If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
IVIRMA New Jersey, Basking Ridge NJ 07920, USADepartment of Reproductive Endocrinology and Infertility, Sidney Kimmel Medical College, Thomas Jefferson University, Philadelphia PA 19107, USA
IVIRMA New Jersey, Basking Ridge NJ 07920, USADepartment of Obstetrics, Gynecology, and Reproductive Sciences, Yale School of Medicine, New Haven CT 06510, USA
IVIRMA New Jersey, Basking Ridge NJ 07920, USADepartment of Reproductive Endocrinology and Infertility, Sidney Kimmel Medical College, Thomas Jefferson University, Philadelphia PA 19107, USA
Can cumulus cells be used as a non-invasive target for the study of determinants of preimplantation embryo quality?
Design
Cumulus cells were collected from monosomy 21, trisomy 21 and euploid embryos and subjected to RNA sequencing analysis and real-time polymerase chain reaction assays. The differential gene expression was analysed for different comparisons.
Results
A total of 3122 genes in monosomy 21 cumulus cells and 19 genes in trisomy 21 cumulus cells were differentially expressed compared with euploid cumulus cells. Thirteen of these genes were differentially expressed in both monosomy and trisomy 21, compared with euploid, including disheveled segment polarity protein 2 (DVL2), cellular communication network factor 1 (CCN1/CYR61) and serum response factor (SRF), which have been previously implicated in embryo developmental competence. In addition, ingenuity pathway analysis revealed cell–cell contact function to be affected in both monosomy and trisomy 21 cumulus cells.
Conclusions
These findings support the use of cumulus cell gene expression analysis for the development of biomarkers evaluating oocyte quality for patients undergoing fertility preservation of oocytes.
). Investigations into factors affecting oocyte quality have been pursued to elucidate underlying mechanisms of this inefficiency. A target of these studies is the pre-ovulatory follicular environment, which has been shown to be critical in the development and condition of its corresponding oocyte (
A novel approach to quantifying ovarian cell lipid content and lipid accumulation in vitro by confocal microscopy in lean women undergoing ovarian stimulation for in vitro fertilization (IVF).
Investigators have reported that regulatory factors (such as amphiregulin) act on cumulus cells to promote secretion of factors (interleukin-7) from oocytes that in turn regulate cumulus cell proliferation and function, thereby completing a feedback regulatory loop resulting in competent oocytes. Therefore, the function of cumulus cells is thought to be a major determinant of oocyte quality (
); the oocyte, in turn, controls much of an embryo's preimplantation development. As cumulus cells are discarded after oocyte retrieval, they pose an attractive, non-invasive target for deeper study into the determinants of preimplantation embryo quality.
Studies have revealed an association between cumulus cell function and subsequent embryo development. For example, optimal cumulus cell metabolism is required to ensure the fertilizing capability of the oocyte (
). A study evaluating genome-wide gene expression in cumulus cells found seven genes, significantly altered at early cleavage stage: CCND2, CXCR4, GPX3, CTNND1 DHCR7, DVL3, HSPB1 and TRIM28) (
). Subsequently, several genes were found to be implicated in blastocyst stage development, including ANG (angiogenin), PLIN2 (perilipin 2), RGS2 (regulator of G-protein signalling 2) (
An integrated investigation of oocyte developmental competence: expression of key genes in human cumulus cells, morphokinetics of early divisions, blastulation, and euploidy.
). Several other studies have reported on cumulus cell gene expression and embryonic competence. Reduced expression of NFIB and upregulation of BCL2L 11 and PCK1 were associated with the cumulus cells of embryos yielding live births (
evaluated 13 genes and found VCAN, PTGS2, GREM1 and PFKP to be upregulated in cumulus cells of embryos leading to successful pregnancy. Notably, gene expression patterns were not predictive of embryo morphology.
analysed the expression of 11 genes, describing an association between pregnancy success and four genes: EFNB2, CAMK1D, STC1 and STC2. The expression patterns of these genes were not related to cleavage stage morphology or subsequent blastocyst development; however, other genes examined (TRPM7, ITPKA, STC2, CYP11A1 and HSD3B1), did show a relationship with these parameters. Prediction model based on the gene expression pattern of cumulus cells associated with embryos that resulted in live birth compared with those that did not were later created. FGF12, GPR137B, SLC2A9, ARID1B, NR2F6, ZNF132 and FAM36A were significantly upregulated in the cumulus cells of embryos that resulted in a successful pregnancy and several others were downregulated (ZNF93, RHBDL2, DNAJC15, MTUS1, NUP133) (
identified another 30 genes with differential expression associated with embryos yielding live births. Unfortunately, these prior studies evaluated pregnancy outcomes after multiple embryo transfer and failed to control for embryo ploidy status, limiting the applicability of the findings to current practice.
) evaluated fertilization and implantation rates after elective single embryo transfer in relation to cumulus cell gene expression. Microarray analysis followed by quantitative real-time polymerase chain reaction (RT-PCR) validation showed no significant difference in gene expression according to these outcomes. Green et al. (2018) used a paired design to evaluate the cumulus cell transcriptome of euploid embryos in double embryo transfers. Using preimplantation genetic testing for aneuploidy (PGT-A) and embryo fingerprinting, the investigators identified which of the two embryos that were transferred implanted and resulted in a livebirth and which one did not. They then compared cumulus cell transcriptome of embryos that resulted in a live birth to the transcriptome of the cumulus cell transcriptome of their siblings that failed to implant. No difference was identified (Green et al., 2018).
