IL-1 Receptor Antagonist Chimeric Protein: Context-Specific and Inflammation-Restricted Activation

This article is featured in In This Issue, p.1337

Interleukin-1α and IL-1β are pluripotent proinflammatory cytokines that promote acute and chronic inflammation (1). The IL-1R antagonist (IL-1Ra) is a naturally occurring member of the IL-1 family, and concentrations of endogenous IL-1Ra increase during inflammatory conditions in an attempt to block IL-1 activities (2). IL-1Ra is a pure receptor antagonist for the IL-1 receptors. For more than 15 y, a recombinant form of IL-1Ra (anakinra) has been used to treat a broad spectrum of diseases, particularly autoinflammatory diseases (3). For example, anakinra is a validated therapeutic approach for blocking IL-1 in gouty arthritis (4), postinfarction cardiac failure (5), pancreatic β cell injury in type 1 diabetes (6), insulin resistance in type 2 diabetes (7), smoldering myeloma (8), and facilitation of tumor growth, angiogenesis, and metastases (9). Any blockade of IL-1 activities comes with the risk of reduced host defenses against infections.

We developed a novel therapy for reducing IL-1–mediated local inflammation with IL-1Ra while limiting widespread IL-1R1 blockade and the risk for decreasing host defenses. For example, we sought to deliver IL-1Ra in such a manner that it would block IL-1 in inflamed joints (e.g., gouty arthritis) without reducing the ability of IL-1 to facilitate entry of neutrophils into the lung upon encountering bacterial pneumonia. Therefore, we designed a form of an inactive IL-1Ra that is not constitutively able to bind to the IL-1R1 until it is converted to active IL-1Ra, primarily at sites of inflammation.

We took advantage of the precursor form of IL-1β, which is constitutively inactive but is converted to active IL-1β extracellularly by serine proteases. Unless cleaved, the N-terminal 116 aa (termed N-terminal peptide; NTP) of the IL-1β precursor protein prevent binding to the IL-1R1. We therefore fused the IL-1β NTP to the active form of IL-1Ra to prevent binding to IL-1R1, essentially creating an inactive IL-1Ra. Thus, we hereby describe a chimeric IL-1Ra and its conversion to active IL-1Ra by proteases, which are abundant in inflammatory cell infiltrates.

IL-1β NTP and IL-1Ra were amplified by conventional PCR from human cDNA using primers 5′-CACGGATCCGCCACCATGGCAGAAGTACCTGAGCTC-3′ with 5′-′CACAAGCTTTCGTACAGGTGCATCGTGCACATAAG-3′ and 5′-CACAAGCTTCGACCCTCTGGGAGAAAATCCAG-3′ with 5′-TCACAGTGCGGCCGCCTACTCGTCCTCCTGGAAGTAG-3′. The two PCR products were digested with HindIII (New England Biolabs, Beverly, MA) to create adhesive linker and were ligated using T4 ligase (New England Biolabs). For IL-1Ra alone, the same reverse primer with a forward primer 5′-CAGGATCCGCCACCATGCGACCCTCTGGGAGAAAATCCAG-3′ was used in PCR. Ligated chimeric IL-1Ra or IL-1Ra alone PCR products were digested and cloned into BamHI and NotI (both from New England Biolabs) sites of pCDNA3.1+ plasmid (Invitrogen, Carlsbad, CA). To create deletion mutations of IL-1β NTP in the chimeric IL-1Ra, 5′-phosphorylated primers 5′-CGACCCTCTGGGAGAAAATCCA-3′ and 5′-TTCTTCTTCAAAGATGAAGGGAAAGAAGGT-3′ flanking codons encoding residues 102–120; for residues 1–70, 5′-phosphorylated primers 5′-ATGGACAAGCTGAGGAAGATGCTG-3′ and 5′-GCAGATCGTCTCTGAATGGAACAGG-3′ were used. The template used was 100 pg pCDNA3.1+ chimeric IL-1Ra plasmid, amplified by Phusion DNA polymerase (New England Biolabs) in PCR program: initial denaturation at 98°C for 30 s then 25 cycles of 98°C for 10 s; 65°C for 10 s; 72°C for 3 min and 30 s; and a final extension of 72°C for 5 min. Sequences were verified by sequencing analysis from both sides of the inserts using T7 promoter and BGH-Reverse primers. Cloning into pET30a (Novagen, Madison, WI) of chimeric IL-1Ra and Δ102–120 mutated chimeric IL-1Ra was performed with primers 5′-CACGGATCCGCCACCATGGCAGAAGTACCTGAGCTC-3′ and 5′-TCACAGTGCGGCCGCCTACTCGTCCTCCTGGAAGTAG-3′, IL-1Ra alone with primers 5′-CAGGATCCGCCACCATGCGACCCTCTGGGAGAAAATCCAG-3′ and 5′-TCACAGTGCGGCCGCCTACTCGTCCTCCTGGAAGTAG-3′, whereas Δ1–70 mutated chimeric IL-1Ra coding sequence was amplified using the same reverse primer with forward primer 5′-CACGGATCCATGGACAAGCTGAGGAAGATGCTG-3′. Using BamHI and NotI sites, the products were cloned into pET30a in frame with the N-terminal His tag. The cloning was verified by sequencing analysis using T7 promoter and T7 termination primers.

