- Email: [email protected]
Multimerisation of A disintegrin and metalloprotease protein-17 (ADAM17) is mediated by its EGF-like domain
Biochemical and Biophysical Research Communications 415 (2011) 330–336
Contents lists available at SciVerse ScienceDirect
Biochemical and Biophysical Research Communications journal homepage: www.elsevier.com/locate/ybbrc
Multimerisation of A disintegrin and metalloprotease protein-17 (ADAM17) is mediated by its EGF-like domain Inken Lorenzen, Ahmad Trad, Joachim Grötzinger ⇑ Biochemisches Institut der Christian-Albrechts-Universität Kiel, Olshausenstr. 40, 24118 Kiel, Germany
a r t i c l e
i n f o
Article history: Received 28 September 2011 Available online 18 October 2011 Keywords: ADAM17 Metalloprotease Metzincins EGF-like domain Cysteine-rich domain Multimerisation Dimer
a b s t r a c t A disintegrin and metalloprotease protein 17 (ADAM17) is a transmembrane zinc dependent metalloprotease. The catalytic activity of the enzyme results in the shedding of a broad range of membrane proteins. The release of the corresponding ectodomains induces a switch in various physiological and pathophysiological processes. So far there is not much information about the molecular mechanism of ADAM17 activation available. As for other transmembrane proteases, multimerisation may play a critical role in the activation and function of ADAM17. The present work demonstrates that ADAM17 indeed exists as a multimer in the cell membrane and that this multimerisation is mediated by its EGF-like domain. Ó 2011 Elsevier Inc. All rights reserved.
1. Introduction The zinc dependent transmembrane membrane metalloprotease ADAM17 (a disintegrin and metalloproteinase 17) is responsible for the shedding of a large number of substrates, so far more than 75 are described [1]. Among these proteins are growth factors, like TGF-a and adhesion molecules, like ICAM-1 as well as cytokines and cytokine receptors, like Tumor Necrosis Factor-a (TNF-a) nd the interleukin-6 receptor (IL-6R) [1]. Shedding of these proteins results not only in the release of their ectodomains from the cell surface, but also in the generation of soluble mediators, like cytokines and agonistic receptors like TNF-a [2,3] and the IL-6R, respectively [4]. The broad substrate spectrum reflects the importance of ADAM17 as a key regulator in various physiological and pathophysiological processes, like development, inflammation and cancer progression. The importance of this enzyme is underlined by the fact that ADAM17 knockout mice are embryonic lethal [5]. The knowledge about the activation and function of ADAM17 is still incomplete. Members of the ADAM-family are multi-domain type-I transmembrane proteins with an extracellular part consisting of up to five extracellular domains; a pro-, a catalytic-, a disintegrin-, a cysteine rich- and an Epidermal Growth Factor- (EGF-)like-domain [6]. It is well known that the pro-domain is needed for the proper Abbreviations: ADAM17, a disintegrin and metalloprotease 17; TNFa, tumor necrosis factor a; IL-6R, interleukin-6 receptor; DN-ADAM17, dominate-negative ADAM17; GPI, glycosylphosphatidylinositol; Mef, mouse embryonic fibroblasts. ⇑ Corresponding author. Fax: +49 431 880 2007. E-mail addresses: [email protected] (I. Lorenzen), atrad@biochem. uni-kiel.de (A. Trad), [email protected] (J. Grötzinger). 0006-291X/$ - see front matter Ó 2011 Elsevier Inc. All rights reserved. doi:10.1016/j.bbrc.2011.10.056
folding and inhibition of the catalytic domain. Disintegrin domains are involved in cell-adhesion [7], whereby the disintegrin domain of ADAM17 is able to interact with a5b1 integrin [8]. Furthermore, disintegrin domains are supposed to act as a scaffold for its C-terminal neighboring domain [6]. ADAM17 and ADAM10 are atypical members of the ADAM-family due to their lack of a cysteine-rich domain whose function is probably taken over by their EGF-like domains [6,9]. It has been supposed that the cysteine-rich-, or in case of ADAM10 and ADAM17 the EGF-like-domain, are participating in activation, thereby supporting the removal of the pro-peptide from the catalytic domain [10] and in the specific recognition of substrates [9,11]. Also the transmembrane region has been described to participate in the interaction with substrates [12]; whereas the cytoplasmic region is known to be involved in regulatory events [1,13,14]. Members of the meprin- and membrane-type matrix metalloproteases (MT-MMPs) are, like the ADAMs, multi-domain transmembrane zinc depended metalloproteases and members of the metzincins superfamily. All members of this superfamily contain a conserved (HEXXHXXG/NXXH/D) consensus sequence in the active site of the catalytic domain which coordinates the zinc ion [15]. Members of the Meprins and MT-MMPs exist as multimers. Meprins are known to generate disulfide linked homo- or heterodimers. The mature meprin-a subunit is not membrane anchored and forms high aggregated protease complexes in solution [15–17]. The dimerisation of the meprin-a subunits with the meprin-b subunit leads to the generation of disulfide-linked membrane anchored heterodimers of meprin A. This heterodimer tends to further noncovalent aggregation leading to membrane anchored tetramers. In contrast, meprin B consistent of two meprin-b subunits and exist
TCAGCTCAGCGTAATCCGGAACATCGTAT GGGTAG TCAGCTCAGCGTAATCCGGAACATCGT ATGGGTAG
GATCCTACCCATACGATGTTCCGGATTACGCT TGAGC
humane DNADAM17 humane DNADAM17 humane DNADAM17 DN-ADAM17 GPI
DN-ADAM17 D DisGPI
humane DNADAM17 DN-ADAM17 D cyto
DN-ADAM17 D Dis
murine ADAM17 GPI
ADAM17
murine flulllength ADAM 17
ADAM17
Tag Primer 30 Tag Primer 50 Primer 30
GGGGTACCATGAGGCGGCGTCTCC CGGGATCCGCACTCTGTCTCTTTGCTGTCAACTCG GATCCGATTACAAGGATGACGACGA TAAGTGAGC FLAG GGGGTACCATGAGGCGGCGTCTCC CGGGATCCGCACTCTGTCTCTTTGCTGTCAACTCG GGCCGCTCACTTGCCATCGATC AGCCGTGGATCGACCTGGTCTTCG PC GGGGTACCATGAGGCGGCGTCTCC GCCGTGGATCGACCTGGTCTTCGGATCCGTC GGGAATTCGCTCAGCTCCTTGCCAT AATGAAATCCCAAAATCGC CGATCAGC CGTGGATCGACCTGG HA GGGGTACCGCCACCATGAACTC ATAGTTTAGCGGCCGCTTTATCCAATTTCTTA GGCCGCTCAAGCGTAATCCGG CTTCTCC TCCACACAATGG AACATCGTA TGGGTAG HA GGGGTACCGCCACCATGAACTC CGGGATCCCTGGTCAATGAAATCCCAAAATCG CTTCTCC Myc TCCGCTCGAGTTCTGCGAGAGG CCCAAGCTTTCAGCACTCTGTTTCTTTGCTGTC GATCCTACCCATACGATGTTCCG GAACAGC AACACG GATTACGCTGAGC HA TCCGGGGCCCTTCTGCGAGAGG CGGGATCCCTGGTCAATGAAATCCCAAAATCG GATCCTACCCATACGATGTTCCG GAACAGC GATTACGCTGAGC
Primer 50 Tag
Pc ADAM17 murine flulllength ADAM 17
Tamplate Origine Construct
Table 1 Oligonucleotides used for cloning of ADAM17 variants.