Preimplantation genetic testing for aneuploidy screening is used to select an embryo for transfer among a cohort of embryos. Since its inception in the early 1990s, PGT-A has undergone many iterations to reach its present-day form. Despite this, sustained implantation and live birth rates seem to have plateaued at around 65%, even for embryos tested using PGT-A and diagnosed as euploid (
Evidence of impaired endometrial receptivity after ovarian stimulation for in vitro fertilization: a prospective randomized trial comparing fresh and frozenthawed embryo transfer in normal responders.
Blastocyst biopsy with comprehensive chromosome screening and fresh embryo transfer significantly increases in vitro fertilization implantation and delivery rates: a randomized controlled trial.
Comprehensive chromosome screening of trophectoderm with vitrification facilitates elective single-embryo transfer for infertile women with advanced maternal age.
). The non-invasive nature of analysing otherwise discarded cumulus cells as a marker of embryonic competence is of interest. Overall, the extent to which the follicular environment correlates with subsequent embryonic ploidy status is largely unknown. A study evaluating cumulus cell gene expression and oocyte ploidy identified the under-expression of genes involved in DNA damage response (tumour protein p53 inducible protein 3 [TP5313]) and ubiquitination regulation (splA/ryanodine receptor domain and suppressor of cytokine signalling [SOCS] box containing 2 [SPSB2]) linked with the development of oocyte aneuploidy (
). A recent study evaluating cumulus cell gene expression according to the ploidy status of embryos found no difference between embryos diagnosed as euploid and aneuploid by quantitative polymerase chain reaction (PCR) (
An integrated investigation of oocyte developmental competence: expression of key genes in human cumulus cells, morphokinetics of early divisions, blastulation, and euploidy.
). Alternatively, reduced cumulus cell expression of HSD3B mRNA was found to be associated with a higher rate of euploidy in embryos in another recent investigation using PGT-A based on next-generation sequencing (NGS) (
Despite a consensus that cumulus cell function is vital to oocyte competence and, therefore, likely embryo competence, little consensus on the specific genes implicated has been achieved. Additionally, although many have evaluated genes associated with a successful pregnancy outcome, only a limited number of previously published studies specifically evaluated the difference in gene expression between euploid and aneuploid embryos. Most do not use a contemporary PGT-A platform, and, therefore, such comparisons are less relevant to current practice. Additionally, the chromosomal aneuploidies, e.g. multiple versus single whole chromosome aneuploidies, as well as specification of the chromosomes involved, are not explicitly indicated in published studies, possibly explaining the lack of consensus in genes implicated in aneuploidy. In humans, the most common aneuploidies are trisomies, some of which are not compatible with life. Of all the trisomies known, the most common human trisomy involves chromosome 21 (Rafii et al., 2019). As it is largely unknown that a diverse range of whole chromosome aneuploidies may be seen within a cohort of sibling embryos, evaluating the cumulus cell gene expression of embryos with one specific whole chromosome (in this case, chromosome 21) aneuploidy seems to be a logical starting point. Therefore, the present proof-of-concept study sought to determine if the cumulus cell transcriptomic profile of embryos predicted to be euploid by NGS-based PGT-A differs from that of embryos with a single whole chromosome aneuploidy.
Materials and methods
Ethics statement
Ethical approval was obtained on 17 May 2018 by the Institutional Review Board (Protocol RMA-2018-01) and registered with clinicaltrials.gov (NCT03604107). Eligible patients were contacted and counselled about the study procedures. Written informed consent was obtained from all participants before study enrollment.
Study population
Women aged between 18 and 44 years with normal ovarian reserve (FSH ≤12 IU/ml and antral follicle count ≥8) undergoing their first IVF and intracytoplasmic sperm injection cycle were screened for eligibility. Patients were excluded if they had a body mass index of 35 kg/m2 or above, or possessed a single gene disorder or structural chromosomal rearrangement requiring preimplantation genetic testing for monogenetic disorders or preimplantation genetic testing for structural chromosome rearrangements, respectively.
Study design
The present study evaluated the transcriptome of cumulus cells from individual oocytes obtained prospectively from consenting patients undergoing their first IVF cycle from August 2018 to November 2019 at Reproductive Medicine Associates of New Jersey (RMANJ), New Jersey, USA. Cumulus cells were collected as part of a larger prospective, blinded, non-selection trial evaluating the predictive value of an aneuploid PGT-A diagnosis using a targeted NGS-based assay (
). All oocytes and embryos were individually cultured to study cumulus cells associated with a specific aneuploidy. The PGT-A diagnoses were unblinded for all included embryos at the time of the present study.
Once PGT-A results became available, cumulus cells from trisomy 21 and monosomy 21 embryos were selected and compared with samples from euploid embryos. Non-sibling embryos were used for analysis for two reasons. First, the statistical likelihood of having all three genotypes in a single cohort of embryos (generated from a single retrieval) is extremely low. Second, this study aimed to investigate clinically relevant differences between aneuploidies and not differences related to individual variations without pathologic significance, such as genetic background, lifestyle and treatments.
In the present study, instead of comparing a mixed group of trisomies and monosomies, aneuploidies associated with chromosome 21 were compared. The reason for this was to increase homogeneity within the groups and decrease the likelihood of a type II error (non-rejection of a false null hypothesis). Chromosome 21 was selected as its aneuploidies (trisomy 21 and monosomy 21 are among the most common aneuploidies observed in blastocysts generated through IVF) (
). In addition, trisomy 21 is the most common chromosomal anomaly observed in humans (Rafii et al., 2019), whereas monosomy 21 is not compatible with life, suggesting a significant difference between their associated transcriptomes.
