icon-folder.gif   Conference Reports for NATAP  
 
  65th Annual Meeting of the
American Association for the
Study of Liver Diseases
Boston, MA Nov 7-11 2014
Back grey_arrow_rt.gif
 
 
 
Curing HBV....attacking cccDNA "Suppression of hepatitis B virus DNA accumulation in chronically infected cells using a bacterial CRISPR/Cas RNA-guided DNA endonuclease"
 
 
  Dowload the PDF Here
 
"While this article was in preparation, Lin et al. (2014) reported that HBV-specific Spy Cas9/sgRNA combinations were able to cleave HBV DNA when Cas9 and HBV expression plasmids were co-transfected into Huh7 cells. Neither HBV-infected human cells nor HBV cccDNA levels were examined in this previous work. Here, we substantially extend these recent findings by demonstrating HBV DNA cleavage and suppression, including repression of cccDNA accumulation, in both a model of chronic HBV infection (HepAD38 cells) and during de novo infection (HepaRG cells) in culture. These novel observations therefore provide substantial support for the idea that bacterial CRISPR/Cas systems have the potential to function as an effective treatment for DNA virus infections."
 
"..... in order to effectively clear an HBV infection and produce full remission, the viral cccDNA intermediates need to be destroyed.......Impressively, the HBV RT specific sgRNA was therefore able to reduce HBV cccDNA formation by ~90% and total HBV intracellular DNA accumulation by ~99%.......all three sgRNAs reduced HBV DNA levels inside cells by ~10 to ~125-fold and in the supernatant media by ~100-fold to ~800-fold, as measured by real-time PCR (Fig. 2A). Even more impressively, the RT sgRNA suppressed HBV DNA replication to the same extent as Tet addition, which essentially entirely blocks virus replication.....
 
.......Indeed, the surface Ag-specific sgRNA suppressed HBsAg production to nearly undetectable levels (Fig. 2C), possibly consistent with the mutational inactivation of this viral gene due to editing by the cognate Cas9/sgRNA combination. We also observed a statistically significant reduction in HBeAg secretion in the cells transduced with the RT specific sgRNA, indicating perhaps a global reduction in cccDNA in this case (Fig. 2D). In contrast, we did not observe a significant reduction in HBeAg for the Core and Surface sgRNAs, and this likely results from this protein's secretion and exceptional stability. In contrast to the HBsAg ORF, it should be noted that none of the sgRNAs employed directly target this viral ORF for mutagenic inactivation......
 
........expression of a Cas9/sgRNA combination specific for the HBV RT gene not only dramatically reduced viral DNA levels, as expected (Figs. 2A and 2B), but also mutationally inactivated the majority of the low level of residual viral DNA......
 
..............Both HBsAg and HBeAg) secretion was reduced by up to ~80% by 17 days after infection in the presence of HBV-specific sgRNAs (Fig. 4A and 4B) and this reduction was highly significant.......
 
..........all three sgRNAs targeting the HBV genome also induced a reduction in HBV total DNA levels in the supernatant media and inside infected cells, arguing that not only are HBe- and HBs-antigen gene expression specifically inhibited, but that HBV replication is also globally reduced by the HBV specific Cas9/sgRNA combinations tested. To further investigate if this reduction is due to degradation of the viral genome in the infected cells, intracellular DNA was subjected to qPCR using primers that specifically amplify cccDNA. All three sgRNAs led to reductions in cccDNA levels in infected HepaRG cells (up to 67%) (Fig. 4E), providing evidence that the Cas9/sgRNA combinations directly act on viral cccDNA intermediates. As expected, ETV treatment had no effect on cccDNA levels, and the observed almost complete inhibition of overall HBV DNA production by ETV in the supernatant media and intracellularly (Figs. 4D and 4F) further confirms the high specificity of the primers used for cccDNA detection. Overall, these data therefore suggest that HBV DNA cleavage by Cas9/ sgRNA combinations has the potential to not only disrupt HBV replication but also viral cccDNA stability.......
 
