Decreased Stability and Translation of T Cell Receptor ζ mRNA with an Alternatively Spliced 3′-Untranslated Region Contribute to ζ Chain Down-regulation in Patients with Systemic Lupus Erythematosus*

The molecular mechanisms involved in the aberrant expression of T cell receptor (TCR) ζ chain of patients with systemic lupus erythematosus are not known. Previously we demonstrated that although normal T cells express high levels of TCR ζ mRNA with wild-type (WT) 3′ untranslated region (3′ UTR), systemic lupus erythematosus T cells display significantly high levels of TCR ζ mRNA with the alternatively spliced (AS) 3′ UTR form, which is derived by splice deletion of nucleotides 672–1233 of the TCR ζ transcript. Here we report that the stability of TCR ζ mRNA with an AS 3′ UTR is low compared with TCR ζ mRNA with WT 3′ UTR. AS 3′ UTR, but not WT 3′ UTR, conferred similar instability to the luciferase gene. Immunoblotting of cell lysates derived from transfected COS-7 cells demonstrated that TCR ζ with AS 3′ UTR produced low amounts of 16-kDa protein. In vitro transcription and translation also produced low amounts of protein from TCR ζ with AS 3′ UTR. Taken together our findings suggest that nucleotides 672–1233 bp of TCR ζ 3′ UTR play a critical role in its stability and also have elements required for the translational regulation of TCR ζ chain expression in human T cells.

The TCR chain is a component of the T cell receptor (TCR) 1 complex that plays a critical role in the assembly and transport of the TCR complex to the cell surface and signal transduction through the TCR that leads to T cell activation. The TCR gene is located in chromosome 1q23.1 (1)(2)(3), and genome-wide screening for systemic lupus erythematosus (SLE) susceptibility genes demonstrated a linkage between chromosome 1 and SLE susceptibility (4 -6). SLE T cells express decreased amounts of TCR mRNA and protein (7)(8)(9). However, the TCR/CD3-initiated signaling events are increased in intensity compared with those in normal T cells (10). Specifically, crosslinking of the CD3-TCR complex in SLE T cells leads to in-creased free intracytoplasmic calcium levels (11) and tyrosine phosphorylation of cytosolic proteins (7). This "overexcitability" of the SLE T cell has been attributed to the fact that the missing TCR chain is replaced by the FcR␥ chain (12) and the surface membrane lipid rafts are aggregated (13). Because replenishment of the TCR chain in SLE T cells corrects signaling aberrations and increases interleukin-2 production (14), it is clear that the decreased TCR chain expression is central in the disease process.
The molecular mechanisms of diminished expression of the TCR mRNA and protein in SLE T cells remain unknown. The TCR gene spans at least 31 kb, and the transcript is generated as a spliced product of 8 exons that are separated by distances of 0.7 to more than 8 kb (15,16). Nucleotide sequence analysis of the TCR gene showed increased frequency of alternatively spliced (AS) forms missing various exons as well as splice insertion in SLE T cells (17). Analysis of the 3Ј untranslated region (UTR) showed a novel 344-bp AS form with a deletion of nucleotides from 672 to 1233 of exon VIII of the TCR mRNA. Using a specific primer that spans either side of the newly identified alternatively spliced site, we recently reported that the AS form of TCR mRNA with 344 bp 3Ј UTR was predominantly expressed in SLE T cells compared with normal T cells (14).
In SLE T cells, defective TCR protein expression inversely correlates with the level of TCR mRNA with AS 3Ј UTR (18) and directly with WT 3Ј UTR. Thus, a molecular switching of TCR mRNA with WT 3Ј UTR to AS 3Ј UTR in T cells contributes to the regulation of TCR chain expression. Several AS isoforms of the TCR mRNA with different nucleotide sequences of the 3Ј UTR have been recently identified in murine T cells (19). The expression of these TCR mRNA isoforms with alternatively spliced 3Ј UTR can be modulated by exposure to pharmacologic agents (20,21).
