©2007 Baishideng Publishing Group Inc. All rights reserved.
Persistent occult hepatitis B virus infection: Experimental findings and clinical implications
Patricia M Mulrooney-Cousins, Tomasz I Michalak, Molecular Virology and Hepatology Research, Division of BioMedical Science, Faculty of Medicine, Health Sciences Centre, Memorial University, St. John’s, NL A1B 3V6, Canada
Author contributions: All authors contributed equally to the work.
Supported by operating research grants from the Canadian Institutes of Health Research, Canada and the Canada Research Chair Program, and the Canada Foundation for Innovation
Correspondence to: Tomasz I Michalak, MD, PhD, Molecular Virology and Hepatology Research, Faculty of Medicine, Health Science Centre, Memorial University, 300 Prince Philip Drive, St. John’s, NL A1B 3V6, Canada. firstname.lastname@example.org
Telephone: +1-709-7777301 Fax: +1-709-7778279
Received: July 24, 2007
Revised: September 5, 2007
Accepted: October 12, 2007
Published online: November 21, 2007
Despite the availability of effective prophylactic vaccines, hepatitis caused by hepatitis B virus (HBV) remains one of the most troublesome infectious diseases responsible for significant morbidity and mortality in the world. HBV-induced chronic hepatitis B (CHB), defined as a prolonged liver inflammation accompanied by HBV surface antigen (HBsAg) in serum for longer than 6 mo, develops in up to 10% of individuals infected as adults and in about 85% of individuals exposed in the neonatal period and early childhood[1,2]. Consequently, up to 400 million people worldwide are afflicted with serum HBsAg-positive chronic infection, which in high endemic areas is mainly acquired by vertical virus transmission. Furthermore, it is estimated that as much as one third of the world population have been exposed to HBV. Many of these individuals may unknowingly carry the virus. It is assumed that, due to low virus loads in many of the cases, HBV may not be identifiable by commonly used serological assays based on detection of HBsAg and that the infection is often asymptomatic or has non-specific symptoms. This serologically and clinically unapparent infection makes the true global incidence of HBV hard to estimate. Consequently, the pathogenic significance of HBV might be much greater than that currently assumed.
OCCULT HBV INFECTION
The clinical data and the findings from animal models accumulated in the past ten years strongly argue that persistent, serologically silent, i.e., serum HBsAg-negative, HBV infection is a natural consequence of resolved acute hepatitis B, but it may also occur following an asymptomatic exposure to HBV, and may have serious pathological and epidemiological repercussions. Another important aspect of infections with HBV and other related viruses, that has been known for a long time but remained controversial and/or not fully realized, is the fact that the immune system, in addition to the liver, is the site of active hepadnaviral replication and long-term persistence.
Occult HBV infection is defined as the existence of HBV DNA in the serum, cells of the lymphatic (immune) system, and/or hepatic tissue in the absence of serum HBsAg. Most frequently, occult HBV infection follows resolution of acute hepatitis and continues indefinitely after clearance of HBsAg and biochemical improvement in liver function[5-9]. For this reason, this form of infection is termed as residual or secondary occult infection or SOI, in opposite to primary occult infection or POI, which will be discussed later. Serological testing in SOI normally reveals the presence of antibodies to HBV core antigen (anti-HBc), which are now recognized not only as a valuable marker of prior exposure to HBV but also as an indicator suggesting actually progressing occult HBV infection. Nonetheless, recent estimates suggest that up to 20% of individuals with occult HBV carriage evidenced by HBV DNA detection could be nonreactive for anti-HBc or any other serological indicator of exposure to HBV. It also is of note that the detection of naturally acquired antibodies to HBsAg (anti-HBs) does not exclude the existence of occult infection[5-7,9].