............we also tested whether expression of Cas9/sgRNA combinations would be able to inhibit HBV replication more effectively when used together with known inhibitors of the HBV life cycle, including 3TC, TDF, ETV and HAP-12. As shown in Fig. 5 and Table 1, we were indeed able to observe enhanced inhibition of HBV replication by HBV-specific Cas9/sgRNA combinations in both HepAD38 cells and in the chronically HBV-infected cell line HBV2.2.15 (Sells et al., 1987) in the presence of various antiviral drugs. We therefore believe that the use of CRISPR/Cas system to directly target HBV cccDNA in the infected liver clearly merits further testing to assess whether there is real potential for a possible cure for chronic HBV infections when effective Cas9/sgRNA combinations are used together with current antiviral drugs......Whether Cas9/sgRNA combinations are entirely safe, and whether their efficiency can be high enough to effectively clear the HBV cccDNA pool, are issues that will next need to be carefully addressed in animal models of HBV infection"
 
----------------------
 
Suppression of hepatitis B virus DNA accumulation in chronically infected cells using a bacterial CRISPR/Cas RNA-guided DNA endonuclease
 
Virology Dec 2014
 
Edward M. Kennedya,*, Leda C. Bassitb,*, Henrik Muellerc, Anand V. R. Kornepatia, Hal P. Bogerda, Ting Nieb, Payel Chatterjeeb, Hassan Javanbakhtc, Raymond F. Schinazib,*, and Bryan R. Cullena,# aDepartment of Molecular Genetics & Microbiology and Center for Virology, Duke University Medical Center, Durham, North Carolina bCenter for AIDS Research, Laboratory of Biochemical Pharmacology, Department of Pediatrics, Emory University School of Medicine, Atlanta, Georgia cInfectious Diseases, F. Hoffmann-LaRoche, Inc., Basel, Switzerland
 
Highlights
 
· We inactivated the HBV genome by targeted DNA editing using CRISPR/Cas.
· While no cytotoxicity was observed, HBV cccDNA was effectively downregulated.
· This paper provides proof of principle for a novel approach to HBV treatment.
 
Abstract
 
Hepatitis B virus (HBV) remains a major human pathogen, with over 240 million individuals suffering from chronic HBV infections. These can persist for decades due to the lack of therapies that can effectively target the stable viral covalently closed circular (ccc) DNA molecules present in infected hepatocytes. Using lentiviral transduction of a bacterial Cas9 gene and single guide RNAs (sgRNAs) specific for HBV, we observed effective inhibition of HBV DNA production in in vitro models of both chronic and de novo HBV infection.
 
Cas9/sgRNA combinations specific for HBV reduced total viral DNA levels by up to ~1000-fold and HBV cccDNA levels by up to ~10-fold and also mutationally inactivated the majority of the residual viral DNA. Together, these data provide proof of principle for the hypothesis that CRISPR/Cas systems have the potential to serve as effective tools for the depletion of the cccDNA pool in chronically HBV infected individuals.
 
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
 
Introduction
 
Hepatitis B virus (HBV) remains an important human pathogen despite the existence of a highly effective vaccine. Particularly in resource-limited contexts, vaccination rates remain well under 100% and vaccination at birth of infants born to HBV positive mothers is not fully protective against vertical transmission (Komatsu, 2014). Moreover, it is estimated that over 240 million individuals, ~4% of the world population, remain chronically infected with HBV and are therefore at high risk for cirrhosis and hepatocellular carcinoma (HCC), leading to ~800,000 deaths per year that are directly linked to HBV infection (Komatsu, 2014).
 
