IL-1R-associated kinase (IRAK) 4 is an essential component of innate immunity. IRAK-4 deficiency in mice and humans results in severe impairment of IL-1 and TLR signaling. We have solved the crystal structure for the death domain of Mus musculus IRAK-4 to 1.7 Å resolution. This is the first glimpse of the structural details of a mammalian IRAK family member. The crystal structure reveals a six-helical bundle with a prominent loop, which among IRAKs and Pelle, a Drosophila homologue, is unique to IRAK-4. This highly structured loop contained between helices two and three, comprises an 11-aa stretch. Although innate immune domain recognition is thought to be very similar between Drosophila and mammals, this structural component points to a drastic difference. This structure can be used as a framework for future mutation and deletion studies and potential drug design.

The mammalian TLRs and the IL-1R are known to play important roles in innate immunity and are characterized by a homologous cytoplasmic Toll/IL-1R domain (1). Disruption of this domain, as in the P712H mutation of Tlr4 (2), where mice are completely resistant to endotoxic shock, perturbs downstream signaling and the production of proinflammatory cytokines. In addition, elimination of downstream signaling molecules can either severely hamper or totally disturb an effective innate immune response (3). These downstream signaling molecules include a set of Toll/IL-1R-domain containing adaptor proteins: MyD88, TIRAP/Mal, Trif/Ticam, and TRAM (4). MyD88 has a modular construction with an N-terminal death domain (DD)3 (5). This domain serves to recruit a family of kinases known as IL-1R- associated kinases (IRAKs). The IRAK family has four known members: IRAK-1 (6), IRAK-2 (7), IRAK-M (8), and IRAK-4 (9). All four IRAKs contain a DD at the N terminus and a kinase domain at the C terminus; however, only IRAK-1 (6) and IRAK-4 (9) have been shown to have kinase activity. IRAK-4 is a unique member of the IRAK family in that it is the only member to have true kinase activity and is the most homologous to the Drosophila IRAK-like kinase, Pelle (3). The central role IRAK-4 plays in TLR/IL-1R signaling and subsequent immunological protection was demonstrated in mouse models (10) and in humans with inherited IRAK-4 deficiencies, who suffer from recurrent pyogenic bacterial infections (11, 12). Modulation of this key mediator in TLR/IL-1R signaling could prove useful in treating disease states of inflammation, septic shock, and in the management of autoimmune disorders, making IRAK-4 an attractive drug target (3, 13). To mediate TLR/IL-1R signaling, IRAK-4 associates with the intermediate domain of the adaptor MyD88 presumably via its own DD (3, 5). Failure to associate with MyD88 disrupts phosphorylation of IRAK-1 (5) by IRAK-4 and subsequent downstream signaling. The DD of IRAK-4 is also necessary in facilitating heterodimerization with other IRAKs (9). To begin to understand the molecular structure of IRAK-4 and its implications for TLR/IL-1R signaling, we solved the crystal structure of the IRAK-4 DD to 1.7 Å resolution. Using the crystal structure as a guide, we describe some important structural components of the IRAK-4 DD, which have implications for its function, molecular recognition, and subsequent TLR/IL-1R signaling.

A cDNA construct encoding Mus musculus IRAK-4 DD (residues 1–113) flanked by NdeI and BamHI restriction sites was created by PCR amplification and cloned into a pGEX6p-1 vector (Amersham Biosciences) in frame with an N-terminal GST tag. The construct was transformed into BL21 (DE3) cells. Transformed cells were grown in Luria broth to an A600 of 0.6–0.8 at 37°C, where overexpression was induced with 0.5 mM isopropyl β-d-thiogalactoside and the temperature was adjusted to 18°C for overnight expression. Purification of GST-IRAK-4 DD from the soluble fraction was conducted by affinity chromatography. Cleavage and removal of the GST tag from IRAK-4 DD was performed by the addition of rhinovirus 3C protease and cation exchange chromatography. Buffer exchange into 20 mM HEPES (pH 7.5) and 125 mM KCl, and subsequent concentration were conducted using a 5-kDa molecular mass limit centrifugal filter device (Millipore).

