Comparison of T Cell Receptor-Induced Proximal Signaling and Downstream Functions in Immortalized and Primary T Cells

Background Human T cells play an important role in pathogen clearance, but their aberrant activation is also linked to numerous diseases. T cells are activated by the concurrent induction of the T cell receptor (TCR) and one or more costimulatory receptors. The characterization of signaling pathways induced by TCR and/or costimulatory receptor activation is critical, since these pathways are excellent targets for novel therapies for human disease. Although studies using human T cell lines have provided substantial insight into these signaling pathways, no comprehensive, direct comparison of these cell lines to activated peripheral blood T cells (APBTs) has been performed to validate their usefulness as a model of primary T cells. Methodology/Principal Findings We used quantitative biochemical techniques to compare the activation of two widely used human T cell lines, Jurkat E6.1 and HuT78 T cells, to APBTs. We found that HuT78 cells were similar to APBTs in proximal TCR-mediated signaling events. In contrast, Jurkat E6.1 cells had significantly increased site-specific phosphorylation of Pyk2, PLCγ1, Vav1, and Erk1/Erk2 and substantially more Ca2+ flux compared to HuT78 cells and APBTs. In part, these effects appear to be due to an overexpression of Itk in Jurkat E6.1 cells compared to HuT78 cells and APBTs. Both cell lines differ from APBTs in the expression and function of costimulatory receptors and in the range of cytokines and chemokines released upon TCR and costimulatory receptor activation. Conclusions/Significance Both Jurkat E6.1 and HuT78 T cells had distinct similarities and differences compared to APBTs. Both cell lines have advantages and disadvantages, which must be taken into account when choosing them as a model T cell line.


Introduction
Human T cells control the extent and focus of the adaptive immune response to pathogens. T cells are activated by the interaction of the cell surface, multi-subunit T cell receptor (TCR) with an antigen-bound major histocompatibility complex present on an antigen presenting cell [1,2]. In addition to TCR induction, T cells also require an activating signal from one or more costimulatory receptors, such as CD28 or the a4b1 integrin VLA-4, to become fully active [1]. Costimulation is critical for the specificity of the immune response because it allows T cells to be activated only during acute infection. This enables the adaptive immune system to mount a response to foreign invaders while tolerating its own cells. The mistaken recognition of self leads to aberrant T cell activation, resulting in numerous human disease, such as autoimmune diseases, cardiovascular disease and allergies/ asthma [3,4,5]. Signaling pathways that are activated by TCR and/or costimulatory receptors are good targets for the development of therapies to these diseases [4,5]. However, before effective therapies can be developed, we must first better understand the intracellular signaling that occurs when a T cell is activated.
An initial event upon TCR activation is the induction of the Src family kinases Lck and Fyn, which then phosphorylate the immunoreceptor tyrosine-based activation motifs (ITAMs) present on several TCR subunits ( Figure 1) [1]. The protein tyrosine kinase ZAP-70 is recruited to the phosphorylated ITAMs and activated upon phosphorylation of tyrosine 319 [1]. Activated Lck, Fyn, and ZAP-70 then phosphorylate multiple downstream substrates, including linker for activation of T cells (LAT) and the tyrosine kinase Pyk2 [6,7,8]. Pyk2 is a member of the Fak family of kinases and appears to control actin cytoskeletal rearrangements that are critical for T cell activation [6]. LAT is a hematopoietic-specific adaptor protein that mediates many downstream events following TCR stimulation. Upon TCR activation, LAT is phosphorylated on five conserved tyrosines, which then bind to several SH2 domain-containing proteins, such as the related adaptor proteins Grb2, Grap, and Gads, as well as PLC-c1 [8]. Once recruited to LAT, PLC-c1 is phosphorylated by the Tec family tyrosine kinase Itk at tyrosine 783. This leads to the increased ability of PLC-c1 to cleave phosphatidlyinositol (4,5) bisphosphate into inositol (1,4,5) trisphosphate, which is important for Ca 2+ influx, and diacylglycerol, which is important for protein kinase C activation [9].
PLC-c1 and the Grb2 family of adaptor proteins all contain SH3 domains that mediate the recruitment of signaling proteins to LAT. Once such protein is SLP-76, an adaptor protein that is brought to the LAT complex via its interaction with the adaptor protein Gads and PLC-c1 ( Figure 1) [10]. Upon TCR stimulation, SLP-76 is phosphorylated and both Vav1 and Itk bind to these phosphorylated tyrosines [10]. Vav1 is a guanine nucleotide exchange factor for Rac, a small GTPase that triggers cytoskeletal rearrangements downstream of TCR induction [11]. Vav1 requires the presence of Itk to bind phosphorylated SLP-76, although it does not appear that Vav1 and Itk interact directly [12]. Similarly, Itk requires binding to phosphorylated SLP-76 to maintain its active conformation [13]. The end result of these interactions at the LAT complex is the induction of important downstream functions needed for T cell activation, such as Ca 2+ flux, Erk1/Erk2 activation, receptor upregulation and the induction of cytokine release.
Much of our understanding of TCR-mediated signaling was discovered using two human T cell lines, the Jurkat E6.1 T cell and the HuT78 T cell. The original Jurkat cells were derived from a 14-year-old boy with T cell acute lymphoblastic leukemia (T-ALL). These cells are thymocytically derived with the characteristics of immature thymocytes. The Jurkat E6.1 subclone cell line was developed in the 1980s [14]. Sublines of Jurkat E6.1 cells have been isolated that are deficient in several TCR signaling molecules, including stable lines lacking the TCRb chain, Lck, LAT, SLP-76, ZAP-70, Vav1, or PLC-c1 [14]. These mutants have yielded much information and insight into the function of many peripheral TCR signaling molecules. The HuT78 T cell line is historically a very important cell line: the H9 subclone of HuT78 T cells is the cell used to isolate HIV. HuT78 cells were derived from a Sezary syndrome patient, and are CD4+, mature, cutaneous, lymphoid T cells [15].
Despite being highly useful for several reasons, such as mutant sublines, high rates of transfection, and ease of growth, Jurkat E6.1 cells have several known abnormalities. These include deficiencies in PTEN and SHIP, lipid phosphatases that regulate phosphoinositide-3 kinase (PI3K) function [16,17]. The lack of PTEN and SHIP leads to several irregularities including constitutive expression of AKT, increased levels of phosphorylated phosphoinositide lipids, and the constitutive association of Itk with the plasma membrane [17,18]. HuT78 cells also have deficiencies; they have abnormal c-myc function and over express Bcl-xL, suggesting problems with anti-apoptotic pathways in these cells [19,20].
Although all cell lines have inherent problems, there are many reasons to use cell lines as model systems for studying the activation and function of human T cells. Cell lines are useful because they offer an unlimited, genetically manipulatable and widely accessible supply of experimental material that can be stored long-term and grown at a low cost. Using continuous cell lines also reduces the human burden and ethical considerations of experimentation and substantially reduces donor to donor variation. It is important however, that the pros and cons of a specific cell line are appreciated to allow for interpretation of experiments using these cell lines. Surprisingly, there are no previous studies that directly compare TCR and/or costimulatorymediated signaling and function in the human T cell lines to human activated peripheral blood T cells (APBTs). Therefore, we chose to compare the Jurkat E6.1 T cells and HuT78 T cells to APBTs in an effort to determine the conduciveness of using these T cell lines as a model for peripheral TCR signaling and function. We found that there was no noteworthy difference among the three cell lines when examining early signaling events such as ZAP-70, LAT, and SLP-76 phosphorylation. However, there were significant differences in downstream signaling events, costimulatory receptor expression and activation and in the cytokines and chemokines released upon TCR induction. Advantages and disadvantages exist for both cell lines, which must be considered during the experimental design process.