Treatment of HBV primarily relies on the use of nucleoside analog chain terminators, such as lamivudine (3TC), tenofovir disoproxil fumarate (TDF) and entecavir (ETV), which act as potent inhibitors of the HBV reverse transcriptase (RT). However, while these drugs can clearly slow the progression of HBV-induced disease, including both cirrhosis and HCC, they do not lead to a full elimination of this viral infection. This is due to the very high stability of an episomal DNA intermediate in the HBV replication cycle, called covalently closed circular DNA (cccDNA), that persists in the nuclei of infected hepatocytes and serves as the template for viral mRNA and pre-genomic RNA synthesis (Werle-Lapostolle et al., 2004). While RT inhibitors can prevent the de novo HBV infection of additional hepatocytes, the cells infected at the time of treatment initiation remain infected and only gradually decrease in number due to cell turnover. Therefore, in order to effectively clear an HBV infection and produce full remission, the viral cccDNA intermediates need to be destroyed. This goal has now become feasible due to the development of programmable RNA-guided DNA endonucleases derived from the Type II bacterial CRISPR (clustered regularly interspaced short palindromic repeats)/Cas mechanism of bacterial adaptive innate antiviral immunity (Hsu et al., 2014).
 
Type II CRISPR/Cas systems rely on a single effector protein, called CRISPR-associated protein 9 (Cas9), that is normally guided to a specific DNA sequence by two small RNA molecules called the tracrRNA and the crRNA (Barrangou and Marraffini, 2014; Hsu et al., 2014). The demonstration that the tracrRNA and crRNA could be combined into a single guide RNA (sgRNA) (Cong et al., 2013; Mali et al., 2013) enhanced the feasibility of using CRISPR/Cas systems as effective gene editing tools in eukaryotic cells. Previously, most research has focused on using the Cas9 protein from Streptococcus pyogenes (Spy), which uses a 20-nt guide sequence to target complimentary DNA sequences that are flanked 3' by a so-called protospacer adjacent motif (PAM), with the sequence 5'-NGG-3'. The PAM, which forms an invariant part of every Spy Cas9-specific DNA target, is directly recognized by the Cas9 protein and initiates DNA targeting guided by the sgRNA (Barrangou and Marraffini, 2014; Cong et al., 2013; Hsu et al., 2014; Mali et al., 2013). Using transfection or lentiviral vector transduction, it is possible to induce cleavage of specific DNA sequences present in a mammalian genome by expression of Spy Cas9 and an sgRNA molecule transcribed by an RNA polymerase III promoter (Cong et al., 2013; Mali et al., 2013). Gene inactivation using this system is both efficient and specific, and Spy Cas9 has in fact been used for not only high throughput genetic screens in human cells in culture (Shalem et al., 2014; Wang et al., 2014; Zhou et al., 2014) but also for the generation of gene knockouts in several species, including mice and even monkeys (Yang et al., 2013). We therefore reasoned that the Spy Cas9 protein, in combination with sgRNAs designed to select a target present on the HBV cccDNA, might represent a powerful method to cleave and eradicate HBV cccDNAs. In the present study, we demonstrate that lentiviral vectors encoding Spy Cas9/sgRNA combinations targeted to the HBV reverse transcriptase (RT), surface antigen (Ag) or core protein genes are all able to effectively repress viral DNA production, including cccDNA accumulation, in chronically HBV infected cells and also inhibit de novo HBV infection. Moreover, these Spy Cas9/sgRNA combinations showed additive inhibition of HBV DNA accumulation when used in combination with known pharmacological inhibitors of the HBV RT enzyme. These experiments provide evidence in support of the hypothesis that HBV-specific Spy Cas9/sgRNA combinations, when delivered to HBV-infected hepatocytes in vivo by viral transduction, might provide a novel and effective approach to facilitate the elimination of HBV in chronically infected individuals.
 
