key: cord-0832127-dwzz9vvf
authors: nan
title: Transferrin receptor mediates uptake and presentation of hepatitis B envelope antigen by T lymphocytes
date: 1992-05-01
journal: J Exp Med
DOI: nan
sha: cf6a16717c4b118d2e8bca72fe7dd28c2f2c2cdb
doc_id: 832127
cord_uid: dwzz9vvf

Human activated T lymphocytes expressing class II molecules are able to present only complex antigens that bind to their own surface receptors, and thus can be captured, internalized, and processed through the class II major histocompatibility complex processing pathway. We have used the antigen-presenting T cell system to identify the viral receptor used by hepatitis B virus (HBV) to enter cells, as well as the sequence of HB envelope antigen (HBenvAg) involved in this interaction. Results show that both CD4+ and CD8+ T clones can process and present HBenvAg to class II-restricted cytotoxic T lymphocytes and that the CD71 transferrin receptor (TfR) is involved in efficient HBenvAg uptake by T cells. Moreover, we provide evidence that the HBenvAg sequence interacting with the T cell surface is contained within the pre-S2 region. Since TfR is also expressed on hepatocytes, it might represent a portal of cellular entry for HBV infection. This system of antigen presentation by T cells may serve as a model to study both lymphocyte receptors used by lymphocytotropic viruses and viral proteins critical to bind them.

C D4 T lymphocytes recognize exogenous antigens as peptides bound to class II MHC molecules expressed on the surface of APC, generally represented by macrophages, dendritic cells, and B cells (1, 2) . APC internalize these antigens, process them into endosomal compartments, and then expresson their membrane short peptides in association with class II antigens (3) .

Unlike mouse T cells, human activated T lymphocytes express class II molecules and have been shown to efficiently present only denatured antigens or peptides not requiring processing (4) (5) (6) (7) . In contrast, they usually fail to present complex antigens that need processing unless they capture antigens by their membrane molecules (8) , as in the case of HIV gp120, which binds monovalently to the CD4 receptor (9, 10) . In this manner, T cells can efficiently internalize antigens, process them through a class II processing pathway, and present their fragments to class II-restricted T lymphocytes, as specific B cells do via their Ig receptors (11) (12) (13) (14) .

In this study, we demonstrate that both CD4 + and CD8 + T cell clones can process and efficiently present hep-atitis B envelope antigen (HBenvAg) t to class II-restricted CTL, which in turn kill the presenting T cells, and that the CD71 transferrin receptor (Tflk) is involved in the uptake of HBenvAg by T cells. Moreover, we have identified the HBenvAg region involved in this interaction.

Reagents. Plasma-derived (native) HBenvAg (Sorin, Saluggia, Italy), containing all three (pre-S1, pre-S2, S) hepatitis B virus (HBV) envelope proteins, was purified from patients with chronic HBV infection as described (15) . This antigen preparation contains high levels of pre-S1, pre-S2, and S reactivity as shown by solid phase ELISA (15) . Recombinant (r) particles containing the entire surface protein ofHBV (pre-S1, pre-S2, S) expressed in yeast (16) were a gift from P. J. Kniskern (Merck Sharp and Dohme Research Laboratories, West Point, PA). A deleted rHBenvAg form containing only the 12-52 sequence of the pre-S1 domain, the 133-145 sequence of the pre-S2 domain, and the entire S domain (S,L antigen [17] ), a recombinant protein displaying antigenic determinants en-A part of this work has appeared in abstract form (]. Cell. Biochera. 15 coded by pre-S2 and S regions (pre-S2, S), and a recombinant protein containing only the S region (S), all expressed in yeast (17) , were kindly donated by P. Hanser andJ. Petre (SmithKline Biologicals, Rixensart, Belgium). Another r(pre-S2, S) protein expressed in CHO cells (18) was provided by P. Adamowicz (Pasteur Vaccines, Marnes-La Coquette, France). Purified native tetanus toxoid (TT) was obtained from Institut Merieux (Marcy-12Etoile, France). The synthetic peptide analogues of pre-S1, pre-S2, and S regions were purchased from American Peptide Company, Inc. (Santa Clara, CA). Soluble Tf was purchased from Calbiochem-Behring Corp. (San Diego, CA).

