piR-1919609 Is an Ideal Potential Target for Reversing Platinum Resistance in Ovarian Cancer

Abstract

Purpose

PIWI-interacting RNAs (piRNAs) are a type of noncoding small RNA that can interact with PIWI-like RNA-mediated gene silencing (PIWIL) proteins to affect biological processes such as transposon silencing through epigenetic effects. Recent studies have found that piRNAs are widely dysregulated in tumors and associated with tumor progression and a poor prognosis. Therefore, this study aimed to investigate the effect of piR-1919609 on the proliferation, apoptosis, and drug resistance of ovarian cancer cells.

Methods

In this study, we used small RNA sequencing to screen and identify differentially expressed piRNAs in primary ovarian cancer, recurrent ovarian cancer, and normal ovaries. A large-scale verification study was performed to verify the expression of piR-1919609 in different types of ovarian tissue, including ovarian cancer tissue and normal ovaries, by RT–PCR and to analyze its association with the clinical prognosis of ovarian cancer. The expression of PIWILs in ovarian cancer was verified by RT–PCR, Western blotting and immunofluorescence. The effects of piR-1919609 on ovarian cancer cell proliferation, apoptosis and drug resistance were studied through in vitro and in vivo models.

Results

(1) piR-1919609 was highly expressed in platinum-resistant ovarian cancer tissues (p < 0.05), and this upregulation was significantly associated with a poor prognosis and a shorter recurrence time in ovarian cancer patients (p < 0.05). (2) PIWIL2 was strongly expressed in ovarian cancer tissues (p < 0.05). It was expressed both in the cytoplasm and nucleus of ovarian cancer cells. (3) Overexpression of piR-1919609 promoted ovarian cancer cell proliferation, inhibited apoptosis, and promoted tumor growth in nude mice. (4) Inhibition of piR-1919609 effectively reversed ovarian cancer drug resistance.

Conclusion

In summary, we showed that piR-1919609 is involved in the regulation of drug resistance in ovarian cancer cells and might be an ideal potential target for reversing platinum resistance in ovarian cancer.

Introduction

Ovarian cancer (OC) accounts for 3.4% and 4.7% of new cancer cases and cancer-related deaths in women worldwide, respectively, ranking eighth in both incidence and mortality. ¹ Its mortality rate ranks first among female reproductive tract malignant tumors, and the five-year overall survival rate is only 30%-40%. Approximately 70% of OC patients are clinically diagnosed at an advanced stage. After standard treatment, approximately 70% of these patients eventually die of drug-resistant recurrent ovarian cancer. ² Platinum resistance is an important factor contributing to the poor prognosis of OC patients. Therefore, it is very important to further explore the molecular mechanism of ovarian cancer chemotherapy resistance and find molecular targets that can be used to predict outcomes or reverse drug resistance, which can ultimately improve the clinical efficacy and prognosis of OC patients.

Drug resistance is related to a decrease in the effective drug concentration, enhancement of detoxification ability and enhancement of DNA damage repair ability in cancer cells, tumor stem cells, the tumor microenvironment, abnormal autophagy and apoptosis, epigenetic changes and so on. ³ Abnormal epigenetic modifications caused by noncoding small RNAs, such as microRNAs (miRNAs) and long noncoding RNAs (lncRNAs), have been confirmed to be associated with drug resistance in cancer cells. As a new type of noncoding small RNA, PIWI-interacting RNAs (piRNAs) are typically 24–31 bp in length. PiRNAs play an important role in inhibiting transposons, ⁴ maintaining reproductive system function, ⁵ and regulating epigenetic processes, genes and proteins. ⁶ Unlike traditional targeting mechanisms, piRNAs do not require extensive base pairing with target mRNAs. ⁷ piRNAs contain 2'-O-methylated structures, which are not easily degraded and can remain relatively stable in serum. Therefore, the discovery of piRNAs has opened a new direction for the study of noncoding small RNAs in cancer. Research has revealed that piRNAs are highly expressed in breast cancer, ⁸ colorectal cancer, ⁹ non-small cell lung cancer, ¹⁰ renal cell carcinoma ¹¹ and other malignant tumors, and they are involved in the regulation of tumor cell proliferation, apoptosis, invasion, and metastasis.

Balaratnam et al ⁷ found that a piRNA utilizes HILI and HIWI2-mediated pathways to downregulate FTH1 expression, and piRNA-mediated FTH1 downregulation increased the sensitivity of breast cancer cells to doxorubicin by 20-fold after treatment compared with that in the control group. Lin Dongxin et al ¹² found that piR-36712 was significantly expressed at low levels in breast cancer tissue through biosignature mining and clinical tissue sample verification, which promoted the expression of SEPW1 through the competing endogenous RNA (ceRNA) mechanism, thereby reducing the level of p53 and leading to tumor proliferation, invasion, metastasis, and chemotherapy resistance. In colon cancer tissue, overexpression of piR-54265 reduced the drug sensitivity of 5-FU and oxaliplatin, resulting in chemoresistance. ¹³ Similarly, overexpression of piR-39980 resulted in a reduction in early and late apoptotic cells in neuroblastoma and reduced neuroblastoma sensitivity to doxorubicin at the same time. ¹⁴ Aberrant expression of piRNAs is a potential signal of cancer development.

