- Email: [email protected]
Selectivity of angiotensin II antisera
Journal of Immunological Methods, 56 (1983) 85-96
85
Elsevier Biomedical Press
Selectivity of Angiotensin II Antisera Jiirg Nussberger, Gary R. Matsueda, Richard Re* and Edgar Haber** Cellular and Molecular Research Laboratory, Cardiac Unit, Massachusetts General Hospital and Harvard Medical School, Boston, MA 02114, U.S.A.
(Received 8 January 1982, accepted 28 June 1982)
A significant problem in the immunoassay of angiotensin II is the cross-reactivity of most available antisera with the peptide's metabolic products, (des-Asp I )-angiotensin I1 and (des-Asp',Arg 2)-angiotensin II. In order to attempt to generate antisera of greater selectivity, a variety of conjugates between angiotensin II or derivative peptides and carrier proteins were examined as immunogens with the aim of generating antisera that would selectively identify the amino terminal region of the peptide. Selectivity for the amino terminus was achieved by either (1) immunization with N-acetylated angiotensin ll-amide which had been coupled to rabbit serum albumin by its carboxy terminus, or (2) immunization with angiotensin-(I-7)-heptapeptide which was randomly coupled to thyroglobulin. The antisera produced with the N-acetylated immunogen cross-reacted with the unacetylated ligand (Asnl-ValS)-angiotensin, but did not recognize the human hormone (Asp',lleS)-angiotensin. Carboxy-terminal coupling of angiotensin without N-acetylation did not induce selectivity for the amino terminus, nor did a conjugate which was linked to the carrier protein via a diazo bond to His6 of the peptide. These findings may be explained by the fact that N-acetylated angiotensin I1 resists degradation by amino Imptidases and thus retains its structure in the immunogen and by the fact that the (I-7)-heptapeptide has lost the immunodominant carboxy-tcrminal epitope, thus emphasizing the desired amino terminal determinant. Key words: angiotensin 11 - - immunodominance - - immunospecificity - - unidirectional conjugation
Introduction A c e n t r a l p r o b l e m in the d e t e r m i n a t i o n o f p e p t i d e - h o r m o n e c o n c e n t r a t i o n in p h y s i o l o g i c fluids is the d i f f e r e n t i a t i o n a m o n g the p r e c u r s o r , the a c t i v e species, a n d t h e d e g r a d a t i o n p r o d u c t . A t times, c h a n c e a p p e a r s to p r o v i d e an a n t i s e r u m o f a p p r o p r i a t e specificity, b u t m o r e o f t e n t h a n n o t p h y s i o l o g i c o b s e r v a t i o n s are c o n f o u n d e d b y an i n a b i l i t y to c o r r e l a t e i m m u n o a s s a y w i t h b i o a s s a y b e c a u s e i n a c t i v e as well as a c t i v e p e p t i d e s a r e b o u n d to a n t i b o d y w i t h n e a r e q u a l affinity. T h i s p r o b l e m is p a r t i c u l a r l y a c u t e in the i n s t a n c e o f a n g i o t e n s i n II d e t e r m i n a t i o n . W h i l e a n t i s e r a * Current address: Ochsner Clinic, 1514 Jefferson Highway, New Orleans, LA 70121, U.S.A. ** Correspondence to: Edgar Haber, MD, Cardiac Unit, Massachusetts General Hospital, Boston, MA 02114, U.S.A. 0022-1759/83/0000-0000/$03.00 © 1983 Elsevier Biomedical Press
86
that show relatively little cross-reaction between the precursor decapeptide, angiotcnsin I, and the octapeptide, angiotensin II, are conventionally obtained by immunization with either material, angiotensin II antisera nearly always cross-react with 2 amino-peptidase-mediated degradation products: (des-Asp~)-angiotensin I! and (des-AspI,ArgZ)-angiotensin II. Because these products have important activity differences with angiotensin II, their independent measurement is of considerable significance. In this paper we present a systematic examination of a variety of immunogens in an attempt to address the question of whether or not specific methods of coupling or orientation of angiotensin II in relation to a carrier molecule or the use of shorter peptides as haptenic determinants would enhance the ability of antisera to bind angiotensin II without significant cross-reaction with other peptides.
Methods
Synthesis of peptides General procedures used for Merrifield solid-phase synthesis have been described by Stewart and Young (1969). The reagents used were 25% trifluoroacetic acid in CH2C12 for deprotection, 10% triethylamine in CH2C! 2 for neutralization, and Boc-amino acid:dicyclohexylcarbodiimide (1 : 1) at a 3-fold excess in CH2CI 2 for coupling. Boc-amino acids of the L-configuration were purchased from either Bachem (Torrance, CA) or Peninsula Labs (San Carlos, CA). (a) Synthesis of (N~-trifluoroacetyl-Asp(NO, Bzl)l)-angiotensin II. Starting with 3.08 g of Boc-Phe-resin (0.13 mmol/g), protected angiotensin was assembled using a Beckman 990B synthesizer to give Boc-Asp(NO2-Bzl)-Arg(Tos)-Val-Tyr(benzyloxycarbonyl)-Ile-His(Tos)-Pro-Phe-OCH2-resin (3.88 g). Following deprotection of the a-Boc group, the neutralized resin was treated with 3 mmol of trifluoracetic acid anhydride in CH 2C12 for 20 min. Trifluoroacetylation was complete as evidenced by a negative result with the Kaiser ninhydrin reagents. HF treatment ( H F : a n i s o l e (9: 1) for 30 min at 0°C) yielded 650 mg of peptide. The peptide was purified by partition-column chromatography on Sephadex G-25 (2 cm x 45 cm) using nBuOH : HOAc : H 2 0 (4 : 1 : 5). The peptide eluted in a 17 ml volume after 72 ml of column effluent had been collected. The R F of the peptide on a silica gel (thin layer) was 0.45 with n-BuOH: HOAc: H 2 0 ( 1 2 : 3 : 5 ) as developing solvent. The amino acid composition of the purified peptide was within 6% of theoretical values. The amino acid sequence was confirmed by solid-phase Edman degradation of a peptidyl-resin sample which was treated previously with I M NaOH in EtOH to remove both the N-trifluoracetyl group and the p-NO 2Bzl group. (b) Synthesis of angiotensin-(1-7)-heptapeptide. Starting with 2.5 g of Boc-Proresin (0.08 mmol/g), Boc-Asp(Bzl)-Arg(Tos)-Val-Tyr(benzyloxycarbonyl)-IieHis(Tos)-Pro-OCH2-resin was assembled using a Beckman 990B peptide synthesizer. H F cleavage of the peptidyl-resin product yielded 164 ~mol of peptide (65% yield). After purification by partition-column chromatography as described above, 105 mg
87 of peptide was obtained which was homogeneous by TLC (R F = 0.42 with solvent system n-BuOH : HOAc: H 2 0 : pyridine (15 : 3 : 12 : 10)). The amino acid composition was within 8% of theoretical values after 6 M HCI hydrolysis for 24 h. (c) Synthesis of angiotensin-(3-8)-hexapeptide. Starting with 1.5 g of Boc-PheOCHE-resin (0.15 mmol/g), Boc-Arg(Tos)-Val-Tyr(benzyloxycarbonyl)-Ile-His(Tos)Pro-Phe-resin was assembled manually. HF treatment of 1.98 g of peptidyl-resin yielded 276 mg of the crude peptide which was subsequently purified by carboxymethyl-cellulose ion-exchange chromatography after partition chromatography failed to give a homogeneous peptide product. The purified peptide (57 rag) gave a single spot by TLC ( R r = 0.56 with n-BuOH : HOAc: H 2 0 (12 : 3 : 5)) and a satisfactory amino acid analysis which was within 4% of theoretical values.
Labeling of peptides The peptides were labeled with 125I by the chloramine-T method (Hunter and Greenwood, 1962).
