Evaluation and validation of reference gene stability during Marek’s disease virus (MDV) infection

Journal of Virological Methods 236 (2016) 111–116

Contents lists available at ScienceDirect

Journal of Virological Methods journal homepage: www.elsevier.com/locate/jviromet

Evaluation and validation of reference gene stability during Marek’s disease virus (MDV) infection Sabari Nath Neerukonda a , Upendra K. Katneni a , Sergey Golovan b , Mark S. Parcells a,∗ a b

Department of Animal and Food Sciences, University of Delaware, Newark, DE 19716, United States Current Address: Spartan Bioscience, Gurdwara Road, Ottawa, Ontario K2H 1B2, Canada

a b s t r a c t Article history: Received 12 January 2016 Received in revised form 18 July 2016 Accepted 19 July 2016 Available online 19 July 2016 Keywords: Reference gene qRT-PCR Data normalization Marek’s disease virus Lymphomas

Quantitative RT-PCR (qRT-PCR) is widely used in the study of relative gene expression in general, and has been used in the field of Marek’s disease (MD) research to measure transcriptional responses to infection and/or vaccination. Studies in the past have either employed cellular ␤-actin (BACT) or glyceraldehyde-3phosphate dehydrogenase (GAPDH) as internal reference genes, although the stability of their expression in the context of Marek’s disease virus (MDV) infection has never been investigated. In the present study, we compared the stability of five reference genes (BACT, 28S RNA, 18S RNA, GAPDH, Peptidyl-prolylisomerase B [PPIB], a.k.a. cyclophilin B) as standard internal controls in chicken embryo fibroblast (CEFs) cultures infected with either MD vaccine or oncogenic MDV1 viruses. We further extend these analyses to reference gene stability in spleen lymphomas induced by infection of commercial broiler chickens with a very virulent plus MDV1 (vv+ TK-2a virus). Two excel based algorithms, (Bestkeeper and Normfinder) were employed to compare reference gene stability. Bestkeeper and Normfinder analysis of reference gene stability in virus- and mock-infected cells, showed that 28S RNA and PPIB displayed higher stability in CEF infections with either oncogenic or vaccine viruses. In addition, both Bestkeeper and Normfinder determined 28S RNA and PPIB to be the most stably-expressed reference genes in vivo in vv+ TK-2ainduced spleen lymphomas. Furthermore, Bestkeeper and Normfinder analyses both determined BACT to be the least stable reference gene during MDV infection of CEF with oncogenic viruses, vaccine viruses, as well as in vv+ TK-2a-induced spleen lymphomas. © 2016 Elsevier B.V. All rights reserved.

1. Introduction Quantitative reverse-transcription primed PCR (qRT-PCR) is widely used for the quantification of cellular mRNA expression, as well as viral RNA replication (Mackay et al., 2002). qRT-PCR offers several advantages over other methodologies for the accurate quantification of mRNA levels in terms of sensitivity, dynamic range, and capacity for multiplexing (Huggett et al., 2005). On the other hand, the sensitivity and robustness of qRT-PCR can be affected by RNA quality and quantity, the presence of inhibitors that co-purify with RNA, the efficiency of the reverse transcription reaction, and the accuracy of pipetting (Bustin and Nolan, 2004). In addition, poor assay design, experimental conditions, and inappropriate normalization strategies compromise the integrity

∗ Corresponding author at: Dept. of Animal and Food Sciences, 052 Townsend Hall, 531 South College Ave, University of Delaware, Newark, DE 19716, United States. E-mail addresses: [email protected] (S.N. Neerukonda), [email protected] (U.K. Katneni), [email protected] (S. Golovan), [email protected] (M.S. Parcells). http://dx.doi.org/10.1016/j.jviromet.2016.07.017 0166-0934/© 2016 Elsevier B.V. All rights reserved.

of the resulting data (Bustin, 2010). Various methods of normalization have been proposed, such as normalization to sample size (cell number), total RNA, and the use of standard internal reference (housekeeping) gene expression (Huggett et al., 2005; Talaat et al., 2002). Normalization to cell number is not possible if the experimental sample is a whole tissue. Normalization to total RNA can be affected by variation in extraction efficiencies, with the final yields being quite low in either quantity or quality. In addition, normalization to total RNA does not take into account the variation in the efficiency of reverse transcription or PCR amplification (Stahlberg et al., 2004). For qPCR, normalization to total DNA poses an additional drawback, due to the presence of multiple haplotypes or genome integrations in tumor cells (Huggett et al., 2005; Kaufer and Flamand, 2014; Talaat et al., 2002) and the presence of multiple copies of particular loci in replicating bacteria in comparison to non-replicating bacteria (Rocha, 2004). Normalization to internal control reference genes has been the most popular and reliable method of normalization, due to the fact that it considers and precludes the error due to initial RNA/cDNA loading, as well as the variation in the efficiency of