DETECTION OF OCCULT HEPADNAVIRAL PERSISTENCE
One of the main current challenges in diagnosis of occult HBV infection is the relatively low sensitivity of HBV DNA detection assays. The present assays are based on different approaches to amplification of HBV DNA by polymerase chain reaction (PCR) and their lowest detection limits range between 200 and 1000 virus genome copies, which are also referred to as virus genome equivalents (vge) per mL of plasma or serum. Although these assays are adequate to detect prodromal and symptomatic HBV infections and to monitor the progress of antiviral therapy, they are yet not sufficiently sensitive to identify occult infection in the majority of the individuals affected, where serum loads are usually below 100 vge per mL, as well as to determine whether virus was indeed completely eradicated following antiviral treatment. In contrast to the clinical assays, the current research tests, which in our laboratory apply nested PCR and detection of resulting amplicons by nucleic acid hybridization (NAH), i.e., PCR/NAH, identify hepadnaviral DNA at loads below 10 vge/mL[10,11]. Furthermore, by testing of serial samples of serum and peripheral blood mononuclear cells (PBMC) and, when feasible, liver biopsies, and by amplifying different non-overlapping regions of the virus genome, detection of occult infection is greatly enhanced. Application of these ultrasensitive assays to investigation of clinical materials and animal models, particularly the woodchuck model of hepatitis B, led to identification in the last decade of occult hepadnavirus infection as a new, distinctive entity and to delineation of its unique characteristics. However, due to yet limited population-based studies and relatively low sensitivity of the assays approved for clinical testing, the full impact of this infection remains to be determined. Nonetheless, with the recent introduction of clinical HBV nucleic acid testing (HBV NAT) in many parts of the world, significant progress in this area is expected.
EXPERIMENTAL OCCULT HBV INFECTION IN THE WOODCHUCK MODEL
Many of the characteristics of occult HBV infection have been originally identified in the woodchuck model of hepatitis B. Woodchuck hepatitis virus (WHV) infecting the Eastern North American woodchuck (Marmota monax) is accepted as the most adequate, natural model of human HBV infection, hepatitis B and HBV-induced hepatocellular carcinoma (HCC). WHV closely resembles HBV in the structure, biology, and in the patterns and outcomes of the liver diseases induced. Much of the basic understanding of the natural course, requirements for transmission, molecular and immunological properties of occult hepadnaviral carriage, and its long-term pathological ramifications is owed to the woodchuck-WHV experimental system[10-15].
Occult WHV infection was first identified as a residual, life-long persistent infection in woodchucks who apparently completely resolved experimental acute hepatitis and developed anti-WHs (antibodies to WHV surface antigen, WHsAg). Analysis of sequential sera, PBMC and liver biopsies collected during the lifespan of these animals, as well as autopsy lymphatic and hepatic tissues, using sensitive PCR/NAH assays, identified WHV DNA in all three compartments, i.e., serum, lymphoid cells and the liver. In addition, the WHV replicating intermediates, covalently closed circular DNA (cccDNA), and virus mRNA were consistently found in the cells of the lymphatic system and the liver. Anti-WHc (equivalent of anti-HBc) persisted in all animals for life. This form of WHV infection become later termed as SOI[5,14]. In addition and unexpectedly, a low grade, intermittent liver inflammation was diagnosed throughout the remaining life in the majority of the woodchucks who apparently completely resolved acute hepatitis. Not less importantly, one fifth of the animals developed HCC in 3-5 years after recovery and despite anti-WHs presence. The same rate of HCC development was reported in woodchucks recovered from acute hepatitis which were examined by other investigators. Most recently, it has also been shown that residual WHV in animals with SOI can be reactivated by treatment with an immunosuppressive agent, Cyclosporin A, leading to transiently serum WHsAg-positive infection. Taken together, the data accumulated indicate that despite resolution of acute hepatitis, the replication of WHV continues for life, albeit at very low levels, both in the liver and the lymphatic system, that limited inflammatory process continues in hepatic tissue, and that the silently persisting virus may retain its oncogenic potency. This implies that although the immune system is able to resolve acute hepatitis and can keep persistently propagating hepadnavirus under relative control, it is unable to eradicate the virus completely[10,14]. The same conclusion was recently reached when bicistronic WHV core-gamma interferon (IFN) DNA vaccine was found to be able to prevent WHV hepatitis, but could not mount sterile immunity and prevent establishment of occult infection. Furthermore, it became apparent that the detection of anti-WHc in the absence of other serological indicators of infection is a reliable indicator of occult WHV persistence. This state, similarly as strong HBV-specific cytotoxic T cell (CTL) and T helper lymphocyte responses detectable years after resolution of acute hepatitis B[19,20], is likely a consequence of sustained restimulation of the immune system with a viral protein produced during low-level virion assembly. The high degree of compatibility between WHV and HBV infections and the data from studies on otherwise healthy anti-HBc-positive individuals[21,22] suggest that the occurrence of isolated anti-HBc could also be of value in identifying occult HBV persistence.
