Extracellular signal-regulated kinase phosphorylates tumor necrosis factor alpha-converting enzyme at threonine 735: a potential role in regulated shedding

Abstract

The ectodomain of certain transmembrane proteins can be released by the action of cell surface proteases, termed secretases. Here we have investigated how mitogen-activated protein kinases (MAPKs) control the shedding of membrane proteins. We show that extracellular signal-regulated kinase (Erk) acts as an intermediate in protein kinase C-regulated TrkA cleavage. We report that the cytosolic tail of the tumor necrosis factor alpha-converting enzyme (TACE) is phosphorylated by Erk at threonine 735. In addition, we show that Erk and TACE associate. This association is favored by Erk activation and by the presence of threonine 735. In contrast to the Erk route, the p38 MAPK was able to stimulate TrkA cleavage in cells devoid of TACE activity, indicating that other proteases are also involved in TrkA shedding. These results demonstrate that secretases are able to discriminate between the different stimuli that trigger membrane protein ectodomain cleavage and indicate that phosphorylation by MAPKs may regulate the proteolytic function of membrane secretases.

Figure 1

(A) Osmotic stress induces TrkA cleavage. CHO^TrkA cells were treated with PMA (1 μM) or high salt (1 M NaCl) for 30 min, followed by immunoprecipitation and Western analysis with the anti-panTrk antiserum. p41/p40 denotes TrkA fragments generated upon holoreceptor cleavage. Ig, immunoglobulin heavy chain. M_r markers are shown at the right. (B) Effect of the Erk1/2 pathway inhibitors PD98059 and U0126 on PMA-induced TrkA cleavage. CHO^TrkA cells were preincubated with the inhibitors (P98059: 50 μM; U0126: 10 μM) for 30 min before PMA addition. Immunoprecipitation and Western analysis was performed with the anti-panTrk antiserum. (C) Effect of a dominant negative form of Erk2 on PMA-induced TrkA cleavage. 293^TrkA cells were transfected with a plasmid encoding an HA-tagged dominant negative form of Erk2 (HA-Erk2^K52R). PMA treatment, immunoprecipitation, and Western analysis of TrkA cleavage was as above. Western blotting of cell lysates with the anti-HA antibody showed the mutated form of Erk2 (HA-Erk2^K52R) in the corresponding transfectants.

Figure 2

Multiple MAPK pathways participate in TrkA ectodomain shedding. (A) Effect of different treatments and MAPK inhibitors on TrkA cleavage. PMA or UV irradiation was used as an activator of Erk1/2 or the p38 and JNK stress pathways, respectively, whereas sorbitol was used as a nonspecific activator of MAPK pathways. Where indicated, CHO^TrkA cells were preincubated with PD98059 (50 μM) or SB203580 (10 μM) for 30 min before PMA, sorbitol, or UV treatment. Cell lysates were analyzed with the corresponding antibodies (indicated at the right of each panel), except for anti-panTrk and anti-pJNK, for which the samples were immunoprecipitated with anti-panTrk and anti-JNK antibodies before Western analysis. (B) Effect of PD98059 and SB203580 on TrkA cleavage. Inhibitors were added to cells, where indicated, before sorbitol treatment. TrkA cleavage was analyzed as described above. (C) Activation of Erk1/2 by sorbitol is independent of PKC. BIM (10 μM) was added to cells for 15 min before PMA or sorbitol and TrkA cleavage was analyzed as above. pErk1/2 was analyzed in cell lysates by Western blotting with an antibody that recognizes the dually phosphorylated forms of Erk1 and Erk2. (D) The metalloprotease inhibitor BB3103 prevents sorbitol and PMA-induced TrkA cleavage. BB3103 (30 μM) was added to cells for 30 min before sorbitol or PMA treatment, and TrkA cleavage was analyzed by immunoprecipitation and Western blotting as described.

