Oncogene (2004) 23, 845–851
& 2004 Nature Publishing Group All rights reserved 0950-9232/04 $25.00
www.nature.com/onc
MMP expression profiling in recurred stage IB lung cancer
Nam Hoon Cho1,2, Kyi Pyo Hong3, Sung Hui Hong1, Suki Kang1, Kyung Young Chung3
and Sang Ho Cho*,1
We aimed to clarify the prime role of recurrence in stage I
lung cancer. To determine the expression profiles,
quantitative RT–PCR and real-time PCR were performed
subsequently to evaluate the validity of meaningful
molecules identified by 0.12 K c-DNA array experiment
surveys. In all, 10 lung cancer patients presenting with
recurrence of stage IB were selected and compared with 10
stage IB lung cancer patients without recurrence since
biopsied 3 years previously. On c-DNA microarray data
analysis using pairs of recurred and the corresponding
nonrecurred patients, the following genes were found to be
upregulated in the recurred cases: matrix metalloproteinase (MMP)-10 in five cases, MMP-12 in two cases,
MMP-11, MMP-14, MMP-15, fos, cyclin E2, E2F3,
TGF-a in each one case. The most frequently upregulated
genes in recurred lung cancers were MMP-10 (stromelysin-2) and MMP-12 (macrophage elastase). On transcriptional assay by quantitative RT–PCR and real-time
RT–PCR analysis to validate those molecules, both
transcripts of MMP-10 and MMP-12 were significantly
more upregulated in recurred stage IB lung cancer than in
the non-recurred stage IB lung cancer (P ¼ 0.004).
Transcript levels were identical to c-DNA array data.
The protein levels of these entities were also evaluated by
immunohistochemistry of archival slides. By immunohistochemistry, MMP-10 monoclonal antibody showed more
intense immunoreactivity in the recurred stage IB lung
cancer than in the nonrecurred stage IB lung cancer
(P ¼ 0.0313). Our approach revealed that MMP-10 plays
an important role in the recurrence in stage IB lung
cancer, irrespective of the histologic type.
Oncogene (2004) 23, 845–851. doi:10.1038/sj.onc.1207140
Published online 1 December 2003
Keywords: recurred stage IB lung cancer; MMP-10;
MMP-12; c-DNA array; real-time RT–PCR
Introduction
The prognosis of stage I non-small-cell lung cancer
(NSCLC) depends on complete surgical removal, and
*Correspondence: SH Cho, Department of Pathology, Yonsei University College of Medicine, Seodaemoon-gu, Shinchon-dong 134,
Seoul, South Korea; E-mail: [email protected]
Received 24 February 2003; revised 13 August 2003; accepted 13 August
2003
the 5-year survival rate is no more than 80%; the
remaining patients dying succumbed to the recurrence
(Martini and Beattic, 1977; Martini et al., 1995; Cai
et al., 2002). Recurrence was defined on the basis of the
clinical and radiological relapse, that is, contralateral
lung, lymph node, or some other organs. The aggressiveness of lung carcinoma is not always related to the
tumor-node-metastasis staging. As a consequence, the
validation of markers of recurrent propensity turns out
to be essential in the therapeutic management. If a
reliable marker could be found that could be used to
predict recurrence, such patients could be administered
aggressive chemotherapy that might improve the likelihood of survival.
