Enzyme Immunoassay of a/3-Adrenergic Agent Using/3-Galactosidase as Label

Susumu lwasa, Koichi Kondo, Tsutomu Miya, and Kazumi Takeda

Abstract: Antisera against tetrahydronaphthalenols, which are conformationally rigid derivatives of adrenergic catecholamine, were produced in rabbits immunized with tmns-5-amino-6-hy- droxy-2-isopropylamino-l,2,3,4-tetrahydronaphthalene-l-ol (I) conjugated to succinyiated bovine serum albumin at the C5 position on the tetralin ring. Antisera were screened by immunodtffusion and further characterized by passive hemagglutination assay using eryth- rocytes sensitized with tmns-l-ovalbumin conjugate and by enzyme immunoassay using trans-l-~-galactosidase conjugate. Cross-reactivity studies indicated that the antiserum was highly specific for the tetralin structure and for substitution at the C2 position. The antiserum also selectively discriminated the stereoisomers about the C1-C2 bond.

The anU-trans-I serum was used to develop EIA for trans-5-hydroxymethyl-6-hydroxy-2-iso- propylamino-l,2,3,4-tetrahydronaphthalene-l-ol (lib), which exhibited strong E-stimulating activity fairly selective to tracheal muscle, since it recognized trans-Ilb to the same degree as tmns-l. The assay could detect as little as 100 pg of this compound. The mean recovery of trans-IIb added to plasma was 105%, and values for plasma trans-IIb determined by this immunoassay correlated well with those determined by gas chromatography-mass spectrometry.

Key Words: Enzyme immunoassay: Passive hemagglutination assay; Tetrahydronaphthalenol; ]3-stimu- lating activity;/~-galactosidase; Double antibody solid phase.

INTRODUCTION

Tetrahydronaphthalenols have been developed as bronchodilators (Oka, 1977; Itoh, 1977; Sugihara, 1977). These compounds have two stereoisomers, and the trans isomer of 5,6-di- hydroxy-2-isopropylamino-l,2,3,4-tetrahydronaphthalene-l-ol (llIc) was reported to be ten times more potent in bronchorelaxing than the cis isomer (Nishikawa, 1975). The present study describes a simple, specific and inexpensive enzyme immunoassay (EIA) for studying the cross-reactivity of the antibody raised against a tetralin derivative and for determining the plasma levels of the drug.

The structure of the tetralin derivatives used in this paper is presented in Table 1. As these small compounds are not immunogenic in themselves, an immunogenic carrier, bovine serum albumin (BSA) was employed to bestow immunogenicity upon the compounds. Trans-5- amino-6-hydroxy-2-isopropylamino-l,2,3,4-te~ahydronaphthalene-l-ol ([), which has an

Received March 13, 1978; accepted June 6, 1978. From the Chemical Research Laboratories, Central Research Division, Takeda Chemical Industries, Ltd.,

Osaka, Japan. Address reprint request to: Susumu Iwasa, Chemical Research Laboratories, Central Research Division,

Takeda Chemical Industries, Ltd., 17-85, Jusohonmachi 2-chome, Yodogawa-ku, Osaka 532, Japan.

© ELsevier North-Holland, Inc., 1978 Immunopharrnacology i, 3 - 1 2 (1978) 0162-3109/78/01000310502.25 3

4 Susumu lwasa et al.

amino group at the C~ position on the tetralin ring, was selected as a hapten. It has been reported that substitutions near the site of attachment of a hapten and carrier protein were not discriminated by antibodies prepared from the hapten-carrier conjugate (Erlanger, 1973). Therefore, antibodies against trans-I coupled to BSA at the amino group of the Cs position were expected to cross-react with trans-lllc and tmns-5-hydroxymethyl-6-hydroxy-2-iso- propylamino-l,2,3,4-tetrahydronaphthalene-1-ol (lib), which exhibited strong /~2-selective activity.

