Difference in mitochondrial gene expression in granulosa cells between recombinant FSH and hMG cycles under in vitro fertilization and transfer

Abstract

Purpose

Examination of the mitochondrial mRNA expression in granulosa cells from an unspecified population of infertile patients to evaluate whether recombinant follicle stimulating hormone (recFSH) is more effective in producing higher quality embryo rates compared with human menopausal gonadotropin (hMG).

Method

Thirty‐nine patients who underwent the in vitro fertilization and embryo transfer program were retrospectively examined. Patients were administered recFSH (n = 18) or hMG (n = 20) in a long protocol where GnRH agonist was used. Granulosa cells were obtained during oocyte retrieval and examined for mitochondria mRNA expression ratio against GAPDH. Expressions of mitochondria mRNA were evaluated by real‐time PCR analysis.

Results

The high‐quality embryo rate in the hMG cycle was higher than in the recFSH cycle, and the total dose of hMG showed a positive correlation with the expression level of mitochondrial genes in granulosa cells. Moreover, mitochondria mRNA expression was higher in the hMG cycle than in the recFSH cycle.

Conclusions

Compared with recFSH, hMG induces a higher mitochondrial gene expression ratio in granulosa cells at the time of oocyte retrieval and, therefore, may lead to higher quality embryo rates.

Introduction

In eukaryotic cells, mitochondria are special organelles responsible for the synthesis of ATP [1]. In recent years, it has been reported that the ATP producing capacity of germ line cells is correlated to reproductive performance, and that the mitochondria in granulosa cells produce and store all of the energy required for cell development [2]. With follicle aging, the production of growth factor in granulosa cells declines, while the apoptosis rate increases. The incidence of apoptotic bodies has been regarded as a predictive marker of assisted reproductive technology (ART) outcome, and the appearance of apoptosis in aspirated granulosa cells has been postulated to reflect the competency of the oocyte in terms of pregnancy rates [3]. In recent years, it has been reported that deleterious mitochondrial DNA rearrangements cause cellular energy deficiencies [4] and result in clinical disorders [5], and women over the age of 38 have granulosa cells containing substantially lower levels of normal mitochondria, as compared with women younger than 34 years of age [6]. The mechanism by which apoptosis occurs in granulosa cells in women of advanced reproductive age is not yet known. As well, the increase of apoptosis may be associated with an increase in mitochondrial DNA (mtDNA) deletions, which have been thought to be a major cause of cellular aging and the resulting decline in mitochondrial oxidative phosphorylation [7, 8]. Although recombinant follicle stimulating hormone (recFSH) is now being used for the treatment of infertility, recent meta‐analysis of 20 trials published in 2003 showed no significant difference in pregnancy rates per cycle between urinary and recombinant gonadotropins in long protocols [9]. Moreover, two large systematic reviews which evaluated the effectiveness of human menopausal gonadotropin (hMG) against recFSH in women undergoing ovarian stimulation for IVF have demonstrated a significant increase in live birth rates with hMG over those with recFSH [10]. Since the release of recFSH, many clinical studies have been undertaken to compare the effectiveness of recFSH with that of hMG. Until now, however, only clinical outcomes have been discussed. In order to evaluate the effectiveness of hMG and recFSH in women under controlled ovarian hyperstimulation (COH), we paid particular attention to the mitochondrial gene expression in granulosa cells, based on the hypothesis that it might be a viable alternative by which to evaluate the follicle environment. We therefore performed real‐time PCR to quantify the expression of mitochondria mRNA in granulosa cells harvested at the time of oocyte retrieval under COH in both the hMG and recFSH cycles.

Materials and methods

Patient population

From 2005 to 2011, granulosa cells were collected after oocyte pick‐up (OPU) from 38 patients of various ages and infertility etiologies who underwent IVF in our outpatient clinic.

Thirty‐eight patients of primary sterility, whose major cause of infertility is shown in Table 1, were treated with recFSH (n = 18) or hMG (n = 20) in a long protocol where gonadotropin releasing hormone (GnRH) agonist was used (Table 1). Before the OPU, the patient's endometriosis and unexplained infertility were confirmed by laparoscopy, and any endometrial cysts were simultaneously removed. The primary aim of this study was to compare the mitochondrial mRNA expression ratio in granulosa cell in an unspecified population of infertile patients. The study was a retrospective cohort study, and our institutional review board approved the protocol and its consent form. Moreover, we received permission from all participants.

