Src-like adaptor protein regulates TCR expression on ... - Nature

© 2006 Nature Publishing Group http://www.nature.com/natureimmunology

ARTICLES

Src-like adaptor protein regulates TCR expression on thymocytes by linking the ubiquitin ligase c-Cbl to the TCR complex Margaret D Myers1, Tomasz Sosinowski2, Leonard L Dragone3, Carmen White3, Hamid Band4, Hua Gu5 & Arthur Weiss1 The adaptor molecule SLAP and E3 ubiquitin ligase c-Cbl each regulate expression of T cell receptor (TCR)–CD3 on thymocytes. Here we provide genetic and biochemical evidence that both molecules function in the same pathway. TCR-CD3 expression was similar in the absence of SLAP and/or c-Cbl. SLAP and c-Cbl were found to interact, and their expression together downregulated CD3e. This required multiple domains in SLAP and the ring finger of c-Cbl. Furthermore, expression of SLAP and c-Cbl together induced TCRf ubiquitination and degradation, preventing the accumulation of fully assembled recycling TCR complexes. These studies indicate that SLAP links the E3 ligase activity of c-Cbl to the TCR, allowing for stage-specific regulation of TCR expression.

Thymocytes progress through multiple developmental stages that ensure the generation of T cells expressing T cell receptors (TCRs) with an appropriate range of avidities for major histocompatibility complex molecules1. At the CD4+CD8+ double-positive (DP) stage of development, thymocytes that have productively rearranged the genes encoding both the TCRa and TCRb chains begin to express small amounts of the mature TCRab on the surface together with the CD3 chains. Expression of the TCR at this stage of development is crucial, as signals through the TCR are required for either survival (positive selection) or deletion (negative selection) of developing thymocytes. Positively selected thymocytes downregulate either CD8 or CD4 expression and upregulate TCR expression to become CD4+ or CD8+ single-positive (SP) thymocytes. Finally, SP thymocytes undergo further maturation and selection processes before exiting the thymus as naive T cells. Whereas DP cells have a strict requirement for signals through the TCR for both positive and negative selection, TCR expression on DP thymocytes is less than 10% of the expression on SP thymocytes and peripheral T cells2,3. That finding suggests that developing thymocytes actively maintain low surface TCR expression, potentially to decrease the sensitivity of thymocytes for interactions with self peptide–major histocompatibility complex molecules so that quantitative differences in signaling can be distinguished more accurately. Thymocytes use multiple mechanisms to regulate TCR expression during thymocyte

development. First, the rate of TCR assembly is kept low because of rapid degradation of newly synthesized TCRa chains4. In the absence of TCRa, assembly of the complex proceeds relatively inefficiently and the remaining chains of the TCR complex either remain in the endoplasmic reticulum or are transported out of the endoplasmic reticulum and are degraded. In addition, the Src-like adaptor protein (SLAP) targets TCRz chains present in mature TCR complexes for degradation, thus preventing the accumulation of fully assembled complexes capable of recycling back to the plasma membrane5. SLAP is an adaptor that shares a considerable degree of similarity with Src family kinases6. Like Src kinases, SLAP has a unique, myristolated N terminus, followed by a Src homology 3 (SH3) domain and an SH2 domain, which are 55% and 50% homologous to the SH3 and SH2 domains of the kinase Lck, respectively. However, unlike the C termini of Src kinases, SLAP contains a unique C terminus of approximately 100 amino acids. Overexpression studies in cell lines have indicated that SLAP inhibits signaling ‘downstream’ of Src family kinases after stimulation through either the TCR or the plateletderived growth factor receptor, perhaps by functioning in a dominant negative way7,8. Consistent with that hypothesis, both the SH3 and SH2 domains of SLAP are required for optimum inhibition of signaling ‘downstream’ of the TCR8. Although it is clear that SLAP can inhibit signaling when overexpressed, it is not known whether SLAP can inhibit Src kinase function when expressed in physiological

1Department of Medicine, Rosalind Russell Medical Research Center for Arthritis, Howard Hughes Medical Institute, University of California San Francisco, San Francisco, California 94143, USA. 2Howard Hughes Medical Institute and National Jewish Medical and Research Center, Denver, Colorado 80206, USA. 3Division of Pediatric Immunology/Rheumatology, Department of Pediatrics, University of California San Francisco, San Francisco, California 94143, USA. 4Division of Molecular Oncology, Department of Medicine, Evanston Northwestern Healthcare Research Institute, Feinberg School of Medicine; Department of Biochemistry, Molecular Biology & Cell Biology and Robert H. Lurie Comprehensive Cancer Center, Northwestern University, Evanston, Illinois 60201, USA. 5Microbiology Department, Columbia University, College of Physicians and Surgeons, New York, New York 10032, USA. Correspondence should be addressed to A.W. ([email protected]).

Received 11 July; accepted 5 October; published online 4 December 2005; doi:10.1038/ni1291

NATURE IMMUNOLOGY

VOLUME 7

NUMBER 1

JANUARY 2006

57

Sla–/–

WT

a

CD3ε

Cbl –/–

RESULTS SLAP and c-Cbl function in the same ‘genetic pathway’ To determine whether SLAP and c-Cbl function in the same ‘genetic pathway’, we crossed mice deficient in either SLAP or c-Cbl to generate Sla–/–Cbl–/– double-deficient mice and assessed surface receptor

Sla–/–Cbl –/–

CD4

CD5

d

CD69

DP

CD4 SP

100 101 102 103 104 100 101 102 103 104 100 101 102 103 104 100 101 102 103 104 100 101 102 103 104

b Expression ('fold WT')

12.5

Percent of total

TCRβ

after TCR stimulation12, is composed of an N-terminal tyrosine kinase–binding domain, a zinc-binding ring finger domain, several proline-rich regions and a C-terminal leucine zipper13. The ring finger of c-Cbl has E3 ubiquitin ligase activity14. E3 ligases are a component of the ubiquitination machinery that catalyzes the transfer of ubiquitin molecules to protein substrates, thus targeting the protein for proteasomal and/or lysosomal degradation15. Therefore, E3 ligases have a crucial function in the ubiquitination process by providing target specificity. Many proteins expressed by T cells, including the TCRz chain, have been identified as targets for ubiquitination13,16–19. In an overexpression system, it has been shown that the tyrosine kinase Zap70 can recruit c-Cbl to the TCRz chains, thus targeting TCRz for ubiquitination19. However, TCR expression is not upregulated on Zap70–/– DP thymocytes, suggesting that additional molecules may be needed to regulate TCR expression in the thymus. Therefore, we hypothesized that SLAP functions as an adaptor for c-Cbl and is required for the recruitment of c-Cbl to the TCR complex. Here we demonstrate that thymocytes deficient in both SLAP and c-Cbl had a phenotype similar to that of thymocytes deficient in either SLAP or c-Cbl alone. In addition, we provide evidence that SLAP and c-Cbl functionally cooperated in the same pathway to regulate TCR expression by targeting the TCRz chain for ubiquitination and degradation.

WT

–/–

Sla

Cbl

–/–

–/–

Sla

Cbl

Thymus

LN

50

10

Percent of total

amounts. Regardless, the C terminus of SLAP is not required for inhibition of Src kinase signaling, and the function of this domain is still unclear8. It has been suggested that the C terminus of SLAP can bind to the E3 ubiquitin ligase c-Cbl9, indicating that SLAP and c-Cbl may function in the same pathway. A further link between SLAP and c-Cbl is suggested from the characterization of mice deficient in either molecule. Mice of each genotype have increased expression of surface TCR-CD3, CD4, CD5 and CD69 molecules specifically on DP thymocytes10,11. Moreover, thymocytes deficient in either SLAP or c-Cbl have an increase in positive selection in the presence of the ovalbumin peptide–specific, major histocompatibility complex class II–restricted DO11.10 transgenic TCR. Notably, the stage-specific thymic phenotypes found in mice deficient in SLAP (Sla–/–) or c-Cbl (Cbl–/–) are consistent with the expression pattern of these molecules. Expression of c-Cbl is detected mainly in hematopoietic cells and the testis, with the highest expression in the thymus. Similarly, whereas Sla mRNA is expressed in a variety of hematopoietic tissues, SLAP protein seems to be even more developmentally restricted, with the highest expression in DP thymocytes. SLAP modulates TCR expression on DP thymocytes by targeting phosphorylated TCRz chains present in fully assembled TCR complexes for degradation5. In the absence of TCRz, the remainder of the complex is either degraded or retained in an intracellular compartment. Although it is clear that SLAP is required to target the TCRz chain for degradation, the mechanism responsible for this process is still unknown. Similarly, it is unclear how c-Cbl modulates TCR expression on DP thymocytes. c-Cbl, a large, multidomain protein containing multiple tyrosine residues that become phosphorylated

