Hepatitis B Virus Large and Middle Glycoproteins A
http://www.100md.com
病菌学杂志 2005年第20期
Department of Biochemistry and Molecular Pharmacology, Thomas Jefferson University College of Medicine, Philadelphia, Pennsylvania
Drexel Institute for Biotechnology and Virology Research, Department of Microbiology and Immunology, Drexel University College of Medicine, Doylestown, Pennsylvania
Institute for Hepatitis and Virus Research (The Pennsylvania Commonwealth Institute), Doylestown, Pennsylvania
Department of Biochemistry, University of Oxford, Oxford, United Kingdom
ABSTRACT
The secretion of hepatitis B virus (HBV) large (LHBs) and middle (MHBs) envelope polypeptides from tissue cultures requires proper protein folding and is prevented by inhibitors of the endoplasmic reticulum (ER) glucosidase. Using competitive inhibitors of the ER glucosidase, here it is shown that the amounts of glycosylated and unglycosylated forms of LHBs and MHBs proteins are all greatly reduced in tissue cultures producing HBV envelope glycoproteins. In contrast, the HBV small (SHBs) protein was not affected. The reduction in secretion of LHBs and MHBs proteins appears to be mediated by proteasomal degradation pathways, since it is prevented by either lactacystin or epoxomicin, two inhibitors of proteasomal degradation. Although there is no detectable proteasomal degradation of LHBs and MHBs in cells with functional glucosidase, the implications of the nearly quantitative sensitivity of glycosylated and unglycosylated forms of LHBs and MHBs proteins, with selective sparing of SHBs protein, in cells in which glucosidase is inhibited is surprising, and its implications are discussed.
INTRODUCTION
Hepatitis B virus (HBV) is the human member of the family Hepadnaviridae and worldwide is associated with more than 350 million chronic infections and nearly one million deaths annually (6, 16, 27). The infectious agent is a small, 42-nm, enveloped particle containing an incompletely double-stranded DNA genome of approximately 3.5 kb (26). Although the replication of the viral genome occurs in the cytoplasm and has been well characterized, viral morphogenesis and secretion are less well understood. As with many viruses, production of infectious viral particles is inefficient, and the management of defective or unused viral gene products is not well studied.
The secretion and morphogenesis of HBV require viral envelope glycoproteins. HBV specifies three envelope proteins, called large ("LHBs"), middle ("MHBs"), and small ("SHBs") (7, 13) that are all derived from the same open reading frame and may exist in the viral particle as either unglycosylated or N-glycosylated forms (11, 14). Secretion of HBV enveloped DNA is prevented by inhibitors such as the endoplasmic reticulum (ER) glucosidase, implying a critical role for glycoprocessing in the trafficking and morphogenesis of viral glycoproteins (3-5, 17, 18, 20-22).
Many nascent N-linked glycoproteins depend upon an interaction with the lectin-like chaperon, calnexin (CNX) to fold properly. CNX recognizes monoglucose residues on the oligosaccharide of the nascent glycoprotein, which are formed by the sequential action of the ER glucosidases (2). Why some, but not other, glycoproteins appear to have an obligate requirement for CNX-mediated folding is unclear, but the extreme sensitivity of HBV secretion to glucosidase function was assumed to be due to an obligate requirement of HBV glycoproteins for CNX-mediated protein folding. Indeed, both LHBs and MHBs proteins have been shown to interact with CNX (24, 32), and the secretion of MHBs is prevented by glucosidase inhibitors (18, 20).
However, the role of MHBs protein in mediating virus secretion is controversial, and there is evidence that MHBs is not essential (7). Thus, it was not clear how prevention of only MHBs biogenesis with glucosidase inhibitors could be responsible for the selective reductions of HBV secretion observed in glucosidase-inhibited cells.
In addition, although the amount of MHBs protein secreted into the culture medium from cells in which glucosidase has been inhibited has been shown to be reduced, the mechanism of reduction and fate of these polypeptides have not been clearly determined. There is even less information about the sensitivity of LHBs protein. Indeed, previous work had suggested that, despite being reduced in secretion, MHBs protein actually accumulated in glucosidase-inhibited cells (18, 19). Those conclusions were largely based upon detection of HBs epitopes using an antigen capture (enzyme-linked immunosorbent) assay. The state of intact LHBs and MHBs proteins was not conclusively explored.
It was therefore of interest to more precisely explore the fates of LHBs and MHBs proteins in HBV-producing cells in which ER glucosidase had been inhibited. In this study, Western blotting and immunoprecipitation (IP) analysis have confirmed that LHBs and MHBs, but not SHBs, are highly sensitive to glucosidase inhibitors. The amounts of LHBs and MHBs proteins became greatly reduced, by Western blot analysis, within 6 days of incubation with glucosidase inhibitors. Surprisingly, both the glycosylated and unglycosylated species were reduced. The reduction in the amounts of LHBs and MHBs proteins in glucosidase-inhibited cells was prevented by inhibition of proteasomes. The implications of these findings for normal HBV particle biogenesis and cellular management of misfolded HBV glycoproteins are discussed.
MATERIALS AND METHODS
Cells and compounds. HepG2 cells, a stable tissue culture line derived from a human hepatoblastoma, were purchased from the American Type Culture Collection (Rockville, MD) and grown in RPMI 1640 (Gibco-BRL, Rockville, MD) containing 10% fetal bovine serum (Gibco-BRL). HepG2 2.2.15 cells, derived from the stable transfection of HepG2 cells with a dimer of the HBV genome producing HBV viral and subviral particles at physiologic conditions, were kindly provided by George Acs (Mt. Sinai Medical College, New York, NY) in 1992 and maintained as HepG2 cells but with the addition of 200 μg/ml of G418 (Gibco-BRL) (29). N-Butyl-deoxynojirimycin (NB-DNJ) and deoxynojirimycin (DNJ) are glucosidase inhibitors and were provided by Monsanto Searle (St. Louis, MO) and Synergy Pharmaceuticals, Inc. (Edison, NJ), respectively. Castanospermine and proteasome inhibitors (lactacystin [LCT] and epoxomicin [EPO]) were purchased from Sigma-Aldrich (St. Louis, MO) and Calbiochem, Inc. (La Jolla, CA), respectively.
Use of inhibitors. In all cases, tissue culture was performed using fully confluent HepG2 or HepG2 2.2.15 cells. Glucosidase inhibitors (NB-DNJ or DNJ) were incubated on cells for a total of 7 days. Medium was changed with fresh compound (where indicated) every 2 days. On day 7, culture media and cells were collected. For the inhibition of proteasome degration, the cells were treated with either 20 μM lactacystin or 1 μM epoxomicin in the absence or presence of a glucosidase inhibitor for 16 h after 6 days of glucosidase inhibition.
Cytotoxicity (cell viability) assay. HepG2 2.2.15 cells were treated with NB-DNJ or DNJ (4.52 mmol/liter) as explained above. The viability of cultures treated with the indicated concentration of imino sugar inhibitor was measured by Trypan blue dye exclusion staining on days 0, 3, 5, and 7 during drug treatment. Based on their color (white), viable cells were counted on a hemocytometer by a light microscope. The cytotoxicity of glucosidase inhibitors was also evaluated by the mitochondrial toxicity test and the lactate dehydrogenase (LDH) assay. The assays show that the glucosidase inhibitor has no observed toxicity (data not shown).
Plasmids and transfection. HepG2 cells were transfected with either the pCMV-M expression vector (23) or pTRE-L expression vector with Lipofectamine or FuGene 6 according to the manufacturer's directions. As a control, pCMV-GFP (a cytomegalovirus-green fluorescent protein plasmid) was used to monitor transfection efficiency. After transfection, the cells, allowed 1 day for recovery, were left untreated or were treated with the glucosidase inhibitor DNJ for 6 days. The cells were harvested in cell lysis buffer containing 1x protease inhibitors, and the LHBs and MHBs proteins and SHBs proteins were detected by Western blotting or Abbott enzyme-linked immunosorbent assay (ELISA), respectively.