BL21(DE3) bacterial cells were transformed with pET30a cloned vectors and were grown in Luria Broth supplemented with kanamycin (30 μg/ml). Bacterial cells were grown to logarithmic phase, induced with 0.1 mM IPTG for 2 h, and then centrifuged and resuspended in 10 mM imidazole followed by sonication (four times for 20 s at 40% amplitude). Recombinant proteins were isolated using affinity chromatography with Ni-NTA agarose beads according to the manufacturer’s protocol (Thermo Scientific, Marietta, OH). Imidazole eluted fractions (250 mM) were desalted using 10 kDa Amicon Ultra centrifugal filters (Millipore, Billerica, MA). To cut the His tag and linker originated from pET30a vector from the N-terminal of the chimeric IL-1Ra, chimeric IL-1Ra^Δ102–120, chimeric IL-1Ra^Δ1–70 or IL-1Ra alone, recombinant enterokinase (Novagen) digestion was performed (20°C for 20 h), followed by enterokinase removal capture beads purification step (Novagen).

HEK-293T cells were cultured in 10-cm Petri culture dishes, and the following day were transfected with 8 μg of pCDNA3.1+ empty vector, IL-1Ra, or chimeric IL-1Ra encoding vector with JetPEI transfection reagent (Polyplus Transfection, Illkirch, France). After 24 h, cells were washed twice in PBS, collected, and centrifuged. Lysates were prepared using 0.5% triton-X100 in PBS. Lysates were then dialyzed against RPMI 1640 medium using a 3.5 kDa molecular weight cutoff–defined membrane (Spectrum Laboratories, New Brunswick, NJ). To obtain mouse neutrophils, C3H mice were injected with 2 ml of thioglycollate (Hylabs, Rehovot, Israel) and 20 h later, the peritoneal exudate cells (PECs) were obtained, cells were washed in PBS, and lysates were made by two cycles of freeze-thawing of the pellets. Primary macrophages were obtained from the peritoneum of mice 4 d after the i.p. injection of 2.5 ml thioglycollate, washed in PBS, and further cultured in 10% FCS-supplemented RPMI 1640. RAW 264.7 or primary macrophages were stimulated with 1 μg/ml LPS (Sigma-Aldrich, Rehovot, Israel) for 20 h, with or without 5 mM ATP for 30 min.

Primary NK (pNK) cells from healthy donors were purified using Human NK cell Enrichment kit (no. 15065, RosetteSep; Stemcell Technologies, Vancouver, Canada) according to the manufacturer’s recommendations and cultured for 1 wk in the presence of 300 U/ml IL-2; following, 1 × 10⁶ pNK cells were incubated overnight in the presence of different concentrations of PMA and 2 h of A23187 (4 μM). Supernatant was then collected and was incubated with chimeric IL-1Ra or chimeric IL-1Ra^Δ102–120 for 4 h at 37°C. The pNK cells were stained for CD3-PE (IQP-519R; IQ Products), CD56-Allophycocyanin (no. 318310; BioLegend, San Diego, CA), CD107a-BV421 (no. 328626; BioLegend) for 30 min on ice and analyzed with BD FACSCanto II.