Results
In order to compare proximal TCR-mediated signaling events in Jurkat E6.1 T cells, HuT78 T cells, and APBTs, equal cell numbers were activated with maximal doses of stimulatory TCR antibodies. The samples were then analyzed by immunoblotting using phosphospecific and pan antibodies to specific signaling molecules. We calculated the ratio of phosphorylated protein to total protein for each specific protein using quantified densities from the immunoblots. By using this method, the total phosphorylation per protein of a specific signaling molecule for each cell type was determined, regardless of the size of the cell or the concentration of protein present in each cell. Thus, the relative extent of activation of key signaling molecules was compared between each cell type.
Jurkat E6.1 T cells, HuT78 T cells, and APBTs had significant differences in the phosphorylation of the tyrosine kinases ZAP-70, but not the adaptor proteins LAT and SLP-76 Before the relative activation of critical signaling molecules can be assessed, the surface expression of the TCR ab chain in the various T cell lines must be determined. This information is needed because differences in the surface expression of the TCR ab chain in Jurkat E6.1 T cells, HuT78 T cells and APBTs may result in variations in receptor-mediated signaling events. To this end, the surface expression of the TCR ab chain in the Jurkat E6.1 T cells, HuT78 T cells and APBTs was examined by flow cytometry. The mean fluorescent intensity (MFI) of the TCR ab chain expression for three separate experiments was 78+/219 for Jurkat E6.1 T cells, 29+/29 for HuT78 T cells and 60+/28 for the APBTs (Figure 2 and data not shown). It was consistently observed that HuT78 T cells had significantly less TCR ab chain expression than Jurkat E6.1 T cells and APBTs, whereas there was no significant difference in the expression of TCR ab chain between the Jurkat E6.1 T cells and APBTs ( Figure 2). Thus, theHuT78 T cells have significantly less TCR surface expression than the Jurkat E6.1 T cell and APBTs. However, whether this difference leads to alterations in the extent of intracellular signaling is unknown.
Upon engagement of the TCR with a specific peptide-MHC complex, the tyrosine kinases ZAP-70 is phosphorylated and activated by the Src family kinases Lck and Fyn [1]. Therefore, it was of interest to determine if there were differences in the TCRinduced phosphorylation of ZAP-70 among Jurkat E6.1 T cells, HuT78 T cells and APBTs. Phosphorylation of ZAP-70 at tyrosine residue 319 is required for its activation and subsequent phosphorylation of the adaptor molecules LAT and SLP-76 [21,22]. The phosphorylation of this tyrosine is an excellent readout for enzymatically active ZAP-70. As seen in Figure 3, the relative levels of tyrosine 319 phosphorylation were similar between HuT78 T cells and APBTs. However, there was a significant and reproducible 2-3 fold decrease in the site-specific phosphorylation of ZAP-70 in Jurkat E6.1 T cells ( Figure 3B). This suggests that Jurkat E6.1 T cells have reduced levels of TCRinduced ZAP-70 activation and potentially have altered downstream signaling.
We next examined the phosphorylation of signaling proteins downstream of ZAP-70 to see if the diminished phosphorylation of this important early kinase translated to later differences in TCRmediated signaling. ZAP-70 is the known kinase for the adaptor proteins LAT and SLP-76 [23]. The phosphorylation of LAT at tyrosine 191 is essential for the formation of multi-protein signaling complexes that transmit the signal from the TCR complex to downstream effectors [7,8]. When the relative levels of LAT tyrosine 191 phosphorylation were examined, no differences were observed between Jurkat E6.1 and APBT samples or between HuT78 and APBT samples ( Figure 4A and 4B). However, there was a small but significant difference in the phosphorylation of LAT tyrosine 191 between Jurkat E6.1 and HuT78 T cell lines ( Figure 4A and 4B). These data show that the phosphorylation of LAT is not substantially different among the cell lines, indicating the binding capacity of LAT in each cell is similar.
Another important adaptor protein for functions downstream of TCR activation is SLP-76. This protein is phosphorylated on three tyrosine residues, including tyrosine 128. Phosphorylated SLP-76 is essential for bringing Vav1 and Itk to the LAT signaling complex [10], which is vital for linking proximal TCR signaling to events such as calcium influx and actin cytoskeletal rearrangement. As seen for LAT, no difference was seen among the two cell lines and APBTs in the phosphorylation of SLP-76 tyrosine 128 ( Figure 4C and 4D), suggesting that SLP-76 has a similar signaling capacity in each type of cell. Together, these data indicate that the reduced levels of TCR a/b chain surface expression in the HuT78 T cells compared to other T cell lines and the decrease in ZAP-70 phosphorylation observed in Jurkat E6.1 T cells compared to HuT78 T cells and APBTs does not translate into differences in the activation of signaling proteins immediately downstream of ZAP-70.
