- Email: [email protected]
B cell antigen receptor assembly and Syk activation in the S2 cell reconstitution system
Immunology Letters 92 (2004) 67–73
B cell antigen receptor assembly and Syk activation in the S2 cell reconstitution system Thomas Wossning, Michael Reth∗ a
Department of Molecular Immunology, Biology III, University of Freiburg, Stübeweg 51, 79108 Freiburg, Germany b Max-Planck-Institute for Immunobiology, Freiburg, Germany Received 17 September 2003; accepted 15 October 2003
Abstract Signal transduction from the B cell antigen receptor (BCR) involves a multitude of signaling molecules often organized in dynamic protein complexes. The molecular mechanisms operating during signaling are difficult to study solely by loss-of-function analysis. For a better understanding of the transient interaction of signaling molecules and their regulation by feedback loops, as well as their dynamic behavior in living cells, new techniques are required. We have developed a method allowing the reconstitution of the BCR complex and several of its key signaling elements in the evolutionary distant environment of the Drosophila S2 Schneider cell line. With this gain-of-function approach, we study here the assembly of the BCR complex and the control of its transport to the cell surface of S2 cells. We find that without binding to a light chain, the membrane-bound m heavy chain (mHC) homodimer, together with the Ig-␣/Ig- heterodimer, can come to the cell surface where it is signaling competent. This finding could have implications for potential signaling functions of such a receptor molecule during pro-/pre-B cell development. We also studied the activation of the BCR-proximal kinase Syk. We found that a truncated Syk mutant lacking the first (N-terminal) SH2 domain and the linker regions, is still regulated by autoinhibition and can only become activated in the presence of the BCR. This indicates that the C-terminal SH2 domain of Syk is the dominant regulatory subunit of this kinase. © 2004 Published by Elsevier B.V. Keywords: Drosophila S2 cells; BCR assembly; Syk activation
1. Introduction Mature B lymphocytes recognize and react to antigen via their B cell antigen receptor (BCR). This multiprotein complex includes the membrane-bound immunoglobulin (mIg) and the Ig-␣/Ig- heterodimer, which mediate antigen-binding and signal transduction, respectively. The mIg molecule is a symmetric disulfide-linked structure of two identical heavy (HCs) and two identical light chains (LCs). The Ig-␣/Ig- heterodimer is associated non-covalently with the mIg molecule in a complex with a 1:1 stoichiometry. This complex is stable in 1% digitonin lysates and results with other detergents like 0.2% NP40 or 0.1% thesit suggest that on the B cell surface, the BCR has an oligomeric structure [1,2]. The folding and assembly of BCR subunits in the endoplasmic reticulum (ER) is supported by ER-resident chap∗ Corresponding author. Tel.: +49-761-5108-420; fax: +49-761-5108-423. E-mail address: [email protected] (M. Reth).
0165-2478/$ – see front matter © 2004 Published by Elsevier B.V. doi:10.1016/j.imlet.2003.10.015
erones. For example, the HC-binding protein (BiP) binds to exposed hydrophobic sequences on the V and C domains of unpaired HC and LC [3,4]. Upon assembly, these chains loose their interaction with BiP thus allowing export of Ig from the ER [5,6]. In addition, the mIg molecule requires the assembly with the Ig-␣/Ig- heterodimer for the efficient transport to the cell surface [1,7]. The mIg molecules are retained in the ER by membrane-associated chaperons like calnexin, calreticulin and BAP29/BAP31 (BCR associated protein) heterodimers [8,9]. However, the requirements for the assembly and export of mIg molecules are less well studied than those for the secreted Ig molecules. The cytoplasmic tails of Ig-␣ and Ig- carry two tyrosines within the amino acid (aa) consensus sequence motif (D/E)x7 (D/E)xxYxx(L/I)x6−7 Yxx(L/I) known as immunoreceptor tyrosine-based activation motif (ITAM) [10]. These two tyrosines play an essential role in the activation of the BCR proximal protein tyrosine kinase (PTK) Syk [11,12]. At its N-terminus, Syk carries two tandem SH2 domains (SH2N and SH2C) interrupted by a flexible linker named interdomain A [13–15]. The two SH2 domains play
68
T. Wossning, M. Reth / Immunology Letters 92 (2004) 67–73
an important role in the regulation of Syk kinase activity. In the resting state the SH2 domains inhibit the kinase activity of Syk presumably by blocking access of ATP to the kinase domain [16,17]. Many signaling elements display similar intramolecular domain interactions that result in autoinhibition. When bound to the BCR, Syk can assume an open conformation and phosphorylate the two ITAM tyrosines. The tandem SH2 domains of Syk then bind to the doubly phosphorylated ITAM tyrosines (ppITAM), thus fixing the kinase in an open and active conformation [16,18]. This results in rapid phosphorylation of neighboring ITAM sequences, further Syk recruitment and the amplification of the BCR signal by a positive enzyme/product (Syk/ITAM) feedback [17]. However, this process is efficiently interrupted by BCR proximal protein tyrosine phosphatases (PTP) like SHP-1 [17]. Presumably the engagement of the BCR by antigen results not only in PTK activation but more importantly in PTP inhibition, which allows amplification of the BCR-signal. Fully activated Syk phosphorylates several cytosolic substrates, thereby leading to the activation of downstream signaling pathways including the mitogen-activated protein kinase (MAPK) and the NFKB pathway [19–21]. A prominent substrate of Syk is the adaptor protein SLP-65 (SH2 domain-containing leukocyte protein of 65 kDa), also known as BLNK (B cell linker protein) and BASH (B cell adaptor protein carrying an SH2 domain) [22–25]. Phosphorylated SLP-65 organizes a signaling complex that mediates the opening of intracellular Ca2+ stores in activated B cells [26–28]. The Src-family kinase Lyn also interacts with the BCR. However, Lyn phosphorylates dominantly only the first ITAM tyrosine and presumably supports the activity of Syk primarily via PTP inhibition rather than ppITAM phosphorylation [17]. The interactions of the BCR with its signaling elements are often transient and do not survive cellular lysis thus making it difficult to study these interactions by biochemical means. Therefore, in recent years, signaling systems have been most successfully studied by genetic means such as knock out techniques and mutational screens [29,30]. With these loss-of-function approaches, a signaling element can be attributed to a certain signaling pathway. However, the molecular details of a signaling process and its regulation by critical feedback cannot be elucidated by such loss-of-function analyses. A method to circumvent this problem is the reconstitution of signaling pathways in a different cell type as it has been shown for TCR and FcεRI signaling elements in non-lymphoid cell lines [31,32]. However, it is often impossible to express signaling elements in vertebrate cells without interfering with the growth and survival of the transfected cells. To solve this problem we have established a method allowing the transient and inducible coexpression of many mammalian genes in the genetically distant environment of S2 Schneider cells from Drosophila melanogaster [17,33] (Fig. 1). Here we use this system to study the molecular requirements for expression of the m
Fig. 1. The S2 signaling pathway reconstitution system allows the transient and inducible expression of 10–20 mammalian genes in the evolutionary distant environment of Drosophila Schneider cells.
heavy chain (mHC) on the cell surface. Furthermore we use mutations and truncations of the tandem SH2 domains of Syk to study the mechanism of Syk kinase activation. 2. Materials and methods 2.1. DNA constructs For inducible expression in S2, cells cDNAs of BCR components and signaling elements were each separately cloned into the S2 plasmid pRmHa-3 containing a copper-inducible metallothionein promoter [34,35]. For reconstitution of the IgD-BCR, we used vectors encoding an N-terminally flag-tagged Ig-a (pDflmb-1), HA-tagged Ig-b (pDHAB29) and a single chain ␦mHC (␦m) in which the VH and CH 1 domains are replaced by a VL /VH single chain [8,17]. Thus no separate LC is required to rebuild this BCR, here referred to as scIgD-BCR. For reconstitution of the IgM-BCR in S2 cells, we used PCR to amplify the mHC and the 1LC cDNAs from the m 3–11 myeloma cell line and cloned the PCR products into the S2 plasmid pRmHa-3 to generate the expression vectors pDm and pD1, respectively. The coexpression of these vectors with pDflmb-1 and pDHAB29 results in the expression of fully assembled IgM-BCR complexes. For reconstitution of BCR signaling pathways, we used vectors encoding human Syk (R. Scheuermann, University of Texas, Dallas; pDhSyk) and murine SLP-65 (pDSLP-65) [17,24]. Syk mutants were expressed using vectors pDSykmSH2N, pDSykmSH2C and pDSykSH2N coding for human Syk bearing a RQS/GGI mutation within the N-terminal SH2 domain, a RAR/GAL mutation within the C-terminal SH2 domain and a deletion of the N-terminal SH2 domain, respectively [17]. For construction of a plasmid encoding a truncated Syk that lacks both the N-terminal SH2 domain and interdomain A (pDDDSyk) amplified by PCR a Syk fragment using the oligos SykSH2Nlinker+ (aaaaaccggttggttccatggaaaaatctctcgg) and pRmHa-3-mcs-rev (tctagaggatccccgggtacc). The PCR product was cut with AgeI/BamHI and ligated into pDSykSH2N cut with the same enzymes, thereby yielding pDSyk.