112

S.N. Neerukonda et al. / Journal of Virological Methods 236 (2016) 111–116

reverse transcription reaction (Huggett et al., 2005). Normalization to the geometric mean of multiple reference genes has also been suggested as an accurate way of normalization, although it is not always possible to use multiple reference genes due to limitations in sample availability, increased expense, and the possible variability in one of the selected reference genes. This method, however, is considered to be the most rigorous method of normalization. The concept of using reference genes for normalization is based on the assumption that the expression of these genes remains constant under any experimental condition, although this may not be the case in reality. Hence, fluctuations in the expression of reference genes selected as internal controls produce erroneous biological results, leading to the misinterpretation of data (Thellin et al., 1999). Based on the published literature, expression of various reference genes varies among different tissues, under different experimental conditions, at various developmental stages, or during various physiological or disease states (Dheda et al., 2004). Therefore, identification and validation of ideal reference genes with constant or stable expression levels under the given experimental condition is important to produce biologically-relevant data (Schmittgen and Zakrajsek, 2000). qRT-PCR has been widely used to define cellular responses to Marek’s disease virus (MDV) infection or vaccination, and also to elucidate mechanisms of MDV pathogenesis and tumor development (Abdul-Careem et al., 2008a,b,c, 2007, 2009; Garcia-Camacho et al., 2003; Kaiser et al., 2003; Lian et al., 2012; Morgan et al., 2001). Additionally, qRT-PCR has also been used for the relative or absolute quantification of viral genome loads or viremia levels, in an attempt to correlate the amount of lytic or latent virus to various factors associated with the outcome of disease, such as host genotype, vaccination status, cellular or viral gene expression, and viral shedding via feather follicular epithelium (Abdul-Careem et al., 2006; Baigent et al., 2005; Gimeno et al., 2011; Islam et al., 2006). Chicken embryo fibroblast (CEF) cultures or fibroblast-derived cell lines (DF1, OU2, SOgE) have been used for the propagation of MDVs, and for the study of MDV-encoded genes (Hunt et al., 2001; Levy et al., 2005; Parcells et al., 1994; Schumacher et al., 2002). Given the highly cell-associated nature of MDV, infection with MDV is associated with global expression changes in both mRNA and ribosomal RNA levels (Morgan et al., 2001). Target gene normalization based on varying housekeeping gene can therefore produce erroneous results. Hence, it is important to validate the stability of a housekeeping gene, to ensure that the selected reference gene is unaffected by MDV infection both in cell culture and in vivo. In the current study, we have evaluated the relative stability of five commonly-used reference genes: chicken ␤-actin (BACT), 28S RNA, 18S RNA, glyceraldehyde-3-phosphate dehydrogenase (GAPDH) and peptidyl-prolyl-isomerase B (PPIB) as standard internal controls by assessing consistency of their expression in MDV-infected CEF cultures and in MDV-induced splenic lymphomas.

2. Materials and methods 2.1. Cells and viruses For the propagation of MDV oncogenic and vaccine viruses, secondary chicken embryo fibroblasts (CEF) were prepared from 10-day-old specific pathogen free (SPF) embryos (Sunrise Farms, Inc. Catskill, NY). Secondary CEF were propagated in M199 medium supplemented with 3% filtered calf serum, l-glutamine, 1X PSN (Penicillin/Streptomycin/Neomycin) and 1X fungizone (Life Technologies, Carlsbad, CA) and maintained at 37 ◦ C in 5% CO2 . CEF cultures were infected with 5000 PFU of each virus, in triplicate, and harvested upon the appearance of plaques at five days post-

