CA2100355A1 - Endothelin converting enzyme - Google Patents

Abstract

ABSTRACT OF THE INVENTION
Endothelin converting enzyme has been isolated in pure form.

Description

2 1L ~ 5 `

ENDOTHB~IN CONVER~ING ~N~YM~

Backqround of the I~vention Endothelin (ET) is a 21-amino acid vasoconstrictor peptide originally isolated from porcine endothelial cells, and having the amino acid sequence shown in Figure la and a molecular weight of 2492 (Yanagisawa et al., Nature 332, 411 ~1988)). Porcine endothelin is synthe-sized first in a preproendothelin form of 203 amino acids, which is cleaved by known processing endopepti-dases to a precursor form of 39 amino acids, which iscalled Big ET or BET. BET is then further cleaved to form ET by a putative enzyme designated by Yanagisawa et al. as "endothelin converting enzyme" (ECE), which cleaves off the C-terminal 18 amino acids from BET to form the active 21 amino acid peptide ET. The existence of ECE was postulated because the cleavage which occurs to produce ET from BET occurs at an amino acid sequence for which there is no known protease having that speci-ficity, e.g., to cut only between Trp-Val within such a sequence.
The human precursor form of ET consists of 38 amino acids, which must be cleaved by an enzyme recognizing an essentially identical cleavage site (Y. Itoh, et al. FEBS
Lett. 231, 440, 1988). Three human isoforms of ~T have 2r, been identified, and an ECE activity was inferred for the cleavage of all isoform precursors ~A. Inoue, et al.
Proc. Natl. Acad. Sci. USA 86, 2863, 1989).

-` 2~3~ ~

ET is one of the most potent vasoconstrictors known.
It is postulated to have an important function in the control of the cardiovascular system. For example, ele-vated levels have been implicated in various cardio vascular-related diseases such as essential hypertension, vasospastic angina, acute myocardial infarction, conges-tive heart failure, pulmonary hypertension, renal failure and shock. These data support the proposal that one or more of the endothelin isoforms, e.g., endothelin-l, may contribute to the pathogenesis of these diseases. There-fore, the putative "endothelin converting enzyme" t which produces active ET from its precursor BET, is also pre-sumed to play an important role in the regulation of ET.
Thus, for example, a pharmacologically active inhibitor of endothelin production, e.g., an ECE inhibitor, would be clinically useful in treating such cardiovascular diseases. For example, even in the absence of isolating and identifying the enzyme, researchers are studying the inhibition of the putative ECE enzyme in vivo, in whole cells and in cell-free extracts, as a means of providing therapies for the above diseases, based theoretically on preventing the formation of the contractilely active ET
from the precursor BET. Other cardiovascular therapies based upon administration of ECE itself are also contem-plated. Furthermore, the enzyme itself would be usefulas an endopeptidase having an amino acid specificity which is not otherwise available to protein biochemistry.
Various enzymes have been purified to varying degrees, characterized and shown to cleave BET to ET;
however, in each case, the cleavage of ET is not spe-cific, in that further cleavages of ET occur, resulting in a degraded product. Some of these enzymes which non-specifically cleave BET to ET have been identified as already-known peptidases such as pepsin, cathepsin D and E, and other aspartic proteases.
However, despite the interest in the field~ the putative ECE has thus far not been isolated or charac-~~~ 2~33a terized. Thus very valuable information is lacking, especially as it relates to the control of production of ET from its precursor forms.

Summary of the Inventio~l This invention provides a proteinaceous substance having the biological activity of cleaving big endothelin specifically to endothe]in at pH 7.2, with substantially no further cleavage of endothelin, and having a specific activity of at least about 30 U/mg of protein, and preferably at least 500 U/mg.
In particular, the invention provides a metalloendo-proteinase having said properties. In addition, the invention provides a metalloendoproteinase containing a metal ion, capable of binding strongly to an anion exchange resin and capable of being substantially completely inhibited by 1 ~M phosphoramidon, as well as being less than 50% inhibited by 10 ~M 1-[(phenyl-methoxy)carbonyl]-L-prolyl-L-leucyl N-hydroxyglycinamide (CK4919), in particular wherein the metal ion is a catalytically effective metal ion, e.g., zinc.
Another aspect of this invention provides a process for isolating endothelin converting enzyme from cells containing said enzyme, comprising disrupting the cells;
isolating the high speed membrane-containing fraction;
solubilizing the membrane-bound proteins;
separating the solubilized proteins by anion exchange, separating by size the thus-obtained fractions of similar charge density proteins containing endothelin converting enzyme activity; and isolating the thus-obtained fraction having an apparenk molecular weight as measured on a Superose 12 column o~ between 232 and 440 kilodalkons.

~ 2~3~

Still another aspect of the invention relates to additional purification steps in the purification of ECE.
In these cases, ECE activity-containing pools are ob~
tained which generally contain protein bands correspond-ing to molecular weights of, respectively~ 100 and 140 kD, when analyzed on SDS PAGE. These additional steps can be performed in any order after solubilization of the membrane fractions:
Heparin-Sepharose fractionation and/or ConA-Sepharose fractionation.
Yet another aspect of the invention provides endo-thelin converting enzyme, preparable by a process as described above, in particular when prepared from human bronchiolar smooth muscle cells or human lung fibroblast-like cells.
Still another aspect of this invention provides a method o~ screening compounds suspected of having endo-thelin converting enzyme inhibitory activity, comprising determining the amount of conversion of big endothelin to endothelin by a protein having the hiological activity of converting big endothelin to endothelin in the presence of said compound and, e.g., comparing with the amount of such conversion in the absence of such compound.
Upon further study of the specification and appended claims, further objects and advantages of this invention will become apparent to those skilled in the art.
Various other objects, features and attendant advan-tages of the present invention will be more ~ully appre-ciated as the same becomes better understood when con-sidered in conjunction with the accompanying drawings, and wherein:
Figure l shows the proposed proteolytic processing pathway for the conversion of preproendothelin to endo-thelin. The preproform of porcine endothelin-1 which contains 203 amino acids is believed to be converted to the 39 amino acid form referred to as big endothelin-1 by dibasic endopeptidases (cross-hatched arrows) and car-21~3 ~ 3 ~ ~

boxypeptidases (small curved arrows) as shown. Big endothelin-1 is then cleaved at the Trpn-Val74 bond by a specific endopeptidase referred to as endothelin convert-ing enzyme (cross-hatched arrow). The final product is the 21 amino acid peptide, endothelin, containing amino acids Cys53 to Trp~. Figure la shows the amino acid sequence of endothelin-l.
Figure 2 shows ECE elution from a Mono Q FPLC
column. A 170 ~1 aliquot containing 1.411 mg of deter-gent solubilized 150,000 x g HBSM cell pellet is loaded in 50 mM Tris, pH 8.0 (20C), 25 mM n-octyl-B~D-gluco-pyranoside. The sample is eluted with a 0 to 1.0 M NaCl gradient at 0.2 ml/min at 4C and 0.5 ml fractions are collected and assayed for ECE activity in the presence of 1 mM PMSF. ECE activity is expressed as pmol of ET pro-duced in a 200 ~1 assay.
Figure 3 shows ECE elution from a Superose 12 FPLC
column. Separate 100 ~1 aliquots of Mono Q fractions 14, 15 or 16 aré run on a Pharmacia Superose 12 column (10 mm x 30 cm; 10 ~ particle size) and eluted with 50 mM MOPS
pH 7.2, 250 mM NaCl and 25 mM n-octyl-~-glucopyranoside with a flow rate of 0.5 ml/min at 4C. One minute frac-tions (0.5 ml) are collected over a 60 min period and assayed for ECE activity in the presence of 1 mM PMSF as described below.
Figure 4 shows ECE activity in individual Superose 12 column fractions. Fractions in the high molecular weight range (>100 kDa) from each of the individual frac-tions making up the pooled samples in Figure 2 are assayed in the presence of 1 mM PMSF.
Figure 5 shows separation of Mono Q fraction 14 on Superose 12. Top trace: absorbance at 280 nm monitored during run, scale 1% = 0.002 OD. Middle trace: activity profile of fractions 14 and 15 separated on Superose 12.
(Data offset o~e fraction for detector-to-collector delay.) Bottom trace. size standards for this Superose column.