T. Wossning, M. Reth / Immunology Letters 92 (2004) 67–73
69
2.2. Cell culture, transient transfection and FACS analysis Drosophila S2 Schneider were obtained as a generous gift from K. Karjalainen, Belinzona Institute for Immunology, Switzerland. They were grown in Schneider’s Drosophila medium (Life Technologies Inc.) supplemented with 10% fetal calf serum (FCS) at 27 ◦ C in a water-saturated atmosphere with atmospheric CO2 levels. For transient transfection, 2 × 106 cells from an exponentially growing culture were spread in six-well plates and cultured for 24 h. An amount of 0.3 g of each plasmid were diluted in 100 l serum-free medium and the amount of DNA was equalized with the empty vector pRmHa-3. An amount of 100 l serum-free medium containing 10 l CellFectin (Life Technologies Inc.) was added and the sample was incubated at RT for 10–15 min. The mixture was adjusted to 1 ml with serum-free medium and added to cells previously washed with serum-free medium. Following 18 h incubation, the solution was replaced with 3 ml medium and the cells were cultured for another 24 h. For induction of gene expression, CuSO4 was added to a final concentration of 1 mM for 24 h. For FACS analysis, cells were harvested using a cell scraper, washed with PBS and stained with anti-IgM ( chain specific, CyTM -5 conjugated, Jackson). To check for transfection efficiency and to recognize transfected cells, we cotransfected the S2 cells with the vector pDEGFP containing the EGFP coding sequence [17]. 2.3. Preparation of cell lysates and immunoblotting CuSO4 -induced cells were harvested using a cell scraper, washed with PBS and lysed in 0.5 ml Triton X-100 lysis buffer (50 mM Tris–HCl (pH 8.0), 140 mM NaCl, 0.5 mM EDTA (pH 8.0), 1 mM Na3 VO4 , 10% (v/v) glycerol, 1% (v/v) Triton X-100 and 10% (v/v) protease inhibitor cocktail (Sigma)). Protein concentrations in postnuclear supernatants were determined with protein assay dye reagent (Bio-Rad). Standardized lysates were boiled in 1× Laemmli buffer [36], separated by SDS-PAGE and analyzed by immunoblotting using the enhanced chemoluminescence detection system (ECL, Amersham). The antibodies used were anti-Syk (4D10, Santa Cruz), anti-SLP-65 [24], anti-flag (M2, Sigma), anti-HA (rat, Roche), anti-phosphotyrosine (4G10, UBI), anti-IgM ( chain specific, peroxidase conjugated, Pierce) and anti- (peroxidase conjugated, Southern).
3. Results 3.1. Requirements for mIgM expression on the surface of Drosophila S2 cells To investigate the requirements for the expression of a mHC on the cell surface we cloned cDNAs coding for murine mHC and 1LC each separately into the
Fig. 2. FACScan analysis of the expression of mHC on the surface of S2 cells transiently cotransfected with plasmids coding for mHC, Syk, SLP-65, EGFP in different combinations with plasmids for 1LC (indicated by − and +), Ig-␣ and/or Ig-. Transfected (EGFP+ ) cells were analyzed for mHC expression on the cell surface 24 h after induction of protein production. Numbers indicate the percentage of transfected cells expressing mHC on the surface.
vector pRmHa-3 containing a copper-inducible metallothionein promoter [34,35,37]. S2 cells were cotransfected with different combinations of plasmids encoding m, 1, N-terminally flag-tagged Ig-␣ (flag-Ig-␣) and HA-tagged Ig- (HA-Ig-) [8,17]. To investigate signal transduction through the IgM-BCR, we cotransfected vectors encoding Syk and SLP-65 as well as EGFP to identify transfected cells by FACS analysis [17]. The expression of EGFP and that of m on the cell surface were examined by FACS analysis 24 h after induction of gene expression with copper sulfate (Fig. 2). In a first series of experiments, we expressed the mHC without a LC either alone or in combination with Ig-␣ and/or Ig- (Fig. 2, upper panels). The efficiency of the transient transfection was controlled by the cotransfection of EGFP and was expressed in roughly 30% of the S2 cells. When the mHC was expressed in the absence of any other BCR components, only 2% of the transfected (EGFP+ ) S2 cells carry low amounts of m on the cell surface demonstrating that most of the unassembled m is retained in the ER (Fig. 2, upper left panel). The same is true for cells coexpressing m with Ig-␣ indicating that Ig-␣ alone is not able to overcome the ER retention of the mHC in S2 cells. In contrast to these findings, coexpression of m together with Ig- leads to a low expression of m on the surface of 42% of the transfected S2 cells. This indicates that Ig-, presumably in the form of an Ig-/Ig- homodimer, can bind to and overcome the ER retention of the mHC. Evidence for such a homodimer has been reported previously [7]. The coexpression of m, Ig-␣ and Ig- resulted in strong expression on the surface of 86% of the transfected cells. Thus the Ig-␣/Ig- heterodimer is more efficient than the Ig-/Ig- homodimer in the binding and transport of the mHC onto the surface of S2 cells (Fig. 2, upper right panel). The same analysis of mHC expression was repeated with cells coexpressing a 1LC (Fig. 2, lower panels). Low amounts of mIgM are found on the surface of 13% of the
70
T. Wossning, M. Reth / Immunology Letters 92 (2004) 67–73
transfected (EGFP+ ) S2 cells coexpressing the mHC and 1LC. This percentage is only slightly increased by the coexpression of Ig-␣, whereas Ig- is more efficient to promote the transport of the mIgM molecule in 75% of transfected S2 cells. The highest mHC expression is achieved by the coexpression of all BCR components and most (98%) of the EGFP+ transfected S2 cells carry the completely assembled IgM-BCR on their cell surface. These data confirm the unusually high cotransfection rate of S2 cells, which ensures that most transfected S2 cells express protein from all the vectors used. The S2 transfectants where also analyzed by Western blotting of total cellular lysates using antibodies specific for the BCR components, the signaling elements Syk and SLP-65 as well as for phosphotyrosine (pY) (Fig. 3). The mHC is expressed in similar amounts in all S2 transfectants independent of its coexpression with other BCR components (Fig. 3, third panel). In contrast, the Ig-␣ and Ig- components are more prominently expressed in cells allowing the assembly of the Ig-␣/Ig- heterodimer together with either the mHC or the mIgM molecule (Fig. 3, fifth and sixth panels, lanes 8 and 9). This suggests that in S2 cells, the mHC is a more stable protein than the Ig-␣/Ig- heterodimer and that the assembly with the mIg molecule can stabilize the Ig-␣/Ig- heterodimer and prevent its degradation. This notion is supported by a pulse/chase experiment in which S2 cells express either Ig-␣ alone, the Ig-␣/Ig- heterodimer or a scBCR containing a sc␦m in which the VH and CH 1 domains are replaced by an VL /VH single chain. Twenty four hours after induction of gene expression the cells were washed and cultured without further expression of the transfected vectors (Fig. 3b). After a 12 h chase period, no Ig-␣ protein is detected in S2 cell transfected with only the Ig-␣ or the Ig-␣ and Ig- vectors (Fig. 3b, first and second panels, lane 3). The S2 cells carrying the vectors for all three scBCR components, however, still contain some of the mature (lower molecular weight) form of the Ig-␣ protein 24 h after the start of the chase period (Fig. 3b, third panel, lane 4). Similar to the mHC, the sc␦m is more stable and can be detected after a 74 h chase period (Fig. 3b, fourth panel, lane 6). We previously have shown that coexpression of the kinase Syk with its in vivo substrate, the adaptor protein SLP-65, does not result in SLP-65 phosphorylation unless the cells coexpress an ITAM containing protein like Ig-␣ or Ig- [17]. Similarly, coexpression of Syk and SLP-65 together with either the mHC or the mIgM molecule does not result in tyrosine phosphorylation of SLP-65 (Fig. 3a, first panel, lanes 1 and 2). However, coexpression of these Ig molecules with Ig-␣ and/or Ig- leads to Syk activation and SLP-65 phosphorylation (Fig. 3, first panel). The most efficient Syk activation occurred when all components of the IgM-BCR were coexpressed (Fig. 3, lane 9). This may be due to the fact that the BCR complex carries several ITAM containing proteins which can facilitate the Syk/ITAM signal amplification at the receptor.
