- Email: [email protected]
Domain mapping and comparative binding features of eight dog IgE-specific reagents in ELISA, immunoblots, and immunohistochemistry
Veterinary Immunology and Immunopathology 70 (1999) 117±124
Domain mapping and comparative binding features of eight dog IgE-specific reagents in ELISA, immunoblots, and immunohistochemistry M.E. Griot-Wenka,b,*, E. Martia, D.J. DeBoerc, A.L. de Weckd, S. Lazarya a
Division of Immunogenetics, Institute of Animal Breeding, Bremgarten-Strasse 109a, Switzerland b Small Animal Clinics, LaÈnggass-Strasse 128; University of Bern, 3012 Bern, Switzerland c Department of Medical Sciences, School of Veterinary Medicine, University of Wisconsin, 2015 Linden Drive West, Madison, WI 53706, USA d CMG-Heska Allergy Products, 16 Grand'Places, 1700 Fribourg, Switzerland Received 31 December 1998; received in revised form 16 April 1999; accepted 5 May 1999
Abstract Eight dog IgE-specific reagents including monoclonal and polyclonal antibodies (Ab) and a cross-reactive alpha chain of the human high affinity IgE receptor were mapped to recombinant fragments of the second (IgEf2) and third/fourth (IgEf3/4) domains of the dog IgE heavy chain. In ELISA, five out of eight reagents reacted to solid-phase bound IgEf2, of which two polyclonal Ab bound in addition to IgEf3/4. All Ab which recognized at least one recombinant IgE fragment, also bound to IgE in ELISA, immunoblots, and immunohistochemistry. In contrast, only one monoclonal Ab, that did not bind to the recombinant IgE fragments, reacted with immunoblots of serum and immunohistochemistry. The alpha chain could only be applied to ELISA with serum IgE. Furthermore, there was a wide range of heat-lability of binding reactions. Comparative analysis of available dog IgE-specific reagents enables more in-depth functional studies on IgE-mediated phenomena in dogs, and helps to further establish the dog as an animal model for allergy research. # 1999 Elsevier Science B.V. All rights reserved. Keywords: Immunoglobulin E; Domain; Recombinant IgE; Dog
* Corresponding author. Present address: Novartis Nutrition Research AG, Fabrikstrasse 10, 3176 Neuenegg, Switzerland; Tel.: +41-31-377-24-85; fax: +41-31-377-24-34 E-mail address: [email protected] (M.E. Griot-Wenk)
0165-2427/99/$ ± see front matter # 1999 Elsevier Science B.V. All rights reserved. PII: S 0 1 6 5 - 2 4 2 7 ( 9 9 ) 0 0 0 7 0 - 7
118
M.E. Griot-Wenk et al. / Veterinary Immunology and Immunopathology 70 (1999) 117±124
1. Introduction Atopic conditions are common in dogs, out of which atopic dermatitis has been by far best characterized (Scott et al., 1995). For in-depth investigations of these presumed immunoglobuin E- (IgE) mediated immune reactions in the dog, a diversity of well characterized dog IgE-specific reagents are needed. Although dog IgE has been described more than two decades ago (Schwartzman et al., 1971; Halliwell et al., 1972, 1975), it was not until recently that these necessary tools have been developed, namely dog IgEspecific murine monoclonal antibodies (mAb; DeBoer et al., 1993; Peng et al., 1993a; Groeben et al., 1997; de Weck et al., 1998), cross-reactive recombinant alpha chain of the human high affinity IgE receptor (Wassom and Grieve, 1998), and several IgE-specific polyclonal Ab (Peng et al., 1993a, b; Hammerberg et al., 1997; Griot-Wenk et al., 1998). Furthermore, sequence analysis of dog IgE (Patel et al., 1995) enabled the production of recombinant dog IgE fragments (Griot-Wenk et al., 1998), which allows domain mapping of IgE-specific reagents. The objectives of the present study were to map eight dog IgE-specific reagents to recombinant dog IgE domains and to compare binding characteristics of these reagents for use in enzyme-linked immunosorbent assay (ELISA), immunoblots, and immunohistochemistry. 2. Materials and methods 2.1. Dog IgE-specific reagents The following reagents were applied in this study: polyclonal Ab raised in hens against the recombinant canine IgE fragments of the second (IgEf2) and third/fourth (IgEf3/4) constant domains (Ab1-2; Griot-Wenk et al., 1998) and in goat against purified dog serum IgE (Ab3; Bethyl Laboratories, Montgomery TX, USA), four mAb raised against purified dog or human serum IgE (mAb4-7; DeBoer et al., 1993; de Weck et al., 1998), and a cross-reactive, biotinylated recombinant alpha chain of the human IgE receptor (R8, Wassom and Grieve, 1998). 2.2. ELISA Capture ELISAs for domain mapping of IgE-specific reagents and for detection of peanut-specific serum IgE were applied as described by us previously (Griot-Wenk et al., 1998). Briefly, for domain mapping, the plates were coated with IgEf2 or IgEf3/4 (10 ng/ well), blocked and then subjected to the IgE-specific reagents (0.7±1.8 mg reagent/well). For visualization of binding the following corresponding alkaline phosphatase-(AP) labeled reagents were used: goat anti-chicken IgG Ab (1 : 1000; Kirkegaard and Perry Laboratories, Gaithersburg, MA, USA), goat anti-mouse IgG Ab (1 : 2000; Jackson ImmunoResearch Laboratories, West Grove, PA, USA), rabbit anti-goat IgG Ab
M.E. Griot-Wenk et al. / Veterinary Immunology and Immunopathology 70 (1999) 117±124
119
