Oral Oncology 45 (2009) e134–e139

Contents lists available at ScienceDirect

Oral Oncology
journal homepage: www.elsevier.com/locate/oraloncology

Clinicopathological significance of ubiquitin-specific protease 2a (USP2a), fatty

acid synthase (FASN), and ErbB2 expression in oral squamous cell carcinomas
Sabrina Daniela da Silva a,b, Isabela Werneck Cunha c, Inês Nobuko Nishimoto b, Fernando Augusto Soares c,
Dirce Maria Carraro d, Luiz Paulo Kowalski b, Edgard Graner a,*
a
Department of Oral Diagnosis, School of Dentistry of Piracicaba, State University of Campinas (UNICAMP), Av. Limeira 901, CP52, Areão, Piracicaba, CEP 13414-018, São Paulo, Brazil
b
Department of Head and Neck Surgery and Otorhinolaryngology, A.C. Camargo Hospital, Rua Prof. Antonio Prudente 211, São Paulo, CEP 01509-900, São Paulo, Brazil
c
Department of Anatomic Pathology, A.C. Camargo Hospital, Rua Prof. Antonio Prudente 211, São Paulo, CEP 01509-900, São Paulo, Brazil
d
Molecular Biology Laboratory, A.C. Camargo Hospital, Rua Prof. Antonio Prudente 211, São Paulo, CEP 01509-900, São Paulo, Brazil

a r t i c l e i n f o s u m m a r y

Article history: Overexpression of fatty acid synthase (FASN) and ErbB2 has been described in oral squamous cell carci-
Received 11 December 2008 nomas (OSCC). FASN is the key lipogenic enzyme responsible for the endogenous synthesis of fatty acids
Received in revised form 12 February 2009 and its expression can be regulated by ErbB2. The deubiquitinating enzyme (DUB) ubiquitin-specific pro-
Accepted 12 February 2009
tease 2a (USP2a) plays a critical role in prostate cancer cell survival by stabilizing the FASN protein. This
Available online 9 April 2009
study investigates whether the gene expression and the immunohistochemical status of FASN, ErbB2, and
USP2a correlate with the clinicopathological characteristics of OSCC cases. A strong positive correlation
Keywords:
among ErbB2, FASN, and USP2a expression (p = 0.001) was observed by qRT-PCR in laser capture micro-
ErbB2
Fatty acid synthase
dissected OSCC samples. Perineural infiltration was associated with ErbB2 mRNA expression
USP2a (p = 0.046). The presence of metastatic cervical lymph nodes was associated with FASN (p = 0.002), ErbB2
Oral squamous cell carcinoma (p = 0.001), and USP2a (p = 0.006) mRNA levels. ErbB2 staining at the cell membranes was stronger in
well-differentiated lesions while a cytoplasmic positivity was found in poorly differentiated tumors. Most
of the OSCC (97.06%) that showed a high positivity for FASN were also labeled for ErbB2 at the cell mem-
branes (p = 0.001). FASN and ErbB2 positivity was associated with tumor thickness and lymphatic embo-
lization (p = 0.006 and p = 0.035, p = 0.006 and p = 0.024 respectively). The membrane expression of
ErbB2 as well as FASN and Ki-67 staining were significantly associated with a high risk of recurrence
by predicting both disease free survival (log-rank test, p = 0.0056, p = 0.0011, and p = 0.0004, respectively)
and overall survival (log-rank test, p = 0.0005, p = 0.0062, and p = 0.0001, respectively). Taken together,
the results presented here suggest a molecular connection among FASN, ErbB2, and USP2a in OSCC since
their mRNA and protein levels were associated with tumor progression and poor prognosis.
Ó 2009 Elsevier Ltd. All rights reserved.

