Identification of optimal reference genes in golden Syrian hamster with ethanol- and palmitoleic acid-induced acute pancreatitis using quantitative real-time polymerase chain reaction.

All authors were involved in data analysis, read and approved the final manuscript, and agreed to be accountable for all aspects of the work. Jinxin Miao is an editorial board member of and a coauthor of this article. To minimize bias, he was excluded from all editorial decision making related to the acceptance of this article for publication.

All animal procedures were approved by the Institutional Animal Care and Use Committee at Henan University of Chinese Medicine (IACUC protocol: DWLL201905321).

Sponsored by the Program for Science and Technology Innovation Talents in Universities of Henan Province (no. 23HASTIT045), Scientific Research of Traditional Chinese Medicine Specialized in Henan Province (no. 2022ZY1172), and China Postdoctoral Science Foundation (no. 2021T140184).

The induction and development of AP were confirmed in the ethanol‐ and POA‐treated group compared to controls. All golden Syrian hamsters survived by intraperitoneal injections of ethanol and POA. After induction of AP, pancreatic tissues of the ethanol‐ and POA‐induced group exhibited edema and paleness for 24 h. The pathological characteristics of the pancreases were then examined by H&E staining. Pancreatic tissues of AP hamsters exhibited spacing broadening, acinar hemorrhage, necrosis, and inflammatory infiltration (Figure  1C–F ). Consistent with the results of visual inspection, the ethanol‐ and POA‐induced model for 24 h exhibited much more severe pathological alteration than in models treated with shorter times (1, 3, 6, and 9 h; Figure  1 ). No significant pathological changes were observed in the control group treated with saline, ethanol, or POA (Figure  1A ; Figure  S1 ). Meanwhile, serum amylase levels in animals were significantly enhanced at 12 h after ethanol and POA treatment compared with those in the control group (Figure  S2A ). Histological changes in the pancreases of Syrian hamsters after ethanol and POA (palmitoleic acid) treatment. (A) Normal pancreas and (B–F) pancreatic histology at 1, 3, 6, 9, and 24 h after ethanol and POA treatment, respectively. Hematoxylin stain; magnification, ×200; bar, 100 μm. The performance of all primer pairs was assessed by primer specificity and efficiency. Under the premise of equal degradation of total RNA and RNA, their C t values were directly compared to analyze the transcript abundance of reference genes in pancreatic samples. To validate primer specificity and exclude the nonspecific product, we performed agarose gel electrophoresis, with the products amplified using pancreatic tissues (Figure  S2B ). Further, a single peak in the melting‐curve analysis of all primer pairs exhibited high specificity (Figure  S2C–M ). The amplification efficiency of all candidate reference gene primers ranged from 92% to 116%, with correlation coefficients ( R 2 ) equal to or greater than 0.99 (Table  2 ). Amplification efficiency for reference genes and target gene. The mRNA expression of 10 reference genes was investigated using pancreases in the absence or presence of ethanol and POA stimulation (Figure  2 ). Raw C t values at discrete time points after intraperitoneal injection of ethanol and POA indicated mRNA expression of the alternative reference genes. The distribution of raw C t values for each reference gene is shown as column means and error lines. Column means show the distribution of raw C t values for each tested gene. In addition, we analyzed the expression of these 10 reference genes in blood samples. Unfortunately, only the expression of B2m and Rpl13a were detected, with C t values >25 (Figure  S3A–J ). Effect of ethanol (EtOH) and POA (palmitoleic acid) stimulation on reference gene expression. The expression of reference genes, namely (A) Actb , (B) Gapdh , (C) Eef2 , (D) Ywhaz , (E) Rps18 , (F) Hprt1 , (G) Tubb , (H) Rpl13a , (I) Nono , and (J) B2m , was analyzed in discrete time points and control groups ( n  = 5 per group). C t values are inversely proportional to the number of templates. To select appropriate reference genes for ranking, four different tools were used to calculate the