ANATOMIC PATHOLOGY
Original Article
Detection of K-ras Point Mutations in the Supernatants of
Peritoneal and Pleural Effusions for Diagnosis
Complementary to Cytologic Examination
KAZUYA YAMASHITA, CT, MT, 1 - 2 TATSURU KUBA, MT, 1 H I R O S H I S H I N O D A , MT, 1 EIKO T A K A H A S H I , MD, 1 - 2
AND I S A O OKAYASU, M D 1 ' 2
To determine whether DNA analysis can be performed using the
supernatants of body fluids after centrifugation at 2,000 rpm for 10
minutes, peritoneal or pleural effusions or bile were examined for
K-ras mutations in 34 cases of pancreatic, colorectal, gastric,
esophageal, or hepatocellular carcinoma and 15 noncancer cases.
The polymerase chain reaction products for K-ras gene codons 2 to
97 of exons 1 and 2 were generated with 41 (93%) of 44 body cavity
fluid and 5 (100%) of 5 bile samples. By the single strand conformation polymorphism method, point mutations were detected in the
ascites supernatants of 8 (89%) of 9 cases of pancreatic carcinoma. In
the remaining case, no point mutation was demonstrated because
few malignant cells were present in the ascites fluid. Furthermore,
K-ras point mutations were observed in the ascites supernatants of
2 cases of colorectal carcinoma and 1 case of gastric carcinoma. The
DNA analysis of the supernatant of ascites fluid showed a K-ras
point mutation in 3 cases of false-negative cytologic diagnosis (2
cases of pancreatic carcinoma and 1 case of colorectal carcinoma).
Direct sequencing confirmed identical point mutations in the
supernatants, whole cell pellets, malignant cells from the cytologic
smears of ascites fluid, and cancer tissues. This novel method
allows simultaneous testing for genetic abnormalities in supernatants of body fluid, after removing cells for cytologic diagnosis.
(Key words: Supernatant; Polymerase chain reaction; Ascites; K-ras;
DNA analysis) Am J Clin Pathol 1998;109:704-711.
Body fluid specimens are generally used for cytologic
diagnoses, but the differential diagnosis between neoplastic cells and reactive mesothelial cells is one of
the most frequent problems encountered in cytologic
examination. 1-5 DNA analysis with polymerase chain
reaction (PCR) amplification has been introduced for
the detection of K-ras gene mutations in pancreatic
juice and feces and for microsatellite and p53 suppressor gene c h a n g e s in u r i n e a n d s p u t u m . 6 - 1 3
However, these analyses have been performed basically with cells scraped from slides or with cells in
sediment. Material is scarce when cytologic diagnosis
is applied, and when DNA analysis is performed
with cytologic smears, materials are lost owing to the
p r o c e s s e s n e c e s s a r y for e x t r a c t i o n of D N A .
Furthermore, this method requires considerable time
before results are obtained. We performed a study to
ascertain whether DNA analysis using PCR amplification is feasible using supernatants of body fluids in
peritoneal and pleural effusions, as well as in bile,
after centrifugation to remove cells.
MATERIALS AND METHODS
Patients and Diagnosis
Fresh samples of peritoneal or pleural effusions in
41 cases (pancreatic carcinoma, 9; colorectal carcinoma,
8; gastric carcinoma, 8; hepatic cirrhosis, 6; esophageal
carcinoma, 4; chronic hepatitis, 1; hepatic cyst, 1;
chronic pancreatitis, 1; hepatocellular carcinoma, 1;
ovarian carcinoma, 1; malignant lymphoma, 1) and
bile in 5 cases (gallstones, 3; gallbladder carcinoma, 1;
gastric carcinoma, 1) were collected at Kitasato
University East Hospital, Sagamihara, Kanagawa,
Japan. Histopathologic diagnosis was used as the preferred criterion for inclusion in the study. In cases in
From the ^Department of Pathology, Kitasato University East
Hospital, and the 'Department of Pathology, Kitasato University School
of Medicine, Sagamihara, Kanagazva, Japan.
Supported in part by the Sumitomo Kinzoku Bio-Medical
Center Grant, Sumitomo Kinzoku Bioscience, Sagamihara, Japan.
Manuscript received April 18, 1997; revision accepted August
27,1997.
Address reprint requests to Mr Yamashita: Department of
Pathology, Kitasato University East Hospital, 2-1-1, Asamizodai,
Sagamihara, Kanagawa 228, Japan.
704
YAMASHITA ET AL
K-ras Mutations inAscites Supernatants
which histologic materials were not available, cytologic
diagnoses were accepted in conjunction with the
results of laboratory examinations, including gastrofiberscopy, colonofiberscopy, computed tomography, echography, and endoscopic retrograde cholang i o p a n c r e a t o g r a p h y . All cytologic slides were
reviewed, and cell numbers were counted. The ratio of
malignant to nonmalignant cells was estimated in 1 mL
of body fluid. Malignant cell clusters for DNA extraction were designated by marking on the slides.