Purified and concentrated (10 mg/ml) IRAK-4 DD was crystallized using the hanging drop method. Crystals were grown at 8°C in 2.5-μl drops in 100 mM HEPES (pH 7.5), 25% polyethylene glycol 3350, and using 10 mM MnCl2 as an additive. Crystals took ∼5 days to form.

IRAK-4 DD crystals were cryoprotected by sequential 30-min soaks at 4°C in mother liquor containing 10, 20, and 30% glycerol and then plunged into liquid nitrogen. For halide-soaked crystals, the above process was conducted in addition to a brief soak in the last cryoprotectant solution containing 1 M sodium bromide.

Native amplitudes were collected, under standard cryogenic conditions, at a high-flux insertion device synchrotron beam line (Argonne National Laboratory, sector 14-ID), to a limiting resolution of 1.7 Å. A multiwavelength anomalous dispersion (MAD) data set was collected on bromide-soaked crystals (Argonne National Laboratory, sector 32-ID). The halide soak resulted in the specific incorporation of four ordered bromide ions into the structured solvent region and anomalous scattering from these bromide ions were exploited for phase determination.

The substructure of bromide ions present in the asymmetric unit was determined using Shake and Bake (〈www.hwi.buffalo.edu/SnB〉) (14). These bromide positions were used for heavy atom refinement and to derive estimates of protein phases using SOLVE (〈www.solve.lanl.gov〉) (15). The overall figure of merit (FOM) calculated to a resolution of 2.0 Å was 0.41 (Table I). Density modification and solvent flattening using RESOLVE (〈www.solve.lanl.gov〉) (15) dramatically improved the quality of the phase estimates (mean FOM = 0.67) as judged by the interpretability of the resultant electron density maps.

Table I.

Statistics for native data collection and refinement, MAD data collection, and phasing

Cell parameters     
 Space group  C2   
 Unit cell dimensions  a = 76.1 Å, b = 38.0 Å, c = 51.6 Å   
Data collection statisticsa     
 Native λ1 λ2 λ3 
 Wavelength (Å) 0.97888 0.91947 0.91919 0.91270 
 Resolution range (Å) 50–1.7 35–2.0 50–2.0 50–2.0 
 Unique reflections 11,988 8,336 8,382 8,441 
 Redundancy 6.1 7.4 7.5 7.5 
 Completeness (%) 100 (100) 95.0 (95.4) 95.0 (95.5) 95.4 (94.8) 
 I/ς 50.6 (8.9) 39.0 (23.8) 45.7 (23.5) 51.2 (21.4) 
Rmergeb (%) 3.2 (19.6) 3.6 (8.6) 3.2 (8.8) 2.9 (9.8) 
 Mosaicity 0.4–0.5    
Bromide MAD phases     
 Number of bromide sites    
 FOM   0.41  
 FOM following density modifications   0.67  
Model statistics     
 Resolution   50–1.7  
Rwork/Rfreec   0.1767/0.2226  
 rmsd bond lengths (Å)   0.017  
 rmsd bond angles (°)   1.719  
Ramachandran plot     
 Most favored φ-ψ (%)   92.1  
 Additionally allowed   7.9  
 Generously allowed   0.0  
 Disallowed   0.0  
Average B factor (Å2  28.9  
Number of nonhydrogen protein atoms per asymmetric unit   823  
Mn2+ ions per asymmetric unit    
Water molecules per asymmetric unit   145  
Cell parameters     
 Space group  C2   
 Unit cell dimensions  a = 76.1 Å, b = 38.0 Å, c = 51.6 Å   
Data collection statisticsa     
 Native λ1 λ2 λ3 
 Wavelength (Å) 0.97888 0.91947 0.91919 0.91270 
 Resolution range (Å) 50–1.7 35–2.0 50–2.0 50–2.0 
 Unique reflections 11,988 8,336 8,382 8,441 
 Redundancy 6.1 7.4 7.5 7.5 
 Completeness (%) 100 (100) 95.0 (95.4) 95.0 (95.5) 95.4 (94.8) 
 I/ς 50.6 (8.9) 39.0 (23.8) 45.7 (23.5) 51.2 (21.4) 
Rmergeb (%) 3.2 (19.6) 3.6 (8.6) 3.2 (8.8) 2.9 (9.8) 
 Mosaicity 0.4–0.5    
Bromide MAD phases     
 Number of bromide sites    
 FOM   0.41  
 FOM following density modifications   0.67  
Model statistics     
 Resolution   50–1.7  
Rwork/Rfreec   0.1767/0.2226  
 rmsd bond lengths (Å)   0.017  
 rmsd bond angles (°)   1.719  
Ramachandran plot     
 Most favored φ-ψ (%)   92.1  
 Additionally allowed   7.9  
 Generously allowed   0.0  
 Disallowed   0.0  
Average B factor (Å2  28.9  
Number of nonhydrogen protein atoms per asymmetric unit   823  
Mn2+ ions per asymmetric unit    
Water molecules per asymmetric unit   145  
a

