| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
,
,,
,,||
* Institute of Immunology, Medical Faculty Carl Gustav Carus, Technical University Dresden, Dresden, Germany;
Arthritis and Immunology Program, Oklahoma Medical Research Foundation, and
University of Oklahoma Health Sciences Center, Oklahoma City, OK 73104;
New York University School of Medicine, Hospital for Joint Diseases, New York, NY 10003;
¶
Medical and Health Science Center, Third Department of Medicine, University of Debrecen, Debrecen, Hungary; and
||
U.S. Department of Veterans Affairs Medical Center, Oklahoma City, OK 73104
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionDisclosuresReferences |
|---|
30% of sera from anti-La-positive patients, we expected that mutant La mRNAs circumvent RNA control and the expression of mutant La protein could become harmful. Indeed, real-time PCR, immunostaining, and immunoblotting data of mice transgenic for the mutant La form show that mutant La mRNAs are not repressed in these animals and are translated to mutant La protein. In addition to the mutant La protein, we detected a minor portion of native human La in the mutant La-transgenic mice. Therefore, ribosomal frame shifting may allow the mutant La mRNA to escape from RNA control. Interestingly, expression of the mutant La mRNA results in a lupus-like disease in the experimental mice. Consequently, escape of mutant La mRNA from RNA control can have two effects: it 1) results in the expression of an immunogenic (neo)epitope, and 2) predisposes to autoimmunity. |
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionDisclosuresReferences |
|---|
-thalassemias (3, 4, 5). The core NMD machinery is phylogenetically conserved from yeast to humans. Depletion of upf1 in mice–a factor which is assumed to be involved in the recognition of a PTC–leads to embryonic death at the implantation stage (6). This implies the fundamental and vital role of the NMD mechanism.
At the molecular level, NMD is not yet fully understood. According to recent models, so-called exon junction complexes (EJCs) assemble at the splice junctions where they remain even after transport to the cytoplasm. During the first round of translation, these EJCs will be removed. If a mRNA contains a PTC, the ribosome does not pass all splice junctions and the EJCs downstream of the PTC will remain associated with the mutant mRNA. mRNAs containing EJCs after the initial round of translation will be degraded.
In previous studies, we prepared an expression cDNA library constructed from PBMC mRNA taken from an autoimmune patient (index patient) with systemic lupus erythematosus (SLE) (7, 8, 9, 10). The cDNA library was screened with the patient’s own autoimmune serum which contained predominantly Abs to the autoantigen La/SS-B. During screening of the patient’s cDNA library, we identified a mutant La cDNA. Genomic DNA analysis of serial blood drawings of the index patient shows that the mutation persists for more than a decade (M. P. Bachmann, unpublished data). The mutation occurred in a region sensitive for mutations and can be found in >15% of anti-La positive autoimmune patients (I. Semsei, unpublished observation).
Native La protein is thought to be involved in essential housekeeping functions. These functions include a role in transcription/termination of RNA synthesized by polymerase III, 3'-RNA processing, and nuclear RNA import and retention (11, 12, 13, 14, 15, 16, 17). Furthermore, this autoantigen may participate in stabilization of mRNAs and translation of viral and mitogen-stimulated mRNAs (18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29). e.g., La protein was shown to be involved in the cap-independent translation of the mRNAs encoding the X-linked inhibitor of apoptosis and the double minute gene product mdm2 which is a regulator of the stability of p53 (30, 31, 32). Thus, an impaired expression of La protein could even have a direct effect on the survival of autoreactive cells and could, thereby, contribute to the development of autoimmunity.
