Menstrual blood-derived stem cells and its mitochondrial treatment improve the ovarian condition of aged mice

Macroscopic ovarian sizes of the recipient mice treated by MenSCs and mitochondria

Ovaries of aged recipient CD-1 (ICR) mice (10–12 months old) were treated with, either MenSCs 3D gel (MO-MenSCs) incubation or mitochondrial injection combined with mitochondrial 3D gel (MO-MITO) incubation. After 7 days following the treatment, ovaries were dissected from the recipient mice and their overall morphology and size was examined. The blank gel control group and MO-MITO treatment group had comparable ovarian volume. However, the MO-MenSCs treatment group had higher ovary size and volume compared to control group. (Figure 1A, 1B).

Figure 1. Macroscopic ovarian sizes (A) and representative bar graph summarizing the ovarian volumes (B) in the three groups after seven days of treatment.

Morphological changes in the recipient ovaries treated by MenSCs and mitochondria

We further observed the histomorphology of the recipient ovaries by Hematoxylin-eosin (HE) staining. HE staining showed that the ovaries of 2–3 months young control group (MY-Control) were properly structured, with different stages of developing follicles, and layers of granulosa cells around the secondary follicle (Figure 2A). The aged control mice (MO-Control) showed disordered ovarian structure, with fewer number of growing follicles at each stage and granulosa cells as well (Figure 2A). MO-control ovaries had more atretic follicles, and enlarged intercellular space. MO-MenSCs and MO-MITO treatment moderately improved the morphology of aged ovaries (Figure 2A). The number of growing follicles at each stage and corpus lutea were counted in the MO-control, MO-MenSCs, and MO-MITO treatment groups. MO-MenSCs and MO-MITO treatment groups showed moderately higher number of primary and secondary follicles compared to the MO-control group (Figure 2B).

Figure 2. MO-MenSCs and MO-MITO treatment groups showed similar effects on the aged ovary. Representative images showing H&E-stained ovary sections in each group after seven days (A). Bar graph summarizing the number of growing follicles at various stages and corpus lutea in each group (B).

Tracking donor MenSCs in the ovaries of recipient mice

We labelled donor MenSCs with red fluorescent CM-Dil and tracked them in the recipient mouse ovaries 7 days after treatment with MenSCs 3D gel. The ovaries from recipient mice were harvested after sacrifice, and paraffin sections were prepared to visualize the location of labelled MenSCs. CM-Dil red labelled MenSCs were clearly visible inside the recipient ovaries 7 days after treatment with 3D MenSCs gel (Figure 3). This result indicates that MenSCs can migrate into aged ovaries in vivo, which may repair the injury and improve the ovarian condition in aged mice.

Figure 3. Visualizing the red fluorescent CM-Dil labelled MenSCs in the ovaries of two individual recipient mice.

RNA expression analysis and related gene functions and pathways

Furthermore, to link ovary morphology changes and altered gene expression, RNA sequencing (RNA-seq) was performed to analyze the global transcriptomic profiles from MO-control and MO-MenSCs and MO-MITO treated ovaries. A sizable number of genes (429 upregulated and 677 downregulated) showed at least two-fold differential expression between MO-MenSCs and MO-control samples (Figure 4A). These differentially upregulated expressed genes were significantly enriched in cellular component (CC) categories related to proteinaceous extracellular matrix, extracellular matrix, collagen trimer, basement membrane, extracellular exosome, nucleoplasm, cytoplasm, mitochondrial respiratory chain complex I, growth cone, mitochondrial intermembrane space, mitochondrion, and mitochondrial inner membrane (Figure 5A).

Figure 4. Identification differently expressed genes by RNA-seq analysis. Volcano plots showing differently expressed genes between MO-MenSCs vs. MO-control (A) and MO-MITO vs. MO-control (B) samples identified using transcriptomic data at day 7. Red dots denote the genes passing our p value and fold difference thresholds. The genes gene names of significant genes are marked. Venn diagram depicting the overlap of up-regulated genes between MO-MenSCs vs. MO-control and MO-MITO vs. MO-control comparisons (C).

