Biomedicine & Pharmacotherapy 78 (2016) 66–73

Available online at

ScienceDirect www.sciencedirect.com

Original article

Selection of reliable reference genes in eutopic and ectopic endometrium for quantitative expression studies  skaa , Mirosław Andrusiewicza,* , Bartosz Słowikowskia,b , Izabela Skibin a c  -Cholewa , Anna Dera-Szymanowska Maria Wołun a b c

Department of Cell Biology, Health Sciences Faculty, Poznan University of Medical Sciences, Rokietnicka str. 5D, 60-806 Poznan, Poland Department of Biochemistry and Molecular Biology, Faculty of Medicine I, Poznan University of Medical Sciences, Swiecickiego str. 6, 60-781 Poznan, Poland Department of Perinatology and Gynecology, Faculty of Medicine II, Poznan University of Medical Sciences, Polna Street 33, 60-535 Poznan, Poland

A R T I C L E I N F O

A B S T R A C T

Article history: Received 5 October 2015 Received in revised form 16 December 2015 Accepted 21 December 2015

Purpose: Physiological changes during menstrual cycle cause the endometrium and endometriosis to develop specific kind of tissues, especially in regard to the gene expression profiles, which may include also housekeeping genes, commonly used as reference genes (RGs) in quantitative studies. Reverse transcription, followed by quantitative polymerase chain reaction (RT-qPCR) is the most precise and commonly used method in gene expression studies. In order to reduce effects of technical approaches and biological variability of gene’s expression level, the studies often employ RGs in experimental data normalization. However, the expression of RGs is not always stable and depends on several variables. Thus, the selection of appropriate RG is one of the most significant steps to obtain reliable results in RTqPCR-based methods. Material and methods: With the usage of RT-qPCR, we researched the expression of seven genes (ACTB, B2M, G6PD, GAPD, GUSB, HPRT and PPIA) as reliable reference genes in eutopic and ectopic endometrial tissue specimens obtained during standard surgery of women of reproductive age. Stability of expression level was analyzed by the most universal MS Excel plug-ins including: geNorm, NormFinder and BestKeeper. The descriptive statistics were evaluated using Statistica software. Results: The distribution of threshold (Ct) values was not equal. We identified genes with higher expression level (referring to Ct values) such as ACTB and B2M, medium e.g., GAPD and low expression level, e.g., G6PD and HPRT. We demonstrated that the stability of the analyzed reference genes was not homogenous, and different algorithms pointed to PPIA, GAPD and B2M as the most stable ones in eutopic and ectopic endometrium. On the contrary to these, GUSB and G6PD were the most unstable ones. Conclusions: In RT-qPCR-based analyses of gene expression level in eutopic and ectopic endometrium, we strongly recommend that a minimum of two reference genes are to be used and we determined that the most suitable seem to be PPIA and GAPD. ã 2015 Elsevier Masson SAS. All rights reserved.

Keywords: Reference genes Housekeeping genes Eutopic endometrium Ectopic endometrium RT-qPCR

1. Introduction Endometrium is particularly specific kind of tissue. Its periodic alterations require proper interaction between many factors. Expression of various genes depends on the menstrual cycle phase. Expression of genes relies not only on physiological changes which occur during menstrual cycle but also on several

* Corresponding author at: Department of Cell Biology, Poznan University of Medical Sciences, Rokietnicka str. 5D, 60-806 Poznan, Poland. Fax: +48 61 854 71 69. E-mail addresses: [email protected] (M. Andrusiewicz),  ska), [email protected] (B. Słowikowski), [email protected] (I. Skibin  -Cholewa), [email protected] [email protected] (M. Wołun (A. Dera-Szymanowska). http://dx.doi.org/10.1016/j.biopha.2015.12.028 0753-3322/ ã 2015 Elsevier Masson SAS. All rights reserved.

gynecological pathologies, such as endometrial cancer and endometriosis which affect the genes’ expression, also potentially including the so called housekeeping genes (HKGs). Endometriosis (ectopic endometrium) is a chronic disease characterized by implantation and overgrowth of endometrial cells outside the uterine cavity. Eutopic and ectopic cells demonstrate functional similarity but additionally manifest significant structural and molecular differences. Endometriosis can lead to dysfunction of the female reproductive system and underlie fertility disorders. Pathogenesis of infertility in endometriosis is based on pathological changes of the ovaries and the fallopian tubes as well as on influence of the hormonal, biochemical and immunological alterations in the eutopic endometrium [1–4]. Multiple factors have so far been implicated and pathogenesis of endometriosis