Lysates of transfected HEK-293T cells or recombinant chimeric IL-1Ra, chimeric IL-1Ra^Δ102–120, chimeric IL-1Ra^Δ1–70, and recombinant IL-1Ra proteins were analyzed for protein concentration by Bradford reagent (Bio-Rad, Hercules, CA) and were quantified using IL-1Ra ELISA (R&D Systems, Minneapolis, MN). Cell lysates or recombinant proteins were boiled in 95°C dry bath together with Laemmli sample buffer and were separated over PAGE, stained with Coomassie brilliant blue, or transferred to PVDF membrane followed by Western blot using goat anti–IL-1Ra Ab (AF-280-NA, R&D Systems). Mouse primary macrophage supernatants were assayed for IL-6 by ELISA kit (PeproTech, Rocky Hill, NJ) and for LDH release (Promega, Madison, WI).

A549 cells were cloned by limiting dilutions to select a sensitive clone to test the activity of human IL-1–induced IL-6. The most sensitive clone was selected and further cultured in 96-well plates (4 × 10⁴/well). On the following day, recombinant IL-1Ra, chimeric IL-1Ra, anakinra (Amgen, Thousand Oaks, CA), or transfected HEK-293T cell lysates either untreated or following enzymatic digestion with elastase were added to the cell culture media for 1 h, followed by 6 h of stimulation with 100 pg/ml recombinant human IL-1β. Culture media of stimulated A549 cells were collected and analyzed for the levels of IL-6 by ELISA Kit (R&D Systems).

Recombinant protein or HEK-293T cell lysates were digested with 40 μg/ml elastase (Athens Research & Technology, Athens, GA) at 37°C for 10 min and then added to A549 cells for testing IL-1R1 inhibition. Chimeric IL-1Ra was incubated (1 h, 37°C) with 7 μg of lysates obtained from 20-h thioglycollate-mouse PECs, pNK supernatants (4 h, 37°C), or macrophages supernatants (16 h, 37°C). To inhibit the elastase, 0.5 mg/ml α-1 antitrypsin (AAT; Kamada, Ness Ziona, Israel) was added prior to elastase to the reaction tubes. For elastase/chimeric IL-1Ra concentration response assays, serial dilutions of elastase (1:5) or chimeric IL-1Ra (1:3) were performed and then mixed and incubated for 15 min in 37°C. Later, each enzymatic reaction was added to A549 cells and measurement of IL-1β–induced IL-6 was performed. RAW264.7 cell supernatants (50 μl) or lysates (5 μg) were incubated with 500 ng of chimeric IL-1Ra, and were analyzed by Western blot.

Matrigel (BD Biosciences, San Jose, CA) was thawed and mixed with 1 μg LPS with or without 1 μg recombinant chimeric IL-1Ra or anakinra, and 500 μl was injected s.c. into C57BL/6 mice (Harlan Laboratories, Jerusalem, Israel). After 24 or 48 h, Matrigel plugs were removed and dissolved using Collagenase IV (Sigma) for 90 min at 37°C with a magnetic stirrer. The cells were filtered using a 70-μM cell strainer (BD Biosciences), washed twice in PBS, and then counted with a Countess automated cell counter (Invitrogen).

Chimeric IL-1Ra in vivo Matrigel results are from three different experiments. Significance of results of the in vivo experiments was determined using two-tailed Mann–Whitney U test. Significance of results of in vitro experiments was analyzed using Student two-tailed unpaired t test; all statistics were calculated using GraphPad Prism 6.0 software (La Jolla, CA).