GGCCGCTCACTTGCCATCGATCAGCCGTG GATCGACCTGGTCTTCG GGCCGCTCACTTATCGTCGTCATCCTTGTAATCG
I. Lorenzen et al. / Biochemical and Biophysical Research Communications 415 (2011) 330–336
331
as a homodimer in the plasma membrane [16,17]. Also MT1-MMP is a multimer as the formation of dimers is mandatory for its biological activities, like proMMP2 activation and collagen degradation [18]. The close relation of those multi-domain proteases to ADAM17 and the fact that ADAM17, lacking the catalytic domain, the so called dominant-negative (DN) ADAM17, acts as inhibitor of the wildtype molecule [19], implicate that ADAM-proteases might also exist as dimers or multimers. In order to study the multimerisation of ADAM17 differently tagged molecules were generated and used for co-immunoprecipitation experiments. The resulting data show that ADAM17 indeed exists as multimer. By using different deletion variants of ADAM17, we identified the EGF-like domain as the minimal requirement for multimerisation. 2. Material and methods 2.1. Generation of C-terminal tagged full-length ADAM17 and deletion variants The murine full-length ADAM17 cDNA was amplified by PCR, flanked 50 -end by a KpnI and 30 -end by a BamHI restriction site. Oligonucleotides encoding the PC- or FLAG-tag (Table 1) containing a 50 -end BamHI- and a 30 -end NotI restriction site were phosphorylated and hybridized (Eurofins MWG Operon, Germany). Both fragments were ligated into pcDNA3.1 neo (+) digested with KpnI and NotI (Fermentas, St. Leon-Rot, Germany). The resulting vectors were used to generate the C-terminal tagged deletion variants. In order to generate HA-tagged variants the C-terminal tag was removed using BamHI and NotI restriction enzymes, and oligonucleotides coding for the HA-tag and the respective restriction sites were ligated into these vectors (Table 1). DN-ADAM17 was used as template for constructs lacking the metalloprotease domain [20]. The DN-ADAM17DDis construct containing a C-terminal myc-tag (Table 1) was cloned behind the IL-6R signal peptide into pcDNA3.1 zeo() using XhoI and HindIII restriction enzymes (Fermentas, St. Leon-Rot, Germany). Glycosylphosphatidylinositol (GPI) anchored variants were cloned into pcDNA3.1 (+), which comprises the coding sequence for a myc-tag and the GPI-signal sequence of TRAIL behind a KpnI and a BlpI restriction sites. To introduce an additional tag in front of the GPI anchor, the sequence of a PC-tag flanked 50 -end by a BamHI- and 30 -end by a BlpI site was attached to the extracellular domain of ADAM17 by two successive PCR reactions (Table 1), and cloned into the GPI-vector using KpnI and BlpI sites. The resulting plasmid was used for cloning GPI deletion variants between KpnI and BamHI sites, and for introduction of the HA-tag between BamHI and BlpI sites. 2.2. Co-immunoprecipitation HEK-293 cells and MefADAM17ex/ex [20] were cultured in DMEM high-glucose containing 10% fetal calf serum (FCS), penicillin (60 mg/l) and streptomycin (100 mg/l) (PAA Laboratories, Marburg, Germany). Transfection was performed using 15 ll Turbofect (Fermentas, St. Leon-Rot, Germany) and 5 or 6 lg DNA (HEK296 or MefADAM17ex/ex) according to manufacturer’s instructions. For co-transfection experiments 2.5 lg or 3 lg (HEK296 or MefADAM17ex/ex) of each construct was used. After transfection the cells were cultured for 2 days in DMEM high-glucose containing 5 % FCS, harvested, lysed in 250 ll lysis buffer containing 20 mM Tris/HCl pH 7.6, 150 mM NaCl, 2 mM EDTA and ‘‘complete’’ protease inhibitor mixture without EDTA (Roche applied science, Mannheim, Germany) and afterwards centrifuged for 15 min at 14.000 rpm. For co-immunoprecipitation experiments the lysates were divided into three parts. Two times 100 ll were mixed with
332
I. Lorenzen et al. / Biochemical and Biophysical Research Communications 415 (2011) 330–336
100 ll lysis buffer and 40 ll of protein-G agarose (Thermo Scientific-Pierce, Perbio Science, Bonn, Germany). The lysis buffer used for precipitation with the HPC4 antibody contained additional 4 mM CaCl2. In one of these tubes 2–3 lg of the respective antibody was added and both samples incubated over night at 4 °C by gentle agitation. Afterwards protein-G agarose was washed five-times with 750 ll of lysis buffer containing 4 mM CaCl2 and mixed with 40 ll of 2.5-fold concentrated SDS-sample buffer. As input control the remaining 50 ll of the pre-cleared lysate were mixed with 40 ll 2.5-fold concentrated SDS-sample buffer. 2.3. Western blotting To detect the co-immunoprecipitated proteins, 25 ll of the sample was loaded onto a 12.5% or 15% SDS–PAGE. As loading controls 2.5–5% of the inputs of the co-immunoprecipitated samples were loaded on a distinct gel. After separation by SDS–PAGE proteins were transferred onto a polyvinylidene fluoride membrane (GE Healthcare, Munich, Germany). The membrane was blocked with 5% skimmed milk powder in 50 mM Tris/HCl, pH 7.6, 200 mM NaCl for 2 h. Depending on the tag, the membrane was probed with rabbit anti-HA (C29F4), rabbit anti-FLAG, murine anti-myc antibody (New England, Biolabs – Cell Signaling, Germany), or murine antiPC antibody (HPC4) in 50 mM Tris/HCl, pH 7.6, 200 mM NaCl, 0.05 % Triton X-100 containing 1% BSA, in case of HPC4 additional 3 mM CaCl2 was added. After incubation over night (4 °C) the membranes were washed and incubated with anti-mouse HRP antibody or anti-rabbit HRP antibody for 1 h at room temperature (Thermo Scientific-Pierce, Perbio Science, Bonn, Germany). After washing the antigen was visualized using the ECL detection system (GE Healthcare, Munich, Germany) according to the manufacturer’s instructions. 2.4. Flow cytometry Two days after transfection HEK293 cells were harvested and prepared for FACS analysis as described earlier [21]. Thereby 106 cells were stained for 1 h at 4 °C with 2 lg/ml of purified A300E antibody [21], washed and incubated with 1 lg of allophycocyanin-conjugated goat anti-mouse antibody (Jackson ImmunoResearch Laboratories, USA). After an additional washing-step cells were analyzed with a FACSCanto flow cytometer (BD Biosciences, USA). 3. Results 3.1. Construction and expression of ADAM17 variants To analyze whether ADAM17 exists in a multimeric state in the cell membrane, co-immunoprecipitation experiments of differently tagged ADAM17 variants were performed. To this end, full-length ADAM17 was C-terminal tagged either with a FLAG-, a PC- or a HA-tag (Fig. 1A). To clarify the role of the different domains, constructs with successive deletions were generated (Fig. 1 first column). Deletion of the catalytic domain results in a construct called DN-ADAM17, which was described earlier and contains a C-terminal myc-tag (Fig. 1B) [22]. The variant lacking the catalytic- and disintegrin-domain (DN-ADAM17DDis), thereby consisting solely of the EGF-like domain, contains the same tag (Fig 1C). To distinguish whether the cytoplasmic- or the transmembraneregion is involved in the multimerisation process, additional variants were constructed lacking either the cytoplasmic-, but still containing the transmembrane-region (Fig. 1D and E); or missing both, the cytoplasmic- and transmembrane-region, but fused to a
glycosylphosphatidylinositol (GPI) anchor in order to ensure membrane localization (Fig. 1F–H). Both constructs containing the whole extracellular part, but missing either the cytoplasmic- (ADAM17Dcyto) or the cytoplasmic- and transmembrane-region (ADAM17– GPI) are labeled with a PC-tag (Fig. 1D and F). Variants with the deletion of the catalytic domain lacking the cytoplasmic tail (DN-ADAM17Dcyto) or the corresponding GPI construct (DNADAM17-GPI) where either tagged with a PC- or a HA-tag (Fig. 1E and G). The same tags were used for the two constructs encoding the sole EGF-like domain fused by a GPI-anchor (Fig. 1H). As it is shown in Fig. 1 (second column) all constructs are expressed and detectable by Western Blotting using the corresponding antibodies. The cell-surface expression was confirmed by FACS analysis using an antibody directed against the EGF-like domain of ADAM17 (Fig. 1 third column). 