- Email: [email protected]
Identification of critical regions within the TIR domain of IL-1 receptor type I
The International Journal of Biochemistry & Cell Biology 68 (2015) 15–20
Contents lists available at ScienceDirect
The International Journal of Biochemistry & Cell Biology journal homepage: www.elsevier.com/locate/biocel
Identification of critical regions within the TIR domain of IL-1 receptor type I Jürgen Radons a,∗ , Werner Falk a , Stefan Dove b a b
Department of Internal Medicine I, University Clinic Regensburg, D-93042 Regensburg, Germany Institute of Pharmacy, University of Regensburg, D-93040 Regensburg, Germany
a r t i c l e
i n f o
Article history: Received 3 June 2015 Received in revised form 5 August 2015 Accepted 11 August 2015 Available online 14 August 2015 Keywords: IL-1 IL-1RI TIR domain NF-B Site-directed mutagenesis Homology modeling
a b s t r a c t Interleukin-1 receptor type I (IL-1RI) belongs to a superfamily of proteins characterized by an intracellular Toll/IL-1 receptor (TIR) domain. This domain harbors three conserved regions called boxes 1-3 that play crucial roles in mediating IL-1 responses. Boxes 1 and 2 are considered to be involved in binding of adapter molecules. Amino acids possibly crucial for IL-1RI signaling were predicted via homology models of the IL-1RI TIR domain based on the crystal structure of IL-1RAPL. The role of ten of these residues was investigated by site-directed mutagenesis and a functional luciferase assay reflecting NFB activity in transiently transfected Jurkat cells. In particular, the mutants E437K/D438K, E472A/E473A and S465A/S470A/S471A/E472A/E473A showed decreased and the mutant E437A/D438A increased IL1 responsiveness compared to the mouse IL-1RI wild type. In conclusion, the ␣C helix (Q469-E473 in mouse IL-1RI) is probably involved in heterotypic interactions of IL-1RI with IL-1RAcP or MyD88. © 2015 Elsevier Ltd. All rights reserved.
1. Introduction IL-1 is a pleiotropic cytokine playing a crucial role in inflammation and regulation of immune responses thereby affecting almost any cell type (Garlanda et al., 2013). It is also involved in the regulation of other homeostatic functions and the pathogenesis of different diseases such as chronic inflammatory and neurodegenerative diseases, atherosclerosis, and cancer (Lukens et al., 2012; Multhoff et al., 2012). IL-1 is expressed by tumor, stromal and endothelial cells and the host‘s infiltrating immune cells (Multhoff et al., 2012). IL-1 signal transduction is initiated by binding of either form of IL-1 to IL-1 receptor type I (IL-1RI), which undergoes a conformational change allowing association with IL-1 receptor accessory protein (IL-1RAcP). IL-1RAcP does not bind IL-1 but represents an essential component in the IL-1 signaling pathway
Abbreviations: DEAE, diethylaminoethyl; FACS, fluorescence-activated cell sorting; FCS, fetal calf serum; IL, interleukin; IL-1RI, IL-1 receptor type I; IL-1RAcP, IL-1 receptor accessory protein; IL-1RAPL, IL-1RAcP-like; LPS, lipopolysaccharide; Mal, MyD88 adaptor-like protein; MyD88, myeloid differentiation primary response protein MyD88; NF-B, nuclear factor kappa B; PI3K, phosphatidylinositol 3-kinase; PBS, phosphate-buffered saline; PKB, protein kinase B; SI, stimulation index; TBS, Tris-buffered saline; TILRR, Toll-like and IL-1R regulator; TIR, Toll/IL-1 receptor; TLR, Toll-like receptor; WT, wild type. ∗ Corresponding author at: Mühldorfer Str. 64, D-84503 Altötting, Germany. Tel.: +49 86714034200. E-mail address: [email protected] (J. Radons). http://dx.doi.org/10.1016/j.biocel.2015.08.009 1357-2725/© 2015 Elsevier Ltd. All rights reserved.
(Korherr et al., 1997; Radons et al., 2002; Wesche et al., 1997). The crystal structure of the ectodomains of an IL-1/IL-1RI/IL-1RAcP ternary complex has been solved recently (Thomas et al., 2012). IL-1-mediated receptor heterodimerization leads to recruitment of the intracellular adapter MyD88 via TIR-TIR domain interactions culminating in activation of intracellular signaling cascades including NF-B (Boraschi and Tagliabue, 2013; Gay et al., 2011; Watters et al., 2007). Recently, a novel IL-1RI co-receptor called Toll-like and IL-1R regulator (TILRR) was identified to associate with IL-RI thus amplifying IL-1-mediated NF-B activation and inflammatory responses (Zhang et al., 2010; Zhang et al., 2012). IL-1 signaling also involves recruitment of phosphatidylinositol 3-kinase (PI3K) to the IL-1 receptor complex (Reddy et al., 1997) and subsequent activation of NF-B via AKT/PKB (Cahill and Rogers, 2008). IL-1 binding also activates numereous G proteins resulting in an IB␣independent NF-B transactivation (Jefferies and O’Neill, 2000; Singh et al., 1999). IL-1 signaling finally regulates gene expression of a great variety of other genes involved in immune defence, repair and others (Multhoff et al., 2012; O’Neill, 2000). For successful initiation of signaling, oligomerization of TIR domains and association of adapter molecules was shown to be required (Jain et al., 2014; Multhoff et al., 2012; Narayanan and Park, 2015; Ve et al., 2015). Within the TIR domain three typical regions of conservation can be found: box 1 (D-K-YDAF-SY), box 2 (GYKLCI–RD–PG), and box 3 (-FWKx-) (Bowie and O’Neill, 2000; O’Neill and Dinarello, 2000). Bold amino acids are identical in nearly all members while conservative substitutions may occur in the other positions. In
16
J. Radons et al. / The International Journal of Biochemistry & Cell Biology 68 (2015) 15–20
dominant negative TLR-4, the proline in box 2 is substituted by histidine in C3H/HeJ mice abolishing the host immune response to LPS (Poltorak et al., 1998). Only IL-1RI and TLR-3 do not harbor a proline at this position. Results on human IL-1RI mutants indicated the conserved arginine in box 2 of IL-1RI and the FWKxxRY sequence including box 3 to be crucial for IL-1 signaling (Heguy et al., 1992). By contrast, in another study single point alanine mutation of R, W and K (sequence RFWK) did not affect function (Slack et al., 2000). Boxes 1 and 2 are considered to be involved in binding of adapter molecules (Bowie and O’Neill, 2000; Slack et al., 2000). Our previous investigations identified certain amino acids within the loop preceding box 3 in IL-1RAcP as being crucial for IL-1 responsiveness (Radons et al., 2003). The present study was to identify amino acids of the IL-1RI TIR domain contributing to receptor signaling. Since the intracellular TIR domains are not part of the crystal structure of the IL-1/IL-1RI/IL-1RAcP heterotrimer (Wang et al., 2010), a homology model of the mouse IL-1RI TIR domain (mIL-1RI-TIR) based on the structure of the human IL-1R accessory protein-like (IL-1RAPL) (Khan et al., 2004) was generated and used for suggesting promising positions to be mutated. The role of these residues was identified by site-directed mutagenesis and a functional luciferase assay reflecting NF-B activity in transiently transfected Jurkat cells. 2. Materials and methods 2.1. Homology modeling of the mIL-1RI TIR domain An initial model of mIL-1RI-TIR was generated by the program COMPOSER (Blundell et al., 1988) as part of the molecular modeling suite SYBYL 7.3 (Certara, L.P., Princeton, NJ). Homology modeling is based on known crystal structures serving as templates for structurally conserved regions (SCRs). Among structures of TIR domains present in the Brookhaven Protein Databank (PDB) – human TLR-1 (PDB ID 1fyv), TLR-2 (1fyw) (Xu et al., 2000) and IL-1RAPL (1t3g) (Khan et al., 2004) – chain A of the latter was selected as single template since only in this case the degree of homology with mIL-1RI was sufficient (ca. 30% sequence identity compared to ca. 20% with TLR-1 and TLR-2). Fig. 1 presents the optimal sequence alignment and the SCRs determined by COMPOSER. In the present case of a single template structure, the backbone frameworks of the SCRs of mIL-1RI-TIR and of IL-1RAPL are identical. Side chains were added to the target SCR backbone
by a knowledge-based approach, taking in account the backbone secondary structure and the side chain conformation at the corresponding residues of the template. The structurally variable regions (SVRs) of the model were constructed by the loop search algorithm within SYBYL. For each SVR, appropriate fragments from a binary PDB database with the same length as the SVR of the target are proposed on the base of distances and superpositions of the anchor residues. After adding hydrogens, the model was energy minimized using the AMBER-FF99 force field (Cornell et al., 1995) with AMBERFF99 charges (distance dependent dielectricity constant 4, first 50 cycles with constrained backbone and steepest descent method) up to a root mean square gradient of 0.05 kcal/mol Å (Powell conjugate gradient). Surfaces and electrostatic potentials of the model were calculated and visualized by the program MOLCAD (MOLCAD GmbH, Darmstadt, Germany) implied in SYBYL 7.3. 2.2. Construction of the reporter plasmid and mutated forms of IL-1RI For determination of transcriptional activity of NF-B, five tandemly arranged NF-B binding sites from the HIV long terminal repeat enhancer (5 -GGGACTTTCC-3 ) (Sen and Baltimore, 1986) were cloned via BglII and HindIII into the luciferase reporter vector pGL3-Basic (pGL3–5xNF-B). Mutated forms of mIL-1RI were generated by site-directed mutagenesis using the QuickChangeTM site-directed mutagenesis kit (Stratagene, Amsterdam, The Netherlands). The vector pFLAG-IL-1RI, a derivative of pFLAG-CMV-1 (Sigma–Aldrich GmbH, Steinheim, Germany) was used as a template in all mutagenesis experiments. All mutations were verified by sequence analysis. Expression and protein production by all mutants was analyzed by FACS detection of the Flag epitope in transfected cells using anti-Flag antibody which was done with permeated cells. 2.3. Detection of NF-B activity in transiently transfected Jurkat cells The human T cell leukaemia cell line Jurkat, lacking a receptor for IL-1 (Kuno et al., 1993), was purchased from the American Type Culture Collection (ATCC; Manassas, USA) and maintained in RPMI 1640 medium (PAN Biotech GmbH, Aidenbach, Germany) containing 10% FCS, 100 U/ml penicillin, 100 mg/ml streptomycin and 2 mM glutamine at 37 ◦ C in humidified air with 5% CO2 . For
Fig. 1. Sequence alignment of TIR domains of hIL-1RI (UniProtKB P14778), mIL-1RI (P13504) and IL-1RAPL (Q9NZN1, the template for homology modeling). ␣-Helices (blue) and -sheets (green) are labeled according to Khan et al. (2004). Red frames indicate boxes 1, 2, and 3. mIL-1RI positions subjected to mutation are highlighted in yellow. SCR1–SCR5: structurally conserved regions determined by COMPOSER. 7x–seven additional residues not resolved in the IL-1RAPL structure (PDB ID 1t3g).