Detection of LHBs and MHBs surface antigens by Western blotting. HBV large and middle glycoproteins, from either clarified culture medium or lysed cells, were dissolved in loading buffer, resolved by electrophoresis through 12% polyacrylamide sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE), transferred to polyinylidene difluoride (Millipore) membranes, and blocked with 5% powdered milk in 0.1% Tween 20-phosphate-buffered saline (PBS) at room temperature for 1 h. After incubation with polyclonal pre-S2 antibody (Research Diagnostics, Inc., Flanders, NJ) at a 1:1,000 dilution in 0.1% Tween 20 with 2% bovine serum albumin for 2 h at room temperature, the blot was washed three times with 0.1% Tween 20-PBS for 10 min/each wash. The blot was incubated with secondary antibody (peroxidase-conjugated donkey anti-rabbit immunoglobulin G serum) at a 1:5,000 dilution in 0.1% Tween 20 with 2% bovine serum albumin for 1 h at room temperature. Blots were washed three times with 0.1% Tween 20-PBS, and proteins were detected by enhanced chemiluminescence (ECL; Amersham Corporation, Arlington Heights, IL).
Detection of HBsAg by ELISA. Two hundred microliters of media from the same studies was used to determine the level of hepatitis B surface antigen (HBsAg) in the culture medium of untreated (UNT) and treated cells. Analysis was performed using the Abbott Diagnostics AUSZYME Monoclonal Diagnostic kit according to the manufacturer's directions (Abbott Laboratories, North Chicago, IL).
Detection of SHBs glycoprotein and ATT by immunoprecipitation. HepG2 and HepG2 2.2.15 cells were used to detect the secretion of a cellular glycoprotein such as SHBs and alpha 1 anti-trypsin (AAT) after treatment with and removal of the glucosidase inhibitor. Briefly, the cells were treated with 4.52 mmol/liter DNJ for 6 days. After the treatment, the cells were incubated with [35S]methionine (200 μCi/ml) for 16 h in the absence or presence of glucosidase inhibitor. The culture supernatant was aspirated and clarified at 13,000 rpm in a bench top centrifuge to remove cellular debris. The cells were harvested in cell lysis buffer containing 1x protease inhibitors. Either the lysed cells or the clarified medium containing radiolabeled proteins was adjusted to 50 mM Tris-HCl, pH 7.4, 150 mM NaCl, 1 mM EDTA, 0.05% SDS, and 0.2% NP-40 (final concentrations) and used in IP assays. The clarified radiolabeled medium was incubated with either human monoclonal antitrypsin antibody (Sigma-Aldrich) or anti-HBsAg monoclonal antibody (DakoCytomation, Inc., Carpinteria, CA) at 4°C overnight. The immune complexes were then precipitated using protein G-agarose (Roche Diagnostics Corp., Indianapolis, IN). Following washing with lysis buffer, the immune complexes were released from protein G-agarose using SDS-PAGE denaturation buffer and boiled for 5 min. Samples were fractionated by SDS-PAGE (12% polyacrylamide), and the labeled proteins were detected with a Bio-Rad Personal FX phosphorimager.
RESULTS
Steady-state levels of unglycosylated and N-glycosylated LHBs and MHBs proteins as a function of glucosidase inhibition. Previous studies have shown that the secretion of enveloped HBV DNA and MHBs protein from tissue cultures is prevented by inhibition of the ER glucosidase (3-5, 17, 18, 20-22). Based upon ELISA-type analysis, it was also suggested that SHBs epitopes, derived from MHBs, were detained within the glucosidase-inhibited cells and accumulated. It was assumed that MHBs protein was accumulating, although MHBs polypeptide was not specifically identified. Therefore, to confirm the sensitivity of MHBs protein biogenesis to glucosidase inhibitors and to more precisely study the extracellular and intracellular fates of the LHBs and MHBs polypeptides, Western blot analysis was performed on cell lysates and culture medium of HepG2 2.2.15 cells as a function of glucosidase inhibition. Briefly, HepG2 2.2.15 cells were cultured for 6 days in the absence or presence of 4.52 mmol/liter DNJ. DNJ is a competitive inhibitor of ER glucosidases I and II, and 4.52 mmol/liter is sufficient to inhibit at least 95% of the glycan processing activity in these cells (3, 20).
As shown in Fig. 1A (lane UNT), LHBs and MHBs proteins and their glycoforms are well resolved in this gel (Fig. 1A). That the lower-molecular-mass band (faster mobility) of the MHBs-specific bands is the unglycosylated species (30 kDa) was confirmed by PNGase digestion where the high-molecular-mass MHBs species (33 and 36 kDa) are converted into the lower-molecular-mass form (30 kDa), following enzymatic removal of N-glycan (as shown in Fig. 1B).
As shown in Fig. 1A, after 6 days of incubation in the presence of the glucosidase inhibitor DNJ, compared with untreated cells (UNT) the intracellular amounts of full-length glycosylated and unglycosylated LHBs and MHBs proteins are reduced by 56% (±17%; n = 6) and 86% (±8%; n = 6), respectively, in treated cells (DNJ). The reductions in the amounts of LHBs and MHBs were not the result of cell loading, as shown by analysis of actin. Moreover, that actin levels are similar in glucosidase-inhibited and uninhibited cells is also evidence for a selective effect upon LHBs and MHBs. Reductions in the amounts of LHBs and MHBs proteins in glucosidase-inhibited cells were not the result of a loss in cell viability, as determined by a number of assays, including mitochondrial toxicity test, LDH, and detection of other cell biomarkers, as shown Fig. 1C.
The amounts of LHBs and MHBs proteins detected in the culture medium were also significantly less from HepG2 2.2.15 cells in which glucosidase was inhibited, as shown in Fig. 2. Specifically, the LHBs secretion was reduced 67% (±17%; n = 6) and the MHBs secretion was reduced 77% (±8.4%; n = 6). These data are consistent with previous reports that the secretion of HBV DNA and MHBs protein is significantly reduced from cells in which glucosidase has been inhibited (3, 18, 20). Not all secreted polypeptides are affected by glucosidase inhibition. Accumulation and secretion of albumin, as detected by a human albumin-specific monoclonal antibody, is not influenced by glucosidase inhibition (Fig. 2). Since human albumin is not N glycosylated, this result is not surprising but serves as a control for sample loading, viability, and selectivity of the glucosidase inhibitors.
Since AAT, a triply N-glycosylated protein secreted by HepG2 cells (the parental line to HepG2 2.2.15), has been shown to bind calnexin following glycoprocessing in the ER and its mobility in cells following glucosidase inhibition is altered (25, 30, 31), AAT was examined as a control. AAT was resolved by SDS-PAGE following immunoprecipitation from the culture medium of HepG2 2.2.15 cells radiolabeled with [35S]Met, as a function of glucosidase inhibition. The results are shown in Fig. 3. Clearly, the mobility of AAT derived from glucosidase-inhibited cells has been altered in a way characteristic of glycan processing inhibition of the AAT glycan (as in references 25, 30, and 31). This serves as evidence that the ER glucosidases have been inhibited. In addition, compared to control (UNT) cultures, there is a modest 40% reduction in the amount of AAT secreted into the culture medium from glucosidase-inhibited cells (DNJ).
The HBV small envelope protein is not detectably affected by glucosidase inhibition. Previous reports suggest that SHBs secretion is not affected by glucosidase inhibition (18, 20, 22). Since the amounts of LHBs and MHBs protein accumulation were dramatically reduced by glucosidase inhibition (Fig. 1), it was of interest to know the sensitivity of SHBs to glucosidase inhibition in this system. However, antibodies that recognize the SHBs protein in Western blots are not readily available. To detect the SHBs protein, it was necessary to perform an IP using antibodies specific for the "a" epitope of the HBV S domain. Therefore, HepG2 2.2.15 cells were incubated in the absence or presence of 4.52 mmol/liter glucosidase inhibitor, as shown in Fig. 1 and 2. After 6 days, cultures were continued in the absence or presence of NB-DNJ and metabolically labeled with [35S]methionine for 16 h (see Materials and Methods). The amount of intracellular SHBs protein was determined by immunoprecipitation of cell lysates, followed by resolution by SDS gel electrophoresis and autoradiography. The results are shown in Fig. 4A. Clearly, although HBV unglycosylated and glycosylated species of the SHBs were not well resolved, there appear to be no detectable differences in the amounts of resolved SHBs protein in lysates of untreated (Fig. 4A, UNT) and glucosidase-inhibited (Fig. 4A, DNJ) cells. These data are consistent with the results of an ELISA detection of SHBs protein, using culture medium and lysate, prior to labeling (Fig. 4B). The SHBs protein level, therefore, does not appear to be significantly affected by glucosidase inhibition.
Consistent with the work of others, and as shown in Fig. 4, the abundance of MHBs proteins, as detected in HepG2 2.2.15 cells, is less than that of the SHBs proteins (18, 20). Nevertheless, on the basis of molecular mass and immunoprecipitation, as shown in Fig. 4A, left panel, a pair of polypeptides were identified as MHBs (and this was confirmed by a Western blot, Fig. 4A, right panel) and their abundance is, as expected, reduced in the glucosidase-inhibited cells.