Jurkat E6.1 T cells have hyperphosphorylated PLC-c1 and exhibited exaggerated Ca 2+ signaling compared to HuT78 T cells and APBTs Next, differences in several important signaling events downstream of LAT and SLP-76 were characterized in Jurkat E6.1 T cells, HuT78 T cells, and APBTs. PLC-c1 activation is important for both intracellular calcium flux and protein kinase C activation [1,8]. Tyrosine 783 on PLC-c1 is phosphorylated by the Tec family kinase Itk and is essential for PLC-c1 activation and enzymatic function [13,24,25]. Therefore, the phosphorylation of PLC-c1 tyrosine 783 was characterized to determine if there was a difference in the phosphorylation of this important residue among the T cell lines. Surprisingly, a 10-fold increase in the relative phosphorylation of PLC-c1 tyrosine 783 was observed in the Jurkat E6.1 T cell line, compared to the HuT78 T cells and APBTs ( Figure 5A and 5B). In contrast, no difference in the sitespecific phosphorylation of PLC-c1 was found between the HuT78 T cells and APBTs ( Figure 5A and 5B). This suggests that PLCc1 is substantially more activated in Jurkat E6.1 T cells compared to HuT78 T cells or APBTs.
The phosphorylation of tyrosine 775 on PLCc1 is also known to be critical for TCR-induced Ca 2+ influx and transcription factor activation [26]. Therefore, we examined whether the TCRmediated phosphorylation of this site is also increased in Jurkat E6.1 T cells compared to other T cell lines. Interestingly, Jurkat E6.1 T cells and APBTs had no detectable difference in the TCRinduced phosphorylation of tyrosine 775 on PLCc1 ( Figure 5C and 5D). In contrast, HuT78 T cells had significantly less phosphorylation of this site compared to both Jurkat E6.1 T cells and APBTs ( Figure 5C and 5D). This indicates that the phosphorylation of tyrosines 775 and 783 on PLCc1, which are both required for optimal TCR-induced function of this protein, may be differentially regulated.
The observation that PLCc1 Y783 is hyperphosphorylated in Jurkat E6.1 T cells compared to HuT78 T cells and APBTs lead us to further examine the kinetics of PLC-c1 phosphorylation. This was done in order to determine if the large differences in phosphorylation were seen not only in the amount of TCRinduced phosphorylation of PLC-c1 tyrosine 783 at a single timepoint, but also in the timing and duration of this phosphorylation event. A substantial difference was seen in the kinetics of phosphorylation of PLC-c1 tyrosine 783 between Jurkat E6.1 T cells, and HuT78 T cells. HuT78 T cells reached maximal phosphorylation at 1 minute post-TCR stimulation, followed by a steady decrease in phosphorylation over 30 minutes ( Figure 6A). APBTs and HuT78 T cells had similar TCR-induced phosphorylation kinetics of PLC-c1 Y783 (data not shown). In contrast, the Jurkat E6.1 T cell line reached maximal phosphorylation at 2 minutes and the phosphorylation of tyrosine 783 decreased only slightly over the course of 30 minutes ( Figure 6A). This indicates that Jurkat E6.1 T cell have not only greater relative PLC-c1 tyrosine 783 phosphorylation than HuT78 T cells and APBTs, but that they also sustain that level of hyperphosphorylation for a much longer time.
The TCR-induced influx of Ca 2+ into human T cells is controlled by PLC-c1 activation. The influx of Ca 2+ is important for many effectors functions in T cells, including NFAT activation and cytokine and chemokine production [1,27]. Due to its importance in downstream T cell functions, we examined whether the hyperphosphorylation of PLC-c1 in Jurkat E6.1 T cells lead to alterations in the levels of TCR-induced Ca 2+ influx in the T cell lines and APBTs. To this end, we utilized a quantitative, real-time measurement method for intracellular Ca 2+ developed by Tsien and coworkers [28]. Using this method, it was observed that both cell lines and APBTs showed TCR inducible Ca 2+ influx ( Figure 6B). Jurkat E6.1 T cells exhibited approximately 7-10fold higher Ca 2+ flux immediately following stimulation than HuT78 T cells or APBTs, as well as a higher level and longer time of sustained Ca 2+ ( Figure 6B). HuT78 T cells also showed higher levels of initial Ca 2+ signaling than APBTs, although they returned to similar levels as APBTs much more quickly than Jurkat E6.1 T cells ( Figure 6B). These data indicate that the increased phosphorylation of PLC-c1 tyrosine 783 in Jurkat E6.1 T cells also translated into an amplification in Ca 2+ flux, as compared to HuT78 and APBT cells.

