Three newly identified Immediate Early Genes of Bovine herpesvirus 1 lack the characteristic Octamer binding motif- 1

Only three immediate early genes (IE) BICP0, BICP4 and BICP22 of Bovine herpesvirus 1 (BoHV-1) are known. These genes are expressed coordinately and their promoters are well characterized. We provide evidence for expression of three additional IE genes of BoHV-1 i.e. UL21, UL33 and UL34. These genes are expressed in the presence of cycloheximide (CH) at the same time as known IE genes. Surprisingly, the promoters of newly identified IE genes (UL21, UL33, UL34) lack the OCT-1 binding site, a considered site of transactivation of the BoHV-1 IE genes. The other difference in the promoters of the newly identified IE genes is the presence of TATA box at near optimal site. However, all the IE genes have similar spatial placements of C/EBPα, DPE and INR elements.

binding of OCT-1 protein with α-TIF brings the complex into the proximity of the transcription start sites [18][19][20][21] . Katan et al. have shown that deletion in POU domain of OCT-1 protein affects complex formation with α-TIF 22 .
There is a report which indicates that OCT-1 protein is not the only cellular factor that can transactivate BoHV-1 gene expression 25 . The study suggested the CCAAT enhancer-binding protein alpha (C/EBPα) has the ability to transactivate IE genes. C/EBPs are a family of transcription factors that contains a conserved, basicleucine zipper domain involved in DNA binding 26 and belongs to the family of leucine zipper class DNA binding protein 27 .
The transcriptional regulation and computational studies of BoHV-1 in the past have been focused on the three major IE genes [3][4][5]8 . A random insertional mutagenesis study by Robinson et al. mentioned seven genes of unknown regulatory control 28 . This study aimed to ascertain the temporal expression patterns of three BoHV-1 genes viz. UL21, UL33 and UL34 and also evaluated their possible transcriptional regulatory elements.

Results
The IE nature of UL21, UL33 and UL34 genes. To assess if UL21, UL33 and UL34 genes are IE genes, the expression of these genes was studied in BoHV-1 infected cells in the presence and absence of cycloheximide (CH). The transcription of the known IE gene, BICP0 as a control was not inhibited by CH (Fig. 1E) while CH effectively inhibited the expression of early gene thymidine kinase (TK) and late gene glycoprotein B (gB) without affecting the expression of IE genes (Fig. 1D). Figure 1A-C shows the amplification of UL21, UL33, UL34. BICP0, TK and gB genes at different time point in BoHV-1 infected cells in the presence and absence of CH. The sequence of the PCR amplified product was identical to that of the targeted template region (data not shown) 3,6,29 . The GAPDH control confirmed the quantity of the total RNA templates ( Supplementary Fig. 1G).
The experiment was repeated three times and found similar results. Results indicate that UL21, UL33 and UL34 are IE genes like BICP0, BICP4 and BICP22 since their expression is not inhibited in presence of CH.
Promoters of UL21, UL33 and UL34 genes. To delineate the promoter regions of UL21, UL33 and UL34 genes, the transcription start sites (TSS) of these genes were first identified by 5′-RLM-RACE. Each gene gave one amplification product differing in size in RLM-RACE ( Fig. 2A-C) suggesting one TSS for each gene. The BICP0, BICP4 and BICP22 also have one transcription start site 9 . The distance between TSS and putative translation start site (ATG) varied among these three genes. This distance was 21 bp in UL21, 127 bp in UL33 and 48 bp in UL34.
Comparative computational analysis of UL21, UL33 and UL34 genes with BICP0, BICP4 and BICP22. The comparative computational analysis between the putative promoters of UL21, UL33, UL34 and BICP0, BICP4, BICP22 genes revealed some interesting findings. For computational analysis, we took 1000 bp upstream and 100 bp downstream of the identified TSS sites. The UL21, UL33 and UL34 group showed presence of TATA boxes at optimal position 30-32 (−30, −28 and −25 bp respectively) from the TSS as shown in Fig. 3 while the TATA box was absent in BICP4, BICP0 and BICP22 at appropriate position though TATA box like sequence was observed at position far upstream from the TSS (−251, −143 and −359 bp respectively). The OCT-1 like sequences were absent in the UL group of IE genes despite intensive examination with relaxed parameters keeping three base mismatch. The OCT-1 motifs were identified in promotor regions of the three known IE genes (Fig. 3). INR (Initiator elements) and DPE (Downstream Processing elements) are important regions in core promoters required for RNA polymerase binding [33][34][35] . Their presence aids in the prediction of probable promoter sites. INR and DPE were present in both the groups of IE genes and their spatial positioning was almost similar. The optimum position of DPE is at +28 to +32 from the TSS 32,34 . The maximum variation from the optimum position observed was 10 bp.
Comparative expression of basal promoters of UL 21and BICP0 genes. The UL group of IE genes contains TATA box in optimal positions 30 while TATA boxes identified in promoter regions of the BICP group were not at the optimal position suggesting that they have no role in RNA polymerase binding. To test if TATA box in optimal position influences basal promoter activity of UL and BICP promoters, the promoter region 300 bp upstream from TSS of one representative from each group (UL21 TATA box at −30 and BICP0 TATA box at −148 base upstream) was cloned in promoterless pDSred Express 2-1 vector and transfected into MDBK cells. The expression of reporter gene was measured in flow cytometer (Fig. 4). The data shows higher expression of UL21 promoter as compared to BICP0 promoter. However, this was not reflected in the temporal expression by real time PCR. Real Time PCR showed almost similar expression of both UL and BICP group of IE genes (Fig. 5). The expression study results suggest that the factors upstream of 300 bp may influence expression in vivo. Flow cytometry (translation) and real time PCR (transcription) data reflect two different parameters which are not necessarily comparable. Since, we have used same reporter gene having identical sequence, it is assumed that they will have same degradation, stability, translation efficiency etc. and so may reflect mRNA copy number.