3.2. ADAM17 exist as a multimer in the cell membrane To analyze the multimerisation of ADAM17 the above described ADAM17 constructs were applied in co-immunoprecipitation experiments. First, two differently C-terminal tagged full-length ADAM17 variants were co-transfected into HEK293 cells. As shown in Fig. 2A the PC-tagged full-length ADAM17 was able to co-immunoprecipitate the HA-tagged ADAM17, demonstrating that ADAM17 exists as a multimer. To clarify whether the cytoplasmic- and/or the transmembraneregion of two molecules are involved in the observed multimerisation, the HA-tagged full-length ADAM17 was co-transfected either with ADAM17 lacking the cytoplasmic- (Fig. 2B) or lacking the cytoplasmic- and the transmembrane-region (Fig. 2C), both labeled with a PC-tag. The pull down of the PC-tagged variants precipitated in both cases the full-length HA-tagged ADAM17, implying that at least one cytoplasmic- and transmembrane region is dispensable for the formation of the multimers. The minimal version of a multimeric ADAM17 is a dimer. To study which one of the extracellular domains is responsible for multimerisation HEK293 cells were co-transfected with full-length FLAG-tagged ADAM17 together with DN-ADAM17, the variant missing the catalytic domain. As presented in Fig. 2D, at least one catalytic domain is dispensable for multimerisation, because DN-ADAM17 precipitates full-length ADAM17. To exclude mutual effects resulting from the transmembrane-/intracellular-region, DN-ADAM17 constructs without the cytoplasmic- (DNADAM17Dcyto) or without the transmembrane- and cytoplasmic-region (DN-ADAM17-GPI) were used (Fig. 2E and F). In both experiments the deletion variants were precipitated with the full-length FLAG-tagged molecule. In a next step we examined whether the disintegrin- or the EGFlike domain or both are needed for the multimerisation. Therefore DN-ADAM17DDis and DN-ADAM17DDis-GPI, both lacking the catalytic and the disintegrin domain, were used for co-transfection with the wild-type molecule. The first construct, DN-ADAM17DDis, contain the whole cytoplasmic- and transmembrane-region whereas the second construct, DN-ADAM17DDis-GPI, is missing these regions (Fig. 2G and H). Both constructs were able to precipitate full-length ADAM17, implying that the sole EGF-like domain is the minimal requirement needed for multimerisation of these molecules. The next question we asked is which domain/s is/are needed on the second molecule. To this end, cells were co-transfected with two different tagged identical deletion variants. In the first experiment two DN-ADAM17 variants without the cytoplasmic-region (DN-ADAM17Dcyto) either tagged with a HA- or a PC-tag were transfected. The PC-tagged protein was precipitated from the lysates using HPC4 antibodies and the HA-tagged variant was co-precipitated as shown in Fig. 3A. To exclude the possibility that
I. Lorenzen et al. / Biochemical and Biophysical Research Communications 415 (2011) 330–336
333
Fig. 1. Variants of ADAM17. The first column present a schematic illustration of ADAM17 constructs. (A) Present the full-length ADAM17, followed by DN-ADAM17 (B), DNADAM17DDis (C), ADAM17Dcyto (D), DN-ADAM17Dcyto (E), ADAM17-GPI (F), DN-ADAM17-GPI (G) and DN-ADAM17DDis-GPI (H). The expression analyses by Western blot and FACS analyses are presented in the second and third column. For Western blot analysis HEK293 cells were harvested and lysed 3 days after transfection and the expression was detected using corresponding antibodies (right lane presents the sample of interest and the left lane an unrelated sample as negative control). Using A300E as primary antibody the cell surface expression was monitored by FACS analysis.
334
I. Lorenzen et al. / Biochemical and Biophysical Research Communications 415 (2011) 330–336
Fig. 2. ADAM17 has the property to multimerise. HEK293 cells were co-transfected with either HA- or FLAG-tagged full-length ADAM17 and a following variants: PC-tagged full-length ADAM17 (A), PC-tagged ADAM17Dcyto (B), PC-tagged ADAM17-GPI (C), myc-tagged DN-ADAM17 (D), PC-tagged DN-ADAM17Dcyto (E), PC-tagged DN-ADAM17GPI (F), myc-tagged DN-ADAM17DDis (G) and PC-tagged DN-ADAM17DDis-GPI (H). Three days later cells were harvested, lysed and subjected to co-immunoprecipitation experiments. The presents of ADAM17 constructs was confirmed by Western blotting using the corresponding antibodies.
I. Lorenzen et al. / Biochemical and Biophysical Research Communications 415 (2011) 330–336
335
mice [20], which show a reduction of the ADAM17 mRNA level of more than 95%, were co-transfected with the HA- and PC-tagged DN-ADAM17DDis-GPI constructs. Consistently with the above described experiments, the PC tagged variant of DN-ADAM17DDis-GPI was able to precipitate the HA-tagged variant. This result confirms that the sole EGF-like domain is indeed responsible for multimerisation of ADAM17 and that multimerisation is mediated by a minimum of two of these domains. 4. Discussion
Fig. 3. Multimerisation of ADAM17 is mediated by its EGF-like domain. HEK293 cells were co-transfected using the same ADAM17 variants either tagged with a HAor a PC-tag: DN-ADAM17Dcyto (A), DN-ADAM17-GPI (B) and DN-ADAM17DDisGPI (C). The cells were harvested and lysed 3 days after transfection. After coimmunoprecipitation the samples were analyzed by Western blotting using HPC4 and anti-HA (C29F4). (D) MefADAM17ex/ex were transfected with the HA- and PCtagged DN-ADAM17DDis-GPI constructs and 3 days later subjected to co-immunoprecipitation experiments.
the multimerisation is mediated by the transmembrane region we performed the same experiment using GPI anchored DN-ADAM17 (DN-ADAM17-GPI) containing only the disintegrin- and EGF-like domain. Again the HA-tagged variant was detectable by precipitating the PC-tagged variant (Fig. 3B). To distinguish whether multimerisation is mediated via an EGF-like/disintegrin or an EGF-like/EGF-like interaction cells were co-transfected with the HA- and PC-tagged DN-ADAM17DDis-GPI variants. Both constructs consists solely of the EGF-like domain and are anchored by GPI in the membrane. As shown in Fig. 3C the PC-tagged variant was able to precipitate the HA-tagged DN-ADAM17DDis-GPI. From these experiments we conclude that the EGF-like domain alone is responsible for multimerisation of ADAM17. In order to confirm the responsibility of the EGF-like domain for the observed multimerisation we aimed to perform the same experiment in cells without ADAM17 expression. HEK293 cells express endogenous ADAM17 at moderate levels, therefore mouse embryonic fibroblasts (Mef) generated from ADAM17ex/ex (MefADAM17ex/ex)
The aim of this work was to gain insight in the multimerisation properties of the ADAM17. By co-immunoprecipitation experiments we demonstrate that ADAM17 has the potential to form multimers. Multimerisation of transmembrane proteases of the metzincins superfamily was described for meprin A and B as well as for MT1-MMP [15,16,18]. In case of MT1-MMP the minimal formation of a dimer is needed for its biological activity [18]. Mutations within the interaction site of the domain that mediates dimerisation, namely the hemopexin-domain, abrogates the ability of proMMP2 activation and collagen degradation [18]. The data presented here do not allow a conclusion about the number of molecules which form the ADAM17 multimer, but demonstrate that multimerisation is mediated by only one domain, the EGF-like domain. Additionally to its function in multimerisation, the EGF-like domain is involved in substrate recognition [9,11] and enzyme activation by facilitating the removal of the inhibitory pro-peptide from the catalytic domain [10]. The use of deletion variants clearly demonstrates that the transmembrane- as well as the cytoplasmicregion of ADAM17 is dispensable for this process. According to this, the intracellular region of ADAM17 is not involved in the generation of multimers, but it cannot be excluded that it takes part in the separation of ADAM17 multimers due to the induction of a conformational change, like it is described for the inside-out signaling of integrins [23]. The inside-out signaling transformed the low-affinity state of integrin heterodimers to a high-affinity state by the separation of the intracellular parts in response to an extra-cellular stimulus. The GPI anchored ADAM17 variant is unable to shed its substrates although it contains the complete ectodomain [12]. This effect is not due to the loss of ADAM17 multimers, in accordance with the presented data. Therefore it should be addressed whether multimerisation of ADAM17 is mandatory for shedding of its substrates, or whether multimers constitute an inactive state. In such a scenario ADAM17 has to become monomeric to be active implicating a conformational change that additional support the release of the pro-peptide from the catalytic active site. Acknowledgments The authors would like to thank Christoph Garbers for the full-length murine ADAM17 and murine GPI anchored construct and Athena Chalaris for the DN-ADAM17 and MefADAM17ex/ex. This study has been supported by the Deutsche Forschungsgemeinschaft (SFB 877, A6, Z3) and the Excellence Cluster ‘Inflammation at Interfaces’. References [1] J. Scheller, A. Chalaris, C. Garbers, S. Rose-John, ADAM17: a molecular switch to control inflammation and tissue regeneration, Trends Immunol. 32 (2011) 380–387. [2] R.A. Black, C.T. Rauch, C.J. Kozlosky, J.J. Peschon, J.L. Slack, M.F. Wolfson, B.J. Castner, K.L. Stocking, P. Reddy, S. Srinivasan, N. Nelson, N. Boiani, K.A. Schooley, M. Gerhart, R. Davis, J.N. Fitzner, R.S. Johnson, R.J. Paxton, C.J. March, D.P. Cerretti, A metalloproteinase disintegrin that releases tumour-necrosis factor-alpha from cells, Nature 385 (1997) 729–733.