Introduction
Fatty Acid Synthase (FASN) is a multifunctional cytosolic en-
zyme responsible for de novo endogenous synthesis of saturated
long-chain fatty acids.1,2 Its expression is up-regulated in several
human malignancies3–11 including oral squamous cell carcinomas
(OSCC).12–17 In some human cancers, the FASN overexpression is
associated with poor clinical outcome by predicting increased risk
of recurrence, metastases, or shorter survival.3,5,8,9,11,18–23 Recent
studies have revealed a bi-directional connection between FASN
* Corresponding author. Tel.: +55 19 2106 5318; fax: +55 19 2106 5218.
E-mail addresses: sabrina.silva@hcancer.org.br (Sabrina Daniela da Silva),
iwerneck@ig.com.br (I.W. Cunha), ines@hcancer.org.br (I.N. Nishimoto), fasoares@
hcancer.org.br (F.A. Soares), dirce.carraro@hcancer.org.br (D.M. Carraro), lp_
kowalski@uol.com.br (L.P. Kowalski), egraner@fop.unicamp.br, edgardgraner@
yahoo.com (E. Graner)sabrina.silva@hcancer.org.br (Sabrina Daniela da Silva),
iwerneck@ig.com.br (I.W. Cunha), ines@hcancer.org.br (I.N. Nishimoto), fasoares@
hcancer.org.br (F.A. Soares), .
and the tyrosine kinase orphan receptor ErbB2.24–27 It was experi-
mentally demonstrated that overexpression of human ErbB2 in
mouse fibroblasts stimulates FASN protein through a PI3 K-depen-
dent pathway.24,26 Importantly, both pharmacological and RNAi-
mediated inhibition of FASN specifically down-regulated ErbB2
expression in breast and ovarian cancer cells through the up-regu-
lation of its transcriptional repressor PEA3.25
The selective degradation of many proteins in eukaryotic cells is
carried out by the ubiquitin (Ub)-proteasome system. In this path-
way, proteins are targeted for degradation by the covalent addition
of Ub, a highly conserved 76-amino acid polypeptide.28 Proteolysis
via the Ub-proteasome system plays important role in a variety of
basic cellular mechanisms and abnormalities in Ub-mediated pro-
cesses have been associated with malignant transformation.29,30
Protein ubiquitination is reversible and deubiquitinating enzymes
(DUBs) are cysteine proteases that specifically cleave Ub from
Ub-conjugated protein substrates.31,32 The DUB ubiquitin-specific
protease 2a (USP2a) interacts with and stabilizes FASN in LNCaP

1368-8375/$ - see front matter Ó 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.oraloncology.2009.02.004
 Sabrina Daniela da Silva et al. / Oral Oncology 45 (2009) e134–e139 e135

prostate cancer cells and may confer selective advantage to cancer
cells through FASN overexpression.33
In this report, we describe for the first time the expression of
USP2a in OSCC, which is associated with FASN levels in ErbB2 po-
sitive cases. In addition, we show that their expression correlates
with clinicopathological features of the tumors.
2: strong) in a blinded analysis performed by the two of the
authors (SDS and IWC). The reactions for ErbB2 were further ana-
lyzed according to the localization of the immunodeposits and
classified in membrane or cytoplasmatic staining. The percentage
of Ki-67 positive nuclei was calculated with an image computer
analyzer (Kontron 400, Carl Zeiss, Germany).