OKTll (anti-CD2), OKT3 (anti-CD3), OKT4A (anti-CD4), OKT8 (anti-CDS), OKT1 (anti-CD5), OKT26a (anti-CD25), OKT10 (anti-CD38), and OKT9 (anti-CD71) mAbs were purchased from Ortho Pharmaceutical (Raritan, NJ). Anti-Leu-M5 (anti-CD18/LFA-1 fl chain) and anti-Leu-19 (anti-CD56) mAbs were purchased from Becton Dickinson & Co. (Mountain View, CA); anti-CD54 (anti-ICAM-1) mAb was from Immunotech (Marseille, France); anti-CD58 (anti-LFA-3; TS 2/9 clone) was obtained from T. A. Springer (19, 20) . The W6/32 mAb to monomorphic determinants of class I Ag was purchased from Cappel Laboratories (Malvern, PA). The Q5/13 mAb to framework determinants of HLA-DR was kindly provided by P. Giacomini (Regina Elena Cancer Institute, Rome, Italy). Anti-HLA-DQ and anti-HLA-DP were purchased from Becton Dickinson & Co. The purchased anti-MHC molecule mAbs were used after dialysis.

Antigen-specific T Cell Clones. Antigen-specific T cell clones were derived from PBL of two HB vaccine (HEVAC-B; Pasteur Institut, Paris, France) recipients, whose HLA-DR typings were DR5 and DR2,3, respectively. Briefly, PBL were plated (10 s per well) in 96well flat-bottomed plates (Falcon Labware, Oxnard, CA) in the presence of 10 #g/ml HBenvAg, and 20 U/ml rib2 (Hoffmann-La Roche, Basel, Switzerland) were added after 5 d. After an additional 5-6 d, growing cultures were expanded with IL-2, and a 15-d cycle of restimulation with antigen plus autologous irradiated (i)PBL was used as APC. Cultures chosen for their capacity to proliferate in response to HBenvAg in the presence of autologous iAPC were cloned by limiting dilution at 0.3 cells per well with 1/~g/ml PHA (Wellcome, Dartford, UK), rib2, and allogeneic iAPC, as described (14, 15) . Table I shows antigen specificity and HLA restriction of the selected T clones. Random T clones, obtained by cloning autologous or aUogeneic PBL with PHA and HLA-incompatible feeder cells, or EBV-transformed B (EBV-B) cells were used as APC.

Cytotoxicity Assay. S~Cr-labeled EBV-B cells or T clones (target cells) were pulsed at 37~ (4 h) with increasing concentrations of entire HBenvAg (native or recombinant) or synthetic peptides. In specific cytotoxicity inhibition experiments, S~Cr-labeled target cells were incubated with 2/~g/ml of various mAbs for 1 h at 37~ before or after pulsing with either antigen or peptide. In some experiments, labeled target cells were treated with increasing concentrations of Tf for 2 h at 37~ washed, and pulsed with HBenvAg or TT for another 2 h at 37~ Viability of target cells before and after incubations with mAbs or Tf was 100%.

Target cells were washed, and cultured (5 x 103) for 4 h with antigen-specific T clones, at a 10:1 E/T ratio into 96-weU roundbottomed plates in triplicate. Plates were then centrifuged, cell-free supernatants were collected from each well, and SlCr release was evaluated in a gamma counter. Percent specific lysis is expressed as the mean of triplicate determinations and is calculated as follows: 100x (experimental release -spontaneous release)/(maximal release -spontaneous release).

