Mol. Cells 2015; 38(4): 304-311
Published online March 31, 2015
https://doi.org/10.14348/molcells.2015.2122
© The Korean Society for Molecular and Cellular Biology
Correspondence to : *Correspondence: 2011105017@njau.edu.cn, liuhonglin@263.net
Most follicles in the mammalian ovary undergo atresia. Granulosa cell apoptosis is a hallmark of follicle atresia. Our previous study using a microRNA (miRNA) microarray showed that the let-7 microRNA family was differentially expressed during follicular atresia. However, whether the let-7 miRNA family members are related to porcine (
Keywords apoptosis, follicle atresia, let-7, MicroRNA, MiRNA, pig, stem-loop
Ovarian follicles grow from primordial follicles into primary, secondary and tertiary follicles. Although follicle development is an ongoing process, atresia prevents follicles from ovulating to control reproductive activity. Previous studies have indicated that ovarian follicular atresia is mediated primarily by granulosa cell apoptosis (Tilly et al., 1991). Since then, additional studies have investigated the molecular mechanisms responsible for granulosa cell apoptosis during follicle atresia. Death-ligand-receptor-induced and mitochondria-controlled apoptotic pathways are the main hypotheses regarding the activation of granulosa cell apoptosis (Manabe et al., 2004). Tumor necrosis factor (TNF) alpha and receptors, fatty acid synthase (FAS) ligand (FasL) and receptors, and TNF-related apoptosis-inducing ligand (TRAIL), are death ligand receptors known to trigger apoptosis (Lin et al., 2010; Sasson et al., 2002; Wada et al., 2002). In mitochondrial apoptosis, members of the BCL-2 family, including BAX and BAD, target the mitochondrial membrane to release cytochrome
MicroRNAs (miRNAs), which are approximately 22 nucleotides (nt) in length, have emerged as pivotal post-transcriptional regulators that cause gene silencing. MiRNAs regulate several biological and cellular processes, including cell development, proliferation and apoptosis (Dai et al., 2011). Mature miRNAs are incorporated into the miRNA induced-silencing complex (miRISC) with argonaute family proteins (AGOs), which target the 3′-untranslated region (UTR) of the target gene, causing mRNA decay in plants or protein degradation in mammals (Eulalio et al., 2009). In general, miRNAs recognize complementary binding sites in the coding region, as is the case for plants, or, in mammals, the mature miRNA partially complements the 3′-UTR, resulting in translational repression (Huntzinger and Izaurralde, 2011). MiRNA lethal-7 (let-7) was one of the first two miRNAs discovered in
Our previous research identified differentially regulated miRNAs during follicular atresia in the porcine (
Mature porcine ovaries were collected at a local slaughterhouse (TianHuan, China). Porcine ovaries were transferred to the laboratory within 1 h (maintained above 30°C) and washed by pre-warmed phosphate buffered saline (PBS, 37°C) supplemented with penicillin (100 IU/ml), streptomycin (100 μg/ml). Follicles of 3?5 mm in diameter were separated and classified by their morphology as H, EA or PA. Small scissors were used to cut the individual porcine ovary follicles and the cumulusoocyte complexes (COCs) were removed under a surgical dissecting microscope (SZ40, Olympus, Japan). Follicular fluid and granulosa cells, including mural granulosa cells, were collected into 1.5 ml tubes and centrifuged at 3,000 ×
The follicular fluid was immediately sent to the General Hospital of the Nanjing Military Command to quantify the P4 and E2 levels using P4 and E2 radioimmunoassay kits (Beijing North Institute of Biological Technology).
In our previous study, 23 differentially expressed miRNAs were obtained under conditions of at least one signal value > 1000 with an EA/H > 2 or an EA/PA < 0.7 and the role of miR-26b in granulose cell apoptosis was characterized (Lin et al., 2012). However, follicle atresia and granulosa cell apoptosis are highly orchestrated processes controlled by many extrinsic and intrinsic factors, and the miRNA molecular regulatory mechanisms in granulosa cell death during follicular atresia may involve multiple miRNAs. Therefore, we statistically analyzed the chip signals in the H, EA and PA groups using a one-way ANOVA test and analyzed the expression patterns of let-7 family members systematically in this study. Moreover, TargetScan (
Follicles and granulosa cells whose P4/E2 values matched their morphology were selected to analyze the expression patterns of the let-7 family. The Trizol Reagent (Invitrogen, USA) was used to extract total RNA from H, EA and PA follicles, respectively, following the manufacturer’s instructions.
