ORIGINAL PAPER

The role of Cyclin G1 in cellular proliferation and apoptosisof human epithelial ovarian cancer

Lifei Jiang1,2 • Rong Liu3 • Yingying Wang2 • Chunmiao Li4 • Qinghua Xi1,2 •

Jianxin Zhong1,2 • Jian Liu1,2 • Shuyun Yang5 • Juan Wang1,2 • Menghui Huang1,2 •

Chunhui Tang1,2 • Zheng Fang2

Received: 11 March 2015 / Accepted: 11 May 2015 / Published online: 17 May 2015

� Springer Science+Business Media Dordrecht 2015

Abstract Cyclin G1 plays an essential role in the de-

velopment of human carcinoma. Here, we characterized the

clinical significance of Cyclin G1 and investigated its role

in cellular proliferation and apoptosis of epithelial ovarian

cancer (EOC). Western blot was used to evaluate the ex-

pression of Cyclin G1 in nine fresh EOC tissues and three

fresh normal ovarian tissues. Immunohistochemistry ana-

lysis was performed on formalin-fixed paraffin-embedded

section of 119 cases of EOCs. Using cell counting kit

(CCK)-8 and colony formation assays, we analyzed the

effect of Cyclin G1 in cellular proliferation of EOC. Be-

sides, the immunofluorescence and flow cytometry analysis

was performed to study the role of Cyclin G1 in cellular

apoptosis of EOC. We found Cyclin G1 was up-regulated

in EOC tissues compared with the normal ovary tissues.

Cyclin G1 expression in EOC was closely correlated with

differentiation grade (P = 0.009) and malignant tumor

cells in ascites (P = 0.009). The Kaplan–Meier curve

showed that higher expression of Cyclin G1 was associated

with significantly shorter survival in EOC patients. Multi-

variate analysis suggested Cyclin G1 expression was an

independent prognostic factor for overall survival. CCK-8

and colony formation assays revealed that depletion of

Cyclin G1 inhibited the proliferation and clone formation.

Combined immunofluorescence and flow cytometry ana-

lysis showed that silencing of Cyclin G1 with shRNA could

promote apoptosis of ovarian cancer cells. Additionally,

the result of immunoprecipitation test showed Cyclin G1

interacted with CDK2 in EOC cells. In summary, our

findings suggest that Cyclin G1 may be involved in the

prognosis of EOC patients and be a useful therapeutic

target for EOC.

Keywords Epithelial ovarian cancer (EOC) � Cyclin G1 �Proliferation � Apoptosis � Prognosis

Introduction

The most common ovarian malignant neoplasms are ep-

ithelial ovarian cancers (EOC) which may develop from the

epithelium of the fimbrial portion of the fallopian tube and/or

the ovarian surface epithelium (Koshiyama et al. 2014).

Every year 220,000 women develop epithelial ovarian can-

cer worldwide. Most EOC patients are clinically character-

ized by an asymptomatic course in early stage disease,

followed by diagnosis at an advanced stage (Aust and Pils

2014). During the last decades, mortality statistics is barely

improved (Vargas-Hernandez et al. 2014). With advance in

Lifei Jiang and Rong Liu have contributed equally to this work.

& Chunhui Tang

ntfytch@163.com

& Zheng Fang

fznt@163.com

1 Department of Obstetrics and Gynecology, Affiliated

Hospital of Nantong University, Nantong, China

2 Jiangsu Province Key Laboratory for Information and

Molecular Drug Target, Nantong University,

Nantong 226001, Jiangsu Province, China

3 Department of Gynecologic Oncology, Nantong University

Cancer Hospital, Nantong University,

Nantong 226001, Jiangsu Province, China

4 Clinical Laboratory, Qilu Children’s Hospital of Shandong

University, Jinan 250022, Shandong, China

5 Department of Pathology, Nantong University Cancer

Hospital, Nantong University,

Nantong 226001, Jiangsu Province, China

123

J Mol Hist (2015) 46:291–302

DOI 10.1007/s10735-015-9622-7

research of the molecular biology of EOC, some studies

provide that clinical and molecular biological factors alone

are insufficiently useful as prognostic parameters (Aust and

Pils 2014). Thus, understanding themolecular mechanism of

EOC and focusing on newly characterized specific biologic

targets become more and more important.

Cyclin G1 was firstly found in 1996, located to chro-

mosome 5q-32-q34 with six exons and the cDNA length of

3.17 kb (Ye et al. 2012). It is a subtype of Cyclin G of the

cyclin family that has a positive and negative regulator of

cell growth for different kinds of cells. Although the pre-

cise function of Cyclin G1 remains unclear, accumulating

evidence has showed that Cyclin G1 abnormally expressed

in many types of malignant cancers such as cervical car-

cinoma (Liang et al. 2006), hepatocellular carcinoma (Wen

et al. 2013), lung carcinoma (Zhao et al. 2015) and breast

cancer (Reimer et al. 1999). Unlike other cyclins, Cyclin

G1 has neither a destruction box nor PEST sequences that

are responsible for cyclin degradation (Tamura et al. 1993).

Cyclin G1 levels remain relatively constant throughout the

cell cycle both in vitro and in vivo. Thus, Cyclin G1 may

inhibit proliferation and induce apoptosis in cancer cell

lines and tumor xenografts, while in others it can increase

cell proliferation and even tumorigenesis.

