Phosphatidylinositol 3,4,5-trisphosphate-dependent Rac exchanger (P-Rex) proteins are RacGEFs that are synergistically activated by phosphatidylinositol 3,4,5-trisphosphate and Gβγ subunits of G-protein-coupled receptors. P-Rex1 and P-Rex2 share similar amino acid sequence homology, domain structure, and catalytic function. Recent evidence suggests that both P-Rex proteins may play oncogenic roles in human cancers. P-Rex1 and P-Rex2 are altered predominantly via overexpression and mutation, respectively, in various cancer types, including breast cancer, prostate cancer, and melanoma. This review compares the similarities and differences between P-Rex1 and P-Rex2 functions in human cancers in terms of cellular effects and signalling mechanisms. Emerging clinical data predict that changes in expression or mutation of P-Rex1 and P-Rex2 may lead to changes in tumour outcome, particularly in breast cancer and melanoma.

Introduction

Phosphatidylinositol 3,4,5-trisphosphate-dependent Rac exchanger (P-Rex) proteins belong to a family of Rac guanine nucleotide exchange factors (RacGEFs) [13], which includes three members: P-Rex1, P-Rex2 (also known as P-Rex2a), and the C-terminal-truncated splice isoform of P-Rex2 known as P-Rex2b [4]. P-Rex proteins are synergistically activated by phosphatidylinositol 3,4,5-trisphosphate (PI(3,4,5)P3) generated by the class I phosphoinositide 3-kinase (PI3K) and Gβγ subunits from activated G-protein-coupled receptors (GPCRs) [1,5]. RacGEFs promote Rac activation by facilitating the release of guanosine diphosphate (GDP), thereby allowing guanosine-5′-triphosphate (GTP) to bind. Active Rac regulates many essential cellular responses including actin cytoskeletal rearrangement, cell migration, adhesion, and the production of reactive oxygen species (ROS) [6,7]. Recent evidence links both Rac and RacGEFs, including P-Rex1 and P-Rex2, to increased cell migration and proliferation in various human cancers [823].

P-Rex1 and P-Rex2 have different tissue distributions and physiological functions. In humans, P-Rex1 mRNA is predominantly expressed in the brain and peripheral blood leukocytes [1], whereas P-Rex2 mRNA is highly expressed in the brain, heart, skeletal muscle, placenta, and lymph node [2]. At the protein level in mice, P-Rex1 is expressed in the brain and neutrophils [2426] and P-Rex2 is expressed in the brain (particularly the cerebellum), lung, and liver [27]. Limited information regarding P-Rex1 and P-Rex2 protein distribution in other human tissues is available, aside from a report of P-Rex2 protein expression in human adipose tissue [28]. P-Rex1 regulates leukocyte functions, including the production of ROS [1], platelet aggregation [29], neuronal migration [25], and neurite differentiation [30]. P-Rex2 is an important regulator of Purkinje cell morphogenesis and motor coordination [27]. There are only limited data available regarding the expression and functions of P-Rex2b, other than evidence of its expression in endothelial cells and the heart, as well as a role in the regulation of endothelial cell migration [2,31]. Prex1−/−, Prex2−/− and Prex1−/− ;Prex2−/− mice have been generated and characterised [24,2628]. All strains are fertile but display several specific phenotypes, which are summarised in Table 1. Interestingly, emerging evidence has revealed that P-Rex1 and P-Rex2 are overexpressed or mutated in many human cancers [8,9,11,17,18,32,33], which is the focus of this review.

Table 1
Physiological effects of Prex knockout in mice
Knockout modelEffectsReferences
Prex1−/− Mild neutrophilia [24
Slight body size reduction [24
Slight liver size reduction [24
Skin depigmentation [24
Reduced ROS production in neutrophil [24,26
Defect in neutrophil chemotaxis [24,26
Defective neutrophil recruitment to inflammation sites [24
Mild haemostasis defect due to defective platelet aggregation and granule secretion [29
Impaired macrophage chemotaxis and reduced superoxide production [60
Prex2−/− Reduction in body weights of ageing females [27
Impaired Purkinje cell morphology affecting 3- to 20-month-old mice [27
Mild motor coordination defect that worsens with age particularly in females [27
Glucose intolerance and insulin resistance [28
Prex1−/−;Prex2−/− Enhanced motor coordination defect compared with Prex2−/− mice from a younger age, including impaired posture and gait [27
Worse Purkinje cell morphology defects compared with Prex2−/− [27
Knockout modelEffectsReferences
Prex1−/− Mild neutrophilia [24
Slight body size reduction [24
Slight liver size reduction [24
Skin depigmentation [24
Reduced ROS production in neutrophil [24,26
Defect in neutrophil chemotaxis [24,26
Defective neutrophil recruitment to inflammation sites [24
Mild haemostasis defect due to defective platelet aggregation and granule secretion [29
Impaired macrophage chemotaxis and reduced superoxide production [60
Prex2−/− Reduction in body weights of ageing females [27
Impaired Purkinje cell morphology affecting 3- to 20-month-old mice [27
Mild motor coordination defect that worsens with age particularly in females [27
Glucose intolerance and insulin resistance [28
Prex1−/−;Prex2−/− Enhanced motor coordination defect compared with Prex2−/− mice from a younger age, including impaired posture and gait [27
Worse Purkinje cell morphology defects compared with Prex2−/− [27

P-Rex1 and P-Rex2 structure and regulation

P-Rex1 and P-Rex2 proteins are 185 and 183 kDa, respectively, and exhibit 59% amino acid sequence homology with highly conserved domain structures [1,2,9]. As depicted in Figure 1, both P-Rex proteins contain an N-terminal Dbl-homologous-Pleckstrin-homology (DH-PH) tandem domain, followed by Dishevelled, Egl-10, and Pleckstrin (DEP) and post-synaptic density protein, Drosophila disc large tumour suppressor, and zonula occludens-1 protein (PDZ) tandem domains, and a C-terminal inositol polyphosphate 4-phosphatase (IP4P) domain.

Binding partners that regulate P-Rex1 and P-Rex2 protein activity.

Figure 1.
Binding partners that regulate P-Rex1 and P-Rex2 protein activity.

↑ indicates activation; ⏉ indicates inactivation.

Figure 1.
Binding partners that regulate P-Rex1 and P-Rex2 protein activity.

↑ indicates activation; ⏉ indicates inactivation.

The P-Rex DH domain encodes the RacGEF catalytic activity and mediates P-Rex activation by Gβγ, whereas the PH domain is required for PI(3,4,5)P3 binding [5]. The recently reported crystal structure of the P-Rex1 DH-PH tandem domain bound to the GTPase Rac1 reveals that contact between P-Rex1 and Rac1 occurs exclusively through the DH domain with no molecular contributions from the PH domain (Figure 2) [34]. Multiple point mutations throughout the GTPase-interacting surface have been shown to abrogate P-Rex1-dependent Rac1 signalling in breast cancer cells [34], indicating the surface of the DH domain is the active site in nucleotide exchange. A second crystal structure of the P-Rex1 DH-PH domains confirmed the PI(3,4,5)P3 binding site resides within the PH domain (Figure 2), and point mutations in this region abrogate lipid binding [35]. P-Rex1 also interacts with Norbin, a GPCR-adaptor protein, via its PH domain, whereby Norbin recruits inactive P-Rex1 to the plasma membrane and facilitates the activation of its RacGEF activity [36]. Additional to PI(3,4,5)P3 and Gβγ binding, the DH and PH domains of P-Rex2 also interact with the tumour suppressor phosphatase and tensin homologue (PTEN), resulting in mutual inhibition [19,28,37]. PTEN is a dual specificity protein and phosphoinositide phosphatase that hydrolyses PI(3,4,5)P3 and thereby inhibits PI3K signalling.BST-2016-0269CF3 

Structure of P-Rex1.

Figure 2.
Structure of P-Rex1.

(A) crystal structure of the P-Rex1 DH-PH tandem domains bound to its cognate GTPase, Rac1 (PDB ID 4YON; [34]) and (B) crystal structure of the P-Rex1 PH domain bound to the PI(3,4,5)P3 analogue, inositol-1,3,4,5-tetrakisphosphate (PDB ID 5D3X; [35]).

Figure 2.
Structure of P-Rex1.

(A) crystal structure of the P-Rex1 DH-PH tandem domains bound to its cognate GTPase, Rac1 (PDB ID 4YON; [34]) and (B) crystal structure of the P-Rex1 PH domain bound to the PI(3,4,5)P3 analogue, inositol-1,3,4,5-tetrakisphosphate (PDB ID 5D3X; [35]).

P-Rex proteins signalling in cancer.

Figure 3.
P-Rex proteins signalling in cancer.