7.5 5 2.5

40 30 20 10

0 Zap70 + – – – – Sla + + – + – Cbl + + + – – CD4 SP

+ – – – – + + – + – + + + – – CD8 SP

0 Zap70 Sla Cbl

+ – – – – + + – + – + + + – – CD4

+ – – – + + – + + + + – CD8

– – –

–/–

5 TCRβ

4

CD3ε

CD4

CD5

CD69

3 2 1 0

DP

SP

DP

SP

DP

c 7.8

SP

DP

Zap70 –/– Sla–/–

Zap70 –/–

WT 0.8

SP

DP

SP

Zap70 –/– Sla –/– Cbl –/–

Zap70 –/– Cbl –/– 6.5

4.7

2.5

Thymus

32.4

0.7

0.2

1.2

2.8

2.7

1.1

0.6

0.6 5.6 LN

CD4


ARTICLES

18.4

0.4

0.1

0.1

Figure 1 Sla–/–, Cbl–/– and Sla–/–Cbl–/– DP thymocytes have a similar phenotype. (a) Expression of cell surface molecules on DP and CD4+ SP thymocytes, assessed by flow cytometry. WT, wildtype. (b) Quantification of flow cytometry staining in a, presented as the ‘fold’ mean fluorescent intensity ± s.e.m. compared with that of DP or SP thymocytes from wild-type thymi (‘fold WT’). Data are from one experiment of three mice per genotype and are representative of three independent experiments. (c) Loss of SLAP and/or c-Cbl partially restores the development of Zap70–/– SP thymocytes and peripheral T cells. Thymocytes (Thymus) or lymph node T cells (LN) were stained for CD4 and CD8 and were analyzed by flow cytometry (genotypes, above dot plots). Numbers in quadrants indicate the frequency of SP thymocytes and lymph node T cells. (d) Average frequency ± s.e.m. of CD4+ and CD8+ SP thymocytes or lymph node T cells for eight mice per genotype (genotypes, below graphs).

0.2

CD8

58

VOLUME 7

NUMBER 1

JANUARY 2006

NATURE IMMUNOLOGY

ARTICLES

c

C

Sla –/– WT Sla –/– Cbl –/– Cbl –/– Time (h) 0 2 4 6 0 2 4 6 0 2 4 6 0 2 4 6

–/ – –/ –

W T Sl a C b Sl l a

α-tubulin

α-tubulin

CD3ε

CD3ε

TCRζ

TCRζ

WT Sla –/– –/– Cbl –/– Sla Cbl –/–

4 3 2 1

d

75 50 25

e

CD3ε Sla –/–

WT

0

2 4 Time (h)

Cbl –/–

100

6

75 50 25 0

0

2 4 Time (h)

6

Sla –/–Cbl –/–

DP RPMI

Sla –/–Cbl –/–

125

100

0 TCRζ

Cbl –/–

Sla –/–

WT

125 TCRζ (% t = 0)

TCR-CD3 ('fold WT')

5

0

SP Sucrose

RPMI

250

Sucrose

1,500 CD3ε (geoMFI)

CD3ε (geoMFI)


b

CD3ε (% t = 0)

–/ –

bl – /–

a

200 150 100 50 0

1,200 900 600 300 0

0

2 4 6 0 2 4 6 Time (h) Time (h)

0

2 4 6 0 2 4 6 Time (h) Time (h)

expression by flow cytometry. As reported before10,11, TCRb and CD3e expression was approximately three- to fourfold higher on Sla–/– or Cbl–/– DP cells than on wild-type control cells (Fig. 1a,b). In contrast, TCR-CD3 expression was not altered on SP thymocytes regardless of genotype. Expression of TCR-CD3 on DP Sla–/–Cbl–/– thymocytes was also increased; however, the increase in TCR-CD3 expression was neither additive nor synergistic. In addition, Sla–/–Cbl–/– mice had a modest 1.5-fold increase in CD4 expression on DP but not SP thymocytes, which was similar to CD4 expression on DP thymocytes deficient in either molecule alone (Fig. 1a,b). CD5 and CD69 are cell surface molecules that are upregulated on thymocytes in response to TCR signals. CD5 is a negative regulator of signal transduction whose surface expression reflects the amount of TCR signaling received by developing thymocytes20. CD69 is an early activation marker that is induced on thymocytes in response to TCR signaling and is associated with thymocytes that have been positively selected21. Expression of CD5 and CD69 is increased on both Sla–/– and Cbl–/– DP but not SP thymocytes10,11. Expression of CD5 and CD69 was also increased on Sla–/–Cbl–/– DP thymocytes and was similar to the expression on thymocytes deficient in either molecule alone (Fig. 1a,b). In contrast, expression of CD5 and CD69 was similar on SP thymocytes regardless of genotype. Zap70 is essential for the transmission of TCR signals in both thymocytes and peripheral T cells22. Because signals through the TCR are required for T cell development, Zap70–/– thymocytes are blocked at the DP stage of development23,24. Consequently, Zap70–/– mice have very few SP thymocytes and peripheral T cells. Loss of SLAP expression can partially ‘rescue’ the developmental defect noted in the absence of Zap70 (ref. 11). The development of Zap70–/– CD4+ SP thymocytes and peripheral T cells was also partially restored in the absence of c-Cbl alone and in the absence of both SLAP and c-Cbl (Fig. 1c,d). ‘Rescue’ of CD4+ SP cells was somewhat additive in the thymus; however, loss of both SLAP and c-Cbl could not completely compensate for loss of Zap70, especially in the peripheral T cell compartment.

NATURE IMMUNOLOGY

VOLUME 7

NUMBER 1

JANUARY 2006

Figure 2 Increased TCRz expression and CD3e recycling pool due to a defect in TCRz degradation. (a) Immunoblot analysis of CD3e and TCRz in purified CD8+ thymocytes (more than 95% DP). a-tubulin, loading control. (b) Quantification of immunoblots in a by quantitative luminescence. Data represent the mean ( ± s.e.m.) of CD3e or TCRz expression by thymocytes compared with that of wild-type control cells (‘fold WT’). Expression of TCR and CD3 is normalized to that of a-tubulin. (c) Immunoblot analysis of CD3e and TCRz expression by CD8+ thymocytes (more than 95% DP) after treatment with cycloheximide (time, above lanes). (d) Quantification of CD3e and TCRz expression in c. Data are presented as the mean ( ± s.e.m.) percent expression relative to expression at ‘time zero’ (t ¼ 0). One wild-type and one Sla–/– mouse are included as controls (c,d). (e) Mean fluorescent intensity ( ± s.e.m.) of CD3e expression on DP or SP thymocytes (genotypes, key) incubated in cell culture medium with (Sucrose) or without (RPMI) 0.45 M sucrose. geoMFI, geometric mean fluorescence intensity. For each experiment, three Cbl–/– mice and three Sla–/–Cbl–/– mice were analyzed and one wild-type mouse and one Sla–/– mouse were used for comparison.

