Gain of function mutant p53 protein activates AKT through the Rac1 signaling to promote tumorigenesis

ABSTRACT

Tumor suppressor p53 is the most frequently mutated gene in human cancer. Mutant p53 (mutp53) not only loses the tumor suppressive activity of wild type p53, but often gains new oncogenic activities to promote tumorigenesis, defined as mutp53 gain of function (GOF). While the concept of mutp53 GOF is well-established, its underlying mechanism is not well-understood. AKT has been suggested to be activated by mutp53 and contribute to mutp53 GOF, but its underlying mechanism is unclear. In this study, we found that the activation of the Rac1 signaling by mutp53 mediates the promoting effect of mutp53 on AKT activation. Blocking Rac1 signaling by RNAi or a Rac1 inhibitor can inhibit AKT activation by mutp53. Importantly, targeting Rac1/AKT can greatly compromise mutp53 GOF in tumorigenesis. Results from this study uncover a new mechanism for AKT activation in tumors, and reveal that activation of AKT by mutp53 via the Rac1 signaling contributes to mutp53 GOF in tumorigenesis. More importantly, this study provides Rac1 and AKT as potential targets for therapy in tumors containing mutp53.

KEYWORDS:

• Mutant p53
• gain of function
• tumorigenesis
• Rac1
• AKT

Introduction

p53 plays a key role in maintaining genomic stability and tumor suppression [1]. p53 is the most frequently mutated gene in human cancers; more than half of all human cancers contain mutations in the p53 gene [2,3]. Majority of tumor associated-p53 mutations are missense mutations occurring in the DNA binding domain (DBD) of the p53 protein. These p53 missense mutations not only abrogate the tumor suppressive function of wild type (WT) p53, but also render mutant p53 (mutp53) proteins oncogenic activities, defined as the mutp53 gain of function (GOF) [3,4]. Studies have demonstrated an array of mutp53 GOFs, including promoting cell proliferation, migration, invasion and metastasis of tumor cells [5–7].

While it is well-established that many tumor-associated mutp53 proteins show GOF activities, the underlying mechanism of mutp53 GOFs is not well-understood. It has been shown that some GOF mutp53 proteins can interact with transcriptional factors and co-activators, including p63, p73, SREBP, ETS2, etc., and regulate their transcriptional activities, which contributes to mutp53 GOFs [4,8,9]. In addition, mutp53 has been reported to exert its GOFs through activation of some oncogenic signaling pathways by interaction with some proteins in these oncogenic pathways. For instance, mutp53 (R248Q) can interact with STAT3 to activate STAT3, promoting tumorigenesis [10]. Intriguingly, our recent study showed that mutp53 directly binds to small GTPase Rac1 and activates the Rac1 signaling in cancer cells [11]. Aberrant activation of the Rac1 signaling in cancer cells has been reported to promote proliferation, migration, invasion and metastasis of cancer cells [12,13]. Previous reports have shown that some tumor-associated mutp53 proteins activate AKT in lung, breast and endometrial cancer cell lines [6,14,15]. p53 mutations are positively correlated with elevated AKT phosphorylation levels in clinical colon and breast cancer specimens [6,15,16]. However, the mechanism whereby mutp53 activates AKT is not well-understood.

In this study, we identified a new mechanism underlying AKT activation by mutp53. Rac1 is a positive regulator of AKT [17]. Rac1 was reported to activate AKT through its critical downstream effectors, PAKs (p21 activated kinases) [18,19]. We found that the activation of the Rac1 signaling by mutp53 mediates the promoting effect of mutp53 on AKT activation. Blocking the Rac1 signaling by RNAi or a Rac1 inhibitor inhibits AKT activation by mutp53. Importantly, targeting the Rac1/AKT axis greatly compromises mutp53 GOFs. Results from this study demonstrate that Rac1 plays a critical role in mediating mutp53-induced AKT activation, which in turn promotes tumorigenesis. This study reveals a new mechanism whereby mutp53 exerts its GOF and provides Rac1 and AKT as potential targets for therapy in tumors containing mutp53.

