Recruitment of activated p56lck on endosomes of CD2-triggered T cells, colocalization with ZAP-70.

We have previously established that upon CD2 activation of T cells, p56lck showed a transient increase in its kinase activity and was partially internalized. Here we studied the possibility that p56lck could retain its kinase activity in the endosomes of CD2-triggered cells. T cells were fractionated on a sucrose gradient, and the endosomal fraction was isolated. In CD2-triggered cells, part of Lck was internalized and presented a maximal kinase activity in the endosome-enriched fraction after 5 min, decreasing thereafter. In the endosomal fraction of activated cells, four tyrosine-phosphorylated proteins of apparent molecular masses of 30, 40, 56, and 70 kDa were detected. We demonstrated that the protein tyrosine kinase ZAP-70 was recruited to the endosomal fraction upon CD2 stimulation with kinetics similar to that of p56lck, suggesting that recruitment of protein tyrosine kinases to endosomal vesicles could promote specific transduction signals at the intracellular level.

We have previously established that upon CD2 activation of T cells, p56 lck showed a transient increase in its kinase activity and was partially internalized. Here we studied the possibility that p56 lck could retain its kinase activity in the endosomes of CD2-triggered cells. T cells were fractionated on a sucrose gradient, and the endosomal fraction was isolated. In CD2-triggered cells, part of Lck was internalized and presented a maximal kinase activity in the endosome-enriched fraction after 5 min, decreasing thereafter. In the endosomal fraction of activated cells, four tyrosine-phosphorylated proteins of apparent molecular masses of 30, 40, 56, and 70 kDa were detected. We demonstrated that the protein tyrosine kinase ZAP-70 was recruited to the endosomal fraction upon CD2 stimulation with kinetics similar to that of p56 lck , suggesting that recruitment of protein tyrosine kinases to endosomal vesicles could promote specific transduction signals at the intracellular level.
p56 lck is a nonreceptor protein tyrosine kinase (PTK) 1 of the Src family that is expressed primarily in lymphoid cells. The implication of p56 lck in T lymphocyte signal transduction has been first suggested by the observations that it is associated with either CD4 or CD8 coreceptors (Rudd et al., 1988;Veillette et al., 1988a) and that p56 lck kinase activity is increased in vitro in T cells stimulated with anti-CD4 mAbs (Veillette et al., 1988a). Similarly, incubation of T cells with anti-CD3 mAbs resulted in an increase in Lck kinase activity  and in the appearance of a 62-kDa form of this protein . The increase in apparent molecular mass of Lck was shown to be due to phosphorylation on serine and tyrosine residues and correlated with the return of tyrosine kinase activity to the basal level following activation (Veillette et al., 1988b;Danielian et al., 1989;Marth et al., 1989;Soula et al., 1994). Recently, we have shown that triggering T cells with a combination of anti-CD2 or anti-CD45 mAbs increased the kinase activity of p56 lck (Danielian et al., 1991;Marie-Cardine et al., 1992;Marie-Cardine et al., 1994).
It is well established that upon fixation of their ligand, receptor tyrosine kinases such as epidermal growth factor or platelet-derived growth factor receptor are dimerized, internalized, and processed within the endocytic pathway, where they become degraded (Carpenter and Cohen, 1976;Rosenfeld et al., 1984). Furthermore, a functional tyrosine kinase activity is necessary for the efficient degradation of the epidermal growth factor receptor after its internalization (Honegger et al., 1987;Glenney et al., 1988;Felder et al., 1990). Although internalization of receptor tyrosine kinases from the plasma membrane to endosomes has been extensively studied, delocalization of Src family PTKs has been reported recently. In platelets, p60 c-src is mainly associated with the cytoplasmic face of the plasma membrane (Ferrell et al., 1990;Anand et al., 1993) but is also found in dense granules (Rendu et al., 1989) or in the cytosol (Dhar and Shukla, 1991). Additionally, c-src is particularly enriched in a population of late endosomes in fibroblasts (Kaplan et al., 1992). In activated platelets, p60 c-src was redistributed from the plasma membrane to the cytoskeletal fraction (Grondin et al., 1991;Horvath et al., 1992;Clark and Brugge, 1993;Pumiglia and Feinstein, 1993) or to the cytosol (Walker et al., 1993), and this correlated with enhanced kinase activity. Other PTKs of the Src family, p55 fgr and p59 hck , were found to associate with secretory granules in resting neutrophils and translocate to the plasma membrane and the phagosomal membrane, respectively, upon cell activation (Gutkind and Robbins, 1989;Möhn et al., 1995). In T lymphocytes, we have described that a fraction of the plasma membrane-associated p56 lck was internalized upon cell stimulation via CD2 receptors (Marie-Cardine et al., 1992. This redistribution is specifically observed upon CD2 or CD45 triggering (Marie-Cardine et al., 1994) but not upon CD3 stimulation (Marie-Cardine et al., 1992;Ley et al., 1994), suggesting that internalization of Lck may be part of the process controlling the transduction of signals along these stimulation pathways.
