Transcriptional Activation of the cAMP-responsive Modulator Promoter in Human T Cells Is Regulated by Protein Phosphatase 2A-mediated Dephosphorylation of SP-1 and Reflects Disease Activity in Patients with Systemic Lupus Erythematosus*

Systemic lupus erythematosus (SLE) is a complex autoimmune disease with numerous abnormalities recorded at the cellular, molecular, and genetic level. Expression of the basic leucine zipper transcription factor cAMP-responsive element modulator (CREM)α was reported to be abnormally increased in T cells from SLE patients. CREMα suppresses IL-2 and T cell receptor ζ chain gene transcription by direct binding to the respective promoters. Here, we show that increased CREM expression is the result of enhanced promoter activity. DNA binding analyses reveal direct binding of transcription factor specificity protein-1 (SP-1) to the CREM promoter resulting in enhanced transcriptional activity and increased CREM expression. Protein phosphatase 2A is known to activate SP-1 through dephosphorylation at its serine residue 59. Our results show that nuclei from SLE T cells contain lower levels of Ser59-phosphorylated SP-1 protein and a stronger SP-1 binding to the CREM promoter. We conclude that protein phosphatase 2A accounts for enhanced SP-1 dephosphorylation at Ser59 in SLE T cells. More importantly, CREM promoter activity mirrors reliably disease activity in SLE patients, underscoring its potential role as a biomarker for the prediction of flares in SLE patients.

Systemic lupus erythematosus (SLE) 3 is a chronic inflammatory disease characterized by an autoimmune process affecting every organ, including joints, kidneys, skin, and the central nervous system (1). T cells have been demonstrated to be important in the pathogenesis of SLE (2). Functional ab-normalities in SLE T cells are largely underwritten by altered gene transcription, and recent studies have revealed a vast number of transcription factors in SLE T cells displaying changes in their expression levels and function. Among them, the ␣-isoform of the transcriptional regulator cAMP-responsive element modulator (CREM) has emerged as a potent regulator of target genes that are crucially involved in T cell physiology, e.g. interleukin (IL)-2, T cell receptor chain, and transcription factor c-fos (3)(4)(5).
The CREM protein family constitutes a group of transcription factors that belong to the superfamily of basic domain/ leucine zipper domain proteins such as cAMP-responsive element-binding protein (CREB) and activating transcription factors. All of them share high structural similarities and possess the ability to bind specifically cis-regulatory DNA sequences as homo-or heterodimers. The target motifs are palindromic DNA sequences denoted cAMP response elements (CRE, TGACGTCA) as well as the 5Ј-TGAC half-sites of this palindrome, both of which are highly conserved within numerous promoters of eukaryotic genes (6). Increased intracellular cAMP levels stimulate a multitude of kinases, e.g. calmodulin kinase IV, that subsequently phosphorylate and thereby activate CREM (7).
Previous studies from our group have provided evidence that CREM␣ expression is increased in T cells from SLE patients (8). We have shown that CREM␣ binds to the Ϫ180 site of the IL-2 promoter in SLE T cells and thereby suppresses IL-2 production in these cells. Decreased IL-2 levels render SLE patients more susceptible to infections (9). In turn, suppression of CREM␣ expression in SLE T cells by introducing a plasmid encoding antisense CREM␣ into these cells sufficiently restores their IL-2 content and normalizes effector cell function (10). Transcriptional down-regulation of T cell receptor chain by CREM␣ binding to the corresponding promoter contributes to the structural abnormalities of the CD3 complex in SLE T cells. Because CREM␣ also represses transcription of the AP-1 family member c-fos, it limits the activity of this transcription enhancer family and thereby indirectly contributes to the defective IL-2 production in SLE T cells (2).
The CREM gene was initially isolated almost 20 years ago and comprises multiple exons (11). Since that time, a large number of different CREM isoforms has been identified, and the composition of the individual exons in these isoforms determines the functional propensities, e.g. whether they act as transcriptional activators or repressors (12). In contrast to CREB, expression and activity of the various CREM isoforms are finely regulated, not only post-transcriptionally, but also at the level of gene transcription. The CREM gene contains at least two distinct functional promoters. The inducible cAMP early repressor (ICER) isoforms of the CREM family are transcriptionally controlled through the intronic ICER promoter (11). ICER inducibility is highly cell type-specific, and its kinetics is characteristic of the early response genes. So, for example, ICER expression peaks rapidly upon stimulation with corticotrophin-releasing hormone in pituitary gland cells (13). In contrast, the CREM promoter located at the 5Ј-end of the CREM gene regulates transcription of the other CREM isoforms and was initially thought to be constitutively activated as expression of most CREM isoforms is ubiquitous and not inducible. However, because we found an increased amount of the transcriptional repressor CREM␣ mRNA in T cells from patients with SLE compared with healthy individuals (8) we asked whether elevated CREM␣ mRNA levels may represent the result of enhanced transcriptional activation through the CREM promoter in these cells and what key factors may be involved in CREM gene transcription.

