Open Access Highly Accessed Research article

Systematic identification of regulatory proteins critical for T-cell activation

Peter Chu1, Jorge Pardo1, Haoran Zhao1, Connie C Li14, Erlina Pali1, Mary M Shen1, Kunbin Qu1, Simon X Yu1, Betty CB Huang1, Peiwen Yu14, Esteban S Masuda1, Susan M Molineaux1, Frank Kolbinger2, Gregorio Aversa3, Jan de Vries3, Donald G Payan1* and X Charlene Liao15*

Author Affiliations

1 Rigel Pharmaceuticals Inc., 1180 Veterans Blvd., South San Francisco, CA 94080, USA

2 Novartis Pharma AG, S-386.6.25, CH-4002 Basel, Switzerland

3 Novartis Forschungsinstitut GmbH, Brunner Strasse 59, A-1235 Vienna, Austria

4 Current address: Exelixis Inc., 170 Harbor Way, South San Francisco, CA 94083, USA

5 Current address: Genentech Inc., 1 DNA Way, South San Francisco, CA 94080, USA

For all author emails, please log on.

Journal of Biology 2003, 2:21  doi:10.1186/1475-4924-2-21

Published: 15 September 2003

Additional files

Additional data file 1:

(a) Jurkat cells were stimulated with anti-TCR antibody C305 for 24 h and then stained with anti-CD69-APC and analyzed on a FACSCalibur. The dashed line indicates CD69 level before stimulation and the solid line after stimulation. (b) A diagram of the TRA-Lyt2 construct and the Dox-off system. (c) A Jurkat clone (4D9) with optimal CD69 expression profile was infected with pTRA-Lyt2 and a retroviral construct that constitutively expresses tTA to obtain the Jurkat-tTA cell clone 4D9#32. The solid line indicates Lyt2 level with Dox (10 ng/ml) and the dashed line without Dox, in clone 4D9#32.

Format: PDF Size: 64KB Download file

This file can be viewed with: Adobe Acrobat Reader

Open Data

Additional data file 2:

(a) Diagram of the retroviral library vector used to make the pTRA-cDNA libraries (top) and after it has stably integrated into the genome (bottom). (b) Efficiencies of transfection (top panels) and infection (bottom panels). Phoenix A cells were transfected with TRA-GFP in parallel to the pTRA-cDNA libraries; 24 h later, TRA-GFP-transfected Phoenix A cells were analyzed for GFP expression by FACS. The percentage of GFP-positive cells (66%) from TRA-GFP-transfection was taken as the efficiency of library transfection (top left panel: mock-transfected Phoenix A cells; top right panel: TRA-GFP-transfected Phoenix A cells). Two days after transfection, the viral supernatant harboring pTRA-cDNA libraries or TRA-GFP were used to infect Jutkat-tTA cells (clone 4D9#32) in parallel. Then, 60h after infection, TRA-GFP-infected Jurkat-tTA cells were analyzed for GFP expression by FACS. The percentage of GFP-positive cells (52%) from TRA-GFP-infection was taken as the efficiency of library infection (bottom left panel: mock-infected Jurkat-tTA cells; bottom right panel: TRA-GFP-infected Jurkat-tTA cells).

Format: PDF Size: 86KB Download file

This file can be viewed with: Adobe Acrobat Reader

Open Data

Additional data file 3:

Stimulation of Jurkat cells with anti-TCR antibody (C305) reduced cell surface expression of CD3. Jurkat-tTA cells were stained with anti-CD3-PE antibody before and after overnight stimulation with C305 (300 ng/ml). The dashed line indicates the CD3 level of unstimulated cells and the solid line the CD3 level of stimulated cells. TCR stimulation reduced cell-surface expression of CD3, as expected for antibody-mediated receptor internalization [52]. Importantly, it was possible to separate the cells expressing a low level of CD3 after stimulation (CD3low) from those not expressing CD3 at all (CD3-); both unstimulated and stimulated Jurkat-tTA cells have a minor population (around 2%) of CD3- cells.

Format: PDF Size: 28KB Download file

This file can be viewed with: Adobe Acrobat Reader

Open Data

Additional data file 4:

(a) Distribution of the Dox ratio (limited to between 1.5 to 10) among 1,186 cell clones. A total of 2,828 cell clones selected based on the CD69lowCD3+ criteria of the T-cell activation screen were assayed for CD69 expression after stimulation in the presence or absence of Dox, resulting in 1,323 clones with a Dox-regulatable phenotype. CD69 geometric mean fluorescent intensity in the presence of Dox was divided by that in the absence of Dox to give rise to the Dox ratio for each individual clone. The majority of Dox-regulatable clones had a Dox ratio between 1.5 and 10 (1,186 clones out of 1,323, or 89.6%). (b) Distribution of Dox ratios among all 2,828 cell clones is shown using the number of clones on the y axis and the Dox ratio on the x axis. The mean Dox ratio for the 2,828 clones was 3.02 ± 10.38 (standard deviation). There were rare clones with a Dox ratio up to 70. These few clones with high Dox ratios contributed to the higher standard deviation for the collection of 2,828 clones. Importantly, the mean Dox ratio for the 2,828 clones was significantly different from the mean of unsorted cells (3.02 versus 1.00), indicating that the sorting and the regulated expression of cDNA libraries worked efficiently. Interestingly, a number of clones gave a Dox ratio that was < 1. This implies that these clones may express proteins that positively regulate TCR function. While we have focused on the dominant-negative regulators of T-cell signaling in this report, we expect that a fuller investigation of these clones would be warranted.