M. Andrusiewicz et al. / Biomedicine & Pharmacotherapy 78 (2016) 66–73

continues to be discussed. Although eutopic endometrium seems to have an important role in the physiopathology of the disease, the origin of this disease still remains unknown [5]. Thus, the last few years have seen the appearance of increasing number of papers on topic of the genetic and epigenetic determinants of endometriosis. Due to presence of a substantial number of factors, which modulate physiology of endometrium, quantitative analyses in gene expression studies encounter many problems, including the difficulty of selecting appropriate reference genes (RGs) [6,7]. Reverse transcription followed by quantitative polymerase chain reaction (RT-qPCR) is the most scrupulous and commonly used method for a gene of interest (GOI) expression study. The main advantages of this technique include high sensitivity and reliability along with fast rate of receiving the results and wide range of applications. These features have resulted in RT-qPCR outclassing other, conventional methods of measuring gene expression [8]. Two different ways in which level of RNA expression can be measured exist. The first is the absolute quantification method which generates a standard curve. It is calculated by comparing GOI with the known number of copies to the sample where the level of transcript is unknown. The second one, which is the relative quantification allows analysis of gene expression by normalization of the fluorescence signal of investigated sample to internal control reference genes, the expression of which is considered to be constitutive under various experimental conditions [8,9]. As a matter of fact, the expression of RG is not always stable and depends on various factors given that small changes in technical factors may have extensive effects on experimental outcomes. Therefore it is crucial for qPCR data to be normalized in order to reduce this variability. However, normalization of relative quantification may provide obstacles, and choice of appropriate internal control gene needs to be determined empirically, especially if small changes in gene expression are expected. Hence, the selection of appropriate RG is one of the most important steps to obtain reliable results in RT-qPCR-based methods. Accurate quantitation of gene expression levels using RT-qPCR is highly dependent on normalization of the GOI against the most suitable reference gene and following the “Minimal Information for the Publication of real-time PCR guidelines” (MIQE), more than one RG is required for optimum normalization during RT-qPCR. Normalization against a single reference gene should not be acceptable unless a clear evidence is provided which confirms its stable expression level under changing physiological and experimental conditions [10]. The most commonly used reference genes in quantitative PCRbased methods belong to the housekeeping genes (HKGs) which are the large group of genes encoding proteins responsible for maintenance of basic cellular functions. In addition, the genes should manifest constant expression in all cells and tissues of the body, independently of its physiological or pathophysiological condition [11]. There are many HKGs/RGs which are used in gene expression studies. Genes such as those coding for beta-actin (ACTB), glyceraldehyde-3 phosphate dehydrogenase (GAPDH), hypoxanthine phosphoribosyltransferase 1 (HPRT), beta-2-microglobulin (B2M), glucose-6-phosphate dehydrogenase (G6PD), betaD-glucuronidase (GUSB), peptidylprolyl isomerase A (PPIA) and other, by way of their crucial role in cell function and constitutive nature are the most frequently used reference internal controls [12]. However, several studies have proved that expression of the commonly used reference genes, including the above-mentioned ones, did not remain constant under many variables, such as distinct kind or condition of tissue and metabolic state [8,12–14]. Regarding the MIQE guidelines, the RGs used in experimental models should be met with some specific requirements [10]. Most importantly, their expression should not be affected by tentative

67

conditions and remain constant, regardless of the type of examined tissue. Moreover, detection of fluorescence signal has to be specific only for RNA. Primers should be properly and carefully designed in order to avoid DNA and pseudogene amplification, which might interfere with the results [15,16]. The choice of proper internal control gene is crucial for normalizing RT-qPCR data obtained during normalization GOI against RGs and leading to reduction of possible gene quantification errors and sample to sample variability [17,18]. Expression of HKGs/RGs may be influenced by many factors, such as nutritional state, tissue pathogenesis, systemic or local disease and experiment validation [19–21]. In this study we investigated the expression of housekeeping genes as reliable reference genes in eutopic and ectopic endometrial tissue specimens obtained during standard surgical procedures of women of reproductive age. 2. Material and methods Eutopic and ectopic endometrial tissue specimens were obtained from 188 females of Western European descent (51 non-pathological endometrial tissue samples vs. 137 pathological). Later it was limited to 23 cases because of strict selection criteria applied to the analysis (e.g., the rate of decay of mRNA for majority of genes is not known [22], and quality and quantity of RNA can vary among samples regarding to the time and storage conditions and the tissue itself [14]). Non-pathological endometrial tissue samples (n = 13) were collected from patients having a surgery for reasons other than endometriosis, during standard hysterectomy (diagnosed with cervical intra-epithelial neoplasia). The pathologically altered tissue fragments (n = 10) were acquired from patients during laparoscopy or laparotomy in the Department of Mother’s and Child’s Health, Poznan University of Medical Sciences. The age of patients ranged from 28 to 42 years. Endometriosis was confirmed by histopathological macroscopic and microscopic examinations. All patients had not been treated by hormonal therapy for at least 6 months before the surgery and showed regular menstrual cycles. Every specimen of tissue was obtained during the mid-secretory phase of menstrual cycle which was confirmed by the measurement of serum hormone levels. The material was used for epithelial cell isolation, as well as for histological and immunohistochemical analysis. Immediately after the surgery, the tissue specimens were placed in RNA protection buffer, RNA Later buffer (Sigma–Aldrich, St. Louis, MO, USA) at
80  C until isolation of genetic material was conducted. The study protocol was approved by the Institutional Review Board of the Poznan University of Medical Sciences (No 163/08). All patients provided written informed consent to participate in this study under a protocol approved by the Local Ethics Review Board of Poznan University of Medical Sciences. 2.1. RNA isolation and reverse transcription Total cellular RNA was extracted using TriPure Isolation Reagent (Roche Diagnostic GmbH, Mannheim, Germany), according to manufacturer’s protocol. The procedure was carried out twice. The concentration of total RNA was determined spectrophotometrically (NanoDrop ND-1000 spectrophotometer; Thermo Fisher Scientific, Waltham, MA). Denaturing agarose gel electrophoresis was carried out to check integrity of the RNA. Only those samples were used, where the total RNA concentration ranged from 1 mg/ml to 2 mg/ml, and the OD(260/280) and OD(260/230) ranged from 1.8 to 2.0 (1.91  0.039) and from 1.9 to 2.0 (1.94  0.054), respectively. The integrity was confirmed electrophoretically in 1.2% agarose gel containing 1.5% formaldehyde (Sigma–Aldrich, USA) in FA buffer (20 mM MOPS, 5 mM sodium acetate, 1 mM EDTA, 200 mM

