Gastric Cancer (2010) 13: 177–185
Offprint requests to: K. Shomori
Received: August 6, 2009 / Accepted: May 12, 2010
© 2010 by
Geminin, Ki67, and minichromosome maintenance 2 in gastric
hyperplastic polyps, adenomas, and intestinal-type carcinomas:
pathobiological signifi cance
KOHEI SHOMORI1, KEISUKE NISHIHARA1, TAKAYUKI TAMURA1, SHIGERU TATEBE2, YASUSHI HORIE3, KANAE NOSAKA1,
TOMOHIRO HARUKI1, YUKI HAMAMOTO1, TATSUSHI SHIOMI1, MOTOKI NAKABAYASHI1, and HISAO ITO1
1 Division of Organ Pathology, Department of Microbiology and Pathology, Faculty of Medicine, Tottori University, 86 Nishi-cho, Yonago,
Tottori 683-8503, Japan
2 Division of Surgical Oncology, Department of Surgery, Faculty of Medicine, Tottori University, Yonago, Tottori , Japan
3 Department of Pathology, Tottori University Hospital, Yonago, Tottori, Japan
Key words Geminin · Intestinal-type · Gastric carcinoma ·
Adenoma · Ki67 · Mcm
Gastric cancer is the world’s second leading cause of
cancer mortality after lung cancer, with an estimated
933 000 new cases and 699 000 deaths per year .
Growing evidence suggests that chemotherapy can
improve disease-free intervals and overall survival [2, 3].
Further studies are needed to identify molecular
markers to help improve patients’ prognoses.
Hyperplastic polyps (HPPs) and tubular adenomas
are proliferative lesions of the gastric mucosa. Low-
grade adenomas (LGAs) are tubular structures lined
by enlarged columnar cells with homogeneously blue
pseudostratifi ed nuclei in the superfi cial portion of
the dysplastic tubules. High-grade adenomas (HGAs)
are tubular structures with architectural distortion and
pseudostratifi ed cigar-shaped nuclei throughout the epi-
thelium . High-grade adenoma (HGA) is often
diffi cult to differentiate from adenocarcinoma (ACA)
[5–7]. LGA and HGA form a morphological spectrum
with ACA. Considerable overlap currently exists
between the histopathological diagnostic criteria for
LGA, HGA, and ACA . The possible subclassifi ca-
tion of gastric mucosal proliferative lesions on the basis
of their cell kinetics could aid in their differential diag-
nosis and offer new ways of predicting lesional growth
Geminin is a 25-kDa nuclear protein with a DNA
replication-inhibitory function . Through negative
regulation of Cdt1, geminin induces the formation of
prereplication complexes by loading minichromosome
maintenance proteins (Mcm) onto chromatin . DNA
polymerase binds to this complex and initiates DNA
Background. Geminin negatively regulates Cdt1 and induces
the formation of prereplicative complexes by loading mini-
chromosome maintenance proteins (Mcm) onto chromatin
and limiting DNA replication to once per cell cycle. Recent
studies have suggested that geminin expression is a marker of
the S/G2/M phase of the cell cycle and is associated with a
poor prognosis in various human malignancies. This study
aimed to clarify the pathobiological role of geminin in intes-
tinal-type gastric carcinoma, and its relationships with mini-
chromosome maintenance 2 (Mcm2) and Ki67 expression.
Methods. We performed western blot analysis of seven human
gastric cancer cell lines, and immunohistochemical analysis of
72 gastric mucosal lesions and 128 surgically removed advanced
intestinal-type gastric carcinomas. Double-labeling immuno-
fl uorescence was performed to identify the coexpression of
geminin and Ki67.
Results. Geminin was detected in all cell lines. Geminin label-
ing indices (LIs) in hyperplastic polyps, low-grade adenomas,
high-grade adenomas, and intestinal-type adenocarcinomas
were 3.9%, 10.5%, 18.6%, and 27.2%, respectively. The equiv-
alent LIs for Ki67 and Mcm2 were 17.7%, 42.2%, 52.6%, and
59.7%; and 26.7%, 70.0%, 67.8%, and 77.8%, respectively.
Double-labeling immunofl uorescence revealed coexpression
of geminin and Ki67 in both normal and tumor cells. The LI
for geminin was signifi cantly correlated with N stage, Interna-
tional Union Against Cancer (UICC) stage, Mcm2 LI, and
Ki67 LI. Patients in stages I-IV and stage III with higher LIs
for geminin (>25%) had signifi cantly worse prognoses (P <
0.05 and P < 0.04, respectively). Univariate Cox regression
analysis indicated that the overall survival of stage I-IV tumors
was signifi cantly correlated with high geminin LIs (relative
risk [RR] = 1.94; P = 0.04).