Laser capture microdissection and RNA extraction/amplification

Materials and methods
Samples were laser capture microdissected with the aid of a
Study population PixCell II Laser Capture Microdissection System (Arcturus Engi-
neering, Mountain View, CA). Approximately 10,000 cells were
Forty-one frozen samples were retrieved from the Tumor Tissue captured from 5 lm frozen sections mounted onto glass slides
Biobank of the A.C. Camargo Hospital, São Paulo, Brazil. Tumor and and stained with toluidine blue. Total RNA was isolated using Pico-
matched morphologically normal areas of the same cases were Pure RNA Isolation Kit (Arcturus Engineering) and submitted to
microscopically selected and 5 lm thick sections were cut and one-round RNA amplification based on template switch and T7-
mounted onto glass slides, stained with hematoxylin and eosin, driven amplification34 with some modifications for microdissected
and reviewed. This study was carried out with approval of the Hu- samples, as described by Saraiva et al. (2006).35
man Research Ethics Committee of A.C. Camargo Hospital and
School of Dentistry of Piracicaba, UNICAMP. Demographic, lifestyle, Primers and quantitative RT-PCR (qRT-PCR)
clinical, and pathological factors were analyzed as previously de-
scribed (14). The clinical characteristics of these patients are The primers were designed using the web tool Human BLAT
shown in Table 1. Search (http://genome.ucsc.edu/cgi-bin/hgBlat). Only secondary
structures with temperature below 10 °C were allowed (OLIGO-
Immunohistochemistry TECH vs 1.00 software, Oligos Etc. Inc. and Oligos Therapeutics
Inc., Wilsonville, USA). The list of assayed genes and primer se-
Immunohistochemical reactions were performed as previously quences are shown in Table 2. The most stable reference genes
described.15 The incubations with the primary antibodies were were beta-actin (ACTB) and glyceraldehyde-3-phosphate dehydro-
made overnight at 4 °C: anti-ErbB2 (Dako, Carpinteria, CA) 1:200, genase (GAPDH) from 4 tested controls (ACTB, GAPDH, hypoxan-
anti-FASN (Transduction Laboratories, Lexington, USA) 1:3000, thine phosphoribosyl transferase – HPRT, and breakpoint cluster
and anti-Ki-67 (Dako) 1:200. Positive and negative controls were region – BCR) with the aid of GeNorm software.36 Fold differences
included in all reactions. The intensity score for FASN and ErbB2 in the relative gene expression between tumors and matched mor-
represented the estimated staining intensity (0: negative, 1: weak, phologically normal epithelia were calculated using Pfaffl model
(2001).37 cDNAs were synthesized with Superscript II reverse
transcriptase (Invitrogen, Carlsbad, CA) and oligo-dT primers
Table 1
(Invitrogen) and qRT-PCR performed in an ABI PrismTM 7500 Se-
Distribution of the 41 OSCC cases according to demographic, lifestyle, and clinico-
pathological variables. quence Detection System (Applied Biosystems, Foster City, CA).
qRT-PCR reactions were carried out in triplicate using SYBR Green
Variable Categories n (%)
PCR MasterMix (Applied Biosystems) for 2 min at 95 °C followed
Age 659.2 years 22 (53.7) by 40 cycles at 95 °C for 15 s, the annealing temperature of each
>59.2 years 19 (46.3)
primer pair (Table 2) for 30 s, and 72 °C for 30 s.
Gender Male 35 (85.4)
Female 6 (14.6)
Race Caucasians 36 (87.8)
Statistical analysis
Non- caucasians 5 (12.2)
Smoking habit Yes 30 (76.3)
For frequency analysis in contingency tables, statistical analyses
No 9 (23.7) of associations between variables were performed by the Fisher’s
Alcohol consumption Yes 29 (76.3)
exact test (with significance set for p < 0.05) and for continuous
No 9 (23.7) variables the non-parametric Mann–Whitney U test. The overall
Clinical stage T1+T2 21 (52.5)
survival (OS) was defined as the interval between the beginning
T3+T4 19 (47.5) of treatment (surgery) and the date of death or the last information
Lymph nodes N0 20 (48.8)
for censored observations. The disease free interval (DFI) was mea-
N+ 21 (51.2) sured from the date of the treatment to the date when recurrence
Extension to adjacent sites No 10 (33.3)
was diagnosed. OS and DFI probabilities were estimated by the
Yes 20 (66.7) Kaplan–Meier method, and the log-rank test was applied to assess
Histological grade I 13 (35.2)
the significance of differences among actuarial survival curves with
II 21 (56.7) the confidence interval of 95%.
III 3 (8.1)
Vascular embolization No 36 (90.0)
Results
Yes 4 (10.0)
Lymphatic permeation No 29 (72.5)
The studied population consisted of 41 patients with OSCC with-
Yes 11 (27.5)
out a second primary tumor diagnosed and treated at the A.C. Cam-
Perineural infiltration No 18 (45.0)
argo Hospital, São Paulo, Brazil. Thirty-five patients (85.4%) were
Yes 22 (55.0)
male and 6 (14.6%) female, with the mean age of 59.2 years, ranging
Status Alive 22 (53.6)
from 35 to 83 years. History of alcohol consumption was observed
Dead 19 (46.3)
in 29 (76.3%) of the patients and tobacco smoking reported by 30
e136 Sabrina Daniela da Silva et al. / Oral Oncology 45 (2009) e134–e139

Table 2
Primer sequences, amplicon sizes, and PCR conditions.

Gene Accession number Foward primer (50 -30 ) Reverse primer (50 -30 ) Amplicon (pb) PCR conditions
AT (°C) Primer concentration nM
GAPDH NM_002046.2 GAAGGTGAAGGTCGGA GGGTCATTGATGGCAAC 102 63 200
ACTB NM_001101.2 GCACCCAGCACAATGAAG CTTGCTGATCCACATCTGC 117 64 200
HPRT1 NM_000194 GAACGTCTTGCTCGAGATGTGA TCCAGCAGGTCAGCAAAGAAT 116 60 400
BCRP NM_004327. CCTTCGACGTCAATAACAAGGAT CCTGCGATGGCGTTCAC 112 60 400
FASN NM_004104 AACTCCTTGGCGGAAGAGA TAGGACCCCGTGGAATGTCA 182 60 400
AR NM_000044 GCTCCTGGACTCCGTGCA GGTGAGCGTGGACTTTCCG 94 60 800
ERBB2 NM_001005862. CCCTCTGAGACTGATGGCTACG GCCGAACATCTGGCTGGTT 79 60 400
USP2a NM_004205. GCCGCTACACACTGTGGGA AGCATCCTGCTGATTATAGC 382 60 400

