doi:10.1016/j.jmb.2011.09.024 J. Mol. Biol. (2011) 413, 790–803

Contents lists available at www.sciencedirect.com Journal of Molecular Biology journal homepage: http://ees.elsevier.com.jmb

The NLRP12 : Structure, Dynamics, and Functional Insights

Anderson S. Pinheiro 1, Clarissa Eibl 2, Zeynep Ekman-Vural 1, Robert Schwarzenbacher 2 and Wolfgang Peti 1⁎

1Department of Molecular Pharmacology, Physiology and Biotechnology, Brown University, Providence, RI 02903, USA 2Department of Molecular Biology, University of Salzburg, A-5020 Salzburg, Austria

Received 6 June 2011; The initial line of defense against infection is sustained by the innate received in revised form . Together, membrane-bound Toll-like receptors and 12 September 2011; cytosolic -binding domain and leucine-rich repeat-containing accepted 14 September 2011 receptors (NLR) play key roles in the innate immune response by detecting Available online bacterial and viral invaders as well as endogenous stress signals. NLRs are 28 September 2011 multi-domain with varying N-terminal effector domains that are responsible for regulating downstream signaling events. Here, we report Edited by M. F. Summers the structure and dynamics of the N-terminal pyrin domain of NLRP12 (NLRP12 PYD) determined using NMR spectroscopy. NLRP12 is a non- Keywords: NLR that has been implicated in the regulation of Toll-like NLR proteins; receptor-dependent nuclear factor-κB activation. NLRP12 PYD adopts a ; typical six-helical bundle death domain fold. By direct comparison with pyrin domain; other PYD structures, we identified hydrophobic residues that are essential NLRP12; for the stable fold of the NLRP PYD family. In addition, we report the first in FAF-1 vitro confirmed non-homotypic PYD interaction between NLRP12 PYD and the pro-apoptotic Fas-associated factor 1 (FAF-1), which links the innate immune system to apoptotic signaling. Interestingly, all residues that participate in this protein:protein interaction are confined to the α2–α3 surface, a region of NLRP12 PYD that differs most between currently reported NLRP PYD structures. Finally, we experimentally highlight a significant role for tryptophan 45 in the interaction between NLRP12 PYD and the FAF-1 UBA domain. © 2011 Elsevier Ltd. All rights reserved.

*Corresponding author. Brown University, 70 Ship Street, GE-3, Providence, RI 02903, USA. E-mail address: [email protected]. Present address: A. S. Pinheiro, Departamento de Bioquímica, Instituto de Química, Universidade Federal do Rio de Janeiro, Rio de Janeiro, RJ 21941909, Brazil. Abbreviations used: HSQC, heteronuclear single quantum coherence; NOE, nuclear Overhauser enhancement; NLR, nucleotide-binding domain and leucine-rich repeat-containing receptor; DD, death domain; PYD, pyrin domain; FAF-1, Fas-associated factor 1; TLR, Toll-like receptor; NF-κB, nuclear factor-κB; NOESY, nuclear Overhauser enhancement spectroscopy; PDB, ; FID, Fas-interacting domain; 2D, two-dimensional; 3D, three-dimensional; TEV, tobacco etch ; TCEP, tris(2-carboxiethyl-phosphine); TOCSY, total correlated spectroscopy; CSP, chemical shift perturbation.

0022-2836/$ - see front matter © 2011 Elsevier Ltd. All rights reserved. The NLRP12 Pyrin Domain 791

Introduction Furthermore, in NLRP12 are linked to hereditary periodic fevers and atopic dermatitis, have evolved numerous strategies to stressing its role in immunity and .16,17 defend against environmental pathogens, including Through the use of silencing, it was recognized the innate and adaptive immune responses. Innate that NLRP12 fine-tunes downstream signaling via immunity is the first line of host defense against down-regulation of TLR-dependent NF-κB activa- pathogenic infection and relies on the detection of tion. Subsequently, NLRP12 suppresses the produc- pathogen- and danger-associated molecular pat- tion of pro-inflammatory cytokines by inhibiting terns by a set of germ-line-encoded receptors noncanonical NF-κB activation. This regulation is collectively called pattern-recognition receptors. likely achieved via the inhibition of necessary Among pattern-recognition receptors, the mem- processing kinases, especially the noncanonical brane-anchored Toll-like receptors (TLRs) have MAP3 kinase NIK (NF-κB-inducing kinase).13,15 In emerged as essential components of innate immu- addition, NLRP12 has been reported to interact with nity. They recognize diverse pathogen-derived the chaperone Hsp90 and postulated to play a role in molecules, causing the activation of intracellular the proteasome pathway.18 Although the precise signaling cascades that ultimately lead to an mechanism by which NLRP12 inhibits NIK-induced – inflammatory response.1 4 NF-κB activation is unknown, it is thought that During the last 10 years, it has become clear that NLRP12 drives the degradation of NIK via a TLRs are not the only pathogen- and danger- proteasome-dependent pathway. Recent reports associated signals sensors in innate immunity. speculate that this modulation can be achieved via Nucleotide-binding domain and leucine-rich repeat- the interaction of effector domains with “adaptor- containing receptors (NLRs) are expressed in the like” proteins, such as Fas-associated factor 1 (FAF- cytoplasm, and their importance in pathogen- 1),19–21 and thus links the innate immune system to associated molecular pattern and danger-associated apoptotic signaling.22 – molecular pattern recognition is rapidly growing.5,6 We23,24 and others25 29 have recently shown that Additionally, numerous human auto-inflammatory NLRP PYDs can vary in structure and dynamics disorders have been associated with mutations in and, as a result, in surface charge, a critical NLRs, emphasizing their role as central regulators of parameter directing their protein:protein interac- immunity and inflammation.7 NLRs possess a tions. Thus, to understand the biological function of tripartite architecture. They are composed of a C- PYDs in general and NLRP12 PYD specifically, we terminal leucine-rich repeat domain, necessary for determined the structure and auto-correlated fast ligand binding; a central NACHT domain, capable of timescale backbone dynamics of NLRP12 PYD. ATP binding and responsible for oligomerization; Interestingly, we identified differences among all and an N-terminal effector domain, linking the NLR currently published NLRP PYD structures in regions to downstream signaling cascades.8 The human described to be central for homotypic PYD:PYD genome encodes for 22 NLRs, which are further interactions. Finally, we tested the previously grouped into subfamilies according to their N- proposed heterotypic interaction of NLRP12 PYD terminal effector domain [pyrin domain (PYD), with the pro-apoptotic protein FAF-1 in vitro.By activation and recruitment domain (CARD), testing several different FAF-1 single- and multi- or baculovirus inhibitor repeat domain (BIR)]. NLRs domain fragments, we identified a direct interaction containing a pyrin domain (NLRP) constitute the of NLRP12 PYD with the UBA domain of FAF-1. largest subfamily of NLRs (NLRP1-14).9 Two members of the NLRP family, NLRP1 and NLRP3, activate the innate immune response via Results their interaction with the adaptor protein ASC (-associated speck-like protein containing a caspase activation and recruitment domain) and, Protein expression and purification subsequently, procaspase-1. This multicomponent complex is called the inflammasome,10,11 a macro- The N-terminal effector PYD of NLRP12 (residues molecular machine that activates the cleavage of 1–98, two N-terminal cloning artifacts; herein pro-inflammatory cytokines such as pro-IL-1β and referred as NLRP12 PYD) showed high levels of pro-IL-18 into their mature, secreted forms.2,11 soluble overexpression in Escherichia coli and was Other members of the NLRP family regulate key readily purified to homogeneity. NLRP12 PYD is a functions in the immune system without forming an monomer in solution as verified by size-exclusion inflammasome. For example, some NLRPs including chromatography. Under low salt conditions NLRP12 (previously called Monarch-1 or PYPAF-7) (100 mM NaCl), NLRP12 PYD precipitates at a – regulate nuclear factor-κB (NF-κB) activation.12 14 concentration higher than 0.2 mM. This is not NLRP12 was one of the earliest identified NLRPs, surprising, as PYDs are known for their poor and its expression is restricted to myeloid cells.15 solubility and tendency to aggregate.26 To achieve 792 The NLRP12 Pyrin Domain concentrations of NLRP12 PYD sufficient for accu- Table 1. Structural and CNS refinement statistics rate structural and dynamics studies, 500 mM NaCl NLRP12 PYD was necessary. Under these conditions, NLRP12 PYD is readily concentrated to 0.6 mM without Number of restraints Unambiguous distance restrains (all) 1859 precipitation or aggregation and stable during the Intra-residue 490 course of several weeks. Short range 458 FAF-1 constructs (FAF-11–99,FAF-11–57, and FAF- Medium range 466 Long range 445 199–180) were overexpressed in E. coli and purified to homogeneity. An maltose-binding protein (MBP) tag Deviations from idealized covalent geometry was used to enhance the solubility of FAF-11–57 during Bonds (Å) 0.009±0.0004 expression. All FAF-1 constructs are monomers in Angles (°) 1.24±0.05 solution as verified by size-exclusion chromatography. Impropers (°) 1.39±0.09