The recurrent propensity is linked to the cell capacity
of degrading basement membranes and extracellular
matrix (Chambers and Matrisian, 1997; Curran and
Murray, 2000; Murray, 2001). Matrix metalloproteinases (MMPs) are believed to be the main tissue-specific
physiologically relevant mediators of matrix degradation (Chambers and Matrisian, 1997; Curran and
Murray, 2000; Murray, 2001). In NSCLC, MMP-2
(gelatinase A) (Brown et al., 1993; Soini et al., 1993;
Passlick et al., 2000), MMP-9 (gelatinase B) (Brown
et al., 1993; Nagakawa and Yagihashi, 1994) MMP7(matrilysin) (Muller et al., 1991), MMP-11 (stromelysin-3) (Urbanski et al., 1992; Delebecq et al., 2000;
Thomas et al., 2000), MMP-14 (MT1-MMP) (Tokuraku
et al., 1995) and rarely MMP-10 (stromelysin-2) (Muller
et al., 1991) have been reported to be associated with the
tumor progression. The expression of some of these
proteases has been correlated with the histopathological
and clinical features of the tumors. The majority of
studies, however, have estimated mRNA expression by
RT–PCR, or mRNA in situ hybridization, or have
estimated the protein activity by gelatin or casein
zymography of tumor tissue homogenates, and rarely
only by immunohistochemistry. Currently, the gene
expression profiling technique, such as microarray,
allows us to test the expression of numerous genes
simultaneously and to identify genes of interest. This
study aimed to identify those genes upregulated using a
cDNA array including the MMP superfamily, to
compare these expressions between recurred stage IB
lung cancer and nonrecurred stage IB lung cancer. In an
attempt to assess whether MMPs may be an indicator of
recurrence and prognosis of stage I lung cancer, we
quantified the expression of the candidate genes selected
ONCOGENOMICS
1
Department of Pathology, Yonsei University College of Medicine, Seoul, Korea; 2BK 21 project for Medical Science,
Yonsei University, Seoul, Korea; 3Department of Thoracic Surgery, Yonsei University College of Medicine, Seoul, Korea
MMP expression in stage IB lung cancer
NH Cho et al
846
in c-DNA array by using quantitative RT–PCR and
real-time RT–PCR and immunohistochemistry in 20
cases with NSCLC at stage I without lymph node
metastasis.
This may lead to the development of novel tumor
markers that will allow better prediction of clinical
behavior and ultimately may discover new therapeutic
approaches.
Materials and methods
instructions (Digital Genomics, Seoul, Korea; http://www.
digital-Genomics.co.kr). Hybridization (hybridization buffer;
formamide 25%, SSC 5 , SDS 0.1%, polyA 0.5 mg/ml, Cot-1
DNA 0.5 mg/ml) was performed in a hybridization oven at
421C for 16 h. After washing (2 SSC/0.1% SDS for 5 min at
421C, 0.1 SSC/0.1% SDS for 10 min at RT, 0.1 SSC for
1 min at RT), the slides were scanned with a DNA chip scanner
(Scanarray lite; GSI Lumonics, Ottawa, Canada) at two
wavelengths to detect emissions from both Cy3 (red) and Cy5
(green). The overall intensities were normalized using a
correction coefficient obtained from the ratios of five
endogenous housekeeping genes.
Case selection
Differentially expressed MMP genes in the c-DNA array
In total, 20 stage I pulmonary NSCLCs were selected from
1999 to 2001 at the tissue bank of the Department of
Pathology, Yonsei University College of Medicine.
All archived microscopic slides were re-examined and reclassified. Clinical records were reviewed retrospectively. The tumor
stages were classified according to the international union
against cancer’s tumor–node–metastasis classification. Clinical
parameters, such as sex, age, pathology, recurrence, diseasefree time and survival time were compared between the
recurrent and nonrecurrent groups. Recurrence was defined
on the basis of the clinical and radiological relapse metastasizing in the contralateral lung, lymph node, or some
other organs. The work-ups for ruling out recurrence included
routine CBC, SMA (13 items including liver function
test, renal function test, cholesterol etc.), tumor markers
(CEA, NSE, Cyfra 21), chest CT scan, whole body scan,
abdominal sonography, PET, brain CT scan, or MRI every 3
or 6 months.
Six pairs of samples showing a 28s to 18s RNA ratio of
more than 1.0 was chosen, and checked by lab-on-a-chip
and spectrophotometry. Those samples which were proven
to be adequate in the quality of mRNA (c-DNA of recurrent
lung cancer labeled with Cy3-dUTP and equivalent c-DNA
of nonrecurred lung cancer with Cy5-dUTP) were applied
on the c-DNA array. Microarray was constructed, containing
a total of 120 genes, including 27 members of the MMP
superfamily, 83 cellular proto-oncogenes related to the
cell cycle and 10 human and yeast internal control genes.
To avoid misinterpretation of the results that could result
from variation in the hybridizations, any two compared filters
were normalized using five human internal control genes
(ACTB, PDK2, UBC, YWHAZ, RPL13A), and a normalization coefficient calculated for each comparison was used to
correct the signal intensities. Our analysis of any two
hybridizations obtained from a given sample showed no
difference using a ratio of 1.5 for overexpression and 0.75
for underexpression. However, the threshold values were
stricter: 2.0 for overexpression and 0.5 for underexpression.
In addition, if the intensity of any signal failed to be at least
three times above the background value, it was discarded from
our analysis.