Ligand binding to RTK and/or GPCR leads to the activation of PI3K class I or PI3K class II, respectively [61]. Active PI3K phosphorylates PI(4,5)P2 at the D3 position to form PI(3,4,5)P3 at the inner leaflet of the plasma membrane [62]. PI(3,4,5)P3 can be dephosphorylated at the D5 position to form PI(3,4)P2 by 5-phosphatases [63]. Both PI(3,4,5)P3 and PI(3,4)P2 are activators of Akt and are required for maximal Akt activation. Activated Akt promotes cell survival and proliferation via multiple downstream targets [61]. The PI3K/Akt signalling is regulated by PTEN, which dephosphorylates PI(3,4,5)P3 back to PI(4,5)P2, and the 4-phosphatase INPP4B, which hydrolyses PI(3,4)P2 to PI(3)P, thus suppressing Akt activation [39, 56]. PI(3,4,5)P3, in synergy with the βγ subunits of activated GPCR, activates P-Rex proteins [1, 5]. P-Rex1 is also positively regulated by norbin and PP1α and P-Rex2 by PP1α and PP2A [36, 41]. PKA inhibits P-Rex1 [45]. Active P-Rex1 activates Rac, a promoter of cell cytoskeletal rearrangement, cell migration, and invasion [1, 7, 8]. Rac, in turns, activates PAK and the RAF/MEF/ERK signalling pathway leading to cell survival and proliferation [7, 10, 53]. P-Rex1 can, in some context, indirectly activate the PI3K/Akt axis [11]. P-Rex1 interaction with EHBP1 is proposed to facilitate the inhibition of cell invasion in a prostate cancer cell line (DU145) [47]. P-Rex1 also colocalises with PTEN and hence may potentially interact with PTEN, although this interaction has not been demonstrated [47]. P-Rex2 also activates Rac [41], which conceivably leads to the activation of PAK, RAF, MEK, and ERK as in the case for P-Rex1, although P-Rex2 signalling pathway downstream from Rac has not been investigated. P-Rex2 interacts with PTEN leading to mutual inhibition [19]. Via inhibition of PTEN, P-Rex2 activity augments PI(3,4,5)P3 levels, hence increasing Akt activation [20, 28, 37]. Both P-Rex1 and P-Rex2 activate mTOR, which facilitates Akt activation [42].

Figure 3.
P-Rex proteins signalling in cancer.

Ligand binding to RTK and/or GPCR leads to the activation of PI3K class I or PI3K class II, respectively [61]. Active PI3K phosphorylates PI(4,5)P2 at the D3 position to form PI(3,4,5)P3 at the inner leaflet of the plasma membrane [62]. PI(3,4,5)P3 can be dephosphorylated at the D5 position to form PI(3,4)P2 by 5-phosphatases [63]. Both PI(3,4,5)P3 and PI(3,4)P2 are activators of Akt and are required for maximal Akt activation. Activated Akt promotes cell survival and proliferation via multiple downstream targets [61]. The PI3K/Akt signalling is regulated by PTEN, which dephosphorylates PI(3,4,5)P3 back to PI(4,5)P2, and the 4-phosphatase INPP4B, which hydrolyses PI(3,4)P2 to PI(3)P, thus suppressing Akt activation [39, 56]. PI(3,4,5)P3, in synergy with the βγ subunits of activated GPCR, activates P-Rex proteins [1, 5]. P-Rex1 is also positively regulated by norbin and PP1α and P-Rex2 by PP1α and PP2A [36, 41]. PKA inhibits P-Rex1 [45]. Active P-Rex1 activates Rac, a promoter of cell cytoskeletal rearrangement, cell migration, and invasion [1, 7, 8]. Rac, in turns, activates PAK and the RAF/MEF/ERK signalling pathway leading to cell survival and proliferation [7, 10, 53]. P-Rex1 can, in some context, indirectly activate the PI3K/Akt axis [11]. P-Rex1 interaction with EHBP1 is proposed to facilitate the inhibition of cell invasion in a prostate cancer cell line (DU145) [47]. P-Rex1 also colocalises with PTEN and hence may potentially interact with PTEN, although this interaction has not been demonstrated [47]. P-Rex2 also activates Rac [41], which conceivably leads to the activation of PAK, RAF, MEK, and ERK as in the case for P-Rex1, although P-Rex2 signalling pathway downstream from Rac has not been investigated. P-Rex2 interacts with PTEN leading to mutual inhibition [19]. Via inhibition of PTEN, P-Rex2 activity augments PI(3,4,5)P3 levels, hence increasing Akt activation [20, 28, 37]. Both P-Rex1 and P-Rex2 activate mTOR, which facilitates Akt activation [42].

P-Rex proteins exhibit low basal RacGEF activity; however, truncation of domains’ C-terminal to the DH domain increases exchange kinetics [5,38]. This suggests that the C-terminus obstructs the DH domain or alters its conformation, such that it is largely unavailable to bind and activate GTPases. Molecular modelling proposes possible binding sites for Gβγ and PI(3,4,5)P3 on the opposite face of the DH-PH domains to that of the Rac1-binding site [34]. It is likely that PI(3,4,5)P3 and Gβγ synergistically activate P-Rex proteins by relieving autoinhibitory intramolecular interactions.

P-Rex1 and P-Rex2 possess a C-terminal IP4P domain, which contains the CX5R dual specificity phosphatase catalytic motif and shares ∼30% amino acid sequence homology with the inositol polyphosphate 4-phosphatase Types I and II (INPP4A and INPP4B) [1,2,39]. Perplexingly, P-Rex proteins do not exhibit detectable inositol polyphosphate 4-phosphatase activity [1,5]. The IP4P domain plays a regulatory role in both P-Rex proteins via binding to protein phosphatase 1α (PP1α), which activates P-Rex RacGEF activity [40,41]. P-Rex2 interacts with protein phosphatase 2A (PP2A) and PTEN via its IP4P domain, resulting in P-Rex2 activation and mutual inhibition, respectively [19,28,41].

In addition to their roles in intramolecular inhibition, the DEP domains of P-Rex1 and P-Rex2 bind to mechanistic target of rapamycin (mTOR), implicating the mTOR pathway in Rac-mediated cell migration [42]. The tandem PDZ domains of P-Rex1 interact with sphingosine-1-phosphate 1 (S1P1), a GPCR, which is involved in endothelial cell migration [43]; however, it has not been reported whether full-length P-Rex1 protein can engage in this interaction. P-Rex1 PDZ domains are also implicated in the interaction between P-Rex1 and ephrin-B1, a transmembrane ligand for the Eph receptor tyrosine kinases involved in neuronal development, as P-Rex1 mutant lacking PDZ domains could not be coimmunoprecipitated with ephrin-B1 in mouse embryonic cortex lysates [44]. Of note, the P-Rex1 PDZ domains also interact with protein kinase A (PKA) [42,45]. PKA potently inhibits P-Rex1 RacGEF activity by phosphorylating the P-Rex1 DEP1 domain and promoting an autoinhibitory conformation, in which the DEP domain and the C-terminal tail obstruct the DH domain catalytic activity [45]. Binding partners that regulate P-Rex protein activity are shown in Figure 1. Platelet-derived growth factor receptor beta (PDGFβ) has also been reported as a P-Rex1-interacting partner in P-Rex1 overexpressing immortalised human fibroblasts [46]; however, the specific P-Rex1 domain(s) involved in this interaction remain(s) unclear. PTEN and EH domain-binding protein 1 (EHBP1) have also been reported as putative P-Rex1-binding partners via colocalisation studies [47], although these interactions have not been confirmed via other approaches.

P-Rex2 is commonly mutated in various human cancers and these mutations are likely to overcome a similar mechanism of autoinhibition. Identified mutations include truncation mutations (Table 2), which result in N-terminal fragments of the protein that are likely to be relieved from autoinhibition or PTEN inhibition. This will be discussed in further detail below. However, structures of full-length P-Rex proteins in the autoinhibited conformation and in complex with PI(3,4,5)P3, Gβγ, or PTEN will be required to effectively map, or predict, the effects of somatic P-Rex mutations. The full suite of functional effects of P-Rex2 mutations have yet to be shown and the known reported effects are summarised in Table 2.

