| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Regular Articles |
From the Departments of Pathology *
and
Surgery,
University of Kiel, Kiel; and the
Department of Internal Medicine,
University of
Bochum, Bochum, Germany
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Genetic data on the various PanIN grades would conceivably provide the necessary means to further refine the histopathological PanIN classification and may also be useful for early detection of high-risk lesions. Mutations in the K-ras oncogene have been identified in PanINs of all grades,5-9 and are therefore not useful in discriminating PanINs according to their grade and malignant potential. In a small series of lesions, p16 gene mutations were almost exclusively identified in PanIN-3 lesions,6 whereas loss of p16 protein expression was already found at the PanIN-2 stage.10 Furthermore, abnormal expression of the p53 protein and lack of Dpc4 protein expression was frequently seen in PanIN-3 lesions but rarely in PanIN-1 or PanIN-2 lesions,11-15 suggesting that monitoring the expression of p53 and Dpc4 would provide additional criteria to identify PanIN-3 lesions. Two recently published studies reported low to moderate frequencies of allelic losses at 9p, 17p, and 18q in PanIN-1 lesions and moderate to high allelic loss frequencies of these regions for PanIN-3 lesions.13,15 Although these molecular analyses of PanINs established a first genetic progression model for pancreatic carcinoma, no criteria have yet been generated to further categorize the PanIN-2 lesions, which may represent a very important step in the preinvasive development of pancreatic carcinoma. We therefore attempted to identify such criteria through an extensive analysis of p16, p53, and DPC4 in a series of microdissected PanINs from 22 patients, using a combination of genetic and immunohistochemical techniques.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Whipple resection specimens from 21 patients (11 men and 10 women; mean age, 60.1 years) with ductal adenocarcinoma of the pancreas and one male patient with noncancerous pancreatic disease were obtained from the Department of Surgery, University of Kiel, Germany. The specimens were macroscopically examined in the unfixed state, cut along the duct level, and immediately fixed in buffered 4% formalin. An average of three blocks of tumor tissue were excised, embedded in paraffin, cut into 5-µm sections, and stained with hematoxylin and eosin. Ductal lesions were classified according to the recently established PanIN classification.4 Furthermore, PanIN-2 lesions were subdivided into lesions exhibiting signs of low- and moderate-grade dysplasia. The criteria for low-grade dysplasia were a slight nuclear enlargement (not more than twice that in normal duct epithelium) and a slight increase in nuclear hyperchromasia with only few coarse chromatin granules. Advanced nuclear enlargement and numerous coarse dark chromatin granules were classified as moderate-grade dysplasia.
For each lesion an unstained 10-µm parallel section was
microdissected, deparaffinized, and overlaid with
phosphate-buffered saline. Microdissection was performed using a
micromanipulator (Leica, Narishige Micromanipulator, Wetzlar, Germany),
as has been described.8
In a few instances, microdissected
cells from the same lesion present in two to three serial sections were
pooled to obtain the minimum of 
100 cells required for genetic
analysis.
Whole Genome Amplification by Degenerate Oligonucleotide Primed (DOP)-Polymerase Chain Reaction (PCR)
Approximately 100 cells per lesion were microdissected and transferred to 0.2-ml PCR tubes, containing 40 µl of lysis buffer (10 mmol/L Tris-HCl, pH 8.0, 10 mg/ml Proteinase K, 1% Tween 20), and overlaid with mineral oil. Samples were subsequently incubated at 48°C for 15 hours, followed by 10 minutes at 95°C. The samples were then divided into two 0.2-ml PCR tubes, and whole genome amplification by DOP-PCR16 was simultaneously performed for each of the two samples.
