Critical Role for Cryopyrin/Nalp3 in Activation of Caspase-1 in Response to Viral Infection and Double-stranded RNA*>

Viral infection induces the production of interleukin (IL)-1β and IL-18 in macrophages through the activation of caspase-1, but the mechanism by which host cells sense viruses to induce caspase-1 activation is unknown. In this report, we have identified a signaling pathway leading to caspase-1 activation that is induced by double-stranded RNA (dsRNA) and viral infection that is mediated by Cryopyrin/Nalp3. Stimulation of macrophages with dsRNA, viral RNA, or its analog poly(I:C) induced the secretion of IL-1β and IL-18 in a cryopyrin-dependent manner. Consistently, caspase-1 activation triggered by poly(I:C), dsRNA, and viral RNA was abrogated in macrophages lacking cryopyrin or the adaptor ASC (apoptosis-associated speck-like protein containing a caspase-activating and recruitment domain) but proceeded normally in macrophages deficient in Toll-like receptor 3 or 7. We have also shown that infection with Sendai and influenza viruses activates the cryopyrin inflammasome. Finally, cryopyrin was required for IL-1β production in response to poly(I:C) in vivo. These results identify a mechanism mediated by cryopyrin and ASC that links dsRNA and viral infection to caspase-1 activation resulting in IL-1β and IL-18 production.

Innate immunity is the initial line of host defense against microbial pathogens, including viral infection. The early recognition of viruses by the host initiates signaling pathways leading to the induction of anti-viral responses, including the secretion of type I interferons (IFNs) 5 ␣ and ␤ and pro-inflammatory cytokines (1)(2)(3). Recognition of pathogens including viruses by the host immune system relies on the detection of conserved molecular structures that are shared by large pathogens (pathogen-associated molecular patterns or PAMPs). In the case of viruses, genomic DNA and single-stranded RNA (ssRNA) or double-stranded RNA (dsRNA) produced during viral replication are sensed by host cells to induce anti-viral responses (3). Viral DNA, ssRNA, and dsRNA are recognized by a subfamily of Toll-like receptors (TLRs 3, 7, 8, and 9) in endosomes after the endocytosis of viral particles in mammalian cells (4 -9). Upon activation, these TLRs recruit the adaptor proteins MyD88 or TRIF (Toll/IL-1 receptor domain-containing adaptor-inducing interferon-␤) to activate the IkB kinase (IKK) complex and the IKKrelated kinases TBK1 and IKK⑀ (1,10). Viral dsRNA is recognized by TLR3 and the two cytosolic RNA helicases, the retinoic acid inducible gene I (RIG-I), and MDA5 (which induces activation of IKK and TBK1/IKK⑀) (4,11,12). Activation of the IKK complex that includes the catalytic subunits IKK␣ and IKK␤ as well as the regulatory subunit IKK␥/NEMO mediates NFB activation, whereas that of TBK1 and IKK⑀ induces phosphorylation and activation of IRF3 or IRF7 (10,13,14). The nuclear translocation of NFB and IRF3/IRF7 mediates the transcriptional activation of interferon and cytokine genes that limit viral replication and promote adaptive immune responses (15)(16)(17).
Signaling pathways other than NFB or IFN that are activated upon viral recognition and mediate anti-viral responses are poorly understood. Previous studies have shown that infection of macrophages with certain viruses including influenza A and Sendai virus induce interleukin (IL)-1␤ and IL-18 secretion (18,19), but the mechanism by which host cells sense viruses to induce caspase-1 activation is unknown. Both IL-1␤ and IL-18 are synthesized as inactive cytoplasmic precursors that are processed into biologically active mature forms in response to various pro-inflammatory stimuli, including viruses by caspase-1, a cysteine protease (18,20,21). Caspase-1 is synthesized as an inactive zymogen that becomes activated by cleavage at aspartic residues to generate an enzymatically active heterodimer composed of a 10-and 20-kDa chain (20). Recent studies have implicated members of the NOD-like receptor (NLR) family of proteins (also called NOD-LRR (nucleotide binding oligomerization domain-(leucine-rich repeat) or CATERPILLER) in the regulation of caspase-1 activation in response to microbial pathogens (22,23). The NLR family is composed of 23 cytosolic proteins, including Nod1, Nod2, Cryopyrin/Nalp3, and Ipaf. The structure of NLRs include an amino-terminal effector binding region that consists of protein-protein interaction domains, such as caspase-recruitment domains or pyrin (a central nucleotide binding oligomerization domain that acts to oligomerize these proteins) and carboxyl-terminal leucine-rich repeats that are required to detect specific PAMPs (22). Cryopyrin forms an endogenous multiprotein complex containing ASC (apoptosis-associated speck-like protein containing a caspase-activating and recruitment domain) and caspase-1 dubbed "the inflammasome," which promotes caspase activation and processing of pro-IL-1␤ (24,25). Notably, missense mutations in the CIAS1 gene that encodes Cryopyrin cause three autoinflammatory disorders characterized by deregulated production of IL-1␤ (26). Cryopyrin senses bacterial RNA, synthetic anti-viral purine analogs, and monosodium urate or calcium pyrophosphate dehydrate crystals (27,28). In addition, other results indicate that Cryopyrin regulates caspase-1 activation in response to factors that induce intracellular K ϩ efflux, such as certain toxins and high concentrations of extracellular ATP (29,30).