Numbers in parentheses represent values for the highest resolution shell.

b

Rmerge = Σhkl |〈I〉 − I|/Σhkl |I|.

c

Rwork = Σhkl|FoFc|/ Σhkl |Fo|, and Rfree is equivalent to Rwork but is calculated for a randomly chosen 7% of reflections excluded from model refinement.

Model building with XtalView (16) was guided by prior knowledge that the structure likely contained a six-helical bundle as seen in structures of other DDs. REFMAC (17) was used for maximum-likelihood structure refinement. Several cycles of model building, interspersed with refinement, resulted in a final model containing residues 4–106 of IRAK-4, 145 well-ordered solvent molecules, and a single manganese ion. The geometry of all residues falls within the allowed regions of the Ramachandran plot. The structure has been refined to a final R factor of 17.7% (free R factor of 22.3%) against data to a resolution of 1.7 Å (Table I).

Atomic coordinates for IRAK-4 DD have been deposited within the Protein Data Bank and can be downloaded using the accession code 2A9I.

DDs typically consist of a hexahelical bundle (18). Because there is no signature motif that indicates initiation or termination of the DD, we performed a primary sequence alignment with other IRAK family members to find a region of conservation in IRAK-4 that might be amenable to crystallization. This primary sequence alignment suggested that the initial 113 aa of IRAK-4 represents the core structure of the DD. The resulting crystal structure revealed no electron density for the first three and last seven amino acids, indicating these regions are disordered and unstructured. This suggests that amino acid residues 4–106 represent the true core structure of the IRAK-4 DD. The structure of the IRAK-4 DD did indeed reveal a hexahelical bundle with the first α helix beginning with valine 16, designated α1-1 (Fig. 1,A). This residue numbering system will be used throughout. A highly structured N-terminal tail containing two proline residues (P4 and P7) precedes this helix. The N-terminal tail also makes several hydrogen bond contacts with the loop located between α4 and α5. A rigid tail was also observed at the C terminus with two proline residues (P102 and P106). These C-terminal proline residues are strictly conserved in all mouse IRAKs, Pelle (Fig. 1 B), and human IRAK-4 and IRAK-1. The rigid nature of proline residues is potentially required for orienting the C-terminal kinase domain so that proper molecular recognition can take place.

FIGURE 1.

Schematic representation of the hexahelical bundle formed by IRAK-4 DD and a primary sequence alignment. A, A ribbon diagram of IRAK-4 DD with the tail regions and interconnecting loops denoted in gray. N and C termini are represented by the letters N and C, respectively. This ribbon diagram was generated with PyMOL (DeLano Scientific, 〈www.pymol.org〉). B, Protein sequences of murine IRAK-4, murine IRAK-1, murine IRAK-2, murine IRAK-M, and Drosophila melanogaster Pelle were aligned using ClustalW (24 ). Amino acid residues strictly conserved throughout all five sequences are highlighted in magenta. Identical residues present in at least three of the aligned sequences are represented in yellow. Similar residues are tinted in blue and gaps are denoted by dashes. A schematic representation of the structural components of IRAK-4 DD is shown above the alignment. Dashed lines denote unstructured regions of IRAK-4 DD, whereas solid lines demonstrate structured tails or loops. The cylinders correspond to the helices with the same color scheme given in the IRAK-4 DD ribbon diagram.

