Evaluation of reference genes in mouse preimplantation embryos for gene expression studies using real-time quantitative RT-PCR (RT-qPCR)

Background Real-time quantitative reverse-transcriptase polymerase chain reaction (RT-qPCR) is the most sensitive, and valuable technique for rare mRNA detection. However, the expression profiles of reference genes under different experimental conditions, such as different mouse strains, developmental stage, and culture conditions have been poorly studied. Results mRNA stability of the actb, gapdh, sdha, ablim, ywhaz, sptbn, h2afz, tgfb1, 18 s and wrnip genes was analyzed. Using the NormFinder program, the most stable genes are as follows: h2afz for the B6D2F-1 and C57BL/6 strains; sptbn for ICR; h2afz for KOSOM and CZB cultures of B6D2F-1 and C57BL/6 strain-derived embryos; wrnip for M16 culture of B6D2F-1 and C57BL/6 strain-derived embryos; ywhaz, tgfb1, 18 s, 18 s, ywhaz, and h2afz for zygote, 2-cell, 4-cell, 8-cell, molular, and blastocyst embryonic stages cultured in KSOM medium, respectively; h2afz, wrnip, wrnip, h2afz, ywhaz, and ablim for zygote, 2-cell, 4-cell, 8-cell, molular, and blastocyst stage embryos cultured in CZB medium, respectively; 18 s, h2afz, h2afz, actb, h2afz, and wrnip for zygote, 2-cell, 4-cell, 8-cell, molular, and blastocyst stage embryos cultured in M16 medium, respectively. Conclusions These results demonstrated that candidate reference genes for normalization of target gene expression using RT-qPCR should be selected according to mouse strains, developmental stage, and culture conditions. Electronic supplementary material The online version of this article (doi:10.1186/1756-0500-7-675) contains supplementary material, which is available to authorized users.


Background
The culture media used for in vitro propagation of mouse preimplantation embryos plays an important role in maintaining their efficacy and survival rates. Extensive efforts to establish appropriate culture conditions has led to the development of media such as the modified version of Whittingham's original medium 16 {M16; [1]}, Chatot-Ziomek-Bavister [CZB; [2]], and Potassium Simplex Optimized Medium {KSOM; [3]} for in vitro culture of mouse preimplantation embryos. These culture media contain seven inorganic ions: Na + , K + , Cl − , Ca 2+ , Mg 2+ , SO 4 2− , and PO 4 2− [4]: M16 was formulated in the 1970s, whereas, in the late 1980s, CZB and KSOM media were first developed in order to overcome 2-cell blocks, which are sensitive to osmolarity [5,6]. Therefore, these media had much lower osmolalities than M16 medium, mainly due to lower inorganic ion concentrations. In contrast, the concentration of KCl, sodium lactate, sodium pyruvate, and glucose were higher in M16 than KSOM. However, M16 has a lower concentration of KH 2 PO 4 , NaCl, and antibiotics. Also, M16 has no ethylenediaminetetraacetic acid (EDTA) or L-Glutamine. EDTA was shown to affect embryonic development by inhibiting glycolysis at the two-cell stage, thereby preventing the premature stimulation of glycolysis [7,8]. In mice, the potential for embryonic development during in vitro culture may differ among strains. For example, the ability of one-cell mouse embryos to develop into blastocysts in vitro has been shown to be a function of mouse strains, media components, and culture conditions [9][10][11][12]. In this regard, there is a need to assess the relative roles of the mouse strain background and culture environment in the modification of gene regulation during in vitro culture.
The scarcity of the mRNA amounts obtained from preimplantation embryos has hampered the molecular analysis of preimplantation embryos [13][14][15]. Recent progress in RNA amplification methods and microarray platforms, including genes unique to preimplantation embryos, allow us to apply global gene expression profiling to the study of preimplantation embryos [16][17][18][19]. Initially, the majority of studies focused on gene expression analysis of preimplantation embryos, which were based on conventional reverse-transcriptase polymerase chain reaction (RT-PCR) [20,21]. Compared to conventional RT-PCR, several studies reported that real-time quantitative reverse-transcriptase polymerase chain reaction (RT-qPCR), in which data are accurately normalized, is significantly less variable than conventional RT-PCR procedures [22,23]. Therefore, it is very important to establish an accurate normalization procedure to control for variability in RT-qPCR data. Generally, glyceraldehyde-3phosphate dehydrogenase (gapdh), beta-actin, and ribosomal RNA are commonly used as internal control RNA. However, it has not been examined systematically whether the amount of mRNA in preimplantation embryos is variable for most genes, including reference genes, due to culture environment, media components, and mouse strains. The differences in reference gene expression among mouse strains during in vitro culture may introduce a considerable bias if the values of the target genes were normalized to the values of inconsistent housekeeping genes. Therefore, in order to quantify the limited quantities of mRNA contained in each embryo, a reference gene with stable expression across preimplantation embryos is required.
The expression stability and validation of reference genes suitable for the normalization of RT-qPCR data have been investigated in detail in various organisms including mouse [24][25][26], rabbit [27], cat [28], pig [29], bovine [22,[30][31][32], horse [33], and human [34,35]. In gene expression studies on animal preimplantation embryos, normalization is generally accomplished using a single housekeeping gene. In the most recent mouse preimplantation study reported, 12 housekeeping genes were tested across in vitrovs. in vivo-derived preimplantation embryos, and three (ppia, h2afz, and hprt genes) of them were used for normalization of target gene expression [25]. Although several studies have proven that the expression level of reference genes in different conditions varies, the aim of this study is to investigate the expression profiles of reference genes under different experimental conditions, such as different strain backgrounds (C57BL/6 for inbreed, B6D2F-1 for hybrids, and ICR for outbreed), culture conditions, and different development stages during the preimplantation of mouse embryos. In addition, we focused on the identification and selection of the best stable genes for normalization of gene expression analysis in different developmental stages, culture condition, and strains.