The regulation of mRNA decay rate is an important control point in determining the abundance of cellular transcripts (22). The 3Ј UTRs of eukaryotic mRNAs are known to play a crucial role in post-transcriptional regulation of gene expression by modulating nucleocytoplasmic mRNA transport, polyadenylation status, subcellular targeting, translation efficiency, stability, and rates of degradation (23)(24)(25). Selective expression of TCR mRNA with very short AS 3Ј UTR in SLE suggests that it plays an important role in the down-regulation of TCR chain expression in SLE T cells. Therefore, we conducted studies to determine the stability of TCR mRNA with an AS 3Ј UTR in SLE T cells and the transport and subsequent trans-lation of TCR mRNA with AS 3Ј UTR in normal and SLE T cells. In this report we demonstrate that AS 3Ј UTR confers instability to the TCR mRNA, as it does to the mRNA of other genes, and thus contributes to decreased levels of TCR mRNA. In addition, we show that the TCR mRNA with AS 3Ј UTR displayed poor translation efficiency leading to the production of low amounts of protein. Therefore, the production of TCR mRNA with short 3Ј UTR represents a molecular mechanism that contributes to decreased expression of TCR chain in SLE T cells.

MATERIALS AND METHODS
SLE Subjects and Controls-Patients fulfilling the American College of Rheumatology Classification criteria for SLE (26) were chosen for the study. Subjects who were on prednisone were asked not to take this medication at least 24 h before drawing the blood. Disease activity for the SLE patients was scored by the SLE disease activity index (SLEDAI) system. Our studies included patients with SLEDAI scores ranging from 0 to 20. The protocol of the study was approved by the Human Use Committees of the involved institutes. Written informed consent was obtained from all participating subjects.
Cells and Antibodies-Peripheral blood mononuclear cells from heparinized venous blood were obtained by density gradient centrifugation over Ficoll. T cells were isolated by magnetic separation of non-T cells using a mixture of hapten-conjugated antibodies and MACS microbeads coupled to anti-hapten mAb as described earlier (17) (Miltenyi Biotech, Auburn, CA). In all cases, the percentage of T cells in the isolated subpopulation was Ͼ97% as determined by fluorescence-activated cell sorter analysis. The TCR chain mAb 6B10.2 (Santa Cruz Biotechnology, Santa Cruz, CA) recognizing amino acids 31-45 of the polypeptide (N-terminal mAb) was purchased from BD Biosciences. The C-terminal TCR chain mAb recognizing amino acids 145-161 (28) and horseradish peroxidase-conjugated anti-phosphotyrosine mAb 4G10 were purchased from Upstate Biotechnology (Lake Placid, NY). CD3-PE was from Sigma. OKT3 was from Orthoclone Biotech (Raritan, NJ). CD4-PE, CD8-PE, and CD14-PE were from Immunotech (Coulter Corp., Miami, FL), CD16-PE was from BD Biosciences. Fluorescent isotype controls were purchased either from Sigma, Jackson ImmunoResearch Laboratories (West Grove, PA), or BD Biosciences.
RT-PCR of TCR mRNA with Wild-type and Alternatively Spliced 3Ј Untranslated Region-Total RNA was isolated using RNeasy mini kit (Qiagen) from 5 million T cells from SLE or normal volunteers according to the supplier's directions. Single-stranded cDNA was synthesized by using the AMV reverse transcriptase-based reverse transcription system from Promega (Madison, WI) and oligo(dT) primer as instructed by the manufacturer. The primers for the amplification were synthesized by Sigma-Genosys (The Woodlands, TX). The 5Ј primer for the PCR amplification of WT and AS TCR mRNA was 5Ј-AGC CTC TGC CTC CCA GCC TCT TTC TGA G-3Ј (sense bp 34 -62 according to the numbering of Weissman et al.; Ref. 2) (Sigma-Genosys). The 3Ј primer for the specific amplification of TCR mRNA with WT 3Ј UTR was 5Ј-CAA CAG TCT GTG TGT GAA GGT TTG GAG-3Ј (antisense bp 737-711). This primer is located in the splice-deleted region of the TCR mRNA ( Fig. 1) and is absent in TCR mRNA with AS 3Ј UTR. TCR mRNA with AS 3Ј UTR was specifically amplified by a primer that spans both sides of the AS site, making it non-complementary, and will not anneal with the WT TCR mRNA. The sequence of this primer is 5Ј-CAT CTT CTG GCC CTT CAG TGG CTG AGA AGA GTG AA-3Ј (antisense bp 1235-1234 and 671-649). ␤-actin was used as a control, and the primers were forward, 5Ј-CAT GGG TCA GAA GGA TTC CT-3Ј, and reverse, 5Ј-AGC TGG TAG CTC TTC TCC A-3Ј. The reverse transcription product was diluted 2-to 5-fold and used for PCR amplification of TCR mRNA or ␤-actin control. The amplification was carried out with a high fidelity PCR system from Roche Applied Science in a Biometra T-3 thermal cycler after initial denaturation at 94°C for 4 min, 33 cycles at 94°C, 45 s; 67°C, 1 min; 72°C for 2 min; and a final extension at 72°C for 7 min. 10 -15 l of the PCR products were electrophoresed on 1.2-1.5% SeaKem-agarose gel (FMC BioProducts, Rockland, ME) and visualized with ethidium bromide or SYBR green staining. Quantitation of the RT-PCR product was done using the software GEL-PRO (Media Cybernetics, Silver Spring, MD).