Fig. 3. Western blot analysis of tyrosine phosphorylation and protein production in S2 cells. (a) Cells were transiently transfected with plasmids coding for mHC, Syk, SLP-65, EGFP and different combinations of 1LC, Ig-␣ and/or Ig-. Twenty four hours after induction cells were lysed in Triton X-100 lysis buffer. The total cellular lysates were separated by SDS-PAGE and analyzed by immunoblotting using the indicated antibodies. (b) Stability of the BCR components tested in a pulse/chase experiment. S2 cells transiently transfected with plasmids coding for Ig-␣ (first panel), Ig-␣ and Ig- (second panel) as well as sc␦m, Ig-␣ and Ig- (third and fourth panels) were induced for 24 h and then washed and further cultured for the indicated time.
3.2. Regulation of Syk kinase activity by its SH2 domains A key to understanding of signal initiation at the BCR is the regulation of Syk kinase activity by the ITAM. It has been shown that Syk binds to ppITAMs via its tandem SH2
T. Wossning, M. Reth / Immunology Letters 92 (2004) 67–73
71
Fig. 4. Regulation of Syk kinase activity by its SH2 domains. (A) Schematic representation of wild type Syk and the analyzed Syk mutants. (B) Immunoblot analysis of tyrosine phosphorylation and protein production in S2 cells transiently transfected with plasmids coding for SLP-65 and either Syk wt or the Syk mutants, SykmSH2C, SykmSH2N, SykSH2N, and Syk () in the absence (−) or presence (+) of the scBCR. 24 h after induction cells were lysed in Triton X-100 lysis buffer and total lysates were separated by SDS-PAGE and analyzed by immunoblotting using the antibodies indicated. Arrows with an asterisk indicate the position of Syk wt, SykmSH2N and SykmSH2C. (C) Model of the autoinhibition structure of Syk wt and the Syk truncation mutant.
domains and that both domains are required for propagation of downstream signaling [16,18,38,39]. Deletion of the tandem SH2 domains generates a constitutively active enzyme with low substrate specificity [17]. Therefore, it has been suggested that Syk is autoinhibited by the intramolecular binding of the tandem SH2 domains to the kinase domain [39]. Since it has been shown that each individual SH2 domain exhibits different binding properties for the ppITAM sequence, we asked whether each domain contributes equally to kinase inhibition [17,38]. For this purpose, we generated different Syk mutants bearing either a RQS/GGI aa exchange within the N-terminal SH2 domain (SykmSH2N) or a RAR/GAL aa exchange in the C-terminal SH2 domain (SykmSH2C). We also constructed vectors for Syk deletion mutants lacking the N-terminal SH2 domain (SykSH2N) or both the N-terminal SH2 domain and the interdomain A (Syk) (Fig. 4A). Either wild type (wt) or mutant (m) Syk were expressed in S2 cells, together with SLP-65 alone or with SLP-65 and the scBCR (Fig. 4B). The Western blot analysis of total cellular lysates revealed almost no SLP-65 phosphorylation in cells lacking the scBCR demonstrating that the wt and mutant Syk are still autoinhibited (Fig. 4B, lanes 1, 3, 5, 7 and 9). In contrast, coexpression of Syk wt and SLP-65 in the presence of a BCR leads to strong phosphorylation of Ig-␣ and SLP-65 showing that the scBCR is able to induce Syk kinase activity (Fig. 4B, lane 2). As described previously [17], SykmSH2C does not become active in the presence of the scBCR (Fig. 4B, lane
4). In contrast, the SykmSH2N mutant is still activated by the scBCR albeit less strongly (Fig. 4B, lane 6). Together, these data indicate the SH2C domain of Syk is the dominant structure in the binding of ppITAM and the amplification of the BCR signal. However, the SH2N domain has some contribution to the autoinhibition of Syk since its deletion either alone or together with the interdomain A results in a stronger Syk activation than the mutation of the SykSH2N binding site (Fig. 4B, lanes 8 and 10). This also demonstrates that the autoinhibition of Syk requires the presence but not the phosphotyrosyl-binding capacity of the Syk SH2 domains (Fig. 4C).
4. Discussion 4.1. Assembly of the IgM-BCR In B cells, the export of BCR subunits from the ER is tightly controlled, thus allowing surface expression of only completely assembled BCRs [3,5–7,40,41]. Single or partially assembled BCR components are retained in the ER and this retention relies on at least two different mechanisms: BiP binding to hydrophobic patches on unpaired Ig domains, and binding of the BAP29/BAP31 heterodimer to conserved hydrophilic aa within the HC transmembrane region [3,4,8,9]. Both BiP and the BAP proteins are ubiquitously expressed and highly conserved [9]. Drosophila cells express BiP but there is only one BAP protein, namely BAP31 expressed
72
T. Wossning, M. Reth / Immunology Letters 92 (2004) 67–73
in this species. Nevertheless, it can be assumed that similar ER retention mechanisms for the BCR components are operating in S2 cells [9]. Indeed, the unpaired mHC chain is retained in the ER of S2 cells. This retention can be partially released by coexpression of either the 1LC or the Ig-␣/Ig- heterodimer, directly demonstrating the existence of the two different retention mechanisms mentioned above. Thus, it is likely that the association of endogenous BAP31 with the m transmembrane region and binding of BiP to the CH 1 domains of mHC dimer is involved in the retention of the mHC in the ER of S2 cells. However, in contrast to B cells, this retention is leaky in S2 cells since m surface expression is also detectable in the absence of one or more receptor subunits. For example, a partially assembled mIg/Ig- complex is expressed in detectable amounts on S2 cells but not on B cells. This leakiness may be due to the absence of BAP29 or other evolutionary alterations which does not allow the chaperone system of Drosophila to interact as efficiently with the BCR components as their murine counterparts. The amount of BCR complexes residing on the cell surface is not only determined by the rate of export but also by its stability and import rate. In this respect, it is feasible that the transport of an mIgM/Ig-/Ig- homodimer complex to the cell surface is as efficient as that of the complete BCR containing the Ig-␣/Ig- heterodimers but the former protein is less stable and thus more rapidly internalized and degraded [7]. This notion can be directly tested in the S2 system by pulse/chase experiments. A partially assembled BCR complex with the strongest expression on the cell surface of S2 cells is the m Ig-␣/Ig- complex and we show here that this complex is signaling competent. Such a complex has not been found so far on murine B cells most of which always express an LC together with a HC. This is different, however, in pre-B cells which have not yet assembled their LC genes. Although it is well known that the pre-BCR on these cells carry a surrogate LC, it remains possible that the m Ig-␣/Ig- complex is an alternative form of the pre-BCR and we have now found some indication of this on normal murine pre-B cells [42]. However, although S2 cells express foreign proteins to an amount comparable that in B cells, it cannot be completely excluded that overexpression of m, Ig-␣ and Ig- serves to circumvent ER retention in the absence of a LC as reported previously [17,43]. 4.2. Regulation of Syk kinase activity Upon antigen stimulation of the BCR, the tandem SH2 domains of Syk to the ppITAMs sequence and Syk becomes activated [17,18,38,39]. In contrast, in resting cells, the enzymatic activity of Syk is inhibited by a mechanism that involves the N-terminal part of the protein including both SH2 domains [17,39]. Thus, the tandem SH2 domains of Syk are not only involved in ITAM binding but also in binding to the kinase domain and the autoinhibition of Syk. Autoinhibition by intramolecular interactions between the SH2 and
kinase domain have been shown to regulate the activity of several other kinases like Src and abl [44]. Furthermore, this mechanism has also been suggested for regulation of the Syk family kinase ZAP70 [45]. In these kinases, formation of the closed conformation relies on intramolecular binding of the SH2 domain to a particular phosphotyrosyl residue. However, our data show that Syk kinase activity is not regulated by an analogous mechanism since the SH2N and SH2C domains do not require their phosphotyrosyl binding capacity for Syk inhibition. In this respect, the regulation of Syk is similar to that of the phosphatase SHP-2 the activity of which is also inhibited by its two tandem SH2 domains independently of their phosphotyrosyl binding capacity [46]. However, whereas the inhibition of SHP-2 mostly involves its SH2N domain, the deletion of the Syk SH2N domain only reduces but does not abolish the autoinhibition of Syk. Furthermore we have shown here that Syk is dominantly regulated by its SH2C domain. In previous studies, it has been reported that interdomain A of Syk shares the sequence Lxx(D/E)Y with two Syk substrate proteins, namely ␣-tubulin and erythrocyte band 3 that are tyrosine phosphorylated at this site [47,48]. This finding led to the suggestion that interdomain A acts as pseudosubstrate and blocks the substrate binding site of Syk thereby mediating the autoinhibition of Syk. However, our study of an interdomain A deletion mutant of Syk shows that this sequence does not play an important role in autoinhibition which mostly relies on SH2C and perhaps the interdomain B. How exactly these domains regulate Syk activity has to await the determination of the crystal structure of the complete Syk protein.
