T Cell Receptor ζ Allows Stable Expression of Receptors Containing the CD3γ Leucine-based Receptor-sorting Motif*

The leucine-based motif in the T cell receptor (TCR) subunit CD3γ constitutes a strong internalization signal. In fully assembled TCR this motif is inactive unless phosphorylated. In contrast, the motif is constitutively active in CD4/CD3γ and Tac/CD3γ chimeras independently of phosphorylation and leads to rapid internalization and sorting of these chimeras to lysosomal degradation. Because the TCRζ chain rescues incomplete TCR complexes from lysosomal degradation and allows stable surface expression of fully assembled TCR, we addressed the question whether TCRζ has the potential to mask the CD3γ leucine-based motif. By studying CD4/CD3γ and CD16/CD3γ chimeras, we found that CD16/CD3γ chimeras associated with TCRζ. The CD16/CD3γ-TCRζ complexes were stably expressed at the cell surface and had a low spontaneous internalization rate, indicating that the leucine-based motif in these complexes was inactive. In contrast, the CD4/CD3γ chimeras did not associate with TCRζ, and the leucine-based motif in these chimeras was constitutively active resulting in a high spontaneous internalization rate and low expression of the chimeras at the cell surface. Thus, our data demonstrate that TCRζ allows stable cell surface expression of receptors containing CD3γ leucine-based motifs by its potential to mask such motifs.

The T cell receptor (TCR) 1 is a multimeric receptor composed of the ligand binding Ti␣␤ dimer and the signal transducing subunits CD3␥⑀ and CD3␦⑀ and the TCR dimer (1,2). Only completely assembled octameric Ti␣␤CD3␥⑀␦⑀ complexes are efficiently expressed at the cell surface of mature T cells to ensure proper TCR functions. Thus, very selective mechanisms that only allow expression of completely assembled and functional TCR must exist in T cells. Several studies have indicated that the TCR chain plays important roles in such mechanisms. Thus, the TCR is not or is only very weakly expressed at the cell surface of T cells from TCR knock-out mice and in the TCRdeficient T cell variant MA 5.8 (3)(4)(5)(6). In the MA 5.8 variant, hexameric Ti␣␤CD3␥⑀␦⑀ complexes are assembled in the endoplasmic reticulum in the absence of TCR and subsequently transported via the Golgi apparatus to the lysosomes for degradation (6). Furthermore, association of TCR to the rest of the TCR seems to be critically dependent on the assembly of hexameric Ti␣␤CD3␥⑀␦⑀ complexes, and TCR does not associate with partial TCR complexes in T cell variants lacking Ti␣, ␤, or CD3␥ (7)(8)(9)(10). Thus, the selective expression of only completely assembled TCR at the T cell surface seems to be ensured by the TCR chain (11,12). However, it still remains to be explained how the TCR chain redirects the sorting of incomplete TCR from a degradative pathway to the cell surface.
The observation that incompletely assembled TCR complexes are sorted to a degradative compartment and not expressed at the cell surface suggests that receptor-sorting motifs with the capacity to sort receptors to the lysosomes must be active in incomplete TCR (6). We and others have recently described a leucine-based (L-based) receptor-sorting motif (S 126 DKQTLL 132 ) in the cytoplasmic tail of CD3␥ (13)(14)(15). When active, the L-based motif is recognized and bound by clathrin-coated vesicle adaptor proteins either at the trans-Golgi network or at the plasma membrane (15)(16)(17). This leads to sorting of receptors to the lysosomes and to rapid receptor internalization, respectively. In completely assembled TCR, the CD3␥ L-based motif is inactive and not accessible for adaptor proteins unless phosphorylated (14). In contrast, in chimeric Tac/CD3␥ and CD4/CD3␥ molecules, the motif is constitutively active independently of phosphorylation, and like hexameric Ti␣␤CD3␥⑀␦⑀ complexes, these chimeras are rapidly transported to the lysosomes for degradation (6,13,15). From these observations it may be suggested that the CD3␥ L-based motif is active in incompletely assembled TCR and that TCR allows cell surface expression of completely assembled Ti␣␤CD3␥⑀␦⑀ complexes by masking this motif. In this study, we addressed the question whether TCR has the potential to mask the CD3␥ leucine-based motif.
