1
Scientific RepoRts | 6:36873 | DOI: 10.1038/srep36873
www.nature.com/scientificreports
Protein tyrosine phosphatase
PTPN3 promotes drug resistance
and stem cell-like characteristics in
ovarian cancer
Shuqin Li
1,*
, Jian Cao
2,*
, Wei Zhang
3
, Fan Zhang
3
, Guantai Ni
1
, Qian Luo
1
, Man Wang
4
,
Xiang Tao
5
& Hongping Xia
3,6
The current standard treatment for ovarian cancer is aggressive surgery followed by platinum-
based combination chemotherapy. Recurrence and chemotherapeutic drug resistance are the two
main factors that account for the high mortality of most ovarian cancers. Liposomal doxorubicin is
primarily used for the treatment of ovarian cancer when the disease has progressed after platinum-
based chemotherapy. However, relatively little is known about the genomic changes that contribute
to both cisplatin and doxorubicin resistance in high-grade serous ovarian cancer (HGSC) under the
selective pressure of chemotherapy. Here, we found that protein tyrosine phosphatase PTPN3 gene
expression was substantially increased in both cisplatin and doxorubicin-resistant ovarian cancer
cells. Silencing of PTPN3 restored sensitivity to cisplatin and doxorubicin in resistant ovarian cancer
cells. Down-regulation of PTPN3 also inhibited cell cycle progression, migration, stemness in vitro and
the tumorigenicity of resistant ovarian cancer cells in vivo. Meanwhile, the expression of PTPN3 was
found to be regulated by miR-199 in resistant ovarian cancer cells. These ndings suggest that PTPN3
promotes tumorigenicity, stemness and drug resistance in ovarian cancer, and thus is a potential
therapeutic target for the treatment of ovarian cancer.
Ovarian cancer, the most common cause of gynaecologic cancer-associated death, is responsible for approxi-
mately 14,000 deaths in the United States annually
1
. Patients with advanced-stage high-grade serous ovarian can-
cer (HGSC) have experienced little improvement in overall survival
2
. e standard treatment is aggressive surgery
followed by platinum-based combination chemotherapy. Recurrence and chemotherapeutic drug resistance are
the two main factors that account for the high mortality of most ovarian cancers. Aer chemotherapy, platinum
resistance occurs in approximately 25% of patients within six months
3
. A report on the integrated genomic anal-
yses of ovarian carcinoma by e Cancer Genome Atlas Research Network showed that HGSC is characterised
by TP53 mutations in almost all tumours (96%). BRCA1 and BRCA2 are mutated in 22% of tumours, owing to a
combination of germline and somatic mutations. Low prevalence mutations include RB1, NF1, FAT3, CSMD3,
GABRA6 and CDK12. e commonly dysregulated signalling pathways, such as RB, RAS/PI3K, FOXM1 and
Notch, may provide promising opportunities for the treatment of advanced ovarian cancer
4
.
A recent study on the whole-genome characterisation of chemoresistant ovarian cancer by e Australian
Ovarian Cancer Study Group showed that CCNE1 amplication is common in primary resistant and refrac-
tory disease. Acquired chemotherapy resistance is mainly associated with the inactivation of tumour suppressors
such as RB1, NF1, RAD51B and PTEN, germline BRCA1 or BRCA2 mutations, loss of BRCA1 promoter meth-
ylation, and overexpression of the drug eux pump MDR1
2
. Doxil, a pegylated (polyethylene glycol coated)
1
Department of Gynecology and Obstetrics, Yijishan Hospital of Wannan Medical College, Wuhu, China.
2
Department
of Gynecology and Obstetrics, Nanjing Maternal and Children Care Hospital Aliated to Nanjing Medical University,
Nanjing, China.
3
Department of Pathology, Yijishan Hospital of Wannan Medical College, Wuhu, China.
4
Department
of Gynecology and Obstetrics, Xiaogan Central Hospital, Xiaogan, China.
5
Department of Pathology, Obstetrics
and Gynecology Hospital of Fudan University, Shanghai, China.
6
Department of Pathology, Sir Run Run Hospital &
Nanjing Medical University, Nanjing, China.
*
These authors contributed equally to this work. Correspondence and
requests for materials should be addressed to X.T. (email: tutufoliage@hotmail.com) or H.X. (email: xiahp82@gmail.
com)
Received: 16 April 2016
Accepted: 21 October 2016
Published: 11 November 2016
OPEN
www.nature.com/scientificreports/
2
Scientific RepoRts | 6:36873 | DOI: 10.1038/srep36873
liposome-encapsulated form of doxorubicin, is primarily used for the treatment of ovarian cancer where the
disease has progressed or recurred aer platinum-based chemotherapy
5
. Liposomal doxorubicin can be used
together with platinum if the recurrence occurs aer more than 6 months, or liposomal doxorubicin can be
given alone if the recurrence occurs within 6 months of the initial or subsequent complete clinical response to
platinum-containing chemotherapy
6
. However, relatively little is known of the genomic changes that contribute
to both cisplatin and doxorubicin resistance of HGSC under the selective pressure of chemotherapy.