Structural quality Three-dimensional structure of NLRP12 PYD Ramachandran plot (10–91; NMR-PROCHECK) (%) Most favored region 89.1 The solution structure of NLRP12 PYD was Additionally allowed region 9.7 determined using heteronuclear NMR spectroscopy. Generously allowed region 0.8 Assignments were obtained for 95% of the backbone Disallowed region 0.5 α α Pairwise RMSD (Å) ′ α nuclei (N, HN, C ,C ,H ) and 95% of the side-chain Backbone (N, C , C, and O) (10–91) 0.62±0.11 13 CHn moieties. Of the 96 expected backbone amide All heavy atoms (10–91) 1.20±0.14 NH pairs (3 prolines), 93 were identified; the missing assignments correspond to the N-terminal two-residue cloning artifact Gly2 and His1, as well allowed regions of the Ramachandran diagram, 0.8% as Thr4 (Supplemental Fig. S1). All aliphatic and of residues in the generously allowed region, and aromatic side-chain resonances that are routinely 0.5% of residues in the disallowed region (Table 1). observed were assigned, except those of residues The NLRP12 PYD structure is well defined, with the Gly2, His1, Arg3, Ser13, Lys29, and Glu100, where exception of the N- and C-termini, residues −2to9 confident assignments were uncertain. An ensemble and 92 to 100, respectively, which are flexible of 100 structures was calculated from 1859 nuclear (Supplemental Fig. S2a and b). The root-mean-square Overhauser enhancement spectroscopy (NOESY)- deviation (RMSD) value about the mean coordinate derived distance constraints [∼18 nuclear Over- positions of the backbone atoms for residues 10–91 of hauser enhancement (NOE) constraints/residue] NLRP12 PYD is 0.62±0.11 Å (Table 1). using a simulated annealing protocol within the As expected, NLRP12 PYD folds into a tightly program CYANA30 and refined in explicit solvent packed helical bundle (residues 10–91) consisting using CNS.31 The 20 lowest-energy structures of of six helices (α1–α6) arranged in an antiparallel NLRP12 PYD are shown in Fig. 1a. All structures have fashion (Fig. 1b), characteristic of PYDs. However, excellent geometry, with no violations of distance unlike other PYD structures, NLRP12 PYD helix α restraints greater than 0.5 Å and no dihedral angle 3 only forms a short 310 helix. The residues violations greater than 5° (Table 1). In addition, all forming the six helices are 10–19 (α1), 22–34 (α2), – α — – α – α structures have excellent stereochemistry, with 98.8% 47 50 ( 3 310 helix), 53 64 ( 4), 66 80 ( 5), and of residues in the most favored and additionally 83–91 (α6).

(a) (b) (c) α6 α5 α4 α2 x α 90o α1 3-310 α1 α3-3 α4 10 C

α2 α6 N α5

Fig. 1. NMR structure of NLRP12 PYD. (a) Ensemble of the 20 lowest-energy structures calculated for NLRP12 PYD superimposed on the backbone atoms of residues 10–91 (PDB ID 2L6A). The six helices, characteristic of the DD fold, are highlighted in light pink, while loops are highlighted in gray. (b) Ribbon representation of the lowest-energy conformer of NLRP12 PYD in an orientation identical with that shown in (a). The N- and C-termini as well as the six helices are labeled. (c) Top view of NLRP12 PYD [rotated by 90° about the x-axis relative to (a) and (b)]. Residues forming the central hydrophobic core are shown as cyan sticks. The NLRP12 Pyrin Domain 793

The NLRP12 PYD six-helical bundle is stabilized PYD shows a small increase in fast timescale by a central hydrophobic core formed by residues dynamics in the α2–α3 loop, but no slow-interme- Leu13, Tyr16, Leu17, and Leu20 from helix α1; diate timescale dynamics, when compared to Leu25, Phe28, and Leu32 from helix α2; Met56, NLRP7 PYD. Indeed, the dynamics detected for Leu60, and Phe64 from helix α4; Ala69, Trp70, NLRP12 PYD seem to be most similar to the Ala73, Phe77, and Ile80 from helix α5; and Leu85 dynamics reported for the PYD of the inhibitor from helix α6(Fig. 1c). All α-helices in the structure protein ASC2.27 of NLRP12 PYD are connected by short well-defined A ∼10% cis/trans proline isomerization for Pro33 loops, with the exception of the loop that connects and Pro42 in NLRP7 PYD and Pro42 in NLRP1 PYD α α α –α helices 2 and 3(310 helix), herein called the 2 3 was reported. These proline residues flank the loop. The α2–α3 loop comprises residues 35–46. α2–α3 loop. In contrast, no significant cis/trans Despite the fact that it does not contain any regular proline isomerization was detected in NLRP12 PYD. secondary structure, the α2–α3 loop is quite ordered in the structure of NLRP12 PYD. This order in the Structural and dynamics comparison of NLRP12 α2–α3 loop is due to hydrophobic interactions of PYD with members of the NLRP family residues Leu38 and Ile43 both within the α2–α3 loop itself and with neighboring residues, especially Despite a relatively low sequence identity be- residue Phe64 in helix α4(Supplemental Fig. S3). tween NLRP12 PYD and the PYDs of NLRP7 [29% sequence identity; Protein Data Bank (PDB) ID Backbone dynamics of NLRP12 PYD 2KM6] and NLRP1 (36% sequence identity; PDB ID 1PN5), their structures superimpose well with The PYDs of NLRP1 and NLRP7 show distinct RMSD values of 2.4 Å for the backbone atoms dynamics behaviors, especially in the α2–α3 loop, between NLRP12 and NLRP7 PYDs and 3.1 Å which exhibits substantially increased fast timescale between NRLP12 and NLRP1 PYDs. Interestingly, a dynamics in NLRP1 PYD (picosecond–nanosecond structure-based sequence alignment identified that as detected by 15N[1H]-NOE measurement) and the conserved residues between all three PYD slow timescale dynamics in NLRP7 PYD (microsec- structures form the hydrophobic core of the PYD ond as detected by 15N CSA, 15N–1H dipole–dipole fold. This hydrophobic core is highly conserved cross-correlation relaxation rates, and on-resonance among the entire NLRP family (NLRP1–14) and, R1ρ measurements).24,26 To investigate the dynam- therefore, defines the overall fold of NLRP PYDs ics in NLRP12 PYD, we measured auto-correlated (Fig. 3). 15N-relaxation data, including measurements of 15N Superposition of the structure of NLRP12 PYD 1 15 [ H]-NOE, as well as N longitudinal (R1) and with those of NLRP1 and NLRP7 PYDs revealed transverse (R2) relaxation rates. Using a model-free small differences in the length and orientation of the τ α α analysis, we calculated a correlation time ( c)of helices. In particular, helices 1 and 6 differ in ∼7 ns for NLRP12 PYD at 298 K. Comparison of the overall length between the three PYD structures. 15N[1H]-NOE data, which reports on fast timescale Furthermore, the relative angle of helix α2 with (picosecond–nanosecond) backbone motions, of the respect to the overall helical bundle shows the PYDs of NLRP1, NLRP7, and NLRP12 is shown in largest difference between the structures of NLRP12 Fig. 2. The dynamics of the α2–α3 loop in NLRP12 and NLRP7 PYDs. PYD are much more similar to the dynamics of Clearly, the most significant differences among the NLRP7 PYD than those of NLRP1 PYD. NLRP12 three PYD structures are helix α3 and the preceding