Two encoding sequences were cloned in-frame to form the chimera: the N-terminus of the chimera coding for the IL-1β NTP and the C-terminal to encode IL-1Ra (Fig. 1A). The NTP was composed of residues 1 to 120 of the IL-1β precursor. This domain contains several cleavage sites for proteases (Fig. 1A). Cleavage of the NTP generates a mature, active IL-1Ra. Chimeric IL-1Ra sequence, as well as the sequence for active IL-1Ra, were cloned into a pET30a vector downstream to the His tag and linker sequences. The proteins were expressed in BL21(DE3) bacterial cells and isolated using Ni-NTA affinity chromatography (Fig. 1B).

To determine the activity of recombinant IL-1Ra and chimeric IL-1Ra, A549 cells were incubated with increasing concentrations of these proteins (measured by ELISA), as well as with clinical grade IL-1Ra (anakinra), for 1 h before the addition of IL-1β. Six hours later, IL-6 levels were determined in cell supernatants. As shown in Fig. 1C and Supplemental Fig. 1A and 1B, upon removal of the His tag linker, IL-1Ra exhibited comparable size as seen in Western blot as well as comparable activity to anakinra. In contrast, uncut chimeric IL-1Ra lacked IL-1 blocking activity (Fig. 1C, Supplemental Fig. 1B).

A nonbacterial expression system was also evaluated. Chimeric IL-1Ra and IL-1Ra were cloned into pCDNA3.1+ vector and transfected into HEK-293T cells. As shown, the expected products were found in cell lysates, as determined by IL-1Ra ELISA and by Western blot analysis using a specific anti–IL-1Ra Ab (Fig. 1D).

To confirm that the lysates also contain the expected IL-1 blocking activities, equal amounts of either mock-transfected, IL-1Ra transfected, or chimeric IL-1Ra transfected cell lysates were added to A549 cells. After 1 h, the cells were stimulated with IL-1β; 6 h later, IL-6 levels were determined in the supernatants. As shown in Fig. 1E, HEK-293T–derived IL-1Ra inhibited IL-1 activity in A549 cells, whereas the chimeric IL-1Ra was comparable to mock-transfected cell lysates in that it lacked IL-1 blocking activity.

Chimeric IL-1Ra was incubated with neutrophil elastase, and the resulting products were analyzed using Western blot. As shown in Fig. 2A, chimeric IL-1Ra was cleaved to form a fragment measuring ∼20 kDa, comparable in size to IL-1Ra. Moreover, the cleavage was inhibited by the highly effective inhibitor of elastase, AAT.

To examine whether the cleaved product contains IL-1 receptor blocking activity, A549 cells were incubated with the products of elastase treatment. After 1 h, the cells were stimulated with IL-1β, and after 6 h, the supernatants were assessed for IL-6 levels. As shown in Fig. 2B, elastase-treated chimeric IL-1Ra exhibited the ability to block IL-1–induced IL-6 production. Similar data were obtained with HEK-293T–derived chimeric IL-1Ra (Fig. 2C, 2D). Furthermore, the antagonizing capacity of chimeric IL-1Ra positively correlated with the levels of added serine protease elastase (Fig. 2E), exhibiting increased capacity to inhibit IL-1–induced IL-6 as the concentrations of chimeric IL-1Ra and of elastase were increased.

To generate a cell lysate preparation that represents enzymatically rich local inflammatory sites, peritonitis was evoked in mice using thioglycollate; PECs were obtained 20 h later. The composition of PECs was predominantly CD11b⁺/Ly6G⁺ neutrophils (Supplemental Fig. 2A). Chimeric IL-1Ra was then incubated with PEC lysates, and cleaved products were examined. As shown in Fig. 2F, PEC lysates cleaved chimeric IL-1Ra to form a 25-kDa product. In contrast, in the presence of AAT, chimeric IL-1Ra remained at full length. Whereas most of the experiments using proteolytic conditions resulted in a product similar in size to IL-1Ra (Fig. 2A), a 25-kDa product was also detected by Western blot (Supplemental Fig. 2B). Despite its different size, the 25-kDa cleavage product inhibited IL-1β–induced IL-6 in the A549 cell assay (Supplemental Fig. 2C).