The genome of poxviruses encode Pyrin-containing proteins that interact with components of the inflammasome and inhibit caspase-1 activation and the processing of IL-1␤ and IL-18 induced by diverse stimuli (31). These results suggest that certain viruses target components of caspase-1 activation pathways to circumvent host anti-viral responses. However, the signaling pathways that link viral infection to caspase-1 activation and IL-1␤/IL-18 are unknown. Here we show that viral dsRNA and its analog poly(I:C) as well as viral infection activate caspase-1 through cryopyrin, resulting in the production of active IL-1␤ and IL-18. The signaling pathway stimulated by dsRNA and cryopyrin was independent of TLR3 or TLR7. These results identify a novel TLR-independent signaling pathway that is mediated by cryopyrin and ASC and leads to the secretion of pro-inflammatory cytokines in response to viral infection.
Macrophages-Bone marrow was prepared from the leg bones of 5-20-week-old mice. The legs were dissected and the bone marrow flushed out. Bone marrow cells were cultured with Iscove's modified Dulbecco's medium supplemented with 30% L929 supernatant containing macrophage-stimulating factor, glutamine, sodium pyruvate, 10% heat-inactivated fetal bovine serum (Invitrogen), 50 g/ml penicillin, and 50 g/ml streptomycin at 37°C in 5% CO 2 for 5 days (bone marrow differentiation medium). Bone marrow macrophages were then harvested with rubber scrapers and seeded. Peritoneal macrophages were elicited to the peritoneum of mice and isolated 4 days after the injection of 4% thioglycolate broth. After 1 day, non-adherent cells were removed, and the remaining macrophages were incubated in Iscove's modified Dulbecco's medium supplemented with 10% heat-inactivated fetal bovine serum, 50 g/ml penicillin, and 50 g/ml streptomycin.
In Vitro RNA Transcription, Viruses, and Viral RNA-The dsRNA for hsp90 and LacZ were made according to the protocol described in Ref. 34. To prepare flock house virus (FHV) dsRNA, the plasmids pFHV2(0,0) and p2VHF(2,0) (35) were used to generate viral (ϩ) and (Ϫ) ssRNAs, respectively. Plasmids were linearized with RsrII, in vitro transcripts were synthesized using T7 polymerase and an Ambion Megascript kit per the manufacturer's instructions, and RNA was purified by phenol-chloroform extraction and ethanol precipitation and quantitated by spectrophotometry. Equal amounts of (ϩ) and (Ϫ) RNA transcripts were mixed in RNase-free water, heated to 75°C for 30 min, and allowed to cool gradually at room temperature to form dsRNA. Viral RNA integrity and formation of dsRNA were assessed by non-denaturing agarose gel electrophoresis. Murine Sendai virus (strain Cantell) was purchased from the American Type Culture Collection and influenza A/Puerto Rico/8/34 virus (H1N1) was a gift from James R. Becker, Jr. (University of Michigan). Rotavirus (strain SA11-4F) dsRNA was isolated by phenol-chloroform extraction from virus grown in monkey kidney (MA014) cells and purified by CsCl centrifugation (36).
