Structure and function of the intracellular portion of the mouse interleukin 1 receptor (type I). Determining the essential region for transducing signals to activate the interleukin 8 gene.

The structural and functional relationships of the intracellular portion of mouse interleukin 1 receptor (muIL-1R) type I were examined with regard to activation of the human IL-8 gene in the Jurkat T cell line. C-terminal deletion mutations of muIL-1R revealed that the C-terminal boundary for receptor function is localized between 28 and 42 amino acids from the C-terminal end. The internal deletion mutants between amino acids 364 and 474 had a loss of activity, demonstrating the requirement for a large region of the mIL-1R cytoplasmic portion for receptor function. Amino acid substitution revealed that the putative nuclear localization elements (amino acids at 429-433, 523-527, and 507-519) and putative protein kinase C or A acceptor sites (Ser-431, Ser-509, Ser-528) do not participate in IL-1 signaling to induce IL-8 gene expression. A truncated mutation within the segment, which possesses homology with gp130, beta chain of IL-6R, or a point mutation of box 1- and box 2-like elements within the gp130 homologous segment, abolished the capacity to induce IL-8 gene expression, suggesting similar structural requirements in the cytoplasmic portion of several cytokine receptors.


IL-6R, or a point mutation of box 1-and box 2-like elements within the gp130 homologous segment, abolished the capacity to induce IL-8 gene expression, suggesting similar structural requirements in the cytoplasmic portion of several cytokine receptors.
Interleukin 1 (IL-1)' activity is a property of two polypeptides (IL-la and IL-1p) that have a number of biological functions including immunoregulatory, proinflammatory, and hematopoietic activities (1,2). Interleukin 8 (IL-8) is a recently identified potent chemotactic cytokine for neutrophils and T-lymphocytes (3). IL-8 is induced by stimulating various types of cells with IL-1 as well as TNF and mitogens (4), and an IL-I-, TNF-, and phorbol 12-myristate 13-acetate-responsive element in the 5"flanking region of the IL-8 gene has been identified (5). However, the mechanism by which IL-1 activates IL-8 gene expression remains to be elucidated.
IL-1 elicits its activities by binding to one of two distinct * The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. specific receptors (IL-1R) expressed on the surface of responsive cells. On T cells and fibroblasts, an 80-kDa IL-1R (type I) has been identified (6-8), whereas the major type of IL-1R on B cells, monocytes, neutrophils, and bone marrow cells seems to be a 60-kDa IL-1R (type 11), which differs antigenically from type I receptors on T cells (9,10). Both types of the IL-1R have been cloned from mouse and human cell lines (11)(12)(13). Analysis of the deduced amino acid sequence revealed that these receptors possess a single membrane-spanning segment and belong to the immunoglobulin (Ig) super gene family. A major difference between the type I and I1 receptors is in the cytoplasmic region, which is much shorter in the type I1 receptor.
Transfection of type I IL-1R cDNA into CHO cells results in functional receptor expression capable of initiating responses that lead to prostaglandin synthesis and cytokine secretion (14). The extracellular region of type I IL-1R binds IL-1 with the same affinity as the complete receptor (14, 15), while the cytoplasmic region is essential for the signal-transducing activity (14). The IL-1R type I is phosphorylated soon after IL-1 binding (16). However, the cytoplasmic region of type I IL-1R does not possess homology to the catalytic subunit of the protein kinases (11,12), and the IL-1R does not possess intrinsic protein kinase activity (16). The presence of a putative PKC acceptor site has been deduced; however, its function remains unclear (11,12). The type I receptor is internalized after IL-1 binding (17, 18). IL-1-receptor complexes are found in nuclei (18,19). The cytoplasmic portion of the IL-1R possesses four elements of homology to nuclear localization sequences (NLS) of the glucocorticoid receptor or polyoma virus large T-antigen (19). The cytoplasmic portion of IL-1Rs shows homology with Drosophila Toll protein, which is claimed to function in forming a dorsal-ventral polarity (20, 21). Recently, Heguy et al. (22) reported that several amino acids of the human IL-1R cytoplasmic portion that are conserved between human IL-1R and Toll protein are essential for activation of the IL-2 gene by IL-1 and phytohemagglutinin stimulation.
