Caspase-mediated interleukin-18 maturation promotes membrane-associated unconventional secretion and signal transduction in a teleost fish
Liming Yuan, Hang Xu, Shuai Jiang
TL;DR
This study explores how a proinflammatory cytokine, IL-18, is processed and secreted in a type of fish, revealing a unique immune regulation mechanism.
Contribution
The study identifies caspase-mediated maturation and unconventional secretion of IL-18 in a teleost fish, revealing a novel immune regulatory mechanism.
Findings
Pro-IL-18 is cleaved by multiple caspases to generate mature IL-18 forms.
Mature IL-18 is secreted via membrane-associated microvesicles and activates immune signaling.
IL-18 signaling enhances antimicrobial immunity and reduces mortality in fish.
Abstract
Interleukin-18 (IL-18) is an important proinflammatory cytokine essential for immune modulation. Unlike most cytokines, it is synthesized as an inactive precursor, with its maturation and secretion being critical for its functionality. As an evolutionarily ancient cytokine, it can be traced back to teleosts, but not zebrafish. However, the regulatory mechanism of IL-18 in early vertebrates remains largely elusive. The present study reports the maturation and secretion of IL-18 along with its role in signal transduction in a teleost fish half-smooth tongue sole (Cynoglossus semilaevis). We found that pro-IL-18 was cleaved by caspase-1, caspase-3/7, and caspase-6 at different N-terminal sites, generating three forms of the mature IL-18. In contrast to the negatively charged pro-IL-18, the positively charged mature IL-18 is highly enriched in the cytoplasmic membrane. It is enclosed within…
Genes, proteins, chemicals, diseases, species, mutations and cell lines named across the full text — each resolved to its canonical identifier and authoritative record.
Click any figure to enlarge with its caption.
Figure 1
Figure 2
Figure 3
Figure 4
Figure 5
Figure 6Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
Taxonomy
TopicsInflammasome and immune disorders · Marine Invertebrate Physiology and Ecology · Psoriasis: Treatment and Pathogenesis
Introduction
Cytokines play essential roles in immune regulation and countering pathogenic infections (Liongue et al. 2016; Xu et al. 2022a, b). As an important proinflammatory cytokine, interleukin (IL)-18 is primarily known for its potent interferon-γ-inducing activity and significance in innate immune activation (Kaplanski 2018). Unlike many other cytokines, IL-18 is initially synthesized as an inactive precursor (pro-IL-18) that is proteolytically cleaved for activation (Afonina et al. 2015). In mammals, pro-IL-18 is typically cleaved by proinflammatory caspase-1, a central component of the canonical inflammasome pathway, at the N-terminal, tetrapeptide motif “33_LESD_36” (Exconde et al. 2023). This process removes a short propeptide from the N-terminus, generating a long C-terminal polypeptide that contains the minimal active domain (Kim et al. 2001). In addition to caspase-1, recent studies have reported that the human-specific, noncanonical, inflammasome activator caspase-4 was also capable of processing pro-IL-18 at the same motif, producing the bioactive IL-18 (Devant et al. 2023; Shi et al. 2023). In contrast to most cytokines that are secreted through the conventional endoplasmic reticulum (ER)–Golgi secretory pathway, the mature IL-18 lacks an ER-signal peptide and has to exit the cells through the ER- and Golgi-independent pathways (Gracie et al. 2003). To date, studies on such unconventional IL-18 secretion pathways are limited, with the general assumption being that IL-18 shares a similar secretion route with IL-1β (Monteleone et al. 2015). For example, canonical and noncanonical inflammasome-induced cell death facilitates IL-1β and IL-18 secretion (Yi 2017). In such a scenario, the pore-forming protein, gasdermin D (GSDMD), is activated, which perforates and lyses the cytoplasmic membrane (Ding et al. 2016; Kayagaki et al. 2015). Membrane rupture eventually leads to a massive release of the intracellular contents, including IL-18 and IL-1β, bypassing the conventional ER–Golgi secretory pathway. Apart from such secretion, the release of membrane-associated vesicles by living cells serves as an alternate route for unconventional IL-18 secretion. In human cells, endosome-derived exosomes and endolysosomes enclose the cytokines and a few other proinflammatory factors during secretion (Cypryk et al. 2018; Noonin and Thongboonkerd 2021; Piccioli and Rubartelli 2013).
After release into the extracellular milieu, interaction with the IL-18 receptor (IL-18R) complex is critical to activate the downstream immune responses (Sims 2002). In humans, this complex consists of two subunits: IL-18Rα and IL-18Rβ. IL-18Rα binds to IL-18 via its extracellular immunoglobulin domain (Kato et al. 2003; Tsutsumi et al. 2014). The interaction induces conformational changes of the IL-18Rα/IL-18 heterodimer, allowing the recruitment of IL-18Rβ to form a high-affinity, ligand–receptor heterotrimeric complex (Wu et al. 2003). This complex triggers the activation of intracellular signaling cascades, primarily through the recruitment of the myeloid differentiation primary response 88 (MyD88) protein, subsequently activating the members of the IL-1R-associated kinase (IRAK) family of serine/threonine kinases (Novick et al. 2013). These signaling pathways converge to activate the nuclear factor κB (NF-κB) and mitogen-activated protein kinase cascades, resulting in the production of proinflammatory cytokines and immune-reaction mediators (Rex et al. 2020; Sedimbi et al. 2013). IL-18 signaling is particularly crucial for innate immunity against pathogen infection (Vecchie et al. 2021). In mice, infection by the fungal pathogen Paracoccidioides brasiliensis activates the nucleotide-binding leucine-rich repeat receptors family of pyrin domain-containing 3 (Nlrp3) inflammasome that mediates IL-18 maturation and secretion, whereas IL-18-knockout exacerbates infection and reduces survivability (Ketelut-Carneiro et al. 2015; Rodrigues et al. 2014). Stimulation by the bacterial protein, flagellin, elevates IL-18 production, inducing the death of rotavirus-infected cells and disrupting viral replication (Zhang et al. 2020). In addition to immune cells, intestinal neurons can also express and secrete IL-18, which enhances the synthesis of antimicrobial proteins and promotes mucosal barrier-based immunity against Salmonella typhimurium (Jarret et al. 2020).