2 1 ~ ~ 3 ~ 3 Figure 6 shows the elution profile of ECE activity and optical density at 280 nm of a solubilized membrane ECE preparation passed over Heparin-Sepharose as described in Example l(g)(i);
Figure 7 shows the elution profile of ECE activity and optical density at 280 nm of a solubilized membrane ECE preparation passed over ConA-Sepharose as described in Example l(g)(ii);
Figure 8 shows the elukion profile of ECE activity and optical density at 280 nm off of a Superose 12 column of an ECE preparation passed over Heparin-Sepharose before the Superose 12 purification step as described in Example l(g)(iii);
Figure 9 shows the kinetics of ET production by ECE
and NEP.

The present invention provides purified endothelin converting enzyme, as well as a process for the isolation and purification of an enzyme which specifically converts human proendothelin-l (amino acids 1-38) or Big ET (B~T) to endothelin-l (amino acids 1-21) or ET. The enzyme ob-tained by this procedure produces only one major detect-able product which coelutes with genuine ET from a C-18 ~PLC column. The enzvme also gives only one major detectable band when run on a 7.5% SDS-PAGE gel. These data imply that the enzyme is homogeneous, e.g., substan-tially free of other proteases, inter alia, and that its major enzymatic activity towards the substrate, human BET-1, is conversion to the product, ET-1. By these criteria, the enzyme purified by this process is the proteolytic activity referred to by Yanagisawa et al. as "endothelin convertiny enzyme" or ECE.
By "proteinaceous substance" is meant a single molecular species or a complex, comprising one or more covalently or otherwise linked amino acid or protein sequences, with or without covalent modifications such as glycosylation or fatty acid acylation, with or without .: ' '. :
, 2 .1 -- 7 ~

one or more bound metal ions, etc., which have the indicated biological activity. Examples include dimers or higher oligomers of a single molecular species, or covalent or tightly bound non-covalent complexes of two or more different protein sequences.
With respect to the several proteinaceous substances described herein, the term "an apparent monomeric molecular weight of greater than x kilodaltons", is meant that the proteinaceous material has a property usually associated with proteins of that size on, e.g., electrophoresis or gel filtration, but is not necessarily actually a single protein chain of that size. For exam-ple, a protein may have an anomalous shape which causes it to behave similarly to a higher molecular weight spe-cies when analyzed on, e.g., gel filtration or electro-phoresis, which anomalous shape may be endogenous to the protein itself or may be due to covalent modifications such as glycosylation; it may be a single gene product that is later processed into two or more covalently linked protein chains, analogously to, e.g., insulin, etc.; it may result from proteolytic fractionation of a larger to one or more smaller species; from dimeric vs.
monomeric molecular weights, etc., or may be due to different gene products. Whether or not the activity in these various bands is due to different molecular spe-cies, in each case the proteins detected are present in fractions containing ECE activity as measured by the various assays disclosed herein. The molecular weight of the species as disclosed herein, therefore, is a physical measurement which is only one property of the claimed polypeptides. In every case, it is the enzymatic and other functional properties as disc]osed below which particularly define the claimed polypeptides.
Several proteinaceous substances have been purified according to the methods disclosed herein. The apparent molecular weights of these materials ~ary according to the method used for its detection. Thus, the material 2 ~0~3 prepared from HBSMC cells which is eluted from a Superose-12 column elutes between molecular weight standards of 232 and 440 kilodaltons, corresponding to an apparent molecular weight of about 400 k:ilodaltons. This matarial eluted from the Superose 12 column and analyzed using electrophoresis on a 7.5% sodium dodecyl sulfate polyacrylamide gel under reducing condit:ions ran as an about 205 kilodalton molecular weight species.
Subsequent analysis of material isolated the same way, but prepared from MRC-5 cells, and analyzed on SDS~PAGE
under reducing conditions demonstrated two bands corresponding to proteins having molecular weights of lO0 and 140 kilodaltons. Each of the proteins which correspond to the bands detected on SDS-PAGE, as well as the proteinaceous material detected in the 232-440 kilodalton molecular weight eluate on Superose-12 possess the enzymatic activities described herein. The 100 and 140 kilodalton proteins are further isolated ~rom each other using routine protein purification techniques. The 100 kilodalton protein has the biological activity disclosed herein. The 140 kilodalton protein has the biological activity disclosed herein.
The phrase "cleaving big endothelin specifically to endothelin at a pH of 7.2", of course, does not imply that the proteinaceous substance or protein of this invention will not cleave at other lower or higher pHs, only that the proteinaceous substance or protein of this invention will cleave at a pH of 7.2 and will have the other characteristics described herein.
As a measure of purity, the term "with substantially no further cleavage of endothelin" as used herein means that a sample having a specific activity of at least 30 U/mg, and preferably at least 500 U/mg, and most preferably greater than 10,000 U/mg, has an absence of detectable amounts of proteolytic activity which cleaves ET to pieces smaller than 21 amino acids, i.e., upon cleavage of 3 ~M BET for one hour, essentially no peaks 2 ~

g of material containing a (tryptophan) fluorescence-emitting group other than BET and ET are detectable after HPLC analysis of the reaction mixture, as disclosed below for analysis of the ECE reaction. Similarly, purity of ECE in this respect can be defined as the absence of substantial ET degradation in the ECE assay conditions described herein with ET-1 added at 0.2 ~M in lieu of BET-1.
As another indication of purity, the protein of this invention has a specific activity of at least about 30 U/mg of protein, and preferably at least 500 U/mg, e.g., values greater than 1000, 2000, 3000, 4000, 5000, 6000, 7000, 8000, 9000, etc., and most preferably greater than 10,000 U/mg, e.g., up to 50,000, 100,000, 500,000 U/mg, i.e., the activity of a 100% purified protein, wherein a unit of activity is defined as the amount of enzyme which converts 1 nmol of substrate, e.g., human BET-l (hBET 1), to product, e.g., ET-1, in one hour, at a starting concentration of 3 ~M of substrate, at 37C, pH 7.2.
These activities can be obtained by further purification steps, e.g., using affinity chromatography wherein an ECE
inhibitor is bound to a support resin; using preparative gel electrophoresis; using chromatofocusing; etc., and especially wherein the ECE is produced by recombinant techniques, as discussed below.
As a further measure of purity, the protein of this invention is characterized by the presence of substantially only a single proteinaceous substance as determined by SDS-PAGE, in that there is no more than one defined stained band using a Pharmacia PHAST System silver stain kit, when > 4 ng of protein is loaded in a lane, or, if more than one band is present, every band corresponds to a polypeptide involved in ECE activity.
As is true of other proteases, which can be identified with families of proteins having similar enzymatic, biochemical and structural properties, the activity of ECE has similarities with other proteases, 21~3 '~

notably ACE (angiotensin converting enzyme; EC 3.4.15.1) and NEP (neutral endopeptidase, enkephalinase; EC