Due to the presence of the PTC in the mutant La mRNA, however, the mutant La mRNA should be eliminated by the NMD mechanism and should not be translated to protein. Consequently, it should not become a target of an immune response. However, in this study, we report that the serum of this patient and 30% of anti-La-positive sera of SLE patients in general contain Abs to a mutant La-specific neoepitope suggesting that mutant La mRNAs were transcribed in these patients, circumvented NMD, were translated to protein, and became a target of an immune response. To analyze how mutant La mRNAs escape NMD, we established mice transgenic for the mutant form of La. We show that mutant La mRNAs are not only repressed and translated to mutant La protein, but, unexpectedly, expression of mutant La favors the development of systemic autoimmunity.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionDisclosuresReferences |
|---|
The serum samples from patients with SLE or primary Sjögren’s syndrome-positive for IgG Abs to La protein were obtained from a routine autoantibody testing laboratory (Institute of Immunology, University of Dresden, Dresden, Germany). In addition, we used serum samples from mothers of children with either cardiac or cutaneous manifestations of neonatal lupus (ages 20–50 years) that were obtained from the Research Registry for neonatal lupus (33). The sera which were originally characterized as La positive by commercial routine ELISAs and immunoblotting were re-evaluated by ELISA and immunoblotting using recombinantly expressed human La protein (see also Proteins and immunoassays). Age- and sex-matched control sera from healthy individuals and anti-La-negative sera from patients with SLE or scleroderma were acquired from the Oklahoma Medical Research Foundation (OMRF) Serum Repository and the European Routine Autoantibody Testing Laboratory (Institute of Immunology, University of Dresden). The OMRF Institutional Review Board approved the use of all specimens in compliance with National Institutes of Health guidelines. As mAbs, we used the anti-La mAbs SW5, 4B6, and 3B9. SW5 is a well-established anti-La mAb which recognizes the N-domain of human La protein but not mouse La protein (15). The anti-La mAb 4B6 was described previously and shown to be directed to the C-domain of human La (34). The anti-La mAb 3B9 was established previously in our lab and recognizes both human and mouse La protein (M. P. Bachmann, unpublished data).
Mice
We housed all mice in specific pathogen-free conditions, and the OMRF Institutional Animal Care and Use Committee approved all mouse-related studies. Mice transgenic for the wild-type human La gene including its natural promoter have been described (35), and we used them as heterozygotes at the 
F12 backcross generation. We genotyped these mice as previously described (36). Mutant La-transgenic mice were generated as follows. We initiated the creation of the mutant La transgene with a construct containing the native human La gene that we had also used to establish the mouse 3T3 cell line expressing native human La (8). First, we deleted an AvrII fragment from the native La genomic clone. The remaining construct contained a unique KpnI site in exon 3 and a unique BstEII site in exon 10 of La. We replaced the KpnI/BstEII fragment from this genomic native La construct with the respective mutant La cDNA fragment and then restored the AvrII fragment in the AvrII site. To prepare linearized DNA that was devoid of a bacterially derived sequence for FVB/N-fertilized oocyte injection, we released the mutant La gene from the vector with XhoI. We identified four positive FVB/N founder mice (one male and three females) by PCR analysis of tail DNA from a total of 81 animals and crossed them with FVB/N mice obtained from The Jackson Laboratory. The male founder was used to establish the mouse strain M1. Two of the female mice did not transfer the transgene to their offspring. The remaining female founder produced a single positive F1 mouse. This mouse was used to establish the mouse strain M2. We checked the presence of the transgene by PCR using three primer pairs. The primer pairs were specific for regions either inside or at the 5'- or 3'-end of the human La gene (data not shown). We confirmed the presence of the transgene by Southern blotting (data not shown) and, ultimately, by analysis of protein expression.
Immunostaining and confocal microscopy
For immunohistochemistry, tissue samples (0.25 cm2) were extracted from mice following CO2 euthanasia, fixed overnight in chilled 2% paraformaldehyde in PBS, then allowed to equilibrate in 30% sucrose at 4°C. Sucrose-infiltrated samples were cryopreserved in Tissue Freezing Medium (Triangle Biomedical Sciences) and 5-µm serial sections were cut and placed on positively charged slides. Slides were allowed to come to room temperature, then sections were treated 30 min with 0.3% hydrogen peroxide to block endogenous peroxidase activity. Sections were stained using the Histomouse SP DAB staining kit (Zymed Laboratories) according to manufacturer’s suggested guidelines. Briefly, sections were blocked for 40 min with BEAT blocking reagent, then rinsed with water and PBS. Either PBS or a 1/50 dilution of the anti-human La Ab SW5 was added to the sections and incubated at room temperature in a humidified chamber for 1 h. After rinsing with PBS, sections were incubated for 10 min with biotinylated enzyme conjugate, developed with the substrate-chromagen mixture for 5 min, and rinsed with distilled water. Slides were counterstained with hematoxylin for 1 min, rinsed thoroughly in tap water followed by 30 s in PBS then distilled water. Sections were dehydrated through a graded alcohol series and three changes of xylene and coverslipped. Slides were evaluated on the Zeiss Axioplan 2i light microscope (Carl Zeiss Microimaging) at x63 and x100 magnifications and photodocumented.