336
I. Lorenzen et al. / Biochemical and Biophysical Research Communications 415 (2011) 330–336
[3] M.L. Moss, S.L. Jin, M.E. Milla, D.M. Bickett, W. Burkhart, H.L. Carter, W.J. Chen, W.C. Clay, J.R. Didsbury, D. Hassler, C.R. Hoffman, T.A. Kost, M.H. Lambert, M.A. Leesnitzer, P. McCauley, G. McGeehan, J. Mitchell, M. Moyer, G. Pahel, W. Rocque, L.K. Overton, F. Schoenen, T. Seaton, J.L. Su, J.D. Becherer, et al., Cloning of a disintegrin metalloproteinase that processes precursor tumour-necrosis factor-alpha, Nature 385 (1997) 733–736. [4] J. Müllberg, E. Dittrich, L. Graeve, C. Gerhartz, K. Yasukawa, T. Taga, T. Kishimoto, P.C. Heinrich, S. Rose-John, Differential shedding of the two subunits of the interleukin-6 receptor, FEBS Lett. 332 (1993) 174–178. [5] J.J. Peschon, J.L. Slack, P. Reddy, K.L. Stocking, S.W. Sunnarborg, D.C. Lee, W.E. Russell, B.J. Castner, R.S. Johnson, J.N. Fitzner, R.W. Boyce, N. Nelson, C.J. Kozlosky, M.F. Wolfson, C.T. Rauch, D.P. Cerretti, R.J. Paxton, C.J. March, R.A. Black, An essential role for ectodomain shedding in mammalian development, Science 282 (1998) 1281–1284. [6] S. Takeda, Three-dimensional domain architecture of the ADAM family proteinases, Semin. Cell Dev. Biol. 20 (2009) 146–152. [7] J.M. White, ADAMs: modulators of cell–cell and cell–matrix interactions, Curr. Opin. Cell Biol. 15 (2003) 598–606. [8] D.V. Bax, A.J. Messent, J. Tart, M. van Hoang, J. Kott, R.A. Maciewicz, M.J. Humphries, Integrin alpha5beta1 and ADAM-17 interact in vitro and colocalize in migrating HeLa cells, J. Biol. Chem. 279 (2004) 22377–22386. [9] P.W. Janes, N. Saha, W.A. Barton, M.V. Kolev, S.H. Wimmer-Kleikamp, E. Nievergall, C.P. Blobel, J.P. Himanen, M. Lackmann, D.B. Nikolov, Adam meets Eph: an ADAM substrate recognition module acts as a molecular switch for ephrin cleavage in trans, Cell 123 (2005) 291–304. [10] M.E. Milla, M.A. Leesnitzer, M.L. Moss, W.C. Clay, H.L. Carter, A.B. Miller, J.L. Su, M.H. Lambert, D.H. Willard, D.M. Sheeley, T.A. Kost, W. Burkhart, M. Moyer, R.K. Blackburn, G.L. Pahel, J.L. Mitchell, C.R. Hoffman, J.D. Becherer, Specific sequence elements are required for the expression of functional tumor necrosis factor-alpha-converting enzyme (TACE), J. Biol. Chem. 274 (1999) 30563–30570. [11] P. Reddy, J.L. Slack, R. Davis, D.P. Cerretti, C.J. Kozlosky, R.A. Blanton, D. Shows, J.J. Peschon, R.A. Black, Functional analysis of the domain structure of tumor necrosis factor-alpha converting enzyme, J. Biol. Chem. 275 (2000) 14608– 14614. [12] X. Li, L. Pérez, Z. Pan, H. Fan, The transmembrane domain of TACE regulates protein ectodomain shedding, Cell Res. 17 (2007) 985–998.
[13] S.M. Soond, B. Everson, D.W. Riches, G. Murphy, ERK-mediated phosphorylation of Thr735 in TNFalpha-converting enzyme and its potential role in TACE protein trafficking, J. Cell Sci. 118 (2005) 2371–2380. [14] P. Xu, R. Derynck, Direct activation of TACE-mediated ectodomain shedding by p38 MAP kinase regulates EGF receptor-dependent cell proliferation, Mol. Cell 37 (2010) 551–566. [15] E.E. Sterchi, W. Stocker, J.S. Bond, Meprins, membrane-bound and secreted astacin metalloproteinases, Mol. Aspects Med. 29 (2008) 309–328. [16] P. Marchand, J. Tang, J.S. Bond, Membrane association and oligomeric organization of the alpha and beta subunits of mouse meprin A, J. Biol. Chem. 269 (1994) 15388–15393. [17] G.P. Bertenshaw, M.T. Norcum, J.S. Bond, Structure of homo- and heterooligomeric meprin metalloproteases, dimers, tetramers, and high molecular mass multimers, J. Biol. Chem. 278 (2003) 2522–2532. [18] A. Tochowicz, P. Goettig, R. Evans, R. Visse, Y. Shitomi, R. Palmisano, N. Ito, K. Richter, K. Maskos, D. Franke, D. Svergun, H. Nagase, W. Bode, Y. Itoh, The dimer interface of the membrane type 1 matrix metalloproteinase hemopexin domain: crystal structure and biological functions, J. Biol. Chem. 286 (2011) 7587–7600. [19] K.A. Solomon, N. Pesti, G. Wu, R.C. Newton, Cutting edge: a dominant negative form of TNF-alpha converting enzyme inhibits proTNF and TNFRII secretion, J. Immunol. 163 (1999) 4105–4108. [20] A. Chalaris, N. Adam, C. Sina, P. Rosenstiel, J. Lehmann-Koch, P. Schirmacher, D. Hartmann, J. Cichy, O. Gavrilova, S. Schreiber, T. Jostock, V. Matthews, R. Hasler, C. Becker, M.F. Neurath, K. Reiss, P. Saftig, J. Scheller, S. Rose-John, Critical role of the disintegrin metalloprotease ADAM17 for intestinal inflammation and regeneration in mice, J. Exp. Med. 207 (2010) 1617–1624. [21] A. Trad, N. Hedemann, M. Shomali, V. Pawlak, J. Grötzinger, I. Lorenzen, Development of sandwich ELISA for detection and quantification of human and murine a disintegrin and metalloproteinase17, J. Immunol. Methods 371 (2011) 91–96. [22] A. Chalaris, B. Rabe, K. Paliga, H. Lange, T. Laskay, C.A. Fielding, S.A. Jones, S. Rose-John, J. Scheller, Apoptosis is a natural stimulus of IL6R shedding and contributes to the pro-inflammatory trans-signaling function of neutrophils, Blood 110 (2007) 1748–1755. [23] C.L. Abram, C.A. Lowell, The ins and outs of leukocyte integrin signaling, Annu. Rev. Immunol. 27 (2009) 339–362.
