https://orcid.org/0000-0002-7114-0970
Abstract
Insulin-like growth factor (IGF) signaling is known to affect human ovarian follicular function during growth and development. However, the role of the IGF system is unknown during the ovulatory peak, which is characterized by profound changes in granulosa cell (GCs) mitosis and function.
How is the IGF system expressed and regulated during the midcycle surge in women?
Follicular fluid (FF) and GCs were collected during the ovulatory peak from 2 specific time points. One sample was obtained before oocyte pickup (OPU): before ovulation trigger (OT) (T = 0 hours) or at 12, 17, or 32 hours after OT, and 1 sample was obtained at OPU 36 hours after OT. Fifty women undergoing ovarian stimulation at a university hospital were included. Gene expression profiles were assessed by microarray analysis of GCs. IGF-related proteins in the FF were assessed by immunoassay or by determination of activity with a proteinase assay.
Gene expression of proteins promoting IGF activity (ie, IGF2, PAPP-A, and IRS1) together with proliferation markers were downregulated on a transcriptional level in GCs after OT, whereas proteins inhibiting the IGF signal (ie, IGFBPs, IGF2, and STC1) were upregulated. STC1 gene expression and protein levels were greatly upregulated after OT with a parallel steep downregulation of PAPP-A proteolytic activity.
These data suggest that downregulation of IGF signaling mediated by increased STC1 expression is instrumental for the sudden cessation in GC proliferation and onset of differentiation during the ovulatory peak.
The insulin-like growth factor (IGF) signaling pathway plays a pivotal role in cell proliferation and survival in a variety of tissues and organs, including the ovaries (1-4). The pathway involves several levels of extracellular regulatory proteins that finely regulate IGF bioactivity. The IGF signaling pathway comprises the 2 ligands IGF-1 and -2, which share structural similarities with proinsulin. IGFs bind with different affinities to and transduce their signal through the IGF1 receptor (IGF1R), the insulin receptor (INSR), and hybrids of these (5-7). Both IGF1R and INSR belong to the receptor tyrosine kinase family, and their activation by IGFs leads to the phosphorylation of various intracellular proteins, including insulin-receptor substrate-1 and -2 (IRS1 and IRS2) (8, 9). These phosphorylated IRS proteins initiate downstream signaling cascades, such as the Ras/Raf/Erk pathway and the phosphatidylinositol-3 kinase (PI3 K)/Akt pathway, ultimately promoting cell proliferation, metabolism, and growth (8-10).
The bioactivity of IGFs is tightly controlled by 6 extracellular IGF binding proteins (IGFBPs) that can neutralize or prolong the half-life of IGFs (11-13). Additionally, soluble IGF2R functions as an extra binding protein, reducing IGF bioavailability. Moreover, membrane-bound IGF2R can internalize IGFs and transport them to lysosomes (14).
Several proteolytic enzymes have been identified that in vitro release bound IGF through the proteolytic cleavage of IGFBPs, thereby facilitating IGF signaling (15). However, only a subset of these proteolytic enzymes has been investigated in vivo (16). One notable enzyme in this context is pregnancy-associated plasma protein-A (PAPP-A) (17), which exhibits robust proteolytic activity toward IGFBP-2, -4, and -5 (18-21). PAPP-A can reversibly bind to target cells through heparan sulphate proteoglycans, resulting in a unique, locally confined release of IGF adjacent to IGF receptors on the cell surface (22).
In addition, PAPP-A's proteolytic activity can be counteracted by 2 glycoproteins, stanniocalcin-1 and -2 (STC1 and STC2), which attenuates IGF bioactivity (23, 24). STC1 and STC2 have earlier been observed to play roles in, for example, calcium regulation, proliferation, carcinogenesis, and angiogenesis (25-28). STC1 forms a high-affinity noncovalent complex with PAPP-A (KD = 75 pM and Ki = 68 pM), while STC2 forms a covalent complex with PAPP-A (24, 29).