For detection of immune complexes in kidney sections, kidneys were fixed in chilled 2% paraformaldehyde in PBS and equilibrated in 30% sucrose in PBS before cryopreservation. We then blocked 5-µm sections with 5% powdered milk in PBS and subsequently stained with optimized dilutions of primary conjugates of Alexa Fluor 546 rabbit anti-mouse IgG (H + L) (Molecular Probes) and goat anti-mouse complement C3-FITC (ICN-Cappel) in PBS. We rinsed the samples twice with PBS containing 0.1% Triton X-100, twice with PBS alone, then incubated with an optimized dilution of 4',6-diamidino-2-phenylindole (DAPI; Sigma-Aldrich), and rinsed twice in PBS. We then visualized fluorescence using a Zeiss LSM 510 Meta Confocal Microscope (Carl Zeiss).
Proteins and immunoassays
For immunoassays, we expressed the recombinant 6x histidine-tagged human La from a pET3d vector. The C-domain of La protein and the N-domain of La either lacking or containing the neoepitope sequence were cloned in a pET28a vector and also expressed as 6x his-tagged fusion proteins (34). Bovine Ro was obtained from K. Clark (Immunovision, Springdale, AR), BSA was purchased from Sigma-Aldrich. The University of Oklahoma Health Sciences Center Molecular Biology Core Facility prepared synthetic peptides as eight-branched multiantigenic peptide (neoepitope: AKKMKKENKIKWKLN) and purified them by HPLC. Screening of human and mouse sera for Abs to extractable nuclear autoantigens (Sm, RNP, Ro, La, Scl-70, Jo-1) was performed using commercial ELISA kits such as the RELISA multiparameter screening test (Immuno Concepts). For further evaluation, we coated ELISA plates with 50 µl of the respective purified native or recombinant Ag including La, Ro, Sm, RNP, Scl-70, Jo-1 (5 µg/ml in 0.15 M Na2CO3, 0.35 M NaHCO3 carbonate buffer (pH 9.6)) or synthetic peptides (1 µg/ml in carbonate buffer) for 2 h at room temperature or overnight at 4°C, washed them briefly with 0.1% Tween 20 in PBS, then blocked them with 0.1% gelatin in PBS (pH 7.4), overnight at 4°C. After washing with 0.1% Tween 20 in PBS, we added 50 µl of sera diluted (1/1000) in PBS and incubated the plates for 2 h at room temperature. We then washed the wells with 0.1% Tween 20 in PBS, incubated the wells with alkaline phosphatase-labeled anti-human (Jackson ImmunoResearch Laboratories) or anti-mouse IgG (Sigma-Aldrich), and detected bound Abs by addition of the substrate p-nitrophenylphosphate. As positive controls, we analyzed monoclonal (if available) and patient Abs to La, NeoLa, Ro, RNP, and Sm in parallel on each ELISA plate. The absorbance was measured at A410.
Serum samples for Abs to dsDNA were assayed with a commercial ELISA kit (cat. no. 5100; Alpha Diagnostic International). This kit was specifically developed for detecting anti-dsDNA Abs in mouse serum. According to the manufacturer, serum samples known to be positive for anti-extractable nuclear Abs, anti-ssDNA, anti-rheumatoid factor, anti-Toxoplasma gondii IgG, IgM, and anti-CMV IgG are negative.
Immunoprecipitation and SDS-PAGE/immunoblotting
For immunoprecipitation of total extracts, we used the immunoprecipitation system from NatuTec. Immunoprecipitated proteins were subjected to SDS/PAGE immunoblotting. The NatuTec system uses a secondary anti-mouse Ab which is directed to a conformational epitope that is destroyed under reducing conditions. Consequently, heating of the immunoprecipitated material before electrophoresis under reducing conditions destroys the ability to detect coprecipitated mouse Igs. Blots were qualitatively and quantitatively evaluated using the ECL system (Amersham Biosciences) and the CHEMI DOC XRS system (Bio-Rad).