Table 1.

Characteristics of patients in the agonist and antagonist groups

Diagnosis	hMG (n = 20)	recFSH (n = 18)	p
Endometriosis	4 (20 %)	2 (11.1 %)
Tubal infertility	7 (35 %)	5 (27.8 %)
Male infertility	3 (15 %)	6 (33.3 %)
Unexplained infertility	3 (15 %)	3 (16.7 %)
PCO	3 (15 %)	2 (11.1 %)
Type of fertilization
IVF	16 (80 %)	12 (66.6 %)
ICSI	4 (20 %)	6 (33.3 %)
Total treatment cycles	20	18
Age	35.5 ± 6.1	35.1 ± 4.3	NS
Total dose (IU)	3593.8 ± 1749.2	3566.7 ± 1429.6	NS
LH (IU/l) at the time of OPU	4.1 ± 7.5	0.5 ± 0.58	0.019
FSH (IU/l) at the time of OPU	17.5 ± 13.7	13.2 ± 7.8	NS
E2 (pg/ml) at the time of OPU	1903.9 ± 1594.2	1761.5 ± 1410.8	NS
P4 (ng/ml) at the time of OPU	1.7 ± 1.6	1.5 ± 1.2	NS
Number of retrieved oocyte	7.4 ± 4.0	8.5 ± 5.3	NS
Fertilization rate (%)	66.2 (98/148)	68.5 (111/162)	NS
High quality embryo rate (%)	27.5 (27/98)	16.3 (19/116)	0.003
Number of transferred embryo	2.2 ± 0.7	2.3 ± 0.8	NS
Endometrial thickness (mm)	11.2 ± 3.2	10.2 ± 3.7	NS
Implantation rate (%)	11.3 (5/40)	9.0 (4/44)	NS
Pregnancy rate (%)	22.5 (5/20)	22.2 (4/18)	NS
Multiple pregnancy rate (%)	0	0

Values are mean ± SD

NS not significant, OPU oocyte pick up

Protocol for COH

The GnRHa agonist (Leuprolide acetate, 1 mg subcutaneously, Leupline™; Takeda Ltd., Tokyo, Japan) was injected at the mid‐luteal phase of the preceding menstrual cycle. Before starting COH, pituitary suppression was confirmed with a baseline assessment to ensure that both serum luteinizing hormone (LH) and FSH levels were <5 IU/l. COH with hMG (Teizou HMG^®:150 IU FSH and 150 IU LH per ampoule; Asuka Pharmaceutical Co., Tokyo, Japan) or recFSH (Follistim^® 75 IU FSH per ampoule; Schering‐Plough Corporation, Kenilworth, NJ, USA) was started in an individual step‐down protocol until a dominant follicle reached a diameter of 18 mm. Ovulation was triggered with 10,000 IU of human chorionic gonadotropin (hCG) (Pregnyl; Schering‐Plough Corporation, Kenilworth, NJ, USA) which was administered 35 h before oocyte retrieval. Follicular maturation was monitored by an ultrasound examination (Sonovista‐EX; Mochida, Tokyo, Japan) from day 6 to the day of the hCG administration. Hormone assays, follicle monitoring, oocyte retrieval, insemination, embryo culture and embryo transfers were performed as previously reported [11]. Embryo quality was evaluated based on Veek's classification [12], and pregnancy was confirmed during an ultrasound examination by the identification of an intrauterine gestational sac. In this study, “implantation” means the percentage of embryos which were implanted, compared to the number of embryos transferred, and “pregnancy rate” means the percentage of pregnancies achieved in relation to the number of embryo transfer cycles.

OneStep realtime PCR

After the retrieval of oocytes, granulosa cells in the aspirated follicular fluid from mature follicles larger than 18 mm in diameter in individual patients were isolated and suspended in a tube according to the method reported by Toya et al. [13]. The pellet was washed with phosphate‐buffered saline and centrifuged at 1000 rpm for 30 min. After removing the supernatant, the pellet was suspended with 4 mL of phosphate‐buffered saline and layered carefully on 6 mL of Ficoll‐Paque (Amersham Pharmacia Biotech, Uppsala, Sweden). After centrifugation at 1000 rpm for 30 min, the granulosa cell layer was transferred to another tube for total RNA extraction. Total RNA was extracted using RNA STAT 60 (TEL‐TEST, INC., Friendswood, TX, USA), followed by a series of phenol–chloroform extractions and ethanol precipitations. Contaminating residual genomic DNA was removed by digestion with RNase‐free DNase (Promega Corp., Madison, WI, USA). Complementary DNAs (cDNAs) were prepared with at least 2 μg of total RNA and Superscript II reverse transcriptase (Gibco Brl Life Technologies, Inc., Gaithersburg, MD, USA), with random hexamers as primers, and were finally dissolved in diethyl pyrocarbonate‐treated water and frozen at −20 °C until use.