Studies of the PIWI protein in the drug resistance of ovarian cancer have also attracted increasing attention. Overexpression of PIWIL2 leads to increased resistance to cisplatin in ovarian cancer cells, and the downregulation of PIWIL2 expression could enhance the sensitivity of cancer cells to cisplatin by inhibiting the repair of cisplatin-induced DNA intrastrand crosslinks (Pt-GG). ¹⁵ Previous research has shown that ¹⁶ PIWIL3 was less expressed in malignant EOC and benign tissues than in normal ovaries, while PIWIL1, PIWIL2, and PIWIL4 were significantly expressed in malignant EOC. In the analysis of the correlation between the prognosis of multiple tumors and the expression of PIWIL3, no significant correlation was found, ¹⁷ indicating that PIWIL3 may have a limited impact on the occurrence and development of tumors. In conclusion, PIWI protein may be involved in the regulation of OC progression, but the specific regulatory mechanism has not been elucidated.

In this study, small RNA sequencing was performed on different types of ovarian tissue samples to identify the differentially expressed piRNAs, and then the relationship between piR-1919609 and clinicopathological factors was analyzed. Next, we performed further cell experiments in vitro and tumor xenograft experiments in nude mice to verify the role of piR-1919609 in regulating platinum resistance in OC cells.

Materials and Methods

Patients and Tissue Samples

In this study, samples from 85 patients with primary ovarian cancer and 49 patients with recurrent ovarian cancer (patients who achieved clinical remission after satisfactory cytoreductive surgery and sufficient standard clinical chemotherapy but experienced clinical tumor recurrence within six months or more and underwent reoperation) were collected for piRNA-related research. The 49 patients with recurrent ovarian cancer included 29 patients with relapsed platinum-sensitive ovarian cancer (patients who responded to platinum-based drug treatment and achieved clinical remission but experienced lesion recurrence after stopping chemotherapy for more than 6 months) and 20 patients with recurrent platinum-resistant ovarian cancer (patients who responded to initial chemotherapy but relapsed within 6 months after completing chemotherapy). Forty normal ovaries (normal ovary tissues obtained from benign nonovarian tumor patients) were also collected for piRNA-related research. All patients were hospitalized in the Department of Gynecology, Guangxi Medical University Cancer Hospital for surgery, and the postoperative pathological diagnosis was epithelial ovarian cancer. The samples were objectively collected according to the above standards, and as serous ovarian cancer is the most common pathological type in epithelial ovarian cancer, it accounts for the majority of the collected samples. All fresh frozen tissue specimens were stored in RNA storage solution and immediately stored at −80 °C. The original study was reviewed by the Guangxi Medical University Cancer Hospital Ethical Review Committee (approval number: LWB2019001), and relevant informed consent forms of all participants were signed prior to all specimens collection.

RNA Extraction and Small RNA Sequencing

Total RNA was extracted from tissues using TRIzol reagent (GIBCO-BRL). A Fragment Analyzer 5200 was used to identify the integrity of the RNA samples and obtain the RNA integrity score (RIN) (completed by BGI). RNA samples that met the standard (total RNA: OD260/280 value: between 1.8 2.2, 28S:18S ≥ 0.8, RIN ≥ 6.5) were selected for small RNA sequencing. All selected RNA samples were sent to the BGI Gene Research Department (Shenzhen, China) for BGISEQ UMI Small RNA sequencing. The small RNA sequencing flow chart is shown in Supplemental Figure S1.

RT–PCR for Detection of piR-1919609

The 9 drug resistance-related piRNAs obtained by sequencing and screening were validated using an expanded set of clinical samples. One microgram of total RNA containing piRNAs was extracted by TRIzol reagent (GIBCO-BRL, RNeasy FFPE Kit 50, Qiagen) and used as starting material. cDNA was prepared by using the miScript II Reverse Transcription Kit (Qiagen). Fluorescent quantitative PCR was performed using the miScript SYBR Green PCR Kit (Qiagen). The cDNA was diluted to a certain multiple and added to the reaction system for amplification. PCR was performed on a Q5-Studio RT‒PCR instrument. The expression of piRNAs was relatively standardized with U6 as the internal reference gene, and the relative expression level was expressed as 2^−△△Ct. The primer sequences for the piRNAs and U6 are shown in Supplemental Table S1.

RNA Isolation and Quantification of PIWILs

We detected the expression of PIWIL1, PIWIL2, and PIWIL4 mRNA in ovarian cancer tissues by RT‒PCR. After total RNA extraction from OC tissues, cDNA preparation was performed using the PrimeScript™ RT Reagent Kit (Takara) and controlled template RNA ≤ 500 ng. PCR was performed on a Q5-Studio RT‒PCR instrument. The expression of PIWILs was relatively standardized with GAPDH as the internal reference gene, and the relative expression level was expressed as 2^−△△Ct. The specific sequences of the PIWIL and GAPDH primers are shown in Supplemental Table S2.

Western Blot Analysis

Total protein was extracted from OC tissues, normal ovarian tissues, and OC cell lines (SKOV3/DDP, SKOV3/DDP-piR-NC, SKOV3/DDP-piR-Inhibitor) by using RIPA + PMSF (Thermo), controlling the amount of each protein sample to 40 µg (BCA Kit, Thermo). Then, the PierceTM BCA Protein Assay Kit (Thermo) was used to quantify the protein concentration. Western blot analysis was performed by the laboratory's scanning developer (Bio-Rad). Anti-PIWIL1 (ab12337), anti-PIWIL2 (ab181340), and anti-PIWIL4 (ab111714) antibodies were from Abcam, anti-Bcl-2 antibody (WL01556) was from Wanleibio.Antibody Diluent (A1800) was from Solarbio. The dilution ratio was 1:1000.