DOP-PCR was performed in 20 mmol/L Tris-HCl, pH 8.8, 10 mmol/L KCl, 10 mmol/L (NH4)2SO4, 2 mmol/L MgSO4, 1% Triton X-100, 1 mg/ml bovine serum albumin, 3 µmol/L universal primer (5'-CCG ACT CGA GNN NNN NAT GTG G-3'), 200 µmol/L dNTPs, and 3.75 U Pfu Turbo DNA polymerase (Stratagene, La Jolla, CA), in a volume of 50 µl. DNA was amplified using a temperature cycler (PTC-200; MJ Research, Watertown, MA). The initial amplification stage comprised eight cycles at 94°C for 1 minute, 30°C for 1.5 minutes, followed by temperature ramping to 72°C (ramping rate 0.2°C/second), and a final extension at 72°C for 2 minutes. The subsequent cycling stage comprised 40 cycles at 94°C for 1 minute, 62°C for 1 minute, and 72°C for 2 minutes. Per cycle 5 seconds were added to each 72°C extension step, and a final extension step was performed at 72°C for 10 minutes.
Microsatellite PCR
Microsatellite markers D9S319, D9S157, D9S304, D9S171, D17S1832, D17S786, D17S796, TP53-PCR15, D18S877, D18S535, D18S363, D18S46, and D18S474 were selected within 4 cM proximity to the p16 (9p21), p53 (17p13), and DPC4 (18q21) gene loci. PCR primers were selected such that the fragment length was below 250 bp for all markers. Primer sequences are available on request.
Microsatellite-PCR reactions were performed in 96-well microtiter plates, in 20 mmol/L Tris-HCl, pH 8.4, 5 mmol/L KCl, 1.5 mmol/L MgCl2, 100 ng of each primer, 200 µmol/L dNTPs, 60 mmol/L TMAC (Sigma, Taufkirchen, Germany), 1.5% formamide, 2 µl DOP-PCR product as template, and 1.5 units Taq DNA polymerase (Gibco BRL, Karlsruhe, Germany), in a final volume of 15 µl. Reactions were performed in a Hybaid Touchdown temperature cycler (MWG-Biotech, Ebersberg, Germany), for 40 cycles of 94°C for 15 seconds, 55°C for 30 seconds, and 72°C for 30 seconds, and a final extension at 72°C for 5 minutes. PCR products were separated on 6% polyacrylamide, 8 mol/L urea gels and DNA fragments were visualized by silver staining. Gels were independently scored by two investigators (HG and SAH), who were unaware of the histological grade of the lesions. Normal tissues were considered informative if two bands of the expected size were detected, usually separated by at least two bases. Loss of an allele was determined when the corresponding normal tissue was informative and one band (allele) was missing in the neoplastic tissue. In a few ambiguous cases, gels were scanned by an imaging system (Alpha Imager, Biozym, Hessisch Oldendorf, Germany) and analyzed using image analysis software (Gel-Pro Analyzer, Version 3;Media Cybernetics, Spring Field, MD). A reduction in relative intensity of >75% was required for allelic loss.
Immunohistochemistry
Immunostaining was performed using the anti-p53 monoclonal antibody DO1 (dilution 1:50; Calbiochem, Oncogene Research Products, Cambridge, MA) and anti-Dpc4 clone A8 (dilution 1:50; Santa Cruz Biotechnology, Santa Cruz, AZ). Antigen retrieval and immunostaining for p53 and Dpc4 were performed as previously described.17,18 Briefly, antigen retrieval was performed by the pressure cooker method for 3.5 minutes for the p53 antibody and 20 minutes for the Dpc4 antibody. To improve the staining sensitivity a DAB-enhanced detection kit (Ventana Medical System, Tucson, AZ) and an amplification kit (Ventana) were used. All immunostainings were independently assessed by two of the authors who were unaware of the molecular data (JL and VB).
