| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Short Communications |
From the Department of Obstetrics and Gynecology,*
Helsinki University Central Hospital; and the Departments of
Pathology
and Medical
Genetics,
University of Helsinki,
Helsinki, Finland
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Ovarian carcinoma shows several different histological types, including
serous, mucinous, endometrioid, and clear cell carcinomas. There is an
increasing amount of biological and molecular evidence that different
histological types of ovarian carcinoma should be regarded as distinct
entities.10-14
The most common form of ovarian carcinoma
is the serous histological type, which accounts for 
55% of all
cases. In our previous comparative genomic hybridization studies, loss
of distal 18q was characteristic of serous ovarian
carcinoma,2,15
suggesting the presence of a tumor
suppressor gene(s) at distal 18q involved particularly in the serous
histological type of ovarian carcinoma.
To define the putative tumor suppressor locus/loci more precisely we performed loss of heterozygosity (LOH) analysis with 27 polymorphic microsatellite markers located at 18q12.3-q23 in 64 serous ovarian carcinomas. The expression of SMAD4, SMAD2, and DCC was studied by immunohistochemistry in normal ovarian tissue and the tumor samples, and their expression was correlated with LOH results. To compare the pattern of allelic loss at this region in serous and mucinous ovarian carcinomas, we also performed allelic analysis with the same 27 markers in 9 microdissected mucinous ovarian carcinomas.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Tumor Samples, Microdissection, and DNA Extraction
Tumor and blood samples were taken from 73 patients undergoing primary surgery for ovarian carcinoma at the Department of Obstetrics and Gynecology, Helsinki University Central Hospital. The tumors were staged according to the classification scheme of the International Federation of Gynecologists and Obstetricians. All of the specimens were reviewed by the same investigator (RB) as regards histological subtype and grade. There were 64 tumor specimens of serous histology (13 stage I, 3 stage II, 42 stage III, and 6 stage IV; 15 grade 1, 18 grade 2, and 31 grade 3) and 9 specimens of mucinous histology (8 stage I and one stage II; 4 grade 1, three grade 2, and two grade 3). After removal, the tissues were snap-frozen. In mucinous carcinomas, as a rule the amount of nonneoplastic cells was high and a laser microbeam microdissection technique16 was used to separate carcinoma cells before DNA extraction. In serous carcinoma, only tissue samples containing >40 to 50% of tumor cells were included in the study (range, 40 to 95%; median, 70%) and no microdissection was needed. Tumor DNA from serous carcinomas was extracted from fresh-frozen tumor tissue blocks after mechanical disruption and DNA from mucinous carcinomas from frozen sections. Normal DNA was extracted from blood lymphocytes of these patients. A standard proteinase K-phenol-chloroform method was used for DNA extraction.
Microsatellites
A set of 27 highly polymorphic microsatellite markers at
18q12.3-q23 were used. Primer sequences and reaction conditions for
dinucleotide markers were obtained from the Genethon human linkage map
(D18S58, D18S61, D18S64, D18S65, D18S68, D18S70, D18S462, D18S465,
D18S468, D18S469, D18S474, D18S483, D18S1009, D18S1118, D18S1119,
D18S1130, and D18S1131) and for tri- and tetranucleotide markers from
Genome Database (D18S539, D18S815, D18S844, D18S845, D18S857, D18S858,
D18S871, D18S969, D18S977, and D18S979). The genetic order of the
markers was based on the Genethon map, the Genome Database, and
GeneMap’99 (Figure 1)
. The
oligonucleotides were labeled fluorescently with one of three dyes
(6-FAM, TET, HEX; Institute of Biotechnology, University of Helsinki,
Finland). A fourth dye (TAMRA; Perkin-Elmer, Foster City, CA) was
reserved for the size standard.
|
The polymerase chain reaction reactions for genotyping were performed in a volume of 10 µl and included GeneAmp 1x polymerase chain reaction buffer (Perkin-Elmer), each dNTP at 50 µmol/L, 60 ng DNA (5 to 10 ng DNA from the microdissected samples), 0.5 U AmpliTaq Gold polymerase (Perkin-Elmer), and 5 pmol of each primer (one of them fluorescently labeled). The reaction mixtures were given 30 to 35 cycles of 5 seconds at 96°C, 59 seconds at 92°C, 1 minute 15 seconds at 55°C (60°C for D18S474, D18S815, D18S844, and D18S845), and 45 seconds at 72°C, preceded by a 10-minute hot start at 96°C for enzyme activation and followed by final extension at 72°C for 30 minutes.