Results

Mutp53 activates AKT

To validate the association of mutp53 with AKT activation, we analyzed the association of p53 mutation status and the phosphorylation levels of AKT at Ser473 (p-AKT^S473), which reflects AKT activity [20], in a breast invasive carcinoma cohort from TCGA (TCGA-BRCA). The mutation status of the p53 gene of these specimens was determined by sequencing and the levels of p-AKT^S473 were determined by Reverse phase protein arrays (RPPA) [21]. Tumors containing mutp53 (n = 689) had higher p-AKT^S473 levels than those containing WT p53 (n = 251) (Figure 1(a)). The average levels (log2) of p-AKT^S473 in tumors containing mutp53 and WT p53 are 0.3 ± 0.06 and 0.2 ± 0.03, respectively (p = 0.0128, Student’s t-test) (Figure 1(a)).

Figure 1. Mutp53 is associated with increased AKT phosphorylation levels in a breast cancer cohort and mutp53 activates AKT in cancer cells. (a) The mutation status of the p53 gene is associated with the phosphorylation levels of AKT at Ser473 (p-AKT^S473) in human breast cancer specimens. An invasive breast cancer cohort from TCGA with known p53 mutation status (WT: n = 689; and mutp53: n = 251) and the levels of p-AKT^S473 was used to analyze the association of p53 mutations with p-AKT^S473 levels. Reverse phase protein arrays (RPPA) signals of p-AKT^S473 were transformed to Log2 value. (b) Mutp53 increased AKT activity in human cancer cells as reflected by the levels of pAKT^S473. Ectopic expression of mutp53 (R175 H, R248 W and R273 H) in p53 null H1299 and HCT116 p53^−/- cells by transient transfection of expression vectors increased p-AKT^S473 levels as determined by western-blot assays. (c) Higher levels of p-AKT^S473 in cells with stable ectopic mutp53 expression. Isogenic cell lines, including HCT116 p53^−/-, HCT116 p53^R248 W/-, Saos2-Con and Saos2-mutp53 R175 H cells, were employed to analyze the levels of p-AKT^S473 by western-blot assays. (d) Knockdown of endogenous mutp53 by siRNA in SK-BR-3, MDA-MB-468 and SW480 cells decreased p-AKT^S473 levels.

To confirm that mutp53 increases AKT activity, the levels of p-AKT^S473 were determined in two p53 null human cancer cell lines, HCT116 p53-/- and H1299, with or without transient transfection of vectors expressing several tumor-associated hot spot mutp53 (R175 H, R248 W and R273 H). As shown in Figure 1(b), ectopic expression of mutp53 (R175 H, R248 W and R273 H) in both cell lines increased the levels of p-AKT^S473 but did not affect the levels of total AKT as determined by western-blot assays, indicating that mutp53 enhances AKT activity but not total AKT levels in cells. The levels of p-AKT^S473 were measured in HCT116 p53-/- cells and HCT116 p53^R248 W/- cells, which contain a single copy of p53 gene with R248 W mutation, and p53 null osteosarcoma human Saos2 cells with or without stable expression of mutp53 (R175 H). The levels of p-AKT^S473 were much higher in cells with mutp53 expression than their isogenic p53 null cells (Figure 1(c)). Further, in a panel of human cancer cell lines containing a single allele of endogenous mutp53 gene, including breast SK-BR-3 (R175 H mutp53), MDA-MB-468 (R273 H mutp53) and colorectal SW480 (R273 H/P309 S mutp53) cells, knockdown of the endogenous mutp53 by siRNA in all three cell lines decreased the levels of p-AKT^S473 (Figure 1(d)). Together, these results demonstrate that mutp53 activates AKT in tumor cells.