(1 : 2000; Sigma, St. Louis, MO, USA), and extravidin (1 : 20,000; Sigma). ELISA results were expressed as mean optical density of duplicate wells after subtraction of background controls (coated antigen without primary developing Ab) measured 30 min after adding enzyme substrates at a wavelength of 405 nm (OD405). Reactivity was defined as an OD405 of at least three times above background (OD405 of background 0.2). It was further subjectively designated as absent (0, OD405 < 0.6), weak (1+, OD405 0.6 and <2.0), or strong (2+, OD405 2.0). For detection of peanut-specific serum IgE, peanut-coated plates (100 ng/well) were incubated with serum (1 : 50 to 1 : 800) and developed as described for domain mapping (0.2 mg IgE-specific reagents/well). Additionally, results obtained with peanut-specific serum were compared to the ones achieved with serum heated at 568C, in which samples for analysis were taken at time points from 15 min to 4 h before application onto plates. 2.3. Immunoblotting Ammonium sulfate precipitated dog serum and IgEf3/4-specific Ab affinity-enriched dog serum fraction were analyzed by 5 and 10% PAGE (100 g/preparative gel) run under native or denaturing-reducing conditions (Griot-Wenk et al., 1998). The immunoblots were analyzed with the IgE-specific (2±10 mg reagent/ml) and secondary developing reagents as described for ELISA. Binding was visualized using 5-bromo-4-chloroindolyl-phosphate/4-nitro blue tetrazolium chloride (Boehringer Mannheim GmbH, Mannheim, Germany). Negative controls included application of secondary developing Ab alone. Immunoblot results were graded subjectively as follows: no band visible (0), visible but faint reaction (1+), and strong binding reaction (2+). 2.4. Immunohistochemistry All reagents were applied to histologic sections of a mesenteric lymph node from a dog as described (Griot-Wenk et al., 1998). In short, deparaffinized tissue sections were trypsin-treated and blocked. The IgE-specific reagents were applied at (6 mg/section) followed by incubation with the corresponding developing Ab: biotinylated rat antimouse IgG Ab (1 : 1000; Jackson Immuno Research Laboratories), AP-labeled rabbit anti-goat IgG Ab (1 : 2000; Sigma), or AP-labeled goat anti-chicken IgG Ab (1 : 1000; Kirkegaard and Perry Laboratories). The endogenous peroxidase and phosphatase were blocked with 1% hydrogen peroxide in methanol, respectively, by addition of levamisol to the secondary developing Ab (1 mg/ml, Sigma). To detect immunoreactivity of the biotinylated mAb the Vectastain1 ABC-PO kit was utilized according to the manufacturer's instructions (Vector Lab., Burlingame, CA, USA). Enzyme activity was detected with diaminobenzidine (Sigma) and hydrogen peroxide. Immunoreactivity of the IgE-specific goat and chicken Ab was detected using naphtol-AS-Mx-phosphate (Sigma) and fast red (Chroma Gesellschaft Schmid GmbH + Co., KoÈngen, Germany). As negative control all secondary developing Ab were applied alone, omitting IgE-specific reagents.
120
M.E. Griot-Wenk et al. / Veterinary Immunology and Immunopathology 70 (1999) 117±124
3. Results 3.1. Reactivity with IgEf2 and IgEf3/4 The two recombinant dog IgE fragments rendered it possible to investigate affinity of IgE-specific reagents to the second and third/fourth domains of dog IgE. In the ELISA applied, five out of eight reagents reacted to solid-phase bound IgEf2 and two out of eight with IgEf3/4, respectively (background OD405 0.1, Table 1). Antibody 3 and mAb6 reacted with high affinity to dog IgEf2; mAb4 also recognized IgEf2 but to a lesser extent. Moreover, Ab3 bound to IgEf3/4. As expected, Ab1 and Ab2, which had been raised against the recombinant IgE fragments, displayed high affinity to the corresponding immunogen and IgEf3/4-specific chicken Ab cross-reacted against IgEf2 (Griot-Wenk et al., 1998). 3.2. Affinity to peanut±specific IgE in ELISA and heat-lability of binding reactions The IgE-specific reagents were applied in an ELISA with solid-phase bound peanut as antigen followed by incubation with peanut-positive dog serum (heat-treated versus nonheated and serum dilution series) to compare affinity and heat-lability of the identified IgE-epitopes. In our test system, a wide range of affinity of the dog IgE-specific reagents to peanut-specific IgE was found. All but two mAb displayed positive reactions (Table 1). The OD405 ranged from 1.1 for Ab1 to 3.5 for Ab3. Different affinity to peanut-specific IgE was further reflected by serum dilution: Ab3, mAb4, and mAb6 resulted in an OD405 of at least 0.9 even at a serum dilution of 1 : 400. Comparative analysis of binding to serum that had been heated for up to 4 h resulted in at least 50% decrease in ELISA reactivity compared with non-heated serum in all reagents studied (Fig. 1). There was a wide range of heat sensitivity in the recognized epitopes. Increased binding after brief heat treatment of dog serum was evident only in Ab2; more prolonged heat treatment then resulted in diminished binding too. The most Table 1 Summary of binding features of eight dog IgE-specific antibodies (Ab)/reagent (R) available for this study in ELISA, immunoblots, and immunohistochemistry. Reactivity was subjectively designated as strong (2+), weak (1+), or absent (0) Designation
Reactivity in ELISA against a
Ab1 Ab2 Ab3 mAb4 mAb5 mAb6 mAb7 R8 a
a
Immunoblotting of serum
IgEf2
IgEf3/4
Serum IgE
Native
Denatured
2+ 2+ 2+ 1+ 0 2+ 0 0
0 2+ 2+ 0 0 0 0 0
1+ 2+ 2+ 2+ 0 2+ 0 2+
1+ 2+ 1+ 2+ 1+ 2+ 0 0
2+ 2+ 2+ 1+ 0 2+ 0 0
Immunohistochemistry
1+ 2+ 2+ 2+ 1+ 2+ 0 0
Recombinant canine IgE fragments of the second (IgEf2) and third/fourth (IgEf3/4) constant domains.
