3–Gab1 Signaling Complex for Keratinocyte Growth Factor–Induced PI3K–AKT–mTORC1 Activation

ORIGINAL ARTICLE

Requirement of Gai1/3–Gab1 Signaling Complex for Keratinocyte Growth Factor–Induced PI3K–AKT–mTORC1 Activation Yi-ming Zhang1, Zhi-qing Zhang2, Yuan-yuan Liu2, Xin Zhou1, Xiao-hua Shi1, Qin Jiang3, Dong-li Fan1 and Cong Cao2 Keratinocyte growth factor (KGF), also termed as fibroblast growth factor-7, promotes proliferation, migration, and adhesion of skin keratinocytes via binding to keratinocyte growth factor receptor (KGFR) and subsequent activation of downstream signaling including the PI3K-AKT-mTORC1 pathway. Here, we found that the a-subunits of the G proteins (Gai1/3) and growth factor receptor binding 2-associated binding protein 1 (Gab1) are required for this activation process. With KGF stimulation, Gai1/3 formed a complex with KGFR and was required for subsequent Gab1 recruitment, phosphorylation, and following PI3K-p85 activation. In addition, Gai1/3 short hairpin RNA knockdown largely inhibited KGF-induced cell proliferation, migration, and the accumulation of cyclin D1/fibronectin in cultured skin keratinocytes. Furthermore, we observed increased expression of Gai1/3 in wounded human skin and keloid skin tissues, suggesting the possible involvement of Gai1/3 in wound healing and keloid formation. Overall, we suggest that Gai1/3 proteins lie downstream of KGFR, but upstream of Gab1-mediated activation of PI3K  AKT-mTORC1 signaling, thus revealing a role for Gai proteins in mediating KGFR signaling, cell migration, and possible wound healing. Journal of Investigative Dermatology (2015) 135, 181–191; doi:10.1038/jid.2014.326; published online 18 September 2014

INTRODUCTION Keratinocyte growth factor (KGF), also termed as fibroblast growth factor-7 (FGF-7), belongs to the FGF family (Werner, 1998). KGF expression is relatively weak in normal skin. However, it is significantly upregulated in skin fibroblasts 1

Department of Plastic and Cosmetic Surgery, Xinqiao Hospital, Third Military Medical University, Chongqing, China; 2Jiangsu Key Laboratory of Translational Research and Therapy for Neuro-Psycho-Diseases and Institute of Neuroscience, Soochow University, Suzhou, China and 3The Affiliated Eye Hospital, Nanjing Medical University, Nanjing City, China Correspondence: Qin Jiang, The Affiliated Eye Hospital, Nanjing Medical University, 138 Han-zhong Road, Nanjing City, Jiangsu 210029, China. E-mail: [email protected]; or Dong-li Fan, Department of Plastic and Cosmetic Surgery, Xinqiao Hospital, Third Military Medical University, Xinqiao Road, Sha Ping Ba District, Chongqing 400037, China. E-mail: [email protected] or [email protected]; or Cong Cao, Jiangsu Key Laboratory of Translational Research and Therapy for Neuro-Psycho-Diseases and Institute of Neuroscience, Soochow University, Suzhou, Jiangsu 215021, China. E-mail: [email protected] Abbreviations: bFGF, basic fibroblast growth factor; DKO, double knockout; DN, dominant negative; EGF, epidermal growth factor; FGF, fibroblast growth factor; FRS-2, FGF receptor substrate 2; Gab1, growth factor receptor binding 2-associated binding protein 1; GPCR, G-protein–coupled receptor; KGF, keratinocyte growth factor; KGFR, keratinocyte growth factor receptor; MEF, mouse embryonic fibroblast; mTOR, mammalian target of rapamycin; mTORC, mTOR complex; PDGF, platelet-derived growth factor; PI3K, phosphatidylinositol 3-kinase; RNAi, RNA interference; RTK, receptor tyrosine kinase; WT, wild type Received 30 March 2014; revised 24 June 2014; accepted 14 July 2014; accepted article preview online 31 July 2014; published online 18 September 2014

& 2015 The Society for Investigative Dermatology

with a skin wound (Werner et al., 1992; Beer et al., 2000), which regulates proliferation, differentiation, migration, and adhesion of surrounding skin keratinocytes to facilitate wound healing (Werner et al., 1992; Tsuboi et al., 1993; Werner et al., 1994; Danilenko et al., 1995; Werner, 1998). Thus, dermalderived KGF regulates wound skin reepithelialization in a paracrine manner. KGF regulates mitogenic effect by binding to its receptor: keratinocyte growth factor receptor (KGFR), a splicing transcript variant of the FGF receptor 2 (Shaoul et al., 1995). KGFR is also a high-affinity receptor for fibroblast growth factor-10 (FGF-10 or KGF-2) and fibroblast growth factor-1 (FGF-1; Miki et al., 1991; Igarashi et al., 1998). KGFR is a cell surface receptor tyrosine kinase (RTK; Ceridono et al., 2005). Following KGF binding, KGFR is phosphorylated at multiple tyrosine residues in its intracellular domain, providing docking sites of scaffold proteins (i.e., phospholipase C-g and FGF receptor substrate 2 (FRS-2)) to activate downstream signaling pathways including the mitogen-activated protein kinases (MAPKs) and phosphatidylinositol 3-kinase (PI3K)/AKT/mammalian target of rapamycin (mTOR) signaling. Activation of these pathways eventually mediates the effects of KGF on cell proliferation, migration, differentiation, and survival (Shaoul et al., 1995; Braun et al., 2006; Belleudi et al., 2011a, 2011b). The heterotrimeric guanine nucleotide–binding proteins (G proteins) of the Gai family of proteins, including Gai1, Gai2, and Gai3, were originally identified by their ability to couple to G protein–coupled receptors (GPCRs) and to inhibit www.jidonline.org

181

Y-m Zhang et al. Gai1/3–Gab1 Complex Mediates KGF Signaling

adenylate cyclase (Alessi et al., 1997). Our previous study has identified an unexpected role of Gai proteins in transducing EGFR signaling (Cao et al., 2009). Gai1 and Gai3 are essential for activation of PI3K  AKT-mammalian mTOR complex 1 (mTORC1) signaling by the EGFR ligands. Upon EGF stimulation, Gai proteins associate with EGFR and the adaptor protein growth factor receptor binding 2-associated binding protein 1 (Gab1) to mediate subsequent activation of PI3K-AKT-mTORC1 signaling (Cao et al., 2009). Recently, Wang et al. (2014) showed that Gai1/3 proteins differentially regulate AKT  mTORC1 and Erk1/2 activations in response to different growth factors (Wang et al., 2014). Here, we show that Gai1/3 and adaptor protein Gab1 are required for KGFinduced activation of PI3K-AKT-mTORC1 signaling and are important for keratinocyte migration and proliferation. RESULTS Gai proteins are required for the activation of AKT-mTORC1, but not Erk, by KGFR family members in MEFs