Quantitative RT-PCR
The expression of human exon 1, exon1', 1'', mouse exon 1a,b,c La mRNA, and actin mRNA was analyzed by a light cycler-based PCR assay. The relative mRNA quantity was determined applying a real-time PCR protocol based on SYBR Green I detection (LC-FastStart DNA Master SYBR Green I; Roche Diagnostics). Serial dilutions of plasmid DNA containing the respective specific human or mouse La mRNA or actin fragment over seven log scales (102–108 molecules/capillary) were used as internal template standards (calculation via LC quantification software version 3.5; Roche Diagnostics). Each determination was conducted twice for each cDNA sample as independent PCR runs. The mean values for the La mRNA forms were normalized to the actin control. The assays were first established on normalized commercial human (I, II, fetal, and blood fractions) and mouse cDNA panels (I and II; all from BD Biosciences/BD Clontech). After subcloning of the PCR products in pGEM-Teasy, the specificity of the PCRs was determined by sequencing. Finally, mRNAs were isolated from shock-frozen tissues of native human La-, mutant human La-, and nontransgenic mice. mRNA isolation and cDNA synthesis was performed as described previously (15). As forward primer, we used for amplification of human exon 1 La (5'-GGAGTCGTTGCTGTTGCTGTTTGTG), human exon 1' La (5'-GGGGTAAACGCCGGAGGGTTC), mouse exon 1a,b La (5'-GGAAGTCCAGGCGCTTCTGTCG), mouse exon 1c La (5'-GGGAAACCTGTAAGGT TAGGAATTC), and actin (5'-GCCGTCTTCCCCTCCATCGTG) mRNA. As reverse primer, we used for amplification of human exon 1 and 1' La (5'-CACTGATTTCCATGAGTTCTGCCTTGG), mouse exon 1a,b/c La (5'-GGGAGTGGTTGCTTGGTGATCTT), and actin (5'-GGAGCCACACGCAGCTCATTGTAGA) mRNA. The PCR protocol consisted of a predenaturation step (10 min at 95°C) and 40 amplification cycles (15 s at 95°C for all samples; for human exon 1 La: 5 s at 74°C; for human exon 1' La and actin mRNA: 5 s at 70°C; for mouse exon 1a,b La mRNA: 5 s at 68°C; for mouse exon 1c La mRNA: 5 s at 64°C; 14 s at 72°C for all samples).
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionDisclosuresReferences |
|---|
|
As mentioned above, downstream of the mutation, the mutant La amino acid sequence MKKENKIKWKLN (Fig. 1b, muLa, aa 193–204) differs from the native La amino acid sequence EERKQNKVEAK (Fig. 1b, nLa, aa 193–203). Full-length La protein, the native 29-kDa form of the N-domain of La (Fig. 2a: LaN; aa 1–203), the mutant form of the N-domain of La (Fig. 2a: LaN-Neo; aa 1–204), and the native 25-kDa C-domain of La (Fig. 2a, LaC; aa 204–408) were recombinantly expressed and blotted against serum samples from either healthy donors (50 samples) or anti-La-negative (25 samples) or anti-La-positive autoimmune patients (100 samples). The respective anti-La Abs were visualized with anti-human IgG specific secondary Abs which did not cross-react with human IgM Abs. Approximately15% of the anti-La-positive sera reacted with LaN-Neo but not with LaN, suggesting a specific reactivity to the La neoepitope in these serum samples. One example is shown in Fig. 2a (lane 4). In addition to the anti-LaN-Neo reactivity, some sera reacted with the C-domain of La (e.g., in Fig. 2a, lane 9). Of most interest are the results obtained for the serum samples shown in Fig. 2a (lanes 4–6). These serum samples represent serial drawings from the same patient. The second sample (Fig. 2a, lane 5) was taken 6 mo, the third sample (Fig. 2a, lane 6) 1 year, after the first drawing (Fig. 2a, lane 4). These immunoblotting data show that the anti-mutant La response preceded the anti-La response in this patient.
|
Taken together, our data show that 1) a mutant La response can precede an anti-La response, 2) Abs to mutant La can be detected in 30% of anti-La-positive autoimmune patients, and 3) mutant La can become a target of an immune response.
Development of two independent mouse lines transgenic for a mutant human La gene
The expression of mutant La protein requires that mutant La mRNA is capable of escaping NMD. To provide experimental evidence for an escape of mutant La mRNA from NMD, we established mice transgenic for mutant human La. The genomic construct encoding mutant human La was cloned as schematically summarized in Fig. 3a and described in detail in Materials and Methods. In summary, we replaced a KpnI/BsTEII fragment in a reconstructed human La gene clone with the respective KpnI/BsTEII fragment from a mutant La cDNA clone. The mutant exon7 was part of this KpnI/BsTEII fragment.
|
Two independent transgenic mouse lines (M1 and M2) were established from the mutant La gene construct (Fig. 3a). The integration of the transgene was detected by PCR and Southern blotting (data not shown).