M.E. Griot-Wenk et al. / Veterinary Immunology and Immunopathology 70 (1999) 117±124
Fig. 1. Variation of heat-lability of recognized epitope of six dog IgE-specific reagents in an ELISA. Peanut-specific IgE-positive dog serum was diluted 1 : 50. Data are expressed as percentage of mean of duplicate OD405 values after subtraction of buffer controls obtained in heated serum as compared to non-heated serum.
121
122
M.E. Griot-Wenk et al. / Veterinary Immunology and Immunopathology 70 (1999) 117±124
heat-labile epitope was recognized by R8: heating serum for merely 15 min at 568C almost abolished binding (97% reduction compared to untreated serum). 3.3. Immunoblotting of native and denatured dog serum In immunoblots of native dog serum, all but mAb7 and the R8 bound to a protein of approximately 180 kDa (Table 1), the expected size of dog IgE (Griot-Wenk et al., 1998). Except for mAb5, binding to immunoblots of native serum corresponded to results obtained with immunoblots run under denaturing conditions: five out of six Ab recognized a 70 kDa-protein in enatured serum (expected size of heavy chain of IgE; Griot-Wenk et al., 1998). Semiquantitative assessment of strength of binding revealed that Ab1 and Ab3 reacted more strongly with the denatured than the native protein. 3.4. Immunohistochemistry IgE-bearing cells could be detected by our assay system in a mesenteric lymph node of a dog with all reagents except for mAb7 and R8 (Table 1). Cells were stained by the reagents with differing intensities. The IgE-bearing cells were predominantly located in the medulla and paracortical area. 4. Discussion This study describes domain mapping of eight different dog IgE-specific reagents to recombinant dog IgE heavy chain fragments, IgEf2 and IgEf3/4 (Griot-Wenk et al., 1998), in ELISA in comparison to their binding features to natural IgE in ELISA, immunoblots, and immunohistochemistry. Five out of eight reagents tested reacted with IgEf2 and only two with IgEf3/4. Being a polyclonal Ab, it is not surprising that Ab3 raised in goats recognized both recombinant IgE fragments. However, it cannot be elucidated whether this broad reactivity reflects shared or multiple epitopes as these Ab were raised against the whole IgE protein. In contrast, shared epitopes on IgEf2 and IgEf3/4 is indicated by the cross-reactivity seen with Ab2, which had been produced using IgEf3/4 (Griot-Wenk et al., 1998). The mAb4 and mAb6 which both recognize IgEf2 must be directed against an epitope only present on IgEf2 as they do not cross-react with IgEf3/4. Diversity of identified epitopes by IgE-specific reagents is further evident by the wide range of heat-sensitivity of binding (Fig. 1). There was only one Ab, namely Ab2, which must be directed against a rather heat resistant epitope(s) as more IgE could be detected at time points 15 and 30 min after heat-treatment of the serum. This Ab could be used to study immune complexes when disrupted by heat. The increase in IgE-specific IgG post disruption of immune complexes has been reported to be most profound in dogs affected with atopic dermatitis and demodectic acariasis (Hammerberg et al., 1997). Interestingly, the cross-reacting recombinant alpha chain of the human IgE receptor (R8) binds to an extremely heat-labile epitope; within 15 min, binding was almost completely abolished. Heat-lability is commonly considered a hallmark sign of an IgE-specific binding reaction,
M.E. Griot-Wenk et al. / Veterinary Immunology and Immunopathology 70 (1999) 117±124
123
and was first described for the human IgE molecule (Ishizaka et al., 1967). A variety of heat-sensitive epitopes have thus far been described for the dog (Schwartzman et al., 1971; DeBoer et al., 1993; Peng et al., 1993a; Hammerberg et al., 1997; Griot-Wenk et al., 1998). Alternatively, the variety in heat sensitivity may further support the existence of polymorphism of IgE in dogs as proposed by DeBoer et al. (1993) and recently supported by experimental data (Peng et al., 1997; Halliwell et al., 1998). Affinity is another defining feature of Ab, which may, however, be affected by the method of detection. The IgE-specific reagents studied here displayed a great variation in affinity in the ELISA utilized to detect peanut-specific IgE. Interestingly, all Ab but mAb7 and R8 resulted in IgE-detection in immunoblotting of serum run under native conditions and immunohistochemistry of paraffin-embedded lymph node tissue (Table 1). The latter observation is striking as all Ab have thus far been described for the usage on frozen sections only except for Ab2 (Griot-Wenk et al., 1998). Binding of mAb5 to native IgE in immunoblot and immunohistochemistry but not when it is bound to peanut antigen suggests that the epitope it recognizes is a conformational one that is changed when IgE `bends' as it binds to antigen. Immunoglobulin E binds to the alpha chain of the high-affinity IgE-receptor, Fc"RI. The actual binding site of IgE has been mapped to the first 12 amino acids of C"3 (reviewed by Sutton and Gould, 1993). The recombinant dog IgE fragment, IgEf3/4, however, does not include this first part of C"3 (Griot-Wenk et al., 1998), which probably explains lack of binding in ELISA against IgEf3/4 (Table 1). The alpha chain herein employed was constructed based on the human DNA sequence (Wassom and Grieve, 1998). The predicted gene product of the conserved C"3 constant region of dog IgE shares 62% amino acid identity with the corresponding human part potentially allowing cross-reactivity (Patel et al., 1995). In conclusion, these studies illustrate for the first time epitope mapping of dog IgEspecific reagents applying recombinant dog IgE fragments combined with data obtained from binding to native and denatured whole IgE. The diversity in binding features of the reagents supports a variety in epitopes on IgE or polymorphism of IgE. These in vitro observations may help to elucidate differences in clinical expression of atopic syndromes in individuals and at various body sites, which will further contribute to establishing the dog as an animal model for allergy. Acknowledgements We thank Corinne MuÈller (Institute of Animal Breeding) for technical assistance. This work was supported by the Swiss National Science Foundation (Grant Nos. 3232044501.95, 3100-045984.95, 31-49618.96).
