head and neck squamous cell growth suppression using...

Vol. 3. 185-191, February 1997 Clinical Cancer Research 185

Head and Neck Squamous Cell Growth Suppression Using

Adenovirus-p53-FLAG: A Potential Marker for Gene

Therapy Trials’

S. Mark Overholt,2 Ta-Jen Liu,2

Dorothy L. Taylor, Mary Wang,

Adel K. El-Naggar, Ed Shillitoe,

Karen Adler-Storthz, Lisa St. John,

Wei-Wei Zhang, Jack A. Roth, and

Gary L. Clayman3

Departments of Head and Neck Surgery [S. M. 0.. T-J. L., D. L. 1.,

M. W.. G. L. Cl, Pathology [A. K. E-N.l, Section of Thoracic

Molecular Oncology. and Department of Thoracic and Cardiovascular

Surgery IW-W. Z.. J. A. Ri. The University of Texas M. D. Anderson

Cancer Center, Houston, Texas 77030; Department of Microbiology

and Immunology. State University of New York. Syracuse, New

York 13210 IE. 5.1: Department of Basic Sciences. The University of

Texas Health Science Center Dental Branch, Houston, Texas 77030

1K. A-SI: and University of Texas. Houston, Graduate School of

Biomedial Sciences, Houston. Texas 77030 IL. S. J.I

ABSTRACT

The recombinant wild-type p53 adenovirus has been

proven effective against the growth of human head and neck

squamous cell cancer (SCCHN) cell lines in vitro and in a

nude mouse model. The addition of a FLAG peptide se-

quence was used in this study, along with the p53 adenovirus

vector as a marker of the site of the gene therapy activity. It

provides clear evidence of the exogenous gene product

within the transduced carcinoma cells. No alterations in

transcription or translation of the p53 gene product were

noted with the addition of the FLAG sequence to the original

p53 adenovirus vector. Immunohistochemical analysis dis-

played simultaneous expression of the p53 and FLAG pro-

teins in the infected cells. The p53 protein remained local-

Received 9/16/96: revised I 1/7/96: accepted I 1/8/96.

The costs of publication of this article were defrayed in part by the

payment of page charges. This article must therefore be hereby marked

advertisement in accordance with 18 U.S.C. Section 1734 solely toindicate this fact.

I This work was supported in part by American Cancer Society Career

Development Award 93-9 (to G. L. C.): M. D. Anderson Cancer Center

Core Grant NIH-National Cancer Institute CA- I6672: National CancerInstitute Grant ROl CA-45l87 (to J. A. R.): Training Grant CA0961 I

(to J. A. R.); National Institute Grant RS5DE/OD10846 (to K. A-S.):National Institute Grant ROl DE-l0842 (to E. S.): National Science

Foundation Graduate Research Fellowship (to L. S. J.): gifts to the

Division of Surgery from Tenneco and Exxon for the Core Lab Facility:

and a generous gift from the Mathers Foundation.2 Contributed equally to the concept. investigation. and writing of this

work.

3To whom requests for reprints should be addressed. at Department ofHead and Neck Surgery. The University of Texas M. D. Anderson

Cancer Center, I 5 15 Holcombe Boulevard. Box 69, Houston, TX

77030. Fax: (713) 794-4662.

ized to the nucleus, whereas the FLAG protein was

additionally noted in the cytoplasm. In vitro growth suppres-

sion assays and in vivo microscopic residual tumor model

experiments in nude mice showed a similar tumoricidal

effect with the p53-FLAG adenovirus vector to that with the

previously studied p53 adenovirus vector without the addi-

tion of the FLAG sequence. We conclude that the addition of

the FLAG octapeptide sequence allows identification of

those cells that have been affected by the molecular therapy

independent of the endogenous gene expression of the cells.

This novel molecular tracer may prove useful in character-

izing infection efficiency and in gene therapy trials.

INTRODUCTION

Locoregional control of disease, despite improvement over

the past several decades, continues to impact significantly on

patients with SCCHN.4 Distant metastases fortuitously occur

late in the course of these tumors and. overall, occur in only

approximately I I % ( I ). Although long-term goals remain fo-

cused on the complicated sequence of molecular events leading

to carcinoma, the immediate goals that challenge researchers are

2-fold: (a) detect changes in epithelium before overt malignancy

occurs; and (b) develop strategies to eradicate local and regional

disease with minimal host morbidity. Brennan et ai. (2) have

recently described the use of p53 satellite mapping around the

resection margins of SCCHNs. They demonstrated that muta-

tions of the p53 gene in the margins of resection increased the

risk of local recurrence. Boyle et a!. (3) found that p53 muta-

tions precede invasion in SCCHN. Mutations of the p53 gene

are present in up to 60% ofSCCHNs (4-8). As new markers for

premalignancy develop with greater site and tumor specificity

and sensitivity. we may be able to predict early progression

toward malignancy.

Gene therapy strategies have shown promise in controlling

local and regional tumor burden in both in vitro and in animal

models (9-15). Specifically. in a head and neck squamous cell

carcinoma. growth suppression and cytotoxicity have been doe-

umented using the wild-type p53 gene delivered via a replica-

lion-defective adenovirus vector (9). A dose-dependent tumori-

cidal effect independent of the host endogenous p53 status has

additionally been shown. The mechanism of action appears to be

apoptosis (16. 17). Furthermore. a recent microscopic residual

disease model that mimics postsurgical resection situations in

humans has been described ( 10). Such a model lays the ground-

work for carryover application in a human trial.