Jurkat E6.1 T cells have hyperphosphorylated Pyk2 and Vav1 compared to HuT78 T cells and APBTs
The intracellular tyrosine kinase Pyk2 is critical for integrating receptor mediated signals that control the rearrangement of the actin cytoskeleton [6]. The levels of the enzymatic activity of Pyk2 is controlled by the phosphorylation of tyrosine 580 in the activation loop of the kinase domain [6]. Therefore, the phosphorylation of this site can be used as a marker for the induction of the kinase activity of Pyk2. To investigate the relative activation of Pyk2 in Jurkat E6.1 T cells, HuT78 T cells, and APBTs, the cells were stimulated for 1 minute (Jurkat E6.1 cells) or 30 seconds (HuT78 T cells and APBTs), which corresponded to the maximal activation time of Pyk2 in these cell types (M. Collins and J.C.D. Houtman, unpublished results). Interestingly, the relative level of Pyk2 tyrosine 580 phosphorylation was significantly different between Jurkat E6.1 T cells, HuT78 T cells and APBTs, with Jurkat E6.1 T cells having the highest relative level of Pyk2 tyrosine 580 phosphorylation and HuT78 T cells having the lowest ( Figure 7A and Figure 5B). These experiments show that the enzymatic activity of Pyk2 is different in the individual T cell lines, suggesting that there are differences in the ability of the TCR to control the actin cytoskeleton between Jurkat E6.1 T cells, HuT78 T cells and APBTs.
One potential mechanism for these differences is the differential expression of PTP-PEST, a tyrosine phosphatase known to dephosphorylate Pyk2 in T cells [29]. To investigate this possibility, the expression of PTP-PEST in the three T cell lines was characterized. Unfortunately, the expression of PTP-PEST was not detectable in cell lysates (data not shown). Therefore, PTP-PEST was immunoprecipitated from unstimulated Jurkat E6.1 T cells, HuT78 T cells and APBTs. As seen in Figure 7C, HuT78 T cells express PTP-PEST, whereas Jurkat E6.1 T cells and APBTs have no detectable expression of this phosphatase. These results agree with a recent study that also observed the Jurkat E6.1 T cells and APBTs do not express PTP-PEST [30]. These data suggest that the expression of PTP-PEST in HuT78 T cells results in reduced levels of Pyk2 phosphorylation compared to APBTs. In contrast, the significant differences in the TCR-induced phosphorylation of Pyk2 in Jurkat E6.1 T cells and APBTs are not due to differential expression of this phosphatase.
Vav1 is a guanine nucleotide exchange factor for the small GTPase Rac1 and is important for receptor-mediated actin cytoskeletal rearrangement in human T cells [11]. An obligate step in the enzymatic activation of Vav1 is the phosphorylation of tyrosine 174 [11]. In contrast to Pyk2, Vav1 tyrosine 174 had a moderate 2-fold hyperphosphorylation in Jurkat E6.1 T cells compared to both HuT78 T cells and APBTs ( Figure 8A and Figure 6B). These data suggest that Vav1 is hyperactivated in Jurkat E6.1 T cells, which, in conjunction with the hyperphosphorylation of Pyk2, indicates a potential distortion of cytoskeletal rearrangement in these cells. Itk, the common link between PLC-c1 and Vav1, has increased expression in Jurkat E6.1 T cells The hyperphosphorylation of both Vav1 and PLC-c1 at important tyrosine residues, as well as exaggerated Ca 2+ flux, in Jurkat E6.1 T cells led us to examine Itk, the common link among these events. Itk requires Vav1 to bind SLP-76, an event necessary for Itk to phosphorylate PLC-c1 Y783 [12]. Phosphorylation of tyrosine residue 511 is required for the activation of Itk [31], therefore the phosphorylation of this site serves as a marker for the induction of Itk kinase activity. Unfortunately, we were unable to directly detect any TCR-induced site-specific phosphorylation of Itk in stimulated T cell lysates (data not shown). Consequently, Itk immunoprecipitations were performed to examine differences in the phosphorylation of Itk tyrosine 511. Interestingly, there was observable TCR-induced phosphorylation of Itk tyrosine 511in Jurkat E6.1 T cells, but little detectable phosphorylation of this site in HuT78 T cells and APBTs ( Figure 9A and Figure 7B). In addition, there was an increase in the amount of immunoprecipitated Itk in Jurkat E6.1 T cells compared to HuT78 T cells and APBTs ( Figure 9A