Routine clinical and laboratory procedures
All patients underwent ovarian stimulation with injectable recombinant or purified urinary gonadotrophins (150–450 IU total daily dose) in a protocol as per their primary physician's discretion. Stimulation protocols included gonadotrophin-releasing hormone antagonist (Ganirelix acetate 250 µg or cetrorelix acetate 0.25 mg) and micro-dose leuprolide flare cycles. Transvaginal ultrasound assessing for follicular growth as well as serum oestradiol and progesterone levels were obtained every 1–3 days. When at least two follicles measured 17 mm or wider, final oocyte maturation was induced by injecting either gonadotrophin-releasing hormone agonist, purified urinary HCG (10000 IU) or recombinant HCG (250 mg). About 36 h later, all follicles were aspirated via transvaginal oocyte retrieval. Cumulus oophorus complexes were isolated from the follicular fluid and each was labelled. To maintain identity in relation to the oocyte, supernumerous cumulus cells were removed before incubating the oocytes for later denudation by hyaluronidase in preparation for intracytoplasmic sperm injection (ICSI). Once oocyte identity was maintained, the cumulus cells were then aspirated from the media and centrifuged for 90 s at 15,000 g. The supernatant was removed from each set of cells, washed with phosphate-buffered saline and again centrifuged at 15,000 g for an additional 90 s. The supernatant was removed, and the labelled pellet was frozen at –80°C.
Oocytes were individually cultured to ensure accurate correlation with embryo development and associated outcomes. The embryos were cultured in the IVF laboratory following standard protocols. Intracytoplasmic sperm injection was carried out in all cases. After ICSI, oocytes were cultured in pre-equilibrated culture dishes with Quinn's Advantage Cleavage culture media drops, covered with OVOIL. The incubators were set to 37°C, 5.0% CO2 and 5.0% O2. Routine laboratory procedures included laser-assisted hatching of the zona pellucida, culture to the expanded blastocyst stage and daily morphologic assessment started on day 5. Additionally, all embryos considered usable, i.e. well-developed blastocoel cavity with adequate cellularity of the inner cell mass and trophectoderm, underwent trophectoderm biopsy followed by vitrification.
Preimplantation genetic testing for aneuploidy assay: library preparation and sequencing for next-generation sequencing based PGT-A
Next-generation sequencing based PGT-A (PGTseq-A) was used for PGT-A. A two-step PCR strategy was used to incorporate sequencing library adapters and indices for PGTseq-A, following the manufacturer's instruction (PGTseq Technology). The first PCR was carried out with 1 µl of PCR1 primer pool, 5 µl of PCR1 master mix, and 9 µl of molecular biological grade water with 10 µl of lysates. Applied Biosystems PCR machine (model number 2720) was used for PCR cycling and conditions were set to 10 min at 95°C, followed by 20 cycles of 15 s at 95°C, and 4 min at 60°C. The second PCR was set up to add indices to the amplified DNA by adding 2 µl of first PCR product, 4 µl of PCR2 master mix, and 16 µl of molecular biological grade water to the index plate. The reaction was set to 10 min at 95°C, followed by nine cycles of 15 s at 95°C, and 4 min at 60°C. Seventy-two samples were quantified, pooled and purified for one NextSeq 500/550 Mid Output Kit v2.5 (Illumina Inc, San Diego, CA, USA) using single 150 base pair reads. PGTseq-A software (PGTseq Technology) was used for copy number analysis.
Messenger RNA isolation and sequencing
The selected cumulus cells were thawed and assessed. Each sample contained about 100–200 cells. The SMART-Seq v4 Ultra Low Input RNA Kit for Sequencing was used to prepare RNA according to the user manual. The aim was to convert the mRNA into cDNA and use long distance PCR for amplification (17 cycles for 100 cells). Agilent Bioanalyser (Agilent Technologies, Santa Clara, CA, USA) was used before sequencing to test RNA integrity of samples. Samples meeting the stringent conditions for RNA quality, as represented by the RIN number (8 and above), and concentration, were processed for sequencing. The Agencourt AMPure XP Kit (Beckman Coulter, Brea, CA) was used to isolate cDNA and the Agilent High Sensitivity DNA Kit on an Agilent 2100 Bioanalyser (Agilent Technologies, Santa Clara, CA) was used for quantification. Nextera XT DNA Library Preparation Kit (Illumina Inc., San Diego, CA, USA) was used for library preparation.
Samples were sequenced to depths of up to 44 million read pairs, 75 nucleotide length reads per sample using the Illumina Rapid v2 kit (75 cycles) on an Illumina HiSeq2500 Sequencing System (Illumina Inc., San Diego, CA, USA). HiSeq Control Software v2.0 (NCS) and Real-Time Analysis Software (RTA) were used for image analysis, base calling and generation of sequence reads. Data were converted to FASTQ files using the bcl2fastq2 v1.8.4 software (Illumina Inc, San Diego, CA, USA.).
Data analysis
The reads were trimmed for quality and aligned with the reference human genome hg38 with GENCODE annotation (GENCODE reference annotation for the human and mouse genomes; Nucleic Acid Research, October 24, 2018). The normal annotation has about 50,000 entries, but the gencode annotation has over 100,000 annotated regions on the genome. HiSAT2 was used for alignment and StringTie and BallGown for transcript abundance estimation (transcript level expression analysis of RNA-sequencing experiments with HISAT, String Tie and Ball gown (
The genes were identified as differentially expressed if adjusted P ≤ 0.05. The statistical programme R was used for downstream processing and visualization of data. Three comparisons of cumulus cell gene expression were made: euploid versus monosomy 21, euploid versus trisomy 21 and monosomy 21 versus trisomy 21.
