FEBS 11869 Volume 314, number 3, 229-231 8 1992 Federation of European L3iochcmicalSocieties 00145793/92/$5.00
Inhibition
IDecember1992
of passive cutaneous anaphylaxis by synthetic human imrnunoglobulin E peptide fragments
Noriki Nio”, Katsuya Seguro’, Yasuo Ariyoshi”, Katsumi Ishiib and Mideo Nakamarah Tentrul
Research Laborrtmricv. Ajirlonloto Co., 1-I Suzuki~cho, Kmasahi-ku, Kuwusaki 210, Japan and bResearc/l L&oratories, Dainipport Pf~amaceurical Co,. 33-93 hoki, Suira, Osaka 564, Japan
Received 14 September 1992; revised version received 28 October 1992 In an attempt to determine the regions responsible for type 1 immdiate hypcrsensilivily, a tolal of42 peptidc frawcnts, which cover the CH3_CW4 domains in human immunoglobulin E (I~EJ, were clicmicallysynkzkcd. Several peptide fragments located in the amino acid sequencesA@:9Thrs5’ and Arg”9-Ala”3, inhibited passive cutaneous anaphylaxis (PCA) in viva. In order 10 pinpoint the sites responsible for the inhibition of the PCA reaction, various fragment pcplidL%in these Lwo regions were synthcsizcd. As a result, residues Pro”‘-LeuUn, Pr~~~-Tbr’~‘, and Ser’56-‘fhf10’were suggesled 10 be involved in type I immediate hypersensitivity. Immunoglobulin
E; Immunoglobulin
E peptide fragmellt; i’assive cuianeous anaphylaxis; immediate hypersnsitivity;
1. INTRODUCTION Immunoglobulin E (IgE) is a glycoprotein of motecular weight of approximately 190 kDa which binds reversibly and with high afinity to the specific membranebound Fee receptor on mast cells and basophils [I]. Bridging of the receptor-bound IgE by a specific antigen triggers the secretion of chemical mediators such as histamine, slow-reacting substance of anaphylaxis, and platelet activating factor, which are responsible for type I immediate hypersensitivity [2,3]. By binding to and blocking the FCE receptor, peptide containing the binding site(s) will inhibit the release of the chemical mediators. To identify such an IgE receptor-binding peptide would lead to the development of an antiallergic agent for immediate hypersensitivity such as asthma, hay fever and food allergy [4]. It has been suggested by studies with proteolytic peptide fragments of human IgE [S] and its recombinant gene products [6-S], as well as conformational studies [9], that the binding sites in the human IgE molecule are located in the CH3 and/or CH4 domains. The exact binding sites, however, remain unidentified. In an attempt to precisely locate the binding sites, we chemically synthesized a number of peptide fragments which cover the CH3-CH4 domains and tested their
Correspondence address: N. Nio, Ccntrul Research Laboratories, Ajinomoto Co., I-1 Suzuki-ho, Kawasaki-ku, Kawasaki 210, Japan. Fax: (81) (44) 244 9G17. Abbreviarims: These follow the recommendations oflhe IUPAC-iUB Joint Commission on Biochemical Nomenclature 1(19&I) Eur. J. Biothem. 138, 9-391.
PuDlishcd by Elscvier Scietrce Ptlblishrrs B, V.
FCE raptor.
capacity to inhibit the passive cutaneous anaphylaxis (PCA) reaction in vivo. 2. MATERIALS AND METHODS All pcptidcs used were synthesized by &c stepwise solid-phase method using Fmoc chemistry [IO,1 l] as described prc~iously [12,13]. The purity of each pepdde was confirmed by thin-layer chromatogaphy, high-performance liquid chromalography, fan atom bombardment mass spectrometry. sequencing with a gus-phase microsequcncer, and amino acid analysis after hydrolyzing in constant-boiling WCI at 110°C for 24 h. The sequences of compounds are ahown in Table I. 2.2. PCA br rurs The capacity oflhe synlhcsized peplides to inhibit degranulation of masl cells in rats in vivo was measured by the PCA test. Diluted mouse an&es albumin antiserum [14] (100 ~1) was injected imndermally (i.d.) inlo the back skins of 5 male Wistar rats. After 48 h, I ml of a 0.5% Evans’ blue solution conlaining 2 mgof egg albumin was injected intravenously (i.v.). The PCA reaclion was determined as a blueing arcr of lhe skin 30 min after the antigen challenge. To dcterminc the inhibition of the degranulation by the synthciicpepddcs, 3 consecutive i.d. injections of each peptide solution (loO@) were cnrried out inlo lhe same she of the back skin of the rat; 3 and 24. h before the passive sensitization with the andserum. and at the same lime with the antiserum injection. The anti-PCA activity of fragmem peptides was expressed as percent inhibition of the blucing area compared with the vehicle control, inlo which a physiological saline solution was injected three limes. Studenl’s r-lest was used for Lhc statisrical analysis.