FIGURE 1.

Schematic representation of the hexahelical bundle formed by IRAK-4 DD and a primary sequence alignment. A, A ribbon diagram of IRAK-4 DD with the tail regions and interconnecting loops denoted in gray. N and C termini are represented by the letters N and C, respectively. This ribbon diagram was generated with PyMOL (DeLano Scientific, 〈www.pymol.org〉). B, Protein sequences of murine IRAK-4, murine IRAK-1, murine IRAK-2, murine IRAK-M, and Drosophila melanogaster Pelle were aligned using ClustalW (24 ). Amino acid residues strictly conserved throughout all five sequences are highlighted in magenta. Identical residues present in at least three of the aligned sequences are represented in yellow. Similar residues are tinted in blue and gaps are denoted by dashes. A schematic representation of the structural components of IRAK-4 DD is shown above the alignment. Dashed lines denote unstructured regions of IRAK-4 DD, whereas solid lines demonstrate structured tails or loops. The cylinders correspond to the helices with the same color scheme given in the IRAK-4 DD ribbon diagram.

Close modal

Previous studies of IRAK-1 have shown residues T66 and W73, both of which are strictly conserved in the IRAK family (Fig. 1 B), to serve important functional roles (19). The overexpression of wild-type IRAK-1 has been shown to spontaneously activate NF-κB (8), whereas IRAK-1 point mutants T66A, T66E, and W73A all show reduced spontaneous activation (19). In addition, the T66A and T66E point mutations were shown to change the intracellular localization of IRAK-1 (19). Confocal microscopy demonstrated wild-type IRAK-1 to be present in punctate cytoplasmic complexes, whereas the T66A and T66E IRAK-1 mutants were found to be diffuse in the cytoplasm (19). These results indicate that perturbation of T66 abrogates the ability of IRAK-1 to form high molecular mass complexes. Mutation of T66 also prevents the normal autophosphorylation activity of IRAK-1 (19).

Given the profound functional consequences of these IRAK-1 mutations, we examined whether corresponding residues in IRAK-4 DD represent critical structural components. Residue W73 in IRAK-1 corresponds to W74, α4-8, in the IRAK-4 DD and serves as a central residue in the hydrophobic core. This hydrophobic core is made up of I11, L19, L22, L35, I39, L70, L71, L84, and L87. Among these residues, L71 and L84 are strictly conserved throughout the IRAK family and in Pelle kinase. The other residues are homologous aliphatic residues (Fig. 1 B). Disruption of this hydrophobic core by a W74-directed point mutation would easily perturb the association of helix 4 with helices 2 and 5. W74 acts as a central residue mediating interactions between many helices in the IRAK-4 DD, whereas T67, corresponding to residue T66 of IRAK-1 (19), has a very different role in maintaining structural integrity. T67, α4-2, instead forms a hydrogen bond to D27, α1-12, effectively linking the end of the first helix to the beginning of the fourth helix. This aspartate residue is strictly conserved in the IRAK family and Pelle. The conservation of a negatively charged residue in this position is a critical structural component that maintains hydrogen bonds with the hydroxylated side chain and the backbone of the fourth helix. However, introduction of a negatively charged side chain at T67 (as in a T67E/D mutation) would not only disrupt this hydrogen bond, but also cause a charge-charge repulsion with the D27 residue and thus destabilize the fourth helix. An electrostatic repulsion such as this could be initiated by a negatively charged posttranslational modification such as phosphorylation. In fact, IRAK-1-derived peptides containing the T66 residue have been shown to be phosphorylated by ι protein kinase C (20) using an in vitro kinase assay. Considering the functional roles the corresponding residue T66 assumes in IRAK-1 and the critical structural role T67 plays in the IRAK-4 DD, this amino acid position is a putative site of posttranslational modification.