IL-18 is an evolutionarily ancient IL-1 family member that can be traced back to teleost fishes (Wang and Secombes 2013; Zou and Secombes 2016). As a typical proinflammatory cytokine, IL-18 has an intimate association with fish immunity (Huising et al. 2004; Ogryzko et al. 2014). In turbots, IL-18 is mainly expressed in the immune function-related tissues, and is significantly upregulated in response to bacterial infections (Pereiro et al. 2021; Yu et al. 2024). In snakeheads, IL-18-mediated stimulation promotes the expression of IFN-γ and a few other proinflammatory cytokines (Cui et al. 2021). In addition to IL-18, its receptors associated with signal transduction have also been identified in teleost (Rivers-Auty et al. 2018). IL-18Rα and IL-18Rβ in rainbow trout possess the same conserved functional domains as their mammalian homologs (Yang et al. 2023; Zou et al. 2004). When expressed in cells, the rainbow trout IL-18Rβ can interact with elements of the MyD88-, IRAK-, and NF-κB-based signaling pathways, indicating evolutionary conservation in intracellular signal transduction in teleost. Although IL-18 exerts critical effects on immune responses, the regulation of IL-18-based signaling in teleost remains largely elusive.
The present study investigated the maturation, secretion, and receptor-mediated immune functions of IL-18 in the teleost, half-smooth tongue sole (Cynoglossus semilaevis). In contrast to the human IL-18, which is activated by proinflammatory caspase-1 and inactivated by apoptotic caspase-3 (Akita et al. 1997), tongue sole IL-18 could be proteolytically activated by proinflammatory caspase-1 as well as apoptotic caspase-3/7 and caspase-6. The cleavages occurred at three distinct N-terminal sites and generated three mature IL-18 forms with increasing electrostatic charges. Such electrostatic complementarity-based association promotes the recruitment of the mature IL-18 to the cytoplasmic membrane and its secretion via membrane-derived microvesicles. We further illustrated the interactions of IL-18 with its receptors and found that the mature IL-18 specifically interacted with IL-18Rα, followed by association with IL-18Rβ, in a way similar to that in humans. Bacterial infection increases the production, maturation, and secretion of IL-18, while IL-18-based signaling pathway activation enhances antimicrobial immunity and reduces infection-associated mortality.
Materials and methods
Ethics statement
Clinically healthy half-smooth tongue sole were obtained from a local fish farm in Shandong Province, China, and maintained in the laboratory. All animal experiments were approved by the Ethics Committee of the Institute of Oceanology, Chinese Academy of Sciences, Qingdao, China, and followed the prescribed guidelines.
Cells and bacterial strains
Tongue sole peripheral blood leukocytes were prepared as previously described (Li et al. 2023). Briefly, the blood was collected and laid on 61% Percoll and centrifuged at 400* g* for 15 min. The white layer containing the leukocytes was collected, washed with PBS, and resuspended in L-15 medium. HEK293T and HeLa cells were cultured in DMEM + 10% FBS (Life Technologies, Renfrewshire, UK). Vibrio harveyi strains were cultured in Luria–Bertani (LB) broth at 28 °C with continuous shaking.
Sequence and phylogenetic analysis
Sequences were analyzed using the ClustalW program (http://www.clustal.org/), and the results were visualized utilizing ESPript 3.0 (https://espript.ibcp.fr/ESPript/ESPript/). Phylogeny was analyzed using the neighbor-joining algorithm in MEGA 11 (https://www.megasoftware.net/) with 1000 bootstrap replicates. The protein 3D structure was modeled using SWISS-MODEL (https://swissmodel.expasy.org/), and the surface electrostatic potentials were mapped with APBS-PDB2PQR (https://server.poissonboltzmann.org/). The isoelectric points (pIs) of the pro- and mature IL-18 were predicted employing ProtParam (https://web.expasy.org/protparam/). The IL-18 proteins from different species selected in this study, along with their GenBank accession numbers, were Homo sapiens, Q14116.1; Mus musculus, P70380.2; Gallus gallus, Q8QFQ8.1; Chelonia mydas, EMP35134.1; Andrias davidianus, QBG64301.1; Oncorhynchus mykiss, CAD89352.2; and C. semilaevis, XP_024910848.1. The IL-1α proteins examined comprised H. sapiens, NP_000566.3; M. musculus, NP_034684.2; Oryctolagus cuniculus, NP_001095154.1; Bos taurus, NP_776517.1; and Equus caballus, NP_001075969.2. The IL-1β proteins analyzed were H. sapiens, NP_000567.1; M. musculus, NP_032387.1; G. gallus, NP_989855.1; C. mydas, EMP30021.1; Xenopus laevis, AAI70521.1; and O. mykiss, CAB93352.1. The IL-33 homologs studied included H. sapiens, O95760.1; M. musculus, Q8BVZ5.1; B. taurus, NP_001068765; Sus scrofa, NP_001272907.1; and Canis lupus, NP_001003180.1.
Gene cloning and mutagenesis
Gene cloning was as described previously (Xu et al. 2022a, b). Briefly, total RNA was extracted with Trizol reagent (Thermo Fisher Scientific, MA, USA), and cDNA was synthesized with the ReverTra Ace qPCR RT Master Mix (TOYOBO, Osaka, Japan). The coding sequences (CDSs) of tongue sole IL-18 (GenBank accession No. XP_024910848.1), IL-18Rα (XP_016887838.1), IL-18Rβ (XP_016888205.1), and caspase-6 (XP_008315389.1) were PCR-amplified employing the PrimeSTAR Max DNA Polymerase (Takara Bio, Beijing, China). For site-directed mutagenesis of the tongue sole IL-18, the Fast Mutagenesis System (TransGen Biotech, Beijing, China) was used. Trelief 5α Chemically Competent Cells (Tsingke Biotech, Beijing, China) were transformed with the recombinant plasmid. All sequences were verified by DNA sequencing. The primers utilized for gene cloning are listed in Supplementary Table S1.