Thus, we were interested in the characterization of the intracellular fraction of Lck and in its codistribution with potential substrates in CD2-triggered cells. To biochemically study the internalized fraction of Lck, a previously described preparation of endosomes was adapted to T lymphocytes (Gorvel et al., 1991). We report that Lck progressively accumulated in the endosome-enriched fraction of CD2-triggered Jurkat T cells, where its kinase activity was increased. Additionally, by Western blot analysis, we showed a parallel recruitment of the PTK ZAP-70 at the endosomal level in CD2-stimulated cells.
Cell Stimulation and mAbs-Cells were cultured at a density of 4 ϫ 10 5 cells/ml and stimulated (10 7 cells/ml) at 37°C using a combination of anti-CD2 mAbs (T11 2 , IgG2a, and T11 3 ; IgG3 was kindly provided by Dr. E. Reinherz, Dana Farber Cancer Institute, Boston, MA). Both antibodies were used at a 1/200 dilution of ascites fluid. Polyclonal anti-p56 lck was raised and used as reported previously (Fischer et al., 1987). Anti-ZAP-70 antiserum was a generous gift of Dr. O. Acuto (Pasteur Institute, Paris, France) and has been generated as described previously (Duplay et al., 1994). The peptide used for immunization comprised amino acids 485-499, located in the kinase domain of ZAP-70. These antibodies do not cross-react with the homologous PTK p72 syk , they specifically recognize ZAP-70 in immunoprecipitates from CD3-stimulated cells, and binding of the antibodies to ZAP-70 is quantitatively inhibited by preincubation with the peptide used for immunization (Duplay et al., 1994). Cell Fractionation and Endosome Isolation-Control or stimulated cells (8 ϫ 10 7 ) were resuspended in 0.6 ml of homogenization buffer (250 mM (8%) sucrose, 3 mM imidazole, pH 7.4, 1 mM sodium vanadate, 1 mM phenylmethylsulfonyl fluoride, 1 g/ml leupeptin, 1 g/ml pepstatin, 1% aprotinin) and disrupted by cavitation at 4°C in a nitrogen bomb for 4 min at 250 p.s.i. resulting in the disruption of 90% of the cells (Maridonneau-Parini et al., 1991). The homogenates were centrifuged to remove debris, nuclei and intact cells. The postnuclear supernatant (PNS) was collected and adjusted to 40.6% sucrose using 62% (w/w) sucrose. The PNS was then loaded at the bottom of a SW55 tube and sequentially overlaid with 2 ml of 35% (w/w) sucrose and 1 ml of homogeneization buffer. The tubes were centrifuged at 35,000 rpm for 90 min at 4°C in a Beckman L-80 ultracentrifuge using a SW55-Ti rotor. Fractions were collected from the top of the tube and assayed for enzymatic markers.
Immunoprecipitation, Kinase Activity and Western Blot Analysis-The endosome-enriched fractions were diluted in phosphate-buffered saline and centrifuged at 200,000 ϫ g for 45 min at 4°C to remove the sucrose.

Preparation of an Endosome-enriched Fraction from Jurkat
T Cells-To test the kinase activity of p56 lck in the endocytic fraction of CD2-triggered T cells, endosomes were isolated by cellular fractionation on sucrose gradient. Control Jurkat cells, which had internalized the fluid phase marker HRP to label the endocytic compartment, were disrupted by cavitation in a nitrogen bomb. After centrifugation, the PNS was adjusted to 40.6% sucrose and placed at the bottom of a sucrose step gradient consisting of 2 ml of 35% sucrose and 1 ml of 8% sucrose homogeneization buffer. After centrifugation, markers were assayed for the lysosomes (␤-D-glucuronidase), Golgi apparatus (galactosyltransferase), and endosomes (HRP) in each of the sucrose layers and interfaces (Fig. 1). We determined that the endocytic marker HRP was enriched at the 8/35% interface (Fig. 1A). The yield of endosomes, calculated as the percentage of the marker (HRP) present in the corresponding homogenate, is 15 Ϯ 2% (n ϭ 5). The lysosomal marker ␤-Dglucuronidase was predominantly localized in the 40.6% sucrose phase and in the 35/40.6% interface (Fig. 1B), whereas the Golgi marker galactosyltransferase was found throughout the gradient, with a peak in the 35% sucrose phase (Fig. 1C).