3. RESULTS A three-dimensional IgE structure could not be used for designing active peptides, since it has not yet been determined by X-ray crystallography. Therefore, we began with peptides which possibly bind to the FCE receptor, and finally covered the CH3-CH4 domains 229
December 1992
FEBB LETTERS
Volume 314, number 3 Table I
Effect of the synthetic human 1g.Epeptidc fragments on heterologous passive cutaneous anaphylaxis (WA) in rnts Pcptide no. I (329-348)
Compoundb
ADSNPRGVSAYLSRPSPFDL PFDLFIRKSPTI FIRKSPTITSLWDLAPSKGd GTVNLTWSRASGKPV'NHSTR KPVNHSTRKEEKQRNGTLTV LTVTSTLPVGTRDWIEGETY QSRVTHPHLPRALMRSTTKTSd PRALMRSTTKTSGPR GPRAAPEVYAFATPEWPGS PGSRDKRTLASLId SRDKRTLASLIQNFM LIQNFLPEDISVQWLHNEV LENEVQLPDARHSTTQPR LPDARHSTTQPRKTKGSGFF KTKGSGFFVFSRLEVTRAE AEWEQXDEFISRAVHEA ASPSQTVQRAVSVNPGKd ADSNPRGVSAYLSRPSPFDLFIRKSPTIT RVTHPHLPRALMRSTTKTSGPRAAPEVYA FATPEWPGSRDKRTLA 20 (32 l-332) FEDSTXKSADSNd 2 1 (329-338) ADSNPRGVSA 22 (330-334) DSNPR 23 (333-344) PRGVSAYLSRPS 24 (339-348) YLSRPSPFDL PSPFDL 25 (343-348) 26 (41 l-425) IEGETYQSRVTHPHLd RVTHPHLPRALMRST 27 (419433) PRALMRST 28 (426-433) PRALMRSTT 29 (42cd34) 30 (432-438) STTKTSG 31 (434-448) TKTSG?RAAPEVYAF 32 (441-455) AAPEVYAFATPEWPG 33 (c)49-%3) ATPSWPGSYDYRTLA . . EWPGSRDKRT 34 (452-461) WPGSRDKRT 35 (453-461) PGSRDKRT 36 (454-461) GSROKRT 37 (455-461) SRDKRTLA 38 (456-4G3) SRDKRT 39 (456dGl) SRDKR 40 (456-460) RDKRT 41 (457-461) DKRT 42 (458461)
2 (345-356) 3 (34%3G8) 4 (368-387) 5 (380-399) 6 (397-416) 7 (417-437) 8 (42G-440) 9 (438-45G) 10 (454--166) 11 (456-470) 12 (4G5--183) I3 (479-496) 14 (485-504) 15 (497-5 15) 1G(514-530) I7 (53 l-547) IS (329-357) 19 (41%163)
Inhibition (8) 23.2* 15.9” 13.2 8.2 9.3 -6.1 20.0* 33.2* 25.4* 25.3** 25.2** I 1.6c -2.7 5.3 -4.8 -4.8 -2.7 32.4* 32.Y 12.3 2.4 -12.2 13.7 -3.9 17.9* 7.0 -6.2 25.Y 9.5 -3.8 -18.9* 12.1 S.0 13.9 1.8 -5.8 16.gp* 23.0 20.1’* 15.1 23.2 9.4
in parentheses indicate the position in human IgE 191. ‘The single-letter notation for amino acids is used. ‘Each peptidc (100 ,ug/site) was injected intradermally. ‘Cys was replaced with Ser. Verccnt inhibition of PCA at 30 ,udsits i.d. *P =z0.05, **P < 0.01, significnntly different from the matched vehicle control. “Nmbws
which partially overlap each other. The possible binding regions in the CH3 and CH4 domains for the FW rccep+A.. n~.-l;~+nrlby s~fi~nrl~~\~ EE..,I.+,.~PL.-J," I161 opne fnr rruAArrr., "..9a".ba.W W&&Y.". LY. II,PPLI ..I._ p.wH.w.wH hydrophobic [16] and hydrophilic residues [17], the model of the three-dimensional structure of human IgE constructed by Padlan and Davies [lS], and the study 230
Fig. 1. The amino acid sequence of the CM3 and 014 domains oT the human IgE molecule, and the positions of the two regions (1 and 2) with inhibitory activity on the PCA reaction.
