209 Veterinaria Italiana 2021, 57 (3), 209-214. doi: 10.12834/VetIt.2607.16055.1 Accepted: 31.08.2021 | Available on line: 31.12.2021 1Istituto Zooprofilattico Sperimentale del Piemonte, Ligura e Valle d’Aosta, via Bologna 148, 10154 Turin, Italy. 2LLP ‘Kazakh Research Institute for Livestock and Fodder Production’, st. Zhandosova 51, Almaty, Kazakhstan. 3LLP ‘Scientific Production Center of Microbiology and Virology’, Bogenbai Batyr str., 105, Almaty, Kazakhstan. *Corresponding author at: LLP ‘Scientific Production Center of Microbiology and Virology’, Bogenbai Batyr str. 105, Almaty, Kazakhstan. Tel.: +7 870 58732835, e-mail: saule.daugalieva@mail.ru. Chiara Beltramo1, Talgat Karymsakov2, Alessandro Dondo1, Berik Aryngaziyev2, Aida Daugaliyeva2, Katia Varello1, Pier Luigi Acutis1, Saule Daugaliyeva3* and Simone Peletto1 Keywords Cattle, Normalization, Real-time qPCR, Reference genes, Small intestine. Summary The use of reference genes is commonly accepted as the most reliable approach to normalize qRT-PCR and to reduce possible errors in the quantification of gene expression. The aim of this study was to identify a set of reference genes suitable for gene expression analysis in the distal portion of small intestine and ileocecal valve in cattle. These sites of intestine are of interest in veterinary science as they are the main sites of inflammation caused by Mycobacterium avium subsp. paratuberculosis, agent of paratuberculosis. We employed ten PCR assays for commonly used reference genes belonging to various functional classes and then determined their expression stability. The most stable genes were RPL13A and HMBS, followed by TFRC and B-ACT. NormFinder analysis provided similar results with B-ACT as the best reference gene, followed by RPL13A and TFRC. This validated gene panel may be useful for studies on paratuberculosis aiming to identify genes differentially expressed by qRT-PCR. Validation of suitable reference genes for quantitative expression analysis by qPCR in bovine terminal ileum and ileocecal valve reference genes varies in different tissues (Lisowski et  al. 2008) and should be evaluated a priori to avoid biased findings (Bas et al. 2004). Accordingly, a proper evaluation of several reference genes should be performed to validate which and how many genes are needed before any gene expression study (Bustin et al. 2009, Huggett et al. 2005). The use of a single reference gene is strongly discourage and the use of genes traditionally considered stable has to be carefully considered. Vandesompele and colleagues (Vandesompele et  al. 2002) demonstrated that the use of a single reference gene can imply an error of up to 20-fold in expression data. Additionally, the expression of genes assumed to be stable (i.e., GAPDH and B-ACT) can vary considerably (Bustin 2000). The aim of this study was to identify a set of reference genes suitable for gene expression analysis in the distal portion of small intestine and ileocecal valve in cattle. These sites of intestine are of interest in veterinary science as they are the main sites of inflammation caused by Mycobacterium avium subsp. paratuberculosis (MAP), the agent of paratuberculosis, a chronic and debilitating Introduction Quantitative PCR (qPCR) is the election technique for accurate expression profiles determination of selected gene of interest being characterized by high sensitivity, specificity and reproducibility (Bustin et  al. 2009). Nevertheless, several variables can concur to misleading conclusions, which include biological and technical variance in the gene expression analysis: the amount and quality of the starting material, the RNA integrity, the efficiency of retrotranscription and PCR reaction, the differences in biological samples. Moreover, the presence of pseudogenes, alternative