MATERIALS AND METHODS

Drugs and Chemicals

Tetralin derivatives, I to llle, were prepared in this Division (Oka, 1977; Itoh, 1977; Sugihara, 1977; Miyake, 1977). Propranolol hydrochloride (Sumitomo Chemical Co., Ltd., Japan), L-iso- proterenol hydrochloride (Nikken Chemical Co., Ltd., Japan), and other chemicals used in the present experiment were obtained from commercial sources and used without further purification.

~-Galactosidase (~- D -galactoside galactohydrolase, EC 3.2.1.23), prepared from Escherich ia coli, was purchased from Boehringer Mannheim, Ltd., West Germany. The specific activity of the preparation used was about 30 U/mg (25°C; lactose as substrate) (Wallenfels, 1959).

Administration of Trans-llb

Female guinea pigs weighing 290-370 g were given a single oral dose of 5 mg/kg of trans-llb. Three animals were examined at each of the following intervals after administration, 0, 0.5, 1, 2, 4, 8, and 24 hr, and blood was collected from the abdominal aorta under ether anesthesia.

Preparation of Plasma for EIA

Extraction of free trans.llb (Evans, 1973). Plasma, 0.2 rnl, was extracted with I ml of alkaline methanol and the extract evaporated in vacuo. Added to the residue were I g of NaCI and I ml of ammonia water (2.5%) and the mixture was extracted three times with 2 ml of methyl acetate. The pooled methyl acetate layer was evaporated in vacuo, redissolved in 0.2 ml of 0.01 M phosphate buffer (pH 7.0) containing 0.1% NAN3, 0.1% BSA, 1 mM MgCI~, and 0.1 M NaCI (buffer A) (Kitagawa, 1976), and assayed for trans-llb in EIA.

Hydrolysis of conjugated trans.llb (Martin, 1971). Added to 0.2 ml of plasma was 0.05 ml of 0.2 M phosphate buffer (pH 6.8) containing 0.5 mg of 13-glucuronidase (EC 3.2.1.31, from E. coli, Sigma Chemical Co., St. Louis, Mo.) and sulfatase (EC 3.1.6.1, from Helix pomatia, Sigma Chemical Co., St. Louis, Mo.). After incubation at 30°C for 3 hr, the enzyme reaction was terminated by adding i ml of alkaline methanol. The solution was centrifuged at 2500 rpm for 5 min and the supematant evaporated in vacuo. The residue was dissolved in 0.2 ml of buffer A and the solution assayed for trans-llb in EIA.

Immunogen

To obtain the conjugate of trans-I and BSA, a succinylated BSA (SucBSA) was initially prepared from succinic anhydride according to the usual procedure (Klotz, 1967), and coupled to trans-I through the COOH group by the following procedure. A mixture of SucBSA (700 mg) and trans-I (320 mg) was dissolved in 100 ml of 0.1 M phosphate buffer, pH 6.0. Gradually added to the solution was 380 mg of 1-ethyl-3-(3-dimethylaminopropyl)-car- bodiimide hydrochloride (ECDI) (Fluka AG, Buchs SG, Switzerland). After incubation at 4°C for 3 days protected from light, the mixed solution was dialyzed against water and freeze- dried in vacuo.

Labeled Immunoassay of ~-Adrenergic Agent 5

Table I Specificity of antiserum: cross-reactivity of various compounds relative to tetrahydronaphthalene derivatives in passive hemagglutination-inhibition assay

Structure

RI

HO . ~ ~ Stereoisomer Cross-

NHR 2 about the HI reactivity Compound OH C1-C2 bond titer a (%)