Figure 5. Functional annotation of up-regulated genes between MO-MenSCs vs. control (A) and MO-MITO vs. control samples (B).

Similarly, a total of 1,969 genes were upregulated and 2,322 genes were downregulated with at least two-fold change in MO-MITO samples compared to MO-control samples (Figure 4B). Of the genes that were upregulated in MO-MenSCs group, nearly 30% of them were also upregulated in MO-MITO group (Figure 4C). Significant enrichment of these 1,969 upregulated genes was observed in the following cellular components: ribosome, intracellular ribonucleoprotein complex, mitochondrion, mitochondrial respiratory chain complex I, respiratory chain, mitochondrial inner membrane, catalytic step 2 spliceosome, centrosome, spindle, and mitochondrial ribosome (Figure 5B). Interestingly the differentially expressed genes in both treatments were significantly enriched in mitochondria and their components (Figure 5A, 5B). This suggests that treatment of aged ovaries with MenSCs or their mitochondria may affect mitochondria related pathways.

We also found that several genes involved in mitochondrial pathways were upregulated upon MO-MenSCs or MO-MITO treatment in aged ovaries. Genes involved in mitochondrial ATP synthesis such as mt-Atp8, mt-Co2, mt-Nd3, mt-Nd4l, Got2-ps1, Tomm7, mt-Atp6, mt-Tl2, Atp5k, mt-Te, and NADH dehydrogenase were significantly upregulated in the MO-MenSCs treatment group compared to MO-control group (Figure 6A). Similarly, in the MO-MITO treatment group, mt-Nd1, Immp2l, Mrpl39, mt-Ti, Mrpl45, Slc25a29, Slc25a14, Mrpl48-ps, Ucp3, Mrps17, Tomm20 and other mitochondrial-related genes were upregulated, whereas the autophagy pathway molecule Prkn was significantly down-regulated compared to the MO-control group (Figure 6).

Figure 6. Heatmap showing the differential expression of mitochondrial related genes between MO-MenSCs vs. MO-control (A) and MO-MITO vs. MO-control samples (B).

Major reagents

The major reagents and antibodies used in this study were Collagenase I (Sigma, USA), DMEM (GIBCO, USA), 0.25% Trypsin-EDTA (GIBCO, USA), FITC Anti-CD34 (Ebioscience, USA), FITC Anti-CD45 (Ebioscience, USA), PE Anti-CD29 (Ebioscience, USA), FITC Anti-CD90 (Ebioscience, USA), PE Anti-CD73 (Ebioscience, USA).

Animals

Experiments involving animals were approved by the General Hospital Animal Ethics Committee of PLA and were performed in accordance with the NIH Guidelines on the Care and Use of Laboratory Animals. Female CD-1 (ICR) mice aged 10–12 months were used as donors and recipients in the experiment.

Isolation of MenSCs

For preparation of MenSCs, 10–12 months CD-1 (ICR) female mice were sacrificed for cervical dissection, and the bilateral uterus was isolated under aseptic conditions. The fat tissue and blood vessels surrounding the uterine horn were removed and the uterus washed with PBS. The uterus was then cut longitudinally along the uterine cavity to expose the endometrium under microscope, and the endometrium was carefully isolated at 4° C. The tissues were minced and digested for 30–60 minutes with gentle agitation in serum-free medium using 0.1% type I collagenase (Sigma, USA). Then, the enzyme was inactivated and a 200-μm mesh filter was used to the filter samples. The filtered samples were centrifuged at 1200rpm for 5 minutes to pellet the cells. The isolated cells were cultured in DEME low-glucose medium with 10% FBS, 50 U/mL penicillin, and 50 μg/mL streptomycin in a humidified incubator at 37° C with 5% CO₂ [26]. To ensure the purity of the MenSCs, we identified MenSCs by flow cytometry (Supplementary Figure 1).