The entire IGF system is active within the human ovary, and its components exhibit differential expression and strict regulation during follicle development (30-33). Gene expression profiling of IGF components at various stages of human follicle development and functional studies both point toward an increase in IGF activity with growing follicle size (34-36). Particularly during the preovulatory stage, IGF activity appears to be highly significant (33, 35) and has been proposed as one of the primary factors contributing to the remarkable mitotic activity observed in the granulosa cells (GCs), leading to roughly 50 to 60 million GCs in the dominant follicle at ovulation (37).
When ovulation takes place, the high rate of GC proliferation comes to an end, the cells undergo transformation and luteinization, and contribute to the formation of the corpus luteum. The regulatory mechanism responsible for this rapid shift from proliferation to differentiation remains unknown.
The objective of the present study was to test the hypothesis that downregulation of IGF signaling is a crucial event in the transition from proliferation to differentiation during the midcycle surge of gonadotropins leading to ovulation, possibly explaining a key underlying mechanism. This was achieved by analyzing human GCs and follicular fluids (FFs) collected at 5 different time points during the ovulatory peak.
Materials and Methods
Human Ovarian Granulosa Cells and Follicular Fluid
FF and GCs were collected at 5 different time points during the ovulatory peak from 50 women receiving fertility treatment at the Fertility Clinic, Department of Gynaecology and Obstetrics, Holbæk Hospital, Denmark, as described previously (38, 39). The women underwent a standard antagonist protocol introduced at cycle day 2 or 3 with individually dosed recombinant follicle-stimulating hormone (n = 42, rFSH, Puregon, MSD) or human menopausal gonadotropin (n = 8, hMG, Menopur, Ferring) and a gonadotropin-releasing hormone (GnRH) antagonist (ganirelix 0.25 mg, Fyremadel, SUN Pharma) at stimulation day 5. Ovulation was triggered with recombinant human chorionic gonadotropin (rhCG; n = 17; 6500 IU; Ovitrelle, Merck Serono) in women in whom 14 or fewer follicles 12 mm or smaller developed, or with GnRH agonist (GnRHa; n = 33; buserelin, 0.5 mg; Suprefact, Sanofi-Aventis) when the woman developed more than 14 follicles 12 mm or larger or showed clinical signs of ovarian hyperstimulation syndrome. When at least 3 follicles reached 17 mm in diameter, final maturation of follicles was induced followed by oocyte pickup (OPU) 36 hours later. For ethical reasons, only women who had developed more than 8 mature follicles at their final control visit before ovulation trigger (OT) were approached and asked for possible participation. Each woman donated the content of 1 follicle at 1 specific time point prior to OPU: before OT (group 1, n = 23), 12 hours (group 2, n = 10), 17 hours (group 3, n = 6), or 32 hours (group 4, n = 11) after OT, and in addition, FF and GCs were collected from 1 follicle at OPU from all participants (n = 50). The women included in this study received fertility treatment due to either male factor infertility, tubal disease, or unexplained infertility, including 6 women with nonhyperandrogenic polycystic ovary syndrome (PCOS). Women with elevated androgens, diseases of the lung, heart, bowel, kidney, dysregulated thyroid disease, and women older than 35 years were excluded from this study.
Ethical Approval
The Scientific Ethical Committee of Region Zealand, Denmark, and the Danish Data Protection Agency approved the study (SJ-530). All study participants provided informed consent before inclusion, and the study was conducted in accordance with the Helsinki Declaration II (38).
Microarray Analysis
Microarray data from a previous study were extracted and analyzed (38), in which the expression profiles of genes related to the IGF system have not been presented.
The Arcturus PicoPure RNA Isolation Kit (Applied Biosystems) was used to isolate total RNA from the GCs according to the manufacturer's instructions. The quality and quantity of the purified RNA were analyzed with NanoDrop (Thermo Fisher) and the Bioanalyzer RNA 6000 Pico Kit (Agilent).