Discussion
Presence of an OCT-1 motif and its other possible variants is an important feature in all the IE genes in HHV-1 identified to date. It is assumed that binding of OCT-1 by bTIF helps in the recruitment of this protein complex to the OCT-1 motif (IFTATGRAAT sequence motif) to initiate the transcription of IE genes in the BoHV-1 infection process 22 . This is the standard model for the beginning of transcription in HHV-1 and many other herpesviruses as well 4,5,10,12,14,36 . There are reports showing that C/EBPα can transactivate IE genes of BoHV-1 25,[37][38][39] , in vitro and also that bTIF can activate promoters that lack OCT-1 motifs provided an alternate mechanism for DNA binding is available 10 . We recently reanalyzed the promoters of known IE gene of BoHV-1 9 and observed that instead of two 4,5,40 , there are three IE transcription units and each of the transcription units have similarly placed INR, DPE and OCT-1 motifs thus reinforcing the hypothesis of OCT-1 having a role in the transactivation of IE genes in BoHV-1.
In this study, we identified three BoHV-1 IE genes that lack OCT-1 site but have a similarly located C/EBPα site as present in known IE genes. Since C/EBPα has the ability to transactivate genes, it is possible that the newly identified IE genes are getting transactivated through it 25 . This is a reasonable assumption which would need further experimentation to prove it. The second point to consider is the functional role if any for these newly identified IE genes. At present, the expression of these genes in virus infection in vitro or in vivo is not known consequently no functional role is ascribed to them. Putatively, they code for late proteins (UL21: tegument protein, UL3: DNA packaging, UL34: nuclear egress protein) 28,41 . Based on this information, it is difficult to make any guess on their functional role at present.
Identifying gene expression cascade is important in virus multiplication and infection and can potentially be used for identifying targets for intervention. A new gene expression cascade has been identified in this study. Further work is needed to explore the importance of these IE genes in BoHV-1 and possibly other herpesviruses.    Cycloheximide treatment. Sub-confluent (60-70% confluency) MDBK (maintained at Immunochemistry laboratory, Biochemistry Division of IVRI) monolayer in 12 well culture plate were infected with BoHV-1 at 10 multiplicity of infection (moi) along with CH at a concentration of 100 µg/mL and incubated at 37 °C until harvesting of RNA at different time intervals (1, 2, 3, 4, 5 hr.). CH was used to identify IE transcripts and added as described by Misra et al. 3 and Wirth et al. 3,8 . BoHV-1 infected MDBK cells (without CH) were used as control.
To check the efficiency of CH treatment, known early and late genes of BoHV-1 were used as controls in PCR. RNA was harvested at 6, 8, 10 and 12, 14, 16 hours post infection for TK for early gene and gB for late gene marker respectively. GAPDH primer pair (GAPDH/F and GAPDH/R as mentioned in Table 1), was used as an internal control for RNA quantity and amplification efficiency. RNA Isolation. RNA was extracted at different time intervals i.e., 1, 2, 3, 4 and 5 hour post infection. For extraction of total RNA, 1 mL TRIZOL (Life technologies) was added directly to the virus infected monolayer and mixed by pipetting. The rest of the RNA extraction was as per the manufacturer's protocols. The RNA pellet recovered after centrifugation at 10,000 × g for 15 minutes was washed once with 70% ethanol and then resuspended in 20 µL of nuclease free water. Before cDNA preparation and PCR amplification, isolated total RNA was treated with DNase I (1µ/µL) (Thermo scientific) at 37 °C for 30 min, followed by enzyme inactivation with 50 mM EDTA (2 µL) at 65 °C for 10 min. Same method for RNA isolation was followed for 6, 8, 10 and 12, 14 and 16 hour post infection, DNase treated RNA was used as negative control in PCR.
Oligonucleotide primers. BoHV-1 complete genome (AJ004801) from NCBI GenBank was used as reference to design gene specific forward and reverse primers to amplify the 5′ regions of the UL21, UL33 and UL34 genes. For RACE, universal primers (outer and inner) were provided with the RLM-RACE kit to recognize the 5′ end of the cDNA. Two sets of outer and inner (nested) gene specific reverse primers were designed to amplify the part of UL21, UL33 and UL34 genes. All the primer sets for PCR reactions were designed using GeneTool Lite (Biotools Incorporated Ltd.). The primer sequences and other relevant details of all the primers used in this study are given in Tables 1 and 2    primer (outer) provided in the kit and gene-specific reverse primer. The reaction was carried out in a thermal cycler (Biometra, U.K.). The cycling conditions were denaturation at 95 °C for 2 minutes, followed by 34 cycles of denaturation at 95 °C for 30 seconds, annealing touchdown (from 65 to 55 °C) for 45 seconds and extension at 72 °C for 1 minute, with a final extension at 72 °C for 5 minutes. After amplification, the reaction products were diluted 20-fold with water, and 3 µl DNA were then re-amplified for 30 cycles using the same reaction conditions with nested and adaptor primer (inner primer provided in the kit). Positive PCR products were eluted from 1% agarose gel and ligated into pGEM-TEasy vector (Promega) and sequenced through vendors (Bioserves).

