Proteasome inhibitor PS-341 causes cell growth arrest and apoptosis in human glioblastoma multiforme...

Proteasome inhibitor PS-341 causes cell growth arrest and apoptosis

in human glioblastoma multiforme (GBM)

Dong Yin*,1, Hong Zhou1, Takashi Kumagai1, Gentao Liu2, John M Ong2, Keith L Black2

and H Phillip Koeffler1

1Division of Hematology/Oncology, Cedars-Sinai Medical Center, UCLA School of Medicine, Los Angeles, CA 90048, USA;2Maxine Dunitz Neurosurgical Institute, Cedars-Sinai Medical Center, UCLA School of Medicine, Los Angeles, CA 90048, USA

The proteasome plays a pivotal role in controlling cellproliferation, apoptosis, and differentiation in a variety ofnormal and tumor cells. PS-341, a novel boronic aciddipeptide that inhibits 26S proteasome activity, hasprominent effects in vitro and in vivo against several solidtumors. We examined its antiproliferation, proapoptoticeffects using three human glioblastoma multiforme(GBM) cell lines and five primary GBM explants.PS-341 markedly inhibited proliferation of GBM celllines and explants in liquid and soft agar culture. Thesecells developed a G2/M cell cycle arrest with aconcomitant decreased percentage of cells in S phase(E2-fold), associated with an increased expression ofp21WAF1, p27KIP1, as well as cyclin B1 and decreased levelsof CDK2, CDK4, and E2F4. About 35–40% of the cellsbecame apoptotic when exposed to PS-341 (10�7

M,24–48 h) as shown by Annexin V analysis; in concert withthese findings, immunobloting showed a C-terminal85 kDa apoptotic fragment of poly ADP-ribose polymer-ase (PARP), and a decreased level of Bcl2 and Bcl-xl.PS-341 downregulated the expression of Bcl-2 and Bcl-xlin protein levels at an early time of treatment. These changesoccurred irrespective of the p53 mutational status of thecells. PS-341 activated JNK/c-Jun signaling in GBMcells, and the JNK inhibitor SP600125 blocked the JNKsignaling to reverse partially the PS-341 growth inhibi-tion. PS-341 (10�7 M, 24 h) decreased nuclear NF-jBlevels as shown by Western blot, and reduced transcrip-tional activity of NF-jB as measured by reporter assays inthese transformed cells. Also, PS-341 enhanced TRAIL(TNF-related apoptosis-inducing ligand) and TNFa (tu-mor necrosis factor alpha) induced cell death andapoptosis (two- to five-fold) in GBM cells. In summary,PS-341 has profound effects on growth and apoptosis ofGBM cells, suggesting that PS-341 may be an effectivetherapy for patients with gliomas.Oncogene (2005) 24, 344–354. doi:10.1038/sj.onc.1208225Published online 8 November 2004

Keywords: proteasome inhibitor; glioblastoma multi-forme; PS-341; apoptosis; NF-kB

Introduction

The ubiquitin–proteasome pathway is involved inproteolysis of most nuclear and cytosolic proteins, andin particular, many of the short-lived regulatory proteinsthat govern growth, activation, and signaling. Theproteasome, therefore, represents a novel target forcancer therapy. PS-341 (Velcade/bortezomib) is adipeptidyl boronic acid inhibitor with high specificityfor the proteasome (Adams et al., 2000). The antitumoreffects of PS-341 have been extensively studied inmutiple myeloma (MM) with inactivation of nuclearfactor-kB (NF-kB) being one of its major modes ofaction. The disease improved in many MM patientstaking the drug (Hideshima et al., 2001, 2002, 2003;LeBlanc et al., 2002; Mitsiades et al., 2002; Orlowskiet al., 2002).Preclinical studies have shown that this compound

has antitumor activity against a variety of solid tumors,including carcinomas of the breast (Teicher et al., 1999),lung (Ling et al., 2003), colon (Cusack et al., 2001),bladder (Johnson et al., 2003), ovary (Frankel et al.,2000), pancreas (Nawrocki et al., 2002), and prostate(Frankel et al., 2000). Sensitivity of glioblastoma multi-forme (GBM) to PS-341 has not been extensivelystudied.GBM is the most malignant, invasive, difficult to treat

primary brain tumor. This form of cancer has a rapidgrowth rate, being capable of doubling in size within10–20 days. Successful treatment of GBM is rare;standard therapies result in a mean survival of only about10–12 months. In the present study, the antigliomaactivity of PS-341 was examined using three humanGBM cell lines and five primary GBM explants. Thedrug (10�7 M, 2 days) caused GBM cells to growth arrestin the G2/Mphase of the cell cycle, which was associatedwith 35–40% cells becoming apoptotic. Our datasuggest that besides inhibition of NF-kB, activation ofthe JNK/c-Jun signaling pathway may be another

Received 19 May 2004; revised 21 September 2004; accepted 21September 2004; published online 8 November 2004

*Correspondence: D Yin, Division of Hematology and Oncology,Cedars-Sinai Medical Center, UCLA School of Medicine, 8700Beverly Boulevard, Davis Building 5022 Room, Los Angeles, CA90048, USA; E-mail: [email protected]

Oncogene (2005) 24, 344–354& 2005 Nature Publishing Group All rights reserved 0950-9232/05 $30.00

www.nature.com/onc

mechanism by which PS-341 causes apoptosis of GBMcells. When combined with either tumor necrosis factoralpha (TNFa) or TNF-related apoptosis-inducing ligand(TRAIL), PS-341 enhanced the antiproliferation effectsof these apoptotic stimuli. Taken together, PS-341 ishighly effective against GBM in vitro, and may have arole in the therapy of this difficult to manage disease.