Analysis of expression of mutant La at the RNA level in mutant La-transgenic mice
Real-time PCR assays were established allowing us to measure the copy numbers of human and mouse La mRNAs. In previous studies, we had shown that three La mRNA isoforms can be transcribed from the human La gene which we termed exon 1, exon 1', and exon 1'' La mRNAs. Comparable studies were not available for the mouse La gene. Therefore, we analyzed the expression of the mouse La gene (M. P. Bachmann, unpublished data). From these studies, we know that three mouse La mRNA isoforms exist. We termed these isoforms exon 1 a, b, and c. Like all the human La mRNA isoforms, all the murine La mRNAs are also functional. Despite their different 5' starts, all La mRNAs encode the same human or mouse La protein, as the initiation codon locates in the exon 2.
Quantitative PCR was performed for RNAs isolated from tissues of either mutant La-transgenic (Fig. 4, a, M1, M2, and b, muLa), or nontransgenic (Fig. 4b, non-tg), or native La-transgenic mice (Fig. 4b, nLa). In addition, we estimated the copy number of La mRNAs for commercially available multitissue-cDNA panels (MTC) from mouse (Fig. 4b, mMTC) and human (Fig. 4b, hMTC) tissues. We analyzed a total of eight mutant La-transgenic mice, four native La-transgenic mice, and four nontransgenic FVB mice. RNAs were obtained from brain, liver, spleen, gut, and kidney. In addition, we analyzed the RNAs from thymus tissue when available. Results for liver and spleen tissues are shown in Fig. 4. Fig. 4a shows the data for individual animals. Fig. 4b gives the mean values of all analyzed animals. Our data shows: 1) the mouse La mRNAs are expressed at similar levels in nontransgenic and transgenic animals, 2) the human La mRNAs in the transgenic mice are expressed in addition to the endogenous mouse La mRNAs, and 3) the expression level of the respective La transgene is comparable to the endogenous mouse La mRNA level. Consequently, 1) the expression of the respective transgene does not alter the expression of the mouse La gene and 2), the mutant La transgene is expressed at similar levels as the native La transgene. Thus, the mutant La mRNA escapes NMD.
|
Using the anti-human La-specific mAb SW5, we detected the truncated human La protein in total tissue extracts of both mutant La-transgenic mouse lines M1 and M2 (Fig. 5a, muLa-tg, M1, and M2). To compare the expression levels, extracts of mutant La-transgenic mice (Fig. 5a, muLa-tg) and nontransgenic mice (Fig. 5a, non-tg) were normalized with anti-actin Abs. For some total tissue extracts, the detection of the mutant La transgene was difficult because of a high content of mouse Igs in these extracts which cross-reacted with the secondary Abs (data not shown). To improve the quality of the immunoblots and to allow a quantitative comparison of the mutant La protein expression in different tissues, we immunoprecipitated the extracts with the anti-human La-specific mAb SW5 and detected the immunoprecipitated Ag with a secondary Ab which does not bind to reduced Igs (see Materials and Methods). As shown in Fig. 5b, the mutant La protein was equally expressed in all organs tested including in thymus, spleen, liver, brain, kidney, and intestine (Fig. 5b, muLa).
|
When tissues were stained with the anti-human La-specific anti-La mAb SW5, the mutant La protein which lacks the NLS was found in the cytoplasm (Fig. 5c, muLa-tg) while the native La transgene was found in the nucleus (Fig. 5c, nLa-tg).