All four IRAKs are homologues of the Drosophila protein kinase Pelle. Similar to the IRAK family, Pelle has been shown to play an important role in host defense. Pelle also is a bipartite molecule with an N-terminal DD and a C-terminal serine-threonine kinase domain. The crystal structure of the Pelle DD has been solved previously (21) and forms a six-helical bundle indicative of a member of the DD super family. Although the DD of Pelle and IRAK-4 have only a 13% sequence identity, their three-dimensional structures are highly homologous (Fig. 2). When the hexahelical bundles of the two structures are superimposed on each other, the core root-mean-square deviation is calculated to be just 1.97 Å. However, the similarity between these two structures is met with one striking exception, that of a loop between helices two and three. This sizeable loop in the IRAK-4 DD structure, designated α2-α3, appears quite small in the Pelle-DD structure (Fig. 2). The loop encompasses an 11-aa stretch, residues 39–49, and is highly structured by maintaining five pairs of polar contacts within the loop itself. Upon further analysis using a sequence alignment of IRAK-4 with other IRAK family members, it became evident that this loop is unique to IRAK-4 (Fig. 1 B).

FIGURE 2.

Structural comparison of Pelle-DD and IRAK-4 DD. A ribbon diagram of Pelle-DD, IRAK-4 DD, and Tube-DD is represented in orange, cyan, and blue, respectively. The structural data file for Pelle and Tube was taken from the Protein Data Bank (1D2Z) and the figure was generated with PyMOL (DeLano Scientific, 〈www.pymol.org〉). The α2-α3 loop in Pelle-DD and IRAK-4 DD is colored in magenta. A red dashed ellipse represents the area the IRAK-4 α2-α3 loop would encompass if it were superimposed on Pelle.

FIGURE 2.

Structural comparison of Pelle-DD and IRAK-4 DD. A ribbon diagram of Pelle-DD, IRAK-4 DD, and Tube-DD is represented in orange, cyan, and blue, respectively. The structural data file for Pelle and Tube was taken from the Protein Data Bank (1D2Z) and the figure was generated with PyMOL (DeLano Scientific, 〈www.pymol.org〉). The α2-α3 loop in Pelle-DD and IRAK-4 DD is colored in magenta. A red dashed ellipse represents the area the IRAK-4 α2-α3 loop would encompass if it were superimposed on Pelle.

Close modal

Because the IRAK-4 DD three-dimensional structure is nearly identical with the Pelle-DD structure, excluding the sizeable α2-α3 loop, we decided to analyze the Pelle-DD/Tube-DD structure in more detail. Tube is an adaptor molecule in Drosophila, which is known to bind to Pelle, and plays a critical role in mediating an innate immune response toward fungal infection (22). From the Pelle-DD/Tube-DD complex crystal structure, solved previously (21), it was shown that the C-terminal tail of Tube fits within a groove of Pelle (Fig. 2). This groove is formed by a cavity between the Pelle α4-α5 and α2-α3 loops. Both loops are rather small and lack any secondary structure or polar contact pairs within the loops. In comparison, the α2-α3 loop of IRAK-4 DD has a twist and five polar contact pairs within the loop. The twist within the α2-α3 loop of IRAK-4 DD precludes the presence of any analogous groove or cavity as in Pelle (Fig. 2). Thus, a steric clash with the α2-α3 loop would prevent an interaction like Pelle-DD/Tube-DD taking place with the IRAK-4 DD. Although no mammalian counterpart has been found for Tube, Pelle is most homologous to IRAK-4 (9). IRAK-4 is known to bind to a number of different molecules including: MyD88, IRAK-1, TNF-related activation factor 6, and Pellino-1, a protein known to play a role in IL-1-mediated signaling (23). If indeed this cognate adaptor interface is evolutionarily conserved in IRAK-4 from Pelle, then there are two obvious possibilities for interaction to take place. One is that the adaptor molecule for IRAK-4 has an interface, which accommodates the sizeable α2-α3 loop. An alternative is that the α2-α3 loop could undergo a conformational change. However, this may require breaking many polar contacts and considerable rearrangement.