Real-time Quantitative PCR of TCR mRNA with Wild-type and Alternatively Spliced 3Ј UTR-Real-time quantitative PCR was carried out with a Cepheid Smart Thermocycler by adding SYBR green to the reaction mixture. The SYBR green-based real-time quantitative PCR technique was used to detect the expression of TCR mRNA with WT and AS 3Ј UTR. PCR beads were used for amplification (Amersham Biosciences), and the amplification was carried out in 25-l tubes. Total RNA was prepared from normal and SLE T cells, and 1 g of RNA was reversed transcribed into cDNA and diluted 10-fold for real-time quantitative PCR. The PCR mixture consists of 2 l of diluted cDNA, 5 l of SYBR green diluted (1:500) containing PCR master mixture, and 150 M each primer in a total volume 25 l. The specific primers for real-time quantitative PCR were designed as above by Sigma-Genosys. Real-time quantitative PCR was carried out with a Cepheid Smart Thermocycler by adding SYBR green (1:500 dilution) to the reaction mixture. The condition for real-time quantitative PCR was initial denaturation of 94°C for 60 s followed by cycles of 94°C for 15 s, 67°C for 15 s, and 72°C for 30 s for 45 cycles, followed by a melting point determination that results in a single peak if the amplification is specific. Real-time quantitative PCR products were also separated on a 1.5% agarose gel to visualize the formation of correct PCR products. The expression level for each gene was shown by the cycle numbers needed for the cDNA to be amplified to reach a threshold. The cycle numbers were also converted into arbitrary amounts of DNA using standard curves generated for each pair of primers, and the results were normalized to the housekeeping gene GAPDH mRNA in the same sample. The cycle threshold value was generated by ABI PRISM 7700 SDS software version 1.7 and then exported to an Excel spreadsheet where equations from the standard curve were generated. Using the cycle threshold values, concentrations of TCR mRNA with WT and AS 3Ј UTR as well as GAPDH mRNAs were calculated from the equations. Each sample was analyzed at two different concentrations, and the result from the linear portion of the standard curve was presented. Samples were analyzed in triplicate at each concentration, and TCR mRNA with WT and AS 3Ј UTR levels were normalized to the corresponding GAPDH.
PCR Amplification, Cloning, and Expression of the Wild-type and Alternatively Spliced TCR mRNA from T Cells-cDNA was synthesized from total RNA by oligo(dT) primer and AMV reverse transcriptase as described above. The full-length TCR mRNA with WT and AS 3Ј UTR was amplified by PCR using primers 5Ј-AGC CTC TGC CTC CCA GCC TCT TTC TGA G-3Ј (sense bp 34 -62) and 5Ј-CCC TAG TAC ATT GAC GGG TTT TTC CTG-3Ј (antisense bp 1472-1446). The amplification was carried out using a high fidelity PCR system from Roche Applied Science in a Biometra T-3 thermal cycler after initial denaturation at 94°C for 6 min, 33 cycles at 94°C, 1 min; 67°C, 1 min; 72°C for 2 min, and a final extension at 72°C for 7 min. The PCR products containing 1438 bp WT and 916 bp AS TCR were ligated to unidirectional pcDNA 3.1 His TOPO vector (Invitrogen). Minipreps were prepared from 12 recombinant clones, and the direction of the insert was verified by restriction mapping using BamHI 1. Wild-type and AS TCR clones with proper orientation were subjected to DNA sequencing from both orientations on an ABI 377 sequencer using the ABI dye terminator cycle sequencing kit (ABI PRISM; Applied Biosystems Inc., Foster City, CA).
