Signaling mechanisms and physiological functions of G‐protein Gα₁₃ in blood vessel formation, bone homeostasis, and cancer

Abstract

Heterotrimeric G‐proteins are cellular signal transducers. They mainly relay signals from G‐protein‐coupled receptors (GPCRs). GPCRs function as guanine nucleotide‐exchange factors to active these G‐proteins. Based on the sequence and functional similarities, these G‐proteins are grouped into four subfamilies: G_s, G_i, G_q, and G_12/13. The G_12/13 subfamily consists of two members: G₁₂ and G₁₃. G_12/13‐mediated signaling pathways play pivotal roles in a variety of physiological processes, while aberrant regulation of this pathway has been identified in various human diseases. Here we summarize the signaling mechanisms and physiological functions of Gα₁₃ in blood vessel formation and bone homeostasis. We further discuss the expanding roles of Gα₁₃ in cancers, serving as oncogenes as well as tumor suppressors.

Introduction

Heterotrimeric G‐proteins have three subunits: α, β, and γ. Gβ and Gγ subunits are associated together and function as one functional unit. G‐proteins are classified based on their Gα subunits. In the Gα_12/13 subfamily, both Gα₁₂ and Gα₁₃ are expressed ubiquitously. As with other G‐proteins, Gα_12/13 undergoes a GTPase cycle. In the inactive form, guanine diphosphate (GDP) bound Gα subunit binds to heterodimer Gβγ. Upon ligand binding to a G‐protein‐coupled receptor (GPCR), the receptor acts as a guanine nucleotide‐exchange factor (GEF), promoting the release of bound GDP from Gα. Nucleotide‐free Gα then binds to guanine triphosphate (GTP) and dissociates from Gβγ.1 Both Gα and Gβγ can signal to downstream effectors.

The most familiar action of the G_12/13 subfamily is to activate the Ras‐superfamily small GTPase RhoA in response to a GPCR that influences a number of cellular responses including actin cytoskeletal reorganization, cell migration, phospholipase D, protein kinase D, Na⁺–H⁺ exchange, JNK activation and serum response factor (SRF)‐mediated gene transcription.2, 3, 4 The mechanism for Gα_12/13 signaling to RhoA is through RhoGEFs which contain a RGS (regulator of G‐protein signaling)‐like domain (thus, called RGS‐RhoGEFs) that binds to G‐protein α subunits.5 Three different RGS‐RhoGEFs have been identified including p115‐RhoGEF, LARG, and PDZ‐RhoGEF.2 Gα_12/13 also interacts with a variety of other interacting proteins including Btk/Tec‐family non‐receptor tyrosine kinases, cadherin, radixin, socius, endothelial nitric oxide synthase (eNOS), A‐kinase anchoring proteins, protein phosphatase 5, and Abl.2, 3, 4, 6, 7 Additionally, Gα₁₂ and Gα₁₃ appear to have their own separate interacting proteins. Gα₁₂ interacts with Ras‐GAP1^m (Ras‐specific GTPase‐activating protein), Axin, RGS1, AKAP‐Lbc (A‐kinase anchoring protein‐lymphoid blast crisis oncogenes), ZO‐1 (Zonula Occludens), PP2A (protein phosphatase 2A), p120‐catenin, αSNAP (soluble NSF attachment protein), and Hsp90 (heat‐shock protein 90).3, 8 Gα₁₃ interacts with RGS16, Pyk2 (proline‐rich tyrosine kinase 2), Hax‐1 (HS1‐associating protein 1), JLP1 (JNK‐interacting leucine zipper protein), AKAP110 (A‐kinase anchoring protein 110), ASK‐1 (apoptosis signal‐regulating kinase 1), and GRK4γ (G‐protein‐coupled receptor kinase 4γ).3 Although the list of the interacting partners continues to increase, the physiological importance of these interactions is, in many cases, still unclear. Nevertheless, it becomes more and more apparent that Gα_12/13‐mediated signaling operates in almost every tissue and manifests itself in a context specific manner. In this review, we will focus on our understandings of the signaling mechanisms and physiological/pathophysiological functions of Gα₁₃ in specific cell types including endothelial cells (ECs), osteoclasts (OCs), and tumor cells.