MiRNA let-7a/b/c/g/i stem-loop primers and forward primers were designed according to the method provided by Chen et al. (2005). Briefly, stem-loop primers for let-7 family mature sequences were designed independently. Considering that the continuous eight nucleotides that began from the 3′ region of the let-7 mature sequences was the most variable region within the highly conserved let-7 family, the stem-loop RT primers comprised a universal stem-loop sequence (5′-CTCAACTGGTGTCGTGGAGTCGGCAATTCAGTTGAG-3′) and eight nucleotides that were the reverse complement to the last eight nucleotides of the let-7 family. The forward primers and universal reverse primers are presented in Table S1. The length of the designed products was 64 bp. PrimeScript™ RT Master Mix (Takara, China) was used to synthesize the cDNA copy of the miRNA.
To evaluate the specificity of the stem-loop RT-PCR, we isolated total RNA from granulosa cells of individual porcine ovary follicle, regardless of their morphology. The total RNA from granulosa cells were used to test whether let-7 miRNA expressed in pig granulosa cells. MiRNA let-7a, let-7b, let-7c, let-7i and let-7g were reversed transcribed using a specific stem-loop primer. PCR was performed on a Thermal Cycler PCR Detection System. 4% agarose gel and 20 bp DNA ladder marker (Takara) was used to detect whether the lengths of the PCR products were as expected. It is difficult to purify such short DNA fragments from PCR mixtures; therefore, the let-7 family PCR products were directly inserted into vector pMD18-T, and transformed into
Stem-loop RT-PCR based SYBR Green I Real-time PCR analysis was used to examine the expression patterns of let-7 family mature miRNAs in follicles or granulosa cells. Real-time quantitative PCR (qPCR) was performed on a Step-One Plus Real-Time PCR System (Applied Biosystems) with SYBR Premix ExTaq II (Takara) following the manufacturer’s instructions. Analysis of each sample was repeated in triplicate. The miRNA expression fold change was calculated with the reference to the expression of the U6 gene using the 2-ΔΔCt method.
To collect granulosa cells, we used syringes to puncture the follicles (3?5 mm). Granulosa cells were washed by preheated PBS (37°C) twice and placed into 25 cm2 flask with culture medium DMEM/F12 (1:1) containing penicillin (100 IU/ml), streptomycin (100 μg/ml) and 15% fetal bovine serum in an incubator at 37°C and 5% CO2. Let-7 mimics and negative controls (NC) were purchased from GenePharma (China). Let-7 mimics (75 nM) and NC (75 nM) were transfected into granulosa cells using Lipofectamine 2000 (Invitrogen) following the manufacturer’s protocol.
Granulosa cells transfected with let-7 mimics or NC were harvested after 72 h incubation and washed in cold PBS. We use Annexin V-FITC Apoptosis Detection Kit (KeyGEN, China) to detect apoptosis of the transfected cells. After staining a cell population (1?5 × 105) with Annexin V-FITC and propidium iodide (PI), according to the manufacturer’s instructions, we analyzed the stained granulosa cells by flow cytometry within 1 h. Meanwhile, nuclei were first fixed in 4% paraformaldehyde for 1hour at room temperature, then stained with chromatin dye Hoechst33342 (Beyotime, China) for 1 h at 37°C after 72 h transfections with let-7 mimics. The changes of cellular morphology were confirmed under a confocal laser microscope (Carl Zeiss, Germany).
Three independent samples from porcine H, EA, and PA ovary follicles were collected to analyze the expression patterns of let-7 family. All statistical evaluation was performed using SPSS for windows version 18.0. All data were presented as the mean ± SE. A one-way ANOVA test was used for multiple comparisons. A value of
On the basis of morphology, visible COCs (Bortul et al., 2003), clear follicular fluid, and well vascularized characterized H follicles. COCs and granulosa cells layer were somewhat separated and slightly turbid follicular fluid could be observed in EA follicles. PA follicles had turbid follicular fluid and were barely vascularized (Carson et al., 1979; Hay et al., 1976). The phenotype of H, EA and PA follicles is shown in Fig. 1.
Meanwhile, the P4/E2 (P/E) value was used to confirm the classification of the follicles accurately (Manabe et al., 1996). A P/E ratio less than 5 was considered H. A P/E ratio between 5 and 15 was considered EA and > 15 was considered PA (Table. 1) (Cheng et al., 2008). The individual follicles whose morphology matched with the P/E value were chosen for expression patterns analysis of let-7 family members.
