Jian-Pi-Yi-Shen decoction inhibits mitochondria-dependent granulosa cell apoptosis in a rat model of POF

Chemical reagents

JPYS decoction (Beijing Kang-ren-tang Pharmaceutical Co., Ltd in China) were dissolved in deionized water (60° C) to acquire stock solution (1.1 g/mL). The stock solution was conserved in 4° C refrigerator. CTX, triptorelin, and selonsertib were obtained from Solarbio Life Sciences Co., Ltd. (China).

Chemical component analysis of JPYS decoction

JPYS contains eight different medicinal herbs (Table 1). HPLC combined with UPLC-QQQ-MS were used to detect main chemical components of JPYS in previous research, 71 compounds were tentatively identified in JPYS and then using HPLC–QQQ–MS/MS to further qualitative and quantitative 12 compounds [8]. 5 μL injection quantity, 20° C, 0.3 ml/min flow rate) (HPLC) and authenticated with standards. The process was operated in both positive and negative ionization forms (Scan: 100-1500 Da; Fragment: 80-185 V; eV: 4-80 eV). The data were analyzed via SCIEX OS software. Major components of JPYS were validated by real standards. Forms of mass spectrometry and chromatographic are presented in Figure 1.

Figure 1. The extracted ion chromatograms of JPYS decoction. The extracted ion chromatograms of mixed standards (A); The extracted ion chromatograms of JPYS sample: 1.Calycosin-7-O-β-D-glucoside (CAS: 20633-67-4, MRM: 447.01/285.1, retention time: 4.343 min); 2. liquiritin (CAS: 551-15-5, MRM: 419.01/257.2, retention time: 4.463 min); 3. nobiletin CAS: 478-01-3, MRM: 403.01/373.2, retention time: 7.615 min); 4.hesperetin (CAS: 69097-99-0, MRM: 303.01/177.1, retention time: 6.591 min); 5. Tanshinone IIA (CAS: 568-72-9, MRM: 295.01/189.2, retention time: 10.986 min); 6.calycosin (CAS: 20575-57-9, MRM: 285.01/270.1, retention time: 5.767 min); 7. formononetin (CAS: 485-72-3, MRM: 269.01/197.1, retention time: 7.042 min); 8. atractylenolide I (CAS: 73069-13-3, MRM: 231.01/77.2, retention time: 10.185 min); 9. stachydrine (CAS: 471-87-4, MRM: 144.01/58.3, retention time: 1.000 min); 10. betaine (CAS: 478-01-3, MRM: 118.01/58.3, retention time: 1.092 min); 11. dioscin (CAS: 19057-60-4, MRM: 867.99/867.5, retention time: 8.656 min); 12. astragaloside IV (CAS: 96574-01-5, MRM: 783.99/783.5, retention time: 6.442 min); 13. acteoside (CAS: 61276-17-3, MRM: 622.99/161.1, retention time: 4.395 min); 14. rosmarinic acid (CAS: 20283-92-5, MRM: 358.99/161.1, retention time: 4.989 min); 15. rhein (CAS: 478-43-3, MRM: 282.99/239.1, retention time: 7.507 min) (B).

Experimental animals

SD rats (body weight, 180–220 g; age, 8 weeks old, 40 female rats and 16 male rats) were used. Rats were housed in cages in room maintained humidity of 45-65 % and a temperature of 20° C, 12-h light/12-h dark (dark at 18:00 h), and administered a pellets and water freely.

40 female rats were randomly divided into 4 groups, namely, control group, POF group, JPYS treatment group, and triptorelin treatment group. For triptorelin group, rats were intramuscularly injected with triptorelin (1.5 mg/kg body weight) on day 1. Triptorelin was dissolved into saline to obtain a 0.75 mg/mL stock solution of. All rats in the other groups were received 0.5 mL saline. Dissolving CTX into saline to obtain stock solutions with 0.5/5 mg/mL. On day 11, all rats, except control group rats, were intraperitoneally injected with CTX (50 mg/kg). For days 12 to 25, all rats, except control group rats, were intraperitoneally injected with CTX (50 mg/kg body weight), whereas control group rats were received saline. With regard to concentration conversion between humans and rats, in terms of the surface area, it was approximately 6.17, and we calculated the dose for intragastric administration. For days 26 to 85, all rats in the JPYS treatment group were intragastrically treated with JPYS (11.0 g/kg body weight), whereas other groups rats were received saline. The treatment protocol is shown in Table 2.