4 The abbreviations used are: SCCHN, head and neck squamous cell

carcinoma: Ad, adenovirus: CMV. cytomegalovirus: pfu. plaque-form-

ing unit(s): TNF. tumor necrosis factor.

Research. on July 14, 2018. © 1997 American Association for Cancerclincancerres.aacrjournals.org Downloaded from

http://clincancerres.aacrjournals.org/

186 p53-FLAG: Potential Marker for Gene Therapy Trials

As more promising therapies develop and we entertain the

application of gene therapy in the head and neck, the ability to

trace the sites affected by the therapy and identify which cells

have been affected by treatment becomes more important. De-

livery of a novel tracer would, therefore, provide definitive

evidence for infection efficiency and extent of effect, both

locally and distantly.

The FLAG biosystem (18), originally described in 1988,

allows one to tag and examine the eft’eet of an exogenously applied

treatment. Originally, it was used in the purification of large pro-

teins, but its design lends itself to other applications. This strategy

generates a fusion protein of the FLAG peptide and the gene of

interest. A monoclonal antibody against the polypeptide enables us

to detect the location of a fusion protein that would correspond to

clonal expression of the delivered gene. The polypeptide is eight

amino acids in length, and its small size should not disrupt the

expression of the delivered gene therapy protein. If the biological

activity remains unchanged, the ability to co-deliver such a marker

holds great implication for future human trials. The goal of this

investigation was to compare the biological effect of the previously

described wild-type pS3 adenovirus to that of the vector modified

with the FLAG octapeptide sequence. We demonstrated that the

biological effect was not changed by the addition of the FLAG

sequence to the adenovirus p53 vector (AdCMV-p53), and that by

using immunohistochemical techniques, we have the ability to trace

the location of the effect following the adenovirus administration.

MATERIALS AND METHODS

Cell Lines and Culture Conditions. Human SCCHN

cell lines Tu-l 38 and 686-LN were both established by the De-

partment of Head and Neck Surgery. The University of Texas

M. D. Anderson Cancer Center, and have been characterized pre-

viously (19, 20). These cells were grown in DMEM/Fl2 medium

supplemented with 10% heat-inactivated fetal bovine serum with

streptomycin/penicillin at 37#{176}Cand 5% carbon dioxide.

Recombinant Adenovirus Preparation and Infection.

The recombinant p53 adenovirus (AdCMV-p53) contains the

CMV promoter, wild-type p53 eDNA, and SV4O polyadenyl-

ation signal in a minigene cassette inserted into the El-deleted

region of modified AdS (21). Viral stocks were propagated in

293 cells. Cells were harvested 36-40 h after infection, pel-

leted, resuspended in PBS, and lysed. Cell debris was then

removed by subjecting the cells to CsCl gradient purification.

Concentrated virus was dialyzed, aliquoted, and stored at

-80#{176}C.Infection was carried out by the addition of virus to the

DMEMIFI2 medium with 10% fetal bovine serum to the cell

monolayers. The cells were incubated at 37#{176}Cfor 60 mm with

agitation. Then complete medium (DMEMIF12 with 10% fetal

bovine serum) was added to the cells in the appropriate volume

for the given Petri dish and incubated for the desired time.

Generation of the p53-FLAG Adenovirus. The p53

eDNA sequence was excised from the pCS3-SN [kindly pro-

vided by Dr. G. Lozano (The University of Texas M. D. Ander-

son Cancer Center, Houston, IX)1 by digestion with BwnHI and

cloned into the BainHI site of pGEM7Z. A recombinant plasmid

with the proper insert orientation was then digested with AccI

and Kpnl to remove 21 amino acids from the 3’ end of p53

eDNA. A linker with AccI-KjmI compatible ends containing the

sequence of the FLAG peptide including a stop codon was then

ligated into the digested plasmid to create the p53-FLAG fusion

gene. The resulting p53-FLAG fusion gene was then cloned into

an expression vector with the human CMV promoter and SV

polyadenylation signal. The final construct was subsequently

inserted into a shuttle vector pXCJL. I (2 1 ) to generate a recom-

binant p53-FLAG adenovirus.

Northern Blot Analysis. Total RNA was isolated by the

acid-guanidinium thiocyanate method of Chomczynski and Sacehi

(22). Northern analyses were performed on 20 �.tg oftotal RNA run

on a 1% agarose gel containing 2.2 M formaldehyde. The mem-

brane was hybridized overnight at 65#{176}Cwith a p53 eDNA probe

labeled by the random primer method in 5 X SSC, 5 X Denhardt’s

solution, 0.5% SDS denatured salmon sperm DNA (20 j.tg/ml).