Preparation of immunogens Peptides were not used as immunogens but were linked to carrier proteins employing several different methods. An amide linkage was formed using the carbodiimide (CDI) method as described previously (Goodfriend et al., 1964; Haber et al., 1965). Carboxy-terminal coupling of the peptide was obtained by protecting the N-terminal reacting groups prior to CDI treatment. (1) N-acetylation of (Asnl)-angiotensin. 17.44 /xmol of [AsnI,ValS]angiotensin II (Hypertensin Ciba), a gift of Ciba (Newark, N J), were suspended in 3 ml distilled water and 3 ml of saturated sodium acetate. With continuous stirring, 1.5 mmol of [laC]acetic anhydride (2.45 pCi per mmol) were added and the insoluble peptide dissolved. After 60 min the fluorescamine test (Hoffmann-LaRoche, Nutley, N J) gave blank values, indicating complete acetylation of the alpha amino group of the peptide. The unreacted acetic anhydride was separated from the peptide by gel filtration on Sephadex G-25 (medium). Paper chromatography (n-butanol:glacial acetic acid:water (12:3:5)) showed an R E of 0.7 for the reaction product as compared to 0.5 for the starting material. The reaction product was positive by the Pauly stain for histidine but negative when tested by the ninhydrin reaction. Its absorbance at 280 nm did not change at pH II indicating that the tyrosine of angiotensin II was not acetylated. Amino acid analysis confirmed that the product was a derivative of angiotensin II. Acetylated peptide (7.64 /xmol) and rabbit serum albumin (0.18 pmol) were dissolved in 2.5 ml of distilled water, and 2.61 mmol of 1-ethyl-3-(3-dimethylaminopropyl)carbodiimide hydrochloride (Sigma, St. Louis, MO) (CDI) were added. Immunoassays were carried out with 125I-labeled acetylhypertensin as tracer. (2) Na-trifluoroacetylation of (Asp(p-NO2Bzl)1)-angiotensin. The synthesis of (Na-trifluoroacetyl-Asp(NO2Bzl)t-angiotensin II is described above. This peptide (57.6 #tool) dissolved in 4 ml of dimethylformamide and 154 #mol bovine thyroglobulin were mixed in 12 ml of distilled water, and 10.4 mmol of CDI was added.
88 The reaction was allowed to proceed with shaking at room temperature for 20 h. Eighteen milliliters of 1.8 N N a O H were then added to remove the amino-terminal protecting group, and 7 min later 18 ml of !.86 N acetic acid was added to bring the pH to 6. When the protected peptide was treated alone, this brief alkaline treatment yielded a product which was indistinguishable from angiotensin 11 (by TLC). A diazo link was effected by coupling p-aminophenylethylamine (Aldrich. Milwaukee, WI) to the carrier protein using the carbodiimide method and then forming the diazonium salt, which was in turn coupled to the peptide (reacting at either histidine, tyrosine or amino groups) essentially as described by lnman (1974). This conjugate was synthesized by mixing 14.7 mmol of p-aminophenylethylamine (Sigma), 154 nmol of bovine thyroglobulin and 5.2 mmol of CDI in 3 ml 0.1 M NaCI, 0.01 M Na phosphate, pH 7.5. to which was added 2 ml of glacial acetic acid. The final pH of the reaction mixture was 6. The thyroglobulin derivative was isolated and purified as described above. Fifteen milligrams of this thyroglobulin derivative were dissolved in 1 ml of 0.01 N HCI at 0°C. Six milligrams NaNO2 were then added. 9.47 ~mol of angiotensin II (purchased from Peninsula Labs.) were dissolved in i.5 ml of 0.1 M Na2CO 3. Fifteen minutes later the peptide solution was added to the thyroglobulin solution and maintained at 0°C for 30 min. The pH of the reaction mixture was 9.
Purification and characterization of the immunogens The conjugates were dialyzed against 1.5 mM N a 2 H P O 4 containing 3 mM sodium azide for 24 h at room temperature. Coupling was monitored by measuring the incorporation of a tracer quantity of ~25I-labeled or 3H-labeled peptide added to the reaction mixture. Non-dialyzed tracer was assumed to be coupled peptide. The molar incorporation of peptide into the conjugate was calculated. The final determination of the incorporation of peptide was accomplished by amino acid analysis (Durrum 500 analyzer).
Immunization Immunogens were dissolved in isotonic saline at a concentration of 2 m g / m i and emulsified with an equal volume of complete Freund's adjuvant (containing 1 m g / m l killed tubercle bacilli). New Zealand White rabbits (1.5-2.5 kg) were injected intradermally, either in the toe pads or other body sites. The primary immunizing dose was 1 mg. Booster injections were administered at 2-4-week intervals using 0.5 mg of conjugate administered intradermaily (in complete Freund's adjuvant). Blood (40 ml) was collected weekly from the ear arteries beginning on the third week after immunization.
Characterization of antisera Immunoassays were carried out utilizing the dextran-coated charcoal method as described previously (Haber et al., 1969). All antisera were tested with the radiolabeled haptenic peptides. Titer was defined as the dilution of antiserum at which 50% of homologous tracer peptide was bound to antibody (3000 cpm representing 3 fmol ~2sI-labeled peptide). Sensitivity was defined as the reciprocal of the quantity of
89
unlabeled peptide that displaced 50% of the tracer bound to antibody in the absence of any inhibitor. Specificity of an antiserum in discriminating between two peptides was defined as the quotient of their respective sensitivities.
Results
The characteristics of the various angiotensin-carrier conjugates (I-VI) are shown in Table I. The highest coupfing efficiency was obtained with the diazo linkage: 89% of the peptide was linked to thyroglobulin resulting in a peptide-to-cartier ratio of 426. Although less efficient, the carbodiimide procedure was reliable, giving 50-80% coupling when the peptide-chloride salt was used, but only 10-30% coupling when peptide-acetate salt was used. Regardless of the coupling procedure, all conjugates elicited antibodies in all rabbits with titers ranging from 100 to 100,000. We believe that this success was due in part to thorough emulsification of the conjugate-adjuvant mixture and to the use of the hind footpad as a site of immunization. When a conjugate was insoluble in saline, ultrasonication was helpful in obtaining a fine suspension.
(I) Antisera against [Na-acetyl-Asnt, ValS]-angiotensin II rabbit serum albumin (A c-hypertensin-RSA ) Five rabbits (nos. 1800-1804) were immunized with Ac-hypertensin-RSA (conjugate I) and were boosted 4, 6, 8 and 12 weeks later. The sera reached titers between
o- --o (NoAcetyi-AsnI ValS)-An~ioten~n_(1_8).C)ctope~ide
501-
;
; (N-~..-AsP')-~s~-(t-.4)-Te.o~o~p~
-
_ _ -- ,-~- . . . .
40'-
-
"''-
30
-
;
"'-.
\~
20
10
~
I
I
t0 2
103
k'~ t - I 0
"~ ~
" "
.//-
///Sz_ 8 --~--- 1-7
~ \\
4-8
', \"
1
[
I
I
104
105
10s
107
CO~NI"/gAT/ON OF/NHIB/IOR(Mx "10-~ Fig. 1. Specificity of rabbit antiserum no. 1803 directed against (N=-acetyl-Asnt,ValS)-angiotensin.(I.8)octapeptide which was carboxy-terminal coupled to rabbit serum albumin. Cross-reactivity could be demonstrated only with (Asnm,VaiS)-angiotensin-(l-8)-octapeptide at a level of I%.
IN”-Acetyl-Asn’,Val’j angiotensin-( I-8)-octapeptide angiotensin-( I-8)-octapeptide angiotensin-( I-7)-heptapeptide angiotensin-( I-7)-heptapeptide angiotensin-( I-7)-heptapeptide angiotensin-(3-8>hexapeptide angiotensin-(2-8)-heptapeptide angiotensin-( I-l>octapeptide
I
’ Determined by amino acid analysis. b RSA = rabbit serum albumin. ’ Tg = thyroglobulin.
