The JI
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     
 

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow Request Permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Pollard, K. M.
Right arrow Articles by Tan, E. M.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Pollard, K. M.
Right arrow Articles by Tan, E. M.
The Journal of Immunology, 2000, 165: 2263-2270.
Copyright © 00 by The American Association of Immunologists

Proteolytic Cleavage of a Self-Antigen Following Xenobiotic-Induced Cell Death Produces a Fragment with Novel Immunogenic Properties1

K. Michael Pollard2, Deborah L. Pearson, Martin Blüthner3 and Eng M. Tan

Department of Molecular and Experimental Medicine, The Scripps Research Institute, La Jolla, CA 92037


   Abstract
Top
 Abstract
Introduction
 Materials and Methods
 Results
 Discussion
 References
 
The heavy metal mercury elicits a genetically restricted autoantibody response in mice that targets the nucleolar autoantigen fibrillarin. HgCl2-induced cell death of macrophages resulted in the proteolytic cleavage of fibrillarin. A prominent feature of mercury-induced cell death was the generation of a 19-kDa fragment of fibrillarin that was not found following apoptotic or nonapoptotic cell death induced by stimuli other than mercury. Proteolysis of fibrillarin lacking cysteines, and therefore unable to bind mercury, also produced the 19-kDa fragment, suggesting that a mercury-fibrillarin interaction was not necessary for the unique cleavage pattern of this self-Ag. In contrast to immunization with full-length fibrillarin, the 19-kDa fragment produced anti-fibrillarin Abs with some of the properties of the HgCl2-induced anti-fibrillarin response. We propose that cell death following exposure to an autoimmunity-inducing xenobiotic can lead to the generation of novel protein fragments that may serve as sources of antigenic determinants for self-reactive T lymphocytes.


   Introduction
Top
 Abstract
 Introduction
Materials and Methods
 Results
 Discussion
 References
 
Cell death is accompanied by proteolysis of intracellular proteins, including nuclear autoantigens (1, 2, 3). Apoptosis, a morphological description of cell death elicited by exposure to numerous stimuli (4), results in nuclear fragmentation and the accumulation of nuclear autoantigens into cell surface blebs, or apoptotic bodies (5). It has been argued that apoptotic cell death, as a result of stimuli such as UV irradiation or viral infection, could be a mechanism for the release of intracellular immunogens to the surrounding tissue, leading to self-Ag presentation and autoimmunity (6). The observation that apoptotic bodies contain multiple autoantigens, as well as many of the components that comprise autoantigenic macromolecular complexes (5), may explain why certain autoantibody responses, particularly those to multiple components of the same complex, are linked (7). However, apoptosis is thought to be a noninflammatory process (4), and the processing and presentation of self-Ags as a result of apoptotic cell death may contribute more to deletion and tolerance, than stimulation, of autoreactive cells (8, 9). Autoimmunity is more likely to be associated with defective apoptotic cell death, such as the acceleration of systemic autoimmunity found in mice with the lpr or gld mutations (10).

There has been little discussion of the possible role of nonapoptotic cell death in the induction of autoimmunity, even though it is associated with proteolysis of autoantigens (2) and inflammatory reactions (11). Inhibitors of apoptosis-associated proteolysis are ineffective at blocking proteolysis due to nonapoptotic cell death, suggesting that the two forms of cell death activate different proteases (2). This is highlighted by the finding that proteolysis of nuclear autoantigens during nonapoptotic cell death results in cleavage patterns that differ from those found in apoptotic cell death (2). The generation of unique protein fragments during nonapoptotic cell death proteolysis suggests a mechanism for the production of novel antigenic determinants that may lead to autoimmune responses (12, 13).

The heavy metal mercury, which can induce nonapoptotic cell death in vitro (2, 11), elicits a genetically restricted autoimmune response that targets the nucleolar autoantigen fibrillarin (14). Mercury-induced cell death is accompanied by modification of the molecular properties of fibrillarin (15). However mercury-modified fibrillarin is a poor Ag for HgCl2-induced anti-fibrillarin autoantibodies (15), implying that the metal-modified protein may function as a source of T cell determinants. As mercury-protein interaction has been shown to alter protease sensitivity (16), it is possible that mercury modification of fibrillarin may influence Ag processing, leading to the production of cryptic T cell determinants. Abnormal Ag processing of metal-modified protein is supported by recent studies that have identified cryptic peptides that stimulate IL-2 production by T cell hybridomas obtained from mice immunized with bovine RNase A complexed to gold (17, 18). The cryptic nature of these determinants has been suggested by studies showing that T cell hybridomas from mice immunized with bovine RNase alone do not respond to the same peptides (17, 18). Although these studies did not elucidate the process of peptide generation from a metal-modified Ag, APCs are involved as the presence of metal-exposed macrophages contributed to T cell activation (17). The observation that lysates of peritoneal cells from HgCl2-exposed mice enhance lymphadenopathy in HgCl2-primed mice suggests that mercury-exposed APCs may contain antigenic material (19).

In this study we report that the proteolysis of fibrillarin in macrophages following mercury-induced cell death leads to unique cleavage fragments. Using radiolabled fibrillarin as substrate, we observed that lysates from mercury-killed cells produced a cleavage pattern that differed from that of apoptotic cell death. A prominent feature of mercury-induced cell death was the generation of a 19-kDa fragment of fibrillarin that was not found following apoptotic cell death. Significantly, the 19-kDa fragment was not associated with nonapoptotic cell death due to stimuli other than mercury. Proteolysis of fibrillarin lacking cysteines, and therefore unable to bind mercury, also produced the 19-kDa fragment, suggesting that a mercury-fibrillarin interaction was not necessary for abnormal cleavage of this self-Ag. Immunization with the 19-kDa fragment resulted in anti-fibrillarin Abs that recognized evolutionarily conserved determinants of fibrillarin, a feature of HgCl2-induced anti-fibrillarin Abs (14). We propose that cell death following exposure to an autoimmunity-inducing xenobiotic can lead to the generation of novel protein fragments that may serve as sources of antigenic determinants for self-reactive T lymphocytes.