Results

PS-341 inhibited cell proliferation and clonal growthindependent of p53 mutational status in GBM cells

The mutational status of p53 in the GBM cells wasdetected by single-strand conformation polymorphism(SSCP) analysis and nucleotide sequencing: Those with

0 10-9 0

25

50

75

100

125

YK163T98G

U87MM156U343

cell

surv

ival

(%

of c

ontr

ol)

10-8 10-7 10-6

PS-341 (M)

Figure 1 Effect of the PS-341 protease inhibitor on the growth ofGBM cells. T98G (p53 mt), U87 (p53 wt), U343 (p53 mt) GBM celllines, and MM156 (p53 wt) and YK163 (p53 mt) GBM explantswere cultured in the presence of various concentrations (10�9–10�6 M) of PS-341 for 48 h. Cell numbers were measured by MTT.Results are expressed as a mean percentage of cell counts in controlwells. Each point represents a mean7standard deviation (s.d.) ofthree independent experiments carried out in triplicate

a T98G GBM cell line

Clonal growth of T98G

control 10-11 10-10 10-9 10-8 10-7

0

25

50

75

PS-341 (10-7 M)Control

d DA66 GBM explant

Clonal growth of DA66

cont

rol0

25

50

75

100

colo

nise

Control

PS-341 (10-7 M)

c JM94 GBM explant

Control PS-341 (10-7 M)

Clonal growth of U87

cont

rol0

10

20

30

40

colo

nies

b U87 GBM cell line

Control

PS-341 (10-7M)

colo

nies

0

100

200

300Clonal growth of JM94

colo

nies

PS

-341

(10

-7 M

)

PS

-341

(10

-7 M

)

PS-341 (M)

control 10-11 10-10 10-9 10-8 10-7

PS-341 (M)

Figure 2 PS-341 inhibits clonal growth of GBM cells in soft agar. GBM cells were cultured in two-layer soft agar with PS-341 (10–11–10�7 M),colonies were counted after 14 days. Panel a: T98G (p53 mt); Panel b: U87 (p53 wt); Panel c: JM94 (p53 mt); Panel d: DA66 (p53 wt).Photographs are representative results of the colony assay. Graphs show the mean7s.d. of quadruplicate wells. The experiments were repeatedtwice with similar results

PS-341 induces growth arrest and apoptosis in GBMD Yin et al

345

Oncogene

p53 mutations included T98G (M237I), U118 (R273H),U343 (R213Q), YK163 (V203L), JM94 (R273C); andthose with wild-type p53 were U87, DA66, and MM156(data not shown). We examined the effect of PS-341(10�9–10�6 M, 48 h) on cell proliferation of GBM cells inliquid culture using the MTT (3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl tetrazolium bromide) assay. PS-341produced a dose-dependent inhibition of growth of thethree permanent GBM cell lines (U87 (p53 wt), T98G(p53 mt), and U343 (p53 mt)) and the two GBMexplants (MM156 (p53 wt), YK163 (p53 mt)). Amongthese cell lines, T98G with mutant p53 was the mostsensitive (ED50D9� 10�9 M), while the ED50 for theother p53 mutant lines, YK163 and U343, were 1� 10�7

and 8� 10�7 M, respectively. The ED50 for U87 andMM156 with wild-type p53 were approximately 8� 10�8

and 10�7 M, respectively (Figure 1).When the GBM cell lines (U87 (p53 wt), T98G (p53

mt)) and GBM explants (DA66 (p53 wt), JM94 (p53mt)) were cultured in soft agar either with or withoutPS-341, clonal growth was inhibited significantly byPS-341 (Figure 2). Dose–response studies using T98Gand JM94 showed an ED50 of approximately 6� 10�9

and 1.6� 10�8 M, respectively (Figure 2). In summary,PS-341 inhibited growth of GBM cell lines and explantsin liquid and soft agar culture with mean ED50 forPS-341 ranging from 10�8 to 10�6 M, and sensitivity ofthe GBM cells to the compound was independentof their p53 status.

PS-341 increased percentage of GBM cells in G2/M,decreased those in S phase and induced expression of cellcycle-related proteins

To understand the mechanism of action of PS-341,we initially studied the cell cycle distribution by flowcytometry in three GBM cell lines and three GBMexplants. The percentage of cells in the G2/Mphaseincreased about two-fold in all six GBM cell linesand explants after their exposure to PS-341 for 6–12 h(Figure 3). The percentage of cell in S phasedecreased about two-fold in five of the six celllines and explants after their exposure to PS-341 for24–48 h.PS-341 (10�7 M, 3, 9, 24 h) caused the accumulation

of cyclin-dependent kinase inhibitors p21WAF1, p27KIP1,and G2/Mphase control protein cyclin B1 in theGBM cell lines (U87 and T98G, Figure 4a and b)and GBM explants (DA66 and JM94, Figure 4c andd). Also, expression of CDK2, CDK4, as well asE2F4 decreased in these cells after their culture withPS-341 (Figure 4). Accumulation of p53 wt in U87cells did not change during exposure to PS-341(Figure 4a).