IV V VI
Ilk
II llla IIIb
Peptide
CONJUGATES
Conjugate number
ANGIOTENSIN-CARRIER
TABLE I
Tg Tg Tg Tg Tg Tg Tg
’ (bovine) (porcine) (bovine) (bovine) (porcine) (porcine) (bovine)
RSA h
Carrier
CD1 CD1 CD1 CD1 CD1 CD1 Diazo
CD1
Coupling method 2
Ill*24 47* 3 63+24 93+ 16 17+ 5 37* 3 426*46
12+
Moles peptide per mole carrier a
C-terminal C/N-terminal C/N-terminal C/N-terminal C/N-terminal C/N-terminal His/Tyr
C-terminal
Coupling through (site of peptide)
II
ANG b (l-8) ANG(I-8)
ANG( l-8)
ANG(I-8)
ANG(l-8)
ANG( l-7)
ANG(I-7)
ANG(I-7)
ANG(3-8) ANG(3-8) ANG(2-8) ANG(I-8)
II II
II
II
II
IIla
IIIa
IlIb
IIIC
IV IV V VI
I 826 1827
1828
1829
1830
1810
1811
1817
1820
I832 1833 1835 1837
’ PI, primary immunization. b ANG, angiotensin. ’ No cross-reaction ( < 0.01 I). d No displacement of tracer (sensitivity
ANG(I-7)
‘*’ I-labeled angiotensin analog
Conjugate
OF RABBIT
Serum number
CHARACTERIZATION
TABLE
2000 100 30000 100 4000 16000 50 4000 50 600 20 2500 IOOOO 20 800 1500 50 2500 25 3ooo 600 800 15wO 5000
Dilution
> 100,000 nM).
8 3 8 3 8 16 3 8 3 8 3 7 I5 3 7 I5 3 II 4 II 4 4 4 4
Weeks after PI ’
ANTI-ANGIOTENSIN
loo IO0 100 100 loo 100 IO0 100 100 100 61 50 4 60 40 2 75 1 80 75 I40 40 65 100
1-8
Specificity
3 44 56 29 46 75 18 53 3 5 nc’ 7 I nc 7 0.7 nc 0.1 nc 20 117 50 100 86
I
0.5
I
0.5
I
0.01 nc 10 100 100 50 61
nc
nc
nc
2 nc 77 2 38 II 0.4 67 nc
3-8
(% cross-reaction) 2-8
SERA
0.2
I
0.1
I
I 47 1 39 I I 41 1 65 I5 I5 13 0.2 24 I6 0.2 30 0.1 31 5
I-10
_
16 100 100 100 100 100 100 100 100 100 IO0
_ _
1 _ _
1-7
8 IO 2 42 4 42 10 0.8 0.50 0.15 0.17 1.9
I
0.05 80 0.1 0.43 0.02 0.02 2.2 0.18 20 2.8 33
l-8
Sensitivity
1.8 180 0.18 1.5 0.05 0.03 I2 0.34 670 62 nd d 7 30 nd II 150 nd 520 nd 3 0.60 0.12 0.11 2.20
2-8
(nM)
3 0.13 21 0.06 0.20 600 0.27 nd 280 nd 110 25 nd 150 105 nd 4200 nd 6 0.70 0.06 0.22 3.10
nd
3-8 8 170 I4 I.1 3.5 3.5 5.4 20.0 31 19 I30 4 140 25 5 450 IO 1100 26 II 70 62 22 900
l-10
_
I7 20 0.5 0.3 6 0.8 I 3 6 8 0.6
_ _
8
l-7
92 400 and 4000. They were tested for sensitivity and specificity beginning 9 weeks following primary immunization. Specificity of these (and subsequent) sera was tested utilizing various angiotensin analogs. Fig. 1 shows the very high specificity of serum (no. 1803); only hypertensin cross-reacted at a level of I%. Very similar characteristics could be demonstrated in the other sera of this group. Sensitivities for Ac-hypertensin between 0.1 and 2.0 nM were always associated with a cross-reactivity of 1-2% for unacetylated hypertensin, and no significant cross-reactivity was found with any other angiotensin peptide. Thus N-terminus specificity for hypertensin was achieved although acetylated hypertensin as the homologous hapten was clearly better recognized by the antisera. This result is consistent with both unidirectional carboxy-terminal coupling of the peptide to the carrier protein and with long persistence of the aminopeptidase-resistant immunogen.
(II) Antisera against angiotensin I1 coupled by its carboxy terminus to bovine thyroglobulin Table II characterizes 5 rabbit antisera (nos. 1826-1830) obtained by immunization (boosted 2 weeks after primary immunization) with angiotensin II that was linked to thyroglobulin by its carboxy terminus (conjugate II). These sera reached titers between 4000 and 40,000. One serum (no. 1828) was very sensitive to the octapeptide (0.02 nM) but exhibited 75% cross-reactivity with the (2-8)-heptapeptide and 11% cross-reactivity with the (3-8)-hexapeptide. The cross-reactivity with angiotensin I was 1%. This cross-reactivity pattern was characteristic for this group of antisera though there appeared to be more amino terminal specificity in early than in late antisera. (III) Antisera against angiotensin-(1- 7)-heptapeptide-thyroglobulin Three sets of 3 rabbits each (nos. 1809-11, 1817-19, 1820-22) were immunized with angiotensin-(l-7)-heptapeptide that was randomly coupled to thyroglobulin by carbodiimide (Table 1, conjugates Ilia-c). Porcine thyroglobulin was found to be more soluble than bovine thyroglobulin. The insoluble conjugates were successfully suspended with ultrasonication. Conjugate IIIb differed from lllc only by a lower peptide incorporation. Under identical conditions with booster injections 2 and 4 weeks after the primary immunization, conjugate Illb induced maximal titers of 30,000, 50,000 and 100,000; whereas lllc induced maximal titers of only 400, 2000, and 3000. The sera of all 9 rabbits showed similar characteristics. Representative examples of each group are demonstrated in Table II. In early sera following immunization with any one of conjugates Ilia, Illb or IIIc, there appeared to be preferential recognition of amino terminal determinants: cross-reactivity with angiotensin I and II was significant; whereas little or no cross-reactivity was found with the (2-8)-heptapeptide and the (3-8)-hexapeptide. In later sera, the cross-reactivity with peptides 2-8 and 3-8 increased, while the cross-reactivity with I-8 and 1-10 decreased, thus decreasing amino terminal specificity. Despite some very high titers, the sensitivity for angiotensin II (> 0.7 nM) was inadequate for immunoassay in plasma. Preliminary results with the immunoperoxidase technique, however, suggest that these sera (1820) might be very useful in localizing angiotensin in brain tissue (E. Zimmerman, personal communication).
93
(IV) A ntisera against angiotensin-( 3.8)-hexapeptide-thyroglobulin Three rabbits (nos. 1831-33) were immunized with angiotensin-(3-8)-hexapeptide that was randomly coupled to porcine thyroglobulin by the CDI method (conjugate IV). The antiserum titers reached between 100 and 2000 with sensitivities between 0.15 and 0.5 riM. All antisera showed significant cross-reactivity with angiotensin II and the peptides 2-8 and 3-8, but less than 2% cross-reactivity with angiotensin I. Serum no. 1832 turned out to be very useful in irnmunocytochemistry. (V) Antisera against angiotensin-(2-8)-heptapeptide-thyroglobulin Three rabbits (nos. 1834-36) were immunized with angiotensin-(2-8)-heptapeptide (purchased from Chemalog, South Plainfield, N J) that was randomly coupled to porcine thyroglobulin by the CDI method (conjugate V). The titers of the antisera reached between 1900 and 30,000. As in antisera raised by the (3-8)-hexapeptide conjugate, the antisera showed significant (> 50%) cross-reactivity with angiotensin II and peptides 2-8 and 3-8, but little reactivity with angiotensin I. (V1) A ntisera against angiotensin 11 diazo linked to thyroglobulin Four rabbits (nos. 1836-40) were immunized with angiotensin II that was diazo linked to bovine thyroglobulin (conjugate VI). One serum (no. 1840) reached a titer of only 100 and was not tested further. The remaining 3 sera reached titers of 800, 24,000, and 34,000 without booster injections. The sensitivity for angiotensin II was low (0.5-1.9 nM), and there was little resolution between the (1-8)-octapeptide and the peptides 2-8 (45-86% cross-reaction) and 3-8 (23-61% cross-reaction). The cross-reaction with angiotensin I was 0.2-1.0% (see Fig. 2).