of homologous sequences among immunoglobulins. In the synthesis, the cysteinc residues of the disulfide bridges in IgE [19], which are assumed not to be involved in the binding between IgE and its receptor, were replaced by serine residues. Table I summarizes the results of the inhibitory activity of the synthetic peptides on PCA. Compounds 1, 2, 7, 8, 9, 10 and 11 in a series of compounds l-l?, which cover the CH3-CH4 domains, significantly inhibited the PCA reaction (Table I). These peptides were located in the amino acid sequences Ala319-Thr357(Region 1) atid Arb*“‘9-A1a463 (Region 2), as shown in Fig. 1. Therefore, compounds 18 and 19, which correspond to Regions 1 and 2, respectively, were synthesized and they significantly inhibited the PCA reaction (Table I). In addition, compounds 20-42 in these two regions were synthesized in crder to define the sites responsible for the inhibition of PCA. Compounds 25, 28, 37 and 39 inhibited the PCA reaction. Therefore, three sites (Pro343-Leu348, Prod2G-Thr”33, and Ser”3G-Thr’“‘) were suggested to be involved in the PCA reaction. 4. DISCUSSION In the present study. we synthesized a series of 42 peptide fragments of human IgE, which cover the CH3CH4 domains, and assayed them for their capacity to inhibit the PCA reaction in rats. The peptide fragments which inhibited PCA were located in the sequences Ala3’g-Thr357 in the .CH3 domain (Region 1) and Arg“19-Ala~63in the junction between the CM3 and CM4 domains (Region 2). This finding supported the previous idea that the CH3 and/or CH4 domains played an important role in the binding of IgE to its receptor. Recently, Helm et al. prepared a series of overlapping recombinant FCE gene products and suggested that the IgE receptor-binding site on human IgE was located within a sequence of 76 amino acids at the CH21CH3 junction (containing the Gln30’-Arg376 sequence) 1201. Region 1 is included in Helm’s peptide fragment. One of the FCE receptor binding sites would probably be ln.w-A+nA .YIH.eY
:q ..
?*&
:‘-bn 1 s .Y.. I.
In addition, various fragment peptides in these regions were synthesized in order to locate the sites responsible for PCA inhibition. Residues Pro3’3-Leu34M,
Volume 314, number 3
FEES
Pro’26-Thr’33, and SerJSG-Thr”6’were identified as being responsible for PCA inhibition in rats. These three sites in human IgE were not homologous with those in human IgAl, IgD, IgGl and IgM, as determined by sequence homology analysis [21]. Since among human immunoglobulins only human IgE binds to its Fcs receptor with high affinity, the amino acid sequences of the binding sites of IgE should not be homologous with those of other immunoglobulins. On the other hand, these three sites in both human and rodent IgE have a comparatively high degree of homology and the same type of amino acids with regard to charge or hydrophobicity. In general, physiologically important amino acids are conserved. It has also been suggested that human IgE and rodent IgE share common structures in the Fc portion of molecules [22]. The three sites in human IgE were indeed relatively homologous with the corresponding sites in rodent IgE [6,23,24], but not with those in human IgAl, IgD, lgG1 and IgM [21]. In addition, the three active sites were located on the surface of the stereoview model of the FCE portion of human IgE [18], which was constructed using a known threedimensional structure of the IgG Fc site. This result indicates that the Fcs receptor binding site(s) is exposed to the surface of the IgE molecule. From the results, we have tentatively concluded that the three sites (Pro343-Leu3’8,Pro4’6-Thr’33, and Ser*“6Thti6’) in the CM3 and CH4 domains are responsible for type I immediate hypersensitivity, and may also contribute to binding of the IgE molecule to its receptor. Ac~rro~l~f~~~crtnrfz~~~ We thank Dr. K. Hirayama, Dr. S. Akashi and Ms. M. Furuyo for measuring mass spL%.rtra,Mr. S. Ozawa and Mr. T. Scino for amino acid analyses, and Mr. K. Iijima for elemental analyses.
