Comparation of mitochondrial morphology, quantity, and function among primary autologous cells

The isolation and culture process of primary USC, GC, BMSC and ADSC was shown in Additional file 3: Fig. S1. After identifying the GC-specific surface marker (Additional file 3: Fig. S2), the in vitro differentiation capacity of USC (Additional file 3: Fig. S3), and the surface-specific markers of MSCs (Additional file 3: Fig. S4), experimental studies were conducted on these four types of primary cells.

For mitochondrial morphology (Fig. 1A, B), we found that USC and GC demonstrated obviously round-like mitochondria with relative immature cristae, which were similar to oocyte stage. However, BMSC and ADSC mitochondria had more dynamically connected tubular structures, which were sausage-like, elongated in shape with relative mature cristae at the cross sections. All primary cells from the old adults showed strikingly impaired cristae structure and more swollen mitochondria with decreased matrix density (Fig. 1A and C, indicated by arrows), suggesting that advanced age may significantly impair normal cristae structure. Among all types of aged primary cells, GC showed the most severe impaired mitochondrial cristae (vs. USC, P < 0.01; Fig. 1C), while USC showed the least severe damage (vs. BMSC and ADSC, Ps < 0.05; Fig. 1C).

Fig. 1

USC mitochondria resemble oocyte morphologically with higher quantities, MMP and lower ROS less affected by age. A Mitochondrial morphology of different types of human primary GC and MSCs under transmission electron microscopy (TEM). Arrows indicate mitochondria with abnormal cristae. Scale bars, 500 nm. B Percentage of round, sausage-like, elongated shape mitochondria among different cell types. C Percentage of mitochondria with normal cristae among different cell types. D Absolute quantification of mtDNA copy number by RT-PCR. E Mitochondrial membrane potential (MMP, indicated by TMRM) and cytosolic reactive oxygen species (ROS, indicated by DCF) were observed under 3D confocal microscope. Scale bars, 10 μm. F The relative abundance of average MMP fluorescence intensity was quantified. G The relative abundance of average ROS fluorescence intensity was quantified. Data are shown as means ± SEM. Each scatter represents an independent biological individual. One-way ANOVA, LSD test. *P < 0.05, **P < 0.01

The results of mtDNA quantification showed that, no significant differences were found among different types of primary cells in the young population. Compared with young adults, mtDNA copy number of the elderly demonstrated a downward trend, among which GC and BMSC reached significance (GC, P < 0.01; BMSC, P < 0.05; Fig. 1D), but USC still kept at a relative higher level of mtDNA content in the elderly (vs. GC and BMSC, Ps < 0.05; Fig. 1D).

MMP was used to indicate mitochondrial activity, as shown in Fig. 1E, we found that MMP of young USC was significantly higher than GC (P < 0.01; Fig. 1F). The MMP of the elderly was significantly decreased in all cell types (Ps < 0.05; Fig. 1F), but USC still showed a relatively higher MMP than GC and BMSC (Ps < 0.01; Fig. 1F).

ROS reflects the level of cytoplasmic oxidative stress. The ROS level of GC was found to be higher than that of MSCs in young adults (Ps < 0.01; Fig. 1G). In the elderly, the ROS levels were significantly increased in all cell types (Ps < 0.05; Fig. 1G), but GC and BMSC demonstrated relative higher levels than USC and ADSC (Ps < 0.05; Fig. 1G).

Together, these results demonstrated that, USC mitochondria resembled oocytes morphologically, and that its mitochondrial content and activity were relatively high and less affected by age than other primary cell types. Besides, its cytosolic oxidative stress is kept at a relatively low level which may be attributed to its immature mitochondrial state, similar to oocytes.

Comparation of metabolic pattern among primary autologous cells

We further assessed the metabolic capacity of different types of primary cells. As shown in Fig. 2A, most of the glycolytic genes (GLUT1, PFK, GAPDH, LDHA) were highly expressed in MSCs, most notably in USC. Compared with the young group, the expression levels of glycolytic genes of the elderly were significantly up-regulated in all cell types. The extracellular acidification rate (ECAR) detected by Seahorse showed that the glycolytic level and capacity were stronger in USC and ADSC (Fig. 2C–D). Compared with the young group, the elderly showed a trend of increase in glycolysis which was consistent with gene expression results, and GC reached significance in glycolysis level (P < 0.05; Fig. 2C).

Fig. 2

USC shows vigorous metabolism on glycolysis, OXPHOS and mitochondrial genome expression activity. A Analysis of glycolytic mRNA expression levels by RT-PCR. B Normalized extracellular acidification rate (ECAR) detected by Seahorse XFe96 analyzer. C Quantification of glycolysis level. D Quantification of glycolytic capacity. E Analysis of tricarboxylic acid cycle (TCA) mRNA expression levels by RT-PCR. F Normalized oxygen consumption rate (OCR) detected by Seahorse XFe96 analyzer. G Quantification of ATP turnover. H Quantification of maximal respiration. I The mRNA expression level of 13 electron transport chain (ETC) genes encoded by mitochondrial genomes was analyzed by RT-PCR. Data are shown as means ± SEM. Each scatter represents an independent biological individual. One-way ANOVA, LSD test. *P < 0.05, **P < 0.01