6O
5o
(b
40~-
20
•---- ~ i ~ e , ~ - 1 ~ - 8 ~
\"~ .
o
I
I
I
L
lo ~
to2
to 3
lo 4
.
I ~ 1o~
CONCENTRATIONOF ilVS~Bl~ (/fix 10-~ Fig. 2. Specificity of rabbit antiserum no. 1837 directed against angiotensin-(l-B)-octapeptide which was diazo linked to bovine thyroglobulin. Strong cross-reaction with the peptides 2-8 (86%) and 3-8 (61%). Weak cross-reaction with angiotensin-( I- 10)-decapeptide (0.2%).
94 Discussion
The value of a peptide immunoassay is markedly enhanced when it is possible to resolve the compound of interest from precursors or metabolites that may also be present in physiologic fluids. Special problems have arisen in the differentiation of parathyroid hormone from its long-lived metabolite (Segre et al., 1972) and in the differentiation of angiotensin II from its degradation products, (des-AspJ)-hepta peptide and (des-AspI,Arg 2)-hexapeptide. We have examined a variety of strategies for raising antibodies that have an enhanced ability to recognize the amino-terminal region of angiotensin II and thus differentiate between the active hormone and its 2 principal metabolites. Carbodiimide conjugation of angiotensin II and a carrier protein results in a product in which the peptide is coupled either by its amino- or carboxy-terminal end to the protein. Immunization with conjugates of this nature most often give rise to antisera that recognize the carboxy-terminal end of the peptide and thus do not differentiate between angiotensin II and its metabolites. Rarely. for reasons that are not understood, an animal will yield an amino terminal-specific antiserum. The frequency of this seemingly serendipitous event is so low that specific immunoassays for angiotensin II are not available at present. We examined several different strategies for overcoming this problem that emphasized presentation of the peptide on the conjugate so that the amino terminal end would be available to recognition by the immune system and the carboxy-terminal end masked. This is based on the observations by Sela et al. (1962), who utilized synthetic co-polymers of amino acids as antigens. They demonstrated that when a branch chain co-polymer of poly-L-lysine and D.L-alanine was used as an immunogen, amino acids near the distal ends of the branches were always recognized by the antibodies produced, while those near the backbone were only recognized when the density of side chains on the backbone was low and not when it was high. Thus it was possible to mask antigenic determinants, making use of the principle of steric hindrance. This principle was previously applied, though with only modest success, by building branch-chain polymers of poly-L-lysine and angiotensin II or bradykinin, in which the peptides were coupled to the backbone either by their carboxy- or amino-terminal ends (Haber et al., 1965; Talamo et al., 1968; Fischer et al., 1969). It was not possible to obtain peptides that exclusively recognized either end of the peptide using this approach, presumably because a sufficiently high density of substitution on the polymer could not be achieved. We have now examined the following approaches. (I) Coupling amino terminal-acetylated angiotensin to carrier protein using the carbodiimide reaction. Permanent blockade of the amino terminus by the acetyl group results in coupling of the carboxy terminus of the peptide exclusively to the carrier protein. (2) Reversible blockade of the amino terminus of angiotensin II by trifluoroacetylation of the a-amino group and by protecting the fl-carboxyl group of the aspartic acid in position 1 of the peptide with NO2Bzl. This permits exclusive coupling of the
95 carboxy terminus to the carrier protein but allows subsequent removal of the blocking group prior to immunization so that the amino terminus of the peptide is exposed to the immune system. (3) Immunization with peptide conjugates in which the peptide is angiotensin II lacking the carboxy-terminal phenylalanine residue. In this instance it is hoped that carboxy-terminal determinants that appear to dominate the immune response will be deleted. (4) Coupling to tyrosine or histidine via a diazo linkage. This fixes the center of the peptide to the carrier molecule and exposes both amino- and carboxy-terminal ends. None of these approaches was successful in yielding angiotensin II-specific antisera that were suitable for immunoassay. Amino terminal-acetylated angiotensin conjugates gave rise to antibodies that were highly specific for acetylated angiotensin II amide and cross-reacted very little with human angiotensin II or any of the smaller peptides. A cross-reactivity of 1% with unacetylated angiotensin amide showed that apparently amino-terminal specificity was achieved, but it was of no value in producing a useful antiserum since the human hormone of interest was not recognized. It was of interest, however, that reversible blockade of the amino terminus did not result in the same degree of specificity. Antisera to such a conjugate cross-reacted extensively with smaller peptides. The reason for the apparent inconsistency in these results may be explained by possible degradation of the immunogen by aminopeptidases in the case of conjugate IX while the presence of an amino-terminal acetyl group in conjugate I prevented amino peptidase action. The acetyl groups also introduced a new antigenic specificity which did not contribute to the solution of the problem at hand. Lee et al. (1980) were able to show enhanced specificity for the amino-terminal end of substance P when specifically carboxy-terminal coupled. Substance P has the sequence, Arg-Pro-..., and is thus not a substrate for mammalian aminopeptidase. This result is consistent with the aminopeptidase hypothesis in our experiments. The decreasing specificity in late sera is consistent with antigen degradation favoring antibody production against metabolites. Because both ends of the molecule are free in conjugate XII, no enhanced specificity for the amino-terminal end is expected, and none was found. The conjugates containing peptides that lack the carboxy-terminal phenylalanine (3, above) seem to be sufficiently different in their conformation from angiotensin II so that antibodies raised in response to them recognize angiotensin I but do effectively differentiate between angiotensin II and its metabolites. This hypothesis is supported by the observation that antibodies to peptides that lacked the amino-terminal residues of angiotensin II (2-8 and 3-8) did not recognize angiotensin I, suggesting that the carboxy-terminal phenylalanine of angiotensin II may be of importance in the immunologic differentiation between angiotensin I and II. The quest for a reproducible way of obtaining angiotensin II-specific antibodies that do not recognize the metabolites of this peptide that lack the first or second amino-terminal amino acid has not been successful via antigen modification since it is likely that the antigen itself is modified by the same process that degrades the
96
physiologic peptide. Stabilization of the amino terminus of the antigen without changing antigenic identity is unlikely to be possible. Since rare clones of antibody-forming cells are known to exist that do have the requisite specificity, combining the technique of somatic-cell fusion with a sophisticated means of selection of the appropriate rare antibody-secreting clone may yield a solution to this difficult problem.
Acknowledgements
The authors thank J. DePaolis, V. Lucas, J. Wechsler and S. Gaehde for excellent technical assistance and R. Rubin for editorial assistance. This work has been supported by the Swiss National Funds and the Swiss Society for Biological and Medical Grants.
References Fischer. J., J. Spragg, R.C. Talamo, R.V. Pierce, K. Suzuki, K.F. Austen and E. Haber, 1969. Biochemistry 8, 3750. Goodfriend, T.L.. L. Levine and G.D. Fasman, 1964, Science 144. 1344. Haber, E., L.B. Page and G.A. Jacoby. 1965, Biochemistry 4. 693. Haber. E., T. Koerner. L.B. Page, B. Khman and A. Purnode. 1969. J. Clin. Endocr. Metab. 29. 1349. Hunter, W.M. and F.C. Greenwood, 1962, Nature (London) 194. 495. Inman, J.K., 1974, in: Methods in Enzymology. XXXIV B, eds. W.B. Jakoby and M. Wilchek (Academic Press. New York) p. 49. Lee, C.M.. P.C. Emson and L.L. Iversen. 1980, Life Sci. 27, 535. Segre. G.V.. J.F. Habener, D. Powell, G.W. Tregear and J.T. Potts, 1972, J. Clin. Invest. 51. 3163. Sela, M.. S. Fuchs and R. Arnon, 1962, B&hem. J. 85, 223. Stewart, J.M. and J.D. Young, 1969, in: Solid-phase Peptide Synthesis (Freeman, San Francisco. CA). Talamo. R.C., E. Haber and K.F. Austen, 1968, J. Immunol. 101, 333.