68

M. Andrusiewicz et al. / Biomedicine & Pharmacotherapy 78 (2016) 66–73

paraformaldehyde; pH 7.0; Sigma–Aldrich); throughout visible 18S and 28S rRNA bands were used for RT-qPCR reactions. The isolated RNA was stored at
80  C until further analysis. Complementary DNA was synthetized following the reverse transcriptase manufacturer’s protocol, using 50 ng/ml of total RNA, 5pm/ml universal oligo(d)T10 primer, 10U/ml Expand Reverse Transcriptase, 1 Expand Reverse Transcriptase Buffer, 10 U/ml RNasin (RNase inhibitor) and 1 pm/ml of each dNTPs (deoxinucleotide triphosphate) (Roche). The cDNA was immediately used for the qPCR analysis. 2.2. Control genes and TaqMan1 hydrolysis probes Primers and TaqMan1 hydrolysis probe assays were obtained from the Universal ProbeLibrary (UPL) Reference Gene Assay (Roche). Each of the UPL assays provides UPL reference gene probe labelled with LightCycler1 Yellow-555 at the 50 end and a dark quencher dye near the 30 end, and the corresponding reference gene-specific primer pair. Excitation filters for the probe of 470 nm to 530 nm and emission filters of 550 nm to 610 nm were employed. Each Universal ProbeLibrary Reference Gene Assay provides prevalidated, specifically designed 12-mer UPL reference gene probe and the corresponding reference gene-specific primer pair. The genes and assay catalogue numbers are listed in Table 1. 2.3. Real-time quantitative PCR Real-time qPCR reaction was performed in a reaction volume of 20 ml as recommended in the LightCycler1 TaqMan1 Master manufacture’s protocol (Roche). The reaction mixture included final concentration of 1 LightCycler1 TaqMan1 Master mix, 200 nm hydrolysis probe, primers’ set and 5 ml of cDNA pro reaction mixture. qPCR reaction was performed in triplicates in LightCycler1 2.0 instrument (Roche Diagnostic GmbH), using preincubation at 95  C, 10 min followed by 45 quantitation cycles consisting of denaturation at 95  C for 10 s, and annealing/ extension step at 60  C for 30 s, and a final step of 1 s at 72  C for fluorescence data acquisition. Consequently, the reaction mix was cooled down to 40  C. Standard curves were generated for each gene to calculate PCR efficiency using serial decimal dilutions of cDNA library, constructed from archival RNA isolated from placenta, ranging from undiluted cDNA and ending at the dilution of 10
5. The standard curve cycling reactions were conducted in triplicates for each gene and the efficiency values were obtained from the standard curves using the efficiency correction. Each set of reactions included the negative no template control (randomly selected RNA mix from different RNAs isolation in reverse transcription reaction without

reverse transcriptase). Because no contamination had been detected, the uracil-DNA glycosylase incubation step was omitted. 2.4. Data collection and statistical analysis The qPCR results were assembled using the LightCycler1 Data Analysis (LCDA) Software version 4.0.5.415, dedicated for the LightCycler1 2.0 instrument. Baseline and threshold cycles were automatically set by the software. The number of PCR cycles required to reach fluorescence over the background was defined as threshold cycle (Ct). Each sample was analyzed in a triplicate and the average Ct value was calculated. The stability of expression of selected RGs was analyzed by the most popular and exact statistical MS Excel plug-ins including: geNorm v3 [16], NormFinder v0.953 [23] and BestKeeper v1.0 [24]. The descriptive statistics for Ct values, showed the biological variation of threshold cycle values of candidate RGs, were calculated and aligned by scatter box-whiskers plot using Statistica ver. 10.0 software (Statsoft, Poland). Samples for the reference gene stability analyses were divided into three groups (eutopic and ectopic endometrium, and both tissue types together). Data from each set was entered into the above mentioned tools, used to select the most stably expressed reference gene. Candidate RGs using geNorm was ranked firstly, followed by the comparison of the obtained data with the NormFinder and BestKeeper results. 3. Results A total of seven candidate genes were selected, based on literature review for determination of the most stable reference gene across eutopic and ectopic endometrial tissue. The selected RGs were classical housekeeping genes used for the normalization of GOI in the qPCR experiments using LightCycler 2.0 (Roche). The probe assays were obtained from the Universal ProbeLibrary Human Reference Gene Assay (Roche). The biological role of the selected genes was described in the Section 1 of this paper. These genes are involved in various cell metabolic processes (cell structure, metabolic enzymes and other processes). The majority of the RGs are involved in cellular and metabolic processes. 3.1. PCR amplification efficiencies of primers and TaqMan hydrolysis probes A standard curve was generated for each of the candidate reference genes, as described in the Section 2. The estimated qPCR amplification efficiency values ranged from 1.850 to 2.051 which is