Animals
The mice were housed in wire cages at 22 ± 1°C under a 12 L:12D cycle with 70% humidity and fed ad libitum. All experiments were performed in the Institutional Animal Care and Use Committee at Konkuk University (IACUC approval number: KU12079), Seoul, Korea.

Preparation and procurement of media
All chemicals used for media preparation were purchased from Sigma Chemical Co. (St. Louis, MO, USA). Embryo culture media, such as M16 [1], was purchased from Sigma and contained no EDTA. CZB [2] and KSOM [3] were purchased from Millipore (St. Charles, MO, USA). Amino acids purchased from Sigma were added to KSOM. Each media composition is described in Additional file 1: Table  S1. All embryo manipulations outside the incubator were performed in CZB-HEPES medium (Sigma, USA).

Embryo recovery and culture
Female ICR, B6D2F-1 and C57B/6 mice (age 6-8 wk) were superovulated by injection of 5 IU of equine chorionic gonadotropin (eCG), followed by the injection of 5 IU of hCG 48 h later, and then mated with male ICR, B6D2F-1, and C57BL/6 mice. Day 0 of gestation was defined as the day a vaginal plug was found. Plug-positive females were separated for experimentation. Zygotes were obtained by opening the ampulla at 20 h post-hCG administration using CZB-HEPES medium. Approximately 20 embryos were transferred into a 30-μL drop of fresh medium (M16, KSOM, CZB) covered with mineral oil. Embryo culture in each media was performed according to manufacturer's protocols and using 5% CO 2 in atmospheric oxygen at 37°C. Embryo development rates in vitro under three different culture media are shown in Additional file 2: Table S2. For recovery of in vivo-derived embryo, ICR or B6D2F-1 and C57B/6 female mice (4 to 6 wk old) were superovulated using PMSG/hCG and mated with a proven fertile male of the same strain. One cell, 2 cells, 4 cells, 8

RNA isolation and reverse transcription
Twenty embryos were washed in Ca +2 -and Mg +2 -free PBS, snap-frozen in liquid nitrogen, and stored at −70°C. mRNA was extracted from groups of embryos using the Dynabeads mRNA Direct Kit (Dynal Ase) according to the manufacturer's instructions (r = 6). For reverse transcription, total RNAs in a final volume of 20 μL (containing 0.5 mg oligo-dT, RT buffer [1×], 10 mM dithiothreitol, and 10 mM dNTP) was subjected to reverse transcription at 37°C for 50 min, followed by 70°C for 15 min, and products were stored at 4°C until use.

