Effects of gut microbiota on omega-3-mediated ovary and metabolic benefits in polycystic ovary syndrome mice

Female (3 weeks old) C57BL/6 mice were purchased from Jiangsu ALF Biotechnology Company and housed in specific pathogen-free (SPF) conditions with free access to regular rodent chow and water. The SPF animal facility, including germ-free isolators at Jinling Hospital, was maintained in a 12-hour light and dark cycle at 23 ± 2 °C and 50 ± 10% humidity. Use a microisolator (CBC filter bonneted) and individually ventilated cages (PIV/IVC). Sterilize or disinfect food, water, cages and anything that will come in contact with the mice. The cage and water were changed twice a week. All animal protocols were approved by the Animal Care and Use Committee of Jinling Hospital. All of the experiments were carried out in accordance with the approved guidelines by Jinling Hospital. They were randomly divided into different groups and housed in individually ventilated cages (3–5 mice in each cage).

The DHEA-induced PCOS mouse model was established based on a previous report [37]. Mice in the experimental groups were subcutaneously injected with DHEA dissolved in sesame oil (D4000; Sigma–Aldrich; 60 mg/kg body weight per day for 21 days, n = 6 each).

For the omega-3-treated mice, after 3 weeks of DHEA treatment, the mice received another 8 weeks of treatment by gavage with omega-3 PUFAs (2 g/kg every 2 days) for the omega-3 group (n = 6 each), while PCOS mice were treated with an equal volume of corn oil at 2 g/kg for comparison. The mice were anesthetized with ether and killed by cervical dislocation. Cervical dislocation after anesthesia is a humane method of euthanasia according to the Guide for the Care and Use of Laboratory Animals (Eighth Edition) made by the Institutional Animal Care & Use Committee.

Serum was collected to measure sex hormone levels at diestrus stages. The ovaries were processed for hematoxylin and eosin staining. The remaining ovaries were quickly frozen and stored at − 80 °C. For H&E staining, ovaries were fixed in 4% paraformaldehyde (PFA) for 24 h and embedded in paraffin. Section (5 μm) were prepared and dewaxed prior to straining. The sera were collected for hormone level analysis; ovarian tissues were collected for H&E staining and real-time quantitative PCR analysis.

Vaginal smears were taken daily at 9:00 a.m. from the 75th to the 84th day after the first day of treatment. The stage of the estrous cycle was determined by microscopic analysis of the predominant cell type in vaginal smears following H&E staining. Proestrus consists of round, nucleated epithelial cells; estrus consists of cornified squamous epithelial cells; metestrus consists of epithelial cells and leukocytes; and diestrus consists of nucleated epithelial cells and a predominance of leukocytes. The estrous cycle was determined by classic criteria [38].

Blood samples were collected in 1.5 mL EP tubes, blood clotting was performed for 1 h at 4 °C, and the tubes were centrifuged at 500 × g/4°C for 10 min to harvest serum. Serum samples and conditional cell culture media were stored at -80 °C for subsequent serum determinations. The levels of follicle-stimulating hormone (FSH) (CEA830Mu; Cloud-Clone), luteinizing hormone (LH) (CEA441Mu; Cloud-Clone), estradiol 2 (E2) (ml001962; mlbio) and testosterone (T) (ml001954; mlbio) were determined by enzyme-linked immunosorbent assay kits for mice.

Total RNA was extracted using a Total RNA Purification Kit (B0004-plus; EZBioscience), and cDNA was reverse-transcribed from 1 µg total RNA using HiScript III RT SuperMix (R312-01; Vazyme) according to the manufacturer’s protocol. RT–qPCR was performed in triplicate in 96-well plates using SYBR qPCR Master Mix (Q111-02; Vazyme) with gene-specific forward and reverse primers. cDNA was amplified for 40 cycles using a Roche LC480 Real-time PCR system. β-actin was used as a reference gene. The qPCR primers used in this study can be found in Table 1.