LETTERS
December
1992
REFERENCES [II Ishizaka, T. and Ishizaka, K. (1975) Prog. Allergy 19, 60-121. [2] Segal, D.M., Taurog, J.D. and Mctzger, H. (i977) Proc. Natl. Acad. Sci. USA 74, 2993-2997. [J] Ishizaka, T., Chang. T.H., Taggart. M. and Ishizaka, K. (1977) J. Immunol. 119. 1589-1596. I41 Blank, U., Ra, d., Miller, L.; White. K., Metzger, H. and Kinct. J.-P. (1989) Nature 337. 187-189. PI Stanworth, D.R. (1971) Nature 233, 310-316. El Liu, F.-T., Alhrandt, K.A.. Bry, C.G. and Ishizaka. T. (1984) Proc. Natl. Acad. Sci. USA 81, 5369-5373. [71 Guha, R.S., Helm, B. and Gould, H. (1985) Nature 315,577-578. PI Coleman, J.W., Helm. B., Stanworth, D.R. and Gould, H.J. (1985) Eur. J. Immunol. IS, 966-969. Y91Dorrington, K.J. and Bcnnick, H.H. (1978) Immunol. Rev. 41, 3-25. DOI Carpino, L.A. and Han, G.Y. (I 970) 1. Am. Chem. Sot. 92.5748. iI11 Fields, G.B. and Noble, R.L. (1990) Int. J. Pqidc Pralein Rcs. 35, 161-214. (For a review.) 1121 Kohmura, M., Nio, N., Kubo, K., Minoshima, Y., Munekata, E. and Ariyoshi, Y. (1989) Agric. Dial. Chcm. S3.2107-2114. 1131 Kohmura, M., Nio, N. and Ariyoshi, Y. (1990) Agric. Biol. Chcm. 54, 1521-1530. [14] Levine, B.B. and Vaz, N.M. (1970) Inl. Arch. Allergy Appl. Immunol. 39, 156-l 7 1, [I 51 Chou, P.Y. and Fasman, G.P. (1978) Adv. Enymol. 47,45-148. [16] Rose. G.D. and Roy, S. (1980) Droc. Natl. Acad. Sci. USA 77, 4643-4647. [17] Hopp, T.P. and Woods, K.R. (1981) Proc. NatI. Aced. 5ci. USA 78, 3824-3828. [18] Padlan, E.A. and Davies, DR. (1986) Mol. lmmunol. 23,1063-
1075. [I91 Bahr-Lindstrom, H. and Bennich, I-1.(1974) FEBS Lat. 40, 57Gl. PO1 Helm, B., Marsh, P., Vcrcelli, D., Padlan. E., Gould, H. and Geha, R. (1988) Nature 331, 180-183. WI Go@, N.M. and Bernard, 0. (1981) Proc. Notl. Acad. Sci. USA 78, 509-513. WI Conrad, D.H.. Wingard, J.R. and Ishlzaka. T. (1983) J. Immunol. 130,327-333. u31 Ishida. N., Ucda. S.. Hayashida, 1-I.. Miyau, T. and Honjo, T. (1982) EMBO J. I, 1117-1123. 1241Hellman, L., Pettcrson, U., Engstrom, A., Karlson, T. and Bcnnich, H. (1982) Nucleic Acids Res. IO. 6041-6049.