I R1 = NH2 Tmns 3200 100 R2 = CH(CHz)2 Cis nt ~ - -

lla R1 = CH~OH Trans 2 0.06 R2 = H Cis nf" - -

lib R, = CH20H Trans 6400 200 R2 = CH(CH3)~ C/s 32 1.0

llc RI = CH~OH Trans 64 2.0 CH~

/ \ R2 = CH CH~ Cis nt b - -

\ / CH~

Ilia R~ = OH Trans 2 0.06 R2 = H C/s <1 <0.03

IIlb R~ = OH Trans 16 0.5 R2 = CH3 Cis nt ° - -

Illc R~ = OH Trans 4096 128 Rz = CH(CH3)~ Cis 64 2.0

llId R~ = OH Trans 1 0.03 Rz = C(CH3)3 Cis <1 <0.03

llle R1 = OH Trans 128 4.0 CHz

/ \ R2 = CH CH2 Cis nt ° - -

\ / CH2

Reciprocal value of sample dilution. Initial concentration of the sample solution was 1.0 mg/ml.

nt = not tested,

I m m u n i z a t i o n

Antibodies were produced in rabbits by repeated immunizations. The immunogen was dissolved in saline and emulsified with an equal amount of complete Freund's adjuvant (Dffco Laboratories, Inc., Detroit, Mich.). An emulsion (I ml) containing 5 mg of immunogen was injected once every 10 days for 1 month and then once every 2 - 4 weeks intramuscularly into both thighs except that the second injection consisted of 1 ml of the alum suspension containing 5 mg of immunogen given intraperitoneally. Bleedings were taken from the central ear artery 6 - 8 days after the booster injections. Blood was allowed to clot overnight at 4°C and centrifuged at 2000 rpm for 15 min to separate the serum. The sera were screened by immunodiffusion (Ouchterlony, 1973) against the hapten coupled to ovalbumin (OA). Coupling of hapten to OA was by the same procedure as for the preparation of hapten-BSA conjugates. Following the detection of precipitin lines indicative of antibody to the hapten, antisera were further characterized by passive hemagglutination assay (PHA) and EIA.

6 Susumu lwasa et al.

PHA Procedures

One volume of a 10% (v/v) chick erythrocyte suspension in phosphate-buffered saline (PBS), pH 7.3, was mixed with ten volumes of an antigen solution in PBS of 0.5 mg/ml and incubated at room temperature for 30 min. To the antigen-erythrocyte mixture was added one volume of physiological saline containing 2.5% glutaraldehyde. Incubation continued at room tempera- ture for another 60 rain with frequent shaking. The sensitized cells were collected by centrifugation and washed three times with PBS.

Hemagglutination assay (HA) and hemagglutination-inhibition assay (HI) tests were performed by the microtiter method of Sever (1962), using V-bottomed plates of the permanent type (Tomy Seiko Co., Ltd., Japan) and spiral loops (Microbiological Associates, Inc., Bethesda, Md.). PBS containing 0. I% OA and 0.001% gelatin was used throughout as a diluent for antisera, hapten solutions, and the sensitized erythrocyte suspensions. The sensitized cells were used as 0.25% suspensions.

Preparation of Hapten-Enzyme Conjugate

A suspension of ~-galactosidase in 2.2 M ammonium sulfate was desalted by gel filtration on Sephadex G-100. To 2 ml of the enzyme fraction containing 10 mg protein, were added 0.5 mg of trans-I and I mg of ECDI. After incubation at 4°C for 20 hr protected from light, the mixture was dialyzed against water and purified by Sephadex G-50 (fine) column chromatography. Conjugates thus prepared were stored at 4°C after tenfold dilution with buffer A.

Insolubilization of Goat Antibodies Against Rabbit lmmunoglobulin G (RIgG)

IgG fractions of anti-RlgG serum (Miles Laboratories, Inc., Kankakee, IlL) were prepared by a combination of Na2SO4 precipitation and DEAE-cellulose column chromatography (Ley, 1966). These IgG fractions were coupled to microcrystalline cellulose activated with CNBr (Axen, 1967). The immunoadsorbents were suspended at a concentration of 5% in buffer A and used for EIA.