Ingenuity pathway analysis
Ingenuity Pathway Analysis (IPA) Ingenuity Systems (QIAGEN, content version: 51963813, 2020, Redwood City, CA, USA) was used to carry out pathway analysis for differentially expressed genes (DEG) across samples. Each gene symbol was mapped to its corresponding gene object in the Ingenuity Pathways Knowledge Base. The DEG used in pathway analysis were determined between experimental and control groups by using a filtering criteria fold change 1.5 or above and Benjamini–Hochberg (B–H) false discovery rate (FDR) 0.05 or lower (
). IPA Core Analysis was used to generate a network showing the overlap between functions and differentially expressed genes (fold-change ≥1.5, FDR P ≤ 0.05) resulting from the comparison between the different groups, in which FDR (or adjusted P-value) refers to the P-value (calculated using Fisher's exact test) that is used in the overrepresentation analysis. This analysis calculates the overlap (P ≤ 0.05) between the list of DEG and pathways to determine if subsets of genes associated with specific pathways are over-represented (or enriched) among DEG. The fold change refers to the cut-off used to identify a gene as differentially expressed, e.g. fold change modular value of 1.5 or above, and to be included in the pathway analysis. In addition, IPA calculates the z-score to infer the activation states (increased or decreased) of implicated pathways and biological functions. This inference is based on the experimentally observed causal relationships found in the biomedical literature between genes and those functions (
Real-time polymerase chain reaction was carried out to confirm differential gene expression of any genes identified as altered between monosomy and trisomy samples compared with euploid samples. RNA was amplified, and cDNA was prepared using established protocol of SMART-Seq v4 Ultra Low Input RNA Kit for Sequencing. Qubit was used to measure cDNA concentration. A total of 5 ng cDNA was used per reaction and RT-PCR using SYBR Green on the ViiA7 real time PCR machine (Applied Biosystems, Waltham, MA, USA) was used for gene expression analysis. The PCR reaction was prepared using 5 µl PowerUpTM SYBRTM Green Master Mix (2x) (Thermo Fisher, Waltham, MA, USA) 1 µl of (800 nM final concentration) forward and reverse primers, 1 µl (5 ng) of DNA template and 3 µl of nuclease-free water. The PCR cycling conditions were 95°C for 5 min followed by 45 cycles of 10 s at 95°C, 58°C for 15 s, and 72°C for 15 s. Hypoxanthine guanine phosphoribosyl transferase (HPRT) and beta-actin (ACTB) were used for normalization, based on a recent study published on human cumulus cells in which these genes were used as the reference genes (
Human Cumulus Cells in Long-Term In Vitro Culture Reflect Differential Expression Profile of Genes Responsible for Planned Cell Death and Aging-A Study of New Molecular Markers.
). To the best of our knowledge, no other quantitative RT-PCR studies have been conducted on cumulus cells with these specific aneuploidies, assessing HPRT and ACTB expression. In our dataset, no significant difference was found in the expression of these genes between the groups. ΔΔCt method was used for calculating the difference in the expression of genes. A melt curve analysis was subsequently carried out to confirm the specificity of products.
Results
Study population
A total of 24 cumulus cell samples collected from 24 unique, non-sibling oocytes from 23 different patients were evaluated. Of these, eight samples were associated with euploid embryos, eight associated with monosomy 21 and eight associated with trisomy 21. One patient donated cumulus cells from one euploid as well as one monosomy 21 embryo. All included patients obtained more than one embryo (range 2–26) during a single IVF cycle. Of note, all patients’ embryo cohorts included euploid and whole chromosome aneuploid embryos. RNA-sequencing alignment metrics of cumulus cells from one euploid embryo and one monosomy 21 embryo were noted to be poor, so the remaining 22 cumulus cell samples used for analysis included seven associated with euploid embryos, seven associated with monosomy 21 embryos and eight associated with trisomy 21 embryos.
Baseline and IVF cycle characteristics of the patients associated with the evaluated embryos are presented in Table 1. No statistically significant difference was found between the mean age of patients from whom euploid, monosomy 21 and trisomy 21 were derived (33.6 ± 4.7; 35.5 ± 4.5; and 36.3 ± 4.0, respectively). Similarly, no statistically significant difference was found between their body mass indices (25.2, 25.1 and 24.6 kg/m2, respectively). Overall, the percentage of good-quality embryos among euploids (75%) was higher than those among monosomic (12.5%) and trisomic (25%) embryos (P = 0.042). Similarly, euploid embryos were obtained from younger patients (33.6 ± 4.7) compared with monosomic and trisomic embryos (35.5 ± 4.5 and 36.3 ± 4.0, P = 0.037). All other characteristics were not statistically different between groups. Blastocyst quality is described according to the simplified Society for Assisted Reproductive Technology (SART) scoring system (
TABLE 1BASELINE AND IVF RETRIEVAL CYCLE CHARACTERISTICS OF INCLUDED PATIENTS
Patient baseline and cycle characteristics
Patients associated with euploid embryos (n = 8)
Patients associated with monosomy 21 embryos (n = 8)
Patients associated with trisomy 21 embryos (n = 8)
Mean female age, years
33.6 ± 4.7
35.5 ± 4.5
36.3 ± 4.0
Median BMI, kg/m2
25.2 (IQR 21.4–28.7)
25.1 (IQR 21.7–29.5)
24.6 (IQR 23.3–28.4)
Median antral follicle count
22.0 (IQR 13.7–24.6)
19.5 (IQR 12.4–22.8)
20.0 (IQR 16.8–23.8)
Indication for IVF, % (n)
Diminished ovarian reserve
25 (2)
25 (2)
0 (0)
Male factor
0 (0)
25 (2)
12.5 (1)
Tubal factor
0 (0)
12.5 (1)
12.5 (1)
Ovulatory dysfunction
25 (2)
37.5 (3)
12.5 (1)
Unexplained
50 (4)
0 (0)
37.5 (3)
Same sex or single
0 (0)
0 (0)
25 (2)
Mean total motile sperm count, million/ml
75.8 ± 55.7
60.1 ± 58.3
75.8 ± 18.1
Median number of usable embryos obtained
7.0 (IQR 3.7–8.2)
6.5 (IQR 4.4–9.6)
8.0 (IQR 5.3–11.2)
Mean proportion of euploid embryos obtained per cohort, %
53.9 ± 18.5
56.8 ± 10.5
54.1 ± 17.7
Proportion of blastocysts deemed:
Good quality, % (n)
75 (6)
12.5 (1)
25 (2)
Fair quality, % (n)
25 (2)
75 (6)
62.5 (5)
Poor quality, % (n)
0 (0)
12.5 (1)
12.5 (1)
Oocyte retrievals were carried out at a single, large infertility centre between 2018 to 2019. Means are expressed with (±) SD. Medians are expressed with interquartile ranges (IQR). BMI, body mass index.