Contents lists available at SciVerse ScienceDirect
Biochemical and Biophysical Research Communications journal homepage: www.elsevier.com/locate/ybbrc
Multimerisation of A disintegrin and metalloprotease protein-17 (ADAM17) is mediated by its EGF-like domain Inken Lorenzen, Ahmad Trad, Joachim Grötzinger ⇑ Biochemisches Institut der Christian-Albrechts-Universität Kiel, Olshausenstr. 40, 24118 Kiel, Germany
a r t i c l e
i n f o
Article history: Received 28 September 2011 Available online 18 October 2011 Keywords: ADAM17 Metalloprotease Metzincins EGF-like domain Cysteine-rich domain Multimerisation Dimer
a b s t r a c t A disintegrin and metalloprotease protein 17 (ADAM17) is a transmembrane zinc dependent metalloprotease. The catalytic activity of the enzyme results in the shedding of a broad range of membrane proteins. The release of the corresponding ectodomains induces a switch in various physiological and pathophysiological processes. So far there is not much information about the molecular mechanism of ADAM17 activation available. As for other transmembrane proteases, multimerisation may play a critical role in the activation and function of ADAM17. The present work demonstrates that ADAM17 indeed exists as a multimer in the cell membrane and that this multimerisation is mediated by its EGF-like domain. Ó 2011 Elsevier Inc. All rights reserved.
1. Introduction The zinc dependent transmembrane membrane metalloprotease ADAM17 (a disintegrin and metalloproteinase 17) is responsible for the shedding of a large number of substrates, so far more than 75 are described [1]. Among these proteins are growth factors, like TGF-a and adhesion molecules, like ICAM-1 as well as cytokines and cytokine receptors, like Tumor Necrosis Factor-a (TNF-a) nd the interleukin-6 receptor (IL-6R) [1]. Shedding of these proteins results not only in the release of their ectodomains from the cell surface, but also in the generation of soluble mediators, like cytokines and agonistic receptors like TNF-a [2,3] and the IL-6R, respectively [4]. The broad substrate spectrum reflects the importance of ADAM17 as a key regulator in various physiological and pathophysiological processes, like development, inflammation and cancer progression. The importance of this enzyme is underlined by the fact that ADAM17 knockout mice are embryonic lethal [5]. The knowledge about the activation and function of ADAM17 is still incomplete. Members of the ADAM-family are multi-domain type-I transmembrane proteins with an extracellular part consisting of up to five extracellular domains; a pro-, a catalytic-, a disintegrin-, a cysteine rich- and an Epidermal Growth Factor- (EGF-)like-domain [6]. It is well known that the pro-domain is needed for the proper Abbreviations: ADAM17, a disintegrin and metalloprotease 17; TNFa, tumor necrosis factor a; IL-6R, interleukin-6 receptor; DN-ADAM17, dominate-negative ADAM17; GPI, glycosylphosphatidylinositol; Mef, mouse embryonic fibroblasts. ⇑ Corresponding author. Fax: +49 431 880 2007. E-mail addresses: [email protected] (I. Lorenzen), atrad@biochem. uni-kiel.de (A. Trad), [email protected] (J. Grötzinger). 0006-291X/$ - see front matter Ó 2011 Elsevier Inc. All rights reserved. doi:10.1016/j.bbrc.2011.10.056
folding and inhibition of the catalytic domain. Disintegrin domains are involved in cell-adhesion [7], whereby the disintegrin domain of ADAM17 is able to interact with a5b1 integrin [8]. Furthermore, disintegrin domains are supposed to act as a scaffold for its C-terminal neighboring domain [6]. ADAM17 and ADAM10 are atypical members of the ADAM-family due to their lack of a cysteine-rich domain whose function is probably taken over by their EGF-like domains [6,9]. It has been supposed that the cysteine-rich-, or in case of ADAM10 and ADAM17 the EGF-like-domain, are participating in activation, thereby supporting the removal of the pro-peptide from the catalytic domain [10] and in the specific recognition of substrates [9,11]. Also the transmembrane region has been described to participate in the interaction with substrates [12]; whereas the cytoplasmic region is known to be involved in regulatory events [1,13,14]. Members of the meprin- and membrane-type matrix metalloproteases (MT-MMPs) are, like the ADAMs, multi-domain transmembrane zinc depended metalloproteases and members of the metzincins superfamily. All members of this superfamily contain a conserved (HEXXHXXG/NXXH/D) consensus sequence in the active site of the catalytic domain which coordinates the zinc ion [15]. Members of the Meprins and MT-MMPs exist as multimers. Meprins are known to generate disulfide linked homo- or heterodimers. The mature meprin-a subunit is not membrane anchored and forms high aggregated protease complexes in solution [15–17]. The dimerisation of the meprin-a subunits with the meprin-b subunit leads to the generation of disulfide-linked membrane anchored heterodimers of meprin A. This heterodimer tends to further noncovalent aggregation leading to membrane anchored tetramers. In contrast, meprin B consistent of two meprin-b subunits and exist
TCAGCTCAGCGTAATCCGGAACATCGTAT GGGTAG TCAGCTCAGCGTAATCCGGAACATCGT ATGGGTAG
GATCCTACCCATACGATGTTCCGGATTACGCT TGAGC
humane DNADAM17 humane DNADAM17 humane DNADAM17 DN-ADAM17 GPI
DN-ADAM17 D DisGPI
humane DNADAM17 DN-ADAM17 D cyto
DN-ADAM17 D Dis
murine ADAM17 GPI
ADAM17
murine flulllength ADAM 17
ADAM17
Tag Primer 30 Tag Primer 50 Primer 30
GGGGTACCATGAGGCGGCGTCTCC CGGGATCCGCACTCTGTCTCTTTGCTGTCAACTCG GATCCGATTACAAGGATGACGACGA TAAGTGAGC FLAG GGGGTACCATGAGGCGGCGTCTCC CGGGATCCGCACTCTGTCTCTTTGCTGTCAACTCG GGCCGCTCACTTGCCATCGATC AGCCGTGGATCGACCTGGTCTTCG PC GGGGTACCATGAGGCGGCGTCTCC GCCGTGGATCGACCTGGTCTTCGGATCCGTC GGGAATTCGCTCAGCTCCTTGCCAT AATGAAATCCCAAAATCGC CGATCAGC CGTGGATCGACCTGG HA GGGGTACCGCCACCATGAACTC ATAGTTTAGCGGCCGCTTTATCCAATTTCTTA GGCCGCTCAAGCGTAATCCGG CTTCTCC TCCACACAATGG AACATCGTA TGGGTAG HA GGGGTACCGCCACCATGAACTC CGGGATCCCTGGTCAATGAAATCCCAAAATCG CTTCTCC Myc TCCGCTCGAGTTCTGCGAGAGG CCCAAGCTTTCAGCACTCTGTTTCTTTGCTGTC GATCCTACCCATACGATGTTCCG GAACAGC AACACG GATTACGCTGAGC HA TCCGGGGCCCTTCTGCGAGAGG CGGGATCCCTGGTCAATGAAATCCCAAAATCG GATCCTACCCATACGATGTTCCG GAACAGC GATTACGCTGAGC
Primer 50 Tag
Pc ADAM17 murine flulllength ADAM 17
Tamplate Origine Construct
Table 1 Oligonucleotides used for cloning of ADAM17 variants.
