Shb Gene Knockdown Increases the Susceptibilityof SVR Endothelial Tumor Cells to Apoptotic StimuliIn Vitro and In VivoNina S. Funa1, Kalpana Reddy2, Sulochana Bhandarkar2, Elena V. Kurenova3, Lily Yang4, William G. Cance3,Michael Welsh1 and Jack L. Arbiser2

The Shb adapter protein is an Src homology 2-domain containing signaling intermediate operating downstreamof several tyrosine kinase receptors, including vascular endothelial growth factor receptor-2. Shb ismultifunctional and apoptosis is one response that Shb regulates. Inhibition of angiogenesis can be used incancer therapy, and one way to achieve this is by inducing endothelial cell apoptosis. The angiosarcoma cell lineSVR is of endothelial origin and can be used as a tool for studying in vivo inhibition of angiogenesis, and wethus employed an Shb-knockdown strategy using an inducible lentiviral system to reduce Shb levels in SVRcells and to study their responses. Shb knockdown increases the susceptibility of SVR cells to the apoptoticagents, cisplatin and staurosporine. Simultaneously, Shb knockdown causes reduced focal adhesion kinase(FAK) activation, monitored as phosphorylation of the regulatory residues tyrosines 576/577. No detectableeffects on Akt or extracellular signal-regulated kinase activity were noted. The altered FAK activity coincidedwith an elongated cell phenotype that was particularly noticeable in the presence of staurosporine. In order torelate the effects of Shb knockdown to in vivo tumorigenicity, cells were exposed to the angiogenesis inhibitorhonokiol, and again the cells with reduced Shb content exhibited increased apoptosis. Tumor growth in vivowas strongly reduced in the Shb-knockdown cells upon honokiol treatment. It is concluded that Shb regulatesapoptosis and cell shape in tumor endothelial cells via FAK, and that Shb is a potential target for inhibition ofangiogenesis.

Journal of Investigative Dermatology (2008) 128, 710–716; doi:10.1038/sj.jid.5701057; published online 4 October 2007

INTRODUCTIONInhibition of angiogenesis comprises a viable strategy forcancer treatment. Angiogenesis is a multi-step processinvolving endothelial cell proliferation, survival, migrationand altered cell shape and inhibition of any of theseindividual components may result in reduced vasculariza-tion. The angiosarcoma SVR cells are of endothelial originand can be employed to screen for inhibitors of angiogenesis.One compound with antiangiogenic and antitumor propertiesis honokiol (Battle et al., 2005; Ishitsuka et al., 2005), whichhas been found to cause apoptosis both in vitro and in vivo in

SVR cells and thus can cause tumor reduction upon in vivoadministration. Honokiol was found to inhibit phosphoryl-ation of vascular endothelial growth factor receptor-2(VEGFR-2) and block downstream signaling in SVR cells(Bai et al., 2003).

Shb is an Src homology 2 domain containing adapterprotein operating downstream of several tyrosine kinasereceptors (Anneren et al., 2003). The N-terminus containsproline-rich motifs, followed by a phosphotyrosine-bindingdomain, four putative tyrosine phosphorylation sites and aC-terminal Src homology 2 domain. The proline-rich motifsbind SH3 domain proteins such as Src (Anneren et al., 2003)and c-Abl (Hagerkvist et al., 2007), the phosphotyrosinebinding domain binds focal adhesion kinase (FAK), thetyrosine phosphorylated residues bind Crk (Anneren et al.,2003) and c-Abl (Hagerkvist et al., 2007) and the Srchomology 2 domain receptors such as fibroblast growthfactor receptor-1 (Anneren et al., 2003) and VEGFR-2(Holmqvist et al., 2004). Shb generates signaling complexesin response to tyrosine kinase receptor activation, and thebiological consequences of these events differ betweendifferent cells. One prominent response to Shb overexpres-sion is increased apoptosis, which has been observed infibroblasts, beta cells and endothelial cells (Anneren et al.,2003). Particularly, in response to the angiogenesis inhibitors

ORIGINAL ARTICLE

710 Journal of Investigative Dermatology (2008), Volume 128 & 2007 The Society for Investigative Dermatology

Received 18 May 2007; revised 13 June 2007; accepted 16 June 2007;published online 4 October 2007

1Department of Medical Cell Biology, Uppsala University, Uppsala, Sweden;2Winship Cancer Institute, Department of Dermatology, Emory UniversitySchool of Medicine, Atlanta, Georgia, USA; 3Department of Surgery,University of Florida School of Medicine, Gainesville, Florida, USA and4Department of Surgery and Winship Cancer Institute, Emory UniversitySchool of Medicine, Atlanta, Georgia, USA

Correspondence: Professor Jack L. Arbiser, Department of Dermatology,Winship Cancer Institute, Emory University School of Medicine, WMB 5309,101 Woodruff Circle, Atlanta, Georgia 30322, USA.E-mail: jarbise@emory.edu

Abbreviations: ERK, extracellular signal-regulated kinase; FAK, focal adhesionkinase; FRNK, FAK-related non-kinase; shRNA, short hairpin RNA; VEGFR,vascular endothelial growth factor receptor

angiostatin and endostatin, apoptosis was increased inShb-overexpressing endothelial cells (Dixelius et al., 2000).Although numerous studies have demonstrated apoptosis as aconsequence of Shb overexpression, less is known about theeffects of reduced Shb content on apopotosis. In one recentstudy, the Shb content was suppressed by short hairpin RNA(shRNA)-knockdown in an insulin-producing cell line, andthis caused a reduction in c-Abl activity and cell death uponexposure to tunicamycin and cisplatin (Siddik 2002; Hagerk-vist et al., 2007).