J. Radons et al. / The International Journal of Biochemistry & Cell Biology 68 (2015) 15–20
17
transfection, 100 l of cells (106 /ml) in TBS were mixed with 30 l of freshly prepared DEAE-dextran solution (10 mg/ml) and 470 l TBS with 1 g of vector DNA and 0.5 g of the luciferase reporter plasmid pGL3–5xNFB harboring five NF-B binding sites (Radons et al., 2002). The total amount of DNA in all transfections was kept constant by adding empty vector DNA. The mixture was incubated for 10 min with agitation and washed once with RPMI 1640 containing 10% FCS, mercaptoethanol (3 × 10−5 M) and gentamycin (50 g/ml). The pellet was resuspended in 1.5 ml of medium, and 50 l were used for stimulation with either recombinant human IL-1␣ (rhIL-1␣; Pan Biotech, Aidenbach, Germany; 1 or 10 ng/ml) or recombinant human TNF (rhTNF; 10 ng/ml) for 6 h or remained unstimulated. NF-B activation was determined by measuring the luciferase activity in transiently transfected cells using the LucLite Plus luciferase reporter gene assay kit (Perkin Elmer Life Sciences, Shelton, USA). Data from corresponding cultures from all experiments were pooled and the highest and the lowest value in all groups were deleted. Afterwards, a mean cpm was established for unstimulated cultures which was then used to convert all cpm to stimulation indices (SI, stimulated/mean unstimulated). A mean SI was then calculated for the WT transfectants stimulated with TNF and IL-1. This made it possible to determine an SI in % of WT for all values; (SI sample/mean SI WT) x 100). 2.4. Statistical analysis Level, spread, and symmetry of data distribution were conveyed by a box-and-whisker plot. Statistical differences between medians were analyzed using the two-tailed non-parametric WilcoxonMann-Whitney test. A P value of less than 0.05 was considered to reflect statistical significance. 3. Results and discussion 3.1. Structure-based selection of IL-1RI mutants The homology model of mIL-1RI-TIR based on the crystal structure of human IL-1RAPL (Khan et al., 2004) comprises all characteristic elements of TIR domains (see sequence alignment, Fig. 1, and structural overview, Fig. 2A). A five-strand parallel -sheet is surrounded by seven ␣-helices. Three typical segments assumed to be functionally significant, boxes 1-3, and moreover, the whole BB-loop belong to the structurally conserved regions (SCRs) and are probably well approximated by the mIL-1RI-TIR model. Differences in the structures of IL-1RAPL (Khan et al., 2004) on one hand as well as TLR-1 and TLR-2 on the other hand (Xu et al., 2000) indicate variable regions possibly not correctly reproduced by the model. In particular, (i) the short helices ␣C and ␣C are followed by a wellordered helix ␣C in the case of IL-1RAPL and by a highly disordered region in the case of TLR-1 and TLR-2, respectively, (ii) ␣C is slightly shifted in IL-1RAPL vs. TLR-1, and (iii) helix ␣D in TLR-1 and TLR-2 is nearly perpendicular to that in IL-1RAPL (Khan et al., 2004). The most striking specific feature of both human and mouse IL-1RI-TIR compared to TIR domains of IL-1RAPL, IL-1RAcP, TLR-1 and TLR-2 is a cluster of acidic amino acids at the C-terminus of the BB-loop and in helix ␣C (Fig. 2B). Additionally, three serines before (S465) and within ␣C (S470, S471) are typical. Moreover, ␣C and the BB-loop belong to the interaction interfaces in homodimers (asymmetric units) of IL-1RAPL (Khan et al., 2004), TLR-1 (Xu et al., 2000) and TLR-2 (Tao et al., 2002) although the crystallographic interfaces of these TIR domains are different and probably not biologically relevant. In a model based on the crystal structure of the dimeric TLR-10 TIR domain, dimerization of TLR-4 TIR domains was shown to require formation of a large conserved platform containing the BB-loop and box 1 motifs and thereby
Fig. 2. Homology model of the mIL-1RI TIR domain. Secondary structure elements are labeled and colored (␣-helices blue, -strands green). Specific conserved regions (boxes 1–3) are drawn in red. (A) Ribbon and tube representation of the tertiary structure. Ocher balls indicate mutated positions. (B) Detailed model of the region with all but one (S519) amino acids subjected to mutation (ocher C atoms). Other atom colors: O – red, N – blue, C – gray.
forming a potential binding site for intracellular adapters (Bovijn et al., 2012). Although the involvement of the BB-loop in TLR-4 TIR dimerization has been described previously, further data from this group suggest that this region establishes contact with helix ␣E of TLR-4, rather than helix ␣C as found in the structure of TLR-10 TIR crystals (Toshchakov et al., 2011). Noteworthy, the crystal structure of the TIR domain of TLR-6 recently revealed that helix ␣C rather than the BB-loop plays a crucial role in homodimeric TIR-TIR interactions (Jang and Park, 2014). Accordingly, homodimerization of Mal-TIR domains lacking the BB-loop involves residues of the ␣C and ␣D helices of both monomers (Lin et al., 2012), implying that the functional dimer does not inevitably represent the “most stable” among possible dimer conformations (Toshchakov et al., 2011). In Mal-TIR dimers, E190 (␣D) of one monomer forms a hydrogen bond with K158 (␣C ) of the other monomer, while P189 and F193 in ␣D of one monomer form stacking interactions with P155 and
18
J. Radons et al. / The International Journal of Biochemistry & Cell Biology 68 (2015) 15–20
Fig. 3. Effect of different mutant variants of IL-1RI on NF-B activation. Jurkat cells were transiently co-transfected with 500 ng of luciferase reporter and 1 g of the indicated IL-1RI constructs. 6 h after transfection, cells were stimulated with different concentrations of rhIL-1␣ or 10 ng/ml rhTNF. After incubation for an additional 6 h, luciferase activity was determined in triplicate as a measure of NF-B activation and normalized to the TNF response. Data were visualized by a box-and-whisker plot. The boxes show the 25th and 75th percentiles together with the medians while the whiskers show the 5th and 95th percentiles, respectively. Data represent the medians of three to six separate experiments performed in at least triplicates. *, p < 0.05 compared with wild type (WT).