α 3-310 α1 α2 α4 α5 α6 1.0 Fig. 2. Comparison of fast time- scale backbone dynamics among 0.8 the PYDs of NLRP12, NLRP7, and 15 1 0.6 NLRP1. Comparison of N[ H]- NOE (hetNOE) values measured H]-NOE 1 0.4 for the PYDs of NLRP12 (light

N[ 0.2 pink), NLRP7 (light blue), and 15 NLRP1 (light green). Experimental- 0.0 ly derived secondary structure ele- 0 20406080100 ments of NLRP12 PYD are depicted Residue by gray cylinders above the figure. A hetNOE of ∼0.8 is typical for well-formed, stable secondary structural elements. While the α2–α3 loop and helix α3 in NLRP1 PYD show a substantial increase in fast timescale backbone dynamics, these motions are missing in NLRP7 PYD. The α2–α3 loop and helix α3in NLRP12 PYD show increased flexibility when compared to NLRP7 PYD. 794 The NLRP12 Pyrin Domain

Fig. 3. Sequence alignment of human NLRP PYDs. Residues contributing to the hydrophobic core of NLRP12 PYD are conserved among the entire NLRP family and are highlighted by gray boxes. Experimentally derived secondary structure elements of NLRP12 PYD are depicted by gray cylinders on top of the figure.

α2–α3 loop. The overall length of NLRP12 PYD is PYD. Furthermore, this secondary structure element shorter than that of NLRP7 and NLRP1 PYDs, and is entirely missing in NLRP1 PYD, where α3is the structure-based sequence comparison (Fig. 3) replaced by a flexible disordered loop that connects identified a two-amino-acid deletion in the α2–α3 helix α2 to helix α4. Second, the dynamics of the loop region of NLRP12 PYD. This deletion has α2–α3 loop in all three NLRP PYDs differ. While the multiple consequences. First, helix α3 (two turns) in α2–α3 loop in NLRP1 PYD shows significantly α –α NLRP7 PYD is replaced by a 310 helix in NLRP12 increased fast timescale dynamics, the 2 3 loop The NLRP12 Pyrin Domain 795 is rigid in NLRP7 PYD. In NLRP12 PYD, the α2–α3 of the α2–α3 loop. Strikingly, NLRP1 PYD lacks all loop is slightly more flexible than in NLRP7 PYD but six hydrophobic residues entirely (Fig. 4a and b). much more rigid than in NLRP1 PYD. Previously, Thus, the dynamics of the α2–α3 loop in NLRP12 we identified a hydrophobic cluster (consisting of six PYD are more similar to those of NLRP7 PYD than to hydrophobic residues) that stabilizes both the α2–α3 those of NLRP1 PYD, and these differences are loop and helix α3 in NLRP7 PYD. Interestingly, this directly correlated with the size of the hydrophobic hydrophobic cluster only contains three hydropho- cluster. bic residues in NLRP12 PYD. The other three The hydrophobic cluster in NLRP7 PYD that hydrophobic residues in NLRP12 PYD are replaced stabilizes the α2–α3 loop is anchored by two by two glycines and one alanine, smaller nonpolar stacking Trp side chains (residue Trp30 and residues, that likely account for the reduced rigidity Trp43). However, Trp30 of NLRP7 PYD is replaced

NLRP12 NLRP7 NLRP1 α (a) 2 α 3 310

α2-α3 loop

(b)

W30 W43 F32 T41 I43 L38 L34 L38

(c)