uSI-Tf uSI-HBenvAg, and Competitive Binding Assays. Purified human Tf (>99%) was conjugated with lzsI by the solid-phase lactoperoxidase system (New England Nuclear, Boston, MA). The binding reactions were performed in 12 x 75-mm polypropylene tubes in RPMI 1640 containing 0.2% BSA (Fraction V; Sigma Chemical Co., St. Louis, MO). Cell concentration was 5 x 106 cells/ml; labeled Tf, 500 ng/ml; and unlabeled Tf at different concentrations (0-500 #g/ml). In some experiments, unlabeled Tf was replaced by different molecular constructs of HBenvAg added at a final concentration ranging from 0 to 200 #g/ml. The cells were incubated for 120 min at 4~ and unbound ligand was removed by passaging cells through a density cushion, as described (21) . After incubation, 200-/zl aliquots of the cell suspension, layered over 150 #1 of a mixture of dibutyl phthalate (Merck, Darmstadt, Germany) up to a final density of 1.025 in 400-#1 plastic microfuge tubes, were centrifuged in a microfuge (Jouan Inc., Winchester, VA) (10,000g, 2 rain). The resulting supernatant and most of the phthalate oil cushion were aspirated. The tips of the vials containing the cell pellet were severed with a scalpel and transferred into plastic vials, and the radioactivity was measured in a gamma counter. '~I-HBenvAg binding was performed under similar experimental conditions, except for labeled ligand concentration (1.25/~g/ml). After incubation, cells were processed for '~I-Tf binding assay.

HBenvAg. Fig. 1 shows that random CD4 + and CD8 + T clones, used as APC, were able to present HBenvAg to class II-restricted specific cytotoxic T clones. The specific CTL used in these experiments were a DRS-restricted T clone (B41) and two DR2-restricted T clones (VB26 and VB27) recognizing the synthetic 1-21 peptide (MGGWSSKPRKGMGTNLSVPNP) of the pre-S1 region, the 120-130 (MQWNSTAFHQT) of pre-S2 region, and the 193-207 (FFLLTRILTIPQSLD) of S region, respectively ( Table 2 ). Lysis of presenting T cells was observed with a significantly lower concentration of antigen than that needed to induce a comparable level of EBV-B killing ( Fig. 1 ). T cell clones, selected for presenting HBenvAg, were isolated and stimulated with PHA using aUogeneic iPBL as feeders, which lacked the appropriate class II molecules, to exclude the presence of HLA-restricted conventional APC. These CD4 + and CD8 + clones were cytotoxic in a lectin-dependent cellmediated cytotoxicity (LDCC) assay (data not shown). In some experiments, the HBenvAg-specific CD4 + T clone, used as source of APC, was able to present HBenvAg to itself (data not shown). In this case, to exclude the presence of autologous conventional APC, the T clone was restimulated with PHA using allogeneic iPBL as feeders as described above. HBenvAg presentation by T cells required processing since cold pulsing resulted in a drastic inhibition of their APC capacity (Table 3 ). This ability was restored after at least 60 rain of incubation at 37~ except for T cells incubated in the presence of the lysosomotropic agent chloroquine (100 #M) (Sigma Chemical Co.), which inhibits the class II endosomal processing pathway (22) (Table 3 ). In contrast, recognition of 1-21 peptide, which presumably directly binds class II molecules without processing (23), was not affected by either cold or chloroquine treatments.

To define the putative receptor used by HBenvAg to interact with T cells, we attempted to inhibit specific lysis of presenting T lymphocytes by treatment with various randomly selected mAbs to lymphocyte surface molecules. Target cells were incubated with each mAb for 1 h at 37~ either before or after antigen pulsing. We assumed that the blocking of the HBenvAg uptake by T lymphocytes should occur only when the mAb to the presumed HBenvAg receptor was added before antigen pulsing. The B41 T clone was used as effector.

Results show that only the anti-CD71/TfR mAb (OKTg) was able to inhibit HBenvAg capture by T cells: indeed, only this mAb blocked specific lysis when added before, but not after, antigen pulsing (Table 4) . Moreover, OKT9 also blocked, to a lesser degree, the specific lysis of EBV-B cells when added before HBenvAg pulsing (Table 5 ). In contrast, anti-TfR mAb did not inhibit either presentation of 1-21 peptide by T or EBV-B cells (Table 4 and 5), or the presentation of an unrelated antigen, the TT, by EBV-B cells to a TT-specific T line (Table 5 ).