Conclusions. Geminin expression might refl ect the biological
nature of gastric intramucosal neoplasms and could be a pos-
sible prognostic marker in advanced intestinal-type gastric
178 K. Shomori et al.: Geminin in intestinal gastric carcinomas
synthesis in S phase. This process repeats and limits
DNA replication to once per cell cycle. To maintain this
process, the level of geminin fl uctuates throughout the
cell cycle, being almost absent in G1 phase, but high in
S and G2 phases and in the initial stage of M phase .
Geminin is thus responsible not only for regulation of
the cell cycle, but also for the genomic integrity of cells.
Geminin has been suggested to have tumor-suppressive
functions, but previous reports have demonstrated
that its expression was correlated with increasing
tumor grade [12–15], and with poor prognosis in breast
cancer patients . Thus, discrepancies exist between
geminin molecular function and the outcomes of clinical
No study has examined the association between
geminin expression and cell kinetics in gastric tumors
and their prognostic signifi cance in advanced gastric
cancers. The aim of this study was to determine whether
expression of the replication-inhibitory protein, geminin,
could be useful for evaluating tumor proliferative
potential and predicting outcome in patients with gastric
carcinomas. We also aimed to clarify the relationships
between the expression of geminin and the expression
of minichromosome maintenance 2 (Mcm2) and Ki67.
Patients, materials, and methods
Cell culture and immunoblotting
Human gastric cancer cell lines TMK-1 (poorly dif-
ferentiated adenocarcinoma; PDA), MKN-1 (ade-
nosquamous carcinoma), MKN-7 (well-differentiated
adenocarcinoma; WDA), MKN-45 (PDA), MKN-74
(WDA), KATO III (signet-ring cell carcinoma), and
HSC-39 (PDA) were routinely cultured in RPMI 1640
(Nissui Pharmaceutical, Tokyo, Japan) supplemented
with 10% fetal bovine serum (JRH Biosciences, Lenexa,
KS, USA) and 100 U/ml penicillin-streptomycin-L-glu-
tamine (Wako Pure Chemical Industries, Osaka, Japan)
at 37 °C in a 5% CO2 atmosphere. Cells were harvested
for protein extraction as they reached subconfl uence.
Each 50 μg of proteins was electrophoresed on sodium
dodecyl sulfate-polyacrylamide gels, and transferred to
polyvinylidene fl uoride membranes. The membranes
were incubated with primary antibodies at the following
dilutions: rabbit anti-geminin antibody (1: 500; Santa
Cruz Biotechnology, Santa Cruz, CA, USA) and mouse
anti-β-actin antibody (1: 2000; Sigma-Aldrich, St. Louis,
MO, USA). The membranes were then incubated with
horseradish peroxidase-conjugated anti-rabbit IgG
(1: 2000; MBL, Nagoya, Japan) and anti-mouse IgG
(1: 2000; MBL). Immunoconjugates were detected by
enhanced chemiluminescence (Amersham Pharmacia
Biotech, Piscataway, NJ, USA).
Patients and tissue specimens
Seventy-two patients underwent endoscopic submuco-
sal dissection (ESD) of gastric mucosal proliferative
lesions. The lesions included 14 HPPs, 22 LGAs, 10
HGAs, and 26 ACAs. All but three depressed HGA
lesions formed grossly fl at or elevated lesions.
A total of 128 intestinal-type gastric carcinomas were
selected on the basis of the cancer invasion being deeper
than the muscularis propria (International Union
Against Cancer [UICC]; T2–T4). Intestinal-type gastric
carcinomas were defi ned as distinct glandular lumina
often accompanied by papillary fold formation or solid
components with distinct epithelial cords, as described
in Lauren’s classifi cation . All patients underwent
radical en-bloc gastrectomy. No patients received pre-
operative radiochemotherapy. The median age was 70.0
years (range, 48–89 years). The mean follow-up period
was 33.6 months, with a maximum of 161 months. There
were 100 male and 28 female patients. Slides of each
tumor were reviewed by pathologists with no knowl-
edge of the clinical outcome, who assigned a pathologi-
cal diagnosis according to the TNM staging system ,
the Japanese classifi cation of gastric carcinoma , and
the Lauren classifi cation . Details of the tumor series
and clinicopathological parameters were obtained from
clinical records. The study was approved by the Local
Research Ethics Committee of Tottori University,
Faculty of Medicine (No.283).