PS-341 induced apoptosis in GBM cells

PS-341 (10�7 M, 3, 24, 48 h) caused apoptosis and/ordeath of all tested GBM cell lines and explants in a

0%

20%

40%

60%

80%

100%

0%

20%

40%

60%

80%

100%

G2/M

S

G0/G1

JM94

G2/M

S

BB184

0%

20%

40%

60%

80%

100%

0 3 6 12 24 48 0 3 6 12 24

0 3 6 12 24 36 48

0 3 6 12 24 36 48

G2/M

G0/G1

DA66

0%

20%

40%

60%

80%

100%G2/M

S

G0/G1

T98G

0%

20%

40%

60%

80%

100%

exposure (hours) exposure (hours)

cell

cycl

e (%

of

cells

)ce

ll cy

cle

(% o

f ce

lls)

cell

cycl

e (%

of

cells

)

cell

cycl

e (%

of

cells

)ce

ll cy

cle

(% o

f ce

lls)

cell

cycl

e (%

of

cells

)

G2/M

S

G0/G1

U118

0%

20%

40%

60%

80%

100%

G2/MS

G0/G1

U87

G0/G1

S



0 3 6 12 24 48

0 3 6 12 24 36 48

Figure 3 Cell cycle analysis of GBM cells treated with PS-341. Cell cycle distribution of T98G (p53 mt), U118 (p53 mt), U87 (p53 wt)GBM cell lines and BB184 (p53 wt), DA66 (p53 wt), JM94 (p53 mt) GBM explants was analysed after culturing for 0, 3, 6, 12, 24, 36,and 48 h with PS-341 (10�7M). Cell cycle was assessed by propidium iodide (PI) staining and FACS analysis. For each sample, thepercentage of cells in the G0/G1, S, or G2/Mphase of the cell cycle is indicated. The experiments were repeated three times with similarresults


346

Oncogene

time-dependent manner as measured by Annexin Vassay (Figure 5). On the second day of culture with PS-341, 12–23% of the cells were apoptotic (Annexin Vpositive) and an additional 15–26% of the cells weredead (PI and Annexin V positive). Immunoblottinganalysis of poly ADP-ribose polymerase (PARP)showed that the C-terminal 85 kDa PARP apoptoticfragment was generated in U87, T98G, and JM94 cellstreated with PS-341 for 9–24 h (Figure 6a–c). Inaddition, the levels of the antiapoptotic proteins, Bcl2and/or Bcl-xl, markedly decreased in these three celllines (Figure 6a–c). The mRNA expression of Bcl-2and Bcl-xl in U87 cells was measured by real-timereverse transcription–PCR (RT–PCR) (Figure 6d).Western blotting showed that protein levels of Bcl-2decreased about 50% at 3 h, and 90% at 9 and 24 h.However, the mRNA of Bcl-2 as measured by real-timeRT–PCR decreased about 50% at 9 and 24 h. Parallelfindings were observed for Bcl-xl. These results suggestthat the early effect of PS-341 is to decrease proteinlevels of Bcl-2 and Bcl-xl followed by a decrease ofboth protein and RNA expression of these genes.

PS-341 activated JNK/c-Jun signaling in GBM cells, andJNK inhibitor SP600125 blocked signaling of JNKprotecting cells from growth inhibition mediatedby PS-341

We explored by Western blot analysis whether PS-341could induce phosphorylation of JNK and c-Jun in GBMcells. PS-341 (10�7M) caused the phosphorylation of bothJNK and c-Jun (Figure 7). Phosphorylation was noted by3 h but massive phosphorylation of both proteinsoccurred by 24h in all four GBM cell lines and explants.To study the role of the JNK/c-Jun signal pathway in

the PS-341-induced growth arrest of GBM cells, thispathway was blocked using the JNK-specific inhibitorSP600125. In the presence of SP600125, the phosphor-ylation of JNK and c-Jun mediated by PS-341 wasmarkedly decreased in T98G cells (Figure 8a). Theseresults suggest that the JNK inhibitor SP600125effectively blocked the JNK/c-Jun signal pathway. Infurther experiments, T98G cells were cultured withPS-341 (10�7 M) either alone or in combination withSP600125 (4� 10�5 M) for 1 day. SP600125 blunted the

p21

0 3h 9 24 hrs

p27

CDK2

CDK4

E2F4

GAPDH

p21

0 3 9 24 hrs

p27

GAPDH

b T98G

p21

p27

Cyclin B1

0 3 9 24 hrs

DA66

0 3 9 24 hrs

p21

p27

GAPDH

JM94

a

c d

U87

GAPDH

Cyclin B1

E2F4CDK2

CDK2

Cyclin B1

E2F4

Cyclin B1

p53

Figure 4 Effect of PS-341 on expression of cell cycle-related proteins in GBM cell lines and explants. After culturing cells with PS-341(10�7 M) for 0, 3, 9, 24 h, total cell lysates were Western blotted and probed with antibodies against p21waf1, p27KIP1, cyclin B1, CDK2,CDK4, E2F4, p53, and GADPH to control for equal loading. Panel a: U87 (p53 wt) GBM cell line; Panel b: T98G (p53 mt) GBM cellline; Panel c: DA66 (p53 wt) GBM explant; Panel d: JM94 (p53 mt) GBM explant


347

Oncogene

ability of PS-341 to induce growth arrest of T98G cellsas measured by the MTT assay (Figure 8b). PS-341reduced MTT activity by about 50%; on the other handunder similar conditions, only about 20% reduction wasobserved when the cells were cultured with thecombination of PS-341 and the JNK inhibitorSP600125 (10 or 20 mM), Po0.05 (t-test) (Figure 8b).

Effect of PS-341 on NF-kB in GBM cells

One of the major effects of proteasome inhibition inmany cell types is the reduction of NF-kB translocation

to the nucleus, due to blocking of the cytosolicdegradation of IkB. In order to assess the effects ofPS-341 on NF-kB activation in GBM cells, the IkBprotein level was examined by Western blotting, andwas found to accumulate in each of the two GBM celllines and the two GBM explants that we examined(Figure 9a–d). Furthermore, immunoblotting resultsshowed that nuclear NF-kB levels decreased in T98G,which had been cultured with PS-341 (1� 10�7 M, 24 h)compared to control untreated cells (Figure 9e). We alsotransfected the T98G GBM cells with an NF-kBpromoter coupled to a luciferase reporter construct.