   Materials and Methods
Top
 Abstract
 Introduction
 Materials and Methods
Results
 Discussion
 References
 
Cell culture

J774A.1, EL-4, and P3 (P3 x 63Ag8.653) murine cell lines were obtained from the American Type Culture Collection (Manassas, VA). Cell lines were cultured in complete RPMI 1640 medium containing 10% FCS (HyClone, Logan, UT), 2 mM L-glutamine (Sigma, St. Louis, MO), 10 mM HEPES (pH 7.2) (Flow Laboratories, Irvine, Scotland), 1% nonessential amino acids (Irvine Scientific, Santa Ana, CA), 20 mM sodium carbonate (Fisher Scientific, Fair Lawn, NJ), and 20 µg/ml gentamicin sulfate (Irvine Scientific). Cell lines were maintained in log phase growth by splitting to 5 x 104 cells/ml every 2 days.

Xenobiotic additions to cell cultures

Cells were seeded at a density of 5 x 104/ml and grown for 2 days before addition of xenobiotics. HgCl2 was added from a 20 mM stock solution in sterile PBS into complete RPMI 1640 to achieve final concentrations ranging from 0 to 100 µM. Control cultures received media-only additions. Etoposide (VP-16; Sigma) was added from a 20 mg/ml stock solution in DMSO to give a final concentration of 100 µM. Control cultures received an equivalent volume of DMSO. Cultures were incubated for up to 12 h at 37°C before harvesting. Ethanol-exposed cultures received ethyl alcohol (200 proof; Quantum Chemical, Tuscola, IL) at a final concentration of 10% (v/v) for 3 h. Heat-induced cell death was achieved by culturing cells for 3 h at 50–55°C before harvesting.

Harvesting of cell lysates

After culture in the presence of individual xenobiotics or heat treatment, cells were washed in 127 mM NaCl, 5.5 mM KCl, 1 mM CaCl2, 2 mM MgSO4, 10 mM dextrose, and 20 mM HEPES (pH 7.0), and then were lysed in 1% Nonidet P-40, 50 mM NaCl, and 10 mM HEPES (pH 7.0) (20) at 107 cells/ml. Lysates were aliquoted and stored at -70°C.

Cell viability and morphology

Cell viability was determined by trypan blue exclusion. To harvest J774A.1 cells for viability studies, cultures were incubated in 0.05% trypsin and 0.02% EDTA (Sigma) for 15 min. Examination of cell morphology by light microscopy was performed using Leukostat stain (Fisher Scientific) as previously described (15).

Insertion of T7 Tag and M2 Flag peptide sequences onto mouse fibrillarin

cDNAs containing wild-type and mutant (Cys105,274 to Ala105,274) mouse fibrillarin were as previously described (15). T7 Tag and M2 Flag peptide sequences were added to the amino and carboxy ends of fibrillarin, respectively, by PCR as follows. The forward primer contained an XhoI restriction site and the T7 Tag sequence, and the reverse primer had the complement of the M2 Flag sequence and an NcoI restriction site. The PCR product was purified by agarose gel electrophoresis followed by passage over Spin-X columns (Sigma). The purified PCR product and the vector pET28b (Novagen, Madison, WI) were cut with XhoI and NcoI, purified as described above, and ligated together using T4 ligase (Boehringer Mannheim, Indianapolis, IN). Following transformation into XL-1 Blue super competent cells (Stratagene, San Diego, CA), kanamycin-resistant colonies were expanded and plasmid DNA was isolated by miniprep (Qiagen, Valencia, CA). Inserts were subjected to double-strand DNA sequencing to confirm the correct sequence and orientation of the T7/fibrillarin/M2 constructs containing either wild-type or mutant fibrillarin.

Analysis of the proteolysis of fibrillarin

Proteolysis of endogenous fibrillarin was analyzed by immunoblot using human anti-fibrillarin autoantibodies as previously described (15). Proteolysis of exogenous fibrillarin was assayed by addition of 2 µl of 35S-labeled mouse fibrillarin (21) to 50 µl of cell lysate in 1% Nonidet P-40, 50 mM NaCl, and 10 mM HEPES (pH 7.0). Following incubation at 37°C samples were mixed with an equal volume of 2x Laemmli sample buffer, boiled, and separated on 15 or 20% SDS-PAGE gels. Radiolabeled bands were enhanced by fluorography and recorded on x-ray film (Kodak, New Haven, CT) at -70°C.

Further identification of fragments resulting from proteolytic cleavage of fibrillarin was achieved using specific Abs to immunoprecipitate the T7/fibrillarin/M2 constructs as follows. Anti-T7 Tag (Novagen) or anti-M2 Flag (Kodak) Abs were adsorbed onto 100 µl Protein A-Sepharose beads (3% w/v; Amersham Pharmacia Biotech, Piscataway, NJ) in 500 µl of 50 mM Tris (pH 7.4), 150 mM NaCl, 5 mM EDTA, 0.5% Nonidet P-40, 0.5% SDC, 0.1% SDS, and 0.02% sodium azide (NET2) (22) for 2 h at 4°C. Unabsorbed Ab was removed by washing the beads with NET2. Wild-type and mutant 35S-labeled mouse T7/fibrillarin/M2 was prepared by in vitro transcription and translation as previously described (15). To prepare mercury-modified 35S-labeled mouse T7/fibrillarin/M2, the radiolabeled protein was incubated with HgCl2 as described previously (15). Free mercury was removed by passage over a Bio-Rad P-30 Bio-Spin Chromatography column (Bio-Rad, Hercules, CA). For immunoprecipitation of cleavage fragments, 10 µl of radiolabeled protein was incubated with 60 µl of cell lysate and 330 µl of 1% Nonidet P-40, 50 mM NaCl, and 10 mM HEPES (pH 7.0) at 37°C for 1 h, and then was added to Protein A-Sepharose beads adsorbed with either anti-T7 or anti-M2 Abs. After incubation for 2 h at 4°C, unbound material was removed by washing in NET2, and the beads were suspended in 30 µl of 2x Laemmli sample buffer. Bound material was analyzed by SDS-PAGE and autoradiography as described above.

Inhibition of protease activity

Protease inhibitors were added individually or as a mixture to cell lysates for 30 min before addition of 35S-labeled mouse fibrillarin: 5 µg/ml aprotinin (Boehringer Mannheim), 1 mg/ml Pefabloc (Boehringer Mannheim), 5 µg/ml leupeptin (Boehringer Mannheim), 2 µg/ml pepstatin A (Sigma), 50 µg/ml L-1-chloro-3-[4-tosylamido]-7-amino-2-heptanone-HCl (Sigma), 50 µg/ml CMBS (4(chloro-mercuric)benzene-sulfonic acid) Na salt (Aldrich Chemical, Milwaukee, WI), 125 µg/ml N-ethylmalemide (Sigma), and 10 mM EDTA (Fisher Scientific).

