Proteolytic cleavage by caspases is the central event in cells undergoing apoptosis. Cleaved proteins are often targeted by autoantibodies, suggesting that the cleavage of self Ags enhances immunogenicity and is prone to induce an autoimmune response. We found autoantibodies that immunoprecipitated a 140-kDa RNA-associated protein, provisionally designated Pa, in 11 of 350 patient sera that were positive for antinuclear Abs in an immunofluorescence test. The Pa protein gave rise to three fragments with m.w. ranging from 120–130 kDa during anti-Fas-activated apoptosis. Pure caspase-3 cleaved the Pa protein into a 130-kDa fragment corresponding to the largest of these three products. Peptide sequence analysis of a tryptic digest from immunoaffinity-purified Pa showed 100% identity to human RNA helicase A (RHA). The identity of Pa with RHA was further confirmed by immunoblotting with rabbit anti-RHA Ab using anti-Pa immunoprecipitates as substrates. All 10 anti-RHA-positive patients who were clinically analyzed were diagnosed as having systemic lupus erythematosus, and 7 of them had lupus nephritis. RHA is a multifunctional protein with roles in cellular RNA synthesis and processing. Inactivation of RHA by cleavage may be an important part of the process leading to programmed cell death. The cleaved RHA fragments that are produced during apoptosis may trigger an autoimmune response in systemic lupus erythematosus.
The presence of autoantibodies against cellular components in patient sera is one of the important clinical characteristics of systemic lupus erythematosus (SLE) 3<;9690f3;10;ZPICKFOOT;> and other systemic rheumatic diseases. The clinical significance of these autoantibodies in rheumatic diseases lies in the distinct correlation between their specificity and clinical manifestations (1, 2). Moreover, clarification of mechanisms leading to autoantibody production could provide a key to the etiology of these diseases. Identification and characterization of molecules recognized by the autoantibodies are important steps toward elucidating these mechanisms. Recent studies have revealed that many autoantigens are cleaved during apoptosis (3, 4, 5, 6, 7, 8, 9), which suggests that inactivation by cleavage of these proteins is part of a biochemical pathway that leads cells toward apoptosis. In addition, attention has been focused recently on the possibility that apoptotic cleavage of proteins might play a role in triggering the autoimmune response (4, 5).
In the present report we describe a new autoantibody system that recognizes human RNA helicase A (RHA) and show for the first time that this enzyme is cleaved by caspase-3 as an early event in apoptosis. RHA is a nuclear helicase that unwinds dsRNA and RNA:DNA duplexes in a reaction driven by hydrolysis of ATP (10, 11). RHA is identical over most of its length with the independently reported DNA helicase II, which unwinds both dsRNA and dsDNA (12, 13, 14, 15). RHA mediates molecular interactions between RNA polymerase II and the CREB binding protein (CBP) and is required for activation of transcription in response to cAMP (16). RHA also links the breast cancer-specific tumor suppressor protein, BRCA1, to the RNA polymerase II holoenzyme (17). Interestingly, RHA is a human homologue of the Drosophila maleless protein (MLE), which increases expression of X-linked genes in male flies (11). The homology with MLE indicates that RHA may play an evolutionarily conserved role in transcriptional regulation. Study of the RHA knockout mouse has revealed that homozygosity for the null allele leads to embryonic lethality (18). Recently, it has been shown that RHA promotes the export of partially spliced or unspliced RNA of certain SIV and HIV retroviruses (19, 20). These reports suggest that RHA is associated with RNA processing as well as transcription. Thus, RHA is a multifunctional protein that plays an important role in cellular RNA biogenesis and metabolism.
The present results suggest that RHA may be a useful model for investigation of the link among apoptotic cell death, RNA synthesis and processing, and initiation of specific humoral autoimmune responses.
Materials and Methods
Patient sera were obtained from the Antinuclear Antibody Laboratory of the University of Missouri (Columbia, MO). Immunoprecipitation using [35S]methionine-labeled HeLa cell extracts was performed on 350 sera that were positive for antinuclear Abs in a fluorescent antinuclear Ab (FANA) test (21) at a dilution greater than 1/160.