G33 W30 W43

W45

Fig. 4. Structural comparison of NLRP PYDs. (a) NLRP12 PYD (light pink) overlaid with the PYDs of NLRP7 (light blue; RMSD, 2.4 Å) and NLRP1 (light green; RMSD, 3.1 Å). The pairwise RMSD between NLRP12 PYD and all other PYDs was calculated by superposition of helices α1–α6. The largest structural difference is localized to the α2–α3 loop as well as helix α3, which is illustrated in the current orientation. (b) Hydrophobic residues in the α2–α3 loop as well as helix α3 are highlighted as dark-blue sticks and labeled. A six-residue hydrophobic cluster stabilizes the α2–α3 loop as well as helix α3 in NLRP7 PYD. The corresponding cluster in NLRP12 PYD consists only of two hydrophobic residues. NLRP1 PYD lacks all six hydrophobic residues. (c) Residues Gly33 and Trp45 in NLRP12 PYD, as well as Trp30 and Trp43 in NLRP7 PYD, are highlighted as dark-red sticks and labeled. While Trp43 in NLRP7 PYD is buried and forms stacking interactions with Trp30, the corresponding Trp45 in NLRP12 PYD is surface exposed. 796 The NLRP12 Pyrin Domain in NLRP12 PYD by a Gly residue (Gly33). This perturbations (CSPs) were identified in the titration substitution has two effects. First, it plays a key role experiments, showing that there is, as expected, no in the increased flexibility of the α2–α3loop interaction between NLRP12 PYD and ASC PYD. displayed by NLRP12 PYD. Second, it also changes the role of NLRP12 PYD Trp45 (Trp43 is the NLRP12 PYD interacts directly with the UBA corresponding residue in NLRP7 PYD). Instead of domain of FAF-1 being highly buried, as it is in NLRP7 PYD, it becomes surface exposed in NLRP12 PYD (Fig. 4c). FAF-1 is a 74-kDa multi-domain protein that has beenshowntofunctionindiversebiological Structural comparison of NLRP12 PYD among processes, such as the regulation of apoptosis and the death domain superfamily NF-κB activity, as well as ubiquitination and proteasomal degradation. FAF-1 consists of multiple The most similar structures to NLRP12 PYD, based protein-interaction domains, including a Fas-inter- on Dali z-scores, are the structures of ASC2 PYD (z- acting domain (FID), a - score, 10.5) and ASC PYD (z-score, 10.4). The overall interacting domain, and multiple -related length of NLRP12 PYD is shorter than that of the domains. Recently, Kinoshita et al. used yeast two- ASC2 PYD, and a structure-based sequence compar- hybrid screening to search for novel FAF-1-interact- ison also identified a two-amino-acid deletion in the ing proteins and identified the PYDs of NLRP3, α2–α3 loop. Similar assessments are possible for the NLRP7, and NLRP12 as potential FAF-1 targets.19 In ASC PYD. Nevertheless, this has very little influence their work, they used the entire FAF-1 FID (residues in the overall structure, as well as dynamics of the 1–180) as bait. The FID of FAF-1 consists of a UBA α2–α3 loop. (ubiquitin associated) domain (residues 1–57) and The electrostatic surface of PYDs has two distinct an UB1 (ubiquitin related) domain (residues 99–180) faces. One surface is formed by helices α2 and α3, connected by a flexible linker sequence (residues 58– and a second surface is composed of helices α1 98). We recently tested the interaction of NLRP7 and α4. In both the ASC and ASC2 PYDs, these PYD with the FAF-1 FID (residues 1–180) using surfaces are highly charged and complementary NMR titrations but were unable to detect a direct with one another. The α2–α3 electrostatic surface interaction.24 This was expected, as no heterotypic of NLRP12 PYD is also charged but also has a interactions had been previously reported for a PYD significant hydrophobic patch clustered around domain. Trp45 on helix α3. Nevertheless, we performed the identical titra- Figure 5 compares the electrostatic surface of tion study with NLRP12 PYD and FAF-1 FID. NLRP12 PYD to those of ASC, ASC2, NLRP7, and Surprisingly, we detected small CSPs, indicating a NLRP1. We used the PIPSA (Protein Interaction direct interaction between the two proteins. In Property Similarity Analysis) server32 to quantify order to investigate this interaction in more detail, the similarities between the electrostatic surfaces of we subcloned the FAF-1 FID into multiple the PYDs from NLRP1, NLRP7, NLRP12, ASC, and shorter, structurally and biologically meaningful ASC2. The PIPSA server uses the UHBD program sub-domains: (1) FAF-11–57 (UBA domain only), to calculate the electrostatic potential of a protein. (2) FAF-11–99 (UBA domain and flexible linker), Pairwise calculations comparing the NLRP12 PYD and (3) FAF-199–180 (UB1 domain only). All FAF-1 to the PYDs of ASC, ASC2, NLRP7, and NLRP1 constructs expressed solubly and were purified to were used to assess overall electrostatic similarity, homogeneity, and one-dimensional 1HNMR where 0 indicates identical and 2 indicates spectra were recorded to confirm their folded completely different electrostatic surfaces. Based state (Supplemental Fig. S5). Because it has been on an overall electrostatic distance, NLRP12 PYD is reported that death domain (DD) interactions are most similar to ASC PYD (electrostatic distance, more robustly identified in higher pH solution,34 0.634) followed by NLRP7 PYD (electrostatic all titrations were performed at pH 7 or 7.5. NMR distance, 0.994), ASC2 PYD (electrostatic distance, CSPs (detected using 15N-labeled NLRP12 PYD) 1.324), and NLRP1 PYD (electrostatic distance, showed that NLRP12 PYD interacts with the N- 1.339). Therefore, based on this analysis, ASC terminal UBA domain of FAF-1 (residues 1–57) PYD can be classified as identical/highly similar (Supplemental Fig. S6). Interestingly, when the to the NLRP12 PYD (i.e., an electrostatic distance NMR titration experiment was carried out with between 0 and 0.75). FAF-11–99 (UBA domain and linker region), the Numerous reports have argued that differences in resulting spectrum was identical with that of the surface charge/hydrophobicity are the key drivers FAF-11–57 titration (Supplemental Fig. S7). This for the interaction specificity of NLRP PYDs.28,33,34 shows that the UBA domain of FAF-1 is both We have tested the direct interaction of NLRP12 necessary and sufficient for the interaction with PYD with ASC PYD at various ratios using NMR NLRP12 PYD and that the residues comprising spectroscopy (data not shown), but no chemical shift the flexible linker region (58–98) do not play a The NLRP12 Pyrin Domain 797

(a) NLRP12 ASC2 ASC α2 α 3-310

α2-α3 loop

(b) NLRP12 NLRP7 NLRP1 α2 α 3-310

α2-α3 loop

Fig. 5. Electrostatic surface potential of NLRP12 PYD and other PYDs with known structures. (a) Top row: ribbon representation of the PYDs of NLRP12 and the ones of the adaptor protein ASC and its inhibitor, ASC2, facing the α2–α3 surface. Bottom row: electrostatic surface representation of the PYDs of NLRP12, ASC and ASC2. Positive surface charge is colored blue; negative surface charge is colored red; and neutral surface, white. (b) Corresponding electrostatic surface representation of the PYDs of NLRP12, NLRP7, and NLRP1. role in this interaction. We also show that this surface, composed of helix α2, the α2–α3 loop, interaction is specific, as no shifts were observed and helix α3 of NLRP12 PYD. It is interesting to when NLRP12 PYD was titrated with FAF-199–180 note that this surface has been previously implicat- (UB1 domain) (Supplemental Fig. S8). ed in mediating homotypic interactions of PYDs. Using the results from the NMR titrations, we Specifically, the interaction between NLRP12 PYD identified the residues of NLRP12 PYD that and FAF1 UBA is mediated by residues Lys27, mediate binding with FAF-11–57. The majority of Thr34, Thr36, Glu40, Lys42, Ile43, Trp45, Gly46, the perturbed residues (greater than 2 SD from the and Lys50 (Fig. 6). The largest chemical shift mean) were located exclusively to the α2–α3 changes were detected for residues Glu40, Lys42, 798 The NLRP12 Pyrin Domain

α (a) 3-310 α1 α2 α4 α5 α6 0.10 0.08 0.06

(ppm) 0.04 δ Δδ Δ 0.02 0.00 0 20 40 60 80 100 Residue (b)α2 (c) K27 α 3-310

K50

G46 T34 W45

T36 I43 E40 K42 α2-α3 loop

Fig. 6. Mapping the interaction of NLRP12 PYD with FAF-11–57. (a) CSPs calculated for NLRP12 PYD upon titration of FAF-11–57 at a molar ratio of 1:10 [10 mM Na-phosphate buffer (pH 7.0), 100 mM NaCl, and 0.5 mM TCEP]. The color scheme denotes the intensity of the shifts observed in the titration experiment. CSP values higher than 2 SD from the mean are colored light blue; 3 SD, marine; and 4 SD, purple. Experimentally derived secondary structure elements of NLRP12 PYD are depicted by gray cylinders on top of the figure. (b) NLRP12 PYD structure displaying the residues that show the highest CSP values upon titration with FAF-11–57. Side chains are depicted in stick model, labeled and colored according to (a). (c) Surface representation of the NLRP12 PYD structure in the same orientation as (b), displaying the residues that show the highest CSP values. These residues are clustered on the α2–α3 surface, which has been previously implicated in homotypic interactions of PYDs.