Quantitation of proteolytic activity

Proteolytic activity was quantified by digestion of resorufin-labeled casein. Fifty microliters of casein-resorufin (4 mg/ml; Boehringer Mannheim) and 50 µl of incubation buffer (200 mM Tris-HCl (pH 7.8) and 20 mM CaCl2) were added to 100 µl of cell lysate. After incubation at 37°C, the reaction was stopped by the addition of 480 µl of 5% TCA. Following a 10-min incubation at 37°C, insoluble material was precipitated by centrifugation and 400 µl of supernatant mixed with 600 µl of assay buffer (500 mM Tris-HCl, pH 8.8). Released resorufin was determined by adsorbance at 574 nm. Trypsin was used as a positive control to construct standard curves to determine protease activity (µg substrate digested over time). Preliminary experiments revealed that 50 µl of lysate (5 x 105 cell equivalents) from HgCl2-killed J774A.1 cells could digest ~40 µg of substrate over 12 h at 37°C.

Immunization studies

Recombinant wild-type and mutant T7/fibrillarin/M2 proteins were expressed and purified as previously described (23). cDNA constructs containing the N-terminal 19-kDa of both wild-type and mutant fibrillarin in pET28b were produced by restriction of T7/fibrillarin/M2 with BSU361 and XhoI to remove nucleotides 787-1041. Following ligation with olignucleotide linkers to restore the M2 Flag and poly-His peptide tag sequences, the fragments were expressed and purified as described for full-length fibrillarin (23). Mercury-modified recombinant proteins were produced as previously described (23). B10.S (H-2s) mice were immunized s.c. at the base of the tail with 50 µg of recombinant protein in CFA, and boosted 10 days later by a similar injection in IFA. Mice were bled ~3 and 8 wk after the initial immunization, and anti-fibrillarin Abs were detected by immunofluorescence, immunoblot, and immunoprecipitation methods as described previously (14).


   Results
Top
 Abstract
 Introduction
 Materials and Methods
 Results
Discussion
 References
 
Mercury-induced cell death and proteolysis of fibrillarin

Mercury-induced cell death is associated with molecular modification of fibrillarin (15) and protease activation (2). To determine whether mercury-induced cell death results in proteolytic activity that cleaves fibrillarin, lysates were prepared from the murine macrophage cell line J774A.1 that had been cultured in the presence of increasing concentrations of HgCl2 for 3 h. As described previously (2, 15), in the presence of 10 and 20 µM HgCl2 J774A.1 cells remained viable and continued to proliferate. At 40 µM HgCl2 all cells were dead within 2 h as judged by trypan blue exclusion. Immunoblotting using autoantibodies to fibrillarin did not detect fragments of proteolytic cleavage at any HgCl2 concentration (Fig. 1GoA), confirming previous observations (2). Cell death-associated mercury modification of fibrillarin (15), identified by the characteristic shift in migration of fibrillarin from 34 to 32 kDa under nonreducing conditions (Fig. 1GoB), confirmed that 40 µM HgCl2 induced cell death. The minor immunoreactive bands at ~23–25 kDa in the 40 µM HgCl2 lysate in Fig. 1Go, A and B, were not seen in subsequent immunoblotting experiments (see below).



View larger version (37K):
[in this window]
[in a new window]
  		FIGURE 1. Detection of fibrillarin proteolysis following HgCl2-induced cell death. The macrophage cell line J774A.1 was exposed to increasing concentrations of HgCl2 for 3 h to induce cell death, and cell lysates were used to detect cleavage of fibrillarin. A, Immunoblot with anti-fibrillarin autoantibodies of lysates separated by SDS-PAGE under reducing conditions (+2-ME). B, Immunoblot with anti-fibrillarin autoantibodies of lysates separated by SDS-PAGE under nonreducing conditions (-2-ME) showing the accumulation of the faster migrating mercury-modified fibrillarin in the lysate from cells exposed to 40 µM HgCl2. C, Proteolysis of 35S-labeled fibrillarin added to cell lysates for 1 h at 37°C showing almost complete proteolysis of fibrillarin in the lysate from cells exposed to 40 µM HgCl2. As described previously (15 ), in vitro transcribed and translated fibrillarin migrates as a doublet of 34–36 kDa.

 
The inability of immunoblotting to detect proteolytic fragments of fibrillarin following cell death supports previous observations (1, 2, 3). However most anti-fibrillarin autoantibodies react poorly, if at all, with less than full-length fragments of fibrillarin (14), making immunoblotting an inappropriate technique to detect proteolytic fragments. As an alternative strategy for the detection of proteolytic fragments, murine 35S-labeled fibrillarin was added to cell lysates and proteolysis was revealed by cleavage of radiolabeled exogenous fibrillarin. Using this procedure, the lysate from cells incubated in 40 µM HgCl2 for 3 h contained proteolytic activity capable of almost complete digestion of full-length fibrillarin (Fig. 1GoC). The greater sensitivity of this technique in detecting proteolysis of fibrillarin was confirmed by additional immunoblotting experiments using lysates from J774A.1 cells cultured in 40 µM HgCl2 for 0–7 h. Although little loss of fibrillarin was detected after 1 or 3 h, longer exposures to HgCl2 revealed loss of full-length endogenous fibrillarin (Fig. 2Go). These experiments identify the activation of proteolytic activity following HgCl2-induced cell death that cleaves fibrillarin.



View larger version (60K):
[in this window]
[in a new window]
  		FIGURE 2. Detection of fibrillarin proteolysis using anti-fibrillarin autoantibodies. J774A.1 cells were incubated in 40 µM HgCl2 for up to 7 h, lysed, and subjected to immunoblotting using anti-fibrillarin autoantibodies. Loss of immunoreactivity is evident after 5 h, with complete loss of Ab reactivity following 7-h exposure to HgCl2.

 
Lysates from untreated J774A.1, P3, and EL-4 cell lines as well as HgCl2-killed P3 and EL-4 cell lines did not contain protease activity (Fig. 3Go), indicating that mercury-induced cell death did not elicit protease activity in all cell types. Addition of HgCl2 to lysates from untreated J774A.1 cells did not lead to proteolysis (data not shown), suggesting that mercury was not activating an existing protease, but that other events (i.e., cell death) need to occur to generate protease activity.