Based on RNA integrity number (RIN) and concentration, RNA was processed using the Clariom DTM Pico Assay (Applied Biosystems, ThermoFisher) according to the manufacturer's protocol. All arrays were washed and stained with phycoerythrin-conjugated streptavidin using the Affymetrix Fluidics Station 450, and afterward scanned in the Affymetrix Gene Array 30007 G scanner as previously described (38).
Cell intensity files (.CEL files) were generated with GeneChip Command Console Software (AGCC, Affymetrix, ThermoFisher). After initial RNA quality control analysis (ie, RIN test), 83 of the 100 samples qualified for further transcriptome analysis. Transcriptome Analysis Console (TAC 4.0.1, Thermo Fisher Scientific) was used to process the 83 CEL files, presenting the 5 time points; 0 hours (n = 17), 12 hours (n = 7), 17 hours (n = 6), 32 hours (n = 9), and 36 hours (n = 44), including data summarization, quantile normalization, gene summaries, and statistical analysis. Normalization was performed by the signal space transduction—robust multiarray average (SST-RMA) approach (Affymetrix 2019). Subsequently, fold changes were calculated from the SST-RMA–normalized data. The differential expression analysis was setup using analysis of variance eBayes comparisons with an advanced random factor for “patientID,” accounting for the pairing of samples. Differential expression with a false discovery rate (FDR) less than 0.01 combined with a gene expression fold change greater than 2 was considered significant. For the present study, only members of the IGF family together with 4 common proliferation markers were evaluated.
Proteinase Assay
The proteolytic activity of PAPP-A was measured in FF samples isolated at the 5 different time points (n = 25) during the ovulatory peak using a proteinase assay based on autoradiography as previously described in detail (18). The output measures from the proteinase assay are based on intact and cleaved radiolabeled IGFBP-4 products obtained after terminating the proteolytic reaction at 0, 1, 2, and 4 hours. Following separation by sodium dodecyl sulfate–polyacrylamide gel electrophoresis, the gels were dried and exposed to a storage phosphor screen (Molecular Dynamics) for a minimum of 20 hours. The screens were scanned on a Typhoon Trio scanner (GE Healthcare), and band intensities were quantified using the ImageQuant software package (GE Healthcare). The PAPP-A concentration was adjusted to 25 pM to achieve cleavage percentages below 30%, where a linear relationship between time and cleavage can be assumed, and the cleavage rates are hereby determined by the slopes (18).
Enzyme-linked Immunosorbent Assay Measurements
Quantification of total IGFBP-1, IGFBP-3, IGFBP-4, IGFBP-5, PAPP-A, and STC1 protein levels from FF samples collected from the 5 time points during the ovulatory peak was performed with commercial enzyme-linked immunosorbent assay (ELISA) kits and completed according to the manufacturer's instructions (catalog No. EHIGFBP1, RRID:AB_2608723, Thermo Fisher Scientific, and from Ansh Labs catalog No. AL-120, RRID:AB_2783671, catalog No. AL-126, RRID:AB_2783676, catalog No. AL-127, RRID:AB_2783677, catalog No. AL-101, RRID:AB_2783656, Ansh Labs, and catalog No. DY2958, RRID:AB_2893122, R&D Systems). FF were diluted prior to analysis in 1% bovine serum albumin in 1×phosphate-buffered saline for the following proteins: 1:1000 or 1:100 for IGFBP-1, 1:100 for IGFBP-3, 1:3 for IGFBP-5, 1:10 or 1:20 for STC1, and 1:10 for PAPP-A measurements. Measurement of total IGFBP-4 protein did not require FF sample dilution. 0 hours (n = 20), 12 hours (n = 9), 17 hours (n = 6), 32 hours (n = 11), and 36 hours (n = 33).
Statistical Analyses
R studio (version 1.4.1106) was used to perform the statistical tests on data not associated with the microarray analysis. To assess differential protein levels in FF between paired time points, a mixed-effect model was performed on transformed data. To account for patient factor, the “patient-ID” was included as a random factor in the statistical model and “time” as a repeated, fixed factor. To compare mixed-effect models, analysis of variance was applied.
