- Email: [email protected]
Heteroclitic behaviour of some monoclonal antibodies against bovine growth hormone
Molec~ar immun&gy, Vol. 23, No. 2, pp. I Wl23, 1986 Printed in Great Britain. AI1rights reserved
0161-58~~86$3.00+ 0.00 Copyright C 1986Pergamon Press Ltd
HETEROCLITIC BEHAVIOUR OF SOME MONOCLONAL ANTIBODIES AGAINST BOVINE GROWTH HORMONE LILIA A. RETEGUI and ALEJANIIRO C. PALADINI* Instituto de Quimica y Fisicoquimica Biologicas (UBA-CQNICET), Facultad de Farmacia y Bioquimica, Universidad de Buenos Aires, Junin 956, 1113 Buenos Aires, Argentina (First received 6 May 1985; accepted in revised form 26 July 1985)
Abstract-Five monoclonal antibodies (MAbs) prepared against bovine growth hormone (bGH) were found to be directed against an immunodominant antigenic domain in bGH. MAbs D8 and H3 reacted equally well with bGH and ovine growth hormone and to a lesser extent with equine (eGH) or porcine (pGH) growth hormones while MAbs HI, H2 and Cl2 behaved as hetero~litic antibodies, i.e. they bound better a cross-reacting antigen (eGH on pGH) than the immunogen. The reactivity of bGH with the heteroclitic MAbs increased proportionally to the time that the native protein was kept frozen. Deamidation of bGH by treatment with alkali also increased its reactivity. Circular dichroism measurements indicated the occurrence of changes in the conformation of the bGH molecule by these treatments which presumably uncover normally buried or non-accesible epitopes. bGH shares epitopes with eGH and pGH which are immunologically expressed by bGH only when its native conformation is modified.
biological activities were as indicated by these authors. pGH was obtained from NIH, NIAMDD Hormone Distribution Program. bGH and eGH were labelled with lz51 by the chloramine-T method of Roth (1975). Their specific radioactivities ranged between 50 and 160 pCi/pg for 12’1-bGH and between 70 and 180 pCi/pg for ?-eGH. All GH stock solutions to be used in the radioimmunoassay (RIA) were prepared in 0.1 mM NaOH and kept at - 20°C. They were thawed only once just before using. Their concn, which was established by aminoacid analysis, was usually 0.7-l .Omg/ml. Deamidation of bGH was carried out by incubating the protein dissolved in 1 M NaOH (1.5 mg/ml) for 20 hr at room temp (Scoffone and Fontana, 1970). The solution was then dialyzed against 0.1 m M NaOH until the pH was 9.5. Circular dichroism (CD) measurements were performed at 25°C with a Jasco J-20 automatic recording spectropolarimeter.
INTRODUCTION
Some of the antibodies
in a polyclonal antiserum react with more affinity with heterologous antigens than with the immunogen. This classical finding made by Mlkela (1965) was called heterocliticity and was established with an antibody (Ab) fraction from
anti-bacte~ophage T2 serum which showed a signi~cantly higher inactivation constant for phage T4 than for T2. Other examples of the same phenomenon were Abs elicited in mice immunized with different haptens (Imanishi and Makelti, 1974; Karjalainen et al., 1980; Makela et al., 1976; Hammer and Steiner, 1982). An “in vitro” heteroclitic behaviour of T-cells immunized with pigeon cytochrome c was described by Solinger et al. (1979). Continuing a series of investigations devoted to the study of the antigenic structure of growth hormones (GHs) [see review by Paladini et ~2. (1983)] we prepared monoclonal Abs (MAbs) against bovine GH (bGH). Competition experiments showed that some MAbs bound equine GH (eGH) and porcine GH (pGH) with higher affinity than bGH, i.e. they were heteroclitic Abs.
Preparation
MATERIALSAND METHODS Hormones
bGH, ovine GH (oGH) and eGH were prepared in our laboratory following the methods of Dellacha and Sonenberg (1964), Dellacha et al. (1970) and Conde et al. (1973), respectively. Their purity and *Author to whom correspondence should be addressed at: Facuftad de Farmacia y Bioquimica, Junin 956, 1113 Buenos Aires, Argentina.
of the MAbs
The MAbs used in this work were obtained from two different hyb~dizations using 129/SV mice (MAbs D8 and H3) and BALB/c mice (MAbs Hl, H2 and C12). Spleen cells were fused with cells of the SP2/0 non secretor myeloma cell line as described by Retegui et al. (1982). Cloned cells were injected intraperitoneally into irradiated (500 rad) BALB/c mice primed with 2,6,10,14-tetramethyl-pentadecane (pristane, Aldrich). Ascitic fluids were collected and stored frozen at -20°C until used. MAbs D8, H3, Hl and H2 belong to the IgGl subclass, whereas MAb Cl2 is an IgG2b. Each of the MAbs used in this study derive from independent clones obtained after the
119
LILIA A. RETEGLJI and ALEJANDROC. PALADINI
120 hybridization. same original
Hence, they are not subclones hybridoma.