Transfection of TCR with WT and AS 3Ј UTR to COS-7 cells-The day before transfection, COS-7 cells were subcultured in RPMI 1640 containing 10% fetal bovine serum and penicillin streptomycin at 37°C in a 5% CO 2 incubator. For transfection, cells were trypsinized, washed, and resuspended in 250 l of Opti-MEM serum-free medium (Invitrogen). 15 g of plasmid pcDNA 3.1 V5 HIS TOPO containing TCR with WT or AS 3Ј UTR was added and electroporated at 250 V, 960 F in a 0.4-cm cuvette (Bio-Rad). Electroporated cells were immediately washed in medium and cultured. Transfected cells were lysed at different time points after incubation with actinomycin D (5 g/ml), and the mRNA was isolated after lysis of the cell membrane by Nonidet P-40 (7).
In Vitro Transcription and Translation-The WT and AS TCR mRNA were transcribed and translated using the TNT T7 quick-coupled rabbit reticulocyte lysate transcription/translational system as recommended by the manufacturer (Promega). 10 g of plasmids containing WT or AS TCR was incubated with the transcription/translation system in the presence of Transcend Botin-Lysyl-tRNA for 60 min at 30°C. The translated product was electrophoresed and transferred to polyvinylidene difluoride membranes. The incorporated biotinylated lysine was detected non-radioactively by blotting with streptavidinhorseradish peroxidase and developed using an ECL chemiluminescent kit (Amersham Biosciences).
Reporter Gene Construction and Transient Transfection-The fulllength 3Ј UTRs of WT and AS of TCR mRNA were amplified by PCR and cloned into the XbaI site downstream of the luciferase reporter gene in the pGL3-Basic and Enhancer vector(s) (Promega). Transcription from this construct is driven by an SV40 promoter. The correct orientations of the clones were verified by restriction mapping and sequencing. Plasmid DNA transfections of Jurkat cells, COS-7 cells, and T cells were carried out in 24-well plates (Corning Inc., Corning, NY) using Lipofectamine TM 2000 reagent (Invitrogen) following the manufacturer's protocol. The day before transfection, 6 ϫ 10 4 COS-7 cells or 0.6 ϫ 10 6 Jurkat cells were plated in 0.5 ml of medium/well. For each well, Lipofectamine TM 2000 reagent (2-3 l) was mixed with plasmid DNA (1.5 g) in serum-free Opti-MEM to allow DNA-Lipofectamine TM reagent complexes to form. The complexes were added to respective wells and mixed by gently rocking the plate back and forth. The transfected cells were washed with 1ϫ phosphate-buffered saline three times.
Luciferase Activity Assays-Luciferase activity was determined using a luciferase assay system (Promega) following the manufacturer's protocol. Briefly, the transfected cells were incubated in a CO 2 incubator at 37°C for 18 h and then lysed with 60 l of reporter lysis buffer (Promega). Cellular debris was removed by centrifugation for 2 min at 12,000 rpm. Luciferase activity was assayed with 20 l of lysate and 80 l of luciferase assay reagent (Promega) in a TD20/20 luminometer (Turner Designs) using a commercially available kit (Promega). Light output was measured over a 10-s time period in triplicate for each sample. Relative luciferase activity was calculated by averaging the readings. Transfection efficiency was established in all samples by cotransfection with 1 g of a plasmid encoding the cytomegalovirus promoter-driven ␤-galactosidase gene. Luciferase activity was normalized for transfection efficiency using the corresponding ␤-galactosidase activity. The data are presented as -fold increase in luciferase activity compared with that obtained from cells transfected with vector alone.
Densitometry and Statistical Analysis-Densitometry analysis of the Western blot was performed with the software program GelPro (Media Cybernetics). Statistical analysis was done using the software Minitab Version 13 (Minitab Inc., State College, PA) by applying the Student's t test.