Previous studies mainly focused on the role of Cyclin G1

via several complicated mechanisms. The major findings

are: (1) Some studies found evidence that as a transcriptional

target of p53, Cyclin G1 expression upon DNA damage was

regulated by p53, initiating a feedback regulation of p53

through a mechanism that involves MDM2. It was demon-

strated that the interaction between Cyclin G1 and enzy-

matically active PP2A led to the dephosphorylation of

MDM2 followed by p53 degradation (Okamoto et al. 2002;

Yuan et al. 2014). In addition, a study showed that Cyclin G1

could interact with hepatocyte-specific miR-122a, which

might abrogate p53-mediated inhibition of HBV replication

and therefore contribute to carcinogenesis (Wang et al.

2012). (2) Cyclin G1was found in complex with a number of

proteins involved in cell cycle checkpoint regulation. For

example, Cyclin G1 could interact with cyclin-dependent

kinase 5 (CDK5) and GAK (cyclin G-associated kinase),

although the physiological significance of these interactions

remained unclear (Kanaoka et al. 1997). (3) Cyclin G1 was

suggested to act as an oncogenic protein, because of its

overexpression in human tumor cells (Kang et al. 2013). (4)

Cyclin G was also involved in G2/M arrest in response to

DNA damage (Piscopo and Hinds 2008) or in the facilitation

of TNF-induced apoptosis (Okamoto and Prives 1999), im-

plicating tumor suppressive function. Despite these findings,

understanding of the biochemical function of Cyclin G1 in

complexwith associated partners and themolecular function

of association partners in cellular physiology has been

limiting.

Although Cyclin G1 has been extensively examined in a

wide range of cancers, the issue of whether Cyclin G1 is

regulated and its role in proliferation and apoptosis of

human epithelial ovarian cancer cells remains unclear.

Russell et al. suggested Cyclin G1 amplification was as-

sociated with significantly shorter post-surgical survival in

patients with ovarian cancer (Russell et al. 2012). In this

study, we aimed to investigate Cyclin G1 expression in

EOC and clarify its possible molecular mechanisms. Ac-

cording to our results, it was reasonable to consider that

Cyclin G1 could be a novel therapeutic target for EOC.

Materials and methods

Ovarian cancer samples collection

We collected 119 formalin-fixed, paraffin-embedded EOC

tissue samples from the archival files of the Department

of Pathology, affiliated Hospital of Nantong University.

All clinical samples were provided using protocols ap-

proved by the Ethics Committee of Affiliated Hospital of

Nantong University. The clinical and pathologic data re-

lating to 119 EOC patients were presented in Table 1,

including age, Figo stage, histologic subtype, lymph node

status and malignant tumor cells in peritoneal fluid, and

so on. All specimens were available for review between

2004 and 2009. Median age of the corresponding EOC

patients at cancer diagnosis was 55 years (range 24–80).

These samples were used for IHC to analyze the rela-

tionship between Cyclin G1 expression and clinico-

pathological parameters. Nine fresh EOC specimens and

three fresh normal ovarian epithelial tissues were imme-

diately snap-frozen in liquid nitrogen after surgical re-

moval and stored at -80 �C until using for western blot.

And the normal fresh specimens were obtained from pa-

tients with normal ovaries who underwent the hysterec-

tomy to treat other diseases such as cervical cancer,

Adenomysis, and so on.

Western blot analysis

The frozen normal and tumor tissues were rapidly lysed with

a homogenization buffer containing protease inhibitors (1 %

NP-40, 50 mmol/l Tris, pH 7.5, 5 mmol/l EDTA, 1 % SDS,

1 % sodium deoxycholate, 1 % Triton X-100, 1 mmol/l

PMSF, 10 mg/ml aprotinin, and 1 mg/ml leupeptin) on ice.

Ovarian cancer cells were disrupted on ice with RIPA buffer

(150 mMNaCl, 1 % Nonidet P-40, 0.5 % deoxycholic acid,

0.1 % sodium dodecyl sulfate, 50 mM Tris–HCl, pH 8.0).

After centrifugation at 12,000 rpm for 15 min at 4 �C, thesupernatant fraction was harvested. 50 lg of total protein

was resolved by using 10 %SDS-PAGE and transferred onto

292 J Mol Hist (2015) 46:291–302

123

polyvinylidene difluoride (PVDF) membranes (Millipore,

Bedford, MA). Membranes were first blocked with 5 % fat-

free driedmilk for 2 h and incubatedwith primary antibodies

overnight at 4 �C. After washed with TBST (20 mM Tris,

150 mM NaCl, 0.05 % Tween-20) three times for 5 min

each, the membranes were blotted with the peroxidase-

conjugated secondary antibody (1:10,000 dilution) for 2 h.

Immunoreactive bands were detected by enhanced chemi-

luminescence (ECL) detection kit (Pierce, Rockford, IL,

USA).

Antibodies

The monoclonal antibodies (mAb) were used in this study

as follows: rabbit anti-human Cyclin G1 antibody (1:1000

dilution), rabbit anti-human Ki-67 antibody (1:500 dilu-

tion), rabbit anti-human CDK2 antibody (1:100, dilution),

mouse anti-human PCNA antibody (1:500 dilution), mouse

anti-human Cleaved Caspase-9 antibody (1:100 dilution),

rabbit anti-human GAPDH antibody (1:1000 dilution) and

the peroxidase-conjugated secondary antibody (1:10,000

dilution). All the antibodies were purchased from Santa

Cruz Biotechnology, USA.