Western Blot Analysis. Total cell lysates were prepared

by sonicating the cells 24 h after infection in RIPA buffer ( 150

mM NaC1, 1.0% NP4O, 0.5% sodium deoxycholate. 0.1% SDS,

and 50 filM Iris, pH 8.0). Fifty �.tg of protein from samples were

subjected to 10% PAGE and transferred to a Hybond-ECL

membrane (Amersham Corp.). The membrane was blocked with

Blotto/Tween (5% nonfat dry milk and 0.2% Tween 20 in PBS)

and probed with the primary antibodies. anti-FLAG M2 mono-

clonal antibody (IB13002), mouse anti-human p53 monoclonal

antibody (PAb I 80 1 ), and mouse anti-human B-actin mono-

clonal antibody (Amersham). The secondary antibody, horse-

radish peroxidase-conjugated goat anti-mouse IgG (Boehringer

Mannheim, Indianapolis, IN), was then exposed to the mem-

brane. The membrane was then processed and developed as the

manufacturer specified.

Cell Growth Assay. Cells were plated at a concentration

of 2 X l0� cells/mI in six-well plates in triplicate, and cells were

infected with AdCMV-p53-FLAG, AdCMV-p53. or the repli-

cation-defective adenovirus control (DL3I2), at an multiplicity

of infection of I 00. A mock infection was also used as a control,

against which to compare the effect of the replication-deficient

adenovirus and AdCMV-p53. Cells were then trypsinized and

counted at specified time points. Viability was checked using

trypan blue exclusion.

Immunohistochemical Staining of in Vitro Cell Layers.

Infected cell monolayers were fixed with 3.8% buffered forma-

lin and treated with H,O� in methanol for 5 mm. Immunohis-

tochemical staining was performed using the Vectastain Elite kit

(Vector Laboratories, Burlingame. CA). The primary antibody

was either the mouse anti-FLAG M2 monoclonal antibody

(IB13002) or mouse anti-human p53 monoclonal antibody

(PAbl8Ol), and the secondary antibody was an avidin-labeled

anti-mouse IgG (Vector). The biotinylated horseradish peroxi-

dase avidin-biotin complex reagent was used to detect the anti-

gen-antibody complex. Preadsorption controls were used in each

experiment. The cells were then counterstained with Harris

hematoxylin (Sigma Chemical Co., St. Louis, MO).

Immunohistochemical Staining of Tissue from in Vivo

Studies. Formalin-fixed, paraffin-embedded in vito animal

experimental tissues were cut at 4-5 p.si. dried at 60#{176}C.depar-

af’finized, and hydrated with distilled water. Sections were then

treated with 0.5% saponin in distilled water and rinsed in several

changes of distilled water; endogenous peroxidase activity was

blocked with 3% hydrogen peroxide in methanol, followed by

rinsing in several changes of distilled water. Sections were



1234567 89

2.8 Kb

1.9 Kb

Endogenous

Exogenous

Clitiical Cancer Research 187

Fig. I Expressiots of exogenous p53

mRNA 24 h after infection with

AdCMV-p53 and AdCMV-p53-

FLAG. La,u’ I. RNA molecular

weight marker. LAUl(’S’ 2-5, Tv- I 38

(niutated PS3) cell line. Lane.s 6-9.

686-LN (wild-type PS3) cell line.

Laiies 2 and 6. mock infection. Lanes

3 and 7. DL3 I 2 infection. Lotus 4 and

8. AdCMV-p53 infection. Lanes’ 5 and

9. AdCMV-p53-FLAG infection.

microwave-irradiated in distilled water for 3 mm using a Sharp

model R9H8 I microwave oven operating at a frequency of 2450

MHz at 700 W. After cooling, sections were washed in several

changes of distilled water and placed in PBS; immunochemical

studies were performed by using the avidin-biotin-peroxidase

complex method of Hsu ci ai. (23) in the following manner.

Sections were blocked with normal horse serum and incubated

overnight at 4#{176}Cwith anti-FLAG M2 monoclonal antibody at a

concentration of 60 p.g/ml (International Biotechnologies, New

Haven, CT) and rabbit antihurnan p53 polyclonal antibody,

clone OM-1. 1:80 (Signet Laboratories, Denham. MA). An

antirabbit IgG Elite kit (Vector) was then used to apply bioti-

nylated antirabbit IgG and avidin-biotin-peroxidase complexes

that were incubated for 45 mm each. The immunostaining

reaction was visualized by using 0.5% dimethylaminoazoben-

zene in PBS containing 0.01% H,02 (pH 7.6), counterstained

with 0.01% toluidine blue. dehydrated. cleared. and mounted in

Permount.

In Vivo Microscopic Residual Disease Experiments. In

t’ivo experiments were performed in a defined pathogen-free

environment using athymic nude mice. The microscopic resid-

ual disease model described previously by our laboratory was

applied ( 10). Experiments were reviewed and approved by in-

stitutional review committees for both animal care and utiliza-

tion and the Biosafety Committee for recombinant DNA re-

search. Briefly, 4-7-week-old female nude mice were

anesthetized using i.p. ketamine/acepromazine at a dose of 70

mg/kg of body weight. The bodies were prepared with alcohol

wipes and incisions were made on four dorsal flanks. s.c. pock-

ets were elevated, and the desired number of tumor cells was

delivered in 100 pi of medium into the pocket. The flap was

then sealed with an interrupted horizontal mattress 5.0 nylon

suture, insuring a water-tight seal. At 48 h, the flaps were

reopened, and 100 ii of an appropriate concentration of virus

were delivered into the same bed inoculated previously with the

tumor cells. Two different sets of repeated experiments were

performed. The first was a dose-response experiment using the

AdCMV-p53-FLAG virus in three of the flaps at descending

concentrations ( l0�, l0�, 106 pfu). The fourth flap served as a

control and was randomized to either PBS or the replication-

defective adenovirus (DL312). The second study was performed

using l0� pfu of AdCMV-p53-FLAG. AdCMV-p53, and repli-

cation-defective adenovirus in three separate flaps. The fourth

flap was inoculated with the same volume ( 100 p.1) of sterile

PBS. Forty-eight h after treatment with virus, two animals were

sacrificed, and the flaps were harvested for immunohistochem-

ical analysis. The remaining animals were observed for 2 1 days

and then sacrificed. Tumor volumes were measured for corn-

parison using calipers.

