Production of mice deficient in genes for interleukin (IL)-1alpha, IL-1beta, IL-1alpha/beta, and IL-1 receptor antagonist shows that IL-1beta is crucial in turpentine-induced fever development and glucocorticoid secretion

Abstract

Interleukin (IL)-1 is a major mediator of inflammation and exerts pleiotropic effects on the neuro-immuno-endocrine system. To elucidate pathophysiological roles of IL-1, we have first produced IL-1alpha/beta doubly deficient (KO) mice together with mice deficient in either the IL-1alpha, IL-1beta, or IL-1 receptor antagonist (IL-1ra) genes. These mice were born healthy, and their growth was normal except for IL-1ra KO mice, which showed growth retardation after weaning. Fever development upon injection with turpentine was suppressed in IL-1beta as well as IL-1alpha/beta KO mice, but not in IL-1alpha KO mice, whereas IL-1ra KO mice showed an elevated response. At this time, expression of IL-1beta mRNA in the diencephalon decreased 1.5-fold in IL-1alpha KO mice, whereas expression of IL-1alpha mRNA decreased >30-fold in IL-1beta KO mice, suggesting mutual induction between IL-1alpha and IL-1beta. This mutual induction was also suggested in peritoneal macrophages stimulated with lipopolysaccharide in vitro. In IL-1beta KO mice treated with turpentine, the induction of cyclooxygenase-2 (EC 1.14.99.1) in the diencephalon was suppressed, whereas it was enhanced in IL-1ra KO mice. We also found that glucocorticoid induction 8 h after turpentine treatment was suppressed in IL-1beta but not IL-1alpha KO mice. These observations suggest that IL-1beta but not IL-1alpha is crucial in febrile and neuro-immuno-endocrine responses, and that this is because IL-1alpha expression in the brain is dependent on IL-1beta. The importance of IL-1ra both in normal physiology and under stress is also suggested.

Figure 1

Targeted disruption of the IL-1 gene family by homologous recombination. Genomic loci, targeting vectors, and mutated loci of IL-1α, IL-1β, and IL-1ra genes are indicated. (A) Targeting strategy for the IL-1α gene. A DNA fragment including exon 5 was replaced with the lacZ-PGK-hph-pA cassette (LacZ-hph). MC1-DT was ligated at the 5′ end of the vector for negative selection. (B) Targeting strategy for the IL-1β gene. A DNA fragment including exon 3–5 was replaced with the lacZ-PGK-neo-pA cassette (LacZ-neo). (C) Targeting strategy for the IL-1ra gene. A DNA fragment containing all the coding regions for the IL-1ra secretion form was replaced with the PGK-neo-pA cassette (neo). DT was ligated at the 5′ end of the vector for negative selection. Arrowheads, Primers (P1–P5) used for PCR. Striped lines, Probes for Southern blot analysis are indicated as 5′ and 3′ probes. Arrows, Expected sizes of DNA fragments obtained by restriction enzyme digestions. Black boxes, Exons. White boxes, Introns. Restriction enzyme sites are also indicated in the maps. w.t., Wild-type.

Figure 2

Southern blot analysis of DNA from cloned ES cells and from transgenic mouse tails bred by heterozygous intercross. Hybridization was carried out using 5′ probes of each gene as shown in Fig. 1. All the bands detected correspond to either the wild-type (W) or mutant allele (M), as expected from the genome structure shown in Fig. 1 (arrows). (A) DNA from IL-1α KO ES cells digested with EcoRI. The number shows the ES clone number, and R1 shows the parental ES cells. (B) DNA from IL-1β KO ES cells digested with KpnI. (C) DNA from IL-1ra KO offspring digested with BamHI. +/+, +/−, and −/−, Expected genotypes.

Figure 2

Southern blot analysis of DNA from cloned ES cells and from transgenic mouse tails bred by heterozygous intercross. Hybridization was carried out using 5′ probes of each gene as shown in Fig. 1. All the bands detected correspond to either the wild-type (W) or mutant allele (M), as expected from the genome structure shown in Fig. 1 (arrows). (A) DNA from IL-1α KO ES cells digested with EcoRI. The number shows the ES clone number, and R1 shows the parental ES cells. (B) DNA from IL-1β KO ES cells digested with KpnI. (C) DNA from IL-1ra KO offspring digested with BamHI. +/+, +/−, and −/−, Expected genotypes.

Figure 2

Southern blot analysis of DNA from cloned ES cells and from transgenic mouse tails bred by heterozygous intercross. Hybridization was carried out using 5′ probes of each gene as shown in Fig. 1. All the bands detected correspond to either the wild-type (W) or mutant allele (M), as expected from the genome structure shown in Fig. 1 (arrows). (A) DNA from IL-1α KO ES cells digested with EcoRI. The number shows the ES clone number, and R1 shows the parental ES cells. (B) DNA from IL-1β KO ES cells digested with KpnI. (C) DNA from IL-1ra KO offspring digested with BamHI. +/+, +/−, and −/−, Expected genotypes.