Stability of TCR mRNA with AS 3Ј UTR in SLE T Cells-
Cloning and sequence analysis of TCR mRNA revealed the presence of a novel TCR chain transcript with alternatively spliced 3Ј UTR in human T lymphocytes (Fig. 1). Although the TCR mRNA with AS 3Ј UTR was detected at low levels in normal human T cells, it was found to be predominantly expressed in SLE T cells (18). Because a vast majority of lupus patients display decreased expression of TCR mRNA (7, 17), we considered the TCR mRNA with AS 3Ј UTR to be more unstable than the TCR mRNA with WT 3Ј UTR. Accordingly, purified T cells were incubated with transcription inhibitor actinomycin D (5 g/ml) for different periods of time (0, 1, 2, and 4 h), and the levels of expression of WT and AS 3Ј UTR TCR mRNA were quantitated by semiquantitative RT-PCR using specific primers as described under "Materials and Methods." By calculating the quantity of mRNA as a percentage of the amount at 0 h, we determined that the levels of TCR mRNA with AS 3Ј UTR were reduced compared with the WT mRNA in SLE T cells (Fig. 2, A and B). We also compared the stability of the TCR mRNA with WT 3Ј UTR to that with the AS 3Ј UTR after calculating the ratio of the levels of each mRNA to ␤-actin mRNA. As shown in Fig. 2C, the stability of AS TCR mRNA is decreased compared with the WT 3Ј UTR in SLE T cells (Fig. 2C).
Stability of TCR mRNA with AS 3Ј UTR in Normal T Cells-To address the question whether the decreased half-life of TCR mRNA with AS 3Ј UTR is restricted to SLE T cells, we analyzed the stability of TCR mRNA in normal T cells. Purified normal T cells were incubated with transcription inhibitor actinomycin D (5 g/ml) for different periods of time, and the levels of TCR mRNA expression with WT and AS 3Ј UTR were measured in parallel. We observed that the mRNA stability of TCR mRNA with AS 3Ј UTR was also low in normal T cells (Fig. 3, A and B). Although the stability of TCR mRNA with AS 3Ј UTR was low in normal cells, it was not as prominent as it was recorded in SLE T cells. After 4 h of incubation with actinomycin D, the amount of TCR mRNA with AS 3Ј UTR was 17 and 33% in SLE and normal T cells, respectively (Figs. 2B and 3B). However, the rate of degradation of TCR mRNA with WT 3Ј UTR was not significantly different between SLE and normal T cells.
Real-time Quantitative PCR Analysis of TCR mRNA with AS 3Ј UTR in SLE and Normal T Cells-We used a quantitative real-time RT-PCR to measure the stability of TCR mRNA with WT and AS 3Ј UTR in SLE T cells and normal T cells treated with actinomycin D (5 g/ml) for 0, 1, 2, or 4 h to confirm the observed differences in the stability of the TCR mRNA with WT or AS 3Ј UTR. The mRNA from the samples were converted to cDNA by reverse transcriptase. The reverse transcription product of cDNA was PCR amplified using TCR WT-and AS 3Ј UTR-specific primers in the presence of SYBR green in a Cepheid thermocycler as described under "Materials and Methods." To validate the real-time quantitative PCR, the standard curves for TCR mRNA with WT and AS 3Ј UTR and GAPDH cDNA were constructed first (data not shown). The cycle threshold values for TCR mRNA with WT/AS 3Ј UTR and GAPDH mRNA in the sample were measured in triplicate by real-time quantitative PCR, and the amounts of these mRNA were determined from the standard curves. TCR mRNA with WT or AS 3Ј UTR was evaluated as the relative quantity against GAPDH mRNA in the cells before treatment with actinomycin D (Fig. 4). Real-time quantitative PCR demonstrated that WT 3Ј UTR mRNA was more stable than AS 3Ј UTR mRNA during the first 4 h. These results also show that TCR mRNA with AS 3Ј UTR (AS mRNA: GAPDH mRNA ϭ 0 h, p ϭ 0.037; 4 h, p ϭ 0.005) in SLE degraded to a greater extent compared with WT 3Ј UTR (WT mRNA: GAPDH mRNA ϭ 0 h, p ϭ 0.017; 4 h, p ϭ 0.009) after 4 h of incubation with actinomycin D in SLE T cells (WT 3Ј UTR, p ϭ 0.042 and AS 3Ј UTR, p ϭ 0.007) (Fig. 4A). The stability of TCR mRNA with WT 3Ј UTR (WT mRNA: GAPDH mRNA ϭ 0 h, p ϭ 0.06; 4 h, p ϭ 0.03) and AS 3Ј UTR (AS mRNA: GAPDH mRNA ϭ 0 h, p ϭ 0.005; 4 h, p ϭ 0.001) was also analyzed in normal T cell by real-time quantitative PCR (WT 3Ј UTR, p ϭ 0.05 and AS 3Ј UTR, p ϭ 0.015) (Fig. 4B). The results show that TCR mRNA with WT 3Ј UTR degraded to a lower extent compared with the TCR mRNA with AS 3Ј UTR after 4 h of incubation with actinomycin D (Fig. 4).