of the
RIA The screening of MAbs in the supernatants of the hybridoma cultures was carried out using disposable polyvinyl flex microtiter plates (Dynatech Laboratories Inc., Alexandria, VA) coated with bGH as indicated previously for hGH (Retegui et al., 1982). Cotitration of MAbs (Retegui et al., 1985) was performed as follows: the plates were coated with 200 ~1 of a bGH solution (5 pg/ml in 5 mM glycine buffer, pH 9.2) by incubating overnight at room temp and were washed with phosphate-buffered saline (PBS) containing 0.125 ml Tween 20/l (PBS-Tween). For the assay, serial dilutions of either individual MAbs or a 1: 1 combination of two different MAbs were prepared in PBS containing 5% (v/v) fetal calf serum (PBS-FCS) and 200~~1 aliquots of the dilutions were added to the coated plates. After incubating for 4 hr at 37°C and overnight at room temp, the plates were washed with PBS-Tween and ‘251-labelled Abs against mouse immunoglobulins (purified by immunoadsorption from rabbit antisera) were added. Before counting, the plates were incubated and washed as indicated. To study the specificities of the MAbs by competitive inhibition we used the solid-phase RIA described in Retegui et al. (1982). Briefly, the microplates were coated with the MAb diluted 1:500 in 5 mM glycine buffer, pH 9.2, and competition experiments carried out by addition of increasing concns of unlabelled hormones to the tracers, dissolved in PBS-FCS. The plates were incubated for 4 hr at 37°C and then overnight at room temp.; Their radioactivity was measured after being washed with PBSTween. RESULTS AND DISCUSSION
An immunodominant antigenic domain is recognized by bGH by the various MAbs In order to investigate the specificity of the five MAbs obtained we carried out the cotitration experiTable I. Cotitration of MAbs: measurement of the amount of MAb bound to insolubilized bGH or hGH (see Materials and Methods for details\ Radioactivity bound (cpm)” Insolubilized hormone hGH
Anti-bCH D8 D8 24,900 H3 HI H2 Cl2 I OD6 3Cll
hGH Anti-hGH MAbh
MAb
H3
HI
H2
Cl2
23,000 22,000
24,900 22,600 24,900
25,000 22,800 25,000 25,400
25,200 23,000 25,000 25,800 25,500
I OD6
3Cll
7900
2 1,400 9900
“cpm bound in the plateau region of the curve (I :20&l :3200 ascitic fluid dilutions). ‘MAbs lOD6 and 3Cl I recognize two different antigenic sites on the hGH molecule (Retegui ef al., 1982).
ments shown in Table 1. The radioactivity measured is an indication of the amount of MAb bound to the insolubilized hormones. A constant value was obtained throughout no matter which one of two MAbs was present in the incubation mixture. It is thus reasonable to assume that all of them are directed against a restricted antigenic domain of the bGH molecule which contains the implicated epitopes. As a control the experiment includes two anti-human GH (hGH) MAbs which are known (Retegui et al., 1982) to recognize two different antigenic sites in hGH. The results shown in Table 1 clearly indicates that their binding to insoluble hGH is additive.
Binding of various GHs to insolubilized MAbs studied using ‘25/-bGH or 12’I-eGH as tracers The curves obtained using increasing concns of bGH, oGH, eGH and pGH as competitors and ‘251-bGH as tracer showed that the MAbs could be classified into two distinct groups. The first one included MAbs D8 and H3 and the second one was formed by MAbs H 1, H2 and C12. An example of the behaviour of MAbs from the first group is shown in Fig. 1A for MAb D8. This antibody reacted equally well and with high affinity with bGH and oGH, whereas its reaction with pGH and eGH was 20- or 40-fold weaker, respectively. Identical results were obtained with MAb H3 (Table 2). The immunological properties of MAbs from the second group are exemplified by the results obtained with MAb Hl (Fig. 1B). This Ab bound eGH and pGH with higher affinity than bGH and oGH. Similar results were obtained with MAbs H2 and Cl2 (Table 2). When “‘1-eGH was used as a tracer the relative order of the inhibition curves was different, as shown for MAb Hl (Fig. 1C). The strongest inhibitor was eGH, followed by pGH. The extent of binding of bGH and oGH was poor. Essentially the same results were obtained with MAbs H2 and Cl2 (Table 2). According to these results MAbs D8 and H3 behaved as homoclitic Abs, i.e. they reacted with high affinity with the immunogen and to a lesser extent with other proteins from the GH family. On the other hand MAbs Hl, H2 and Cl2 were shown to be heteroclitic antibodies (Makell, 1965) since they bound cross-reacting antigens (eGH and pGH) better than the immunogen. Their reactions with the different GHs was found to be a function of the tracer used in the RIA. The affinity order was eGH = pGH > bGH > oGH when ‘251-bGH was but used the order changed to eGH > pGH > bGH = oGH when these proteins were competing for the binding of 12’I-eGH (Fig. 1). These observations indicate the subtle specificity of the heteroclitic MAbs: they were even able to discern between bGH and oGH, whose sequences differ in only one amino acid: glycine in position 13 1, in bGH, is replaced by valine in oGH (Paladini et al., 1983).
