Use of Crispr/Cas9 to Investigate CAR Isoform Function in AdV Infection

Paper: Readler, J. M., AlKahlout, A. S., Sharma, P., & Excoffon, K. J. (2019). Isoform specific editing of the coxsackievirus and adenovirus receptor. Virology, 536, 20-26.

Coxsackievirus infects a range of hosts, including humans, and is linked to a number of diseases such as myocarditis and pancreatic inflammation [2]. Of a different viral family, adenovirus also infects humans, inducing mild diseases such as gastroenteritis and upper respiratory tract infection, in addition to more severe conditions when paired with host immunosuppression. Coxsackievirus is made up of single-stranded RNA, while adenovirus is comprised of double-stranded DNA. Despite having different genomic structures, both coxsackievirus and adenovirus utilize a cell-adhesion protein encoded by the CXADR gene as their primary receptor [2]. This transmembrane receptor exists in two isoforms that are dependent on alternative splicing, a gene expression regulatory process that differentially includes or excludes particular exons of a transcript. The two isoforms, Car^Ex7 and Car^Ex8, have distinct C-terminal ends that lead to different localization in polarized epithelial cells. These polarized cells, as depicted in the image below, are characterized by apical and basal faces that are distinct from one another in structure and function at sites of cell-cell contact [4]. The Car^Ex7  isoform encodes the first seven exons of CXADR and is localized at the basolateral cell surface, while Car^Ex8 includes an additional eighth exon and is localized at the apical cell surface.

Epithelial Cell Polarity: Apical and basolateral surfaces are associated with distinct structures and locations within the cell.
Image source: Figure 4-1. "The Polarity of Epithelial Cells." Adapted from Pearson Education Inc., 2015, p. 116.

            In this study, Readler et al. aimed to better illuminate the functional differences between these two isoforms of the coxsackievirus and adenovirus receptor (CAR). Specifically, researchers chose to focus on the importance of the Car^Ex8 isoform to adenovirus entry at apical cell surfaces. Two key components of this investigation were the Madin Darby Canine Kidney (MDCK) cell line and Crispr/Cas9 gene editing technology. Readler et al. previously created the MDCK cells to stably express human Car^Ex8 cDNA with a promoter that is inducible with the antibiotic doxycycline. With this model cell line established to manipulate the induction of Car^Ex8 expression, researchers then utilized Crispr/Cas9 to knockdown endogenous expression of Car^Ex8 in the MDCK cells. This novel knockdown cell-line thereby isolated the functionality of the Car^Ex8 receptor to a level never achieved before.

           

            To ensure that they successfully targeted the eighth exon of the MDCK CXADR gene, researchers first performed a screen using fluorescence-activated cell sorting and PCR. From this screen, one clone was identified to be a successful knockout, as PCR primers set to recognize the inside of the eighth exon sequence produced no bands. Furthermore, sequencing this clone, renamed JR1-Car^Ex8-KO, verified that the intended deletions were indeed present when compared to MDCK-Car^Ex8 parental cells. After successfully completing this exonic knockout, researchers then investigated total CAR and Car^Ex8  expression levels within the JR1-Car^Ex8-KO cells via western blot analysis. As shown in Figure 3 below, these cells exhibited significantly less expression of Car^Ex8, while showcasing partially reduced levels of total CAR. In both cases, treatment with the antibiotic (Dox), which induced expression of human Car^Ex8 in this cell line, restored expression levels to that of the parental cells. This demonstrated the role of the Car^Ex8 knockdown on the decrease of CAR expression levels, as well as that the insert of human Car^Ex8 was not impacted by the endogenous MDCK Car^Ex8 deletion.

 

Figure 3. CAR expression in JR1-Car^Ex8-KO cells as compared to parental cells. Western blots were performed using polyclonal antibodies detecting A) total CAR expression and B) Car^Ex8 expression alone. Total CAR expression partially decreased while Car^Ex8 reduced significantly in knockout cells (KO) as compared to parental cells (Par). With treatment of 200 ng/mL doxycycline (Dox), KO showed expression restored to the level of Par.