Stability of TCR mRNA with AS 3Ј UTR in COS-7 Cells-Next we studied the stability of TCR mRNA with WT and AS 3Ј UTR transfected in COS-7 cells. TCR with WT or AS 3Ј UTR was cloned into a eukaryotic expression vector, pcDNA3.1/V5-HIS-TOPO, and transfected to COS-7 cells. After 18 h of transfection, the cells were incubated with actinomycin D for 0 or 4 h, and the mRNA were measured by RT-PCR analysis. Although in COS-7 cells there was no significant degradation of TCR mRNA by 4 h, by 8 h we recorded a significant decrease in the level of TCR mRNA with AS 3Ј UTR compared with WT 3Ј UTR (Fig. 5). To rule out that the difference was due to differences in export from nucleus or transport of TCR mRNA with WT and AS 3Ј UTR from nucleus to cytoplasm as a result of alternative splicing, the transfected cells were lysed at different time points after incubation with actinomycin D and mRNA was isolated after lysis of the cell membrane by Nonidet P-40. Levels of TCR mRNA with WT and AS 3Ј UTR were measured by semiquantitative RT-PCR. The data showed that the ratio of TCR mRNA between the cytoplasm and nucleus was similar at different points, suggesting that alternative splicing of TCR 3Ј UTR does not affect the transport of the TCR mRNA (data not shown).

Translational Regulation of TCR Expression in COS-7 Cells Transfected with TCR Carrying WT and AS 3Ј UTR-Our
previous data showed that the down-regulation of TCR protein in SLE T cells inversely correlated with the level of TCR mRNA with AS 3Ј UTR (18). This finding suggested that, although the coding region is intact in TCR mRNA with AS 3Ј UTR, the translational regulation of TCR mRNA is defective. We analyzed the translational efficiency of TCR mRNA with WT and AS 3Ј UTR in transfected COS-7 cells by immunoblotting experiments. Eighteen hours after transfection with equal amounts of the constructs, the cells were lysed, electrophoresed, and immunoblotted with a TCR chain-specific mAb. As shown in Fig. 6A, TCR with WT 3Ј UTR expressed a single band with a molecular mass of 16 kDa as reported previously (14,29). Cells transfected with TCR carrying AS 3Ј UTR produced very low amounts of the 16-kDa form of TCR protein. Densitometric analysis showed a 10-fold decrease in the protein level in COS-7 cells transfected with AS TCR (Fig.  6B). However, RT-PCR showed equal amounts of mRNA in WTand AS TCR -transfected cells. This suggests that alternative splicing of TCR 3Ј UTR has a major impact on the translation of TCR mRNA. Indirectly the data also suggest that the splice-deleted 562-bp region contains some elements that play important roles in the translational regulation of TCR chain.
In Vitro Transcription and Translation of TCR with WT and AS 3Ј UTR-To provide further evidence that the TCR mRNA with AS 3Ј UTR results in lower production of protein and rule out the possibility of transfection variability in the above experiments, we transcribed and translated the TCR with AS 3Ј UTR in vitro using biotinylated lysine as a nonradioactive label. The in vitro transcribed and translated products were electrophoretically separated and blotted with streptavidin-horseradish peroxidase conjugate. Similar to the expression in COS-7 cells, in vitro transcription and translation of TCR with WT 3Ј UTR produced a protein of 16-kDa TCR chain (Fig. 7A). The TCR with WT 3Ј UTR produced high levels of 16-kDa protein, whereas the TCR with AS 3Ј UTR produced significantly low levels of the 16-kDa TCR protein. The results were also confirmed by stripping and reprobing the blots with TCR chain-specific antibody (Fig. 7B). Densitometric analysis showed that transfected COS-7 cells with TCR AS 3Ј UTR resulted in an 8-fold reduction in the protein level (Fig. 7C).

