CA1240987A - Urinary kallikrein assay: specific substrates and assay method - Google Patents

Abstract

URINARY KALLIKREIN ASSAY:
SPECIFIC SUBSTRATES AND ASSAY METHOD

Abstract Substrates and method for the assay of urinary kallikrein are provided. The substrates have the general formula R-L-Pro-L-phe-L-Arg-NH-(CH2)n wherein R is H, acetyl, benzoyl, cyclopentanecarbonyl, succinyl, R1-L-Ser or R1-L-Phe-L-Ser where R1 is H, acetyl, benzoyl, cyclopentanecarbonyl or succinyl, X is H, tritium, 3-iodo or 4-iodo, and n is 0 or 1.
Ratioactive label may be incorporated in the anilide or benzylamide moiety. Hydrolysis catalyzed by urinary kallikrein yields labeled aniline or benzylamine as product.
The assay method includes mixing the enzyme with substrate in a buffered solution, pH 7.5-10.5, the substrate being present preferrably at a concentration substantially below Km. After incubating to allow the reaction to proceed, the reaction is terminated and the radioactive hydrolysis product is separated and counted.
Also disclosed are compounds of the type R-L-Pro-L-Phe-L-Arg-NH-(CH2)n wherein R, R1 and n are as defined above. X is 3-OH or 4-OH.
Y1 is 3-iodo if X is 4-OH or 6-iodo if X is 3-OH. Y2 may be H
or 5-iodo if Y1 is 3-iodo, or 4-iodo if Y1 is 6-iodo.

Description

~2~

The present invention relates to certain novel com-pounds and assays using the same.
Kallikrein is an enzyme which functions physiologically in the formation of the hypotensive peptides bradykinin and kallidin by hydrolytic cleavage of the precursor peptide kininogen. The enzyme hydrolyzes peptide bonds on the carboxyl side of arginine or lysine residues and thus resembles trypsin and other proteases having a serine at the active site, such as thrombin, urokinase, and plasmin. There are two main groups of kallikreins, plasma and glandular. Urinary kallikrein ha~ s~milar characteristics to a glandular kallikrein. The urinary enzyme i8 of special interest because it is implicated in blood pressure regulation and regulation of sodium balance.
The assay of human urinary kallikrein is of clinical significance in the diagnosis of hypertension, in determining an appropriate course of treatment and in monitoring the effects of medication.
Serum and urinary kallikreins are immunologically dis-tinguishable and have different pH optima and substrate specificities. Human urinary kallikrein is immunologically distinct from rat or dog urinary kallikrein. The enzyme is found only in minute amounts in human urine. Development of an assay for human urinary kallikrein activity requlres the specific development of a suitable substrate capable of providing sufficient specificity and sensitivity. Compounds such as benzoylarginine ethyl ester and tosylarginine methyl ester, although hydrolyzed by the enzyme, are unsuitable because of the many known enzymes of trypsin-like specificity which also hydrolyze the substrates. In general, substrates whose hydrolysis is measured spectrophotometrically are unsuitable since such

- 2 - ;~

~,~

,~ ¦mea~ ements do n~t providel~u~ic~ent sensltivit9 for convenlent ¦
measurement of the minute amounts of human urinary kallikrein encountered in clinical practice.
The following abbreviations are employed herein:
Aoc amyloxycarbonyl Arg arginine Boc t-butyloxycarbonyl Bz benzoyl Ile isoleucine ~ Leu - leucine ; MCA 7-amino-4-methylcoumarin amide OHA O-hydroxyanilide Phe phenylalanine PIA p-iodoanilide pNA p-nitroanilide Pro proline Ser serine Tos tosyl Val valine 0 Z carbobenzoxy Amundsen et al., in Chemistry and Biology of the Kallikrein-Kinin System in Health and Disease, Pisano and Austin, Ed.J DHEW
Publication No. (NIH) 76-796, Washington, p. 215 (1977), disclo8e the use of Bz-L-Pro-L-Phe-L-Arg-pNA as a chromogenic sub8trate for plasma kallikrein. These authors reported at page 218 that glandular kallikrein, specifically bovine pancreati kallikrein, was "virtually inactive towards the tripeptide derivative". They reported that 20 ug of this enzyme had an activity on the substrate corresponding to 0.09 umole substrate 0 8plit/minute/mg dry protein. Their analysis is believed to I indicate that this substrate is not sensitive enough to be uçeful for assaying glandular kallikreins. Urinary kallikrein is a glandular kallikrein, and would therefore be expected to show little activity toward this substrate. Aasen et al., Eur. Surg. Res. 10,-50 (1978), state "glandular kallikrein is inactive on the substrate (Bz-L-Pro-L-Phe-L-Arg-p~A)" citing the hmundsen et al. reference.
; Claeson et al., Haemostasis 7, 62 (1978), disclose synthetic, chromogenic peptide substrates for several proteolytic enzymes.
0 The authors report that substrates containing Pro-Phe-Arg, e.g.