According to miRBase 21.0 (
To evaluate the specificity of Stem-loop RT-PCR for the let-7 family, we chose let-7a/b/c/g/i as candidates and PCR was conducted to amplify miRNA cDNA. Agarose gel electrophoresis confirmed the length of the PCR products (64 bp). As shown in Fig. 3A, all the PCR products were slightly above 60 bp, without hairpins or dimers. Furthermore, TA cloning and sequencing confirmed the let-7 family mature sequences. In eight successful sequencing samples, seven samples’ sequences matched with let-7a mature sequences (Fig. 3B). Let-7b was successfully identified in six out of 11 sequenced samples (Fig. 3C), let-7c was successfully identified in eight out of nine sequenced samples (Fig. 3D), let-7i was successfully identified in six out of eight sequenced samples (Fig. 3E) and let-7g was successfully identified in 19 out of 21 sequenced samples (Fig. 3F). All the unmatched sequences did not match the other members of let-7 family.
We discriminated H, EA and PA porcine ovary follicles by morphology and the ratio of Progesterone (P4)/17b-estradiol (E2) and extracted total RNA from each group. We then tested whether SYBR Green I real-time-PCR could quantify the let-7 family levels in these follicles. U6 snRNA was used as normalization control. The results are shown in Fig. 4. MiRNA let-7a, let-7b, let-7c and let-7i showed decreased abundance in EA and PA (Figs. 4A, 4B, 4C, 4D); however, let-7g was significantly increased in EA and PA relative to its expression in H (Fig. 4E). These results suggested that the expression patterns of the let-7 family members mostly correlated with the miRNA microarray data. The exception was let-7c, whose mature form did not show significantly reduced expression in EA relative to H based on our microarray data (Fig. 2C); however, stem-loop real-time-PCR showed that the relative abundance of mature let-7c was much lower in EA than in H (Fig. 4C).
We extracted total RNA of granulosa cells from H, EA and PA follicles to test whether let-7 family were expressed in granulosa cells. The results were consistent with the expression pattern of let-7 family in follicles and microarray (Fig. 5). Previous studies showed that apoptosis triggered granulosa cell death and apoptotic granulosa cells increased in number during follicular atresia (Hughes and Gorospe, 1991; Sugimoto et al., 1998). Our results suggested that let-7 family were associated with the process of follicular apoptosis.
To explore the function of let-7 family in cultured granulosa cells, we divided the let-7 family into two groups: let-7a/b/c/i and let-7g. Let-7a, let-7b, let-7c and let-7i mimic were co-transfected into granulosa cells
Targets prediction, GO analysis and KEGG pathway analysis indicated that let-7a, let-7b, let-7c and let-7i co-regulated 496 genes (Fig. 7A), while let-7g was predicted to target 501 genes. Furthermore, the most relevant biological process involving the target genes of both let-7a/b/c/i (Fig. 7B) and let-7g (Fig. 7D) was regulation of transcription, followed by positive regulation of signal transduction and posttranscriptional regulation of gene expression. These predicted targets of downregulated let-7a/b/c/i (Fig. 7C) and let-7g (Fig. 7E) during follicular atresia may be involved in MAPK, TGF- beta and p53 signaling pathways, which affect cell proliferation, cell cycle or cell survival during follicle atresia.
According to morphological changes and differences in P4 and E2 levels, porcine ovary follicles were divided into healthy (H), early atretic (EA), and progressively atretic (PA) follicles. Hormones (Hsueh et al., 1994), nutrient deprivation (Diskin et al., 2003) and oxidative stress (Tilly and Tilly, 1995) contribute to follicular fate. Moreover, in pigs, these factors affect breeding and reproduction, which are important economic traits for domestic animals. FSH, LH/hCG and estrogens suppress ovary follicle apoptosis (Chun et al., 1996). In contrast, gonadotropins, androgens and gonadal hormones are involved in follicular atresia (Craig et al., 2007; Hillier and Tetsuka, 1997). Moreover, the estrogen receptor (ER) inhibits miRNA processing, which is dependent on the RNA helicases p68/p72. In this process, the E2/ERα complex interacts with p68/p72 and dissociates Drosha from pri-miRNAs. ER-mediated inhibition of pri-miR-145 and pri-miR-195 is stimulated by E2. Our previous microarray data showed that levels of these mature miRNAs were increased in follicle atresia, while E2 was reduced (Guthrie et al., 1995; Yamagata et al., 2009). Meanwhile, E2 suppresses the expression of mature let-7g though E2-ERα in MCF-7 cells (Qian et al., 2011). The relative abundance of mature let-7b in mouse endometrial cells increases after treatment with steroid hormones, including P4 or E2 (Fu et al., 2011). These studies suggested that hormonal signals also regulate the differential abundance of miRNAs during follicle atresia, and can activate or suppress intracellular apoptotic pathways. Therefore, use of a miRNA microarray was proposed to examine miRNA expression profiles among healthy (H), early atretic (EA), or progressively atretic (PA) porcine ovary follicles.