231
Inhibition
IDecember1992
of passive cutaneous anaphylaxis by synthetic human imrnunoglobulin E peptide fragments
Noriki Nio”, Katsuya Seguro’, Yasuo Ariyoshi”, Katsumi Ishiib and Mideo Nakamarah Tentrul
Research Laborrtmricv. Ajirlonloto Co., 1-I Suzuki~cho, Kmasahi-ku, Kuwusaki 210, Japan and bResearc/l L&oratories, Dainipport Pf~amaceurical Co,. 33-93 hoki, Suira, Osaka 564, Japan
Received 14 September 1992; revised version received 28 October 1992 In an attempt to determine the regions responsible for type 1 immdiate hypcrsensilivily, a tolal of42 peptidc frawcnts, which cover the CH3_CW4 domains in human immunoglobulin E (I~EJ, were clicmicallysynkzkcd. Several peptide fragments located in the amino acid sequencesA@:9Thrs5’ and Arg”9-Ala”3, inhibited passive cutaneous anaphylaxis (PCA) in viva. In order 10 pinpoint the sites responsible for the inhibition of the PCA reaction, various fragment pcplidL%in these Lwo regions were synthcsizcd. As a result, residues Pro”‘-LeuUn, Pr~~~-Tbr’~‘, and Ser’56-‘fhf10’were suggesled 10 be involved in type I immediate hypersensitivity. Immunoglobulin
E; Immunoglobulin
E peptide fragmellt; i’assive cuianeous anaphylaxis; immediate hypersnsitivity;
1. INTRODUCTION Immunoglobulin E (IgE) is a glycoprotein of motecular weight of approximately 190 kDa which binds reversibly and with high afinity to the specific membranebound Fee receptor on mast cells and basophils [I]. Bridging of the receptor-bound IgE by a specific antigen triggers the secretion of chemical mediators such as histamine, slow-reacting substance of anaphylaxis, and platelet activating factor, which are responsible for type I immediate hypersensitivity [2,3]. By binding to and blocking the FCE receptor, peptide containing the binding site(s) will inhibit the release of the chemical mediators. To identify such an IgE receptor-binding peptide would lead to the development of an antiallergic agent for immediate hypersensitivity such as asthma, hay fever and food allergy [4]. It has been suggested by studies with proteolytic peptide fragments of human IgE [S] and its recombinant gene products [6-S], as well as conformational studies [9], that the binding sites in the human IgE molecule are located in the CH3 and/or CH4 domains. The exact binding sites, however, remain unidentified. In an attempt to precisely locate the binding sites, we chemically synthesized a number of peptide fragments which cover the CH3-CH4 domains and tested their
Correspondence address: N. Nio, Ccntrul Research Laboratories, Ajinomoto Co., I-1 Suzuki-ho, Kawasaki-ku, Kawasaki 210, Japan. Fax: (81) (44) 244 9G17. Abbreviarims: These follow the recommendations oflhe IUPAC-iUB Joint Commission on Biochemical Nomenclature 1(19&I) Eur. J. Biothem. 138, 9-391.
PuDlishcd by Elscvier Scietrce Ptlblishrrs B, V.
FCE raptor.
capacity to inhibit the passive cutaneous anaphylaxis (PCA) reaction in vivo. 2. MATERIALS AND METHODS All pcptidcs used were synthesized by &c stepwise solid-phase method using Fmoc chemistry [IO,1 l] as described prc~iously [12,13]. The purity of each pepdde was confirmed by thin-layer chromatogaphy, high-performance liquid chromalography, fan atom bombardment mass spectrometry. sequencing with a gus-phase microsequcncer, and amino acid analysis after hydrolyzing in constant-boiling WCI at 110°C for 24 h. The sequences of compounds are ahown in Table I. 2.2. PCA br rurs The capacity oflhe synlhcsized peplides to inhibit degranulation of masl cells in rats in vivo was measured by the PCA test. Diluted mouse an&es albumin antiserum [14] (100 ~1) was injected imndermally (i.d.) inlo the back skins of 5 male Wistar rats. After 48 h, I ml of a 0.5% Evans’ blue solution conlaining 2 mgof egg albumin was injected intravenously (i.v.). The PCA reaclion was determined as a blueing arcr of lhe skin 30 min after the antigen challenge. To dcterminc the inhibition of the degranulation by the synthciicpepddcs, 3 consecutive i.d. injections of each peptide solution (loO@) were cnrried out inlo lhe same she of the back skin of the rat; 3 and 24. h before the passive sensitization with the andserum. and at the same lime with the antiserum injection. The anti-PCA activity of fragmem peptides was expressed as percent inhibition of the blucing area compared with the vehicle control, inlo which a physiological saline solution was injected three limes. Studenl’s r-lest was used for Lhc statisrical analysis.
