Production of monoclonal antibodies against human growth hormone releasing hormone and their use in an enzyme-linked immunosorbent assay (ELISA)

Journal of Immunological Methods, 94 (1986) 257-262 Elsevier

257

JIM04131

Production of monoclonal antibodies against human growth hormone releasing hormone and their use in an enzyme-linked immunosorbent assay (ELISA) * Ora Hirsch Pescovitz 1, Marie C. Gelato, Mary Bundy, D. L y n n Loriaux, George R. Merriam and Mark D. Pescovitz 2 1 Department of Pediatrics (OHP) and 2 Department of Surgery (MDP), University of Minnesota, Minneapolis, MN 55455, and Developmental Endocrinology Branch, National Institute of ChiM Health and Human Development and Immunology Branch, National Cancer Institute, National Institutes of Health, Bethesda, MD 20892, U.S.A. (Received 16 April 1986, accepted 8 July 1986)

Two murine monoclonal antibodies (mAbs) specific for human growth hormone releasing hormone (GHRH-44-NH2) were produced from a fusion of spleen cells from a BALB/c mouse immunized with GHRH-conjugated BSA with SP 2/0 myeloma cells. The antibodies were of the IgGlr, and IgG2b-x isotypes. The binding of both antibodies to GHRH-coated plates was inhibited by a 30-44 amino acid fragment but not by a 1-26 fragment. Thus, both antibodies are directed against the carboxy terminus of the peptide. Furthermore, both antibodies bind to the same epitope on the 30-44 amino acid portion since they cross-inhibit each other's binding to intact GHRH. Using these mAbs, a direct binding GHRH enzyme-linked immunosorbent assay (ELISA) was developed which had a least detectable dose of 0.3 pg. The availability of these antibodies and their use in ELISA methodology permits consistent and specific detection of GHRH in a non-isotope assay. They should prove of value in screening acromegalic patients for ectopic sources of GHRH secretion and in studies of ontogenic analysis of GHRH production. Key words: Growth hormone releasing hormone; Monoclonal antibody; ELISA

Introduction

Human growth hormone releasing hormone (GHRH), a 44 amino acid polypeptide that stimulates the release of growth hormone, is the most recent of the hypothalamic releasing hormones to be characterized and sequenced (Guillemin et al., 1982; Rivier et al., 1982). It has been

* Supported in part by grants from the Viking Children's Fund (OHP), the Minnesota Medical Foundation (OHP) and the University of Minnesota Graduate School (OHP).

shown to be of diagnostic value in studies of short stature (Gelato et al., 1986), obesity (Williams et al., 1984), and acromegaly (Thorner et al., 1984). It has also been reported to be of therapeutic value in the treatment of growth hormone deficiency (Thorner et al., 1985; Gelato et al., 1985b). Determination of plasma GHRH levels would be valuable in screening new acromegalics, some of whom may have ectopic GHRH production. We report here the development and characterization of two monoclonal antibodies to GHRH and demonstrate their usefulness in a non-radioisotope assay using ELISA technology.

0022-1759/86/$03.50 © 1986 Elsevier Science Publishers B.V. (Biomedical Division)

258 Methods

of incubation at room temperature, excess biotin was removed by dialysis against PBS.

Immunization and establishment of hybridoma lines G H R H , purchased from Bachem Fine Chemicals, (Torrance, CA) was conjugated to bovine serum albumin (BSA) using carbodiimide (Goodfriend et al., 1969). Female B A L B / c mice, 8-10 weeks old, were immunized ip with 100 #g of the G H R H - B S A conjugate in complete Freund's adjuvant on day 0. They were boosted ip on days 7 and 14, and iv on day 37, with 100 /~g of the conjugate in saline. On day 40, splenocytes were fused to the SP 2/0.Ag.14 myeloma cell line as previously described (Pescovitz et al., 1984). Hybrids were selected in hypoxanthine-aminopterinthymidine medium. Growth positive wells were screened for anti-GHRH antibody by binding to G H R H in an ELISA described below. Hybridomas of interest were cloned at 0.5 cells per well on irradiated rat fibroblast feeder layers.