W156 in ␣C of the other monomer, respectively (Lin et al., 2012). In the case of IL-1RAPL, S488 (␣C ) of one monomer forms a hydrogen bond with S488 of the other monomer, an interaction that is reproduced by a disulfide bridge of the corresponding cysteines in the TLR-1 dimer. Thus, it may be suggested that the whole BB-␣C face is involved in heterodimerization in the case of IL-1RI, too. Consequently, the specific acidic and polar amino acids in this region are particularly interesting for in-vitro mutagenesis experiments, considering also Q469 at the beginning of ␣C , the only amino acid of the whole BB-␣C face which differs between mouse and human IL-1RI (G466). The conserved proline in box 2 (BB-loop) of nearly all TIR domains is replaced by valine in IL-1RI species orthologs (mIL-1RI: V435). Substitution of proline by polar residues – histidine in TLR-4 (Poltorak et al., 1998), histidine and aspartate in IL-1RAcP (Radons et al., 2002) – impaired function, whereas an IL-1RAcP P446A mutation was without effect (Radons et al., 2002). Comparison of the BB-loops in IL-1RAPL and the mIL-1RI model suggests that proline and valine do not lead to different backbone conformations. 3.2. Effects of IL-1RI mutants on NF-B activation NF-B activation mediated by several intracellular signaling pathways was determined by measuring the luciferase activity of
our reporter plasmid pGL3–5xNFB harboring five NF-B binding sites (Radons et al., 2002). This NF-B activation assay does not distinguish between activation via the MyD88-dependent and AKT/PKB-dependent pathways. Significant changes in IL-1 responsiveness of transfected Jurkat cells were observed in eight out of 13 tested mutants (Fig. 3). As expected and as control, the mutant with the complete intracytoplasmatic region missing (Stop380) showed no response to IL-1. In case of the five mutants E437K, D438K, D438A, S519K and Q469P, responses were not significantly different from that of the wild type. The four mutants E437K/D438K, E437A, E472A/E473A and 5xA showed decreased IL-1 responsiveness, albeit in case of E437K/D438K and E437A only at the lower IL-1 concentration. Effects of the 5xA mutant were exceptionally strong. Increased responsiveness to IL-1 was observed for the mutants E437A/D438A, V435P and Q469A (for V435P and Q469A only at the higher IL-1 concentration). It has been shown previously that substitution of the conserved R within box 2 of IL-1RI by alanine caused a massive reduction in IL-1-induced NF-B activation (Slack et al., 2000). Moreover, this box 2 receptor mutant was found to strongly decrease IL-1mediated receptor activation after blockage of TILRR expression identifying box 2 as being a putative interaction site for this cytosolic adapter (Zhang et al., 2010). The observed decreased IL-1 responsiveness of our C-terminal box 2 mutants (E437K,
J. Radons et al. / The International Journal of Biochemistry & Cell Biology 68 (2015) 15–20
Fig. 4. Electrostatic potential mapped on a MOLCAD Connolly surface of the mIL-1RI TIR domain. The potential is based on AMBER-FF99 charges. Colors: most negative – violet, negative – blue, neutral – cyan, green, positive – yellow, most positive – brown. Labeled are amino acids subjected to mutation.
E437K/E438K) might rely, at least in part, on restrictions in TILRR recruitment. Since the E472A/E473A mutation reduced IL-1-induced NF-B activation by 25% and the 5xA variant even by 75%, the specific ␣C helix in mIL-1RI (and hIL-1RI) is functionally relevant. Fig. 4 presents the electrostatic potential map of the mIL-1RI model. The acidic nature of ␣C leads to a high negative potential. The region around helix ␣E including the conserved box 3 is positively charged due to four basic amino acids (K530, R532, K535, R538). Two of these basic amino acids (R537 and K540 corresponding to mIL1RI R532 and K535, respectively) are present in IL-1RAcP as well so that its EE-loop and box 3 are positively charged, too. The TIR domain of the adaptor protein MyD88 shows a similar charge distribution (Li et al., 2005). Our previous results on IL-1RAcP mutants indicated that the EE-loop and box 3 are involved in TIR domain interactions (Radons et al., 2002). Docking of IL-1RAcP and MyD88 based on homology models with human TLR-2 as template and in-vitro mutagenesis results suggested electrostatic interactions between the DRD-motif in box 2 (BB-loop) of MyD88 and the lysine patch in the EE-loop of IL-1RAcP (Li et al., 2005). Heterodimerization of IL-1RAcP and MyD88 precedes homotypic oligomerization of MyD88 (Akira, 2003; Akira and Sato, 2003; Yamamoto et al., 2004) again based on box 2-box 3 interactions (Li et al., 2005), and recruitment of downstream signaling transducers by the MyD88 oligomers. Common for all these TIR-TIR interactions is that a negatively charged pocket in the region of box 2 and ␣C/␣C may provide a binding site for protruding positively charged residues in or close to box 3. The shape of the corresponding regions of IL-1RI shows this general pattern, too (Fig. 4). Heterodimerization or oligomerization of IL-1RI with IL-1RAcP and/or MyD88 may rely on these interactions. Thus, IL-1RI is probably involved in transient or stable TIR-TIR interactions essential for downstream signaling. Strikingly, mIL-1RI Q469 in the IL-1R specific region (mIL-1RI amino acids 465, 467-473) may be exchanged by glycine (hIL-1RI G466). To investigate the significance of a polar amino acid at this position, Q469 (␣C ) was mutated to alanine and proline, respectively. Whereas Q469P was without any effect on NF-B activation, mutation of Q469 to alanine led to a significant increase of the IL-1
19
response at the higher IL-1 concentration, suggesting that a small neutral amino acid like glycine or alanine is more favorable than the polar glutamine in mIL-1RI. Due to the IL-1RI-specific amino acids E437 and D438, the Cterminus of the BB-loop contributes to the large negatively charged region spanned by parts of box 2, ␣C , the DD-loop and ␣D. Lysine and alanine mutation of these acidic amino acids behaved quite contradictory. Only E437A and the E437K/D438K double mutant led to a significant reduction in IL-1-induced NF-B activation, but in both cases only at the lower IL-1 concentration. Surprisingly, the double-A mutant even induced increased effects at both 1 and 10 ng/ml IL-1. Thus, a negative potential at the C-terminal end of the BB-loop is not necessary for IL-1RI functionality. At least one basic amino acid is well tolerated. Among the members of the TIR family, a highly conserved proline can be found in box 2. This residue corresponds to a naturally occurring missense point mutation resulting in the substitution of P712 by histidine (the LPSd mutation) in dominant negative TLR-4 in C3H/HeJ mice selectively leading to LPS hyporesponsiveness (Poltorak et al., 1998). Equivalent substitutions in other TLR molecules also abrogate inflammatory responses (Ozinsky et al., 2000; Underhill et al., 1999). In contrast to the other TIR family members, the conserved proline as part of box 2 within the BBloop is substituted by valine in IL-1RI (mIL-RI V435). Exchange of V435 by proline improved IL-1-mediated NF-B activation, but only at the higher IL-1 concentration. In accordance with the model in Fig. 2, the V435P mutation probably stabilizes the BB-loop without major structural effects. Interestingly, TLR-3 bears an alanine (A795) in place of the conserved BB-loop proline. Mutation of A795 to proline also improved NF-B activation (Verstak et al., 2013) most likely due to the proposed effects on BB-loop stability thereby modulating recruitment of downstream adapter molecules. In nearly all TIR domains, a conserved WX motif can be found at the E-EE junction with WK in IL-1RAcP, WH in IL-1RAPL and WS in IL-1RI. To test for specificity, S519 in mIL-1RI was mutated to lysine thereby reconstituting the IL-1RAcP motif. This mutation did not affect IL-1-induced NF-B activation. In our previous study on IL-1RAcP, mutation of the conserved W526 to phenylalanine and the neighbouring K527 to glycine was without effect on IL1 responsiveness, too (Radons et al., 2002). Taken together, these findings argue against a crucial role of this motif within the NF-B signaling process. In conclusion, site-directed mutagenesis and a functional luciferase assay reflecting NF-B activity in transiently transfected Jurkat cells identified the specific ␣C helix of the IL-1RI to be crucial for signaling. Homology modeling of the mouse IL-1RI also indicated putative TIR-TIR domain interfaces for heterotypic interactions. The negative electrostatic potential and the shape of this region including the BB-loop (box 2) predestine interactions with the positively charged area around box 3 of IL-1RAcP or MyD88. The acidic C-terminal end of the BB-loop is not necessary with respect to function and dimerization. Back mutation of valine in the BB-loop into the highly conserved proline of other TIR domains did only slightly improve NF-B activity, indicating minor stabilization of the BB-loop. Acknowledgements pFLAG-IL-1RI was a generous gift from Michael Martin (Gießen, Germany). The excellent technical assistance of Daniela Aigner is gratefully acknowledged. References Akira, S., 2003. Toll-like receptor signaling. J. Biol. Chem. 278, 38105–38108.