The virus recovered from woodchucks with SOI remains infectious. It was observed that WHV harvested from PBMC isolated during SOI and ex vivo stimulated with lipopolysaccharide (LPS) induced classical acute WHV hepatitis in virus-naive animals. Furthermore, woodchuck dams with SOI were found to transmit virus to offspring, however, the infection induced frequently displays characteristics different than those of SOI. Thus, WHV DNA and WHV cccDNA was harboured only in the lymphatic system, but not in the liver, and in the absence of serological markers of infection, including anti-WHc antibodies. This form of hepadnaviral infection was later named as primary occult infection or POI. Interestingly, POI, in contrast to SOI, is not associated with protection from reinfection and hepatitis after challenge with liver pathogenic doses of WHV. In subsequent studies, the same profile of occult infection was experimentally induced in adult animals by intravenous injections with WHV doses lower than 1000 virions. At the same time, it was also established that WHV quantities higher than 1000 virions, in addition to infecting the lymphatic system, consistently cause classical acute hepatitis. This revealed that doses greater than 1000 virions are liver pathogenic in woodchucks. From these data it was inferred that the threshold level of WHV required to infect lymphoid cells is 100 to 1000-fold lower than that to infect hepatocytes. WHV sequence variation was not responsible for the induction of these two contrasting forms of infection.
At the present time, POI is the entity which has not yet been convincingly identified in humans. However, the detection of occult HBV DNA-positive, anti-HBc-negative infection may suggest its existence. The persistence of trace amounts of replicating virus, especially at the extrahepatic locations, could have an impact in terms of unforeseen virus transmission and induction of disorders which are not yet considered to be related to hepadnaviral infection. Small quantities of the virus carried across the placenta may induce POI, similarly to CHB developing in neonates born to mothers with serum HBsAg-positive infection.
OCCULT HEPADNAVIRAL INFECTION AND THE HOST’S IMMUNE SYSTEM
Since the cells of the immune system are clearly the site of WHV propagation, independent of whether infection is symptomatic or occult, it became of interest to enumerate the cells carrying the virus in different forms of experimentally induced infection. By applying in situ PCR combined with flow cytometric quantification of cells containing amplified WHV core gene sequences, the proportion of WHV DNA-positive circulating lymphoid cells was found to range from 3.4% to 20.4% (mean 9.6%) in serum WHsAg-positive hepatitis, as compared to 1.1% to 14.6% (mean 4.8%) in occult WHV infection. There was no difference between SOI and POI in the numbers of infected cells.
In terms of WHV-specific T cell responses, recent studies have shown that animals with POI display weak but evident, persistent and multispecific virus-specific T cell proliferative reactivity and that this cellular response does not provide protection against challenge with liver pathogenic doses of WHV (Gujar et al, manuscript in preparation). The existence of active peripheral WHV-specific T cell response and innate immune cell activation in the livers of animals with SOI have also been established (Gujar et al, manuscript in preparation). Overall, the findings in experimental SOI in woodchucks are closely compatible with those reported for humans convalescent from acute hepatitis B[19,20].