splicing, health status, storage conditions of the samples may affect the level of gene expression and the PCR efficiency (Rekawiecki et  al. 2012). The selection of suitable reference genes is crucial to mitigate any variations arising during the experiment, because they are assumed to be characterized by constitutive and uniform expression level in all the analyzed samples, regardless of tissue differentiation, treatments or experimental design (McNeill et  al. 2007). However, several studies indicate that the expression of 210 xxxxxxxxxxxxxxxx Beltramo et al. Veterinaria Italiana 2021, 57 (3), 209-214. doi: 10.12834/VetIt.2607.16055.1 using the Bioanalyzer 2100 (Agilent Technologies). cDNA synthesis was carried out starting from 1 µg of RNA using the QuantiTect Reverse Transcription kit (Qiagen). Reference genes, primer design and qPCR Ten genes usually used as references in qPCR experiments were selected as candidate normalizers (Table I). They were tested on a serial 10-fold-dilutions of pooled cDNA to determine primer efficiency, slope value and correlation coefficient. cDNAs of the eight samples were pooled and 10-fold diluted until 1:100,000 then a five-point calibration curve was constructed for each gene. The qPCR reactions were performed in 25 µl-total volume containing: 1X iTaq Universal SYBR Green Supermix (Biorad), 11 µl of Nuclease free water, 200 nM of each primer and 1 µl of cDNA. The reactions were set up in triplicates and loaded on a Mx3005P qPCR System (Agilent Technologies) with the following thermal cycle: 95 °C for 5 min; 44 cycles of 95 °C for 30 sec, 56 °C for 45 sec and 72 °C for 30 sec; a melting curve analysis of 95  °C for 1 min and 56  °C for 30  sec. Amplification directly from genomic DNA and no-template control were included to recognize / granulomatous enteritis of ruminants, characterized by persistent diarrhea, progressive wasting, and eventual death (Coussens 2001, Dorshorst et  al. 2006, Fanelli et al. 2020). Materials and methods Samples collection, RNA extraction and cDNA synthesis Bovine tissue samples were collected during routine slaughtering procedures from eight Holstein Friesian cows, at least 2 years old. The distal tract of small intestine and the ileocecal valve were immediately sampled from each animal, rinsed in sterile PBS and preserved in RNAlater (Qiagen). All samples were stored at - 80 °C before RNA extraction. Total RNA was extracted from 50 mg of bovine intestine tissue in RNAlater with the RNeasy Lipid Tissue Mini kit (Qiagen), following the manufacturer’s instructions and performing the optional on-column DNase digestion with the RNase-Free DNase Set (Qiagen). Concentration and purity of the RNA were determined by spectrophotometer and fluorescent measurements, while RNA integrity was evaluated Table I. Details of candidate reference genes assays. Gene symbol Gene name Function Accession No. Reference Primer Sequence Amplicon size B-ACT Beta-actine Cytoskeletal structural protein AY141970.1 Liu et al. 2016 F:GCACAATGAAGATCAAGATCATC 173 R:CTAACAGTCCGCCTAGAAGCA GAPDH Glyceraldehyde 3-phosphate dehydrogenase Glycolytic enzyme, Oxidoreductase in glycolysis and gluconeogenesis U85042.1 Reist et al. 2003 F: GTCTTCACTACCATGGAGAAGG 197 R: TCATGGATGACCTTGGCCAG HMBS Hydroxymethyl-bilane synthase Heme biosynthesis NM_001046207.1 Lecchia et al. 2012 F: GAGAGGAATGAAGTGGACCTAG 110 R: GCATCATAGGGGCTCTCCC PGK1 Phosphoglycerate kinase 1 Glycolytic enzyme, polymerase α cofactor protein NM_001034299.1 Modesto et al. 2013 F: GGAAGGGAAGGGAAAAGATGC 92 R: TCCCCTAGCTTGGAAAGTGA PPIA Peptidylprolyl isomarase A (cyclophilin A) Accelerate the folding of proteins NM_178320.2 De Maria et al. 2010 F: GCCCCAACACAAATGGTTCC 95 R: CCCTCTTTCACCTTGCCAAAG RPL13A Ribosomial protein L13A Member of ribosome proteins NM_001076998.2 