Interestingly, deep sequencing of adult Tibetan ovary and testis miRNA libraries indicated that let-7i, let-7f, let-7a, and let-7c were represented in the top 10 unique miRNAs with the highest expression levels (Li et al., 2011), suggesting that they play important roles in reproductive system development.
Based on miRBase 21.0, the mature sequence of porcine let-7i (ssc-let-7i) is 5′-UGAGGUAGUAGUUUGUGCU-3′, and the mature sequence of human and mouse let-7i is 5′-UGAGGUAGUAGUUUGUGCUGUU-3′. According to miRBase, ssc-let-7i is recorded as a 19 nt sequence that lacks three nucleotides (GUU) at the 3′ end. Nevertheless, in 2010, Xie et al. identified microRNAs from porcine skeletal muscle and found that the length of ssc-let-7i was 21 nt (Xie et al., 2010). In the same year, Li et al. used whole body tissue from pig or a mixture of 60 tissue types to identify the microRNAome of porcine preand postnatal development. They showed that the most abundant microRNA for ssc-let-7i was 22 nt in length, consistent with nearly all mammalian mature let-7i in miRBase (Li et al., 2010). Further, Lian et al. profiled porcine testes using Solexa deep sequencing and found that the most abundant ssc-let-7i was 22 nt in length, similar to the results of Li et al. (Lian et al., 2012). Thus, the ssc-let-7i sequence remains disputed. Our mature let-7i sequencing results indicate that ssc-let-7i is most likely 22 nt in length.
According to our prediction, let-7a/b/c/i may target TP53, CASP3 and FAS to prevent apoptosis, whereas let-7g may induce apoptosis by binding to CCND2 or BCL-XL. TP53, CASP3, FAS, CCND2, and BCL-XL represent pro- or anti-apoptotic genes that regulate granulosa cell fate. It has been reported that let-7a suppresses therapeutics-induced cancer cell death by targeting CASP3 (Tsang and Kwok, 2008). Additionally, let-7a and let-7b contribute to survival signaling by repressing interleukin (IL)-6 in human hepatocellular cancer stem cells (Meng et al., 2012). However, let-7g inhibits BCL-XL expression to promote sorafenib-induced apoptosis in human hepato-cellular carcinoma (Shimizu et al., 2010). Notably, let-7g (Fig. 7E) may also function as a potent antagonist regulating the same pathways as let-7a/b/c/i (Fig. 7C). However, further research is required to identify let-7 family functions, target genes, and associated signal pathways. Moreover, it is possible that the let-7 miRNA family forms a conserved regulatory network during granulosa cell apoptosis, similar to the conserved p53 regulatory network mediated by miRNA-125b (Le et al., 2011). However, the mechanisms of let-7 miRNA family mediated regulation of porcine ovarian follicular atresia require further validation.
. Detection of 17b-estradiol (E2) and progesterone (P4) levels in follicles
Follicle number | Progesterone (P) (pg/ml) | 17b-estradiol (E) (pg/ml) | P / E | Morphology | |
---|---|---|---|---|---|
H | 1 | 7450 | 1563.52 | 4.76 | H |
2 | 5110 | 1575.37 | 3.24 | H | |
3 | 6250 | 1261.49 | 4.95 | H | |
EA | 1 | 7493 | 1042.35 | 7.18 | EA |
2 | 6610 | 1188.25 | 5.56 | EA | |
3 | 8420 | 889.37 | 9.46 | EA | |
PA | 1 | 9753 | 510.48 | 19.11 | PA |
2 | 12560 | 532.10 | 23.6 | PA | |
3 | 11760 | 471.76 | 24.9 | PA |
H: The ratio of P/E less than 5 was considered as H;
EA: The ratio of P/E between 5 ? 15 was considered as EA;
PA: The ratio of P/E more than 15 was considered as PA
Mol. Cells 2015; 38(4): 304-311
Published online April 30, 2015 https://doi.org/10.14348/molcells.2015.2122
Copyright © The Korean Society for Molecular and Cellular Biology.