An inhibitory effect was also obtained with adhesion molecule mAbs (anti-CD2, anti-ICAM-1, anti-LFA-1, anti-LFA-3) (24) (Tables 4 and 5). These mAbs showed an inhibitory effect both before and after HBenvAg pulsing of APC. They also EBV-B APC slCr-labeled autologous target EBV-B cells pulsed with 50 #g/ml antigen or 10/~g/ml peptide at 37~ (4 h) were washed and then incubated with cloned T cells (4 h) at an E/T ratio of 10:1 for the CTL assay. Each peptide pool consisted of a mixture of single peptides (5 #g/ml) of pre-S1, pre-S2, and S proteins, respectively. The pre-S2/S peptide included the 172-175 sequence of pre-S2 protein and the 176-184 of S protein. Percent specific lysis is expressed as the mean of triplicate determinations.

inhibited presentation of the 1-21 peptide and the unrelated TT antigen.

To further clarify the role of TfK in HBenvAg uptake by T cells, we tested the effect of soluble Tf on specific lysis of presenting T and EBV-B cells. Incubation of APC with Tf for 2 h before antigen pulsing resulted in a drastic inhibition of their specific lysis. In control experiments, Tf did not block specific lysis of TT-pulsed EBV-B cells by the TT-specific T line (Fig. 2) .

Is Contained in the Pre-S2 Region. To identify the viral envelope sequence relevant for virus/T cell interaction, we tested the ability of both T and EBV-B ceils to present different HBV envelope antigen constructs. The VB27 T clone recognizing the 193-207 of S region, present in all tested proteins, was used as effector. EBV-B cells presented all HBenvAg preparations tested with the same efficiency. T cells presented the entire recombinant or native (pre-S1, pre-S2, S) and the r(pre-S2, S) proteins with high efficiency (Fig. 3) ; they presented, to a lesser degree, the r(12-21 pre-S1, 133-145 pre-S2, S) S,L protein, and, not at all, the r(S) protein. These experiments were performed several times with equivalent results.

CD4* CTL-mediated Lysis Does Not Induce Nonspecific Bystander Lysis. In this study, the antigen-specific CD4 + T clones killed T cells presenting HBenvAg. To examine whether they killed by a contact-dependent mechanism and not aspecifically by production of soluble factors (25) , bystander lysis experiments were performed.

When unlabeled (cold) HBenvAg-pulsed target T cells were mixed with SlCr-labeled (hot) unpulsed T cells, class II-restricted CTL did not lyse the bystander targets (Table 6 ). In contrast, when pulsed T cells were added as hot targets to cold unpulsed T cells, strong lysis of the former was observed, thus suggesting that they killed by direct T cell/T cell contact and did not induce bystander lysis.

