphorylation for in cells stably expressing EGFR mutants G735S, G796S, E804G and R841K in the absence of EGF ligand. (d) Graphical representation of Akt, MAPK and STAT3 phosphorylation for cells stably expressing EGFR mutants G735S, G796S, E804G and R841K in the presence of EGF ligand.
Oncogene
EGFR activation in prostate cancer CQ Cai et al
3208
(Figure 6b, right panel and 6d) among these mutations. In the absence of EGF, phosphorylated Akt was not significantly elevated in cells stably expressing all three oncogenic EGFR mutants, compared to that in cells harboring WT EGFR and pBabe vector alone (Figure 6b, left panel and 6c). In the presence of EGF ligand (Figure 6b, right panel and 6d), cells expressing all four mutants showed increases in Akt activation of 1.8-, 1.7-, 1.6- and 1.3-fold for G735S, G796S, E804G and R841K compared to WT EGFR, respectively. Thus, three phosphorylated tyrosines (Y992, Y1068 and Y1173) can presumably transmit synergistic signals via activated Akt, which may contribute to the oncogenic phenotypes observed for all three oncogenic mutants in a model cell culture system. Signal transducers and activators of transcription (STAT) and MAPK are also important kinase pathways activated by EGFR (Sordella et al., 2004). Thus we also examined activation of STAT3 and MAPK by our EGFR mutants using western blot analysis with total and phospho antibodies to these two proteins. Once again, the levels of expression in total STAT3 and MAPK were at equivalent levels for cells expressing all mutants (Figures 6b–d). In the absence of EGF ligand, there was an increase in activation of phospho-STAT3 by G796S, R841K and E804G versus WT with fold increases as 2.2, 1.9 and 1.8 respectively (Figure 6b, left panel and 6c). The level of phospho-STAT3 in cells harboring G735S was at similar level with WT. The phospho-MAPK pathway showed constitutive activation with significant increases in band intensity of 5.4-, 3.9-, 2.7- and 1.5-fold, respectively, for cells expressing mutants G735S, R841K, G796S and E804G compared to WT and pBabe vector alone. However, in the presence of EGF, there was no obvious increase of both phospho-STAT3 and phospho-MAPK for cells expressing the three oncogenic mutations compared to WT (Figure 6b, right panel and 6d).
Discussion This is the first study on the functional relevance of somatic EGFR kinase domain mutations, that is, G735S, G796S, E804G, R841K, in prostate cancer. Most EGFR mutations in lung and head/neck cancer reside in the activation loop or alpha helix C region (Lynch et al., 2004; Pao et al., 2004; Lee et al., 2005; Nagahara et al., 2005). These mutations disrupt the autoinhibition of interaction between intracellular kinase domains and render the mutated kinase constitutively active (Greulich et al., 2005; Zhang et al., 2006; Choi et al., 2007). Interestingly, while they occur in the kinase domain of EGFR in prostate cancer, none of the four novel somatic EGFR mutations are located in the activation loop or alpha helix C region. Therefore, it was imperative to determine whether these mutations are cancer epiphenomena or are oncogenic in nature. In this study, we have shown that three (G735S, G796S and E804G) of the four EGFR mutations Oncogene
identified in prostate cancer patients are oncogenic in NIH3T3 cells. NIH3T3 cells harboring each of the three mutations showed increased growth, transformation and invasion compared to cells expressing WT EGFR in the absence of EGF stimulation. As the three oncogenic mutations demonstrated increased phosphorylation of four tyrosine residues compared to WT EGFR in the absence of ligand, this indicates that these three kinase domain mutations are constitutively active. Another common feature for these mutations is that they are all hyperactive in response to ligand stimulation, and hyperphosphorylation of EGFR caused increased growth, transformation and invasion. The most active and significant somatic