Expression of the TCRz chain is increased in both Sla–/–and Cbl–/– thymocytes5,25. The increase in TCRz expression by Sla–/– thymocytes is due to a defect in TCRz degradation5. To analyze TCRz expression by Sla–/–Cbl–/– DP thymocytes, we enriched thymocyte suspensions for CD8+ cells using magnetic beads. This method allowed us to obtain thymocytes that were more than 95% DP, as determined by flow cytometry (data not shown). Immunoblot analysis of TCRz expression by purified CD8+ Sla–/–Cbl–/– thymocytes demonstrated that total TCRz expression was increased three- to fourfold compared with that of wild-type control thymocytes (Fig. 2a,b). In contrast, expression of CD3e remained similar in all of the genotypes tested, presumably because of the large intracellular pool of unassembled CD3e chains in the endoplasmic reticulum of DP thymocytes4. To analyze degradation of the TCRz chain in Cbl–/– and Sla–/–Cbl–/– mice, we incubated purified CD8+ thymocytes in the presence of the protein synthesis inhibitor cycloheximide and monitored TCRz expression over time. Like Sla–/– thymocytes, Cbl–/– and Sla–/–Cbl–/– thymocytes failed to substantially degrade TCRz over the 6-hour course of the assay (Fig. 2c,d). In contrast, the minor degradation of CD3e in wildtype thymocytes was not substantially affected regardless of genotype. Failure to degrade TCRz leads to an increased number of TCR complexes in the recycling pool of Sla–/–Cbl–/– thymocytes. Therefore, we tested whether the size of the CD3e recycling pool was increased in the absence of c-Cbl or both SLAP and c-Cbl. The size of the recycling pool can be determined by incubation of thymocytes in hypertonic medium, such as 0.45 M sucrose5. Hypertonic medium inhibits clathrin-mediated endocytosis by depleting the cell of free clathrin monomers26,27. Because the TCR complex is internalized via clathrincoated pits, incubation of cells in hypertonic medium blocks the internalization of CD3e28,29. Over time, CD3e accumulation on the cell surface (in the absence of new receptor synthesis) is indicative of the number of CD3e molecules present in the recycling pool. As reported before5, CD3e upregulation by thymocytes incubated in hypertonic medium was increased for Sla–/– DP but not SP thymocytes compared with wild-type control thymocytes (Fig. 2e). CD3e upregulation in hypertonic medium was also increased for both Cbl–/– as well as Sla–/–Cbl–/– DP thymocytes, demonstrating that the number of fully assembled TCR complexes present in the recycling pool of Cbl–/– and Sla–/–Cbl–/– DP thymocytes was increased to an extent similar to that for Sla–/– DP thymocytes. The increase in CD3e expression over the 6-hour assay was dependent on hypertonicity, as there was little CD3e upregulation on DP thymocytes incubated in the absence of 0.45 M sucrose, indicating that the increase in CD3e expression was due mainly to recycled CD3e (Fig. 2e). Furthermore, the upregulation of CD3e on DP thymocytes

59

ARTICLES GFP–

a Vector

c-Cbl

GFP+

b

c

Cbl-b

GFP

– –

SLAP-GFP

+ – – +

– –

+ –

– +

Xpress 75

SLAPGFP

50 GFP 25 GFP 0 Vector

60

α-tubulin

f Extracellular

Figure 3 Coexpression of SLAP and c-Cbl downregulates CD3e expression on Jurkat T cells. (a) Flow cytometry of CD3e expression on GFP+ versus GFP– Jurkat T cells transiently transfected with Xpress-Cbl or SLAP-GFP expression constructs. Data are representative of six independent experiments. (b) Quantification of data in a, presented as percent CD3e expression (± s.e.m.) on GFP+ cells versus GFP– cells, normalized to the ‘GFP plus vector’ control. (c) Immunoblot analysis of c-Cbl, Cbl-b (Xpress) and SLAP-GFP expression in Jurkat T cells transfected with various constructs (above lanes). a-tubulin, loading control. (d) CD8z chimera, with the extracellular and transmembrane (TM) regions of CD8 fused to the cytoplasmic domain of TCRz. (e) Flow cytometry of CD8 expression on GFP+ versus GFP– JbCD8.z14 T cells transiently transfected with either SLAP-GFP plus vector or SLAP-GFP plus c-Cbl. Data are representative of three independent experiments. (f) Quantification of data in e, presented as percent CD8 expression (± s.e.m.) on GFP+ cells versus GFP– cells, normalized to the ‘GFP + vector’ control. (g) Immunoblot analysis of c-Cbl and SLAP-GFP expression in JbCD8.z14 T cells transfected with various constructs (above lanes). a-tubulin, loading control.

SLAP and c-Cbl coexpression downregulates CD3e expression Ligand-induced internalization of the CD3e is not altered in Sla–/– thymocytes5. Similarly, thymocytes deficient in c-Cbl30 or both SLAP and c-Cbl showed normal rates of CD3e internalization induced by antibody to CD3e (anti-CD3e; Supplementary Fig. 1 online), suggesting that SLAP and c-Cbl regulate TCR expression independently of TCR internalization. To elucidate the mechanism of TCR downregulation by SLAP and c-Cbl, we developed a model system to induce CD3e downregulation by SLAP and c-Cbl. In this model system, we used constructs expressing SLAP fused to green fluorescent protein (GFP) at the C terminus or constructs expressing GFP alone as a control. We transfected Jurkat T cells with these constructs with or without an expression construct encoding c-Cbl. After transfection, we assessed CD3e expression on transfected (GFP+) and untransfected (GFP–) cells by flow cytometry. CD3e expression on Jurkat T cells was not substantially altered by the expression of SLAP-GFP or c-Cbl alone. In contrast, expression of SLAP-GFP together with c-Cbl substantially downregulated surface CD3e expression on GFP+ Jurkat T cells in the absence of TCR ligation (Fig. 3a,b). CD3e downregulation was specific for c-Cbl, as expression of Cbl-b together with SLAP-GFP had no effect on CD3e downregulation, despite similar expression of Cbl-b (Fig. 3c). The phosphorylated TCRz cytoplasmic domain is needed to prevent the accumulation of TCR complexes in the recycling pool in DP thymocytes5. Therefore, we tested whether the TCRz cytoplasmic domain in isolation could be downregulated by SLAP-GFP and

Cbl-b

d

100 101 102 103 104 100 101 102 103 104 100 101 102 103 104 CD3ε

was due to CD3e recycling rather than new synthesis, as the protein synthesis inhibitor cycloheximide had no effect on CD3e expression (data not shown).

c-Cbl

TM

CD8 TCRζ CD8ζ

e

–

GFP

SLAP + vector SLAP + c-Cbl

Cytoplasmic

+

GFP

75 50 25 0

SLAP + c-Cbl

SLAP + vector

100

g

Ve ct c- or C bl

SLAP-GFP

CD8 (% untransfected)


100 101 102 103 104 100 101 102 103 104 100 101 102 103 104

CD3ε (% untransfected)

100 GFP

GFP c-Cbl Cbl-b

SLAP-GFP

c-Cbl SLAP-GFP 100

101

102

103

104

100

101

102

103

104

α-tubulin

CD8

c-Cbl. JbCD8.z14 is a TCRb-deficient Jurkat T cell line stably transfected with a construct encoding a chimeric CD8z molecule31 (Fig. 3d). JbCD8.z14 T cells do not express endogenous TCR-CD3 at the cell surface. Surface expression of the CD8z chimera occurs independently of the remaining TCR-CD3 chains and can recapitulate TCR signaling after crosslinking of CD8 (ref. 31). Expression of SLAP-GFP together with c-Cbl in JbCD8.z14 T cells downregulated surface expression of the CD8z chimera (Fig. 3e–g), demonstrating that the TCRz cytoplasmic domain was a sufficient target for downregulation by SLAP and c-Cbl. Lck is needed to downregulate CD3e The development of a Jurkat T cell system provided us with a powerful tool for studying both the cellular and the molecular requirements for CD3e downregulation. We first tested which signaling molecules are required for CD3e downregulation mediated by SLAP and c-Cbl using Jurkat T cell lines deficient in Lck, Zap70 or the adaptor SLP-76 (Lcp2–/–). Expression of SLAP-GFP and c-Cbl in Jurkat T cells lacking either Zap70 or SLP-76 downregulated CD3e specifically on GFP+ cells (Fig. 4). In contrast, the Lck-deficient Jurkat T cell J.CaM1 failed to substantially downregulate CD3e after expression of SLAP-GFP together with c-Cbl (Fig. 4a,b). However, stable reintroduction of Lck into JCaM.1 T cells restored the ability of SLAP and c-Cbl to downregulate CD3e expression. The failure of SLAP and c-Cbl to downregulate expression of CD3e in J.CaM1 T cells was not due to lower expression of SLAP and/or c-Cbl, as expression of both molecules was similar to their expression in J.CaM1 stably reconstituted with Lck (Fig. 4c). These data indicated that Lck but not Zap70 or SLP-76 is required for downregulation of CD3e by SLAP and c-Cbl.