Results
 
To target the HBV genome (subtype AYW) for elimination in cell culture, we first designed three sgRNAs specific for Spy Cas9 targeting the viral open reading frames (ORFs) encoding HBV surface Ag, core, and/or RT. For the HBV structural proteins surface Ag and core, N-terminal targets were selected to induce frame-shift mutagenesis. The surface Ag specific sgRNA was also predicted to cleave the RT gene (Fig. 1A). RT was also targeted in the highly conserved "YMDD" motif required for enzyme catalysis (Fig. 1A) using an sgRNA predicted to also cleave near the end of the surface Ag ORF. To verify effective sgRNA function, we employed a previously described (Kennedy et al., 2014) indicator assay based on co-transfection of Spy Cas9 and sgRNA expression plasmids, along with a cognate indicator plasmid, into human 293T cells. This assay quantitatively measures sgRNA efficacy by monitoring the firefly luciferase (FLuc) activity or steady state expression level of an HIV-1 Rev-sgRNA target-FLuc fusion protein; high cleavage activity results in a substantial knockdown of both. As shown in Figure 1B and 1C, all three sgRNAs induced an effective and specific knockdown of Fluc expression, as measured by dual luciferase assay, and Rev fusion protein expression, as measured by Western blot analysis, respectively. To directly test these three sgRNA constructs in relevant HBV cell-culture models, we next shuttled them into a lentiviral Cas9/sgRNA expression vector (Shalem et al., 2014) capable of efficient transduction of the cell lines HepAD38 and HepaRG (Fig. S1).
 
Suppression of HBV replication by Cas9/sgRNA combinations specific for the HBV RT, surface Ag and core genes
 
In the HepAD38 cell line, transcription initiation of an integrated HBV linear DNA genome is tightly regulated by a Tet repressed promoter (Ladner et al., 1997). In the absence of Tet in the culture medium, the system mimics HBV replication by transcribing HBV mRNAs and pre-genomic RNA, leading to the initiation of a robust HBV replication cycle, including the synthesis of substantial levels of HBV cccDNA, and the release of DNA-containing viral particles into the supernatant medium. Consequently, this cell line represents an excellent model to test suppression of HBV replication and cccDNA synthesis by Spy Cas9 loaded with sgRNAs specific for HBV DNA targets. Therefore, we tested the ability of this system to suppress HBV replication by transducing HepAD38 cells with lentiviral vectors encoding HBV-specific Cas9/sgRNA combinations. HBV transcription was activated 48 h prior to transduction by removal of Tet from the medium to ensure that an active HBV replication cycle was in progress prior to Cas9/sgRNA expression. Subsequently, Tet was added back to the medium again, repressing the chromosomal HBV locus, to ensure that cccDNA was the primary source of viral transcription. The lentiviral vectors used also encode a puromycin resistance gene (Shalem et al., 2014), thus permitting selection of the transduced cells. As a positive control for HBV gene expression, HepAD38 cells were also transduced with a nonspecific (N.S.) sgRNA construct that does not target any HBV sequence. As a control for maximally repressed HBV expression, HepAD38 cells were also tested in the continuous presence of Tet, which in the HepAD38 cells blocks HBV transcription (Ladner et al., 1997). After transduction and selection, all three sgRNAs reduced HBV DNA levels inside cells by ~10 to ~125-fold and in the supernatant media by ~100-fold to ~800-fold, as measured by real-time PCR (Fig. 2A). Even more impressively, the RT sgRNA suppressed HBV DNA replication to the same extent as Tet addition, which essentially entirely blocks virus replication.
 
Suppression of cccDNA formation by HBV specific Cas9/sgRNA combinations
 
To test the capacity of HBV-specific Cas9/sgRNA combinations to eliminate cccDNA accumulation, transduced HepAD38 cells were also analyzed for the level of cccDNA accumulation by real-time PCR after 14 days in culture using previously described cccDNAspecific primers (Chen et al., 2004). Remarkably, the HBV RT-specific sgRNA suppressed cccDNA formation by ~10-fold, while the surface Ag and core-specific sgRNAs repressed cccDNA levels by ~4-fold (Fig. 2B). Total intracellular HBV DNA accumulation was reduced by ~8-fold by the HBV surface Ag and core-specific sgRNAs, and were repressed by ~80-fold by the RT specific sgRNA (Fig. 2B). Impressively, the HBV RT specific sgRNA was therefore able to reduce HBV cccDNA formation by ~90% and total HBV intracellular DNA accumulation by ~99%.
 