The results of these structural studies show that the core structure of the IRAK-4 DD comprises amino acid residues 4–106 and forms a hexahelical bundle. The hexahelical bundle is followed by two conserved proline residues, P102 and P106, whereas the hydrophobic core is centralized around W74. The end of the first helix is tethered to the beginning of the fourth helix by means of a hydrogen bond between D27 and T67. A loop, α2-α3, unique to IRAK-4 within the IRAK family, forms a twist and five pairs of polar contacts with itself. This twist reduces the possibility that a groove is formed as in the Pelle crystal structure. Given that humans with mutations in IRAK-4 are susceptible to recurrent pyogenic infections, these results have considerable implications on innate immune defense in mammals and will hopefully serve as a guide for future biochemical and cellular studies.

We thank the staff of beamlines 14-ID BIOCARS (K. Brister) and 32-ID COMCAT (J. Brunzelle) at Advanced Photon Source for their support. In addition, we are also grateful to R. Tapping, R. Huang, J. Shisler, D. Kranz, T. Graham, P. Besant, J. Yu, and D. Lorenz for helpful discussions and critical comments on this manuscript.

The authors have no financial conflict of interest.

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

1

This work was supported by start-up funds and the Institute for Genomic Biology, University of Illinois, Urbana-Champaign (to S.K.N.). M.V.L. was supported by an individual National Institutes of Health National Research Service Award 1 F30 NS 048779-01 from the National Institute of Neurological Disorders and Stroke.

3

Abbreviations used in this paper: DD, death domain; IRAK, IL-1R-associated kinase; MAD, multiwavelength anomalous dispersion; FOM, figure of merit.