DNA extraction and molecular analysis of the cellular components from cytologic smears was performed by the microdissection method. 5 ' 14,15 Primary
tumor and nontumor tissue specimens for molecular
analysis were obtained from freshly resected material
(cases 10, 23, 24, 27, 34, and 41) or from formalin-fixed
paraffin-embedded tissue blocks (cases 1, 7-9, 11, 12,
16,17-19, 25, 26, 28-33, 35-40,42^6).
DNA
Extraction
DNA extraction from fresh frozen tissue and formalin-fixed paraffin-embedded tissue was performed following routine procedures. 14-17 Approximately 10-mL
samples of peritoneal and pleural effusions were collected in 15-mL centrifuge tubes and separated into
supernatant and cell pellet components by centrifugation at 2,000 rpm for 10 minutes. Bile samples were
diluted fivefold with phosphate-buffered saline and
then centrifuged at 2,000 rpm for 10 minutes. Five-milliliter aliquots of supernatant were gently mixed with 5
mL of ethyl alcohol to prevent DNA degeneration.
From each sample, a precipitated pellet was produced
by centrifugation, followed by DNA extraction after
addition of 1 mL of distilled water. One-milliliter
aliquots of supernatant were transferred into 1.5-mL
centrifuge tubes, mixed with 100 (iL of lysis buffer (a
100-mmol/L concentration of tromethamine hydrochloride [Tris HC1] pH 8.7; a 500-mmol/L concentration of
potassium chloride; 3 m g / m L of protein kinase K; 4.5%
NONIDET P-40 (Nacalai Tesque, Kyoto, Japan); 4.5%
polysorbate 20; and a 200-mmol/L concentration of disodium dihydrogen ethylendiaminetetraacetate dihydrate (EDTA)16 and incubated at 55°C for 3 hours. Each
sample was then purified with phenol-chloroform, precipitated with ethyl alcohol, and resuspended in 50 uL
of distilled water. The purity of DNA in the samples
was examined by electrophoresis in 0.3 % agarose gels
(Seaplaque, FMC Bioproducts, Rockland, Me). To avoid
the possibility of cross-contamination, DNA extraction
of tissue, cell pellets, and supernatants was performed
in separate rooms on different days.
Multiplex
705
PCR
Two-microliter aliquots of DNA were amplified
with a thermal cycler (Takara Shuzo, Shiga, Japan) in
a total volume of 50 |J.L; the reaction mixture contained a 5 0 - m m o l / L concentration of KC1; a 10m m o l / L c o n c e n t r a t i o n of Tris HC1, p H 8.5; a 3m m o l / L concentration of magnesium chloride; 1%
Triton X-100 (polyethylene glycol mono-p-isooctylphenyl ether, Nacalai Tesque); 6.25 pmol of each of the
p r i m e r p a i r s ; 200 umol each of d e o x y a d e n o s i n e
triphosphate, deoxyguanosine triphosphate, deoxycytidine triphosphate, and deoxythymidine triphosphate; and 2.5 U recombinant Taq polymerase, overlaid with 50 \\L of mineral oil. The primers used for
multiplex amplification 18 were 5'-GACTGAATATAA
CTTGTGG-3', 5'-GCTATTGTTGGATCAATATTC-3'
(Takara Shuzo), 5'-GATTCCTACAGGAAGCAAGT-3'
and 5'-CTATAATGGTGAATATCTTC-3' yielding the
108-base pair (bp) and 185-bp amplified DNA fragments in the K-ras gene codons 2 to 37 of exon 1 and
codons 38 to 97 of exon 2. The cycle conditions were
an initial denaturation at 94°C for 3 minutes, two
cycles of annealing at 56°C for 1 minute, extension at
72°C for 1 minute, denaturation at 94°C for 30 seconds, and 35 cycles of 45 seconds at 56°C, 30 seconds
at 72°C, and 20 seconds at 94°C, followed by final
extension at 72°C for 3 minutes. After amplification,
10-|iL reaction mixture samples were separated on 4%
agarose gels and exposed to UV light (365 nm) after
staining with ethidium bromide. In cases in which
amplification could not be confirmed, 5 |0.L of the 100fold PCR product was reamplified with fresh reagents
under the same conditions as used in the first amplification. For all PCR runs, distilled water was used as a
negative control.
Nonradioisotopic Single Strand Conformation
Polymorphism Analysis and Nonradioisotopic
Direct Sequencing
For analysis with nonradioisotopic (non-RI) single
strand conformation p o l y m o r p h i s m (SSCP), PCR
product samples of 1 uL were diluted with 49 aL of
gel-loading buffer (95% formamide; a 4 mmol/L concentration of EDTA; 0.05% bromophenol blue; and
0.05% xylene cyanol FF) and heated at 94°C for 3
minutes, followed by quenching on ice. Then 3-uL
aliquots were immediately loaded on 18% polyacrylamide gels containing 4% glycerol, 10% formamide,
2.5% sucrose, and 0.5 x TBE (tris-HCl, bolic acid,
EDTA) buffer. 16 Electrophoresis was performed at
vol. i a I • No. 6
ANATOMIC PATHOLOGY
706
! Article
550 V for 6 hours followed by silver staining with a
Silver Stain Plus Kit (Bio-Rad Laboratories, Hercules,
Calif).