RNA-sequencing defined repertoire of differentially expressed genes in cumulus cells associated with aneuploid embryos
Principal component analysis of RNA-sequencing data revealed substantial overlap between monosomy and trisomy and no clustering of the distinct euploid embryos (Supplementary Figure 1A). Qlucore Omics Explorer v3.6 software (QlucoreAB, Lund, Sweden) was used to generate three-dimensional principal component analysis. The differentially expressed genes (fold change ≥1.5, adjusted P ≤ 0.05) were used to distinguish the three different groups (Supplementary Figure 1B).
RNA-sequencing analysis revealed a total of 3122 genes that were significantly differentially expressed when comparing monosomy 21 with euploid cumulus cell gene expression. Genes with significant difference in expression in the monosomy patients compared with the euploid patients are presented in Supplementary Table 1. Of these 3122 identified genes, the associated protein products have been identified in all areas of the cell, i.e. cytoplasm, nucleus, plasma membrane, and extracellular space; however, no association has been found between most of the genes and a protein product. The regularized-logarithm transformation of gene counts was used to evaluate the Euclidean distances between euploid and monosomy 21cumulus cell samples. As shown in the heatmap (Figure 1a), a differential pattern of gene expression is identified between the euploid and monosomy 21 cumulus cells. The volcano plot (Figure 1b) shows the significantly changed expression of genes –log10 (P-value) in red.
Figure 1Heatmap, volcano plot and over-represented pathways of comparison between cumulus cells of euploid versus monosomy 21 embryos. (A) Heatmap of the sample-to-sample Euclidean distances displaying the significantly differentially expressed genes when comparing cumulus cells of euploid with monosomy 21 embryos (n = 3122 genes). The colour spectrum ranging from red to blue indicates normalized levels of gene expression from high to low. The dendrogram demonstrates hierarchical sample clustering. The significantly differentially expressed genes are listed in Supplementary Table 1; (B) volcano plot comparing the expression of genes in euploid with monosomy 21 cumulus cell samples. Red represents genes that reached statistical significance (–log10 [P-value]); (C) major pathways found to be significantly altered in monosomy 21 cumulus cell samples. The over-represented pathways are ranked according to the calculated Benjamini–Hochberg [B–H] false discovery rate, P < 0.05. Embryos are identified by their study number. E, euploid embryo; M, monosomy 21 embryo.
The comparison of euploid versus trisomy 21 cumulus cell gene expression revealed a total of 19 significantly differentially expressed genes (Supplementary Table 2). Of these, six genes (DVL2, SRF, CYR61, ZC3H4, MYO1E and CTNNB1) were associated with known protein products that have been localized to the cytoplasm (n = 2/6), nucleus (n = 3/6), and extracellular space (n = 1/6). A heat map showing the differences in expression of the identified genes according to ploidy status is presented in Figure 2a. The volcano plot shows the significantly increased expression of genes –log10 (P-value) in trisomy compared with euploid embryos (Figure 2b).
Figure 2Heatmap, volcano plot and over-represented pathways of comparison between cumulus cells of euploid versus trisomy 21 embryos. (A) Heatmap of the sample-to-sample Euclidean distances displaying the significantly differentially expressed genes when comparing cumulus cells of euploid versus trisomy 21 embryos (n = 19 genes). The dendrogram demonstrates hierarchical sample clustering; (B) volcano plot compares the expression of genes in euploidy versus monosomy 21 samples. Red represents genes that reached statistical significance (–log10 [P-value]); (C) ingenuity pathway analysis showed major pathways that were found to be significantly altered in trisomy 21 cumulus cell samples. The over-represented pathways are ranked according to the calculated Benjamini–Hochberg (B–H) false discovery rate, P < 0.05. E, euploid embryo; T, trisomy 21 embryo.