HBenvAg. Experiments were performed to evaluate the possible presence of high affinity membrane receptors for HBenvAg on T cells. Thus, T cells were incubated in the presence of a fixed amount of 12SI-HBenvAg (1.25/~g/ml S, L construct) and increasing amounts of cold HBenvAg (0-200 /zg/ml). Scatchard plot of binding data shows that T cells possess high affinity receptors for HBenvAg (KD 1.3 x 10-7 M) (Fig. 4) . The binding of lZSIHBenvS,L Ag to T cells was inhibited by an excess concentration of some cold HBenv con- structs, such as native HBenvAg, or S,L, or r(pre-S2, S), but not by the form containing only the S protein (data not shown). Furthermore, using mI-HBenvAg, the level of nonspecific binding (i.e., the binding observed when the cells were incubated with labeled antigen and a large excess of cold antigen) is high and corresponds to 40-50% of the total binding. Inhibitor (soluble Tf ~g ml -t) The Binding of HBenvAg to Cells Is Inhibited by Some mAbs to TfR. Experiments were performed to evaluate whether 125I-HBenvAg is able to interact with TfK. Cells were incubated with a panel of mAbs to human TfRs (42/6, B3/25, T9, T56/14, and 43/31), washed to remove excess unbound mAb, and then incubated for 120 min at 4~ in the presence of nSI-HBenvAg. These experiments showed that: (a) on T cells, 43/31 anti-CD71 mAb produced a strong inhibition of nSI-HBenvAg binding, whereas the other anti-CD71 mAbs induced only a slight inhibitory effect; and (b) on EBV-B cells, 43/31 mAb elicited a marked inhibition of HBenvAg binding, whereas only a moderate decrease of HBenvAg binding was observed with the remaining anti-CD71 mAbs (Fig. 5) . Although OKT9 anti-TfR mAb had little effect in binding inhibition experiments performed at 4~ it downregulated surface TfKs when incubated at 37~ with a subsequent decrease of both t~SI-Tf and nSI-HBenvAg binding (after a 60-rain preincubation of T cells with OKT9, the binding of Tf and HBenvAg decreased by 60 and 50%, respectively).

The Binding of Tf to Its Receptor Is Partially Inhibited by HBenvAg. Competitive binding experiments were carried out to directly evaluate the ability of HBenvAg to interact with TfR. Recombinant HBenvAg S,L construct competed with radiolabeled Tf for TfR binding sites on both T and EBV-B cells, though to a significantly lesser extent than Tf (Fig. 5) . Interestingly, rHBenvAg S,L construct exhibited a greater capacity to inhibit the binding of labeled Tf on T lymphocytes as compared with EBV-B ceUs (0.25 and 2.5 mg/ml of HBenvAg S,L construct was required to inhibit 50% of labeled Tf binding on T and EBV-B cells, respectively). This finding may explain why T ceils present HBenvAg more efficiently than EBV-B lymphocytes.

Additional studies using different HBenvAg constructs provided results in line with those obtained on the capacity of T cells to present different forms of HBenvAg. Indeed, entire native HBenvAg and r(pre-S2, S) protein were able to inhibit Tf binding to TfR to a higher degree than rHBenvAg S,L; the S protein alone had no significant effect (Fig. 4 ).

Human activated T lymphocytes express class II MHC molecules, but they usually fail to function as APC in presenting soluble antigens. Since T cells are able to present denatured antigens or peptides (4-7), it has been hypothesized that the defect was at the level of antigen processing, or alternatively at the level of antigen internalization due to the inability of T cells to take up antigens aspecifically (27) . This last possibility has recently been supported by studies indicating that these cells are unable to present conventional antigens but internalize, process, and present antigens that bind to their membrane receptors with high efficiency (27) . These studies show that T cells need receptor-mediated interactions with antigens to effectively capture and present them.

In light of these results, we evaluated whether the system of antigen presentation by T cells could be used to identify receptors used by lymphocytotropic viruses to enter T cells (28) . In particular, the possibility was considered that HBV, demonstrated to infect T lymphocytes (28, 29) , is captured by T cells via a receptor-mediated mechanism to be subsequently processed and presented to specific T lymphocytes.