Protein immunoprecipitation was performed as described previously (22). For each protein immunoprecipitation, 3 × 106 HeLa cells were cultured with 10 μCi of [35S]methionine (translation grade; NEN Life Science Products, Boston, MA) overnight in methionine-free RPMI 1640 medium containing 10% FBS. The labeled cells were sonicated in 100 μl of IPP buffer (10 mM Tris-HCl (pH 7.4), 0.5 M NaCl, and 0.1% Nonidet P-40). The lysates were centrifuged at 12,000 × g for 30 min in the microcentrifuge (Eppendorf, Westbury, NY), and the supernatants were incubated with protein A-Sepharose CL-4B (Pharmacia Biotech, Piscataway, NJ) that had been preincubated with patient serum (3 μl with 3 mg of protein A beads). After a 3-h incubation at 4°C, the beads were washed with IPP buffer three times, and proteins bound to the beads were eluted with 30 μl of 2× SDS-sample buffer (0.125 M Tris-HCl (pH 6.8), 20% glycerol, 4% SDS, 5% 2-ME, and 0.005% bromophenol blue). The proteins were separated by SDS-PAGE and analyzed by autoradiography.
Protein immunoprecipitation was also performed with sucrose density gradient fractions of 35S-labeled HeLa cell extracts. For these experiments, HeLa cells were labeled with [35S]methionine (10 μCi/3 × 106 cells) overnight, then sonicated in IPP buffer (2 × 107 cells/100 μl). After centrifugation of the lysate, the supernatant was incubated for 30 min at 37°C with DNase I (Pharmacia Biotech; final concentration, 20 U/ml) or RNase A (Roche Molecular Biochemicals, Indianapolis, IN; final concentration, 100 μg/ml). The RNase A was boiled for 15 min before use to remove DNase and protease contamination. After digestion, the cell extracts were clarified by centrifugation, and 500 μl of supernatant was layered onto 4.2 ml of 20–60% sucrose density gradient in 10 mM Tris-HCl (pH 7.5), 100 mM NaCl, and 2.5 mM MgCl2. The sucrose density gradient sedimentation was performed at 100,000 × g for 16 h at 4°C. After ultracentrifugation, the tube was punctured on the bottom, and fractions were collected into nine tubes. Each fraction was incubated with protein A beads preincubated with patient serum, and immunoprecipitation was conducted as described above.
For nucleic acid immunoprecipitations, unlabeled HeLa cells (1 × 107 cells/sample) were sonicated in NET-2 buffer (50 mM Tris-HCl (pH 7.4), 0.15 M NaCl, and 0.05% Nonidet P-40). The lysates were centrifuged, and the supernatants were incubated with protein A beads that had been preincubated with patient sera as described previously. After incubation, the beads were washed with NET-2 three times, and nucleic acids were extracted with phenol-chloroform-isoamyl alcohol (25/24/1, v/v/v), followed by precipitation with 100% ethanol. Precipitated nucleic acids were digested by DNase I (final concentration, 20 U/ml) or RNase A (final concentration, 50 μg/ml) for 30 min at 37°C, again extracted with phenol-chloroform-isoamyl alcohol, and then precipitated with ethanol. The precipitated nucleic acids were electrophoresed in a 7 M urea-6% polyacrylamide gel and were visualized by silver staining (Silver Stain Plus Kit, Bio-Rad, Hercules, CA).
Immunoblotting using immunoprecipitated samples as substrates
Unlabeled HeLa cell extracts (1 × 107 cells in 500 μl of IPP buffer) were incubated with Ab-bound protein A beads (3 μl of serum/3 mg of protein A). After washing with IPP buffer, precipitated proteins were eluted from the beads with SDS-sample buffer. Immunoblotting was performed as described previously (22, 23) using the eluted proteins as substrates. Briefly, the proteins were separated by SDS-PAGE and transferred to a nitrocellulose membrane. The membrane was blocked with 3% BSA, then incubated with primary Abs (patient serum or immunized rabbit serum) at the dilutions indicated in the figure legends. After incubation with primary Abs, the membrane was incubated with alkaline phosphatase-conjugated secondary Abs (anti-human IgG or anti-rabbit IgG), and the bands were visualized by incubation of the membrane with nitro blue tetrazolium and 5-bromo-4-chloro-3-indolyl phosphate (Sigma, St. Louis, MO) according to the manufacturer’s instructions.