Rui Cao, Wang Jun Wu, Xiao Long Zhou, Peng Xiao, Yi Wang, and Hong Lin Liu*
Department of Animal Genetics, Breeding and Reproduction, College of Animal Science and Technology, Nanjing Agricultural University, Nanjing, 210095, China
Correspondence to:*Correspondence: 2011105017@njau.edu.cn, liuhonglin@263.net
Most follicles in the mammalian ovary undergo atresia. Granulosa cell apoptosis is a hallmark of follicle atresia. Our previous study using a microRNA (miRNA) microarray showed that the let-7 microRNA family was differentially expressed during follicular atresia. However, whether the let-7 miRNA family members are related to porcine (
Keywords: apoptosis, follicle atresia, let-7, MicroRNA, MiRNA, pig, stem-loop
Ovarian follicles grow from primordial follicles into primary, secondary and tertiary follicles. Although follicle development is an ongoing process, atresia prevents follicles from ovulating to control reproductive activity. Previous studies have indicated that ovarian follicular atresia is mediated primarily by granulosa cell apoptosis (Tilly et al., 1991). Since then, additional studies have investigated the molecular mechanisms responsible for granulosa cell apoptosis during follicle atresia. Death-ligand-receptor-induced and mitochondria-controlled apoptotic pathways are the main hypotheses regarding the activation of granulosa cell apoptosis (Manabe et al., 2004). Tumor necrosis factor (TNF) alpha and receptors, fatty acid synthase (FAS) ligand (FasL) and receptors, and TNF-related apoptosis-inducing ligand (TRAIL), are death ligand receptors known to trigger apoptosis (Lin et al., 2010; Sasson et al., 2002; Wada et al., 2002). In mitochondrial apoptosis, members of the BCL-2 family, including BAX and BAD, target the mitochondrial membrane to release cytochrome
MicroRNAs (miRNAs), which are approximately 22 nucleotides (nt) in length, have emerged as pivotal post-transcriptional regulators that cause gene silencing. MiRNAs regulate several biological and cellular processes, including cell development, proliferation and apoptosis (Dai et al., 2011). Mature miRNAs are incorporated into the miRNA induced-silencing complex (miRISC) with argonaute family proteins (AGOs), which target the 3′-untranslated region (UTR) of the target gene, causing mRNA decay in plants or protein degradation in mammals (Eulalio et al., 2009). In general, miRNAs recognize complementary binding sites in the coding region, as is the case for plants, or, in mammals, the mature miRNA partially complements the 3′-UTR, resulting in translational repression (Huntzinger and Izaurralde, 2011). MiRNA lethal-7 (let-7) was one of the first two miRNAs discovered in
Our previous research identified differentially regulated miRNAs during follicular atresia in the porcine (
Mature porcine ovaries were collected at a local slaughterhouse (TianHuan, China). Porcine ovaries were transferred to the laboratory within 1 h (maintained above 30°C) and washed by pre-warmed phosphate buffered saline (PBS, 37°C) supplemented with penicillin (100 IU/ml), streptomycin (100 μg/ml). Follicles of 3?5 mm in diameter were separated and classified by their morphology as H, EA or PA. Small scissors were used to cut the individual porcine ovary follicles and the cumulusoocyte complexes (COCs) were removed under a surgical dissecting microscope (SZ40, Olympus, Japan). Follicular fluid and granulosa cells, including mural granulosa cells, were collected into 1.5 ml tubes and centrifuged at 3,000 ×
The follicular fluid was immediately sent to the General Hospital of the Nanjing Military Command to quantify the P4 and E2 levels using P4 and E2 radioimmunoassay kits (Beijing North Institute of Biological Technology).
In our previous study, 23 differentially expressed miRNAs were obtained under conditions of at least one signal value > 1000 with an EA/H > 2 or an EA/PA < 0.7 and the role of miR-26b in granulose cell apoptosis was characterized (Lin et al., 2012). However, follicle atresia and granulosa cell apoptosis are highly orchestrated processes controlled by many extrinsic and intrinsic factors, and the miRNA molecular regulatory mechanisms in granulosa cell death during follicular atresia may involve multiple miRNAs. Therefore, we statistically analyzed the chip signals in the H, EA and PA groups using a one-way ANOVA test and analyzed the expression patterns of let-7 family members systematically in this study. Moreover, TargetScan (
Follicles and granulosa cells whose P4/E2 values matched their morphology were selected to analyze the expression patterns of the let-7 family. The Trizol Reagent (Invitrogen, USA) was used to extract total RNA from H, EA and PA follicles, respectively, following the manufacturer’s instructions.