Table 2
Published P-Rex2 non-synonymous mutations in various cancers and their known cellular effects
CancerMutationDomainFrequencySample sizeSample tissueEffectsReference
Breast cancer R155Q DH 7.7% 560 Human breast tumours – [51
Melanoma K278* PH – 25 Human metastatic melanoma Increased tumour incidence, accelerated tumorigenesis, and decreased tumour-free survival in mice xenografts
Increased GEF activity, activation of PI3K/AKT pathway, and cellular proliferation 
[17,18
Melanoma E730* IP4P – 25 Human metastatic melanoma – [17
Melanoma E824* IP4P – 25 Human metastatic melanoma Increased tumour incidence, accelerated tumorigenesis, and decreased tumour-free survival in mice xenografts
Increased GEF activity, activation of PI3K/AKT pathway, and cellular proliferation 
[17,18
Melanoma Q1430* IP4P – 25 Human metastatic melanoma Increased tumour incidence, accelerated tumorigenesis, and decreased tumour-free survival in mice xenografts [17
Melanoma G106E DH – 25 Human metastatic melanoma – [17
Melanoma E213K DH-PH – 25 Human metastatic melanoma – [17
Melanoma E421K DEP1 – 25 Human metastatic melanoma – [17
Melanoma I534M DEP2 – 25 Human metastatic melanoma – [17
Melanoma P948S IP4P – 25 Human metastatic melanoma No significant difference in tumour growth and survival in mice xenografts
Resistant to PTEN C2-tail inhibition 
[17,19
Melanoma S1052F IP4P – 25 Human metastatic melanoma – [17
Melanoma G1196E IP4P – 25 Human metastatic melanoma – [17
Melanoma V1441F IP4P – 25 Human metastatic melanoma – [17
Melanoma G1581R IP4P – 25 Human metastatic melanoma – [17
Melanoma G844D IP4P – 107 Melanoma short-term culture Increased tumour incidence, accelerated tumorigenesis, and decreased tumour-free survival in mice xenografts
Escape PTEN inhibition to retain GEF activity 
[17,19
Melanoma E12* DH – 107 Melanoma short-term culture – [17
Melanoma S413N DEP1 – 107 Melanoma short-term culture – [17
Melanoma A1549V IP4P – 107 Melanoma short-term culture – [17
Melanoma S771 Frameshift IP4P – 107 Melanoma short-term culture – [17
Melanoma E1356K IP4P – 107 Melanoma short-term culture – [17
Melanoma R1557C IP4P – 107 Melanoma short-term culture – [17
Melanoma P775S IP4P – 107 Melanoma short-term culture – [17
Melanoma E1389K IP4P – 107 Melanoma short-term culture – [17
Melanoma A1376V IP4P – 107 Melanoma short-term culture – [17
Melanoma P665L PDZ2 – 107 Melanoma short-term culture – [17
Melanoma R463C DEP1 – 107 Human metastatic melanoma – [17
Melanoma P776S IP4P – 107 Human metastatic melanoma – [17
Melanoma E354D PH – 107 Human metastatic melanoma – [17
Melanoma T994A IP4P – 107 Human metastatic melanoma – [17
Lung cancer R280T PH – 97 Japanese lung cancer samples – [58
Lung cancer T302A PH – 97 Japanese lung cancer samples – [58
Lung cancer A315D PH – 97 Japanese lung cancer samples – [58
Lung cancer H451N DEP1 – 97 Japanese lung cancer samples – [58
Lung cancer E557Q DEP2 – 97 Japanese lung cancer samples – [58
Lung cancer R562L DEP2 – 97 Japanese lung cancer samples – [58
Lung cancer K1177 IP4P – 97 Japanese lung cancer samples – [58
Pancreatic cancer V432M DEP1 – – – Resistant to PTEN C2-tail inhibition [19
Squamous cell carcinoma D269H PH 38% 29 Human metastatic cSCC samples – [59
CancerMutationDomainFrequencySample sizeSample tissueEffectsReference
Breast cancer R155Q DH 7.7% 560 Human breast tumours – [51
Melanoma K278* PH – 25 Human metastatic melanoma Increased tumour incidence, accelerated tumorigenesis, and decreased tumour-free survival in mice xenografts
Increased GEF activity, activation of PI3K/AKT pathway, and cellular proliferation 
[17,18
Melanoma E730* IP4P – 25 Human metastatic melanoma – [17
Melanoma E824* IP4P – 25 Human metastatic melanoma Increased tumour incidence, accelerated tumorigenesis, and decreased tumour-free survival in mice xenografts
Increased GEF activity, activation of PI3K/AKT pathway, and cellular proliferation 
[17,18
Melanoma Q1430* IP4P – 25 Human metastatic melanoma Increased tumour incidence, accelerated tumorigenesis, and decreased tumour-free survival in mice xenografts [17
Melanoma G106E DH – 25 Human metastatic melanoma – [17
Melanoma E213K DH-PH – 25 Human metastatic melanoma – [17
Melanoma E421K DEP1 – 25 Human metastatic melanoma – [17
Melanoma I534M DEP2 – 25 Human metastatic melanoma – [17
Melanoma P948S IP4P – 25 Human metastatic melanoma No significant difference in tumour growth and survival in mice xenografts
Resistant to PTEN C2-tail inhibition 
[17,19
Melanoma S1052F IP4P – 25 Human metastatic melanoma – [17
Melanoma G1196E IP4P – 25 Human metastatic melanoma – [17
Melanoma V1441F IP4P – 25 Human metastatic melanoma – [17
Melanoma G1581R IP4P – 25 Human metastatic melanoma – [17
Melanoma G844D IP4P – 107 Melanoma short-term culture Increased tumour incidence, accelerated tumorigenesis, and decreased tumour-free survival in mice xenografts
Escape PTEN inhibition to retain GEF activity 
[17,19
Melanoma E12* DH – 107 Melanoma short-term culture – [17
Melanoma S413N DEP1 – 107 Melanoma short-term culture – [17
Melanoma A1549V IP4P – 107 Melanoma short-term culture – [17
Melanoma S771 Frameshift IP4P – 107 Melanoma short-term culture – [17
Melanoma E1356K IP4P – 107 Melanoma short-term culture – [17
Melanoma R1557C IP4P – 107 Melanoma short-term culture – [17
Melanoma P775S IP4P – 107 Melanoma short-term culture – [17
Melanoma E1389K IP4P – 107 Melanoma short-term culture – [17
Melanoma A1376V IP4P – 107 Melanoma short-term culture – [17
Melanoma P665L PDZ2 – 107 Melanoma short-term culture – [17
Melanoma R463C DEP1 – 107 Human metastatic melanoma – [17
Melanoma P776S IP4P – 107 Human metastatic melanoma – [17
Melanoma E354D PH – 107 Human metastatic melanoma – [17
Melanoma T994A IP4P – 107 Human metastatic melanoma – [17
Lung cancer R280T PH – 97 Japanese lung cancer samples – [58
Lung cancer T302A PH – 97 Japanese lung cancer samples – [58
Lung cancer A315D PH – 97 Japanese lung cancer samples – [58
Lung cancer H451N DEP1 – 97 Japanese lung cancer samples – [58
Lung cancer E557Q DEP2 – 97 Japanese lung cancer samples – [58
Lung cancer R562L DEP2 – 97 Japanese lung cancer samples – [58
Lung cancer K1177 IP4P – 97 Japanese lung cancer samples – [58
Pancreatic cancer V432M DEP1 – – – Resistant to PTEN C2-tail inhibition [19
Squamous cell carcinoma D269H PH 38% 29 Human metastatic cSCC samples – [59

More unpublished and uncharacterised P-Rex2 mutations can be found on Oncomine and CBioPortal. indicates frequency/sample size/sample tissue/effects not reported.

P-Rex1 and P-Rex2 alterations in human cancers

P-Rex1 is overexpressed in several human cancers via gene amplification [11] or changes in epigenetic regulation [32,33]. P-Rex1 mRNA is up-regulated in breast, prostate, thyroid, lymphoid, ovarian, adrenal, and kidney cancers [8,32], as well as melanoma relative to normal tissues [8,13,32]. At the protein level, P-Rex1 overexpression has only been confirmed in breast cancer, prostate cancer, and melanoma [8,9,13,15].

P-Rex1 shows low or no expression in the normal breast lobules or ducts [8,9] but in breast cancer, P-Rex1 overexpression correlates strongly with oestrogen receptor (ER) expression and the luminal subtype [8,11,32]. The Cancer Genome Atlas (TCGA) database analysis shows that P-Rex1 gene methylation, which suppresses transcription, correlates with better patient survival after 5-year post-diagnosis [32]. However, the cohort size involved in this study was not specified and the analysis was not stratified for breast cancer subtype.

High P-Rex1 protein expression has been linked to decreased disease-free breast cancer patient survival, although the patient numbers involved in this study were very small (n = 32 low P-Rex1-expressing and n = 4 high P-Rex1-expressing tumours) [9]. Paradoxically, P-Rex1 mRNA expression was associated with an improved breast cancer patient outcome in a METABRIC data set (n = 1981) [48]. This model predicts that a patient with breast cancer expressing low levels of P-Rex1 will be 61.7% more likely to have a shorter survival time compared with a patient with a higher P-Rex1-expressing tumour [48,49]. The reason for the differences between these two findings is not clear; however, it is possible that P-Rex1 mRNA and P-Rex1 protein expression are not equivalent. More studies evaluating this in the context of breast cancer subtypes are required. Another study showed that metastatic breast tumours (n = 46) and lymph node metastases (n = 24) exhibited increased P-Rex1-positive immunohistochemical staining compared with non-metastatic tumours (n = 52) [8]. As P-Rex1 regulates cell migration through its RacGEF activity [8,9], it is interesting to speculate that P-Rex1 may facilitate metastatic dissemination of cancer cells [8]. It is also possible that further up-regulation of P-Rex1 proteins occurs after metastasis. There have been no reported studies to date, such as using genetically modified murine models that demonstrate the role P-Rex proteins play in breast cancer progression or metastasis.

Only limited evidence is available regarding P-Rex1 expression in human prostate cancer. One study identified that in eight primary human prostate cancer samples and matched lymph node metastases, P-Rex1 protein was only slightly and heterogeneously up-regulated in localised prostate tumours compared with adjacent normal tissue, but was more uniformly and highly increased in metastatic samples [15]. Supporting this, P-Rex1 protein levels were significantly increased in metastatic compared with non-metastatic human prostate cancer cell lines (PC3 versus 22Rv1 and LNCaP cells) [15,33]. Similar to breast cancer, there are no reports detailing the role P-Rex1 plays in prostate cancer initiation or progression to date. In addition, an association between P-Rex1 expression levels and prostate cancer patient survival, disease progression, or stage has not been reported.

P-Rex1 protein up-regulation has been shown in benign melanocyte nevi [14], primary melanoma, and lymph node metastases, and the relative expression of P-Rex1 increases with disease progression [13]. The majority of human cell lines derived from NRAS- or BRAF-mutated melanomas exhibit up-regulation of P-Rex1 protein compared with normal melanocytes or NRAS-/BRAF-wild-type cell lines [13,14], suggesting subset-specific involvement of P-Rex1 in melanoma. Interestingly, the in vivo role P-Rex1 plays in melanoma metastasis has been examined experimentally by crossing Prex1−/− mice with an NrasQ61K-driven mouse model of melanoma (Tyr::NrasQ61K/° transgenic mice) [13]. This study revealed that P-Rex1 is essential for melanocyte migration and that the absence of P-Rex1 protects against melanoma metastasis in this experimental model [13].