and Trp45. Based on the poor solubility of both FAF-11–57, as no chemical shift changes were proteins and the limitations of sample concentra- observed in the titration experiments (Supplemen- tions necessary for reliable NMR measurements, tal Figs. S10–S12). In contrast, chemical shift the highest feasible titration ratio used was 1:10 changes were identified for W45F and W45I (NLRP12 PYD:FAF-11–57). This ratio does not mutations (Supplemental Figs. S13 and S14). identify a plateau in the titration curve that is This highlights a role for the surface-exposed characteristic of saturation of the interaction and hydrophobic Trp45 residue in the binding of necessary for a defined calculation of a Kd value. NLRP12 PYD to FAF-11–57. Nevertheless, based on these chemical shift Recently, a crystal structure of FAF-15–47 was changes, we estimated a dissociation constant reported (PDB ID 3E21), as were the chemical shift ≥ μ 21 (Kd) 150 M between NLRP12 PYD and FAF-11–57 assignments for FAF-11–81. While the domain as well as FAF-11–99 (Supplemental Fig. S9). boundaries for these constructs differ from those To elucidate the importance of hydrophobic used in our studies, we were nevertheless able to contacts in the interaction between NLRP12 PYD rapidly transfer the reported sequence-specific and FAF-11–57, we generated numerous point backbone assignments to FAF-11–57, allowing the mutants of NLRP12 PYD residue Trp45 (W45A, detection of NLRP12-PYD-caused CSPs on 15N- W45E, W45R, W45F, and W45I). The ability of labeled FAF-11–57 in a two-dimensional (2D) 1 15 the NLRP12 PYD mutants to bind FAF-11–57 was [ H, N] heteronuclear single quantum coherence then assayed using NMR titration experiments. (HSQC) spectrum. However, these reverse titration Substitution of NLRP12 PYD W45 by an A, E, or experiments (unlabeled NLRP12 PYD into 15N- R residue completely abolished binding with labeled FAF-11–57) did not result in any significant The NLRP12 Pyrin Domain 799

CSPs, up to the protein ratios technically possible Discussion (Supplemental Fig. S15). The most likely explana- tion for the lack of detectable CSPs is that this PYDs are the most common members of the DD interaction occurs in an intermediate exchange superfamily.35 PYDs play a key role in the control of regime when detected by FAF-11–57. Therefore, it signaling pathways in the immune system, as well does not lead to progressive chemical shift as in many apoptotic pathways.36 While numerous changes, which are the hallmarks of CSP mapping structures of PYDs have been reported during the experiments in a fast or slow exchange regime. last few years, the differences, in both structure and In order to test this hypothesis, we measured the dynamics, between them are larger than the peak intensities of free FAF-11–57 and FAF-11–57 similarities. Moreover, the currently available struc- bound to NLRP12 PYD (at the highest titration tures have neither revealed the differential specific- ratio of 1:10) under identical buffer, concentration, ities for their homotypic interactions, nor have and NMR conditions. If NLRP12 PYD binds to heterotypic interactions of these domains been FAF-11–57, the intensities of the peaks correspond- reported in vitro. Here, we report the three-dimen- ing to those residues that interact directly with sional (3D) structure of human NLRP12 PYD, one of NLRP12 PYD will change. As shown in Fig. 7a, I/ the first NLRPs discovered that functions to down- κ I0 intensity changes that are 2× or 3× higher than regulate TLR-dependent NF- B activation via direct the standard deviation were detected. These interaction with processing kinases. Furthermore, changes, which define the NLRP12 PYD interac- NLRP12 plays also a role in the proteasome tion surface on FAF-1, occur in three α-helices that pathway.13,15 Thus, NLRP12 is considered as an form the core structure of FAF-11–57. FAF-1 also essential regulatory bridge between innate immunity interacts with ubiquitin and ubiquitin-like proteins, and apoptotic signaling pathways. and the residues involved in these interactions The structure of NLRP12 PYD folds into an differ from those that interact with NLRP12 expected DD fold. Through the use of a structure- PYD,21 showing that these proteins bind to distinct based sequence alignment of the currently three surfaces on FAF-1. available NLRP PYD structures (NLRP1, NLRP7,