Figure 3

Participation of TACE in TrkA cleavage. (A) A polyclonal affinity-purified antibody to the C terminus of TACE or the monoclonal M220 antibody was used to immunoprecipitate TACE from lysates of 293 cells. Where indicated, 10 μg of the peptide used for the generation of the polyclonal antibody (TACE peptide) were added to the immunoprecipitates. Western analysis was performed with the affinity-purified anti-TACE antibody. (B) Distribution of TACE and TrkA in 293^TrkA cells. Monolayers were fixed, permeabilized, and incubated with affinity-purified anti-TACE or anti-TrkA ectodomain antibody MGR12, and images were taken in a confocal microscope. Bar, 15 μm. (C) Expression of TACE and TACE-ΔZn. Lysates from fibroblasts derived from wild-type (TACE^+/+) and TACE^ΔZn/ΔZn animals were analyzed by immunoprecipitation and Western blotting with the affinity-purified anti-TACE antibody. Where indicated, 10 μg of the TACE peptide were included in the immunoprecipitates. TACE immunoprecipitated from 293 cells is shown at the left. (D) Accumulation of cell-bound TrkA-truncated fragments in TACE^ΔZn/ΔZn−TrkA cells. Lysates from TACE^ΔZn/ΔZn−TrkA cells were treated with PMA (1 μM), high salt (1 M NaCl), or sorbitol (0.5 M) for 30 min, immunoprecipitated, and analyzed by Western blotting with the anti-panTrk antiserum. (E) Biochemical evidence that TACE can rescue TrkA cleavage. TACE^ΔZn/ΔZn−TrkA cells were infected with retrovirus containing pLZR-TACE-IRES-GFP or pLZR-IRES-GFP and treated with PMA where indicated. Lysates were analyzed for TrkA cleavage (top) or TACE content by Western blotting. The asterisks denote two additional bands present in cells infected with pLZR-TACE-IRES-GFP. (F) Reconstitution of TrkA ectodomain cleavage in TACE^ΔZn/ΔZn−TrkA cells. Cells were infected with the retrovirus with the pLZR-TACE-IRES-GFP vector and treated (bottom) or not (top) with PMA for 30 min. Cells were then fixed, permeabilized, and incubated with the anti-TrkA ectodomain and endodomain antibodies, followed by secondary antibodies labeled with Cy5 (blue, anti-mouse) or Cy3 (red, anti-rabbit). Triple color images were obtained in a confocal microscope. Notice that cells with a high GFP (and thus TACE, asterisks) amount had lost most of the blue (ectodomain epitope) fluorescence but still kept high amounts of red (endodomain epitope) signal in cells treated with PMA. The central cell in the bottom (arrowhead), which did not capture pLZR-TACE-IRES-GFP, did not respond to PMA by a decrease in ectodomain staining. Bar, 30 μm.

Figure 4

Phosphorylation of TACE by Erk. (A) Primary sequence of a region of the mouse TACE intracellular domain indicating (arrow) a potential MAPK phosphorylation site. Bottom, an autoradiogram of an in vitro phosphorylation of TACE by Erk. Assays were performed by using bacterially produced GST-Erk or anti-Erk immunoprecipitates (IP anti-Erk) as the enzyme and 10 μg of GST, GST-Elk or GST-TACE as the substrate. To obtain activated Erk from HeLa cell lysates, cells were treated with PMA for 30 min before lysis and immunoprecipitated with the anti-Erk antibody. (B) Results from an in vitro kinase assay using bacterially produced GST-TACE or GST-TACE-T⁷³⁵A. Lysates from control or PMA-treated HeLa cells were immunoprecipitated with the anti-Erk antibody, and the immunoprecipitates were tested for their ability to phosphorylate in vitro 10 μg of GST-TACE or GST-TACE-T⁷³⁵A. The amounts of the recombinant proteins used in the kinase assays were verified by Coomassie staining of gels run in parallel (Díaz-Rodríguez, Montero, Esparís-Ogando, Yuste, and Pandiella, unpublished data). (C) Phosphoamino acid analysis of an experiment identical to that shown in B. The phosphorylated bands were excised, proteins were eluted and hydrolyzed, and phosphoamino acids were identified by two-dimensional electrophoresis. (D) In vivo phosphorylation of TACE by PMA. Top, 293 cells labeled with ³²P were treated with PMA for 30 min where indicated. TACE was immunoprecipitated with the anti-TACE antibody, and the immunoprecipitates were analyzed by SDS-PAGE, followed by autoradiography. An additional PMA-treated sample was incubated with an excess (10 μg) of the peptide used to raise the anti-TACE antiserum. The asterisks indicate the position of two bands of unknown identity. Bottom, effect of the Erk pathway inhibitor PD98059 on TACE phosphorylation. Where indicated, cells labeled with ³²P were incubated with PD98059 (50 μM) for 30 min before PMA treatment. Lysates were immunoprecipitated with the anti-TACE antibody, followed by SDS-PAGE and autoradiography.