Immunogloblin (Ig) class The Ig class was determined by the Ouchterlony double diffusion technique with classspecific anti-mouse Ig purchased from Litton Bionetics (Kensington, MD).

Purification and biotin conjugation of mAb mAb 3-2-2 was purified by adsorption to a column of protein A-Sepharose (Pharmacia Fine Chemicals, Piscataway, N J) and eluted with 3 M KSCN. mAb 3-3-7 was purified by adsorption to a column of pig anti-mouse Ig-coupled Sepharose (kindly provided by Dr. J. Bluestone, NIH, Bethesda, MD) and eluted with 1 M KSCN. Both eluates were then dialyzed against phosphatebuffered saline (PBS) before being concentrated to 0.5-1.0 m g / m l in an Amicon ultrafiltration unit (Amicon Corporation, Lexington, MA) with a PM30 filter. Protein concentration was determined by measuring OD280 with an extinction coefficient of 1.4. The purified proteins were conjugated to biotin by using the method of Jackson et al. (1982). Briefly, the protein was dialyzed into 0.1 M NaHCO3; 125 /tl of 1 m g / m l (+)-biotin-n-hydroxysuccinimide ester (Calbiochem, San Diego, CA) in dimethyl sulfoxide was then added for each ml of a 1 m g / m l protein solution. After 4 h

Epitope binding analys& ELISA plates were coated with 200 #1 of 0.1 # g / m l G H R H at 4°C overnight. The G H R H was diluted in PBS at pH 7.4. Serial dilutions of the culture supernatant from the two anti-GHRH mAbs, or Ig-isotype-matched control antibodies, in 200/~1 were added to the plates. Without washing 20 /~1 of diluted biotin-conjugated 3-2-2 or 3-3-7 were added to the appropriate wells. After a 1 h incubation at room temperature, the plates were washed three times. Avidin-coupled horseradish peroxidase was added for 15 min at room temperature. Finally, the plates were again washed, substrate was added, and the plates read in the Titertek Multiskan ELISA reader at 492 nm.

,nAb screens Nunc immunoassay plates were prepared by an overnight incubation with 200 ~tl/well of a 0.1 /~g/ml G H R H solution. The plates were washed three times with PBS Tween (0.05%) immediately before use. For the initial fusion screen and cloning screen, 200 /~l/well of hybridoma culture supernatant or 200 /~l/well of two-fold serial dilutions of mAb in PBS-Tween were added to the plate for a 1 h incubation at room temperature. The plates were again washed three times with PBS Tween. 200 #l/well of peroxidase-coupled goat anti-mouse Ig (Cappel Laboratories, Cochranville, PA) at a dilution of I : 750 was added and allowed to bind for 1 h at room temperature. After a final wash, 200 /~1 of the substrate ophenylene-diamine (0.10 m g / m l distilled water and 2.5 /~l/ml of 30% hydrogen peroxide) were added to each well. The reaction was stopped after 1 h with 8 N H2SO 4 and the optical density (492 nm) of each well was determined with a Titertek Multiskan ELISA reader.

Plasma extraction Plasma was extracted by modification of a previously described method (Frohman, 1986). Briefly, 5 ml of plasma were acidified with TFA to a final concentration of 0.2 M TFA and extracted over a C-18 Sep-Pak Cartridge (Waters Assoc.) with successive 4 ml washes of 0.01 M TFA, 80%

259 A C N / 2 0 % 0.01 M T F A ( v / v ) and 0.01 M TFA. The extract was lyophilized and then reconstituted in 1 ml of PBS buffer. This extraction decreased the non-specific binding presumed to result from plasma proteins and concentrated the samples five-fold.