missense mutation in the kinase domain was E804G. In the absence of EGF ligand, NIH3T3 cells harboring the E804G mutation showed an increased ability for growth and invasion due to the constitutive activation of four tyrosine residues Y845, Y992, Y1068 and Y1173. Upon ligand stimulation, E804G showed the most hyperactive activation and subsequently greatest aggressiveness in growth and invasion. Though all three oncogenic mutations have aggressive behavior in cell proliferation and invasion assays, they have different degrees of aggressiveness in biological activities and signal transduction pathway activation. Mutations E804G and G796S conferred the greatest growth potential, highest transformation and invasion ability. Of note, the NIH3T3 cells stably expressing mutant G796S demonstrated a greater invasion ability than NIH3T3 cells bearing mutation L858R, a frequent mutation in lung cancer (data not shown). Epidermal growth factor receptor (EGFR) exerts its function through a complex network of signaling pathways including Akt, pSTAT3 and MAPK. We have elucidated the important pathways affected by these oncogenic EGFR mutants. In the presence of EGF, we found that in all three oncogenic mutants, phosphorylated Akt was substantially increased, compared to WT EGFR. Thus, inhibition of phosphorylated Akt might be a potential therapeutic target for prostate cancer patients who harbor this distinct set of mutations in their EGFR kinase domain. Although it has been reported that STAT3 activation inhibits LNCaP cell growth (Spiotto and Chung, 2000), many reports indicate that activated STAT3 is expressed in prostate cancer and that STAT3 activation in prostate cancer cells results in cell proliferation (Giri et al., 2001; DeMiguel et al., 2002; Huang et al., 2005). Furthermore, inhibitors of Stat 3 lead to the apoptosis of prostate cancer cells (Barton et al., 2004; Bellezza et al., 2006). Of the three mutations, EGF-independent activation of STAT3 in cells stably expressing the mutants G796S and E804G was notably observed. In addition, phosphorylation of MAPK in cells harboring mutants G735S and G796S was also observed, which may contribute to the increased colony formation and invasive potential conferred by these mutants. Of the three oncogenic mutations, cells expressing the mutant E804G were phosphorylated at the highest levels (at Y992, Y1068 and Y1173), which presumably leads
EGFR activation in prostate cancer CQ Cai et al
3209
to signal transduction and downstream activation of mitogenic and/or anti-apoptotic pathways (Hanahan and Weinberg, 2000; Jorissen et al., 2003; Shao et al., 2003; Sordella et al., 2004; Citri and Yarden, 2006). However, we did not observe the expected extent of activation for MAPK and/or STAT3 with cells that expressed mutant E804G. Substantially elevated activation of MAPK and STAT3 seen in an EGF-independent manner disappeared upon addition of 50 ng ml1 EGF, indicating signaling other than STAT3 or MAPK might be involved. Several lines of evidence indicate that the role of EGFR kinase domain mutations in response to EGFR kinase inhibitors may be clinically relevant. In lung cancer, cells whose growth depends on EGFR with mutations in exons 19 and 21 are sensitive to tyrosine kinase inhibitors of EGFR (Lynch et al., 2004; Paez et al., 2004; Pao et al., 2004; Bell et al., 2005; Choi et al., 2007). Glioblastoma cells transformed by expression of these EGFR mutants were sensitive to small-molecule EGFR kinase inhibitors (Lee et al., 2006). We are in the process of testing anti-EGFR drugs, including several classes of EGFR inhibitors (small molecules, monoclonal antibodies and the HSP90 inhibitor 17AAG) to determine whether these mutations in prostate cancer have differential responses to EGFR inhibitors. In summary, we have demonstrated the oncogenic nature of three EGFR mutations in prostate cancer due to constitutive and hyperactive functions of the mutated EGFR.