Representative tissue blocks containing a sample of
peripheral tumor were selected for each patient. For-
malin-fi xed, paraffi n-embedded tissues were cut onto
silane-coated slides at 4-μm thickness. The slides were
dewaxed in xylene and rehydrated through graded alco-
hols. For antigen retrieval, the sections for Ki67 were
incubated in 0.01% (w/v) trypsin including 0.01% (w/v)
CaCl2 in 0.05 M Tris-buffered saline (pH 7.6) for 20 min
at 37 °C. The slides were then heated in a microwave in
0.1 M citrate buffer (pH 6.0) at 95 °C for 20 min. Fol-
lowing antigen retrieval, endogenous peroxidase activ-
ity was quenched with 0.6% H2O2 in absolute methanol
for 30 min at room temperature. The primary antibody
was applied at 4 °C overnight at the following dilutions:
rabbit anti-geminin antibody (1: 250), mouse anti-Ki67
antibody (1: 100; MIB-1; DAKO, Glostrup, Denmark),
and rabbit anti-Mcm2 antibody (1: 500; Santa Cruz
Biotechnology). Antibodies were detected using the
streptavidin-biotin immunoperoxidase method using
Histofi ne SAB reagents (Nichirei, Tokyo, Japan). Sec-
tions were then stained with 3,3-diaminobenzideine tet-
rahydrochloride (MP Biomedicals, Solon, OH, USA),
counterstained with Meyer’s hematoxylin, differenti-
K. Shomori et al.: Geminin in intestinal gastric carcinomas 179
ated in water at 37 °C, dehydrated, and cleared in xylene.
Incubation without the primary antibody was used as a
Evaluation of geminin, Ki67, and Mcm2 labeling
To determine the LIs in each tumor, slides were evalu-
ated in a low-power fi eld (×40) to identify regions with
the most intense staining. Because of the uneven distri-
bution of positive cells in the ESD specimens, all the
tumor cells were counted in at least seven glands. In
the advanced gastric cancers, LIs were evaluated at the
tumor’s invasive front, because of lower LIs in the
center of the tumor. High-power fi elds (×400) were cap-
tured from the selected areas using a charge-coupled
device camera. Counts were performed using a Flovel
Image Filing System (FlvFs; Flovel, Tachikawa, Japan).
More than 700 nuclei were counted for each case. Quan-
titative analysis was performed by an assessor unaware
of the clinicopathological variables. The LI was calcu-
lated by dividing the number of positive cells by the
total number of cells counted.
Double-labeling immunofl uorescence
Paraffi n-embedded tissue sections were processed for
double-labeling immunofl uorescence, as described pre-
viously . Sections were incubated with a mixture of
anti-geminin (1: 100) and anti-Ki67 antibodies (1: 100).
They were incubated with R-phycoerythrin-conjugated
goat anti-rabbit IgG (1: 200; Molecular Probes, Eugene,
OR, USA) followed by Alexa Fluor 488-conjugated
rabbit anti-mouse IgG (1: 200; Molecular Probes) and
observed under a fl uorescence microscope (Eclipse 80i;
Nikon Instruments, Tokyo, Japan).
Correlations among LIs for geminin, Mcm2, and Ki67
were analyzed using Pearson’s correlation coeffi cient
test. Correlations between LIs and clinicopathological
variables were analyzed using the Mann-Whitney U-test
for the analysis of two values and the Kruskal-Wallis
rank test for more than two values. Kaplan-Meier cumu-
lative survival curves were constructed for the LIs .
A log-rank test was used to assess the statistical signifi -
cance of the curves. Univariate and multivariate Cox
regression models for survival were used to evaluate the
contributions of various factors. P < 0.05 was considered
to indicate statistical signifi cance. The analyses were
carried out using Statcel ver. 2 for Windows (OMS Pub-
lishing, Saitama, Japan) and Dr. SPSS II for Windows
(SPSS, Tokyo, Japan).
Geminin protein expression in human gastric cancer
Geminin protein was detected at various levels in
all seven gastric cancer cell lines, regardless of their
histological type, as two clear bands of 25 and 28 kDa
Immunohistochemistry and double-labeling
In the 72 ESD specimens, geminin, Mcm2, and Ki67
expressions were noted exclusively in the nucleus in
both normal and tumor cells, except for prophase mitotic
cells, where geminin immunoreactivity was found in the
cytoplasm. All protein-positive cells were observed in
the neck of both non-neoplastic fundic and pyloric
glands, in the lower portion of intestinal metaplasia,
and in the germ center of lymphoid follicles in the
lamina propria that corresponded to the proliferative
zone. Geminin was localized in the nucleus, but in
prophase and metaphase it was diffusely distributed in
cytoplasm, and it was not shown in anaphase cells.