References DeBoer, D.J., Ewing, K.M., Schultz, K.T., 1993. Production and characterization of mouse monoclonal antibodies directed against canine IgE and IgG. Vet. Immunol. Immunopathol. 37, 183±199.
124
M.E. Griot-Wenk et al. / Veterinary Immunology and Immunopathology 70 (1999) 117±124
de Weck, A.L., Derer, M., Mayer, P., Stumper, B., SchoÈni, B., HaÈmmerling, R., Kristensen, F., Itaya, H., Yamasity, Y., Hasegawa, A., 1998. RuÈckblick und Perspektiven der Allergiediagnostik beim Hund: Ein neues serologisches Verfahren zur Bestimmung des allergen-spezifischen IgE mit Streifentesten. Der praktische Tierarzt 1, 6±23. Griot-Wenk, M.E., Marti, E., Racine, B., Crameri, R., Zurbriggen, A., de Weck, A.L., Lazary, S., 1998. Characterization of two dog IgE-specific antibodies elicited by different recombinant fragments of the epsilon chain in hens. Vet. Immunol. Immunopathol. 64, 15±32. Groeben, H., Lindeman, K.S., Davis, F.M., Cheng, W.I., Sampson, H.A., Hirshman, C.A., 1997. Dermal and airway responses to monoclonal antibodies specific for canine IgE. Vet. Immunol. Immunopathol. 58, 209± 218. Halliwell, R.E.W., Schwartzman, R.M., Rockey, J.H., 1972. Antigenic relationship between human IgE and canine IgE. Clin. Exp. Immunol. 10, 399±407. Halliwell, R.E.W., Schwartzmann, R.M., Montgomery, P.C., Rockey, J.H., 1975. Physicochemical properties of canine IgE. Transplant. Proc. 7, 537±543. Halliwell, R.E.W., Gilbert, S.M., Lian, T.M., 1998. Induced and spontaneous IgE antibodies to dermatophagoides-farinae in dogs and cats ± evidence of functional-heterogeneity of IgE. Vet. Dermatology 9, 179±184. Hammerberg, B., Bevier, D., DeBoer, D., Olivry, T., Orton, S.M., Gebhard, D., Vaden, S.L., 1997. Auto IgG anti-IgE and IgG x IgE immune complex presence and effect on ELISA-based quantitation of IgE in canine atopic dermatitis, demodectic acariasis and helminthiasis. Vet. Immunol. Immunopathol. 60, 33±46. Ishizaka, K., Ishizaka, T., Menzel, A.E.O., 1967. Physicochemical properties of reaginic antibody VI. Effect of heat on gE-, gG- and gA- antibodies in the sera of ragweed sensitive patients. J. Immunol. 99, 610±618. Patel, M., Selinger, D., Mark, G.E., Hickey, G.J., Hollis, G.F., 1995. Sequence of the dog immunoglobulin alpha and epsilon constant region genes. Immunogenetics 41, 282±286. Peng, Z., Simons, F.E.R., Becker, A.B., 1993. Measurement of ragweed-specific IgE in canine serum by use of enzyme-linked immunosorbent assays, containing polyclonal and monoclonal antibodies. Am. J. Vet. Res. 54, 239±243. Peng, Z., Arthur, G., Kierek-Jaszezuk, D., Thomas, E., Sehon, A.H., Rector, E.S., Simons, F.E.R., Becker, A.B., 1993. Purification and identification of polyclonal IgE antibodies from ragweed-sensitized dog sera. Int. Arch. Allergy Immunol. 102, 176±184. Peng, Z., Arthur, G., Rector, E.S., Kierek-Jaszczuk, D., Simons, F.E.R., Becker, A.B., 1997. Heterogeneity of polyclonal IgE characterized by differential charge, affinity to protein A, and antigenicity. J. Allergy Clin. Immunol. 100, 87±95. Scott, D.W., Miller, W.H., Griffin, C.E., 1995. Immunologic skin diseases. In: Muller, G.H., Kirk, R.W. (Eds.), Small Animal Dermatology, 5th ed., WB Saunders, Philadelphia, pp. 500±518. Schwartzman, R.M., Rockey, J.H., Halliwell, R.E., 1971. Canine reaginic antibody. Characterization of the spontaneous anti-ragweed and induced anti-dinitrophenyl reaginic antibodies of the atopic dog. Clin. Exp. Immunol. 9, 549±569. Sutton, B.J., Gould, H.J., 1993. The human IgE network. Nature 366, 421±428. Wassom, D.L., Grieve, R.B., 1998. In-vitro measurement of canine and feline IgE ± a review of Fc-epsilon-R1alpha-based assays for detection of allergen-reactive IgE. Vet. Dermatology 9, 173±178.