Acknowledgements We thank Christa Kalmbach-Zürn for technical support and Dr. Peter Nielsen for critical reading of the manuscript. We also thank Anna Erdei and János Gergely for organizing a marvelous signaling meeting and for their support throughout these years. This work was supported by the Deutsche Forschungsgesellschaft through SFB388.
References [1] Hombach J, Tsubata T, Leclercq L, Stappert H, Reth M. Nature 1990;343:760–2. [2] Schamel WWA, Reth M. Immunity 2000;13:5–14. [3] Knittler MR, Dirks S, Haas IG. Proc Natl Acad Sci USA 1995;92:1764–8. [4] Knittler MR, Haas IG. EMBO J 1992;11:1573–81. [5] Hendershot L, Bole D, Kohler G, Kearney JF. J Cell Biol 1987;104:761–7. [6] Hendershot LM. J Cell Biol 1990;111:829–37. [7] Brouns GS, de Vries E, Borst J. Int Immunol 1995;7:359–68. [8] Schamel WWA. Ph.D. Thesis, University of Freiburg, Freiburg, 1999. [9] Adachi T, Schamel WW, Kim KM, Watanabe T, Becker B, Nielsen PJ, et al. EMBO J 1996;15:1534–41.
T. Wossning, M. Reth / Immunology Letters 92 (2004) 67–73 [10] Reth M. Nature 1989;338:383–4. [11] Gold MR, Matsuuchi L, Kelly RB, DeFranco AL. Proc Natl Acad Sci USA 1991;88:3436–40. [12] Pao LI, Famiglietti SJ, Cambier JC. J Immunol 1998;160:3305–14. [13] Taniguchi T, Kobayashi T, Kondo J, Takahashi K, Nakamura H, Suzuki J, et al. J Biol Chem 1991;266:15790–6. [14] Muller B, Cooper L, Terhorst C. Immunogenetics 1994;39:359–62. [15] Law CL, Sidorenko SP, Chandran KA, Draves KE, Chan AC, Weiss A, et al. J Biol Chem 1994;269:12310–9. [16] Shiue L, Green J, Green OM, Karas JL, Morgenstern JP, Ram MK, et al. Mol Cell Biol 1995;15:272–81. [17] Rolli V, Gallwitz M, Wossning T, Flemming A, Schamel WW, Zurn C, et al. Mol Cell 2002;10:1057–69. [18] Rowley RB, Burkhardt AL, Chao HG, Matsueda GR, Bolen JB. J Biol Chem 1995;270:11590–4. [19] Takada Y, Mukhopadhyay A, Kundu GC, Mahabeleshwar GH, Singh S, Aggarwal BB. J Biol Chem 2003;278:24233–41. [20] Takata M, Sabe H, Hata A, Inazu T, Homma Y, Nukada T, et al. EMBO J 1994;13:1341–9. [21] Jabril-Cuenod B, Zhang C, Scharenberg AM, Paolini R, Numerof R, Beaven MA, et al. J Biol Chem 1996;271:16268–72. [22] Fu C, Turck CW, Kurosaki T, Chan AC. Immunity 1998;9:93– 103. [23] Goitsuka R, Fujimura Y-I, Mamada H, Umeda A, Morimura T, Uetsuka K, et al. J Immunol 1998;161:5804–8. [24] Wienands J, Schweikert J, Wollscheid B, Jumaa H, Nielsen PJ, Reth M. J Exp Med 1998;188:791–5. [25] Zhang Y, Wienands J, Zurn C, Reth M. EMBO J 1998;17:7304–10. [26] Hashimoto S, Iwamatsu A, Ishiai M, Okawa K, Yamadori T, Matsushita M, et al. Blood 1999;94:2357–64. [27] Kurosaki T, Tsukada S. Immunity 2000;12:1–5. [28] Su YW, Zhang Y, Schweikert J, Koretzky GA, Reth M, Wienands J. Eur J Immunol 1999;29:3702–11. [29] Kurosaki T. Ann Rev Immunol 1999;17:555–92.
73
[30] Rajewsky K, Gu H, Kuhn R, Betz UAK, Muller W, Roes J, et al. J Clin Investig 1996;98:600–3. [31] Scharenberg AM, Lin S, Cuenod B, Yamamura H, Kinet JP. EMBO J 1995;14:3385–94. [32] Yamamoto T, Matsuda T, Junicho A, Kishi H, Yoshimura A, Muraguchi A. FEBS Lett 2001;491:272–8. [33] Schneider I. J Embryol Exp Morphol 1972;27:353–65. [34] Bunch TA, Grinblat Y, Goldstein LS. Nucl Acids Res 1988;16:1043– 61. [35] Olsen MK, Rockenbach SK, Fischer HD, Hoogerheide JG, Tomich CS. Cytotechnology 1992;10:157–67. [36] Laemmli UK. Nature 1970;227:680–5. [37] Mahiouz DL, Aichinger G, Haskard DO, George AJ. J Immunol Methods 1998;212:149–60. [38] Kurosaki T, Johnson SA, Pao L, Sada K, Yamamura H, Cambier JC. J Exp Med 1995;182:1815–23. [39] Shiue L, Zoller MJ, Brugge JS. J Biol Chem 1995;270:10498–502. [40] Grupp SA, Mitchell RN, Schreiber KL, McKean DJ, Abbas AK. J Exp Med 1995;181:161–8. [41] Hombach J, Sablitzky F, Rajewsky K, Reth M. J Exp Med 1988;167:652–7. [42] Su Y, Flemming A, Wossning T, Hobeika E, Reth M, Jumaa H. J Ex Med 2003;198:1699–706. [43] Schamel WW, Kuppig S, Becker B, Gimborn K, Hauri HP, Reth M. Proc Natl Acad Sci USA 2003;100:9861–6. [44] Pluk H, Dorey K, Superti-Furga G. Cell 2002;108:247–59. [45] Visco C, Magistrelli G, Bosotti R, Perego R, Rusconi L, Toma S, et al. Biochemistry 2000;39:2784–91. [46] Hof P, Pluskey S, Dhe-Paganon S, Eck MJ, Shoelson SE. Cell 1998;92:441–50. [47] Harrison ML, Isaacson CC, Burg DL, Geahlen RL, Low PS. J Biol Chem 1994;269:955–9. [48] Peters JD, Furlong MT, Asai DJ, Harrison ML, Geahlen RL. J Biol Chem 1996;271:4755–62.