Table 1 Genes and probes used in this study. Gen’s symbol

Name

Major protein function

Roche assay catalogue number

Cell structure ACTB

Actin, beta

Cell motility, structure, and integrity

05 046 165 001

Enzymes G6PD GAPD GUSB HPRT PPIA

Glucose-6-phosphate dehydrogenase Glyceraldehyde-3 phosphate dehydrogenase Beta-D-glucuronidase Hypoxanthine phosphoribosyltransferase 1 Peptidylprolyl isomerase A

Role in NADPH synthesis Role in carbohydrate metabolism Degrades glycosaminoglycans Role in the purine nucleotide metabolism Cis-trans isomerization of proline imidic peptide bonds

05 046 246 001 05 190 541 001 05 190 525 001 05 046 157 001 05 189 268 001

Other B2M

Beta-2-microglobulin

Association with the major histocompatibility complex

05 189 390 001

M. Andrusiewicz et al. / Biomedicine & Pharmacotherapy 78 (2016) 66–73

69

Table 2 Efficiency and Ct values for standard curves of examined genes. Gene’s symbol

Average lowest Ct

Average highest Ct

Efficiency value

Mean squared error

ACTB B2M G6PD GAPD GUSB HPRT PPIA

20.51 16.50 25.75 20.73 26.29 28.50 18.78

36.71 34.10 40.25 39.23 40.10 37.09 35.14

1.981 2.051 2.010 1.850 1.897 1.956 1.918

0.0418 0.0162 0.0465 0.0217 0.00278 0.00495 0.0118

Note: Ct—threshold value.

equivalent to 92.5% and 102.5%, respectively. The mean squared error in each case was under 0.05 (Table 2.). 3.2. Expression levels of candidate reference genes The geNorm calculates the gene expression stability value (M value) for each RG. The calculation is performed by comparing average expression ratio of each analyzed gene. The gene which has a high value of M is defined as a gene with a variability in expression. The procedure requires stepwise exclusion of the least stable gene with highest M value. The quantities are needed for input in analyses instead of Ct values. The quantities were obtained using the comparative delta-Ct method. In order to calculate the quantities, the average Ct values were transformed into relative expression quantities using the comparative delta-Ct method (E^ (minCt-sampleCt), where the efficiency (E) means amplification efficiency (2 = 100%, or the efficiencies derived from standard curves) and the minCt = lowest Ct value = Ct value of sample with highest expression. The geNorm suggested the two RGs with the lowest average stability M values (highest expression stability), obtained during stepwise exclusion of more variable RGs [16]. We have also used the NormFinder algorithm to analyze our data. This plug-in tested the stability of reference gene across groups. This software uses a model-based approach to calculate stability values based both on inter- and intragroup expression

variation [23]. The NormFinder software suggested the best RG or the pair of RGs, in case of more than one experimental subgroup. The Ct values were transformed into relative quantities as described in the case of geNorm for NormFinder analyses. The BestKeeper, another Microsoft1 Excel-based plug-in, used pair-wise correlations and the comparative delta Ct method to rank the stability of candidate genes according to repeatability of gene expression differences [24]. The BestKeeper utilized average Ct values to calculate coefficient of variance (CV) but expressed it as a percentage of process capability index (CP) and SD with CP for each of the RGs, similarly as it was done before with the Statistica software. Ct values derived from the amplification curves were used to measure the expression levels of candidate reference genes. Between the two tissue kinds, in most cases the maximum values were lower in ectopic endometrium, but the respective mean and median values were higher (these Ct values were more homogenous). The distribution of Ct values was not equal. There were genes with higher expression level (referring to Ct values) in the group such as ACTB and B2M, medium e.g., GAPD and low expression level, e.g., G6PD and HPRT (Table 3 and Fig. 1). The coefficient of variation for analyzed genes computed using Statistica software reveals that the most stable and most variable Ct values for the eutopic and ectopic endometrium were manifested by B2M and GAPD, respectively, for eutopic endometrium tissue samples the most

Table 3 Descriptive statistics for analyzed genes. Gene’s symbol

Median Ct

Minimum Ct

Maximum Ct

SD

SEM

Eutopic and ectopic endometrium ACTB 18.11 B2M 16.77 G6PD 25.53 GAPD 19.74 GUSB 25.02 HPRT 25.35 PPIA 18.15