Mutp53 activates AKT through the Rac1 signaling in vitro and in vivo

Our recent study showed that tumor-associated GOF mutp53 proteins activate small GTPase Rac1 through promoting Rac1 SUMOylation [11]. The Rac1 signaling is involved in a diverse array of cellular processes. The Rac1 signaling is frequently activated in human cancers, which plays a crucial role in promoting tumor development and progression [22]. Rac1 has been shown to be an upstream regulator of the PI3 K signaling [17]. Rac1 cycles between an inactive Rac1-GDP form and an active Rac1-GTP form in cells [12,13]. Activation of Rac1 was reported to promote the phosphorylation PAKs, the critical downstream effectors of Rac1, which in turn activate AKT [18,19]. Here, we investigated whether the activation of Rac1 by mutp53 mediates AKT activation by GOF mutp53. Ectopic expression of Rac1 in H1299 cells by Rac1 vectors increased the levels of Rac1-GTP and the phosphorylation levels of PAK1 at Ser144 and PAK2 at Ser141 (p-PAK1/2), respectively, which reflect the Rac1 activity in cells [23,24], and the levels of p-AKT^S473, but did not affect the levels of total PAK1/2 or AKT (Figure 2(a)). Knockdown of endogenous Rac1 by shRNA vectors decreased the levels of Rac1-GTP, p-PAK1/2 and p-AKT^S473, but did not have a clear effect on the levels of total PAK1/2 and AKT (Figure 2(b)). Further, compared with cells with ectopic expression of Rac1 alone, simultaneous expression of Rac1 and mutp53 (R175 H) increased the levels of p-PAK1/2 and p-AKT^S473 to a much greater extent, which supports the hypothesis that mutp53 activates the Rac1 signaling to increase AKT activity (Figure 2(c)). Importantly, knockdown of endogenous Rac1 by shRNA vectors largely abolished the promoting effect of mutp53 (R175 H) on AKT activation as reflected by the levels of p-AKT^S473 in H1299 cells (Figure 2(d)). Consistently, blocking Rac1 signaling by ectopic expression of the dominant-negative Rac1 (DN-Rac1; Rac1-T17 N) decreased the levels of p-PAK1/2 and p-AKT^S473, and notably, largely abolished the promoting effect of mutp53 (R175 H) on AKT activation in H1299 cells (Figure 2(e)).

Figure 2. Mutp53 activates AKT through the Rac1 signaling in vitro and in vivo. (a) Ectopic Rac1 expression activated PAKs and AKT reflected by the increased phosphorylation levels of PAK1 at Ser144 and PAK2 at Ser141 (p-PAK1/2) and increased p-AKT ^S473 levels. Ectopic expression of Rac1 in p53 null H1299 cells by transient transfection of vectors increased Rac1-GTP, p-PAK1/2 and p-AKT ^S473 levels. (b) Knockdown of endogenous Rac1 by shRNA in H1299 cells decreased Rac1-GTP, p-PAK1/2 and p-AKT ^S473 levels. (c) The levels of p-PAK1/2 and p-AKT ^S473 were determined in H1299 cells with ectopic expression of Rac1 and mutp53 (R175 H) individually or simultaneously. (d) Knockdown of endogenous Rac1 abolished the increase of p-PAK1/2 and p-AKT ^S473 levels induced by mutp53 (R175 H) in H1299 cells. (e) Ectopic expression of the dominant-negative Rac1 (DN-Rac1) largely abolished the increase of p-PAK1/2 and p-AKT ^S473 levels induced by mutp53 (R175 H) in H1299 cells. (f) & (g) Mutp53 (R715 H) increased p-PAK1/2 and p-AKT ^S473 levels in H1299 xenograft tumors. Knockdown of Rac1 by shRNA (F) or expression of DN-Rac1 (G) largely abolished the increase of p-PAK1/2 and p-AKT ^S473 levels induced by mutp53 (R175 H) in H1299 xenograft tumors.

We further investigated the role of Rac1 in mediating the promoting effect of mutp53 on AKT activation in vivo. H1299 cells with or without ectopic expression of mutp53 (R175 H) (H1299-Con and H1299-R175 H cells) were infected with control shRNA vectors or shRNA targeting Rac1 and employed for xenograft tumor formation in nude mice. Tumors were harvested when the tumor volume reached ~200 mm³. H1299-R175 tumors displayed a much faster proliferation rate with higher IHC staining of Ki-67, a marker for cell proliferation, than H1299-Con tumors (Supplementary Fig 1). Knocking down endogenous Rac1 by shRNA vectors largely abolished the promoting effect of mutp53 on cell proliferation in H1299 xenograft tumors (Supplementary Fig 1). The activities of the Rac1 signaling and AKT were much higher in xenograft tumors formed by H1299-R175 H cells compared with tumors formed by H1299-Con cells, indicating that mutp53 increases the activities of Rac1 and AKT in vivo (Figure 2(f)). Knockdown of Rac1 by shRNA vectors clearly decreased AKT activity (Figure 2(f)). Notably, knockdown of Rac1 largely abolished the promoting effect of mutp53 on AKT activation (Figure 2(f)). Similar results were obtained when Rac1 activity was blocked by ectopic expression of the DN-Rac1 in H1299-Con and H1299-R175 H tumors. DN-Rac1 expression largely abolished the promoting effect of mutp53 on cell proliferation in H1299 xenograft tumors (Supplementary Fig 1). Tumors with DN-Rac1 expression showed decreased levels of p-PAK1/2 and p-AKT^S473 (Figure 2(g)). Importantly, DN-Rac1 expression largely abolished the promoting effect of mutp53 on AKT activation in tumors (Figure 2(g)). Together, these results strongly suggest that the activation of Rac1 by mutp53 mediates the promoting effect of mutp53 on AKT activation in tumor cells.