P-Rex proteins, in particular P-Rex2, are also mutated in human cancers. TCGA database shows a higher incidence of P-Rex2 mutations than P-Rex1 in both breast cancer and melanoma. No P-Rex2 mutations were identified in a small cohort of human breast tumours (n = 11) [50], but a recent genome-wide screen of 560 human breast cancer patient samples identified a common P-Rex2 mutation (P-Rex2 R155Q located in the DH domain) in 7.7% of samples [51]. In melanoma, P-Rex1 mutations are mostly located in the C-terminal domains. In contrast, P-Rex2 mutations are distributed throughout the protein [14]. Sequence analysis has identified non-synonymous P-Rex2 mutations in 14% of 107 human melanoma samples, and 13 non-synonymous mutations spanning the length of P-Rex2 were detected in 25 human metastatic melanomas [17]. P-Rex2 is the third most mutated protein in human primary melanomas after NRAS and BRAF [17]. Published P-Rex2 mutations and their associations in breast cancer and melanoma, as well as other cancers, are presented in Table 2.

To date, the functional effects of seven P-Rex2 mutations (K278, E730, E824, G844D, Q1430, P948S, G106E) have been examined, and most (aside from P-Rex2 P948S, G106E) appear to be activating mutations whereby an increase in RacGEF activity (K278, E824) [18], cell proliferation (E824) [18], and xenograft tumour incidence (K278, E824, G844D, Q1430) were observed [17,18]. Also, tumour-free survival is decreased in mice xenografted with immortalised human melanocytes carrying P-Rex2 K278, E824, G844D, or Q1430 truncating mutations [17]. Interestingly, 10 P-Rex2 mutations identified in pancreatic cancers display RacGEF inactivation [52], but there is a lack of information regarding the functional effects of these mutations in terms of tumour biology.

Several studies have been undertaken in human cancer cell lines to elucidate the functional effects of P-Rex1 and P-Rex2 expression (summarised in Table 3). P-Rex1 and P-Rex2 both enhance cell migration and invasion in multiple cancer types. P-Rex1 depletion by shRNA or siRNA in oestrogen receptor-positive (ER+) breast cancer cell lines (MCF-7 and T-47D) and in a metastatic prostate cancer cell line (PC3) impaired cell migration and invasion [8,9,15]. P-Rex1 knockdown via siRNA in multiple melanoma cell lines (CHL1, Wm793, Mel224, WM2664, Wm852) resulted in reduced invasive phenotypes [13,14,46]. P-Rex1 overexpression in breast and prostate cancer cell lines with negligible endogenous P-Rex1 expression (MDA-MB-231 and 22Rv1, respectively) increased migratory and invasive phenotypes [15,53]. However, one study reported that siRNA depletion of P-Rex1 in the DU145 prostate cancer cell line resulted in increased invasiveness, correlating with increased levels of matrix metalloproteinase 9 (MMP9) and matrix metalloproteinase 2 (MMP2), which play a role in cell invasion via extracellular matrix degradation [47]. The opposing effects of P-Rex1 depletion observed in DU145 and PC3 prostate cancer cells are proposed to result from P-Rex1 interaction with PTEN, the latter phosphatase is expressed in DU145, but not in PC3 cells [47]. Colocalisation between P-Rex1 and PTEN in response to insulin stimulation in DU145 cells was reported [47]; however, a direct interaction between the two proteins has not been confirmed by coimmunoprecipitation or other protein interaction assays. Studies using various human cancer cell lines derived from breast, osteosarcoma, neuroblastoma, and hepatocellular carcinoma suggest that P-Rex2 promotes cancer cell migration and/or invasion [19,2123,37]. However, the role P-Rex2 plays in cancer metastasis in vivo has not been reported. Further investigation is required to fully elucidate whether P-Rex proteins regulate metastasis and also whether they affect the overall process or only at specific stages.

Table 3
Cellular effects of P-Rex1 and P-Rex2 knockdown and overexpression in human cancer cell lines
P-Rex1P-Rex2
Breast cancerProstate cancerMelanomaBreast cancerGliomaOsteosarcomaGastric cancerNeuroblastomaHCCMelanoma
ER+TNMetNon-metWTBRAFNRASNormalER+TN
Proliferation KD↓ OE↑    – –  KD↓        
Cell-cycle progression        KD↓   KD↓  KD↓    
Anchorage-independent growth KD↓ OE↑      OE↑      KD↓   
Serum-independent growth        OE↑         
Apoptosis KD↑ OE↓         KD↑  KD↑    
Xenograft tumour growth KD↓ OE↑ – –            OE↑ 
Migration KD↓ OE↑ KD↓, OE↑ OE↑        KD↓, OE↑  KD↓ KD↓  
Lamellipodia formation KD↓  KD↓ OE↑             
Invasion KD↓  KD? OE↑ KD↓ KD↓ KD↓ OE↑  KD↓, OE↑  KD↓, OE↑  KD↓ KD↓  
Xenograft tumour cell metastasis    OE↑             
Adhesion KD↓                
References [8,9,53[53[15,47[15[13[13,14[14,46[19,28,37[12[23[20[21[22[17,18    
P-Rex1P-Rex2
Breast cancerProstate cancerMelanomaBreast cancerGliomaOsteosarcomaGastric cancerNeuroblastomaHCCMelanoma
ER+TNMetNon-metWTBRAFNRASNormalER+TN
Proliferation KD↓ OE↑    – –  KD↓        
Cell-cycle progression        KD↓   KD↓  KD↓    
Anchorage-independent growth KD↓ OE↑      OE↑      KD↓   
Serum-independent growth        OE↑         
Apoptosis KD↑ OE↓         KD↑  KD↑    
Xenograft tumour growth KD↓ OE↑ – –            OE↑ 
Migration KD↓ OE↑ KD↓, OE↑ OE↑        KD↓, OE↑  KD↓ KD↓  
Lamellipodia formation KD↓  KD↓ OE↑             
Invasion KD↓  KD? OE↑ KD↓ KD↓ KD↓ OE↑  KD↓, OE↑  KD↓, OE↑  KD↓ KD↓  
Xenograft tumour cell metastasis    OE↑             
Adhesion KD↓                
References [8,9,53[53[15,47[15[13[13,14[14,46[19,28,37[12[23[20[21[22[17,18    

KD, knockdown study (via shRNA/siRNA); OE, overexpression study; ↑, increased effect; ↓, decreased effect; ?, effects are context dependent; ER+, oestrogen receptor-positive; TN, triple-negative; Non-met, non-metastatic; WT, wild-type; BRAF, BRAF-mutated; NRAS, NRAS-mutated; HCC, hepatocellular carcinoma.

The role P-Rex1 plays in cancer cell proliferation and xenograft tumour growth may vary depending on the cancer type. P-Rex1 shRNA knockdown in an ER+ breast cancer cell line (MCF-7) resulted in a marked reduction in xenograft tumour size [8,9,53]. Conversely, P-Rex1 overexpression in a triple-negative breast cancer cell line (MDA-MB-231) increased cell proliferation, anchorage-independent cell growth, and xenograft tumour growth [53]. However, P-Rex1 siRNA depletion in various melanoma cell lines and overexpression in prostate cancer cell lines (PC3-LN4 and CWR22Rv1) did not affect cell proliferation and xenograft tumour growth, respectively [1315]. In comparison, P-Rex2 mutations appear to play a more consistent role in promoting cancer cell proliferation, survival, anchorage-independent cell growth, and xenograft tumour growth in different cancer types, particularly in melanoma (Table 3). Mouse xenografts of immortalised human melanocytes overexpressing P-Rex2 mutants (listed prior) elicited increased tumour incidence, accelerated tumorigenesis, and decreased tumour-free survival [17,18]. Aside from its regulation of Rac1 in response to S1P1 that contributes to cell migration [31], P-Rex2b RacGEF activity has not been recorded in human cancers.

Oncogenic signalling

P-Rex1 signalling in cancer has been studied more extensively than P-Rex2 and mostly in the context of breast cancer. P-Rex1 is activated downstream from the ErbB family of receptor tyrosine kinases and CXCR4, the latter is a GPCR for the chemokine SDF-1, and both receptors have been implicated in breast cancer progression [8]. P-Rex1 depletion abrogated ErbB ligand-induced Rac activation in ER+ MCF-7 and T-47D breast cancer cell lines, which correlated with impaired cell migration, proliferation, and transformation [8,9]. Similarly, CXCR4 stimulation by its ligand (SDF-1) led to Rac activation in a P-Rex1-dependent manner and P-Rex1 shRNA knockdown impaired breast cancer cell migration in this context [8]. As proposed by the authors, these findings indicate that ErbB and CXCR4 signalling in breast cancer intersects with P-Rex1. Interestingly, CXCR4 is reported to be transactivated by ErbB receptors, which suggests that CXCR4 may also mediate ErbB activation of the P-Rex1/Rac pathway [8]. Prolonged treatment of breast cancer cell lines (MCF-7, BT-474, and MDA-MB-361) with heregulin (an ErbB3- and ErbB4-specific ligand) led to up-regulation of CXCR4 on the cell surface, which sensitises the cell to SDF-1-induced Rac1 activation and cell migration in a P-Rex1-dependent manner, supporting the model that CXCR4 is transactivated by ErbB receptors [54]. Figure 3 depicts proposed P-Rex1 and P-Rex2 signalling pathways in cancer.