800 ' 600 '

OD(nm)

400

200

GHRH assay G H R H standard curves were developed using G H R H dosed into both PBS-Tween buffer and into pooled blood bank plasma, which was then extracted as described above. Serial dilutions of these standards were made from 200 n g / m l down to 10 p g / m l . 100/~1 of each of these was added to each well of a Nunc immunoassay plate. 100/~1 of neat mAb 3-2-2 or 3-3-7 culture supernatant were added to each well and allowed to incubate for 1 h at room temperature. The plates were washed three times as above and 200 /xl of a 1:4000 dilution of peroxidase-conjugated goat anti-mouse was added for 1 h at room temperature. The plates were again washed three times and 200 /~1 of substrate o-phenylene-diamine (0.4 m g / m l citrate buffer and 2.5/xl of 30% H202) were added per well. The reaction was stopped with 8 N H2SO 4 after five minutes.

Results Fusion, screening and cloning of hybridornas Two B A L B / c mice were immunized with G H R H - c o n j u g a t e d BSA in CFA. Prior to fusion, sera from each mouse were tested for a n t i - G H R H antibodies in the ELISA binding assay. Both animals had serum binding titers of 1:640. The spleen of one mouse was fused to the SP 2/0.Ag.14 myeloma line. All 72 growth positive wells were screened for a n t i - G H R H m A b production. Of the 72 wells, three were positive. Two of these, 3-2-2 and 3-3-7, remained stable during the cloning.

0

....................... 10

100

Antibody Dilution

:... 1000

1400 ] 1200 t 1000 1 800 t OD(nm) coo t 400 t

20~! b

' 1000

Antibody Dilution

Fig. la: Competitive inhibition of the biotin-conjugated mAb 3-2-2 by unconjugated mAb 3-2-2 (closed circles) and unconjugated mAb 3-7-7 (open circles), b: Competitive inhibition of the biotin-conjugated mAb 3-3-7 by unconjugated mAb 3-2-2 (closed circles) and unconjugated mAb 3-3-7 (open circles).

Localization of the mAb binding site on the GHRH polypeptide To identify the site on the G H R H molecule to which the mAbs bind, m A b binding to a G H R H 1-44-coated ELISA plate was inhibited by various G H R H fragments. Complete inhibition of binding of both 3-3-7 and 3-2-2 was seen with G H R H 30-44. There was no inhibition of binding by the 2OOO

1500 OD(nm) lOO0 500

Immunoglobulin class Ig class was determined by Ouchterlony double diffusion on eight times concentrated hybridoma culture supernatant, m A b 3-3-7 was determined to be IgG1-K and 3-2-2 was IgG2b-x.

10

100 GHRH(pg)

1000

10000

Fig. 2. GHRH standard curves. Serial dilutions of GHRH in buffer are in the open circles and serial dilutions of GHRH extracted from plasma are in the closed circles.

260 1-26 G H R H fragment. The reactivity of both mAbs was therefore mapped to the carboxy terminus of the G H R H polypeptide. There was also no inhibition of binding with the 1-44 polypeptide. In fact, even further binding of antibody occurred. This resulted from the very high affinity of G H R H 1-44 for the immunoassay plates. Even in the presence of up to 1% Tween and coating with an irrelevant protein such as ovalbumin, the Ag-Ab complex bound to the plate. It was this feature that made the direct binding assay possible. Further confirmation of the related binding site of the two mAbs is shown in Fig. 1. Utilizing, the biotinylated antibodies, both mAbs completely inhibited the binding of themselves and each other. There was no inhibition of this binding with two irrelevant mAbs. Thus, both mAbs, 3-2-2 and 3-3-7 react with the same epitope on the 30-44 portion of the G H R H molecule. Cross-reaction with other human polypeptide hormones No inhibition of binding of either mAb to GHRH-coated plates was observed with up to 10 /~g/ml of thyroid stimulating hormone, thyrotropin releasing hormone, human placental lactogen, human chorionic gonadotropin, prolactin, somatostatin or human growth hormone (data not shown). Measurement of G H R H in plasma Recovery of G H R H from plasma following extraction averaged 80-90%. Sample standard curves for G H R H in buffer and in extracted plasma using the neat monoclonal antibody are seen in Fig. 2. The least detectable dose was 30 pg in both PBS and extracted plasma. The intra- and inter-assay coefficients of variation at the EDs0 were 9.9% and 15.7%, respectively. Equal sensitivity was achieved with both 3-2-2 and 3-3-7.