To examine the estrous cycle, vaginal smears were obtained from rats in the morning hours (estrous interval period: mostly white blood cells; preestrus period: mostly nuclear epithelial cells; estrous cycle: mostly keratinized epithelial cells; and late estrous cycle: keratinized epithelial cells and white blood cells) (Figure 2A). Approximately 1–5 days following the last irrigation, collecting abdominal aorta blood from all groups rats during the non-estrus period with anesthesia (isoflurane via inhalation anesthesia), followed by oophorectomy. Rats were euthanized. Blood was let to coagulate (1 h) in room temperature, and then serum was acquired through centrifugation. Samples were conserved at -80° C. Ovarian tissues were placed in paraformaldehyde (4%), and some ovarian tissues were frozen by liquid nitrogen and then conserved at -80° C for molecular/protein studies. Lastly, four female rats from each group were mated with male rats with 1:1 ratio for 12 h. A vaginal plug was an indicator of successful pregnancy, and this was considered day 0.5 of gestation. Pregnant rats were euthanized on day 15.5 of gestation.

Figure 2. JPYS improved the abnormal estrous cycle and rate of pregnancy in premature ovarian failure (POF) rats. Rats were treated with JPYS (11.0 g/kg.d) and pre-treated with triptorelin (1.5 mg/kg) followed by intraperitoneally injected cyclophosphamide (50 mg/kg). (A) Estrous cycle of the rats (Estrous interval period: Vagina smear with white blood cells mainly; Preestrus period: With nuclear epithelial cells mainly; Estrous period: With Keratinized epithelial cells mainly; Late estrous period: See Keratinized epithelial cells and white blood cells) (n=6); (B) The rate of pregnancy (Female rats of each groups were sent to mate with male rats with ratio of 1:1 for 12 h. The mixture of sperm and vaginal smears seen on the next morning indicated the success of pregnancy, and this was considered as the 0.5 th day of the gestation. The pregnant rats were euthanized on the 13.5 th day of the gestation) (n=4); (C) The rate of abnormal estrous cycle (n=6). Data are shown as mean ± SD. *p < 0.05 versus control group, ^#p < 0.05 versus POF group, ^△p < 0.05 versus JPYS group.

Diameter of bilateral ovaries and ovarian index

After anesthesia, the rats were euthanized, and bilateral ovaries were harvested and placed in ice-cold buffer. Fatty tissue surrounding the ovaries was removed, and ovaries were measured the diameter and weighed. The ovarian index was calculated as follows: wet weight of bilateral ovaries (mg)/body weight (g) × 100 % [9].

Histological assessment of ovarian tissues by HE

Ovarian tissues were fasten by paraformaldehyde (4 %) for 24 h, dehydrated in a graded ethanol series, and then embedded in paraffin. Tissue blocks were cross-sectioned to generate 5-μm-thick sections, and then sections were stained by HE [10]. Sections were observed by Olympus CX33 light microscope and follicle development (primary, secondary, and atretic follicles), as well as the integrity of the corpora lutea were assessed.

Measurement of serum FSH and E₂ levels

Arterial blood samples were collected to and levels of FSH and E₂ were detected using ELISA (Cusabio Biotech Co., Ltd.).

Morphological evaluation of ovarian tissues by electron microscopy

Ovarian tissues were harvested immediately, cut into smaller pieces of 1 mm³, fixed in 4% glutaraldehyde at 4° C for approximately 2 h, and washed with 0.1 M sodium dimethyl arsenate buffer three times. Specimens were placed in osmium tetroxide (1 %) at 4° C for 90 min and washed with distilled water three times. Specimens were dehydrated, and then dehydrated in 100% propionaldehyde two times. The dewatering time of each concentration gradient was 10–15 min. Specimens were embedded in a graded epoxy series (1:3, 1:1, and 3:1) and polymerized at 35° C for 24 h, 45° C for 24 h, and 60° C for 24 h. Tissue blocks were sectioned to generate 70–90 nm-thick sections, which were double stained with lead citrate/uranyl acetate, and then observed with a Hitachi H-7650 electron microscope.