~L240987 Bz-L-pro-L-phe-L-Arg-pNA~ are the best substrates for plasma kallikrein. They found ~hat D-Pro-L-Phe-L-Arg-pNA was the best substrate for plasma kallikrein. The authors found that plasma kallikrein was approximately 1800 and 300 times as active as glandular kallikrein using D-Pro-L-Phe-L-Arg-pNA and Bz-L-Pro-L-Phe-L-Arg-pNA, respec~ively, as the substrate. These authors further report at p. 64 "none of the glandular substrates in Table 1 (including Bz-L-Pro-L-Phe-L-Arg-pNA) are sufficiently good, i.e. sensitive enough, to be of any practical value."
0 The best substrate for glandular kallikrein, as determined by these authors, was D-Val-L-Leu-L-Arg-pNA. This substrate is not structurally related to those invented by the applicants.
Morita et al., J. Biochem. 82, 1495 (1977), disclose new fluorogenic substrates for several proteolytic enzymes. The authors report that Z-L-Pro-L-Phe-L-Arg-MCA and L-Pro-L-Phe-L-Arg-MCA are susceptible to hydrolysis by plasma kallikrein and glandular kallikreins. The glandular kallikreins examined were proci~e pancreatic kallikrein and procine urinary kallikrein.
Porcine urinary kallikrein hydrolyzed L-Pro-L-Phe-L-Arg-MCA
0 the best. They report that 10 ul of this enzyme (no indication of amount) had an activity on the substrate corresponding to 9.7 umole split/minute/mg protein. Porcine urinary kallikrein shows about a 100 fold increase in sen~itivity to L-Pro-L-Phe-L-Arg-MCA than that shown by bovine pancreatic kallikrein to Bz-L-Pro-L~Phe-L-Arg-pNA as reported by Amundsen et al. In view of Amundsen et al. and Claeson et al., it is doubtful that this increase in sensitivity would make this substrate a sufficiently good one for assaying urinary kallikrein which is usually present in minute quantities.
0 Day et al., Agents and Actions, 6, 421 (1976), disclose several substrates for urinary kallikrein which were designed .

~LÆ~L(3987 to be radiolabelled and to yield reaction products having strong W absorptions or capable of being made fluorescent. The ¦
substrates designed by the authors included Boc-Pro-Phe-A~g-OHA
and Boc-Pro-Phe-Arg-PIA. There is no express indication in th~s paper of the optical activity of the amino acids in the disclosed substrates. It is virtually impossible to predict differences between optical isomers o~ the amino acids in substrates and the effect this may have on enzyme activity towards that substrate. The paper reports that the authors had begun 0 to 6ynthesize these substrates. According to the paper, if radiolabelled substrates were desired, the -OHA derivative -could be labelled with 125I using the chloramine T method and the -PIA derivatives could be labelled by catalytic tritiation via dehalogenation. This reference does not indicate whether any of the radiolabelled substrates had been made. In fact, the authors found that the catalytic tritiation of the PIA derivatives wa~ a dif~icult reaction with low yields and this not suitable for preparing a radiolabelled substrate.
In support of this fact, applicants herein describe the o results o a catalytic tritiation of L-Pro-L-Phe-L-Arg-PIA.
Applicant~ prepared the named compound by following the procedure tescribed in the appropriate part of Example 1 of this applicatior .
inra. The material produced behaved as a pure substance as i judged by paper electrophoresis at pH 2 and pH 5 and by thin layer chromatography using three separate solvent systems. 2 mg of the L-Pro-L-Phe-L-Arg-p-iodoanilide was sent to New England Nuclear Corporation, Boston, Massachusetss. New England Nuclear carried out the catalytic tritiation and returned 5 mCi of material to applicants. This material was then analyzed by ,0 applicants. If the material was L-Pro-L-Phe-L-Arg-13H]analide, it was expected to be retained by a CM-Sephadex column since it should have two (+) charges. The radiolabelled material was placed on a CM-Sephadex column which had been developed with ~240~3~'7 50% acetic acid. 90% of the [3H] was excluded by the column and behaved as a mixture of neutral compounds. That is, the excluded material did not migrate on paper electrophoresis at pH 2. 5% of the [3H] was retained and later eluted from the column using acetic acid. This material behaved as a mixture of (+) charged compounds as judged by paper electrophoresis at pH 2. This latter material was then analyzed to determine if it was hydrolyzed by human urinary kallikrein by uliizing the assay described in this application, infra. It was found that there 0 was not hydrolysis to yield 13H]analide and therefore, was not a substrate for this enzyme. Thus, a radiolabelled substrate for urinary kallikrein having an anilide moiety was not produced by this method. The Day et al. reference is thus prophetic and not descriptive of facts in existence at the time the paper was published.
EkenRtam et al., German Offenlegungsscrift 26 ~9 067, tisclose chromogenic substrates for serine proteases, which group of proteases includes the kallikreins-plasma and glandular.
The substrates described in this patent have a D-amino acid at the N-terminal end of the peptide, i.e. D-Al-l-A2-1-A3-. Thé
8ubstrates te~ted agalnst kallikrein were D-Val-L-Leu-L-Arg-pNA
and D-Ile-L-Leu-L-Arg-pNA as indicated on page 19. There is no clear indication in the patent whether plasma or glandula~
i kallikrein was examined. However, the Claeson et al. reference indicates that these two substrates are suitable for glandular kallikrein. These substrates are not structurally related to those invented by the applicants.
Svendsen, U.S. Patent No. 4,016,042, discloses a broad ; group of chromogenic or fluorogenic substrates for proteolytic S0 enzymes of class E.C. 3.4.4, which are essentially serine proteases. This patent discloses that Bz-L-Pro-L-Phe-Arg-pNA
is a substrate for plasma kallikrein and useful for measuring ~ ~2~9~3~
, minute quantities of it. It appears that glandular kallikreins were not examined using the disclosed substrates. In view of Amundsen et al., and Claeson et al. this substrate would not be useful for glandular kallikreins.
Applicants have prepared substrates for urinary kallikrein which are derivatives of L-Pro-L-Phe-L-Arg-anilide or L-Pro-L-Phe-L-Arg-benzylamide. These derivatives are radio-labelled in the anilide or benzylamide portion of the substrates.
, Applicants found that it is essential that all of the am~no acids D of these substrateæ be in the L-form. It is not possible to predict the effect which different optical isomers will have on a partlcular substrate as to the enzyme activity towards that subs~rate. Applicants found that urinary kallikrein -- a glandular kallikrein -- i8 very active towards the disclosed sub8trate~ having all the amino acids in the L-for,m. This is contrary to the teachings of the prior art as shown by Amundsen et al., and Claeson et al. The prior art as discussed above found that the best substrates for assaying any kallikrein species had the first amino acid in the D-form and the others 0 in the L-form. Applicants have found that it is critical that all of the amino acids of their substrates be in the L-f~rm for assaying urinary kallikrein. Furthermore, Claeson ét al.
found that the best substrate for glandular kallikrein was D-Val-L-Leu-L-Arg-pNA. Applicants have found that the best sub~trates for urinary kallikrein are I.-Pro-Phe-L-Arg-X where X ls a benzylamide or anilide compound. Applicants found that they could rapidly assay minute quantities of urinary kallikrein--in fact, as little as 5 ng of urinary kallikrein--using the substrates of this invention. This is not possible using the 0 sub,strates of the prior art. Thus, applicants have developed a highly sensitive assay for urinary kallikrein which was not available in the prior art. ' ~ ~2~398t~