Heteroclitic antibodies to growth hormone
A
B
C
-.-A+
‘251-eGH 20t
--_-
__
10
1
Hormone
cvcentrcllon
1oc ‘nM)
Fig. 1. Competitive inhibition of binding of 9-bGH (0.2 nM) to insolubilized MAbs D8 (A) and Hl (B) and of binding of ‘251-eGH(0.3 nM) to insolubilized MAb Hl (C). The inhibitors were bGH (a), oGH (0) eGH (+) and pGH (A). The results are expressed as percentages of maximal binding (the 100% values were 75,263 cpm in A, 20,671 cpm in B and 27,023cpm in C).
121
and a positive correlation was found between aging and the effectiveness of the inhibition (Fig. 2). eGH did not behave in the same way as bGH: its competitive behaviour remained unaltered up to 1 yr of aging under the same conditions as bGH (Fig. 2). The experiments shown in Fig. 1C were done with a l-month-aged solution of bGH. Identical results were obtained with MAbs H2 and Cl2 (data not shown). A chemical modification of the protein during storage, coupled perhaps to a conformation change, could be responsible for the phenomenon described. In fact, alkaline deamidation of the hormone closely duplicated the effect of 1 yr of storage (Fig. 2). The changes in secondary structure and conformation of the various GH preparations were studied by CD measurements. The CD spectra in the near-u.v. of I-yr-aged or deamidated bGH (Fig. 3) showed that the protein in both preparations had suffered substantial conformation changes in the neighbourhood of the aromatic residues and disulfide bridges (Bewley, 1979). The behaviour of eGH was strikingly different: a freshly prepared or a I-yr-old solution gave quite similar CD spectra in the same region of the spectrum. The cc-helix content of each hormone at pH 9.5 was calculated from their respective CD spectra in the far-u.v. and the values obtained are collected in Table 3. Prolonged storage decreased the cc-helix content in bGH but a drastic reduction only occurred after the alkaline deamidation treatment. The x-helix content of eGH was mostly unaffected after 1 yr of storage. It is not possible with present data to define the region of bGH presumably uncovered by aging, or chemical treatment, responsible for its greater competitive effectiveness towards the binding of eGH by the heteroclitic MAbs. The simplest interpretation
Effect of aging of frozen solutions of bGH or a deamidation treatment of the hormone on its effectiveness as a competitive inhibitor in the binding of ‘=‘I-eGH by heteroclitic MAbs Alkaline bGH solutions, that had been kept frozen for different periods, were used to competitively inhibit the binding of iz51-eGH to insolubilized MAbs Table 2. Reactivities
with various
GHs of MAbs against
Binding
Tracer = ‘*‘I-eGH
Tracer = “1-bGH MAb D8 H3 HI H2 Cl2
bGH
1 1 1 I
I
oGH
I I 6 2 3
bGH
to’
eGH
pGH
bGH
oGH
40 45 0.4 0.2 0.2
21 19 0.4 0.3 -
7 12 9
7 9 I
eGH
I I
I
pGH
I 4 3
“Expressed as the relative GH cmcn required to obtain 50% binding inhibition. Absolute bGH concns to obtain 50% binding inhibition were 7nM for MAb D8, 8nM for MAb H3 3x108M~‘),42nMforMAbHI,68nMforH2and Km, ” 41 nM for MAb Cl2 (K,,,,=O.2-0.6 x 108Mm’). Absolute eGH concns to obtain 50% binding inhibition were 32 nM for MA\, HI, II nM for MAb H2 and 12nM for MAb Cl2 V&o, = 2-3 x IO” W’).
Fig. 2. Competitive inhibition of binding of ‘Z51-eGH (0.2nM) to insolubilized MAb Hl. The inhibitors were eGH (open symobls) and bGH (filled symbols) whose solutions were kept for different times at -20°C. Deamidated bGH (+) was prepared as indicated in Materials and Methods. All stock solutions were approx. 1mg/ml in 0.1 mM NaOH, using a unique batch of either eGH or bGH. Final dilutions in PBS-FCS were done a few minutes prior to development of the RIA. The results are expressed as percentages of maximal binding (100% = 30,158 cpm).
LILIAA. RETEGUI and ALEJANDRO C. PALADINI
122
It is also known that when hGH is injected in human beings it is possible to detect in their serum Abs against non-human GHs (Poskus et al., 1982) which were obviously originated by hGH. These results and those described in the present paper indicate that the mammalian GHs examined have in common a number of epitopes either available or masked in the native proteins. Acknowledgements-We are grateful to Prof. P. L. Masson (ICP, MEXP, Brussels, Belgium) in whose laboratory LA Retegui prepared the MAbs used in this work. The secretarial assistance of Miss M. Ramirez is also gratefully acknowledged. This work was supported in part by grants from the Regional Program of Scientific and Technological Development of O.A.S. LAR and ACP are Career Investigators, CONICET.