TUNEL assay

TUNEL was applied to assess cell apoptosis. Five-micron-thick cross-sections were deparaffinized, rehydrated, and treated with protease K (10 μg/mL) for 15 min. Cross-sections were incubated in TUNEL reaction mixture in 37° C for 60 min, then followed by washing. Nuclei were stained with 0.1 μg/mL DAPI, and cross-sections were mounted. Cross-sections were examined under a Canon fluorescent microscope. Eight random fields per sample were examined by investigators, and then number of TUNEL-positive cells was determined.

Measurement of caspase-3/9 activity

Caspase-3/9 activity of ovarian tissues was measured using related kit. A piece of ovarian tissue (10 mg) was put into the reaction buffer, and then incubated (37° C) for 2 h. A fluorimeter was used to quantify the release of the catalyzed enzyme at an absorbance of 405 nm.

Preparation of mitochondrial suspension and detection of mitochondrial function in vivo

After anesthesia, the rats were killed, and ovaries were harvested, and then placed in PBS. After trimming, approximately 50 mg of ovarian tissues were homogenized in isolation buffer, following centrifugation with 700 × g (10 min). The supernate was gathered, and then centrifuged with 7000 × g (10 min). Discarding the supernate, using isolation buffer to wash mitochondrial pellet, and then sample was centrifuged with 7000 × g (10 min) two times. The purified mitochondrial pellet was resuspended to obtain a protein solution (5 mg/mL). The protein concentration was in the range of 100 μg/mL–1000 μg/mL, then mitochondrial suspension was emploied to measure MMP [11], opening of mPTP [12], ROS production, injuried mtDNA [13], mitochondrial oxygen consumption rate [14], RCR [14], mitochondrial respiratory chain complex enzymes [15], and ATP [15].

RNA extraction, cDNA synthesis, and real-time PCR

Total RNA was separated with Trizol, then the RNA integrity was assessed by spectrophotometry at a wavelength of 260 nm. For quantitative polymerase chain reactions (PCR), and the reaction volume was 40 μL. Cycling conditions were performed. For semi-quantitative PCR [16], and the reaction system was 20 μL. Cycling conditions were according to following: 95° C for 10 min and 95° C for 10 s, then 40 cycles of 60° C for 15 s, 72° C for 20 s, and 72° C for 10 min. The GAPDH mRNA level was emploied for target gene normalization [17]. Primer sequences are shown in Table 3.

Total protein isolation and Western blotting

Total proteins were extracted, and then protein level was detected with BCA. Equivalent quantity of total protein were subjected to 8–12%. Following blocking membranes with skim milk, they were incubated separately overnight (Table 4). Immunoreactive proteins were observed with chemiluminescence kit and analyzed with software.

Preparation and compound analysis of JPYS-containing serum

Female SD rats were intragastrically treated with saline or JPYS (11.0 g/kg/d) for 7 days. 2 hours after the last treatment, then the blood was extracted from the aorta ventralis, reserved in 4° C (1 h), and then centrifuged (2000 rpm/min) for 30 min. Serum samples were inactivated in a water bath (56° C) for 30 min.

Cell culture and establishment of POF model

The KGN human granulosa cell line was purchased from Fenghui Bio-Company (cat. no. CL0544, Changsha, China). Cells (2 × 10⁶ cells) were plated in 100-mm culture dishes and cultured in medium with incubator at 37° C with humidified air containing 5% CO₂. Medium was replaced every other day. Cells were treated with CTX with different concentrations (20, 40, and 60 μg/mL) and/or JPYS-containing serum at different concentrations (2, 4, and 8 %).

Assignment of cell groups

Cells were divided into 4 groups including control group, CTX (20, 40, and 60 μg/mL) -induced POF model groups, POF + JPYS (JPYS-containing serum, 2, 4, and 8 %) treatment groups, and POF + GS-49977 (5 μM) treatment group and used for in vitro experiments. Then we detect the cell viability through CCK-8.

Hoechst staining

Cells were fixed in paraformaldehyde (4 %) for 10 min, followed by washing with PBS and stained by Hoechst 33258 for 5 min. Images were observed with fuorescent microscope with an 350 nm excitation wavelength and an 460 nm emission wavelength.

Assessment of apoptosis with flow cytometry

Cells (1 × 10⁵ cells/mL) were seeded and incubated overnight, and then cells were intervened with JPYS (2, 4, and 8 %) and GS-49977 (5 μM) for 24 h, followed by CTX with different levels (20, 40, and 60 μg/mL) for 24 h. Cells were analyzed via flow cytometry.