3.4.24.11). Thus, both ACE and NEP can cleave hB~T 1;
however, these proteases do not have the speci~icity nor activity for hBET-l of ECE. The activity of ECE is distinguished from these enzymatic activities using various biochemical and kinetic tests. For example, ECE
is clearly distinguished from ACE by the fact that it is not inhibited by 10 ~M of the potent ACE inhibitors enalaprilat and benzaprilat. Distinguishing ECE from NEP
is less unambiguous, in that kinetic, biochemical and immunological properties are used to distinguish them comparatively. See Table VI, in which the comparative properties of the two enzymes are summarized. In particular:
1. The specific activity, and thus KCat/Km of ECE for hBET-1 is very much higher than the specific activity of NEP ~or hBET-1: see Table IV;
2. NEP will cleave BET to ET (and other products), but unlike the result using ECE, the amount of ET
produced by NEP is limited by the greater intrinsic rate of ET degradation by this enzyme: see Fig. 9;
3. NEP does not show the strict dependence on Cl which ECE shows, similarly to the distinction between ACE
and NEP (J.P. Swerts et al., Eur. J. Pharm. 53, 209-210 (1979)); and 4. Immunological crossreactivity using antibodies raised to NEP ([provided by Prof. A.J. Turner, Univ. of Leed, UK) can be used to show that preparations of ECE
contain little if any crossreacting material as a percent of activity present and these amounts at worst correlate with an insignificant amount of NEP activity.
In addition, two commercially available compounds, CK4590 (= carboxyphenylpropionyl-Leu; N-([R,S]-2-carboxy-3-phenylpropionyl)-L-leucine, a NEP inhibitor) and CK4919 (=Cbz-Pro-Leu-Cly-hydroxamate; l-[(phenylmethoxy3-carbonyl]-L-prolyl L-leucyl-N-hydroxyglycinamide, a 2~

collagenase inhibitor) (both compounds from Sigma, St.
Louis, M0) have been shown to inhibit NEP significantly more than they inhibit ECE, when tested in the ECE assay under conditions in which NEP generates ET from BET
linearly. See Table VII.
Substantially purified EOE according to this invention has numerous advantages over the unpurified ECE
disclosed in the prior art, e.g., the material as it occurs in nature or in partially purified form, in all cases at substantially lower purity. This invention for the first time provides ECE at specific activities at which the following advantages become apparent, e~g., at specific activities of about 20 U/mg and higher:
~1) substantially purified ECE can be used to screen potential ECE inhibitors most effectively at purities where the measurable results of the conversion of BET to ET result primarily from the specific cleavage of BET by ECE, as opposed to measuring the non-specific cleavage by other proteases which may be present in a less purified sample;
(~) substantially purified material can be used to prepare antibodies to ECE, whereas the use of less purified material would make this process much more difficult;
t3) substantially purified protein can be sequenced, which is extremely difficult if not impossible to do with impure proteins, in order to provide information that can be used to tailor nucleic acid probes for identifying and retrieving a clone containing the ECE gene from a genomic or cDNA library for further cloning and expression of the ECE gene; etc.
The proteinaceous substance or protein of this invention is isolated using a sequence of steps which are each per se routine in the isolation of membrane bound proteins of high molecular weight. The following description is intetlded to be illustrative of one method for isolating ECE.

2~3~

Suitable cell or tissue types for use as a source of ECE are any cell or tissue types which contain ECE that is expressed; for examples of cell or tissue types which have already been shown to contain ET, and thus by inference, ECE, see Table A:

~, 2~3 Table A
~I8~UE DIS~RIB~ION OF ~NDOT~BLIN
Oriqin Tissue 8pecies ~ef.
8rain Cortex Pig Spinal Cord Pig 2 Hypothalamus Pig Cerebellum Rat 3 Cerebrum Rat 3 Medulla Oblongata Rat 3 Pituitary Rat 3 ~eart Atrium Pig Atrium Rat 4 Ventricle Pig Ventricle Rat 4 Blood Ve~sel Aorta Rat 3 Umbilical Artery Human 5 Umbilical Vein Human 5 Ridney Inner Medulla Pig Outer Medulla Pig Inner Medulla Rat 6 Cortex Rat 6 Cortex Pig Kidney Rat 3 ~astrointe~tinal Duodenum Pig Traat Duodenum Rat 3 Stomach Rat 3 Intestine Rat 3 Colon Rat 3 Pancreas Rat 3 Other Lung Pig Lung Rat 3 Liver Pig Liver Rat 3 Spleen Pig Spleen Rat 3 Adrenal Gland Rat 3 Urinary Bladder Rat 3 Testis Rat 3 Placenta Human 5 Refs.:
1. Kitamura, K., et al., Biochem. Biophys. Res.
Commun. 161, 348-352 (1989).
2. Shinmi, O., et al., Biochem. Biophys. Res.
Commun. 164, 587-593 (1989).-4S 3. Matsumoto, H., et al , Biochem. Biophys. Res.
Commun. 164, 74-80 (1989)-.
4. Yanagisawa, M., et al., Proc. Natl. Acad. Sci.
USA 85, 6964-6968 ~1988~.

5. Haegerstrand, A., et al., Acta Physiol. Scand.
137, 541-542 (1989).

6. Xitamura, K , et al., Biochem. Biophys. Res.
- Commun. 162, 38-44 ~1989).

2~3~

In addition to HBSM cells, ~RC-5 cells, which are a human fetal lung "fibroblast-like" cell line, publicly available from ATCC, Rockville, MD as CCL171, have been used as a source of ECE in the results disclosed herein.
Other cell types can be routinely screened by one of skill in the art for the presence of ECE to determine if they would be suitable as a source of the enzyme.
In order to screen cell types for the presence of ECE and monitor samples for the presence of ECE during purifications, any suitable ECE assay can be used, e.g., an enzymatic assay, e.g., in which the production of ET
from BET is detected, an immunoassay, an HPLC
chromatogram, an electrophoretic gel analysis, etc. A
preferred method is to perform an assay in which the production of ET from BET is detected. This assay is performed by placing a sample suspected to contain ECE in an assay mixture containing a suitable buffer, e.g., any biologically suitable buffer effective in the neutral pH
range, e.g., about pH 6.5-8.0, e~g., MOPS, TRIS, phosphate, etc., and a predetermined amount of BET for a certain amount of time, and then detecting the formation of ET from BET by ECE. Suitable concentrations of NaCl or other source of chloride ion are also required for maximal ECB activity. As controls are used blanks containing no ECE and samples containing known amounts of ECE for quantifying the results. The detection can be performed by a number of means, as noted above. In a preferred method, the detection of ET is monitored by passing the assay mixture, after the reaction is terminated, e.g., by the addition of EDTA, over an appropriate liquid chromatography column, e.g., a Vydac C-l~ reverse phase HPLC column, and monitoring the eluted material, e.g., by fluorescence spectroscopy at, e.g, an excitation wavelength of 225 nm and an emission wavelength of 340 nm, for the presence of the peptide peak characteristic of the smaller ET polypeptide. Any of the usual methods of eluting material from colu~ns can ,,, 3 ;.~ ~