Role of Gα₁₃ signaling in endothelial cells and blood vessel formation

Formation of blood vessels during mammalian embryonic development is a complex and highly regulated process. Angioblasts proliferate, migrate, and differentiate to form primitive vascular structures composed of ECs. These structures remodel by sprouting, branching, growing and regressing, and mature by the recruitment and differentiation of pericytes and smooth muscle cells.9 The essential role of Gα₁₃ signaling for proper blood vessel formation during mammalian embryonic development as well as adult angiogenesis has been recognized for 20 years now, however, the underlying molecular mechanisms are not completely understood. It was shown that Gα₁₃ deficient (Gα₁₃ ^−/−) mice died at embryonic Day 9.5.10 The yolk sac of Gα₁₃ ^−/− mouse embryos (at E9.5) did not show any blood vessels. Importantly, the lack of Gα₁₃ did not affect the differentiation of progenitor cells into ECs, which were present throughout the embryo. It was further confirmed that EC‐specific Gα₁₃ knockout (KO) mouse embryos also died at E9.5 with the same phenotypes as conventional Gα₁₃ KO embryos.11 Notably, Gα₁₃ deficiency could be rescued by a transgene expressing Gα₁₃ under the control of an EC‐specific promoter.11 These mouse genetic studies demonstrate that Gα₁₃ is essential for proper blood vessel formation in vascular development. On the other hand, mice lacking Gα₁₂ appeared normal.12 However, Gα₁₂ and Gα₁₃ double‐deficient mice died at embryonic day 8.5, and the Gα₁₂‐deficient mice that carried only one intact Gα₁₃ allele also died in utero.12 These observations suggest that Gα₁₃ has an overlapping as well as distinct functions with Gα₁₂ in early mouse development.

In addition to the essential role for Gα₁₃ signaling in embryonic vascular function, Gα₁₃ also plays critical roles in adult angiogenesis in various physiological and pathological processes. The ovary, for example, undergoes hormonally regulated changes manifested by growth of the ovarian follicles accompanied by the expansion of blood vessels.13 Gα₁₃ signaling contributes to ovarian angiogenesis, which is necessary for proper fertility.14 In Gα₁₃ ^+/− heterozygous mice, tumor angiogenesis is also impaired.14 The third example of postnatal angiogenesis for Gα₁₃ signaling is in the retina which is avascular at birth and a single superficial layer of blood vessels grows progressively from the center toward the periphery from postnatal Day (P) 1 until P7.15 In this physiological condition, Gα₁₃ controls angiogenesis through regulation of vascular endothelial growth factor‐2 (VEGFR‐2) expression.16 Gα₁₃ –mediated VEGFR‐2 expression involves activation of the small GTPase RhoA and transcription factor NF‐κB. Importantly, EC‐specific Gα₁₃ KO mice showed impaired retinal angiogenesis and tumor angiogenesis.16

In canonical G‐protein signaling pathways, Gα₁₃ signaling is initiated by GPCRs (Fig. 1). It has been proposed that protease‐activated receptor 1 (PAR1) and sphingosine‐1‐phosphate receptors (such as S1PR1) have important roles in regulating EC functions during vascular development via signaling though Gα₁₃.17, 18 Gα₁₃ signaling can also be initiated through non‐canonical signaling pathways, which are less‐well defined. In ECs and fibroblasts, Gα₁₃ is required for receptor tyrosine kinase (RTK)‐induced cell migration and actin cytoskeleton reorganization8, 19 (Fig. 1). Ric‐8A, a non‐GPCR GEF for some Gα proteins, is critical in relaying RTK signals to Gα₁₃ suggesting that this non‐canonical Gα₁₃ signaling pathway maybe GPCR‐independent.19, 20, 21 Cell migration and cell remodeling including cell shape changes are necessary for proper EC coordinated behavior and therefore functional blood vessel formation and angiogenesis,9 illustrating the importance of this non‐canonical Gα₁₃ signaling pathway for vascular functions. Cell remodeling requires dynamic control of actin cytoskeletal reorganization. Gα₁₃ has been shown to control actin cytoskeletal reorganization by regulating the disassembly of dorsal ruffles.8, 20, 21 Dorsal ruffles are induced in response to growth factors such as platelet‐derived growth factor (PDGF). They are intense bursts of ruffling of the dorsal surface plasma membranes as seen under the phase‐contrast microscope.22 It is thought dorsal ruffles are needed to reorganize the actin cytoskeleton to prepare a static cell for motility.23 Dorsal ruffles are also involved in other cell physiological functions such as micropinocytosis, invasion and receptor internalization.24 Although, Gα₁₃ could act through Rho‐GEF to activate RhoA linking to actin cytoskeleton,5 RhoA activation seemed not to contribute to EC remodeling since, in Gα₁₃‐deleted ECs, RhoA activation was observed to be normal.11 On the other hand, the duration of Rac activation is extended in Gα₁₃‐deleted cells.8 Notably, re‐expression of Gα₁₃ shortens the duration of Rac activity in cells.8 It is not clear how Gα₁₃ is linked to Rac. However, Rac is activated by Abl tyrosine kinase in PDGF‐induced dorsal ruffle formation in mammalian cells.25, 26, 27 Abl is a multi‐functional kinase that regulates signaling pathways implicated in cytoskeleton reorganization that are important for cell migration, morphogenesis and adhesion, among others cellular functions (reviewed in Ref. 28). Abl could potentially provide such a link to connect Gα₁₃ to Rac (Fig. 1). Gα₁₃ directly interacts with Abl.7 Indeed, this direct interaction is critical for Gα₁₃‐induced dorsal ruffle turnover, endothelial cell remodeling, and cell migration.28 Future investigation will be needed to explore the Gα₁₃ –Abl axis signaling pathways in ECs and their contributions to blood vessel formation.