Mean Ct

17.50 16.62 24.66 19.72 25.75 25.63 17.92

16.10 15.22 23.13 17.18 20.66 22.60 16.47

22.29 18.63 31.10 23.35 27.65 28.24 20.68

1.546 0.790 2.102 1.625 2.162 1.517 1.092

0.32237 0.16468 0.43825 0.33876 0.45091 0.31626 0.22767

Eutopic endometrium ACTB B2M G6PD GAPD GUSB HPRT PPIA

17.71 16.58 25.02 19.10 24.07 24.71 17.88

17.04 16.36 24.00 18.26 23.98 24.12 17.42

16.10 15.22 23.13 17.18 20.66 22.60 16.47

22.29 18.63 31.10 23.35 27.50 28.24 20.68

1.678 0.881 2.269 1.693 2.406 1.629 1.215

0.46534 0.24437 0.62941 0.46957 0.66719 0.45189 0.33696

Ectopic endometrium ACTB B2M G6PD GAPD GUSB HPRT PPIA

18.64 17.02 26.19 20.57 26.24 26.18 18.50

18.56 16.96 26.20 20.36 26.28 26.15 18.38

16.94 16.30 24.16 18.76 24.64 24.86 17.34

20.84 18.14 28.91 22.49 27.65 27.44 19.81

1.243 0.605 1.752 1.126 0.873 0.857 0.842

0.39313 0.19128 0.55413 0.35610 0.27597 0.27105 0.26641

Note: SD—standard deviation, SEM—standard error of mean.

70

M. Andrusiewicz et al. / Biomedicine & Pharmacotherapy 78 (2016) 66–73

Fig. 1. Distribution of Ct values of candidate reference genes. This aligned box-whiskers plot depicts the biological variation of threshold cycle (Ct) values of candidate reference genes in endometrium and endometriosis. Lower and upper ends of the box represent 25th and 75th percentiles, respectively. Horizontal line inside the box represents median value. Whiskers below and above the box represent minimum and maximum values.

stable Ct was in the case of B2M in contrast to GUSB, which had very high CV. In ectopic endometrium the most stable HPRT had CV at the level of 3.3%, in contrast with G6PD, where the coefficient of variation was 6.69%. Therefore, further analyses were performed to select the best combination of RGs for an acute and reliable normalization of GOI in ectopic and eutopic endometrium expression data. Table 4 geNorm, NormFinder and BestKeeper stability values.

Note: the most stable gene—highlighted areas, unstable genes—italic, bold. *CV[%CP]  SD [CP].

3.3. Selection of reference genes for eutopic and ectopic endometrium Throughout the previous analysis based on the coefficient of variation, we expected the most stable expression to be able to characterize B2M and HPRT and the variable GUSB and G6PD but the results derived from geNorm suggested, that for the analysis of eutopic, ectopic and both group together different RGs should be

M. Andrusiewicz et al. / Biomedicine & Pharmacotherapy 78 (2016) 66–73

chosen. Importantly, the efficiency value derived from standard curves had influence on selection of some reference genes (Table 4 and Fig. 2). Only in cases where both endometrial tissue types were analyzed, the geNorm pointed the same RGs (B2M and PPIA) as the best match, whereas if subgrouping was present, the choice of efficiency 2 (100%) or efficiency derived from the standard curves (described in Table 2) gave different results especially, regarding to ectopic endometrium, where no common RG was typed. In the case of using NormFinder, most of the RGs with the high and low stability were the same as in the case of geNorm. The gene with the highest stability rate was PPIA for endometriosis and total endometrium and GAPD for eutopic endometrial tissue but when the comparative analysis was made, the algorithm pointed to the pair of PPIA and HPRT as the best matches for normalization between both groups, the eutopic and ectopic endometrium (Table 4, bolded frame). Again, the GUSB and G6PD were classified as most variable. But that data were not conclusive enough. The BestKeeper Excel based software which utilizes, as mentioned above, original raw data (Ct values) instead of quantities was used next.

71

With the usage of BestKeeper, the genes with higher variation were classified as the least stable, in contrast to the RGs with lower variation, which appeared to be more stable. This demonstrated that B2M represented one of the most stable reference genes with the PPIA being the second choice in eutopic endometrium. This plug-in has been a third one to pointed to the same as in case of geNorm and NormFinder genes, namely indicated GUSB and G6PD as the most variable genes. Based on the three of the Excel plug-in’s, we indicated the PPIA, as a gene with high expression level stability in most of the cases. Two of the algorithms indicated B2M and GAPD as stable. GUSB and G6PD manifested the most unstable expression, indicated by three of the software’s plug-ins (Table 4). The obtained results take usage of only one reference gene for RT-qPCR expression data normalization under consideration. Furthermore, the selection of RGs should be very sustainable and balanced not only for the gene expression analysis in eutopic and ectopic endometrium, but overall in experimental models, which use quantitative expression study at the RNA and, in our opinion protein level as well.

Fig. 2. Determination of the most stable expressed reference genes across eutopic and ectopic endometrium using geNorm plug-in. Average expression stability values (M) were calculated for each RGs candidate and the least stable genes (with higher M values) were stepwise excluded until two most stable genes were obtained.