HBV-specific Cas9/sgRNA combinations reduce HBV surface Ag secretion
 
In HepAD38 cells, analysis of the level of HBV surface Ag (HBsAg) and "e" Ag (HBeAg) secretion represents a useful assay for screening for antivirals that inhibit HBV replication and/or viral gene expression. To determine if HBV-specific Cas9/sgRNA combinations can suppress secretion of these viral antigens, we performed an ELISA assay using the supernatant media of transduced HepAD38 cells harvested after 12 days in culture. Low levels of secreted HBsAg were observed in the supernatant of HepAD38 cells transducer with HBV-specific sgRNAs targeting the RT, surface and core genes but not the N.S. sgRNA. Indeed, the surface Ag-specific sgRNA suppressed HBsAg production to nearly undetectable levels (Fig. 2C), possibly consistent with the mutational inactivation of this viral gene due to editing by the cognate Cas9/sgRNA combination. We also observed a statistically significant reduction in HBeAg secretion in the cells transduced with the RT specific sgRNA, indicating perhaps a global reduction in cccDNA in this case (Fig. 2D). In contrast, we did not observe a significant reduction in HBeAg for the Core and Surface sgRNAs, and this likely results from this protein's secretion and exceptional stability. In contrast to the HBsAg ORF, it should be noted that none of the sgRNAs employed directly target this viral ORF for mutagenic inactivation.
 
Mutational inactivation of HBV by Cas9/sgRNA combinations
 
While the observed depletion of total HBV DNA and cccDNA accumulation was extensive (Figs. 2A and 2B), we were also curious as to the mutational status of the residual viral DNA as Cas9 cleavage of chromosomal targets usually results in the introduction of small sequence insertions or deletions (indels) (Shalem et al., 2014). For this purpose, we focused on HepAD38 cells transduced with the sgRNA specific for the active site YMDD motif of the HBV RT gene (Fig. 1A). We harvested total DNA from the HepAD38 cells transducer with the RT-specific Cas9/sgRNA combination that we had previously shown effectively knocked down viral DNA levels (Figs. 2A and 2B) and subjected the DNA to PCR amplification using HBV RT specific primers followed by cloning and sequencing of the resultant HBV DNA fragments. Seventy-three deletion mutations located at or immediately adjacent to the predicted Cas9 cleavage site present 3 bp 5' to the target DNA PAM sequence were recovered, and we also detected five insertion mutations at this same location in the HBV genome (Fig. 3). Of the 103 recovered viral sequences, 78 (76%) were found to be mutated and, interestingly, the bulk of the amplicons recovered were predicted to be lethally mutagenized; even an in-frame insertion or deletion would be highly detrimental as these residues are required for HBV RT function and therefore highly conserved. In conclusion, expression of a Cas9/sgRNA combination specific for the HBV RT gene not only dramatically reduced viral DNA levels, as expected (Figs. 2A and 2B), but also mutationally inactivated the majority of the low level of residual viral DNA.
 
Targeting HBV with Cas9/sgRNA combinations reduced antigen secretion and viral replication and leads to cccDNA degradation in infected HepaRG cells
 