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
One hundred and thirty-six PanINs derived from 21 cases of
pancreatic ductal adenocarcinoma and one patient without neoplastic
disease of the pancreas were microdissected from paraffin-embedded
tissue specimens. After whole genome amplification, allelic losses at
chromosomal regions 9p21 (the location of the p16 gene),
17p13 (p53), and 18q21 (DPC4/SMAD4)
were examined by microsatellite analysis using 13 markers that mapped
to these regions. To exclude amplification artifacts, we performed two
DOP-PCRs per lesion and two microsatellite PCRs for each marker from
both DOP-PCR templates, thus generating a data set of four PCRs per
marker. A representative example of the microsatellite analysis is
shown in Figure 1
.
|
The number of lesions that were excluded from the study either because of insufficient DOP-PCR amplification or because of discordant PCR results for the various PanIN grades revealed no obvious selection bias. PCR was ineffective in 6 of 14 (43%), 7 of 22 (32%), 14 of 28 (50%), 5 of 21 (24%), 7 of 24 (29%), and 16 of 28 (57%) of PanIN-1A, PanIN-1B, PanIN-2 with low-grade dysplasia, PanIN-2 with moderate-grade dysplasia, PanIN-3 lesions, and carcinomas, respectively. Discordant results were observed in 10 of 41 (24%), 11 of 98 (11%), 16 of 86 (19%), 25 of 80 (31%), 21 of 89 (24%), and 10 of 65 (15%) PCRs performed with DNA derived from PanIN-1A, PanIN-1B, PanIN-2 with low-grade dysplasia, PanIN-2 with moderate-grade dysplasia, PanIN-3 lesions, and carcinomas, respectively. Importantly, in 67 data sets the larger allele was lost and in 63 data sets the smaller allele was lost, thus further confirming the reliability of the amplification procedure and the stringency of our inclusion criteria.
Allelic Losses in PanINs
The results of the allelic loss study are detailed in Figure 2
and summarized in Figure 3A
. Of the 81 lesions included in the
study, eight were classified as PanIN-1A. None of these early lesions
revealed loss of an allele at any of the three loci analyzed. A low
frequency of allelic loss was detected for chromosomal regions 17p (2
of 12) and 18q (1 of 14) in PanIN-1B lesions. A low to moderate
frequency of allelic loss was found at 9p (2 of 10), 17p (4 of 12), and
18q (2 of 12) in PanIN-2 lesions with low-grade dysplasia. The most
significant increase in allelic losses occurred in PanIN-2 lesions with
moderate-grade dysplasia (6 of 13 lesions at 9p, 10 of 13 at 17p, and 7
of 12 at 18q), which further increased in the PanIN-3 lesions (13 of 15
at 9p, 6 of 10 at 17p, and 14 of 16 at 18q) and the carcinomas (8 of 8
at 9p, 10 of 11 at 17p, 9 of 11 at 18q). The observed progressive
increase in allelic losses in low- to high-grade PanIN lesions support
the PanIN classification and the proposed tumor progression model for
pancreatic carcinoma.19
|
|
. The nine
PanIN-2 lesions with low-grade dysplasia had either no (6 of 9) or
one (3 of 9) alteration. The nine PanIN-2 lesions with moderate-grade
dysplasia had no (1 of 9), one (4 of 9), two (2 of 9), or three (2 of
9) alterations. The eight PanIN-3 lesions with high-grade dysplasia had
one (2 of 8), two (4 of 8), or three (2 of 8) alterations, and the
seven carcinomas had one (1 of 7), two (1 of 7), or three (5 of 7)
alterations. Again these findings support the PanIN
classification and the established tumor progression
model.19
Clonal heterogeneity was assessed in 15 cases for which more than one
lesion was available (Figure 2)
. Clonal heterogeneity was postulated
when allelic loss data were incompatible, that is, different alleles
were lost in lesions from a single case (red and green squares in
Figure 2
), or earlier lesions revealed allelic loss, whereas later
stages were heterozygous (Figure 2
, yellow squares) at the same locus.