Selection of reference genes and primer design
Reference genes were chosen from those used routinely in studies of pre-implantation embryonic stages [11,36,37]. Other potentially suitable reference genes were selected among those used in published literature on the reproductive system (Table 1) [38][39][40][41]. Whenever possible, primers fulfilled the following recommended criteria: amplicon length of 80 bp -130 bp, location of primers on two different exons, primer sequence length of 18-25 bp, melting temperature of 58°C ± 2°C and GC content of 40% -60%. Primer specificity was checked in silico (Primer-BLAST Tool from http://www.ncbi. nlm.nih.gov/tools/primer-blast/). All oligonucleotides were supplied unmodified and desalted (Cosmogenetech, Korea).

Real-time quantitative reverse transcriptase PCR (RT-qPCR)
The PCR reactions were performed according to the instructions of the real-time PCR machine manufacturer (ABI 7800, Applied Biosystems, Foster City, CA). The threshold cycle (Ct) value represents the cycle number at which sample fluorescence rises to a statistically significant level above the background. Each well contained 1 μL of a 10-fold dilution of cDNA, 10 μL of 2× Maxima® SYBR Green/ROX qPCR Master Mix (Thermo Scientific Fermentas, Göteborg, Sweden), 2 μL of each primer 1 -3 μM and 7 μL water. We optimized qPCR conditions on the ViiA TM 7 real-time PCR machine according to manufacturer's instructions and by testing different concentrations of primers and templates. The PCR program was as follows: denaturation (95°C for 10 min), amplification and quantification repeated 40 times (95°C for 10 sec, 55 -60°C for 30 sec, and 72°C for 30 sec with a single fluorescent measurement), melting curve analysis (65 -95°C, with a heating rate 0.2°C/sec and continuous fluorescence measurement), and final cooling to 12°C.
We confirmed the amplification of specific RT-qPCR products by performing a melting-curve step at the end of each run. Serial dilution curves for each primer allowed us to calculate RT-qPCR efficiencies. The 10-fold diluted cDNA that was used for all the amplifications was within the linear dynamic range of the calibration curvebetween 1-and 1000-fold dilution. Across all the assays, none of the quantification cycle (Cq) values were higher than 44. No-template and no-reverse transcription controls were run to determine any contamination or the generation of primer dimers. All amplifications were run in triplicate, and any doubtful curves were excluded. To minimize technical variation between samples through different runs, we preferred the sample maximization method, i.e., a run contained all the samples for one gene of interest respective to one reference gene.

Statistical analysis
GenEX qPCR data analysis software (Lotsgatan, Göteborg, Sweden) was used for implementation of quality controls and the calculation of optimal endogenous controls. This program uses the comparative Ct method for relative quantitative analysis, and the results are expressed as a fold change of expression levels. The mean value of triplicates was applied for all calculations. Medians were used to replace missing values that occurred due to inconsistencies between replicates rather than from low expression. To measure the expression stability of the candidate endogenous control genes, the commonly used program NormFinder was employed. All data are expressed as means ± SD.

