-
HBV is the type member of the Hepadnaviridae and its replication cycle starts with binding of the PreS1 domain (amino acids 8–18) of the large envelope protein to the sodium taurocholate cotransporting polypeptide (NTCP) receptor (Yan et al., 2012). Infectious virions contain a relaxed circular DNA (RC-DNA), and the RC-DNA is converted into covalently closed circular DNA (cccDNA) in the host cell nucleus. Serving as a transcription template, cccDNAs are transcribed into several genomic and subgenomic RNAs by cellular RNA polymerase II. Of these RNA transcripts, the pregenomic RNA (pgRNA) is further reverse transcribed by the reverse transcriptase (RT) activity of the P protein into a new RC-DNA genome (Feng and Hu, 2009). Matured RC-DNA-containing nucleocapsids are subsequently enveloped and finally released from host hepatocytes as progeny virions, or re-enter into the nucleus supplementing the cccDNA reservoir (Beck and Nassal, 2007) (Figure 1A).
Figure 1. (A) The HBV life cycle (modified from Nassal, 2015; Gut; with the author’s permission) and the corresponding target interfering strategies (red). I. Attachment and entry. II. Uncoating. (a) Transcription of viral RNAs. (b) RNA nuclear export. (c) Translation. (d) Epigenetic regulation of cccDNA by HBx. (e) Encapsidation with co-packaging of P and pgRNA. (f) Negative strand synthesis (inhibited by nucleoside analogues), pgRNA degradation and positive strand synthesis. (g) Nuclear transport. (h) The conversion of RC-DNA to cccDNA. (i) Envelopment of mature RC-DNA containing nucleocapsids. L, large; M, middle; NA, nucleoside analogue; NTCP, sodium taurocholate cotransporting polypeptide; S, small. (B) The conversion of RC-DNA to cccDNA and the targeting strategy in this process.
-
cccDNA is a mini-chromosome in the infected cell nucleus and is the key element in the establishment of a persistent infection. Upon successful infection, the RC-DNA is converted into cccDNA, which is highly stable and acts as a virus transcription template. For viral RNA transcription, cccDNA is transcribed to pgRNA and other viral mRNAs. There are 3–50 copies of cccDNA per infected cell. The number of copies of the cccDNA decreases during the process of host cell division. So, recycling of the new RC-DNA to the nucleus occurs in order to replenish the relatively stable cccDNA copy number (Figure 1A). There are three main steps (Figure 1B) in the RC-DNA to cccDNA conversion: (1) P protein that is covalently linked to the 5′ end of the (–)-DNA is removed; (2) the 5′ end of the (+)-strand consisting of an RNA oligonucleotide is removed; (3) both strands are covalently ligated. To date, the mechanism of RC-DNA transformation into cccDNA is little understood (Nassal, 2015). A study reported that a "P-free RC-DNA" may be the intermediate of RC-DNA conversion into cccDNA (Gao and Hu, 2007). According to Guo et al., P-free (or deproteinized) RC-DNAs are distributed almost equally between the cytoplasm and nucleus (Guo et al., 2007). Other studies showed more deproteinized RC-DNA in the nucleus than in the cytoplasm, which was perhaps due to nuclease treatment of cytoplasmic lysate before Hirt extraction. Guo et al. showed that inhibiting the nuclear importation machinery resulted in accumulation of deproteinized RC-DNAs in the cytoplasm and the reduction of such DNAs in nucleus, which suggested that the deproteinized RC-DNA is a functional precursor of cccDNA formation (Guo et al., 2010).