Clonal heterogeneity was demonstrated in 6 of these 15 cases (40%; eg,
cases 14, 16, 19, 20, 21, and 22; Figure 2
). For four of these cases,
samples of at least two PanIN-3 and/or invasive carcinomas were
available (eg, cases 14, 19, 20, and 22; Figure 2
). In all four cases
there were PanIN-3 lesions with allelic losses that were incompatible
with the allelic loss pattern of the carcinoma or other PanIN-3 lesions
(eg, lesions E85 and E73 from case 14; lesions F33 and all other
lesions from case 19; lesions F29 and F17 from case 20; and lesions E92
and E93 from case 22). These results indicate that these PanIN-3
lesions cannot simply be classified as "cancerization of the
ducts."19
p53 and Dpc4 Protein Expression in PanINs
p53 protein expression was analyzed immunohistochemically in 192
PanINs (Figure 3A)
. Thirty-three PanIN-1A and 62 PanIN-1B lesions
did not stain for p53. Nuclear accumulation of p53 was found in 2 of 22
(9%) PanIN-2 lesions with low-grade dysplasia, in 4 of 36 (11%)
PanIN-2 lesions with moderate-grade dysplasia, and in 16 of 39 (41%)
PanIN-3 lesions. Concordant with the
literature,11-13,15,20
14 of 21 (67%) carcinomas showed
overexpressed p53, including all carcinomas from patients with
p53-positive precursor lesions. Our data suggest that abnormal p53
protein expression patterns essentially do not arise until the PanIN-3
stage of pancreatic tumor development.
Dpc4 protein expression was analyzed immunohistochemically in 180 PanIN
lesions (Figures 3A and 4)
. Dpc4 protein
expression was detected in all 92 PanIN-1 lesions and in all 23 PanIN-2
lesions with low-grade dysplasia, but was not detected in 4 of 31
(13%) PanIN-2 lesions with moderate-grade dysplasia and in 14 of 34
(41%) PanIN-3 lesions. Nine of 18 (50%) carcinomas did not express
Dpc4, which is in agreement with reported
frequencies.14,18
In two cases, the carcinoma tissue had
retained Dpc4 expression whereas the PanIN-3 lesions were
Dpc4-negative. This observation substantiates the clonal heterogeneity
that was observed in the allelic loss patterns of these lesions (Figure 2)
. Our results suggest that abnormal Dpc4 protein expression patterns
also tend to arise at the PanIN-3 stage.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
We found a progressive increase in allelic losses at all three
chromosomal regions in PanINs from low to high grade, with the highest
rate of allelic loss identified in carcinomas (Figure 3A)
. The majority
(75%) of PanIN-3 lesions revealed losses at two or three of the
chromosomal regions tested, whereas 67% of PanIN-2 lesions with
moderate dysplasia had losses at one or two of the three regions
(Figure 3B)
. These data are in agreement with earlier molecular studies
showing that the successive accumulation of genetic changes paralleled
the severity of ductal dysplasia.6,13,15
Our data also
support the previously suggested precursor nature of the ductal lesions
and the proposed tumor progression model for pancreatic
neoplasia.14,19
Recently, two independent reports also described allelic losses at 9p, 17p, and 18q in various grades of PanINs.13,15 The allelic loss frequencies that we identified at these three regions in PanIN-1 and PanIN-3 lesions are in good agreement with those reported by Yamano and colleagues.15 PanIN-2 lesions were not included in their study.15 The allelic loss frequencies reported by Heinmöller and colleagues,13 however, are discordant with ours and those of Yamano and colleagues,15 in that they found a significant rate of allelic loss at chromosomes 9p and 18q (32 and 35%, respectively) and a somewhat lower rate at 17p (15%) in PanIN-1 lesions. In addition, their frequencies of loss at 9p and 18q did not differ significantly among PanIN-1B, -2, and -3 lesions. Finally, they did not find an accumulation of clonal genetic changes with increasing PanIN grades.13 The discrepancy in the allelic losses reported by Heinmöller and colleagues13 and those reported here and by Yamano and colleagues15 may very well be explained by differences in the histological classification systems applied to the investigated lesions.