By analyzing receptor expression and sorting of mutated TCR and chimeric CD4/CD3␥ and CD16/CD3␥ molecules, we found that similar to the TCR, the CD16/CD3␥ chimera was stably expressed at the cell surface in association with TCR. Furthermore, both the TCR and the CD16/CD3␥-TCR complexes had low spontaneous internalization rates indicating that the CD3␥ L-based motif in these multimeric complexes was inactive. As is true for the TCR, the CD3␥ L-based motif in the CD16/CD3␥-TCR complexes was activated following protein kinase C (PKC) activation, which resulted in a rapid internalization of the complexes from the cell surface. In contrast, the CD4/CD3␥ chimera did not associate with the TCR chain and the CD3␥ L-based motif in these chimeras was constitutively active resulting in a high spontaneous internalization rate and low cell surface expression of the chimera. Thus, our data demonstrate that the TCR chain allows stable cell surface expression of receptors containing a CD3␥ L-based motif most probably by masking this motif.
Receptor Internalization and Recycling-To determine the spontaneous internalization rates, cells were incubated in RPMI 1640 ϩ 10% FCS at a cell density of 2 ϫ 10 5 cells/ml at 37°C or 4°C with PE- For PKC-induced receptor internalization, cells were adjusted to 2 ϫ 10 5 cells/ml medium (RPMI 1640 ϩ 10% FCS) and incubated at 37°C with various concentrations of the phorbol ester PDB. At the time indicated, cells were transferred to ice-cold PBS containing 2% FCS and 0.1% NaN 3 and washed twice. The cells were stained directly with PE-conjugated anti-CD3⑀ or anti-CD16 mAb and analyzed by flow cytometry. MFI was recorded and used in the calculation of percent mAb binding: (MFI of phorbol ester treated cells) divided by (MFI of untreated cells) ϫ 100%. For each construct at least three different clones were analyzed.

RESULTS AND DISCUSSION
The CD16/CD3␥ Chimera Is Stably Expressed at the Cell Surface in Association with TCR-We and others have previously demonstrated that the CD3␥ L-based motif is constitutively active in chimeric CD4/CD3␥ and Tac/CD3␥ molecules independently of phosphorylation (13,15). The active L-based motif results in very low expression of these chimeras at the cell surface. To determine whether the TCR chain had the potential to mask the CD3␥ L-based motif and thereby allow stable receptor cell surface expression, we took advantage of the observation that the Fc receptor Fc␥RIIIA-␣ chain (CD16) is only expressed at the cell surface of Jurkat cells in association with TCR (23). In contrast to CD16, CD4 is expressed as a monomer at the cell surface (24). By comparing chimeric CD4/CD3␥ and CD16/CD3␥ molecules either with or without an intact CD3␥ L-based motif, this enabled us specifically to examine if and how the TCR chain influenced the activity of the CD3␥ L-based motif. Six different constructs were made. CD3␥-tS126 and CD3␥-tP133 coded for the CD3␥ chain with a truncated cytoplasmic tail immediately before and after the L-based motif, respectively. CD4/CD3␥-tS126 and CD4/CD3␥-tP133 coded for chimeric molecules composed of the extracellular and transmembrane domains of CD4 and the cytoplasmic tail of CD3␥ truncated immediately before and after the Lbased motif, respectively. CD16/CD3␥-tS126 and CD16/CD3␥-tP133 coded for chimeric molecules composed of the extracellular and transmembrane domains of CD16 and the cytoplasmic tail of CD3␥ truncated immediately before and after the L-based motif, respectively (Fig. 1A). These constructs were separately transfected into the CD3␥ negative Jurkat variant JGN (10) and G418-resistant transfectants were tested for cell surface expression of the transfected molecules by FACS analysis. As shown in Fig. 1B, the TCR and the CD16/ CD3␥ chimera were all highly expressed at the cell surface independent of the presence or absence of the CD3␥ L-based motif. In agreement with previous studies, the CD4/CD3␥-tS126 chimera was highly expressed, whereas the CD4/CD3␥-tP133 chimera with an active CD3␥ L-based motif was only weakly expressed at the cell surface although highly expressed intracellularly ( Fig. 1B and data not shown) (15). To analyze whether the chimeras actually associated with TCR, cells were lysed in digitonin lysis buffer and immunoprecipitated with either anti-CD4 or anti-CD16 mAb. The precipitates were resolved by SDS-polyacrylamide gel electrophoresis, and Western blot analysis was performed using the anti-TCR mAb. As shown in Fig. 1C, TCR clearly co-precipitated with the CD16/ CD3␥ chimeras but did not co-precipitate with the CD4/CD3␥ chimeras.