Domain mapping and comparative binding features of eight dog IgE-specific reagents in ELISA, immunoblots, and immunohistochemistry M.E. Griot-Wenka,b,*, E. Martia, D.J. DeBoerc, A.L. de Weckd, S. Lazarya a
Division of Immunogenetics, Institute of Animal Breeding, Bremgarten-Strasse 109a, Switzerland b Small Animal Clinics, LaÈnggass-Strasse 128; University of Bern, 3012 Bern, Switzerland c Department of Medical Sciences, School of Veterinary Medicine, University of Wisconsin, 2015 Linden Drive West, Madison, WI 53706, USA d CMG-Heska Allergy Products, 16 Grand'Places, 1700 Fribourg, Switzerland Received 31 December 1998; received in revised form 16 April 1999; accepted 5 May 1999
Abstract Eight dog IgE-specific reagents including monoclonal and polyclonal antibodies (Ab) and a cross-reactive alpha chain of the human high affinity IgE receptor were mapped to recombinant fragments of the second (IgEf2) and third/fourth (IgEf3/4) domains of the dog IgE heavy chain. In ELISA, five out of eight reagents reacted to solid-phase bound IgEf2, of which two polyclonal Ab bound in addition to IgEf3/4. All Ab which recognized at least one recombinant IgE fragment, also bound to IgE in ELISA, immunoblots, and immunohistochemistry. In contrast, only one monoclonal Ab, that did not bind to the recombinant IgE fragments, reacted with immunoblots of serum and immunohistochemistry. The alpha chain could only be applied to ELISA with serum IgE. Furthermore, there was a wide range of heat-lability of binding reactions. Comparative analysis of available dog IgE-specific reagents enables more in-depth functional studies on IgE-mediated phenomena in dogs, and helps to further establish the dog as an animal model for allergy research. # 1999 Elsevier Science B.V. All rights reserved. Keywords: Immunoglobulin E; Domain; Recombinant IgE; Dog
* Corresponding author. Present address: Novartis Nutrition Research AG, Fabrikstrasse 10, 3176 Neuenegg, Switzerland; Tel.: +41-31-377-24-85; fax: +41-31-377-24-34 E-mail address: [email protected] (M.E. Griot-Wenk)
0165-2427/99/$ ± see front matter # 1999 Elsevier Science B.V. All rights reserved. PII: S 0 1 6 5 - 2 4 2 7 ( 9 9 ) 0 0 0 7 0 - 7
118
M.E. Griot-Wenk et al. / Veterinary Immunology and Immunopathology 70 (1999) 117±124
1. Introduction Atopic conditions are common in dogs, out of which atopic dermatitis has been by far best characterized (Scott et al., 1995). For in-depth investigations of these presumed immunoglobuin E- (IgE) mediated immune reactions in the dog, a diversity of well characterized dog IgE-specific reagents are needed. Although dog IgE has been described more than two decades ago (Schwartzman et al., 1971; Halliwell et al., 1972, 1975), it was not until recently that these necessary tools have been developed, namely dog IgEspecific murine monoclonal antibodies (mAb; DeBoer et al., 1993; Peng et al., 1993a; Groeben et al., 1997; de Weck et al., 1998), cross-reactive recombinant alpha chain of the human high affinity IgE receptor (Wassom and Grieve, 1998), and several IgE-specific polyclonal Ab (Peng et al., 1993a, b; Hammerberg et al., 1997; Griot-Wenk et al., 1998). Furthermore, sequence analysis of dog IgE (Patel et al., 1995) enabled the production of recombinant dog IgE fragments (Griot-Wenk et al., 1998), which allows domain mapping of IgE-specific reagents. The objectives of the present study were to map eight dog IgE-specific reagents to recombinant dog IgE domains and to compare binding characteristics of these reagents for use in enzyme-linked immunosorbent assay (ELISA), immunoblots, and immunohistochemistry. 2. Materials and methods 2.1. Dog IgE-specific reagents The following reagents were applied in this study: polyclonal Ab raised in hens against the recombinant canine IgE fragments of the second (IgEf2) and third/fourth (IgEf3/4) constant domains (Ab1-2; Griot-Wenk et al., 1998) and in goat against purified dog serum IgE (Ab3; Bethyl Laboratories, Montgomery TX, USA), four mAb raised against purified dog or human serum IgE (mAb4-7; DeBoer et al., 1993; de Weck et al., 1998), and a cross-reactive, biotinylated recombinant alpha chain of the human IgE receptor (R8, Wassom and Grieve, 1998). 2.2. ELISA Capture ELISAs for domain mapping of IgE-specific reagents and for detection of peanut-specific serum IgE were applied as described by us previously (Griot-Wenk et al., 1998). Briefly, for domain mapping, the plates were coated with IgEf2 or IgEf3/4 (10 ng/ well), blocked and then subjected to the IgE-specific reagents (0.7±1.8 mg reagent/well). For visualization of binding the following corresponding alkaline phosphatase-(AP) labeled reagents were used: goat anti-chicken IgG Ab (1 : 1000; Kirkegaard and Perry Laboratories, Gaithersburg, MA, USA), goat anti-mouse IgG Ab (1 : 2000; Jackson ImmunoResearch Laboratories, West Grove, PA, USA), rabbit anti-goat IgG Ab
M.E. Griot-Wenk et al. / Veterinary Immunology and Immunopathology 70 (1999) 117±124
119