Rac1 inhibitor inhibits AKT activation by mutp53

We further investigated whether targeting Rac1 by employing a small molecule inhibitor targeting Rac1 can block AKT activation promoted by mutp53. Small molecule compound NSC23766 is a widely-used potent Rac1 inhibitor, which can competitively block the binding loop of Rac1-specific guanine nucleotide exchange factors Rac1 activators to prevent the conversion of the inactive Rac1-GDP form to the active Rac1-GTP form [25,26]. H1299 cells with ectopic expression of Rac1 and mutp53 (R175 H) individually or simultaneously were treated with NSC23766 or vehicle. NSC23766 treatment decreased the levels of p-PAK1/2, indicating the inhibition of the Rac1 signaling (Figure 3). Notably, NSC23766 clearly decreased the levels of p-AKT^S473 and largely abolished the promoting effect of Rac1 and mutp53 on AKT activation (Figure 3). These results support that mutp53 activates AKT through the Rac1 signaling, and targeting Rac1 can inhibit AKT activation by mutp53.

Figure 3. The Rac1 inhibitor abolishes AKT activation by mutp53 (R175 H). H1299 cells with or without ectopic expression of Rac1 and mutp53 were treated with the Rac1 inhibitor (NSC23766; 1 µM).

Rac1 SUMOylation promoted by mutp53 is crucial for mutp53-induced AKT activation

Our recent report revealed that mutp53 activates Rac1 through increasing Rac1 SUMOylation levels [11]. Rac1 SUMOylation is critical in maintaining the active Rac1-GTP form to enhance Rac1 activity in cells [27]. Here, we investigated whether Rac1 SUMOylation promoted by mutp53 contributes to mutp53-induced AKT activation. Vectors expressing wild type Rac1 (WT-Rac1-R) or SUMOylation site mutated Rac1 (Rac1 -DSR) that are resistant to Rac1 shRNA were transfected into H1299 cells with knockdown of endogenous Rac1. The expression levels of WT-Rac1-R and Rac1-DSR were comparable to endogenous Rac1 levels in cells (Figure 4(a)). In H1299-R175 H cells, knockdown of endogenous Rac1 abolished AKT activation induced by mutp53, which can be restored by expression of WT-Rac1-R but not Rac1-DSR (Figure 4(a)). These results suggest that promoting Rac1 SUMOylation by mutp53 plays an important role in mutp53-induced AKT activation.

Figure 4. Mutp53 activates AKT through Rac1 activation and Rac1 activation by mutp53 inhibits DAB2IP-AKT interaction. (a) H1299 cells with or without knockdown of endogenous Rac1 were transfected with vectors as indicated. Ectopic expression of mutp53 increased the levels of p-PAK1/2 and p-AKT ^S473 in cells which was abolished by knockdown of endogenous Rac1. The effect of mutp53 on increasing of p-PAK1/2 and p-AKT ^S473 levels was restored in cells with knockdown of endogenous Rac1 by expression of wild type Rac1 resistant to Rac1 shRNA (WT-Rac1-R) but not SUMOylation site mutant Rac1 resistant to Rac1 shRNA (Rac1-DSR). (b) AKT interacted with DAB2IP in H1299 cells transfected with vectors expressing AKT and DAB2IP, respectively. The DAB2IP-AKT interaction was determined by co-IP assays using anti-Flag and anti-DAB2IP antibodies, respectively, followed by western-blot assays. (c) H1299 cells were transfected with vectors expressing DAB2IP together with vectors expressing Rac1 or mutp53 (R175 H) as indicated. The protein levels of DAB2IP were determined by western-blot assays. (d) Rac1 reduced the DAB2IP-AKT interaction, especially in the presence of mutp53. H1299 cells were transfected with vectors as indicated. The DAB2IP-AKT interaction was determined by co-IP assays followed by western-blot assays. (e) Quantification of the DAB2IP-AKT interaction by measuring the relative intensity of DAB2IP pulled down by AKT-Flag (left panel) and the relative intensity of AKT-Flag pulled down by DAB2IP (right panel) in co-IP assays. Data are presented as mean ± SD, n = 3. * p < 0.05, ** p < 0.01, Student’s t-test.