expression stability of 10 reference genes, namely BestKeeper, Comprehensive Delta CT, NormFinder, and geNorm. As shown in Figure  3A , the expression stability of the aforementioned 10 reference genes was calculated using the BestKeeper software. BestKeeper showed that Ywhaz had the highest stability in the acute pancreatitis model, whereas Actb was listed as the least stable gene (Figure  3A ). Reference gene evaluation by comparative delta CT is shown in Figure  3B . Lower mean expression stability values indicated more stable and optimally expressed genes. The results by comparative delta CT showed that Gapdh , Rpl13a , Nono , Ywhaz , and Hprt1 were the most stable genes, whereas B2m , Rps18 , Tubb , Eef2 , and Actb were less stable (Figure  3B ). In the current study, the average expression stability of the reference genes was calculated using NormFinder (Figure  3C ). Ywhaz was most stable gene, whereas Actb had the highest expression instability (Figure  3C ). Based on the M values, the most stable reference genes with the lowest M values were B2m and Hprt1 , whereas the least stable was Actb using geNorm (Figure  3D ). The calculation using the four tools showed that Actb was the least stable gene in ethanol‐ and POA‐induced acute pancreatitis (Figure  3 ). Expression stability of the 10 reference genes was scored by different algorithms. (A) BestKeeper, (B) Delta CT method, (C) NormFinder analysis, and (D) geNorm analysis for 30 pancreatic samples from the control group and ethanol‐ and POA‐induced pancreatitis group, n   =  5 per group. Lower values refer to higher stability, and higher values refer to lower stability. The stability of the reference genes was calculated using RefFinder software, which is a comprehensive ranking system. Figure  4 shows the comprehensive gene ranking from the most stable to the least stable genes: Ywhaz , Gapdh , Hprt1 , Rpl13a , Nono , Rps18 , B2m , Tubb , Eef2 , and Actb (Table  3 ). Based on the comprehensive rank, Ywhaz and Gapdh showed higher expression stability in the AP model (Table  3 ). However, Actb , Eef2 , and Tubb were the least stable in the AP model, because their stability was affected by the presence of inflammation and experimental conditions. Comprehensive ranking for gene stability of the 10 reference genes in ethanol‐ and POA (palmitoleic acid)‐induced pancreatitis and control groups. RT‐qPCR was carried out for each of the 10 reference genes using the same RNA samples from the control group and ethanol‐ and POA‐induced pancreatitis group. RefFinder was used to calculate the comprehensive gene stability. Comprehensive ranking of reference genes. To determine whether the selection of the reference gene affects the outcome of the target genes, the ΔΔ C t method was used to study the standardization of TNF‐α mRNA expression in the AP model (Figure  5 ). This study showed that the selection of the reference gene significantly affects the expression level of TNF‐α mRNA. Our result could significantly alter the outcomes and their associated interpretation. The magnitude of the relative fold change and the standard misinterpretation of the mean amplified when the target gene was standardized using the most unstable gene candidate. When the expression was normalized against the least stable genes, TNF‐α expression levels showed higher variability in the AP model, which shifted the results from a significant upregulation to a nonsignificant upregulation. In the present study, p ‐values varied with different reference genes (Figure  5 ). When using reference genes associated with stability reference genes ( Ywhaz , Gapdh , Hprt1 , Rpl13a , and Nono ), the relative TNF‐α mRNA expression levels of the inflamed pancreas were significantly upregulated (Figure  5A,B,E,F,I ). However, there was no significant difference in TNF‐α mRNA levels using the unstable genes ( Rps18 , B2m , Tubb , Eef2 , and Actb ) (Figure  5C,D,G,H,J ). Effect of different reference genes on the relative expression of pancreatic TNF‐α . Target gene expression was calculated against the 10 reference genes, namely (A) Ywhaz , (B) Gapdh , (C) Rps18 , (D) B2m , (E) Hprt1 , (F) Rpl13a , (G) Tubb , (H) Eef2 , (I) Nono , and (J) Actb using comparative ΔΔ C t method. *** p  < 0.001, ** p  < 0.01, * p  < 0.05.