Detection of native human La in extracts from tissues of mutant La-transgenic mice
Extracts from tissues of mutant La-transgenic (Fig. 6, a–d), native La-transgenic (Fig. 6, lanes e, g, and i), and nontransgenic (Fig. 6, lanes f, h, and j) mice were precipitated with an anti-human La serum. The precipitated La peptides were analyzed with the anti-La mAbs 3B9 (Fig. 6, lanes a, e, and f), SW5 (Fig. 6, lanes b, g, and h), and 4B6 (Fig. 6, lane c, i, and j). The anti-La mAb 3B9, which recognizes the N-domain of both human and mouse La protein, reacted with the endogenous mouse La protein in all analyzed transgenic and nontransgenic extract samples (Fig. 6, lanes a, e, and f). In addition, it reacted with the mutant human La protein in the extract of the mutant La-transgenic mouse (Fig. 6, lane a) and the native human La protein in the extract of the native La-transgenic mouse (Fig. 6, lane e). The human anti-La-specific mAb SW5, which is directed to the N-domain of human La, also reacted with the mutant human La protein (Fig. 6, lane b) and the native human La protein (Fig. 6, lane g) but failed to react with the endogenous mouse La protein (e.g., Fig. 6, lane h). The anti-human La-specific anti-La mAb 4B6, which is directed to the C-domain of human La, also reacted with the native human La protein in the extract of the native La-transgenic mouse (Fig. 6, lane i). However, it failed to react with the mutant human La protein (Fig. 6, lane c) and the endogenous mouse La protein (e.g., Fig. 6, lane j). Unexpectedly, all three monoclonal anti-La Abs reacted with an additional protein in the extract of the mutant La-transgenic mice (Fig. 6, lanes a–c). This protein, which has the same molecular weight as full-length human La protein, is absent in the extract of the nontransgenic mouse (Fig. 6, lanes f, h, and j). Taken together, the protein 1) has the same molecular weight as human La protein, 2) is coprecipitated by the anti-La serum, and 3) is recognized by three anti-La mAbs. From previous studies, we know that the three anti-La mAbs recognize independent epitopes. The two epitopes recognized by the anti-La mAbs 3B9 and SW5 are upstream of the frame shift mutation. The epitope recognized by the anti-La mAb 4B6 is downstream of the frame shift mutation. Therefore, we conclude that this protein represents full-length human La protein. Full-length human La protein can only be translated from the mutant La mRNA if ribosomal frame shifting occurs upstream of the PTC.
|
As mentioned above, during the immunoblotting experiments, we detected high levels of Igs in extracts from tissues of older mutant La-transgenic mice, especially in kidney tissue extracts. Similar results were not seen for age-matched FVB/N controls (data not shown). Clinical SLE is characterized by the presence of anti-nuclear autoantibodies (ANA) and immune complex deposits in tissues.
Immune complex nephritis in mice transgenic for mutant human La. Therefore, we checked mutant La-transgenic mice for proteinuria and analyzed kidney tissues of mutant La-transgenic mice for Ab deposits. Six of ten animals showed mild to moderate proteinuria. Two animals had severe proteinuria (+++). Tissues from these animals were further analyzed. Their histology data are shown in Fig. 7. We detected glomeruli which were positively stained with Abs to Ig and complement (Fig. 7, a–g). Similar deposits were not seen in age-matched FVB/N controls (Fig. 7, h–j).
|
|
R knockout mice (data not shown). In none of the control mice which were older than 5 mo and younger than 8 mo did we detect autoantibodies to the nuclear Ags Ro, La, Sm, and dsDNA. One of six nontransgenic FVB/-N mice older than 8 mo developed ANAs. However, these Abs were IgM-type Abs and were not specific for the lupus Ags La, Ro, Sm, RNP, or dsDNA.
Finally, we wanted to learn whether or not the overexpression of native La can also result in an autoimmune phenotype. Therefore, we collected the sera of 25 mice which were transgenic for the native La Ag. All the sera were negative for ANAs including for Abs to La, Ro, Sm, RNP, Scl-70, and Jo-1 (data not shown). Twelve of these sera were from animals older than 8 mo, six sera were from mice between 5 and 6 mo, and seven sera were from mice of 2–3 mo of age. Taken together, these data support our interpretation that the autoimmune phenotype of the mutant La-transgenic mice is the result of the transgenic expression of the mutant La form but not an overexpression of La.
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionDisclosuresReferences |
|---|
Since our development of the autologous cDNA expression library approach to detect self Ags, Sahin et al. (37) and many others have extended this technique to sera and tissues from tumor patients and termed it as serological identification of Ags by recombinant expression cloning (SEREX) (37). And, the list of autoantigens that are detected by the SEREX technique is still growing (
www2.licr.org/CancerImmunomeDB
). Paradoxically, autoantibodies to tumor Ags can be found more frequently in patients with missense mutations than in patients with nonsense mutations although nonsense mutations such as stop, splice/stop, splice, or frameshift mutations encode more severely altered proteins (38). One explanation may be that the RNA quality control mechanism NMD allows eukaryotic cells to recognize and degrade mRNAs which contain a nonsense mutation. Consequently, this elimination of mRNAs containing nonsense mutations helps to protect cells against harmful dominant-negative effects which could be caused by the encoded truncated mutant protein if translated. In addition, the absence of a translation product of the mutant mRNA may protect the mutant cell against its detection and destruction by the immune system.