Induction of apoptosis
Jurkat cells were used for induction of apoptosis by anti-Fas Ab, since they have been shown to express Fas on the cell surface (24). Jurkat cells were labeled with [35S]methionine overnight. After labeling, the medium was changed to RPMI 1640 complete medium containing 10% FBS. Anti-human Fas mouse mAb (clone CH11, Upstate Biotechnology, Lake Placid, NY) was added at a final concentration of 100 ng/ml. Cells were collected before adding anti-Fas and at 1, 3, 6, and 24 h of incubation after adding anti-Fas. Extracts of the cells collected were used for protein immunoprecipitation.
In vitro cleavage by caspase-3
[35S]Methionine-labeled Jurkat cell extracts were incubated with protein A beads that had been preincubated with patient serum as described above. After washing with IPP buffer three times, the beads were further washed with caspase reaction buffer (50 mM PIPES (pH 6.5) and 2 mM EDTA) three times. Twenty microliters of reaction buffer containing caspase-3 (Upstate Biotechnology) and DTT (final concentration, 5 mM) were added to the beads and incubated for 1 h at 37°C. The reaction was stopped by adding 20 μl of SDS-sample buffer. In some reactions, caspase-3 inhibitor (DEVD-CHO, Calbiochem, San Diego, CA) or caspase-1 inhibitor (YVAD-CHO, Calbiochem) was also added.
Unlabeled HeLa cell extracts (3 × 107cells in 500 μl of IPP buffer) were incubated with 10 mg of protein A beads that had been preincubated with 10 μl of patient serum. After a 3-h incubation, the beads were washed three times with IPP buffer, and 30 μl of SDS-sample buffer was added. Immunoprecipitated proteins were separated in SDS-PAGE and were visualized by Coomassie blue staining. Selected protein was excised for trypsin digestion. Trypsin digests were fractionated by HPLC, and a selected peak fraction was subjected to Edman degradation.
Immunoprecipitation using 35S-labeled HeLa cell extracts
Immunoprecipitation using 35S-labeled HeLa cell extracts was performed with 350 sera that were positive for antinuclear Abs in a FANA test at a titer of greater than 1/160. Among those tested, 11 sera precipitated a protein that migrated at a position of 140 kDa in SDS-PAGE (Fig. 1). All these sera showed a nuclear speckled pattern in immunofluorescent staining of cells (data not shown). Because the 140-kDa protein did not appear to be a known autoantigen, we provisionally named it Pa based on the identifier of the prototype serum.
Sera positive for anti-Pa autoantibodies immunoprecipitate RNAs
To test whether the anti-Pa-positive sera precipitate nucleic acids, immunoprecipitation was performed using unlabeled HeLa cell extracts, and nucleic acids were isolated from the immune complexes. Denaturing gel electrophoresis with silver staining revealed that all the anti-Pa-positive sera precipitated a large, heterogeneous set of nucleic acids (Fig. 2, A and B, lanes marked (−)). To determine whether these were DNA or RNA, the eluted samples were subjected to DNase or RNase digestion (Fig. 2). As shown in Fig. 2,A, the nucleic acids precipitated with anti-Pa Abs were resistant to DNase digestion. As a positive control, we used a sample immunoprecipitated with anti-Ku-positive serum. Because Ku autoantigen is a DNA binding protein (25), nucleic acids extracted from anti-Ku immunoprecipitates should contain DNA, and indeed, this sample was DNase sensitive (Fig. 2,A, Ku(+)). By contrast, samples from anti-Pa precipitates were sensitive to RNase, while the sample from anti-Ku immunoprecipitate was resistant (Fig. 2 B). These results suggest that Pa is an RNA binding protein. However, we cannot rule out the possibility that immunoprecipitation was attributable to anti-RNA Abs or Abs against other RNA binding proteins present in the anti-Pa-positive sera.