View larger version (76K):
[in this window]
[in a new window]
  		FIGURE 3. HgCl2-induced cell death of macrophages leads to proteolysis of fibrillarin. J774A.1, P3, and EL-4 cell lines were exposed to 40 µM HgCl2 for 3 h and 35S-labeled fibrillarin was added to the cell lysates for 1 h at 37°C to assess proteolysis. Only lysates from the macrophage cell line J774A.1 contained proteolytic activity that cleaved fibrillarin. Control signifies lysates from cell cultures receiving media only during the exposure period.

 
Comparison of the mercury-induced and apoptotic proteolytic cleavage of fibrillarin

To differentiate between HgCl2-induced nonapoptotic cell death and apoptotic cell death, studies were done using the apoptosis-inducing cancer drug etoposide (VP-16) (24). J774A.1 cells incubated with 100 µM VP-16 show the typical morphology of apoptosis (15). Lysates from such cells contained protease activity that cleaved 35S-labeled fibrillarin (see below), although the activity, as judged by digestion of casein-resorufin was about one-third of that of lysates from HgCl2-killed cells (data not shown). Attempts to differentiate mercury-induced and apoptotic proteolysis of fibrillarin with protease inhibitors revealed that the serine protease inhibitors aprotinin and, to a lesser extent, Pefabloc-inhibited proteolysis resulting from both mercury-induced and apoptotic cell death (data not shown). Other protease inhibitors, including the serine protease inhibitors leupeptin and L-1-chloro-3-[4-tosylamido]-7-amino-2-heptanone-HCl, were ineffective. The metalloproteinase inhibitor EDTA did not inhibit proteolysis (data not shown).

As autoantigen proteolysis during mercury-induced cell death and apoptosis has revealed differences in cleavage patterns (2), experiments were done to compare the cleavage pattern of 35S-labeled fibrillarin in lysates from HgCl2- and VP-16-killed J774A.1 cells. Preliminary experiments showed that proteolysis in lysates from HgCl2- and VP-16-killed cells were time and lysate-concentration dependent (data not shown). Comparison of the respective cleavage patterns showed subtle differences between the two types of cell death, with additional fragments present after proteolysis in lysates from HgCl2-killed cells.

To help distinguish the cleavage patterns further, 35S-labeled T7/fibrillarin/M2 was used as a substrate for fibrillarin proteolysis, and anti-T7 Tag or anti-M2 Flag Abs were used to immunoprecipitate cleavage fragments. Although this strategy only identified fragments with intact amino or carboxyl termini, it did allow use of larger amounts of substrate. As seen later in Fig. 5Go, this procedure greatly enhanced the clarity of some cleavage fragments, as evidenced by the detection of proteolytic activity in the control cell cultures, which appears due to constitutive apoptotic cell death. Comparison of the cleavage patterns produced by VP-16- and HgCl2-induced cell death showed a number of differences, particularly in anti-M2 immunoprecipitates of wild-type fibrillarin (Fig. 4Go) with several bands appearing more intensely after cleavage in lysate from HgCl2-killed cells. However the most striking difference was the appearance of a 19-kDa band in the anti-T7 immunoprecipitate of fibrillarin cleaved in lysate from HgCl2-killed cells. This band was not found in immunoprecipitates following incubation of fibrillarin substrate in control or apoptotic cell lysates. The 19-kDa band was also produced when mutant fibrillarin, lacking cysteines, was used as substrate (Fig. 4Go), suggesting that the appearance of this fragment may not be dependent upon mercury-modification of fibrillarin.



View larger version (50K):
[in this window]
[in a new window]
  		FIGURE 5. The 19-kDa fragment of fibrillarin is not a common product of nonapoptotic proteolysis. 35S-labeled T7/fibrillarin/M2 was added to lysates from J774A.1 cells killed by ethanol, heat, or HgCl2, and cleavage fragments were analyzed by immunoprecipitation using anti-T7 Tag Abs. Only the immunoprecipitate of fibrillarin cleaved in the lysate from HgCl2-killed cells contained the 19-kDa fragment (arrow). Control is lysate from cells incubated in media only.

 


View larger version (120K):
[in this window]
[in a new window]
  		FIGURE 4. Comparison of the cleavage of wild-type and mutant 35S-labeled T7/fibrillarin/M2 in lysates from J774A.1 cells killed with HgCl2 or VP-16. Wild-type or mutant 35S-labeled T7/fibrillarin/M2 was added to lysates from J774A.1 cells killed by HgCl2 or VP-16 and cleavage fragments analyzed by immunoprecipitation using anti-T7 Tag or anti-M2 Flag Abs. Only the anti-T7 Tag immunoprecipitate of fibrillarin cleaved in the lysate from HgCl2-killed cells contained the 19-kDa fragment (arrows). Control is lysate from cells incubated in media only.

 
To determine whether interaction with mercury could further alter the proteolysis of fibrillarin, 35S-labeled T7/fibrillarin/M2 was modified by prior incubation with HgCl2 before addition to cell lysates. Immunoprecipitation of the cleavage fragments of mercury-modified and unmodified fibrillarin again revealed the presence of the 19-kDa fragment, arguing that its genesis could also be from mercury-modified fibrillarin. Comparison of the cleavage patterns of mercury-modified and unmodified fibrillarin using lysates from mercury-killed cells did not reveal the presence of fragments unique to the proteolysis of mercury-modified protein (data not shown). However, it must be noted that the immunoprecipitation protocol being used is selective in recognizing only fragments containing either the amino (T7) or carboxy (M2) peptide tags, and consequently may not identify cleavage fragments composed of internal protein sequence that may or may not require mercury for their generation.

Absence of 19-kDa fragment in non-mercury-induced cell death

It was possible that the 19-kDa fragment of fibrillarin was a common feature of proteolysis during nonapoptotic cell death. To determine whether this was the case, J774A.1 cells were killed by either ethanol or heat, and cell lysates were used as sources of protease to cleave 35S-labeled T7/fibrillarin/M2. Although lysates from heat- and ethanol-killed cells contained proteolytic activity, generation of the 19-kDa fragment was only possible with lysate from HgCl2-killed cells (Fig. 5Go).