VOLUME 7

NUMBER 1

JANUARY 2006

NATURE IMMUNOLOGY

ARTICLES

b

100 80 60 40 20 0 100 101 102 103 104

100 100 80 80 60 60 40 40 20 20 0 0 100 101 102 103 104 100 101 102 103 104

100 80 60 40 20 0 100 101 102 103 104

P116 + Zap70

100 80 60 40 20 0 100 101 102 103 104

100 100 80 80 60 60 40 40 20 20 0 0 100 101 102 103 104 100 101 102 103 104

100 80 60 40 20 0 100 101 102 103 104

J14 –/– (Lcp2 )

100 80 60 40 20 0 100 101 102 103 104

100 100 80 80 60 60 40 40 20 20 0 0 100 101 102 103 104 100 101 102 103 104

100 80 60 40 20 0 100 101 102 103 104

J14 + SLP-6

α-tubulin

0

Lck

d

76

100 80 60 40 20 0 100 101 102 103 104

GFP

25 4

100 100 80 80 60 60 40 40 20 20 0 0 100 101 102 103 104 100 101 102 103 104

50

P-

100 80 60 40 20 0 100 101 102 103 104

J.CaM1 + Lck

75

J1

100 80 60 40 20 0 100 101 102 103 104

100

SL

100 100 80 80 60 60 40 40 20 20 0 0 100 101 102 103 104 100 101 102 103 104

J.CaM1 + J.CaM1 Lck GFP SLAP GFP SLAP c-Cbl – + – + – + – + c-Cbl

+

100 80 60 40 20 0 100 101 102 103 104

J.CaM1 (Lck –/–)

70

100 80 60 40 20 0 100 101 102 103 104

c

125

Za p

100 100 100 80 80 80 60 60 60 40 40 40 20 20 20 0 0 0 100 101 102 103 104 100 101 102 103 104 100 101 102 103 104

SLAP-GFP c-Cbl

16

c-Cbl

SLAP-GFP + vector

+

pEF

GFP + vector GFP + c-Cbl

P1

c-Cbl

J. C aM 1


pEF


SLAP-GFP

GFP

Lc k

GFP+

+

–

GFP

a

P116 + Zap70 GFP SLAP GFP SLAP c-Cbl – + – + – + – + c-Cbl

e

P116

P116 (Zap70 –/–)

J14

J14 + SLP-76

GFP SLAP GFP SLAP c-Cbl – + – + – + – + c-Cbl

GFP

GFP

α-tubulin

α-tubulin

Zap70

SLP-76

Figure 4 CD3e downregulation requires Lck. (a) Flow cytometry of CD3e expression on Jurkat T cell lines (right margin) transiently transfected with GFP or SLAP-GFP with or without c-Cbl expression constructs. Data are representative of at least four independent experiments. (b) Quantification of data in a, presented as percent CD3e expression (± s.e.m.) on GFP+ versus GFP– cells, normalized to the ‘GFP + vector’ control for each cell line. (c–e) Immunoblot analysis of c-Cbl and SLAP-GFP expression and expression of Lck (c), Zap70 (d) and SLP-76 (e) in various cell lines (above blots). a-tubulin, loading control.

CD3ε

CD3e downregulation requires multiple domains of SLAP To test which domains of SLAP are required to downregulate CD3e expression in Jurkat T cells, we transfected a panel of GFP-tagged SLAP constructs (Fig. 5a) together with wild-type c-Cbl and assessed CD3e expression on GFP+ versus GFP– cells by flow cytometry. Mutation of the SH3 domain of SLAP had little effect on CD3e downregulation on GFP+ cells after expression of SLAP and c-Cbl compared with Jurkat T cells expressing wild-type SLAP-GFP and c-Cbl together. In contrast, mutation of the myristolation site or the SH2 domain of SLAP prevented the downregulation of CD3e on GFP+ cells after expression of SLAP-GFP together with c-Cbl (Fig. 5b). In addition, a C-terminal truncation of SLAP failed to downregulate CD3e on GFP+ cells when expressed together with c-Cbl (Fig. 5b). Failure to downregulate CD3e by the various mutants was not due to lower expression of the SLAP mutants, as protein expression was similar to or higher than expression of wild-type SLAP (Fig. 5c). CD3e downregulation requires the ring finger of c-Cbl To determine which domains of c-Cbl are needed to downmodulate CD3e expression in Jurkat T cells, we transfected a panel of hemagglutinin-tagged c-Cbl expression constructs (Fig. 5d) together with an expression construct encoding wild-type SLAP-GFP and assessed CD3e expression on GFP+ versus GFP– cells by flow cytometry. Mutation of the ring finger of c-Cbl, either by partial deletion or by multiple point mutations in the ring finger domain, eliminated the ability of c-Cbl to downregulate CD3e when expressed together

NATURE IMMUNOLOGY

VOLUME 7

NUMBER 1

JANUARY 2006

with SLAP-GFP (Fig. 5e). In contrast, mutation of the c-Cbl tyrosine kinase–binding domain had no effect on the ability of c-Cbl to downregulate CD3e when expressed together with SLAP. Notably, the conserved N terminus of c-Cbl encompassing the tyrosine kinase–binding domain and ring finger domains was sufficient to downregulate CD3e, albeit less completely, even when the tyrosine kinase–binding domain of c-Cbl was mutated. These data suggest that only the ring finger domain of c-Cbl was absolutely required for downregulation of CD3e in the presence of SLAP. The observed CD3e downregulation was not a consequence of higher expression of the truncated c-Cbl proteins, as similar expression of a similar truncation mutant of c-Cbl containing multiple point alterations in the ring finger domain or a mutant of c-Cbl truncated N-terminal to the ring finger had no effect on CD3e expression (Fig. 5e,f). SLAP and c-Cbl interact in Jurkat T cells Interaction between SLAP and c-Cbl could not be detected previously because of the insolubility of SLAP in nonionic detergents8. However, a substantial proportion of SLAP is soluble if lysates are prepared in radioimmunoprecipitation assay (RIPA) lysis buffer (RIPA lysates; data not shown). In transiently transfected Jurkat T cells, SLAP-GFP and c-Cbl immunoprecipitated together from RIPA lysates (Fig. 5g,h). The interaction was specific, as shown by the trace interaction between c-Cbl and the GFP control. The C-terminal truncation of SLAP also coimmunoprecipitated with c-Cbl, although much less than wild-type SLAP (Fig. 5g,h). Similarly, the C terminus of SLAP alone interacted

61

Myr

Myr

GFP

GFP

SH3

SH2

COOH

GFP

SLAP

SH3

SH2

COOH

GFP

G2A

*

SH2

COOH

GFP

SH3*

c

125

G F SL P G AP 2 SHA SH3* ∆C 2*

b

a


ARTICLES

12CA5

100 75

GFP 50 25

α-tubulin

d TKB HA

4H EF

SH2

RF

PRR

LZ

c-Cbl

HA

4H EF

*

RF

PRR

LZ

G306E

HA

4H EF

SH2

*

PRR

LZ

C3AHN

HA

4H EF

SH2

RF

PRR

LZ

70/Z

HA

4H EF

SH2

CbIN

HA

4H EF

SH2

RF 1–436.WT

100 75 50

12CA5

25 SLAP-GFP

0

g

α-tubulin

h c-Cbl

c-Cbl

4H EF

*

HA

4H EF

SH2

IgH

IgH GFP

GFP HA

FP SL A ∆C P ∆N

∆N

f

125

G

GFP

e

FP SL A ∆C P ∆N

∆C

G

COOH

GFP

Ve c c- tor C G bl 30 C 6E 3 70 AH N C /Z b 1– IN 4 1– 36 4 .W 1– 36 T 43 .G 6. 30 C 6E 3A H N

SH2

SH2*

∆C

SH3

GFP

G F SL P A ∆C P ∆N

Myr

COOH

G FP SL AP G 2A SH 3* SH 2*

*

G F SL P A ∆C P ∆N

SH3


Myr

Ve ct c- or G Cb 3 l C 06 3A E H N 70 / 1 C 1– –4 b Z I 1– 436 36. N 43 .G WT 6. 30 C 6E 3A H N