HepAD38 cells are an excellent model of post-entry chronic infection, but to test whether Cas9/sgRNA combinations specific for HBV are also capable of inhibiting de novo infection we employed a second cell culture model, HepaRG cells. These cells are permissive for infection with live HBV virus subsequent to differentiation (Gripon et al., 2002) and are an excellent complement to the HepAD38 system, which models the late stage of the HBV lifecycle (Ladner et al., 1997). To assess the effect of HBV-specific sgRNAs on HBV replication in HepaRG cells, we transduced these cells with the lentiviral constructs described above, expressing HBV RT, surface Ag, core, or control N.S. sgRNAs. After selection for puromycin resistance and differentiation, these cell lines were infected with HBV at a high MOI of ~30 and the antiviral effects of the sgRNAs evaluated (Fig. 4). Both HBsAg and HBeAg) secretion was reduced by up to ~80% by 17 days after infection in the presence of HBV-specific sgRNAs (Fig. 4A and 4B) and this reduction was highly significant. This effect could be due either to a direct effect of the sgRNA on the expression of the specific viral gene or, perhaps more likely, a global reduction in cccDNA production and HBV transcription. To further investigate the impact and specificity of the panel of HBV-specific Cas9/sgRNA combinations, we isolated DNA from the supernatant media and cell lysates of the HepaRG RT, surface Ag, core, and N.S. sgRNA-expressing cells following infection with HBV in the presence or absence of ETV treatment (see schematic in Fig. 4C). ETV, a nucleoside analogue, has been shown to reduce HBV DNA levels by inhibiting reverse transcription, but does not directly affect cccDNA stability in HepaRG cells (Lucifora et al., 2014). Moreover, in the context ETV treatment and viral inhibition, Cas9/sgRNAs might represent an effective means to eliminate cccDNA from the treated and suppressed cells. We therefore isolated total DNA from ETV-treated and untreated HepaRG cells and used these for qPCR analysis of total HBV DNA levels present either in viral particles released into the supernatant media (Fig. 4D) or present intracellularly (Fig. 4F). As expected, in the presence of ETV, HBV DNA was greatly reduced both in the supernatant (by ~96%) and intracellularly (by ~80%). In addition, all three sgRNAs targeting the HBV genome also induced a reduction in HBV total DNA levels in the supernatant media and inside infected cells, arguing that not only are HBe- and HBs-antigen gene expression specifically inhibited, but that HBV replication is also globally reduced by the HBV specific Cas9/sgRNA combinations tested. To further investigate if this reduction is due to degradation of the viral genome in the infected cells, intracellular DNA was subjected to qPCR using primers that specifically amplify cccDNA. All three sgRNAs led to reductions in cccDNA levels in infected HepaRG cells (up to 67%) (Fig. 4E), providing evidence that the Cas9/sgRNA combinations directly act on viral cccDNA intermediates. As expected, ETV treatment had no effect on cccDNA levels, and the observed almost complete inhibition of overall HBV DNA production by ETV in the supernatant media and intracellularly (Figs. 4D and 4F) further confirms the high specificity of the primers used for cccDNA detection. Overall, these data therefore suggest that HBV DNA cleavage by Cas9/sgRNA combinations has the potential to not only disrupt HBV replication but also viral cccDNA stability.
 
Enhanced inhibition of HBV DNA accumulation by antivirals in combination with Cas9/ sgRNAs
 
To further assess the possibility that other potent RT inhibitors (TDF, ETV or 3TC) or a viral assembly inhibitor (HAP12) could enhance elimination of residual virus from HBV-specific Cas/sgRNA expressing cells, we treated HepAD38 cells at concentrations that only partly inhibit HBV DNA replication in this system, as measured by real-time PCR of secreted HBV DNA. First, we assessed the level of viral inhibition in cells expressing Cas9 and the control N.S. sgRNA and all inhibitors tested exhibited the expected dose response (Fig. 5A). Remarkably, and as previously observed (Fig. 2A), HepAD38 cells expressing the HBV RT-specific sgRNA showed essentially complete suppression of secreted HBV DNA accumulation and gained no added benefit from the viral replication inhibitors employed (Fig. 5B). However, with the less effective surface Ag and core-specific sgRNAs, there was a modest but readily detectable enhancement of the level of induced inhibition of secreted HBV DNA in the presence of the antivirals, as shown in Fig. 5C and 5D. To further confirm these results, we used a constitutively HBV-expressing cell system, HBV2.2.15 cells (Liu et al., 2004; Sells et al., 1987). After transduction with Cas9/sgRNA expressing lentiviral constructs and selection, we observed that extracellular levels of HBV DNA were suppressed by ~98%, ~80%, and ~90%, respectively, by the RT, surface Ag, and core specific Cas9/sgRNA combinations (Table 1). Importantly, treatment of these cells with 1 μM of TDF, ETV, HAP12 or 3TC demonstrated a clear at least additive effect, leading to more efficient elimination of residual HBV DNA replication, for the less effective surface Ag- and core-specific sgRNA in 2.2.15 cells (Table 1). In the case of the RT-specific sgRNA, the detection of a possibly synergistic level of inhibition was again largely prevented by the extremely high efficacy of this Cas9/sgRNA combination.
 
