Production of IL-1 receptor antagonist by hepatocytes is regulated as an acute-phase protein in vivo

2.1 Hepatocytes are a major source of circulating IL-1Ra in response to LPS

We recently demonstrated that sIL-1Ra mRNA is produced in different organs following systemic administration of LPS to mice 11. In order to examine the relative importance of the liver as a source of IL-1Ra in the circulation, C57BL / 6 mice were injected with LPS. The results show that circulating levels of IL-1Ra were elevated in mice 4 h after injection with LPS (9.1 ± 1.8 ng / ml, four mice), whereas IL-1Ra remained undetectable in the plasma of mice injected with saline (< 0.3 ng / ml). sIL-1Ra mRNA was not present in the lung, liver and spleen of mice injected with saline, but its production was strongly up-regulated after LPS administration (Fig. 1). In contrast, icIL-1Ra1 mRNA was not detected in any of the examined organs.

To compare the relative amounts of IL-1Ra synthesized in each organ in response to LPS, we determined the production of total IL-1Ra protein by the liver, spleen and lung. The results show that the total content of IL-1Ra in the liver was six- and tenfold higher than in the lung and spleen, respectively (Table 1). This result indicates for the first time that the liver is the major source of IL-1Ra following LPS injection in vivo.

Different cells within the liver such as macrophages and fibroblast-like cells may produce IL-1Ra in response to LPS 12, 13. By in situ hybridization, we determined the cellular source of IL-1Ra in the liver and which IL-1Ra isoforms were produced. As shown in Fig. 2, a positive signal for sIL-1Ra, but not icIL-1Ra1, mRNA was observed in parenchymal hepatocytes of mice injected with LPS. In contrast, no signal was detected with control sense probes or in the liver of mice injected with saline (data not shown). These results demonstrate that hepatocytes contribute to the circulating levels of IL-1Ra following in vivo stimulation by LPS.

The presence of the different IL-1Ra isoforms can also be distinguished by Western blot analysis, as shown in Fig. 3 A. Cell lysates of resident peritoneal macrophages and of RAW 264.7 murine macrophage cells stimulated with LPS (lanes 1 and 3) contained both 18-kDa icIL-1Ra1 and 16-kDa icIL-1Ra3, a recently characterized low molecular mass intracellular isoform. However, only icIL-1Ra1 but not icIL-1Ra3 was constitutively present in skin extracts (lane 2). As recently described 14, the cell lysates of LPS-stimulated peritoneal macrophages contained two additional bands (lane 1): a higher molecular mass band corresponding to pro-sIL-1Ra and a very faint band of 17 kDa corresponding to mature sIL-1Ra (detected only on the original autoradiograph). Western blot analysis was performed on proteins extracted from the liver and spleen of mice injected with LPS. Three different bands were present in extracts of the two organs (Fig. 3 B, lanes 2 and 3). A weak middle band migrated at the same level as recombinant 17 kDa murine sIL-1Ra and represented unglycosylated mature sIL-1Ra. The presence of the lower and the higher molecular mass bands is consistent with 16-kDa icIL-1Ra3 and pro-sIL-1Ra, respectively. As opposed to RAW 264.7 cells and proteins extracted from the skin (Fig. 3 B, lanes 1 and 4), icIL-1Ra1 protein was not detected in extracts of the spleen and liver, thus further supporting the findings at the mRNA level.

2.2 Production of sIL-1Ra mRNA by the liver in response to turpentine-induced local inflammation

Plasma concentrations of IL-1β, IL-6, IL-1Ra and SAA as well as IL-1Ra and SAA1 mRNA levels in the liver were determined at different time-points after local injection of turpentine. In contrast to mice treated with LPS, circulating IL-1Ra and IL-1β levels remained undetectable after turpentine-induced inflammation 11. Plasma levels of IL-6 increased with a peak occurring 24 h after the injection of turpentine (Fig. 4 A). However, the levels of IL-6 were much lower than those observed in response to LPS 11. Plasma levels of SAA were also elevated and remained high up to 48 h after stimulation (Fig. 4 B). Production of sIL-1Ra and SAA1 mRNA by the liver were also up-regulated in response to local turpentine-induced inflammation (Fig. 5). In contrast to IL-1Ra, neither IL-6 nor IL-1β mRNA were present in the liver of mice injected locally with either saline or turpentine (data not shown). Thus, although sIL-1Ra mRNA was present in the liver of mice with local turpentine inflammation, no IL-1Ra protein could be detected in the circulation.