Ingenuity pathway analysis identified pathways over-represented in differentially expressed genes in cumulus cells associated with aneuploid embryos
The DEG were analysed by over-representation using the Ingenuity Pathway Analysis Knowledge Base. On the basis of the expression pattern of the DEG in the present data set, inhibition of several pathways, including the cyclic adenosine cyclic adenosine monophosphate (cAMP)-mediated pathway (B–H P = 0.005), was seen in the monosomy 21 samples based on the z-score value (Figure 1c). Significant changes in the expression of other pathways linked to embryonic developmental competence, such as the insulin-like growth factor 1 (IGF-1) (B–H P = 0.005) and Wnt/β-catenin-signalling pathways (B–H P = 0.009), was seen in trisomy 21 samples (Figure 2c). The activation status of some pathways, however, could not be predicted owing to the lack of data on the knowledge base for genes over-represented in these pathways, i.e. bars without colour in Figure 1c and Figure 2c.
On the basis of the results of IPA Core Analysis, we generated a network showing the overlap between functions and differentially expressed genes (fold change ≥1.5, FDR P ≤ 0.05) in various groups. When comparing monosomy 21 aneuploid with euploid cumulus cells, predictions show significant inhibition of important functions associated with embryo growth and development, such as cell-to-cell contact (P = 8.5e-161), cell-to-cell adhesion (P = 4.89e-32) and cell movement (P = 1.07e-18) (Figure 3a). Interestingly, when trisomy 21 was compared with the euploid cumulus cell samples, the cell-to-cell adhesion process was also significantly affected, although the effect was less prominent (Figure 3b).
Figure 3Ingenuity pathway analysis of differentially expressed genes in monosomy 21, trisomy 21 versus euploid groups. Ingenuity Pathway Analysis Core Analysis-generated network showing the overlap between functions and differentially expressed genes (fold change ≥1.5, false discovery rate [FDR] P ≤ 0.05) resulting from the comparison between monosomy 21 and euploid groups. P-value of the overlap is calculated by the fisher's exact test; (B) Ingenuity Pathway Analysis Core Analysis-generated network showing the overlap between functions and differentially expressed genes (fold change ≥1.5, FDR P ≤ 0.05) resulting from the comparison between trisomy 21 versus euploid cumulus cell samples. P-value of the overlap is calculated by Fisher's exact test. Green represents differentially expressed (downregulated) genes that passed the cut-off fold-change ≥1.5 (FDR P ≤ 0.05) when comparing monosomy 21 and euploid groups.
Significantly increased expression of CYR61/CCN1, DVL2 and SRF in cumulus cells associated with aneuploid embryos
Thirteen genes were identified with significantly differential expression common to both the monosomy 21 versus euploid and trisomy 21 versus euploid comparisons (Figure 4a). In aneuploid compared with euploid embryos, expression of three of these 13 genes are implicated in embryo developmental competence: cellular communication network factor 1 (CCN1/CYR61, adjusted P = 0.04), disheveled segment polarity protein 2 (DVL2, adjusted P = 0.04) and serum response factor (SRF, adjusted P = 0.01) (Supplementary Table 2). The box plot from RNA sequencing shows significantly increased expression of all three genes in the aneuploid samples compared with the euploid samples (Figure 4b).
Figure 4Venn diagram representing the overlapping genes in the comparison of euploid versus monosomy versus trisomy samples. Box plot and real time polymerase chain reaction graphs of the three of the overlapping genes. (A) Venn diagram demonstrating the number of significantly differentially expressed cumulus cell genes across all comparisons identified via RNA Seq analysis, i.e. euploid versus monosomy 21, euploid versus trisomy 21 and monosomy 21 versus trisomy 21. Numbers indicate the number of genes significantly differentially expressed in the indicated comparison; (B) box plot representation of three genes (of the 13 total overlapping genes differentially expressed in the euploid-monosomy and euploidy-trisomy comparisons as seen in (A) that significantly increased in expression in both the monosomy 21 and trisomy 21 samples. DESeq2 uses the Wald's test (modified chi-squared test) to determine statistical difference; (C) quantitative real-time polymerase chain reaction of DVL2, CYR61/CCN1, and SRF was carried out on euploid (n = 12), monosomy 21 (n = 9), and trisomy 21 (n = 9) cumulus cell samples to confirm RNA sequencing data. PCR results were consistent with RNA sequencing analysis, demonstrating significant increase in CCN1/CYR61 (P = 0.01monosomy versus euploid; P = 0.005 trisomy versus euploid), DVL2 (P = 0.02 monosomy versus euploid; P = 0.06 trisomy versus euploid). SRF showed increased expression approaching significant values in cumulus cells associated with monosomy 21 embryos (P = 0.1) and a significant increase was observed in trisomy 21 embryos (P = 0.007) as compared with euploid embryos.
Quantitative RT-PCR for CCN1/CYR61, DVL2 and SRF was conducted using euploid, monosomy 21, and trisomy 21 samples to validate the RNA-sequencing findings. Reference genes HPRT and ACTB were used as the internal controls. Quantitative RT-PCR confirmed a significant increase in CCN1/CYR61 (P = 0.01) in monosomy and trisomy samples (P = 0.005) compared with euploid cumulus cells. A significant increase of DVL2 was also observed in monosomy (P = 0.02) and trisomy (P = 0.06) Cumulus cells compared with euploid cumulus cells. SRF showed increased expression approaching significant values in cumulus cells associated with monosomy 21 embryos (P = 0.1) and a significant increase was observed in trisomy 21 embryos (P = 0.007) compared with euploid embryos (Figure 4c). Quantitative RT-PCR results confirmed the sequencing findings. Because of the limited amount of RNA available from cumulus cell samples, genes that showed a decreased or unchanged expression in individual aneuploidies compared euploids could not be included in the validation analysis. Therefore, the expression levels of the selected genes were verified with potential biological relevance that were differentially expressed (increased) in both monosomy 21 and trisomy 21 samples compared with euploids.