We next examined whether products of stimulated macrophages can also convert chimeric IL-1Ra into an active antagonist. Mouse RAW264.7 cells were stimulated with either LPS or LPS plus ATP; 30 min later, activation of caspase-1 was depicted by Western blot (Fig. 3A). Supernatants were collected and added directly to chimeric IL-1Ra for 16 h. As depicted in Fig. 3B (left), chimeric IL-1Ra was cleaved by activated RAW264.7 cell products.

To confirm the requirement of protease cleavage sites, a deletion-mutation was generated, which lacks the protease target sites (chimeric IL-1Ra^Δ102–120; Fig. 3C). Indeed, as shown in Fig. 3B (right), chimeric IL-1Ra^Δ102–120 displayed decreased levels of IL-1Ra release by exposure to supernatants from activated RAW264.7. However, lysates prepared from RAW264.7 cells did cleave chimeric IL-1Ra^Δ102–120 in a time-dependent manner (Fig. 3D). In addition, LPS activation was sufficient to elicit a protease-rich lysate that cleaved chimeric IL-1Ra, but a supernatant that lacks the ability to cleave chimeric IL-1Ra^Δ102–120.

Following LPS/ATP activation, peritoneal macrophages also cleaved chimeric IL-1Ra (Fig. 3E). However, in the absence of added ATP, LPS was sufficient to induce IL-6 in a similar manner (Supplemental Fig. 3A). Using extracellular LDH as a measure of cell death, we consider the activation of chimeric IL-1Ra to be independent of cell necrosis (Supplemental Fig. 3B).

We were able to detect in Western blots of transfected HEK-293T an additional band ∼9 kDa smaller in molecular size than the chimeric IL-1Ra (Fig. 4A). The NTP is prone to cleavage because the precursor of IL-1β has a trypsin cleavage site around position 76–77, which results in a shorter precursor protein with m.w. of ∼22 kDa (10, 11). Therefore, to examine whether a shorter intermediate product can still serve as a chimeric IL-1Ra, we designed a truncated version of chimeric IL-1Ra by deleting the first 70 aa within the NTP (chimeric IL-1Ra^Δ1–70). The sequence was cloned and expressed in BL21(DE3) bacterial cells (Fig. 4B). Chimeric IL-1Ra^Δ1–70 was digested from the His-tag and linker (Supplemental Fig. 1C), and its activity was determined using A549 cells (Supplemental Fig. 1C, Fig. 4C). Similar to chimeric IL-1Ra, the chimeric IL-1Ra^Δ1–70 protein was cleaved into IL-1Ra products by elastase (Fig. 4D) and mouse PEC lysate (Fig. 4E).

Human NK cells were isolated from peripheral blood, cultured with IL-2, and then stimulated with PMA and A23187 ionophore (Fig. 5). The level of activation was assessed by surface expression of the CD107a degranulation marker (Fig. 5A). Supernatants from activated NK cells were then directly added to chimeric IL-1Ra (4 h, 37°C). As shown in Fig. 5B, activated NK cell supernatants cleaved chimeric IL-1Ra, but not chimeric IL-1Ra^Δ102–120 (Fig 5C).

To test activation of chimeric IL-1Ra in an animal model that is primarily composed of inflammatory conditions, sterile Matrigel was mixed with LPS and either anakinra or chimeric IL-1Ra. Matrigel plugs were then implanted under the skin of wild-type mice. After 24 or 48 h, plugs were collected, and the cellular infiltrate was isolated and quantified (Fig. 6A). At 24 h, there was a nonsignificant reduction in the infiltrates, although, as shown in Fig. 6B, incorporation of anakinra into the Matrigel reduced the cellular infiltrates at this time point. However, after 48 h with the chimeric IL-1Ra, significantly reduced numbers of infiltrating cells were present (Fig. 6A). As demonstrated by flow cytometer analysis, infiltrating cells were composed of CD11b⁺/GR1⁺/Ly6G⁺ myeloid cells (Supplemental Fig. 4), suggesting that chimeric IL-1Ra requires neutrophil-derived enzymes to be activated at an inflammatory site.