IL-6 is known to play a major role in the induction of APP by the liver. To investigate the role of IL-6 in production of sIL-1Ra and SAA by the liver, IL-6^{– / –} and IL-6^+ / + littermates were injected locally with turpentine or saline. Steady-state levels of sIL-1Ra mRNA remained unchanged in the livers of IL-6^{– / –} mice after the injection of turpentine, whereas the levels were significantly up-regulated in IL-6^+ / + mice (Fig. 6 A). The levels of hepatic SAA1 mRNA also were significantly lower in IL-6^{– / –} than IL-6^+ / + mice after turpentine injection as well as in controls injected with saline solution (Fig. 6 B). Consistent with this result, plasma levels of SAA were elevated in IL-6^+ / + mice, but remained undetectable in IL-6^{– / –} mice following turpentine injection (data not shown). These results indicate that IL-6 is required for hepatic production of both IL-1Ra and SAA1. In addition, basal production of SAA but not of IL-1Ra is dependent on IL-6 (Table 2).

2.3 Production of IL-1Ra at the site of inflammation after local injection of turpentine

The production of IL-1Ra in the local tissue in turpentine-induced inflammation was next examined and compared with the levels of IL-1β. In contrast to hepatocytes, both sIL-1Ra and icIL-1Ra1 mRNA were produced in inflamed tissues (Fig. 7 A). The result of the Western blot analysis confirmed that each IL-1Ra protein isoform was present in the inflamed tissues and also increased between 8 h and 48 h (Fig. 7 B). Local IL-1β mRNA and protein production was also up-regulated following turpentine injection. However, in contrast to IL-1Ra, local levels of IL-1β mRNA peaked at 8 h and decreased after 48 h (Fig. 8). Production of IL-6 mRNA exhibited a similar kinetics as IL-1β (Fig. 8). The IL-1Ra / IL-1β protein ratio in the inflamed tissue increased from 10 : 1 at 8 h to over 1000 : 1 at 48 h (data not shown).

To determine whether turpentine-induced IL-1Ra production was regulated in a similar manner in the liver and in inflamed tissues, local IL-1Ra mRNA levels were also measured in IL-6^{– / –} and in IL-6^+ / + mice. As opposed to the liver, local production of IL-1Ra after turpentine injection was not significantly different in IL-6^{– / –} and IL-6^+ / + mice (Fig. 9). Local tissue levels of IL-1Ra protein were normalized to total protein concentrations. In accordance with the results at the mRNA level, the concentrations of IL-1Ra protein were not significantly different in IL-6^{– / –} and IL-6^+ / + mice (72 ± 12 vs. 76 ± 11 pg / μg total protein, p > 0.05).

2.4 icIL-1Ra3 is an in vivo product ofsIL-1Ra mRNA but not of icIL-1Ra1 mRNA

Recently, 16-kDa icIL-1Ra3 has been characterized as a smaller molecular mass isoform of IL-1Ra. Using an in vitro translation assay, icIL-1Ra3 was demonstrated to be produced by alternative translation initation from both sIL-1Ra and icIL-1Ra1 mRNA 6. To examine the production of icIL-1Ra3 in vivo, we used transgenic mice overexpressing either human sIL-1Ra or icIL-1Ra1. We previously showed that IL-1Ra is not detected in normal tissues except for the constitutive presence of icIL-1Ra1 in the skin and intestine 14. As shown in Fig. 10 A, both glycosylated sIL-1Ra and icIL-1Ra3 were present in different tissues of mice possessing the sIL-1Ra transgene and in the lysates of murine macrophage RAW 264.7 cells transfected with the transgenic construct containing human sIL-1Ra. In contrast, only icIL-1Ra1 and not icIL-1Ra3 was detected in mice overexpressing icIL-1Ra1 mRNA (Fig. 10 B). Similar results were obtained in the different transgenic mouse lines. This result indicates that icIL-1Ra3 is primarly produced from sIL-1Ra mRNA in vivo.