T cells have hyperphosphorylated Erk1 and Erk2 compared to HuT78 T cells and APBTs
The MAP kinases Erk1 and Erk2 are important for the regulation of the expression and function of the transcription factor AP-1 [1]. The activation of these kinases is downstream of LAT phosphorylation, Vav1 activation, and Ca 2+ influx [1]. The phosphorylation of Erk1 and Erk2 on threonine 202 and tyrosine 204 in the activation loop of their kinase domain is required for full enzymatic activity of these proteins. Thus, the dual phosphorylation of these sites is a measure of the enzymatic activity of these kinases. Due to its important role in linking cytoplasmic signaling events to changes in gene transcription, the phosphorylation of Erk1 and Erk2 was determined in the three T cell lines. A significant increase in the TCR-induced phosphorylation of Erk1 and Erk2 was observed in Jurkat E6.1 T cells compared to HuT78 T cells and APBTs ( Figure 10A and 10B). This is consistent with the increase in the phosphorylation of other important signaling molecules in Jurkat E6.1 cells and indicates that these T cells have exaggerated distal signaling events compared to HuT78 T cells and APBTs.
Jurkat E6.1 T cells, HuT78 T cells, and APBTs have distinct patterns of TCR-induced cytokine release and costimulatory receptor expression and function TCR activation leads to the increased production and release of numerous cytokines and chemokines that affect the function and localization of immune cell types. Because of the importance of cytokines and chemokines in the host immune response, we examined the release of these molecules from Jurkat E6.1 T cells, HuT78 T cells and APBTs upon TCR activation. For these studies, the T cell lines were stimulated for 24 hours using platebound stimulatory anti-TCR antibody and the release of 28 chemokines and cytokines was quantitatively measured using a cytokine multiplex assay. As seen in Table 1, APBTs released a wide range of cytokines and chemokines upon TCR stimulation. These include Th1-and Th2-associated cytokines and chemokines and the Th17 associated cytokine, IL-17 (Table 1) [32,33,34,35]. The release of such a wide range of cytokines and chemokines was due to the varied nature of the APBTs, which contain mixture of CD4+ and CD8+ T cells. In contrast, Jurkat E6.1 T cells release a more limited pool of cytokines and chemokines, with only IL-2 and the chemokines, IL-8, MIP-1a, MIP-1b and MCP-1 being produced in detectable quantities (Table 1). HuT78 T cells released a wider range of chemokines and cytokines than Jurkat E6.1 T cells, including a mixture of Th1-and Th-2 associated chemokines and cytokines. These data suggest that HuT78 T cells are more similar to APBTs than Jurkat E6.1 T cells in their TCRinduced production of cytokines and chemokines.
One important cytokine produced by TCR-stimulated T cells is IL-2, which controls T cell growth and differentiation [34]. As IL-2 production is considered an excellent way to measure cumulative activation of a T cell and it was the only cytokine produced by all three cell lines, we decided to examine the production of IL-2 by TCR-stimulated Jurkat E6.1 T cells, HuT78 T cells, and APBTs. For these studies, the T cell lines were stimulated with plate-bound stimulatory TCR antibody alone or in combination with soluble stimulatory CD28 antibody or plate-bound fibronectin, the ligand for VLA-4. These activators were chosen to investigate the coordinate induction of two well studied costimulatory receptors, CD28 and the VLA-4 integrin, and the TCR. To this end, APBTs produced IL-2 when stimulated with TCR antibody, but cotreatment with either CD28 antibody or fibronectin lead to a significant 106-126-fold increase in IL-2 production ( Figure 11). There was no significant difference between the IL-2 production in APBTs costimulated with either the CD28 antibody or fibronectin ( Figure 11). Jurkat E6.1 T cells also produced IL-2 in response to TCR antibody treatment and had approximately 36-fold more IL-2 production upon costimulation with the CD28 antibody ( Figure 11). In contrast, costimulation of Jurkat E6.1 T cells with fibronectin resulted in a small, but not significant, increase in IL-2 production over TCR stimulation alone. HuT78 T cells showed no statistically significant increase in IL-2 production when co-treated with the CD28 antibody or fibronectin compared to TCR antibody alone ( Figure 11). It is of note, however, that both Jurkat E6.1 T cells and HuT78 T cells consistently produced substantially more IL-2 than APBTs upon treatment with similar concentrations of TCR antibody alone, suggesting both cell lines have increased expression of IL-2 upon TCR activation.
To further investigate the surprising unresponsiveness of Jurkat E6.1 and HuT78 T cells to CD28 and/or VLA-4-mediated signals, we characterized the expression of several critical cell surface receptors on Jurkat E6.1 T cells, HuT78 T cells, and APBTs by flow cytometry. As expected, all three cell lines had high expression of the TCR (Table 2). In contrast, there was variable expression of the co-receptor CD4. Jurkat E6.1 T cells had little to no detectable CD4 expression whereas the majority of HuT78 T cells had a low but detectable expression of this protein ( Table 2).
APBTs had substantial percentage of cells that expressed CD4, with levels between 50%-80% depending on the donor ( Table 2). All three cells lines had high surface levels of the costimulatory receptor CD7 and the integrins LFA-1 and VLA-4 (Table 2). Interestingly, Jurkat E6.1 T cells and APBTs also had substantial expression of the costimulatory receptors CD28 and CD2, whereas HuT78 T cells had no detectable expression of these molecules ( Table 2). APBTs expressed both the activation-induced receptors CD44 and CD137 (Table 2), as expected since these cells were stimulated upon expansion. In contrast, Jurkat E6.1 cells had no detectable levels of either CD44 or CD137 (Table 2). Interestingly, HuT78 T cells had low but detectable levels of CD44 but did not expression CD137 in the unactivated state (Table 2). However, TCR stimulation could upregulated CD137 in HuT78 T cells but not in Jurkat E6.1 T cells (data not shown). These data show that Jurkat E6.1 T cells, HuT78 T cells and APBTs have different patterns of costimulatory receptor expression, indicating that Jurkat E6.1 and HuT T cells would be distinctly useful for the study of costimulatory receptors. In addition, these data suggest that the unresponsiveness of HuT78 T cells to CD28 stimulatory signals is due the lack of surface