Microbial DNA was extracted using HiPure StoolDNA Kits (Magen, Guangzhou, China) according to the manufacturer’s protocols. High-throughput sequencing was used to research the V3 − V4 region of 16S rDNA with a sequencing depth of 60,000 in the control, PCOS and omega-3 groups (n = 3 each). The 16S rDNA V3-V4 region of the ribosomal RNA gene was amplified by PCR (94 °C for 2 min, followed by 30 cycles at 98 °C for 10 s, 62 °C for 30 s, and 68 °C for 30 s and a final extension at 68 °C for 5 min) using primers 341 F: CCTACGGGNGGCWGCAG; 806R: GGACTACHVGGGTATCTAAT. PCRs were performed in triplicate in a 50 µL mixture containing 5 µL of 10. KOD buffer, 5 µL of 2 mM dNTPs, 3 µL of 25 mM MgSO4, 1.5 µL of each primer (10 µM), 1 µL of KOD polymerase, and 100 ng of template DNA. Noisy sequences of raw tags were filtered by the QIIME (version 1.9.1) pipeline under specific filtering conditions to obtain high-quality clean tags. Clean tags were searched against the reference database (version r20110519) to perform reference-based chimera checking using the UCHIME algorithm. All chimeric tags were removed, and finally, the obtained effective tags were used for further analysis. The effective tags were clustered into operational taxonomic units (OTUs) of ≥ 97% similarity using the UPARSE (version 9.2.64) pipeline. Taxonomic classification was conducted by BLAST (version 2.6.0), searching the representative OTU sequences against the NCBI 16S ribosomal RNA Database (Bacteria and archaea) (http://www.ncbi.nlm.nih.gov↗) (version 202,101) using the best hit with strict criteria (E value < e-5, query coverage ≥ 60% and the following identity thresholds: a hit with sequence identity ≥ 92% was considered to belong to the same species; with sequence identity ≥ 88% as indicators of belonging to the same genus; with sequence identity ≥ 85% as indicators of the same family; with sequence identity ≥ 80% as indicators of the same order; the classes were inferred when sequence identity ≥ 75%; the phylum were inferred when sequence identity ≥ 70%). The tag sequence with the highest abundance was selected as a representative sequence within each cluster. The representative sequences were classified into organisms by a naive Bayesian model using the RDP classifier (version 2.2) with confidence threshold values ranging from 0.8 to 1.

We evaluated the adequacy of sequencing by plotting the rarefaction curve for each index and indirectly reflected the richness of species in the sample. The rarefaction curve calculates the expected value of each index by sampling n tags several times and then makes a curve according to a set of n values (i.e., the number of samplings, which is generally a group of arithmetic sequences less than the total number of sequences) and the corresponding expected value of the index. When the curve flattens or reaches a plateau, it can be considered that the increase in sequencing depth has no effect on species diversity. This indicated that the amount of sequencing was sufficient. The OTU rarefaction curve was plotted in the R project ggplot2 package (version 2.2.1).

Sequences were rarefied prior to calculation of alpha and beta diversity statistics. Alpha diversity indexes were calculated in QIIME (version 1.9.1) from rarefied samples using the Shannon index for diversity and the Chao1 index for richness. Beta diversity was calculated using the unweighted pair group method using the arithmetic average (UPGMA) clustering method and principal component analysis (PCA). The total OTUs are shown in Supplementary Table. 1

The mice were treated with antibiotics for 7 d before transplantation experiments. Six-week-old PCOS C57BL/6 mice were supplemented with either omega-3 PUFAs or an equal volume of corn oil at 2 g/kg in the presence of the antibiotic vancomycin (0.5 g/liter), neomycin sulfate (1 g/liter), metronidazole (1 g/liter), and ampicillin (1 g/liter) [39].

Briefly, 6-week-old female donor PCOS mice were treated with omega-3 PUFAs or an equal volume of corn oil (2 g/kg). The recipient mice were treated with fresh transplant material from either PCOS mice or omega-3 PUFA-treated mice. Stool was collected daily from donor mice and pooled. Donor stool (100 mg) was diluted with saline, homogenized for 1 min using a vortex to achieve a liquid slurry, and then centrifuged at 500 × g for 3 min to remove particulate matter to facilitate administration. A total of 100 µl of the fecal suspension from either PCOS mice or omega-3 PUFA-treated mice was administered to mice by oral gavage. Fresh transplant material was prepared within 10 min before oral gavage to prevent changes in bacterial composition. Oral gavage with fecal transplant material was conducted every 2 days throughout the 8-week experiment.