Rac1 activation by mutp53 inhibits DAB2IP-AKT interaction

DAB2IP (DODC-2/DAB2-interacting protein) is a Ras-GAP protein that negatively regulates multiple oncogenic pathways and functions as a tumor suppressor [28]. DAB2IP was reported to bind to AKT and inhibit its activation [16,29]. DAB2IP is frequently down-regulated or inactivated in cancers and its underlying mechanism remains unclear. A previous study showed that DAB2IP can be functionally inactivated through physical interaction with mutp53 proteins [30]. A recent study reported that mutp53 can increase insulin-induced AKT activation by binding to and inhibiting the function of DAB2IP [16]. We confirmed the interaction of AKT with DAB2IP in H1299 cells transfected with expression vectors of AKT1 and DAB2IP by co-Immunoprecipitation (co-IP) assays (Figure 4(b)). Neither Rac1 nor mutp53 appears to change DAB2IP levels; expression of Rac1 or mutp53 R175 H did not affect DAB2IP protein levels in H1299 cells (Figure 4(c)). Interestingly, we found that Rac1 inhibits the interaction of DAB2IP and AKT in H1299 cells. H1299 cells were co-transfected with vectors expressing AKT, DAB2IP, along with vectors expressing Rac1 or control vectors. The interaction of DAB2IP and AKT was determined by co-IP assays. Rac1 clearly decreased the interaction between DAB2IP and AKT (Figure 4(d), left panels). Results from co-IP assays also confirmed the interaction between mutp53 and DAB2IP (Figure 4(d), right panels), and showed no obvious interaction between mutp53 and AKT1 (Figure 4(d), right panels). Notably, the inhibitory effect of Rac1 on the interaction between DAB2IP and AKT was much stronger in H1299 cells with mutp53 expression compared with cells without mutp53 expression. These results suggest that Rac1 activation by mutp53 contributes to the inhibition of DAB2IP-AKT interaction, which in turn activates AKT (Figure 4(d) and Figure 4(e)).

Targeting the Rac1/AKT axis by inhibitors largely blocks mutp53 GOFs

Tumor-associated mutp53 proteins often gain new oncogenic activities to promote tumor development and progression, such as promoting proliferation, migration and invasion of cancer cells [31–34]. The AKT signaling is often activated in many types of cancers, and plays an important role in promoting survival, proliferation and migration of cancer cells [35–37]. Given that mutp53 activates AKT, we tested whether AKT activation by mutp53 contributes to mutp53 GOF in tumorigenesis. To this end, inhibitors targeting the Rac1/AKT axis, including NSC23766 and Wortmannin, which are specific inhibitors for Rac1 and AKT, respectively, were employed to treat H1299 cells with or without ectopic expression of mutp53 (R715 H) (H1299-Con and H1299-R175 H cells). Expression of mutp53 (R175 H) in H1299 cells significantly promoted cell proliferation, colony formation and anchorage-independent cell growth (Figure 5(a–c)). NSC23766 and Wortmannin significantly inhibited cell proliferation, colony formation and anchorage-independent growth in both H1299-Con and H1299-R175 H cells. Most importantly, these inhibitors largely abolished mutp53 GOF in promoting cell proliferation, colony formation and anchorage-independent growth (Figure 5(a–c)).