B cell antigen receptor assembly and Syk activation in the S2 cell reconstitution system Thomas Wossning, Michael Reth∗ a
Department of Molecular Immunology, Biology III, University of Freiburg, Stübeweg 51, 79108 Freiburg, Germany b Max-Planck-Institute for Immunobiology, Freiburg, Germany Received 17 September 2003; accepted 15 October 2003
Abstract Signal transduction from the B cell antigen receptor (BCR) involves a multitude of signaling molecules often organized in dynamic protein complexes. The molecular mechanisms operating during signaling are difficult to study solely by loss-of-function analysis. For a better understanding of the transient interaction of signaling molecules and their regulation by feedback loops, as well as their dynamic behavior in living cells, new techniques are required. We have developed a method allowing the reconstitution of the BCR complex and several of its key signaling elements in the evolutionary distant environment of the Drosophila S2 Schneider cell line. With this gain-of-function approach, we study here the assembly of the BCR complex and the control of its transport to the cell surface of S2 cells. We find that without binding to a light chain, the membrane-bound m heavy chain (mHC) homodimer, together with the Ig-␣/Ig- heterodimer, can come to the cell surface where it is signaling competent. This finding could have implications for potential signaling functions of such a receptor molecule during pro-/pre-B cell development. We also studied the activation of the BCR-proximal kinase Syk. We found that a truncated Syk mutant lacking the first (N-terminal) SH2 domain and the linker regions, is still regulated by autoinhibition and can only become activated in the presence of the BCR. This indicates that the C-terminal SH2 domain of Syk is the dominant regulatory subunit of this kinase. © 2004 Published by Elsevier B.V. Keywords: Drosophila S2 cells; BCR assembly; Syk activation
1. Introduction Mature B lymphocytes recognize and react to antigen via their B cell antigen receptor (BCR). This multiprotein complex includes the membrane-bound immunoglobulin (mIg) and the Ig-␣/Ig- heterodimer, which mediate antigen-binding and signal transduction, respectively. The mIg molecule is a symmetric disulfide-linked structure of two identical heavy (HCs) and two identical light chains (LCs). The Ig-␣/Ig- heterodimer is associated non-covalently with the mIg molecule in a complex with a 1:1 stoichiometry. This complex is stable in 1% digitonin lysates and results with other detergents like 0.2% NP40 or 0.1% thesit suggest that on the B cell surface, the BCR has an oligomeric structure [1,2]. The folding and assembly of BCR subunits in the endoplasmic reticulum (ER) is supported by ER-resident chap∗ Corresponding author. Tel.: +49-761-5108-420; fax: +49-761-5108-423. E-mail address: [email protected] (M. Reth).
0165-2478/$ – see front matter © 2004 Published by Elsevier B.V. doi:10.1016/j.imlet.2003.10.015
erones. For example, the HC-binding protein (BiP) binds to exposed hydrophobic sequences on the V and C domains of unpaired HC and LC [3,4]. Upon assembly, these chains loose their interaction with BiP thus allowing export of Ig from the ER [5,6]. In addition, the mIg molecule requires the assembly with the Ig-␣/Ig- heterodimer for the efficient transport to the cell surface [1,7]. The mIg molecules are retained in the ER by membrane-associated chaperons like calnexin, calreticulin and BAP29/BAP31 (BCR associated protein) heterodimers [8,9]. However, the requirements for the assembly and export of mIg molecules are less well studied than those for the secreted Ig molecules. The cytoplasmic tails of Ig-␣ and Ig- carry two tyrosines within the amino acid (aa) consensus sequence motif (D/E)x7 (D/E)xxYxx(L/I)x6−7 Yxx(L/I) known as immunoreceptor tyrosine-based activation motif (ITAM) [10]. These two tyrosines play an essential role in the activation of the BCR proximal protein tyrosine kinase (PTK) Syk [11,12]. At its N-terminus, Syk carries two tandem SH2 domains (SH2N and SH2C) interrupted by a flexible linker named interdomain A [13–15]. The two SH2 domains play
68
T. Wossning, M. Reth / Immunology Letters 92 (2004) 67–73
an important role in the regulation of Syk kinase activity. In the resting state the SH2 domains inhibit the kinase activity of Syk presumably by blocking access of ATP to the kinase domain [16,17]. Many signaling elements display similar intramolecular domain interactions that result in autoinhibition. When bound to the BCR, Syk can assume an open conformation and phosphorylate the two ITAM tyrosines. The tandem SH2 domains of Syk then bind to the doubly phosphorylated ITAM tyrosines (ppITAM), thus fixing the kinase in an open and active conformation [16,18]. This results in rapid phosphorylation of neighboring ITAM sequences, further Syk recruitment and the amplification of the BCR signal by a positive enzyme/product (Syk/ITAM) feedback [17]. However, this process is efficiently interrupted by BCR proximal protein tyrosine phosphatases (PTP) like SHP-1 [17]. Presumably the engagement of the BCR by antigen results not only in PTK activation but more importantly in PTP inhibition, which allows amplification of the BCR-signal. Fully activated Syk phosphorylates several cytosolic substrates, thereby leading to the activation of downstream signaling pathways including the mitogen-activated protein kinase (MAPK) and the NFKB pathway [19–21]. A prominent substrate of Syk is the adaptor protein SLP-65 (SH2 domain-containing leukocyte protein of 65 kDa), also known as BLNK (B cell linker protein) and BASH (B cell adaptor protein carrying an SH2 domain) [22–25]. Phosphorylated SLP-65 organizes a signaling complex that mediates the opening of intracellular Ca2+ stores in activated B cells [26–28]. The Src-family kinase Lyn also interacts with the BCR. However, Lyn phosphorylates dominantly only the first ITAM tyrosine and presumably supports the activity of Syk primarily via PTP inhibition rather than ppITAM phosphorylation [17]. The interactions of the BCR with its signaling elements are often transient and do not survive cellular lysis thus making it difficult to study these interactions by biochemical means. Therefore, in recent years, signaling systems have been most successfully studied by genetic means such as knock out techniques and mutational screens [29,30]. With these loss-of-function approaches, a signaling element can be attributed to a certain signaling pathway. However, the molecular details of a signaling process and its regulation by critical feedback cannot be elucidated by such loss-of-function analyses. A method to circumvent this problem is the reconstitution of signaling pathways in a different cell type as it has been shown for TCR and FcεRI signaling elements in non-lymphoid cell lines [31,32]. However, it is often impossible to express signaling elements in vertebrate cells without interfering with the growth and survival of the transfected cells. To solve this problem we have established a method allowing the transient and inducible coexpression of many mammalian genes in the genetically distant environment of S2 Schneider cells from Drosophila melanogaster [17,33] (Fig. 1). Here we use this system to study the molecular requirements for expression of the m
Fig. 1. The S2 signaling pathway reconstitution system allows the transient and inducible expression of 10–20 mammalian genes in the evolutionary distant environment of Drosophila Schneider cells.
heavy chain (mHC) on the cell surface. Furthermore we use mutations and truncations of the tandem SH2 domains of Syk to study the mechanism of Syk kinase activation. 2. Materials and methods 2.1. DNA constructs For inducible expression in S2, cells cDNAs of BCR components and signaling elements were each separately cloned into the S2 plasmid pRmHa-3 containing a copper-inducible metallothionein promoter [34,35]. For reconstitution of the IgD-BCR, we used vectors encoding an N-terminally flag-tagged Ig-a (pDflmb-1), HA-tagged Ig-b (pDHAB29) and a single chain ␦mHC (␦m) in which the VH and CH 1 domains are replaced by a VL /VH single chain [8,17]. Thus no separate LC is required to rebuild this BCR, here referred to as scIgD-BCR. For reconstitution of the IgM-BCR in S2 cells, we used PCR to amplify the mHC and the 1LC cDNAs from the m 3–11 myeloma cell line and cloned the PCR products into the S2 plasmid pRmHa-3 to generate the expression vectors pDm and pD1, respectively. The coexpression of these vectors with pDflmb-1 and pDHAB29 results in the expression of fully assembled IgM-BCR complexes. For reconstitution of BCR signaling pathways, we used vectors encoding human Syk (R. Scheuermann, University of Texas, Dallas; pDhSyk) and murine SLP-65 (pDSLP-65) [17,24]. Syk mutants were expressed using vectors pDSykmSH2N, pDSykmSH2C and pDSykSH2N coding for human Syk bearing a RQS/GGI mutation within the N-terminal SH2 domain, a RAR/GAL mutation within the C-terminal SH2 domain and a deletion of the N-terminal SH2 domain, respectively [17]. For construction of a plasmid encoding a truncated Syk that lacks both the N-terminal SH2 domain and interdomain A (pDDDSyk) amplified by PCR a Syk fragment using the oligos SykSH2Nlinker+ (aaaaaccggttggttccatggaaaaatctctcgg) and pRmHa-3-mcs-rev (tctagaggatccccgggtacc). The PCR product was cut with AgeI/BamHI and ligated into pDSykSH2N cut with the same enzymes, thereby yielding pDSyk.