Messenger ribonucleic acid (mRNA) expression
Total cellular RNA was extracted from 20 fresh frozen tumor
tissues. The validity of the fresh frozen tissues was checked by
the examination of the corresponding tissue slides. The total
RNA was prepared using Ultraspec reagent (Biotecx Lab, TX,
USA) according to the manufacturer’s chloroform extraction,
isopropanol purification and ethanol elution protocol. The
quality and integrity of mRNA extracted were confirmed using
a Lab-on-a-chip and an Agilent 2100 bioanalyzer (Agilent,
Palo Alto, CA, USA), and by spectrophotometry. Only six
pairs of the 20 samples from recurrent and nonrecurrent
patients with a 28s to 18s RNA ratio of more than 1.0 were
chosen for the next procedure. The other four pairs of samples
with less than 1.0 ratio of 28s to 18s RNA on lab-on-a-chip
were directly skipped to RT–PCR procedure without surveillence of c-DNA array.
Construction of a microarray with hybridization
c-DNA array of 0.12 K genes (Digital Genomics, Seoul,
Korea; http://www.digital-Genomics.co.kr), containing the
MMP superfamily and cell-cycle-related genes, were used for
the study. Total RNA (approximately 5 mg) was converted into
c-DNA and labeled with aminoallyl dUTP using an amino
allyl cDNA labeling kit (Ambion, Austin, TX, USA). The
RNA samples from fresh-frozen tissues of recurrent lung
cancer were labeled with Cy3-dUTP (Amersham), and those of
nonrecurrent lung cancer with Cy5-dUTP (Amersham). The
two color probes were then mixed, precipitated with ethanol,
and dissolved in 45 ml DMSO (Cy dye 2.22 mM). Hybridization
and washes were performed according to the manufacturer’s
Oncogene
Quantitative RT–PCR
A measure of 200 ng of RNA was transcribed to c-DNA
in a final volume of 20 ml. Primers for MMP-10 and MMP-12
were selected using the ENTREZ computer-analysis package.
Each primer sequence is as follows: MMP10 FW ACC TGG
GCT TTA TGG AGA TAT TC, RV TCA TAT GCA GCA
TCC AAA TAT GAT G and MMP12 FW GAA GCC AGA
AAT CAA GTT, RV TCA AGC AAG AAT GGA CAA.
Homologies with other human proteinases were excluded.
The PCR products were sequenced and showed no homology
with other known gene sequences. PCR was performed
using the primers in a volume of 100 ml containing first-strand
cDNA, primer, buffer, MgCl2 and Taq polymerase.
MMP-10 and MMP-12 were amplified using 35 cycles of the
following conditions: denaturation at 941C for 60 s, annealing
at 551C and 57.51C for 60 s, extension at 721C for 60 s, and a
final extension of 10 min at 721C. To determine the gene
expression level, b-globin was used as an internal standard in
accordance with the manufacturer’s (Gibco BRL) instructions.
For the interpretation of electrophoresis, TINA (Raytest,
Wilmington, NC, USA) software was utilized as a densitometer.
Real-time RT–PCR
TaqMan probes are designed using PrimerExpress software,
Ver 2.0 (Applied Biosystems), MMP-10 (NM_002425 NCBI
gene reference) ACAGGCATTTGGATTTTTCTAC TTC
MMP expression in stage IB lung cancer
NH Cho et al
847
and MMP-12 (NM_002425 NCBI gene reference) TTCAGGAGGCACAAACTTGTTCCTC. To normalize results, we
used a VIC-labeled glyceraldehyde 3-phosphate dehydrogenase (GAPDH) TaqMan PDAR endogenous control reagent set
(Applied Biosystems). Each probe was quenched by 6carboxy-tetramethylrhodamine (TAMRA) at the 30 end. TaqMan PCR assays were performed on an ABI Prism 7900 HT
Sequence Detection System (Perkin-Elmer Applied Biosystems). To quantify each ABC transporter gene, we prepared a
master mixture containing 10 ml of 2 TaqMan Universal
PCR Master Mix, 1 ml of gene-specific forward and 1 ml of
reverse primer (each at 18 mM), 1 ml of the gene-specific probe
(5 mM), and 2 ml of sterile water, and aliquoted this mixture into
the wells of a 384-well optical plate. A master mixture for the
endogenous control GAPDH, containing 10 ml of 2 TaqMan
Universal PCR Master Mix, 1 ml of predeveloped TaqMan
assay reagents (PDAR) from the endogenous control reagent
set and 4 ml of sterile water, was treated similarly. Finally,
triplicates of cDNA templates equivalent to 50 ng of RNA
were added to a final volume of 20 ml. The thermal cycling
conditions used were: 2 min at 501C and 10 min at 951C,
followed by 45 cycles of 15 s at 951C and 1 min at 601C.