Direct sequencing was performed according to the
procedure of Werle et al 19 with slight modifications
for 14 cases (10-19, 23-25, and 27), in which abnormal
bands were detected and 21 cases (6, 9, 20-22, 26,
28-41, and 46), in which wild type bands were seen
with SSCP analysis. PCR products (10 uL), including
exonuclease I and shrimp alkaline phosphatase (USB,
Cleveland, Ohio), were incubated for 30 minutes at
37°C and heated at 80°C for 10 minutes.
Non-RI detection with chemiluminescence was
performed with a Sequencing High Plus Kit (Toyobo,
Osaka, Japan) and the previously described nonlabeling primers. The cycle sequencing steps were initial
denaturation at 94°C for 3 minutes, 35 cycles (56°C 15
seconds, 60°C 3 minutes, 94°C 20 seconds), and final
extension at 60°C for 10 minutes.
f^
RESULTS
DNA Preparation From
Supernatants
Extractable DNA was detected in all supernatants
of peritoneal or pleural effusion and bile samples
after centrifugation at 2,000 rpm for 10 minutes. High
molecular weight nucleic acids were well maintained
(Fig 1, A) with 10 ng to 30 |J,g of DNA obtained from
1-mL samples.
Multiplex
PCR
Products of exons 1 and 2 of the K-ras gene were
obtained with the supernatants in 46 of 49 cases (body
cavity fluids, 41/44 [93%]; bile, 5/5 [100%]); all had
pellets rich in cells (more than 500 x 10 3 /mL; Table 1,
Fig 1, B). In contrast, few cells (<250 x 10 3 /mL) were
found in the pellets in the remaining 3 cases in which
PCR products were not available (data not shown).
Case No.
L
M
LH
N
/c
9
9
10
10
11
11
12
12
20
20
21
21
28
28
44
44
45
45
46
46
FIG 1. A, Demonstration of extractable DNA in the supernatants of
ascites (cases 9-12, 20, 21, and 28) and bile (cases 44—46). Ten-microliter aliquots of undigested DNA were electrophoresed in an
agarose gel and stained with ethidium bromide. Lane L, undigested
X DNA; lane LH, molecular marker of HmdIII-digested X DNA. B,
Multiplex polymerase chain reaction detection of the exons 1 and 2
of the K-ras gene in the ascites supernatants in cases 9 through 12,
20, 21, 28, and 44 through 46. Lane M, molecular marker of Hinfldigested <|>xI74 DNA; Lane N / C , distilled water as a negative control. Arrows indicate the amplified products of 108-base pair (bp)
(exon 1) and 185-bp (exon 2) fragment size, kb = kilobase.
Non-RI SSCP and Non-RI Direct Sequencing
SSCP analysis of tumor tissues or malignant cell
clusters of cytologic smears s h o w e d abnormally
shifted bands in 14 cases (numbers 10-19, 23-25, and
28), including 9 (100%) of 9 cases of pancreatic carcinoma, 4 (50%) of 8 cases of colorectal adenocarcinoma,
and 1 (12%) of 8 cases of gastric adenocarcinoma
(Table 2; see Table 1).
In addition, SSCP analysis of supernatants showed
abnormally shifted bands in 8 of 9 cases of pancreatic
carcinoma (cases 10-15,17, and 18), 2 cases of colorectal adenocarcinoma (cases 19 and 23), and 1 case of
gastric adenocarcinoma (case 27). Samples in the positive cases contained more than 1% cancer cells in the
cell pellets. Identically shifted bands in the supernatant and tumor tissues or malignant cell clusters of
cytologic smears were detected in cases 10 through 15,
17 through 19, 23, and 27 (Fig 2, A; see Table 1). The
point mutations responsible were confirmed to be in
codon 12 (GGT -» GAT, Gly -> Asp in cases 10,15,17,
23 and 27; GGT -» GTT, Gly -> Val in cases 12 through
14 and 19; GGT -» CGT, Gly -» Arg in case 11) by
direct sequencing (Fig 2, B).