Figure 5 Apoptotic analysis of GBM cell lines and explants treated with PS-341. U87 (p53 wt) and T98G (p53 mt) GBM cell lines,and DA66 (p53 wt) and JM94 (p53 mt) GBM explants were cultured for 0, 3, 24, 48 h with PS-341 (10�7M); apoptosis was assessed byAnnexin V (FL-1) and PI (FL-3) staining and FACS analysis. Three experiments were carried out for each cell type; and the resultswere similar. Panels a–d show a representative experiment


348

Oncogene

BCL2

BCL-xl

PARP

Control 3 9 24 hrs

a

b

c

d

U87

PARP

Control 3 9 24 hrsT98G

JM94

PARP

GAPDH

GAPDH

GAPDH

Control 3 9 24 hrs

P85/PARP

P85/PARP

BCL2

BCL-xl

P85/PARP

100% 45% 3% 12%

100% 24% 1% 4%

control 3 6 9 24

control 3 6 9 24

0

50

100

150

Bcl

-2m

RN

A le

vels

(%

of

con

tro

l)

0

50

100

150

hours

hours

Bcl

-xl m

RN

A le

vels

(%

of

con

tro

l)

U87

Figure 6 Effect of PS-341 on expression of apoptotic-related proteins in GBM cell lines and explant. U87 (p53wt) (panel a) and T98G (p53 mt) (panel b) GBM cell linesand JM94 (p53 mt) (panel c) GBM explant were culturedwith PS-341 (10�7 M) for 0, 3, 9, and 24 h, lysates weremade and subjected to Western blot analysis for Bcl-2,Bcl-xl, and poly ADP-ribose polymerase (PARP), gen-erating a C-terminal 85 kDa apoptotic fragment (arrow).Panel d: real-time RT–PCR examined the mRNA levels ofBcl-2 and Bcl-xl in U87 GBM cells treated with PS-341(10�7 M) for 0, 3, 6, 9, and 24 h

JNK1JNK2

P-c-jun

C-jun

GAPDH

P-c-jun

C-jun

GAPDH

P-c-jun

GAPDH

P-c-jun

GAPDH

Control 3 9 24 hrs

Control 3 9 24 hrs

a U87

b T98G

d JM94

c DA66

P-JNK2P-JNK1

P-JNK2P-JNK1

P-JNK2P-JNK1

P-JNK2P-JNK1

JNK1JNK2

Control 3 9 24 hrs

Control 3 9 24 hrs

Figure 7 PS-341 activates JNK/c-Jun signaling in GBMcells. After culturing U87 (p53 wt) and T98G (p53 mt)GBM cell lines, as well as, DA66 (p53 wt) and JM94 (p53mt) GBM explants with PS-341 (10�7 M) for 0, 3, 9, 24 h,total cell lysates were subjected to Western blotting andprobed for phosphorylated and unphosphorylated JNK-1/2 and c-Jun, as well as GADPH as loading control


349

Oncogene

Exposure of these cells to PS-341 (10�7 M, 24 h)decreased their levels of NF-kB activity by about 50%(Figure 9f).

Drug interactions between the PS-341 and either TNFa orTRAIL in GBM

The ability of TRAIL and TNFa to induce apoptoticdeath of GBM cells was enhanced by PS-341 (Figure 10).T98G cells were cultured with PS-341 (10�8 M or5� 10�8 M) combined with either TRAIL (10, 20 ng/ml) (Figure 10a, Po0.001 compared to either agentalone, t-test) or TNFa (50, 100 ng/ml) (Figure 10b,Po0.01 compared to either agent alone, t-test). Both theMTT and annexin-V assay showed that the ability ofTRAIL and TNFa to induce cell death and apoptosiswas enhanced two- to five-fold by PS-341 in these GBMcells (Figure 10).

Discussion

Boronated proteasome inhibitors, such as PS-341, aremore potent than their structurally related aldehydes(Ling et al., 2003). In vitro screening data using a panel

of 60 human tumors cell lines indicated that PS-341 hassignificant growth inhibitory activity against a widevariety of tumors (Adams et al., 1999). Other protea-some inhibitors, such as the calpain inhibitor, lactacys-tin, MG132, and AcLLNal, also inhibit glioma cellgrowth (Kitagawa et al., 1999; Wagenknecht et al., 1999;Tani et al., 2001). However, little has been reported ofthe effect of PS-341 on GBM cells. We found that cellgrowth of three of the GBM cell lines and five of theGBM explants were growth inhibited in liquid and softagar culture by PS-341 (ED50, 10�8–10�6 M). Priorinvestigations have shown that PS-341 blocked cells ofseveral types of solid tumors in the G2/Mphase of thecell cycle (Adams et al., 1999; Sunwoo et al., 2001; Linget al., 2002, 2003; Nawrocki et al., 2002). In this study,both the GBM cell lines and explants cultured withPS-341 accumulated in the G2/Mphase associated withtheir increased levels of p21WAF1, p27KIP1, and cyclin B1.Previous studies using other proteasome inhibitorsfound that expression of p21WAF1 and p27KIP1 increasedin glioma cells (Kitagawa et al., 1999; Wagenknechtet al., 1999). Furthermore, we found that levels ofCDK2 and CDK4, as well as, E2F4 decreased (Figure 4)consistent with the cells exiting the cell cycle.The p53 protein plays an important role in cell cycle