Stability and Expression of Luciferase Gene Fused with TCR
WT and AS 3Ј UTR-To determine whether the AS 3Ј UTR conferred instability to genes other than that of TCR , we introduced the WT and AS 3Ј UTRs downstream of the luciferase gene. Luciferase constructs containing TCR WT and AS 3Ј UTR were made by PCR amplification by engineered primers containing the XbaI site (Fig. 8A). PCR-amplified products with the XbaI site (Fig. 8B) were cloned in pGL3 basic and enhancer vector(s) after digestion with XbaI, which is located immediately after the luciferase gene at the 3Ј region (Fig. 8A). The luciferase clones were confirmed by restriction mapping that resulted in appropriate sizes of inserts (Fig. 8C). Finally luciferase clones were reconfirmed by sequence analysis (data not shown). The resulting luciferase reporter constructs containing TCR WT and AS 3Ј UTR were individually co-transfected with ␤-galactosidase to COS-7 cells, Jurkat cells, and normal T cells. Cells were harvested after 18 h of transfection with phosphate-buffered saline, and the luciferase activity of the transfected cells was measured. As shown in Fig. 8, D and E, luciferase activity is significantly decreased in cells transfected with the AS 3Ј UTR construct compared with WT 3Ј UTR in Jurkat, COS-7, and normal T cells (COS-7 cells, p ϭ 0.001 and Jurkat cells, p ϭ 0.049). Maximal decrease was found in COS-7 cells (2.8-fold) (Fig. 8D), followed by Jurkat cells (2.1-fold) (Fig.   8E) and T cells (data not shown). These data indicate that AS 3Ј UTR confers instability that is not gene-or cell type-specific. DISCUSSION One of the major outcomes of the present study is the finding that TCR mRNA with AS 3Ј UTR that is selectively expressed in SLE T cells is less stable compared with the WT 3Ј UTR in T cells of SLE patients as well as normal subjects (Figs. 2 and  3). These results are consistent with the general notion that the shorter the 3Ј UTR, the less stable is the message (25, 30 -32). However, compared with normal subjects, the level of degradation of the TCR mRNA with AS 3Ј UTR is more severe in SLE T cells (Fig. 4), suggesting that the SLE T cells have additional factors that promote TCR mRNA degradation. However, the stability of TCR mRNA with WT 3Ј UTR is similar between SLE and normal T cells. These data, along with the fact that TCR mRNA with AS 3Ј UTR is selectively expressed in SLE T cells and TCR chain protein expression inversely correlates with TCR mRNA with AS 3Ј UTR, suggest that SLE T cells preferentially produce TCR mRNA with the AS 3Ј UTR form that is unstable. This process represents a new pathologic mechanism for the down-regulation of TCR mRNA in SLE. The levels of AS 3Ј UTR TCR mRNA are not increased in T cells from patients with other systemic inflammatory rheumatic diseases such as rheumatoid arthritis (18).
The 3Ј UTR of mRNAs play important roles in regulating gene expression at the post-transcriptional level (23,24,(33)(34)(35). To date, the minimum and maximum lengths of 3Ј UTR observed in human mRNAs are 21 nucleotides and 8.5 kb, respectively, and average length is 0.3-0.7 kb (32,36). The 3Ј UTR of TCR is ϳ1 kb, which is considerably longer, suggesting that it may have one or more important roles in regulation of gene expression. The molecular mechanisms that lead to destabilization of the TCR mRNA with AS 3Ј UTR are currently unknown. 3Ј UTR of mRNA contains cis-acting elements, for example, adenosine-uridine (AU)-rich elements that bind to trans-activating factors and either stabilize or destabilize the transcripts (37,38). Sequence analysis indicates the presence of one AU-rich sequence in the splice-deleted 562 bp of TCR mRNA with AS 3Ј UTR. The TCR mRNA with AS 3Ј UTR also does not contain a 31-nucleotide sequence (from 973 to 1003) that is conserved across the TCR mRNA 3Ј UTR of several species. If this conserved region is involved in the regulation of mRNA stability, then the absence of this region would explain the decreased stability of the TCR mRNA with AS 3Ј UTR. Alternatively, the new splicing introduces exonexon boundaries more than 50 base pairs from the stop codon and likely to be subjected to nonsense-mediated mRNA decay (39,40). Nonsense-mediated mRNA decay has been implicated in the regulation of AUF1 expression via conserved alternatively spliced elements in the 3Ј UTR (41,42).