Electrophoresis and Allele Scoring
The 27 products were pooled in three groups, each consisting of 9
µl. One µl of this mixture was added to 12.5 µl of formamide and
0.5 µl of TAMRA 500 size standard and denatured at 96°C for 3
minutes before loading the samples into an ABI Prism 310 Genetic
Analyzer (Perkin-Elmer), which introduces the samples into a
polymer-filled capillary for electrophoresis. Analysis of raw data and
assessment of LOH were performed with GeneScan and Genotyper software
(Perkin-Elmer). The peaks of the normal DNA sample were used to
determine whether the sample was homozygous (one peak only) or
heterozygous (two peaks). The sizes of the allele peaks were assigned
according to the area under the highest peak. When two alleles were
present in normal tissue and one was absent in the tumor, the result
was determined to be LOH (Figure 1B)
. In cases in which the assessment
was not clear-cut, the ratio of alleles was calculated for each normal
and tumor sample, and the tumor ratio was divided by the normal ratio,
ie, T2:T1/N2:N1 (T1 and N1 are the area values for the shorter length
alleles and T2 and N2 are the values for the longer length alleles, for
tumor and normal tissue, respectively). If the ratio was <0.6 or
>1.67, the result was determined to be LOH.17
In
ambiguous cases, the polymerase chain reaction, electrophoresis, and
scoring were repeated.
SMAD4, SMAD2, and DCC Immunohistochemistry
Sixty of the 64 serous tumors analyzed for LOH and 20 normal ovarian samples were included in a tissue microarray, which was constructed as described previously.18 In brief, core tissue biopsy specimens (diameter, 0.8 mm) were taken from representative areas of individual donor blocks and precisely arrayed into a new recipient paraffin block with a custom-built precision instrument (Beecher Instruments, Silver Spring, MD). Four core tissue biopsies were obtained from each carcinoma specimen. After block construction was completed, 5-µm sections were cut with a microtome. The presence of tumor tissue in the arrayed samples was verified in hematoxylin and eosin-stained sections.
The presence of SMAD4, SMAD2, and DCC protein in the samples was analyzed by immunohistochemistry using antibodies and protocols described earlier (R Salovaara, et al, Frequent loss of SMAD4/DPC4 protein in colorectal cancers, submitted).19-21 The primary antibodies were: monoclonal anti-human SMAD4 (final concentration 2 µg/ml, sc-7966; Santa Cruz Biotechnology Inc., Santa Cruz, CA), goat polyclonal anti-SMAD2-peptide (6 µg/ml, sc-6200; Santa Cruz Biotechnology Inc.), and monoclonal anti-human DCC (5 µg/ml, clone G97-499; Pharmingen, San Diego, CA). The sections were pretreated in a microwave oven in buffered sodium citrate before SMAD4 and DCC immunohistochemistry. An avidin-biotin immunoperoxidase system was used to visualize the bound antibody. For SMAD4, the procedure was run in a Techmate automated machine (Peroxidase DAB detection kit; DAKO ChemMate, Glostrup, Denmark). For SMAD2 and DCC, the procedure was performed manually (Vectastain Elite ABC kit; Vector Laboratories, Burlingame, CA) and 3-amino-9-ethylcarbazole was used as the chromogen. The sections were counterstained with Mayer’s hematoxylin. Blocking of the antibody by peptide preincubation (SMAD2) or omitting the primary antibody were used as negative controls. Normal ovaries have previously been shown to express SMAD2 by Northern blotting22 and DCC by immunohistochemistry and reverse transcriptase-polymerase chain reaction.9 Thus, normal ovarian samples were used as positive controls for SMAD2 and DCC in the present study. For SMAD4, colon carcinoma cell lines shown to express SMAD4 by Western blotting and reverse transcriptase-polymerase chain reaction, were used as positive controls (R Salovaara, S Roth, A Loukola, V Launonen, P Sistonen, E Avizienyte, P Kristo, H Jarvinen, S Souchelnytsky, M Sarloma-Rikala, LA Aaltonen, submitted for publication). Absence of any reactivity or very weak staining that diverged from that observed in the surface epithelium of normal ovaries and the general pattern of positive staining of the tumor samples were interpreted as negative.