0

RF 1–436.G306E

*

1–436.C3AHN

WCL

GFP IP

WCL

c-Cbl IP

Figure 5 TCR-CD3 downregulation requires multiple domains in SLAP and the ring finger of c-Cbl. (a) SLAP-GFP construct. Myr, myristolation; *, point mutations in the SH3 or SH2 domain (SH3* and SH2*); G2A, myristolation site; DC, C-terminal truncation; DN, C terminus alone. (b) Quantification of CD3e expression on Jurkat T cells transiently transfected with wild-type c-Cbl and SLAP-GFP constructs. Data are presented as percent CD3e expression (± s.e.m.) on GFP+ versus GFP– cells for three independent experiments and are normalized to the ‘GFP + vector’ control. (c) Immunoblot analysis of hemagglutinin-tagged c-Cbl (12CA5) or SLAP-GFP expression by Jurkat T cells transfected with various constructs (above lanes). a-tubulin, a loading control. (d) c-Cbl construct. HA, hemagglutinin; 4H, four-helix bundle; EF, EF hand; *, point mutations in tyrosine kinase–binding (TKB) or ring finger (RF) domains; PRR, proline-rich repeat; LZ, leucine zipper; G306E, tyrosine kinase–binding domain mutant; C3AHN, multiple point mutations in ring finger domain; 70/Z, partial deletion of ring finger; CblN, truncation N-terminal to ring finger; 1–436.WT, conserved N terminus encompassing tyrosine kinase–binding and ring finger domains; 1–436.G306E, mutation of tyrosine kinase–binding domain; 1-436.C3AHN, truncation mutant containing multiple point mutations in ring finger domain. (e) Quantification of CD3e expression on Jurkat T cells transiently transfected with wild-type SLAP-GFP and c-Cbl constructs. Data are presented as in b. (f) Immunoblot analysis of hemagglutinin-tagged c-Cbl and SLAP-GFP expression by Jurkat T cells transfected with various constructs (above lanes). a-tubulin, loading control. (g) Coimmunoprecipitation of wild-type c-Cbl with GFP or SLAP-GFP constructs (above lanes). (h) Coimmunoprecipitation of GFP or SLAP-GFP constructs (above lanes) with wild-type c-Cbl. WCL, protein expression in whole-cell lysates; IP, immunoprecipitation; IgH, immunoglobulin heavy chain. Data in g,h are representative of five independent experiments.

only weakly with c-Cbl (Fig. 5g,h). These data indicated that residues present in both the N and C termini of SLAP are required for its interaction with c-Cbl. TCRf is ubiquitinated and degraded by SLAP and c-Cbl As both SLAP5 and c-Cbl (Fig. 2c,d) are required for efficient TCRz degradation by DP thymocytes, we next tested whether TCRz is degraded after transfection of SLAP-GFP together with c-Cbl. We transfected Jurkat T cells with c-Cbl plus either SLAP-GFP or GFP (as a control) and assessed TCRz expression by intracellular flow cytometry staining with a phycoerythrin-labeled monoclonal antibody to TCRz. Expression of either SLAP or c-Cbl alone failed to downregulate expression of TCRz (Fig. 6a,b). In contrast, TCRz expression was substantially downregulated in cells expressing both SLAP-GFP and c-Cbl (Fig. 6a,b). As with CD3e downregulation, degradation of TCRz required the myristolation site, the SH2 domain and the C terminus of SLAP as well as the ring finger of c-Cbl (Fig. 6b,c), indicating that downregulation of surface CD3e expression may be coupled to the degradation of TCRz.

62

We next tested whether TCRz is ubiquitinated after transfection of SLAP together with c-Cbl. We immunoprecipitated TCRz from lysates of Jurkat T cells transfected with either SLAP or c-Cbl or with both SLAP and c-Cbl. We then analyzed the immunoprecipitates by immunoblot with either an antibody specific for the TCRz chain or an antibody specific for ubiquitin. Transfection of both SLAP and c-Cbl, but not SLAP or c-Cbl alone, led to an increase in highermolecular-weight species of the TCRz chain (Fig. 6d, right, lane 4). These higher-molecular-weight species migrated together with ubiquitin-reactive bands that were immunoprecipitated with TCRz (Fig. 6d, middle, lane 4), demonstrating that the ‘laddering’ of TCRz detected after expression of SLAP together with c-Cbl was due to ubiquitin modification of the TCRz chain. DISCUSSION Here we have provided both genetic and functional evidence that SLAP and c-Cbl cooperatively regulate TCR expression on DP thymocytes. Thymocytes deficient in SLAP and/or c-Cbl had increased expression of TCRb, CD3e, CD4, CD5 and CD69 at the DP stage of

VOLUME 7

NUMBER 1

JANUARY 2006

NATURE IMMUNOLOGY

ARTICLES

a

GFP–

Figure 6 SLAP and c-Cbl target TCRz for ubiquitination and degradation. (a) Intracellular flow cytometry staining for TCRz expression in GFP+ versus GFP– Jurkat T cells coexpressing c-Cbl and GFP (left) or c-Cbl and SLAPGFP (right). Data are representative of six independent experiments. (b,c) Quantification of TCRz expression in Jurkat T cells transiently transfected with wild-type c-Cbl and SLAP-GFP constructs (b) or with SLAP-GFP and c-Cbl constructs (c). Data are presented as percent CD3z expression (± s.e.m.) on GFP+ versus GFP– cells and are normalized to the vector control. (d) Detection of TCRz ubiquitination by immunoblot analysis. Left, expression of hemagglutinin-tagged c-Cbl (HA.c-Cbl), GFP and SLAPGFP in whole-cell lysates (WCL) of Jurkat T cells transfected with expression constructs (above lanes). a-tubulin, loading control. Middle and right, TCRz immunoprecipitates from Jurkat T cells transfected with various constructs (above lanes). Middle, ubiquitin-reactive species that migrate together with higher-molecular-weight TCRz species at right. A shorter exposure of the p16 form of TCRz is included as a control for immunoprecipitation. Left margin, molecular weight markers (in kDa). Immunoblots are representative of eight independent experiments.

GFP+

GFP + c-Cbl

SLAP-GFP + c-Cbl

100 75 50 25

125 100 75 50 25

AP FP

AP

SL

G

SL

–

–

+

+

12CA5

G FP SL AP G FP SL AP

FP G

HA.c-Cbl

–

–

+

G FP SL AP G FP SL AP

TCRζ IP

WCL

d

c-

to r C G bl 30 C 6E 3 1– AH 4 1– 3 N 43 6. 1– 6.G WT 43 3 6. 06 C E 3A H N

0 Ve c

∆C

3* SH 2*

SH

AP G 2A

G FP

0

TCRζ (% untransfected)

c

125

SL

b

TCRζ (% untransfected)


100 101 102 103 104 100 101 102 103 104 TCRζ

+

–

–

+

+

93.0

53.5

IgH

GFP

36.1 α-tubulin

29.5

IgL p21-23

21.3

p16 Ubiquitin

TCRζ

thymocyte development. In addition, the failure to express either SLAP and/or c-Cbl partially restored the development of Zap70–/– SP thymocytes and peripheral T cells, presumably by increasing TCR avidity due to increased TCR cell surface expression in a compromised signaling system. Furthermore, both SLAP and c-Cbl were needed to target the TCRz chain for degradation in thymocytes. Loss of either molecule leads to an increase in the number of fully assembled TCR complexes present in the recycling pool5. Notably, except for a small increase in CD4+ SP cells, the phenotype of doubly deficient mice is neither additive nor synergistic, suggesting that SLAP and c-Cbl function in the same pathway to regulate surface expression of the TCR. In addition to our studies in DP thymocytes, we developed a model system to study CD3e downregulation in Jurkat T cells after expression of SLAP together with c-Cbl. In this system, expression of SLAPGFP together with c-Cbl, but not SLAP-GFP or c-Cbl alone, led to downregulation of surface CD3e as well as total TCRz expression, thus mimicking the phenotype of DP thymocytes. Moreover, the ring finger activity of c-Cbl was required not only for downregulation of surface CD3e expression but also for TCRz degradation. Therefore, similar to our observations in DP thymocytes, CD3e downregulation in Jurkat T cells after expression of SLAP and c-Cbl together seems to result from accelerated degradation of the TCRz chain. The SH2 domain of SLAP was required for surface CD3e downregulation and TCRz degradation. The SH2 domain of SLAP is required for the interaction of SLAP with phosphorylated TCRz8,9. Furthermore, Lck is required for interaction of SLAP with TCRz5. Phosphorylation of TCRz immunoreceptor tyrosine-based activation