EXPERIMENTAL PROCEDURES
Study Subjects-All SLE patients included in our studies were diagnosed according to the American College of Rheumatology classification criteria (14) and were recruited from the Division of Rheumatology at Beth Israel Deaconess Medical Center, Boston, MA, and the Department of Medicine at the Uniformed Services University of the Health Sciences, Bethesda, MD. All 12 SLE patients included in the CREM promoter assessment (as depicted in Fig. 2) were females at a median age of 36.5 years (range 22-62 years), four were of black ethnicity, and two were African-Americans. Six patients with rheumatoid arthritis or other autoimmune diseases as well as healthy individuals were chosen as controls. The study protocol was approved by the institutional review boards of all involved institutions, and written informed consent was obtained from all participating subjects.
T Cell Purification-Peripheral venous blood was collected in heparin-lithium tubes, and T cells were purified as described before (8).
Luciferase Assays-Three million T cells were transfected with a total of 3 g of the indicated plasmids and 10 ng of Renilla control vector by an Amaxa transfection system (Amaxa, Cologne, Germany). Whenever an effector:reporter co-transfection was performed, the molar ratio between the two was 3:1. Five hours after transfection, cells were collected and lysed, and luciferase activity was quantified using the Dual Luciferase Assay system (Promega) and a TD-20/20 luminometer (Turner Designs, Sunnyvale, CA). Transfection efficiency was quantified by measurement of Renilla enzyme activity. Each effector:reporter transfection experiment was repeated at least three times, and values in the bar diagrams are given as mean Ϯ S.E.
mRNA Extraction, RT-PCR, and Real-time Quantitative RT-PCR-Total RNA was isolated from three million T cells using the RNeasy Mini kit (Qiagen). Residual genomic DNA contamination was removed by using DNase I (Qiagen). Sequences for PCR primers were as follows: CREM forward, 5Ј-tcgcgaacttgggacga-3Ј; CREM reverse, 5Ј-ctgagctaaagcagggattg-3Ј; 18 S rRNA forward,: 5Ј-actcaacacgggaaacctca-3Ј; 18 S rRNA reverse, 5Ј-aaccagacaaatcgctccac-3Ј. Real-time PCR amplification of CREM was carried out using the same primers on a LightCycler 480 (Roche Applied Science), and threshold cycles (Ct) values were used to calculate relative mRNA expression by the ⌬⌬Ct relative quantification method against 18 S rRNA expression.
Preparation of Nuclear Protein Extracts-Five million T cells from individual subjects were collected and used for preparation of nuclear protein extracts as described before (16).
Chromatin Immunoprecipitation (ChIP) Assay-ChIP analysis was performed using the ChIP assay kit (Millipore) according to the manufacturer's instructions. T cells (3 ϫ 10 6 / immunoprecipitating antibody) were fixed with 1% formaldehyde for 10 min, lysed, and sonicated. The protein⅐DNA complexes were immunoprecipitated with either a polyclonal anti-SP-1 antibody (Abcam) or an isotype-matched control antibody (Abcam), subjected to a series of washing steps, reverse-cross-linked, and the protein was digested with proteinase K (Qiagen). The DNA was extracted by the QIAamp DNA mini kit (Qiagen). Immunoprecipitated and purified DNA was analyzed by PCR using the following primers detecting the relevant sequence of the CREM promoter: forward, 5Ј-ggggagatagaggttgcag-3Ј; reverse, 5Ј-gaccaaaagtagcgctgcag-3Ј. The corresponding nonimmunoprecipitated DNA (input DNA) was also analyzed by the same method.