(76.9%). With regard to the ethnic group, 36 (87.8%) were cauca-

sians and 5 (12.2%) non-caucasians (Table 1). A total of 21 cases
(52.5%) were at initial clinical stages (T1 + T2) and 19 (47.5%) at
advanced clinical stages (T3 + T4) and twenty-one patients
(51.2%) had metastatic cervical lymph nodes at the moment of
the diagnosis (Table 1). Of the 41 eligible cases, vascular emboliza-
tion was found in four cases (10.0%), perineural infiltration in 22
(55.0%), and lymphatic permeation in 11 (27.5%). Thirteen cases
(31.7%) were histologically well-differentiated (Grade I), 25 cases
(61.0%) moderately differentiated (Grade II), and three cases
(7.3%) poorly differentiated (Grade III) (Table 1). Twenty (48.8%) pa-
tients had tumor recurrence after initial treatment within a mean
time of 11.4 months. At the end of the follow-up period, 22
(53.7%) patients were alive and 19 (46.3%) dead (Table 1). Eighteen
(43.9%) patients died because of the OSCC and one (2.4%) died dur-
ing the treatment. The 5-year rates for OS and DFI were 47.0% and
52.5%, respectively.
We sought to determine, in the patients described above, the
mRNA expression levels of ERBB2, FASN, and USP2a by qRT-PCR.
Significant differences (p < 0.01) were found in the relative mRNA
expression of FASN, ERBB2, and USP2a when OSCC and normal adja-
cent epithelium were compared after normalization by GAPDH
(Fig. 1). The tumor/normal fold differences were 80.3, 68.5, and
24.5, respectively. Metastatic cervical lymph nodes were associ-
ated with ERBB2 (p = 0.001), FASN (p = 0.002), and USP2a expres-
sion (p = 0.006) (Table 3, Supplementary material), and advanced
clinical stage (T3 + T4) and perineural infiltration with ERBB2
expression (p = 0.001 and p = 0.046) (Table 3, Supplementary
material). In addition, ERBB2, FASN, and USP2a were over expressed
in patients with smoking habit (p = 0.036, p = 0.019, and p = 0.036,
respectively) (Table 3, Supplementary material) and a strong posi-
tive correlation among their mRNA levels was observed (p = 0.001).

200
180
Relative mRNA levels

160
140
120
100
80
60
40
20
0
T N T N T N Figure 2 Representative immunohistochemical reactions for ErbB2 and FASN in
FASN ErbB2 USP2a OSCC samples. A clearly demarcated membrane staining was observed in well-
differentiated tumors, mainly in the cells closely associated with the formation of
Figure 1 Relative mRNA expression levels of FASN, ERBB2, and USP2a in primary keratin pearls (A); intracytoplasmic ErbB2 labeling (*) was detected in poorly
OSCCs and morphologically normal epithelium adjacent to the tumors, after differentiated tumor cells (B). (C) Strong intracytoplasmic reaction for FASN in
normalization by the GAPDH mRNA levels (p < 0.01, t-test). OSCC. Original magnification: (A) and (C): 400, (B): 200.
 Sabrina Daniela da Silva et al. / Oral Oncology 45 (2009) e134–e139 e137

Similar results were obtained after normalization by both GAPDH tic permeation (p = 0.035), metastatic lymph nodes (p = 0.039), and
and ACTB reference genes. thickness higher than 2.9 mm (p = 0.006) (Table 4, Supplementary
Cell membrane staining for ErbB2 was found in the adjacent material). Most of the OSCC (97.1%) that showed a high expression
morphologically normal epithelium as well as in well-differenti- of FASN were also positive for ErbB2 at the cell membrane
ated tumors, mainly close to the keratin pearls (Fig. 2A). Intracyto- (p = 0.001). The cytoplasmic ErbB2 staining was associated with
plasmic ErbB2 was characteristic of poorly differentiated tumor the Ki-67 positivity (p = 0.001). Interestingly, ErbB2 at the cell
cells (Fig. 2B). FASN positivity was cytoplasmic and weak in the membrane as well as FASN were associated with lymphatic embo-
normal epithelium, where it was restricted to the lower cell layers lization (p = 0.024 and p = 0.006, respectively) and thickness higher
(not shown). Thirty-four OSCC cases (82.9%) were strongly positive than 2.9 mm (p = 0.006 and p = 0.035, respectively) (Table 4, Sup-
for FASN (Fig. 2C), which was significantly correlated with lympha- plementary material). ErbB2 at the cell membrane was inversely

A 1.00
Membrane ErbB2 (negative or weak) B 1.00

FASN (negative or weak))

Survival Probability
Survival Probability

0.75 0.75

0.50 Membrane ErbB2 (strong) 0.50 FASN (strong))

0.25 0.25

p= 0.0005 p=0.0062
0.00 0.00
0 12 24 36 48 60 0 12 24 36 48 60

C 1.00 D 1.00
Survival Probability

Survival Probability
Membrane ErbB2 (negative or weak)) 0.75 FASN (negative or weak))
Disease- Free

0.75
Disease- Free

0.50 Membrane ErbB2 (strong) 0.50

0.25 FASN (strong))

0.25
p=0.0056
p=0.0011
0.00 0.00
0 12 24 36 48 60 0 12 24 36 48 60

Figure 3 OSCC cases with strong membrane ErbB2 and FASN immunolabeling had shorter survival rates (A) and (B) and higher risk of recurrence (C) and (D) in comparison
with the negative and weakly stained ones. (——) Membrane ErbB2 and FASN negative or weak; (———) Membrane ErbB2 and FASN strongly positive. Kaplan–Meier method,
log-rank test.