VI
I
250
I
I
I
I
260 270 280 290 Wavelength (nml
REFERENCES
300
Fig. 3. CD spectra in the near U.V. of fresh or aged solutions of bGH or eGH, and of alkali-treated bGH. The hormones were dissolved in 0.1 mM NaOH (pH 9.5) at a concn of 1 mg/ml. bGHa is the spectrum of the solution recently prepared, bGHb is the spectrum obtained with the same solution after being kept frozen for 1 yr, and bGHc is the spectrum of bGH treated with alkali as indicated in Materials and Methods. For eGH (---) and (-) indicate the spectra of solutions freshly prepared, and 1 yr of being kept frozen, respectively.
of
the
facts is that the eGH epitope by the heteroclitic Abs is unmasked in
recorded
recognized bGH
after
through
prolonged
a more
storage
drastic
in alkaline
alkaline
solution
or
treatment.
It is noteworthy that homoclitic MAbs D8 and H3 did not detect any difference between aged and fresh bGH solutions in spite of the fact that they are directed against the same antigenic domain of the bGH molecule recognized by the heteroclitic MAbs. Further, the results shown in Table 1 indicate that all five MAbs bind to bGH in the same proportion when the protein is adsorbed on the polyvinyl microplates. Hence this surface adsorption quite probably brings about a conformation change that expresses all these epitopes in the protein. Friguet et al. (1984) have recently described a similar phenomenon with MAbs against the &subunit of E. coli tryptophan synthase and have reviewed several other instances which support the hypothesis that a protein may undergo some denaturation with partial or total unfolding when it is adsorbed to a solid surface as occurs in a solid-phase RIA.
Table 3.
a-Helix
content
of bGH and eCH in solutions
(0.1 mM NaOH) recently prepared or after
at pH 9.5
1yr at -20°C”
o.-Helix (%) Treatment Recent solution
37 31
45 33
bGH eGH “Deaminated
Methods.
bGH
of the hormone
Solution after lyrat -20°C
was prepared
as indicated
Deamidation 17 in Materials
and
Bewley T. A. (1979) Circular dichroism of pituitary hormones. Rec. Prog. Horm. Res. 35, 155-213. Conde R. D., Paladini A. C., Santomi. J. A. and Dellacha J. M. (1973) Isolation, purification and characterization of equine growth hormone. Eur. J. Biochem. 32,563%568. Dellacha J. M., Enero M. A., Santomk J. A. and Paladini A. C. (1970) Isolation, characterization and C-terminal sequence of ovine growth hormone. Eur. J. Biochem. 12, 289-295. Dellacha J. M. and Sonenberg M. (1964) Purification of bovine growth hormones. J. hiol. Chem. 239, 1515-1520. Friguet B., Djavadi-Ohaniance L. and Goldberg M. E. (1984) Some monoclonal antibodies raised with a native protein bind preferentially to the denatured antigen. Molec. Immun. 21, 673-677. Hammer P. and Steiner L. A. (1982) Specificity and heterocliticity of rabbit antisera to the 2,4-dinitrophenyl determinant. J. Immun. 128, 343-350. Imanishi T. and Mlkela 0. (1974) Inheritance of antibody specificity I. Anti-(4-hydroxy-3-nitrophenyl)acetyl of the mouse primary response. J. exp. Med. 140, 1498-1510. Karjalainen K., BPng B. and Mikell 0. (1980) Fine specificity and idiotypes of early antibodies against (4-hydroxy-3-nitrophenyl)acetyl (NP). J. Immun. 125, 313-317. MBkell 0. (1965) Single lymph node cells producing heteroclitic bacteriophage antibody. J. Immun. 95, 378-386. MLkell O., Julin M. and Becker M. (1976) Inheritance of antibody specificity. III. A new V, gene controls fine specificity of anti-p-azobenzene-arsonate coupled to a carbon atom 5 of hydroxy-phenylacetic acid in the mouse. J. exp. Med. 143, 3lti328. Paladini A. C., Pefia C. and Poskus E. (1983) In Molecular Biology of Growth Hormone (Edited by Hurrel J. G. R.), Vol. 215, pp. 25~56. CRC Press, Cleveland, OH. Poskus E., Pefia C., P&rez A. R., Vita N., Heinrich J. J. and Paladini A. C. (1982) Human, bovine and equine growth hormone antibodies in patients treated with human growth hormone. J. clin. Endocr. Metab. 55, 13-17. Retegui L. A., Masson P. L. and Paladini A. C. (1985) Specificities of antibodies to hGH in patients treated with hGH. Longitudinal study and comparison with the specificities of animal antisera. J. clin. Endocr. Metab. 60, 184190. Retegui L. A., Milne R. W., Cambiaso C. L. and Masson P. L. (1982) The recognition by monoclonal antibodies of various portions of a major antigenic site of human growth hormone. Molec. Immun. 19, 865-875. Roth J. (1975) Methods for assessing immunologic and biologic properties of iodinated peptide hormones. In Methods in Enzymology (Edited by Colowick S. 0. and Kaplan N. O.), Vol. 37, part B, Peptide Hormones, p. 223. Scoffone E. and Fontana A. (1970) In Protein Sequence
Heteroclitic antibodies to growth hormone Defermination (Edited by Needleman S. B.), Vol. 18, pp.
185-210. Chapman & Hall, London. Solinger A. M., Ultee M. E., Margohash E. and Schwartz R. H. (1979) T-lymphocyte response to cytochrome c. I. Demonstration of a T-cell heteroclitic proliferative re-
123
sponse and identification of a topographic antigenic determinant on pigeon cytochrome c whose immune recognition requires two complementing major histocompatibility complex-linked immune response genes. J. exp. Med. 150, 83&848.