P-Rex regulation of the extracellular signal-regulated kinase signalling pathway

Extracellular signal-regulated kinase 1/2 (ERK1/2), a member of the mitogen-activated protein kinase (MAPK) family, has been identified as an effector of P-Rex1 downstream from Rac. The MAPK signalling pathway is characterised by the transmission of signals through a cascade of kinases (MAPK kinase kinase, MAPK kinase, and MAPK), eventually leading to transcriptional alterations and various cellular effects such as cell survival, growth, and proliferation. P-Rex1/Rac signalling has been linked to the ERK pathway in multiple cancer types.

A study aiming to elucidate the molecular mechanisms underlying the heterogeneity of patient's response to PI3K inhibitors revealed that inhibition of PI3K impairs ERK, as well as AKT activation in breast cancer cell lines harbouring PIK3CA mutations and/or ErbB2 amplification (MCF-7, T-47D, MDA-MB-361, EFM-192A, BT-474, HCC1419, and AU565) [10]. PI3K regulation of ERK was dependent on Rac activity and PI3K inhibitors reduced pERK levels only in breast cancer cell lines and xenograft tumours in which P-Rex1 was expressed, suggesting that Rac-mediated activation of ERK is facilitated by P-Rex1 [10]. Consistent with this contention, P-Rex1 shRNA knockdown in T-47D cells suppressed ERK phosphorylation, as well as the activation of Rac1 and its downstream effectors protein 21-activated kinase (PAK), rapidly accelerated fibrosarcoma (RAF), and mitogen-activated protein kinase kinase (MEK), thereby linking PI3K and P-Rex1 to the ERK signalling pathway [10]. This finding was confirmed in subsequent studies which showed that P-Rex1 shRNA knockdown or overexpression in various breast cancer cell lines resulted in decreased or elevated ERK phosphorylation, respectively [11,53]. As ERK signalling has been widely described as pro-proliferative and pro-survival, it is postulated that P-Rex1 mediates its oncogenic functions, at least in part, through ERK activation. Furthermore, P-Rex1-mediated ERK activation and increase in breast cancer cell xenograft tumour growth are dependent on its RacGEF activity [53].

ERK is also activated downstream from P-Rex1/Rac1 signalling in the context of prostate cancer. P-Rex1/Rac1 is required for ERK activation downstream from the neuropilin (NRP) receptor in response to vascular endothelial growth factor (VEGF) stimulation in bevacizumab (anti-VEGF therapy)-resistant prostate cancer cell lines (PC3 and C4-2), which have stem cell-like properties [16]. This mechanism is proposed to contribute to prostate cancer resistance to VEGF/VEGF receptor-targeted therapies, which inhibit the interaction between VEGF and the VEGF receptor but not the NRP receptor, and therefore do not effectively target prostate cancer stem cells [16].

Recent genetic profiling to identify genes essential for the survival of cells harbouring mutated Ras revealed P-Rex1 as an essential upstream promoter of the ERK signalling pathway in Ras-driven human acute myeloid leukaemia (AML) cell lines [55]. Interestingly, the requirement of P-Rex1 for cell survival was specific to Ras-mutated AML cell lines, as there was no difference in P-Rex1-dependence between Ras-mutated and Ras-wild-type cell lines derived from other haematological cancer types [55]. The dependence of P-Rex1 reflects the requirement for active Rac signalling, as expression of constitutively active Rac1 or MEK1 relieved the P-Rex1-dependency of human SKM-1 AML cells [55]. The cancer type specificity of P-Rex1 is proposed to be due to the lack of other active RacGEFs in AML cells [55].

As P-Rex2 also activates Rac, it is interesting to speculate that ERK also functions downstream from P-Rex2, although this has not been reported. Interestingly, P-Rex1 functions downstream from ERK in melanoma and is up-regulated in a dose-dependent manner with increasing pERK levels [14]. As the RAF–ERK pathway is aberrantly activated in 80% of cases of melanoma due to NRAS and BRAF mutations, it has been proposed that P-Rex1 may be a driver of ERK-dependent melanoma progression [14].

P-Rex in AKT signalling pathway

An interaction between P-Rex1 and PTEN has not been shown; however, P-Rex1 may, in some contexts, regulate the PI3K/AKT axis. One study showed that P-Rex1 overexpression or siRNA knockdown in MCF-7 and T-47D breast cancer cell lines resulted in a corresponding Rac-dependent increase or decrease in the levels of pAKT and insulin-like growth factor 1R (IGF-1R), respectively [11]. As IGF-1R activates PI3K and subsequently P-Rex1, P-Rex1 hyperactivation may result in a positive feedback loop that leads to further PI3K activation [11]. This model predicts P-Rex1 amplification augments’ AKT activation, both directly and via PI3K, thus promoting tumour growth [11]. However, this may depend on context specificity, as several lines of evidence suggest that P-Rex1 knockdown/overexpression does not affect AKT phosphorylation in MCF-7, T-47D, and MDA-MB-231 cells [810,53]. P-Rex1 expression inversely correlates with PI3K activity, as determined by phosphorylation of its effectors (including pAKT, p4E-BP1, pmTOR, and pGSK3), in a large cohort of 712 breast tumours [11]. Based on this model, PI3K inhibition in tumours may lead to increased P-Rex1 expression, which sustains AKT activation despite low PI3K activity [11]. Hence, it is proposed that this may be a possible mechanism for breast cancer resistance to PI3K-targetted therapies [11].

PTEN is a tumour suppressor that negatively regulates the PI3K signalling pathway by dephosphorylating PI(3,4,5)P3 to form PI(4,5)P2, thereby preventing activation of downstream effectors, such as AKT, which promote cellular proliferation and survival [56]. P-Rex2 colocalises with PTEN at the plasma membrane, where it inhibits PTEN activity via its DH-PH domain and this interaction leads to increased AKT phosphorylation [37]. Abrogation of P-Rex2 activity in human cancer cell lines via shRNA/siRNA, or expression of GEF-dead mutants, or in Prex2−/− mouse embryonic fibroblasts resulted in reduced AKT phosphorylation and cellular proliferation [20,28,37]. In tumours carrying P-Rex2 mutations, elevated pAKT levels correlated with higher expression of IGF2, a growth factor that activates the PI3K pathway [57]. Indeed, higher levels of pAKT were observed in melanocytes upon exogenous IGF2 expression [18]. It should be noted that the ability of P-Rex2 to inhibit PTEN is independent of its GEF activity [37]. Interestingly, P-Rex2 itself is also susceptible to PTEN inhibition by direct binding between the PTEN PDZ domain and the P-Rex2 IP4P domain, which reduces its RacGEF activity [19]. A weaker interaction occurs between the PTEN catalytic and C2 domains and the P-Rex2 PH domain [19,28]. The PDZ-containing C2 tail of PTEN alone is sufficient for suppressing P-Rex2-driven invasion in breast cancer cells [19]. This model proposes that when relieved of inhibition from PTEN, P-Rex2 activation results increased PI3K activity and subsequently increased AKT activation [20,28,37]. It is possible that P-Rex2 and PTEN act as scaffolds to inhibit each other, but the mechanisms behind this mutual inhibition require further elucidation.

P-Rex2 truncation mutations in the IP4P domain result in increased RacGEF activity, PI3K/AKT pathway activation, and cellular proliferation in vitro, as well as xenograft tumour incidence in mice [17,18], suggesting the possible relief of P-Rex2 from PTEN inhibition. Moreover, melanoma cells expressing truncated P-Rex2 mutants (K278*, E824*, and Q1430*) demonstrated increased phosphorylation of AKT [18]. In PTEN-deficient BT549 and SUM149 breast cancer cells overexpressing P-Rex2 harbouring IP4P point mutations, exogenous PTEN failed to suppress cellular invasion [19], indicating that certain residues in this domain are critical in facilitating inhibition by PTEN. Furthermore, P-Rex2 mutations are associated with increased PTEN mRNA expression, and thus, it can be speculated that high PTEN protein expression may select for P-Rex2 mutants that can escape PTEN inhibition [19]. The PH domain is involved in P-Rex2 activation by interacting with PI(3,4,5)P3. Interestingly, truncation of the PH domain at residue K278 did not hinder P-Rex2 activation, but rather increased RacGEF activity and xenograft tumour growth in mice [17,18]. It is plausible that despite truncation at the PH domain, the PI(3,4,5)P3-interacting residues (K254 and R263) [19] are preserved in the K278 mutant, and thus, P-Rex2 can still be activated. To date, the K278 truncation is the only P-Rex2 PH domain mutation studied, but truncation at both the PH and the IP4P domain leads to increased xenograft tumour formation, which could be attributed to the abrogation of IP4P domain in both cases. More evidence is needed to support the importance of the P-Rex2 IP4P domain in facilitating P-Rex2 negative regulation by PTEN.