MiRNA let-7a/b/c/g/i stem-loop primers and forward primers were designed according to the method provided by Chen et al. (2005). Briefly, stem-loop primers for let-7 family mature sequences were designed independently. Considering that the continuous eight nucleotides that began from the 3′ region of the let-7 mature sequences was the most variable region within the highly conserved let-7 family, the stem-loop RT primers comprised a universal stem-loop sequence (5′-CTCAACTGGTGTCGTGGAGTCGGCAATTCAGTTGAG-3′) and eight nucleotides that were the reverse complement to the last eight nucleotides of the let-7 family. The forward primers and universal reverse primers are presented in Table S1. The length of the designed products was 64 bp. PrimeScript™ RT Master Mix (Takara, China) was used to synthesize the cDNA copy of the miRNA.
To evaluate the specificity of the stem-loop RT-PCR, we isolated total RNA from granulosa cells of individual porcine ovary follicle, regardless of their morphology. The total RNA from granulosa cells were used to test whether let-7 miRNA expressed in pig granulosa cells. MiRNA let-7a, let-7b, let-7c, let-7i and let-7g were reversed transcribed using a specific stem-loop primer. PCR was performed on a Thermal Cycler PCR Detection System. 4% agarose gel and 20 bp DNA ladder marker (Takara) was used to detect whether the lengths of the PCR products were as expected. It is difficult to purify such short DNA fragments from PCR mixtures; therefore, the let-7 family PCR products were directly inserted into vector pMD18-T, and transformed into
Stem-loop RT-PCR based SYBR Green I Real-time PCR analysis was used to examine the expression patterns of let-7 family mature miRNAs in follicles or granulosa cells. Real-time quantitative PCR (qPCR) was performed on a Step-One Plus Real-Time PCR System (Applied Biosystems) with SYBR Premix ExTaq II (Takara) following the manufacturer’s instructions. Analysis of each sample was repeated in triplicate. The miRNA expression fold change was calculated with the reference to the expression of the U6 gene using the 2-ΔΔCt method.
To collect granulosa cells, we used syringes to puncture the follicles (3?5 mm). Granulosa cells were washed by preheated PBS (37°C) twice and placed into 25 cm2 flask with culture medium DMEM/F12 (1:1) containing penicillin (100 IU/ml), streptomycin (100 μg/ml) and 15% fetal bovine serum in an incubator at 37°C and 5% CO2. Let-7 mimics and negative controls (NC) were purchased from GenePharma (China). Let-7 mimics (75 nM) and NC (75 nM) were transfected into granulosa cells using Lipofectamine 2000 (Invitrogen) following the manufacturer’s protocol.
Granulosa cells transfected with let-7 mimics or NC were harvested after 72 h incubation and washed in cold PBS. We use Annexin V-FITC Apoptosis Detection Kit (KeyGEN, China) to detect apoptosis of the transfected cells. After staining a cell population (1?5 × 105) with Annexin V-FITC and propidium iodide (PI), according to the manufacturer’s instructions, we analyzed the stained granulosa cells by flow cytometry within 1 h. Meanwhile, nuclei were first fixed in 4% paraformaldehyde for 1hour at room temperature, then stained with chromatin dye Hoechst33342 (Beyotime, China) for 1 h at 37°C after 72 h transfections with let-7 mimics. The changes of cellular morphology were confirmed under a confocal laser microscope (Carl Zeiss, Germany).
Three independent samples from porcine H, EA, and PA ovary follicles were collected to analyze the expression patterns of let-7 family. All statistical evaluation was performed using SPSS for windows version 18.0. All data were presented as the mean ± SE. A one-way ANOVA test was used for multiple comparisons. A value of
On the basis of morphology, visible COCs (Bortul et al., 2003), clear follicular fluid, and well vascularized characterized H follicles. COCs and granulosa cells layer were somewhat separated and slightly turbid follicular fluid could be observed in EA follicles. PA follicles had turbid follicular fluid and were barely vascularized (Carson et al., 1979; Hay et al., 1976). The phenotype of H, EA and PA follicles is shown in Fig. 1.
Meanwhile, the P4/E2 (P/E) value was used to confirm the classification of the follicles accurately (Manabe et al., 1996). A P/E ratio less than 5 was considered H. A P/E ratio between 5 and 15 was considered EA and > 15 was considered PA (Table. 1) (Cheng et al., 2008). The individual follicles whose morphology matched with the P/E value were chosen for expression patterns analysis of let-7 family members.