Statistical analysis was performed using GraphPad Prism version 8 for Windows. The normality of the data was analyzed first, and then statistical tests were performed. Student’s t test, the Mann–Whitney test, and one-way analysis of variance (ANOVA) followed by Tukey’s post hoc test were used. All other data are presented as the mean ± standard error of the mean. The statistical significance threshold: alpha = 0.05 for hypothesis testing, as p value < alpha, we conclude that the null hypothesis should be rejected. And p values are denoted in figures and indicate the following: *P < 0.05; **P < 0.01; ***P < 0.001, ****P < 0.0001.

Because PCOS is closely associated with endocrine and metabolic disorders, to investigate the effect of omega-3 PUFAs on metabolism and endocrine function, we established a PCOS model (Fig. 1A) and assessed serum hormone levels, glucose homeostasis and insulin sensitivity in diverse groups (Fig. 1). During the 8-week treatment, the body weight of the PCOS mice was similar to that of the control group. In comparison with mice treated with DHEA alone, omega-3 PUFA treatment did not change body weight (Fig. 1B). No differences were observed in FSH levels among the three groups (Fig. 1C). The levels of LH, E2, T and the LH/FSH ratio were elevated in the PCOS mice, but these effects were reversed by omega-3 PUFA treatment (Fig. 1D-G). In addition, compared with control mice, the PCOS mice displayed higher fasting blood glucose levels, in which omega-3 PUFAs significantly decreased (Fig. 1H). Compared with healthy controls, mice treated with DHEA displayed insulin resistance, as revealed by glucose tolerance tests (GTTs) and insulin tolerance tests (ITTs). This impairment in insulin resistance was improved by omega-3 PUFA treatment (Fig. 1I, J). Together, these results demonstrate that omega-3 PUFA supplementation improved insulin resistance and sex hormone disorders in DHEA-induced PCOS mice.

Infertility and metabolic syndrome are major concerns of women with PCOS. Next, we measured the effects of omega-3 PUFAs on ovarian function in PCOS-like mouse models. H&E staining was used to determine the alteration of ovarian pathology in diverse groups (Fig. 2A). The PCOS group exhibited a higher number of cystic follicles and a lower number of corpus lutea than the control group. On the other hand, omega-3 PUFA supplementation decreased the number of cystic follicles and increased the formation of corpus lutea (Fig. 2A). Compared with the PCOS group, omega-3 PUFA-treated mice displayed a decreased ovarian index and reversed disrupted estrous cycles (Fig. 2B-G). Together, these results suggest that omega-3 PUFAs ameliorated ovarian dysfunction in PCOS mice.

To assess the overall composition of the bacterial community in different groups, we analyzed the degree of bacterial taxonomic similarity at the genus level (Fig. 4A). Compared to control mice, PCOS mice displayed an increase in the relative abundance of Alloprevotella and a decrease in the relative abundance of Akkermansia and Alistipes, while omega-3 PUFA treatment protected against these effects (Fig. 4B-K). A heatmap was further used to identify the specific bacterial phylotypes that were altered by omega-3 PUFAs (Fig. 4L). In control mice, treatment with DHEA significantly increased 6 operational taxonomic units (OTUs) and decreased 15 OTUs (Fig. 4L). Among these 21 OTUs, omega-3 PUFA treatment regulated 20 OTUs, which were altered in PCOS mice compared with control mice (Fig. 4L). These results indicate that omega-3 PUFAs modulated the gut microbiota of PCOS mice.