Microbial Ligands and Antibodies-Pam2CGDPKHPKSF (FSL-1), Pam3CSK4, poly(I:C), and CpG oligonucleotide (5Ј-TCCATGACGTTCCTGACGTT-3Ј) were purchased from Invivogen. poly(A), (C), (G), and (U) ssRNA and poly(I)⅐poly(C) dsRNA were from Sigma. Purified total RNA from Escherichia coli was purchased from Ambion. Bacillus anthracis protective antigen and lethal factor were obtained from List Biologicals Laboratories and were used at 1 g/ml concentration. Aliquots of RNA samples were incubated with RNases (Ambion) as suggested by the manufacturer. Rabbit anti-mouse caspase-1 was a generous gift of P. Vandenabeele (Ghent University, Ghent, Belgium). The antibodies for mouse IB␣, phospho-IB␣, p38, phospho-p38, ERK, and phospho-ERK were from Cell Signaling Technologies.

Immunoblotting-Cells
were washed twice with phosphate-buffered saline and scraped in lysis buffer solution (150 mM NaCl, 10 mM Tris, pH 7.4, 5 mM EDTA, 1 mM EGTA, 0.1% Nonidet P-40) supplemented with 1ϫ protease inhibitor mixture (Roche Diagnostics). For analysis of caspase-1 activation, macrophages were cultured with stimuli for 1-3 h and then with medium containing 5 mM ATP (Sigma) for 30 min. Extracts were prepared from cells and culture supernatants by adding lysis buffer containing 1% Nonidet P-40 supplemented with complete protease inhibitor mixture (Roche Diagnostics, Mannheim, Germany) and 2 mM dithiothreitol. Samples were clarified, denatured with SDS buffer, and boiled for 5 min, separated by SDS-PAGE, and transferred to nitrocellulose membranes. The membranes were immunoblotted with primary antibodies and proteins detected with appropriate secondary anti-rabbit antibody conjugated to horseradish peroxidase followed by enhanced chemiluminescence.
Measurements of Cytokines-Macrophages were stimulated with various microbial and synthetic ligands for 24 h and the supernatants were analyzed for IL-1␤, IL-18, TNF␣, and IL-6 secretion. The ligand concentrations used were FLS-1 and Pam3CSK4 (at 1 g/ml) and poly(I:C) (at 2.5 g/ml). Mouse cytokines were measured in culture supernatants by enzyme-linked immunosorbent assay (ELISA) kits (R&D Systems, Minneapolis, MN).

Cryopyrin Is Required for IL-1␤ and IL-18 Secretion in Response
to the dsRNA Analog Poly(I:C)-To assess a role for cryopyrin in immune responses induced by viruses, macrophages from wildtype (WT) and cryopyrin-deficient mice were stimulated with polyinosinic-polycytidylic acid (poly(I:C), a synthetic dsRNA analog that mimics viral infection and a synthetic diacylated lipopeptide (FSL-1, TLR2 agonist) or E. coli RNA as controls. Incubation of wild-type macrophages with poly(I:C) induced IL-1␤, but this response was abolished in cryopyrin-null macrophages (Fig. 1, A  and B). Consistent with previous results, production of IL-1␤ in response to E. coli RNA (but not FSL-1) was deficient in cryopyrinnull macrophages (Fig. 1A). Additionally, cryopyrin was required for IL-18 induced by poly(I:C), but dispensable for IFN␣ produc- Macrophages from WT and Cryopyrin Ϫ/Ϫ mice (A and B) or WT and TLR3 Ϫ/Ϫ mice (C and D) were stimulated with the indicated concentrations of poly(I:C) for 24 h, and cell-free supernatants were analyzed by ELISA for the production of TNF␣ and IL-6. Macrophages from WT and cryopyrin Ϫ/Ϫ mice (E) or WT and TLR3 Ϫ/Ϫ mice (F) were stimulated with poly(I:C) for the indicated times. Cell extracts were immunoblotted with antibodies that detect total and phosphorylated (activated) forms of the indicated proteins. The data are means Ϯ S.D. of triplicates and are representation of at least two independent experiments. KO, knock-out; Unst, unstimulated. tion (Fig. 1, C and D). These results indicate that cryopyrin is specifically required for IL-1␤/IL-18 secretion induced by poly(I:C).