Construction of reporter plasmid constructs and Transient Transfection. Basal promoter regions
of BICP0 and UL21 were PCR amplified and cloned into promoter less pDSRED Express 2-1 (Clonetech) vector. All plasmids used for transfection were prepared with Sureprep Plasmid Endofree Maxi Kit (Genetix) according to manufacturer's instructions. The recombinant plasmids were transfected into 70-80% confluent MDBK cells in a 24 well plate using PolyFect transfection reagent (QIAGEN, Germany) as per the manufacturer's instructions. A total of 600 ng of recombinant plasmids were mixed with serum free Dulbecco's Minimal Essential Medium (DMEM, optimum) and 3 µL of Polyfect reagent. All samples were kept in triplicates and incubated for 24 hours. Flow Cytometry. Transfected cells were harvested and washed with prewarmed (37 °C) Phosphate Buffer Saline (PBS). Cells were dislodged with 50 µL of prewarmed Trypsin Versene Glucose solution (TVG) and pelleted by adding 300 µL PBS and centrifuged at 3,000 rpm for 5 min. The pellet was resuspended in 300 µL PBS and analyzed in FACScalibur (BD Dickinson, USA) in FL2 channel.

Computational identification of promoter elements.
To analyze the putative promoter regions, 1000 nucleotide upstream and 100 nucleotide downstream of TSS was selected for searching the putative promoter core elements. Promoter regions were identified using the online web server PROMOTER SCAN (http://www-bimas. cit.nih.gov/molbio/proscan/) and the downstream core elements were identified using YAPP online webserver (http://www.gene-regulation.com/cgi-bin/pub/programs/patch/bin/patch.cgi). The promoter sequences were also checked with other softwares like TRANSFAC, BDGP-Neural Network Promoter Prediction.