Modesto et al. 2013 F: CCCTGGAGGAGAAGAGAAAGG 104 R: AATTTTCTTCTCGATGTTCTTTTCG RPS9 Ribosomial protein S9 Member of ribosome proteins DT860044.1 Janovick-Guretzky et al. 2007 F: CCTCGACCAAGAGCTGAAG 62 R: CCTCCAGACCTCACGTTTGTTC SF3A1 Splicing factor 3 subunit 1 Structural component of the splicing system NM_001081510.1 Lecchia et al. 2012 F: CCTTACCATGCCTACTACCGG 144 R: CACTTGGGCTTGAACCTTCTG TFRC Transferrin receptor Transferrin receptor NM_001206577.1 Modesto et al. 2013 F: TGGAAAAATCAGTTTTGCTGAA 124 R: GTCCAAAAACTGGAAGATTTGC YWHAZ Tyrosine 3-monooxygenase Signal transduction by binding to phosphorylated serine residues on a variety of signaling molecules NM_174814.2 Modesto et al. 2013 F: CTGAACTCCCCTGAGAAAGC 165 R: CTGCTTCAGCTTCGTCTCCT 211 Beltramo et al. xxxxxxxxxxxxxxxx Veterinaria Italiana 2021, 57 (3), 209-214. doi: 10.12834/VetIt.2607.16055.1 90%, respectively. Cq values for the reference genes varied from 18 to 33. Data analysis with geNorm ranked the reference genes according to their M-values in decreasing order (Figure 1, Table II). All the genes reached M-values less than 1.5 except for YWHAZ and GAPDH. Most of the genes showed stable expression with M-values less than 1. For adequate data analysis, reference genes with M-values less than 1 should be use for a comparison of minor differences in gene expression (Hellemans et  al. 2007). The most stable genes were RPL13A and HMBS, followed by TFRC and B-ACT, while the less stable were GAPDH and YWHAZ. NormFinder analysis provided similar results with B-ACT as the best reference gene, followed by RPL13A and TFRC; YWHAZ and GAPDH showed the highest SD values (Figure 2). NormFinder showed that the lowest number of reference genes for the best evaluation of gene expression is three: in fact, the accumulated SD from B-ACT, RPL13A and TFRC was 0.4446 and its increment with the addition of the fourth gene PGK1 was negligible (Figure 3). Discussion RNA yields and quality were adequate for qPCR and similar to results obtained by other studies on bovine intestinal RNA (Weber et  al. 2016, Lecchi et  al. 2012). A good quality RNA is fundamental for gene expression analysis to avoid wrong conclusion, but it could be a problem for matrices such as intestine. It is very important the preservation of the tissue immediately after the sampling: for this reason, samples are often immediately frozen in liquid nitrogen (Hempel et  al. 2016, De Luca et  al. 2014, Weber et  al. 2016). The samples for this study could not be snap-frozen, but RNA was preserved by immediately adding RNAlater solution that avoids RNA degradation with acceptable quality and yield for qPCR, as it was done also by Lecchi and colleagues (Lecchi et al. 2016). to exclude genomic DNA contamination. Stability of the ten candidate reference genes was evaluated by algorithms geNorm and NormFinder available in the GenEx software (bioMCC). GeNorm sequentially eliminates the gene that shows the highest variation relative to all other genes, generating an M-value that should be lower than 1.5 (Vandesompele et al. 2002). Specifically, the lower is the M-value, the more stable is the gene. NormFinder compares the individual genes to a global average expression of all the genes in all samples, estimating a standard deviation (SD) for each reference gene. It also calculates the accumulated SD by using multiple reference genes: in this case the random variation among their expression is partially cancelled by SD reduction. Plotting the SD from different number of reference genes according to their stability, allows to identify a minimum in the accumulated SD, indicating the number of reference genes giving the lowest SD. Results The concentration of the RNA extracted from the eight samples by fluorimeter analysis ranged from 550 ng/µl to 1,000 ng/µl. RNA