Detection of release of ROS by flow cytometry

Using flow cytometry was emploied to measure ROS levels of different groups. Cells in every group were intervened, collected and suspended by DCFH-DA (10 μmol/L) with a final cell (1 × 10⁶ cells/mL), then cells were incubated for 30 min in 37° C with CO₂ (5 %), and blended every 5 min. The flow cytometer was used to measure the fluorescence intensity.

Electron microscopy

Cells were fixed with glutaraldehyde (4 %) in 4° C (2 h), and then washed with 0.1 M sodium dimethyl arsenate, and then centrifuged between the washing steps. The remaining steps was similar to in vivo study.

Preparation of mitochondrial suspension and assessment of mitochondrial function in vitro

Mitochondria were seperated via related kit and reserved on ice for using. Mitochondria were resuspended in buffer. KGN cells (1 × 10⁵ cells/mL) were plated in 6-well plates, and then centrifuged (600 × g) for 3–4 min. Throwing away supernate. Finally, cells were stained by JC-1 staining buffer, and then supernate was throwed away. Detecting MMP and mPTP [18, 11].

Statistical analysis

Statistical analysis was carried out by SPSS 17.0 Software. ANOVA was emploied to compare 4/5 independent groups. Two-to-two comparison among groups was emploied to analyze variance. The LSD-t test was emploied to compare multiple comparisons between 4/5 groups. We defineded P < 0.05 represent having statistically significance.

Data availability

The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.

Chemical components of JPYS decoction

The chemical components of JPYS were measured by HPLC-Q-TOF-MS/MS. The 15 chemical components including: Calycosin-7-O-β-D-glucoside (contents: 12.58 μg/g; CAS: 20633-67-4), liquiritin (contents: 64.56 μg/g; CAS: 551-15-5), nobiletin (contents: 1.41 μg/g; CAS: 478-01-3), hesperetin (contents: 1.04 μg/g; CAS: 69097-99-0), Tanshinone IIA (contents: 18.40 μg/g; CAS: 568-72-9), calycosin (contents: 6.64 μg/g; CAS: 20575-57-9), formononetin (contents: 7.26 μg/g; CAS: 485-72-3), atractylenolide I (contents: 0.40 μg/g; CAS: 73069-13-3), stachydrine (contents: 0.90 μg/g; CAS: 471-87-4), betaine (contents: 17.79 μg/g; CAS: 478-01-3), dioscin (contents: 9.82 μg/g; CAS: 19057-60-4), astragaloside IV (contents: 28.88 μg/g; CAS: 96574-01-5), acteoside (contents: 24.58 μg/g; CAS: 61276-17-3), rosmarinic acid (contents: 28.86 μg/g; CAS: 20283-92-5), and rhein (contents: 6.17 μg/g; CAS: 478-43-3). The extracted-ion chromatograms are described in Figure 1A, and ESI-Q-TOF MS/MS spectra of the five components are described in Figure 1B. 2D structure of chemical components were seeked in https://pubchem.ncbi.nlm.nih.gov/.

JPYS improved the estrous cycle length, pregnancy rate, and abnormal estrous cycle rate in POF rats

To evaluate degree of ovarian injury and protective effects of JPYS, we examined the estrous cycle length (Figure 2A), rate of pregnancy (Figure 2B), and abnormal estrous cycle rate (Figure 2C). The estrous cycle length increased in POF group (p < 0.05) but reduced by treatment with JPYS and pretreatment with triptorelin (p < 0.05) (Figure 2A). The pregnancy rate decreased in POF group (p < 0.05) but increased after JPYS treatment and triptorelin pretreatment (p < 0.05) (Figure 2B). The abnormal estrous cycle rate increased in the POF group (p < 0.05) but decreased after JPYS treatment and triptorelin pretreatment (p < 0.05) (Figure 2C).

JPYS improved the ovarian index, diameter of bilateral ovaries, ovarian function, and follicle development in POF rats

To further evaluate degree of ovarian injury and protective effects of JPYS, we examined the diameter of bilateral ovaries and ovarian index (Figure 3A), ovarian function (Figure 3B), and follicle development (Figure 3C). The diameter of bilateral ovaries and ovarian index decreased in the POF group (p < 0.05) but increased by treatment with JPYS and pretreatment with triptorelin (p < 0.05) (Figure 3A). Ovarian function decreased (FSH increased; E₂ decreased) in the POF group (p < 0.05) but increased (FSH decreased; E₂ increased) after JPYS treatment and triptorelin pretreatment (p < 0.05) (Figure 3B). The number of primary and secondary follicles, as well as the surface area of the corpora lutea, decreased in POF group (p < 0.05) but increased after JPYS treatment and triptorelin pretreatment (p < 0.05). The number of atretic follicles increased in POF group (p < 0.05) but reduced following JPYS treatment and triptorelin pretreatment (p < 0.05) (Figure 3C).