0161-58~~86$3.00+ 0.00 Copyright C 1986Pergamon Press Ltd
HETEROCLITIC BEHAVIOUR OF SOME MONOCLONAL ANTIBODIES AGAINST BOVINE GROWTH HORMONE LILIA A. RETEGUI and ALEJANIIRO C. PALADINI* Instituto de Quimica y Fisicoquimica Biologicas (UBA-CQNICET), Facultad de Farmacia y Bioquimica, Universidad de Buenos Aires, Junin 956, 1113 Buenos Aires, Argentina (First received 6 May 1985; accepted in revised form 26 July 1985)
Abstract-Five monoclonal antibodies (MAbs) prepared against bovine growth hormone (bGH) were found to be directed against an immunodominant antigenic domain in bGH. MAbs D8 and H3 reacted equally well with bGH and ovine growth hormone and to a lesser extent with equine (eGH) or porcine (pGH) growth hormones while MAbs HI, H2 and Cl2 behaved as hetero~litic antibodies, i.e. they bound better a cross-reacting antigen (eGH on pGH) than the immunogen. The reactivity of bGH with the heteroclitic MAbs increased proportionally to the time that the native protein was kept frozen. Deamidation of bGH by treatment with alkali also increased its reactivity. Circular dichroism measurements indicated the occurrence of changes in the conformation of the bGH molecule by these treatments which presumably uncover normally buried or non-accesible epitopes. bGH shares epitopes with eGH and pGH which are immunologically expressed by bGH only when its native conformation is modified.
biological activities were as indicated by these authors. pGH was obtained from NIH, NIAMDD Hormone Distribution Program. bGH and eGH were labelled with lz51 by the chloramine-T method of Roth (1975). Their specific radioactivities ranged between 50 and 160 pCi/pg for 12’1-bGH and between 70 and 180 pCi/pg for ?-eGH. All GH stock solutions to be used in the radioimmunoassay (RIA) were prepared in 0.1 mM NaOH and kept at - 20°C. They were thawed only once just before using. Their concn, which was established by aminoacid analysis, was usually 0.7-l .Omg/ml. Deamidation of bGH was carried out by incubating the protein dissolved in 1 M NaOH (1.5 mg/ml) for 20 hr at room temp (Scoffone and Fontana, 1970). The solution was then dialyzed against 0.1 m M NaOH until the pH was 9.5. Circular dichroism (CD) measurements were performed at 25°C with a Jasco J-20 automatic recording spectropolarimeter.
INTRODUCTION
Some of the antibodies
in a polyclonal antiserum react with more affinity with heterologous antigens than with the immunogen. This classical finding made by Mlkela (1965) was called heterocliticity and was established with an antibody (Ab) fraction from
anti-bacte~ophage T2 serum which showed a signi~cantly higher inactivation constant for phage T4 than for T2. Other examples of the same phenomenon were Abs elicited in mice immunized with different haptens (Imanishi and Makelti, 1974; Karjalainen et al., 1980; Makela et al., 1976; Hammer and Steiner, 1982). An “in vitro” heteroclitic behaviour of T-cells immunized with pigeon cytochrome c was described by Solinger et al. (1979). Continuing a series of investigations devoted to the study of the antigenic structure of growth hormones (GHs) [see review by Paladini et ~2. (1983)] we prepared monoclonal Abs (MAbs) against bovine GH (bGH). Competition experiments showed that some MAbs bound equine GH (eGH) and porcine GH (pGH) with higher affinity than bGH, i.e. they were heteroclitic Abs.
Preparation
MATERIALSAND METHODS Hormones
bGH, ovine GH (oGH) and eGH were prepared in our laboratory following the methods of Dellacha and Sonenberg (1964), Dellacha et al. (1970) and Conde et al. (1973), respectively. Their purity and *Author to whom correspondence should be addressed at: Facuftad de Farmacia y Bioquimica, Junin 956, 1113 Buenos Aires, Argentina.
of the MAbs
The MAbs used in this work were obtained from two different hyb~dizations using 129/SV mice (MAbs D8 and H3) and BALB/c mice (MAbs Hl, H2 and C12). Spleen cells were fused with cells of the SP2/0 non secretor myeloma cell line as described by Retegui et al. (1982). Cloned cells were injected intraperitoneally into irradiated (500 rad) BALB/c mice primed with 2,6,10,14-tetramethyl-pentadecane (pristane, Aldrich). Ascitic fluids were collected and stored frozen at -20°C until used. MAbs D8, H3, Hl and H2 belong to the IgGl subclass, whereas MAb Cl2 is an IgG2b. Each of the MAbs used in this study derive from independent clones obtained after the
119
LILIA A. RETEGLJI and ALEJANDROC. PALADINI
120 hybridization. same original
Hence, they are not subclones hybridoma.
