Characterization of immunoreactive endothelin in human urine

Vol. 182, No. 3, 1992 February 14, 1992

BIOCHEMICAL

CHARACTERIZATION

AND BIOPHYSICAL RESEARCH COMMUNICATIONS Pages 1506-1513

OF IMMUNOREACTIVE IN HUMAN URINE

ENDOTHELIN

Koichi Am?, Tomoyuki Koshi’, Yuriko Ehara’, Toshiyuki Edano’, Masao Ohkuchi’, Mitsuteru Hirata’ and Tetsuro Okabe’ ‘Tokyo Research Laboratories, Kowa Co. Ltd., Tokyo 189, Japan ?he Third Department of Internal Medicine, Faculty of Medicine, Tokyo 113, Japan Received

January

University of Tokyo,

9, 1992

Summary: We developed three antibodies, specific and sensitive to endothelin-1 (ET-l), and established two sandwich and three competitive enzyme immunoassays (EL%). By using these EIAs, large immunoreactive ET (IR-ET) of molecular weight 10 k Da was identified as a main component of IR-ETs in human urine. This large IR-ET, which reacted with two antibodies specific for N-terminal region of ET-l but not with the antibody against Cterminal peptide of ET-l, was partially purified by six-step procedure and examined by Western blotting after SDS polyacrylamide gel electrophoresis. The large IR-ET was detected as a single band at molecular weight of 10 k Da both in reduced and non-reduced conditions. From these results, the large IR-ET was thought to consist of a single polypeptide chain and possess the steric restricted N-terminal region of ET-l. 0 1992 Academic Press, Inc. To investigate pathophysiological

role of endothelin (ET), which is the most potent

vasoconstrictor peptide consisted of 21-amino

acid residues (l), specific and sensitive

antibodies for ET were developed and several kinds of enzyme immunoassays @IAs) for evaluation ET concentration were established in many groups. Elevated levels of ET-l

in

body fluids of some patients have been reported from many institutes, however, the values of ET concentration in body fluids, especially in plasma, were different among institutes. These controversial results were thought to be attributed to the specificity of every antibody for ET and the extraction methods of immunoreactive ET (IR-ET) hand, some groups reported that IR-ET-like Abbreviations; ET, endothelin, B-GAL, B-D-gahtOSidase, horseradish peroxidase. 0006-291X/92 Copyright All rights

from plasma. On the other

substances (large H&ET), other than ET-1 and

IR-, immunoreactive-; ElA, enzyme immunoassay; HPLC, high-pressure liquid chromatography; HRP,

$1.50

0 1992 by Academic Press, Inc. of reproduction in any form reserved.

1506

Vol.

182,

No.

big ET-l, filtration

3, 1992

BIOCHEMICAL

AND

BIOPHYSICAL

RESEARCH

COMMUNICATIONS

existed in plasma at molecular weights of 6 k Da (2, 3) or 11.6 k Da (4) by gel chromatography.

Although large IR-ET has not yet been well defined, it reacted

with ET specific antibodies and the dilution curve of large lR-ET was parallel to the ET-l standard curve (4).

Therefore, large IR-ET

results of EIAs for ET concentration.

was also one of the candidates affecting the

To determine the exact levels of ET-l,

it is urgent to

purify and identify these IR-ETs. We identified large IR-ET as a main component of IR-ETs

in human urine by our ET,

and here, we describe partial purification and several characteristics

MATERIALS

of this large IR-ET.