SSCP and direct sequencing results of samples,
including supernatants, whole cell pellets, noncancer
tissues, nonmalignant cases (cases 1-9), and microdissected mesothelial cells of cytologic smears (cases
10-41) were all wild types. The SSCP analysis showed
the same results as direct sequencing for the identification of K-ras point m u t a t i o n s in s u p e r n a t a n t s ,
AJCP • j une 1998
707
YAMASHITA ET AL
K-ras Mutations in Ascites Supematants
TABLE 1. CLINICOPATHOLOGIC DATA AND K-ras GENE MUTATIONS
K-ras Mutation
Case Clinical
No. Diagnosis
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
Histologic
Diagnosis
Hepatic
Hepatic
cirrhosis
cirrhosis
NA
Hepatic
cirrhosis
NA
Hepatic
cirrhosis
NA
Hepatic
cirrhosis
NA
Hepatic
cirrhosis
NA
Hepatic
cirrhosis
Chronic
Chronic
hepatitis
hepatitis
Hepatic
Hepatic
cyst
cyst
Chronic
Chronic
pancreatitis pancreatitis
AdenoPancreatic
carcinoma carcinoma
AdenoPancreatic
carcinoma carcinoma
AdenoPancreatic
carcinoma carcinoma
NA
Pancreatic
carcinoma
NA
Pancreatic
carcinoma
NA
Pancreatic
carcinoma
AdenoPancreatic
carcinoma carcinoma
AdenoPancreatic
carcinoma carcinoma
AdenoPancreatic
carcinoma carcinoma
AdenoColorectal
carcinoma carcinoma
NA
Colorectal
carcinoma
NA
Colorectal
carcinoma
NA
Colorectal
carcinoma
AdenoColorectal
carcinoma carcinoma
AdenoColorectal
carcinoma carcinoma
AdenoColorectal
carcinoma carcinoma
AdenoColorectal
carcinoma carcinoma
AdenoGastric
carcinoma carcinoma
AdenoGastric
carcinoma carcinoma
AdenoGastric
carcinoma carcinoma
Conclusive
Stage
Material
No. of Cells
(x l&lmL)
Tumor Cell
Ratio (%)
Cytologic
Diagnosis
Supematants
Wlwle Cell Maligtiant
Pellets
Cells'
Tissues
Wild type
(GGT)
Wild type
NA
Negative
Wild type
(GGT)
Wild type
NA
Wild type
(GGT)
NA
1214
Negative
Wild type
Wild type
NA
NA
Ascites
3450
Negative
Wild type
Wild type
NA
NA
Ascites
883
Negative
Wild type
Wild type
NA
NA
Ascites
979
Negative
Wild type
Wild type
NA
NA
Ascites
698
Negative
Wild type
Wild type
NA
Wild type
Ascites
512
Negative
Wild type
Wild type
NA
Wild type
Ascites
5430
Negative
Wild type
Wild type
NA
Wild type
IVb
Ascites
1277
30
IVb
Ascites
1784
24
IVb
Ascites
962
21
IVb
Ascites
1540
17
IVb
Ascites
677
9
IVa
Ascites
1521
1
IVb
Ascites
16430
lO"7
GAT
(Asp): 12
CGT
(Arg): 12
GTT
(Val): 12
GTT
(Val): 12
GTT
(Val): 12
GAT
(Asp): 12
Wild type
GAT
(Asp): 12
CGT
(Arg): 12
GTT
(Val): 12
GTT
(Val): 12
GTT
(Val): 12
GAT
(Asp): 12
Wild type
IVb
Ascites
930
Adenocarcinoma
Adenocarcinoma
Adenocarcinoma
Adenocarcinoma
Adenocarcinoma
Adenocarcinoma
Adenocarcinoma
Negative
IVa
Ascites
21994
Negative
GAT
(Asp): 12
Wild type
rv
Ascites
3642
3
IV
Ascites
1148
5
GTT
(Val): 12
Wild type
IV
Ascites
943
10
Wild type
Wild type
Wild type
NA
rv
Ascites
2431
62
Wild type
Wild type
Wild type
NA
HI
Ascites
2040
Adenocarcinoma
Adenocarcinoma
Adenocarcinoma
Adenocarcinoma
Negative
GAT
(Asp): 12
GTT
(Val): 12
GTT
(Val): 12
Wild type
GAT
GAT
(Asp): 12 (Asp): 12
CGT
CGT
(Arg): 12 (Arg): 12
GTT
GTT
(Val): 12 (Val): 12
GTT
NA
(Val): 12
GTT
NA
(Val): 12
GAT
NA
(Asp): 12
GAT
GAT
(Asp): 12 (Asp): 12
NA
GAT
(Asp): 12
NA
GTT
(Val): 12
GTT
GTT
(Val): 12 (Val): 12
Wild type
NA
Ascites
843
Negative
GAT
(Asp): 12
Wild type
NA
i
GAT
(Asp): 12
Wild type
i
Ascites
968
Negative
Wild type
Wild type
NA
rv
Ascites
715
Negative
Wild type
Wild type
NA
GAT
(Asp): 12
GAC
(Asp): 13
GAC
(Asp): 13
Wild type
rv
Ascites
4193
90
rv
Ascites
1413
30
GAT
(Asp): 12
Wild type
GAT
(Asp): 12
Wild type
GAT
GAT
(Asp): 12 (Asp): 12
Wild type Wild type
rv
Ascites
1453
8
Adenocarcinoma
Adenocarcinoma
Adenocarcinoma
Wild type
Wild type
Wild type Wild type
Ascites
770