72

M. Andrusiewicz et al. / Biomedicine & Pharmacotherapy 78 (2016) 66–73

4. Discussion Endometriosis is a common gynecologic disorder characterized by the displacement of endometrial tissue to ectopic locations. In most cases the relocation affects the pelvic peritoneum, ovaries and bowel. Regarding the women in reproductive age, the prevalence is estimated at 6–10% and reaches the level of 35– 50% in women with pain and/or idiopathic/unexplained infertility [1]. Pelvic pain remains to be the cardinal symptom of endometriosis remains, however the mechanisms leading to the development of the disease are still inadequately understood. The successful implantation of endometrial tissue in non-physiological, ectopic location involves some mechanisms, such as apoptosis avoidance (or inhibition), immune resistance, invasiveness features, relocation-habitual proliferation and vascularization and supply with nerves [25]. There are many data related to genetics and biochemistry of endometriosis, reviewed e.g., by Burney [26], which suggested strong participation of genetic and epigenetic mechanisms in development of the disease. One of the commonly used methods related to analyses of expression level manifested by various genes is the reverse transcription, followed by quantitative polymerase chain reaction (RT-qPCR). The fluorescence-based qPCR and especially RT-qPCR are effective methods of detection and measurement of amount of nucleic acid in a wide range of samples using different application methods. Nevertheless, the technical deficiencies may affect the obtained experimental data and lead to confusing results. Even before experiment’s design and during the analysis of the expression level, one of the most important factors is the appropriate choice of the reference gene or, regarding MIQE guidelines, more than one of them [10]. One of the RT-qPCR methods based on the TaqMan1 Universal Probe Library offers a ready to use set of reference genes, which can be used for normalization of the gene of interest expression level. In this paper we were trying to determine, which of seven genes (namely ACTB, B2M, G6PD, GAPD, GUSB. HPRT and PPIA), chosen on the basis of literature on RGs could be the best to serve for normalization of the raw data obtained during RT-qPCR analysis of GOI expressed in eutopic and ectopic endometrium. Statistical software as well as three different analytical plug-ins dedicated for gene’s expression stability measurement were used as a mathematical research tool for the descriptive statistics. Some of genes analyzed in this paper have been widely used for investigation of disease patterns. They have been extensively utilized in experimental work, as they are assumed to have a minimum variation in gene expression, allowing the reception of reliable and repeatable results. In contrast to them, there have been studies questioning the reliability of these RGs in experiments [27]. In fact, these reference genes are in most cases resistant to the environmental and tissue specific conditions. The problem of establishing an appropriate reference gene in endometrium and endometriosis is further complicated due to changes in gene’s expression level in accordance with the phases of the menstrual cycle. This was one of the reason for investigating usage of homogenous group of tissue samples obtained from patients in mid-secretory phase. Additionally there are few studies available which compared gene’s expression level in eutopic and ectopic endometrium. The reason may simply be the difficulty in finding a suitable reference gene. Even though the investigators intend to apply the commonly used housekeeping genes as a reference for quantitative analysis, Ejskjaer et al. wanted to use HGKs (B2M and 18SRNA as well as ACTB) as a reference in their work regarding the endometrium and endometriosis gene’s expression level. As they admitted, there was a significant cyclic variation during the menstrual cycle, which made them chose to quantify and use equal amounts of RNA in all

samples which could involve an additional factor affecting the obtained results [28]. Despite the fact that the tissue samples were collected at the same phase of the menstrual cycle, the ACTB and B2M revealed lower or higher instability of expression level between subgroups during our experiments. The most stable of RGs investigated by us seems to be PPIA, encoding peptidylprolyl isomerase A. This gene has been successfully applied in some investigations on cancers of different origin in vivo and in vitro [29,30]. Although we did not find papers containing quantitative analyses of expression level of PPIA in endometrium and endometriosis, it appears to be stable in various physiological and pathological conditions, not only in cancers and cancer cell line, but also, as described in this paper, in eutopic and ectopic endometrial tissue. We found two less stable HGKs during our investigation, GUSB and G6PD, which encode proteins responsible for glycosaminoglycans degradation and NADH synthesis, respectively. Even though these two genes were successfully applied in cancer research expression study, G6PD shows genetic and expression instability [31–33]. According to results both of these genes should not be used in experiments employing endometrium as well as endometriosis. The HPRT housekeeping gene commonly used as an reference gene, was applied in studies related to concentration of serum mRNA markers in diagnosis of endometriosis, but the authors used only this RG and the reason of such a choice was not explained [34]. Only the NormFinder pointed HPRT, in our study but it was combined with PPIA and was not a single reference gene. 5. Conclusions To obtain conclusive results, which could be comparable to other analyzes based on qPCR using reference genes as an internal control, it is crucial to use the RNA of the best quality and quantity, because the results could be affected by half-life of mRNA. We strongly recommend that a minimum of two reference genes should be used in RT-qPCR-based analyses of gene expression level in eutopic and ectopic endometrium and we demonstrated that PPIA and GAPD appear to be the most suitable. Disclosure of interest The authors declare that they have no conflicts of interest concerning this article. Acknowledgements We would like to thank professor Krzysztof Szymanowski from the Gynecological Endoscopy and Minimally Invasive Surgery Laboratory for histopathological examination of the tissue specimens. This paper was supported by The Polish Committee for Scientific Research Award nr 2011/01/B/NZ4/03487 (M.A.) and Faculty of Health Science Young Scientist Award nr 501-14044410519-08342 (M.A.). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. References [1] B. Eskenazi, M.L. Warner, Epidemiology of endometriosis, Obstet. Gynecol. Clin. North Am. 24 (1997) 235–258. [2] W.-C.V. Yang, H.-W. Chen, H.-K. Au, C.-W. Chang, C.-T. Huang, Y.-H. Yen, et al., Serum and endometrial markers, Best Pract. Res. Clin. Obstet. Gynaecol. 18 (2004) 305–318, doi:http://dx.doi.org/10.1016/j.bpobgyn.2004.03.003. [3] M. Ulukus, H. Cakmak, A. Arici, The role of endometrium in endometriosis, J. Soc. Gynecol. Investig. (2006) , doi:http://dx.doi.org/10.1016/j. jsgi.2006.07.005.

