Abundance of cell‐free mitochondrial DNA in spent culture medium associated with morphokinetics and blastocyst collapse of expanded blastocysts

Abstract Purpose This retrospective observational study investigated relationships between the abundance of cell‐free mitochondrial DNA (cf‐mtDNA) in spent culture medium (SCM) of human‐expanded blastocysts and their morphokinetics to address the question of whether the abundance of cf‐mtDNA in SCM could predict the quality of blastocysts. Methods Embryos (n = 53) were individually cultured in a time‐lapse incubator until they reached the expanded blastocyst stage (5 or 6 days), following which copy numbers of cf‐mtDNA in SCM (20 μL) of expanded blastocysts were determined using real‐time PCR. Results The duration between start of blastulation to expanded blastocyst (tEB–tSB) and between that of the blastocyst stage to expanded blastocyst (tEB–tB) significantly and positively correlated with the abundance of cf‐mtDNA in the SCM (tEB–tSB: r = .46; P < .01; tEB–tB: r = .47; P < .01). The abundance of cf‐mtDNA in the SCM was significantly greater in blastocysts with blastocyst collapse (BC), than without BC, and significantly and positively correlated with the number of BC. Conclusions The abundance of cf‐mtDNA in the SCM was associated with expansion duration and BC. Thus, cf‐mtDNA abundance in the SCM serves as a marker to predict the quality of expanded blastocysts.

select chromosomally normal embryos before SBT, which improves CPR and reduces spontaneous abortion rates. Although these evaluation methods are useful for embryo selection, the LBR after elective SBT remains low, at 40%-70%. [8][9][10] Therefore, other parameters are needed for embryo evaluation. Spent culture medium (SCM) contains cell-free DNA (cf-DNA) derived from embryos and might serve as a noninvasive marker of embryo status. However, a study of the relationship between cf-DNA derived from embryonic nuclei and aneuploidy found that contamination with embryo-associated structures led to a low probability of cf-DNA serving as a marker of embryo status. 11 Moreover, more mitochondrial (mtDNA) than nuclear DNA is detected in SCM because cells and embryos harbor more copies of the mitochondrial, than the nuclear genome. Sequencing of DNA extracted from porcine follicular fluid and real-time PCR of DNA extracted from SCM of porcine granulosa cells has revealed that the abundance of cf-mtDNA in SCM exceeds that of cf-nucleic DNA by factors ranging from the tens to the hundreds, 12,13 indicating that cf-mtDNA in SCM could serve as a marker for embryonic evaluation. The abundance of cf-mtDNA in SCM reflects the mitochondrial status of porcine granulosa cells and bovine early cleaved embryos. 14,15 Furthermore, the abundance of cf-mtDNA in SCM of human embryos (day 3 post-insemination) reflects a higher developmental rate and raises pregnancy expectations. 16 However, how abundance of cf-mtDNA in SCM relates to the developmental events and quality of the blastocysts remains unknown. In the present study, we assessed embryo development till the expanded blastocyst stage using a time-lapse incubator to determine cf-mtDNA abundance in SCM, and its relationship to morphokinetic data.

| Ethical approval
The Institutional Review Board at Kanagawa Ladies Clinic approved this retrospective observational study (Approval no. 003; 2017). All involved patients provided informed consent to participate in this study, following which samples were collected and analyzed.

| Participating patients and ovarian stimulation
Twenty couples underwent in vitro fertilization (IVF) at the Kanagawa Ladies Clinic in Kanagawa, Japan, during 2019. All patients were treated with an ovarian stimulation protocol using a gonadotropin releasing hormone (GnRH) agonist or antagonist. The effects of follicle-stimulating hormone (FSH) stimulation were monitored by measuring serum estradiol (E2) levels and follicle growth. Cycles without proper follicle growth and E2 elevation were excluded from further treatment. Human chorionic gonadotropin (hCG) was administered when follicles reached a diameter of ≥18 mm, and oocytes were retrieved 35-36 h later, rinsed with Sydney IVF Fertilization Medium (Cook Medical), and incubated until insemination.

| Insemination and embryo culture, frozenthawed transfer
We counted the numbers, assessed the motility, and measured concentrations of spermatozoa in fresh semen, and selected those with high motility based on swim-up protocols. Using the standard IVF protocol, oocytes were co-incubated with 5 × 10 6 /mL spermatozoa in fertilization medium for 5 hours, and then cumulus cells were denuded from the oocytes by mechanical pipetting.

| Spent culture medium collection
When the culture was finished, embryos were removed from the wells and 20 μL of the applied 25 μL of SCM was collected and stored at -20°C. SCM was centrifuged (2500 g for 1 minute) to remove potentially contaminated cellular debris and sperm and the supernatant was used for DNA extraction.

| Defined blastocysts used in this study
For the present analysis, SCM was used only from blastocyst that reached the expanded blastocyst stage (≥170 µm in inner diameter). Blastocysts with an inner diameter at the <170 µm, hatched and hatching blastocysts, and degeneration blastocysts were excluded.