According to miRBase 21.0 (
To evaluate the specificity of Stem-loop RT-PCR for the let-7 family, we chose let-7a/b/c/g/i as candidates and PCR was conducted to amplify miRNA cDNA. Agarose gel electrophoresis confirmed the length of the PCR products (64 bp). As shown in Fig. 3A, all the PCR products were slightly above 60 bp, without hairpins or dimers. Furthermore, TA cloning and sequencing confirmed the let-7 family mature sequences. In eight successful sequencing samples, seven samples’ sequences matched with let-7a mature sequences (Fig. 3B). Let-7b was successfully identified in six out of 11 sequenced samples (Fig. 3C), let-7c was successfully identified in eight out of nine sequenced samples (Fig. 3D), let-7i was successfully identified in six out of eight sequenced samples (Fig. 3E) and let-7g was successfully identified in 19 out of 21 sequenced samples (Fig. 3F). All the unmatched sequences did not match the other members of let-7 family.
We discriminated H, EA and PA porcine ovary follicles by morphology and the ratio of Progesterone (P4)/17b-estradiol (E2) and extracted total RNA from each group. We then tested whether SYBR Green I real-time-PCR could quantify the let-7 family levels in these follicles. U6 snRNA was used as normalization control. The results are shown in Fig. 4. MiRNA let-7a, let-7b, let-7c and let-7i showed decreased abundance in EA and PA (Figs. 4A, 4B, 4C, 4D); however, let-7g was significantly increased in EA and PA relative to its expression in H (Fig. 4E). These results suggested that the expression patterns of the let-7 family members mostly correlated with the miRNA microarray data. The exception was let-7c, whose mature form did not show significantly reduced expression in EA relative to H based on our microarray data (Fig. 2C); however, stem-loop real-time-PCR showed that the relative abundance of mature let-7c was much lower in EA than in H (Fig. 4C).
We extracted total RNA of granulosa cells from H, EA and PA follicles to test whether let-7 family were expressed in granulosa cells. The results were consistent with the expression pattern of let-7 family in follicles and microarray (Fig. 5). Previous studies showed that apoptosis triggered granulosa cell death and apoptotic granulosa cells increased in number during follicular atresia (Hughes and Gorospe, 1991; Sugimoto et al., 1998). Our results suggested that let-7 family were associated with the process of follicular apoptosis.
To explore the function of let-7 family in cultured granulosa cells, we divided the let-7 family into two groups: let-7a/b/c/i and let-7g. Let-7a, let-7b, let-7c and let-7i mimic were co-transfected into granulosa cells
Targets prediction, GO analysis and KEGG pathway analysis indicated that let-7a, let-7b, let-7c and let-7i co-regulated 496 genes (Fig. 7A), while let-7g was predicted to target 501 genes. Furthermore, the most relevant biological process involving the target genes of both let-7a/b/c/i (Fig. 7B) and let-7g (Fig. 7D) was regulation of transcription, followed by positive regulation of signal transduction and posttranscriptional regulation of gene expression. These predicted targets of downregulated let-7a/b/c/i (Fig. 7C) and let-7g (Fig. 7E) during follicular atresia may be involved in MAPK, TGF- beta and p53 signaling pathways, which affect cell proliferation, cell cycle or cell survival during follicle atresia.
According to morphological changes and differences in P4 and E2 levels, porcine ovary follicles were divided into healthy (H), early atretic (EA), and progressively atretic (PA) follicles. Hormones (Hsueh et al., 1994), nutrient deprivation (Diskin et al., 2003) and oxidative stress (Tilly and Tilly, 1995) contribute to follicular fate. Moreover, in pigs, these factors affect breeding and reproduction, which are important economic traits for domestic animals. FSH, LH/hCG and estrogens suppress ovary follicle apoptosis (Chun et al., 1996). In contrast, gonadotropins, androgens and gonadal hormones are involved in follicular atresia (Craig et al., 2007; Hillier and Tetsuka, 1997). Moreover, the estrogen receptor (ER) inhibits miRNA processing, which is dependent on the RNA helicases p68/p72. In this process, the E2/ERα complex interacts with p68/p72 and dissociates Drosha from pri-miRNAs. ER-mediated inhibition of pri-miR-145 and pri-miR-195 is stimulated by E2. Our previous microarray data showed that levels of these mature miRNAs were increased in follicle atresia, while E2 was reduced (Guthrie et al., 1995; Yamagata et al., 2009). Meanwhile, E2 suppresses the expression of mature let-7g though E2-ERα in MCF-7 cells (Qian et al., 2011). The relative abundance of mature let-7b in mouse endometrial cells increases after treatment with steroid hormones, including P4 or E2 (Fu et al., 2011). These studies suggested that hormonal signals also regulate the differential abundance of miRNAs during follicle atresia, and can activate or suppress intracellular apoptotic pathways. Therefore, use of a miRNA microarray was proposed to examine miRNA expression profiles among healthy (H), early atretic (EA), or progressively atretic (PA) porcine ovary follicles.