Statistical Analyses
Differences in LOH and lost expression were tested by using Fisher’s exact test (two-tailed P values) and allelic loss of informative markers by using the nonparametric Mann-Whitney U test. The product-limit method was used to construct survival curves and statistical significance was tested by log-rank analysis. Multivariate analysis was performed by using the Cox proportional hazards model.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
To refine the deletion map of the distal part of chromosome arm
18q we performed allelic analysis of 64 serous ovarian carcinomas using
27 microsatellite markers at 18q12.3-q23. The samples were informative
on average at 18 loci (range, 13 to 25). Replication error of three
microsatellite markers was seen in one tumor (case 1097) and three
tumors (cases 810, 852, and 1106) showed replication error at only one
marker each. Thirty-eight of 64 (59%) serous ovarian carcinomas
studied showed allelic loss at one or more markers at distal 18q.
Fifteen of the tumors (23%) showed loss of all informative markers,
suggesting complete loss of distal 18q. Twenty-three samples showed
partial deletion and they were used to construct a deletion map of this
region (Figure 1)
. In this deletion map 12 microsatellite markers
showed allelic loss of >50% of informative alleles (D18S845, D18S539,
D18S977, D18S64, D18S969, D18S483, D18S979, D18S815, D18S469, D18S844,
D18S871, and D18S70) that were located at 18q21-23. There were three
markers that showed LOH in >75% of informative alleles: D18S483
(18q22), D18S979 (18q22), and D18S871 (18q23). Two minimal common
regions of loss could be defined around these markers with the highest
percentage of LOH: MCRL1 between markers D18S465 and D18S61 (18q22) and
MCRL2 between markers D18S462 and D18S70 (18q23).
Allelic Loss at Distal 18q in Mucinous Versus Serous Ovarian Carcinomas
In mucinous carcinomas LOH was found in one of the nine specimens (11%) and it showed LOH at 12 of 15 informative markers. The number of cases showing allelic loss in mucinous versus serous (59%) carcinomas was statistically significant (P = 0.018). The mean degree of LOH of informative alleles in mucinous carcinomas was 8.3% and in serous carcinomas it was 42% (P = 0.013). When only grade 1 and 2 (mucinous, 0%; serous, 36%) or stage I and II (mucinous, 8.3%; serous, 31%) tumors were taken into account, the difference in the degree of LOH of informative alleles still remained significant (P = 0.022 and 0.035, respectively).
Clinicopathological Characteristics and LOH at Distal 18q in Serous Ovarian Carcinomas
In serous carcinomas LOH at distal 18q was detected in 7.1% of grade 1 tumors, 72% of grade 2 tumors, and 77% of grade 3 tumors (grade 1 versus grades 2 and 3, P < 0.001). There was no correlation between LOH and stage of the serous tumors: LOH was detected in 56% of stage I and II tumors and in 60% of stage III and IV tumors.
Follow-up data for >24 months was available for 52 of the 64 serous
ovarian carcinomas analyzed for allelic loss at 18q. Two patients were
excluded from the analysis: the cause of death of one patient was
uncertain and the other patient died of a postoperative complication.
The remaining 50 patients that were included in the analysis were
either alive or had died of the ovarian carcinoma (Figure 2)
. There were 27 carcinomas that showed
LOH at 18q and 18 (67%) of these patients had died of the carcinoma,
whereas of the 23 patients showing no LOH at 18q, 7 (30%) had died of
their disease. The mean follow-up times of live patients were 37 and 38
months, respectively. The difference in survival between these two
groups was statistically significant (P =
0.044). Tumor grade was also associated with survival
(P = 0.0009), but the association between tumor
stage and survival did not reach statistical significance
(P = 0.058). In multivariate analysis, including
grade, stage, and LOH status, only grade was an independent prognostic
factor.
|
In normal ovarian tissue positive immunoreactivity of SMAD4
(moderate to strong), SMAD2 (weak), and DCC (focally weak to moderate)
was observed in surface epithelial cells and a proportion of stromal
cells. Forty-two of 60 serous carcinomas (70%) showed positive SMAD4
immunostaining, 17 (28%) were negative, and one not interpretable.