Negative
Ascites
798
Ascites
NA
Continued on page 708
Vol. 109 • No. 6
708
ANATOMIC PATHOLOGY
Original Article
TABLE 1. CLINICOPATHOLOGIC DATA AND K-RAS GENE MUTATIONS (CONTINUED)
K-ras Mutation
Case Clinical
No. Diagnosis
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
Histologic
Diagnosis
Conclusive
Material
Stage
Gastric
Adenocarcinoma carcinoma
Gastric
Adenocarcinoma carcinoma
Gastric
Adenocarcinoma carcinoma
Gastric
Adenocarcinoma carcinoma
Gastric
Adenocarcinoma carcinoma
Hepatocellular Hepatocellular
carcinoma carcinoma
Esophageal Squamous cell
carcinoma carcinoma
Esophageal Squamous cell
carcinoma carcinoma
Esophageal Squamous cell
carcinoma carcinoma
Esophageal Squamous cell
carcinoma carcinoma
Ovarian
Adenocarcinoma carcinoma
Malignant Malignant
lymphoma lymphoma
Gallstone Cholecystitis
Gallstone Cholecystitis
Gallstone Cholecystitis
AdenoGastric
carcinoma carcinoma
Gallbladder
Adenocarcinoma carcinoma
No. of Cells
(x lfilmL)
Tumor Cell
Ratio (%)
Cytologic
Diagnosis
Wild type
Wild type
Wild type Wild type
Wild type
Wild type
Wild type Wild type
Wild type
Wild type
NA
Wild type
Supematants
Whole Cell Malignant
Pellets
Cells'
Tissues
rv
Ascites
788
72
rv
Ascites
1661
0.1
m
Ascites
1686
Adenocarcinoma
Adenocarcinoma
Negative
m
Ascites
987
Negative
Wild type
Wild type
NA
Wild type
m
Ascites
2341
Negative
Wild type
Wild type
NA
Wild type
IV
Ascites
762
Negative
Wild type
Wild type
NA
Wild type
rv
Ascites
840
Negative
Wild type
Wild type
NA
Wild type
rv
Pleural
effusion
Pleural
effusion
Pleural
effusion
Pleural
effusion
Pleural
effusion
Bile
Bile
Bile
Bile
930
Negative
Wild type
Wild type
NA
Wild type
824
Negative
Wild type
Wild type
NA
Wild type
Squamous cell
carcinoma
Adenocarcinoma
Malignant
lymphoma
Negative
Negative
Negative
Negative
Wild type
Wild type
Wild type Wild type
Wild type
Wild type
Wild type Wild type
Wild type
Wild type
Wild type Wild type
Wild type
Wild type
Wild type
Wild type
Wild type
Wild type
Wild type
Wild type
Wild type
Wild type
i
rv
Ic
rv
ra
i
Bile
1283
20
1189
90
8536
17
ND
ND
ND
ND
ND
Adenocarcinoma
NA
NA
NA
NA
Wild type
Wild type
Wild type
Wild type
Wild type Wild type
NA = not available; ND = not determined.
"Obtained from cytologic smears by microdissection.
TABLE 2. COMPARISON OF K-ras GENE MUTATIONS AND CYTOLOGIC DIAGNOSIS IN ASCITES
K-ras Gene Mutations*
Histologic or
Cytologic Diagtiosis
Cytologic
Diagnosis
Hepatic cirrhosis and other benign lesions (n = 9)
Positive, 0
Negative, 9
Positive, 7
Negative, 2
Positive, 4
Negative, 4
Pancreatic carcinoma (n = 9)
Colorectal carcinoma (n = 8)
Gastric carcinoma (n = 8)
Supematants
Positive, 5
Negative, 3
Wliole Cell
Pellets
Malignant
Cells
Mutant type, 0 Mutant type, 0 Mutant type, 0
Wild type, 9
Wild type, 9
NA, 9
Mutant type, 8 Mutant type, 7 Mutant type, 7
Wild type, 1
Wild type, 2
NA, 2
Mutant type, 2 Mutant type, 2 Mutant type, 1
Wild type, 6
Wild type, 6
Wild type, 3
NA,4
Mutant type, 1 Mutant type, 1 Mutant type, 1
Wild type, 7
Wild type, 7
Wild type, 4
NA,3
Tissues
(Primary carcinoma)
Mutant type, 0
Wild type, 4; NA, 5
Mutant type, 6
NA,3
Mutant type, 4
Wild type, 1
NA,3
Mutant type, 1
Wild type, 7
NA = not available.
*K-rflS gene including exons 1 and 2 was confirmed by polymerase chain reaction-single strand conformation polymorphism analysis and direct sequencing, both of which showed completely consistent results.
AJCP • June 1998
YAMASHITA ET AL
709
K-ras Mutations in Ascites Supernatants
Case 12
B
Case 12
A
Codon12 .
GGT/GTT
(Gly/Val) I
G
C
i
whole cell pellets, malignant cell clusters of cytologic
smears, and tumor tissues in 34 cases (see Table 2).