Another important finding of this research was that the less pronounced impact of the knockdown on total CAR expression is likely due to the endogenous levels of Car^Ex7 still maintained in the cell line. As the specificity of CRISPR editing targeted the intronic regions on either side of the eighth exon alone, the seventh exon of CXADR remained intact to generate the majority of CAR expression levels still seen. Another component of this figure worth highlighting is the slight expression of Car^Ex8, despite a seemingly complete knockdown, that was detected via western blot. As compared to the PCR screen (shown below) that showed no amplification of exon 8, the presence of this protein expression in Figure 3B above raises questions as to the true completeness of this knockout. In addressing this discrepancy, Readler et al. offer the possibility that one allele remains, although with primer recognition sites altered, as an explanation for the lack of this Car^Ex8 presence in their PCR analysis. Additionally, they consider non-specific binding within the western blot as a possible reason the fully complete knockdown of Car^Ex8 was not demonstrated within the protein expression levels seen.

 

Figure 1C. PCR knockout screening using primers inside the exon 8 cut site. JR1-Car^Ex8-KO cells (KO) exhibited no band as compared to untreated parental cells (Par) and two haploid knockouts (Hap 1 and 2).

            Despite this inconsistency in their results, Readler et al. determined their knockdown of Car^Ex8 significant enough to investigate the receptor isoform’s importance to adenovirus infection. Using a virus reporter gene, β-galactosidase, researchers quantified the levels of viral-induced luminescence per milligram of protein for MDCK-Car^Ex8 parental and knockout cells. In knockout cells, viral transduction was subjected to a two-fold decrease, suggesting the importance of Car^Ex8 to adenovirus entry. This is supported by Car^Ex8’s known tendency to localize the CAR receptor at apical cell surfaces that are exposed to cellular lumen   where adenovirus is able infect host cells [3]. Further, to look at this apical entry alone, researchers targeted infection to the apical surface of fully polarized MDCK cells and measured genomic levels of adenovirus via qPCR. As shown in Figure 4C below, at 24 hours post-infection a three-fold reduction in adenovirus genome was seen in Car^Ex8-knockouts as compared to parental cells. This difference was restored with the reestablishment of Car^Ex8 induced through Dox antibiotic treatment. This greater fold-change in adenovirus infection when targeted to the apical cell surface supports the notion that Car^Ex8 is essential for the mediation of viral entry at this polarized cell surface, a function that was less pronounced when Car^Ex7 was still allowing adenovirus entry at the basolateral cell surface in these knockouts.

Figure 4C. Cell associated adenovirus genomes in MDCK-Car^Ex8 parental and JR1-Car^Ex8-KO cells. Detected via qPCR, KO cells exhibited a significant reduction in cell-associated viral genomes, as compared to parental and KO cells treated with 200 ng/mL doxycycline.

            Ultimately, this novel Car^Ex8 knockout cell line revealed the importance of this receptor isoform to adenovirus entry at the apical surface of polarized cells. With attenuated expression of Car^Ex8, MDCK cells showed reduced levels of adenovirus transduction and genome levels, which were restored with antibiotic-induced Car^Ex8 expression from a gene insert. The methods utilized here offer a model system for future studies to elucidate further functionalities of the Car^Ex8 receptor using this knockout cell line. Additionally, this methodology could be applied to other genes, such as CD46, which has 14 different receptor isoforms [1]. The use of Crispr/Cas9 editing to create exonic knockouts of this receptor could showcase the interaction between specific CD46 isoforms and the entry of measles or herpesvirus strains [1]. In conclusion, the most notable result of this research is the clinical relevance this method holds in narrowing down receptor isoform correlations to viral entry, as regulation of particular exonic expression could work to inhibit certain strains of viral infection.

References

[1] Excoffon, K. J., Bowers, J. R., & Sharma, P. (2014). 1. Alternative splicing of viral receptors: A review of the diverse morphologies and physiologies of adenoviral receptors. Recent research developments in virology, 9, 1.

[2] He, Y., Chipman, P. R., Howitt, J., Bator, C. M., Whitt, M. A., Baker, T. S., ... & Rossmann, M. G. (2001). Interaction of coxsackievirus B3 with the full length coxsackievirus-adenovirus receptor. Nature Structural & Molecular Biology, 8(10), 874.

[3] Kotha, P. L., Sharma, P., Kolawole, A. O., Yan, R., Alghamri, M. S., Brockman, T. L., ... & Excoffon, K.J. (2015). Adenovirus entry from the apical surface of polarized epithelia is facilitated by the host innate immune response. PLoS pathogens, 11(3), e1004696.

[4] Rappel, W. J., & Edelstein-Keshet, L. (2017). Mechanisms of cell polarization. Current opinion in systems biology, 3, 43-53.

[5] Readler, J. M., AlKahlout, A. S., Sharma, P., & Excoffon, K. J. (2019). Isoform specific editing of the coxsackievirus and adenovirus receptor. Virology, 536, 20-26.