We have previously determined that a part of the plasma membrane-associated Lck was internalized. Therefore, it was necessary to ascertain that the endosome-enriched fraction was not contaminated by the plasma membrane. The protein tyrosine phosphatase CD45, which is located at the plasma membrane and, to a lesser extent, in the Golgi apparatus in control cells, remained associated with these compartments even after CD2 stimulation (Marie-Cardine et al., 1994). On the other hand, under the same conditions of stimulation, CD2 receptors were completely internalized (Marie-Cardine et al., 1992). Thus, we examined by Western blot analysis the distribution of both glycoproteins throughout gradients obtained from control or CD2-stimulated cells (Fig. 1D). The partitioning of CD45 along the gradient did not change upon cell stimulation (Fig.  1D, upper panel); it was recovered under both conditions in the 40.6% sucrose phase and in the 35/40.6% interface. Part of CD45 was also detected in the 35% sucrose phase, which is probably due to the location of the Golgi in this fraction. Similarly, in control cells, CD2 was predominantly found in the 40.6% sucrose phase and in the 40.6/35% interface (Fig. 1D, lower panel), confirming that these fractions contained the plasma membrane. No CD2 was detected in the endosomal fraction from nonstimulated cells. Striking changes were observed after CD2 stimulation. Indeed, CD2 was no longer detected in the higher sucrose density fractions but was almost exclusively recovered in the endosomal fraction of activated cells, confirming the complete relocation of the protein from the plasma membrane to endosomal vesicles. Taken together, our results showed that we had obtained an endosome-enriched fraction that was not contaminated by the plasma membrane or by lysosomes and only minimally by the Golgi apparatus.
Association of p56 lck with Endosome-enriched Fraction in CD2-activated Jurkat Cells-To determine if p56 lck was efficiently internalized upon CD2 activation, control cells or cells stimulated with the combination of anti-CD2 mAbs (T11 2 ϩ T11 3 ) were fractionated. The endosome-enriched fractions were collected in each tube and centrifuged at 200,000 ϫ g to remove the sucrose. The pellets were resuspended in sample buffer, and an equal amount of proteins from each sample was submitted to Western blot analysis using anti-p56 lck antibodies followed by 125 I-protein A. The results presented in Fig. 2 show that a small amount of p56 lck was present in the endosomal fraction of control cells, consistent with recent data indicating that a fraction of Lck is present on vesicular structures surrounding the centrosome (Ley et al., 1994). After 1 min of stimulation, a slight increase in the amount of Lck in this intracellular compartment was observed. The maximum level of Lck in the endosomal fraction was observed after 5 min of stimulation, remaining constant at later time points of 30 min (Fig. 2) and 1 h (data not shown). These results are in agreement with our previous studies using indirect immunofluorescence in which intracellular staining of p56 lck was found to persist for 1 h after activation of T cells via CD2 (Marie-Cardine et al., 1992). Additionally, the higher molecular mass form of Lck, migrating at approximately 62 kDa, appeared in the endosomal fraction between 1 and 5 min after stimulation (Fig. 2). The bands corresponding to Lck were cut out from the nitrocellulose, and the incorporated radioactivity was measured. By comparison with the total amount of p56 lck present in the cells, we estimated that the fraction of Lck in the endosomes was 3.2 Ϯ 0.5% in control cells (t ϭ 0 min) and 12.8 Ϯ 2.6% (n ϭ 3) 5 min after CD2 activation. This indicates that roughly 10% of Lck was endocytosed in CD2-triggered T cells.