Definite peritoneal dissemination was found in cases
10 through 12, 19, and 27, in which cancer cells were
observed in the histologic sections and cytologic smears
of ascites fluid. An identical point mutation in the K-ras
gene was also evident in the supernatant of ascites
fluid, the primary tumor, or the whole cell pellets in
cases 17,18, and 23, in which gross and cytologic examinations at autopsy failed to suggest peritoneal dissemination. In these cases, positive cytologic diagnoses were
not available because aggregation of fibrin occurred or
inflammatory cells were prominent. In addition, SSCP
and direct sequencing of the supernatant of ascites fluid
revealed a K-ras point mutation in case 18, although it
was not detected in the whole cell pellet (Fig 3). In case
16, the mutant type was confirmed in the primary
tumor and microdissected malignant cell clusters of
cytologic smears, while DNA analysis of whole cell pellets and supernatants of ascites fluid failed to show a
point mutation (Fig 4). In this case, the ratio of malignant to nonmalignant cells was low (10~7%).
Case 23
T
**>
Case 23
A G
C
T
Codon 12r-^ . A:
GGT/GAT
f
(Gly/Asp)^
FIG 2. A, Nonradioisotopic single strand conformation polymorphism (SSGP) analysis of amplified products of exon 1 of the K-ras
gene from the supernatants (S), cell pellets (C), nontumor tissues
(N), and primary tumor tissues (T) of cases 12 and 23. Arrows indicate the mutant allele bands. Polymerase chain reaction (PCR)
products were adjusted to the same quantity in this analysis. B,
Nonradioisotopic direct sequencing analysis of PCR-amplified
products from supernatants of specimens with mutations detected
by SSCP analysis. Arrows indicate the positions of mutations.
DNA analysis showed a point mutation of exon 1
of the K-ras gene (codon 13, GCC -> GAC, Gly -»
Asp) only in the primary tumor and the wild type in
cells and supernatant of ascites fluid in cases 24 and
25, in which the primary tumors were well-differentiated adenocarcinomas of the colon, Tl NO MO and T2
NO MO, respectively. In these cases, the cytologic diagnoses of ascites fluid were negative, and the colorectal
carcinomas did not reach the peritoneum.
B
S
C
N
T
A
G
C
T
A
G C
T
G°T
FIG 3. A, Negative cytologic diagnosis of ascites fluid in case 18, in which a K-ras gene mutation was detected by single strand conformation
polymorphism (SSCP) and direct sequencing. The cytologic smear shows aggregated cells and fibrin (Papanicolaou, x250). B,
Nonradioisotopic SSCP analysis of amplified products of exon 1 of the K-ras gene from the supernatant (S), whole cell pellets (C), nontumor
tissues (N), and primary tumor tissues (T) in case 18. Arrows indicate the mutant allele bands. C, Nonradioisotopic direct sequencing analysis
of polymerase chain reaction-amplified products of exon 1 of the iC-ras gene from the supernatant and whole cell pellets in case 18. Arrows
indicate the positions of mutations.
Vol. 109 • No. 6
710
ANATOMIC PATHOLOGY
Orivh
Article
DISCUSSION
B
The present study demonstrated that although all
cells had been removed, highly stable DNA is obtainable from supernatants of peritoneal and pleural effusions and bile after centrifugation at 2,000 rpm for 10
minutes. Indeed, DNA prepared in this way proved
stable after centrifugation at 1,000, 1,500, 2,000, 3,000,
5,000, a n d 10,000 r p m for 10 m i n u t e s a n d w h e n
passed through a millipore filter (0.22 um in diameter;
data not shown). However, the amount of DNA was
relatively small in comparison with that from similar
quantities of serum or blood plasma. 20-24 Since DNA
recovery and amplification from cell-poor samples
containing less than 250 x 10 3 cells per milliliter were
unsuccessful in three cases (data not shown), the best
results can be expected with samples derived from
lesions exhibiting cell desquamation and destruction.
C
Cell
S
Tissue
CP MC NM NT
T
S
MC
A G C T
A G C T
a
m
m --
FIG 4. A, The positive cytologic smear of the peritoneal effusion
from a patient with pancreatic carcinoma (case 16); DNA analysis of the supernatant failed to demonstrate a K-ras gene mutation. From these two malignant cells, a mutation of exon 1 of the
K-ras gene was revealed by a combination of the microdissection m e t h o d a n d D N A a n a l y s i s ( P a p a n i c o l a o u , x250). B,
Nonradioisotopic single strand conformation p o l y m o r p h i s m
analysis of amplified products of exons 1 and 2 of the K-ras gene
from the supernatants (S), whole cell pellets (C), microdissected
malignant cells of the cytologic smear (MC), microdissected
nonmalignant cells from the cytologic smear (NM), microdissected nontumor tissues from the histologic section (NT), and
microdissected primary tumor tissues from the histologic section (T) in case 16. Arrows indicate the mutant allele bands. C,
N o n r a d i o i s o t o p i c direct sequencing analysis of p o l y m e r a s e
chain reaction-amplified products of exon 1 of the K-ras gene
from the supernatant (S) and the microdissected malignant cells
(MC) in case 16. The wild type was revealed by DNA analysis of
the supernatant, while microdissected malignant cells showed a
mutant sequence (codon 12: GGT to GAT). Arrow indicates the
position of the mutation.