A nonparametric Kruskal-Wallis test was used for analyzing cleaved IGFBP-4 levels. A P value below .05 was considered statistically significant.
Results
Microarray Analysis of Key Insulin-like Growth Factor Family Members During the Ovulatory Peak
Insulin-like growth factor 1 and 2
IGF2 was highly expressed in human GCs before OT and was subsequently observed to decrease steeply (∼1000-fold reduction) between 0 hours and 36 hours after OT (Fig. 1A and Supplementary Table S1 (40)). In contrast, the expression level of IGF1 in human GCs was low and did not differ between the 5 time points during the ovulatory peak (see Supplementary Table S1 (40)).
Gene expression profiles of insulin-like growth factor (IGF) members and proliferation markers during the ovulatory peak. This figure displays the key A to H, IGF genes, and I to L, proliferation markers that are differentially expressed during the ovulatory peak in human granulosa cells collected at 0 to 36 hours after ovulation trigger (OT). The mean expression intensity signal from the microarray is shown on the y-axis on a log2 scale. Error bars indicate SEM values. Statistical significance is shown in Supplementary Table S1 (40).
Insulin-like growth factor receptor type 1 and insulin receptor
The expression level of IGF1R was observed to increase significantly in human GCs 12 hours after OT, but thereafter the expression level was shown to decrease 17 hours after OT (see Supplementary Table S1 (40)). The expression levels of INSR did not show any change during the ovulatory peak.
Intracellular signaling proteins
The IRS1 and IRS2 genes showed opposite expression profiles in human GCs during the ovulatory peak with high levels of IRS1 and low levels of IRS2 before OT, and with a subsequent significant decrease of IRS1 and increase in IRS2 expression levels (Fig. 1B and Supplementary Table S1 (40)).
Insulin-like growth factor binding proteins and insulin-like growth factor receptor type 2
Four of the IGFBPs and the IGF2R were significantly differentially expressed in human GCs during the ovulatory peak (Fig. 1C-1F and Supplementary Table S1 (40)), especially IGFBP3, which increased 428-fold 12 hours after OT. The expression levels of IGFBP1, -3, -4, -5 and IGF2R increased significantly 12 and 17 hours after OT, whereas IGFBP2 and IGFBP6 did not differ between the 5 time points (see Supplementary Table S1 (40)).
Pregnancy-associated plasma protein-A
The expression level of PAPP-A in human GCs showed a tendency to decrease after OT (time point FDR = 0.0283), but increased again after 36 hours to the same levels as T = 0 hours (Fig. 1G and Supplementary Table S1 (40)).
Stanniocalcin-1 and -2
The expression level of STC1 in human GCs was observed to increase with a huge change of 43 000-fold between 0 hours and 12 hours after OT (time point FDR = 7.22E-23) (Fig. 1H and Supplementary Table S1 (40)). In addition, 17 hours after OT the expression level of STC1 was almost 9000-fold higher compared to 0 hours after OT. In contrast, the STC2 gene expression was low and did not differ between the 5 time points during the ovulatory peak (see Supplementary Table S1 (40)).
Proliferation markers (MKI67, E2F1, E2F8, and FOXM1)
All 4 proliferation markers, MKI67, E2F1, E2F8, and FOXM1, were observed to decrease significantly between 0 hours and 36 hours after OT (Fig. 1I-1L and Supplementary Table S1 (40)).
Intrafollicular Insulin-like Growth Factor Extracellular Binding Protein Levels During Ovulatory Peak
Total protein levels of IGFBP-1, -3, -4, and -5 did not show statistically significant differences in FF obtained from the 5 included time points during the ovulatory peak (Fig. 2). However, the total IGFBP-1 protein level shows a tendency to increase 12 hours after OT (NS) compared to 0 hours after OT and a subsequent decrease 17 hours after OT (P value = .02) (Fig. 2A), suggesting a strict regulation of IGFBP-1 shortly after the observed increase in transcriptional level (Fig. 1D). Total IGFBP-3 shows a tendency to increase after OT and peaks at 36 hours after OT (P value = .012) (Fig. 2B), indicating a delayed upregulation of the protein relative to transcription of the messenger RNA.