Cryopyrin-dependent IL-1␤ Secretion by Poly(I:C) Is Independent of NFB and MAPK Activation-Stimulation of macrophages with poly(I:C) induces the secretion of several pro-inflammatory cytokines (4). We found that cryopyrin was dispensable for the production of TNF␣ and IL-6 induced by poly(I:C) (Fig. 2, A and B). In contrast, production of TNF␣ and IL-6 required TLR3 (Fig. 2, C and D), as previously reported (4). The induction of IL-1␤ secretion is thought to involve the up-regulation of pro-IL-1␤ through transcriptional mechanisms via NFB and then a second stimulus that leads to the activation of caspase-1, processing of pro-IL-1␤, and release of mature IL-1␤ (37,38). Stimulation with poly(I:C) induced comparable levels of NFB, ERK, and p38 activation in WT and cryopyrin Ϫ/Ϫ macrophages (Fig. 2E). By contrast, the activation of NFB and MAPKs was abolished in TLR3deficient macrophages (Fig. 2F). These results demonstrate that cryopyrin-mediated IL-1␤ secretion induced by poly(I:C) is independent of NFB and MAPK as well as TLR3.
Cryopyrin (but Not Nod2/TLR3/ TLR7) Is Essential for Activation of Caspase-1 in Response to Poly(I:C)-Proteolytic activation of procaspase-1 is a critical step in the  induction of IL-1␤ secretion (39,40). Importantly, processing of pro-caspase-1 was induced rapidly (by 1 h) and in a dose-dependent manner after stimulation of WT macrophages with poly(I:C), as determined by the detection of the mature 20-kDa subunit of caspase-1 (Fig. 3, A and B). Such activation of caspase-1 was abrogated in macrophages lacking cryopyrin (Fig. 3, A and B) or ASC (Fig. 3C), an adaptor that links cryopyrin to caspase-1 (25,41). In contrast, activation of caspase-1 induced by poly(I:C) was unimpaired in Nod2-and TLR3-or TLR7-deficient macrophages (Fig. 3,  D-F). These results demonstrate that cryopyrin is essential for caspase-1 processing in response to poly(I:C). Furthermore, TLR3 is required for NFB and MAPK activation but dispensable for caspase-1 activation.
Cryopyrin-dependent IL-1␤ and IL-18 Secretion and Caspase-1 Activation by Poly(I:C) Requires dsRNA Structure -We next tested several ssRNA and dsDNA compounds to determine the structural requirements for IL-1␤ and IL-18 secretion and caspase-1 activation induced by poly(I:C). Importantly, the synthetic ssRNA analogs (polycytidylic acid (poly(C), polyuri-dylic acid (poly(U), or polyinosinic (poly(I)) neither induce IL-1␤ and IL-18 secretion (Fig. 4, A and B) nor caspase-1 activation (Fig. 4C). Moreover, the dsDNA analogs polydeoxyinosinic-deoxycytidylic acid (poly(dI: dC)) and polydeoxyinosinic-deoxcytidylic acid (poly(dG:dC)) did not have any effect on IL-1␤ secretion or caspase-1 activation (Fig. 4, A-C). To further verify these results, we treated poly(I:C) with a panel of RNases that cleave ssRNA, dsRNA, or both. Digestion of poly(I:C) with RNase A and RNase T1 that are specific for ssRNA had little or no effect on caspase-1 activation induced by poly(I:C) (Fig. 4D). In contrast, treatment of poly(I:C) with RNase V1 that cleaves dsRNA or with benzonase that digests both ssRNA and dsRNA abolished its ability to induce processing of caspase-1 (Fig.  4D). Together, these results indicate that the dsRNA structure of poly(I:C) is essential for cryopyrindependent induction of IL-1␤ secretion and caspase-1 activation in macrophages.