85
Elsevier Biomedical Press
Selectivity of Angiotensin II Antisera Jiirg Nussberger, Gary R. Matsueda, Richard Re* and Edgar Haber** Cellular and Molecular Research Laboratory, Cardiac Unit, Massachusetts General Hospital and Harvard Medical School, Boston, MA 02114, U.S.A.
(Received 8 January 1982, accepted 28 June 1982)
A significant problem in the immunoassay of angiotensin II is the cross-reactivity of most available antisera with the peptide's metabolic products, (des-Asp I )-angiotensin I1 and (des-Asp',Arg 2)-angiotensin II. In order to attempt to generate antisera of greater selectivity, a variety of conjugates between angiotensin II or derivative peptides and carrier proteins were examined as immunogens with the aim of generating antisera that would selectively identify the amino terminal region of the peptide. Selectivity for the amino terminus was achieved by either (1) immunization with N-acetylated angiotensin ll-amide which had been coupled to rabbit serum albumin by its carboxy terminus, or (2) immunization with angiotensin-(I-7)-heptapeptide which was randomly coupled to thyroglobulin. The antisera produced with the N-acetylated immunogen cross-reacted with the unacetylated ligand (Asnl-ValS)-angiotensin, but did not recognize the human hormone (Asp',lleS)-angiotensin. Carboxy-terminal coupling of angiotensin without N-acetylation did not induce selectivity for the amino terminus, nor did a conjugate which was linked to the carrier protein via a diazo bond to His6 of the peptide. These findings may be explained by the fact that N-acetylated angiotensin I1 resists degradation by amino Imptidases and thus retains its structure in the immunogen and by the fact that the (I-7)-heptapeptide has lost the immunodominant carboxy-tcrminal epitope, thus emphasizing the desired amino terminal determinant. Key words: angiotensin 11 - - immunodominance - - immunospecificity - - unidirectional conjugation
Introduction A c e n t r a l p r o b l e m in the d e t e r m i n a t i o n o f p e p t i d e - h o r m o n e c o n c e n t r a t i o n in p h y s i o l o g i c fluids is the d i f f e r e n t i a t i o n a m o n g the p r e c u r s o r , the a c t i v e species, a n d t h e d e g r a d a t i o n p r o d u c t . A t times, c h a n c e a p p e a r s to p r o v i d e an a n t i s e r u m o f a p p r o p r i a t e specificity, b u t m o r e o f t e n t h a n n o t p h y s i o l o g i c o b s e r v a t i o n s are c o n f o u n d e d b y an i n a b i l i t y to c o r r e l a t e i m m u n o a s s a y w i t h b i o a s s a y b e c a u s e i n a c t i v e as well as a c t i v e p e p t i d e s a r e b o u n d to a n t i b o d y w i t h n e a r e q u a l affinity. T h i s p r o b l e m is p a r t i c u l a r l y a c u t e in the i n s t a n c e o f a n g i o t e n s i n II d e t e r m i n a t i o n . W h i l e a n t i s e r a * Current address: Ochsner Clinic, 1514 Jefferson Highway, New Orleans, LA 70121, U.S.A. ** Correspondence to: Edgar Haber, MD, Cardiac Unit, Massachusetts General Hospital, Boston, MA 02114, U.S.A. 0022-1759/83/0000-0000/$03.00 © 1983 Elsevier Biomedical Press
86
that show relatively little cross-reaction between the precursor decapeptide, angiotcnsin I, and the octapeptide, angiotensin II, are conventionally obtained by immunization with either material, angiotensin II antisera nearly always cross-react with 2 amino-peptidase-mediated degradation products: (des-Asp~)-angiotensin I! and (des-AspI,ArgZ)-angiotensin II. Because these products have important activity differences with angiotensin II, their independent measurement is of considerable significance. In this paper we present a systematic examination of a variety of immunogens in an attempt to address the question of whether or not specific methods of coupling or orientation of angiotensin II in relation to a carrier molecule or the use of shorter peptides as haptenic determinants would enhance the ability of antisera to bind angiotensin II without significant cross-reaction with other peptides.
Methods
Synthesis of peptides General procedures used for Merrifield solid-phase synthesis have been described by Stewart and Young (1969). The reagents used were 25% trifluoroacetic acid in CH2C12 for deprotection, 10% triethylamine in CH2C! 2 for neutralization, and Boc-amino acid:dicyclohexylcarbodiimide (1 : 1) at a 3-fold excess in CH2CI 2 for coupling. Boc-amino acids of the L-configuration were purchased from either Bachem (Torrance, CA) or Peninsula Labs (San Carlos, CA). (a) Synthesis of (N~-trifluoroacetyl-Asp(NO, Bzl)l)-angiotensin II. Starting with 3.08 g of Boc-Phe-resin (0.13 mmol/g), protected angiotensin was assembled using a Beckman 990B synthesizer to give Boc-Asp(NO2-Bzl)-Arg(Tos)-Val-Tyr(benzyloxycarbonyl)-Ile-His(Tos)-Pro-Phe-OCH2-resin (3.88 g). Following deprotection of the a-Boc group, the neutralized resin was treated with 3 mmol of trifluoracetic acid anhydride in CH 2C12 for 20 min. Trifluoroacetylation was complete as evidenced by a negative result with the Kaiser ninhydrin reagents. HF treatment ( H F : a n i s o l e (9: 1) for 30 min at 0°C) yielded 650 mg of peptide. The peptide was purified by partition-column chromatography on Sephadex G-25 (2 cm x 45 cm) using nBuOH : HOAc : H 2 0 (4 : 1 : 5). The peptide eluted in a 17 ml volume after 72 ml of column effluent had been collected. The R F of the peptide on a silica gel (thin layer) was 0.45 with n-BuOH: HOAc: H 2 0 ( 1 2 : 3 : 5 ) as developing solvent. The amino acid composition of the purified peptide was within 6% of theoretical values. The amino acid sequence was confirmed by solid-phase Edman degradation of a peptidyl-resin sample which was treated previously with I M NaOH in EtOH to remove both the N-trifluoracetyl group and the p-NO 2Bzl group. (b) Synthesis of angiotensin-(1-7)-heptapeptide. Starting with 2.5 g of Boc-Proresin (0.08 mmol/g), Boc-Asp(Bzl)-Arg(Tos)-Val-Tyr(benzyloxycarbonyl)-IieHis(Tos)-Pro-OCH2-resin was assembled using a Beckman 990B peptide synthesizer. H F cleavage of the peptidyl-resin product yielded 164 ~mol of peptide (65% yield). After purification by partition-column chromatography as described above, 105 mg
87 of peptide was obtained which was homogeneous by TLC (R F = 0.42 with solvent system n-BuOH : HOAc: H 2 0 : pyridine (15 : 3 : 12 : 10)). The amino acid composition was within 8% of theoretical values after 6 M HCI hydrolysis for 24 h. (c) Synthesis of angiotensin-(3-8)-hexapeptide. Starting with 1.5 g of Boc-PheOCHE-resin (0.15 mmol/g), Boc-Arg(Tos)-Val-Tyr(benzyloxycarbonyl)-Ile-His(Tos)Pro-Phe-resin was assembled manually. HF treatment of 1.98 g of peptidyl-resin yielded 276 mg of the crude peptide which was subsequently purified by carboxymethyl-cellulose ion-exchange chromatography after partition chromatography failed to give a homogeneous peptide product. The purified peptide (57 rag) gave a single spot by TLC ( R r = 0.56 with n-BuOH : HOAc: H 2 0 (12 : 3 : 5)) and a satisfactory amino acid analysis which was within 4% of theoretical values.
Labeling of peptides The peptides were labeled with 125I by the chloramine-T method (Hunter and Greenwood, 1962).