Immunohistochemistry and tissue microarray

analysis

The 119 EOC tissues were sliced in 4 lm thick, dewaxed

in xylene for 15 min twice and rehydrated through graded

ethanols. The sections were performed by heating to

Table 1 Cyclin G1 and Ki-67 expression and clinicopathologic parameters of 119 EOC patients

Patient and tumor characteristics Total Cyclin G1 P Ki-67 P

High n (%) Low n (%) High n (%) Low n (%)

Age

B 50 35 12 (34.3) 23 (65.7) 0.182 16 (45.7) 19 (54.3) 0.208

[ 50 84 40 (47.6) 44 (52.4) 49 (58.3) 35 (41.7)

Histologic subtype

Serous 61 30 (49.2) 31 (50.8) 0.201 37 (60.7) 24 (39.3) 0.030*

Mucinous 5 0 (0) 5 (100) 0 (0) 5 (100)

Endometrioid 10 3 (30) 7 (70) 8 (80) 2 (20)

Clear cell 10 4 (40) 6 (60) 5 (50) 5 (50)

Undifferentiated 33 14 19 15 (45.5) 18 (54.5)

Differentiation grade

G1 7 0 (0) 7 (100) 0.009* 3 (42.9) 4 (57.1) 0.045*

G2 36 12 (33.3) 24 (66.7) 14 (38.9) 22 (61.1)

G3 76 40 (52.6) 36 (47.4) 48 (63.2) 28 (36.8)

FIGO stage

Stage I 52 17 (32.7) 35 (67.3) 0.068 23 (44.2) 29 (55.7) 0.040*

Stage II 13 8 (61.5) 5 (38.5) 11 (84.6) 2 (18.4)

Stage III 49 23 (46.9) 26 (53.1) 27 (55.1) 22 (44.9)

Stage IV 5 4 (80) 1 (20) 4 (80) 1 (20)

Malignant tumor cells

Present 28 19 (67.9) 9 (32.1) 0.009* 19 (67.9) 9 (32.1) 0.041*

Absent 91 33 (36.3) 58 (63.7) 44 (48.4) 47 (51.6)

Lymph node status

Positive 23 12 (52.2) 11 (47.8) 0.362 12 (52.2) 11 (47.8) 0.256

Negative 96 40 (41.7) 56 (58.3) 51 (53.1) 45 (46.9)

Ascites

Present 47 28 (59.6) 19 (40.4) 0.005 27 (57.4) 20 (42.6) 0.21

Absent 72 24 (33.3) 48 (66.7) 36 (50) 36 (50)

Total 119

Statistical analyses were performed by the Pearson’s v2 test

* P\ 0.05 was considered significant

J Mol Hist (2015) 46:291–302 293

123

121 �C for 3 min in 10 mmol/l citrate buffer (pH 6.0). The

endogenous peroxidase activity was blocked by incubation

with 3 % hydrogen peroxide in methanol for 20 min. The

following panel of antibodies was used: diluted anti-Cyclin

G1 antibody (dilution 1:50) or anti-Ki67 antibody (dilution

1:500) at 4 �C for 12 h (Turan et al. 2014). All slides were

processed using the peroxidase antiperoxidase method

(Dako, Hamburg, Germany). After washing with PBS, the

peroxidase reaction was visualized by incubating the sec-

tions with the liquid mixture DAB (0.1 % phosphate buffer

solution, 0.02 % diaminobenzidine tetrahydrochloride, and

0.03 % H2O2). All slides were processed using the per-

oxidase–anti-peroxidase method (Dako, Hamburg, Ger-

many). After rinsing in water, the sections were

counterstained with hematoxylin, dehydrated, and cover

slipped. Two professional observers who were blind to the

background of patients evaluated immunoreactivity by in-

tensity and percentage of epithelial stained by a Leica

fluorescence microscope (Germany). Staining intensity in

every tumor section was scored from 0 to 3 [0, nega-

tive(-); 1, weak(?); 2, medium(??); 3, strong(???)],

and the proportion of epithelial cell staining positively was

scored from 0 to 4 (0, 0–5 %; 1, 6–25 %; 2, 26–50 %; 3,

51–75 %; 4, 76–100 %) (Ramsay et al. 2007). The ex-

pression of Cyclin G1 and Ki-67 was interpreted and

graded according to the sum of the intensity and extent

score. Based on the results from immunohistochemistry, all

patients were divided into two groups: low expression

group (score B 3) and high expression group (score[ 3).

Cell culture

We purchased three human ovarian cancer cell lines

(OVCA3, HO8910 and SKOV3) from Shanghai Institute of

Cell Biology. Cells were cultured in RPMI 1640 (Gibco

BRL, Grand Island, NY) supplemented with 10 % fetal

bovine serum (FBS). They were maintained at 37 �C in a

humidified atmosphere with 5 % CO2.

Transfection

HO8910 cells were transfected with short hairpin RNA

(shRNAs) targeting the nucleotide sequences (50-AAATGTTCAGAAGTTGAAA-30,50-ATTGTCTATCATTGCATTA-30,50-TCAACTGAAGGCATGTCAT-30, and 50-GGACAGATTCCTGTCTAAA-30 respectively) and control-sh RNA, which

were purchased from Genechem (Shanghai, China). The

transfectionof shCyclinG1wasperformedwithLipofectamine

2000 (Invitrogen) according to the manufacture’s instructions.

After 48 h, transfected cells were harvested and analyzed in the

indicated assays. The efficacy of Cyclin G1 inhibition was

tested by western blot analysis.

Cell proliferation assay

After transfected with sh Cyclin G1 or control-sh RNA

24 h, cells were seeded into 96-well plates at a concen-

tration of 2 9 104 cells per well in volumes of 90 ll andgrew overnight (Chen et al. 2015). 10 ll CCK8 reagents

(Dojindo, Kumamoto, Japan) were then added into every

well at different time points and then incubated for 2 h in

dark environment. The blanks and samples were measured

at a test wavelength of 450 nm and a reference wavelength

of 630 nm using a microplate reader (Bio-Rad). The ex-

periments were repeated at least thrice.