Discussion
The altered expression of several genes plausibly implicated in embryo development was identified upon comparison of cumulus cell transcriptomic profiles of euploid embryos compared with embryos with whole chromosome aneuploidy for chromosome 21 (monosomy and trisomy). A recent study by
, which investigated differentially expressed genes in embryos with chromosome 21 aneuploidies (monosomy 21 and trisomy 21), found that 1232 genes were differentially expressed in monosomy 21 embryos compared with euploid embryos, whereas no gene was differentially expressed in trisomy 21 embryos. Their findings are in parallel with ours, as we observed a high number of genes (n = 3122) that were differentially expressed in cumulus cells associated with monosomy 21, compared with only 19 genes in cumulus cells of oocytes that give rise to trisomy 21 embryos.
Specifically, enhanced expression of CCN1/CYR61, DVL2 and SRF were found in cumulus cells of whole chromosome 21 aneuploid compared with euploid embryos. These genes are known to be associated with protein products found in several different areas of the cell, e.g. in mouse embryos the DVL2 fusion protein expression in single blastomeres of four-cell stage embryo resulted in progeny that showed reduced cell adhesion and rounded shape. Therefore, it is possible that the differentially expressed genes act in concert with one another to produce the associated aneuploidy. This, unsurprisingly, implies that a multitude of factors influence oocyte ploidy. Additionally, the fact that a single gene was significantly differentially expressed between monosomy and trisomy cumulus cells compared with the many significantly differentially expressed genes identified between euploid and trisomy, monosomy 21 cumulus cells, or both, lends credence to the possibility that cumulus cell gene expression may be implicated in the determination of ploidy status of the embryo, especially as it relates to the segregation of chromosome 21.
The possibility that the follicular environment may affect meiotic events within the oocyte has long been suggested. As reviewed by
, optimal bi-directional communication between the oocyte and its surrounding cumulus cells is critical to the competence of the oocyte. Additionally, the follicular environment may influence oocyte ploidy via expression of genes involved in cellular metabolism, transport, signalling and apoptosis (Fragouli and Wells, 2014). Specifically, an altered expression pattern of several of the genes implicated in these critical events was identified in the cumulus cells in relation to euploidy, both in the present study as well as others. For example, protein SPSB2 plays an important role in cell signalling, and was significantly downregulated in cumulus cells from chromosomally abnormal oocytes (
These data reveal enhanced expression of three genes with previously described functions in the cumulus cells of aneuploid embryos: disheveled segment polarity protein 2 (DVL2), cellular communication network factor 1 (CCN1/CYR61) and serum response factor (SRF). Studies show that when DVL2-GFP fusion proteins are overexpressed in single blastomeres of the four-cell stage embryo, the progeny of the affected cells show reduced cell adhesiveness and a rounded shape at the blastocyst stage. This suggests that perturbing DVL2 function interferes with cell–cell adhesion through the non-canonical Wnt pathway in blastocysts (
). DVL2 encodes a protein that undergoes post-translational phosphorylation to form a 95 kDa cytoplasmic protein after, which possibly plays a role in the WNT signalling pathway. The WNT/β–catenin signal transduction pathway is involved in the regulation of cell proliferation and thought to be critical in mediating normal folliculogenesis, as described by
. CCN1/ CYR61 encodes for a protein localized within the extracellular space that promotes endothelial cell adhesion to integrins and heparan sulfate proteoglycans, and is required for angiogenesis and tissue repair. Other functions of this protein product include cell proliferation, differentiation, apoptosis and extracellular matrix formation (
). A previous study showed that SGHPL-5 trophoblast cells treated with the glycosylated forms of recombinant CCN1 and CCN3 decreased cell proliferation by bringing about G0/G1 cell cycle arrest (Kipkeew et al., 2016). Finally, SRF is localized to the cell nucleus and is instrumental in cell–cycle regulation, by binding to target genes’ serum response elements. SRF plays a role in apoptosis, cell growth and cell differentiation (
). Increased expression of SRF promoted reprogramming in NPC cell lines and other cell lines, such as hepatoblasts and ureteric bud cells, and promotes the loss of original cell identity (
), which may translate to altered IVF laboratory outcomes, although further study is needed. Additionally, decreased SRF has recently been described in malignant thyroid tumours (
Clinical Significance and Potential Regulatory Mechanisms of Serum Response Factor in 1118 Cases of Thyroid Cancer Based on Gene Chip and RNA-Sequencing Data.
Interference of neuronal activity-mediated gene expression through serum response factor deletion enhances mortality and hyperactivity after traumatic brain injury.
), further suggesting its important role in regulation of cell proliferation.
The IPA analysis showed inhibition of multiple pathways, including the cAMP-mediated signalling pathway, which has demonstrated utility in trophoblast fusion (
) have established roles in human embryo development and showed altered activity in cumulus cells from trisomy 21 patients in the present study. Taken together, aberrant expression of these genes and pathways would logically adversely affect embryonic development, perhaps including early meiotic events leading to aneuploidy.