infection. The MDV1 strains used were: CU2 (a mildly-virulent MDV, obtained originally from Dr. K.A. Schat, Cornell University), RB-1B (a vvMDV, originally obtained from Dr. K.A. Schat), rMd5 (originally obtained from Dr. Sanjay Reddy, Texas A & M University), and T KING (TK-2a, a vv + MDV, originally obtained from Dr. John K. Rosenberger, University of Delaware) (Tavlarides-Hontz et al., 2009). Vaccine viruses used were commercially-produced HVT, SB-1, and CVI-988, all provided by Merial, Inc., Gainesville, GA. 2.2. In vivo lymphoma isolation Solid lymphoma masses were isolated from unvaccinated commercial broiler chickens (Hubbard × Cobb) infected via contact with the vv + MDV (TK-2a strain)-inoculated shedder chickens during a vaccine efficacy study. The bird experiment described here followed a natural MDV infection model known as “shedder model” that has already been described elsewhere (Tavlarides-Hontz et al., 2009). Briefly, one-day-old commercial broiler chickens, grouped as “shedders” were infected intra-abdominally with 200 PFU of TK-2a-infected CEF, and were placed on wood shaving-based litter in a multi-room chicken house equipped with independent heating, ventilation, feeders, and waterers. Two weeks post-placement of shedders, one-day-old unvaccinated and vaccinated contact chickens were placed in contact with infected shedders. Frank lymphomas were obtained from spleens (n = 4) during necropsy of the contact-exposed chickens at the end of the seventh week post-placement, and were excised from the surrounding nonlymphomatous spleen tissue. In addition, as an infected positive control, phenotypically non-lymphomatous adjacent tissue sections (n = 4) were also isolated from the lymphoma-containing spleens. To serve as negative controls, healthy normal spleens (n = 3) were isolated from unvaccinated and unchallenged chickens, housed separately. Spleen samples were collected into RNA later (Ambion Inc., Austin, TX) and stored at −80 ◦ C for further RNA purification. The animal experiment from which these samples were obtained was conducted in accordance to Ag Animal Care and Use Committee (AACUC) guidelines of University of Delaware registered as Vaccine Trial protocol: (22) 04-15-10a. 2.3. qRT-PCR analysis Total RNA was isolated from mock- or MDV-infected cells and spleen tissues using Qiagen RNA/DNA/Protein Kit according to the manufacturer’s instructions (Qiagen, USA). Total RNA quality (260/280 ratio) and quantity (at 260 nm absorbance) were measured using an Agilent Nanodrop spectrophotometer. RNA samples with OD 260/280 and OD 260/230 > 2 only were selected for further experimentation. For each sample, 1 ␮g of total RNA was reverse transcribed with random hexamers using the High Capacity cDNA Reverse Transcription Kit (Applied Biosystems, USA), as recommended by the manufacturer’s protocol (Step 1: 25 ◦ C for 10 min; Step 2: 37 ◦ C for 120 min; Step 3: 85 ◦ C for 5 min; Step 4: 4 ◦ C storage). The final cDNAs were diluted 10 fold with nuclease-free water and 1 ␮l of the final diluted cDNA was used in a 20 ␮l reaction composed of 10 ␮l iQTM SYBR® Green Supermix (Bio-Rad, Hercules, CA), 8.2 ␮l of nuclease free water, and 250 nM of each forward and reverse primer. Quantitative Real Time PCR (qRT-PCR) was performed using SYBR green chemistry, as recommended by the manufacturer (Bio-rad Laboratories) on the MyiQ2 Two Color Real Time PCR Detection System (Bio-rad Laboratories, Hercules, CA). Following amplification and collection of raw fluorescence data, melt curve analysis was performed to exclude the possibility of

S.N. Neerukonda et al. / Journal of Virological Methods 236 (2016) 111–116

113

Table 1 Primers used for qRT-PCR Analysis. Reference gene

Tm

Primer sequence

Amplicon size (bp)

BACT

58 ◦ C

101

28 S RNA

58 ◦ C

18 S RNA

55 ◦ C

GAPDH

55 ◦ C

CACAGATCATGTTTGAGACCTT (Fwd) CATCACAATACCAGTGGTACG (Rev) GGTATGGGCCCGACGCT (Fwd) CCGATGCCGACGCTCAT (Rev) TCAGATACCGTCGTAGTTCC (Fwd) TTCCGTCAATTCCTTTAAGTT (Rev) GGAGTCCACTGGTGTCTTCA (Fwd) AGCACCACCCTTCAGATGAG (Rev) GAGAAAGGGTTCGGCTTCA (Fwd) GAGAAAGGGTTCGGCTTCA (Rev)

PPIB

◦

55 C

non-specific amplification. All reactions bore similar PCR amplification efficiencies. 2.4. Primer design Primer sequences for the chicken reference genes were either obtained from other studies (BACT, 28S RNA, 18S RNA) (De Boever et al., 2008; Li et al., 2005; Bestkeeper, 2015; Normfinder, 2015; Reffinder, 2015), or were designed across intronic regions for these studies (GAPDH and PPIB) (see Table 1, below).

144 154 60 144

Table 2 Bestkeeper Analysis of Housekeeping Genes. Treatment

BACT

28S RNA

18S RNA

GAPDH

PPIB

Oncogenic viruses (n = 15) Vaccine viruses (n = 12) TK-2a spleen lymphomas (n = 12) Overall SD average

0.85 0.83 1.20 0.96

0.30 0.39 0.62 0.44

0.61 0.38 0.84 0.61

0.54 0.44 0.91 0.63

0.37 0.35 0.72 0.48

Table 3 Pairwise correlation analyses of Reference Genes with Bestkeeper index. Bestkeeper coefficient of correlation (HKG vs Bestkeeper index)