Preparation of immunogens Peptides were not used as immunogens but were linked to carrier proteins employing several different methods. An amide linkage was formed using the carbodiimide (CDI) method as described previously (Goodfriend et al., 1964; Haber et al., 1965). Carboxy-terminal coupling of the peptide was obtained by protecting the N-terminal reacting groups prior to CDI treatment. (1) N-acetylation of (Asnl)-angiotensin. 17.44 /xmol of [AsnI,ValS]angiotensin II (Hypertensin Ciba), a gift of Ciba (Newark, N J), were suspended in 3 ml distilled water and 3 ml of saturated sodium acetate. With continuous stirring, 1.5 mmol of [laC]acetic anhydride (2.45 pCi per mmol) were added and the insoluble peptide dissolved. After 60 min the fluorescamine test (Hoffmann-LaRoche, Nutley, N J) gave blank values, indicating complete acetylation of the alpha amino group of the peptide. The unreacted acetic anhydride was separated from the peptide by gel filtration on Sephadex G-25 (medium). Paper chromatography (n-butanol:glacial acetic acid:water (12:3:5)) showed an R E of 0.7 for the reaction product as compared to 0.5 for the starting material. The reaction product was positive by the Pauly stain for histidine but negative when tested by the ninhydrin reaction. Its absorbance at 280 nm did not change at pH II indicating that the tyrosine of angiotensin II was not acetylated. Amino acid analysis confirmed that the product was a derivative of angiotensin II. Acetylated peptide (7.64 /xmol) and rabbit serum albumin (0.18 pmol) were dissolved in 2.5 ml of distilled water, and 2.61 mmol of 1-ethyl-3-(3-dimethylaminopropyl)carbodiimide hydrochloride (Sigma, St. Louis, MO) (CDI) were added. Immunoassays were carried out with 125I-labeled acetylhypertensin as tracer. (2) Na-trifluoroacetylation of (Asp(p-NO2Bzl)1)-angiotensin. The synthesis of (Na-trifluoroacetyl-Asp(NO2Bzl)t-angiotensin II is described above. This peptide (57.6 #tool) dissolved in 4 ml of dimethylformamide and 154 #mol bovine thyroglobulin were mixed in 12 ml of distilled water, and 10.4 mmol of CDI was added.
88 The reaction was allowed to proceed with shaking at room temperature for 20 h. Eighteen milliliters of 1.8 N N a O H were then added to remove the amino-terminal protecting group, and 7 min later 18 ml of !.86 N acetic acid was added to bring the pH to 6. When the protected peptide was treated alone, this brief alkaline treatment yielded a product which was indistinguishable from angiotensin 11 (by TLC). A diazo link was effected by coupling p-aminophenylethylamine (Aldrich. Milwaukee, WI) to the carrier protein using the carbodiimide method and then forming the diazonium salt, which was in turn coupled to the peptide (reacting at either histidine, tyrosine or amino groups) essentially as described by lnman (1974). This conjugate was synthesized by mixing 14.7 mmol of p-aminophenylethylamine (Sigma), 154 nmol of bovine thyroglobulin and 5.2 mmol of CDI in 3 ml 0.1 M NaCI, 0.01 M Na phosphate, pH 7.5. to which was added 2 ml of glacial acetic acid. The final pH of the reaction mixture was 6. The thyroglobulin derivative was isolated and purified as described above. Fifteen milligrams of this thyroglobulin derivative were dissolved in 1 ml of 0.01 N HCI at 0°C. Six milligrams NaNO2 were then added. 9.47 ~mol of angiotensin II (purchased from Peninsula Labs.) were dissolved in i.5 ml of 0.1 M Na2CO 3. Fifteen minutes later the peptide solution was added to the thyroglobulin solution and maintained at 0°C for 30 min. The pH of the reaction mixture was 9.
Purification and characterization of the immunogens The conjugates were dialyzed against 1.5 mM N a 2 H P O 4 containing 3 mM sodium azide for 24 h at room temperature. Coupling was monitored by measuring the incorporation of a tracer quantity of ~25I-labeled or 3H-labeled peptide added to the reaction mixture. Non-dialyzed tracer was assumed to be coupled peptide. The molar incorporation of peptide into the conjugate was calculated. The final determination of the incorporation of peptide was accomplished by amino acid analysis (Durrum 500 analyzer).
Immunization Immunogens were dissolved in isotonic saline at a concentration of 2 m g / m i and emulsified with an equal volume of complete Freund's adjuvant (containing 1 m g / m l killed tubercle bacilli). New Zealand White rabbits (1.5-2.5 kg) were injected intradermally, either in the toe pads or other body sites. The primary immunizing dose was 1 mg. Booster injections were administered at 2-4-week intervals using 0.5 mg of conjugate administered intradermaily (in complete Freund's adjuvant). Blood (40 ml) was collected weekly from the ear arteries beginning on the third week after immunization.
Characterization of antisera Immunoassays were carried out utilizing the dextran-coated charcoal method as described previously (Haber et al., 1969). All antisera were tested with the radiolabeled haptenic peptides. Titer was defined as the dilution of antiserum at which 50% of homologous tracer peptide was bound to antibody (3000 cpm representing 3 fmol ~2sI-labeled peptide). Sensitivity was defined as the reciprocal of the quantity of
89
unlabeled peptide that displaced 50% of the tracer bound to antibody in the absence of any inhibitor. Specificity of an antiserum in discriminating between two peptides was defined as the quotient of their respective sensitivities.
Results
The characteristics of the various angiotensin-carrier conjugates (I-VI) are shown in Table I. The highest coupfing efficiency was obtained with the diazo linkage: 89% of the peptide was linked to thyroglobulin resulting in a peptide-to-cartier ratio of 426. Although less efficient, the carbodiimide procedure was reliable, giving 50-80% coupling when the peptide-chloride salt was used, but only 10-30% coupling when peptide-acetate salt was used. Regardless of the coupling procedure, all conjugates elicited antibodies in all rabbits with titers ranging from 100 to 100,000. We believe that this success was due in part to thorough emulsification of the conjugate-adjuvant mixture and to the use of the hind footpad as a site of immunization. When a conjugate was insoluble in saline, ultrasonication was helpful in obtaining a fine suspension.
(I) Antisera against [Na-acetyl-Asnt, ValS]-angiotensin II rabbit serum albumin (A c-hypertensin-RSA ) Five rabbits (nos. 1800-1804) were immunized with Ac-hypertensin-RSA (conjugate I) and were boosted 4, 6, 8 and 12 weeks later. The sera reached titers between
o- --o (NoAcetyi-AsnI ValS)-An~ioten~n_(1_8).C)ctope~ide
501-
;
; (N-~..-AsP')-~s~-(t-.4)-Te.o~o~p~
-
_ _ -- ,-~- . . . .
40'-
-
"''-
30
-
;
"'-.
\~
20
10
~
I
I
t0 2
103
k'~ t - I 0
"~ ~
" "
.//-
///Sz_ 8 --~--- 1-7
~ \\
4-8
', \"
1
[
I
I
104
105
10s
107
CO~NI"/gAT/ON OF/NHIB/IOR(Mx "10-~ Fig. 1. Specificity of rabbit antiserum no. 1803 directed against (N=-acetyl-Asnt,ValS)-angiotensin.(I.8)octapeptide which was carboxy-terminal coupled to rabbit serum albumin. Cross-reactivity could be demonstrated only with (Asnm,VaiS)-angiotensin-(l-8)-octapeptide at a level of I%.
IN”-Acetyl-Asn’,Val’j angiotensin-( I-8)-octapeptide angiotensin-( I-8)-octapeptide angiotensin-( I-7)-heptapeptide angiotensin-( I-7)-heptapeptide angiotensin-( I-7)-heptapeptide angiotensin-(3-8>hexapeptide angiotensin-(2-8)-heptapeptide angiotensin-( I-l>octapeptide
I
’ Determined by amino acid analysis. b RSA = rabbit serum albumin. ’ Tg = thyroglobulin.