The Pa autoantigen is associated with RNAs
To investigate whether the Pa autoantigen is directly associated with RNA, we compared the sedimentation rate of Pa autoantigen in sucrose density gradients using cell extracts that had been digested with DNase or RNase (Fig. 3). In DNase-treated cell extracts the Pa autoantigen migrated as a large heterogeneous complex throughout the gradient (Fig. 3,A). The pattern was the same for untreated extracts (data not shown). In RNase-treated extracts, the Pa autoantigen was found in a sharp peak near the top of the gradient (Fig. 3 B). It is most likely that this difference in sedimentation pattern is due to digestion of RNAs associated with the Pa protein. Therefore, these results indicate that Pa autoantigen has a molecular interaction with RNA.
The Pa autoantigen is cleaved during apoptosis
Recently, a number of autoantigens have been shown to be cleaved during apoptosis (3, 4, 5, 6, 7, 8, 9), and it is possible that the resulting products contribute to the presentation of immunocryptic epitopes to induce an autoimmune response (4). Therefore, we were interested in learning whether Pa is cleaved during apoptosis. After labeling Jurkat cells with [35S]methionine, apoptosis was induced by the anti-human Fas mAb CH11 (Fig. 4). Immunoprecipitation with anti-Pa Abs showed that the Pa autoantigen was cleaved into several fragments of approximately 120–130 kDa during apoptosis (Fig. 4,A). In time-course studies, the larger 130-kDa fragment appeared first (labeled F1 in Fig. 4,A), followed by the appearance of two smaller fragments after 3–6 h (labeled F2 and F3). Almost 100% of the Pa protein was cleaved by 6 h. Fig. 4 B shows the result of a control immunoprecipitation using patient serum containing Abs against the catalytic subunit of DNA-dependent protein kinase (DNA-PKcs) and Ku. DNA-PKcs is known to be cleaved into approximately 250- and 165-kDa fragments during apoptosis (4, 5). As described previously, DNA-PKcs was cleaved during anti-Fas-induced apoptosis. This serum mainly immunoprecipitated the 165-kDa fragment of DNA-PKcs from anti-Fas-induced apoptotic cell extracts. Ku70 and Ku80 were not cleaved under these conditions, as expected.
The Pa autoantigen is cleaved by caspase-3
A number of autoantigens that are cleaved during apoptosis are targets of the caspase-3 protease (5, 6, 7, 8, 26). This protease cleaves DNA-PKcs, the 70-kDa protein of U1 small nuclear ribonucleoprotein, poly(ADP-ribose) polymerase, and DNA topoisomerase I (5, 6, 7, 8, 26). As shown in Fig. 4, the rates of Pa and DNA-PKcs cleavage during anti-Fas-activated apoptosis were similar. Therefore, we tested whether caspase-3 also cleaves Pa autoantigen (Fig. 5,A). Incubation of Pa autoantigen with 1.5 U/μl of caspase-3 resulted in generation of a 130-kDa cleaved fragment. This cleavage was completely inhibited by 80 nM caspase-3 inhibitor peptide (DEVD-CHO), whereas it was not inhibited by a 5 times greater concentration (400 nM) of caspase-1 inhibitor peptide (YVAD-CHO). Fig. 5 B compares the position of the fragment produced by in vitro caspase-3 digestion and those of in vivo cleavage products of Pa autoantigen during apoptosis. The position of the fragment produced by in vitro caspase-3 digestion matched the position of F1, the largest in vivo cleavage product. One faint band could be seen in the caspase-3 digest at a position of approximately 120 kDa. This band may correspond to F3 in apoptotic cells. These results suggest that caspase-3 contributes to cleavage of Pa autoantigen in the early stages of apoptosis. It is uncertain, however, whether F2 and F3 can be produced by further caspase-3 digestion or whether other proteases are involved in these cleavages.