Figure 3. JPYS improved the ovarian function in premature ovarian failure (POF) rats. Rats were treated with JPYS (11.0 g/kg.d) and pre-treated with triptorelin (1.5 mg/kg) followed by intraperitoneally injected cyclophosphamide (50 mg/kg). (A) Diameter of bilateral ovaries and Ovarian index (Ovarian index= wet weight of bilateral ovaries (mg)/ body weight (g)×100 %, the size of the ovary in POF group were significantly reduced compared with control group), the scale bars represents a length of 1 cm on histology; (B) Ovarian function (follicle stimulating hormone (FSH) and oestradiol (E2)); (C) Histological assessment of the ovarian tissue using hematoxylin-eosin (HE) staining (All phases of follicles (primary follicles, secondary follicles, and atretic follicles) and corpora lutea were counted), the scale bars represents a length of 20 μm on histology. Data are shown as mean ± SD. *p < 0.05 versus control group, ^#p < 0.05 versus POF group, ^△p < 0.05 versus JPYS group. (n=6).

JPYS alleviated mitochondrial damage in POF rats

To assess mitochondrial function and examine the protective effects of JPYS, electron microscopy was used (Figure 4). Compared with the ovarian tissues of the control group, which showed normal mitochondria, the results of electron microscopy revealed that ovarian tissues of the POF group showed abnormal mitochondria, with membrane swelling and rupture. The percentage of damaged mitochondria in the POF group was higher than that of mitochondria in the control group (p < 0.05). JPYS treatment and triptorelin pretreatment decreased the percentage of damaged mitochondria (Figure 4A). Next, we measured MMP, mPTP opening (%), ROS production, damaged mtDNA, the mitochondrial oxygen consumption rate, RCR, ATP, and mitochondrial respiratory chain complex enzyme in isolated mitochondria. Compared to the control group, ROS production and mPTP opening (%) increased in the POF group (p < 0.05). However, ROS production decreased by JPYS treatment and triptorelin pretreatment (p < 0.05). Compared to the control group, the mitochondrial oxygen consumption rate, RCR, and the MMP decreased in the POF group (p < 0.05). However, JPYS treatment and triptorelin pretreatment increased these indices (p < 0.05). Real-time qPCR was used to measure the degree of damaged mtDNA. Compared to the control group, ratio of long/short fragments decreased in the POF group (p < 0.05), whereas JPYS treatment and triptorelin pretreatment increased the ratio (p < 0.05). Furthermore, the activity of mitochondrial respiratory chain complex enzymes and ATP level were decreased in POF group compared with control group (p < 0.05), whereas JPYS treatment and triptorelin pretreatment increased these indices (p < 0.05) (Figure 4B).

Figure 4. JPYS improved mitochondrial function in premature ovarian failure (POF) rats. Rats were treated with JPYS (11.0 g/kg.d) and pre-treated with triptorelin (1.5 mg/kg) followed by intraperitoneally injected cyclophosphamide (50 mg/kg). (A) Electron microscope pictures (10,000×; 20,000×; 40 000×) of ovary in POF rats, the scale bars represents a length of 2 μm, 1 μm, and 500 nm on histology respectively. Abnormal mitochondrial (paired yellow arrow) morphology show that mitochondrial membrane rupture or swellings, normal mitochondrial (single yellow arrow) morphology type show that mitochondrial membrane smooth and inner carinulae distinct and percentage of damaged mitochondria; (B) The MMP (ratio of red/green), the opening of mPTP (%), the mitochondrial ROS, the mtDNA damage (ratio of long/short fragments), the mitochondrial RCR, mitochondrial oxygen consumption rate, the mitochondrial respiratory chain complex enzymes (I, II, III, IV, and V), and ATP were recorded above. Data are shown as mean ± SD. *p < 0.05 versus control group, ^#p < 0.05 versus POF group, ^△p < 0.05 versus JPYS group. (n=6).