Viral and Non-viral dsRNA Produced in Vitro Induce Cryopyrin-dependent Caspase-1 Activation-Poly(I:C) does not represent all viral and non-viral dsRNA. To assess whether in vitro transcribed dsRNAs induce caspase-1 activation, we produced ϳ700 bp dsRNA fragments from the 5Ј coding region of the lacZ and Drosophila Hsp83 genes using the in vitro transcription system (34,35). In addition, complementary plus sense (ϩ) and minus sense (Ϫ) ssRNAs of FHV were produced by in vitro transcription and annealed to form viral dsRNA. Stimulation of macrophages with FHV dsRNA as well as nonviral dsRNAs induced caspase-1 activation as effectively as the poly(I:C) that was used as a control (Fig. 5A). As expected, neither the (ϩ) ssRNA nor the (Ϫ) ssRNA used in preparing FHV dsRNA was able to induce processing of caspase-1 (Fig. 5B). As observed with poly(I:C), digestion of FHV dsRNA with RNases that digest dsRNA (but not ssRNA) abolished their ability to induce caspase-1 activation (Fig. 5, C and D). Additionally, treatment of dsRNAs with E. coli RNase III, an endoribonuclease that cleaves dsRNA into 10 -18-bp dsRNA fragments, had no effect on their ability to induce caspase-1 activation (Fig. 5E).
Cryopyrin/ASC-dependent Activation of Caspase-1 by Naturally Produced Viral dsRNA-We next tested the ability of genomic dsRNA purified from rotavirus grown in monkey kidney cells to induce caspase-1 activation. Incubation of macro- A and B, macrophages from WT and cryopyrin Ϫ/Ϫ mice were stimulated with in vitro produced viral RNA, non-viral dsRNA from LacZ, and Hsp83 gene plus sense (ϩ) and minus sense (Ϫ) ssRNAs and poly(I:C) for 3 h, and then cells were pulsed transiently with ATP for 30 min. Cell extracts were immunoblotted with a caspase-1 antibody. Arrows denote procaspase-1 and its processed p20 subunit. C-E, macrophages from WT mice were stimulated for 3 h with dsRNA preparations digested with the indicated RNases and then pulsed transiently with ATP for 30 min. Cell extracts were immunoblotted with a caspase-1 antibody. Results are representative of three separate experiments. KO, knock-out; Unstim, unstimulated; Hsp, heat shock protein.
phages with naturally produced rotavirus dsRNA induced caspase-1 activation in WT but not cryopyrin-deficient cells (Fig. 6A). Similarly, cryopyrin was required for the activation of caspase-1 induced by viral dsRNA purified from plant cells infected with brome mosaic virus as well as by in vitro produced brome mosaic viral dsRNA. (Fig. 6B). In contrast, caspase-1 activation triggered by the lethal toxin of B. anthracis, consisting of the protective antigen and lethal factor that depends on Nalp1b (42), proceeded normally in the absence of cryopyrin (Fig. 6A). Consistent with the results obtained with poly(I:C), activation of caspase-1 induced by rotavirus dsRNA required ASC but not TLR3 or TLR7 (Fig. 6,  D-F). Furthermore, treatment of naturally produced rotavirus and dsRNA with RNase V1 (but not RNase III) abolished their ability to induce caspase-1 activation (Fig.  6G).
Infection of Macrophages with Sendai and Influenza Viruses Induce Cryopyrin-dependent Caspase-1 Activation-We next tested the ability of Sendai and influenza A virus, two viruses known to stimulate IL-1␤ secretion in human macrophages (18,21), to induce caspase-1 activation in WT and cryopyrin-deficient macrophages. Macrophages were infected with Sendai or influenza A virus at a low multiplicity of infection, and cell extracts were examined for pro-caspase-1 processing by immunoblotting. The analysis revealed that both viruses induced caspase-1 activation in WT but not cryopyrin-deficient macrophages (Fig. 7A). As it was observed with viral dsRNA, activation of caspase-1 triggered by viral infection was independent of TLR3 and TLR7 (Fig. 7B).
Cryopyrin Is Required for Production of IL-1␤ and IL-18 after Administration of Poly(I:C) in Vivo-We next examined whether cryopyrin plays a role in the production of pro-inflammatory cytokines in the animal. Intraperitoneal administration of poly(I:C) induced the production of IL-1␤, IL-18, TNF␣, and IL-6 in the serum of wild-type mice. Similarly, the bacterial lipopeptide Pam3CSK4 (TLR2 agonist) also induced the production of IL-1␤, TNF␣, and IL-6 in the serum of wild-type mice. (Fig.  8). In contrast, the serum levels of IL-1␤, IL-18, and to lesser extent IL-6 were greatly reduced, whereas those of TNF␣ were unimpaired in cryopyrin knock-out mice after injection with poly(I:C) (Fig. 8, A-D and G, and supplemental Fig. 1). The reduced levels of IL-6 detected in the serum of cryopyrinnull mice after stimulation with poly(I:C) is presumably due to the induction of IL-6 by IL-1␤ in vivo (43). Notably, the production of pro-inflammatory cytokines including IL-1␤ was unimpaired in cryopyrin-deficient mice after intraperitoneal administration of Pam3CSK4, demonstrating the specific role of cryopyrin in regulating poly(I:C) responses in vivo.