Immunization with full-length and 19-kDa fibrillarin results in Abs with different antigenic properties

To determine the immunogenic potential of the 19-kDa fragment, B10.S (H-2s) mice were immunized with the product of a construct containing the N-terminal 19-kDa fragment of fibrillarin. The unmodified wild-type 19-kDa protein elicited anti-fibrillarin Abs in three of four mice as judged by immunoblot; however, immunization with the mercury-modified 19-kDa fragment did not elicit an immunoblot-positive response (Fig. 6Go, left). Comparative immunization studies with the full-length protein elicited a similar response; unmodified full-length protein was immunogenic in all four immunized mice, but mercury modification reduced immunogenicity with only one mouse being immunoblot positive (Fig. 6Go, right).



View larger version (113K):
[in this window]
[in a new window]
  		FIGURE 6. Immunization with wild-type full-length and 19-kDa mouse fibrillarin elicits immunoblot-positive Abs. B10.S mice were immunized with unmodified or mercury-modified (+Hg) wild-type (Wt) or mutant (Mut) 19-kDa (left panel) or full-length (right panel) mouse fibrillarin. Each lane represents serum from an individual mouse. Anti-fibrillarin Abs were detected by immunoblot on SDS-PAGE separated rat liver nuclei. Positive (+) anti-fibrillarin control was from a HgCl2-treated B10.S mouse as described previously (14 ). The arrow indicates 34-kDa fibrillarin. The identity of the band at ~22 kDa in the left panel is unknown.

 
Mutant 19-kDa (Fig. 6Go, left) and full-length protein (Fig. 6Go, right), which do not bind mercury due to mutation of Cys to Ala (15), elicited immunoblot-positive anti-fibrillarin Abs in only one or two of the four mice immunized. As expected, mercury-modification did not appear to alter the immunogenicity of these Ags.

Further analysis of antigenic specificity revealed that Abs raised by immunization with full-length fibrillarin were negative by immunofluorescence for antinucleolar reactivity on HEp-2 cells (data not shown), and could not immunoprecipitate 35S-labeled mouse fibrillarin (Fig. 7Go, upper). Sera from mice immunized with the various forms of the 19-kDa fragment were also negative by immunofluorescence (data not shown), however they did immunoprecipitate 35S-labeled mouse fibrillarin (Fig. 7Go, lower). Although this response was quite variable and a number of the reactions were weak, they were clearly more reactive than the negative response found following immunization with full-length protein.



View larger version (74K):
[in this window]
[in a new window]
  		FIGURE 7. Immunization with the 19-kDa fragment of mouse fibrillarin elicits immunoprecipitation-positive Abs. B10.S mice were immunized with unmodified or mercury-modified (+Hg) wild-type (Wt) or mutant (Mut), full-length (upper) or 19-kDa (lower) mouse fibrillarin. Each lane represents serum from an individual mouse. Anti-fibrillarin Abs were detected in sera 8 wk after initial immunization by immunoprecipitation of 35S-labeled mouse fibrillarin. Positive (POS) and negative (NEG) controls were mAbs 17C12 and 7G3 as described previously (14 ). The arrow indicates 35S-labeled fibrillarin.

 
Several of the sera from mice immunized with the 19-kDa protein were also able to immunoprecipitate 35S-labeled yeast fibrillarin (Fig. 8Go). This was particularly true of those sera (Nos. 7, 10, and 16) that reacted most strongly with 35S-labeled mouse fibrillarin, suggesting that Ab titer influenced the number of sera found positive for 35S-labeled yeast fibrillarin immunoprecipitation. Recognition of evolutionarily conserved antigenic determinants is a feature of HgCl2-induced anti-fibrillarin Abs (14). These observations imply that an immunogen constructed to represent the 19-kDa fragment produced following proteolysis of fibrillarin during HgCl2-induced cell death elicits anti-fibrillarin Abs with novel antigenic properties.



View larger version (74K):
[in this window]
[in a new window]
  		FIGURE 8. Immunization with the 19-kDa fragment of mouse fibrillarin elicits Abs that recognize evolutionarily conserved antigenic determinants. B10.S mice were immunized with unmodified or mercury-modified (+Hg), wild-type (Wt) or mutant (Mut) 19-kDa mouse fibrillarin. Each lane represents serum from an individual mouse. Anti-fibrillarin Abs were detected in sera 8 wk after initial immunization by immunoprecipitation of 35S-labeled yeast fibrillarin. Positive (POS) and negative (NEG) controls were mAbs 72B9 and 7G3 as described previously (14 ). The arrow indicates 35S-labeled fibrillarin.

 

   Discussion
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
References
 
Xenobiotic-induced autoimmunity is characterized by autoantibody responses against native self-Ag, thus interaction between xenobiotic and Ag is not required for autoantibody reactivity (15, 25, 26). The genetically restricted anti-fibrillarin autoantibody response of mercury-induced autoimmunity is not directed against a fibrillarin-mercury complex even though a metal-protein interaction does occur (15). In attempting to uncover the significance of the metal-self-Ag complex in this highly restricted autoantibody response, the present study has demonstrated that mercury-induced cell death leads to a proteolytic cleavage of fibrillarin that appears specific to this xenobiotic. Generation of a unique (19 kDa) proteolytic cleavage fragment did not require interaction between mercury and fibrillarin. Rather, it appears that mercury-induced cell death was associated with a protease activity not found in other forms of induced cell death. Anti-fibrillarin Abs elicited by immunization with a 19-kDa recombinant protein displayed some properties of HgCl2-induced anti-fibrillarin Abs. Although this study does not exclude a role for xenobiotic-protein interaction in mercury-induced autoimmunity, it does highlight the potential of xenobiotic-induced cell death to produce novel protein fragments that stimulate self-reactivity that differs from that elicited by full-length protein.

Reports that a specific cellular morphology, and proteolytic cleavage of a subgroup of autoantigens, are associated with apoptotic cell death have suggested possible sources for self-immunogens (1, 3, 5, 6, 7). However, attempts to elicit autoimmunity by immunization with apoptotic cells have not proven more successful than immunization with cells killed by freezing and thawing (27). The observation that autoantibodies are reactive with the proteolytic fragments generated during apoptotic cell death (1, 3, 5, 6, 28) has been argued as evidence for a role of cleavage fragments as immunogens. However not all autoantigens are cleaved during cell death (1, 28). In addition, it must be noted that very few autoantibody-reactive cleavage fragments are less than 30 kDa (see Ref. 2), suggesting a size limit to Ab recognition of fragments produced during cell death. This may explain why previous studies have suggested that the 34-kDa fibrillarin is resistant to proteolytic cleavage during cell death (1, 2, 3, 28). In this study detection of proteolytic cleavage using a radiolabeled substrate showed that fibrillarin is cleaved during apoptotic cell death.