Discussion
The inappropriate activation or suppression of T cells can be implicated in nearly all human diseases, including cancer, heart disease, multiple sclerosis, asthma, allergies, diabetes, and many others [3,4,36,37,38]. Early signaling pathways induced by both the TCR and/or costimulatory receptors are excellent targets for therapies to these diseases. Although a great deal is known about the induction of signaling by these receptors, much more must be learned before therapies can be effectively developed to target these pathways. T cell lines are an integral part of the research of T cell signaling due to their ease of use, low cost, and reduced human burden. However, not all cell lines are equally useful as a model system of human T cells. In this paper, we investigated differences in peripheral TCR signaling, cytokine release and costimulatory receptor function and expression among the Jurkat E6.1 T cell line, the HuT78 T cell line, and APBTs.
To examine the relative extent of activation of key signaling molecules in Jurkat E6.1 T cells, HuT78 T cells, and APBTs, we used a previously described method for quantifying the amount of phosphorylation per protein present in an activated T cell [24,39]. By using this method, the total phosphorylation per protein of a specific signaling molecule for each cell type was determined and compared, regardless of the total amount of protein present. This is important since there were substantial differences between the T cell lines and APBTS in the expression of several of the examined signaling proteins, including ZAP-70, Pyk2 and Vav1. By normalizing the phosphorylation level of each protein to the absolute amount expressed in each cell type, we were able to eliminate potential artifacts due to differences in expression levels.
The first signaling event to occur upon TCR activation is the stimulation of the Src family kinases Lck and Fyn [1]. We investigated the phosphorylation of Lck and Fyn at the activating tyrosine residue 394 and 420, respectively, and found that these sites had constitutive phosphorylation in Jurkat E6.1 T cells, but moderate TCR-inducible phosphorylation in HuT78s and APBTs (data not shown). However, the low level of inducible phosphorylation of Lck and Fyn made the immunoblots difficult to quantify using the described method. These differences in the regulation of Lck and Fyn did not appear to matter since only a slight decrease was seen in the phosphorylation of tyrosine 319 on the tyrosine kinase ZAP-70 in Jurkat E6.1 T cells compared to HuT78 T cells and APBTs (Figure 3). This difference in ZAP-70 activation did not translate into changes in the phosphorylation of the ZAP-70 substrates, LAT and SLP-76 ( Figure 4). Together, these results indicate that although there are small differences in events immediately following TCR activation, the relative levels of activated LAT and SLP-76 in each cell type remains the same.
The similarities between Jurkat E6.1 T cells, HuT78 T cells and APBTs ended at signaling events downstream of LAT and SLP-76. Jurkat E6.1 T cells have deficiencies in the expression of lipid phosphatases PTEN and SHIP, leading to the constitutive activation of AKT, a serine-threonine kinase important for cell survival, and other downstream signaling molecules [16,17,18]. In addition to this defect, we observed three new defects in Jurkat E6.1 T cells compared to APBTs. First, Jurkat E6.1 cells have a significant increase in the relative phosphorylation of PLCc1 tyrosine 783 compared to HuT78 T cells and APBTs ( Figure 5). In contrast, HuT78 T cells had reduced levels of PLCc1 tyrosine 775 phosphorylation compared to Jurkat E6.1 and APBTs ( Figure 5). Interestingly, these data indicate these two sites on PLCc1 may have differential regulation by intracellular kinases and phosphatases. The end result of alterations in PLCc1 phosphorylation in the various cell lines was that Jurkat E6.1 T cells exhibited a nearly 76-fold greater TCR-induced Ca 2+ flux than HuT78 T cells and APBTs ( Figure 6). In addition, Jurkat E6.1 T cells also had increased levels of site-specific phosphorylation of Vav1 and Erk1/ Erk2 (Figure 8 and Figure 10).
The reason for the distorted phosphorylation of PLCc1 and Vav1 does not appear to be due to deficiencies in the expression of PTEN and SHIP, since we have recently shown that inhibition of PI3K does not alter the site specific phosphorylation of these proteins (NCO and JCDH, manuscript submitted). The hyperphosphorylation of Erk1/Erk2 and the increased levels of TCRinduced Ca 2+ influx could, in part, be due to decreased expression of PTEN and SHIP, because PI3K signaling is needed for the optimal activation of these events in human T cells (NCO and JCDH, manuscript submitted). However, a probable mechanism for this distorted signaling appears to involve Itk, which had both higher levels of expression and phosphorylation at tyrosine 511 in Jurkat E6.1 T cells than in HuT78 T cells and APBTs (Figure 9). There was some variation in the Itk expression levels in various sublines of Jurkat E6.1 T cells, with most expressing high levels of Itk (data not shown). This indicates that genetic drift in the various sublines of this T cell line leads to differences in the expression of intracellular signaling proteins. Importantly, this is the first time that the over-expression of Itk, an crucial molecule at the intersection of TCR and costimulatory receptor-mediated signaling in T cells [40], has been observed in Jurkat E6.1 cells.
Next, we found that Pyk2 was hyperphosphorylated in TCRstimulated Jurkat E6.1 T cells compared to HuT78 T cells and APBTs (Figure 7). Again, this is the first description of the increased phosphorylation of Pyk2 in Jurkat E6.1 T cells. The mechanism for the increased phosphorylation of Pyk2 is unknown, but does not appear to be due to alterations in the expression of PTP-PEST. Pyk2 is important since it integrates receptor signals controlling actin cytoskeletal rearrangement [6]. The increased relative phosphorylation of Pyk2, in conjunction with the augmented phosphorylation of Vav1 (Figure 8), suggests that the actin cytoskeletal rearrangement in Jurkat E6.1 T cells may be altered compared to HuT78 T cells and APBTs.
Finally, Jurkat E6.1 T cells release substantially fewer cytokines and chemokines upon TCR activation than HuT78 T cells and APBTs (Table 1). However, the amount of IL-2 and IL-8 released by Jurkat E6.1 T cells is substantially greater than the levels of these hormones produced by APBTs. These differences are likely due to the thymocytic origin of the Jurkat E6.1 T cells and the fact that the cells were selected for their ability to produce large amounts of IL-2 [14].
Although this study suggests that Jurkat E6.1 T cells differ in signaling from APBTs, they do have advantages ( Table 3). The cells are incredibly resilient and easy to grow and maintain. There are also many useful mutant lines of Jurkat E6.1 T cells, each lacking a specific molecule in the peripheral TCR signaling pathway. Jurkat E6.1 T cells are remarkably easy to transfect, with transfection efficiencies routinely reaching 80-90%. Finally, a hidden advantage pointed to in this study is that they have exaggerated signaling, making changes much easier to detect. Even with the increased proximal signaling, studies using Jurkat E6.1 are likely valid since many of the discoveries and observations initially found in Jurkat E6.1 T cells have been subsequently shown in APBTs. Together, the data show that although the Jurkat E6.1 T cell line has substantially altered TCR-induced signaling, this cell line can still be a useful model for TCR signaling if deficiencies and problems are taken into account and the results are confirmed in APBTs.
Conversely, this study points to the many similarities between HuT78 T cells and APBTs. One significant difference in the relative levels of early TCR-induced signaling events between HuT78 T cells and APBTs was the reduced TCR-induced phosphorylation of Pyk2 in HuT78 T cells (Figure 7). This is likely due to the expression of PTP-PEST, a known phosphatase for Pyk2, in HuT78 T cells and not activated APBTs. However, this difference may be less important than it seems, since a recent study has shown that PTP-PEST is expressed in naïve primary T cells and not activated T cells, the subtype of APBTs used for these studies [30]. In addition, HuT78 T cells produce a greater number of cytokines and chemokines upon TCR stimulation than Jurkat E6.1 T cells. Finally, HuT78 T cell can express several costimulatory receptors found in activated T cells, allowing investigation of these costimulatory pathways in a model T cell line. HuT78 T cells also have disadvantages, however. They do not express the costimulatory receptors CD2 or CD28 (Table 2), and are unable to be co-stimulated by fibronectin, suggesting that although VLA-4 is expressed, it may not be functional in these cells ( Figure 11). In addition, HuT78 cells have elevated levels of IL-2 production without costimulation as compared to APBTs ( Figure 11). HuT78 T cells also have abnormal c-myc function and over express Bcl-xL, which indicates possible problems with the anti-apoptotic pathway in these cells [19,20]. All of this information must be kept in mind when choosing to use HuT78 T cells as a model system for T cell biology.
The results of this study indicate that HuT78 T cells are similar to APBTs in peripheral TCR signaling, whereas Jurkat E6.1 T cells exhibit exaggerated signaling (Table 3). In addition, there were differences in cytokine release and costimulatory receptor expression and function between the various cell lines. These results are significant in that, for the first time, a direct comparison of commonly used cell lines and APBTs has been performed. Though it is clear that APBTs are the ideal cells to use for examining TCR signaling, this is not always feasible. Both Jurkat E6.1 T cells and HuT78 T cells have advantages and disadvantages, but HuT78 T cells are clearly the most like APBTs in the phosphorylation of important peripheral TCR signaling molecules and cytokine release. The relative advantages and disadvantages of each cell line must be taken into account when choosing a model cell line for individual experiments.