GGCCGCTCACTTGCCATCGATCAGCCGTG GATCGACCTGGTCTTCG GGCCGCTCACTTATCGTCGTCATCCTTGTAATCG
I. Lorenzen et al. / Biochemical and Biophysical Research Communications 415 (2011) 330–336
331
as a homodimer in the plasma membrane [16,17]. Also MT1-MMP is a multimer as the formation of dimers is mandatory for its biological activities, like proMMP2 activation and collagen degradation [18]. The close relation of those multi-domain proteases to ADAM17 and the fact that ADAM17, lacking the catalytic domain, the so called dominant-negative (DN) ADAM17, acts as inhibitor of the wildtype molecule [19], implicate that ADAM-proteases might also exist as dimers or multimers. In order to study the multimerisation of ADAM17 differently tagged molecules were generated and used for co-immunoprecipitation experiments. The resulting data show that ADAM17 indeed exists as multimer. By using different deletion variants of ADAM17, we identified the EGF-like domain as the minimal requirement for multimerisation. 2. Material and methods 2.1. Generation of C-terminal tagged full-length ADAM17 and deletion variants The murine full-length ADAM17 cDNA was amplified by PCR, flanked 50 -end by a KpnI and 30 -end by a BamHI restriction site. Oligonucleotides encoding the PC- or FLAG-tag (Table 1) containing a 50 -end BamHI- and a 30 -end NotI restriction site were phosphorylated and hybridized (Eurofins MWG Operon, Germany). Both fragments were ligated into pcDNA3.1 neo (+) digested with KpnI and NotI (Fermentas, St. Leon-Rot, Germany). The resulting vectors were used to generate the C-terminal tagged deletion variants. In order to generate HA-tagged variants the C-terminal tag was removed using BamHI and NotI restriction enzymes, and oligonucleotides coding for the HA-tag and the respective restriction sites were ligated into these vectors (Table 1). DN-ADAM17 was used as template for constructs lacking the metalloprotease domain [20]. The DN-ADAM17DDis construct containing a C-terminal myc-tag (Table 1) was cloned behind the IL-6R signal peptide into pcDNA3.1 zeo() using XhoI and HindIII restriction enzymes (Fermentas, St. Leon-Rot, Germany). Glycosylphosphatidylinositol (GPI) anchored variants were cloned into pcDNA3.1 (+), which comprises the coding sequence for a myc-tag and the GPI-signal sequence of TRAIL behind a KpnI and a BlpI restriction sites. To introduce an additional tag in front of the GPI anchor, the sequence of a PC-tag flanked 50 -end by a BamHI- and 30 -end by a BlpI site was attached to the extracellular domain of ADAM17 by two successive PCR reactions (Table 1), and cloned into the GPI-vector using KpnI and BlpI sites. The resulting plasmid was used for cloning GPI deletion variants between KpnI and BamHI sites, and for introduction of the HA-tag between BamHI and BlpI sites. 2.2. Co-immunoprecipitation HEK-293 cells and MefADAM17ex/ex [20] were cultured in DMEM high-glucose containing 10% fetal calf serum (FCS), penicillin (60 mg/l) and streptomycin (100 mg/l) (PAA Laboratories, Marburg, Germany). Transfection was performed using 15 ll Turbofect (Fermentas, St. Leon-Rot, Germany) and 5 or 6 lg DNA (HEK296 or MefADAM17ex/ex) according to manufacturer’s instructions. For co-transfection experiments 2.5 lg or 3 lg (HEK296 or MefADAM17ex/ex) of each construct was used. After transfection the cells were cultured for 2 days in DMEM high-glucose containing 5 % FCS, harvested, lysed in 250 ll lysis buffer containing 20 mM Tris/HCl pH 7.6, 150 mM NaCl, 2 mM EDTA and ‘‘complete’’ protease inhibitor mixture without EDTA (Roche applied science, Mannheim, Germany) and afterwards centrifuged for 15 min at 14.000 rpm. For co-immunoprecipitation experiments the lysates were divided into three parts. Two times 100 ll were mixed with
332
I. Lorenzen et al. / Biochemical and Biophysical Research Communications 415 (2011) 330–336
100 ll lysis buffer and 40 ll of protein-G agarose (Thermo Scientific-Pierce, Perbio Science, Bonn, Germany). The lysis buffer used for precipitation with the HPC4 antibody contained additional 4 mM CaCl2. In one of these tubes 2–3 lg of the respective antibody was added and both samples incubated over night at 4 °C by gentle agitation. Afterwards protein-G agarose was washed five-times with 750 ll of lysis buffer containing 4 mM CaCl2 and mixed with 40 ll of 2.5-fold concentrated SDS-sample buffer. As input control the remaining 50 ll of the pre-cleared lysate were mixed with 40 ll 2.5-fold concentrated SDS-sample buffer. 2.3. Western blotting To detect the co-immunoprecipitated proteins, 25 ll of the sample was loaded onto a 12.5% or 15% SDS–PAGE. As loading controls 2.5–5% of the inputs of the co-immunoprecipitated samples were loaded on a distinct gel. After separation by SDS–PAGE proteins were transferred onto a polyvinylidene fluoride membrane (GE Healthcare, Munich, Germany). The membrane was blocked with 5% skimmed milk powder in 50 mM Tris/HCl, pH 7.6, 200 mM NaCl for 2 h. Depending on the tag, the membrane was probed with rabbit anti-HA (C29F4), rabbit anti-FLAG, murine anti-myc antibody (New England, Biolabs – Cell Signaling, Germany), or murine antiPC antibody (HPC4) in 50 mM Tris/HCl, pH 7.6, 200 mM NaCl, 0.05 % Triton X-100 containing 1% BSA, in case of HPC4 additional 3 mM CaCl2 was added. After incubation over night (4 °C) the membranes were washed and incubated with anti-mouse HRP antibody or anti-rabbit HRP antibody for 1 h at room temperature (Thermo Scientific-Pierce, Perbio Science, Bonn, Germany). After washing the antigen was visualized using the ECL detection system (GE Healthcare, Munich, Germany) according to the manufacturer’s instructions. 2.4. Flow cytometry Two days after transfection HEK293 cells were harvested and prepared for FACS analysis as described earlier [21]. Thereby 106 cells were stained for 1 h at 4 °C with 2 lg/ml of purified A300E antibody [21], washed and incubated with 1 lg of allophycocyanin-conjugated goat anti-mouse antibody (Jackson ImmunoResearch Laboratories, USA). After an additional washing-step cells were analyzed with a FACSCanto flow cytometer (BD Biosciences, USA). 3. Results 3.1. Construction and expression of ADAM17 variants To analyze whether ADAM17 exists in a multimeric state in the cell membrane, co-immunoprecipitation experiments of differently tagged ADAM17 variants were performed. To this end, full-length ADAM17 was C-terminal tagged either with a FLAG-, a PC- or a HA-tag (Fig. 1A). To clarify the role of the different domains, constructs with successive deletions were generated (Fig. 1 first column). Deletion of the catalytic domain results in a construct called DN-ADAM17, which was described earlier and contains a C-terminal myc-tag (Fig. 1B) [22]. The variant lacking the catalytic- and disintegrin-domain (DN-ADAM17DDis), thereby consisting solely of the EGF-like domain, contains the same tag (Fig 1C). To distinguish whether the cytoplasmic- or the transmembraneregion is involved in the multimerisation process, additional variants were constructed lacking either the cytoplasmic-, but still containing the transmembrane-region (Fig. 1D and E); or missing both, the cytoplasmic- and transmembrane-region, but fused to a
glycosylphosphatidylinositol (GPI) anchor in order to ensure membrane localization (Fig. 1F–H). Both constructs containing the whole extracellular part, but missing either the cytoplasmic- (ADAM17Dcyto) or the cytoplasmic- and transmembrane-region (ADAM17– GPI) are labeled with a PC-tag (Fig. 1D and F). Variants with the deletion of the catalytic domain lacking the cytoplasmic tail (DN-ADAM17Dcyto) or the corresponding GPI construct (DNADAM17-GPI) where either tagged with a PC- or a HA-tag (Fig. 1E and G). The same tags were used for the two constructs encoding the sole EGF-like domain fused by a GPI-anchor (Fig. 1H). As it is shown in Fig. 1 (second column) all constructs are expressed and detectable by Western Blotting using the corresponding antibodies. The cell-surface expression was confirmed by FACS analysis using an antibody directed against the EGF-like domain of ADAM17 (Fig. 1 third column). 