Cytotoxicity analysis
 
For both HepAD38 and HepaRG cells transduced with lentiviral sgRNA/Cas9 expression vectors cytotoxicity assays were performed, and no cellular toxicity was observed using two distinct assays at late timepoints (see Materials and Methods and Table S1).
 
Discussion
 
Despite an effective vaccine, HBV remains a serious human pathogen, with over 240 million infected individuals worldwide leading to ~800,000 deaths annually, primarily from cirrhosis and HCC (Komatsu, 2014). Nucleoside inhibitors such as ETV and TDF are effective inhibitors of the HBV RT enzyme and can prevent the release of infectious visions from HBV infected cells, but they are unable to cure HBV due to the remarkable stability of the viral cccDNA intermediate, which is not a direct target for these drugs (Lucifora et al., 2014). Recently, RNA-guided DNA endonucleases belonging to the CRISPR/Cas family (Hsu et al., 2014) were shown to effectively cleave and inactivate viral DNA in both HIV-1 and HPV-infected human cells (Ebina et al., 2013; Hu et al., 2014; Kennedy et al., 2014), and we therefore wished to test whether Cas9/sgRNA combinations specific for the HBV DNA genome might be able to also cleave and reduce HBV DNA levels, including levels of cccDNA, in infected human cells in culture.
 
The system we have used relies on the delivery of the Spy Cas9 protein and an HBVspecific sgRNA to HBV-infected hepatoma cells using a lentiviral vector system (Shalem et al., 2014). We selected three distinct target sites in the HBV genome, focused on the viral surface Ag, core and RT ORFs (Fig. 1A) and were able to confirm that all three sgRNAS were active using indicator constructs containing cognate HBV DNA segments (Fig. 1). We next introduced each Cas9/sgRNA combination, or a control non-specific sgRNA, into the cell line HepaRG (Ladner et al., 1997) by lentiviral transduction, selected transducer cells using puromycin and then quantified the ability of the HBV DNA genome chromosomal integrated in these cells to support a complete viral replication cycle. To ensure that any observed inhibition did not reflect simply the mutational activation of the single integrated HBV copy, we removed Tet, which blocks HBV transcription in HepAD38 cells, 48 h prior to transduction resulting in the production of readily detectable levels of unintegrated HBV DNA, including cccDNA. Nevertheless, we observed a strong inhibition of viral DNA accumulation upon introduction of any of the three HBV-specific Cas9/sgRNA combinations, with the RT-specific sgRNA, which targets the essential "YMDD" motif in the HBV RT ORF for cleavage, being particularly effective. In fact, the RT-specific sgRNA reduced total viral DNA released into the supernatant media, by ~1000-fold and total intracellular HBV DNA levels by ~100-fold, which is comparable to the inhibitory effect seen upon retention of Tet in the culture media (Fig. 2A). The Cas9/sgRNA combinations used also inhibited the accumulation of cccDNA in this system, with the RT-specific sgRNA again most effective, though the maximal reduction in this case was ~10-fold (Fig. 2B). Measurement of viral surface Ag production also revealed a clear inhibition caused by all three Cas9/sgRNA combinations, though in this case the surface Ag-specific sgRNA was most effective, possibly due to mutational inactivation of the surface Ag ORF due to frameshifting (Fig. 2C). In contrast, only a modest inhibitory effect on HBeAg secretion was observed, likely in part because no sgRNA specific for HBV ORF was tested.
 