(1 : 2000; Sigma, St. Louis, MO, USA), and extravidin (1 : 20,000; Sigma). ELISA results were expressed as mean optical density of duplicate wells after subtraction of background controls (coated antigen without primary developing Ab) measured 30 min after adding enzyme substrates at a wavelength of 405 nm (OD405). Reactivity was defined as an OD405 of at least three times above background (OD405 of background 0.2). It was further subjectively designated as absent (0, OD405 < 0.6), weak (1+, OD405 0.6 and <2.0), or strong (2+, OD405 2.0). For detection of peanut-specific serum IgE, peanut-coated plates (100 ng/well) were incubated with serum (1 : 50 to 1 : 800) and developed as described for domain mapping (0.2 mg IgE-specific reagents/well). Additionally, results obtained with peanut-specific serum were compared to the ones achieved with serum heated at 568C, in which samples for analysis were taken at time points from 15 min to 4 h before application onto plates. 2.3. Immunoblotting Ammonium sulfate precipitated dog serum and IgEf3/4-specific Ab affinity-enriched dog serum fraction were analyzed by 5 and 10% PAGE (100 g/preparative gel) run under native or denaturing-reducing conditions (Griot-Wenk et al., 1998). The immunoblots were analyzed with the IgE-specific (2±10 mg reagent/ml) and secondary developing reagents as described for ELISA. Binding was visualized using 5-bromo-4-chloroindolyl-phosphate/4-nitro blue tetrazolium chloride (Boehringer Mannheim GmbH, Mannheim, Germany). Negative controls included application of secondary developing Ab alone. Immunoblot results were graded subjectively as follows: no band visible (0), visible but faint reaction (1+), and strong binding reaction (2+). 2.4. Immunohistochemistry All reagents were applied to histologic sections of a mesenteric lymph node from a dog as described (Griot-Wenk et al., 1998). In short, deparaffinized tissue sections were trypsin-treated and blocked. The IgE-specific reagents were applied at (6 mg/section) followed by incubation with the corresponding developing Ab: biotinylated rat antimouse IgG Ab (1 : 1000; Jackson Immuno Research Laboratories), AP-labeled rabbit anti-goat IgG Ab (1 : 2000; Sigma), or AP-labeled goat anti-chicken IgG Ab (1 : 1000; Kirkegaard and Perry Laboratories). The endogenous peroxidase and phosphatase were blocked with 1% hydrogen peroxide in methanol, respectively, by addition of levamisol to the secondary developing Ab (1 mg/ml, Sigma). To detect immunoreactivity of the biotinylated mAb the Vectastain1 ABC-PO kit was utilized according to the manufacturer's instructions (Vector Lab., Burlingame, CA, USA). Enzyme activity was detected with diaminobenzidine (Sigma) and hydrogen peroxide. Immunoreactivity of the IgE-specific goat and chicken Ab was detected using naphtol-AS-Mx-phosphate (Sigma) and fast red (Chroma Gesellschaft Schmid GmbH + Co., KoÈngen, Germany). As negative control all secondary developing Ab were applied alone, omitting IgE-specific reagents.
120
M.E. Griot-Wenk et al. / Veterinary Immunology and Immunopathology 70 (1999) 117±124
3. Results 3.1. Reactivity with IgEf2 and IgEf3/4 The two recombinant dog IgE fragments rendered it possible to investigate affinity of IgE-specific reagents to the second and third/fourth domains of dog IgE. In the ELISA applied, five out of eight reagents reacted to solid-phase bound IgEf2 and two out of eight with IgEf3/4, respectively (background OD405 0.1, Table 1). Antibody 3 and mAb6 reacted with high affinity to dog IgEf2; mAb4 also recognized IgEf2 but to a lesser extent. Moreover, Ab3 bound to IgEf3/4. As expected, Ab1 and Ab2, which had been raised against the recombinant IgE fragments, displayed high affinity to the corresponding immunogen and IgEf3/4-specific chicken Ab cross-reacted against IgEf2 (Griot-Wenk et al., 1998). 3.2. Affinity to peanut±specific IgE in ELISA and heat-lability of binding reactions The IgE-specific reagents were applied in an ELISA with solid-phase bound peanut as antigen followed by incubation with peanut-positive dog serum (heat-treated versus nonheated and serum dilution series) to compare affinity and heat-lability of the identified IgE-epitopes. In our test system, a wide range of affinity of the dog IgE-specific reagents to peanut-specific IgE was found. All but two mAb displayed positive reactions (Table 1). The OD405 ranged from 1.1 for Ab1 to 3.5 for Ab3. Different affinity to peanut-specific IgE was further reflected by serum dilution: Ab3, mAb4, and mAb6 resulted in an OD405 of at least 0.9 even at a serum dilution of 1 : 400. Comparative analysis of binding to serum that had been heated for up to 4 h resulted in at least 50% decrease in ELISA reactivity compared with non-heated serum in all reagents studied (Fig. 1). There was a wide range of heat sensitivity in the recognized epitopes. Increased binding after brief heat treatment of dog serum was evident only in Ab2; more prolonged heat treatment then resulted in diminished binding too. The most Table 1 Summary of binding features of eight dog IgE-specific antibodies (Ab)/reagent (R) available for this study in ELISA, immunoblots, and immunohistochemistry. Reactivity was subjectively designated as strong (2+), weak (1+), or absent (0) Designation
Reactivity in ELISA against a
Ab1 Ab2 Ab3 mAb4 mAb5 mAb6 mAb7 R8 a
a
Immunoblotting of serum
IgEf2
IgEf3/4
Serum IgE
Native
Denatured
2+ 2+ 2+ 1+ 0 2+ 0 0
0 2+ 2+ 0 0 0 0 0
1+ 2+ 2+ 2+ 0 2+ 0 2+
1+ 2+ 1+ 2+ 1+ 2+ 0 0
2+ 2+ 2+ 1+ 0 2+ 0 0
Immunohistochemistry
1+ 2+ 2+ 2+ 1+ 2+ 0 0
Recombinant canine IgE fragments of the second (IgEf2) and third/fourth (IgEf3/4) constant domains.