Here, we undertook a search for candidate genes that could confer both cisplatin and doxorubicin resist-
ance in established resistant ovarian cancer cells. Using genomic analyses, we found that protein tyrosine
phosphatase non-receptor type 3 (PTPN3) gene expression was substantially increased in both cisplatin and
doxorubicin-resistant cells. Moreover, PTPN3 silencing restored the sensitivity of resistant ovarian cancer cells to
cisplatin and doxorubicin. Silencing of PTPN3 also inhibited cell cycle progression, migration and stemness, and
reduced the tumorigenicity of resistant ovarian cancer cells. ese ndings identify PTPN3 as a potential thera-
peutic target for ovarian cancer treatment and for overcoming cisplatin and doxorubicin resistance.
Results
PTPN3 is highly expressed in cisplatin and doxorubicin resistant ovarian cancer cells. To iden-
tify the critical factors associated with both cisplatin and doxorubicin resistance in ovarian cancer, we com-
pared the gene expression dierences in A2780, A2780CIS and A2780ADR cells obtained from the European
Collection of Cell Cultures (ECACC) using ovarian cancer cell line panel gene expression data
7
. Among the panel
of signicantly dierent genes, PTPN3 was signicantly overexpressed in both A2780CIS and A2780ADR cells
compared to A2780 cells (Table S1). e resistance of A2780CIS and A2780ADR to cisplatin and doxorubicin
was conrmed by the MTS assay (Fig.1A,B). e signicantly increased expression of PTPN3 in both A2780CIS
and A2780ADR was further validated by real-time quantitative reverse transcription PCR (qRT-PCR) and west-
ern blotting (Fig.1C,D). Importantly, the highly increased expression of PTPN3 was also detected in approxi-
mately one third of the clinical ovarian cancer tissue samples, but not detectable in normal ovarian tissue samples
(Fig.1E). e high expression of PTPN3 was signicantly associated with poor overall survival of ovarian cancer
patients (Fig.1F). ese data suggest that overexpression of PTPN3 may play a critical role in mediating both
cisplatin and doxorubicin resistance in ovarian cancer cells.
Silencing of PTPN3 inhibits cell cycle progression in resistant ovarian cancer cells. Cell cycle
progression has been shown to play important roles in drug resistance. To investigate the molecular mechanism
of PTPN3-mediated cisplatin and doxorubicin resistance in ovarian cancer cells, cell cycle analysis was used
to examine the eect of silencing PTPN3 on cell signalling pathways. Aer transfection with PTPN3 esiRNAs
(esiPTPN3), an endoribonuclease-prepared siRNA pool comprised of a heterogeneous mixture of siRNAs that
all target the same mRNA sequence of PTPN3, a signicant silencing eect of PTPN3 in both cisplatin and dox-
orubicin resistant ovarian cancer cells was conrmed by qRT-PCR (Fig.2A,B). Subsequent cell cycle analysis by
ow cytometry indicated that silencing of PTPN3 signicantly increased the G0-G1 phase cell population and
decreased the S phase cell population in both cisplatin and doxorubicin resistant ovarian cancer cells. ese data
suggest that silencing PTPN3 inhibits cell cycle progression in resistant ovarian cancer cells.
Silencing of PTPN3 inhibits resistant ovarian cancer cell growth, migration and drug resist-
ance. To investigate the critical role of PTPN3 in ovarian cancer drug resistance and cell cycle progression, we
examined the eects of silencing of PTPN3 on resistant ovarian cancer cell growth, migration and drug resistance.
e cell growth curves indicated that cell growth was dramatically reduced by silencing PTPN3 in both cisplatin
and doxorubicin resistant ovarian cancer cells (Fig.3A,B). We next examined whether PTPN3 is a critical mole-
cule for resistant ovarian cancer cell migration using the Transwell migration assay. Silencing PTPN3 signicantly
suppressed the migration rates of both cisplatin and doxorubicin resistant ovarian cancer cells (Fig.3C,D). We
then assayed drug sensitivity to cisplatin and doxorubicin. e results showed that silencing PTPN3 signicantly
contributed to increasing the sensitivity of resistant ovarian cancer cells to cisplatin and doxorubicin (Fig.3E,F).
ese data collectively indicate that silencing of PTPN3 inhibits resistant ovarian cancer cell growth, migration
and drug resistance.