SMAD2 expression was positive in 42 (70%), negative in 17 (28%), and
not interpretable in one tumor. DCC expression was positive in 41
(68%), negative in 18 (30%), and not interpretable in one tumor
(Table 1)
. Comparison of allelic loss
results and loss of expression of SMAD4, SMAD2, or DCC is shown in
Table 2
. There was a tendency toward a
higher amount of lost expression of SMAD4, SMAD2, and DCC in tumors
with LOH at 18q21.1 compared with the tumors with no LOH at 18q21.1 (42
to 46% and 17 to 20%, respectively). However, only 41 to 46% of
tumors with LOH at 18q21.1 had lost SMAD4, SMAD2, or DCC expression,
suggesting that none of these genes alone could be the sole target of
the frequent allelic loss at 18q21.1. When analyzing the additive
effect of all three factors, a total of 83% of the tumors with LOH at
18q21.1 had lost SMAD4, SMAD2, and/or DCC expression, whereas 40% of
the tumors with no LOH at 18q21.1 had lost expression of one or more of
these proteins. Examples of SMAD4, SMAD2, and DCC immunohistochemistry
are shown in Figure 3
.
|
|
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
In contrast to the frequent allelic loss (59%) found in serous carcinomas, only one of nine (11%) mucinous carcinomas showed LOH at 18q. Despite the limited number of mucinous carcinomas included in the study, the present results showed a significant difference in the frequency of allelic loss at 18q in these two types of ovarian carcinoma. Previously, serous and mucinous ovarian carcinomas have been found to differ, for example in respect to LOH at 17q and 8p, mutations of p53 and K-ras, genomic alterations as observed in comparative genomic hybridization, and cytogenetic changes.10-14,26 Our finding of a distinct pattern of LOH at 18q adds to the evidence of a different molecular pathogenesis of these common types of ovarian carcinoma.
To our knowledge this is the first detailed mapping of distal 18q in
ovarian carcinoma. Fifteen of 38 (40%) serous tumors showing LOH at
18q presented with loss of all informative markers, suggesting loss of
the whole of distal 18q. A deletion map of 18q12.3-q23 was constructed
based on the 23 tumors showing partial or interstitial losses (Figure 1)
. The pattern of allelic loss at distal 18q was complex, but based on
the LOH frequencies and the deletion map of tumors showing partial
losses, two minimal common regions of loss could be defined: MCRL1
between markers D18S465 and D18S61 at 18q22 (3.9 cM) and MCRL2 between
markers D18S462 and D18S70 at 18q23 (5.8 cM). Both of the MCRLs are
located distal to SMAD4, SMAD2, and
DCC loci (18q21.1). There are no well-known tumor suppressor
genes at these regions, but the cadherin 7 gene
(CDH7), at 18q22.1, could be one
candidate.27
CDH7 has not been studied in
cancers, but a mutation in the E-cadherin gene (CDH1)
has been detected in ovarian carcinoma.28
In a previous
allelotype study of ovarian carcinoma, using six markers at 18q, the
smallest region of overlap was also found distal to 18q21. In that
study only two markers were located at 18q22-q23 and more detailed
mapping of the distal region was not possible.25
Our
results suggest that there are as yet unknown tumor suppressor genes at
18q22-q23 involved in serous ovarian carcinoma and we have defined two
loci for further studies.