However, these experiments were too insensitive to permit conclusions about LHBs protein. Taken together, although conclusions about the influence of glucosidase inhibitors upon LHBs and MHBs proteins do not depend upon the immunoprecipitation results, the results are consistent with the Western blot data on the question of sensitivity of LHBs and MHBs proteins to glucosidase inhibition.
Reductions in the amounts of LHBs or MHBs proteins are independent of other viral proteins. HepG2 2.2.15 cells contain the entire HBV genome (29), which specifies at least five open reading frames (10). The reduction of LHBs and MHBs protein levels in glucosidase-inhibited HepG2 2.215 cells could occur independently of other viral proteins or could require other viral proteins, such as core, polymerase, or the "X" protein. To distinguish between these possibilities, HepG2 cells were transfected to express either the MHBs protein or the LHBs protein in the absence of the HBV proteins. Therefore, HepG2 cells were transfected with either the "MHBs-only" plasmid (pCMV-M) or the "LHBs-only" plasmid (pTRE-L) (23). The transfected HepG2 cells were incubated for 6 days in the absence or presence of glucosidase inhibitor, and the amounts of MHBs and LHBs proteins in the intracellular compartment were detected by Western blotting. Figure 5A and B show that, in cells transfected with either pCMV-M (Fig. 5A) or pTRE-L (Fig. 5B), the amounts of both glycosylated and unglycosylated MHBs and LHBs proteins are dramatically reduced as a function of glucosidase inhibitor. Specifically, MHBs is reduced 93% (±4%; n = 3) and LHBs is reduced 73% (±12%; n = 4). These data are consistent with the conclusion that HBV sensitivity to glucosidase inhibition does not require expression of HBV gene products other than LHBs or MHBs.
Proteasome inhibitors prevent the glucosidase inhibitor-mediated reductions of LHBs and MHBs proteins in HepG2 2.2.15 cells. Inhibition of ER glucosidases interferes with CNX-mediated folding of some, but not all, nascent polypeptides bearing N-linked glycan (24, 28). Interference with CNX-mediated folding would be expected to result in misfolded polypeptides. One pathway for the disposal of misfolded glycoproteins is through the cytosolic proteasome system (1, 15). Therefore, the possibility that proteasomal degradation was responsible for the decrease in detectable HBV LHBs and MHBs was pursued. Briefly, HepG2 2.2.15 cells were either left untreated or treated with glucosidase inhibitor, for 6 days, as before. In some cultures, as indicated in Fig. 6A, LCT or EPO (inhibitors of cytosolic proteasomes) was included. The amounts of LHBs and MHBs proteins, as a function of glucosidase inhibition in the absence or presence of functioning proteasomes, were then determined by Western blots.
As shown in Fig. 6A and seen in previous experiments, the amounts of detectable unglycosylated and glycosylated LHBs and MHBs proteins are significantly smaller in HepG2 2.2.15 cells in which glucosidase is inhibited with NB-DNJ. Parenthetically, similar results have been obtained using castanospermine and DNJ, two other inhibitors of glucosidase, in place of NB-DNJ (data not shown). On the other hand, the reduction of LHBs and MHBs is not seen in glucosidase-inhibited cells in which either LCT or EPO has been included in the culture (compare lane 4 with lanes 5 and 6 of Fig. 6A). That is, degradation of both the unglycosylated and glycosylated species of LHBs and MHBs proteins seen in glucosidase-inhibited cells is apparently prevented if irreversible inhibitors of the proteasomes (lactacystin or epoxomicin) are included in the culture medium.
AAT mobility and levels were examined for comparison to the effects upon HBV proteins. Inhibition of proteasomes caused a modest, but reproducible, increase in the amount of AAT (Fig. 6B) but had no detectable effect upon constitutive levels of LHBs and MHBs proteins, compared to cells in which proteasomes were not inhibited (Fig. 6B). A small amount of constitutively produced AAT would be expected to be misfolded and degraded by proteasomes. Therefore, the slight increase in AAT in cells in which proteasomes are inhibited is expected. The apparent lack of an increase in the amount of LHBs and MHBs in cells in which proteasomes are inhibited was somewhat surprising, since some constitutive amount of misfolding and proteasomal degradation was expected, even in the absence of glucosidase inhibition. This is considered further in the Discussion. It was noted that, although the mobility of AAT derived from glucosidase-inhibited cells was altered, consistent with the expected effect upon glycan processing of AAT, the levels of AAT were not reduced. This, again, is in contrast with the effect of glucosidase inhibition upon LHBs and MHBs. Taken together, these data are consistent with the notion that, in glucosidase-inhibited cells, LHBs and MHBs proteins are degraded and that the proteasomes are the major pathway by which this degradation occurs.
DISCUSSION
This report helps resolve two previously unanswered questions regarding the selective sensitivity of the HBV envelope proteins to glucosidase inhibition. First, the question as to whether or not both glycosylated and unglycosylated HBV envelope proteins are sensitive to glucosidase inhibition has been addressed. Second, we have determined whether or not LHBs is sensitive to the glucosidase inhibitors. Here it is shown that, compared with untreated cells, there is a striking reduction in the amount of unglycosylated as well as glycosylated species of LHBs and MHBs proteins, under conditions where glycan processing is inhibited and other secreted proteins are not significantly affected. This implies that both glycoforms of the LHBs and MHBs proteins are linked and that they are sent for degradation as an oligomer, and maybe even a particle. These results also suggest that the presence of a small amount of misfolded protein can prevent the proper morphogenesis of the particle. Indeed, this would suggest that misfolded LHBs and MHBs can exert a dominant-negative effect on the particle. It is even feasible to say that the ER quality control mechanism passes judgment on the entire misfolded HBV subviral particle and not on each individual HBV glycoprotein. Although it is possible that the compounds utilized could have unknown effects, since three different glucosidase inhibitors representing two distinct chemical families had similar effects (data not shown) and the selective reduction occurred with cells where LHBs protein or MHBs protein was the only viral protein produced, these data are consistent with the conclusion that ER glucosidase inhibition in this system is centrally responsible for the selective reduction in the amounts of LHBs and MHBs proteins.
It is also important to note that in all of these experiments the SHBs was not altered by the glucosidase inhibitors, as determined by ELISA or by immunoprecipitation. This indicates that the effect on LHBs and MHBs is specific and not the result of some general inhibition of translation. It also suggests that either some level of sorting occurs that removes LHBs and MHBs from SHBs containing subviral particles or the LHBs and MHBs primarily interact with each other.
The other significant question answered in this report relates to how LHBs and MHBs proteins were metabolized in glucosidase-inhibited cells. Since two different proteasome inhibitors prevented the reductions of LHBs and MHBs proteins that would otherwise have occurred in glucosidase-inhibited cells, the results strongly suggest that cytosolic proteasomes are responsible for their degradation. Proteasomes are large multiprotein complexes that include a number of different proteases and are thought to be part of the pathway of degradation of misfolded glycoproteins (1, 9). Many misfolded proteins become ubiquitinated, and this serves as the basis by which many misfolded proteins are recognized by the proteasomes (8, 12, 15). We are currently exploring the possibility that LHBs and MHBs proteins, produced in glucosidase inhibited cells, become ubiquitinated prior to their degradation. Based on amino acid sequence, LHBs protein contains only three possible ubiquitination sites and MHBs protein contains only one possible site. Experimental results with antiubiquitin antibodies so far, although preliminary, suggest that LHBs and MHBs proteins do not become detectably ubiquitinated (data not shown), suggesting that nonubiquitination recognition signals might be operating. Nevertheless, the degradation of LHBs and MHBs proteins in glucosidase-inhibited cells seemed to be completely dependent upon proteasomes, since the degradation was quantitatively reversed by inclusion of proteasome inhibitors.
Although the results are modest, one consistent observation worth noting is that, in HepG2 2.2.15 cells incubated with proteasome inhibitors under normal conditions (where there is no glucosidase inhibitor), the amounts of LHBs and MHBs proteins (and SHBs protein) did not noticeably increase. This was somewhat surprising, since even under constitutive conditions of LHBs and MHBs synthesis, some misfolding would be expected to occur, and an amount of proteasomal degradation would be expected. Thus, inhibition of proteasomes would result in a commensurate increase in polypeptide abundance. This was seen with AAT, for example.