In gp130, the signal-transducing component of several cytokines, including IL-6, leukemia inhibitory factor, and oncostatin M (23), the cytoplasmic segment near the transmembrane region is sufficient to generate the growth signal (24). This segment shows sequence similarity with several other members of the cytokine receptor family including the IL-2RP, the erythropoietin receptor, and the granulocyte colonystimulating factor receptor. Two stretches of amino acids, boxes 1 and 2, are highly conserved in the homologous segment (24). The importance of box 2 for signal transduction has been demonstrated in IL-ZRP as well as in gp130 (25.26). We found that the cytoplasmic portion of the mouse type I IL-1R also possesses this homologous segment, amino acid region 435-484, containing box 1-and box 2-like elements.
To establish whether or not the domains mentioned above in the mouse IL-1R cytoplasmic portion are essential for transduction of the IL-1 signal for activation of the human IL-8 gene, we have examined the effects of serial deletions or amino acid substituted mutants in the cytoplasmic portion of the receptor. We found that NLS-like elements and putative phosphorylation sites are not required for IL-8 gene activation and that the large part of the cytoplasmic segment conserved between muIL-1R and IL-6R gp130 is important for the receptor function.
Isolation of the Mouse IL-1R Type I cDNA and Construction of Expression Vectors-A fragment corresponding to the region from the N-terminal to AccI site on mouse IL-1R type I cDNA was obtained by reverse transcriptase polymerase chain reaction (PCR), using mouse NIH3T3 total RNA and specific primers, which were synthesized according to the reported sequences (11). Using this PCR product as a probe, two IL-IR cDNA clones were isolated from an EL-4 mouse T-cell thymoma cDNA library which was constructed using the A ZAP system (Stratagene, La Jolla, CA). The phagemid clone, C7, contained 1.5-kb fragments (EcoRI-EcoRI) of mouse IL-1R type I cDNA but lacked 0.2 kb of the N-terminal region and about 50 bp of the C-terminal region. Another clone, C1, containing a 1.8kb IL-1R cDNA fragment (EcoRI-EcoRI), lacked some of the Nterminal region but possessed the entire C-terminal region of the IL-1R gene.
To construct an expression vector for the mouse IL-1R type I, an incomplete N-terminal fragment (EcoRI-AccI) of clone C7 was replaced with the PCR fragment (EcoRI-AccI) containing the entire Nterminal as described above, and inserted into the EcoRI site of the vector PUCl9 to make PUC-C7. Next, a 1.7-kb cDNA fragment (EcoRI-EcoRI) of plasmid PUC-C7 was transferred into the EcoRI site of PSG5UT, which is a modified version of PSG5 (Pharmacia LKB Biotechnology Inc., Uppsala, Sweden), containing a universal terminator just downstream of the integration site of PSG5. The resultant plasmid, pIR7, had an entire N-terminal region of IL-1R cDNA but still lacked about 50 bp of its C-terminal region. Therefore, another PCR fragment, containing the region from HindIII to the Cterminal end of IL-1R cDNA, was prepared using specific primers and phagemid C1 clone as a template, and an incomplete C-terminal fragment (HindIII-BamHI) of pIR7 was replaced with this PCR fragment (HindIII-BgDI). The resulting plasmid was named pIR8 and contained the full-length mouse IL-1R type I coding region.
To construct deletion mutants of the C-terminal region, pIR9-16, various 3'-PCR primers were designated which have an antisense of stop codon following the sequence of deleted sites described in Fig. 2, and PCR was performed utilizing these 3"primers and 5'-primer R7. PCR products digested by HindIII and BglII were inserted into HindIII and BamHI digested pIR7 in a similar manner to the construction of pIR8. Similarly, serial deletion mutants from the HindIII site near the transmembrane portion, pIR71 and pIR73, were constructed using various 3"primers that correspond to each deletion site.