Immunofluorescence of PIWILs in Ovarian Cancer Cells

The presence and localization of PIWILs in the OC cell line SKOV3 were determined by immunofluorescence. SKOV3 cells were cultured at a density of 1 × 10⁴ cells/ml with L-lysine-coated coverslips in a CO₂ incubator for 24 h. The next day, the cells were fixed with 4% paraformaldehyde solution for 20 min, permeabilized with 0.3% Triton-X-100 (1×PBS diluted) solution for 10 min, and finally blocked in 1% BSA for 30 min. The above steps were carried out at room temperature. Then, we added anti-PIWIL1, anti-PIWIL2, and anti-PIWIL4 (Abcam) primary antibodies (dilution ratio 1:500) to the well plate and incubated overnight at 4 °C. In the next step, the cells were incubated with fluorescent secondary antibody for 2 h in the dark. After washing again, the cells were redyed with DAPI solution. Finally, the cells were observed with an ImageXpress Micro Confocal instrument (Molecular Devices) to obtain images.

Cell Culture

The human OC cell lines SKOV3 and A2780 were purchased from the National Collection of Authenticated Cell Cultures. The DDP-resistant cell lines SKOV3/DDP and A2780/DDP were obtained by cisplatin induction from our laboratory. ¹⁸ The above cell lines were cultured in 10% FBS (Gibco) + 1640 (Gibco) supplemented with a 1% penicillin‒streptomycin mixture (Solarbio) in a 37 °C, 5% CO₂ incubator. When the cells grew to an exponential growth phase and the bottom of the culture flask was 90% confluent, they were passaged. The expression of piR-1919609 in each cell line was detected by RT‒PCR.

Lentivirus Preparation and Transfection

PiR-1919609 was overexpressed or knocked down by transfection with GFP-labeled LV-piR-1919609-inhibition lentivirus, LV-piR-1919609-mimics lentivirus, and their negative control (NC) lentivirus, which were packaged and synthesized by Shanghai Jikai Gene Co., Ltd (Shanghai, China). The virus carried a puromycin resistance gene and expressed GFP fluorescent protein. The LV-piR-1919609-inhibitor lentivirus and its NC lentivirus were transfected into SKOV3/DDP and A2780/DDP cells to construct the cell line with silenced piR-1919609 (piR-Inhibitor-AD/SD) and the NC cell line (piR-LV-CON-SD/AD). The sequence of LV-piR-1919609-inhibitor was as follows: 5’-ACTAAACCCAGTAATGGTAACGGTTT-3’. LV-piR-1919609-mimics lentivirus and its NC lentivirus were transfected into SKOV3 and A2780 cells to construct the cell line that overexpressed piR-1919609 (piR-mimic-SKOV3/A2780) and the NC cell line (piR-LV-CON-SKOV3/A2780). The sequence of the LV-piR-1919609-mimics was as follows: 5’-AAACCGTTACCATTACTGGGTTTAGT-3’. The stably transfected strains were screened with puromycin. We detected the expression of piR-1919609 in the above cell lines by RT‒PCR.

Cell Counting Kit-8 (CCK8) Analysis

Cell viability was measured in all groups. The cells in the growth log phase were taken from the inhibition group (piR-Inhibitor-AD/SD) and the mimic group (piR-mimic-SKOV3/A2780), digested with 0.25% trypsin, centrifuged, and resuspended into a single cell suspension. They were inoculated in a 96-well culture plate at a cell density of 3 × 10³ cells/well and set to 3 replicate wells, which were incubated in a 37 °C, 5% CO₂ incubator environment for 24 h. A row of cells was selected every 24 h, 100 μL of 1640 medium + 10 μL of CCK8 reagent (Meilunbio) was added, and the cell-free wells were used as blank wells and incubated for 2 h with 5% CO₂ at 37 °C. An automatic microplate reader (BioTek, USA) was used to measure the optical density (OD) values at an excitation wavelength of 450 nm. Continuous detection was performed for 6 days, and the growth curve of each group of cells was drawn.

Diamminedichloroplatinum (DDP) induction: After the DDP concentration gradient test, the mimic group and the NC cells were treated with 1 µg/ml DDP (100 µg/well), and the inhibitor group (SD-inhibitor, AD-inhibitor) and its NC group were treated with 2 µg/ml DDP (100 µl/well). The above experiment was repeated after 24 h of culture.

Examining the Chemosensitivity and Half-Maximal Inhibitory Concentration (IC50) of DDP by CCK8 Assay

The sensitivity of OC cells to cisplatin in the piR-1919609 mimic group or inhibition group was detected by CCK-8 assay. Cells from the mimic group (piR-mimic-SKOV3/A2780) and the inhibition group (piR-Inhibitor-AD/SD), as well as the respective NC groups, were seeded separately in 96-well plates at 3 × 10³ cells/well. Fresh culture medium containing DDP at 0.5 µg/ml, 1 µg/ml, 2 µg/ml, 4 µg/ml, and 8 µg/ml was added to the experimental wells, and the blank wells contained the same amount of fresh medium without DDP. All of the above conditions were set up with 5 replicate wells. The plates were cultured in a 37 °C, 5% CO₂ incubator for 48 h. Then, 100 L of 1640 medium + 10 μL of CCK-8 was added to each well, and the cell-free wells were used as blank wells and incubated for 2 h. The OD value of each well was detected by an automated microplate reader at a wavelength of 450 nm. The IC50 of DDP was calculated by SPSS 20.0 statistical software, and statistical graphs were drawn by GraphPad Prism 5.0.