Accordingly, in one aspect, the present invention provides a compound having the formula:
R - L - Pro - L - Phe - L - Arg - NH - (CH2)n - R2 ( I ) wherein R is hydrogen, acetyl, benzoyl, cyclopentanecarbonyl, succinyl, Rl-L-Ser or Rl-L-Phe-L-Ser where Rl is hydrogen, acetyl, benzoyl, cyclopentanecarbonyl or succinyl, n is 0 or 1, and R2 is the moiety ~ X (II) wherein X is hydrogen, tritium, 3-iodo or 4-iodo, or ~ Xl (III) wherein X1 i5 3-hydroxy or 4-hydroxy, Yl is 3-iodo if X1 is 4-hydroxy, or 6-iodo if Xl is 3-hydroxy, and Y2 is hydrogen or 5-iodo if Y1 is 3-iodo, or 4-iodo if Yl is 6-iodo.
These compounds are useful as substrates for an assay of human urinary kallikrein. The present invention provides, therefore, in another aspect, a method for the assay of human urinary kallikrein, comprising (1) mixing the kallikrein with a substrate compound as defined above, (2) incubating the kallikrein I with the substra-te to permit kallikrein-catalyzed hydrolysis of the substrate, (3) terminating the kallikrein-catalyzed hydrolysis, (4) separating the hydrolysis product from the substrate, and (5) measuring the amount of the hydrolysis product formed by counting the radioactivity thereof, whereby an assay of the kallikrein is obtained.
All amino acid residues are in the L-configuration herein, unless otherwise specified.

'~ ~LZ 4~ g 8 ~

The substrates may be radioactively labeled by substitution of tritium or [14C] in the anilide or benzylamide moiety, or by iodination of the anilide or benzylamide moiety with [125I]
in the 3- or 4-position. The 3-iodo and 4-iodo derivatives are also active substrates for human urinary kallikrein. The substrates are highly selective for human and other mammilian urinary kallikreins and have a very low Km~ a fact which is especially useful for radioassay.
Hydrolysis of the radioactively labeled substrates by human urinary kallikrein results in formation of radioactively labeled benzylamine or labeled aniline, respectively. In the assay method, the products are extracted from the reaction mixture with an aprotic solvent and counted by standard techniques. Using the substrates and assay method of the present invention, the kallikrein activity of 50 ul dialyzed human urine is readily measured in a 15 minute incubation at 37C. At little as S ng of enzyme can be measured under these conditions.
The compounds of the present invention have been discovered to be excellent substrates for human urinary kallikrein. ProPhe-Arg-13H] anilide has a Km of approximately 0.8 uM and a Vmax of - 8a -~ 0 98'7 approximately 0.39 mmol/min/mg. ProPheArg-l3H] benzylamide has a ~ of approximately 3.0 uM and a Vmax of approximately 0.~8 mmol/min/mg. The low Km renders these substrates especially suitable for a radioassay. A small amount of substrate is desirable in a radioassay, since sensitivity is maximized when the substrate has a high specific radioactivity. The specific radioactivity is a function of the proportion of labeled to unlabeled molecules in the substrate preparation. Generally, the cost of the substrate preparation will increase as the specific radioactivity is increased. Therefore, substrates that can be used at low concentration, such as those of the present invention, are advantageous. The low Km values exhibited by the substrates of the present invention are ~ndlcative of a highly specific binding between the enzyme and substrate. Such specific binding generally increases the likelihood that substrate will only be hydrolyzed by human urinary kallikrein rather than some other enzyme with trypsin-like activity, especially when substrate is present in relatively small amounts in the assay mixture.
The present substrates have the unexpected property of inhibiting human urinary kallikrein when present at çoncen-trations greater than about 10 uM, or greater than about