The internal deletion mutants (ID33-ID38) and some amino acid substitution mutants were constructed by site-directed mutagenesis using PCR as described (29). For example, ID35, the mutated 5'primer, R35a (5"TACTAATGAAGTGAGAGATATGGGA GGCTT-3'), and the 3'-primer, R35b (5"TATCTCTCACTTCATTAGTA-ACCTCGATGG-3') were designated to delete amino acids 427-438 of the IL-1R. The initial PCR was performed between primers R7 and R35b and between primers R8 and R35a. Using a mixture of each initial PCR product as a template and primers R7 and R8, a second PCR reaction was performed to prepare a mutated cytoplasmic fragment. The DNA sequences of the mutational sites were confirmed by dideoxy chain termination (30). The -185/IL-8-CAT and -133/IL-8-CAT plasmid is a derivative of PUX-CAT (31) containing from -185 or -133 to +44 bp of the 5'-flanking region of the human IL-8 gene (5). Both plasmids possess the AP-1 binding site (-120 to "126) and the NFIB site (-80 to "70) of the regulatory region of the IL-8 gene.
" PCR primers used were prepared using a DNA synthesizer (PCR Mate, Applied Biosystems).
Transfection-Jurkat cells were collected after 3 days of incubation at 37 "C in RPMI medium containing 5% fetal calf serum and washed. The cells (4 X IO6) were transfected with 10 pg of plasmid (2 pg of wild type or mutant IL-1R expression plasmid, 4 pg of -185/IL-8-CAT and 4 pg of PUC119) using DEAE-dextran (32), and incubated in RPMI medium without fetal calf serum for 24 h. Jurkat cells were then stimulated with 100 ng/ml IL-la or 151Y for 24 h.
CHO cells were briefly treated with trypsin and washed with serumfree DMEM, and then cells (4 X lo6) were transfected with 10 pg of plasmid DNA (5 pg of receptor plasmid and 5 pg of PUC119 for receptor binding assay; or 5 pg of receptor plasmid, 4 pg of -133/IL-8-CAT and 1 pg of PUC119 for CAT assay) using DEAE-dextran in suspension (32). CHO cells were incubated for 48 h in DMEM containing 5% fetal calf serum, prior to the receptor binding assay. For CAT assay, transfected cells were incubated for 24 h and then stimulated with 100 ng/ml 151Y for 24 h. CAT Assay-Jurkat cells collected by centrifugation or CHO cells collected by a cell scraper (Costar, Cambridge, MA), were washed with phosphate-buffered saline and 0.25 M Tris-HC1 (pH 7.5) and lysed by sonication in 100 p1 of 0.25 Tris-HC1 (pH 7.5). Cell lysates were then centrifuged at 15000 rpm in 1-ml Eppendorf tubes, and supernatants were collected. CAT activity was determined using the whole cell extract by the method described by Gorman et a2. (33). To avoid errors due to variation in sample collection, the protein concentration of cell extracts was measured using the method described by Bradford (34), and the same amount of protein was used for each CAT assay. After thin layer chromatography, the radioactivity was determined using a Bioimaging analyzer (BA100; Fuji Co. Ltd., Tokyo, Japan), and the percentage of conversion of chloramphenicol into acetylated forms was calculated.
Receptor Binding Assay of Transient cDNA Transfectants-CHO monolayers in dishes were rinsed twice with serum-free RPMI containing 0.1% bovine serum albumin, 10 mM HEPES (pH 7.4), 0.1% Na2N3, and incubated in a 1-ml volume of the same buffer in the presence of various amounts of '2SI-IL-la (89.1 pCi/pg, Du Pont-New England Nuclear) at 37 "C for 1 h. After two washes with the same solution, cells were collected by cell scraper and the cell-associated radioactivity was counted using a y-radiation counter. Nonspecific binding was determined in the presence of a 1000-fold excess of unlabeled IL-la. Scatchard analyses were performed to calculate the receptor number and dissociation constant (&) (35).

Transfected Murine IL-1R Type I Is Functional in a Human
T-cell Line, Jurkat-IL-1R is found on many cell types including T-cells, fibroblasts, endothelial cells, and hepatocytes. Therefore, it is difficult to identify a cell which does not express IL-1R molecule. The human IL-la mutant, 151Y, which has a Tyr-151 substitution for Asp (28) is useful, since 151Y functions as an agonist for mouse IL-1R, whereas binding to human IL-1R does not cause signal transduction. Therefore, 151Y can act as a competitive antagonist for human IL-1R (28). Consequently, expression of the cDNA for the mouse IL-1R gene on a human cell line can selectively stimulate the expressed mouse IL-1R without any signals from the human IL-1R, which is natively expressed or induced by DNA transfection in various types of human cells. This assay system is unique and useful for the functional analysis of the IL-1R in human cells.