regulation and apoptosis (Kuerbitz et al., 1992; Wu andLevine, 1994). Several studies suggested that proteasomeinhibitors induced apoptosis and inhibited growth oftumor cells in a p53-independent manner (Shinoharaet al., 1996; Kitagawa et al., 1999; Wagenknecht et al.,1999); others have suggested these effects occurred in ap53-dependent (Lopes et al., 1997) or partly dependentfashion (Ling et al., 2003). We found that both thep53-wt and -mutant GBM cell lines and explants weresensitive to growth inhibition and apoptosis mediated byPS-341. About 35% of GBMs have mutant p53 (Gallouet al., 2001), emphasizing the importance of selecting atherapy that can function independent of the p53mutational status of the tumors.Stress stimuli can activate either the stress-activated

protein kinase (SAPK) or JNK to induce apoptosis(Verheij et al., 1996). Previous studies have shown thatPS-341-induced apoptosis was associated with activa-tion of JNK in MM cells and lung cancer cells(Hideshima et al., 2003; Yang et al., 2004). BlockingJNK by a specific JNK inhibitor, SP600125, abrogatesPS-341 induction of apoptosis in prostate cancer (Yanget al., 2004). We found that PS-341 markedly stimulatedphosphorylation of JNK and c-Jun in GBM cells, andthe JNK inhibitor SP600125 blocked both this phos-phorylation and partly blocked the cell death mediatedby PS-341. These results suggested that JNK/c-Junsignaling pathway is in part involved in the apoptosisinduced by PS-341.One common feature of most GBM is a hyperactive

NF-kB pathway (Hayashi et al., 2001), and the rate ofgrowth of these cells paralleled the activity of thispathway in these tumors (Nagai et al., 2002). Therefore,this pathway becomes an important therapeutic targetfor these tumors. NF-kB activation depends on thesignal-induced phosphorylation and ubiquitination of

0.0

50

100

cell

surv

ival

(%

of c

ontr

ol)

PS-341 (10-7M)

P-JNK2P-JNK1

P-c-jun

GAPDH

40-

--

1+

-+

20+

40+

5+

10+

Sp600125 (40 uM)PS-341 (1 x 10-7M)

+-

--

++

-+

T98G

T98G

** ***

a

b

Sp600125 (uM)

Figure 8 JNK inhibitor Sp600125 blocks PS-341 inducible celldeath. Panel a: T98G GBM cell line was cultured with PS-341(10�7M) either with or without JNK inhibitor Sp600125 (40 mM,24 h). Western blot of the cells showed that phosphorylation ofJNK1/2 and c-jun was blocked by SP600125. Panel b: T98G cellswere cultured with either Sp600125 or PS-341 alone or incombination. MTT assay showed that blockade of JNK signalingby SP600125 partially protected cells from growth inhibitioncaused by PS-341. **Po0.01; *Po0.05


350

Oncogene

the inhibitory protein IkBa. PS-341 can stabilize IkBa byinhibiting the chymotryptic activity of the 26s proteasome.Our study demonstrated that PS-341 could significantlydecrease the nuclear activity of NF-kB in GBM.Previous studies have shown that activation of NF-kB

can inhibit apoptosis induced by a number of stimuli(Wang et al., 1998). Both TRAIL and TNFa can bind totheir respective receptors, and activate the caspaseapoptotic pathway, as well as, stimulate NF-kB activa-tion (Franco et al., 2001; Dai et al., 2003). Proteasomeinhibitors can block the stimulation of NF-kB whileleaving the activation of the caspase pathway intact,explaining that PS-341 may enhance TNFa and TRAILcell death (Franco et al., 2001; An et al., 2003) (Figure 11).In our studies, PS-341 markedly increased apoptosis ofthe GBM cell lines treated with either TNFa or TRAIL,suggesting that PS-341 may be a clinically useful adjunctto therapy with TRAIL and TNFa-like drugs.In summary, we have demonstrated that PS-341 has

profound effects on GBM cells including inhibiting their

growth in liquid culture and their clonogenic prolifera-tion in soft agar. This was accompanied by theiraccumulation in the G2/Mphase of the cell cycle withincreased levels of p21WAF1, p27KIP1, and cyclin B1decreased expression of CDK2, CDK4, and E2F4 in ap53-independent fashion. In conjunction with thesechanges, activation of NF-kB was blunted, and JNKand c-Jun underwent marked phosphorylation. Theseevents cumulated in apoptosis. These findings suggestthat PS-341 may have a niche in the therapy ofglioblastomas.

Materials and methods

Glioma cell lines and primary cell cultures

Human GBM cell lines U87 (p53 wild type (p53 wt)), T98G(p53 mutant (p53 mt)), U118 (p53 mt) were maintained inDulbecco’s modified Eagle’s medium (Gibco, BRL) with 10%fetal calf serum (Gemini Bio-Products, Calabasas, CA, USA),

hnRNP

NF-kB

PS-341 - +

T98G

T98G T98G + PS-3410

25

50

75

100

125

luci

fera

se a

ctiv

ity (

%)

4 x NF-kB sites Luc

T98G0 3h 9 24 hrs

GAPDH

0 3 9 24 hrs

GAPDH

T98G

GAPDH

0 3 9 24 hrsDA66

0 3 9 24 hrs

GAPDH

JM94

a

b

c

d

e

f

U87

IkB

IkB

IkB

IkB

Figure 9 PS-341 increased IkB and decreased both the level of nuclear NF-kB and its transcriptional activity in GBM cells. Panels a–d: U87 (p53 wt) and T98G (p53 mt) GBM cell lines as well as DA66 (p53 wt) and JM94 (p53 mt) GBM explants were cultured with PS-341 (10–7M) for 0, 3, 9, 24 h and Western blotted for levels of IkB. Panel e: T98G cells were cultured either without (�) or with (þ ) PS-341 (10�7 M, 24 h). Nuclear protein was extracted, Western blotted and probed for NF-kB and hnRNP (nuclear protein examined ascontrol for equal loading). Panel f: The construct (NF-kB-Luc) containing four copies of NF-kB-binding sites attached to pGL3luciferase reporter plasmid (top panel) was transiently transfected with p-RL-SV40-Luciferase (renilla luciferase) plasmid into T98Gcells. After 3 h, cells were treated with either PS-341 (10�7M) or control diluent for 24 h, and luciferase activity was measured. Resultsrepresent mean7s.d. of three experiments with triplicate dishes per experimental point