100
A 100 p = 0.0002
B 100 p = 0.0001 C p = 0.0481
USP2a mRNA levels
FASN mRNA levels
ErbB2 mRNA levels

0 0 0
Negative Weak Strong Negative Weak Strong Negative Weak Strong
ErbB2 immunohistochemistry ErbB2 immunohistochemistry ErbB2 immunohistochemistry

D 100 p = 0.0004 E 100 p = 0.0022

FASN mRNA levels
ErbB2 mRNA levels

0 0
Negative Weak Strong Negative Weak Strong
FASN immunohistochemistry FASN immunohistochemistry

Figure 4 Relationship between mRNA levels and immunohistochemical labeling in OSCC samples. The box plots indicate mRNA expression levels according to the
immunolabeling intensities (negative, weak, or strong). Horizontal lines in each box represent the median and vertical bars the spread of the interquartile range, dots
represent outliers. The ErbB2 protein levels were significantly associated with ERBB2, FASN, and USP2a mRNA levels (p = 0.0002, p = 0.0001, and p = 0.0481, respectively) (A)–
(C). The FASN protein levels were correlated with ERBB2 and FASN gene expression (p = 0.0004 and p = 0.0022, respectively) (D) and (E) (Mann–Whitney test).
e138 Sabrina Daniela da Silva et al. / Oral Oncology 45 (2009) e134–e139
associated with local extension to adjacent structures (p = 0.009)
(Table 4, Supplementary material).
Our immunohistochemical results also evidenced that ErbB2 at
the tumor cell membranes as well as FASN staining were able to
predict the survival probability (Fig. 3A and B). The OS for ErbB2
and FASN negative or weak cases were 58.5% and 54.1% while
strongly positive cases had 25.5% and 31.1% of OS, respectively,
with significant differences among the survival curves (log-rank
test: p = 0.0005, p = 0.0062, respectively) (Fig. 3A and B). In the
same way, there was a significant association in the DFI, which
was 29.9% and 22.4% for membrane ErbB2 and FASN positive and
65.0% and 69.0% for negative or weak cases (log-rank test:
p = 0.0056, p = 0.0011, respectively) (Fig. 3C and D). Additionally,
Ki-67 was also significantly associated with a higher risk of recur-
rence because it predicted both OS and DFI (log-rank test:
p = 0.0001 and p = 0.0004, respectively). USP2a mRNA levels were
not predictors of OS or DFI (log-rank test: p = 0.1546 and
p = 0.4237, respectively).
The correlations between transcriptional levels and immuno-
histochemistry were represented by box-and-whisker plots
(Fig. 4). The ErbB2 protein levels were associated with ERBB2, FASN,
and USP2a gene expression (p = 0.0002, p = 0.0001, and p = 0.0481,
respectively) (Fig. 4A–C), and FASN protein levels were correlated
with ERBB2 and FASN mRNA levels (p = 0.0004 and p = 0.0022,
respectively) (Fig. 4D and E).
Discussion
Despite tremendous progresses in surgical techniques, radio-
therapy, and chemotherapy, the prognosis for patients with OSCC
has been slightly improved during the past three decades38 and
major improvement in patient survival will require an increased
understanding of its pathogenesis. In the present study, we have
applied the qRT-PCR technology in association with RNA linear
amplification39 to verify ERBB2, FASN, and USP2a mRNA expression
in laser microdissected OSCC and matched morphologically normal
epithelia from the same patients.
The oncoprotein ErbB2 is overexpressed in several human
malignancies including OSCC, in which it is associated with short-
er survival, and local or distant metastasis.40,41 A functional con-
nection between ErbB2 and FASN was first described in human
breast epithelial cells.27 In addition, overexpression of ErbB2 in
mouse fibroblasts stimulates FASN protein expression through a
PI3 K-dependent pathway and the inhibition of FASN by means
of RNAi or cerulenin down-regulates ErbB2 mRNA in cancer cell
lines.25,26
The production of FASN is higher in OSCC in comparison with
the normal oral epithelium and epithelial dysplasias.12–17 In con-
trast with several FASN-overexpressing tumors, we have previ-
ously shown that FASN expression is higher in well-differentiated
than in poorly differentiated OSCC and associated with the pres-
ence of ErbB2 at the cell membrane.12,13,15 These findings can be