Our study demonstrates that both CD4 + and CD8 + T clones can efficiently present HBenvAg only after processing in endosomal compartments, because presentation is inhibited by chloroquine. The finding that HBenvAg presentation was drastically inhibited when presenting T cells were cold-pulsed argues against the possibility that denatured HBenvAg fragments directly bind MHC molecules without processing and indicates that both internalization and processing are required. These results suggest that HBenvAg uses a surface receptor to enter T cells. Our investigations support the hypothesis that TfR represents one of the membrane molecules, which play a fundamental role in HBenvAg internalization and presentation. Indeed, only the anti-TfR/CD71 (OKTg), from a large series of randomly selected mAbs, was able to inhibit killing of both CD4 + and CD8 § T cells by specific class II-restricted CTL when added before and not after antigen pulsing. This suggests that OKT9 was the only mAb blocking the HBenvAg capture by T cells. OKT9 also inhibited antigen uptake by EBV-B cells, but to a lesser degree than that observed with T cells, thereby suggesting that HBenvAg may in part use the CD71 receptor to enter B lymphocytes. However, B lymphocytes may also capture the HBenvAg either aspecificaUy (27) or through interaction with other membrane structures. The anti-TfR mAb selectively interfered with HBenvAg uptake by APC, since it did not inhibit the presentation of either the 1-21 peptide (which does not need processing) or the unrelated TT antigen. Conversely, the aspecific interactions between APC and T cells, rather than the initial HBenvAg uptake, were inhibited by adhesion molecule mAbs (24) . They indiscriminately inhibited specific lysis when added before or after antigen pulsing of APC. These mAbs also blocked presentation of both the 1-21 peptide and unrelated TT antigen.

The selective blocking of specific lysis, observed by pulsing the presenting T or EBV-B cells with HBenvAg after treatment with soluble Tf, further suggests that TfK/HBenvAg interaction is a crucial, initial step required for antigen processing and presentation. Endocytosis and recycling of the TfK with either its ligand or anti-TfK mAbs have been extensively analyzed in both lymphoid and nonlymphoid cells (30) (31) (32) (33) . In agreement with these findings, we found that OKT9 does not inhibit Tfbinding to its receptors, but downregulates surface Till on T cells when incubated at 37~ This is followed by a decrease of iodinated Tf and HBenvAg bindings, confirming that this mAb blocked HBenvAg presentation by interfering with its internalization (data not shown).

The involvement of TfK in HBenvAg interaction with lyre-phocytes is also supported by binding experiments. In this regard, HBenvAg only partially inhibits Tf binding. Furthermore, among the anti-CD71 mAbs inhibiting the binding of Tf to its receptor, only one (43/31 mAb) markedly decreases iodinated HBenvAg binding to the cells. This clearly suggests that the epitope involved in the interaction between HBenvAg and Tf receptor is similar but not identical to that involved in the binding of Tf. To the best of our knowledge, this is the first study indicating that TfR may be involved in the binding of a ligand different from Tf. We evaluated whether the system of antigen presentation by T cells could also be used for identification of the sequence relevant for HBenvAg interaction with T cells. Our results dearly indicate that this sequence mapped to the pre-S2 region, based on the ability of T cells to present both the entire (pre-S1, pre-S2, S) protein and the protein containing pre-S2/S regions without pre-S1 to a S-specific T clone (VB27), and on their failure to present the S protein alone to the same clone. Moreover, the finding that T cells presented the r(S,L) protein, containing the only 133-145 sequence of pre-S2 region, to lesser extent than r(pre-S2, S) containing the entire pre-S2, suggests that HBenvAg epitope binding to T cells overlaps the 133-145 pre-S2 peptide. Comparative analysis of the capacity of different HBenvAg molecular constructs to inhibit "If binding provided evidence strictly in line with results derived from antigen-presentation studies, i.e., the HBenvAg sequence interacting with TfR on T cells is contained in the pre-S2 region. More precise mapping of this epitope may be useful for the design of innovative synthetic vaccines. Studies on T and B cell recognition of pre-S2 epitopes suggest that the 120-132 and 132-149 peptides are particularly crucial for virus neutralization (26, 34, 35) . Moreover, the pre-S2 domain, as well as the pre-S1 region, have been proposed to be involved in attachment of HBV to hepatocytes and to other cells that are susceptible to infection (26) .

It has been hypothesized that antigen presentation by T cells plays a role in immunoregulation (36, 37) . In our system, antigen-specific CD4 + T clones killing presenting T cells may participate in generalized immunosuppression by elimination of T lymphocytes. In particular, CD71 § cells expressing class II molecules, as macrophages or activated T cells, could be susceptible to lysis by HBenvAg presentation to class II-restricted CTL.