Flow cytometry

HO8910 cells transfected with sh Cyclin G1 and control-sh

RNA were cultured for 48 h and harvested. Then, add

60 ll of Muse TM Annexin V and Dead Cell Reagent (Part

No. 4700-1485, 100 tests/bottle) and 60 ll of cells in

suspension to each tube. After incubated for 20 min in the

dark, the apoptosis assay was performed by Muse TM Cell

Analyser (EMD Millipore corporation). The experiments

were repeated at least thrice.

Colony formation assays

After transfection, cells were seeded in culture dishes (500

cells/plate) and cultured for two weeks. After washed with

PBS, cells were stained with 1 % crystal violet for 30 s after

fixation with 10 % formaldehyde for 2 h. Experiments were

performed in at least three independent replicates.

Immunoprecipitation

Cells were washed three times with PBS and lysed in

immunoprecipitation buffer (50 mM HEPES pH 7.6,

150 mM NaCl, 5 mM EDTA, 0.1 % NP-40). After cen-

trifugation (10 min at 15,000 g), to remove cell debris,

Samples were precleared with nonspecific IgG and albumin

controls. The lysates were incubated with antibodies

(1:200) against anti-Cyclin G1, CDK2 or nonspecific IgG

and rotated overnight. The immunocomplexes were added

with 40 ll of a protein A-Sepharose (Sigma) with rotation

for 1 h at 4 �C. The last supernatant liquid was removed,

and the protein was collected. A 50-ll quantity of

294 J Mol Hist (2015) 46:291–302

123

Fig. 1 Expression changes of Cyclin G1 in normal ovary and EOC

tissues. a Expression of Cyclin G1 in normal ovarian tissues (N1, N2, N3)

and EOC tissues (C1-C9) from grade 1 (G1) to grade 3 (G3). b The bar

chart demonstrated the ratio of Cyclin G1 protein to GD for the above by

densitometry. The relative expression levels were showed by density

photometry. The data were mean ± SEM (*P\ 0.05 compared Cyclin

G1 expression levels between normal ovarian samples and EOC tissues).

The sameexperimentwasrepeatedat least three times.c–hRepresentativephotographs of Cyclin G1 and Ki-67 immunohistochemistry in paraffin-

embedded EOC tissue samples. c and f low concentrations of Cyclin G1

and Ki-67 in EOC tissue with grade 1. d and gmedium concentrations of

Cyclin G1 and Ki-67 in EOC tissue with grade 2. e and h high

concentrations of Cyclin G1 and Ki-67 expression in EOC tissue with

grade 3. GD, GAPDH. c–h were imaged in Magnification 9 400

J Mol Hist (2015) 46:291–302 295

123

2 9 loading buffer was added, and the samples were

boiled at 95–100 �C for 15 min to denature the protein.

Finally, the samples were analyzed by SDS–polyacry-

lamide gel electrophoresis using enhanced chemilumines-

cence detection (Amersham International).

Immunofluorescence

The cells transiently transfected with sh Cyclin G1 and

control-sh RNA were washed with phosphate-buffered sal-

ine, fixed with 4 % paraformaldehyde (4 h), and permeabi-

lizedwith 0.1 %TritonX-100 (10 min). Then, the cells were

incubated with both primary antibodies for Cyclin G1 (anti-

rabbit, 1:200; Santa Cruz Biotechnology) and Cleaved

Caspas-9 (a marker of apoptosis, anti-mouse, 1:100; Santa

Cruz Biotechnology) overnight at 4 �C. After washing threetimes with phosphate-buffered saline, samples were incu-

bated with Hoechst 33342 dye (1 lg/ml, 2 h) for DNA

staining at room temperature. The fluorescence was exam-

ined with a Leica fluorescence microscope (Solms, Ger-

many). All assays were performed three times in duplicate.

Statistical analysis

Statistical Analysis in this study was performed with the

SPSS 17.0 software version (Chicago, USA). Chi square

(v2) and Fisher’s exact test were used to measure the re-

lationship between the Cyclin G1 and Ki-67 expression and

the clinicopathological features. Multivariate analysis was

performed on the Cox proportional hazards regression

model. Student’s t test was used to determine significant

differences between 2 means. A P value\0.05 was con-

sidered significant. Values are represented as mean ±

SEM. Significant differences are marked in the graphs.