One possibility is that the formation of large viral lipoprotein particles immediately after synthesis of LHBs, MHBs, and SHBs proteins renders HBV envelope proteins refractory to retrotranslocation to proteasomes. In this case, since proteasomal degradation is a source of peptides for major histocompatibility complex class I loading (33), the rapid multimer formation of HBsAgs and consequent natural resistance to proteasomal degradation would be a mechanism of immune evasion. The implications are substantial. It is also possible the lack of proteasomal degradation in the absence of the glucosidase inhibitors is the result of the slow turnover of the HBV envelope proteins. However, in the experiments presented in Fig. 1 and 2, proteasome inhibitors were incubated with cells for 16 h. Indeed, we have also performed pulse-chase experiments with chase times of 48 h (in the presence of proteasome inhibitors) with little observed proteasomal degradation (data not shown). Although more work is needed to determine the method by which HBV subviral particles are degraded, the evidence presented here suggests that normally the HBV proteins may be poor substrates for ER-associated proteasomal degradation. These issues are the subjects of active investigation.
ACKNOWLEDGMENTS
Preparation of the manuscript and the work described herein were supported by The Hepatitis B Foundation of America, The Commonwealth of Pennsylvania, and grants from the National Institutes of Health (AI 53884 and AI054763).
Ju-Tao Guo is thanked for careful reading of the manuscript.
REFERENCES
Baumeister, W., J. Walz, F. Zuhl, and E. Seemuller. 1998. The proteasome: paradigm of a self-compartmentalizing protease. Cell 92:367-380.
Bergeron, J. J., M. B. Brenner, D. Y. Thomas, and D. B. Williams. 1994. Calnexin: a membrane-bound chaperone of the endoplasmic reticulum. Trends Biochem. Sci. 19:124-128.
Block, T. M., X. Lu, F. M. Platt, G. R. Foster, W. H. Gerlich, B. S. Blumberg, and R. A. Dwek. 1994. Hepatitis B virus secretion is inhibited by the imino sugar N-butyl-deoxynojirimycin. Proc. Natl. Acad. Sci. USA 91:2235-2239.
Block, T. M., X. Lu, A. S. Mehta, B. S. Blumberg, B. Tennant, M. Ebling, B. Korba, D. M. Lansky, G. S. Jacob, and R. A. Dwek. 1998. Treatment of chronic hepadnavirus infection in a woodchuck animal model with an inhibitor of protein folding and trafficking. Nat. Med. 4:610-614.
Block, T. M., and R. Jordan. 2002. Imino sugars as possible broad spectrum anti hepatitis virus agents: the alkovirs and glucovirs. Antiviral Chem. Chemother. 12:317-326.
Block, T. M., A. S. Mehta, C. J. Fimmel, and R. Jordan. 2003. Molecular viral oncology of hepatocellular carcinoma. Oncogene 22:5093-5107.
Bruss, V., and D. Ganem. 1991. The role of envelope proteins in hepatitis B virus assembly. Proc. Natl. Acad. Sci. USA 88:1059-1063.
De Virgilio, M., H. Weninger, and N. E. Ivessa. 1998. Ubiquitination is required for the retro-translocation of a short-lived luminal endoplasmic reticulum glycoprotein to the cytosol for degradation by the proteasome. J. Biol. Chem. 273:9734-9743.
Foster, A., and C. P. Hill. 2003. Proteasome degradation: enter the substrate. Trends Cell Biol. 13:550-553.
Ganem, D. 1991. Assembly of hepadnaviral virons and subviral particles. Curr. Top. Microbiol. Immunol. 168:61-84.
Gerlich, W. H., and K. H. Heermann. 1991. Functions of hepatitis B virus proteins and virus assembly, p. 121-131. In F. B. W. Hollinger, S. M. Lemon, and H. S. Margolis (ed.), Viral hepatitis and liver disease. Williams and Wilkins, Baltimore, Md.
Glickman, M. H., and A. Ciechanover. 2002. The ubiquitin-proteasome proteolytic pathway: destruction for the sake of construction. Physiol. Rev. 82:373-428.
Heermann, K. H., F. Kruse, M. Seiger, and W. H. Gerlich. 1987. Immunogenicity of S and pre-S domains in hepatitis virions and HBsAg filaments. Intervirology 28:14-25.
Heermann, K. H., and W. H. Gerlich. 1992. Surface proteins of hepatitis B viruses, p. 109-143. In A. MacLachlan (ed.), A molecular biology of hepatitis viruses. CRC, Boca Raton, Fla.
Hirsch, C., D. Blom, and H. L. Ploegh. 2003. A role for N-glycanase in the cytosolic turnover of glycoproteins. EMBO J. 22:1036-1046.
Hoofnagle, J. H., and A. M. Di Bisceglie. 1997. Drug therapy: the treatment of chronic viral hepatitis. N. Engl. J. Med. 336:347-356.
Lu, X., A. Mehta, R. Dwek, T. Butters, and T. Block. 1995. Evidence that N-linked glycosylation is necessary for hepatitis B virus secretion. Virology 213:660-665.
Lu, X., A. Mehta, M. Dadmarz, R. Dwek, B. S. Blumberg, and T. M. Block. 1997. Aberrant trafficking of hepatitis B virus glycoproteins in cells in which N-glycan processing is inhibited. Proc. Natl. Acad. Sci. USA 94:2380-2385.
Lu, X., Y. Lu, R. Geschwindt, R. A. Dwek, and T. M. Block. 2001. Hepatitis B virus MHBs antigen is selectively sensitive to glucosidase-mediated processing in the endoplasmic reticulum. DNA Cell Biol. 20:647-656.
Mehta, A., X. Lu, T. M. Block, B. S. Blumberg, and R. A. Dwek. 1997. Hepatitis B virus (HBV) envelope glycoproteins vary drastically in their sensitivity to glycan processing: evidence that alteration of a single N-linked glycosylation site can regulate HBV secretion. Proc. Natl. Acad. Sci. USA 94:1822-1827.
Mehta, A., P. M. Rudd, T. M. Block, and R. A. Dwek. 1997. A strategy for anti-viral intervention: the use of alpha-glucosidase inhibitors to prevent chaperone-mediated folding of viral envelope glycoproteins. Biochem. Soc Trans. 25:1188-1193.
Mehta, A., N. Zitzmann, P. M. Rudd, T. M. Block, and R. A. Dwek. 1998. Alpha-glucosidase inhibitors as potential broad based anti-viral agents. FEBS Lett. 430:17-22.
Norton, P. A., B. Conyers, Q. Gong, L. F. Steel, T. M. Block, and A. S. Mehta. 2005. Assays for glucosidase inhibitors with potential antiviral activities: secreted alkaline phosphatase as a surrogate marker. J. Virol. Methods 124:167-172.
Prange, R., M. Werr, and M. H. Loffler. 1999. Chaperons involved in hepatitis B virus morphogenesis. Biol. Chem. 380:305-314.
Rabouille, C., and R. G. Spiro. 1992. Nonselective utilization of the endomannosidase pathway for processing glycoproteins by human hepatoma (HepG2) cells. J. Biol. Chem. 267:11573-11578.
Robinson, W. S., D. A. Clayton, and R. L. Greenman. 1974. DNA of a human hepatitis B virus candidate. J. Virol. 14:384-391.
Robinson, W. S. 1990. Hepadnaviridae and their replication, p. 2137-2169. In B. N. Fields, D. M. Knipe et al. (ed.), Fields virology. Raven Press, Ltd., New York, N.Y.
Rudd, P. M., T. Elliott, P. Cresswell, I. A. Wilson, and R. A. Dwek. 2001. Glycosylation and the immune system. Science 291:2370-2375.
Sells, M. A., A. Z. Zelent, M. Shvartsman, and G. Acs. 1998. Replicative intermediates of hepatitis B virus in HepG2 cells that produce infectious virions. J. Virol. 62:2836-2844.
Tan, A., L. van den Broek, S. van Broeckel, H. Ploegh, and J. Bolscher. 1991. Chemical modification of the glucosidase inhibitor 1-deoxynojirimycin. J. Biol. Chem. 22:14504-14510.
Tan, A., L. van den Broek, J. Bolscher, D. J. Vermaas, L. Pastoors, C. van Boeckel, and H. Ploegh. 1994. Introduction of oxygen into the alkyl chain of N-decyl-dNM decreases lipophilicity and results in increased retention of glucose residues on N-linked oligosaccharides. Glycobiology 4:141-149.
Xu, Z., V. Bruss, and T. S. Benedict Yen. 1997. Formation of intracellular particles by hepatitis B virus large surface protein. J. Virol. 71:5487-5494.