SDS-PAGE and Immunoblotting-Whole T cell lysates obtained from SLE patients and healthy controls were analyzed by SDS-PAGE and subsequent immunoblotting as described previously (17). Phosphorylated SP-1 protein was visualized using an anti-SP-1 antibody kindly provided by B. Kahn-Perles (Marseille, France (18)) that specifically detects SP-1 phosphorylated at its serine 59. Total SP-1 protein was detected using a monoclonal anti-SP-1 antibody (Santa Cruz Biotechnology). A polyclonal ␤-actin antibody (Sigma) was used to show equal protein loading. Band intensities were quantified by OptiQuant TM software.
Statistical Analysis-The paired two-tailed Student's t test and the Pearson product moment correlation coefficient r were used for statistical analysis. Statistical significance was defined as p Ͻ 0.05.

CREM-500 Promoter Activity Is Elevated in SLE T Cells-
We generated luciferase reporter constructs containing serially truncated human CREM promoters ranging ϳ400, 500, 800, or 1,000 bp from the transcription initiation site (as schematically depicted in Fig. 1B). T cells obtained from at least three pairs of SLE patients and healthy control individuals were transfected with these reporters. Luciferase activity for each reporter was calculated relative to the basal empty reporter activity (Fig. 1B). Activity of the CREM promoter spanning ϳ500 bp, denoted CREM-500, was significantly elevated when introduced into SLE T cells compared with T cells from healthy individuals. (3.08 Ϯ 1.90 versus 0.87 Ϯ 0.33, p Ͻ 0.05). The activity of other CREM promoter constructs was much lower than that of the CREM-500 reporter, in both normal and SLE T cells. Furthermore, there were no significant differences in these constructs between normal and SLE T cells. We concluded that the CREM-500 promoter contains the crucial cis-regulatory elements mediating transcriptional CREM up-regulation in SLE T cells.
CREM-500 Promoter Activity Reflects SLE Disease Activity-We next increased the number of SLE patients and healthy controls to 12 individuals/group and included six patients with rheumatoid arthritis and other autoimmune diseases ( Fig. 2A). We purified T cells from these patients and transfected them with the CREM-500 reporter plasmid. The relative CREM-500 promoter activity level was significantly elevated in SLE T cells compared with T cells from healthy controls (3.54 Ϯ 2.23 versus 1.51 Ϯ 1.31, respectively, p Ͻ 0.05). In T cells from patients with other autoimmune diseases CREM-500 promoter activity was enhanced as well (2.16 Ϯ 1.66); however, this was not significant compared with T cell from normal individuals (p ϭ 0.39). As there was a fairly broad range of CREM-500 promoter activity levels among T cells from SLE patients, we addressed the question whether disease activity influences CREM promoter activation. Indeed, CREM-500 promoter activity levels correlate significantly with SLE disease activity indices of the respective patients (r ϭ 0.87, p Ͻ 0.001, Fig. 2B).

SP-1 Trans-activates CREM Promoter by Binding to Its
Ϫ468 Site-To elucidate the molecular mechanism underlying CREM-500 promoter activation in SLE T cells we analyzed the promoter sequence with particular focus on those nucleotides that are different between the CREM-500 and CREM-400 construct because the latter did not show any inducibility in SLE T cells (as shown in Fig. 1A). Within this sequence we identified a DNA motif that shares high similarity to the consensus binding site of a transcription factor denoted SP-1 (Fig. 3A). To test whether SP-1 protein participates in the nucleocomplex formation of the proteins bound to this portion of the CREM promoter we performed DNAbinding assays with nuclear T cell extracts and a synthetic double-stranded oligonucleotide harboring this element. The latter was denoted CREM-500 (Ϫ468) because the identified sequence (-ccacggc-) starts 468 bp upstream of the transcription initiation site. As can be seen in Fig. 3B (lane 2), addition of an SP-1-specific antibody to the binding reaction decreased the intensity of one particular band (indicated by an arrow) whereas inclusion of an unrelated antibody had no effect on complex formation (Fig. 3B, lane 3).
Next, we set out to confirm this finding by competition assays including the unlabeled wild-type (comp wt ) and an irrelevant (comp irrel ) probe at a 10-and 50-fold molar excess, respectively. The detected complex was not affected by inclusion of the irrelevant competitor but completely abrogated by the cold wild-type probe (Fig. 3C, lanes 2-5). To prove more specifically that SP-1 binds to this site we extended our competition assays by inclusion of two oligonucleotides, one of which was mutated at the predicted Ϫ468 site (comp mut1 ) and the other one mutated at an adjacent position (comp mut2 ). As expected, inclusion of the first one did not significantly alter band intensity whereas the latter one decreased band intensity in a concentration-dependent manner (Fig. 3C, lanes  6 -9). Taken together, these DNA-binding analyses indicate that endogenous SP-1 protein specifically interacts with the Ϫ468 site within the CREM promoter and participates in the complexes observed with nuclear protein extracts from human T cells.