be used, e.g., a decreasing gradient of a buffer, e.g., sodium acetate, and an increasing gradient of an organic solvent, e.g., acetonitrile.
Since ECE is associated with the membrane ~raction of cells which express it, cells containing ECE are disrupted by routine methods, e.g., by high pressure, low pressure, grinding, osmotic disruption, etc., followed by removal of large material, such as unbroken and only partially disrupted cells and lysosomes, from the suspension, e.g., by low speed centrifugation, e.g., at 20,000 x g for 1 hr. The ECE-containing plasma membrane fraction is then separated from the soluble fraction by, e.g., high speed centrifugation for a suitable period of time; this is called the high speed membrane fraction, and is distinct from other membrane-containing fractions which contain additional undesired components, e.g., the lysosomes, or which do not contain the plasma membrane.
Thus, by high speed fraction is meant the fraction pelleted, by centrifugation at 150,000 x g for about 1 hr, from the supernatant remaining after pelleting the crude homogenate at 20,000 x g for about 1 hr. The membrane-bound proteins are then solubilized from the lipid portion of the membrane using a solubilization buffer containing a detergent suitable for gently solubilizing membrane proteins, e.g., n-octyl-~-D-glucopyranoside. Other suitable detergents are those which gently solubilize proteins from membranes without irreversibly denaturing the protein of interest, including, e.g., non-ionic detergents such as Triton X-100, Nonidet P-40, digitonin, Lubrol PX, C12E8, zwitterionic CHAPS, etc. In particular, Triton X-lOOR is useful when purification over Con~-Sepharose is used, as the sugar on ~-octylglucoside interferes with the enzyme binding to this matrix. Other suitable detergents can be determined according to methods known to one of skill in the art, e.g., according to methods in Methods in Enzymology 182, Guide to Protein Purification, M.P.
"

2~

Deutscher, ed., Academic Press, NY (1990), especially pages 247-255. The solubilized membrane fraction is then subjected to a preliminary fractionation on the basis of a particular physical characteristic, e.g., charge density, e.g., on a column, e.g., a Pha~macia MONO Q HR
5/5 column, and eluted from the column, ~e.g., with a detergent-containing buffer in a gradient of an agent suitable for eluting proteins from such a column, e.g., NaCl. Fractions are collected and tested for the presence of ECE, and ECE-containing fractions are pooled.
Other preliminary fractionation methods may also be used, e.g., based on other physical methods, e.g., size~ or based on affinity, e.g., by antibodies specific for ECE
bound to a column or by binding to a non-cleavable reversible inhibitor for ECE.
If analysis of the pooled ECE-containing fractions from said preliminary fractionation, e.g., by HPLC or gel electrophoresis, indicates that the ECE is not substantially purified, e.g., to at least a specific activity of 30 U/mg, and preferably at least 500 U/mg, most preferably greater than 10,000 U/mg, the pooled fractions are then subjected to a further purification step, preferably based upon a different physical property, e.g., a size fractionation on a column having a suitable size differentiative ability consistent with the unusually large apparent size (about 400 kDa) of ECE, e.g., on a Superose 12 HR 10/30 column, and eluted with a suitable buffer, again containing a suitable detergent to maintain the solubility of membrane proteins, such as the above-mentioned n-octyl-~-D-glucopyranoside. Fractions are collected and assayed for the presence of ECE, and EC~-containing fractions pooled. Similarly, other purification matrices can be used, preferably based upon different physical, biological or enzymatic properties, e.g., Heparin-Sepharose or ConA-Sepharose, each of which provides a differentiative ability based upon a property of ECE. In each case, the partially purified ECE is 2 ~

added to a column containing the matrix and eluted with a suitable buffer, again containing a suit:able detergent to maintain the solubility o~ membrane proteins. Depending on the origin of the material, the size of the batch to be purified, and other routinely determinable factors, these various purification steps can be used in any order to best effectuate the purification required.
The pooled material from each step is tested for purity, e.g., on a polyacrylamide-SDS gel, run against molecular weight standards, and protein bands detected by, e.g., silver stain or Coomassie Brilliant Blue dye.
It was found in this assay to be a monomeric protein of a very high molecular weight of greater than about 205 kilodaltons, substantially free of any other protein-stainable material, i.e., when 4-40 ng of protein was loaded on the gel, and using a stain which is capable of detecting bands containing > O.S ng or more of protein, no band of protein not in the larger-than 205 kDa band was detectable.
Further purifications of HBSM ECE as well as ECE
isolated from MRC-5 cells, when analyæed on SDS gels under reducing conditions, have resulted in the detection of essentially pure protein bands which migrate on the gel in positions which indicate molecular weights of 100 and 140 kilodaltons.
0.5 ng of the purified enzyme was subjected to 1 ~M
phosphoramidon, which is known to inhibit metallo-proteinases, and assayed for ECE as above, and was found to be completely inhibited thereby. It is also inhibited by other metal chelating agents, e.g., EDTA, EGTA and o-phenanthroline. Therefore, the puxified enzyme is also a metalloproteinase. Since ECE is a metalloenzyme, containing one or more metal ions, eOg., Zn'~ ions, it can also be prepared in an apoprotein form by removing (or, in the case of synthetic ECE prepared by recombinant DNA
methods, by not adding) zinc, e.g., by chelation usin~ a 2~

chelator such as, e.g., EDTA. Glycosylated and unglycosylated forms of the protein can also be prepared.
The conditions employed in each of these steps individually are conventional and routinely optimizable using conventional considerations, e.g., as described in Methods in Enzymology 182, Guide to Protein Purification, M.P. Deutscher, ed., Academic Press, NY (1990).
The purified ECE obtained by this process can be used for various purposes. In particular, it can be used to screen compounds which are suspected of having E OE
inhibitory activity. Such compounds are of interest as they may be useful in treating conditions in which excess levels of ET have been implicated, e.g., various cardiovascular related diseases such as essential hypertension, vasospastic angina, acute myocardial infarction, congestive heart failure, pulmonary hypertension, renal failure and shock. As a screening test, various methods may be employed to determine if ECE
is being inhibited, for example, a modification of an ECE
assay as described above, whereinl instead of a sample suspected of containing ECE being added to an assay mixture containing BET and buffer, a sample of the compound suspected of having ECE-inhibitory activity is added to a sample containing either a known amount of ECE
in buffer or a known amount of BET in buffer, and the reaction started by adding the missing ingredient (i.e., BET or ECE, respectively). The production (or lack thereof) of ET is then monitored in the usual ways described above, e.g., by comparison with blanks not containing any inhibitor. The ECE of this invention can also be used as a tool to characterize the optimum conditions for preparing ET from BET.
Suitable methods for performing screening tests for inhibitors of ECE according to this invention are analogous to those for screening for inhibitors of other enzymes, given the ECE of this invention, and particular protocols can be routinely optimized-~Y one of ordinary skill in the art. For example, 0.08 U/ml of E OE are incubated for 1 hr at 37C in a buffer c:ontaining 100 ~M
Na-MOPS, pH 7.2, 150 mM NaCl, 2.5 mM ~-octylglucoside, 30 ~M CaCl2, and 3 ~M BET, in the presence or absence of the compound to be screened. The reaction is stopped by addition of EDTA to 3 mM, and the resulting mixture is analyzed for the production of ET by separation o~ the mixture on HPLC, and quantification of the resulting ET
peak area.
Another use of the purif~ed ECE includes using it as an antigen for the production of antibodies, using routine immunological protocols, as well as for the production of monoclonal cell lines producing such antibodies, for use in immunoassays for detecting the presence of ECE in, e.g., clinical samples. All of the methods for this use are routine to one of ordinary skill in the art, using standard protocols, e.g., as described in Galfre, G. and Milstein, C., Preparation of Monoclonal Antibodies: Strategies and Procedures, in Methods in Enzymology 73, 3-46 (1973).
Yet another use for purified ECE is for the det~rmination of the amino acid sequence of the protein.
Either the entire sequence can be determined using standard protocols, or, particularly in view of the large size of the protein, partial amino acid sequences can be determined sufficient to enable one of ordinary skill in the art to prepare one or more probes, i.e., nucleic acid sequences, or sequences complementary to such nucleic acid sequences, which encode the amino acid partial sequences of ECE; such probes can then be used to probe a genomic or cDNA library, e.g. any of the standard available libraries, for clones containing the DNA
sequence for the ECE gene. Due to its size, probes ~rom more than one location in the protein, e.g.~ three, will preferably be used to ensure that the entire correct gene is present in a given clone; it may be necessary ~o join together two or more clones to Qht~ir~ a full~length gene, but such procedures are routine to one of ordinary skill in the art without undue experimentation. After a full-length cloned gene is obtained, the entire sequence of the gene and thereby also of the protein may be determined using routine DNA sequence analysis.
A further use for ECE, after the above protocol has been performed and a cloned full-length gene is obtained, is to transfer the cloned gene into an expression vector capable of expressing such a cloned protein in a host cell, e.g., using a baculovirus expression system ancl insect cells or a vector with a retroviral promoter and a mammalian cell system, in order to produce large amounts o purified ECE using standard genetic engineering protocols.
Methods for performing the sequencing and genetic engineering aspects of this invention are routine for one of ordinary skill in the art, e.g., by reference to any one of a number of standard references, e.g., Fritsch, E.F. and Maniatis, T., Molecular Cloning: A Laboratory Manual (2nd. ed.), Cold Spring Harbor, Cold Spring Harbor Laboratory Press (1990).
The ECE gene(s) so obtained can be used to prepare transgenic animals using fully conventional methods, e.g., according to the methods disclosed in U.S.
4,736,866. These cloning techniques can be further used, e.g., to provide gene therapies, and to create experimental animal models for disease states.
Still another use of ECE is for administration to patients to treat conditions characterized by lack of ECE
or insufficient ET, or a condition for which increased amounts of ECE or ET would be effective; for example, low blood pressure. The corresponding pharmaceutical preparations could be prepared and administered analogously to other cardiovascularly active enzymes which are administered for various purposes, e.g., tissue plasminogen activator, which can be used to regulate blood pres.sure, particularly for intravenous ~,, 2 1 0 ~ 3 ~ ~