M.E. Griot-Wenk et al. / Veterinary Immunology and Immunopathology 70 (1999) 117±124
Fig. 1. Variation of heat-lability of recognized epitope of six dog IgE-specific reagents in an ELISA. Peanut-specific IgE-positive dog serum was diluted 1 : 50. Data are expressed as percentage of mean of duplicate OD405 values after subtraction of buffer controls obtained in heated serum as compared to non-heated serum.
121
122
M.E. Griot-Wenk et al. / Veterinary Immunology and Immunopathology 70 (1999) 117±124
heat-labile epitope was recognized by R8: heating serum for merely 15 min at 568C almost abolished binding (97% reduction compared to untreated serum). 3.3. Immunoblotting of native and denatured dog serum In immunoblots of native dog serum, all but mAb7 and the R8 bound to a protein of approximately 180 kDa (Table 1), the expected size of dog IgE (Griot-Wenk et al., 1998). Except for mAb5, binding to immunoblots of native serum corresponded to results obtained with immunoblots run under denaturing conditions: five out of six Ab recognized a 70 kDa-protein in enatured serum (expected size of heavy chain of IgE; Griot-Wenk et al., 1998). Semiquantitative assessment of strength of binding revealed that Ab1 and Ab3 reacted more strongly with the denatured than the native protein. 3.4. Immunohistochemistry IgE-bearing cells could be detected by our assay system in a mesenteric lymph node of a dog with all reagents except for mAb7 and R8 (Table 1). Cells were stained by the reagents with differing intensities. The IgE-bearing cells were predominantly located in the medulla and paracortical area. 4. Discussion This study describes domain mapping of eight different dog IgE-specific reagents to recombinant dog IgE heavy chain fragments, IgEf2 and IgEf3/4 (Griot-Wenk et al., 1998), in ELISA in comparison to their binding features to natural IgE in ELISA, immunoblots, and immunohistochemistry. Five out of eight reagents tested reacted with IgEf2 and only two with IgEf3/4. Being a polyclonal Ab, it is not surprising that Ab3 raised in goats recognized both recombinant IgE fragments. However, it cannot be elucidated whether this broad reactivity reflects shared or multiple epitopes as these Ab were raised against the whole IgE protein. In contrast, shared epitopes on IgEf2 and IgEf3/4 is indicated by the cross-reactivity seen with Ab2, which had been produced using IgEf3/4 (Griot-Wenk et al., 1998). The mAb4 and mAb6 which both recognize IgEf2 must be directed against an epitope only present on IgEf2 as they do not cross-react with IgEf3/4. Diversity of identified epitopes by IgE-specific reagents is further evident by the wide range of heat-sensitivity of binding (Fig. 1). There was only one Ab, namely Ab2, which must be directed against a rather heat resistant epitope(s) as more IgE could be detected at time points 15 and 30 min after heat-treatment of the serum. This Ab could be used to study immune complexes when disrupted by heat. The increase in IgE-specific IgG post disruption of immune complexes has been reported to be most profound in dogs affected with atopic dermatitis and demodectic acariasis (Hammerberg et al., 1997). Interestingly, the cross-reacting recombinant alpha chain of the human IgE receptor (R8) binds to an extremely heat-labile epitope; within 15 min, binding was almost completely abolished. Heat-lability is commonly considered a hallmark sign of an IgE-specific binding reaction,
M.E. Griot-Wenk et al. / Veterinary Immunology and Immunopathology 70 (1999) 117±124
123
and was first described for the human IgE molecule (Ishizaka et al., 1967). A variety of heat-sensitive epitopes have thus far been described for the dog (Schwartzman et al., 1971; DeBoer et al., 1993; Peng et al., 1993a; Hammerberg et al., 1997; Griot-Wenk et al., 1998). Alternatively, the variety in heat sensitivity may further support the existence of polymorphism of IgE in dogs as proposed by DeBoer et al. (1993) and recently supported by experimental data (Peng et al., 1997; Halliwell et al., 1998). Affinity is another defining feature of Ab, which may, however, be affected by the method of detection. The IgE-specific reagents studied here displayed a great variation in affinity in the ELISA utilized to detect peanut-specific IgE. Interestingly, all Ab but mAb7 and R8 resulted in IgE-detection in immunoblotting of serum run under native conditions and immunohistochemistry of paraffin-embedded lymph node tissue (Table 1). The latter observation is striking as all Ab have thus far been described for the usage on frozen sections only except for Ab2 (Griot-Wenk et al., 1998). Binding of mAb5 to native IgE in immunoblot and immunohistochemistry but not when it is bound to peanut antigen suggests that the epitope it recognizes is a conformational one that is changed when IgE `bends' as it binds to antigen. Immunoglobulin E binds to the alpha chain of the high-affinity IgE-receptor, Fc"RI. The actual binding site of IgE has been mapped to the first 12 amino acids of C"3 (reviewed by Sutton and Gould, 1993). The recombinant dog IgE fragment, IgEf3/4, however, does not include this first part of C"3 (Griot-Wenk et al., 1998), which probably explains lack of binding in ELISA against IgEf3/4 (Table 1). The alpha chain herein employed was constructed based on the human DNA sequence (Wassom and Grieve, 1998). The predicted gene product of the conserved C"3 constant region of dog IgE shares 62% amino acid identity with the corresponding human part potentially allowing cross-reactivity (Patel et al., 1995). In conclusion, these studies illustrate for the first time epitope mapping of dog IgEspecific reagents applying recombinant dog IgE fragments combined with data obtained from binding to native and denatured whole IgE. The diversity in binding features of the reagents supports a variety in epitopes on IgE or polymorphism of IgE. These in vitro observations may help to elucidate differences in clinical expression of atopic syndromes in individuals and at various body sites, which will further contribute to establishing the dog as an animal model for allergy. Acknowledgements We thank Corinne MuÈller (Institute of Animal Breeding) for technical assistance. This work was supported by the Swiss National Science Foundation (Grant Nos. 3232044501.95, 3100-045984.95, 31-49618.96).