In accord with this reasoning, we expected that the identified frame shift mutation in the patient’s La cDNA should not result in an immune response. However, we detected anti-NeoLa Abs in the serum of this patient and even more surprising in 30% of all analyzed anti-La-positive autoimmune patients. Anti-NeoLa Abs were detected with two independent techniques. Using the immunoblotting technique, one cannot identify anti-NeoLa Abs in serum samples which react with the native N-domain of La protein. Consequently, the immunoblotting technique results in an underestimation of the anti-NeoLa frequency. However, the ELISA specific for the NeoLa peptide allowed us to overcome this limitation.
At present, we do not know whether the occurrence of the mutant La was the event for triggering of the anti-NeoLa immune response in the autoimmune patient(s) or the mutant La epitope became a target of an (auto)immune response because it was expressed just at the time when the immune response occurred. At least in the mutant La-transgenic mice, which should be tolerant to the mutant La form, the latter mechanism seems to work. In patients who express the mutant La epitope after establishing tolerance, both possibilities or a combination of them could also be considered.
Either way, the development of an IgG type anti-NeoLa response requires the expression of mutant La or a mutant La-related epitope at the time when the immune response started. One prerequisite for the expression of mutant La protein, however, is an escape of the mutant La mRNA from RNA control. In a previous study, we had shown that Escherichia coli and insect cells can partially correct the mutant La-reading frame by ribosomal frame shifting. Ribosomal frame shifting most likely occurs in the oligo(A)-region upstream of the PTC in exon7 (34). Due to this, ribosomal frame-shifting expression of the mutant La mRNA does not only result in the expression of the prematurely terminated truncated form of La. In addition to the mutant La protein, a portion of correct human La was made. However, whether this ribosomal frame shifting could also occur in mammalian cells was unclear. If so, the ribosomes translating the mutant La mRNA have to slip into the native La-reading frame. The shift has to occur upstream of the PTC and, thus, the ribosomes will bypass the PTC. During translation of the native reading frame downstream of the PTC, these ribosomes can pass all exon splice junctions downstream of the PTC in the mutant La mRNA and terminate translation at the native La mRNA stop codon. Thereby, all EJCs downstream of the PTC will be removed. Consequently, NMD cannot detect the frame shift mutation in the mutant La mRNA, and the mutant La mRNA will, thus, circumvent NMD.
To experimentally verify that mutant La mRNA can escape from NMD by ribosomal frame shifting, we established two independent mouse lines transgenic for mutant La. Southern blotting verified the integration of the respective transgene into independent genomic regions in these two lines of mice (data not shown). Quantitative PCR, immunoblotting analysis, and immunohistology show that the mutant La mRNA is not eliminated by NMD. In addition to mutant La, a protein is made which 1) has the same size as native human La protein, 2) was coprecipitated by patient anti-La Abs, and 3) reacted with three monoclonal anti-La Abs on immunoblots. The used anti-La mAbs recognize unrelated epitopes in either the N- or the C-domain of La. Therefore, the anti-La-reactive protein contains La epitopes upstream and downstream of the PTC in the mutant La mRNA. This strongly supports the interpretation that ribosomal frame shifting upstream of the PTC in mutant La mRNA is not restricted to bacterial and insect cells but can also occur in mammalian cells.
In summary, the capability to correct the mutant La reading frame into the native La reading frame by ribosomal frame shifting may be the reason for escape of mutant La mRNAs from NMD.
Both mouse lines develop ANAs including to dsDNA and an immune complex nephritis. The observation of a lupus phenotype in both transgenic lines strongly argues that this phenotype is critically dependent upon the mutant La transgene itself. Thus, expression of mutant La seems to result in an increased risk for developing an autoimmune phenotype.