We next determined the trans-regulatory capacities of SP-1 on the CREM promoter using the wild-type CREM-500 luciferase reporter and a CREM-500 plasmid containing a Ϫ468 site mutation (Fig. 3D). Human T cells were transiently cotransfected with the respective promoter constructs and either pCDNA3 vector or an expression plasmid encoding SP-1 and subsequently assayed for luciferase activity. Notably, SP-1 significantly trans-activated the CREM-500 promoter (2.00 Ϯ 0.24; p Ͻ 0.05), most likely through its binding to the Ϫ468 site as mutation at this site renders the CREM-500 reporter not inducible by SP-1 (1.13 Ϯ 0.31; p ϭ 0.69).

CREM Gene Transcription in SLE T Cells Is Up-regulated through Enhanced SP-1 Binding to the Ϫ468 Site of the CREM
Promoter-After we identified SP-1 as a transcriptional regulator at the Ϫ468 site within the CREM promoter we next wanted to clarify whether there is a difference in the binding affinity of endogenous SP-1 protein to this sequence between nucleoprotein from SLE patients and normal individuals. Therefore, we obtained T cell nucleoprotein lysates from 11 SLE patients and 8 controls and performed EMSAs using the CREM-500 (Ϫ468) oligonucleotide as a probe (Fig. 4A). Indeed, using SLE T cell nucleoprotein resulted in a significantly stronger band intensity of the specific protein⅐DNA complex (1.52 Ϯ 0.25; p Ͻ 0.05) than using nucleoprotein from control T cells. ChIP analyses were performed in normal and SLE T cell samples (n ϭ 2) using a polyclonal SP-1 antibody. Results from these experiments confirmed the EMSA data and indicate increased SP-1 binding to the CREM promoter in SLE T cells in vivo (a representative result of the ChIP is shown in Fig. 4B).
SP-1 Activation Is Mediated through Dephosphorylation by Phosphatase PP2A-SP-1 function is highly dependent on its phosphorylation status, and its serine 59 has been identified as a major target for phosphatase PP2A during cell cycle interphase (18). Because we already proved that SP-1 by itself trans-activates CREM-500 promoter (see Fig. 3D), we next asked whether addition of PP2A to this reaction can further increase CREM-500 activation. Human T cells were transfected with the wild-type CREM-500 promoter construct or its mutant CREM-500 mut as well as the SP-1-encoding plasmid with and without overexpression of PP2A (Fig. 5A). Inclusion of PP2A enhanced trans-activational capacity of SP-1 on the CREM-500 promoter (1.60 Ϯ 0.26, p Ͻ 0.05). As expected, there was no activation of the CREM-500 mut reporter (0.69 Ϯ 0.16, p ϭ 0.13). This transcriptional effect was confirmed at the mRNA level using T cells transfected with either SP-1, PP2A, or both expression plasmids (Fig. 5B). Co-expression of SP-1 and its activating phosphatase PP2A yielded a robust 40.5-fold increase of CREM mRNA as quantified by real-time PCR (Fig. 5B, bar diagram). Specificity of PP2A effect was further substantiated by using a PP2A construct un- able to dephosphorylate substrates due to a site-directed mutation within its active enzymatic domain (15). This defective PP2A protein was not able to increase SP-1 effect on the wildtype CREM-500 promoter (Fig. 5C).
We have previously reported that the PP2A protein amount is increased in T cells from lupus patients to a degree that reflects the respective SLE disease activity status (19). PP2A is known to activate specifically SP-1 through dephosphorylation at its serine residue 59. Accordingly, we set out to determine the phosphorylation status of SP-1 protein at Ser 59 in SLE T cells versus normal T cells (Fig. 5D). We analyzed nucleoprotein lysates obtained from eight lupus T cell and five normal T cell samples by Western blotting using antibodies against either Ser 59 -phosphorylated SP-1 or total SP-1 protein. Fig. 5D shows representative results from two normal and two lupus T cell analyses. Densitometry of the respective band intensities of all samples demonstrated a significantly lower ratio of Ser(P) 59 SP-1 over total SP-1 protein in T cells from lupus patients compared with normal T cells (0.30 Ϯ 0.18, n ϭ 8 versus 0.76 Ϯ 0.32, n ϭ 5; p Ͻ 0.05). This supports the hypothesis that increased PP2A amounts in SLE T cells mediate Ser 59 dephosphorylation and thus functional activation of SP-1.