Mcm2 and Ki67 were localized in the nucleus and nucle-
olus. In mitotic cells, Mcm2 was localized in the cyto-
plasm, and Ki67 was localized in the cytoplasm and
microscopy demonstrated that all geminin-positive cells
(Fig. 2A; red) also expressed Ki67 (Fig. 2B; green),
giving a yellow signal when the images were merged
(Fig. 2C). No geminin-only-positive cells were detected.
Similar results were found in non-neoplastic mucosa
Geminin- (Fig. 3A-D), Ki67- (Fig. 3 E-H), and Mcm2
(Fig. 3 I-L)-positive cells were variably located in the
intramucosal proliferative lesions. In HPPs, there were
a few cells expressing the three proteins in the bottom
of the glands (Fig. 3A, E, I). In LGAs, geminin-positive
Fig. 1. Western blot analysis of geminin expression in seven
human gastric cancer cell lines. β-Actin was used as an internal
control. The molecular markers are indicated on the right.
Geminin is noted at 25 kDa and 28 kDa
180 K. Shomori et al.: Geminin in intestinal gastric carcinomas
cells were localized only in the middle zone of the glands
(Fig. 3B), while Ki67- and Mcm2-positive cells were
localized from the surface to the middle zone of the
glands (Fig. 3F, J). In HGAs, protein-positive cells were
located from the surface to the middle zone of the
glands (Fig. 3C, G, K). In ACAs, positive cells were dif-
fusely distributed (Fig. 3D, H, L).
The mean LIs of the proteins in the ESD lesions are
shown in Table 1. When all lesion types were compared,
the Mcm2 LIs were signifi cantly higher than those for
Fig. 2A–C. Double-labeling immunofl uorescent staining for geminin and Ki67 in human gastric cancer specimens. Geminin-
positive cells (A; red) coexpressed Ki67 (B; green). When the images were merged, many tumor cells showed yellow signals,
indicating combined geminin and Ki67 positivity (C). ×400
Fig. 3A–L. Immunohistochemical analysis of geminin, Ki67, and minichromosome maintenance 2 (Mcm2). Each of the fi elds
used for analysis of geminin (A–D), Ki67 (E–H), and Mcm2 (I–L) were photographed from the same area of each lesion for
hyperplastic polyps (HPP; A, E, I), low-grade adenomas (LGA; B, F, J), high-grade adenomas (HGA; C, G, K), and adenocar-
cinomas (ACA; D, H, L). ×100
K. Shomori et al.: Geminin in intestinal gastric carcinomas 181
Ki67, which were signifi cantly higher than those for
geminin (P < 0.0001). The geminin LIs of HPPs, LGAs,
HGAs, and ACAs were 3.9%, 10.5%, 18.6%, and 27.2%,
respectively (P < 0.001). The Ki67 LI in HPPs was
17.7%, which was signifi cantly lower than that in LGAs
(42.2%), HGAs (52.6%), and ACAs (59.7%). The Ki67
LI in LGAs was signifi cantly lower than those in HGAs
and ACAs. The Mcm2 LI in HPPs (26.7%) was also
signifi cantly lower than those in all the other lesions
(LGA; 70.0%, HGA; 67.8%, ACA; 77.8%) (P < 0.001).
Correlations between geminin LI and
clinicopathological profi le in advanced gastric cancer
The mean LIs for Mcm2, Ki67, and geminin were 64.3%
(median, 66.1%), 47.4% (median, 45.5%), and 24.6%
(median, 23.7%), respectively, in the 128 advanced
gastric cancers. Table 2 shows the relationship between
geminin LIs and clinicopathological profi les. We chose
LI cutoff values for geminin (25%), Mcm2 (65%), and
Ki67 (45%), the values being almost equal to the mean/
Table 1. Geminin, Ki67, and Mcm2 expression in endoscopic submucosal dissection
Data shown are mean percentages of positive cells (interquartile ranges)
Mcm2, minichromosome maintenance 2; HPP, hyperplastic polyp; LGA, low-grade adenoma;
HGA, high-grade adenoma; ACA, adenocarcinoma
Table 2. Correlation between the geminin labeling index (LI) and clinicopathological
P value Mean LI (%)LI 25%LI >25%
tub1, well-differentiated type; tub2, moderately differentiated; por1, poorly differentiated and
solid type; UICC, International Union Against Cancer
182 K. Shomori et al.: Geminin in intestinal gastric carcinomas
median LIs. We divided the patients into two groups
with LIs either above (high) or below (low) the cutoff
values. High LIs for geminin, Mcm2, and Ki67 were
noted in 55, 68, and 65 patients, respectively. The LIs for
geminin were signifi cantly higher in papillary and solid-
type adenocarcinomas than in well- and moderately
differentiated tubular adenocarcinomas. The LIs for
geminin were signifi cantly correlated with N stage (P <
0.02), UICC stage (P < 0.04), high Mcm2 LIs (>65%; P
< 0.0001), and high Ki67 LIs (>45%; P < 0.0001), but not
with age, sex, or T stage.