| Morphokinetic evaluation
Images of individual blastocysts were retrospectively analyzed using Embryo Viewer, an external computer workstation (Vitrolife), and the timing of embryonic developmental events during culture from postinsemination to the expanded blastocyst stage was evaluated as described. 17 Morphokinetic parameters included pronuclear fading (tPNf), onset of 2-9 cell divisions (t2, t3, t4, t5, t6, t7, t8, t9+), timing of morula formation (tM), start of blastulation (tSB; first signs of a visible blastocoel), full blastocyst (tB; just before ZP thinning, blastocyst has not reached an inner diameter of 170 µm), and expanded blastocyst (tEB; time of reached an inner diameter of 170 µm). Based on the findings, we determined durations between the following developmental events: period of second cell cycle from 2 to 3 cells (CC2), third cell cycle from 3 to 5 cells (CC3), blastulation from tSB to tB (tB-tSB), duration of blastocyst expansion from tSB to tEB (tEB-tSB), and from tB to tEB (tEB-tB). The primers were 5′-ccctaaaacccgccacatct -3′ and 5′-ggcctaggttgaggttgacc-3′, which targeted short (126 bp) mitochondrial sequences (3486-3612). The PCR cycling program comprised initial denaturation at 95°C for 3 minutes, followed by 40 cycles at 97°C for 6 s and 60°C for 10 s. The PCR efficiency was confirmed using a melting curve and electrophoresis. A standard curve was generated for each run using 10-fold serial dilutions of representative PCR products that were cloned using Zero Blunt TOPO PCR Cloning Kits (Invitrogen). Copy numbers of standard PCR products were obtained from their concentrations using Avogadro numbers. The PCR products were confirmed by sequencing before use. Based on the copy number of mitochondrial genomes in the sample, cf-mtDNA copy numbers were calculated per 20 μL of culture medium.

| Definition of blastocysts collapse
Blastocyst collapse (BC) was defined as described by Bodri et al and Sciorio et al. 18,19 During collapse, volumes of embryos at maximum expansion and of embryos with the lowest volume during collapse were compared using the EmbryoViewer drawing tool. Reductions in embryonic volume of ≥50% and <50% were respectively defined as BC and contraction. Figure 1 shows representative embryos at BC or contraction.

| Statistical Analysis
Data were statistically analyzed using JMP statistical software, ver-

| Morphokinetic data and abundance of cf-mtDNA in the SCM
The cf-mtDNA abundance in the SCM did not significantly correlate with any morphokinetic parameters during all incubation periods (Table 2). However, duration of tEB-tSB (r = .46, P < .01) and tEB-tB (r = .47, P < .01) significantly and positively correlated with the abundance of cf-mtDNA in the SCM (Table 3). Figure 4 shows that the mean (±SD) cf-mtDNA abundance in SCM was significantly greater in blastocysts with, than without BC (14.9 ± 7.7 vs. 7.67 ± 3.9, P < .01), and that it significantly correlated with the number of BC (r = .31, P < .01) ( Figure 5). Figure 6A,B shows relationships between BC and the duration of the blastocyst stage in embryos with and without BC. The mean (±SD) duration of tEB-tSB (22.9 ± 6.0 vs. 17.2 ± 4.5, P < .01) and tEB-tB (15.9 ± 5.9 vs. 11.3 ± 4.1, P < .01) was significantly longer in embryos with, than those without BC.  Figure 7B). Furthermore, a significant positive correlation was observed between the cf-mtDNA content in SCM and tEB-tB duration in good blastocysts (r = .35, P = .025; Figure 8B). While the correlation was not significant, there was also a positive trend between cf-mtDNA content and tEB-tSB duration (r = .28, P = .07; Figure 8A). Moreover, good blastocyst with BC had a greater abundance of cf-mtDNA in the SCM than those without BC (14.9 ± 7.7 vs. 7.67 ± 3.9, P = .02; Figure 9).

| D ISCUSS I ON
The present study demonstrated significant relationships between abundance of cf-mtDNA in the SCM and morphokinetic parameters of the embryos including duration of blastocyst stages and presence of blastocyst collapse which were obtained using time-lapse incubator. Cumulative ET cycles (times) 0.8 0 0-7 1.8 .94 Note: Associations between abundance of cf-mtDNA in SCM and age, BMI, AMH level, cumulative OPU, and ET cycles (including frozen-thawed ET) were analyzed by multiple linear regression models. Data are shown as means, medians, range, and SD.
Abbreviations: AMH, anti-Müllerian hormone; BMI, body mass index; cf-mtDNA, cell-free mitochondrial DNA; ET, embryo transfer; OPU, ovum pick-up; range, minimum to maximum; SD, standard deviation.  21 The present study found that cf-mtDNA abundance in the SCM significantly and positively correlated with the duration of tEB-tSB and of tEB-tB. In addition, when the data were filtered by embryonic grade, the difference was also observed in good blastocysts. These results indicate that higher abundance of cf-mtDNA in the SCM is a marker of poor blastocyst quality.