NATURE IMMUNOLOGY

VOLUME 7

NUMBER 1

JANUARY 2006

motifs is presumably mediated by tonic signaling via Lck32,33. Therefore, we hypothesize that the defect in CD3e downregulation and TCRz degradation in Lck-deficient Jurkat T cells results from the inability of SLAP to interact with unphosphorylated TCRz chains. However, Jurkat T cells deficient in Lck have a defect in both constitutive and ligand-induced TCR internalization due to a defect in clathrin heavy-chain phosphorylation34. Therefore, we cannot exclude the possibility that CD3e downregulation after transfection of Jurkat T cells with SLAP plus c-Cbl may be due to a defect in TCR internalization. Based on our studies, we propose the following model: TCRz chains present in fully assembled TCR complexes are phosphorylated by Lck, and the TCR is internalized via clathrin-coated pits. TCR complexes are transported to an endosomal compartment where phosphorylated TCRz binds to the SH2 domain of SLAP. SLAP binds and recruits c-Cbl to the TCR complex, after which c-Cbl ubiquitinates the TCRz chain, thereby targeting TCRz for degradation. In the absence of TCRz, the remainder of the TCR-CD3 complex is either degraded or retained in an intracellular compartment. Notably, failure to ubiquitinate and/or degrade TCRz in the presence of SLAP alone allowed CD3e to recycle back to the cell surface, suggesting that the TCR was not actively bound and retained by SLAP in the absence of c-Cbl. Conversely, SLAP seemed to be required for the recruitment of c-Cbl to the TCR. Thus, according to our model, SLAP functions as an adaptor to target the E3 ubiquitin ligase activity of c-Cbl to phosphorylated TCRz chains in DP thymocytes. Notably, we have been able to coimmunoprecipitate both SLAP-GFP and c-Cbl with the CD8z chimera from Jurkat T cells (M.M. and A.W., unpublished observations), suggesting that SLAP, c-Cbl and TCRz may form a trimolecular (or larger) complex. The ability of SLAP to regulate CD3e expression on Jurkat T cells is dependent on c-Cbl. In contrast, expression of SLAP together with Cbl-b failed to downregulate CD3e expression. That finding was somewhat unexpected, as c-Cbl and Cbl-b are highly homologous (84% identity), especially in the N-terminal tyrosine kinase–binding and ring finger domains that together are sufficient to downregulate CD3e expression. However, unlike c-Cbl, Cbl-b does not seem to regulate basal TCR expression, as neither thymocytes nor peripheral T cells have increased receptor expression in the absence of Cbl-b35,36. Instead, Cbl-b may be involved in ligand-induced TCR downregulation, as T cells deficient in both c-Cbl and Cbl-b have a reduced capacity to downregulate TCR expression in response to TCRb crosslinking30. Cbl-b also has E3 ubiquitin ligase activity for many other molecules, including the adaptor molecule Vav and phosphati-

63


ARTICLES dylinositol-3-kinase37,38. Therefore, it is possible that Cbl-b targets molecules not included in the TCR complex for ubiquitination and degradation. Further studies are needed to determine whether SLAP and Cbl-b function together to regulate expression of other signaling molecules. Moreover, the existence of multiple SLAP (SLAP and SLAP-2) and Cbl (c-Cbl, Cbl-b and Cbl-3) family members raises the possibility that various combinations of SLAP and Cbl family members function independently to regulate the expression of unique protein targets. In particular, SLAP-2 interacts with c-Cbl and promotes ligand-induced CD3e downregulation in Jurkat T cells39–41, suggesting that both SLAP family members may be required for optimal regulation of TCR expression. The C terminus of SLAP has been reported to interact with the tyrosine kinase–binding domain of c-Cbl9. Similarly, we have been able to coimmunoprecipitate the N terminus of c-Cbl with SLAP-GFP from Jurkat T cells (data not shown). In contrast, we found that the interaction between SLAP and c-Cbl was reduced but not absent when the SLAP C terminus was truncated. Moreover, an N-terminal deletion construct of SLAP interacted only weakly with c-Cbl. Our preliminary results suggest that the interaction between SLAP and c-Cbl is complex and may involve multiple domains of SLAP and/or c-Cbl. Both the SH2 and the SH3 domains of Src family kinases interact with c-Cbl12,42–44. Therefore, it is possible that the SH2 domain and/or SH3 domain of SLAP contribute to the interaction with c-Cbl. Alternatively, a third molecule may participate in the interaction of SLAP and c-Cbl, particularly in the absence of the SLAP N or C terminus. Regardless, it is apparent that the residual interaction between the C-terminal deletion mutant of SLAP and c-Cbl is not sufficient for CD3e downregulation and TCRz degradation. Therefore, it is possible that the C terminus of SLAP may accomplish additional functions that are independent of its interaction with c-Cbl. We have shown that SLAP and c-Cbl function in the same pathway to regulate TCR expression on DP thymocytes. Loss of SLAP and/or c-Cbl leads to similar alterations in thymocyte development10,11, emphasizing that strict regulation of TCR expression on developing thymocytes is essential during thymic ‘education’. The function of SLAP and c-Cbl in maintaining low TCR expression may be critical for the positive selection of an appropriate TCR repertoire. Consequently, the TCR repertoire in Sla–/– and/or Cbl–/– T cells may be inappropriately biased for less efficient self-recognition. The outcome of SLAP and/or c-Cbl deficiency on the peripheral TCR repertoire in nontransgenic mice has not yet been tested. Therefore, future studies in this area may eventually lead to a greater understanding of T cell development and T cell repertoire selection. METHODS Cell lines and cell culture. Cell lines were maintained in RPMI medium supplemented with 10% FBS, 2 mM glutamine, penicillin and streptomycin. Cell lines were either wild-type Jurkat T cells or were derived from a parental Jurkat T cell line: JCaM1 (Lck–/–)45, JCaM1 + Lck (J.CaM1 stably expressing Lck)46, P116 (Zap70–/–), P116.C39 (P116 stably expressing Zap70)47, J14 (SLP76 deficient), J14-76-11 (J14 stably expressing SLP-76)48 and JbCD8.z14 (TCRb-deficient stably expressing a chimeric CD8z molecule)31. Expression constructs. Plasmids encoding GFP-tagged SLAP were constructed by PCR amplification of mouse SLAP from pEF-BOS and ligation into pN1-GFP (Clontech) via EcoRI and BamHI sites. GFP constructs were subcloned into pCDEF3 with EcoRI and NotI sites. Xpress-tagged c-Cbl and Cbl-b were a gift from Y.-C. Liu (La Jolla Institute for Allergy and Immunology, San Diego, California). Hemagglutinin-tagged c-Cbl constructs