Preparation of recombinant proteins
Recombinant proteins were expressed and purified according to established protocols (Qin et al. 2023). The IL-18 and caspase-6 CDSs were cloned into the pET30a vector, and the recombinant vectors were used to transform Escherichia coli Transetta DE3 cells (TransGen Biotech, Beijing, China). They were then cultured in LB broth containing kanamycin and shaken at 220 rpm and 37 °C until the OD_600_ reached 0.4–0.6. Recombinant protein expression was induced by adding isopropyl-β-D-thiogalactopyranoside (IPTG) to a final concentration of 0.1 mmol/L and overnight incubation at 16 °C. The cells were harvested by centrifugation at 12,000 g for 30 min and lysed by sonication. The recombinant proteins were purified using nickel–nitrilotriacetic acid (Ni–NTA) columns. The expression and purification of tongue sole caspase-1/3/7 were performed as described previously (Jiang et al. 2019). The purified protein was dialyzed extensively at 4 °C overnight. Triton X-114 was added as previously reported (Wang et al. 2014) to remove the endotoxin. The proteins were concentrated using Ultrafree centrifugal filters with a 10 kDa cutoff (Millipore, MA, USA) at 4 °C. The protein concentration was assessed using the BCA Protein Assay Kit (Pierce Chemical, IL, USA).
Quantitative reverse transcription PCR (qRT-PCR)
qRT-PCR was performed as elaborated previously (Dang and Sun 2011). Briefly, to examine IL-18-induced gene expression, peripheral blood leukocytes were collected and incubated with 1 μg/mL pro- or mature IL-18 at 24 °C for 6 h. Total RNA was extracted with the TRIzol reagent (Thermo Fisher Scientific, CA, USA), and cDNA was synthesized with the ReverTra Ace qPCR RT Master Mix (TOYOBO, Osaka, Japan) as described above. The relative levels of TNF-α, IL-1β, and IFN-γ mRNA were determined by qRT-PCR utilizing a ChamQ Universal SYBR qPCR Master Mix (Vazyme Biotech, Nanjing, China) on a QuantStudio 3 (Thermo Fisher Scientific, MA, USA). The comparative threshold cycle method (2^−ΔΔCT^) was employed to analyze the relative mRNA levels. To examine bacterial infection-induced gene expression, the tongue sole fish were injected intraperitoneally (i.p.) with either 100 μL of 1 × 10^7^ CFU/mL V. harveyi or PBS (control). At 6, 12, 24, and 48 h post-infection (hpi), the relative IL-18 mRNA levels in the spleen, peripheral blood leukocytes, head kidney, and liver were measured by qRT-PCR. β-Actin was used as the internal reference gene. The primers used for qRT-PCR are listed in Supplementary Table S1.
Confocal microscopy
Cells were plated overnight on 35 mm glass-bottomed culture dishes (Nest Scientific, NJ, USA). For the subcellular localization of IL-18, pCAGGS-Flag-N (Honor Gene, Changsha, China) vectors expressing either the pro- or mature IL-18 were transfected using the PolyJet In Vitro Transfection Reagent (SignaGen Laboratories, MD, USA). After 24 h, the cells were fixed with 4% paraformaldehyde and permeabilized with 0.1% Triton X-100, each for 15 min. The cells were then blocked with 5% BSA and incubated with mouse anti-DDDDK tag antibody (Abcam, MA, USA) at a 1:1000 dilution for 1 h. After extensive washing, the cells were incubated with Alexa Fluor 594-labeled goat anti-mouse IgG antibody (Abcam, MA, USA) at a 1:1000 dilution, stained for 1 h, and re-washed extensively. To identify the subcellular location of IL-18Rα and IL-18Rβ, the pEGFP-N1 and pmCherry-N1 vectors (Takara, CA, USA) expressing IL-18Rα and IL-18Rβ, respectively, were used to transfect the cells, as described above. To investigate the interaction of IL-18 with its receptors, 20 μg/mL of recombinant IL-18 was incubated with the IL-18Rα and/or IL-18Rβ-expressing cells for 1 h and blocked with 5% BSA. These cells were then incubated with the mouse anti-His tag antibody (Abcam, MA, USA) at a 1:1000 dilution for 1 h. After extensive washing, the cells were incubated with the Alexa Fluor 647-labeled goat anti-mouse IgG antibody (Abcam, MA, USA) at a 1:1000 dilution and stained for 1 h. For cell membrane and nuclear staining, the cells were pre-fixed in 4% paraformaldehyde and permeabilized with 0.1% Triton X-100, each for 15 min. The cell membrane and nuclei were stained with FITC-labeled phalloidin (Sigma-Aldrich, MO, USA) and Hoechst 33342 (Sigma-Aldrich, MO, USA), respectively, following the manufacturer’s instructions. The images were captured with an LSM 710 confocal microscope (Carl Zeiss, Jena, Germany).
Caspase specificity and activity assay
To determine the activity and cleavage specificity of the caspases studied, recombinant caspases were incubated with various colorimetric substrates as described previously (Jiang et al. 2019), and ρ-nitroanilide (ρNA) release was monitored at OD_405_ nm. One unit of caspase proteolytic activity was defined as the amount of enzyme that produces 1 nmol of ρNA at 37 °C for 1 h.
IL-18 cleavage by caspases
Caspase cleavage specificity toward IL-18 was assayed as previously described (Xu et al. 2024). Briefly, 5 μg aliquots of recombinant Flag-tagged IL-18 were incubated with different caspases in a 50 μL reaction system containing 50 mmol/L HEPES (pH 7.5), 3 mmol/L EDTA, 150 mmol/L NaCl, 0.005% (v/v) Tween 20, and 10 mmol/L DTT. After incubation at 37 °C for 1 h or 26 °C for 6 h, the fragments obtained were separated by SDS-PAGE and immunoblotted using the anti-flag antibody (Abcam, MA, USA).