2.5. Reference Gene Stability Analysis Candidate reference gene stability was evaluated using two excel-based algorithms Bestkeeper and Normfinder (Andersen et al., 2004; Stahlberg et al., 2004). Separate analyses were carried out for cell culture infections, and MDV-induced splenic lymphomas. The results obtained from Bestkeeper and Normfinder were compared and confirmed with analyses using geNorm, delta Ct (Ct) and a web-based tool known as Reffinder. Bestkeeper determines gene expression stability by calculating the standard deviation [SD (±Ct)] and coefficient of variance [CV] from the input raw Ct values. Therefore the reference gene having the lowest standard deviation is considered to be the most stably-expressed. Reference genes with standard deviation >1 are considered less stable. In addition, the geometric mean of Ct values for stably-expressed reference genes in a given sample is combined into a Bestkeeper Index (BI). Bestkeeper then performs pairwise correlation analyses for all pairs of candidate genes and assigns a Pearson correlation coefficient (r value), and probability value (p value) for each gene combination in order to determine the relationship between them. Genes with high correlation are combined into an index, and the correlation between each gene and index is calculated as a correlation coefficient (r value). The genes with significantly higher correlation coefficients (r value) are considered to be the most stable reference genes. The average raw Ct values of technical duplicates were imported into Bestkeeper (version 1) to calculate the stability. Normfinder calculates inter-group and intra-group variations and combines these to calculate the stability value for each gene. Since the stability value is a combination of two sources of variation, it represents a practical measure of systematic error introduced by a gene when used as a reference gene. The gene with lowest stability value (M number) is considered to be the most stably-expressed. For each reference gene of interest, the Ct values were converted into relative quantities of the lowest Ct value, which is set to 1. The log-transformed data is then analyzed by Normfinder to calculate the stability values. The Bestkeeper algorithm is resistant to sampling errors, while it is a requisite that none of the selected genes for Bestkeeper analysis are co-regulated. Normfinder is relatively less sensitive to co-regulated genes, while it could be sensitive to sampling errors

Treatment Oncogenic viruses Vaccine viruses TK-2a spleen lymphomas a

BACT 0.668a 0.021 0.945a

28S RNA 0.892a 0.727a 0.892a

18S RNA 0.812a 0.467a 0.967a

GAPDH 0.746a 0.735 a 0.859a

PPIB 0.655a 0.249 0.931a

Indicates significance at p < 0.05 level.

(Andersen et al., 2004). Hence the use of a combination of both Bestkeeper and Normfinder provides a more robust analysis in evaluating the stability of candidate reference genes. geNorm calculates stability values (M) based on arithmetic mean of all pair-wise variations between a particular gene and all other reference genes in the study panel (Andersen et al., 2004). 2−Ct method was used calculate relative expression of each housekeeping gene in virus infected cells or tissues compared to uninfected controls (Livak and Schmittgen, 2001). Reffinder is a web-based tool that combines geNorm, Normfinder, BestKeeper and Ct methods to compare and rank each reference gene. 3. Results According to Bestkeeper analysis, none of the candidate reference genes displayed a SD ± Ct value greater than 1 except BACT in TK-2a-induced lymphomas [±1.2], indicating that all of the reference genes in the current study, with the exception of BACT, were suitable to be considered as reference genes (Table 2). BACT was determined to be the least stable reference gene with a higher standard deviation [SD (±Ct)] during cell culture infections with either oncogenic [±0.85] or vaccine viruses [±0.83]. Although the standard deviation was not greater than 1, BACT had the greatest standard deviation among the validated reference genes during cell culture infections with either oncogenic or vaccine viruses. Corroborating this, BACT displayed a significant 1.78 fold increase in expression in oncogenic virus infected cell treatments (Fig. 1A). Furthermore, reffinder ranking displayed BACT as the least stable reference gene in each study treatment (Table 5). Among the most stable reference genes, 28S RNA displayed the greatest stability with the lowest standard deviation during CEF infection with oncogenic viruses [±0.3] and in TK-2a-induced spleen lymphomas [±0.62], while PPIB displayed the greatest stability during vaccine virus CEF infection [±0.35]. Correspondingly, reffinder ranked 28S

114

S.N. Neerukonda et al. / Journal of Virological Methods 236 (2016) 111–116

Fig. 1. Relative expression of five reference genes via Ct analysis. The graph above depicts relative gene expression in oncogenic MDV-infected (black bars), vaccine virus-infected (gray bars) CEF, and MDV-induced spleen tumors (diagonallyshaded bars) compared to uninfected controls. The dotted line represents baseline expression set to 1. The Y-axis represents fold change in expression (Ct method). The X-axis shows the reference genes tested. The asterisks denote significance at p<0.05 level.

Fig. 2. Graph representing Normfinder analysis of stability in their order of ranking from least stable to most stable. The Y-axis represents stability number or M value, while the X-axis lists the reference genes included in the analysis.