4.1 Animals and treatments

IL-6^{– / –} and IL-6^+ / + C57BL / 6 control mice were purchased from Jackson Laboratory (Bar Harbor, ME). Transgenic mice overexpressing either human sIL-1Ra or icIL-1Ra1 under the control of a β-actin-CMV promoter were generated as follows. Human sIL-1Ra and icIL-1Ra1 coding regions were subcloned into pCAGGSβ. High levels of protein expression were obtained with this vector in transgenic mice 27. The fragments containing the IL-1Ra sequence, the promoter, and the polyA signal regions were gel purified and injected into fertilized eggs obtained after crossing FVB females with DBA / 1 males, as described 28. The presence of the transgene was identified by PCR in founder mice and their offspring, and expression of human IL-1Ra isoforms was then examined in the transgenic offspring.

Six- to eight-week old C57BL / 6 mice were used to examine systemic inflammation. LPS (Escherichia coli 055 : B5, Difco, Detroit, MI) was injected i. p. at a dose of 100 μg, as previously described 11. To create a model of local tissue inflammation, turpentine was injected s. c. at a dose of 100 μl in each hindlimb, as previously described 29. All procedures using mice were approved by the University of Colorado Health Sciences Center Committee on Care and Use of Animals.

4.2 Determination of plasma levels of cytokines and SAA

Plasma and tissue levels of IL-1β and IL-6 were measured using ELISA kits specific for murine cytokines (Endogen, Inc, Cambridge, MA). Murine IL-1Ra levels were measured using a sandwich ELISA, as previously described 11. The detection limits for IL-1β, IL-6 and IL-1Ra were 10, 30, and 300 pg / ml, respectively. Plasma levels of SAA were determined using a direct ELISA 30. The SAA ELISA sensitivity was 15 μg / ml.

4.3 RNA isolation and RNase protection assays

Total RNA from the different tissues was extracted with Triazol (Life Technologies, Gaithersburg, MD). Riboprobes complementary to murine IL-1Ra, IL-1β and glyceraldehyde-3-phosphate dehydrogenase (GAPDH) mRNA were created as recently described 31. The riboprobe for IL-1Ra is able to distinguish between the two murine IL-1Ra mRNA isoforms as different protected size products following RNase treatment. Riboprobes for IL-6 and SAA1 mRNA were created using the following primers: 5′IL-6, GCA TGA ATT CTT AAT TAC ACA TGT TCT CTG; 3prime;IL-6, GCA TAA GCT TGT TCT TCA TGT ACT CCA GG; 5′SAA, GCA TGA ATT CAA CTC AGA CAA ATA CTT CCA TGC TCG; and 3prime;SAA, GCA TAA GCT TCT GAG CTA ATA GGA GGA CGC TCA GT. The PCR products were cleaved with EcoRI and HindIII, gel purified, and cloned into pBluescript SK + (Stratagene, San Diego, CA). The plasmids containing the probes were linearized with EcoRI and transcribed with T7 RNA polymerase and ³²P-labeled CTP. The RNase protection assays were performed as recently described 10.

4.5 In situ hybridization

Mice were injected i. p. with either LPS or saline and killed by CO₂ asphyxiation. The liver was dissected and immediately frozen in OCT embedding matrix on dry ice and isopentane, then stored at − 80 °C until use. Sections (5 μm) were cut in a cryostat and mounted on Superfrost-plus slides (Eric Scientific Company, Portsmouth, NH). The in situ hybridization studies were performed as recently described 14.

4.6 Protein preparation and Western blot analysis

For Western blot analysis, total proteins from different tissues were prepared using Triazol, as recently described 14. Concentration of total proteins was determined using a protein assay kit (Bio-Rad). For IL-1Ra and IL-1β ELISA, organs were weighed and homogenized in PBS-0.1 % Tween-20 in a 1 : 4 dilution (weight : volume). The Western blot analysis was performed using a horseradish peroxidase-labeled goat anti-murine IL-1Ra antibody that recognizes all the described IL-1Ra isoforms. For Western blot analysis on protein extracts from transgenic mice expressing human IL-1Ra, a mouse anti-human IL-1Ra mAb was used 10.