We examined the cumulative survival of the high and
low LI groups using the Kaplan-Meier method and
log-rank tests. There were no signifi cant correlations
between geminin LIs and survival for stages I and stage
II (P < 0.54 and P < 0.73, respectively; Fig. 4A,B).
However, high geminin LIs were associated with poorer
survival in the 70 stage III patients (P < 0.05; Fig. 4C).
Moreover, the 55 patients in stages I–IV with high
geminin LIs had signifi cantly poorer prognoses than
the 73 patients in stages I–IV with low geminin LIs
(P < 0.04; Fig. 4D). Mcm2 and Ki67 LIs were not cor-
related with survival in stages I-IV (Fig. 4E, F), and in
stage III (P > 0.93 and P > 0.57, respectively) (fi gure not
Univariate Cox regression analysis indicated that the
overall survival of stage I-IV tumors was signifi cantly
correlated with high geminin LIs (relative risk [RR] =
1.94; P = 0.04; Table 3). Multivariate Cox regression
analysis was used to determine the prognostic value of
the markers, age, sex, histological differentiation, T
stage, N status, and UICC stage. N status was identifi ed
as an independent prognostic factor.
The estimated molecular weight of geminin protein
is 25 kDa, but the immunoblotting of proteins
from seven gastric cancer cell lines in the present study
produced two bands, of 25 and 28 kDa. The oncosup-
pressor, retinoblastoma protein, also shows multiple
bands on western blots , and this is due to the
phosphorylation of serine/threonine sites by cdc2
kinase. Geminin is thought to be phosphorylated at both
serine/threonine and tyrosine residues by cell cycle-
dependent kinases [22, 23] and it is therefore possible
that the 28-kDa geminin band could have resulted from
Geminin blocks re-replication of the genome in the
same cell cycle. It is present during the S/G2/M phases of
the cell cycle but absent during G1, and thus acts as an
S/G2/M marker. Mcm2-positive cells are thought to
occur not only during cell cycling, but also during the G0
phase, or “dormancy” . Ki67 is expressed during all
phases of the cell cycle, except for G0 , and may be
present during cell-cycle arrest. In accordance with these
features, we found that the geminin LI was signifi cantly
lower than that of Ki67, and the latter was lower than
the Mcm2 LIs in both normal and tumor cells.
In the present study, the LIs of the three markers in HPP,
LGA, HGA, and ACA were: geminin, 3.9%, 10.5%,
18.6%, and 27.2%; Ki67, 17.7%, 42.2%, 52.6%, and
59.7%; Mcm2, 26.7%, 70.0%, 67.8%, and 77.8%. These
fi ndings are consistent with the published data [12, 19].
The low Mcm2 LIs could be explained if most HPP
cells were differentiated or out of the cell cycle, while
LGAs could have more G0 phase or dormant cells
with potential proliferative activity (high Mcm2 and low
Ki67 and geminin LIs). HGAs could include more
arrested G1 cells (high Mcm2 and Ki67, and low geminin
LIs), which would be diffi cult to differentiate from
ACA. ACAs showed the highest geminin LI among the
intramucosal neoplasms, indicative of high proliferative
activity. The ratio of geminin/Ki67 represents the per-
centage of active cycling cells in all the proliferating
cells. In the present study, the ratio of ACA (0.46) was
defi nitely higher than those of HPP (0.22), LGA (0.19),
and HGA (0.33).
Our results suggest that geminin could refl ect the
population of “active” cycling cells in S/G2/M phase in
neoplastic lesions. Cho et al.  studied the cell kinetics
Table 3. Univariate and multivariate Cox regression analyses for predictors of overall
survival in patients with advanced intestinal-type gastric carcinoma
RR95% Confi dence interval
UICC stage III
vs N (−)
vs Stage I
vs N (−)
RR, relative risk; ly, lymphatic invasion
K. Shomori et al.: Geminin in intestinal gastric carcinomas 183
Fig. 4A–F. Kaplan-Meier curves showing cumulative overall
survival rates according to the markers. Cutoff values were
chosen at 25% for geminin, 45% for Ki67, and 65% for Mcm2.