Oligonucleotide primers for TaqMan probes (Parkin‐Elmer Applied Biosystems, Tokyo, Japan) were designed with the use of Primer Express (version 1.0, Parkin‐Elmer Applied Biosystems, Tokyo, Japan) from the GeneBank database, based on published sequences of mitochondrial DNA, and primers for a specific sequence of mitochondrial DNA were chosen. As internal standards, human GAPDH was purchased from Perkin‐Elmer Applied Biosystems. The primers used in this study and their expected sizes from the reported complementary DNA sequences are shown in Table 2. The cDNA template was amplified by real‐time PCR in a 20 μl reaction containing 1× TaqMan Universal PCR Master Mix (Perkin‐Elmer Applied Biosystems), 200 nM of forward and reverse primer, and 100 nM of TaqMan probe. TaqMan PCR conditions were 95 °C for 15 s, followed by 60 °C for 1 min, for 45 cycles in each case on real‐time PCR (OneStep realtime PCR: Parkin‐Elmer Applied Biosystems, Tokyo, Japan). Amplification of mitochondrial mRNA relative to GAPDH was compared using the ΔΔC _t method.

Table 2.

Primers

Gene	Primer	Temp (°C)	Product size (bp)
Mitochondria (granulosa cells)	5′‐CCTCTAGAGCCCACTGTAAAGCTAAC‐3′ 5′‐TTTAGTTGGGTGATGAGGAATAGTGTA‐3′	60.0	148
Mitochondria TaqMan probe (FAM‐MGB)	5′‐CTACCGTATGGCCCACC‐3′	70.0

Statistical analysis

All experiments were performed in duplicates. Statistical calculations were performed with Stat View statistical software (SAS Institute Inc., Cary, NC, USA), and the statistical significance of each difference was determined using the Mann–Whitney U test or χ² tests. The Spearman's rank correlation coefficient was also used to analyze the relation between two different values. A p < 0.05 was considered statistically significant.

Results

There was no statistical difference of the total dose of injection in both groups. Although the serum LH level was higher in the hMG cycle than in the recFSH cycle (p = 0.019), serum P4 revealed no statistical difference. The level of serum E₂, the number of retrieved oocytes, endometrial thickness, the pregnancy rate and the implantation rate showed no statistical difference. The fertilization rate showed no statistical difference; however, high‐quality embryo ratios (Grade I + Grade II) in the hMG cycle were statistically higher than in the recFSH cycle (p = 0.03) (Table 1). The total dose of hMG showed a positive correlation with the expression level of mitochondrial genes in granulosa cells (Fig. 1); however, no such correlation was found with recFSH (Fig. 2).

Figure 1.

Mitochondrial gene expression ratio in patients who received hMG showed positive correlation with the total dose of hMG

Figure 2.

There was no correlation between total dose of recFSH and mitochondrial gene expression ratio

As well, the mitochondrial gene expression ratio in patients who received hMG was statistically higher than that in patients under recFSH (Fig. 3). No patients experienced ovarian hyperstimulation syndrome when hospitalization became necessary.

Figure 3.

Mitochondrial gene expression ratio in patients who received hMG was statistically higher than that of patients under recFSH

Discussion

Ovarian stimulation to retrieve oocytes has played a key role in ART, and many clinical trials to compare the effectiveness of hMG and recFSH in COH in IVF have been reported [9, 10]. Previous studies have claimed a better performance for recombinant gonadotropins and, moreover, it has been suggested that recFSH containing more relatively basic isoforms allows for a better interaction between FSH and its receptor [14]. Out et al. described that COH with recombinant FSH leads to statistically and significantly higher ongoing pregnancy rates compared with those with hMG [15]. In addition, Mannaerts et al. [16] reported that recFSH bioactivity tends to be higher than that of hMG, whereas immunoactivities display an opposite trend. Furthermore, more FSH basic isoforms, with lower sialic acid moieties, are produced in the preovulatory phase under higher estrogen levels [17]. On the other hand, other studies have revealed that hMG has a longer half‐life with decreased clearance rates due to the higher moieties of sialic acid.