To further investigate whether the beneficial effects of omega-3 PUFAs were mediated by the gut microbiota, we transferred the microbiota from omega-3 PUFA-treated mice to PCOS recipient mice (Fig. 5A). After 8 weeks of colonization, horizontal fecal transfer from omega-3 PUFA-treated mice (omega-3→PCOS) demonstrated some protective effects similar to those observed in the omega-3 PUFA-treated group, such as fasting blood glucose (Fig. 5C) and LH/FSH ratio (Fig. 5F). However, the body weight (Fig. 5B), ovarian index, serum FSH, LH, E2 and T levels were not significantly different between the t-PCOS and t-Omega-3 groups (Fig. 5D-I). Compared with t-PCOS mice, t-Omega-3 mice also displayed disrupted estrous cycles (Fig. 5J, K). To further confirm whether the protective effects of omega-3 were conducted by microbes, we treated omega-3 PUFA-treated PCOS mice with a cocktail of antibiotics, which included vancomycin, neomycin sulfate, metronidazole, and ampicillin. When the gut microbiota was suppressed by the antibiotic cocktail, only the protective effects of LH levels were abolished (Fig. 1). Collectively, these results demonstrated that the gut microbiota mediates part of the beneficial effects of omega-3 PUFAs on the ovaries.

The possible mechanisms by which omega-3 regulates insulin resistance in PCOS were investigated. White adipose tissue browning increases the metabolic rate and improves insulin resistance [40, 41]. Transplantation of brown adipose tissue reversed anovulation, polycystic ovaries, insulin resistance and infertility in PCOS rats [42]. Therefore, the morphology of adipose tissues and thermogenic markers were measured in subcutaneous and brown adipose tissues from PCOS mice. Omega-3 PUFA-treated mice showed a decrease in multilocular cells within adipose tissues (Fig. 6A). PCOS mice exhibited significantly lower levels of Ucp1, Pgc1a, Cited1 and Cox8b mRNAs than controls in subcutaneous adipose tissues. Notably, the relative expression of these thermogenic markers in subcutaneous fat was dramatically elevated after omega-3 administration in the PCOS mice (Fig. 6B-E). However, although the expression of Pgc1a and Cited1 in brown fat tissues was decreased in PCOS mice, omega-3 PUFA treatment did not change these thermogenic markers (Fig. 6F-I). Interestingly, different results were observed in omega-3 + ABX mice, that is, browning of adipose tissues was decreased, as revealed by the reduced expression of Ucp1, Cited and Cox8b in subcutaneous fat tissues and the increase in multilocular cells within the adipose tissues (Fig. 6A-E). However, transferring feces from omega-3 PUFA-treated mice to PCOS mice only increased Ucp1 and Pgc1a expression in both subcutaneous and brown adipose tissues, and other thermogenic markers did not change significantly (Fig. 7A-I). These data indicated that the mechanisms underlying omega-3 PUFA-regulated improvements in the insulin resistance associated with PCOS likely involve effects on adipose tissue to decrease thermogenic markers and multilocular cells within the subcutaneous adipose tissues. Furthermore, the gut microbiota played a key role in the improvement of adipose tissue function and morphology by omega-3 in PCOS mice.

The possible mechanisms by which omega-3 regulates ovarian function in PCOS were investigated. Inflammation and oxidative damage play an essential role in the pathogenesis of PCOS [43]. We next investigated whether omega-3 plays an anti-inflammatory role in the ovaries of PCOS mice. Supplementation with omega-3 PUFAs significantly reversed the enhanced expression of IL-1b, TNF-a and IL-18 mRNAs in PCOS mouse ovaries (Fig. 8A-C). However, the mRNA levels of CCL2 did not increase in PCOS mice (Fig. 8D). There was no difference between the omega-3 and omega-3 + ABX groups. Together, RT‒qPCR analyses confirmed that t-omega-3 mice showed decreased expression of Il1b and Ccl2 (Fig. 8E, H). The mRNA levels of TNF-α did not change between t-PCOS and t-Omega-3 groups (Fig. 8F). There was a trend toward decreased Il18 (Fig. 8G), but this difference was not statistically significant. Moreover, we investigated some antioxidants, such as Ho1, Nqo1, Sod3 and Cat. However, they did not change (Fig. 8I-L). Transferring feces from omega-3 PUFA-treated mice to PCOS mice only increased Ho1 expression in PCOS mouse ovaries, but other antioxidants did not change significantly (Fig. 8M-P). Our data indicated that the mechanisms underlying omega-3 PUFA-regulated improvements in the ovarian dysfunction associated with PCOS likely involve direct effects on the ovary to inhibit inflammation.

Additional file 1.