1
Imler, J. L., J. A. Hoffmann.
2000
. Toll and Toll-like proteins: an ancient family of receptors signaling infection.
Rev. Immunogenet.
2
:
294
.-304.
2
Poltorak, A., X. He, I. Smirnova, M. Y. Liu, C. Van Huffel, X. Du, D. Birdwell, E. Alejos, M. Silva, C. Galanos, et al
1998
. Defective LPS signaling in C3H/HeJ and C57BL/10ScCr mice: mutations in Tlr4 gene.
Science
282
:
2085
.-2088.
3
Janssens, S., R. Beyaert.
2003
. Functional diversity and regulation of different interleukin-1 receptor-associated kinase (IRAK) family members.
Mol. Cell.
11
:
293
.-302.
4
O’Neill, L. A., K. A. Fitzgerald, A. G. Bowie.
2003
. The Toll-IL-1 receptor adaptor family grows to five members.
Trends Immunol.
24
:
286
.-290.
5
Burns, K., S. Janssens, B. Brissoni, N. Olivos, R. Beyaert, J. Tschopp.
2003
. Inhibition of interleukin 1 receptor/Toll-like receptor signaling through the alternatively spliced, short form of MyD88 is due to its failure to recruit IRAK-4.
J. Exp. Med.
197
:
263
.-268.
6
Cao, Z., W. J. Henzel, X. Gao.
1996
. IRAK: a kinase associated with the interleukin-1 receptor.
Science
271
:
1128
.-1131.
7
Muzio, M., J. Ni, P. Feng, V. M. Dixit.
1997
. IRAK (Pelle) family member IRAK-2 and MyD88 as proximal mediators of IL-1 signaling.
Science
278
:
1612
.-1615.
8
Wesche, H., X. Gao, X. Li, C. J. Kirschning, G. R. Stark, Z. Cao.
1999
. IRAK-M is a novel member of the Pelle/interleukin-1 receptor-associated kinase (IRAK) family.
J. Biol. Chem.
274
:
19403
.-19410.
9
Li, S., A. Strelow, E. J. Fontana, H. Wesche.
2002
. IRAK-4: a novel member of the IRAK family with the properties of an IRAK-kinase.
Proc. Natl. Acad. Sci. USA
99
:
5567
.-5572.
10
Suzuki, N., S. Suzuki, G. S. Duncan, D. G. Millar, T. Wada, C. Mirtsos, H. Takada, A. Wakeham, A. Itie, S. Li, et al
2002
. Severe impairment of interleukin-1 and Toll-like receptor signalling in mice lacking IRAK-4.
Nature
416
:
750
.-756.
11
Picard, C., A. Puel, M. Bonnet, C. L. Ku, J. Bustamante, K. Yang, C. Soudais, S. Dupuis, J. Feinberg, C. Fieschi, et al
2003
. Pyogenic bacterial infections in humans with IRAK-4 deficiency.
Science
299
:
2076
.-2079.
12
Medvedev, A. E., A. Lentschat, D. B. Kuhns, J. C. Blanco, C. Salkowski, S. Zhang, M. Arditi, J. I. Gallin, S. N. Vogel.
2003
. Distinct mutations in IRAK-4 confer hyporesponsiveness to lipopolysaccharide and interleukin-1 in a patient with recurrent bacterial infections.
J. Exp. Med.
198
:
521
.-531.
13
Wietek, C., L. A. O’Neill.
2002
. IRAK-4: a new drug target in inflammation, sepsis, and autoimmunity.
Mol. Interv.
2
:
212
.-215.
14
Weeks, C. M., R. Miller.
1999
. The design and implementation of SnB v2.0.
J. Appl. Cryst.
32
:
120
.-124.
15
Terwilliger, T. C..
2003
. SOLVE and RESOLVE: automated structure solution and density modification.
Methods Enzymol.
374
:
22
.-37.
16
McRee, D. E..
1999
. XtalView/Xfit–a versatile program for manipulating atomic coordinates and electron density.
J. Struct. Biol.
125
:
156
.-165.
17
Murshudov, G. N., A. A. Vagin, E. J. Dodson.
1997
. Refinement of macromolecular structures by the maximum-likelihood method.
Acta Crystallogr. D Biol. Crystallogr.
53
:
240
.-255.
18
Weber, C. H., C. Vincenz.
2001
. The death domain superfamily: a tale of two interfaces?.
Trends Biochem Sci.
26
:
475
.-481.
19
Ross, K., L. Yang, S. Dower, F. Volpe, F. Guesdon.
2002
. Identification of threonine 66 as a functionally critical residue of the interleukin-1 receptor-associated kinase.
J. Biol. Chem.
277
:
37414
.-37421.
20
Mamidipudi, V., C. Lin, M. L. Seibenhener, M. W. Wooten.
2004
. Regulation of interleukin receptor-associated kinase (IRAK) phosphorylation and signaling by ι protein kinase C.
J. Biol. Chem.
279
:
4161
.-4165.
21
Xiao, T., P. Towb, S. A. Wasserman, S. R. Sprang.
1999
. Three-dimensional structure of a complex between the death domains of Pelle and Tube.
Cell
99
:
545
.-555.
22
Lemaitre, B., E. Nicolas, L. Michaut, J. M. Reichhart, J. A. Hoffmann.
1996
. The dorsoventral regulatory gene cassette spatzle/Toll/cactus controls the potent antifungal response in Drosophila adults.
Cell
86
:
973
.-983.
23
Jiang, Z., H. J. Johnson, H. Nie, J. Qin, T. A. Bird, X. Li.
2003
. Pellino 1 is required for interleukin-1 (IL-1)-mediated signaling through its interaction with the IL-1 receptor-associated kinase 4 (IRAK4)-IRAK-tumor necrosis factor receptor-associated factor 6 (TRAF6) complex.
J. Biol. Chem.
278
:
10952
.-10956.
24
Thompson, J. D., D. G. Higgins, T. J. Gibson.
1994
. CLUSTAL W: improving the sensitivity of progressive multiple sequence alignment through sequence weighting, position-specific gap penalties and weight matrix choice.
Nucleic Acids Res.
22
:
4673
.-4680.