To examine whether HBV-specific Cas9/sgRNA combinations were indeed inducing inactivating indels at their predicted cleavage site in the HBV genome, as expected (Shalem et al., 2014), we cloned and sequenced PCR-generated DNA fragments that traversed the predicted cleavage site in the HepAD38 cells expressing the highly effective RT-specific sgRNA and observed deletions or insertions in 78 out of 103 sequenced clones. The RTspecific sgRNA already reduced HBV DNA levels by 100-1000-fold, and it is therefore clear that not only is very little HBV DNA present in the cells transduced with the RTspecific Cas9/sgRNA combination but also what little viral DNA is left has almost all been mutationally inactivated by error-prone repair of the Cas9-induced cleavage site.
 
To extend these studies to a system that supports de novo HBV infection, we also analyzed the effect of these Cas9/sgRNA combinations on HBV replication in the permissive HepaRG cell line (Gripon et al., 2002). As may be observed in Fig. 4, we again were able to detect a substantial inhibition in the accumulation of HBV DNA in these cells, including cccDNA (Fig. 4E) as well as a clear, statistically significant reduction in both HBgAg and HBeAg production at later time points (Fig. 4A and B). Finally, we also tested whether expression of Cas9/sgRNA combinations would be able to inhibit HBV replication more effectively when used together with known inhibitors of the HBV life cycle, including 3TC, TDF, ETV and HAP-12. As shown in Fig. 5 and Table 1, we were indeed able to observe enhanced inhibition of HBV replication by HBV-specific Cas9/sgRNA combinations in both HepAD38 cells and in the chronically HBV-infected cell line HBV2.2.15 (Sells et al., 1987) in the presence of various antiviral drugs. We therefore believe that the use of CRISPR/Cas system to directly target HBV cccDNA in the infected liver clearly merits further testing to assess whether there is real potential for a possible cure for chronic HBV infections when effective Cas9/sgRNA combinations are used together with current antiviral drugs.
 
While this current study used the Spy Cas9 protein, and cognate sgRNAs, delivered by lentiviral vector transduction (Shalem et al., 2014) we believe that future efforts should focus on the use of adeno-associated virus (AAV) vectors, as these can potentially be generated at very high titers, up to 1014 virion particles per ml, and have a high degree of tropism for hepatocytes. However, the Spy Cas9 gene, at ~4.2 kb, is too large for incorporation into AAV vectors, as these can only accommodate ~4.6 kb of DNA, which would need to include not only the Cas9 ORF but also the sgRNA, as well as the transcriptional regulatory regions required for Cas9 mRNA and sgRNA expression and, finally, the AAV inverted terminal repeats. However, there are several smaller (~3.2 kb) Cas9 genes known in other bacterial species of which at least one, expressed by Neisseria meningitidis (Nme) has been characterized (Esvelt et al., 2013). Our initial data indeed suggest that Nme Cas9 is as effective as Spy Cas9 in inducing target DNA cleavage and inactivation (data not shown), thus raising the clear possibility of generating single AAV vectors able to deliver complete, fully functional Cas9/sgRNA combinations specific for HBV to the liver in vivo at high efficiency. Whether Cas9/sgRNA combinations are entirely safe, and whether their efficiency can be high enough to effectively clear the HBV cccDNA pool, are issues that will next need to be carefully addressed in animal models of HBV infection (Dandri and Lutgehetmann, 2014).
 
While this article was in preparation, Lin et al. (2014) reported that HBV-specific Spy Cas9/sgRNA combinations were able to cleave HBV DNA when Cas9 and HBV expression plasmids were co-transfected into Huh7 cells. Neither HBV-infected human cells nor HBV cccDNA levels were examined in this previous work. Here, we substantially extend these recent findings by demonstrating HBV DNA cleavage and suppression, including repression of cccDNA accumulation, in both a model of chronic HBV infection (HepAD38 cells) and during de novo infection (HepaRG cells) in culture. These novel observations therefore provide substantial support for the idea that bacterial CRISPR/Cas systems have the potential to function as an effective treatment for DNA virus infections.