EIA Procedures

EIA of tmns-llb was performed by the double antibody solid phase (DASP) method (van Weemen, 1971; Comoglio, 1976). To 0.1 ml of standard trans-llb or plasma sample, was added 0.1 ml of anti-trans-I serum, 1:20,000, diluted with buffer A and allowed to stand at 4°C for 20 hr. To this mixture was added 0.1 ml of the hapten-enzyme conjugate solution containing about one p.U 13-galactosidase activity. After incubation at 30°C for 2 hr, 0.1 ml of the anti-RlgG antibody-coupled cellulose suspension was added and the mixture was shaken at 30°C for another 2 hr. After centrifugation, the solid phase was suspended in 0.5 ml of the substrate reagent containing 10 p.g of 4-methylumbelliferyl-~-D-galactoside (P-L Biochemicals, Inc., Milwaukee, Wis.) as substrate (Woollen, 1965), and incubated at 30°C for 60 min. The reaction was terminated by adding 4 ml of 0.05 M glycine buffer (pH 10.5) and the suspension was centrifuged. The fluorescence intensity of the supernatant was measured with a fluorometer at an excitation wavelength of 365 nm and an emission wavelength of 450 nm.

Gas Chromatography-Mass Spectrometry (GC-MS) of Transolib

Trans-llb in guinea-pig plasma was extracted as described above. After trimethylsilylation by bis(trimethylsilyl)-trifluoroacetoamide (Tokyo Kasei Kogyo, Japan) in acetonitrile, it was subjected to mass spectrometric analysis in conjugation with gas chromatography.

The mass spectrometer used was a double-focusing type, JEOL-JMS-01SC (Nihon Denshi Co., Ltd., Japan). The chromatograph was equipped with a 2-m × 2-mm glass column

Labeled Immunoassay of fl-Adrenergic Agent 7

packed with 4% OV-17 on Gas-Chrom Q (Shimadzu Ind. Co., Japan), 100-120 mesh, and was interfaced to the mass spectrometer inlet system. The column was operated at 230°C and the injection port and interface were maintained at 280°C. Helium was used as carrier gas. The ion of role 382 which showed the presence of trans-IIb was selectively monitored by the ion detector (Hammar, 1968).

RESULTS

Antiserum Spectral analysis was performed to determine the number of haptenic molecules per carrier protein (Grota, 1976). SucBSA gave soluble conjugates with an apparent molar ratio greater than 20, whereas intact BSA gave those with an apparent molar ratio of only about 10.

The sera obtained were tested for antibody production by the Ouchteriony method and PHA. Precipitin lines were observed opposite wells containing the hapten-OA conjugates, while there was no reaction with OA alone. HA reactions occurred at a serum dilution of more than 6000 when conjugate-sensitized erythrocytes were used, while there was no reaction with OA- sensitized erythrocytes. These data indicated that the antiserum was sensitive to the hapten compound trans-I.

The specificity of the anti-trans-I serum was studied by both PHA and EIA (Table 1, Figures 1 and 2). Antiserum against trans-I bound to the trans isomer of lib or lllc much more selectively than to the c/s isomer. HI titer for the trans isomer of lib or lllc was 60-200 times that for the cis isomer. Trans-llb and -Illc were the most cross-reactive tetralin derivatives tested, while trans-IIa, -Ilia, -lllb, and -llld were not significantly recognized unless their concentrations were 100 times greater. Trans-llc and -llIe showed a cross-reactivity of 2-4% of trans-l. These results demonstrated that the antibody against trans-I recognized the substitutions at the C2 position on the tetralin ring but not the substitutions at the C5 position. Other fl-adrenoceptor direcUng agents, such as propranolol and isoproterenol, which have the same

Figure I Cross-reactivity of trans- and cis-llb compounds in the enzyme immunoassay using antibody raised against trans-/.

60

~I 4o .£

u C

~ 20 LL

0

- - - - - - ~ o o Cis-IIb

Trans- I i b ~

\ . \

10 0 10 2

Tetralin compd.~ lib) ( ng/ml )

104

6 0 -~J

C

r-

r- oJ

O

14_

40

20

~ ~ ; ' " ~ ; Z I I I I I I " " S x ~ ° L ~ ~ a ..... b,d - ~ .

o x.

o

o .