RESULTS

Expression of mRNA after Infection with AdCMV-p53

and AdCMV-p53-FLAG Virus. Both To- I 38 and 686-LN

were examined for expression of p53 mRNA. Total RNA was

isolated after adenovirus infection. Northern blot analysis was

performed. Similar levels of exogenous AdCMV-p53 rnRNA

were detected between AdCMV-p53- and AdCMV-p53-FLAG-

infected cells. Fig. I shows the comparable levels of p53 mRNA

expression af’ter infection with AdCMV-p53 and AdCMV-p53-

FLAG in Lanes 4 and S for Iu-138 and in Lanes 8 and 9 for

MDA 686-LN. Variation of intensity is felt to be related to

loading dose. Endogenous expression of p53 rnRNA is seen in

Lanes 2 and 3 in the mutated p53 cell line, Iu-138. Lanes 6 and

7 show no endogenous p53 rnRNA in the 686-LN cell line,

which is wild-type for the p53 gene. These data suggest that the

AdCMV-p53-FLAG virus, like the AdCMV-p53 virus, is sue-

cessfully transduced and efficiently transcribed. Northern anal-

ysis did not reveal evidence of AdCMV-p53 DNA contarnina-

tion.

Expression of Exogenous p53 Protein in AdCMV-p53-

and AdCMV-p53-FLAG-infected SCCHN Cell Lines.

Western blot analysis was performed to compare the amount of

protein expressed by the AdCMV-p53- and AdCMV-p53-

FLAG-infected cells. Protein bands were identified using the

monospecifie p53 antibody (PAbl8Ol) and the anti-FLAG M2

antibody (1B13002) on two simultaneously run gels. Fig. 2 (top)

showed, using the p53 antibody (PAbl8Ol ), a similar high level

of p53 protein expression in both cell lines that were infected

with the AdCMV-p53 and AdCMV-p53-FLAG. Lanes 3 and 7

correspond to the To- I 38 and MDA 686-LN cells infected with

AdCMV-p53, respectively. Lanes 4 and S of the blot represent

protein expression from those cells infected with the AdCMV-

p53-FLAG. No change in p53 protein expression was noted in

either the replication-defective, adenovirus-infeeted cells or in

the mock group. The bottom blot demonstrates a similarly

executed gel that was probed with the mouse anti-FLAG M2

antibody. The level of p53-FLAG protein expression appeared

to be similar to that expressed following p53 antibody probing.

but no detectable band was noted in those cells infected with the

AdCMV-p53 virus (Fig. 2. lanes 3 and 7). The mock and

DL3I2-infeeted cells exhibited no detectable level of the im-

munoreactive p53 protein in either cell line.



A

0 2

B

6 8

tn

L

.nE

2:

LI

Days

Fig. 3 Inhibition of SCCHN cell growth in vitro. A, Tu-138; B, 686-

LN. At each indicated time point, three dishes of cells were trypsinized

and counted. The mean cell counts per triplicate wells following infec-

tion were plotted against the number of days since infection; bars, SE.

0, mock; A, DL312; EL PS3; U, p53-FLAG.

2 4 6 8


A

B

I 2345678

�*p53

,�. actin

.� FLAG

..�- actin

Fig. 2 Expression of p53 and

p53-FLAG fusion protein from

cellular extracts isolated 24 h

after infection with AdCMV-

p53 or AdCMV-p53-FLAG.

Lanes 1-4, Tu-138 (mutated

pS3) cell line; Lanes 5-8,

686-LN (wild-type p53) cell

line. Lanes I and 5, mock in-

fection. Lanes 2 and 6, DL3 12infection. Lanes 3 and 7,

AdCMV-p53 infection. Lanes

4 and 8, AdCMV-p53-FLAGinfection. A, probed with p53(PAbl8Ol) antibody; B,

probed with FLAG M2 anti-

body.

Effect of AdCMV-p53 and AdCMV-p53-FLAG on

SCCHN Cell Growth. We have described previously the

cytotoxic effect of p53 therapy in Tu-l38 and 686-LN cell lines.

The Iu-l38 cell line has an endogenously mutated p53 gene,

and the 686-LN cell line possesses the wild-type p53 gene. This

study sought to determine if any difference in efficacy would be

seen after manipulation of the AdCMV-p53 virus by inserting

the FLAG sequence.

Cells infected with the replication-defective adenovirus had a

similar growth rate to the mock infection cells. A mild cytotoxic

effect may be seen with the replication-defective adenovirus (Fig.