3 x Km~ Substrate inhibition has not previously been I reported for human urinary kallikrein, but may account for the previously mentioned lack of activity of the p-nitroanillde compounds with this enzyme. The latter are typically used at higher concentrations for spectrophotometric assay. The discovery of substrate inhibition with the compounds of the present invention raises the possibility of therapeutic use in reduction of excessive glandular kallikrein activity, in vivo.

, _ g _ ' ~2~(~9~7 According to the foregoing principles, the substrates of the present invention may be used at concentrations ranging from about 1 nM to approximately 10 uM, depending on the amount of enzyme present, and the specific activit~
of the substrate. Preferably, the assay is carried out using 10 nM to 40 nM substrate.
Previous attempts to synthesize the [3H]-anilide compcunds of the present invention by catalytic-dehalogenation of the 4-iodoanilide in the presence of [3H2] resulted in poor 0 yields, at best. Surprisingly, it has been found that catalytic trit~ation of the 3-iodobenzy~amide provides an excellent yield of the [3H]-benzylamide. Synthesis of [3H]-anilides is preferably accomplished by coupling ~3H]-aniline with Boc-ProPheArg(nitro), yielding substrate having a ~pecific radioactivity of about O.l to 2 Cilmmol, depending on available ~pecific radioactivity of [3H]-aniline starting material. The catalytic-[3H] of 3-iodobenzylamides yields specific radioactivities on the order of 25 Ci/mmol.
It has been noted that spectrophotometric assays, such O as developed for serum kallikrein, depend upon accumulation of sufficient product to produce measurable change in optical density at a chosen wavelength. Optimal substrates for such i assays exhibit high Vmax values with the enzyme, and optimal reaction conditions are chosen so as to achieve a high Vmax.
The desirability of the D-amino acid analogs of the prior art is most likely related to an enhancement of Vmax, possibly accompanied by increased Km, attendant upon D-amino acid substitution. In contrast, the present substrates are composed of L-amino acids, to maximize the binding affinity O of the enzyme for the substrate. High binding affinity (as manifested by a low Km) permits the use of substrates at low concentration in the assay, yielding the advantages discussed, supra.
.

~ o~
¦ Another unexpected feature of the present substrates ¦is the fact that unprotected derivatives (where R is H) are ¦as effective, or more effective, than N ~-substituted ¦derivatives. A protecting group was previously considered ¦essential to avoid degradation of substrate by amino peptidases.
¦BY contrast, we have now found that no significant degradation ¦of this type occurs in the assay of urinary kallikrein, under Ithe assay conditions employed herein.
¦ The pH optimum for the assay of human urinary kallikrein 0 ¦using substrates of the present invention was f~und to be ¦unusually high, with maximal activity in the range pH 9.0 ¦to 10Ø Enzyme activity was approximately one-tenth ¦optimal at pH 7.0 and approximately one-half optimal at pH 8.0 ¦and pH 10.7. Above pH 10.7, there was a rapid loss of activity ¦with increasing pH. Tris(2-amino-2-hydroxyethyl-1,3-propanediol) ¦buffer is preferred to either Hepes (N-2-hydroxyethylpiperazine-¦N'-2-ethanesulfonic acid) or phosphate buffers. Accordingly, for the assay method, 0.05 M tris buffer pH 9.5 is preferred.
At pH g.5, the reaction of human urinary kallikrein with the substrates of the present invention at 37C or 15 minutes is linear for kallikrein in the range of 20 ng/ml to 700 ng/ml.
Much lower amounts of kallikrein can be measured by extending the incubation time or reducing the volume of the reaction mixture.
Human urine is treated to remove low molecular weight impurities, either by overnight dialysis against O.g%(w/v) NaCl or by gel filtration. A typical reaction mixture may contain 50 ul dialyzed urine and 0.1 uCi of substrate at a final concentration of 20 ng/ml buffered with 0.1 M tris-HCl, pH 9.5, in a total reaction volume of 100 ul. Reaction tubes are incubated at 37C for two hours. Variations of ~L2~098t7 incubation time and temperature for the purpose of increasing ¦ or decreasing the amount of product formed are known expedients within the scope of ordinary skill in the art. Reactions may ¦be stopped by any technique which causes rapid termination of ¦ the enzyme-catalyzed reaction with significant alteration of ¦ the product concentration. Use of an excess volume of base, ¦ e.g , 1.0 ml of 0.1 N NaOH, is preferred. The product is separated from the reaction mixture by extraction with a ¦ measured volume of an aprotic solvent such as toluene or ether, O ¦or by other suitable means such as ion-exchange chromatography.
¦Where solvent extraction is employed, an aliquot of the solvent ¦ (containing reaction protuct) may be sampled for radioactivity ¦counting. Scintillation countiffg is preferred for counting ¦tritium, and use of an extracting solvent such as toluene or ¦ether permits direct transfer of a solvent aliquot to the ¦scint~llation fluid.
A standard curve may be constructed substituting purified ¦kallikrein for dialyzet urine. A unit of e~zyme is defined ¦herein as that a unt requiret to hydrolyze substrate at an 0 initial rate of lZ per minute at 37C. The foregoing definition ¦is appli~able where the concentration of substrate is well ¦below Km such that hytrolysis is first orter with respect to ¦substrate concentration. The number of kallikrein units/ml ¦of dialyzet urine is glven by 2(Test c.p.m. - Blank c.P.m.) Total Substrate c.~.~ X 100 ~ncubation time (min) X vol. of enzyme in ml where 1.0 ml of solvent is used for the extraction and an aliquot of 0.5 ml is taken for countin~. The blank c.p.m.
value is determined in a control reaction lacking enzyme but otherwise treated identically.
.