In fact, a large amount of data show that there is a close relationship between viral replication and the intracellular repair system (Lilley et al., 2007). The formation of cccDNA requires multiple enzymatic activities (Figure 2). It has been proven that tyrosyl-DNA phosphodiesterases 2 (TDP2) can remove P proteins from authentic HBV and duck hepatitis B virus (DHBV) RC-DNAs in vitro (Königer et al., 2014). The gap requires filling-in by a DNA polymerase. DNA polymerase κ is a key cellular factor involved in cccDNA formation (Qi et al., 2016). HBV RC-DNA, with incomplete incision double chain structure, is likely to be the substrate of the 5′-flap endonuclease (FEN1). As a structurally specific nuclease, FEN1, can remove the free ends of the 5′ single-stranded nucleic acids and is an important member of the DNA repair system (Wood et al., 2001). In addition, it was found that DNA repair system proteins such as tyrosyl-DNA phosphodiesterases (TDPs), DNA polymerases (Qi et al., 2016; Königer et al., 2014), and DNA ligases are involved in this process (Haitao Guo, personal communication, in 2016 International HBV Meeting).
Figure 2. A detailed view of the 5′ ends of (–)-DNA (upper strand) and (+)-DNA is shown. The distance between DR1 and DR2 is not drawn to scale. The (–)-DNA end forms a 5′ flap structure, including the linked P protein and a 10 nt terminal redundancy. P is removed by TDP2, followed by a downstream incision by a 5′ flap endonuclease, and repair synthesis and ligation; cleavage exactly after the 5′ proximal nt c3100 (c for complement to nt 3100 in the (+)-sense) would allow direct ligation without new DNA synthesis. Like 5′ flap endonucleases in TOP1cc repair, 3′ endonucleases may play a role despite their seemingly improper substrate specificity (e.g. acting as 5′ flap endonuclease by association with a different partner, or providing an accessible upstream 3′ OH end to start displacement synthesis). The (+)-DNA 5′ end with the bound RNA primer is topologically equivalent; however, no protein “processing” (TDP2 or proteolysis) would be required. The incomplete (+)-DNA 3′ end can be filled-in by cellular DNA polymerase (Qi et al., 2016). The involvement of cellular DNA ligases is definite, and of end-polishing enzymes is possible also.
-
Few experimental in vitro and in vivo models for the study of HBV cccDNA exist. At present, three kinds of in vitro cell models are available: primary tupaia hepatocytes (PTH), primary human hepatocytes (PHH) and differentiated HepaRG (dHepaRG). These cell types can be applied for HBV infection. In infected cells, cccDNAs are produced and serve as templates for HBV replication. However, these models have their own limitations that need to be improved under special experimental conditions (Lucifora and Protzer, 2016). In subsequent studies, some new cell lines were developed. HepAD38 is a tetracycline-inducible HBV stable-transfected cell line (Ladner et al., 1997). The expression of pgRNAs is inhibited by tetracycline. Once the tetracycline is removed into the extracellular supernatant, viral pgRNAs in the cells are expressed again. Unlike HepG2.2.15, in which there is no detectable cccDNA during HBV replication, HepAD38 cccDNAs could be found within cellular nuclei, which is similar to the authentic viral life cycle in vivo. In addition, due to its high replicative levels, HepAD38 is widely used for the screening of antivirals (Ladner et al., 1997). Zhu Haizhen’s group established a new liver cancer cell line, HLCZ01, which supports the entire life cycle of HBV and hepatitis C virus (HCV) replication, providing a new platform for the study of cccDNA (Yang et al., 2014). Furthermore, Yamanaka and colleagues reported that pluripotent stem cells could be in vitro induced into hepatocyte-like cells (HLCs) in which a panel of transcription factors, which usually are found only in adult-derived hepatocytes, were expressed (Takahashi et al., 2007 ; Takahashi and Yamanaka, 2006). The HLCs have been shown to be self-renewing in culture, remain genetic stability (Si-Tayeb et al., 2010; Touboul et al., 2010; Schwartz et al., 2014) and are susceptible to HBV infection (Schwartz et al., 2014). Recent research revealed that the HBV functional receptor is NTCP (Yan et al., 2012). A NTCP-transduced in vitro infection model, HepG2-NTCP and Huh7-NTCP cell lines, thus, has been applied to study HBV infection and cccDNA formation (Hayes and Chayama, 2016; Ni et al., 2014; Qi et al., 2016; Yan et al., 2012).