Determination of p56 lck Kinase Activity in the Endosomal Fraction of CD2-stimulated Jurkat Cells-The kinase activity of p56 lck was analyzed in parallel in PNS (Fig. 3A) and in endosome-enriched fractions (Fig. 3B) of control and CD2-stimulated cells. After addition of the mAbs to the cell suspensions, the kinase activity of immunoprecipitated p56 lck from cell lysates (PNS) presented a transient increase as previously observed (Danielian et al., 1991;Marie-Cardine et al., 1992) After 1 min of stimulation (Fig. 3A), phosphorylation levels of p56 lck and enolase reached a maximum value and then decreased (t ϭ 5 min) and returned to basal level at 15 min. The higher molecular mass form of the protein appeared after 5 min of activation (Fig. 3A, open arrowhead) and is predominant at 30 min, thus correlating with the return of the kinase activity to basal level.
Next, the kinase activity of Lck was analyzed in the endosome-enriched fraction. As shown in Fig. 3B, p56 lck presented a delayed activation in this fraction when compared to the PNS. In control cells (t ϭ 0), almost no kinase activity was detected. Upon stimulation, autophosphorylation of Lck and phosphorylation of enolase reached a maximum value after 5 min of CD2 triggering. Between 15 and 30 min it began to decrease; at these time points, the higher molecular mass form of p56 lck was visualized. An additional phosphorylated band migrating at 50 kDa was also detected, probably corresponding to the heavy chain of immunoglobulin.
The specific activity of p56 lck was estimated in both PNS and endosomal fractions of CD2-triggered cells (Fig. 3C). The bands corresponding to enolase and to Lck (in immunoblots) were cut out from the gels and nitrocellulose sheets, respectively, and the incorporated radioactivity was counted (n ϭ 3). After 1 min of cell stimulation, Lck specific activity presented a 2.7 Ϯ 0.31-fold increase in the PNS, while, at the same time, it was not modified in the endosome-enriched fraction (0.95 Ϯ 0.26).

FIG. 1. Isolation of an endosomal-enriched fraction on a sucrose gradient.
Jurkat cells (8 ϫ 10 7 ) were incubated with HRP for 20 min at 37°C and disrupted, and the PNS was isolated and loaded on the bottom of a sucrose step gradient. After centrifugation, fractions were collected and analyzed for HRP activity (endosomal marker) (A), ␤-D-glucuronidase activity (lysosomal marker) (B), and galactosyltransferase activity (Golgi marker) (C). The relative specific activity of HRP was calculated from its specific activity (OD units per mg of protein) in the fraction divided by its specific activity in the PNS. D, control cells (8 ϫ 10 7 ) or cells stimulated via CD2 for 15 min (8 ϫ 10 7 ) were disrupted and fractionated on sucrose gradient. Western blot analyses were performed for each fraction using anti-CD45 (upper panels) and anti-CD2 (lower panels) antibodies. I1, 8/35% interface; I2, 35/40.6% interface.
FIG. 2. Association of p56 lck with endosomes upon CD2 stimulation. Cells were incubated with medium alone (t ϭ 0) or with the combination of anti-CD2 mAbs (T11 2 ϩ T11 3 ) for 1, 5, 15, and 30 min. Endosome-enriched fractions were prepared and equal amounts of proteins were separated on a 10% SDS-polyacrylamide gel, transferred to nitrocellulose, and analyzed with anti-p56 lck antibodies followed by 125 I-protein A. Positions of molecular mass markers (kDa) are indicated on the left of the autoradiogram.
This suggested that at this time point, the activated fraction of the kinase was still at the plasma membrane level. After 5 min of activation, the specific activity of p56 lck in both the PNS and the endosomal fraction increased by 1.6-fold (1.6 Ϯ 0.22 and 1.6 Ϯ 0.12, respectively), indicating that an activated fraction of Lck was internalized upon cell triggering.
Colocalization of p56 lck with Proteins Phosphorylated on Tyrosine Residues and with the PTK ZAP-70 -Taken together, our results indicate that p56 lck is transiently under its activated form in the endosomal fraction of CD2-stimulated T cells. Western blot analysis was then performed on endosomal proteins from CD2-stimulated cells (t ϭ 0, 1, 5, 15, 30, and 60 min) using antiphosphotyrosine mAbs (Fig. 4A). In control cells (t ϭ 0 min), the antiphosphotyrosine antibodies reacted with two proteins of apparent molecular mass of approximately 56 and 70 kDa (p56 and p70, respectively). In cells triggered for 1 min with the combination of anti-CD2 mAbs, p56 and p70 presented increased labeling with antiphosphotyrosine antibodies that remained constant until 60 min. Additionally, two other proteins of approximately 40 and 30 kDa (p40 and p30) became detectable after 1 min. The phosphotyrosine staining of p40 decreased thereafter, whereas the reactivity of anti-phosphotyrosine antibodies toward p30 increased, reaching a maximum after 5 min of activation and decreasing thereafter. The same pattern of tyrosine-phosphorylated proteins was obtained after preclearing of the fractions with protein A-Sepharose before protein separation (data not shown), indicating that anti-CD2 mAbs did not interfere with the assay.