York, I. A., and K. L. Rock. 1996. Antigen processing and presentation by the class I major histocompatibility complex. Annu. Rev. Immunol. 4:369-396.(Ender Simsek, Anand Mehta)
Drexel Institute for Biotechnology and Virology Research, Department of Microbiology and Immunology, Drexel University College of Medicine, Doylestown, Pennsylvania
Institute for Hepatitis and Virus Research (The Pennsylvania Commonwealth Institute), Doylestown, Pennsylvania
Department of Biochemistry, University of Oxford, Oxford, United Kingdom
ABSTRACT
The secretion of hepatitis B virus (HBV) large (LHBs) and middle (MHBs) envelope polypeptides from tissue cultures requires proper protein folding and is prevented by inhibitors of the endoplasmic reticulum (ER) glucosidase. Using competitive inhibitors of the ER glucosidase, here it is shown that the amounts of glycosylated and unglycosylated forms of LHBs and MHBs proteins are all greatly reduced in tissue cultures producing HBV envelope glycoproteins. In contrast, the HBV small (SHBs) protein was not affected. The reduction in secretion of LHBs and MHBs proteins appears to be mediated by proteasomal degradation pathways, since it is prevented by either lactacystin or epoxomicin, two inhibitors of proteasomal degradation. Although there is no detectable proteasomal degradation of LHBs and MHBs in cells with functional glucosidase, the implications of the nearly quantitative sensitivity of glycosylated and unglycosylated forms of LHBs and MHBs proteins, with selective sparing of SHBs protein, in cells in which glucosidase is inhibited is surprising, and its implications are discussed.
INTRODUCTION
Hepatitis B virus (HBV) is the human member of the family Hepadnaviridae and worldwide is associated with more than 350 million chronic infections and nearly one million deaths annually (6, 16, 27). The infectious agent is a small, 42-nm, enveloped particle containing an incompletely double-stranded DNA genome of approximately 3.5 kb (26). Although the replication of the viral genome occurs in the cytoplasm and has been well characterized, viral morphogenesis and secretion are less well understood. As with many viruses, production of infectious viral particles is inefficient, and the management of defective or unused viral gene products is not well studied.
The secretion and morphogenesis of HBV require viral envelope glycoproteins. HBV specifies three envelope proteins, called large ("LHBs"), middle ("MHBs"), and small ("SHBs") (7, 13) that are all derived from the same open reading frame and may exist in the viral particle as either unglycosylated or N-glycosylated forms (11, 14). Secretion of HBV enveloped DNA is prevented by inhibitors such as the endoplasmic reticulum (ER) glucosidase, implying a critical role for glycoprocessing in the trafficking and morphogenesis of viral glycoproteins (3-5, 17, 18, 20-22).
Many nascent N-linked glycoproteins depend upon an interaction with the lectin-like chaperon, calnexin (CNX) to fold properly. CNX recognizes monoglucose residues on the oligosaccharide of the nascent glycoprotein, which are formed by the sequential action of the ER glucosidases (2). Why some, but not other, glycoproteins appear to have an obligate requirement for CNX-mediated folding is unclear, but the extreme sensitivity of HBV secretion to glucosidase function was assumed to be due to an obligate requirement of HBV glycoproteins for CNX-mediated protein folding. Indeed, both LHBs and MHBs proteins have been shown to interact with CNX (24, 32), and the secretion of MHBs is prevented by glucosidase inhibitors (18, 20).
However, the role of MHBs protein in mediating virus secretion is controversial, and there is evidence that MHBs is not essential (7). Thus, it was not clear how prevention of only MHBs biogenesis with glucosidase inhibitors could be responsible for the selective reductions of HBV secretion observed in glucosidase-inhibited cells.
In addition, although the amount of MHBs protein secreted into the culture medium from cells in which glucosidase has been inhibited has been shown to be reduced, the mechanism of reduction and fate of these polypeptides have not been clearly determined. There is even less information about the sensitivity of LHBs protein. Indeed, previous work had suggested that, despite being reduced in secretion, MHBs protein actually accumulated in glucosidase-inhibited cells (18, 19). Those conclusions were largely based upon detection of HBs epitopes using an antigen capture (enzyme-linked immunosorbent) assay. The state of intact LHBs and MHBs proteins was not conclusively explored.
It was therefore of interest to more precisely explore the fates of LHBs and MHBs proteins in HBV-producing cells in which ER glucosidase had been inhibited. In this study, Western blotting and immunoprecipitation (IP) analysis have confirmed that LHBs and MHBs, but not SHBs, are highly sensitive to glucosidase inhibitors. The amounts of LHBs and MHBs proteins became greatly reduced, by Western blot analysis, within 6 days of incubation with glucosidase inhibitors. Surprisingly, both the glycosylated and unglycosylated species were reduced. The reduction in the amounts of LHBs and MHBs proteins in glucosidase-inhibited cells was prevented by inhibition of proteasomes. The implications of these findings for normal HBV particle biogenesis and cellular management of misfolded HBV glycoproteins are discussed.
MATERIALS AND METHODS
Cells and compounds. HepG2 cells, a stable tissue culture line derived from a human hepatoblastoma, were purchased from the American Type Culture Collection (Rockville, MD) and grown in RPMI 1640 (Gibco-BRL, Rockville, MD) containing 10% fetal bovine serum (Gibco-BRL). HepG2 2.2.15 cells, derived from the stable transfection of HepG2 cells with a dimer of the HBV genome producing HBV viral and subviral particles at physiologic conditions, were kindly provided by George Acs (Mt. Sinai Medical College, New York, NY) in 1992 and maintained as HepG2 cells but with the addition of 200 μg/ml of G418 (Gibco-BRL) (29). N-Butyl-deoxynojirimycin (NB-DNJ) and deoxynojirimycin (DNJ) are glucosidase inhibitors and were provided by Monsanto Searle (St. Louis, MO) and Synergy Pharmaceuticals, Inc. (Edison, NJ), respectively. Castanospermine and proteasome inhibitors (lactacystin [LCT] and epoxomicin [EPO]) were purchased from Sigma-Aldrich (St. Louis, MO) and Calbiochem, Inc. (La Jolla, CA), respectively.
Use of inhibitors. In all cases, tissue culture was performed using fully confluent HepG2 or HepG2 2.2.15 cells. Glucosidase inhibitors (NB-DNJ or DNJ) were incubated on cells for a total of 7 days. Medium was changed with fresh compound (where indicated) every 2 days. On day 7, culture media and cells were collected. For the inhibition of proteasome degration, the cells were treated with either 20 μM lactacystin or 1 μM epoxomicin in the absence or presence of a glucosidase inhibitor for 16 h after 6 days of glucosidase inhibition.
Cytotoxicity (cell viability) assay. HepG2 2.2.15 cells were treated with NB-DNJ or DNJ (4.52 mmol/liter) as explained above. The viability of cultures treated with the indicated concentration of imino sugar inhibitor was measured by Trypan blue dye exclusion staining on days 0, 3, 5, and 7 during drug treatment. Based on their color (white), viable cells were counted on a hemocytometer by a light microscope. The cytotoxicity of glucosidase inhibitors was also evaluated by the mitochondrial toxicity test and the lactate dehydrogenase (LDH) assay. The assays show that the glucosidase inhibitor has no observed toxicity (data not shown).
Plasmids and transfection. HepG2 cells were transfected with either the pCMV-M expression vector (23) or pTRE-L expression vector with Lipofectamine or FuGene 6 according to the manufacturer's directions. As a control, pCMV-GFP (a cytomegalovirus-green fluorescent protein plasmid) was used to monitor transfection efficiency. After transfection, the cells, allowed 1 day for recovery, were left untreated or were treated with the glucosidase inhibitor DNJ for 6 days. The cells were harvested in cell lysis buffer containing 1x protease inhibitors, and the LHBs and MHBs proteins and SHBs proteins were detected by Western blotting or Abbott enzyme-linked immunosorbent assay (ELISA), respectively.
Detection of LHBs and MHBs surface antigens by Western blotting. HBV large and middle glycoproteins, from either clarified culture medium or lysed cells, were dissolved in loading buffer, resolved by electrophoresis through 12% polyacrylamide sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE), transferred to polyinylidene difluoride (Millipore) membranes, and blocked with 5% powdered milk in 0.1% Tween 20-phosphate-buffered saline (PBS) at room temperature for 1 h. After incubation with polyclonal pre-S2 antibody (Research Diagnostics, Inc., Flanders, NJ) at a 1:1,000 dilution in 0.1% Tween 20 with 2% bovine serum albumin for 2 h at room temperature, the blot was washed three times with 0.1% Tween 20-PBS for 10 min/each wash. The blot was incubated with secondary antibody (peroxidase-conjugated donkey anti-rabbit immunoglobulin G serum) at a 1:5,000 dilution in 0.1% Tween 20 with 2% bovine serum albumin for 1 h at room temperature. Blots were washed three times with 0.1% Tween 20-PBS, and proteins were detected by enhanced chemiluminescence (ECL; Amersham Corporation, Arlington Heights, IL).