_ . _,. _ ............. , .. ,.. ,. ... ,.. ...... ........ . _ . __ . __ ._.. __,._. .___ ... .....

r~
1 ~2~0987 ¦ Example 1 Synthesis of H-ProPheArg [3H]-benzylamide.
l Boc-ProPhe-OH was synthesized by conven~ional solution ¦phase peptide syn~hesis, coupling Boc-Pro-N-hydroxy succinimide ¦ester with Phe-benzyl ester toluene sulfonic acid salt using ¦N-ethylmorpholine and l-hydroxybenzotriazole. An oily product ¦was obtained in 98% yield. The product was deprotected by ¦hydrogenolysis in ethanol with 5~/O palladium on barium sulfate ¦aæ catalyst. The peptide gave white crystals from ethyl acetate 0 ¦in 51.9% yield, m.p. 140.5-141C. The product behaved as a ¦pure compound in 3 thin-layer chromatographic systems and on ¦electrophoresis at pH 5Ø The peptide ~as not reactive with ¦ninhydrin reagent but was reactive with o-tolidine/chloride ¦(C12) reagents.
l Aoc-Arg(Tos)-3-iodobenzylamide was prepared by coupling ¦Aoc-Arg(To~)-OH .~ith meta-iodobenzylamine by using dicyclo hexylcarbo~iimide and l-hydroxybenzotriazole. The coupling reaction product ~m.p. 80C) was deprotected with ethanolic HCl to yield H Arg(Tos)-3-iodobenzylamide HCl.
0 The protected dipeptide Boc-ProPhe-OH was coupled to H-Arg(Tos)-3-iodobenzylamide HCl in the presence of l-hydroxy-benzotriazole in dimethylformamide (DMF) with dicyclohexyl-carbodiimide at about 0C with N-ethyl morpholine as base.
The reaction yielded a foam-like materialj Boc-ProPheArg(To8)-3-iodobenzylamide. The product was deprotected by treatment with anhydrouæ HF to yield H-ProPheArg-3-iodobenzylamide.
The catalytic tritiation was performed essentially as follows: H-ProPheArg-3-iodobenzylamide (10 mg) was dissolved in 2.0 ml of DMF:water (1:1 by volume). To this was added 0 10 mg of 10%(wtw) palladium on CaCO3 and 10 Ci of tritium gas.
The reaction was stirred at room temperature for 4 hours, ~2 4~ 98~7 filtered, the filter washed with DMF:water, lyophilized to remove labile tritium and lyophilized again after addition of water. The product, H-ProPheArg-[3H~ benzylamide, was then dissolved in 25 ml water. Total radioactivity was 295 mCi, at a specific radioac~ivity of 25.6 Ci/mmole.
The labeled peptide was purified by chromatography in Bio-Rex 70 ttrademark, BioRad Laboratories, Fullerton, ` California) using a gradient of acetic acid from 10~L(v/v) to glacial. The purified compound could not be separated 0 from authentic unlabeled ProPheArg-benzylamide in four thin layer chromatography systems nor upon electrophoresis at pH l.9 or pH 5Ø Acid hydrolysis with 6 N HCl at 115C
for 20 hours yiélded benzylamide, proline, phenylalanine and arginine, identified by chromatography on silica gel plates eluted with CHC13-methanol-H2O (12:9:4 parts by vol~me) having Rf vfllues of 0.86, 0.20, 0.48, and 0.08, respectively. 98.5% of the tritium was contained in the benzylamine fraction, and the amino acid products each contained less than 0.8% of the total radioactivity.
0 Example 2 Synthesis of H-SerProPheAr~-[3H] benzylamide and N-PheSerProPheAr~-~3H] benzYlamide.
Essentially, the method of Example 1 is employed except I that the starting material Boc-Pro-N-OH-succinimide ester was eplaced by Boc-Ser(Benzoyl)Pro-N-OH-succinimide ester or oc-PheSer(Benzoyl)Pro-N-OH-succinimide ester, respectively, synthesized by standard solution phase peptide synthesis echniques. The starting materials are either coupled to a rotected phenylalanine, as in Example 1, and thence to 0 Arg(Tos)-3-~or-4-)iodobenzylamide, in a two-step reaction, or ~dlrectly -Phe-Arg(N02)-3-~Or-4)-iodo-benzylamide formed , .
_ ~ - 14 -11 ~2~0~3'7 by coupling H-Arg(NO2)-3(or -4)-iodo-benzylamide with Boc-Phe-N-OH_succinimide ester. The coupling reactions are followed by standard deprotection reactions. Catalytic tritiation is performed essentially as described in Example 1.
The compounds H-SerProPheArg-3-iodobenzylamide, H-SerProPheArg-4-iodobenzylamide, H-SerProPheArg-[3H~-benzylamide, H-PheSerProPheArg-3-iodobenzylamide, H-PheSerPro--- PheArg-4-iodobenzylamide and H-PheSerProPheArg-t3H]benzy}amide are suitable substrates for the assay of human urinary 0 kallikrein.
Example 3 Synthesis of H-ProPheAr~-[3H]-analide.
The synthesis was achieved by the coupling of [3H]-aniline (commercially available) with Boc-ProPheArg(NO2)-OH. The starting compound, Boc-ProPhe-OH was made es~entially a~
described in Example 1. The dipeptide was coupled to H-Arg-(NO2)-benzylester 2-toluene sulfonic acid salt in the presence of l-hydroxybenzotriazole in DMF with dicyclohexylcarbodiimide at about 0C with N-ethyl morpholine as base. The reaction yielded a yellow oil, Boc-ProPheArg(NO2)-benzylester, in 85%
yielt. The protuct behaved as a pure compound in three thin layer chromatography systems and on electrophoresis at pH 5.
It was not reactive with ninhydrin reagent but was reactive with o-tolidine/chlorine reagents.
The product was saponified with 1.1 equivalents o~
1 N KOH in methanol at room temperature for 1 hour. The reaction yielded a gum-like product, Boc-ProPheArg(NO2)-OH, in 61.5% yield. The product was homogeneous in three thin-layer systems and on electrophoresis at pH 5. The peptide was not reactive with ninhydrin but was reactive with o-t~lid hlorine reagents.
. .