purity was assessed as A260/280 ratio (range: 1.5-2.0) and A260/230 ratio (range: 1.0-1.7). RNA integrity was checked by determining RIN values, which were between 6.6 and 7.9. qPCR experiments showed that all the reference geneassays were expressed in the eight samples and provided a single sharp peak in the melting curve profile corresponding to an amplicon of expected size on agarose gel. Only the melting curve for SF3A1 gene assay showed a secondary peak for the last three dilution points, due to primer dimer formation. Correlation coefficients and efficiencies for all the standard curves were higher than 0.97 and above Table II. Gene expression stability measures determined by geNorm for each candidate reference gene. Gene name M-value YWHAZ 1.959 GAPDH 1.712 PPIA (cyclophilin A) 1.364 RPS9 1.153 PGK1 0.799 B-ACT 0.714 TFRC 0.643 HMBS 0.552 RPL13A 0.552 Genes 2.2 M -V al u e (A n ti L o g 2) 2.0 1.8 1.6 1.4 1.2 1.0 0.8 0.6 0.4 0.2 0 RPL13A HMBS TFRC B-ACT PGK1 RPS GAPDH PPIA YWHAZ Figure 1. Ranking of candidate reference genes according to their expression stability by geNorm after stepwise exclusion of the worst scoring genes. 212 xxxxxxxxxxxxxxxx Beltramo et al. Veterinaria Italiana 2021, 57 (3), 209-214. doi: 10.12834/VetIt.2607.16055.1 genes in a collection of 33 different bovine tissues comprising small intestine, cecum and colon. They analyzed the expression of seven genes, and SF3A1, HMBS, B-ACT were the three most stable. This study differed from the previous because it focused on a specific portion of the small intestine, near the ileocecal valve, and on the ileocecal valve itself. Our results confirmed B-ACT and HMBS as optimal reference genes; SF3A1 was also included in our gene panel, but it was discarded for the presence of an unspecific peak in the melt-curve profile of the higher dilution points, probably due to primer dimer formation. This result is apparently in contrast with the results by Lecchi and colleagues (Lecchi et  al. 2016), but in that study the standard curve for the qPCR analysis was performed with a 4-fold serial dilution, while, in our study, 10-fold serial dilutions were used. This technical difference in the experiment setup results in Cqs  >  32 at the last dilution points, and the SF3A1 expression is probably low in the studied intestinal sites, thus resulting in unspecific amplification products at the highest dilutions. So SF3A1 is not helpful as reference gene for qPCR analysis of genes with a low level of expression in bovine terminal ileum tract and ileocecal valve. Our analysis demonstrated that YWHAZ and GAPDH are not suitable for optimal normalization, but they were instead acceptable for Lecchi and colleagues (Lecchi et  al. 2012). It is important to take into consideration that these conclusions derived by the analysis of the stability of reference genes on 33  different bovine tissues together and are not focused on a particular one, as in this work. Considering the gastrointestinal tract, GAPDH resulted stable in abomasums, duodenum, jejunum and cecum of lactating and not lactating cows (Connor et  al. 2010): it could be considered a good reference gene for gene expression analysis in intestinal tissues, but this conclusion is not true for For the set up of optimal qPCR conditions, the ten reference genes were tested on pooled cDNA samples and genomic DNA. The selected primer pairs spanned two exons, in order to generate specific melt-curve profile for cDNA and genomic DNA amplifications and allow the identification of possible DNA contamination in the samples. The presence of a single peak on the melt-curve profile and of a single band of the expected size on agarose gel electrophoresis confirmed the gene specific amplifications (Peletto et al. 2011). The identification of a list of appropriate