administration. Typically, these compositions would be formulated with carriers usual in galenic pharmacy, such as, e.g., water and serum albumin for intravenous administration, in the presence of the usual additives, e.g., buffers, etc.

Without further elaboration, it is believed that one skilled in the art can, using the preceding description, utilize the present invention to its fullest extent. The following preferred specific embodiments are, therefore, to be construed as merely illustrative and not limitative of the remainder of the disclosure in any way whatsoever.
In the foregoing and in the following examples, all temperatures are set forth in degrees Celsius; and, unless otherwise indicated, all parts and percentages are by weight.

The entire disclosures of all applications, patents and publications, cited above and below, if any, are hereby incorporated by reference~

2~0~3~ja E ~ A M P L E 8 Exam~le 1: PuriPicatio~ of ~n~othelin Converting E~zyme ~EC~
a. Crude ~ell preparation The source of ECE is cultured human bronchiolar smooth muscle cells obtained according to the procedure of Twort and Van sreemen (Tissue ~ Cell 20 (3), 339-344 (1988)).
Cultured bronchiolar smooth muscle (HBSM) cells at passage 16 are grown to confluency in DMEM media containing 10~ fetal bovine serum, 0.05 mg gentamicin/ml, 105 U penicillin/ml, and 105 ~g streptomycin/ml in eight 850 cm2 (surface area) roller bottles. At this time, the cells are washed twice with 50 ml of physiological saline solution (PSS) containing 140 mM NaCl, 10 mM glucose, 5.3 mM KCl, 13 ~M EDTA, 1.5 mM CaCl2 and 5 mM HEPES, pH 7.4.
The cells in each roller bottle are scraped from the surface of the roller bottles into 25 ml of PSS, pooled, and then pelleted in a tabletop centrifuge at 100 x g.
The supernatant is discarded and the cell pellet weighed.
The pellet is then resuspended to a single cell suspension in PSS containing 10% sucrose and 0.1% sodium azid~ (10 ml solution/1.5 g wet cell pellet weight.) The cells are disrupted using a Parr Cell Disruption Bomb under 700 psi nitrogen for 5 min at 4C. The broken cells are centrifuged at 370 x g ~2000 rpm, Sorvall SS-34 rotor) at 4C. The supernatant is saved and the pellet resuspended in 2 ml PSS containing 10% sucrose and 0.1%
sodium azide. The cell disruption is repeated and the 3Q broken cells centrifuged at 370 x g. The supernatants are pooled-and centrifuged at 20,000 x g (16,000 rpm, Sorvall T-865 rotor) for 45 min at 4C. The resulting supernatant is centrifuged again at 150,000 x g (45,000 rpm, Sorvall T-865 rotor) for 60 min at 4C. The resulting pellet is resuspended in 0.5 ml PSS containing ---- - I0~-s~cros~ --o ~ diu~ azide. The protein concentration is dete~nined using the PIERCE Protein Assay Reagent with BSA as the skandard protein.
b. Solubilization of the 150,000 ~ g cruaa Nembrane ~ssM Cell Pellet _ _ The 150,000 x g crude HBSM cell membrane preparation is solubilized by adding Tris Solubilization Buffer containing 100 mM Tris-Cl pH 8.0 ~at 4C), 1% n-octyl-~-D-glucopyranoside and 1 mM PMSF. 170 ~1 containing 1.411 mg protein of the 150,000 x g HBSM cell pellet is added to 330 ~1 of Tris Solubilization Buffer and vortexe~.
The solution is centrifuged at 19,500 x g for 10 min.
The pellet is saved, and purification continued with the supernatant, which is loaded onto a Pha~nacia FPLC Mono Q
HR 5/5 (5 X 50 mm) column, pre-equilibrated with 50 mM
Tris-Cl, pH 8.0 (20C), 25 mM n-octyl-~-D-glucopyranoside. Unbound material is washed through the column using this initial buffer at a flow rate of 0.2 ml/min for 15 min. At the end of the 15 min period, a 25 min linear gradient from 0 to 1 M NaCl in 50 mM TRIS-Cl, pH 8.0 (20C) and 25 mM n-octyl-~-D-ylucopyranoside is run. The 1 M NaCl buffer is then pumped through the column for 5 min. Eighteen fractions of 0.5 ml each are collected throughout the run and assayed for endothelin converting enzyme (ECE) activity. The results are shown in Figure 2.
c. ECE A~say of Mono Q Fractio~s from the ~olubilized 150,000 x ~ ~B~M Cell Pellet Each fraction is assayed at a 1:10 dilution in the presence of 50 mM MO~S pH 7.4, 30 ~M CaCl2 and in the presence or absence of 1 mM phenylmethylsulfonyl fluoride (PMSF). The diluted column fractions and the assay buffer are pre-incubated for 30 min at room temperature before starting the reaction by the addition of 3 ~M
hwnan big endothelin (hBET). The reaction is terminated after 1 h at 37C by adding EDTA to 3 ~M and placing the samples on ice. Quantitation of product is done by loading 200 ~1 of the assay mixture onto a Vydac c 18 2 ~

Reverse Phase HPLC column. The program for sample elution is shown in Table I~

~ABLE~ IPLC GRl~DIENq! I~LUTION PROGR2~M
.. ,,,.,~ , -~.,.. _,, .. ~

5-5 75 _ lO

I . .

I .
1 26.01 90_ 10_ 35.0 90 10 , Flow rate is 1.5 ml/min, Vydac 218TP54 RP, 4.6 x 250 mm, 5 ~, 300 A pore size.
2 Time at which condition is reached.
3 Buffer A - 50 mM sodium acetate pH 6.5.
Buffer B - 100% acetonitrile.