3. RESULTS A three-dimensional IgE structure could not be used for designing active peptides, since it has not yet been determined by X-ray crystallography. Therefore, we began with peptides which possibly bind to the FCE receptor, and finally covered the CH3-CH4 domains 229
December 1992
FEBB LETTERS
Volume 314, number 3 Table I
Effect of the synthetic human 1g.Epeptidc fragments on heterologous passive cutaneous anaphylaxis (WA) in rnts Pcptide no. I (329-348)
Compoundb
ADSNPRGVSAYLSRPSPFDL PFDLFIRKSPTI FIRKSPTITSLWDLAPSKGd GTVNLTWSRASGKPV'NHSTR KPVNHSTRKEEKQRNGTLTV LTVTSTLPVGTRDWIEGETY QSRVTHPHLPRALMRSTTKTSd PRALMRSTTKTSGPR GPRAAPEVYAFATPEWPGS PGSRDKRTLASLId SRDKRTLASLIQNFM LIQNFLPEDISVQWLHNEV LENEVQLPDARHSTTQPR LPDARHSTTQPRKTKGSGFF KTKGSGFFVFSRLEVTRAE AEWEQXDEFISRAVHEA ASPSQTVQRAVSVNPGKd ADSNPRGVSAYLSRPSPFDLFIRKSPTIT RVTHPHLPRALMRSTTKTSGPRAAPEVYA FATPEWPGSRDKRTLA 20 (32 l-332) FEDSTXKSADSNd 2 1 (329-338) ADSNPRGVSA 22 (330-334) DSNPR 23 (333-344) PRGVSAYLSRPS 24 (339-348) YLSRPSPFDL PSPFDL 25 (343-348) 26 (41 l-425) IEGETYQSRVTHPHLd RVTHPHLPRALMRST 27 (419433) PRALMRST 28 (426-433) PRALMRSTT 29 (42cd34) 30 (432-438) STTKTSG 31 (434-448) TKTSG?RAAPEVYAF 32 (441-455) AAPEVYAFATPEWPG 33 (c)49-%3) ATPSWPGSYDYRTLA . . EWPGSRDKRT 34 (452-461) WPGSRDKRT 35 (453-461) PGSRDKRT 36 (454-461) GSROKRT 37 (455-461) SRDKRTLA 38 (456-4G3) SRDKRT 39 (456dGl) SRDKR 40 (456-460) RDKRT 41 (457-461) DKRT 42 (458461)
2 (345-356) 3 (34%3G8) 4 (368-387) 5 (380-399) 6 (397-416) 7 (417-437) 8 (42G-440) 9 (438-45G) 10 (454--166) 11 (456-470) 12 (4G5--183) I3 (479-496) 14 (485-504) 15 (497-5 15) 1G(514-530) I7 (53 l-547) IS (329-357) 19 (41%163)
Inhibition (8) 23.2* 15.9” 13.2 8.2 9.3 -6.1 20.0* 33.2* 25.4* 25.3** 25.2** I 1.6c -2.7 5.3 -4.8 -4.8 -2.7 32.4* 32.Y 12.3 2.4 -12.2 13.7 -3.9 17.9* 7.0 -6.2 25.Y 9.5 -3.8 -18.9* 12.1 S.0 13.9 1.8 -5.8 16.gp* 23.0 20.1’* 15.1 23.2 9.4
in parentheses indicate the position in human IgE 191. ‘The single-letter notation for amino acids is used. ‘Each peptidc (100 ,ug/site) was injected intradermally. ‘Cys was replaced with Ser. Verccnt inhibition of PCA at 30 ,udsits i.d. *P =z0.05, **P < 0.01, significnntly different from the matched vehicle control. “Nmbws
which partially overlap each other. The possible binding regions in the CH3 and CH4 domains for the FW rccep+A.. n~.-l;~+nrlby s~fi~nrl~~\~ EE..,I.+,.~PL.-J," I161 opne fnr rruAArrr., "..9a".ba.W W&&Y.". LY. II,PPLI ..I._ p.wH.w.wH hydrophobic [16] and hydrophilic residues [17], the model of the three-dimensional structure of human IgE constructed by Padlan and Davies [lS], and the study 230
Fig. 1. The amino acid sequence of the CM3 and 014 domains oT the human IgE molecule, and the positions of the two regions (1 and 2) with inhibitory activity on the PCA reaction.