As shown in Fig. 2E, the highest expression levels of tricarboxylic acid cycle (TCA) genes (PDHE1α, CS, IDH2, IDH3B, αKGDH, SDHA, FH, MDH2) were found in USC, while the lowest expression levels were found in GC. Compared with the young, the TCA genes of people with advanced age were significantly down-regulated in all cell types. The results of the oxygen consumption rate (OCR) detected by Seahorse also showed that USC demonstrated the strongest maximal respiratory capacity in the young population (P < 0.01; Fig. 2H). With age increasing, the ATP turnover of GC and BMSC decreased (Ps < 0.01; Fig. 2G), and the maximal respiration capacity of all cell types decreased significantly (Ps < 0.05; Fig. 2H). Among all aged cell types, GC reached the lowest in both ATP turnover and maximal respiration, but USC and ADSC still maintained relatively higher levels (vs. GC, P < 0.01; vs. BMSC, P < 0.05; Fig. 2 G, H).

The mRNA expression levels of 13 electron transport chain (ETC) genes (ND1, ND2, ND3, ND4, ND4L, ND5, ND6, CYTB, COX1, COX2, COX3, ATP6, ATP8) encoded by mitochondrial genomes were analyzed in different types of primary cells (Fig. 2I). Among all cell types, people with advanced age showed significant decreased expression levels compared to the young group, but USC still kept in the highest level regardless of age.

Together, the results showed that the metabolic transformation from oxidative phosphorylation (OXPHOS) to glycolysis is generally occurring in primary cells with increasing age, especially in GC. In addition, the overall cellular metabolism of USC, including glycolysis, OXPHOS, and mitochondrial expression activity, was the most vigorous among all cell types, regardless of age.

Aged oocytes show aberrant mitochondrial physiology

To examine the effect of age on mitochondria in germ cells, we performed experiments on the oocytes of the young and advanced age. TEM showed that (Fig. 3A), mitochondria of young oocytes were observed to be round in shape, with a high matrix density and immature cristae. However, mitochondria of aged oocytes showed heterogeneous matrix density and cord-like cristae, and the ratio of normal cristae was significantly reduced compared to the young oocytes (P < 0.01; Fig. 3A). In addition, Confocal showed that mitochondria content was significantly reduced in aged oocytes (P < 0.01; Fig. 3B), accompanied by significantly decreased MMP, increased cytosolic ROS and Ca2+ levels (Ps < 0.01; Fig. 3C–E). These evidences demonstrate the detrimental effects of age on mitochondrial quantity and quality in oocytes.

Fig. 3

Aberrant phenotype of mitochondrial physiology in aged oocytes. A Mitochondrial morphology of human young and aged oocytes under transmission electron microscopy (TEM). Scale bars, 500 nm. B Mitochondrial content (indicated by Mitotracker green) was observed under the confocal microscope. Scale bars, 50 μm. C Cytosolic Ca2+ levels (indicated by Fluo-4) were observed under the confocal microscope. Scale bars, 50 μm. D Mitochondrial membrane potential (MMP, indicated by TMRM) were observed under 3D confocal microscope. Scale bars, 50 μm. E Cytosolic reactive oxygen species (ROS, indicated by DCF) were observed under 3D confocal microscope. Scale bars, 50 μm. The relative abundance of average fluorescence intensity or area was quantified. Data are shown as means ± SEM. Each scatter represents an independent biological individual. Unpaired two-tailed Student’s t-test. *P < 0.05, **P < 0.01

In previous studies, we compared various types of autologous primary cells. Among all cell types, USC demonstrated round shaped mitochondria and a relative low ROS level similar to oocytes, as well as abundant mitochondrial content, relative vigorous mitochondrial activity and cellular metabolism.

To further verify the mitochondrial genomic biosecurity of USC, we performed whole mitochondrial genome sequencing on 2 cases of young USC and 2 cases of aged USC (Additional file 3: Fig. S5). We did not observe obvious differences in the number of SNVs and InDels in the D-loop, Gene, tRNA, and rRNA regions of the USC mitochondrial genome, and there was no significant difference in the number of different types of SNVs and InDels in the coding region (Additional file 3: Table S1). Pathogenicity prediction analysis by MutPred and Polyphen-2 HumVar found no pathogenic SNVs and InDels loci with high heterogeneity scores.

Considering USC can also be easily obtained by non-invasive procedure and large-scale expanded, it can act as a superior autologous donor cell for oocyte cytoplasmic mitochondria transfer.

Autologous non-invasively USC-derived mitochondrial transfer improves mitochondrial content and function of human early embryos, especially in the advanced age

Mature oocytes (MII stage) were obtained through in-vitro maturation (IVM) culture of immature oocytes (GV or MI stage) from IVF/ICSI patients. Clinical characteristics of donors were listed in Additional file 3: Table S2. Finally, 42 mature oocytes of the young population and 29 mature oocytes of the population with advanced age were included in the following study, and oocytes were further randomized to the corresponding conventional ICSI group and Mito ICSI group. The proportions of GV and MI stage-derived oocytes between the control and treatment groups remained generally consistent. For different age groups, USC mitochondria from young or aged populations were extracted, and transferred during the ICSI process (Fig. 4A, a). For the detailed Mito ICSI operation process, see Additional file 2: Movie S1.