of the
RIA The screening of MAbs in the supernatants of the hybridoma cultures was carried out using disposable polyvinyl flex microtiter plates (Dynatech Laboratories Inc., Alexandria, VA) coated with bGH as indicated previously for hGH (Retegui et al., 1982). Cotitration of MAbs (Retegui et al., 1985) was performed as follows: the plates were coated with 200 ~1 of a bGH solution (5 pg/ml in 5 mM glycine buffer, pH 9.2) by incubating overnight at room temp and were washed with phosphate-buffered saline (PBS) containing 0.125 ml Tween 20/l (PBS-Tween). For the assay, serial dilutions of either individual MAbs or a 1: 1 combination of two different MAbs were prepared in PBS containing 5% (v/v) fetal calf serum (PBS-FCS) and 200~~1 aliquots of the dilutions were added to the coated plates. After incubating for 4 hr at 37°C and overnight at room temp, the plates were washed with PBS-Tween and ‘251-labelled Abs against mouse immunoglobulins (purified by immunoadsorption from rabbit antisera) were added. Before counting, the plates were incubated and washed as indicated. To study the specificities of the MAbs by competitive inhibition we used the solid-phase RIA described in Retegui et al. (1982). Briefly, the microplates were coated with the MAb diluted 1:500 in 5 mM glycine buffer, pH 9.2, and competition experiments carried out by addition of increasing concns of unlabelled hormones to the tracers, dissolved in PBS-FCS. The plates were incubated for 4 hr at 37°C and then overnight at room temp.; Their radioactivity was measured after being washed with PBSTween. RESULTS AND DISCUSSION
An immunodominant antigenic domain is recognized by bGH by the various MAbs In order to investigate the specificity of the five MAbs obtained we carried out the cotitration experiTable I. Cotitration of MAbs: measurement of the amount of MAb bound to insolubilized bGH or hGH (see Materials and Methods for details\ Radioactivity bound (cpm)” Insolubilized hormone hGH
Anti-bCH D8 D8 24,900 H3 HI H2 Cl2 I OD6 3Cll
hGH Anti-hGH MAbh
MAb
H3
HI
H2
Cl2
23,000 22,000
24,900 22,600 24,900
25,000 22,800 25,000 25,400
25,200 23,000 25,000 25,800 25,500
I OD6
3Cll
7900
2 1,400 9900
“cpm bound in the plateau region of the curve (I :20&l :3200 ascitic fluid dilutions). ‘MAbs lOD6 and 3Cl I recognize two different antigenic sites on the hGH molecule (Retegui ef al., 1982).
ments shown in Table 1. The radioactivity measured is an indication of the amount of MAb bound to the insolubilized hormones. A constant value was obtained throughout no matter which one of two MAbs was present in the incubation mixture. It is thus reasonable to assume that all of them are directed against a restricted antigenic domain of the bGH molecule which contains the implicated epitopes. As a control the experiment includes two anti-human GH (hGH) MAbs which are known (Retegui et al., 1982) to recognize two different antigenic sites in hGH. The results shown in Table 1 clearly indicates that their binding to insoluble hGH is additive.
Binding of various GHs to insolubilized MAbs studied using ‘25/-bGH or 12’I-eGH as tracers The curves obtained using increasing concns of bGH, oGH, eGH and pGH as competitors and ‘251-bGH as tracer showed that the MAbs could be classified into two distinct groups. The first one included MAbs D8 and H3 and the second one was formed by MAbs H 1, H2 and C12. An example of the behaviour of MAbs from the first group is shown in Fig. 1A for MAb D8. This antibody reacted equally well and with high affinity with bGH and oGH, whereas its reaction with pGH and eGH was 20- or 40-fold weaker, respectively. Identical results were obtained with MAb H3 (Table 2). The immunological properties of MAbs from the second group are exemplified by the results obtained with MAb Hl (Fig. 1B). This Ab bound eGH and pGH with higher affinity than bGH and oGH. Similar results were obtained with MAbs H2 and Cl2 (Table 2). When “‘1-eGH was used as a tracer the relative order of the inhibition curves was different, as shown for MAb Hl (Fig. 1C). The strongest inhibitor was eGH, followed by pGH. The extent of binding of bGH and oGH was poor. Essentially the same results were obtained with MAbs H2 and Cl2 (Table 2). According to these results MAbs D8 and H3 behaved as homoclitic Abs, i.e. they reacted with high affinity with the immunogen and to a lesser extent with other proteins from the GH family. On the other hand MAbs Hl, H2 and Cl2 were shown to be heteroclitic antibodies (Makell, 1965) since they bound cross-reacting antigens (eGH and pGH) better than the immunogen. Their reactions with the different GHs was found to be a function of the tracer used in the RIA. The affinity order was eGH = pGH > bGH > oGH when ‘251-bGH was but used the order changed to eGH > pGH > bGH = oGH when these proteins were competing for the binding of 12’I-eGH (Fig. 1). These observations indicate the subtle specificity of the heteroclitic MAbs: they were even able to discern between bGH and oGH, whose sequences differ in only one amino acid: glycine in position 13 1, in bGH, is replaced by valine in oGH (Paladini et al., 1983).