JPYS affected mitochondrial biogenesis and ovarian function in POF rats

Real-time PCR and Western blotting were used to examine different aspects of mitochondrial function. OPA1, Mfn1, and Mfn2 were selected as markers of mitochondrial biogenesis, PGC-1α was selected as a marker of mitochondrial fusion, and Drp1 and Fis1 were selected as markers of mitochondrial fission. Compared to control group, OPA1, Mfn1, and Mfn2 levels decreased in POF group (p < 0.05), whereas JPYS treatment and triptorelin pretreatment increased their mRNA/protein levels (p < 0.05). Compared to control group, Drp1 and Fis1 levels increased in POF group (p < 0.05), whereas JPYS treatment and triptorelin pretreatment decreased mRNA/protein levels (p < 0.05). Compared to the control, PGC-1α levels decreased in POF group (p < 0.05), whereas JPYS treatment and triptorelin pretreatment their mRNA/protein levels (p < 0.05) (Figure 5).

Figure 5. JPYS improved mitochondrial biogenesis and dynamics in premature ovarian failure (POF) rats. Rats were treated with JPYS (11.0 g/kg.d) and pre-treated with triptorelin (1.5 mg/kg) followed by intraperitoneally injected cyclophosphamide (50 mg/kg). We used real-time qPCR and western blot to detect mitochondrial biogenesis and dynamics. We chose OPA1, Mfn1, and Mfn2 to represent mitochondrial biogenesis function, and PGC-1α to represent the dynamic mitochondrial fusion, and Drp1 and Fis1 to represent mitochondrial fission. The expression of OPA1, Mfn1, Mfn2, PGC-1α, Drp1, and Fis1 in mRNA (A) and protein (B) levels. Data are shown as mean ± SD. *p < 0.05 versus control group, ^#p < 0.05 versus POF group, ^△p < 0.05 versus JPYS group. (n=6).

JPYS modulated the ASK1/JNK pathway in vivo

Real-time qPCR and western blotting were used to examine the levels of target genes/proteins within the ASK1/JNK pathway. Compared to the control group, caspase-3/9, Bax, ASK1, JNK, and Cyt-c expression increased (p < 0.05) in the POF group. However, the mRNA level of these genes decreased after JPYS treatment and triptorelin pretreatment (p < 0.05). Compared to the control, Bcl-2 expression decreased in the POF group (p < 0.05). However, the mRNA level of Bcl-2 increased after JPYS treatment and triptorelin pretreatment (p < 0.05). JPYS treatment and triptorelin pretreatment inhibited ASK1 and JNK phosphorylation (decreased p-ASK1/ASK1 and p-JNK/JNK) and caspase-3/9 activation (decreased cleaved/caspase-3/9) (Figure 6B, 6C).

Figure 6. JPYS improved mitochondrial function and inhibited apoptosis via inhibiting ASK1/JNK pathway. Rats were treated with JPYS (11.0 g/kg.d) and pre-treated with triptorelin (1.5 mg/kg) followed by intraperitoneally injected cyclophosphamide (50 mg/kg). Graphical abstract (A): Mitochondrial dysfunction of the ovarian GC occurs following POF and can lead to apoptosis. JPYS can relieve GC apoptosis by improving mitochondrial function via inhibiting the ASK1/JNK pathway in vivo and vitro. We used real-time qPCR and western blot to detect the target genes of ASK1/JNK pathway in mRNA (B) and protein (C) levels. TUNEL positive cells (D), the scale bars represents a length of 50 μm on histology. The activity of caspase-9 (E). The activity of caspase-3 (F). Data are shown as mean ± SD. *p < 0.05 versus control group, ^#p < 0.05 versus POF group, ^△p < 0.05 versus JPYS group. (n=6).

JPYS reduced ovarian cell apoptosis in POF rats

The TUNEL assay was emploied to evaluate protective effects of JPYS on ovarian cell apoptosis in POF rats (Figure 6D). The number of apoptotic ovarian cells in the POF group increased compared with control group (p < 0.05). However, treatment with JPYS and pretreatment with triptorelin decreased the number of apoptotic ovarian cells (Figure 6D). The activity of caspase-3/9 in ovarian tissues in the POF group increased compared with control group (p < 0.05) but decreased after JPYS treatment and triptorelin pretreatment (Figure 6E, 6F).