Sequence Detector Software SDS 2.0 (Applied Biosystems)
was used for the data analysis. From each amplification plot, a
threshold cycle (Ct) value was calculated (defined as the cycle
at which a statistically significant increase in DRn is first
detected). The threshold was calculated automatically by SDS
at the 10-fold s.d. of Rn in the first 15 cycles. The Ct values
obtained were then exported to a Microsoft Excel for further
analysis. We compared the expressions of each ABC transporter in the tissue panel. Therefore, for each ABC transporter, the normalized amount of expression in the tissues that
showed the lowest expression was used as a calibrator and the
remaining tissue samples were displayed as fold changes.
Immunohistochemistry
Pulmonary archival tissue sections (4 mm) corresponding to
the fresh frozen tissues were cut from the 10 pairs of formalinfixed, paraffin-embedded tissue blocks. Three to five paraffin
blocks taken from lung cancer tissue, at the same time when
we prepared the tissue bank of lung cancer tissue, were
used for immunohistochemistry. Sections were dewaxed in
xylene, rehydrated through graded ethanol solutions, rinsed in
PBS for 5 min, and immersed in 0.3% hydrogen peroxide
in methanol for 30 min to block endogenous peroxidase
activity. The slides were then rinsed in PBS for 5 min, blocked
with a solution of 10% normal rabbit serum in PBS at room
temperature for 10 min and incubated at 41C overnight
with the primary monoclonal antibody of MMP-10/stromelysin-2 (NeoMarkers, Fremont, CA, USA), which contain an
epitope at the hinge region of MMP-10 at a dilution ratio
of 1 : 200 without any pretreatment. Nonspecific mouse
immunoglobulin G1 monoclonal antibody (03001 D, PharMingen, 1 mg/ml) was used as the negative control. Tissues
were incubated with biotinylated horse anti-mouse secondary
antibody, at a 1 : 500 dilution (Vector Laboratories, Burlingame, CA, USA), and washed extensively with avidin–biotin
peroxidase complex, at a 1 : 25 dilution. Diaminobenzene was
used as the chromogen, and hematoxylin as the nuclear
counterstain. Cytoplasmic immunoreactivities were classified
as a continuum extending from the undetected level (0%) to
diffuse and homogeneously strong staining (100%), and then
recategorized into a three-tier grading system based on the
percentage of tumor cells staining positively in the entire tumor
boundary (0, 1: minimal o5%, 2: moderately stained o50%,
3: markedly stained 450%).
Statistical analysis
The standard w2 test or Fisher’s exact test was used for
comparative analysis. The level of significance was Po0.05.
The duration of the free-from-recurrence period was measured
from the date of operation until the first evidence of recurrence
from primary NSCLC or the last date of follow-up for patients
who remained alive and with no recurrence. Survival was
calculated from the date of operation until death of the date of
the last follow-up. The survival rates were estimated by the
Kaplan–Meier method, and the statistical analysis was
performed by the log-rank test (SAS Windows).
Results
Clinical records
A total of 10 pairs of stage I lung cancer were classified
as recurred or nonrecurred (Table 1). The primary
histologic types of the recurred carcinoma were four
squamous cell carcinomas, four adenocarcinomas and
two other variants (one adenosquamous cell carcinoma
and one large cell carcinoma), whereas the histologic
types of the nonrecurred carcinoma were four squamous
cell carcinomas and five adenocarcinomas and one large
cell carcinoma. The recurred group manifested metastasis in the contralateral lung in four cases, with more
remote metastasis, which included metastasis to the
bone in three cases, liver, brain and scalene node in each
one case 9 months after the initial operation. The 10
cases of nonrecurred stage I NSCLC were observed at
34.9 months on average (26.8–41.6 months). Survival
time of the recurred group in the present study was 19.9
months, which was significantly lower (P ¼ 0.0021) than
that of the 10 nonrecurred group (Figure 1).