Results

Expression of Cyclin G1 was up-regulated in EOC

tissues

To investigate the role of Cyclin G1 in EOC development,

we used western blot to detect Cyclin G1 expression in nine

frozen EOC and three normal ovarian tissues. As shown in

Fig. 1a, b, Cyclin G1 expression was dramatically increased

in nine tumor samples compared with the three normal

ovarian samples (P\ 0.05). To explore the clinicopatho-

logical significance of Cyclin G1 in EOC progression, we

Fig. 2 The relation between Ki-67 and Cyclin G1 expression in

EOC. Scatter plot of Ki-67 versus Cyclin G1 with regression line

showing a correlation using Spearman correlation coefficient

Table 2 Survival status and clinicopathologic parameters of 119

EOC patients

Patient and tumor

characteristics

Total Survival

status

P value v2

Alive Dead

Age

B 50 35 19 16 0.08 3.064

[ 50 84 31 53

Histologic subtype

Serous 61 21 40 0.026* 11.02

Mucinous 5 4 1

Endometrioid 10 5 5

Clear cell 10 8 2

Undifferentiated 33 12 21

Differentiation grade

G1 7 6 1 0.007* 9.793

G2 36 19 17

G3 76 25 51

FIGO stage

Stage I 52 34 18 \0.001* 21.258

Stage II 13 2 11

Stage III 49 13 36

Stage IV 5 1 4

Malignant tumor cells

Absent 91 42 49 0.099 2.717

Present 28 8 20

Lymph node status

Negative 96 44 52 0.085 2.970

Positive 23 6 17

Ascites

Absent 72 33 39 0.296 1.090

Present 47 17 30

Cyclin G1 expression

Low 67 36 31 0.003* 9.042

High 52 14 38

Ki-67 expression

Low 54 28 26 0.048* 3.925

High 65 22 43

Total 119

Statistical analyses were performed by the Pearson’s v2 test

* P\ 0.05 was considered significant

296 J Mol Hist (2015) 46:291–302

123

investigated the reactivity for Cyclin G1 and Ki-67 by im-

munohistochemical staining. Representative examples of

reactivity for Cyclin G1 (Fig. 1c–e) and the proliferation

index Ki-67 (Fig. 1f–h) were shown. Interestingly, elevated

expression of Cyclin G1 was observed in high grade tissues

compared with that in low grade ones, which was further

confirmed by western blot assay (Fig. 1a). These results

indicated the potential role of Cyclin G1 in EOC

development.

Cyclin G1 expression was associated

with clinicopathological parameters and prognosis

of EOC patients

We evaluated the association of Cyclin G1 and Ki-67

expression with clinicopathological variables in Table 1.

The statistical analysis showed that expression of Cyclin

G1 was significantly associated with differentiation grade

and malignant tumor cells in ascites (P\ 0.05), but there

Fig. 3 Kaplan–Meier survival curves EOC patients showed a highly

significant separation between curves. a–e Survival rates of EOC

patients in low versus high Cyclin G1 expression, low versus high Ki-

67 expression, serous versus non serous ovarian cancers, Grade 1–2

and Grade 3 and Figo I–II and III–IV separately are exhibited

(P\ 0.05)

J Mol Hist (2015) 46:291–302 297

123

was no relation between Cyclin G1 and other clinico-

pathological variables such as lymph node status and

histologic subtype. According with Cyclin G1 and Ki-67

expression in the nuclei, we evaluated the proportion of

Cyclin G1 and Ki-67 positive tumor cells. After analysis,

we found that there was a positive correlation between

Cyclin G1 expression and Ki-67-based proliferative ac-

tivity (P\ 0.001; Fig. 2). When all variables were com-

pared separately with survival status, we found Cyclin G1

and Ki-67 expression, histologic subtype, differentiation

grade, and FIGO stage (P\ 0.05) could significantly in-

fluenced survival (Table 2 and Fig. 3). Only 14 (26.9 %)

of 52 patients in the Cyclin G1 high expression group

were alive versus 36 (53.7 %) of 67 in the Cyclin G1 low

expression group. More importantly, the patients in the

Cyclin G1-high group exhibited shorter survival

(P\ 0.05) than those in the Cyclin G1-low group

(Fig. 3a). Multivariate analysis suggested Cyclin G1 ex-

pression (P = 0.009) as well as FIGO stage (P = 0.001)

were independent prognostic factors for EOC (Table 3).

Thus, Cyclin G1 expression could serve as a valuable

predicting factor for poor survival of EOC patients.

The depletion of Cyclin G1 inhibited

the proliferation of ovarian cancer cells and Cyclin

G1 interacts with CDK2

In cell level, we first evaluated the different expression

level of Cyclin G1 in three human ovarian cancer cells:

HO8910, SKOV3, and OVCA3. After normalizing to

GAPDH, the endogenous Cyclin G1 was higher in HO8910

and SKOV3 than OVCA3 cells, especially HO8910 cells

(Fig. 4a). To further investigate the potential effects of

Cyclin G1 in EOC cells, a series of short hairpin RNAs (sh

RNAs) were used to inhibit the expression of endogenous

Cyclin G1 in the HO8910 cells. After transfection for 48 h,

western blot analysis showed Cyclin G1 protein level was

markedly decreased in HO8910 cells transfected with sh

Cyclin G1#2 compared with others sh RNAs

(Fig. 4b).Therefore, sh Cyclin G1#2 was used for all the

subsequent experiments. In HO8910 cells treated with sh

Cyclin G1#2, PCNA protein level was significantly de-

creased than the control group (Fig. 4c).

Some studies found Cyclin G1 promoted G2/M cell

cycle arrest in response to DNA damage (Russell et al.

2012; Kimura et al. 2001). In our study, Cyclin G1 ex-

pression was significantly associated with differentiation

grade and malignant tumor cells in ascites. Therefore,

Cyclin G1 might affect EOC cells proliferation. Using

cell counting kit (CCK)-8 assays, we further investigated

if Cyclin G1 activity affected cell proliferation. After

transfected with sh Cyclin G1#2, the cells proliferation

rate of SKOV3 and HO8910, especially HO8910, exhib-

ited a significant decrease compared with the other groups

(Fig. 4d). Besides, the colony formation assay showed

inhibition of the expression of Cyclin G1 suppressed

clone formation of HO8910 cells (Fig. 4e). Moreover,

genome-wide two-hybrid screens suggested that Cyclin G

interacted with different CDKs, including CDK2 and

CDK4 (Stanyon et al. 2004). To further study the

mechanism of Cyclin G1 in EOC, we found Cyclin G1

interacted with CDK2 in ovarian cancer cells by im-

munoprecipitation (Fig. 4f). Taken together, we consid-

ered that Cyclin G1 might promote ovarian tumor cell

proliferation by interacting with CDK2.