explained by the participation of FASN in both cell proliferation
and differentiation, since its expression during keratinization was
experimentally described.12 In this study, a significant difference
was found in the relative FASN mRNA levels between primary OSCC
and adjacent normal epithelium. Moreover, FASN expression was
significantly correlated with lymphatic permeation, tumor thick-
ness, and presence of metastatic lymph nodes, whereas FASN imu-
nohistochemical status was able to predict the survival probability.
Previous results from our laboratory have demonstrated that OSCC
cases with strong intracytoplasmic ErbB2 expression and high Ki-
67 index had a significantly shorter survival and recurrence time
than those with low or negative expression of these proteins.13,15
Interestingly, the results here presented show that the presence
of ErbB2 at the tumor cell membranes was associated with reduced
survival probability and despite the fact that 91.9% of the studied
cases were histologically well or moderately differentiated 20
(48.8%) patients had tumor recurrence during the study and a short
overall survival time. In our previous work, most patients did not
have metastatic lymph nodes that are very strong predictors of
survival. On the contrary, in the present series 51.2% of the cases
had metastatic lymph nodes at the moment of diagnosis.
Despite of the large number of DUBs already identified, little is
known about their physiological roles or substrates. DUBs may
‘‘edit” the number of ubiquitin moieties in the polyubiquitin chain
of erroneously ubiquitinated proteins or generate free ubiquitin
from polyubiquitin chains released after proteasomal activity.
Importantly, the pre-proteasomal action of DUBs results in the
cleavage of the polyubiquitin tag from specific substrates, thus pre-
venting their degradation.42,43 USP2a plays a critical role in pros-
tate cancer cell survival through FASN stabilization. It has been
found that FASN colocalizes and physically interacts with USP2a
in LNCaP cells,33 which suggest that this isopeptidase rescues FASN
from degradation and thereby prevents apoptosis. Moreover, FASN
downregulation and induction of apoptosis were achieved by tar-
geting USP2a in prostate cancer cells.33 Here, USP2a mRNA expres-
sion in OSCC was not significantly correlated with FASN protein
levels. The expression of ERBB2, FASN and USP2a was higher in
OSCC than in the adjacent normal epithelium and associated with
clinicopathological findings such as perineural infiltration (ERBB2)
and metastatic regional lymph nodes (FASN, ERBB2, and USP2a). Ta-
ken together, these results suggest that USP2a is expressed in OSCC
and may have a role in FASN accumulation. Despite the fact that
USP2a is an androgen-regulated gene in prostate cells, its expres-
sion in OSCC is responsive to epidermal growth factor (EGF)
(unpublished results; Dr M. Agostini, University of Campinas).
Functional studies will be necessary in order to better understand
the role of USP2a/FASN interaction as well as ERBB2 signaling in
OSCC. In addition, the association of ERBB2, FASN, and USP2a
expression with smoking habit may be an effect of tobacco carcin-
ogens on oral epithelial cells.
In summary, our results show that the expression of ERBB2,
FASN, and USP2a is higher in OSCC than in normal epithelium
and associated with clinicopathological characteristics. Further-
more, the ErbB2, FASN, and Ki-67 immunohistochemical expres-
sion correlates with poor prognosis, as evidenced by the higher
risk of recurrence and shorter survival.
Conflict of Interest Statement
None declared.
Acknowledgements
This work was supported by Fundação de Amparo à Pesquisa do
Estado de São Paulo (FAPESP 02/08030-1 and CEPID/FAPESP
9814335). SDS is supported by a FAPESP fellowship (04/06398-7).
Appendix A. Supplementary material
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.oraloncology.2009.02.004.
References
1. Baron A, Migita T, Tang D, Loda M. Fatty acid synthase: a metabolic oncogene in