H O ~ o C

N H R

O H

0 I0 ° 102 104

Trans- III (nglml)

a : R=H (: : R= C H ( C H 3 ) 2

b : R = C H 3 d : R= C ( C H 3 ) 3

C H 2 e : R=CI~ CH 2

C H 2 /

Figure 2 Cross-reactivity of various trans-lll compounds in the enzyme immunoassay using antibody raised against trans-/.

Figure 3 Standard curve for trans-llb (©) and curve for serially diluted plasma (@) measured by enzyme immunoassay.

60

<..

c

u e-

u

0 m g..

40

20

0

° ~ o * ~ Sample dilution

Transollb ~ '~oo * ~ . curve standard ~ \ c u r v e ~ o \ \ .

°~o I L , L

1 / 1 6 1 / 4 1/1

Sample dilution factor

, L

100 101 102

Tetralin compd. ~ lib a (ng /ml )

103

Labeled Immunoassay of P-Adrenergic Agent 9

Table 2 Recoue y of trans-lib added to guinea-pig plasmaa

Added

(w/ml)

Recovered (gross)

Found Expected

hs/mD

Recovered (net)

Mmll I%)

- 7.9 - 0 -

5.0 13.2 12.9 5.3 106 10.0 17.6 17.9 9.7 97 20 31 28 23 115 40 53 48 45 113 80 82 88 74 93

a Mean k SD = 105 + 8%.

functional groups on the side chain as trans-1, showed a cross-reactivity of less than 0.1%. The probable reason was that the antibody recognized the tetralin moiety as well as the substitutions at the Cz position.

Standard and Sample Dilution Curves

Figure 3 shows a typical standard curve of EIA of tmns-IIb. Anti-trans-I serum was used at a dilution of l:ZO,OOO. The results showed that standard trans-IIb was measurable over the range of 1 - 1000 &ml.

Plasma specimen from guinea pig given a single oral dose of trans-IIb was serially diluted 1:l to 1:16 with control plasma and extracted as described above. As shown in Figure 3, the curves for standard trans-IIb and for the serially diluted plasma were parallel.

The coefficients of variation were 1.3-5.1% (within assay) and 4.5-10.4% (between assay). These results showed that the immunoassay was reproducible.

Measurement of Trans-IIb Added to Plasma

Various amounts of trans-IIb were added to the plasma to give final concentrations of 5.0- 80 ng/ml. The mixtures were extracted and assayed for trans-IIb. The mean value was 105% of the amount of trans-IIb added (Table 2).

Comparison of EIA and GC-MS for Measurement of Trans-IIb

The concentration of trans-IIb in 18 plasma samples was determined by both EIA and GC-MS. In three samples, the concentration was less than the lower limit for detection by EIA (0.1 &dl). In the other 15 samples, there was a good correlation (r = 0.99) between the values for trans-IIb determined by the two methods (Figure 4).

Plasma Levels of Trans-IIb in Guinea Pigs

Figure 5 shows the results obtained from EIA analyses of trans-IIb in guinea-pig plasma. Free tmns-IIb levels became maximum in the first 30 min and decreased to an undetectable level by 8 hr after administration. The conjugated trans-IIb level, however, became maximum l-4 hr after administration, followed by a much slower decline. These results agreed with those of metabolic studies of 3H-labeled trans-IIb (Kanai, Y., personal communication), which indicated that the radioactivity in plasma became maximum 1.5 hr after administration with a half-life of about 4 hr, and that the radioactivity was mostly due to 3H-trans-IIb glucuronide.

10

m "D &

a

o e-

E E

¢)

E N e-

U,I

200

160

120

80

40

0

O / ..." o / . . ."

..."

o,.

0

i i

4 0 8 0 120 i

160 200

G C - M S method ( u g t d l )

Figure 4 Correlation between trans-llb concentrations in plasma measured by the GC-MS method and enzyme imrnunoassay, y = 1.04x - 3.52; r = 0.99.