Figure 5. Targeting the Rac1/AKT axis by inhibitors greatly compromises mutp53 GOF in promoting cell proliferation, anchorage-independent cell growth and migration in H1299 cells. (a) Mutp53 (R175 H) promoted proliferation of H1299 cells, which was greatly compromised by Rac1 inhibitor NSC23766 and AKT inhibitor Wortmannin. H1299 cells with or without ectopic mutp53 (R175 H) expression were treated with NSC23766 (1 µM) or Wortmannin (5 µM). (b) Mutp53 (R175 H) promoted the colony formation of H1299 cells, which was greatly compromised by NSC23766 (100 nM) and Wortmannin (500 nM). Left panel: representative images. Right panel: quantification of the area of colonies/field. (c) Mutp53 (R175 H) promoted the anchorage-independent growth of H1299 cells, which was greatly compromised by NSC23766 (100 nM) and Wortmannin (500 nM). Left panel: representative images. Right panel: quantification of colony numbers per field. (d) Mutp53 (R175 H) promoted the migration ability of H1299 cells, which was greatly compromised by NSC23766 and Wortmannin. Left panel: representative images. Right panel: quantification of average number of migrated cells/field. H1299 cells with or without ectopic mutp53 (R175 H) were treated with NSC23766 (2 µM) or Wortmannin (1 µM) for 24 h. For B-D, data were present as mean ± SD. n ≥ 3. *** p < 0.001, Student’s t-test.

Furthermore, mutp53 (R175 H) significantly promoted migration of H1299 cells as determined by trans-well assays (Figure 5(d)). Treatment of NSC23766 or Wortmannin significantly inhibited the migration of both H1299-Con and H1299-R175 H cells, and notably, largely abolished mutp53 GOF in promoting migration of cells (Figure 5(d)).

We further investigated whether AKT functions as a down-stream effector of the Rac1 signaling to promote mutp53 GOF. In H1299-Con and H1299-R175 cells, blocking Rac1 function by knockdown of Rac1 inhibited the proliferation, colony formation, and anchorage-independent cell growth, and importantly, largely abolished mutp53 GOF in promoting proliferation and anchorage-independent cell growth (Figure 6(a–c)). Notably, Wortmannin treatment exhibited limited inhibitory effects on cell proliferation, colony formation and anchorage-independent growth in H1299-Con and H1299-R175 H cells with knockdown of endogenous Rac1 (Figure 6(a–c)). Similarly, knockdown of Rac1 inhibited migration of H1299-Con and H1299-R175 H cells, and importantly, largely abolished mutp53 GOF in promoting migration of cells (Figure 6(d)). While Wortmannin treatment greatly inhibited the migration of H1299-Con and H1299-R175 H cells and largely abolished mutp53 GOF in promoting migration, it exhibited a limited inhibitory effect on migration of both cell lines with knockdown of endogenous Rac1 (Figure 6(d)). These results indicate that AKT functions as an important down-stream effector of the Rac1 signaling to promote mutp53 GOF.

Figure 6. AKT activation is a downstream effector of the Rac1 signaling to promote mutp53 GOF in cell proliferation (a), colony formation (b), anchorage-independent cell growth (c) and migration (d). H1299 cells with or without ectopic expression of mutp53 (R175 H) were subjected to knockdown of endogenous Rac1 by shRNA vectors and/or treatment of Wortmannin. For B-D, data were present as mean ± SD. n ≥ 3. *** p < 0.001, Student’s t-test.

Together, these results strongly suggest that AKT activation by mutp53 via the Rac1 signaling contributes to mutp53 GOF in tumorigenesis, and targeting the Rac1/AKT axis suppresses mutp53 GOF in tumorigenesis.

Discussion

AKT, a serine/threonine kinase, is a major player in the PI3 K/AKT signaling that regulates many important biological functions, including cell proliferation, survival and metabolism, etc [20,38]. AKT is frequently activated in many human cancers, which contributes to tumor initiation and progression [20,38]. Activation of certain oncogenes, such as ErbB2 and EGFR, and inactivation of some tumor suppressor genes, such as PTEN that regulate PI3 K/AKT signaling, are important mechanisms underlying AKT activation in human cancers [20,39,40]. The phosphorylation of AKT at Ser473 and Thr308 leads to its activation and regulation of its downstream substrates, such as FOXOs, GSK3, p27, to exert its biological functions [41,42].

Previous studies on the cross-talk between AKT and p53 and its underlying mechanisms were mainly focused on WT p53 [43–45]. It was reported that the activation of AKT can phosphorylate MDM2, a key negative regulator of p53, at Ser166 and Ser186, which results in the translocation of MDM2 to the nucleus to promote ubiquitination and degradation of WT p53 [43–45]. At the same time, WT p53 negatively regulates AKT via its target gene PTEN [46,47]. The coexistence of p53 mutations and AKT activation in cancers has been reported previously [6,14,15]. Previous reports have also shown that some tumor-associated mutp53 proteins activate AKT in cancer cells [6,14,15]. Results from this study demonstrated that GOF mutp53 activates AKT in cancer cells, and analysis of a breast cancer cohort in TCGA database shows a significant correlation between p53 mutation and AKT activation, which is consistent with previous reports [6,14,15]. However, the mechanism whereby mutp53 activates AKT is not well-understood and cannot be explained by the cross-talk between AKT and WT p53.