, _ 15 -~, 11 ~
Boc-ProPheArg(NO2)-OH was coupled with [3H]-aniline in DMF by dicyclohexylcarbodiimide in the presence of l-hydroxy-benzotriazole at 4C overnight. The anilide was then deprotected with anhydrous ~F and further purified by chromatography, first on Bio-Rex 70 eluted with an acetic acid gradient, 1%-50%(v/v), then on Sephadex G-10 (trademark, Pharmacia, Inc., Uppsale, Sweden) using 10%(v/v) acetic acid as eluent. The [3H]-anilide was homogeneous in two thin-layer chromatography systems and on electrophoresis at pH 2.
0 H-ProPheArg-13H~anilide is an excellent substrate for human urinary kallikrein. The enzymatic hydrolysis products are H-ProPheArg-OH and [3H]aniline.
The foregoing procedure i8 used to synthesize H-SerPro-PheArg-13H]anilide and H-PheSerProPheArg-13H] anilide by coupling [3H]aniline to the respective peptides. The resulting compounds are substrates for human urinary kallikrein.
Example 4 Preparation of 114C]-labeled substrates.
The synthetic procedure of Example 3 i8 adapted to make 114Cl labeLed substrates by substituting ~14C~-aniline for 13H]aniline in the reaction. The same reaction is also used to produce 114C]-benzylamide substrates by substituting [14C]_ benzylamine for 13H]-analine.
I Example 5 Synthesis of 125I-labeled substrates.
When 3-hydroxyaniline, 4-hydroxyaniline, 3-hydroxy-benzylamine or 4-hydroxybenzylamine are substituted for 13H]-aniline in the synthesis procedure of Example 3, the corresponding 3-(or 4-)hydroxy anilides or benzylamines are 0 produced. These compounds are readily iodinated with [125I] or by the chloramine-T method to yield the corresponding [131I]-.

9~ 7 ¦or [125I]-OH-anilides or OH-benzylamides. These iodinated ¦compounds are effective substrates for human urinary kallikrein.
¦3-(or 4-)iodo-anilines or 3- (or 4-iodobenzyl2mines are ~available commercially. Iodination of 4-OH-anilides or ¦benzylamides yields 3- and 3,5-iodo-derivatives. Iodination ¦of the 3-OH anilides or benzylamides yields 6- or 4,6-iodo-¦derivatives.
I Example 6 I
NC~-acetyl-, benzoyl-, and cyclopentanecarbonyl-derivatives ¦of H-ProPheArg-[3H]benzylamide were prepared by reacting the ¦respective acid chlorides with the labeled peptide in dioxane ¦and 1 M NaHCO3 (1:1 by volume) buffered at pH 8.5 with 1.5 M
¦Na2CO3. Succinyl-ProPheArg-[3H]benzylamide was prepared using ¦mono-N-succinimidyl succinate in 1 M NaHC03 and DMF (1:2 by ¦volume). The acylated derivatives were purified by chromatography ¦on Sephadex G-10 developed with 5%(v/v) acetic acid.
N -acylated derivatives of H-P~oPheArg-[3H]anilide are prepared in the manner described, suPra, for the benzylamides.
Urinary kallikrein-catalyzed hydrolysis of the cyclopentane-0 carbonylated and acetylated substrates proceeded at generally similar rates as for the non-acylated compounds. Benzoyl-¦ProPheArg-[3H]benzylamide was hydrolyzed at about 1/2 the rate and the succinyl derivative was hydrolyzed at 1/10 the rate. The foregoing results were obtained under assay conditions optimized for the non-acylated substrate, using approximately 20 nM substrate and 60 ng purified urinary kallikrein (Oza, N.B. and Ryan, J.W., Biochem.J. 171, 285 (1978)), in 100 ul of 0.1 M tris HCl, pH 9.S, incubated at 37C for 15 minutes. These conditions may not be optimal 0 for the acylated substrates. However, even under possibly subQptimal conditions, the N -acyl derivatives are clearly active substrates.