In stage I and stage II patients, a higher geminin labeling index
(LI; >25%) did not correlate with poorer survival (A and B;
P < 0.54 and P < 0.73, respectively). The cumulative survival
of the 55 patients in stage I–IV with a high geminin LI (>25%)
was signifi cantly poorer than that of the 73 patients in stage
I–IV with a low LI (≤25%) (P < 0.04; D). Patients in stage III
with a high geminin LI showed poorer survival (P < 0.05; C).
High Ki67 LI (>45%) and Mcm2 LI (>65%) in stage I–IV
patients did not correlate with poorer survival (E and F; P >
0.9 and P > 0.65, respectively)
184 K. Shomori et al.: Geminin in intestinal gastric carcinomas
of proliferative gastric lesions and found that human
growth and transformation-dependent protein expres-
sion occurred during the progression from normal
gastric mucosa to intestinal-type carcinoma and might
be associated with tumor cell proliferation. The present
study suggests that the assessment of cell kinetics using
anti-geminin, Ki67, and Mcm2 antibodies might permit
differentiation among HPPs, LGAs, HGAs, and ACAs.
Quaglia et al.  reported that the combination of
Mcm2, geminin, and Ki67 could represent a valuable
tool in differentiating hepatocellular carcinoma from
regenerative and dysplastic nodules in liver cirrhosis.
Montanari et al.  suggested that increased expres-
sion of geminin promoted cell-cycle progression and
proliferation in normal and cancerous breast cells in
vitro, suggesting that higher geminin LIs could be caus-
ally associated with the early stages of gastric carcino-
genesis. These results are consistent with our data, which
showed higher geminin, Ki67, and Mcm2 LIs in intramu-
cosal gastric cancer than in advanced gastric cancers.
Higher geminin LIs were associated with more pro-
liferative tumors and poorer prognoses. We previously
demonstrated the prognostic signifi cance of geminin in
human colorectal and oral cancers [19, 27]. Other studies
have reported the potential prognostic signifi cance of
geminin in breast, kidney, brain, and oral tumors [15, 12,
28, 29], in line with our results.
We recently demonstrated the prognostic value of a
combination of Mcm2 and Ki67 for the assessment of
stage III diffuse-type gastric cancer, but not the intesti-
nal type . Thus, the addition of geminin increased
the predictive prognostic value of Mcm2 and Ki67.
Previous reports found that geminin expression was
a poor prognostic marker of overall survival in patients
with breast, kidney, and colorectal cancers [15, 12, 19].
However, Shrestha et al.  reported that higher
geminin LIs might be associated with longer survival in
patients with high-grade astrocytic tumors with pre-
operative chemotherapy. We also found that higher
geminin LIs were associated with better prognoses in
patients with stage III–IV oral squamous cell carcinoma
receiving preoperative radiochemotherapy . These
confl icting results could be explained by an increase in
radiochemosensitivity associated with geminin overex-
pression; some studies have shown that cells synchro-
nized in G1 phase were more radioresistant than cells
in the S, late G2, and M phases [31, 32].
There are some limitations of the present study. We
omitted diffuse-type gastric cancers from advanced
gastric cancers because of diffi culties in counting the
poorly cohesive tumor cells mixed with fi broblasts and
infl ammatory cells. Second, it is possible that there were
too few cases to demonstrate the relationship between
cell kinetics, including those of geminin, and patient
outcomes in the multivariate Cox regression analysis.
Many studies have evaluated the relationships between
advanced gastric cancer prognosis and known prolifera-
tive markers, such as Ki67. Some studies found that high
Ki67 LIs were correlated with poorer survival [33, 34],
whereas others did not . In our previous study,
Mcm2 and Ki67 expression did not correlate with any
clinicopathological factors, e.g., lymph node status or
vascular invasion in stage III intestinal-type gastric
cancers , which is consistent with our present data.
Although both Mcm2 and Ki67 occur in active cycling
cells, it is possible that they also occur in slow/arrested
G1, and Mcm2 is expressed even in G0 phase [36, 37].
These markers may not therefore refl ect the prolifera-
tive activity of tumor cells and may not be suitable
predictive markers in patients with malignant tumors.
In conclusion, geminin expression might facilitate the
proliferation of intramucosal neoplasms of the stomach.