IL-18 secretion analysis
Cells were incubated in six-well plates overnight and transiently transfected with pCAGGS-Flag-N plasmids (1 μg/well) expressing the pro- or mature IL-18 as described above. After transfection for 24 h, the cells were collected by centrifugation at 600 g for 5 min and lysed on ice for 15 min. The cytosol and crude membranes were separated by centrifugation as elaborated previously (Liu et al. 2016). Proteins were separated by SDS-PAGE and transferred onto a nitrocellulose membrane (Cytiva, MA, USA). The membranes were incubated overnight at 4 °C with anti-β-actin primary antibodies, Flag, Myc, or Na^+^/K^+^-ATPase, all from Abcam. The anti-IL-18 primary antibody was used as previously described (Yu et al. 2024). The membranes were then incubated with the corresponding horseradish peroxidase-labeled secondary antibodies for 1 h. The proteins were detected using an enhanced chemiluminescent kit (Sparkjade Biotechnology, Shandong, China).
Co-immunoprecipitation
Cells were cultured in six-well plates and transfected with the plasmids described above for 24 h. The cells were then lysed with RIPA (Beyotime Biotechnology, Shanghai, China) on ice for 30 min. After centrifugation at 12,000 g at 4 °C for 30 min, the supernatant was collected and incubated with either anti-Flag or anti-Myc affinity gels (Abcam, MA, USA) at 4 °C overnight. After extensive washing, the proteins bound to the affinity gels were heat denatured and immunoblotted with either anti-Flag or anti-Myc antibodies. Cytosolic β-actin was used as an internal reference.
Detection of IL-18 secretion and maturation
V. harveyi (1 × 10^7^ CFU/mL) were injected intraperitoneally (i.p.) into the tongue sole fish. Blood was drawn at 0, 12, 24, and 48 hpi, and the serum was collected by centrifugation at 600* g* for 10 min. The secretion and maturation of IL-18 were determined by immunoblotting with anti-IL-18 polyclonal antibodies as described previously (Yu et al. 2024). The total serum proteins were separated by SDS-PAGE and stained with Coomassie Brilliant Blue R250 as a loading control.
In vivo infection
To investigate the impacts of IL-18 on fish survival post-bacterial infection, the fish were i.p. injected with mature IL-18 Δ49 or the control protein at 0.1 μg/g body weight. After 12 h, the fish were infected with V. harveyi as mentioned above and monitored daily for mortality. Each group consists of 20 individuals. The control fish were pre-injected with the recombinant Trx tag protein, and mock fish with only PBS.
Statistical analysis
A two-sample Student’s t-test was used to compare the groups. Statistical analysis was performed with the GraphPad Prism 7 software (https://www.graphpad.com/). Statistical significance was defined as P < 0.05.
Results
Structural and phylogenetic analysis of tongue sole IL-18
We identified an IL-18 homolog in half-smooth tongue sole by searching the genome database. Multiple sequence alignment showed that the tongue sole IL-18 shared an overall sequence identity of 23–32% with its homologs from the vertebrates: humans, mouse, domestic chicken, green turtles, and Chinese giant salamanders (Fig. 1A). Among mammals, the human and mouse IL-18 serve as the substrates for caspase-1, which recognizes and cleaves at a conserved N-terminal “33_LESD_36” motif. Notably, the caspase-1-targeted consensus motif “LxxD” (where, x represents any amino acid) is not conserved among birds, reptiles, amphibians, and teleost fish, suggesting a different IL-18 maturation mode in non-mammalian vertebrates (Fig. 1A). Despite its low sequence homology, the tongue sole IL-18 has a structurally similar conformation to its mammalian counterparts. Homology modeling demonstrated that the tongue sole IL-18 represents a β-trefoil structure, consisting of three twisted four-stranded anti-parallel β-sheets (Fig. 1B), which is a structural signature typical of the IL-1 family (Huising et al. 2004). The IL-18 structure modeled could be readily superimposed with its human homolog, except for the less conserved N-termini (Fig. 1B). This finding supported the notion that fish cytokines generally differ in amino acid sequence, but share a relatively well-conserved structural conformation (Wang et al. 2009; Zou and Secombes 2016). Cytokines other than IL-18, including IL-1α, IL-1β, and IL-33, are functionally related IL-1 family members (Arend et al. 2008). Phylogenetic analysis reveals that IL-18 and IL-1β are ubiquitous, from teleosts to mammals, whereas IL-1α and IL-33 are identified only in mammals. Specifically, tongue sole IL-18 was initially clustered with its teleost homologs and then formed a larger branch with other tetrapod IL-18 homologs (Fig. 1C). In contrast, IL-1α, IL-1β, and IL-33 clustered with their corresponding vertebrate homologs and formed three distinct branches parallel to that of IL-18.Fig. 1. Structural and phylogenetic analysis of tongue sole IL-18. A Tongue sole IL-18 was aligned with its vertebrate homologs. The conserved amino acid residues are highlighted in orange (100% identical) or in boxes. Secondary crystal structures of human IL-18, including 2 α-helices and 12 β-strands, are indicated above the sequences. The caspase-1 cleavage site “33_LESD_36” of human and mouse IL-18 is in boxes and indicated by arrows. The minimal active domain of IL-18 is indicated by a purple line beneath the sequences. Hs, Homo sapiens; Mm, Mus musculus; Gg, Gallus gallus; Cm, Chelonia mydas; Ad, Andrias davidianus; and Cs, Cynoglossus semilaevis. B Structural comparison of the tongue sole and human IL-18. The 3D structural model of the tongue sole IL-18 represents a β-trefoil conformation consisting of two α-helices and three twisted four-stranded β-sheets indicated by successive rainbow colors (left panel). The crystal structure of human IL-18 is shown (PDB: 3WO2, middle panel) and superimposed with the tongue sole IL-18 (right panel). C Phylogenetic analysis of IL-1α, IL-1β, IL-18, and IL-33 was conducted employing the neighbor-joining method. The tongue sole IL-18 is labeled with a red asterisk. Hs, Homo sapiens; Mm, Mus musculus; Gg, Gallus gallus; Cm, Chelonia mydas; Ad, Andrias davidianus; Om, Oncorhynchus mykiss; Cs, Cynoglossus semilaevis; Oc, Oryctolagus cuniculus; Bt, Bos taurus; Ec, Equus caballus; Xl, Xenopus laevis; Ss, Sus scrofa; and Cl, Canis lupus