Acute pancreatitis is a severe inflammatory disease that affects many people globally. Studies comparing different gene expressions of AP provide critical insights into the pathogenesis of disease onset and progression. 28 , 29 Traditionally, single housekeeping genes such as Actb , Gapdh , or RPL18 are taken as an internal control for qPCR analysis. 30 , 31 To avoid using unsuitable reference genes that may result in misleading conclusions and improve the analysis of gene expression in pancreatic tissues of golden Syrian hamsters with AP, the present study examined the expression stability of Actb , Gapdh , Eef2 , Ywhaz , Rps18 , Hprt1 , Tubb , Rpl13a , Nono , and B2m in ethanol‐ and POA‐treated hamsters. Our study evaluated their expression at different time points after ethanol and POA induction and showed that Ywhaz and Gapdh were the most stable genes and Tubb , Eef2 , and Actb were the least stable genes. Ywhaz is a gene that encodes a signal protein and involves in cellular signal transduction. 32 Of note, the stability of Ywhaz expression has been reported by other researchers, which makes it as a suitable reference gene in bovine liver and rat oligodendrocytes. 33 , 34 In this study, Ywhaz ranked first among the 10 reference genes when the expression stability was analyzed using BestKeeper and NormFinder algorithms. Gapdh is known for its function in glycolysis and is also a commonly used housekeeping gene for mRNA expression analysis. 35 , 36 In one study, the expression of Gapdh mRNA in 72 human tissues was compared, and their results demonstrated that Gapdh mRNA expression varied slightly within the same type of tissue but varied significantly among different tissues. 37 Gapdh expression is stable in human reticulocyte, 38 HEK293 cells, 39 and cervical tissues. 40 However, Gapdh expression changed in the pancreas of mouse after caerulein treatment. 41 Here, Gapdh expression was the most stable when calculated using the delta CT method. We recommend Gapdh as one of the suitable reference genes in gene expression comparison in AP. Our result also found that Hprt1 was the most stable gene when calculated using geNorm algorithm. Other studies demonstrated that Hprt1 was one of the optimal reference genes for gene expression analysis in meniscus injury 42 and chronic obstructive pulmonary disease. 43 In contrast, our data showed that Actb was the least stable gene in AP. The variation in Actb expression was also observed in the mouse model of AP. 24 Moreover, Actb was one of the least stable genes in other disease models, such as the mouse model of sulfonic acid–induced colitis. 21 These results suggested that Actb might not be suitable for normalizing gene expression in AP. Additionally, we used these 10 reference genes to normalize the expression of TNF‐α , a pancreatic inflammatory factor in AP. Consistent with the results obtained using different algorithms, significant variation in TNF‐α mRNA level was detected when Ywhaz and Gapdh , the most stable genes, were taken as internal controls, whereas a negative result was obtained when Tubb , Eef2 , and Actb were taken as internal controls. These results further supported that Ywhaz and Gapdh were suitable housekeeping genes in the pancreas of golden Syrian hamsters. In conclusion, this study was the first to analyze the stability of 10 reference genes used in the ethanol‐ and POA‐induced Syrian hamster model of AP, which affected the final conclusions (Figure  6 ). Ywhaz and Gapdh were the suitable reference genes for gene expression comparison in the golden Syrian hamster model of AP. The limitation of this study is that the animal models were established using only one method (alcohol and POA). It is challenging to exclude the possibility that other treatments (e.g., caerulein or l ‐arginine) would lead to distinct results. Effect of ethanol and palmitoleic acid (POA) on the expression stability of reference genes in the pancreas and their effect when applied to calculate the expression level of target genes using RT‐qPCR. Moreover, the immune cellular composition can vary greatly and contribute to the variation in the expression of a given gene.