Initially, the 90% 50 mM sodium acetate, pH 6.5, and 10%
acetonitrile is pumped ~hrough the column at a flow rate ~of 1.5 ml/min at 30C for 1 min. For the next 7.5 min, a linear gradient of 90% to 75% 50 mM sodium acetate and 10% to 25% acetonitrile is run, followed by 5 min of 75%
50 mM sodium acetate and 25% acetonitrile. The linear gradient is then continued from 75% 50 mM sodium acetate and 25% acetonitrile to 50% 50 mM sodium acetate and 50%
acetonitrile over 12.5 min. Finally, over the last 9 min the column is reequilibrated with 90% 50 mM sodium acetate and 10% acetonitrile.
a. Application of ~v~o Q Fractions 13, 1~, 15, 1~ on ~upero~e 1~ _ Separate 100 ~l aliquots of Mono Q fractions 13, 14, 15 and 16 are loaded individually onto a Pharmacia Superose 12 HR 10/30 (10 mm x 30 cm, 10 ~ particle size) -" 2 ~ 3 ~ ~

column connected to an FP~C and eluted with 50 m~ ~OPS, pH 7.2 containing 250 mM~NaCl and 25 mM n-octyl-~-D-glucopyranoside at a flow rate of 0.5 ml/min at 4C. One minute fractions of 0.5 ml are collected over a 60 min period and assayed for ECE activity as described below.
o. ECE As~aY o~ Su~erose Column Fractio~s To determine which fractions contain ECE activity, 50 ~1 of each succassive five or six fractions are pooled and assayed. The void volume of the column is approximately 9 ml; therefore, the fractions are pooled as follows: 14-19, 20-24, 25-30, 31-34, 35-40, 46-50, 51-55, and 56-60. These fractions are assayed undiluted in the presence of 30 ~M CaCl2 and 1 mM PMSF. The assay mixture is pre-incubated for 30 min at room temperature and the reaction initiated with the addition of 3 ~M
hBET . The reaction is incubated for 6 h at 37C and finally terminated with the addition of EDTA to 3 mM and placed on ice. 200 ~1 of each assay mixture is loaded onto a Vydac C-18 Reverse Phase HPLC column and analyzed as described previously. The fractionation scheme, yields and purification are shown in Table II:

~ABLE II - FRaCTIONATION 8C~ENE FOR ~CE

100 xg Cell 0.15 g/ml 10 ml 4 370 xg Disrupted 12 ml Cell Supernatants2 I _ I
20,000 xg 12 ml Supernatants 150,000 xg Pellet 8.3 mg/ml0.17 ml 0.3175 1 19,500 xg 2.38 mg/ml O.S ml Supernatants of the 150,000 xg Solubillzed Pellet3 Mono Q Fraction0.088 0.5 ml 2.41 7.6 14 mg/ml Mono Q Frac~on4-40 ng/ml 4.0 ml 588-58801855-18,550 14 (100 ~1), Superose Fractions 20-27 .
Cell pellet from 8 roller bottles of human bronchiolar smooth muscle cells, total yield was 1.5 g net weight of cells.
2 Cells were disrupted with a Parr Cell Disruption Bomb 3 One volume of the 150,000 xg pellet was solubilized with 2 volumes of 100 mM Tris-Cl pH 8.0 at 4C, 1~ n-octyl-~-D-glucopyranoside, 1 mM PMSF.
4 --, Not determined.
5 These data represent an apparent specific activity due to substantial ET breakdown in the impure stages of preparation.

2:~t~3~5 ECE activity is found in the pooled fractions containing Superose tubes 20-24 and 25-29 from Mono Q
fractions 14, and 15 and 16 (Figure 3). Each of the individual fractions making up the above pools is assayed individually at a 1:2.5 dilution in the presence of 30 ~M
CaCl2 and 1 mM PMSF. ~he assay mixture is pre-incubated for 30 min at room temperature and the reaction started ~y the addition of 3 ~M hBET. The reaction is incubated for 6 h at 37C and stopped by adding EDTA to 3mM and placing the samples on ice. 200 ~1 is loaded onto a Vydac C-18 column and run as stated previously. The results are shown in Figure 4. Comparison of the activity profile with the elution locations at standard proteins, shown in Figure 5, indicates a molecular weight of approximately 400 kDa.
f. SD~-PAGE of Mono O 14 8uperose Fraction~ 20-27 The Mono Q 14 Superose fractions 20 through 27 are pooled for a total volume of 2.5 ml and placed in an Amicon Centricon 10 centrifugal microconcentrator with a molecular weight cut off of 10,000 Da. It is centrifuged at 5000 xg for 90 min at 4C until a deadstop volume of 100 ~1 is reached. Two ml of 0.02M TRIS pH 8.0 is added in order to dilute the remaining detergent. The solution is again centrifuged at 5000 xg for 90 min at 4C until a final volume of 100 ~1 is reached. The 100 ~1 is then quick frozen in dry ice/ethanol and lyophilized.
The lyophilized powder is resuspended in 2.5 ~1 PHAST SDS-PAGE sample buffer which contains 10 mM TRIS-- HCl pH 8.0, 1 mM EDTA, 2.5% SDS t 20 mM DTT and 0.01%
Bromophenol blue. The sample is loaded onto a 7.5%
homogeneous SDS gel and run on a Pharmacia PHAST System.
The gel is stained with Pharmacia PHAST System silver stain kit. The number of proteins is noted and their molecular weights are estimated by comparison with SDS-PAGE standards. Only one prominent band of a molecular weight substantially greater than 205 kDa is observed.

2 ~

g. ~urther affinity ~ri~ic~tion In addition to the above-described ]purification procedure, which results in purifications of much higher specific activity than was previously possible, other affinity methods have been used to even further purify ECE.
i. Heparin-Sepharose purification:
HBSMC cells were processed as described above, and the membrane proteins solubilized. Solubilized membrane protein, 6 ml, was added to 120 ml of Heparin-Sepharose CL-6B (Pharmacia), in 50 mM Tris-HCl, pH 8.0, 50 mM NaCl, 25 mM ~-octylglucoside, and incubated overnight at 4 C
with gentle rotation. The solution was then poured into a chromatographic column and allowed to equilibrate and settle for three days. ~Excess buf~`er was removed from the top, and the column was attached to a gradient elution and UV detection system. The column was washed through with starting buffer, then eluted with a linear gradient of 0.05 to 1.5 M NaCl in 50 mM Tris-HCl, pH 8.0, 25 mM ~-octylglucoside, over 5 column volumes (280 ml) at 0.67 ml/min. Fraction of 10 ml each were collected, then assayed for ECE activity, with the results shown in Fig.
6.
ii. ConA-Sepharose purification:
About 400 mg of MRC-5 membrane proteins, solubilized as described above, except that Triton X-lOOR was used in place of ~-octylglucoside, was mixed with 54 ml of ConA-Sepharose (Sigma) in a total volume of 210 ml o~ 50 mM Tris-HCl, pH 8.0, 0.5% Triton X-lOOR, and equilibrated in the cold for 65 h. The slurry was poured into an FPLC
column and attached to an FPLC system (Pharmacia). The column was washed through, then eluted with a dual linear gradient of O to 0.~ M NaCl plus O to 0.5 M
~-methylmannoside over 9.4 column volumes (510 ml~ at O.3 ml/min. Fractions of 7 ml each were collected and assayed for ECE activity, with the results shown in Fig.