Ethics statement
This study was conducted according to the principles expressed in the Declaration of Helsinki. The study was approved by the Institutional Review Board of University of Iowa. All donors provided written informed consent for the collection of samples and subsequent analysis.

Cell line stimulation
The Jurkat E6.1 T cells were a kind gift of Dr. Lawrence Samelson. Jurkat E6.1 T cells were grown at 37uC in 5% CO 2 using complete RPMI 1640 media (RPMI 1640 supplemented with 10% FBS, 2 mM L-glutamine, 50 U/mL penicillin, and 50 mg/mL streptomycin). HuT78 T cells were purchased from ATCC. HuT78 T cells were grown in IMDM media supplemented with 10% FBS, 2 mM L-glutamine, 50 U/mL penicillin, and 50 mg/mL streptomycin. Both cell lines were grown to a concentration of 2-5610 5 cells/mL. Cells were washed with RPMI 1640 before use and then resuspended at a density of 5610 7 cells/mL. Each sample of 4610 6 cells was then treated with 10 mg/mL CD3 antibody (OKT3) for various times. Samples were lysed with 4-fold excess hot 26 sample buffer (20 mM Tris pH 8.0, 2 mM EDTA, 2 mM Na 3 VO 4 , 20 mM DDT, 2% SDS, and 20% glycerol), heated at 95uC for 4 minutes, and sonicated to reduce viscosity.

APBT isolation and stimulation
All the studies described below were approved by the Human Subjects Institutional Review Board at the University of Iowa. Healthy blood donors were recruited by word-of-mouth from the students and staff of biomedical laboratories at the Carver College of Medicine at the University of Iowa. Informed consent was obtained from all donors before commencement of blood draws. Peripheral blood T cells were isolated from the venous blood of healthy human donors as previously described [24]. The cells were then grown in culture for 5-7 days in complete RPMI 1640 media supplemented with 10 ng/mL IL-2. After incubation, the T cells were .90% positive for TCR a/b chains and .97% positive for CD28, CD2, CD7, CD11a and CD49d. Depending on the donor, the T cells were 30-70% positive for CD4, with the typical donor having ,65% CD4 positive T cells.
To stimulate the cells, the T cells were washed with RPMI 1640 and resuspended to a concentration of 1610 8 cells/mL. After a short incubation on ice, cells were treated with 10 mg/mL CD3 (OKT3) and 10 mg/mL CD4 antibodies and returned to ice for 30 minutes. Both CD3 and CD4 antibodies were used since only concurrent stimulation with these antibodies, and not CD3 antibody treatment alone, could induce detectable TCR-mediated signaling [24]. The cells were warmed at 37uC for 10 minutes and immediately stimulated with 25 mg/mL mouse specific goat IgG antibody for various times. Samples were lysed with a 2-fold excess of hot 26 sample buffer, heated for 4 minutes at 95uC, and sonicated to reduce viscosity.

Immunoblotting
The protein components of 2610 5 cell equivalents/ sample were separated by SDS-PAGE using 4-15% or 15% Criterion pre-cast polyacrylamide gels. The separated proteins were then transferred to PVDF and blocked with 5% milk in TBST for 45 minutes at room temperature. The membranes were then incubated at room temperature with primary antibody diluted in 5% milk for 1 hour, followed by the appropriate secondary antibody diluted in 5% milk for 30 minutes. The immunoblots were then visualized by chemiluminescence using a Fuji imager.

Analysis of immunoblotting
Immunoblots were analyzed using NIH Image J. The ratio of phosphospecific to pan expression of each protein was determined for stimulated and unstimulated samples in each cell type, allowing the determination of the phosphorylation per protein in each cell. Stimulated sample ratios for each cell type were then averaged and compared to the other cell types using a two-tailed t-test assuming equal variance. Whole cell lysate samples used to compare Itk expression were also analyzed using NIH Image J and normalized using actin expression. The ratio of protein: actin was compared between cell lines using a two-tailed t-test assuming equal variance.