To evaluate whether the IL-1 signal was transduced through the expressed mouse IL-lR, the reporter plasmid, -185/IL-8-CAT, which contains from -185 to +44 of the 5"flanking region of the human IL-8 gene fused to the bacterial chloramphenicol acetyltransferase (CAT) gene, was co-transfected with the expression plasmid for the mouse IL-1R into the human T-cell line, Jurkat.
IL-la or 151Y had no effect on the induction of the IL-8 gene in Jurkat cells (Fig. 1, left), indicating that no functional receptors were expressed on human Jurkat cells. In contrast, when the full-length mouse IL-1R type I cDNA was transiently expressed in Jurkat cells, the IL-8 gene was induced in response to IL-la or 151Y (Fig. 1, right). The fact that the Jurkat cell was converted into an IL-1-responsive cell by engineered expression of the mouse IL-1R gene indicates that the post-receptor factors required for the activation of the IL-8 gene are present in Jurkat cells. The IL-la mutant 151Y activated the IL-8 gene more efficiently than wild type IL-la.
Critical Boundary of IL-1R Cytoplasmic C-terminal Portion-To investigate the requirement of the C-terminal region in the cytoplasmic portion of muIL-1R to transduce an IL-1 signal, several plasmids that express receptors deleted in the C-terminal region were constructed as described under "Materials and Methods" and were co-transfected with -185/IL-8-CAT into Jurkat cells. Mutant receptors with a deletion of up to 28 amino acids in the C-terminal region (pIR15) conferred the induction of CAT activity by 151Y, whereas the deletion of 42 amino acids (pIR16) abolished the transducing activity (Fig. 2B). Similarly, mutants (pIR10, -11, -12, and -14) having further C-terminal deletions were also defective ( Fig. 2B and data not shown). These results indicate that the C-terminal region between amino acids 515 and 529 is critical for transducing IL-1 the signal.
This C-terminal critical boundary region (515-529 amino acids) possesses the entire second element (523-527 amino acids) and part of the first element (507-519 amino acids) of the C-terminal NLS-like elements, which are characterized by the presence of Pro and basic amino acids. To examine the contribution of these elements, amino acid substitution mutants were constructed. The substitution of Pro and Arg residues in the second C-terminal NLS-like element by Gly and Ala (NL-2) (Fig. 3A), did not reduce the 151Y dependent induction of CAT activity (Fig. 3B). Similarly, mutant NL-3, possessing two amino acid substitutions in the first C-terminal NLS-like element (Fig. 3A), retained the ability to induce CAT activity (Fig. 3B). These results demonstrated that the C-terminal NLS-like element does not participate in the receptor function. The element RRS (526-528), located in the C-terminal boundary, is presumed to be susceptible to phosphorylation by either of protein kinases A or C (36), but substitution of Ser-528 by Ala (MS3) did not affect the transducing activity of 151Y (Fig. 3C). A NPXY element required for internalization has been identified in the human low density lipoprotein receptor. A similar element, NLXY, is found at amino acids 517-520 of the C-terminal boundary region of muIL-1R. However, M11, containing substitutions of Asn-517 and Tyr-520 by Gly in this element (Fig. 3A), retained the ability to induce CAT activity (data not shown). These results indicate that another crucial element may be present within the C-terminal boundary region.
TNF and IL-1 share many biological activities, and IL-8 can be induced by TNF as well as IL-1. Another type of PKC acceptor motif, (S/T)-X-(R/K), which is found in both TNFRl and TNFR2 (37), is also present at amino acids 509-511 of muIL-1R C-terminal region. However, the replacement of Ser-509 by Ala in the S-X-K motif (MS2) (Fig. 3A) increased the induction level of CAT activity (Fig. 3C), indicating that this element is not essential for transducing activity.
Requirements of the gp130 Homologous Region for Receptor Function-To clarify the requirement for the cytoplasmic segment near the transmembrane portion, a series of deletion mutants from the Hind111 site (corresponding to Ser-364) were prepared (Fig. 4A). The mutants pIR71 and pIR73, with deletions from 35 to 139 amino acids from Ser-364, lacked transducing activity (Fig. 4B), indicating that the region near the transmembrane portion is also essential.