As expected, in our study abnormal p53 and Dpc4 protein expression
patterns were most prevalent in the invasive carcinomas (67 and 50% of
carcinomas, respectively).12,14,18,20-22
Importantly,
abnormal expression of both proteins seemed to arise mainly at the
PanIN-3 stage (each near 40% of lesions, Figure 3A
). Our genetic
analyses, however, had revealed that the majority of allelic losses
occurred at the PanIN-2 stage, with 88% of moderate-grade PanIN-2
lesions harboring allelic losses (Figures 2 and 3B)
. The combined
genetic and immunohistochemical data thus support a model for
pancreatic carcinogenesis in which allelic loss precedes the mutational
event in the biallelic inactivation of the p53 and
DPC4 tumor suppressor genes. Moreover, our data suggest that
allelic loss analysis may be useful in separating PanIN-2 lesions with
low-grade dysplasia from those PanIN-2 lesions with moderate-grade
dysplasia, each potentially representing a distinct progression step
toward invasive carcinoma.
A major challenge for the near future lies in the development of early
diagnosis strategies to identify those patients at the precarcinoma
in situ stage with a high risk for developing pancreatic
carcinoma. Our identification of allelic loss as the first hit in the
inactivation of the p53 and DPC4 genes together
with the observed marked increase in allelic losses at the transition
between PanIN-2 lesions with low-grade dysplasia and those with
moderate-grade dysplasia (15 to 35% and 45 to 80% losses for the
three chromosomal regions, respectively; Figure 3A
) could form the
basis for designing studies to test the positive predictive value of
allelic losses for the development of pancreatic carcinoma in risk
patients. In this context, recent advances in the application of
fluorescence in situ hybridization technology suggest that
the analysis of pancreatic juice for allelic losses as a means for
early diagnosis may indeed become feasible.23
|
Acknowledgements |
|---|
|
Footnotes |
|---|
Supported by grants from the Deutsche Krebshilfe (to S. A. H., I. S.-W., W. S., J. L., and G. K.), from the European Community Biomed 2 Project, BMH4-CT98-3085 (to S. A. H., I. S.-W., W. S.), from the BMBF (01GB9708; to W. S., I. S. W.), and from the Ruhr-Universität-Bochum, FORUM (to S. A. H., I. S.-W.).
J. Lüttges and H. Galehdari both made equal contributions to this work.
Accepted for publication January 23, 2001.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  G. Feldmann and A. Maitra Molecular Genetics of Pancreatic Ductal Adenocarcinomas and Recent Implications for Translational Efforts J. Mol. Diagn., March 1, 2008; 10(2): 111 - 122. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. Abe, N. Fukushima, K. Brune, C. Boehm, N. Sato, H. Matsubayashi, M. Canto, G. M. Petersen, R. H. Hruban, and M. Goggins Genome-Wide Allelotypes of Familial Pancreatic Adenocarcinomas and Familial and Sporadic Intraductal Papillary Mucinous Neoplasms Clin. Cancer Res., October 15, 2007; 13(20): 6019 - 6025. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Z. Ji, F. C. Mei, J. Xie, and X. Cheng Oncogenic KRAS Activates Hedgehog Signaling Pathway in Pancreatic Cancer Cells J. Biol. Chem., May 11, 2007; 282(19): 14048 - 14055. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. Wang, H. Han, and D. D. Von Hoff Identification of an Agent Selectively Targeting DPC4 (Deleted in Pancreatic Cancer Locus 4)-Deficient Pancreatic Cancer Cells Cancer Res., October 1, 2006; 66(19): 9722 - 9730. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 F Bergmann, S Aulmann, M N Wente, R Penzel, I Esposito, J Kleeff, H Friess, and P Schirmacher Molecular characterisation of pancreatic ductal adenocarcinoma in patients under 40 J. Clin. Pathol., June 1, 2006; 59(6): 580 - 584. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. F. Hezel, A. C. Kimmelman, B. Z. Stanger, N. Bardeesy, and R. A. DePinho Genetics and biology of pancreatic ductal adenocarcinoma. Genes & Dev., May 15, 2006; 20(10): 1218 - 1249. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. E. Dowen, T. Crnogorac-Jurcevic, R. Gangeswaran, M. Hansen, J. J. Eloranta, V. Bhakta, T. A. Brentnall, J. Luttges, G. Kloppel, and N. R. Lemoine Expression of S100P and Its Novel Binding Partner S100PBPR in Early Pancreatic Cancer Am. J. Pathol., January 1, 2005; 166(1): 81 - 92. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A Khalid, R Pal, E Sasatomi, P Swalsky, A Slivka, D Whitcomb, and S Finkelstein Use of microsatellite marker loss of heterozygosity in accurate diagnosis of pancreaticobiliary malignancy from brush cytology samples Gut, December 1, 2004; 53(12): 1860 - 1865. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. X. Liang, S. K. Chambers, L. Cheng, S. Zhang, Y. Zhou, and W. Zheng Endometrial Glandular Dysplasia: A Putative Precursor Lesion of Uterine Papillary Serous Carcinoma. Part II: Molecular Features International Journal of Surgical Pathology, October 1, 2004; 12(4): 319 - 331. [Abstract] [PDF]
|
|
|
|
|
|
H. Takahashi, T. Oda, T. Hasebe, Y. Aoyagi, T. Kinoshita, M. Konishi, T. Nakagohri, K. Inoue, S. Takahashi, H. Kawahira, et al. Biologically Different Subgroups of Invasive Ductal Carcinoma of the Pancreas: Dpc4 Status According to the Ratio of Intraductal Carcinoma Components Clin. Cancer Res., June 1, 2004; 10(11): 3772 - 3779. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. J. Aguirre, N. Bardeesy, M. Sinha, L. Lopez, D. A. Tuveson, J. Horner, M. S. Redston, and R. A. DePinho Activated Kras and Ink4a/Arf deficiency cooperate to produce metastatic pancreatic ductal adenocarcinoma Genes & Dev., December 15, 2003; 17(24): 3112 - 3126. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
W. H. de Vos tot Nederveen Cappel, G. J. A. Offerhaus, M. van Puijenbroek, E. Caspers, N. A. Gruis, F. A. de Snoo, C. B. H. W. Lamers, G. Griffioen, W. Bergman, H. F. A. Vasen, et al. Pancreatic Carcinoma in Carriers of a Specific 19 Base Pair Deletion of CDKN2A/p16 (p16-Leiden) Clin. Cancer Res., September 1, 2003; 9(10): 3598 - 3605. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 B. TYCKO Genetic and Epigenetic Mosaicism in Cancer Precursor Tissues Ann. N.Y. Acad. Sci., March 1, 2003; 983(1): 43 - 54. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
N. T. van Heek, A. K. Meeker, S. E. Kern, C. J. Yeo, K. D. Lillemoe, J. L. Cameron, G. J. A. Offerhaus, J. L. Hicks, R. E. Wilentz, M. G. Goggins, et al. Telomere Shortening Is Nearly Universal in Pancreatic Intraepithelial Neoplasia Am. J. Pathol., November 1, 2002; 161(5): 1541 - 1547. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
B. Terris, E. Blaveri, T. Crnogorac-Jurcevic, M. Jones, E. Missiaglia, P. Ruszniewski, A. Sauvanet, and N. R. Lemoine Characterization of Gene Expression Profiles in Intraductal Papillary-Mucinous Tumors of the Pancreas Am. J. Pathol., May 1, 2002; 160(5): 1745 - 1754. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Hiroyasu, M. Ozeki, H. Kohda, M. Echizenya, T. Tanaka, H. Hiai, and S. Toyokuni Specific Allelic Loss of p16 INK4A Tumor Suppressor Gene after Weeks of Iron-Mediated Oxidative Damage during Rat Renal Carcinogenesis Am. J. Pathol., February 1, 2002; 160(2): 419 - 424. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. E. Wilentz, P. Argani, and R. H. Hruban Loss of Heterozygosity or Intragenic Mutation, Which Comes First? Am. J. Pathol., May 1, 2001; 158(5): 1561 - 1563. [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