Figure 5

Erk phosphorylates TACE at T⁷³⁵ in vivo. (A) Immunoreaction of the anti-pTP antibody with phosphorylated GST-TACE and GST-Elk. GST-TACE or GST-Elk was subjected to in vitro phosphorylation using cold ATP as the phosphate donor and GST-Erk as the enzyme. Reactions were stopped by the addition of sample buffer and divided into two aliquots that were subjected to electrophoresis in 12% SDS-PAGE gels. The gel was blotted to an Immobilon membrane that was divided in two identical parts. One was probed with the anti-pTP antibody, and the other was probed with the same antibody that had previously been preadsorbed with 10 mM phosphothreonine. (B) The anti-pTP antibody failed to react with GST-TACE-T⁷³⁵A. GST-TACE or GST-TACE-T⁷³⁵A was subjected to in vitro phosphorylation using cold ATP as the phosphate donor and immunoprecipitated Erk as the enzyme. Reactions were stopped by the addition of sample buffer and divided into two aliquots that were subjected to electrophoresis in 12% SDS-PAGE gels. One of the gels was blotted to an Immobilon membrane that was probed with the anti-pTP antibody (top); the other gel was stained with Coomassie (bottom). (C) PMA increased phosphorylation of TACE at T⁷³⁵. Where indicated, 293 cells were treated with PMA and okadaic acid (1 μM) for 30 min and then lysed and immunoprecipitated with the anti-TACE antibody. An additional PMA-treated sample was incubated during the immunoprecipitation with an excess (10 μg) of the peptide used to raise the anti-TACE antiserum. The immunoprecipitates were analyzed by SDS-PAGE, followed by Western blotting with the anti-pTP antibody (top). The amount of TACE present in the immunoprecipitated samples was analyzed after stripping and reprobing of the blot with the anti-TACE antibody (bottom). The signals were quantitated using the NIH Image 1.6 software, and the anti-pTP/anti-TACE ratio was calculated. A ratio of 1 was considered that of the untreated control sample. (D) Effect of PD98059 on TACE T⁷³⁵ phosphorylation. Where indicated, cells were incubated with PD98059 (50 μM) for 30 min before PMA treatment. Lysates were immunoprecipitated with the anti-TACE antibody, and Western blots were probed with anti-pTP (top) or anti-TACE antibodies (bottom). (E) In vivo phosphorylation of TACE at T⁷³⁵ detected by Western blotting with the anti-pTP antibody. 293 cells transfected with HA-TACE or HA-TACE-T⁷³⁵A were treated with PMA, immunoprecipitated with the anti-HA antibody, and Western blots were probed with anti-pTP (top). After stripping, the blot was reprobed with anti-HA (bottom). (F) Effect of a dominant negative form of Erk2 on TACE phosphorylation at T⁷³⁵. 293^TrkA cells were infected with retrovirus containing pLZR-IRES-GFP or pLZR-HA-Erk2^K52R-IRES-GFP and treated with PMA where indicated. Cell lysates were lysed in 1 ml of lysis buffer and 20 μl were analyzed for HA-Erk2^K52R (top) or total Erk2 (endogenous+ HA-Erk2^K52R, second panel from top) by Western blotting. In parallel, blots were also probed with anti-pErk (third panel from top). A 200-μl aliquot was immunoprecipitated with anti-Erk2 antibodies, and the immunoprecipitates were used for in vitro kinase reactions (fourth panel from top). Another 500-μl aliquot was immunoprecipitated with the anti-TACE C terminus antibody, and the blot was probed with anti-pTP (fifth panel from top). The amount of TACE in the samples was analyzed by Western blotting with anti-TACE (bottom). (G) Phosphorylation of TACE at T⁷³⁵ upon receptor tyrosine kinase activation. 293^TrkA cells were incubated with EGF (10 nM) or NGF (50 ng/ml) for 20 min and then lysed and immunoprecipitated with anti-TACE. Blots were probed with anti-pTP (top) or anti-TACE (bottom) antibodies. In parallel, 20-μl aliquots of the samples were run in a 10% gel and the blot was probed with anti-pErk antibodies (middle). (H) Expression of TACE-T⁷³⁵A inhibits NGF-induced TACE phosphorylation at T⁷³⁵. 293^TrkA cells were infected with retrovirus containing pLZR-TACE-IRES-GFP or pLZR-TACE-T⁷³⁵A-IRES-GFP. Cells were stimulated with NGF (50 ng/ml) for 20 min and lysed, and equal amounts of protein were immunoprecipitated with anti-TACE. Blots were probed with anti-pTP (top), stripped and reprobed with anti-TACE (bottom).