20
J. Radons et al. / The International Journal of Biochemistry & Cell Biology 68 (2015) 15–20
Akira, S., Sato, S., 2003. Toll-like receptors and their signaling mechanisms. Scand. J. Infect. Dis. 35, 555–562. Blundell, T., Carney, D., Gardner, S., Hayes, F., Howlin, B., Hubbard, T., et al., 1988. 18th Sir Hans Krebs lecture. Knowledge-based protein modelling and design. Eur. J. Biochem. 172, 513–520. Boraschi, D., Tagliabue, A., 2013. The interleukin-1 receptor family. Semin. Immunol. 25, 394–407. Bovijn, C., Ulrichts, P., De Smet, A.S., Catteeuw, D., Beyaert, R., Tavernier, J., et al., 2012. Identification of interaction sites for dimerization and adapter recruitment in Toll/interleukin-1 receptor (TIR) domain of Toll-like receptor 4. J. Biol. Chem. 287, 4088–4098. Bowie, A., O’Neill, L.A., 2000. The interleukin-1 receptor/Toll-like receptor superfamily: signal generators for pro-inflammatory interleukins and microbial products. J. Leukoc. Biol. 67, 508–514. Cahill, C.M., Rogers, J.T., 2008. Interleukin (IL) 1beta induction of IL-6 is mediated by a novel phosphatidylinositol 3-kinase-dependent AKT/IkappaB kinase alpha pathway targeting activator protein-1. J. Biol. Chem. 283, 25900–25912. Cornell, W.D., Cieplak, P., Bayly, C.I., Gould IRMKM, Ferguson, D.M., Spellmeyer, D.C., et al., 1995. A second generation force field for the simulation of proteins, nucleic acids, and organic molecules. J. Am. Chem. Soc. 117, 5179–5197. Garlanda, C., Dinarello, C.A., Mantovani, A., 2013. The interleukin-1 family: back to the future. Immunity 39, 1003–1018. Gay, N.J., Gangloff, M., O’Neill, L.A., 2011. What the Myddosome structure tells us about the initiation of innate immunity. Trends Immunol. 32, 104–109. Heguy, A., Baldari, C.T., Macchia, G., Telford, J.L., Melli, M., 1992. Amino acids conserved in interleukin-1 receptors (IL-1Rs) and the Drosophila toll protein are essential for IL-1R signal transduction. J. Biol. Chem. 267, 2605–2609. Jain, A., Kaczanowska, S., Davila, E., 2014. IL-1 receptor-associated kinase signaling and its role in inflammation, cancer progression, and therapy resistance. Front. Immunol. 5, 553. Jang, T.H., Park, H.H., 2014. Crystal structure of TIR domain of TLR6 reveals novel dimeric interface of TIR-TIR interaction for toll-like receptor signaling pathway. J. Mol. Biol. 426, 3305–3313. Jefferies, C.A., O’Neill, L.A., 2000. Rac1 regulates interleukin 1-induced nuclear factor kappaB activation in an inhibitory protein kappaBalpha-independent manner by enhancing the ability of the p65 subunit to transactivate gene expression. J. Biol. Chem. 275, 3114–3120. Khan, J.A., Brint, E.K., O’Neill, L.A., Tong, L., 2004. Crystal structure of the Toll/interleukin-1 receptor domain of human IL-1RAPL. J. Biol. Chem. 279, 31664–31670. Korherr, C., Hofmeister, R., Wesche, H., Falk, W., 1997. A critical role for interleukin-1 receptor accessory protein in interleukin-1 signaling. Eur. J. Immunol. 27, 262–267. Kuno, K., Okamoto, S., Hirose, K., Murakami, S., Matsushima, K., 1993. Structure and function of the intracellular portion of the mouse interleukin 1 receptor (type I). Determining the essential region for transducing signals to activate the interleukin 8 gene. J. Biol. Chem. 268, 13510–13518. Li, C., Zienkiewicz, J., Hawiger, J., 2005. Interactive sites in the MyD88 Toll/interleukin (IL) 1 receptor domain responsible for coupling to the IL1beta signaling pathway. J. Biol. Chem. 280, 26152–26159. Lin, Z., Lu, J., Zhou, W., Shen, Y., 2012. Structural insights into TIR domain specificity of the bridging adaptor Mal in TLR4 signaling. PLoS ONE 7, e34202. Lukens, J.R., Gross, J.M., Kanneganti, T.D., 2012. IL-1 family cytokines trigger sterile inflammatory disease. Front. Immunol. 3, 315. Multhoff, G., Molls, M., Radons, J., 2012. Chronic inflammation in cancer development. Front. Immunol. 2, 98. Narayanan, K.B., Park, H.H., 2015. Toll/interleukin-1 receptor (TIR) domain-mediated cellular signaling pathways. Apoptosis 20, 196–209. O’Neill, L.A., 2000. The interleukin-1 receptor/Toll-like receptor superfamily: signal transduction during inflammation and host defense. Sci STKE 2000, re1. O’Neill, L.A., Dinarello, C.A., 2000. The IL-1 receptor/toll-like receptor superfamily: crucial receptors for inflammation and host defense. Immunol. Today 21, 206–209. Ozinsky, A., Underhill, D.M., Fontenot, J.D., Hajjar, A.M., Smith, K.D., Wilson, C.B., et al., 2000. The repertoire for pattern recognition of pathogens by the innate
immune system is defined by cooperation between toll-like receptors. Proc. Natl. Acad. Sci. USA 97, 13766–13771. Poltorak, A., He, X., Smirnova, I., Liu, M.Y., Van Huffel, C., Du, X., et al., 1998. Defective LPS signaling in C3H/HeJ and C57BL/10ScCr mice: mutations in Tlr4 gene. Science 282, 2085–2088. Radons, J., Dove, S., Neumann, D., Altmann, R., Botzki, A., Martin, M.U., et al., 2003. The interleukin 1 (IL-1) receptor accessory protein Toll/IL-1 receptor domain: analysis of putative interaction sites by in vitro mutagenesis and molecular modeling. J. Biol. Chem. 278, 49145–49153. Radons, J., Gabler, S., Wesche, H., Korherr, C., Hofmeister, R., Falk, W., 2002. Identification of essential regions in the cytoplasmic tail of interleukin-1 receptor accessory protein critical for interleukin-1 signaling. J. Biol. Chem. 277, 16456–16463. Reddy, S.A., Huang, J.H., Liao, W.S., 1997. Phosphatidylinositol 3-kinase in interleukin 1 signaling. Physical interaction with the interleukin 1 receptor and requirement in NFkappaB and AP-1 activation. J. Biol. Chem. 272, 29167–29173. Sen, R., Baltimore, D., 1986. Multiple nuclear factors interact with the immunoglobulin enhancer sequences. Cell 46, 705–716. Singh, R., Wang, B., Shirvaikar, A., Khan, S., Kamat, S., Schelling, J.R., et al., 1999. The IL-1 receptor and Rho directly associate to drive cell activation in inflammation. J. Clin. Invest. 103, 1561–1570. Slack, J.L., Schooley, K., Bonnert, T.P., Mitcham, J.L., Qwarnstrom, E.E., Sims, J.E., et al., 2000. Identification of two major sites in the type I interleukin-1 receptor cytoplasmic region responsible for coupling to pro-inflammatory signaling pathways. J. Biol. Chem. 275, 4670–4678. Tao, X., Xu, Y., Zheng, Y., Beg, A.A., Tong, L., 2002. An extensively associated dimer in the structure of the C713S mutant of the TIR domain of human TLR2. Biochem. Biophys. Res. Commun. 299, 216–221. Thomas, C., Bazan, J.F., Garcia, K.C., 2012. Structure of the activating IL-1 receptor signaling complex. Nat. Struct. Mol. Biol. 19, 455–457. Toshchakov, V.Y., Szmacinski, H., Couture, L.A., Lakowicz, J.R., Vogel, S.N., 2011. Targeting TLR4 signaling by TLR4 Toll/IL-1 receptor domain-derived decoy peptides: identification of the TLR4 Toll/IL-1 receptor domain dimerization interface. J. Immunol. 186, 4819–4827. Underhill, D.M., Ozinsky, A., Hajjar, A.M., Stevens, A., Wilson, C.B., Bassetti, M., et al., 1999. The Toll-like receptor 2 is recruited to macrophage phagosomes and discriminates between pathogens. Nature 401, 811–815. Ve, T., Williams, S.J., Kobe, B., 2015. Structure and function of Toll/interleukin-1 receptor/resistance protein (TIR) domains. Apoptosis 20, 250–261. Verstak, B., Arnot, C.J., Gay, N.J., 2013. An alanine-to-proline mutation in the BB-loop of TLR3 Toll/IL-1R domain switches signalling adaptor specificity from TRIF to MyD88. J. Immunol. 191, 6101–6109. Wang, D., Zhang, S., Li, L., Liu, X., Mei, K., Wang, X., 2010. Structural insights into the assembly and activation of IL-1beta with its receptors. Nat. Immunol. 11, 905–911. Watters, T.M., Kenny, E.F., O’Neill, L.A., 2007. Structure, function and regulation of the Toll/IL-1 receptor adaptor proteins. Immunol. Cell Biol. 85, 411–419. Wesche, H., Korherr, C., Kracht, M., Falk, W., Resch, K., Martin, M.U., 1997. The interleukin-1 receptor accessory protein (IL-1RAcP) is essential for IL-1-induced activation of interleukin-1 receptor-associated kinase (IRAK) and stress-activated protein kinases (SAP kinases). J. Biol. Chem. 272, 7727–7731. Xu, Y., Tao, X., Shen, B., Horng, T., Medzhitov, R., Manley, J.L., et al., 2000. Structural basis for signal transduction by the Toll/interleukin-1 receptor domains. Nature 408, 111–115. Yamamoto, M., Takeda, K., Akira, S., 2004. TIR domain-containing adaptors define the specificity of TLR signaling. Mol. Immunol. 40, 861–868. Zhang, X., Pino, G.M., Shephard, F., Kiss-Toth, E., Qwarnstrom, E.E., 2012. Distinct control of MyD88 adapter-dependent and Akt kinase-regulated responses by the interleukin (IL)-1RI co-receptor, TILRR. J. Biol. Chem. 287, 12348–12352. Zhang, X., Shephard, F., Kim, H.B., Palmer, I.R., McHarg, S., Fowler, G.J., et al., 2010. TILRR, a novel IL-1RI co-receptor, potentiates MyD88 recruitment to control Ras-dependent amplification of NF-kappaB. J. Biol. Chem. 285, 7222–7232.