To address the role of Shb knockdown on apoptosis inother cell system, and to relate this to tumor treatment byinhibition of angiogenesis, we generated SVR cells withreduced Shb content. We note increased apoptosis inresponse to several stress-inducing agents, as well as reducedFAK activity and altered cell shape. In vivo treatment of Shb-depleted SVR cells with honokiol potentiated the antitumoreffect and resulted in reduced tumor growth.

RESULTSShb knockdownWe have recently described a system for cre-dependentknockdown of Shb in murine cells (Hagerkvist et al., 2007).Employing this system on the angiosarcoma cell line, SVR ofendothelial origin yielded efficient Shb knockdown (Figure 1).The SVR cells were infected with the lentivirus Sico-Shb witha sequence inserted that produces an shRNA that recognizesthe Shb mRNA upon cre-dependent recombination. Thesecells were denoted control, as knockdown had not beeninitiated in these. Subsequent infection with adenovirus-creinduces recombination of the integrated lentivirus and

activation of Shb shRNA (Ventura et al., 2004; Hagerkvistet al., 2007). This results in a significant reduction of the Shbprotein content (Figure 1a). The Shb protein was tested in amass population of adenovirus-cre-treated cells (Shb shRNA)or in a subclone from the adenovirus-cre-treated cells (ShbshRNA-1), and both showed a reduced Shb protein content.

Cell death

The cells were used to examine their apoptotic responses tovarious stress agents (Figure 1b). The cell death rates of thethree cell lines (control, Shb shRNA, and Shb shRNA-1) weresimilar under basal conditions. However, addition ofstaurosporine, an agent that induces apoptosis in many celltypes due to serine/threonine kinase inhibition (Kabir et al.,2002), caused a significant increase in apoptosis of both themass population of Shb-knockdown cells (Shb shRNA) andthe knockdown clone (Shb shRNA-1). This increase inapoptosis was apparent after both 18 and 40 hours exposureto the drug.

The cells were also exposed to cisplatin, an agent thatcauses apoptosis due to DNA damage (Siddik 2002). Bothgroups of Shb-knockdown cells exhibited increased apoptosisafter 18 hours exposure to cisplatin (Figure 1b). However, thisdifference disappeared at 40 hours, suggesting that loss of Shbprotein accelerates the apoptotic response, but in casedamage is irreparable, apoptosis will eventually ensue.

FAK activity

We decided to search for changes in signaling occurring as aconsequence of Shb knockdown in SVR cells. As Shb haspreviously been shown to affect FAK activity (Holmqvistet al., 2003, 2004), FAK-activity was analyzed in the presentcontext. A phosphospecific antibody that recognizes phos-phorylated FAK at tyrosines 576/577 was used to monitorFAK activity. These phosphorylation sites are present in aregulatory region of the catalytic domain of FAK and arethought to reflect activation of FAK kinase (Withers et al.,1996; Ilic et al., 1998, 2001). Analysis of FAK phosphoryla-tion at these sites revealed less phosphorylation in the Shb-knockdown cells, regardless of whether these were main-tained in 0.1% serum, in 10% serum, or after incubation withcisplatin (Figure 2a). The level of FAK phosphorylation wassomewhat higher in the presence of 10% serum than in 0.1%serum, although the reduction induced by Shb knockdownwas still maintained. After exposure to staurosporine, themost dramatic effect was noted. The combination stauros-porine and Shb knockdown reduced FAK phosphorylation toa level that was barely detectable (Figure 2a). It is thusconcluded that Shb regulates FAK activity also in SVR cells.

We also studied Akt and extracellular signal-regulatedkinase (ERK) activity in the Shb-knockdown SVR cells usingphosphospecific antibodies and noted no differences in theactivities of these between the control or the cells withreduced Shb protein content (Figure 2b). The activity of c-Ablassessed as phosphorylation of regulatory tyrosines 245 or418 could not be detected in the SVR cells (results notshown). We conclude that the main consequences of reducedShb content in SVR cells involve FAK signaling.