References DeBoer, D.J., Ewing, K.M., Schultz, K.T., 1993. Production and characterization of mouse monoclonal antibodies directed against canine IgE and IgG. Vet. Immunol. Immunopathol. 37, 183±199.
124
M.E. Griot-Wenk et al. / Veterinary Immunology and Immunopathology 70 (1999) 117±124
de Weck, A.L., Derer, M., Mayer, P., Stumper, B., SchoÈni, B., HaÈmmerling, R., Kristensen, F., Itaya, H., Yamasity, Y., Hasegawa, A., 1998. RuÈckblick und Perspektiven der Allergiediagnostik beim Hund: Ein neues serologisches Verfahren zur Bestimmung des allergen-spezifischen IgE mit Streifentesten. Der praktische Tierarzt 1, 6±23. Griot-Wenk, M.E., Marti, E., Racine, B., Crameri, R., Zurbriggen, A., de Weck, A.L., Lazary, S., 1998. Characterization of two dog IgE-specific antibodies elicited by different recombinant fragments of the epsilon chain in hens. Vet. Immunol. Immunopathol. 64, 15±32. Groeben, H., Lindeman, K.S., Davis, F.M., Cheng, W.I., Sampson, H.A., Hirshman, C.A., 1997. Dermal and airway responses to monoclonal antibodies specific for canine IgE. Vet. Immunol. Immunopathol. 58, 209± 218. Halliwell, R.E.W., Schwartzman, R.M., Rockey, J.H., 1972. Antigenic relationship between human IgE and canine IgE. Clin. Exp. Immunol. 10, 399±407. Halliwell, R.E.W., Schwartzmann, R.M., Montgomery, P.C., Rockey, J.H., 1975. Physicochemical properties of canine IgE. Transplant. Proc. 7, 537±543. Halliwell, R.E.W., Gilbert, S.M., Lian, T.M., 1998. Induced and spontaneous IgE antibodies to dermatophagoides-farinae in dogs and cats ± evidence of functional-heterogeneity of IgE. Vet. Dermatology 9, 179±184. Hammerberg, B., Bevier, D., DeBoer, D., Olivry, T., Orton, S.M., Gebhard, D., Vaden, S.L., 1997. Auto IgG anti-IgE and IgG x IgE immune complex presence and effect on ELISA-based quantitation of IgE in canine atopic dermatitis, demodectic acariasis and helminthiasis. Vet. Immunol. Immunopathol. 60, 33±46. Ishizaka, K., Ishizaka, T., Menzel, A.E.O., 1967. Physicochemical properties of reaginic antibody VI. Effect of heat on gE-, gG- and gA- antibodies in the sera of ragweed sensitive patients. J. Immunol. 99, 610±618. Patel, M., Selinger, D., Mark, G.E., Hickey, G.J., Hollis, G.F., 1995. Sequence of the dog immunoglobulin alpha and epsilon constant region genes. Immunogenetics 41, 282±286. Peng, Z., Simons, F.E.R., Becker, A.B., 1993. Measurement of ragweed-specific IgE in canine serum by use of enzyme-linked immunosorbent assays, containing polyclonal and monoclonal antibodies. Am. J. Vet. Res. 54, 239±243. Peng, Z., Arthur, G., Kierek-Jaszezuk, D., Thomas, E., Sehon, A.H., Rector, E.S., Simons, F.E.R., Becker, A.B., 1993. Purification and identification of polyclonal IgE antibodies from ragweed-sensitized dog sera. Int. Arch. Allergy Immunol. 102, 176±184. Peng, Z., Arthur, G., Rector, E.S., Kierek-Jaszczuk, D., Simons, F.E.R., Becker, A.B., 1997. Heterogeneity of polyclonal IgE characterized by differential charge, affinity to protein A, and antigenicity. J. Allergy Clin. Immunol. 100, 87±95. Scott, D.W., Miller, W.H., Griffin, C.E., 1995. Immunologic skin diseases. In: Muller, G.H., Kirk, R.W. (Eds.), Small Animal Dermatology, 5th ed., WB Saunders, Philadelphia, pp. 500±518. Schwartzman, R.M., Rockey, J.H., Halliwell, R.E., 1971. Canine reaginic antibody. Characterization of the spontaneous anti-ragweed and induced anti-dinitrophenyl reaginic antibodies of the atopic dog. Clin. Exp. Immunol. 9, 549±569. Sutton, B.J., Gould, H.J., 1993. The human IgE network. Nature 366, 421±428. Wassom, D.L., Grieve, R.B., 1998. In-vitro measurement of canine and feline IgE ± a review of Fc-epsilon-R1alpha-based assays for detection of allergen-reactive IgE. Vet. Dermatology 9, 173±178.