3.2. ADAM17 exist as a multimer in the cell membrane To analyze the multimerisation of ADAM17 the above described ADAM17 constructs were applied in co-immunoprecipitation experiments. First, two differently C-terminal tagged full-length ADAM17 variants were co-transfected into HEK293 cells. As shown in Fig. 2A the PC-tagged full-length ADAM17 was able to co-immunoprecipitate the HA-tagged ADAM17, demonstrating that ADAM17 exists as a multimer. To clarify whether the cytoplasmic- and/or the transmembraneregion of two molecules are involved in the observed multimerisation, the HA-tagged full-length ADAM17 was co-transfected either with ADAM17 lacking the cytoplasmic- (Fig. 2B) or lacking the cytoplasmic- and the transmembrane-region (Fig. 2C), both labeled with a PC-tag. The pull down of the PC-tagged variants precipitated in both cases the full-length HA-tagged ADAM17, implying that at least one cytoplasmic- and transmembrane region is dispensable for the formation of the multimers. The minimal version of a multimeric ADAM17 is a dimer. To study which one of the extracellular domains is responsible for multimerisation HEK293 cells were co-transfected with full-length FLAG-tagged ADAM17 together with DN-ADAM17, the variant missing the catalytic domain. As presented in Fig. 2D, at least one catalytic domain is dispensable for multimerisation, because DN-ADAM17 precipitates full-length ADAM17. To exclude mutual effects resulting from the transmembrane-/intracellular-region, DN-ADAM17 constructs without the cytoplasmic- (DNADAM17Dcyto) or without the transmembrane- and cytoplasmic-region (DN-ADAM17-GPI) were used (Fig. 2E and F). In both experiments the deletion variants were precipitated with the full-length FLAG-tagged molecule. In a next step we examined whether the disintegrin- or the EGFlike domain or both are needed for the multimerisation. Therefore DN-ADAM17DDis and DN-ADAM17DDis-GPI, both lacking the catalytic and the disintegrin domain, were used for co-transfection with the wild-type molecule. The first construct, DN-ADAM17DDis, contain the whole cytoplasmic- and transmembrane-region whereas the second construct, DN-ADAM17DDis-GPI, is missing these regions (Fig. 2G and H). Both constructs were able to precipitate full-length ADAM17, implying that the sole EGF-like domain is the minimal requirement needed for multimerisation of these molecules. The next question we asked is which domain/s is/are needed on the second molecule. To this end, cells were co-transfected with two different tagged identical deletion variants. In the first experiment two DN-ADAM17 variants without the cytoplasmic-region (DN-ADAM17Dcyto) either tagged with a HA- or a PC-tag were transfected. The PC-tagged protein was precipitated from the lysates using HPC4 antibodies and the HA-tagged variant was co-precipitated as shown in Fig. 3A. To exclude the possibility that
I. Lorenzen et al. / Biochemical and Biophysical Research Communications 415 (2011) 330–336
333
Fig. 1. Variants of ADAM17. The first column present a schematic illustration of ADAM17 constructs. (A) Present the full-length ADAM17, followed by DN-ADAM17 (B), DNADAM17DDis (C), ADAM17Dcyto (D), DN-ADAM17Dcyto (E), ADAM17-GPI (F), DN-ADAM17-GPI (G) and DN-ADAM17DDis-GPI (H). The expression analyses by Western blot and FACS analyses are presented in the second and third column. For Western blot analysis HEK293 cells were harvested and lysed 3 days after transfection and the expression was detected using corresponding antibodies (right lane presents the sample of interest and the left lane an unrelated sample as negative control). Using A300E as primary antibody the cell surface expression was monitored by FACS analysis.
334
I. Lorenzen et al. / Biochemical and Biophysical Research Communications 415 (2011) 330–336
Fig. 2. ADAM17 has the property to multimerise. HEK293 cells were co-transfected with either HA- or FLAG-tagged full-length ADAM17 and a following variants: PC-tagged full-length ADAM17 (A), PC-tagged ADAM17Dcyto (B), PC-tagged ADAM17-GPI (C), myc-tagged DN-ADAM17 (D), PC-tagged DN-ADAM17Dcyto (E), PC-tagged DN-ADAM17GPI (F), myc-tagged DN-ADAM17DDis (G) and PC-tagged DN-ADAM17DDis-GPI (H). Three days later cells were harvested, lysed and subjected to co-immunoprecipitation experiments. The presents of ADAM17 constructs was confirmed by Western blotting using the corresponding antibodies.
I. Lorenzen et al. / Biochemical and Biophysical Research Communications 415 (2011) 330–336
335
mice [20], which show a reduction of the ADAM17 mRNA level of more than 95%, were co-transfected with the HA- and PC-tagged DN-ADAM17DDis-GPI constructs. Consistently with the above described experiments, the PC tagged variant of DN-ADAM17DDis-GPI was able to precipitate the HA-tagged variant. This result confirms that the sole EGF-like domain is indeed responsible for multimerisation of ADAM17 and that multimerisation is mediated by a minimum of two of these domains. 4. Discussion
Fig. 3. Multimerisation of ADAM17 is mediated by its EGF-like domain. HEK293 cells were co-transfected using the same ADAM17 variants either tagged with a HAor a PC-tag: DN-ADAM17Dcyto (A), DN-ADAM17-GPI (B) and DN-ADAM17DDisGPI (C). The cells were harvested and lysed 3 days after transfection. After coimmunoprecipitation the samples were analyzed by Western blotting using HPC4 and anti-HA (C29F4). (D) MefADAM17ex/ex were transfected with the HA- and PCtagged DN-ADAM17DDis-GPI constructs and 3 days later subjected to co-immunoprecipitation experiments.
the multimerisation is mediated by the transmembrane region we performed the same experiment using GPI anchored DN-ADAM17 (DN-ADAM17-GPI) containing only the disintegrin- and EGF-like domain. Again the HA-tagged variant was detectable by precipitating the PC-tagged variant (Fig. 3B). To distinguish whether multimerisation is mediated via an EGF-like/disintegrin or an EGF-like/EGF-like interaction cells were co-transfected with the HA- and PC-tagged DN-ADAM17DDis-GPI variants. Both constructs consists solely of the EGF-like domain and are anchored by GPI in the membrane. As shown in Fig. 3C the PC-tagged variant was able to precipitate the HA-tagged DN-ADAM17DDis-GPI. From these experiments we conclude that the EGF-like domain alone is responsible for multimerisation of ADAM17. In order to confirm the responsibility of the EGF-like domain for the observed multimerisation we aimed to perform the same experiment in cells without ADAM17 expression. HEK293 cells express endogenous ADAM17 at moderate levels, therefore mouse embryonic fibroblasts (Mef) generated from ADAM17ex/ex (MefADAM17ex/ex)
The aim of this work was to gain insight in the multimerisation properties of the ADAM17. By co-immunoprecipitation experiments we demonstrate that ADAM17 has the potential to form multimers. Multimerisation of transmembrane proteases of the metzincins superfamily was described for meprin A and B as well as for MT1-MMP [15,16,18]. In case of MT1-MMP the minimal formation of a dimer is needed for its biological activity [18]. Mutations within the interaction site of the domain that mediates dimerisation, namely the hemopexin-domain, abrogates the ability of proMMP2 activation and collagen degradation [18]. The data presented here do not allow a conclusion about the number of molecules which form the ADAM17 multimer, but demonstrate that multimerisation is mediated by only one domain, the EGF-like domain. Additionally to its function in multimerisation, the EGF-like domain is involved in substrate recognition [9,11] and enzyme activation by facilitating the removal of the inhibitory pro-peptide from the catalytic domain [10]. The use of deletion variants clearly demonstrates that the transmembrane- as well as the cytoplasmicregion of ADAM17 is dispensable for this process. According to this, the intracellular region of ADAM17 is not involved in the generation of multimers, but it cannot be excluded that it takes part in the separation of ADAM17 multimers due to the induction of a conformational change, like it is described for the inside-out signaling of integrins [23]. The inside-out signaling transformed the low-affinity state of integrin heterodimers to a high-affinity state by the separation of the intracellular parts in response to an extra-cellular stimulus. The GPI anchored ADAM17 variant is unable to shed its substrates although it contains the complete ectodomain [12]. This effect is not due to the loss of ADAM17 multimers, in accordance with the presented data. Therefore it should be addressed whether multimerisation of ADAM17 is mandatory for shedding of its substrates, or whether multimers constitute an inactive state. In such a scenario ADAM17 has to become monomeric to be active implicating a conformational change that additional support the release of the pro-peptide from the catalytic active site. Acknowledgments The authors would like to thank Christoph Garbers for the full-length murine ADAM17 and murine GPI anchored construct and Athena Chalaris for the DN-ADAM17 and MefADAM17ex/ex. This study has been supported by the Deutsche Forschungsgemeinschaft (SFB 877, A6, Z3) and the Excellence Cluster ‘Inflammation at Interfaces’. References [1] J. Scheller, A. Chalaris, C. Garbers, S. Rose-John, ADAM17: a molecular switch to control inflammation and tissue regeneration, Trends Immunol. 32 (2011) 380–387. [2] R.A. Black, C.T. Rauch, C.J. Kozlosky, J.J. Peschon, J.L. Slack, M.F. Wolfson, B.J. Castner, K.L. Stocking, P. Reddy, S. Srinivasan, N. Nelson, N. Boiani, K.A. Schooley, M. Gerhart, R. Davis, J.N. Fitzner, R.S. Johnson, R.J. Paxton, C.J. March, D.P. Cerretti, A metalloproteinase disintegrin that releases tumour-necrosis factor-alpha from cells, Nature 385 (1997) 729–733.