Figure 5 Plasma leuels of trans-llb determined by enzyme imrnunoassay after a single oral administration of 5 mg/kg of trans-llb. Each plasma was assayed for trans-llb before (×) and after (0) treatment with fl-glucuronidase and sulfatase. Each point represents the average of a group of three guinea pigs.

. . . . . . r i -

100l o I ° ~ o /: 10 ~ o

-~ l "" " " . ,

a " Free trans-IIb " ' ' " " , , , . ,,,.. ,

0.1 ...... x. , , ,

f~

0 2 4 8 24

Time (hours)

Labeled lmmunoassay of ~-Adrenergic Agent 11

DISCUSSION

As described in the present paper, a method was developed for measuring a jff-adrenergic agent by immunoassay using/~-galactosidase.

~Galactosidase has been used to label antigens or antibodies because of its high turnover number, availability, and stability (Dray, 1975; Kato, 1975; Exley, 1977). The enzyme was coupled to the hapten under mild conditions with the help of carbodiimide and the enzyme activity decreased to 20-30% of the original value by these procedures. As this hapten-enzyme conjugate, when stored at 4°C, was stable for more than 6 months, the same preparation was used throughout the present study and resulted in good reproducibility of this immuno- assay.

For the preparation of antiserum, trans-1 was chosen as the hapten and coupled to BSA at the C5 position on the tetralin ring. The anti-tmns-I serum completely cross-reacted with trans-llb and -I]lc which are different from trans-I only at the Cs position, but the antiserum was highly sensitive to the stereospecificity at the C1-Cz bond and to the substitution at the C2 position responsible for the /3-stimulating activity. These results were compatible with the evidence that the carrier molecule caused steric hindrance between the drug and antibody- producing sites, directing antibody specificity to that portion of the hapten furthest removed from the functional group linked to the carrier protein (Ertanger, 1973; London, 1976).

In order to study the specificity of the antiserum, PHA was employed as well as EIA. The former immunoassay has the advantages of rapidity, simplicity of operation, and specificity (Adler, 1971), although it is not as accurate as EIA. Therefore, it is expected to be qseful for screening antisera and studying their cross-reactivity. Chick erythrocytes were used in the PHA since they gave reproducible results and sharply contrasting patterns between agglutinated and nonagglutinated cells (Itoh, 1976; Iwasa, 1977a). The sensitized chick erythrocytes main- tained their initial sensitivity to the antiserum after storage at 4°C for 3 months more by freeze- drying in vacuo (lwasa, 1977b).

Separation of antibody-bound and free conjugate is very important in EIA. We used the DASP method, which resulted in 3-10-fold higher sensitivity and more reproducibility than the single antibody solid phase (SASP) method using anti-trans-I antibody-coupled cellulose. The probable reasons for the higher sensitivity and more reproducibility of the DASPmethod are that the amount of anti-RlgG antibody-coupled cellulose in DASP systems is not critical since it has to be used in excess, and that excess of anti-trans-I antibody-coupled cellulose in SASP systems impairs the sensitivity of the test systems. Van Weemen and Schuurs (1971) also showed that DASP systems were more sensitive and reproducible than SASP systems in the EIA of human chorionic gonadotropin.

Plasma specimens containing trans-llb were deproteinized and extracted with organic solvents. These procedures were necessary to prevent a loss of sensitivity due to impurities in plasma and to separate free trans-IIb from conjugated ones.

Enzyme activity was monitored using 4-methy]umbellifery]/3-D-galactoside as substrate in this paper. The chromogenic substrate, o-nitrophenyl-/3-D-galactoside, was also used and it resulted in 5-20-fold lower sensitivity of the test system. But it contributes to the possible utilization of simple spectrophotometric equipment which is available in any clinical laboratories.