F I G U R E 2
The major unresolved issues regarding cf-mtDNA in the SCM are its origin and its secretion mechanism from embryos into the SCM.
That is, cf-mtDNA might be generated from dead or fragmented blastomeres or arise due to specific events in blastocysts or actively secreted blastocysts.
Furthermore, the present study found no significant differences in cf-mtDNA abundance in the SCM with respect to morphological blastocyst quality (ICM and TE grades).
However, whether cf-mtDNA abundance in the SCM reflects mitochondrial numbers in corresponding blastocysts remains to be clarified.
More mitochondria in oocytes is a marker of good quality, as oocytes with fewer mitochondria have poor fertilization and developmental outcomes. 22,23 However, this is not true of blastocysts, as mtDNA abundance in TE cells and the quality of blastocysts are negatively associated; that is, mtDNA is more abundant in TE cells in aneuploid blastocysts. 24,25 After fertilization, the number of mitochondria in embryos decreases and those in embryos at the early developmental stage reflect the number in oocytes, because active mitochondria are not generated until the blastocyst stage. [26][27][28] We did not assess relationships between mitochondrial DNA copy numbers in blastocysts and cf-mtDNA abundance in the SCM, but higher mtDNA content in blastocysts might be the background for the increased abundance of cf-mtDNA in the corresponding SCM.
The notion that cf-mtDNA abundance reflects poor embryo quality  Note: Data are shown as means, range, correlation coefficient, and P value.
Abbreviations: CC2, second cell cycle from 2 to 3 cells; CC3, third cell cycle from 3 to 5 cells; h, hours; r, Spearman rank-order correlation coefficient; tB-tSB, duration of tSB to tB; tEB-tB, duration of tB to tEB; tEB-tSB, duration of tSB to tEB. and on corresponding blastocysts. Here, the mitochondrial genome was easily detected in the SCM, but the copy number of nucleic DNA was lower than that in the mitochondrial genome (data not shown).
These data are consistent with previous findings showing up to several hundred-fold more mitochondrial genomic DNA than nuclear DNA in the SCM. 14 F I G U R E 6 Comparison of tEB-tSB and tEB-tB durations between blastocysts with and without BC. Boxplots show blastocysts with (BC; n = 39) and without (Non-BC; n = 14) BC. Y-axis, durations between tEB-tSB (A) and tEB-tB (B) (hours). P < .01. BC, blastocyst collapse; tEB-tB, duration between blastocyst stage (tB) and expanded blastocyst (tEB); tEB-tSB, duration between start of blastulation (tSB) and expanded blastocyst (tEB)

F I G U R E 7
Comparison of cf-mtDNA abundance in SCM at day 5 and day 6 and between good and poor blastocysts. A, Cell-free mitochondrial DNA (cf-mtDNA) copy numbers in spent culture medium (SCM) were compared between SCM at culture day 5 (n = 18) and culture day 6 (n = 35). Y-axis, copy number of cf-mtDNA in 20 μL of SCM P = .18. B, Cell-free mtDNA copy numbers in SCM were compared between morphologically good (AA, AB, BA and BB, n = 40) and poor (BC and CC, n = 13) blastocysts. Y-axis, copy number of cf-mtDNA in 20 μL of SCM P = .20. cf-mtDNA, cell-free mitochondrial DNA; SCM, spent culture medium The main limitation of the present study is the sample size, which was too small to demonstrate a statistically significant relationship between cf-mtDNA abundance in SCM and clinical outcomes.
However, we found a trend toward greater abundance of cf-mtDNA in the SCM of embryos that failed to implant. In addition, the data obtained on culture duration and embryonic stage improve the understanding of the significance of cf-mtDNA and should be examined further in future studies.
In conclusion, this study revealed more abundant cf-mtDNA in the SCM of embryos with delayed blastocyst expansion and BC. This suggests that cf-mtDNA abundance in the SCM could serve as an additional or alternative marker of blastocyst evaluation.

ACK N OWLED G M ENTS
This study received no financial support. We thank Vitrolife Japan for providing the Embryoscope. 02. cf-mtDNA, cell-free mitochondrial DNA; SCM, spent culture medium; tEB-tB, duration between blastocyst stage (tB), and expanded blastocyst (tEB); tEB-tSB, duration between start of blastulation (tSB) and expanded blastocyst (tEB)

F I G U R E 9
Comparison of cf-mtDNA abundance in SCM between blastocysts with and without BC using the data from good blastocysts. Cf-mtDNA copy numbers of SCM were compared between blastocysts with BC (BC; n = 27) and without BC (Non-BC; n = 13). Y-axis, copy number of cf-mtDNA in 20 μL of SCM P = .02. BC, blastocyst collapse; cf-mtDNA, cell-free mitochondrial DNA; SCM, spent culture medium