PRIMARY VERSUS SECONDARY OCCULT WHV INFECTION
In summary, experimental occult hepadnaviral infection occurs in two distinctive forms, as primary and secondary infection. POI is induced upon exposure to WHV doses at or below 1000 vge and engages the lymphatic system, although the infection with time may also spread to the liver[11,14]. Animals with POI carry virus in serum and lymphoid cells at levels which are comparable with those occurring in SOI and mount virus-specific T cell proliferative response, but they do not have serological (immunovirological) markers of infection and they are not protected from challenge with liver pathogenic doses of the virus. On the other hand, in SOI, a low-level replication of the virus progresses in both the liver and the lymphatic system, anti-WHc and frequently anti-WHs are detectable, and the animals are protected from hepatitis when re-exposed to liver pathogenic doses of WHV. Moreover, while SOI can be accompanied by low grade, protracted liver inflammation and HCC may finally develop, the liver is entirely normal in POI and a possible development of hepatoma has not yet been assessed.
IMPLICATIONS OF OCCULT HEPADNAVIRAL PERSISTENCE
Given the high degree of similarity between WHV and HBV and infections induced by these viruses, it is important to recognize the impact of silent HBV persistence in terms of its overall incidence, infectivity and pathogenic consequences. The previous clinical and research focus on the liver as essentially the only reservoir of HBV has somehow restricted characterization of extrahepatic sites of hepadnaviral replication and its potential pathological consequences. Recent studies are bringing occult HBV infection and the virus’ lymphotropic nature to the forefront. Some of the cardinal works on occult HBV infection were accomplished by examining lymphoid cells of patients with resolved acute hepatitis B where HBV DNA sequences were found[19,24,25]. Furthermore, immunological studies showed that vigorous CTL and T cell proliferative responses specific against HBV antigens persisted for years after recovery[19,20]. Many case reports also indicate that immunosuppression caused by chemotherapy[26,27], immunomodulatory agents, or immune deficiencies, such as HIV infection or hematological malignancies, can reactivate occult infection. As in woodchucks with SOI, mild necroinflammation has been documented in liver samples obtained many years after recovery from acute hepatitis B[31,32]. Liver fibrosis and cirrhosis of unknown origin has now been explained by occult HBV infection in many retrospective studies[33,34]. There also is strong evidence for the risk of HCC development in patients with occult HBV[35-38] and this risk is further elevated in alcoholics and in patients with other liver ailments, like hepatitis C. Interestingly, convincing evidence points to the conservation of the HBV genome sequence in occult infection, negating the possibility that faulty HBsAg production accounts for all incidences of occult infection[32,41]. On the other hand, studies demonstrating the passage of HBV in non-liver cells, such as stem cells[42,43] and bone marrow, and by blood donations collected from individuals with occult HBV are increasingly frequent in the literature. Recent estimates suggest that transient HBV DNA levels of up to 104 vge/mL can be found in sera from apparently healthy individuals with occult infection, which evade detection by current clinical assays, especially when testing is done on single serum or plasma samples[9,46].
The seriousness of the consequences of occult HBV infection is yet not fully recognized. Accumulated evidence indicate that occult HBV can be both a source of virus contamination in blood and organ donations, as well as the reservoir from which full blown hepatitis can arise. The oncogenic potency of occult HBV persistence becomes progressively evident. This silent infection can also affect the progression and outcomes of other viral diseases, particularly hepatitis C. It can be expected that while the burden of CHB will decrease with time due to introduction of more effective antiviral therapies, occult HBV infection could become a main concern. Understanding in its entirety of the nature and consequences of this form of HBV persistence should be of a high priority and this quest has already been initiated.
S- Editor Liu Y L- Editor Rippe RA E- Editor Lu W