Figure 1.

Signaling pathways regulated by Gα₁₃ in endothelial cells (ECs). Both GPCRs, including protease‐activated receptor 1 (PAR1) or sphingosine‐1‐phosphate receptor (S1PR), and RTKs, including VEGFR2, can signal to Gα₁₃ in ECs. β‐cat, β‐catenin; Abl, Abl tyrosine kinase; Abi, Abl interactor; GDP, guanosine diphosphate; GTP, guanosine triphosphate; NF‐kB, nuclear factor kappa‐light‐chain‐enhancer of activated B; RhoGEF, Rho‐specific guanine nucleotide exchange factor; RhoA, Ras homolog gene family, member A; ROCK, Rho‐associated protein kinase; Ric8A, resistance to inhibitors of cholinesterase‐8A; RTK, receptor tyrosine kinase; VEGF, vascular endothelial growth factor; VEGFR, Vascular endothelial growth factor receptor; VASP, vasodilator‐stimulated phosphoprotein; WAVE, Wiskott–Aldrich syndrome protein family member.

Role of Gα₁₃ signaling in osteoclast cells and bone homeostasis

Osteoclasts (OCs) are multi‐nucleated bone resorptive cells derived from fusion of bone marrow cells of the monocyte/macrophage lineage.29 In response to extracellular stimuli, OCs can switch from resorptive to migratory states (Fig. 2). OC activation is initiated with matrix recognition, adhesion to the bone surface followed by actin cytoskeletal reorganization. The actin cytoskeleton in OCs is a unique structure that polarizes the resorptive machinery to the bone‐cell interface where it creates an isolated resorptive microenvironment consisting of an actin ring surrounding a ruffled border.30 The actin ring is also known as sealing zone (SZ) that is a prerequisite for the function of OCs. Inhibition of the actin ring leads to the inhibition of bone reabsorption.31 The SZ is composed of tightly packed dot‐like actin‐ and integrin‐rich attachment structures called podosomes. Podosomes are similar to dorsal ruffles, and they differ in their cellular locations (basal versus dorsal cell surfaces).32 Podosomes in OCs first appear as small actin dots/patches which are then reorganized into small rings or rosettes with a diameter of 0.5‐1 μm.33 Those small rings transform into sealing zones by fusing and positioning the podosomes at the basal periphery of the OCs. A particular feature of the OC podosome ring is that the SZ contains organized cortical actin cytoskeleton, which consists of a dense circumferential band of actin filaments.32 Once done with degrading a specific spot of bone, the OC then disassembles SZ and adopts a migratory state characterized by the formation of a “front‐to‐back” migratory polarity.33 Thus, organization of the OC actin cytoskeleton is an essential component of its capacity to resorb bone (Fig. 2).

Figure 2.

Regulation of the actin cytoskeleton in osteoclast resorption and migration states. In the stationary state, OCs are apical‐basally polarized with sealing zone (SZ) that contains actin and integrins, and area of membrane tightly juxtaposed to the bone surface that defines the resorption lacuna. SZ delineates the ruffled border where bone degradation takes places. Once the resorption stops, OCs enter into the migration state where they disassemble the SZ, flatten, and start the migration by becoming polarized with a leading and a trailing edge. Rac and Rho contribute to different states of OC biology. N, nuclei.

Given the important roles of actin cytoskeleton to OC resorptive and migratory states, it is not surprising that Gα₁₂ and Gα₁₃ have been shown to play critical roles in the biology of OCs. Gα₁₂ KO (Gα₁₂ ^−/−) and Gα₁₃ KO (Gα₁₃ ^−/−) mice present defective bone phenotypes.34, 35 Whereas Gα₁₂ ^−/− mice show osteopetrotic phenotype with increased trabecular bone volumes,35 Gα₁₃ ^−/− mice have severe osteoporosis phenotype with weak bone density.34 It is unclear what led to such an opposite phenotype considering that Gα₁₂ and Gα₁₃ share an ~67% amino acid identity, and many of the same upstream GPCRs and downstream effector molecules.4 The differences might be due to that Gα₁₂ ^−/− mice were with a conventional KO while Gα₁₃ ^−/− mice were with osteoclast‐lineage‐specific conditional KO.34, 35

Gα₁₃‐deficiency triggers an increase in both osteoclast number and activity. Mechanistically, it is proposed that Gα₁₃ antagonizes osteoclast formation and activity by attenuating the Akt‐GSK3β‐NFATc1 signaling axis.34 In addition to the increase in bone resorption, Gα₁₃‐deficient OCs exhibit an increase in the actin ring formation and those actin rings were observed to be three‐fold larger in Gα₁₃ ^−/− cells than in wild‐type cells, similar to the observation in fibroblast cells.17 Moreover, Gα₁₃‐deficient OCs secreted more of Cathepsin K, a protease that is necessary to degrade bone.36 This suggests a possibility that Gα₁₃‐deficient osteoclasts display hyper‐activation due to a larger actin ring which may allow for more surface areas for those cells to secrete more bone‐degrading molecules and thus resorb more bones, leading to osteoporosis. Similar to Gα₁₃ ^−/− fibroblasts, Gα₁₃ ^−/− OCs have larger cell sizes. In the future, it is necessary to further investigate the actin ring formation and disassembly to understand the actin cytoskeletal dynamics in OCs, for example, by live‐cell microscopy.

It has been shown that both RhoA and Rac GTPases play important roles in OC biology37 (Fig. 2). The migratory state of OCs, where OCs acquire a thicker and less spread‐out morphology on bones, is correlated with high basal RhoA activity.38 Additionally, Rho GTPase activity is proposed to decrease as podosome pattern changes in OCs from dots/patches to small actin rings at the migratory state to stable actin ring at the SZ at the stationary state.33 In contrast, Rac contributes to resorptive state of OCs, where OCs are in a stationary and spread‐out morphology on bones.37 Injecting anti‐Rac antibodies into OCs to inhibit Rac1 or Rac2 disrupts actin ring formation, reduces OC resorption and causes retraction of OCs.39 Potentially, Gα₁₃ may regulate the activation of both RhoA and Rac, serving as a decisive hub to regulate different states of OC functions (Fig. 2). In the future, it is important to demonstrate that RhoA and Rac activity is indeed perturbed in Gα₁₃‐deficient OCs and to correlate this information to the states of OCs.

Role of Gα₁₃ signaling in cancer cells

Gα₁₃ as an oncogene

Gα₁₂ and Gα₁₃ were initially identified as oncogenes with the potential for neoplastic transformation of fibroblast cell lines.40, 41, 42, 43 They were further confirmed to stimulate mitogenic responses and cell growth in many different cell lines.43 Additionally, Gα_12/13 subfamily has been shown to regulate cell migration, invasion and metastasis,2 which are all critical cellular processes for tumor progression. Hence, Gα_12/13 proteins have been designated as “gep” oncogenes and are of great interested to cancer biologists.43 To date, the signaling pathways downstream of Gα_12/13 that promote oncogenic transformation, tumor cells growth, cell migration, invasion and metastasis have been studied extensively.2, 44 Gα_12/13 activation can promote tumorigenesis and cell growth in ovarian, small‐cell lung and hepatocellular cancer cells, but not in breast and prostate cancer cells.45, 46, 47, 48, 49, 50 Recently, it has also been shown that Gα_12/13 signaling regulates proliferation of ovarian cancer cells via the Hippo pathway through activation of the transcriptional coactivator Yes‐associated protein (YAP).50 YAP promotes tissue growth and cell viability by regulating the activity of multiple transcription factors.50, 51 The precise mechanism is unknown, but it is clear that Gα_12/13 regulation of YAP depends on both RhoA and F‐actin.50 Together, these suggest that a diverse repertoire of Gα_12/13‐mediated pathways that preferentially activate downstream effectors in cell type specific /context‐dependent matters leading to differential effects in different types of cells.