Table 5 Reffinder ranking of reference gene stability. Comprehensive ranking as determined by Reffinder

Table 4 geNorm Analysis of Housekeeping Genes. Average expression stability value (M value) Treatment Oncogenic viruses Vaccine viruses TK-2a spleen lymphomas

BACT 0.830 1.100 0.710

28S RNA 0.162 0.455 0.625

18S RNA 0.589 0.384 0.670

GAPDH 0.589 0.384 0.670

PPIB 0.384 0.396 0.407

and PPIB on the top in terms of stability in oncogenic and vaccine virus treatments respectively. Overall, 28S RNA and PPIB displayed the greatest stability during CEF infection with oncogenic MDVs [0.3 < SD < 0.37] and vaccine strains [0.35 < SD < 0.39]. In TK-2a-induced spleen tumors, PPIB [SD ± 0.72] ranked second in stability after 28S RNA [SD ± 0.62]. With respect to correlation coefficients between Bestkeeper index and the studied reference genes, there was some inconsistency (Table 3). This could be due to the finding that Bestkeeper’s correlation coefficients and the corresponding p values are limited to groups not having heterogeneous variance between gene expression levels. However, there was a difference in the expression levels (Ct values) of selected reference genes, contributing to the observed significant variances. The Ct values in our study varied from ∼13 (28 S RNA), ∼16 (18 S RNA), ∼22 (BACT), to ∼25 (GAPDH and PPIB). In general, the most stably-expressed reference gene according to Bestkeeper analysis was 28S RNA, which displayed a significantly higher correlation with the Bestkeeper index. According to the Normfinder analysis (Fig. 2), both 28S RNA and PPIB were considered to be the most stably-expressed reference genes, with the lowest M values or stability numbers among all the treatments and reference genes under consideration. Consistent with the Bestkeeper analysis, BACT was determined to be the least stable reference gene, having higher M values. Furthermore, analogous to Bestkeeper and Normfinder results, geNorm analysis displayed BACT as least stable with highest average gene stability M value (Table 4). While geNorm algorithm displayed greater stability for 28S RNA and PPIB in oncogenic viruses treatment, GAPDH was found to be comparably stable along with PPIB and 28S RNA in vaccine viruses treatment. Intriguingly, geNorm determined 18S RNA to possess higher stability in TK-2a-induced spleen lymphomas treatment. PPIB was comparably stable but fell only after 18S RNA. Altogether, both Bestkeeper and Normfinder determined 28S RNA and PPIB to be the most stable reference gene candidates among the five reference genes tested, under the given experimen-

Rank # 1 2 3 4 5

Oncogenic viruses 28S RNA PPIB GAPDH 18S RNA BACT

Vaccine viruses PPIB 28S RNA GAPDH 18S RNA BACT

TK-2a spleen lymphomas PPIB 18S RNA 28S RNA GAPDH BACT

tal conditions, while BACT proved to be the least stable reference gene. Furthermore, we observed discrete levels of variation among CEFs infected with oncogenic MDVs, vaccine viruses and TK-2ainduced spleen lymphomas (Table 2). The levels of variation (SD) being higher in TK-2a-induced spleen lymphomas among all the reference genes tested (Table 2). Although the Normfinder stability values (M values) among the three treatments in the present study displayed similar trends for every reference gene, stability (M) values were highest for TK-2a induced lymphomas (Fig. 2), compared to vaccine virus and oncogenic virus CEF infections. These trends further corroborated the observed discrete levels of variation with Bestkeeper analysis. A possible explanation of differences in variation could be ascribed to the cell type and/or phase of infection. Spleen lymphomas are comprised of mixed lymphocytes, the transformed component of which is latently-infected CD4+ T lymphocytes (Burgess and Davison, 2002). CEFs are more homogeneous and are lytically-infected. The transformed tumor cells showed higher levels of metabolic deregulation (GAPDH ∼0.91), cytoskeletal changes (BACT ∼1.2), increased rates of protein translation (28S RNA ∼0.62, 18S RNA ∼0.84) and cis-prolyl isomerization (PPIB ∼0.72), ostensibly to meet the demands of rapid proliferation, which is not the case during lytic replication in cell culture. 4. Discussion Our data are consistent with previous reports, where BACT was found to be the least stable reference gene in the context of herpesvirus infection (Radonic et al., 2005; Watson et al., 2007). Cellular actin content plays an important role during herpesvirus infection, from viral entry to egress. Herpes simplex virus 1 (HSV1) entry via fusion is dependent on cortical actin (Clement et al., 2006). During entry, HSV-1 stimulates Rho GTPase signaling (Rho1, CDC42, Rac1), altering the structure of cortical actin. For instance, the movement of nucleocapsids towards the inner nuclear membrane is ATP-dependent and sensitive to the actin depolymerizing