reference genes is basilar for a proper determination of gene expression, to reduce the variability introduced during the different steps of the experiment (Sahu et  al. 2018). The ideal reference geneshould show a stable expression in the tissue under different experimental conditions, so it is necessary to identify and validate reference genes for each type of sample and for each condition (Bustin et al. 2009). Genes such as GAPDH, B-ACT and 18S  rRNA have been used as reference genes in a great number of studies in the last decades without validation, but their expression can vary considerably, influencing the validity of the results (Thelling et  al. 1999, Vandesompele et  al. 2002). NormFinder identified B-ACT, RPL13A and TFRC as the best reference genes for qPCR analyses in the terminal tract and ileocecal valve of the bovine small intestine, while for geNorm RPL13A and HMBS, followed by TFRC and B-ACT, were the most stable (Figures 2 and 3). Both the software gave similar results, with small differences in the ranking of the reference genes, due to the different algorithms used to calculate variability. Therefore, the comparison of reference gene rankings from different software/algorithms is advisable for robust results (Modesto et al. 2012). The results of this study partially confirmed the conclusions by Lecchi and colleagues (Lecchi et  al. 2012), regarding the evaluation of suitable reference Genes 2.8 SD 2.6 2.4 2.2 2.0 1.8 1.6 1.4 1.2 1.0 0.8 0.6 0.4 0.2 0 RPL13A HMBS TFRC B-ACT PGK1 RPS GAPDH PPIA YWHAZ Figure 2. Ranking of candidate reference genes according to their expression stability by NormFinder. No. of Genes 0.60 A cc . S D 0.55 0.50 0.45 0.40 0.35 0.30 0.25 0.20 0.15 0.10 0.05 0 1 2 3 4 5 6 7 8 9 Figure 3. NormFinder: accumulated standard deviation (SD) for the determination of the optimal number of reference genes. 213 Beltramo et al. xxxxxxxxxxxxxxxx Veterinaria Italiana 2021, 57 (3), 209-214. doi: 10.12834/VetIt.2607.16055.1 independently from the infection by Histomonas meleagridis (Mitra et al. 2016). As reported by several studies, the expression stability of genes to be used as reference genes needs to be carefully validated to ensure reliable data, and the use of software such as geNorm and NormFinder could help in the selection of suitable genes for each specific sample type/experimental condition. An universal reference gene did not exist and changing sampling site or experimental condition could lead to different conclusions: actually, not all the reference genes previously identified in the bovine small intestine were confirmed as stable in the sampling sites of our study. On the basis of our results, B-ACT, RPL13A, TFRC and HMBS are stable reference genes for normalization of gene expression data in the terminal tract of the bovine small intestine and ileocecal valve. This validated gene panel may be useful for studies on paratuberculosis aiming to identify differentially expressed genes by quantitative PCR. Acknowledgements This research was funded by the Italian Ministry of Health (Ricerca Corrente 2014 - grant IZS PLV 10/14 RC) and by the Science Committee of the Ministry of Education and Science of the Republic of Kazakhstan (Grant No. AP09259133). the ileocecal tract. So, these evidences highlighted the importance of stability analysis of the reference genes, in order to take in account all the possible biological variation related to the expression level of genes for normalization. In this study, the analysis on the selected samples identified other two stable genes, TFRC and RPL13A. These genes are not frequently used as reference genes for gene expression analysis in bovine samples and there is few information in literature. Both of them were chosen for stability analysis in bovine neutrophils: RPL13A showed