Materials and methods Expression constructs for EGFR mutants Retroviral EGFR expression constructs containing puromycin (pBabe-Puro-EGFR) (Greulich et al., 2005) (a gift from Dr Meyerson) were used for site-directed mutagenesis using the Quick-Change Mutagenesis XL kit (Stratagene, La Jolla, CA, USA) with the following primers: for mutant G735S: 50 GGACTCTGGATCCCAGAAAGTGAGAAAGTTAAAAT TCCC-30 and 50 -GGGAATTTTAACTTTCTCACTTTCTGG GATCCAGAGTCC-30 ; for mutant G796S: 50 -GCAGCTCAT GCCCTTCAGCTGCCTCCTGGACTATG-30 and 50 -CATA GTCCAGGAGGCAGCTGAAGGGCATGAGCTGC-30 ; for mutant E804G: 50 -CTCCTGGACTATGTGCCGGGACACA AAGACAATATTGGC-30 and 50 -GCCAATATTGTCTTTG TGTCCCCGGACATAGTCCAGGAG-30 ; for mutant R841K: 50 -CCGCGACCTGGCAGCCAAGAACGTACTGGTGAAA AC-30 and 50 -GTTTTCACCAGTACGTTCTTGGCTGCCA GGTCGCGG-30 . Construction of clonal NIH3T3 cells stably expressing wild-type and mutated EGFR Replication incompetent retroviruses were produced from pBabe-Puro-based (pBabe) vector with four kinase domain mutants, that is, G735S, G796S, E804G and R841K mutants, by transfection into the Phoenix 293T packaging cell line using lipofectamine (Invitrogen, Carlsbad, CA, USA). The packaging cells were grown in Dulbecco’s modification of Eagle’s medium (DMEM; Gibco, Carlsbad, CA, USA) with 10% fetal bovine serum. NIH3T3 cells (ATCC) were infected with retroviruses in the presence of 8.0 mg ml1 polybrene (Sigma
Aldrich, St Louis, MO, USA). Twenty-four hours post infection, 2.5 mg ml1 polybrene was added to the media and pooled cell lines were selected. Three single-clonal cell lines for each condition were derived using a 96-well plate format. Cell proliferation assays WT EGFR and pBabe vector transfected cells were used as controls. Cells (2.5 104) with WT and mutated EGFR G735S, G796S, E804G and R841K were seeded into 24-well plates in the presence or absence of EGF. For experiments with EGF (Invitrogen), 50 ng ml1 EGF was added day at the time of cell plating and day 3. Cells were counted microscopically at three random fields from each condition on days 2, 3, 4 and 5 after plating. Cell counts were performed using three independent clones for each condition. Anchorage-independent growth assay EGFR-expressing NIH3T3 cells or kinase domain-mutated EGFR (5 104) were suspended in a top layer of DMEM containing 10% bovine calf serum (BCS) and 0.35% SeaPlaque GTG Agarose (Cambrex, Rockland, ME, USA) and plated on a bottom layer of DMEM containing 10% BCS and 0.5% SeaPlaque GTG agarose in 24-well plates. EGF (50 ng ml1) was added to the top agar. Three weeks after plating, colonies (defined as a group of 50 or more cells) were counted in triplicate wells from nine random fields. Matrigel invasion assays Cells (5 104 cells) were suspended in 0.5 ml, and 0.5% bovine serum albumin was plated on the insert of Matrigel invasion chamber (Becton Dickinson/Biocoat, Bedford, MA, USA). Inserts were placed on top of 0.75 ml DMEM with 10% BCS (as chemoattractant) located in the lower chambers ±50 ng ml1 EGF. After 24 h of incubation, cells on the other side of Matrigel membranes were stained using a Diff-Quik stain (Richard-Allan Scientific, Kalamazoo, MI, USA) and counted microscopically on the entire field at 40 objective magnification. Immunoblotting Cells were stimulated with EGF (50 ng ml1) for 30 min at 37 1C. Whole cell extracts were prepared from NIH3T3 cells with either WT or mutant EGFR or pBabe vector construct alone with lysis buffer (50 mM HEPES, 150 mM NaCl, 1 mM EDTA, 1 mM EGTA, 10% glycerol, 1% Triton, 25 mM NaF, 10 mM ZnCl) in the presence of protease inhibitor cocktail (Roche Applied Science, Indianapolis, IN, USA) at pH 7.5. Lysates were run on 12% SDS–PAGE gels then transferred to a nitrocellulose membrane for western blot analysis. Primary antibodies specific for phosphorylation at Tyr 845, Tyr 992, Tyr 1045, Tyr 1068 and Tyr 1173 (Santa Cruz, Santa Cruz, CA, USA), phospho-Akt (Ser 473), phospho-STAT3 (Tyr705) and phospho-STAT5 (Tyr694), total Akt and STAT3 (Cell Signaling Technology, Danvers, MA, USA), ERK1/2 and phospho-ERK1/2 (Thr202/Tyr204) (Santa Cruz) were used. Immunoblots were incubated with the primary antibodies for 2 h and secondary antibody (Amersham Biosciences, Piscataway, NJ, USA) for 1.5 h at room temperature. Protein bands were detected by enhanced chemiluminescence (Amersham Biosciences), and analysis of net band intensity was determined using Kodak Molecular Imaging Software version 4.0. Statistical analysis Cell proliferation, transforming and invasion experiments were performed at least three times with similar results. All Oncogene
EGFR activation in prostate cancer CQ Cai et al
3210 statistical analysis was performed using GraphPad Instat version 3.0 (Graph Pad Software Inc., San Diego, CA, USA). Statistical significance was calculated using the Student’s two-tailed t-test, where Po0.05 was considered significant.