~ 1 3L240~8~7 Example 7 Kinetic binding constant (Km) of human urinary kallikrein with H-ProPheArg-[H]benzylamide and H-ProPheArg-[3H]anilide.
The double reciprocal plot method of H. Lineweaver and D. Burk (J.Am.Chem.Soc. 56, 658 (1934)) was used to estimate Km and Vmax for these substrates. Purified human urinary kallikrein prepared as described by Oza, N.B. and Ryan, J.,.
suPra, was used in this study. Reactions were conducted at 37C for 15 minutes with 62.5 ng of enzyme and the indicated 0 amount of substrate, in 0.05 M tris-HCl pH 9.5. React;ans ¦ were terminated by the addition of a nine-fold excess of 0.1 N NaOH. The 3H-labeled product was separated from substrate by extraction with an equal volume (1 ml) of toluene. A 0.5 ml aliquot of the toluene phase was trans-ferred to a scintillation vial containing a toluene-based scintillation fluid and counted. Results for H-ProPheArg-~3H]-benzylamide are shown in FIG. 1, and for H-ProPheArg-~3H]anilite in FIG. 2. ~he units of l/v were nmol 1 min.
Units of l/S were umole liter. For H-ProPheAr~-[3H]
0 benzylamide, ~ was measured as 3.0 uM and Vmax was 0.88 ¦ mmol/min/mg protein. For H-ProPheArg-[3H]-anilide, ~ was 0.8 uM and Vmax was 0.39 mmol/min/mg protein-Example 8 I Relative activity of halo~enated substrates.
. Unlabeled halogenated 4-iodoanilide substrates, prepared essentially as described in Example 5, were compared with respect to reaction rate with human urinary kallikrein in a standard assay. Each substrate, 150 nmol, was incubated with purified human urinary kallikrein in 0.7 ml of 0.05 tris-HCl, ~0 pH-8.1 at 37C. At timed intervals, 100 ul of reaction . miXture was added to 1.0 ml of 0.1 N NaOH. The resulting ..

~ - 18 -~: '' solution was extracted with CHC13 and the organic phase was evaporated to dryness. The residue was dissolved in buffer and examined spectrophotometrically in the 240-300 nm range, where a difference spectrum between substrate and product is measurable. The results are shown in Table l,-with rates of hydrolysis expressed as nmol/min/mg enæyme.

RELATIVE AFFINITY OF HALOGENATED SUBSTRATES FOR
~UMAN URINARY KALLIKREIN

_ _ l0Substrate Rate of Hydrolysis Phe-Arg-4-iodoanilide ~il Pro-Phe-Arg-4-iodoanilide 0.73 Ser-Pro-Phe-Arg-4-iodoanilide 0.62 Phe-Ser-Pro-Phe-Arg-4-iodoanilide ~ 1.12 Example 9 Reaction specificity.
The hydrolysis of H-ProPheArg-[3H]benzylamide by a series of serine proteases was measured. Reactions were carried out using 20 nM substrate in 0.1 M tris-HCl buffer, pH 7.5 or ~0 pH 9.5, at 37C for 15 minutes in the presence of varied . amounts of enzyme. The results are shown in Table 2, expressed as the amount of enzyme necessary to hydrolyze 10% of the substrate under the given conditions. "No hydrolysis" is used to denote no measurable reaction with a miximum of 0.5 mg/ml enzyme.
.. . .

g~

SPECIFICITY STUDIES: HYDROLYSIS OF
PRO-PHE-ARG-[3H]BENZYLAMIDE
BY SERINE PROTEASE ENZYMES

Amount of Enzyme Required for 10%
Enzyme pH Hydrolysis Human Urinary Kallikrein9.5 45 ng (10 nM) 7.5 265 ng Plasmin . 9.5 30,500 ng .
7.5 17,200 ng Trypsin 9 5 1,030 ng O~-Chymotrypsin - 9 5 No hydrolysis .

