| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
LETTERS TO THE EDITOR |
Virus Host Interactions Unit, Center for Vaccinology, Department of Clinical Biology, Microbiology, and Immunology, Ghent University Hospital, Ghent, Belgium
In an article published September 1, 2005, Cooper et al. (1) demonstrated that HBV nucleocapsids bind to PMA-treated THP-1 cells, an interaction that resulted in the secretion of cytokines. A critical role for the C-terminal arginine-rich domains of nucleocapsids in both processes was suggested. It was claimed that capsids only interacted with surface-expressed heparan sulfates (HS) and not chondroitin sulfates (CS) because binding of capsids and induction of cytokines was not inhibited by coincubations with 15 µg/ml CS-A, CS-B, or CS-C. Additionally, binding to Chinese hamster ovary cells, which only express CS, was not observed. However, we previously showed that the addition of 50 µg/ml CS-B strongly reduced binding of nucleocapsids to different cell lines (2). As shown in Fig. 1A, binding of nucleocapsids to ELISA plates coated with HS and CS-B and, to a minor degree, to heparin (H) was detected, thus providing clear proof for an interaction between CS-B and nucleocapsids (the low signal obtained with H is probably due to a low coating efficiency because of the very high negative charge of H).
|
B-luciferase reporter plasmid responded toward stimulation with nucleocapsids. We previously showed that nucleocapsids purified from Escherichia coli (HBcAg-c1) and Saccharomyces cerevisiae (HBcAg-y) bind to surface-expressed heparan sulfates (2). However, these preparations differed in their stimulatory capacity due to differences in levels of contaminating LPS and TLR2 stimulating ligands (3). The latter was concluded because HEK293T cells transfected with CD14, TLR2, and an NF-B promoter-based luciferase reporter only responded to HBcAg-c1. The use of up to 5 µg/ml HBcAg-y did not induce luciferase activity. Cells transfected with TLR3 and an IFN-luciferase reporter plasmid were unresponsive toward both preparations (Fig. 1B). Cooper et al. (1) performed several experiments to exclude that induction of cytokines was due to contaminating TLR4 and TLR2 ligands. The use of boiled E. coli-purified nucleocapsids ruled out that even the very low levels of LPS were responsible. Mammalian nucleocapsids were used to exclude that contaminating lipopeptides in their E. coli-derived nucleocapsids were inducing cytokines. However, such an approach requires absolute certitude about the lack of contamination with TLR-activating ligands that may occur during production (were cells checked for mycoplasma?) or purification.
Taken together, important differences exist in stimulatory activities of different nucleocapsid preparations. It is clear that well controlled studies are required to demonstrate whether this activity is really a genuine characteristic of nucleocapsids.
References
* Department of Molecular Genetics
Department of Immunology, Weizmann Institute of Science, Rehovot, Israel
In our article published on September 1 (R1 ), we demonstrate that the arginine-rich region of HBcAg specifically interacts with membrane heparan sulfate to mediate capsid attachment to cells and that this interaction plays an important role in cytokine induction in human macrophages. We further demonstrate that the cytokine response involves TLR2-mediated NF-B activation.
Vanlandschoot and Leroux-Roels suggest that in addition to binding to heparan sulfate, HBc capsids also bind to chondroitin sulfate (CS)-B. The suggestion is based on their observations that 50 µg/ml soluble CS-B reduces capsid association with cells (R2 ) and that the capsids bind plates coated with CS-B.
Because all glycosaminoglycans (GAG) are highly negatively charged, conclusions based on soluble GAG as competitive antagonists and on in vitro binding assays should be considered cautiously. Therefore, experiments with cells in which the expression of the different GAG was genetically or biochemically modified are required to provide direct assessment of the specific GAG serving as the capsid receptor in a biological context. In our article, we demonstrate that mutant Chinese hamster ovary cells that lack heparan sulfate but express excessive amounts of CS completely fail to bind the capsids (R1 ). Furthermore, we found that enzymatic removal of membrane heparan sulfate by heparinase I strongly diminishes capsid attachment to COS-1 cells while removal of CS by chondroitinase ABC has no effect (our unpublished observation). Taken together, our results clearly demonstrate that under physiological conditions heparan sulfate, but not CS-B, functions as the cell surface receptor for the capsids. Notably, data presented by Vanlandschoot and Leroux-Roels reveal that binding of the capsids to heparan sulfate in vitro is stronger compared with CS-B and that the difference is particularly pronounced at concentrations <50 µg/ml. These data are consistent with our finding that at a concentration of 15 µg/ml, soluble heparan sulfate strongly inhibits capsid binding to cells while CS-B has only a mild effect (R1 ).
The second concern raised by Vanlandschoot and Leroux-Roels relates to the possibility that contamination with LPS and other TLR-activating ligands may be involved in the responses provoked by HBc capsids. In our study, we show that Escherichia coli-derived capsids induce cytokine synthesis in THP-1 cells (R1 ). The capsid preparation used in our study contained very minute amounts of LPS (5.6 pg of LPS/µg protein), and the response was abolished following boiling of the capsids, ruling out the possibility that LPS is involved in the immunostimulation. To exclude further the possibility that TLR-activating ligands other than LPS (e.g., lipopeptides) are involved in the response, we used capsids produced in a human cell line (mycoplasma-free cells). We show that these capsids provoke cytokine responses in THP-1 macrophages while no response is observed with an extract from human cells not expressing the capsids. Collectively, our results provide strong evidence that the immunostimulatory activity is an intrinsic, genuine property of the capsids.
In addition, we show that both E. coli-derived and human cell line-derived capsids activate NF-B in cells expressing TLR2 but not in cells expressing TLR4 or TLR9. In the experiment presented by Vanlandschoot and Leroux-Roels, E. coli-derived capsids, but not capsids produced in Saccharomyces cerevisiae, activate NF-B in cells expressing TLR2. Data presented in a previous study by Vanlandschoot and Leroux-Roels demonstrate cell binding by E. coli-derived capsids is substantially stronger than by S. cerevisiae-derived capsids (R2 ). Therefore, it is possible that the diminished immunostimulatory activity of S. cerevisiae-derived capsids is due to their compromised binding to cells.
References
This article has been cited by other articles:
|
|
 |
|
  P. Vanlandschoot, F. Van Houtte, B. Serruys, and G. Leroux-Roels Contamination of a Recombinant Hepatitis B Virus Nucleocapsid Preparation with a Human B-Cell Activator J. Virol., March 1, 2007; 81(5): 2535 - 2536. [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
), heparan sulfate (
), chondroitin sulfate B (•), or PBS only. The wells were washed with PBS and blocked with 0.5% BSA in PBS, followed by washing three times (0.05% Triton X-100). Nucleocapsids (1 µg/ml) dissolved in 0.5% BSA in PBS were added. A biotinylated HBcAg-specific mAb (1 µg/ml) was added, and the plates were incubated for 1.5 h at room temperature. The mAb was detected with streptavidin labeled with peroxidase. After three washes, 3,3',5,5'-tetramethylbenzidine (Sigma-Aldrich) was added, and 30 min later, the reaction was stopped with 1 N H2SO4. B, Only E. coli-derived nucleocapsids activate NF-