Heteroclitic antibodies to growth hormone
A
B
C
-.-A+
‘251-eGH 20t
--_-
__
10
1
Hormone
cvcentrcllon
1oc ‘nM)
Fig. 1. Competitive inhibition of binding of 9-bGH (0.2 nM) to insolubilized MAbs D8 (A) and Hl (B) and of binding of ‘251-eGH(0.3 nM) to insolubilized MAb Hl (C). The inhibitors were bGH (a), oGH (0) eGH (+) and pGH (A). The results are expressed as percentages of maximal binding (the 100% values were 75,263 cpm in A, 20,671 cpm in B and 27,023cpm in C).
121
and a positive correlation was found between aging and the effectiveness of the inhibition (Fig. 2). eGH did not behave in the same way as bGH: its competitive behaviour remained unaltered up to 1 yr of aging under the same conditions as bGH (Fig. 2). The experiments shown in Fig. 1C were done with a l-month-aged solution of bGH. Identical results were obtained with MAbs H2 and Cl2 (data not shown). A chemical modification of the protein during storage, coupled perhaps to a conformation change, could be responsible for the phenomenon described. In fact, alkaline deamidation of the hormone closely duplicated the effect of 1 yr of storage (Fig. 2). The changes in secondary structure and conformation of the various GH preparations were studied by CD measurements. The CD spectra in the near-u.v. of I-yr-aged or deamidated bGH (Fig. 3) showed that the protein in both preparations had suffered substantial conformation changes in the neighbourhood of the aromatic residues and disulfide bridges (Bewley, 1979). The behaviour of eGH was strikingly different: a freshly prepared or a I-yr-old solution gave quite similar CD spectra in the same region of the spectrum. The cc-helix content of each hormone at pH 9.5 was calculated from their respective CD spectra in the far-u.v. and the values obtained are collected in Table 3. Prolonged storage decreased the cc-helix content in bGH but a drastic reduction only occurred after the alkaline deamidation treatment. The x-helix content of eGH was mostly unaffected after 1 yr of storage. It is not possible with present data to define the region of bGH presumably uncovered by aging, or chemical treatment, responsible for its greater competitive effectiveness towards the binding of eGH by the heteroclitic MAbs. The simplest interpretation
Effect of aging of frozen solutions of bGH or a deamidation treatment of the hormone on its effectiveness as a competitive inhibitor in the binding of ‘=‘I-eGH by heteroclitic MAbs Alkaline bGH solutions, that had been kept frozen for different periods, were used to competitively inhibit the binding of iz51-eGH to insolubilized MAbs Table 2. Reactivities
with various
GHs of MAbs against
Binding
Tracer = ‘*‘I-eGH
Tracer = “1-bGH MAb D8 H3 HI H2 Cl2
bGH
1 1 1 I
I
oGH
I I 6 2 3
bGH
to’
eGH
pGH
bGH
oGH
40 45 0.4 0.2 0.2
21 19 0.4 0.3 -
7 12 9
7 9 I
eGH
I I
I
pGH
I 4 3
“Expressed as the relative GH cmcn required to obtain 50% binding inhibition. Absolute bGH concns to obtain 50% binding inhibition were 7nM for MAb D8, 8nM for MAb H3 3x108M~‘),42nMforMAbHI,68nMforH2and Km, ” 41 nM for MAb Cl2 (K,,,,=O.2-0.6 x 108Mm’). Absolute eGH concns to obtain 50% binding inhibition were 32 nM for MA\, HI, II nM for MAb H2 and 12nM for MAb Cl2 V&o, = 2-3 x IO” W’).
Fig. 2. Competitive inhibition of binding of ‘Z51-eGH (0.2nM) to insolubilized MAb Hl. The inhibitors were eGH (open symobls) and bGH (filled symbols) whose solutions were kept for different times at -20°C. Deamidated bGH (+) was prepared as indicated in Materials and Methods. All stock solutions were approx. 1mg/ml in 0.1 mM NaOH, using a unique batch of either eGH or bGH. Final dilutions in PBS-FCS were done a few minutes prior to development of the RIA. The results are expressed as percentages of maximal binding (100% = 30,158 cpm).
LILIAA. RETEGUI and ALEJANDRO C. PALADINI
122
It is also known that when hGH is injected in human beings it is possible to detect in their serum Abs against non-human GHs (Poskus et al., 1982) which were obviously originated by hGH. These results and those described in the present paper indicate that the mammalian GHs examined have in common a number of epitopes either available or masked in the native proteins. Acknowledgements-We are grateful to Prof. P. L. Masson (ICP, MEXP, Brussels, Belgium) in whose laboratory LA Retegui prepared the MAbs used in this work. The secretarial assistance of Miss M. Ramirez is also gratefully acknowledged. This work was supported in part by grants from the Regional Program of Scientific and Technological Development of O.A.S. LAR and ACP are Career Investigators, CONICET.
VI
I
250
I
I
I
I
260 270 280 290 Wavelength (nml
REFERENCES
300
Fig. 3. CD spectra in the near U.V. of fresh or aged solutions of bGH or eGH, and of alkali-treated bGH. The hormones were dissolved in 0.1 mM NaOH (pH 9.5) at a concn of 1 mg/ml. bGHa is the spectrum of the solution recently prepared, bGHb is the spectrum obtained with the same solution after being kept frozen for 1 yr, and bGHc is the spectrum of bGH treated with alkali as indicated in Materials and Methods. For eGH (---) and (-) indicate the spectra of solutions freshly prepared, and 1 yr of being kept frozen, respectively.
of
the
facts is that the eGH epitope by the heteroclitic Abs is unmasked in
recorded
recognized bGH
after
through
prolonged
a more
storage
drastic
in alkaline
alkaline
solution
or
treatment.