Urokinase 9 5 No hydrolysis Thrombin 9 5 No hydrolysis While the invention has been described in connection with specific embodiments thereof, it will be understood that it is capable of further modifications and this application is intended to cover any variations, uses, or adaptations of the invention following, in general, the principles of the invention and including such departures from the present disclosure as come within known or customary practice within the art to which the invention pertains and as may be applied to the essential features hereinbefore set forth, and as follows in the scope of the appended claims.
.", _~

Claims (27)

The embodiments of the invention in which an exclusive property or privilege is claimed are defined as follows:

1. A compound having the formula:
R - L - Pro - L - Phe - L - Arg - NH - (CH2)n - R2 I
wherein R is hydrogen, acetyl, benzoyl, cyclopentanecarbonyl, succinyl, R1-L-Ser or R1-L-Phe-L-Ser where R1 is hydrogen, acetyl, benzoyl, cyclopentanecarbonyl or succinyl, n is 0 or 1, and R2 is the moiety II

wherein X is hydrogen, tritium, 3-iodo or 4-iodo, or III

wherein X1 is 3-hydroxy or 4-hydroxy, Y1 is 3-iodo if X1 is 4-hydroxy, or 6-iodo if X1 is 3-hydroxy, and Y2 is hydrogen or 5-iodo if Y1 is 3-iodo, or 4-iodo if Y1 is 6-iodo.

2. A compound as described in claim 1 wherein R2 is moiety II.

3. A compound as described in claim 2 wherein X is tritium.

4. A compound as described in claim 2 wherein X is 3-iodo or 4-iodo, at least a portion of said iodo substituents consisting of [125I].

5. A compound as described in claim 2 wherein the benzene ring to which X is attached contains at least one [14c].

6. A compound as described in claim 2 wherein R is hydrogen, H-L-Ser or H-L-Phe-L-Ser, X is hydrogen or tritium and n is 0.

7. A compound as described in claim 6 wherein R is hydrogen.

8. A compound as described in claim 2 wherein R is hydrogen, H-L-Ser or H-L-Phe-L-Ser, X is hydrogen or tritium and n is 1.

9. A compound as described in claim 8 wherein R is hydrogen.

10. H-L-Pro-L-Phe-L-Arg-[3H]benzylamine.

11. The N -acetyl, -benzoyl, -cyclopentanecarbonyl or -succinyl derivative of H-L-Pro-L-Phe-L-Arg-[3H]benzylamide.

12. H-L-Pro-L-Phe-L-Arg-[3H]anilide.

13. H-L-Pro-L-Phe-L-Arg-4-iodoanilide.

14. H-L-Ser-L-Pro-L-Phe-L-Arg-4-iodoanilide.

15. H-L-Phe-L-Ser-L-Pro-L-Phe-L-Arg-4-iodoanilide.

16. A compound as described in claim 1 wherein R2 is moiety III.

17. A compound as described in claim 16 wherein R
is hydrogen, H-L-Ser or H-L-Phe-L-Ser.

18. A compound as described in claim 17 wherein at least a portion of the iodo substituents are [125I] or [131I].

19. A compound as described in claim 18 wherein R

is hydrogen.

20. A method for the assay of human urinary kallikrein comprising:
(1) mixing said kallikrein with a substrate compound having the formula R-L-Pro-L-Phe-L-Arg-NH-(CH2)n-R2 I
wherein R is hydrogen, acetyl, benzoyl, cyclopentanecarbonyl, succinyl, R1-L-Ser or R1-L-Phe-L-Ser where R1 is hydrogen, acetyl, benzoyl, cyclopentanecarbonyl or succinyl, n is 0 or 1, and R2 is the moiety II

wherein X is hydrogen, tritium, 3-iodo or 4-iodo, or III

wherein X1 is 3-hydroxy or 4-hydroxy, Y1 is 3-iodo if X1 is 4-hydroxy, or 6-iodo if X1 is 3-hydroxy, and Y2 is hydrogen or 5-iodo if Y1 is 3-iodo, or 4-iodo if Y1 is 6-iodo, at least a portion of the molecules of said substrate having a [3H], [14C] or [125I] substituted for the carbon, hydrogen or iodine atoms in that portion of said substrate connected by amide linkage to the carboxy group of the arginine moiety, said substrate being present at a concentration less than in a buffer having a pH in the range of 7.5-10.5, (2) incubating said kallikrein with said substrate to permit kallikrein-catalyzed hydrolysis of said substrate, (3) terminating said kallikrein-catalyzed hydrolysis, (4) separating the hydrolysis product from said substrate, and (5) measuring the amount of said hydrolysis product formed by counting the radioactivity thereof, whereby an assay of said kallikrein is obtained.

21. A method as described in claim 20 wherein R2 is moiety II.

22. A method as described in claim 21 wherein said buffer has a pH in the range 8.5-10.5.

23. A method as described in claim 21 wherein said kallikrein-catalyzed hydrolysis is terminated by the additon of excess base.

24. A method as described in claim 21 wherein said hydrolysis product is separated from said substrate by extraction with an aprotic solvent.

25. A method as described in claim 23 wherein said beffer has a pH in the range of 9.0-10.0, said kallikrein-catalyzed hydrolysis is terminated by the addition of excess 0.1 N NaOH, and said aprotic solvent is toluene.

26. A method as described in claim 20 wherein R2 is moiety III.

27. A method as described in claim 26 wherein said buffer has a pH in the range of 9.0-10.0, said kallikrein-catalyzed hydrolysis is terminated by the addition of excess 0.1 N NaOH, and said hydrolysis product is separated from said substrate by extraction with toluene.