3). In contrast, those cells infected with either the AdCMV-p53 or

AdCMV-p53-FLAG experienced virtual total tumor cell death by

day three. Histological examination revealed bleb formation by the

plasma membrane, which is one of the characteristic features of

apoptosis. As noted previously in our laboratory, the effect was

more prominent for the Tu-138 cell line (mutated p53) than it was

for the MDA 686-LN cell line (wild-type p53). Growth curve

assays were reproducible in three repeated experiments without a

significant difference being noted between the effect of the Ad-

CMV-p53 and the AdCMV-p53-FLAG viruses, suggesting that the

addition of the FLAG peptide did not affect the ability of p53 in

suppression of cell growth.

Immunohistochemical Staining of SCCHN Cell Lines

Infected with Adenovirus. Infected cell monolayers were corn-

pared for expression of the p53 arid p53-FLAG protein using

standard immunohistochemical techniques. As controls, both the

mock-infected cells and the DL3 12-infected cells showed similar

staining (Fig. 4). Neither the p53 nor FLAG protein could be

clearly identified in the control group. However, when cells were

infected with the AdCMV-p53 virus, a strong staining was noted.

Those cells infected with the AdCMV-p53-FLAG virus showed

virtually identical intensity of staining and number of positive cells

with PAbl8Ol for antibody as compared to the cells infected with

the AdCMV-p53 virus. The cells infected with the AdCMV-p53-

FLAG virus also showed strong immunohistochemical positivity

with the M2 FLAG antibody. The quality of staining was different,

although both were within the nucleus and to a lesser degree in the

cytoplasm.

In Vivo Suppression of Growth. Dose-response studies

using 106, l0�, and 108 pfu of AdCMV-p53-FLAG virus com-

pared to a control flap that was either PBS or DL3 12 were

performed using the microscopic model method (10). The mean

tumor size for the mock infection was 1205 mm (3). Tumor size

decreased in a linear fashion with increasing concentration of

virus used in the molecular intervention. Mean tumor size was



Clinical Cancer Research 189

A .. .‘ . 1’ ;-: .� #{149}..

‘, I � #{149}� � �‘ �, �, ‘ -

- � �A � �

� �‘ ,‘-.-�.. -- .. ,1 #{149}� � :‘,,.�‘T)I.,..

-‘. . I ‘�: � � �1

.(a. � � .‘s%�., ‘#{149}-.

. s-... .-‘

‘4 .% . : . L��#{149} #{149}‘:�� . �:

0#{149}0

0�

�

c� �

#{149}.:‘� � � g’:�t-��e .

Fig. 4 In vitro immunohistochemical staining of 686-LN cells infected

with AdCMV-p53-FLAG. A, endogenous p53 staining. B, p53 antibody

(PAbl8Ol). C, FLAG M2 antibody (1B13002). B, nuclear p53 staining

is prominent. C’, cytoplasmic component to staining noted with the

FLAG antibody. x 100.

637 mm (3), 392 mm (3), and 193 mm (3) for those flaps treated

with 106, l0�, and l0� pfu of the AdCMV-p53-FLAG, respec-

tively. Each animal was compared against itself using a paired

t test, and a significant dose-response effect was noted at P <

0.05 in all comparisons, except between the flap treated with l0�

and l0� pfu. Clearly, the greater the amount of virus, the greater

the tumor growth inhibition. In an additional study, the in vito

effects of AdCMV-p53 were compared to that of AdCMV-p53-

FLAG. No difference in growth suppression activity was noted

(data not shown).

Immunohistochemical Demonstration of Exogenous

Tumor Suppression Effect in the Microscopic Residual Dis-

ease Animal Model. After proving comparable in vitro and in

t,ivo activity of AdCMV-p53 and the AdCMV-p53-FLAG. we

applied immunohistoehemical techniques to demonstrate p53-

FLAG fusion protein product in vito. Microscopic residual

disease flaps were harvested 48 h after treatment, fixed in

formalin, and paraffin embedded. On neighboring sections of

tumor cells treated with the AdCMV-p53-FLAG virus, staining

for both the p53 and FLAG protein was applied. Staining

intensity and the number of cells staining positively was directly.� . proportional to the amount of virus used in the infection. Con-

trols were negative for staining with both p53 and FLAG anti-

bodies. Fig. S shows a histological specimen stained with H&E

(A) the p53 antibody (B) and the FLAG antibody (C). The

characteristic intranuclear staining with the p53 antibody is

similarly expressed as the intranuclear staining with the FLAG

throughout multiple layers of the gene transfer site. This also

demonstrates that the FLAG M2 antibody is effective on paraf-

fin-embedded fixed tissue. The staining demonstrates that the

tumor-suppressive effect is directed by the exogenous therapy,

and that in an in vito model, one can identify the exogenous

therapy using the applied FLAG system. lime course protein

expression experiments using Western blotting show peak pro-

tein at 3 days and low detectable levels at 15 days. When

animals were sacrificed at 21 days, tissue sections were, there-

fore, uninformative using immunohistoehemical methods (data

not shown).