Our findings show that K-ras mutation analysis of
cancer cell DNA p r e p a r e d from s u p e r n a t a n t s , in
which cancer cells account for more than 1% of the
total cell pellet, can be readily achieved with the PCRSSCP method. We also confirmed that the PCR-SSCP
method can give completely consistent results for Kras gene mutations with direct sequencing.
On the other hand, DNA analysis on the basis of the
PCR method failed to show K-ras gene mutations in the
supernatant or whole cell pellet of ascites fluid that contained few malignant cells (10"7%) within numerous
n o n t u m o r cells in case 16. The limitations of this
method suggest that DNA analysis of body cavity fluid
may have false-negative results only when very few
malignant cells are included. However, even with a
lower proportion of cancer cells, it should be possible to
gain reliable results by performance of mismatch PCR
or restriction fragment length polymorphism-PCR, as
reported for cell smears and suspensions. 5,8,25-27
The identical point mutations in codons 2 to 97 of exons
1 and 2 of the K-ras gene in primary pancreatic adenocarcinomas, colorectal carcinomas, or gastric carcinoma, malignant cell clusters of cytologic smears, whole cell pellets,
and supernatants of peritoneal effusions, confirmed in
cases 10 through 12,19, and 27 with peritoneal dissemination histologically detected at autopsy or surgical operation,
provide strong evidence of the efficacy of the present
approach for DNA analysis. While a point mutation of the
K-ras gene was detected in primary colorectal carcinomas
in cases 24 and 25, the wild type gene was found in the
supernatant and whole cell pellet of ascites fluid, consistent with the negative result for peritoneal cussemination
from the cytologic diagnosis. However, the detection in
the supernatant and/or cell pellet from the ascites fluid in
AJCP • June 1998
YAMASHITA ET AL
711
K-ras Mutations it Ascites Supernatants
cases 17,18, and 23 of the same K-ras point mutation as in
the primary tumor indicates that use of this approach may
help avoid false-negative diagnoses, which are more likely
with diagnoses made on the basis of cytologic examination alone. Furthermore, because SSCP and direct
sequencing analyses showed a relatively stronger intensity
of abnormally shifted bands from the supernatant than the
whole cell pellet in case 18 (Fig 3), the DNA fragment
derived from destroyed malignant cells may be available
in the supernatant of the body cavity fluid.
The present study results are in accordance with
the reports in the literature of point mutations of
exons 1 and 2 of the K-ras gene in most pancreatic carcinomas. 5/8_11 ' 28 Because no K-ras gene point mutations were detected in mesothelial cells in patients
with or without malignant tumors, the positive results
for ascites fluid imply that cancer invasion or metastasis to the peritoneum had occurred.
The application of this method for analysis of various
genes, including p53^9 as well as microsatellite21,22 and
imrnunoglobulin heavy chain rearrangement,30 using the
supernatant of body cavity fluid as the material for study,
should facilitate definite diagnosis of minute infiltration or
metastasis by cancers of different types. Thus, we conclude that DNA analysis of supernatants from ascites,
pleural effusions, and bile is a useful tool with the following merits: (1) tests can be performed with no loss of cytologic smear samples in cases in which tissue material is not
available for histologic diagnosis; and (2) comparison of
oncogene changes allows a check for false-negative cytologic diagnoses. In addition to ordinary histopathologic
and cytologic examinations, this approach should find
broad application for molecular-pathologic diagnoses.
REFERENCES
1. El-Habashi A, El-Morsi B, Freeman SM, et al. Tumor oncogenic
expression in malignant effusions as a possible method to
enhance cytologic diagnostic sensitivity: an immunocytochemical study of 87 cases. Am ] Clin Pathol. 1995;103:206-214.
2. Bailey ME, Brown RW, Mody DR, et al. Ber-EP4 for differentiating adenocarcinoma from reactive and neoplastic mesothelial cells in serous effusions. Acta Cytol. 1996;40:1212-1216.
3. Zoppi JA, Pellicer EM, Sundblad AS. Diagnostic value of p53
p r o t e i n in the s t u d y of s e r o u s effusions. Acta Cytol.
1995;39:721-724.
4. Nance KV, Silverman JF. Immunocytochemical panel for the
identification of malignant cells in serous effusions. Am J
Clin Pathol. 1991;95:867-874.
5. Apple SK, Hecht JR, Novak JM, et al. Polymerase chain reaction-based K-ras mutation detection of pancreatic adenocarcinoma in r o u t i n e cytology s m e a r s . Am } Clin Pathol.
1996;105:321-326.