7.
,,, 2~ 3 iii. Heparin-Sepharose purification followed by Superose 12 purification:
Superose 12 (10/30, Pharmacia) was equilibrated with 50 mM Tris-HCl, pH 8 0, 50 mN NaCl, 25 ~M
~-octylglucoside. 0.5 ml of the Heparin-Sepharose pool isolated in Example l(g)(i), above was :Loaded and eluted in this buffer at 0.5 ml/min. Fractions of 0.5 ml each were collected and assayed for ECE activity, with the results shown in Fig. 8.
These exemplary additional purification steps can be used in any order after solubilization of the membrane-bound proteins from the crude cell homogenate. Thus, depending upon the technical requirements of the size of the purification, etc., one or more of these additional steps can be conducted before or after Superose-12 purification, or before or after Mono Q purification.
A summary of a pilot purification of ECE from ~BSMC
is shown in Table III, showing a specific activity yield of 25000 U/mg protein. (The step of using a Blue-Sepharose column purification is omitted in subsequent purifications due to its negligible effect on enhancing purification.) . : . . . . . . ... .

2 ~

T~BL~ III
Pilot Purification of ~CB fro~ 30 ~oller Bottles ~s~

Tatal Protein Total Specif1c Activity ¦ Total Act. h:tivity Puri-Vol. (ml) Cnnc. (mg/ml)Prote1n (~9) (Units/~93 ¦ (Ihlits) Reoovery (X) Fold _ . _ - _ _ _ -Membrane 0.41 16.9 7.0 _4.6 _ 32.2 Solub11i~ed _ 1.36 3.0 4.1 4.2 ~ 17.2 = 100 Q-Seph Pool 20. 0.0115 0.23 43.0 9.9 57 10.2 Blue-Seph Pool 3. 0.038 0.08 66.2 5.D 29 15.8 Dialy2ed Pool 3. ND* ND ND 4.2 _ 25 _ ND
Heparin-Seph 7. 0.0000660,000471 1062. O.S 3 253.
IPool_ l Superose 12 Pool 19 ND ND ~ S000.(est) 0.2 1 ~es~c) l , * ND = Not Determined. Protein concentration is too low to detect by standard methods.
= Estimated from SDS-PAGE

BxamPle 2: Characterizatio~ Of ECE
Gel filtration studies as disclosed herein show that ~CE has an usually high apparent molecular weight of approximately 400 kDa under these conditions. However, SDS-PAGE under reducing conditions on various preparations have indicated that ECE preparations of high specific activity contain proteins migrating on such gels which correspond to molecular weight markers of approximately 205, 140 and/or 100 kilodaltons. The SDS-PAGE data indicate this is a monomeric molecular weight.
The purified enzyme is completely inhibited by 1 ~M
phosphoramidon, which indicates that it is a metallo-(Zn2-containing) endoproteinase.

Exa~ple 3: Distinguishing ~CB from N~P
Studies were conducted to distinguish ECE from NEP
(neutral endopeptidase, enkephalinase, EC 3.4.24.11).
NEP was obtained from purified recombin~nt rabbit 2.~3~

preparations. ECE was prepared from HB~C,MC cells, as disclosed above.
a. Ri~eticq NEP has properties which are superficially similar to ECE; e.g. NEP will produce ET, as well as other products, under conditions identical to those for ECE
assay; however, the kinetic rate constant for B~T --> ET
is ~uch lower than for ET --> degradation products. See Table IV (KCat and Km are determined from least-squares fit of concentraton-dependence of the rates, according to methods routine in the art):
TAsLE IV
~INæTIC8 OF r~P A~D h~C~
ENZYME-- SUBSTRATE - (~) C(Amin~1) K ~ M
NEP Enkephalin l.0 24332433 NEP ET-l 2.3 131 57 NEP big ET-l 12. l.2 0.l ECE big ET-l 12.0 83 7 ______.__________________________________________________ aEstimated bAssumptions: ~AX = 50, 000 nmol hr~1mg~
M.W. = l00,000 daltons Another comparison which can be made is based upon the kinetic prediction of the activity of NEP as an endothelin converting enzyme. Thus, the question that was posed was: using rNEP, at what % conversion of BET
- will the yield of ET-l plateau? This was answered using the ~ollowing calculations (Table V), by finding the concentration of ET at which the rate of BET --> ET-l equals the rate of further hydrolysis of ET-l:

~;

:

?,~0~
- 32 ~

TABLB V
r2a~P ~ ~C~: pR~DIcTIo~æ F~O~ ~TIC8 [bET-l] ET-l Synth [ET-l] ET-l Degr.
UM nm/min/mq ~M nm~min/mq INITIAL 3.0 26 0 0 AFTER 1% 3.0 26 0.03 19 CONVERSION
__________________ ________________~_~_ _________________ Therefore, ET-1 production from NEP should plateau at about 1% conversion = 6 pmol ET-l produced.
ET reaches an equilibrium amount in the reaction with NEP, but continues linear production with ECE (see Fig. 9). Furthermore, the rate of ET production from BET
by NEP is very much lower than that for ECE (50 U/mg vs >10,000 U/mg). Another important distinction is the NaCl-dependence of the activity. ECE is completely dependent on the presence of chloride ions (this has also been observed for metalloproteinase ACE) for substantial activity. Maximal activity is obtained at 150-200 mM
NaCl. For NEP with the same substrate and same assay conditions, much of the activity is present in the absence of salt, and maximal activity is not reached until 1.5 M NaCl. Finally, immunological assays (ELISA) for NEP in the above ECE preparations, both from HBSMC
and MRC-5 cells, were performed. Amounts reported were converted to "units of ECE" and compared to our measurements of the units actually observed. Less than 0.03% of the activity could be accounted for by any NEP
contamination. These results are summarized in Table VI:

2 ~ 3 ~ ~

~rABLE VI
Properties o~ B~ ~C,B Versu~
Purifie~l ~eco~ nt R~bbilt ~JEP

BET Cleavage Products ET E~ ETIS-21 ET(S.6 _ __ 1 Specific Activity (U/mg) _ 10 000 _ 50 _ ¦
Km for BET-1 12~ -10 aM
Time Course of ETIncreases linearly for Plateaus at 2h Production at Equivalent ~6 h ~1% conversion Initial Rates 3% conversion Molecular Weight Gel Filtration ~232-440 kDa 232 kDa SDS-PA6E (reduced)100 140 205 kDa 94 kDa Behavior on anion exchange Binds; eluted 0.2-0.3M Binds; eluted by 0.2- l (pH 8) . salt0.3M salt ._ _ _ Behavior on heparinBinds; eluted by 0.6-0.7M Binds; eluted by 0.6-agarose (pH B) _ ~ 0.7M salt _ Behavlor on Con A agarose Binds weakly; eluted by Binds; eluted by a-~pH 8) low a-methylmannoside methylmannoside plus NaCl Snake Plasma Inhibits contaminating No inhibiton of any proteases leaving ET as of the NEP products the sole ECE product from from BET
_ _ '3ET _ Salt Dependence No Activity: 0 M NaCl 25-50X Max Activity: Max Activity: 150-20û mM 0 M NaCl NaCl Max Activity:
40% Inhibition: ~ 1.5 M l.S M NaCl NaCl _ Inmunoprecipitation by Mouse Anti-human CALLA <20X ppt ND2 Mouse Anti-rabbit NEP 0% ppt ND
18BS 0% ppt ND

% Activ3ity accounted for ~0.03 100 by NEP _ _ 1 Assayed according to ECE protocol with 3 aM BET as substrate 2 ND = Not Determined 3 NEP by ELISA (AJ Turner) 9~