The balance between IL-1 and its endogenous inhibitor, IL-1Ra, can affect disease severity. Infants who have a deficiency in IL-1Ra because of genetic mutations develop intense inflammatory conditions accompanied by high mortality (12–14). IL-1β is highly active in humans, and barely elevated levels of circulating IL-1β are sufficient to account for autoinflammatory diseases, that include several inflammatory and metabolic manifestations (15). Thus, inhibition of IL-1 activities using recombinant IL-1Ra (anakinra) has been shown to be effective in type 2 diabetes, gout, cardiac remodeling, rheumatoid arthritis and osteoarthritis, as well as in a broad list of other autoinflammatory conditions (reviewed in Ref. 3). However, blocking IL-1 activity is not without its drawbacks. For example, treatment with anakinra, anti–IL-1β (canakinumab) or monoclonal anti–IL-1R1 Abs can result in neutropenia (16) and an increased risk of infections (17–19). Anakinra is contraindicated in immune-compromised patients and in patients with a history of recurrent infections. Safety thus becomes a major concern, particularly in an aging population.

Our objective was to develop an approach that allows one to block IL-1 activities primarily at sites of active inflammation, without global blockade of IL-1R. For example, in gouty arthritis, inflammation is primarily in a single joint and current therapy for refractory gout—that is, nonresponsive to standard therapies—is systemic anakinra or canakinumab (4, 20, 21). Because refractory gout is found mostly in older subjects, infections are increased in patients treated with IL-1 blocking agents. Indeed, it is well established that gouty arthritis is an IL-1β–mediated disease, and inflammation in the affected joint is nearly entirely due to the activity of neutrophils (22). As such, the milieu of a gout-affected joint is ideal for activating chimeric IL-1Ra. The neutrophil-rich gouty joint will also be rich in PR3, elastase, and cathepsin, which are released from dying neutrophils and can therefore process the chimera to provide active IL-1Ra at the site (22). Based on the chimeric approach, unaffected tissues, such as the lung and the skin, are spared from excessive IL-1R blockade and can fully participate in host defense against invading organisms.

In the current study, we used the native protein sequence of the NTP of the IL-1β precursor, fused in tandem with the active protein sequence of IL-1Ra. Thus, chimeric IL-1Ra is composed of two native molecules. Because the NTP of IL-1β precursor prevents the binding of the precursor to IL-1R1, we anticipated that IL-1Ra would remain inactive and unable to bind IL-1 receptor. Indeed, chimeric IL-1Ra appears to be biologically inert until the NTP is cleaved. We also demonstrate that upon adding elastase in vitro, chimeric IL-1Ra is converted to active IL-1Ra, and it reduces IL-1 activity in a concentration-dependent manner. It is presumed that intermediate forms of digested chimeric IL-1Ra, for example, because of trypsin digestion, are further cleaved by local elastase or neutrophil lysate to produce active IL-1Ra. This was demonstrated using the chimeric IL-1Ra^Δ1–70 mutant protein.

Activation of chimeric IL-1Ra was also demonstrated in vivo using a Matrigel model. Expectedly, the addition of anakinra to LPS-containing Matrigel plugs lowered the number of infiltrating cells 24 h following implantation; at this time, chimeric IL-1Ra was minimally active, consistent with the low levels of inflammatory proteases. However, as invading neutrophils died and released proteases into the immediate environment, chimeric IL-1Ra became active. At 48 h, there was a highly significant reduction in subsequent cell infiltration. The Matrigel model has limitations in demonstrating the dynamics of ongoing inflammation. The plugs accumulate inflammatory agents upon infiltration, but not at the immediate point of implantation, creating conditions in which anakinra is immediately active, but the chimeric IL-1Ra is pending neutrophil infiltration. Indeed, the data support the concept that as inflammation increases at a local site, chimeric IL-1Ra is activated and inflammation is limited; noninflamed sites lack the ability to activate chimeric IL-1Ra.