Acknowledgements We thank Jessie Yu for editorial assistance. This work is supported by an MSKCC SPORE grant to IO, DOD grants to IO and PL.
References Arteaga CL. (2006). EGF receptor mutations in lung cancer: from humans to mice and maybe back to humans. Cancer Cell 9: 421–423. Barton BE, Karras JG, Murphy TF, Barton A, Huang HF. (2004). Signal transducer and activator of transcription 3 (STAT3) activation in prostate cancer: direct STAT3 inhibition induces apoptosis in prostate cancer lines. Mol Cancer Ther 3: 11–20. Bell DW, Lynch TJ, Haserlat SM, Harris PL, Okimoto RA, Brannigan BW et al. (2005). Epidermal growth factor receptor mutations and gene amplification in non-small-cell lung cancer: molecular analysis of the IDEAL/INTACT gefitinib trials. J Clin Oncol 23: 8081–8092. Bellezza I, Neuwirt H, Nemes C, Cavarretta IT, Puhr M, Steiner H et al. (2006). Suppressor of cytokine signaling-3 antagonizes cAMP effects on proliferation and apoptosis and is expressed in human prostate cancer. Am J Pathol 169: 2199–2208. Choi SH, Mendrola JM, Lemmon MA. (2007). EGF-independent activation of cell-surface EGF receptors harboring mutations found in gefitinib-sensitive lung cancer. Oncogene 26: 1567–1576. Chu YW, Runyan RB, Oshima RG, Hendrix MJ. (1993). Expression of complete keratin filaments in mouse L cells augments cell migration and invasion. Proc Natl Acad Sci USA 90: 4261–4265. Citri A, Yarden Y. (2006). EGF-ERBB signalling: towards the systems level. Nat Rev Mol Cell Biol 7: 505–516. Danielsen AJ, Maihle NJ. (2002). Ligand-independent oncogenic transformation by the EGF receptor requires kinase domain catalytic activity. Exp Cell Res 275: 9–16. DeMiguel F, Lee SO, Lou W, Xiao X, Pflug BR, Nelson JB et al. (2002). Stat3 enhances the growth of LNCaP human prostate cancer cells in intact and castrated male nude mice. Prostate 52: 123–129. Di Fiore PP, Pierce JH, Fleming TP, Hazan R, Ullrich A, King CR et al. (1987). Overexpression of the human EGF receptor confers an EGF-dependent transformed phenotype to NIH 3T3 cells. Cell 51: 1063–1070. Douglas D, Zhong H, Ro J, Oddoux C, Berger A, Pincus M et al. (2006). Racial difference in somatic mutations of epidermal growth factor receptor in prostate cancer. Front Biosci 11: 2518–2525. Giri D, Ozen M, Ittmann M. (2001). Interleukin-6 is an autocrine growth factor in human prostate cancer. Am J Pathol 159: 2159–2165. Greulich H, Chen TH, Feng W, Janne PA, Alvarez JV, Zappaterra M et al. (2005). Oncogenic transformation by inhibitor-sensitive and -resistant EGFR mutants. PLoS Med 2: e313. Hanahan D, Weinberg RA. (2000). The hallmarks of cancer. Cell 100: 57–70. Huang HF, Murphy TF, Shu P, Barton AB, Barton BE. (2005). Stable expression of constitutively-activated STAT3 in benign prostatic epithelial cells changes their phenotype to that resembling malignant cells. Mol Cancer 4: 2. Hubbard SR. (2005). EGF receptor inhibition: attacks on multiple fronts. Cancer Cell 7: 287–288. Jorissen RN, Walker F, Pouliot N, Garrett TP, Ward CW, Burgess AW. (2003). Epidermal growth factor receptor: mechanisms of activation and signaling. Exp Cell Res 284: 31–53. Lee JC, Vivanco I, Beroukhim R, Huang JH, Feng WL, Debiasi RM et al. (2006). Epidermal growth factor receptor activation in