Discussion

We report here the development of two monoclonal antibodies against human G H R H . The development of monoclonal antibodies to rat G H R H was reported by Luben et al. (1982). We know of

no previous reports on the development of monoclonal antibodies against human G H R H . The antibodies are both directed against the carboxy terminus of the polypeptide. This was demonstrated by inhibition of binding to intact G H R H by the carboxy terminal fragment. The absence of inhibition by the amino terminus adds further support to this although no positive control for the inhibition by 1-26 was available. Originally, we were perplexed by the lack of inhibition and the actual excessive binding we observed with the intact 1-44 peptide. However, further study lead us to conclude that this was due to the extreme affinity of the intact polypeptide for the plates. It was this feature which ultimately was crucial in our ability to develop an assay. In addition to binding to the same portion of the molecule the two mAbs bind to the same epitope of the G H R H molecule as demonstrated by the cross-inhibition of binding. This is perhaps not surprising in light of the short region (14 amino acids) to which they bind. The two antibodies are clearly not sister clones as they are of different Ig classes. It is unknown if they use the same variable region. A polyclonal antibody raised in rabbits also recognizes the carboxy terminus portion of the G H R H molecule (Gelato et al., 1985a). Table I (a and b) summarizes the published structures of human (B'ohlen et al., 1983a), rat (Spiess et al., 1983b), ovine (Ling et al., 1984), bovine (Esch et al., 1983) and porcine (B'ohlen et al., 1986) growth hormone releasing factors and suggests one possible explanation for the predominance of carboxy terminal reactivity. There is a high degree of interspecies homology in the first 1-30 amino acids, while the 30-44 terminus is more divergent. Therefore, the carboxy terminus is more likely to be immunogenic. Alternatively, the three-dimensional structure of G H R H might preferentially expose the carboxy terminus to antigenic recognition. Either factor would make the generation of amino terminus specific mAbs more difficult. Using these mAbs we have developed an ELISA for h G H R H . Initially an attempt was made to develop a competitive inhibition assay in which the unknown samples or those containing known amounts of G H R H were preincubated with the monoclonal antibody prior to transfer to a

261 TABLE Ia COMPARISON OF HUMAN, RAT, OVINE, BOVINE AND PORCINE GHRH CARBOXY TERMINUS 30 Human Rat Ovine Bovine Porcine

31

32

33

34

35

36

37

38

39

40

41

42

43

44

Gin- Gin- Gly- Glu- Ser- Asn- Gin- Glu- Arg- Gly- Ala- Arg- Ala- Arg- Leu-NH 2 Gin- Gin- Gly- Glu- Arg- Asn- Gin- Glu- Gin- Arg- Ser- Arg- Phe- Asn-OH Gin- Gin- Gly- Glu- Arg- Asn- Gin- Glu- Gin- Gly- Ala- Arg- Val- Arg- Leu-NH 2 Gin- Gin- Gly- Glu- Arg- Asn- Gin- Gly- Gin- Gly- Ala- Lys- Val- Arg- Leu-NH 2 Gin- Gln- Gly- Glu- Arg- Asn- Gin- Glu- (;In- Gly- Ala- Arg- Val- Arg- Leu-NH 2

TABLE Ib NUMBER OF AMINO ACID DIFFERENCES BETWEEN HUMAN AND RAT, OVINE, BOVINE AND PORCINE GHRH AT 1-29 AND 30-44 REGIONS

Human vs. rat Human vs. ovine Human vs. bovine Human vs. porcine

1-29

30-44

8 0 1 0

7 3 5 3

Acknowledgements We acknowledge the excellent support a n d assistance of Drs. Joan L u n n e y , Jeffrey Bluestone a n d David Sachs a n d the excellent technical assistance of Ms. D i a n a H e r n a n d e z .

References G H R H - c o a t e d plate. It was anticipated that the G H R H in the sample would compete with the G H R H - c o a t e d o n the plate for b i n d i n g with the m A b . U n f o r t u n a t e l y , as m e n t i o n e d above, the G H R H b o u n d so avidly to the plates or tubes in which the initial p r e i n c u b a t i o n was performed that u n m e a s u r e a b l e quantities were actually transferred. Therefore, we took a d v a n t a g e of this peculiarity to develop a direct b i n d i n g assay. The assay is of similar sensitivity to previously described h G H R H r a d i o i m m u n o a s s a y s ( F r o h m a n et al., 1984; A u d h y a et al., 1985; G e l a t o et al., 1985b). A n E L I S A offers several advantages over radioimmunoassays. First, this sytem does n o t use radioactive isotopes. Second, the tagged second antib o d y is less expensive a n d more stable t h a n a radioactive trace. Third, the assay can be performed in 1 day as c o m p a r e d to the 3 - 5 days n e e d e d for the most sensitive RIAs. W e conclude that these m o n o c l o n a l antibodies, which are specific for h u m a n G H R H a n d are available in u n l i m i t e d supply, will be useful in the developm e n t of future assays as well as i n biological a n d i m m u n o h i s t o c h e m i c a l studies.

Audhya, T., M.M. Manzione, T. Nakane, N. Kanie, J. Passarelli, M. Russo and C.S. Hollander, 1983, Proc. Natl. Acad. Sci. U.S.A. 82, 2970. B'ohlen, P., P. Brazeau, B. Bloch, N. Ling, R. Gaillard and R. GuiUemin, 1983a, Biochem. Biophys. Res. Commun. 114, 930. B'ohlen, P., F. Esch, P. Brazeau, N. Ling and R. Guillemin, 1983b, Biochem. Biophys. Res. Commun. 116, 726. Esch, F., P. B'ohlen, N. Ling, P. Brazeau and R. Guilleman, 1983, Biochem. Biophys. Res. Commun. 117, 772. Frohman, L.A. and T.R. Downs, 1986, Methods Enzymol. 124, 371. Frohman, L.A., J.L. Thominet, C.B. Webb, M.L. Vance, H. Uderman et al., 1984, J. Clin. Invest. 73, 1304. Gelato, M.C., R.S. Rittmaster, O.H. Pescovitz, R.D. D'Agata, W.E. Nixon, D.L. Loriaux and G.R. Merriam, 1985a, J. Clin. Endocrinol. Metab. 61,223. Gelato, M.C., J.L. Ross, D. Malozowski, O.H. Pescovitz, M. Skerda, F. Cassorla, D.L. Loriaux and G.R. Merriam, 1985b, J. Clin. Endocrinol. Metab. 61,444. Geiato, M.C., S. Malozowski, M.C. Nicoletti, J.R. Ross, O.H. Pescovitz, S. Rose, D.L. Loriaux, F. Cassorla and G.R. Merriam, 1986, J. Clin. Endocrinol. Metab. 63, 174. Goodfriend, T.L., L. Levine and G.H. Fasman, 1969, Science 144, 1344. Guillemin, R. P.Brazeau, P. B'ohlen, F. Esch, N. Ling and W.B. Wehrenberg, 1982, Science 218, 585. Jackson, S., J. Sogn and T.J. Kindt, 1982, J. Immunol. Methods 48, 299. Ling, N., F. Esch, P. B'ohlen, P. Brazeau, W.B. Wehrenberg and R. Guillemin, 1984, Proc. Natl. Acad. Sci. 81, 4302.

262 Luben, R.A.P. Brazeau, P. B'ohlen and R. Guillemin, 1982, Science 218, 887. Pescovitz, M.D., J.K. Lunney and D.H. Sachs, 1984, J. Immunol. 133, 368. River, J., J. Spiess, M. Thorner and W. Vale, 1982, Nature 300, 276. Spiess, I., J. Rivier and W. Vale, 1983, Nature 303, 532.

Thorner, M.O., L.A., Frohman et al., 1984, J. Clin. Endocrinol. Metab. 59, 846. Thomer, M.O., J. Reschke, J. Chitwood, A.D. Rogol, R. Furlanetto, J. Rivier, W. Vale and R.M. Blizzard, 1985, N. Engl. J. Med. 312, 4. Williams, T., M. Berelowitz, S.N. Joffe, et al., 1984, N. Engl. J. Med. 311, 1403.