In addition to the above mentioned in vitro HBV infection models, there are in vivo models for studying HBV cccDNA, but these are still limited. Animal models of HBV replication, such as transgenic mice, human liver chimeric mice, the woolly monkey model and the chimpanzee model have been reported. Even though HBV transgenic mice offer many possibilities for performing preclinical antiviral testing in vivo, the cccDNA is not built within the murine hepatocytes, whereas transcriptionally active cccDNA in human liver chimeric mice was reported (Allweiss and Dandri, 2016). However, the technique for the construction of chimeric liver is expensive and time-consuming. Chimpanzees have been proven to be an invaluable model for investigations on HBV-driven disease pathogenesis and the testing of novel antiviral therapies, as well as having detectable cccDNA in infected livers (Wieland, 2015). Although they mimic a natural human HBV infection to the closest degree, their large-scale application has been restricted, not only due to the cost, but also because of the demanding ethical concerns (Table 1).
Models Advantages Disadvantages Cell models PTH/PHH -intracellular environment close
to that in vivo-poor accessibility;
-limited cell numbers;
-lot-to-lot variation;
-genetic manipulation impossibleHepaRG -homogeneous source -license protected, which limits wide application;
-induction of differentiation time-consuming
and complicatedHepAD38 -homogeneous source;
-genetic manipulation possible-intracellular environment quite different
from that in vivoHLCZ01 -homogeneous source;
-genetic manipulation possible-not yet widely used NTCP-HepG2/Huh7 -homogeneous source;
-genetic manipulation possible-intracellular environment quite different
from that in vivoAnimal models Transgenic mice -wide applications;
-simple techniques-no detectable cccDNAs in liver;
-no associated hepatitis disease for mouseHuman liver chimeric mice -cccDNAs detectable in liver -high costs;
-unstable results;
-low cccDNA copy numbers;
-complicated techniquesWoolly monkey -cccDNAs detectable in liver -high costs;
-long time periods for study;
-ethical concernsChimpanzee -cccDNAs detectable in liver;
-situation in vivo similar to that in human-high costs;
-long time periods for study;
-ethical concernsTable 1. A comparison of the advantages and disadvantages for cellular and animal models with detectable cccDNAs
Recently, a minicircle HBV cccDNA was engineered with a Gaussia Luciferase reporter (mcHBV-GLuc cccDNA), which serves as a substitute for the detection of cccDNA activity. This model is independent of HBV infection and can be used to study the biology of cccDNA and to screen anti-cccDNA drugs (Li et al., 2016). There is a similar, but different, cccDNA system. A recombinant cccDNA (rcccDNA) produced by using minicircle technology resembles closely its wild-type counterpart, both in structure and function. It is a convenient technology to produce a large quantity of rcccDNA as a surrogate for wild-type cccDNA for investigating HBV biology (Guo et al., 2016). The group of Deng Qing produced a rcccDNA that established HBV persistence with sustained liver injury in immunocompetent mice. The HBV genome was engineered with a Cre/LoxP intron, which could be distinguished using polymerase chain reaction (PCR). This model system provides an innovative approach for testing antivirals directly targeting HBV cccDNA (Qi et al., 2014). Through minicircle technology, HBV circle, a rcccDNA, was easily generated and extracted from a genetically engineered Escherichia coli strain. Upon transduction of mouse liver through hydrodynamic injection, HBV circle formed cccDNA-like molecules in vivo and persistent virus replication was successfully established. The HBV circle is a close mimic of cccDNA and it represents a novel tool for addressing HBV cccDNA issues and for anti-HBV drug discovery (Yan et al., 2017).
At present, the currently available infection-dependent systems still have some technical limitations, for example, proper HepaRG differentiation requires a long and complex process. However, the infection-independent cccDNA model system is becoming a useful tool for finding cccDNA-relevant host factors involved in the intracellular steps of hepadnavirus replication and for screening chemical inhibitors of cccDNA formation. Minicircle technology is able to produce cccDNA-like molecules, but de novo biogenesis of cccDNA cannot be investigated through this system. Thus, more robust infection-susceptible cellular and animal models need to be found.