Many tumor-associated mutp53 proteins exhibit GOFs, including promoting cell proliferation, survival, migration, invasion, metabolic reprogramming, and chemoresistance [5–7,48]. Mutp53 proteins often accumulate to high levels in tumors, which is crucial for its GOF activities [3,49–51]. Mutp53 can accumulate in both nuclei and cytoplasm and interact with different proteins to regulate their functions, which is an important mechanism by which mutp53 exerts its GOFs [4,8,9]. For example, in the nucleus, mutp53 interacts with a group of transcriptional factors and/or cofactors and regulate their transcription activities to promote tumor progression [4,9,52–55]. Our recent study showed that mutp53 binds to small GTPase Rac1 in the cytoplasm [11]. The mutp53-Rac1 interaction activates Rac1 through inhibiting de-SUMOylation of Rac1. SUMOylation of Rac1 is critical to maintain Rac1 activity in cells [27]. The inhibition of Rac1 de-SUMOylation by mutp53 leads to increased Rac1 activity in cells [11]. Rac1 is an upstream regulator of the PI3 K/AKT signaling and can activate AKT via its downstream effectors, PAKs [17]. These findings suggest that the activation of AKT through the Rac1 signaling could be an important mechanism whereby mutp53 activates AKT to exert its GOF.

Results from this study showed that mutp53 does not interact with AKT in cells as examined by co-IP assays (Figure 4(d)). Interestingly, our results show that blocking Rac1 by RNAi or inhibitors largely inhibited AKT activation by mutp53. We found that Rac1 SUMOylation promoted by mutp53 is crucial for mutp53-induced AKT activation; in H1299 cells with knockdown of endogenous Rac1 and ectopic expression of the Rac1 with a mutation at the SUMOylation site, mutp53 showed a minimal effect on AKT activation. We further found that Rac1 activation by mutp53 inhibits the interaction of DAB2IP, a negative regulator of AKT, with AKT, which contributes to AKT activation by mutp53. These results suggested that the activation of Rac1 by mutp53 mediates mutp53-induced AKT activation. Results from this study further demonstrated that the activation of AKT by mutp53 contributes to mutp53 GOF. Wortmannin, the inhibitor targeting AKT, largely abolished mutp53 GOF in promoting cell proliferation, colony formation, anchorage-independent growth and migration ability in H1299 cells (Figure 5). Combined treatment using inhibitors targeting Rac1 and AKT did not lead to an additive or synergistic effect on inhibiting mutp53 GOF (Figure 6). These results suggest that the Rac1/AKT axis contributes to mutp53 GOF.

Together, this study reveals that the activation of AKT through the Rac1 signaling is an important mechanism whereby mutp53 exerts its GOF. AKT is an important signaling hub with over 100 downstream target substrates [56,57]. Further studies to characterize the levels and functions of these substrates in tumors containing mutp53 will help further understand the mechanism of mutp53 GOF and the contribution of AKT activation to mutp53 GOF. It is worth noting that p53 mutations are scattered throughout the DNA binding domain, and it is still unclear whether p53 mutations at different residues have a similar effect on AKT activation. Further studies on mutp53 at different residues will help address this question. Results from this study also suggest that Rac1 and AKT are potential therapeutic targets in tumors containing GOF mutp53.

Materials and methods

Cell cultures, vectors and reagents

Human cancer cell lines H1299, SK-BR-3, SW480, Saos2 and MDA-MB-468 were purchased from ATCC. HCT116 p53^−/- and HCT116 p53^R248W/- were gifts from Dr. Bert Vogelstein at Johns Hopkins University. H1299-Con and H1299-R175 H cells were established as previously described [11]. Vectors (mutp53, Rac1, DN-Rac1, WT-Rac1-R and Rac1-DSR) and shRNAs (p53 and Rac1) were constructed previously [11]. Flag-tagged AKT1 plasmid was a gift from Dr. William Sellers (Addgene plasmid #9021). pLPC-mDAB2IP plasmid was a gift from Dr. Collavin’s lab (National Laboratory CIB, Italy). siRNA oligos targeting p53 were purchased from Dharmacon. The Rac1 inhibitor (NSC23766) and the AKT inhibitor (Wortmannin) were purchased from Sigma.