351

Oncogene

10U/ml penicillin-G, and 10mg/ml streptomycin (GeminiBio-Products, Calabasas, CA, USA). Primary GBM explants(DA66 (p53 wt), JM94 (p53 mt), MM156 (p53 wt), YK163(p53 mt), BB184 (p53 wt)) were established from patientswith GBM undergoing surgery. Written informed consent forresearch use of tumor tissue was obtained from each patientprior to surgery, according to a protocol approved by theinstitutional ethics committee. Tumor specimens were im-mediately transported to the laboratory, finely minced tosingle-cell suspension and cultured in complete medium(Ham’s F-12/DME High Glucose medium (Irvine Scientific,Santa Ana, CA, USA) containing 10% fetal calf serum(Gemini Bio-Products, Calabasas, CA, USA), 10U/mlpenicillin-G, and 10mg/ml streptomycin (Gemini Bio-Pro-ducts, Calabasas, CA, USA) and 2mM glutamax-1 (Invitro-gen, CA, USA)) into 100 cm2 tissue culture plastic dishes.Cells were then collected and aliquots were cryopreserved inliquid nitrogen. One aliquot of cells was kept in culture andgrown to confluence. Cells used in these experiments weresubcultured for no more than 10 additional passages. All cellswere incubated at 371C in 5% CO2. For clonogenic assays,

cells were plated into 24-well flat-bottomed plates using atwo-layer soft agar system with a total of 1� 103 cells/well ina volume of 400 ml/well, as described previously (Munkeret al., 1986). After 14 days of incubation, the colonies werecounted.

Chemicals

Proteasome inhibitor PS-341 was obtained from MillenniumPharmaceuticals (Cambridge, MA, USA), and dissolved inDMSO at a stock concentration of 10�2 M and stored at�201C. Fresh dilutions in medium were made for eachexperiment.

Cell proliferation

The cells were placed into 96-well plates at 2.0� 103 cells/well,and cell proliferation was measured at various times by MTTassay according to the protocol provided by Roche MolecularBiochemicals (Basel, Switzerland).

Figure 10 PS-341 enhances antiproliferation activity of TRAIL and TNFa. Panels a and b: T98G GBM cells were cultured in PS-341(10�8 M or 5� 10�8M) either alone or combined with either TRAIL (10 or 20 ng/ml) or TNFa (50 or 100 ng/ml). MTT assay was carriedout after 24 h. Results represent the mean7s.d. of three experiments with each experimental point having three wells. *Po0.05;**Po0.01; ***Po0.001, as compared to either agent alone. Panels c and d: T98G GBM cells were cultured with either PS-341(5� 10�8 M), TRAIL (10 ng/ml), or TNFa (100 ng/ml) alone or combined for 12 h, and apoptosis (Annexin V positive) and death(Annexin V and PI positive) was measured. Experiments were repeated three times with similar results. Panels c and d show arepresentative experiment


352

Oncogene

Western blot

Cells were harvested for total cell lysates with RIPA buffer(1% Nonidet P-40, 0.5% sodium deoxycholate, 0.1% SDS,50mM Tris-HCl, pH 7.5) containing a mixture of proteaseinhibitors (Roche Diagnostics GmbH, Mannheim, Germany)as well as 1mM NaF and 1mM NaVO4. Cell lysates werecentrifuged at 13 000 r.p.m. for 10min at 41C. Supernatant wascollected, and the protein concentration was measured. Theprocedure for the nuclear protein extraction was carried outaccording to the manufacturer’s instructions (CelLytic NuclearExtraction Kit (Sigma)). Cells were swelled with hypotonicbuffer, and then disrupted. The cytoplasmic fraction wasremoved, and the nuclear protein was released from the nucleiby a high salt buffer. The lysates (30 mg) were denatured in thesample buffer (10% glycerol, 5% b-mercaptoethanol, 2.3%SDS, 62.5mM Tris-HCl, pH 6.8) by boiling and then subjectedto 4–15% SDS–PAGE followed by electrotransfer to poly-vinylidene difluoride membrane. The immune complexes werevisualized with Supersignal West Pico ChemiluminescentSubstrate or Supersignal west dura extended duration sub-strate (Pierce, Rockford, IL, USA) and normalized by internalcontrol (GAPDH). The antibodies were bought from SantaCruz Biotechnology Inc. (CA, USA).

Detection of p53 mutations

Presence of p53 gene mutations was detected by single-strandconformation polymorphism (SSCP) as description in a prior

study (Imai et al., 1994). The p53 gene was analysed formutations in exons 5–8, which encompasses the region that ismost frequently mutated, and all aberrantly migrating bandswere DNA sequenced.

Cell cycle analysis

GBM cells (BB184, DA66, JM94, T98G, U118, U87) wereplaced in six-well dishes, cultured for 0, 3, 6, 12, 24, 36, 48 hwith PS-341 (10�7M). After being trypsinized and washed twicewith ice-cold phosphate-buffered saline (PBS), cells were fixedin 70% ice-cold ethanol overnight. The samples were treatedby RNAase A and exposed to propidium iodide (PI), thenanalysed by flow cytometry (FACScan, Becton Dickinson).