Selection of ten reference genes
For RT-qPCR analysis, total RNA was extracted from 20 zygotes or embryos. To identify the best reference genes for gene expression studies in mouse preimplantation embryos, an RT-qPCR assay based on large amounts of transcriptome data from the mammalian preimplantation embryos was designed for the transcription profiling of the ten genes (actb, gapdh, sdha, ablim, ywhaz, sptbn, h2afz, tgfb1, 18 s and wrnip; Table 1). The specificity of the amplifications was confirmed by the presence of a single band of the expected size for each primer pair in agarose gels following electrophoresis, and by visualizing the single-peak melting curves of the PCR products (Additional file 3: Figure  S1). The melting temperatures of all PCR products are shown in Table 1. The cycle threshold values of candidate genes in different experimental subsets/conditions showed in Additional file 4: Table S3. All the genes tested in this study, using the NormFinder analysis, presented gene stability values (SD-value) acceptable enough for them to qualify as potential reference genes in both mouse strains and culture media. Among them, in all different experimental conditions including culture conditions, development stages and strains, the 18 s gene had a lower average Ct value, whereas the tgfb1 gene had a higher average Ct value. The RT-qPCR test suggested that tgfb1 should be selected as an internal reference gene when analyzing a high-abundance target gene, while 18 s can serve as an internal reference gene when analyzing a low-abundance target gene.

Validation of reference genes
Gene expression during the 2-cell, 4-cell, 8-cell, morulae, and blastocyst embryonic stages were analyzed in three differently derived mouse preimplantation embryos grown under three different culture media, and in the in vivo-derived embryos (Figure 1; Additional file 5: Figure S2 and Additional file 6: Figure S3). The pairwise comparison of all the potential reference genes (actb, gapdh, sdha, ablim, ywhaz, sptbn, h2afz, tgfb1, 18 s and wrnip) calculated using NormFinder resulted in SD-values above 1.0 for all except the h2fz gene, which showed 0.92 and 0.84 for the B6D2F-1 and C57BL/6 strains, respectively ( Figure 2). In the ICR strains, however, both h2afz and ablim genes were below the SD-value cut-off of 1.0.

NormFinder analysis of in Vivo vs. in vitro derived embryos
We used NormFinder analysis as a model-based approach to identify the optimal reference genes among a set of candidates. A lower average expression stability value indicates that the gene expression is more stable. NormFinder ranks all reference gene candidates based on intra-and inter-group variations and combines both results into a stability value for each reference gene candidate [42]. In B6D2F-1 embryos, h2afz, actb, and tgfb1 are ranked as good reference genes in M16 and KSOM media, whereas h2afz and 18 s are ranked as good reference genes in CZB ( Table 2). Even though actb and tgfb1 are the most stably expressed genes across strains cultured in M16 and KSOM media, these reference genes did not fulfill the stability criteria in CZB medium. By NormFinder analysis, the recommended comprehensive ranking of the stability of gene expression in the in vivoderived embryos was determined as h2afz > gapdh > tgfb1 > wrnip > sptbn > 18 s > ablim > ywhaz > sdha > actb. In C57B/6 embryos, h2afz is the most suitable gene for studies in the three different media tested and in the in vivo-derived embryos. Of note, sdha in KSOM media is the most suitable reference gene, whereas this reference gene was identified as the least stable gene for CZM and M16 media, and in vivo-derived embryos ( Table 2). The average gene expression stability in order of most stable to least stable genes for in vivo-derived embryos was determined as h2afz > gapdh > 18 s > wrnip > tgfb1 > sptbn > actb > ablim > ywhaz > sdha. In ICR embryos, the gene with the most stable expression for in vitroand in vivo-derived embryos was h2afz (Table 2). Of note, the gene stability of in vivo-derived embryos contrasted with in vitro culture-derived embryos: most of the reference genes, except sdha, showed stable expression. In conclusion, the best reference gene for in vitroor in vivo-derived embryos is h2afz, regardless of culture media used.
NormFinder analysis of inbred vs. hybrid or outbred embryos h2afz was found to be most stable in B6D2F-1 and C57BL/6 embryos cultured in KSOM and CZB media ( Table 2), while gapdh was the best reference gene for analyzing B57BL/6 ( Table 2) and ICR strains ( Table 2) cultured in M16 and KSOM media. The wrnip gene was the most stable for evaluating the B6D2F-1 strain cultured in M16 medium ( Table 2) and for the ICR strain propagated in CZB and M16 media ( Table 2). The albim and sptbn genes proved to be the most stable. On the other hand, ablim exhibited poor stability in B6D2F-1 and B57BL/6 cultured in M16 medium, and CZB or M16 media, respectively ( Figure 3B and Additional file 7: Figure S4B and C). The other genes that consistently ranked poorly included sdha for B6D2F-1 in KSOM and M16 media (Table 2), 18 s for C57BL/6 and ICR cultured in KSOM and CZB, respectively (Table 2). Also, the actb and wrnip genes were least stable in ICR strain embryos cultured in KSOM and M16 media ( Table 2). Even though gapdh and actb for the ICR and B6D2F-1 strains in KSOM medium emerged as the most stably expressed, consolidated evaluation under different conditions estimated that these genes were least stable in the B6D2F-1 and ICR strains ( Table 2). When evaluated across three different strains, h2afz (B6D2F-1, C57BL/6, ICR), 18 s (B6D2F-1), sptbn (B6D2F-1), gapdh (C57BL/6), ablim (ICR), and actb (C57BL/6, ICR) in CZB medium are the most stable, whereas 18 s (ICR), sptbn (ICR), actb (B6D2F-1) and ablim (B6D2F-1) have been excluded as good candidate reference genes. The results showed that the best-ranked reference genes differed across culture conditions or mouse strains.