Differential expression by cDNA microarray
In all, 10 pairs of tissue bank lung cancer tissues were
used for microarray, but only six pairs of cases, including
three pairs of squamous cell carcinoma, two pairs of
adenocarcinomas and a pair of large cell carcinoma,
were evaluated with microarray due to poor RNA
sample quality. We investigated altered gene expression
profiles in pairs for the ‘recurred’ and ‘nonrecurred’ stage
I NSCLC. The expressions of MMP-10, MMP-11,
MMP-12, MMP-14, MMP-15, MMP-16 and MMP-19
with fos, cyclin E2, E2F3 and TGF-a were upregulated in
recurred stage I NSCLC as compared with the nonrecurred stage I NSCLC, MMP-10 was most frequently
upregulated five out of six recurrent cases and MMP-12
was upregulated in two out of six recurrent cases. The
other genes, except MMP10 or MMP12, showed an
upregulated expression in only one recurred case
(Figure 2).
Transcriptional levels of MMP- 10 and MMP-12
Assessments of MMP-10 and MMP-12 transcriptional
levels were available in 20 stage I NSCLC. MMP-10
mRNA (163 bp) revealed relatively stronger expression
in the recurred samples than in the nonrecurred samples
Oncogene
MMP expression in stage IB lung cancer
NH Cho et al
848
Table 1 Clinical summary of patients with stage I lung cancer
Recurrence
(n ¼ 10)
No recurrence (n ¼ 10)
8
2
7
3
Age (mean7s.e.)
66.774.7
61.0710.8
Pathological type
SCC
AC
Large cell
Adenosquamous
4
4
1
1
4
5
1
0
Tumor size (cm)
5.071.3
5.672.5
50%
10%
40%
5.941.2
19.973.7
9.075.1
100%
26.841.6
34.974.3
IB (T2N0M0)
IB (T2N0M0)
Sex
Male
Female
Invasion to visceral pleura
Survival
(month)
(mean7s.e.)
DFT (mean)
Stage
Significance
(P-value)
NS
(P ¼ 1.00)
NS (P ¼ 0.15)
NS
NS
NS (P ¼ 0.14)
Significant
(P ¼ 0.0021, log-rank test)
Same
construct calibration curves for relative quantification,
we performed real-time amplification of several cDNA
dilutions with ABC transporter genes and a reference
gene. Figure 4a–c shows representative examples of
MMP-10 and MMP-12. The Ct values plotted vs the log
concentrations of these calibration samples showed an
inverse linear correlation; the correlation coefficient for
each calibrator should thus be close to one. The ratios of
MMP-10 or MMP-12 to 18s mRNA was converted to a
bar graph in an end point of cycles (Figure 4d). The
increased ratios of 18s mRNA of MMP-10 or MMP-12
in the recurrent group was compatible with those
upregulated in cDNA array. However, there was no
difference of MMP-10 expression according to the
histologic type.
Figure 1 The Kaplan–Meier survival curve for 10 pairs of early
lung cancer patients. The survival rate of the recurrent group was
significantly lower than that of the nonrecurrent group (P ¼ 0.0021,
log-rank test). X-axis means follow-up period (months)
by quantitative RT–PCR (Figure 3), which was highly
significant (P ¼ 0.004, Wilcoxon ranked score test).
However, there was no difference of MMP-10 expression according to the histologic type. MMP-12 mRNA
(594 bp) was constitutively expressed, regardless of the
histologic type (Figure 3), which was not significant
(P ¼ 0.269).
Immunohistochemistry
MMP-10 was variably expressed in tumor cells as well as
faintly expressed in normal bronchial epithelial cells.
MMP-10 was relatively intense in the recurrent group as
compared with the nonrecurrent group (Figure 5), the
difference of which was statistically significant
(P ¼ 0.0313, Fisher’s exact test). There was no difference
of MMP-10 expression according to the histologic type
(P ¼ 0.847). The poorer overall survival was statistically
significant (P ¼ 0.0396) for cases with intense expression
of MMP-10 (Figure 6).