64

(excluding 436wild-type, 436GE and 436RF) have been described49–51. The 436wild-type, 436GE and 436RF constructs were generated by PCR amplification from full-length c-Cbl, G306E or C3AHN and ligation into pAlter-MAX. Hemagglutinin-tagged c-Cbl constructs were subcloned into pCDEF3 via EcoRI and NotI sites. Mice. Sla–/– mice have been described and have been backcrossed onto C57BL/6 at least five generations5,11. C57BL/6 mice (Taconic) were used as wild-type controls. Cbl–/– mice have been described10 and were crossed with Sla–/– mice to generate Sla–/–Cbl–/– mice. Zap70–/– mice have been described24 and were crossed with Sla–/–, Cbl–/–and Sla–/–Cbl–/– mice. All animal experiments were in accordance with protocols approved by the University of California San Francisco Institutional Animal Care and Use Committee (San Francisco, California). Flow cytometry staining. After being washed in PBS, thymocytes were stained for 30 min with antibodies to CD4 (RM4-5; eBioscience), CD8a (53-6.7; Pharmingen), TCRb (H57-197; eBioscience), CD3e (145-2C11; eBioscience), CD5 (53-7.3; Pharmingen) or CD69 (H1.2F3; Pharmingen) in flow cytometry buffer (PBS with 1% BSA and 0.01% azide). Thymocytes were washed and then analyzed by flow cytometry. All data presented are representative of at least six mice per genotype. Jurkat T cells transfected overnight with various constructs were washed in PBS and stained with antibody to human CD3e (UCHT1, BD Biosciences) or CD8 (3B5, Caltag Laboratories) as described above. For intracellular flow cytometry staining, transiently transfected Jurkat T cells were washed in PBS and fixed in 4% paraformaldehyde for 20 min. After being washed, Jurkat T cells were permeabilized for 20 min at 24 1C in flow cytometry buffer containing 0.5% saponin. Jurkat T cells were stained for 1 h at 4 1C with a phycoerythrin-conjugated antibody to TCRz (6B10.2; Santa Cruz Biotechnology) in 0.5% saponin containing 1% mouse serum. Cells were washed and then analyzed by flow cytometry. All data presented are representative of at least six independent experiments. Hypertonic recycling assay. Freshly isolated thymocyte single-cell suspensions were cultured at a density of 5 106 cells/ml in primary cell culture media (RPMI medium containing 10% FBS, 2 mM glutamine, 50 mM b-mercaptoethanol, penicillin and streptomycin) in the presence or absence of 0.45 M sucrose. At each time point, cells were mixed with 100 ml ice-cold PBS containing 1% BSA and 0.1% NaN3. Cells were maintained on ice for the remainder of the assay and then stained and analyzed by flow cytometry. Immunoblot. For analysis of total amounts of CD3e and TCRz, CD8+ thymocytes were purified by magnetic cell sorting (Miltinyi Biotec) according to the manufacturer’s protocol. Recovered cells were 95% or more DP, as assessed by flow cytometry. CD8+ thymocytes were lysed for 30 min on ice at a density of 200 106 cells/ml in RIPA lysis buffer (1% Nonidet P-40, 1% sodium deoxycholate, 0.1% SDS, 50 mM 3-(N-morpholino) propanesulfonic acid, pH 7.0, 150 mM NaCl and 2 mM EDTA) supplemented with protease inhibitors (leupeptin, aprotinin, phenylmethyl sulfonyl fluoride and pepstatin A). Postnuclear supernatants were prepared by centrifugation of samples for 30 min at 16,000g and 4 1C. Samples were separated by 12.5% SDS-PAGE and then transferred to Immobilon membranes (Millipore), followed by immunoblot for a-tubulin (B-5-1-2; Sigma), CD3e (M20; Santa Cruz Biotechnology) or TCRz (8D3; Pharmingen). Membranes were incubated with secondary antibodies coupled to horseradish peroxidase (Amersham Biosciences) followed by detection with enhanced chemiluminescence (Amersham Biosciences). Immunoblots were quantified on a Kodak Imaging Station with Kodak 1D image analysis software version 3.5 (Eastman Kodak). For TCRz degradation experiments, CD8+ thymocyte samples (enriched for DP thymocytes) were cultured at a density of 20 106 cells/ml in primary cell culture media containing 100 mg/ml of cycloheximide (Sigma). At each time point, thymocytes were lysed in RIPA lysis buffer and kept on ice for the remainder of the experiment. Lysates were analyzed by immunoblot for a-tubulin (B-5-1-2), CD3e (M20) or TCRz (8D3). Lysis, immunoblots and quantification were done as described above.

VOLUME 7

NUMBER 1

JANUARY 2006

NATURE IMMUNOLOGY

ARTICLES


For cell line experiments, cells were transfected overnight with various constructs, washed twice in PBS and then lysed at a density of 50 106 cells/ml in RIPA lysis buffer as described above. Postnuclear supernatants were separated by 10% SDS-PAGE and then transferred to Immobilon membranes, followed by immunoblot for a-tubulin (B-5-1-2), GFP (JL8; Clonetech) or c-Cbl (12CA5, Boehringer Mannheim; Xpress, Invitrogen; C15, Santa Cruz Biotechnology) as described above. Coimmunoprecipitation of SLAP and c-Cbl. Jurkat T cells (20 106) were transiently transfected with hemagglutinin-tagged c-Cbl and various SLAP-GFP constructs. Then, 4.5 h after transfection, cells were washed and lysed as described above. Lysates were immunoprecipitated for 30 min at 4 1C with anti-hemagglutinin (12CA5) or anti-GFP (JL8, Clontech) crosslinked to protein G (Amersham Biosciences). Immunoprecipitates were washed four times with ice-cold lysis buffer, resuspended in SDSPAGE loading buffer and then boiled. Samples were separated by electrophoresis, transferred to Immobilon membranes and analyzed by immunoblot as described above. For each transfection, postnuclear supernatants from 5 105 cell equivalents were analyzed by immunoblot as described above to control for expression. Data presented are representative of six independent experiments. TCRf ubiquitination. Jurkat T cells (80 106) were transiently transfected with various SLAP-GFP and hemagglutinin-tagged c-Cbl constructs. Then, 4.5 h after transfection, cells were washed and lysed in RIPA lysis buffer supplemented with 1.25 mg/ml of N-ethylmalemide (Sigma). Lysates were immunoprecipitated for 1 h at 4 1C for TCRz (6B10.2) with protein G. Samples were washed four times with ice-cold lysis buffer, resuspended in SDS loading buffer and then boiled for 5 minutes. Samples were separated by 10% SDSPAGE and transferred to Immobilon membranes, followed by immunoblot for hemagglutinin (12CA5), GFP (JL8), a-tubulin (B-5-1-2), ubiquitin (P4D1; Santa Cruz Biotechnology) or TCRz (8D3) as described above. Data are representative of at least ten independent experiments. Note: Supplementary information is available on the Nature Immunology website.

ACKNOWLEDGMENTS We thank F. Brodsky, M. von Zastrow and members of the Weiss lab for comments and suggestions. Supported by the National Institutes of Health (CA72531 to A.W. and CA987986 to H.B.). COMPETING INTERESTS STATEMENT The authors declare that they have no competing financial interests. Published online at http://www.nature.com/natureimmunology/ Reprints and permissions information is available online at http://npg.nature.com/ reprintsandpermissions/

1. Love, P.E. & Chan, A.C. Regulation of thymocyte development: only the meek survive. Curr. Opin. Immunol. 15, 199–203 (2003). 2. Finkel, T.H., McDuffie, M., Kappler, J.W., Marrack, P. & Cambier, J.C. Both immature and mature T cells mobilize Ca2+ in response to antigen receptor crosslinking. Nature 330, 179–181 (1987). 3. Havran, W.L. et al. Expression and function of the CD3-antigen receptor on murine CD4+8+ thymocytes. Nature 330, 170–173 (1987). 4. Kearse, K.P., Roberts, J.P., Wiest, D.L. & Singer, A. Developmental regulation of ab T cell antigen receptor assembly in immature CD4+CD8+ thymocytes. Bioessays 17, 1049–1054 (1995). 5. Myers, M., Dragone, L.D. & Weiss, A. Src-like adaptor protein down-regulates T cell receptor (TCR)/CD3 expression by targeting TCRz for degradation. J. Cell Biol. 170, 285–294 (2005). 6. Pandey, A., Duan, H. & Dixit, V.M. Characterization of a novel Src-like adapter protein that associates with the Eck receptor tyrosine kinase. J. Biol. Chem. 270, 19201– 19204 (1995). 7. Roche, S. et al. Src-like adaptor protein (SLAP) is a negative regulator of mitogenesis. Curr. Biol. 8, 975–978 (1998). 8. Sosinowski, T., Pandey, A., Dixit, V.M. & Weiss, A. Src-like adaptor protein (SLAP) is a negative regulator of T cell receptor signaling. J. Exp. Med. 191, 463–474 (2000). 9. Tang, J., Sawasdikosol, S., Chang, J.H. & Burakoff, S.J. SLAP, a dimeric adapter protein, plays a functional role in T cell receptor signaling. Proc. Natl. Acad. Sci. USA 96, 9775–9780 (1999).

