4. DISCUSSION

We have shown that CRISPR-mediated deletion of distal ovalbumin promoter in DF1 cells (DF1 +/OVA Pro ∆) induces the expression of Ovalbumin mRNA (Figures 1, 2). In addition, in cells with this kind of promoter, we inserted a promoterless reporter in theOvalbumin gene (DF1 +/OVA Pro ∆-Tg (promoterless dsRed)) and registered the expression of the reporter protein (DsRed2 fluorescence) (Figure 3). In this study, we showed that a chicken non-oviduct cell line with deletion of distal promoter sequences can serve as a chicken production model for steroid-independent expression of a transgene driven by endogenous ovalbumin promoter (Figure 4).
Transgenesis has become an important technique for generating biopharmaceutical products. The application of effective promoters is essential for achieving high expression levels and well-structured recombinant proteins. Although constitutive strong promoters have been extensively used to drive the expression of transgenes, they increase the metabolic burden of host cells, resulting in cell debilitation and cell population reduction in culture. Utilization of constitutive strong promoters might also lead to toxicity for the host cell due to the activation of unfolded protein response and aggregation of misfolded proteins in the host cells (Z. Liu, Tyo, Martinez, Petranovic, & Nielsen, 2012). In this regard, researchers have tried to discover the proper promoters for the continuous production of recombinant proteins at a convenient rate, exclusive to a preferred cell kind, and with appropriate posttranslational modifications and proper protein folding.
Tissue-specific ovalbumin promoter has been one of the novel candidates for the large-scale production of pharmaceutical proteins. The synthesis of several therapeutic proteins under the control of regulatory sequences from the chicken Ovalbumin gene has been reported (Byun et al., 2011; Cao et al., 2015; Herron et al., 2018; Kodama et al., 2012; M. S. Kwon et al., 2018; S. C. Kwon et al., 2010; S G Lillico et al., 2007; T. Liu et al., 2015; Oishi et al., 2018; T. S. Park et al., 2015; Zhu et al., 2005). Although the regulatory elements in theOvalbumin gene are well characterized out of their genomic context (Dougherty et al., 2009; Dougherty & Sanders, 2005; Haecker et al., 1995; Kato et al., 1992; Kaye et al., 1984, 1986; Monroe & Sanders, 2000; H. M. Park et al., 2000; Sanders & McKnight, 1988; Schimke et al., 1975; Schweers et al., 1990; Sensenbaugh & Sanders, 1999; Wang et al., 1989), it is not clear what regulatory sequences of the ovalbumin promoter are sufficient and efficient enough for inducing oviduct-specific expression of exogenous genes in the bioreactor chickens. In plasmid constructs, various lengths of chicken ovalbumin promoter fragments and, 5’ and 3’ flanking regions have been fused to the exogenous genes in order to induce gene expression. Some reports suggest that the inclusion of two major regulatory elements residing in the chicken ovalbumin promoter, a steroid-dependent regulatory element (SDRE, −900 to −732) and a negative regulatory element (NRE, −308 to −88) is sufficient to induce oviduct-specific expression of a therapeutic protein (S. C. Kwon et al., 2010; S G Lillico et al., 2007). These two regulatory elements are critical for appropriate regulation ofOvalbumin gene expression (Gaub, Dierich, Astinotti, Touitou, & Chambon, 1987; Nordstrom, Dean, & Sanders, 1993; Sanders & McKnight, 1988; Schweers et al., 1990; Schweers & Sanders, 1991). The SDRE is required for responsiveness to steroid hormones (i.e., estrogen, progesterone, androgen, and glucocorticoids) (Schimke et al., 1975) and the NRE, acts as a bifunctional element, cooperating with SDRE to activate Ovalbumin gene expression in the presence of steroids in the oviduct tissue, and repressing the Ovalbumin gene transcription in the absence of steroids in the oviduct and non-oviduct cells (Gaub et al., 1987; Haecker et al., 1995; Sanders & McKnight, 1988; Sensenbaugh & Sanders, 1999).
In an attempt to improve the expression level of the transgene ex situ (out of the native genomic context), additional regulatory sequences comprising the ovalbumin exon 1, intron 1, and the beginning of exon 2 were included in the promoter construct (S G Lillico et al., 2007). Zhu et al. utilized either 7.5 kb and 15 kb of the 5’ flanking region, and 15.5 kb of the 3’ flanking region from the Ovalbumingene to direct transgene expression ex situ . Although these regions contained all oviduct-specific regulatory elements, the ectopic expression of the transgene was detected in non-oviduct tissue of the chimeric chicken, and also germline transmission did not occur under the conditions of this study (Zhu et al., 2005). In the other studies, it was assumed that inclusion of the estrogen-responsive enhancer element (ERE), normally located approximately 3.3 kb upstream from the transcription start site (Figure 1A) (Kato et al., 1992) in the ovalbumin promoter-driven construct would increase the expression level of transgene (M. S. Kwon et al., 2018; S G Lillico et al., 2007). On the contrary, the results of the study failed to prove any increase in the level of recombinant protein produced in the transgenic chickens (S G Lillico et al., 2007). Herron et al. reintroduced an additional regulatory sequence between ERE and SDRE in their construct to enhance the expression level of protein in the egg white (Herron et al., 2018). The ovalbumin promoter (ranging from 1.35 kb to 3.0 kb) which have been used in most of ex situ (in a non-native site of the genome, or in a plasmid construct) studies so far, contains five main conserved sites which have been identified in chicken and other avian species (Woodfint, Hamlin, & Lee, 2018). However, the progressive identification of other farther regulatory elements associated with oviduct specificity (Kodama et al., 2012) and the complexity of gene expression regulation, have inevitably led to the use of ovalbumin promoter in situ (in its original genomic position).