We found that LOH at distal 18q was associated with high tumor grade. Furthermore, it was also found to be associated with poor survival: 67% of patients with tumors showing LOH at distal 18q had died of the ovarian carcinoma, in contrast to 30% of patients who had tumors showing no LOH at 18q. Previously, LOH at 18q has been associated with tumor progression and poor survival in other types of cancer, including colorectal and gastric carcinomas.29,30 In an allelotype study of ovarian tumors, which included use of one marker at 18q, there was frequent allelic loss at D18S50 (18q23) in high-grade and -stage ovarian carcinomas, but not in well-differentiated carcinomas, borderline or benign tumors.31 Interestingly, in a recent genome-wide analysis of copy number changes in ovarian carcinoma, an association between loss of 18q and reduced survival duration was found.32 However, multivariate analysis of our data showed only tumor grade to be an independent prognostic factor, not tumor stage or LOH status. Because tumor grade was a strong predictor of survival, the association between LOH and grade may partly explain the correlation of LOH with poor survival. In addition, LOH at distal 18q may be associated with other characteristics of the tumors (eg, fractional allelic loss or genomic instability) that may directly affect tumor behavior and patient survival. Thus, loss at distal 18q seems to be associated with a more aggressive phenotype of serous ovarian carcinoma, but to evaluate the role of LOH at 18q as an independent prognostic factor, studies with larger numbers of cases are needed.
To date, three candidate tumor suppressor genes have been identified on
chromosome arm 18q (18q21.1): SMAD4
(DPC4), SMAD2, and
DCC.3-5
SMAD2 and SMAD4 are known to reside in
a pathway of transforming growth factor-ß signaling. SMAD4
inactivation is frequent in pancreatic carcinomas, but relatively
uncommon in other cancer types.6
In ovarian carcinoma,
mutations of SMAD4 have been observed in 3 of 78 (3.8%)
primary tumors studied6-8
and deletion in intron 3 of
SMAD2 has been detected in 12 of 32 ovarian
carcinomas.8
DCC has been identified as a gene
frequently deleted in colorectal carcinomas,5
and it
encodes a transmembrane protein that functions as a receptor for the
axonal chemoattractant netrin-1.33
No DCC
mutations have been reported in ovarian carcinoma. We found allelic
loss at SMAD4, SMAD2, and DCC loci
(18q21.1) in 40% of serous ovarian carcinomas, which is similar to
previous findings.7
The present study, to our knowledge,
is the first in which the expression of SMAD4 and SMAD2 in ovarian
carcinoma has been evaluated by immunohistochemistry. We found lost or
very weak expression of SMAD4, SMAD2, and DCC in 28, 28, and 30% of
tumor samples, respectively. Our results are in agreement with earlier
studies in which decreased expression of DCC in a subset of ovarian
carcinomas has been reported.9,34
In contrast to our
results, in one study in which Western blotting was used, no abnormal
expression of SMAD2 in ovarian carcinomas was found.8
We
found expression of SMAD2 in the stromal cells of normal ovaries and
carcinomas. Thus, possible normal cell contamination in samples used
for Western blotting would result in a positive signal, which might
explain the discrepancy between the present and previous results.
Mutant SMAD2 and SMAD4 proteins are degraded more rapidly than their
wild-type counterparts,35
and SMAD4 immunohistochemistry
has been found to be a sensitive and specific marker for gene
alterations detected in SMAD4.19
If one allele
of these genes is inactivated by mutation and the other allele by
allelic loss, one would expect to see decreased or lost protein
expression. We found a tendency toward a higher amount of lost
expression of SMAD4, SMAD2, and DCC in tumors showing LOH compared with
tumors showing no LOH at 18q21.1. However, 20% of tumors with no
LOH showed negative immunostaining of SMAD4, SMAD2, or DCC. In these
tumors small deletions may reside between the markers used in the
study, or down-regulated expression may be because of other mechanisms
such as biallelic mutations, altered transcriptional regulation, or
epigenetic events. Furthermore, <50% of the tumors showing LOH at
18q21.1 had lost SMAD4, SMAD2, or DCC expression, which suggests that
none of these genes alone is the main target of frequent allelic loss
at 18q. Together these genes could explain up to 80% of the cases
showing LOH at 18q21.1. On the basis of these results there are still
at least 20% of cases in which LOH at 18q21.1 cannot be explained by
the effect of SMAD4, SMAD2, and DCC,
suggesting the existence of other tumor suppressor gene(s). Consistent
with this, the deletions in several tumors were large and also included
more distal parts of 18q. Furthermore, the highest frequencies and
minimal common regions of allelic loss were found at 18q22-q23.