S.N. Neerukonda et al. / Journal of Virological Methods 236 (2016) 111–116

agent latrunculin A, but not microtubule depolymerizing agent cytochalasin D (Forest et al., 2005). During HSV-1 infection, the Us3-encoded protein kinase alters actin dynamics. Deletion of the US 3 protein kinase gene from HSV1, pseudorabies virus (PRV) and MDV-1 results in the accumulation of nucleocapsids between the two leaflets of nuclear membrane, with a subsequent reduction in viral titers (Klupp et al., 2001; Purves et al., 1991; Reynolds et al., 2002; Ryckman and Roller, 2004; Schumacher et al., 2008, 2005). Upon MDV-1 infection of chicken embryo cells, US 3 null mutants have a decreased burst size, with an increased accumulation of enveloped virions at the perinuclear space, corresponding to a decrease in number of viral particles in cytoplasm (Schumacher et al., 2005). The US 3 kinase was demonstrated to mediate cytoskeletal rearrangement during earlier phases via transient actin stress fiber break down with a subsequent regeneration, possibly due to the cellular stress response (Schumacher et al., 2005). Regarding the stable expression of other housekeeping genes, Watson et al., noted PPIA (Peptidyl-prolyl-isomerase A), GAPDH, and SDHA (Succinate dehydrogenase complex subunit A) were stable reference genes in decreasing order of stability in HSV-, CMVand VZV-infected cells (Watson et al., 2007). A study by Radonic et al., noted that TBP and PPIA were the most stably-expressed reference genes during cellular infection with CMV, HHV-6A, CAMP (Camelpox virus), SARS and YF (Yellow fever) viruses (Radonic et al., 2005). In addition, a recent study identified PPIA as the most stable reference gene with TBP being the least stable in HHV-6B infected Molt-3T cell line (Engdahl et al., 2016). In our study, we did not include PPIA as a reference gene; however, we did include a closelyrelated cyclophilin family member, PPIB, which was found to be equally stable along with 28S RNA in the context of MDV infection (this study). In conclusion, we report that 28S RNA and PPIB are the most stably expressed reference genes during MDV infection, while BACT proved to be the least stable reference gene, with GAPDH and 18S RNA being only slightly more stable during MDV infection of CEFs in vitro or in vv+ MDV1 induced lymphomas in vivo. Due to lack of a poly-A tail, 28S RNA cannot be employed as a reference gene when reverse transcription priming is via oligo-dT. Under those circumstances, PPIB can serve as a stable reference gene, when the reverse transcription priming is via either random oligos or oligo dT. Normalization based on the geometric mean of the reference gene pair, 28S RNA and PPIB, may provide a more robust control for expression data during the course of MDV infection.

Competing interests The authors have no competing interests in the performance or reporting of this work.

Author contribution Sabari Nath Neerukonda: Mr. Neerukonda performed RNA isolation, primer research and design, qRT-PCR, data analysis and wrote drafts of the manuscript. Dr. Upendra K. Katneni: Aided in sample preparation, RNA isolation and data analysis. Dr. Sergey Golovan: Initiated and initially directed the project and provided funding. Dr. Mark S. Parcells: Provided all viruses, primary cell cultures, tumor samples and finished directing and funding the project, edited and submitted the manuscript.