The performance and validity of this EIA for trans-IIb were compared with those of the GC-MS method. Both assays correlated well and had a similar detection limit. We concluded that by using a ~-galactosidase-labeled hapten, quantitative EIA can be performed satisfactorily, as pointed out by Lauer and Erlanger (1974). In addition, such an immunoassay can be done in any routine laboratory without radio-labeled compounds, specialized equipment and expensive technology.

This EIA model can be extended to many haptens by linking them to ~-galactosidase and thus used for metabolic studies of drugs in medical laboratories.

12 Susumu lwasa et al.

The authors are grateful for the helpful advice offered by Drs. M. Nishikawa, M. Hori, and T. Matsumoto during the course of this study. They also thank Messrs. Y. Oka, K. Itoh, H. Sugihara, M. Motohashi, and A. Miyake for the generous gift of various tetralin compounds, and Messrs. T. Saijyo and I. Yoshida for the technical advice.

REFERENCES

Adler FL, Liu C-T (1971) J Immunol 106: 1684. Axen R, Porath J, Ernback S (1967) Nature (Lond) 214: 1302. Comoglio S, Celada F (1976) J Immunol Methods 10: 161. Dray F, Andrieu JM, Renaud F (1975) Biochim Biophys Acta 403: 131. Erlanger BF (1973) Pharmacol Rev 25: 271. Evans ME, Walker SR, Brittain RT, Paterson JW (1973) Xenobiotica 3: 113. Exley D, Abuknesha R (1977) FEBS Lett 79: 301. Grota LJ, Brown GM (1976) Endocrinology 98: 615. Hammar C-G, Holmstedt B, Ryhage R (1968) Anal Biochem 25: 532. Itoh H, lwasa S (1976) J Clin Microbiol 4: 188. Itoh K, Motohashi M, Kuriki H, Sugihara H, Inatomi N, Nishikawa M, Oka Y (1977) Chem

Pharm Bull (Tokyo) 25: 2917. lwasa S, Itoh H, Hori M (1977a) Arch Virol 54: 241. lwasa S, Yoshida I, Itoh H, Hori M (1977b) J Clin Microbiol 6: 176. Kato K, Hamaguchi Y, Fukui H, Ishikawa E, (1975) J Biochem (Tokyo) 78: 235. Kitagawa T, Aikawa T, (1976) J Biochem (Tokyo) 79: 233. Klotz IM (1967) In Methods in Enzymology, Vol. XI. Ed., CHW Hirs. New York and London:

Academic Press, p. 576. Lauer, RC, and Erlanger BF (1974)Irnmunochemistry 11: 533. Ley HB, Sober HA (1966) Proc Soc Exp Biol Med 103: 250. London J, Moffat AC (1976) Analyst 101: 225. Martin LE, Hobson JC, Page JA, Harrison C (1971) Eur J Pharmacol 14: 183. Miyake A, Kuriki H, Tada N, Nishikawa M, Oka Y (1977) Chem Pharm Bull (Tokyo) 25: 3066. Nishikawa M, Kanno M, Kuriki H, Sugihara H, Miyashita O, Itoh K, Motohashi M, Oka Y,

Sanno Y (1975) Life Sci 16: 305. Oka Y, Motohashi M, Sugihara H, Miyashita O, Itoh K, Nishikawa M, Yurugi S (1977) Chem

Pharm Bull (Tokyo) 25: 632. Ouchterlony O, Nilsson LA (1973) In Handbook of Experimental Immunology, 2nd ed. Ed.,

DM Weir. Oxford: Blackwell Scientific Publications, chap 19. Sever JL (1962) J Immunol 88: 320. Sugihara H, Ukawa K, Kuriki H, Nishikawa M, Sanno Y (1977) Chem Pharm Bull (Tokyo)

25: 2988. Van Weemen BK, Schuurs AHWM (1971) FEBS Left 15: 232. Wallenfels K, Zamitz ML, Laule G, Blender H, Keser M (1959) Biochem Z 331: 459. Wolberg G, Liu C-T, Adler FL (1969) J Irnmunol 103: 879. Woollen JW, Walker PG (1965) Clin Chim Acta 12: 647.