IV V VI
Ilk
II llla IIIb
Peptide
CONJUGATES
Conjugate number
ANGIOTENSIN-CARRIER
TABLE I
Tg Tg Tg Tg Tg Tg Tg
’ (bovine) (porcine) (bovine) (bovine) (porcine) (porcine) (bovine)
RSA h
Carrier
CD1 CD1 CD1 CD1 CD1 CD1 Diazo
CD1
Coupling method 2
Ill*24 47* 3 63+24 93+ 16 17+ 5 37* 3 426*46
12+
Moles peptide per mole carrier a
C-terminal C/N-terminal C/N-terminal C/N-terminal C/N-terminal C/N-terminal His/Tyr
C-terminal
Coupling through (site of peptide)
II
ANG b (l-8) ANG(I-8)
ANG( l-8)
ANG(I-8)
ANG(l-8)
ANG( l-7)
ANG(I-7)
ANG(I-7)
ANG(3-8) ANG(3-8) ANG(2-8) ANG(I-8)
II II
II
II
II
IIla
IIIa
IlIb
IIIC
IV IV V VI
I 826 1827
1828
1829
1830
1810
1811
1817
1820
I832 1833 1835 1837
’ PI, primary immunization. b ANG, angiotensin. ’ No cross-reaction ( < 0.01 I). d No displacement of tracer (sensitivity
ANG(I-7)
‘*’ I-labeled angiotensin analog
Conjugate
OF RABBIT
Serum number
CHARACTERIZATION
TABLE
2000 100 30000 100 4000 16000 50 4000 50 600 20 2500 IOOOO 20 800 1500 50 2500 25 3ooo 600 800 15wO 5000
Dilution
> 100,000 nM).
8 3 8 3 8 16 3 8 3 8 3 7 I5 3 7 I5 3 II 4 II 4 4 4 4
Weeks after PI ’
ANTI-ANGIOTENSIN
loo IO0 100 100 loo 100 IO0 100 100 100 61 50 4 60 40 2 75 1 80 75 I40 40 65 100
1-8
Specificity
3 44 56 29 46 75 18 53 3 5 nc’ 7 I nc 7 0.7 nc 0.1 nc 20 117 50 100 86
I
0.5
I
0.5
I
0.01 nc 10 100 100 50 61
nc
nc
nc
2 nc 77 2 38 II 0.4 67 nc
3-8
(% cross-reaction) 2-8
SERA
0.2
I
0.1
I
I 47 1 39 I I 41 1 65 I5 I5 13 0.2 24 I6 0.2 30 0.1 31 5
I-10
_
16 100 100 100 100 100 100 100 100 100 IO0
_ _
1 _ _
1-7
8 IO 2 42 4 42 10 0.8 0.50 0.15 0.17 1.9
I
0.05 80 0.1 0.43 0.02 0.02 2.2 0.18 20 2.8 33
l-8
Sensitivity
1.8 180 0.18 1.5 0.05 0.03 I2 0.34 670 62 nd d 7 30 nd II 150 nd 520 nd 3 0.60 0.12 0.11 2.20
2-8
(nM)
3 0.13 21 0.06 0.20 600 0.27 nd 280 nd 110 25 nd 150 105 nd 4200 nd 6 0.70 0.06 0.22 3.10
nd
3-8 8 170 I4 I.1 3.5 3.5 5.4 20.0 31 19 I30 4 140 25 5 450 IO 1100 26 II 70 62 22 900
l-10
_
I7 20 0.5 0.3 6 0.8 I 3 6 8 0.6
_ _
8
l-7
92 400 and 4000. They were tested for sensitivity and specificity beginning 9 weeks following primary immunization. Specificity of these (and subsequent) sera was tested utilizing various angiotensin analogs. Fig. 1 shows the very high specificity of serum (no. 1803); only hypertensin cross-reacted at a level of I%. Very similar characteristics could be demonstrated in the other sera of this group. Sensitivities for Ac-hypertensin between 0.1 and 2.0 nM were always associated with a cross-reactivity of 1-2% for unacetylated hypertensin, and no significant cross-reactivity was found with any other angiotensin peptide. Thus N-terminus specificity for hypertensin was achieved although acetylated hypertensin as the homologous hapten was clearly better recognized by the antisera. This result is consistent with both unidirectional carboxy-terminal coupling of the peptide to the carrier protein and with long persistence of the aminopeptidase-resistant immunogen.
(II) Antisera against angiotensin I1 coupled by its carboxy terminus to bovine thyroglobulin Table II characterizes 5 rabbit antisera (nos. 1826-1830) obtained by immunization (boosted 2 weeks after primary immunization) with angiotensin II that was linked to thyroglobulin by its carboxy terminus (conjugate II). These sera reached titers between 4000 and 40,000. One serum (no. 1828) was very sensitive to the octapeptide (0.02 nM) but exhibited 75% cross-reactivity with the (2-8)-heptapeptide and 11% cross-reactivity with the (3-8)-hexapeptide. The cross-reactivity with angiotensin I was 1%. This cross-reactivity pattern was characteristic for this group of antisera though there appeared to be more amino terminal specificity in early than in late antisera. (III) Antisera against angiotensin-(1- 7)-heptapeptide-thyroglobulin Three sets of 3 rabbits each (nos. 1809-11, 1817-19, 1820-22) were immunized with angiotensin-(l-7)-heptapeptide that was randomly coupled to thyroglobulin by carbodiimide (Table 1, conjugates Ilia-c). Porcine thyroglobulin was found to be more soluble than bovine thyroglobulin. The insoluble conjugates were successfully suspended with ultrasonication. Conjugate IIIb differed from lllc only by a lower peptide incorporation. Under identical conditions with booster injections 2 and 4 weeks after the primary immunization, conjugate Illb induced maximal titers of 30,000, 50,000 and 100,000; whereas lllc induced maximal titers of only 400, 2000, and 3000. The sera of all 9 rabbits showed similar characteristics. Representative examples of each group are demonstrated in Table II. In early sera following immunization with any one of conjugates Ilia, Illb or IIIc, there appeared to be preferential recognition of amino terminal determinants: cross-reactivity with angiotensin I and II was significant; whereas little or no cross-reactivity was found with the (2-8)-heptapeptide and the (3-8)-hexapeptide. In later sera, the cross-reactivity with peptides 2-8 and 3-8 increased, while the cross-reactivity with I-8 and 1-10 decreased, thus decreasing amino terminal specificity. Despite some very high titers, the sensitivity for angiotensin II (> 0.7 nM) was inadequate for immunoassay in plasma. Preliminary results with the immunoperoxidase technique, however, suggest that these sera (1820) might be very useful in localizing angiotensin in brain tissue (E. Zimmerman, personal communication).
93
(IV) A ntisera against angiotensin-( 3.8)-hexapeptide-thyroglobulin Three rabbits (nos. 1831-33) were immunized with angiotensin-(3-8)-hexapeptide that was randomly coupled to porcine thyroglobulin by the CDI method (conjugate IV). The antiserum titers reached between 100 and 2000 with sensitivities between 0.15 and 0.5 riM. All antisera showed significant cross-reactivity with angiotensin II and the peptides 2-8 and 3-8, but less than 2% cross-reactivity with angiotensin I. Serum no. 1832 turned out to be very useful in irnmunocytochemistry. (V) Antisera against angiotensin-(2-8)-heptapeptide-thyroglobulin Three rabbits (nos. 1834-36) were immunized with angiotensin-(2-8)-heptapeptide (purchased from Chemalog, South Plainfield, N J) that was randomly coupled to porcine thyroglobulin by the CDI method (conjugate V). The titers of the antisera reached between 1900 and 30,000. As in antisera raised by the (3-8)-hexapeptide conjugate, the antisera showed significant (> 50%) cross-reactivity with angiotensin II and peptides 2-8 and 3-8, but little reactivity with angiotensin I. (V1) A ntisera against angiotensin 11 diazo linked to thyroglobulin Four rabbits (nos. 1836-40) were immunized with angiotensin II that was diazo linked to bovine thyroglobulin (conjugate VI). One serum (no. 1840) reached a titer of only 100 and was not tested further. The remaining 3 sera reached titers of 800, 24,000, and 34,000 without booster injections. The sensitivity for angiotensin II was low (0.5-1.9 nM), and there was little resolution between the (1-8)-octapeptide and the peptides 2-8 (45-86% cross-reaction) and 3-8 (23-61% cross-reaction). The cross-reaction with angiotensin I was 0.2-1.0% (see Fig. 2).