a stable expression in these cells (Crookenden et  al. 2017), while TFRC was discarded from subsequent analyses because of a high standard deviation (Vorachek et al. 2013). In a recent gene expression study aiming to identify putative biomarkers in MAP-infected cattle, TFRC resulted up-regulated in bovine whole-blood, suggesting caution in the use of this gene as reference gene. Also, the use of RPL13A had to be carefully considered, because rRNA gene expression may not be a good estimation of the total mRNA (Vandesompele et  al. 2002), so the exclusive use of rRNA genes as reference genes should be avoided. Analyses of expression stability on somatic cells from goat milk (Modesto et  al. 2013) and ovine whole-blood (Peletto et  al. 2011) defined TFRC and RPL13A as the least reliable controls. In contrast, the high stability of TFRC and RPL13A was demonstrated in spleen, liver, cecum and cecal tonsil of turkeys, Bas A., Forsberg G., Hammarstrom S. & Hammarstrom M.L. 2004. Utility of the housekeeping genes 18S rRNA, beta-actin and glyceraldehyde-3-phosphate-dehydrogenase for normalization in real-time quantitative reverse transcriptase-polymerase chain reaction analysis of gene expression in human T lymphocytes. Scand J Immunol, 59, 566-573. Bustin S.A. 2000. Absolute quantification of mRNA using real-timereverse transcription polymerase chain reaction assays. J Mol Endocrinol, 25, 169-193. Bustin S.A., Benes V., Garson J.A., Hellemans J., Huggett J., Kubista M., Mueller R., Nolan T., Pfaffl M.W., Shipley G.L., Vandesompele J. & Wittwer C.T. 2009. The MIQE guidelines: minimum information for publication of quantitative real time PCR experiments. Clin Chem, 55, 611-622. Connor E.E., Baldwin VI R.B., Capuco A.V., Evock-Clover C.M., Ellis S.E. & Sciabica K.S. 2010. Characterization of glucagon-like peptide 2 pathway member expression in bovine gastrointestinal tract. J Dairy Sci, 93, 5167-5178. References Coussens P.M. 2001. Mycobacterium paratuberculosis and the bovine immune system. Anim Health Res Rev, 2, 141-161. Crookenden M.A., Walker C.G., Kuhn-Sherlock B., Murray A., Dukkipati V.S.R., Heiser A. & Roche J.R. 2017. Evaluation of endogenous control gene expression in bovine neutrophils by reverse-transcription quantitative PCR using microfluidics gene expression arrays. J Dairy Sci, 100, 6763-6771. De Luca A., Vassalotti G., Pelagalli A., Pero M.E., Squillacioti C., Mirabella N., Lombardi P. & Avallone L. 2015. Expression and localization of aquaporin‐1 along the intestine of colostrum suckling buffalo calves. Anat Histol Embryol, 44, 391-400. Dorshorst N.C., Collins M.T. & Lombard J.E. 2006. Decision analysis model for paratuberculosis control in commercialdairy herds. Prev Vet Med, 75, 92-122. Hellemans J., Mortier G., De Paepe A., Speleman F. & Vandesompele J. 2007. qBase relative quantification framework and software for management and automated analysis of real-time quantitative PCR data. Genome Biol, 8, R19. 214 xxxxxxxxxxxxxxxx Beltramo et al. Veterinaria Italiana 2021, 57 (3), 209-214. doi: 10.12834/VetIt.2607.16055.1 M.G., Bellino C., Ariello D., Quasso A., Caramelli M. & Acutis P.L. 2011. Evaluation of internal reference genes for quantitative expression analysis by real-time PCR in ovine whole blood. Int J Mol Sci, 12, 7732-7747. Pérez R., Tupac-Yupanqui I. & Dunner S. 2008. Evaluation of suitable reference genes for gene expression studies in bovine muscular tissue. BMC Mol Biol, 9, 79. Reist M., Pfaffl M.W., Morel C., Meylan M., Hirsbrunner G., Blum J.W. & Steiner A. 2003. Quantitative mRNA analysis of eight bovine 5-HT receptor subtypes in brain, abomasum, and intestine by real‐time RT‐PCR. J Recept Signal Transduct Res, 23, 271-287. Rekawiecki R., Rutkowska J., & Kotwica J. 2012. Identification of optimal housekeeping genes for