Interestingly, deep sequencing of adult Tibetan ovary and testis miRNA libraries indicated that let-7i, let-7f, let-7a, and let-7c were represented in the top 10 unique miRNAs with the highest expression levels (Li et al., 2011), suggesting that they play important roles in reproductive system development.
Based on miRBase 21.0, the mature sequence of porcine let-7i (ssc-let-7i) is 5′-UGAGGUAGUAGUUUGUGCU-3′, and the mature sequence of human and mouse let-7i is 5′-UGAGGUAGUAGUUUGUGCUGUU-3′. According to miRBase, ssc-let-7i is recorded as a 19 nt sequence that lacks three nucleotides (GUU) at the 3′ end. Nevertheless, in 2010, Xie et al. identified microRNAs from porcine skeletal muscle and found that the length of ssc-let-7i was 21 nt (Xie et al., 2010). In the same year, Li et al. used whole body tissue from pig or a mixture of 60 tissue types to identify the microRNAome of porcine preand postnatal development. They showed that the most abundant microRNA for ssc-let-7i was 22 nt in length, consistent with nearly all mammalian mature let-7i in miRBase (Li et al., 2010). Further, Lian et al. profiled porcine testes using Solexa deep sequencing and found that the most abundant ssc-let-7i was 22 nt in length, similar to the results of Li et al. (Lian et al., 2012). Thus, the ssc-let-7i sequence remains disputed. Our mature let-7i sequencing results indicate that ssc-let-7i is most likely 22 nt in length.
According to our prediction, let-7a/b/c/i may target TP53, CASP3 and FAS to prevent apoptosis, whereas let-7g may induce apoptosis by binding to CCND2 or BCL-XL. TP53, CASP3, FAS, CCND2, and BCL-XL represent pro- or anti-apoptotic genes that regulate granulosa cell fate. It has been reported that let-7a suppresses therapeutics-induced cancer cell death by targeting CASP3 (Tsang and Kwok, 2008). Additionally, let-7a and let-7b contribute to survival signaling by repressing interleukin (IL)-6 in human hepatocellular cancer stem cells (Meng et al., 2012). However, let-7g inhibits BCL-XL expression to promote sorafenib-induced apoptosis in human hepato-cellular carcinoma (Shimizu et al., 2010). Notably, let-7g (Fig. 7E) may also function as a potent antagonist regulating the same pathways as let-7a/b/c/i (Fig. 7C). However, further research is required to identify let-7 family functions, target genes, and associated signal pathways. Moreover, it is possible that the let-7 miRNA family forms a conserved regulatory network during granulosa cell apoptosis, similar to the conserved p53 regulatory network mediated by miRNA-125b (Le et al., 2011). However, the mechanisms of let-7 miRNA family mediated regulation of porcine ovarian follicular atresia require further validation.
. Detection of 17b-estradiol (E2) and progesterone (P4) levels in follicles.
Follicle number | Progesterone (P) (pg/ml) | 17b-estradiol (E) (pg/ml) | P / E | Morphology | |
---|---|---|---|---|---|
H | 1 | 7450 | 1563.52 | 4.76 | H |
2 | 5110 | 1575.37 | 3.24 | H | |
3 | 6250 | 1261.49 | 4.95 | H | |
EA | 1 | 7493 | 1042.35 | 7.18 | EA |
2 | 6610 | 1188.25 | 5.56 | EA | |
3 | 8420 | 889.37 | 9.46 | EA | |
PA | 1 | 9753 | 510.48 | 19.11 | PA |
2 | 12560 | 532.10 | 23.6 | PA | |
3 | 11760 | 471.76 | 24.9 | PA |
H: The ratio of P/E less than 5 was considered as H;.
EA: The ratio of P/E between 5 ? 15 was considered as EA;.
PA: The ratio of P/E more than 15 was considered as PA.
Soo-Jin Lee, Sung-E Choi, Seokho Park, Yoonjung Hwang, Youngho Son, and Yup Kang*
Mol. Cells 2023; 46(8): 496-512 https://doi.org/10.14348/molcells.2023.0045Juyong Kim, Sangwoo Seo, Jung Han Yoon Park, Ki Won Lee, Jiyoung Kim, and Jin-Chul Kim
Mol. Cells 2023; 46(5): 319-328 https://doi.org/10.14348/molcells.2023.2156Kwangho Kim, Tae Young Ryu, Jinkwon Lee, Mi-Young Son, Dae-Soo Kim, Sang Kyum Kim, and Hyun-Soo Cho
Mol. Cells 2022; 45(9): 622-630 https://doi.org/10.14348/molcells.2022.0014