Figure 3

Northern blot analysis of RNA from macrophages from IL-1α (A), IL-1β (B), and IL-1ra (C) KO mice. Thioglycollate-induced peritoneal macrophages were prepared from wild-type (+/+), heterozygous (+/−), or homozygous (−/−) mice. RNA was isolated from the macrophages after incubation with 10 μg/ml LPS for 12 h (A) or 6 h (B and C). Total RNA (5–10 μg) was loaded on a denatured agarose gel and hybridized with either an IL-1α, IL-1β, IL-1ra, or β-actin probe.

Figure 4

Expression of inflammatory cytokine genes in peritoneal macrophages from IL-1α and IL-1β KO mice. Thioglycollate-induced peritoneal macrophages were prepared from wild-type (open triangles) and homozygous IL-1α (filled circles) and IL-1β (filled squares) KO mice. Macrophages were treated with 10 μg/ml LPS for 0, 3, 6, and 12 h, and the total RNA was isolated. Northern blot analysis was carried out using each cytokine probe and a β-actin probe. Relative radioactivity of each cytokine mRNA was calculated as a percentage of β-actin mRNA. (A) IL-1α; (B) IL-1β; (C) IL-1ra; (D) IL-6; and (E) TNF-α mRNA. The values represent the mean ± SD of three independent experiments.

Figure 5

Fever induction after turpentine injection. Wild-type mice (Cont.) and homozygous IL-1 KO mice (KO) (male) were implanted with transmitters to monitor body temperature. Mice were injected subcutaneously with either saline (Sal.) or turpentine oil (Tur.) in both hindlimbs. Fever development was monitored in IL-1α/β (A), IL-1α and IL-1β (B), and IL-1ra (C) KO mice. Arrowheads, Time point of turpentine injection. The number of mice used for each experiment is shown in parentheses. Fever development was compared between these KO mice and wild-type mice injected with turpentine. *P <0.001, ^‡ P <0.01, ^§ P <0.05. These experiments were repeated twice with similar results.

Figure 5

Fever induction after turpentine injection. Wild-type mice (Cont.) and homozygous IL-1 KO mice (KO) (male) were implanted with transmitters to monitor body temperature. Mice were injected subcutaneously with either saline (Sal.) or turpentine oil (Tur.) in both hindlimbs. Fever development was monitored in IL-1α/β (A), IL-1α and IL-1β (B), and IL-1ra (C) KO mice. Arrowheads, Time point of turpentine injection. The number of mice used for each experiment is shown in parentheses. Fever development was compared between these KO mice and wild-type mice injected with turpentine. *P <0.001, ^‡ P <0.01, ^§ P <0.05. These experiments were repeated twice with similar results.

Figure 5

Fever induction after turpentine injection. Wild-type mice (Cont.) and homozygous IL-1 KO mice (KO) (male) were implanted with transmitters to monitor body temperature. Mice were injected subcutaneously with either saline (Sal.) or turpentine oil (Tur.) in both hindlimbs. Fever development was monitored in IL-1α/β (A), IL-1α and IL-1β (B), and IL-1ra (C) KO mice. Arrowheads, Time point of turpentine injection. The number of mice used for each experiment is shown in parentheses. Fever development was compared between these KO mice and wild-type mice injected with turpentine. *P <0.001, ^‡ P <0.01, ^§ P <0.05. These experiments were repeated twice with similar results.

Figure 6

Time course of mRNA expression specific for the IL-1α, IL-1β, COX-1, and COX-2 genes in whole brain after turpentine injection. Wild-type (+/+) or homozygous (−/−) mice were injected with 100 μl turpentine subcutaneously in the left hindlimb. Mice were killed at 0, 6, 12, and 24 h after turpentine injection, and poly A⁺ RNA was isolated from whole brain. Poly A⁺ RNA (5–10 μg) was electrophoresed on a denatured agarose gel and hybridized with specific probes. β-actin was used as a control. (A) IL-1α and IL-1β KO mice; (B) IL-1ra and IL-1α/β KO mice.

Figure 7

Region specificity of mRNA expression in the brain at 12 h after turpentine injection. Samples were pooled from four mice of each genotype. Lane 1, Cerebral cortex; lane 2, diencephalon; lane 3, hippocampus; lane 4, cerebellum. Poly A⁺ RNA (5–10 μg) was purified from each part of the brain and hybridized with probes as indicated. (A) IL-1α and IL-1β KO mice; (B) IL-1ra and IL-1α/β KO mice.

Figure 8

Turpentine-induced serum corticosterone levels in IL-1α/β KO mice. Control (Cont) and IL-1α/β KO mice were injected subcutaneously with saline (white bars) or turpentine (black bars), and serum samples (n = 5) were obtained at 2 h (A) and 8 h later (B). Corticosterone levels were then measured by radioimmunoassay. Although corticosterone levels at 2 h were similar, the levels at 8 h were significantly higher in wild-type than in IL-1α/β KO mice. *P <0.01.

Figure 9

Turpentine-induced serum corticosterone levels in IL-1α (A), IL-1β (B), and IL-1ra (C) KO mice. IL-1α KO (α KO), IL-1β KO (β KO), and IL-1ra KO (ra KO) mice were injected subcutaneously with saline (white bars) or turpentine (black bars), and serum samples (n = 5, except for saline-injected IL-1α KO mice) were obtained 8 h later. The corticosterone level in IL-1α or IL-1ra KO mice is similar to that in wild-type mice, whereas that in IL-1β KO mice is significantly lower. *P <0.001.