To investigate the potential role of Gai1 and Gai3 in KGFinduced activation of the AKT-mTORC1 pathway, we initially utilized mouse embryonic fibroblasts (MEFs) derived from wild-type (WT) or mouse embryos deficient in Gai1 and Gai3 (termed double knockout (Gai1/3 DKO) cells; (Cao et al., 2009). As there is no KGFR expression in regular MEFs, we exogenously expressed KGFR into MEFs (MEFs/ KGFR, both WT and DKO) and generated stable cell lines (see Materials and Methods). Reverse transcriptase–PCR and western blotting results confirmed KGFR mRNA and protein expression in stable MEFs (Figure 1a and b). Kinetic and dose–response experiments demonstrated that MEFs/KGFR lacking Gai1 and Gai3 were severely impaired in KGFinduced AKT activation, which was detected by phosphorylation of AKT-308T, AKT-473S and GSK-3b (9S; Figure 1c and Supplementary Figure S1 online). We also tested the requirement of Gai proteins in the activation of mTOR by KGF. Activated AKT phosphorylates and inhibits TSC2, interfering with the ability of TSC2 to inhibit mTORC1, leading to mTORC1 activation and 4E-BP1/ p70S6K1 (S6K) phosphorylation. We found that KGF-induced mTORC1 activation, indicated by phosphorylation of 4E-BP1 (65S) and S6 (235/236S), was significantly inhibited by Gai1 and Gai3 deficiency (Figure 1d and ,Supplementary Figure S1 online). As expected, EGF-induced phosphorylation of AKT, S6K, and S6 was abolished by Gai1 and Gai3 double knockouts (Figure 1c and d). To study the individual role of Gai1 or Gai3 in the regulation of KGF-induced AKT-mTORC1 signaling, Gai1 or Gai3 single-knockout MEFs were applied. Similarly, KGFR was exogenously expressed into Gai1 or Gai3 single-knockout MEFs. As demonstrated, KGF-induced AKT-S6 phosphorylation was inhibited in the above singleknockout MEFs/KGFR, whereas Gai1 and Gai3 DKO almost abolished AKT-S6 phosphorylation by KGF (Figure 1e). Gai3 deficiency exerted more inhibitory effect on KGF-induced AKT-S6 phosphorylation than did Gai1 deficiency (Figure 1e, quantification). Note that the basal phosphorylation of the above proteins was low and almost undetectable in the above cell lines (data not shown). In contrast, loss of Gai1 and Gai3 182

Journal of Investigative Dermatology (2015), Volume 135

had no effect on platelet-derived growth factor (PDGF)-BBinduced phosphorylation of AKT (473S; Figure 1h). Significantly, bFGF- and another KGFR ligand FGF-10-induced AKT activation was also markedly inhibited in Gai1/3 DKO MEFs/ KGFR (Figure 1g). To assess whether the defect in the activation of the AKTmTORC1 signaling by KGF in DKO MEFs/KGFR occurred at the level of the receptor, we examined the expression and activation of KGFR in the above MEFs. Activation of KGFR was detected by measuring phosphorylation of KGFR and its physiological substrate FRS-2 after KGF stimulation (Ceridono et al., 2005). Results showed that KGFR and FRS-2 expression in DKO MEFs/KGFR was equivalent to that of WT MEFs/KGFR (Figure 1c and f). KGF-induced phosphorylation of KGFR (653/ 654Y) and FRS-2 (196Y) was almost the same between WT and DKO MEFs (Figure 1f), indicating that Gai proteins possibly lie downstream of KGFR and FRS-2. A recent study by Wang et al. (2014) demonstrated that Gai1/3 deficiency impaired aFGF- and bFGF-induced phosphorylation of AKT (S473) and S6 (S235/236) but not Erk1/2 (T202/Y204; Wang et al., 2014). Here we found that KGF- and FGF-10-induced Erk1/2 activation was not affected by Gai protein deficiency (Figure 1e-g), indicating the selective requirement of Gai proteins for activation of AKT-mTORC1 signaling by KGFR family members. Together, these results suggest that Gai1 and Gai3 proteins probably act downstream of the KGFR-FRS-2 to regulate activation of AKT-mTORC1 by KGF, FGF-10, and bFGF. Note that the expression of Gai2 in Gai1/3 DKO MEFs was lower than that in WT MEFs (Figure 1g and h), and similar results were also seen in the recent study (Wang et al., 2014). Gai1 and Gai3 proteins are vital for the phosphorylation of AKT-473S and S6 (65S) in response to KGF in KGFR-expressing MEFs

To further rule out the possibility that the DKO MEFs/KGFR behaved differently from WT MEFs/KGFR in response to KGF, we verified these observations using RNA interference (RNAi) and dominant-negative (DN) strategies (Cao et al., 2009). Consistent with the above results, we observed that combined knockdown of Gai1 and Gai3 in MEFs largely inhibited the phosphorylation of AKT (473S) and GSK-3b (9S), as well as 4E-BP1 (65S) and S6 (235/236S) by KGF. Gai1 and Gai3 RNAis did not affect the expression or phosphorylation of KGFR, FRS-2, or Erk (Figure 2a and b). PDGF-BB-induced phosphorylation of AKT, S6K, S6, and Erk was not affected by Gai1 and Gai3 RNAis (data not shown). Significantly, we observed that reconstitution of WT-Gai3 into DKO MEFs/KGFR restored phosphorylation of AKT (473S) and S6 (235/236S) in response to KGF (Figure 2c). In contrast, the DN mutant Gai3 was not able to rescue KGFR-induced AKT/S6 phosphorylation (Figure 2c). In this DN Gai3 mutation, we substituted a conserved Gly(G) residue with Thr (T) in the G3 box of Gai3 (Cao et al., 2009). Gai3 with this mutation was unable to bind to other proteins (Hubbard and Hepler, 2006; Barren and Artemyev, 2007; Cao et al., 2009). Notably, KGFR expression and phosphorylation after KGF stimulation were

Y-m Zhang et al.

MEFs/KGFR

5′ 10′ 20′

0′

5′ 10′ 20′

0′

-1,019 bp

KGFR

KGFR GAPDH

p-Akt s473

-175 kDa

KGFR

-120 kDa

β-Actin

-45 kDa

5′ 10′

20′

0′

5′ 10′ 20′

D KO

S6 p-4E-BP1 s65

Akt

EGFR

Gαi-1+3-DKO

p-S6 s235/236

p-Akt t308

Keratinocytes

MEFs/KGFR-C2

MEFs/KGFR-C1

MEFs

-983 bp

EGF

KGF (50 ng ml–1)