DISCUSSION
Infection of monocytes and macrophages with a variety of viruses is known to induce the secretion of IL-1␤ and IL-18 through the activation of caspase-1, but the molecular mechanisms involved have remained largely unknown. These studies demonstrate that cryopyrin plays an essential role in the secretion of IL-1␤ and IL-18 by sensing viral dsRNA and inducing the activation of caspase-1. Viral dsRNA also triggers the production of type I IFNs, TNF␣, and IL-6 through TLR3 and the RIG-1 stimulation (4,11), but these immune responses were independent of cryopyrin. The latter responses are mediated through the activation of the transcriptional factors IRF3/IRF7 and NFB (16,44). Thus, dsDNA induces at least three distinct defense-signaling pathways in host cells to limit viral infection.
IL-1␤ and IL-18 are induced by a variety of microbial stimuli through the activation of caspase-1. IL-1␤ is considered to be a master cytokine in that it mediates several innate and adaptive immune responses directly or through the induction of other cytokines such as IL-6 (43). In addition, IL-1␤ is a potent pyrogen that is involved in the development of fever in response to pathogen infection (45). However, there is little evidence for a role of IL-1␤ in host defense against viral infection. In contrast, IL-18, a cytokine that stimulates nature killer cells and CD8ϩ T cells and is potently synergistic with IL-12 for this function (46), is known to play an important  role in viral clearance (47,48). Furthermore, administration of IL-18 can protect the host against infection with herpes simplex and vaccinia virus (49,50). A role for IL-18 in viral infection is also supported by the finding that the genome of several poxviruses encodes proteins that bind and inactivate IL-18 (51).
Functional analyses of RNA compounds and enzymatic studies revealed that dsRNA (but not ssRNA) activate caspase-1 through cryopyrin. The ability of cryopyrin to discriminate between dsRNA and ssRNA provides a mechanism to sense viral RNA and to avoid harmful activation of the cryopyrin inflammasome by endogenous RNA. Treatment of 700 bp dsRNA fragments with RNase III, an endoribonuclease that cleaves both synthetic and natural dsRNA into small duplex products averaging 10 -18 bp in length (52), had no effect on its ability to induce caspase-1 activation. It is known that mammalian RNase III enzymes, such as Dicer, process dsRNA into 21-24-nt dsRNAs that can induce degradation of homologous mRNAs and specific gene silencing (53). It is also known that certain siRNAs can induce IFN responses and toxic effects in mammalian cells (54). Our results suggest both long viral and endogenous dsRNA fragments or their shorter dsRNA products generated by RNase III enzymes might lead to cryopyrin activation and IL-1␤/IL-18 secretion.
There are several mechanisms by which dsRNA could induce the activation of caspase-1 through cryopyrin. Viral dsRNA is produced during viral infection in the cytosol of host cells and could be sensed directly by cryopyrin. There is evidence that several NLR proteins, including Nod1, Nod2, Ipaf, and cryopyrin, can recognize microbial structures (22,23). However, there is no evidence that the microbial products physically interact with the leucine-rich repeats of these NLR proteins (55). The sensing of dsRNA by cryopyrin is most likely to be indirect, as it has been proposed for the recognition of microbial components by plant NLR homologs (55,56). Cryopyrin has been shown to form a large multiprotein complex containing ASC that promotes the activation of caspase-1 in cell-free systems (25). Thus, a possible model is that the leucine-rich repeats of cryopyrin sense dsRNA produced during viral infection, which induces the oligomerization of cryopyrin and recruitment of caspase-1 via ASC and possibly other factors involved in caspase-1 activation.
In summary, our results have shown a novel role for cryopyrin in the activation of inflammasome by dsRNA/viral RNA and by viral infection. These findings provide important insight into the role of NLR proteins in the host response to dsRNA. The possibility that certain siRNAs might activate the cryopyrin inflammasome warrant further investigation, as such siRNAs are being explored for their potential therapeutic use (57,58).