(a) 0

0 -1 I/I -2

-3 01020 30405060 Residue L37 A40 (b) (c) x R7 180o

E8 E31 R7 E8 M9 L29 I27 A12 I27 F14 A26 A16 Q15

Fig. 7. Mapping the interaction of FAF-11–57 with NLRP12 PYD. (a) Chemical shift intensity differences for unbound FAF-11–57 and NLRP12-PYD-bound FAF-11–57 [1:10 ratio; 10 mM Na-phosphate buffer (pH 7.0), 100 mM NaCl, and 0.5 mM TCEP; black bars]. Cyan denotes values of I/I0 2× higher than the standard deviation and blue I/I0 values 3× higher than SD. The color scheme denotes the intensity of the shifts observed in the titration experiment. Experimentally derived secondary structure elements of FAF-11–57 are depicted by gray cylinders above the figure. (b) FAF-15–47 structure (PDB ID 3E21) displaying the residues that show the largest intensity changes upon titration with NLRP12 PYD. Side chains are depicted in stick model, labeled and colored according to (a). (c) Surface representation of the FAF-11–57 structure, with residues that show the highest I/I0 differences colored according to (a). 800 The NLRP12 Pyrin Domain and NLRP12), it was possible to identify conserved completely buried in the hydrophobic core of this residues that form the hydrophobic core of these protein. This likely explains the differential interac- three PYDs. Critically, this hydrophobic core is tion with the FAF-1 UBA domain. nearly perfectly conserved among the NLRP family Taken together, our analysis revealed initial (NLRP1–14) and, thus, enabled us to define the insights into a non-homotypic PYD interaction overall hydrophobic core of the NLRP PYD family. between NLRP12 PYD and the FAF-1 UBA domain. However, because of a two-amino-acid deletion in In contrast to homotypic PYD:PYD interactions, α –α α the 2 3 loop, helix 3 is replaced by a 310 helix in which are mainly driven by electrostatic contacts NLRP12 PYD. Furthermore, analysis of the dynam- and anchored by hydrophobic residues, this inter- ics in the α2–α3 loop of NLRP12 PYD shows slightly action is mainly driven by a weak hydrophobic increased fast timescale dynamics when compared contact. The interaction between NLRP12 PYD and to NLRP7 PYD. Interestingly, the PYD domain that FAF-1 indicates a possible mechanism for inhibition is structurally most similar to that of NLRP12 is that of TLR-dependent NF-κB activation. of ASC2. However, the ASC2 PYD has highly charged electrostatic surfaces formed by helices Materials and Methods α2/α3 (positive) and α1/α4 (negative), which are indicated to be essential for the interaction with ASC PYD.25,27,37 These oppositely charged surfaces are Protein expression and purification mostly missing in NLRP PYDs. NLRP12 PYD has a higher surface charge than NLRP7 PYD (which NLRP12 PYD (residues 1–98) was subcloned into 41 explains why it was necessary to add 500 mM NaCl pHisparallelSTOP, which encodes an N-terminal His6- to stabilize the NLRP12 PYD sample for NMR purification tag and a tobacco etch virus (TEV) protease cleavage site. FAF-1 constructs FAF-11–99 (UBA domain measurements). However, NLRP12 PYD has fewer – and linker region; residues 1 99) and FAF-199–180 (UB1 hydrophobic patches than NLRP7 PYD. Based on 42 domain; residues 99–180) were subcloned into RP1B, the differential electrostatic surfaces, it seems un- which encodes a Thio His expression/purification tag likely that the charge-driven NLRP1 PYD:ASC PYD 6 6 and a TEV protease cleavage site. FAF-11–57 (UBA domain; interaction can be formed between NLRP12 PYD: residues 1–57) was cloned into pETM30-MBP, which ASC PYD, which is typically required for the encodes an N-terminal His6-MBP purification/solubility formation of an inflammasome.38 tag and a TEV cleavage site. The plasmids were While numerous homotypic and heterotypic in- transformed into E. coli BL21-Codon-Plus (DE3)-RIL (Stratagene) cells. The expression of uniformly 13C/15N- teractions have been reported for NLRP PYDs, very 15 few have been confirmed in vitro. Here, we tested the labeled and N-labeled protein was carried out by previously reported interaction with FAF-1, a vital growing freshly transformed cells in M9 minimum medium containing 4 g/L [13C]glucose and/or 1 g/L protein involved in ubiquitination and proteasomal 15 19 NH4Cl (Cambridge Isotope Laboratory) as the sole degradation regulation. We were able to use NMR source of carbon and nitrogen, respectively. Cell cultures chemical shift mapping to detect a very weak ≥ μ were grown at 37 °C under vigorous shaking (250 rpm) in interaction (estimated a Kd of 150 Mbasedon the presence of 34 μg/mL chloramphenicol and 50 μg/mL chemical shift analysis) with the N-terminal UBA kanamycin in the case of all FAF-1 constructs or in the (ubiquitin associated) domain (residues 1–57) of presence of 34 μg/mL chloramphenicol and 50 μg/mL FAF-1. This is the first in vitro confirmed heterotypic ampicillin in the case of NLRP12 PYD until they reached interaction of a PYD. The inflammatory response to an OD600 of 0.6. Expression of all proteins was induced by danger signals is a difficult balancing act for the host addition of 1 mM IPTG to the culture medium, and and therefore must be tightly regulated. Whereas the cultures were allowed to grow overnight (18 h) at 18 °C most prominent NLRPs, including NLRP1 and under vigorous shaking (250 rpm). Cells were harvested by centrifugation and stored at −80 °C. NLRP3, are positive regulators of inflammatory Purification of NLRP12 PYD and all FAF-1 constructs responses, NLRP2, NLRP4, and NLRP12 were was performed as follows. Cells were resuspended in lysis shown to be negative regulators of pro-inflammatory buffer [50 mM Tris (pH 8.0), 500 mM NaCl, 5 mM 12,13,18,39,40 signaling by inhibiting NF-κB activation. imidazole, and 0.1% Triton-X 100, supplemented with As FAF-1 itself is indicated in inhibiting NF-κB ethylenediaminetetraacetic-acid-free protease inhibitor signaling, it is likely that this interaction can lead to a tablets (Roche)] and lysed by high-pressure homogeniza- synergistic effect between NLRP12 PYD and FAF-1 tion (Avestin C-3 Emulsiflex). Cell debris was removed by UBA. Thus, NLRP12 together with FAF-1 modulates centrifugation at 35,000g for 40 min at 4 °C, and the the pro-inflammatory response, where NF-κB pos- supernatant containing soluble proteins was loaded onto a sesses a key role. Therefore, to understand the HisTrap HP column (GE Healthcare) equilibrated with 50 mM Tris (pH 8.0), 500 mM NaCl, and 5 mM imidazole. molecular basis of this interaction, we performed a – His6-tagged proteins were eluted with a 5 500 mM number of mutagenesis experiments and showed imidazole gradient. Fractions containing the protein of that the hydrophobic nature of a surface-exposed interest, as identified by SDS-PAGE, were pooled, incu- Trp (Trp45) is critical for the interaction. Interest- bated with His6-TEV NIa (S219V) protease, and dialyzed at ingly, the identical Trp residue in NLRP7 PYD is 4 °C against 50 mM Tris (pH 7.5) and 200 mM NaCl until The NLRP12 Pyrin Domain 801 cleavage was complete. The untagged proteins were the study were performed using the lowest-energy separated from the enzymatically cleaved His6- or His6- conformer of the ensemble of structures. +2 MBP tag, as well as from His6-TEV by Ni -affinity subtraction purification. Proteins were subsequently puri- Relaxation measurements and analysis fied by size-exclusion chromatography using a Superdex 75 26/60 column (GE Healthcare). Fractions containing the 15 pure proteins, as identified by SDS-PAGE, were pooled N relaxation experiments were performed under the and concentrated. All purifications were performed at 4 °C. exact same conditions as described for the structure 15 determination of NLRP12 PYD. N longitudinal (R1) 15 1 and transverse (R2) relaxation rates and N[ H]-NOE NMR spectroscopy (hetNOE) measurements were acquired using sensitivity- enhanced pulse sequences. T1 experiments were acquired NMR experiments were acquired at 298 K on a Bruker with relaxation delays (T) of 20, 100, 200, 400, 550, 700, 850, 13 Avance 500-MHz spectrometer. In addition, a 3D C- and 1000 ms. T2 experiments were acquired with relaxation resolved [1H,1H] NOESY spectrum was recorded on a delays (T) of 20, 80, 110, 140, 180, 220, 250, and 350 ms. A Bruker Avance 800-MHz spectrometer. Both spectrometers recycle delay of 3 s between scans was used for all T1 and T2 are equipped with a TCI HCN z-gradient cryoprobe. Proton experiments. 15N[1H]-NOEs were measured from a pair of chemical shifts were referenced directly to internal 2,2- spectra acquired with and without presaturation recorded dimethyl-2-silapentane-5-sulfonate. 13Cand15N chemical in an interleaved manner. A recycle delay of 5 s between shifts were referenced indirectly to 2,2-dimethyl-2-silapen- scans was used for the heteronuclear NOE experiments. tane-5-sulfonate using the absolute frequency ratios. All spectra were processed with NMRPipe47 and 48 analyzed with NMRView. R1 and R2 relaxation rates were determined by fitting the peak intensities as a Chemical shift assignments and structure calculation function of the relaxation delays using an exponential − decay function, I(T)=Ae( r/T), where I(T) is the peak All NMR experiments for chemical shift assignments intensity after a time delay T, A is the intensity at time and structure determination were performed with either a zero, and r=R or R . 15N[1H]-NOEs were calculated by 15 15 13 1 2 N- or a N/ C-labeled NLRP12 PYD sample at a final dividing the intensity of the peaks in the spectra recorded concentration of 0.6 mM in 20 mM sodium phosphate without presaturation by the intensity of the peaks in the (pH 6.5), 500 mM NaCl, and 0.5 mM tris(2-carboxiethyl- presaturated spectra. phosphine) (TCEP). The following spectra were used to achieve the sequence-specific backbone and side-chain resonance assignments of NLRP12 PYD: 2D [1H,15N] NLRP12 PYD:FAF-1 interactions HSQC, 3D HNCA, 3D HNCACB, 3D CBCA(CO)NH, 3D HNCO, 3D HN(CA)CO, 3D CC(CO)NH [CC-total correlat- The interaction of NLRP12 PYD with FAF-1 was ed spectroscopy (TOCSY) mixing time of 12 ms], 3D HBHA investigated by NMR titration experiments using a 15N- (CO)NH, and 3D HC(C)H-TOCSY (CC-TOCSY mixing time labeled NLRP12 PYD or FAF-11–57 sample at a final of 12 ms). TopSpin 2.1 (Bruker) was used for data acquisition concentration of 35 μM in 10 mM Na-phosphate buffer and processing. NMR spectra were analyzed using the (pH 7.0 or 7.5), 100 mM NaCl, and 0.5 mM TCEP (pH † program CARA . varied depending on the pI of the FAF-1 constructs used). The following spectra were used for the structure Titration experiments were performed with three different 15 1 1 calculation of NLRP12 PYD: 3D N-resolved [ H, H] FAF-1 constructs (FAF-1 – , FAF-1 – , and FAF-1 – ) 13 1 1 1 99 99 180 1 57 NOESY (mixing time of 80 ms), 3D C-resolved [ H, H] at 1:2, 1:5, 1:10, and 1:15 molar ratios, with the highest 1 1 NOESY (mixing time of 80 ms), and 2D [ H, H] NOESY concentration ratio dependent on the solubility of the FAF- 15 (mixing time of 80 ms; acquired in 100% D2O solution). 1 construct in the case of detection by N-labeled NLRP12 NOESY peak picking, NOESY peak assignment, and 3D PYD. CSPs in the 2D [1H,15N] HSQC spectrum of NLRP12 structure calculation were performed automatically using PYD upon titration with FAF-1 constructs were used to 43,44 the ATNOS/CANDID module in the Unio software. monitor binding. CSPs were calculated using the follow- The input for the structure calculations of NLRP12 PYD ing equation: CSP=[(ΔNH)2 +(Δ15N/10)2]1/2,where was the sequence, the complete chemical shift ΔNH and Δ15N represent the difference between free 45 lists, and the 3D and 2D NOESY spectra. Default and bound 1H and 15N chemical shifts, respectively. A program parameters were used for all calculations. dissociation constant (Kd) for the interaction between Constraints for backbone dihedral angles derived from NLRP12 PYD and FAF-1 – was calculated using the 13 1 57/99 C chemical shifts were only used in the initial structure chemical shift changes in the NMR titration experiments. 30 calculation. The 20 conformers from the final CYANA Only the chemical shift resonances of NLRP12 PYD that cycle with the lowest residual CYANA target function exhibited the most significant change upon titration with values were energy minimized in a water shell using FAF-1 – were used for this calculation. CSPs were 31 46 1 57/99 CNS and the RECOORD script package (Table 1). The fitted to a one-site saturation binding equation using structure quality was assessed by PSVS ( Sigmaplot 11.0 (Systat Software Inc.). Validation Suite‡). All structural comparisons throughout