Cre:

Cell death induced by staurosporine

908070

70

60

60

5040302010

0

5040302010

00 hour 18 hours 40 hours 0 hour 18 hours 40 hours

Cell death induced by cisplatin

– (ctrl)

+ (Shb shRNA)

+ (Shb shRNA-1)

– Shb

– ERK

Ctrl Shb shRNA

Shb shRNA-1

Ctrl Shb shRNA

Shb shRNA-1

Figure 1. Effect of Shb-knockdown on cell death in response to staurosporine

and cisplatin. (a) SVR cells were infected with the lentivirus containing the

Shb shRNA sequence (control, cre �) and subjected to a second infection

with adenovirus containing cre (cre þ ) to initiate Shb knockdown

(Shb shRNA). A subclone from the mass population of Shb shRNA cells was

isolated (Shb shRNA-1, cre þ ). Lysates were prepared and tested for Shb

protein content by Western blotting. ERK is shown as loading control.

(b) Control, Shb shRNA, and Shb shRNA-1 cells were plated and exposed to

1mM staurosporine for the time periods indicated. Percent cell death after

staining for propidium iodide and FACS analysis is shown. n¼ 8. (c) Control,

Shb shRNA, and Shb shRNA-1 cells were plated and exposed to 10mM

cisplatin for the time periods indicated. Percent cell death after staining for

propidium iodide and FACS analysis is shown. n¼ 4. *, **, and *** indicate

Po0.05, 0.01, and 0.001, respectively, when tested against control with

analysis of variance. Means7SEM are given.

www.jidonline.org 711

NS Funa et al.SVR Susceptibility to Apoptotic Agents

Cell morphology

Altered FAK activity is likely to result in changed cellmorphology (Hong et al., 2003) and thus we studied cellmorphology of control and Shb-knockdown cells by staining thecytoskeleton with phalloidin (Figure 3). Both populations ofcells with reduced Shb protein content exhibited an elongatedcell shape, regardless of whether the cells were maintained in10% serum or 0.1% serum. Stress fibers were commonly seen inall groups of cells, whereas microspikes and membrane ruffleswere uncommon. In the staurosporine-treated cells, the effectsof Shb protein reduction were the most pronounced. Theelongated cell phenotype of these cells became more accen-tuated by a contraction of the cytoplasm towards the centerlinelocated at the cellular axis, giving the Shb-knockdown cells aneurite-like appearance. Blinded scoring of the changes inFigure 3a was performed and revealed the differences due toShb knockdown to be statistically significant (Figure 3b).

Honokiol-induced death is accentuated by reduced Shb protein

As Shb knockdown increased stress-induced apoptosis, wewanted to investigate the effect of the natural compoundhonokiol on apoptosis in these cells (Figure 4). Honokiol is anantitumor agent that inhibits angiogenesis by interfering withVEGF signaling (Bai et al., 2003). A part of the honokiol-effectinvolves increased apoptosis as detected in SVR cells (Baiet al., 2003). The rates of cell death were increased in bothgroups of cells with reduced Shb protein (Shb shRNA, ShbshRNA-1) upon exposure to honokiol.

Reduced Shb protein content inhibits in vivo tumor growth inhonokiol-treated mice

To investigate the consequences of Shb knockdown onin vivo tumor growth of SVR cells, control and Shb-

knockdown cells were injected subcutaneously into nudemice. The tumor growth without additional treatment was notaffected by reduced Shb protein content (Figure 5a and b). Alow-dose regimen of honokiol administration was provided todetermine whether the increased susceptibility of the Shb-knockdown cells to honokiol in vitro was conferred to an invivo tumor growth situation. The low-dose honokiol admin-istration protocol did not reduce growth of the control celltumors (Figure 5a and b). However, the Shb-knockdown cellsdramatically decreased their tumor growth rate, indicating anincreased susceptibility to the antitumor compound honokiolupon Shb knockdown (Figure 5c and d). Cell death wasdramatically increased in Shb-knockdown tumors treatedwith honokiol (Figure 5d).

DISCUSSIONTreatment of solid tumors still poses a substantial therapeuticchallenge. Other than a few examples (choriocarcinoma,testicular cancer), solid tumors cannot be eliminated bychemotherapy alone. Chemotherapeutic agents can causetumor cell death through apoptosis, and tumor cells havedeveloped multiple antiapoptotic defenses, including pro-teins, which protect against both intrinsic and extrinsic formsof apoptosis. In addition, proapoptotic genes such as Apaf-1are silenced through methylation, transcriptional silencingby Id and polycomb proteins, as well as genetic deletion(Foreman et al., 1996; Hakem et al., 1998; Pan et al., 1998;Lyden et al., 1999; Polsky et al., 2001; Wang et al., 2001;Arbiser 2004; Elstrom et al., 2004; Govindarajan et al., 2005;Soengas et al., 2006).

Shb is an adapter protein that is associated with numeroustyrosine kinases, including platelet-derived growth factorreceptor beta, and VEGFR-2 (Anneren et al., 2003; Holmqvist

Staurosporine: Staurosporine:

pY576/577-FAK-

FAK-

p-Akt-

p-ERK-

ERK-

Akt-

p-Akt-

p-ERK-

ERK-

Akt-

ERK-

pY576/577-FAK-

FAK-

ERK-

Control

10% serum Cisplatin

10% serum Cisplatin

Con

trol

Shb shRNA

Shb

shR

NA

Shb shRNA-1

Control Shb shRNA Shb shRNA-1

Shb

shR

NA

-1

Con

trol

Shb

shR

NA

Shb

shR

NA

-1

Con

trol

Shb

shR

NA

Shb

shR

NA

-1

Con

trol

Shb

shR

NA

Shb

shR

NA

-1

– + – + – + – +– + – +

Figure 2. Effect of Shb knockdown on FAK, Akt, and ERK activity. (a) Control, Shb shRNA, and Shb shRNA-1 cells were plated and exposed to 0.1%

serum, 10% serum, 0.1% serum plus 1 mM staurosporine, and 0.1% serum plus 10 mM cisplatin for 1 hour. Cells were scraped off the dishes, lysates

prepared that were subjected to Western blot analysis for p576/577 FAK, total FAK, and ERK as loading control. (b) Control, Shb shRNA, and Shb shRNA-1 cells

were plated and exposed to 0.1% serum, 10% serum, 0.1% serum plus 1mM staurosporine, and 0.1% serum plus 10 mM cisplatin for 1 hour. Cells were

scraped off the dishes, lysates prepared that were subjected to Western blot analysis for pAkt, total Akt, pERK, and total ERK. The same blots as in (a) were used.

712 Journal of Investigative Dermatology (2008), Volume 128

NS Funa et al.SVR Susceptibility to Apoptotic Agents

et al., 2003, 2004). Overexpression of Shb has in manyinstances been found to be associated with increasedapoptosis, and with respect to the inhibitors of angiogenesisangiostatin and endostatin this may involve activation of FAK(Holmqvist et al., 2003; Hess et al., 2005). FAK is a proteinkinase that is associated with focal adhesions, and mediatessignaling from extracellular matrix proteins and integrins(Polte et al., 1994; Shattil et al., 1994; Hungerford et al.,1996; Sieg et al., 1999), thereby being a mediator of crosstalkbetween receptor mediated signaling and cues from extra-cellular matrix. Complete loss of FAK results in embryoniclethality, but organ-mediated deletion of FAK is compatiblewith life when deleted in the skin by cre-mediatedrecombination (Ilic et al., 1995; Essayem et al., 2006). FAK

10% serum

10% serum

Shb-shRNA

ShbshRNA-1

Control

Shb shRNA Shb shRNA-1Ctrl Shb shRNA Shb shRNA-1Ctrl

Shb shRNA Shb shRNA-1CtrlShb shRNA Shb shRNA-1Ctrl

0.1% serum

0.1% serum

+Staurosporine

+Staurosporine

+Cisplatin

+Cisplatin

Percentage of elongated cells Percentage of elongated cells

Percentage of elongated cellsPercentage of elongated cells

50

40

30

20

10

0

50

40

30

20

10

0

40

40

60

80

100

30

20

20

10

0

0

Figure 3. Effects of Shb knockdown on the cytoskeleton. (a) Control, Shb shRNA, and Shb shRNA-1 cells were cultured on multiwell microscopic slides.

After exposure to 0.1% serum, 10% serum, 0.1% serum plus 1 mM staurosporine, and 0.1% serum plus 10mM cisplatin for 1 hour, the cells were fixed

and stained for phalloidin–rhodamine. Pictures show fluorescence photographs of typical cells from each condition. Original magnification: �400.

(b) Scoring of the elongated phenotype seen in a. Percentage elongated cells (the cell length more than twice the maximum width) are shown. *, **,

and *** indicate Po0.05, 0.01, and 0.001, respectively, when compared with analysis of variance against control. Means7SEM are given for n¼ 8–11.

Cell death induced by honokiol

Ctrl Shb shRNA Shb shRNA-1454035302520151050

0 hour 48 hours

Figure 4. Effect of Shb knockdown on honokiol-induced cell death. Control,

Shb shRNA, and Shb shRNA-1 cells were exposed to 10mg/ml honokiol

for 48 hours, after which cell death rates in percent were determined.

* indicates Po0.05 with analysis of variance. Means7SEM are given for n¼ 4.

www.jidonline.org 713

NS Funa et al.SVR Susceptibility to Apoptotic Agents

is also highly expressed and amplified in human neoplasia,and is associated with metastatic growth. Interestingly,expression of a dominant-negative FAK, called FAK-relatednon-kinase (FRNK), resulted in decreased metastases, yet hadno effect on primary tumor growth (Hauck et al., 2002). FAKactivation mediates, in part, integrin- and receptor-mediatedactivation of phosphoinositol-3 kinase/Akt activation (Schal-ler et al., 1994; Siesser and Hanks, 2006). Disturbed FAK-signaling has been shown to cause apoptosis in manysystems, particularly in endothelial cells where numerousconditions giving rise to apoptosis show reduced FAKactivity, including that of staurosporine treatment (Kabiret al., 2002; Dai et al., 2004; Kurenova et al., 2004; Ryuet al., 2006).

In this study, we knocked out Shb in SVR angiosarcomacells through a cre-mediated mechanism. This allowed theexamination of the phenotype of Shb knockout without thepotential complicating factor of developmentally mediatedcompensation to Shb/FAK blockade. Consistent with priordata, elimination of Shb led to reduced phosphorylation ofY576/577 of FAK, thus showing that the blockade of Shb hada biological effect. SVR cells are driven by functional VEGFR-2 and oncogenic H-ras. Phenotypically, SVR cells lackingShb were elongated with stress fibers and showed increasedsensitivity to apoptotic agents, including cisplatin, stauros-porine, and honokiol, demonstrating a nonredundant func-tion of Shb on these parameters. On the other hand, levels ofactivated Akt and ERK were not affected. One potentialexplanation of this observation is that cells may have severalinputs that give rise to basal expression of Akt and ERKactivation, including oncogenic ras, tyrosine kinase recep-tors, and extracellular matrix–integrin interactions. Elimina-tion of Shb–FAK interactions may lead to compensation fromoncogenes or tyrosine kinase receptors in terms of maintain-

ing baseline expression of aberrant signaling, but makes theresulting tumor cells more dependent, or ‘‘addicted’’ tosignaling from tyrosine kinases and oncogenes and thus itstotal survival potential is reduced. The situation in tumorsmay differ from primary cells, in which targeted deletion ofFAK in endothelial cells leads to embryonic lethality (Brarenet al., 2006). In these mice, cues from extracellular matrix arelikely required for correct vascular patterning, and cannot becompensated by other genes.

We decided to test the hypothesis that the total input ofsurvival signals in Shb-depleted SVR cells is reduced in vivo.Honokiol is a small molecule that is known to target VEGFR-2,and was initially discovered to have antiangiogenic activity ina screen using SVR cells (Bai et al., 2003). If blockade ofShb–FAK signaling makes tumor cells more dependent onFAK-independent signaling, one would predict that the cellswould be more sensitive to an agent that targets VEGFR-2.Loss of Shb had no effect on tumor growth in vivo. We thentested the effect of a subtherapeutic dose of honokiol ontumor cells that either possess or lack wild-type Shb.Treatment of control mice with 40 mg/kg honokiol led to anincrease in tumor growth in wild-type SVR cells, but asignificant decrease in tumor volume in Shb knockout mice.A potential explanation for increased tumor growth in thelow-dose control mice is that blockade of NF-kB signaling inendothelial cells initially stimulates tumor growth, butsensitizes endothelial cells to apoptotic stimuli (Kisselevaet al., 2006). The Shb-knockout cells are more dependent onVEGFR2 and other compensatory stimuli, and thus are highlysensitive to honokiol in vivo, leading to decreased tumorgrowth.

Our findings have clinical applications. Integrin signalinghas been considered a druggable target, both at the level ofthe cell membrane, as well as downstream of the cell surface

Control Shb shRNA

Control+honokiol

Control+honokiol

Control–honokiol

Control Shb shRNA

Tumor size4,000

3,000

2,000

1,000

0

Shb shRNA+honokiol

Shb shRNA+honokiol

Shb shRNA–honokiol

a

b

c

d

Figure 5. Effect of Shb knockdown on tumor growth in vivo. (a) Nude mice were injected subcutaneously with control or Shb shRNA cells. Tumors after

9 days are shown. (b) Effect of honokiol on tumor growth of control or Shb shRNA cells. Tumors after 9 days are shown. (c) Tumor size of control or

Shb shRNA SVR cells after 9 days with or without honokiol treatment. * indicates Po0.05 with a Students’ t-test when compared against corresponding

control. Means7SEM for n¼ 3. (d) Histologic analysis of tumors: note the massive cell death occurring in the Shb knockout tumor (right) compared

with the control tumor in the presence of honokiol (left).

714 Journal of Investigative Dermatology (2008), Volume 128

NS Funa et al.SVR Susceptibility to Apoptotic Agents

integrins. In addition, although endostatin was unsuccessfulin clinical trials as monotherapy, endostatin administrationmay select for cells that have decreased Shb expression(Dixelius et al., 2000; Eder et al., 2002). These resistant cellsmay be hypersensitive to agents such as cisplatin or honokiol,providing a basis for combination or sequential therapy. Ourfindings suggest that blockade of Shb–FAK signaling isunlikely to be effective in patients as monotherapy, yet maybe clinically useful in causing tumors to become dependenton a more limited repertoire of signaling pathways. Combi-nation or sequential blockade of Shb–FAK signaling withtyrosine kinase inhibitors and chemotherapy may provebeneficial in the treatment of aggressive solid tumors.

MATERIALS AND METHODSAll experiments contained herein were approved by the Emory

University Institutional Review Board.

Shb knockdown

An inducible lentiviral system (Ventura et al., 2004) was used to

reduce the Shb protein content (Hagerkvist et al., 2007) of SVR cells

(Arbiser et al., 1997). SVR cells were infected with the lentivirus

containing the Shb shRNA sequence at 500 multiplicity of infection.

This yielded a cell population with 99% green fluorescent protein-

positivity (indicating lentiviral infection efficiency). The cells were

subjected to a second infection with an adenovirus containing cre-

recombinase Ad-Cre (5� 108 PFU/ml) (gift of RS Johnson, UCSD).

This will cause deletion of lentiviral sequences that inhibit

expression of the shRNA and simultaneously delete green fluor-

escent protein. From this population of cells with less than 5%

expressing green fluorescent protein, clones were isolated that

were homogeneously deficient in green fluorescent protein and

thus expressing the Shb shRNA. Consequently, experimentation

was carried out on SVR infected with the shRNA lentivirus

only (control), a mass population of cells subjected to a sequential

adenoviral infection to activate knockdown (Shb shRNA) and

one subclone obtained from the mass population of knockdown

cells (Shb shRNA-1).

Western blot analysis of FAK, Akt, and ERK activity

Cells (control, Shb shRNA, and Shb shRNA-1) were plated at

100,000 in 3 cm tissue culture plates. After 18 hours at 371C, cells

were maintained in 0.1% serum for 4 hours, after which they were

incubated for 1 hour in 10% serum, 1mM staurosporine, 10 mM

cisplatin or maintained in low serum. The cells were washed in

phosphate-buffered saline, scraped off the dishes, collected, and

lysed in SDS-sample buffer. Aliquots were subjected to Western blot

analysis after SDS-PAGE on 7.5% acrylamide gels, probing for

pY576/577-FAK, total FAK, pAkt (S473), total Akt, pERK (T202/

Y204), and total ERK as loading control (antibodies from Cellular

Signaling, Beverly, MA).

Cell death

Cells (control, Shb shRNA, and Shb shRNA-1) were plated at

100,000 in 3 cm diameter tissue culture dishes and maintained for

18 hours at 371C. The cells were then exposed to honokiol (10mg/

ml), staurosporine (1mM), or cisplatin (10 mM) for the time periods

indicated in the figures. Cells were then incubated with 50 mg/ml

propidium iodide for 10 minutes, washed with phosphate-buffered

saline, and subjected to flow cytometry to analyze for cell size and

uptake of propidium iodide. Dying cells were smaller than viable

cells with normal or increased propidium iodide uptake.

Cell morphology

Cells (control, Shb shRNA, and Shb shRNA-1) were plated at a

number of 5,000 in each chamber in a six-chamber microscopical

slide. After 18 hours, the serum concentration was reduced to 0.1%.

Following an incubation of an additional 4 hours at 371C, the

chambers were incubated with 10% serum, 1 mM staurosporine,

10 mM cisplatin, or maintained for an additional hour in 0.1% serum.

The glass slides were then washed twice in phosphate-buffered

saline, fixed for 15 minutes in 4% paraformaldehyde, washed again

twice in phosphate-buffered saline and permeabilized in 0.1%

Triton-X100 for 10 minutes. After additional washing for three times

with phosphate-buffered saline, the cells were blocked with 10%

FBS for 1 hour, washed again and then stained with phalloidin–rho-

damine A (Molecular Probes, Eugene, OR) for 30 minutes. After

mounting with anti-fade mounting medium, the morphology was

examined in a Nikon Eclipse TE2000 inverted fluorescence

microscope and photographed. Photographs were scored blindly

for morphology and actin staining.

Tumor growthOne million cells (control and Shb shRNA) were injected sub-

cutaneously into 5-week-old nude mice. For honokiol studies, mice

were treated daily with 40 mg/kg honokiol intraperitoneally dis-

solved in 20% Intralipid. Tumor volume was measured with a

Vernier caliper and calculated using the formula (w2� l)0.52, where

w represents the shortest dimension (Arbiser et al., 1997, 2000;

Bai et al., 2003).

CONFLICT OF INTERESTThe authors state no conflict of interest.

ACKNOWLEDGMENTSThe study was supported (MW) by the Swedish Cancer Foundation, theSwedish Research Council (31X-10822), the Juvenile Diabetes ResearchFoundation, the Swedish Diabetes Association and the Family Ernfors Fund or(JA) NIH Grant RO1 AR02030, Veterans Administration Merit Award, grantsfrom Jamie-Rabinowitch-Davis Foundation and the Minsk Foundation.

REFERENCES

Anneren C, Lindholm CK, Kriz V, Welsh M (2003) The FRK/RAK-SHBsignaling cascade: a versatile signal-transduction pathway that regulatescell survival, differentiation and proliferation. Curr Mol Med 3:313–24

Arbiser JL (2004) Molecular regulation of angiogenesis and tumorigenesis bysignal transduction pathways: evidence of predictable and reproduciblepatterns of synergy in diverse neoplasms. Semin Cancer Biol 14:81–91

Arbiser JL, Larsson H, Claesson-Welsh L, Bai X, LaMontagne K, Weiss SWet al. (2000) Overexpression of VEGF 121 in immortalized endothelialcells causes conversion to slowly growing angiosarcoma and high levelexpression of the VEGF receptors VEGFR-1 and VEGFR-2 in vivo. Am JPathol 156:1469–76

Arbiser JL, Moses MA, Fernandez CA, Ghiso N, Cao Y, Klauber N et al. (1997)Oncogenic H-ras stimulates tumor angiogenesis by two distinct path-ways. Proc Natl Acad Sci USA 94:861–6

Bai X, Cerimele F, Ushio-Fukai M, Waqas M, Campbell PM, Govindarajan Bet al. (2003) Honokiol, a small molecular weight natural product,

www.jidonline.org 715

NS Funa et al.SVR Susceptibility to Apoptotic Agents

inhibits angiogenesis in vitro and tumor growth in vivo. J Biol Chem278:35501–7

Battle TE, Arbiser J, Frank DA (2005) The natural product honokiol inducescaspase-dependent apoptosis in B-cell chronic lymphocytic leukemia(B-CLL) cells. Blood 106:690–7

Braren R, Hu H, Kim YH, Beggs HE, Reichardt LF, Wang R (2006) EndothelialFAK is essential for vascular network stability, cell survival, andlamellipodial formation. J Cell Biol 172:151–62

Dai Y, Pei XY, Rahmani M, Conrad DH, Dent P, Grant S (2004) Interruption ofthe NF-kappaB pathway by Bay 11-7082 promotes UCN-01-mediatedmitochondrial dysfunction and apoptosis in human multiple myelomacells. Blood 103:2761–70

Dixelius J, Larsson H, Sasaki T, Holmqvist K, Lu L, Engstrom A et al. (2000)Endostatin-induced tyrosine kinase signaling through the Shb adaptorprotein regulates endothelial cell apoptosis. Blood 95:3403–11

Eder JP Jr, Supko JG, Clark JW, Puchalski TA, Garcia-Carbonero R, Ryan DPet al. (2002) Phase I clinical trial of recombinant human endostatinadministered as a short intravenous infusion repeated daily. J Clin Oncol20:3772–84

Elstrom RL, Bauer DE, Buzzai M, Karnauskas R, Harris MH, Plas DR et al.(2004) Akt stimulates aerobic glycolysis in cancer cells. Cancer Res64:3892–9

Essayem S, Kovacic-Milivojevic B, Baumbusch C, McDonagh S, Dolganov G,Howerton K et al. (2006) Hair cycle and wound healing in mice with akeratinocyte-restricted deletion of FAK. Oncogene 25:1081–9

Foreman KE, Wrone-Smith T, Boise LH, Thompson CB, Polverini PJ, SimonianPL et al. (1996) Kaposi’s sarcoma tumor cells preferentially express Bcl-xL. Am J Pathol 149:795–803

Govindarajan B, Shah A, Cohen C, Arnold RS, Schechner J, Chung Jet al. (2005) Malignant transformation of human cells by constitutiveexpression of platelet-derived growth factor-BB. J Biol Chem 280:13936–43

Hagerkvist R, Mokhtari D, Lindholm C, Farnebo F, Mostoslavsky G, MulliganRC et al. (2007) Consequences of Shb and c-Abl interactions for celldeath in response to various stress stimuli. Exp Cell Res 313:284–91

Hakem R, Hakem A, Duncan GS, Henderson JT, Woo M, Soengas MS et al.(1998) Differential requirement for caspase 9 in apoptotic pathways invivo. Cell 94:339–52

Hauck CR, Hsia DA, Puente XS, Cheresh DA, Schlaepfer DD (2002) FRNKblocks v-Src-stimulated invasion and experimental metastases withouteffects on cell motility or growth. EMBO J 21:6289–302

Hess AR, Postovit LM, Margaryan NV, Seftor EA, Schneider GB, Seftor REet al. (2005) Focal adhesion kinase promotes the aggressive melanomaphenotype. Cancer Res 65:9851–60

Holmqvist K, Cross M, Riley D, Welsh M (2003) The Shb adaptor proteincauses Src-dependent cell spreading and activation of focal adhesionkinase in murine brain endothelial cells. Cell Signal 15:171–9

Holmqvist K, Cross MJ, Rolny C, Hagerkvist R, Rahimi N, Matsumoto T et al.(2004) The adaptor protein shb binds to tyrosine 1175 in vascularendothelial growth factor (VEGF) receptor-2 and regulates VEGF-dependent cellular migration. J Biol Chem 279:22267–75

Hong SY, Lee H, You WK, Chung KH, Kim DS, Song K (2003) The snakevenom disintegrin salmosin induces apoptosis by disassembly of focaladhesions in bovine capillary endothelial cells. Biochem Biophys ResCommun 302:502–8

Hungerford JE, Compton MT, Matter ML, Hoffstrom BG, Otey CA (1996)Inhibition of pp125FAK in cultured fibroblasts results in apoptosis. J CellBiol 135:1383–90

Ilic D, Almeida EA, Schlaepfer DD, Dazin P, Aizawa S, Damsky CH (1998)Extracellular matrix survival signals transduced by focal adhesion kinasesuppress p53-mediated apoptosis. J Cell Biol 143:547–60

Ilic D, Furuta Y, Kanazawa S, Takeda N, Sobue K, Nakatsuji N et al. (1995)Reduced cell motility and enhanced focal adhesion contact formation incells from FAK-deficient mice. Nature 377:539–44

Ilic D, Genbacev O, Jin F, Caceres E, Almeida EA, Bellingard-Dubouchaud Vet al. (2001) Plasma membrane-associated pY397FAK is a marker ofcytotrophoblast invasion in vivo and in vitro. Am J Pathol 159:93–108

Ishitsuka K, Hideshima T, Hamasaki M, Raje N, Kumar S, Hideshima H et al.(2005) Honokiol overcomes conventional drug resistance in humanmultiple myeloma by induction of caspase-dependent and -independentapoptosis. Blood 106:1794–800

Kabir J, Lobo M, Zachary I (2002) Staurosporine induces endothelial cellapoptosis via focal adhesion kinase dephosphorylation and focaladhesion disassembly independent of focal adhesion kinase proteolysis.Biochem J 367:145–55

Kisseleva T, Song L, Vorontchikhina M, Feirt N, Kitajewski J, Schindler C(2006) NF-kappaB regulation of endothelial cell function during LPS-induced toxemia and cancer. J Clin Invest 116:2955–63

Kurenova E, Xu LH, Yang X, Baldwin AS Jr, Craven RJ, Hanks SK et al. (2004)Focal adhesion kinase suppresses apoptosis by binding to the deathdomain of receptor-interacting protein. Mol Cell Biol 24:4361–71

Lyden D, Young AZ, Zagzag D, Yan W, Gerald W, O’Reilly R et al. (1999) Id1and Id3 are required for neurogenesis, angiogenesis and vascularizationof tumour xenografts. Nature 401:670–7

Pan G, O’Rourke K, Dixit VM (1998) Caspase-9, Bcl-XL, and Apaf-1 form aternary complex. J Biol Chem 273:5841–5

Polsky D, Young AZ, Busam KJ, Alani RM (2001) The transcriptional repressorof p16/Ink4a, Id1, is up-regulated in early melanomas. Cancer Res61:6008–11

Polte TR, Naftilan AJ, Hanks SK (1994) Focal adhesion kinase is abundant indeveloping blood vessels and elevation of its phosphotyrosine content invascular smooth muscle cells is a rapid response to angiotensin II. J CellBiochem 55:106–19

Ryu SJ, Cho KA, Oh YS, Park SC (2006) Role of Src-specific phosphorylationsite on focal adhesion kinase for senescence-associated apoptosisresistance. Apoptosis 11:303–13

Schaller MD, Hildebrand JD, Shannon JD, Fox JW, Vines RR, Parsons JT(1994) Autophosphorylation of the focal adhesion kinase, pp125FAK,directs SH2-dependent binding of pp60src. Mol Cell Biol 14:1680–8

Shattil SJ, Haimovich B, Cunningham M, Lipfert L, Parsons JT, Ginsberg MHet al. (1994) Tyrosine phosphorylation of pp125FAK in platelets requirescoordinated signaling through integrin and agonist receptors. J BiolChem 269:14738–45

Siddik ZH (2002) Biochemical and molecular mechanisms of cisplatinresistance. Cancer Treat Res 112:263–84

Sieg DJ, Hauck CR, Schlaepfer DD (1999) Required role of focal adhesionkinase (FAK) for integrin-stimulated cell migration. J Cell Sci 112(Part16):2677–91

Siesser PM, Hanks SK (2006) The signaling and biological implications of FAKoverexpression in cancer. Clin Cancer Res 12:3233–7

Soengas MS, Gerald WL, Cordon-Cardo C, Lazebnik Y, Lowe SW (2006)Apaf-1 expression in malignant melanoma. Cell Death Differ 13:352–3

Ventura A, Meissner A, Dillon CP, McManus M, Sharp PA, Van PL et al.(2004) Cre-lox-regulated conditional RNA interference from transgenes.Proc Natl Acad Sci USA 101:10380–5

Wang J, Mager J, Chen Y, Schneider E, Cross JC, Nagy A et al. (2001)Imprinted X inactivation maintained by a mouse Polycomb group gene.Nat Genet 28:371–5

Withers BE, Hanks SK, Fry DW (1996) Correlations between the expression,phosphotyrosine content and enzymatic activity of focal adhesionkinase, pp125FAK, in tumor and nontransformed cells. Cancer BiochemBiophys 15:127–39

716 Journal of Investigative Dermatology (2008), Volume 128

NS Funa et al.SVR Susceptibility to Apoptotic Agents