The geminin LI could serve as a marker for active
cycling tumor cells and have prognostic signifi cance in
intestinal-type gastric carcinomas. The role of geminin
might also have signifi cant implications for the develop-
ment and application of targeted therapies.
Acknowledgments We thank Mr. N. Itaki, Ms. M. Iwatani,
and Ms. C. Yamasaki (Division of Organ Pathology,
Faculty of Medicine, Tottori University) for their skillful
technical assistance and Ms. Y. Tokuoka for assistance
with collecting patient information. This work was sup-
ported in part by a Tottori University Faculty of Medi-
cine Research Grant for 2008 (No. 310530).
1. Kamangar F, Dores GM, Anderson WF. Patterns of cancer inci-
dence, mortality, and prevalence across fi ve continents: defi ning
priorities to reduce cancer disparities in different geographic
regions of the world. J Clin Oncol 2006;24:2137–50.
2. Cunningham D, Allum WH, Stenning SP, Thompson JN, Van de
Velde CJ, Nicolson M, et al. Perioperative chemotherapy versus
surgery alone for resectable gastroesophageal cancer. N Engl J
3. Ajani JA, Moiseyenko VM, Tjulandin S, Majlis A, Constenla M,
Boni C, et al. Quality of life with docetaxel plus cisplatin and fl uo-
rouracil compared with cisplatin and fl uorouracil from a phase III
trial for advanced gastric or gastroesophageal adenocarcinoma:
The V-325 study group. J Clin Oncol 2007;25:3210–6.
4. Fenoglio-Preiser C, Carneiro F, Correa P, Guilford P, Lambert R,
Megraud F, et al. Gastric carcinoma. In: Hamilton SR, Aaltonen
LA, editors. World Health Organization classifi cation of tumours.
Pathology and genetics of tumours of the digestive system. Lyon,
France: IARC; 2000. p. 47–8.
5. Kamiya T, Morishita T, Asakura H, Miura S, Munakata Y, Tsuchiya
M. Long-term follow-up study on gastric adenoma and its relation
to gastric protruded carcinoma. Cancer 1982;50:2496–503.
6. Cristallini EG, Ascani S, Bolis GB. Association between histologic
type of polyp and carcinoma in the stomach. Gastrointest Endosc
7. Lee JH, Abraham SC, Kim HS, Nam JH, Choi C, Lee MC, et al.
Inverse relationship between APC gene mutation in gastric ade-
K. Shomori et al.: Geminin in intestinal gastric carcinomas 185
nomas and development of adenocarcinoma. Am J Pathol 2002;
8. Misdraji J, Lauwers GY. Gastric epithelial dysplasia. Semin Diagn
9. McGarry TJ, Kirschner MW. Geminin, an inhibitor of DNA rep-
lication, is degraded during mitosis. Cell 1998;93:1043–53.
10. Wohlschlegel JA, Dwyer BT, Dhar SK, Cvetic C, Walter JC, Dutta
A. Inhibition of eukaryotic DNA replication by geminin binding
to Cdt1. Science 2000;290:2309–12.
11. Nishitani H, Lygerou Z. Control of DNA replication licensing in
a cell cycle. Genes Cells 2002;7:523–34.
12. Wohlschlegel JA, Kutok JL, Weng AP, Dutta A. Expression of
geminin as a marker of cell proliferation in normal tissues and
malignancies. Am J Pathol 2002;161:267–73.
13. Dudderidge TJ, Stoeber K, Loddo M, Atkinson G, Fanshawe T,
Griffi ths DF, et al. Mcm2, geminin, and Ki67 defi ne proliferative
state and are prognostic markers in renal cell carcinoma. Clin
Cancer Res 2005;11:2510–7.
14. Montanari M, Boninsegna A, Faraglia B, Coco C, Giordano A,
Cittadini A, et al. Increased expression of geminin stimulates the
growth of mammary epithelial cells and is a frequent event in
human tumors. J Cell Physiol 2005;202:215–22.
15. Shetty A, Loddo M, Fanshawe T, Prevost AT, Sainsbury R,
Williams GH, et al. DNA replication licensing and cell cycle
kinetics of normal and neoplastic breast. Br J Cancer 2005;93:
16. Gonzalez MA, Tachibana KE, Chin SF, Callagy G, Madine MA,
Vowler SL, et al. Geminin predicts adverse clinical outcome in
breast cancer by refl ecting cell-cycle progression. J Pathol 2004;
17. Lauren P. The two histological main types of gastric carcinoma.
Diffuse and so-called intestinal-type carcinoma. Acta Pathol
Microbiol Scand 1965;64:31–49.