In addition to neutrophil proteases, stimulated macrophages also activate chimeric IL-1Ra. Upon cell death, macrophages release proteases such as PR3, which cleave the NTP of the IL-1β precursor (22, 23). As such, these same proteases cleave chimeric IL-1Ra. Indeed, in the current study, lysates prepared from macrophages activated with LPS and ATP cleaved chimeric IL-1Ra. Moreover, activated macrophages converted chimeric IL-1Ra independent of cell death, indicating that cleavage of chimeric IL-1Ra is also due to secretion of proteases and not solely due to the passive release of necrotic cell content. In addition, ATP-dependent activation of caspase-1 in the macrophage also results in pyroptosis (24). In this particular cell death pathway, dying macrophages rich in IL-1β, IL-18, and proteases, release cell content through ruptured membranes (25), resulting in protease-rich milieu that can activate chimeric IL-1Ra. In addition, Granzyme A, an NK cell–related enzyme, cleaves IL-1β precursor four residues downstream to the caspase-1 cleavage site (26, 27); we therefore also observed that proteases derived from activated human NK cells cleaved chimeric IL-1Ra. Furthermore, this cleavage correlated with increased NK cell degranulation.

Anakinra has been used to treat acute myocardial infarction (AMI) (28, 29). The cleavage of chimeric IL-1Ra is therefore highly relevant to suppressing local inflammation in AMI. In AMI mouse models of left coronary artery occlusion, infiltrating neutrophils dominate the first 24 h, whereas macrophage infiltration reaches a peak at 3 d and then persists (30). Thus, chimeric IL-1Ra will be presumably generated early after AMI by neutrophils and later by infiltrating macrophages. Anakinra has also been used to treat occlusive stroke, demonstrating improved recovery rates (31); high-dose systemic anakinra (2 mg/kg/h for 3 d) was used to treat stroke patients to facilitate drug entry into the neutrophil-rich infarcted area. However, patients with stroke are vulnerable to infections, particularly pneumonia (32). According to the current study, high systemic levels of chimeric IL-1Ra will also be required to penetrate into the brain, but because IL-1 serves host defense by representing a critical upstream mediator of neutrophil extravasation to phagocytize invading bacteria, high systemic levels of chimeric IL-1Ra will probably not interfere with this important host defense property.

Importantly, although we do not anticipate that host defense will be impaired by high systemic concentrations of chimeric IL-1Ra, we also provide a rescue mechanism for generating large amounts of active IL-1Ra. As we demonstrate in this study, the naturally occurring protease inhibitor, AAT prevented elastase-mediated activation of chimeric IL-1Ra. AAT is an approved therapeutic with a 30-y history of safety and i.v. infusions of AAT are explored in an expanding number of inflammatory conditions, as reviewed (33).

Chimeric IL-1Ra inflammation-dependent activity may also apply to the treatment of individuals with various types of cancer. Chronic inflammation is a consistent characteristic of sites of rapid tumor growth (34). Surrounded by myeloid cells, such as neutrophils and macrophages, these foci are plentiful with enzymes and cytokines, including NK cell–derived granzymes. Indeed, IL-1R blockade was shown to be effective in reducing tumor size, angiogenesis, and metastases in several setups (9, 35, 36), and there is now documented benefit of blocking IL-1 in patients with cancer (37). However, cancer patients are often treated for prolonged periods with bone-marrow suppressing drugs and, therefore, are exposed to an increased risk of infection. In addition to IL-1β, many tumors produce the precursor of IL-1α, which promotes angiogenesis and tumor growth (38). Anti–IL-1α monotherapy has been used successfully to treat patients refractory to antitumor therapies and those losing weight (39). Thus, blocking IL-1R with anakinra is also a therapeutic target in these patients. However, despite its attractive capacity to inhibit all IL-1 activity, anakinra therapy places cancer patients with a suppressed bone marrow at a high risk for a septic event. Thus, chimeric IL-1Ra has potential utility in treating tumor-mediated inflammation in the context of cancer by allowing systemic administration without increasing the risk of infection.

The authors have no financial conflicts of interest.