~ 34 -b. ~nhibition ECE and the other mammalian neutral neutral metallopeptidases NEP and ACE can also b~ distinguished by their inhibition by various standard metalloproteinase inhibitors. Neither NEP nor ECE is inhibited by the ACE
inhibitor enalaprilat (Merck). ECE is inhibited about 5-fold less potently, under identical assay conditions, by the NEP inhibitor CK4590 (= carboxyphenylpropionyl-Leu; N-([R,S]-2-carboxy-3-phenylpropionyl~-L-leucine, purchased from Si~ma: Ref.: M.-C. Fournie-Zaluski et al., J. Med. Chem. 26, 60 (1983)). It is inhibited about 14-fold less potently by the collagenase inhibitor CK4919 (=Cbz-Pro-Leu-Cly-hydroxamate; 1-[(phenylmethoxy)-carbonyl]-L-prolyl-L-leucyl-N-hydroxyglycinamide, Si~ma:
Ref. W.M. Moore et al., Biochem. 25, 5189 (1986)). In the latter case, a clear distinction can be made between the inhibition of ECE and NEP by use of 10 ~M CK4919:
under these conditions, ECE is less than 50% inhibited, while NEP is more than 50% inhibited. See Table VII:
T~BLE VII
INHIBITOR POTENCY - ICso's in "ECE" SCREEN
ICso (nanomolar) Ratio CK # ECE NEP ECEJNEP Structure 4590 3600 (4) 770 (2) 4.7 O ~ '-P' HO2~J~` Nli C02H
i'h 4919 33~000 (1) 2300 (2) 14.4 ' O
N~
-e.
1 Value determined by assay of NEP according to ECE
protocol, with 3 ~M BET as substrate, lin2ar ran~e of ET
~oductii~ ._._ _ ---------- ---- -- ~-~
____________ ____________________________ _________ ____ 2 ~

Although a variety of screening protocols can be used to find new inhibitors of ECE (see above), a suitable screening protocol is provided iLn Table VIII, below:
TABLB VII
~creeni~g Protocol Criteria which should be met by each ECE prep before screening:
1. ET-1 production from hBET-l is linear over time 2. ET degradation rate is <10% of production rate at final [ET] produced in (1) 3. ICso for phosphoramidon is approximately 1 nM
Assay conditions (and why):
1. 50 mM MOPS-, pH-7.2 (physiological pH, pH
optimum) 2. 2.5 mM ~-octylglucoside (detergent solubilized membrane enzyme) 3. 30 ~m CaC12 (enzyme is calcium-activated) 4. 150 mM NaCl (enzyme is chloride-activated;
essentially no activity in absence; compare to ACE) 5. 1 mM PMSF (inhibit serine proteinases) 6. 1 ~M amastat=i-n--~-inhibit aminopeptidases; allows ---easier observation of breakdown products) 7. 3 ~M hBET-l (substrate by definition; time and cost effective amount which is significantly below Km of 12 ~M) . 37'C, time and enzyme sufficient for 2-3%
reaction (initial rate measurement; typically 3-5 milliunits, 4-5 h; 1 unit = 1 nmol/hour ET
produced under these conditions) 9. Stop reaction with 3 mM EDTA, 4OC; quantify by HP~C, tryptophan fluorescence (sensitivity <1 pmole; <5% of control reaction) .
:

,. .~ 2 ~ r~ ,r~

It is also noted that screening for various inhibitors can be based upon either apparent IC50 or upon apparent K~: See Table IX:

AsL~ IX
CO~PARI~ON OF APPA~$~ IC50 vs APPAR~NT
~ASSUMPTION: Reversible, competitive inhibition) FO~ULA: Kl = ---------(1 + [S]/XM~
KM K, ENZYME SUBSTRATE r s ] UM UM
NEP ENKEPHALIN 0. 5 1. 0 0. 67*IC50 ECE Big ET-l 3.0 12.0 o. 80*IC50 ________________________________________._________ Therefore, expressing screening data as apparent Kl will lower the reported values for both enzymes, but will have little influence on the evaluation of the selectivity of individual CK compounds for these two enzymes. The calculation for BET as a substrate for NEP
can be calculated correspondingly.

The preceding examples can be repeated with similar success by substituting the generically or specifically described reactants and/or operating conditions of this invention for those used in the preceding examples.

From the foregoing description, one skilled in the art can easily ascertain the essential characteristics of this invention, and without departing from the spirit and scope thereof, can make various changes and modifications of the invention to adapt it to various usages and conditions.

Claims (22)

WHAT IS CLAIMED IS:

1. A proteinaceous substance having the biological activity of cleaving big endothelin specifically to endothelin at pH 7.2, with substantially no further cleavage of endothelin, and having a specific activity of at least 500 U/mg of protein.

2. A metalloendoproteinase containing a metal ion, capable of binding strongly to an anion exchange resin and capable of being substantially completely inhibited by 1 µM phosphoramidon, but less than 50% inhibited by 10 µM 1-[(phenylmethoxy)carbonyl]-L-prolyl-L-leucyl-N-hydroxyglycinamide.

3. A metalloendoproteinase of claim 2, having the biological activity of cleaving big endothelin specifically to endothelin at pH 7.2, with substantially no further cleavage of endothelin.

4. A protein of claim 1, further comprising a metal ion.

5. A protein of claim 4, wherein the metal ion is zinc.

6. A metalloendoproteinase of claim 2, wherein the metal ion is zinc.

7. An apoprotein derived from a protein of claim 4 by removal of the metal ion.

8. An apometalloendoproteinase derived from a metalloendoproteinase of claim 2 by removal of the metal ion.

9. A process for isolating endothelin converting enzyme from cells containing said enzyme, comprising disrupting the cells;
isolating the high speed membrane-containing fraction;
solubilizing the membrane-bound proteins;
separating the solubilized proteins by anion exchange, separating by size the thus-obtained fractions of similar charge density proteins containing endothelin converting enzyme activity; and isolating the thus-obtained fraction having an apparent molecular weight as measured on a Superose 12 column of between 232 and 440 kilodaltons.

10. Endothelin converting enzyme, preparable by a process of claim 9.

11. Endothelin converting enzyme, prepared from human bronchiolar smooth muscle cells by a process of claim 9.

12. Endothelin converting enzyme, preparable by a process of claim 9, wherein said cells are human bronchiolar smooth muscle cells.

13. A method of screening compounds suspected of having endothelin converting enzyme inhibitory activity, comprising determining the amount of conversion of big endothelin to endothelin by a protein of claim 1 in the presence of said compound.

14. A protein having the biological activity of cleaving big endothelin specifically to endothelin at pH
7.2, with substantially no further cleavage of endothelin to smaller peptides, and having a specific activity of at least 500 U/mg of protein.

15. A method of claim 9, further comprising, after solubilizing the membrane-bound proteins, the additional steps of:
fractionating the proteins on Heparin-Sepharose and/or fractionating the proteins on ConA-Sepharose;
said step or steps being performed in any order after solubilization of the membrane-bound proteins.

16. A proteinaceous substance of claim 1, wherein the specific activity is at least 1000 U/mg of protein.

17. A proteinaceous substance of claim 1, wherein the specific activity is at least 5000 U/mg of protein.

18. A proteinaceous substance of claim 1, wherein the protein moiety which effects the cleavage has an amino acid sequence corresponding to that of the human enzyme.

19. A metalloendoproteinase of claim 3, wherein the protein moiety which effects the cleavage has an amino acid sequence corresponding to that of the human enzyme.

20. The proteinaceous substance preparable by a process of claim 9.

21. A proteinaceous substance of claim 1, wherein the specific activity is at least 10,000 U/mg of protein.

22. A proteinaceous substance of claim 2, wherein the specific activity is at least 10,000 U/mg of protein.