Contents lists available at ScienceDirect
The International Journal of Biochemistry & Cell Biology journal homepage: www.elsevier.com/locate/biocel
Identification of critical regions within the TIR domain of IL-1 receptor type I Jürgen Radons a,∗ , Werner Falk a , Stefan Dove b a b
Department of Internal Medicine I, University Clinic Regensburg, D-93042 Regensburg, Germany Institute of Pharmacy, University of Regensburg, D-93040 Regensburg, Germany
a r t i c l e
i n f o
Article history: Received 3 June 2015 Received in revised form 5 August 2015 Accepted 11 August 2015 Available online 14 August 2015 Keywords: IL-1 IL-1RI TIR domain NF-B Site-directed mutagenesis Homology modeling
a b s t r a c t Interleukin-1 receptor type I (IL-1RI) belongs to a superfamily of proteins characterized by an intracellular Toll/IL-1 receptor (TIR) domain. This domain harbors three conserved regions called boxes 1-3 that play crucial roles in mediating IL-1 responses. Boxes 1 and 2 are considered to be involved in binding of adapter molecules. Amino acids possibly crucial for IL-1RI signaling were predicted via homology models of the IL-1RI TIR domain based on the crystal structure of IL-1RAPL. The role of ten of these residues was investigated by site-directed mutagenesis and a functional luciferase assay reflecting NFB activity in transiently transfected Jurkat cells. In particular, the mutants E437K/D438K, E472A/E473A and S465A/S470A/S471A/E472A/E473A showed decreased and the mutant E437A/D438A increased IL1 responsiveness compared to the mouse IL-1RI wild type. In conclusion, the ␣C helix (Q469-E473 in mouse IL-1RI) is probably involved in heterotypic interactions of IL-1RI with IL-1RAcP or MyD88. © 2015 Elsevier Ltd. All rights reserved.
1. Introduction IL-1 is a pleiotropic cytokine playing a crucial role in inflammation and regulation of immune responses thereby affecting almost any cell type (Garlanda et al., 2013). It is also involved in the regulation of other homeostatic functions and the pathogenesis of different diseases such as chronic inflammatory and neurodegenerative diseases, atherosclerosis, and cancer (Lukens et al., 2012; Multhoff et al., 2012). IL-1 is expressed by tumor, stromal and endothelial cells and the host‘s infiltrating immune cells (Multhoff et al., 2012). IL-1 signal transduction is initiated by binding of either form of IL-1 to IL-1 receptor type I (IL-1RI), which undergoes a conformational change allowing association with IL-1 receptor accessory protein (IL-1RAcP). IL-1RAcP does not bind IL-1 but represents an essential component in the IL-1 signaling pathway
Abbreviations: DEAE, diethylaminoethyl; FACS, fluorescence-activated cell sorting; FCS, fetal calf serum; IL, interleukin; IL-1RI, IL-1 receptor type I; IL-1RAcP, IL-1 receptor accessory protein; IL-1RAPL, IL-1RAcP-like; LPS, lipopolysaccharide; Mal, MyD88 adaptor-like protein; MyD88, myeloid differentiation primary response protein MyD88; NF-B, nuclear factor kappa B; PI3K, phosphatidylinositol 3-kinase; PBS, phosphate-buffered saline; PKB, protein kinase B; SI, stimulation index; TBS, Tris-buffered saline; TILRR, Toll-like and IL-1R regulator; TIR, Toll/IL-1 receptor; TLR, Toll-like receptor; WT, wild type. ∗ Corresponding author at: Mühldorfer Str. 64, D-84503 Altötting, Germany. Tel.: +49 86714034200. E-mail address: [email protected] (J. Radons). http://dx.doi.org/10.1016/j.biocel.2015.08.009 1357-2725/© 2015 Elsevier Ltd. All rights reserved.
(Korherr et al., 1997; Radons et al., 2002; Wesche et al., 1997). The crystal structure of the ectodomains of an IL-1/IL-1RI/IL-1RAcP ternary complex has been solved recently (Thomas et al., 2012). IL-1-mediated receptor heterodimerization leads to recruitment of the intracellular adapter MyD88 via TIR-TIR domain interactions culminating in activation of intracellular signaling cascades including NF-B (Boraschi and Tagliabue, 2013; Gay et al., 2011; Watters et al., 2007). Recently, a novel IL-1RI co-receptor called Toll-like and IL-1R regulator (TILRR) was identified to associate with IL-RI thus amplifying IL-1-mediated NF-B activation and inflammatory responses (Zhang et al., 2010; Zhang et al., 2012). IL-1 signaling also involves recruitment of phosphatidylinositol 3-kinase (PI3K) to the IL-1 receptor complex (Reddy et al., 1997) and subsequent activation of NF-B via AKT/PKB (Cahill and Rogers, 2008). IL-1 binding also activates numereous G proteins resulting in an IB␣independent NF-B transactivation (Jefferies and O’Neill, 2000; Singh et al., 1999). IL-1 signaling finally regulates gene expression of a great variety of other genes involved in immune defence, repair and others (Multhoff et al., 2012; O’Neill, 2000). For successful initiation of signaling, oligomerization of TIR domains and association of adapter molecules was shown to be required (Jain et al., 2014; Multhoff et al., 2012; Narayanan and Park, 2015; Ve et al., 2015). Within the TIR domain three typical regions of conservation can be found: box 1 (D-K-YDAF-SY), box 2 (GYKLCI–RD–PG), and box 3 (-FWKx-) (Bowie and O’Neill, 2000; O’Neill and Dinarello, 2000). Bold amino acids are identical in nearly all members while conservative substitutions may occur in the other positions. In
16
J. Radons et al. / The International Journal of Biochemistry & Cell Biology 68 (2015) 15–20
dominant negative TLR-4, the proline in box 2 is substituted by histidine in C3H/HeJ mice abolishing the host immune response to LPS (Poltorak et al., 1998). Only IL-1RI and TLR-3 do not harbor a proline at this position. Results on human IL-1RI mutants indicated the conserved arginine in box 2 of IL-1RI and the FWKxxRY sequence including box 3 to be crucial for IL-1 signaling (Heguy et al., 1992). By contrast, in another study single point alanine mutation of R, W and K (sequence RFWK) did not affect function (Slack et al., 2000). Boxes 1 and 2 are considered to be involved in binding of adapter molecules (Bowie and O’Neill, 2000; Slack et al., 2000). Our previous investigations identified certain amino acids within the loop preceding box 3 in IL-1RAcP as being crucial for IL-1 responsiveness (Radons et al., 2003). The present study was to identify amino acids of the IL-1RI TIR domain contributing to receptor signaling. Since the intracellular TIR domains are not part of the crystal structure of the IL-1/IL-1RI/IL-1RAcP heterotrimer (Wang et al., 2010), a homology model of the mouse IL-1RI TIR domain (mIL-1RI-TIR) based on the structure of the human IL-1R accessory protein-like (IL-1RAPL) (Khan et al., 2004) was generated and used for suggesting promising positions to be mutated. The role of these residues was identified by site-directed mutagenesis and a functional luciferase assay reflecting NF-B activity in transiently transfected Jurkat cells. 2. Materials and methods 2.1. Homology modeling of the mIL-1RI TIR domain An initial model of mIL-1RI-TIR was generated by the program COMPOSER (Blundell et al., 1988) as part of the molecular modeling suite SYBYL 7.3 (Certara, L.P., Princeton, NJ). Homology modeling is based on known crystal structures serving as templates for structurally conserved regions (SCRs). Among structures of TIR domains present in the Brookhaven Protein Databank (PDB) – human TLR-1 (PDB ID 1fyv), TLR-2 (1fyw) (Xu et al., 2000) and IL-1RAPL (1t3g) (Khan et al., 2004) – chain A of the latter was selected as single template since only in this case the degree of homology with mIL-1RI was sufficient (ca. 30% sequence identity compared to ca. 20% with TLR-1 and TLR-2). Fig. 1 presents the optimal sequence alignment and the SCRs determined by COMPOSER. In the present case of a single template structure, the backbone frameworks of the SCRs of mIL-1RI-TIR and of IL-1RAPL are identical. Side chains were added to the target SCR backbone
by a knowledge-based approach, taking in account the backbone secondary structure and the side chain conformation at the corresponding residues of the template. The structurally variable regions (SVRs) of the model were constructed by the loop search algorithm within SYBYL. For each SVR, appropriate fragments from a binary PDB database with the same length as the SVR of the target are proposed on the base of distances and superpositions of the anchor residues. After adding hydrogens, the model was energy minimized using the AMBER-FF99 force field (Cornell et al., 1995) with AMBERFF99 charges (distance dependent dielectricity constant 4, first 50 cycles with constrained backbone and steepest descent method) up to a root mean square gradient of 0.05 kcal/mol Å (Powell conjugate gradient). Surfaces and electrostatic potentials of the model were calculated and visualized by the program MOLCAD (MOLCAD GmbH, Darmstadt, Germany) implied in SYBYL 7.3. 2.2. Construction of the reporter plasmid and mutated forms of IL-1RI For determination of transcriptional activity of NF-B, five tandemly arranged NF-B binding sites from the HIV long terminal repeat enhancer (5 -GGGACTTTCC-3 ) (Sen and Baltimore, 1986) were cloned via BglII and HindIII into the luciferase reporter vector pGL3-Basic (pGL3–5xNF-B). Mutated forms of mIL-1RI were generated by site-directed mutagenesis using the QuickChangeTM site-directed mutagenesis kit (Stratagene, Amsterdam, The Netherlands). The vector pFLAG-IL-1RI, a derivative of pFLAG-CMV-1 (Sigma–Aldrich GmbH, Steinheim, Germany) was used as a template in all mutagenesis experiments. All mutations were verified by sequence analysis. Expression and protein production by all mutants was analyzed by FACS detection of the Flag epitope in transfected cells using anti-Flag antibody which was done with permeated cells. 2.3. Detection of NF-B activity in transiently transfected Jurkat cells The human T cell leukaemia cell line Jurkat, lacking a receptor for IL-1 (Kuno et al., 1993), was purchased from the American Type Culture Collection (ATCC; Manassas, USA) and maintained in RPMI 1640 medium (PAN Biotech GmbH, Aidenbach, Germany) containing 10% FCS, 100 U/ml penicillin, 100 mg/ml streptomycin and 2 mM glutamine at 37 ◦ C in humidified air with 5% CO2 . For
Fig. 1. Sequence alignment of TIR domains of hIL-1RI (UniProtKB P14778), mIL-1RI (P13504) and IL-1RAPL (Q9NZN1, the template for homology modeling). ␣-Helices (blue) and -sheets (green) are labeled according to Khan et al. (2004). Red frames indicate boxes 1, 2, and 3. mIL-1RI positions subjected to mutation are highlighted in yellow. SCR1–SCR5: structurally conserved regions determined by COMPOSER. 7x–seven additional residues not resolved in the IL-1RAPL structure (PDB ID 1t3g).