115

Acknowledgements This work was supported, in part, by the State of Delaware through its allocation to Poultry Disease Research (PDR). Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.jviromet.2016. 07.017. References Abdul-Careem, M.F., Hunter, B.D., Nagy, É., Read, L.R., Sanei, B., Spencer, J.L., Sharif, S., 2006. Development of a real-time PCR assay using SYBR Green chemistry for monitoring Marek’s disease virus genome load in feather tips. J. Virol. Methods 133, 34–40. Abdul-Careem, M.F., Hunter, B.D., Parvizi, P., Haghighi, H.R., Thanthrige-Don, N., Sharif, S., 2007. Cytokine gene expression patterns associated with immunization against Marek’s disease in chickens. Vaccine 25, 424–432. Abdul-Careem, M.F., Hunter, B.D., Lee, L.F., Fairbrother, J.H., Haghighi, H.R., Read, L., Parvizi, P., Heidari, M., Sharif, S., 2008a. Host responses in the bursa of Fabricius of chickens infected with virulent Marek’s disease virus. Virology 379, 256–265. Abdul-Careem, M.F., Hunter, B.D., Sarson, A.J., Parvizi, P., Haghighi, H.R., Read, L., Heidari, M., Sharif, S., 2008b. Host responses are induced in feathers of chickens infected with Marek’s disease virus. Virology 370, 323–332. Abdul-Careem, M.F., Hunter, D.B., Lambourne, M.D., Read, L.R., Parvizi, P., Sharif, S., 2008c. Expression of cytokine genes following pre- and post-hatch immunization of chickens with herpesvirus of turkeys. Vaccine 26, 2369–2377. Abdul-Careem, M.F., Read, L.R., Parvizi, P., Thanthrige-Don, N., Sharif, S., 2009. Marek’s disease virus-induced expression of cytokine genes in feathers of genetically defined chickens. Dev. Comp. Immunol. 33, 618–623. Andersen, C.L., Jensen, J.L., Orntoft, T.F., 2004. 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, 5245–5250. Baigent, S.J., Petherbridge, L.J., Howes, K., Smith, L.P., Currie, R.J.W., Nair, V.K., 2005. Absolute quantitation of Marek’s disease virus genome copy number in chicken feather and lymphocyte samples using real-time PCR. J. Virol. Methods 123, 53–64. Bestkeeper. http://www.gene-quantification.de/bestkeeper.html. (Date accessed 5.5.15). Bustin, S.A., Nolan, T., 2004. Pitfalls of quantitative real-time reverse-transcription polymerase chain reaction. J. Biomol. Tech.: JBT 15, 155–166. Bustin, S.A., 2010. Why the need for qPCR publication guidelines? – The case for MIQE. Methods 50, 217–226. Clement, C., Tiwari, V., Scanlan, P.M., Valyi-Nagy, T., Yue, B.Y., Shukla, D., 2006. A novel role for phagocytosis-like uptake in herpes simplex virus entry. J. Cell Biol. 174, 1009–1021. Burgess, S.C., Davison, T.F., 2002. Identification of the neoplastically transformed cells in Marek’s disease herpesvirus-induced lymphomas: recognition by the monoclonal antibody AV37. J. Virol. 76, 7276–7292. De Boever, S., Vangestel, C., De Backer, P., Croubels, S., Sys, S.U., 2008. Identification and validation of housekeeping genes as internal control for gene expression in an intravenous LPS inflammation model in chickens. Vet. Immunol. Immunopathol. 122, 312–317. Dheda, K., Huggett, J.F., Bustin, S.A., Johnson, M.A., Rook, G., Zumla, A., 2004. Validation of housekeeping genes for normalizing RNA expression in real-time PCR. BioTechniques 37, 112–114, 116, 118–119. Engdahl, E., Dunn, N., Fogdell-Hahn, A., 2016. Investigation of reference gene expression during human herpesvirus 6B infection indicates peptidylprolyl isomerase A as a stable reference gene and TATA box binding protein as a gene up-regulated by this virus. J. Virol. Methods 227, 47–49. Forest, T., Barnard, S., Baines, J.D., 2005. Active intranuclear movement of herpesvirus capsids. Nat. Cell Biol. 7, 429–431. Garcia-Camacho, L., Schat, K.A., Brooks Jr., R., Bounous, D.I., 2003. Early cell-mediated immune responses to Marek’s disease virus in two chicken lines with defined major histocompatibility complex antigens. Vet. Immunol. Immunopathol. 95, 145–153. Gimeno, I.M., Witter, R.L., Cortes, A.L., Reed, W.M., 2011. Replication ability of three highly protective Marek’s disease vaccines: implications in lymphoid organ atrophy and protection. Avian Pathol.: J. W.V.P.A 40, 573–579. Huggett, J., Dheda, K., Bustin, S., Zumla, A., 2005. Real-time RT-PCR normalisation; strategies and considerations. Genes Immun. 6, 279–284. Hunt, H.D., Lupiani, B., Miller, M.M., Gimeno, I., Lee, L.F., Parcells, M.S., 2001. Marek’s disease virus down-regulates surface expression of MHC (B Complex) Class I (BF) glycoproteins during active but not latent infection of chicken cells. Virology 282, 198–205. Islam, A., Cheetham, B.F., Mahony, T.J., Young, P.L., Walkden-Brown, S.W., 2006. Absolute quantitation of Marek’s disease virus and Herpesvirus of turkeys in