WT

+3 -

+3 -

Gαi-1+3-DKO

W T G αi -1

0′

MEFs/KGFR

D KO

EGF

KGF (50 ng ml–1) WT

W T G αi -1

MEFs

MEFs/KGFR

Keratinocytes

Gai1/3–Gab1 Complex Mediates KGF Signaling

p-Erk1/2

p-GSK3β s9

Erk1/2

GSK3β

β-Actin Gαi-1 Gαi-3

MEFs/KGFR

KGFR

2′

5′

0′

2′

5′

p-KGFR

Akt Gαi-3

KGFR

p-Akt s473

p-Akt s473

p-Akt t308

Gαi-1

FGF-10

0′

bFGF

Gαi1+3-DKO

WT KGF

bFGF

WT Gαi-1+3-DKO

MEFs/KGFR

FGF-10

Gαi1+3-DKO

Gαi1 KO

WT

KGF, 10′

Gαi3 KO

MEFs/KGFR

MEFs/KGFR Gαi-1+3-DKO

WT PDGF

0′ 10′

0′ 10′

Gαi-1 Gαi-3

Gαi-3 p-Akt s473 Akt p-Erk1/2

p-Erk1/2

Erk1/2

Erk1/2

Gαi-2

p-Erk1/2

Akt

Erk1/2

β-Actin

β-Actin

p-Akt s473

Gαi-1 p-S6

p-FRS2

Akt

Gαi-2

S6 FRS2

0 p-Akt/Akt

p-S6/S6

KGF (50 ng ml–1), 10′

0.5

*

3 2 1

*

*

0 0 p-KGFR/ p-Akt/ p-Erk/ p-FRS2/ KGFR Akt Erk FRS2 –1

KGF (50 ng ml ), 2′

p-Akt/Akt

p-Erk/Erk

Relative intensity

#

P >0.05

P >0.05 P >0.05

FGF-10

0.2

#

P >0.05

WT Gαi1+3-DKO

bFGF

#

0.4

1

4

FGF-10

#

WT Gαi1+3-DKO

P >0.05

bFGF

0.6

* **

*

Relative intensity

0.8

5

**

Relative intensity

Relative intensity

1

WT Gαi3 SKO Gαi1 SKO Gαi1+3DKO

1.4 1.2 1 0.8 0.6 0.4 0.2 0

WT Gαi1+3-DKO

*

P >0.05

p-Akt/Akt

Gαi-2/Akt

PDGF, 10′

Figure 1. Gai1 and Gai3 are required for the activation of AKT-mTORC1 by keratinocyte growth factor (KGF) in mouse embryonic fibroblast (MEFs). (a) Wild type (WT) MEFs were infected with virus particles containing keratinocyte growth factor receptor (KGFR)-pSUPER-puro-retro vector for 2 days, followed by 1.0 mg ml  1 of puromycin selection to generate KGFR-expressing MEFs (termed MEFs/KGFR), and KGFR and glyceraldehyde-3-phosphate dehydrogenase mRNA expression was tested by reverse transcriptase–PCR. (b) KGFR, b-actin, and EGFR expression in MEFs, MEFs/KGFR (Clone-1/-2), and primary keratinocytes were tested by western blotting. (c–h) WT MEFs/KGFR, Gai1 or Gai3 single-knockout (SKO) MEFs/KGFR, and Gai1/3 double knockout (DKO)-MEFs/KGFR were treated with indicated KGF (50 ng ml  1), EGF (50 ng ml  1, 10 min), fibroblast growth factor (FGF)-10 (25 ng ml  1, 10 min), basic fibroblast growth factor (bFGF) (25 ng ml  1, 10 min), or platelet-derived growth factor (PDGF)-BB (25 ng ml  1, 10 min), and listed proteins were analyzed by western blotting. Indicated proteins or protein phosphorylation was quantified (for e–h; n ¼ 3, same for all blot quantifications). *Po0.05. #Po0.05 versus Gai1/3 DKO MEFs/KGFR group (e).

comparative in DKO MEFs/KGFR-expressing WT or mutant Gai3 (Figure 2c), further confirming that Gai proteins lie downstream of KGFR. Notably, knockout of Gai2 had no significant effect on KGF-induced phosphorylation of AKT-473S and AKT-308T

(Figure 2d). Pertussis toxin, which ADP-ribosylates Gai proteins to interfere with GPCR-dependent signaling, also had no effect on the AKT (473S) phosphorylation by KGF (Figure 2e). Significantly, bFGF-/FGF-10-induced, but not PDGF-BBinduced, AKT (473S) phosphorylation was inhibited by www.jidonline.org

183

Y-m Zhang et al. Gai1/3–Gab1 Complex Mediates KGF Signaling

Gai proteins are required for the activation of AKT and mTORC1 by KGF in primary cultured skin keratinocytes

Gai1/3 small interfering RNAs in MEFs/KGFR (Figure 2f). Our observations confirm that Gai1 and Gai3 proteins have a vital role in KGF-mediated activation of the AKT-mTORC1 signaling.

Using the lentiviral short hairpin RNA infection strategy, we successfully generated stable Gai1 and Gai3

Gαi-1+3-RNAi

KGF

sc-RNAi

sc-RNAi

Gαi-1+3-RNAi

MEFs/KGFR

KGF

DKO MEFs DKO MEFs/KGFR Plasmid

pKGFR

p-S6

p-KGFR

KGFR

S6

KGFR

p-Akt s473

p-4E-BP1

Gi3

Akt

p-Erk1/2

Erk1/2

p-GSK3β

Erk1/2

p-Akt s473

GSK3β

β-Actin

Akt

Gαi-1

p-FRS2

p-S6

Gαi-3

FRS2

S6

0′

WT-Gαi3 DN-Gαi3

10′

0′

10′

2 * P>0.05

2 * 1 0

p-S

p-A

k

Er rk/

p-E

tin -ac

p-Akt/Akt p-S6/S6 KGF (50 ng ml–1), 10′

β i3/

Gα

R GF p-K GFR /K

MEFs/KGFR

KGF (50 ng ml–1), 10′

5′

p-Akt/Akt

10′

Relative intensity

p-Akt t308 Akt Gαi-2

MEFs/KGFR

p-Akt s473

0.4 0.2 0

Gαi-1

Akt KGF, 5′

β-Actin

Journal of Investigative Dermatology (2015), Volume 135

1.2 0.8 0.4 0

p-Akt/Akt 2

*

1.5 1

P>0.05

*

0.5 0

KGF, 10′

sc-RNAi Gαi-1+3-RNAi

bFGF

Akt

DMSO PTX P>0.05

Relative intensity

p-Akt s473

Relative intensity

1.6 PTX

DMSO

p>0.05

0.6

Gαi-3

p-Akt/Akt

KGF PTX

DMSO

0.8

Gαi-1+3-RNAi

PDGF

Gαi2-KO 0′

WT Gαi2-KO

Ctrl

bFGF

MEFs/KGFR WT 0′ 5′ 10′

FGF-10

sc-RNAi

PDGF

6 6/S

*

FGF-10

kt

A kt/

184

10′

PDGF

*

WT-Gαi3 DN-Gαi3

bFGF

1

3

FGF-10

*

Relative intensity

sc-RNAi Gαi-1+3-RNAi

0

KGF p-Akt s473

0′

P>0.05

3 Relative intensity

YFP

KGF

Y-m Zhang et al. Gai1/3–Gab1 Complex Mediates KGF Signaling

double-knockdown keratinocytes. Results in Figure 3a showed that expression of Gai1 and Gai3 was significantly downregulated in the stable keratinocytes. Gai1/3 knockdown did not affect KGFR expression or KGF-induced KGFR phosphorylation in keratinocytes (Figure 3a). However, KGF-induced phosphorylation of AKT (473S) and S6 (235/236S) was largely inhibited by Gai1/3 stable knockdown in keratinocytes, although phosphorylation of Erk1/2 (202T/204Y) by KGF was not significantly affected (Figure 3a). Thus, Gai1/3 depletion only inhibited KGF-induced phosphorylation of AKT and S6,