Apoptosis Study via Flow Cytometry

Cell apoptosis rates were detected by annexin V-PE staining. The SKOV3/DDP, piR-Inhibitor-SD, and piR-LV-CON-SD cell lines were digested with EDTA-free trypsin, and the trypsin was aspirated after 5–10 min. Then, the cells were mixed with culture solution and centrifuged, and the cell precipitate was collected. The cells were stained with PE Annexin V and PE using the PE Annexin V Apoptosis Detection Kit (BD). Early apoptotic cells were showed (PE+/7-AAD –) in the first quadrant, when combined with PE and 7-AAD staining, late apoptotic cells and necrotic cells showed dual positivity (PE+/7–AAD+) in the fourth quadrant, both of which quadrant was used to calculate apoptosis. The percentage of apoptotic cells was detected by flow cytometry (CvtoFLEX, Beckman Coulter).

Tumor Xenografts in Nude Mice

All animal protocols were approved by the Medical Ethics Committee of Guangxi Medical University Cancer Hospital (approval number: LWB2019002), and animal experiments were performed in accordance with the International Convention on Laboratory Animal Ethics and the ‘Guide for the Care and Use of Laboratory Animals, eighth Edition’. ¹⁹ The reporting of this study conformed to ARRIVE 2.0 guidelines. ²⁰ Referring to previous experiment, ²¹ we set the sample size included a total of 20 nude mice, with 10 in each group. For hypothesis-testing studies, we computed achieved power through post hoc test by Gpower. It is estimated that a significant level <0.05 and a power >85% between the two groups are considered appropriate for this experiment. A total of 20 SPF-grade BALB/C female nude mice (aged 4 weeks, weight of 16-18 g) were purchased from the Animal Experiment Center of Guangxi Medical University (Nanning, China). They were housed in a sterile, constant temperature, special pathogen-free SPF grade animal laboratory and provided standard laboratory feed. Twenty nude mice were randomly divided into a mimic group (piR-mimic-SKOV3) and a negative control group (piR-NC-SKOV3) by Simple sorting randomization, with 10 mice in each group. Two hundred microliters of piR-mimic-SKOV3 and piR-NC-SKOV3 single-cell suspensions (5 × 10⁷/ml) were subcutaneously injected into the armpits of nude mice in each group. When masses were observed, the volume was calculated every other day as follows: tumor volume (V) = 1/2 length (mm)×width (mm)². When the tumor-like masses reached 200 mm³, the nude mice were treated with DDP intervention treatment: (1) 10 nude mice in each group were randomly divided into the DDP group and the NS (normal saline) group, with 5 mice in each group; (2) the DDP group was injected with cisplatin (0.25 mg/ml concentration, the dose was 2.5 mg/kg ²¹ ) every other day, while the NS group was injected with the same amount of normal saline. After a total of 8 injections, the tumor volume of each group was assessed before each injection, and the tumor inhibition rate was calculated (the tumor inhibition rate = (mean V_nc-V_DDP)/V_nc× 100%). The flowchart was created in The Experimental Design Assistant (EDA, https://eda.nc3rs.org.uk:443/eda/modelPublicExport/index/FA5E544651173BAE6E3C6A9CC22DBDD0). Mice were euthanized by cervical dislocation the day after the last injection, we confirmed death by observing if animals were unable to move, unresponsive to external stimuli, or had no heartbeat. The subcutaneous tumor-like tissues were stripped, photographed, and stored at −80 °C.

Statistical Analysis

SPSS 20.0 was used for statistical analysis. The measurement statistics were expressed as the mean ± standard error (mean ± SEM) of the three independent experimental datasets. GraphPad Prism 5.0 was used for statistical mapping. The data between different groups were analyzed by t test and analysis of variance. The post hoc test was conducted in Gpower to compute achieved power. Bartlett's statistic analysis was used following ANOVA as post hoc test. P < 0.05 was considered statistically significant. The Mann‒Whitney test and Spearman correlation analysis were used to analyze the correlation between piR-1919609 expression and clinicopathological factors (*p < 0.05; **p < 0.01; ***p < 0.001).

Result

Identification of Differentially Expressed piRNA Profiles in Ovarian Cancer

After RNA quality inspection, a total of 11 samples were sent for small RNA sequencing. The tissue sample information is shown in Supplemental Table S3. The number of piRNAs detected in each sample from small RNA sequencing is shown in Supplemental Table S4. To further expand the verification of tissue samples, we collected tissue sample information, as shown in Supplemental Table S5. From the sequencing results, we divided the samples into 4 groups:

C: primary ovarian cancer, R: recurrent ovarian cancer (which included RS: relapsed platinum-sensitive ovarian cancer and RR: recurrent platinum-resistant ovarian cancer), and N: normal ovary. Through pairwise comparison by the DESeq2 algorithm, four groups of differentially expressed piRNA volcano plots and histograms were obtained, as shown in Figures 1 and 2.

Figure 1.

Differentially expressed piRNAs between each pair of groups (A. C vs N; B. R vs N; C. C vs R; D. RR vs RS)

(notes: the red points indicate the upregulated genes screened based on |fold change|>2 and a corrected P value < 0.05. The green points indicate the downregulated genes screened based on |fold change|>2.0 and a corrected P value < 0.05. Black dots indicate genes with no significant difference. FC is a fold change.).

Figure 2.