To identify more precisely the cytoplasmic elements involved in the IL-1 signal transduction, plasmids were constructed to express several mutated receptors having internal deletions in the cytoplasmic region, including an ID35 mutant that deleted the putative PKC acceptor site (429-433) overlapping the NLS-like element (Fig. 4A). Unexpectedly, all internal deletions including ID35 lost transducing activity (Fig. 4C), indicating the extensive requirement of the cytoplasmic portion of the muIL-1R for signal-transducing activity.
Interestingly, the segment from amino acid 435 to 484 of muIL-1R shows sequence similarity with the IL-6R gp130 cytoplasmic portion (Fig. 4, A and D). Moreover, two hydrophobic stretches, box 1 and 2 elements of gp130, are well conserved in muIL-1R. Besides these elements, several amino acids conserved between gp130 and other cytokine receptors including Trp-447 are also found in IL-1R. Thus, loss of activities for mutants ID35-ID38 may be explained by the involvement of the gp130 homologous segment for IL-1 signal transmission.
The deleted region of ID35 contains a potential PKC acceptor site (position 431), the NLS-like element (positions 429-433) characterized by basic amino acid-rich sequences, and the box 1-like element. However, as shown in Fig. 5B, replacement of Ser-431 with Ala (MS1) did not affect the capacity to induce the IL-8 gene. Moreover, conversion of three basic amino acids in the NLS-like element to Ala (NL-1) (Fig. 5A) did not reduce the transducing activity (Fig. 5C). These results demonstrate that neither the PKC acceptor site (position 431) nor the NLS-like element (positions 429-433) is required for the receptor function in activation of the IL-8 gene.  In contrast, substitution of hydrophobic amino acids within the box 1-like element by Asn (M12) (Fig. 5A), abolished their transducing activity (Fig. 5 0 ) , consistent with the defectiveness of the partial truncation mutant ID35.
In addition, conversion of hydrophobic amino acids in the box 2-like element by Gln or Asn (M13) (Fig. 5A) caused a loss of the ability to activate the IL-8 gene (Fig. 50). Furthermore, detailed analyses of the box 2-like element revealed that Leu-472 is critical for receptor function (Fig. 5, A and   D). These results demonstrate that the hydrophobic elements within the gp130 homologous segment are important for the receptor function of activating the IL-8 gene.
Receptor Binding Analysis of Mutated IL-1Rs-Receptor binding assays were performed to establish whether mutated receptors were expressed on the cell surface of Jurkat transfectants, but the level of ligand binding was too low to assess the effect of mutations on the cell surface expression in Jurkat cells, possibly because of a low efficiency of DNA transfection. An expression system using CHO cells, which has been used for functional studies of cytokine receptors, allowed the efficiency of expression and the function of mutated muIL-1R on the cell surface to be examined. As shown in Table 1, significant levels of receptor binding sites were detected in CHO cells expressing mutated mouse IL-1Rs. Consistent with the previous observation that the mutant of the cytoplasmic region of mouse IL-1R completely truncated does not affect IL-1 binding affinity (14), no significant difference in binding affinity could be observed between mutants and wild type IL-1R (Table I).
The receptor number of CHO transfectant expressing IL-1R mutants ranged from 0.1 to 1.2 times that of the wild type receptor. To further confirm the results obtained in Jurkat cells, the abilities of each mutated IL-1R to induce the IL-8 gene were also examined using CHO transient expression system. Similar results were obtained regarding the requirement of the Cterminal region and several cytoplasmic segments in the muIL-1R to induce CAT activity in CHO cells, as summarized in Table 1. Partially truncated mutants including ID35-ID38 and those altered in the two hydrophobic regions, M12 and M13, did not induce the IL-8 gene. There was no differences in the relative CAT-inducing activities of the mutants when they were transfected into either Jurkat or CHO cells. Mutant B1, expressing a low level of receptors as shown in Table I, also exhibited transducing activity in CHO cells, supporting our notion that the surface receptor levels on CHO transfectants of each mutants are not limiting factor for transducing the IL-1 signal.