Acute pancreatitis (AP) is an inflammatory disorder of the pancreas characterized by acinar cell death, local or systemic inflammation, and organ failure. 1 Bile duct obstruction caused by gallstone (38%) and alcohol abuse (36%) are two common reasons that lead to AP. 2 The global incidence of AP ranges from 5 to 30 affected individuals per 100 000 person‐years, and it continues to increase globally. 3 , 4 It has been documented that about one of four patients develops severe AP, which results in high morbidity and mortality with prolonged hospital stay and increased health‐care expenses. 5 , 6 Unfortunately, there is no effective therapeutic drug to treat or prevent AP. Therefore, it is necessary for researchers to explore and develop rational treatments that may reduce pain for patients with AP and improve their quality of life. Various animal models of AP have been established to investigate the pathogenesis and examine the therapeutic effects on AP. 7 , 8 Although mice and rats are the most widely used animals for the pancreatitis model, their disease may not be the most relevant to human disease. 7 Golden Syrian hamster ( Mesocricetus auratus ), a choice model of cancer, 9 , 10 , 11 infectious diseases, 12 , 13 , 14 and pancreatic diseases, 15 has been identified as an emerging optimal animal model for studying human disease. Four conventional methods, including treatment with caerulein, l ‐arginine or sodium taurocholate alone, or palmitoleate and ethanol together, have been used to establish the AP models of golden Syrian hamsters. 15 The presentations of pathological injury such as edema, inflammatory cell infiltration, necrosis, hemorrhage, and vacuolar degeneration could be observed in these models. 15 The similarities in anatomical features (e.g., the junction of adipose tissue to pancreatic tail) between golden Syrian hamsters and humans make it an appropriate model for translational study of AP. 15 However, the lack of antibodies and other tools limits the use of this animal model. Quantitative real‐time polymerase chain reaction (qPCR) can solve current problems. qPCR represents a powerful technique that can perform quantification and analysis processes during the amplification of target genes. Its high sensitivity, relatively low cost, and time‐saving feature enable it to be extensively used for experimental and diagnostic purposes. To obtain accurate gene expression patterns from qPCR, a reliable normalization process is required using appropriate reference genes. Reference genes as internal control are widely used in normalizing messenger ribonucleic acid (mRNA) data. Although the use of reference genes is regarded as the most appropriate normalization strategy, necessary measures are needed to validate their utility considering the expression of reference genes may vary in different tissues or cell types or during different experimental treatments. 16 Several algorithms, such as BestKeeper, 17 Comprehensive Delta CT method, 18 NormFinder, 19 and geNorm 20 analysis methods, are developed for the selection of stable reference genes, which have been employed by some researchers to examine appropriate reference genes in different disease models. 21 , 22 , 23 A previous study has investigated the stability of several potential reference genes in pancreatic tissues from the mouse model of AP induced by caerulein and Lipopolysaccharides injection. 24 Their study suggested that Rpl13a , Ywhaz , and Hprt1 were the most stable reference genes and Actb , Tubb , and B2m were the least stable reference genes. 24 However, it remains unknown whether these internal controls are suitable for gene expression normalization in golden Syrian hamsters with AP. Therefore, the present study sought to elucidate the expression stability of several reference genes, including Actb , Gapdh , Eef2 , Ywhaz , Rps18 , Hprt1 , Tubb , Rpl13a , Nono , and B2m , after the establishment of the golden Syrian hamster model of AP induced by ethanol and palmitoleic acid (POA). We hope our study would provide appropriate reference genes for evaluating the mechanism of AP induced by alcohol.