Assessment of apoptosis

To study induction of apoptosis, annexin V assays (AnnexinV-FITC Apoptosis Detection Kit; Pharmingen, San Diego,CA, USA) were performed according to the manufacturer’sinstructions. Briefly, cells were harvested after exposure toreagents, washed twice with PBS, incubated with FITC-conjugated annexin V and PI for 15min, and measured byFACScan (Becton Dickinson).

RNA isolation and RT–PCR

U87 cells were treated by PS-341 for 0, 3, 6, 9, and 24 h.Extraction of total RNA was carried out using Trizol reagent(Life Technologies, Inc., Grand Island, NY, USA) accordingto the manufacturer’s instructions. DNA was removed byDNase. RNA (2 mg) was reverse-transcribed with randomprimers (Life Technologies Inc.) and Superscriptt II reversetranscriptase (Life Technologies Inc.). The cDNA was used forreal-time PCR.

Real-time RT–PCR

Real-time PCR was performed on the iCycler iQt system (Bio-Rad, Hercules, CA, USA). The PCR product is measured via afluorescent signal generated by the binding of the fluorophoreSybrGreen during the PCR. The amplification was followed by40 cycles with 941C for 20 s, 601C for 10 s, 651C for 25 s, and afluorescence read step at 831C for 25 s. We used the GAPDHcDNA to normalize the expression data of Bcl-2 and Bcl-xl.The gene expression in control sample was given 100%, andthe PS-341-treated samples were compared to the controlsample. GAPDH primers: gagtcaacggatttggtcg (forward),tggaagatggtgatggga (reverse); Bcl-2 primers: ccccgttacttttcctctg(forward), ccctctgcgacagcttata (reverse); Bcl-xl primers:ggagctggtggttgactttc (forward), ctccgattcagtcccttctg (reverse).All experiments were carried out in triplicate.

Transfections and reporter assays

The NF-kB reporter construct (pGL3-NF-kB) containing fourcopies of the NF-kB site was cloned into the pGL3-basicplasmid (Promega) (generous gift of Moshe Arditi, MD,Department of Pediatrics Infectious Diseases, Cedars-SinaiMedical Center). T98G cells were placed into 24-well platesand incubated until they reached 60–80% confluence. Cellswere transfected with pGL3-NF-kB and pRL-SV40-Luciferase(Renilla luciferase) plasmids using the Geneporter transfectionreagent (Qiagen, Valencia, CA, USA). After 3 h, cells wereexposed to PS-341 (10�7 M, 24 h). Luciferase activity in celllysates was measured by Dual Luciferase assay system(Promega, Madison, WI, USA), and this was normalized byrenilla activity.

IKB

Apoptotic stimuli (TNF, Trail)pr

otea

som

e

NFkB

nuclear

Apoptosis pathway

Ub-Ub-Ub-

Apoptosis

PS-341

Cell survival, proliferation

Figure 11 Drug interactions between the PS-341 and either tumornecrosis factor (TNF) a or TNF-related apoptosis-inducing ligand(TRAIL) in GBM. Both TRAIL and TNFa bind to their respectivereceptors, and activate the caspase apoptotic pathway, as well as,stimulate NF-kB activation (Franco et al., 2001; Dai et al., 2003).Activation of NF-kB can attenuate the apoptosis induced by thestimuli (Wang et al., 1998). The proteasome inhibitor, PS-341,blocks degradation of IkB and thus inhibits the nuclear transloca-tion of NF-kB; therefore, the activation of the caspase pathway isunfettered by the NF-kB progrowth effects. (Franco et al., 2001;An et al., 2003)


353

Oncogene

Statistical analysis

Differences between the results of experimental treatmentswere evaluated by the t-test. Differences were consideredsignificant at values of Po0.05.

Abbreviations

GBM, glioblastoma multiforme; SSCP, single-strand confor-mation polymorphism; PARP, poly ADP-ribose polymerase;MM, mutiple myeloma; MTT, 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl tetrazolium bromide; TNFa, tumor necrosisfactor alpha; TRAIL, TNF-related apoptosis-inducing ligand;

wt, wild type; mt, mutant; PS-341, Velcade/bortezomib; PI,propidium iodide.

Acknowledgements

This work was supported in part by National Institutes ofHealth, Ko-So Foundation, Parker Hughes Trust, and theInger Fund, as well as a grant from the Maxine DunitzNeurosurgical Institute, Cedars-Sinai Medical Center. HPK isa member of the Molecular Biology Institute and JonssonComprehensive Cancer Center at UCLA, and holds theendowed Mark Goodson Chair of Oncology Research atCedars-Sinai Medical Center/UCLA School of Medicine. Thiswork is in loving memory of Matt Shreck.

References

Adams J, Palombella VJ and Elliott PJ. (2000). Invest. NewDrugs, 18, 109–121.

Adams J, Palombella VJ, Sausville EA, Johnson J, Destree A,Lazarus DD, Maas J, Pien CS, Prakash S and Elliott PJ.(1999). Cancer Res., 59, 2615–2622.

An J, Sun YP, Adams J, Fisher M, Belldegrun A and RettigMB. (2003). Clin. Cancer Res., 9, 4537–4545.

Cusack JCJ, Liu R, Houston M, Abendroth K, Elliott PJ,Adams J and Baldwin ASJ. (2001). Cancer Res., 61,

3535–3540.Dai Y, Rahmani M and Grant S. (2003). Oncogene, 22,7108–7122.