Comparative studies between HgCl2- and VP-16-induced cell death showed that both nonapoptotic and apoptotic protease activity were sensitive to the same inhibitors. Nonetheless, the cleavage pattern of fibrillarin was sufficiently different to suggest the action of different proteases. This supports a previous observation showing that the cleavage patterns for a number of autoantigens differ between nonapoptotic- (HgCl2, heat, ethanol) and apoptotic- (anti-Fas) induced cell death (2). Significantly, autoantibodies to these proteins, which included DNA topoisomerase 1, lamin B1, and the 70-kDa U1-snRNP, are not found in HgCl2-induced autoimmunity (29), nor do they interact with mercury (15). It also appears that the cleavage patterns produced by nonapoptotic stimuli (HgCl2, heat, or ethanol) do not differ for proteins like poly(ADP-ribose) polymerase and DNA topoisomerase 1 (2). An MHC-restricted autoantibody response and interaction with HgCl2 are features that distinguish fibrillarin as an autoantigen in HgCl2-induced autoimmunity. The observation that specific cleavage fragment(s) of fibrillarin result from HgCl2 induced cell death, and not other forms of cell death, suggests that novel cleavage fragments may function as autoimmunogens.

Mice were immunized with both full-length fibrillarin and the N-terminal 19-kDa fragment in the presence and absence of mercury to determine the immunogenic potential of the presumptive cleavage fragment. Both unmodified wild-type proteins elicited Ab responses that recognized fibrillarin by immunoblot. In contrast, the mercury-modified wild-type proteins were much less effective at eliciting immunoblot-positive reactions, suggesting that bound mercury may not contribute to the immunogenicity of fibrillarin. Mutant fibrillarin, which does not bind mercury due to replacement of Cys by Ala (15), elicited immunoblot-positive reactions in a minority of immunized mice. The reason for this reduced response is not clear but may reflect the subtle difference in sequence between the mutated immunogen and the wild-type Ag used in the immunoblot.

The anti-fibrillarin Ab response elicited by the N-terminal 19-kDa fibrillarin differed from the full-length protein by the presence of Abs that immunoprecipitated 35S-labeled mouse fibrillarin. Unlike immunoblot-positive Abs that were elicited primarily by unmodified wild-type protein, immunoprecipitation-positive Abs were found in mice immunized with either wild-type or mutant 19-kDa protein. Although the strength of the immunoprecipitation results following 19-kDa immunization were quite variable, they were clearly different from the uniformly negative result obtained with full-length fibrillarin. This difference in reactivity between Abs raised by immunization with 19-kDa and full-length protein imply that different populations of B cells are responding to the different Ags. The presence of bound mercury did not influence the response to the 19-kDa fragment, suggesting that immunogenicity was not influenced by mercury. The ability of high titer sera to immunoprecipitate yeast fibrillarin suggests recognition of a conserved conformational determinant, which is one of the distinguishing features of the anti-fibrillarin response in murine HgCl2-induced autoimmunity and human scleroderma (14). However, immunization with the 19-kDa fragment did not result in an Ab response that could be detected by indirect immunofluorescence. The characteristic "clumpy" nucleolar staining pattern is another distinguishing feature of anti-fibrillarin autoantibodies (14, 15) and at least one HgCl2-induced monoclonal anti-fibrillarin autoantibody is positive in immunofluorescence, immunoblot, and immunoprecipitation assays (15). Thus while the anti-fibrillarin response raised by the 19-kDa fragment is tantalizingly similar to the HgCl2-induced response, it is not identical.

The failure of immunization to exactly mimic the specificities of a bona fide autoantibody response is not uncommon (30) and most reports of Ab responses that recognize self-Ag have required the use of foreign homologues as immunogens (31, 32). In the case of the murine anti-fibrillarin response, simple administration of mercury produces a response that very closely mimics the human response (14). The target of the autoantibodies is not a fibrillarin-mercury complex (15) and, as immunization with mercury-modified full-length fibrillarin demonstrates, a fibrillarin-mercury complex is not an effective immunogen. Instead it appears that cleavage of the protein to mimic a fragment produced by mercury-induced nonapoptotic cell death provides more appropriate immunogenic material. However the availability of immunogenic peptides is unlikely to be the only requirement to imitate the HgCl2-induced anti-fibrillarin response. Other effects of mercury on the immune system (29), including specific cytokine requirements (33) and inhibition of Fas-mediated cell death (34), are likely contributors to the breaking of self-tolerance.

In this and a previous study (15), mercury-induced cell death demonstrated features of a nonapoptotic process based primarily on cellular morphology following exposure. Although the exact mechanism remains to be determined, the prelethal phase resembles oncosis (35) and is consistent with the nonapoptotic HgCl2-induced cell death reported by others (11). Cell death, as judged by trypan blue staining, occurs within the first 2 h. Proteolysis of fibrillarin, which is most evident after cell death, is thus a part of oncotic necrosis which leads to phagocytosis of necrotic debris and inflammation (11). Proteolysis of fibrillarin was found in macrophages suggesting that the most likely source of novel fibrillarin cleavage products might be phagocytes that accumulate necrotic debris following toxic exposure to mercury. This would result in the simultaneous accumulation of mercury, leading to cell death and cleavage by cell death protease(s). The resulting cellular debris would be subject to further phagocytosis, and potentially processing and presentation to T cells to elicit autoimmunity. This is supported by a number of studies showing T cell proliferation and cytokine expression, including IL-1 secretion by peritoneal macrophages, following exposure to HgCl2 (reviewed in Ref. 29). As HgCl2-induced autoimmunity is elicited by repeated exposure to mercury (29) it is conceivable that immunogenic material is produced by successive rounds of cell death, phagocytosis, and the processing and presentation of novel fragments of self-Ag to T cells. The observation that necrotic rather than apoptotic cell debris activates murine macrophages (36) and dendritic cells (37) supports the likelihood that HgCl2-induced cell death contributes immunogenic material.