prostate cancer? J Cell Biochem 2004;91(1):47–53.
2. Jayakumar A, Tai MH, Huang WY, al-Feel W, Hsu M, Abu-Elheiga L, Chirala SS,
Wakil SJ. Human fatty acid synthase: properties and molecular cloning. Proc
Natl Acad Sci USA 1995;92(19):8695–9.
 Sabrina Daniela da Silva et al. / Oral Oncology 45 (2009) e134–e139
3. Witkiewicz AK, Nguyen KH, Dasgupta A, Kennedy EP, Yeo CJ, Lisanti MP, Brody
JR. Co-expression of fatty acid synthase and caveolin-1 in pancreatic ductal
adenocarcinoma: implications for tumor progression and clinical outcome. Cell
Cycle 2008;7(19):3021–5.
4. Ogino S, Nosho K, Meyerhardt JA, Kirkner GJ, Chan AT, Kawasaki T, et al. Cohort
Study of Fatty Acid Synthase Expression and Patient Survival in Colon Cancer. J
Clin Oncol 2008;26(35):5713–20.
5. Horiguchi A, Asano T, Asano T, Ito K, Sumitomo M, Hayakawa M. Fatty acid
synthase over expression is an indicator of tumor aggressiveness and poor
prognosis in renal cell carcinoma. J Urol 2008;180(3):1137–40.
6. Notarnicola M, Altomare DF, Calvani M, Orlando A, Bifulco M, D’Attoma B, et al.
Fatty acid synthase hyperactivation in human colorectal cancer: relationship
with tumor side and sex. Oncology 2006;71(5–6):327–32.
7. Kawamura T, Kanno R, Fujii H, Suzuki T. Expression of liver-type fatty-acid-
binding protein, fatty acid synthase and vascular endothelial growth factor in
human lung carcinoma. Pathobiology 2005;72(5):233–40.
8. Innocenzi D, Alo PL, Balzani A, Sebastiani V, Silipo V, La Torre G, et al. Fatty acid
synthase expression in melanoma. J Cutan Pathol 2003;30(1):23–8.
9. Takahiro T, Shinichi K, Toshimitsu S. Expression of fatty acid synthase as a
prognostic indicator in soft tissue sarcomas. Clin Cancer Res 2003;9(6):2204–12.
10. Kusakabe T, Nashimoto A, Honma K, Suzuki T. Fatty acid synthase is highly
expressed in carcinoma, adenoma and in regenerative epithelium and
intestinal metaplasia of the stomach. Histopathology 2002;40(1):71–9.
11. Alo PL, Visca P, Framarino ML, Botti C, Monaco S, Sebastiani V, et al.
Immunohistochemical study of fatty acid synthase in ovarian neoplasms.
Oncol Rep 2000;7(6):1383–8.
12. Silva SD, Cunha IW, Rangel AL, Jorge J, Zecchin KG, Agostini M, et al. Differential
expression of fatty acid synthase (FAS) and ErbB2 in nonmalignant and
malignant oral keratinocytes. Virchows Arch 2008;453(1):57–67.
13. Silva SD, Perez DE, Alves FA, Nishimoto IN, Pinto CA, Kowalski LP, et al. ErbB2
and fatty acid synthase (FAS) expression in 102 squamous cell carcinomas of
the tongue: correlation with clinical outcomes. Oral Oncol 2008;44(5):484–90.
14. Silva SD, Perez DE, Nishimoto IN, Alves FA, Pinto CA, Kowalski LP, et al. Fatty
acid synthase expression in squamous cell carcinoma of the tongue:
clinicopathological findings. Oral Dis 2008;14(4):376–82.
15. Silva SD, Agostini M, Nishimoto IN, Coletta RD, Alves FA, Lopes MA, et al.
Expression of fatty acid synthase, ErbB2 and Ki-67 in head and neck squamous
cell carcinoma. A clinicopathological study. Oral Oncol 2004;40(7):688–96.
16. Agostini M, Silva SD, Zecchin KG, Coletta RD, Jorge J, Loda M, et al. Fatty acid
synthase is required for the proliferation of human oral squamous carcinoma
cells. Oral Oncol 2004;40(7):728–35.
17. Krontiras H, Roye GD, Beenken SE, Myers RB, Mayo MS, Peters GE, et al. Fatty
acid synthase expression is increased in neoplastic lesions of the oral tongue.
Head Neck 1999;21(4):325–9.
18. Sebastiani V, Visca P, Botti C, Santeusanio G, Galati GM, Piccini V, et al. Fatty
acid synthase is a marker of increased risk of recurrence in endometrial
carcinoma. Gynecol Oncol 2004;92(1):101–5.
19. Visca P, Sebastiani V, Pizer ES, Botti C, De Carli P, Filippi S, et al.
Immunohistochemical expression and prognostic significance of FAS and
GLUT1 in bladder carcinoma. Anticancer Res 2003;23(1A):335–9.
20. Alo PL, Visca P, Trombetta G, Mangoni A, Lenti L, Monaco S, et al. Fatty acid
synthase (FAS) predictive strength in poorly differentiated early breast
carcinomas. Tumori 1999;85(1):35–40.
21. Gansler TS, Hardman 3rd W, Hunt DA, Schaffel S, Hennigar RA. Increased