5′ 10′ 0′

Anti-sense

5′ 10′

KGF

sc-RNAi

Gαi-1/3-RNAi

5′

5′

0′

10′ 0′

0’

10′

Gαi-1 Gαi-1

Gαi-3-GFP Gαi-3

Gαi-3

Gαi-1-GFP Gαi-1

Gαi-3 β-Actin

β-Actin

p-KGFR

p-Akt 473 p-Akt 473

p-Akt 473

Akt

Akt

p-S6

p-S6

Akt p-S6 S6

p-Erk1/2

2.5

3

Relative intensity

Relative intensity

Erk1/2 sc-shRNA P >0.05 Gαi-1/3-shRNAs

2

*

1

*

P >0.05

0 p-Akt /Akt

p-S6 p-KGFR /β-actin /KGFR

p-Erk /Erk

p-Akt 473 β-Actin

0.5

*

P >0.05

KGF Gαi-2 p-Akt s473

0.3 0.2

Akt

0.1

p-S6/S6

0.4 0 KGF (50 ng ml–1), 5′

sc-shRNA

0.4

p-Akt/Akt

0.8

p-Akt/Akt p-Erk/β-actin

Vector DN-Gαi3

*

1.2

KGF (50 ng ml–1), 5′

PDGF

0′

KGF

PDGF

*

1

Empty vector

Relative intensity

Akt

KGF

0′

2 1.5

*

1.6

0

KGF (50 ng ml–1), 5′

DN-Gαi-3

Vector Gαi-1/3

sc-siRNA Gαi-1/3-siRNAs P >0.05

Relative intensity

p-Erk1/2

0′

5′

10′

Gαi-2-shRNA 0′

5′

sc-shRNA Gαi-2-shRNA

10′ 2.5 Relative intensity

KGFR

DN-Gαi-3 Gαi-3

KGF Gαi-1/3

0′

Vector

KGF

Vector

Primary keratinocytes sc-shRNA Gαi-1/3-shRNA

shRNA

but not KGFR or Erk. Next, we dissected the individual role of Gai1 or Gai3 in KGF signaling in keratinocytes. Results showed that stable depletion of Gai1 or Gai3 by single lentiviral shRNA only alleviated KGF-induced AKT/S6 phosphorylation in primary keratinocytes, whereas combined knockdown of Gai1 and Gai3 showed an additive inhibitory effect on AKT/S6 phosphorylation (Supplementary Figure S2 online). Again, although both are required, Gai3 appears more important than Gai1 in mediating AKT-mTORC1 activation by KGF in primary keratinocytes (Supplementary Figure S2

*

2 1.5 1

P >0.05

0.5 0

0 KGF PDGF p-Akt/Akt

KGFR

p-Akt/Akt Gαi-2/KGFR KGF (50 ng ml–1), 5′

Figure 3. Gai1 and Gai3 proteins are required for keratinocyte growth factor (KGF)-induced AKT-mTOR complex 1 (mTORC1) activation in primary keratinocytes. (a, e) KGF (50 ng ml  1)-induced signaling in stable primary keratinocytes expressing scramble or Gai1/3 short hairpin RNAs or Gai2 short hairpin RNA was tested. (b) Primary keratinocytes were introduced with anti-sense RNAs against Gai1/3 or scramble RNA sequence, and KGF (50 ng ml  1)-induced signaling was tested in the above cells. (c) KGF (50 ng ml  1)-induced signaling in stable primary keratinocytes expressing empty vector or GFP-Tagged Gai1/3 was detected. (d) KGF (50 ng ml  1, 10 min) or platelet-derived growth factor (PDGF)-BB (25 ng ml  1, 10 min)-induced AKT activation in stable primary keratinocytes expressing empty vector or GFP-dominant-negative (DN)-Gai3 was shown. All protein phosphorylations in this figure were quantified. *Po0.05.

Figure 2. Gai1 and Gai3 proteins are vital for the phosphorylation of AKT (473 S) and S6 (235/236 S) in response to keratinocyte growth factor (KGF) in keratinocyte growth factor receptor (KGFR)-expressing mouse embryonic fibroblast (MEFs). (a, b, f) Wild type (WT) MEFs/KGFR transfected with scramble small interfering RNA or Gai1/3 small interfering RNAs were treated with KGF (50 ng ml  1), basic fibroblast growth factor (bFGF) (25 ng ml  1), fibroblast growth factor (FGF)-10 (25 ng ml  1), or platelet-derived growth factor (PDGF)-BB (25 ng ml  1) for 10 min, and western blotting was utilized to test indicated proteins. (c) Double knockout (DKO) MEFs/KGFR were transfected with the vector encoding WT or dominant-negative (DN) Gai3 (G202T; 2 mg each), and KGF (50 ng ml  1)-induced signaling in DKO MEFs (no KGFR, first two lanes) and above MEFs/KGFR (last four lanes) was tested. (d) KGF (50 ng ml  1)-induced AKT activation in WT and Gai2 KO MEFs/KGFR was treated. (e) WT MEFs were pretreated overnight with pertussis toxin (PTX, 100 ng ml  1) followed by indicated KGF treatment, and pAKT (473S) and AKT were detected by western blotting. Indicated proteins or protein phosphorylation was quantified. *Po0.05.

www.jidonline.org

185

Y-m Zhang et al. Gai1/3–Gab1 Complex Mediates KGF Signaling

KGF 0′

MEFs/KGFR Gab1-KO

5′ 10′ 20′

0′

Primary keratinocytes

d

5′ 10′ 20′ Gab1-shRNA

sc-shRNA

KGFR KGF 0′

5′ 10′ 20′ 0′

Gab1-shRNA

WT

sc-shRNA

a

5′ 10′ 20′

KGF 5′

p-Gab1 Gab1

Gab1 p-Akt

p-Akt

p-Akt s473 p-Akt t308

pFRS2

Akt

Gab1

FRS2 Akt

p-GSK

pERK1/2 p-S6

Gαi-1

Erk1/2

Gαi-3

Erk1/2

WT

Gab1-KO

5′ 10′ 20′

0′

5′ 10′ 20′

e

*

3 Relative intensity

KGF 0′ p-p85 p85

sc-shRNA Gab1-shRNA

2 1

P >0.05

P >0.05

p-FRS2/ FRS2

p-Erk/ Erk

0

*

WT Gab1 KO

*

KGF (50 ng ml–1), 5′ Primary keratinocytes sc-shRNA Gαi-1/3-shRNA KGF 0′ 5′ 10′ 0′ 5′ 10′