Regulation of mRNA stability is often mediated by elements within the 3Ј UTR (23,25,37). To facilitate identification of regulatory elements within 3Ј UTRs, reporter systems using constitutive promoters have been extensively used. We tested a constitutive luciferase reporter model to determine whether the WT 3Ј UTR and AS 3Ј UTR of the TCR could affect mRNA stability. The luciferase activity in the transfected cells with the AS 3Ј UTR construct was significantly decreased compared with WT 3Ј UTR (Fig. 8). These data demonstrate that the destabilizing effect of the AS 3Ј UTR that was identified as part of the TCR mRNA is not gene-specific and may confer instability to other genes. In addition, the effect was found not to be cell type-specific, suggesting that either trans factors are not required and the destabilizing effect is simply 3Ј UTR lengthdependent or the required trans factors are non-cell type-specific and rather universal in nature. FIG. 7. In vitro transcription and translation of TCR with WT and AS 3 UTR. The TCR with WT and AS 3Ј UTR was transcribed and translated using the TNT T7 quick-coupled rabbit reticulocyte lysate transcription/translational system from Promega. A, the translated product was electrophoresed and transferred to polyvinylidene difluoride membranes. The incorporated biotinylated lysine was detected with streptavidin-horseradish peroxidase and developed using an ECL chemiluminescent kit. B, the blots were stripped and reprobed with TCR chain-specific antibody. C, densitometric analysis of TCR protein produced from WT and AS 3Ј UTR (mean Ϯ S.E., n ϭ 3) We have previously reported that in SLE T cells the mutations/polymorphisms and splice variation of TCR mRNA including a single exon or combination of exons are significantly higher compared with normal subjects (17,18). These abnormalities could provide an additive effect on the degradation of TCR mRNA with AS 3Ј UTR in SLE T cells. It has been reported that in patients with multiple sclerosis the regulation of major histocompatibility class II transactivator expression is associated with alternative splicing in the 3Ј UTR (43,44). It is also possible that abnormalities in other cellular factors may contribute to increased degradation of TCR mRNA with 3Ј UTR in SLE T cells.
The second major finding is that the TCR mRNA with AS 3Ј UTR produces low amounts of 16-kDa TCR protein compared with the chain with WT 3Ј UTR (Figs. 5 and 6). This result strongly suggests that the presence of translational regulatory elements in the splice-deleted 562 bp region is required for the proper and efficient translation of TCR chain mRNA. Clearly, more experiments are needed to explain how the splice-deleted region influences TCR chain protein expression. There are many reports describing the regulation of protein expression by AS elements (45,46). This regulation is mediated by the binding of protein factors to the 3Ј UTR, and splice deletion of TCR mRNA with AS 3Ј UTR may abate the binding of these factors. Identification of the proteins or factors that interact with the 562-bp splice-deleted region will support this hypothesis.

FIG. 8. Luciferase activity of TCR WT and AS 3 UTR in reporter gene transfected into COS-7 and Jurkat cells.
A, the 3Ј UTR WT and AS of TCR were cloned into the XbaI site downstream of the luciferase gene in the sense orientation and verified by restriction digestion. Luciferase construct containing TCR WT 3Ј UTR or AS 3Ј UTR was transfected into COS-7 and Jurkat cells. Schematic representation of constitutive luciferase reporter constructs (not drawn to scale). Transcription (bent arrow) was driven by the constitutive SV40 promoter upstream of the luciferase reporter gene. B, PCR-amplified products of WT or AS 3Ј UTR of TCR containing the XbaI site. C, verification of luciferase clones by restriction digestion using XbaI. D, luciferase activity in transfected COS-7 cells. E, luciferase activity in transfected Jurkat cells. Data shown are the mean Ϯ S.E. of three independent determinations.
We and others have shown that T cells from SLE patients are deficient in TCR chain and display abnormal TCR/CD3-induced early signaling events (7,27,47). The molecular basis of the TCR chain decrease in SLE T cells, and its significance in the pathogenesis of the disease has not been defined. Our present findings suggest that selective expression of TCR mRNA with AS 3Ј UTR is associated with the abnormalities of TCR chain expression in SLE patients. These results also infer an important pathway for the regulation of gene expression involved in the immune response by exploiting splicing processes that destabilize mRNAs or modulate its expression. In conclusion, we have reported herein that the instability and limited translation of the AS TCR mRNA contribute to the decreased expression of TCR chain in SLE T cells. We propose that the 562-bp splice-deleted region of the TCR 3Ј UTR contains crucial cis elements that bind proteins that confer stability and sufficient translation rate and when spliced out render the mRNA unstable and with limited translational efficiency.