DISCUSSION
In this paper we present evidence that increased expression of the transcriptional repressor CREM␣ in T cells from SLE FIGURE 3. SP-1 protein binds to and trans-activates CREM-500 promoter. A, CREM promoter encodes multiple putative SP-1 binding sites. DNA analysis revealed at least six SP-1-binding elements within the CREM-500 promoter region, which are indicated as black vertical bars. Only one of the six (Ϫ468) sites is located outside the Ϫ400 region. B and C, DNA binding analyses were performed with a radiolabeled oligonucleotide containing the (Ϫ468) site of CREM-500 promoter, nuclear protein extracts from human T cells, and poly(dI)⅐poly(dC) or poly(dG)⅐poly(dC) as indicated below. The specific protein⅐DNA complex containing SP-1 (marked by an arrow) was supershifted by an anti-SP-1 antibody (B, lane 2). For competition assay, the experiments were conducted by adding either one of the following unlabeled oligonucleotides at a dose as indicated: the unlabeled wild type probe (comp wt , C, lanes 2 and 3), an irrelevant oligonucleotide (comp irrel , C, lanes 4 and 5), and oligonucleotides with specific mutations at the (Ϫ468) site (comp mut1 , C, lanes 6 and 7) or at an adjacent location (comp mut2 , C, lanes 8 and 9). D, scheme of the constructs of both the wild-type CREM-500 promoter reporter and a mutant CREM-500 reporter harboring a mutation at the Ϫ468 SP-1 site (CREM-500 mut ) is shown. The CREM-500 reporter plasmid or its mutant was introduced into human T cells, respectively, in the presence or absence of SP-1 protein.
The experiment was repeated three times, and values are given as mean Ϯ S.E. (error bars). n.s., not significant.
patients is due to enhanced activity of the CREM promoter. We show that PP2A, which is also known to be up-regulated in SLE T cells, mediates dephosphorylation and thereby activation of transcription factor SP-1. SP-1 directly binds and trans-activates the CREM promoter. Moreover, CREM promoter activity levels significantly reflect disease activity in SLE patients.
We reported previously on CREM␣ overexpression in SLE T cells (8). Increased CREM␣ levels account for suppressed expression of a multitude of T cell-related genes, including IL-2 and T cell receptor chain (3,4). IL-2 has well been demonstrated to be important in 1) the generation and function of T regulatory cells, 2) activation-induced cell death, and 3) the generation of cytotoxic cells that are necessary for the defense against infections (20).
We found that increased transcription activity, as assayed by CREM promoter activity, is responsible for the increased CREM expression in SLE patients. Notably, our analyses displayed a statistically strong correlation between the CREM promoter activity levels and SLE disease activity indices, which are a marker of disease activity capturing both clinical and laboratory manifestations in SLE patients. This result indicates that modulating CREM expression may be beneficial for the control of autoimmune processes and inflammation in SLE patients. Furthermore, CREM promoter activity represents a new candidate biomarker for lupus disease activity. Longitudinal studies need to be conducted to define the temporal relationship between CREM promoter activity and clinical activity. In contrast to most biomarker studies that assay candidate molecules on either mRNA or protein expression level, to the best of our knowledge, this study is the first one to utilize the promoter activity of a gene as read-out system. Replacement of the luciferase gene by a GFP-encoding sequence should allow for an easier assessment of CREM promoter activity.
Our current study points to protein phosphatase PP2A as a key player in the CREM␣ up-regulation in SLE T cells. We demonstrated previously that expression levels of PP2A are increased in SLE T cells. In addition, higher PP2A expression levels are present in patients with higher disease activity. This finding complements our previous results in which we showed that PP2A mediates dephosphorylation of p-CREB, which contributes to the defective production of IL-2 in SLE T cells (19). Both CREB and CREM are members of the basic domain/leucine zipper domain proteins transcription factor family, although their transcriptional effects on IL-2 regulation are antithetic. Thus, PP2A employs two distinct mechanisms to suppress IL-2 expression in SLE patients: activating a suppressor (CREM␣) and concurrently suppressing an activator (p-CREB). Similar mechanisms may also take place in the abnormal expression of other genes in SLE T cells as Ͼ3000 genes encode the CRE site on their promoters (5).