AND METHODS

Pep&&: ET-2, ET-3, porcine big ET-1[1-391 and Sarafotoxin S6b were obtained from Peptide Institute Inc. (Osaka). ET-l and following ET-l related peptides, ET-1[16-211, ET-1[4-lo], porcine big ET-1[22-391 and type B ET-l (ET-1 isomer linked with Cys’-Cys” and Cys3-Cysl’ bonds) and nicked ET-l (ET-1 cleaved at Lys9-Glu” bond by lysylendopeptidase) were obtained as previously described (5). ET-1[1-161, ET-l[l-191 and ET-1[1-201 (all containing Cys’-Cys” and Cys3-Cys” bonds) were prepared by cathepsin D and/or carboxypeptidase Y digestion and confirmed by amino acid analysis. . . -of ET-l and ET-1[16-211, conjugated to porcine thyroglobulin and emulsified with complete Freund’s adjuvant, was respectively administered subcutaneously in foot pad of rabbits and boosted 5 times at 2-week interval. One week after the last injection, anti-ET-1-IgG (ET-PA20) and anti-ET-l[lbal]-IgG (ET-PACl) were respectively obtained and purified with the use of Protein A-cellulofine column. A monoclonal antibody for ET-l (ET-MA51, IgG,) was prepared from BALBlc mice immunized ET-l conjugate (50 pg protein) as previously described (5). .. Qmpetrtrve EIG: Competitive EL4 was carried out as described previously (5). Briefly, ET-MA51, ET-PA20 or ET-PACl coated microtiter plates were incubated at 4°C for 16h with 100 pl of standard ET-l or samples in phosphate buffered saline (PBS) containing 0.05% (W/V) Tween 20 (PBS-T) and 100 yl of f3-D-galactosidase (g-GAL)-conjugated ET-l, diluted lOOO-fold with PBS-T. After washing each well with PBS-T, the bound enzyme activity was measured using 4-methylumbelliferyl-8-D-galactoside as a fluorescent substrate (5). &n&i&E& ET-MA51 or ET-PA20 coated microtiter plates were incubated at 4’C for 16h with standard ET-l (or sample, 200 pl) in PBS-T. After washing, 200 pl of B-GAL-conjugated ET-PACl solution diluted 500-fold with PBS-T, was added to the wells and the adsorbed conjugates were measured. Concentration Urine was collected from normal subjects and concentrated with a hollow fiber ultrafiltration system using HlP3-20 type cartridge (Amicon CO., USA). Gel The concentrated urine from 300 1, was precipitated with 50 % saturation of ammonium sulfate and collected by centrifugation (8,000 g x 30 min). The precipitate was dialyzed using SPECTRA/POR 3 against 20 mM Tris-HCl buffer, pH 7.4 (buffer A). The dialyzed solution was run on a Sephacryl S-200 HR column (2.6 x 94 cm) at 1.2 mUmin and eluted with buffer A containing 0.2 M NaCl. : The large IR-ET fractions obtained by gel exclusion chromatography were collected, loaded on a column of formyl- cellulofine coupled with 1507

Vol.

182, No. 3, 1992

BIOCHEMICAL

AND BIOPHYSICAL RESEARCH COMMUNICATIONS

ET-PA20 (5 mg/ml resine, 2 x 6.4 cm), washed with PBS containing 1 M K!XN

until & became less than 0.1 and eluted with 1 M citrate solution. The IR-ET fractions were dialyzed against PBS, and was passed through a formyl-cellulofine coupled with normal rabbit y-globulin column (5 mg/ml resin, 2 x 6.4 cm) in order to exclude some proteins non-specifically bound to IgG. . . wchromatoera~hv All the passed-through &verse-wure solution described above was then acidified with 2 N of HCl to pH 4.0 and applied on an Amprep C, column (500 mg). After washing with 0.1 % trifluoroacetic acid (TFA), large IR-ET was eluted with 80 % acetonitrile containing 0.1 % TFA. The eluate, evaporated in vacua, was performed on a wondasphere C,, column (3.9 mm x 15 cm) and eluted with a linear gradient from 0 to 80 % acetonitrile in 0.1 % TFA at 0.4 ml/min. SDS-PAGE We&m.!&U&blottine: The samples were analyzed by the method of Laemmli et al. (6), in 20 % acrylamide containing 30 % ethylene glycol gel. The proteins on the SDS gel were transferred to PVDF membrane (7), and detected by chromogenic dye after treatment with ET-MA51 conjugated with horseradish peroxidase (HRP).

RESULTS The cross-reactivities of three antibodies with ET-related peptides were shown in Table 1. ET-MA51

and ET-PA20 were specific for the structure of ET-1[1-161

to type B or nicked ET-l. residues from C-terminal

and less reactive

They also reacted with the peptides lacking several amino acid of ET-l,

but only weakly with big ET-l.

ET-PACl

recognized

Table 1. Cross-reactivities of threecompetitiveEIAs for ET-relatedpeptides cross-reactivity(%) ET-relatedpeptides ET-MA51

ET-PA20

ET-PACl

ET-l ET-2 ET-3 porcinebig ET-1[1-391 SarafotoxinS6b

100 158 50.1 15.1 0.43

100 16.4 co.02 0.15 0

100 100 100 0.51 10.4

ET-1[1-161 (cyd-Cys’S, cys’-cysll) ET-l[l-19 (cys’-cys”, cys3-cys”) ET-1[1-201 (Cyd-cys”, cys3-Cys”)

100

100

0

100

100

0

100

100

0

ET-1[16-211 ET-1[4-lo]

type B ET-1[1-211 (cyd-cys”, cy&cys15) nickedET-1[1-9, lo-211 (cysl-cysls, cys3-Cys”) porcinebig ET-1[22-391

1508

0 0 9.5

0 0 5.6

100 0 100

0.1

0.31

100

0

0

0

Vol.

182,

No.