Discussion
RT-qPCR is an invaluable technique for investigating changes in gene expression during preimplantation embryonic stages [43][44][45]. Since it has be performed on limited quantities of mRNA contained in each embryo, the reliability of this method mainly depends on the use of validated, stably expressed reference genes for the normalization of mRNA expression [46]. However, to the best of our knowledge, no such study on gene expression and stability in different strains cultured under different conditions has been published. The present work was thus undertaken to emphasize the need to validate the expression stability of reference genes in preimplantation embryos using different mouse strains and varied culture conditions. An ideal reference gene is one that is stably expressed within the samples to be compared, regardless of tissue differences, experimental conditions or treatments [47,48]. 'Housekeeping' genes are often supposed to have a steady expression pattern, and have been used extensively as reference genes [49]. However, many reports have shown that the expression levels of internal standards, including some housekeeping genes such as gapdh, actb, or 18 s, can alter considerably in response to alterations in the experimental conditions [50][51][52]. In this study, we also reconfirmed that several of the commonly used reference genes, including gapdh, 18 s rRNA, and beta-actin, are unsuitable for normalization during in vitro culture of some of mouse strain-derived embryos.
In this study, we selected ten genes (actb, gapdh, sdha, ablim, ywhaz, sptbn, h2afz, tgfb1, 18 s and wrnip) because they play different cellular roles: actb, ablim and sptbn encode cytoskeletal components expressed in various types of cells [53][54][55]; gapdh encodes an enzyme that catalyzes glycolysis for energy and carbon molecules [56]; ywhaz is implicated in the protection of cells from apoptosis through binding to the pro-apoptotic protein [57]; sdha encodes the enzyme that catalyzes oxidation in the succinate pathway [58]; h2afz encodes a component of the nucleosome structure of the chromosomal fiber [59]; 18 s encodes a part of the ribosomal RNA [60]; tgfb1 is a multifunctional component that controls proliferation and differentiation in several cell types [61]; and wrnip interacts with the N-terminus of the Wener protein containing the exonuclease domain [62]. KSOM and CZB media were chosen because they were optimized for inbred and outbred strains [3]. M16 medium, on the other hand was chosen, because it had higher osmolalities than KSOM and CZB media: CZB medium differs from the M16 medium, since it contains a high lactate:pyruvate ratio, 1 mM glutamine and lacks glucose. In addition, the B6D2F-1, C57BL/6, and  ICR strains were chosen because these strains are welldefined and frequently used in embryological research.
In this study, we compared the candidate genes and established a stability ranking using the NormFinder software. The stability of gene expression and, therefore, the choice of reference gene for ICR strains varied considerably based on the culture media used. The most stable reference genes for KSOM, M16, and CZB media or in vivo embryos are the gapdh, ablim, and sptbn genes, respectively (Additional file 8: Figure S5). The gene h2afz was the most stable reference for B6D2F-1, except that wrnip was more stable than h2afz, when cultured in M16 medium ( Table 2). In the B6D2F-1 and C57BL/6 strains (Table 2), the best stable reference for CZB medium is h2afz, whereas ablim is the best suitable gene for the ICR strain (Table 2). In the mouse embryos cultured in KSOM medium, the results of NormFinder software indicate that the h2afz gene is the most stable reference gene in the zygote and 8-cell stages, whereas wrnip is the most stable reference gene in the 2-cell and 4-cell stages, ywhaz for the morulae stage, and ablim for the blastocyst stage ( Figure 3). In the mouse embryos cultured in CZB medium, 18 s is the most stable reference gene in the zygote stage, whereas h2afz is the most stable in the 2-cell, 4-cell and morulae stages, actb for the 8-cell stage, and wrnip for the blastocyst stage (Table 3). In M16 medium, ywhaz is the most stable housekeeping gene in the zygote and 8-cell stages. The h2afz gene is the most stable of the reference genes in the 2-cell, 4-cell, morulae, and blastocyst stages (Table 3). This difference may be caused by culture media or the difference in developmental stages. Also, the reference gene expression levels in each of the developmental stages were shown in the different mouse strains tested (Figure 1; Additional file 5: Figure S2 and Additional file 6: Figure S3). Taken together, our observed data suggested that candidate reference genes to normalize and analyze target gene expression should be selected according to mouse strain, culture conditions, and developmental stages of the embryos.
The majority of gene expression studies on preimplantation embryos have been performed using only one housekeeping gene [63][64][65]. Contrary to our results, Chang et al. [66] reported significantly lower stability values for four reference genes (sdha, sptbn, ablim and wrnip); for example, sptbn had a higher stability value in our experiments. The differences in expression stability may be the result of different media or developmental stages analyzed in the compared experiments. The second observed difference relates to the ranking of the stability of the reference genes; for example, wrnip was found to be one of the two most stable genes in a recent study, but is one of the least stable genes in our study. The regulation of reference genes is not only variety/cultivar specific, but may also be developmental stage-or strain-specific and influenced by the experimental conditions [67][68][69]. Some genes have a relatively constant expression level across tissues while others do not [23]. This calls for validation and selection of appropriate housekeeping genes for specific strains and various experimental conditions.