J. Radons et al. / The International Journal of Biochemistry & Cell Biology 68 (2015) 15–20
17
transfection, 100 l of cells (106 /ml) in TBS were mixed with 30 l of freshly prepared DEAE-dextran solution (10 mg/ml) and 470 l TBS with 1 g of vector DNA and 0.5 g of the luciferase reporter plasmid pGL3–5xNFB harboring five NF-B binding sites (Radons et al., 2002). The total amount of DNA in all transfections was kept constant by adding empty vector DNA. The mixture was incubated for 10 min with agitation and washed once with RPMI 1640 containing 10% FCS, mercaptoethanol (3 × 10−5 M) and gentamycin (50 g/ml). The pellet was resuspended in 1.5 ml of medium, and 50 l were used for stimulation with either recombinant human IL-1␣ (rhIL-1␣; Pan Biotech, Aidenbach, Germany; 1 or 10 ng/ml) or recombinant human TNF (rhTNF; 10 ng/ml) for 6 h or remained unstimulated. NF-B activation was determined by measuring the luciferase activity in transiently transfected cells using the LucLite Plus luciferase reporter gene assay kit (Perkin Elmer Life Sciences, Shelton, USA). Data from corresponding cultures from all experiments were pooled and the highest and the lowest value in all groups were deleted. Afterwards, a mean cpm was established for unstimulated cultures which was then used to convert all cpm to stimulation indices (SI, stimulated/mean unstimulated). A mean SI was then calculated for the WT transfectants stimulated with TNF and IL-1. This made it possible to determine an SI in % of WT for all values; (SI sample/mean SI WT) x 100). 2.4. Statistical analysis Level, spread, and symmetry of data distribution were conveyed by a box-and-whisker plot. Statistical differences between medians were analyzed using the two-tailed non-parametric WilcoxonMann-Whitney test. A P value of less than 0.05 was considered to reflect statistical significance. 3. Results and discussion 3.1. Structure-based selection of IL-1RI mutants The homology model of mIL-1RI-TIR based on the crystal structure of human IL-1RAPL (Khan et al., 2004) comprises all characteristic elements of TIR domains (see sequence alignment, Fig. 1, and structural overview, Fig. 2A). A five-strand parallel -sheet is surrounded by seven ␣-helices. Three typical segments assumed to be functionally significant, boxes 1-3, and moreover, the whole BB-loop belong to the structurally conserved regions (SCRs) and are probably well approximated by the mIL-1RI-TIR model. Differences in the structures of IL-1RAPL (Khan et al., 2004) on one hand as well as TLR-1 and TLR-2 on the other hand (Xu et al., 2000) indicate variable regions possibly not correctly reproduced by the model. In particular, (i) the short helices ␣C and ␣C are followed by a wellordered helix ␣C in the case of IL-1RAPL and by a highly disordered region in the case of TLR-1 and TLR-2, respectively, (ii) ␣C is slightly shifted in IL-1RAPL vs. TLR-1, and (iii) helix ␣D in TLR-1 and TLR-2 is nearly perpendicular to that in IL-1RAPL (Khan et al., 2004). The most striking specific feature of both human and mouse IL-1RI-TIR compared to TIR domains of IL-1RAPL, IL-1RAcP, TLR-1 and TLR-2 is a cluster of acidic amino acids at the C-terminus of the BB-loop and in helix ␣C (Fig. 2B). Additionally, three serines before (S465) and within ␣C (S470, S471) are typical. Moreover, ␣C and the BB-loop belong to the interaction interfaces in homodimers (asymmetric units) of IL-1RAPL (Khan et al., 2004), TLR-1 (Xu et al., 2000) and TLR-2 (Tao et al., 2002) although the crystallographic interfaces of these TIR domains are different and probably not biologically relevant. In a model based on the crystal structure of the dimeric TLR-10 TIR domain, dimerization of TLR-4 TIR domains was shown to require formation of a large conserved platform containing the BB-loop and box 1 motifs and thereby
Fig. 2. Homology model of the mIL-1RI TIR domain. Secondary structure elements are labeled and colored (␣-helices blue, -strands green). Specific conserved regions (boxes 1–3) are drawn in red. (A) Ribbon and tube representation of the tertiary structure. Ocher balls indicate mutated positions. (B) Detailed model of the region with all but one (S519) amino acids subjected to mutation (ocher C atoms). Other atom colors: O – red, N – blue, C – gray.
forming a potential binding site for intracellular adapters (Bovijn et al., 2012). Although the involvement of the BB-loop in TLR-4 TIR dimerization has been described previously, further data from this group suggest that this region establishes contact with helix ␣E of TLR-4, rather than helix ␣C as found in the structure of TLR-10 TIR crystals (Toshchakov et al., 2011). Noteworthy, the crystal structure of the TIR domain of TLR-6 recently revealed that helix ␣C rather than the BB-loop plays a crucial role in homodimeric TIR-TIR interactions (Jang and Park, 2014). Accordingly, homodimerization of Mal-TIR domains lacking the BB-loop involves residues of the ␣C and ␣D helices of both monomers (Lin et al., 2012), implying that the functional dimer does not inevitably represent the “most stable” among possible dimer conformations (Toshchakov et al., 2011). In Mal-TIR dimers, E190 (␣D) of one monomer forms a hydrogen bond with K158 (␣C ) of the other monomer, while P189 and F193 in ␣D of one monomer form stacking interactions with P155 and
18
J. Radons et al. / The International Journal of Biochemistry & Cell Biology 68 (2015) 15–20
Fig. 3. Effect of different mutant variants of IL-1RI on NF-B activation. Jurkat cells were transiently co-transfected with 500 ng of luciferase reporter and 1 g of the indicated IL-1RI constructs. 6 h after transfection, cells were stimulated with different concentrations of rhIL-1␣ or 10 ng/ml rhTNF. After incubation for an additional 6 h, luciferase activity was determined in triplicate as a measure of NF-B activation and normalized to the TNF response. Data were visualized by a box-and-whisker plot. The boxes show the 25th and 75th percentiles together with the medians while the whiskers show the 5th and 95th percentiles, respectively. Data represent the medians of three to six separate experiments performed in at least triplicates. *, p < 0.05 compared with wild type (WT).