Isolation of mitochondria

Mitochondria were isolated from MenSCs by differential centrifugation. The MenSCs were suspended in PBS at 10⁶ cells per mL. Then the cells were homogenized 4–6 times on ice using a cell homogenizer. The homogenate was then centrifuged for 20 min at 2000 rpm, the supernatant was collected and again centrifuged for 20 min at 10000 rpm for the mitochondria to sediment [26]. All the processes were carried out at 4° C, and the extracts were maintained at 4° C until use.

Preparation of MenSCs 3D alginate gel and mitochondrial 3D alginate gel

The preparation alginate gel was performed as described previously [27, 28]. Sodium alginate powder (Sigma-Aldrich, USA) was dissolved in PBS to form 2% sodium alginate solution at room temperature. The MenSCs were digested with 0.25% trypsin and added into the alginate solution, the cell concentration was adjusted to 10⁶ cells per mL. Then, the cell suspension droplets were added into 1% calcium chloride solution to form a 3D stem cell gelatinous substance. Similarly, we extract mitochondria from 10⁶ stem cells to prepare 3D mitochondria gel. The whole encapsulation process is carried out under aseptic conditions.

Preparation of fluorescent labeled MenSCs

Cultured MenSCs were incubated with 4 μL/mL of CM-Dil and incubated for 30 min at 37° C, shaken once every 10 minutes. After incubation, the cells were washed twice with PBS to remove the excess dye [29, 30].

Treatment of MenSCs gel and MITO gel to the ovaries of recipient mice

Recipient mice (10–12 months old) were individually anesthetized with 0.2 mL of pentobarbital (0.2%). After the mice were anesthetized, they were placed flat on their abdomen, fixed and the dorsal side of the abdomen was disinfected with 75% alcohol. Incision was made to cut the skin and muscle to expose the ovaries. 20–50 μL of control gel, MenSCs gel, or mitochondrial gel was applied it around both ovaries After surgery, the mice were injected with 100,000 U of penicillin to prevent infection. Three recipient mice were used in each of the control and MO-MenSCs and MO-MITO treatment groups.

Morphological examination of mouse ovaries

Seven days after treatment with the control or MenSCs or MITO 3D gel, both ovaries of the recipient mice were isolated and fixed in 4% paraformaldehyde overnight. Following fixation, the ovaries were dehydrated and embedded in paraffin blocks and sections were immobilized on slides. Ovarian sections were stained with hematoxylin and eosin. The CM-Dil red fluorescence of mesenchymal stem cells in the treated ovaries were also imaged within the paraffin section under a fluorescence microscope.

RNA library preparation and sequencing

Ovary samples were homogenized in TRIzol (Invitrogen) and processed according to the manufacturer’s instructions. 2 ug of total RNA from each sample was processed using the NEBNext®Ultra II kit (New England Biolabs, Inc.) according to the manufacturer's instructions. Briefly, mRNAs were captured with oligo-dT magnetic beads and fragmented for 6 min at 94° C. First strand synthesis was performed on fragmented RNA, followed by second strand synthesis and dA-tailing. Adapter ligation was performed using KAPA single-indexed adapters (KAPA, Pleasanton, CA). The adapter-ligated fragments were amplified with 10 PCR cycles. Final libraries were then pooled and sequenced on a HiSeq4000 single-end 50 nt system (Illumina).

RNA-seq data analysis

FASTQ sequence files from biological replicates were concatenated and aligned to the M. musculus reference genome (mm10) using Hisat2 [31]. FeatureCounts software was used for counting reads to each exon or each gene for read summarization [32]. Differentially expressed genes with two-fold or higher change, and p < 0.05 (unpaired t-test) were identified using the DESeq2 program [33]. Enrichment analysis of differentially expressed genes was performed using the web-based software DAVID and including the cellular component Gene Ontology vocabulary. Statistical parameters to consider significant enrichment were p < 0.01, and fold enrichment > 2.