Amino acid sequence of a tryptic peptide from Pa autoantigen matches a sequence in human RNA helicase A
To determine the molecular identity of the Pa protein, we determined the amino acid sequence of a peptide isolated from a tryptic digest of immunoaffinity-purified material. The amino acid sequence, DVVQAYPEVR, showed 100% identity to the sequence of human RHA at amino acid positions 529–538 (11). Because the amino acid preceding position 529 in RHA is arginine (Fig. 6), theoretically one would expect trypsin cleavage at this position, as we observed for Pa.
Rabbit Ab against RNA helicase A recognizes Pa autoantigen
To provide further proof that the Pa autoantigen is identical with human RHA, we performed immunoblotting using immunoprecipitates of anti-Pa Abs probed with rabbit anti-human RHA serum at a dilution 1/20,000 (Fig. 7). Rabbit anti-human RHA Abs recognized Pa autoantigen, whereas normal rabbit serum did not. Rabbit anti-RHA Ab did not show any bands in immunoblotting using immunoprecipitates of anti-Sm-positive serum and normal human serum. These results confirm that Pa autoantigen is identical with human RHA.
Autoantibodies against RNA helicase A are associated with SLE
The clinical significance of autoantibodies against RHA was studied. We obtained clinical information from 10 anti-RHA-positive patients; all 10 patients were diagnosed as having SLE (27) (Table I). Of these 10 patients, seven patients had renal involvement, including five with biopsy-proven lupus nephritis, one with proteinuria, and one with renal failure.
|Patient .||Diagnosis .||Renal Involvement .|
|1||SLE||WHOa class IV nephritis|
|3||SLE||WHO class III nephritis|
|4||SLE||WHO class III, IV, VI nephritis|
|5||SLE||WHO class II nephritis|
|7||SLE||WHO class IV nephritis|
|Patient .||Diagnosis .||Renal Involvement .|
|1||SLE||WHOa class IV nephritis|
|3||SLE||WHO class III nephritis|
|4||SLE||WHO class III, IV, VI nephritis|
|5||SLE||WHO class II nephritis|
|7||SLE||WHO class IV nephritis|
WHO, World Health Organization.
In this study we describe a novel autoantigen targeted by sera from patients with SLE. This autoantigen is cleaved during apoptosis. Sequencing of a tryptic peptide, m.w., an experimentally demonstrated RNA binding property, and immunological cross-reactivity reveal that this autoantigen is identical with human RHA. This study, to our knowledge, is the first to report that RHA is cleaved during apoptosis. Our results show that RHA is cleaved into at least three fragments (F1, F2, and F3 in Figs. 4 and 5) within 6 h after anti-Fas activation. The largest band, F1, appears in an early stage of apoptosis, and the size of this fragment matches the size of the in vitro product after digestion of RHA with caspase-3. These results suggest that caspase-3 contributes to the cleavage of RHA in an early stage of apoptosis. It is possible that other proteases participate in the production of F2 and F3, which appear later in apoptosis. Indeed, a recent study revealed the possibility that both caspase-3 and caspase-6 are involving in cleavage of DNA topoisomerase I during apoptosis (26). Further studies are needed to determine the entire array of proteases that contribute to cleavage of RHA during apoptosis.
Cleavage of RHA into the 120- to 130-kDa fragments involves the release of 90–180 aa (10–20 kDa) from the RHA molecule. RHA has a DTPD sequence at aa position 93–96 that corresponds to the consensus cleavage motif for caspase-3, DXXD (6). If caspase-3 cleaves at this site, 96 aa at the N-terminus would be removed during apoptosis, which is compatible with the cleavage pattern seen experimentally. The DXXD sequence is found in three other sites in RHA: DKDD at aa positions 594–597, DALD at positions 839–842, and DAND at positions 842–845. These sites theoretically give rise to fragments of about 75 and 65 kDa or 95 and 45 kDa, respectively, which are not seen experimentally.