W156 in ␣C of the other monomer, respectively (Lin et al., 2012). In the case of IL-1RAPL, S488 (␣C ) of one monomer forms a hydrogen bond with S488 of the other monomer, an interaction that is reproduced by a disulfide bridge of the corresponding cysteines in the TLR-1 dimer. Thus, it may be suggested that the whole BB-␣C face is involved in heterodimerization in the case of IL-1RI, too. Consequently, the specific acidic and polar amino acids in this region are particularly interesting for in-vitro mutagenesis experiments, considering also Q469 at the beginning of ␣C , the only amino acid of the whole BB-␣C face which differs between mouse and human IL-1RI (G466). The conserved proline in box 2 (BB-loop) of nearly all TIR domains is replaced by valine in IL-1RI species orthologs (mIL-1RI: V435). Substitution of proline by polar residues – histidine in TLR-4 (Poltorak et al., 1998), histidine and aspartate in IL-1RAcP (Radons et al., 2002) – impaired function, whereas an IL-1RAcP P446A mutation was without effect (Radons et al., 2002). Comparison of the BB-loops in IL-1RAPL and the mIL-1RI model suggests that proline and valine do not lead to different backbone conformations. 3.2. Effects of IL-1RI mutants on NF-B activation NF-B activation mediated by several intracellular signaling pathways was determined by measuring the luciferase activity of
our reporter plasmid pGL3–5xNFB harboring five NF-B binding sites (Radons et al., 2002). This NF-B activation assay does not distinguish between activation via the MyD88-dependent and AKT/PKB-dependent pathways. Significant changes in IL-1 responsiveness of transfected Jurkat cells were observed in eight out of 13 tested mutants (Fig. 3). As expected and as control, the mutant with the complete intracytoplasmatic region missing (Stop380) showed no response to IL-1. In case of the five mutants E437K, D438K, D438A, S519K and Q469P, responses were not significantly different from that of the wild type. The four mutants E437K/D438K, E437A, E472A/E473A and 5xA showed decreased IL-1 responsiveness, albeit in case of E437K/D438K and E437A only at the lower IL-1 concentration. Effects of the 5xA mutant were exceptionally strong. Increased responsiveness to IL-1 was observed for the mutants E437A/D438A, V435P and Q469A (for V435P and Q469A only at the higher IL-1 concentration). It has been shown previously that substitution of the conserved R within box 2 of IL-1RI by alanine caused a massive reduction in IL-1-induced NF-B activation (Slack et al., 2000). Moreover, this box 2 receptor mutant was found to strongly decrease IL-1mediated receptor activation after blockage of TILRR expression identifying box 2 as being a putative interaction site for this cytosolic adapter (Zhang et al., 2010). The observed decreased IL-1 responsiveness of our C-terminal box 2 mutants (E437K,
J. Radons et al. / The International Journal of Biochemistry & Cell Biology 68 (2015) 15–20
Fig. 4. Electrostatic potential mapped on a MOLCAD Connolly surface of the mIL-1RI TIR domain. The potential is based on AMBER-FF99 charges. Colors: most negative – violet, negative – blue, neutral – cyan, green, positive – yellow, most positive – brown. Labeled are amino acids subjected to mutation.
E437K/E438K) might rely, at least in part, on restrictions in TILRR recruitment. Since the E472A/E473A mutation reduced IL-1-induced NF-B activation by 25% and the 5xA variant even by 75%, the specific ␣C helix in mIL-1RI (and hIL-1RI) is functionally relevant. Fig. 4 presents the electrostatic potential map of the mIL-1RI model. The acidic nature of ␣C leads to a high negative potential. The region around helix ␣E including the conserved box 3 is positively charged due to four basic amino acids (K530, R532, K535, R538). Two of these basic amino acids (R537 and K540 corresponding to mIL1RI R532 and K535, respectively) are present in IL-1RAcP as well so that its EE-loop and box 3 are positively charged, too. The TIR domain of the adaptor protein MyD88 shows a similar charge distribution (Li et al., 2005). Our previous results on IL-1RAcP mutants indicated that the EE-loop and box 3 are involved in TIR domain interactions (Radons et al., 2002). Docking of IL-1RAcP and MyD88 based on homology models with human TLR-2 as template and in-vitro mutagenesis results suggested electrostatic interactions between the DRD-motif in box 2 (BB-loop) of MyD88 and the lysine patch in the EE-loop of IL-1RAcP (Li et al., 2005). Heterodimerization of IL-1RAcP and MyD88 precedes homotypic oligomerization of MyD88 (Akira, 2003; Akira and Sato, 2003; Yamamoto et al., 2004) again based on box 2-box 3 interactions (Li et al., 2005), and recruitment of downstream signaling transducers by the MyD88 oligomers. Common for all these TIR-TIR interactions is that a negatively charged pocket in the region of box 2 and ␣C/␣C may provide a binding site for protruding positively charged residues in or close to box 3. The shape of the corresponding regions of IL-1RI shows this general pattern, too (Fig. 4). Heterodimerization or oligomerization of IL-1RI with IL-1RAcP and/or MyD88 may rely on these interactions. Thus, IL-1RI is probably involved in transient or stable TIR-TIR interactions essential for downstream signaling. Strikingly, mIL-1RI Q469 in the IL-1R specific region (mIL-1RI amino acids 465, 467-473) may be exchanged by glycine (hIL-1RI G466). To investigate the significance of a polar amino acid at this position, Q469 (␣C ) was mutated to alanine and proline, respectively. Whereas Q469P was without any effect on NF-B activation, mutation of Q469 to alanine led to a significant increase of the IL-1
19
response at the higher IL-1 concentration, suggesting that a small neutral amino acid like glycine or alanine is more favorable than the polar glutamine in mIL-1RI. Due to the IL-1RI-specific amino acids E437 and D438, the Cterminus of the BB-loop contributes to the large negatively charged region spanned by parts of box 2, ␣C , the DD-loop and ␣D. Lysine and alanine mutation of these acidic amino acids behaved quite contradictory. Only E437A and the E437K/D438K double mutant led to a significant reduction in IL-1-induced NF-B activation, but in both cases only at the lower IL-1 concentration. Surprisingly, the double-A mutant even induced increased effects at both 1 and 10 ng/ml IL-1. Thus, a negative potential at the C-terminal end of the BB-loop is not necessary for IL-1RI functionality. At least one basic amino acid is well tolerated. Among the members of the TIR family, a highly conserved proline can be found in box 2. This residue corresponds to a naturally occurring missense point mutation resulting in the substitution of P712 by histidine (the LPSd mutation) in dominant negative TLR-4 in C3H/HeJ mice selectively leading to LPS hyporesponsiveness (Poltorak et al., 1998). Equivalent substitutions in other TLR molecules also abrogate inflammatory responses (Ozinsky et al., 2000; Underhill et al., 1999). In contrast to the other TIR family members, the conserved proline as part of box 2 within the BBloop is substituted by valine in IL-1RI (mIL-RI V435). Exchange of V435 by proline improved IL-1-mediated NF-B activation, but only at the higher IL-1 concentration. In accordance with the model in Fig. 2, the V435P mutation probably stabilizes the BB-loop without major structural effects. Interestingly, TLR-3 bears an alanine (A795) in place of the conserved BB-loop proline. Mutation of A795 to proline also improved NF-B activation (Verstak et al., 2013) most likely due to the proposed effects on BB-loop stability thereby modulating recruitment of downstream adapter molecules. In nearly all TIR domains, a conserved WX motif can be found at the E-EE junction with WK in IL-1RAcP, WH in IL-1RAPL and WS in IL-1RI. To test for specificity, S519 in mIL-1RI was mutated to lysine thereby reconstituting the IL-1RAcP motif. This mutation did not affect IL-1-induced NF-B activation. In our previous study on IL-1RAcP, mutation of the conserved W526 to phenylalanine and the neighbouring K527 to glycine was without effect on IL1 responsiveness, too (Radons et al., 2002). Taken together, these findings argue against a crucial role of this motif within the NF-B signaling process. In conclusion, site-directed mutagenesis and a functional luciferase assay reflecting NF-B activity in transiently transfected Jurkat cells identified the specific ␣C helix of the IL-1RI to be crucial for signaling. Homology modeling of the mouse IL-1RI also indicated putative TIR-TIR domain interfaces for heterotypic interactions. The negative electrostatic potential and the shape of this region including the BB-loop (box 2) predestine interactions with the positively charged area around box 3 of IL-1RAcP or MyD88. The acidic C-terminal end of the BB-loop is not necessary with respect to function and dimerization. Back mutation of valine in the BB-loop into the highly conserved proline of other TIR domains did only slightly improve NF-B activity, indicating minor stabilization of the BB-loop. Acknowledgements pFLAG-IL-1RI was a generous gift from Michael Martin (Gießen, Germany). The excellent technical assistance of Daniela Aigner is gratefully acknowledged. References Akira, S., 2003. Toll-like receptor signaling. J. Biol. Chem. 278, 38105–38108.