As La protein is hypothesized to be involved in so many different cellular functions, there are several possibilities as to how the expression of mutant La enhances the risk of an autoimmune response in the transgenic mice. For example, La protein was hypothesized to work as a RNA chaperone and to be involved in the correct folding of ribonucleoproteins (RNPs) including for example in RNA quality control of 5S ribosomal RNA. An impaired assembly of ribosomes and/or an increased rate of incorrectly folded RNPs could easily lead to neoepitopes which could become the target of an immune response. Alternatively, the mutant La protein could indirectly interfere with the apoptosis pathway allowing perhaps the survival of autoreactive T and B cells (31, 32). Many recently described transgenic or knockout mice which interfere with the expression of components of the immune system or the apoptosis pathway develop autoimmune phenotypes (39). Although a direct link between La protein, apoptosis resistance, and the immune system has not yet been experimentally shown, there are reports supporting such a function. As mentioned in the introduction section La protein was shown to be involved in the cap-independent translation of the X-linked inhibitor of apoptosis. Moreover, La protein is involved in the translation of the mdm2 protein (21, 22). The mdm2 protein is a negative regulator of p53. p53 is a key regulator in DNA repair, the cell cycle and the apoptosis pathway, e.g., p53 is a transcription factor of the cell cycle inhibitor p21. Consequently, a down-regulation of mdm2 would result in an up-regulation of p21. Interestingly, p21 is known to be up-regulated in many experimental autoimmune animal models, and is assumed to stabilize autoreactive memory cells in SLE patients (35). Consequently, an altered expression of La protein could affect the apoptosis resistance and the survival of autoreactive immune cells. Indeed, preliminary data show that La-overexpressing cells have an increased apoptosis resistance while knocking down the expression of La increases the apoptosis sensitivity of cells (M. P. Bachmann, unpublished data). Furthermore, real-time PCR shows a 100-fold up-regulation of p21 including in thymus in the mutant La-transgenic mice but not in control animals (M. P. Bachmann, unpublished data).
In summary, our data suggest that mutant La mRNAs can circumvent NMD most likely via ribosomal frame shifting. Expression of mutant La can have two effects: 1) it can become a target of an immune response in patients and mutant La-transgenic mice, and 2) it increases the risk of developing systemic autoimmunity, at least in mutant La-transgenic mice. The expression of mutant La may interfere with one or all of the functions of the La Ag which finally leads to a predisposition toward an autoimmune response. In contrast to the occurrence of autoimmunity in many other transgenic mice, the autoimmune response in the mutant La-transgenic mice may reflect a true risk factor in human SLE, as suggested by the escape mechanism described here and as suggested by the detection of Abs to the NeoLa form in a significant fraction of SLE patients.
|
Disclosures |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionDisclosuresReferences |
|---|
|
Footnotes |
|---|
1 This work was supported by grants to: M.P.B. (National Institutes of Health (NIH) Grants GM63497 and IP20RRI5577, and Oklahoma Center for the Advancement of Science and Technology Grant HR01-003, Fritz-Thyssen Stiftung (Germany), and Bundesministerium für Forschung und Technik, Germany); J.B.H. (NIH Grants AI31584, DE015223, AI053747, AI054117, AR049084, AR48940, AR42460, AI24717, and RR020143); and A.D.F. (NIH Grants AI51647, AI48097, and AR48940). M.P.B. was a recipient of a Greenberg-Scholarship (Oklahoma Medical Research Foundation). 
2 Address correspondence and reprint requests to Dr. Michael P. Bachmann, Institute of Immunology, Medical Faculty Carl Gustav Carus, Technical University Dresden, Fiedlerstrasse 42, 01307 Dresden, Germany. E-mail address: michael.bachmann{at}mailbox.tu-dresden.de
3 Abbreviations used in this paper: NMD, nonsense-mediated decay; PTC, premature termination codon; EJC, exon junction complex; SLE, systemic lupus erythematosus; DAPI, 4',6-diamidino-2-phenylindole; NLS, nuclear location signal; ANA, anti-nuclear Ab; MTC, multitissue-cDNA panel; RNP, ribonucleoprotein.
Received for publication February 3, 2006. Accepted for publication April 21, 2006.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionDisclosuresReferences |
|---|
Related articles in The JI:
This article has been cited by other articles:
|
|
 |
|
  K. Nakata, B. K. Lipska, T. M. Hyde, T. Ye, E. N. Newburn, Y. Morita, R. Vakkalanka, M. Barenboim, Y. Sei, D. R. Weinberger, et al. DISC1 splice variants are upregulated in schizophrenia and associated with risk polymorphisms PNAS, September 15, 2009; 106(37): 15873 - 15878. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