The in vivo relevance of cytotoxic activity mediated by CD4 + T cells is controversial, but the finding that these cells act as CTL has been reported for a number of viral infection models, including HBV, measles, herpes, and coronavirus (14, 15, (38) (39) (40) (41) (42) . In our hands, CD4 + CTL did not elicit bystander lysis, indicating that they killed by a direct target/CTL contact mechanism, and not aspecifically by secreted soluble cytotoxic mediators (25, 43) . The finding that CD4 + CTL are able to lyse both CD4* and CD8 + CTL clones in a HLA-restricted manner are in contrast with reports suggesting that CTL are resistant to self-mediated lysis (44) . These studies used murine CTL instead of human CD4 + CTL, utilized in the present study, thus, possibly explaining this discrepancy (6) . Experiments with human class II-restricted CTL, specific for the HIV gp120, indicated that lysis of gp120-presenting CD4 + T cells represent an immunosuppression mechanism (10, 37) . Analogously, class II-restricted CTL specific for HBenvAg can lyse themselves or other HBenvAg-presenting CD4 + or CD8 + T cells, thereby downregnlating the T cell response and posing a selective advantage for HBV persistence (45) .

In conclusion, our data indicate that both CD4 § and CD8 + activated T cells can process and present HBenvAg with high efficiency. More importantly, HBenvAg uptake via TfP, seems to be an essential initial step for the presentation of this antigen. Human TfR (20) is expressed on a variety of cell types, including hepatocytes (21, (46) (47) (48) (49) (50) (51) (52) (53) , which represent the privileged target cells of HBV infection. The wide TfP, distribution may thus explain the broad tissue tropism of HBV (54) . Other pathogens enter cells using the binding site for a known physiological ligand (55-58); e.g., HIV uses the CD4 receptor, rhinoviruses enter through ICAM-1 molecules, herpes simplex virus I through the fibroblast growth factor receptor, and Pneumocystis carinii through the mannose receptor (59) (60) (61) (62) .

The identification of both the receptor used by HBV to enter cells and the HBenvAg epitope relevant for this interaction may have fundamental implications, not only for the design of alternative vaccines, but also on therapeutic approaches to chronic HBV infection.

Antigen processing

Antigen presenting function of the macrophage

The cell biology of antigen processing and presentation

Activated human T cells can present denaturated antigen

Human T cell clones present antigen

Specific killing ofcytotoxic T cells and antigen-presenting cells by CD4 § cytotoxic T cell clones. A novel potentially immunoregulatory T-T cell interaction in man

Presentation of autoantigens by human T cells

Antibodies as antigens. The use of mouse monoclonal antibodies to focus human T cells against selected targets

T cells can present antigens such as HIV gp120 targeted to their own surface molecules

Analysis of host-virus interactions in AIDS with anti-gp120 T cell clones: effect of HIV sequence variation and a mechanism of CD4 + cell depletion

Antigen presentation by antigen-specific B lymphocytes

Antigen-specific interactions between T and B cells

Antigen presentation by B cells and its significance in T-B interactions

Selective killing of hepatitis B envelope antigenspecific B cells by class I-restricted, exogenous antigen-specific T lymphocytes

Recognition of hepatitis B virus envelope proteins by liver-infiltrating T lymphocytes in chronic HBV infection

A candidate vaccine for hepatitis B containing the complete virus surface protein

A new hepatitis B vaccine containing pre-S1 and pre-S2 epitopes from saccharomyces cerevisiae

Synthesis in animal cells of hepatitis B surface antigen partides carrying a receptor for polymerized human serum albumin

Monoclonal antibodies to transferrin receptors

Transferrin receptors: structure and function

Transferrin binding to K562 cell line: effect of heine and sodium butyrate induction

Decrease in macrophage antigen catabolism caused by ammonia and chloroquine is associated with inhibition of antigen presentation to T cells

The relation between major histocompatibility complex (MHC) restriction and the capacity of Ia to bind immunogenic peptides