DISCUSSION

The tumor suppressor gene p.53 has been shown previously

to be an effective molecular therapy against SCCHN in vitro and

in a nude mouse model (10). In addition, another gene therapy

strategy using the HSV-TK gene has demonstrated effectiveness

in this epithelial carcinoma (24). Despite rapid progress directed

toward the development of gene therapy, an effective unique

marker, which may establish the transduction oftargeted tissues,

has not been established. A potential marker gene product is the

FLAG peptide. So et ai. (25) used this method to isolate a

TNF-a protein. The FLAG M2 monoclonal antibody was ap-

plied to identify the FLAG-INF-a fusion protein. Then the

FLAG peptide was cleaved using enterokinase, leaving the

specific NH2 terminus to the TNF-a molecule. Importantly, the

fusion FLAG-TNF-a protein and the TNF-a protein alone had

similar biological function in cytotoxic assays. We hypothesized

that we could co-deliver the p53-FLAG fusion gene via an

adenovirus vector, identify it using the FLAG M2 monoclonal

antibody, and express the tumor suppressor gene product, p53.

At the same time, identification of those cells transduced by the

virus could be identified independently of the endogenous p53

expression of the infected cells by determining the expression of

the FLAG gene product.

Initial studies sought to establish the induction of p53

mRNA following infection with either the AdCMV-p53 or

AdCMV-p53-FLAG and clearly demonstrate similar exogenous

mRNA induction. Western blot analysis of p53 and FLAG

protein expression demonstrated similar levels of p53 protein

expression in cells infected with the AdCMV-p53 and AdCMV-



B� “ “ . �.

. .


‘� “ ‘ ‘ .‘:“ “.- ‘ ‘ . . “ � ,�4 ‘

.. . . ..‘ .. . . .. :: : , . . � � t.1��P;t�

� � .‘ ‘�. �

‘ ‘ . .. - .: ,‘. �

�. �‘. I �‘ .: . ; , . ,.- ., .. � .‘-.� , C ��“�‘�-\ A

�;J�.1C.-.’ ‘ ‘ ‘ : , � . , , .

w

� :‘�;‘ � �N�I�’ � :

� .

� ,, . ,‘.

� ‘.- -�‘ . ...

� � ‘� � � , �

Fig. 5 Immunohistochemical staining of formalin-fixed, paraffin-em-bedded tissue from in vito Tu-I38 tumor treated with AdCMV-p53-

FLAG. A, H&E. B, p53 antibody (PAbl8Ol). C. FLAG M2 antibody

(IBI3002). B, strong nuclear staining noted with p53 antibody. C. strong

nuclear staining is also seen utilizing the FLAG antibody. X 100.

p53-FLAG, respectively. Furthermore, the FLAG antibody

demonstrated that the protein expressed by the cells infected

with the AdCMV-p53-FLAG was a unique fusion protein ex-

pressing the FLAG octapeptide moiety that could be differenti-

ated from the native p53 protein.

Northern analysis confirmed that the cells infected with

either the AdCMV-p53 or AdCMV-p53-FLAG exhibited simi-

lar exogenous mRNA induction. Our previous studies using the

AdCMV-p53 had demonstrated marked induction of apoptosis

among infected cancer cell lines ( 16, 26); however, nonmalig-

nant cell lines were spared. The next question was whether the

addition of the FLAG sequence to the putative p53 protein

would alter its inherent growth-inhibitory effect. To this end.

cell growth assays were performed, and no significant difference

could be demonstrated between AdCMV-p53 and AdCMV-p53-

FLAG. As we had described previously, induction of apoptosis

was not observed following mock or replication-defective virus

infection.

In vitro immunohistochemical studies were performed to

establish that the FLAG protein could be identified among

infected SCCHN cell lines. We were able to show a strong

nuclear staining for the p53 protein in those cells infected by

either the AdCMV-p53 or the AdCMV-p53-FLAG. Further-

more, using the FLAG antibody, the novel FLAG octapeptide

sequence was demonstrated in those cells that were infected

with the AdCMV-p53-FLAG only. There appeared to be a slight

qualitative difference in the staining with a probable cytoplas-

mic component when probing with the FLAG monoclonal

antibody.

in t’it’() experiments were carried out to further explore the

efficacy of FLAG immunohistochemical staining. Using the

microscopic residual disease model described previously. s.c.

tumor sites were infected with increasing concentrations of the

AdCMV-p53-FLAG vector. A dose-response curve was noted

with greater tumor kill as the virus concentration was increased.

Flaps infected with AdCMV-p53-FLAG could be effectively

immunohistoehemically analyzed for both the p53 and FLAG

proteins using formalin-fixed, paraffin-embedded tissues. The

large pS3 protein was noted to be primarily intranuclear. as

documented previously. However, the FLAG protein was noted

additionally within the cytoplasm. Because the nuclear translo-

cation signal was altered in the p53 3’-end deletion in this

FLAG construct, both nuclear and cytoplasniic localization of

the transgene product were found (27). We were not able to

demonstrate staining for the FLAG protein outside the region of

treatment, confirming the local expression of these virions and

lack of systemic expression when delivered in this model.

FLAG appears to be a potential marker for tracking protein

product expression in gene therapy. We currently have the

technology to use virus-specific promoters and PCR techniques

to analyze the presence ofdelivered virions. As a screening tool,

this would be an effective and sensitive test for identifyi�ig virus.

However, it does not answer the more important question: is the

protein product encoded by the delivered virus being expressed?

Herein lies the major advantage of the FLAG marker. Fixed

tissue can be analyzed for the presence of the delivered gene

product. This is particularly important in tumor biology, where

endogenous heterogeneity for the genes of interest might mask

locations where a specific gene therapy has been delivered.