Pro-IL-18 is proteolytically activated by caspase-1/3/6/7, generating three mature IL-18 forms
To investigate the IL-18 maturation process, we purified the recombinant pro-IL-18 and incubated it with various commercially available caspases. The human caspase-1/3/6/7/8 were capable of cleaving pro-IL-18 to generate the processed IL-18 fragments (Fig. 2A). To confirm such caspase-mediated pro-IL-18 cleavage, we prepared the recombinant active forms of tongue sole caspase-1/3/6/7/8 and verified their pro-IL-18 cleavage activity (Fig. 2B–E), whereas caspase-8 failed to do so (Supplementary Fig. S1). Since caspases are a family of endoproteases that specifically cleave after an Asp residue, we mutated a series of such residues to determine the pro-IL-18 cleavage sites. When incubated with caspase-1, pro-IL-18 bearing single-site mutations of either D46, D49, D51, or D55 could not prevent proteolysis (Fig. 2F). In contrast, the double-site mutations of D46/D49 and D46/D51, but not D49/D51, rendered pro-IL-18 completely resistant to caspase-1-mediated cleavage (Fig. 2G), suggesting that D46 was the primary determinant for caspase-1 recognizing pro-IL-18. Consistently, the D46/D49/D51 triple mutation prevented caspase-1-based cleavage of pro-IL-18 (Fig. 2G). Pro-IL-18 mutants bearing either D46R or D49R exhibited full resistance to proteolytic processing by caspase-3/7 (Fig. 2H, J). D46 and D49 are located within the same tetrapeptide motif “46_DNID_49,” which thus serves as the recognition motif for caspase-3/7. A single-site D55 mutation conferred the maximal resistance to caspase-6 activity (Fig. 2I). Based on these observations, we conclude that caspase-1, caspase-3/7, and caspase-6 are capable of sequentially cleaving the tongue sole pro-IL-18 at three motifs “43_YFED_46,” “46_DNID_49,” and “52_SFTD_55,” removing the respective N-terminal pro-peptides, and generating three types of fragments containing minimal active domains (Figs. 1A, 2K). Following the positions of the cleavage-associated Asp residues, the three forms of mature IL-18 were named Δ46 (for caspase-1), Δ49 (for caspase-3/7), and Δ55 (for caspase-6).Fig. 2. Pro-IL-18 is processed by caspase-1/3/6/7 to generate three forms of mature IL-18. A Recombinant Flag-tagged tongue sole pro-IL-18 was treated with 1 unit of active human caspase (CASP)-1/2/3/6/7/8/9 at 37 °C for 1 h. The fragments obtained were separated by SDS-PAGE and immunoblotted with the anti-Flag antibody. B–E Recombinant pro-IL-18 was incubated with varying units of the tongue sole CASP1/3/6/7. The fragments were separated by SDS-PAGE and analyzed as above. F–J Pro-IL-18 wild-type and single or double site mutants were incubated with tongue sole CASP1 (F, G), CASP3 (H), CASP6 (I), or CASP7 (J), and the products were assessed as described above. K Diagram of the CASP1/3/6/7-mediated cleavage of the tongue sole pro-IL-18. The mature IL-18 fragments are designated Δ46, Δ49, and Δ55 according to the cleavage points
IL-18 maturation is sufficient for membrane translocation and microvesicle-mediated unconventional secretion
As the N-terminal pro-peptide of tongue sole IL-18 is rich in negatively charged residues, i.e., Asp and Glu, we examined the effects of caspase-mediated cleavage on the electrostatic charges of the final, mature IL-18. The results showed that the pro-IL-18 surface was mainly occupied by negatively charged residues, whereas the mature IL-18 surface (e.g., Δ46), predominantly exhibited positively charged patches, indicating that the maturation process switches IL-18 from a negatively charged precursor to a positively charged functional molecule (Fig. 3A). Similarly, the predicted pI of IL-18 markedly increased post-maturation. For pro-IL-18, the predicted pI was 5.75, much lower than a neutral pH, suggesting a negatively charged IL-18 precursor under physiologically relevant conditions. The caspase-1, caspase-3/7, and caspase-6 mediated cleavages sequentially removed the N-terminal negatively charged residues, and thus increased the pI to a nearly neutral pH of 6.65 for Δ46 IL-18, a slightly higher than neutral pH of 7.13 for Δ49 IL-18, and much greater than neutral pH of 8.11 for Δ55 IL-18 (Fig. 3B). Because the cytoplasmic membrane, especially the inner leaflet, is highly enriched with anionic phospholipids, such as phosphatidic acid, phosphatidylserine, and phosphatidylinositol (Doktorova et al. 2018), we postulated that the positively charged mature IL-18 may be most likely recruited to the cytoplasmic membrane through electrostatic complementarity-based interactions. To test this hypothesis, we expressed the pro- and mature IL-18 in cells and determined if IL-18 could translocate to the cytoplasmic membrane. Immunoblotting could not detect pro-IL-18 in the membrane fraction (Fig. 3C). In contrast, the mature IL-18, especially Δ55, was highly abundant in the membrane fraction, indicating a much greater membrane binding stoichiometry. Δ46 was barely detectable in the membrane fraction, probably due to its comparatively low pI. Confocal microscopy confirmed that pro-IL-18 was mainly distributed across the cytosol, while the mature IL-18 could be transported to the cytoplasmic membrane (Fig. 3D), suggesting an essential role of electrostatic complementarity-based membrane association in the subcellular localization of IL-18. Additionally, we noticed that the cells expressing mature IL-18 formed cytoplasmic membrane extrusions and microvesicles, which enclosed IL-18 and were shed to the exterior (Fig. 3D lower panel, E). This observation indicated that such microvesicles could serve as transport vehicles for the unconventional secretion of IL-18 to the extracellular environment. Accordingly, when