M. Andrusiewicz et al. / Biomedicine & Pharmacotherapy 78 (2016) 66–73 [4] J. Meola, D.B. Dentillo, J.C. Rosa, E. Silva, R.A. Ferriani, L.C. Veiga, C.C. Paro de Paz, et al., Glycodelin expression in the endometrium of healthy women and in the eutopic and ectopic endometrium of women with endometriosis, Fertil. Steril. 91 (2009) 1676–1680, doi:http://dx.doi.org/10.1016/j. fertnstert.2008.02.158. [5] L. Mu, W. Zheng, L. Wang, X.-J. Chen, X. Zhang, J.-H. Yang, Alteration of focal adhesion kinase expression in eutopic endometrium of women with endometriosis, Fertil. Steril. 89 (2008) 529–537, doi:http://dx.doi.org/10.1016/ j.fertnstert.2007.03.060. [6] J.M. Borthwick, D.S. Charnock-jones, B.D. Tom, M.L. Hull, R. Teirney, S.C. Phillips, et al., Determination of the transcript profile of human endometrium, Mol. Hum. Reprod. 9 (2003) 19–33, doi:http://dx.doi.org/10.1093/humrep/gag004. [7] B.W. O’Malley, M.J. Tsai, Molecular pathways of steroid receptor action, Biol. Reprod. 46 (1992) 163–167, doi:http://dx.doi.org/10.1095/biolreprod46.2.163. [8] A.K. Dhar, R.M. Bowers, K.S. Licon, G. Veazey, B. Read, Validation of reference genes for quantitative measurement of immune gene expression in shrimp, Mol. Immunol. 46 (2009) 1688–1695, doi:http://dx.doi.org/10.1016/j. molimm.2009.02.020. [9] T.D. Schmittgen, K.J. Livak, Analyzing real-time PCR data by the comparative CT method, Nat. Protoc. 3 (2008) 1101–1108, doi:http://dx.doi.org/10.1038/ nprot.2008.73. [10] S.A. Bustin, V. Benes, J.A. Garson, J. Hellemans, J. Huggett, M. Kubista, et al., The MIQE guidelines:Minimum Information for publication of quantitative realtime PCR experiments, Clin. Chem. 55 (2009) 611–622, doi:http://dx.doi.org/ 10.1373/clinchem.2008.112797. [11] A.J. Butte, V.J. Dzau, S.B. Glueck, Further defining housekeeping, or maintenance, genes Focus on A compendium of gene expression in normal human tissues, Physiol. Genom. 7 (2001) 95–96 7/2/95 [pii]. [12] A. Radoni c, S. Thulke, I.M. Mackay, O. Landt, W. Siegert, A. Nitsche, Guideline to reference gene selection for quantitative real-time PCR, Biochem. Biophys. Res. Commun. 313 (2004) 856–862, doi:http://dx.doi.org/10.1016/j. bbrc.2003.11.177. [13] F. Ohl, M. Jung, C. Xu, C. Stephan, A. Rabien, M. Burkhardt, et al., Gene expression studies in prostate cancer tissue: Which reference gene should be selected for normalization? J. Mol. Med. 83 (2005) 1014–1024, doi:http://dx. doi.org/10.1007/s00109-005-0703-z. [14] O. Thellin, W. Zorzi, B. Lakaye, B. De Borman, B. Coumans, G. Hennen, et al., Housekeeping genes as internal standards: use and limits, J. Biotechnol. 75 (1999) 291–295, doi:http://dx.doi.org/10.1016/s0168-1656(99) 00163-7. [15] S.A. Bustin, Absolute quantification of mRNA using real-time reverse transcription polymerase chain reaction assays, J. Mol. Endocrinol. 25 (2000) 169–193, doi:http://dx.doi.org/10.1677/jme.0.0250169. [16] J. Vandesompele, K. De Preter, F. Pattyn, B. Poppe, N. Van Roy, A. De Paepe, et al., Accurate normalization of real-time quantitative RT-PCR data by geometric averaging of multiple internal control genes, Genome Biol. 3 (2002) , doi: http://dx.doi.org/10.1186/gb-2002-3-7-research0034. [17] C. Tricarico, P. Pinzani, S. Bianchi, M. Paglierani, V. Distante, M. Pazzagli, et al., Quantitative real-time reverse transcription polymerase chain reaction: normalization to rRNA or single housekeeping genes is inappropriate for human tissue biopsies, Anal. Biochem. 309 (2002) 293–300, doi:http://dx.doi. org/10.1016/s0003-2697(02) 00311-1. [18] K. Dheda, J.F. Huggett, S.A. Bustin, M.A. Johnson, G. Rook, A. Zumla, Validation of housekeeping genes for normalizing RNA expression in real-time PCR, Biotechniques 37 (2004) 112–119. [19] N.A. Janovick-Guretzky, H.M. Dann, D.B. Carlson, M.R. Murphy, J.J. Loor, J.K. Drackley, Housekeeping gene expression in bovine liver is affected by physiological state, feed intake, and dietary treatment, J. Dairy Sci. 90 (2007) 2246–2252, doi:http://dx.doi.org/10.3168/jds.2006-640.