Compound analysis of JPYS-containing serum

The results of compound analysis were as follows: rhein (0.127 μg/mL), salvianolic acid A (1.868 μg/mL), liquiritin (0.068 μg/mL), acteoside (0.000 μg/mL, calycosin-7-O-β-D-glucoside (0.134 μg/mL), rosmarinic acid (0.812 μg/mL), formononetin (0.010 μg/mL), calycosin (0.049 μg/mL), astragaloside IV (24.989 μg/mL), atractylenolide I (0.016 μg/mL), dioscin (1.873 μg/mL), tanshinone IIA (0.043 μg/mL), narirutin (5.145 μg/mL), nobiletin (0.064 μg/mL), hesperetin (0.467 μg/mL), stachydrine (0.015 μg/mL), and betaine (5.508 μg/mL) (Figure 7). 2D structure of chemical components were seeked in https://pubchem.ncbi.nlm.nih.gov/.

Figure 7. Chemical component analysis of the JPYS-containing serum. Preparation of JPYS-Containing Serum. Female SD rats were intragastrically administrated with normal saline (control) or JPYS (11.0 g/kg/d) once daily for seven consecutive days. Then, 17 compounds (Rhein, salvianolic acid A, liquiritin, acteoside, calycosin-7-O-β-D-glucoside, rosmarinic acid, formononetin, calycosin, astragaloside IV, atractylenolide I, dioscin, tanshinone IIA, narirutin, nobiletin, hesperetin, stachydrine, betaine) were further quantified simultaneously by HPLC-QQQ-MS/MS.

JPYS alleviated CTX-induced apoptosis in vitro

To assess effects of JPYS-containing serum and CTX on cell function, a cell viability assay was emploied. Compared to control group, KGN cell viability decreased through intrevening with CTX (20, 40, and 60 μg/mL) (p < 0.05). The results of light microscopy revealed normal cells with long fusiform shapes in control group. Nevertheless, CTX treatment altered cell shape by shrinking cells and disrupting cell–cell interactions. Cell viability was not affected by JPYS-containing serum (2, 4, 8 %) but it was affected by pretreatment with CTX (20 μg/mL) followed by treatment with JPYS (2, 4, 8 %). At the two lowest concentrations but not at the highest concentration, JPYS-containing serum attenuated cell viability dose-dependently (p < 0.05) (Figure 8A–8C).

Figure 8. JPYS improved decreased cell viability, increased cell apoptosis caused by cyclophosphamide (CTX) in vitro. JPYS-containing serum inhibited loss of cells viability induced by CTX. (A) Cells were treated with CTX (20, 40, and 60 μg/ml) for 24 h; (B) Cells were treated with JPYS-containing serum (2, 4, 8 %); (C) Cells were treated with CTX (20 μg/ml) for 24 h, then treated with JPYS-containing serum (2, 4, 8 %); (D) Hoechst 33258 staining was used to detect the apoptosis and counted the percentage of apoptotic cells; (E) Cell apoptosis was measured by fow cytometry and counted the percentage of apoptotic cells: provided 2-dimensional graphical representations of PI/annexin V-FITC plots. ‘Early apoptosis’ was defifined as cells positive for annexin V-FITC only. ‘Late apoptosis’ was defifined as cells positive for annexin V-FITC and PI. ‘Necrosis’ was defifined as cells positive for PI only; (F) Cell apoptosis was observed by electron microscope pictures (10,000×), the scale bars represents a length of 2 μm on cells respectively. Data are shown as mean ± SD. *p < 0.05 versus control group, ^#p < 0.05 versus CTX group, ^△p < 0.05 versus CTX+ JPYS (2 %) group, ^▲p < 0.05 versus CTX+ JPYS (4 %) group. (n=3).

Compared to control group, CTX (20 μg/mL) induced apoptosis. At the two lowest concentrations but not at highest concentration, JPYS-containing serum attenuated the apoptosis dose-dependently (p < 0.05) (Figure 8D).

Compared to control group, the apoptotic rate increased in the POF group (p < 0.05). The apoptotic rate was lower in cells pretreated with JPYS-containing serum at concentrations of 2 and 4 % than that in untreated cells (p < 0.05), and the effects were dose-dependent (p < 0.05). However, JPYS-containing serum at its highest concentration did not protect cells from apoptosis (Figure 8E). The results of electron microscopy revealed that untreated cells were healthy, whereas CTX caused apoptosis, and JPYS-containing serum (2 and 4 %) can inhibit apoptosis induced by CTX (Figure 8F).