DISCUSSION
In this study, we established an assay system for examining recombinant IL-1R function by transient expression of the mouse IL-1R type I cDNA in the human T-cell line, Jurkat, followed by specific stimulation with an IL-la mutant, 151Y. Under our conditions, no functional receptors were detected on native Jurkat cells, which is consistent with a recent report (38). In addition, our system absolutely negates the involvement of endogenously induced or natively undetectable levels of expressed human IL-1R on Jurkat cells, and will also be useful for functional studies in other types of human cells, including fibroblasts and melanoma cells, which express native IL-1R. The fact that the mouse IL-1 type I receptor is functional in a human cell line suggests that the conserved regions of the cytoplasmic portion between human and mouse IL-1Rs, possessing 78% homology, are sufficient to transduce an IL-1 signal.
In a CHO transient expression system, some truncation or point mutations of the muIL-1R cytoplasmic portion resulted in low levels of receptor expression on the cell surface. However, it has been reported that the receptor level of highly IL-1-responsive cells such as T-lymphoma cell lines including LBRM-33-1A5 and HSB2, are very low (about 200 and 80 receptors/cell, respectively) (39,40), and that occupancy of a few receptor is thought to be enough to trigger the signals (8). In this study, more than 100 receptor binding sites were detected in each CHO transfectant expressing mutated IL-lR, on the assumption of a 100% transfection efficiency. Moreover, the actual number of receptor binding sites in each CHO transfectant may be 2-10 times higher than the values shown in Table I, since other investigators have reported a transfection efficiency of 1040% using DEAE-dextran (41,42). Therefore, it is unlikely that the reduced levels of receptor expression observed in several mutations was responsible for the loss of the ability to induce CAT activity. Similar effects of deletions in the cytoplasmic domain on the expression of the surface receptor have been reported in a functional study of the LFA-1 integrin p subunit (43). Differences in the levels of receptors having various cytoplasmic mutations are probably caused by changes in the turnover rate of receptors, efficiency of trafficking processes, or reduction of the protein biosynthesis rate owing to the instability of the mRNA containing mutated sequences. Permanent cell clones obtained by transfection of the same cDNA gene, and selected during cloning, usually exhibit highly varied biological responses or phenotypes among clones (25), probably due to differences in the integration sites of the gene on the chromosome. Therefore, the transient expression system represents a large population of transfected cells and could be considered superior to the permanent expression system, for evaluating not only signal-transducing activities but also the effects of mutations on cell surface expression of receptors.
To specify the pathway of IL-8 gene activation in Jurkat cells, IL-1-responsive elements of the 5' regulatory region in the IL-8 gene were examined. Truncated or point-mutated analysis of the 5' region revealed that both AP-1 binding motif (-120 to "126) and NFKB site (-80 to "70) are required for IL-8 gene expression in Jurkat cells.' Therefore, we assume that similar cytoplasmic regions of muIL-1R are involved in the induction of other genes whose expression is regulated by the AP-1 and/or the NFKB (44,45). Moreover, we have shown previously that the region of DNA from -94 to -71 bp containing NFKB and NF-IL-6 binding motif-like elements is sufficient for IL-8 gene induction by either IL-1, TNF, or phorbol 12-myristate 13-acetate in human fibrosarcoma 8387 cells (5). Thus, it is suggested that IL-1 induces IL-8 gene expression in the T-cell line, Jurkat, by a mechanism different from that in fibroblasts. The requirement of the AP-1 site is not specific for the Jurkat cells but is also observed in gastric cancer cells, which produce IL-8 in response to T N F (46) or IL-l. 3 Analyses using several deleted mutants of the cytoplasmic portion of muIL-1R revealed that there is a critical point between amino acids 515 and 529 in the C-terminal region of muIL-1R for its function in IL-8 gene activation. Furthermore, we showed that an internal region, consisting of about 100 amino acids (364-474), is also essential for muIL-1R function. Taken collectively, a broader cytoplasmic portion of IL-1R is required for signal transmission compared with other cytokine receptors, such as the p chain of IL-2R and gp130 of IL-6R (24, 25). These results suggest that most of the entire cytoplasmic portion may be necessary for IL-1 receptor function, for example by maintaining conformational changes after ligand binding.