Detection of HBsAg by ELISA. Two hundred microliters of media from the same studies was used to determine the level of hepatitis B surface antigen (HBsAg) in the culture medium of untreated (UNT) and treated cells. Analysis was performed using the Abbott Diagnostics AUSZYME Monoclonal Diagnostic kit according to the manufacturer's directions (Abbott Laboratories, North Chicago, IL).
Detection of SHBs glycoprotein and ATT by immunoprecipitation. HepG2 and HepG2 2.2.15 cells were used to detect the secretion of a cellular glycoprotein such as SHBs and alpha 1 anti-trypsin (AAT) after treatment with and removal of the glucosidase inhibitor. Briefly, the cells were treated with 4.52 mmol/liter DNJ for 6 days. After the treatment, the cells were incubated with [35S]methionine (200 μCi/ml) for 16 h in the absence or presence of glucosidase inhibitor. The culture supernatant was aspirated and clarified at 13,000 rpm in a bench top centrifuge to remove cellular debris. The cells were harvested in cell lysis buffer containing 1x protease inhibitors. Either the lysed cells or the clarified medium containing radiolabeled proteins was adjusted to 50 mM Tris-HCl, pH 7.4, 150 mM NaCl, 1 mM EDTA, 0.05% SDS, and 0.2% NP-40 (final concentrations) and used in IP assays. The clarified radiolabeled medium was incubated with either human monoclonal antitrypsin antibody (Sigma-Aldrich) or anti-HBsAg monoclonal antibody (DakoCytomation, Inc., Carpinteria, CA) at 4°C overnight. The immune complexes were then precipitated using protein G-agarose (Roche Diagnostics Corp., Indianapolis, IN). Following washing with lysis buffer, the immune complexes were released from protein G-agarose using SDS-PAGE denaturation buffer and boiled for 5 min. Samples were fractionated by SDS-PAGE (12% polyacrylamide), and the labeled proteins were detected with a Bio-Rad Personal FX phosphorimager.
RESULTS
Steady-state levels of unglycosylated and N-glycosylated LHBs and MHBs proteins as a function of glucosidase inhibition. Previous studies have shown that the secretion of enveloped HBV DNA and MHBs protein from tissue cultures is prevented by inhibition of the ER glucosidase (3-5, 17, 18, 20-22). Based upon ELISA-type analysis, it was also suggested that SHBs epitopes, derived from MHBs, were detained within the glucosidase-inhibited cells and accumulated. It was assumed that MHBs protein was accumulating, although MHBs polypeptide was not specifically identified. Therefore, to confirm the sensitivity of MHBs protein biogenesis to glucosidase inhibitors and to more precisely study the extracellular and intracellular fates of the LHBs and MHBs polypeptides, Western blot analysis was performed on cell lysates and culture medium of HepG2 2.2.15 cells as a function of glucosidase inhibition. Briefly, HepG2 2.2.15 cells were cultured for 6 days in the absence or presence of 4.52 mmol/liter DNJ. DNJ is a competitive inhibitor of ER glucosidases I and II, and 4.52 mmol/liter is sufficient to inhibit at least 95% of the glycan processing activity in these cells (3, 20).
As shown in Fig. 1A (lane UNT), LHBs and MHBs proteins and their glycoforms are well resolved in this gel (Fig. 1A). That the lower-molecular-mass band (faster mobility) of the MHBs-specific bands is the unglycosylated species (30 kDa) was confirmed by PNGase digestion where the high-molecular-mass MHBs species (33 and 36 kDa) are converted into the lower-molecular-mass form (30 kDa), following enzymatic removal of N-glycan (as shown in Fig. 1B).
As shown in Fig. 1A, after 6 days of incubation in the presence of the glucosidase inhibitor DNJ, compared with untreated cells (UNT) the intracellular amounts of full-length glycosylated and unglycosylated LHBs and MHBs proteins are reduced by 56% (±17%; n = 6) and 86% (±8%; n = 6), respectively, in treated cells (DNJ). The reductions in the amounts of LHBs and MHBs were not the result of cell loading, as shown by analysis of actin. Moreover, that actin levels are similar in glucosidase-inhibited and uninhibited cells is also evidence for a selective effect upon LHBs and MHBs. Reductions in the amounts of LHBs and MHBs proteins in glucosidase-inhibited cells were not the result of a loss in cell viability, as determined by a number of assays, including mitochondrial toxicity test, LDH, and detection of other cell biomarkers, as shown Fig. 1C.
The amounts of LHBs and MHBs proteins detected in the culture medium were also significantly less from HepG2 2.2.15 cells in which glucosidase was inhibited, as shown in Fig. 2. Specifically, the LHBs secretion was reduced 67% (±17%; n = 6) and the MHBs secretion was reduced 77% (±8.4%; n = 6). These data are consistent with previous reports that the secretion of HBV DNA and MHBs protein is significantly reduced from cells in which glucosidase has been inhibited (3, 18, 20). Not all secreted polypeptides are affected by glucosidase inhibition. Accumulation and secretion of albumin, as detected by a human albumin-specific monoclonal antibody, is not influenced by glucosidase inhibition (Fig. 2). Since human albumin is not N glycosylated, this result is not surprising but serves as a control for sample loading, viability, and selectivity of the glucosidase inhibitors.
Since AAT, a triply N-glycosylated protein secreted by HepG2 cells (the parental line to HepG2 2.2.15), has been shown to bind calnexin following glycoprocessing in the ER and its mobility in cells following glucosidase inhibition is altered (25, 30, 31), AAT was examined as a control. AAT was resolved by SDS-PAGE following immunoprecipitation from the culture medium of HepG2 2.2.15 cells radiolabeled with [35S]Met, as a function of glucosidase inhibition. The results are shown in Fig. 3. Clearly, the mobility of AAT derived from glucosidase-inhibited cells has been altered in a way characteristic of glycan processing inhibition of the AAT glycan (as in references 25, 30, and 31). This serves as evidence that the ER glucosidases have been inhibited. In addition, compared to control (UNT) cultures, there is a modest 40% reduction in the amount of AAT secreted into the culture medium from glucosidase-inhibited cells (DNJ).
The HBV small envelope protein is not detectably affected by glucosidase inhibition. Previous reports suggest that SHBs secretion is not affected by glucosidase inhibition (18, 20, 22). Since the amounts of LHBs and MHBs protein accumulation were dramatically reduced by glucosidase inhibition (Fig. 1), it was of interest to know the sensitivity of SHBs to glucosidase inhibition in this system. However, antibodies that recognize the SHBs protein in Western blots are not readily available. To detect the SHBs protein, it was necessary to perform an IP using antibodies specific for the "a" epitope of the HBV S domain. Therefore, HepG2 2.2.15 cells were incubated in the absence or presence of 4.52 mmol/liter glucosidase inhibitor, as shown in Fig. 1 and 2. After 6 days, cultures were continued in the absence or presence of NB-DNJ and metabolically labeled with [35S]methionine for 16 h (see Materials and Methods). The amount of intracellular SHBs protein was determined by immunoprecipitation of cell lysates, followed by resolution by SDS gel electrophoresis and autoradiography. The results are shown in Fig. 4A. Clearly, although HBV unglycosylated and glycosylated species of the SHBs were not well resolved, there appear to be no detectable differences in the amounts of resolved SHBs protein in lysates of untreated (Fig. 4A, UNT) and glucosidase-inhibited (Fig. 4A, DNJ) cells. These data are consistent with the results of an ELISA detection of SHBs protein, using culture medium and lysate, prior to labeling (Fig. 4B). The SHBs protein level, therefore, does not appear to be significantly affected by glucosidase inhibition.
Consistent with the work of others, and as shown in Fig. 4, the abundance of MHBs proteins, as detected in HepG2 2.2.15 cells, is less than that of the SHBs proteins (18, 20). Nevertheless, on the basis of molecular mass and immunoprecipitation, as shown in Fig. 4A, left panel, a pair of polypeptides were identified as MHBs (and this was confirmed by a Western blot, Fig. 4A, right panel) and their abundance is, as expected, reduced in the glucosidase-inhibited cells.