Total intrafollicular levels of insulin-like growth factor extracellular binding proteins (IGFBPs) during the ovulatory peak. This figure presents mean (±SEM) levels of total A, IGFBP-1; B, IGFBP-3; C, IGFBP-4; and D, IGFBP-5 in follicular fluid of preovulatory follicles during the ovulatory peak at 0 hours (n = 20), 12 hours (n = 9), 17 hours (n = 6), 32 hours (n = 11) and 36 hours (n = 33) after ovulation trigger (OT). Significant changes: overall: none. Total IGFBP-1 protein levels 0 to 12 hours: P value = .021; total IGFBP-3 protein levels 0 to 36 hours: P value = .012.
Intrafollicular Stanniocalcin-1 and Pregnancy-associated Plasma Protein-A Protein Levels During Ovulatory Peak
The substantial increase in STC1 gene expression level in GCs from preovulatory follicles 12 hours and 17 hours compared to 0 hours after OT was confirmed and manifested at a protein level from the corresponding FF samples (Fig. 3A), showing a 50- to 70-fold increase in STC1 protein level 12 hours (37.4 ± 5.7 ng/mL [mean ± SEM]) and 17 hours (50.7 ± 6.7 ng/mL [mean ± SEM]) after OT compared to 0 hours (0.7 ± 0.1 ng/mL [mean ± SEM]) after OT (P value < .001).
Intrafollicular levels of STC1 and PAPP-A in the granulosa cells (GCs) during the ovulatory peak. This figure presents mean (±SEM) of intrafollicular levels of A, STC1 and B, PAPP-A in preovulatory follicles during the ovulatory peak at 0 hours (n = 20), 12 hours (n = 9), 17 hours (n = 6), 32 hours (n = 11), and 36 hours (n = 33) after ovulation trigger (OT). Significant changes: STC1 protein levels overall P value less than .001; multiple comparisons of means: 0 to 12 hours P value less than .001; 0 to 17 hours P value less than .001; 0 to 32 hours P value = .09; 0 to 36 hours P value = .14.
PAPP-A protein levels did not show statistically significant differences in FF obtained from the 5 included time points during the ovulatory peak (Fig. 3B).
Cleavage rate of Radiolabeled Insulin-like Growth Factor Binding Protein-4 Decreases During Ovulatory Peak
Proteinase activity was assessed in the FF from the 5 included time points during the ovulatory peak to examine whether the observed increased STC1 is associated with inhibition of the proteolytic activity of PAPP-A, thus mirroring downregulation of IGF bioavailability.
FF samples collected after OT showed a significantly lower cleavage rate of radiolabeled IGFBP-4 compared to FF samples collected before OT (P value = .0026) (Fig. 4).
Proteolytic activity in preovulatory follicles during the ovulatory peak. This figure presents mean cleavage rates (pM/min) (±SEM) of radiolabeled recombinant IGFBP-4 (rIGFBP-4) when incubating with follicular fluid from preovulatory follicles collected 0 to 36 hours after ovulation trigger (OT). A significant lower IGFBP-4 cleavage rate was observed after OT (P value = .0026).
Discussion
This study presents a novel insight into the tightly interconnected dynamics of GC mitotic activity and IGF bioavailability within human ovarian follicles during the midcycle surge of gonadotropins. Notably, this investigation reports a significant downregulation of genes promoting IGF signaling, such as IGF-2, PAPPA, and IRS1, along with a concurrent upregulation of genes associated with IGF binding and inactivation, including IGFBPs, IGF2R, and STC1 (see Fig. 1). Particularly, the substantial increase in STC1 at both gene and protein levels (see Fig. 1H and 3A) likely plays a key role in the downregulation of IGF signaling. Additionally, this study reveals a marked reduction in intrafollicular PAPP-A proteolytic activity after OT (see Fig. 4), further emphasizing the role of PAPP-A as a central regulator of IGF bioactivity within human ovaries.