Site-directed mutagenesis

† http://www.nmr.ch NLRPP12 PYD mutants W45A, W45E, W45R, W45F, and ‡ http://psvs-14-dev.nesg.org W45I were generated using the QuickChange Mutagenesis 802 The NLRP12 Pyrin Domain kit (Agilent). All mutated DNA constructs were sequence family: a standard nomenclature. Immunity, 28, verified (Beckman Coulter). 285–287. 10. Martinon, F., Burns, K. & Tschopp, J. (2002). The inflammasome: a molecular platform triggering acti- Accession numbers vation of inflammatory and processing of proIL-beta. Mol. Cell, 10, 417–426. Chemical shift assignments of NLRP12 PYD were 11. Petrilli, V., Dostert, C., Muruve, D. A. & Tschopp, J. deposited in the Biological Magnetic Resonance Data (2007). The inflammasome: a danger sensing complex Bank under accession number 17305, and the atomic triggering innate immunity. Curr. Opin. Immunol. 19, coordinates were submitted to the PDB under accession 615–622. code 2L6A. 12. Fiorentino, L., Stehlik, C., Oliveira, V., Ariza, M. E., Godzik, A. & Reed, J. C. (2002). A novel PAAD- containing protein that modulates NF-kappa B induc- tion by cytokines tumor factor-alpha and interleukin-1beta. J. Biol. Chem. 277, 35333–35340. Acknowledgements 13. Lich, J. D., Williams, K. L., Moore, C. B., Arthur, J. C., Davis, B. K., Taxman, D. J. & Ting, J. P. (2007). The authors thank Dr. Rebecca Page for discussion Monarch-1 suppresses non-canonical NF-kappaB ac- tivation and p52-dependent chemokine expression in and careful reading of the manuscript. This work monocytes. J. Immunol. 178, 1256–1260. was supported by European Union MCEXT-033534 14. Ting, J. P., Duncan, J. A. & Lei, Y. (2010). How the to R.S., Austrian Science Fund W1213 to C.E., and noninflammasome NLRs function in the innate National Institutes of Health R01NS056128 to W.P. immune system. Science, 327, 286–290. NMR data (800 MHz) were recorded on the 800- 15. Williams, J. (2009). Cutting Edge: a novel lab-on-a- MHz spectrometer at Brandeis University funded by tube for multimodality neuromonitoring of patients National Institutes of Health S10-RR017269. with traumatic brain injury (TBI). Lab Chip, 9, 1987. 16. Jeru, I., Duquesnoy, P., Fernandes-Alnemri, T., Cochet, E., Yu, J. W., Lackmy-Port-Lis, M. et al. Supplementary Data (2008). Mutations in NALP12 cause hereditary peri- odic fever syndromes. Proc. Natl Acad. Sci. USA, 105, 1614–1619. Supplementary data associated with this article 17. Macaluso, F., Nothnagel, M., Parwez, Q., Petrasch- can be found, in the online version, at doi:10.1016/ Parwez, E., Bechara, F. G., Epplen, J. T. & Hoffjan, S. j.jmb.2011.09.024 (2007). Polymorphisms in NACHT-LRR (NLR) in atopic dermatitis. Exp. Dermatol. 16, 692–698. 18. Arthur, J. C., Lich, J. D., Aziz, R. K., Kotb, M. & Ting, References J. P. (2007). Heat shock protein 90 associates with monarch-1 and regulates its ability to promote 1. Fritz, J. H., Ferrero, R. L., Philpott, D. J. & Girardin, degradation of NF-kappaB-inducing kinase. J. Immu- S. E. (2006). Nod-like proteins in immunity, inflam- nol. 179, 6291–6296. mation and . Nat. Immunol. 7, 1250–1257. 19. Kinoshita, T., Kondoh, C., Hasegawa, M., Imamura, 2. Kanneganti, T. D., Lamkanfi, M. & Nunez, G. (2007). R. & Suda, T. (2006). Fas-associated factor 1 is a Intracellular NOD-like receptors in host defense and negative regulator of PYRIN-containing Apaf-1-like disease. Immunity, 27, 549–559. protein 1. Int. Immunol. 18, 1701–1706. 3. Meylan, E., Tschopp, J. & Karin, M. (2006). Intracel- 20. Park, M. Y., Moon, J. H., Lee, K. S., Choi, H. I., lular pattern recognition receptors in the host Chung, J., Hong, H. J. & Kim, E. (2007). FAF1 response. Nature, 442,39–44. suppresses IkappaB kinase (IKK) activation by 4. Martinon, F. & Tschopp, J. (2005). NLRs join TLRs as disrupting the IKK complex assembly. J. Biol. innate sensors of pathogens. Trends Immunol. 26, Chem. 282, 27572–27577. 447–454. 21. Song, J., Park, J. K., Lee, J. J., Choi, Y. S., Ryu, K. S., 5. Faustin, B. & Reed, J. C. (2008). Sunburned skin Kim, J. H. et al. (2009). Structure and interaction of activates . Trends Cell Biol. 18,4–8. ubiquitin-associated domain of human Fas-associated 6. Inohara, N. & Nunez, G. (2003). NODs: intracellular factor 1. Protein Sci. 18, 2265–2276. proteins involved in inflammation and apoptosis. Nat. 22. Ting, J. P., Willingham, S. B. & Bergstralh, D. T. (2008). Rev., Immunol. 3, 371–382. NLRs at the intersection of cell death and immunity. 7. Shaw, P. J., McDermott, M. F. & Kanneganti, T. D. Nat. Rev., Immunol. 8, 372–379. (2011). Inflammasomes and autoimmunity. Trends 23. de Sa Pinheiro, A., Ehart, A., Ebner, N., Proell, M., Mol. Med. 17,57–64. Schwarzenbacher, R. & Peti, W. (2009). Backbone and 8. Proell, M., Riedl, S. J., Fritz, J. H., Rojas, A. M. & sidechain (1)H, (15)N and (13)C assignments of the Schwarzenbacher, R. (2008). The Nod-like receptor NLRP7 pyrin domain. Biomol. NMR Assign. 3, (NLR) family: a tale of similarities and differences. 207–209. PLoS One, 3, e2119. 24.Pinheiro,A.S.,Proell,M.,Eibl,C.,Page,R., 9. Ting, J. P., Lovering, R. C., Alnemri, E. S., Bertin, J., Schwarzenbacher, R. & Peti, W. (2010). Three-dimen- Boss, J. M., Davis, B. K. et al. (2008). The NLR gene sional structure of the NLRP7 pyrin domain: insight The NLRP12 Pyrin Domain 803