6O
5o
(b
40~-
20
•---- ~ i ~ e , ~ - 1 ~ - 8 ~
\"~ .
o
I
I
I
L
lo ~
to2
to 3
lo 4
.
I ~ 1o~
CONCENTRATIONOF ilVS~Bl~ (/fix 10-~ Fig. 2. Specificity of rabbit antiserum no. 1837 directed against angiotensin-(l-B)-octapeptide which was diazo linked to bovine thyroglobulin. Strong cross-reaction with the peptides 2-8 (86%) and 3-8 (61%). Weak cross-reaction with angiotensin-( I- 10)-decapeptide (0.2%).
94 Discussion
The value of a peptide immunoassay is markedly enhanced when it is possible to resolve the compound of interest from precursors or metabolites that may also be present in physiologic fluids. Special problems have arisen in the differentiation of parathyroid hormone from its long-lived metabolite (Segre et al., 1972) and in the differentiation of angiotensin II from its degradation products, (des-AspJ)-hepta peptide and (des-AspI,Arg 2)-hexapeptide. We have examined a variety of strategies for raising antibodies that have an enhanced ability to recognize the amino-terminal region of angiotensin II and thus differentiate between the active hormone and its 2 principal metabolites. Carbodiimide conjugation of angiotensin II and a carrier protein results in a product in which the peptide is coupled either by its amino- or carboxy-terminal end to the protein. Immunization with conjugates of this nature most often give rise to antisera that recognize the carboxy-terminal end of the peptide and thus do not differentiate between angiotensin II and its metabolites. Rarely. for reasons that are not understood, an animal will yield an amino terminal-specific antiserum. The frequency of this seemingly serendipitous event is so low that specific immunoassays for angiotensin II are not available at present. We examined several different strategies for overcoming this problem that emphasized presentation of the peptide on the conjugate so that the amino terminal end would be available to recognition by the immune system and the carboxy-terminal end masked. This is based on the observations by Sela et al. (1962), who utilized synthetic co-polymers of amino acids as antigens. They demonstrated that when a branch chain co-polymer of poly-L-lysine and D.L-alanine was used as an immunogen, amino acids near the distal ends of the branches were always recognized by the antibodies produced, while those near the backbone were only recognized when the density of side chains on the backbone was low and not when it was high. Thus it was possible to mask antigenic determinants, making use of the principle of steric hindrance. This principle was previously applied, though with only modest success, by building branch-chain polymers of poly-L-lysine and angiotensin II or bradykinin, in which the peptides were coupled to the backbone either by their carboxy- or amino-terminal ends (Haber et al., 1965; Talamo et al., 1968; Fischer et al., 1969). It was not possible to obtain peptides that exclusively recognized either end of the peptide using this approach, presumably because a sufficiently high density of substitution on the polymer could not be achieved. We have now examined the following approaches. (I) Coupling amino terminal-acetylated angiotensin to carrier protein using the carbodiimide reaction. Permanent blockade of the amino terminus by the acetyl group results in coupling of the carboxy terminus of the peptide exclusively to the carrier protein. (2) Reversible blockade of the amino terminus of angiotensin II by trifluoroacetylation of the a-amino group and by protecting the fl-carboxyl group of the aspartic acid in position 1 of the peptide with NO2Bzl. This permits exclusive coupling of the
95 carboxy terminus to the carrier protein but allows subsequent removal of the blocking group prior to immunization so that the amino terminus of the peptide is exposed to the immune system. (3) Immunization with peptide conjugates in which the peptide is angiotensin II lacking the carboxy-terminal phenylalanine residue. In this instance it is hoped that carboxy-terminal determinants that appear to dominate the immune response will be deleted. (4) Coupling to tyrosine or histidine via a diazo linkage. This fixes the center of the peptide to the carrier molecule and exposes both amino- and carboxy-terminal ends. None of these approaches was successful in yielding angiotensin II-specific antisera that were suitable for immunoassay. Amino terminal-acetylated angiotensin conjugates gave rise to antibodies that were highly specific for acetylated angiotensin II amide and cross-reacted very little with human angiotensin II or any of the smaller peptides. A cross-reactivity of 1% with unacetylated angiotensin amide showed that apparently amino-terminal specificity was achieved, but it was of no value in producing a useful antiserum since the human hormone of interest was not recognized. It was of interest, however, that reversible blockade of the amino terminus did not result in the same degree of specificity. Antisera to such a conjugate cross-reacted extensively with smaller peptides. The reason for the apparent inconsistency in these results may be explained by possible degradation of the immunogen by aminopeptidases in the case of conjugate IX while the presence of an amino-terminal acetyl group in conjugate I prevented amino peptidase action. The acetyl groups also introduced a new antigenic specificity which did not contribute to the solution of the problem at hand. Lee et al. (1980) were able to show enhanced specificity for the amino-terminal end of substance P when specifically carboxy-terminal coupled. Substance P has the sequence, Arg-Pro-..., and is thus not a substrate for mammalian aminopeptidase. This result is consistent with the aminopeptidase hypothesis in our experiments. The decreasing specificity in late sera is consistent with antigen degradation favoring antibody production against metabolites. Because both ends of the molecule are free in conjugate XII, no enhanced specificity for the amino-terminal end is expected, and none was found. The conjugates containing peptides that lack the carboxy-terminal phenylalanine (3, above) seem to be sufficiently different in their conformation from angiotensin II so that antibodies raised in response to them recognize angiotensin I but do effectively differentiate between angiotensin II and its metabolites. This hypothesis is supported by the observation that antibodies to peptides that lacked the amino-terminal residues of angiotensin II (2-8 and 3-8) did not recognize angiotensin I, suggesting that the carboxy-terminal phenylalanine of angiotensin II may be of importance in the immunologic differentiation between angiotensin I and II. The quest for a reproducible way of obtaining angiotensin II-specific antibodies that do not recognize the metabolites of this peptide that lack the first or second amino-terminal amino acid has not been successful via antigen modification since it is likely that the antigen itself is modified by the same process that degrades the
96
physiologic peptide. Stabilization of the amino terminus of the antigen without changing antigenic identity is unlikely to be possible. Since rare clones of antibody-forming cells are known to exist that do have the requisite specificity, combining the technique of somatic-cell fusion with a sophisticated means of selection of the appropriate rare antibody-secreting clone may yield a solution to this difficult problem.
Acknowledgements
The authors thank J. DePaolis, V. Lucas, J. Wechsler and S. Gaehde for excellent technical assistance and R. Rubin for editorial assistance. This work has been supported by the Swiss National Funds and the Swiss Society for Biological and Medical Grants.
References Fischer. J., J. Spragg, R.C. Talamo, R.V. Pierce, K. Suzuki, K.F. Austen and E. Haber, 1969. Biochemistry 8, 3750. Goodfriend, T.L.. L. Levine and G.D. Fasman, 1964, Science 144. 1344. Haber, E., L.B. Page and G.A. Jacoby. 1965, Biochemistry 4. 693. Haber. E., T. Koerner. L.B. Page, B. Khman and A. Purnode. 1969. J. Clin. Endocr. Metab. 29. 1349. Hunter, W.M. and F.C. Greenwood, 1962, Nature (London) 194. 495. Inman, J.K., 1974, in: Methods in Enzymology. XXXIV B, eds. W.B. Jakoby and M. Wilchek (Academic Press. New York) p. 49. Lee, C.M.. P.C. Emson and L.L. Iversen. 1980, Life Sci. 27, 535. Segre. G.V.. J.F. Habener, D. Powell, G.W. Tregear and J.T. Potts, 1972, J. Clin. Invest. 51. 3163. Sela, M.. S. Fuchs and R. Arnon, 1962, B&hem. J. 85, 223. Stewart, J.M. and J.D. Young, 1969, in: Solid-phase Peptide Synthesis (Freeman, San Francisco. CA). Talamo. R.C., E. Haber and K.F. Austen, 1968, J. Immunol. 101, 333.