Conclusions
In the present study, we examined 3 genetic backgrounds (ICR, C57BL/6, and B6D2F-1), 6 different developmental stages (1, 2, 4, 8-cells, morulae and blastocyst), and 4 environment factors (3 culture media and in vivo control) for mRNA stability and abundance of mouse preimplantation embryos. These required so many combinations to validate each embryo. Since we used oligo-dT to reverse transcribe the mRNAs, random hexamers in addition to oliogo-dT and/or reference-free method such as mRNA sequence for evaluation of each embryo were not checked. Therefore, the RNA extraction method, reference-free methods such as mRNA-sequence, and length of the poly-A tail, which has stability as well as regulatory functions (some mRNAs have long or short poly-A tail as part of their posttranscriptional mode of regulation) remain open. In summary, we have evaluated the expression stability of various reference genes using different culture conditions and strain backgrounds to identify suitable reference genes for normalization, and NormFinder was used to calculate the normalization factor for different rankings. The results of this study indicate that h2afz is a better choice than other reference genes when using a single reference gene to assess target gene expression. This study provides the first assessment of new reference genes for gene expression analysis in preimplantation mouse embryos based on culture conditions, mouse strain backgrounds, and embryonic stages. These candidates may serve as better reference genes than the traditional housekeeping genes in achieving valid and reliable analysis of gene expression.

Additional files
Additional file 1: Table S1. Composition of each medium used in this experiment.
Additional file 2: Table S2. Effects of EDTA on the development of 1 cell ICR mouse embryos in vitro.
Additional file 4: Table S3. Cycle threshold values for RT-qPCR of 10 reference genes.
Additional file 5: Figure S2. Selected reference gene expression levels of transcripts in the B6D2F-1 mouse-derived pre-implantation stage embryos. The transcript levels are shown for KSOM-(A), CZB-(B),