Expression analysis of MMP-10 and MMP-12 using
real-time RT–PCR
Discussion
After collecting the correct consensus sequences of all
human ABC transporter mRNAs by GenBank screening, we used Primer Express Software to design the
optimum specific primer and probe combinations. To
We have explored the gene expression profiles of 10
pairs of the representative of recurred and nonrecurred
stage I NSCLC, using a 0.12 K array including the
MMP superfamily. We validated the expression of
Oncogene
MMP expression in stage IB lung cancer
NH Cho et al
849
Figure 2 Image displaying 120 genes and c-DNA array data of the
representative pulmonary NSCLCs. When labeled with Cy3-dUTP
in the recurred samples and Cy5-dUTP in the nonrecurrent
samples, upregulated genes in the recurrent samples are seen in
red fluorescence (Cy-3) and downregulated genes in green
fluorescence (Cy-5). The blanks filled with red color represent
upregulated genes more than 2.0 of threshold value in the recurrent
cancer. MMP-10 (white arrow) and MMP-12 (yellow arrow) spots
are recognized as the red fluorescence, which means that both are
upregulated in the recurrent samples when compared with the
nonrecurrent samples
molecules upregulated in cDNA array by quantitative
RT–PCR and real-time RT–PCR, and immunohistochemistry. On c-DNA expression profiles, we focused on
MMP-10 and MMP-12 genes among several upregulated genes because both of them were found to be
significantly upregulated in more than two cases of
recurred stage I NSCLC. MMP14 and MMP15 with
several proto-oncogenes, such as fos, cyclin E2, TGF-a,
which were upregulated only in one case of recurred
stage I NSCLC, were excluded in the current study.
MMPs are believed to be the main physiologically
relevant mediators of matrix degradation, and are tissue
specific (Chambers and Matrisian, 1997; Curran and
Murray, 2000; Murray, 2001). Certain MMPs might
predict a poor prognosis when they are overexpressed;
for example, MMP-2 expression was correlated with
nodal invasion (Tokuraku et al., 1995), MMP-14 with
advanced stage disease (Nawrocki et al., 1997) and
TIMP-1 with poor survival (Fong et al., 1996). Cellular
Figure 3 Quantitative RT–PCR results for MMP-10 and MMP12. Note the downregulated expression of MMP-10 mRNA in the
nonrecurrent samples (lanes 1–5) vs the recurrent samples (lane 6–
11). In lanes 3 and 5, nonrecurrent cases show relatively weaker
signals than those in other lanes, especially right lanes from 6–11
lanes (recurrent). MMP-12 mRNA was expressed at similar levels
in recurrent and nonrecurrent samples Lanes No. 6, No. 7, No. 9,
and No. 10 are squamous cell carcinomas, lanes 8 and 11 are
adenocarcinomas
localization studies by in situ hybridization and immunohistochemistry in NSCLC have been focused on
MMP-2 and MMP-11 (Ohori et al., 1992; Bolon et al.,
1997). MMP-11 was found to be expressed in the
stromal fibroblasts of tumors and to correlate with
tumor size and nodal metastasis (Bolon et al., 1997). We
utilized a c-DNA array including almost all known
MMPs and cell-cycle-related genes, in order to detect
the genes most related to the recurrence of NSCLCs.
The present study suggests that MMP-10 could play
an important role in the recurrence of stage I NSCLC.
In addition, we found that MMP-12 could play an
important role in the pathogenesis of some cases of
recurred stage I NSCLC. We confirmed the overexpressions of MMP-10 and MMP-12 in recurred
NSCLC by RT–PCR, by the more specific real-time
RT–PCR and by immunohistochemistry. In particular,
real-time RT–PCR analysis has several advantages over
previous, mostly gel-based RT–PCR methods (Langmann et al., 2003). Thus, TaqMan PCR is a closed
homogenous system that does not require post-PCR
processing and which has a low contamination risk.
Moreover, real-time RT–PCR allows precise and
reproducible quantification because it is based on Ct
values rather than end-point detection, when the PCR
components are rate limiting. The results obtained from
the real-time RT–PCR assay are consistent with those
Oncogene
MMP expression in stage IB lung cancer
NH Cho et al
850
Figure 5 Immunohistochemical results of MMP-10 expression in
lung cancer and normal tissue. MMP-10 was diffusely expressed in
recurrent stage I squamous cell carcinoma (a,b), and welldifferentiated adenocarcinomas (c–e), compared with negative
immunoreactivity in nonrecurrent stage I squamous cell carcinoma
(f,g) and adenocarcinoma (h,i). MMP-10 was selectively stained in
tumor cells, not in normal inflammatory cells or stromal cells.
(a: 40, b: 200, c: 40, d: 100, e: 200, f: 100, g: 200,
h: 40, i: 100, diaminobenzene)
Figure 4 Results of TaqMan assays to determine optimum primer
and TaqMan probe concentrations. Normalized fluorescence
values (DRn) on the X-axis. (a) Amplification plot for 18 s RNA
is shown as a calibration curve, depicting normalized fluorescence
value (DRn) plotted against the number of amplification cycles.