Stable silencing of PTPN3 inhibits colony formation and stemness in resistant ovarian cancer
cells. Colony formation in so agar is the most widely used assay to evaluate cellular anchorage-independent
growth in vitro. So agar colony formation assays showed that stable silencing of PTPN3 signicantly inhibited
the colony forming ability of both cisplatin and doxorubicin resistant ovarian cancer cells, suggesting that sta-
ble silencing of PTPN3 inhibits cellular anchorage-independent growth of resistant ovarian cancer cells in vitro
(Fig.4A,B). Increasing evidence has suggested that the stemness of cancer cells is thought to be responsible for
cancer initiation and drug resistance. Previous studies have shown that ALDH+ and CD133+ cells are enriched
with ovarian cancer-initiating (stem) cells, and that ALDH and CD133 may be widely used as reliable markers
to investigate ovarian cancer stem cell biology
8
. Flow cytometry analysis showed that stable silencing of PTPN3
signicantly decreased the ALDH+ , CD133+ and ALDH+ CD133+ cell populations (Fig.4C,D). e tumour
sphere formation assay showed that stable silencing PTPN3 signicantly inhibited the sphere forming ability of
both cisplatin and doxorubicin resistant ovarian cancer cells (Fig.4E,F). ese data suggest that stable silencing
of PTPN3 inhibits colony formation and stemness in resistant ovarian cancer cells.
The expression of PTPN3 is regulated by miR-199 in resistant ovarian cancer cells. Previous
studies have shown that at least one-third of human genes are regulated by miRNAs
9
. Aer demonstrating the
important roles of PTPN3 in resistant ovarian cancer cells, we next investigated whether miRNAs regulate the
expression of PTPN3. To identify the potential posttranscriptional regulation of PTPN3 by miRNAs, we used
www.nature.com/scientificreports/
3
Scientific RepoRts | 6:36873 | DOI: 10.1038/srep36873
two online soware resources, i.e. TargetScan and miRDB, for prediction. A panel of miRNAs was predicted
to be potential regulators of PTPN3 by both miRNA target prediction programs (Table S3). We then used the
luciferase reporter assay to validate that PTPN3 could be potentially regulated by miR-199. To validate whether
miR-199 directly recognises the 3′ -UTR of PTPN3 mRNA, we cloned a sequence containing the predicted target
site and a mutated sequence with the predicted target sites downstream of the pGL3 luciferase reporter gene to
generate pGL3-PTPN3-wt or pGL3-PTPN3-mut vectors (Fig.5A). e vectors were then co-transfected with the
miR-199 mimics or control into HEK293 cells. A Renilla luciferase vector (pRL-TK) was used to normalise the
dierences in transfection eciency. Luciferase activity in cells co-transfected with the miR-199 mimics and the
Figure 1. PTPN3 is highly expressed in cisplatin and doxorubicin resistant ovarian cancer cells.
(A,B) Comparison of the sensitivity of A2780 cells to that of A2780CIS and A2780ADR cells to cisplatin
(A) and doxorubicin (B). *P < 0.05 (t test). (C,D) e expression of PTPN3 in cisplatin and doxorubicin
resistant ovarian cancer cells was examined by qRT-PCR (C) and western blotting (D). PTPN3 was found
to be highly expressed in cisplatin and doxorubicin resistant ovarian cancer cells. **P < 0.01 (t test). e gels
were run under the same experimental conditions. Densitometric analysis was performed to quantify the
bands. (E) Representative immunohistochemical (IHC) images of PTPN3 in ovarian cancer tissue samples.
Scale bar = 50 µ m. (F) e high expression of PTPN3 was signicantly associated with poor overall survival of
ovarian cancer patients.
www.nature.com/scientificreports/
4
Scientific RepoRts | 6:36873 | DOI: 10.1038/srep36873
pGL3-PTPN3-wt vector was signicantly decreased when compared with the control (Fig.5B). Next, we further
detected the protein expression of PTPN3 in resistant ovarian cancer cells aer transfection with miR-199 mim-
ics or control. e results show that overexpression of miR-199 decreased the expression of PTPN3 in resistant
Figure 2. Silencing of PTPN3 inhibits cell cycle progression in resistant ovarian cancer cells. (A,B) e
expression of PTPN3 was signicantly silenced by PTPN3 esiRNA transfection using qRT-PCR analysis in
both A2780CIS (A) and A2780ADR (B) cells. *P < 0.05 (t test). (C,D) Representative cell cycle analysis by ow
cytometry shows that silencing PTPN3 signicantly increased the G0-G1 phase cell population and decreased
the S phase cell population in both A2780CIS (C) and A2780ADR (D) cells.
www.nature.com/scientificreports/
5
Scientific RepoRts | 6:36873 | DOI: 10.1038/srep36873
ovarian cancer cells (Fig.5C). Consistently, overexpression of miR-199 also increased the sensitivity of resistant
ovarian cancer cells (Fig.5D,E). ese data suggest that the expression of PTPN3 is regulated by miR-199 in
resistant ovarian cancer cells.
Figure 3. Silencing of PTPN3 inhibits resistant ovarian cancer cell growth, migration and drug resistance.
(A,B) MTS assay results showing that cell growth was dramatically reduced by silencing PTPN3 in both
A2780CIS (A) and A2780ADR (B) cells. (C,D) e Transwell migration assay showed that silencing PTPN3
signicantly suppressed the migration rates of both cisplatin and doxorubicin resistant ovarian cancer cells.
Scale bar = 100 µ m. (E,F) MTS assay results showing that silencing PTPN3 signicantly contributed to
increasing the sensitivity of resistant ovarian cancer cells to cisplatin (E) and doxorubicin (F). *P < 0.05 (t test).