examination of gene expression in bovine corpus luteum. Reprod Biol, 12, 362-367. Robinson T.L., Sutherland I.A., & Sutherland J. 2007. Validation of candidate bovine reference genes for use with real-time PCR. Vet Immunol and Immunopathol, 115, 160-165. Sahu A.R., Wani S.A., Saxena S., Rajak K.K., Chaudhary D., Sahoo A.P., Khanduri A., Pandey A., Mondal P., Malla W.A., Khan R.I.N., Tiwari A.K., Mishra B., Muthuchelvan D., Mishra B.P., Singh R.K. & Gandham R.K. 2018. Selection and validation of suitable reference genes for qPCR gene expression analysis in goats and sheep under Peste des petits ruminants virus (PPRV), lineage IV infection. Sci Rep, 8, 15969. Thellin O., Zorzi W., Lakaye B., De Borman B., Coumans B., Hennen G., Grisar T., Igout A. & Heinen E. 1999. Housekeeping genes as internal standards: use and limits. J Biotechnol, 75, 291-295. Vandesompele J., De Preter K., Pattyn F., Poppe B., Van Roy N., De Paepe A. & Speleman F. 2002. Accurate normalization of real-time quantitative RT-PCR data by geometric averaging of multiple internal control genes. Genome Biol, 3, research0034-1. Vorachek W.R., Bobe G. & Hall J.A. 2013. Reference gene selection for quantitative PCR studies in bovine neutrophils. Advances in Bioscience and Biotechnology, 4, 6. Weber K.L., Welly B.T., Van Eenennaam A.L., Young A.E., Porto-Neto L.R., Reverter A. & Rincon G. 2016. Identification of gene networks for residual feed intake in Angus cattle using genomic prediction and RNA-seq. PLoS ONE, 11, e0152274. Hempel R.J., Bannantine J.P. & Stabel J.R. 2016. Transcriptional profiling of ileocecal valve of Holstein dairy cows infected with Mycobacterium avium subsp. paratuberculosis. PLoS ONE, 11 (4), e0153932. Huggett J., Dheda K., Bustin S. & Zumla A. 2005. Real-time RT-PCR normalisation; strategies and considerations. Genes Immun, 6, 279-284. Janovick-Guretzky N.A., Dann H.M., Carlson D.B., Murphy M.R., Loor J.J. & Drackley J.K. 2007. Housekeeping gene expression in bovine liver is affected by physiological state, feed intake, and dietary treatment. J Dairy Sci, 90, 2246-2252. Kadegowda A.K.G., Bionaz M., Thering B., Piperova L.S., Erdman R.A. & Loor J.J. 2009. Identification of internal control genes for quantitative polymerase chain reaction in mammary tissue of lactating cows receiving lipid supplements. J Dairy Sci, 92, 2007-2019. Lecchi C., Dilda F., Sartorelli P. & Ceciliani F. 2012. Widespread expression of SAA and Hp RNA in bovine tissues after evaluation of suitable reference genes. Vet Immunol and Immunopathol, 145, 556-562. Lisowski P., Pierzchała M., Goscik J., Pareek C.S. & Zwierzchowski L. 2008. Evaluation of reference genes for studies of gene expression in the bovine liver, kidney, pituitary, and thyroid. J Appl Genet, 49, 367-372. Liu J., Sun Y., Yang C., Zhang Y., Jiang Q., Huang J. & Wang C. 2016. Functional SNPs of INCENP affect semen quality by alternative splicing mode and binding affinity with the target Bta-miR-378 in Chinese Holstein bulls. PloS ONE, 11, e0162730. McNeill R.E., Miller N. & Kerin M.J. 2007. Evaluation and validation of candidate endogenous control genes for real-time quantitative PCR studies of breast cancer. BMC Mol Biol, 8, 107. Mitra T., Bilic I., Hess M. & Liebhart D. 2016. The 60S ribosomal protein L13 is the most preferable reference gene to investigate gene expression in selected organs from turkeys and chickens, in context of different infection models. Vet Res, 47, 105. Modesto P., Peletto S., Pisoni G., Cremonesi P., Castiglioni B., Colussi S., Caramelli M., Bronzo V., Moroni P. & Acutis P.L. 2013. Evaluation of internal reference genes for quantitative expression analysis by real-time reverse transcription-PCR in somatic cells from goat milk. J Dairy Sci, 96, 7932-7944. Peletto S., Bertuzzi S., Campanella C., Modesto P., Maniaci