expressed in cells, pro-IL-18 was mainly retained in the cytosolic fraction and hardly secreted into the cell culture medium, whereas the mature IL-18 with a greater pI (Δ55) was secreted more efficiently than IL-18 with lower pIs (Δ46 and Δ49), into the extracellular environment (Fig. 3F, G).Fig. 3. Mature IL-18 translocates to the cytoplasmic membrane and is secreted via microvesicles. A Analysis of the electrostatic surface potentials of tongue sole pro-IL-18 (left) and mature Δ46 IL-18 (right). The positively and negatively charged surfaces are colored blue and red, respectively, while the electro-neutral or nonpolar/hydrophobic regions are colored white. B The isoelectric points (pIs) of the tongue sole pro- and mature IL-18 (Δ46, Δ49, and Δ55) were predicted. C HEK 293 T cells were transfected with vectors expressing the Flag-tagged tongue sole pro-IL-18 or mature IL-18 (Δ46, Δ49, and Δ55). After 24 h, the cytoplasmic membrane was separated and immunoblotted using the anti-Flag antibody. β-Actin and Na^+^/K^+^-ATPase proteins were used as internal references for the cytoplasm and cytoplasmic membrane, respectively. D Cells expressing Flag-tagged pro-IL-18 or mature IL-18 were subjected to immunofluorescence using the anti-Flag and Alexa Fluor 594-labeled goat anti-mouse antibodies. The cell membrane and nucleus were stained with FITC-labeled phalloidin and Hoechst 33342, respectively. Membrane or microvesicle localization of pro- or mature IL-18 was observed by confocal microscopy and shown in the enlarged images (lower panel). Scale bars, 5 μm. E Cells were transfected with vectors expressing the mCherry-tagged pro- or mature IL-18 (Δ46, Δ49, and Δ55). The microvesicles extruding from the cytoplasmic membrane were observed by confocal microscopy and are indicated with blue arrows. Scale bars, 5 μm. F, G Cells were transfected with Flag-tagged pro- or mature IL-18 as described above. The culture medium and cells were collected and immunoblotted with the anti-Flag antibody to detect IL-18 secretion. β-Actin was used as the internal reference for the cytoplasm
IL-18Rα specifically recognizes mature IL-18 and recruits IL-18Rβ
Genomic analysis suggested the existence of two IL-18 receptor subunits, IL-18Rα and IL-18Rβ, in tongue sole fish. In contrast to IL-18, which had a comparatively lower homology with its vertebrate homologs, IL-18Rα and IL-18Rβ shared a conserved domain architecture with their counterparts in humans (Supplementary Fig. S2). When expressed in cells, IL-18Rα and IL-18Rβ were found to be located within the cell membrane (Fig. 4A), which is required for binding with the extracellular IL-18 ligand. The co-expression of IL-18Rα and IL-18Rβ facilitated their co-localization, suggesting a close functional relationship between these two receptors. To examine the interaction between IL-18 and its receptors, we co-expressed IL-18 with IL-18Rα and/or IL-18Rβ in the same cell system. Confocal microscopy revealed that the cell membrane binding of pro-IL-18 was defective in IL-18Rα-expressing cells, whereas the mature IL-18, i.e., Δ46, Δ49, and Δ55, were capable of cell membrane binding and exhibited a strong co-localization with IL-18Rα (Fig. 4B). A co-immunoprecipitation assay confirmed that the mature IL-18 forms could be recognized by IL-18Rα (Fig. 4C). Compared with Δ46, the Δ49 and Δ55 IL-18 exhibited an even enhanced binding stoichiometry to IL-18Rα, suggesting the possible involvement of electrostatic charge during the IL-18 ligand–receptor interaction. In contrast, pro-IL-18 was unable to bind to IL-18Rα, probably due to N-terminal stereo-hindrance and a negative electrostatic potential. We next examined if IL-18 could bind directly to IL-18Rβ. Confocal microscopy revealed that neither pro-IL-18 nor the mature IL-18 s were able to bind cells expressing IL-18Rβ, suggesting an inability of IL-18Rβ to interact with IL-18 (Fig. 4D). Consistent with this observation, co-immunoprecipitation verified that the pro- and mature forms of IL-18 could not bind to IL-18Rβ (Fig. 4E). When IL-18Rα and IL-18Rβ are co-expressed on the cell membrane, the mature IL-18 (Δ46, Δ49, and Δ55) could bind and co-localize with both the receptors, indicating the formation of IL-18/IL-18Rα/IL-18Rβ heterotrimeric complexes (Fig. 4F). Compared with the robust binding ability of the mature IL-18, pro-IL-18 exerted meager binding to IL-18Rα/IL-18Rβ, supporting the notion that the processing of the precursor into its active form is necessary for IL-18-based signaling.Fig. 4IL-18Rα specifically recognizes mature IL-18 and recruits IL-18Rβ. A HEK 293 T cells were transfected with vectors expressing EGFP-tagged IL-18Rα and/or mCherry-tagged IL-18Rβ. The subcellular location of IL-18Rα and IL-18Rβ was observed by confocal microscopy. Cells transfected with empty vectors were used as controls. Scale bars, 5 μm. B, D, and F Cells were transfected with vectors expressing EGFP-tagged IL-18Rα (B), mCherry-tagged IL-18Rβ (D), or EGFP-tagged IL-18Rα plus mCherry-tagged IL-18Rβ (F), and incubated with recombinant pro- or mature (Δ46, Δ49, and Δ55) IL-18. Alexa Fluor 647-labeled antibody was used to indicate the binding with IL-18. The nucleus was stained with Hoechst 33342. Scale bars, 5 μm. C, E Cells were transfected with vectors expressing Flag-tagged pro- or mature (Δ46, Δ49, and Δ55) IL-18 and Myc-tagged IL-18Rα (C) or IL-18Rβ (E). The cytoplasm was prepared for immunoprecipitation with anti-Myc affinity gels and immunoblotted with anti-Flag and anti-Myc antibodies. Cell lysates were used and immunoblotted as described above. β-Actin was used as an internal reference for the cytoplasm
IL-18 activates the antibacterial immune response during infections