20
J. Radons et al. / The International Journal of Biochemistry & Cell Biology 68 (2015) 15–20
Akira, S., Sato, S., 2003. Toll-like receptors and their signaling mechanisms. Scand. J. Infect. Dis. 35, 555–562. Blundell, T., Carney, D., Gardner, S., Hayes, F., Howlin, B., Hubbard, T., et al., 1988. 18th Sir Hans Krebs lecture. Knowledge-based protein modelling and design. Eur. J. Biochem. 172, 513–520. Boraschi, D., Tagliabue, A., 2013. The interleukin-1 receptor family. Semin. Immunol. 25, 394–407. Bovijn, C., Ulrichts, P., De Smet, A.S., Catteeuw, D., Beyaert, R., Tavernier, J., et al., 2012. Identification of interaction sites for dimerization and adapter recruitment in Toll/interleukin-1 receptor (TIR) domain of Toll-like receptor 4. J. Biol. Chem. 287, 4088–4098. Bowie, A., O’Neill, L.A., 2000. The interleukin-1 receptor/Toll-like receptor superfamily: signal generators for pro-inflammatory interleukins and microbial products. J. Leukoc. Biol. 67, 508–514. Cahill, C.M., Rogers, J.T., 2008. Interleukin (IL) 1beta induction of IL-6 is mediated by a novel phosphatidylinositol 3-kinase-dependent AKT/IkappaB kinase alpha pathway targeting activator protein-1. J. Biol. Chem. 283, 25900–25912. Cornell, W.D., Cieplak, P., Bayly, C.I., Gould IRMKM, Ferguson, D.M., Spellmeyer, D.C., et al., 1995. A second generation force field for the simulation of proteins, nucleic acids, and organic molecules. J. Am. Chem. Soc. 117, 5179–5197. Garlanda, C., Dinarello, C.A., Mantovani, A., 2013. The interleukin-1 family: back to the future. Immunity 39, 1003–1018. Gay, N.J., Gangloff, M., O’Neill, L.A., 2011. What the Myddosome structure tells us about the initiation of innate immunity. Trends Immunol. 32, 104–109. Heguy, A., Baldari, C.T., Macchia, G., Telford, J.L., Melli, M., 1992. Amino acids conserved in interleukin-1 receptors (IL-1Rs) and the Drosophila toll protein are essential for IL-1R signal transduction. J. Biol. Chem. 267, 2605–2609. Jain, A., Kaczanowska, S., Davila, E., 2014. IL-1 receptor-associated kinase signaling and its role in inflammation, cancer progression, and therapy resistance. Front. Immunol. 5, 553. Jang, T.H., Park, H.H., 2014. Crystal structure of TIR domain of TLR6 reveals novel dimeric interface of TIR-TIR interaction for toll-like receptor signaling pathway. J. Mol. Biol. 426, 3305–3313. Jefferies, C.A., O’Neill, L.A., 2000. Rac1 regulates interleukin 1-induced nuclear factor kappaB activation in an inhibitory protein kappaBalpha-independent manner by enhancing the ability of the p65 subunit to transactivate gene expression. J. Biol. Chem. 275, 3114–3120. Khan, J.A., Brint, E.K., O’Neill, L.A., Tong, L., 2004. Crystal structure of the Toll/interleukin-1 receptor domain of human IL-1RAPL. J. Biol. Chem. 279, 31664–31670. Korherr, C., Hofmeister, R., Wesche, H., Falk, W., 1997. A critical role for interleukin-1 receptor accessory protein in interleukin-1 signaling. Eur. J. Immunol. 27, 262–267. Kuno, K., Okamoto, S., Hirose, K., Murakami, S., Matsushima, K., 1993. Structure and function of the intracellular portion of the mouse interleukin 1 receptor (type I). Determining the essential region for transducing signals to activate the interleukin 8 gene. J. Biol. Chem. 268, 13510–13518. Li, C., Zienkiewicz, J., Hawiger, J., 2005. Interactive sites in the MyD88 Toll/interleukin (IL) 1 receptor domain responsible for coupling to the IL1beta signaling pathway. J. Biol. Chem. 280, 26152–26159. Lin, Z., Lu, J., Zhou, W., Shen, Y., 2012. Structural insights into TIR domain specificity of the bridging adaptor Mal in TLR4 signaling. PLoS ONE 7, e34202. Lukens, J.R., Gross, J.M., Kanneganti, T.D., 2012. IL-1 family cytokines trigger sterile inflammatory disease. Front. Immunol. 3, 315. Multhoff, G., Molls, M., Radons, J., 2012. Chronic inflammation in cancer development. Front. Immunol. 2, 98. Narayanan, K.B., Park, H.H., 2015. Toll/interleukin-1 receptor (TIR) domain-mediated cellular signaling pathways. Apoptosis 20, 196–209. O’Neill, L.A., 2000. The interleukin-1 receptor/Toll-like receptor superfamily: signal transduction during inflammation and host defense. Sci STKE 2000, re1. O’Neill, L.A., Dinarello, C.A., 2000. The IL-1 receptor/toll-like receptor superfamily: crucial receptors for inflammation and host defense. Immunol. Today 21, 206–209. Ozinsky, A., Underhill, D.M., Fontenot, J.D., Hajjar, A.M., Smith, K.D., Wilson, C.B., et al., 2000. The repertoire for pattern recognition of pathogens by the innate
immune system is defined by cooperation between toll-like receptors. Proc. Natl. Acad. Sci. USA 97, 13766–13771. Poltorak, A., He, X., Smirnova, I., Liu, M.Y., Van Huffel, C., Du, X., et al., 1998. Defective LPS signaling in C3H/HeJ and C57BL/10ScCr mice: mutations in Tlr4 gene. Science 282, 2085–2088. Radons, J., Dove, S., Neumann, D., Altmann, R., Botzki, A., Martin, M.U., et al., 2003. The interleukin 1 (IL-1) receptor accessory protein Toll/IL-1 receptor domain: analysis of putative interaction sites by in vitro mutagenesis and molecular modeling. J. Biol. Chem. 278, 49145–49153. Radons, J., Gabler, S., Wesche, H., Korherr, C., Hofmeister, R., Falk, W., 2002. Identification of essential regions in the cytoplasmic tail of interleukin-1 receptor accessory protein critical for interleukin-1 signaling. J. Biol. Chem. 277, 16456–16463. Reddy, S.A., Huang, J.H., Liao, W.S., 1997. Phosphatidylinositol 3-kinase in interleukin 1 signaling. Physical interaction with the interleukin 1 receptor and requirement in NFkappaB and AP-1 activation. J. Biol. Chem. 272, 29167–29173. Sen, R., Baltimore, D., 1986. Multiple nuclear factors interact with the immunoglobulin enhancer sequences. Cell 46, 705–716. Singh, R., Wang, B., Shirvaikar, A., Khan, S., Kamat, S., Schelling, J.R., et al., 1999. The IL-1 receptor and Rho directly associate to drive cell activation in inflammation. J. Clin. Invest. 103, 1561–1570. Slack, J.L., Schooley, K., Bonnert, T.P., Mitcham, J.L., Qwarnstrom, E.E., Sims, J.E., et al., 2000. Identification of two major sites in the type I interleukin-1 receptor cytoplasmic region responsible for coupling to pro-inflammatory signaling pathways. J. Biol. Chem. 275, 4670–4678. Tao, X., Xu, Y., Zheng, Y., Beg, A.A., Tong, L., 2002. An extensively associated dimer in the structure of the C713S mutant of the TIR domain of human TLR2. Biochem. Biophys. Res. Commun. 299, 216–221. Thomas, C., Bazan, J.F., Garcia, K.C., 2012. Structure of the activating IL-1 receptor signaling complex. Nat. Struct. Mol. Biol. 19, 455–457. Toshchakov, V.Y., Szmacinski, H., Couture, L.A., Lakowicz, J.R., Vogel, S.N., 2011. Targeting TLR4 signaling by TLR4 Toll/IL-1 receptor domain-derived decoy peptides: identification of the TLR4 Toll/IL-1 receptor domain dimerization interface. J. Immunol. 186, 4819–4827. Underhill, D.M., Ozinsky, A., Hajjar, A.M., Stevens, A., Wilson, C.B., Bassetti, M., et al., 1999. The Toll-like receptor 2 is recruited to macrophage phagosomes and discriminates between pathogens. Nature 401, 811–815. Ve, T., Williams, S.J., Kobe, B., 2015. Structure and function of Toll/interleukin-1 receptor/resistance protein (TIR) domains. Apoptosis 20, 250–261. Verstak, B., Arnot, C.J., Gay, N.J., 2013. An alanine-to-proline mutation in the BB-loop of TLR3 Toll/IL-1R domain switches signalling adaptor specificity from TRIF to MyD88. J. Immunol. 191, 6101–6109. Wang, D., Zhang, S., Li, L., Liu, X., Mei, K., Wang, X., 2010. Structural insights into the assembly and activation of IL-1beta with its receptors. Nat. Immunol. 11, 905–911. Watters, T.M., Kenny, E.F., O’Neill, L.A., 2007. Structure, function and regulation of the Toll/IL-1 receptor adaptor proteins. Immunol. Cell Biol. 85, 411–419. Wesche, H., Korherr, C., Kracht, M., Falk, W., Resch, K., Martin, M.U., 1997. The interleukin-1 receptor accessory protein (IL-1RAcP) is essential for IL-1-induced activation of interleukin-1 receptor-associated kinase (IRAK) and stress-activated protein kinases (SAP kinases). J. Biol. Chem. 272, 7727–7731. Xu, Y., Tao, X., Shen, B., Horng, T., Medzhitov, R., Manley, J.L., et al., 2000. Structural basis for signal transduction by the Toll/interleukin-1 receptor domains. Nature 408, 111–115. Yamamoto, M., Takeda, K., Akira, S., 2004. TIR domain-containing adaptors define the specificity of TLR signaling. Mol. Immunol. 40, 861–868. Zhang, X., Pino, G.M., Shephard, F., Kiss-Toth, E., Qwarnstrom, E.E., 2012. Distinct control of MyD88 adapter-dependent and Akt kinase-regulated responses by the interleukin (IL)-1RI co-receptor, TILRR. J. Biol. Chem. 287, 12348–12352. Zhang, X., Shephard, F., Kim, H.B., Palmer, I.R., McHarg, S., Fowler, G.J., et al., 2010. TILRR, a novel IL-1RI co-receptor, potentiates MyD88 recruitment to control Ras-dependent amplification of NF-kappaB. J. Biol. Chem. 285, 7222–7232.