In recent studies Griem and colleagues (17, 18) have used bovine RNase A as a model Ag to examine how metals might influence T cell recognition of protein Ags. Immunization of mice with unaltered RNase A or gold-altered RNase A (RNase A/Au(III)) led to a number of T cell hybridomas of varying antigenic specificities (17). Irrespective of the immunizing Ag the majority of T cell clones recognized Ags common to both unaltered RNase A and RNase A/Au(III). However five clones from RNase A/Au(III)-immunized mice recognized only RNase A/Au(III), and a further six clones from untreated RNase A immunized mice recognized only untreated RNase A. Thus the presence of gold led to T cell recognition of determinants specific to the metal-protein complex. However the presence of gold was not essential; simple oxidation of methionine side chains of RNase A produced material recognized by RNase A/Au(III)-specific T cell clones. Analysis of the fine specificity of RNase A/Au(III)-specific T cell clones revealed that four clones recognized a peptide containing amino acid residues 94–108, and the fifth clone recognized amino acid residues 7–21. T cell clones obtained following immunization with untreated RNase A did not recognize these peptides. Most significant was the finding that T cell recognition of peptides 94–108 and 7–21 did not require pretreatment of peptide with gold. This suggests that the RNase A-gold complex was being processed so that different peptides were being presented compared with untreated RNase A (17). Treatment of RNase A with a number of other metal ions failed to produce material reactive with the RNase A/Au(III)-specific T cell clones. However a more recent study (18) showed that lead-, platinum-, or nickel-treated bovine RNase A could serve as an Ag for RNase A/Au(III)-specific T cell clones. Although immunization with lead-, platinum-, or nickel-treated bovine RNase A resulted in metal-RNase A specific clones; only one clone, following lead-RNase A immunization, could recognize the cryptic peptide 94–108.

The relevance of the observations of Griem and colleagues (17, 18) to mechanisms of metal-induced autoimmunity remains to be determined, as responses to a foreign Ag like bovine RNase A are unlikely to mimic responses to a self-Ag such as fibrillarin. More importantly, as Ab responses to RNase A were not examined, the role that RNase A/Au(III)-specific T cells play in an Ab response is unclear. In particular, these studies do not explain how Ab responses might differ between protein and protein/metal immunizations when the majority of T cells clones produced recognize peptides common to both Ags. However these studies do demonstrate that immunization with metal-protein complexes can lead to T cell determinants that do not include the metal ion. Instead the presence of the metal ion appears to influence the generation of antigenic determinants so that novel immunogens become available for immune recognition. The ability of mercury to alter the proteolysis of fibrillarin suggests a mechanism whereby an autoimmunity-inducing xenobiotic might generate unique fragments from a self-Ag. The finding that a protein fragment mimicking such a cleavage product can elicit Abs with novel antigenic specificities suggests that altered proteolysis may contribute to the breaking of self tolerance.


   Footnotes
 
1 This work was supported by Grants ES08080, ES07511, and AR32063 from the National Institutes of Health and in part by the Sam and Rose Stein Charitable Trust’s support of the DNA Core Facility of the Department of Molecular and Experimental Medicine, The Scripps Research Institute. This is publication 12346 MEM from The Scripps Research Institute (La Jolla, CA). Back

2 Address correspondence and reprint requests to Dr. K. M. Pollard, Department of Molecular and Experimental Medicine, MEM131, 10550 North Torrey Pines Road, La Jolla, CA 92037. Back

3 Current address: Institute for Molecular Genetics, University of Heidelberg, Im Neuenheimer Fels 230, 69120 Heidelberg, Germany. Back

Received for publication October 15, 1999. Accepted for publication May 24, 2000.