p-Akt/Akt p-S6/Erk p-p85/85 KGF (50 ng ml–1), 5′

Gαi-1

c MEFs/KGFR KGF

0′

Gαi-1+3-DKO

WT

5′ 10′ 20′ 0′

p-Gab1

5′ 10′ 20′

p-Gab1

g

Gab1

IP: KGFR

IP: Gαi-3 KGF

p85 Relative intensity

h Primary keratinocytes

p-p85

0′

2′

5′

KGF

Gαi-3

KGFR Gab1

Gab1

p-Gab1

1

KGFR

FRS-2

0.5

Nonspecific

IgG Gαi-3

2

*

WT Gαi-1+3-DKO

1.5

*

0 p-Gab1/Gab1

sc-shRNA Gαi-1/3-shRNA

*

1.5 1 0.5 0

p-Gab1/Gab1 KGF (50 ng ml–1), 5′

Gab1

KGFR

2

Inputs

Gαi-1/3-shRNA

f

0

sc-shRNA

*

1

Gαi-1/3-shRNA

2

p-Akt/ Akt

Relative intensity

Relative intensity

3

sc-shRNA

b

p-p85/85

KGF (50 ng ml–1), 5′ Figure 4. Gab1, lying downstream of Gai, is required for keratinocyte growth factor (KGF)-induced AKT activation. (a, b) KGF (50 ng ml  1)-induced signaling in keratinocyte growth factor receptor (KGFR)-expressing wild type (WT) or growth factor receptor binding 2-associated binding protein 1 (Gab1) KO mouse embryonic fibroblast (MEFs) was detected by western blotting. (c) KGFR-expressing WT and Gai1/3 double knockout (DKO) MEFs were treated with KGF (50 ng ml  1), and pGab1 (627Y), Gab1, p-p85 (458Y), p85, and KGFR were tested. (d, e) KGF (50 ngml  1)-induced signaling in stable primary keratinocytes expressing scramble or Gab1 short hairpin RNA was tested. (f) Stable primary keratinocytes expressing scramble or Gai1/3 short hairpin RNAs were treated with KGF (50 ng ml  1), and p-Gab1(627Y), Gab1, and Gai1 were tested. (g) Primary keratinocytes were treated with KGF (50 ngml  1) for the indicated time, and the association between Gai3, Gab1, and KGFR was tested by co-immunoprecipitation (Co-IP). (h) Stable primary keratinocytes expressing scramble or Gai1/3 short hairpin RNAs were treated with KGF (50 ngml  1) for 2 minutes, and expression and association of KGFR-FGF receptor substrate 2 (FRS-2)  Gai3-Gab1 were tested. All protein phosphorylations in this figure were quantified. *Po0.05.

186

Journal of Investigative Dermatology (2015), Volume 135

Y-m Zhang et al. Gai1/3–Gab1 Complex Mediates KGF Signaling

Gαi-1

Gαi-1 Wounded skin

Surrounding healthy skin

Gαi-1 Surrounding healthy skin

Gαi-3

Gαi-1

Surrounding healthy skin

Gαi-3 Keloid skin

Gαi-3

Surrounding healthy skin

Wounded skin

Gαi-3 Keloid skin

Figure 5. Increased expression of Gai1/3 in human wounded skin and keloid skin tissues. (a, b) Gai1/Gai3 immunohistochemistry of the granulation tissue of healing skin or surrounding healthy skin from the same patient (male, 35 years old, multiple skin wounds developing for 1 month). (c, d) Gai1/Gai3 immunohistochemistry of the skin keloid tissue and surrounding healthy skin tissue from the same patient (male, 27 years old, keloids develop for 4 months). Similar results were seen in three independent patients. Bar ¼ 100 mm.

online, quantification). To further confirm the role of Gai1 and Gai3 in KGFR signaling, the target small interfering RNAs against Gai1 and Gai3 (targeting nonoverlapping mRNA sequence with the plasmids) were utilized. These anti-sense small interfering RNAs inhibited Gai1 and Gai3 expression (460%) in primary keratinocytes. Further, KGF-induced phosphorylation of AKT (473S) and S6 (235/236S), but not Erk1/2, was markedly inhibited in Gai1/3-silenced cells (Figure 3b). In addition, we established a stable keratinocyte line expressing WT-GFP-tagged Gai3 and WT-GFP-tagged Gai1. KGF-induced AKT (473S) and S6 (235/236S) phosphorylation was enhanced by the exogenous expression of wt-Gai1/3 (Figure 3c). On the other hand, overexpressing of the DN mutant Gai3 inhibited KGF-induced AKT (473S) phosphorylation, whereas AKT (473S) phosphorylation by PDGF-BB was not affected (Figure 3d). These results further confirmed that Gai1/3 proteins are required for the activation of AKT and mTORC1 by KGF in keratinocytes. RNAi knocking down of Gai2 had no significant effect on KGF-induced AKT activation in keratinocytes (Figure 3e). Gab1, lying downstream of Gai proteins, mediates KGF-induced PI3K-AKT-mTORC1 activation

Upon RTK activation, Gab1 is enrolled in the plasma membrane, being phosphorylated, and works as a scaffold to mediate signal transduction from the receptor to downstream events (Holgado-Madruga et al., 1996). We discovered that Gab1 is associated with Gai proteins in response to EGFR activation to transduce signaling from EGFR-Gai to p85 (Cao et al., 2009). The role of Gab1 in KGFR signaling has not been

studied. As demonstrated, KGF induced Gab1 phosphorylation in MEFs/KGFR. Gab1 deficiency severely impaired KGF-induced phosphorylation of AKT (473S), GSK-3b (9S), and S6 (235/236S) in MEFs/KGFR (Figure 4a, upper panel and Figure 4b), indicating that Gab1 is required for KGF-induced AKT-mTORC1 activation. Importantly, KGF-induced p85 phosphorylation was also inhibited with Gab1 deficiency (Figure 4a, lower panel and Figure 4b), indicating that p85-PI3K activation by KGF requires Gab1. To sort out the relationship between Gab1 and Gai in mediating PI3K-AKT activation by KGF, we once again utilized Gai1/3-deficient cells. KGF-induced phosphorylation of Gab1 (627Y) and p85 (458Y) was almost abolished in Gai1/3 DKO MEFs/KGFR (Figure 4c). In primary keratinocytes, Gab1 depletion by shRNA inhibited KGF-induced AKT phosphorylation, whereas Gai1/3 protein expression was unchanged (Figure 4d, left panel and Figure 4e). Significantly, KGF-induced Erk1/2 and FRS-2 phosphorylation was not affected by Gab1 shRNA knockdown in primary keratinocytes (Figure 4d, right panel and Figure 4e). In primary keratinocytes, stable knockdown of Gai1/3 inhibited KGF-induced Gab1 phosphorylation (Figure 4f). These results together demonstrate that Gai1/3 proteins are upstream signals for Gab1 in KGFR signaling. Co-immunoprecipitation results of primary keratinocytes showed that KGFR, Gab1, and Gai3 formed a signaling complex after KGF stimulation (Figure 4g). More importantly, Gai1/3 depletion by targeted shRNAs largely inhibited KGF-induced KGFR-Gab1 association and subsequent Gab1 phosphorylation, but leaving KGFR-FRS-2 association unaffected (Figure 4h). Expression of KGFR, Gab1, and FRS-2 was not affected by Gai1/3 depletion (Figure 4h, www.jidonline.org

187

Y-m Zhang et al.

Primary keratinocytes BrdU incorporation sc-shRNA

1.6

Gi1/3-shRNA

*

sc-shRNA

*

1.4

3

P >0.05

Gi1/3-shRNA

*

1.2

Fold vs. Ctrl

Fold vs. Ctrl

[H3]Thymidine P >0.05

1.0 0.8 0.6

KGF

* 2

Gαi-1

1

Gαi-3

sc-shRNA

1.8

Gi1/3-shRNA

Gai1/3–Gab1 Complex Mediates KGF Signaling

0.4 0.2

β-Actin

0

0 Ctrl

25 ng ml–1

50 ng ml–1

25 ng ml–1

KGF

Ctrl

25 ng ml–1

PDGF

50 ng ml–1 KGF

25 ng ml–1 PDGF

Migrated cells

1

Gαi-1

0.5

Gαi-3

KGF

Ctrl

KGF

KGF

Fold vs. Ctrl

*

1.5

sc-shRNA Gαi-1/3-shRNA Ctrl

*

2

Gi1/3-shRNA

P >0.05 sc-shRNA

2.5

sc-shRNA Gi1/3-shRNA

Cyclin D1 Fibronectin

0 Ctrl

50 ng ml–1

25 ng ml–1

KGF

β-Actin

β-Actin

25 ng ml–1 PDGF

KGF KGFR

P

P P

P FRS2

FRS2 ? Gi1/3

P ?

Gab1

P

Erk mTORC2

473s 308T

P Akt

P

mTORC1 P P S6K1

4E-BP1

P S6

Proliferation Migration

Figure 6. Gai proteins are important for keratinocyte growth factor (KGF)-dependent cell growth and migration in keratinocytes. Stable primary keratinocytes expressing scramble or Gai1/3 short hairpin RNAs were treated with KGF (25/50 ngml  1) or platelet-derived growth factor (PDGF)-BB (25 ngml  1) for 24 hours, and cell proliferation and migration were detected by the BrdU incorporation assay (a), the /[H3] Thymidine DNA assay (b), and the ‘‘Transwell’’ assay (d), and expression of Gai1, Gai3, and b-actin (c and e) as well as of cyclin D1 and fibronectin (f) in the above cells was tested by western blotting. (g) The proposed signaling pathway of this study (see Discussion section). *Po0.05 (a, b, d).

input). Consistent with these findings, Wang et al. (2014) reported that bFGF induced Gai3 association with activated FGF receptor and Gab1, whereas FRS-2-growth factor receptor binding 2 association by bFGF was not affected by 188

Journal of Investigative Dermatology (2015), Volume 135

Gai1/3 deficiency (Wang et al., 2014). On the basis of these data, we propose that Gab1, lying downstream of FRS-2-Gai proteins but upstream of PI3K-p85, mediates KGF-induced AKT-mTORC1 activation.

Y-m Zhang et al. Gai1/3–Gab1 Complex Mediates KGF Signaling

Increased expression of Gai1/3 in human wounded skin and human keloid skin tissues

The above results suggest that Gai1/3 proteins are required for KGF-induced AKT-mTORC1 activation in skin keratinocytes. Next, we examined the expression of these two proteins in different human skin tissues. Western blotting (Supplementary Figure 3 S3 online) and immunohistochemistry (Figure 5a and b) results showed that Gai1/3 expression was higher in wounded human skin than in the surrounding normal skin tissues, indicating a functional role of Gai proteins in wound healing. Interestingly, Gab1 expression in the above-mentioned skin tissues was comparable (Supplementary Figure 3 S3 online). Skin keloid is characterized by excess accumulation of collagen and hyperactivity of proliferation/migration of skin cells within the wound (Singer and Clark, 1999). In these conditions, hyperactivities of inflammations, synthesis and secretion of extracellular matrix proteins (i.e., cytokines and growth factors), and tissue remodeling have all been suggested as pathological causes (Singer and Clark, 1999). As shown in Figure 5c and d, a significant increase in Gai1 and Gai3 expression in keratinocytes of keloid skin was observed, whereas expression of these two proteins was relatively low in surrounding healthy skin tissues from the same individual. Importantly, the expression of Gai1/3 in skin fibroblasts was also higher in keloid skin than in surrounding non-lesional skin (Figure 5c and d). Gai proteins are important for KGF-dependent cell growth and migration in keratinocytes

We next explored the biological role of Gai proteins in the KGF signaling at the physiological level. We found that KGFinduced cell proliferation was greatly alleviated in stable keratinocytes expressing Gai-1/3 shRNA, and keratinocyte proliferation was examined by BrdU incorporation assay (Figure 6a) and the [H3] Thymidine assay (Figure 6b), and expression of Gai1/3 and b-actin in stable cells is shown in Figure 6c. In contrast, PDGF-BB-induced cell proliferation was not affected by Gai1/3 depletions (Figure 6a and b). In addition, stable knocking down of Gai1/3 by targeted shRNAs largely inhibited KGF-induced keratinocyte migration (Figure 6d), which was detected by the ‘‘Transwell’’ assay. PDGF-BB-induced keratinocyte migration was independent of Gai1/3 proteins (Figure 6d and e). Importantly, KGF-induced upregulation of proliferation-associated protein cyclin D1 and migration-associated protein fibronectin (Figure 6f) was almost overturned by Gai1/3 lentiviral shRNA depletion in keratinocytes (Figure 6f). Taken together, our results demonstrate that Gai proteins are critical for KGF-induced cyclin D1/fibronectin expression, cell proliferation, and migration in primary keratinocytes. DISCUSSION Studies have demonstrated a critical role of KGF-KGFR in the wound healing process. The transgenic mice with a DN KGFR mutation suffered a severe delay in wound healing (Werner et al., 1994). Significant upregulation of KGF was seen in mouse and human skin following skin wounds (Werner et al., 1992; Beer et al., 2000). In the animal models of wound

healing, exogenous KGF markedly promoted wound reepithelialization (Braun et al., 2002). In vitro, significant cell migration and proliferation were seen in keratinocytes after KGF stimulation, which was alleviated by the monoclonal neutralizing antibody against KGFR (Tsuboi et al., 1993; Danilenko et al., 1995). KGF-KGFR binding induces receptor dimerization on the cell surface with auto-phosphorylation of tyrosine residues located in the cytosol (Shaoul et al., 1995; Heldin and Ostman, 1996; Ceridono et al., 2005; Belleudi et al., 2011a, 2011b). Phosphorylated KGFR provides docking sites for Src homology 2 domain–containing proteins, including Shc, growth factor receptor binding 2, FRS-2, and PLC-g, to mediate signal transduction (Heldin and Ostman, 1996). Our IP data demonstrated that Gab1 and Gai proteins were recruited to KGFR after KGF stimulation and were required for activation of downstream PI3K-AKT-mTORC1 signaling. Cells with Gab1 deficiency showed defective PI3K-AKT-mTORC1 activation in response to KGF. Our data suggest that Gab1, lying downstream of Gai proteins, mediates KGF-stimulated PI3K-AKT-mTORC1 signaling transduction (Figure 6g). Our results provide genetic evidence that Gai proteins are critical for activation of the AKT-mTORC1 signaling by KGF. Although we found that KGFR signaling and G proteins are intertwined in the regulation of AKT-mTORC1 activation, little is known about the precise molecular mechanisms that regulate this cross talk. Growth factor receptors or RTKs are known to activate G proteins through functional cross-talk termed trans-activation (Marty and Ye, 2010). In that case, activated RTKs trans-activate GPCR signaling through physical association with Gai1/3 proteins, or through upregulation of GPCR ligands (see discussion in Marty and Ye, (2010)). This is unlikely the case here, as pertussis toxin, which ADP-ribosylates the Gai family of proteins and uncouples these G proteins from their GPCR receptors, had no effect on KGF-induced AKT activation. The present study reveals a unique mechanism for the role of Gai proteins in KGFR signaling. We propose that, upon KGF stimulation, Gai1 and Gai3 act downstream of KGFR-FRS-2 to transduce KGFR signaling to Gab1, leading to Gab1 activation and downstream p85-AKT-mTORC1 activation (Figure 6g). Interestingly, loss of Gai2 did not result in a clear defect in KGF-mediated AKT activation; this could be explained by the fact that the expression of Gai1/3 in Gai2-deficient MEFs is higher than that in WT cells ( see Cao et al., 2009). Interestingly, mice with KGF knockout showed normal wound repair (Guo et al., 1996), indicating that other members of the KGFR family, i.e., FGF-10, could compensate the effect by KGF (Igarashi et al., 1998; Jameson et al., 2002). In the current study, we found that FGF-10- and bFGF-induced AKT and/or mTORC1 activation was also dependent on Gai proteins. Significantly, loss of Gai1 and Gai3 largely impaired KGFinduced keratinocyte migration and proliferation. Because PI3K-AKT-mTORC1 signaling is important for cell migration and proliferation, this observation may be explained by a failure of KGF to trigger the activation of this pathway in the Gai-depleted cells. Interestingly, we found that keratinocytes with Gai deficiency did not exhibit a defect in www.jidonline.org

189

Y-m Zhang et al. Gai1/3–Gab1 Complex Mediates KGF Signaling

phosphorylation of Erk by KGF. Further, KGF-induced Erk activation was intact in Gab1-deficient cells. Thus, the Gab1/ Gai signaling complex is not necessary for Erk activation by KGF (Figure 6g). As a matter of fact, studies have shown that KGFR-FRS-2 phosphorylation is required for Erk1/2 activation by KGF (Ceridono et al., 2005). We found that KGF-stimulated KGFR-FRS-2 phosphorylation and association were not affected by Gai1/3 deficiency. Further, Wang et al. (2014) showed that FRS-2-growth factor receptor binding 2 association, which mediates bFGF-induced Erk1/2 activation (Kouhara et al., 1997), was also intact in Gai1/3 DKO MEFs (Wang et al., 2014). In summary, our results demonstrate that Gai1/3 proteins lie downstream of KGFR, but upstream of Gab1-mediated activation of PI3K-AKT-mTORC1 signaling, thus revealing a role for Gai proteins in mediating KGFR signaling, cell migration, proliferation, and possible wound healing. MATERIALS AND METHODS The study was approved by the institutional review board of all authors’ institutions. All clinical investigations were conducted according to the principles expressed in the declaration of Helsinki.

Materials Pertussis toxin was purchased from Sigma (Shanghai, China). KGF (mouse and human), FGF-10, bFGF, EGF, and PDGF-BB were purchased from Calbiochem (Shanghai, China).

Cell culture WT, Gai1 or Gai3 single-knockout MEFs, Gai1 and Gai3 (Gai1/3) DKO MEFs, as well as WT/Gab1-deficient MEFs were described previously (Cao et al., 2009). Primary cultures of human keratinocytes were gifts from Dr Bi’s Lab (Ji et al., 2012) and maintained in Medium 154-CF (Cascade Biologics, Portland, OR) supplemented with Human Keratinocyte Growth Supplement (Cascade Biologics) plus antibiotics and Ca2 þ (0.03 mM; Cascade Biologics Invitrogen, Shanghai, China). Cells were starved in 0.5% FBS medium overnight before any treatment/s.

Exogenous expression of KGFR in MEFs A complementary DNA encoding full-length human KGFR (from primary keratinocytes; synthesized by GeneChem, Shanghai, China) was sub-cloned into the pSUPER-puro-retro vector (Promega, Madison, WI), which was then transfected into HEK-293 cells with plasmids encoding viral packaging proteins VSVG and Hit-60 (Promega; Zecevic et al., 2009) using Lipofectamine 2000 (Invitrogen) reagent according to the manufacturer’s instructions. Two days post transfection, the medium containing virus particles was added to MEFs (both WT and KO MEFs), and infections were allowed to proceed for 24 hours. The vector-expressing cells were selected 10 days post infection in the presence of 1.0 mg ml  1 of puromycin. Expression of KGFR in the transfected MEFs (MEFs/KGFR) was tested by western blotting and reverse transcriptase–PCR.

Generation of Gai1 and Gai3 stable knockdown keratinocytes by lentiviral infection Primary cultured skin keratinocytes were seeded in a six-well plate with 60% confluence in a growth medium with polybrene. A quqntity 190

Journal of Investigative Dermatology (2015), Volume 135

of 20 ml ml  1 of lentiviral particles containing shRNA of human Gai1 (sc-105382-V) and/or human Gai3 (sc-29325-V; Santa Cruz Biotech, Santa Cruz, CA) were added to the cells for 24 hours; the cell culture medium was then replaced by growth medium, and cells were cultured for an additional 24 hours. The stable clones expressing target-shRNA/s were selected with puromycin (0.25 mg ml  1, Sigma). The culture medium was replaced with fresh puromycin-containing culture medium every 2–3 days, until resistant colonies could be identified (12–14 days). The expression of Gai1 and Gai3 protein in stable cells was detected by western blotting. The same amount of scramble nonsense shRNA lentiviral particles (Santa Cruz) was added to the control cells. The Gab1 or Gai2 stable knockdown keratinocytes were generated similarly.

Gai plasmid transit transfection in MEFs The expression vectors encoding EE-tagged Gai3 and EE-tagged DN mutant of Gai3 (G202T) were described previously (Cao et al., 2009). Plasmids were transfected with Lipofectamine 2000 (Invitrogen, Carlsbad, CA) according to the manufacturer’s recommendation. After 24 hours, cells were starved before the indicated treatments.