The increased GEF activity observed in P-Rex2 mutants lacking the IP4P domain may be accounted for by autoinhibition disruption. The P-Rex1 IP4P domain has been implicated in autoinhibitory functions, whereby the C-terminus undergoes intramolecular interactions with the DH-PH domain to inhibit RacGEF catalytic activity [5,34]. As the DH-PH domains of P-Rex1 and P-Rex2 share 71% sequence homology, the P-Rex2 DH-PH domain may exhibit a similar autoinhibition [18]. Without further clarification of the mechanisms behind both P-Rex2 autoinhibition and PTEN interaction, it is difficult to conclude whether up-regulation in GEF activity driven by IP4P mutations are due to abrogated PTEN interaction or disrupted autoinhibition.

Conclusion

P-Rex1 and P-Rex2 share similarities in amino acid sequence, domain structure, activation by PI(3,4,5)P3 and Gβγ, and both function as RacGEFs. Expressions of both P-Rex1 and P-Rex2 are altered in many human cancers: P-Rex1 predominantly via overexpression and P-Rex2 by mutations. Both proteins seem to play similar roles as facilitators of cancer cell migration. This raises the possibility that both P-Rex proteins may facilitate cancer metastasis. It is not known whether P-Rex1 and P-Rex2 are co-expressed in the same cancer type. There are limited data associating P-Rex expression or mutation in human tumour profiles to patient survival. A correlation has been shown between P-Rex1 overexpression and poor survival in breast cancer but no other cancer types. The effect of P-Rex2 mutations on human survival outcomes remains unknown. There is evidence to suggest that P-Rex1 mediates its functional effects through Rac and ERK1/2. Sequence homology suggests that P-Rex2 regulates similar downstream targets; however, P-Rex2 regulation of ERK has not been reported. Interestingly, only P-Rex2 interacts with PTEN, resulting in mutual inhibition. The lack of structural information renders it difficult to speculate why PTEN favours P-Rex2 binding over P-Rex1. A crystal structure of P-Rex2/PTEN could prove useful in further understanding the dynamics between their interaction. It appears that although P-Rex1 and P-Rex2 share functional similarities in cancer, there is non-redundancy in their mechanisms of action across different cancer types. P-Rex1 and P-Rex2 may function as oncogenes via amplification and mutation, respectively, and activation of downstream signalling depending on the RacGEF activity. Future studies of the long-term outcomes of P-Rex1 overexpression or P-Rex2 mutation in a larger cohort are required to reveal whether these RacGEFs function as human oncogenes.

Abbreviations

     
  • 4e-BP1

    eukaryotic translation initiation factor 4E binding protein 1

  •  
  • AML

    acute myeloid leukaemia

  •  
  • BRAF

    B-rapidly accelerated sarcoma

  •  
  • DEP

    Dishevelled, Egl-10, and pleckstrin

  •  
  • DH

    Dbl-homologous

  •  
  • ER

    oestrogen receptor

  •  
  • ER+

    oestrogen receptor-positive

  •  
  • ERK

    extracellular signal-regulated kinases

  •  
  • EHBP1

    EH domain-binding protein 1

  •  
  • GPCR

    G-protein-coupled receptor

  •  
  • GSK3

    glycogen synthase kinase 3

  •  
  • GTP

    guanosine-5′-triphosphate

  •  
  • IGF-1R

    insulin-like growth factor 1R

  •  
  • IP4P

    inositol polyphosphate 4-phosphatase

  •  
  • MAPK

    mitogen-activated protein kinase

  •  
  • MEK

    mitogen-activated protein kinase kinase

  •  
  • MMP

    matrix metalloproteinase

  •  
  • mTOR

    mammalian target of rapamycin

  •  
  • NRAS

    neuroblastoma RAS viral oncogene homolog

  •  
  • NRP

    neuropilin

  •  
  • p4e-BP1

    phosphorylated eukaryotic translation initiation factor 4E binding protein 1

  •  
  • PI3K

    phosphoinositide 3-kinase

  •  
  • PP1α

    protein phosphatase 1α

  •  
  • PP2A

    protein phosphatase 2A

  •  
  • P-Rex

    phosphatidylinositol 3,4,5-trisphosphate-dependent Rac exchanger

  •  
  • PAK

    protein 21-activated kinase

  •  
  • pAKT

    phosphorylated AKT

  •  
  • PDZ

    post-synaptic density protein, Drosophila disc large tumour suppressor, and zonula occludens-1 protein

  •  
  • pGSK3

    phosphorylated glycogen synthase kinase 3

  •  
  • PH

    Pleckstrin-homology

  •  
  • PI(3,4,5)P3

    phosphatidylinositol 3,4,5-trisphosphate

  •  
  • PI(3,4)P2

    phosphatidylinositol 3,4-bisphosphate

  •  
  • PKA

    protein kinase A

  •  
  • pmTOR

    phosphorylated mammalian target of rapamycin

  •  
  • PTEN

    phosphatase and tensin homologue

  •  
  • PDGFβ

    platelet-derived growth factor receptor beta

  •  
  • RacGEFs

    Rac guanine nucleotide exchange factors

  •  
  • RAF

    rapidly accelerated fibrosarcoma

  •  
  • ROS

    reactive oxygen species

  •  
  • S1P1

    sphingosine-1-phosphate 1

  •  
  • SDF-1

    stromal-derived factor 1

  •  
  • TCGA

    The Cancer Genome Atlas

  •  
  • VEGF

    vascular endothelial growth factor

Funding

This work was supported by grants from the National Health and Medical Research Council [APP1104614 and APP1128120].

Competing Interests

The Authors declare that there are no competing interests associated with the manuscript.