JPYS improved CTX-induced mitochondrial dysfunction in vitro

The percentage of damaged mitochondria was higher in the CTX group than that in the control group (p < 0.05). The percentage of damaged mitochondria decreased after treatment with JPYS-containing serum at concentrations of 2 or 4 % but not at a concentration of 8 % (p < 0.05) (Figure 9A).

Figure 9. JPYS improved mitochondrial dysfunction caused by CTX in vitro. Cells were treated with CTX (20 μg/ml) for 24 h, then treated with JPYS-containing serum (2, 4, 8 %). (A) Electron microscope pictures (10,000×; 20,000×; 40,000×) of ovary in POF rats, the scale bars represents a length of 2 μm, 1 μm, and 500 nm on cells respectively. Abnormal mitochondrial (paired yellow arrow) morphology show that mitochondrial membrane rupture or swellings, normal mitochondrial (single yellow arrow) morphology type show that mitochondrial membrane smooth and inner carinulae distinct and percentage of damaged mitochondria; (B) The MMP (ratio of red/green); (C) The opening of the mPTP (%); (D) The ROS levels. Data are shown as mean ± SD. *p < 0.05 versus control group, ^#p < 0.05 versus CTX group, ^△p < 0.05 versus CTX+ JPYS (2 %) group, ^▲p < 0.05 versus CTX+ JPYS (4 %) group. (n=3).

Compared to control group, the MMP decreased in CTX group (p < 0.05). Cells pretreated with 2 or 4 %, but not 8 %, JPYS-containing serum had an increased MMP (p < 0.05) (Figure 9B). Compared to control group, the mPTP (%) increased in CTX group (p < 0.05). Cells pretreated with 2 or 4 %, but not 8 % JPYS-containing serum had a decreased mPTP (%) (p < 0.05) (Figure 9C). Compared to the control group, the ROS levels increased in the CTX group (p < 0.05), cells pretreated with 2 or 4 %, but not 8 % JPYS-containing serum had an decreased ROS levels (p < 0.05) (Figure 9D). As such, 4 % JPYS-containing serum was selected for subsequent experiments.

JPYS inhibited ASK1/JNK-induced apoptosis and mitochondrial dysfunction in vitro

To research whether anti-apoptotic effects of JPYS-containing serum were associated with ASK1/JNK pathway, GS-49977 (ASK1 inhibitor) was emploied. Compared to control group, CTX increased ASK1 and JNK phosphorylation (p < 0.05), whereas JPYS and GS-49977 decreased ASK1 and JNK phosphorylation (p < 0.05). Compared to the control group, CTX up-regulated Bax and cleaved caspase-9 levels (p < 0.05), whereas JPYS and GS-49977 down-regulated Bax and cleaved caspase-9 levels (p < 0.05). Compared to the control group, CTX decreased the Bcl-2 level (p < 0.05), whereas PYSP and GS-49977 increased the Bcl-2 level (p < 0.05) (Figure 10A). Compared to the control group, the MMP level (p < 0.05) was decreased but the mPTP level (p < 0.05) was increased in POF group. Compared with CTX group, the MMP level (p < 0.05) was increased but mPTP was decreased after JPYS and GS-49977 treatment (p < 0.05) (Figure 10B, 10C). Compared to the control group, the ROS levels increased in the CTX group (p < 0.05), cells pretreated with JPYS and GS-49977 had an decreased ROS levels (p < 0.05) (Figure 10D).

Figure 10. JPYS inhibited ASK1/JNK-induced apoptosis and mitochondrial dysfunction in vitro. To investigate whether the anti-apoptotic efect of JPYS-containing serum (4 %) was associated with the ASK1/JNK pathway. GS-49977 (5 μM) was the inhibitor of ASK1. (A) We used western blot to detect the target genes of the ASK1/JNK pathway in protein level; (B) The MMP (ratio of red/green); (C) The opening of mPTP (%); (D) The ROS levels; (E) Hoechst 33258 staining was used to detect the apoptosis and counted the percentage of apoptotic cells. Data are shown as mean ± SD. *p < 0.05 versus control group, ^#p < 0.05 versus CTX group. (n=3).

Compared to control group, CTX induced apoptosis, whereas JPYS and GS-49977 inhibited apoptosis (p < 0.05) (Figure 10D).