Statistics of differentially expressed piRNAs.

Screening of Coexpressed piRNAs in the Different Groups

By Venn analysis of the above four groups of differentially expressed piRNAs, a total of 19 common differentially expressed piRNAs were screened out (Figure. 3). Ten of these 19 piRNAs were significantly upregulated in the drug-resistant group, while another 9 piRNAs were downregulated, as shown in Supplemental Table S6. We further selected three piRNAs, piR-hsa-1888559, piR-hsa-1919609, and piR-hsa-395465, which with differences of more than 10 times for sample validation.

Figure 3.

Screening of common differentially expressed piRNAs among the four groups.

piR-1919609 is Significantly Upregulated in Platinum-Resistant OC, and piR-1919609 Upregulation is Related to Poor Prognosis in OC Patients

Through sample verification in a larger sample set (85 patients with primary ovarian cancer, 49 patients with recurrent ovarian cancer, and 40 normal ovaries), we surprisingly found that piR-1919609 was upregulated in primary ovarian cancer tissues compared with normal ovarian tissues (Figure 4A. p = 0.0068, **); at the same time, it was upregulated approximately 7.18 times in the platinum-resistant group (Figure 4B. p = 0.0002, ***) compared with the platinum-sensitive group, which confirmed that piR-1919609 may be involved in OC tumor progression and drug resistance. However, there was no statistically significant difference in the expression of piR-1888559 or piR-395465 in primary ovarian cancer tissue and recurrent ovarian cancer tissue compared to normal ovarian tissue (Supplemental Figure. S2). We also performed PCR detection for other downregulated piRNAs, but some of them were not expressed, while others were not statistically significant. The correlation analysis between piR-1919609 expression and the clinical pathological factors of the patients was carried out by the Mann‒Whitney test. According to the median expression of piR-1919609, 85 samples from primary ovarian cancer patients were divided into high expression and low expression groups, and the statistical results are shown in Table 1. The results showed that piR-1919609 was highly correlated with OC lymph node and omental metastasis (p < 0.01,*) and FIGO stage (p < 0.05,**), but had no significant correlation with patient age, histological grade, or residual lesion size (P > 0.05). We also analyzed the expression of piR-1919609 in 49 samples from recurrent OC patients by Spearman correlation analysis. The results showed that the expression of piR-1919609 was highly negatively correlated with time to recurrence, and this correlation was statistically significant (p = 0.002, **), as shown in Table 2. For the above results, piR-1919609 was selected for follow-upstudies.

Figure 4.

piR-1919609 is overexpressed in epithelial ovarian cancer.

(piR-1919609 expression in: (A). 40 normal ovary tissues versus 85 primary ovarian cancer versus 49 recurrent ovarian cancer and (B). 19 platinum-resistant tissue versus 30 platinum-sensitive tissue as detected by RT‒PCR. The results were normalized according to U6 expression using the 2^−ΔΔCT method.).

Table 1.

Correlation Analysis Between piR-1919609 Expression and Clinicopathological Features in Epithelial Ovarian Cancer

Clinicopathological features	piR-1919609 expression		Z	P value
	Lower than median n = 42）	Higher than median (n = 43)
Age (median)
<54	17	24	-1.407	0.160
≥54	25	19
FIGO stage
I-II	16	8	-1.984	0.047*
III-IV	26	35
Histological grade
Low grade	4	6	-0.630	0.529
High grade	38	37
Lymph node/omental metastasis
No	21	8	-3.034	0.002**
Yes	21	35
Residual lesion (cm)
≤1 cm	36	39	-0.709	0.478
>1 cm	6	4

Table 2.

Correlation Analysis Between piR-1919609 Expression and Relapse Time in Epithelial Ovarian Cancer

Relapse time	piR-1919609 expression		R	P value
	Lower than median n = 25	Higher than median (n = 24)
<3 months	3	9	-0.430	0.002**
3–6 months	2	6
6–12 months	3	4
>12 months	16	6

PIWIL2 is Significantly Upregulated in Primary Ovarian Cancer

We confirmed that piR-1919609 is significantly related to the recurrence and drug resistance of OC. Next, we explored the influence of PIWILs on OC. The relative expression of PIWIL mRNA was detected by RT‒PCR. The results showed that the expression of PIWIL2 mRNA was significantly increased in primary ovarian cancer tissues (3.070 ± 0.513 vs 1.460 ± 0.267, p < 0.05) compared with normal ovarian tissues. There was no significant difference in PIWIL1 mRNA between the cancer group and the normal group (2.844 ± 1.006 vs 1.300 ± 0.268, p > 0.05), and the same was true for PIWIL4 mRNA (1.202 ± 0.260 vs 1.385 ± 0.202, p > 0.05), as shown in Figure. 5. The expression of PIWIL proteins in OC was detected by Western blot analysis. The results indicated that the expression of PIWIL2 and PIWIL4 in OC tissue was significantly higher than that in normal ovarian tissue (p < 0.05), whereas the expression of PIWIL1 protein was not significantly different between the two groups (p > 0.05), as shown in Figure. 6. Taken together, these findings indicated that PIWIL proteins may be associated with the progression of OC.

Figure 5.

Rt‒PCR of the relative expression of PIWIL1, PIWIL2, and PIWIL4 in different ovarian tissues.

Figure 6.

(A). PIWI protein expression in ovarian tissue. (B). PIWI protein expression statistics

(notes: OC: primary ovarian cancer; primary ovarian cancer: C1-C3; normal ovary: N1-N3.).