into pyrin–pyrin-mediated effector domain signaling 37. Espejo, F. & Patarroyo, M. E. (2006). Determining the in innate immunity. J. Biol. Chem. 285, 27402–27410. 3D structure of human ASC2 protein involved in 25. de Alba, E. (2009). Structure and interdomain dynam- apoptosis and inflammation. Biochem. Biophys. Res. ics of apoptosis-associated speck-like protein contain- Commun. 340, 860–864. ing a CARD (ASC). J. Biol. Chem. 284, 32932–32941. 38. Martinon, F., Hofmann, K. & Tschopp, J. (2001). The 26. Hiller, S., Kohl, A., Fiorito, F., Herrmann, T., Wider, pyrin domain: a possible member of the death G., Tschopp, J. et al. (2003). NMR structure of the domain-fold family implicated in apoptosis and apoptosis- and inflammation-related NALP1 pyrin inflammation. Curr. Biol. 11, R118–R120. domain. Structure, 11, 1199–1205. 39. Bauernfeind, F., Ablasser, A., Bartok, E., Kim, S., 27. Natarajan, A., Ghose, R. & Hill, J. M. (2006). Structure Schmid-Burgk, J., Cavlar, T. & Hornung, V. (2011). and dynamics of ASC2, a pyrin domain-only protein Inflammasomes: current understanding and open that regulates inflammatory signaling. J. Biol. Chem. questions. Cell. Mol. Life Sci. 68, 765–783. 281, 31863–31875. 40. Fontalba, A., Gutierrez, O. & Fernandez-Luna, J. L. 28. Srimathi, T., Robbins, S. L., Dubas, R. L., Chang, H., (2007). NLRP2, an inhibitor of the NF-kappaB Cheng, H., Roder, H. & Park, Y. C. (2008). Mapping of pathway, is transcriptionally activated by NF-kappaB POP1-binding site on pyrin domain of ASC. J. Biol. and exhibits a nonfunctional allelic variant. J. Immu- Chem. 283, 15390–15398. nol. 179, 8519–8524. 29. Liepinsh, E., Barbals, R., Dahl, E., Sharipo, A., Staub, 41. Sheffield, P., Garrard, S. & Derewenda, Z. (1999). E. & Otting, G. (2003). The death-domain fold of the Overcoming expression and purification problems of ASC PYRIN domain, presenting a basis for PYRIN/ RhoGDI using a family of “parallel” expression PYRIN recognition. J. Mol. Biol. 332, 1155–1163. vectors. Protein Expression Purif. 15,34–39. 30. Güntert, P. (2004). Automated NMR structure calcu- 42.Peti,W.&Page,R.(2007).Strategiestomaximize lation with CYANA. Methods Mol. Biol. 278, 353–378. heterologousproteinexpressioninEscherichiacoli 31. Brünger, A. T., Adams, P. D., Clore, G. M., DeLano, with minimal cost. Protein Expression Purif. 51, W. L., Gros, P., Grosse-Kunstleve, R. W. et al. (1998). 1–10. Crystallography & NMR system: a new software suite 43. Herrmann, T., Guntert, P. & Wuthrich, K. (2002). for macromolecular structure determination. Acta Protein NMR structure determination with automated Crystallogr., Sect. D: Biol. Crystallogr. 54, 905–921. NOE-identification in the NOESY spectra using the 32. Richter, S., Wenzel, A., Stein, M., Gabdoulline, R. R. & new software ATNOS. J. Biomol. NMR, 24, 171–189. Wade, R. C. (2008). webPIPSA: a web server for the 44. Herrmann, T., Güntert, P. & Wüthrich, K. (2002). comparison of protein interaction properties. Nucleic Protein NMR structure determination with automat- Acids Res. 36, W276–W280. ed NOE assignment using the new software 33. Qin, H., Srinivasula, S. M., Wu, G., Fernandes- CANDID and the torsion angle dynamics algorithm Alnemri, T., Alnemri, E. S. & Shi, Y. (1999). Structural DYANA. J. Mol. Biol. 319, 209–227. basis of procaspase-9 recruitment by the apoptotic 45. Güntert, P. (2009). Automated structure determina- protease-activating factor 1. Nature, 399, 549–557. tion from NMR spectra. Eur. Biophys. J. 38, 129–143. 34. Esposito, D., Sankar, A., Morgner, N., Robinson, C. V., 46. Nederveen, A. J., Doreleijers, J. F., Vranken, W., Rittinger, K. & Driscoll, P. C. (2010). Solution NMR Miller, Z., Spronk, C. A., Nabuurs, S. B. et al. (2005). investigation of the CD95/FADD homotypic death RECOORD: a recalculated coordinate database of domain complex suggests lack of engagement of the 500+ proteins from the PDB using restraints from CD95 C terminus. Structure, 18, 1378–1390. the BioMagResBank. Proteins, 59, 662–672. 35. Park, H. H., Lo, Y. C., Lin, S. C., Wang, L., Yang, J. K. 47. Delaglio, F., Grzesiek, S., Vuister, G. W., Zhu, G., & Wu, H. (2007). The death domain superfamily in Pfeifer, J. & Bax, A. (1995). NMRPipe: a multidimen- intracellular signaling of apoptosis and inflammation. sional spectral processing system based on UNIX Annu. Rev. Immunol. 25, 561–586. pipes. J. Biomol. NMR, 6, 277–293. 36. Reed, J. C., Doctor, K. S. & Godzik, A. (2004). The 48. Johnson, B. A. (2004). Using NMRView to visualize domains of apoptosis: a genomics perspective. Sci. and analyze the NMR spectra of macromolecules. STKE, 2004, re9. Methods Mol. Biol. 278, 313–352.