Specifically, in tumors with a mutated p53 gene, exogenous

expression of the vector-delivered p53 can be differentiated

from the overexpression of the endogenous p53 of the tumor.

Such a marker would also demonstrate the expression and effect

imparted on normal bystander cells. Both of these lend them-

selves to carryover into human trials, where such questions must

be addressed.

Potential criticisms of FLAG as a marker for gene therapy

are 2-fold: (a) a minimal protein expression is required to



Clinical Cancer Research 191

immunohistochemically identify the gene product. For simple

documentation of the presence of virion, the sensitivity is far

less than that of PCR. However, the transduction efficiency of

the replication-defective adenovirus in SCCHN is excellent, and

p53 protein expression is high. The added advantage of docu-

menting expression of the viral product and its histological

location distinguishes this technique; (b) introduction of novel

genes has the potential to cause deleterious downstream effects.

Using a retroviral vector, these potential effects would be irre-

versible, but with the transient episomal expression of the ade-

novirus, this hazard is avoided. Lastly, the p53-FLAG and

p53-adenoviral vectors both show evidence of inflammatory

infiltrates in animal studies at viral doses of l0� pfu and greater.

Local inflammatory responses may, however, be beneficial in

the local tumor environment. The potential systemic immune

response from viral products and exogenous gene products will

require further investigations.

In conclusion, we feel that the co-delivery of the FLAG

protein along with the desired gene therapy offers potential

utility as a marker of gene therapy. We were able to show that

it was simultaneously promoted along with the p53 gene and

that expression of the mRNA and protein were not decreased.

More importantly, the biological activity of the delivered tumor

suppressor gene was not altered. For the first time, the FLAG

antibody was proven effective when immunohistoehemical

analysis was perf’ormed on formalin-fixed, paraffin-embedded

tissue. These factors suggest the utility of this novel protein as

a tracer in further gene therapy studies.

REFERENCES

1. Calhoun, K. H., Fulmer, P., Weiss, R., and Hokanson, J. A. Distant

metastases from head and neck squamous cell carcinomas. Laryngo-

scope, 104: 1199-1205, 1994.

2. Brennan, J. A., Mao, L., Hruban, R. H., Boyle, J. 0., Eby, Y. J..Koch, W. M., Goodman, S. N., and Sidransky, D. Molecular assessmentof histopathological staging in squamous-cell carcinoma of the head and

neck. N. EngI. J. Med., 332: 429-435. 1995.

3. Boyle. J. 0.. Hakim, J., Koch, W., van der Riet, P., Hruban, R. H.,

Roa, R. A., Correo, R., Eby, Y. J., Ruppert, M., and Sidransky, D. Theincidence of p53 mutation increases with progression of head and neck

cancer. Cancer Res., 53: 4477-4480, 1993.

4. Field, J. K.. Pavelic, Z. P., Spandidos, D. A.. Stambrook, P. J.. Jones.

A. S., and Gluckman. J. L. The role of the p53 tumor suppressor gene

in squamous cell carcinoma of the head and neck. Arch. Otolaryngol.

Head Neck Surg., 119: 1 1 18-1 122, 1993.

5. Watling, D. L., Gown, A. M., and Coltrera, M. D. Overexpression of

p53 in head and neck cancer. Head Neck, 14: 437-444, 1992.

6. Nylander. K., Stenling, R., Gustafsson, H., Zackrisson, B., and Root,G. p53 expression and cell proliferation in squamous cell carcinomas of

the head and neck. Cancer (Phila.), 75: 87-93, 1995.

7. Pavelic, Z. P.. Li, Y-Q., Stambrook, P. J., McDonald, J. S., Munck-

Wikland, E., Pavelic, K., Dacic, S., Danilovic, Z., Pavelic, L., Mugge,

R. E., Wilson, K., Nguyen, C.. and Gluckman, J. L. Overexpression of

p53 protein is common in premalignant head and neck lesions. Anti-

cancer Res., 14: 2259-2266, 1994.

8. Maestro, R., Doleetti, R., Gasparatto, D., Doglioni, C., Pelucchi, S.,

Barzan, L., Grandi, E., and Boiocchi, M. High frequency of p53 gene

alterations associated with protein overexpression in human squamous

cell carcinoma of the larynx. Oncogene, 7: 1 159-1 166, 1992.

9. Liu, T-J., Zhang, W-W., Taylor, D. L., Roth, J. A., Goepfert, H., and

Clayman. G. L. Growth suppression of human head and neck cancer

cells by the introduction of a wild-type p53 gene via a recombinant

adenovirus. Cancer Res., 54: 3662-3667, 1994.

10. Clayman, G. L., El-Naggar, A. K., Roth, J. A.. Zhang. W-W.,Goepfert. H.. Taylor. D. L., and Liu, 1-i. In vito molecular therapy withp53 adenovirus for microscopic residual head and neck squamous car-

cinoma. Cancer Res., 55: 1-6, 1995.

1 1 . Baker, S. J., Markowitz, S., Fearon, E. R., Wilson, J. R., and

Vogelstein, B. Suppression of human colorectal carcinoma cell growth

by wild-type p53. Science (Washington DC), 249.’ 912-915, 1990.

12. Mercer, W. E., Shields, M. 1., Amin, M., Sauve, G. J., Apella, E.,

Romano, J. W., and Ulrich, S. J. Negative growth regulation in aglioblastoma tumor cell line that conditionally expresses human wild-

Type p53. Proc. NatI. Acad. Sci. USA, 87: 6166-6170, 1990.