Immunofluorescence Detection of PIWILs in SKOV3 OC Cells

Next, we observed the cellular location of PIWIL proteins in OC cells by immunofluorescence, aiming to provide a reference for subsequent studies on the mechanism of PIWILs in OC at the cellular level.

The results showed the following: (1) PIWIL1 fluorescence was detected as positively expressed in SKOV3 cells. It was mainly expressed in the nucleus, and the fluorescence distribution was uniform. (2) PIWIL2 had strong positive expression in SKOV3 cells, and it was enriched in both the cytoplasm and the nucleus; the fluorescence distribution was relatively uniform. (3) PIWIL4 was weakly positive in SKOV3 cells; the fluorescence was mainly enriched in the nucleus, and the fluorescence distribution was uneven. All of the above results are shown in Figure. 7. In summary, Western blotting analysis and immunofluorescence assays indicated that PIWIL proteins are expressed in OC and may play a regulatory role in OC by localizing to the nucleus.

Figure 7.

PIWIL protein localization in SKOV3 cells by imageXpress micro confocal (200X).

piR-1919609 Shows High Expression in a DDP-Resistant Cell Line

To further study the effect of piR-1919609 on the malignant phenotype of OC cells, we detected the expression level of piR-1919609 in OC cell lines. We found that piR-1919609 was highly expressed in DDP-resistant cell lines (SKOV3-DDP, A2780-DDP, p < 0.05) compared with the parent strains (SKOV3, A2780), as shown in the Supplemental Figure S3. After virus transfection, the transfection efficiency of the inhibition group (piR-Inhibitor-SD/AD) and the mimic group (piR-mimic-SKOV3, piR-mimic-A2780) as well as their corresponding NC group were detected by RT‒PCR, and the results are shown in Supplemental Figure. S3.

The Effect of piR-1919609 on the Biological Function of OC Cells

piR-1919609 Promotes OC Cell Proliferation

Next, we explored the effect of piR-1919609 on the proliferation of OC cells. The viability of OC cells was measured by CCK-8 assay for one week, and then a proliferation curve was drawn. The results revealed that overexpression of piR-1919609 in OC cells (SKOV3, A2780) significantly increased cell proliferation (Figure. 8), whereas knockdown of piR-1919609 had the opposite result (Figure. 8), which indicated that piR-1919609 could effectively affect the proliferation of OC cells.

Figure 8.

Proliferation curve of ovarian cancer cells after infection with piR-1919609

(A-B: mimic group, C-D: inhibitor group, ns: p > 0.05, * p < 0.05，** p < 0.01, *** p < 0.001, the above experiment repeated three times independently.).

Inhibition of piR-1919609 Induced Apoptosis of SKOV3/DDP Cells

Since the overexpression of piR-1919609 promoted OC cell proliferation, we wondered whether piR-1919609 modulated cell apoptosis in OC. The apoptosis rate of the inhibition group and its NC group was detected by flow cytometry. Surprisingly, the apoptosis rate of the inhibition group was increased significantly compared to that of the NC group (7.373 ± 0.4331 vs 2.653 ± 0.2489, p = 0.0007), as shown in Figure. 9. This result implied that knockdown of piR-1919609 may promote apoptosis of OC drug-resistant cells.

Figure 9.

(A). Flow cytometry detection of cell apoptosis rates in SKOV3/DDP cells after transfection with the piR-1919609 inhibitor; (B). Cell apoptosis rate statistics (* p < 0.05, ** p < 0.01, *** p < 0.001, the above experiment repeated three times independently.).

Effects of piR-1919609 on the Expression of Apoptosis-Related Proteins

The above experiments showed that the inhibition of piR-1919609 had a positive effect on apoptosis. Next, we further studied apoptosis signaling pathway-related proteins. The effect of piR-1919609 on Bcl-2 was detected by Western blotting. The results revealed that Bcl-2 was expressed in SKOV3/DDP cells. When piR-1919609 was inhibited, the expression of Bcl-2 appeared to show a downward trend. However, due to experimental limitations, there was no statistically significant difference, as shown in the Supplemental Figure. S4.

piR-1919609 Reduces DDP Sensitivity in Ovarian Cancer Cells

In previous studies, we demonstrated that the expression of piR-1919609 in drug-resistant OC tissues was significantly increased (p < 0.01) and was obviously correlated with a shorter recurrence time in OC. Thus, we further investigated the effect of piR-1919609 on chemoresistance. The IC50 values of OC cell lines (without viral transfection) are shown in Supplemental Table S7. Cell viability analysis revealed that after inhibition of piR-1919609, the IC50 value of DDP was significantly decreased (p < 0.05). In contrast, the IC50 of DDP in the mimic group was significantly increased (p < 0.05), as shown in Figure. 10. The results proved that piR-1919609 decreased the sensitivity of OC cells to DDP.

Figure 10.

IC50 of DDP in ovarian cancer cells after transfection with piR-1919609

(A. piR-mimic group; B. piR-inhibitor group, * p < 0.05, ** p < 0.01, *** p < 0.001, the above experiment repeated three times independently.).

Next, based on the above drug sensitivity results, cellular proliferation was assessed by CCK-8 assay after 24 h of DDP treatment. The results showed that after DDP induction, the cells in the mimic group still proliferated rapidly (p < 0.01), while cell proliferation in the inhibition group was significantly inhibited (p < 0.01), as shown in Figure. 11. It was concluded that piR-1919609 reduced DDP-mediated cell damage and enabled ovarian cancer cell survival.