Al‐Inany et al. [18] reported that there was no statistically significant difference in clinical pregnancy rates per cycle between recFSH and hMG. In a recent Cochrane library, four randomized trials comparing recFSH against hMG for ovulation induction in women with polycystic ovarian syndrome showed insufficient evidence to recommend one of these treatments over the other [19]. However, the different hormone compositions of recFSH and hMG used for COH had a statistically significant impact on the gene expression profile of preovulatory granulosa cells [20].

For these reasons, analyses of the molecular events occurring during granulosa cell development are pivotal to our understanding of how these cells contribute to the modulation of processes critical for oocyte development.

The frequency of apoptotic bodies has been considered to be a predictive marker of ART outcomes. Human mtDNA is a circular, histone‐free molecule composed of 16.6 kb of DNA, and exists in one or more copies in every mitochondrion and encodes 37 genes in the oxidative phosphorylation pathway [2]. In humans, abnormally fertilized oocytes, as well as embryos arrested to grow in culture, demonstrate a high rate of mtDNA deletions, indicating the crucial role that mitochondria play in oocyte function and early embryo development [1, 21]. Recent studies have shown a relationship between GC apoptosis and ART outcomes, but the data are still controversial [22].

The mechanism by which apoptosis occurs in granulosa cells is still unknown [1, 21]; however, mtDNA deletions and low expressions might cause a decline of mitochondrial oxidative phosphorylation [6]. Fewer granulosa‐lutein cells are apoptotic in women who have an ongoing pregnancy after IVF treatment than in women who do not conceive [23, 24]. Women over the age of 38 have granulosa cells that contain substantially lower levels of normal mitochondria as compared with women 34 years or younger [23, 24]. Moreover, Fatum et al. [25] hypothesized that during the process of reproductive aging, mitochondrial membrane potential may deteriorate as a result of free‐radical accumulation.

Lin [26] reported that using real‐time PCR to evaluate mtDNA copy numbers is more practical than previously popular methods for reducing ATP‐producing capacity parameters. The LH‐induced 30 kDa phosphoprotein directly results in an increase in steroid biosynthesis, and this protein is required in the acute regulation of steroidogenesis in mitochondria. As a result, the necessity of LH has been looked over again [27].

We evaluated the expression level of mitochondrial genes in granulosa cells to discover what impact the two drugs have on mitochondrial gene expression. Our results revealed that there is a positive correlation between the total dose of hMG and the expression level of mitochondrial genes in granulosa cells. Moreover, the high quality embryo rate was higher in the hMG cycle compared with the recFSH cycle in proportion to mitochondrial gene expression.

These results may be explained by the fact that recFSH has no LH activity. Ruvolo [28] suggested that supplementation with recombinant LH (recLH) during the proliferative phase improves some clinical parameters. In particular, the supplementation of recLH greatly reduces the number of picked‐up immature oocytes, thus increasing the number of transferred embryos, which is negatively correlated with the risk of OHSS. In addition, the increase in pregnancy rates and implantation rates may be correlated with the reduction of apoptosis seen in the cumulus cells of patients treated with recLH [28].

Although the mitochondrial gene expression ratio in patients given hMG was statistically higher than that of patients given recFSH in our study, further investigation is necessary to study the relationship between LH and mitochondrial gene expression in granulosa cells. Previously we reported that mitochondrial mRNA expression compensatively elevated in the granulosa cells of patients with severe endometriosis; however, in this study, very few cases of severe endometriosis were involved [24]. Thus, more study is necessary to determine how directly the expression of mitochondrial genes affects oocyte quality and IVF‐ET outcome. In particular, it may be necessary to investigate whether mitochondrial gene expression is influenced by the actual size of the follicle to confirm the relationship between follicle and mitochondrial gene expression in granulose cells.

In conclusion, this study showed no apparent difference in pregnancy rates between recFSH and hMG. However, hMG does induce a higher mitochondrial gene expression ratio in granulosa cells at the time of oocyte retrieval and, thus, may lead to higher quality embryo rates.

Acknowledgment

We have no conflict of interest.