13. Diller, L., Kassel, J., Nelson, C. E., Gryka, M. A., Litwank, G.,

Gebhardt, M., Bressac, B., Ozturk, M., Baker, J. J., and Vogelstein, B.p53 functions as a cell cycle control protein in osteosarcoma. Mol. Cell.

Biol.. 10: 5772-5781. 1990.

14. Shaw, P., Bovey, R., Tardy, S., Sahli, R., Sordet, B., and Costa,

J. Induction of apoptosis by wild-type p53 in human colon tumor-

derived cell line. Proc. NatI. Acad. Sci. USA, 89: 4495-4499, 1992.

15. Fujiwara. 1., Grimm, E. A., Murkhopadhyay, 1., Cal, D. W.,Owen-Schaub, L. B., and Roth, J. A. A retroviral wild-type p53 expres-sion vector penetrates human lung cancer spheroids and inhibits growth

by inducing apoptosis. Cancer Res., 53: 4129-4133, 1993.

16. Liu, T-J., El-Naggar, A. K., McDonnell, 1. J., Steck, K. D., Wang,M., Taylor. D. L.. and Clayman. G. L. Apoptosis induction mediated by

wild-type p53 adenovirus gene transfer in squamous cell carcinoma ofthe head and neck. Cancer Res., 55: 31 17-3122, 1995.

I 7. Hartwell, L. H., and Kastan, M. B. Cell cycle control and cancer.

Science (Washington DC), 266: 1821-1828, 1994.

18. Hopp, 1. P., Prickett, K. S., Price, V., Libby, R. 1., March, C. J.,

Cerretti, P., Urdal, D. L., and Conlon, P. J. A short polypeptide marker

sequence useful for identification and purification of recombinant pro-

teins. Biotechnology, 7: 1205-1210, 1988.

19. Clayman, G. L., Wang, S. W., Nicholson, G. L., El-Naggar, A. K..Mazar, A.. Henkins, J., Blasi, F., Goepfert, H., and Boyd, D. D.

Regulation of urokinase-type plasminogen activator expression in squa-

mous-cell carcinoma ofthe oral cavity. Int. J. Cancer, 54: 73-80, 1993.

20. Sacks, P. G., Parnes, S. M., Gallick, G. E., Mansouri, Z., Lichtner,

R., Satya-Prakash. K. L.. Pathak, S., and Parsons, D. F. Establishmentand characterization of two new squamous cell carcinoma cell lines

derived From tumors of the head and neck. Cancer Res., 48: 2858-

2866, 1988.

21. Zhang. W-W.. Fang, X., Mazur, W., French, B. A., Georges, R. N.,

and Roth, J. A. High-efficiency gene transfer and high-level expression

of wild-type p53. I. Human lung cancer cells mediated by recombinant

adenovirus. Cancer Gene Ther., I: 1-10, 1994.

22. Chomczynski, P., and Sacchi, N. Single-step method of RNA iso-

lation by guanidinium thiocyanate-phenol-chloroform extraction. Anal.

Biochem., 162: 156-159, 1987.

23. Hsu, S. M., Raine, L., and Fanger, H. Use of avidin-biotin-perox-

idase complex (ABC) in immunoperoxidase techniques: a comparison

between ABC and unlabeled antibody (PAP) procedures. J. Histochem.Cytochem.. 29: 577-580, 1981.

24. O’Malley. B. W., Jr., Chen, S-H., Schwartz, M. R., and Woo,

S. L. C. Adenovirus-mediated gene therapy for human head and neck

squamous cell cancer in a nude mouse model. Cancer Res., 55: 1080-

1085, 1995.

25. So, X., Prestwood, A. K., and McGraw, R. A. Production of

recombinant porcine tumor necrosis factor a in a novel E. cohi expres-sion system. Biotechniques, /3: 756-761, 1992.

26. Wang. J., Bucana, C. D., Roth, J. A., and Zhang, W-W. Apoptosisinduced in human osteosarcoma cells is one of the mechanisms for the

cytocidal effect of AdCMV-p53. Cancer Gene Ther., 2: 1-9, 1995.

27. Shaulsky, G., Goldfinger, N., Ben-Ze’ev, A., and Rotter, V. Nuclear

accumulation of p53 protein is mediated by several localization signals

and plays a role in tumorigenesis. Mol. Cell Biol.. 10: 6565-6577, 1990.



1997;3:185-191. Clin Cancer Res S M Overholt, T J Liu, D L Taylor, et al. trials.adenovirus-p53-FLAG: a potential marker for gene therapy Head and neck squamous cell growth suppression using

Updated version

http://clincancerres.aacrjournals.org/content/3/2/185

Access the most recent version of this article at:

E-mail alerts related to this article or journal.Sign up to receive free email-alerts

Subscriptions

Reprints and

[email protected] at

To order reprints of this article or to subscribe to the journal, contact the AACR Publications

Permissions

Rightslink site. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC)

.http://clincancerres.aacrjournals.org/content/3/2/185To request permission to re-use all or part of this article, use this link



http://clincancerres.aacrjournals.org/cgi/alerts

mailto:[email protected]



head and neck squamous cell growth suppression using...

Documents