None. Jinxin Miao is an editorial board member of Animal Models and Experimental Medicine and a coauthor of this article. To minimize bias, he was excluded from all editorial decision making related to the acceptance of this article for publication.

Male golden Syrian hamsters (aged 5–6 weeks) were purchased from Vital River Laboratories and maintained in a 12‐h light–dark cycle at 23°C with standard laboratory chow and water ad libitum. The golden Syrian hamsters were made to fast overnight before the induction of AP and were randomly divided into the control group and experimental group. According to Wang's protocol, 15 POA (Sigma‐Aldrich) and ethanol (EtOH) were used to induce AP. Briefly, golden Syrian hamsters received intraperitoneal injections of ethanol (1.35 g/kg) and POA (2 mg/kg, diluted in dimethyl sulfoxide) twice at 1‐h interval; 1 mL of normal saline (0.9%) was injected before ethanol and POA injections. Controls received the same volume of saline ( n  = 5) or ethanol (1.35 g/kg, n  = 3) or POA (2 mg/kg, n  = 3) 24 h after the final injection. Samples of the pancreas were dissected and immediately stored at −80°C until total RNA extraction. Pancreatic tissues were collected 1, 3, 6, 9, and 24 h after induction of pancreatitis, and the tissues were fixed in formalin for a minimum of 2 days. Paraffin‐embedded tissues from each hamster were sectioned (3 μm) and then stained with hematoxylin and eosin (H&E) solution. Histopathological analysis of the pancreatic injury was performed by two investigators using light microscopy. Approximately 40–50 mg of pancreatic tissue was used for total RNA extraction using RNAiso Plus (Takara). Quantity and quality of total RNA were determined in the ratio of OD260 nm/OD280 nm, which was expected to be between 1.80 and 2.0, using a NanoDrop spectrophotometer (Thermo Fisher Scientific). The integrity of RNA was confirmed by staining 18s and 28s rRNA bands on 1% agarose gel. Total RNA (1 μg) was treated with gDNA Eraser at 42°C for 2 min (Takara) to remove genomic DNA contamination, according to the manufacturer's protocol. Reverse transcription was performed using the Prime Script RT Reagent Kit (Takara), according to the manufacturer's protocol, in an Applied Biosystems thermocycler (Thermo Fisher Scientific) at 37°C for 15 min and at 85°C for 5 min and then cooled to 4°C. cDNA samples were stored at −20°C for Reverse transcription‐qPCR analysis. Ten candidate reference genes were selected from previously published studies to evaluate the stability in ethanol‐ and POA‐induced AP hamster models. The candidate reference genes were Ywhaz , Gapdh , Eef , B2m , Actb , Tubb , Nono , Rpl13a , Hprt1 , and Rps18 . The primers were designed by Primer‐Blast ( https://www.ncbi.nlm.nih.gov/tools/primer‐blast/ ) and a previous study. 25 The specific characteristics of each gene primer are presented in Table  1 . RT‐qPCR reaction was performed in a StepOnePlus Real‐Time PCR System with TB Green Premix Ex Taq II (Takara), following the manufacturer's instructions, in a final volume of 20‐μL reactions. The PCR mix was prepared using 10 μL of TB Green Premix Ex Taq II, 0.8 μL of primers (forward and reverse, 10 μM), 0.4 μL of ROX reference dye, 2 μL of template cDNA, and 6 μL of water. The following amplification program was used: the cycles were set at 15 s at 95°C, 1 min at 60°C, and 15 s at 95°C, followed by 40 rounds of 5 s at 95°C and 30 s at 60°C. A melting‐curve analysis (60–95°C, with fluorescence measured every 0.5°C) was performed to confirm the specificity of each primer. All reactions were performed twice to ensure the technical reproducibility of the assays, and the average standard deviation (SD) within the duplicates studied was 0.5 cycles. The limit of detection was calculated as mean + 3 SD. Reference genes and TNF‐α target gene‐specific RT‐qPCR primers and different parameters derived from RT‐qPCR analysis. F: AGCACTTGTGATGCTGTGGT R: TGAACCGTTTCTGCCCTTGT F: CAGGATCAAGTGGACCGCAA R: ACGCCTCATCAAATCCTGCC F: GTAACGGCCACACTGGAAGA R: TGTTTCCGTAGCCTCACCAG F: TGCTCGGCACCTACCTTGAA R: CCTCCCAGAACTTAGCACCGA F: GTCACCTGGGAACGAGACAT R: TGCTGCTGTGTGCATAGACT F: GCAGTTCAAAGGCACAGTCA R: TGGTGGTGAAGATGCCAGTA F: GCAGACATCGACCTCACCAA R: ACTTTCCGTCCTTCACGTCC F: GGACTCAAGGTCCGCATCAT R: GATGGTTCTCTGGATGGGCT F: GTGCTATGTTGCCCTGGACT R: GCTCGTTGCCAATGGTGATG F: AGCCTGTTGGGCTTACTTCC R: CACTAATCACGACGCTGGGA F: TGAGCCATCGTGCCAATG R: AGCCCGTCTGCTGGTATCAC The guidelines of the Minimum Information for Publication of Quantitative RT‐PCR Experiments recommend the analysis of RT‐qPCR efficiencies for each primer pair using standard curves (5‐point, 10‐fold serial dilution of pooled cDNA including equal quantities from the sample sets). 16 The mean cycle threshold C t values for each serial dilution were plotted against the logarithm of the cDNA dilution factor and calculated according to the equation E  = 10 [−1/slope] , where the slope is the gradient of the linear regression line. 21 The linear dynamic range was determined by the standard curve and correlation coefficients ( R 2 ) for each gene as reported. To evaluate the stability of reference genes in ethanol‐ and POA‐induced AP hamster models and control groups, the comparative analysis of the stability of the candidate reference genes from four software programs, namely BestKeeper, geNorm, NormFinder, and RefFinder, was used. BestKeeper calculates the stability of candidate gene expression based on the C t values. 17 geNorm defines and ranks the reference gene based on an M value, where M is defined as the average pairwise variation in the gene of interest relative to all other control genes, and a high value of M expression refers to being less stable. 20 NormFinder evaluates intra‐ and inter‐group variations and then combines the two to produce stability values; stability values indicate the amount of systematic error introduced when studying genes. 19 The comparative delta C t (ΔCt) method was used to assess the most stable reference genes by comparing the relative expression within each sample. 26 RefFinder ( https://www.heartcure.com.au/reffinder/ ) combines the data from four algorithms to integrate, and the data from each of these analyses were used to rank the overall stability of each gene based on the geometric means. 27 TNF‐α was used to test the suitability of 10 reference genes and highlight the significance of choosing a reference gene that can quantify mRNA expression levels of the target genes. Selected TNF‐α was based on contribution to experimental pancreatitis development. Upregulation of these cytokines in the inflamed pancreas has been extensively reported, and they are also considered to be classical pro‐inflammatory mediators in AP. Ten reference genes to choose the relative TNF‐α mRNA expression in the control and AP groups were determined by calculating the differences in the comparative threshold cycle ( ΔΔ C t ). Statistical analysis was performed using GraphPad Prism 8.0. All data are expressed as means of the SD. The differences between the two groups were compared using Student's t ‐test one‐way analysis. p  < 0.05 was the statistically significant mean.

Figures S1–S3 Click here for additional data file.