However, these experiments were too insensitive to permit conclusions about LHBs protein. Taken together, although conclusions about the influence of glucosidase inhibitors upon LHBs and MHBs proteins do not depend upon the immunoprecipitation results, the results are consistent with the Western blot data on the question of sensitivity of LHBs and MHBs proteins to glucosidase inhibition.
Reductions in the amounts of LHBs or MHBs proteins are independent of other viral proteins. HepG2 2.2.15 cells contain the entire HBV genome (29), which specifies at least five open reading frames (10). The reduction of LHBs and MHBs protein levels in glucosidase-inhibited HepG2 2.215 cells could occur independently of other viral proteins or could require other viral proteins, such as core, polymerase, or the "X" protein. To distinguish between these possibilities, HepG2 cells were transfected to express either the MHBs protein or the LHBs protein in the absence of the HBV proteins. Therefore, HepG2 cells were transfected with either the "MHBs-only" plasmid (pCMV-M) or the "LHBs-only" plasmid (pTRE-L) (23). The transfected HepG2 cells were incubated for 6 days in the absence or presence of glucosidase inhibitor, and the amounts of MHBs and LHBs proteins in the intracellular compartment were detected by Western blotting. Figure 5A and B show that, in cells transfected with either pCMV-M (Fig. 5A) or pTRE-L (Fig. 5B), the amounts of both glycosylated and unglycosylated MHBs and LHBs proteins are dramatically reduced as a function of glucosidase inhibitor. Specifically, MHBs is reduced 93% (±4%; n = 3) and LHBs is reduced 73% (±12%; n = 4). These data are consistent with the conclusion that HBV sensitivity to glucosidase inhibition does not require expression of HBV gene products other than LHBs or MHBs.
Proteasome inhibitors prevent the glucosidase inhibitor-mediated reductions of LHBs and MHBs proteins in HepG2 2.2.15 cells. Inhibition of ER glucosidases interferes with CNX-mediated folding of some, but not all, nascent polypeptides bearing N-linked glycan (24, 28). Interference with CNX-mediated folding would be expected to result in misfolded polypeptides. One pathway for the disposal of misfolded glycoproteins is through the cytosolic proteasome system (1, 15). Therefore, the possibility that proteasomal degradation was responsible for the decrease in detectable HBV LHBs and MHBs was pursued. Briefly, HepG2 2.2.15 cells were either left untreated or treated with glucosidase inhibitor, for 6 days, as before. In some cultures, as indicated in Fig. 6A, LCT or EPO (inhibitors of cytosolic proteasomes) was included. The amounts of LHBs and MHBs proteins, as a function of glucosidase inhibition in the absence or presence of functioning proteasomes, were then determined by Western blots.
As shown in Fig. 6A and seen in previous experiments, the amounts of detectable unglycosylated and glycosylated LHBs and MHBs proteins are significantly smaller in HepG2 2.2.15 cells in which glucosidase is inhibited with NB-DNJ. Parenthetically, similar results have been obtained using castanospermine and DNJ, two other inhibitors of glucosidase, in place of NB-DNJ (data not shown). On the other hand, the reduction of LHBs and MHBs is not seen in glucosidase-inhibited cells in which either LCT or EPO has been included in the culture (compare lane 4 with lanes 5 and 6 of Fig. 6A). That is, degradation of both the unglycosylated and glycosylated species of LHBs and MHBs proteins seen in glucosidase-inhibited cells is apparently prevented if irreversible inhibitors of the proteasomes (lactacystin or epoxomicin) are included in the culture medium.
AAT mobility and levels were examined for comparison to the effects upon HBV proteins. Inhibition of proteasomes caused a modest, but reproducible, increase in the amount of AAT (Fig. 6B) but had no detectable effect upon constitutive levels of LHBs and MHBs proteins, compared to cells in which proteasomes were not inhibited (Fig. 6B). A small amount of constitutively produced AAT would be expected to be misfolded and degraded by proteasomes. Therefore, the slight increase in AAT in cells in which proteasomes are inhibited is expected. The apparent lack of an increase in the amount of LHBs and MHBs in cells in which proteasomes are inhibited was somewhat surprising, since some constitutive amount of misfolding and proteasomal degradation was expected, even in the absence of glucosidase inhibition. This is considered further in the Discussion. It was noted that, although the mobility of AAT derived from glucosidase-inhibited cells was altered, consistent with the expected effect upon glycan processing of AAT, the levels of AAT were not reduced. This, again, is in contrast with the effect of glucosidase inhibition upon LHBs and MHBs. Taken together, these data are consistent with the notion that, in glucosidase-inhibited cells, LHBs and MHBs proteins are degraded and that the proteasomes are the major pathway by which this degradation occurs.
DISCUSSION
This report helps resolve two previously unanswered questions regarding the selective sensitivity of the HBV envelope proteins to glucosidase inhibition. First, the question as to whether or not both glycosylated and unglycosylated HBV envelope proteins are sensitive to glucosidase inhibition has been addressed. Second, we have determined whether or not LHBs is sensitive to the glucosidase inhibitors. Here it is shown that, compared with untreated cells, there is a striking reduction in the amount of unglycosylated as well as glycosylated species of LHBs and MHBs proteins, under conditions where glycan processing is inhibited and other secreted proteins are not significantly affected. This implies that both glycoforms of the LHBs and MHBs proteins are linked and that they are sent for degradation as an oligomer, and maybe even a particle. These results also suggest that the presence of a small amount of misfolded protein can prevent the proper morphogenesis of the particle. Indeed, this would suggest that misfolded LHBs and MHBs can exert a dominant-negative effect on the particle. It is even feasible to say that the ER quality control mechanism passes judgment on the entire misfolded HBV subviral particle and not on each individual HBV glycoprotein. Although it is possible that the compounds utilized could have unknown effects, since three different glucosidase inhibitors representing two distinct chemical families had similar effects (data not shown) and the selective reduction occurred with cells where LHBs protein or MHBs protein was the only viral protein produced, these data are consistent with the conclusion that ER glucosidase inhibition in this system is centrally responsible for the selective reduction in the amounts of LHBs and MHBs proteins.
It is also important to note that in all of these experiments the SHBs was not altered by the glucosidase inhibitors, as determined by ELISA or by immunoprecipitation. This indicates that the effect on LHBs and MHBs is specific and not the result of some general inhibition of translation. It also suggests that either some level of sorting occurs that removes LHBs and MHBs from SHBs containing subviral particles or the LHBs and MHBs primarily interact with each other.
The other significant question answered in this report relates to how LHBs and MHBs proteins were metabolized in glucosidase-inhibited cells. Since two different proteasome inhibitors prevented the reductions of LHBs and MHBs proteins that would otherwise have occurred in glucosidase-inhibited cells, the results strongly suggest that cytosolic proteasomes are responsible for their degradation. Proteasomes are large multiprotein complexes that include a number of different proteases and are thought to be part of the pathway of degradation of misfolded glycoproteins (1, 9). Many misfolded proteins become ubiquitinated, and this serves as the basis by which many misfolded proteins are recognized by the proteasomes (8, 12, 15). We are currently exploring the possibility that LHBs and MHBs proteins, produced in glucosidase inhibited cells, become ubiquitinated prior to their degradation. Based on amino acid sequence, LHBs protein contains only three possible ubiquitination sites and MHBs protein contains only one possible site. Experimental results with antiubiquitin antibodies so far, although preliminary, suggest that LHBs and MHBs proteins do not become detectably ubiquitinated (data not shown), suggesting that nonubiquitination recognition signals might be operating. Nevertheless, the degradation of LHBs and MHBs proteins in glucosidase-inhibited cells seemed to be completely dependent upon proteasomes, since the degradation was quantitatively reversed by inclusion of proteasome inhibitors.
Although the results are modest, one consistent observation worth noting is that, in HepG2 2.2.15 cells incubated with proteasome inhibitors under normal conditions (where there is no glucosidase inhibitor), the amounts of LHBs and MHBs proteins (and SHBs protein) did not noticeably increase. This was somewhat surprising, since even under constitutive conditions of LHBs and MHBs synthesis, some misfolding would be expected to occur, and an amount of proteasomal degradation would be expected. Thus, inhibition of proteasomes would result in a commensurate increase in polypeptide abundance. This was seen with AAT, for example.