We found that a protein of 70 kDa was present in the endosome-enriched fraction. As a recent study reported that p56 lck interacts via its src homology-2 (SH2) domain with ZAP-70 (approximately 62 kDa). C, incorporation of 32 P in enolase from in vitro kinase assay and 125 I-radiolabeled bands corresponding to immunoblotted Lck were counted at each time point in PNS and endosomal fractions; the specific activity of Lck was expressed in arbitrary units. The results shown are mean Ϯ S.D. of three independent experiments. *, p Ͻ 0.05 when compared with controls (0 min) calculated with paired Student's t test.

FIG. 3. Time course of p56 lck kinase activity in lysates (PNS) and in the endosomal fractions of CD2-stimulated cells.
A, Jurkat cells were incubated alone (t ϭ 0) or with the combination of anti-CD2 mAbs at 37°C for 1, 5, 15, and 30 min. At each time point, cells were disrupted by cavitation and aliquots of the PNS were collected. Immunoprecipitation of p56 lck was performed on equal protein amounts (350 g, corresponding to 5 ϫ 10 6 cell eq). Immunoprecipitates were incubated with the exogenous substrate enolase in the presence of [␥-32 P]ATP and 4.5 M of unlabeled ATP and resolved on a 8% SDSpolyacrylamide gel. The autoradiogram was exposed for 4 h at Ϫ70°C. B, p56 lck kinase activity was measured in the endosomal fraction isolated after fractionation on sucrose gradient of the PNS obtained in A. The kinase was assayed after immunoprecipitation of equal protein amounts of each sample (70 g corresponding to 8 ϫ 10 7 cell eq) as  (Duplay et al., 1994), we investigated the possibility that ZAP-70 could associate with the endosomal fraction of CD2triggered cells by performing Western blot analyses with anti-ZAP-70 antibodies. As shown in Fig. 4B, ZAP-70 appeared in the endosome-enriched fraction as early as 1 min after the addition of the anti-CD2 mAbs, with peak levels observed after 5 min of activation. The level of ZAP-70 in the endosomal fraction decreased after 15 min of activation and remained constant between 15 and 30 min.

DISCUSSION
In the present report, we describe that stimulation of T cells via CD2 resulted in a time-dependent accumulation in the endosome-enriched fraction of an active population of p56 lck , which colocalized with the PTK ZAP-70.
In Jurkat cells stimulated with an activating combination of anti-CD2 mAbs, we have previously shown, by indirect immunofluorescence, that Lck was partially relocated from the plasma membrane to endocytic vesicles (Marie-Cardine et al., 1992). To isolate the endosome-enriched fraction from T lymphocytes, we adapted a subcellular fractionation method previously applied to several cell types (Gorvel et al., 1991Escola et al., 1995). After centrifugation of cell lysates on a discontinuous sucrose gradient, the partitioning of cellular organelles was tested using enzymatic markers. A fraction enriched in endosomes, separated from the plasma membrane, was obtained. This was an important point as it allowed us to clearly distinguish the endosome-associated p56 lck from the pool that remained bound to the plasma membrane upon T cell stimulation via CD2.