-
In this review, recent progress in the study of cccDNA is summarized, but many problems have yet to be solved (Table 2); for example, whether chronic hepatitis B can be completely cured, that is, to eliminate cccDNA, or how to achieve a functional cure (control cccDNA expression leading past HBsAg loss) (Revill and Locarnini, 2016). At the same time, none of hepatocytes are damaged during the process of eliminating cccDNA.
Strategies References Advantages Limitations Future directions Host DNA damage response Pommier et al., 2014;
Königer et al., 2014;
Yang et al., 2014;
Cui et al., 2015;
Qi et al., 2016;
Oqi and Lehmann, 2006;
Oqi et al., 2010;
Ran et al. 2017;
Schreiner and Nassal, 2017.The present experimental evidence shows many members of the host DNA damage response, such as TDP2, DNA polymerase, especially DNA polymerase κ, and the host DNA ligases are involved in the conversion of RC-DNA to cccDNA. Interference with DNA repair response may cause severe genetic instability of host hepatocytes. The study of how the host DNA damage response is involved in the conversion of RC-DNA to cccDNA is just at its beginning. Through knockout of a defined DNA repair gene, an individual host factor in the DNA repair pathway can be investigated for its involvement in cccDNA formation. Epigenetic regulation of cccDNA Belloni et al., 2009;
Guo et al., 2014;
Rivière et al., 2015;
Decorsière et al., 2016;
Murphy et al., 2016;
Niu et al., 2017;
Schubeler, 2015;
Vivekanandan et al., 2010;
Hong et al., 2017;
Palumbo et al., 2015.HBx is required for cccDNA transcriptional activity; thus, it represents an attractive anti-cccDNA target. DNMTs could be utilized to inhibit cccDNA transcription. IL6 has a repressive effect on HBV replication. The mechanism of HBx-mediated cccDNA transcription activation is to a large extent unclear. cccDNA functional inactivation, instead of elimination, may be only a transient solution. The role of DNMTs in HBV DNA methylation has not been clearly identified. Epigenetic regulation of cccDNA transcription activity should be addressed in detail, including post-translational modifications of histones, nucleosome spacing, non-coding RNAs, etc. Immune-mediated degradation Lucifora et al., 2014;
Luo et al., 2016;
Kitamura et al., 2013.High doses of IFN-α or LTβR-agonists were able to trigger noncytolytic degradation of cccDNA. This underscores the value of activating the innate immune response. A fraction of the cccDNA pool is refractory to further reduction. Studies, such as of the Toll-like receptor 7 agonist GS-9620, are indicative of the importance of reactivating host innate response. Genome-editing nucleases Seeger and Sohn, 2016;
Lin et al., 2014;
Ramanan et al., 2015;
Zhen et al., 2015;
Lin et al., 2015;
Fu et al., 2013.The application of artificial endonucleases has promoted cccDNA-related studies. More than 90% of HBV DNAs were cleaved by Cas9. The CRISPR/Cas9 system has been used to target HBV DNA, including cccDNA in cell culture and in the murine model. It is so far unconfirmed that the nuclease has been delivered to each persistent infected hepatocyte. The potential off-target effects may have an adverse effect on liver cell function, especially in the presence of long-acting effector enzymes. The CRISPR/Cas9 system can lead to indels in the host genomes. Finally, excessive amounts of pre-existing RC-DNA may supplement the cccDNA source. How nucleases enter into all HBV-infected hepatocytes to attack cccDNA, and how to avoid off-target effects of CRISPR/Cas9 etc. remain major challenges. Table 2. Comparison of four strategies against HBV cccDNA
The clinical application of antiviral drugs against HBV is still very limited. There are two types of antivirals for the treatment of chronic hepatitis B. IFN-α could modulate the host antiviral immune responses and nucleoside analogue could inhibit HBV DNA polymerase (Keeffe et al., 2008; Zoulim and Locarnini, 2009). However, these drugs cannot eliminate cccDNA. By using the HepDE19 system, researchers previously screened a small-molecule compound library and identified two disubstituted sulfonamide compounds that act as inhibitors of cccDNA formation by blocking RC-DNA deproteinization (Cai et al., 2012). So far, no drugs have been found directly targeting cccDNA, although some progress has been made towards developing small molecule anti-HBV compounds to interfere with other steps of HBV replication. HBV nucleocapsid assembly is considered as a potential target. Prototypes of small molecule modulators of HBV nucleocapsid assembly have been developed and partly tested through clinical phase I (Yang and Lu, 2017). Cyclosporin derivatives inhibit HBV entry with no decrease in NTCP transporter activity, so bile acid transport would not be influenced. These anti-HBV molecules may be potential candidates for developing novel drugs with fewer adverse effects in the near future (Shimura et al., 2017).