References

References
1
Welch
,
H.C.E.
,
Coadwell
,
W.J.,
Ellson
,
C.D.
,
Ferguson
,
G.J.
,
Andrews
,
S.R.
,
Erdjument-Bromage
,
H.
et al
(
2002
)
P-Rex1, a PtdIns(3,4,5)P3- and Gβγ-regulated guanine-Nucleotide exchange factor for Rac
.
Cell
108
,
809
821
doi:
2
Donald
,
S.
,
Hill
,
K.
,
Lecureuil
,
C.
,
Barnouin
,
R.
,
Krugmann
,
S.
,
John Coadwell
,
W.
et al
(
2004
)
P-Rex2, a new guanine-nucleotide exchange factor for Rac
.
FEBS Lett.
572
,
172
176
doi:
3
Rosenfeldt
,
H.
,
Vázquez-Prado
,
J.
and
Gutkind
,
J.S.
(
2004
)
P-REX2, a novel PI-3-kinase sensitive Rac exchange factor
.
FEBS Lett.
572
,
167
171
doi:
4
Welch
,
H.C.E.
(
2015
)
Regulation and function of P-Rex family Rac-GEFs
.
Small GTPases
6
,
49
70
doi:
5
Hill
,
K.
,
Krugmann
,
S.
,
Andrews
,
S.R.
,
Coadwell
,
W.J.
,
Finan
,
P.
,
Welch
,
H.C.E.
et al
(
2005
)
Regulation of P-Rex1 by phosphatidylinositol (3,4,5)-trisphosphate and Gβγ subunits
.
J. Biol. Chem.
280
,
4166
4173
doi:
6
Jaffe
,
A.B.
and
Hall
,
A.
(
2005
)
RHO GTPASES: biochemistry and biology
.
Annu. Rev. Cell Dev. Biol.
21
,
247
269
doi:
7
Wertheimer
,
E.
,
Gutierrez-Uzquiza
,
A.
,
Rosemblit
,
C.
,
Lopez-Haber
,
C.
,
Sosa
,
M.S.
and
Kazanietz
,
M.G.
(
2012
)
Rac signaling in breast cancer: a tale of GEFs and GAPs
.
Cell. Signal.
24
,
353
362
doi:
8
Sosa
,
M.S.
,
Lopez-Haber
,
C.
,
Yang
,
C.
,
Wang
,
H.B.
,
Lemmon
,
M.A.
,
Busillo
,
J.M.
et al
(
2010
)
Identification of the Rac-GEF P-Rex1 as an essential mediator of ErbB signaling in breast cancer
.
Mol. Cell
40
,
877
892
doi:
9
Montero
,
J.C.
,
Seoane
,
S.
,
Ocaña
,
A.
and
Pandiella
,
A.
(
2011
)
P-Rex1 participates in Neuregulin-ErbB signal transduction and its expression correlates with patient outcome in breast cancer
.
Oncogene
30
,
1059
1071
doi:
10
Ebi
,
H.
,
Costa
,
C.
,
Faber
,
A.
,
Juric
,
D.
,
Della Pelle
,
P.
,
Song
,
Y.
et al
(
2013
)
Abstract A29: PI3K regulates MEK/ERK signaling in breast cancer via the Rac-GEF, P-Rex1
.
Mol. Cancer Ther.
110
,
21124
21129
doi:10.1158/1535-7163.targ-13-a29
11
Dillon
,
L.M.
,
Bean
,
J.R.
,
Yang
,
W.
,
Shee
,
K.
,
Symonds
,
L.K.
,
Balko
,
J.M.
et al
(
2014
)
P-REX1 creates a positive feedback loop to activate growth factor receptor, PI3K/AKT and MEK/ERK signaling in breast cancer
.
Oncogene
34,
3968
3976
doi:10.1038/onc.2014.328
12
Liu
,
L.
,
Liu
,
Z.
,
Wang
,
H.
,
Chen
,
L.
,
Ruan
,
F.
,
Zhang
,
J.
et al
(
2016
)
Knockdown of PREX2a inhibits the malignant phenotype of glioma cells
.
Mol. Med. Rep.
13
,
2301
2307
doi:10.3892/mmr.2016.4799
13
Lindsay
,
C.R.
,
Lawn
,
S.
,
Campbell
,
A.D.
,
Faller
,
W.J.
,
Rambow
,
F.
,
Mort
,
R.L.
et al
(
2011
)
P-Rex1 is required for efficient melanoblast migration and melanoma metastasis
.
Nat. Commun.
2
,
555
doi:
14
Ryan
,
M.B.
,
Finn
,
A.J.
,
Pedone
,
K.H.
,
Thomas
,
N.E.
,
Der
,
C.J.
and
Cox
,
A.D.
(
2016
)
ERK/MAPK signaling drives overexpression of the Rac-GEF, PREX1, in BRAF- and NRAS-Mutant melanoma
.
Mol. Cancer Res.
14
,
1009
1018
doi:
15
Qin
,
J.
,
Xie
,
Y.
,
Wang
,
B.
,
Hoshino
,
M.
,
Wolff
,
D.W.
,
Zhao
,
J.
et al
(
2009
)
Upregulation of PIP3-dependent Rac exchanger 1 (P-Rex1) promotes prostate cancer metastasis
.
Oncogene
28
,
1853
1863
doi:
16
Goel
,
H.L.
,
Pursell
,
B.
,
Shultz
,
L.D.
,
Greiner
,
D.L.
,
Brekken
,
R.A.
,
Vander Kooi
,
C.W.
et al
(
2016
)
P-Rex1 promotes resistance to VEGF/VEGFR-targeted therapy in prostate cancer
.
Cell Rep.
14
,
2193
2208
doi:
17
Berger
,
M.F.
,
Hodis
,
E.
,
Heffernan
,
T.P.
,
Deribe
,
Y.L.
,
Lawrence
,
M.S.
,
Protopopov
,
A.
et al. (
2012
)
Melanoma genome sequencing reveals frequent PREX2 mutations
.
Nature
485
,
502
506
doi:10.1038/nature11071
18
Lissanu Deribe
,
Y.
,
Shi
,
Y.
,
Rai
,
K.
,
Nezi
,
L.
,
Amin
,
S.B.
,
Wu
,
C.-C.
et al
(
2016
)
Truncating PREX2 mutations activate its GEF activity and alter gene expression regulation in NRAS-mutant melanoma
.
Proc Natl Acad. Sci. U.S.A.
113
,
E1296
E1305
doi:
19
Mense
,
S.M.
,
Barrows
,
D.
,
Hodakoski
,
C.
,
Steinbach
,
N.
,
Schoenfeld
,
D.
,
Su
,
W.
et al
(
2015
)
PTEN inhibits PREX2-catalyzed activation of RAC1 to restrain tumor cell invasion
.
Sci. Signal.
8
,
ra32
doi:
20
Guo
,
B.
,
Liu
,
L.
,
Yao
,
J.
,
Ma
,
R.
,
Chang
,
D.
,
Li
,
Z.
et al
(
2014
)
miR-338-3p suppresses gastric cancer progression through a PTEN-AKT axis by targeting P-REX2a
.
Mol. Cancer Res.
12
,
313
321
doi:
21
Chen
,
X.
,
Pan
,
M.
,
Han
,
L.
,
Lu
,
H.
,
Hao
,
X.
and
Dong
,
Q.
(
2013
)
miR-338-3p suppresses neuroblastoma proliferation, invasion and migration through targeting PREX2a
.
FEBS Lett.
587
,
3729
3737
doi:
22
Lan
,
X.
,
Xiao
,
F.
,
Ding
,
Q.
,
Liu
,
J.
,
Liu
,
J.
,
Li
,
J.
et al
(
2014
)
The effect of CXCL9 on the invasion ability of hepatocellular carcinoma through up-regulation of PREX2
.
J. Mol. Histol.
45
,
689
696
doi:
23
Dai
,
Z.
,
Xie
,
J.
,
Lei
,
P.
and
Hu
,
Y.
(
2016
)
Knockdown of PREX2a inhibits the malignant phenotype of osteosarcoma cells
.
Mol. Med. Rep.
13
,
1930
1936
doi:10.3892/mmr.2015.4728
24
Welch
,
H.C.E.
,
Condliffe
,
A.M.
,
Milne
,
L.J.
,
Ferguson
,
G.J.
,
Hill
,
K.
,
Webb
,
L.M.C.
et al
(
2005
)
P-Rex1 regulates neutrophil function
.
Curr. Biol.
15
,
1867
1873
doi:
25
Yoshizawa
,
M.
,
Kawauchi
,
T.
,
Sone
,
M.
,
Nishimura
,
Y.
,
Terao
,
M.
,
Chihama
,
K.
et al
(
2005
)
Involvement of a Rac activator, P-Rex1, in neurotrophin-derived signaling and neuronal migration
.
J. Neurosci.
25
,
4406
4419
doi:
26
Dong
,
X.
,
Mo
,
Z.
,
Bokoch
,
G.
,
Guo
,
C.
,
Li
,
Z.
and
Wu
,
D.
, (
2005
)
P-Rex1 is a primary Rac2 guanine nucleotide exchange factor in mouse neutrophils
.
Curr. Biol.
15
,
1874
1879
doi:
27
Donald
,
S.
,
Humby
,
T.
,
Fyfe
,
I.
,
Segonds-Pichon
,
A.
,
Walker
,
S.A.
,
Andrews
,
S.R.
et al. (
2008
)
P-Rex2 regulates Purkinje cell dendrite morphology and motor coordination
.
Proc. Natl Acad. Sci. U.S.A.
105
,
4483
4488
doi:
28
Hodakoski
,
C.
,
Hopkins
,
B.D.
,
Barrows
,
D.
,
Mense
,
S.M.
,
Keniry
,
M.
,
Anderson
,
K.E.
et al
(
2014
)
Regulation of PTEN inhibition by the pleckstrin homology domain of P-REX2 during insulin signaling and glucose homeostasis
.
Proc. Natl Acad. Sci. U.S.A.
111
,
155
160
doi:
29
Qian
,
F.
,
Le Breton
,
G.C.
,
Chen
,
J.
,
Deng
,
J.
,
Christman
,
J.W.
,
Wu
,
D.
et al
(
2012
)
Role for the guanine nucleotide exchange factor phosphatidylinositol-3,4,5-trisphosphate–dependent Rac exchanger 1 in platelet secretion and aggregation
.
Arterioscler. Thromb. Vasc. Biol.
32
,
768
777
doi:
30
Waters
,
J.E.
,
Astel
,
M.V.
,
Ooms
,
L.M.
,
Balamatsias
,
D.
,
Gurung
,
R.
and
Mitchell
C. A.
(
2008
)
P-Rex1 — a multidomain protein that regulates neurite differentiation
.
J. Cell Sci.
121
,
2892
2903
doi:
31
Li
,
Z.
,
Paik
,
J.-H.
,
Wang
,
Z.
,
Hla
,
T.
and
Wu
,
D.
(
2005
)
Role of guanine nucleotide exchange factor P-Rex-2b in sphingosine 1-phosphate-induced Rac1 activation and cell migration in endothelial cells
.
Prostaglandins Other Lipid Mediat.
76
,
95
104
doi:
32
Barrio-Real
,
L.
,
Benedetti
,
L.G.
,
Engel
,
N.
,
Tu
,
Y.
,
Cho
,
S.
,
Sukumar
,
S.
et al
(
2014
)
Subtype-specific overexpression of the Rac-GEF P-REX1 in breast cancer is associated with promoter hypomethylation
.
Breast Cancer Res.
16
,
441
doi:
33
Wong
,
C.-Y.A.
,
Wuriyanghan
,
H.
,
Xie
,
Y.
,
Lin
,
M.-F.
,
Abel
,
P.W.
and
Tu
,
Y.
(
2011
)