In conclusion, we found frequent allelic loss at 18q in serous, but not in mucinous ovarian carcinomas. In serous carcinomas the highest frequency of losses was detected at the distal part, 18q22-q23, and two minimal common regions of loss could be defined at this region. Deletions were found to be associated with high-grade tumors and poor survival. The expression of three putative tumor suppressor genes, SMAD4, SMAD2, and DCC (18q21.1), was found to be lost in a subset of serous carcinomas, but none of these genes alone could be the target of the frequent allelic losses at distal 18q. The three genes could account for a substantial proportion of the cases showing LOH at 18q21.1, but not all of them. Thus we propose that the frequent allelic loss at 18q is because of the effect of multiple genes and there is at least one as yet unidentified tumor suppressor gene at 18q residing distal to SMAD4, SMAD2, and DCC involved in serous ovarian carcinoma.
|
Acknowledgements |
|---|
|
Footnotes |
|---|
Supported by grants from Helsinki University Central Hospital, the Finnish Medical Foundation, the Emil Aaltonen Foundation, Helsinki University Science Fund, and the Cancer Society of Finland.
Accepted for publication April 4, 2001.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  A Bamias, Z Yu, P. Weinberger, S Markakis, D Kowalski, R. Camp, D. Rimm, M. Dimopoulos, and A Psyrri Automated quantitative analysis of DCC tumor suppressor protein in ovarian cancer tissue microarray shows association with {beta}-catenin levels and outcome in patients with epithelial ovarian cancer Ann. Onc., December 1, 2006; 17(12): 1797 - 1802. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. Alazzouzi, P. Alhopuro, R. Salovaara, H. Sammalkorpi, H. Jarvinen, J.-P. Mecklin, A. Hemminki, S. Schwartz Jr., L. A. Aaltonen, and D. Arango SMAD4 as a Prognostic Marker in Colorectal Cancer Clin. Cancer Res., April 1, 2005; 11(7): 2606 - 2611. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. D. Pack, O. M. Alper, K. Stromberg, M. Augustus, M. Ozdemirli, A. M. Miermont, G. Klus, M. Rusin, R. Slack, N. F. Hacker, et al. Simultaneous Suppression of Epidermal Growth Factor Receptor and c-erbB-2 Reverses Aneuploidy and Malignant Phenotype of a Human Ovarian Carcinoma Cell Line Cancer Res., February 1, 2004; 64(3): 789 - 794. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T.-L. Erkinheimo, H. Lassus, P. Finne, B. P. van Rees, A. Leminen, O. Ylikorkala, C. Haglund, R. Butzow, and A. Ristimaki Elevated Cyclooxygenase-2 Expression Is Associated with Altered Expression of p53 and SMAD4, Amplification of HER-2/neu, and Poor Outcome in Serous Ovarian Carcinoma Clin. Cancer Res., January 15, 2004; 10(2): 538 - 545. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T.-L. Erkinheimo, H. Lassus, A. Sivula, S. Sengupta, H. Furneaux, T. Hla, C. Haglund, R. Butzow, and A. Ristimaki Cytoplasmic HuR Expression Correlates with Poor Outcome and with Cyclooxygenase 2 Expression in Serous Ovarian Carcinoma Cancer Res., November 15, 2003; 63(22): 7591 - 7594. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 F. Beuschlein, B. D. Looyenga, S. E. Bleasdale, C. Mutch, D. L. Bavers, A. F. Parlow, J. H. Nilson, and G. D. Hammer Activin Induces x-Zone Apoptosis That Inhibits Luteinizing Hormone-Dependent Adrenocortical Tumor Formation in Inhibin-Deficient Mice Mol. Cell. Biol., June 1, 2003; 23(11): 3951 - 3964. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
V.-M. Wasenius, S. Hemmer, E. Kettunen, S. Knuutila, K. Franssila, and H. Joensuu Hepatocyte Growth Factor Receptor, Matrix Metalloproteinase-11, Tissue Inhibitor of Metalloproteinase-1, and Fibronectin Are Up-Regulated in Papillary Thyroid Carcinoma: A cDNA and Tissue Microarray Study Clin. Cancer Res., January 1, 2003; 9(1): 68 - 75. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