Oishi et al. were the first and remain the only group to report the successful pruduction of pharmaceutical proteins driven by endogenous ovalbumin promoter in the egg white of transgenic chickens (Oishi et al., 2018). Low number of reports is due to the challenges in the generation of transgenic chickens. Although transgenic chicken bioreactors are valuable tools for the production of human recompinant proteins containing appropriate posttranslational modifications, generating of founder transgenic chicken is relatively difficult, inefficient and time consuming. Thus, the use of alternative cell production systems, for example chicken non-oviduct cell lines, would seem desirable to overcome these obstacles.
Previous studies on the precise characterization of the regulatory properties of the Ovalbumingene demonstrated that deletion of the SDRE and the NRE, as well as the linker between them, increases chloramphenicol acetyltransferase (CAT) activity on a plasmid (Haecker et al., 1995; Sanders & McKnight, 1988; Sensenbaugh & Sanders, 1999). These studies indicated that a cooperation between multiple distal regulatory and promoter-proximal regions confers oviduct-specific Ovalbuminexpression. Deletion of regulatory elements upstream of −80 abolished the tissue-specific expression of Ovalbumin in primary oviduct cell cultures, while basal expression increased to levels seen with estrogen-induced genes containing a SDRE (Haecker et al., 1995; H. M. Park et al., 2000; Sanders & McKnight, 1988). A few reports showed that the expression of the reporter CAT gene was induced by the ovalbumin proximal promoter (−87 to +9) in primary oviduct cell and non-oviduct cell cultures such as LMH/2A (Table 3) (Dean, Jones, & Sanders, 1996; Haecker et al., 1995; Monroe & Sanders, 2000; Muramatsu et al., 1998; H. M. Park et al., 2000; Schweers et al., 1990; Sensenbaugh & Sanders, 1999).
Although previous transfection experiments with truncated ovalbumin promoter-CAT reporter (OvCAT) constructs have tried to mimic the activity of the endogenous ovalbumin promoter in the oviduct and non-oviduct cells, there is not any report on the in situ deletion of the regulatory sequences of ovalbumin promoter and thier effects on the levels ofOvalbumin gene expression. In this study, we show that thein situ deletion of distal ovalbumin promoter results in the upregulation of Ovalbumin transcript in chicken DF1 cell line. Our RT-qPCR analysis upon deletion of the distal ovalbumin promoter including two major regulatory elements, the SDRE and the NRE (DF1+/OVA Pro ∆ cells), indicated an increased level of expression of ovalbumin, ~104 fold higher than the Ovalbumin transcript levels in WT DF1 (Figures 2). Deletion of a 962-bp region (−1044 to −82 bp) containing the distal promoter elements completely abolished tissue-restricted and hormone-dependent expression of the Ovalbumin gene. It has been reported that chicken ovalbumin upstream promoter (COUP) site (−85 to −73) represses basal Ovalbumin expression in the absent of steroids and is required for induction by steroids (Figure 1A) (H. M. Park et al., 2000). Although previous reports have shown that the deletion of the COUP site in OvCAT constructs increases transcriptional activity in the absence of the NRE and confirm its repression role on the basal gene expression, our data clearly show that, without the NRE, transcriptional activity is increased even when COUP site is present. This finding suggests that the opposing effect of COUP site on the transcriptional activity depends on the native genomic context and perhaps to other regulatory elements in wild-type composition.
In our DF1+/OVA Pro ∆ cells, althogh the core promoter elements (TATA box and the initiator element (INR)containing sufficient information for the initiation of transcription) have been remained intact, we cannot rule out the regulatory role of alternative promoters in the genome (Ayoubi & Van De Ven, 1996). Kodama et al. have found several TATA-like and other promoter motifs located at a position around −1800 bp (Kodama et al., 2012). Muramatsu et al. demonstrated that the sequence from −3200 to −2800 act as a tissue-specific silencer-like which represses the Ovalbumin gene expression in non-oviduct tissue (Figure 1A) (Muramatsu et al., 1998). However, the presence of this sequence did not inhibit the transcriptional activity under our experimental conditions. Our results support this notion that the transcriptional regulation is not determined only by promoter regions, but involves multiple native features in the local genomic context including enhancers, insulators, DNA binding regulatory proteins such as transcription factors and repressors, nucleosome positioning, histone modifications, non-coding RNA, the three-dimensional organization of genes, and epigenetic mechanisms (Andersson & Sandelin, 2019; Gibcus & Dekker, 2012).
In conclusion, our study demonstrates the potential for producing recombinant proteins in chicken cell lines as an appropriate alternative to mammalian cell culture systems. This accomplishment of hormonally-independent expression of the transgene driven by the endogenous regulatory mechanism(s) overcomes the limitation of cloned promoters, where the promoter regulatory sequences have to be taken out of their cis context and spatial organization into a plasmid. Use of CRISPR technology enables precise deletion or mutagenesis of regulatory sequences in the native genomic context, showing great promise to better understand, regulate, and exploit the native biological elements of gene regulation.