Furthermore, this study demonstrates a pronounced decline in GC mitotic activity during the midcycle surge, as evidenced by the downregulation of proliferation markers, including MKI67, E2F1, E2F8, and FOXM1 (see Fig. 1I-1L). These markers displayed expression levels just a few percentage points compared to baseline levels before OT (T = 0 hours), highlighting a sudden and substantial decrease in GC proliferation rate during this critical phase.
Previous research has demonstrated that IGF signaling stimulates the proliferation and differentiation of cultured human GCs (32, 41, 42). Therefore, the significant changes in IGF bioavailability and bioactivity observed in human GCs during the ovulatory surge underscore the essential role of IGF signaling in controlling the high proliferation rate and differentiation of GCs within the preovulatory follicle.
The ovarian follicle, being avascular and isolated from direct circulation, provides an ideal environment to study IGF signaling, but also highlights the complexity of the intricate regulation of IGF bioactivity in vivo. However, this study sheds light on the interaction between STC1 and PAPP-A within the human follicle. The presence of the STC1-PAPP-A complex in human FF isolated before and after OT indicates that PAPP-A activity is regulated within this compartment (33). This study extends these findings by revealing significant changes in STC1 gene expression and protein levels during the ovulatory surge (see Fig. 1H and 3A), suggesting an important function in regulating PAPP-A activity and, consequently, IGF bioactivity. The possibility of additional functions of STC1 within the human follicle, unrelated to PAPP-A inhibition, should be considered (16). STC1 has been reported to suppress inflammatory responses and protect against hypoxia, hypercalcemia, ischemic damage, and oxidative stress, all processes active within the follicle during ovulation (27, 43-45). Ovulation itself has been characterized as a local inflammatory process, marked by the upregulation of various inflammatory factors in the ovary (46, 47). Moreover, human GCs have exhibited characteristics of innate immune cells during ovulation (46), suggesting that STC1 may also play an anti-inflammatory role in the human ovary during ovulation.
The intrafollicular concentrations of total IGFBPs measured in this study during the ovulatory peak reveal that FF contains significantly higher concentrations of total IGFBP-1 and IGFBP-3 in comparison to total IGFBP-4 and IGFBP-5 (see Fig. 2) with approximately 60 times more IGFBP-1 than IGFBP-4 and 25 times more than IGFBP-5. This indicates that IGFBP-1 and -3 are functional IGFBP regulators within the human preovulatory follicle. However, it is important to note that various posttranslational modifications, including proteolysis, phosphorylation, and glycosylation, can alter the binding affinity of IGFBPs. For instance, the proteinase PAPP-A2 specifically cleaves IGFBP-3 and -5 (16), and PAPP-A2 activity is also inhibited by the STCs (24). PAPPA2 was not differentially expressed in this data set (data not shown). It is also important to keep in mind that PAPP-A cleavage of IGFBP-4 at the GC surface may represent a mechanism to ensure efficient delivery of bioactive IGF to these cells (48), potentially required for GC proliferation. Still, however, the results presented here should be interpreted cautiously, emphasizing that the total levels of IGFBPs alone cannot fully reflect IGF bioactivity and bioavailability. Additionally, the increase in gene expression of IGFBP-1, -3, -4, and -5 observed at 12 and 17 hours after OT did not necessarily translate directly to changes in protein levels. This may suggest that either a certain threshold in gene expression must be reached before protein levels are affected or that translational regulatory mechanisms are at play. Previous studies have indicated that transcriptional differences can be influenced by changes in the environment and may be subject to modification or regulation at the translational level (49), which could be the case for IGFBPs in the human preovulatory follicle.