The levels of Gα_12/13 mRNAs and/or proteins are elevated in patients with breast or prostate cancers.45, 46 Moreover, a systematic meta‐analysis of gene‐expression microarray datasets revealed that Gα₁₂ and Gα₁₃ are overexpressed in breast, oral, esophageal and colon cancers.52 In addition, the meta‐analysis of expression signatures from ∼18,000 human tumors with overall survival outcomes across 39 malignancies revealed associations between the expression of Gα₁₂ and Gα₁₃ and poor prognosis in patients with glioblastoma, oral, breast, lung, kidney, and ovarian cancers.53 Despite the transforming capacity of constitutive Gα₁₂ and Gα₁₃ mutants in experimental systems and numerous implications of these G‐proteins in cancers, activating mutants in the Gα₁₂ and Gα₁₃ genes in patient tumor samples have not been described.52

Gα₁₃ as a tumor suppressor

Contrasting with the oncogenic roles of Gα₁₂ and Gα₁₃, loss‐of‐function mutations in Gα₁₃ promote proliferation and dissemination of B‐cell‐derived lymphoma.54, 55 The presence of Gα₁₃ gene loss‐of‐function mutations in Burkitt's lymphoma and diffuse large B‐cell lymphoma (DLBCL) were first identified in large‐scale genomic DNA sequencing studies.56, 57 Germinal center (GC) B cells, unlike most lymphocytes, are tightly confined in lymphoid organs and do not recirculate. Deficiency in Gα₁₃ led to GC B‐cell dissemination into lymph and blood55 (Fig. 3). It has been proposed that the effector pathway for Gα₁₃ in lymphocytes is the activation of ARH‐GEF1 which leads to activation of RhoA.54, 58 This Gα₁₃–RhoA axis has suppressive effect on Akt phosphorylation which leads to the precise control of GC B cell proliferation and migration. Gα₁₃ signaling helps to confine B cells to the GC and to limit B‐cell expansion. Loss of Gα₁₃ would, therefore, allow cells to escape the GC to lymph nodes and blood stream55 (Fig. 3). Frequent mutations in Gα₁₃ gene and RhoA gene are observed in B‐cell lymphomas.59 Interestingly, the expression of wild‐type Gα₁₃ in B‐cell lymphoma cell line (with a mutant Gα₁₃ gene) has limited impact on proliferation of these cells in vitro but results in a remarkable growth inhibition in vivo utilizing a tumor xenograft model.59

Figure 3.

Gα₁₃‐mediated signaling pathway in germinal center (GC) cells and diffuse large B‐cell lymphoma (DLBCL) cells. (A) In GC cells, S1PR2 initiates Gα₁₃‐mediated signaling leading to the suppression of Atk phosphorylation, which leads the control of apoptosis and confinement of B‐cells to the GC. (B) In DLBCL cells, the loss of suppressive pathway via the absence of Gα₁₃ leads to the increase in survival and migration of GC B‐cells.

Furthermore, it has been shown that Gα₁₃‐mediated pathway in DLBCL could also be Atk‐independent.60 In these B‐cells, the signaling can be initiated through S1PR2 which also promotes apoptosis of GC B cells. S1PR2 gene expression is essential for the B‐cell confinement to the GC. Loss of Gα₁₃ in DLBCL, results in a suppression of S1PR2 gene expression by the forkhead box protein (FOX1P) transcription factor leading to loss of B‐cell confinement to the GC and an increase of B‐cell survival.60

Although the detailed mechanisms by which Gα₁₃ inhibits cell growth and migration of B‐cells requires further exploration, it is interesting to note that this signaling elicits the opposite effect in B‐cell lymphomas as compared to other cancer cell types. As discussed above, Gα₁₃ is associated with cellular transformation and tumorigenesis in other cancer cells. The apparent‐opposing effects of a particular signaling molecule, based on cellular context, have been observed with other pathways such as the JNK and transforming growth factor beta (TGFβ) signaling pathways.61, 62 Further investigations will reveal the molecular mechanisms of Gα₁₃ signaling for these different functions in different cancer cell types.

Conclusion

We have briefly reviewed the data on the signaling mechanisms and functions of Gα₁₃ in endothelial cells, osteoclast cells, and cancer cells. In ECs and OCs, Gα₁₃ seems to mainly control the actin cytoskeletal reorganization to contribute to blood vessel formation or bone resorption, respectively. In most types of cancer cells, Gα₁₃ promotes cell proliferation; but in B‐cell lymphoma, Gα₁₃ confines B‐cells in germinal centers and limits B‐cell expansion. Gα₁₃ can be activated by the conical GPCRs, as well as by other signaling pathways, such as RTKs, possible through non‐GPCR GEFs such as Ric‐8A. The exact cellular and physiological or pathological outcomes of Gα₁₃ signaling depend on cell‐types and cell functional states.