Figure 6

Association of TACE and Erk. (A) In vitro association of TACE and Erk. GST, GST-TACE, or GST-TACE-T⁷³⁵A coupled to GSH beads was incubated with extracts from control or PMA-treated 293 cells. After incubation at 4°C, the beads were washed with PBS, and Erk associated with the beads was detected by Western blotting with the anti-Erk antibody (top). Bottom, a Coomassie stain of the GST, GST-TACE, or GST-TACE-T⁷³⁵A loaded. (B) 293^TrkA cells were treated with PMA or vehicle, and lysates were immunoprecipitated with anti-TACE or anti-TGFα antibodies. Western blots were then probed with anti-Erk (left) or anti-TACE (right) antibodies. (C) 293^TrkA cells were treated with PMA or vehicle, and lysates were immunoprecipitated with anti-TACE, anti-p-Erk1/2, or anti-Erk antibodies, followed by Western blotting with anti-TACE antibodies. (D) Active TACE is not required for association with Erk. Extracts from control and PMA-treated TACE^ΔZn/ΔZn cells were lysed and then protein-A-Sepharose beads were added alone (Beads) or together with the indicated antibodies (anti-TACE or anti-TGFα). In parallel, extracts from 293^TrkA cells pretreated with BB3103 (20 μM) were also immunoprecipitated with anti-TACE antibodies. Precipitates were washed and analyzed for Erk presence by Western blotting with anti-p-Erk antibodies.

Figure 7

Effect of TACE and TACE-T⁷³⁵A on TrkA cleavage in TACE^ΔZn/ΔZn-TrkA cells. (A) Cells were infected with pLZR-TACE-IRES-GFP, pLZR-TACE-T⁷³⁵A-IRES-GFP, or pLZR-IRES-GFP. Cells were treated (where indicated) with PMA for 30 min. TrkA cleavage (top) and TACE content (bottom) were analyzed by Western blotting. (B) Effect of U0126 on PMA-induced TrkA cleavage in TACE^ΔZn/ΔZn-TrkA cells. Cells were infected with pLZR-TACE-IRES-GFP or with pLZR-IRES-GFP. Where indicated, U0126 (10 μM) was included in the incubation media for 30 min before PMA treatment. TrkA cleavage, p-Erk1/2, and TACE were analyzed by Western blotting as above.