These experiments demonstrated that in the absence of the CD3␥ L-based motif both the TCR, the CD4/CD3␥, and the CD16/CD3␥ chimeras were highly expressed at the cell surface independent of their capacity to associate with TCR. In contrast, in the presence of an intact CD3␥ L-based motif only molecules capable of forming association with TCR were expressed at the cell surface. Because the CD4/CD3␥-tP133 and CD16/CD3␥-tP133 chimeras had identical cytoplasmic tails with an intact CD3␥ L-based motif and only differed intracellularly by their ability to associate with the TCR chain, these observations suggested that TCR masked and thereby inactivated the CD3␥ L-based motif of the CD16/CD3␥-tP133 chimera.
The CD3␥ L-based Motif Is Inactive in CD16/CD3␥-TCR Complexes but Can Be Activated Following PKC Activation-We have previously demonstrated that CD4/CD3␥ chimeras with an inactive CD3␥ L-based motif have low spontaneous internalization rates and are highly expressed at the cell surface, whereas CD4/CD3␥ chimeras with an active CD3␥ L-based motif have high spontaneous internalization rates and are weakly expressed at the cell surface (15). To analyze whether the CD3␥ L-based motif in the CD16/CD3␥-tP133-TCR complex was active at the cell surface, the spontaneous internalization rate was determined. As expected, the CD4/ CD3␥-tP133 chimera had a high spontaneous internalization rate reflecting the active CD3␥ L-based motif present in this chimera, and the TCR-tP133 had a low spontaneous internalization rate reflecting the inactive CD3␥ L-based motif in the completely assembled TCR. Like the TCR-tP133, the CD16/ CD3␥-tP133-TCR complex had a low spontaneous internalization rate, indicating that the CD3␥ L-based motif in the CD16/ CD3␥-tP133-TCR complex was inactive ( Fig. 2A). Likewise, the CD4/CD3␥-tS126 that did not contain the CD3␥ L-based motif had a low spontaneous internalization rate.
Following PKC-induced phosphorylation the CD3␥ L-based motif in the context of the TCR is activated. This results in an increased internalization rate and a down-regulation of the TCR from the cell surface (14,25). To analyze whether the CD3␥ L-based motif in the CD16/CD3␥-tP133-TCR complex could be activated following PKC activation, cells were treated with the phorbol ester PDB and subsequently analyzed for receptor expression. Interestingly, similar to the TCR-tP133, the CD16/CD3␥-tP133-TCR complex was down-regulated from the cell surface following PKC activation (Fig. 2B). An approximately 5-fold higher PBD concentration was required to induce down-regulation of the CD16/CD3␥-tP133-TCR complex as compared with the TCR, which might indicate that the CD3␥ L-based motif was not as accessible for PKC in the CD16/CD3␥-tP133-TCR complex as in the TCR-tP133.
Taken together, these experiments demonstrated that the CD3␥ L-based motif is constitutively active in monomeric CD4/ CD3␥-tP133 chimeras but inactive in CD16/CD3␥-tP133-TCR complexes. Furthermore, as seen for the CD3␥ L-based motif in the TCR, the CD3␥ L-based motif in the CD16/CD3␥-tP133-TCR complex was activated following phosphorylation. These observations strongly indicated that the TCR chain allows stable cell surface expression of receptors containing CD3␥ L-based motifs by masking this motif and that phosphorylation directly influences the interaction between CD3␥ and TCR in both TCR and CD16/CD3␥-TCR complexes. Previous studies have demonstrated that the cytoplasmic tails of Ti␣, Ti␤, CD3␥, CD3␦, and CD3⑀ are dispensable for TCR expression (18, 26 -28). Thus, neither of these chains seems to be involved in masking any potential internalization/degradation motifs in the TCR. In contrast, to our knowledge the shortest TCR chain that allows TCR cell surface expression previously published contained a cytoplasmic tail of 26 amino acids (29,30). These studies support our present results. Experiments with successive truncations of the TCR chain are presently being performed to determine the minimal length of the TCR cytoplasmic tail that allows TCR expression.