73

[20] N. Silver, E. Cotroneo, G. Proctor, S. Osailan, K.L. Paterson, G.H. Carpenter, Selection of housekeeping genes for gene expression studies in the adult rat submandibular gland under normal, inflamed, atrophic and regenerative states, BMC Mol. Biol. 9 (2008) 64, doi:http://dx.doi.org/10.1186/1471-2199-964. [21] E.W. Kuijk, L. du Puy, H.T.A. van Tol, H.P. Haagsman, B. Colenbrander, B.A.J. Roelen, Validation of reference genes for quantitative RT-PCR studies in porcine oocytes and preimplantation embryos, BMC Dev. Biol. 7 (2007) 58, doi: http://dx.doi.org/10.1186/1471-213x-7-58. [22] A. Raghavan, P.R. Bohjanen, Microarray-based analyses of mRNA decay in the regulation of mammalian gene expression, Brief. Funct. Genomic. Proteomic 3 (2004) 112–124. http://www.ncbi.nlm.nih.gov/pubmed/15355594. [23] C.L. Andersen, J.L. Jensen, T.F. Ørntoft, Normalization of real-time quantitative reverse transcription-PCR data: a model-based variance estimation approach to identify genes suited for normalization, applied to bladder and colon cancer data sets, Cancer Res. 64 (2004) 5245–5250, doi:http://dx.doi.org/10.1158/ 0008-5472.can-04-0496. [24] M.W. Pfaffl, A. Tichopad, C. Prgomet, T.P. Neuvians, Determination of stable housekeeping genes, differentially regulated target genes and sample integrity: BestKeeper-Excel-based tool using pair-wise correlations, Biotechnol. Lett. 26 (2004) 509–515, doi:http://dx.doi.org/10.1023/b: bile.0000019559.84305.47. [25] R.O. Burney, L.C. Giudice, Pathogenesis and pathophysiology of endometriosis, Fertil. Steril. 98 (2012) 511–519, doi:http://dx.doi.org/10.1016/j. fertnstert.2012.06.029. [26] R.O. Burney, The genetics and biochemistry of endometriosis, Curr. Opin. Obstet. Gynecol. 25 (2013) 280–286, doi:http://dx.doi.org/10.1097/ gco.0b013e3283630d56. [27] J. Dundas, M. Ling, Reference genes for measuring mRNA expression, Theory Biosci. 131 (2012) 215–223, doi:http://dx.doi.org/10.1007/s12064-012-0152-5. [28] K. Ejskjaer, B.S. Sorensen, S.S. Poulsen, O. Mogensen, A. Forman, E. Nexo, Expression of the epidermal growth factor system in eutopic endometrium from women with endometriosis differs from that in endometrium from healthy women, Gynecol. Obstet. Invest. 67 (2009) 118–126, doi:http://dx.doi. org/10.1159/000167798. [29] M. Rienzo, C. Schiano, A. Casamassimi, V. Grimaldi, T. Infante, C. Napoli, Identification of valid reference housekeeping genes for gene expression analysis in tumor neovascularization studies, Clin. Transl. Oncol. 15 (2013) 211–218, doi:http://dx.doi.org/10.1007/s12094-012-0904-1. [30] H. Bjerregaard, S. Pedersen, S.R. Kristensen, N. Marcussen, Reference genes for gene expression analysis by real-time reverse transcription polymerase chain reaction of renal cell carcinoma, Diagn. Mol. Pathol. 20 (2011) 212–217. [31] T.A. Almeida, A. Quispe-Ricalde, F. Montes de Oca, P. Foronda, M.M. Hernández, A high-throughput open-array qPCR gene panel to identify housekeeping genes suitable for myometrium and leiomyoma expression analysis, Gynecol. Oncol. 134 (2014) 138–143, doi:http://dx.doi.org/10.1016/j.ygyno.2014.04.012. [32] Y.-L. Li, F. Ye, Y. Hu, W.-G. Lu, X. Xie, Identification of suitable reference genes for gene expression studies of human serous ovarian cancer by real-time polymerase chain reaction, Anal. Biochem. 394 (2009) 110–116, doi:http://dx. doi.org/10.1016/j.ab.2009.07.022. [33] F. Ohl, M. Jung, A. Radoni c, M. Sachs, S.A. Loening, K. Jung, Identification and validation of suitable endogenous reference genes for gene expression studies of human bladder cancer, J. Urol. 175 (2006) 1915–1920, doi:http://dx.doi.org/ 10.1016/s0022-5347(05) 00919-5. [34] A. Agic, S. Djalali, M.M. Wolfler, G. Halis, K. Diedrich, D. Hornung, Combination of CCR1 mRNA, MCP1, and CA125 measurements in peripheral blood as a diagnostic test for endometriosis, Reprod. Sci. 15 (2008) 906–911, doi:http:// dx.doi.org/10.1177/1933719108318598.