RHA is one of several proteins that are involved in RNA synthesis and processing and have been shown to undergo proteolytic cleavage during apoptosis (3, 8, 28, 29). Functional inactivation of RHA in apoptotic cells may be important, either directly or indirectly, for the process of apoptosis. According to studies on the domain structure of human RHA (15), the first 100 aa at the amino-terminal end contain one of the two dsRNA binding domains (30). Loss of this region significantly decreases dsRNA binding affinity (15). A study of the molecular interaction between RHA and CBP revealed that a fragment of RHA containing aa 1–250 binds to CBP (16). Therefore, cleavage of the N-terminal 96 aa might be expected to disrupt both dsRNA binding and CBP binding. Loss of these functions at the early stage of apoptosis might contribute to the process of programmed cell death. Detection of precise cleavage sites in RHA during apoptosis and biochemical characterization of the cleaved fragments will be important to obtain a better understanding of the mechanisms by which this cleavage may contribute to programmed cell death. These investigations are in progress.
Recently, a number of autoantigens have been found to be cleaved during apoptosis (3, 4, 5, 6, 8, 9). It has been shown that dendritic cells have the ability to take up apoptotic cells, and they efficiently process and present peptides from proteins inside the apoptotic cells on their MHC class I and II molecules (31). Thus, dendritic cells might be the main APC that participate in processing and presentation of cleaved autoantigens in apoptotic cells. Due to changes in their conformation, it is possible that these cleaved molecules are processed in a different manner than the intact molecules in the APC. Possibly, this leads to presentation of immunocryptic epitopes that stimulate autoreactive T cells, followed by activation of autoantibody-producing B cells (32, 33, 34, 35, 36, 37, 38, 39). It is interesting that most autoantigens that are cleaved during apoptosis are normally, like RHA, structurally associated with nucleic acids. These autoantigens could lose their ability to bind to nucleic acids as a result of apoptotic cleavage. This may in itself lead to altered processing patterns in the APC. In this respect, it will be intriguing to test whether RHA loses its RNA-binding ability when it is cleaved during apoptosis.
The association of linked sets of autoantibodies with a particular disease or symptom in systemic rheumatic diseases has been well established (1). In this study we screened 350 FANA-positive patient sera that were selected solely on the basis of having an Ab titer greater than 1/160 in the FANA test. Therefore, patients with various rheumatic conditions were included in this group. Although the initial group was clinically heterogeneous, all 10 of the anti-RHA Ab-positive patients who were tested for clinical manifestation were classified as having SLE. Of these, seven had renal disorders associated with SLE. This homogeneous clinical manifestation strongly suggests that anti-RHA Abs are associated with SLE, especially with lupus nephritis. Thus, the presence of anti-RHA autoantibodies in patient sera might be clinically useful for diagnosis of SLE and for the prediction of the development of lupus nephritis. To confirm the association of anti-RHA autoantibodies with SLE and to determine more precisely the frequency of this autoantibody system, a more systematic clinical study will be needed.
We are grateful to Dr. J. Hurwitz (Sloan-Kettering Institute) and Dr. C.-G. Lee (University of Medicine and Dentistry of New Jersey) for the generous gift of rabbit antiserum against human RNA helicase A. We are indebted to The W. M. Keck Foundation Biotechnology Resource Laboratory at Yale University for their excellent technical help in peptide sequencing. We thank Dr. K. McConnell and colleagues in the Program in Gene Regulation, Institute of Molecular Medicine and Genetics, for invaluable advice. We also thank Dr. Rhea Markowitz for a critical reading of the manuscript.
This work was supported by a grant from Medical College of Georgia Research Institute.
Abbreviations used in this paper: SLE, systemic lupus erythematosus; RHA, RNA helicase A; CREB, cAMP responsive element binding protein; CBP, CREB-binding protein; MLE, maleless protein; FANA, fluorescent antinuclear Ab; DNA-PKcs, catalytic subunit of DNA-dependent protein kinase.