Franco AV, Zhang XD, Van Berkel E, Sanders JE, Zhang XY,Thomas WD, Nguyen T and Hersey P. (2001). J. Immunol.,166, 5337–5345.

Frankel A, Man S, Elliott P, Adams J and Kerbel RS. (2000).Clin. Cancer Res., 6, 3719–3728.

Gallou C, Beroud C and Soussi T. (2001). http://p53.curie.fr.Hayashi S, Yamamoto M, Ueno Y, Ikeda K, Ohshima K,Soma G and Fukushima T. (2001). Neurol. Med. Chir.(Tokyo), 41, 187–195.

Hideshima T, Chauhan D, Richardson P, Mitsiades C,Mitsiades N, Hayashi T, Munshi N, Dang L, Castro A,Palombella V, Adams J and Anderson KC. (2002). J. Biol.Chem., 277, 16639–16647.

Hideshima T, Mitsiades C, Akiyama M, Hayashi T, ChauhanD, Richardson P, Schlossman R, Podar K, Munshi NC,Mitsiades N and Anderson KC. (2003). Blood, 101,

1530–1534.Hideshima T, Richardson P, Chauhan D, Palombella VJ,Elliott PJ, Adams J and Anderson KC. (2001). Cancer Res.,61, 3071–3076.

Imai Y, Strohmeyer TG, Fleishhacker M, Slamon DJ andKoeffler HP. (1994). Mod. Pathol., 7, 766–770.

Johnson TR, Stone K, Nikrad M, Yeh T, Zong WX,Thompson CB, Nesterov A and Kraft AS. (2003). Oncogene,22, 4953–4963.

Kitagawa H, Tani E, Ikemoto H, Ozaki I, Nakano A andOmura S. (1999). FEBS Lett., 443, 181–186.

Kuerbitz SJ, Plunkett BS, Walsh WV and Kastan MB. (1992).Proc. Natl. Acad. Sci. USA, 89, 7491–7495.

LeBlanc R, Catley LP, Hideshima T, Lentzsch S, MitsiadesCS, Mitsiades N, Neuberg D, Goloubeva O, Pien CS,Adams J, Gupta D, Richardson PG, Munshi NC andAnderson KC. (2002). Cancer Res., 62, 4996–5000.

Ling YH, Liebes L, Jiang JD, Holland JF, Elliott PJ, Adams J,Muggia FM and Perez-Soler R. (2003). Clin. Cancer Res., 9,1145–1154.

Ling YH, Liebes L, Ng B, Buckley M, Elliott PJ, Adams J,Jiang JD, Muggia FM and Perez-Soler R. (2002). Mol.Cancer Ther., 1, 841–849.

Lopes UG, Erhardt P, Yao R and Cooper GM. (1997). J. Biol.Chem., 272, 12893–12896.

Mitsiades N, Mitsiades CS, Poulaki V, Chauhan D, FanourakisG, Gu X, Bailey C, Joseph M, Libermann TA, Treon SP,Munshi NC, Richardson PG, Hideshima T and AndersonKC. (2002). Proc. Natl. Acad. Sci. USA, 99, 14374–14379.

Munker R, Norman A and Koeffler HP. (1986). J. Clin.Invest., 78, 424–430.

Nagai S, Washiyama K, Kurimoto M, Takaku A, Endo S andKumanishi T. (2002). J. Neurosurg. JID-0253357, 96, 909–917.

Nawrocki ST, Bruns CJ, Harbison MT, Bold RJ, Gotsch BS,Abbruzzese JL, Elliott P, Adams J and McConkey DJ.(2002). Mol. Cancer Ther., 1, 1243–1253.

Orlowski RZ, Stinchcombe TE, Mitchell BS, Shea TC,Baldwin AS, Stahl S, Adams J, Esseltine DL, Elliott PJ,Pien CS, Guerciolini R, Anderson JK, Depcik-Smith ND,Bhagat R, Lehman MJ, Novick SC, O’Connor OA andSoignet SL. (2002). J. Clin. Oncol., 20, 4420–4427.

Shinohara K, Tomioka M, Nakano H, Tone S, Ito H andKawashima S. (1996). Biochem. J., 317 (Part 2), 385–388.

Sunwoo JB, Chen Z, Dong G, Yeh N, Crowl BC, Sausville E,Adams J, Elliott P and Van Waes C. (2001). Clin. CancerRes., 7, 1419–1428.

Tani E, Kitagawa H, Ikemoto H and Matsumoto T. (2001).FEBS Lett., 504, 53–58.

Teicher BA, Ara G, Herbst R, Palombella VJ and Adams J.(1999). Clin. Cancer Res., 5, 2638–2645.

Verheij M, Bose R, Lin XH, Yao B, Jarvis WD, Grant S,Birrer MJ, Szabo E, Zon LI, Kyriakis JM, Haimovitz-Friedman A, Fuks Z and Kolesnick RN. (1996). Nature,380, 75–79.

Wagenknecht B, Hermisson M, Eitel K and Weller M. (1999).Cell Physiol. Biochem., 9, 117–125.

Wang CY, Mayo MW, Korneluk RG, Goeddel DV andBaldwin ASJ. (1998). Science, 281, 1680–1683.

Wu X and Levine AJ. (1994). Proc. Natl. Acad. Sci. USA, 91,3602–3606.

Yang Y, Ikezoe T, Saito T, Kobayashi M, Koeffler HP andTaguchi H. (2004). Cancer Sci., 95, 176–180.


354

Oncogene

Proteasome inhibitor PS-341 causes cell growth arrest and apoptosis in human glioblastoma multiforme...

Documents

Transcript of Proteasome inhibitor PS-341 causes cell growth arrest and apoptosis in human glioblastoma multiforme...