18. Japanese Gastric Cancer Association, editors. The Japanese clas-
sifi cation of gastric carcinoma. 14th ed. Tokyo: Kanehara; 2010. p.
19. Nishihara K, Shomori K, Tamura T, Fujioka S, Ogawa T, Ito H.
Immunohistochemical expression of geminin in colorectal cancer:
implication of prognostic signifi cance. Oncol Rep 2009;21:
20. Kaplan EL, Meier P. Nonparametric estimation for incomplete
observation. J Am Stat Assoc 1958;53:457–81.
21. Lin BT, Gruenwald S, Morla AO, Lee WH, Wang JY. Retinoblas-
toma cancer suppressor gene product is a substrate of the cell
cycle regulator cdc2 kinase. EMBO J 1991;10:857–64.
22. Kulartz M, Kreitz S, Hiller E, Damoc EC, Przybylski M, Knippers
R. Expression and phosphorylation of the replication regulator
protein geminin. Biochem Biophys Res Commun 2003;305:
23. Nakuci E, Xu M, Pujana MA, Valls J, Elshamy WM. Geminin is
bound to chromatin in G2/M phase to promote proper cytokine-
sis. Int J Biochem Cell Biol 2006;38:1207–20.
24. Duchrow M, Ziemann T, Windhövel U, Bruch HP, Broll R.
Colorectal carcinomas with high MIB-1 labelling indices but low
pKi67 mRNA levels correlate with better prognostic outcome.
25. Cho YE, Kim JY, Kim YW, Park JH, Lee S. Expression and
prognostic signifi cance of human growth and transformation-
dependent protein in gastric carcinoma and gastric adenoma.
Hum Pathol 2009;40:975–81.
26. Quaglia A, McStay M, Stoeber K, Loddo M, Caplin M, Fanshawe
T, et al. Novel markers of cell kinetics to evaluate progression
from cirrhosis to hepatocellular carcinoma. Liver Int 2006;26:
27. Tamura T, Shomori K, Haruki T, Nosaka K, Hamamoto Y, Shiomi
T, et all. Minichromosome maintenance-7 and geminin are reli-
able prognostic markers in patients with oral squamous cell car-
cinoma: immunohistochemical study. J Oral Pathol Med 2010;
28. Shrestha P, Saito T, Hama S, Arifi n MT, Kajiwara Y, Yamasaki F,
et al. Geminin: a good prognostic factor in high-grade astrocytic
brain tumors. Cancer 2007;109:949–56.
29. Loddo M, Kingsbury SR, Rashid M, Proctor I, Holt C, Young J, et
al. Cell-cycle-phase progression analysis identifi es unique pheno-
types of major prognostic and predictive signifi cance in breast
cancer. Br J Cancer 2009;100:959–70.
30. Tokuyasu N, Shomori K, Nishihara K, Kawaguchi H, Fujioka S,
Yamaga K, et al. Minichromosome maintenance 2 (MCM2)
immunoreactivity in stage III human gastric carcinoma: clinico-
pathological signifi cance. Gastric Cancer 2008;11:37–46.
31. Hama S, Matsuura S, Tauchi H, Yamasaki F, Kajiwara Y, Arita K,
et al. Gene transfer increases cell killing with abnormal nucleation
after ionising radiation in glioma cells. Br J Cancer 2003;89:
32. Terasima T, Tolmach LJ. Variations in several responses of HeLa
cells to x-irradiation during the division cycle. Biophys J 1963;3:
33. Yonemura Y, Ohoyama S, Kimura H, Kamata T, Yamaguchi A,
Miyazaki I. Assessment of tumor cell kinetics by monoclonal anti-
body Ki-67. Eur Surg Res 1990;22:365–70.
34. Chen L, Li X, Wang GL, Wang Y, Zhu YY, Zhu J. Clinicopathologi-
cal signifi cance of overexpression of TSPAN1, Ki67 and CD34 in
gastric carcinoma. Tumori 2008;94:531–8.
35. Oshima CT, Iriya K, Forones NM. Ki-67 as a prognostic marker
in colorectal cancer but not in gastric cancer. Neoplasma 2005;
36. Stoeber K, Tlsty TD, Happerfi eld L, Thomas GA, Romanov S,
Bobrow L, et al. DNA replication licensing and human cell pro-
liferation. J Cell Sci 2001;114:2027–41.
37. Gerdes J, Lemke H, Haisch H, Wacker HH, Schwab U, Stein H.
Cell cycle analysis of a cell proliferation-associated human nuclear
antigen defi ned by the monoclonal antibody Ki-67. J Immunol