T. Wossning, M. Reth / Immunology Letters 92 (2004) 67–73
69
2.2. Cell culture, transient transfection and FACS analysis Drosophila S2 Schneider were obtained as a generous gift from K. Karjalainen, Belinzona Institute for Immunology, Switzerland. They were grown in Schneider’s Drosophila medium (Life Technologies Inc.) supplemented with 10% fetal calf serum (FCS) at 27 ◦ C in a water-saturated atmosphere with atmospheric CO2 levels. For transient transfection, 2 × 106 cells from an exponentially growing culture were spread in six-well plates and cultured for 24 h. An amount of 0.3 g of each plasmid were diluted in 100 l serum-free medium and the amount of DNA was equalized with the empty vector pRmHa-3. An amount of 100 l serum-free medium containing 10 l CellFectin (Life Technologies Inc.) was added and the sample was incubated at RT for 10–15 min. The mixture was adjusted to 1 ml with serum-free medium and added to cells previously washed with serum-free medium. Following 18 h incubation, the solution was replaced with 3 ml medium and the cells were cultured for another 24 h. For induction of gene expression, CuSO4 was added to a final concentration of 1 mM for 24 h. For FACS analysis, cells were harvested using a cell scraper, washed with PBS and stained with anti-IgM ( chain specific, CyTM -5 conjugated, Jackson). To check for transfection efficiency and to recognize transfected cells, we cotransfected the S2 cells with the vector pDEGFP containing the EGFP coding sequence [17]. 2.3. Preparation of cell lysates and immunoblotting CuSO4 -induced cells were harvested using a cell scraper, washed with PBS and lysed in 0.5 ml Triton X-100 lysis buffer (50 mM Tris–HCl (pH 8.0), 140 mM NaCl, 0.5 mM EDTA (pH 8.0), 1 mM Na3 VO4 , 10% (v/v) glycerol, 1% (v/v) Triton X-100 and 10% (v/v) protease inhibitor cocktail (Sigma)). Protein concentrations in postnuclear supernatants were determined with protein assay dye reagent (Bio-Rad). Standardized lysates were boiled in 1× Laemmli buffer [36], separated by SDS-PAGE and analyzed by immunoblotting using the enhanced chemoluminescence detection system (ECL, Amersham). The antibodies used were anti-Syk (4D10, Santa Cruz), anti-SLP-65 [24], anti-flag (M2, Sigma), anti-HA (rat, Roche), anti-phosphotyrosine (4G10, UBI), anti-IgM ( chain specific, peroxidase conjugated, Pierce) and anti- (peroxidase conjugated, Southern).
3. Results 3.1. Requirements for mIgM expression on the surface of Drosophila S2 cells To investigate the requirements for the expression of a mHC on the cell surface we cloned cDNAs coding for murine mHC and 1LC each separately into the
Fig. 2. FACScan analysis of the expression of mHC on the surface of S2 cells transiently cotransfected with plasmids coding for mHC, Syk, SLP-65, EGFP in different combinations with plasmids for 1LC (indicated by − and +), Ig-␣ and/or Ig-. Transfected (EGFP+ ) cells were analyzed for mHC expression on the cell surface 24 h after induction of protein production. Numbers indicate the percentage of transfected cells expressing mHC on the surface.
vector pRmHa-3 containing a copper-inducible metallothionein promoter [34,35,37]. S2 cells were cotransfected with different combinations of plasmids encoding m, 1, N-terminally flag-tagged Ig-␣ (flag-Ig-␣) and HA-tagged Ig- (HA-Ig-) [8,17]. To investigate signal transduction through the IgM-BCR, we cotransfected vectors encoding Syk and SLP-65 as well as EGFP to identify transfected cells by FACS analysis [17]. The expression of EGFP and that of m on the cell surface were examined by FACS analysis 24 h after induction of gene expression with copper sulfate (Fig. 2). In a first series of experiments, we expressed the mHC without a LC either alone or in combination with Ig-␣ and/or Ig- (Fig. 2, upper panels). The efficiency of the transient transfection was controlled by the cotransfection of EGFP and was expressed in roughly 30% of the S2 cells. When the mHC was expressed in the absence of any other BCR components, only 2% of the transfected (EGFP+ ) S2 cells carry low amounts of m on the cell surface demonstrating that most of the unassembled m is retained in the ER (Fig. 2, upper left panel). The same is true for cells coexpressing m with Ig-␣ indicating that Ig-␣ alone is not able to overcome the ER retention of the mHC in S2 cells. In contrast to these findings, coexpression of m together with Ig- leads to a low expression of m on the surface of 42% of the transfected S2 cells. This indicates that Ig-, presumably in the form of an Ig-/Ig- homodimer, can bind to and overcome the ER retention of the mHC. Evidence for such a homodimer has been reported previously [7]. The coexpression of m, Ig-␣ and Ig- resulted in strong expression on the surface of 86% of the transfected cells. Thus the Ig-␣/Ig- heterodimer is more efficient than the Ig-/Ig- homodimer in the binding and transport of the mHC onto the surface of S2 cells (Fig. 2, upper right panel). The same analysis of mHC expression was repeated with cells coexpressing a 1LC (Fig. 2, lower panels). Low amounts of mIgM are found on the surface of 13% of the
70
T. Wossning, M. Reth / Immunology Letters 92 (2004) 67–73
transfected (EGFP+ ) S2 cells coexpressing the mHC and 1LC. This percentage is only slightly increased by the coexpression of Ig-␣, whereas Ig- is more efficient to promote the transport of the mIgM molecule in 75% of transfected S2 cells. The highest mHC expression is achieved by the coexpression of all BCR components and most (98%) of the EGFP+ transfected S2 cells carry the completely assembled IgM-BCR on their cell surface. These data confirm the unusually high cotransfection rate of S2 cells, which ensures that most transfected S2 cells express protein from all the vectors used. The S2 transfectants where also analyzed by Western blotting of total cellular lysates using antibodies specific for the BCR components, the signaling elements Syk and SLP-65 as well as for phosphotyrosine (pY) (Fig. 3). The mHC is expressed in similar amounts in all S2 transfectants independent of its coexpression with other BCR components (Fig. 3, third panel). In contrast, the Ig-␣ and Ig- components are more prominently expressed in cells allowing the assembly of the Ig-␣/Ig- heterodimer together with either the mHC or the mIgM molecule (Fig. 3, fifth and sixth panels, lanes 8 and 9). This suggests that in S2 cells, the mHC is a more stable protein than the Ig-␣/Ig- heterodimer and that the assembly with the mIg molecule can stabilize the Ig-␣/Ig- heterodimer and prevent its degradation. This notion is supported by a pulse/chase experiment in which S2 cells express either Ig-␣ alone, the Ig-␣/Ig- heterodimer or a scBCR containing a sc␦m in which the VH and CH 1 domains are replaced by an VL /VH single chain. Twenty four hours after induction of gene expression the cells were washed and cultured without further expression of the transfected vectors (Fig. 3b). After a 12 h chase period, no Ig-␣ protein is detected in S2 cell transfected with only the Ig-␣ or the Ig-␣ and Ig- vectors (Fig. 3b, first and second panels, lane 3). The S2 cells carrying the vectors for all three scBCR components, however, still contain some of the mature (lower molecular weight) form of the Ig-␣ protein 24 h after the start of the chase period (Fig. 3b, third panel, lane 4). Similar to the mHC, the sc␦m is more stable and can be detected after a 74 h chase period (Fig. 3b, fourth panel, lane 6). We previously have shown that coexpression of the kinase Syk with its in vivo substrate, the adaptor protein SLP-65, does not result in SLP-65 phosphorylation unless the cells coexpress an ITAM containing protein like Ig-␣ or Ig- [17]. Similarly, coexpression of Syk and SLP-65 together with either the mHC or the mIgM molecule does not result in tyrosine phosphorylation of SLP-65 (Fig. 3a, first panel, lanes 1 and 2). However, coexpression of these Ig molecules with Ig-␣ and/or Ig- leads to Syk activation and SLP-65 phosphorylation (Fig. 3, first panel). The most efficient Syk activation occurred when all components of the IgM-BCR were coexpressed (Fig. 3, lane 9). This may be due to the fact that the BCR complex carries several ITAM containing proteins which can facilitate the Syk/ITAM signal amplification at the receptor.