The curves represent triplicate measurements at each sample
dilution. (b) Amplification plot for MMP-10 is shown. Lanes 1–5
show a tendency toward higher Ct and lower DRn than lanes 6–13
(lanes 1–5, the nonrecurred samples; lanes 6–13, recurrent samples).
(c) An amplification plot for MMP-12 is shown. Lanes 1–5 show a
tendency toward higher Ct and lower DRn than lanes 6–13 (lanes
1–5, the nonrecurrent samples; lanes 6–13, recurrent samples). The
two bar graphs are diagrams of real-time RT–PCR of MMP-10
and MMP-12. The nonrecurrent samples (L1–L5) showed relatively lower MMP-10 expression than recurrent samples (L6–L13).
(d) Note the high expression end-point ratio of MMP-10 in L9
(MMP-10/18s ratio; 22) and L10 (MMP-10/18s ratio;18). (e) In all
but one case, lane 11, recurrent samples showed comparatively high
expressions of MMP-12-mRNA. L6, L7, L9 and L10 are squamous
cell carcinomas, L8, L11, L12 and L13 are adenocarcinomas
Oncogene
obtained using the c-DNA array, suggesting that realtime RT–PCR offers a reliable method of quantifying
ABC transporter gene expression. Ct value of real-time
RT–PCR analysis was significantly increased in recurred
cases, such as L6, L7, L8, L9, L10, L11, L12 and L13, as
shown in Figure 4, whereas the relative value was
equivocal in RT–PCR alone.
MMP-10 (stromelysin-2), which is located on chromosome 11q22–q23, consists of a propeptide (82 amino
acids), a catalytic domain (165 aa), a proline-rich ‘hinge
region’ (25 aa) and a C-terminal domain (188 aa)
(Nagase, 1998). The biological role of MMP10 has been
little studied, but it is suggested to be linked with
reproduction, such as endometrial shedding during
menstruation, and wound healing (Nagase, 1998;
Rechardt et al., 2000). Although the exact role of
MMP-10 in tumor progression is not clearly understood, transformed rat embryo cell lines with high
metastatic potential were found to produce higher levels
of MMP-10 than nonmetastatic lines (Sreenath et al.,
1992). These studies suggest a correlation between
invasive behavior of tumor cells and the expression of
stromelysins. In terms of tumor invasion and metastasis,
a few malignant tumors including head and neck
MMP expression in stage IB lung cancer
NH Cho et al
851
Figure 6 The Kaplan–Meier survival curve for 10 pairs of early
lung cancer patients according to MMP-10 expression. The
survival rate with strong expression of MMP-10 was significantly
lower than that with weak expression of MMP-10 (P ¼ 0.0396,
log-rank test). X-axis means follow-up period (months)
carcinoma and lung cancer (Muller et al., 1991), and
skin cancer (Kerkela et al., 2001), express MMP-10
along with MMP-1 and MMP-7.
MMP-12 (macrophage elastase) incorporates the
principal cellular source of this enzyme and it is able
to degrade insoluble elastin (Shapiro and Senior, 1998).
Since macrophages can express several elastolytic
proteinases other than macrophage elastase, that is,
cystein proteinases, metalloproteinases, particularly
MMP-9, MMP-7 and perhaps MMP-3 (Shapiro and
Senior, 1998), MMP-12 most closely resembles the
stromelysins MMP-3, MMP-10 and matrilysin (MMP7), with respect to their ability to hydrolyse a broad
variety of extracellular matrix components (Shapiro and
Senior, 1998). MMP-12 is unique with respect to its
predominantly macrophage-specific pattern of expression and perhaps its ability to readily shed its C-terminal
domain on processing (Nagase, 1998; Shapiro and
Senior, 1998). Recently, MMP-12 has been shown to
play a central role in the pathogenesis of pulmonary
emphysema (Hautamaki et al., 1997) and of atherosclerotic lesions (Carmeleit et al., 1997). The molecular
bases for the MMP-12 expressions of macrophages are
unknown.
We conclude that MMP-10 is mainly upregulated
during the process of recurrence of stage I NSCLC at
the transcriptional and protein level. Furthermore,
the determination of MMP-10 may provide a clue to
the preselection of patients with lung cancer in clinical
trials to investigate the benefits of MMP-inhibitory
therapy.
Acknowledgements
This study was financially supported by a faculty research
grant (CMB-YUHAN 2002-02) of Yonsei University College
of Medicine for 2002 [NHC & SHC]. This study was supported
by BK 21 project for Medical Science, Yonsei University
[NHC].