As a typical proinflammatory cytokine, IL-18 is highly expressed in the immune system-related organs, e.g., peripheral blood leukocytes, spleen, head kidney, and gill, compared to the liver and intestines (Fig. 5A). To explore the immunological function of IL-18, we incubated peripheral blood leukocytes with endotoxin-free IL-18 and examined the gene expression patterns. The results revealed that pro-IL-18 is incapable of upregulating the expression of IFN-γ, IL-1β, and TNF-α, unlike all three forms of mature IL-18 (Δ46, Δ49, and Δ55), which significantly increased the expression (Fig. 5B–D). Since IL-18 was initially identified as an IFN-γ-inducing factor due to its ability to enhance IFN-γ expression in human immune cells (Golab 2000), similar upregulation pattern in tongue sole suggests an evolutionarily conserved role of fish IL-18 in regulating gene expression. V. harveyi is a major bacterial pathogen that causes severe diseases and even massive mortality in a broad spectrum of marine fishes (Austin and Zhang 2006). When infected with* V*. harveyi, the tongue sole expressed significantly higher levels of IL-18 in the spleens (6 and 12 h), peripheral blood leukocytes (6 h), head kidney (24 and 48 h), and liver (48 h) than control fish (Fig. 5E–H). To determine if infection induces IL-18 maturation and secretion, we collected the serum of the infected fish and detected the IL-18 levels by immunoblotting. The results revealed that V. harveyi infection remarkedly increased the production and secretion of IL-18, and a majority of the IL-18 proteins were processed into a mature form in a time-dependent manner (Fig. 5I). During infection, caspase-3 was markedly upregulated in peripheral blood leukocytes (Supplementary Fig. S3), suggesting its vital role in IL-18 maturation. We further examined the protective role of IL-18 in infected fish. Tongue sole fish injected with control proteins exhibited severe clinical signs, including muscle necrosis, skin ulcers, and tail rot (Fig. 5J), and ~ 70% of them died within 5 days (Fig. 5K). In contrast, injection with mature IL-18 reinforced the immunity against V. harveyi infection, leading to milder disease and a much higher survival rate than the control group.Fig. 5IL-18 induces proinflammatory cytokine expression and promotes antimicrobial immunity. A The expression patterns of the tongue sole IL-18 in different tissues were determined by qRT‐PCR. The mean expression levels of IL-18 in fish intestine samples were normalized as 1. n = 3. B, C, and D The tongue sole peripheral blood leukocytes were incubated with or without pro- and mature IL-18 (Δ46, Δ49, and Δ55). The relative expression levels of IFN-γ (B), IL-1β (C), and TNF-α (D) were determined by qRT‐PCR. Values are the means ± SD of three replicates. E–H Tongue sole fish were infected with V. harveyi, and the expression levels of IL-18 in the spleen (E), peripheral blood leukocytes (F), head kidney (G), and liver (H) at different time points were determined by qRT-PCR. The control group of fish were injected with PBS. In each case, the mean expression level of IL-18 in the control fish samples was normalized as 1. n = 3. I Tongue sole were infected with V. harveyi, and the serum was collected at different hours post-infection. The secreted IL-18 was detected by immunoblotting with anti-IL-18 antibodies (upper panel). The total serum protein content was assessed by SDS-PAGE and Coomassie Brilliant Blue R250 staining (lower panel). J, K Tongue soles were injected with the control protein thioredoxin (Trx) or mature IL-18 Δ49, and infected with V. harveyi as described above. J The clinical syndromes of mock (upper), Trx- (middle), and IL-18- (lower) treated fish were photographed. The infection-induced tissue damage is indicated by blue arrows. K The infected fish were monitored daily, and the survival curve was analyzed by employing a log-rank test. The mock fish were injected with PBS. n = 20.Values in B–H are the means ± SD. *P < 0.05, **P < 0.01
Discussion
Processing to remove the N-terminal propeptide is essential for the maturation of the IL-1 family members, particularly IL-18 and IL-1β (Sims and Smith 2010). In humans, IL-18 is typically activated by the proinflammatory caspase-1/4, which recognize the same N-terminal tetrapeptide motif “33_LESD_36,” generating a mature IL-18 polypeptide that contains the C-terminal minimally active IL-18 domain (Devant et al. 2023; Shi et al. 2023). In the teleost fish half-smooth tongue sole, the conserved “LxxD” motif of IL-18 is lacking, suggesting the possible existence of a cleavage site alternative to that of human IL-18. Indeed, we found that caspase-1 could cleave tongue sole IL-18, but at a nonconserved N-terminal motif, “43_YFED_46,” indicating an evolutionarily crucial role of caspase-1 in IL-18 maturation. Notably, although caspase-1 exhibits high cleavage specificity, tongue sole caspase-1 could cleave at alternative sites once this motif was mutated, suggesting the flexibility of teleost caspase-1 in substrate recognition and cleavage. In contrast to caspase-1-mediated IL-18 maturation, some other proteases, e.g., apoptotic caspase-3, could inactivate human IL-18 by cleaving at the “73_DCRD_76 motif,” which generated a truncated IL-18 functional domain (Akita et al. 1997). This finding supported the notion that caspase-3 is capable of proteolytically degrading a large number of substrates. Similarly, IL-33 can be cleaved by caspase-3 and -7 at the C-terminal “175_DGVD_178,” to generate a biologically inactive IL-33 (Martin et al. 2012). Caspase-3- and/or caspase-7-mediated IL-18 and IL-33 inactivation represents a negative regulation of the proinflammatory response during apoptosis. In contrast, we observed that both caspase-3 and caspase-7 could activate tongue sole IL-18 by cleaving at the N-terminal “46_DNID_49 motif,” a typical caspase-3/7 consensus motif located near the caspase-1 cleavage site. Moreover, another apoptotic protease, caspase-6, also participated in the maturation of tongue sole IL-18. This cleavage occurs at a site downstream of the caspase-3/7 recognition site and generates a bioactive IL-18 polypeptide. These findings collectively suggest the essential roles of both proinflammatory and apoptotic caspases in regulating IL-18 maturation in teleosts.