The underlying mechanism of cccDNA formation is not yet well understood, except that cccDNA starts with an RC-DNA molecule as a precursor prior to a series of biochemical steps. In the conversion of RC-DNA to cccDNA, many details such as the removal of P protein, the capped RNA primer and finally formation of closed loops are not yet clear. What we can be sure of is that HBV could not perform the multiple steps of changing RC-DNA to cccDNA without exploiting host nucleic acid enzymes that are derived from the DNA repair system. However, the host DNA repair system is a complex process and how it interferes with HBV or the formation of cccDNA has not been studied thoroughly. Understanding the molecular mechanism of the conversion of RC-DNA into cccDNA is critical, and may aid the development of strategies to cure chronic hepatitis B, and the host DNA repair system represents a direction that can be dug deep. cccDNA transcriptional activity is positively affected by HBx, which is suspected to be an oncogene. HBV expression influences DNA repair owing to HBx, which is believed to interact with DNA repair proteins (Guo et al., 2014; Murphy et al., 2016). So, HBx connects HBV to the host DNA damage response and becomes an attractive target. Two studies in HepaRG cells demonstrated that HBV could suppress innate responses, possibly via its P protein (Liu et al., 2015; Luangsay et al., 2015). However, other studies were suggestive that HBV is a stealth virus that simply goes undetected and, hence, it possesses the potential to escape immune system attacks (Niu et al., 2017). A fraction of the cccDNA pool seems to be refractory to further reduction from the results of recent studies (Lucifora et al., 2014; Luo et al., 2016; Kitamura et al., 2013). Anyway, for the consideration of re-waking host immune responses to eliminate cccDNA, there are still many difficulties to be overcome. At present, the unambiguous detection of HBV cccDNA is still difficult. The results of different methods for quantifying cccDNA are widely divergent in the same system. For example, Southern blotting and cccDNA-specific PCR have limitations (Nassal, 2015). The results of two methods are in many cases inconsistent and we don’t know which of the results is believable. The main issue is how to distinguish cccDNA from sequence-identical non-cccDNA forms (Schreiner and Nassal, 2017). This situation has severely limited the assessment of cccDNA. Hence, it is urgently necessary to develop an internationally recognized standard test method.
In conclusion, the development of more effective, better-tolerated and affordable antiviral treatment is crucial for controlling the progression of hepatitis, cirrhosis and hepatocellular carcinoma. To achieve the ultimate cure of a true chronic hepatitis B, further basic studies and a more detailed understanding of cccDNA formation mechanisms are still needed.
Recent advances in the study of hepatitis B virus covalently closed circular DNA
- Received Date: 28 April 2017
- Accepted Date: 27 November 2017
- Published Date: 21 December 2017
Abstract: Chronic hepatitis B infection is caused by hepatitis B virus (HBV) and a total cure is yet to be achieved. The viral covalently closed circular DNA (cccDNA) is the key to establish a persistent infection within hepatocytes. Current antiviral strategies have no effect on the pre-existing cccDNA reservoir. Therefore, the study of the molecular mechanism of cccDNA formation is becoming a major focus of HBV research. This review summarizes the current advances in cccDNA molecular biology and the latest studies on the elimination or inactivation of cccDNA, including three major areas: (1) epigenetic regulation of cccDNA by HBV X protein, (2) immune-mediated degradation, and (3) genome-editing nucleases. All these aspects provide clues on how to finally attain a cure for chronic hepatitis B infection.