Table 3 Contribution of various potential prognostic factors to sur-

vival by Cox regression analysis in 119 EOC patients

Hazard ratio 95 % CI P

Histologic subtype 0.996 0.862–1.150 0.951

Differentiation grade 1.575 0.943–2.630 0.083

FIGO stage 1.638 1.237–2.169 0.001*

Cyclin G1 expression 1.944 1.183–3.194 0.009*

Ki-67 expression 0.940 0.511–1.729 0.842

Histologic subtype (non-serous vs serous); Differentiation grade

(Grade 1–2 vs. Grade 3); FIGO stage (FIGO I–II vs. FIGO III–IV);

Cyclin G1 expression (High expression vs Low expression); Ki-67

expression (High expression vs Low expression)

* P\ 0.05 was considered significant

cFig. 4 The depletion of Cyclin G1 inhibited the proliferation of

ovarian cancer cells and Cyclin G1 interacts with CDK2. a Western

blot analysis showed that Cyclin G1 protein level was higher in

HO8910 and SKOV3 than OVCA3 cells, especially HO8910 cells.

The bar charts were quantifications of the western blots. Experiments

were repeated at least three times. b HO8910 cells were transfected

with control-sh RNA and sh Cyclin G1#1, 2, 3, 4. The protein levels

of Cyclin G1 were detected by western blot 48 h after transfection.

The bar charts were quantifications of the western blots. Experiments

were repeated at least three times. cWestern blot analysis showed that

sh Cyclin G1#2 markedly decreased Cyclin G1 and PCNA levels. The

bar charts were quantifications of the western blots. Experiments

were repeated at least three times. d The proliferation of HO8910 and

SKOV3 cells was measured using CCK-8 assay. Data showed sh

Cyclin G1#2 knockdown represses HO8910 (left) and SKOV3 (right)

cells proliferation, especially HO8910. e Cyclin G1 knockdown in

HO8910 cells suppressed clone formation. The bar chart demon-

strated the number of colonies of HO8910 cells treated with sh Cyclin

G1#2 markedly decreased. f The immunoprecipitation experiments

showed Cyclin G1 was able to interact with CDK2. HO8910 cells

were harvested for IP. The lysates were incubated with antibodies

(1:200) against anti-Cyclin G1, CDK2 or nonspecific IgG and rotated

overnight. The immune complexes were analyzed by western

blotting. All experiments were repeated at least three times using

independent samples and data were shown as mean ± SEM

(*P\ 0.05). Asterisks (*) indicated a significant difference, com-

pared with the control group. GD, GAPDH

298 J Mol Hist (2015) 46:291–302

123

J Mol Hist (2015) 46:291–302 299

123

Loss of Cyclin G1 expression promoted EOC cells

apoptosis is involved in many of the functions

regulated by p53 such as apoptosis

Recent reports indicated that Cyclin G1 was involved in

diverse cellular processes, including regulation of the

cell cycle, apoptosis, and so on (Kimura and Nojima

2002; Kimura et al. 2001). Thus, we presumed that the

changes in Cyclin G1 expression might involve ovarian

cancer cells apoptosis regulation. To further address this

hypothesis, we used sh RNA to knockdown Cyclin G1

expression in HO8910 cells and detected the expression

of the apoptosis -related protein Cleaved Caspase-9. As

shown in Fig. 5a, western blot analysis showed a sig-

nificant increase in the expression of Cleaved Caspase-9

after transfected with sh Cyclin G1#2, which was op-

posite to the protein level of Cyclin G1. Besides, we

investigated whether Cyclin G1 induced apoptosis in

EOC cells by the flow cytometry assay. The results re-

vealed that there were more apoptotic cells in the sh

Cyclin G1#2 group compared with the control group

(Fig. 5b). Considering the pattern of Cyclin G1 local-

ization in the nucleus and the potential functions of

Cyclin G1, we explored the possibility of the association

between Cyclin G1 and Cleaved Caspase-9. Moreover,

Combined immunofluorescence and flow cytometry

analysis showed that silencing of Cyclin G1 with shRNA

could promote apoptosis of ovarian cancer cells

(Fig. 5c–k). The immunofluorescent staining and flow

cytometry analysis showed that silencing of Cyclin G1

with shRNA could promote apoptosis of ovarian cancer

cells.

Discussion

Tumor development results from a multistep process.

Despite increasing knowledge in the etiology of EOC,

there has been only a slight change in the mortality of

EOC patients in the past 30 years. Recent studies show

disrupted cell cycle regulations and targeting these cell

cycle proteins may bring beneficial effects for EOC pa-

tients (Aust and Pils 2014). Cyclin G1 was described as

both a positive and negative regulator of human cancer

cells growth (Li et al. 2009). Although exact mechanism

of Cyclin G1 remains unclear, Cyclin G1 may be asso-

ciated with oncogenic potential and involve growth con-

trol, DNA repair, and apoptosis (Kimura et al. 2001). Our

study showed Cyclin G1 expression level in EOC was

closely correlated with differentiation grade (P = 0.009)

and malignant tumor cells (P = 0.009). Multivariate

analysis showed Cyclin G1 expression was an indepen-

dent prognostic factor for EOC patients. Thus, Cyclin G1

expression could serve as a valuable predicting factor for

EOC patients.

Cyclin G1 belongs to cyclin family, which contain a

highly conserved motif called the cyclin box. The abnor-

mal expression of many cyclin proteins has been found in

many human cancer tissues, such as Cyclin A, B, D1, D3,

and E (Ye et al. 2012). Cyclins have been classified into

different groups on the basis of their structural similarity

and functional period in the cell division cycle. Currently

the cyclin family comprises 17 subtypes denoted A–M, O,

T and Y in mammalian cells. The Cyclin G subfamily

includes Cyclin G1 and Cyclin G2. Though the structures

of Cyclin G1 and G2 proteins are similar, their functions

are separated. In contrast to Cyclin G2, Cyclin G1 is the

only known cyclin that is activated transcriptionally by the

p53 tumor suppressor gene and apparently constitutive

throughout the cell cycle.