Immunoprecipitations
Jurkat E6.1 T cells, HuT78 T cells, and APBTs were grown as indicated above. The cells were washed in RPMI 1640 and resuspended to a concentration of 5610 7 cells/mL. 100 mL of cells/sample were stimulated as described above for various times and lysed in 400 mL lysis buffer (25 mM TRIS pH 8.0, 150 mM NaCl, 1% Brij 97, 0.5% n-Octyl-beta-D-glucopyranoside, 5 mM EDTA, 1 mM Na 3 VO 4 , and protease inhibitor). Samples were incubated on ice with mixing for 20 minutes and centrifuged. Supernatant was added to 20 mL protein A/G agarose beads and rotated at 4uC for 30 minutes. Beads were pelleted and the supernatant was applied to 25 mL protein A/G agarose beads and 2 mg anti-Itk antibody and rotated at 4uC overnight. Pelleted beads were washed 4 times with 800 mL lysate buffer (25 mMTRIS pH 8.0, 150 mM NaCl, 1% Brij 97, and 0.5% octyl glucoside) and then the bound proteins were eluted with 30 mL hot 26 sample buffer. Samples were heated at 95uC for 4 minutes.

Calcium influx
Calcium influx was measure quantitatively as previously described [34,35]. Briefly, Jurkat E6.1, HuT78, and APBTs were grown as indicated above. The cells were washed in RPMI 1640 without phenol red and resuspended to a concentration of 1610 6 cells/mL in 1 mL RPMI 1640 without phenol red, 1 mM probenecid, and 5 mM indo-1-AM. The cells were incubated for 1 hour at 37uC. After incubation the cells were washed twice with RPMI 1640 without phenol red. The pellet was then resuspended in RPMI 1640 without phenol red and 1 mM probenecid. The cells were kept on ice and in the dark until use.
Measurements were done at 37uC with constant stirring in acrylic cuvettes using a Fluorolog fluorimeter. The cells were excited at 350 nm and emission was measured at l 1 = 405 nm and l 2 = 485 nm. Fluorescence at l 1 and l 2 was measured in unstimulated cells every 10 seconds for a total of 60 seconds. The cells were then stimulated with 10 mg/mL CD3 antibody and fluorescence at l 1 and l 2 measured every 10 seconds for a total of 360 seconds. Cells were then lysed with 0.1% Triton x-100 and florescence was measured at both l 1 and l 2 . 2 mM EGTA was then added to sequester the Ca 2+ from the dye and again, fluorescence was measured at both emission wavelengths. Ca 2+ influx was then determined using the equation [Ca 2+ ] = K D [(R2Rmin )/R max 2R)](S f2 /S b2 ), where K D is the dissociation constant for indo-1 and Ca 2+ , R is the ratio of l 1 and l 2 at a specific time, R max is the ratio of l 1 and l 2 after the cells are lysed with 0.1% Triton x-100, R min is the ratio of l 1 to l 2 after treatment with 2 mM EGTA, and S f2 /S b2 is the ratio of free to bound dye as measured at l 2 [34,35].

Cytokine and chemokine measurements
Jurkat E6.1 T cells, HuT78 T cells, and APBTs were grown as indicated above. The cells were washed in RPMI 1640 and resuspended in complete RPMI 1640 media to a concentration of 1610 6 cells/mL. 0.5610 6 cells were then added to 24 well tissue culture plates containing 2 mg/mL plate-bound OKT3 in the absence or presence of 2 mg/mL soluble CD28 antibody or 1 mg/ mL human fibronectin. The cells were then incubated for 24 hours at 37uC in 5% CO 2 .
The presence of cytokines and chemokines in the media were assessed using a human cytokine 30-plex luminex assay. Briefly, a mixture of fluorescent beads, each containing an antibody to one of 30 different human cytokines or chemokines, were incubated with standards of known quantities of each cytokine and chemokine or supernatants from treated cells for two hours at RT. A mixture of biotinylated antibodies to each cytokine or chemokine was added to the beads for 1 hour at RT and then streptavidin labeled with the fluorescent dye RPE was added for 30 minutes at RT. The amount of bound streptavidin to each type of fluorescent bead was then assessed using Bio-Rad Bio-Plex system.
The presence of IL-2 in the media was determined by standard sandwich ELISA. Briefly, a monoclonal IL-2 antibody was coated onto a RIA 96-well plate overnight at 4uC. Triplicates of IL-2 standards and supernatants from treated cells were then added to the plate and incubated overnight at 4uC. A biotinylated polyclonal IL-2 antibody was added to the plate for 2 hours at RT and then streptavidin-horse radish peroxidase (HRP) was added for 1 hour at RT. The amount of bound streptavidin was then assessed using TMB peroxidase that was acidified using 0.67 N H 2 SO 4 . The absorbance of each well at 450 nM was then measured using spectrophotometric plate reader.

Flow cytometry
Jurkat E6.1 T cells, HuT78 T cells, and APBTs were grown as indicated above. The cells were washed in FACS Buffer (PBS with 10% fetal calf serum and 0.05% sodium azide) and resuspended in FACS Buffer to a concentration of 1610 6 cells/mL. 0.5610 6 cells were then incubated with isotype control or indicated primary antibodies for 30 minutes on ice. The cells were washed in FACS Buffer and again resuspended in FACS Buffer to a concentration of 1610 6 cells/mL. The cells were then incubated with a PElabeled mouse IgG antibody for 30 minutes on ice. This step was eliminated for the TCR a chain, CD3 and CD28 antibodies, which were all directly conjugated to fluorescent dyes. The cells were washed in FACS Buffer and again resuspended in FACS Buffer to a concentration of 1610 6 cells/mL. The amount of primary and secondary antibody bound to each cell was these examined using a Becton Dickinson FACScan flow cytometer. The flow cytometry data was then analyzed using the FloJo analysis program. The percentage of positive cells were identified by determining the percentage of cells with fluorescence .98% of the isotype control. Mean fluorescence intensity was determined using FloJo analysis program. The differences in TCR surface expression was statistically compared between cell lines using a two-tailed t-test assuming equal variance.