of homologous sequences among immunoglobulins. In the synthesis, the cysteinc residues of the disulfide bridges in IgE [19], which are assumed not to be involved in the binding between IgE and its receptor, were replaced by serine residues. Table I summarizes the results of the inhibitory activity of the synthetic peptides on PCA. Compounds 1, 2, 7, 8, 9, 10 and 11 in a series of compounds l-l?, which cover the CH3-CH4 domains, significantly inhibited the PCA reaction (Table I). These peptides were located in the amino acid sequences Ala319-Thr357(Region 1) atid Arb*“‘9-A1a463 (Region 2), as shown in Fig. 1. Therefore, compounds 18 and 19, which correspond to Regions 1 and 2, respectively, were synthesized and they significantly inhibited the PCA reaction (Table I). In addition, compounds 20-42 in these two regions were synthesized in crder to define the sites responsible for the inhibition of PCA. Compounds 25, 28, 37 and 39 inhibited the PCA reaction. Therefore, three sites (Pro343-Leu348, Prod2G-Thr”33, and Ser”3G-Thr’“‘) were suggested to be involved in the PCA reaction. 4. DISCUSSION In the present study. we synthesized a series of 42 peptide fragments of human IgE, which cover the CH3CH4 domains, and assayed them for their capacity to inhibit the PCA reaction in rats. The peptide fragments which inhibited PCA were located in the sequences Ala3’g-Thr357 in the .CH3 domain (Region 1) and Arg“19-Ala~63in the junction between the CM3 and CM4 domains (Region 2). This finding supported the previous idea that the CH3 and/or CH4 domains played an important role in the binding of IgE to its receptor. Recently, Helm et al. prepared a series of overlapping recombinant FCE gene products and suggested that the IgE receptor-binding site on human IgE was located within a sequence of 76 amino acids at the CH21CH3 junction (containing the Gln30’-Arg376 sequence) 1201. Region 1 is included in Helm’s peptide fragment. One of the FCE receptor binding sites would probably be ln.w-A+nA .YIH.eY
:q ..
?*&
:‘-bn 1 s .Y.. I.
In addition, various fragment peptides in these regions were synthesized in order to locate the sites responsible for PCA inhibition. Residues Pro3’3-Leu34M,
Volume 314, number 3
FEES
Pro’26-Thr’33, and SerJSG-Thr”6’were identified as being responsible for PCA inhibition in rats. These three sites in human IgE were not homologous with those in human IgAl, IgD, IgGl and IgM, as determined by sequence homology analysis [21]. Since among human immunoglobulins only human IgE binds to its Fcs receptor with high affinity, the amino acid sequences of the binding sites of IgE should not be homologous with those of other immunoglobulins. On the other hand, these three sites in both human and rodent IgE have a comparatively high degree of homology and the same type of amino acids with regard to charge or hydrophobicity. In general, physiologically important amino acids are conserved. It has also been suggested that human IgE and rodent IgE share common structures in the Fc portion of molecules [22]. The three sites in human IgE were indeed relatively homologous with the corresponding sites in rodent IgE [6,23,24], but not with those in human IgAl, IgD, lgG1 and IgM [21]. In addition, the three active sites were located on the surface of the stereoview model of the FCE portion of human IgE [18], which was constructed using a known threedimensional structure of the IgG Fc site. This result indicates that the Fcs receptor binding site(s) is exposed to the surface of the IgE molecule. From the results, we have tentatively concluded that the three sites (Pro343-Leu3’8,Pro4’6-Thr’33, and Ser*“6Thti6’) in the CM3 and CH4 domains are responsible for type I immediate hypersensitivity, and may also contribute to binding of the IgE molecule to its receptor. Ac~rro~l~f~~~crtnrfz~~~ We thank Dr. K. Hirayama, Dr. S. Akashi and Ms. M. Furuyo for measuring mass spL%.rtra,Mr. S. Ozawa and Mr. T. Scino for amino acid analyses, and Mr. K. Iijima for elemental analyses.