Oncogene
glioblastoma through novel missense mutations in the extracellular domain. PLoS Med 3: e485. Lee JW, Soung YH, Kim SY, Nam HK, Park WS, Nam SW et al. (2005). Somatic mutations of EGFR gene in squamous cell carcinoma of the head and neck. Clin Cancer Res 11: 2879–2882. Lynch TJ, Bell DW, Sordella R, Gurubhagavatula S, Okimoto RA, Brannigan BW et al. (2004). Activating mutations in the epidermal growth factor receptor underlying responsiveness of non-small-cell lung cancer to gefitinib. N Engl J Med 350: 2129–2139. Nagahara H, Mimori K, Ohta M, Utsunomiya T, Inoue H, Barnard GF et al. (2005). Somatic mutations of epidermal growth factor receptor in colorectal carcinoma. Clin Cancer Res 11: 1368–1371. Paez JG, Janne PA, Lee JC, Tracy S, Greulich H, Gabriel S et al. (2004). EGFR mutations in lung cancer: correlation with clinical response to gefitinib therapy. Science 304: 1497–1500. Pao W, Miller V, Zakowski M, Doherty J, Politi K, Sarkaria I et al. (2004). EGFR receptor gene mutations are common in lung cancers from ‘never smokers’ and are associated with sensitivity of tumors to gefitinib and erlotinib. Proc Natl Acad Sci USA 101: 13306–13311. Ratan HL, Gescher A, Steward WP, Mellon JK. (2003). ErbB receptors: possible therapeutic targets in prostate cancer? BJU Int 92: 890–895. Sato H, Takino T, Okada Y, Cao J, Shinagawa A, Yamamoto E et al. (1994). A matrix metalloproteinase expressed on the surface of invasive tumour cells. Nature 370: 61–65. Schreiber AB, Libermann TA, Lax I, Yarden Y, Schlessinger J. (1983). Biological role of epidermal growth factor-receptor clustering. Investigation with monoclonal anti-receptor antibodies. J Biol Chem 258: 846–853. Shao H, Cheng HY, Cook RG, Tweardy DJ. (2003). Identification and characterization of signal transducer and activator of transcription 3 recruitment sites within the epidermal growth factor receptor. Cancer Res 63: 3923–3930. Sherwood ER, Van Dongen JL, Wood CG, Liao S, Kozlowski JM, Lee C. (1998). Epidermal growth factor receptor activation in androgen-independent but not androgen-stimulated growth of human prostatic carcinoma cells. Br J Cancer 77: 855–861. Sordella R, Bell DW, Haber DA, Settleman J. (2004). Gefitinibsensitizing EGFR mutations in lung cancer activate anti-apoptotic pathways. Science 305: 1163–1167. Spiotto MT, Chung TD. (2000). STAT3 mediates IL-6-induced growth inhibition in the human prostate cancer cell line LNCaP. Prostate 42: 88–98. Torring N, Dagnaes-Hansen F, Sorensen BS, Nexo E, Hynes NE. (2003). ErbB1 and prostate cancer: ErbB1 activity is essential for androgen-induced proliferation and protection from the apoptotic effects of LY294002. Prostate 56: 142–149. Ushiro H, Cohen S. (1980). Identification of phosphotyrosine as a product of epidermal growth factor-activated protein kinase in A-431 cell membranes. J Biol Chem 255: 8363–8365. Yarden Y, Sliwkowski MX. (2001). Untangling the ErbB signalling network. Nat Rev Mol Cell Biol 2: 127–137. Zhang X, Gureasko J, Shen K, Cole PA, Kuriyan J. (2006). An allosteric mechanism for activation of the kinase domain of epidermal growth factor receptor. Cell 125: 1137–1149.