336
I. Lorenzen et al. / Biochemical and Biophysical Research Communications 415 (2011) 330–336
[3] M.L. Moss, S.L. Jin, M.E. Milla, D.M. Bickett, W. Burkhart, H.L. Carter, W.J. Chen, W.C. Clay, J.R. Didsbury, D. Hassler, C.R. Hoffman, T.A. Kost, M.H. Lambert, M.A. Leesnitzer, P. McCauley, G. McGeehan, J. Mitchell, M. Moyer, G. Pahel, W. Rocque, L.K. Overton, F. Schoenen, T. Seaton, J.L. Su, J.D. Becherer, et al., Cloning of a disintegrin metalloproteinase that processes precursor tumour-necrosis factor-alpha, Nature 385 (1997) 733–736. [4] J. Müllberg, E. Dittrich, L. Graeve, C. Gerhartz, K. Yasukawa, T. Taga, T. Kishimoto, P.C. Heinrich, S. Rose-John, Differential shedding of the two subunits of the interleukin-6 receptor, FEBS Lett. 332 (1993) 174–178. [5] J.J. Peschon, J.L. Slack, P. Reddy, K.L. Stocking, S.W. Sunnarborg, D.C. Lee, W.E. Russell, B.J. Castner, R.S. Johnson, J.N. Fitzner, R.W. Boyce, N. Nelson, C.J. Kozlosky, M.F. Wolfson, C.T. Rauch, D.P. Cerretti, R.J. Paxton, C.J. March, R.A. Black, An essential role for ectodomain shedding in mammalian development, Science 282 (1998) 1281–1284. [6] S. Takeda, Three-dimensional domain architecture of the ADAM family proteinases, Semin. Cell Dev. Biol. 20 (2009) 146–152. [7] J.M. White, ADAMs: modulators of cell–cell and cell–matrix interactions, Curr. Opin. Cell Biol. 15 (2003) 598–606. [8] D.V. Bax, A.J. Messent, J. Tart, M. van Hoang, J. Kott, R.A. Maciewicz, M.J. Humphries, Integrin alpha5beta1 and ADAM-17 interact in vitro and colocalize in migrating HeLa cells, J. Biol. Chem. 279 (2004) 22377–22386. [9] P.W. Janes, N. Saha, W.A. Barton, M.V. Kolev, S.H. Wimmer-Kleikamp, E. Nievergall, C.P. Blobel, J.P. Himanen, M. Lackmann, D.B. Nikolov, Adam meets Eph: an ADAM substrate recognition module acts as a molecular switch for ephrin cleavage in trans, Cell 123 (2005) 291–304. [10] M.E. Milla, M.A. Leesnitzer, M.L. Moss, W.C. Clay, H.L. Carter, A.B. Miller, J.L. Su, M.H. Lambert, D.H. Willard, D.M. Sheeley, T.A. Kost, W. Burkhart, M. Moyer, R.K. Blackburn, G.L. Pahel, J.L. Mitchell, C.R. Hoffman, J.D. Becherer, Specific sequence elements are required for the expression of functional tumor necrosis factor-alpha-converting enzyme (TACE), J. Biol. Chem. 274 (1999) 30563–30570. [11] P. Reddy, J.L. Slack, R. Davis, D.P. Cerretti, C.J. Kozlosky, R.A. Blanton, D. Shows, J.J. Peschon, R.A. Black, Functional analysis of the domain structure of tumor necrosis factor-alpha converting enzyme, J. Biol. Chem. 275 (2000) 14608– 14614. [12] X. Li, L. Pérez, Z. Pan, H. Fan, The transmembrane domain of TACE regulates protein ectodomain shedding, Cell Res. 17 (2007) 985–998.
[13] S.M. Soond, B. Everson, D.W. Riches, G. Murphy, ERK-mediated phosphorylation of Thr735 in TNFalpha-converting enzyme and its potential role in TACE protein trafficking, J. Cell Sci. 118 (2005) 2371–2380. [14] P. Xu, R. Derynck, Direct activation of TACE-mediated ectodomain shedding by p38 MAP kinase regulates EGF receptor-dependent cell proliferation, Mol. Cell 37 (2010) 551–566. [15] E.E. Sterchi, W. Stocker, J.S. Bond, Meprins, membrane-bound and secreted astacin metalloproteinases, Mol. Aspects Med. 29 (2008) 309–328. [16] P. Marchand, J. Tang, J.S. Bond, Membrane association and oligomeric organization of the alpha and beta subunits of mouse meprin A, J. Biol. Chem. 269 (1994) 15388–15393. [17] G.P. Bertenshaw, M.T. Norcum, J.S. Bond, Structure of homo- and heterooligomeric meprin metalloproteases, dimers, tetramers, and high molecular mass multimers, J. Biol. Chem. 278 (2003) 2522–2532. [18] A. Tochowicz, P. Goettig, R. Evans, R. Visse, Y. Shitomi, R. Palmisano, N. Ito, K. Richter, K. Maskos, D. Franke, D. Svergun, H. Nagase, W. Bode, Y. Itoh, The dimer interface of the membrane type 1 matrix metalloproteinase hemopexin domain: crystal structure and biological functions, J. Biol. Chem. 286 (2011) 7587–7600. [19] K.A. Solomon, N. Pesti, G. Wu, R.C. Newton, Cutting edge: a dominant negative form of TNF-alpha converting enzyme inhibits proTNF and TNFRII secretion, J. Immunol. 163 (1999) 4105–4108. [20] A. Chalaris, N. Adam, C. Sina, P. Rosenstiel, J. Lehmann-Koch, P. Schirmacher, D. Hartmann, J. Cichy, O. Gavrilova, S. Schreiber, T. Jostock, V. Matthews, R. Hasler, C. Becker, M.F. Neurath, K. Reiss, P. Saftig, J. Scheller, S. Rose-John, Critical role of the disintegrin metalloprotease ADAM17 for intestinal inflammation and regeneration in mice, J. Exp. Med. 207 (2010) 1617–1624. [21] A. Trad, N. Hedemann, M. Shomali, V. Pawlak, J. Grötzinger, I. Lorenzen, Development of sandwich ELISA for detection and quantification of human and murine a disintegrin and metalloproteinase17, J. Immunol. Methods 371 (2011) 91–96. [22] A. Chalaris, B. Rabe, K. Paliga, H. Lange, T. Laskay, C.A. Fielding, S.A. Jones, S. Rose-John, J. Scheller, Apoptosis is a natural stimulus of IL6R shedding and contributes to the pro-inflammatory trans-signaling function of neutrophils, Blood 110 (2007) 1748–1755. [23] C.L. Abram, C.A. Lowell, The ins and outs of leukocyte integrin signaling, Annu. Rev. Immunol. 27 (2009) 339–362.