Stable expression of Gai proteins in primary cultured keratinocytes The plasmid (0.25 mg ml  1) encoding wt-Gai1 (Gai1-GFP-puro), wt- Gai3 (Gai3-GFP-puro), and DN Gai3 (G202T; dominant-negative-Gai3-GFP-puro), were obtained from GeneChem (Shanghai, China), or the empty vector (pGCL-GFP-puro, GeneChem) was transfected to primary cultured skin keratinocytes with the Lipofectamine 2000 as described. After 24 hours, cells were switched back to culture medium for an additional 48 hours. Stable clones expressing target plasmid were selected using puromycin (0.25 mg ml  1) for 12– 14 days. The target protein (Gai-GFP) in stable cells was tested by western blotting. Western blotting, antibodies used, the co-Immunoprecipitation assay,RNAi, the ‘‘Transwell’’ migration assay, reverse transcriptase– PCR, and statistical analysis have been reported in Cao et al. (2009, 2013) and Chen et al. (2013), and are described in Supplementary Information S1 online. The [H3] Thymidine incorporation assay and the BrdU incorporation assay as well as immunohistochemistry staining are also described in Supplementary Information S1 online. CONFLICT OF INTEREST The funders had no role in the study design, data collection and analysis, decision to publish, or preparation of the manuscript. The authors state no conflict of interest.

ACKNOWLEDGMENTS This work was funded by a grant from the National Natural Science Foundation of China (81101436). This project was funded by the Priority Academic Program Development of Jiangsu Higher Education Institutions. SUPPLEMENTARY MATERIAL Supplementary material is linked to the online version of the paper at http:// www.nature.com/jid

REFERENCES Alessi DR, James SR, Downes CP et al. (1997) Characterization of a 3-phosphoinositide-dependent protein kinase which phosphorylates and activates protein kinase Balpha. Curr Biol 7:261–9

Y-m Zhang et al. Gai1/3–Gab1 Complex Mediates KGF Signaling

Barren B, Artemyev NO (2007) Mechanisms of dominant negative G-protein alpha subunits. J Neurosci Res 85:3505–14

Hubbard KB, Hepler JR (2006) Cell signalling diversity of the Gqalpha family of heterotrimeric G proteins. Cell Signal 18:135–50

Beer HD, Gassmann MG, Munz B et al. (2000) Expression and function of keratinocyte growth factor and activin in skin morphogenesis and cutaneous wound repair. J Investig Dermatol Symp Proc 5:34–9

Igarashi M, Finch PW, Aaronson SA (1998) Characterization of recombinant human fibroblast growth factor (FGF)-10 reveals functional similarities with keratinocyte growth factor (FGF-7). J Biol Chem 273:13230–5

Belleudi F, Purpura V, Scrofani C et al. (2011a) Expression and signaling of the tyrosine kinase FGFR2b/KGFR regulates phagocytosis and melanosome uptake in human keratinocytes. FASEB J 25:170–81

Jameson J, Ugarte K, Chen N et al. (2002) A role for skin gammadelta T cells in wound repair. Science 296:747–9

Belleudi F, Purpura V, Torrisi MR (2011b) The receptor tyrosine kinase FGFR2b/KGFR controls early differentiation of human keratinocytes. PLoS One 6:e24194 Braun S, Hanselmann C, Gassmann MG et al. (2002) Nrf2 transcription factor, a novel target of keratinocyte growth factor action which regulates gene expression and inflammation in the healing skin wound. Mol Cell Biol 22:5492–505

Ji C, Yang YL, Yang Z et al. (2012) Perifosine sensitizes UVB-induced apoptosis in skin cells: new implication of skin cancer prevention? Cell Signal 24:1781–9 Kouhara H, Hadari YR, Spivak-Kroizman T et al. (1997) A lipid-anchored Grb2-binding protein that links FGF-receptor activation to the Ras/MAPK signaling pathway. Cell 89:693–702 Marty C, Ye RD (2010) Heterotrimeric G protein signaling outside the realm of seven transmembrane domain receptors. Mol Pharmacol 78:12–8

Braun S, Krampert M, Bodo E et al. (2006) Keratinocyte growth factor protects epidermis and hair follicles from cell death induced by UV irradiation, chemotherapeutic or cytotoxic agents. J Cell Sci 119:4841–9

Miki T, Fleming TP, Bottaro DP et al. (1991) Expression cDNA cloning of the KGF receptor by creation of a transforming autocrine loop. Science 251:72–5

Cao C, Huang X, Han Y et al. (2009) Galpha(i1) and Galpha(i3) are required for epidermal growth factor-mediated activation of the Akt-mTORC1 pathway. Sci Signal 2:ra17

Shaoul E, Reich-Slotky R, Berman B et al. (1995) Fibroblast growth factor receptors display both common and distinct signaling pathways. Oncogene 10:1553–61

Cao C, Rioult-Pedotti MS, Migani P et al. (2013) Impairment of TrkB-PSD-95 signaling in Angelman syndrome. PLoS Biol 11:e1001478

Singer AJ, Clark RA (1999) Cutaneous wound healing. N Engl J Med 341: 738–46

Ceridono M, Belleudi F, Ceccarelli S et al. (2005) Tyrosine 769 of the keratinocyte growth factor receptor is required for receptor signaling but not endocytosis. Biochem Biophys Res Commun 327:523–32

Tsuboi R, Sato C, Kurita Y et al. (1993) Keratinocyte growth factor (FGF-7) stimulates migration and plasminogen activator activity of normal human keratinocytes. J Invest Dermatol 101:49–53

Chen MB, Wei MX, Han JY et al. (2013) MicroRNA-451 regulates AMPK/ mTORC1 signaling and fascin1 expression in HT-29 colorectal cancer cells. Cell Signal 26:102–9

Wang Z, Dela Cruz R, Ji F et al. (2014) G(i)alpha proteins exhibit functional differences in the activation of ERK1/2, Akt and mTORC1 by growth factors in normal and breast cancer cells. Cell Commun Signal 12:10

Danilenko DM, Ring BD, Yanagihara D et al. (1995) Keratinocyte growth factor is an important endogenous mediator of hair follicle growth, development, and differentiation. Normalization of the nu/nu follicular differentiation defect and amelioration of chemotherapy-induced alopecia. Am J Pathol 147:145–54

Werner S (1998) Keratinocyte growth factor: a unique player in epithelial repair processes. Cytokine Growth Factor Rev 9:153–65

Guo L, Degenstein L, Fuchs E (1996) Keratinocyte growth factor is required for hair development but not for wound healing. Genes Dev 10:165–75 Heldin CH, Ostman A (1996) Ligand-induced dimerization of growth factor receptors: variations on the theme. Cytokine Growth Factor Rev 7:3–10 Holgado-Madruga M, Emlet DR, Moscatello DK et al. (1996) A Grb2-associated docking protein in EGF- and insulin-receptor signalling. Nature 379:560–4

Werner S, Peters KG, Longaker MT et al. (1992) Large induction of keratinocyte growth factor expression in the dermis during wound healing. Proc Natl Acad Sci USA 89:6896–900 Werner S, Smola H, Liao X et al. (1994) The function of KGF in morphogenesis of epithelium and reepithelialization of wounds. Science 266: 819–22 Zecevic A, Menard H, Gurel V et al. (2009) WRN helicase promotes repair of DNA double-strand breaks caused by aberrant mismatch repair of chromium-DNA adducts. Cell Cycle 8:2769–78

www.jidonline.org

191