LETTERS
December
1992
REFERENCES [II Ishizaka, T. and Ishizaka, K. (1975) Prog. Allergy 19, 60-121. [2] Segal, D.M., Taurog, J.D. and Mctzger, H. (i977) Proc. Natl. Acad. Sci. USA 74, 2993-2997. [J] Ishizaka, T., Chang. T.H., Taggart. M. and Ishizaka, K. (1977) J. Immunol. 119. 1589-1596. I41 Blank, U., Ra, d., Miller, L.; White. K., Metzger, H. and Kinct. J.-P. (1989) Nature 337. 187-189. PI Stanworth, D.R. (1971) Nature 233, 310-316. El Liu, F.-T., Alhrandt, K.A.. Bry, C.G. and Ishizaka. T. (1984) Proc. Natl. Acad. Sci. USA 81, 5369-5373. [71 Guha, R.S., Helm, B. and Gould, H. (1985) Nature 315,577-578. PI Coleman, J.W., Helm. B., Stanworth, D.R. and Gould, H.J. (1985) Eur. J. Immunol. IS, 966-969. Y91Dorrington, K.J. and Bcnnick, H.H. (1978) Immunol. Rev. 41, 3-25. DOI Carpino, L.A. and Han, G.Y. (I 970) 1. Am. Chem. Sot. 92.5748. iI11 Fields, G.B. and Noble, R.L. (1990) Int. J. Pqidc Pralein Rcs. 35, 161-214. (For a review.) 1121 Kohmura, M., Nio, N., Kubo, K., Minoshima, Y., Munekata, E. and Ariyoshi, Y. (1989) Agric. Dial. Chcm. S3.2107-2114. 1131 Kohmura, M., Nio, N. and Ariyoshi, Y. (1990) Agric. Biol. Chcm. 54, 1521-1530. [14] Levine, B.B. and Vaz, N.M. (1970) Inl. Arch. Allergy Appl. Immunol. 39, 156-l 7 1, [I 51 Chou, P.Y. and Fasman, G.P. (1978) Adv. Enymol. 47,45-148. [16] Rose. G.D. and Roy, S. (1980) Droc. Natl. Acad. Sci. USA 77, 4643-4647. [17] Hopp, T.P. and Woods, K.R. (1981) Proc. NatI. Aced. 5ci. USA 78, 3824-3828. [18] Padlan, E.A. and Davies, DR. (1986) Mol. lmmunol. 23,1063-
1075. [I91 Bahr-Lindstrom, H. and Bennich, I-1.(1974) FEBS Lat. 40, 57Gl. PO1 Helm, B., Marsh, P., Vcrcelli, D., Padlan. E., Gould, H. and Geha, R. (1988) Nature 331, 180-183. WI Go@, N.M. and Bernard, 0. (1981) Proc. Notl. Acad. Sci. USA 78, 509-513. WI Conrad, D.H.. Wingard, J.R. and Ishlzaka. T. (1983) J. Immunol. 130,327-333. u31 Ishida. N., Ucda. S.. Hayashida, 1-I.. Miyau, T. and Honjo, T. (1982) EMBO J. I, 1117-1123. 1241Hellman, L., Pettcrson, U., Engstrom, A., Karlson, T. and Bcnnich, H. (1982) Nucleic Acids Res. IO. 6041-6049.
231