Without further validation, the present findings have limited utility for clinical practice. Several potential applications for this technology, however, exist. For example, analysis of the cumulus cell transcriptome may better inform the quality of retrieved oocytes for patients undergoing oocyte vitrification for fertility preservation. Such information may encourage patients to undergo additional cycles if the transcriptomic profiles are associated with whole chromosome aneuploidy. Moreover, using otherwise-discarded cumulus cell material for such a test does not require exposure of the oocyte to conditions outside of the incubator. Such practice contrasts with other proposed tests of oocyte quality, such as polar body biopsy (
). Additionally, an improved understanding of the cumulus cell transcriptome may lead to enhanced in-vitro maturation techniques, potentially improving the efficiency of IVF cycles in women of all ages. Furthermore, although PGT-A via trophectoderm biopsy is currently the standard for preimplantation assessment of embryo ploidy, evaluation of cumulus cell gene expression has the potential to serve as a non-invasive marker of ploidy, possibly allowing for enhanced embryo selection for patients choosing not to proceed with PGT-A.
The present study has several strengths, including the identification of altered expression of cumulus cell genes with previously described major regulatory functions in cell processes, lending relevance to the present findings. Confirmation of the identified genes using RT-PCR is an additional important step to ensuring the accuracy of the present analyses. Additionally, evaluation of the cumulus cell transcriptomes associated with euploid embryos versus embryos with a single whole chromosome aneuploidy is a unique assessment not previously reported with associated novel findings. In contrast to many previously published studies, a contemporary PGT-A platform was used for copy number assessment in the present study.
Limitations of the present analysis relate to its small sample size and design as a proof-of-concept study. As such, replication of these analyses in a larger cohort of embryos is necessary to confirm the findings. Additionally, evaluation of the cumulus cell transcriptome within and across patient cohorts is required to gain a deeper understanding of cumulus cell gene expression patterns associated with varying chromosome copy number. It is possible that the chromosome assessed in association with embryonic aneuploidy in the present study, chromosome 21, has a unique cumulus cell transcriptome. Analysis of other whole chromosomal aneuploidies is warranted. In addition, the comparison between the aneuploid embryos compatible and incompatible with birth is worth pursuing. Unfortunately, that comparison was beyond the scope of the present study. Evaluating the cumulus cell transcriptome of embryos with PGT-A predictions of mosaicism and segmental aneuploidies in correlation with clinical outcomes may also be of interest in future studies.
Although promising, transcriptomic profiling is presently costly of both time and financial resources. Additionally, many of the techniques and analyses involved require highly trained personnel (
). As such, cumulus cell transcriptome profiling is not practical for every patient undergoing oocyte retrieval at the present time, but future technologies may reduce the cost and increase the ease with which such procedures are carried out. Further research is, therefore, needed to evaluate whether whole chromosome 21 aneuploidy, euploidy, or both, can be reliably determined via the differential expression of the genes identified in the present investigation. Notably, a non-selection study, in which the cumulus cell transcriptome is analysed after the clinical outcome of the embryo transfer is known, will need to be conducted for clinical validation of the identified cumulus cell gene expression patterns associated with ploidy status before this technology can be used in clinical practice.
Acknowledgements
The authors would like to thank the embryologists at Reproductive Medicine Associates of New Jersey for their diligent work in collecting the samples used in the present study. This study was funded by the Foundation for Embryonic Competence and IVI-America.
Human Cumulus Cells in Long-Term In Vitro Culture Reflect Differential Expression Profile of Genes Responsible for Planned Cell Death and Aging-A Study of New Molecular Markers.
Interference of neuronal activity-mediated gene expression through serum response factor deletion enhances mortality and hyperactivity after traumatic brain injury.
Clinical Significance and Potential Regulatory Mechanisms of Serum Response Factor in 1118 Cases of Thyroid Cancer Based on Gene Chip and RNA-Sequencing Data.
An integrated investigation of oocyte developmental competence: expression of key genes in human cumulus cells, morphokinetics of early divisions, blastulation, and euploidy.
Comprehensive chromosome screening of trophectoderm with vitrification facilitates elective single-embryo transfer for infertile women with advanced maternal age.
Blastocyst biopsy with comprehensive chromosome screening and fresh embryo transfer significantly increases in vitro fertilization implantation and delivery rates: a randomized controlled trial.
Evidence of impaired endometrial receptivity after ovarian stimulation for in vitro fertilization: a prospective randomized trial comparing fresh and frozenthawed embryo transfer in normal responders.
A novel approach to quantifying ovarian cell lipid content and lipid accumulation in vitro by confocal microscopy in lean women undergoing ovarian stimulation for in vitro fertilization (IVF).
Ashley W Tiegs, MD, graduated from Wake Forest School of Medicine, joined the Jefferson-RMA fellowship programme and completed her residency training in obstetrics and gynaecology at NYU School of Medicine. She works at Atlanta Center for Reproductive Medicine, has received awards for her research on PGT-A and published several studies.
Key message
Transcriptomic signature difference in the cumulus cells of euploid and chromosome 21 aneuploid embryos were identified. Key genes, DVL2, CCN1/CYR61 and SRF, implicated in embryo developmental competence, changed in monosomy and trisomy. Ingenuity Pathway Analysis revealed that cell–cell contact function was affected in monosomy and trisomy 21 cumulus cells.
Article info
Publication history
Published online: June 25, 2021
Accepted:
June 16,
2021
Received in revised form:
May 19,
2021
Received:
March 23,
2021
Declaration: ES is a consultant, for and receives research funding from, the Foundation for Embryonic Competence. He is also cofounder and a shareholder of ACIS LLC and co-holds patent US2019/055906 issued for using electrical resistance measurement for assessing cell viability and cell membrane piercing.
00299-6&id=ga1.jpg){kind=link}
00299-6&id=gr1.jpg){kind=link}
00299-6&id=gr2.jpg){kind=link}
00299-6&id=gr3.jpg){kind=link}
00299-6&id=gr4.jpg){kind=link}