Fig. 4

Improvement of mitochondrial content and function in embryos after USC mitochondria transfer. A Representative images of early embryo development in the Control and MITO ICSI groups. a Co-injection of single sperm with USC-derived mitochondria during ICSI; b-c Grade IV embryos on day 3 (b) and Grade 6CB blastocysts on day 6 (c) in the Control group; d-f Fertilization and 2PN formation on day 1 (d), Grade I 8-cells embryos on day 3 (e), and Grade 3BC blastocysts on day 5 (f) in the MITO ICSI group. B Mitochondrial content (indicated by Mitotracker green) and cytosolic Ca2+ (indicated by Fluo-4) were observed under confocal microscope. Scale bars, 50 μm. C MMP (indicated by TMRM) and cytosolic ROS (indicated by DCF) were observed under 3D confocal microscope. The relative abundance of average fluorescence intensity or area was quantified. Data are shown as means ± SEM. Each scatter represents an independent biological individual. Unpaired two-tailed Student’s t-test. *P < 0.05, **P < 0.01

As shown in Fig. 4A, we observed the general embryonic development in the control group (Grade IV embryos on day 3, b; Grade 6CB blastocysts on day 6, c) and the Mito ICSI group (fertilization and 2PN formation on day 1, d; Grade I 8-cells embryos on day 3, e; Grade 3BC blastocysts on day 5, f). Confocal analysis showed that the mitochondrial content of early embryos in the population with advanced age was significantly lower than that in the young, but after USC mitochondrial transfer, the mitochondrial content of both young and aged embryos increased significantly (Ps < 0.01; Fig. 4B), revealing the effective mitochondrial transfer process during ICSI. Futher, we analyzed cytoplasmic physiological state and mitochondrial function. Compared with young embryos, aged embryos demonstrated lower MMP, and higher levels of cytosolic ROS and Ca2+ (Ps < 0.01; Fig. 4B, C). After USC mitochondrial transfer, significant increase of MMP and decrease of Ca2+ were observed in both young and aged embryos (Ps < 0.01; Fig. 4B, C), however, the improvement was generally greater in the advanced age. Besides, the aged embryos also showed significantly decreased cytosolic ROS after mitochondrial transfer, but no difference was observed in the young embryos (Fig. 4C).

Autologous non-invasively USC-derived mitochondrial transfer improves morphological development and euploidy rate of human early embryos with advanced age

Morphological evaluation of early embryos is listed in Table 1. Although not statistically significant due to the limited number of available embryos, we observed an upward trend in the rate of 7–10 cell embryos at EDD3 in both of the young and old population after mitochondria transfer, with the young group rose from 9.5% (2/21) to 23.8% (5/21), and the elderly group rose from 26.7% (4/15) to 50.0% (7/14). In addition, in old populations with mitochondria transfer, we also found an upward trend in the rate of good-quality embryos at EDD3 which increased from 20.0% (3/15) to 42.9% (6/14), and the rate of blastocyst formation at EDD5 which increased from 13.3% (2/15) to 21.4% (3/14). Moreover, one high-quality blastocyst (Grade 4BB) was obtained only in the advanced age with mitochondria transfer (1/14).

Table 1 Evaluation of embryo morphological and ploidy between the Control and Mito ICSI group

Interestingly, in the conventional ICSI groups, decreased formation rates of good-quality embryos and 7–10 cell embryos at EDD3, as well as blastocyst formation rates at EDD5, were found in the young population when compared to the advanced age. Actually, clinical immature oocytes obtained from the young and elderly patients have their own developmental problems, which may be related to the aging factor in the population with advanced age, and may also be caused by a number of complex factors in the young population that we do not yet fully understand, despite our inclusion criteria have excluded a range of basic diseases. This phenomenon of improvement in the early embryonic development after mitochondria transfer in the elderly may explain the dominant role of age-mitochondrial related abnormalities in this population, while the young population seems to be associated with more abnormal non-mitochondrial related factors, which needs to be further explored in future studies.

Further, we detected the euploidy rate of early embryos in the Control and the MITO ICSI group of both the young and elderly population (Table 1). Due to the poor quality of the embryos derived from clinically discarded immature oocytes, the average detection rates were around 50%. Encouragingly, consistent with our previous observations in embryo morphology, the highest detection rate was found in the advanced age with mitochondria transfer (77.8%, 7/9). Moreover, among all embryos, 2 euploid embryos were found only in the advanced age with mitochondria transfer (28.6%, 2/7), however, the other groups were all aneuploid or mosaic embryos. Although not statistically significant due to the limited number of available embryos, the results suggest that mitochondria transfer in IVF populations with advanced age may exert favorable effects on the restoration of embryo euploidy. Representative results of euploidy, aneuploidy (segmental aneuploidy, complex aneuploidy), and mosaicism were shown in Additional file 3: Fig. S6.