Most cyclins have been shown to associate with and

regulate the cyclin dependent kinases (cdks) to promote

the progression of the cell cycle (Peyressatre et al. 2015).

Previous reports suggested that control of cell growth and

cell cycle by Cyclin G might be achieved via interaction

with the cyclin-dependent kinases CDK2. Piscopo and

Hinds found co-expression of dominant-negative CDK2

with cyclin G1 in U2OS cells resulted in a substantial

increase in the half-life of cyclin G1 (Piscopo and Hinds

2008). Faradji et al. built a Cyclin G interactor network

based on some data, which indicated that top interactors

could include the cyclin-dependent kinases CDK4 and

CDK2 (Faradji et al. 2011). In our study, cell counting kit

(CCK)-8 and colony formation assays showed Cyclin G1

played a critical role in cells proliferation of EOC. The

results of immunoprecipitation showed Cyclin G1 was

able to bind CDK2 in EOC cells, Hence, Cyclin G1 might

promote ovarian tumor cell proliferation by interacting

with CDK2.

In a recently published paper, Cyclin G1 and MEF2D, a

transcriptional factor established as an oncogenic gene for

liver cancers, have been shown to mediate the anti-tumor

activity of oleanolic acid, reinforcing the notion that Cyclin

G1 is an important gene involved with cancer biology

(Zhao et al. 2015). Interestingly, MEF2D can promote or

depress the expression of target genes by directly binding

the recognition site located within their promoter regions.

Furthermore, Ma et al. showed that MEF2D could regulate

the progression of cell cycle progression by modulating the

expression of G2/M transition gene in liver cancers (Ma

et al. 2014), indicating that MEF2D was closely associated

with cell cycle progression. In combination with the co-

involvement of Cyclin G1 and MEF2D in the antitumor

function of oleanolic acids, we hypothesize that MEF2D

may be associated with the expression of Cyclin G1. Thus,

further study can be done to establish the link between

300 J Mol Hist (2015) 46:291–302

123

MEF2D and Cyclin G1, in order to elucidate the

mechanism of Cyclin G1 overexpression in EOC.

In conclusion, these findings provide multiple lines of

evidence to Cyclin G1 implicate as a key determinant of EOC

development. Cyclin G1 expression is associated with

clinicopathological parameters and prognosis of EOC pa-

tients. The down-regulation of Cyclin G1 by sh RNA could

inhibit ovarian cancer cells proliferation and promote cell

apoptosis. Thus, Cyclin G1 might be a useful biomarker in

diagnosis and provide a new treatment strategy for EOC

patients.

Acknowledgments This work was supported by the National Nat-

ural Science Foundation of China (No. 81302285), and A Project

Funded by the Priority Academic Program Development of Jiangsu

Higher Education Institutions (PAPD).

Conflict of interest All authors declare that they have no conflict of

interests.

References

Aust S, Pils D (2014) Epithelial ovarian cancer—more data, more

questions? Wien Med Wochenschr 164(21–22):479–486

Chen C, Deng Y, Hua M, Xi Q, Liu R, Yang S, Liu J, Zhong J, Tang

M, Lu S, Zhang Z, Min X, Tang C, Wang Y (2015) Expression

and clinical role of TCTP in epithelial ovarian cancer. J Mol

Histol 46(2):145–156

Faradji F, Bloyer S, Dardalhon-Cumenal D, Randsholt NB, Peronnet

F (2011) Drosophila melanogaster Cyclin G coordinates cell

growth and cell proliferation. Cell Cycle 10(5):805–818

Kanaoka Y, Kimura SH, Okazaki I, Ikeda M, Nojima H (1997) GAK:

a cyclin G associated kinase contains a tensin/auxilin-like

domain. FEBS Lett 402(1):73–80

Fig. 5 Loss of Cyclin G1 expression promoted EOC cells apoptosis.

a The expression of Cleaved Caspase-9 in HO8910 transfected with

sh Cyclin G1#2 had a significant increase. b HO8910 cells were

treated with control-sh RNA and sh Cyclin G1#2, respectively, and

then assayed by flow cytometry. c–j Double-immunofluorescence

analysis of Cyclin G1 (red) and the apoptosis-related protein Cleaved

Caspase-9 (green) in ovarian cancer cells. c–f HO8910 cells

transfected with control-sh RNA; G-J, HO8910 cells transfected with

sh Cyclin G1#2. Cells were immunostained with anti- Cyclin G1 and

Cleaved Caspase-9 antibodies. Cell nucleus was stained with Hochest.

k Quantitative analysis of Cyclin G1 and Cleaved Caspase-9-positive

cells in HO8910 cells transfected with control-sh RNA or sh Cyclin

G1#2. All results were representative of at least three repeated

experiments and data were shown as mean ± SEM (*P\ 0.05).