Fig. 3. Western blot analysis of tyrosine phosphorylation and protein production in S2 cells. (a) Cells were transiently transfected with plasmids coding for mHC, Syk, SLP-65, EGFP and different combinations of 1LC, Ig-␣ and/or Ig-. Twenty four hours after induction cells were lysed in Triton X-100 lysis buffer. The total cellular lysates were separated by SDS-PAGE and analyzed by immunoblotting using the indicated antibodies. (b) Stability of the BCR components tested in a pulse/chase experiment. S2 cells transiently transfected with plasmids coding for Ig-␣ (first panel), Ig-␣ and Ig- (second panel) as well as sc␦m, Ig-␣ and Ig- (third and fourth panels) were induced for 24 h and then washed and further cultured for the indicated time.
3.2. Regulation of Syk kinase activity by its SH2 domains A key to understanding of signal initiation at the BCR is the regulation of Syk kinase activity by the ITAM. It has been shown that Syk binds to ppITAMs via its tandem SH2
T. Wossning, M. Reth / Immunology Letters 92 (2004) 67–73
71
Fig. 4. Regulation of Syk kinase activity by its SH2 domains. (A) Schematic representation of wild type Syk and the analyzed Syk mutants. (B) Immunoblot analysis of tyrosine phosphorylation and protein production in S2 cells transiently transfected with plasmids coding for SLP-65 and either Syk wt or the Syk mutants, SykmSH2C, SykmSH2N, SykSH2N, and Syk () in the absence (−) or presence (+) of the scBCR. 24 h after induction cells were lysed in Triton X-100 lysis buffer and total lysates were separated by SDS-PAGE and analyzed by immunoblotting using the antibodies indicated. Arrows with an asterisk indicate the position of Syk wt, SykmSH2N and SykmSH2C. (C) Model of the autoinhibition structure of Syk wt and the Syk truncation mutant.
domains and that both domains are required for propagation of downstream signaling [16,18,38,39]. Deletion of the tandem SH2 domains generates a constitutively active enzyme with low substrate specificity [17]. Therefore, it has been suggested that Syk is autoinhibited by the intramolecular binding of the tandem SH2 domains to the kinase domain [39]. Since it has been shown that each individual SH2 domain exhibits different binding properties for the ppITAM sequence, we asked whether each domain contributes equally to kinase inhibition [17,38]. For this purpose, we generated different Syk mutants bearing either a RQS/GGI aa exchange within the N-terminal SH2 domain (SykmSH2N) or a RAR/GAL aa exchange in the C-terminal SH2 domain (SykmSH2C). We also constructed vectors for Syk deletion mutants lacking the N-terminal SH2 domain (SykSH2N) or both the N-terminal SH2 domain and the interdomain A (Syk) (Fig. 4A). Either wild type (wt) or mutant (m) Syk were expressed in S2 cells, together with SLP-65 alone or with SLP-65 and the scBCR (Fig. 4B). The Western blot analysis of total cellular lysates revealed almost no SLP-65 phosphorylation in cells lacking the scBCR demonstrating that the wt and mutant Syk are still autoinhibited (Fig. 4B, lanes 1, 3, 5, 7 and 9). In contrast, coexpression of Syk wt and SLP-65 in the presence of a BCR leads to strong phosphorylation of Ig-␣ and SLP-65 showing that the scBCR is able to induce Syk kinase activity (Fig. 4B, lane 2). As described previously [17], SykmSH2C does not become active in the presence of the scBCR (Fig. 4B, lane
4). In contrast, the SykmSH2N mutant is still activated by the scBCR albeit less strongly (Fig. 4B, lane 6). Together, these data indicate the SH2C domain of Syk is the dominant structure in the binding of ppITAM and the amplification of the BCR signal. However, the SH2N domain has some contribution to the autoinhibition of Syk since its deletion either alone or together with the interdomain A results in a stronger Syk activation than the mutation of the SykSH2N binding site (Fig. 4B, lanes 8 and 10). This also demonstrates that the autoinhibition of Syk requires the presence but not the phosphotyrosyl-binding capacity of the Syk SH2 domains (Fig. 4C).
4. Discussion 4.1. Assembly of the IgM-BCR In B cells, the export of BCR subunits from the ER is tightly controlled, thus allowing surface expression of only completely assembled BCRs [3,5–7,40,41]. Single or partially assembled BCR components are retained in the ER and this retention relies on at least two different mechanisms: BiP binding to hydrophobic patches on unpaired Ig domains, and binding of the BAP29/BAP31 heterodimer to conserved hydrophilic aa within the HC transmembrane region [3,4,8,9]. Both BiP and the BAP proteins are ubiquitously expressed and highly conserved [9]. Drosophila cells express BiP but there is only one BAP protein, namely BAP31 expressed
72
T. Wossning, M. Reth / Immunology Letters 92 (2004) 67–73
in this species. Nevertheless, it can be assumed that similar ER retention mechanisms for the BCR components are operating in S2 cells [9]. Indeed, the unpaired mHC chain is retained in the ER of S2 cells. This retention can be partially released by coexpression of either the 1LC or the Ig-␣/Ig- heterodimer, directly demonstrating the existence of the two different retention mechanisms mentioned above. Thus, it is likely that the association of endogenous BAP31 with the m transmembrane region and binding of BiP to the CH 1 domains of mHC dimer is involved in the retention of the mHC in the ER of S2 cells. However, in contrast to B cells, this retention is leaky in S2 cells since m surface expression is also detectable in the absence of one or more receptor subunits. For example, a partially assembled mIg/Ig- complex is expressed in detectable amounts on S2 cells but not on B cells. This leakiness may be due to the absence of BAP29 or other evolutionary alterations which does not allow the chaperone system of Drosophila to interact as efficiently with the BCR components as their murine counterparts. The amount of BCR complexes residing on the cell surface is not only determined by the rate of export but also by its stability and import rate. In this respect, it is feasible that the transport of an mIgM/Ig-/Ig- homodimer complex to the cell surface is as efficient as that of the complete BCR containing the Ig-␣/Ig- heterodimers but the former protein is less stable and thus more rapidly internalized and degraded [7]. This notion can be directly tested in the S2 system by pulse/chase experiments. A partially assembled BCR complex with the strongest expression on the cell surface of S2 cells is the m Ig-␣/Ig- complex and we show here that this complex is signaling competent. Such a complex has not been found so far on murine B cells most of which always express an LC together with a HC. This is different, however, in pre-B cells which have not yet assembled their LC genes. Although it is well known that the pre-BCR on these cells carry a surrogate LC, it remains possible that the m Ig-␣/Ig- complex is an alternative form of the pre-BCR and we have now found some indication of this on normal murine pre-B cells [42]. However, although S2 cells express foreign proteins to an amount comparable that in B cells, it cannot be completely excluded that overexpression of m, Ig-␣ and Ig- serves to circumvent ER retention in the absence of a LC as reported previously [17,43]. 4.2. Regulation of Syk kinase activity Upon antigen stimulation of the BCR, the tandem SH2 domains of Syk to the ppITAMs sequence and Syk becomes activated [17,18,38,39]. In contrast, in resting cells, the enzymatic activity of Syk is inhibited by a mechanism that involves the N-terminal part of the protein including both SH2 domains [17,39]. Thus, the tandem SH2 domains of Syk are not only involved in ITAM binding but also in binding to the kinase domain and the autoinhibition of Syk. Autoinhibition by intramolecular interactions between the SH2 and
kinase domain have been shown to regulate the activity of several other kinases like Src and abl [44]. Furthermore, this mechanism has also been suggested for regulation of the Syk family kinase ZAP70 [45]. In these kinases, formation of the closed conformation relies on intramolecular binding of the SH2 domain to a particular phosphotyrosyl residue. However, our data show that Syk kinase activity is not regulated by an analogous mechanism since the SH2N and SH2C domains do not require their phosphotyrosyl binding capacity for Syk inhibition. In this respect, the regulation of Syk is similar to that of the phosphatase SHP-2 the activity of which is also inhibited by its two tandem SH2 domains independently of their phosphotyrosyl binding capacity [46]. However, whereas the inhibition of SHP-2 mostly involves its SH2N domain, the deletion of the Syk SH2N domain only reduces but does not abolish the autoinhibition of Syk. Furthermore we have shown here that Syk is dominantly regulated by its SH2C domain. In previous studies, it has been reported that interdomain A of Syk shares the sequence Lxx(D/E)Y with two Syk substrate proteins, namely ␣-tubulin and erythrocyte band 3 that are tyrosine phosphorylated at this site [47,48]. This finding led to the suggestion that interdomain A acts as pseudosubstrate and blocks the substrate binding site of Syk thereby mediating the autoinhibition of Syk. However, our study of an interdomain A deletion mutant of Syk shows that this sequence does not play an important role in autoinhibition which mostly relies on SH2C and perhaps the interdomain B. How exactly these domains regulate Syk activity has to await the determination of the crystal structure of the complete Syk protein.