Format: PDF Size: 30KB Download file

This file can be viewed with: Adobe Acrobat Reader

Open Data

Additional data file 5:

(a) Details of selected cell clones from which various TCRβ hits were identified. Nineteen randomly chosen cell clones that contributed to various TCRβ hits and five randomly chosen cell clones that contributed to various other hits (given in parentheses) were left unstimulated, or stimulated with C305 (anti-TCRβ), anti-CD3 or PMA for 24 h. CD69 surface expression was measured with anti-CD69-APC before and after stimulation, whereas CD3 surface expression was measured only with resting cells. The fold induction was obtained by dividing CD69 geometric mean after stimulation by that of the same clone before stimulation. In control Jurkat-tTA cells, C305 gave a much stronger induction of CD69 expression (47.5-fold) than did anti-CD3 (6.4-fold). With cells identified to have a CD69-inhibitory phenotype and later shown to harbor other hits such as Csk, Nucleolin, SHP-1, Moesin, and Ku70, C305 stimulation was severely compromised (1.5-2.6-fold). Nevertheless, with the exception of one, these cells responded better to C305 stimulation than to anti-CD3 stimulation. With cells harboring the TCRβ hits, however, a reverse trend was seen. With these cells, anti-CD3 stimulation gave a higher induction of CD69 expression than did C305 stimulation. This observation supported our hypothesis that various TCRβ hits may compete with the endogenous Jurkat clonotypic TCRβ to form a new TCR-CD3 complex that is no longer recognized by C305 but may still be recognized by anti-CD3. (b) FACS data for four clones harboring TCRβ hits. Left panels are CD3 staining patterns (filled peaks). Central panels are CD69 histograms before (thin lines) and after (thick red lines) C305 stimulation. Right panels are CD69 histograms before (thin lines) and after stimulation with anti-CD3 (thick lines) or PMA (dashed lines).

Format: PDF Size: 113KB Download file

This file can be viewed with: Adobe Acrobat Reader

Open Data

Additional data file 6:

Summary of the genetic screen for inhibitors of TCR-induced CD69 expression.

Format: ZIP Size: 49KB Download file

Open Data

Additional data file 7:

Diagrams of proteins predicted from the cDNA inserts and those from the corresponding wild-type genes are shown above the histograms. Each histogram shows the Dox-regulatable phenotype of the original cell clones. These original cell clones were grown in the presence or absence of Dox for 6 days and then stimulated overnight with the anti-TCR antibody. Cells were stained with anti-CD69-APC and analyzed by FACS. The filled peaks indicate CD69 expression level in the absence of Dox, when the cDNA hits were expressed. The open peaks indicate CD69 expression level in the presence of Dox, when the cDNA hits were not expressed. The Dox ratio is shown above the histograms. The following cDNA hits are shown: (a) Syk; (b) SHP-1 (PTP1C); (c) PAG; (d) A-Raf-1; (e) TCPTP; (f) IL-10Ra; (g) integrin a2; (h) SH2-B; (i) GG2-1; (j) moesin; and (k) vimentin. Open file for more details.

Format: PDF Size: 96KB Download file

This file can be viewed with: Adobe Acrobat Reader

Open Data

Additional data file 8:

Stronger CD69 inhibitory phenotypes correlate with higher expression levels of the cDNA hits. The cDNA inserts such as (a) Lck and PLCγ1 or (b) TCPTP and Grb7 were subcloned in front of IRES-GFP in the Tet-regulated retroviral vector pTRA-IRES-GFP where the expression of hits was positively correlated with the expression of GFP. After retroviral infection, the Jurkat-tTA cells were left unstimulated (left panels), or stimulated with C305 (300 ng/ml, center panels) or PMA (5 ng/ml, right panels), and analyzed by FACS for CD69 induction after staining with anti-CD69-APC. The dot plots show CD69 expression along the y axis and GFP expression along the x axis. A diagonal shape of the dots in C305-stimulated samples indicates that cells expressing higher GFP have a lower CD69 induction (for Lck, PLCγ1 and TCPTP). Other hits (except Grb7) showed the same trend. In addition, all of the hits, with the exception of A-Raf-1, had no inhibitory effect on PMA induction of CD69 expression, suggesting that they affect membrane-proximal signaling.

A dozen or so hits did not confer CD69 inhibition when transferred to naïve Jurkat cells. There may be a few reasons for this observation. The hits may be false positives from the genetic screen. They may, on the other hand, be genuine (positive) hits from the screen. In this case a lack of phenotype upon transferring the cDNA hits to naïve Jurkat cells could result from one or a combination of the following reasons: first, the hit may be inserted in a favorable chromosomal region that allowed for high-level expression whereas the subclone in front of IRES-GFP has a compromised expression level correlating to the lack of phenotype on transfer; second, the hit may contribute to the CD69-inhibition phenotype in the original cell clone where its chromosomal integration may also accidentally disrupt a locus important for CD69 upregulation; third, there may be more than one retroviral insert in the original cell clone where the CD69-inhibition phenotype was seen, and one of them was cloned and listed in Table 1, but perhaps such a clone alone was not sufficient to confer CD69 inhibition phenotype. Given the above circumstances, it is quite possible that the hits that failed to produce a transfer phenotype may still be somehow important to TCR signal transduction leading to CD69 upregulation.

Format: PDF Size: 102KB Download file

This file can be viewed with: Adobe Acrobat Reader

Open Data