One possibility is that the formation of large viral lipoprotein particles immediately after synthesis of LHBs, MHBs, and SHBs proteins renders HBV envelope proteins refractory to retrotranslocation to proteasomes. In this case, since proteasomal degradation is a source of peptides for major histocompatibility complex class I loading (33), the rapid multimer formation of HBsAgs and consequent natural resistance to proteasomal degradation would be a mechanism of immune evasion. The implications are substantial. It is also possible the lack of proteasomal degradation in the absence of the glucosidase inhibitors is the result of the slow turnover of the HBV envelope proteins. However, in the experiments presented in Fig. 1 and 2, proteasome inhibitors were incubated with cells for 16 h. Indeed, we have also performed pulse-chase experiments with chase times of 48 h (in the presence of proteasome inhibitors) with little observed proteasomal degradation (data not shown). Although more work is needed to determine the method by which HBV subviral particles are degraded, the evidence presented here suggests that normally the HBV proteins may be poor substrates for ER-associated proteasomal degradation. These issues are the subjects of active investigation.
ACKNOWLEDGMENTS
Preparation of the manuscript and the work described herein were supported by The Hepatitis B Foundation of America, The Commonwealth of Pennsylvania, and grants from the National Institutes of Health (AI 53884 and AI054763).
Ju-Tao Guo is thanked for careful reading of the manuscript.
REFERENCES
Baumeister, W., J. Walz, F. Zuhl, and E. Seemuller. 1998. The proteasome: paradigm of a self-compartmentalizing protease. Cell 92:367-380.
Bergeron, J. J., M. B. Brenner, D. Y. Thomas, and D. B. Williams. 1994. Calnexin: a membrane-bound chaperone of the endoplasmic reticulum. Trends Biochem. Sci. 19:124-128.
Block, T. M., X. Lu, F. M. Platt, G. R. Foster, W. H. Gerlich, B. S. Blumberg, and R. A. Dwek. 1994. Hepatitis B virus secretion is inhibited by the imino sugar N-butyl-deoxynojirimycin. Proc. Natl. Acad. Sci. USA 91:2235-2239.
Block, T. M., X. Lu, A. S. Mehta, B. S. Blumberg, B. Tennant, M. Ebling, B. Korba, D. M. Lansky, G. S. Jacob, and R. A. Dwek. 1998. Treatment of chronic hepadnavirus infection in a woodchuck animal model with an inhibitor of protein folding and trafficking. Nat. Med. 4:610-614.
Block, T. M., and R. Jordan. 2002. Imino sugars as possible broad spectrum anti hepatitis virus agents: the alkovirs and glucovirs. Antiviral Chem. Chemother. 12:317-326.
Block, T. M., A. S. Mehta, C. J. Fimmel, and R. Jordan. 2003. Molecular viral oncology of hepatocellular carcinoma. Oncogene 22:5093-5107.
Bruss, V., and D. Ganem. 1991. The role of envelope proteins in hepatitis B virus assembly. Proc. Natl. Acad. Sci. USA 88:1059-1063.
De Virgilio, M., H. Weninger, and N. E. Ivessa. 1998. Ubiquitination is required for the retro-translocation of a short-lived luminal endoplasmic reticulum glycoprotein to the cytosol for degradation by the proteasome. J. Biol. Chem. 273:9734-9743.
Foster, A., and C. P. Hill. 2003. Proteasome degradation: enter the substrate. Trends Cell Biol. 13:550-553.
Ganem, D. 1991. Assembly of hepadnaviral virons and subviral particles. Curr. Top. Microbiol. Immunol. 168:61-84.
Gerlich, W. H., and K. H. Heermann. 1991. Functions of hepatitis B virus proteins and virus assembly, p. 121-131. In F. B. W. Hollinger, S. M. Lemon, and H. S. Margolis (ed.), Viral hepatitis and liver disease. Williams and Wilkins, Baltimore, Md.
Glickman, M. H., and A. Ciechanover. 2002. The ubiquitin-proteasome proteolytic pathway: destruction for the sake of construction. Physiol. Rev. 82:373-428.
Heermann, K. H., F. Kruse, M. Seiger, and W. H. Gerlich. 1987. Immunogenicity of S and pre-S domains in hepatitis virions and HBsAg filaments. Intervirology 28:14-25.
Heermann, K. H., and W. H. Gerlich. 1992. Surface proteins of hepatitis B viruses, p. 109-143. In A. MacLachlan (ed.), A molecular biology of hepatitis viruses. CRC, Boca Raton, Fla.
Hirsch, C., D. Blom, and H. L. Ploegh. 2003. A role for N-glycanase in the cytosolic turnover of glycoproteins. EMBO J. 22:1036-1046.
Hoofnagle, J. H., and A. M. Di Bisceglie. 1997. Drug therapy: the treatment of chronic viral hepatitis. N. Engl. J. Med. 336:347-356.
Lu, X., A. Mehta, R. Dwek, T. Butters, and T. Block. 1995. Evidence that N-linked glycosylation is necessary for hepatitis B virus secretion. Virology 213:660-665.
Lu, X., A. Mehta, M. Dadmarz, R. Dwek, B. S. Blumberg, and T. M. Block. 1997. Aberrant trafficking of hepatitis B virus glycoproteins in cells in which N-glycan processing is inhibited. Proc. Natl. Acad. Sci. USA 94:2380-2385.
Lu, X., Y. Lu, R. Geschwindt, R. A. Dwek, and T. M. Block. 2001. Hepatitis B virus MHBs antigen is selectively sensitive to glucosidase-mediated processing in the endoplasmic reticulum. DNA Cell Biol. 20:647-656.
Mehta, A., X. Lu, T. M. Block, B. S. Blumberg, and R. A. Dwek. 1997. Hepatitis B virus (HBV) envelope glycoproteins vary drastically in their sensitivity to glycan processing: evidence that alteration of a single N-linked glycosylation site can regulate HBV secretion. Proc. Natl. Acad. Sci. USA 94:1822-1827.
Mehta, A., P. M. Rudd, T. M. Block, and R. A. Dwek. 1997. A strategy for anti-viral intervention: the use of alpha-glucosidase inhibitors to prevent chaperone-mediated folding of viral envelope glycoproteins. Biochem. Soc Trans. 25:1188-1193.
Mehta, A., N. Zitzmann, P. M. Rudd, T. M. Block, and R. A. Dwek. 1998. Alpha-glucosidase inhibitors as potential broad based anti-viral agents. FEBS Lett. 430:17-22.
Norton, P. A., B. Conyers, Q. Gong, L. F. Steel, T. M. Block, and A. S. Mehta. 2005. Assays for glucosidase inhibitors with potential antiviral activities: secreted alkaline phosphatase as a surrogate marker. J. Virol. Methods 124:167-172.
Prange, R., M. Werr, and M. H. Loffler. 1999. Chaperons involved in hepatitis B virus morphogenesis. Biol. Chem. 380:305-314.
Rabouille, C., and R. G. Spiro. 1992. Nonselective utilization of the endomannosidase pathway for processing glycoproteins by human hepatoma (HepG2) cells. J. Biol. Chem. 267:11573-11578.
Robinson, W. S., D. A. Clayton, and R. L. Greenman. 1974. DNA of a human hepatitis B virus candidate. J. Virol. 14:384-391.
Robinson, W. S. 1990. Hepadnaviridae and their replication, p. 2137-2169. In B. N. Fields, D. M. Knipe et al. (ed.), Fields virology. Raven Press, Ltd., New York, N.Y.
Rudd, P. M., T. Elliott, P. Cresswell, I. A. Wilson, and R. A. Dwek. 2001. Glycosylation and the immune system. Science 291:2370-2375.
Sells, M. A., A. Z. Zelent, M. Shvartsman, and G. Acs. 1998. Replicative intermediates of hepatitis B virus in HepG2 cells that produce infectious virions. J. Virol. 62:2836-2844.
Tan, A., L. van den Broek, S. van Broeckel, H. Ploegh, and J. Bolscher. 1991. Chemical modification of the glucosidase inhibitor 1-deoxynojirimycin. J. Biol. Chem. 22:14504-14510.
Tan, A., L. van den Broek, J. Bolscher, D. J. Vermaas, L. Pastoors, C. van Boeckel, and H. Ploegh. 1994. Introduction of oxygen into the alkyl chain of N-decyl-dNM decreases lipophilicity and results in increased retention of glucose residues on N-linked oligosaccharides. Glycobiology 4:141-149.
Xu, Z., V. Bruss, and T. S. Benedict Yen. 1997. Formation of intracellular particles by hepatitis B virus large surface protein. J. Virol. 71:5487-5494.
York, I. A., and K. L. Rock. 1996. Antigen processing and presentation by the class I major histocompatibility complex. Annu. Rev. Immunol. 4:369-396.(Ender Simsek, Anand Mehta)