Analysis of our amino acid-substituted muIL-1R, revealed that NLS-like elements located a t 429-433,523-527, and 507-519 were not essential for transducing activity. Moreover, pIR15, lacking most of the C-terminal third NLS-like eiement (529-534), retained the activity to induce IL-8 gene activation. Recently, it has been claimed that human IL-1R mutants with most of the cytoplasmic portion deleted still allowed nuclear localization of IL-1 (47). It remains to be determined K. Kuno, S. Okamoto, N. Mukaida, and K. Matsushima  whether muIL-1R mutants of the NLS-like elements retain the ability to translocate to nuclei in a permanent expression system. Heguy et al. (22) showed that three basic residues, Arg-431, Lys-515, and Arg-518 in the NLS-like elements of human IL-lR, which are conserved in IL-1Rs and the Drosophila Toll protein, are essential for IL-1-mediated signal transduction to induce the IL-2 gene. Our results using mutant murine IL-lR, NL-1, and NL-3, however, indicated that the corresponding basic amino acids, Arg-432, Lys-516, and Arg-519 in muIL-lR, are not required for activation of the IL-8 gene, suggesting that different cytoplasmic elements are required for IL-2 and IL-8 gene activation. Alternatively, it is possible that this situation is caused by a difference in substituted amino acids. Even in the latter, these basic amino acids could be substituted by Ala without affecting their transducing activity, suggesting that basic side chains of these amino acids play little, if any, role in the receptor functions, including interactions with putative second messenger molecule(s) to activate the IL-8 gene.
Phosphorylation of hormone receptors following ligandreceptor interaction is presumed to play a role in regulating The number of receptor binding sites was calculated on the assumption that 100% of CHO cells were transfected and that then expressed the mutated receptors.
receptor signal transduction (48). The phosphorylation of Ser/Thr residues in type I muIL-1R has been suggested in CHO cells soon after exposure to IL-1 (16). On the contrary, Rosoff et al. (49) reported IL-1-dependent diacylglycerol production in Jurkat cells, which might lead to PKC activation, and other groups have reported the IL-1-induced activation of the PKA pathway (50). Thus, we investigated the involvement of several putative phosphorylation sites of PKC and PKA. Our mutational analysis negated the contribution of the putative PKC or PKA acceptor sites, Ser-431, Ser-509, and Ser-528, for IL-1 signal transmission. However, IL-1 stimulates the phosphorylation of several cytosolic proteins through the activation of a PKC/PKA-independent pathway (51)(52)(53). It remains unclear which protein kinase(s) phosphorylates the IL-1R and which Ser/Thr residues in the cytoplasmic portion are actually phosphorylated by the activated Ser/Thr kinase. The physiological relevance of receptor phosphorylation will be clarified by further analyses of the effects of several truncated and mutated receptors on IL-1-induced phosphorylation.
In gp130 of the 1L-6 receptor, the cytoplasmic segment including the box 1 and box 2 elements, which are conserved in the cytokine receptor family, are critical for signal transduction (24). A segment homologous to gp130 is also found at amino acids 435-484 of muIL-1R cytoplasmic portion. Truncation mutants within it and point mutations in the box 1and 2-like elements revealed that the gp130 homologous segment is essential for IL-1 receptor function, suggesting that the common structure among cytokine receptors may play a role in IL-1R function. Since the biological activities of IL-6 differ from those of IL-1, including that of inducing the IL-8 gene (54), gp130 homologous region of IL-1R might contribute to the maintenance of a basic structure of the receptor rather than to the interaction with a putative second messenger molecule(s). Alternatively, box 1and box 2-like elements are also conserved between IL-1R and Drosophila Toll protein.
Thus, it is possible that these hydrophobic elements play a role in functions common between IL-1R and Drosophila Toll protein. It is necessary to determine how this homologous segment is involved in the receptor function.
In this study, by extensive mutational analysis, we revealed that a broad cytoplasmic region, particularly a gp130 homologous segment in the cytoplasmic portion of muIL-lR, is required to activate the IL-8 gene. The properties of the signals that induce the second messenger from the receptor are still unclear. Detailed studies of the effect of a truncated or mutated receptor on either G protein activation or phosphorylation in a permanent expression system, will further clarify the structure and function of the type I IL-1 receptor.