3, 1992

BIOCHEMICAL

AND

BIOPHYSICAL

RESEARCH

COMMUNICATIONS

B l/100 l/20 l/5 l/200 l/)50 l/,10 y; . -.._-._. *-A__L_A

100

250 2 25

.:-. I

100

200 ET-1

300 ( rig/ml

400

I

0

500

1

0.1

)

ET-l

10 ( ngfml

100 )

Fig. 1. Dose response curves for ET-l in sandwich EIAs using ET-MA51 ( l ) and ETPA20 ( o ) (A) and competitive EIAs using ET-MA51 ( l ), ET-PA20 ( o ) and ET-PACl ( A ) (B) and serial difution curves of large IR-ET (broken line) (B).

ET-1[16-211

and was not affected with N-terminal

sequence or the steric structure, but C-

terminal Trp residue was very important. Using these antibodies, two sandwich (fig. 1A) and three competitive

EIAs (fig. 1B) were established.

The minimal detectable concentrations

were calculated 10 and 40 pg/ml, respectively, from the ET-1 standard curves.

Ten liters of human urine were concentrated with ultrafiltration

using YM-2

membrane,

and loaded on a Sephacryl S-100 HR column (1.6 x 98 cm). Fig. 2 illustrates those results

15

-10 F P L5 ":

0

50

100 Elution

150 Volume

200

( ml )

Fig. 2. Gel filtration profile of JR-ET in human urine measured by competitive EIA ( l ) and sandwich EIA ( o ) using ET-MA51. Arrows indicate the elution Positions of BSA (l), ovalbumin (2), chymotrypsinogen A (3), ribonuclease (4) and synthetic ET-1 (5). 1509

Vol.

182,

No.

3, 1992

BIOCHEMICAL

AND

BIOPHYSICAL

RESEARCH

COMMUNICATIONS

?

1.5.

i ET-l

I I

,' t

1.0'

0.5' ,'

I'

0

3

O

lo Fraction

0

2o 3o 4o 5o) Number ( 5 ml/tube

4

'60

:

'40

;

-20

h '; ::

,,'

L

0

10

,'

.'

,'

,'

,'

20

Retention

,,'

30 Time

f 46

5b

( min

)

=r

Fip. 3. Affinity chromatography of large IR-ET on ET-PA20 coupled formyl-cellulofine. Fig. Reverse-phase HPLC profile of large IR-ET. of synthetic ET-l.

measured by sandwich

Arrow indicates the elution position

and competitive EIAs using ET-MA51.

Two major IR-ET

peaks

were observed and the former peak (fr. 120-150) was detected only by competitive EL4 and calculated 1Ok Da of molecular weight from marker proteins. this peak reacted with ET-MA51 ET-l,

but not with ET-PACl

position of synthetic ET-l

and ET-PA20

in parallel to standard solution of synthetic

(fig. 1B). The latter peak (fr. 185-200) was eluted at the same and reacted not only in competitive but also in sandwich

To investigate the former IR-ET (large IR-ET), further purification

Serially diluted fractions from

as described in “MATERIALS

300 1 of human urine was conducted to AND METHODS”.

More than 80 %

of large IR-ET was tiactionated by ammonium sulfate, and large IR-ET from other IR-ET ET-PA20

could be separated

by gel filtration as same as illustrated in fig. 2. Large IR-ET

an Amprep C, column.

The following

reverse-phase

was eluted earlier than authentic ET-l

was recovered from

HPLC yielded over 60 % and large

(fig. 4).

After six steps of purification

procedure, the sample solution still contained large amounts of brown-black 5 shows the results of Western blotting of the partially purified IR-ET. non-reduced

bound to

tightly enough to be retained on the ligand even after a washing with 1 M K!XN

and was eluted with 1 M citrate (fig. 3). About 40 % of large IR-ET

IR-ET

EIAs.

conditions, large IR-ET

detected as a single band, while ET-l

pigments.

Fig.

Both in reduced and

migrated to the same position of 10 k Da and was didn’t react with ET-MA51 1510

in reduced condition.

Vol.

182,

No.

3, 1992

BIOCHEMICAL

AND

BIOPHYSICAL

RESEARCH

COMMUNICATIONS

,16949 14404 8159 6214

2512

Fig. 5. Western blotting of ET-l (A, B) and large IR-ET (C, D). Samples were applied without heating or treatment with 2-mercaptoethanol (A, C) and were heated for 3 min at 90°C in the presence of 5 % (V/V) 2-mercaptoethanol (B, D). Lane E contained molecular weight markers.

DISCUSSION It has been reported that the presence of IR-ETs, ET-1[1-391,

affect measurement of ET concentration

hemodialysis

patients (4) and rats (3).