References
Bolon I, Devouassoux M, Robert C, Moro D, Brambilla C
and Brambilla E. (1997). Am. J. Pathol., 150, 1619–1629.
Brown PD, Bloxidge RE, Stuart NSA, Gatter KC and
Carmichael J. (1993). J. Natl. Cancer Inst., 85, 574–578.
Cai M, Onoda K, Takao M, Kyoko I, Shimpo H, Yoshida T
and Yada I. (2002). Clin. Cancer Res., 8, 1152–1156.
Carmeleit P, Moons L, Lijnen Rn, Crawley J, Tipping P, Drew
A, Eeckhout Y, Shapiro SD, Lupu F and Collen D. (1997).
Nat Genet, 17, 439–444.
Chambers AF and Matrisian LM. (1997). J. Natl. Cancer Inst.,
89, 1260–1270.
Curran S and Murray GI. (2000). Eur. J. Cancer, 36, 1621–1630.
Delebecq TJ, Porte H, Zerimech F, Copin M, Gouyer V,
Dacquembronne E, Balduyck M, Wirtz A and Huer G.
(2000). Clin. Cancer Res., 6, 1086–1092.
Fong KM, Kida Y, Zimmerman PV and Smith PJ. (1996).
Clin. Cancer Res., 2, 1369–1372.
Hautamaki RD, Kobayashi DK, Senior RM and Shapiro SD.
(1997). Science, 277, 2002–2004.
Kerkela E, Jeskanen L, Lohi J, Grenman R, Kahari VM and
Saarialho-Kere U. (2001). Br. J. Cancer, 84, 650–669.
Langmann T, Mauerer R, Zahn A, Moehle C, Probst M,
Stermmel W and Schmitz G. (2003). Clin. Chem., 49, 230–238.
Martini N, Bains MS, Burt ME, Zakowski MF, Mocormack
P, Rusch VW and Ginberg RJ. (1995). J. Thorac.
Cardiovasc. Surg., 109, 120–129.
Martini N and Beattic Jr EJ. (1977). J. Thorac. Cardiovasc.
Surg., 74, 499–505.
Muller D, Breathnach R, Millon R, Bronner G, Flesch H,
Dumont P, Eber M and Abecassis J. (1991). Int. J. Cancer,
48, 550–556.
Murray GI. (2001). J. Pathol., 195, 135–137.
Nagakawa H and Yagihashi S. (1994). Jpn. J. Cancer Res., 85,
934–938.
Nagase H (ed). (1998). Stromelysin 1 and 2. Matrix Metalloproteinases, Parks WC and Mecham RP (eds). Academic
Press: San Diego, pp. 43–84.
Nawrocki B, Polette M, Marchand V, Monteau M, Gillery P,
Tournier JM and Birembaut P. (1997). Int. J. Cancer, 72,
556–564.
Ohori P, Yousem SA, Griffin J, Stanis K, Stetler-Stevenson
WG, Colby TV and Somez-Alpan E. (1992). Am. J. Surg.
Pathol., 16, 675–686.
Passlick B, Sienel W, Seen-Hibler R, Wockel W, Thetter O,
Mutschler W and Pantel K. (2000). Clin. Cancer Res., 6,
3944–3948.
Rechardt O, Elomaa O, Vaalamo M, Paakkonen K, Jahkola
T, Hook-Nikane J, Hembry RM, Hakkinen L, Kere J and
Saarialho-Kere U. (2000). J. Invest. Dermatol., 115, 778–787.
Shapiro SD and Senior RM. (1998). Macrophage Elastase:
Matrix Metalloproteinases, Parks WC Mecham RP (eds).
Academic Press: San Diego, pp. 185–197.
Soini Y, Paakko P and Autio-Harmainen H. (1993). Am. J.
Pathol., 142, 1622–1630.
Sreenath T, Matrisian LM, Stetler-Stevenson W, Gattoni-Celli
S and Pozzatti RO. (1992). Cancer Res., 52, 4942–4947.
Thomas P, Khokha R, Shepherd FA, Feld R and Tsao M.
(2000). J. Pathol., 190, 150–156.
Tokuraku M, Sato H, Murakami S, Okada Y, Watanabe Y
and Seiki M. (1995). Int. J. Cancer, 64, 355–359.
Urbanski SJ, Edwards DR, Maitland A, Leco KJ, Watson A
and Kossakowska AE. (1992). Br. J. Cancer, 66, 1188–1194.
Oncogene