Acknowledgements We thank Christa Kalmbach-Zürn for technical support and Dr. Peter Nielsen for critical reading of the manuscript. We also thank Anna Erdei and János Gergely for organizing a marvelous signaling meeting and for their support throughout these years. This work was supported by the Deutsche Forschungsgesellschaft through SFB388.
References [1] Hombach J, Tsubata T, Leclercq L, Stappert H, Reth M. Nature 1990;343:760–2. [2] Schamel WWA, Reth M. Immunity 2000;13:5–14. [3] Knittler MR, Dirks S, Haas IG. Proc Natl Acad Sci USA 1995;92:1764–8. [4] Knittler MR, Haas IG. EMBO J 1992;11:1573–81. [5] Hendershot L, Bole D, Kohler G, Kearney JF. J Cell Biol 1987;104:761–7. [6] Hendershot LM. J Cell Biol 1990;111:829–37. [7] Brouns GS, de Vries E, Borst J. Int Immunol 1995;7:359–68. [8] Schamel WWA. Ph.D. Thesis, University of Freiburg, Freiburg, 1999. [9] Adachi T, Schamel WW, Kim KM, Watanabe T, Becker B, Nielsen PJ, et al. EMBO J 1996;15:1534–41.
T. Wossning, M. Reth / Immunology Letters 92 (2004) 67–73 [10] Reth M. Nature 1989;338:383–4. [11] Gold MR, Matsuuchi L, Kelly RB, DeFranco AL. Proc Natl Acad Sci USA 1991;88:3436–40. [12] Pao LI, Famiglietti SJ, Cambier JC. J Immunol 1998;160:3305–14. [13] Taniguchi T, Kobayashi T, Kondo J, Takahashi K, Nakamura H, Suzuki J, et al. J Biol Chem 1991;266:15790–6. [14] Muller B, Cooper L, Terhorst C. Immunogenetics 1994;39:359–62. [15] Law CL, Sidorenko SP, Chandran KA, Draves KE, Chan AC, Weiss A, et al. J Biol Chem 1994;269:12310–9. [16] Shiue L, Green J, Green OM, Karas JL, Morgenstern JP, Ram MK, et al. Mol Cell Biol 1995;15:272–81. [17] Rolli V, Gallwitz M, Wossning T, Flemming A, Schamel WW, Zurn C, et al. Mol Cell 2002;10:1057–69. [18] Rowley RB, Burkhardt AL, Chao HG, Matsueda GR, Bolen JB. J Biol Chem 1995;270:11590–4. [19] Takada Y, Mukhopadhyay A, Kundu GC, Mahabeleshwar GH, Singh S, Aggarwal BB. J Biol Chem 2003;278:24233–41. [20] Takata M, Sabe H, Hata A, Inazu T, Homma Y, Nukada T, et al. EMBO J 1994;13:1341–9. [21] Jabril-Cuenod B, Zhang C, Scharenberg AM, Paolini R, Numerof R, Beaven MA, et al. J Biol Chem 1996;271:16268–72. [22] Fu C, Turck CW, Kurosaki T, Chan AC. Immunity 1998;9:93– 103. [23] Goitsuka R, Fujimura Y-I, Mamada H, Umeda A, Morimura T, Uetsuka K, et al. J Immunol 1998;161:5804–8. [24] Wienands J, Schweikert J, Wollscheid B, Jumaa H, Nielsen PJ, Reth M. J Exp Med 1998;188:791–5. [25] Zhang Y, Wienands J, Zurn C, Reth M. EMBO J 1998;17:7304–10. [26] Hashimoto S, Iwamatsu A, Ishiai M, Okawa K, Yamadori T, Matsushita M, et al. Blood 1999;94:2357–64. [27] Kurosaki T, Tsukada S. Immunity 2000;12:1–5. [28] Su YW, Zhang Y, Schweikert J, Koretzky GA, Reth M, Wienands J. Eur J Immunol 1999;29:3702–11. [29] Kurosaki T. Ann Rev Immunol 1999;17:555–92.
73
[30] Rajewsky K, Gu H, Kuhn R, Betz UAK, Muller W, Roes J, et al. J Clin Investig 1996;98:600–3. [31] Scharenberg AM, Lin S, Cuenod B, Yamamura H, Kinet JP. EMBO J 1995;14:3385–94. [32] Yamamoto T, Matsuda T, Junicho A, Kishi H, Yoshimura A, Muraguchi A. FEBS Lett 2001;491:272–8. [33] Schneider I. J Embryol Exp Morphol 1972;27:353–65. [34] Bunch TA, Grinblat Y, Goldstein LS. Nucl Acids Res 1988;16:1043– 61. [35] Olsen MK, Rockenbach SK, Fischer HD, Hoogerheide JG, Tomich CS. Cytotechnology 1992;10:157–67. [36] Laemmli UK. Nature 1970;227:680–5. [37] Mahiouz DL, Aichinger G, Haskard DO, George AJ. J Immunol Methods 1998;212:149–60. [38] Kurosaki T, Johnson SA, Pao L, Sada K, Yamamura H, Cambier JC. J Exp Med 1995;182:1815–23. [39] Shiue L, Zoller MJ, Brugge JS. J Biol Chem 1995;270:10498–502. [40] Grupp SA, Mitchell RN, Schreiber KL, McKean DJ, Abbas AK. J Exp Med 1995;181:161–8. [41] Hombach J, Sablitzky F, Rajewsky K, Reth M. J Exp Med 1988;167:652–7. [42] Su Y, Flemming A, Wossning T, Hobeika E, Reth M, Jumaa H. J Ex Med 2003;198:1699–706. [43] Schamel WW, Kuppig S, Becker B, Gimborn K, Hauri HP, Reth M. Proc Natl Acad Sci USA 2003;100:9861–6. [44] Pluk H, Dorey K, Superti-Furga G. Cell 2002;108:247–59. [45] Visco C, Magistrelli G, Bosotti R, Perego R, Rusconi L, Toma S, et al. Biochemistry 2000;39:2784–91. [46] Hof P, Pluskey S, Dhe-Paganon S, Eck MJ, Shoelson SE. Cell 1998;92:441–50. [47] Harrison ML, Isaacson CC, Burg DL, Geahlen RL, Low PS. J Biol Chem 1994;269:955–9. [48] Peters JD, Furlong MT, Asai DJ, Harrison ML, Geahlen RL. J Biol Chem 1996;271:4755–62.