We developed three ET-l competitive

EIAs.

which

specific antibodies, and established two sandwich

HPLC so that only ET-l

our EIAs, we detected large IR-ET

or big

in plasma of healthy human (2),

Ando et al. has reported that urinary IR-ET

itself on reverse-phase

differ from ET-l

and three

was co-eluted with ET-l

was present in human urine (8). By using

as a main component of IR-ETs

in human urine at the

molecular weight of 10 k Da by gel filtration chromatography (fig. 2). This discrepancy leads us to consider that our large IR-ET covalently bound to ET-l. filtration chromatography

would consist of ET-l

and a particular protein non-

We tried to separate the possible binding protein from ET-l.

was performed using several kinds of detergents and dissociation

reagents such as TRITON X-100, urea, guanidine hydrochloride, solution.

However,

Gel

large IR-ET

KSCN and others as eluting

couldn’t be divide into smaller components by any above 1511

Vol.

182,

No.

3, 1992

BIOCHEMICAL

AND

BIOPHYSICAL

RESEARCH

solution (data not shown). These results suggested that the large IR-ET two or more components but was a single chain polypeptide. ET-MA51

and ET-PA20,

The large IR-ET

standard

possessing two

and Cys3-Cys”) was necessary to be recognized by ET-MA51

and ET-PACl

that the large IR-ET

reacted with

(fig. 1B). From the results of cross-reactivities of three

antibodies with ET-related peptides (Table l), the structure of ET-1[1-161

and ET-PA20,

did not consist of

the dilution curves of which are parallel to the ET-l

curve, but not with the ET-PACl

disulfide bonds (Cysr-Cys”

COMMUNICATIONS

was specific for C-terminal region of ETs. These data suggest

would possess the similar structure to N-terminal

of ET-l

in the

molecule. To confirm these results, the large IR-ET

was further purified by gel filtration, affinity

chromatography using ET-PA20 and reverse-phase HPLC. Totsune et al. have reported that the C,* cartridge treatment could be used for the elimination

of large IR-ET

Tom plasma (4).

The low recovery of large IR-ET from the C, column in our experiment was consistent with their results, but in the following 2nd reverse-phase HPLC with the C,* column, the recovery became over 60 %.

The partially purified large IR-ET

was subjected to SDS-PAGE,

transferred to PVDF membrane and examined by Western blotting using HRP-conjugated ET-MA.51.

The large IR-ET

Interestingly,

the large IR-ET

migrated to a position of 10 k Da as a single band. reacted with ET-ml

both in reduced and non-reduced

conditions, while ET-1 reacted only in non-reduced condition (fig. 5). These results suggest that large IR-ET region to ET[l-163,

is a single chain polypeptide and possesses the similar steric restricted but is not affected by reducing agents.

Further studies are necessary to elucidate the complete structure of large IR-ET clarify its pathophysiological

role .

REFERENCES 1. Yanagisawa, M., Kurihara, H., Khnura, S., Tomobe, Y., Kobayashi, M., Mitsui, Y., Yaxaki, Y., Goto, K. and Masaki, T. (1988) Nature 332, 411-41s. 1512

and to

Vol.

182,

No.

3, 1992

BIOCHEMICAL

AND

BIOPHYSICAL

RESEARCH

COMMUNICATIONS

2. Saito, Y., Nakao, K., Itoh, H., Yamada, T., Mukoyama, M., Arai, H., Hosoda, K., Shirakami, G., Suga, S., Jougasaki, M., Morichika, S. and Imura, H. (1989) Biochem.Biophys.Res.Commun. 161, 320-326. 3. Saito, Y., Nakao, K., Shirakami, G., Jougasaki, M., Yamada, T., Itoh, H., Mukoyama, M., Arai, H., Hosoda, K., Suga, S., Ogawa, Y. and Imura, H. (1989) Biochem.Biophys.Res.Commun. 163, 1512-1516. 4. Totsune, K., Mom-i, T., Takahashi, K., Ohneda, M., Sone, M., Saito, Y. and Yoshinaga, K. (1989) FEBS Lett. 249, 239-242. 5. Koshi, T., Torii, T., Arai, K., Edano, T., Hirata, M., Ohkuchi, M. and Okabe, T. (1991) Chem.Phann.Bull. 39, 1295-1297. 6. Laemmli, U. K (1970) Nature 227, 680-685. 7. Towbin, H., Staeheiin, T. and Gordon, J. (1979) Proc.Natl.Acad.Sci. U.S.A. 76, 4350-4354. 8. Ando, K., Hirata, Y., Takai, Y., Kawakami, M. and Marumo, F. (1991) Nephron 57, 36-39.

1513