Asterisks (*) indicated a significant difference, compared with the

control group. GD, GAPDH

J Mol Hist (2015) 46:291–302 301

123

Kang D, Cho HS, Toyokawa G, Kogure M, Yamane Y, Iwai Y,

Hayami S, Tsunoda T, Field HI, Matsuda K, Neal DE, Ponder

BA, Maehara Y, Nakamura Y, Hamamoto R (2013) The histone

methyltransferase Wolf-Hirschhorn syndrome candidate 1-like 1

(WHSC1L1) is involved in human carcinogenesis. Genes

Chromosomes Cancer 52(2):126–139

Kimura SH, Nojima H (2002) Cyclin G1 associates with MDM2 and

regulates accumulation and degradation of p53 protein. Genes

Cells 7(8):869–880

Kimura SH, Ikawa M, Ito A, Okabe M, Nojima H (2001) Cyclin G1 is

involved in G2/M arrest in response to DNA damage and in

growth control after damage recovery. Oncogene

20(25):3290–3300

Koshiyama M, Matsumura N, Konishi I (2014) Recent concepts of

ovarian carcinogenesis: type I and type II. Biomed Res Int

934261

Li H, Okamoto K, Peart MJ, Prives C (2009) Lysine-independent

turnover of cyclin G1 can be stabilized by B’alpha subunits of

protein phosphatase 2A. Mol Cell Biol 29(3):919–928

Liang J, Bian ML, Chen QY, Liu X, Ou H, Li M, Liu J (2006)

Relationship between cyclin G1 and human papilloma virus

infection in cervical intraepithelial neoplasia and cervical

carcinoma. Chin Med Sci J 21(2):81–85

Ma L, Liu J, Liu L, Duan G, Wang Q, Xu Y, Xia F, Shan J, Shen J,

Yang Z, Bie P, Cui Y, Bian XW, Prieto J, Avila MA, Qian C

(2014) Overexpression of the transcription factor MEF2D in

hepatocellular carcinoma sustains malignant character by sup-

pressing G2-M transition genes. Cancer Res 74(5):1452–1462

Okamoto K, Prives C (1999) A role of cyclin G in the process of

apoptosis. Oncogene 18(32):4606–4615

Okamoto K, Li H, Jensen MR, Zhang T, Taya Y, Thorgeirsson SS,

Prives C (2002) Cyclin G recruits PP2A to dephosphorylate

Mdm2. Mol Cell 9(4):761–771

Peyressatre M, Prevel C, Pellerano M, Morris MC (2015) Targeting

cyclin-dependent kinases in human cancers: from small

molecules to Peptide inhibitors. Cancers 7(1):179–237

Piscopo DM, Hinds PW (2008) A role for the cyclin box in the

ubiquitin-mediated degradation of cyclin G1. Cancer Res

68(14):5581–5590

Ramsay AG, Keppler MD, Jazayeri M, Thomas GJ, Parsons M,

Violette S, Weinreb P, Hart IR, Marshall JF (2007) HS1-

associated protein X-1 regulates carcinoma cell migration and

invasion via clathrin-mediated endocytosis of integrin alphav-

beta6. Cancer Res 67(11):5275–5284

Reimer CL, Borras AM, Kurdistani SK, Garreau JR, Chung M,

Aaronson SA, Lee SW (1999) Altered regulation of cyclin G in

human breast cancer and its specific localization at replication

foci in response to DNA damage in p53 ?/? cells. J Biol Chem

274(16):11022–11029

Russell P, Hennessy BT, Li J, Carey MS, Bast RC, Freeman T,

Venkitaraman AR (2012) Cyclin G1 regulates the outcome of

taxane-induced mitotic checkpoint arrest. Oncogene

31(19):2450–2460

Stanyon CA, Liu G, Mangiola BA, Patel N, Giot L, Kuang B, Zhang

H, Zhong J, Finley RL Jr (2004) A Drosophila protein-interac-

tion map centered on cell-cycle regulators. Genome Biol

5(12):R96

Tamura K, Kanaoka Y, Jinno S, Nagata A, Ogiso Y, Shimizu K,

Hayakawa T, Nojima H, Okayama H (1993) Cyclin G: a new

mammalian cyclin with homology to fission yeast Cig1.

Oncogene 8(8):2113–2118

Turan G, Usta CS, Usta A, Kanter M, Tavli L, Karacan M, Celik C,

Eser M (2014) The expression of HER-2/neu (c-erbB2), survivin

and cycline D1 in serous ovarian neoplasms: their correlation

with clinicopathological variables. J Mol Histol 45(6):679–687

Vargas-Hernandez VM, Moreno-Eutimio MA, Acosta-Altamirano G,

Vargas-Aguilar VM (2014) Management of recurrent epithelial

ovarian cancer. Gland surg 3(3):198–202

Wang S, Qiu L, Yan X, Jin W, Wang Y, Chen L, Wu E, Ye X, Gao

GF, Wang F, Chen Y, Duan Z, Meng S (2012) Loss of

microRNA 122 expression in patients with hepatitis B enhances

hepatitis B virus replication through cyclin G(1)-modulated P53

activity. Hepatology 55(3):730–741

Wen W, Han T, Chen C, Huang L, Sun W, Wang X, Chen SZ, Xiang

DM, Tang L, Cao D, Feng GS, Wu MC, Ding J, Wang HY

(2013) Cyclin G1 expands liver tumor-initiating cells by Sox2

induction via Akt/mTOR signaling. Mol Cancer Ther

12(9):1796–1804

Ye XX, Liu CB, Chen JY, Tao BH, Zhi-Yi C (2012) The expression

of cyclin G in nasopharyngeal carcinoma and its significance.

Clin Exp Med 12(1):21–24

Yuan DZ, Ding XL, Yu HL, Cheng Z, Tang XR, He YP, Zhang JH,

Blok LJ, Hanifi-Moghaddam P, Burger CW, Yue LM (2014)

Progesterone-induced cyclin g1 inhibits the proliferation of

endometrial epithelial cell and its possible molecular mechan-

ism. Horm Metab Res 46(11):761–767

Zhao X, Liu M, Li D (2015) Oleanolic acid suppresses the

proliferation of lung carcinoma cells by miR-122/Cyclin G1/

MEF2D axis. Mol Cell Biochem 400(1–2):1–7

302 J Mol Hist (2015) 46:291–302

123