Figure 11.

Proliferation curve of cells transfected with the piR-1919609 mimic or inhibitor after treatment with 1 µg DDP (A-B: mimic group, C-D: inhibitor group, ns p > 0.05, * p < 0.05，** p < 0.01, *** p < 0.001, the above experiment repeated three times independently.).

piR-1919609 Promotes Tumor Growth and DDP Resistance in Vivo

Through in vitro experiments, we confirmed that piR-1919609 could effectively promote cell proliferation and induce drug resistance in OC cells. Next, we assessed the effect of piR-1919609 on tumors in vivo by tumor xenograft experiments in nude mice.

piR-1919609 Promoted Subcutaneous Tumor Growth in Nude Mice

piR-mimic-SKOV3 and piR-NC-SKOV3 cells were subcutaneously injected into the armpits of nude mice in each group. After 10 days, obvious masses were observed at the inoculation site, and the tumor formation rate was 95% (1 nude mouse in the piR-NC-SKOV3 group did not form tumors, which was excluded). The tumor volume was measured every other day. We observed that 12–18 days after injection, the size of tumor-like masses in the piR-mimic-SKOV3 group was significantly increased compared with that in the piR-NC-SKOV3 group (P < 0.01), indicating that overexpression of piR-1919609 promoted subcutaneous tumor growth in nude mice, as shown in Figure. 12. The Post hoc test by Gpower demonstrated that the power was >85%.

Figure 12.

(A). Tumor volume of nude mice after inoculation with ovarian cancer cells at 12–16 days; (B). Tumor inhibition rate after DDP injection in nude mice C. Nude mouse tumor samples

(tumor inhibition rate: used to objectively evaluate the effectiveness of DDP in tumor treatment * p < 0.05，** p < 0.01, *** p < 0.001).

piR-1919609 Increased DDP Resistance in Tumors

Next, we performed experimental chemotherapy by adding DDP to the mouse xenograft tumor model. By intraperitoneal injection of DDP every other day, we observed that subcutaneous tumors in the piR-mimic-SKOV3 group were more resistant to DDP than tumors formed from the piR-NC-SKOV3 group, while the tumors in the piR-NC-SKOV3 group were significantly reduced. Statistics of the tumor volume of nude mice after injection of DDP 1–8 were recorded. We calculated the tumor inhibition rate (the tumor inhibition rate = (mean V_nc-V_DDP)/V_nc× 100%), and the difference was significant between the two groups (p < 0.05), as shown in Figure. 12. After the nude mice in each group were sacrificed, the specimens were collected for later experiments, as shown in Figure. 12.

Discussion

Ovarian cancer ranks first among the three major gynecological malignancies in terms of mortality and recurrence rates. Due to the lack of tumor markers that can effectively identify OC in the early stage, the diagnosis and treatment of early OC are difficult to achieve. ²² The current standard of treatment for OC is the combination therapy of surgery and platinum-based chemotherapy, but it is only effective for most early-stage patients; for advanced-stage patients, it only gradually shortens the disease-free survival period, eventually leading to drug resistance, intestinal obstruction, and death. ²³ In OC treatment, the exploration of cancer-specific potential markers is important for the prediction and diagnosis of early-stage OC.

Recent studies have shown that aberrantly expressed noncoding RNAs induce OC tumorigenesis and chemoresistance by mediating multiple cellular phenotypes, including promotion of cell proliferation, inhibition of apoptosis, and EMT.^24,25 The study of the epigenetic regulatory mechanism of noncoding small RNAs may provide new research ideas for the treatment of recurrent OC. PiRNAs are a distinct class of small noncoding RNAs that have better stability and clearer mRNA regulatory mechanisms than miRNAs. When PIWIL homologs are highly expressed, piRNAs can effectively exert epigenetic effects, resulting in aberrant methylation and excessive silencing of the genome (including tumor suppressor genes), creating a ‘stem-like state’, leading to tumorigenesis. ²⁶

Initially, piRNAs were found to be abnormally expressed in gastrointestinal tumors. Later, abnormally expressed piRNAs were successively found in many types of tumors, such as breast cancer, ²⁷ hepatocellular carcinoma, ²⁸ and renal clear cell carcinoma. ²⁹ However, the role and specific regulatory mechanisms of piRNAs in OC tumor progression have not been elucidated. This study provides the differential expression profiles of piRNAs in primary OC and recurrent OC for the first time. We found that there were many abnormally expressed piRNAs, including upregulated and downregulated piRNAs in OC. Through expanded clinical sample validation, it was confirmed that piR-1919609 was related to a shorter recurrence time and a poor prognosis of epithelial ovarian cancer, we also found that other three subtypes of ovarian cancer were all overexpressing piR-1919609, which implied that piR-1919609 may be involved in tumor progression and the drug resistance of OC.

PiRNAs play an epigenetic role by interacting with PIWIL proteins. Studies have found that PIWIL proteins can mediate transposon silencing through DNA methylation, indicating that PIWIL may be localized in the nucleus. We performed preliminary research on the expression of PIWILs in OC and confirmed that the mRNA and protein levels of PIWIL2 were both high in OC (p < 0.05). Our immunofluorescence experiments revealed that PIWIL2 was enriched in the cytoplasm and nucleus, while PIWIL1 was mainly expressed in the nucleus. In short, our study confirmed the existence and complexity of PIWIL in OC, whether it plays a role through localized into the nucleus to regulate OC progression needs to be studied in further progression.