6. Mao L, Schoenberg MP, Scicchitano M, et al. Molecular detection of primary bladder cancer by microsatellite analysis.
Science. 1996;271:659-662.
7. Ricevuto E, Ficorella C, Fusco C, et al. Molecular diagnosis of
p53 mutations in gastric carcinoma by touch preparation. Am
I Pathol. 1996;148:405^13.
Vol. 109
8. Van Laethem JL, Vertongen P, Deviere J, et al. Detection of c-Kiras gene codon 12 mutations from pancreatic duct brushings
in the diagnosis of pancreatic tumours. Cut. 1995;36:781-787.
9. Watanabe H, Miyagi C, Yamaguchi Y, et al. Detection of K-ras
point mutations at codon 12 in pancreatic juice for the diagnosis of pancreatic cancer by hybridization protection assay:
a simple method for the determination of the types of point
mutations, jpn J Cancer Res. 1996;87:466-474.
10. Tada M, Omata M, Kawai S, et al. Detection of ras gene mutations in pancreatic juice and peripheral blood of patients with
pancreatic adenocarcinoma. Cancer Res. 1993;53:2472-2474.
11. van Es IM, Polak MM, van den Berg FM, et al. Molecular markers for diagnostic cytology of neoplasms in the head region
of the pancreas: mutation of K-ras and overexpression of the
p53 protein product. / Clin Pathol. 1995;48:218-222.
12. Nollau P, Moser C, Wagener C. Isolation of DNA from stool and
b o d i l y fluids for PCR amplification.
Biotechniques.
1996;20:784-788.
13. Caldas C, Hahn SA, Hruban RH, et al. Detection of K-ras mutations in the stool of patients with pancreatic adenocarcin o m a and p a n c r e a t i c ductal h y p e r p l a s i a . Cancer Res.
1994;54:3568-3573.
14. Goelz SE, Hamilton SR, Vogelstein B. Purification of DNA from
formaldehyde fixed and paraffin embedded human tissue.
Biochem Biophys Res Comm. 1985;130:118-126.
15. Zhuang Z, Bertheau P, Emmert-Buck MR, et al. A microdissection technique for archival DNA analysis of specific cell populations in lesions <1 mm in size. Am J Pathol. 1995;146:620-625.
16. Yamashita K, Tatebayashi T, Shinoda H, et al. Simplified rapid
non-radioactive PCR-SSCP method applied to K-ras mutation analysis. Pathol Int. 1996;46:801-804.
17. Iwamoto KS, Mizuno T, Ito T, et al. Feasibility of using decadesold archival tissues in molecular oncology/epidemiology.
Am J Pathol. 1996;149:399-406.
18. Cama A, Palmirotta R, Curia MC, et al. Multiplex PCR analysis
and genotype-phenotype correlations of frequent APC mutations. Hum Mutat. 1995;5:144-152.
19. Werle E, Schneider C, Renner M, et al. Convenient single-step,
one tube purification of PCR products for direct sequencing.
Nucleic Acids Res. 1994;22:4354-4355.
20. Nawroz H, Koch W, Anker P, et al. Microsatellite alterations in
serum DNA of head and neck cancer patients. Nat Med.
1996;2:1035-1037.
21. Chen XQ, Stroun M, Magnenat JJ, et al. Microsatellite alterations in plasma DNA of small cell lung cancer patients. Nat
Med. 1996;2:1033-1034.
22. Vasioukhin V, Anker P, Maurice P, et al. Point mutations of the
N-ras gene in the blood plasma DNA of patients with
myelodysplastic syndrome or acute myelogenous leukaemia.
Br] Haematol. 1994;86:774-779.
23. Stroun M, Anker P, Maurice P, et al. Neoplastic characteristics
of the DNA found in the plasma of cancer patient. Oncology.
1989;46:318-322.
24. Stroun M, Anker P, Lyautey J, et al. Isolation and characterization of DNA from the plasma of cancer patients. Eur ] Cancer
Clin Oncol. 1987;23:707-712.
25. Levi S, Urbano-Ispizua A, Gill R, et al. Multiple K-ras codon 12
mutations in cholangiocarcinomas demonstrated with a sensitive polymerase chain reaction technique. Cancer Res.
1991;51:3497-3502.
26. Tada M, Omata M, Kawai S, et al. Detection of ras gene mutations in pancreatic juice and peripheral blood of patients with
pancreatic adenocarcinoma. Cancer Res. 1993;53:2472-2474.
27. Load M. Polymerase chain reaction-based methods for the
detection of mutations in oncogenes and tumor suppressor
genes. Hum Pathol. 1994;25:564-571.
28. Bos IL. Ras oncogenes in human cancer: A review. Cancer Res.
1989;49:4682-4689.
29. Hollstein M, Sidransky D, Vogelstein B, et al. P53 mutations in
human cancer. Science. 1991;253:49-53.
30. Cano SD. PCR-based alternative for diagnosis of immunoglobulin heavy chain gene rearrangement. Diagn Mol Pathol.
1996;5:3-9.