Western-blot assays

Protein lysates from cells and xenograft tumors were used for western-blot assays. The following antibodies were used in this study: anti-p53 (Santa Cruz Biotechnology), anti-Rac1 (05–389, Millipore), anti HA (Roche), anti-PAK1 (Cell Signaling Technology), anti-PAK2 (Cell signaling technology), anti-p-PAK1/2 (Cell Signaling Technology), anti-AKT (Santa Cruz Biotechnology), anti-p-AKT (s473) (Cell Signaling Technology) and anti-DAB2IP (Abcam) antibodies.

Rac1 activity analysis

Rac1 activity was analyzed by using a Rac1 activation assay kit (Millipore) to perform the GST-p21-binding domain of PAK1 pull-down assays which specifically pull down Rac1-GTP in cells as previously described [24]. The p21-binding domain of the Rac1 effector protein PAK1 specifically binds to Rac1-GTP [58]. The levels of precipitated Rac1-GTP were measured by western-blot assays.

Co-IP assays

Co-IP assays were performed as previously described [48]. Anti-Flag and anti-DAB2IP antibodies were used to pull down Flag-AKT and DAB2IP, respectively. The protein complex pulled-down by above-mentioned antibodies was used for western-blot assays.

Cell proliferation assays

Cells (1 × 10⁴) were seeded in 24-well plates, and treated with DMSO, Rac1 inhibitor NSC23766 (1 µM) or AKT inhibitor Wortmannin (5 µM). Cell proliferation was monitored by IncuCyte ZOOM live-cell analysis system (Essen Bioscience).

Colony formation

Exponentially growing cells (1 × 10³) were seeded in 6-well plates, and treated with DMSO, Rac1 inhibitor NSC23766 (100 nM) or AKT inhibitor Wortmannin (500 nM) for 14 days. Colonies were then stained by crystal violet. Colony area was analyzed by using Image J (vendor) software.

Anchorage-independent cell growth

Anchorage-independent cell growth assays were performed as previously described [11]. Briefly, 6-well plates were coated with 0.6% agarose, and cells (1 × 10³) were cultured in it with medium containing 0.3% agarose. Cells were treated with DMSO, Rac1 inhibitor NSC23766 (100 nM) or AKT inhibitor Wortmannin (500 nM) for 14 days. The colonies were stained using crystal violet, and quantified by Image J (vendor) software.

Migration assays

Migration assays were performed by using the trans-well system (BD Biosciences) as previously described [11]. Briefly, cells were seeded on the top chamber with FBS-free medium and the bottom chamber was filled with medium containing 10% FBS. Rac1 inhibitor NSC23766 (2 µM) or AKT inhibitor Wortmannin (1 µM) were added into the medium in both chambers. Cells on the lower surface of the top chambers were fixed by methanol, stained by violet crystal for counting after culturing for 24 h.

Xenograft tumor formation

Cells were inoculated subcutaneously (s.c.) into BALB/c athymic male nude mice (Taconic) to form xenograft tumors as previously described [59]. Tumors were harvested when the tumor volume reached ~200 mm³ for western-blot assays. The mouse experiments were performed with the approval of the Institutional Animal Care and Use Committee of University.

Statistical analysis

Quantitative data were presented as mean ± SD. The Student’s t-test was used to analyze the difference between two groups. Values of p < 0.05 were considered to be significant.

Author Contributions

X.Y. carried out the experiments, analyzed data and wrote manuscript; F.W., Y.L., J.L. C.Z., M.B., K.M. carried out experiments; Z.F. and J.G. analyzed data and wrote manuscript; W.H. designed experiments, analyzed data and wrote manuscript.

Acknowledgments

We thank Dr. Bert Vogelstein for HCT116 p53^−/- and HCT116 p53^R248W/- cell lines, Dr. William Sellers for Flag-tagged AKT1 plasmid; and Dr. Collavin for pLPC-mDAB2IP plasmid.

Disclosure statement

Authors declare no competing interests.

Supplementary materials

Supplemental data for this article can be accessed here.