It is noteworthy that homoclitic MAbs D8 and H3 did not detect any difference between aged and fresh bGH solutions in spite of the fact that they are directed against the same antigenic domain of the bGH molecule recognized by the heteroclitic MAbs. Further, the results shown in Table 1 indicate that all five MAbs bind to bGH in the same proportion when the protein is adsorbed on the polyvinyl microplates. Hence this surface adsorption quite probably brings about a conformation change that expresses all these epitopes in the protein. Friguet et al. (1984) have recently described a similar phenomenon with MAbs against the &subunit of E. coli tryptophan synthase and have reviewed several other instances which support the hypothesis that a protein may undergo some denaturation with partial or total unfolding when it is adsorbed to a solid surface as occurs in a solid-phase RIA.
Table 3.
a-Helix
content
of bGH and eCH in solutions
(0.1 mM NaOH) recently prepared or after
at pH 9.5
1yr at -20°C”
o.-Helix (%) Treatment Recent solution
37 31
45 33
bGH eGH “Deaminated
Methods.
bGH
of the hormone
Solution after lyrat -20°C
was prepared
as indicated
Deamidation 17 in Materials
and
Bewley T. A. (1979) Circular dichroism of pituitary hormones. Rec. Prog. Horm. Res. 35, 155-213. Conde R. D., Paladini A. C., Santomi. J. A. and Dellacha J. M. (1973) Isolation, purification and characterization of equine growth hormone. Eur. J. Biochem. 32,563%568. Dellacha J. M., Enero M. A., Santomk J. A. and Paladini A. C. (1970) Isolation, characterization and C-terminal sequence of ovine growth hormone. Eur. J. Biochem. 12, 289-295. Dellacha J. M. and Sonenberg M. (1964) Purification of bovine growth hormones. J. hiol. Chem. 239, 1515-1520. Friguet B., Djavadi-Ohaniance L. and Goldberg M. E. (1984) Some monoclonal antibodies raised with a native protein bind preferentially to the denatured antigen. Molec. Immun. 21, 673-677. Hammer P. and Steiner L. A. (1982) Specificity and heterocliticity of rabbit antisera to the 2,4-dinitrophenyl determinant. J. Immun. 128, 343-350. Imanishi T. and Mlkela 0. (1974) Inheritance of antibody specificity I. Anti-(4-hydroxy-3-nitrophenyl)acetyl of the mouse primary response. J. exp. Med. 140, 1498-1510. Karjalainen K., BPng B. and Mikell 0. (1980) Fine specificity and idiotypes of early antibodies against (4-hydroxy-3-nitrophenyl)acetyl (NP). J. Immun. 125, 313-317. MBkell 0. (1965) Single lymph node cells producing heteroclitic bacteriophage antibody. J. Immun. 95, 378-386. MLkell O., Julin M. and Becker M. (1976) Inheritance of antibody specificity. III. A new V, gene controls fine specificity of anti-p-azobenzene-arsonate coupled to a carbon atom 5 of hydroxy-phenylacetic acid in the mouse. J. exp. Med. 143, 3lti328. Paladini A. C., Pefia C. and Poskus E. (1983) In Molecular Biology of Growth Hormone (Edited by Hurrel J. G. R.), Vol. 215, pp. 25~56. CRC Press, Cleveland, OH. Poskus E., Pefia C., P&rez A. R., Vita N., Heinrich J. J. and Paladini A. C. (1982) Human, bovine and equine growth hormone antibodies in patients treated with human growth hormone. J. clin. Endocr. Metab. 55, 13-17. Retegui L. A., Masson P. L. and Paladini A. C. (1985) Specificities of antibodies to hGH in patients treated with hGH. Longitudinal study and comparison with the specificities of animal antisera. J. clin. Endocr. Metab. 60, 184190. Retegui L. A., Milne R. W., Cambiaso C. L. and Masson P. L. (1982) The recognition by monoclonal antibodies of various portions of a major antigenic site of human growth hormone. Molec. Immun. 19, 865-875. Roth J. (1975) Methods for assessing immunologic and biologic properties of iodinated peptide hormones. In Methods in Enzymology (Edited by Colowick S. 0. and Kaplan N. O.), Vol. 37, part B, Peptide Hormones, p. 223. Scoffone E. and Fontana A. (1970) In Protein Sequence
Heteroclitic antibodies to growth hormone Defermination (Edited by Needleman S. B.), Vol. 18, pp.
185-210. Chapman & Hall, London. Solinger A. M., Ultee M. E., Margohash E. and Schwartz R. H. (1979) T-lymphocyte response to cytochrome c. I. Demonstration of a T-cell heteroclitic proliferative re-
123
sponse and identification of a topographic antigenic determinant on pigeon cytochrome c whose immune recognition requires two complementing major histocompatibility complex-linked immune response genes. J. exp. Med. 150, 83&848.