   References
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 References
 

  1. Casiano, C. A., S. J. Martin, D. R. Green, E. M. Tan. 1996. Selective cleavage of nuclear autoantigens during CD95(Fas/APO-1)-mediated T cell apoptosis. J. Exp. Med. 184:765.[Abstract/Free Full Text]
  2. Casiano, C. A., R. L. Ochs, E. M. Tan. 1998. Distinct cleavage products of nuclear proteins in apoptosis and necrosis revealed by autoantibody probes. Cell Death Differ. 5:183.[Medline]
  3. Casciola-Rosen, L. A., G. J. Anhalt, A. Rosen. 1995. DNA-dependent protein kinase is one of a subset of autoantigens specifically cleaved early during apoptosis. J. Exp. Med. 182:1625.[Abstract/Free Full Text]
  4. Wyllie, A. H., J. F. R. Kerr, A. R. Currie. 1980. Cell death: the significance of apoptosis. Int. Rev. Cytol. 68:251.[Medline]
  5. Casciola-Rosen, L. A., G. Anhalt, A. Rosen. 1994. Autoantigens targeted in systemic lupus erythematosus are clustered in two populations of surface structures on apoptotic keratinocytes. J. Exp. Med. 179:1317.[Abstract/Free Full Text]
  6. Rosen, A., L. Casciola-Rosen, J. Ahearn. 1995. Novel packages of viral and self-antigens are generated during apoptosis. J. Exp. Med. 181:1557.[Abstract/Free Full Text]
  7. Tan, E. M.. 1994. Autoimmunity and apoptosis. J. Exp. Med. 179:1083.[Free Full Text]
  8. Earnshaw, W. C.. 1995. Nuclear changes in apoptosis. Curr. Opin. Cell Biol. 7:337.[Medline]
  9. Mountz, J. D., J. Wu, J. Cheng, T. Zhou. 1994. Autoimmune disease: a problem of defective apoptosis. Arthritis Rheum. 37:1415.[Medline]
  10. Cohen, P. L., R. A. Eisenberg. 1991. lpr and gld: single gene models of systemic autoimmunity and lymphoproliferative disease. Annu. Rev. Immunol. 9:243.[Medline]
  11. Trump, B. F., I. K. Berezwsky. 1998. The reactions of cells to lethal injury: oncosis and necrosis–the role of calcium. R. A. Lockshin, and Z. Zakeri, and J. L. Tilly, eds. When Cells Die 57.-96. Wiley-Liss, Inc, New York.
  12. Lanzavecchia, A.. 1995. How can cryptic epitopes trigger autoimmunity?. J. Exp. Med. 181:1945.[Free Full Text]
  13. Moudgil, K. D., E. E. Sercarz. 1998. Antigenic determinants involved in induction and propagation of autoimmunity. N. R. Rose, and I. R. Mackay, eds. The Autoimmune Diseases 45.-58. Academic, New York.
  14. Takeuchi, K., S. J. Turley, E. M. Tan, K. M. Pollard. 1995. Analysis of the autoantibody response to fibrillarin in human disease and murine models of autoimmunity. J. Immunol. 154:691.
  15. Pollard, K. M., D. K. Lee, C. A. Casiano, M. Blüthner, M. M. Johnston, E. M. Tan. 1997. The autoimmunity-inducing xenobiotic mercury interacts with the autoantigen fibrillarin and modifies its molecular and antigenic properties. J. Immunol. 158:3521.[Abstract]
  16. Dolginova, E. A., E. Roth, I. Silma, L. M. Weiner. 1992. Chemical modification of Torpedo acetylcholinesterase by disulfides: appearance of a "molten globule" state. Biochemistry 31:12248.[Medline]
  17. Griem, P., K. Panthel, H. Kalbacher, E. Gleichmann. 1996. Alteration of a model antigen by Au(III) leads to T cell sensitization to cryptic peptides. Eur. J. Immunol. 26:279.[Medline]
  18. Griem, P., C. von Vultée, K. Panthel, S. L. Best, P. J. Sadler, III. C.F. Shaw. 1998. T cell cross-reactivity to heavy metals: identical cryptic peptides may be presented from protein exposed to different metals. Eur. J. Immunol. 28:1941.[Medline]
  19. Kubicka-Muranyi, M., O. Behmer, M. Uhrberg, H. Klonowski, J. Bister, E. Gleichmann. 1993. Murine systemic autoimmune disease induced by mercuric chloride (HgCl2): Hg-specific helper T-cells react to antigen stored in macrophages. Int. J. Immunopharmacol. 15:151.[Medline]
  20. Casciola-Rosen, L. A., D. K. Miller, G. J. Anhalt, A. Rosen. 1994. Specific cleavage of the 70-kDa protein component of the U1 small nuclear ribonucloeprotein is a characteristic biochemical feature of apoptotic cell death. J. Biol. Chem. 269:30757.[Abstract/Free Full Text]
  21. Turley, S. J., E. M. Tan, K. M. Pollard. 1993. Molecular cloning and sequence analysis of U3 snoRNA-associated mouse fibrillarin. Biochim. Biophys. Acta 1216:119.[Medline]
  22. Tan, E. M., C. L. Peebles. 1993. Immunoprecipitation of labeled proteins. Manual Biol. Markers Dis. A 6:1.
  23. Pearson, D. L., R. D. Reimonenq, K. M. Pollard. 1999. Expression and purification of recombinant mouse fibrillarin. Protein Expression Purif. 17:49.[Medline]
  24. Catchpoole, D. R., B. W. Stewart. 1993. Etoposide-induced cytotoxicity in two human T-cell leukemia lines: delayed loss of membrane permeability rather than DNA fragmentation as an indicator of programmed cell death. Cancer Res. 53:4287.[Abstract/Free Full Text]
  25. Pollard, K. M., E. K. L. Chan, R. L. Rubin, E. M. Tan. 1996. Autoimmunity and autoantibodies. R. A. Meyers, ed. The Encyclopedia of Molecular Biology and Molecular Medicine 84. VCH, New York.
  26. Rubin, R.L.. 1997. Drug-induced lupus. D. J. Wallace, and B. H. Hahn, eds. Dubois’ Lupus Erythematosus 5th Ed.871.-901. Williams & Wilkens, Baltimore.
  27. Mevorach, D., J. L. Zhou, X. Song, K. B. Elkon. 1998. Systemic exposure to irradiated apoptotic cells induces autoantibody production. J. Exp. Med. 188:387.[Abstract/Free Full Text]
  28. Casiano, C. A., E. M. Tan. 1996. Antinuclear autoantibodies: probes for defining proteolytic events associated with apoptosis. Mol. Biol. Rep. 23:211.[Medline]
  29. Pollard, K. M., P. Hultman. 1997. Effects of mercury on the immune system. A. Sigel, and H. Sigel, eds. In Metal Ions in Biological Systems: Mercury and its Effects on Environment and Biology Vol. 3:421.-440. Dekker, New York.
  30. Rubin, R. L., F.-L. Tang, G. Tsay, K. M. Pollard. 1990. Pseudoautoimmunity in normal mice: antihistone antibodies elicited by immunization versus induction during graft-versus-host reaction. Clin. Immunol. Immunopathol. 54:320.[Medline]
  31. Mamula, J. M., S. Fatenejad, J. Craft. 1994. B cells process and present lupus autoantigens that initiate autoimmune T-cell responses. J. Immunol. 152:1453.[Abstract]
  32. Topfer, F., T. Gordon, J. McCluskey. 1995. Intra- and intermolecular spreading of autoimmunity involving the nuclear self antigens La (SS-B) and Ro (SS-A). Proc. Natl. Acad. Sci. USA 92:875.[Abstract/Free Full Text]
  33. Kono, D. H., D. Balomenos, D. L. Pearson, M. S. Park, B. Hildebrandt, P. Hultman, K. M. Pollard. 1998. The prototypic Th2 autoimmunity induced by mercury is dependent on IFN-{gamma} and not Th1/Th2 imbalance. J. Immunol. 161:234.[Abstract/Free Full Text]
  34. Whitekus, M. J., R. P. Santini, A. J. Rosenspire, M. J. McCabe. 1999. Protection against CD95-mediated apoptosis by inorganic mercury in jurkat T cells. J. Immunol. 162:7172.
  35. Levin, S.. 1998. Apoptosis, necrosis, or oncosis: what is your diagnosis? A report from the cell death nomenclature committee of the society of toxicologic pathologists. Toxicol. Sci. 41:155.[Free Full Text]
  36. Reiter, I., B. Krammer, G. Schwamberger. 1999. Differential effect of apoptotic versus necrotic tumor cells on macrophage antitumor activities. J. Immunol. 163:1730.[Abstract/Free Full Text]
  37. Gallucci, S., M. Lolkema, P. Matzinger. 1999. Natural adjuvants: endogenous activators of dendritic cells. Nat. Med. 5:1249.[Medline]




This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow Request Permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Pollard, K. M.
Right arrow Articles by Tan, E. M.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Pollard, K. M.
Right arrow Articles by Tan, E. M.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS