Indian Journal of Experimental Biology Vol. 37, July 1999, pp. 650-654

Monoclonal antibodies against bovine growth hormone

Pradeep Kumar & Yudhishthir S. Raj put*

Animal Biochemistry Divis ion, National Dairy Research Institute, Karnal 132 00 I. Indi a

Received 9 November 1998; revised 17 March 1999

A number of mouse x mouse hybridomas producing monoclonal antibodies (MAbs) against bovine growth hormone (bGH) were prepared by fusion of spleen cells from bGH-primed mice (Balb/c) with non-secretory mouse myeloma cells (PAIOP3) and characterized. MAbs obtained from three fusion experiments belonged to IgM, IgG I and IgG2b class/subclass of antibodies. Cross-reaction studies indicated that generated antibodies were against three di fferent epitopes of bGH. VI A6E8 (lgG 1) and VlIB2 E II C9 (lgM) did not cross-react with ovine prolactin (oPRL), ovine leutinizing hormone (oUn and porcine fo llicle stimulating hormone. Antibody VIB3C9E8 (lgM) exhibited cross-reaction with oPRL and oLH. Antibody VI C I F9 (lgG2b) cross reacted with oPRL. All MAbs were against conformati,onal epitopes ofbGH.

Bovine growth hormone (bGH) is a single chain polypeptide of 191 amino acids produced by anterior pituitary gland!. It is a single domain molecule that folds into a four antiparallel a -helix bundle2

. bGH has been completely sequenced' and it has two disulphide bridges, one connecting cystine residues at position 53 to 164 and other connecting cystine residues pos ition 181 to 1894

. These bridges configure the molecule into two protein loops. The major immunologically active zones are confined to 87-124 and 125-149 regions5

. Somatogenic as well as lactogehic activities are present in 22 kD bGH. Although molecular events how bGH triggers these activities are not completely elucidated, hut it has been observed that exogenously introd uced bG H results in enhanced level of insulin like growth factors I (lGF-I )' , a molecule central to somatogenic as well as lactogenic response. These activities are not located in one portion of molecule suggesting distant amino acid residues might be involved in creating a site through which bGH binds to its receptor. The involvement of distant amino acid res idues in an epitope is also known and an ep itopic site may or may not overlap the site which binds to receptor. The antibodies can modulate activities of bGH depending upon to which epitopes these bind . With the advent of deve lopment of protocol for monoclonal antibodies (MAbs) production of pre­defined spec ificity in 197? by Kohler and Milste in6

,

' Correspondent author fax No. : 0184-250042 [ -mail : NDRI@ X400.NICGW.NI C IN

MAbs against different epitopes of bGH can be generated. In the present investi gation, it has been shown with the help of generated MAbs against bGH that there are atleast three epitopes in bGH and all of them being conformational epitopes.

Materials and! Methods Balb/c mice and mouse mye loma ce ll line PAIOP3

(non-secretory) were obtained from Institute of Microbial Technology, Chandigarh . bGH was kindly prov ided by USDA-Animal Hormone Programme, USA.

Immunisation ofmice-Bal b/c mice (2 months old) were immunised intraperitoneally. Equal vo lumes of bGH (100 /-lgi IOO /-ll PBS) and Freund's adjuvant were mixed. In each immunisation , I 00 ~lg bGH was ad mini stered. The first inj ection was given in Freund's complete adjuvant (FCA) followed by two injection in Freund's incomplete adjuvant (FIA) at 15 days interva ls. For booster injecti on, bGH ( I 00 ~lg)

disso lved in 200 /-ll PBS was given intraperitoneal ly for three consecutive days before removal of spleen for ce ll fusion .

Preparation of hybridoma-The ce ll fusion s were performed according to standard method as described by Ga lfre and Mil stein 7

. The sensi tized mouse spleen cells were fu sed with PAIOP3 mye loma cell s in 10 : I rati o using 50% polyethyleneglyco l (Mw 4000 dalton) . After fusion , the ce ll s were suspended in a se lective growth medium contain ing hypoxanthi ne, aminopterin and thymidine (HAT ) and were cul tured in 24 we ll plates at 37°C in humidifi ed 5% C01195%

KUMAR & RAJPUT MO OCLONA L ANTrBOOIES AGA INST BOVI E GROWTH " ORMO E 65 1

air. The hybrids were regul arly observed under in verted mi croscope and based on ce ll den sity, medium was repleni shed with HAT on around day 7. The cells were a llowed to grow in HAT medium for at least 10 days before these are shifted to HT medium. Afte r 1 week of growth of cell s in HT mediu m, ce ll s were shifted to normal medium. At about ha lf con fluency, supernatants from grow ing hybridomas were screened for the presence of antibody us ing enzyme-I inked-immunosorbent-assay (E LI SA). The hybridomas producing anti-bGI-1 antibody were cloned by limiting dilution method where peritonea l ce ll s were used as feeder-layer. The clones were screened for anti-bGH antibodies in ELI SA . C lones tested pos iti ve were se lected and recl oned if necessary. One such se lected hybridoma was expanded in 25 cm2-t issue culture fl ask and ce ll suspended in PBS were injected to pristane or FIA­primed Balb/c mouse for asc ite productions.

ELISA- ELI SA was used for screenlllg of hybridomas 96-well s fl at bottomed polystyren e plates (Costar, US A) were used in ELI SA and method desc ribed by Engva ll and Perlman9 with some modificati ons was adopted . Brie fl y. multiwell plates

were coated \\ it h bG H ( 1Ilg/ I OO ~t1 PBS pH 7.4/ \.\ cll)

by overn ight i ncubat ion at 4°C. A II subscqucnt in cubati ons wcrc ca rri ed out at room temperature (RT) wi th mil d shaking on orbi tal shaker. The plates 'vve re washed wi th PBS-Tween 20 (0 .05% ) ::lIld then bloc ked whic h was ach ie'v'ed by compl cte ly fi ll ing well s with block in g so lution ( I % BSt\ -PI1S-T\'vee n 20) fo r 2 hr. Pl ates \\ erc washed fnur times with PB S­Twcen 20 and then I 00 ~t1 of hybridoma su pernatant \\ as added to different wel ls of ELISA plate. Thc plates were in cubated fo r 2 hr and then washed four times wiih PB S-Tween 20 . I 00 ~t1 of I: 1000 diluted rabbit anti-mouse IgG-perox idasc conj ugate (Banga lorc (jene i) in 1% BSA-PBS-Tween 20 was added to indi vidual we ll and plates incubated for 2 hr. The pl ates wen.: washed fi ve times with PBS-twee Jl 20.Subsequentl y 100 III o f substrate so lution (4 mg 0 -

ph eny lenediam ine hydrochl oride di sso lved in 10 1111 o f 50 m M sodium c itrate bufrer p H 5.0 conta ining 0.0 I % H cO~ was added to indi vidua l well s. Aft er 30 min .. th e reacti on was termin ated by addition of 100 pi 4N H ~SO , to each we ll. Absorbance was recorded at ~20 nm in ELISA pl atc rcadcr (Flow Lab. USA) .

1.I'()l)IJillg--C lasscs and subc lasses of antibodics from cultllre supern atall t \\CIT deterillined by uSIng

mouse monoc lonal sub-i sotyping kit from Hyc lone laboratories , USA and in structi ons gIven by Illanufacturer were foll owed.

DOI-Blof- Dot-blot was done on nitro ce llulose (NC) membrane (Costar, USA). I ~t1 of prote in hormone d isso lved in PBS was applied to g lossy sid e of NC membrane. All incubations were carri ed out at RT with mild shaki ng on orbital shaker. After a ir dry in g, NC membrane was pl aced in blocking so lution (3 % BSA-PBS-0 .05% Tween 20) for I hr. The membrane was washed once with PB S-Tween 20 and subsequently in cubated with e ither I: I d ilu ted hybridoma supernatant with 10 mM Tris-HCI, pH 8.0 containing 150 mM NaC I and 0 .05% Tween 20 (T BST) or diluted a citi c fluid (I :500 in TBST) for I hr. Membrane was washed with PBS-Tween 20 (3 x 10 min ) and then incubated with rabbit anti-m ouse IgG-peroxidase conjugate ( I :500 diluted in 1% BSA­PBS-0 .05%. Tween 20) for 2 hr. NC membranes \\ cre washed first with PBS-Tween 20 (3 x 10 min ) and then with PBS ( I x 10 min. ). Co lour was deve loped by incubating membrane in substrate so luti on (6mg d iaminobenzidine di sso lved in 10 ml or 50 mAl Tri s­HC!. ; IH 7.6 containin g 10 pi o f' 30% II ~O~ til l bro \\ n

hl(lt~ appeared on the mC!l1branes . Reaction \\ as termin ated by washin g membra nes \\ it h PBS . Mem branes were then air-d ri cd and photographed.

IVc.I'lcrn-h/ol- SDS-PACIE in slab !.!,e ls \\as carried out as pCI' the meth od of L.aen1Jn l i l o~ Di sco ntinu ous gc l was prepa red Ill/' mini dua l ve rti cle s lab ge l L1nit (M is Banga lore Genei). 4% stackin g ge l \va s laye red (we r 15% sepa ratin g ge lS. Mw-m arke/'s-samplc and bG11 wc re prepared by bo iling them for 5 min . in sa mple buffer (2% SDS. 5% f3-me rcaptoetllG no l and 10% glycero l). Other hormo ne and a lso bG H sa mplc we re prepared in sa mp le buller lac k i ng f3- mcrcapto­cthanol (f3- IE) without bo ili ng. Afte r e lectro­phores is. pro te in s fro m gel were transferred to !e lIIembrane using mini tran s-blot asse mbly from 8io­rad USA II. The porti on of membrane co ntaining Mw markers was cut and sta in cd with Poncea u SX. The remaining porti on of NC membrane was bl oc ked with BSA- PBS Tween 20 and separated prote in hormones wcre visua li sed as per th e meth od desc ribed f(l r dot­blot.

Results From three fu s ion expcrim ent s, 188 \\ell s out l, r

266 well s from 24-we ll s pl ates shO'v'ved grO\\ til or fused ce lls co rrespondin !.!, to 70.7% fu sion crfic ien..:y .

652 INDIAN 1. EXP. BlOL., JULY 1999

Screening of hybridoma supernatant showed that initially 26 out of f88 hybrids were tested positive against bGH in ELISA. Some of these hybrids grew ' poorly or lost the ability to secrete antibodies on subsequent culturing. Nine hybrids tested positive on sub-culturing _were cloned. It was observed that cloning of five hybrids did not result in production of cloned hybridoma producing antibodies against bGH. However, cloning of four hybrids resulted in 11 cloned cells producing anti-bGH antibodies (Table 1).

The class or sub-class of antibodies produced by above hybridomas is shown in Table 1. The antibodies belonged to IgGJ, IgG2b and IgM class/subclass. The relative signal in ELISA as an index for production of anti bGH-antibodies is also shown in Table I. The signal in ELISA in supernatant of hybridomas is rather high indicating high level of antibody production. Since cloning of fused cells obtained from one particular well of 24-wells plate resulted in cloned cells producing antibodies of same class or subclass, it is likely that these could be

CO W <.D <! ..... >

bGH

Reduced bGH

oPRL

oLH

pFSH

Iodination grade bGH

identical clones. This allowed to select VIA6E8, \ VIB3C9B8, VIC I F9 and VIIB2E 11 C9 hybridomas for studying cross-reaction. The result of cross

reaction of VIA6E8, VIB3C9B8, VIC 1 F9 and VIIB2E 11 C9 MAbs are shown in dot blot (Fig. I) and Table 2. Two MAbs namely VIA6E8 and VIIB2E 11 C9 did not cross-react with ovine prolactin

Table I---{::Iass or subclass and relative signals of antibody produced by hybridomas

Name of hybrid om as Class/Subclass A420m~

VIA6D4 IgG , 1.585 VIA6E8 IgG I 1.700 V183C9B8 IgM 2.232 V183C9FII IgM 2.238 VICIB4 IgG2b 2.255 VICIBII IgG2b 2.201 VICID2 IgG 2b 2.213 VICID9 IgG 2b 2.151 VICIF9 IgG2b 2.188 VI182EIIC9 IgM 0.768 V182D8D5 IgM 1.133

"

........ 1:J :J

LL u .~ u en lfl CO U <! CO -- .... CO en en W W U li- N <.D (Y) .... CD <! CD U --.-.

> > > >

Fig. I-Detection of cross-reaction of MAbs using dot-blot. 2 Ilg of each hormone was spotted on nitrocellulose strips and blots were

developed using different MAbs

KUMAR & RAJPUT: MONOCLONAL ANTIBODIES AGAINST BOVINE GROWTH HORMONE 653

(oPRL), ovine leutinizing hormone (oLH) and porcine follicle stimulating hormone (pFSH). MAb VIB3C9B8 cross-reacted with oPRL and oLH but not with pFSH. VIC I F9 antibody cross-reacted with oPRL and not with oLH and pFSH. None of the MAbs reacted with bGH when its disulfide bonds were reduced with /3-ME. The possibility of minor contamination of growth hormone (GH) in oPRL and oLH preparations cannot be ruled out and low level of GH can be detected in dot-blot and then net result would be false cross-reaction. Thus, cross-reaction of VIB3C9B8 and VIC I F9 MAbs were also studied in Western blot where protein hormones (including impurities if any) were separated on the basis of their molecular weigh.ts and results are shown in Figs 2 and 3. Both these antibodies did not show cross-reaction with pFSH in both dot-blot and Western-blot. Similarly, these antibodies did not react in dot-blot and Western blot with bGH when its disulphide bonds were reduced. Blots in lane 2 and lane 4 from both figures (Figs 2 and 3) indicated that antibodies VIB3C9B8 and VIC I F9 truely cross-reacted with oPRL. However, in Fig. 3, an additional blot at position corresponding to dimer of oPRL could be observed indicating that VIC I F9 antibody reacted with dimer of oPRL. A faint additional band (not visible in Fig. 2) at similar position was also observed with antibody VIB3C9B8. The ant ibody VIB3C9B8 exhibited cross-reaction with oLH in dot-blot but not in Western blot.

Discussion Based on cross-reaction results, there appears to be

at least three epitopes against which antibodies were generated. VIA6E8 and VIIB2E II C9 did not cross­react with oPRL, oLH and pFSH and perhaps these antibodies could be against identical epitope. VIB3C9B8 and VIC I F9 antibodies are against different epitopes as cross-reaction results were not

Table 2--Cross-reaction of MAbs in dot-blot

Hybridoma Hormones tested bGH bGH* bGH** oPRL oLH pFSH

VI A6 E8 + + VI A6 E8 (ascite) + + VI B3 C9 B8 + + + + VI CI F9 + + + VII B2 El l C9 + +

*Iodination grade

**Reduced

2 J 4 5 6

2 3 4 5 6

Fig. 2-Detection of cross-reaction of VII33C9B8 using Western­blot. 5 ~g of each hormone was separated on 15% SDS-PAGE and transferred to nitrocellulose membrane. Lane I. Mw markers (97.4, 66. 45. 31 and 21.5 kD). Lane 2. bGH. Lane 3. reduced bGH. Lane 4. oPRL. Lane 5 oLH. Lane 6. pFSH

Fig. 3-Detection of cross-reaction of VIC I F9 using Western­blot. 5 ~g of each hormone was separated on 15% SDS-PAGE and transferred to nitrocellulose membrane. Lane I. Mw markers (97.4 , 66. 45. 31 and 21.5 kD). Lane 2. bGH. Lane 3. reduced bGH. Lane 4. oPRL. Lane SoLi-I. Lane 6. pFSH

identical in dot-blot (Fig. I) for the all the hormones tested. All the four antibodies were against conformational epitopes as all failed to react with denatured bGH (Fig. I). Conformational epitopes in bGH are also shown by other workers 12 and our results support this view. It has been reported that bGH maintained a high degree of secondary structure

without intra-chain disulfide bond13. Perhaps, minor alterations in secondary structure may result in distortion in epitopes. On the other hand report also exists where it has been shown that, on reduction of

654 INDIAN 1. EXP. BlOl , JULY 1999

disulphide bond In bGH , antigenic determinants are not lost5 and perhaps, these could be sequential epitopes. YIB3C9B8 antibody cross-reacted with oLH in dot-blot but not in Western blot. This is not unlikely as oLH in Western blot during electrophoresis comes in contact with SDS which may change conformation of oLH .

References I Prosser C G & Mepham T B in Use oj somatotropin in

livestock production, edited by K. Sejrsen , M. Yestergaard. and A. Neimann-Sorensen (Elsevier App ~ ied Science. London) 1989, I.

2 Abdel-Meguid S S Shieh H S Smith W W Dayringer II E Yioland B N & Bentle L A, IJroc Natl Acad Sci, 84 ( 1987) 6434.

3 Lehrman S R Tuls J L & Lund M. Biochemi.l'lIy, 29 ( 1990) 5590.

4 Have l H A Chao R S Hastell R J & Thamann T J. Anal ('hem. 6 1 (1989) 642.

5 Ferrara P Zakin M M Pena C & Paladini !\ C. EliI' J Im lllunol. 9 (1979) 1020.

6 Kohler G & Mi lstei n C, Nature, 256 (1975 ) 495 . 7 Galfre G & Milstein C. Met;' En::,l'f1/ol 73 (1981) I.

8 Ilariow E & Lane D, in Antibodies : A laboratory manual (Co ld Spring Harbor Laboratory. New York) 1988, 245.

9 Engvall E & Perlmann P. immunochemistr)" 8 ( 1971 ) 871. 10 Laemmli U K. Nature, 227 (1970) 680. II Towb in H Staehelin T & Gordon J, Proc Natl Acad Sci 72

( 1979) 4350. 12 Secchi C Beliini A l3orromeo Y Gamba D & I3rocchi E,

Hy bridoma. 10:4 ( 199 1) 499. 13 Holzman T F Brems D N & Doughertly J Jr, Biochemistry, 25

(1986) 6907.