Adhesion receptors of the immune system

Self recognition by T cells. I. Bystander killing by targets cells bearing syngeneic MHC antigens

Nature and display of hepatitis B virus envelope proteins and the humoral immune response

Receptor-mediated antigen uptake and its effect on antigen presentation to class II-restricted T lymphocytes

Viral persistence

Hepatitis B virus infects cultured human lymphoblastoid cells

Internalization of lymphocyte membrane components

Intracellular class II HLA antigens are accessible to transferrin-neuraminidase conjugates internalized by receptor-mediated endocytosis

Exposure of K562 cells to anti-receptor monoclonal antibody OKT9 results in rapid redistribution of the transferrin receptor

Role of the human transferrin receptor cytoplasmic domain in endocytosis: localization of a specific signal sequence for internalization

Genetic and molecular basis for T and B cell recognition of hepatitis B viral antigens

A synthetic peptide vaccine involving the product of the pre $2 region of hepatitis B virus DNA: protective ef~cacy in chimpanzees

Is suppression a function of class IIrestricted cytotoxic T cells?

Antigen processing and CD4 § T cell depletion in AIDS

Establishment of cytotoxic T-cell clones specific for cells infected with mouse hepatitis virus

Recognition of hepatitis B virus surface antigen by human T lymphocytes: proliferative and cytotoxic responses to a major antigenic determinant defined by synthetic peptides

Measles virus-specific T4 + human cytotoxic T cell clones are restricted by class II HLA antigens

Helper activity in antigen-specific antibody production mediated by CD4 + human cytotoxic T cell clones directed against herpes simplex virus

Nucleocapsid or spike protein-specific CD4 § T lymphocytes protect against coronavirus-induced encephalomyelitis in the absence of CD8 + T cells

Perforin dependent and independent pathways of cytotoxicity mediated by lymphocytes

How do cytotoxic T lymphocytes avoid self lysis?

Immunologic and molecular aspects of viral persistence: hepatitis B virus as a model

Role of iron in T cell activation: TH1 clones differ from TH2 clones in their sensitivity to inhibition of DNA synthesis caused by IgG mAbs against the transferrin receptor and the iron chelator deferoxamine

Role oftransferrin receptors and endocytosis in iron uptake by hepatic and erythroid cells

An epitope of the transferrin receptor is exposed on the cell surface of high-grade but not low-grade human lymphomas

Human cell surface glycoprotein related to cell proliferation is the receptor for transferrin

Transferrin receptor induction in mitogen-stimulated human T lymphocytes is required for DNA synthesis and cell division and is regulated by interleukin 2

Expression of the transferrin receptor gene during the process of mononuclear phagocyte maturation

Expression of transferrin receptor in phytohemoagglutinin-stimulated human T lymphocytes. Evidence for a three-step model

Iron up-modulates the expression of transferrin receptors during monocyte-macrophage maturation

The molecular biology of the hepatitis B viruses

Epstein-Barr virus receptor of human B lymphocytes is the C3d receptor CR2

Is the acetylcholine receptor a rabies virus receptor?

Isolation and biochemical characterization of the mammalian reovirus type 3 cell-surface receptor

Epidermal growth factor receptor occupancy inhibits vaccinia virus infection

Delineation of a region of the human immunodeficiency virus type 1 gp120 glycoprotein critical for interaction with the CD4 receptor

A cell adhesion molecule, ICAM-1, is the major surface receptor for rhinoviruses

Mediation of virion penetration into vascular cells by association of basic fibroblast growth factor with herpes simplex virus type 1

Uptake of Pneumocystis carinii mediated by the macrophage mannose receptor

We thank L. Adorini, A. Alberti, M. Ferrarini, and A. Lanzavecchia for criticism and suggestions.This work was supported in part by grants from Ministero dell'Universit~ della Ricerca Scientifica and by "Programma Terapia dei Tumori:' Istituto Superiore di SanitL Rome, Italy.