Due to the lack of a signal peptide, IL-18 is presumed to exit cells by an unconventional secretion pathway similar to that of IL-1β (Lopez-Castejon and Brough 2011). In humans, multiple molecular pathways, including exosome release and endolysosome exocytosis, have been proposed to explore the route for IL-1β and possibly IL-18 exudation (Eder 2009). Exosomes are endosome-associated membrane vesicles that pack IL-1β and a few other cytosolic contents, enabling the transport of intracellular components to the extracellular environment (Qu et al. 2007). In contrast, another study could not detect any mature IL-1β in the exosomes of human macrophages (Öhman et al. 2014), probably due to a varied cellular context and diversified unconventional secretion scenarios. Similar to the exosome-mediated secretion pathway, late endolysosomes could wrap IL-1β and exit cells after plasma membrane shedding (Andrei et al. 2004). Whether exosomes and late endolysosomes could serve as the secretory vehicles for IL-18 remains elusive. This study found that the cleavage of tongue sole IL-18 specifically removed the N-terminal negatively charged peptide and elevated the surface electrostatic potential from a negatively charged precursor to positively charged mature IL-18 isoforms. Because of electrostatic repulsion by the cytoplasmic membrane, pro-IL-18 is prone to be retained within the cytosol, while a substantially large amount of positively charged, mature IL-18, especially Δ55, was recruited to the membrane. Although the possibility of IL-18 directly translocating across the plasma membrane cannot be ruled out, the highly enriched mature IL-18 shed within membrane-enclosed microvesicles supports the notion that these could specifically facilitate mature IL-18 secretion. In addition to secretion from live cells, cytokines and other damage-associated molecular patterns could also be released passively from the membrane-perforated or lysed cells. For example, in humans, the inflammasome activates caspase-1, which affects IL-18 maturation and GSDMD-dependent pyroptosis, leading to the release of mature IL-18 and massive amounts of intracellular contents from membrane-ruptured pyroptotic cells (Baik et al. 2023; Exconde et al. 2023). A recent study reported that IL-1α could also be released into the extracellular environment in a pyroptosis-dependent manner (Tsuchiya et al. 2021). In contrast to humans, teleosts lack GSDMD, but possess a pyroptosis-inducing Gasdermin E (GSDME), which serves as a rapid secretory mechanism for intracellular contents (Chen et al. 2021; Li et al. 2020).
Receptor-mediated signaling is critical for IL-18-regulated immune response (Gerdes et al. 2002). In humans, IL-18Rα, but not IL-18Rβ, initially recognizes mature IL-18 (Landy et al. 2024). The formation of an IL-18/IL-18Rα heterodimer is insufficient for triggering signal transduction, unless the IL-18Rβ also binds to form a functional heterotrimer (Harel et al. 2022). Unlike the tongue sole IL-18 ligand, which has low sequence analogy with its human homologs, IL-18Rα and IL-18Rβ exhibit a conserved architecture and a comparatively high similarity, suggesting an evolutionarily conserved mode of IL-18 receptor recognition. Indeed, we found that the tongue sole IL-18Rα could directly bind to the mature IL-18, with the capacity of IL-18 (Δ49 and Δ55) being much greater than that of IL-18 (Δ46), suggesting a role of electrostatic charges during IL-18 and IL-18Rα binding. Pro-IL-18 is completely ineffective in IL-18Rα binding, probably because of steric hindrance and a negatively charged N-terminus. The formation of an IL-18/IL-18Rα heterodimer complex subsequently enables IL-18Rβ recruitment like that in humans. IL-18 was initially named the IFN-γ-inducing factor for its ability to promote IFN-γ expression in humans (Arimitsu et al. 2006). Similarly, IL-18 stimulation in tongue sole fish enhanced IFN-γ expression, indicating a conserved role of teleost IL-18 during immune activation. Owing to a conserved intracellular domain, teleost IL-18 receptors might activate a downstream signaling pathway similar to that in humans. In rainbow trout, IL-18Rβ could bind to a series of signaling elements, including MyD88, IRAK4, IRAK1, TRAF6, and TAB2, thereby activating NF-κB-based signal transduction (Yang et al. 2023). The potent immune activation of tongue sole IL-18 confers robust protection against bacterial infections, emphasizing the evolutionarily vital and conserved role of IL-18 in immune response regulation.
In conclusion, the findings presented reveal the unique molecular mechanisms underlying IL-18 maturation, secretion, and receptor-mediated immune regulation in the teleost fish, half-smooth tongue sole (Fig. 6). Although the teleost IL-18 lacked a conserved caspase-1 recognizing consensus motif found in human IL-18, the tongue sole IL-18 harbored an alternate cleavage site specific to caspase-1. Furthermore, pro-IL-18 could be proteolytically activated by apoptosis-related caspases, i.e., caspase-3, caspase-6, and caspase-7, which were reported to proteolytically inactivate IL-18 and a few other cytokines in mammals. Although membrane lysis can provide a rapid secretion route for intracellular contents, our study reveals that maturation increased the positive electrostatic charge of the IL-18 surface, thus facilitating membrane-based translocation and microvesicle-mediated secretion without cell lysis. Similar to the IL-18R-mediated ligand binding mode in humans, the IL-18 in tongue sole fish is initially recognized by IL-18Rα, forming a binary complex that further associates with IL-18Rβ to form a functional heterotrimeric complex. Bacterial infection promotes IL-18 maturation and secretion in tongue sole fish. The activation of the IL-18-based signaling pathway induces proinflammatory cytokine expression and confers protection against bacterial infections.Fig. 6. Schematic representation of IL-18 maturation, secretion, and receptor-mediated immune response in the teleost fish, half-smooth tongue sole. Tongue sole pro-IL-18 is processed sequentially by caspase (CASP)1, CASP3/7, and CASP6 at three different N-terminal motifs, generating three C-terminal forms of mature IL-18. The cleavages removed the negatively charged N-termini and switched the electrostatic potential from a negatively charged precursor to positively charged mature IL-18 isoforms. Due to the negative charge of the cytoplasmic membrane, pro-IL-18 is repelled and retained mainly in the cytosol. In contrast, the cytoplasmic membrane is highly enriched with the mature IL-18 (e.g., Δ55) and secreted via membrane-associated microvesicles. IL-18Rα specifically recognizes IL-18 to form a heterodimer, which subsequently recruits IL-18Rβ to form a functional heterotrimeric complex. The ligand–receptor interaction induces conformation change in the IL-18Rα/IL-18Rβ intracellular domain, which activates downstream signaling, probably through the NF-κB cascade, and promotes proinflammatory cytokine expression. Bacterial infection induces the maturation and secretion of IL-18, while the activation of IL-18-based signaling reinforces immunity against bacterial infections
Supplementary Information
Below is the link to the electronic supplementary material.Supplementary file1 (DOCX 128 KB)