Recent experimental data have highlighted the critical role of PAPP-A as a physiological regulator of IGF bioactivity (33, 35, 36). This study investigates further this role within human preovulatory follicles collected at 5 distinct time points during the ovulatory peak. While the levels of PAPP-A in FF remained relatively stable during the midcycle surge of gonadotropins, its proteolytic activity was significantly downregulated after OT by the local ovarian environment. It is worth mentioning that one outlier was observed at T = 0 hours (before OT), showing a notably lower cleavage rate (see Fig. 4). This could be attributed to a potential early endogenous surge of luteinizing hormone in this individual, resulting in preprimed preovulatory follicles and an altered intrafollicular milieu, which may subsequently lead to decreased proteolytic activity, as observed after the hCG trigger for OT.
This study builds on previous findings showing that IGF2 is the primary ligand within the human ovarian follicle, given that the expression of IGF1 in GCs from the analyzed time points is notably lower. However, it is worth noting that IGF1, akin to insulin, may be supplied to the ovarian follicle either through circulation or secretion by stroma or theca cells surrounding the follicles.
Final maturation of follicles was induced with GnRH agonist in women representing 51 of 83 samples used in the microarray analysis. Therefore, the results most likely present a combined effect of both luteinizing hormone and FSH, mimicking the natural midcycle surge. In this study, differences between the two groups having either hCG or agonist for final maturation of follicles were insignificant (contribution to sample variation <3%), probably because the study was not powered to study these differences. In the microarray analysis, the time-point differences exceeded that of the treatment differences, and only the last time-point group at 36 hours after OT (n = 44) provided the power to show some differences between the protocols, as mentioned in the previous study by Poulsen et al, from which our data are extracted (38). Thus, future studies should be designed to focus on potential differences in the genes of interest between the different modes of ovulation trigger.
In summary, our data suggest a possible mechanism that accounts for the abrupt cessation of GC proliferation and differentiation during the ovulatory peak, a critical process for the formation of a functional corpus luteum. Specifically, we suggest that elevated levels of STC1 inhibit PAPP-A–induced IGF-2 release. These findings highlight the need for future functional studies to confirm a direct mechanism. Furthermore, we highlight the importance of IGF-independent functions of STC1 within the human ovary, especially during the ovulatory peak, where elevated levels of this regulatory protein are observed.
Acknowledgments
Marjo Westerdahl at the Laboratory of Reproductive Biology, Copenhagen, is acknowledged for her excellent technical assistance.
Funding
The work was supported by ReproUnion (www.reprounion.eu) and by a grant from the Region Zealand, Denmark, as well as The Independent Research Fund Denmark.
Author Contributions
J.A.B. performed ELISA measurements of IGFBPs, analyzed the data, conducted the statistical analysis on data not associated with the microarray analysis, and wrote the paper. L.C.P. executed the cohort study, included the patients, isolated the GCs and FF, and performed the microarray analysis. P.R.N. performed the proteinase assay, ELISA measurements of PAPP-A and STC1, and data analysis. A.L.M.E. included patients and performed or supervised follicle punctures. S.F., K.H., and C.O. participated in interpreting the results and writing the paper. M.L.G. and C.Y.A. conceived the idea, designed the cohort study, and interpreted the results. All authors critically revised the manuscript and approved the final version.
Disclosures
The authors have nothing to disclose.
Data Availability
The data sets generated during the current study are not publicly available but are available from the corresponding author on reasonable request.
References
Abbreviations
- ELISA
enzyme-linked immunosorbent assay
- FDR
false discovery rate
- FF
follicular fluid
- GC
granulosa cell
- GnRH
gonadotropin-releasing hormone
- IGF
insulin-like growth factor
- IGF1R
insulin-like growth factor 1 receptor
- IGFBP
insulin-like growth factor binding protein
- INSR
insulin receptor
- IRS1
insulin-receptor substrate-1
- IRS2
insulin-receptor substrate-2
- OPU
oocyte pickup
- OT
ovulation trigger
- PAPP-A
pregnancy-associated plasma protein-A
- rFSH
recombinant follicle-stimulating hormone
- rhCG
recombinant human chorionic gonadotropin
- STC1
stanniocalcin-1
- STC2
stanniocalcin-2