NATURE IMMUNOLOGY

VOLUME 7

NUMBER 1

JANUARY 2006

10. Naramura, M., Kole, H.K., Hu, R.J. & Gu, H. Altered thymic positive selection and intracellular signals in Cbl-deficient mice. Proc. Natl. Acad. Sci. USA 95, 15547– 15552 (1998). 11. Sosinowski, T., Killeen, N. & Weiss, A. The Src-like adaptor protein downregulates the T cell receptor on CD4+CD8+ thymocytes and regulates positive selection. Immunity 15, 457–466 (2001). 12. Donovan, J.A., Wange, R.L., Langdon, W.Y. & Samelson, L.E. The protein product of the c-cbl protooncogene is the 120-kDa tyrosine-phosphorylated protein in Jurkat cells activated via the T cell antigen receptor. J. Biol. Chem. 269, 22921–22924 (1994). 13. Rao, N., Dodge, I. & Band, H. The Cbl family of ubiquitin ligases: critical negative regulators of tyrosine kinase signaling in the immune system. J. Leukoc. Biol. 71, 753–763 (2002). 14. Joazeiro, C.A. et al. The tyrosine kinase negative regulator c-Cbl as a RING-type, E2dependent ubiquitin-protein ligase. Science 286, 309–312 (1999). 15. Weissman, A.M. Regulating protein degradation by ubiquitination. Immunol. Today 18, 189–198 (1997). 16. Cenciarelli, C. et al. Activation-induced ubiquitination of the T cell antigen receptor. Science 257, 795–797 (1992). 17. Cenciarelli, C., Wilhelm, K.G., Jr., Guo, A. & Weissman, A.M. T cell antigen receptor ubiquitination is a consequence of receptor-mediated tyrosine kinase activation. J. Biol. Chem. 271, 8709–8713 (1996). 18. Hou, D., Cenciarelli, C., Jensen, J.P., Nguygen, H.B. & Weissman, A.M. Activationdependent ubiquitination of a T cell antigen receptor subunit on multiple intracellular lysines. J. Biol. Chem. 269, 14244–14247 (1994). 19. Wang, H.Y. et al. Cbl promotes ubiquitination of the T cell receptor z through an adaptor function of Zap-70. J. Biol. Chem. 276, 26004–26011 (2001). 20. Azzam, H.S. et al. CD5 expression is developmentally regulated by T cell receptor (TCR) signals and TCR avidity. J. Exp. Med. 188, 2301–2311 (1998). 21. Merkenschlager, M. et al. How many thymocytes audition for selection? J. Exp. Med. 186, 1149–1158 (1997). 22. Kane, L.P., Lin, J. & Weiss, A. Signal transduction by the TCR for antigen. Curr. Opin. Immunol. 12, 242–249 (2000). 23. Negishi, I. et al. Essential role for ZAP-70 in both positive and negative selection of thymocytes. Nature 376, 435–438 (1995). 24. Kadlecek, T.A. et al. Differential requirements for ZAP-70 in TCR signaling and T cell development. J. Immunol. 161, 4688–4694 (1998). 25. Thien, C.B., Bowtell, D.D. & Langdon, W.Y. Perturbed regulation of ZAP-70 and sustained tyrosine phosphorylation of LAT and SLP-76 in c-Cbl-deficient thymocytes. J. Immunol. 162, 7133–7139 (1999). 26. Heuser, J.E. & Anderson, R.G. Hypertonic media inhibit receptor-mediated endocytosis by blocking clathrin-coated pit formation. J. Cell Biol. 108, 389–400 (1989). 27. Daukas, G. & Zigmond, S.H. Inhibition of receptor-mediated but not fluidphase endocytosis in polymorphonuclear leukocytes. J. Cell Biol. 101, 1673–1679 (1985). 28. Telerman, A. et al. Internalization of human T lymphocyte receptors. Eur. J. Immunol. 17, 991–997 (1987). 29. Dallanegra, A., Schaffar, L., Breittmayer, J.P., Carpentier, J.L. & Fehlmann, M. Effect of hypertonicity and monensin on CD3/TCR surface expression in human T cells. Immunol. Lett. 19, 115–120 (1988). 30. Naramura, M. et al. c-Cbl and Cbl-b regulate T cell responsiveness by promoting ligandinduced TCR down-modulation. Nat. Immunol. 3, 1192–1199 (2002). 31. Irving, B.A. & Weiss, A. The cytoplasmic domain of the T cell receptor z chain is sufficient to couple to receptor-associated signal transduction pathways. Cell 64, 891–901 (1991). 32. Qian, D., Mollenauer, M.N. & Weiss, A. Dominant-negative z-associated protein 70 inhibits T cell antigen receptor signaling. J. Exp. Med. 183, 611–620 (1996). 33. van Oers, N.S., Killeen, N. & Weiss, A. ZAP-70 is constitutively associated with tyrosine-phosphorylated TCRz in murine thymocytes and lymph node T cells. Immunity 1, 675–685 (1994). 34. Crotzer, V.L., Mabardy, A.S., Weiss, A. & Brodsky, F.M. T cell receptor engagement leads to phosphorylation of clathrin heavy chain during receptor internalization. J. Exp. Med. 199, 981–991 (2004). 35. Bachmaier, K. et al. Negative regulation of lymphocyte activation and autoimmunity by the molecular adaptor Cbl-b. Nature 403, 211–216 (2000). 36. Chiang, Y.J. et al. Cbl-b regulates the CD28 dependence of T-cell activation. Nature 403, 216–220 (2000). 37. Miura-Shimura, Y. et al. Cbl-mediated ubiquitinylation and negative regulation of Vav. J. Biol. Chem. 278, 38495–38504 (2003). 38. Fang, D. & Liu, Y.C. Proteolysis-independent regulation of PI3K by Cbl-b-mediated ubiquitination in T cells. Nat. Immunol. 2, 870–875 (2001). 39. Loreto, M.P., Berry, D.M. & McGlade, C.J. Functional cooperation between c-Cbl and Src-like adaptor protein 2 in the negative regulation of T-cell receptor signaling. Mol. Cell. Biol. 22, 4241–4255 (2002). 40. Holland, S.J. et al. Functional cloning of Src-like adapter protein-2 (SLAP-2), a novel inhibitor of antigen receptor signaling. J. Exp. Med. 194, 1263–1276 (2001). 41. Pandey, A. et al. A novel Src homology 2 domain-containing molecule, Src-like adapter protein-2 (SLAP-2), which negatively regulates T cell receptor signaling. J. Biol. Chem. 277, 19131–19138 (2002). 42. Andoniou, C.E. et al. The Cbl proto-oncogene product negatively regulates the Srcfamily tyrosine kinase Fyn by enhancing its degradation. Mol. Cell. Biol. 20, 851–867 (2000).

65


ARTICLES 43. Sanjay, A. et al. Cbl associates with Pyk2 and Src to regulate Src kinase activity, avb3 integrin-mediated signaling, cell adhesion, and osteoclast motility. J. Cell Biol. 152, 181–195 (2001). 44. Tsygankov, A.Y., Mahajan, S., Fincke, J.E. & Bolen, J.B. Specific association of tyrosine-phosphorylated c-Cbl with Fyn tyrosine kinase in T cells. J. Biol. Chem. 271, 27130–27137 (1996). 45. Goldsmith, M.A. & Weiss, A. Isolation and characterization of a T-lymphocyte somatic mutant with altered signal transduction by the antigen receptor. Proc. Natl. Acad. Sci. USA 84, 6879–6883 (1987). 46. Straus, D.B. & Weiss, A. Genetic evidence for the involvement of the lck tyrosine kinase in signal transduction through the T cell antigen receptor. Cell 70, 585–593 (1992).

66

47. Williams, B.L. et al. Genetic evidence for differential coupling of Syk family kinases to the T-cell receptor: reconstitution studies in a ZAP-70-deficient Jurkat T-cell line. Mol. Cell. Biol. 18, 1388–1399 (1998). 48. Yablonski, D., Kuhne, M.R., Kadlecek, T. & Weiss, A. Uncoupling of nonreceptor tyrosine kinases from PLC-g1 in an SLP-76-deficient T cell. Science 281, 413–416 (1998). 49. Rao, N. et al. The linker phosphorylation site Tyr292 mediates the negative regulatory effect of Cbl on ZAP-70 in T cells. J. Immunol. 164, 4616–4626 (2000). 50. Rao, N. et al. The non-receptor tyrosine kinase Syk is a target of Cbl-mediated ubiquitylation upon B-cell receptor stimulation. EMBO J. 20, 7085–7095 (2001). 51. Ota, S. et al. The RING finger domain of Cbl is essential for negative regulation of the Syk tyrosine kinase. J. Biol. Chem. 275, 414–422 (2000).

VOLUME 7

NUMBER 1

JANUARY 2006

NATURE IMMUNOLOGY