Epigenetic regulation of phosphatidylinositol 3,4,5-triphosphate-dependent Rac exchanger 1 gene expression in prostate cancer cells
.
J. Biol. Chem.
286
,
25813
25822
doi:
34
Lucato
,
C.M.
,
Halls
,
M.L.
,
Ooms
,
L.M.
,
Liu
,
H.-J.
,
Mitchell
,
C.A.
,
Whisstock
,
J.C.
et al
(
2015
)
The phosphatidylinositol (3,4,5)-trisphosphate-dependent Rac exchanger 1·Ras-related C3 botulinum toxin substrate 1 (P-Rex1·Rac1) complex reveals the basis of Rac1 activation in breast cancer cells
.
J. Biol. Chem.
290
,
20827
20840
doi:
35
Cash
,
J.N.
,
Davis
,
E.M.
and
Tesmer
,
J.J.G.
(
2016
)
Structural and biochemical characterization of the catalytic core of the metastatic factor P-Rex1 and its regulation by ptdIns(3,4,5)P3
.
Structure
24
,
730
740
doi:
36
Pan
,
D.
,
Barber
,
M.A.
,
Hornigold
,
K.
,
Baker
,
M.J.
,
Toth
,
J.M.
,
Oxley
,
D.
et al
(
2016
)
Norbin stimulates the catalytic activity and plasma membrane localization of the guanine-Nucleotide exchange factor P-Rex1
.
J. Biol. Chem.
291
,
6359
6375
doi:
37
Fine
,
B.
,
Hodakoski
,
C.
,
Koujak
,
S.
,
Su
,
T.
,
Saal
,
L.H.
,
Maurer
,
M.
et al
(
2009
)
Activation of the PI3K pathway in cancer through inhibition of PTEN by exchange factor P-REX2a
.
Science
325
,
1261
1265
doi:
38
Barber Donald
,
S.
,
Thelen
,
S.
,
Anderson
,
K.
,
Thelen
,
M.
and
Welch
,
H.
. (
2007
)
Membrane translocation of P-Rex1 is mediated by G protein betagamma subunits and phosphoinositide 3-kinase
.
J. Biol. Chem.
282
,
229976
299967
doi:10.1074/jbc.M701877200
39
Rynkiewicz
,
N.K.
,
Liu
,
H.-J.
,
Balamatsias
,
D.
and
Mitchell
,
C.A.
(
2012
)
INPP4A/INPP4B and P-Rex proteins: related but different?
Adv. Biol. Regul.
52
,
265
279
doi:
40
Barber
,
M.A.
,
Hendrickx
,
A.
,
Beullens
,
M.
,
Ceulemans
,
H.
,
Oxley
,
D.
,
Thelen
,
S.
et al
(
2012
)
The guanine-nucleotide-exchange factor P-Rex1 is activated by protein phosphatase 1α
.
Biochem. J.
443
,
173
183
doi:
41
Barrows
,
D.
,
Schoenfeld
,
S.M.
,
Hodakoski
,
C.
,
Silkov
,
A.
,
Honig
,
B.
,
Couvillon
,
A.
et al
(
2015
)
p21-activated kinases (PAKs) mediate the phosphorylation of PREX2 protein to initiate feedback inhibition of Rac1 GTPase
.
J. Biol. Chem.
290
,
28915
28931
doi:
42
Hernandez-Negrete
,
I.
,
Carretero-Ortega
,
J.
,
Rosenfeldt
,
H.
,
Hernandez-Garcia
,
R.
,
Calderon-Salinas
,
J.V.
,
Reyes-Cruz
,
G.
et al
(
2007
)
P-Rex1 links mammalian target of rapamycin signaling to Rac activation and cell migration
.
J. Biol. Chem.
282
,
23708
23715
doi:
43
Ledezma-Sánchez
,
B.A.
,
García-Regalado
,
A.
,
Guzmán-Hernández
,
M.L.
and
Vázquez-Prado
,
J.
(
2010
)
Sphingosine-1-phosphate receptor S1P1 is regulated by direct interactions with P-Rex1, a Rac guanine nucleotide exchange factor
.
Biochem. Biophys. Res. Commun.
391
,
1647
1652
doi:
44
Dimidschstein
,
J.
,
Passante
,
L.
,
Dufour
,
A.
,
van den Ameele
,
J.
,
Tiberi
,
L.
,
Hrechdakian
,
T.
et al
(
2013
)
Ephrin-B1 controls the columnar distribution of cortical pyramidal neurons by restricting their tangential migration
.
Neuron
79
,
1123
1135
doi:
45
Chávez-Vargas
,
L.
,
Adame-García
,
S.R.
,
Cervantes-Villagrana
,
R.D.
,
Castillo-Kauil
,
A.
,
Bruystens
,
J.G.H.
,
Fukuhara
,
S.
et al
(
2016
)
Protein kinase A (PKA) type I interacts with P-Rex1, a Rac guanine nucleotide exchange factor: effect on PKA localization and P-Rex1 signaling
.
J. Biol. Chem.
291
,
6182
6199
doi:
46
Campbell
,
A.D.
,
Lawn
,
S.
,
McGarry
,
L.C.
,
Welch
,
H.C.
,
Ozanne
,
B.W.
,
Norman
,
J.C.
et al
(
2013
)
P-Rex1 cooperates with PDGFRβ to drive cellular migration in 3D microenvironments
.
PLoS ONE
8
,
e53982
doi:
47
Ghalali
,
A.
,
Wiklund
,
F.
,
Zheng
,
H.
,
Stenius
,
U.
and
Högberg
,
J.
(
2014
)
Atorvastatin prevents ATP-driven invasiveness via P2X7 and EHBP1 signaling in PTEN-expressing prostate cancer cells
.
Carcinogenesis
35
,
1547
1555
doi:
48
Cheng
,
W.-Y.
,
Yang
,
T.-H.O.
and
Anastassiou
,
D.
(
2013
)
Development of a prognostic model for breast cancer survival in an open challenge environment
.
Sci. Transl. Med.
5
,
181ra50
doi:
49
Yang
,
C.
,
Liu
,
Y.
,
Lemmon
,
M.A.
and
Kazanietz
,
M.G.
(
2006
)
Essential role for Rac in heregulin β1 mitogenic signaling: a mechanism that involves epidermal growth factor receptor and is independent of ErbB4
.
Mol. Cell. Biol.
26
,
831
842
doi:
50
Sjöblom
,
T.
,
Jones
,
S.
,
Wood
,
L.D.
,
Parsons
,
D.W.
,
Lin
,
J.
,
Barber
,
T.D.
et al
(
2006
)
The consensus coding sequences of human breast and colorectal cancers
.
Science
314
,
268
274
doi:
51
Nik-Zainal
,
S.
,
Davies
,
H.
,
Staaf
,
J.
,
Ramakrishna
,
M.
,
Glodzik
,
D.
,
Zou
,
X.
et al
(
2016
)
Landscape of somatic mutations in 560 breast cancer whole-genome sequences
.
Nature
534
,
47
54
doi:
52
Waddell
,
N.
,
Pajic
,
M.
,
Patch
,
A.-M.
,
Chang
,
D.K.
,
Kassahn
,
K.S.
,
Bailey
,
P.
et al
(
2015
)
Whole genomes redefine the mutational landscape of pancreatic cancer
.
Nature
518
,
495
501
doi:
53
Liu
,
H.-J.
,
Ooms
,
L.M.
,
Srijakotre
,
N.
,
Man
,
J.
,
Vieusseux
,
J.
,
Waters
,
J.E.
et al
(
2016
)
PREX1 Rac-GEF activity promotes breast cancer cell proliferation and tumor growth via activation of extracellular-signal-regulated kinase 1/2 (ERK1/2) signaling
.
J. Biol. Chem.
doi:10.1074/jbc.M116.743401
54
Lopez-Haber
,
C.
,
Barrio-Real
,
L.
,
Casado-Medrano
,
V.
and
Kazanietz
,
M.G.
(
2016
)
Heregulin/ErbB3 signaling enhances CXCR4-Driven Rac1 activation and breast cancer cell motility via hypoxia-inducible factor 1α
.
Mol. Cell. Biol.
36
,
2011
2026
doi:
55
Wang
,
T.
,
Yu
,
H.
,
Hughes
,
N.W.
,
Liu
,
B.
,
Kendirli
,
A.
,
Klein
,
K.
et al
(
2017
)
Gene essentiality profiling reveals gene networks and synthetic lethal interactions with oncogenic Ras
.
Cell
168
,
890
903.e15
doi:
56
Carracedo
,
A.
and
Pandolfi
,
P.P.
(
2008
)
The PTEN–PI3K pathway: of feedbacks and cross-talks
.
Oncogene
27
,
5527
5541
doi:
57
Soroceanu
,
L.
,
Kharbanda
,
S.
,
Chen
,
R.
,
Soriano
,
R.H.
,
Aldape
,
K.
,
Misra
,
A.
et al
(
2007
)
Identification of IGF2 signaling through phosphoinositide-3-kinase regulatory subunit 3 as a growth-promoting axis in glioblastoma
.
Proc. Natl Acad. Sci. U.S.A.
104
,
3466
3471
doi:
58
Suzuki
,
A.
,
Mimaki
,
S.
,
Yamane
,
Y.
,
Kawase
,
A.
,
Matsushima
,
K.
,
Suzuki
,
M.
et al
(
2013
)
Identification and characterization of cancer mutations in Japanese lung adenocarcinoma without sequencing of normal tissue counterparts
.
PLoS ONE
8
,
e73484
doi:
59
Li
,
Y.Y.
,
Hanna
,
G.J.
,
Laga
,
A.C.
,
Haddad
,
R.I.
,
Lorch
,
J.H.
and
Hammerman
,
P.S.
(
2015
)
Genomic analysis of metastatic cutaneous squamous cell carcinoma
.
Clin. Cancer Res.
21
,
1447
1456
doi:
60
Wang
,
Z.
,
Dong
,
X.
,
Li
,
Z.
,
Smith
,
J.D.
and
Wu
,
D.
(
2008
)
Lack of a significant role of P-Rex1, a major regulator of macrophage Rac1 activation and chemotaxis, in atherogenesis
.
Prostaglandins Other Lipid Mediat.
87
,
9
13
doi:
61
Vivanco
,
I.
and
Sawyers
,
C.L.
(
2002
)
The phosphatidylinositol 3-kinase-AKT pathway in human cancer
.
Nat. Rev. Cancer
2
,
489
501
62
Yuan
,
T.L.
and
Cantley
,
L.C.
(
2008
)
PI3K pathway alterations in cancer: variations on a theme
.
Oncogene
27
,
5497
5510
63
Dyson
,
J.M.
et al (
2012
)
Phosphoinositide phosphatases: just as important as the kinases
.
Subcell. Biochem.
58
,
215
279

Author notes

*

These authors contributed equally to this review.