Studies have reported that piRNAs often act as tumor promoters or suppressors in the process of tumor progression, which affects tumor proliferation, apoptosis, invasion, and metastasis. ³⁰ After successfully constructing piR-1919609-inhibitor and piR-1919609-mimic cell lines by lentiviral transfection, we detected cell proliferation and confirmed that overexpression of piR-1919609 effectively promoted the proliferation of OC cells. Apoptosis and proliferation are basic biological processes. The occurrence of tumors is commonly caused by cell mutations that enable the cells to escape apoptosis, resulting in infinite proliferation, invasion and metastasis, ultimately leading to drug resistance. Studies have confirmed that piRNAs mediate cell apoptosis. ¹⁴ The flow cytometry results showed that the inhibition of piR-1919609 increased the apoptosis rate of OC cells. The above results indicated that piR-1919609 affected cell proliferation and apoptosis in OC.

Next, the effect of piR-1919609 on apoptosis-related proteins was investigated. Data have shown that the antiapoptotic protein Bcl-2 is a regulator of tumor drug resistance, and its dysregulation could lead to tumor development and resistance to tumor therapy.^31,32 Bcl-2 family proteins are divided into antiapoptotic proteins (Bcl-2, Bcl-XL) and proapoptotic proteins (Bax, Bak). The antiapoptotic Bcl-2 protein has prosurvival properties and maintains the integrity of the mitochondrial outer membrane (MOM) by inhibiting proapoptotic proteins and inhibiting apoptosis. ³³ Yang et al showed that cisplatin resistance in ovarian cancer was positively correlated with the upregulation of Bcl-2 expression. ³⁴ In our experiment, the expression of Bcl-2 in SKOV3/DDP ovarian cancer cells was detected by Western blotting. After inhibiting piR-1919609, the expression of Bcl-2 showed a downward trend, which indicated that piR-1919609 may have a regulatory effect on apoptotic proteins. However, the mechanism by which it regulates the apoptotic pathway may be complicated and may affect cell proliferation, potentially being related to the altered sensitivity of tumor cells to chemotherapy.

Chemotherapy resistance is the greatest challenge faced by patients with advanced OC and eventually leads to intestinal obstruction, ascites, etc Cisplatin, also known as diamminedichloroplatinum (DDP), is a first-line chemotherapeutic drug for OC tumor treatment. DDP can bind to DNA, resulting in cross-linking between DNA strands, inhibiting DNA function, and inhibiting cell mitosis and proliferation. ³⁵ The dysregulated expression of small noncoding RNAs, such as miRNAs, is thought to be significantly associated with chemotherapy resistance in many kinds of tumors. Recently, a piRNA-like small RNA, piR-L138, was identified as a key factor in cisplatin resistance in lung cancer patients with squamous cell carcinoma. ³⁶ We speculated that dysregulated piRNAs in OC were associated with chemoresistance. In the early stage of this study, we confirmed that piR-1919609 was highly expressed in platinum-resistant OC tissues and DDP-resistant OC cells (SKOV3-DDP and A2780-DDP). After DDP induction, changes in cell proliferation in vitro and tumor volume in nude mice in vivo were observed. The results showed that when piR-1919609 was overexpressed, the cell proliferation of the mimic group showed an upward trend in vivo, while the tumor volume of the piR-mimic-SKOV3 group in vitro in nude mice was significantly greater than that of the NC group. The above results indicated that piR-1919609 not only promoted tumor cell growth but also increased tumor cell chemoresistance in the cellular environment in vivo. In the in vitro tumor environment, overexpression of piR-1919609 could also effectively promote tumor growth and increase DDP resistance in nude mice. We also confirmed the above points by reverse verification. Thus, we concluded that regulation of piR-1919609 could effectively reverse the drug resistance of OC cells.

Despite piR-1919609 has significant influence in OC clinical prognosis progression,the results of the present study should be viewed cautiously because of the relatively limited sample size. In our future work, we intend to go futher study on proving the role of piR-1919609 in OC tumor progression.

In summary, this study provided the differential expression profiles of piRNAs in primary ovarian cancer and recurrent ovarian cancer by high-throughput sequencing for the first time and showed that piR-1919609 is an oncogenic piRNA. It can be used as a potential indicator for evaluating the clinical prognosis of OC. The results of this study are meaningful and provide a new direction for the early diagnosis and treatment of OC. This study also proved that piR-1919609 could effectively reverse drug resistance, which provided a research model and experimental basis for further study of the drug resistance mechanism of piR-1919609 and treatment in OC.

Abbreviations

Primary ovarian cancer

refers to ovarian cancer patients who have undergone primary surgery without chemotherapy or radiotherapy;Recurrent ovarian cancer refers to patients who have achieved clinical remission after satisfactory cytoreductive surgery and sufficient regular clinical chemotherapy but have clinical tumor recurrence within six months or six months later and undergo reoperation;Relapsed platinum sensitivity refers to the response to platinum-based drug treatment, clinical remission is achieved, and the lesion recurs after stopping chemotherapy for more than 6 months;Recurrent platinum resistance refers to patients who responded to initial chemotherapy but relapsed within 6 months after completing chemotherapy.

OC

ovarian cancer

PIWIL1

piwi-like RNA-mediated gene silencing 1

PIWIL2

piwi-like RNA-mediated gene silencing 2

PIWIL4

piwi-like RNA-mediated gene silencing 4

DDP

diamminedichloroplatinum

IC50

50% inhibitory concentration/half-maximal inhibitory concentration