expression of fatty acid synthase (OA-519) in ovarian neoplasms predicts
shorter survival. Hum Pathol 1997;28(6):686–92.
22. Alo PL, Visca P, Marci A, Mangoni A, Botti C, Di Tondo U. Expression of fatty acid
synthase (FAS) as a predictor of recurrence in stage I breast carcinoma patients.
Cancer 1996;77(3):474–82.
e139
23. Shurbaji MS, Kalbfleisch JH, Thurmond TS. Immunohistochemical detection of a
fatty acid synthase (OA-519) as a predictor of progression of prostate cancer.
Hum Pathol 1996;27(9):917–21.
24. Menendez JA, Lupu R, Colomer R. Targeting fatty acid synthase: potential for
therapeutic intervention in her-2/neu-overexpressing breast cancer. Drug News
Perspect 2005;18(6):375–85.
25. Menendez JA, Vellon L, Mehmi I, Oza BP, Ropero S, Colomer R, et al. Inhibition
of fatty acid synthase (FAS) suppresses HER2/neu (erbB-2) oncogene
overexpression in cancer cells. Proc Natl Acad Sci USA 2004;101(29):10715–20.
26. Menendez JA, Mehmi I, Verma VA, Teng PK, Lupu R. Pharmacological inhibition
of fatty acid synthase (FAS): a novel therapeutic approach for breast cancer
chemoprevention through its ability to suppress Her-2/neu (erbB-2) oncogene-
induced malignant transformation. Mol Carcinog 2004;41(3):164–78.
27. Kumar-Sinha C, Ignatoski KW, Lippman ME, Ethier SP, Chinnaiyan AM.
Transcriptome analysis of HER2 reveals a molecular connection to fatty acid
synthesis. Cancer Res 2003;63(1):132–9.
28. Ciechanover A. The ubiquitin-proteasome proteolytic pathway. Cell
1994;79(1):13–21.
29. Ciechanover A, Orian A, Schwartz AL. The ubiquitin-mediated proteolytic
pathway: mode of action and clinical implications. J Cell Biochem Suppl
2000;34:40–51.
30. Hershko A, Ciechanover A. The ubiquitin system. Annu Rev Biochem.
1998;67:425–79.
31. Chung CH, Baek SH. Deubiquitinating enzymes: their diversity and emerging
roles. Biochem Biophys Res Commun 1999;266(3):633–40.
32. Wilkinson KD. Regulation of ubiquitin-dependent processes by
deubiquitinating enzymes. FASEB J 1997;11(14):1245–56.
33. Graner E, Tang D, Rossi S, Baron A, Migita T, Weinstein LJ, et al. The isopeptidase
USP2a regulates the stability of fatty acid synthase in prostate cancer. Cancer
Cell 2004;5(3):253–61.
34. Gomes LI, Silva RL, Stolf BS, Cristo EB, Hirata R, Soares FA, et al. Comparative
analysis of amplified and nonamplified RNA for hybridization in cDNA
microarray. Anal Biochem 2003;321:244–51.
35. Saraiva TF, Castro NP, Pineda PHB, Osório CABT, Camargo LP, Brentani HP, et al.
Effects of Oligo dT-T7 RNA primer in RNA amplification from paraffin-
embedded tissue for microarray experiments. Appl Cancer Res
2006;26(1):14–20.
36. Vandesompele J, De Preter K, Pattyn F, Poppe B, Van Roy N, De Paepe A, et al.
Accurate normalization of real-time quantitative RT-PCR data by geometric
averaging of multiple internal control genes. Genome Biol 2002;3(7)
RESEARCH0034.
37. Pfaffl MW. A new mathematical model for relative quantification in real-time
RT-PCR. Nucleic Acids Res 2001;29(9):e45.
38. Zhou X, Temam S, Oh M, Pungpravat N, Huang BL, Mao L, et al. Global
expression-based classification of lymph node metastasis and extracapsular
spread of oral tongue squamous cell carcinoma. Neoplasia 2006;8(11):925–32.
39. Wang E, Miller LD, Ohnmacht GA, Liu ET, Marincola FM. High-fidelity mRNA
amplification for gene profiling. Nat Biotechnol 2000;18(4):457–9.
40. Xia W, Lau YK, Zhang HZ, Xiao FY, Johnston DA, Liu AR, et al. Combination of
EGFR, HER-2/neu, and HER-3 is a stronger predictor for the outcome of oral
squamous cell carcinoma than any individual family members. Clin Cancer Res
1999;5(12):4164–74.
41. Xia W, Lau YK, Zhang HZ, Liu AR, Li L, Kiyokawa N, et al. Strong correlation
between c-erbB-2 overexpression and overall survival of patients with oral
squamous cell carcinoma. Clin Cancer Res 1997;3(1):3–9.
42. Kim JH, Park KC, Chung SS, Bang O, Chung CH. Deubiquitinating enzymes as
cellular regulators. J Biochem 2003;134(1):9–18.
43. Wing SS. Deubiquitinating enzymes–the importance of driving in reverse along
the ubiquitin-proteasome pathway. Int J Biochem Cell Biol 2003;35(5):590–605.