. CREM gene transcription is induced in SLE T cells by enhanced SP-1 binding to the (؊468) site within the CREM promoter.
A, radiolabeled CREM-500 (Ϫ468) probe was incubated with nuclear T cell extracts obtained either from 11 SLE patients or 8 healthy controls followed by electrophoresis on a DNA retardation gel. Band intensity of the specific SP-1⅐DNA complex was quantified using densitometry and is shown as mean value Ϯ S.E. (error bar). B, in vivo SP-1 binding to the CREM promoter was analyzed by ChIP. Sonicated lysates from 3 ϫ 10 6 T cells from SLE patients and healthy controls were incubated with a polyclonal anti-SP-1 antibody or a control IgG and subsequently subjected to PCR using CREM promoter-specific primers. Nonimmunoprecipitated input DNA is shown on the left side. Two independent experiments were performed, and a representative result is presented here. FIGURE 5. PP2A cooperates with SP-1 in the induction of CREM promoter. A, primary human T cells were co-transfected with expression plasmids of SP-1 with or without PP2A as indicated in the presence of either a wild-type CREM-500 reporter or its mutant CREM-500 mut . The experiment was repeated three times, and values are given as mean Ϯ S.E. (error bars). n.s., not significant. B, PP2A cooperates with SP-1 in the induction of CREM mRNA. T cells were transfected with either SP-1 or PP2A alone or a combination of both plasmids. Total plasmids used for transfection were kept at the same amount by using control plasmid, pcDNA3. 5 h after transfection, the endogenous CREM mRNA levels were assayed against 18 S rRNA expression either by regular RT-PCR or real-time PCR (bar diagram). C, wild-type PP2A or a catalytically defective PP2A mutant was co-introduced into human primary T cells together with the CREM-500 promoter reporter as indicated. The experiment was repeated three times, and values are given as mean Ϯ S.E. (error bar). D, decreased phosphorylation of SP-1 at Ser 59 in SLE T cells is shown. Nuclear proteins were obtained from eight SLE patients and five healthy controls and analyzed sequentially by immunoblotting using antibodies detecting SP-1 protein phosphorylated at its serine 59, total SP-1 protein, and ␤-actin. Western blot images show representative results from two healthy controls and two lupus patients. The bar diagram gives the ratio between Ser 59 -phosphorylated and total SP-1 protein as quantified by densitometric analysis (mean Ϯ S.E.).
Transcription factor SP-1 is ubiquitously expressed in numerous cell types, and its binding site is present in the promoter of most genes. A closer analysis of the core region of the human CREM promoter revealed at least six sites highly homologous to the consensus SP-1 binding site. Interestingly, our studies indicate that SP-1 binding to one of these sites (at position Ϫ468 relative to the transcription initiation site) is responsible for transcriptional activation of the CREM promoter and thus for increased CREM expression in primary human T cells. At this point, we cannot exclude that other SP-1 sites may also become functional for CREM expression in a different context such as different cell development stages.
SP-1 function is highly regulated through a number of posttranslational modifications, e.g. phosphorylation occurring at multiple serine and threonine residues; however, only serine 59 is specifically dephosphorylated by PP2A (18). This finding corroborates with our previous results that expression of PP2A was increased in SLE T cells and the current finding that nuclear SP-1 protein phosphorylated at serine 59 is decreased in SLE T cells whereas cellular amounts of total SP-1 protein are comparable between T cells from lupus patients and healthy individuals. It was shown previously that dephosphorylation of SP-1 increased its affinity to DNA whereby this enhanced its transcriptional activity. In addition to phosphorylation, O-linked glycosylation has also been reported to regulate the SP-1 activity. Whether O-linked glycosylation is also involved in the PP2A-mediated SP-1 effects on the human CREM promoter deserves further investigation as reciprocal phosphorylation and O-linked glycosylation are common mechanisms regulating the activity of a number of intracellular proteins.
Dysregulated function of T cells is central to the disordered immune system in SLE patients. The presented characterization of PP2A Ͼ SP-1 Ͼ CREM signaling pathway (Fig. 6) unravels another key component of lupus T cell abnormalities and extends our understanding of the pathogenic potential of SLE T cells.