116

S.N. Neerukonda et al. / Journal of Virological Methods 236 (2016) 111–116

chicken lymphocyte, feather tip and dust samples using real-time PCR. J. Virol. Methods 132, 127–134. Kaiser, P., Underwood, G., Davison, F., 2003. Differential cytokine responses following Marek’s disease virus infection of chickens differing in resistance to Marek’s disease. J. Virol. 77, 762–768. Kaufer, B.B., Flamand, L., 2014. Chromosomally integrated HHV-6: impact on virus, cell and organismal biology. Curr. Opin. Virol. 9, 111–118. Klupp, B.G., Granzow, H., Mettenleiter, T.C., 2001. Effect of the pseudorabies virus US3 protein on nuclear membrane localization of the UL34 protein and virus egress from the nucleus. J. Gen. Virol. 82, 2363–2371. Levy, A.M., Gilad, O., Xia, L., Izumiya, Y., Choi, J., Tsalenko, A., Yakhini, Z., Witter, R., Lee, L., Cardona, C.J., Kung, H.J., 2005. Marek’s disease virus Meq transforms chicken cells via the v-Jun transcriptional cascade: a converging transforming pathway for avian oncoviruses. Proc. Natl. Acad. Sci. U. S. A. 102, 14831–14836. Li, Y.P., Bang, D.D., Handberg, K.J., Jorgensen, P.H., Zhang, M.F., 2005. Evaluation of the suitability of six host genes as internal control in real-time RT-PCR assays in chicken embryo cell cultures infected with infectious bursal disease virus. Vet. Microbiol. 110, 155–165. Lian, L., Qu, L.J., Sun, H.Y., Chen, Y.M., Lamont, S.J., Liu, C.J., Yang, N., 2012. Gene expression analysis of host spleen responses to Marek’s disease virus infection at late tumor transformation phase. Poult. Sci. 91, 2130–2138. Livak, K.J., Schmittgen, T.D., 2001. Analysis of relative gene expression data using real-time quantitative PCR and the 2(-Delta Delta C(T)) Method. Methods 25, 402–408. Mackay, I.M., Arden, K.E., Nitsche, A., 2002. Real-time PCR in virology. Nucleic Acids Res. 30, 1292–1305. Morgan, R.W., Sofer, L., Anderson, A.S., Bernberg, E.L., Cui, J., Burnside, J., 2001. Induction of host gene expression following infection of chicken embryo fibroblasts with oncogenic Marek’s disease virus. J. Virol. 75, 533–539. Normfinder. http://moma.dk/normfinder-software. (Date accessed 5.5.15). Parcells, M.S., Anderson, A.S., Cantello, J.L., Morgan, R.W., 1994. Characterization of Marek’s disease virus insertion and deletion mutants that lack US1 (ICP22 homolog), US10, and/or US2 and neighboring short-component open reading frames. J. Virol. 68, 8239–8253. Purves, F.C., Spector, D., Roizman, B., 1991. The herpes simplex virus 1 protein kinase encoded by the US3 gene mediates posttranslational modification of the phosphoprotein encoded by the UL34 gene. J. Virol. 65, 5757–5764. Radonic, A., Thulke, S., Bae, H.G., Muller, M.A., Siegert, W., Nitsche, A., 2005. Reference gene selection for quantitative real-time PCR analysis in virus infected cells: SARS corona virus, Yellow fever virus, Human Herpesvirus-6, Camelpox virus and Cytomegalovirus infections. Virol. J. 2, 7. Reffinder. http://fulxie.0fees.us/?ckattempt=1. (Date accessed 5.5.15).

Reynolds, A.E., Wills, E.G., Roller, R.J., Ryckman, B.J., Baines, J.D., 2002. Ultrastructural localization of the herpes simplex virus type 1 UL31, UL34, and US3 proteins suggests specific roles in primary envelopment and egress of nucleocapsids. J. Virol. 76, 8939–8952. Rocha, E.P., 2004. The replication-related organization of bacterial genomes. Microbiology 150, 1609–1627. Ryckman, B.J., Roller, R.J., 2004. Herpes simplex virus type 1 primary envelopment: UL34 protein modification and the US3-UL34 catalytic relationship. J. Virol. 78, 399–412. Schmittgen, T.D., Zakrajsek, B.A., 2000. Effect of experimental treatment on housekeeping gene expression: validation by real-time, quantitative RT-PCR. J. Biochem. Biophys. Methods 46, 69–81. Schumacher, D., Tischer, B.K., Teifke, J.P., Wink, K., Osterrieder, N., 2002. Generation of a permanent cell line that supports efficient growth of Marek’s disease virus (MDV) by constitutive expression of MDV glycoprotein E. J. Gen. Virol. 83, 1987–1992. Schumacher, D., Tischer, B.K., Trapp, S., Osterrieder, N., 2005. The protein encoded by the US3 orthologue of Marek’s disease virus is required for efficient de-envelopment of perinuclear virions and involved in actin stress fiber breakdown. J. Virol. 79, 3987–3997. Schumacher, D., McKinney, C., Kaufer, B.B., Osterrieder, N., 2008. Enzymatically inactive U(S)3 protein kinase of Marek’s disease virus (MDV) is capable of depolymerizing F-actin but results in accumulation of virions in perinuclear invaginations and reduced virus growth. Virology 375, 37–47. Stahlberg, A., Kubista, M., Pfaffl, M., 2004. Comparison of reverse transcriptases in gene expression analysis. Clin. Chem. 50, 1678–1680. Talaat, A.M., Howard, S.T., Hale, W.t., Lyons, R., Garner, H., Johnston, S.A., 2002. Genomic DNA standards for gene expression profiling in Mycobacterium tuberculosis. Nucleic Acids Res. 30, e104. Tavlarides-Hontz, P., Kumar, P.M., Amortegui, J.R., Osterrieder, N., Parcells, M.S., 2009. A deletion within glycoprotein L of Marek’s disease virus (MDV) field isolates correlates with a decrease in bivalent MDV vaccine efficacy in contact-exposed chickens. Avian Dis. 53, 287–296. 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. Watson, S., Mercier, S., Bye, C., Wilkinson, J., Cunningham, A.L., Harman, A.N., 2007. Determination of suitable housekeeping genes for normalisation of quantitative real time PCR analysis of cells infected with human immunodeficiency virus and herpes viruses. Virol. J. 4, 130.