Fractionation experiments showed that in Jurkat cells triggered via CD2, the internalized fraction of p56 lck presented a 1.6-fold increase in its specific activity. The kinase activity of Lck recovered in the endosome-enriched fraction reached a maximum with delayed kinetics when compared to that of PNS. One min after activation, an increase in the amount of internalized Lck was detectable, whereas the activity of the enzyme was similar to that obtained in control cells. On the other hand, the kinase activity detected in the PNS at this time point was maximum, and the maximal kinase activity in the endosomal-fraction was reached 5 min after the addition of the anti-CD2 mAbs. It has been estimated that in control Jurkat cells, 2% of p56 lck interacts with CD2 (Bell et al., 1992;Carmo et al., 1993). In stimulated cells, we have estimated that about 13% of the Lck expressed in the cells was recovered in the endosome-enriched fraction. Therefore, changes in the degree of interaction between CD2 and p56 lck could occur upon activation. As CD2 is devoid of enzymatic activity, it has been proposed that activation of Lck through CD2 involves other molecules. One appropriate candidate is the protein tyrosine phosphatase CD45, since it has been shown that its expression at the cell surface is necessary for Lck activation upon CD2 stimulation  and that triggering of the cells with either anti-CD2 or anti-CD45 mAbs induced similar internalization of Lck and increase in its kinase activity (Marie-Cardine et al., 1994). Thus, it has been proposed that a functional complex between CD2, CD45, and p56 lck is formed and involved in the activation of the kinase. It is possible that shortly after incubation with the anti-CD2 mAbs, pre-existing CD2/p56 lck complexes (representing 2% of cellular p56 lck ) were endocytosed, leading to the presence of nonactivated Lck at the endosomal level after 1 min of stimulation. Simultaneously, at the cell surface, free molecules of CD2 could be triggered and functional complexes consisting of CD2, CD45, and Lck could be formed, which reflects the maximal activation of p56 lck observed in the PNS 1 min after cell stimulation. This activated fraction of Lck could then be internalized, accounting for the peak of the kinase activity in the endosomal-enriched fraction at 5 min. At this point, CD2 receptors are endocytosed while CD45 phosphatase remains at the plasma membrane (Marie-Cardine et al., 1994), suggesting that the transduction complexes are in part disrupted after enhancement of p56 lck activity. Further identification of the molecular basis of the interaction between CD2, CD45, and p56 lck will allow a better understanding of the mechanism of CD2-mediated signal transduction.
As Lck is a nonreceptor PTK, it is presumably associated with the inner face of the plasma membrane and the cytoplasmic face of endocytic vesicles. A consequence of the redistribution of Lck in CD2-stimulated cells may be the distinct accessibility of the kinase to regulatory proteins and substrates. We report here that the internalized fraction of p56 lck colocalized with a few proteins phosphorylated on tyrosine residues. By Western blot analysis, four proteins of 70, 56, 40, and 30 kDa were detected. It is possible that the 56-kDa protein was p56 lck itself. Indeed, p56 lck can be phosphorylated on tyrosine residues independently of its activity. The negative regulation site, Tyr-505, is phosphorylated when the enzyme activity is repressed and its autophosphorylation site, Tyr-394, is generally phosphorylated when the kinase is activated. The tyrosinephosphorylated protein p70 could be the PTK ZAP-70. In endosomal fractions from control cells, ZAP-70 was not detected, even after long time exposure of the blot, whereas the protein p70 was observed with the antiphosphotyrosine antibodies. This suggests that the bulk of the 70-kDa signal is probably not ZAP-70. A functional interaction between p56 lck and tyrosinephosphorylated ZAP-70 has been described previously (Duplay et al., 1994). Upon CD2 stimulation, the amounts of Lck and ZAP-70 in the endosomal fraction increased in parallel. Whether these two kinases interact at the endosomal level remains to be established. We also detected the presence of two tyrosine-phosphorylated proteins of 30 and 40 kDa in the endosomal fraction of CD2-activated cells. The level of p40 phosphotyrosine staining did not seem to be related to the activation state of p56 lck , suggesting that Lck is not directly responsible for p40 phosphorylation. On the contrary, tyrosine phosphorylation detected on p30 was modulated according to the time of stimulation. Several proteins of similar molecular mass interacting with p56 lck have been described, but none have been shown to be tyrosine phosphorylated in vivo upon CD2 stimulation (Schraven et al., 1991(Schraven et al., , 1992Ross et al., 1994). Identification of this protein will be of interest because it appeared in the endosomal-enriched fraction upon CD2 activation and was tyrosine phosphorylated in parallel to the activation of Lck. However, at present, we have been unable to detect any protein coimmunoprecipitating with Lck, probably because of the low amount of material present in the endosomal fractions.
In conclusion, we showed that a fraction of p56 lck is recruited at the endosomal level in CD2-triggered T cells. In the endosomes, the kinase activity of Lck and tyrosine phosphorylation of proteins increased in parallel. These observations might be critical in the transduction of CD2-mediated activation signals.